Pharmacovigilance in Psychiatry Edoardo PDF

Edoardo Spina
Gianluca Trifirò Editors
Pharmacovigilance
in Psychiatry
Pharmacovigilance in Psychiatry
Edoardo Spina • Gianluca Trifirò
Editors
Pharmacovigilance in
Psychiatry
Editors
Edoardo Spina Gianluca Trifirò
Department of Clinical and Experimental Department of Clinical and Experimental
Medicine Medicine
University of Messina University of Messina
Messina Messina
Italy Italy
ISBN 978-3-319-24739-7 ISBN 978-3-319-24741-0 (eBook)

DOI 10.1007/978-3-319-24741-0
Library of Congress Control Number: 2015957257
Springer Cham Heidelberg New York Dordrecht London

© Springer International Publishing Switzerland 2016
This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of
the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation,
broadcasting, reproduction on microfilms or in any other physical way, and transmission or information
storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology
now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication
does not imply, even in the absence of a specific statement, that such names are exempt from the relevant
protective laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in this book
are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the
editors give a warranty, express or implied, with respect to the material contained herein or for any errors
or omissions that may have been made.
Printed on acid-free paper
Adis is a brand of Springer

Springer International Publishing AG Switzerland is part of Springer Science+Business Media (www.
springer.com)
Preface and Acknowledgements
Primum non nocere (first, do no harm) is a guiding principle for all physicians
including psychiatrists that, whatever the intervention or procedure, the patient’s
well-being is the primary consideration.
Rational use of psychotropic drugs may improve the quality of life and the func-
tional status of patients with neuropsychiatric diseases while minimizing adverse
effects potentially associated to the pharmacological treatment. However, currently,
psychotropic medications are often misused and overused, especially in elderly
patients; thus, exposing this frail population to unmotivated and potentially life-
threatening risks.
In general terms, it is well known that premarketing randomized clinical trials
are designed to investigate, primarily, the efficacy of the drugs and may only partly
explore the drug safety profile. Therefore, the risks associated with newly marketed
drugs can be properly quantified and characterized only after their use in clinical
practice (i.e. post-marketing phase). For this reason, post-marketing pharmacovigi-
lance monitoring has been long recognized as the last phase of drug development.
Pharmacovigilance is a discipline that entails both cultural and scientific aspects.
On one hand, traditional pharmacovigilance activities based on spontaneous report-
ing of adverse drug reactions from patients and healthcare professionals are essen-
tial to increase the awareness of prescribers and users about the potential risks
associated to exposure to medicines. On the other, in the last few years, pharmaco-
vigilance as a science is rapidly evolving towards a more proactive approach in
terms of emerging safety issues detection, strengthening and validation, thanks to
innovative methodologies that have been developed in the context of outstanding
initiatives, such as US FDA-endorsed Sentinel and European FP-7-funded EU-ADR
and IMI-funded PROTECT.
It is nowadays well acknowledged that continuously growing availability of
databases with longitudinal electronic health records of millions of persons world-
wide offers the opportunity to get better insight, rapidly and cheaply, into real-life
psychotropic drug use and the benefit-risk profile of those medications in the gen-
eral population, as well as in specific frail categories of patients such as older peo-
ple, children and pregnant women.
v
vi Preface and Acknowledgements
A number of large-size observational studies have been conducted in several

continents in the last decades, exploring and confirming the associations of poten-
tially serious safety outcomes and use of drugs commonly prescribed in psychiatry,
ranging from risk of hemorrhagic stroke and use of antidepressants, especially
selective serotonin reuptake inhibitors, to risk of all-cause mortality and use of both
atypical and conventional antipsychotics in older people with dementia, to risk of
falls and benzodiazepine use and so on. Those data sources can complement the
traditional spontaneous adverse drug reaction reporting system in drug safety signal
management, also regarding psychotropic drugs.
Knowledge in pharmacovigilance is dynamic and requires rapid (re)assessment
of risk associated with currently marketed drugs, including psychotropic drugs.
This book presents a timely overview of updated evidence about the safety pro-
file of drugs that are commonly used in psychiatry, as well as of established and
advanced methodologies that have been used up to now for the conduct of post-
marketing pharmacovigilance studies.
We gratefully acknowledge the continuous advice from Prof. Bruno Stricker
(Erasmus University Medical Center of Rotterdam, the Netherlands) and Prof.
Achille P. Caputi (University of Messina, Italy), who inspired relevant parts of this
book through their innovative research in the field of pharmacovigilance.
Messina, Italy Gianluca Trifirò

Messina, Italy Edoardo Spina
Contents
Part I General Aspects of Pharmacovigilance in Psychiatry
1 Pharmacovigilance in Psychiatry: An Introduction . . . . . . . . . . . . . . 3

Edoardo Spina, Gianluca Trifirò, and Achille Patrizio Caputi
2 Adverse Drug Reactions: Definitions, Classifications
and Regulatory Aspects. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
Paola Maria Cutroneo and Giovanni Polimeni
3 Methods for the Post-Marketing Monitoring of Psychotropics
Safety: Interests and Pitfalls . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
Francesco Salvo, Annie Fourrier-Réglat, Nicholas Moore,
Bernard Bégaud, and Antoine Pariente
4 Contribution of UK Prescription-Based Event Monitoring
Methods in the Pharmacovigilance of Psychotropic Medications . . . 45
Deborah Layton
5 The Role of Healthcare Databases in Pharmacovigilance
of Psychotropic Drugs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
Gianluca Trifirò and Janet Sultana
6 Monitoring of Plasma Concentrations of Psychotropic
Drugs in Pharmacovigilance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
Christoph Hiemke and Ekkehard Haen
7 The Role of Pharmacogenetics in Pharmacovigilance
of Psychotropic Drugs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121
Alessandro Serretti and Chiara Fabbri
Part II Pharmacovigilance of Specific Psychotropic Drug Classes
8 Safety and Tolerability of Antidepressants . . . . . . . . . . . . . . . . . . . . . 149

Chris Manson, Robert Gordon, and David Baldwin
vii
viii Contents
9 Safety and Tolerability of Antipsychotics . . . . . . . . . . . . . . . . . . . . . . 167

Michael W. Jann and William K. Kennedy
10 Safety and Tolerability of Anxiolytics/Sedative-Hypnotics . . . . . . . . 191
Vincenzo Arcoraci and Edoardo Spina
11 Safety and Tolerability of Mood Stabilisers . . . . . . . . . . . . . . . . . . . . 209
Michele Fabrazzo and Alfonso Tortorella
12 Safety and Tolerability of Medications for ADHD . . . . . . . . . . . . . . . 233
Antonio Clavenna and Maurizio Bonati
Part III Pharmacovigilance of Psychotropic Drugs in Special Populations
13 Safety of Psychotropic Drugs in Children and Adolescents. . . . . . . . 257

Florentia Kaguelidou and Eric Acquaviva
14 Safety of Psychotropic Drugs in the Elderly . . . . . . . . . . . . . . . . . . . . 285
Kate L. Lapane, Anne Hume, Christine Ulbricht,
and Giovanni Gambassi
15 Safety of Psychotropic Drugs in Pregnancy and Breastfeeding. . . . . 299
Olav Spigset and Hedvig Nordeng
List of Contributors
Editors
Edoardo Spina, MD, PhD Department of Clinical and Experimental Medicine,

University of Messina, Messina, Italy
Gianluca Trifirò, MD, PhD Department of Clinical and Experimental Medicine,
Contributors
Eric Acquaviva Department of Child and Adolescent Psychiatry, Assistance

Publique - Hôpitaux de Paris, Robert Debré University Hospital, Paris, France
Vincenzo Arcoraci Department of Clinical and Experimental Medicine,
David Baldwin University Department of Psychiatry, Clinical and Experimental
Sciences Academic Unit, Faculty of Medicine, University of Southampton,
Southampton, UK
Bernard Bégaud Department of Medical Pharmacology, INSERM U657,
Bordeaux University, Bordeaux, France
University Hospital Centre de Bordeaux, Bordeaux, France
Maurizio Bonati Department of Public Health, Laboratory for Mother and Child
Health, Mario Negri IRCCS Institute of Pharmacological Research, Milan, Italy
Achille Patrizio Caputi Department of Clinical and Experimental Medicine,
ix
x List of Contributors
Antonio Clavenna Department of Public Health, Laboratory for Mother and

Child Health, Mario Negri IRCCS Institute of Pharmacological Research,
Milan, Italy
Paola Maria Cutroneo Department of Clinical and Experimental Medicine,
Chiara Fabbri Department of Biomedical and Neuromotor Sciences,
University of Bologna, Bologna, Italy
Michele Fabrazzo Department of Psychiatry, University of Naples SUN, Naples,
Italy
Annie Fourrier-Réglat Department of Medical Pharmacology, INSERM U657,
Giovanni Gambassi Department of Medical Sciences, Università Cattolica del
Sacro Cuore, Rome, Italy
Robert Gordon University Department of Psychiatry, Clinical and Experimental
Southampton, UK
Ekkehard Haen Department of Psychiatry and Psychotherapy and Department
of Pharmacology and Toxicology, University of Regensburg, Regensburg,
Germany
Institute AGATE e.V., Pentling, Germany
Christoph Hiemke Department of Psychiatry and Psychotherapy, University
Medical Center of Mainz, Mainz, Germany
Anne Hume College of Pharmacy, Rhode Island University, Kingston,
RI, USA
Department of Quantitative Health Sciences, University of Massachusetts
Medical School, Worcester, MA, USA
Michael W. Jann Department of Pharmacotherapy, University of North Texas
System College of Pharmacy (UNTSCP), Fort Worth, TX, USA
Department of Pharmacotherapy, University of North Texas Health Science Center
(UNTHSC), Fort Worth, TX, USA
Florentia Kaguelidou Department of Pediatric Pharmacology and
Pharmacogenetics, Assistance Publique - Hôpitaux de Paris, Robert Debré
University Hospital, Paris, France
University Paris 7 - Diderot, Sorbonne Paris Cité, Paris, France
William K. Kennedy Department of Pharmacy Practice, Mercer University
College of Pharmacy, Atlanta, GA, USA
List of Contributors xi
Department of Pharmacotherapy, University of North Texas Health Science Center

(UNTHSC), Fort Worth, TX, USA
Kate L. Lapane Department of Quantitative Health Sciences, Division of
Epidemiology of Chronic Diseases and Vulnerable Populations, University of
Massachusetts Medical School, Worcester, MA, USA
Deborah Layton Drug Safety Research Unit, Southampton, UK
Chris Manson University Department of Psychiatry, Clinical and Experimental
Southampton, UK
Nicholas Moore Department of Medical Pharmacology, INSERM U657,
Hedvig Nordeng Pharmacoepidemiology and Drug Safety Research Group,
School of Pharmacy, PharmaTox Strategic Research Initiative, Faculty of
Mathematics and Natural Sciences, University of Oslo, Oslo, Norway
Division of Mental Health, Norwegian Institute of Public Health, Oslo, Norway
Antoine Pariente Department of Pharmacology, INSERM U657, BP 36,
Giovanni Polimeni Department of Clinical and Experimental Medicine,
Francesco Salvo Department of Pharmacology, INSERM U657, University of
Bordeaux, Bordeaux, France
Alessandro Serretti Department of Biomedical and Neuromotor Sciences,
University of Bologna, Bologna, Italy
Olav Spigset Department of Clinical Pharmacology, St. Olav University Hospital,
Trondheim, Norway
Department of Laboratory Medicine, Children’s and Women’s Health, Norwegian
University of Science and Technology, Trondheim, Norway
Janet Sultana Department of Clinical and Experimental Medicine,
Alfonso Tortorella Department of Psychiatry, University of Naples SUN,
Naples, Italy
Christine Ulbricht Department of Quantitative Health Sciences, Division of
Epidemiology of Chronic Diseases and Vulnerable Populations,
University of Massachusetts Medical School, Worcester, MA, USA
Part I
General Aspects of Pharmacovigilance
in Psychiatry
Chapter 1
Pharmacovigilance in Psychiatry:
An Introduction
Edoardo Spina, Gianluca Trifirò, and Achille Patrizio Caputi
Abstract The discipline of pharmacovigilance is of particular importance in the

field of psychiatry. Pharmacotherapy is the principal modality of management in
several psychiatric disorders and psychotropic drugs are associated with a variety of
adverse drug reactions. Over the past decades, pharmacovigilance activity has led to
the identification of several adverse drug reactions caused by psychotropic drugs,
resulting in their withdrawal from the market or restrictions in use. Psychotropic
medications are often administered for longer periods and are commonly prescribed
in combination with other drugs and, therefore, may be involved in clinically rele-
vant drug interactions. Psychotropic agents may be prescribed to populations at
higher risk of developing adverse effects. In particular, they are increasingly used to
treat psychiatric disorders in children and adolescents, as well as in the elderly, and
may be used by pregnant or lactating women. The main source of knowledge on
tolerability and safety of psychotropic drugs comes from clinical trials, but this is
associated with several limitations. Pharmacovigilance programs are designed to
gather information on what effects drugs have in the real world rather than in groups
of carefully selected clinical trial populations. For the abovementioned reasons, it is
important that psychiatrists become familiar with the concepts and methods of phar-
macovigilance as they have a key role in identifying and reporting new or serious
adverse drug effects.
Keywords Pharmacovigilance • Psychiatry • Psychotropic drugs • Adverse drug

reactions
The World Health Organization (WHO) defines pharmacovigilance as “science and

activities relating to the detection, assessment, understanding and prevention of
adverse effects or any other drug-related problem” (WHO 2000). As a result, a wide
range of healthcare professionals, including doctors, nurses, and pharmacists can
E. Spina (*) • G. Trifirò • A.P. Caputi

Department of Clinical and Experimental Medicine,
e-mail: espina@unime.it; trifirog@unime.it; achille.caputi@unime.it
© Springer International Publishing Switzerland 2016 3

E. Spina, G. Trifirò (eds.), Pharmacovigilance in Psychiatry,
DOI 10.1007/978-3-319-24741-0_1
4 E. Spina et al.
contribute to pharmacovigilance activities. Such activities are a critical aspect of

medical practice and have the potential to reduce and possibly prevent drug-related
harm and associated costs.
Pharmacovigilance is of particular importance in the field of psychiatry
(Rajkumar and Melvin 2014). Most psychiatric disorders are primarily treated with
drugs, all of which are associated with their own adverse drug reactions (ADRs).
Soon after the advent of modern psychopharmacology in the early 1950s, it became
clear that the use of first available psychotropic drugs was sometimes associated
with life-threatening or disabling adverse effects. This is the case with neuroleptic
malignant syndrome and tardive dyskinesia associated with traditional antipsychot-
ics or the potentially fatal hypertensive crises occurring when irreversible mono-
amine oxidase inhibitors, the first clinically effective antidepressants, are given in
combination with tyramine-rich food.
The birth of pharmacovigilance is closely related to psychiatry. In the late 1950s,
thalidomide was widely used in several countries as an antiemetic and a “safe seda-
tive,” particularly effective when given to pregnant women. It was soon noticed by
physicians that babies exposed to the drug in utero developed congenital malforma-
tions (McBride 1961). Widespread use of thalidomide in Europe, Australia, and
Japan resulted in approximately 10,000 children born with phocomelia, leading to
the ban of thalidomide in most countries. The thalidomide tragedy is generally
regarded as the beginning of legislated pharmacovigilance (Moore 2013). This
disaster marked a turning point in toxicity testing, as it prompted international regu-
latory agencies to develop systematic toxicity testing protocols. From such a trag-
edy, national pharmacovigilance centers emerged, as well as the development of
spontaneous reporting to identify ADRs that could justify or mandate regulatory
action.
Over the years that followed, pharmacovigilance activity has led to the identifi-
cation of several adverse drug reactions caused by psychotropic drugs, resulting in
their withdrawal from the market or restrictions in use. In 1973, a number of fatal
cases of agranulocytosis cases occurred with clozapine treatment which led to the
withdrawal of this efficacious antipsychotic drug in some countries and to restric-
tion of use in many others (McKenna and Bailey 1993). Zimelidine was the first
selective serotonin reuptake inhibitor (SSRI) antidepressant to be marketed in the
early 1980s. Within a year and a half of its introduction, rare case reports of Guillain-
Barré syndrome emerged that appeared to be caused by the drug, prompting its
manufacturer to remove it from the market (Fagius et al. 1985). The antidepressant
nomifensine, a noradrenaline and dopamine reuptake inhibitor, was withdrawn
worldwide by the manufacturer in January 1986 following a rising incidence of
reports of acute immune hemolytic anemia with serious clinical sequelae, including
a number of fatalities (Stonier 1992). The clinical use of the atypical antipsychotic
remoxipride was severely restricted in 1993, soon after its marketing, due to reports
of aplastic anemia (Nadal 2001). In 1994, alpidem, an anxiolytic drug from the
imidazopyridine family, was withdrawn from the market following reports of severe
liver damage (Berson et al. 2001). The atypical tricyclic antidepressant (TCA)
amineptine, a selective inhibitor of the reuptake of dopamine and, to a lesser extent,
1 Pharmacovigilance in Psychiatry: An Introduction 5
noradrenaline, was marketed in some European countries as an antidepressant and

soon gained a reputation for abuse potential due to its short-lived, but pleasant,
stimulant effect experienced by some patients. The emergence of severe cases of
hepatotoxicity, along with the potential for abuse, led to the suspension of market-
ing authorization in 1999 (Lazaros et al 1996, No authors listed 1999). More
recently, there were concerns on the propensity of some antipsychotics to prolong
corrected QT (QTc) interval and to cause severe cardiac arrhythmias including tor-
sades de pointes and sudden death. This led to the withdrawal of some antipsychot-
ics from the market (e.g., thioridazine), temporary suspension (e.g., sertindole), or
restriction of use (e.g., pimozide) (Glassman 2005).
A study investigating FDA ADR reports from 1998 to 2005 found that several of
the drugs in question were psychotropic agents such as antipsychotics (clozapine,
olanzapine, and risperidone), antidepressants (duloxetine, sertraline, paroxetine,
bupropion), mood stabilizers (carbamazepine, valproate, and lamotrigine), and also
anti-ADHD medication such as atomoxetine (Moore et al. 2007). Another study
reviewing nine major ADRs reported in Europe from 1995 to 2008 found that psy-
chotropic drugs were implicated in two ADRs, seizures caused by bupropion, and
suicidality in children prescribed SSRIs (Harmark and van Groothest 2008).
Interestingly, the latter was identified on reanalyzing data provided from pharma-
ceutical companies, while the former was identified from analysis of physician
reports. A more recent investigation concluded that of the 19 drugs withdrawn from
the European market between 2002 and 2011, 4 were psychotropic agents (nefazo-
done, thioridazine, veralipride, and aceprometazine+acepromazine) (McNaughton
et al. 2014).
There are several reasons why pharmacovigilance is important to psychiatry
(Rajkumar and Melvin 2014). As psychiatric disorders are often chronic, prolonged
pharmacological treatment may be required, thus increasing the possibility of long-
term ADRs. Psychotropic medications are commonly prescribed in combination
with other drugs used to treat comorbid psychiatric or somatic disorders and, there-
fore, may be involved in clinically relevant drug interactions. Psychotropic agents
are increasingly used to treat psychiatric disorders in children and adolescents, as
well as in the elderly, and may be used by pregnant or lactating women. Side effects
of psychotropic medications can mimic psychiatric symptoms (anxiety, insomnia,
somnolence, suicidality) and distinguishing between ADRs and symptoms of the
underlying illness can be difficult. Moreover, chronic use of many psychiatric medi-
cations may lead to physical dependence, and abrupt discontinuation may result in
withdrawal symptoms.
The main source of knowledge on tolerability and safety of psychotropic drugs
comes from clinical trials, but this is associated with several limitations (Stricker
and Psaty 2004). Clinical trials are inherently limited in their ability to produce data
regarding adverse effects, especially when these are rare and unexpected. In brief,
clinical trials recruit patients who are not representative of the persons who will use
the drug/s under study in clinical practice and expose these trial participants to the
study drug/s in a similarly ideal manner. Recruited adult patients are likely to be
healthier, younger, and not on multidrug regimens. On the other hand, elderly
6 E. Spina et al.
persons, children, and pregnant or lactating women and ethnic minorities are likely
to be underrepresented. Drug use in a clinical trial is necessarily controlled, with the
result that there is likely to be high drug adherence and drug use at optimal doses. A
consequence of all of this is that the safety profile of a drug in the clinical setting
where patients are much more heterogeneous and drug use may involve low adher-
ence, incorrect dose and/or administration and drug interactions is not known.
Furthermore, clinical trials are typically of short duration, whereas psychiatric
drugs are often used for long periods of time. Based on these considerations, at the
time of marketing of a drug, the knowledge of its tolerability is inevitably incom-
plete. The more complete safety profile of a newly marketed drug must therefore be
discovered over time as it is used in clinical practice and is dependent on pharmaco-
vigilance activities that identify, report, and describe ADRs as they occur.
For all these reasons, it is important that psychiatrists become familiar with the
concepts and methods of pharmacovigilance as they have a key role in identifying
and reporting new or serious adverse drug effects. The first step in this process is the
identification of potential ADRs. ADR detection can take place through various
methods, all of which start with signal detection or generation. The gold standard of
signal generation is spontaneous reporting, with reports being submitted to a central
authority; however, signals can also be generated from pharmaceutical company
data or from hospital or academic center data. Spontaneous reporting has several
advantages, for example, it is cost-effective and quick. On the other hand, it is
known that this approach is associated with significant underreporting. Signals
derived from spontaneous reports are not necessarily authentic and signal strength-
ening is required to confirm the causal link between drug and adverse effect.
Causality is a central issue in identifying authentic signals in spontaneous reports.
The assigning of causality by clinicians is guided by general considerations such as
whether the putative cause (i.e., the drug) precedes the effect, whether a greater
exposure results in a more severe reaction, whether the ADR disappears on discon-
tinuing the drug and reappears on rechallenge, whether there is a known similar
effect caused by a similar drug, and whether there is a biologically plausible mecha-
nism for the ADRs. Such considerations undoubtedly have their limitations, particu-
larly in the case of previously unknown ADRs and/or newly marketed drugs. Once
a causal link between a drug and an ADR is confirmed and the use of a drug is dif-
fuse, different methodological approaches, such as population-based studies, can be
used to evaluate the frequency as well as the risk of an ADR within a population.
Such studies provide information on different aspects of causality, such as the
strength of an association between a drug and an ADR as well as whether the occur-
rence of an ADR after drug exposure is consistent across different populations.
Population-based studies also have a role in the drug regulatory sector and are often
the way in which pharmaceutical companies satisfy legislation such as EU guidance
EMA/813938/2011 Revision 1, in which the European Medicines Agency (EMA)
made the request for post-authorization safety studies (PASS) legally binding. In
this way, carrying out pharmacovigilance activities is given a regulatory-legal
framework and can be enforced. PASS studies are an example of an increasing
awareness and commitment of regulatory agencies towards understanding the risks
associated with drug use as fully as possible.
1 Pharmacovigilance in Psychiatry: An Introduction 7
The constant developments in pharmacovigilance and psychiatry led to the

development of this book, which aims to describe the most important themes in
pharmacovigilance within the field of psychiatry. The first section gives a broad
overview of the topic, including important definitions and ADR pathogenesis and
methods and data sources used for pharmacovigilance activities. The second section
focuses on specific psychotropic drug class, namely, antidepressants, antipsychot-
ics, anxiolytics and sedatives, mood stabilizers, and medications for attention deficit
hyperactivity disorder. The final section focuses further on psychotropic drug use in
special populations such as children and adolescents, the elderly, and pregnant or
breast-feeding women.
References
Berson A, Descatoire V, Sutton A, Fau D, Maulny B, Vadrot N, Feldmann G, Berthon B, Tordjmann

T, Pessayre D (2001) Toxicity of alpidem, a peripheral benzodiazepine receptor ligand, but not
zolpidem, in rat hepatocytes: role of mitochondrial permeability transition and metabolic
activation. J Pharmacol Exp Ther 299:793–800
Fagius J, Osterman PO, Sidén A, Wiholm BE (1985) Guillain-Barré syndrome following zimeldine
treatment. J Neurol Neurosurg Psychiatry 48:65–69
Glassman AH (2005) Schizophrenia, antipsychotic drugs, and cardiovascular disease. J Clin
Psychiatry 66(Suppl 6):5–10
Harmark L, van Groothest AC (2008) Pharmacovigilance: methods recent developments and
future perspectives. Eur J Pharmacol 64:743–752
Lazaros GA, Stavrinos C, Papatheodoridis GV, Delladetsima JK, Toliopoulos A, Tassopoulos NC
(1996) Amineptine induced liver injury. Report of two cases and brief review of the literature.
Hepatogastroenterology 43:1015–1019
McBride WG (1961) Thalidomide and congenital abnormalities. Lancet 2:1358
McKenna PJ, Bailey PE (1993) The strange story of clozapine. Br J Psychiatry 162:32–37
McNaughton R, Huet G, Shakir S (2014) An investigation into drug products withdrawn from the
EU market between 2002 and 2011 for safety reasons and the evidence used to support the
decision-making. BMJ Open 4:e004221
Moore N (2013) The past, present and perhaps future of pharmacovigilance: homage to Folke
Sjoqvist. Eur J Clin Pharmacol 69(Suppl 1):S33–S41
Moore TJ, Cohen MR, Furberg CD (2007) Serious adverse drug events reported to the Food and
Drug Administration, 1998–2005. Arch Intern Med 167:1752–1759
Nadal R (2001) Pharmacology of the atypical antipsychotic remoxipride, a dopamine D2 receptor
antagonist. CNS Drug Rev 7:265–282
No authors listed (1999) Welcome withdrawal of amineptine in France. Prescrire Int 8:51
Rajkumar RP, Melvin G (2014) Pharmacovigilance for psychiatrists: an introduction. Ind J
Psychiatry 56:176–181
Stonier PD (1992) Nomifensine and hemolytic anemia – experience of a post-marketing alert.
Pharmacoepidemiol Drug Saf 1:177–185
Stricker BH, Psaty BM (2004) Detection, verification, and quantification of adverse drug reac-
tions. BMJ 329:44–47
World Health Organization (WHO) (2000) Safety monitoring of medicinal products: guidelines
for setting up and running of a pharmacovigilance centre. World Health Organization, Geneva;
2000. Available from: http://apps.who.int/medicinedocs/en/d/Jh2934e. Accessed on 11 Dec
2011
Chapter 2
Adverse Drug Reactions: Definitions,
Classifications and Regulatory Aspects
Paola Maria Cutroneo and Giovanni Polimeni
Abstract Adverse drug reactions (ADRs) are a common and important cause of
morbidity and mortality that represent a major health problem worldwide, with high
social costs for communities. Several studies have shown that ADR-related hospital
admissions comprise up to 10 % of the total number of hospitalizations.
Owing to the well-known limitations of pre-marketing research, it is now gener-
ally accepted that part of the process of evaluating drug safety needs to take place in
the post-marketing (approval) phase. Thus, once approval is granted, it becomes
essential to detect and to evaluate unrecognized ADRs related to medicines for pro-
tecting the public health. This activity, known as post-marketing surveillance or
“pharmacovigilance,” can lead to the identification of important safety problems,
which may even result in the withdrawal of drugs from the market. The main goal
of pharmacovigilance is the early detection of new, rare, or serious ADRs and the
communication of these risks to the public.
ADRs occur by a number of mechanisms, some of which remain unclear. Besides
the intrinsic danger associated with the drug, patients might have a particular,
unpredictable hypersensitivity to certain drugs, which requires careful monitoring.
Furthermore, several risk factors are important in determining susceptibility to
ADRs. Knowledge and use of ADR classification systems can give the health pro-
fessional greater clarity about an ADR and in some cases suggest ways of managing
or avoiding a future event.
Keywords Adverse drug reactions • Pharmacovigilance • Adverse events • Safety

withdrawals
P.M. Cutroneo (*) • G. Polimeni

Department of Clinical and Experimental Medicine, University of Messina, Messina, Italy
e-mail: pmcutroneo@unime.it; gpolimeni@unime.it

DOI 10.1007/978-3-319-24741-0_2
10 P.M. Cutroneo and G. Polimeni
2.1 Adverse Drug Reactions: Terminologies and Definitions
An adverse drug reaction (ADR) is described by the World Health Organization

(WHO) as “a response to a drug which is noxious and unintended and which occurs
at doses normally used in man for the prophylaxis, diagnosis, or therapy of disease,
or for the modification of physiological function.” The term “response” in this con-
text means that a clear pharmacological link between the drug and the event is not
present, but a causal relationship between a medicinal product and an adverse event
is at least a reasonable possibility. Importantly, this definition underlines the fact
that the phenomenon is noxious (differentiating between adverse drug reaction and
side effects which can also be beneficial) and that it includes doses prescribed clini-
cally, excluding accidental or deliberate overdose.
According to the new European pharmacovigilance legislation, which came into
effect in July 2012, an adverse drug reaction is “a response to a medicinal product
which is noxious and unintended” (Directive 2010/84/EU). Compared to the defini-
tion from the WHO, this definition also covers ADRs resulting from doses other
than those normally used in therapy, including off-label use, overdose, misuse,
abuse, medication errors, and those caused by occupational exposure. Furthermore,
the term “medicinal products” implicitly encompasses also reactions caused by
products not classically included in the definition of “drugs,” such as herbal sub-
stances or homeopathic products.
An alternative definition of ADR, proposed by Edwards and Aronson (2000), is
the following: “an appreciably harmful or unpleasant reaction, resulting from an
intervention related to the use of a medicinal product, which predicts hazard from
future administration and warrants prevention or specific treatment, or alteration of
the dosage regimen, or withdrawal of the product.”
Adverse drug reactions are further classified for regulatory purposes according to
their seriousness and expectedness. The seriousness of an ADR is the extent to
which the reaction causes harm to the patient. A serious suspected adverse reaction
is defined as “an adverse reaction which results in death, is life-threatening, requires
inpatient hospitalization or prolongation of existing hospitalization, results in per-
sistent or significant disability or incapacity, or is a congenital anomaly/birth
defect.”
However, besides these specific situations, with regard to the seriousness of an
ADR, the medical individual opinion is also taken into account; in fact, as specified
in current international guidelines, “medical and scientific judgment should be exer-
cised in deciding whether other situations should be considered serious reactions,
such as important medical events that might not be immediately life threatening or
result in death or hospitalization but might jeopardize the patient or might require
intervention to prevent one of the other outcomes listed above.” (European Medicines
Agency and Heads of Medicines Agencies 2014 GVP-Annex I). Examples of such
events are allergic bronchospasm intensively treated at home, blood dyscrasias or
convulsions that do not result in hospitalization, or development of drug depen-
dency or drug abuse. In contrast to seriousness, which is based on the patient and
2 Adverse Drug Reactions: Definitions, Classifications and Regulatory Aspects 11
event outcome, the “severity” of an adverse reaction is often used to describe the
intensity of a medical event, as in the grading “mild,” “moderate,” and “severe.”
Thus, a severe reaction can be of relatively minor medical significance, such as a
severe headache.
With regard to the preventability of ADRs, an unexpected suspected adverse
reaction has been defined by the WHO as “an adverse reaction, the nature or sever-
ity of which is not consistent with domestic labeling or market authorization, or
expected from characteristics of the drug” (WHO 2002) or according to EMA, “an
adverse reaction, the nature, severity or outcome of which is not consistent with the
summary of product characteristics” (European Medicines Agency and Heads of
Medicines Agencies 2014 GVP-Annex I). Reports that add significant information
on specificity or severity of a known, already documented serious ADRs constitute
unexpected events (e.g., an event more specific or more severe than described in the
reference document would be considered as “unexpected”).
A side effect is “any unintended effect of a pharmaceutical product occurring at
doses normally used by a patient, which is related to the pharmacological properties
of the drug.” This definition was formulated to include side effects that, although are
not the main aim of the therapy, may be beneficial rather than harmful. For example,
a tricyclic antidepressant may incidentally also relieve symptoms of irritable bowel
syndrome in a depressed patient.
2.2 Social and Economic Impact of Adverse Drug Reactions
In spite of the apparent rarity of serious reactions to individual drugs, ADRs are
effectively a global epidemic, with large economic effects on healthcare. Several
epidemiological studies have been conducted in order to determine the frequency of
ADRs and the related healthcare costs in both in- and outpatient settings, mainly
focusing on drug-related hospital admission, prolongation of hospital stay, and
emergency department visits due to ADRs (Sultana et al. 2013).
Based on 37 studies mostly conducted in the USA, Taché et al. reviewed the
prevalence of adverse drug events in ambulatory care and reported that 5.1 % of
hospital admissions were due to ADRs (Taché et al. 2011).
Budnitz et al. reported the USA estimates of emergency department (ED) visits
for adverse drug events, derived from data of 58 hospitals participating in the
National Electronic Injury Surveillance System–Cooperative Adverse Drug Event
Surveillance (NEISS-CADES) System (Budnitz et al. 2006). Over a 2-year study
period, 21,298 adverse drug event cases were reported. The estimated annual popu-
lation rate of adverse drug events treated in EDs was 2.4 per 1000 individuals (95 %
CI, 1.7–3.0). Furthermore, adverse drug events led to hospitalization of 3487 indi-
viduals (annual estimate, 117,318 [16.7 %]; 95 % CI, 13.1–20.3 %).
Several studies focused on more vulnerable populations such as geriatric patients.
People aged 65 years or older are in fact more likely than younger individuals to
develop adverse drug events. In another study, Budnitz et al. estimated that 99,628
emergency hospitalizations (95 % CI 55,531–143,724) occurred annually for

adverse drug events in elderly patients. Nearly half of these hospitalizations involved
adults 80 years of age or older. Most hospitalizations were attributed to four medica-
tions or medication classes alone or in combination: warfarin (33.3 %), insulins
(13.9 %), oral antiplatelet agents (13.3 %), and oral hypoglycemic agents (10.7 %).
The majority of these cases resulted from unintentional overdoses (Budnitz et al.
2011).
The frequency of hospital admissions caused by ADRs in general adult popula-
tion has been estimated at about 6.5 % of all admissions to UK general medical
wards, with a mortality rate of 2 % (Pirmohamed et al. 2004). Interestingly, although
media attention often focuses on the safety of new drugs, the major causes of these
admissions to hospital were older-established drugs with relatively well-described
safety profiles, the most common being caused by NSAIDs, aspirin, warfarin, or
diuretics (Pirmohamed et al. 2004). In the same study, the projected economic costs
of ADR-related admissions to UK hospitals were estimated to be £466 million.
A recent systematic review reported the percentage of hospitalizations resulting
from medicine-related problems and identified a median of prevalence rates of hos-
pitalization due to ADRs corresponding to 7 % (interquartile range, 2.4–14.9 %) (Al
Hamid et al. 2014). Another review showed that ADRs account for 4.2–30 % of
hospital admissions in the USA and Canada, 5.7–18.8 % of admissions in Australia,
and 2.5–10.6 % of admissions in Europe (Howard et al. 2007). More than 6 % of
inpatients may also experience an adverse drug event (Krähenbühl-Melcher et al.
2007).
Preventable adverse drug reactions are a significant burden to healthcare. A
meta-analysis in 2012, including studies on outpatients with emergency visits or
hospital admissions and inpatients, showed that 2.0 % of outpatients (95 % CI, 1.2–
3.2 %) had preventable ADR and 52 % (95 % CI, 42–62 %) of detected ADR were
preventable. Among inpatients, 1.6 % (95 % CI, 0.1–51 %) had preventable ADR
and 45 % (95 % CI, 33–58 %) of ADR was preventable (Hakkarainen et al. 2012).
Apart from the medical impact, pharmacoeconomic studies on the costs of ADRs
suggest that governments pay considerable amounts from health budgets toward
covering costs associated with them. In most countries, the extent of this expendi-
ture has not been measured. It has been suggested that patients who developed
adverse effects during hospitalization were hospitalized in an average of 1.2–
3.8 days longer than patients who did not, with a substantial increase of the health-
care costs (Rodriguez-Monguio et al. 2003). Furthermore, up to 57 % of the
community-acquired ADRs are not being recognized by the attending physician
upon hospital admission, leading to inappropriate management of the adverse event,
exposure of the patient to additional hazards of the drug, and prolonged hospitaliza-
tion (Dormann et al. 2003).
In the European Union (EU), on the basis of a report published by the European
Commission in 2008, 197,000 deaths per year are caused by ADRs, which also
accounted for the fifth most common cause of hospital death. Furthermore, it has
been estimated that ADRs are responsible for approximately 5 % of all hospital
admissions, and almost 5 % of hospitalized patients will experience an ADR during
their hospital stay in the EU. Overall, the total cost to society of ADRs is estimated
to be €79 billion per year (European Commission 2008). Similarly, ADRs are listed
as one of the top ten causes of death in the USA, with more than 100,000 deaths
annually attributed to various ADRs (Riedl and Casillas 2003).
2.3 Pharmacovigilance for Evaluating Adverse Drug

Reactions
A medicinal product is authorized by competent authorities if the benefit–risk bal-

ance is judged to be positive for the target population at the time of authorization.
Before medicinal products are marketed, they are extensively tested in animals and
in clinical trials in humans, and sufficient evidence is required to show the new drug
to be of good quality, effective, and safe. Pre-approval studies include double-blind
randomized controlled trials (RCT) that are considered to be the most rigorous
approach to establish whether a cause–effect relationship exists between a treatment
and an outcome.
However, the design of pre-marketing clinical trials is not optimal to monitor the
safety of a drug. There are several issues that limit the generalizability of RCT
results to clinical practice. The relatively short duration of clinical trials and narrow
study population size makes it difficult to detect rare ADRs, or those with a long
latency. Furthermore, the selective recruitment of patients (with resulting exclusion
of special subgroups of patients or those receiving certain concurrent medicines)
and the consideration of few predefined ADRs will considerably limit the generaliz-
ability of results from the pre-approval phase. Consequently, once a product is mar-
keted, new information will be generated from clinical practice, which can have an
impact on the benefits or risks of the product.
These issues lead to uncertainties about the safety of a new drug once it is mar-
keted and used in a wider population, over longer periods of time, in patients with
comorbidities and concomitant medications and for off-label indications not previ-
ously evaluated. Therefore, the true picture of a product safety actually evolves over
the months and even years during the product’s lifetime in the marketplace.
In the course of medical history, there have been many examples of the potential
danger associated with medication use. During the last century, the most striking
was the thalidomide disaster emerged in 1961, with the sudden upsurge in the num-
ber of severe deformities (phocomelia or micromelia) occurred in thousands of
newborns whose mother had taken the drug during pregnancy. As a result of this
catastrophic epidemic, it has become evident that continuing, post-marketing drug
safety monitoring, often referred to as “pharmacovigilance,” is essential for protect-
ing the public health by assuring the security of human medicines and helping the
public and healthcare providers get the accurate science-based information that they
need to use medicines properly (Kavitha 2010).
Once approval is granted, drug regulatory authorities, such as the European
Medicines Agency (EMA) and the Food and Drug Administration (FDA) in the
USA, are responsible for post-marketing safety evaluation. This risk assessment
activity encompasses the entire period the drug is on the market and is mainly focused
on collecting and analyzing case reports of ADRs, distinguishing signals from back-
ground “noise,” making regulatory decisions based on strengthened signals, and
alerting prescribers, manufacturers, and the public to new risks of adverse reactions.
2.3.1 Pharmacovigilance: Definitions and Objectives
The World Health Organization has defined pharmacovigilance as “the science and
activities relating to the detection, assessment, understanding and prevention of
adverse effects or any other medicine-related problem” (World Health Organization
2002).
The main goal of pharmacovigilance is the early detection of new, rare, and seri-
ous ADRs. Other objectives encompass the identification of the increases of the
frequency of known adverse effects, the recognition of mechanisms or risk factors
relevant for the occurrence of ADRs, and the communication of information about
adverse effects to healthcare professionals and consumers (World Health
Organization 2002).
An important cornerstone in further clarifying the risk profile of a medical prod-
uct is the detection of “safety signals” in the post-marketing phase. There are several
different definitions of “signal” in pharmacovigilance. According to WHO, a signal
is defined as a “reported information on a possible causal relationship between an
adverse event and a drug, the relationship being unknown or incompletely docu-
mented previously” (World Health Organization 2002). Usually more than a single
report is required to generate a signal, depending upon the seriousness of the event
and the quality of the information. On the basis of the Council for International
Organizations of Medical Sciences (CIOMS) definition (2010), a signal is “an
information arising from one or multiple sources, including observations and exper-
iments, which suggests a new potentially causal association, or a new aspect of a
known association between an intervention and an event or set of related events,
either adverse or beneficial, that is judged to be of sufficient likelihood to justify
verificatory action.” For the purpose of monitoring data in pharmacovigilance, only
signals related to an adverse reaction are generally considered.
Specific sources for pharmacovigilance data include spontaneous ADR reporting
systems, active surveillance systems, post-authorization non-interventional studies,
clinical trials, and other sources of information.
A spontaneous reporting system (SRS) relies primarily on unsolicited reports by
healthcare professionals or consumers to established national or regional pharmaco-
vigilance centers or alternatively to marketing authorization holders (MAHs) that
describe one or more suspected adverse reactions in a patient who was given one or
more medicinal products (European Medicines Agency and Heads of Medicines
Agencies, GVP – Module VI 2014).
Safety signals may be detected from monitoring of SRS databases collecting
spontaneous ADR reports or from review of information provided by MAHs in the
context of regulatory procedures.
However, the presence of a safety signal does not actually mean that a medicine
has caused the reported adverse event. The adverse event could be a symptom of
another illness or caused by other medicines taken by the patient. Thus, a deep
evaluation of safety signals is required to establish the likelihood of a causal rela-
tion between the suspected medicine and the reported adverse event, and a system-
atic process of verification and quantification should follow the detection of signals
of a potential adverse event. Any signal should be validated taking into account all
other relevant sources of information, such as aggregated data from active surveil-
lance systems or pharmacoepidemiological studies and medical literature.
Despite the vast underreporting of adverse effects, spontaneous reporting remains
the most frequent source of safety signals in pharmacovigilance (Ishiguro et al.
2012; Pacurariu et al. 2014). Nevertheless, such systems cannot reliably quantify
incidence rates, confirm causality, understand risk factors, or elucidate patterns of
use. For hypothesis testing and quantification of risks, observational studies have
proved useful.
2.3.2 Examples of Safety Regulatory Actions and Drug

Withdrawals for Safety Reasons
As discussed above, the prescribing of medications in the “real world” often yields
safety concerns related to new drugs or to medicines on the market for many years,
probably due to a change in the patterns of their use or to a better implementation of
safety monitoring (Pacurariu et al. 2014).
Once an adverse reaction is detected, appropriate measures will be taken, and
regulatory authorities have responsibility to safeguard public health in such a
situation.
Safety concerns referred back to the regulatory authorities may lead to revision
of drug labeling (i.e., restriction of an indication, new contraindication, change in
the recommended dose, major warnings, or precautions for use), addition of a black
box warning (available only in the USA), a direct communication to healthcare
professionals, or the suspension/withdrawal from the market.
The regulatory authorities may withdraw the approval of an application with
respect to any drug for reasons of safety or efficacy, taking account of the serious-
ness of the condition and the range of optional treatments available. Safety issues
are largely responsible for decisions to remove pharmaceutical products from the
market. Nineteen drugs were discontinued, throughout the EU, for safety reasons
from 2002 to 2011. The main therapeutic categories represented among banned
pharmaceutical products were “nervous system,” “musculoskeletal system,” and
nonsteroidal anti-inflammatory drugs. Cardiovascular adverse reactions were the
main reason for withdrawal, followed by hepatic disorders and neuropsychiatric
conditions (McNaughton et al. 2014).
In the USA from 1980 to 2009, safety reasons accounted for 26 drug discontinu-
ations (3.5 % of the drugs approved in the study period). Severe cardiovascular
effects and hepatic toxicity were the major problems that caused the withdrawal of
these products (Qureshi et al. 2011).
A total of 22 pharmaceuticals were removed from the market in France between
2005 and 2011. The number of drug discontinuations increased during this period
(Paludetto et al. 2012).
Several nervous system drugs were withdrawn for safety reasons (Table 2.1)
(Aronson 2012; Jones and Kingery 2014; McNaughton et al. 2014). For instance,
pemoline, a CNS stimulant approved in 1975 for treatment of Attention-Deficit/
Hyperactivity Disorder (ADHD), was removed from the US market because of fatal
hepatotoxicity in 2005.
Rimonabant is a selective CB1 endocannabinoid receptor antagonist indicated
for the treatment of obesity. Although available in Europe since 2006 for use as an
adjunct to diet and exercise for obese or overweight patients with associated risk
factors, rimonabant failed to secure FDA approval in the USA. Concerns had been
growing that patients taking rimonabant were at increased risk of psychiatric adverse
events, including suicidality. In October 2008, following a review of post-marketing
data, the EMEA recommended suspension of the drug’s marketing authorization on
safety grounds. Data had shown a doubling of the risk of psychiatric disorders in
patients taking rimonabant in comparison with placebo.
Another drug that has been banned in the USA is pergolide, an ergot-derived
dopamine receptor agonist used for the treatment of Parkinson’s disease. In 2007, it
was discontinued due to increased rates of cardiac valve regurgitation, along with
pleuropulmonary or retroperitoneal fibrosis related to its use.
The suspension of the marketing authorizations of tetrazepam-containing medi-
cines across the European Union in 2013 is derived from the evidence of a low but
Table 2.1 Examples of psychotropic drugs removed from the market for safety reasons, 1980–
2013 (Aronson 2012; Jones and Kingery 2014; McNaughton et al. 2014)
Year withdrawn Drug (generic name) Reason withdrawn
1982 Clomacron Hepatotoxicity (UK)
1983 Zimeldine Hypersensitivity, Guillain-Barre
syndrome (worldwide)
1986 Nomifensine Hemolytic anemia (worldwide)
1991 Triazolam Depression, amnesia (UK, France,
and other countries)
1994 Remoxipride Aplastic anemia (worldwide)
1995 Alpidem Hepatotoxicity (worldwide)
1999 Amineptine Hepatotoxicity, abuse (France and
other countries)
2005 Thioridazine Cardiotoxicity (UK and other
countries)
2005 Pemoline Hepatotoxicity (USA)
2011 Aceprometazine/acepromazine/ Cumulative adverse effects, misuse,
clorazepate fatal side effects (EU)
2013 Tetrazepam Serious cutaneous reactions (EU)
increased risk of serious skin reactions of tetrazepam compared with other benzodi-
azepines (European Medicines Agency 2013).
In addition, psychotropic medications represent a group of drugs for which sev-
eral boxed warnings or other product labeling changes were performed by all the
regulatory agencies (Table 2.2). For instance, the US FDA issued a series of adviso-
ries culminating in a black box warning for all antidepressants for patients under
age 18 (Food and Drug Administration 2004), following a consistent finding of an
Table 2.2 Examples of safety warnings issued by the drug regulatory agencies concerning
psychotropic drugs (Clavenna and Bonati 2009; Foy et al. 2014)
Medicine Adverse events Actions taken
ADHD drugs Cardiovascular adverse events, Black box warning, medication
neuropsychiatric symptoms guides, contraindications in at-risk
patients
Agomelatine Hepatotoxicity Monitoring of liver function, warning
in patients aged 75 years or over,
medication guide for patients
Antidepressants Increased risk of suicidality in Boxed warnings or other product
children, adolescents, and young labeling changes
adults
Antiepileptics Adverse effects on the bone Vitamin D supplementation for
at-risk patients
Antipsychotics Venous tromboembolic events SpC updated
Atomoxetine Increased risk of suicidal thinking Boxed warning, SpC updated
in children and adolescents
Buproprion Seizures Improved warnings and revised
dosing instructions
Citalopram/ QT interval prolongation Daily dose restrictions,
escitalopram contraindications
Codeine Risk of severe ADRs for infants SpC updated, warnings
with ultra-rapid metabolizer
breastfeeding mothers
Respiratory depression Restrictions on the use in children
contraindicated in at-risk patients,
boxed warning
Lamotrigine Increased risk of potentially fatal Warnings, SpC updated
rash, particularly in children
Topiramate Oligohydrosis, hyperthermia SpC updated
Varenicline Depression, increased risk of Boxed warnings, SpC updated
suicidal thinking and behavior
Ziprasidone Drug reaction with eosinophilia and SpC updated
systemic symptoms (DRESS)
Zoplicone Risk of next-day impairment Dosage recommendations, SpC
updated
Zonisamide Oligohydrosis and hyperthermia in SpC updated, monitoring for
pediatric patients evidence of decreased sweating and
increased body temperature
SpC summary of product characteristics
increased risk of suicidality in children and adolescents. Moreover, epidemiologic

findings led to an update to the 2004 boxed warnings to include information about
an increased risk of suicidality in young adults ages 18–24 during initial treatment
with all antidepressants in 2007 (Food and Drug Administration 2007). The EMA
issued analogous warnings for SSRIs and, additionally, contraindicated prescrip-
tions of SSRIs in youths (European Medicines Agency 2005).
Similarly, in 2005 FDA and Health Canada warned of increased suicidality in
children and adolescents being treated with atomoxetine for ADHD. Furthermore,
all ADHD medications were involved in warnings about their cardiac risks
(Clavenna and Bonati 2009).
Post-marketing safety regulatory actions frequently concern liver injury related
to drug treatment. Hepatotoxic reactions, even severe forms with fatal outcomes,
have been reported for many antidepressants, like nefazodone, which was sus-
pended from the US market in 2003 (Choi 2003), or amineptine, which was with-
drawn from the market in France in 1999 due to its potential of abuse (Prescrire
Editorial Staff 1999).
More recently, liver problems have been reported in patients taking agomelatine,
a novel antidepressant, first approved in 2009 by the EMA for major depression in
adults. The attention to its hepatic side effect profile has gradually increased, lead-
ing the EMA to recommend in 2014 further measures aimed to minimize the risk of
liver toxicity, such as monitoring of liver function during treatment and contraindi-
cation in patients aged 75 years or above, since they might be at an increased risk of
severe hepatic effects (European Medicines Agency 2014).
Safety issues have also appeared recently for psychotropic drugs that have been
on the market for more than 50 years (e.g., thiopental, codeine). An example is the
signal concerning codeine and life-threatening toxicity (in particular, respiratory
depression) in specific subgroups of patients at special risk of such side effects, such
as cytochrome P450 2D6 ultra-rapid metabolizers or children below 12 years
(European Medicines Agency 2013). This safety issue, due to a more striking con-
version of codeine into morphine in the body of at-risk patients, resulted in several
restrictions or contraindications being introduced in order to minimize the risk of
serious side effects.
Among recent safety communications related to nervous system drugs, an exam-
ple is the advisory issued in 2014 by Health Canada to healthcare professionals to
inform about the risks of next-day impairment after exposure to the hypnotic zopi-
clone and ways to minimize it, including new dosing recommendations (Health
Canada 2014). A recent warning issued by FDA in 2014 concerns the association
between the use of ziprasidone and the onset of drug rash with eosinophilia and
systemic symptom (DRESS) syndrome (Food and Drug Administration 2014).
Finally, several alerts released by regulatory agencies in 2011 on the potential
cardiac toxicity (dose-dependent QTc prolongation) of citalopram and escitalopram
resulted in daily dose restrictions (including in elderly patients) and contraindica-
tions of use of these drugs (Food and Drug Administration 2011; Italian Medicines
Agency 2011).
2.4 Classification of ADRs
Adverse drug reactions can be difficult and sometimes impossible to distinguish

from the patient’s disease as they act through the same physiological and pathologi-
cal pathways. Moreover, some distinctive and specific physical signs can be consid-
ered with a high probability drug-related (e.g., extrapyramidal disorders or
Stevens-Johnson syndrome). Because adverse drug reactions can mimic or precipi-
tate different pathological conditions, if the physician does not consider medica-
tions as a potential cause of the patient’s symptoms, additional drug therapy may be
prescribed to treat the adverse effect of the original drug, causing what is called a
“prescribing cascade.” The use of metoclopramide may induce parkinsonism, which
will be treated with levodopa. Drug-induced cognitive impairment is among the
most common causes of reversible dementia (e.g., narcotics, antihistamines). Falls
can be precipitated by a wide variety of drugs (e.g., psychotropics, antihyperten-
sives), and the anticholinergic effect of many drugs (e.g., amitriptyline, oxybutynin)
can result in dry mouth, constipation, urinary retention, blurred vision, and
confusion.
Knowledge and use of ADR classification systems can give the health profes-
sional greater clarity about an ADR and suggest ways of managing or avoiding a
future event.
Adverse drug reactions can be classified in various ways: immunologic or non-
immunologic, predictable or unpredictable, and common or rare.
The most common classification, proposed by Rawlings and Thompson (1977),
divides ADRs into type A and type B reactions on the basis of the mechanism of
action.
2.4.1 Type A Adverse Event
Type A reactions are due to an exaggerated, but otherwise normal, pharmacologi-

cal action of a drug (A indicates augmented) given in the usual therapeutic doses.
They are therefore largely predictable on the basis of the drug’s known pharmaco-
logic action (Riedl and Casillas 2003) and usually reversible on either adjusting
the dose or withdrawing the drug. Examples of a type A reaction include
antipsychotic-induced parkinsonism (a known and predictable side effect caused
by the block of dopamine receptors), or daytime somnolence after a sedative-hyp-
notic taken for sleep. Furthermore, these reactions are expected to possibly occur
in a certain percentage of individuals based on current scientific evidence. More
than 80 % of all occurring ADRs are type A reactions, which include toxic effects
(such as digoxin toxicity and serotonin syndrome caused by selective serotonin
reuptake inhibitors), side effects, secondary effects (e.g., antibiotic-associated
diarrhea), and drug interactions (e.g., lithium toxicity due to NSAID-induced inhi-
bition of its excretion).
The effect of type A ADRs is generally less severe than type B events and is dose
related; for example, some degree of anticholinergic symptoms can be observed in
nearly everyone taking tricyclic antidepressants, provided the dose is large enough.
However, they are not necessarily caused by overdosage but can be also seen after a
normal dose is administered in a susceptible subject (e.g., constipation due to mor-
phine or gastrointestinal irritation with nonsteroidal anti-inflammatory drugs).
Generally, type A reactions can be reproduced and studied experimentally and
are often already identified during the clinical trials done before marketing.
2.4.2 Type B Adverse Event
Unlike type A reactions, type B ADRs cannot be explained based on the pharmaco-
logic actions of the offending agent, are not dose related in most patients, and may
develop quite unpredictably in susceptible individuals (the B indicates bizarre)
(Riedl and Casillas 2003; Pillans 2008). They are generally serious and notoriously
difficult to study. The majority of type B reactions can occur in predisposed patients
as a result of an immune-mediated mechanism (allergic or hypersensitivity reac-
tions, pseudoallergy or anaphylactoid reaction), where the drug acts as an antigen
or allergen. Less frequently, type B ADRs may occur with a mechanism not yet
understood (idiosyncratic reactions), generally due to a genetic or acquired enzyme
abnormality with the formation of toxic metabolites. Neuroleptic malignant syn-
drome, hyperthermia of anesthesia, and tardive dyskinesia caused by neuroleptic
drugs fall into this category.
ADRs classified as type B reactions are usually quite rare, representing about
10–15 % of all ADRs but are the more troublesome than type A. Differently from
type A reactions, in patients with a type B ADR, it is usually necessary to withdraw
therapy. They are not identified in pre-marketing trials and are only exhibited when
the drug has been on the market for some time and used in a wide variety of patient
populations. These side effects came to light after further studies were conducted or
various case reports were supplied via post-marketing surveillance documenting the
adverse events.
Type B drug allergies can be further subdivided in several ways and can induce
severe responses that are deadly to susceptible individuals.
• Type I allergic reactions are classified as immunoglobulin E (IgE) mediated. An
example of this type of allergy is anaphylaxis induced by beta-lactam antibiotics
(penicillin allergy).
• Type II reactions are cytotoxic. An example of this is a specific type of thrombo-
cytopenia induced by heparin, which can be quite severe.
• Type III reactions are categorized as immune complexes and occur when anti-
gens and antibodies (immunoglobulin G [IgG] or immunoglobulin M [IgM])
accumulate in the body in equal amounts, causing extensive cross-linking. An
example of this reaction is hydralazine-induced systemic lupus erythematosus
(SLE).
• Type IV reactions are described as delayed or hypersensitivity reactions. These

generally take 2–3 days to develop and are not described as antibody mediated
but instead are induced by a cell-mediated response. Contact dermatitis caused
by an offending skin product is an excellent example of this type of reaction.
There are other groups in this system of classification, but these may also be
considered as subclasses or hybrids of type A and B ADRs. These are type C ADRs
(chronic reactions, dose and time related), type D (delayed reactions, time related),
type E (end of use reactions), and type F (failure of therapy) (Edwards and Aronson
2000; Meyboom et al. 2000; Rehan et al. 2009).
The characteristics, some examples, and the management of these ADRs are
listed in Table 2.3. Another ADR classification scheme is based on frequency of the
occurrence of a particular drug-induced reaction. Based on pre-approval studies and
post-marketing reports, the frequency of a particular ADR can also be hypothesized.
ADR frequency categories for adverse events are generally defined as very common
(≥1/10), common (≥1/100 but <1/10), uncommon (≥1/1000 but <1/100), rare
Table 2.3 Classification of adverse drug reactions

Examples relevant to
Type of reaction Mnemonic Features psychiatry
A: Dose related Augmented Common Serotonin syndrome with
Related to pharmacological SSRIs
action of the drug Anticholinergic effects of
Predictable tricyclic antidepressants
Low mortality Phenytoin toxicity
B: Nondose Bizarre Uncommon Anticonvulsant syndrome
related Not related to Clozapine-induced
pharmacological action of agranulocytosis
the drug
Unpredictable
High mortality
C: Dose related Chronic Uncommon Propofol infusion syndrome
and time related Related to the cumulative Malignant hyperthermia with
dose halothane
D: Time related Delayed Uncommon Tardive dyskinesia with
Usually dose related chlorpromazine
Occurs or becomes
apparent some time after
the use of the drug
E: Withdrawal End of use Uncommon Opiate withdrawal syndrome
Occurs soon after the
withdrawal of the drug
F: Unexpected Failure Common Unwanted pregnancies
failure of therapy Dose related following interaction
Often caused by drug between oral contraceptives
interactions and St. John’s wort
(hypericum)
Adapted from Edwards and Aronson (2000)
SSRIs serotonin selective reuptake inhibitors
(≥1/10,000 but <1/1000), and very rare (<1/10,000). These basic definitions are
utilized to describe and classify adverse drug events. These data attempt to quantify
adverse drug reactions to some degree and may simply refer to the general popula-
tion. However, a particular ADR may be common in specific groups of patients.
Stratifying those patients in whom ADRs may be more prevalent is important, as
some ADRs are restricted to a small segment of the population at greatest risk.
ADRs are usually more prevalent in the elderly, because of potential interactions
with other agents, as medication use is much more prevalent in the elderly than in
the general population. Geriatric patients might also be at greater risk for develop-
ing a Type A ADR because of the potential for reduced hepatic or renal metabolism
due to age, which results in higher drug concentrations. In some cases, age-related
changes may involve not only pharmacokinetics but also pharmacodynamics, with
significant clinical consequences. An excellent example of pharmacodynamic
changes in the older adult has been demonstrated with benzodiazepines. Older
adults have more sedation and lower performance than younger persons at the same
plasma concentration (Reidenberg et al. 1978).
A more recent classification (Table 2.4) accounts for the dose relatedness, time
course, and susceptibility of the patient (DoTS) to a reaction, and this classification
is increasingly used (Aronson and Ferner 2003). Malignant hyperpyrexia, for exam-
ple, occurs at any dose in susceptible individuals (Do) and occurs on first dose (T),
Table 2.4 DoTS (dose relatedness, time course, and susceptibility) classification of adverse drug
reactions
Classification Subclassification Explanation/further classification
Dose Toxic Reactions occurring at supratherapeutic doses
Collateral Reactions occurring at therapeutic doses
Hypersusceptibility Reactions occurring at subtherapeutic doses in
susceptible individuals
Time course Time independent Occur at any time during therapy
Time dependent Occur:
Due to rapid administration
After the first dose of a medication but not always
after subsequent doses
Early reactions that resolve (i.e., due to tolerance)
Intermediate reactions after some time (i.e.,
nonallergic hypersensitivity reactions)
Late reactions (incidence increases with longer
duration of therapy)
Delayed reactions much later after administration,
even after cessation of the drug (i.e.,
carcinogenesis)
Susceptibility Genetic Single or multiple factors associated with risk of an
Age adverse drug reaction
Gender
Physiological variation
Exogenous factors
Diseases
Adapted from Aronson and Ferner (2003)
and individual susceptibility factors are an inherited mutation for the ryanodine
receptor (S). The advantage of this system consists in the fact that an ADR can be
profiled in such a way that implications for its management may be more obvious.
For example, the dystonic reactions to metoclopramide can be characterized as a
collateral time-dependent reaction, with both sex and age acting as susceptibilities.
Future management would therefore focus on avoiding the use of the drug in sus-
ceptible groups such as children and young women.
It is important to realize, however, that it is not always possible to classify an
adverse drug reaction into one of these categories; furthermore, in any one individ-
ual, more than one mechanism may be responsible for any particular adverse effect.
References
Al Hamid A, Ghaleb M, Aljadhey H et al (2014) A systematic review of hospitalization resulting

from medicine-related problems in adult patients. Br J Clin Pharmacol 78:202–217
Aronson JK (2012) Adverse drug reactions: history, terminology, classification, causality, fre-
quency, preventability. In: Talbot J, Aronson JK (eds) Stephens’ detection and evaluation of
adverse drug reactions, 6th edn. Wiley-Blackwell, Chichester (UK), pp 1–119
Aronson JK, Ferner RE (2003) Joining the DoTS: new approach to classifying adverse drug reac-
tions. BMJ 327:1222–1225
Budnitz DS, Pollock DA, Weidenbach KN et al (2006) National surveillance of emergency depart-
ment visits for outpatient adverse drug events. JAMA 296:1858–1866
Budnitz DS, Lovegrove MC, Shehab N et al (2011) Emergency hospitalizations for adverse drug
events in older Americans. N Engl J Med 365:2002–2012
Choi S (2003) Nefazodone (Serzone) withdrawn because of hepatotoxicity. CMAJ 169(11):1187
Clavenna A, Bonati M (2009) Adverse drug reactions in childhood: a review of prospective studies
and safety alerts. Arch Dis Child 94:724–728
Council for International Organizations of Medical Sciences (CIOMS) (2010) Practical aspects of
signal detection in pharmacovigilance: report of CIOMS Working Group VIII. CIOMS, Geneva
Directive 2010/84/EU of the European Parliament and of the Council of 15 December 2010
amending, as regards pharmacovigilance, Directive 2001/83/EC on the Community code relat-
ing to medicinal products for human use. Official Journal of the Eur Union, 31 Dec 2010
Dormann H, Criegee-Rieck M, Neubert A et al (2003) Lack of awareness of community-acquired
adverse drug reactions upon hospital admission. Drug Saf 26:353–362
Edwards IR, Aronson JK (2000) Adverse drug reactions: definitions, diagnosis, and management.
Lancet 356:1255–1259
European Commission (2008) Proposal for a regulation amending, as regards pharmacovigi-
lance of medicinal products for human use. Regulation (EC) No 726/2004. Impact
assessment. Available at: http://ec.europa.eu/health/files/pharmacos/pharmpack_12_2008/
pharmacovigilance-ia-vol1_en.pdf
European Medicines Agency (2005) European medicines agency finalises review of antidepres-
sants in children and adolescents. EMEA/CHMP/128918/2005, Press release April 25, 2005.
Available at: http://www.ema.europa.eu/docs/en_GB/document_library/Referrals_document/
European Medicines Agency (2013) Recommendation to suspend tetrazepam-containing medi-
cines endorsed by CMDh, EMA/256383/2013, Press release, April 29 2013. Available at:
http://www.ema.europa.eu
European Medicines Agency (2013) Restrictions on use of codeine for pain relief in children –
CMDh endorses PRAC recommendation. EMA/385716/2013, Press release, June 28 2013.
Available at: http://www.ema.europa.eu
European Medicines Agency (2014) EMA confirms positive benefit-risk for antidepressant
Valdoxan/Thymanax (agomelatine). EMA/577812/2014, Press release September 26, 2014.
Available at: http://www.ema.europa.eu
European Medicines Agency and Heads of Medicines Agencies (2014) Guideline on good phar-
macovigilance practices (GVP) – Annex I (Rev 3) EMA/876333/2011 Rev 3. Available at:
www.ema.europa.eu
European Medicines Agency and Heads of Medicines Agencies (2014) Guideline on good phar-
macovigilance practices (GVP) – Module VI. EMA/873138/2011. Available at: www.ema.
europa.eu
FDA (2004) FDA launches a multi-pronged strategy to strengthen safeguards for children treated
with antidepressant medications. FDA News Release October 15, 2004. Available at: http://
www.fda.gov/NewsEvents/Newsroom/PressAnnouncements/
FDA (2007) FDA proposes new warnings about suicidal thinking, behavior in young adults who
take antidepressant medications. News Release May 2, 2007. Available at: http://www.fda.gov/
NewsEvents/Newsroom/PressAnnouncements/
FDA (2011) FDA drug safety communication: abnormal heart rhythms associated with high doses
of Celexa (citalopram hydrobromide). Safety Announcement August 24, 2011. Available at:
http://www.fda.gov/Drugs/DrugSafety
FDA (2014) FDA drug safety communication: FDA reporting mental health drug ziprasidone
(Geodon) associated with rare but potentially fatal skin reactions. Safety Announcement,
November 12, 2014. Available at: http://www.fda.gov/Drugs/DrugSafety
Foy M, Barrow P, Raine JM (2014) Spontaneous reporting: United Kingdom. In: Andrews EB,
Moore N (eds) Mann’s pharmacovigilance, 3rd edn. Wiley-Blackwell, Chichester (UK),
pp 185–201
Hakkarainen KM, Hedna K, Petzold M et al (2012) Percentage of patients with preventable adverse
drug reactions and preventability of adverse drug reactions – a meta-analysis. PLoS One
7(3):e33236
Health Canada (2014) IMOVANE (zopiclone) – new dosage recommendations to minimize the
risk of next-day impairment – for health professionals. RA-42253, Recalls and alerts, November
19, 2014. Available at: http://healthycanadians.gc.ca
Howard RL, Avery AJ, Slavenburg S et al (2007) Which drugs cause preventable admissions to
hospital? A systematic review. Br J Clin Pharmacol 63:136–147
Ishiguro C, Hall M, Neyarapally GA et al (2012) Post-market drug safety evidence sources: an
analysis of FDA drug safety communications. Pharmacoepidemiol Drug Saf 21:1134–1136
Italian Medicines Agency (2011) Associazione tra Seropram®/Elopram® (Citalopram) e prolun-
gamento dose dipendente dell’intervallo QT. Nota Informativa Importante October 2011.
Available at: http://www.agenziafarmaco.gov.it
Jones JK, Kingery E (2014) History of pharmacovigilance. In: Andrews EB, Moore N (eds)
Mann’s pharmacovigilance, 3rd edn. Wiley-Blackwell, Chichester (UK), pp 11–24
Kavitha D (2010) Adverse drug reaction (ADR) monitoring and pharmacovigilance. J Pharm Res
Health Care 2:127–134
Krähenbühl-Melcher A, Schlienger R, Lampert M et al (2007) Drug-related problems in hospitals:
a review of the recent literature. Drug Saf 30:379–407
McNaughton R, Huet G, Shakir S (2014) An investigation into drug products withdrawn from the
EU market between 2002 and 2011 for safety reasons and the evidence used to support the
decision-making. BMJ Open 4:e004221
Meyboom RH, Lindquist M, Egberts AC (2000) An ABC of drug-related problems. Drug Saf
22:415–423
Pacurariu AC, Coloma PM, van Haren A et al (2014) A description of signals during the first 18
months of the EMA pharmacovigilance risk assessment committee. Drug Saf 37:1059–1066
Paludetto MN, Olivier-Abbal P, Montastruc JL (2012) Is spontaneous reporting always the most
important information supporting drug withdrawals for pharmacovigilance reasons in France?
Pharmacoepidemiol Drug Saf 21:1289–1294
Pillans PI (2008) Clinical perspectives in drug safety and adverse drug reactions. Expert Rev Clin
Pharm 1:695–705
Pirmohamed M, James S, Meakin S et al (2004) Adverse drug reactions as cause of admission to
hospital: prospective analysis of 18 820 patients. BMJ 329:15–19
Prescrire Editorial Staff (1999) Welcome withdrawal of amineptine in France. Prescrire Int
8(40):51
Qureshi ZP, Seoane-Vazquez E, Rodriguez-Monguio R et al (2011) Market withdrawal of new
molecular entities approved in the United States from 1980 to 2009. Pharmacoepidemiol Drug
Saf 20:772–777
Rawlings TJP (1977) Pathogenesis of adverse drug reactions. In: Davies DM (ed) Textbook of
adverse drug reactions. Oxford University Press, Oxford, p 44
Rehan HS, Chopra D, Kakkar AK (2009) Physician’s guide to pharmacovigilance: terminology
and causality assessment. Eur J Intern Med 20:3–8
Reidenberg MM, Levy M, Warner H et al (1978) Relationship between diazepam dose, plasma
level, age, and central nervous system depression. Clin Pharmacol Ther 23:371–374
Riedl MA, Casillas AM (2003) Adverse drug reactions: types and treatment options. Am Fam
Physician 68:1781–1790
Rodriguez-Monguio R, Otero MJ, Rovira J (2003) Assessing the economic impact of adverse drug
effects. Pharmacoeconomics 21:623–650
Sultana J, Cutroneo P, Trifirò G (2013) Clinical and economic burden of adverse drug reactions. J
Pharmacol Pharmacother 4(suppl 1):s73–s77
Taché SV, Sönnichsen A, Ashcroft DM (2011) Prevalence of adverse drug events in ambulatory
care: a systematic review. Ann Pharmacother 45(7–8):977–989
World Health Organization (2002) The importance of pharmacovigilance: safety monitoring of
medicinal products. World Health Organization, Genève, pp 40–43, Available at: http://www.
who.int/
Chapter 3
Methods for the Post-Marketing Monitoring
of Psychotropics Safety: Interests and Pitfalls
Francesco Salvo, Annie Fourrier-Réglat, Nicholas Moore, Bernard Bégaud,

and Antoine Pariente
Abstract Psychotropics are among the most widely used drugs, which makes their
safety monitoring absolutely crucial. This monitoring is also especially important
given the frequency and potential seriousness of adverse effects related to psycho-
tropic drug use. The impact of benzodiazepine use, for instance, has been estimated
to more than 20,000 serious falls per year in French elderly, around 2000 of these
being fatal. None of the antidepressants existing is considered completely safe. If
serotonin reuptake inhibitors have extremely low cardiac toxicity compared to tricy-
clic antidepressants, they can still be incriminated for QT prolongation and have
been shown to significantly increase the risk of bleeding. Finally antipsychotics are
associated with numerous serious side effects that only appear acceptable with
regard to the seriousness of their indication. However, drug monitoring is not an end
in itself; it has to be associated to a specific goal. Apart from safety signal detection
and risk confirmation, the usual objectives of pharmacovigilance and pharmacoepi-
demiology, it needs to help clinicians to choose which therapeutic they should con-
sider within those presenting a benefit in a given indication. The tools that can be
used for the post-marketing monitoring of psychotropics safety include (i) pharma-
covigilance and individual case analysis, (ii) pharmacoepidemiology and popula-
tion data analysis and (iii) meta-analysis and clinical trial data analysis. In the
F. Salvo
Department of Pharmacology, INSERM U657, University of Bordeaux,
Bordeaux, 33076, France
A. Fourrier-Réglat (*) • N. Moore • B. Bégaud
Department of Medical Pharmacology, INSERM U657, Bordeaux University,
Bordeaux, France
e-mail: annie.fourrier-reglat@u-bordeaux.fr; nicholas.moore@u-bordeaux.fr;
bernard.begaud@u-bordeaux.fr
A. Pariente (*)
Department of Pharmacology, INSERM U657, BP 36, Bordeaux University,
Bordeaux, 33076, France
e-mail: antoine.pariente@u-bordeaux.fr

DOI 10.1007/978-3-319-24741-0_3
28 F. Salvo et al.
following sections, we will present the interests, pitfalls and complementarity of

these three approaches for this monitoring activity.
Keywords Psychotropic drugs • Safety • Monitoring • Limitations •

Pharmacovigilance
3.1 Introduction
Psychotropics are among the most widely used drugs, which makes their safety
monitoring absolutely crucial. This monitoring is also especially important given
the frequency and potential seriousness of adverse effects related to psychotropic
drug use. The impact of benzodiazepine use, for instance, has been estimated to
more than 20,000 serious falls per year in French elderly, around 2000 of these
being fatal. None of the antidepressants existing is considered completely safe. If
serotonin reuptake inhibitors have extremely low cardiac toxicity compared to tricy-
clic antidepressants, they can still be incriminated for QT prolongation and have
been shown to significantly increase the risk of bleeding. Finally antipsychotics are
associated with numerous serious side effects that only appear acceptable with
regard to the seriousness of their indication. However, drug monitoring is not an end
in itself; it has to be associated to a specific goal. Apart from safety signal detection
and risk confirmation, the usual objectives of pharmacovigilance and pharmacoepi-
demiology, it needs to help clinicians to choose which therapeutic they should con-
sider within those presenting a benefit in a given indication. The tools that can be
used for the post-marketing monitoring of psychotropics safety include (i) pharma-
covigilance and individual case analysis, (ii) pharmacoepidemiology and popula-
tion data analysis and (iii) meta-analysis and clinical trial data analysis. In the
following sections, we will present the interests, pitfalls and complementarity of
these three approaches for this monitoring activity.
3.2 Strengths and Limitations of Pharmacovigilance

for Psychotropic Safety Monitoring
Post-marketing pharmacovigilance is mostly based on spontaneous reporting,

through which physicians, health professionals and patients can notify adverse
experiences related to medical drugs to their healthcare system.
Pharmacovigilance relies on the analysis of the information contained in these
reports. The analysis of this information can consist on an individual case analysis
aiming to determine the likelihood of the drug causality in the occurrence of a given
adverse event. It can also consist in comparative statistical analyses of the frequencies
of adverse event reporting observed for the different drugs, the so-called dispropor-
tionality analyses.
3 Methods for the Post-Marketing Monitoring of Psychotropics Safety 29
3.2.1 Strengths and Limitations of Individual Case Reports
The analysis of individual case reports is historically how clinicians specialised in

drug safety initially identified unknown drug adverse effects. This exercise is what
led to the identification of the link between thalidomide and a/phocomelia and is
also related to the classical etiological activities that clinicians routinely engage
when trying to identify the cause of a disease. In drug safety monitoring, individual
case report analysis is mostly useful for the detection of unknown adverse events,
mostly in the early stages of drug marketing, even though it also allows the identifi-
cation of drug adverse events long after their marketing date. The most notable
psychotropic-related adverse effects identified through individual case report analy-
sis include amnesia automatism with benzodiazepines (Hugues et al. 1987), serious
cutaneous reactions with tetrazepam (Breuer et al. 2009; Sanchez-Morillas et al.
2008), bleeding with antidepressants (Aarts et al. 2014) and/or metabolic syndrome
with atypical antipsychotics (Falissard et al. 2011; Tournier et al. 2012).
However, this approach has some limitations. Reporting varies with time and it
is almost impossible, among all reports, to identify those that clearly constitute a
signal. Individual case analysis, which needs robust and extensive data to be per-
formed, can be carried out by comparative statistical analyses of reporting frequen-
cies, which does not require high data quality. The latter can indeed be performed
only if drug use and reaction type are well documented, in addition to basic infor-
mation on patient age and sex.
3.2.2 Strengths and Limitations of Disproportionality Analyses
The principles of disproportionality analyses are simple: using data from a sponta-
neous reporting database (i.e. a collection of data compiled from individual reports),
the reporting rate of one event for one drug is compared to the reporting rate of the
same event for other drugs. The potential safety signals identified by applying sta-
tistical methods to spontaneous reporting data are called Signals of Disproportionate
Reporting (SDRs). SDR detection is currently routinely performed by numerous
drug agencies (EMA, WHO, FDA etc.) as part of their drug safety signal detection
activities. The main limitations of SDR monitoring concern event-competition bias
and drug-competition bias.
3.2.2.1 Event-Competition Bias
The reporting rate of an event for a given drug corresponds to the proportion of
reports for that drug (among all reports present for that drug in a spontaneous report-
ing database) that mentions the event of interest. The more an event is reported for
a drug, the higher its reporting rate will be and the lower the reporting rates of other
30 F. Salvo et al.
events. If the event of interest is reported significantly more with one drug than with
another, its reporting rate has a differential impact on the reporting rates of other
events for that drug of interest as well as for all other drugs. When the reporting rate
is high for a drug, it will lower its reporting rates for other events significantly, while
it will not decrease reporting rates of other events for other drugs with a lower
reporting rate. Thus, the difference in reporting for one event between drugs can
actually artificially induce numerous other differences for the drugs and lead to
biased associations when performing disproportionality analyses. This bias has
been called event-competition bias, or masking effect (Salvo et al. 2013; Pariente
et al. 2010, 2012a). It can affect disproportionality analysis and safety signal detec-
tion for psychotropics in many ways but was mostly demonstrated for antipsychot-
ics considering the potential competition induced by reporting of extrapyramidal
syndrome which hinders signal detection that could otherwise be reliably performed
though disproportionality analyses.
In the study that demonstrated the impact of event-competition bias due to
antipsychotic-related safety signalling (Pariente et al. 2012a), data was used from
the French Pharmacovigilance Database, which includes all adverse drug reactions
reported to the 31 French regional pharmacovigilance centres by health profession-
als but not those reported to manufacturers. The regional centre reviews and assesses
each report before entering them into the database and ensures that the diagnosis
and associated coding for all ADRs are accurate. The potential event-competition
bias induced by antipsychotic-related reports of extrapyramidal syndrome was
explored by performing two data-mining analyses to identify SDRs related to anti-
psychotics. In the first analysis, SDR detection was performed in the whole data-
base. In the second, it was performed on a restricted database obtained after
removing all the reports mentioning extrapyramidal syndromes, irrespective of the
drugs mentioned in the omitted reports. In doing so, the differences in extrapyrami-
dal syndrome reporting between drugs could no longer differentially affect the
reporting rates of other events related to other drugs. The results of the two proce-
dures were then compared to see if new SDRs appeared for antipsychotics once the
influence of reporting of extrapyramidal syndrome for any drugs had been removed
from the database. The data-mining algorithm chosen for the SDR detection was the
Reporting Odds Ratio (ROR) of the case/non-case method (Moore et al. 2005). To
focus only on potential signals that would have been considered relevant for the
examination in routine pharmacovigilance practice, only SDRs with three or more
exposed cases were selected.
In the French Pharmacovigilance Database, extrapyramidal syndrome repre-
sented 3.1 % of all reports entered for the 15-year period of reporting considered
and 16.3 % of all reports mentioning antipsychotics. Removing all reports of
extrapyramidal syndrome from the database (i.e. the related 3.1 % of all the data-
base reports, including those representing 16.3 % of reports mentioning antipsy-
chotics) allowed the identification of six SDRs previously undetected using the
standard procedure as currently applied in pharmacovigilance systems performing
SDR detection. The safety issues already widely identified for three SDRs were
gynecomastia or galactorrhoea, metabolic disorders, and hepatic disorders, while
one potential spurious SDR (perception disturbance) was likely related more to the
indication of antipsychotics than to their adverse effects. The two potential new
safety signals identified were cleft palate abnormalities and congenital gastrointes-
tinal abnormalities. Few data are currently available from the literature for the latter
two, which need to be further investigated.
This study illustrates one of the pitfalls associated with the use of spontaneous
reporting data for the post-marketing assessment of antipsychotics. Dealing with
this pitfall is however easy and should be systematically performed, at least as a
sensitivity analysis, when performing pharmacovigilance studies aiming to identify
or explore SDRs concerning antipsychotics.
3.2.2.2 Drug-Competition Bias
Drug-competition bias is a bias similar to that previously described for events, by

which reports for a given drug for one event will hamper the detection of SDRs
associating this event to other drugs. This bias has been demonstrated, for instance,
for rhabdomyolysis, for which very few SDRs can be detected other than for statins
and fibrates, except when excluding all reports related to these drugs from spontane-
ous reporting datasets, whatever the event reported. To which extent this bias could
affect SDR detection for psychotropics is still being investigated.
Among other biases affecting SDR detection, most concern differences in under-
reporting between drugs and events, which can lead to statistically significant sig-
nals of disproportion that do not reflect differences in true incidence of adverse
events in any way. Among the many possible reasons for differential under-
reporting, notoriety bias and differences associated to time of marketing are of spe-
cific importance as they vary with time and can even combine to create challenges
regarding SDR detection activity. This was especially illustrated for psychotropic
drugs regarding differences in SDR detected for the risk of suicide with antidepres-
sants, with the bias in SDR detection being designated as “dilution bias” (Pariente
et al. 2009). On May 2003, a TV show was aired on the British Broadcasting
Corporation channel (BBC) focussing on the risk of suicide associated with antide-
pressants, which was considered a prominent issue at that time. The show was
highly publicised and drew much attention to these drugs from the general popula-
tion as well as from health professionals. Consequently, a significant rise in the
reporting of suicide with antidepressants occurred. Interestingly, this was respon-
sible for the detection of SDR for suicide not for all antidepressants but only for
escitalopram, the most recently introduced drug of this class at that time. The mech-
anism leading to this difference in SDR detection was very simple: escitalopram
had been involved in very few reports at the time the programme was aired. Thus,
the stimulated high number of reports associating suicides with antidepressants
(and escitalopram) following the broadcasting of the show constituted a very impor-
tant proportion of all reports for escitalopram. Conversely, despite being similar to
other antidepressants, it was hidden among all cases of adverse events that had been
notified since their launching. Results of SDR detection were no longer different
32 F. Salvo et al.
2.5% 9.0%
TV show 8.0%
2.0% 7.0%
6.0%
1.5%
5.0%
4.0% TV show
1.0%
3.0%
0.5% 2.0%
1.0%
2.5% 0.0%
Mar-Apr Mar-Jun Jul-Aug Sep-Oct Nov-Dec Mar-Apr Mar-Jun Jul-Aug Sep-Oct Nov-Dec
Escitalopram Paroxetine Venlafaxine Escitalopram Paroxetine Venlafaxine
Fig. 3.1 Dilution bias: cumulative (left) and instantaneous (right) reporting rates for death by
suicide among all reports for three antidepressants in the UK. After a TV show discussed this pos-
sible issue, a signal of disproportionate reporting (cumulative) was found for the newest antide-
pressant, escitalopram, but not the older drugs (Adapted from Pariente et al. 2009)
among antidepressants when considering all only the reports corresponding to a

comparable time period. In the case of our study, the chosen period was that between
escitalopram marketing and the 6 months following the BBC programme airing.
Figure 3.1 illustrates this phenomenon of dilution of cases within an important pre-
existing amount of reports for three of the studied drugs with different marketing
time: escitalopram (year of marketing, 2002), paroxetine (year of marketing, 1991)
and venlafaxine (year of marketing, 1997). When looking at cumulative amount of
reports, no increase in the spontaneous reporting of suicides can be observed for
paroxetine or venlafaxine after the show. This is completely unmasked when look-
ing at instantaneous reporting rate (that of the 2-month period and not that of all
reports from marketing to the end of the period). This observation excluded the
potential role of escitalopram in inducing a higher risk of suicides than any other
antidepressants in the UK and prevented the drug from being erroneously
withdrawn.
3.2.2.3 Strategies to Improve Drug Monitoring Using Spontaneous

Reporting Data
Used in a more sophisticated way, disproportionality analyses can be used, in addi-

tion to other methods, to rank drugs from the same therapeutic/pharmacologic class
for the risk of reporting of one event. This innovative method has been developed
recently in the context of the ARITMO project aiming to study the ventricular
arrhythmogenic potential of drugs (i.e. all events of this type from simple QT pro-
longation to torsade de pointes or sudden cardiac death). Using this approach, anti-
psychotics were ranked for the risk of such conduction disorders. As this ranking
also considered the global number of reports (number of cases), the global number
of cases for which no other at-risk drug was mentioned in the report and the serious-
ness of the reported event, this approach integrated all aspects of reporting informa-
tion for the assessed drugs (Salvo et al. 2014).
3.3 Strengths and Limitations of Pharmacoepidemiology

for Psychotropic Drug Safety Monitoring
Pharmacoepidemiology studies consist in the analysis of information from a popu-

lation, either performed from data collected prospectively specifically for the study
or performed from data of an existing database. These existing databases can con-
sist in a specific cohort database or in electronic healthcare record database, of
which there are different types, such as claim databases, hospital databases etc.
Pharmacoepidemiology research mostly consists of drug utilisation studies and of
drug safety studies aiming to identify and quantify health risks associated with the
use of marketed drugs. The aim of drug utilisation studies is to facilitate the rational
use of drugs in populations. To achieve this goal, it is necessary to have information
on how drugs are being prescribed and used, in order to identify potential misuse, to
improve drug use and to design interventions such as educational programmes and
drug safety communications, thus optimising the use of marketed drugs. Drug
safety studies follow two main types of designs: the cohort design that allows the
investigation of an initial exposure on multiples outcomes, provided that the out-
comes are sufficiently frequent to be observed in the studied population, and case-
control design that allows the investigation of potential effects of various exposures
on a given outcome, provided that these exposures are sufficiently frequent in the
studied population. In the following sections, we will describe the use and limita-
tions of (i) drug utilisation studies and (ii) drug safety studies performed using data-
bases for the post-marketing monitoring of psychotropics. Examples will be taken
from studies performed in paediatric and adult psychiatric populations, as well as
from populations of elderly dementia patients.
3.3.1 Strengths and Limitations of Drug Utilisation Studies
The main role of drug utilisation studies in the context of psychotropic drugs is to
describe the use of these drugs, including misuse, abuse or otherwise inappropriate
use through which the prescribed drugs can be harmful or ineffective. Studying
psychotropic drug use is also an indirect way to investigate the psychiatric health of
a region/state/country and how this changes over time, for instance, in relation to
changes in society, employment rate within a population, financial difficulties etc.
Drug utilisation studies targeting psychotropic drugs are thus powerful tools to
study and monitor population health indirectly.
Considering benzodiazepines, the use of which is a concern in most European
countries (especially France and Northern Europe), and drug utilisation studies per-
formed from cohort databases demonstrated that almost one third of persons aged
65 and over were frequent users and that approximately 60 % of persons aged 70
and over were frequent benzodiazepines users (i.e. several days per week) for at
least 2 years. There are several studies on this topic, all of which illustrate that the
34 F. Salvo et al.
most important problem related to benzodiazepine use in the elderly is the difficulty
of discontinuing these drugs. This in turn leads to prolonged risks associated with
the use of these drugs, which have only short-term efficacy in sleep disorders or
anxiety. A particular limitation of these studies is that nowadays most are performed
using electronic healthcare databases that lack detailed medical information. Due to
the biases inherent to each pharmacoepidemiology safety study (see following sec-
tion), the lack of medical information on psychotropic drug users constitutes an
important limitation. If proxies of this medical information can be found from data
entered in these electronic databases, they imply drawing hypotheses instead of
making direct observations and significantly limit the potential of pharmacoepide-
miology studies to establish causal associations. An example is given below con-
cerning the patterns of use of cholinesterase inhibitors in dementia patients. In a
study published in 2009, persistence to cholinesterase inhibitors was found to be
higher in patients younger than 80, exposed to antidepressants at the time cholines-
terase inhibitors were initiated and also exposed to antipsychotics. The assumptions
were thus made that there was a greater interest in pursuing treatment among
younger dementia patients as well as in those suffering behavioural disorders
(explaining the use of antipsychotics and antidepressant). However, the alternative
hypothesis would be that these younger patients just receive more intensive care
than older ones, independently of the seriousness of symptoms. Not being able to
conclude which hypothesis really reflects the truth hampers the provision of recom-
mendations concerning the use of psychotropic drugs in this case as in many others.
If the first hypothesis was true, recommendations would focus on improving drug
persistence (in this case, to cholinesterase inhibitors) whatever the age and mostly
in early stage dementia. If the second hypothesis was true, recommendations would
have completely differed, focusing instead on the more judicious use of psychotro-
pics irrespective of age, reserving them only for patients with mood or behaviour
disorders, as their use in dementia is associated with serious adverse events.
3.3.2 Strengths and Limitations of Pharmacoepidemiological

Drug Safety Studies
Pharmacoepidemiological drug safety studies are performed to provide information

which complements that provided by RCTs with respect to the risks associated to
drug use in routine care because RCTs are too short and their populations too small
and homogeneous to allow the investigation of harmful drug effects. However, drug
safety studies cannot provide information on causal associations. Among their
many limitations, the ones inherent to these studies that have to be addressed are the
following: (i) protopathic bias, (ii) indication bias and (iii) confounding. Protopathic
bias can link drug to disease occurrence when the drug is actually used to treat the
very early symptoms of an unrecognised disease. Indication bias can link drug to
disease occurrence when the indication in which the drug is used is itself a risk fac-
tor for the disease being studied as a potential drug side effect. Finally confounders
can link drug to disease when the risk of a drug being used is increased in people
presenting with confounding conditions increasing the risk of developing the dis-
ease being studied. In these three situations, the drug is not at all involved in the
development of the disease but can however be statistically associated with it in
the sense that drug users can be found to present with higher risk of developing the
disease than drug non-users.
3.3.2.1 Protopathic Bias
Antidepressants are mainly prescribed to treat depressive episodes, but their pre-
scription should preferentially concern major depressive episodes as their efficacy
has been shown to be very limited in the management of mild to moderate depressive
disorders. Anxiety and depressive symptomatology can constitute early symptoms
of dementia. For this reason, it is very difficult to assess properly the causal associa-
tion between antidepressants or anxiolytic drug and the subsequent occurrence of
dementia. This situation is made even more complex by the very long prodromal
phase of dementia, in which pathological lesion and prodromal symptoms can be
observed more than 20 years and more than 10 years before diagnosis, respectively.
Two different studies were published in the BMJ specifically regarding the potential
increase in the risk of dementia associated with the use of benzodiazepines. To deal
with the potential protopathic bias, the authors used different methods. The first
study (Billioti de Gage et al. 2012) excluded people with dementia at benzodiaze-
pine initiation and carried out systematic screening of included subjects for demen-
tia allowing the identification of numerous dementia cases in subjects without a
dementia diagnosis. This also allowed the authors to take into account the existence
and seriousness of cognitive decline, depressive symptomatology and anxiety at the
time benzodiazepine was initiated, which cannot be done presently from electronic
healthcare databases. This study demonstrated a potential 50 % increase in the risk
of dementia for incident benzodiazepine users aged 70 and over, over a follow-up of
15 years, most of which was observed more than 10 years after starting benzodiaz-
epine. A similar increase was demonstrated by the second study (Billioti de Gage
et al. 2014), which used data from the Quebec reimbursement database, that pro-
vides more power due to the larger number of patients included. To eliminate pre-
scriptions potentially related to prodromal symptoms of dementia, the information
on benzodiazepine use was censored for the 5–6 year period preceding the diagnosis
of cases and the selection date of controls. In doing so, exposure to benzodiazepines
during the 5–10 year period preceding the first prescription of benzodiazepines was
found to be associated with a 50 % increase in the risk of developing dementia, with
the risk being higher for drugs with long half-life compared to short-acting ones.
3.3.2.2 Indication Bias
Indication bias leads to spurious associations when the indication of a drug corre-
sponds to a situation increasing or decreasing the risk of an event. The drug indi-
cated can thus be associated to an increase or a decrease in the risk of an event not
36 F. Salvo et al.
because of its effect but only because of its indication. Regarding psychotropics,
depression is a risk factor of cognitive decline, dementia or suicide in the patient and
cognitive development disorder in children having parents. Anxiety is a risk factor
for cardiovascular disorders, mostly for ischemic heart disease but also for
depression.
Cholinesterase inhibitors are indicated in patients with mild to moderate demen-
tia, who are of course at lower risk of institutionalisation than patients with advanced
dementia. The use of cholinesterase inhibitors should therefore be found to be asso-
ciated with a decrease in the risk of institutionalisation, not because of its pharma-
cological effect but because of its indication in demented in a condition being at low
risk of this event compared to others. When trying to evaluate the risk of institution-
alisation associated with nonpersistence to cholinesterase inhibitors, Pariente et al.
had to take indication bias into account, even though the data available from the
electronic healthcare database used lacked information on the stage of dementia
(Pariente et al. 2012b). To do so, they considered for their analysis all information
potentially indicating a more advanced disease (as the use of antipsychotics or anti-
depressants, for instance). Another example can be given in which despite the
advanced study design, controlling for indication bias was incomplete (Pariente
et al. 2012c). For numerous reasons that have been cited before, patients taking
antipsychotics have increased risk of presenting with an MI, independently of the
potential role of antipsychotics. To limit this potential bias in a study performed
among dementia patients, we used different methods including traditional adjust-
ment, propensity score adjustment and the use of a self-controlled case series design.
This was done to take into account confounding related to any time-independent
risk factor but not time-dependent ones. In the case of incident antipsychotic treat-
ment, we had to acknowledge that residual confounding by indication could not be
excluded. Indeed, if a dementia patient was prescribed an antipsychotic, it is possible
that the main factors contributing to the acute risk of MI are symptoms of delusion
and agitation which increase stress and anxiety, thus potentially accounting for a
temporary indication bias.
3.3.2.3 Confounding and Confounding Factors
Grimes and Schultz give a very clear definition of confounding in their paper on
biases published in the Lancet series on epidemiology. They explain it as follows:
“Confounding is a mixing or blurring of effects. A researcher attempts to relate an
exposure to an outcome, but actually measures the effect of a third factor, termed a
confounding variable. A confounding variable is associated with the exposure and
it affects the outcome, but it is not an intermediate link in the chain of causation
between exposure and outcome.” They finally state: “Confounding is often easier to
understand from examples than from definitions.” We will therefore give some
famous example of confounding that occurred, were suspected or are still present in
the context of psychotropic post-marketing assessment.
One such example concerns the potential association between antipsychotic use
and myocardial infarction (MI) in patients with psychosis. Clearly, these patients
have an increased baseline risk of MI compared to nonpsychotic patients but
whether antipsychotics add to this risk remains unclear. Psychotic patients tend to
be heavy smokers and have several deleterious life habits regarding their cardiovas-
cular health as compared to nonpsychotic patients. They also tend to have a gener-
ally much lower socioeconomic status and are thus less likely to have an ideal
lifestyle in terms of diet and physical activity. All the above factors clearly expose
them to an increased risk of MI. Adding antipsychotics to the risk factors for MI, the
real risk of MI attributable to antipsychotics is unclear. Is the use of antipsychotics
really responsible for the increased risk of MI observed in antipsychotic users? The
proper way to deal with this confounding is to try to compare the risk of MI in anti-
psychotic users and non-users with similar status regarding smoking, diet, physical
activity and so on. This was attempted but was nevertheless not completely convinc-
ing either to clearly incriminate antipsychotics or to clearly dismiss their potential
harmful effects in that situation. The fact that their use has been found to increase
the risk of thrombotic events in other contexts (e.g. stroke or MI in dementia patient)
for which such confounders would not act supports the likelihood of a causal effect,
but there is still insufficient evidence to reach a conclusion. Confounding occurs in
every non-randomised study and is always treated with caution to avoid reaching
conclusions on the basis of spurious associations. Large electronic healthcare data-
bases, particularly those lacking detail on medical history, are powerful tools as they
include numerous variables that allow a large number of confounders to be consid-
ered. This can be addressed using traditional adjustment in the study of psychotro-
pics (Trifirò et al. 2007, 2010a, b) but also the earlier mentioned self-controlled
design in which a case constitutes its own control, as is done in crossover trials
(Douglas and Smeeth 2008; Whitaker et al. 2009; Maclure 2014).
As one can moreover consider that the information contained in such databases
reflects many characteristics of subjects’ lifestyle and propensity to use healthcare
services, etc., it has been demonstrated that taking this information into account
could control for confounding. This is mostly the principle underlying the use of
propensity scores or disease risk scores that have been widely used in the pharma-
coepidemiologic assessment of psychotropics (Rassen et al. 2013; Wang et al. 2005)
and have allowed this research to provide more compelling evidence about the asso-
ciations under investigation.
3.4 Strengths and Limitations of Meta-analyses

for Psychotropic Drug Safety Monitoring
Meta-analysis is a research method in which studies, rather than people, are sur-
veyed; it represents one way to summarise, integrate and interpret selected studies
related to one or more outcomes of interest. In doing so, it circumvents the sample
38 F. Salvo et al.
power limitations of individual studies and increases the generalisability of the

results. Of course, it is necessary that included studies share enough characteristics
as to allow a meaningful comparison, especially in terms of outcome(s) of interest
and in terms of study design. As an example, it is well known that antipsychotics
can induce QT prolongation. Although this could be a causal factor for potentially
fatal cardiac arrhythmias, both cardiac arrhythmias and sudden cardiac death can be
due to other causes. Thus, pooling events such as QT prolongation and cardiac
arrhythmias in a single meta-analysis are inappropriate.
From a general point of view, two main sources of data are available to carry out
meta-analyses aiming to study the safety of drugs: clinical trials and observational
studies. These two settings have important differences and different limitations
that have to be taken into account. Even if the information they provide is interest-
ing, it is generally inappropriate to pool the results from clinical trials and observa-
tional studies.
3.4.1 Strengths and Limitations of Meta-analysis

of Clinical Trials
Randomised clinical trials (RCTs) are without doubt the best study design for the
assessment of health intervention efficacy. However, existing clinical trials are not
the best sources of information for the evaluation of the safety of drugs, especially
for heterogeneous drug classes such as psychotropic drugs. Among these drugs,
antipsychotics are probably the most heterogeneous. They have several definitions
and classifications, the most common of which refers to them as first- and second-
generation antipsychotics. This classification has several limitations, as, for exam-
ple, it does not take into account the differences in the molecular structure of these
drugs or in their pharmacological action. Aside from a potential difference in the
safety profile of antipsychotics regarding antipsychotic-induced Parkinsonism, the
concept of first- and second-generation antipsychotics also implies important differ-
ences regarding drug monitoring and assessment: development and marketing of
these drugs relate to a very different era of drug safety evaluation and to significant
differences in evidence-based standards. This is important to take into account when
planning a meta-analysis: comparing clinical trials performed about 50 years ago
with studies performed in the last 10 years can be inappropriate, whether evaluating
efficacy or safety. Safety was not a concern for the first antipsychotics initially, as
no pharmacological alternatives were available. Over time, safety aspects became
more and more central and were used to develop and market the so-called second-
generation antipsychotics, which were supposed to reduce the frequency and sever-
ity of induced Parkinsonism (Simpson and Varga 1974). After the marketing of
clozapine and other second-generation antipsychotics, Parkinsonism gradually
stopped being a drug safety issue. At the same time, other events were discovered
that became the key to the benefit/risk ratio of antipsychotics, such as agranulocyto-
sis, metabolic syndrome and diabetes and cardiovascular events (Newcomer 2005).
Haematological adverse events and cardiovascular events were neither searched for
nor reported systematically in clinical trials performed on first-generation antipsy-
chotics. Thus, searching for related information in the original RCTs will certainly
be time-consuming and will also lead to the retrieval of biased information relating
to a screening or diagnostic bias between first- and second-generation antipsychot-
ics, relating in turn to differences in procedures for the detection and reporting of
these events between RCTs evaluating first-generation antipsychotics and RCTs
evaluating second-generation ones.
Randomisation remains the ideal tool to evaluate differences among two or more
interventions. Nevertheless, a critical eye on randomisation of studies included in a
safety meta-analysis needs to be maintained, in particular for small-size studies.
When a large number of subjects are included in the study, we can assume that the
distribution of risk factors for the event(s) of interest will have made the group com-
parable for any measured or unmeasured characteristics. When the study size is
small, this is unlikely to be true. Thus, performing a meta-analysis to assess the risk
of drug-related harm in small trials (e.g. first-generation antipsychotics) together
with other drugs studied through large studies (e.g. second-generation antipsychot-
ics) could lead to a wrong interpretation of data if this aspect is not correctly taken
into account.
The technique of network meta-analysis (also referred to as multiple treatment
comparison meta-analysis) has advantages over conventional pairwise meta-
analysis, as the technique borrows strength from indirect evidence to gain certainty
about all treatment comparisons. In other words, network meta-analysis allows indi-
rect comparison for the estimation of comparative effects of drugs and yields more
reliable and definitive results than would a pairwise meta-analysis (Mills et al.
2013). As for conventional meta-analysis, the sample size of the included studies is
per se crucial, as the chance to find outcome(s) of interest has to be taken into
account when a meta-analysis is planned. The more rare the event, the more difficult
it will be to find in clinical trials. If the principle of meta-analysis is to aggregate
data to increase the power of the analysis in order to find any differences, a meta-
analysis will be useless without a sufficient number of cases. It will thus have to be
postponed until the likelihood of identifying a minimal number of events from
existing RCTs is judged acceptable. This can be assessed before starting the time-
consuming exercise of systematic review that precedes each meta-analysis. If the
aim of a meta-analysis is to evaluate the risk of sudden cardiac death associated with
the use of antipsychotics using data from RCTs, for instance, this likelihood could
currently be estimated as follows: consider a scenario where 212 RCTs are being
considered for a meta-analysis where 43,049 patients are treated with 15 different
antipsychotics or placebo for 6–14 weeks (Leucht et al. 2013). Under the conserva-
tive hypothesis that all patients were followed for 14 weeks, this would correspond
at best to a total follow-up of 11,590 patient-years. The incidence of sudden cardiac
death in the general population is 1.4 per 1000 person-years (95 % CI 1.3–1.5) (Ray
et al. 2009). Assuming that antipsychotics as a class increase the risk of sudden
cardiac death by 300 %, no more than 45–50 cases of sudden cardiac death would
be expected in these RCTs for the 15 antipsychotics considered together. This is
40 F. Salvo et al.
clearly insufficient to allow the comparison of risk of sudden cardiac death between
individual antipsychotics using RCT data.
One can assume that meta-analysis of RCTs is of limited interest when rare or
very rare events are studied. Another limitation regarding meta-analysis of RCTs is
that they highly depend on the quality of the RCTs involved. Reporting bias is an
aspect of note when meta-analyses aim to evaluate safety of treatments from RCTs,
as safety is mostly a secondary objective of the RCTs, which is often affected by
under-reporting. At the international level, there is currently no clear rule for the
reporting of adverse events in the manuscripts issued from a RCT. This does not
allow the exclusion of reporting bias, as the quality and exhaustiveness of the listed
adverse events detailed in the publications mainly depends on the authors’ choices
or the journal instructions. Sometimes this selectivity in reporting is transparent
(e.g. adverse event with ≥2 % of frequency), and sometimes it is more complex to
picture: the reader can find no mention of adverse events (serious or otherwise),
these is a lack of a clear definition of the considered adverse events, it is not possible
to clearly distinguish between number of events and number of patients with an
event, groups of events are considered together, etc. For old RCTs, even if a request
for additional data to authors or drug companies is successful, these data can fre-
quently be impossible to interpret. The obligation to annex to the principal publica-
tion or public report a list of every adverse events which occurred in every included
patient would easily solve this issue and allow a much more efficient meta-analysis
process. Regarding psychotropics, this approach would allow a much more efficient
response to patient expectations as well as helping clinicians to make informed
therapeutic choices (Tiihonen et al. 2006).
3.4.2 Role and Limitations of Meta-analysis

of Observational Studies
Meta-analyses of observational studies are considered to provide “weaker” evi-

dence than those using clinical trial data. The absence of randomisation in the
source studies has a key role in this relative downgrading of the evidence (Guyatt
et al. 2011). A certain degree of residual confounding remains in all real-life stud-
ies, even if matching or adjustment techniques are used. As a consequence, per-
forming a meta-analysis only using observational studies for frequent events, which
could be easily retrieved from clinical trial data, is certainly inadequate. Thus, once
again, the expected frequency of the event has to be taken into account when plan-
ning a systematic review and a meta-analysis. For a frequent event, a larger system-
atic review is preferable in order to include evidence from randomised studies. The
principal results will come from the randomised studies, while the observational
data will be an additional data source for a deeper interpretation of the results.
Conflicting results will be due to residual confounding or different pattern of drug
use, which are not taken into account in observational studies or not represented in
clinical trials.
For rarer events, meta-analysis of observational studies is the only source of data
and has to be considered as the best option to investigate them. As a direct conse-
quence, the quality of the included studies is the most crucial aspect in the interpreta-
tion of the meta-analysis results. The use of the Newcastle-Ottawa scale (Wells et al.
2014) is recommended by the Cochrane Collaboration to evaluate the quality of non-
randomised studies (Higgins 2011). This tool scores each study from zero to nine stars
for the selection and the comparability of the groups and for the ascertainment of
either the exposure for case control or outcome of interest for cohort studies.
Unfortunately, this guidance is too unspecific to investigate the methodological issues
of observational studies with respect to a specific event. Its use may produce highly
arbitrary results (Stang 2010), and even studies that have received the maximum score
with this scale have to be interpreted with caution, in particular where the appropriate-
ness (and consistency) of timeframe between drug exposure and event is questionable
and where for the article-by-article evaluation of residual confounders is involved.
Meta-analysis of observational studies can pool data coming from prospective or
retrospective study design. Cohort or case-control approaches, the most frequently
used observational study designs, have some intrinsic differences. In general, cohort
studies are supposed to provide a higher level of scientific evidence than case-
control studies (Tiihonen et al. 2006), and the latter tend to overestimate the strength
of associations as compared to cohort studies. The influence of study design as a
source of heterogeneity has thus to be carefully investigated.
Publication bias is a significant concern for meta-analyses of observational stud-
ies. It is particularly difficult to detect and to evaluate in this setting as registries for
the recording of observational studies are not as used as are those of clinical trials.
In contrast to clinical trials, observational studies that do not find an increase in the
risk of an event have more difficulty in getting published than RCTs which do not
find an increased risk. This publication bias can be explored using funnel plots when
at least ten estimates for the same drug are available (Higgins 2011). More than
using statistic tools, the existence of publication bias for meta-analyses using obser-
vational studies could be assessed by following points:
1. Number of studies privately funded on a specific drug: the more published stud-
ies are sponsored studies, the higher the risk of publication bias is
2. Number of studies which investigated a psychotropic drug class: the more pub-
lished studies investigate a drug class as a whole without a priori hypotheses of
differences between drugs, the more the results can be considered unbiased
3. Effect size of a meta-analysis: when a large effect size is found from a meta-
analysis, the chance that new published evidence changes substantially the
results is remote
4. Consistency of results and pharmacological profile of the drugs under investiga-
tion: it will be always be more convincing if preclinical data corroborate meta-
analysis of observational studies (and vice versa)
Classically, heterogeneity is another important issue of meta-analyses: it relates
to significant differences in risk estimates between the studies pooled in the
meta-analysis. Heterogeneity in risk for a drug (intra-drug heterogeneity) could be
42 F. Salvo et al.
related to its different pattern of use or in terms of dose or indication of use or both.
Thus, this heterogeneity could be considered as a source of a potential safety signal
and has to be further analysed, by meta-regressions or by further and more specific
studies (even randomised, if required). Furthermore, heterogeneity among different
drugs of the same classes (inter-drug heterogeneity) could reveal a different risk
profile and merit even more attention than results that are entirely homogeneous. If
the pharmacological profile supports the hypothesis of a differential risk, heteroge-
neity could be more important than the pooled risk estimate; thus, it could be the
principal result of a meta-analysis and drive regulatory decisions on drugs.
As reporting bias is an issue in clinical trial settings that could be easily solved
by single patient data availability, the meta-analysis of observational studies would
have increased value if raw data could be used. This would allow the harmonisation
of the event definition, adjustment and matching and even allow the construction of
a network of observational studies starting from the pseudo-randomised patient
numbers to gain certainty about treatment safety in real life.
3.5 Conclusion
Post-marketing monitoring of psychotropics is an essential and difficult activity.

Powerful tools exist for such ends, all having their advantages and disadvantages.
Pharmacovigilance can provide clinically meaningful information on the risk associ-
ated with the use of a psychotropic drug, but it is plagued by under-reporting and so
many biases that it has to be used with extreme caution, even in the modest exercise
of signal detection. Pharmacoepidemiology safety studies became very powerful
with the use of large electronic healthcare databases which allow the direct or indi-
rect control of confounding. However, the lack of medical information especially
regarding the seriousness of symptoms limits their ability to fully exclude proto-
pathic or indication bias. The best remedy for this is to perform such safety studies
using large cohorts. Finally, the data on drug safety available from RCTs should be
kept in mind, even if mostly relating to pre-marketing information. While a single
RCT is insufficient to draw robust conclusions on drug safety, the pooling and meta-
analysis of such studies are informative as well as being free from confounding.
Network meta-analyses are also a powerful tool in the investigation of drug safety,
although little has been published in this field concerning psychotropics to date.
References
Aarts N, Akoudad S, Noordam R, Hofman A, Ikram MA, Stricker BH et al (2014) Inhibition of

serotonin reuptake by antidepressants and cerebral microbleeds in the general population.
Stroke J Cereb Circ 45(7):1951–1957
Billioti de Gage S, Begaud B, Bazin F, Verdoux H, Dartigues JF, Peres K et al (2012) Benzodiazepine
use and risk of dementia: prospective population based study. BMJ 345:e6231
Billioti de Gage S, Moride Y, Ducruet T, Kurth T, Verdoux H, Tournier M et al (2014) Benzodiazepine

use and risk of Alzheimer’s disease: case-control study. BMJ 349:g5205
Breuer K, Worm M, Skudlik C, Schroder C, John SM (2009) Occupational airborne contact allergy
to tetrazepam in a geriatric nurse. J Dtsch Dermatol Ges J Ger Soc Dermatol JDDG
7(10):896–898
Douglas IJ, Smeeth L (2008) Exposure to antipsychotics and risk of stroke: self controlled case
series study. BMJ 337:a1227
Falissard B, Mauri M, Shaw K, Wetterling T, Doble A, Giudicelli A et al (2011) The METEOR
study: frequency of metabolic disorders in patients with schizophrenia. Focus on first and sec-
ond generation and level of risk of antipsychotic drugs. Int Clin Psychopharmacol
26(6):291–302
Guyatt GH, Oxman AD, Schunemann HJ, Tugwell P, Knottnerus A (2011) GRADE guidelines: a
new series of articles in the Journal of Clinical Epidemiology. J Clin Epidemiol
64(4):380–382
Higgins JPT (2011) Cochrane handbook for systematic reviews of interventions: the cochrane col-
laboration. Available from: http://www.cochrane-handbook.org
Hugues FC, Jouglard J, Le Jeunne C, Moulin M, Begaud B (1987) Psychotic disorders linked to
the inhibition of benzodiazepine catabolism. La Rev Med Interne/fondee Soc Natl Fr Med
Interne 8(4):433–436
Leucht S, Cipriani A, Spineli L, Mavridis D, Orey D, Richter F et al (2013) Comparative efficacy
and tolerability of 15 antipsychotic drugs in schizophrenia: a multiple-treatments meta-
analysis. Lancet 382(9896):951–962
Maclure M (2014) Case-crossover designs for more patient-centred epidemiology. Paediatr Perinat
Epidemiol 28(2):77–78
Mills EJ, Thorlund K, Ioannidis JP (2013) Demystifying trial networks and network meta-analysis.
BMJ 346:f2914
Moore N, Thiessard F, Begaud B (2005) The history of disproportionality measures (reporting
odds ratio, proportional reporting rates) in spontaneous reporting of adverse drug reactions.
Pharmacoepidemiol Drug Saf 14(4):285–286
Newcomer JW (2005) Second-generation (atypical) antipsychotics and metabolic effects: a com-
prehensive literature review. CNS Drugs 19(S 1):1–93
Pariente A, Daveluy A, Laribiere-Benard A, Miremont-Salame G, Begaud B, Moore N (2009)
Effect of date of drug marketing on disproportionality measures in pharmacovigilance: the
example of suicide with SSRIs using data from the UK MHRA. Drug Saf 32(5):441–447
Pariente A, Didailler M, Avillach P, Miremont-Salame G, Fourrier-Reglat A, Haramburu F et al
(2010) A potential competition bias in the detection of safety signals from spontaneous report-
ing databases. Pharmacoepidemiol Drug Saf 19(11):1166–1171
Pariente A, Avillach P, Salvo F, Thiessard F, Miremont-Salame G, Fourrier-Reglat A et al (2012a)
Effect of competition bias in safety signal generation: analysis of a research database of spon-
taneous reports in France. Drug Saf 35(10):855–864
Pariente A, Fourrier-Reglat A, Bazin F, Ducruet T, Dartigues JF, Dragomir A et al (2012b) Effect
of treatment gaps in elderly patients with dementia treated with cholinesterase inhibitors.
Neurology 78(13):957–963
Pariente A, Fourrier-Reglat A, Ducruet T, Farrington P, Beland SG, Dartigues JF et al (2012c)
Antipsychotic use and myocardial infarction in older patients with treated dementia. Arch
Intern Med 172(8):648–653; discussion 54–5
Rassen JA, Shelat AA, Franklin JM, Glynn RJ, Solomon DH, Schneeweiss S (2013) Matching by
propensity score in cohort studies with three treatment groups. Epidemiology 24(3):
401–409
Ray WA, Chung CP, Murray KT, Hall K, Stein CM (2009) Atypical antipsychotic drugs and the
risk of sudden cardiac death. N Engl J Med 360(3):225–235
Salvo F, Leborgne F, Thiessard F, Moore N, Begaud B, Pariente A (2013) A potential event-
competition bias in safety signal detection: results from a spontaneous reporting research data-
base in France. Drug Saf 36(7):565–572
44 F. Salvo et al.
Salvo F, Raschi E, Moretti U, Chiarolanza A, Fourrier-Reglat A, Moore N et al (2014)

Pharmacological prioritisation of signals of disproportionate reporting: proposal of an algo-
rithm and pilot evaluation. Eur J Clin Pharmacol 70(5):617–625
Sanchez-Morillas L, Laguna-Martinez JJ, Reano-Martos M, Rojo-Andres E, Ubeda PG (2008)
Systemic dermatitis due to tetrazepam. J Investig Allergol Clin Immunol 18(5):404–406
Simpson GM, Varga E (1974) Clozapine – a new antipsychotic agent. Curr Ther Res Clin Exp
16(7):679–686
Stang A (2010) Critical evaluation of the Newcastle-Ottawa scale for the assessment of the quality
of nonrandomized studies in meta-analyses. Eur J Epidemiol 25(9):603–605
Tiihonen J, Lonnqvist J, Wahlbeck K, Klaukka T, Tanskanen A, Haukka J (2006) Antidepressants
and the risk of suicide, attempted suicide, and overall mortality in a nationwide cohort. Arch
Gen Psychiatry 63(12):1358–1367
Tournier M, Begaud B, Cougnard A, Auleley GR, Deligne J, Blum-Boisgard C et al (2012)
Influence of the drug exposure definition on the assessment of the antipsychotic metabolic
impact in patients initially treated with mood-stabilizers. Br J Clin Pharmacol 74(1):189–196
Trifirò G, Verhamme KM, Ziere G, Caputi AP, Ch Stricker BH, Sturkenboom MC (2007) All-cause
mortality associated with atypical and typical antipsychotics in demented outpatients.
Trifirò G, Dieleman J, Sen EF, Gambassi G, Sturkenboom MC (2010a) Risk of ischemic stroke
associated with antidepressant drug use in elderly persons. J Clin Psychopharmacol
30(3):252–258
Trifirò G, Gambassi G, Sen EF, Caputi AP, Bagnardi V, Brea J et al (2010b) Association of
community-acquired pneumonia with antipsychotic drug use in elderly patients: a nested case-
control study. Ann Intern Med 152(7):418–425, W139-40
Wang PS, Schneeweiss S, Avorn J, Fischer MA, Mogun H, Solomon DH et al (2005) Risk of death
in elderly users of conventional vs. atypical antipsychotic medications. N Engl J Med
353(22):2335–2341
Wells G, Shea B, O’Connell D, Peterson J, Welch V, Losos M et al (2014) The Newcastle-Ottawa
Scale (NOS) for assessing the quality of nonrandomised studies in meta-analyses2010.
Available from: http://www.ohri.ca/Programs/clinical_epidemiology/oxford.asp
Whitaker HJ, Hocine MN, Farrington CP (2009) The methodology of self-controlled case series
studies. Stat Methods Med Res 18(1):7–26
Chapter 4
Contribution of UK Prescription-Based Event
Monitoring Methods in the Pharmacovigilance
of Psychotropic Medications
Deborah Layton
Abstract Psychotropic drugs are widely used in clinical practice. The balance
between drug safety and efficacy is affected by many factors including the prescriber,
the patient, the disease as well as other environmental effects. Physician factors
include failure to diagnosis conditions because of the complex nature and pattern of
presenting symptoms. Thus patients may not receive the treatment they need, whether
that is a particular medication or the appropriate medication at sub-therapeutic doses.
Patient factors include failure to recognize existence of a condition accompanied by
denial and reluctance to seek medical attention. In addition, many patients prescribed
antipsychotic medications do not take their medicines in accordance with instruc-
tions. Such factors, accompanied by large inter- and intra- individual manifestation
of mental health disease severity contribute to difficulties in appropriately managing
such patients, especially long -term. In this chapter, examples of real-life studies
about some of these challenging issues will be described, as reported from the post-
marketing event-monitoring systems, now known as Modified Prescription-Event
Monitoring (M-PEM) and Specialist Cohort Event Monitoring (SCEM).
Keywords Prescription-based event monitoring • Psychotropic drugs • Observational

research • Drug utilization • Methodology
4.1 Introduction
In the past three decades, a host of new psychotropic medicines have been
introduced for the treatment of mental health disorders. An examination of trends
in prescriptions and costs of all classes of psychiatric medication in England
between 1998 and 2010 reported that prescribing of psychotropics has increased by
D. Layton
Drug Safety Research Unit, Southampton, UK
e-mail: deborah.layton@dsru.org

DOI 10.1007/978-3-319-24741-0_4
46 D. Layton
7 % per year on average (in line with other drugs of other therapeutic classes) from
8.3 % of all prescription items in 1998 to 8.6 % in 2010. However there were highly
statistically significant upward trends in prescriptions of all classes of psychiatric
medications, except for hypnotics and anxiolytics (Ilyas and Moncrieff 2012). In
contrast the corresponding costs of psychiatric medications increased by 5 % per
year on average (adjusted for inflation) during the same period, which was statisti-
cally significantly different to the costs of other prescription drugs (which rose by
3 %). Rising prescription numbers and drug costs do not imply a rising number of
psychotropic medication users in the UK. However, according to the UK Adult
Psychiatric Morbidity Survey (APMS) conducted in 2007, the prevalence of at
least one common mental health disorder in people aged 16–64 is increasing
(15.5 % 1993, 17.5 % in 2000, 17.6 % 2007) (National Health Service 2009).
Psychotropic medications have been shown to be effective in the treatment of
manifestations of mental health conditions, yet a delicate balance exists between
efficacy and safety. This balance is affected by many factors including the pre-
scriber, the patient, the disease as well as other environmental effects. Physician
factors include failure to diagnosis conditions because of the complex nature and
pattern of presenting symptoms. Thus, patients may not receive the treatment
required, whether that is a particular medication or the appropriate medication but
given at subtherapeutic doses. Patient factors include failure to recognise existence
of a condition accompanied by denial and reluctance to seek medical attention. In
addition, many patients prescribed antipsychotic medications do not take their med-
icines in accordance with instructions. Such factors, accompanied by large inter-
and intra-individual manifestation of mental health disease severity, contribute to
difficulties in appropriately managing such patients, especially long term. In this
chapter, examples of real-life studies about some of these challenging issues will be
described, as reported from the post-marketing event-monitoring systems, now
known as Modified Prescription-Event Monitoring (M-PEM) (Layton and Shakir
2014) and Specialist Cohort Event Monitoring (SCEM) (Layton and Shakir 2015).
4.2 Event Monitoring: Fulfilling a Need for Observational

Research
The limitations of randomised controlled trials (RCTs) in terms of statistical power

to detect rare but serious adverse events and exclusion of high-risk populations which
can impact on generalisability of results to real-world clinical practice are well
known (Singh and Loke 2012). Observational studies can help examine natural vari-
ation in disease and treatment effects. Issues that are fundamental in making pre-
scribing decisions include understanding the risk of potential adverse events in
different types of patients that are treated with psychoactive medications, particularly
those where there is limited information available from premarketing development
programmes. The reason for this relates partly to the process by which a medicinal
product is approved and partly to design. The initial marketing authorisation (MA)
for a product may be applied for by a pharmaceutical company on the basis of a
4 Contribution of UK Prescription-Based Event Monitoring Methods 47
particular clinical indication, which may be in a narrowly defined population. Clinical

trials may have been conducted in a relatively healthy patient population to obtain
sufficient information on efficacy and safety in that population to satisfy a regulatory
authority’s requirements. The patient population selected may exclude those groups
that are difficult to recruit for whom issues of ethics and consent arise, such as chil-
dren, the elderly, patients with comprehension and learning difficulties or those who
are acutely ill. In terms of study design, clinical trials are experiments where the
circumstances of treatment (such as setting, treatment regimens, duration and inten-
siveness of monitoring) are often controlled by the study investigator.
For the last five decades, since the thalidomide disaster (Lenz et al. 1962), there
has been recognition for the need for post-marketing observational research to iden-
tify safety issues associated with the use of a new drug in all populations likely to
receive it in routine clinical practice (Finney 1965). This is because the populations
likely to be studied are a more heterogeneous population than those observed in
clinical trials. The importance of observational studies has increased dramatically
since the revised EU Pharmacovigilance Legislation came into force, July 2012
(European Commission 2012). Accordingly every new medicinal product must
have comprehensive Risk Management Plan (RMP) in place as part of its approval
and to retain its approved status (European Medicines Agency 2014). M-PEM, pre-
viously known as PEM, is a national surveillance system whose origins stem from
the recognition that spontaneous reporting systems may fail to identify all safety
hazards post-marketing. M-PEM is an enhanced application of PEM that offers a
targeted safety surveillance system that can be readily adapted to meet the specific
needs of a RMP and systematically collect information on large cohorts about
patient baseline characteristics in relation to prespecified risks and can quantify the
incidence and prevalence of risks of adverse events after treatment initiation (Layton
and Shakir 2014). SCEM is a recent adaptation of the technique, which focuses on
surveillance in secondary care setting (Layton and Shakir 2015). This application
was developed in the recognition that safety studies conducted exclusively in the
primary care setting may be at risk of biased conclusions about the prevalence of the
types of patients prescribed new medications and also the frequency of adverse
events, because of the potential exclusion of patients who are managed predomi-
nantly within the secondary care setting. These patients, who may be initiated under
the care of a specialist health care professional, may have different characteristics
and health experiences to those treated by physicians in the primary care setting for
similar indications. Since the adoption of a new medicine into clinical practice in
the UK is often initially facilitated by hospital specialists, there is a need for data
capture across both the primary and secondary care setting, to ensure all relevant
exposed populations are characterised and monitored. SCEM enables cohorts of
patients prescribed a new medicine in the secondary care setting to be monitored.
Both M-PEM and SCEM collect additional information that can inform on
patient and physician behaviours and also highlight ‘off-label’ prescribing which
refers to the use of a drug in situations where a medicinal product is intentionally
used for a medicinal purpose not in accordance with the authorised product infor-
mation (European Medicines Agency 2012). Both M-PEM and SCEM are recog-
nised as tools to conduct real-world post-authorisation safety studies (PASS) to
48 D. Layton
gather additional safety monitoring information or assess patterns of drug utilisation

(European Medicines Agency 2013). Each study is conducted in accordance with
national and international guidelines (British Medical Association Board of Science
and British Medical Association Science and Education 2006; Council for
International Organizations of Medical Sciences and World Health Organisation
2002; General Medical Council 2009a, b).
The general methodology uses a retrospective non-interventional observational
cohort design to provide active surveillance of targeted medicines on a national scale
in England. Specific details of the study methodologies have been provided else-
where (Layton and Shakir 2011). In brief, for M-PEM, primary care dispensed pre-
scription data sources, and medical record-based data sources are used to provide
data for eligible patients (Table 4.1). For all studies, exposure (index/exit) data are
derived from dispensed prescriptions issued by general practitioners (GPs) immedi-
ately after market launch until sufficient numbers have been identified (per protocol
often several thousands). In contrast, for SCEM, networks of specialists are estab-
lished (facilitated by the UK National Institute of Health Research (NIHR) Clinical
Research Network (CRN)) (Department of Health 2014) and patients are identified
by these practitioners subsequently. For all approaches, outcome data (indication,
events and reasons for stopping) that have been recorded in the patient’s medical
records may be derived from questionnaires sent to each physician at some pre-
defined period after the date of each patient’s first prescription, but at a minimum of
3 months to allow information to be shared between patient and prescriber. This
duration of follow-up is often driven by the expected pattern of risk for events identi-
fied within the study primary objectives. However, for general surveillance in
M-PEM, the desire is to capture at least 6 months observation. There are no specific
exclusion criteria. Specific to M-PEM and SCEM, more detailed information on
posology, identified and potential risks, selected risk factors-prior medical history/
concomitant medicines and selected prescriber or patient behaviours may be
collected.
For each approach, events of interest (e.g. identified risks, such as suicide events)
may be followed up for purposes of further evaluation. For each patient, trained cod-
ing staff prepared a computerised, longitudinal, chronological record of demo-
graphic, exposure and outcome data (including additional follow-up). Quantitative
and qualitative data analyses are conducted. Quantitative analyses include descrip-
tive statistics to summarise patient and prescriber (where provided) characteristics
such that drug utilisation and compliance with recommended prescribing regimens
may be described. Calculations of event risk and rates (incidence densities for a fixed
period (t) – IDt- usually expressed in units of first event reports per 1000 patient-
months) can give estimates of real-world frequency. Calculation ID differences
between periods of observation are effective methods by which disproportionality in
risk or ID may be observed, suggestive of signals of treatment effects. Inter-cohort
IDs and ID difference estimates may be calculated, as well as intra-cohort estimates
for special populations of interest. Other complementary analyses include describ-
ing frequency of events given as reasons for stopping and also application of survival
methods to provide further detail on pattern and time to onset of an event after start-
ing treatment with a study drug. Qualitative analyses are undertaken for exploration
Table 4.1 Characteristics of standard Prescription-Event Monitoring (PEM), Modified PEM

(M-PEM) and Specialist Cohort Event Monitoring (SCEM) studies
Study
Characteristics PEM M-PEM SCEM
Setting Primary care Secondary care
Prescribers General Practitioner Specialist
Design Non-interventional observational cohort
Data Source Secondary use of data from existing medical records
Ethics Waiver under Section 251 of NHS Act 2006 Yes
Sample size 10,000+ Bespoke to targeted events
Period of 6–12 months 3–12+ months ≤3 months
Observation
Questionnaire Standard (single A5) across Bespoke design (multiple A4), by study
all studies
Drug Utilization All populations Targeted sub-populations considered at risk
Surveillance General surveillance General and targeted events (identified,
potential and missing risks)a
Risk factors Event (case) specific on For targeted identified and potential risks
follow-up for all patients
Analysis Crude and age/sex adjusted Multiple covariate adjusted measures of
measures of occurrence and occurrence and association; survival
association methods for signal generation; intra- and
inter- cohort comparisons possible
PASS RMP No Yes
ENCEPP No Yes
registration
Prescriber No Yes
remuneration
Grey highlights indicate common aspects of the method
ENCEPP European Network of Centres of Excellence in Pharmacoepidemiology and
Pharmacovigilance, NHS National Health Service, PASS Post-authorisation safety study, RMP
Risk Management Plan. A4: page dimension 210 by 297 mm; A5: page dimension 148 by 210 mm
a
Identified risks (known from clinical trials), potential risks (effects not observed in trials but
expected e.g. class effects) and missing risks (identified and potential effects that may occur in
populations not studied)
of signals or events of interest. Through medical evaluation, individual or clusters of

event reports are examined and important attributes identified. Examples of these
approaches will be presented in the remainder of this chapter.
4.3 Completed Studies
A wide range of drugs used to manage mental health conditions have been studied
using standard and modified approaches. These include agents to manage depres-
sive disorders, anxiety disorders, attention deficit hyperactivity disorder (ADHD),
schizophrenia and dementia (Table 4.2). Treatments used to manage conditions
Table 4.2 List of completed studies (standard (PEM)/modified (M-PEM)) of psychotropic medicines, by ATC code (World Health Organisation 2015)
50
Therapeutic class and ATC code Name: generic [UK proprietary] Design/cohort size Collection period
Antiobesity preparations (excl. diet products)
A08AA10 Sibutramine [REDUCTIL™] Standard (N = 12,336) Oct 2001–Jun 2002
A08AX01 Rimonabant [ACOMPLIA™] Modified (N = 10,008) Jun 2006–Oct 2008
Opioids
N02AB03 Fentanyl buccal [EFFENTORA™] Modified (N = 556) Mar 2009–Jun 2011
N02AB03 Fentanyl nasal [PECFENT™] Modified (N = 63) Dec 2010–Sep 2012
Antipsychotics
N05AE05 Sertindole [SERDOLECT™] Standard (N = 436) Dec 1996–Dec 1997
N05AH03 Olanzapine [ZYPREXA™] Standard (N = 8858) Dec 1996–May 1998
N05AH04 Quetiapine [SEROQUEL™] Standard (N = 1728) Jun 1998–Jan 2000
N05AH04 Quetiapine [SEROQUEL XL™] Modified (N = 13,276) Sep 2008–Feb 2013
N05AX08 Risperidone [RISPERDAL™] Standard (N = 7684) Jul 1993–Dec 1996
Anxiolytics
N05BE01 Buspirone [BUSPAR™] Standard (N = 11,113) Mar 1988–Feb 1989
Hypnotics and sedatives
N05CF01 Zopiclone [ZIMOVANE™] Standard (N = 11,543) Mar 1991–Jul 1991
N05CF02 Zolpidem [STILNOCT™] Standard (N = 13,460) Jul 1994–Jan 1996
Antidepressants
N06AB03 Fluoxetine [PROZAC™] Standard (N = 12,692) Mar 1989–Mar1990
N06AB05 Paroxetine [SEROXAT™] Standard (N = 13,741) Mar 1991–Mar1992
N06AB06 Sertraline [LUSTRAL™] Standard (N = 12,734) Jan 1991–Sep1992
N06AB08 Fluvoxamine [FAVERIN™] Standard (N = 10,983) Feb 1987–Feb 1988
N06AG02 Moclobemide [MANERIX™] Standard (N = 10,835) Jun 1993–Aug 1995
N06AX06 Nefazodone [DUONIN™] Standard (N = 11,834) Jan 1996–Feb 1997
D. Layton
N06AX11 Mirtazapine [ZISPIN™] Standard (N = 13,554) Sep 1997–Feb 1999
N06AX12 Bupropion [ZYBAN™] Standard (N = 11,735) Jul 2000–Aug 2000
N06AX16 Venlafaxine [EFEXOR™] Standard (N = 12,642) May 1995–Jun 1996
N06AX21 Duloxetine [CYMBALTA/YENTREVE™] Standard (N = 19,485) Sep 2004–Apr 2005
Psychostimulants (agents used for ADHD and nootropics)
N06BA07 Modafinil [PROVIGIL™] Modified (N = 2092) Jul 2004–Aug 2005
N06BA09 Atomoxetine [STRATTERA™] Modified (N = 5079) May 2004–Oct 2004
Antidementia drugs
N06DA02 Donepezil [ARICEPT™] Standard (N = 1762) Apr 1997–Feb 1999
Drugs used in addictive disorders
N07BA03 Varenicline [CHAMPIX™] Modified (N = 12,135) Dec 2006–March 2007
4 Contribution of UK Prescription-Based Event Monitoring Methods
51
52 D. Layton
associated with addictive behaviours such as smoking habit, obesity and chronic
non-cancer pain are also presented for purposes of demonstrating particular meth-
odological considerations. All these medicines were all intended for wide-spread,
long-term use in primary care. Other medications have also been studied (Layton
and Shakir 2011), but these are not the focus of this chapter and so will not be
described further.
The sample sizes of standard, M-PEM and SCEM studies of psychotropics
are remarkably different. The median cohort size of the 19 standard PEM studies
of psychotropic medicines presented is 11735 (IQR 9847, 12713), whilst that of
the 7 M-PEM studies is 5079 (IQR 1324,11072). One SCEM study has been
completed to date (N = 869). The explanation is related to the difference in prin-
ciple study objective: standard PEM studies were intended for general surveil-
lance with a target sample size of at least 10,000 patients to allow for the
detection of rare events occurring with a frequency of at least 1 in 2000 patients
(assuming the background rate is zero) with 85 % power (Machin et al. 1997a,
b). The lower sample size for M-PEM and SCEM reflects a bespoke study spe-
cific need, which generally requires fewer numbers than the general surveillance
studies.
4.4 Exploring Drug Utilisation Factors Within

Pharmacovigilance
Drug utilisation research is an essential part of pharmacoepidemiology, as it

describes the extent, nature and determinants of drug exposure at the patient level.
Collectively standard, M-PEM and SCEM approaches permit the examination of
the characteristics of prescriber and new drug user populations and contribute to the
accumulation of safety data, particularly with regard to vulnerable populations for
whom off-label prescribing has occurred.
There are several ways in which this may occur: by prescribing a dose in excess
of that specified by the MA, prescribing for an unlicenced indication, prescribing
for a special group outside of the MA specification and altering the dosage form.
Off-label prescribing is permitted in circumstances where a physician concludes
that for medical reasons, treatment with the product is necessary to meet the specific
needs of the patient (General Medical Council 2015).
Situations leading to a warning or precaution use include conditions to be ful-
filled before or during use, special populations at increased risk, risks associated
with starting or stopping the product and possible medication errors. These situa-
tions represent vulnerable populations that require more careful management in
order to avoid harm. They also represent areas of potential risk for pharmacovigi-
lance because the populations being exposed have not been studied, or there is lim-
ited information.
4.4.1 Off-Label Use
As an example, the summary demographic information on the study populations

identified for the four antipsychotic studies conducted using standard methods is
presented in Table 4.3 (Mackay et al. 1999). It is notable that all drugs listed first
gained MA for the treatment of schizophrenia. This is consistent with the narrowly
defined patient population studied premarketing. Also notable in Table 4.3 is the
frequency of apparent ‘off-label’ prescribing. In terms of indication, quetiapine IR
was the antipsychotic for which the proportion of use was highest (47.4 %) fol-
lowed by risperidone (44.9 %) for indications other than within the MA at launch.
The observation that quetiapine IR appeared to have the highest frequency is not
unexpected, given it was the last of the four drugs listed to be marketed.
Nevertheless, off-label prescribing in terms of indication was very common. This
observation is supported by similar findings elsewhere (Hodgson and Belgamwar
2006).
None of these products studied were indicated for the use in children <15 years.
However, the frequency of prescribing appeared to be consistently of the order of
1 %. The literature suggests that in the UK, antipsychotics are prescribed generally
for aggressive behaviour in this population (Doerry and Kent 2003). Indeed the
study for risperidone reported that of the 98 children aged <15 years, 49 (50 %) had
been prescribed risperidone for hyperactivity. For all the antipsychotics studied, the
use in the elderly was considered a special warning for use, and the proportions of
each cohort aged >65 years was consistently between 10 % and 20 %. Whilst this in
itself does not constitute off-label prescribing, the frequency of use in elderly
patients for the treatment of behavioural and psychological manifestations of
dementia is. In the majority of these studies, such prescribing was common (>1 %
of population studied) although the frequency was the highest for quetiapine (4 %).
Further discussion on the contribution of use and safety antipsychotics in patients
with dementia is given later in this chapter.
One of the limitations of standard PEM methodology was that information on
baseline characteristics, such as prior medical history, concurrent morbidities, as
well as treatment patterns was limited. M-PEM retains all the strengths of the origi-
nal method, with the same underlying process, but also tried to overcome these
limitations through bespoke targeted surveillance, enhanced data collection and
application of new analytical methods. This permits more detailed exploration of
the heterogeneity of the new user population, as well as appropriate or inappropri-
ate use.
There are a number of examples in the published literature where off-label pre-
scribing has been reported within M-PEM studies conducted on psychotropic medi-
cines. Davies et al. reported on the prescribing of modafinil (Provigil™),which was
marketed in the UK in 1998 to promote wakefulness in the treatment of narcolepsy
(Davies et al. 2013). Its licence was extended in 2004 to include chronic pathologi-
cal conditions. Following a review of the safety of modafinil (European Medicines
Table 4.3 Completed standard PEM studies of antipsychotics (Mackay et al. 1999)
54
Drug (n) Study population

MA Sex Age
Min age Male Female Mean (SD) years MA
Indication [year] years n (%a) n (%a) [Range] Indication n (%a) Comments on use in elderly and children
Risperidone (n = 7684)
Acute and chronic 15 4124/7624 3500/7624 44.1 (19.4) 3376/6129 (55.1) 98/6678 (1.5 %) <15 years; 1271/6678 (19.0 %)
schizophrenic (54.1) (45.9) [4–99] ≥65 years
psychoses, other N = 6678 Indication: 1120/6129 (18.3 %) had psychosis;
psychoses and affective 87/6129 (1.4 %) dementiab
symptoms [1993]
Sertindole (n = 462)
Schizophrenia [1996] 18 230/459 229/459 40.8 16.4) 210/311 (67.5) 2/402 (0.5 %) <18 years; 48/402 (11.9 %) ≥65 years
(50.1) (49.9) [16–95] Indication: 54/311 (17.4 %) had psychosis; 3/311
N = 402 (1.0 %) dementiab
Olanzapine (n = 8858)
Schizophrenia [1996] 18 4972/8810 3838/8810 42.2 (17.1) 3470/5902 (58.8) 81/7455 (1.1 %) <18 years; 995/7455 (13.3 %)
(56.4) (43.6) [14–97] ≥65 years
N = 7455 Indication: 1104/5902 (18.7 %) had psychosis;
47/5902 (0.8 %) dementiab
Quetiapine immediate release (n = 1728)
Schizophrenia [1997] 18 807/1725 918/1725 44.6 (19.3) 596/1133 (52.6) 20/1425 (1.4 %) <18 years; 266/1425 (18.6 %)
(46.8) (53.2) [15–98] ≥65 years
N = 1425 Indication: 215/1133 (18.9 %) had psychosis;
45/1133 (4.0 %) dementiab
MA Marketing authorisation, SD standard deviation
a
% where specified
b
Reported terms: Alzheimers, dementia, dementia senile, dementia presenile, Lewy body dementia, multifarct dementia, Pick’s disease, vascular dementia,
arteriosclerotic dementia, alcoholic dementia
D. Layton
Agency 2011), risk minimisation measures were introduced. These included updates
to the SPC to reflect nature of adverse reactions, additions to types of patients with
conditions contraindicated for use and restriction of indication to patients with shift
work sleep disorder, narcolepsy and obstructive sleep apnoea/hypopnoea syndrome.
The M-PEM study looked specifically at use post the 2004 extension in 1096
patients prescribed modafinil in primary care. Study results reported that the preva-
lence of use in multiple sclerosis was very common (n = 372, 33.9 %) and use in
children aged 16 years or under was uncommon (n = 9, 0.8 %) – both regarded as
off-label (Davies et al. 2013).
4.4.2 Contraindications and Special Warnings and Precautions

for Use
M-PEM studies were recently conducted to support a regulatory requirement to

examine the use in general medical practice in England of two novel formulations
of an opioid analgesic. Fentanyl citrate buccal tablets (Effentora™; Cephalon),
were approved in the EU on April 2008 for the management of breakthrough pain
in patients with cancer who are already receiving and who are tolerant to opioid
therapy for their underlying persistent cancer pain. Effentora™ was launched in the
UK in January 2009 (Cephalon 2010). Fentanyl pectin nasal spray (PecFent™;
Archimedes Development Ltd.), licenced for the same indication, gained marketing
approval in the EU in August 2010 and was launched in the UK in October 2010
(Archimedes Pharma UK Ltd. 2012). An objective of both studies study was to
examine the frequency of inappropriate use, i.e. use without long-term opioid ther-
apy and off-label use. Prescribing indicators were developed based on the SPC
Sects. 4.3 and 4.4. The results from both studies regarding the number of reports of
contraindications are summarised in Table 4.4. The majority of patients within
either Effentora™ or PecFent™ cohorts had no contraindications for use (n = 482,
87.5 % vs n = 56, 88.9 %). However, the remainder were contraindicated because
they were reported to be opioid naïve or not receiving maintenance opioids and had
pre-existing COPD or respiratory depression. The prevalence was similar for both
products (n = 69, 12.5 % vs n = 7, 11.1 %) albeit the study cohort size was small for
PecFent™.
These two studies also demonstrate that vulnerable sub-groups within the pri-
mary care setting can also be identified; the number of reports of patients requiring
special warnings for use are also summarised in Table 4.4. These include elderly,
those with particular conditions and or those receiving treatments that place them at
risk for important identified adverse events. In the studies of the two fentanyl prod-
ucts, nearly half of the Effentora™ or PecFent™ cohorts could be considered vul-
nerable (n = 285, 51.7 % vs n = 34, 54.0 %).
56 D. Layton
Table 4.4 Characteristics of Effentora™ or PecFent™ M-PEM cohorts

Characteristic n (% cohort) Effentora™ (N = 551) PecFent™(N = 63)
Age at start of treatment (years): median (IQR) 62 (50–72) 62 (49–73)
Sex:
Male 248 (45.1) 22 (34.9)
Female 302 (54.6) 41 (65.1)
Not specified 1 –
Indication:
Break through pain 341 (61.9) 41 (65.1)
Break through pain and other 2 (0.4) 2 (3.2)
Other indications 133 (24.1) 16 (25.4)
Not specified 75 (13.6) 4 (6.3)
Contraindications:
Opioid naïve/non-tolerant 31 (5.6) 3 (4.8)
Age <18 years 1 (0.2) 0 (0)
COPD 35 (6.4) 3 (4.8)
Respiratory depression 7 (1.3) 1 (1.6)
Respiratory failure 3 (0.5) 0 (0)
Obliterative bronchiolitis 1 (0.2) 0 (0)
Breastfeeding 0 (0) 0 (0)
MAOI use <15 days of starting (or concomitant) 1 (0.2) 0 (0)
Indicators of special warning/precautions for use:
65+ years 235 (42.6) 28 (44.4)
CKD stage >=3 43 (7.8) 7 (11.1)
Liver disease (moderate/severe) 33 (6.0) 2 (3.2)
Hypovolaemia 3 (0.5) 0 (0)
Hypotension 4 (0.7) 2 (3.2)
Bradyarrhythmia 0 (0) 0 (0)
Increased ICP 4 (0.7) 0 (0)
Pregnancy 0 (0) 0 (0)
Concomitant CYP3A4 moderate/strong inhibitors 13 (2.4) 1 (1.6)
CKD Chronic kidney disease, CYP Cytochrome P450, COPD Chronic obstructive pulmonary dis-
ease, ICP Intracranial pressure, IQR Interquartile range, MAOI Monoamine oxidase inhibitor
4.4.3 Prescriber Characteristics
Alternatively, SCEM studies permit the characterisation not only of vulnerable

patients within secondary care setting but the prescribing physicians. The
Observational Assessment of Safety in Seroquel (OASIS) study was also conducted
to support part of the PV requirements to extend the post-authorisation safety
knowledge of quetiapine extended release (Seroquel XL™), with particular empha-
sis on short-term (within 12 weeks of starting) safety during titration and at higher
doses (European Network of Centres of Excellence for Pharmacoepidemiology and
Pharmacovigilance 2014a). An internal comparator cohort of patients prescribed
Fig. 4.1 Site of OASIS

psychiatrist and patient
recruitment within UK
Government Office regions
(GOR) which define
percentage of adults
Key:
accessing specialist mental % patients aged
health services (Figure 18+ years accessing
Specialist Mental
adapted from Patterns of health services
Specialist Mental Health 3.2–3.4
Service usage in England, 3.0–3.1
2010 (Office for National
2.8–2.9
Statistics 2010))
2.5–2.7
2.3–2.4
GOR boundaries
Patients
Psychiatrists
quetiapine immediate release (IR) was collected concurrently. Seroquel XL™ was
launched in the UK in 2008 (AstraZeneca UK Limited 2014).
From over 50 trusts throughout England over 3 years from December 2009 to
December 2012, a total of 407 psychiatrists identified 948 patients who consented. Of
these 948, 869 (91.6 %) with a clinical diagnosis of schizophrenia or bipolar mania,
newly initiated on quetiapine XL or IR were eligible for inclusion and had evaluable
data for analysis. The distribution of participating psychiatrists’ trusts and consented
patients is provided in Fig. 4.1. The pattern and spread reflected density of mental
health care service utilisation in England, with the densest areas around London,
Birmingham and North West of England. The majority of psychiatrists were male
[68.9 % (275/399)] and had been qualified for 10 years or more [94.6 % (336/355)].
The most frequent area of psychiatrist specialisation, where reported, was general
psychiatry [65.4 % (193/295)]. The most frequently reported prescribing setting was
in the community (n = 83, 20.4 %) or within community mental health teams (n = 80,
19.7 %). Investigators were asked to provide the supporting reasons for prescribing
quetiapine for the primary diagnosis indicated. Reasons (other than indication) for
prescribing were also collected. Although multiple reasons for prescribing could be
reported, the most frequently provided reason for prescribing in all indications and for
the whole cohort was ‘prescriber clinical decision’ [90.2 % (767/850)].
Such information is important to understand the drivers of adoption of a new
medicine within the UK NHS. The process by which patients are referred, treated
and monitored affects the availability and accessibility of relevant information to
support safety surveillance for PV purposes. Within the UK, mental health service
provision crosses many boundaries in terms of clinical settings and the type of
healthcare professionals providing services. Only the most severely ill patients (prev-
alence estimated to be (<10 %)) (Health and Social Care Information Centre 2013)
58 D. Layton
will be admitted to hospital as an in-patient. Reports from psychiatrists participating

in OASIS indicated that seeking informed consent from such acutely ill patients was
not possible at the time of quetiapine initiation. Therefore, systematic surveillance of
this high-risk population may not be possible, and the responsibility for routine PV
activities remains with the responsible physician. In prescribing newly licenced
drugs, hospital specialist prescribers can be classified as adopters of innovation.
Generally, early adopters have been classified as opinion leaders who greatly influ-
ence the prescribing habits of others (Rogers 2004). However, they are also described
as risk takers. As such, there have been assertions that early adopters may prescribe
irrationally (Inman and Pearce 1993). This may be true for some, but further study is
needed to investigate relationships between prescriber and patient characteristics.
4.4.4 Reasons for Treatment Discontinuation
Quantifying and exploring reasons for stopping treatment are regarded as one of the
complementary and unique signal generation processes in standard, M-PEM and
SCEM. Specific data are requested in all of these studies regarding clinical or non-
clinical events that lead to treatment withdrawal. Clinical events inform on possible
issues associated with short- or long-term tolerability, whilst non-clinical event may
highlight patient adherence issues and external influences on persistence, such as
media reports of adverse effects. Early treatment discontinuation of psychotropic
medication will have a negative effect on a medications benefit risk profile, particu-
larly if discontinued before the positive effects of the psychotropic treatment can be
established. Two examples are described below.
Rimonabant (Acomplia™, Sanofi-Aventis) is a centrally acting medication first
marketed in UK in June 2006. It was licenced as an adjunct to diet and exercise for
the treatment of obese patients (BMI ≥30 kg/m2) or overweight patients (BMI >27 kg/
m2) with associated risk factors such as type 2 diabetes mellitus or dyslipidaemia. In
October 2008, the MA was suspended because of psychiatric safety concerns
(European Medicines Agency 2008). One of the objectives of the M-PEM study was
to explore utilisation and duration of treatment (Buggy et al. 2011). A supplementary
analysis (Willemen et al. 2012) examined these data according to four categories: (1)
discontinuation due to any clinical event (with a focus on psychiatric events); (2)
discontinuation due to lack of effectiveness; (3) discontinuation due to target weight
loss achieved (as a proxy for effectiveness); (4) other events reported as reasons for
stopping; and (5) stopped but no reason specified. The study results reported that
7204 (72.0 %) of patients stopped treatment, of which 50 % stopped within 1 year
(323 (IQR 279, 371)). Where specified (n = 5763, 80.0 % patients who stopped), at
least one clinical event was given as the reason for stopping for 1896 patients (32.9 %).
Of these 1896 patients, the most frequent reason was depression (n = 284 cases,
15.0 %). The most frequent non-medical reason for stopping was due to lack of effec-
tiveness was reported for 2480 (3.0 % where reason specified) patients, whilst stop-
ping due to achieving target weight loss was reported for only 215 (3.7 %) patients.
Factors that were associated with stopping (all reasons) included female gender, prior
history of antiobesity drug use and prior history of psychiatric illness. Those with a
prior history of psychiatric disease had a higher rate of stopping because of psychiat-
ric events (hazard ratio 1.8 (95 % CI 1.5, 2.09)) than those with no history.
A second example looks at reasons for and time to antipsychotic treatment discon-
tinuation. As for treatment of obesity, there are also specific determinants that affect
choice and maintenance of antipsychotic agents. Accordingly, such issues can be
explored using M-PEM data. In addition to the OASIS SCEM study with quetiapine
described earlier, an M-PEM study was conducted to further understand safety of que-
tiapine extended release (Seroquel XL™) during titration and use at higher doses as
prescribed in primary care (Gilchrist et al. 2011). Of 13,276 patients, 3753 (28.4 %)
were reported to have stopped treatment within 12 months of starting. Ten percent of
patients were no longer on treatment by day 56 and 25 % by day 220. In total 4844
events were given as reasons for stopping were specified for 3086 patients (82.2 % who
stopped within 12 months of starting). Of 1750 clinical (medical) reasons for stopping,
the most frequent was sedation (269, 15.4 %), whilst of the other 3094 non-medical
reasons given for stopping, the most frequent was drug ineffective (581, 18.8 %).
In both the Acomplia™ and Seroquel XL™ M-PEM studies, high proportions of
patients stopped treatment within 12 months of starting, and reasons for stopping
were varied. Certain patient characteristics were also associated with a higher risk
of stopping. M-PEM can provide provides information on frequency of medication
discontinuation, reasons thereof and patients who may require more intense moni-
toring. Accordingly, these data can help support physicians decisions on prescribing
and management to prevent premature treatment cessation.
4.5 Describing the Safety of Psychotropic Medicines
Psychotropicmedicines are a heterogeneous group of chemically unrelated

compounds which share a broad range of pharmacological effects, principally as a
result of broad receptor-blocking properties. These effects can involve every system
of the body, although many are non-serious. Drugs which possess anticholinergic
activity are associated adverse effects such as dry mouth, constipation, visual distur-
bances and urine hesitancy, whilst the use of drugs that affect noradrenaline activity
can, for example, cause sexual dysfunction and postural hypotension. Interference
with the dopaminergic activity can lead to endocrinological adverse effects such as
hyperprolactinaemia, conditions associated with hyperglycaemia, weight gain and
extrapyramidal symptoms. As described earlier, non-serious adverse effects play an
important role in persistence with medication. Other more serious adverse effects
include seizures, blood dyscrasias, neuroleptic malignant syndrome and enhanced
propensity for suicidal acts. Such serious effects have played an important role in
the restriction, suspension or ultimately withdrawal of some psychotropic medi-
cines. Some examples of key findings from the PEM and M-PEM studies of psycho-
tropic medicines are provided below.
In each of these examples, as described previously, the investigations are possi-
ble because the methodology (standard, M-PEM or SCEM) permits a more detailed
60 D. Layton
examination of particular risks. The underlying architecture of each study is that of

survival data – therefore treatment effect over time can be explored. Furthermore,
data are collected in an identical manner, and, for the majority of studies, cohorts of
new user subjects are identified at the same stage of a drug’s lifecycle in the imme-
diate post-marketing period. By excluding prevalent users, bias introduced by (1)
under-ascertainment of events early in therapy and (2) inability to control for dis-
ease risk factors before treatment started is minimised (Ray 2003). As such, inter-
cohort comparisons of risk and rate can be undertaken. Where no suitable comparator
may be found, intra-cohort comparisons are generally used. The limitations of PEM
and M-PEM should however be acknowledged. These include: loss of statistical
power where observed final sample size is different to that expected because of
dependency on level of prescribing of a study drug by GPs in England, missing data
arising from incomplete questionnaires or non-response and assumptions about data
for signal detection purposes. These limitations are described in further detail else-
where (Layton and Shakir 2014).
4.5.1 General Surveillance
National Institute of Clinical Excellence (NICE) guidelines for the prescribing of

antidepressants in adults advocate that choice be determined, taking into account
anticipated adverse events and concomitant medication for physical health prob-
lems (National Institute of Clinical Excellence 2009). Tricyclic antidepressants are
one of the oldest classes of antidepressants and are still used extensively. However,
these are now used second line in favour of newer antidepressants associated with a
low propensity for anticholinergic effects.
As shown in Table 4.2, ten products used in the treatment of depression have
been studied using either the standard or M-PEM approach. A study was conducted
to compare the general tolerability and safety profile of six antidepressants (fluox-
etine, sertraline, paroxetine, moclobemide, venlafaxine and nefazodone), as reported
within the PEM study conducted for each (Mackay et al. 1999). A minimum of
6 months observation was collected. Analysis included calculation of incidence
densities in (ID – per 1000 patient-months) all events, ID ratios for events of spe-
cific interest (adjusted by age, sex and indication) and odds ratios for deaths
(adjusted by age, sex and indication). The most frequently reported event consistent
across all drugs in the first month was nausea and vomiting [range 26.3 (fluoxetine)
to 71.9 (venlafaxine)], followed by malaise [range 9.9 (moclobemide) to 25.0
(nefazodone)]. The results of the comparisons for selected events are presented in
Fig. 4.2. Compared to fluoxetine (as reference group), the variability of event rates
of agitation/anxiety, drowsiness/sedation and hypertension across the different
drugs examined is evident. There were no significant differences in odds of death
between drugs after adjustment. A limitation of this particular study was that the
products were marketed sequentially between 1989 and 1996; therefore changes in
clinical practice are possible. Nevertheless, this study demonstrates the contribution
of the methodology as a whole to inform on event profiles between drugs.
Fig. 4.2 Relative risks for selected events observed in users of non-tricyclic antidepressants (ser-
traline, paroxetine, nefazodone, venlafaxine and moclobemide) compared to fluoxetine as refer-
ence in first 6 months of treatment, adjusted for age, sex and indication. Adjusted incidence density
ratios (+95 % CI) are presented for the events of agitation/anxiety, drowsiness/sedation and hyper-
tension. Adjusted odds ratios (+95 % CI) are presented for the event of death (all-cause)
4.5.2 Antipsychotics and Stroke
Initial concerns about an increased risk of cerebrovascular (CVA) adverse effects

such as stroke and transient ischemic attack initially focused on risperidone in 2002
and olanzapine in 2004 (Committee on Safety of Medicine 2004). In particular, the
number of fatalities in RCTs of elderly patients prescribed these drugs for dementia
and also in patients with no pre-existing cardiovascular problems was of concern
(Dear Healthcare Professional 2004). Prescribing restrictions were added to the
labels in the EU, US and Canada. In 2005 the European PV Working Party assessed
additional evidence from three observational studies and concluded that the risk of
CVA adverse events was possibly a class effect (European Medicines Agency 2005).
In support of this review in 2005, the DSRU conducted a retrospective analysis
to examine and compare the incidence rates of CVA and/or transient ischaemic
attack (TIA) as reported in the PEM studies conducted on risperidone, olanzapine
62 D. Layton
and quetiapine (Layton et al. 2005). Crude rate ratios (RR) and RR adjusted for age,
gender and indication (dementia vs other) were calculated using Poisson regression
modelling and time to event was modelled using survival methods. For comparative
purposes, the reference cohort was olanzapine. Person-time was calculated between
date of first dispensed prescription and date of first event or, for non-cases, censored
at end of 180 days or date of loss to follow-up, which ever occurred sooner.
When risperidone was compared to olanzapine, the age and sex-adjusted RR was
non-significant [RR1.2 (95 % CI 0.4, 3.0); n = 23 (0.3 %) vs n = 10 (0.11 %)].
Similarly, the RR was non-significant for the comparison between quetiapine and
olanzapine [RR 2.1 (95 % CI 0.6, 7.7); n = 6 (0.35 %) vs n = 10 (0.11 %)]. Likewise,
when risperidone was compared to quetiapine, the adjusted RR was not signifi-
cantly different. When data were stratified by indication, there were no cases of
CVA/TIA in the olanzapine cohort treated for indication, so this could not be exam-
ined. The age and sex-adjusted RR was non-significantly different [2.1 (95 % CI
0.5, 10.1)] between those prescribed risperidone for dementia and those prescribed
quetiapine for dementia. Of the three drugs, the time to onset was shortest for ris-
peridone and the subgroup with dementia. The intra-cohort analysis which com-
pared indication groups identified a sixfold difference in rate of CVA/TIA in patients
with dementia compared to those treated for other indications. This retrospective
comparison did not reveal any difference in rates between the three antipsychotics
but did support the observations that dementia is an important risk factor.
Irrespective of treatment, dementia is associated with a two- to fourfold increased
risk of mortality compared to those without such diagnosis (Xie et al. 2008). Yet,
despite additional regulatory activity since 2005, and the introduction of best
practice guidelines for rational and safe use of antipsychotics in people with
dementia, there remain concerns that antipsychotic medications are being overpre-
scribed as first line choice of pharmacological management of BPSD. In the UK,
estimates from the published literature range from 5 % to 15 %, despite limited
clinical effectiveness (IMS Health 2009; Child et al. 2012). In the UK, NICE recom-
mends that antipsychotics can be used in elderly patients, but under strict guidelines
(Oxford Health NHS Foundation Trust 2013). Risperidone is recommended as the
antipsychotic of first choice, plus a low dose, with slow dose titration is advocated.
4.5.3 Cardiovascular Safety
Sibutramine is a serotonin-noradrenaline re-uptake inhibitor, indicated for the treat-

ment of obesity. It was licenced in the UK in May 2001. A safety signal regarding
adverse cardiovascular effects was reported shortly after marketing and a review con-
ducted by the European Medicines Agency in 1999 and 2002 in light of these con-
cerns. At that time, CHMP concluded that the benefits of sibutramine for the
management of obsess and overweight patients outweighed its risks (European
Medicine Agency 2002). However the MAH was asked to start a study of use of
sibutramine in patients with cardiovascular risk factors – the Sibutramine
Cardiovascular Outcome Trial (SCOUT) study. This study started in 2002; its aim
was to determine the impact of weight loss with sibutramine on cardiovascular prob-
lems in a large group of overweight and obese patients at high risk for cardiovascular
disease.
During that period, a PEM study of sibutramine was undertaken between October
2001 and June 2002 (Perrio et al. 2007). This study provided information on the
‘real-world’ use of sibutramine in general practice, irrespective of co-morbidities and
concomitant use of other medicines. As the prescribing recommendations already
had contraindications and precaution for use in patients with cardiovascular risk fac-
tors, prescribers were asked to provide additional information specifically regarding
prior medical history of diabetes mellitus, hypertension or ischaemic heart disease,
plus prior use of other antiobesity drugs. Patient characteristics, events and reasons
for stopping were examined according to the prescribing guidelines. The full details
of this analysis are published elsewhere. In summary, the PEM cohort comprised
12,336 patients. In terms of cardiovascular risk factors, the prevalence of pre-existing
IHD was 1.9 % (n = 238), this being a contraindication for use. Three cardiovascular
events were considered signals associated with starting treatment (cardiovascular
tests, faintness and palpitations). Within 3 months of starting treatment, 5157 patients
(41.8 % of total cohort) had stopped taking sibutramine. Of 5280 reasons specified
for 4554 patients (88.3 % who stopped within 3 months), the events of hypertension
and raised blood pressure were commonly reported as reasons for stopping (n = 203
(3.8 % of reasons) and n = 179 (3.4 % of reasons), respectively), whilst palpitation,
disorders of heart rate, chest pain and raised pulse were uncommon. Causality assess-
ment of reports of arrhythmia, cerebrovascular events, angina and myocardial infarc-
tion identified 14 events assessed as possibly (n = 13) or probably (n = 1) related to
sibutramine: ‘arrhythmia’ (n = 3), ‘fibrillation atrial’ (n = 4), angina (n = 4) syncope
(n = 2) and myocardial infarction (n = 1). Although there is limited information on
other risk factors for cardiovascular safety, this study gave an indication of how par-
ticular safety aspects highlighted within prescribing recommendations, including the
identification of vulnerable populations at risk can be investigated.
Notably in January 2010, the European Agency’s Committee for Medicinal
Products for Human (CHMP) recommended suspension of the marketing authorisa-
tion of medicinal products containing sibutramine, following a safety review of
interim data from the SCOUT study which indicated that sibutramine is associated
with more cardiovascular problems than placebo (European Medicines Agency
2010; Maggioni 2009).
4.5.4 Sudden Unexpected Deaths
PEM study data has also been used to investigate cardiovascular concerns including
sudden unexpected deaths for an atypical antipsychotic sertindole (Serdolect™)
(Wilton et al. 2001). This product was voluntarily suspended in the EU in 1998, fol-
lowing regulatory concerns over reports of serious cardiac dysrhythmias and sudden
64 D. Layton
unexpected deaths. The reported causes of death, their frequency, prolongation of

the rate corrected QT interval (QTc) and cardiac dysrhythmias in patients within the
sertindole PEM cohort were compared with PEM data obtained those for patients
treated with two other atypical antipsychotics (olanzapine and risperidone) (Wilton
et al. 2001). Age and gender Standardise Mortality Ratios (SMRs) (+95 % CI) were
calculated using the indirect method (dos Santos Silva 1999). There was no statisti-
cally significant difference in all-cause mortality rates between sertindole and the
comparator cohort according to the SMR (0.87 95 % CI; 0.35, 1.80) based on n = 7
vs 397 deaths. Similarly, there was no statistically significant difference in CV mor-
tality (SMR 0.73 (0.009, 2.63)) based on n = 2 vs n = 133 CV deaths, though confi-
dence intervals were wide due to small numbers in the sertindole cohort. Six cases
of prolongation of QTc interval were identified in 462 patients (1.3 %, 95 % CI;
0.5 %, 2.8 %) treated with sertindole and one with unspecified electrocardiogram
changes in the comparator cohort of 16,542 patients. Although no statistically sig-
nificant difference was shown in mortality rates between sertindole and comparator
cohort, the sertindole cohort was too small to rule out an association between the
use of this drug and cardiovascular deaths.
This example shows how calculations of SMRs area an alternative method for
inter-cohort comparisons when an external comparator cannot be identified. In this
example, because the mortality rate in schizophrenic patients was known to be dif-
ferent to that of the general population (Brown 1997), national mortality figures
were not considered appropriate for comparison purposes. Therefore, data from the
cohorts of two PEM studies of other atypical antipsychotic drugs were considered
most appropriate to calculate the ‘expected’ mortality rates because the same meth-
odology had been used for all three studies.
4.6 Methodological Developments in Design to Enhance

Evaluation of Safety Concerns
4.6.1 Treating Dose Flexibly
When initiating pharmacological treatments for mental health conditions, prescrib-

ing guideline recommend individualisation of the dosing regimen. This in turn
introduces added complexity in exploring associations between dose and an event
because dose may be rapidly changing over time. Traditional approaches to model-
ling such associations need to extend beyond simple categorical measures based on
arbitrary values at a fixed time point (such as start dose) and assumptions of homo-
geneity within each stratum (Greenland 1995). In studies of psychotropic medi-
cines, particularly antipsychotics, nonlinear dose patterns should be anticipated
with appropriate analytical methods to describe such patterns of change. As
described earlier, targeted questions within M-PEM and SCEM design permit infor-
mation on dose to be collected at multiple intervals during the observation period
for each individual patient. In treatments where sub- or supra-optimal dose regimes
may significantly affect the benefit: risk ratio, knowledge of possible changes risk
of clinical or adherence outcomes due to variation in dynamics of dose over time are
extremely important.
Such methods to analyses repeated dose measurement over time in M-PEM and
SCEM studies are being explored. One of the objectives of the OASIS SCEM study
was to examine posology and titration patterns over the 12-week observation period
(European Network of Centres of Excellence for Pharmacoepidemiology and
Pharmacovigilance 2014a). The relationship of dose with time, by formulation, was
explored by such methods; however the results have not yet been published. This
type of analysis can be used to model general trend at group level and permits sim-
ple univariate explorations of associations between dose, predictors and outcomes
in defined periods for signal detection purposes.
In addition to the challenge of modelling changes of dose over time, there is a
need to be able to better account for multiple combinations of treatment or where
multiple indications exist, particularly as seen in psychiatric medicine. In the UK,
the Ready Reckoner is a tool used in clinical practice to monitor antipsychotic doses
in high-risk patients with complex dosing regimens. It was developed by the UK
Prescribing Observatory for Mental Health (POMH-UK), based within the Centre
for Quality Improvement at the Royal College of Psychiatrists (Prescribing
Observatory for Mental Health 2015). The principal is that at any stage of treatment,
the total daily dose of each individual antipsychotic is converted to a percentage of
the maximum dose for the indication being treated. Where the total sum of percent
exceeds 100 %, then the patient is identified as a high-risk patient that requires more
intensive monitoring (Royal College of Psychiatrists 2006). The potential advan-
tages of modelling dose using this alternative metric include automatic adjustment
by indication and loss of power due to creation of multiple strata of small sample
size. The limitations include possible underestimating where dose data are missing,
or concomitant medications are not reported. This Ready Reckoner is currently
being applied to model antipsychotic treatment effects using M-PEM data.
4.6.2 Modelling Patterns of Events
In this chapter, examples have been provided of signals generated through quantita-
tive means, such as calculation of crude incidence rates and risks. It is acknowl-
edged in signal detection that the artificial segregation of time periods may not be
appropriate for all events with respect to their most relevant time periods of excess
(Guess 2006). More recently survival methods have been used to support signal
detection and strengthening in M-PEM to more appropriately model the hazard
function with the aim of detecting adverse drug reactions (Cornelius et al. 2012).
Both non-parametric and parametric regression modelling of time to event data are
considered to be useful adjuncts to signal detection methods and been used to
explore safety signals with M-PEM and SCEM studies. A key difference between
the two modelling approaches is that parametric methods allow more flexibility
66 D. Layton
when the associated hazard is not (expected to be) constant with respect to time. The
construction of the M-PEM cohort is suitable for modelling time to event because
the time origin is unambiguously defined (prescription start date), there is a metric
for measuring time (person-time exposed), and failures are defined according to
events. Since study data are collected over a finite period of time, such that not all
‘time to event’ may be observed for all patients, the study data are right censored.
This means that the distribution of time to event cannot be described usual summary
statistics because such sample parameters may no longer be unbiased estimators of
the population parameter.
One example was the evaluation of the association between neuropsychiatric
symptoms (including depression, suicidal thoughts and behaviour) and use of var-
enicline (Champix™) – indicated for the treatment of smoking cessation in adults
(≥18 years of age) (Buggy et al. 2013). This study was conducted as part of the
M-PEM study in response to the regulatory warnings issued in 2008 (Medicines and
Healthcare products Regulatory Agency 2008). It was known that patients with a
prior history of psychiatric disorders or users of psychotropic medicines were
excluded from premarketing clinical trials. Therefore, the possibility existed that
prior history of psychiatric disease could have had a profound effect on risk of neu-
ropsychiatric events after starting treatment. For five events (depression, anxiety,
aggression, suicidal ideation and nonfatal self-harm), analysis included using a the
semi-parametric method to provide a smoothed estimate of the empirical hazard
function and the parametric Weibull model to test for non-constant hazard using the
shape parameter estimate in the first 3 months after starting treatment (Cornelius
et al. 2012). Events reported on prescription start date were excluded, as they were
considered pre-existing. The shapes of the smoothed hazards for depression and
anxiety were suggestive of a non-constant hazard over time, but this was not
observed for aggression, suicidal ideation and nonfatal self-harm. The correspond-
ing predicted hazard function shape parameter estimated using the Weibulll model
suggested that the hazard of anxiety increased over time (p = 0.009), whilst there
was no evidence of change for the other four events. In this particular example, the
parametric model was crude, and the purpose was simply to detect whether hazards
for prespecified events were non-constant or not, and not to model hazards accu-
rately. Further exploration of the feasibility of integrating age and sex-adjusted
parametric regression models as an additional tool for general surveillance purposes
to support the identification of multiple safety signals within M-PEM studies is cur-
rently underway (Layton and Kimber 2014).
4.6.3 Assessing Suicidality
Assessing causal relationships between a drug and a suicidal event is complicated

by the high background rate of such events in the treated populations and the high
prevalence of known risk factors for Suicidal Behaviour (SB). Furthermore, a lack
of well-defined terminology in clinical practice and failure to correctly identify sui-
cidal behaviour and ideation has negative implications on appropriate management
of suicidality by health care professionals as well as decreasing reliability in com-

parisons of rates of such events across studies as a result of misclassification. Tools
to support identification in PASS are vital to help quantify incidence in at-risk
groups. The FDA has endorsed standard suicide terms and the application of the
Columbia-Classification Algorithm for Suicide Assessment (C-CASA) approach to
identify and categorise suicidal events in clinical trials (Posner et al. 2007), but its
application in PASS has been limited. The C-CASA instrument includes nine codes
within three categories (suicidal events, indeterminate or potentially suicidal events
and non-suicidal events) that assist in classifying and distinguishing those events
representing suicidal intent from to those without. These domains have been
expanded and incorporated into a new instrument used to prospectively assess
Suicidal Ideation (SI) and SB in clinical trials, the Columbia Suicide Severity Rating
Scale (C-SSRS) with five subtypes (Posner et al. 2011). Although the C-CASA and
C-SSRS criteria were both developed for the use in a different population in clinical
trials, the instruments were considered relevant to the exploratory objective of four
PASS studies for two antipsychotics [Seroquel XL™ and asenapine (Sycrest™)
(European Network of Centres of Excellence for Pharmacoepidemiology and
Pharmacovigilance 2013, 2014b)] since reports of suicidal events were anticipated.
Since there is uncertainty in predicting which patients are at risk, data are being col-
lected on risk factors possibly associated with an increased risk of suicidality.
Potential suicide events are being classified according to C-CASA expanded
C-SSRS domains by DSRU clinicians. As of January 2015, the studies for asenap-
ine are ongoing, whilst the Seroquel XL™ studies are complete, but data are not yet
published.
Such systematic assessment of suicidal events using C-CASA/C-SSRS domains
can help support post-marketing benefit: risk evaluation of treatment choices in cat-
egories of patients according to risk of suicide when starting treatment with psycho-
tropic medicines.
4.6.4 Indicators of Drug-Seeking Aberrant Behaviour
Problematic prescription drug use includes misuse (‘non-medical use’) and addic-
tion as well as unsanctioned diversion to third parties and is reflected by or associ-
ated with drug-seeking aberrant behaviours. However, important research gaps
include missing information on the incidence of such events and the patients likely
to be at high risk of dependency. Such data are needed in order to support the effec-
tiveness of risk minimisation strategies for psychoactive medications aimed at miti-
gating problematic use, particularly in patients for whom long-term treatment may
be necessary. The feasibility of estimating the prevalence of risk factors for depen-
dence and aberrant behaviours inpatients prescribed psychotropic products with
misuse potential was first explored in the M-PEM post-study for Effentora™ in
response to a regulatory requirement (Osborne et al. 2014). A number of criteria and
instruments used in clinical practice in various setting were examined. Surrogate
markers of indicators of aberrant behaviour suggestive of addiction were proposed
68 D. Layton
Table 4.5 Indicators Overwhelming focus on opioid-related issues

of aberrant behaviours
Escalating drug use (early refills/larger amounts for longer
during treatment
periods) unexplained by change in clinical condition
Reports lost, spilled stolen medication
Unclear aetiology and/or exaggeration of pain
Requests for treatment from multiple prescribers
Accidental/unsanctioned diversion to third parties
based on the Chabal criteria (Chabal et al. 1997) that reflected behavioural rather
than clinical manifestations (Table 4.5).
In addition, information on known risk factors strongly associated with sub-
stance dependence was collected on prior history of: psychiatric disorders, sub-
stance misuse, alcohol misuse and smoking. Information was also requested on
clinical diagnosis of opioid withdrawal syndrome within 14 days of starting treat-
ment since its manifestation would suggest pre-existing dependence. Simple (non-
weighted) risk scores were constructed on aggregate counts of the indicators of
dependence and aberrant behaviours (score >3 suggested a patient to be at high
risk). In the Effentora™ M-PEM study, the prevalence of at least one pre-existing
risk factor for dependence was 26 % (n = 145), whilst the frequency of aberrant
behaviours observed during treatment was 8 % (n = 46). Patients with aberrant
behaviours appeared to have different characteristics to those who did not.
Such reports do not confirm misuse but are potential signals of such. Since the
systematic collection of physician reports of aberrant behaviours has shown to be
feasible, these criteria have also been applied to the study of other psychotropic
medicines using M-PEM and SCEM.
4.7 Final Remarks
This chapter describes the usefulness of standard, M-PEM and SCEM methodology
in revealing insight into important characteristics of users of psychotropic medicines,
including vulnerable populations, and prescribing patterns as well as providing further
information on important safety issues. Through examples, a demonstration has been
presented on how the methodology of the prescription-based event-monitoring sur-
veillance system and pharmacovigilance as a whole has evolved and still is evolving.
References
Archimedes Pharma UK Ltd (2012) Summary of product characteristics: PecFent. eMC Available
from: http://www.medicines.org.uk/emc/medicine/23962
AstraZeneca UK Limited (2014) Seroquel XL 50 mg, 150mg, 200 mg, 300 mg, 400 mg prolonged-
release tablets: summary of product characteristics. Ref type: Pamphlet
British Medical Association Board of Science, British Medical Association Science & Education
(2006) Reporting adverse drug reactions. A guide for healthcare professionals. Available at
: http://bmaopac.hosted.exlibrisgroup.com/exlibris/aleph/a21_1/apache_media/
GYVFNJ1RT2PFNDUJ8IUKPUNTFVRF8G.pdf. London
Brown S (1997) Excess mortality of schizophrenia. A meta-analysis. Br J Psychiatry 171:
502–508
Buggy Y, Cornelius V, Wilton L, Shakir SA (2011) Risk of depressive episodes with rimonabant:
a before and after modified prescription event monitoring study conducted in England. Drug
Saf 34(6):501–509
Buggy Y, Cornelius V, Fogg C, Kasliwal R, Layton D, Shakir SA (2013) Neuropsychiatric events
with varenicline: a modified prescription-event monitoring study in general practice in England.
Drug Saf 36(7):521–531
Cephalon (2010) Summary of product characteristics: Effentora 100, 200, 400, 600 and 800 micro-
grams buccal tablets. eMC Available from: http://www.medicines.org.uk/EMC/medi-
cine/21401/SPC/Effentora+100%2c+200%2c+400%2c+600+and+800+micrograms+buccal+t
ablets/
Chabal C, Erjavec MK, Jacobson L, Mariano A, Chaney E (1997) Prescription opiate abuse in
chronic pain patients: clinical criteria, incidence, and predictors. Clin J Pain 13(2):150–155
Child A, Clarke A, Fox C, Maidment I (2012) A pharmacy led program to review anti-psychotic
prescribing for people with dementia. BMC Psychiatry 12:155
Committee on Safety of Medicine LD (2004) Atypical antipsychotic drugs and stroke. Web
address:http://www.mhra.gov.uk/news/antipsystroke_9304.htm. Ref type: Internet
Communication
Cornelius VR, Sauzet O, Evans SJ (2012) A signal detection method to detect adverse drug reactions
using a parametric time-to-event model in simulated cohort data. Drug Saf 35(7):599–610
Council for International Organizations of Medical Sciences(CIOMS), World Health Organisation
(WHO) (2002) International ethical guidelines for biomedical research involving human
subjects. [online] Available at: http://www.cioms.ch/publications/layout_guide2002.pdf.
Geneva
Davies M, Wilton L, Shakir S (2013) Safety profile of modafinil across a range of prescribing
indications, including off-label use, in a primary care setting in England: results of a modified
prescription-event monitoring study. Drug Saf 36(4):237–246
Dear Healthcare Professional (2004) The risk of cerebrovascular adverse events (CVAE) in elderly
patients with dementia:- important changes to the summary of product characteristics relating
to use of risperidone. Janssen-Cilag Ltd., High Wycombe
Department of Health (2014) NIHR Clinical Research Network Support. Available at: http://www.
nihr.ac.uk/about/clinical-research-network.htm. Date Accessed: 17/11/2014
Doerry UA, Kent L (2003) Prescribing practices of community child and adolescent psychiatrists.
Br J Psychiatry Bull 27:407–410
European Commission(2012) Question and answers on transitional arrangements con-
cerning the entering into force of the new pharmacovigilance rules provided by direc-
tive 2010/84/EU amending directive 2001/83/EC and regulation (EU) no 1235/2010
amending regulation (EC) no 726/2004. Available from: http://ec.europa.eu/health/files/
pharmacovigilance/2012-07_qa_transitional_en.pdf
European Medicines Agency (2002) Opinion of the committee for proprietary medicinal products
opinion following an Article 31 referral sibutramine. Doc Ref: EMEA/CPMP/4514/02/Final
Available at: http://www.ema.europa.eu/docs/en_GB/document_library/Referrals_document/
Sibutramine_31/WC500013923.pdf)
European Medicines Agency (2005) European pharmacovigilance working party, 2005.
Antipsychotics and cerebrovascular accident. Available at: www.mhra.gov.uk/home/
idcplg?IdcService=GET_FILE&dDocName=CON2024914&RevisionSelectionMethod
European Medicines Agency (2008) The European Medicines Agency recommends suspension of
the marketing authorisation of Acomplia. Doc Ref: EMEA/CHMP/537777/208. Available at:
http://www.ema.europa.eu/docs/en_GB/document_library/Press_release/2009/11/
WC500014774.pdf. Date accessed: 30/01/2015
70 D. Layton
European Medicines Agency (2010) Questions and answers on the suspension of medicines
containing sibutramine Doc Ref: EMA/808179/2009rev Available at http://www.ema.europa.
eu/docs/en_GB/document_library/Referrals_document/Sibutramine_107/WC500094238.pdf
European Medicines Agency (2011) Press release. European Medicines Agency recommends
restricting the use of modafinil. Doc Ref: EMEA/459173/2010 Web address: http://www.ema.
europa.eu/docs/en_GB/document_library/Press_release/2010/07/WC500094976.pdf. Date
accessed: 31/01/2015. The European Medicines Agency (EMEA)
European Medicines Agency (2012) Guideline on good pharmacovigilance practices (GVP) annex
I – definitions EMA/876333/2011 (Rev 1)
European Medicines Agency (2013) Guideline on good pharmacovigilance practices (GVP) mod-
ule VIII – post-authorisation safety studies (Rev 1). EMA/813938/2011
European Medicines Agency (2014) Guideline on good pharmacovigilance practices (GVP): mod-
ule V – risk management systems (Rev 1)
European Network of Centres of Excellence for Pharmacoepidemiology and Pharmacovigilance
(ENCEPP) (2013) E-register of studies. An observational post-authorization modified
prescription-event monitoring safety study to monitor the safety and utilization of asenapine
(Sycrest) in the primary care setting in England. Ref type: Online Source
(ENCEPP) (2014a) E-register of Studies. A Cohort study to monitor the safety and use of
prolonged-release quetiapine. Ref type: Online Source
(ENCEPP) (2014b) E-register of studies. An observational post-authorisation safety specialist
cohort monitoring study (SCEM) to monitor the safety and utilisation of asenapine (Sycrest) in
the mental health trust setting in England. Ref type: Online Source
Finney DJ (1965) The design and logic of a monitor of drug use. J Chronic Dis 18:77–98
General Medical Council (2009a) Confidentiality. Web address: http://www.gmc-uk.org/static/
documents/content/Confidentiality_0910.pdf Date accessed: 08/07/2011
General Medical Council (2009b) Confidentiality: disclosing information for education and
training purposes. http://www.gmc-uk.org/Confidentiality_disclosing_info_education_2009.
pdf_27493403.pdf
General Medical Council (2015) Good practice in prescribing and managing medicines and
devices. Web address: http://www.gmc-uk.org/Prescribing_guidance.pdf_59055247.pdf
Gilchrist A, Davies M, Layton D, Shakir SAW (2011) Abstract no: 773. Drug utilisation data for
Seroquel XL – results of a first interim modified PEM study report abstract no: 773 27th
international conference on pharmacoepidemiology & therapeutic risk management.
14th–17th Aug 2011 -Chicago
Greenland S (1995) Dose-response and trend analysis in epidemiology: alternatives to categorical
analysis. Epidemiology 6(4):356–365
Guess HA (2006) Exposure-time-varying hazard function ratios in case-control studies of drug
effects. Pharmacoepidemiol Drug Saf 15(2):81–92
Hodgson R, Belgamwar R (2006) Off-label prescribing by psychiatrists. Psychiatr Bull 30:55–57
Ilyas S, Moncrieff J (2012) Trends in prescriptions and costs of drugs for mental disorders in
England, 1998–2010. Br J Psychiatry 200(5):393–398
IMS Health (2009) The use of medicines in dementia care. IMS, London
Inman W, Pearce G (1993) Prescriber profile and post-marketing surveillance. Lancet
342(8872):658–661
Layton D, Kimber A (2014) Abstract no: 802. Feasibility of adjusted parametric methods to model
survival data as a tool for signal strengthening: an example in Modified Prescription-Event
Monitoring (M-PEM). 2014 30th international conference on pharmacoepidemiology & ther-
apuetic risk management (ISPE). 24th–27th Oct 2014 – Teipei. Pharmacoepidmiology Drug
Safety (S1), pp 1–497
Layton D, Shakir SAW (2011) Prescription-event monitoring. In: Strom BL, Kimmel SE, Hennessy
S (eds) Pharmacoepidemiology, 5th edn. Wiley, Chichester, pp 301–330
Layton D, Shakir SAW (2014) Prescription-event monitoring (PEM): the evolution to the new
modified PEM and its support of risk management. In: Mann RD, Andrews EB (eds)
Pharmacovigilance, 3rd edn. Wiley Blackwell, Chichester, pp 359–384
Layton D, Shakir SA (2015) Specialist cohort event monitoring studies: a new study method for
risk management in pharmacovigilance. Drug Saf 38(2):153–163
Layton D, Harris S, Wilton LV, Shakir SA (2005) Comparison of incidence rates of cerebrovascu-
lar accidents and transient ischaemic attacks in observational cohort studies of patients
prescribed risperidone, quetiapine or olanzapine in general practice in England including
patients with dementia. J Psychopharmacol 19(5):473–482
Lenz W, Pfeiffer R, Kosenow W, Hayman D (1962) Thalidomide and congenital abnormalities.
Lancet 279(7219):45–46
Machin D, Campbell M, Fayers P et al (1997a) Sample size tables for clinical studies. Table 7.1.
Blackwell Science Ltd, Oxford
Machin D, Campbell M, Fayers P et al (1997b) Sample size tables for clinical studies. Table 7.2.
Blackwell Science Ltd, Oxford
Mackay FR, Dunn NR, Martin RM, Pearce GL, Freemantle SN, Mann RD (1999) Newer antide-
pressants: a comparison of tolerability in general practice. Br J Gen Pract 49(448):892–896
Maggioni AP (2009) SCOUT trial reports on the safety profile of sibutramine in patients with
cardiovascular diseases. Phys Sportsmed 37(3):95–97
Measures of occurrence of disease and other health-related events (1999) In: dos Santos Silva I
(ed) Cancer epidemiology: principles and methods. Lyon: International Agency for Research
on Cancer Press, pp 57–82
Medicines and Healthcare products Regulatory Agency (2008) Varenicline: suicidal thoughts and
behaviour. Drug Safety Update 1(12):2–3
National Health Service, The Information Centre (2009) Adult psychiatric morbidity in England,
2007. Results of a householde survey Available at: http://www.hscic.gov.uk/catalogue/
PUB02931/adul-psyc-morb-res-hou-sur-eng-2007-rep.pdf. Date accessed: 31/01/2015
National Institute of Clinical Excellence (2009) Clinical guideline 90: depression in adults: the
treatment and management of depression in adults. Department of Health, London
Office for National Statistics (2010) Patterns of specialist mental health service usage in England.
Available at: http://www.neighbourhood.statistics.gov.uk/HTMLDocs/images/Mental-Health-
Service-usage-small-analysis1_tcm97-97434.pdf. Date accessed 31/01/2015
Osborne V, Layton D, Fogg C, Al-Shukri M, Shakir SA (2014) Utilization of fentanyl buccal tab-
lets in England: exploring off-label use reported in a cohort study. Eur J Pain 18(4):506–512
Oxford Health NHS Foundation Trust (2013) Prescribing antipsychotic medications for the
treatment of behavioural symptoms in people with dementia, Milton Keynes 2012. NICE,
London
Perrio MJ, Wilton LV, Shakir SA (2007) The safety profiles of orlistat and sibutramine: results of
prescription-event monitoring studies in England. Obesity (Silver Spring) 15(11):2712–2722
Posner K, Oquendo MA, Gould M, Stanley B, Davies M (2007) Columbia Classification Algorithm
of Suicide Assessment (C-CASA): classification of suicidal events in the FDA’s pediatric sui-
cidal risk analysis of antidepressants. Am J Psychiatry 164(7):1035–1043
Posner K, Brown GK, Stanley B, Brent DA, Yershova KV, Oquendo MA et al (2011) The
Columbia-Suicide Severity Rating Scale: initial validity and internal consistency findings from
three multisite studies with adolescents and adults. Am J Psychiatry 168(12):1266–1277
Prescribing Observatory For Mental Health (POMH – UK) (2015) The antipsychotic dosage
‘Ready Reckoner’. Royal College of Psychiatrists. Ref type: Online Source
Ray WA (2003) Evaluating medication effects outside of clinical trials: new-user designs. Am J
Epidemiol 158(9):915–920
Rogers EM (2004) A prospective and retrospective look at the diffusion model. J Health Commun
9(S 1):13–19
Royal College of Psychiatrists (2006) Council Report CR138. Concensus statement on high-dose
antipsychotic medictaion. London, Royal College of Psychiatrists. Ref type: Pamphlet
72 D. Layton
Singh S, Loke YK (2012) Drug safety assessment in clinical trials: methodological challenges and
opportunities. Trials 13:138
Health and Social Care Information Centre (2013) Mental health bulletin: annual report from
MHMDS returns – England 2011–12, initial national figures
Willemen MJ, Mantel-Teeuwisse AK, Buggy Y, Layton D, Straus SM, Leufkens HG et al (2012)
Reasons for and time to discontinuation of rimonabant therapy: a modified prescription-event
monitoring study. Drug Saf 35(12):1147–1158
Wilton LV, Heeley EL, Pickering RM, Shakir SA (2001) Comparative study of mortality rates and
cardiac dysrhythmias in post-marketing surveillance studies of sertindole and two other atypi-
cal antipsychotic drugs, risperidone and olanzapine. J Psychopharmacol 15(2):120–126
World Health Organisation (WHO) (2015) Collaborating centre for drug statistics methodology.
ATC/DDD Index 2015. Ref type: Online Source
Xie J, Brayne C, Matthews FE (2008) Survival times in people with dementia: analysis from popu-
lation based cohort study with 14 year follow-up. BMJ 336(7638):258–262
Chapter 5
The Role of Healthcare Databases
in Pharmacovigilance of Psychotropic Drugs
Gianluca Trifirò and Janet Sultana
Abstract Recent years have seen a rapidly growing number of healthcare data-
bases that have been used worldwide for post-marketing assessment of use and
safety of medicines including psychotropic drugs, as the result of the conversion of
healthcare data storage systems from paper-based to electronic formats. Currently
used databases include general practitioners’ electronic medical records as well as
administrative/claims healthcare databases which are, respectively, used for regis-
tering clinically relevant information of patients during routine care and for docu-
mentation of healthcare services provided to citizens for reimbursement reasons. In
both cases data are collected irrespective of any research purpose. Nevertheless,
secondary use of these data sources allowed the evaluation of prescribing pattern of
and safety outcomes associated with a number of medicines widely used in psychia-
try routine care such as antidepressants, antipsychotics (especially when used in
older people with dementia), benzodiazepines, cholinesterase inhibitors, and others.
In these databases, the collection of many clinical details for large populations and
long follow-up period offers the opportunity to investigate even rare adverse events
potentially associated to psychotropic drugs. On the other hand, observational data-
base studies have inherent limitations including confounding by indication, proto-
pathic bias, outcome, and exposure misclassification which need to be taken into
account and addressed in the study design and analysis. In this chapter, examples of
observational studies investigating the use and safety of drugs commonly used in
psychiatry and limitations and solutions adopted in these studies are provided and
discussed.
Keywords Healthcare databases • Observational studies • Drug utilization research

• Drug safety studies • Healthcare interventions • Pharmacovigilance • Psychotropic
drugs
G. Trifirò (*) • J. Sultana

e-mail: trifirog@unime.it; jaysultana@gmail.com

DOI 10.1007/978-3-319-24741-0_5
74 G. Trifirò and J. Sultana
5.1 Introduction
Recent years have seen a rapidly growing number of observational studies that
explored emerging safety issues associated with psychotropic drugs through the use
of electronic healthcare databases. The increase in the utilization of healthcare data-
bases worldwide is due to the conversion of healthcare data storage systems from
paper-based to electronic formats as well to improved skills in automatic patient-
level data extraction and elaboration. Administrative healthcare data such as data
registered for health insurance purposes that may have been underused in the past is
now often collected with great attention and subjected to scrutiny in the form of
audits.
The type of information recorded in a healthcare database depends on the reason
for which the database was set up. For example, data recorded in general practice
databases during daily routine care is more likely to contain details of medical his-
tory, including information that is not associated with the use of healthcare resources,
such as lifestyle information (e.g., cigarette smoking, alcohol use, medical diagno-
ses that do not require pharmacotherapy, medical procedures or hospitalization,
etc.). In contrast, this type of data is unlikely to be registered in health insurance/
claims databases, which are instead more likely to have accurately registered codes
related to healthcare resource utilization (e.g., emergency department visit, hospital
discharge diagnoses, outpatient diagnostic tests), as this is required for billing pur-
poses. Using a (generally anonymized) unique patient identifier, several claims
databases can be linked together and even general practice databases can be linked
to claims databases as well as disease registries and surveys capturing information
of the same catchment area while not violating any country-specific data privacy
issues, thus enormously enlarging the potential of these sources for pharmacoepide-
miology research.
5.2 Healthcare Databases
5.2.1 General Practice Databases
General practice (GP) databases have been extensively used in pharmacoepidemio-

logical research. Examples of GP and other databases commonly used for pharma-
coepidemiology studies are found in Table 5.1. GP databases are composed of
electronic patient data (i.e., electronic medical records) registered during routine
care; the purpose of such records is primarily related to healthcare (e.g., recording
of patient medical and drug history for healthcare practitioner’s perusal). In general,
GP databases contain an extensive amount of information such as medical diagno-
sis, drug prescription data, and laboratory data. Data on smoking and socioeco-
nomic status and obesity are not systematically recorded, even though in several
countries educational interventions have been promoted to increase GP awareness
5
Table 5.1 Examples of commonly used general practice and family pediatrician databases used for safety studies of psychotropic drugs
CPRD THIN IPCI BIFAP HSD Arianna Pedianet
Country/region UK UK The Netherlands Spain Italy Italy Italy
Population General population, including adults and children Pediatric and Pediatric
adult population population
Disease codes used ICD-10 CM and READ codes ICPC codes ICD-10 ICD-9 ICD-9 CM ICD-9 CM codes
READ codes codes CM codes
codes
Drug codes used BNF/Multilex codes BNF/Multilex codes ATC codes ATC ATC ATC codes ATC codes
codes codes
Abbreviations: ATC anatomic and therapeutic classification, BIFAP Base de datos Informatizada para estudios Farmacoepidemiologicós en Atención Primaria
database, BNF British National Formulary, CPRD Clinical Practice Research Datalink database, HSD Health Search Database, ICD-10 CM 10th International
Classification of Disease codes, clinical modification, ICD-9 CM 9th International Classification of Disease codes, Clinical Modification, ICPC International
Classification for Primary Care, IPCI Integrated Primary Care Information database, THIN The Health Improvement Network database, UK United Kingdom
The Role of Healthcare Databases in Pharmacovigilance of Psychotropic Drugs
75
about benefits related to the registration of information on lifestyle of patients

affected by specific chronic diseases (e.g., body mass index and cigarette smoking
in patients with chronic obstructive pulmonary disease or history of myocardial
infarction). GP databases used in pharmacoepidemiology research are often nation-
wide and representative of the geographical area where they are based and, although
they are expensive to be initially set up, they are often cost-effective when used for
research purposes. This is especially true for retrospective studies where the
expenses related to data collection are significantly curtailed.
Electronic medical records from general practitioners are a relatively recent
development, mainly due to the increasing computerization of health records in
general practice, which is gradually replacing paper medical charts. These data-
bases have several advantages, the most apparent of which is their large size: popu-
lations sampled from those databases can be in the order of millions, as with CPRD
and THIN. This advantage is of particular importance in drug safety studies where
the incidence of adverse drug effects or drug use (or both) can be very low. Patient
data (e.g., on diagnosis, drug prescription, etc.) are generally recorded through
structured and coded information. Nevertheless, the additional availability of
unstructured free-text clinical notes, as in the Dutch GP database “IPCI,” holds
potential for manually validating and better characterizing clinical outcomes.
5.2.2 Administrative/Claims Databases
Data in claims/administrative databases are recorded when a patient uses healthcare

resources that are provided free of charge to citizens as they are reimbursed by
National Health Service (NHS) or when a commercial (e.g., Kaiser Permanente in
various North American states) or national social insurance provider (e.g., Medicare
in the USA) must be billed for the resources provided. The use and availability of
these databases depend on the healthcare system used in each country. The avail-
ability of commercial insurance databases is more prominent in countries where
healthcare is not provided publicly free of charge and persons must be insured/
registered with a health plan to access healthcare resources. Administrative data-
bases generally include pharmacy dispensing and hospital discharge diagnoses
which can be linked through patient unique identifier to several other administrative
databases (e.g., emergency department visits, outpatient diagnostic test, etc.).
Pharmacy claims contain data collected when a patient is dispensed a drug from a
pharmacy and its cost is covered by insurance/NHS. The advantage of this data
source as opposed to prescription data from a GP database is that, while the latter
records only the prescription of a drug, the former can at least ensure that prescribed
drug has been dispensed. Neither of the database types can however ensure that a
drug is ultimately taken, thus potentially leading to exposure misclassification in
case of drug therapies with low adherence. Likewise pharmacy databases, hospital
discharge diagnosis databases are built for claims/administrative purposes: health-
care resources related to the management of hospitalized medical event and eventu-
ally medical procedures undergone during hospital admission are recorded in order
5 The Role of Healthcare Databases in Pharmacovigilance of Psychotropic Drugs 77
to allow reimbursement by the insurance provider or National Healthcare Service.

However, because the billing depends on the type of information that is registered
as primary and secondary causes of hospital admission as well as hospital proce-
dures, it is possible that diagnoses in administrative and claims data are recorded
less accurately/rigorously than for medical record databases not used for billing
purposes. An example of a hospital claims/administrative database is the Pediatric
Health Information System (PHIS), which contains administrative data for over six
million patients from 44 children’s hospitals in the USA.
5.3 Observational Studies Concerning Psychotropic Drugs

Using Healthcare Databases
Observational drug-related studies in psychiatry conducted using healthcare data-

bases can be divided into two major types: drug utilization (e.g., exploring changes in
prevalence and incidence of drug use over time and across different settings, adher-
ence or persistence to drug use and drug switching patterns, etc.) and drug safety
studies (risk of adverse outcomes associated with the use of a specific drug). Drug
utilization studies are descriptive studies that can be conducted by analyzing an expo-
sure of interest in the underlying general population or selected cohort of patients
identified from a database. In the context of pharmacovigilance activities using
healthcare databases, it is possible to evaluate the implementation and impact of pub-
lic health interventions such as drug safety warnings issued by drug agencies and
other risk minimization measures (RMMs) rapidly and cost-effectively. This kind of
investigation is aimed at exploring any change, if at all, in the drug prescribing pattern
occurred as a result of a drug-related public health intervention (i.e., implementation
of RMMs) and measuring if the risk of an adverse outcome associated to specific drug
treatment was actually minimized after the implementation of the RMM in routine
care (Fig. 5.1). This assessment is essential to ensure that the RMMs led to the
achievement of expected results in terms of drug-related risk prevention/dilution.
Public health intervention

(e.g. drug safety warning)
Persons exposed to drug over time
Indirect assessment of intervention impact:Drug utilization studies

Direct assessment of intervention impact: Drug safety studies
Fig. 5.1 The role of healthcare databases in the assessment of public health interventions
Drug safety studies can be conducted using analytical methods that can estimate
the risk of an outcome following exposure to a drug. Nowadays, the assessment of
specific associations of drugs and adverse events is mostly carried out using well-
established study designs such as propensity score-matched new user cohorts or
case-control designs, nested in a cohort of new users of the drugs under study. More
recently, advance statistical techniques have been applied to data from healthcare
databases also for post-marketing assessment and comparison of benefits of psycho-
tropic drugs in the context of comparative effectiveness research.
5.3.1 Drug Utilization Studies
Healthcare databases have been widely used to analyze the prescribing pattern of
psychotropic drugs in clinical practice, mainly antidepressants, antipsychotics, and
benzodiazepines. Drug utilization studies have been carried out in the general popu-
lation as well as in “special” subsets of population such as elderly or pediatric
patients and pregnant women. Data on the prevalence/incidence of drug treatment,
adherence, and persistence to drug therapy provides an indication of quality of care.
An example of how the drug prevalence proves useful can be seen in antidepres-
sant utilization studies, which are commonly carried out at regional (Poluzzi et al.
2013; Parabiaghi et al. 2011), national (Sultana et al. 2014b; Chollet et al. 2013;
Lam et al. 2013), or multinational (Abbing-Karahagopian et al. 2014) level. Such
studies can give an indication of appropriate drug prescribing by confirming whether
first-line agents are indeed most commonly used and if the drugs are used in agree-
ment with international treatment guidelines and Summary of Products
Characteristics. If the electronic data source also has the indication of use included
in the prescription data, researchers can further investigate whether drug prescribing
is prescribed appropriately or off-label. Another example of the usefulness of preva-
lence and incidence data concerns the assessment of the distribution of prescription
of generic or branded drugs for drugs that are off- patent. Several clinical guidelines
suggest that generic antidepressants, selective serotonin reuptake inhibitors in par-
ticular, should be prescribed as first-line agents to reduce the use of healthcare
resources and promote sustainability of national healthcare systems. An example of
a basic evaluation of prescribing appropriateness would be to compare the preva-
lence of generic and brand antidepressant drug use using pharmacy data (Ubeda
et al. 2007). Such investigations have been put to use by national drug observatories
to find out the extent of generic antidepressant penetration. A more complex inves-
tigation of generic antidepressant drug use might involve a comparison of selected
outcomes that reflect drug safety such as mortality and other health resource con-
sumption (hospitalizations, specialist examinations, other drugs) between generic
and brand antidepressant users (Colombo et al. 2013).
Information on adherence/persistence complements prevalence data. A detailed
analysis of antidepressant use consistently shows that adherence levels are usually
low (Hansen et al. 2004). In fact, most of the drug utilization database studies on
antidepressants found that these drugs are frequently withdrawn as early as 3 months
after the therapy starts in a large proportion of depressed patients even though
guidelines suggest at least 1 year treatment (Poluzzi et al. 2013; Parabiaghi et al.
2011; Sultana et al. 2014b; Chollet et al. 2013). Analysis of persistence to drug
treatment can be useful as it may indirectly imply tolerability or lack of efficacy: in
the absence of economic reasons, a drug which is switched more often than others
may be less effective and/or less tolerable than other drugs. For example, a US-based
study using claims data found that patients prescribed with escitalopram were more
likely to persist with their antidepressant treatment and less likely to switch to other
antidepressants, compared to other SSRI users (Esposito et al. 2009).
Another class of drugs that has been the subject of many drug utilization studies
is that of antipsychotics. This is particularly important given the diverse pharmaco-
logical and safety profile of antipsychotic drugs as well as their several indications
of use apart from schizophrenia and mania, which currently include motor tics, nau-
sea, and vomiting in palliative care and intractable hiccups for haloperidol and per-
sistent aggression in pediatric conduct disorders and persistent aggression in
Alzheimer’s disease for risperidone. Drug utilization studies highlighted the increas-
ing use of second-generation or atypical antipsychotics in the general population,
particularly in elderly persons (Trifirò et al. 2005, 2010a). Studies investigating anti-
psychotic adherence were also pivotal, suggesting that nonadherence is a significant
problem in schizophrenia and is associated with negative outcomes such as hospital-
ization (Tiihonen et al. 2006; Ascher-Svanum et al. 2006; Gilmer et al. 2004). Some
examples of psychotropic drug utilization database studies are reported in Table 5.2.
5.3.2 Evaluation the Impact of Drug-Related Risk

Minimization Measures in Psychiatry
Understanding prescriber behavior is important particularly from a public health

perspective. Knowledge of drug safety is constantly expanding and clinical guide-
lines are continuously evolving in response to this. Sometimes rapid changes in
prescriber behavior are critical, as when prescribers are expected to respond
promptly to drug safety warnings. For example, studies of prescriber behavior are
necessary to see whether citalopram-related drug safety warnings on the risk of
cardiac arrhythmia resulted in a change of citalopram prescribing among persons at
higher risk of such adverse effects. Similarly, healthcare databases have also been
used to investigate changes in prescribing pattern after the antipsychotic-related
safety warnings in older people with dementia. In short, Health Canada, the
European Medicines Agency, the Food and Drug Administration, and other national
drug agencies issued several warnings from 2002 onward on the risk of adverse
events such as all-cause mortality and stroke when antipsychotics are used mostly
off-label in elderly persons with dementia, highlighting their unfavorable risk-
benefit ratio. A summary of studies investigating changes in antipsychotic prescrib-
ing pattern among elderly dementia patients is shown in Table 5.3. Such investigations
can be said to be indirect evaluations of drug-related public health interventions,
because the primary aim of such interventions is not the reduction of drug
Table 5.2 Examples of psychotropic drug utilization studies carried out using healthcare databases
Author
(year) Setting Population Exposure Outcomes
Marston THIN General APs APs prescribing rate in
et al. (2014) database population primary care
Petersen THIN Pregnant APs Discontinuation of APs in
et al. (2014) database women pregnancy
Wijlaars THIN Children ADs Trends in depression and
et al. (2012) database and antidepressant prescribing
adolescents in children and
adolescents
Man et al. THIN Pregnant AED Prevalence of AED use in
(2012) database women pregnancy
Hayes et al. THIN General Psychotropic drugs Prescribing trends of
(2011) database population used in bipolar drugs used in bipolar
disorder disorder
Prah et al. THIN General APs Prescribing trends of APs
(2011) database population for schizophrenia
Petersen THIN Pregnant ADs AD discontinuation
et al. (2011) database women
Aguglia HSD General SSRI and SNRI SSRI/SNRI prescribing
et al. (2012) population pattern
Parabiaghi HSD Elderly ADs AD utilization among
et al. (2011) population elderly in Lombardy
Trifirò et al. HSD Elderly APs AP prescribing pattern
(2010a) persons
with
dementia
Savica et al. HSD General AEDs AED prescribing pattern
(2007) population
Trifirò et al. HSD General AP AP prescribing pattern
(2005) population
Koopman IPCI General Drugs used to treat Pharmacological
et al. (2010) population neuropathic pain treatment of neuropathic
facial pain
Kraut et al. GePaRD Pediatric Methylphenidate Comorbidities in ADHD
(2013) population children treated with
methylphenidate
Lindemann GePaRD Children ADHD drugs Age-specific prevalence,
et al. (2012) and incidence of new
adolescents diagnoses, and drug
treatment of attention-
deficit/hyperactivity
disorder
Jacobsen Aarhus Pediatric AEDs Prenatal exposure to
et al. (2014) University population antiepileptic drugs
Prescription
Database
Table 5.2 (continued)

Author
Laugesen Aarhus Female SSRIs Use of SSRIs and lifestyle
et al. (2013) University general among women of
Prescription population childbearing age
Database
Hsia et al. PEDIANET, Pediatric AEDs AED prescribing among
(2010) IPCI, and IMS population children
disease
analyzer
AD antidepressant, ADHD attention-deficit/hyperactivity disorder, AED antiepileptic drug, AP
antipsychotic, GeParD German Pharmacoepidemiological Research Database, HSD Health Search
Database, IPCI Integrated Primary Care Information database, SNRI serotonin-norepinephrine
reuptake inhibitors, SSRI selective serotonin reuptake inhibitor, THIN The Health Improvement
Network
prescribing per se but a reduction in the drug-related risk as a consequence of a

more cautious use of the drug (Fig. 5.1). Healthcare databases can also be used to
carry out a direct evaluation of a safety warning impact by measuring the occur-
rence of a known adverse drug reaction before and after a warning.
More recently, healthcare databases have been increasingly used to carry out
postauthorization safety studies (PASS) also concerning psychotropic drugs. A
PASS is a scientific investigation of the safety profile of a drug that has already been
marketed with the aim of identifying and describing the safety profile of that drug,
as indicated in the risk management plan (RMP) supporting the premarketing docu-
mentation. In this case, healthcare databases can be used to estimate the frequency
of known ADRs within a given exposed population or to identify predictors of drug-
related risks, thus identifying categories of patients at high risk of developing ADRs;
on the other hand, it is unlikely for healthcare databases to have a major role in de
novo signal detection (hypothesis generation) in the immediate future, despite the
several international initiatives that have been carried out to better explore this issue
in the last years (Trifirò et al. 2014a). A PASS can also be carried out with the aim
of measuring whether any risk minimization measures implemented as part of a
RMP were effective or otherwise. Among other things, RMPs also cover planning
for how drug-related risks can be prevented or minimized as well as plans for stud-
ies aimed at increasing awareness on drug safety; healthcare databases have a piv-
otal role in both these aspects.
5.3.3 Drug Safety Studies
The safety profile of therapeutic drugs including psychotropic drugs is initially

investigated in randomized clinical trials (RCTs) in the premarketing phase.
However, drug safety data from RCTs is fraught with limitations such as small and
Table 5.3 Overview of database studies evaluating the impact of antipsychotic drug safety
warnings on the pattern of use of these drugs in routine care
Author
Gallini EGB database Elderly All APs, APs by Monthly prevalence of
et al. (France) patients class and AP use
(2014) with and olanzapine and
without risperidone
dementia individually
Schulze GEK database Elderly All APs and APs Yearly prevalence of AP
et al. (Germany) dementia by class use, number of AP
(2013) patients packages, and DDD per
person per year
Guthrie PCCIU database Elderly All APs and Quarterly prevalence of
et al. (Scotland) dementia risperidone, oral AP prescribing,
(2013) patients olanzapine, initiation, and
quetiapine discontinuation;
individually prescription of
hypnotics, anxiolytics, or
antidepressants
Franchi Lombardy Region Elderly All APs, APs by Number of AP
et al. Drug dementia class and prescriptions per person
(2012) Administrative patients olanzapine, and gap between AP
Database (Italy) treated with quetiapine, prescriptions; yearly
AChEIs haloperidol, prevalence of AP use,
clotiapine, and probability of continuing
risperidone antipsychotic treatment
individually
Dorsey IMS Health’s Elderly All APs and APs Monthly prevalence of
et al. National Disease dementia by class AP use, monthly change
(2010) and Therapeutic patients in AP drug use, and
Index (USA) annual growth rate of AP
use
Sanfélix- Valencia Health Elderly Risperidone and Monthly prevalence of
Gimeno Agency pharmacy patients and olanzapine use AP use
et al. claims database younger (stratified by
(2009) (Spain) adults strength)
Valiyeva Ontario Drug Elderly All APs, APs by Monthly prevalence of
et al. Benefit database dementia class and AP use; change in the
(2008) (Canada) patients olanzapine, number of patients
quetiapine, and prescribed an AP and
risperidone ratio of change of
individually prescription rate after the
safety warnings
AChEI acetylcholinesterase inhibitor, AIFA Agenzia Italiana del Farmaco (Italian Drug Agency),
AP antipsychotic, EGB Échantillon Généraliste de Bénéficiaires, EMA European Medicines
Agency, DDD defined daily dose, FDA Food and Drug Administration, GEK Gmünder Ersatzkasse,
a German nationwide health insurance company database, HSD-CSD LPD Health Search
Database – Cegedim Strategic Data, Longitudinal Patient Database, MHRA Medicines and
Healthcare Products Regulatory Agency, PCCIU Primary Care Clinical Informatics Unit
highly selective study populations as well as limited duration of the observation

period which preclude the possibility of identifying potentially severe adverse drug
reactions with long-term onset or which are more likely to occur when used in
patients affected by several comorbidities or receiving concomitant medications at
interaction risk (Sultana et al. 2013). In fact RCT populations may be healthier than
patients in real clinical practice and less likely to be prescribed a multidrug regimen
as RCTs in the drug development phase tend to exclude the frailest patient catego-
ries such as very old patients who are generally affected by multiple comorbidities.
Since the number of patients enrolled in a RCT is small compared to the number of
patients that will be exposed to a drug in clinical practice, RCTs are unlikely to
detect even life-threatening ADRs which occur rarely. These limitations can be
addressed by observational studies, including those making use of databases, since
such studies typically include data from long follow-up of large populations that
are representative of real clinical settings, i.e., patients of all ages (pediatric and/or
geriatric populations) and those at higher risk of an adverse drug reaction (the old-
est old, patients with several comorbidities, and those with on a multidrug
regimen).
The use of healthcare databases to monitor drug safety can be illustrated using
the association of antipsychotic drug use in dementia patients and the risk of all-
cause mortality as an example (Table 5.4). The employed study designs for studying
this association were either cohort or (nested) case-control and were conducted with
various types of healthcare databases, including government administrative, general
practice, and health insurance databases. The setting of each study might limit its
generalizability and should be considered in the interpretation of study findings. For
example, veterans’ affairs database populations are mostly white males and may not
be generalizable to the general population and government claims databases cover-
ing Medicaid beneficiaries, who as persons having a lower socioeconomic status
compared to non-Medicaid beneficiaries, might not be generalizable to the general
US population.
Safety studies can be used to investigate various outcomes other than mortality.
In keeping with the previous example of antipsychotic use in dementia, risk of
adverse reactions such as cerebrovascular events, pneumonia, and venous thrombo-
embolism has been investigated with AP use in elderly persons with/without demen-
tia using healthcare databases (Trifirò et al. 2014b). Such studies were particularly
useful from a clinical perspective as they explored the differential risk associated
with individual antipsychotics, building on previous findings suggesting class-spe-
cific findings, such as that conventional antipsychotics are more likely to be associ-
ated specific outcomes (e.g., bacterial infections, myocardial infarction, and hip
fractures) but less likely to be associated with cerebrovascular events compared
with atypical antipsychotics when used in elderly dementia patients (Huybrechts
et al. 2012a). For example, emerging evidence suggests that quetiapine appears to
have a lower risk of all-cause mortality compared to risperidone, while haloperidol
has much higher risk all-cause mortality than risperidone (Huybrechts et al. 2012b).
Table 5.4 Examples of safety studies on antipsychotic drugs carried out using healthcare databases
84
Author (year) Study design Setting Population Outcome Exposure

Wang et al. (2005) Retrospective cohort Pennsylvania Medicare Patients ≥65 All-cause mortality CAPs vs. AAPs
study (N = 22,890)
Trifirò et al. (2007) Nested case-control Dutch general practice Dementia patients All-cause mortality Current use of AAP or CAP
study database (IPCI) ≥65 (N = 2,385) (nonuse of respective class as
comparator unless otherwise
specified)
Gill et al. (2007) Retrospective cohort Four administrative Dementia patients All-cause mortality at Incident use of AAPs
study databases in Ontario: ≥66 (N = 27,259) 30, 60, 120, and (nonuse as comparator) and
Ontario Drug Benefit 180 days after the CAPs (AAP as comparator)
program initial dispensing of stratified in CD LTC cohorts
Canadian Institute for antipsychotic
Health Information medication
Discharge Abstract
Database
Ontario Health
Insurance Plan
Registered Persons
Database
Schneeweiss et al. Retrospective cohort Linked administrative Patients ≥65 year old 180-day all-cause Incident use of AAPs and
(2007) study data from the British with an AP mortality CAPs (risperidone as
Columbia Ministry of prescription comparator)
Health, PharmaNet (N = 37,241)
database British
Columbia
Kales et al. (2007) Retrospective cohort US Department of Dementia patients 1-year all-cause Incident use of AAPs, CAPs,
study Veteran Affairs registries ≥65 year old with an mortality from and combination of both
AP prescription National Death Index types (CAP as comparator)
(N = 10,615)
G. Trifirò and J. Sultana
5
Hollis et al. (2007a) Retrospective cohort Australian Department Veterans and war All-cause mortality Incident use of
study of Veteran Affairs widows ≥65 antipsychotics,
claims-based (N = 16,634) carbamazepine and valproate
pharmaceutical database (incident use of olanzapine
as comparator)
Hollis et al. (2007b) Cohort study Department of Veterans’ Incident users of APs All-cause mortality CAPs (chlorpromazine,
Affairs database and haloperidol, pericyazine,
Medicare Australia trifluoperazine) AAPs
(quetiapine, olanzapine,
risperidone) (olanzapine as
comparator)
Aparasu et al. Retrospective cohort Medicare and Medicaid Nursing home All-cause mortality AAPs vs. CAPs (AAP as
(2012) design matched on data from Texas, residents ≥65 years comparator)
propensity score Florida, New York, and (N = 7,218)
California
Kales et al. (2012) Retrospective cohort US Department of Dementia patients 180-day mortality Risperidone, haloperidol,
study Veteran Affairs database ≥65 years old olanzapine, quetiapine
(N = 1,932) (risperidone as comparator)
Huybrechts et al. Cohort study Linked data from Nursing home patients All-cause mortality Incident use of haloperidol,
(2012a) Medicaid, Medicare, the ≥65 (N = 75,445) (excluding cancer aripiprazole, olanzapine,
Minimum Data Set, the mortality) and quetiapine, risperidone,
National Death Index, cause-specific ziprasidone (risperidone as
and a national mortality comparator)
assessment of nursing
home quality
(continued)
The Role of Healthcare Databases in Pharmacovigilance of Psychotropic Drugs
85
86

Rafaniello et al. Prospective cohort Dementia Evaluation Dementia patients All-cause mortality Rate per 100 PY:
(2014) study Unit of Campania with BPSD C65 years Quetiapine: 5.8 (4.4–7.7)
who were incident Risperidone: 7.3 (4.8–11.1)
antipsychotic users
Olanzapine: 4.6 (2.5–8.3)
(N = 1,618)
Clozapine: 7.7 (1.9–30.9)
Aripiprazole:
17.6 (2.5–125.0)
Sultana et al. Retrospective cohort South London and Vascular dementia All-cause mortality AAPs (nonuse as
(2014a) study Maudsley NHS patients (N = 1,531) comparator)
Foundation Trust
(SLaM), Biomedical
Research Centre
database, Clinical
Record Interactive
Search (CRIS)
AAPs atypical antipsychotics, Adj adjusted, AP antipsychotic, BSPD behavioral and psychological symptoms of dementia, CAPs conventional antipsychotics,
CD community dwelling, HR hazard ratio, LTC long-term care, NHS National Health Service, OR odds ratio, RR relative risk, PY person-years
G. Trifirò and J. Sultana
5.3.4 Limitations of Healthcare Databases for the Conduct

of Observational Studies Concerning Psychotropic Drugs
Studying the relationship between psychotropic drug use and new onset of adverse
effects in electronic health record databases is extremely challenging due to a
variety of potential biases and confounders (Brookhart et al. 2010). In all observa-
tional studies, various types of bias, e.g., selection bias, protopathic bias, informa-
tion bias due to outcome and exposure misclassification and confounding by
indication, may influence the study findings. The potential for selection bias in
database safety studies is minimal as all data are obtained from prospectively col-
lected medical records/claims that are maintained for patient care purposes or
administrative/reimbursement reasons on a population-based level and indepen-
dent of the patients’ health status. All the other biases represent a real concern for
this type of studies as discussed more into detail below.
5.3.4.1 Outcome Misclassification
Outcome misclassification may present more of a problem in claims healthcare data

compared to general practice data registered for non-claims purposes, since the for-
mer may be less likely to register diagnosis codes accurately if this is not relevant to
claims, compared to the latter. On the other hand, GP databases are less likely to
register clinical outcomes which lead directly to emergency department visit/hospital
admission or death.
All healthcare databases may be prone to the misclassification of disease for
which symptoms may be missed. An example of this is the registration of transient
ischemic attack (TIA) when assessing the risk of cerebrovascular accidents (CVAs)
in association to the use of antipsychotics in older people with dementia. TIA may
be the trigger of stroke but potentially misdiagnosed by clinicians (thus not properly
registered in any type of healthcare databases), especially in patients suffering
already cerebrovascular disorders such as dementia patients.
In general, in drug safety database studies, it is necessary to identify outcomes by
using coding algorithms that have been previously validated in the same or a similar
database or to manually validate outcome through revision of medical charts (in
claims databases) or electronic medical records (in GP database) including unstruc-
tured free-text information. It may be helpful also to conduct sensitivity analyses
while using different outcome definitions having different level of accuracy with the
aim of exploring the possible effect and magnitude of bias due to outcome misclas-
sification. This approach has been frequently used in most of observational database
studies which explore the risk of CVAs, myocardial infarction, pneumonia, and oth-
ers in association to several psychotropic drugs such as antidepressant and antipsy-
chotics (Pariente et al. 2012; Trifirò et al. 2010b, c).
5.3.4.2 Exposure Misclassification
Prescription data in healthcare databases is free of recall bias as information on

exposure is systematically registered for all drugs reimbursed by NHS, as it is the
case for most of psychotropic drugs (except for benzodiazepines in several coun-
tries such as Italy where these drugs are directly charged to the citizens and thus not
traceable with healthcare databases). General practice databases may have greater
risk of exposure misclassification since, even if the prescribed drugs are registered
correctly within the database, they may not necessarily be filled by patients. For this
reason, pharmacy dispensing data is perhaps less likely to be associated with such
exposure misclassification, even though exposure assessment is based on the
assumption that patients take their medication as prescribed, once dispensed. A
patient might prolong the duration of a prescription by tablet splitting but based on
the prescription would be classified as exposed. Misclassification may also occur if
patients discontinue a drug before their drug supply is finished. Drugs taken inter-
mittently on an as-needed basis are at risk of being misclassified, as may be the case
for antipsychotics or drugs for insomnia or benzodiazepines as drops formulation.
Another issue concerns over-the-counter or privately purchased medicines which
would not be captured by healthcare databases; however, this is not a significant
issue for psychotropic drugs which are prescription drugs. Claims healthcare data
may lead to exposure misclassification as far as prescriptions for drugs that are not
covered by an insurance are not available, for example, if a patient pays for a medi-
cation himself without requesting reimbursement (Schneeweiss and Avorn 2005).
As for outcome misclassification, in database safety studies on psychotropic
drugs, sensitivity analyses changing the criteria for exposure status measurement
should be carried out to investigate the possible role and magnitude of the effect of
bias on the risk estimates.
5.3.4.3 Confounding by Indication
In routine care the decision to prescribe a specific drug for the treatment of a certain
indication in individual patients is based on several factors such as demographic
characteristics of the patients, presence of concomitant medications and comorbidi-
ties, overall patients’ health status, severity of the disease/symptom that is intended
to be treated, as well as patients’ and physicians’ preference and, in some situations,
affordability of the drug therapy by the patient if the purchase of the medicine is
charged directly to the patient, and so on. As a result, it is possible that patients
prescribed a given drug for a certain indication of use differ substantially from the
patients treated with other drugs or not treated at all in terms of baseline risk of the
studied safety outcome, which may lead to the so-called confounding by
indication.
Confounding by indication is a commonly used term that refers to an extraneous
determinant of the outcome parameter that is present if a perceived high risk or poor
prognosis is an indication for intervention. The indication is a confounder of the
drug-adverse event association because it correlates with the intervention and is a

risk indicator for the illness (Salas et al. 1999). Confounding by indication is likely
in safety studies concerning the association of commonly investigated safety out-
comes and psychotropic drugs, since the choice of psychotropic drug is often asso-
ciated with the prognosis or condition of a patient, which in itself can be a risk factor
for studies outcomes (e.g., stroke, pneumonia, all-cause mortality, etc.). As an
example, older people with dementia requiring antipsychotic prescriptions for the
treatment of behavioral and psychotic disturbances of dementia (BPSD) have an
increased risk of dying as compared to those who do not require such a treatment as
the occurrence of BPSD in dementia patients is per se a strong risk factor of death
(Tschanz et al. 2004). It is essential to keep this in mind when selecting the com-
parator in the assessment of association of any outcome and specific drug use in
observational database studies.
In observational studies using healthcare databases, this confounding may bias
the drug-adverse event risk estimate and particular attention should be paid in the
design (i.e., selection of proper comparator, nesting a case-control study in a new
user cohort) as well as the analytical phase, i.e., carrying out sensitivity analyses
changing comparator to explore possible effect and magnitude of the bias effect on
the risk estimate due to confounding by indication.
5.3.4.4 Protopathic Bias
Protopathic bias occurs when a drug is used to treat prodromic symptoms of the
study outcome; as a result, it may mistakenly appear that the drug is causing the
occurrence of the outcome under investigation (Horwitz and Feinstein 1980). For
example, SSRIs are frequently prescribed for the treatment of late life depression,
which may represent a manifestation of subtle cerebrovascular disorders leading to
stroke in the elderly (Krishnan 2000). Similarly, severe pneumonia may induce
delirium and trigger subsequent antipsychotic drug use in elderly patients (Marrie
2000). In both situations, wrong assessment of the date of onset of study outcome
could result in protopathic bias, thus mistakenly attributing stroke and pneumonia
onset to SSRIs and antipsychotics, respectively. To deal with this protopathic bias in
database safety studies, sensitivity analyses can be carried out by excluding from
the analysis those patients who started the therapy within a short period prior to the
occurrence of the outcome.
5.3.4.5 Residual Confounding
The impact of confounding in healthcare database research can be reduced through

the use of a suitable study design and/or a statistical strategy to adjust for known
confounders, such as propensity score matching (Patorno et al. 2013b).
Nevertheless, is it difficult to fully exclude the possibility of bias due to unknown
and unmeasured confounders in all database safety studies, irrespective of the
drug under investigation. Such confounders may include lifestyle factors (diet,
level of exercise, etc.), the use of over-the-counter medications, and disease sever-
ity, which is often not registered, particularly in claims databases (Patorno et al.
2013a). If data is available on a single unmeasured variable suspected to be a
confounder, for example, data from a subsample or supplementary data from
another data source, it may be possible to remove or adjust for residual confound-
ing due to that variable. If, on the other hand, data on several unmeasured variables
that may be confounders is available, the propensity score method can be used to
adjust estimates.
5.4 Concluding Remarks and Future Directions
Healthcare databases such as GPs’ electronic medical records as well as administra-

tive/claims databases are important data sources to carry out observational studies
aimed at quantifying and describing emerging safety issues associated with the use
psychotropic drugs, as shown by the large amount of database safety studies that
have been published worldwide in the last decades. As observational studies lack
randomization in the assignment of the drug treatment under study, confounding
and bias are issues that should be taken carefully into consideration. Nevertheless,
it is to be considered that in any case a single observational study (not being experi-
mental) cannot establish the causal pathway of a drug-(adverse) event association.
Instead, confirmation of the results from a significant number of observational stud-
ies that have been carried out using different methodologies and data sources to
explore the same safety issues, together with evidence coming from different
sources (e.g., biological plausibility, existence of randomized clinical trials or case
reports, etc.), can support causal association. It is important however to acknowl-
edge limitations of observational studies and databases being used so to interpret the
study results properly and, if necessary, cautiously, in addition to, whenever possi-
ble, the use of sensitivity analyses which may contribute to strengthen the robust-
ness of the findings.
In the context of safety studies using electronic healthcare databases, new oppor-
tunities are open through the carrying out of multiple databases’ safety studies
which have been developed in different international initiatives which created dis-
tributed database networks. Multiple database safety studies are extremely chal-
lenging from a logistical point of view, especially in case of combination of data
sources from different countries with different underlying healthcare systems and
healthcare data privacy and management legislations, as well as time- and resource-
consuming from a methodological perspective. However, this approach may be par-
ticularly useful for the investigation of association of rare adverse events and rare
drug exposure (e.g., cardiac valve regurgitation and use of ergot-derived dopamine
agonists for the treatment of Parkinson’s disease) or in very specific patients catego-
ries (e.g., older people with specific type of dementia) which requires large base
populations to gain enough statistical power to investigate such associations.
References
Abbing-Karahagopian V, Huerta C, Souverein PC, de Abajo F, Leufkens HG, Slattery J, Alvarez Y,

Miret M, Gil M, Oliva B, Hesse U, Requena G, de Vries F, Rottenkolber M, Schmiedl S,
Reynolds R, Schlienger RG, de Groot MC, Klungel OH, van Staa TP, van Dijk L, Egberts AC,
Gardarsdottir H, De Bruin ML (2014) Antidepressant prescribing in five European countries:
application of common definitions to assess the prevalence, clinical observations, and method-
ological implications. Eur J Clin Pharmacol 70(7):849–857
Aguglia E, Ravasio R, Simonetti M, Pecchioli S, Mazzoleni F (2012) Use and treatment modalities
for SSRI and SNRI antidepressants in Italy during the period 2003–2009. Curr Med Res Opin
28(9):1475–1484
Aparasu RR, Chatterjee S, Mehta S, Chen H (2012) Risk of death in dual-eligible nursing home
residents using typical or atypical antipsychotic agents. Med Care 50(11):961–969
Ascher-Svanum H, Faries DE, Zhu B, Ernst FR, Swartz MS, Swanson JW (2006) Medication
adherence and long-term functional outcomes in the treatment of schizophrenia in usual care.
J Clin Psychiatry 67(3):453–460
Brookhart MA, Stürmer T, Glynn RJ, Rassen J, Schneeweiss S (2010) Confounding control in health-
care database research: challenges and potential approaches. Med Care 48(6 Suppl):S114–S120
Chollet J, Saragoussi D, Clay E, François C (2013) A clinical research practice datalink analysis of
antidepressant treatment patterns and health care costs in generalized anxiety disorder. Value
Health 16(8):1133–1139
Colombo GL, Agabiti-Rosei E, Margonato A, Mencacci C, Montecucco CM, Trevisan R (2013)
Off-patent generic medicines vs. off-patent brand medicines for six reference drugs: a retro-
spective claims data study from five local healthcare units in the Lombardy Region of Italy.
PLoS One 8(12), e82990
Dorsey ER, Rabbani A, Gallagher SA, Conti RM, Alexander GC (2010) Impact of FDA black box
advisory on antipsychotic medication use. Arch Intern Med 170(1):96–103
Esposito D, Wahl P, Daniel G, Stoto MA, Erder MH, Croghan TW (2009) Results of a retrospec-
tive claims database analysis of differences in antidepressant treatment persistence associated
with escitalopram and other selective serotonin reuptake inhibitors in the United States. Clin
Ther 31(3):644–656
Franchi C, Tettamanti M, Marengoni A, Bonometti F, Pasina L, Cortesi L, Fortino I, Bortolotti A,
Merlino L, Lucca U, Riva E, Nobili A (2012) Changes in trend of antipsychotics prescription
in patients treated with cholinesterase inhibitors after warnings from Italian Medicines Agency.
Results from the EPIFARM-Elderly Project. Eur Neuropsychopharmacol 22(8):569–577
Gallini A, Andrieu S, Donohue JM, Oumouhou N, Lapeyre-Mestre M, Gardette V (2014) Trends
in use of antipsychotics in elderly patients with dementia: impact of national safety warnings.
Eur Neuropsychopharmacol 24(1):95–104
Gill SS, Bronskill SE, Normand SL, Anderson GM, Sykora K, Lam K, Bell CM, Lee PE, Fischer
HD, Herrmann N, Gurwitz JH, Rochon PA (2007) Antipsychotic drug use and mortality in
older adults with dementia. Ann Intern Med 146(11):775–786
Gilmer TP, Dolder CR, Lacro JP, Folsom DP, Lindamer L, Garcia P, Jeste DV (2004) Adherence to
treatment with antipsychotic medication and health care costs among Medicaid beneficiaries
with schizophrenia. Am J Psychiatry 161:692–699
Guthrie B, Clark SA, Reynish EL, McCowan C, Morales DR (2013) Differential impact of two
risk communications on antipsychotic prescribing to people with dementia in Scotland: seg-
mented regression time series analysis 2001–2011. PLoS One 8(7), e68976
Hansen DG, Vach W, Rosholm JU, Søndergaard J, Gram LF, Kragstrup J (2004) Early discontinu-
ation of antidepressants in general practice: association with patient and prescriber characteris-
tics. Fam Pract 21(6):623–629
Hayes J, Prah P, Nazareth I, King M, Walters K, Petersen I, Osborn D (2011) In bipolar disorder:
cohort study in the United Kingdom THIN primary care database 1995–2009. PLoS One
6(12):e28725
Hollis J, Grayson D, Forrester L, Brodaty H, Touyz S, Cumming R (2007a) Antipsychotic medica-

tion dispensing and risk of death in veterans and war widows 65 years and older. Am J Geriatr
Psychiatry 15(11):932–941
Hollis J, Forrester L, Brodaty H, Touyz S, Cumming R, Grayson D (2007b) Risk of death associ-
ated with antipsychotic drug dispensing in residential aged care facilities. Aust N Z J Psychiatry
41(9):751–758
Horwitz RI, Feinstein AR (1980) The problem of “protopathic bias” in case-control studies. Am J
Med 68(2):255–258
Hsia Y, Neubert A, Sturkenboom MC, Murray ML, Verhamme KM, Sen F, Giaquinto C, Ceci A,
Wong IC, TEDDY Network of Excellence (2010) Comparison of antiepileptic drug prescribing
in children in three European countries. Epilepsia 51(5):789–796
Huybrechts KF, Schneeweiss S, Gerhard T, Olfson M, Avorn J, Levin R, Lucas JA, Crystal S
(2012a) Comparative safety of antipsychotic medications in nursing home residents. J Am
Geriatr Soc 60(3):420–429
Huybrechts KF, Gerhard T, Crystal S, Olfson M, Avorn J, Levin R, Lucas JA, Schneeweiss S
(2012b) Differential risk of death in older residents in nursing homes prescribed specific anti-
psychotic drugs: population based cohort study. BMJ 344:e977
Jacobsen PE, Henriksen TB, Haubek D, Østergaard JR (2014) Prenatal exposure to antiepileptic
drugs and dental agenesis. PLoS One 9(1):e84420
Kales HC, Valenstein M, Kim HM, McCarthy JF, Ganoczy D, Cunningham F, Blow FC (2007)
Mortality risk in patients with dementia treated with antipsychotics versus other psychiatric
medications. Am J Psychiatry 164(10):1568–1576
Kales HC, Kim HM, Zivin K, Valenstein M, Seyfried LS, Chiang C, Cunningham F, Schneider LS,
Blow FC (2012) Risk of mortality among individual antipsychotics in patients with dementia.
Am J Psychiatr 169(1):71–79
Koopman JS, Huygen FJ, Dieleman JP, de Mos M, Sturkenboom MC (2010) Pharmacological
treatment of neuropathic facial pain in the dutch general population. J Pain 11(3):264–272
Kraut AA, Langner I, Lindemann C, Banaschewski T, Petermann U, Petermann F, Mikolajczyk
RT, Garbe E (2013) Comorbidities in ADHD children treated with methylphenidate: a database
study. BMC Psychiatry 13:11
Krishnan KR (2000) Depression as a contributing factor in cerebrovascular disease. Am Heart
140(4 Suppl):70–76
Lam D, Gorman DA, Patten S, Pringsheim T (2013) The pharmacoepidemiology of selective sero-
tonin reuptake inhibitors for children and adolescents in Canada from 2005 to 2009: a database
analysis. Pediatr Drugs 15(4):319–327
Laugesen K, Telén Andersen AB, Nørgaard M, Nielsen RB, Thomsen RW, Larsen FB, Sørensen
HT (2013) Use of selective serotonin reuptake inhibitors and lifestyle among women of child-
bearing age: a Danish cross-sectional survey. BMJ Open. Jul 31;3(7). pii: e003024.
Lindemann C, Langner I, Kraut AA, Banaschewski T, Schad-Hansjosten T, Petermann U,
Petermann F, Schreyer-Mehlhop I, Garbe E, Mikolajczyk RT (2012) Age-specific prevalence,
incidence of new diagnoses, and drug treatment of attention-deficit/hyperactivity disorder in
Germany. J Child Adolesc Psychopharmacol 22(4):307–314
Man SL, Petersen I, Thompson M, Nazareth I (2012) Antiepileptic drugs during pregnancy in
primary care: a UK population based study. PLoS One 7(12):e52339
Marrie TJ (2000) Community-acquired pneumonia in the elderly. Clin Infect Dis
31(4):1066–1078
Marston L, Nazareth I, Petersen I, Walters K, Osborn DP2 (2014) Prescribing of antipsychotics in
UK primary care: a cohort study. BMJ Open 4(12):e006135
Parabiaghi A, Franchi, Tettamanti M, Barbato A, D’Avanzo B, Fortino I, Bortolotti A, Merlino L,
Nobili A (2011) Antidepressants utilization among elderly in Lombardy from 2000 to 2007:
dispensing trends and appropriateness. Eur J Clin Pharmacol 67(10):1077–1083
Pariente A, Fourrier-Réglat A, Ducruet T, Farrington P, Béland SG, Dartigues JF, Moore N, Moride
Y (2012) Antipsychotic use and myocardial infarction in older patients with treated dementia.
Arch Intern Med 172(8):648–653
Patorno E, Glynn RJ, Hernandez-Diaz S, Avorn J, Wahl PM, Bohn RL, Mines D, Liu J, Schneeweiss
S (2013a) Risk of ischemic cerebrovascular and coronary events in adult users of anticonvul-
sant medications in routine care settings. J Am Heart Assoc 2(4), e000208
Patorno E, Grotta A, Bellocco R, Schneeweiss S (2013b) Propensity score methodology
for confounding control in health care utilization databases. Epidemiol Biostat Public
Health 10(3), e8940:1–16
Petersen I, Gilbert RE, Evans SJ, Man SL, Nazareth I (2011) Pregnancy as a major determinant for
discontinuation of antidepressants: an analysis of data from The Health Improvement Network.
J Clin Psychiatry 72(7):979–985
Petersen I, McCrea RL, Osborn DJP, Evans S, Pinfold V, Cowen PJ, Gilbert R, Nazareth I (2014)
Discontinuation of antipsychotic medication in pregnancy: a cohort study. Schizophr Res
159(1):218–225
Poluzzi E, Piccinni C, Sangiorgi E, Clo M, Tarricone I, Menchetti M, De Ponti F (2013) Trend in
SSRI-SNRI antidepressants prescription over a 6-year period and predictors of poor adherence.
Eur J Clin Pharmacol 69(12):2095–2101
Prah P, Petersen I, Nazareth I, Walters K, Osborn D (2011) National changes in oral antipsychotic
treatment for people with schizophrenia in primary care between 1998 and 2007 in the United
Kingdom. Pharmacoepidemiol Drug Saf 21(2):161–169
Rafaniello C, Lombardo F, Ferrajolo C, Sportiello L, Parretta E, Formica R, Potenza S, Rinaldi B,
Irpino A, Raschetti R, Vanacore N, Rossi F, Capuano A (2014) Predictors of mortality in atypi-
cal antipsychotic-treated community-dwelling elderly patients with behavioural and psycho-
logical symptoms of dementia: a prospective population-based cohort study from Italy. Eur J
Clin Pharmacol 70(2):187–195
Salas M, Hofman A, Stricker BH (1999) Confounding by indication: an example of variation in the
use of epidemiologic terminology. Am J Epidemiol 149:981–983, 72
Sanfélix-Gimeno G, Cervera-Casino P, Peiró S, López-Valcarcel BG, Blázquez A, Barbera T
(2009) Effectiveness of safety warnings in atypical antipsychotic drugs: an interrupted time-
series analysis in Spain. Drug Saf 32(11):1075–1087
Savica R, Beghi E, Mazzaglia G, Innocenti F, Brignoli O, Cricelli C, Caputi AP, Musolino R, Spina
E, Trifiro G (2007) Prescribing patterns of antiepileptic drugs in Italy: a nationwide population-
based study in the years 2000–2005. Eur J Neurol 14(12):1317–1321
Schneeweiss S, Avorn J (2005) A review of uses of health care utilization databases for epidemio-
logic research on therapeutics. J Clin Epidemiol 58(4):323–337
Schneeweiss S, Setoguchi S, Brookhart A, Dormuth C, Wang PS (2007) Risk of death associated
with the use of conventional versus atypical antipsychotic drugs among elderly patients. CMAJ
176(5):627–632
Schulze J, van den Bussche H, Glaeske G, Kaduszkiewicz H, Wiese B, Hoffmann F (2013) Impact
of safety warnings on antipsychotic prescriptions in dementia: nothing has changed but the
years and the substances. Eur Neuropsychopharmacol 23(9):1034–1042
Sultana J, Cutroneo P, Trifirò G (2013) Clinical and economic burden of adverse drug reactions. J
Pharmacol Pharmacother 4(Suppl 1):S73–S77
Sultana J, Chang CK, Hayes RD, Broadbent M, Stewart R, Corbett A, Ballard C (2014a)
Associations between risk of mortality and atypical antipsychotic use in vascular dementia: a
clinical cohort study. Int J Geriatr Psychiatry 29(12):1249–1254
Sultana J, Italiano D, Spina E, Cricelli C, Lapi F, Pecchioli S, Gambassi G, Trifirò G (2014b)
Changes in the prescribing pattern of antidepressant drugs in elderly patients: an Italian,
nationwide, population-based study. Eur J Clin Pharmacol 70(4):469–478
Tiihonen J, Wahlbeck K, Lönnqvist J, Klaukka T, Ioannidis JP, Volavka J, Haukka J (2006)
Effectiveness of antipsychotic treatments in a nationwide cohort of patients in community care
after first hospitalisation due to schizophrenia and schizoaffective disorder: observational fol-
low-up study. BMJ 333(7561):224
Trifirò G, Spina E, Brignoli O, Sessa E, Caputi AP, Mazzaglia G (2005) Antipsychotic prescribing
pattern among Italian general practitioners: a population-based study during the years 1999–
2002. Eur J Clin Pharmacol 61(1):47–53
Trifirò G, Verhamme KM, Ziere G, Caputi AP, Ch Stricker BH, Sturkenboom MC (2007) All-cause
mortality associated with atypical and typical antipsychotics in demented outpatients.
Trifirò G, Sini G, Sturkenboom MC, Vanacore N, Mazzaglia G, Caputi AP, Cricelli C, Brignoli O,
Aguglia E, Biggio G, Samani F (2010a) Prescribing pattern of antipsychotic drugs in the Italian
general population 2000–2005: a focus on elderly with dementia. Int Clin Psychopharmacol
25(1):22–28
Trifirò G, Dieleman J, Sen EF, Gambassi G, Sturkenboom MC (2010b) Risk of ischemic stroke
associated with antidepressant drug use in elderly persons. J Clin Psychopharmacol
30(3):252–258
Trifirò G, Gambassi G, Sen EF, Caputi AP, Bagnardi V, Brea J, Sturkenboom MC (2010c)
Association of community-acquired pneumonia with antipsychotic drug use in elderly patients:
a nested case-control study. Ann Intern Med 152(7):418–2
Trifirò G, Coloma PM, Rijnbeek PR, Romio S, Mosseveld B, Weibel D, Bonhoeffer J, Schuemie
M, van der Lei J, Sturkenboom M (2014a) Combining multiple healthcare databases for post-
marketing drug and vaccine safety surveillance: why and how? Intern Med J 275(6):551–561
Trifiró G, Sultana J, Spina E (2014b) Are the safety profiles of antipsychotic drugs used in demen-
tia the same? An updated review of observational studies. Drug Saf 37(7):501–520
Tschanz JT, Corcoran C, Skoog I, Khachaturian AS, Herrick J, Hayden KM, Welsh-Bohmer KA,
Calvert T, Norton MC, Zandi P, Breitner JC, Cache County Study Group (2004) Dementia: the
leading predictor of death in a defined elderly population: the Cache County Study. Neurology
62(7):1156–1162
Ubeda A, Cardo E, Sellés N, Broseta R, Trillo JL, Fernández-Llimós F (2007) Antidepressant
utilization in primary care in a Spanish region: impact of generic and reference-based pricing
policy (2000–2004). Soc Psychiatry Psychiatr Epidemiol 42(3):181–188
Valiyeva E, Herrmann N, Rochon PA, Gill SS, Anderson GM (2008) Effect of regulatory warnings
on antipsychotic prescription rates among elderly patients with dementia: a population-based
time-series analysis. CMAJ 179(5):438–446
Wang PS, Schneeweiss S, Avorn J, Fischer MA, Mogun H, Solomon DH, Brookhart MA (2005)
Risk of death in elderly users of conventional vs. atypical antipsychotic medications. N Engl J
Med 353(22):2335–2341
Wijlaars LPMM, Nazareth I, Petersen I (2012) Trends in depression and antidepressant prescribing
in children and adolescents: a cohort study in The Health Improvement Network (THIN). PLoS
One 7(3):e33181
Chapter 6
Monitoring of Plasma Concentrations
of Psychotropic Drugs in Pharmacovigilance
Christoph Hiemke and Ekkehard Haen
Abstract The primary aims of pharmacovigilance are supervision and prevention

of medication-related problems under everyday conditions. Pharmacovigilance is
an indication for monitoring plasma concentrations, i.e. conducting therapeutic
drug monitoring (TDM). Using TDM, it can be clarified if observed unwanted drug
effects may be attributed to abnormally high or low drug concentrations. Utmost
benefits from TDM are obtained for pharmacovigilance when the method is ade-
quately integrated into the clinical treatment process. How to do this is described in
consensus guidelines for TDM in psychiatry. During the last 20 years, TDM was
very successful for detection of multiple pharmacokinetic drug-drug interactions.
Many of them were discovered in individual cases. This gave rise to systematic
prospective studies to verify or falsify such observations. Confirmatory studies,
however, are critical when drug combinations are potentially harmful. Then TDM
databases should be used for retrospective analysis. They enable to study retrospec-
tively the safety and tolerability of psychotropic drugs and drug combinations taken
in a broad spectrum of patients, including risk patients like children or adolescent
patients, old-aged patients or patients with comorbid diseases. For such studies,
however, TDM databases must contain not only laboratory but also clinical data.
This is actually quite rare. Work in this regard is necessary. When functioning TDM
software is available and TDM is widely used in psychiatry, data can be pooled to
C. Hiemke (*)
Department of Psychiatry and Psychotherapy, University Medical Center of Mainz,
Untere Zahlbacher Str. 8, Mainz 55101, Germany
e-mail: hiemke@uni-mainz.de
E. Haen
Department of Psychiatry and Psychotherapy and Department of
Pharmacology and Toxicology, University of Regensburg, Universitätsstr. 31,
Regensburg 93053, Germany
Institute AGATE e.V., Nelkenweg 10, Pentling 93080, Germany
e-mail: ekkehard.haen@klinik.uni-regensburg.de

DOI 10.1007/978-3-319-24741-0_6
96 C. Hiemke and E. Haen
emerge large databases for data mining. TDM has thus the potential to be and
become a valuable part of pharmacovigilance.
Keywords Therapeutic drug monitoring • Side effect • Plasma concentration •

Reference range • Laboratory alert level • Data mining • Computerized physician
order entry • Clinical decision systems • Drug-drug patients • Risk patients
6.1 Introduction
Drug therapies are commonly controlled by the dose. With the “right” dose, optimal
clinical improvement is to be achieved without adverse effects. Therefore, a crucial
factor is the concentration of the drug that reaches the site of action, in case of psy-
chotropic drugs, the brain. In particular, because of interindividual differences in the
equipment of the liver with drug-degrading enzymes or drug transporters, drug con-
centrations under the same dose can vary considerably from patient to patient. This
is shown exemplary in Fig. 6.1 for 169 schizophrenic patients who were treated with
amisulpride. Each point in this figure shows the concentration of the antipsychotic
drug that was attained under the prescribed dose.
Similar patterns of dose-related drug concentrations are known for any psycho-
tropic drug, e.g. antidepressants (Ostad Haji et al. 2012) or antipsychotic drugs
(Hiemke et al. 2004; Hefner et al. 2013). Since the drug concentration in the blood
usually correlates well with concentrations at the site of action, it has been proven
Recommended dose range

1800
1600
Plasma concentration (ng/ml)
1400
1200
1000
800
600
400
Recommended
200 therapeutic reference range
0
200 400 600 800 1000 1200 1400 1600 1800
Amisulpride dose (mg/day)
Fig. 6.1 Dose-dependent plasma concentrations of amisulpride in 179 schizophrenic patients

treated and observed under everyday conditions. Drug concentrations are highly variable. Under
recommended doses of 400–800 mg/day, many patients exhibit drug concentrations above or
below the recommended therapeutic reference range of 100–320 ng/mL (Hiemke et al. 2011)
6 Monitoring of Plasma Concentrations of Psychotropic Drugs in Pharmacovigilance 97
Pharmacokinetic Variability Pharmacodynamic Variability
Drug concentration in blood Clinical effect Risk for side

effects
Dose range
TDM
Therapeutic Reference Range

Clinical improvement
TDM
Lack of therapeutic effect, morbogenic nisks
Daily dose Drug concentration in blood
Fig. 6.2 The concept of therapeutic drug monitoring (TDM) for guidance of pharmacotherapy.
Under recommended doses and steady concentration, drug in the blood should be within the
therapeutic reference range to attain highest probability of drug response and good tolerability.
In psychiatry, too low or too high drug concentrations are associated with risks. Too low doses/
concentrations bear morbogenic risks due to treatment failure or relapse like psychotic exacerba-
tion or self-harming. Too high drug concentrations can lead to reduced tolerance and poor
compliance or adverse drug reactions
for many drugs to use blood level measurements to guide pharmacotherapy, i.e. to
use therapeutic drug monitoring (TDM). Using TDM, the dose should be corrected
individually as shown in Fig. 6.2 when the drug concentrations in the blood are not
within the therapeutic reference range and the patient did not improve or suffers
from side effects. At concentrations within the therapeutic range, the highest prob-
ability of response and good tolerability is expected. In psychiatry, both too low and
too high drug doses or concentrations are associated with risks. Too low doses or
drug concentrations can be associated with treatment failure or relapse which bear
morbogenic risks such as psychotic exacerbation or self-harming. Too high doses or
drug concentrations can lead to adverse drug reactions or intoxications.
TDM consists not only of determination and reporting of drug concentrations in
blood. It is a process that begins with the indication for TDM and ends with the
treatment decision for the patient. TDM was originally established for drugs with a
narrow therapeutic index, such as digoxin or cyclosporin A, and in psychiatry, for
lithium (Fry and Marks 1971), tricyclic antidepressants (Åsberg et al. 1971; Preskorn
and Fast 1991) or the antipsychotic clozapine (Simpson and Cooper 1978). Primary
aims of TDM in psychiatry were originally related to safety, drug defaulting and
side effects (Preskorn and Fast 1992; Sjöqvist et al. 1980; Touw et al. 2005).
Since the first application of TDM between 1970 and 1980, enormous progress
was made in psychopharmacotherapy not only by the development of new drugs but
also by many new findings on the pharmacokinetics and pharmacodynamics of
available drugs. Cytochromes P450 (CYP) enzymes have been recognized as a
major source of variability in drug pharmacokinetics and response (Zanger et al.
2014). Of 57 putatively functional human CYP enzymes, seven were found to be
relevant for about 90 % of phase I metabolism of psychoactive drugs, CYP1A2,
CYP2B6, CYP2C9, CYP2C19, CYP2D6, CYP2E1 and CYP3A4. The expression
of each CYP is influenced by multiple mechanisms and factors including genetic
polymorphisms, induction by xenobiotics, regulation by cytokines and hormones
and alterations during disease states, as well as sex, age and others (Zanger and
Schwab 2013). Genetic polymorphisms play a major role for the function of
CYP2D6 and CYP2C19 and lead to distinct pharmacogenetic phenotypes termed
poor (PM, no active alleles), intermediate (IM), extensive (EM, basic genotype) and
ultrarapid metabolizers (UM).
Co-medication and other xenobiotics are another important determinant that has
been found to affect the pharmacokinetics of drugs. Fluoxetine, fluvoxamine, par-
oxetine, bupropion, duloxetine, moclobemide and multiple other drugs were identi-
fied as potent inhibitors of distinct CYP enzymes. To a lesser degree, but also
clinically relevant, carbamazepine, compounds in St. John’s wort and other drugs
were found to enhance the expression of CYP3A4. Marked pharmacokinetic drug-
drug interactions can result when combining such inhibitors or inducers with drugs
that are metabolized by inhibited or induced enzymes (victim drugs). The escalating
use of prescribed drugs has led to polypharmacy as a “normal” condition of phar-
macotherapy, in psychiatry as well as in other disciplines (Guthrie et al. 2015), espe-
cially in elderly patients.
Modern TDM uses the above-mentioned psychopharmacology knowledge
gained during the last 40 years in practice for the best possible pharmacotherapy
of individual patients. It thus enhances patient care and patient safety. Modern
pharmacovigilance, which is defined as the science and activities relating to the
detection, assessment, understanding and prevention of adverse effects or any
other drug-related problem (WHO 2002; EMA 2014), also aims to enhance patient
care and patient safety in relation to the use of medicines. Spontaneous reporting
of suspected adverse drug reactions (ADRs) has long been the cornerstone of
pharmacovigilance worldwide for the identification of early signals of problems
of safety related to the use of medicines. However, data generated from spontane-
ous reporting systems are often fragmentary. They are of limited use for the evalu-
ation of observed events. Spontaneously reported adverse drug reactions are
therefore considered as a hint that must be followed by systematic studies. Since
TDM supervises pharmacotherapy under everyday conditions and thus aims to
prevent safety problems related to the use of medicines (Haen 2011; Jaquenoud-
Sirot et al. 2006), TDM and pharmacovigilance thus have the same primary aims.
When applied appropriately, i.e. as recommended in the consensus guidelines for
TDM in psychiatry, collected information is more complete and better structured
(Baumann et al. 2004; Hiemke et al. 2011) than reports in the pharmacovigilance
programs. TDM is therefore potentially a highly suitable tool and source for
pharmacovigilance.
6.2 Monitoring of Plasma Concentrations to Guide

and Supervise Psychopharmacotherapy
The benefits of monitoring of plasma concentrations regarding the optimization of

pharmacotherapy can only be obtained if the method is adequately integrated into
the clinical treatment process. The TDM group of the Arbeitsgemeinschaft für
Neuropsychopharmakologie und Pharmakopsychiatrie (AGNP) issued best practice
guidelines for TDM in psychiatry in 2004 (Baumann et al. 2004) which were
updated in 2011 (Hiemke et al. 2011).
6.2.1 Therapeutic Reference Range and Laboratory

Alert Level
TDM is based on the assumption that there is a relationship between plasma con-
centrations and clinical effects, i.e. therapeutic improvement, side effects and
adverse effects (Hefner et al. 2013). It also assumes that there is a plasma concentra-
tion range of the drug which is characterized by maximal effectiveness and maximal
safety, the “therapeutic reference ranges” (often called “therapeutic window”). The
therapeutic reference range is a key target value to perform TDM.
The therapeutic reference ranges reported in the AGNP guideline define ranges
of medication concentrations which specify a lower limit below which a drug-
induced therapeutic response is relatively unlikely to occur and an upper limit above
which tolerability decreases or above which it is relatively unlikely that therapeutic
improvement may be still enhanced. It is an orienting, population-based range
which may not necessarily be applicable to all patients. Individual patients, e.g.
patients under drug combinations, may show optimal therapeutic response under a
drug concentration that differs from the therapeutic reference range. The therapeutic
reference ranges as recommended by the TDM group of the AGNP for antidepres-
sant, antipsychotic and mood-stabilizing drugs are given in Table 6.1. They were
evidence based and derived from the literature by a structured review process. For
only 15 neuropsychiatric drugs, therapeutic reference ranges based on randomized
clinical trials were found in the literature. For most drugs, reference ranges were
obtained from studies with therapeutically effective doses.
If prospective or retrospective studies on the therapeutic reference range are
lacking, the drug concentration that is expected under steady-state conditions (Css)
can be calculated for the given dose as follows:
Css = DD / CLt
The calculation is based on the direct correlation of the daily dose (DD, constant
dose per day under steady state) to its blood concentration Css, with the total
clearance of the drug (CLt) being the correlation coefficient:
DD × F / t = Css × CLt
Table 6.1 Therapeutic reference ranges and laboratory alert levels of antidepressant, antipsychotic and mood-stabilizing drugs in blood plasma or serum used
100
for therapeutic drug monitoring (TDM)-guided pharmacotherapy

Drug and active metabolite Therapeutic reference range Laboratory alert level Comment
Antidepressant drugs
Amitriptyline plus 80–200 ng/mL 300 ng/mL Elevated concentration in PM of CYP2D6
nortriptyline
Bupropion plus 225–1,500 ng/mL 2,000 ng/mL Bupropion is unstable at room temperature; reference range and
hydroxybupropion alert level refer to hydroxybupropion only
Citalopram 50–110 ng/mL 220 ng/mL Elevated concentration in PM of CYP2C19
Clomipramine plus 230–450 ng/mL 450 ng/mL Elevated concentration in PM of CYP2D6 and CYP2C19
norclomipramine
Desipramine 100–300 ng/mL 300 ng/mL Elevated concentration in PM of CYP2D6
Desvenlafaxine 100–400 ng/mL 600 ng/mL Elevated concentration in PM of CYP2C19
Doxepin plus nordoxepin 50–150 ng/mL 300 ng/mL Elevated concentration in PM of CYP2D6 or CYP2C19
Duloxetine 30–120 ng/mL 240 ng/mL Lower concentration in smokers than in nonsmokers, moderate
inhibitor of CYP2D6
Escitalopram 15–80 ng/mL 160 ng/mL Higher concentration in PM of CYP2C19
Fluoxetine plus norfluoxetine 120–500 ng/mL 1,000 ng/mL Long elimination half-life of norfluoxetine (mean 14 days) and
long-lasting inhibition of CYP2D6
Fluvoxamine 60–230 ng/mL 500 ng/mL Inhibitor of CYP1A2 and CYP2C19
Imipramine plus desipramine 175–300 ng/mL 300 ng/mL Elevated concentration in PM of CYP2D6
Maprotiline 75–130 ng/mL 220 ng/mL Active metabolite N-desmethylmaprotiline
Mianserine 15–70 ng/mL 140 ng/mL
Milnacipran 50–110 ng/mL 220 ng/mL No phase I metabolism
Mirtazapine 30–80 ng/mL 160 ng/mL
Moclobemide 300–1,000 ng/mL 2,000 ng/mL Inhibitor of CYP2C19 and CYP2D6
C. Hiemke and E. Haen
6
Nortriptyline 70–170 ng/mL 300 ng/mL Elevated concentration in PM of CYP2D6

Paroxetine 30–60 ng/mL 120 ng/mL Inhibitor of CYP2D6
Reboxetine 60–350 ng/mL 700 ng/mL
Sertraline 10–150 ng/mL 300 ng/mL
Tianeptin 30–80 ng/mL 160 ng/mL
Tranylcypromin ≤50 ng/mL 100 ng/mL Irreversible inhibition of monoamine oxidase (MAO); drug
actions correlate with MAO activity but not with drug
concentration in blood
Trazodone 700–1,000 ng/mL 1,200 ng/mL
Trimipramine 150–300 ng/mL 600 ng/mL Active metabolites
Venlafaxine plus 100–400 ng/mL 800 ng/mL Elevated concentration of venlafaxine in PM of CYP2D6 or
O-desmethylvenlafaxine CYP2C19
Vortioxetine 10–60 ng/mL 120 ng/mL Elevated concentration in PM of CYP2D6
Antipsychotic drugs
Amisulpride 100–320 ng/mL 640 ng/mL
Aripiprazole 150–500 ng/mL 1,000 ng/mL Active metabolite dehydroaripiprazole
Asenapine 2–5 ng/mL 10 ng/mL
Benperidol 1–10 ng/mL 20 ng/mL Higher levels may be required in patients under long-term
high-dose therapy due to adaptive changes
Bromperidol 12–15 ng/mL 30 ng/mL
Chlorpromazine 30–300 ng/mL 600 ng/mL
Chlorprothixene 20–300 ng/mL 400 ng/mL
Clozapine 350–600 ng/mL 1,000 ng/mL Major metabolite N-desmethylclozapine
Flupenthixol 0.5–5 ng/mL (cis-Isomer) 15 ng/mL
Fluphenazine 1–10 ng/mL 15 ng/mL
Monitoring of Plasma Concentrations of Psychotropic Drugs in Pharmacovigilance
Fluspirilen 0.1–2.2 ng/mL 4.4 ng/mL

(continued)
101
102
Drug and active metabolite Therapeutic reference range Laboratory alert level Comment
Haloperidol 1–10 ng/mL 15 ng/mL Higher levels can be tolerated in patients under long-term
high-dose therapy due to adaptive changes
Iloperidone 5–10 ng/ml 20 ng/ml
Levomepromazine 30–160 ng/mL 320 ng/mL Inhibitor of CYP2D6
Lurasidone 40–120 ng/mL 240 ng/mL
Melperone 30–100 ng/mL 200 ng/mL Inhibitor of CYP2D6
Olanzapine 20–80 ng/mL 100 ng/mL Under olanzapine pamoate, patients exhibited a postinjection
syndrome when drug concentrations exceeded 100 ng/mL
Paliperidone 20–60 ng/mL 120 ng/mL Paliperidone = 9-hydroxyrisperidone
Perazine 100–230 ng/mL 460 ng/mL
Perphenazine 0.6–2.4 ng/mL 5 ng/mL
Pimozide 15–20 ng/mL 20 ng/mL
Pipamperone 100–400 ng/mL 500 ng/mL
Prothipendyl 5–10 ng/mL 20 ng/mL
Quetiapine 100–500 ng/mL 1,000 ng/mL When the patient has taken the extended release (XR)
formulation in the evening and blood was withdrawn in the
morning, plasma concentrations are by mean twofold higher
than trough levels
Risperidone plus 20–60 ng/mL 120 ng/mL Elevated concentration of risperidone in PM of CYP2D6
9-hydroxyrisperidone
Sertindole 50–100 ng/mL 200 ng/mL Active metabolite dehydrosertindole (concentration at therapeutic
doses 40–60 ng/mL)
Sulpiride 200–1,000 ng/mL 1,000 ng/mL Renal elimination
Thioridazine 100–200 ng/mL 400 ng/mL Contraindicated in poor metabolizers of CYP2D6, elevated
concentration
Ziprasidone 50–200 ng/mL 400 ng/mL The drug should be taken with a meal; otherwise absorption is
reduced and plasma concentration will be lower than expected
C. Hiemke and E. Haen
6
Zotepine 10–150 ng/mL 300 ng/mL

Zuclopentixol 4–50 ng/mL 100 ng/mL
Mood-stabilizing drugs
Carbamazepine 4–10 μg/mL 20 μg/mL Active 10,11-epoxide metabolite contributes to clinical effects
Lamotrigine 3–14 μg/mL 30 μg/mL So far no specific reference range for mood-stabilizing effect;
valproate increases elimination half-life to 48–70 h
Lithium 0.5–1.2 mmol/l 1.2 mmol/l Renal elimination
(4–8 μg/mL) (8 μg/mL)
Valproic acid 50–100 μg/mL 120 μg/mL In individual cases, 120 μg/mL is also tolerated in acute mania
Values given are those recommended in the AGNP consensus guidelines for TDM in psychiatry (Hiemke et al. 2011)
CYP cytochrome P450, PM poor metabolizer
103
with DD as dose (mg), F as bioavailability and τ as dosing interval (h). Based on this
information, it is possible to calculate the plasma concentration of a drug that may
be expected in blood under a defined dose (Haen et al. 2008).
In addition to the therapeutic reference range and especially with regard to safety
aspects, the so-called laboratory alert level is another relevant target value for TDM-
guided drug therapies. For most psychotropic drugs, plasma concentrations with an
increased risk of toxicity are normally much higher than the upper threshold levels
of the therapeutic reference ranges shown in Table 6.1. In the TDM guidelines
(Hiemke et al. 2011), a “laboratory alert level” mostly above the upper plasma con-
centration limit of the therapeutic reference range was defined as follows: “The
“laboratory alert levels” (Table 6.1) indicate drug concentrations that cause the
laboratory to feedback immediately to the prescribing physician. The alert levels are
based on reports on intolerance or intoxications and plasma concentration measure-
ments. In most cases, however, it was arbitrarily defined as a plasma concentration
that is twofold higher than the upper limit of the therapeutic reference range. The
laboratory alert should lead to dose reduction when the patient exhibits signs of
intolerance or toxicity. When the high drug concentration is well tolerated by the
patient and if dose reduction bears the risk of symptom exacerbation, the dose
should remain unchanged. The clinical decision, especially in case of unchanged
dose, needs to be documented in the medical file.
6.2.2 TDM Request
TDM should only be requested when there is evidence that the result will provide
an answer to a specific question. Typical indications are shown in Fig. 6.3.
TDM requests must include a completed request form, especially when informa-
tion is to be used for pharmacovigilance issues. An adequate interpretation of the
results is essential to support clinical decision making. The form should contain the
following:
• Patient name or code
• Demographic data
• Diagnosis
• Medication including co-medications
• Reason for the request
• Commercial and the generic name of the drug and its dose
• Galenic formulation
• Time of the last change of the dose
• Time of blood withdrawal
Moreover,
• A brief clinical rating score for evaluation of therapeutic improvement
• A brief side effect rating score
Suggested
Renal or liver non-compliance
dysfunction Intoxication
Aged patients
>65 y Side effect
Children TDM Insufficient

adolescent response
Distinct Drug
genotype combinations
Suggested
Relapse drug-drug
Relapse interaction
prevention
Fig. 6.3 Typical indications to request therapeutic drug monitoring (TDM). Indications related to
pharmacovigilance are marked in grey
should be on the request form (Hiemke et al. 2011).

For objective symptom rating, the clinical global impression (CGI) scale, which
measures severity of illness and therapeutic improvement (Guy 1976), is useful. For
safety purposes, the summary form of the UKU scale (Lingjærde et al. 1987) is use-
ful to evaluate the occurrence and severity of side effects.
A brief comment on the clinical situation must also be given for interpretation of
the results.
6.2.3 Blood Sample Collection, Storage and Shipment
Generally, TDM is carried out in plasma or serum samples. There is no consensus

whether plasma or serum should be preferred. The few available comparisons indi-
cate that values obtained from serum or plasma can be used interchangeably.
Analysis of drugs in other materials such as urine, spinal fluid, tears, hairs or mater-
nal milk has not been introduced for TDM purposes, and no validated data are avail-
able which deal with therapeutic concentrations. Saliva offers the advantage of
non-invasive collection.
With few exceptions, TDM relies on trough steady-state plasma concentrations.

One exception is the indication “side effects”. When they occur, it is useful to know
if they can be attributed to elevated drug concentration and if the dose may be
reduced without loss of efficacy. For concentration measurements under steady
state, blood should be collected after four to five drug elimination half-lives after the
start of or change of dosage. For most psychotropic drugs, elimination half-lives
vary between 12 and 36 h. Notable exceptions are quetiapine, trazodone or venla-
faxine, which display elimination half-lives around 6 h. Fluoxetine and aripiprazole
have longer elimination half-lives. In clinical practice, the appropriate sampling
time for most psychoactive drugs is 1 week after stable daily dosing and immedi-
ately before ingestion of the morning dose, which usually is 12–16 h (or 24 h if the
drug is given once daily in the morning) after the last medication. It is always rec-
ommended to indicate exactly the time of administration of the last dose for
interpretation.
With few exceptions, serum or plasma samples can be stored in the dark (at 4 °C)
for at least 24 h, and most drug samples can be sent without freezing. An exception
is bupropion. Blood serum or plasma must be frozen after blood withdrawal.
Olanzapine must also be stored frozen (−20 °C) if not analysed within 72 h.
6.2.4 Laboratory Measurements
Selective and sensitive analytical methods for the quantitative evaluation of drugs
and their metabolites (analytes) are essential for the successful conduct of
TDM. Methods must be validated. Fundamental parameters for validation include
(1) inaccuracy, (2) imprecision, (3) selectivity, (4) sensitivity, (5) reproducibility
and (6) stability.
For psychoactive drugs, high-performance liquid chromatography (HPLC), in
combination with suitable detection methods or liquid chromatography coupled
with mass spectroscopy (LC-MS) especially tandem MS (LC-MS/MS) methods
are preferred. They are most sensitive and selective and can be used without time-
consuming sample preparation. Many compounds can be analysed simultaneously.
In case of suspected intoxications, TDM methods should enable drug analysis
within 1–2 h. For this purpose, automated methods are advantageous.
The laboratory should not only analyse the drug but also its active metabolites.
The determination of metabolites that do not contribute to the overall clinical
effect can also be useful to monitor drug adherence of the patient, to get informa-
tion on his/her capacity to metabolize drugs or to interpret drug-drug interactions
when drugs are involved exhibiting enzyme-inhibiting or enzyme-inducing prop-
erties. Within the therapeutic reference range, intraday and interday precision
should not exceed 15 % (coefficient of variation), and accuracy should not deviate
more than 15 % from the nominal value. To ensure quality and reliability of
plasma concentrations assays, internal and external quality control procedures are
mandatory.
6.2.5 Reporting of Results
Reporting of results should contain the following information:

• Concentration of the psychoactive drug as well as that of metabolites contribut-
ing to the therapeutic action
• Reference range
• Interpretation and pharmacologic advice
The results should be available for decision making within a clinically meaning-
ful time. A 24 h TDM service is desirable; 48 h is sufficient in most cases. In case
of suspected intoxications, a few hours of service is necessary. To assist rapid inter-
vention in patients at risk for toxicity or loss of tolerability, prompt information
(phone call) of the treating physician is required when the laboratory measures drug
concentrations above the “laboratory alert level”.
Expert interpretation of a drug concentration measurement and the adequate use
of the information are essentials to ensure the full clinical benefit of TDM. Reporting
of results with inclusion of dose recommendations and other comments must be
guided by the best available evidence. Diagnosis and drug dose are important for
interpretation, since they permit a judgement on whether a result is plausible or not.
For the interpretation of the results, it should not only be considered whether the
plasma concentration of the drug is within the “therapeutic reference range”. It must
also be considered if the drug plasma concentration is consistent with the dose. A
plasma concentration may be outside the therapeutic reference range, just because a
low or high dose was taken. Often it is necessary to deal with pharmacokinetic prop-
erties such as metabolic pathways, enzymes involved and substrate and inhibitor
properties of all drugs taken by the patient for interpretation of the results. Supportive
information is given in the TDM guidelines (Hiemke et al. 2011).
Any drug concentration outside its dose-related reference range should alert the
TDM laboratory to actively look for non-average pharmacokinetic drug disposition
of the patient, drug-drug-interactions, gene polymorphisms that give rise to poor or
ultra rapid metabolism, altered function of the excretion organs liver and kidneys,
age and/or disease-related changes in the patient’s pharmacokinetics, compliance
(adherence) problems, a nonsteady state and even signal interference from other
medications that the patient may not have declared to the prescribing physician (e.g.
St. John’s wort) in the laboratory analysis.
Plasma concentrations must be interpreted with the clinical presentation in mind.
Recommendations on dosage changes constitute the most frequent advice. Other
information which could be of help for the physician are those related to genetic
polymorphisms, risks of pharmacokinetic interactions in the case of polypragmasy
and pharmacokinetic properties of the drug in patients belonging to a “special popu-
lation”, e.g. elderly patients, or patients with hepatic or renal insufficiency.
Since interpretation of TDM results may be complex, training in clinical psycho-
pharmacology and pharmacokinetics is essential. Regular conferences with discussion
of the interpretation of real cases are most helpful for learning. It is also recommended
that junior psychiatrists interpret the results under supervision of an expert.
6.3 Computerized Data Entry Ordering and Reporting

Systems for Therapeutic Drug Monitoring
Documented feedback to questionnaires indicates that clinicians often do not want

to put clinical information on the form (Vuille et al. 1991). Moreover, the filled-in
information is often not accurate. Therefore, it is advantageous to use computerized
physician order entry and reporting of results (Bates 1998; Haen 2011), especially
when TDM is to be used for pharmacovigilance research. This technology guides
the ordering physician to give the relevant information required for interpretation in
a comfortable way. Actual laboratory information systems are so far of limited use
for TDM. They collect, record, organize and archive laboratory results. However,
they do not have access to information that is required for interpretation of TDM
results. Pharmacokinetic, pharmacodynamic and marketing characteristics of all
drugs taken by the patient must be considered.
This can be made available by modern information technology: Konbest is a
web-based laboratory information management system (LIMS) for TDM-
laboratories (see www.konbest.de). Konbest consists of a server, a web client, the
software for local hospital servers (electronic patient chart, clinic information sys-
tem, laboratory information system LIS) and expansive pharmacological databases.
The server hosts the following pharmacological databases:
• A database for pharmacokinetic characteristics of drugs
• A database for metabolic pathways of drugs
• A database for composition of commercial drug products
• A database for guidelines of pharmacotherapy
• The implementation of the database www.psiac.de for drug-drug-interactions
The web client coordinates the data flow from hospital to laboratory and allows
interpretation by clinical pharmacological experts via the Internet. A server at hand
for the treating physician stores personalized patient data (demographic data, diag-
nosis, clinical status as assessed by clinical global impression, CGI scale, medica-
tion and clinical variables), checks the medication for potential drug-drug
interactions using a drug interaction database, exports pseudonymized patient data
to be sent to the TDM laboratory and reimports the analysis results together to com-
bine this information with the personalized patient chart.
The Internet platform Konbest supports clinical pharmacological experts in set-
ting up the clinical pharmacological report. Some parts of the report are created
automatically; for other parts, text sequences are suggested to the clinical pharma-
cologist who has to choose one and adapt it to the particular case; a third part of the
report cannot be created by the computer at all; it has to be entered via the clipboard
into Konbest. Finally Konbest creates a pdf file that is send back to the treating
physician as laboratory result report.
Konbest classifies the drug concentration according to these relations in a TDM
9-field-board (Haen 2011): In row 1, all drug concentrations are found that were
influenced by an metabolic pathway induced by drug-drug interaction or by
genetically fast metabolizers; in row 2, all drug concentrations are found that were
quantified in “normal” patients; in row 3, all drug concentrations are found that
were influenced by an metabolic pathway inhibited by drug-drug interaction or by
genetically slow metabolizers.
TDM software like Konbest not only supports the practice of TDM but also
assembles multiple data on the pharmacotherapy under everyday clinical condi-
tions. It thus builds up over time a mine that is suitable for research. Systematic
collection of data on drug exposure, serum concentrations and clinical character-
istics as well as outcomes can generate practice-based evidence. A German-
Swiss-Austrian competence network for TDM in child and adolescent psychiatry
used such approach (Taurines et al. 2013). They compiled a multicentre Internet-
based data infrastructure to document and collect demographic, safety and effi-
cacy data as well as blood concentrations of psychotropic drugs in children and
adolescents (for further information, see www.tdm-kjp.com). More recently, a
large multicenter clinical trial («TDM-VIGIL»), funded by the German Federal
Institute for Drugs and Medical Devices, has begun to collect epidemiological
prescription and safety data of psychotropic drugs in children and adolescents
(Egberts et al. 2015).
6.4 Drug Level Monitoring for Pharmacovigilance
When using drug level monitoring for pharmacovigilance purposes, supervision of

the medication can directly prevent the occurrence of adverse drug reactions. They
can occur when drug concentrations are abnormally high. Typical reasons are drug-
drug interactions when a critical drug combination that contains an inhibitor of a
drug metabolizing enzyme is prescribed. Using TDM, the dose can be adapted.
Other pharmacovigilance-related indications aiming to prevent problems of tolera-
bility or adverse effects are treatment of old-aged patients, children or adolescent
patients and patients with comorbid diseases. In these patients, pharmacokinetically
relevant alterations of renal or liver function may occur. Another group of risk
patients are those with CYP gene variants that give rise to abnormal drug concentra-
tions. For such patients, the dose can be easily adapted by TDM when plasma levels
are measured regularly to avoid critical drug concentrations that may lead to side
effects or loss of action (Fig. 6.2).
When unwanted side effects occur unexpectedly, drug concentration measure-
ment in the blood is useful for clarification. How to do this step by step is shown
schematically in Fig. 6.4. The algorithm indicates the usefulness of drug monitoring
in case of side effects, intoxications or relapses (Fig. 6.5).
When the drug concentration is within the recommended therapeutic reference
range, evidence is given that a pharmacodynamic or a patient-specific problem.
When the drug concentration is lower than expected, CYP-inducing comedication,
heavy smoking or other factors can be relevant. The most frequent reason, however,
is non-adherence to medication. When the latter can be excluded, the dose can be
adapted or changed even if inducing co-medication is taken. For a few patients, an
ultrarapid metabolizer status of CYP2D6 may be relevant. This must be considered
Patient
Blood withdrawal Blood sample
Order entry TDM request
Lab
TDM report Evaluated report staff
Physician
Lab alert
and staff
TDM
server
Research
data
Webclient in the network
of the hospital
Data mining
Fig. 6.4 Information technology (IT) supported therapeutic drug monitoring (TDM) to individually
optimize pharmacotherapy of psychiatric patients. Computerized physician order entry and report-
ing of results support clinical decision making. Collected data can be used as a mine for retrospec-
tive analysis of pharmacotherapy and treatment outcomes under everyday clinical conditions
Side effect, intoxication or relapse

Drug concentration measurement in blood
Drug concentration Drug concentration Drug concentration

as expected and recommended lower than expected higher than expected
Control of medications and smoking habits
Does the list of medications include drugs with CYP inhibitory or

inducing properties and victim drugs?
Discontinuation or change of Adaptation of dose

medication Change of medication
Use of comedication Replace drugs with CYP inhibitory or inducing properties
Use comedication
Are polymorphic drug metabolizing CYP enzymes involved?
Suggested poor metabolizer (PM) Suggested ultra rapid metabolizer (UM)
(Genotyping of CYP2D6 or CYP2C19)
Dose adaptation, change of medication or comedication
Fig. 6.5 Algorithm for the use of drug concentration measurement in the blood. In case of side
effects, intoxications or relapses, it is advisable to measure drug concentrations for problem solv-
ing and for rationale clinical decision making
when the victim drug is a substrate of CYP2D6. For such cases, genotyping can be
useful for clarification (Spina and de Leon 2015).
Case 1
A depressed male patient, aged 60 years, was treated for months with several
antidepressant drugs without any clinical improvement. Other antidepressant
drugs, especially those with noradrenaline reuptake inhibiting activity (doxe-
pin and venlafaxine), were not tolerated. When prescribing escitalopram (up
to 20 mg/day), the medication was well tolerated, but it did not lead to suffi-
cient clinical improvement. Measuring the escitalopram concentration in
blood revealed 10 ng/mL. This was below the therapeutic reference range of
15–80 ng/mL (Table 6.1). Dose increase up to 80 mg which was fourfold
above the recommended maximal dose was required to attain therapeutic lev-
els. Under these conditions, full remission was reached.
For this patient, plasma level monitoring could clarify an abnormally rapid
elimination of escitalopram. CYP2C19 is the predominant enzyme involved in the
degradation of escitalopram. High CYP2C19 activity could not be explained by
CYP2C19 inducing medication or an abnormal genotype. Under TDM control, it was
possible to apply supra-therapeutic doses and thus attain remission. Reasons for non-
response or tolerability problems under other antidepressant medications remained
obscure since plasma concentration measurement was not used for clarification.
Case 2
A schizophrenic male patient, aged 56 years and heavy smoker, was under
stable daily medication with 10 mg olanzapine and 50 mg amitriptyline.
Because of occurrence of psychotic symptoms, the olanzapine dose was
increased to 20 mg. In parallel, an acute infection occurred which was associ-
ated with a severe cough. The patient stopped smoking and had soon a severe
accident by bike (broken shoulder). In the intensive care unit, medication was
discontinued. The consulting psychiatrist recommended TDM, and blood was
taken 48 h after the last intake of olanzapine. Laboratory analysis revealed
113 ng/mL olanzapine which was above the upper threshold level of the refer-
ence range. Amitriptyline and nortriptyline were not detectable. Considering
the elimination half-life, it was extrapolated that the concentration of olanzap-
ine was above 300 ng/mL and thus above the laboratory alert level of 150 ng/
mL. The accident could thus be explained as a drug-induced delirium. Suicidal
ideations were not present.
For this patient, drug level measurement could clarify that the concentration of
olanzapine in the blood had increased because of two reasons. The dose had been
increased from 10 to 20 mg/day. Moreover, cessation of smoking decreased the activ-
ity of CYP1A2. CYP1A2 is relevant for olanzapine, and it is induced by smoking.
The inducing effect of smoke was interrupted due to cessation of smoking.
6.5 Drug Level Monitoring for Detection

and Characterization of Pharmacokinetic Drug-Drug
Interactions
During the last 30 years of psychopharmacotherapy, the most important finding

brought about by TDM with regard to pharmacovigilance was detection of drug-
drug interactions. In a considerable number of cases supervised by TDM clinically
relevant drug-drug interactions were found. Representative examples are given in
Table 6.2.
Using TDM, it was recognized that most new antidepressants are strong inhibi-
tors of CYPs. Fluoxetine, paroxetine, duloxetine and bupropion inhibit CYP2D6;
fluvoxamine inhibits CYP1A2 and CYP2C19; moclobemide CYP2C19 and
CYP2D6; and nefazodone CYP3A4. For all these drugs, their inhibitory potential
was unknown when the drugs were introduced on the market. Testing of CYP
inhibitory and inducing properties of drugs in the early clinical phase of drug
development has therefore become obligatory for drug development.
Observations on drug-drug interactions gave rise to multiple prospective studies
to verify the case report-based findings on drug-drug interactions and to quantify the
magnitude of inhibitory or inducing properties of drugs. Thus drug-drug interac-
tions were studied prospectively between fluvoxamine and imipramine (Spina et al.
1992, 1993b), fluvoxamine or fluoxetine and carbamazepine (Spina et al. 1993a,
1998), fluoxetine or fluvoxamine and tricyclic antidepressant and antipsychotic
drugs (Vandel et al. 1995), phenobarbital and desipramine Spina et al. (1996), flu-
voxamine and clozapine (Szegedi et al. 1995; Wetzel et al. 1998), fluvoxamine and
clomipramine (Szegedi et al. 1996), ketoconazole and carbamazepine (Spina et al.
1997), paroxetine or sertraline and clozapine (Spina et al. 2000b), moclobemide and
dextromethorphan (Härtter et al. 1998), risperidone with carbamazepine and valpro-
ate (Spina et al. 2000a), fluoxetine or paroxetine and risperidone (Spina et al. 2001a,
2002), reboxetine and clozapine or risperidone (Spina et al. 2001b), fluvoxamine
and olanzapine (Hiemke et al. 2002), sertraline and risperidone (Spina et al. 2004),
lamotrigine and clozapine, olanzapine or risperidone (Spina et al. 2006), and val-
proate and olanzapine (Spina et al. 2009).
Drug-drug interactions identified by case reports should always be verified, since
a single case does not allow generalization of the finding. The best way of verifica-
tion is a prospective, well-controlled clinical trial as mentioned above. When case
reports, however, found severe adverse event by a drug combination, prospective
studies can be ethically problematic. Then it is advantageous to conduct retrospec-
tive analysis using collected TDM data especially when computerized data entry
ordering and reporting systems for TDM. When large TDM databases are available,
it is possible to analyse drug-drug interactions retrospectively. Examples for publi-
cations that used this approach are summarized in Table 6.3.
TDM databases can be most informative for pharmacovigilance. So far such
databases are rare. Actual routine laboratory software is not suitable. Clinical data
on diagnoses, psychopathology or side effects or complete information on
6
Table 6.2 Drug-drug interactions identified by TDM in single cases

Inhibitory or
inducing drug Victim drug Effect Suggested mechanism Reference
Fluvoxamine Imipramine Dramatic increase of imipramine and desipramine plasma Inhibition of CYP1A2, Spina et al. (1992)
concentration and side effects CYP2C19
Fluvoxamine Clozapine Increase of clozapine plasma concentration, side effects, Inhibition of CYP1A2 Hiemke et al. (1994)
improved tolerability after dose reduction and CYP219 Szegedi et al. (1996)
Fluvoxamine Methadone 40–100 % increase of methadone plasma concentration Inhibition of methadone Bertschy et al. (1994)
clearance
Fluvoxamine Trimipramine Increase of trimipramine plasma concentration Inhibition of CYP2C19 Seifritz et al. (1994)
and CYP1A2
Fluvoxamine Clomipramine Increase of clomipramine plasma concentration, improved Inhibition of CYP1A2 Conus et al. (1996)
response and CYP2C19
Fluoxetine Methadone Moderate increase of methadone plasma concentration Inhibition of CYP2D6 Bertschy et al. (1996)
and CYP3A4
Oxybutynin Clomipramine Decreased plasma concentrations of clomipramine and Induction of CYP3A4 Grözinger et al. (1999)
norclomipramine
Efavirenz Methadone Decrease of methadone plasma concentration, withdrawal Induction of CYP2B6 Marzolini et al. (2000)
symptoms
Mianserine Venlafaxine Increase of venlafaxine plasma concentration Inhibition of CYP2D6 Eap et al. (2000)
Propofol
Thioridazine
Fluoxetine Tramadol Increased plasma concentration, serotonin syndrome Inhibition of CYP2D6 Lange-Asschenfeldt et al. (2002)
Terbinafine Amitriptyline Prolonged increase of amitriptyline plasma Inhibition of CYP2D6 Castberg et al. (2005)
concentrations, anticholinergic side effects
Fluoxetine Venlafaxine Increase of venlafaxine plasma concentration and side Inhibition of CYP2D6 Gerbaulet et al. (2012)
effects
Cotrimoxazole Venlafaxine Increase of venlafaxine and side effects (tremor) Inhibition of CYP2C9 in Geber et al. (2013)
a PM of CYP2C19
113
Melperone Nortriptyline Increase of nortriptyline plasma concentration, side effects Inhibition of CYP2D6 Hefner et al. (2014)
Table 6.3 Drug-drug interactions identified and characterized by retrospective analysis of TDM
data
Inhibitory or Suggested
inducing drug Victim drug Effect mechanism Reference
Fluvoxamine Clozapine Increase of clozapine Inhibition of Jerling et al.
plasma concentration, CYP1A2 and (1994b)
occurrence of side CYP219
effects
Levomepromazine, Amitriptyline Increase of Inhibition of Jerling et al.
perphenazine, amitriptyline and CYP1A2 and (1994a)
thioridazine plasma concentration, CYP219
occurrence of side
effects
Valproate Risperidone Decrease of risperidone Induction of Spina et al.
Carbamazepine active moiety plasma CYP3A4 (2000a, b)
concentration
Lamotrigine Olanzapine No effect on Botts et al.
Lorazepam olanzapine plasma (2008)
Mirtazapine concentration
Oxcarbazepine
Topiramate
Valproate
Valproate Clozapine Decrease of clozapine Induction of Diaz et al.
plasma concentration clozapine (2014)
metabolism
Esomeprazole Escitalopram Increase of Inhibition of Gjestadt
Lansoprazole Citalopram escitalopram and CYP2C19 et al. (2015)
Omeprazole Sertraline citalopram but not of
sertraline concentration
Pansoprazole
by omeprazole and
esomeprazole
prescribed drugs and doses are mostly missing. Nevertheless, when information is
restricted to plasma concentrations, dose and co-medication, it is possible to evalu-
ate at least pharmacokinetic aspects. Only few institutions have access to TDM data
mines, and when available, they often do not use their databases for research.
Examples for TDM studies that used the data mining approach are shown in
Table 6.3.
6.6 Drug Level Monitoring in Risk Patients
Another example for the usefulness of TDM data mines and pharmacovigilance is
psychiatric risk patients. Pregnant or breastfeeding patients, children or adolescent
patients, individuals with intellectual disabilities, elderly patients, especially
patients aged above 75 years, patients with co-morbid diseases or patients in
forensic psychiatry are generally excluded from clinical trials. Individuals with
intellectual disabilities or patients in forensic psychiatry clinical trials are not
allowed. For such patients, many psychoactive drugs are not approved for use.
Therefore TDM is highly recommended for these patients (Hiemke et al. 2011). For
pregnant or breastfeeding women, TDM aims to minimize the risk of relapse on the
mother’s side and, at the same time, to minimize risks associated with drug expo-
sure of the foetus or the child. Moreover, pharmacokinetics and pharmacodynamics
change during development. Ageing involves progressive impairments of the func-
tional reserve of multiple organs, especially renal excretion, and body composition
changes significantly. Hepatic clearance can be reduced by up to 30 % with phase I
reactions being more likely to be impaired than phase II reactions.
To raise data on the effectiveness and tolerability of psychoactive drugs in these
patients TDM is most useful. First results are available for children and adolescent
psychiatric patients (Cherma et al. 2011; Taurines et al. 2013; Egberts et al. 2011,
2015). For old-aged patients the medication with citalopram and/or venlafaxine was
analysed (Sigurdsson et al. 2014; Unterecker et al. 2012; Wenzel-Seifert et al.
2014). Old-aged patients exhibited a higher risk for adverse reactions under citalo-
pram (Wenzel-Seifert et al. 2014), and a 42 % higher dose-adjusted plasma concen-
tration was found for venlafaxine (Sigurdsson et al. 2014). Castberg and Spigset
(2008) analysed data in a high-security forensic unit and found higher doses in
forensic patients than in a control group. The dose-related plasma concentrations
were significantly lower for olanzapine but higher for quetiapine in the forensic
patients than in the control group. Pfuhlmann and co-workers used TDM data in
conjunction with routine laboratory data (Pfuhlmann et al. 2009). They could thus
show for clozapine that patients with abnormally high levels of serum levels had
significantly more often pathological C reactive protein (CRP) levels, a common
laboratory parameter indicating signs of inflammation. Logistic regression analysis
revealed CRP elevation as the most relevant predictive factor for an increase of
clozapine serum levels.
Overall the number of publications that used the data mining approach is so far
limited. The rapid development of information technology and the broader use of
TDM in the future will certainly have stimulating effects on such efforts.
6.7 Conclusion and Perspective
The common objectives of TDM and pharmacovigilance are supervision of medica-

tion in everyday clinical practice and improvement of tolerability and safety of
medications. When TDM is used appropriately and more frequently than so far,
treatment efficiency and safety of psychotropic medication will improve. TDM has
the potential to become a highly suitable tool for pharmacovigilance. Computerized
drug monitoring will gather valid information on clinical interventions and out-
comes. In conjunction with computerized decision support systems, it will be able
to analyse treatment strategies and outcomes. In psychiatry, there is an urgent need
for perfection of treatment available and a need for studies analysing treatment under
everyday conditions, especially for aged patients and other risk patients who require
pharmacotherapy but are normally excluded from clinical trials. Commercial labora-
tory software is actually not suitable for TDM and pharmacovigilance. Effective
packages are on the way to become available (e.g. www.konbest.de). Information
technology systems can support knowledge-based optimal drug dosing and clinical
decision making in a user-friendly way and thereby build up a platform suitable for
health-care research on drug treatment, especially for pharmacovigilance research.
References
Åsberg M, Crönholm B, Sjöqvist F, Tuck D (1971) Relationship between plasma level and thera-
peutic effect of nortriptyline. Br Med J 3(5770):331–334
Bates DW (1998) Improving the use of therapeutic drug monitoring. Ther Drug Monit
20:550–555
Baumann P, Hiemke C, Ulrich S, Eckermann G, Gaertner I, Gerlach M, Kuss HJ, Laux G, Müller-
Oerlinghausen B, Rao ML, Riederer P, Zernig G (2004) Arbeitsge-meinschaft fur neuropsy-
chopharmakologie und pharmakopsychiatrie. The AGNP-TDM expert group consensus
guidelines: therapeutic drug monitoring in psychiatry. Pharmacopsychiatry 37(6):243–265
Bertschy G, Baumann P, Eap CB, Baettig D (1994) Probable metabolic interaction between metha-
done and fluvoxamine in addict patients. Ther Drug Monit 16(1):42–45
Bertschy G, Eap CB, Powell K, Baumann P (1996) Fluoxetine addition to methadone in addicts:
pharmacokinetic aspects. Ther Drug Monit 18(5):570–572
Botts S, Diaz FJ, Santoro V, Spina E, Muscatello MR, Cogollo M, Castro FE, de Leon J (2008)
Estimating the effects of co-medications on plasma olanzapine concentrations by using a
mixed model. Prog Neuropsychopharmacol Biol Psychiatry 32(6):1453–1458
Castberg I, Spigset O (2008) Prescribing patterns and the use of therapeutic drug monitoring of
psychotropic medication in a psychiatric high-security unit. Ther Drug Monit 30(5):597–603
Castberg I, Helle J, Aamo TO (2005) Prolonged pharmacokinetic drug interaction between terbin-
afine and amitriptyline. Ther Drug Monit 27(5):680–682
Chermá MD, Ahlner J, Bengtsson F, Gustafsson PA (2011) Antidepressant drugs in children and
adolescents: analytical and demographic data in a naturalistic, clinical study. J Clin
Psychopharmacol 31(1):98–102
Conus P, Bondolfi G, Eap CB, Macciardi F, Baumann P (1996) Pharmacokinetic fluvoxamine-
clomipramine interaction with favorable therapeutic consequences in therapy-resistant depres-
sive patient. Pharmacopsychiatry 29(3):108–110
Diaz FJ, Eap CB, Ansermot N, Crettol S, Spina E, de Leon J (2014) Can valproic acid be an
inducer of clozapine metabolism? Pharmacopsychiatry 47(3):89–96
Eap CB, Bertel-Laubscher R, Zullino D, Amey M, Baumann P (2000) Marked increase of venla-
faxine enantiomer concentrations as a consequence of metabolic interactions: a case report.
Pharmacopsychiatry 33(3):112–115
Egberts KM, Mehler-Wex C, Gerlach M (2011) Therapeutic drug monitoring in child and
adolescent psychiatry. Pharmacopsychiatry 44(6):249–253
Egberts K, Karwautz A, Plener PL, Mehler-Wex C, Kölch M, Dang SY, Taurines R, Romanos M,
Gerlach M (2015) Pharmacovigilance in child and adolescent psychiatry. Z Kinder
Jugendpsychiatr Psychother 43(1):21–28
EMA (2014) Guideline on good pharmacovigilance practices http://www.ema.europa.eu/docs/
en_GB/document_library/Scientific_guideline/2013/05/WC500143294.pdf
Fry DE, Marks V (1971) Value of plasma-lithium monitoring. Lancet 1(7705):886–888
Geber C, Ostad Haji E, Schlicht K, Hiemke C, Tadić A (2013) Severe tremor after cotrimoxazole-
induced elevation of venlafaxine serum concentrations in a patient with major depressive
disorder. Ther Drug Monit 35(3):279–282
Gerbaulet M, Engel A, Hiemke C, Dahmen N (2012) High venlafaxine serum levels after prior
fluoxetine medication. J Clin Psychopharmacol 32(2):295–297
Gjestad C, Westin AA, Skogvoll E, Spigset O (2015) Effect of proton pump inhibitors on the serum
concentrations of the selective serotonin reuptake inhibitorscitalopram, escitalopram, and ser-
traline. Ther Drug Monit 37(1):90–97
Grözinger M, Härtter S, Hiemke C, Röschke J (1999) Oxybutynin enhances themetabolism of
clomipramine and dextrorphan possibly by induction of a cytochrome P450 isoenzyme. J Clin
Guthrie B, Makubate B, Hernandez-Santiago V, Dreischulte T (2015) The rising tide of polyphar-
macy and drug-drug interactions: population database analysis 1995–2010. BMC Med 13(1):74
[Epub ahead of print]
Guy W (1976) Early clinical drug evaluation unit (ECDEU) assessment manual for
psychopharmacology. Revised. NIMH publication (DHEW publ No ADM 76-338). National
Institute of Mental Health, Bethesda, p 217–222
Haen E (2011) Therapeutic drug monitoring in pharmacovigilance and pharmacotherapy safety.
Pharmacopsychiatry 44(6):254–258
Haen E, Greiner C, Bader W, Wittmann M (2008) Expanding therapeutic reference ranges using
dose-related reference ranges. Nervenarzt 79(5):558–566
Härtter S, Dingemanse J, Baier D, Ziegler G, Hiemke C (1998) Inhibition of dextromethorphan
metabolism by moclobemide. Psychopharmacology (Berl) 135(1):22–26
Hefner G, Laib AK, Sigurdsson H, Hohner M, Hiemke C (2013) The value of drug and metabolite
concentration in blood as a biomarker of psychopharmacological therapy. Int Rev Psychiatry
25(5):494–508
Hefner G, Geschke K, Hiemke C (2014) Severe adverse drug events under combination of nortrip-
tyline and melperone due to pharmacokinetic interaction. J Clin Psychopharmacol 34(3):
394–396
Hiemke C, Weigmann H, Härtter S, Dahmen N, Wetzel H, Müller H (1994) Elevated levels of
clozapine in serum after addition of fluvoxamine. J Clin Psychopharmacol 14(4):279–281
Hiemke C, Peled A, Jabarin M, Hadjez J, Weigmann H, Härtter S, Modai I, Ritsner M, Silver H
(2002) Fluvoxamine augmentation of olanzapine in chronic schizophrenia: pharmacokinetic
interactions and clinical effects. J Clin Psychopharmacol 22(5):502–506
Hiemke C, Dragicevic A, Gründer G, Hätter S, Sachse J, Vernaleken I, Müller MJ (2004)
Therapeutic monitoring of new antipsychotic drugs. Ther Drug Monit 26(2):156–160
Hiemke C, Baumann P, Bergemann N, Conca A, Dietmaier O, Egberts K, Fric M, Gerlach M,
Greiner C, Gründer G, Haen E, Havemann-Reinecke U, Jaquenoud Sirot E, Kirchherr H,
Laux G, Lutz UC, Messer T, Müller MJ, Pfuhlmann B, Rambeck B, Riederer P, Schoppek B,
Stingl J, Uhr M, Ulrich S, Waschgler R, Zernig G (2011) AGNP consensus guidelines for
therapeutic drug monitoring in psychiatry: update 2011. Pharmacopsychiatry 44(6):
195–235
Jaquenoud Sirot E, van der Velden JW, Rentsch K, Eap CB, Baumann P (2006) Therapeutic drug
monitoring and pharmacogenetic tests as tools in pharmacovigilance. Drug Saf 29(9):
735–768
Jerling M, Bertilsson L, Sjöqvist F (1994a) The use of therapeutic drug monitoring data to docu-
ment kinetic drug interactions: an example with amitriptyline and nortriptyline. Ther Drug
Monit 16(1):1–12
Jerling M, Lindström L, Bondesson U, Bertilsson L (1994b) Fluvoxamine inhibition and
carbamazepine induction of the metabolism of clozapine: evidence from a therapeutic drug
monitoring service. Ther Drug Monit 16(4):368–374
Lange-Asschenfeldt C, Weigmann H, Hiemke C, Mann K (2002) Serotonin syndrome as a result
of fluoxetine in a patient with tramadol abuse: plasma level-correlated symptomatology. J Clin
Lingjærde O, Ahlfors UG, Bech P, Dencker SJ, Elgen K (1987) The UKU side effect rating scale.
A new comprehensive rating scale for psychotropic drugs and a cross-sectional study of side
effects in neuroleptic-treated patients. Acta Psychiatr Scand Suppl 334:1–100
Marzolini C, Troillet N, Telenti A, Baumann P, Decosterd LA, Eap CB (2000) Efavirenz decreases
methadone blood concentrations. AIDS 14(9):1291–1292
Ostad Haji E, Hiemke C, Pfuhlmann B (2012) Therapeutic drug monitoring for antidepressant
drug treatment. Curr Pharm Des 18(36):5818–5827
Pfuhlmann B, Hiemke C, Unterecker S, Burger R, Schmidtke A, Riederer P, Deckert J, Jabs B
(2009) Toxic clozapine serum levels during inflammatory reactions. J Clin Psychopharmacol
29(4):392–394
Preskorn SH, Fast GA (1991) Therapeutic drug monitoring for antidepressants: efficacy, safety,
and cost effectiveness. J Clin Psychiatry 52(Suppl):23–33
Preskorn SH, Fast GA (1992) Tricyclic antidepressant-induced seizures and plasma drug concen-
tration. J Clin Psychiatry 53(5):160–162
Seifritz E, Holsboer-Trachsler E, Hemmeter U, Eap CB, Baumann P (1994) Increased trimipra-
mine plasma levels during fluvoxamine comedication. Eur Neuropsychopharmacol
4(1):15–20
Sigurdsson HP, Hefner G, Ben-Omar N, Köstlbacher A, Wenzel-Seifert K, Hiemke C, Haen E
(2014) Steady-state serum concentrations of venlafaxine in patients with late-life depression.
Impact of age, sex and BMI. J Neural Transm [Epub ahead of print] PubMed
Simpson GM, Cooper TA (1978) Clozapine plasma levels and convulsions. Am J Psychiatry
135(1):99–100
Sjöqvist F, Bertilsson L, Asberg M (1980) Monitoring tricyclic antidepressants. Ther Drug Monit
2(1):85–93
Spina E, de Leon J (2015) Clinical applications of CYP genotyping in psychiatry. J Neural Transm
122(1):5–28
Spina E, Campo GM, Avenoso A, Pollicino MA, Caputi AP (1992) Interaction between fluvoxamine
and imipramine/desipramine in four patients. Ther Drug Monit 14(3):194–196
Spina E, Avenoso A, Pollicino AM, Caputi AP, Fazio A, Pisani F (1993a) Carbamazepine
coadministration with fluoxetine or fluvoxamine. Ther Drug Monit 15(3):247–250
Spina E, Pollicino AM, Avenoso A, Campo GM, Perucca E, Caputi AP (1993b) Effect of fluvox-
amine on the pharmacokinetics of imipramine and desipramine in healthy subjects. Ther Drug
Monit 15(3):243–246
Spina E, Avenoso A, Campo GM, Caputi AP, Perucca E (1996) Phenobarbital induces the
2-hydroxylation of desipramine. Ther Drug Monit 18(1):60–64
Spina E, Arena D, Scordo MG, Fazio A, Pisani F, Perucca E (1997) Elevation of plasma
carbamazepine concentrations by ketoconazole in patients with epilepsy. Ther Drug Monit
19(5):535–538
Spina E, Avenoso A, Facciolà G, Fabrazzo M, Monteleone P, Maj M, Perucca E, Caputi AP (1998)
Effect of fluoxetine on the plasma concentrations of clozapine and its major metabolites in
patients with schizophrenia. Int Clin Psychopharmacol 13(3):141–145
Spina E, Avenoso A, Facciolà G, Salemi M, Scordo MG, Giacobello T, Madia AG, Perucca E
(2000a) Plasma concentrations of risperidone and 9-hydroxyrisperidone: effect of comedica-
tion with carbamazepine or valproate. Ther Drug Monit 22(4):481–485
Spina E, Avenoso A, Salemi M, Facciolá G, Scordo MG, Ancione M, Madia A (2000b) Plasma
concentrations of clozapine and its major metabolites during combined treatment with
paroxetine or sertraline. Pharmacopsychiatry 33(6):213–217
Spina E, Avenoso A, Facciolà G, Scordo MG, Ancione M, Madia A (2001a) Plasmaconcentrations
of risperidone and 9-hydroxyrisperidone during combined treatment with paroxetine. Ther
Drug Monit 23(3):223–227
Spina E, Avenoso A, Scordo MG, Ancione M, Madia A, Levita A (2001b) No effect ofreboxetine
on plasma concentrations of clozapine, risperidone, and their activemetabolites. Ther Drug
Monit 23(6):675–678
Spina E, Avenoso A, Scordo MG, Ancione M, Madia A, Gatti G, Perucca E (2002) Inhibition of
risperidone metabolism by fluoxetine in patients with schizophrenia: a clinically relevant
pharmacokinetic drug interaction. J Clin Psychopharmacol 22(4):419–423
Spina E, D’Arrigo C, Migliardi G, Morgante L, Zoccali R, Ancione M, Madia A (2004) Plasma
risperidone concentrations during combined treatment with sertraline. Ther Drug Monit
26(4):386–390
Spina E, D’Arrigo C, Migliardi G, Santoro V, Muscatello MR, Micò U, D’Amico G, Perucca E
(2006) Effect of adjunctive lamotrigine treatment on the plasma concentrations of clozapine,
risperidone and olanzapine in patients with schizophrenia or bipolar disorder. Ther Drug Monit
28(5):599–602
Spina E, D’Arrigo C, Santoro V, Muscatello MR, Pandolfo G, Zoccali R, Diaz FJ, de Leon J (2009)
Effect of valproate on olanzapine plasma concentrations in patients with bipolar or schizoaffec-
tive disorder. Ther Drug Monit 31(6):758–763
Szegedi A, Wiesner J, Hiemke C (1995) Improved efficacy and fewer side effects under clozapine
treatment after addition of fluvoxamine. J Clin Psychopharmacol 15(2):141–143
Szegedi A, Wetzel H, Leal M, Härtter S, Hiemke C (1996) Combination treatment with clomip-
ramine and fluvoxamine: drug monitoring, safety, and tolerability data. J Clin Psychiatry
57(6):257–264
Taurines R, Burger R, Wewetzer C, Pfuhlmann B, Mehler-Wex C, Gerlach M, Egberts K (2013)
The relation between dosage, serum concentrations, and clinical outcome in children and ado-
lescents treated with sertraline: a naturalistic study. Ther Drug Monit 35(1):84–91
Touw DJ, Neef C, Thomson AH, Vinks AA (2005) Cost-effectiveness of Therapeutic Drug
Monitoring Committee of the International Association for Therapeutic Drug Monitoring and
Clinical Toxicology. Cost-effectiveness of therapeutic drug monitoring: a systematic review.
Ther Drug Monit 27(1):10–17
Unterecker S, Hiemke C, Greiner C, Haen E, Jabs B, Deckert J, Pfuhlmann B (2012) The effect of
age, sex, smoking and co-medication on serum levels of venlafaxine and O-desmethylvenlafaxine
under naturalistic conditions. Pharmacopsychiatry 45(6):229–235
Vandel S, Bertschy G, Baumann P, Bouquet S, Bonin B, Francois T, Sechter D, Bizouard P (1995)
Fluvoxamine and fluoxetine: interaction studies with amitriptyline, clomipramine and neuro-
leptics in phenotyped patients. Pharmacol Res 31(6):347–353
Vuille F, Amey M, Baumann P (1991) Use of plasma level monitoring in clinical practice. Towards
an analysis of clinical utility. Pharmacopsychiatry 24:190–195
Wenzel-Seifert K, Brandl R, Hiemke C, Haen E (2014) Influence of concomitant medications on
the total clearance and the risk for supra-therapeutic plasma concentrations of Citalopram. A
population-based cohort study. Pharmacopsychiatry 47(7):239–244
Wetzel H, Anghelescu I, Szegedi A, Wiesner J, Weigmann H, Härter S, Hiemke C (1998)
Pharmacokinetic interactions of clozapine with selective serotonin reuptake inhibitors: differ-
ential effects of fluvoxamine and paroxetine in a prospective study. J Clin Psychopharmacol
18(1):2–9
World Health Organization (WHO) (2002) The importance of pharmacovigilance: safety monitor-
ing of medicinal products. WHO, Genève
Zanger UM, Schwab M (2013) Cytochrome P450 enzymes in drug metabolism: regulation of gene
expression, enzyme activities, and impact of genetic variation. Pharmacol Ther 138(1):
103–141
Zanger UM, Klein K, Thomas M, Rieger JK, Tremmel R, Kandel BA, Klein M, Magdy T (2014)
Genetics, epigenetics, and regulation of drug-metabolizing cytochrome p450 enzymes. Clin
Pharmacol Ther 95(3):258–261
Chapter 7
The Role of Pharmacogenetics
in Pharmacovigilance of Psychotropic Drugs
Alessandro Serretti and Chiara Fabbri
Abstract Compliance to psychotropic medications is often reduced by the emer-

gence of unwanted side effects, and rare life-threatening adverse events to these
drugs require strict systems of surveillance. Genetic factors are hypothesized to
contribute significantly to the susceptibility of adverse drug reactions (ADRs).
Indeed, about 50 % of ADRs in central nervous system disorders may be attributed
to pharmacogenomic factors.
Some genotype tests such as HLA-B*1502 when using carbamazepine in popu-
lations of Asian descent aid in the prevention of Stevens-Johnson syndrome. This in
addition to testing for CYP2D6 genotypes when using pimozide is already recom-
mended for use in clinical practice. HLA-DQB1 testing for the susceptibility to
clozapine-induced agranulocytosis and CYP2D6 genotyping in case of polyphar-
macy could be recommended in the near future, at least in patients with nongenetic
risk factors. The most promising findings for future clinical applications include the
association between HTR2C, MC4R, leptin genes, and antipsychotic-induced meta-
bolic side effects; DRD2, HTR2A, CYP2D6, HSPG2, and ZFPM2 genes and
antipsychotic-induced movement disorders; SLC6A4, HTR2A, and genes coding for
cytochrome P450 isoenzymes and the overall risk of antidepressant-induced ADRs.
Given the preliminary results supporting improved outcomes (including
improved tolerability) in case of pharmacogenetic testing and possible cost/benefit
ratio improvement, clinical indications for genotyping are expected to increase in
the near future.
Keywords Psychotropic drug • Antidepressant • Antipsychotic • Adverse drug reac-

tion • Side effect • Gene • Polymorphism • Pharmacogenetic • Pharmacogenomic •
Genotyping
A. Serretti (*) • C. Fabbri

Department of Biomedical and Neuromotor Sciences, University of Bologna, Bologna, Italy
e-mail: alessandro.serretti@unibo.it; chiara.fabbri@yahoo.it

DOI 10.1007/978-3-319-24741-0_7
122 A. Serretti and C. Fabbri
7.1 Introduction
7.1.1 The Genetic Contribution to Psychotropic Drug-Induced

Side Effects
Neuropsychiatric disorders account for 19 % of disability-adjusted life years

(DALYs) in Europe, representing the second most frequent cause of disability after
cardiovascular diseases (WHO 2004). Psychotropic medications are not less effec-
tive than other drugs used in general medicine (Leucht et al. 2012), but treatments
tailored to the individual are lacking and compliance to psychotropic drugs is often
low due to stigma and also due to the issue of unwanted side effects. For example,
treatment nonadherence occurs at a rate between 12 % and 64 % among individuals
with bipolar disorder, with consequent increase in the likelihood of relapse and
reduction of quality of life (Leclerc et al. 2013).
When considering psychotropic medications, one of the most relevant categories
of adverse drug reactions (ADRs) is represented by metabolic ADRs. Patients with
severe psychiatric disorders have twice the risk of obesity compared to the general
population (Dickerson et al. 2006) and a higher risk of dyslipidemia, glucose intoler-
ance, and type II diabetes. Consequently, cardiovascular diseases represent a main
cause of mortality in patients with severe psychiatric diseases (Osborn et al. 2007),
making metabolic ADRs in this population a significant health issue. Extrapyramidal
side effects (EPS) and drug-induced sexual dysfunction are further critical ADRs due
to their high incidence and heavy impact on quality of life. EPS are among the most
frequent side effects reported by patients with schizophrenia (almost 60 % of
patients), followed by sedation and weight gain (∼50 % of patients for each) (Millier
et al. 2014). Sexual dysfunction is reported by almost 30 % of patients with schizo-
phrenia, and over 45 % of patients with depression may experience this side effect
associated with antidepressant drugs (Baldwin and Foong 2013). In recent years, the
issue of antidepressant-induced suicidal ideation also emerged, and efforts have been
make to identify risk factors. Finally, some rare but potentially life-threatening ADRs
are due to immune-mediated cutaneous hypersensitivity reactions such as carbam-
azepine- and lamotrigine-induced Stevens-Johnson syndrome (SJS), clozapine-
induced agranulocytosis (CIA), and antipsychotic/antidepressant-induced cardiac
arrhythmia.
Genetic polymorphisms are hypothesized to play a relevant role in determining
the susceptibility to ADRs. It is estimated that genetics account for 20–95 % of vari-
ability in drug disposition and pharmacodynamics and about 50 % of adverse drug
reactions (ADRs) in the central nervous system (CNS) disorders might be attributed
to pharmacogenomic factors (Cacabelos et al. 2012). Twin studies indicate that
genetic polymorphisms contribute 60–80 % of the variance observed in antipsychotic-
induced weight gain (Gebhardt et al. 2010).
Some clinical applications of pharmacogenetics are already available or will
probably be available in the near future (Table 7.1). Such pharmacogenetic data
could guide drug choice and/or dose titration. Figure 7.1 represents a schema of
current clinical-based and future genotype-clinical-based treatment choice.
7 The Role of Pharmacogenetics in Pharmacovigilance of Psychotropic Drugs 123
Table 7.1 Recommended genotyping test for the prevention of severe psychotropic-induced
adverse reactions (ADRs) and promising genes/polymorphisms for clinical application in the near
future
Comment on
Gene Polymorphism(s) Drug/drug class Type of ADR evidence
HLA-B HLA-B*1502 Carbamazepine Stevens-Johnson Recommended in
syndrome Asian ancestry
CYP2D6 Partially or totally Pimozide Arrhythmia Recommended for
inactive alleles doses >4 mg/day in
adults and >0.05 mg/
kg/day in children
CYP2D6 Partially or totally Polypharmacy Overall risk of High evidence,
inactive alleles ADRs probably will be
recommended
POLG A467T, W748S Valproate Liver toxicity Recommended in
children/adolescents
CPS1 rs1047891 Hyperammonemia Recommended in
case of suspected
urea cycle disorder
HLA- 6672G>C Clozapine Agranulocytosis Probably will be
DQB1 recommended in the
near future
HLA-B 158T Promising
HTR2C −759C/T Antipsychotics Metabolic ADRs Promising
MC4R rs489693 Promising
Leptin −2548A/G Promising
CNR1 rs806378, Promising
rs1049353
HTR2A rs6311, rs6313 Antidepressants Overall Promising
SLC6A4 5-HTTLPR, tolerability Promising
rs25531
DRD2 rs1800497 Antipsychotics Tardive dyskinesia Promising
HTR2A 102CC, −1438GG Promising
CYP2D6 Partially or totally Promising
inactive alleles
HSPG2 rs2445142 Parkinsonism Promising
ZFPM2 rs12678719 Promising
7.1.2 Pharmacogenetics in Focus
Approximately 0.5 % of the DNA sequence is responsible for phenotype (i.e., somatic)
differences among humans. This difference consists in di-, tri-, and tetranucleotide
repeats (satellite sequences) and large variants >1 kbp due to deletions, insertions, or
duplications (copy number variants, CNV) and nucleotide substitutions. Over three
million substitutions distinguish the individual genome, and over 80 % of them are in
the form of single-nucleotide substitution polymorphism (SNP). Therefore, it has
been estimated that SNPs account for over 80 % of the variability between humans,
1. Current approach: drug and dose are chosen according to clinical evaluation.
Drug and dose Switch in case of poor

Laboratory and/or
selection on the response and/or
diagnostic tests (e.g. ECG, Clinical monitoring
basis of the clinical tolerability until
blood count)
experience satisfactory outcome is
obtained (trial and error)
2. Future approach: drug and dose are chosen on the basis of clinical evaluation assisted by genotyping.
Laboratory and/or
diagnostic tests (e.g. ECG, Expected improvement
blood count) in cost/effectiveness
Drug and dose
selection on the basis
of clinical experience Clinical monitoring
+ + genotype
Potential application of
genotype information for
other medications involving
Genotyping the same
pharmacodynamics/
pharmacokinetics
Fig. 7.1 Schematic description of the current approach in the choice of psychotropic medications
(1) and expected future approach based on both clinical judgment and genotype information (2)
including liability to ADRs (Roberts et al. 2010). The Human Genome Project (HGP),
which started in 1990 and was completed in 2003, with further analysis still being
published, has made possible to determine the sequence of chemical base pairs which
make up DNA and to identify and map the approximately 20,000–25,000 genes of the
human genome from both a physical standpoint and functional standpoint.
The largest part of available data regarding the pharmacogenetics of ADRs has
been obtained through candidate gene studies. Candidate genes are selected on the
basis of their biological role and polymorphisms on the basis of their functional role
(i.e., a known impact on gene function resulting in a variation in the level/function
of the product), tagging properties (linkage disequilibrium with near variants), or
position (in regulatory regions). In the last decade, genome-wide association studies
(GWAS) were introduced and have rapidly expanded, since they provide hundreds
of thousands of variants, thanks to the array technology without the need of any a
priori hypothesis. Obviously, GWAS have some limitations (in particular the limited
covering of genetic polymorphisms provided by the currently available platforms
and the common difficulty in explaining the biological meaning of findings).
7.2 Pharmacogenetics of Antidepressant-Induced Side Effects
Antidepressant drugs are among the most frequently prescribed drugs worldwide.
Selective serotonin reuptake inhibitors (SSRIs) and serotonin-noradrenaline reup-
take inhibitors (SNRIs) are usually better tolerated than first-generation
antidepressants (tricyclic antidepressants or TCAs; inhibitors of monoamino oxi-

dase or MAOIs). Nevertheless, some patients require treatment with the latter group
of antidepressants or are genetically predisposed to develop idiosyncratic ADRs at
therapeutic dosages (an abnormal effect due to polymorphisms in metabolic
enzymes or target molecules) or to have an increased risk of overdose from drug
with a well-defined and wide therapeutic index. Some patients are at higher risk of
severe ADRs because of medical comorbidities, concomitant medications, and/or
old age. Finally, the issue of treatment-induced suicidal ideation (TESI) when pre-
scribing antidepressants led regulatory authorities to issue warnings to clinicians
(US Food and Drug Administration 2006). Pharmacogenetics can provide clinically
useful information on the safe use of antidepressant drugs, especially in patients at
increased risk of ADRs due to individual, pharmacological, and/or environmental
factors.
7.2.1 Treatment-Emergent Suicidal Behavior (TESI)
Family studies support a genetic contribution to suicidal behavior (SB) (Brent and
Mann 2005). Candidate gene studies were focused especially on serotonin-related
genes since low central nervous system (CNS) serotonin (5-HT) turnover was dem-
onstrated in SB (Mann 2003). Consistently, the association of SB with variants in
the serotonin transporter gene (SLC6A4) (Li and He 2007) and the tryptophan
hydroxylase 1 gene (TPH1 that codes for the rate-limiting enzyme responsible for
5-HT biosynthesis) (Bellivier et al. 2004) was supported at meta-analytic level. The
noradrenergic system and especially the alpha 2A-adrenergic receptor gene
(ADRA2A) have also been implicated in SB (Escriba et al. 2004; Sequeira et al.
2004) with an effect that may be higher in patients treated with noradrenergic anti-
depressants and in males (Perroud et al. 2009). The noradrenergic system is impli-
cated in the modulation of aggressive and impulsive behaviors, and enhanced
noradrenergic activity may have a role in treatment-emergent suicidal behavior
(TESI). Thus, the enhanced activity or hypersensitivity of ADRA2A receptors may
be associated with higher suicidal ideation during treatment with noradrenergic
antidepressants.
Other candidate gene studies suggested the involvement of brain-derived neuro-
trophic factor gene (BDNF) and the gene coding for its receptor, the neurotrophic
tyrosine kinase receptor type 2 (NTRK2) (Perroud et al. 2008). These genes are
involved in the regulation and growth of 5-HT neurons and are mediators of neural
plasticity in response to acute and chronic stress. The cyclic adenosine monophos-
phate (cAMP) response element binding (CREB1) protein gene is involved in the
regulation of BDNF expression, and it has also been associated with TESI (Perlis
et al. 2007).
In addition to dysregulation in the monoaminergic and neurotrophic systems, the
hypothalamic-pituitary-adrenal (HPA) axis and inflammatory pathways have been
also proposed as modulators of TESI. Indeed, altered sensitivity to glucocorticoids
and increase in proinflammatory cytokines were demonstrated in depression and

particularly in suicide victims. BDNF and 5-HT positively modulate neurogenesis
in the hippocampus (a pivotal area involved in depression pathogenesis), and the
hippocampus in turn regulates HPA axis function and response to stress. In depressed
subjects, hippocampal atrophy (due to reduced cell proliferation, cell survival, and
cell differentiation) promotes impaired regulation of HPA axis activity (Mahar et al.
2014). FK506-binding protein 5 (FKBP5) gene, which codes for a protein that
decreases the sensitivity of the glucocorticoid receptor to the effect of corticoste-
roids, was suggested as a modulator of TESI (Mandelli and Serretti 2013).
The role of glutamate in modulating mood and antidepressant response has been
increasingly recognized, due to observations that existing antidepressants modulate
various glutamatergic pathways. Disrupted glutamatergic-noradrenergic interac-
tions at the level of the stress-sensitive locus coeruleus (LC) were demonstrated in
depression and suicide victims (Chandley et al. 2014). Accordingly, polymorphisms
in glutamate receptor genes GRIK2 and GRIA3 were associated with TESI during
SSRI treatment (Laje et al. 2007).
Genome-wide association studies (GWAS) did not report the above genes among
their top findings but outlined other genes that may involved in TESI. Associations
were found for genetic variants within the loci encoding papilin (PAPLN) and IL-28
α-receptor (IL28RA) genes (Laje et al. 2009) and in the vicinity of the guanine
deaminase (GDA) gene (Perroud et al. 2012). IL28RA encodes a cytokine receptor,
while papilin is involved in the regulation of extracellular matrix remodeling, a
process that affects the release of bioactive fragments that function as immune mod-
ulators (Korpos et al. 2009). The involvement of these genes in TESI is therefore
consistent with the inflammation theory of depression and SB. On the other hand,
GDA encodes an enzyme responsible for the hydrolytic deamination of guanine and
is probably involved in microtubule assembly. No clear biological rationale links
this gene to TESI. A recent development of GWAS is based on multimarker analy-
ses given the hypothesis that multiple genetic variants contribute to complex pheno-
types such as TESI. A cluster of 79 SNPs demonstrated a 94 % probability of
predicting the nonoccurrence of TESI (negative predictive value), even with only a
48 % probability of correctly identifying TESI (positive predictive value) (Menke
et al. 2012), but with no confirmation of this finding so far.
Given the complex and multifactorial pathogenesis of TESI, it seems unlikely
that genotyping could be able to prevent it adequately.
7.2.2 Cardiovascular Side Effects
Tricyclic antidepressants (TCAs) have a higher risk of cardiovascular side effects

compared to other antidepressants, even in patients with no previous cardiovascular
disease (Pacher and Kecskemeti 2004). The most common among such side effects
is the slowing of intraventricular conduction, manifested by prolonged PR, QRS,
and QT intervals on the standard ECG, and orthostatic hypotension. TCAs have
been demonstrated to exert I/A class antiarrhythmic effects and antinoradrenergic

and anticholinergic effects that are responsible for their cardiovascular ADRs.
The identification of genetic polymorphisms predicting the risk of intraventricu-
lar conduction alterations induced by antidepressants could help clinicians to pre-
vent life-threatening side effects. So far, a number of genes have been associated
with arrhythmia: SCN5A, SCN4B, CACNL1AC, KCNH2, KCNQ1, KCNE1, ANK2,
ALG10, KCNJ2, KCNE2, RYR2, KCND3, KCND2, ACE, NOS1AP, CASQ2, and
Rad (Drago et al. 2008). These genes are good candidates for the definition of a
genetic proarrhythmic profile, but evidence is still lacking for antidepressant drugs.
Changes in systemic blood pressure are possible especially during treatment
with antidepressants that affect the activity of the noradrenergic system (norepi-
nephrine (NE) reuptake, transport, and elimination from the synapse). Blood pres-
sure changes are mediated through the autonomic nervous system in part by the
neurotransmitter NE. Selective NE transporter (NET) blockade creates a phenotype
that resembles idiopathic orthostatic intolerance (Schroeder et al. 2002), but the
gene encoding this transporter (SLC6A2) was very marginally associated with blood
pressure changes during treatment with duloxetine (an SNRI antidepressant drug).
Other noradrenergic genes (ADRB2, coding for the adrenergic beta 2 receptor, and
COMT, encoding the main enzyme involved in NE metabolism) do not apparently
play a role in the risk of blood pressure changes during treatment with duloxetine
(Fijal et al. 2013). In addition, 5-HT binding to the serotonin-2A receptor has been
associated with vasoconstriction and hypertension, and HTR2A gene has consis-
tently been associated with essential hypertension in women (Liolitsa et al. 2001).
Nevertheless, evidence suggesting an effect of this gene on blood pressure increase
during treatment with duloxetine appears limited (Fijal et al. 2013).
The SSRIs fluoxetine and paroxetine demonstrated a relatively high receptor
affinity for adrenergic beta receptors in vitro, suggesting an increased propensity of
affecting cardiovascular parameters compared to other SSRIs. Interestingly, a poly-
morphism in the gene encoding the adrenergic beta 1 receptor (ADRB1) was found
to modulate blood pressure and heart rate values in patients treated with these
antidepressants (Thomas et al. 2010), but the potential clinical impact of this appears
to be limited.
Finally, a group of clinically relevant side effects affecting the cardiovascular
system are hemorrhagic complications, whose risk is increased during treatment
with SSRI antidepressants, especially in some at-risk conditions (e.g., concomitant
treatment with anticoagulant or antiaggregant drugs and surgery). 5-HT is a strong
vasoconstrictor and a relatively weak platelet activator. At rest, 5-HT is stored in
platelets, but after platelet activation, it is released into the circulation, together with
other aggregating factors such as adenosine diphosphate (ADP) and adrenaline, and
it acts as a stimulus for platelet aggregation. The 5-HT transporter is present on
platelet surface and is necessary to transport 5-HT into the platelet, since platelets
themselves do not produce 5-HT but are dependent on its uptake from the blood.
Thus, blockade of the 5-HT transporter with an SSRI leads to a lower concentration
of 5-HT in the platelet. Given this mechanisms, SLC6A4 gene (encoding the 5-HT
transporter) appears to be an optimal candidate for affecting the risk of bleeding
during SSRI treatment. An insertion/deletion polymorphism (named 5-HTTLPR) in

the promoter region of SLC6A4 showing two variant alleles (a short (S) allele and a
long (L) allele) has been particularly studied since the S variant is associated with a
nearly 50 % reduction in basal expression of the 5-HT transporter (Heils et al.
1996). Despite the interesting functional impact of this polymorphism, the available
knowledge does not suggest any major impact of this variant on platelet function
(Abdelmalik et al. 2008; Hougardy et al. 2008).
In conclusion, no genetic variants are known to significantly affect the risk of
cardiovascular side effects during antidepressant treatment, but very few studies
were focused on this topic. The identification of polymorphisms affecting the risk of
proarrhythmic effects appears to be the most relevant issue under the clinical point
of view.
7.2.3 Weight Gain, Sexual Dysfunction, and Other Side Effects
Sexual dysfunction, weight gain, and gastrointestinal side effects (dry mouth, con-
stipation, diarrhea, and nausea) are the most frequent side effects induced by antide-
pressant drugs and thus the most frequently responsible for early treatment
discontinuation. On the other hand, antidepressant-related hyponatremia is a rela-
tively rare but potentially fatal antidepressant-induced ADR.
Some polymorphisms (5-HTTLPR, rs25531, intron 2 VNTR, or STin2) of the
5-HT transporter gene (SLC6A4) have been repeatedly investigated for association
with the overall risk of antidepressant-induced side effects. Available evidence
mainly suggests that carriers of the 5-HTTLPR/rs25531 short alleles show lower
treatment tolerability while STin2 probably does not exert a significant influence on
this phenotype (Garfield et al. 2014; Fabbri et al. 2013). HTR1A rs6295 and HTR2A
rs6311/rs6313 are also promising variants that contribute to individual antidepres-
sant tolerability (Fabbri et al. 2013; Garfield et al. 2014).
Weight gain is a quite common side effect of treatment with antidepressant drugs.
Genes pertaining to the serotonergic system have been particularly investigated in
relation to this ADR, since 5-HT has been implicated in the control of eating behav-
ior and body weight by hypothalamic serotonergic receptor mechanisms (De Vry
and Schreiber 2000). HTR2C gene (coding for 5-HT2C receptor) was proposed as a
modulator of the risk of weight gain during both antidepressant and antipsychotic
therapies (Altar et al. 2013). The stimulation of hypothalamic 5-HT2C receptors
leads to a behaviorally specific hypophagic effect by accelerating satiety processes
(De Vry and Schreiber 2000), providing a biological rationale that supports the
pharmacogenetic finding.
Catechol-O-methyltransferase (COMT) and tryptophan hydroxylase type I
(TPH1) encode pivotal enzymes in the catabolism and synthesis of 5-HT, respec-
tively. COMT rs4680 and TPH1 rs18532 were demonstrated to modulate the risk of
weight gain during antidepressant treatment independently from age and gender
(Secher et al. 2009). GNB3 (that encodes the β3 subunit of the G protein complex)
is involved in the downstream signaling cascade following monoamine receptor

activation. A functional polymorphism in this gene (C825T) was associated with the
improvement of neurovegetative symptoms of depression during treatment, and the
TT genotype was found to be a predictor of greater weight gain during treatment
with nortriptyline (a drug of the TCA class) (Keers et al. 2011).
ADRA2A gene (encoding adrenergic α2 receptor) is a primary target of mirtazap-
ine, a noradrenergic and specific serotonergic antidepressant (NaSSA). Mirtazapine
is more likely to cause weight gain or increased appetite compared to SSRIs or
SNRIs (Watanabe et al. 2011; Lee et al. 2009); thus the effect of ADRA2A-1291C/G
polymorphism on the risk of weight gain during treatment with this antidepressant
should be investigated further.
Sexual dysfunction in patients with major depression may be triggered or exac-
erbated by treatment with antidepressants (especially SSRIs, TCAs, and venlafax-
ine), with a prevalence of up to 50–70 % (Fava and Rankin 2002). Serotonergic,
dopaminergic, noradrenergic, and glutamatergic systems are those primarily
hypothesized to be involved in the pathogenesis of this ADR. In general, reduction
of 5-HT function facilitates, whereas enhancement inhibits, sexual behavior, with
the 5-HT transporter, 5-HT1A, 5-HT1B, 5-HT2A/B, and 5-HT7 receptors being the
primary molecular players involved (Olivier et al. 2011). Furthermore, the blockade
of 5-HT transporters may decrease the concentrations of dopamine and noradrena-
line in the mesolimbic system by activating serotonin 5-HT2C receptors (Strohmaier
et al. 2011); these two neurotransmitters play a role in the modulation of sexual
arousal and sexual motivation. Indeed, dopamine antagonists inhibit copulation
among male rats, whereas agonists have the opposite (facilitating) effect. Similar
effects were demonstrated also for glutamate antagonists and agonists, respectively
(Dominguez et al. 2006; Dominguez and Hull 2005).
The promoter polymorphism 5-HTTLPR rs25531 of the 5-HT transporter gene
(SLC6A4) has been investigated as predictor of antidepressant-induced sexual dys-
function. Results suggested a higher risk in high-expressing genotypes (long alleles)
of the polymorphism (Garfield et al. 2014), and the effect appears to be dependent
on age (Strohmaier et al. 2011). HTR1A and HTR2A genes may also contribute to
the development of this ADR (Garfield et al. 2014; Bishop et al. 2006).
Consistently with the involvement of glutamate in the modulation of sexual behav-
ior, in SSRI-treated patients, multiple genes encoding glutamatergic receptors were
associated with a decrease in libido (GRIA3 and GRIK2), difficulty achieving orgasm
(GRIA1), and difficulty developing an erection (GRIN3A) (Perlis et al. 2009).
Currently, no data are available on the contribution of polymorphisms in noradrener-
gic and dopaminergic genes to the risk of antidepressant-induced sexual dysfunction.
Gastrointestinal ADRs are sometimes disabling side effects that usually emerge
in the initial phases of antidepressant treatment. Variants in genes belonging to the
serotonergic system were the most studied in relation to these ADRs, since 80 % of
the body 5-HT stores are located in enterochromaffin cells of the gut and serotonin
plays a central role in the regulation of motility and secreting activity of the gastro-
intestinal tract. HTR3B gene was associated with paroxetine-induced gastrointesti-
nal side effects (Tanaka et al. 2008; Sugai et al. 2006), and HTR2A was shown to
exert a synergistic effect with CYP2D6 gene polymorphisms in the prediction of

fluvoxamine-induced gastrointestinal side effects (Suzuki et al. 2006). CYP2D6
gene was associated also with venlafaxine (a SNRI)-induced gastrointestinal ADRs
(Shams et al. 2006) and with broad antidepressant-induced ADRs (Rau et al. 2004).
Polymorphisms in genes encoding cytochrome P450 enzymes (CYP450) are
responsible for variations in drug metabolizing rapidity, including antidepressant
drugs. Consequently, genetic variations in CYP450 genes could influence the occur-
rence of several ADRs, including the risk of overdose; this issue is discussed in the
section “Toxicity from Overdose.” P-glycoprotein (P-gp), an ATP-driven efflux
pump that regulates the uptake of drugs through organ barriers, is another protein
influencing the pharmacokinetics of antidepressants through the regulation of drug
distribution in the body. Polymorphisms in the gene coding for this transporting
protein (ABCB1) may influence gastrointestinal complaints and sexual side effects
during antidepressant therapy (de Klerk et al. 2013).
Hyponatremia is a potentially fatal side effect of antidepressant drugs. It occurs
in one in 200 elderly patients per year receiving fluoxetine and paroxetine, two com-
monly used SSRIs (Wilkinson et al. 1999). There is little pharmacogenetic evidence
with regard to this ADR, but lower mean serum sodium concentrations were shown
to be present in CYP2D6 poor metabolizers (PMs) in comparison with CYP2D6
extensive metabolizers (EMs). As a result, CYP2D6 PMs might be at increased risk
of developing hyponatremia (Kwadijk-de Gijsel et al. 2009).
GWAS did not outlined particularly interesting findings for any of the above
ADRs, but different genes were proposed as putative candidates compared to candi-
date gene studies. SACM1L (coding for the phosphatidylinositide phosphatase
SAC1) gene was associated with bupropion-induced sexual dysfunction, even though
this antidepressant is not commonly responsible for this ADR, resulting in limited
clinical utility. SAC1 is an integral membrane protein of the endoplasmic reticulum
and the Golgi apparatus that plays a direct role in growth factor signaling; thus alter-
ations in the activity of this enzyme may lead to disruptions in the cellular secretory
machinery and hormone-neurotransmitter secretion, with possible consequences on
sexual functioning (Clark et al. 2012). A later extension of the study by Clark et al.
suggested that EMID2 (EMI domain containing 2) gene may affect SSRI-induced
vision/hearing side effects, LAMA1 (laminin, alpha 1) gene and the rs16965962 SNP
(in a gene desert on chromosome 7) may influence overall SSRI tolerability, while
AOX2P gene may be related with dizziness. EMID2 encodes the protein collagen α-1
chain that is involved in the regulation of corneal collagen fibrillogenesis (Rada et al.
1993). On the other hand, for LAMA1 and AOX2P, it is not so easy to hypothesize a
biological rationale explaining the reported GWAS findings.
7.2.4 Toxicity from Overdose
The risk of overdose from antidepressant drugs was reduced significantly after the
introduction of SSRIs and SNRIs compared to earlier antidepressants (TCAs and
MAOIs). Despite the reduction of life-threatening reactions, 5-HT toxicity can still
result from serotonin excess in the CNS from serotonergic drugs. Serotonin syn-
drome is a potentially life-threatening condition characterized by myoclonus,
hyperreflexia, sweating, shivering, incoordination, and mental status changes.
Furthermore, the risk of cardiac arrhythmia is another potentially fatal manifesta-
tion following overdose especially from TCAs but possibly also from venlafaxine
and IMAOs.
The genetic variants that were hypothesized to influence the risk and severity of
overdose symptoms are mainly those in genes coding for CYP450 enzymes and
P-glycoprotein (P-gp) gene (ABCB1), the products of which are involved in the
distribution and metabolism of antidepressant drugs.
CYP2D6 and CYP2C19 encode the P450 isoenzymes that are mostly involved in
antidepressant metabolism. The level of CYP enzyme activity is dependent on
genetic polymorphisms and allows the distinction of different metabolizing groups.
The wild-type genotype results in extensive metabolizers (EM), while the interme-
diate metabolizer (IM) is characterized by the presence of one wild-type allele plus
a partially or totally defective allele. Poor metabolizers (PMs) have a combination
of two partially or totally defective alleles, and the ultrarapid metabolizer (UM)
category exists only for CYP2D6 and is usually due to multiple copies of normal
alleles.
The available evidence suggests that CYP2D6 PMs have lower tolerance to
TCAs as well as to venlafaxine (a SNRI drug), whereas they have an average
tolerance to other antidepressants. Based on literature, CYP2D6 PMs are
expected to have a concentration-to-dose ratio (C:D ratio) of 4–6 for TCAs,
whereas CYP2D6 EMs are expected to have a C:D ratio of 0.5–1.5. Dose adjust-
ments for different metabolizing groups were calculated, even if prospective
validations should be performed before routine clinical application (Porcelli
et al. 2011). Polypharmacy is likely to represent a valid indication for CYP2D6
genotyping to minimize the risk of toxicity from drug-drug interactions in PMs
(Laje 2013).
Finally, P-gp limits drug uptake into key organs such as the brain, and fatal intox-
ication from venlafaxine overdose was associated with C1236T and C3435T poly-
morphisms in ABCB1 gene (Karlsson et al. 2013).
7.3 Pharmacogenetics of Antipsychotic-Induced Side Effects
Antipsychotic drugs are the mainstay of treatment for schizophrenia and related
disorder and have improved schizophrenia prognosis significantly since their intro-
duction in the 1950s. In recent years, the use of second-generation antipsychotics
(SGAPs) has been indicated also for the treatment of several phases of bipolar dis-
order or augmentation in major depressive disorder with benefits being seen espe-
cially in more severe cases. Both first-generation antipsychotics (FGAPs) and
SGAPs carry the risk of severe and sometimes debilitating ADRs, whose clinical
relevance has been increased in proportion with the expansion of their clinical use
in terms of increasing indications and treatment duration.
7.3.1 Antipsychotic-Induced Weight Gain
Weight gain is a major health problem encountered during treatment with antipsychot-
ics especially SGAPs due to a high risk of obesity and other metabolic conditions
(Dickerson et al. 2006). Weight gain is associated with significant variability among
individuals, and genetic factors play an important role, estimated to be around 60–80 %
through twin and family studies (Gebhardt et al. 2010). Given that cardiovascular dis-
ease is the primary cause of excess of mortality among severe psychiatric diseases
(Osborn et al. 2007), the identification of genetic predictors of antipsychotic metabolic
ADRs could be a turning point in the treatment of schizophrenia and bipolar disorder.
Several pharmacogenetic studies have investigated the genes that could influence
antipsychotic-induced weight gain (AIWG), with focus mainly on homeostatic reg-
ulators expressed in hypothalamic areas that belong to the complex network that
regulates appetite and satiety.
The most replicated pharmacogenetic association is the serotonergic receptor
5-HTC2 (HTR2C) gene that is responsible for 5-HT central anorexigenic action on
the hypothalamic nuclei. Consistently, antagonists of 5-HTC2 receptors, such as
clozapine and olanzapine, promote appetite increase (Bonhaus et al. 1997). Carriers
of the minor T allele of the promoter polymorphism −759C/T appear to be protected
from substantial gain in weight (Sicard et al. 2010).
Leptin and melanocortin receptor 4 (MC4R) are other essential components of
one of the most important hypothalamic satiety signals. Leptin is mainly synthesized
in the adipocytes of white adipose tissue and activates leptin receptors in the arcuate
nucleus of the hypothalamus, resulting in a feeling of satiety. The leptin-2548A/G
polymorphism may interact with the HTR2C −759C/T variant in affecting AIWG
(Reynolds 2012). Neurons of the arcuate nucleus also express MC4R, activation of
which decreases food intake while elevating energy utilization (Fani et al. 2014).
This gene has been consistently associated with AIWG by four independent studies
(Shams and Muller 2014).
Cannabinoid receptor 1 (CNR1) gene and the fatty acid amide hydrolase (FAAH)
gene have been suggested as genetic factors involved in AIWG (Shams and Muller
2014), consistently with the observation that they play an important role in the
mediation of leptin anorexigenic action. Some SNPs within CNR1 in particular pro-
vided encouraging findings.
Other candidate genes provided less convincing evidence for an association with
AIWG, among them being ghrelin (GHRL) and neuropeptide Y (NPY), which act as
antagonists of the leptin-induced satiety signal, other hypothalamic neuroendocrine
regulators (FTO and PMCH genes), and histamine receptor 1 (HRH1 gene, consis-
tently to the observation that central histaminergic transmission contributes to the
modulation of the motivational aspect of appetite and physical activity (Torrealba
et al. 2012)). The T allele of GNB3C825T polymorphism was considered to be par-
ticularly interesting since it is associated with a G protein β3 splice variant and
previously described associated with obesity in several ethnic groups. Nevertheless,
the available evidence mainly does not support the involvement of this variant in
AIWG (Souza et al. 2008). Furthermore, polymorphisms in AMP-activated protein
kinase (AMPK gene, a central molecule integrating nutrient and hormonal signals to
maintain energy homeostasis), relaxin-3 (RLN3, a member of the insulin/relaxin

pathway) and its receptors (RXFP3 and RXFP4), tumor necrosis factor-α (TNF-a, a
proinflammatory cytokine), and methylenetetrahydrofolate reductase (encoded by
MTHFR, involved in nuclear methylation and gene expression regulation) were sug-
gested to modulate AIWG (Muller et al. 2013; Kao and Muller 2013).
A GWAS was aimed to identify genetic risk factors for metabolic side effects in
patients treated with psychopharmacological medications (Athanasiu et al. 2012).
SNP rs7838490 (8q21.3 region) was associated with BMI alterations, while
rs11615724 (12q21) was associated with the effect of medications on decreasing
HDL-C levels. Both markers are in intergenic regions. rs7838490 is located
upstream of the gene matrix metalloproteinase 16 (MMP16), and it may regulate the
expression of MMP16 and affect tumor necrosis factor receptor superfamily, mem-
ber 1A (sTNFRSF1A), which may be involved in lipid regulation.
7.3.2 Antipsychotic-Induced Extrapyramidal Side Effects
The greater part of pharmacogenetic studies focused on genetic risk factors of

antipsychotic-induced tardive dyskinesia, the most severe among extrapyramidal
side effects (EPS) due to its tendency to persist over time, its treatment resistance,
and its high frequency (around 20 % of patients after prolonged treatment (Kane
et al. 1988)). Genes are hypothesized to be a relevant factor in the risk of tardive
dyskinesia, as suggested by increased risk of tardive dyskinesia in affected families
(Muller et al. 2013). Genes that influence the pharmacokinetics, pharmacodynam-
ics, and oxidative stress associated with antipsychotics have been studied for tardive
dyskinesia risk. In particular, the gene coding for the dopamine receptor 2 (DRD2),
one of the main targets of antipsychotics (especially for those drugs with a higher
risk of inducing this ADR), was suggested to be involved in the risk of this EPS by
various independent studies (especially the Taq1A polymorphism or rs1800497).
The minor (T) Taq1A allele has been associated with a 40 % reduction in striatal D2
receptor density (according to in vitro assays and in vivo imaging studies); this
allele appears to be protective against tardive dyskinesia (Lencz and Malhotra
2009). DRD2 may also be implicated in the risk of akathisia, even though only pre-
liminary evidence is available (Muller et al. 2013). A functional missense mutation
in dopamine 3 receptor (DRD3, another target of antipsychotics) gene (ser9gly or
rs6280) has been suggested to modulate the risk of tardive dyskinesia, but meta-
analytic results did not support this hypothesis (Tsai et al. 2010). 5-HT2A receptor
(HTR2A) gene is a target of atypical antipsychotics, and it has been implicated in
their reduced extrapyramidal side effect profile (Meltzer 2012). HTR2A gene has
also been associated with tardive dyskinesia susceptibility by several candidate
gene studies (Segman et al. 2001; Gunes et al. 2007; Wilffert et al. 2009).
The highly polymorphic gene CYP2D6 is responsible for the hepatic metabolism
of several commonly prescribed antipsychotics. CYP2D6 poor metabolizer status
(homozygosity for null alleles) or intermediate metabolizer status (null allele het-
erozygosity) were associated with 1.64- and 1.43-fold greater odds of developing
tardive dyskinesia on the basis of literature meta-analysis (Patsopoulos et al. 2005).
Other genes were investigated in relation to antipsychotic-induced EPS, among

which were the glutamate receptor GRIN2A, BDNF, TNF-a (Muller et al. 2013), and
SOD2 (Lencz and Malhotra 2009). SOD2 encodes for manganese superoxide dis-
mutase, a mitochondrial enzyme involved in oxidative metabolism. SOD2 and TNF-
a may be involved in the risk of tardive dyskinesia through neuron oxidative stress.
GWAS results shed light on the potential contribution of different genes to
antipsychotic-induced tardive dyskinesia. In particular, the gamma-aminobutyric
acid (GABA) receptor signaling pathway (especially SLC6A11, GABRB2, and
GABRG3 genes) may be involved in genetic susceptibility to treatment-resistant
tardive dyskinesia in Asian populations (Inada et al. 2008). On the other hand, the
involvement of HSPG2 gene (coding for the heparan sulfate proteoglycan 2) was
suggested by independent GWAS on both Asian (Syu et al. 2010) and Caucasian
(Greenbaum et al. 2012a) populations, thus representing the strongest candidate
outlined by GWAS. This pharmacogenomic finding was supported by an increase in
HSPG2 expression in subjects with lower risk of developing tardive dyskinesia,
which may exert a protective effect via a cholinergic or basic fibroblast growth fac-
tor (FGF2)-mediated neuroprotective mechanism.
Some data are available also for other types of EPS. A GWAS investigating
antipsychotic-induced Parkinsonism (Alkelai et al. 2009) has been replicated
(Greenbaum et al. 2012b) supporting a role of the rs12678719 SNP in the zinc fin-
ger protein multitype 2 (ZFPM2) gene, especially in patients of African ancestry.
The risk allele (G) was associated with lower FP-CIT uptake, which is indicative of
a higher degree of nigrostriatal terminal degeneration, in Parkinson disease subjects
as assessed by SPECT (Greenbaum et al. 2012b). Interestingly, another gene coding
for a zinc finger protein (ZNF202) was associated with the development of abnor-
mal movements during antipsychotic treatment (Aberg et al. 2010). The ZNF202 is
a transcriptional repressor controlling, among other genes, PLP1, which is the
major protein expressed in myelin. Mutations in PLP1 can determine the develop-
ment of parkinsonism (Willard and Riordan 1985).
7.3.3 Neuroleptic Malignant Syndrome, Clozapine-Induced

Agranulocytosis, and Antipsychotic-Induced QT
Prolongation
Neuroleptic malignant syndrome (NMS) is a severe complication of treatment with

antipsychotic drugs. Initial reports of familial clustering of NMS have suggested a
genetic basis. As a result, some pharmacogenetic case-control association studies
were performed in Japanese populations which were mainly focused on CYP2D6
and DRD2 polymorphisms, though with inconsistent findings (Ferentinos and
Dikeos 2012). Further pharmacogenetic studies should be performed to elucidate
the genetic contribution to NMS.
Clozapine-induced agranulocytosis (CIA) is a rare (incidence 0.8 %) but poten-
tially fatal ADR of clozapine, which limits its use in treatment-resistant schizophre-
nia. Several pharmacogenetic case-control studies focused on CIA have produced
inconsistent findings, mainly regarding the major histocompatibility complex

(MHC) region (including human leukocyte antigen (HLA) class I, II, and III loci
and some non-HLA genes), as well as a few non-MHC genes. HLA-DQB1 poly-
morphisms have been the ones most consistently associated with CIA (Ferentinos
and Dikeos 2012; Goldstein et al. 2014). Among them, a SNP (6672G>C) was
found to confer a 16.9 times increased risk of CIA. A commercial test using this
variant was marketed in 2007, but its clinical application is limited by its low sensi-
tivity and low predictive validity (Ferentinos and Dikeos 2012).
The risk of sudden death for patients receiving antipsychotics has been estimated
to be 2.4 times the risk of untreated controls (Ray et al. 2001). This increased risk is
partly attributable to antipsychotic drugs causing QT prolongation (e.g., pimozide,
thioridazine, sertindole, and haloperidol). Genetic research on the QT interval has
initially identified several rare variants which cause congenital Mendelian QT syn-
dromes (long QT and short QT syndromes), both increasing the risk of syncope and
sudden death by predisposing to torsades de pointes and atrial fibrillation,
respectively (Hedley et al. 2009). Some of these rare mutations have been found in
patients with medication-induced QT prolongation; however, only 5–15 % of per-
sons experiencing drug-induced torsades de pointes carry a mutation in one of the
genes associated with hereditary long QT syndrome (Paulussen et al. 2004).
Two recent GWAS of antipsychotic-induced QT prolongation have identified
potentially involved polymorphisms in several genes, including those coding for
ceramide kinase-like protein (CERKL), solute carrier organic anion transporter 3A1
(SLCO3A1), paladin (PALLD), solute carrier family 22, member 23 (SLC22A23),
nucleotide-binding protein-like (NUBPL) gene, and nitric oxide synthase 1 adaptor
protein (NOS1AP) gene (the most frequently replicated candidate gene for
QT-interval variation in the general population) (Ferentinos and Dikeos 2012).
These findings suggest that some of the genes mediating antipsychotic-induced QT
prolongation are unique, whereas others partially overlap with the genes affecting
normal QT-interval variation.
Finally, given that pimozide is one of the highest-risk antipsychotics regarding
proarrhythmic effects, the impact of CYP2D6 (the CYP450 isoform mainly involved
in its metabolism) genotype on pimozide exposure was investigated. Dose-ranging
analyses revealed CYP2D6 poor metabolizers should not receive more than 4 mg
daily to avoid plasma concentrations in excess of those observed in extensive
metabolizer and intermediate metabolizer receiving 10 mg daily. Thus, CYP2D6
genotyping is now recommended in the pimozide product label before exceeding
4 mg of pimozide daily in adults or 0.05 mg/kg/day in children (Rogers et al. 2012).
7.3.4 Prolactin Elevation
Few pharmacogenetic studies have investigated this frequent antipsychotic side

effect, occurring in 80–90 % of all female subjects treated with the highest-risk
drugs, i.e., risperidone, amisulpride, and sulpiride (Peuskens et al. 2014), and the
conclusions from such studies is unclear. Dopamine D2 receptor gene (DRD2) is an
attractive candidate, given that dopamine 2 antagonism plays an important role in

the increase in prolactin via the tuberoinfundibular pathway. The DRD2Taq1A
polymorphism was associated with increased risk for hyperprolactinemia by four
independent studies, but conflicting negative results also exist (Lencz and Malhotra
2009). The gene coding for the cytochrome P450 2D6 (CYP2D6) enzyme is another
interesting candidate explaining variation in prolactin levels after treatment with
antipsychotic medications. CYP2D6 metabolizer status was associated with plasma
levels of different antipsychotic drugs such as risperidone, but no clear evidence of
an impact on this side effect has been demonstrated (Novalbos et al. 2010).
7.4 Pharmacogenetics of Mood Stabilizer-Induced Side

Effects
Mood stabilizers are used primarily in the treatment of bipolar disorders (for both
acute phase and prophylaxis), but other clinical indications include the augmentation
of antidepressant treatment in unipolar depression and the treatment of symptoms
such as impulsivity in severe personality disorders. Some of them (e.g., valproate,
carbamazepine, and lamotrigine) are also indicated for the treatment of some types
of seizure disorders. The most frequent and potentially severe side effects of these
medications are psychological/neurological (hypersomnia, sedation, retardation,
tremors), endocrine/metabolic (body weight gain and metabolic dysfunction, kidney
and thyroid dysfunction for lithium, hyperammonemia for valproate), gastrointesti-
nal (nausea, abdominal pain, diarrhea, and hepatic dysfunction for valproate), hema-
tologic (leucopenia, anemia, thrombocytopenia), and teratogenic. Immune-mediated
cutaneous hypersensitivity reactions are the most common idiosyncratic reactions to
antiepileptic drugs such as carbamazepine and lamotrigine. These ADRs include a
risk of potentially life-threatening Stevens-Johnson syndrome (SJS), toxic epider-
mal necrolysis (TEN), and drug-related rash with eosinophilia and systemic symp-
toms (DRESS). Unfortunately, fewer pharmacogenetic data are available for mood
stabilizers compared to antipsychotics and antidepressants.
7.4.1 Lithium- and Valproate-Induced Side Effects
Few and mainly negative pharmacogenetic findings are available for lithium-related
ADRs (e.g., Zill et al. 2003). Lithium has been linked to the development of primary
hyperparathyroidism and parathyroid tumors (Dwight et al. 2002). The majority of
lithium-associated parathyroid tumors were reported to be independent of classical
chromosomal alterations associated with this type of cancer (Dwight et al. 2002),
suggesting a potential specific genetic susceptibility. In any case, few data are avail-
able to indicate an increased risk linked to specific genetic mutations.
Regarding valproate-induced ADRs, a missense polymorphism in carbamoyl
phosphate synthase 1 (CPS1) gene (4217C>A or rs1047891) was suggested to
increase the risk and severity of hyperammonemia, even if drug plasmatic levels are
within the therapeutic range (Bezinover et al. 2011; Janicki et al. 2013). In the
Japanese population on the other hand, the association between rs1047891 and
hyperammonemia may not be present (Inoue et al. 2014).
Valproate-induced liver impairment was related to mitochondrial DNA depletion
and mutations in POLG gene that codes for the mitochondrial DNA polymerase
gamma in pediatric patients (Pronicka et al. 2011; Saneto et al. 2010). Thus, POLG
gene testing has been recommended in children/adolescents since they are particu-
larly at risk of developing valproate-induced liver toxicity (FDA 2013a). The
involvement of POLG mutations in the pathogenesis of this ADR was suggested
also by in vitro experiments on human cell lines (Stewart et al. 2010). Other genes
that may be involved in the risk of valproate hepatotoxicity are those responsible for
the metabolism of the drug. The principal pathways of valproate metabolism
involved glucuronidation and β-oxidation and cytochrome P450 system. Glutathione
S-transferases (GSTs) comprise a supergene family of enzymes that catalyze the
inactivation of a variety of endogenous and exogenous products. Polymorphisms in
GSTM1 and GSTT1 genes were suggested to be potential predictors of increase in
serum gamma-glutamyltransferase in patients treated with valproate, but these are
not necessarily markers of valproate-induced liver toxicity (Franciotta et al. 2009).
In rare cases, valproic acid can be metabolized to the active and hepatotoxic metab-
olite, 4-ene-valproic acid, but it is not yet clear whether genetic variants of the
involved enzyme (CYP2C9) are responsible for this problem (Klotz 2007).
Superoxide dismutase 2 (coded by the SOD2 gene) plays a critical role in the detox-
ification of mitochondrial reactive oxygen species, and SNPrs4880 (Val16Ala) was
suggested as a modulator of valproate-induced elevation of gamma-
glutamyltransferase (γ-GT, a hepatic and biliary enzyme) (Ogusu et al. 2014).
Twin studies suggest that genetic factors have an influence on the weight change
induced by valproate (Klein et al. 2005). The C825T variation of GNB3 gene was
associated with higher plasma total cholesterol, triglyceride, leptin levels, and body
mass index in bipolar patients treated with valproate, suggesting that T allele carri-
ers at this locus may have a lower risk of metabolic ADRs during valproate treat-
ment (Chang et al. 2010). The same variant in GNB3 was also associated with
weight gain during treatment with nortriptyline (see paragraph 2.3) and proposed as
a modulator of AIWG (see paragraph 3.1). Unfortunately, no other pharmacoge-
netic data exist for metabolic ADRs during mood stabilizer treatment.
7.4.2 Carbamazepine- and Lamotrigine-Induced

Hypersensitivity Reactions
Severe hypersensitivity reactions (cutaneous or systemic) to carbamazepine and

lamotrigine have a frequency that ranges between 1 and 10 per 10,000 new users
(Zaccara et al. 2007). The risk of developing this type of ADR was initially associ-
ated with a genetic polymorphism in the promoter region of the proinflammatory
cytokine tumor necrosis factor-alpha (TNF-a) and with variants in genes coding for
heat shock protein-70 isoforms (Franciotta et al. 2009). These associations were
nevertheless not independently replicated. A breakthrough came in 2004 when a
100 % prevalence of carbamazepine-induced SJS was reported among Han Chinese
carriers of the human leukocyte antigen HLA-B*1502 allele, compared with a fre-
quency of this allele of only 3 % among carbamazepine-tolerant patients (Chung
et al. 2004). Further case-control studies in Hong Kong Chinese and Thai popula-
tions confirmed the strong association of HLA-B*1502 with carbamazepine-
induced SJS or TEN, but not with carbamazepine-induced maculopapular rash or
DRESS. Ethnicity appears to play an important role in the association, since studies
in Caucasians and in Japanese failed to identify any relationship between HLA-
B*1502 status and SJS or TEN (Franciotta et al. 2009). On the basis of these data,
the US Food and Drug Administration (FDA) recommended that patients with
ancestry from areas in which HLA-B*1502 is present (China, Thailand, Malaysia,
Indonesia, the Philippines, Taiwan, and Vietnam) should be screened for this allele
before starting treatment with carbamazepine. Given that over 90 % of drug-induced
SJS/TEN occur within 2 months of starting treatment, patients who have been tak-
ing carbamazepine for at least a few months without developing severe cutaneous
reactions are at low risk of developing SJS or TEN during continuation of treatment,
even if they carry the HLA-B*1502 allele (FDA 2013b).
Although much attention has been focused on HLA-B*1502, other HLA geno-
types were investigated as potential predictors of cutaneous reactions. In a Chinese
population, maculopapular reactions to carbamazepine were associated with SNPs
in the HLA-E region and with the HLA-A*3101 allele, whereas DRESS was associ-
ated with polymorphisms in motilin (MLN) gene that is located in the MHC class II
terminal region (Hung et al. 2006). In Caucasians, the HLA-B*0702 allele has been
found to be potentially protective against severe carbamazepine-induced hypersen-
sitivity, while the HLA-B*38 allele was identified as a possible risk factor for
lamotrigine-induced SJS or TEN (Franciotta et al. 2009).
7.5 Conclusion
Current knowledge of the genetic factors involved in the risk of adverse drug reac-
tions (ADRs) during treatment with psychotropic drugs is often not univocal, and
clinical applications are still rare. Twin and family studies indicate a genetic contri-
bution of about 50 % to the overall risk of ADRs in CNS disorders (Cacabelos et al.
2012) and a contribution varying from 60 % to 80 % for a clinically relevant ADR
such as antipsychotic-induced weight gain (Gebhardt et al. 2010). As with other
complex phenotypes, pharmacogenetic studies often failed to detect polymorphisms
with the expected effect sizes, possibly due to confounders such as gene-gene or
gene-environment interactions, heterogeneity in ancestry, and age. Despite a num-
ber of preliminary/unconfirmed findings, some pharmacogenetic findings are quite
promising for future clinical applications, and a limited number have been already
recommended for use in clinical practice (Table 7.1). Testing for HLA-B*1502
when using carbamazepine in Asian populations to prevent Stevens-Johnson syn-

drome, CYP2D6 when using pimozide, POLG in children/adolescents treated with
valproate, and CPS1 before prescribing valproate in case of suspected urea cycle
disorder is included in the latter group.
The first studies evaluating the benefits of clinical pharmacogenetic testing were
performed, with promising results. Preliminary estimates in a simulated trial indi-
cated that SLC6A4 genetic testing may improve antidepressant tolerability in terms
of quality-adjusted life weeks and incremental cost/effectiveness ratio in patients
suffering from at least two depressive episodes (Serretti et al. 2011). In patients with
depressive/anxiety disorders treated with antidepressant or antipsychotic drugs, a
genetic test (GeneSight) for variations in cytochrome P450 genes (CYP2D6,
CYP2C19, CYP2C9, and CYP1A2), serotonin transporter gene (SLC6A4), and sero-
tonin 2A receptor gene (5HTR2A) was used to distinguish between them in three
categories according to the predicted outcome. Patients that were classified in the
“red” group (medication status “use with caution and frequent monitoring”) had
69 % more total health care visits, 67 % more general medical visits, over threefold
more medical absence days, and over fourfold more disability claims than subjects
taking drugs categorized in the “green” group (“use as directed”) or “yellow” group
(“use with caution”). The result is attributable to increased side effects and poor
outcome in the “red” group (Winner et al. 2013).
A better definition of relevant genes and polymorphisms with the subsequent
production of targeted genetic chips represents a fundamental step for future devel-
opments in this field. Evidence regarding genetic test cost/benefit ratio and specific
clinical indications for genotyping should be provided through randomized trials.
The benefit deriving from genotype information to guide the choice of different
medications prescribed to the patient lifelong should be considered.
On the basis of the results provided by existing studies and current clinical appli-
cations for a number of commercialized drugs (pharmacogenetic test information is
currently included in over 200 drug labels among those approved in the United
States (Ikediobi et al. 2009)), indications to genotyping are expected to increase in
the psychiatric field in the next years.
References
Abdelmalik N, Ruhe HG, Barwari K, van den Dool EJ, Meijers JC, Middeldorp S, Buller HR,
Schene AH, Kamphuisen PW (2008) Effect of the selective serotonin reuptake inhibitor parox-
etine on platelet function is modified by a SLC6A4 serotonin transporter polymorphism. J
Thromb Haemost JTH 6(12):2168–2174. doi:10.1111/j.1538-7836.2008.03196.x
Aberg K, Adkins DE, Bukszar J, Webb BT, Caroff SN, Miller DD, Sebat J, Stroup S, Fanous AH,
Vladimirov VI, McClay JL, Lieberman JA, Sullivan PF, van den Oord EJ (2010) Genomewide
association study of movement-related adverse antipsychotic effects. Biol Psychiatry
67(3):279–282. doi:10.1016/j.biopsych.2009.08.036
Alkelai A, Greenbaum L, Rigbi A, Kanyas K, Lerer B (2009) Genome-wide association study of
antipsychotic-induced parkinsonism severity among schizophrenia patients.
Psychopharmacology (Berl) 206(3):491–499. doi:10.1007/s00213-009-1627-z
Altar CA, Hornberger J, Shewade A, Cruz V, Garrison J, Mrazek D (2013) Clinical validity of
cytochrome P450 metabolism and serotonin gene variants in psychiatric pharmacotherapy. Int
Rev Psychiatry 25(5):509–533. doi:10.3109/09540261.2013.825579
Athanasiu L, Brown AA, Birkenaes AB, Mattingsdal M, Agartz I, Melle I, Steen VM, Andreassen
OA, Djurovic S (2012) Genome-wide association study identifies genetic loci associated with
body mass index and high density lipoprotein-cholesterol levels during psychopharmacologi-
cal treatment – a cross-sectional naturalistic study. Psychiatry Res 197(3):327–336.
doi:10.1016/j.psychres.2011.12.036
Baldwin DS, Foong T (2013) Antidepressant drugs and sexual dysfunction. Br J Psychiatry J Ment
Sci 202:396–397. doi:10.1192/bjp.bp.112.110650
Bellivier F, Chaste P, Malafosse A (2004) Association between the TPH gene A218C polymor-
phism and suicidal behavior: a meta-analysis. Am J Med Genet B Neuropsychiatr Genet Off
Publ Int Soc Psychiatr Genet 124B(1):87–91. doi:10.1002/ajmg.b.20015
Bezinover D, Postula M, Donahue K, Bentzen B, McInerney J, Janicki PK (2011) Perioperative
exacerbation of valproic acid-associated hyperammonemia: a clinical and genetic analysis.
Anesth Analg 113(4):858–861. doi:10.1213/ANE.0b013e318228a001
Bishop JR, Moline J, Ellingrod VL, Schultz SK, Clayton AH (2006) Serotonin 2A–1438 G/A and
G-protein Beta3 subunit C825T polymorphisms in patients with depression and SSRI-
associated sexual side-effects. Neuropsychopharmacol Off PublAm Coll Neuropsychopharmacol
31(10):2281–2288. doi:10.1038/sj.npp.1301090
Bonhaus DW, Weinhardt KK, Taylor M, DeSouza A, McNeeley PM, Szczepanski K, Fontana DJ,
Trinh J, Rocha CL, Dawson MW, Flippin LA, Eglen RM (1997) RS-102221: a novel high affin-
ity and selective, 5-HT2C receptor antagonist. Neuropharmacology 36(4–5):621–629
Brent DA, Mann JJ (2005) Family genetic studies, suicide, and suicidal behavior. Am J Med Genet
C Semin Med Genet 133C(1):13–24. doi:10.1002/ajmg.c.30042
Cacabelos R, Martinez-Bouza R, Carril JC, Fernandez-Novoa L, Lombardi V, Carrera I, Corzo L,
McKay A (2012) Genomics and pharmacogenomics of brain disorders. Curr Pharm Biotechnol
13(5):674–725
Chandley MJ, Szebeni A, Szebeni K, Crawford JD, Stockmeier CA, Turecki G, Kostrzewa RM,
Ordway GA (2014) Elevated gene expression of glutamate receptors in noradrenergic neurons
from the locus coeruleus in major depression. Int J Neuropsychopharmacol/Off Sci J Coll Int
Neuropsychopharmacol 17(10):1569–1578. doi:10.1017/S1461145714000662
Chang HH, Gean PW, Chou CH, Yang YK, Tsai HC, Lu RB, Chen PS (2010) C825T polymor-
phism of the GNB3 gene on valproate-related metabolic abnormalities in bipolar disorder
patients. J Clin Psychopharmacol 30(5):512–517. doi:10.1097/JCP.0b013e3181f03f50
Chung WH, Hung SI, Hong HS, Hsih MS, Yang LC, Ho HC, Wu JY, Chen YT (2004) Medical
genetics: a marker for Stevens-Johnson syndrome. Nature 428(6982):486. doi:10.1038/428486a
Clark SL, Adkins DE, Aberg K, Hettema JM, McClay JL, Souza RP, van den Oord EJ (2012)
Pharmacogenomic study of side-effects for antidepressant treatment options in
STAR*D. Psychol Med 42(6):1151–1162. doi:10.1017/S003329171100239X
de Klerk OL, Nolte IM, Bet PM, Bosker FJ, Snieder H, den Boer JA, Bruggeman R, Hoogendijk
WJ, Penninx BW (2013) ABCB1 gene variants influence tolerance to selective serotonin reup-
take inhibitors in a large sample of Dutch cases with major depressive disorder.
Pharmacogenomics J 13(4):349–353. doi:10.1038/tpj.2012.16
De Vry J, Schreiber R (2000) Effects of selected serotonin 5-HT(1) and 5-HT(2) receptor agonists
on feeding behavior: possible mechanisms of action. Neurosci Biobehav Rev 24(3):
341–353
Dickerson FB, Brown CH, Kreyenbuhl JA, Fang L, Goldberg RW, Wohlheiter K, Dixon LB (2006)
Obesity among individuals with serious mental illness. Acta Psychiatr Scand 113(4):306–313.
doi:10.1111/j.1600-0447.2005.00637.x
Dominguez JM, Hull EM (2005) Dopamine, the medial preoptic area, and male sexual behavior.
Physiol Behav 86(3):356–368. doi:10.1016/j.physbeh.2005.08.006
Dominguez JM, Gil M, Hull EM (2006) Preoptic glutamate facilitates male sexual behavior. J
Neurosci Off J Soc Neurosci 26(6):1699–1703. doi:10.1523/JNEUROSCI.4176-05.2006
Drago A, De Ponti F, Boriani G, De Ronchi D, Serretti A (2008) Strategy for a genetic assessment
of antipsychotic and antidepressant-related proarrhythmia. Curr Med Chem 15(24):
2472–2517
Dwight T, Kytola S, Teh BT, Theodosopoulos G, Richardson AL, Philips J, Twigg S, Delbridge L,
Marsh DJ, Nelson AE, Larsson C, Robinson BG (2002) Genetic analysis of lithium-associated
parathyroid tumors. Eur J Endocrinol/Eur Fed Endocr Soc 146(5):619–627
Escriba PV, Ozaita A, Garcia-Sevilla JA (2004) Increased mRNA expression of alpha2A-
adrenoceptors, serotonin receptors and mu-opioid receptors in the brains of suicide victims.
Neuropsychopharm Off Publ Am Coll Neuropsychopharm 29(8):1512–1521. doi:10.1038/
sj.npp.1300459
Fabbri C, Di Girolamo G, Serretti A (2013) Pharmacogenetics of antidepressant drugs: an update
after almost 20 years of research. Am J Med Genet B Neuropsychiatr Genet Off Publ Int Soc
Psychiatr Genet 162B(6):487–520. doi:10.1002/ajmg.b.32184
Fani L, Bak S, Delhanty P, van Rossum EF, van den Akker EL (2014) The melanocortin-4 receptor
as target for obesity treatment: a systematic review of emerging pharmacological therapeutic
options. Int J Obes (Lond) 38(2):163–169. doi:10.1038/ijo.2013.80
Fava M, Rankin M (2002) Sexual functioning and SSRIs. J Clin Psychiatry 63(Suppl 5):13–16;
discussion 23–15
FDA (2013a) FDA label for valproic acid and OTC, POLG. https://www.pharmgkb.org/label/
PA166104825. Accessed 14 Dec 2014
FDA (2013b) Information for healthcare professionals: dangerous or even fatal skin reactions –
carbamazepine (marketed as Carbatrol, Equetro, Tegretol, and generics). http://www.fda.gov/
Drugs/DrugSafety/PostmarketDrugSafetyInformationforPatientsandProviders/ucm124718.
htm. Accessed 11 Dec 2014
Ferentinos P, Dikeos D (2012) Genetic correlates of medical comorbidity associated with schizo-
phrenia and treatment with antipsychotics. Curr Opin Psychiatry 25(5):381–390. doi:10.1097/
YCO.0b013e3283568537
Fijal BA, Kohler J, Ostbye K, Ahl J, Houston JP (2013) Association of candidate gene polymor-
phisms with diastolic blood pressure change in patients treated with duloxetine. Psychiatry Res
206(2–3):313–314. doi:10.1016/j.psychres.2012.10.019
Franciotta D, Kwan P, Perucca E (2009) Genetic basis for idiosyncratic reactions to antiepileptic
drugs. Curr Opin Neurol 22(2):144–149. doi:10.1097/WCO.0b013e328328f276
Garfield LD, Dixon D, Nowotny P, Lotrich FE, Pollock BG, Kristjansson SD, Dore PM, Lenze EJ
(2014) Common selective serotonin reuptake inhibitor side effects in older adults associated
with genetic polymorphisms in the serotonin transporter and receptors: data from a randomized
controlled trial. Am J Geriatr Psychiatry Off J Am Assoc Geriatr Psychiatry 22(10):971–979.
doi:10.1016/j.jagp.2013.07.003
Gebhardt S, Theisen FM, Haberhausen M, Heinzel-Gutenbrunner M, Wehmeier PM, Krieg JC,
Kuhnau W, Schmidtke J, Remschmidt H, Hebebrand J (2010) Body weight gain induced by
atypical antipsychotics: an extension of the monozygotic twin and sib pair study. J Clin Pharm
Ther 35(2):207–211. doi:10.1111/j.1365-2710.2009.01084.x
Goldstein JI, Jarskog LF, Hilliard C, Alfirevic A, Duncan L, Fourches D, Huang H, Lek M, Neale
BM, Ripke S, Shianna K, Szatkiewicz JP, Tropsha A, van den Oord EJ, Cascorbi I, Dettling M,
Gazit E, Goff DC, Holden AL, Kelly DL, Malhotra AK, Nielsen J, Pirmohamed M, Rujescu D,
Werge T, Levy DL, Josiassen RC, Kennedy JL, Lieberman JA, Daly MJ, Sullivan PF (2014)
Clozapine-induced agranulocytosis is associated with rare HLA-DQB1 and HLA-B alleles.
Nat Commun 5:4757. doi:10.1038/ncomms5757
Greenbaum L, Alkelai A, Zozulinsky P, Kohn Y, Lerer B (2012a) Support for association of HSPG2
with tardive dyskinesia in Caucasian populations. Pharmacogenomics J 12(6):513–520.
doi:10.1038/tpj.2011.32
Greenbaum L, Smith RC, Lorberboym M, Alkelai A, Zozulinsky P, Lifschytz T, Kohn Y, Djaldetti
R, Lerer B (2012b) Association of the ZFPM2 gene with antipsychotic-induced parkinsonism
in schizophrenia patients. Psychopharmacology (Berl) 220(3):519–528. doi:10.1007/
s00213-011-2499-6
Gunes A, Scordo MG, Jaanson P, Dahl ML (2007) Serotonin and dopamine receptor gene poly-
morphisms and the risk of extrapyramidal side effects in perphenazine-treated schizophrenic
patients. Psychopharmacology (Berl) 190(4):479–484. doi:10.1007/s00213-006-0622-x
Hedley PL, Jorgensen P, Schlamowitz S, Wangari R, Moolman-Smook J, Brink PA, Kanters JK,
Corfield VA, Christiansen M (2009) The genetic basis of long QT and short QT syndromes: a
mutation update. Hum Mutat 30(11):1486–1511. doi:10.1002/humu.21106
Heils A, Teufel A, Petri S, Stöber G, Riederer P, Bengel D, Lesch P (1996) Allelic variation of
human serotonin transporter gene expression. J Neurochem 66(6):2621–2624
Hougardy DM, Egberts TC, van der Graaf F, Brenninkmeijer VJ, Derijks LJ (2008) Serotonin
transporter polymorphism and bleeding time during SSRI therapy. Br J Clin Pharmacol
65(5):761–766. doi:10.1111/j.1365-2125.2008.03098.x
Hung SI, Chung WH, Jee SH, Chen WC, Chang YT, Lee WR, Hu SL, Wu MT, Chen GS, Wong
TW, Hsiao PF, Chen WH, Shih HY, Fang WH, Wei CY, Lou YH, Huang YL, Lin JJ, Chen YT
(2006) Genetic susceptibility to carbamazepine-induced cutaneous adverse drug reactions.
Pharmacogenet Genomics 16(4):297–306. doi:10.1097/01.fpc.0000199500.46842.4a
Ikediobi ON, Shin J, Nussbaum RL, Phillips KA, Translational UCf, Policy Research on
Personalized M, Walsh JM, Ladabaum U, Marshall D (2009) Addressing the challenges of the
clinical application of pharmacogenetic testing. Clin Pharmacol Ther 86(1):28–31. doi:10.1038/
clpt.2009.30
Inada T, Koga M, Ishiguro H, Horiuchi Y, Syu A, Yoshio T, Takahashi N, Ozaki N, Arinami T
(2008) Pathway-based association analysis of genome-wide screening data suggest that genes
associated with the gamma-aminobutyric acid receptor signaling pathway are involved in
neuroleptic-induced, treatment-resistant tardive dyskinesia. Pharmacogenet Genomics
18(4):317–323. doi:10.1097/FPC.0b013e3282f70492
Inoue K, Suzuki E, Takahashi T, Yamamoto Y, Yazawa R, Takahashi Y, Imai K, Miyakawa K, Inoue
Y, Tsuji D, Hayashi H, Itoh K (2014) 4217C>A polymorphism in carbamoyl-phosphate syn-
thase 1 gene may not associate with hyperammonemia development during valproic acid-based
therapy. Epilepsy Res 108(6):1046–1051. doi:10.1016/j.eplepsyres.2014.04.008
Janicki PK, Bezinover D, Postula M, Thompson RS, Acharya J, Acharya V, McNew C, Bowman
JD, Kurkowska-Jastrzebska I, Mirowska-Guzel D (2013) Increased occurrence of Valproic
acid-induced hyperammonemia in carriers of T1405N polymorphism in carbamoyl phosphate
synthetase 1 gene. ISRN Neurol 2013:261497. doi:10.1155/2013/261497
Kane JM, Woerner M, Lieberman J (1988) Tardive dyskinesia: prevalence, incidence, and risk fac-
tors. J Clin Psychopharmacol 8(4 Suppl):52S–56S
Kao AC, Muller DJ (2013) Genetics of antipsychotic-induced weight gain: update and current
perspectives. Pharmacogenomics 14(16):2067–2083. doi:10.2217/pgs.13.207
Karlsson L, Green H, Zackrisson AL, Bengtsson F, Jakobsen Falk I, Carlsson B, Ahlner J,
Kugelberg FC (2013) ABCB1 gene polymorphisms are associated with fatal intoxications
involving venlafaxine but not citalopram. Int J Leg Med 127(3):579–586. doi:10.1007/
s00414-013-0849-0
Keers R, Bonvicini C, Scassellati C, Uher R, Placentino A, Giovannini C, Rietschel M, Henigsberg
N, Kozel D, Mors O, Maier W, Hauser J, Souery D, Mendlewicz J, Schmal C, Zobel A, Larsen
ER, Szczepankiewicz A, Kovacic Z, Elkin A, Craig I, McGuffin P, Farmer AE, Aitchison KJ,
Gennarelli M (2011) Variation in GNB3 predicts response and adverse reactions to antidepres-
sants. J Psychopharmacol 25(7):867–874. doi:10.1177/0269881110376683
Klein KM, Hamer HM, Reis J, Schmidtke J, Oertel WH, Theisen FM, Hebebrand J, Rosenow F
(2005) Weight change in monozygotic twins treated with valproate. Obes Res 13(8):1330–
1334. doi:10.1038/oby.2005.161
Klotz U (2007) The role of pharmacogenetics in the metabolism of antiepileptic drugs: pharmaco-
kinetic and therapeutic implications. Clin Pharmacokinet 46(4):271–279.
doi:10.2165/00003088-200746040-00001
Korpos E, Wu C, Sorokin L (2009) Multiple roles of the extracellular matrix in inflammation. Curr
Pharm Des 15(12):1349–1357
Kwadijk-de Gijsel S, Bijl MJ, Visser LE, van Schaik RH, Hofman A, Vulto AG, van Gelder T, Ch
Stricker BH (2009) Variation in the CYP2D6 gene is associated with a lower serum sodium
concentration in patients on antidepressants. Br J Clin Pharmacol 68(2):221–225.
doi:10.1111/j.1365-2125.2009.03448.x
Laje G (2013) Pharmacogenetics of mood disorders: what clinicians need to know. CNS Spectr
18(5):272–284. doi:10.1017/S1092852913000278
Laje G, Paddock S, Manji H, Rush AJ, Wilson AF, Charney D, McMahon FJ (2007) Genetic mark-
ers of suicidal ideation emerging during citalopram treatment of major depression. Am J
Psychiatry 164(10):1530–1538. doi:10.1176/appi.ajp.2007.06122018
Laje G, Allen AS, Akula N, Manji H, John Rush A, McMahon FJ (2009) Genome-wide association
study of suicidal ideation emerging during citalopram treatment of depressed outpatients.
Pharmacogenet Genomics 19(9):666–674. doi:10.1097/FPC.0b013e32832e4bcd
Leclerc E, Mansur RB, Brietzke E (2013) Determinants of adherence to treatment in bipolar disor-
der: a comprehensive review. J Affect Disord 149(1–3):247–252. doi:10.1016/j.jad.2013.01.036
Lee HY, Kang RH, Paik JW, Jeong YJ, Chang HS, Han SW, Lee MS (2009) Association of the
adrenergic alpha 2a receptor – 1291C/G polymorphism with weight change and treatment
response to mirtazapine in patients with major depressive disorder. Brain Res 1262:1–6.
doi:10.1016/j.brainres.2009.01.013
Lencz T, Malhotra AK (2009) Pharmacogenetics of antipsychotic-induced side effects. Dialogues
Clin Neurosci 11(4):405–415
Leucht S, Hierl S, Kissling W, Dold M, Davis JM (2012) Putting the efficacy of psychiatric and
general medicine medication into perspective: review of meta-analyses. Br J Psychiatry J Ment
Sci 200(2):97–106. doi:10.1192/bjp.bp.111.096594
Li D, He L (2007) Meta-analysis supports association between serotonin transporter (5-HTT) and
suicidal behavior. Mol Psychiatry 12(1):47–54. doi:10.1038/sj.mp.4001890
Liolitsa D, Powell JF, Prince M, Lovestone S (2001) Association study of the 5-HT(2A) receptor
gene polymorphism, T102C and essential hypertension. J Hum Hypertens 15(5):335–339.
doi:10.1038/sj.jhh.1001177
Mahar I, Bambico FR, Mechawar N, Nobrega JN (2014) Stress, serotonin, and hippocampal neu-
rogenesis in relation to depression and antidepressant effects. Neurosci Biobehav Rev 38:173–
192. doi:10.1016/j.neubiorev.2013.11.009
Mandelli L, Serretti A (2013) Gene environment interaction studies in depression and suicidal
behavior: an update. Neurosci Biobehav Rev 37(10 Pt 1):2375–2397. doi:10.1016/j.
neubiorev.2013.07.011
Mann JJ (2003) Neurobiology of suicidal behaviour. Nat Rev Neurosci 4(10):819–828. doi:10.1038/
nrn1220
Meltzer HY (2012) Serotonergic mechanisms as targets for existing and novel antipsychotics.
Handb Exp Pharmacol 212:87–124. doi:10.1007/978-3-642-25761-2_4
Menke A, Domschke K, Czamara D, Klengel T, Hennings J, Lucae S, Baune BT, Arolt V, Muller-
Myhsok B, Holsboer F, Binder EB (2012) Genome-wide association study of antidepressant
treatment-emergent suicidal ideation. Neuropsychopharmacol Off Publ Am Coll
Neuropsychopharmacol 37(3):797–807. doi:10.1038/npp.2011.257
Millier A, Amri I, Boyer L, Auquier P, Toumi M (2014) Utility decrements associated with side
effects in schizophrenia. J Med Econ 17:853–861. doi:10.3111/13696998.2014.964405, 1–9
Muller DJ, Chowdhury NI, Zai CC (2013) The pharmacogenetics of antipsychotic-induced adverse
events. Curr Opin Psychiatry 26(2):144–150. doi:10.1097/YCO.0b013e32835dc9da
Novalbos J, Lopez-Rodriguez R, Roman M, Gallego-Sandin S, Ochoa D, Abad-Santos F (2010)
Effects of CYP2D6 genotype on the pharmacokinetics, pharmacodynamics, and safety of ris-
peridone in healthy volunteers. J Clin Psychopharmacol 30(5):504–511. doi:10.1097/
JCP.0b013e3181ee84c7
Ogusu N, Saruwatari J, Nakashima H, Noai M, Nishimura M, Deguchi M, Oniki K, Yasui-Furukori
N, Kaneko S, Ishitsu T, Nakagaswa K (2014) Impact of the superoxide dismutase 2 Val16Ala
polymorphism on the relationship between Valproic acid exposure and elevation of gamma-
glutamyltransferase in patients with epilepsy: a population pharmacokinetic-pharmacodynamic

analysis. PLoS One 9(11):e111066. doi:10.1371/journal.pone.0111066
Olivier B, Chan JS, Snoeren EM, Olivier JD, Veening JG, Vinkers CH, Waldinger MD, Oosting RS
(2011) Differences in sexual behaviour in male and female rodents: role of serotonin. Curr Top
Behav Neurosci 8:15–36. doi:10.1007/7854_2010_116
Osborn DP, Levy G, Nazareth I, Petersen I, Islam A, King MB (2007) Relative risk of cardiovas-
cular and cancer mortality in people with severe mental illness from the United Kingdom’s
general practice research database. Arch Gen Psychiatry 64(2):242–249. doi:10.1001/
archpsyc.64.2.242
Pacher P, Kecskemeti V (2004) Cardiovascular side effects of new antidepressants and antipsy-
chotics: new drugs, old concerns? Curr Pharm Des 10(20):2463–2475
Patsopoulos NA, Ntzani EE, Zintzaras E, Ioannidis JP (2005) CYP2D6 polymorphisms and the
risk of tardive dyskinesia in schizophrenia: a meta-analysis. Pharmacogenet Genomics
15(3):151–158
Paulussen AD, Gilissen RA, Armstrong M, Doevendans PA, Verhasselt P, Smeets HJ, Schulze-
Bahr E, Haverkamp W, Breithardt G, Cohen N, Aerssens J (2004) Genetic variations of
KCNQ1, KCNH2, SCN5A, KCNE1, and KCNE2 in drug-induced long QT syndrome patients.
J Mol Med 82(3):182–188. doi:10.1007/s00109-003-0522-z
Perlis RH, Purcell S, Fava M, Fagerness J, Rush AJ, Trivedi MH, Smoller JW (2007) Association
between treatment-emergent suicidal ideation with citalopram and polymorphisms near cyclic
adenosine monophosphate response element binding protein in the STAR*D study. Arch Gen
Psychiatry 64(6):689–697. doi:10.1001/archpsyc.64.6.689
Perlis RH, Laje G, Smoller JW, Fava M, Rush AJ, McMahon FJ (2009) Genetic and clinical predic-
tors of sexual dysfunction in citalopram-treated depressed patients. Neuropsychopharmacol
Off Publ Am Coll Neuropsychopharmacol 34(7):1819–1828. doi:10.1038/npp.2009.4
Perroud N, Courtet P, Vincze I, Jaussent I, Jollant F, Bellivier F, Leboyer M, Baud P, Buresi C,
Malafosse A (2008) Interaction between BDNF Val66Met and childhood trauma on adult’s
violent suicide attempt. Genes Brain Behav 7(3):314–322.
doi:10.1111/j.1601-183X.2007.00354.x
Perroud N, Aitchison KJ, Uher R, Smith R, Huezo-Diaz P, Marusic A, Maier W, Mors O, Placentino
A, Henigsberg N, Rietschel M, Hauser J, Souery D, Kapelski P, Bonvicini C, Zobel A,
Jorgensen L, Petrovic A, Kalember P, Schulze TG, Gupta B, Gray J, Lewis CM, Farmer AE,
McGuffin P, Craig I (2009) Genetic predictors of increase in suicidal ideation during antide-
pressant treatment in the GENDEP project. Neuropsychopharmacol Off Publ Am Coll
Neuropsychopharmacol 34(12):2517–2528. doi:10.1038/npp.2009.81
Perroud N, Uher R, Ng MY, Guipponi M, Hauser J, Henigsberg N, Maier W, Mors O, Gennarelli
M, Rietschel M, Souery D, Dernovsek MZ, Stamp AS, Lathrop M, Farmer A, Breen G,
Aitchison KJ, Lewis CM, Craig IW, McGuffin P (2012) Genome-wide association study of
increasing suicidal ideation during antidepressant treatment in the GENDEP project.
Pharmacogenomics J 12(1):68–77. doi:10.1038/tpj.2010.70
Peuskens J, Pani L, Detraux J, De Hert M (2014) The effects of novel and newly approved antipsy-
chotics on serum prolactin levels: a comprehensive review. CNS Drugs 28(5):421–453.
doi:10.1007/s40263-014-0157-3
Porcelli S, Fabbri C, Spina E, Serretti A, De Ronchi D (2011) Genetic polymorphisms of cyto-
chrome P450 enzymes and antidepressant metabolism. Expert Opin Drug Metab Toxicol
7(9):1101–1115. doi:10.1517/17425255.2011.597740
Pronicka E, Weglewska-Jurkiewicz A, Pronicki M, Sykut-Cegielska J, Kowalski P, Pajdowska M,
Jankowska I, Kotulska K, Kalicinski P, Jakobkiewicz-Banecka J, Wegrzyn G (2011) Drug-
resistant epilepsia and fulminant valproate liver toxicity. Alpers-Huttenlocher syndrome in two
children confirmed post mortem by identification of p.W748S mutation in POLG gene. Med
Sci Monit Int Med J Exp Clin Res 17(4):CR203–CR209
Rada JA, Cornuet PK, Hassell JR (1993) Regulation of corneal collagen fibrillogenesis in vitro by
corneal proteoglycan (lumican and decorin) core proteins. Exp Eye Res 56(6):635–648.
doi:10.1006/exer.1993.1081
Rau T, Wohlleben G, Wuttke H, Thuerauf N, Lunkenheimer J, Lanczik M, Eschenhagen T (2004)

CYP2D6 genotype: impact on adverse effects and nonresponse during treatment with antide-
pressants-a pilot study. Clin Pharmacol Ther 75(5):386–393. doi:10.1016/j.clpt.2003.12.015
Ray WA, Meredith S, Thapa PB, Meador KG, Hall K, Murray KT (2001) Antipsychotics and the
risk of sudden cardiac death. Arch Gen Psychiatry 58(12):1161–1167
Reynolds GP (2012) Pharmacogenetic aspects of antipsychotic drug-induced weight gain – a criti-
cal review. Clin Psychopharmacol Neurosci Off Sci J Korean Coll Neuropsychopharmacol
10(2):71–77. doi:10.9758/cpn.2012.10.2.71
Roberts R, Wells GA, Stewart AF, Dandona S, Chen L (2010) The genome-wide association
study – a new era for common polygenic disorders. J Cardiovasc Transl Res 3(3):173–182.
doi:10.1007/s12265-010-9178-6
Rogers HL, Bhattaram A, Zineh I, Gobburu J, Mathis M, Laughren TP, Pacanowski M (2012)
CYP2D6 genotype information to guide pimozide treatment in adult and pediatric patients:
basis for the U.S. Food and Drug Administration’s new dosing recommendations. J Clin
Psychiatry 73(9):1187–1190. doi:10.4088/JCP.11m07572
Saneto RP, Lee IC, Koenig MK, Bao X, Weng SW, Naviaux RK, Wong LJ (2010) POLG DNA
testing as an emerging standard of care before instituting valproic acid therapy for pediatric
seizure disorders. Seizure 19(3):140–146. doi:10.1016/j.seizure.2010.01.002
Schroeder C, Tank J, Boschmann M, Diedrich A, Sharma AM, Biaggioni I, Luft FC, Jordan J
(2002) Selective norepinephrine reuptake inhibition as a human model of orthostatic intoler-
ance. Circulation 105(3):347–353
Secher A, Bukh J, Bock C, Koefoed P, Rasmussen HB, Werge T, Kessing LV, Mellerup E (2009)
Antidepressive-drug-induced bodyweight gain is associated with polymorphisms in genes cod-
ing for COMT and TPH1. Int Clin Psychopharmacol 24(4):199–203. doi:10.1097/
YIC.0b013e32832d6be2
Segman RH, Heresco-Levy U, Finkel B, Goltser T, Shalem R, Schlafman M, Dorevitch A, Yakir
A, Greenberg D, Lerner A, Lerer B (2001) Association between the serotonin 2A receptor gene
and tardive dyskinesia in chronic schizophrenia. Mol Psychiatry 6(2):225–229. doi:10.1038/
sj.mp.4000842
Sequeira A, Mamdani F, Lalovic A, Anguelova M, Lesage A, Seguin M, Chawky N, Desautels A,
Turecki G (2004) Alpha 2A adrenergic receptor gene and suicide. Psychiatry Res 125(2):87–
93. doi:10.1016/j.psychres.2003.12.002
Serretti A, Olgiati P, Bajo E, Bigelli M, De Ronchi D (2011) A model to incorporate genetic testing
(5-HTTLPR) in pharmacological treatment of major depressive disorders. World J Biol
Psychiatry Off J World Fed Soc Biol Psychiatry 12(7):501–515. doi:10.3109/15622975.2011.
572998
Shams TA, Muller DJ (2014) Antipsychotic induced weight gain: genetics, epigenetics, and bio-
markers reviewed. Curr Psychiatry Rep 16(10):473. doi:10.1007/s11920-014-0473-9
Shams ME, Arneth B, Hiemke C, Dragicevic A, Muller MJ, Kaiser R, Lackner K, Hartter S (2006)
CYP2D6 polymorphism and clinical effect of the antidepressant venlafaxine. J Clin Pharm
Ther 31(5):493–502. doi:10.1111/j.1365-2710.2006.00763.x
Sicard MN, Zai CC, Tiwari AK, Souza RP, Meltzer HY, Lieberman JA, Kennedy JL, Muller DJ
(2010) Polymorphisms of the HTR2C gene and antipsychotic-induced weight gain: an update
and meta-analysis. Pharmacogenomics 11(11):1561–1571. doi:10.2217/pgs.10.123
Souza RP, De Luca V, Muscettola G, Rosa DV, de Bartolomeis A, Romano Silva M, Kennedy JL
(2008) Association of antipsychotic induced weight gain and body mass index with GNB3
gene: a meta-analysis. Prog Neuropsychopharmacol Biol Psychiatry 32(8):1848–1853.
doi:10.1016/j.pnpbp.2008.08.014
Stewart JD, Horvath R, Baruffini E, Ferrero I, Bulst S, Watkins PB, Fontana RJ, Day CP, Chinnery
PF (2010) Polymerase gamma gene POLG determines the risk of sodium valproate-induced
liver toxicity. Hepatology 52(5):1791–1796. doi:10.1002/hep.23891
Strohmaier J, Wust S, Uher R, Henigsberg N, Mors O, Hauser J, Souery D, Zobel A, Dernovsek
MZ, Streit F, Schmal C, Kozel D, Placentino A, Farmer A, McGuffin P, Aitchison KJ, Rietschel
M (2011) Sexual dysfunction during treatment with serotonergic and noradrenergic
antidepressants: clinical description and the role of the 5-HTTLPR. World J Biol Psychiatry
Off J World Fed Soc Biol Psychiatry 12(7):528–538. doi:10.3109/15622975.2011.559270
Sugai T, Suzuki Y, Sawamura K, Fukui N, Inoue Y, Someya T (2006) The effect of
5-hydroxytryptamine 3A and 3B receptor genes on nausea induced by paroxetine.
Pharmacogenomics J 6(5):351–356
Suzuki Y, Sawamura K, Someya T (2006) Polymorphisms in the 5-hydroxytryptamine 2A receptor
and CytochromeP4502D6 genes synergistically predict fluvoxamine-induced side effects in
japanese depressed patients. Neuropsychopharmacol Off Publ Am Coll Neuropsychopharmacol
31(4):825–831. doi:10.1038/sj.npp.1300919
Syu A, Ishiguro H, Inada T, Horiuchi Y, Tanaka S, Ishikawa M, Arai M, Itokawa M, Niizato K,
Iritani S, Ozaki N, Takahashi M, Kakita A, Takahashi H, Nawa H, Keino-Masu K, Arikawa-
Hirasawa E, Arinami T (2010) Association of the HSPG2 gene with neuroleptic-induced tar-
dive dyskinesia. Neuropsychopharmacol Off Publ Am Coll Neuropsychopharmacol
35(5):1155–1164. doi:10.1038/npp.2009.220
Tanaka M, Kobayashi D, Murakami Y, Ozaki N, Suzuki T, Iwata N, Haraguchi K, Ieiri I, Kinukawa
N, Hosoi M, Ohtani H, Sawada Y, Mine K (2008) Genetic polymorphisms in the
5-hydroxytryptamine type 3B receptor gene and paroxetine-induced nausea. Int J
Neuropsychopharmacol/Off Sci J Coll Int Neuropsychopharm 11(2):261–267
Thomas KL, Ellingrod VL, Bishop JR, Keiser MJ (2010) A pilot study of the pharmacodynamic
impact of SSRI drug selection and beta-1 receptor genotype (ADRB1) on cardiac vital signs in
depressed patients: a novel pharmacogenetic approach. Psychopharmacol Bull 43(1):11–22
Torrealba F, Riveros ME, Contreras M, Valdes JL (2012) Histamine and motivation. Front Syst
Neurosci 6:51. doi:10.3389/fnsys.2012.00051
Tsai HT, North KE, West SL, Poole C (2010) The DRD3 rs6280 polymorphism and prevalence of
tardive dyskinesia: a meta-analysis. Am J Med Genet B Neuropsychiatr Genet Off Publ Int Soc
Psychiatr Genet 153B(1):57–66. doi:10.1002/ajmg.b.30946
US Food and Drug Administration (2006) Clinical review: relationship between antidepressant
drugs and suicidality in adults. http://www.fda.gov/
Watanabe N, Omori IM, Nakagawa A, Cipriani A, Barbui C, Churchill R, Furukawa TA (2011)
Mirtazapine versus other antidepressive agents for depression. Cochrane Database Syst Rev
(12):CD006528. doi:10.1002/14651858.CD006528.pub2
WHO (2004) Mental Health. http://www.euro.who.int/en/health-topics/noncommunicable-
diseases/mental-health/data-and-statistics
Wilffert B, Al Hadithy AF, Sing VJ, Matroos G, Hoek HW, van Os J, Bruggeman R, Brouwers JR,
van Harten PN (2009) The role of dopamine D3, 5-HT2A and 5-HT2C receptor variants as
pharmacogenetic determinants in tardive dyskinesia in African-Caribbean patients under
chronic antipsychotic treatment: Curacao extrapyramidal syndromes study IX. J
Psychopharmacol 23(6):652–659. doi:10.1177/0269881108091594
Wilkinson TJ, Begg EJ, Winter AC, Sainsbury R (1999) Incidence and risk factors for hyponatrae-
mia following treatment with fluoxetine or paroxetine in elderly people. Br J Clin Pharmacol
47(2):211–217
Willard HF, Riordan JR (1985) Assignment of the gene for myelin proteolipid protein to the X
chromosome: implications for X-linked myelin disorders. Science 230(4728):940–942
Winner J, Allen JD, Altar CA, Spahic-Mihajlovic A (2013) Psychiatric pharmacogenomics pre-
dicts health resource utilization of outpatients with anxiety and depression. Transl Psychiatry
3:e242. doi:10.1038/tp.2013.2
Zaccara G, Franciotta D, Perucca E (2007) Idiosyncratic adverse reactions to antiepileptic drugs.
Epilepsia 48(7):1223–1244. doi:10.1111/j.1528-1167.2007.01041.x
Zill P, Malitas PN, Bondy B, Engel R, Boufidou F, Behrens S, Alevizos BE, Nikolaou CK,
Christodoulou GN (2003) Analysis of polymorphisms in the alpha-subunit of the olfactory
G-protein Golf in lithium-treated bipolar patients. Psychiatr Genet 13(2):65–69. doi:10.1097/01.
ypg.0000057881.80011.45
Part II
Pharmacovigilance of Specific
Psychotropic Drug Classes
Chapter 8
Safety and Tolerability of Antidepressants
Chris Manson, Robert Gordon, and David Baldwin
Abstract Problems abound when assessing the comparative acceptability and

safety of antidepressant drugs. The findings of double-blind randomised placebo-
and comparator-controlled trials provide a reasonable impression of the relative tol-
erability of antidepressants in short-term treatment, but less is confidently known
about their relative acceptability and safety in long-term treatment. The findings of
randomised controlled trials which are conducted in homogenous and largely physi-
cally healthy samples of trial participants do not necessarily generalise well into the
mixed and highly comorbid groups of depressed patients seen in real-world clinical
settings. Furthermore, many of the reported adverse effects of antidepressants
include reduced sexual desire, emotional indifference, and ‘activation’ – which can
make it hard to distinguish the adverse effects of treatment from persistent or emerg-
ing symptoms of the underlying condition. The quality of many reports of possible
adverse drug reactions with antidepressants is poor and this further hinders the accu-
rate assessment of comparative tolerability and potential toxicity. Certain antide-
pressant classes are prescribed preferentially to particular groups of patients with
comorbid physical illnesses, whose presence hinders the interpretation of tolerabil-
ity events in pharmacoepidemiological studies. These problems are addressed
through discussing the role of mixed treatment comparisons in assessing the accept-
ability of antidepressant treatment, are highlighted by referring to methodological
challenges in research into sexual dysfunction and emotional indifference during
treatment with selective serotonin reuptake inhibitors, are illustrated by considering
the quality of case reports of adverse reactions relating to hepatic function with
serotonin-noradrenaline reuptake inhibitors, and are further considered when review-
ing pharmacoepidemiological data on antidepressant tolerability in elderly patients.
Keywords Antidepressant • Tolerability • Safety • Pharmacovigilance • Elderly
C. Manson • R. Gordon • D. Baldwin (*)

University Department of Psychiatry, Clinical and Experimental
Southampton, UK
e-mail: c.c.manson@soton.ac.uk; rpg1822@gmail.com; D.S.Baldwin@soton.ac.uk

DOI 10.1007/978-3-319-24741-0_8
150 C. Manson et al.
8.1 Introduction
Pharmacotherapy, especially in the form of ‘second-generation antidepressants’

such as selective serotonin reuptake inhibitors (SSRIs), has become the mainstay
first-line choice of medical management for many mood and anxiety disorders. This
widespread use of antidepressants remains the source of some controversy. Double-
blind randomised placebo-controlled and comparator-controlled trials provide a
reasonable impression of their absolute and relative short-term tolerability, but less
is confidently known about the relative long-term acceptability and safety of differ-
ing antidepressants. Furthermore randomised controlled trials are typically con-
ducted in homogenous groups of generally physically healthy participants, who do
not fully represent the more heterogeneous and highly comorbid groups of depressed
patients in routine clinical practice.
Reported adverse effects of antidepressants often relate to largely subjective psy-
chological experiences such as reduced sexual desire, emotional indifference, and
‘activation’, which makes the task of distinguishing these experiences from persis-
tent or emerging symptoms of the underlying condition a difficult one. Another
hindrance in assessing the adverse effects of antidepressants lies in the poor quality
of many of the reports of their comparative tolerability and potential toxicity.
Current difficulties in assessing the safety of antidepressant drugs are considered
through a discussion of the role of mixed treatment comparisons in assessing accept-
ability of antidepressant medications, are highlighted through the example of method-
ological challenges in research into ‘treatment-emergent’ sexual dysfunction and
emotional indifference with SSRIs, and are illustrated through the examples of reported
associations of antidepressants with attempted suicide, hepatic dysfunction, persistent
pulmonary hypertension, and an increased risk of adverse outcomes in elderly patients.
8.2 The Range of Antidepressant Drugs
Antidepressants were first introduced in the 1950s and came to gradually be the first-
line medical treatment for patients with moderate to severe depression (Excellence
National Institute for Health and Clinical 2009). There are four main classes, tricy-
clic antidepressants (TCAs), monoamine oxidase inhibitors (MAOIs), selective sero-
tonin reuptake inhibitors (SSRIs), and serotonin-noradrenaline reuptake inhibitor
(SNRIs), and a group of other antidepressants with differing pharmacological prop-
erties, such as the noradrenaline and specific serotonergic antidepressant (NASSA)
mirtazapine, the selective noradrenaline reuptake inhibitor reboxetine, and the mela-
tonin agonist and serotonin receptor antagonist agomelatine. In addition to their
principal use in treating patients with unipolar depressive disorders, antidepressants
are also used to treat patients with anxiety disorders, obsessive-compulsive disorder,
post-traumatic stress disorder, and other conditions including chronic pain.
‘Second-generation’ antidepressants include SSRIs, SNRIs, and others that
selectively target neurotransmitters, as their primary mechanism of action. Some of
these are available as both immediate-release and extended-release formulations
8 Safety and Tolerability of Antidepressants 151
(Nussbaumer et al. 2014). The clinical rationale for extended-release formulations

is to improve patients adherence to prescribed drug regimens because of the need
for less frequent administration (once a day to once a week) (De Vane 2003); they
may also have better overall tolerability through reduced fluctuations of plasma
drug concentrations with lower plasma peaks (Nussbaumer et al. 2014). However,
critics argue that the availability of extended-release formulations represents little
more than a pharmaceutical industry marketing strategy, without improved benefits
for patients but with substantially higher costs (Huskamp 2006). A recent system-
atic review and meta-analysis suggested there are no clear differences between the
two approaches to drug formulation (Nussbaumer et al. 2014).
8.3 Relative Efficacy and Tolerability
The SSRI zimeldine was developed with the intention of it being superior in toler-
ability and having greater safety in overdose when compared to its TCA predeces-
sors. However, it was associated with a ‘hypersensitivity syndrome’, symptoms of
which included fever and myalgia (Nilsson 1983). Thirteen cases of drug reactions
resembling Guillain-Barre syndrome had been reported by 1985 and the drug was
withdrawn the same year (Fagius et al. 1985). Fortunately this has not been associ-
ated with other SSRIs. An early meta-analysis of randomised controlled trials with
SSRIs (with the exception of fluvoxamine) indicated that the proportion of patients
withdrawing from treatment due to adverse effects was significantly lower than the
proportion treated with a TCA (with the exception of dosulepin [dothiepin])
(Anderson 2000). The SSRI class is currently the most widely recommended first-
line pharmacological treatment for patients with major (unipolar) depression or
anxiety disorders, based on their efficacy, tolerability, safety, and acquisition cost.
Until recently, relatively little was known about whether certain SSRIs were
more efficacious or better tolerated than other newer antidepressants. Whether
newer antidepressants such as the SNRIs, reboxetine, and mirtazapine perform bet-
ter in terms of efficacy and side effect profile than existing SSRIs was also uncertain.
However meta-analyses, sometimes utilising ‘mixed treatment comparisons’ based
on network meta-analysis suggest that some newer drugs – for example, mirtazapine
and escitalopram – are more efficacious than their comparators, while others are
better tolerated – for example, escitalopram and sertraline (Cipriani et al. 2009; Nutt
2009). The SNRI reboxetine has disadvantages in both efficacy and tolerability,
leading some to argue for it not to be used as a first-line treatment for patients with
depression and some to suggest it should be withdrawn (Eyding et al. 2010).
Meta-analyses typically focus on overall efficacy (the proportion of patients
responding to treatment or who go into symptomatic remission) and overall toler-
ability (the proportion of patients withdrawing due to adverse effects). However,
some meta-analyses have examined the relative incidence of particular adverse
effects: for example, Gartlehner and colleagues found that the SNRI venlafaxine
is more strongly associated with the risks of nausea and vomiting, the SSRI ser-
traline with diarrhoea, mirtazapine with weight gain and trazodone with somno-
lence. They also found that there is a lower incidence of emergent sexual
dysfunction in patients treated with the predominantly dopaminergic

antidepressant drug bupropion, when compared to other second-generation anti-
depressants (Gartlehner et al. 2011). Most published meta-analyses have been
primarily concerned with the relative risks and benefits of SSRIs and other newer
antidepressants in the acute treatment of patients with major (unipolar) depres-
sion, but some are now emerging which focus on treatment of other conditions
such as generalised anxiety disorder (Baldwin et al. 2011).
8.4 Treatment Emergent Sexual Dysfunction
Depressed patients often experience lowered sexual interest and difficulty in achiev-
ing sexual satisfaction but many antidepressants are associated with ‘treatment-
emergent’ sexual dysfunction (i.e. worsening of pre-existing difficulties or
development of new problems during treatment). Sexual dysfunction is associated
with worsening of mood and diminished self-esteem and is detrimental to interper-
sonal relationships (Williams et al. 2006, 2010). Meta-analyses and randomised
controlled trials indicate that some antidepressants may be less likely to be associ-
ated with treatment-emergent sexual dysfunction than others. For example, Chiesa
and Serretti suggest that agomelatine, bupropion, moclobemide, mirtazapine, and
nefazodone were no different to placebo in associated treatment-emergent sexual
difficulties (Serretti and Chiesa 2009). Mirtazapine may be less likely to cause sex-
ual dysfunction (Serretti and Chiesa 2009) including delayed ejaculation (Taylor
et al. 2013). The latter is a less common side effect with bupropion than with SSRI
treatment, and bupropion is often used in the management of patients with SSRI-
associated sexual dysfunction. However the long-term clinical relevance of these
statistically significant differences seen between drugs in acute treatment studies is
uncertain (Reichenpfader et al. 2014). It is often the case that a reduction in the
severity of depressive symptoms is followed by an improvement in sexual function
and satisfaction, even though patients are taking antidepressant drugs that can affect
sexual function adversely (Baldwin et al. 2008, 2006).
8.5 Differentiating Depressive Symptoms from Adverse

Emotional Effects
Emotional indifference, increased ‘nervousness’, and worsening of suicidal

thoughts and images are all known to be potential adverse effects of antidepressant
treatment. However, these experiences can also be symptoms of the underlying
condition being treated, making it hard to distinguish adverse effects from worsen-
ing depression. Loss of interest and a reduced capacity for pleasure and reward
(anhedonia) are cardinal depressive symptoms, but antidepressants themselves can
sometimes reduce emotional responsiveness, increase apathy, and induce a state of
‘psychic indifference’. A qualitative study of individuals being treated with SSRIs
found that problems such as emotional detachment and reductions in both positive
and negative emotions were not uncommon (Price et al. 2009). Assessing indiffer-
ence is difficult due to the presence of depressive symptoms as well as a lack of
standardised methods. Previous scales and questionnaires have some weaknesses,
but the Oxford Questionnaire on Emotional Side Effects of Antidepressants appears
to have robust psychometric properties (Price et al. 2012) and in a randomised con-
trolled trial was able to differentiate between the emotional effects of the SSRI
escitalopram and the novel antidepressant agomelatine (Corruble et al. 2013).
Similar problems arise when assessing increased ‘nervousness’, worsened anxi-
ety, increased agitation, and suicidal ideation during antidepressant treatment. The
potential adverse effect of increased nervousness with SSRI treatment was recog-
nised a few years after they became available for clinical use (Baldwin et al. 1991).
Randomised controlled clinical trials suggest that SSRIs are often associated with a
transient worsening of anxiety symptoms (Baldwin et al. 1999). However, the ‘jit-
teriness syndrome’, thought to be associated with the early stages of antidepressant
treatment, is poorly characterised and its exact frequency (and optimal manage-
ment) is unknown (Sinclair et al. 2009). An early prescription-event monitoring
survey suggested an incidence of approximately 1 %, but this may represent signifi-
cant underreporting.
‘Discontinuation’ symptoms are not uncommon when antidepressant treatment
is withdrawn, either abruptly or when slowly tapered. Symptoms typically peak in
the first week and then lessen over time. The risk of discontinuation syndromes dif-
fers between antidepressants but predicting which patients will be affected is chal-
lenging. The recommended withdrawal method is slow and stepwise (tapered), but
its value is yet to be established through double-blind staggered discontinuation
design studies (Baldwin et al. 2007). As with emotional indifference and suicidal
thoughts and images, it is challenging to distinguish between pharmacological dis-
continuation symptoms and symptoms arising from an early relapse of anxiety and
depression. This is compounded by the fact that the most widely used instrument for
assessing withdrawal symptoms (the Discontinuation-Emergent Signs and
Symptoms Scale) (Rosenbaum et al. 1998) has 43 items, only 10 of which are
uncommon in untreated depressed patients.
8.6 Antidepressant Drugs and Suicidal Thoughts, Images,

and Behaviour
Meta-analyses suggest that the incidence of suicidality does not markedly differ
between classes of antidepressant: for example, a meta-analysis of 372 randomised
double-blind placebo-controlled trials, involving almost 100,000 patients found
‘little difference’ between drugs, although it found a greater risk of suicidality in
patients younger than 25 years (Stone et al. 2009). This is supported by the findings
of a systematic review of suicide attempts during observational studies of patients
being treated with SSRIs, which found them to be protective against both attempted
and completed suicide in older adults (Barbui et al. 2009).
Research findings suggesting a potential link between antidepressants and

suicide risk in young people have engendered much controversy. However, the evi-
dence on antidepressants and suicide risk is inconsistent (Kaizar et al. 2006;
Leckman and King 2007; Simon 2006) (See Table 8.1). During 2003–2004 the
Food and Drug Administration (FDA) in the United States issued health advisory
warnings stating that children and adolescents were at increased risk of suicidality
(ideation and behaviour) and soon after made it a requirement that a safety warning
be placed on the labels of all antidepressant drugs, this warning later being extended
to include young adults. The reasoning behind the FDA decisions is contentious
(Klein 2006): it appears based on an FDA-solicited meta-analysis, which showed a
relative risk for ‘suicidality’ of 1.95 (95 % confidence interval 1.28–2.98) for young
people taking antidepressants compared with those given placebo (Hammad et al.
2006). However, the trials included within the meta-analysis were not designed to
estimate suicidality risk (Klein 2006; Hammad et al. 2006) and the majority of the
reported adverse events involved suicidal thoughts and images and did not represent
attempted or completed suicide (Klein 2006; Hammad et al. 2006; Baldessarini
et al. 2006). Despite the evidence being inconclusive, the FDA advisories and boxed
warning attracted much media coverage, leading to safety alarms in clinicians and
young patients and their parents (Lu et al. 2014).
Treating young depressed people with antidepressant drugs can be effective in
terms of improved mood (March et al. 2004, 2007; Gibbons et al. 2012) but the
relationship between suicidal behaviour and antidepressant use is complex, and
studies have produced inconsistent results (Lu et al. 2014). Pre-existing suicidality
may be the reason for deciding to treat a depressed patient with an antidepressant,
and suicidal risk is usually reduced with effective treatment (Simon 2006). Managing
suicidality in this way may be associated with different risks in adults and in young
people (Gibbons et al. 2012). In adolescents and young adults, antidepressants may
lead to short-term increases in suicidality (Hammad et al. 2006; Jick et al. 2004;
FDA 2004). Studies undertaken since the FDA warnings have revealed substantial
reductions in the numbers of young people being treated with antidepressants
(Libby et al. 2007, 2009); this effect is also seen in older adults, who were not
included within the warnings (Valuck et al. 2007).
There were also an associated decline in diagnoses of depression in both children
and adults (Libby et al. 2007, 2009; Valuck et al. 2007), but no increase in the use
of alternative approaches in young people nor in the degree to which patients were
monitored, despite this being an explicit part of the warning (Libby et al. 2007,
2009; Pamer et al. 2010).
A large-scale long-term quasi-experimental study by Lu and colleagues (2014)
investigated whether there were changes in nationwide ‘suicidality’ following
the FDA warnings, the study cohorts comprising approximately 1.1 million
adolescents, 1.4 million young adults, and 5 million older adults. They found an
abrupt decline (31 %) in the previously upward trend of adolescent antidepressant
use during the second year after the warnings, but a simultaneous sharp increase in
psychotropic drug poisonings (21.7 %) (a validated measure of suicide attempts),
particularly amongst males; there was also a significant relative increase of 13.9 %
in drug poisonings amongst adolescents. In younger adults there was a reversal of
the upward trend in antidepressant use in the second year following the warnings
and a simultaneous increase in psychotropic drug poisonings (33.7 %): in older

adults, there was an observed relative reduction of 14.5 % in antidepressant use, but
no significant increase in psychotropic drug poisonings.
Contrary to the well-meaning intention of the regulatory warnings, the net effect
may have been to increase suicides, by leaving young people without effective anti-
depressant treatment. Similar findings have been observed in Sweden, where
Isacsson and colleagues assessed trends in antidepressant use by comparing pre-
scription records and post-mortem toxicology for all suicides in the 10–19 year age
group during the periods 1992–2003 (baseline) and 2004–2010 (after the warning).
They found that suicide increased over five consecutive years (by 60.5 %) and that
the increases occurred amongst individuals who were not undergoing antidepres-
sant treatment (Isacsson et al. 2014). The findings from both the American and
Swedish studies therefore reveal the potential for both intended and unintended
outcomes following widely publicised warnings.
8.7 Antidepressants and Hepatic Dysfunction
Certain antidepressants become associated with particular adverse effects,

sometimes without justification. An example is the TCA lofepramine, which became
linked to potential hepatotoxicity during the first few years of its launch; the first
reported case was in October 1987 (Macphee et al. 1987) and a further 57 reports
had been received by the UK Committee on Safety of Medicines 2 months later. A
case report featuring withdrawal of lofepramine and exposure to the metabolite
desipramine implicated the ‘parent drug’ (i.e. lofepramine) as being responsible for
hepatotoxicity (Lack et al. 1990). However, a subsequent case series involving
elderly patients revealed that the effects of lofepramine on hepatic function were
only transient (Kelly et al. 1993), and a further investigation demonstrated that there
was no difference between lofepramine and placebo in the incidence of abnormal
liver function tests (Tan et al. 1994). No further published reports about hepatic
dysfunction with the drug have appeared.
There is much variability in the quality of case reporting of adverse drug reac-
tions; a systematic review of all case reports published in a single psychopharma-
cology journal over 25 years found that only 7 out of 40 reports of presumed adverse
reactions were deemed sufficiently robust for ascribing possible or probable causal-
ity (with no correlation between the quality of the report and its impact as assessed
by citations) (Talat et al. 2013). Similarly, presumed hepatic reactions to SNRI
antidepressants (primarily duloxetine and venlafaxine) have been reported 17
times between 1999 and 2012; but only two reports involved re-exposure, only two
investigated possible ‘dose-response’ relationships, and only three confirmed
presumed hepatic reactions with a biopsy (Pradeep et al. 2004). High likely causal-
ity scores were only seen in reports of acute hepatitis in patients treated with venla-
faxine (Pradeep et al. 2004), of asymptomatic transaminitis in a patient with known
Gilbert’s syndrome treated with venlafaxine (Phillips et al. 2006), and of acute
hepatitis attributed to venlafaxine in a patient who was also taking five other medi-
cations, and who had recently stopped a phytomedicine (Feinberg 2010).
Findings from clinical trials and pharmacoepidemiological studies of duloxetine

suggest that the incidence of elevated liver function test results in patients without
hepatic disease is generally low (being somewhat higher in patients with known
liver pathology) and not significantly greater than when treated with placebo
(Wernicke et al. 2008a). In a large pharmacoepidemiological investigation based on
1.55 million years of patient exposure, there were 406 reported cases of duloxetine-
associated hepatic problems, equating to a cumulative reporting rate of 0.008 % for
all hepatic events and a risk of severe hepatic injury in 0.7/100,000 exposed person-
years (Wernicke et al. 2008b). However, independent analysis of the FDA adverse
event reporting system, and of the i3 Aperio (health insurance) database, suggests
that the reporting of hepatic events with duloxetine may be disproportionately
greater, when compared to other antidepressants (Strombom et al. 2008).
The novel antidepressant agomelatine has been shown to be hepato-protective
against paracetamol-induced liver damage in rats (Karakus et al. 2013). However in
human studies, elevated AST and/or ALT levels were seen in 1.4 % of patients tak-
ing 25 mg per day, and 2.5 % of patients prescribed 50 mg per day. Agomelatine is
therefore contraindicated in patients with pre-existing hepatic disease, and regular
monitoring (at baseline and at weeks 3, 12, and 24 of treatment) of hepatic function
is required (McAllister-Williams et al. 2010). One case of fulminant hepatic failure
in a patient with pre-existing fatty liver disease has been described (Gruz et al.
2014), and the German regulatory agency BfArm has received 58 cases of ‘hepato-
toxic adverse drug reactions’ (Gahr et al. 2013), emphasising the need for further
research into the effects of agomelatine on hepatic function.
8.8 SSRIs and Persistent Pulmonary Hypertension
Prenatal exposure to antidepressant drugs has been suggested to be associated with

a number of adverse effects in the newborn infant, and several variables need to be
considered when deciding whether or not to treat a depressed pregnant woman. One
potential adverse effect that should be considered is persistent pulmonary hyperten-
sion of the newborn (Kieler et al. 2012), (see Table 8.3 and Table 8.4) a relatively
rare condition (with an estimated prevalence of 1.9 per 1,000 live births) (Walsh-
Sukys et al. 2000), in which pulmonary vasculature in the infant fails to ‘relax’
resulting in poor oxygenation (Grigoriadis et al. 2014). Symptoms range in severity
from mild respiratory distress to hypoxia which requires urgent intensive medical
care (Jong et al. 2012). Study findings suggesting an association between maternal
use of SSRIs during pregnancy and persistent pulmonary hypertension in the new-
born led both Health Canada and the FDA to issue advice to clinicians alerting them
to the potential adverse effect (FDA US 2012). However more recent findings have
been inconsistent (Occhiogrosso et al. 2012), with some studies reporting no asso-
ciation (Andrade et al. 2009; Wilson et al. 2011), others some association (Kieler
et al. 2012; Chambers et al. 2006), and others indicating differential effects depend-
ing on the stage of the pregnancy when exposure occurred (Kieler et al. 2012;
Chambers et al. 2006) (See Table 8.2). The FDA has stated, ‘given the conflicting
Table 8.1 Hazard ratios for six adverse outcomes by antidepressant class and adjusted for
confounders (Coupland et al. 2011)
Antidepressant class adjusted hazard ratio (95 % CI)
Not taking
SSRIs TCAs Others antidepressants
Adverse All-cause 1.54 1.16 1.66 1.00
outcomes mortality (1.48–1.59) (1.10–1.22) (1.56–1.77)
Attempted 2.16 1.70 5.16 1.00
suicide/ (1.71–2.71) (1.28–2.25) (3.90–6.83)
self-harm
Myocardial 1.15 1.09 1.04 1.00
infarction (1.04–1.27) (0.96–1.23) (0.85–1.27)
Stroke/transient 1.17 1.02 1.37 1.00
ischaemic (1.10–1.26) (0.93–1.11) (1.22–1.55)
attack
Falls 1.66 1.30 1.39 1.00
(1.58–1.73) (1.23–1.38) (1.28–1.52)
Fracture 1.58 1.26 1.64 1.00
(1.48–1.68) (1.16–1.37) (1.46–1.84)
The * simply refers to the fact that all figures are adjusted vales for age, sex etc. Adjusted for sex,
age (5 year bands), year, severity of depression, depression before age 65, smoking status,
Townsend deprivation score, coronary heart disease, diabetes, hypertension, cancer, dementia,
Parkinson’s disease, hypothyroidism, obsessive-compulsive disorder, epilepsy/seizures, statins,
nonsteroidal anti-inflammatory drugs, antipsychotics, lithium, aspirin, antihypertensive drugs,
anticonvulsant drugs, and hypnotics/anxiolytics; all outcomes except stroke/transient ischaemic
attack also adjusted for stroke/transient ischaemic attack at baseline; fracture outcome also
adjusted for falls at baseline
Table 8.2 Hazard ratios for five adverse outcomes by antidepressant class and adjusted for
confounders (Coupland et al. 2011)
Antidepressant class adjusted hazard ratio (95 % CI)
Not taking
SSRIs TCAs Others antidepressants
Adverse Upper 1.22 1.29 1.37 1.00
outcomes gastrointestinal (1.07–1.40) (1.10–1.51) (1.08–1.74)
bleeding
Epilepsy/seizures 1.83 1.02 2.24 1.00
(1.49–2.26) (0.76–1.38) (1.60–3.15)
Road traffic 0.89 0.86 0.67 1.00
accidents (0.70–1.13) (0.64–1.15) (0.39–1.14)
Adverse drug 1.16 1.06 0.95 1.00
reactions (0.98–1.37) (0.86–1.29) (1.68–1.34)
Hyponatraemia 1.52 1.05 1.28 1.00
(1.33–1.75) (0.87–1.27) (0.98–1.67)
*Adjusted for sex, age (5 year bands), year, severity of depression, depression before age 65,
smoking status, Townsend deprivation score, coronary heart disease, stroke/transient ischaemic
attack, diabetes, hypertension, cancer, dementia, Parkinson’s disease, hypothyroidism, obsessive-
compulsive disorder, statins, nonsteroidal anti-inflammatory drugs, antipsychotics, lithium, aspi-
rin, antihypertensive drugs, anticonvulsant drugs, hypnotics/anxiolytics; all outcomes except
epilepsy/seizures also adjusted for epilepsy/seizures at baseline
results from different studies, it is premature to reach any conclusion about a pos-
sible link between SSRI use in pregnancy’ and persistent pulmonary hypertension
in the newborn (FDA 2012).
Risk factors for persistent pulmonary hypertension (Grigoriadis et al. 2014)
include certain congenital malformations, premature birth, meconium aspiration,
maternal obesity, and caesarean section delivery (Occhiogrosso et al. 2012; Koren
and Nordeng 2012a, b), but many of the studies of SSRIs did not exclude or control
for other known risk factors and more than one risk factor may be a prerequisite for
developing the condition (Occhiogrosso et al. 2012; Galbally et al. 2012). In a sys-
tematic review and meta-analysis of the associations between SSRI treatment and
persistent pulmonary hypertension, Grigoriadis and colleagues found an increased
risk if exposure to SSRIs occurred in late (but not early) pregnancy: study design,
congenital malformation, and meconium aspiration were not significant effect mod-
ifiers, but the possible moderating effects of caesarean section, preterm birth, and
maternal obesity could not be examined (Grigoriadis et al. 2014). There was a sta-
tistically significantly pooled odds ratio of 2.5 for exposure to SSRIs in late preg-
nancy, though the absolute risk is low; between 286 and 351 women in late gestation
would have to be treated with an SSRI for there to be one additional case of persis-
tent pulmonary hypertension in a new born baby. Therefore the meta-analysis indi-
cates that less than 1 infant in 100 will develop persistent pulmonary hypertension
following prenatal exposure to SSRIs. Although this is a serious condition with
death rates between 5 % and 10 % when associated with other conditions (such as
congenital malformations, meconium aspiration, sepsis, and idiopathic disease), it
can be managed successfully (Koran and Nodding 2012). The mortality of infants
with persistent pulmonary hypertension who have been exposed to SSRI is not
established, though one study suggests 9.1 % (3 out of 33 infants) of the infants who
were exposed to an SSRI died, whereas 9.5 % (183 of 1,935 infants) who were not
exposed to an SSRI died (Kieler et al. 2012); however the disparity in group size
makes this finding hard to interpret.
8.9 Adverse Outcomes in Elderly Patients
Depression is common in older people with around 10–15 % of those living in the
community being affected by depressive symptoms (McDougall et al. 2007;
Beekman et al. 1999). Adverse drug events are more common in elderly patients,
due to higher rates of comorbid illness, age-related physiological changes, and
polypharmacy (Cadieux 1999). Despite this they are underrepresented in ran-
domised controlled trials, as these typically exclude older people and those with
comorbid conditions (Giron et al. 2005). Several observational studies have investi-
gated adverse outcome associated with antidepressants (Reid and Barbui 2010),
though few have been dedicated specifically to an older population.
This paucity of data led Coupland and colleagues to undertake a large cohort
study of antidepressant use in older people (aged 65 years and older), involving a
Table 8.3 Exposure to SSRIs in gestational week 20 or later and risk of PPH of the newborn
(Kieler et al. 2012)
Number of infants with PPH of the Adjusted* odds ratio
newborn (per 1,000) (95 % CI)
Not exposed Exposed
Drug Any SSRI 1899 (1.2) 33 (3.0) 2.1 (1.5–3.0)
Fluoxetine 1952 (1.2) 9 (2.7) 2.0 (1.0–3.8)
Citalopram 1936 (1.2) 11 (3.3) 2.3 (1.2–4.1)
Paroxetine 1959 (1.2) 5 (3.9) 2.8 (1.2–6.7)
Sertraline 1949 (1.2) 10 (3.5) 2.3 (1.3–4.4)
Escitalopram 1966 (1.2) 1 (1.8) 1.3 (0.2–9.5)
Table 8.4 Exposure to SSRIs before gestational week 8 or later and risk of PPH of the newborn
(Kieler et al. 2012)
Number of infants with PPH of the Adjusteda odds ratio
newborn (per 1,000) (95 % CI)
Not exposed Exposed
Drug Any SSRI 1899 (1.2) 32 (1.9) 1.4 (1.0–2.0)
Fluoxetine 1952 (1.2) 7 (1.8) 1.3 (0.6–2.8)
Citalopram 1936 (1.2) 17 (2.5) 1.8 (1.1–3.0)
Paroxetine 1959 (1.2) 4 (1.7) 1.3 (0.5–3.5)
Sertraline 1949 (1.2) 9 (2.7) 1.9 (1.0–3.6)
Escitalopram 1966 (1.2) 1 (0.4) 0.3 (0.0–2.2)
a
Adjusted for maternal age, dispensed nonsteroidal anti-inflammatory drugs and antidiabetes
drugs, pre-eclampsia, chronic diseases during pregnancy, country of birth, birth year, level of deliv-
ery hospital, and birth order
total of 60,746 UK patients. The most commonly prescribed antidepressant drugs

were SSRIs (54.7 % of prescriptions), followed by TCAs (31.6 % of prescriptions),
other antidepressants (13.5 %), and MAOIs (0.2 %) (Coupland et al. 2011); the lat-
ter were excluded from analyses due to the small number of prescriptions. All
classes of antidepressant were associated with significantly increased risks of all-
cause mortality, nonfatal self-harm, falls, fractures, and upper gastrointestinal
bleeding, compared with patient groups not taking antidepressants: SSRIs and the
group of ‘other’ antidepressants were associated with greater risks of stroke and/or
transient ischaemic attack and of epilepsy/seizures, and SSRIs were also associated
with increased risk of myocardial infarction and hyponatraemia (Coupland et al.
2011) (see Tables 8.1 and 8.2).
Associations with seven adverse outcomes differed significantly between drug
classes. SSRIs were associated with significantly higher rates of all-cause mortality,
stroke/transient ischaemic attack, falls, fracture, epilepsy/seizures, and hyponatrae-
mia. No significant difference was found for nonfatal self-harm. The group of
‘other’ antidepressants had significantly higher rates for all-cause mortality, nonfatal
self-harm, stroke/transient ischaemic attack, fracture, and epilepsy/seizures when
compared with TCAs. No significant difference was found for falls or hyponatraemia.
TCAs prescriptions were not associated with significantly higher rates of adverse
outcomes than either SSRIs or the group of ‘other’ antidepressants.
When individual antidepressant drugs were examined, trazodone was found to
be associated with the highest adjusted hazard ratio for all-cause mortality and one
of the highest risks for nonfatal self-harm. Mirtazapine had the highest adjusted
hazard ratio for nonfatal self-harm and one of the highest risks for all-cause mortal-
ity and stroke/ischaemic attack. Venlafaxine had higher hazard ratios for stroke/
transient ischaemic attack, fracture, and epilepsy/seizures and is one of the highest
associations with all-cause mortality and nonfatal self-harm. The SSRI citalopram
had the highest association with falls, although risks were similar amongst all the
SSRIs (Coupland et al. 2011). Three SSRIs (citalopram, escitalopram, and fluox-
etine) were associated with significantly increased risks of hyponatraemia. The
TCAs amitriptyline and dosulepin had the lowest associated risk for many adverse
outcomes (Coupland et al. 2011). Significant trend was found in analyses of dosage
relationships with TCAs and SSRIs for all-cause mortality, falls, and epilepsy/sei-
zures. A significant dose relationship was also seen with TCAs and fracture.
Previous evidence has suggested that low-dose TCAs may be similar to higher
doses of TCAs in depressive symptom reduction (NICE 2009; Furukawa et al.
2003). Adverse outcomes were most likely during the first 28 days of use and again
in the first 28 days after stopping treatment (Coupland et al. 2011): adverse events
in the first 28 days of prescription could be due to depression being most severe, and
increased risk after stopping treatment could be due to the onset of supervening ill-
ness, or admission to hospital or residential homes, rather than treatment withdrawal
(Coupland et al. 2011).
Data such as these are hard to interpret. The effectiveness of the differing antide-
pressant classes in relieving depressive symptoms is not described. Given the asso-
ciation of depressive illness with chronic physical ill health, it should be expected
that populations prescribed with antidepressant drugs will have increased mortality.
The known association of TCAs with problems such as seizures and cardiac arrhyth-
mias may lead doctors to avoid prescribing them in patients with epilepsy and car-
diac disease, with a preference for SSRIs, which therefore become associated with
these conditions.
8.10 Conclusion
A detailed analysis of ten drug information resources for the SSRI fluoxetine found
a median of 74.5 reported possible adverse drug reactions (Tan et al. 2014). However,
a recent population survey found that 9 of the 20 most commonly everyday symp-
toms in the general population are also listed as ‘adverse drug reactions’ in more
than half of drug information documents (Petrie et al. 2014): so the misattribution
of everyday symptoms to possible adverse effects of prescribed medication must be
common in clinical settings (Tan et al. 2014). Frequently reported everyday symp-
toms include fatigue, sleeping problems, irritability, difficulty concentrating,
anxiety, depression, and agitation (Petrie et al. 2014), all common in depressed
patients and all of which are also listed as potential adverse drug reactions with
frequently prescribed medications (Tan et al. 2014). The routine provision of long
lists of possible side effects of medication may increase the expectancy of unwanted
treatment-emergent effects (Faasse and Petrie 2013), as informing individuals of
possible symptoms increases the likelihood of those symptoms being reported
(Myers et al. 1987); in randomised placebo-controlled trials with antidepressants,
participants who receive placebo often report adverse events which are remarkably
similar in nature to those seen with active drugs (Rief et al. 2009).
As such, careful communication of accurate information about potential adverse
effects of pharmacological treatment is needed in the management of patients with
chronic diseases, such as depression and anxiety disorders (Colloca and Finniss
2012). It makes sense to provide information on both the absolute and relative risks
of the most important potential side effects, with the greatest emphasis being given
to data from randomised controlled trials (Tan et al. 2014). However the quality of
reporting of data from randomised controlled trials is highly variable (Ioannidis
2009; Zorzela et al. 2014); and observational studies tend to underestimate the risk
of rare but serious adverse reactions (Papanikolau et al. 2006) and overestimate the
risk of common but less important reactions (de Lange et al. 2008). It has been
argued that simply listing the possible adverse effects of medication is probably
unhelpful: rather, efforts should be made to contextualise the potential risks and
benefits of intervention by considering the patient, the treatment being offered and
the underlying condition being treated (Tan et al. 2014). In the pharmacological
treatment of depressive illness, the need to provide understandable information
about the anticipated beneficial effects and possible adverse effects of antidepres-
sant medication has to be balanced with the imperative to encourage patients to
accept potentially life-saving treatment, often for many years.
References
Administration Food and Drugs (2004) Joint Meeting of the Psychopharmacologic Drugs Advisory
Committee and Paediatric Advisory Committee
Administration Food and Drug (2012) FDA drug and safety communication: selective serotonin
inhibitor (SSRI) antidepressant use during pregnancy and reports of a rare heart and lung con-
dition in newborn babies. www.fda.gov/Drugs/DrugSafety/ucm283375.htm
Administration US Food and Drug (2012) Public health advisory: treatment challenges of depres-
sion in pregnancy and the possibility of persistent pulmonary hypertension in newborns
Anderson IM (2000) Selective serotonin reuptake inhibitors versus tricyclic antidepressants: a
meta-analysis of efficacy and tolerability. J Affect Disord 58(1):19–36
Andrade SE, McPhillips H, Loren D, Raebel MA, Lane K, Livingston J et al (2009) Antidepressant
medication use and risk of persistent pulmonary hypertension of the newborn.
Baldessarini RJ, Pompili M, Tondo L (2006) Suicidal risk in antidepressant drug trials. Arch Gen
Baldwin D, Fineberg N, Montgomery S (1991) Fluoxetine, fluvoxamine and extrapyramidal tract

disorders. Int Clin Psychopharmacol 6(1):51–58
Baldwin D, Bobes J, Stein DJ, Scharwachter I, Faure M (1999) Paroxetine in social phobia/social
anxiety disorder. Randomised, double-blind, placebo-controlled study. Paroxetine Study
Group. Br J Psychiatry 175:120–126
Baldwin D, Bridgman K, Buis C (2006) Resolution of sexual dysfunction during double-blind
treatment of major depression with reboxetine or paroxetine. J Psychopharmacol 20(1):91–96
Baldwin DS, Montgomery SA, Nil R, Lader M (2007) Discontinuation symptoms in depression
and anxiety disorders. Int J Neuropsychopharmacol 10(1):73–84
Baldwin D, Moreno RA, Briley M (2008) Resolution of sexual dysfunction during acute treatment
of major depression with milnacipran. Hum Psychopharmacol 23(6):527–532
Baldwin D, Woods R, Lawson R, Taylor D (2011) Efficacy of drug treatments for generalised anxi-
ety disorder: systematic review and meta-analysis. BMJ 342:d1199
Barbui C, Esposito E, Cipriani A (2009) Selective serotonin reuptake inhibitors and risk of suicide:
a systematic review of observational studies. CMAJ: Can Med Assoc J Assoc Med Can
180(3):291–297
Beekman AT, Copeland JR, Prince MJ (1999) Review of community prevalence of depression in
later life. Br J Psychiatry J Ment Sci 174:307–311
Board Royal College of Psychiatrists’ Public Education Editorial (2012) In: Psychiatrists RCo (ed)
Antidepressants. Royal College of Psychiatrists, London
Cadieux RJ (1999) Antidepressant drug interactions in the elderly. Understanding the P-450 sys-
tem is half the battle in reducing risks. Postgrad Med 106(6):231–232, 37–40, 45–49
Chambers CD, Hernandez-Diaz S, Van Marter LJ, Werler MM, Louik C, Jones KL et al (2006)
Selective serotonin-reuptake inhibitors and risk of persistent pulmonary hypertension of the
newborn. N Engl J Med 354(6):579–587
Cipriani A, Furukawa TA, Salanti G, Geddes JR, Higgins JP, Churchill R et al (2009) Comparative
efficacy and acceptability of 12 new-generation antidepressants: a multiple-treatments meta-
analysis. Lancet 373(9665):746–758
Colloca L, Finniss D (2012) Nocebo effects, patient-clinician communication, and therapeutic
outcomes. JAMA 307(6):567–568
Corruble E, de Bodinat C, Belaidi C, Goodwin GM (2013) Agomelatine study group. Efficacy of
agomelatine and escitalopram on depression, subjective sleep and emotional experiences in
patients with major depressive disorder: a 24-wk randomized, controlled, double-blind trial. Int
J Neuropsychopharmacol Off Sci J Coll Int Neuropsychopharmacologicum 16(10):
2219–2234
Coupland C, Dhiman P, Morriss R, Arthur A, Barton G, Hippisley-Cox J (2011) Antidepressant use
and risk of adverse outcomes in older people: population based cohort study. BMJ 343:d4551
de Langen J, van Hunsel F, Passier A, de Jong-van den Berg L, van Grootheest K (2008) Adverse
drug reaction reporting by patients in the Netherlands-three years of experience. Drug Saf
31(6):515–524
DeVane CL (2003) Immediate-release versus controlled-release formulations: pharmacokinetics
of newer antidepressants in relation to nausea. J Clin Psychiatry 64(Suppl 18):14–19
Excellence National Institute for Health and Clinical (2009) Depression: the treatment and man-
agement of depression in adults (update) (National Clinical Practice Guideline 90): NICE
Eyding D, Lelgemann M, Grouven U, Harter M, Kromp M, Kaiser T et al (2010) Reboxetine for
acute treatment of major depression: systematic review and meta-analysis of published and
unpublished placebo and selective serotonin reuptake inhibitor controlled trials. BMJ
341:c4737
Faasse K, Petrie KJ (2013) The nocebo effect: patient expectations and medication side effects.
Postgrad Med J 89(1055):540–546
Fagius J, Osterman PO, Siden A, Wiholm BE (1985) Guillain-barre syndrome following zimeldine
treatment. J Neurol Neurosurg Psychiatry 48(1):65–69
Feinberg SS (2010) Correction of venlafaxine- and duloxetine-induced transaminase elevations
with desvenlafaxine in a patient with Gilbert’s syndrome. CNS Spectr 15(1):53–55
Furukawa T, McGuire H, Barbui C (2003) Low dosage tricyclic antidepressants for depression.
Cochrane Database Syst Rev (3):CD003197
Gahr M, Freudenmann RW, Connemann BJ, Hiemke C, Schonfeldt-Lecuona C (2013) Agomelatine
and hepatotoxicity: implications of cumulated data derived from spontaneous reports of
adverse drug reactions. Pharmacopsychiatry 46(6):214–220
Galbally M, Gentile S, Lewis AJ (2012) Further findings linking SSRIs during pregnancy and
persistent pulmonary hypertension of the newborn: clinical implications. CNS Drugs
26(10):813–822
Gartlehner G, Hansen RA, Morgan LC, Thaler K, Lux L, Van Noord M et al (2011) Comparative
benefits and harms of second-generation antidepressants for treating major depressive disorder:
an update meta-analysis. Ann Intern Med 155(11):772–785
Gibbons RD, Brown CH, Hur K, Davis J, Mann JJ (2012) Suicidal thoughts and behavior with
antidepressant treatment: reanalysis of the randomized placebo-controlled studies of fluoxetine
and venlafaxine. Arch Gen Psychiatry 69(6):580–587
Giron MS, Fastbom J, Winblad B (2005) Clinical trials of potential antidepressants: to what extent
are the elderly represented: a review. Int J Geriatr Psychiatry 20(3):201–217
Grigoriadis S, Vonderporten EH, Mamisashvili L, Tomlinson G, Dennis CL, Koren G et al (2014)
Prenatal exposure to antidepressants and persistent pulmonary hypertension of the newborn:
systematic review and meta-analysis. BMJ 348:f6932
Gruz F, Raffa S, Santucci C, Papale RM, Videla MG, Fernandez MG et al (2014) Agomelatine:
fulminant liver failure in a patient with fatty liver. Gastroenterol Hepatol 37(2):92–94
Hammad TA, Laughren T, Racoosin J (2006) Suicidality in pediatric patients treated with antide-
pressant drugs. Arch Gen Psychiatry 63(3):332–339
Huskamp HA (2006) Prices, profits, and innovation: examining criticisms of new psychotropic
drugs’ value. Health Aff 25(3):635–646
Ioannidis John PA (2009) Adverse events in randomized trials neglected, restricted, distorted, and
silenced. Arch Intern Med 169(19):1737–1739
Isacsson G, Ahlner J (2014) Antidepressants and the risk of suicide in young persons-prescription
trends and toxicological analyses. Acta Psychiatr Scand 129(4):296–302
Jick H, Kaye JA, Jick SS (2004) Antidepressants and the risk of suicidal behaviors. JAMA J Am
Med Assoc 292(3):338–343
Jong GW, Einarson T, Koren G, Einarson A (2012) Antidepressant use in pregnancy and persistent
pulmonary hypertension of the newborn (PPHN): a systematic review. Reprod Toxicol
34(3):293–297
Kaizar EE, Greenhouse JB, Seltman H, Kelleher K (2006) Do antidepressants cause suicidality in
children? A Bayesian meta-analysis. Clin Trials 3(2):73–90; discussion 91–98
Karakus E, Halici Z, Albayrak A, Polat B, Bayir Y, Kiki I et al (2013) Agomelatine: an antidepres-
sant with new potent hepatoprotective effects on paracetamol-induced liver damage in rats.
Hum Exp Toxicol 32(8):846–857
Kelly C, Roche S, Naguib M, Webb S, Roberts M, Pitt B (1993) A prospective evaluation of the
hepatotoxicity of lofepramine in the elderly. Int Clin Psychopharmacol 8(2):83–86
Kieler H, Artama M, Engeland A, Ericsson O, Furu K, Gissler M et al (2012) Selective
serotonin reuptake inhibitors during pregnancy and risk of persistent pulmonary hyperten-
sion in the newborn: population based cohort study from the five Nordic countries. BMJ
344:d8012
Klein DF (2006) The flawed basis for FDA post-marketing safety decisions: the example of anti-
depressants and children. Neuropsychopharmacol: Off Pub Am Coll Neuropsychopharmacol
31(4):689–699
Koren G, Nordeng H (2012a) Antidepressant use during pregnancy: the benefit-risk ratio. Am J
Obstet Gynecol 207(3):157–163
Koren G, Nordeng H (2012b) SSRIs and persistent pulmonary hypertension of the newborn. BMJ
344:d7642
Lack SJ, Baldwin DS, Montgomery SA (1990) Lofepramine, desipramine and abnormal tests of
liver function: a case report. Int Clin Psychopharmacol 5(3):185–190
Leckman JF, King RA (2007) A developmental perspective on the controversy surrounding the use
of SSRIs to treat pediatric depression. Am J Psychiatry 164(9):1304–1306
Libby AM, Brent DA, Morrato EH, Orton HD, Allen R, Valuck RJ (2007) Decline in treatment of
pediatric depression after FDA advisory on risk of suicidality with SSRIs. Am J Psychiatry
164(6):884–891
Libby AM, Orton HD, Valuck RJ (2009) Persisting decline in depression treatment after FDA
warnings. Arch Gen Psychiatry 66(6):633–639
Lu CY, Zhang F, Lakoma MD, Madden JM, Rusinak D, Penfold RB et al (2014) Changes in anti-
depressant use by young people and suicidal behavior after FDA warnings and media coverage:
quasi-experimental study. BMJ 348:g3596
Macphee GJA, Roberts MA, McKean CE (1987) Drug points: association of lofepramine with
abnormal liver function in elderly patients. BMJ 295(6604): 1001
March J, Silva S, Petrycki S, Curry J, Wells K, Fairbank J et al (2004) Fluoxetine, cognitive-
behavioral therapy, and their combination for adolescents with depression: Treatment for
Adolescents with Depression Study (TADS) randomized controlled trial. JAMA 292(7):
807–820
March JS, Silva S, Petrycki S, Curry J, Wells K, Fairbank J et al (2007) The Treatment for
Adolescents with Depression Study (TADS): long-term effectiveness and safety outcomes.
Arch Gen Psychiatry 64(10):1132–1143
McAllister-Williams RH, Baldwin DS, Haddad PM, Bazire S (2010) The use of antidepressants in
clinical practice: focus on agomelatine. Hum Psychopharmacol 25(2):95–102
McDougall FA, Matthews FE, Kvaal K, Dewey ME, Brayne C (2007) Prevalence and symptom-
atology of depression in older people living in institutions in England and Wales. Age Ageing
36(5):562–568
Myers MG, Cairns JA, Singer J (1987) The consent form as a possible cause of side-effects. Clin
Pharmacol Ther 42(3):250–253, <Go to ISI>://WOS:A1987K208900003
Nilsson BS (1983) Adverse reactions in connection with zimeldine treatment – a review. Acta
Psychiatr Scand Suppl 308:115–119
Nussbaumer B, Morgan LC, Reichenpfader U, Greenblatt A, Hansen RA, Van Noord M et al
(2014) Comparative efficacy and risk of harms of immediate- versus extended-release second-
generation antidepressants: a systematic review with network meta-analysis. CNS Drugs
28(8):699–712
Nutt DJ (2009) Prescribing antidepressants post Cipriani et al. J Psychopharmacol 23(8):865–866
Occhiogrosso M, Omran SS, Altemus M (2012) Persistent pulmonary hypertension of the new-
born and selective serotonin reuptake inhibitors: lessons from clinical and translational studies.
Am J Psychiatry 169(2):134–140
Pamer CA, Hammad TA, Wu YT, Kaplan S, Rochester G, Governale L et al (2010) Changes in US
antidepressant and antipsychotic prescription patterns during a period of FDA actions.
Papanikolaou PN, Christidi GD, Ioannidis JPA (2006) Comparison of evidence on harms of
medical interventions in randomized and nonrandomized studies. Can Med Assoc J 174(5):
635–641
Petrie KJ, Faasse K, Crichton F, Grey A. (2014) How common are symptoms? Evidence from a
New Zealand national telephone survey. BMJ Open. 4(6):e005374. doi:10.1136/
bmjopen-2014-005374.
Phillips BB, Digmann RR, Beck MG (2006) Hepatitis associated with low-dose venlafaxine for
postmenopausal vasomotor symptoms. Ann Pharmacother 40(2):323–327
Pradeep RJ, Victor G, Iby N, Kurpad SS, Galgali RB, Srinivasan K (2004) Venlafaxine induced
hepatitis. J Assoc Physicians India 52:340
Price J, Cole V, Goodwin GM (2009) Emotional side-effects of selective serotonin reuptake inhibi-
tors: qualitative study. Br J Psychiatry J Ment Sci 195(3):211–217
Price J, Cole V, Doll H, Goodwin GM (2012) The Oxford questionnaire on the emotional side-
effects of antidepressants (OQuESA): development, validity, reliability and sensitivity to
change. J Affect Disord 140(1):66–74
Reichenpfader U, Gartlehner G, Morgan LC, Greenblatt A, Nussbaumer B, Hansen RA et al
(2014) Sexual dysfunction associated with second-generation antidepressants in patients with
major depressive disorder: results from a systematic review with network meta- analysis. Drug
Saf: Int J Med Toxicol Drug Experience 37(1):19–31
Reid S, Barbui C (2010) Long term treatment of depression with selective serotonin reuptake
inhibitors and newer antidepressants. BMJ 340:c1468
Rief W, Nestoriuc Y, von Lilienfeld-Toal A, Dogan I, Schreiber F, Hofmann SG et al (2009)
Differences in adverse effect reporting in Placebo Groups in SSRI and tricyclic antidepressant
trials a systematic review and meta-analysis. Drug Saf 32(11):1041–1056
Rosenbaum JF, Fava M, Hoog SL, Ascroft RC, Krebs WB (1998) Selective serotonin reuptake
inhibitor discontinuation syndrome: a randomized clinical trial. Biol Psychiatry 44(2):
77–87
Serretti A, Chiesa A (2009) Treatment-emergent sexual dysfunction related to antidepressants: a
meta-analysis. J Clin Psychopharmacol 29(3):259–266
Simon GE (2006) The antidepressant quandary – considering suicide risk when treating adolescent
depression. N Engl J Med 355(26):2722–2723
Sinclair LI, Christmas DM, Hood SD, Potokar JP, Robertson A, Isaac A et al (2009) Antidepressant-
induced jitteriness/anxiety syndrome: systematic review. Br J Psychiatry J Ment Sci
194(6):483–490
Stone M, Laughren T, Jones ML, Levenson M, Holland PC, Hughes A et al (2009) Risk of
suicidality in clinical trials of antidepressants in adults: analysis of proprietary data submitted
to US Food and Drug Administration. BMJ 339:b2880
Strombom I, Wernicke JF, Seeger J, D'Souza DN, Acharya N (2008) Hepatic effects of duloxetine-
III: analysis of hepatic events using external data sources. Curr Drug Saf 3(2):154–162
Talat B, Mayers A, Baldwin DS (2013) Quality of case reports of adverse drug reactions with
psychotropic drugs: a 25-year review. Hum Psychopharmacol 28(5):413–420
Tan RS, Barlow RJ, Abel C, Reddy S, Palmer AJ, Fletcher AE et al (1994) The effect of low dose
lofepramine in depressed elderly patients in general medical wards. Br J Clin Pharmacol
37(4):321–324
Tan K, Petrie KJ, Faasse K, Bolland MJ, Grey A (2014) Unhelpful information about adverse drug
reactions. BMJ 349:g5019
Taylor MJ, Rudkin L, Bullemor-Day P, Lubin J, Chukwujekwu C, Hawton K (2013) Strategies for
managing sexual dysfunction induced by antidepressant medication. Cochrane Database Syst
Rev (5):CD003382
Valuck RJ, Libby AM, Orton HD, Morrato EH, Allen R, Baldessarini RJ (2007) Spillover effects
on treatment of adult depression in primary care after FDA advisory on risk of pediatric suicid-
ality with SSRIs. Am J Psychiatry 164(8):1198–1205
Waldinger MD, Zwinderman AH, Olivier B (2003) Antidepressants and ejaculation: a double-
blind, randomized, fixed-dose study with mirtazapine and paroxetine. J Clin Psychopharmacol
23:467–470
Walsh-Sukys MC, Tyson JE, Wright LL, Bauer CR, Korones SB, Stevenson DK et al (2000)
Persistent pulmonary hypertension of the newborn in the era before nitric oxide: practice varia-
tion and outcomes. Pediatrics 105(1 Pt 1):14–20
Wernicke J, Acharya N, Strombom I, Gahimer JL, D’Souza DN, DiPietro N et al (2008a) Hepatic
effects of duloxetine-II: spontaneous reports and epidemiology of hepatic events. Curr Drug
Saf 3(2):143–153
Wernicke J, Pangallo B, Wang F, Murray I, Henck JW, Knadler MP et al (2008b) Hepatic effects
of duloxetine-I: non-clinical and clinical trial data. Curr Drug Saf 3(2):132–142
Williams VS, Baldwin DS, Hogue SL, Fehnel SE, Hollis KA, Edin HM (2006) Estimating the
prevalence and impact of antidepressant-induced sexual dysfunction in 2 European countries:
a cross-sectional patient survey. J Clin Psychiatry 67(2):204–210
Williams VS, Edin HM, Hogue SL, Fehnel SE, Baldwin DS (2010) Prevalence and impact of
antidepressant-associated sexual dysfunction in three European countries: replication in a
cross-sectional patient survey. J Psychopharmacol 24(4):489–496
Wilson KL, Zelig CM, Harvey JP, Cunningham BS, Dolinsky BM, Napolitano PG (2011) Persistent
pulmonary hypertension of the newborn is associated with mode of delivery and not with
maternal use of selective serotonin reuptake inhibitors. Am J Perinatol 28(1):19–24
Zorzela L, Golder S, Liu Y, Pilkington K, Hartling L, Joffe A et al (2014) Quality of reporting in
systematic reviews of adverse events: systematic review. BMJ 2014:348
Chapter 9
Safety and Tolerability of Antipsychotics
Michael W. Jann and William K. Kennedy
Abstract Antipsychotics are commonly used to treat schizophrenia, bipolar disor-

ders, depression, and other conditions where psychotic symptoms occur.
Antipsychotics are classified as typical first-generation agents (FGAs) or atypical
second-generation agents (SGAs). Regulatory agencies have approved their usage
in the adult population 18–65 years. However, these agents are prescribed “off-
label” for children and adolescents. Antipsychotics have special warnings for use in
the elderly patients with dementia-related psychosis that includes an increased risk
for fatalities and cerebrovascular events. Extrapyramidal side effects (EPS) and tar-
dive dyskinesia are a risk with all antipsychotic classes. Even though antipsychotics
can lower seizure threshold, clozapine is the most commonly known for this adverse
effect. The metabolic syndrome associated with the SGAs is well known, and olan-
zapine most commonly produces weight gain and lipid changes. Thioridazine and
mesoridazine have a “black box” FDA warning due to QTc prolongation and risk
for torsades de pointes. Hematologic disorders such as agranulocytosis are associ-
ated with clozapine and specific guidelines for patient monitoring are established.
Both FGAs and SGAs are known to produce elevations from baseline prolactin
levels with risperidone and paliperidone. Aripiprazole was reported to have the least
effects on plasma prolactin levels. Neuroleptic malignant syndrome is an acute
medical emergency, and antipsychotic therapy must immediately cease and the
patient treated in the hospital.
Keywords Antipsychotics • Safety • Tolerability • Dementia • Elderly • Adverse

effects
M.W. Jann (*)

Department of Pharmacotherapy, University of North Texas System College of Pharmacy
(UNTSCP), Fort Worth, TX, USA
University of North Texas Health Science Center (UNTHSC), Fort Worth, TX, USA
e-mail: michael.jann@unthsc.edu
W.K. Kennedy
Department of Pharmacy Practice, Mercer University College of Pharmacy,
Atlanta, GA, USA
e-mail: kennedy_wk@mercer.edu

DOI 10.1007/978-3-319-24741-0_9
168 M.W. Jann and W.K. Kennedy
9.1 Introduction
Antipsychotics are the mainstay therapeutic treatment for a variety of psychiatric

conditions including schizophrenia, bipolar disorders, and others associated dis-
eases. These psychiatric disorders have acute episodes of psychotic symptoms
requiring immediate therapy. However, usage of antipsychotics often requires
chronic disease management as these conditions are long-term debilitating diseases.
With the exception of clozapine, all other antipsychotics were found to be equally
effective agents (Perel and Jann 2006). Adverse side effects of antipsychotics are
associated with their pharmacologic actions at various receptors located in the cen-
tral nervous system (CNS) and periphery displayed in Table 9.1 (Richelson 1999,
2010; Richelson and Sounder 2000). Pharmacovigilance detects, assesses, and
attempts to understand the adverse effects of medications so that preventative mea-
sures can be employed to maximize their outcomes and tolerability. The adverse
side effect profiles of 15 antipsychotic agents (both first generation and second gen-
eration) were compared using meta-analysis from over 200 clinical trials and pro-
vided an overall profile of each agent relative to the other antipsychotics with six
parameters (e.g., all-cause discontinuation) so that policy decision-makers can for-
mulate practice guidelines (Rajkumar and Melvin 2014). From the practical per-
spective, patients typically experience more than one adverse effect at any particular
time during treatment with different levels of severity (Perel and Jann 2006). These
types of studies focus on the adult population, and antipsychotics are used in all age
groups from the geriatric population to children and adolescents including during
pregnancy. It is beyond the scope of this chapter to fully investigate and present
studies that have assessed the adverse events associated with the antipsychotics as
several topics themselves can be an individual chapter such as the metabolic syn-
drome. References are included with a summary of their impact on health systems,
and this chapter will center on the various adverse side effects reported with anti-
psychotics that influence regulatory agencies and policy makers (Rajkumar and
Melvin 2014; Piparva et al. 2011).
9.2 Specific Populations
9.2.1 Elderly
Unless the patient has an existing diagnosis for schizophrenia or bipolar disorders,
antipsychotic use in the elderly is “off-label” by the regulatory agencies.
Antipsychotics are commonly used in the elderly and present a different set of issues
compared to the adult population as polypathology, polypharmacy, potential drug-
drug interactions, and age-related pharmacokinetic and pharmacodynamics are
important variables in selecting the drug dosages. Due to these multiple variables,
antipsychotics doses are generally lower in the elderly versus the adult group (Gareri
et al. 2003). The adverse event profile for the antipsychotic drugs in the elderly
9 Safety and Tolerability of Antipsychotics 169
Table 9.1 Summary of the antipsychotic receptor pharmacology and adverse side-effect profile
Drug D2 5HT2A M H1 Alpha-1 Adverse side-effect association
Reference compound
Spiperone 4 – – – – EPS, tardive dyskinesia, prolactin
Ketanserin – 4 – – – Rhabdomyolysis
3-quinuclidinyl-4- – – 4 – Anticholinergic, memory
iodobenzilate (QNB) impairment, ↑ narrow angle,
glaucoma, sinus tachycardia,
blurred vision
Pyrilamine – – – 4 – Sedation and weight gain
Prazosin – – – – 4 Cardiovascular, orthostatic
hypotension
First-generation typical antipsychotics
Chlorpromazine 3 2 2 3 3
Fluphenazine 4 2 1 2 2
Haloperidol 3 3 0 1 1
Loxapine 3 2 1 2 2
Mesoridazine 3 3 1 3 2
Perphenazine 4 3 0 3 2
Thioridazine 3 3 2 2 3
Second-generation typical antipsychotics
Aripiprazole 4 3 0 2 1
Asenapine 3 3 0 2 3
Clozapine 1 4 3 4 2
Iloperidone 3 4 0 2 4
Lurasidone 3 3 0 0 4
Olanzapine 2 3 2 4 1
Paliperidone 3 3 0 3 2
Quetiapine 1 2 0 2 2
Risperidone 3 4 0 3 3
Ziprasidone 3 4 0 2 1
0 = negligible, 1 = low, 2 = moderate, 3 = moderate high, 4 = high, reference compound; EPS
extrapyramidal side effects
D2 = dopamine receptor subtype, 5Ht2A = serotoninergic receptor subtype 2A, M = muscarinic
receptor, H1 = histamine receptor subtype 1, Alpha-1 = alpha adrenergic receptor subtype 1
differs than the adult population with specific patient warnings such as regarding
their use for dementia-related psychosis and potential for cerebrovascular disorders.
9.2.2 Children and Adolescents
Most antipsychotic drugs prescribed for children are also “off-label” that include
use in aggression and irritability in patients with autism, conduct disorders, and
pervasive developmental disorders with an increasing frequency (Schneider et al.
2014). Antipsychotic use in clinical practice especially in young children as early as
2 years of age presents on-going challenges to healthcare systems. The overall

adverse side-effect profile of antipsychotics in children and adolescents does not
substantially differ from the adult population, but due to the earlier exposure to
these drugs, the long-term risks and benefits must be taken into consideration by
prescribers. Database systems have been set up to monitor antipsychotic adverse
events in the children and adolescent population. Compared to other medications, it
was reported from the FDA Adverse Event Reporting System (FAERS) that the
atypical agents aripiprazole, risperidone, and quetiapine were among the top 20
medications with adverse events (Lee et al. 2014). Some countries have established
nationwide antipsychotic safety monitoring programs for children and adolescents
to provide system-wide safety information for clinicians (Harrison-Woolrych et al.
2007; Rani et al. 2009; Palanca-Maresca et al. 2014).
9.2.3 Pregnancy
Antipsychotic use in pregnant women continues, and information from the

Tennessee Medicaid program reported use of the typical first-generation agents
(FGAs) decreased while prescribing of the atypical second-generation agents
(SGAs) has increased (Epstein et al. 2013). For the FGAs, the incidence of use
decreased from 7.77 to 0.99 per 1,000 pregnancies between 1995 and 2005. In con-
trast, SGAs use increased from 1.73 to 16.52 per 1,000 pregnancies between 2000
and 2005. The specific warnings for atypical antipsychotic use during pregnancy
were possible risk of abnormal muscle movements and withdrawal symptoms
described for the mother and infant.
9.2.4 Dementia-Related Psychosis Fatality
Psychosis associated with Alzheimer’s disease (PAD) forms part of the behavioral
and psychological symptoms of dementia (BPSD) that include hallucinations, delu-
sions, agitation, paranoia, combativeness, or depression (Madhusoodanan et al.
2007). Delirium was reported to be the third most common cause of psychosis in the
elderly characterized by thought disturbance, poverty of thinking, irrationality, and
usually visual hallucinations (Jeste and Finkel 2000). Major depression was reported
to be the second most common diagnosis in the elderly accounting for most of the
psychosis in this population (Kyonmen and Whitfield 2009). Depression-related
psychosis is typically characterized by themes of somatic troubles, persecution,
guilt, and poor self-esteem. While regulatory agencies have yet to recognize BPSD
as a disease, the diagnostic criteria for the concept of PAD could be acknowledged
(Madhusoodanan et al. 2007; Jeste and Finkel 2000). Yet, these patients can present
the clinician with a complex, multifactorial, and fluctuating nature of psychotic
symptoms (Kyonmen and Whitfield 2009; Devanand 2013).
In April 2005, the FDA issued a “black box” warning for the SGAs and for the
FGAs in June 2008 indicating an increased risk for death in persons with dementia
(Kalapatapu and Schimming 2009). Data reported by the FDA noted a relative risk
(RR) ratio of 1.47 and 1.68 for the SGAs and FGAs, respectively (US Food and
Drug Administration 2014). A large meta-analysis study with 15 placebo-controlled
clinical trials 10–12 weeks of duration treated with the atypical antipsychotics in
patients with dementia reported an increased odds ratio (OR) of 1.54 [95 % C.I.
1.06–2.23, p = 0.02] for death that was similar to the FDA findings with a number
needed to harm (NNH) = 87 (Schneider et al. 2005). A later systematic review and
meta-analysis evaluated adverse events and divided the patients into two categories:
elderly patients with dementia and all other nonelderly adult patients (Simoni-
Wastila et al. 2009). This study examined the combined adverse events that included
cardiovascular symptoms, edema, and vasodilatation. Risperidone (2.10, 95 % C.I.
1.38–3.22) and olanzapine (2.30, 95 % C.I. 1.08–5.61) was found to have a higher
OR compared to placebo for these combined symptoms. Quetiapine and aripipra-
zole were found not associated with cardiovascular outcomes. However, when anti-
psychotics were examined in an outpatient population in patients with probable
Alzheimer’s disease for an extended time period (mean of 4.3 years), the presence
of antipsychotic drugs was found not to increase the risk of death using time-
dependent sensitive models (Lopez et al. 2013). It was reported that nursing home
admission and death were more frequent in patients treated with FGAs than SGAs.
But, both FGAs and SGAs were not associated with nursing home admission or
time to death when including covariate factors such as diabetes, hypertension, pre-
existing heart disease, and cognitive and demographic variables.
The efficacy of SGAs in patients with PAD (N = 423) was assessed with olanzap-
ine, quetiapine, and risperidone, and the main outcomes were discontinuation of
treatment for any reason and the number of patients with at least a minimal improve-
ment on the Clinical Global Impression of Change (CIBIC) at 12 weeks (Schneider
et al. 2006). Significant differences were not found between the SGAs, and improve-
ment based upon CIBIC scores was found with olanzapine 32 %, quetiapine 26 %,
and risperidone 29 % compared to placebo 21 % (p = 0.22). Patients treated with
SGAs had a higher discontinuation rate due to intolerability (p = 0.009). The study
reported that the overall adverse events offset the efficacy from SGAs. While the
risk of fatality was not an outcome, these results pose difficult questions for the use
of SGAs in elderly patients with dementia that display PAD or BPSD (Maher et al.
2011). The American College of Neuropsychopharmacology (ACNP) White Paper
noted that the two most common causes for deaths were due to cardiovascular dis-
ease or infections in this population (Jeste et al. 2008). Also noted that besides
psychosis and agitation commonly occurring in patients with dementia, these symp-
toms lead to significant caregiver distress and accelerated patient placement into the
nursing home environment. Finally, regulatory agencies have not yet “approved”
any medications for the treatment of patients with PAD or BSPD.
Clinicians face unusual challenges in risk management when prescribing anti-
psychotics in this patient population. Risk management procedures recommended
are to avoid unnecessary medications, balancing acute versus long-term risks,
informed consent from the patient and family members, and documentation in the
medical records (Jeste et al. 2008; Recupero and Rainey 2007). The ACNP White
Paper recommended these 11 steps: (1) determine the etiology of psychotic symp-
toms, (2) general treatment considerations, (3) shared decision-making, (4) identify
specific target symptoms for treatment, (5) pharmacotherapy selection, (6) dose, (7)
monitor for efficacy, (8) monitoring for safety, (9) educate patient and caregivers,
(10) know when to discontinue or switch pharmacotherapy, and (11) coordinate
care among the treatment team and family members (Jeste et al. 2008).
9.2.4.1 Cerebrovascular Warning
The FDA black box warning for SGAs in 2003 of increased fatality in patients with
dementia-related psychosis coincided with the cerebrovascular warning (US Food
and Drug Administration 2003). Cerebrovascular adverse events (CVAEs) were spe-
cifically noted for olanzapine, aripiprazole, and risperidone which included strokes,
transient ischemic events, and other undetermined events thought to be vascular in
origin. Whether or not the increased risk of fatality and CVAEs risk have similar
mechanisms or are interrelated remains unknown, but the overall death rate was not
overly represented by the CVAEs (Nelson 2005). The risk for CVAEs is higher
among patients with vascular dementia, vascular disease, or risk factors for stroke.
Added to the pathophysiology are potential drug-induced orthostatic hypotension
and oversedation as both events could result in falls and aspiration pneumonia. A
cohort study compared the OR of CVAEs between FGAs and SGAs in patients with
dementia aged 65 years or older (Laredo et al. 2011). The OR for FGAs was reported
to be 1.16 [95 % C.I. 1.07–1.27], whereas the OR for the atypical agents was 0.62
[95 % C.I. 0.53–0.72]. The results of this study indicate that FGAs have a slightly
higher risk for CVAEs but substantially and that SGAs appear to be safe. A large
systematic meta-analysis reported that the SGAs were associated with an increased
risk of stroke (pooled OR, 3.12 [95 % C.I. 1.32–8.21]) and determined the NNH =
53. The optimal approach to care is that each patient who needs an antipsychotic
drug must be carefully assessed with both short-term and long-term risks and ben-
efits. The ACNP White Paper included the CVAE assessment and noted significant
limitations of employing randomized clinical trial data and attempting to determine
long-term patient outcome with adverse events (Jeste et al. 2008). The identical 11
steps for patient assessment and treatment with SGAs in consideration of the CVAEs
possibility were recommended in the patients with dementia-related psychosis.
9.3 Specific Adverse Events
9.3.1 Extrapyramidal Side Effects (EPS) and Tardive

Dyskinesia (TD)
The advent of the SGAs was viewed as a major improvement in the pharmacother-
apy of psychotic disorders as a lower incidence of extrapyramidal side effects (EPS,
dystonia, pseudoparkinson’s, and akathisia) was found compared to the first-
generation agents (Divac et al. 2014; Peluso et al. 2012). Although not entirely
EPS-free, the use of adjunctive anticholinergic medications was 30-fold lower

among the patients treated with the atypical antipsychotics (Peluso et al. 2012).
CLZ had the reported lowest EPS odds ratio (OR) of 0.3 [95 % C.I. 0.12–0.62], and
lurasidone had the highest for the SGAs with an OR of 2.46 [95 % C.I. 155–3.72]
(Leucht et al. 2013). In comparison to the FGAs, the OR for chlorpromazine and
haloperidol was 2.65 [95 % C.I. 1.33–4.76] and 4.76 [95 % C.I. 3.70–6.04], respec-
tively (Leucht et al. 2013). A 17-year experience of drug-induced Parkinsonism
from a regional pharmacovigilance center reported that antipsychotics had the high-
est incidence compared to all other drugs reported to cause this adverse event (49 %
versus miscellaneous drugs 28.7 % and antidepressants 8.0 %) (Bondon-Guitton
et al. 2011). Although other medications can cause EPS, antipsychotics (both SGAs
and FGAs) are likeliest agents when compared to other medication classes.
A systematic review of long-term studies investigating the annual incidence of
tardive dyskinesia (TD) reported a rate of 3.9 % for the SGAs and 5.5 % for the
FGAs (Correll and Schenk 2008). Two risk factors were identified possibly associ-
ated with an increased risk of TD – mood disorders and age (Keck et al. 2000). The
observed rate of TD from a 15-year time period of atypical use on patients with
mood or anxiety disorder was 5.9 % from a retrospective chart review of 268 patients
(Coplan et al. 2013). The mean length of treatment for TD occurrence was
28.7 months (range 1–83 months) and average chlorpromazine equivalent dose was
350 mg (range 67–969 mg) and symptoms resolved in all but one patient when
antipsychotics were discontinued. Most patients (90.9 %) consented to a second
treatment with an atypical antipsychotic and did not reexperience TD. The risk of
TD with FGAs was found to be three to five times higher in patients >55 years than
younger patients (Woerner et al. 1998). Two large studies in elderly patients with
risperidone and olanzapine reported TD rates comparable to the adult population,
but either drug had an equal or lower incidence than FGAs (Woerner et al. 2011;
Kinon et al. 2014). Careful patient monitoring for EPS and early detection of TD in
all patients are recommended with FGAs and SGAs.
9.3.2 Seizures
Patients with schizophrenia may be more prone to seizures than the general popula-
tion (Hyde and Weinberger 1997). Psychotropic drugs and especially antipsychotics
and some selected antidepressants reduce seizure threshold with a reported range of
0.1–1.5 % in patients with psychiatric disorders versus the general population of
0.07–0.09 % with therapeutic doses (Devinsky et al. 1991). Dose-dependent and
rapid dose increase were factors reported with clozapine and chlorpromazine that
can lower seizure threshold. The overall incidence of seizures reported with CLZ
daily doses were <300 mg 1.0 %, 300–600 mg 2.7 %, and >600 mg 4.4 % (Devinsky
and Pacia 1994). Tonic-clonic seizures were most often noted, but myoclonic sei-
zures can also occur (Pacia and Devinsky 1994). The FDA analysis of seizure inci-
dence from the atypical antipsychotic clinical trial data reported a standard incidence
ratio (SIR) of 2.05 for all antipsychotics, whereas the SIR for CLZ was found to be
9.50 (Pisani et al. 2002). The Spanish Pharmacovigilance network reported a higher
reporting odds ratio (ROR) with the SGAs versus the FGAs of 3.2 [95 % C.I. 2.21–
4.63] with the highest incidence with CLZ (94/169 of the total number of convul-
sions with SGAs) (Alper et al. 2007). Whether or not atypical agents possess a
higher seizure induction compared to first-generation agents should be further eval-
uated (Lertxundi et al. 2013).
9.3.3 Metabolic Syndrome (MetS)
The MetS associated with SGAs is well known and associated with increased
weight gain, diabetes, and hyperlipidemia (Bak et al. 2014). The MetS definition
includes the presence of at least three of five parameters – blood pressure >130/85,
fasting blood glucose (FBS) >110 mg/dL, fasting triglycerides >150 mg/dL, HDL
<40 mg/dL (men) or <50 mg/dL (women), waist circumference >102 cm (men) or
>88 cm (women) (Ganguli and Strassnig 2011). The prevalence of MetS in the
general population was estimated to be about 23 % compared to a higher incidence
of 35–37 % in patients with psychiatric disorders treated with SGAs (Newcomer
and Hennekens 2007; Kagal et al. 2012). The FDA in 2003 issued a warning regard-
ing hyperglycemia and diabetes and recommended monitoring fasting blood glu-
cose in patients with diabetic risk factors, symptoms of hyperglycemia, or diabetes
(Mittal et al. 2013). The consensus panel a year later from the American Diabetic
Association (ADA), American Psychiatric Association (APA), and the Mount Sinai
Summit panel recommended monitoring weight, glucose, or glycosylated hemoglo-
bin and lipids every 12 weeks. Blood pressure was included every 12 weeks while
body weight monitoring every 4 weeks and waist circumference annually (ADA
et al. 2004).
Weight gain is among the strongest factors leading to the MetS (Papanastasiou
2013). Data from the antipsychotic registration clinical trials reported the risk for
clinically significant weight gain (≥7 % than baseline) was about ten times greater
for olanzapine compared to placebo. Ziprasidone, aripiprazole, and paliperidone
were only about twice the risk of placebo. Data reported on weight gain associated
with SGAs from highest to lowest were olanzapine, quetiapine, risperidone, zipra-
sidone, aripiprazole, and paliperidone (Leucht et al. 2013). From the comparative
study with the FGAs and SGAs, the latest SGA lurasidone had a low SMD (0.10,
95 % C.I. −0.02–0.21) for weight gain and was about equal to the SMD for ziprasi-
done (0.10, 95 % C.I. −0.02–0.22) and haloperidol (0.09, 95 % C.I. −0.00–0.17) in
producing weight gain. Asenapine SMD (0.23, 95 % C.I. 0.07–0.39) was slightly
greater than aripiprazole (0.17, 95 % C.I. 0.05–0.28) but less than paliperidone
(0.38, 95 % C.I. 0.27–0.48). The iloperidone SMD (0.62, 95 % C.I. 0.49–0.74) was
above chlorpromazine (0.55, 95 % C.I. 0.34–0.76) and slightly less than clozapine
(0.65, 95 % C.I. 0.31–0.99). Clozapine was slightly lower than olanzapine (0.74,
95 % C.I. 0.67–0.81) in weight gain, and both agents are well known to cause MetS
(Leucht et al. 2013). The underlying pathophysiology for antipsychotics to cause

weight gain is complex but thought to involve genetic variables that include the
5HT2C receptor system and leptin promoter gene variants MTHFR and MC4R
genes. Additional factors reported were HRH1, BDNF, NPY, CNR1, GHRL, FTO,
and the AMPK gene (Kao and Muller 2013). Further research remains underway to
determine biological mechanisms that influence antipsychotic-induced weight gain.
As the monitoring parameters for MetS in patients treated with SGAs were rec-
ommended, implementation remains challenging for healthcare systems. A large
VA retrospective cohort study (N = 12,009) evaluated practices for monitoring MetS
from antipsychotics between April 2008 and March 2009 (Kumra et al. 2008).
Weight, glucose or Hb1Ac, and LDL cholesterol were monitored within 30 days of
antipsychotic therapy initiation (baseline) and every 60–120 days afterward. This
study reported that weight was the most frequently monitored. Patients without a
schizophrenia diagnosis were less likely to be monitored. Antipsychotics with low-
est MetS risk were more likely to have weight monitored versus the higher-risk
agents at baseline. Whereas, patients taking antipsychotics with a moderate MetS
risk were more likely to have baseline glucose and LDL compared to low- and high-
risk antipsychotics. The study concluded that improvements are needed and should
be applied to all patients regardless of the antipsychotic agent being prescribed.
Another factor associated with antipsychotic-induced weight gain and MetS was
age (Papanastasiou 2013). This finding can have significant impact as the antipsy-
chotics are used in the elderly population. Antipsychotics can also cause weight
gain in children and adolescents (Kumra et al. 2008; Kranzler and Cohen 2013). No
appreciable differences were reported between men and women (Papanastasiou
2013). Waist size was found to be a useful factor in predicting MetS with a sensitiv-
ity of 79.4 % and specificity of 78.8 % with the highest rate found with CLZ
(Mitchell et al. 2013). Hospitalized patients with various psychiatric disorders were
evaluated for MetS in which weight gain and BMI were strongly associated
(Centorrino et al. 2012). Risk of MetS was also greater with polytherapy (defined as
>2 antipsychotics or one plus adjunctive mood stabilizer) (Centorrino et al. 2012).
Factors reported to be associated with MetS were sex, ethnicity, and martial and
living status where patients with schizophrenia can develop quickly even within
6 weeks of antipsychotic treatment (Papanastasiou 2013).
As the landscape of MetS associated with FGAs and SGAs became evident for
clinicians and monitoring parameters established, the issue of prediabetes emerged
given that patients with serious mental illnesses had increased incidences of MetS
compared to the general population. The ADA in January 2010 published the pre-
diabetes criteria that includes impaired FBG (>125 mg/dL) or impaired glucose tol-
erance (140–199 mg/dL 2 h after 75 g of glucose) or Hb1Ac in the range of
5.7 %–6.4 % (ADA 2010). Abdominal obesity, hypertension, dyslipidemia,
decreased HDL, and or increased triglycerides are other factors that are associated
with prediabetes. From 2003 to 2007, patients (N = 783) without a history of type 1
or type 2 diabetes admitted to a psychiatric facility treated with antipsychotics were
assessed for metabolic screening (Manu et al. 2012). Using the ADA 2010 criteria
for prediabetes, 413 (52.8 %) subjects had normal glucose, while 290 (37.0 %) were
identified as prediabetic. Newly diagnosed diabetes was found among 80 patients

(10.2 %). Prediabetes diagnosis was established by using one criteria in 209 patients
(72.1 %), 71 patients (24.5 %) had two criteria, and only 10 patients (3.4 %) fulfilled
all three criteria. An abnormal Hb1Ac identified 120 of the 290 (41.4 %) of the
prediabetic patients when used as the sole criteria. When FBS and/or Hb1Ac was
used with the glucose tolerance test, 89.7 % of prediabetic patients were identified.
FGA and SGA use was evaluated among all patients. CLZ and olanzapine had the
highest incidence of prediabetes and diabetes, while the other SGAs and FGAs had
about an equal incidence.
MetS associated with antipsychotic treatment is an important component when
clinicians prescribe these medications when treating patients with various psychiat-
ric conditions. The consequences of patients with prediabetes and then those who
develop diabetes present significant challenges in antipsychotic drug therapy and
lifestyle management in balancing the risks and benefits of short-term and long-
term use. MetS can lead to other significant diseases like cardiovascular, cerebro-
vascular, ophthalmologic, and renal impairment with an increased morbidity and
mortality resulting in significant healthcare expenditures (Riordan et al. 2011).
9.3.4 Cardiovascular
9.3.4.1 Torsades de Pointes (TdP)
A meta-analysis review of 15 antipsychotics revealed that the lowest OR occurred

with lurasidone −0.10 [95 % C.I. −0.21–0.01], a modest OR with ziprasidone of
0.41 [95 % C.I. 0.31–0.51], and the highest OR with sertindole 0.90 [95 % C.I.
0.76–1.02]. Other SGA agents had OR between lurasidone and ziprasidone.
Sertindole and amisulpride had the highest OR; however, these two agents are not
available in the USA (Leucht et al. 2013). From the historical perspective, in 1996,
sertindole was under development in the USA, but due to prolonged QTc interval
and association with 12 sudden unexplained deaths, the FDA did not grant approval.
TdP is a malignant ventricular arrhythmia associated with syncope and sudden
death. The primary form of TdP occurs from congenital courses either familial or
sporadic. The secondary TdP form is usually drug-induced (Beach et al. 2013).
From an electrical cardiovascular pathophysiological approach, the heart rate is
inversely related to the QT interval and can be viewed upon an ECG. The “c” in the
QTc accounts for a “correction” intended to remove the heart rate as a confounding
variable. The Bazett formula is the most method QTc = QT/square root of the R-R
interval (Beach et al. 2013). Clinicians should be concerned when the QTc interval
is between 450 and 500 ms, and >500 ms has increased risk of developing TdP.
The risk factors for TdP include women, elderly age, electrolyte imbalance, pre-
vious history of heart disease, metabolic disorders, stroke, and medications (Beach
et al. 2013; Vieweg 2003). As the elderly population is included, the link with deaths
associated with dementia-related psychosis and cerebrovascular with TdP is
plausible but not directly associated. Elderly patient assessment for antipsychotic
therapy recommendations include ECG, chest x-ray, stress test, and echocardio-
gram (Narang et al. 2010). The echocardiogram may be considered optional, but
periodic ECG monitoring is recommended on an individual patient basis.
The antipsychotics with the highest TdP risk are thioridazine and mesoridazine
which have FDA “black box” warnings. Other agents in the high-risk group are
droperidol, pimozide, and intravenous (IV) haloperidol (Meyer-Massetti et al.
2011). The remaining antipsychotics range from mild to moderate and little or no
risk. Ziprasidone’s drug development led to an established protocol for antipsy-
chotic investigations to assess QTc prolongation and possible TdP including the use
of drug-drug interaction studies with metabolic CYP inhibitors (Glassman and
Bigger 2001). It was found that ziprasidone produced a modest QTc prolongation of
6–10 ms, whereas sertindole and thioridazine prolonged QTc intervals by 20–22 ms
and 30 ms, respectively. Although a modest QTc prolongation occurs with ziprasi-
done, it has a low association with TdP. In contrast, both thioridazine and IV halo-
peridol are associated with significant QTc prolongation and possible TdP
(Glassman and Bigger 2001). All other antipsychotics have less than a moderate
association with QTc prolongation and TdP.
The results over a 7-year time period from 2004 to 2010 of reports from the FDA
FAERS identified 37 different antipsychotics with a reported odds ratio (ROR)
examining cardiovascular adverse events (Poluzzi et al. 2013). Cardiac arrhythmias
were detected in 4,794 cases, and three agents with the highest ROR are amisul-
pride, cyamemazine, and olanzapine. The olanzapine finding was surprising as
other pharmacovigilance studies were not able to separate this agent from other
SGAs in association with prolonged QTc intervals, TdP, or cardiac arrhythmias
(Manu et al. 2011; Ozeki et al. 2010). The risk of sudden cardiac death and antipsy-
chotics in a Medicaid and dual eligible Medicare-Medicaid population of 459,614
patients from five different states reported that haloperidol and chlorpromazine had
less favorable profiles. Among the SGAs, risperidone, quetiapine, and olanzapine
had lower risks of sudden death (Leonard et al. 2013).
9.3.4.2 Myocarditis
Myocarditis is an uncommon but potentially life-threatening condition. Clozapine

(CLZ) has been known to cause myocarditis with the total number of cases that
exceeds all other antipsychotics combined together (Coulter et al. 2001). The total
number of cases reported from this study for myocarditis from CLZ was 231 cases
versus 89 cases from all other antipsychotics. The incidence of CLZ-induced myo-
carditis ranges between 1 in 1,000 and 1 in 10,000 patients with 213 cases (includ-
ing 50 fatalities) that occurred in the first 2 months of therapy (Berardis et al. 2012).
CLZ dose was an independent factor. Recommendations for myocarditis detection
includes weekly ECGs, C-reactive protein (CRP), and troponin laboratory testing
matched with vital signs and clinical symptomatology for the first 4 weeks of CLZ
therapy (Berardis et al. 2012; Munshi et al. 2014). Afterward, vital signs and clinical
symptoms can be closely followed for the next few months. If the patient’s heart rate
becomes ≥120 bpm or increases by >30 bpm with symptoms of shortness of breath,
chest pain, cough, and myalgia, laboratory CRP and troponin should be obtained
and if needed a cardiology consult.
9.3.5 Hematologic
Hematological disorders have been associated with both the FGAs and SGAs that
include leukopenia, neutropenia, agranulocytosis, thrombocytopenia, and anemia
(Flanagan and Dunk 2008). Drug-induced neutropenia usually occurs after
1–2 weeks of treatment and agranulocytosis typically appears 3–4 weeks with the
exception of CLZ (Nooijen et al. 2011). Of all the antipsychotics, CLZ is the most
well recognized for hematological adverse events and extensively studied. The
CLZ-induced agranulocytosis is estimated to occur at 1:10,000 with the weekly
complete blood count (CBC) monitored for the first 18 weeks and then every
2 weeks afterward. In the USA, CBC occurs weekly for the first 6 weeks and then
every 2 weeks (Cohen et al. 2012). Regulatory agencies required CBC monitoring
programs when CLZ was prescribed. The mechanism for CLZ-induced agranulocy-
tosis remains unknown, but formation of a nitrenium cation metabolite from the
flavin-containing monoxygenase-3 (FMO3) is suggested to be the initial step for
hematological toxicity (Flanagan and Dunk 2008; Nooijen et al. 2011). A strong
genetic component is noted with the HLA-DQB1 haplotype that may allow for
identifying a patient subset at exceptionally high risk (5.1 % positive predictive
value). This finding also occurs with carbamazepine especially patients of Jewish
ethnicity in which biomarkers HLA-B38, DR4, and DQw3 provide a positive signal
for potential CLZ-induced agranulocytosis (Flanagan and Dunk 2008). Any CLZ-
treated patient that develops a fever or infection such as laryngitis should be imme-
diately evaluated for possible agranulocytosis and consider immediate medication
discontinuation.
The incidence of CLZ-induced agranulocytosis varied from 3.8 % to 8 % from
four studies that included 130,133 patients. The mortality rate was 0.1–0.3 % and
case-fatality rate was 2.2–4.2 % (Cohen et al. 2012). The Australian monitoring
program reported 209 cases of CLZ-induced white blood cell deficiency (WBCD)
between 2006 and 2010. The program reported with CLZ from 1993 to 2011; 141
cases of CLZ-induced agranulocytosis were recorded (Cohen and Monden 2013).
With US data, the incidence of CLZ-induced agranulocytosis was reported to dra-
matically decrease after 18 weeks from 3.39–6.93/1,000 patient-years to 0.37–
0.40/100 patient-years; agranulocytosis can still occur even years after CLZ
initiation with a recommendation for quarterly CBC monitoring beyond the initial
18 weeks (Drew 2013). Treatment for CLZ-induced agranulocytosis includes sup-
portive therapy, use of antibiotics, and GM-CSF or G-CSF (Flanagan and Dunk
2008). The use of antibiotics significantly lowered the mortality of almost 80 to
5–10 % in Western countries (Nooijen et al. 2011).
9.3.6 Prolactin-Related Adverse Events
Hyperprolactinemia (HPRL) is a well-known common adverse event from antipsy-

chotics since the 1970s. The standardized mean difference (SMD) on prolactin
effects for the various agents reported the lowest with aripiprazole 0.22 [95 % C.I.
−0.46–0.03] and the highest with paliperidone 1.30 [95 % C.I. 1.08–1.51].
Haloperidol and chlorpromazine were registered at 0.70 [95 % C.I. 0.56–0.85] (-)
and 0.16 [95 % C.I. −0.48–0.8], respectively (Leucht et al. 2013). A comprehensive
article on antipsychotic-induced HPRL was recently published (Peuskens et al.
2014). Antipsychotic-induced HPRL was associated with dopamine receptor sub-
type 2 (D2) antagonism at the anterior pituitary gland with changes in serum prolac-
tin concentrations the same between adults and children and adolescents. HPRL
increases are dose related but for RIS and paliperidone, and a small dose increase
may lead to a profound impact compared to other antipsychotics. As D2 receptor
antagonism is associated with HPRL, other endocrine effects are also included such
as galactorrhea, amenorrhea, and gynecomastia (Peuskens et al. 2014). Data mining
from the FDA Adverse Event Reporting System (AERS) for these endocrine prob-
lems associated with antipsychotics was conducted for the time period between
January 1968 and May 2005 (Szarfman et al. 2006). A total of 1,530 cases of endo-
crine problems were reported, and risperidone accounted for the vast majority of
these adverse events (1,247 cases, 81.5 %). The total number of pituitary tumors
was 77 case reports, and risperidone had the vast majority with 54 cases (70 %). As
a group, SGAs cause a lesser elevation in prolactin levels than FGAs except RIS and
paliperidone with a class warning for this condition included in their regulatory
labeling (Peuskens et al. 2014). HPRL may be associated with sexual dysfunction;
however, matching clinical symptoms and ruling out other medical conditions and
medications (besides the antipsychotics) need to be included in patient assessment.
9.3.7 Neuroleptic Malignant Syndrome (NMS)
NMS is a rare and idiosyncratic reaction reported with both FGAs and SGAs
reported to occur between 0.02 % and 0.25 % of patients (Trollor et al. 2012; Zarrouf
and Bhanot 2007; Guanci et al. 2012). Diagnosis of NMS consists of hyperthermia
(>100.4 °F or 38 °C on at least two occasions), muscle rigidity, and two of following
symptoms: diaphoresis, dysphagia, incontinence, altered consciousness, hyperten-
sion (≥25 % above baseline), mutism, labile blood pressure (≥20 mmHg diastolic;
≥ 25 mmHg systolic), creatine kinase (CK) increase (>4 times the upper normal
limit), tachypnea (≥50 % above baseline), and tremor or tachycardia (≥25 % above
baseline) (Guanci et al. 2012). NMS from antipsychotics was recognized since the
late 1970s (Gurrera et al. 2011). The Australian Adverse Drug Reaction Advisory
Committee (ADRAC) identified 208 NMS cases from both FGAs (N = 43) and
SGAs (N = 165) from April 1994 to September 2010 (Trollor et al. 2012). The
overall mortality rate was 5.8 % which is below the 10 % from the historical large-
case series (Caroff 1980). The SGA mortality rate was 3.0 % and lower than the
16.3 % with the FGAs (Trollor et al. 2012). CLZ-induced NMS had less muscle
rigidity than other antipsychotics and could be related to its pharmacologic profile
of weak D2 but potent 5-HT receptor binding affinity (Trollor et al. 2012). The
antipsychotic NMS adverse effect statement is found in the regulatory literature as
a class warning. Additional factors reported that can increase NMS risk were use of
polypharmacy (>2 antipsychotics) and rapid dose escalation (Su et al. 2014; Langan
et al. 2012). NMS symptom duration was reported to be about 7–10 days and longer
when depot antipsychotics are involved. The time period for NMS onset was found
to be 16 % in the first 24 h of antipsychotics therapy, 66 % within the first week, and
all cases by 30 days (Caroff and Mann 1988). NMS treatment involves the support-
ive therapy, use of benzodiazepines, dopamine agonists, and dantrolene (Caroff and
Mann 1993).
9.3.8 Rhabdomyolysis (Rhab) and Acute Kidney Injury (AKI)
Rhab ensues from damaged skeletal muscle fiber breakdown that results in the
release of toxic products from myocytes into the systemic circulation. The mecha-
nism of antipsychotic-induced Rhab remains unknown (Packhard et al. 2014).
Although Rhab is commonly associated with NMS (see Sect. 3.7), Rhab can occur
independently of NMS. Besides in the adult population, Rhab was reported to occur
in children and adolescents treated with antipsychotics, and only six cases of NMS
occurred among the 26 Rhab reports (Star et al. 2012). Significant elevations in
serum CK are often present in patients with Rhab with laboratory findings
>5,000 IU/L (median 9,600 IU/L), whereas in NMS, the serum CK was lower and
ranged from 500 to 3,000 IU/L (Melzter et al. 1996). Rhab and acute renal failure
were noted to occur in case reports since in the 1980s it was initially associated with
antipsychotic overdoses (Tam et al. 1980). However, later case reports included low
to modest antipsychotic doses (Stephanie and Trenkwalder 2010). Clinical symp-
toms that preceded recognition of Rhab were muscle and abdominal pain, general-
ized weakness, and dark urine (Packhard et al. 2014). The pathophysiology of
antipsychotic-induced Rhab may be associated with increased skeletal muscle
membrane permeability involving the 5HT2A receptor antagonism (Packhard et al.
2014). Blockade of the 5HT2A receptor may impact glucose uptake in the skeletal
muscle increasing CK permeability leading to the muscle breakdown. An alterna-
tive mechanism involves the D2 receptor system where excessive muscle stiffness
and rigidity also lead to muscle breakdown. At this time, Rhab has been reported
with each antipsychotic agent except the newer agents lurasidone, asenapine, and
iloperidone as only time on the market for these agents will determine whether or
not Rhab occurs. Two risk factors were identified as contributing to Rhab that
included polytherapy (>2 antipsychotics) and dose as most of the reported cases
took place with antipsychotic overdose situations (Packhard et al. 2014).
A large population-based, retrospective cohort study in the elderly (>65 years,

N = 97,777) who received a prescription for a SGAs was matched with patients who
did not receive this class of medication to examine the risk of AKI (Hwang et al.
2014). AKI is defined as a sudden loss of renal function and can be attributed to
hypotension, acute urinary retention, and NMS or Rhab. Patients with an SGA pre-
scription from June 2003 to December 2011 were included and followed for 90 days
after the antipsychotic start date. The primary outcome was hospitalization with
AKI. The secondary outcomes were known causes of AKI and all-cause mortality.
AKI was defined as a median increase in serum creatinine level of 1.11 mg/dL
(interquartile range of 0.49–2.26 mg/dL) at the time of hospitalization, by an abso-
lute increase of serum creatinine level of 0.31 mg/dL or >50 % increase from base-
line. Patients were matched with others who did not receive antipsychotic
medications, and analysis was conducted where patients were coded for AKI pres-
ence and where possible included the serum creatinine values. The relative risk
(RR) for hospitalization with AKI was 1.73 [95 % C.I. 1.55–1.92] for patients tak-
ing SGAs. In patients with information on serum creatinine levels, a higher risk for
AKI was found in patients with SGA use with an RR of 1.70 [95 % C.I. 1.22–2.38].
The all-cause mortality was higher in patients taking SGAs than nonrecipients
(6.8 % versus 3.1 %) which was similar to the randomized clinical trials reported by
the FDA in 2005 (See Sect. 2.4). It was recommended that proactive patient clinical
monitoring after SGA initiation with serum creatinine be included. If a patient com-
plains of urinary difficulty, a bladder scan to detect urinary retention must be con-
ducted. If AKI is suspected with SGAs, prompt drug discontinuation is suggested.
9.3.9 Gastrointestinal Hypomotility and Pancreatitis
FGAs and SGAs with highly potent anticholinergic pharmacologic properties are
thought to be the mostly likely agents to induce GI hypomotility (Richelson 1999,
2010; Richelson and Sounder 2000). A pharmacovigilance study collected data
from 1997 to 2006 and identified 27 GI hypomotility cases from FGAs and SGAs.
Intestinal colitis was reported in 57 cases that included a variety of FGAs and SGAs
with phenothiazines having the highest incidence of 33 cases. Other antipsychotics
reported were: haloperidol 9 cases and CLZ 7 cases, and all other SGAs reported a
total of 11 cases (Peyriere et al. 2009). Ischemic colitis occurred in 10 patients and
24 required surgery. Of the 27 cases, 14 fatalities were reported with 6 cases during
surgery and only 6 patients fully recovered. Other concomitant medications with
anticholinergic properties were found in 68.4 % of the patients. An analysis of 102
cases of CLZ-induced GI hypomotility (CIGH) was conducted with published
reports from 1950 to 2007 (Palerm et al. 2008). The mortality rate was 27.4 % and
mostly due to bowel resection surgery. The risk factors for GI hypomotility included
high CLZ dose or serum concentrations, concomitant anticholinergic use, and prior
history of GI disturbances. An additional pharmacologic mechanism for CIGH was
suggested to include the serotonergic (5HT) system with the 5HT2, 5HT3, and
5HT7 receptor subtypes that influence GI smooth muscle, colon transit, and visceral
sensation. As antipsychotic-induced GI hypomotility is a very rare but potentially
fatal adverse event, clinicians need to be especially aware of a patient reporting
constipation that continues unabated and when other drugs with anticholinergic
activity are included in the patient’s treatment.
Acute pancreatitis is another rare but potentially fatal adverse event due to anti-
psychotic and is listed in the USA Physician’s Desk Reference (PDR) for CLZ,
OLZ, and RIS (Hauben 2004; Koller et al. 2003; Kawabe and Ueno 2014). The
laboratory tests that assist in diagnosis are increased serum levels of amylase and
lipase with the clinical symptoms of GI pain, nausea, vomiting, and high fever. Most
cases occurred within 6 months of treatment initiation but rare cases have been
reported afterwards (Hauben 2004). The FDA Medwatch surveillance system had
reported 192 cases of antipsychotic-induced pancreatitis with 22 fatalities (Koller
et al. 2003). The antipsychotics reported and their occurrence were CLZ 40 %, OLZ
30 %, RIS 16 %, and haloperidol 12 %. Concomitant valproate use was found in
23 % of the cases with additional laboratory findings of hyperglycemia and meta-
bolic acidosis. A data mining study using Bayesian analysis failed to detect a signal
of disproportional of pancreatitis with these three atypical antipsychotics (Hauben
2004). Nevertheless, clinicians should be aware of this serious medical condition
when prescribing antipsychotics.
9.4 Patient Safety Monitoring
Patients with psychotic symptoms pose a tremendous challenge for clinicians to

provide short-term and long-term benefits while balancing risks and potential
adverse side effects. Selection of antipsychotic agent is based upon the patient’s
history, medical conditions, and use of other concurrent medications to treat the
medical conditions. The antipsychotic pharmacological profile shown in Table 9.1
provides a basic overall information regarding the extrapyramidal side effects, anti-
cholinergics, sedation, and orthostatic hypotensive properties in which to monitor
patients for these adverse effects. However, antipsychotics can have a wide spec-
trum of potential adverse events displayed in Table 9.2. Antipsychotic use in special
populations that include the elderly, children, and adolescents must be carefully
examined. In the elderly, the increased risk of fatalities and cerebrovascular events
must be balanced with efficacy and appropriate documentation of the antipsychotic
usage. The potential long-term consequences of antipsychotic treatment in children
and adolescents can lead to an increased continuum of care to minimize potential
adverse effects compared to the adult and elderly populations.
Clinical observations and early symptom detection of potential adverse effects
with antipsychotics are foundational methods that minimize patient risk and
increases safety. A comprehensive baseline laboratory assessment including ECGs
should be included when antipsychotics are initiated. If clinicians wish to add a
quantitative approach to patient monitoring for extrapyramidal side effects (e.g.,
pseudoparkinson’s, akathisia, and tardive dyskinesia), the use of standardized

patient rating scales can be periodically and systematically used (Simpson and
Angus 1970; Barnes 1989; Guy 1976). Although specific guidelines for routine
patient monitoring with rating scales have not been established for EPS, assess-
ments every 3–6 months can be employed. Unlike EPS, specific recommendations
and guidelines for monitoring patients regarding MetS have been established (See
Sect. 3.3). ECG monitoring to reduce the risk of TdP and identify prolonged QTc
intervals can be accomplished with consideration of incorporating patient’s past
medical history, age, and other medications known to increase QTc interval. If clo-
zapine is selected for a patient, specific hematologic monitoring recommendations
for <1 year are established. Long-term periodic CBC monitoring >1 year must con-
tinue with clozapine. Besides the CBC monitoring, clozapine-treated patients
should be assessed for the potential of myocarditis and gastrointestinal colitis. The
potential for NMS and pancreatitis can occur with any antipsychotic, and clinicians
need to be vigilant in detecting early clinical symptoms.
A special notation needs to be presented regarding long-acting depot antipsy-
chotic administration as a “black box” warning for postinjection delirium/sedation
syndrome (PDSS) that regulatory agencies have required patients treated with
Table 9.2 Summary of adverse events and the associated clinical assessment and/or laboratory
tests
Adverse event Clinical assessment and/or laboratory tests
Dementia-related psychosis fatality Clinical assessment of benefits versus risk
Cerebrovascular risk Clinical assessment of benefits versus risk
Extrapyramidal side effects Clinical observation
Pseudoparkinson’s Simpson-Angus Scale (SAS)
Akathisia Barnes Akathisia Rating Scale (BARS)
Tardive dyskinesia Abnormal Involuntary Movement Scale (AIMS)
Seizures Clinical assessment of benefits versus risk – clozapine
dose and rate of titration
Metabolic syndrome Body weight, FBS, blood pressure, lipid profile, and
waist circumference
Cardiovascular Vital signs and clinical symptoms
Torsade de pointes Electrocardiograph (ECG) monitoring; QTc <500 ms
Myocarditis Clozapine-treated patients – ECGs, CRP, and troponin
Hematologic Vital signs, fever, sore throat, and infection. Complete
blood counts with differential – especially with
clozapine-treated patients
Prolactin Clinical symptoms and prolactin serum concentrations
Neuroleptic malignant syndrome Hyperthermia (>38 °C), muscle rigidity, diaphoresis,
dysphagia, hypertension, serum creatine kinase (>4 times
upper limit), tachypnea, tremor, or tachycardia
Gastrointestinal colitis Constipation – especially with clozapine-treated patients
and use of other anticholinergic drugs
Pancreatitis Hyperglycemia, serum amylase, metabolic acidosis
FBS fasting blood glucose, CRP C-reactive protein
olanzapine pamoate be observed for 3 h after each injection administration (Alphs

et al. 2011). The PDSS symptoms include excessive sedation, altered conscious-
ness, slurred speech, and lethargy in which the patient may need to be taken to a
hospital’s emergency department for close monitoring (Alphs et al. 2011; Detke
et al. 2010). A 6-year multinational study (N = 931 patients) who received depot
olanzapine pamoate injections (N = 45,662) noted 36 (0.08 %) PDSS occurrences in
35 patients. One patient experienced two events and no fatalities were reported. All
patients recovered within 72 h postinjection. The PDSS events occurred at different
ages and BMIs without any significant identifying factors (McDonnell et al. 2014).
Thus, the recommendation for retaining the patient at the facility for 3 h remains
intact as PDSS can occur at any injection in any patient while being treated with
depot olanzapine pamoate injection. The PDSS has not been found with any other
long-acting depot antipsychotic products and is recommended to avoid accidental
injection into the blood vessel (Alphs et al. 2011; Novakovic et al. 2013).
9.5 Conclusions
FGAs and SGAs continue to be used in the management of patients with various
psychiatric conditions. The use of these agents in the elderly are “off-label” as regu-
latory agencies have not yet recognized psychosis associated with Alzheimer’s dis-
ease as a psychiatric or medical disorder. When antipsychotics are prescribed in
children and adolescents, their usage has expanded in addition to the various psychi-
atric illnesses to include other disorders such as autism. Clinicians must always
balance the benefits and risks when prescribing these agents especially in the elderly,
children, and adolescents. Appropriate individual patient safety monitoring for
potential adverse events should be implemented by clinicians taking into consider-
ation each antipsychotic pharmacologic and safety profile while matching the
patient characteristics. Pharmacovigilance surveillance studies and data can provide
important information on specific adverse event features, patterns of clinical symp-
toms, severity of the events, and where applicable, fatality rates of various antipsy-
chotic agents. Healthcare systems at the local, regional, or national levels may wish
to employ patient safety monitoring programs for antipsychotics based upon phar-
macovigilance studies.
References
Alper K, Schwartz KA, Kolts RL, Khan A (2007) Seizure incidence in psychopharmacological
clinical trials: an analysis of Food and Drug Administration (FDA) summary basis of approvals
reports. Biol Psychiatry 62:345–354
Alphs L, Gopal S, Karcher K et al (2011) Are the long-acting intramuscular formulations of ris-
peridone or paliperidone palmitate associated with post-injection delirium/sedation syndrome?
An assessment of safety databases. Curr Drug 6:43–45
American Diabetes Association (2010) Standards of medical care in diabetes – 2010. Diabetes
Care 33(Suppl 1):S11–S61
American Diabetes Association, American Psychiatric Association, American Association of
Clinical Endocrinologists et al (2004) Consensus development conference on antipsychotic
drugs and obesity and diabetes. Diabetes Care 27:596–601
Bak M, Fransen A, Janssen J, van Os J, Drukker M (2014) Almost all antipsychotics result in
weight gain: a meta-analysis. PLoS 9(4):e94112. doi:10.1371/journal.pone.0094112
Barnes TR (1989) A rating scale for drug-induced akathisia. Br J Psychiatry 154:672–676
Beach SR, Celano CM, Noseworthy PA, Januzzi JL, Huffman JC (2013) QTc prolongation,
Torsades de Pointes, and psychotropic medications. Psychosomatics 54:1–13
Berardis D, Serroni N, Campanella D et al (2012) Update on the adverse effects of clozapine: focus
on myocarditis. Curr Drug Saf 7:55–62
Bondon-Guitton E, Perez-Lloret SP, Bagheri H, Brefel C, Rascol O, Montastruc JL (2011) Drug-
induced Parkinsonism: a review of 17 years’ experience in a regional pharmacovigilance center
in France. Mov Disord 26:2226–2231
Caroff SN (1980) The neuroleptic malignant syndrome. J Clin Psychiatry 41:79–83
Caroff SN, Mann SC (1988) Neuroleptic malignant syndrome. Psychopharmacol Bull 24:25–29
Caroff SN, Mann SC (1993) Neuroleptic malignant syndrome. Med Clin N Am 77:185–202
Centorrino F, Masters GA, Talamo A, Baldessarini RJ, Ongur D (2012) Metabolic syndrome in
psychiatrically hospitalized patients treated with antipsychotics and other psychotropics. Hum
Psychopharmacol 27:521–526
Cohen D, Monden M (2013) White blood cell monitoring during long-term clozapine treatment.
Am J Psychiatry 170:366–369
Cohen D, Bogers JPA, van Dijk D, Bakker B, Schulte PFJ (2012) Beyond white blood cell moni-
toring: screening in the initial phase of clozapine therapy. J Clin Psychiatry 73:1307–1312
Coplan J, Gugger JJ, Tasleem H (2013) Tardive dyskinesia from atypical antipsychotic agents in
patients with mood disorders in a clinical setting. J Affect Disord 150:868–871
Correll CU, Schenk EM (2008) Tardive dyskinesia and new antipsychotics. Curr Opin Psychiatry
21:151–156
Coulter DM, Bate A, Meyboom RHB, Lindquist M, Edwards IR (2001) Antipsychotic drugs and heart
muscle disorder in international pharmacovigilance: data mining study. BMJ 322:1207–1209
Detke HC, McDonnell DP, Brunner E et al (2010) Post-injection delirium/sedation syndrome in
patients with schizophrenia treated with olanzapine long-acting injection: I. Analysis of cases.
BMC Psychiatry 10:4
Devanand DP (2013) Psychosis, agitation, and antipsychotic treatment in dementia. Am J
Devinsky O, Pacia SV (1994) Seizures during clozapine therapy. J Clin Psychiatry 55(Suppl
B):153–156
Devinsky O, Honigfeld G, Patin J (1991) Clozapine-related seizures. Neurology 41:369–371
Divac N, Prostran M, Jakovcevski I, Cerovac N (2014) Second-generation antipsychotics and
extrapyramidal adverse effects. BioMed Res Int; http://dx.doi.org/10.1155/2014/656370
Drew L (2013) Clozapine and agranulocytosis. Aust Psychiatry 21:335–337
Epstein RA, Bobo WV, Shelton RC et al (2013) Increasing use of atypical antipsychotics and
anticonvulsants during pregnancy. Pharmacoepidemiol Drug Saf 22:794–801
Flanagan RJ, Dunk L (2008) Haematological toxicity of drugs used in psychiatry. Hum
Psychopharmacol Clin Exp 23:27–41
Ganguli R, Strassnig M (2011) Prevention of metabolic syndrome in serious mental illness.
Psychiatr Clin N Am 34:109–125
Gareri P, De Fazio P, Stilo M, Ferrei G, De Sarro G (2003) Conventional and atypical antipsychot-
ics in the elderly. Clin Drug Invest 23:287–322
Glassman AH, Bigger JT (2001) Antipsychotic drugs: prolonged QTc interval, Torsade de Pointes,
and sudden death. Am J Psychiatry 158:1774–1782
Guanci N, Aggarwal R, Schleifer S (2012) Atypical neuroleptic malignant associated with iloperi-
done administration. Psychosomatics 53:603–604
Gurrera RJ, Caroff SN, Cohen A et al (2011) An international consensus study of neuroleptic
malignant syndrome diagnostic criteria suing the Delphi method. J Clin Psychiatry
72:1222–1228
Guy W (1976) AIMS. Abnormal Involuntary Movement Scale. Early Clinical Drug Evaluation
Unit (ECDEU). In: Rockville MD (ed) Assessment manual for psychopharmacology, vol 338.
NIMH, DHHS No. ADM 91–338, pp 534–537
Harrison-Woolrych M, Garica-Quiroga J, Ashton J, Herbison P (2007) Safety and usage of atypi-
cal antipsychotic medicines in children. Drug Saf 30:569–579
Hauben M (2004) Application of an empiric Bayesian data mining algorithm to reports of pancre-
atitis associated atypical antipsychotics. Pharmacotherapy 24:1122–1129
Hwang YJ, Dixon SN, Reiss JP et al (2014) Atypical antipsychotic drugs and the risk for acute
kidney injury and other adverse outcomes in older adults. Ann Intern Med 161:242–248
Hyde TM, Weinberger DR (1997) Seizures and schizophrenia. Schizophrenia Bull 23:611–622
Jeste D, Finkel S (2000) Psychosis of Alzheimer’s Disease and related: diagnostic for a distinct
syndrome. Am J Geriatr Psychiatry 8:29–34
Jeste DV, Balzer D, Casey D et al (2008) ACNP white paper: update on the use of antipsychotic
drugs in elderly persons with dementia. Neuropsychopharmacology 33:957–970
Kagal UA, Torgal SS, Patil NM, Malleshappa A (2012) Prevalence of the metabolic syndrome in
schizophrenic patients receiving second-generation antipsychotic agents – a cross sectional
study. J Pharm Pract 25:368–373
Kalapatapu RK, Schimming C (2009) Update on neuropsychiatric symptoms of dementia: antipsy-
chotic use. Geriatrics 64:10–18
Kao AC, Muller DJ (2013) Genetics of antipsychotic-induced weight gain: update and current
perspectives. Pharmacogenomics 14:2067–2083
Kawabe K, Ueno SC (2014) A case of acute pancreatitis associated with risperidone treatment.
Clin Psychopharmacol Neurosci 12:67–68
Keck PE, McElroy SL, Stakowski SM, Soutullo CA (2000) Antipsychotics in the treatment of
mood disorders and the risk of tardive dyskinesia. J Clin Psychiatry 61(Suppl 4):33–38
Kinon BJ, Kollack-Walker S, Jeste D et al (2015) Incidence of tardive dyskinesia in older adult
patients treated with olanzapine or conventional antipsychotics. J Geriatr Psychiatry Neurol
28(1):67–79
Koller EA, Cross JT, Doraiswamy PM, Malozowski SN (2003) Pancreatitis associated with atypi-
cal antipsychotics: from the Food and Drug Administration’s MedWatch surveillance system
and published reports. Pharmacotherapy 23:1123–1130
Kranzler HN, Cohen SD (2013) Psychopharmacologic treatment of psychosis in children and ado-
lescents: efficacy and management. Child Adolesc Psychiatry N Am 22:727–744
Kumra S, Oberstar JV, Sikich L et al (2008) Efficacy and tolerability of second-generation antipsy-
chotics in children and adolescents with schizophrenia. Schizopr Bull 34:60–71
Kyonmen HH, Whitfield TH (2009) Psychosis in the elderly. Am J Psychiatry 166:146–150
Langan J, Martin D, Shajahan P, Smith DJ (2012) Antipsychotic dose escalation as a trigger for
neuroleptic malignant syndrome (NMS): literature review and case series report. BMC
Psychiatry 12:214
Laredo L, Vargas E, Blasco AJ, Aguilar MD, Moreno A, Portoles A (2011) Risk of cerebrovascular
accident associated with use of antipsychotics: population-based case-control study. JAGS
59:1182–1187
Lee WJ, Lee TA, Pickhard AS, Caskey RN, Schumock GT (2014) Drugs associated with adverse
events in children and adolescents. Pharmacotherapy 34(9):918–926
Leonard CE, Freeman CP, Newcomb CW et al (2013) Antipsychotics and risk of sudden cardiac
death and all-cause death: cohort studies in Medicaid and dually-eligible Medicaid-Medicare
beneficiaries of five states. J Clin Exp Cardiol Suppl 10:1–9
Lertxundi U, Hernandez R, Medrano J, Domingo-Echaburu S, Garcia M, Aguirre C (2013)
Antipsychotics and seizures: higher risk with atypicals? Seizure 22:141–153
Leucht S, Cipriani A, Spineli L et al (2013) Comparative efficacy and tolerability of 15 antipsy-
chotic drugs in schizophrenia: a multiple-treatments meta-analysis. Lancet 382:951–962
Lopez OL, Becker JT, Chang YF et al (2013) The long-term effects of conventional and atypical
antipsychotics in patients with probable Alzheimer’s Disease. Am J Psychiatry
170:1051–1058
Madhusoodanan S, Shah P, Brenner R, Gupta S (2007) Pharmacological treatment of psychosis of
Alzheimer’s disease. CNS Drugs 21:101–115
Maher AR, Magione M, Bagley S et al (2011) Efficacy and comparative effectiveness of atypical
antipsychotic medications for off-label uses in adults. JAMA 306:1359–1369
Manu P, Kane JM, Correll CU (2011) Sudden deaths in psychiatric patients. J Clin Psychiatry
72:936–941
Manu P, Corrll CU, van Winkel R, Wampers M, De Hert M (2012) Prediabetes in patients treated
with antipsychotic drugs. J Clin Psychiatry 73:460–466
McDonnell DP, Landry J, Detke HC (2014) Long-term safety and efficacy of olanzapine long-
acting injection in patients with schizophrenia or schizoaffective disorder: a 6-year, multina-
tional, single-arm, open-label study. Int Clin Psychopharmacol 29(6):322–331
Melzter HY, Cola PA, Parsa M (1996) Marked elevations of serum creatine kinase activity associ-
ated with antipsychotic drug treatment. Neuropsychopharmacology 15:395–405
Meyer-Massetti C, Vaerini S, Ratz Bravo AE et al (2011) Comparative safety of antipsychotics in
the WHO pharmacovigilance database: the haloperidol case. Int J Clin Pharm 33:806–814
Mitchell AJ, Vancampfort D, Sweers K, van Winkel R, Yu W, De Hert M (2013) Prevalence of
metabolic syndrome and metabolic abnormalities in schizophrenia and related disorders – a
systematic review and meta-analysis. Schizophr Bull 39:306–318
Mittal D, Li C, Williams JS, Viverito K, Landes RD, Owen RR (2013) Monitoring veterans for
metabolic side effects when prescribing antipsychotics. Psychiatr Serv 64:28–35
Munshi TA, Volochniouk D, Hassan T, Mazhar N (2014) Clozapine-induced myocarditis: is man-
datory monitoring warranted for its early recognition? Case Rep Psychiatry 20, 513108
Narang P, El-Refai M, Parlapalli R et al (2010) Antipsychotic drugs: sudden cardiac death among
elderly patients. Psychiatry 7:25–29
Nelson JC (2005) Increased risk of cerebrovascular adverse events and death in elderly demented
patients treated with atypical antipsychotics: what’s a clinician to do? J Clin Psychiatry 66:1071
Newcomer JW, Hennekens CH (2007) Severe mental illness and risk of cardiovascular disease.
JAMA 298:1794–1796
Nooijen PMM, Carvalho F, Flanagan RJ (2011) Haematological toxicity of clozapine and some
other drugs used in psychiatry. Hum Psychopharmacol Clin Exp 26:112–119
Novakovic V, Adel T, Peselow E, Lindenmayer JP (2013) Long-acting injectable antipsychotics
and the development of post injection delirium/sedation syndrome (PDSS). Clin
Neuropharmacol 36:59–62
Ozeki Y, Fujii K, Kurimoto N et al (2010) QTc prolongation and antipsychotic medications in a
sample of 1017 patients with schizophrenia. Prog Neuropsychopharmacol Biol Psychiatry
34:401–405
Pacia SV, Devinsky O (1994) Clozapine-induced seizures: experience with 5,629 patients.
Neurology 44:2247–2249
Packhard K, Price P, Hanson A (2014) Antipsychotic use and risk of rhabdomyolysis. J Pharm
Pract 27(5):501–512
Palanca-Maresca I, Ruiz-Antorian B, Centeno-Soto G et al (2014) SENTA: a systematic online
monitoring registry for children and adolescents treated with antipsychotics. SpingerPlus 3:187
Palerm SE, McLean RM, Ellis PM, Harrison-Woolrych M (2008) Life-threatening clozapine-
induced gastrointestinal hypomotility: an analysis of 102 cases. J Clin Psychiatry 69:759–769
Papanastasiou E (2013) The prevalence and mechanism of metabolic syndrome in schizophrenia:
a review. Ther Adv Psychopharmacol 3:33–51
Peluso MJ, Lewis SW, Barnes TR, Jones PB (2012) Extrapyramidal motor side effects of first-
generation and second-generation antipsychotic drugs. Br J Psychiatry 200:387–392
Perel JM, Jann MW (2006) Antipsychtoics. In: Burton ME, Shaw LM, Schentag JJ, Evans WE
(eds) Applied pharmacokinetics and pharmacodynamics – principles of therapeutic drug moni-
toring, 4th ed. Lippincott Williams and Wilkins, Baltimore, pp 813–838
Peuskens J, Pani L, Detraux J, De Hert M (2014) The effects of novel and newly approved antipsy-
chotics on serum prolactin levels: a comprehensive review. CNS Drugs 28:421–453
Peyriere H, Roux C, Ferard C et al (2009) Antipsychotic-induced ischaemic colitis and gastroin-
testinal necrosis: a review of the French pharmacovigilance database. Pharmacoepidemiol
Drug Saf 18:948–955
Piparva KG, Buch JG, Chandrani KV (2011) Analysis of adverse drug reactions of atypical anti-
psychotic drugs in psychiatry OPD. Indian J Psychol Med 33:153–157
Pisani F, Oteri G, Costa C, Di Raimondo G, Di Perri R (2002) Effects of psychotropic drugs on
seizure threshold. Drug Saf 25:91–110
Poluzzi E, Raschi E, Kochi A et al (2013) Antipsychotics and torsadogenic risk: signals emerging
from the US FDA adverse events reporting system database. Drug Saf 36:467–479
Rajkumar RP, Melvin G (2014) Pharmacovigilance for psychiatrists: an introduction. Ind J
Rani FA, Byrne PJ, Murray ML, Carter P, Wong ICK (2009) Paediatric atypical antipsychotic
monitoring safety (PAMS) study. Drug Saf 32:325–333
Recupero PR, Rainey SE (2007) Managing risk when considering the use of atypical antipsychot-
ics for elderly patients with dementia-related psychosis. J Psychiatr Pract 13:143–152
Richelson E (1999) Receptor pharmacology of neuroleptics: relation to clinical effects. J Clin
Psychiatry 60(Suppl 10):5–14
Richelson E (2010) New antipsychotic drugs: how do their receptor-binding profiles compare? J
Clin Psychiatry 71:1243–1244
Richelson E, Sounder T (2000) Binding of antipsychotic drugs to human brain receptors – focus
on newer generation compounds. Life Sci 68:29–39
Riordan HJ, Antonini P, Murphy MF (2011) Atypical antipsychotics and metabolic syndrome in
patients with schizophrenia: risk factors, monitoring, and healthcare implications. Am Health
Drug Benefits 4:292–302
Schneider LSD, Dagerman KS, Insel P (2005) Risk of death with atypical antipsychotic drug treat-
ment for dementia. JAMA 294:1934–1943
Schneider LS, Tariot PN, Dagerman KS et al (2006) Effectiveness of atypical antipsychotic drugs
in patients with Alzheimer’s disease. NEJM 355:1525–1538
Schneider C, Taylor D, Zalsman G, Frangou S, Kyriakopoulos M (2014) Antipsychotics use in
children and adolescents: an on-going challenge in clinical practice. J Psychopharmacol
28:615–623
Simoni-Wastila L, Ryder PT, Qian J et al (2009) Association of antipsychotic use with hospital
events and mortality among Medicare beneficiaries residing in long-term care facilities. Am J
Geriatr Psychiatry 17:417–427
Simpson GM, Angus JW (1970) A rating scale for extrapyramidal side effects. Acta Psychiatr
Scand Suppl 212:11–19
Star K, Iessa N, Almandil NB et al (2012) Rhabdomyolysis reported for children and adolescents
treated with antipsychotic medicines: a case series analysis. J Child Adolesc Psychopharmacol
22:440–451
Stephanie C, Trenkwalder C (2010) Rhabdomyolysis after low dose quetiapine in a patient with
Parkinson’s disease with drug-induced psychosis. A case report. Mov Disorder 25:790–791
Su YP, Chang CK, Hayes RD et al (2014) Retrospective chart review on exposure to psychotropic
medications associated with neuroleptic malignant syndrome. Acta Psychiatr Scand
130:52–60
Szarfman A, Tonning JM, Levine J, Doraiswamy PM (2006) Atypical antipsychotics and pituitary
tumors: a pharmacovigilance study. Pharmacotherapy 26:748–758
Tam CW, Olin BR, Ruiz AE (1980) Loxapine-associated rhabdomyolysis and acute renal failure.
Arch Int Med 140:975–976
Trollor JN, Chen X, Chitty K, Sachdev PS (2012) Comparison of neuroleptic malignant syndrome
induced by first- and second-generation antipsychotics. Br J Psychiatry 201:52–56
US Food and Drug Administration (2003) Safety alert – risperdal (risperidone) [on line]. Available
at http://www.fda.gov/cder/drug/advisory/antipsychotics.htm. Accessed May 2014
US Food and Drug Administration (2014) FDA public health advisory: deaths with antipsychotics
in elderly patients with behavioral disturbances. Available at http://www.fda.gov/cder/drug/
advisory/antipsychotics.htm. Accessed May 2014
Vieweg WV (2003) New generation antipsychotic drugs and QTc interval prolongation. Primary
Care Companion J Clin Psychiatry 5:2–5, 215
Woerner MG, Alvir JMJ, Saltz BL, Lieberman JA, Kane JM (1998) Prospective study of tardive
dyskinesia in the elderly: rates and risk factors. Am J Psychiatry 155:1521–1528
Woerner MG, Correll CU, Alvir JMJ, Greenwald B, Delman H, Kane JM (2011) Incidence of
tardive dyskinesia with risperidone or olanzapine in the elderly: results from a 2-year, prospec-
tive study in antipsychotic-naïve patients. Neuropsychopharmacology 36:1738–1746
Zarrouf FA, Bhanot V (2007) Neuroleptic malignant syndrome: don’t let your guard down yet.
Curr Psychiatry 6:89–95
Chapter 10
Safety and Tolerability of Anxiolytics/
Sedative-Hypnotics
Vincenzo Arcoraci and Edoardo Spina
Abstract A variety of pharmacological agents are currently available for the

treatment of anxiety disorders and insomnia. The adverse event profile of selective
serotonin reuptake inhibitors (SSRIs) and serotonin and noradrenaline reuptake
inhibitors (SNRIs), first-line drugs for the treatment of many anxiety conditions,
is discussed in Chap. 8. Benzodiazepines are still valuable in the management of
anxiety disorders and transient insomnia. Tolerability and safety issued associated
with benzodiazepines includes sedation, memory and psychomotor impairment
and reduced driving performance. As stated in international guidelines, benzodi-
azepine treatment should be of short duration not exceeding 3 months, as continu-
ous use of benzodiazepines can lead to abuse and dependence. In addition, recent
findings suggest that long-term use of benzodiazepines in the elderly may increase
the risk of developing Alzheimer’s disease. Moreover, warnings released by
European Medicines Agency (EMA) and Food and Drug Administration (FDA)
advised that two benzodiazepines, tetrazepam and clobazam, may cause serious
skin events, including Stevens-Johnson syndrome and toxic epidermal necrolysis.
Emerging evidence indicates that pregabalin, a pharmacological agent approved
for the treatment of generalized anxiety disorder, also has the potential to lead to
abuse and dependence. Recent data in elderly patients has highlighted potential
safety concerns of non-benzodiazepine hypnotics, the so-called Z-drugs, includ-
ing zolpidem, specifically with regard to effects on balance and memory and on
fracture risk.
Keywords Anxiolytics • Sedative-hypnotics • Tolerability • Safety
V. Arcoraci • E. Spina (*)

e-mail: varcoraci@unime.it; espina@unime.it

DOI 10.1007/978-3-319-24741-0_10
192 V. Arcoraci and E. Spina
10.1 Introduction
A variety of agents and drug classes provide anxiolytic and sedative-hypnotic

effects. Benzodiazepines are still among the most widely prescribed agents for the
management of anxiety and insomnia. However, there is great concern regarding
long-term use of benzodiazepines due to their potential for dependence and abuse as
well as negative effects on memory and cognition (Lader 2014). Therefore, other
drugs, which generally have fewer side effects and lower addiction potentials, have
gradually replaced benzodiazepines. Selective serotonin reuptake inhibitors (SSRIs)
and serotonin and noradrenaline reuptake inhibitors (SNRIs) are licenced for the
treatment of many anxiety conditions such as generalised anxiety disorder, panic
disorder, social phobia and obsessive-compulsive disorder. According to recent
guidelines, these agents are first-line treatment for most types of anxiety disorders
(Bandelow et al. 2012; Baldwin et al. 2014). The tolerability and safety profile of
these agents are covered in Chap. 8. Buspirone, an azapirone acting on the serotonin
system, has been available for nearly 30 years to treat generalised anxiety disorders
and appears to be useful especially in patients who have not been previously on a
benzodiazepine (Chessick et al. 2006). The antiepileptic and antineuropathic pain
agent pregabalin has been also approved for the treatment of generalised anxiety
disorder in the EU since 2006 (Frampton 2014). The non-benzodiazepine drugs
zopiclone, eszopiclone (not licenced in the European Union), zolpidem and
zaleplon, commonly known as Z-drugs, have been developed as hypnotics with
improved pharmacokinetics and better tolerability profile in comparison to benzodi-
azepines, the traditional treatments for insomnia (Gunja 2013).
In this chapter, we will review the various adverse effects associated with the use
of benzodiazepines, buspirone, pregabalin and non-benzodiazepine hypnotics.
10.2 Benzodiazepines
Benzodiazepines have been widely used in clinical practice for over five decades
and continue to be among the most commonly prescribed agents to treat anxiety and
insomnia (Greenblatt 2011). All benzodiazepines have anxiolytic, hypnotic, muscle
relaxant, and anticonvulsant properties. The various compounds differ in their
potency and efficacy with regard to each of the pharmacodynamic actions. All the
effects of benzodiazepines are mediated by their interaction with specific binding
sites (benzodiazepine receptors) located on the ionotropic GABA-A receptors, but
distinct from GABA binding site. The binding of benzodiazepines to their receptors
causes an allosteric modification of the GABA-A receptor that results in an enhanced
neurotransmitter affinity for its receptor and an increased frequency of channel-
opening events thereby leading to an increase in chloride ion conductance and inhi-
bition of the action potential.
Benzodiazepines are generally viewed as safe and effective for short-term use,
although cognitive impairment and paradoxical effects such as aggression or
10 Safety and Tolerability of Anxiolytics/Sedative-Hypnotics 193
behavioural disinhibition occasionally occur (Lader 2011). A minority of people

can have paradoxical reactions such as worsened agitation or panic. Long-term use
is controversial due to concerns about adverse psychological and physical effects,
decreasing effectiveness, physical dependence and withdrawal. Due to adverse
effects associated with the long-term use of benzodiazepines, withdrawal from ben-
zodiazepines, in general, leads to improved physical and mental health. Elderly
patients are at an increased risk of suffering from both short- and long-term adverse
effects (Gould et al. 2014; Paquin et al. 2014).
10.2.1 Short-Term Effects
10.2.1.1 Sedation
Sedation is the most common subjective effect of benzodiazepines. Tolerance may

develop within a few weeks of starting treatment, but some residual effects as
increased alertness is reported by patients on stopping treatment with benzodiaze-
pines (Curran et al. 2003). At higher doses, unsteadiness, slurring of speech and
disorientation indicate over-sedation particularly when benzodiazepines are com-
bined with alcohol (Lader 2014).
10.2.1.2 Psychomotor Impairment
Benzodiazepines generally have consistent psychomotor effects in short- and long-

term use (Woods et al. 1992). They impair the performance of simple and more
complex tasks. This effect is due to alteration in the speed of execution, because
persons taking benzodiazepines tend to slow down in order to maintain accuracy of
performance. Benzodiazepines also impair the performance of simple tasks requir-
ing heightened attention. Despite the possible development of tolerance to effects
such as sedation and impaired psychomotor performance, impaired performance of
simple repetitive tasks may persist for periods of up to 1 year (Lader 2011). On the
other hand, the ability to perform tests of attention may remain impaired after sev-
eral years of treatment in long-term benzodiazepine users as compared to
non-users.
10.2.1.3 Learning and Memory Impairment
The short-term use of benzodiazepines significantly impairs multiple areas of cog-

nition, including learning and memory (Lader 2011). These cognitive effects are
enhanced when benzodiazepines are taken concomitantly with alcohol. Memories
formed before drug administration (retrograde memory) are not impaired, but the
formation of new memories after benzodiazepine use (anterograde memory) can be
significantly impaired by benzodiazepines. The more complex a memory is, e.g.
memory regarding high task complexity and delay in recall, the greater the effect of
benzodiazepines on memory formation is expected to be. It should be noted that
there are differences between the propensities of individual benzodiazepine drugs to
impair memory formation. The majority of benzodiazepines do not affect implicit
memory or priming, but lorazepam was found to impair these aspects of memory
formation (Curran et al. 1994). After months or even years of stopping treatment,
the effects of benzodiazepines on episodic memory may still be noticeable and were
not reversed by the action of flumazenil, a GABA-A receptor antagonist (Gorenstein
et al. 1994). Findings from a meta-analysis suggest that benzodiazepine users per-
form worse on most of cognitive tasks assigned, in particular those requiring the use
of verbal memory, compared to non-users (Barker et al. 2004). It should be noted
however that the studies included in the meta-analysis were very diverse with
respect to length of use, dosage and diagnosis.
10.2.1.4 Falls and Hip Fractures
The use of benzodiazepines, as well as other sedative drugs, is often associated with
falls and hip fractures, particularly in elderly patients (Ray et al 1989). Due to
increased sedation and cognitive/psychomotor impairment, benzodiazepines
increase the likelihood of falls, which are commonly the cause of hip fracture, a
devastating event in the life of an older person.
An early meta-analysis of observational epidemiological studies found that use
of benzodiazepines was associated with a 50 % increased risk of falling (Leipzig
et al. 1999). A more recent Bayesian meta-analysis has substantially confirmed this
finding (Woolcott et al. 2009). Cumming and Le Couteur (2003) provided a detailed
review of epidemiological studies on the relationship between the use of benzodiaz-
epines and the risk of hip fracture. The results of these studies were somewhat
inconsistent, and this was almost entirely explained by research design. The studies
that did not find an association between the increased risk of hip fracture and the use
of benzodiazepines were nearly all hospital-based case-control studies. These types
of studies are susceptible to confounding due to the difficulty of finding a suitable
control group. Apart from the hospital-based case-control studies, all of the remain-
ing seven studies included in the meta-analysis found that benzodiazepines use was
associated with an increased risk of hip fracture, except one. No evidence was found
to support a differential risk of hip fracture associated with short- or long-acting
benzodiazepines. The use of higher benzodiazepine doses and the incident use of
benzodiazepines were associated with the highest risk of hip fracture. Some pre-
liminary evidence was found that benzodiazepines which are substrates for oxida-
tion in the liver may be associated with a higher risk of hip fracture in the oldest old.
A recent meta-analysis including 25 studies (19 case-control studies and 6 cohort
studies) investigated the association between use of benzodiazepines and risk of
fractures (Xing et al. 2014). In general, the meta-analysis indicated that benzodiaz-
epine use was associated with a significantly increased risk of fracture (relative risk
(RR) = 1.25; 95 % confidence intervals (CI), 1.17–1.34; p < 0.001). A higher fracture
risk associated with benzodiazepine use was observed in persons ≥65 years old
(RR = 1.26; 95 % CI, 1.15–1.38; p < 0.001). When only hip fractures were consid-
ered the outcome of interest, the risk ratio increased to 1.35. A subgroup meta-
analyses did not find any significant association between long-acting benzodiazepine
use and risk of fractures (RR = 1.21; 95 % CI, 0.95–1.54; p = 0.12). Adjusting for
publication bias, the association between benzodiazepine use and the risk of frac-
ture remained but was slightly weaker (RR = 1.21; 95 % CI, 1.13–1.30) and signifi-
cant. The results of this meta-analysis demonstrated that the use of benzodiazepines,
especially short-acting ones, is associated with a moderate and clinically significant
increase in fracture risk.
In conclusion, evidence from pharmacoepidemiological studies strongly sug-
gests that the use of benzodiazepines by elderly people increases their risk of hip
fracture by up to half. Given the serious sequelae of hip fracture among older people
(including the risk of death), the use of benzodiazepines in this population should
be avoided and older persons already using benzodiazepines should have them
tapered off.
10.2.1.5 Complex Skills and Driving
Sedation and impaired psychomotor function may influence the ability to drive or
operate machinery (Leung 2011). Both simulated driving performance and actual
driving ability can be impaired and accidents are more likely. Early epidemiological
studies have confirmed that road traffic accidents involving injury or death are asso-
ciated with the use of sedative drugs (Barbone et al. 1998), and this appeared to be
related to dose, increased age and concomitant use of alcohol (Hemmelgarn et al.
1997). A systematic literature review has confirmed that exposure to benzodiaze-
pine increases the risk of traffic accidents (Smink et al. 2010). In particular, the
greater accident risk seems to be associated with the use of long half-life benzodi-
azepines, increasing dosages in the first few weeks of use. A meta-analysis of stud-
ies from 1966 to 2000 concluded that benzodiazepines were associated with a
60–80 % higher the risk of accidents (Dassanayake et al. 2011). The impaired abil-
ity to drive was often related to long plasma half-lives of hypnotics, with few excep-
tions. Daytime anxiolytics were found to impair driving independently of their
half-lives. Alcohol significantly potentiates the detrimental effects of benzodiaze-
pines on driving (Maxwell et al. 2010; Orriols et al. 2011).
10.2.1.6 Paradoxical Effects
Although benzodiazepines are traditionally prescribed as anxiolytic and sedating

agents, they may cause paradoxical effects characterised by increased talkativeness,
emotional release, excitement, excessive movement and even hostility, rage, aggres-
sion and violence (Mancuso et al. 2004). The patient may have complete or partial
amnesia for the event. Paradoxical reactions to benzodiazepines are relatively
uncommon and occur in less than 1 % of patients. The pathophysiology behind

these reactions is poorly defined. However, several risk factors such as age, genetic
predisposition, alcoholism and psychiatric and/or personality disorders have been
identified. Children and older patients are more likely to experience paradoxical
reactions with benzodiazepines compared to other patients. It has been speculated
that these patients may have a different pharmacodynamic response to benzodiaze-
pines. However, the exact differences have not been specifically described in the
literature. Some patients may have differences in the benzodiazepine-GABA-
chloride receptor at the genetic level, such changes may result in an abnormal phar-
macodynamic response. There are multiple allelic forms of benzodiazepine
receptors, resulting in different affinities for benzodiazepine drugs. Alcohol has
been found to increase the risk of benzodiazepine-associated violence and aggres-
sion (Daderman and Lidberg 1999). High-risk patients include those with border-
line personality disorders, impulse control disorder, history of substance abuse and
persistent alcohol problems (Mancuso et al. 2004). Disinhibition as a result of seda-
tive drugs is related to the type of benzodiazepine used, the dose and the mode of
administration (Bond 1998). Preoperative intravenous administration of high doses
of high-potency benzodiazepines poses a particularly enhanced risk of
disinhibition.
Paradoxical excitement is an unwanted effect which also has possible legal
implications (Paton 2002). This disinhibition associated with benzodiazepines can
produce increased anxiety, acute excitement and hyperactivity. Aggressive impulses
may be triggered with the emergence of hostility and rage; criminal acts such as
assault and rape have been recorded.
10.2.2 Long-Term Effects
10.2.2.1 Risk of Dementia
While the acute effects of benzodiazepines on memory and cognition are well docu-
mented, the possibility that they increase risk of dementia is still debated. Studies
investigating the association between the use of benzodiazepines and cognitive
decline or dementia in elderly patients have given conflicting results (Verdoux et al.
2005). Some found an increased risk of dementia or cognitive impairment in benzo-
diazepine users (Lagnaoui et al. 2002; Paterniti et al., 2002; Allard et al. 2003; Wu
et al. 2009, 2011; Gallacher et al. 2012), whereas others were not conclusive or even
reported a potential protective effect (Dealberto et al. 1997; Fastbom et al. 1998;
Hanlon et al. 1998; Lagnaoui et al. 2009; Boeuf-Cazou et al. 2011). Such previous
studies had many methodological shortcomings, and, in particular, the timing of
exposure to benzodiazepines in relation to the outcome event allowed for the pos-
sibility of reverse causation. Moreover, it should be acknowledged that insomnia,
depression and anxiety (the main indications for prescribing benzodiazepines) can
be prodromal symptoms of dementia (Amieva et al. 2008).
Two recent studies by the same research group suggested that the use of benzo-
diazepines may increase the risk of developing dementia in the elderly (Bilioti de
Gage et al. 2012, 2014). The first study evaluated the association between the use of
benzodiazepines and incident dementia (Bilioti de Gage et al. 2012). This was a
prospective, population-based study involving 1,063 men and women (mean age
78.2 years) who were free of dementia and did not start taking benzodiazepines
until at least the third year of follow-up. During a 15-year follow-up, 253 incident
cases of dementia were confirmed. New use of benzodiazepines was associated with
an increased risk of dementia (multivariable adjusted hazard ratio 1.60, 95 % confi-
dence interval 1.08–2.38). A secondary analysis consisted of the association between
incident dementia and the initiation of benzodiazepines. The hazard ratio of incident
benzodiazepine users was 1.46 (1.10–1.94). Results of a nested case-control study
suggest that use of benzodiazepines at any time was associated with roughly a 50 %
increase in the risk of dementia (adjusted odds ratio 1.55, 1.24–1.95) compared with
non-users. Similar results were obtained for past users (odds ratio 1.56, 1.23–1.98)
and recent users (1.48, 0.83–2.63), but findings were significant only for past users.
The second study investigated the association between Alzheimer’s disease and
exposure to benzodiazepines starting at least 5 years before benzodiazepine use,
considering both the dose-response relation and prodromes (anxiety, depression,
insomnia) possibly related to treatment (Bilioti de Gage et al. 2014). This was a
case-control study, based on the Quebec health insurance programme database
(RAMQ), involving 1,796 people with a first diagnosis of Alzheimer’s disease and
followed up for at least 6 years before were matched with 7,184 controls on sex, age
group and duration of follow-up. Both groups were randomly sampled from older
people (age >66) living in the community in 2000–2009. The association between
Alzheimer’s disease and benzodiazepine use started at least 5 years before diagnosis
was assessed by using multivariable conditional logistic regression. Exposure any
time to benzodiazepines was first considered and then categorised according to the
cumulative dose expressed as prescribed daily doses (1–90, 91–180, >180) and the
drug elimination half-life. Benzodiazepine use was associated with Alzheimer’s dis-
ease (adjusted odds ratio 1.51, 95 % confidence interval 1.36–1.69). No association
was found for a cumulative dose <91 prescribed daily doses. The association
increased with exposure density (1.32 (1.01–1.74) for 91–180 prescribed daily
doses and 1.84 (1.62–2.08) for >180 prescribed daily doses) and with the drug half-
life (1.43 (1.27–1.61) for short-acting drugs and 1.70 (1.46–1.98) for long-acting
ones). Based on the findings of these studies, the authors suggest to avoid indis-
criminate long-term and widespread use of benzodiazepine in general population.
The possible association between dementia and the use of benzodiazepines
should nevertheless be viewed with caution and the above studies have prompted
some criticism (Barbui et al. 2013; Kmietowicz 2014; Salzman and Shader 2015).
While the above studies by Billioti de Gage et al. (2012, 2014) provide evidence in
favour of such an association, such as an increase in risk with increasing duration of
exposure and increasing benzodiazepine half-life, observational studies on the topic
may be subject to confounding by indication in the case that the drugs were pre-
scribed for anxiety and/or insomnia that were due to prodromal symptoms of
dementia (Barbui et al. 2013). In addition, the follow-up of 6 years may be consid-
ered relatively short to study a neurodegenerative disease such as Alzheimer’s,
which may take much longer to develop (Kmietowicz 2014). It was also pointed out
that the cohort of patients under study may have had mild cognitive impairment
although they did not have Alzheimer disease and that other factors which may be
associated with memory impairment, such as alcohol use, were not controlled for
(Salzman and Shader 2015).
In conclusion, findings on the risk of Alzheimer’s disease with benzodiazepine
use are thought provoking and suggest that these drugs may increase the risk of
Alzheimer’s disease, but such findings should be considered an indication of defini-
tive causality, in particular given that observational studies are not the most suited
study design to investigate causality.
10.2.2.2 Tolerance, Dependence and Withdrawal
The main problem associated with long-term use of benzodiazepines is the develop-
ment of tolerance and dependence (Ashton 2005; O’Brien 2005; Lader 2011).
Tolerance can be defined as a reduced pharmacological response following
repeated administration of the same drug dose. As a consequence, increasing doses
are needed to produce the same response. Benzodiazepine tolerance develops at dif-
ferent rates and to different extents (Ashton 2005). Tolerance to hypnotic effects
develops rapidly, within a few days or weeks of regular use. Although some subjects
report continued efficacy of benzodiazepine hypnotics, clinical experience suggests
that a considerable proportion of hypnotic users must gradually increase their dos-
age to maintain a given level of therapeutic response. Tolerance to the anticonvul-
sant and muscular relaxant effects of benzodiazepines also develops relatively
quickly. On the other hand, tolerance to the anxiolytic effects develops more slowly,
over a few months, and clinical experience shows that long-term use does little to
control anxiety and may even aggravate it (Ashton 2005). There is also evidence of
dosage escalation when benzodiazepines are used for their anxiolytic effect. In gen-
eral, little tolerance develops to the amnesic other cognitive effects caused by ben-
zodiazepines. Studies of long-term users have shown deficits in learning, memory,
attention and visuospatial ability. A meta-analysis of 13 research studies that evalu-
ated the effect on long-term use of benzodiazepines on cognitive function found that
long-term benzodiazepine users had moderate-significant deficits for each of the 12
of the cognitive domains tested compared to controls (Barker et al. 2004).
According to the World Health Organization, dependence is defined as a strong
compulsion to take a substance, difficulty in controlling its use, the development of
tolerance and the presence of a withdrawal state. Withdrawal consists of a group of
symptoms which occur on cessation or reduction of use of a psychoactive substance
that has been used repeatedly, often for prolonged periods and/or in high doses
(Lader 1987). Withdrawal may be accompanied with physiological disturbances.
Withdrawal syndrome is one of the markers of dependence. Benzodiazepine discon-
tinuation or an abrupt reduction in dose may result in rebound and/or withdrawal
syndrome, even after only 3–4 weeks of treatment. Rebound can be considered as
the mildest form of withdrawal (Lader 2011). Rebound symptoms are the return of
the symptoms for which the patient was treated but with a greater intensity than
before. The most common phenomenon with benzodiazepines is the rebound of the
hypnotic effect which is likely when stopping hypnotic benzodiazepines, particu-
larly short-acting ones, even after only a few days or nights of use (Bonnet and
Arand 1999). After stopping the hypnotic benzodiazepine, the insomnia can return
in an exaggerated form, time to sleep onset is prolonged, sleep is more disturbed
and it is shorter in duration. Rebound is generally short lived lasting a night or two,
but can panic the patient into resuming the medication. Withdrawal is a more seri-
ous phenomenon, consisting in a characteristic grouping of signs and symptoms
that occur when the benzodiazepine is stopped or the dose is reduced. Withdrawal
often involves the onset of new symptoms not experienced previously by the patient.
The occurrence of a withdrawal syndrome is the main indicator of physical depen-
dence (Chouinard 2004). Withdrawal syndrome associated with benzodiazepine use
is generally related to high dosage and long-term treatment. Withdrawal symptoms
usually occur after 4–6 weeks of use, but only in about 15–30 % of patients (Lader
1998). Benzodiazepine withdrawal symptoms may be divided into common, less
common and rare symptoms (Table 10.1). Severe symptoms usually occur as a
result of abrupt or over-rapid withdrawal. Abrupt withdrawal can be dangerous;
therefore, the dose of benzodiazepines should be gradually tapered off until they are
discontinued. The symptoms of withdrawal usually subside in 2–4 weeks but can be
prolonged. Withdrawal symptoms may occur even if the dose of benzodiazepines is
reduced gradually, but symptoms tend to be less severe. Nevertheless, they may
persist as a withdrawal syndrome for months after discontinuation of benzodiaze-
pines. A prospective study by Vikander et al. (2010) identified four patterns of with-
drawal symptoms over time: (1) a gradual decrease in symptom severity over
50 weeks, (2) an increase in the severity of symptoms at the beginning of dose
tapering followed by a decrease in severity after tapering the dose off, (3) an increase
Table 10.1 Benzodiazepine withdrawal symptoms

Common Less common Rare
Anxiety Muscle pain Convulsions
Insomnia Vomiting Delirium
Dysphoria Hyperacusis Psychotic symptoms
Excitability Photophobia Delusions
Poor memory and concentration Altered sensation Hallucinations
Dizziness Depersonalization Mania
Gastrointestinal problems Derealization Depression
Palpitations Suicidal ideation
Sweating
Tremor
Nausea
Headache
in the severity of symptoms 4 weeks after the cessation of benzodiazepine tapering,

and (4) no change over 50 weeks.
10.2.2.3 Abuse
A clear distinction should be made between dependence and withdrawal from thera-
peutic or somewhat higher doses within the medical context and abuse of benzodi-
azepines in the context of recreational and illicit use (Lader 2014). Benzodiazepines
are widely misused, although patterns vary from country to country and from region
to region. One type takes the form of binges, say at weekends, another regular sus-
tained high-dose usage. Some misusers keep to oral use, whereas others inject intra-
venously or sniff intranasally like with cocaine use. The abuse of benzodiazepines
depends on their formulation, bioavailability and pharmacokinetics. Temazepam
and flunitrazepam are known to be often misused. Although benzodiazepines may
be misused alone, they may also be misused along with other drugs, for example, to
potentiate the euphorigenic effects of opioids, lessen the impact following the
effects of cocaine or interact in a complex way with amphetamines or other drugs of
abuse. Drug abusers may turn to benzodiazepines if other drugs of abuse become
scarce and expensive. The risks of benzodiazepine abuse such as viral infection or
local tissue necrosis are well known and are associated with intravenous drug use.
Overdose is a hazard, particularly in combination with other psychotropic drugs.
Another danger is related to the potentiation of the depressant effects of alcohol by
benzodiazepines. This has been associated with an increased likelihood of criminal
acts, often accompanied by amnesia. The misuse of benzodiazepines is undoubtedly
dangerous, and the potential for misuse should be a consideration when deciding to
prescribe these drugs.
10.2.2.4 Mortality
Adverse effects of benzodiazepines are generally unpleasant but may not be severe;
most adverse effects are reversible. However, recent data suggest that use of benzo-
diazepines may be associated with excess mortality (Charlson et al. 2009; Kripke
et al. 2012). A systematic review has examined the risk of death associated with
benzodiazepine use in studies published from 1990 onwards (Charlson et al. 2009).
Data from six cohort and three registry studies indicate that regular users and illicit
benzodiazepine users had a higher risk of mortality compared to non-users. A recent
matched cohort study, based on electronic medical records and involving 10,529
people who received hypnotic agents (including both benzodiazepine and non-
benzodiazepine hypnotics) and 23,676 controls with no hypnotic prescriptions, esti-
mated the mortality risks, using proportional hazard regression models (Kripke
et al. 2012). For patients prescribed 0.4–18, 18–132 and >132 pills/year, the hazard
ratios were 3.60 (95 % CI 2.92, 4.44), 4.43 (3.67, 5.36) and 5.32 (4.50, 6.30),
respectively. Thus, even occasional hypnotic users had over three times the
background risk of dying in 2.5 years. Selective prescription of hypnotics for ailing
patients was ruled out as the main explanation. The presence of co-morbidities was
associated with a significant increase in the risk of death among patients receiving
hypnotics, but this accounted for only a small proportion of the excess risk.
10.2.2.5 Skin Reactions
Skin manifestations such as generalised reactions, contact dermatitis, photoderma-

titis and Stevens-Johnson syndrome are rarely associated with benzodiazepine
treatment.
In April 2013, the Pharmacovigilance Risk Assessment Committee (PRAC) of
the European Medicines Agency (EMA) issued a warning about tetrazepam, a ben-
zodiazepine that has been used to treat painful muscle spasms (such as low-back
pain and neck pain) and spasticity (excessive stiffness of muscles) in some European
countries (European Medicines Agency 2013). The alert was prompted by the
occurrence of life-threatening skin reactions (including Stevens-Johnson syndrome,
toxic epidermal necrolysis, and drug rash with eosinophilia and systemic symptoms
syndrome), as described by the French National Agency for the Safety of Medicine
and Health Products (Proy-Vega et al. 2014). After an assessment of available data
on the risk of skin reactions, the PRAC concluded that tetrazepam is indeed associ-
ated with a low but increased risk of serious skin reactions compared to other ben-
zodiazepines. The Coordination Group for Mutual Recognition and Decentralized
Procedures of medicines for human use of the EMA agreed with the PRAC conclu-
sion that the benefits of tetrazepam do not outweigh its risks, and on 29 May 2013,
EMA adopted a final decision and suspended the marketing authorizations of tetraz-
epam across the European Union (EMA 2013). Proy-Vega et al. (2014) have recently
commented on the clinical evidence leading to tetrazepam withdrawal, underlining
the lack of randomised controlled clinical trials evaluating tetrazepam efficiency
and safety. In their conclusion, they claim that ‘it is very important to foster a strong
interaction between pharmacovigilance agencies, scientific publications and health
professionals, in order to improve and optimise exchange of knowledge on clinical
cases of ADRs’.
In December 2013, the FDA released a warning that clobazam, a benzodiazepine
medication approved as adjunctive therapy to treat seizures that accompany Lennox-
Gastaut syndrome, may cause serious skin events, including Stevens-Johnson syn-
drome (SJS), toxic epidermal necrolysis (TEN) and acute generalised exanthematous
pustulosis (AGEP), and consequently approved changes to clobazam label and
medication guide (FDA 2013a). The FDA identified 20 cases of severe skin reac-
tions from its Adverse Event Reporting System database: all cases had resulted in
hospitalisation, one case in blindness and one in death. These reactions can occur at
any time during clobazam treatment, but the risk is greater during the first 8 weeks
of treatment or when clobazam is stopped and then restarted. However, no compara-
tive pharmacovigilance analysis was conducted for other anticonvulsant benzodiaz-
epines, including clonazepam, clorazepate, diazepam and lorazepam.
10.3 Buspirone
Buspirone is an azapirone used in the treatment of generalised anxiety disorder.

While the anxiolytic effects of benzodiazepines usually occur within a few days of
therapy, buspirone requires chronic treatment for effectiveness (Chessick et al.
2006). Buspirone lacks the sedative, muscle relaxant and anticonvulsant properties
of the benzodiazepines. The underlying mechanism of action of buspirone is not
clear; however, it is thought that its anxiolytic effects are mediated through interac-
tions with the serotonin 5HT1A receptor, where it acts as a partial agonist (Loane and
Politis 2012). The most common side effects of buspirone are dizziness, headache
and light-headedness. Buspirone does not impair psychomotor performance or
results in abuse, dependence or withdrawal. Like benzodiazepines, buspirone
appears to be safe even when given in very high doses.
10.4 Pregabalin
Pregabalin is a pharmacological agent approved in many countries for the treatment

of neuropathic pain, partial seizures and generalized anxiety disorder (Frampton
2014). Pregabalin is a structural analogue of GABA that neither interacts with
GABA receptors nor alters GABA uptake or degradation. On the other hand, prega-
balin binds to the α2δ (alpha-2-delta) subunit of the voltage-dependent calcium
channel in the central nervous system so thus decreasing the release of neurotrans-
mitters including glutamate and substance P. The most frequent adverse effects of
pregabalin include dizziness, somnolence, dry mouth, peripheral oedema, blurred
vision and weight gain (Zaccara et al. 2011).
Following its introduction, accumulating evidence from case reports, data-
bases and a limited number of studies have suggested that pregabalin has the
potential to cause abuse and dependence (Gahr et al. 2013). Schwan et al. (2010)
analysed data from the Swedish national register of adverse drug reactions and
concluded that pregabalin is likely to be associated with an abuse potential based
on 16 positive reports. As a result, the prescribing information was changed and
now states that cases of pregabalin abuse have been reported and patients with a
previous history of psychotropic substance abuse should be monitored closely for
signs of pregabalin abuse (Lyrica SPC). However, there is currently limited evi-
dence on this topic. According to recent review articles (Baldwin et al. 2013;
Frampton 2014; Schifano 2014), the potential for abuse of pregabalin is low, as its
positive psychological effects are weak and not maintained over time. Moreover,
unless stopped abruptly, pregabalin seems to carry a limited risk for physical
dependence or withdrawal. In this respect, a recent study documented that gradual
discontinuation of pregabalin after 4–24 weeks of treatment at a dose ranging
from 150 to 600 mg/day was not associated with clinically significant withdrawal
symptoms (Kasper et al. 2014).
Post-marketing surveillance studies are needed to identify risk factors for prega-
balin abuse and dependence. The assessment of pregabalin’s potential to cause
addictive behaviours is also of particular clinical relevance as this agent is currently
under evaluation for the treatment of benzodiazepine and alcohol dependence (Oulis
and Konstantakopoulos 2012).
10.5 Non-benzodiazepine Hypnotics
Non-benzodiazepine hypnotics, the so-called Z-drugs, including zolpidem, zaleplon,

zopiclone and eszopiclone, are a class of drugs structurally unrelated to benzodiaz-
epines, but with a similar mechanism of action. These agents are agonists at the α1
subunit of GABA-A receptors which mediate sedation. They have become preferred
drugs for the treatment of insomnia, in particular among older adults, because of
perceived improved safety profiles compared with traditional benzodiazepines. The
most common adverse events of non-benzodiazepine hypnotics are drowsiness or
fatigue, headache, nightmares and nausea or gastrointestinal disturbances. However,
recent evidence in elderly patients over the age of 65 has highlighted potential safety
concerns of these medications and zolpidem specifically with regard to effects on
balance and memory and on fracture risk (Levy 2014).
On January 2013, the US FDA issued a warning recommending that the bedtime
dose of zolpidem should be lowered based on new data showing that blood levels in
some patients may be high enough the morning after use to impair activities that
require alertness, including driving. Therefore, FDA required manufacturers of zolpi-
dem products to lower the recommended initial dose for women from 10 to 5 mg for
immediate-release products and from 12.5 to 6.25 mg for extended-release products
(Food and Drug Administration 2013b). Driving simulation and laboratory studies
had established a threshold of 50 ng/ml, above which zolpidem is associated with
decreased alertness and increased risk of adverse events. A randomised, placebo-
controlled trial has shown that a single 5 mg dose of zolpidem resulted in clinically
significant balance and cognitive impairments upon awakening from sleep (Frey et al.
2011). In particular, 58 % of older adults (7/12) and 27 % of younger adults (3/11)
tested had a loss of balance after taking zolpidem, whereas none of the same partici-
pants had a loss of balance during ten pre-sleep practice trials. A loss of balance after
zolpidem use was marked and more common in older adults compared to placebo.
Early evidence of zolpidem-associated hip fracture in older adults (Wang et al.
2001) has recently been reinforced by findings from three studies (Finkle et al. 2011;
Kang et al., 2012; Berry et al. 2013). In the first study (Finkle et al. 2011), zolpidem
was found to have a similar risk of hip fractures compared with diazepam and loraze-
pam, but significantly lower compared with alprazolam. A second investigation found
that fracture risk was significantly greater with zolpidem compared with traditional
benzodiazepines (odds ratio [OR]: 1.72 vs. 1.00, respectively) (Kang et al. 2012). A
sub-analysis of 135 patients aged 85 or older found an even more pronounced risk with
zolpidem compared with benzodiazepines (OR: 4.48 vs. 1.13, respectively). A third
study utilised a nursing home population (Berry et al. 2013). Claims data were evalu-
ated in residents who were prescribed zolpidem, eszopiclone or zaleplon during the
‘hazard period’ (within 30 days of a hip fracture) or during the ‘control period’ (60 days
or longer before hip fracture). Residents who were prescribed a non-benzodiazepine
during the hazard period were more likely to experience a hip fracture compared with
residents prescribed a non-benzodiazepine during the control period (OR: 1.66).
Due to a lack of evidence on the optimal selection of zolpidem and other non-
benzodiazepines in the elderly, it would be prudent to use these hypnotic agents
sparingly and cautiously in older adults.
10.6 Conclusion
Benzodiazepines continue to be used in the management of patients with anxiety

and insomnia. Based on cognitive and psychomotor impairment as well as abuse
and dependence liability associated with benzodiazepines, clinicians must always
balance the benefits and risks when prescribing these agents especially in the elderly.
Other pharmacological agents used to treat anxiety disorders or insomnia, including
pregabalin and non-benzodiazepine hypnotics, may share with benzodiazepines
similar tolerability and safety issues. Pharmacovigilance surveillance studies can
provide important information on specific adverse event features, patterns of clini-
cal symptoms, severity of the events and, where applicable, fatality rates of various
anxiolytics/sedative-hypnotics. Healthcare systems at the local, regional or national
levels may wish to employ patient safety monitoring programmes for anxiolytics/
sedative-hypnotics based upon pharmacovigilance studies.
References
Allard J, Artero S, Ritchie K (2003) Consumption of psychotropic medication in the elderly: a re-
evaluation of its effect on cognitive performance. Int J Geriatr Psychiatry 18:874–878
Amieva H, Le Goff M, Millet X, Orgogozo JM, Peres K, Barberger-Gateau P, Jacqmin-Gadda H,
Dartigues JF (2008) Prodromal Alzheimer’s disease: successive emergence of the clinical
symptoms. Ann Neurol 64:492–498
Ashton H (2005) The diagnosis and management of benzodiazepine dependence. Curr Opin
Baldwin DS, Ajel K, Masdrakis VG, Nowak M, Rafiq R (2013) Pregabalin for the treatment of
generalized anxiety disorder: an update. Neuropsychiatr Dis Treat 9:883–892
Baldwin DS, Anderson IM, Nutt DJ, Allgulander C, Bandelow B, den Boer JA, Christmas DM,
Davies S, Fineberg N, Lidbetter N, Malizia A, McCrone P, Nabarro D, O’Neill C, Scott J, van der
Wee N, Wittchen HU (2014) Evidence-based pharmacological treatment of anxiety disorders,
post-traumatic stress disorder and obsessive-compulsive disorder: a revision of the 2005 guide-
lines from the British Association for Psychopharmacology. J Psychopharmacol 28:403–439
Bandelow B, Sher L, Bunevicius R, Hollander E, Kasper S, Zohar J, Moller HJ, WFSBP Task
Force on Mental Disorders in Primary Care, WFSBP Task Force on Anxiety Disorders, OCD
and PTSD (2012) Guidelines for the pharmacological treatment of anxiety disorders, obsessive-
compulsive disorder and posttraumatic stress disorder in primary care. Int J Psychiatry Clin
Pract 16:77–84
Barbone F, McMahon AD, Davey PG, Morris AD, Reid IC, McDevitt DG, MacDonald TM (1998)
Association of road-traffic accidents with benzodiazepine use. Lancet 352:1331–1336
Barbui C, Gastaldon C, Cipriani A (2013) Benzodiazepines and risk of dementia: true association
or reverse causation? Epidemiol Psychiatry Sci 22:307–308
Barker MJ, Greenwood KM, Jackson M, Crowe SF (2004) Cognitive effects of long-term benzo-
diazepine use: a meta-analysis. CNS Drugs 18:37–48
Berry SD, Lee Y, Cai S, Dore DD (2013) Nonbenzodiazepine sleep medication use and hip frac-
tures in nursing home residents. JAMA Intern Med 173:754–761
Billioti de Gage S, Bégaud B, Bazin F, Verdoux H, Dartigues JF, Pérès K, Kurth T, Pariente A
(2012) Benzodiazepine use and risk of dementia: prospective population based study. BMJ
345:e6231
Billioti de Gage S, Moride Y, Ducruet T, Kurth T, Verdoux H, Tournier M, Pariente A, Bégaud B
(2014) Benzodiazepine use and risk of Alzheimer’s disease: case-control study. BMJ 349:g5205
Boeuf-Cazou O, Bongue B, Ansiau D, Marquie JC, Lapeyre-Mestre M (2011) Impact of long-term
benzodiazepine use on cognitive functioning in young adults: the VISAT cohort. Eur J Clin
Pharmacol 67:1045–1052
Bond AJ (1998) Drug-induced behavioural disinhibition. Incidence, mechanisms and therapeutic
implications. CNS Drugs 9:41–57
Bonnet MH, Arand DL (1999) The use of lorazepam TID for chronic insomnia. Int Clin
Charlson F, Degenhardt L, McLaren J, Hall W, Lynskey M (2009) A systematic review of research
examining benzodiazepine-related mortality. Pharmacoepidemiol Drug Saf 18:93–103
Chessick CA, Allen MH, Thase M, Batista Miralha da Cunha AB, Kapczinski FF, de Lima MS, dos
Santos Souza JJ (2006) Azapirones for generalized anxiety disorder. Cochrane Database Syst
Rev 3, CD006115
Chouinard G (2004) Issues in the clinical use of benzodiazepines: potency, withdrawal, and
rebound. J Clin Psychiatry 65(Suppl 5):7–12
Cumming RG, Le Couteur DG (2003) Benzodiazepines and risk of hip fractures in older people: a
review of the evidence. CNS Drugs 17:825–837
Curran HV, Bond A, O’Sullivan G, Bruce M, Marks I, Lelliot P, Shine P, Lader M (1994) Memory
functions, alprazolam and exposure therapy: a controlled longitudinal trial of agoraphobia with
panic disorder. Psychol Med 24:969–976
Curran HV, Collins R, Fletcher S, Kee SCY, Woods B, Iliffe S (2003) Older adults and withdrawal
from benzodiazepine hypnotics in general practice: effects on cognitive function, sleep, mood
and quality of life. Psychol Med 33:1223–1227
Daderman AM, Lidberg L (1999) Flunitrazepam (rohypnol) abuse in combination with alcohol
causes premeditated, grievous violence in male juvenile offenders. J Am Acad Psychiatry Law
27:83–99
Dassanayake T, Michie P, Carter G, Jones A (2011) Effects of benzodiazepines, antidepressants
and opioids on driving: a systematic review and meta-analysis of epidemiological and experi-
mental evidence. Drug Saf 34:125–156
Dealberto MJ, McAvay GJ, Seeman T, Berkman L (1997) Psychotropic drug use and cognitive
decline among older men and women. Int J Geriatr Psychiatry 12:567–574
European Medicine Agency (EMA). Recommendation to suspend tetrazepam-containing medi-
cines endorsed by CMDh. EMA Drug Safety Communication (29 April 2013). Available from:
http://www.ema.europa.eu/ema/index.jsp?curl=pages/news_and_events/news/2013/04/news_
detail_001777.jsp&mid=WC0b01ac058004d5c1. Accessed on 19 May 2015
Fastbom J, Forsell Y, Winblad B (1998) Benzodiazepines may have protective effects against
Alzheimer disease. Alzheimer Dis Assoc Disord 12:14–17
Finkle WD, Der JS, Greenland S, Adams JL, Ridgeway G, Blaschke T, Wang Z, Dell RM, VanRiper
KB (2011) Risk of fractures requiring hospitalization after an initial prescription for zolpidem,
alprazolam, lorazepam, or diazepam in older adults. J Am Geriatr Soc 59:1883–1890
Food and Drug Administration (FDA) (2013a). FDA warns of serious skin reactions with the anti-
seizure drug Onfi (Clobazam) and has approved label changes. FDA Drug Safety Communication
(12 Dec 2013). Available from: http://www.fda.gov/Drugs/DrugSafety/ucm377204.htm.

Accessed on: 19 May 2015
Food and Drug Administration (FDA) (2013b) Risk of next-morning impairment after use of
insomnia drugs; FDA requires lower recommended doses for certain drugs containing zolpi-
dem. FDA Drug Safety Communications (10 Jan 2013). Available from: http://www.fda.gov/
Drugs/DrugSafety/ucm334033.htm. Accessed on: 19 May 2015
Frampton JE (2014) Pregabalin: a review of its use in adults with generalized anxiety disorder.
CNS Drugs 28:835–854
Frey DJ, Ortega JD, Wiseman C, Farley CT, Wright KP (2011) Influence of zolpidem and sleep
inertia on balance and cognition during nighttime awakening: a randomized placebo-controlled
trial. J Am Geriatr Soc 59:73–81
Gahr M, Freudenmann RW, Hiemke C, Kölle MA, Schönfeldt-Lecuona C (2013) Pregabalin abuse and
dependence in Germany: results from a database query. Eur J Clin Pharmacol 2013:1335–1342
Gallacher J, Elwood P, Pickering J, Bayer A, Fish M, Ben-Shlomo Y (2012) Benzodiazepine use
and risk of dementia: evidence from the Caerphilly Prospective Study (CaPS). J Epidemiol
Community Health 66:869–873
Gorenstein C, Bernik MA, Pompeia S (1994) Differential acute psychomotor and cognitive effects
of diazepam on longterm benzodiazepine users. Int Clin Psychopharmacol 9:145–153
Gould RL, Coulson MC, Patel N, Highton-Williamson E, Howard RJ (2014) Interventions for
reducing benzodiazepine use in older people: meta-analysis of randomized controlled trials. Br
J Psychiatry 204:98–107
Greenblatt DJ (2011) Psychotropic drug prescribing in the United States: extent, costs, and expen-
ditures. J Clin Psychopharmacol 31:1–3
Gunja N (2013) The clinical & forensic toxicology of Z-drugs. J Med Toxicol 9:163–171
Hanlon JT, Horner RD, Schmader KE, Fillenbaum GG, Lewis IK, Wall WE Jr, Landerman LR,
Pieper CF, Blazer DG, Cohen HJ (1998) Benzodiazepine use and cognitive function among
community-dwelling elderly. Clin Pharmacol Ther 64:684–692
Hemmelgarn B, Suissa S, Huang A, Boivin JF, Pinard G (1997) Benzodiazepine use and the risk
of motor vehicle crash in the elderly. JAMA 278:27–31
Kang DY, Park S, Rhee CW, Kim YJ, Choi NK, Lee J, Park BJ (2012) Zolpidem use and risk of
fracture in elderly insomnia patients. J Prev Med Public Health 45:219–226
Kasper S, Iglesias-García C, Schweizer E, Wilson J, DuBrava S, Prieto R, Pitman VW, Knapp L
(2014) Pregabalin long-term treatment and assessment of discontinuation in patients with gen-
eralized anxiety disorder. Int J Neuropsychopharmacol 17:685–695
Kmietowicz Z (2014) Benzodiazepines may be linked to Alzheimer’s disease, study finds. BMJ
349:g5555. doi:10.1136/bmj.g5555 (Published 10 September 2014)
Kripke DF, Langer RD, Kline LE (2012) Hypnotic association with mortality or cancer; a matched
cohort study. BMJ Open 2, e000850
Lader M (1987) Long-term anxiolytic therapy: the issue of drug withdrawal. J Clin Psychiatry
48:12–16
Lader M (1998) Withdrawal reactions after stopping hypnotics in patients with insomnia. CNS
Drugs 10:425–440
Lader M (2011) Benzodiazepines revisited—will we ever learn? Addiction 106:2086–2109.
add_3563
Lader M (2014) Benzodiazepine harm: how can it be reduced? Br J Clin Pharmacol 77:295–301
Lagnaoui R, Begaud B, Moore N, Chaslerie A, Fourrier A, Letenneur L, Dartigues JF, Moride Y
(2002) Benzodiazepine use and risk of dementia: a nested case-control study. J Clin Epidemiol
55:314–318
Lagnaoui R, Tournier M, Moride Y, Wolfson C, Ducruet T, Begaud B, Moore N (2009) The risk of
cognitive impairment in older community-dwelling women after benzodiazepine use. Age
Ageing 38:226–228
Leipzig RM, Cumming RG, Tinetti ME (1999) Drugs and falls in older people: a systematic review
and meta-analysis. I: psychotropic drugs. J Am Geriatr Soc 47:30–39
Leung SY (2011) Benzodiazepines, opioids and driving: an overview of the experimental research.
Drug Alcohol Rev 30:281–286
Levy HB (2014) Non-benzodiazepine hypnotics and older adults: what are we learning about zol-
pidem? Expert Rev Clin Pharmacol 7:5–8
Loane C, Politis M (2012) Buspirone: what is it all about? Brain Res 1461:111–118
Mancuso C, Tanzi MG, Gabay M (2004) Paradoxical reactions to benzodiazepines: literature
review and treatment options. Pharmacotherapy 24:1177–1185
Maxwell HG, Dubois S, Weaver B, Bédard M (2010) The additive effects of alcohol and benzodi-
azepines on driving. Can J Public Health 101:353–357
O’Brien CP (2005) Benzodiazepine use, abuse, and dependence. J Clin Psychiatry 66(suppl
2):28–33
Orriols L, Philip P, Moore N, Castot A, Gadegbeku B, Delorme B, Mallaret M, Lagarde E (2011)
Benzodiazepine-like hypnotics and the associated risk of road traffic accidents. Clin Pharmacol
Ther 89:595–601
Oulis P, Konstantakopoulos G (2012) Efficacy and safety of pregabalin in the treatment of alcohol
and benzodiazepine dependence. Expert Opin Investig Drugs 21:1019–1029
Paquin AM, Zimmerman K, Rudolph L (2014) Risk versus risk: a review of benzodiazepine reduc-
tion in older adults. Exp Opin Drug Saf 13:919–934
Paterniti S, Dufouil C, Alperovitch A (2002) Long-term benzodiazepine use and cognitive
decline in the elderly: the epidemiology of vascular aging study. J Clin Psychopharmacol 22:
285–293
Paton C (2002) Benzodiazepines and disinhibition: a review. Psychiatr Bull 26:460–462
Proy-Vega B, Aguirre C, de Groot P, Solis-Garcia del Pozo J, Jordan J (2014) On the clinical evi-
dence leading to tetrazepam withdrawal. Exp Opin Drug Saf 13:705–712
Ray WA, Griffin MR, Downey W (1989) Benzodiazepines of long and short elimination half-life
and risk of hip fracture. JAMA 262:3303–3307
Salzman C, Shader RI (2015) Benzodiazepine use and risk for Alzheimer disease. J Clin
Schifano F (2014) Misuse and abuse of pregabalin and gabapentin: cause for concern? CNS Drugs
28:491–496
Schwan S, Sundström A, Stjernberg E, Hallberg E, Hallberg P (2010) A signal for an abuse liability
for pregabalin–results from the Swedish spontaneous drug reaction reporting system. Eur J
Clin Pharmacol 66:947–953
Smink BE, Egberts AC, Lusthof KJ, Uges DR, de Gier JJ (2010) The relationship between benzo-
diazepine use and traffic accidents: a systematic literature review. CNS Drugs 24:639–653
Verdoux H, Lagnaoui R, Begaud B (2005) Is benzodiazepine use a risk factor for cognitive decline
and dementia? A literature review of epidemiological studies. Psychol Med 35:307–315
Vikander B, Koechling UM, Borg S, Tönne U, Hiltunen AJ (2010) Benzodiazepine tapering: a
prospective study. Nord J Psychiatry 64:273–282
Wang PS, Bohn RL, Glynn RJ, Mogun H, Avorn J (2001) Zolpidem use and hip fractures in older
people. J Am Geriatr Soc 49:1685–1690
Woods JH, Katz JL, Winger G (1992) Benzodiazepine use, abuse and consequences. Pharmacol
Rev 44:151–347
Woolcott JC, Richardson KJ, Wiens MO, Patel B, Marin J, Khan KM, Marra CA (2009) Meta-
analysis of the impact of 9 medication classes on falls in elderly persons. Arch Intern Med
169:1952–1960
Wu CS, Wang SC, Chang IS, Lin KM (2009) The association between dementia and long-term use
of benzodiazepine in the elderly: nested case-control study using claims data. Am J Geriatr
Wu CS, Ting TT, Wang SC, Chang IS, Lin KM (2011) Effect of benzodiazepine discontinuation on
dementia risk. Am J Geriatr Psychiatry 19:151–159
Xing D, Ma XL, Ma JX, Wang J, Yang Y, Chen Y (2014) Association between use of benzodiaze-
pines and risk of fractures: a meta-analysis. Osteoporos Int 25:105–120
Zaccara G, Gangemi P, Perucca P, Specchio L (2011) The adverse event profile of pregabalin:
a systematic review and meta-analysis of randomized controlled trials. Epilepsia 52:
826–836
Chapter 11
Safety and Tolerability of Mood Stabilisers
Michele Fabrazzo and Alfonso Tortorella
Abstract Safety and tolerability of mood stabilisers are major clinical concerns
when used in bipolar patients. Side effects of lithium and some antiepileptics have
been reviewed in the context of a spontaneous reporting database over the last
10 years (FDA database and published reports). During pregnancy, antiepileptics
show great concerns, and adverse events are all related to childbirth, whilst congeni-
tal abnormalities are not higher than previously estimated. Cutaneous adverse reac-
tions are the most prevalent in children and adolescent. In adult lithium-treated
patients, nephrotoxicity is still a major problem; the combination with carbamaze-
pine and valproate can increase the risk of hypothyroidism. Hyperparathormonemia
and hypercalcaemia are unrecognised and underappreciated adverse effects. Acute
exacerbation of psoriasis is still a major problem, and the risk of skin reactions with
eosinophilia and systemic symptoms is higher when mood stabilisers are used con-
comitantly; when associated with antipsychotics, the risk of pneumonia is possible
(highest risk for olanzapine plus carbamazepine). A decrease in total body water
and the decline of glomerular filtration rate represent the main lithium adverse
effects in elderly patients. Long-term treatment is associated with impairment in
immediate verbal learning and memory and creativity performance. Antiepileptics
display significant adverse events (hyponatraemia, cardiac toxicity) and the risk of
multiple drug-drug interactions is very high. Cumulative exposure to antipsychotics
and mood stabilisers can be associated with vascular stiffness (elevated systolic
blood pressure), hypertriglyceridaemia, insulin resistance and low HDL cholesterol.
In light of the above considerations, clinicians should continuously and further
assess risks and benefits of mood stabilisers when treating bipolar patients.
Keywords Mood stabilisers • Lithium • Safety • Tolerability • Bipolar disorder
M. Fabrazzo (*) • A. Tortorella

Department of Psychiatry,
University of Naples SUN, Naples, Italy
e-mail: michele.fabrazzo@tiscali.it; alfonsotortorella@gmail.com

DOI 10.1007/978-3-319-24741-0_11
210 M. Fabrazzo and A. Tortorella
11.1 Introduction
Bipolar disorder (BD) is a dynamic illness that is amongst the top 30 causes of
disability worldwide and is associated with a complex clinical picture characterised
by dramatic changes in mood (typically manic/hypomanic episodes alternate with
episodes of depression), energy, cognition and to multiple psychiatric/nonpsychiat-
ric comorbidities (Strakowski et al. 2011). BD affects 3 % or more of the general
population and ranks second only to unipolar depression as a cause of worldwide
disability (Murray and Lopez 1997; Kupfer 2005).
In this chapter, we will review the various adverse side effects reported for lith-
ium and some antiepileptic drugs (AEDs) indicated as mood stabilisers (sodium
valproate, carbamazepine, oxcarbazepine and lamotrigine). The safety profile of
lithium and individual AEDs will be considered in the context of a spontaneous
reporting database over the last 10 years.
Taken in therapeutic doses, these drugs can cause a range of side effects that can
be divided into five different groups: (1) common (greater than 10 %), (2) uncom-
mon (between 1 and 10 %), (3) rare (between 0.1 and 1 %), (4) very rare (between
0.01 and 0.1 %) and (5) isolated reports (less than 0.01 %). Moreover, when assess-
ing side effects of drugs, many variables can be considered, and the age of patients
is one of the most important. We will therefore divide this chapter on the basis of the
effects reported in different populations of patients.
11.2 Lithium Safety
Although they are rapidly outdated, all major evidence-based guidelines support the
use of lithium as a first-line option for long-term maintenance treatment and pro-
phylaxis of BD (Yatham et al. 2013; Pfennig et al. 2013; Goodwin 2009; Grunze
et al. 2009). Lithium has been demonstrated to be a suitable first-line treatment in
several clinical circumstances including acute manic/hypomanic episodes, relapse
prevention, suicidal ideation and as an augmentation agent in the treatment of uni-
polar major depression (Fountoulakis et al. 2005; Baldessarini and Tondo 2008;
Geddes et al. 2010).
Despite the fact that lithium is still considered the most important mood stabi-
liser, at least for maintaining long-term stability of BD patients (Müller-
Oerlinghausen et al. 2002), it has also been described as the most underutilised
treatment supported by solid evidence-based reports. The narrow therapeutic index
necessitating regular monitoring of therapeutic concentrations (serum levels
between 0.6 and 1.5 mEq/L) and concerns over a number of adverse effects are
responsible for the underutilisation of lithium (Ferrier et al. 2006; McKnight et al.
2012). The most frequent adverse effects found in patients treated with lithium are
(a) reduced urinary concentrating ability (polyuria, often accompanied by polydip-
sia) (Raja 2011), (b) hypothyroidism (Özerdem et al. 2014; Bauer et al. 2014) and
11 Safety and Tolerability of Mood Stabilisers 211
Table 11.1 Most common Lithium N (%)

lithium adverse events
Therapeutic agent toxicity 601 (2.3 %)
reported to the FDA from
2004 to present Tremor 525 (2.01 %)
Drug interaction 477 (1.82 %)
Confusional state 337 (1.29 %)
Completed suicide 299 (1.14 %)
Drug ineffective 264 (1.01 %)
Acute renal failure 259 (0.99 %)
Neuroleptic malignant syndrome 253 (0.97 %)
Vomiting 241 (0.92 %)
Somnolence 240 (0.92 %)
Diarrhoea 222 (0.85 %)
Nausea 209 (0.8 %)
Depression 205 (0.78 %)
Mania 203 (0.78 %)
Drug toxicity 201 (0.77 %)
Agitation 195 (0.74 %)
Dysarthria 193 (0.74 %)
Suicide attempt 190 (0.73 %)
Fatigue 186 (0.71 %)
Dehydration 185 (0.71 %)
The percentage of each adverse event as a proportion of all
adverse events is reported
hyperparathyroidism (Broome and Solorzano 2011), (c) weight gain (Torrent et al.
2008), (d) skin disorders (Jafferany 2008) and (e) risk of malformation and terato-
genic effects (Dols et al. 2013; Galbally et al. 2010).
Given that lithium salts have been used therapeutically for almost 150 years and
that John Cade reported highly successful results in ten manic patients who received
the drug as early as 1949 (Cade 1949), it is not surprising that lithium features heav-
ily in the majority of pharmacovigilance reports published in the past, when lithium
was one of the few psychotropic drug available for the treatment of BD. Present
reports, instead, are often the revival of data that was well established in the scien-
tific literature of recent years. To update reports on lithium pharmacovigilance, we
used the Food and Drug Administration (FDA) database, restricting our search from
2004 to 2014 (Table 11.1).
11.2.1 Pregnancy
Lithium has been assigned to pregnancy category D by the FDA. This category
is assigned on the basis of prospective studies conducted in the 1990s (Jacobson
et al. 1992). Despite these considerations, data regarding adverse effect of lith-
ium treatment during pregnancy are controversial, in particular due to
conflicting findings regarding teratogenicity. In fact, some authors showed high

incidences of teratogenic effects (Kallen and Tandberg 1983; Nora et al. 1974),
even in the first reports carried out during the 1970s and the 1980s amongst
newborn exposed to lithium, whereas this correlation was not reported in other
studies (Schou et al. 1973).
The Registry of Lithium Babies was founded in 1968 in order to monitor the
effects of lithium in infants who were prenatally exposed to the drug after several
case reports of cardiovascular defects, in particular Ebstein’s anomaly, amongst
children born from mothers who were given lithium during the first trimester or
throughout pregnancy (Weinstein and Goldfield 1975). In a retrospective study,
including all the 225 cases of lithium babies of the registry, the same author reported
25 malformed infants (11.1 %). This rate consisted of 18 cardiovascular defects, 6
of which were Ebstein’s anomaly (Weinstein 1980). In a second retrospective study,
Källén and Tandberg (1983) reported 10.2 % neonatal deaths, 11.9 % malforma-
tions and 6.8 % heart defects in a group of 59 babies exposed to either lithium alone
or in combination with other psychotropic drugs. As well as several retrospective
studies showing the teratogenic effect of lithium, the two studies mentioned are
based on voluntary reports and therefore associated with an overestimation of
results due to the sample selection. These differences have become more marked
through the years, mainly if we compare early reports with the more recent ones.
Indeed, the top ten adverse events are all related to childbirth (premature, abortion,
placental disorder, etc.), whilst congenital abnormalities are much rarer. Over the
last few years, we have witnessed the complete reconsideration of the literature and
the revision of the data published in the past.
In a meta-analysis by Yacobi and Ornoy (2008), the authors reviewed all the
studies on teratogenic and embryotoxic effects associated with lithium treatment
during pregnancy. In the conclusions, the authors stated that lithium therapy
throughout pregnancy did not seem to increase the general rate of major anomalies
and apparently added only a small risk of cardiovascular defects, notably Ebstein’s
anomaly. Similarly, a more recent meta-analysis concluded that the odds of lithium
exposure in cases of Ebstein’s anomaly were not significantly elevated (McKnight
et al. 2012), even though the value of this assessment was limited by the small num-
ber of cases. Although the use of lithium in the first trimester of pregnancy should
be avoided, it is fair to say that one cannot rule out completely the teratogenic poten-
tial of the drug and that the prevalence of this risk is not as high as previously
estimated.
Lithium is approved by FDA for the treatment of mania in children aged 12 years
and above and is the first therapeutic option in the acute monotherapy for mania in
this population. The early reports on paediatric lithium treatment were mainly based
on adult studies (Prien et al. 1972), and for this reason, they had no specific
relevance to the topic. Even in later reports, the results were difficult to evaluate due
to the small sample sizes or the methodological limitations.
More recently, a number of case reports, chart reviews and prospective studies
have been published providing better insight into the tolerability of lithium treat-
ment in young people. Unfortunately, the lack of definitive randomised controlled
trials prevented an adequate assessment of the real efficacy and tolerability of lith-
ium treatment in children and adolescents suffering from mania or mixed states.
For this reason, the FDA and the National Institute of Child Health and Human
Development (NICHD) sponsored the Collaborative Lithium Trials (CoLT) to
provide an evaluation of the acute and long-term effectiveness of lithium in pae-
diatric bipolarity and characterise the short and long-term safety of lithium
(Findling et al. 2008). The same authors have recently published a study evaluat-
ing lithium dose strategies and monitoring the potential treatment emergent
adverse events (TEAEs) in 41 children and adolescents suffering from bipolar I
disorder. The most commonly experienced TEAEs reported during the study
were those repeatedly reported in adult patients (vomiting, upper abdominal pain,
nausea, thirst, headache, dizziness). No patients experienced serious TEAEs
(Findling et al. 2013).
11.2.3 Adults
11.2.3.1 Kidney
Lithium-induced nephrotoxicity is a form of chronic tubulointerstitial nephropathy

known since lithium was introduced in the treatment of mood disorders and reported
from the mid-1970s (Lindop and Padfield 1975). Renal complications associated
with lithium treatment include (a) impairment of tubular function and urinary con-
centrating ability with polyuria and polydipsia and development of nephrogenic
diabetes insipidus that may become irreversible in 15 % of patients after long-term
lithium exposure and renal tubular acidosis, (b) chronic kidney disease secondary to
tubulointerstitial nephritis and (c) infrequent and relatively mild renal insufficiency
(Vestergaard and Schou 1981; Boton et al. 1987).
Early lithium reports established that lithium-induced nephropathy was usually
characterised by tubulointerstitial nephritis with minimal glomerular involvement
(Hansen 1981) and rare significant changes in glomerular filtration rate (GFR), even
in presence of a long-term lithium usage (Jensen and Rickers 1984; DePaulo et al.
1986; Schou and Vestergaard 1988). More recently, several reports have clarified
that a very high percentage of patients treated chronically with lithium have low
GFR and that GFR monitoring is frequently neglected, with a risk of progression to
end-stage renal disease (ESRD) (Bassilios et al. 2008; Janowsky et al. 2009). ESRD
was considered an unlikely event in patients taking lithium, and only three case
reports of lithium-induced ESRD were known in the world literature up to the early
nineties (Von Knorring et al. 1990; Gitlin 1993). Since then, pharmacovigilance
reports have uncovered adverse events more accurately, and from 2004 to present,
data recorded by FDA, despite small differences related to age and sex, detect rates
around 0.1 % for the most severe adverse events such as renal impairment (N = 34),
acute renal failure (N = 132), increased blood creatinine (N = 91), renal failure
(N = 60) and haemodialysis (N = 68).
The presence of renal damage induced by long-term lithium treatment in
chronically treated patients was confirmed by Markowitz et al. (2000). Lithium-
induced chronic renal disease is slowly progressive, and its rate of progression is
related to the duration of lithium administration. Regular monitoring of estimated
creatinine clearance is mandatory in long-term lithium-treated patients. A survey
of lithium-induced ESRD conducted in France adds further information on this
topic, demonstrating that lithium-related ESRD represents only 0.22 % of all
causes of ESRD in France and that the rate of progression is related to the dura-
tion of lithium administration (Presne et al. 2003). Lepkifker et al. (2004) reported
similar findings in a retrospective study showing that in long-term lithium ther-
apy, dose reduction or discontinuation of lithium resulted in stabilisation of
plasma creatinine levels and that about 20 % of long-term lithium developed renal
insufficiency.
These results are slightly different from a large nationally representative sample,
such as the Third National Health and Nutrition Examination Survey, estimating the
prevalence and distribution of chronic kidney disease in the United States. The
results of this survey show that the risk of ESRD, in lithium-treated patients, might
be increased compared with healthy controls, but the absolute risk seems to be rela-
tively low (0.53 % compared to 0.2 % of the general population) (Coresh et al.
2003). In contrast with these results, Bendz et al. (2010), reviewing the data of The
Swedish Registry for Active Treatment of Uremia in two Swedish regions (2.7 mil-
lion inhabitants), observed a substantially higher prevalence of lithium-induced
ESRD in patients on renal replacement therapy.
McKnight et al. (2012), in a recent meta-analysis, clarified that GFR impairment
secondary to lithium treatment is not clinically significant in most patients with a
reduction ranging from 0 to 5 mL/min that represents only 5 % of the minimum
normal GFR.
At the moment, two elements, strictly related, seem to be very important to pre-
vent irreversible renal damage and glomerular failure in patients treated with lith-
ium: the duration of lithium treatment and the age of treated patients (Bocchetta
et al. 2013).
Pharmacovigilance reports emphasise that renal function must be carefully
assessed in every patient starting lithium treatment. To minimise the risk of adverse
events during treatment, assessment of renal function at least twice a year is manda-
tory. This assessment must provide (1) complete 24-h urine collection; (2) glomerular
filtration rate either by 24-h creatinine clearance or estimated glomerular filtration
rate; (3) in case of a chronic kidney disease, a nephrologist should be consulted
before lithium treatment as long as the creatinine clearance is >40 mL/min and (4)
the decision should be taken also on the basis of the patient’s age and the duration
of lithium treatment.
11.2.3.2 Thyroid Gland
Long-term treatment with lithium results in its accumulation in the thyroid with
many effects on the physiology of the gland. The pathogenetic mechanism of
lithium-induced hypothyroidism is manifold; lithium acts through five different
mechanisms: (1) inhibition of thyroidal iodine uptake, (2) inhibition of iodotyrosine
coupling, (3) changes of the thyroglobulin structure, (4) inhibition of thyroid hor-
mone (thyroxine) secretion and (5) increase of TSH levels as a result of reduced
availability of thyroxine (Berens et al. 1970; Burrow et al. 1971). Inhibition of thy-
roid hormone secretion associated with high rates of hypothyroidism and thyrotoxi-
cosis is the result of these effects on the gland function (Bocchetta et al. 2001;
Barclay et al. 1994).
Adverse events submitted to the FDA from 2004 to present have revealed 54
cases of hypothyroidism in adult patients treated with lithium and, although minor
differences related to age and sex, rates are around 0.29 % of the total FDA reports.
Hypothyroidism is the most common thyroid disorder caused by lithium treat-
ment. It is very difficult to evaluate its prevalence because records range from 3.3 to
35.4 %. Eight case-control studies reveal a prevalence of clinical and subclinical
hypothyroidism of 9.2 % in patients treated with lithium compared with a preva-
lence in the general population comprised between 0.5 and 1 %. The risk of hypo-
thyroidism increases about six times in patients taking lithium (OR = 5.78)
(McKnight et al. 2012). The main risk factors for the onset of hypothyroidism in
lithium-treated patients are female gender, age between 40 and 60 years, a personal
or family history of thyroid disorders and the presence of autoantibodies (Malhi
et al. 2012). The risk amongst women is greater than in men and in the general
population and is related to the age of patients and the duration of treatment
(Grandjean and Aubry 2009). However, it is important to stress that, irrespective of
lithium treatment, patients with mood disorders show higher rates of thyroid abnor-
malities (hypothyroidism and hyperthyroidism) than the general population
(Chakrabarti 2011).
Goitre is a clinical finding associated with lithium therapy, probably linked to the
inhibition of thyroid hormone synthesis and release, determining an increase of
TSH and a final thyroid enlargement. The prevalence of goitre is highly variable,
ranging from 3.6 to 51 %. This variability is probably due to the presence of various
geographic risk factors, specifically the reduced availability of iodine, the different
duration of the exposure to lithium and different diagnostic methods (Kibirige et al.
2013). The presence of hypothyroidism or goitre is not a contraindication to lithium
treatment; therefore, patients successfully treated should continue treatment, even
in the presence of hypothyroidism, and compensate with a hormone replacement
treatment.
Case reports of hyperthyroidism and thyrotoxicosis have been described in the
literature since the 1970s (Rosser 1976). More recent studies clarified that it is a rare
condition whose incidence is comparable to that of the general population
(Vanderpump et al. 1995). Thyrotoxicosis occurs in the early stages of treatment
and at a young age, especially in women.
11.2.3.3 Parathyroid Gland
Hyperparathormonemia and hypercalcaemia are unrecognised and underappreci-

ated adverse effects of lithium treatment despite the evidence that in healthy volun-
teers a single dose of lithium (600 mg) is enough to induce a transient, and
statistically significant, rise in the serum PTH (Seely et al. 1989) and a prevalence
ranging from 6.3 to 50 % (Livingstone and Rampes 2006).
Health Canada has reviewed the available evidence and scientific literature con-
cerning the association between lithium treatment and hypercalcaemia associated
with hyperparathyroidism. Following this evaluation, Health Canada advised health
professionals about the risk of hypercalcaemia/hyperparathyroidism associated
with lithium treatment and the need of considering calcium and parathormone blood
levels before starting lithium treatment and of repeating this evaluation every
6 months in order to reduce the risk of hypercalcaemia (www.healthycanadians.
gc.ca).
Adverse events submitted to the FDA from 2004 to present have revealed 29
cases of hypothyroidism in adult patients treated with lithium, and, despite minor
differences related to age and sex, rates are around 0.24 % of the total FDA reports
(Table 11.1).
Saunders et al. (2009) reviewed the effect of lithium exposure on parathyroid cell
function confirming that lithium causes hypercalcaemia and isolated hyperparathor-
monemia and the need of a screening of patients on chronic lithium therapy for
hypercalcaemia. More recently, a meta-analysis identified 60 studies (14 case-
control studies, 36 case reports, 6 cross-sectional studies) and a 10 % increase of
calcium and parathormone (PTH) levels in patients treated with lithium. This
adverse effect remains unrecognised by most psychiatrists with an underestimation
of the long-term effects of hypercalcaemia and hyperparathyroidism (McKnight
et al. 2012).
This evidence indicates that the assessment of serum calcium before and during
lithium treatment is mandatory. Currently, only the clinical guideline of The
International Society for Bipolar Disorders (ISBD) suggests this assessment before
starting lithium treatment and, in the absence of clinical elements suggestive of
impaired parathyroid function, a re-evaluation after 6 months and then annually
(Pacchiarotti et al. 2013).
11.2.3.4 Skin
Potential skin changes associated with lithium treatment are generally limited, and
the evidence supporting this association is restricted to descriptions of several case
reports, retrospective studies and few case-control studies. There are only two recent
randomised controlled trials comparing lithium with lamotrigine and placebo for
18 months. The combined analysis of the results of these two studies showed no
significant difference in the prevalence of cutaneous adverse effects between
patients given lithium and those given placebo (Goodwin et al. 2004). It is therefore
not clear whether lithium exposes them to a greater risk of developing cutaneous
adverse effects in general. This statement is confirmed by the difficulty in assessing
the real prevalence rates of this adverse effect. The few controlled studies on preva-
lence seem to suggest an increase of these adverse effects in patients treated with
lithium, compared to the general population, ranging from 13.6 to 34 % in the study
by Sarantidis et al. (1983) and from 25 to 45 % in the study by Chan et al. (2000).
The situation is slightly different for psoriasis since the presence of this skin
disorder in lithium-treated patients is reported in several cases as de novo onset of
psoriasis or as a marked worsening of a previously diagnosed disease (actually,
reports are greater for acute exacerbation of a known disease).
There are only three case-control studies evaluating the prevalence of psoriasis in
patients treated with lithium. The above-mentioned studies found a prevalence of
2.2 % in patients compared to 0 % in controls (Sarantidis and Waters 1983) and 6 %
in patients compared to 0 % in controls (Chan et al. 2000). The third one is a large-
scale epidemiological case-control study that found a small but significant increase
in the risk of psoriasis in lithium-treated patients (Brauchli et al. 2009).
11.2.3.5 Weight Gain
The greater part of pharmacovigilance reports on lithium treatment-associated

weight gain has been published in the 1970s and 1980s (Rockwell et al. 1983), and
present reports are often the continuation of data well established in the scientific
literature of recent years. Reports on adverse events submitted to the FDA from
2004 to present have revealed 93 cases of weight gain representing, despite minor
differences related to age and sex, the 0.66 % of the total FDA reports.
11.2.3.6 Cognition
Bipolar patients display a persistent cognitive impairment in the different stages of

the disease including euthymia. This impairment, evident in a range of neuropsy-
chological tests including memory and executive functioning (Frangou et al. 2005),
is present in unaffected first-degree relatives and is thought to be related to illness
severity, specific neurodevelopmental features, medical co-morbidity and patient’s
lifestyle (Bourne et al. 2013).
Data in this area of research are highly controversial mainly due to the presence
of numerous methodological problems which reduce their reliability. Lithium treat-
ment is associated with mild impairment in psychomotor speed, verbal memory and
psychomotor speed functioning without a clear positive effect on cognition (Pachet
and Wisniewski 2003). In a recent study, Wingo et al. (2009) reviewed the results of
12 studies involving 276 lithium-treated bipolar patients. Long-term lithium treat-
ment was associated with impairment in immediate verbal learning and memory
and creative psychomotor performance with no evidence of cognitive
improvements.
11.2.3.7 Elderly
Concerns about lithium adverse effects in older patients have led to both declining
rates of lithium use and questions regarding the most useful approach to the use of
lithium in this population.
Despite the lack of randomised placebo-controlled trials, it is assumed that lith-
ium is as effective in the elderly as in the younger population for prophylaxis of
affective disorders and for resistant unipolar depression (Bech 2006), but several
concerns about neurotoxicity have led to questions about the effectiveness and
safety of lithium in older bipolar patients.
The decrease in total body water and the decline of glomerular filtration rate are
common amongst older persons, and in bipolar patients treated with lithium, this
can result in a decrease in lithium clearance and increased serum level (Slater et al.
1984; Sproule et al. 2000). In a 2-year study of unipolar and bipolar out-patients
(21–78 years) on long-term lithium treatment, Murray et al. (1983) found polydip-
sia/polyuria in 44 % of patients and hand tremor in 29 % of them. The prevalence
and severity of the tremor tended to increase with age, but polydipsia/polyuria
didn’t. More recently, van Melick et al. (2013) evaluated 759 patients aged 40 or
older and treated with lithium with at least 2 years follow-up in a retrospective study
and assumed that age was not a determinant of serum lithium concentration instabil-
ity and, above all, was not a reason not to initiate or to discontinue lithium therapy.
Lithium is a drug that continues to have a critical role in the treatment of bipolar
disorder in the elderly. It becomes clear that, in the elderly more than in the young
bipolar patients, lithium requires a careful evaluation in order to prevent adverse
effects or toxicity. Lithium treatment in the elderly may be appropriate only after a
clear evaluation of benefits and risks in each individual patient, and if the patient is
monitored correctly, it is possible to avoid commonly reported adverse effects that
can occur even at therapeutic dosages.
Guidelines for lithium concentrations in geriatric bipolar population are based on
limited evidence, and a recent study recommends a low concentration range (0.5–
0.6 mmol/L) for patients of 50 years and over (Wijeratne and Draper 2011).
11.3 Safety of Mood Stabilisers Other Than Lithium:

Antiepileptic Drugs (AEDs)
In different studies, the proportion of patients with side effects from AED therapy
ranged from less than 10 % to over 70 % depending on ascertainment methods, char-
acteristics of the patients, AED dosage and duration of follow-up (Perucca et al. 2000).
The tolerability profiles of AEDs differ substantially from one drug to another,
and it is not straightforward to establish which drug has the best one. However, the
safety profile is often a determining factor in drug selection because efficacy rates
shown by most AEDs are similar (Perucca and Meador 2005). Clinical trials have
provided inconclusive information to evaluate the comparative risk-benefit ratio.
In addition, there is a lack of systematic pharmaco-epidemiological studies

investigating adverse drug reactions (ADRs) to AEDs, which makes it difficult to
accurately assess the incidence of anticonvulsant-related ADRs (Acharya et al.
2005; Wong and Lhatoo 2000). Moreover, clinical exposure to some of the newer
drugs is still relatively limited, and experience shows that it may take many years for
important adverse effects to be discovered, especially when they are rare (Perucca
et al. 2000). We will therefore review the safety profile of individual AEDs in a
spontaneous reporting database over the last 10 years with a particular attention to
carbamazepine, oxcarbazepine, valproic acid and lamotrigine.
11.3.1 Pregnancy
A report by Tica et al. (2013) highlighted the possibility that phenobarbital (PH)/
carbamazepine (CBZ) therapy during foetal organogenesis could induce sirenome-
lia by a synergistic teratogenic effect and supported the recommendation to use only
one drug in pregnant epileptic or bipolar women. At birth, the newborn, delivered
by an epileptic woman after 37 weeks of gestation, weighed 2.2 kg and presented
with sirenomelia type II, with some of its “classic” features: oligohydramnios and
absence of kidneys, bladder, rectum, uterus and a single umbilical artery. Some
other “particularities” included the absence of Potter’s face and no significant
cardio-pulmonary abnormalities. The authors postulated that combined therapy
with PH and CBZ (both strong enzyme inductors, especially PH) had potentiated
their teratogenicity, by producing supplementary quantities of epoxides and/or other
oxides, which accumulated in the foetal tissues who received in the first 4 months of
pregnancy PH (0.1 g/day) and CBZ (0.4 g/day), followed only by PH 0.1 g/day,
until delivery.
Another report presented an infant born with renal and cardiac malformations
who developed a withdrawal syndrome and hyponatraemia following in utero expo-
sure to oxcarbazepine. The infant was born at 35 weeks’ gestation by urgent caesar-
ean section to a mother in status epilepticus who had been treated with oxcarbazepine
throughout her pregnancy. Evaluation for congenital anomalies identified mild aor-
tic stenosis, a bicuspid aortic valve, patent foramen ovale, patent ductus arteriosus
and severe left hydronephrosis due to left ureteropelvic junction stenosis. On the
third day of life, the infant developed clinical signs of a withdrawal syndrome,
which peaked on day 7 and resolved by day 12. Transient hyponatraemia resolved
by day 8 of life. Follow-up showed normal development at 15 months (Rolnitsky
et al. 2013).
The prevalence of neurodevelopmental disorders in children prenatally exposed
to antiepileptic drugs was studied in a prospective cohort of women with epilepsy
and a control group of women without epilepsy. The children of this cohort were
followed longitudinally until 6 years of age (N = 415), and the analysis revealed an
increase in risk for children exposed to monotherapy sodium valproate and in those
exposed to polytherapy with sodium valproate compared to control children (4/214;
1.87 %). Autistic spectrum disorder was the most frequent diagnosis. No significant
increase was found amongst children exposed to carbamazepine (1/50) or lamotrig-
ine (2/30) (Bromley et al. 2013).
Valproate is associated with polycystic ovary syndrome as well as congenital
malformations and developmental delays of infants who were prenatally exposed.
In a study by Wisner et al. (2011), using New York State Medicaid Claims for
Persons with Psychiatric Disorders, the authors concluded that over 20 % of
childbearing-aged women receiving mood stabilisers were treated with valproate.
In women, according to FDA-reported side effects, the use of valproic acid dur-
ing pregnancy led to spontaneous abortion in a different percentage, depending on
the age of patients (0.1 %, 0.02 % and 0.04 %, respectively, for women’s age of
10–19, 20–29 and 30–39 years). Intrauterine deaths were reported, instead, only in
few cases (0.03 %). Carbamazepine and oxcarbazepine use during pregnancy was
associated to spontaneous abortion only in 0.12 % and 0.16 %, respectively, of all
treated patients. Lamotrigine, on the other hand, was reported to FDA to induce
spontaneous abortion in 0.5 % of cases, with a peak of 0.12 % in women ageing
from 20 to 29 years.
Cutaneous adverse drug reactions (CADRs) are the most prevalent ADRs in hospi-
talised children, with an estimated rate of 2–3 % (Ross et al. 2007). An analysis of
reports from the Canadian Pharmacogenomics Network for Drug Safety (CPNDS)
included 326 CADR cases of which 214 (65.6 %) were severe and 112 (34.4 %)
non-severe. Overall, carbamazepine (N = 17, 4.9 %) and lamotrigine (N = 13, 3.7 %)
accounted for almost 9 % of all suspected medications (Castro-Pastrana et al. 2011).
A case-control study of severe cutaneous drug reactions (SCDR) with carbam-
azepine was reported by Chong et al. (2014). In recruited patients, HLA-B*1502
positivity increased the odds of carbamazepine-induced SCDR in children of
Chinese and Malay ethnicity and they occurred within 2 weeks and at low doses.
Stevens-Johnson Syndrome (SJS), on the other hand, was induced by oxcarbaze-
pine, and HLA genotyping showed a HLA-B15 variant in this patient (HLA-
B*1518/B*4001) (Lin et al. 2009). A 4-year-old girl with a life-threatening clinical
course of drug rash with eosinophilia and systemic symptoms syndrome (DRESS)
with massive pulmonary involvement was also reported by Irga et al. (2013).
The use of combined antiepileptic drugs can cause toxicity by affecting the clear-
ance of the drugs, especially in children. A case with SJS triggered by the combina-
tion of clobazam, lamotrigine and valproic acid treatment was reported in a
4-year-old boy admitted to the hospital with a 3-day history of fever, oral mucosa
ulcerations and skin lesions. The patient had been under the treatment of valproic
acid (900 mg/day) for 3 years with the diagnosis of epilepsy. Because of the poor
control of the seizures, lamotrigine (75 mg/day) had been added to the treatment
1 month before and clobazam (20 mg/day) 10 days before. Weintraub et al. (2005)
reported that valproic acid decreased lamotrigine clearance by approximately 60 %

in a study with 570 patients. It has been reported that valproic acid may interfere
with the metabolism of lamotrigine by inhibiting glucuronides, leading to increased
blood levels of the drug, or resulting in accumulation of toxic metabolites of the
drug. New AEDs (clobazam, etc.) are reported not to affect the clearance of
lamotrigine significantly, but the skin lesions of SJS in the above patient appeared
when clobazam was added to valproic acid and lamotrigine treatment. Therefore, it
could be concluded that clobazam can affect the clearance of combined drugs.
Adverse reactions have been reported in children co-treated with lamotrigine and
valproate (Kocak et al. 2007; Levi et al. 2009). Sixty-three cases of interaction
between lamotrigine and other drugs have been reported to FDA from 2004 to
present.
In 2011, the US FDA informed the public that children born to mothers who take
the anti-seizure medication sodium valproate or related products (valproic acid and
divalproex sodium) during pregnancy have an increased risk of lower cognitive test
scores than children exposed to other anti-seizure medications during pregnancy.
This conclusion was based on the results of epidemiologic studies. The largest of
these studies is a prospective cohort study conducted in the United States and
United Kingdom (Meador et al. 2009). They found that children with prenatal expo-
sure to valproate throughout pregnancy had lower Differential Ability Scale (DAS)
scores at age 3 than children with prenatal exposure to the other evaluated antiepi-
leptic drug monotherapy treatments: lamotrigine, carbamazepine and phenytoin.
Although all of the available studies have methodological limitations, the weight of
the evidence supports the conclusion that valproate exposure in utero causes subse-
quent adverse effects on cognitive development in offspring (Gaily et al. 2004;
Adab et al. 2004).
The FDA reported that for AEDs, the most common side effects in children and
adolescents (0–19 years of age) were represented by the following:
• Convulsions, exposure during pregnancy, foetal anticonvulsant syndrome, dys-
morphism for valproic acid
• Atrial septal defect, aggressive behaviour, psychomotor hyperactivity, convul-
sions, pyrexia and rashes for lamotrigine
• Hypertelorism of orbit, micrognathia, drug rashes with eosinophilia and sys-
temic symptoms, pyrexia, convulsions for carbamazepine
11.3.3 Adults
In a recent review, Dols et al. (2013) described the prevalence of neurological, gas-
trointestinal, metabolic, thyroid, dermatological, nephrogenic, cognitive, sexual,
haematological, hepatogenic and teratogenic side effects of lithium, valproate, car-
bamazepine and lamotrigine and discussed their clinical management. The most
common side effects reported to the FDA since 2004 are listed in Table 11.2.
Table 11.2 Most common carbamazepine, valproate and lamotrigine adverse events reported to
the FDA from 2004 to present
Carbamazepine Valproate Lamotrigine
Convulsions 642 (2.15 %) 455 (2.7 %) 993 (2.87 %)
Drug exposure during pregnancy 315 (1.06 %) 226 (1.34 %) 778 (2.25 %)
Rash 236 (0.79 %) 55 (0.33 %) 501 (1.45 %)
Completed suicide 304 (1.02 %) – 480 (1.39 %)
Drug toxicity 234 (0.78 %) 87 (0.52 %) 466 (1.35 %)
Drug interaction 417 (1.4 %) 395 (2.34 %) 384 (1.11 %)
Pyrexia 397 (1.33 %) 163 (0.97 %) 373 (1.08 %)
Dizziness 223 (0.75 %) 65 (0.39 %) 311 (0.9 %)
Somnolence 288 (0.97 %) 195 (1.16 %) 278 (0.8 %)
Vomiting 221 (0.74 %) 133 (0.79 %) 270 (0.78 %)
Maternal drugs affecting foetus 98 (0.33 %) 46 (0.27 %) 257 (0.74 %)
Stevens-Johnson syndrome 221 (0.74 %) 56 (0.33 %) 230 (0.66 %)
Spontaneous abortion – 58 (0.34 %) 179 (0.52 %)
Agitation 81 (0.27 %) 79 (0.47 %) 177 (0.51 %)
Confusional state 178 (0.6 %) 131 (0.78 %) 159 (0.46 %)
White blood cell count increased – 43 (0.25 %) 128 (0.37 %)
Premature baby – – 74 (0.21 %)
Thrombocytopenia 80 (0.27 %) 127 (0.75 %) –
For each event, the number of involved patients is indicated along with the percentage of the
adverse events as a proportion of all reactions reported for the drug
Moreover, Gau et al. (2010) indicated that lithium, carbamazepine and valproate
may dose dependently increase the risk of hypothyroidism with the increasing num-
ber of mood stabilisers used (the risk of hypothyroidism was more prominent when
the combination included lithium and valproate).
The study by Gau et al. (2008) investigated the association between two mood
stabilisers (carbamazepine and valproate) and other medications (including other
anticonvulsants) and the risks of erythema multiforme (EM), SJS and toxic epider-
mal necrolysis (TEN) amongst patients with BD. Results showed that carbamaze-
pine and valproate use significantly predicted EM, SJS or TEN. Other significant
predictors for EM, SJS or TEN included other anticonvulsants (phenytoin, pheno-
barbital and lamotrigine). The most predictive exposures were carbamazepine, val-
proate, other anticonvulsants and acetaminophen. They also found that the
combination of carbamazepine and acetaminophen further increased the risk for the
occurrence of EM, SJS or TEN and that no interaction effect of age and sex was
evident.
Matsuda et al. (2013) reported a case of drug-induced hypersensitivity syndrome
(DIHS)/drug reaction with eosinophilia and systemic symptoms (DRESS) due to
carbamazepine and associated with maculopapular eruptions and systemic skeletal
muscle involvement. Sharma et al. (2013), on the other hand, reported the case of a
patient with seizure who developed SJS and neuroleptic malignant syndrome
(NMS) following administration of carbamazepine.
Hydzik et al. (2011) published the case of acute poisoning with carbamazepine
and quetiapine, which resulted in cardiotoxic effects in the form of arrhythmias and
conduction disorders of the heart. These symptoms disappeared spontaneously after
resolution of the poisoning.
He et al. (2012) reported that the incidence of oxcarbazepine (OXC)-induced
cutaneous drug reactions (cADRs) was low, and no severe reactions occurred
although they observed that patients with a history of allergy were more susceptible
to OXC-cADRs. Moreover, they did not find a significant association between
HLA-B*1502 and OXC-maculopapular eruptions.
In a nationwide cohort of bipolar patients, Yang et al. (2013) highlighted that
some drug combinations were associated with a dose-dependent increase in the risk
of pneumonia. In particular, olanzapine plus carbamazepine had the highest risk,
followed by clozapine and valproic acid, but lithium had a dose-dependent protec-
tive effect.
The use and safety profile of antiepileptic drugs in Italy was evaluated only for
those associated with at least 30 reports by Iorio et al. (2007). Skin reactions were
the most frequently reported ADRs, followed by haematological, general condition,
hepatic and neurological and gastrointestinal adverse reactions. Lamotrigine and
carbamazepine had the highest percentage of skin reactions (67 % and 60 %, respec-
tively). Many haematological reactions were reported for each AED, but the highest
percentage was related to valproic acid (25 %) which was associated also to the
highest percentage of hepatic reactions (20 %).
SJS associated with single high dose of lamotrigine was reported in a 23-year-
old female patient with idiopathic epilepsy and previously taking carbamazepine,
valproic acid and lamotrigine until 1 week prior to referral. Following consultations
with a range of clinicians, the patient was diagnosed with SJS related to lamotrigine
based on her history and physical findings and on consideration of current consen-
sus definitions of this condition (Kocak et al. 2007).
In addition, SJS after lamotrigine treatment was also reported in a 24-year-old
man on lamotrigine (25 mg every other day for the first 2 weeks with the dose
increased to 25 mg/day for the subsequent 2 weeks) for tonic-clonic seizures not
adequately controlled with valproate (600 mg/day). The patient had been receiving
valproate for 3 years without any other medications, whilst in the fourth week of
titration therapy, he mistakenly took lamotrigine 200 mg at one time rather than
25 mg. A few hours later, high-grade fever (40 °C) and painful oral ulcerations
involving much of the oropharynx developed (Famularo et al. 2005). Bicknell et al.
(2012) reported a case of drug reaction with eosinophilia and systemic symptoms
apparently precipitated by the associated use of lamotrigine and cyclobenzaprine.
Side effects are also related to non-adherence, but at a low level (Baldessarini
et al. 2008; Jonsdottir et al. 2012). More than one-half of BD patients either discon-
tinue pharmacotherapy or use it irregularly. Although rates of non-adherence do not
necessarily differ between mood-stabilising medications, the predictors for non-
adherence do. Moreover, adherence to one medication does not guarantee adher-
ence to another, nor does adherence at one time-point ensure later adherence.
Attitudes towards treatment affect adherence to medications as well as psychosocial
treatments and should be repeatedly monitored (Arvilommi et al. 2014). Some side
effects have more impact on adherence than others, and in order of importance,
weight gain, cognitive impairment and severity of depressive symptoms (as an out-
come of medication) were most associated with non-adherence to medication
(Johnson et al. 2007). Experts reported that sedative side effects of medications also
contributed to non-adherence (Velligan et al. 2010).
Somnolence, sedation, fatigue, pyrexia and drug rashes were all reported to the
FDA for carbamazepine (age ranging from 20 to 59 years); convulsions, vomiting,
decreased platelet count, breast abscesses and rashes were those reported for
lamotrigine whilst pyrexia, somnolence, drug interactions, thrombocytopenia and
tremor for valproic acid.
11.3.4 Elderly
Descriptions of geriatric patients with BD tend to include an overview of cognitive

impairment. However, the literature regarding cognitive test performance in this
population is very limited.
Cognitive impairment in elderly (defined by the commonly accepted criterion of
age ≥60 years) bipolar patients persists during euthymic state. At the moment, the
aetiology of cognitive impairment is not well understood, although it has been asso-
ciated with lithium and other mood stabilisers. The dosage of medications has been
reported to influence the cognition of treated patients significantly, especially when
affected by cardiovascular co-morbidities (Schouws et al. 2010). Other studies, on
the other hand, have found no association between exposure to mood stabilisers
(lamotrigine, valproate, lithium, carbamazepine all used at clinical therapeutic dos-
age) and any clinical or cognitive variables in the elderly (Martino et al. 2008).
The use of AEDs (mainly, valproic acid, carbamazepine and oxcarbazepine) in
the management of behavioural and psychological symptoms in Alzheimer’s disease
and related dementias (ADRD) was reported by Dutcher et al. (2014) in a study
where these medications were used in 3 % (N = 571) of the all dementia patients.
They observed a functional decline in activity of daily living (ADL) scores and a
negative effect on cognition over time in most patients. In particular, female users of
mood stabilisers declined most quickly, followed by male non-users, female non-
users and male users thus suggesting a faster ADL decline in women but not in men
when these drugs were used.
The treatment of behavioural and psychological symptoms in patients with
dementia has included several different mood stabilisers. Amongst these medica-
tions, only carbamazepine demonstrated its efficacy in behavioural and psycho-
logical symptoms of dementia (BPSD) in controlled studies (Olin et al. 2001),
but significant adverse events were reported (sedation, hyponatraemia, cardiac
toxicity), and multiple drug-drug interactions occurred probably because this
drug is a strong enzymatic inducer. Valproic acid showed some interesting results
in BPSD within a large number of open studies and case reports. However,
amongst the five controlled studies that have been published (Pinheiro 2008),
none confirmed its efficacy on these symptoms. Regarding its tolerability in the
geriatric population, no notable major side effect was reported (haematologic
and hepatic effects were not more frequent than in the general population),
except for a possible over-sedation. Moreover, it appears that valproic acid could
have neuroprotective effects, even if the contrary has also been observed. More
studies need to be (and are being) conducted, notably on the potentially prophy-
lactic effect of valproic acid in BPSD. Lamotrigine, which may potentially
induce severe cutaneous side effects when administered with valproic acid, has
shown its efficacy in BD, and two recent case reports seem to indicate some
clinical relevance to BPSD. Oxcarbazepine, theoretically, could be an alternative
to carbamazepine, which is, as previously mentioned, the only anticonvulsant
proved to be of clinical benefit in BPSD. However, no clinical study has been
published so far to support this hypothesis. This drug, although inducing severe
and more frequent hyponatraemia than carbamazepine, is better tolerated than
carbamazepine. Polypharmacotherapy and concomitant psychotropic drugs are
also reported as risk factors for falls in long-term care setting for elderly patients
(Baranzini et al. 2009; Landi et al. 2005).
Carbamazepine use was also associated with a nearly tenfold increase in severe
cutaneous drug reactions in Korean elderly patients (Kim et al. 2013). This associa-
tion was consistently high with SCARs in patients who received carbamazepine for
neuropathic pain. On the other hand, other studies reported that elderly patients had
a lower incidence of reported allergic skin reactions with mood-stabilising anticon-
vulsants (carbamazepine, lamotrigine and valproic acid), and the risk for other
ADRs decreases significantly with age, in particular extrapyramidal motor system
(EPMS) symptoms, galactorrhoea, weight gain and increased liver enzymes. In con-
trast, the risk of developing delirium increased with age, and the risk of developing
oedema showed a corresponding trend (Greil et al. 2013).
Koda et al. (2012) reported the case of a 68-year-old woman with Alzheimer’s
disease developing renal dysfunction after starting carbamazepine for epilepsy: the
autopsy found an acute tubulointerstitial nephritis with multiple organ involvement,
including fatal adrenalitis.
Carbamazepine may have negative chronotropic and dromotropic effects on the
cardiac conduction system (Ide and Kamijo 2007). Koutsampasopoulos et al. (2014)
reported the case of an 82-year-old woman who was admitted in the hospital follow-
ing a syncopal episode at home and that developed a cardiac syncope due to atrial
tachycardia combined with complete atrioventricular block as a consequence of
carbamazepine administration for trigeminal neuralgia.
Mood disorders substantially increase the risk of cardiovascular diseases, though
the mechanisms are unclear. Chronicity of mood symptoms contribute to vasculopa-
thy in a dose-dependent fashion. Fiedorowicz et al. (2012) reported that patients
with more manic/hypomanic symptoms had poorer vascular endothelial function.
Moreover, antipsychotic/mood stabiliser and antidepressant exposure were not
associated with flow-mediated dilation (FMD) or nitroglycerin-mediated vasodila-

tion. Cumulative exposure to antipsychotics and mood stabilisers (in particular, val-
proic acid derivatives, lithium, carbamazepine and lamotrigine) was instead
associated with vascular stiffness as shown by elevated aortic systolic augmentation
pressure and total aortic systolic blood pressure. Valproic acid derivative exposure
was also associated with hypertriglyceridaemia, insulin resistance and low HDL
cholesterol.
The adverse events most commonly reported to the FDA for carbamazepine,
from 2004 to present, were mainly those induced by drug interactions, hypona-
traemia and pyrexia in patients older than 60 years lamotrigine induced decreased
haemoglobin level, convulsions and myelodysplastic syndrome more frequently,
whilst valproic acid was associated with confusional state, convulsions and drug
interactions.
11.4 Conclusions
Several combinations of mood stabilisers appear to be safe and effective.

Unfortunately, there is a conspicuous lack of data from randomised controlled trials
regarding mood stabiliser combinations despite their widespread use, and only lim-
ited data from open-label studies are available for many such combinations.
Combination therapy introduces considerations such as drug interactions and addi-
tional side effects. The interactions of combinations of mood stabilisers are some-
times complex, often very useful and potentially dangerous.
As discussed previously, combination therapy may also decrease compliance.
One general rule that may reduce the risks of toxic drug interactions is to add medi-
cation to the patient’s current regimen in modest doses and increase the dose slowly.
This strategy of using low doses is particularly relevant for the elderly bipolar
patients.
Currently, the most useful mood stabiliser combinations are the mixtures of anti-
convulsants and lithium, particularly valproate plus lithium. Carbamazepine,
lamotrigine and gabapentin have also been added to lithium and appear to be well
tolerated. Moreover, combinations of anticonvulsants have been used successfully,
although pharmacokinetic interactions may complicate this strategy, particularly
when valproate and carbamazepine or valproate and lamotrigine combinations are
used. Combinations of lithium or anticonvulsant mood stabilisers with standard or
atypical antipsychotic agents and benzodiazepines have also been used relatively
successfully. In particular, several second-generation antipsychotics have been
approved for the management of BD (olanzapine, risperidone, quetiapine, aripipra-
zole, ziprasidone, asenapine, paliperidone).
Future research should clarify the mechanisms of action of the most widely used/
mood stabilisers as these are currently not well understood. This will give clinicians
the possibility of using combinations of different mood stabilisers more rationally
and improve clinical outcomes in treated patients.
References
Acharya NV, Pickering RM, Wilton LW et al (2005) The safety and effectiveness of newer
antiepileptics: a comparative postmarketing cohort study. J Clin Pharmacol 45:385–393
Adab N, Kini U, Vinten J et al (2004) The longer-term outcome of children born to mothers with
epilepsy. J Neurol Neurosurg Psychiatry 75:1575–1583
Arvilommi P, Suominen K, Mantere O et al (2014) Predictors of adherence to psychopharmaco-
logical and psychosocial treatment in bipolar I or II disorders – an 18-month prospective study.
J Affect Disord 155:110–117
Baldessarini RJ, Tondo L (2008) Lithium and suicidal risk. Bipolar Disord 10:114–115
Baldessarini RJ, Perry R, Pike J (2008) Factors associated with treatment non-adherence among
US bipolar disorder patients. Hum Psychopharmacol Clin Exp 23:95–105
Baranzini F, Poloni N, Diurni M et al (2009) Polypharmacy and psychotropic drugs as risk factors for
falls in long-term care setting for elderly patients in Lombardy. Recenti Prog Med 100:9–16
Barclay ML, Brownlie BE, Turner JG et al (1994) Lithium associated thyrotoxicosis: a report of
14 cases, with statistical analysis of incidence. Clin Endocrinol 40:759–764
Bassilios N, Martel P, Godard V et al (2008) Monitoring of glomerular filtration rate in lithium-
treated outpatients—an ambulatory laboratory database surveillance. Nephrol Dial Transplant
23:562–565
Bauer M, Glenn T, Pilhatsch M et al (2014) Gender differences in thyroid system function:
relevance to bipolar disorder and its treatment. Bipolar Disord 16(1):58–71
Bech P (2006) The full story of lithium (1918–2005). A tribute to Mogens Schou. Psychother
Psychosom 75:265–269
Bendz H, Schön S, Attman PO et al (2010) Renal failure occurs in chronic lithium treatment but is
uncommon. Kidney Int 77(3):219–224, Epub 2009 Nov 25
Berens SC, Wolff J, Murphy DL (1970) Lithium concentration by the thyroid. Endocrinology
87(5):1085–1087
Bicknell LT, Sanches M, Schatte DJ (2012) Erythema multiforme vs. DRESS syndrome associated
with the combined use of lamotrigine and cyclobenzaprine: a case report. Int J Psychiatry Med
44(4):305–314
Bocchetta A, Mossa P, Velluzzi F et al (2001) Ten-year follow-up of thyroid function in lithium
patients. J Clin Psychopharmacol 21:594–598
Bocchetta A, Ardau R, Carta P et al (2013) Duration of lithium treatment is a risk factor for
reduced glomerular function: a cross-sectional study. BMC Med 11:33
Boton R, Gaviria M, Batlle DC (1987) Prevalence, pathogenesis, and treatment of renal dysfunc-
tion associated with chronic lithium therapy. Am J Kidney Dis 10(5):329–345
Bourne C, Aydemir O, Balanza-Martinez V et al (2013) Neuropsychological testing of cognitive
impairment in euthymic bipolar disorder: an individual patient data meta-analysis. Acta
Psychiatr Scand 128(3):149–162
Brauchli YB, Jick SS, Curtin F et al (2009) Lithium, antipsychotics, and risk of psoriasis. J Clin
Psychopharmacol 29(2):134–140. doi:10.1097/JCP.0b013e31819a4b7c
Bromley RL, Mawer GE, Briggs M et al (2013) The prevalence of neurodevelopmental disorders
in children prenatally exposed to antiepileptic drugs. J Neurol Neurosurg Psychiatry
84(6):637–643
Broome JT, Solorzano CC (2011) Lithium use and primary hyperparathyroidism. Endocrinology
17(Suppl 1):31–35
Burrow GN, Burke WR, Himmelhoch JM et al (1971) Effect of lithium on thyroid function. J Clin
Endocrinol Metab 32(5):647–652
Cade JF (1949) Lithium salts in the treatment of psychotic excitement. Med J Aust
2(10):349–352
Castro-Pastrana LI, Ghannadan R, Rieder MJ et al (2011) Cutaneous adverse drug reactions in
children: an analysis of reports from the Canadian pharmacogenomics network for drug safety
(CPNDS). J Popul Ther Clin Pharmacol 18(1):e106–e120
Chakrabarti S (2011) Thyroid functions and bipolar affective disorder. J Thyroid Res
2011:306–367
Chan HL, Wing Y, Su R et al (2000) A control study of the cutaneous side effects of chronic lith-
ium therapy. J Affect Disord 57:107–113
Chong KW, Chan DW, Cheung YB et al (2014) Association of carbamazepine-induced severe
cutaneous drug reactions and HLA-B*1502 allele status, and dose and treatment duration in
paediatric neurology patients in Singapore. Arch Dis Child 99(6):581–584
Coresh J, Astor BC, Greene T et al (2003) Prevalence of chronic kidney disease and decreased
kidney function in the adult US population: Third National Health and Nutrition Examination
Survey. Am J Kidney Dis 41:1–12
De Paulo J, Correa E, Sapir D (1986) Renal function and lithium: a longitudinal study. Am J
Dols A, Sienaert P, van Gerven H et al (2013) The prevalence and management of side effects of
lithium and anticonvulsants as mood stabilizers in bipolar disorder from a clinical perspective:
a review. Int Clin Psychopharmacol 28(6):287–296
Dutcher SK, Rattinger GB, Langenberg P et al (2014) Effect of medications on physical function
and cognition in nursing home residents with dementia. J Am Geriatr Soc 62(6):1046–1055
Famularo G, De Simone C, Minisola G (2005) Stevens-Johnson syndrome associated with single
high dose of lamotrigine in a patient taking valproate. Dermatol Online J 11(1):25
Ferrier N, Ferrier LJ, Macritchie KA (2006) Lithium therapy. Adv Psychiatry Treat 12:256–264
Fiedorowicz JG, Coryell WH, Rice JP et al (2012) Vasculopathy related to manic/hypomanic
symptom burden and first generation antipsychotics in a sub-sample from the Collaborative
Depression Study (CDS). Psychother Psychosom 81(4):235–243
Findling RL, Frazier JA, Kafantaris V et al (2008) The Collaborative Lithium Trials (CoLT): spe-
cific aims, methods, and implementation. Child Adolesc Psychiatry Ment Health 2(1):21
Findling RL, Kafantaris V, Pavuluri M et al (2013) Post-acute effectiveness of lithium in pediatric
bipolar I disorder. Child Adolesc Psychopharmacol 23(2):80–90
Fountoulakis KN, Vieta E, Sanchez-Moreno J et al (2005) Treatment guidelines for bipolar disor-
der: a critical review. J Affect Disord 86(1):1–10
Frangou S, Donaldson S, Hadjulis M et al (2005) The Maudsley bipolar disorder project: executive
dysfunction in bipolar disorder I and its clinical correlates. Biol Psychiatry 58:859–864
Gaily E, Kantola-Sorsa E, Hiilesmaa V et al (2004) Normal intelligence in children with prenatal
exposure to carbamazepine. Neurology 62:28–32
Galbally M, Roberts M, Buist A (2010) Perinatal psychotropic review group. Mood stabilizers in
pregnancy: a systematic review. Aust N Z J Psychiatry 44(11):967–977. doi:10.3109/0004867
4.2010.506637
Gau SS, Chao PF, Lin YJ et al (2008) The association between carbamazepine and valproate and
adverse cutaneous drug reactions in patients with bipolar disorder: a nested matched
case-control study. J Clin Psychopharmacol 28(5):509–517
Gau CS, Chang CJ, Tsai FJ et al (2010) Association between mood stabilizers and hypothyroidism
in patients with bipolar disorders: a nested, matched case-control study. Bipolar Disord
12(3):253–263. doi:10.1111/j.1399-5618.2010.00814.x
Geddes JR, Goodwin GM, Rendell J et al (2010) Lithium plus valproate combination therapy
versus monotherapy for relapse prevention in bipolar I disorder (BALANCE): a randomised
open-label trial. Lancet 375:385–395
Gitlin MJ (1993) Lithium-induced renal insufficiency. J Clin Psychopharmacol 13:276–279
Goodwin GM (2009) Evidence-based guidelines for treating bipolar disorder: revised second
edition – recommendations from the British Association for Psychopharmacology. J
Goodwin GM, Bowden CL, Calabrese JR et al (2004) A pooled analysis of 2 placebo-controlled
18-month trials of lamotrigine and lithium maintenance in bipolar I disorder. J Clin Psychiatry
65:432–441
Grandjean EM, Aubry JM (2009) Lithium: updated human knowledge using an evidence-based
approach: part III: clinical safety. CNS Drugs 23:397–418
Greil W, Häberle A, Schuhmann T et al (2013) Age and adverse drug reactions from
psychopharmacological treatment: data from the AMSP drug surveillance programme in
Switzerland. Swiss Med Wkly 143:w13772
Grunze H, Vieta E, Goodwin GM et al (2009) The World Federation of Societies of Biological
Psychiatry (WFSBP) guidelines for the biological treatment of bipolar disorders: update 2009
on the treatment of acute mania. World J Biol Psychiatry 10(2):85–116
Hansen HE (1981) Renal toxicity of lithium. Drugs 22(6):461–476
He N, Min FL, Shi YW et al (2012) Cutaneous reactions induced by oxcarbazepine in Southern
Han Chinese: incidence, features, risk factors and relation to HLA-B alleles. Seizure 21(8):
614–618
Hydzik P, Balicka-Slusarczyk B, Jastrzebski M (2011) Cardiologic interaction carbamazepine
with atypical neuroleptics. Acute mixed carbamazepine and quetiapine intoxication-case
report. Przegl Lek 68(8):499–502
Ide A, Kamijo Y (2007) Intermittent complete atrioventricular block after long term low-dose
carbamazepine therapy with a serum concentration less than the therapeutic level. Intern Med
46:627–629
Iorio ML, Moretti U, Colcera S et al (2007) Use and safety profile of antiepileptic drugs in Italy.
Eur J Clin Pharmacol 63:409–415. doi:10.1007/s00228-006-0236-6
Irga N, Kosiak W, Jaworski R et al (2013) Pediatrician! Do you know the symptoms of DRESS
syndrome? A case report of a 4-year-old girl. Pediatr Emerg Care 29(4):504–507
Jacobson SJ, Jones K, Johnson K et al (1992) Prospective multicentre study of pregnancy outcome
after lithium exposure during first trimester. Lancet 339(8792):530–533
Jafferany M (2008) Lithium and skin: dermatologic manifestations of lithium therapy. Int J
Dermatol 47(11):1101–1111
Janowsky DS, Soares J, Hatchm JP et al (2009) Lithium effect on renal glomerular function in
individuals with intellectual disability. J Clin Psychopharmacol 29:296–299
Jensen S, Rickers H (1984) Glomerular filtration rate during lithium therapy. Acta Psychiatr Scand
70:235–238
Johnson FR, Ozdemir S, Manjunath R et al (2007) Factors that affect adherence to bipolar disorder
treatments: a stated preference approach. Med Care 45(6):545–552
Jonsdottir H, Opjordsmoen S, Birkenaes AB et al (2012) Predictors of medication adherence in
patients with schizophrenia and bipolar disorder. Acta Psychiatr Scand 127(1):23–33
Kallen B, Tandberg A (1983) Lithium and pregnancy. A cohort study on manic depressive women.
Acta Psychiatr Scand 68:134–139
Källén B, Tandberg A (1983) Lithium and pregnancy. A cohort study on manic-depressive women.
Acta Psychiatr Scand 68(2):134–139
Kibirige D, Luzinda K, Ssekitoleko R (2013) Spectrum of lithium induced thyroid abnormalities:
a current perspective. Thyroid Res 6(1):3
Kim JY, Lee J, Ko YJ et al (2013) Multi-indication carbamazepine and the risk of severe cutaneous
adverse drug reactions in Korean elderly patients: a Korean health insurance databased study.
PLoS One 8(12). doi:10.1371/journal.pone.0083849
Kocak S, Girisgin SA, Gul M et al (2007) Stevens-Johnson syndrome due to concomitant use of
lamotrigine and valproic acid. Am J Clin Dermatol 8(2):107–111
Koda R, Aoyagi R, Okazaki E et al (2012) Acute tubulointerstitial nephritis with multiple organ
involvement including fatal adrenalitis: a case report with autopsy findings. Intern Med
51:2917–2922
Koutsampasopoulos K, Zotos A, Papaioannou K (2014) Carbamazepine induced atrial tachycardia
with complete AV block. Hippokratia 18(2):185–186
Kupfer DJ (2005) The increasing medical burden in bipolar disorder. J Am Med Assoc
293:2528–2530
Landi F, Onder G, Cesari M et al (2005) Silver Network Home Care Study Group. Psychotropic
medications and risk for falls among community-dwelling frail older people: an observational
study. J Gerontol A Biol Sci Med Sci 60:622–626
Lepkifker E, Sverdlik A, Iancu I et al (2004) Renal insufficiency in long-term lithium treatment. J
Clin Psychiatry 65:850–856
Levi N, Bastuji-Garin S, Mockenhaupt M et al (2009) Medications as risk factors of Stevens-
Johnson syndrome and toxic epidermal necrolysis in children: a pooled analysis. Pediatrics
123(2):e297–e304
Lin LC, Lai PC, Yang SF et al (2009) Oxcarbazepine-induced Stevens-Johnson Syndrome: a case
report. Kaohsiung J Med Sci 25(2):82–85
Lindop P, Padfield GL (1975) The renal pathology in a case of lithium-induced diabetes. J Clin
Pathol 28:472–475
Livingstone C, Rampes H (2006) Lithium: a review of its metabolic adverse effects. J
Malhi GS, Tanious M, Das P et al (2012) The science and practice of lithium therapy. Aust N Z J
Markowitz GS, Radhakrishnan J, Kambham N et al (2000) Lithium nephrotoxicity: a progres-
sive combined glomerular and tubulo-interstitial nephropathy. J Am Soc Nephrol 11:
1439–1448
Martino DJ, Igoa A, Marengo E et al (2008) Cognitive and motor features in elderly people with
bipolar disorder. J Affect Disord 105:291–295
Matsuda H, Saito K, Takayanagi Y et al (2013) Pustular-type drug-induced hypersensitivity syn-
drome/drug reaction with eosinophilia and systemic symptoms due to carbamazepine with
systemic muscle involvement. J Dermatol 40(2):118–122
McKnight RF, Adida M, Budge K et al (2012) Lithium toxicity profile: a systematic review and
meta-analysis. Lancet 379:721–728
Meador K, Baker GA, Browning N et al (2009) Cognitive function at 3 years of age after fetal
exposure to antiepileptic drugs. N Engl J Med 360:1597–1605
Müller-Oerlinghausen B, Berghöfer A, Bauer M (2002) Bipolar disorder. Lancet
359(9302):241–247
Murray CJ, Lopez AD (1997) Global mortality, disability, and the contribution of risk factors:
global burden of disease study. Lancet 349:1436–1442
Murray N, Hopwood S, Balfour DJK et al (1983) The influence of age on lithium efficacy and
side-effects in out-patients. Psychol Med 13:53–60
Nora JJ, Nora AH, Toews WH (1974) Letter: lithium, Ebstein’s anomaly, and other congenital
heart defects. Lancet 2:594–595
Olin JT, Fox LS, Pawluczyk S et al (2001) Pilot randomized trial of carbamazepine for behavioral
symptoms in treatment-resistant outpatients with Alzheimer disease. Am J Geriatr Psychiatry
9(4):400–405
Özerdem A, Tunca Z, Çımrın D et al (2014) Female vulnerability for thyroid function abnormality
in bipolar disorder: role of lithium treatment. Bipolar Disord 16(1):72–82
Pacchiarotti I, Bond DJ, Baldessarini RJ et al (2013) The International Society for Bipolar
Disorders (ISBD) task force report on antidepressant use in bipolar disorders. Am J Psychiatry
170(11):1249–1262
Pachet AK, Wisniewski AM (2003) The effects of lithium on cognition: an updated review.
Psychopharmacology 170(3):225–234, Epub 2003 Sep 19
Perucca E, Meador KJ (2005) Adverse effects of antiepileptic drugs. Acta Neurol Scand 112(Suppl
181):30–35
Perucca E, Beghi E, Dulac O et al (2000) Assessing risk to benefit ratio in antiepileptic drug
therapy. Epilepsy Res 41:107–139
Pfennig A, Bschor T, Falkai P et al (2013) The diagnosis and treatment of bipolar disorder:
recommendations from the current S3 guideline. Dtsch Arztebl Int 110(6):92–100
Pinheiro D (2008) Anticonvulsant mood stabilizers in the treatment of behavioral and psychologi-
cal symptoms of dementia (BPSD). Encéphale 34(4):409–415
Presne C, Fakhouri F, Noël LH et al (2003) Lithium-induced nephropathy: rate of progression and

prognostic factors. Kidney Int 64(2):585–592
Prien RF, Caffey EM Jr, Klett CJ (1972) Comparison of lithium carbonate and chlorpromazine in
the treatment of mania. Report of the Veteran Administration and National Institute of Mental
Health Collaborative Study Group. Arch Gen Psychiatry 26(2):146–153
Raja M (2011) Lithium and kidney, 60 years later. Curr Drug Saf 6(5):291–303
Rockwell WJ, Ellinwood EH Jr, Trader DW (1983) Psychotropic drugs promoting weight gain:
health risks and treatment implications. South Med J 76(11):1407–1412
Rolnitsky A, Merlob P, Klinger G (2013) In utero oxcarbazepine and a withdrawal syndrome,
anomalies, and hyponatremia. Pediatr Neurol 48:466–468
Ross CJ, Carleton B, Warn D et al (2007) Genotypic approaches to therapy in children: a national
active surveillance network (GATC) to study the pharmacogenomics of severe adverse drug
reactions in children. Ann N Y Acad Sci 10:177–192
Rosser R (1976) Thyrotoxicosis and lithium. Br J Psychiatry 128(48):61–66
Sarantidis D, Waters BA (1983) A review and controlled-study of cutaneous conditions associated
with lithium-carbonate. Br J Psychiatry 143:42–50
Saunders BD, Saunders EF, Gauger PG (2009) Lithium therapy and hyperparathyroidism: an
evidence-based assessment. World J Surg 33(11):2314–2323
Schou M, Vestergaard P (1988) Prospective studies on a lithium cohort: II renal function. Water
and electrolyte metabolism. Acta Psychiatr Scand 78:427–433
Schou M, Goldfield M, Weinstein MR et al (1973) Lithium and pregnancy, I: report from the
Register of Lithium Babies. BMJ 2:135–136
Schouws SN, Stek ML, Comijs HC, Beekman AT (2010) Risk factors for cognitive impairment in
elderly bipolar patients. J Affect Disord 25:330–335
Seely EW, Moore TJ, LeBoff MS et al (1989) A single dose of lithium carbonate acutely elevates
intact parathyroid hormone levels in humans. Acta Endocrinol 121:174–176
Sharma B, Sannegowda RB, Gandhi P et al (2013) Combination of Steven-Johnson syndrome and
neuroleptic malignant syndrome following carbamazepine therapy: a rare occurrence. BMJ
Case Rep. doi:10.1136/bcr-2013-008908
Slater V, Milanes F, Talcott V et al (1984) Influence of age on lithium therapy. South Med J
77:153–154, 158
Sproule BA, Hardy BG, Shulman KI (2000) Differential pharmacokinetics of lithium in elderly
patients. Drugs Aging 16:165–177
Strakowski SM, Fleck DE, Maj M (2011) Broadening the diagnosis of bipolar disorder: benefits
vs. risks. World Psychiatry 10(3):181–186
Tica OS, Tica AA, Brailoiu CG et al (2013) Sirenomelia after phenobarbital and carbamazepine
therapy in pregnancy. Birth Defects Res A Clin Mol Teratol 97(6):425–428
Torrent C, Amann B, Sánchez-Moreno J et al (2008) Weight gain in bipolar disorder: pharmaco-
logical treatment as a contributing factor. Acta Psychiatr Scand 118(1):4–18
van Melick EJ, Souverein PC, den Breeijen JH et al (2013) Age as a determinant of instability of
serum lithium concentrations. Ther Drug Monit 35(5):643–648
Vanderpump MP, Tunbridge WM, French JM et al (1995) The incidence of thyroid disorders in the
community: a twenty-year follow-up of the Whickham Survey. Clin Endocrinol 43(1):55–68
Velligan DI, Weiden PJ, Sajatovic M et al (2010) Strategies for addressing adherence problems in
patients with serious and persistent mental illness: recommendations from the expert consensus
guidelines. J Psychiatr Pract 16(5):306–324
Vestergaard P, Schou M (1981) Kidney morphology and function in lithium-treated patients. Bibl
Psychiatr 161:104–114
Von Knorring L, Wahlin A, Nystr K et al (1990) Uremia induced by long-term lithium treatment.
Lithium 1:251–253
Weinstein MR (1980) Lithium treatment of women during pregnancy and in the post-delivery
period. In: Johnson FN (ed) Handbook of lithium therapy. MTP, Lancaster, pp 421–429
Weinstein MR, Goldfield M (1975) Cardiovascular malformations with lithium use during
pregnancy. Am J Psychiatry 132:529–531
Weintraub D, Buchsbaum R, Resor SR Jr et al (2005) Effect of antiepileptic drug comedication on

lamotrigine clearance. Arch Neurol 62:1432–1436
Wijeratne C, Draper B (2011) Reformulation of current recommendations for target serum lithium
concentration according to clinical indication, age and physical comorbidity. Aust N Z J
Psychiatry 45(12):1026–1032
Wingo AP, Wingo TS, Harvey PD, Baldessarini RJ (2009) Effects of lithium on cognitive perfor-
mance: a meta-analysis. J Clin Psychiatry 70:1588–1597
Wisner KL, Leckman-Westin E, Finnerty M et al (2011) Valproate prescription prevalence among
women of childbearing age. Psychiatr Serv 62(2):218–220
Wong IC, Lhatoo SD (2000) Adverse reactions to new anticonvulsant drugs. Drug Saf
23(1):35–56
Yacobi S, Ornoy A (2008) Is lithium a real teratogen? What can we conclude from the prospective
versus retrospective studies? A review. Isr J Psychiatry Relat Sci 45(2):95–106
Yang SY, Liao YT, Liu HC et al (2013) Antipsychotic drugs, mood stabilizers, and risk of pneu-
monia in bipolar disorder: a nationwide case-control study. J Clin Psychiatry 74(1):e79–e86
Yatham LN, Kennedy SH, Parikh SV et al (2013) Canadian Network for Mood and Anxiety
Treatments (CANMAT) and International Society for Bipolar Disorders (ISBD) collaborative
update of CANMAT guidelines for the management of patients with bipolar disorder: update
2013. Bipolar Disord 15:1–44
Chapter 12
Safety and Tolerability of Medications
for ADHD
Antonio Clavenna and Maurizio Bonati
Abstract Attention deficit hyperactivity disorder (ADHD) is a common neurobe-

havioral disorder in children and adolescents that comprises core symptoms of
developmentally inappropriate levels of inattention and/or hyperactivity and impul-
sivity. Current drug treatment approaches for ADHD comprise stimulant medica-
tions (methylphenidate, amphetamines) and non-stimulant medications
(atomoxetine, clonidine and guanfacine). Drugs for ADHD appear to be safe and
well tolerated, and most of the adverse events observed in randomised clinical trials
were mild and temporary. Stimulants and atomoxetine are associated with decreased
appetite and gastrointestinal pain. An increased risk of insomnia exists for stimu-
lants, while atomoxetine and alpha-2 agonists are associated with somnolence.
Long-term safety is poorly evaluated. Concerns regarding cardiovascular and psy-
chiatric adverse events have been raised by regulatory authorities in the last decade.
However, there is currently no proven association between ADHD medications and
an increased risk of these events, and their incidence is extremely low.
Keywords Attention deficit hyperactivity disorder • Amphetamines • Atomoxetine

• Methylphenidate
12.1 Attention Deficit Hyperactivity Disorder (ADHD)
Attention deficit hyperactivity disorder (ADHD) is a common neurobehavioral dis-

order in children and adolescents that comprises core symptoms of developmentally
inappropriate levels of inattention and/or hyperactivity and impulsivity (American
Psychiatric Association 2013).
A. Clavenna (*) • M. Bonati

Department of Public Health, Laboratory for Mother and Child Health,
Mario Negri IRCCS Institute of Pharmacological Research,
Milan, Italy
e-mail: antonio.clavenna@marionegri.it

DOI 10.1007/978-3-319-24741-0_12
234 A. Clavenna and M. Bonati
The prevalence of ADHD is estimated to be approximately 7.2 % (95 % CI 6.7–

7.8) in children and 5.0 % (95 % CI 4.1–6.2) in adults according to the results of 175
epidemiological studies, although wide differences were reported across geographi-
cal regions (Thomas et al. 2015; Willcutt 2012). Diagnostic criteria, study setting,
size of population, study design and medical culture of patients as well as healthcare
practitioners can affect these estimates (Thomas et al. 2015; Willcutt 2012).
Several drug therapies for the treatment of ADHD have been evaluated in ran-
domised controlled trials.
Current drug treatment approaches for ADHD comprise stimulant medications
(methylphenidate, amphetamines) and non-stimulant medications (atomoxetine and
alpha-2 agonists clonidine and guanfacine).
Stimulants enhance the neurotransmission of dopamine and norepinephrine, ato-
moxetine selectively blocks the reuptake of norepinephrine, clonidine binds equally
to α 2A-, α 2B-, and α 2C-adrenoceptors, while guanfacine binds preferentially to
postsynaptic α 2A-adrenoceptors in the prefrontal cortex which have been impli-
cated in attentional and organisational functions.
Differences exist between countries in the availability of ADHD medications.
Atomoxetine, immediate and sustained-release formulations of methylphenidate,
amphetamines, guanfacine and clonidine are registered in the USA. In Europe, only
atomoxetine and methylphenidate are commercially available. Guanfacine and
clonidine are not licensed for use in ADHD in Europe (extended release guanfacine
was approved by European Commission in September 2015), while amphetamines
are available only in some countries.
It is widely recognised that interventions in ADHD should be based on multi-
modal treatment combining psychosocial interventions with pharmacological thera-
pies. There is a consensus on the treatment of ADHD in the international guidelines
regarding the recommended use of psychostimulant drugs, particularly methylpheni-
date, as first-line treatment, with a documented efficacy in 80 % of children consider-
ing all psychostimulants as a class. Atomoxetine, although less effective than
stimulants, may be recommended as an alternative to methylphenidate (American
Academy of Pediatrics 2011; National Collaborating Centre for Mental Health 2008).
12.2 Safety of ADHD Medications
There is mounting evidence that many conditions exist concurrently with ADHD,
and each modifies the overall clinical presentation and treatment response. About
two-thirds of children with ADHD have comorbid learning disorders, other mental
health or neurodevelopmental conditions and other nonpsychiatric disorders (Becker
et al. 2012; Chen et al. 2013; Elia et al. 2008). These comorbid conditions should be
considered together in order to broaden our understanding of the clinical picture and
optimise treatment. The significance of this issue is underscored by the fact that a few
disorders (e.g. oppositional defiant disorders and conduct disorders, depression and
12 Safety and Tolerability of Medications for ADHD 235
anxiety disorders, bipolar disorders, tic disorders, obsessive compulsive disorders

and autism spectrum disorders) are relatively common and genetically related, their
occurrence is associated with more severe impairment, and they are often underdiag-
nosed (Gillberg et al. 2004). Moreover, the same disorder can be a comorbid condi-
tion occurring before the start of pharmacological treatment (e.g. tics, sleep disorders),
as well as a drug side effect (Simpson et al. 2011). Thus, in clinical practice, in par-
ticular in psychiatry, it is a challenge to discriminate between an adverse drug reac-
tion and an exacerbation of a disease. Not all the studies investigating the safety of
drugs adequately address the prevention or minimisation of methodological bias, and
these limitations should be taken into account when evaluating adverse events (AEs)
reported for pharmacological treatments described in the following sections.
12.2.1 Findings from Clinical Trials
Since their approval, several issues have affected the use of ADHD medications,
such as tolerability, the presence of comorbidities, potential substance abuse risk
and lack of efficacy (Vaughan and Kratochvil 2012).
Hundreds of clinical studies have reported that ADHD drugs are generally
well tolerated and that most of their adverse effects are mild and/or temporary
(Aagaard and Hansen 2011; Cortese et al. 2013; Graham et al. 2011; Graham and
Coghill 2008).
Several meta-analyses have addressed the safety of methylphenidate, atomox-
etine and alpha-2 agonists in children with ADHD (Aagaard and Hansen 2010;
Cheng et al. 2007; Hirota et al. 2014; Ruggiero et al. 2014; Schachter et al. 2001;
Schwartz and Correll 2014).
According to the findings of clinical trials, irritability, crying, sleeping problems,
daydreams and anxiety were the most frequent AEs reported for stimulant users.
Decreased appetite and emotional disturbances were common AEs in children tak-
ing stimulants or atomoxetine. Headache and gastrointestinal pain were among the
most frequent AEs reported for all ADHD medications, while somnolence and
fatigue were more commonly reported with atomoxetine and alpha-2 agonists
(Table 12.1).
When compared to placebo (Table 12.2), the short-term effects of lisdexamfet-
amine and immediate-release methylphenidate are decreased appetite (number
needed to harm, NNH = 3 and 4, respectively) and insomnia (NNH = 8 and 7)
(Coghill et al. 2014; Schachter et al. 2001), while atomoxetine was associated with
significantly decreased appetite (NNH = 9), increased somnolence (NNH = 19) and
abdominal pain (NNH = 23) (Cheng et al. 2007).
An increased odds ratio was found mainly for somnolence (NNH = 5), fatigue
(NNH = 15), and sedation (NNH = 16) in guanfacine-treated patients (Ruggiero
et al. 2014) and for somnolence (NNH = 4) and fatigue (NNH = 8) in patients
treated with clonidine (Hirota et al. 2014).
Table 12.1 Adverse drug event reporting rates (%) by category

AMP (%) MPH (%) ATX (%) GUA (%) CLO (%)
ADE (5; 1,226)a (16; 2,092)a (21; 3,127)a (6; 1,150)a (1; 154)a
Irritability 7–82 1–80 2–12 6–7 8
Crying 76 2–71 1 – –
Sleeping problems 70 9–64 – – –
Decreased appetite 28–59 3–56 3–50 7 –
Daydreams 62 30–62 – – –
Anxiety 68 5–61 – – –
Emotional disturbances 59 1–56 3–55b – 4
Social withdrawal 64 27–59 – – –
Fingernail biting 40 22–45 – – –
Stomachache 40 4–32 5 – –
Anorexia 17–25 3–15 9–35 – –
Headache 12–30 2–33 4–28 17–27 –
Insomnia 17–28 2–44 3–19 4 4
Tics 26 1–28 31 – –
Somnolence 5 1–2 4–36 27–51 35
Mood alteration – 1–34 44 – –
Dizziness 5–32 1–30 2–13 5–16 –
Gastrointestinal pain 11–19 4–19 5–47 6–14 –
Nightmares 28 16–21 – – 6
Unusually happy 26 28 – – –
Fatigue 2 1–4 3–33 9–22 14
Dry mouth 5 12–24 – 6 3
Blood pressure – 3–18 18 6 –
changes
Nervousness 6 4–10 6–16 – –
Changes in heart rate – 12 11–19 – –
Abnormal behavior – 1–5 2–19 – –
Depression – 5 3–10 – –
Data from Aagaard and Hansen (2011), Ruggiero et al. (2014) and Hirota et al. (2014). Only ADEs
with rate ≥10 % for at least one medication were reported
AMP amphetamine, MPH methylphenidate, ATX atomoxetine, GUA guanfacine, CLO clonidine
a
N of studies; N of children
b
Emotional lability
12.2.2 Findings from Spontaneous Reporting
ADHD medications were among the drugs most commonly associated with sponta-
neously reported ADRs in pediatric populations (Blake et al. 2014; Hawcutt et al.
2012; Lee et al. 2014).
Methylphenidate was the most commonly suggested cause of spontaneously
reported ADRs in the UK (covering 6 % of drug-related ADRs) (Hawcutt et al.
2012), and in the USA (5 %) (Lee et al. 2014) and the second most commonly
Table 12.2 Number needed to harm (NNH) estimated from RCTs

ADE AMP MPH ATX GUA CLO
Discontinuation 26 (15–84) – 50 (33–100) 15 (12–21) 5 (4–10)
≥1 AE 6 (4–10) – 7 (5–13) 7 (5–10) 4 (2–50)
Decreased appetite 3 (3–4) 3 (2–5) 9 (7–12) – –
Insomnia 8 (6–10) 6 (4–12) n.s. – –
Gastrointestinal pain – 11 (6–83) 10 (8–14) 22 (14–48) –
Somnolence – – 19 (12–44) 5 (3.5–4.9) 4 (3–5)
Fatigue – – 62 (38–182) 15 (11–21) 8 (5–15)
Headache – 17 (10–71) – 17 (10–52) –
Irritability 19 (13–35) – – 33 (19–112) –
Dizziness 39 (22–186) 20 (12–45) 53(33–131) – –
Vomiting – – 30 (16–171) – –
Data from Coghill et al. (2014), Schachter et al. (2001), Cheng et al. (2007), Ruggiero et al. (2014),
Hirota et al. (2014). Confidence intervals (95 % CI) are reported under parenthesis
Abbreviations: ADE adverse drug event, AMP amphetamine (lisdexamfetamine), MPH methyl-
phenidate, ATX atomoxetine, GUA guanfacine, CLO clonidine
suggested cause of ADRs in Europe (2 % of all the pediatric ADRs) (Blake

et al. 2014).
Atomoxetine was the drug for which the second largest number of spontaneously
reported ADRs was registered in the UK (Hawcutt et al. 2012).
ADHD medications accounted for 14 % of ADRs collected in the World Health
Organization global individual case safety report database (VigiBase) during the
2005–2010 period occurring in children aged 2–11 years. ADRs concerned mainly
decreased appetite, psychomotor hyperactivity, upper abdominal pain, aggression,
somnolence and suicidal ideation (Star et al. 2011).
12.2.3 Regulatory Authorities’ Warnings
Drug regulatory authorities raised several concerns about cardiovascular and psy-
chiatric adverse drug reactions associated with ADHD medications in the last
decade (Clavenna and Bonati 2009).
In the USA, the Food and Drug Administration (FDA) issued a black box warn-
ing for amphetamines (risk of serious cardiovascular events and risk of drug depen-
dence), methylphenidate (risk of drug dependence), and atomoxetine (risk of
suicidal ideation). Other related warnings not leading to a black box warning con-
cerned psychiatric adverse events, effects on growth, heart rate and blood pressure
(Table 12.3).
The most recent warning issued by the FDA concerned the risk of priapism in
males treated with methylphenidate (Food and Drug Administration 2013).
According to the FDA statement, this risk is also associated with atomoxetine and
amphetamine use.
Table 12.3 Adverse drug events for ADHD medications that were highlighted in warnings issued
by the Food and Drug Administration
MPH AMP ATX GUA CLO
Drug dependencea Drug dependencea Suicidal Hypotension, Hypotension,
ideationa bradycardia bradycardia and
and syncope syncope
Serious Serious Serious Sedation and Sedation and
cardiovascular cardiovascular cardiovascular somnolence somnolence
events eventsa events
↑ Blood pressure ↑ Blood pressure ↑ Blood Cardiac Cardiac
and heart rate and heart rate pressure and conduction conduction
heart rate abnormalities abnormalities
Psychiatric Psychiatric Psychiatric
adverse events adverse events adverse events
(psychosis, (psychosis, (psychosis,
aggression, aggression, aggression,
bipolar disorder) bipolar disorder) bipolar
disorder)
Long-term Long-term Long-term
suppression of suppression of suppression of
growth growth growth
Priapism Priapism Priapism
Seizures Seizures Severe liver
injury
Peripheral Peripheral Allergic events
vasculopathy vasculopathy
Visual Visual Urinary
disturbance disturbance retention
Potential for Potential for
gastrointestinal gastrointestinal
obstruction obstruction
a
Black box warning
Abbreviations: MPH methylphenidate, ATX atomoxetine, GUA guanfacine, CLO clonidine
12.2.4 Adverse Events of Particular Concern
As stated above, despite the fact that most of AEs observed in clinical trials were
mild and temporary, physicians should be aware of some uncommon severe events
(e.g. cardiac AEs, psychiatric AEs, suicidal ideation) and on the potential impact of
drug treatment on child growth and development.
Guidelines on this were published in 2011 by the European Network for
Hyperkinetic Disorders regarding the management of clinically relevant adverse
effects of ADHD medication (Graham et al. 2011).
The following paragraphs address the most relevant AEs that clinicians should
be aware of and monitor when deciding on drug therapy for ADHD.
12.2.4.1 Sleep Disturbances
Sleep disturbances may be associated both with ADHD medication and with ADHD
itself. Higher rates of sleep problems were reported by parents in a systematic
review, but few of these were confirmed by objective sleep data (Cohen-Zion and
Ancoli-Israel 2004; Cortese et al. 2009).
Studies on the relationship between psychostimulants and ‘sleep disturbance’
assessed with objective methods (e.g. polysomnography, actigraphy) reported
inconsistent findings (Stein et al. 2012). The heterogeneity in results may be
explained by several factors, including differences in the length of the trial, the dose
of the drug and duration of exposure. It should be taken into account, however, that
a meta-analysis of eight studies (393 patients) recording homogeneous actigraphic
outcomes found that children taking methylphenidate had a decreased mean activ-
ity, a decreased total sleep time and a longer sleep latency compared with children
taking placebo (De Crescenzo et al. 2014). A history of any sleep problems should
therefore be taken before starting ADHD medication, and if this is a significant
concern, then atomoxetine could be considered as the first choice. Sleep hygiene
should be encouraged and a switch of medication should be considered when sleep
problems persist after dose adjustment and dose scheduling of the original medica-
tion (Graham et al. 2011).
12.2.4.2 Growth
Treating ADHD children with stimulants generally results in a reduction in height

and weight gain. A systematic analysis of psycho-stimulant effects in 18 studies
found height and weight deficits. The height reduction was approximately 1 cm/
year during the first 1–3 years of treatment, while the decrease of weight is esti-
mated to be around 3 kg over a 3-year period (Faraone et al. 2008).
However, the number of children who fall below the fifth growth percentile did
not increase. The initial effect of stimulants on growth appears to slow down over
time, and final adult height does not seem to be affected (Biederman et al. 2010).
A longitudinal study monitoring 243 ADHD cases and 394 controls did not find
an association between ADHD or treatment with stimulants and differences in mag-
nitude of peak height velocity (PHV) during adolescence (Harstad et al. 2014). The
mean age of PHV was slightly later in boys treated with stimulants (13.6 years in
those treated with stimulants for at least 3 years versus 12.9 years in ADHD cases
never receiving stimulants), but no correlation was found between treatment dura-
tion and change in height for age Z scores, and the adult height was not different in
ADHD patients treated with stimulants compared with ADHD stimulant naive and
controls (Harstad et al. 2014). A slower body mass index (BMI) growth in child-
hood was observed in children with ADHD and children undergoing stimulant treat-
ment compared with untreated ADHD and non-ADHD (Schwartz et al. 2014).
However, during adolescence, the rate of BMI increase was more rapid in the first
group, with BMIs eventually exceeding those of controls (Schwartz et al. 2014).
With regard to atomoxetine, a meta-analysis of seven double blind placebo-
controlled and six open-label studies found that height and weight at 24 months of
treatment were 2.5 cm and 2.7 kg lower than the expected values (Kratochvil et al.
2006). Patient weight, height and body mass index should be monitored every
6 months, and a growth chart should be used (Graham et al. 2011; National
Collaborating Centre for Mental Health 2008).
12.2.4.3 Tics
Psychostimulants increase dopamine levels and can therefore theoretically aggra-

vate tic severity.
A meta-analysis including nine double-blind placebo-controlled trials evaluating
the efficacy of medications in treatment of ADHD in patients with comorbid tic
concluded that there is no evidence that methylphenidate worsens tic severity in the
short-term, that supra-therapeutic doses of dextroamphetamine worsen tics and that
atomoxetine improves tics (Bloch et al. 2009).
However, psychostimulants may exacerbate tics in individual cases. Since tics
are common in childhood and tend to wax and wane spontaneously, an observation
period of at least 3 months is needed before making a clinical decision. If tics are
troublesome, clinicians should consider a dose reduction or drug substitution. If
these measures are ineffective, an antipsychotic could be added to control tics
(Graham et al. 2011).
12.2.4.4 Cardiac Adverse Events
In June 2009, the Food and Drug Administration issued a safety communication to
warn health professionals about a possible association between stimulant medica-
tions and an increased risk of sudden deaths in healthy children. The alert was issued
after the completion of a study funded by the FDA and the National Institute of
Mental Health (NIMH) that compared the use of stimulant medications in 564
healthy children with a registration of sudden death and in 564 children who died as
passengers in a motor vehicle accident. Stimulant use was reported by ten out of 564
children with sudden death versus two out of ten children in the control group
(Gould et al. 2009). Due to the limitation of the study, the FDA was unable to evalu-
ate the presence of a causal association. However, subsequent studies did not find an
increased risk of cardiovascular events. In a cohort of 1,200,438 children and young
adults aged 2–24 years, a total of 81 subjects had a serious cardiovascular event (3.1
per 100,000 person-years) including 33 sudden cardiac deaths (1.3 per 100,000 per-
son-years). As compared with the nonusers, the adjusted rate of serious cardiovascu-
lar events did not differ significantly among current users of ADHD drugs (hazard
ratio, HR 0.75; 95 % CI 0.31–1.85) or among former users (HR 1.03; 95 % CI
0.57–1.89) (Cooper et al. 2011). In a cohort of 241,417 incident ADHD medication
users, no statistically significant difference in the rate of validated sudden death or

ventricular arrhythmia was found compared to nonusers (HR 1.60; 95 % CI 0.19–
13.60) or all-cause death (HR 0.76; 95 % CI 0.52–1.12) (Schelleman et al. 2011).
Similar findings, i.e. no increased risk of cardiovascular events, were also found in a
study performed in the adult population (25–64 years) (Habel et al. 2011).
All stimulant medications and atomoxetine are reported to increase blood pres-
sure. An average increase of 1–5 mmHg of blood pressure and an increase of ≤10
heart beats per minute have been observed with stimulants (Hammerness et al.
2015). An average increase of 1–4 mmHg of systolic pressure and 1–2 mmHg in
diastolic pressure was observed.
An increase in blood pressure was observed with atomoxetine, with a stan-
dardised mean difference (SMD) of 0.27 mmHg (95 % CI 0.19–0.35) in diastolic
blood pressure and 0.15 mmHg (95 % CI 0.06–0.23) in systolic blood pressure
(Schwartz and Correll 2014). Elevation of blood pressure above the 95th percentile
occurred in 6.8 % of patients (systolic) and 2.8 % (diastolic) treated with atomox-
etine in comparison with 3 and 0.5 % patients treated with placebo (Hammerness
et al. 2011; Wernicke et al. 2003). Immediate-release clonidine was associated with
a significant SMD drop of 0.52 mmHg (95 % CI −0.15 to −0.89) in systolic blood
pressure and an SMD of – 0.49 mmHg (95 % CI −0.02 to −0.097) in diastolic blood
pressure.
Change in QTc interval showed no significant differences between alpha-2 ago-
nists and placebo (SMD = 0.12, 95 % CI – 0.18–0.43). In a subgroup analyses,
however, extended-release guanfacine significantly prolonged the QTc interval by a
mean of 5.3 ms (95 % CI 2.7–7.9) compared to placebo (SMD = 0.33, 95 % CI
0.12–0.43) (Hirota et al. 2014). Pretreatment monitoring of pulse and blood pres-
sure is recommended with any ADHD medication. Blood pressure should be
measured prior to treatment and at each visit and converted to a percentile score
using the appropriate chart. If BP is elevated and above the 95th percentile after at
least three measurements, patients should be referred to a pediatric cardiologist.
Moreover, before starting any ADHD drug treatment, family history of cardiac dis-
eases and history of exercise syncope, undue breathlessness and any other cardio-
vascular symptoms should be assessed (Graham et al. 2011; National Collaborating
Centre for Mental Health 2008). Routine electrocardiographic screening of ADHD
patients prior to initiation of medication is not recommended (Graham et al. 2011;
National Collaborating Centre for Mental Health 2008).
12.2.4.5 Suicide-Related Events
According to the available evidence, there is a relationship between the presence

of ADHD itself and suicidal attempts (Impey and Heun 2012). Thus it is difficult
to investigate if ADHD medications are associated with an increased risk of sui-
cidal ideation. In September 2005, the US Food and Drug Administration (FDA)
added a black box warning to the product labelling of atomoxetine. A meta-anal-
ysis of clinical trials observed that suicidal ideation was more frequently recorded
among children and adolescents treated with atomoxetine compared to those

treated with placebo (0.37 %, 5/1,357 versus 0 % (0/851) respectively) with an
NNH of 227 (Bangs et al. 2008). A subsequent meta-analysis did not find a sta-
tistically significant increased risk of suicidal behavior/ideation in children or
adults treated with atomoxetine compared with placebo (RR 1.57; 95 % CI 0.53–
4.71 and RR 0.96; 95 % CI 0.19–4.74, respectively) (Bangs et al. 2014) No com-
pleted suicide was reported, while nine out of 2,445 (0.37 %) children in the
atomoxetine group had suicidal behavior/ideation. The analysis of data collected
in the ADHD Italian register identified a total of seven cases of suicidal ideation
or self-harming behaviour were reported among 971 children treated with atom-
oxetine. No cases were reported among 1,268 children receiving methylpheni-
date. The median time to event for the seven cases was 6 months after the first
atomoxetine administration and 2 months after an increase in dosage (Capuano
et al. 2014).
A study evaluating data from the Swedish national patient register found a greater
rate of suicide-related events during treatment periods with ADHD medications
than during nontreatment period (HR = 1.31; 95 % CI 1.19–1.44). On the contrary,
no evidence of an increased risk was found when comparing the rates of suicide-
related events for the same patients over different periods (HR = 0.89; 95 % CI
0.79–1.00) (Chen et al. 2014).
Although the evidence suggests minimal risk with atomoxetine treatment,
patients should be appropriately monitored during treatment, in particular after dose
increase (Graham et al. 2011).
12.2.4.6 Seizures
ADHD is a risk factor for seizures in children. The incidence of unprovoked sei-
zures is two- to threefold greater in ADHD than non-ADHD children (Hesdorffer
et al. 2004). Among children with epilepsy, ADHD is the most common psycho-
pathological comorbidity (Dunn et al. 2009). Concern exists that ADHD medica-
tions may lower the seizure threshold. However, in ADHD patients without epilepsy,
the risk of seizure did not differ among methylphenidate, atomoxetine and placebo
(McAfee et al. 2008).
Adolescents with epilepsy are at increased risk for depression and suicidal ide-
ation. During ADHD treatment, they should be monitored for the emergence of
depression, irritability and suicidal ideation (Graham et al. 2011).
12.2.4.7 Psychosis
Psychotic symptoms are rarely associated with ADHD drug treatment. A review of
49 RCTs performed by the Food and Drug Association found 11 psychosis/mania
events, with a rate per 100 person-years in the pooled active drug group of 1.48
(95 % CI 0.74–2.65 per 100 person-years) (Mosholder et al. 2009). Among all of
the pediatric ADHD patients in placebo treatment groups, corresponding to 420
person-years of placebo exposure, there were no psychosis/mania adverse events.
The highest rate of psychotic adverse events was observed with transdermal meth-
ylphenidate (13.2 per 100 person-years) followed by dexamfetamine (two per 100
person-years) (Mosholder et al. 2009).
A total of 865 unique post-marketing cases were spontaneously reported to
manufacturers in the period 2000–2005. In the vast majority of cases, there was
no previous history of a similar psychiatric condition reported. Many young chil-
dren with hallucinations reported visual and/or tactile sensations of insects,
snakes or worms. When the drug was discontinued, the psychosis/mania-type
symptoms often resolved (Mosholder et al. 2009). According to the guidelines,
caution is needed when prescribing ADHD drugs to children and adolescents
with a past history of psychotic episodes or a family history of psychosis (Graham
et al. 2011).
12.2.4.8 Risk for Substance Use Disorders (SUD)
Children with ADHD have been found to be at increased risk for developing SUDs
(Zulauf et al. 2014).
Why SUD is linked to ADHD is still unclear. Different hypotheses have been
developed: substances may be used as a kind of self-medication with the aim to
decrease symptoms (e.g. depressed mood, insomnia); ADHD and SUD may share
alterations in dopaminergic neurotransmission and abnormalities in reward circuitry
(Cortese et al. 2013).
Concerns have been raised that ADHD medications, in particular stimulants,
may be associated with an increased risk of SUD. However, this hypothesis is not
supported by the available evidence.
A meta-analysis of 15 longitudinal studies (for a total of 2,565 subjects) did not
find any increase or reduction of SUD risk associated with ADHD medication
(Humphreys et al. 2013). Previously, meta-analysis of six studies suggested that
psychostimulant therapy might be associated with a reduced risk of SUD. A 1.9-fold
(95 % CI 1.1–3.6) reduction in risk for SUD in ADHD youths treated with stimu-
lants compared with youths with no pharmacotherapy was estimated (Humphreys
et al. 2013). Even if the risk of SUD was not proved, an increased risk of misuse/
abuse of ADHD medications is reported. Data from a 10-year follow-up study
reported a medication misuse in 22 % of ADHD children compared with 5 % of a
group receiving psychotropic drugs for reasons other than ADHD (Wilens et al.
2006). Current use or previous substance abuse in the family could be a reason to
closely monitor patient treated with ADHD medications or a relative contraindica-
tion for stimulant prescription. Atomoxetine or extended-release formulations of
stimulants should be preferred in high-risk patients, since they are less prone to
diversion (Graham et al. 2011).
12.3 Long-Term Safety
The majority of the clinical trials that monitored the occurrence of AEs in children
and adolescents receiving drug treatment for ADHD were short term (Schachter
et al. 2001; Schwartz and Correll 2014). In a systematic review of the literature
(Clavenna and Bonati 2014), only six prospective studies were found evaluating the
long-term safety of ADHD medications (Donnelly et al. 2009; Findling et al. 2009;
Harfterkamp et al. 2013; Hoare et al. 2005; McGough et al. 2005; Wilens et al.
2005). These studies concerned atomoxetine (two studies, 802 patients), the osmotic
controlled-released oral formulation of methylphenidate (two studies, 512 patients),
the extended-release formulation of mixed amphetamine salts (one study, 568
patients) and transdermal methylphenidate (one study, 326 patients). All studies
were open-label extension that followed patients previously enrolled in a total of 24
short-term randomised clinical trials, with a duration of treatment ranging from 1 to
18 weeks. Vital signs (e.g. heart rate, blood pressure), weight and height were moni-
tored in four studies (Donnelly et al. 2009; Findling et al. 2009; Hoare et al. 2005;
McGough et al. 2005). AEs were collected mainly through spontaneous reporting
by patients and/or caregivers and in one study in combination with investigator
queries. In four studies, investigators evaluated the severity of AEs and their rela-
tionship to drug treatment (Findling et al. 2009; Hoare et al. 2005; McGough et al.
2005; Wilens et al. 2005). Heterogeneity was found in the duration of follow-up
(ranging between 1 and 4 years) and the way data were reported. The rate of discon-
tinuation due to AEs was the only measure reported in all six studies (Table 12.4).
According to the long-term open-label extension studies, the rate of treatment-
related AEs ranged from 58 to 78 %, and most of the AEs were mild or moderate in
severity, with a range of 86–98 %. The rate of discontinuation due to AEs ranged
from 8 to 25 % of the children (Table 12.4). The adverse events most commonly
associated with therapy discontinuation were reported in five out of six studies.
None of these AEs were reported in all the studies. Insomnia and abdominal pain
were among the most common AEs leading to discontinuation in four studies.
Weight loss (32 % of the children who discontinued) and decreased appetite (26 %)
were the most common reasons for suspending extended-release amphetamine
(McGough et al. 2005), tics (24 %) and decreased appetite (24 %) were the most
common reasons for discontinuing osmotic controlled-released oral formulation of
methylphenidate (Wilens et al. 2005), and upper abdominal pain (21 %) and emo-
tional lability (17 %) were the most common AEs for atomoxetine discontinuation
(Donnelly et al. 2009).
Most of the discontinuation occurred during the first year of treatment. The rate
of discontinuation in patients treated with extended-release amphetamine was 7 %
in the first quarter and 6 % in the second quarter. In the third and fourth quarters, it
decreased to 2 %, after which it was <1 % (McGough et al. 2005). The discontinu-
ation rate with osmotic controlled-released oral formulation of methylphenidate
was 7 % in the first 12 months and 8 % after 24 months (Wilens et al. 2003, 2005).
A total of 71 (4 %) out of 1,553 children receiving atomoxetine interrupted the
12
Table 12.4 Number of subjects (%) with at least one adverse event (AE) (Clavenna and Bonati 2014)
N. Treatment-related
short- % Cut-off AEs AEs
term Age Duration for AE Discontinuation due
Drug CTa N. childrenb (years) (years) reporting Any Serious Any Serious to AEs
Atomoxetine 1 88 6–17 12 weeks ≥1 n.r. 2 (2) n.r. 0 11 (25)
Atomoxetine 13 1,533 6–17 ≥4 ≥10 n.r. 78 (15)c n.r. n.r. 71 (9)c
(508)c
Atomoxetine 14 596 6–17 – 82 (14)
MAS XR 2 568 (284) 6–17 2 ≥5 525 18 (3) 440 2 (<1) 84 (15)
(92) (78)
MPH OROS 3 407 (229) 6–13 2 ≥5 363 n.r. 282 n.r. 31 (8)
(89) (69)
MPH OROS 1 101 (56) 6–16 1 ≥2 76 n.r. 61 4 (4) 16 (15)
Safety and Tolerability of Medications for ADHD
(72) (58)
MPH OROS 4 508 6–16 1–2 – 439 – 343 – 47 (9)
(86) (68)
Transdermal MPH 4 326 6–15 1 ≥5 265 3 (1) n.r. 0 29 (9)
(81)
CT controlled trial, MAS XR extended-release amphetamine, OROS MPH osmotic controlled-released oral formulation of methylphenidate (MPH), n.r. not
reported
a
Number of short-term clinical trials that originally enrolled children followed up by each open-label extension study
b
Number of children followed up; number of children who were still on drug treatment at the end of the study period is reported under brackets
c
Number of children treated for at least 4 years was considered as the denominator, with the exception of discontinuation, which was calculated using the 1,533
patients initially enrolled and treated for less than 3 years
245
therapy before 3 years had elapsed. The subsequent rate of discontinuation in those
treated for ≥3 years and ≥4 years was 2 % (Donnelly et al. 2009).
Only for a few studies was it possible to compare the incidence of AEs over time.
Harfterkamp et al. compared the rate of AEs during the first 8 weeks of treatment
with atomoxetine (experimental phase) with the rate during the subsequent 12 weeks
of treatment. A decrease in the rate across time was observed for all the AEs. In
particular, a statistically significant reduction was found for nausea (from 13.6 to
1.1 %, p = 0.003) and fatigue (from 18.2 to 6.8 %, p = 0.04) (Harfterkamp et al.
2013). McGough et al. compared the percentage of the AEs reported with extended-
release amphetamine in four periods, from months 1–6 to months 18–24. For all the
AEs, the percentage of cases was highest in the first 6 months and decreased with
time. In all, 58 % of the AEs were recorded during the first 4 months of therapy
(McGough et al. 2005). In studies concerning osmotic controlled-released oral for-
mulation of methylphenidate, the incidence of AEs increased from 42 to 89 %
between the experimental phase (short-term) and the long-term open-label follow-
up (24 months) (Wilens et al. 2005). No children had tics in the experimental phase,
and no psychiatric AEs (e.g. aggravation reaction, anxiety, emotional lability and
hostility) were reported in the short-term RCTs (Clavenna and Bonati 2014). An
increase in the incidence of AEs was observed comparing short-term RCTs with the
prospective open-label extension trial concerning transdermal methylphenidate
(from 55 to 81 %). The percentage of children with decreased appetite was similar,
the incidence of headache increased from 4 to 17 %, while the incidence of insom-
nia, vomiting and nausea decreased. Tics, affect lability and anorexia were reported
only during short-term trials while abdominal pain and irritability only during the
open-label extension phase. The incidence of the most common AEs observed in
the short- and long-term periods is summarised in Fig. 12.1.
12.4 Conclusions
Drugs for ADHD seem to be safe and well tolerated. Decreased appetite, insomnia,
headache and abdominal pain are the most common adverse events observed both
in the short- and the long-term studies. Long-term safety and tolerability need to be
further investigated. Few trials are available with different follow-up duration and
criteria for defining and reporting adverse events. Many AEs are mild or moderate
in severity, and the incidence of serious events is low. Although the medications for
ADHD are generally well tolerated, with only mild or minor adverse effects in most
cases, their rational use can be guaranteed by implementing and monitoring
evidence-based practice. In this regard, recommendations regarding pretreatment
screening and monitoring of adverse events issued by the National Institute for
Health and Care Excellence (NICE) are reported in Box 12.1.
Fig. 12.1 Incidence (%) of the most common AEs in short versus long periods by drug. Incidence
was calculated using all the monitored children as denominator (Clavenna and Bonati 2014).
Abbreviations: MAS XR extended-release amphetamine
Box 12.1: National Institute for Health and Care Excellence (NICE)
recommendations (National Collaborating Centre for Mental
Health 2008)
Pre-drug treatment assessment
Before starting drug treatment, children and young people with ADHD should
have a full pretreatment assessment, which should include:
• Full mental health and social assessment
• Full history and physical examination, including:
– Assessment of history of exercise syncope, undue breathlessness and
other cardiovascular symptoms
– Heart rate and blood pressure (plotted on a centile chart)
– Height and weight (plotted on a growth chart)
– Family history of cardiac disease and examination of the cardiovascular
system
• An electrocardiogram (ECG) if there is past medical or family history of
serious cardiac disease, a history of sudden death in young family mem-
bers or abnormal findings on cardiac examination
• Risk assessment for substance misuse and drug diversion (where the drug
is passed on to others for nonprescription use).
Monitoring side effects
Healthcare professionals should consider using standard symptom and side
effect rating scales throughout the course of treatment as an adjunct to clinical
assessment for people with ADHD.
• In people taking methylphenidate, atomoxetine or dexamfetamine:
– Height should be measured every 6 months in children and young
people.
– Weight should be measured 3 and 6 months after drug treatment has
started and every 6 months thereafter in children, young people and adults.
– Height and weight in children and young people should be plotted on a
growth chart and reviewed by the healthcare professional responsible
for treatment.
• If there is evidence of weight loss associated with drug treatment in adults
with ADHD, healthcare professionals should consider monitoring body
mass index and changing the drug if weight loss persists.
• Strategies to reduce weight loss in people with ADHD or manage decreased
weight gain in children include:
– Taking medication either with or after food, rather than before meals
– Taking additional meals or snacks early in the morning or late in the
evening when the stimulant effects of the drug have worn off
– Obtaining dietary advice
– Consuming high-calorie foods of good nutritional value
• If growth is significantly affected by drug treatment (i.e. the child or young

person has not met the height expected for their age), the option of a
planned break in treatment over school holidays should be considered to
allow ‘catch-up’ growth to occur.
• In people with ADHD, heart rate and blood pressure should be monitored
and recorded on a centile chart before and after each dose change and rou-
tinely every 3 months.
• For people taking methylphenidate, dexamfetamine and atomoxetine, rou-
tine blood tests and ECGs are not recommended unless there is a clinical
indication.
• Liver damage is a rare and idiosyncratic adverse effect of atomoxetine, and
routine liver function tests are not recommended.
• For children and young people taking methylphenidate and dexamfet-
amine, healthcare professionals and parents or carers should monitor
changes in the potential for drug misuse and diversion, which may come
with changes in circumstances and age. In these situations, modified-
release methylphenidate or atomoxetine may be preferred.
• In young people and adults, sexual dysfunction (i.e. erectile and ejacula-
tory dysfunction) and dysmenorrhoea should be monitored as potential
side effects of atomoxetine.
• People taking methylphenidate, dexamfetamine or atomoxetine who have
sustained resting tachycardia, arrhythmia or systolic blood pressure greater
than the 95th percentile (or a clinically significant increase) measured on
two occasions should have their dose reduced and be referred to a paedia-
trician or adult physician.
• If psychotic symptoms (e.g. delusions and hallucinations) emerge in chil-
dren, young people and adults after starting methylphenidate or dexamfet-
amine, the drug should be withdrawn and a full psychiatric assessment
carried out. Atomoxetine should be considered as an alternative.
• If seizures are exacerbated in a child or young person with epilepsy, or de
novo seizures emerge following the introduction of methylphenidate or
atomoxetine, the drug should be discontinued immediately.
• If tics emerge in people taking methylphenidate or dexamfetamine, health-
care professionals should consider whether:
– The tics are stimulant related (tics naturally wax and wane).
– Tic-related impairment outweighs the benefits of ADHD treatment.
• If tics are stimulant related, reduce the dose of methylphenidate or dexam-
fetamine, consider changing to atomoxetine, or stop drug treatment.
Anxiety symptoms, including panic, may be precipitated by stimulants,

particularly in adults with a history of coexisting anxiety. Where this is an
issue, lower doses of the stimulant and/or combined treatment with an antide-
pressant used to treat anxiety can be used; switching to atomoxetine may be
effective.
References
Aagaard L, Hansen EH (2010) Adverse drug reactions from psychotropic medicines in the paedi-
atric population: analysis of reports to the Danish medicines agency over a decade. BMC Res
Notes 3:176
Aagaard L, Hansen EH (2011) The occurrence of adverse drug reactions reported for attention
deficit hyperactivity disorder (ADHD) medications in the pediatric population: a qualitative
review of empirical studies. Neuropsychiatr Dis Treat 7:729–744
American Academy of Pediatrics (2011) ADHD: clinical practice guideline for the diagnosis,
evaluation, and treatment of attention-deficit/hyperactivity disorder in children and adoles-
cents. Pediatrics 128:1007–1022
American Psychiatric Association (2013) Diagnostic and statistical manual of mental disorders:
DSM-5, 5th edn. American Psychiatric Publishing, Washington, DC
Bangs ME, Tauscher-Wisniewski S, Polzer J et al (2008) Meta-analysis of suicide-related behavior
events in patients treated with atomoxetine. J Am Acad Child Adolesc Psychiatry 47:209–218
Bangs ME, Wietecha LA, Wang S, Buchanan AS, Kelsey DK (2014) Meta-analysis of suicide-
related behavior or ideation in child, adolescent, and adult patients treated with atomoxetine. J
Child Adolesc Psychopharmacol 24:426–434
Becker SP, Luebbe AM, Langberg JM (2012) Co-occurring mental health problems and peer func-
tioning among youth with attention-deficit/hyperactivity disorder: a review and recommenda-
tions for future research. Clin Child Fam Psychol Rev 15:279–302
Biederman J, Spencer TJ, Monuteaux MC, Faraone SV (2010) A naturalistic 10-year prospective
study of height and weight in children with attention-deficit hyperactivity disorder grown up:
sex and treatment effects. J Pediatr 157:635–640, 640
Blake KV, Zaccaria C, Domergue F et al (2014) Comparison between paediatric and adult sus-
pected adverse drug reactions reported to the European medicines agency: implications for
pharmacovigilance. Paediatr Drugs 16:309–319
Bloch MH, Panza KE, Landeros-Weisenberger A, Leckman JF (2009) Meta-analysis: treatment of
attention-deficit/hyperactivity disorder in children with comorbid tic disorders. J Am Acad
Child Adolesc Psychiatry 48:884–893
Capuano A, Scavone C, Rafaniello C et al (2014) Atomoxetine in the treatment of attention deficit
hyperactivity disorder and suicidal ideation. Expert Opin Drug Saf 13(Suppl 1):S69–S78
Chen MH, Su TP, Chen YS et al (2013) Asthma and attention-deficit/hyperactivity disorder: a
nationwide population-based prospective cohort study. J Child Psychol Psychiatry
54:1208–1214
Chen Q, Sjolander A, Runeson B et al (2014) Drug treatment for attention-deficit/hyperactivity
disorder and suicidal behaviour: register based study. BMJ 348:g3769
Cheng JY, Chen RY, Ko JS, Ng EM (2007) Efficacy and safety of atomoxetine for attention-deficit/
hyperactivity disorder in children and adolescents-meta-analysis and meta-regression analysis.
Psychopharmacol (Berl) 194:197–209
Clavenna A, Bonati M (2009) Adverse drug reactions in childhood: a review of prospective studies
and safety alerts. Arch Dis Child 94:724–728
Clavenna A, Bonati M (2014) Safety of medicines used for ADHD in children: a review of pub-
lished prospective clinical trials. Arch Dis Child 99:866–872
Coghill DR, Caballero B, Sorooshian S, Civil R (2014) A systematic review of the safety of lisdex-
amfetamine dimesylate. CNS Drugs 28:497–511
Cohen-Zion M, Ancoli-Israel S (2004) Sleep in children with attention-deficit hyperactivity disor-
der (ADHD): a review of naturalistic and stimulant intervention studies. Sleep Med Rev
8:379–402
Cooper WO, Habel LA, Sox CM et al (2011) ADHD drugs and serious cardiovascular events in
children and young adults. N Engl J Med 365:1896–1904
Cortese S, Faraone SV, Konofal E, Lecendreux M (2009) Sleep in children with attention-deficit/
hyperactivity disorder: meta-analysis of subjective and objective studies. J Am Acad Child
Adolesc Psychiatry 48:894–908
Cortese S, Holtmann M, Banaschewski T et al (2013) Practitioner review: current best practice in

the management of adverse events during treatment with ADHD medications in children and
adolescents. J Child Psychol Psychiatry 54:227–246
De Crescenzo F, Armando M, Mazzone L et al (2014) The use of actigraphy in the monitoring of
methylphenidate versus placebo in ADHD: a meta-analysis. Atten Defic Hyperactivity Disord
6:49–58
Donnelly C, Bangs M, Trzepacz P et al (2009) Safety and tolerability of atomoxetine over 3 to 4
years in children and adolescents with ADHD. J Am Acad Child Adolesc Psychiatry
48:176–185
Food and Drug Administration (2013) Methylphenidate ADHD medications: drug safety
communication – risk of long-lasting erections. http://www.fda.gov/Safety/MedWatch/
SafetyInformation/SafetyAlertsforHumanMedicalProducts/ucm378876.htm?source=gov
delivery&utm_medium=email&utm_source=govdelivery
Dunn DW, Austin JK, Perkins SM (2009) Prevalence of psychopathology in childhood epilepsy:
categorical and dimensional measures. Dev Med Child Neurol 51:364–372
Elia J, Ambrosini P, Berrettini W (2008) ADHD characteristics: I. Concurrent co-morbidity pat-
terns in children & adolescents. Child Adolesc Psychiatry Ment Health 2:15
Faraone SV, Biederman J, Morley CP, Spencer TJ (2008) Effect of stimulants on height and weight:
a review of the literature. J Am Acad Child Adolesc Psychiatry 47:994–1009
Findling RL, Wigal SB, Bukstein OG et al (2009) Long-term tolerability of the methylphenidate
transdermal system in pediatric attention-deficit/hyperactivity disorder: a multicenter, prospec-
tive, 12-month, open-label, uncontrolled, phase III extension of four clinical trials. Clin Ther
31:1844–1855
Gillberg C, Gillberg IC, Rasmussen P et al (2004) Co-existing disorders in ADHD – implications
for diagnosis and intervention. Eur Child Adolesc Psychiatry 13(Suppl 1):I80–I92
Gould MS, Walsh BT, Munfakh JL et al (2009) Sudden death and use of stimulant medications in
youths. Am J Psychiatry 166:992–1001
Graham J, Coghill D (2008) Adverse effects of pharmacotherapies for attention-deficit hyperactiv-
ity disorder: epidemiology, prevention and management. CNS Drugs 22:213–237
Graham J, Banaschewski T, Buitelaar J et al (2011) European guidelines on managing adverse
effects of medication for ADHD. Eur Child Adolesc Psychiatry 20:17–37
Habel LA, Cooper WO, Sox CM et al (2011) ADHD medications and risk of serious cardiovascu-
lar events in young and middle-aged adults. JAMA 306:2673–2683
Hammerness PG, Perrin JM, Shelley-Abrahamson R, Wilens TE (2011) Cardiovascular risk of
stimulant treatment in pediatric attention-deficit/hyperactivity disorder: update and clinical rec-
ommendations. J Am Acad Child Adolesc Psychiatry 50:978–990
Hammerness PG, Karampahtsis C, Babalola R, Alexander ME (2015) Attention-deficit/hyperac-
tivity disorder treatment: what are the long-term cardiovascular risks? Expert Opin Drug Saf
14:543–551
Harfterkamp M, Buitelaar JK, Minderaa RB et al (2013) Long-term treatment with atomoxetine
for attention-deficit/hyperactivity disorder symptoms in children and adolescents with autism
spectrum disorder: an open-label extension study. J Child Adolesc Psychopharmacol
23:194–199
Harstad EB, Weaver AL, Katusic SK et al (2014) ADHD, stimulant treatment, and growth: a lon-
gitudinal study. Pediatrics 134:e935–e944
Hawcutt DB, Mainie P, Riordan A, Smyth RL, Pirmohamed M (2012) Reported paediatric adverse
drug reactions in the UK 2000–2009. Br J Clin Pharmacol 73:437–446
Hesdorffer DC, Ludvigsson P, Olafsson E et al (2004) ADHD as a risk factor for incident unpro-
voked seizures and epilepsy in children. Arch Gen Psychiatry 61:731–736
Hirota T, Schwartz S, Correll CU (2014) Alpha-2 agonists for attention-deficit/hyperactivity disor-
der in youth: a systematic review and meta-analysis of monotherapy and add-on trials to stimu-
lant therapy. J Am Acad Child Adolesc Psychiatry 53:153–173
Hoare P, Remschmidt H, Medori R et al (2005) 12-month efficacy and safety of OROS MPH in
children and adolescents with attention-deficit/hyperactivity disorder switched from MPH. Eur
Child Adolesc Psychiatry 14:305–309
Humphreys KL, Eng T, Lee SS (2013) Stimulant medication and substance use outcomes: a meta-
analysis. JAMA Psychiatry 70:740–749
Impey M, Heun R (2012) Completed suicide, ideation and attempt in attention deficit hyperactivity
disorder. Acta Psychiatr Scand 125:93–102
Kratochvil CJ, Wilens TE, Greenhill LL et al (2006) Effects of long-term atomoxetine treatment
for young children with attention-deficit/hyperactivity disorder. J Am Acad Child Adolesc
Lee WJ, Lee TA, Pickard AS, Caskey RN, Schumock GT (2014) Drugs associated with adverse
events in children and adolescents. Pharmacotherapy 34:918–926
McAfee AT, Holdridge KC, Johannes CB, Hornbuckle K, Walker AM (2008) The effect of phar-
macotherapy for attention deficit hyperactivity disorder on risk of seizures in pediatric patients
as assessed in an insurance claims database. Curr Drug Saf 3:123–131
McGough JJ, Biederman J, Wigal SB et al (2005) Long-term tolerability and effectiveness of once-
daily mixed amphetamine salts (Adderall XR) in children with ADHD. J Am Acad Child
Adolesc Psychiatry 44:530–538
Mosholder AD, Gelperin K, Hammad TA, Phelan K, Johann-Liang R (2009) Hallucinations and
other psychotic symptoms associated with the use of attention-deficit/hyperactivity disorder
drugs in children. Pediatrics 123:611–616
National Collaborating Centre For Mental Health (2008) Attention deficit hyperactivity disorder:
diagnosis and management of ADHD in children, young people and adults. NICE, London
Ruggiero S, Clavenna A, Reale L et al (2014) Guanfacine for attention deficit and hyperactivity
disorder in pediatrics: a systematic review and meta-analysis. Eur Neuropsychopharmacol
24:1578–1590
Schachter HM, Pham B, King J, Langford S, Moher D (2001) How efficacious and safe is short-
acting methylphenidate for the treatment of attention-deficit disorder in children and adoles-
cents? A meta-analysis. CMAJ 165:1475–1488
Schelleman H, Bilker WB, Strom BL et al (2011) Cardiovascular events and death in children
exposed and unexposed to ADHD agents. Pediatrics 127:1102–1110
Schwartz S, Correll CU (2014) Efficacy and safety of atomoxetine in children and adolescents with
attention-deficit/hyperactivity disorder: results from a comprehensive meta-analysis and
metaregression. J Am Acad Child Adolesc Psychiatry 53:174–187
Schwartz BS, Bailey-Davis L, Bandeen-Roche K et al (2014) Attention deficit disorder, stimulant
use, and childhood body mass index trajectory. Pediatrics 133:668–676
Simpson HA, Jung L, Murphy TK (2011) Update on attention-deficit/hyperactivity disorder and tic
disorders: a review of the current literature. Curr Psychiatry Rep 13:351–356
Star K, Noren GN, Nordin K, Edwards IR (2011) Suspected adverse drug reactions reported for
children worldwide: an exploratory study using VigiBase. Drug Saf 34:415–428
Stein MA, Weiss M, Hlavaty L (2012) ADHD treatments, sleep, and sleep problems: complex
associations. Neurotherapeutics 9:509–517
Thomas R, Sanders S, Doust J, Beller E, Glasziou P (2015) Prevalence of attention-deficit/hyper-
activity disorder: a systematic review and meta-analysis. Pediatrics 135(4):e994–e1001
Vaughan B, Kratochvil CJ (2012) Pharmacotherapy of pediatric attention-deficit/hyperactivity dis-
order. Child Adolesc Psychiatr Clin N Am 21:941–955
Wernicke JF, Faries D, Girod D et al (2003) Cardiovascular effects of atomoxetine in children,
adolescents, and adults. Drug Saf 26:729–740
Wilens T, Pelham W, Stein M et al (2003) ADHD treatment with once-daily OROS methylpheni-
date: interim 12-month results from a long-term open-label study. J Am Acad Child Adolesc
Wilens T, McBurnett K, Stein M et al (2005) ADHD treatment with once-daily OROS methylphe-
nidate: final results from a long-term open-label study. J Am Acad Child Adolesc Psychiatry
44:1015–1023
Wilens TE, Gignac M, Swezey A, Monuteaux MC, Biederman J (2006) Characteristics of adoles-
cents and young adults with ADHD who divert or misuse their prescribed medications. J Am
Acad Child Adolesc Psychiatry 45:408–414
Willcutt EG (2012) The prevalence of DSM-IV attention-deficit/hyperactivity disorder: a meta-
analytic review. Neurotherapeutics 9:490–499
Zulauf CA, Sprich SE, Safren SA, Wilens TE (2014) The complicated relationship between atten-
tion deficit/hyperactivity disorder and substance use disorders. Curr Psychiatry Rep 16:436
Part III
Pharmacovigilance of Psychotropic Drugs
in Special Populations
Chapter 13
Safety of Psychotropic Drugs in Children
and Adolescents
Florentia Kaguelidou and Eric Acquaviva
Abstract The recent rise in the use of psychotropic drugs to treat mental disease in
children and adolescents has not been accompanied by quality research evidence on
their efficient and safe use in this population. Currently, 60–70 % of pharmacologi-
cal prescriptions in pediatric psychiatry are considered “off label,” that is, used in
age ranges, doses, and indications that are not approved by regulatory authorities. In
addition, children and adolescents exposed to psychotropic drugs may be at higher
risk than adults for certain adverse events such as metabolic and endocrine abnor-
malities associated with second-generation antipsychotic treatments. They may also
present adverse events not previously assessed in adults such as growth delay with
chronic use of methylphenidate. As psychotropic drugs are prescribed to control
clinical symptoms in long-lasting psychiatric disorders, specific attention should be
paid on the detection of delayed adverse events due to exposure during childhood.
Further research should also elucidate the physiopathological mechanisms of psy-
chotropic-induced toxicity and the potential value of personalized approaches based
on genetics and neurobiology. In conclusion, additional safety data are urgently
needed to clarify the risk-benefit ratio of psychotropic medications in children and
adolescents and to adequately guide medical decision-making.
Keywords Psychotropic drugs • Children • Adolescents • Safety • Off-label
F. Kaguelidou (*)
Department of Pediatric Pharmacology and Pharmacogenetics, Assistance Publique - Hôpitaux
de Paris, Robert Debré University Hospital, Paris, France
University Paris 7 - Diderot, Sorbonne Paris Cité, Paris, France
e-mail: florentia.kaguelidou@rdb.aphp.fr
E. Acquaviva
Department of Child and Adolescent Psychiatry, Assistance Publique - Hôpitaux
de Paris, Robert Debré University Hospital, Paris, France
e-mail: dracquaviva@aol.com

DOI 10.1007/978-3-319-24741-0_13
258 F. Kaguelidou and E. Acquaviva
Abbreviations
ADHD Attention deficit hyperactivity disorder

BMI Body mass index
EPS Extrapyramidal symptoms
FGAs First-generation antipsychotics
MAOIs Monoamine oxidase inhibitors
NaSSA Noradrenergic and specific serotoninergic antidepressant
SCD Sudden cardiac death
SGAs Second-generation antipsychotics
SNRIs Serotonin-norepinephrine reuptake inhibitors
SSRIs Selective serotonin reuptake inhibitors
SUD Substance use disorder
TCAs Tricyclic antidepressants
13.1 Introduction
Several millions of children and adolescents are being exposed to psychotropic

medications worldwide. A five- to ninefold increase on the number of psychotropic
prescriptions has been observed in most countries during the past 25 years
(Steinhausen 2014; Steinhausen and Bisgaard 2014; Thomas et al. 2006).
There are several explanations to this increase in the use of psychiatric drugs.
First, the change in modern psychiatry with the emphasis on a medical model as
compared to a psychosocial interpretation of mental illness led to a greater utiliza-
tion of medical interventions such as pharmacological agents. Moreover, increas-
ing knowledge and awareness about mental disease and its negative social impact
motivated early management in children and adolescents. Consequently, psychiat-
ric disorders that have a childhood onset have been increasingly diagnosed and
treated. Likewise, there has been a clear trend toward greater acceptability and
demand for use of psychotropics in children, especially in those presenting with
profound social impairment. Finally, in countries with limited availability of non-
pharmacological therapeutic resources and inpatient psychiatric services, the use
of these agents is a quick and affordable treatment and the only possibility of
shortening hospital stay and favor outpatient care of these vulnerable patients
(Harrison et al. 2012).
Despite the rise in the use of psychotropics to treat mental disease in children and
adolescents, only a few drugs have a marketing authorization for use in this popula-
tion. Consequently, psychotropics in children and adolescents are prescribed mainly
“off label,” i.e., for indications that are not approved by regulatory authorities or in
children who are younger than the approved age ranges. Currently, 60–70 % of
13 Safety of Psychotropic Drugs in Children and Adolescents 259
prescriptions in pediatric psychiatry are considered “off label” (Koelch et al. 2009;
Winterfeld et al. 2009).
Still, the majority of psychotropic medications are associated with significant
adverse reactions in adults; thus their use in children is a major public health issue.
Children and adolescents may be more susceptible to drug-related adverse events
than adults, and this susceptibility may vary with age, growth, and developmental
state. Recently, several regulatory warnings have been issued because of specific
safety concerns such as the increased risk of suicidal behaviors in depressed chil-
dren and adolescents treated with antidepressants although the reality of this asso-
ciation is still under debate (Stone 2014). Other recent examples include the rising
concerns about the cardiac toxicity of methylphenidate and the potential endocrine
adverse events of second-generation antipsychotics in children.
These considerations highlight the need to clarify the safety profile of psychotro-
pic drugs in children and adolescents to promote their efficient and safe use and
improve medical decision-making in routine practice.
13.2 Safety of Psychotropic Drugs in Children

and Adolescents
13.2.1 Psychostimulants
13.2.1.1 Background
The use of psychostimulants in children with behavioral problems dates back to the
1930s. Currently, methylphenidate is the most frequently prescribed psychostimu-
lant drug worldwide. In most European countries, it is also the only psychostimu-
lant marketed for use in children, whereas in the USA, other molecules such as
dextroamphetamine, lisdexamfetamine, dexmethylphenidate, and amphetamine
mixed salts are also authorized in children and adolescents with attention deficit
hyperactivity disorder (ADHD).
ADHD affects approximately 5 % of school-aged children and about 4 % of
adults. The main symptoms are inattention, hyperactivity, and impulsivity.
Individuals with ADHD may have difficulties in school, troubled relationships with
family and peers, and low self-esteem. Psychostimulants are the most commonly
prescribed medications for this indication although other agents such as alpha2-
adrenergic (clonidine, guanfacine) or selective norepinephrine reuptake inhibitors
(atomoxetine) may also be prescribed. Of course, a multimodal therapeutic approach
is often necessary for this condition including pharmacological but also patient and
family counseling or behavioral therapies. This chapter will be limited to the
description of specific side effects of psychostimulants in children and adolescents
with ADHD.
13.2.1.2 Safety of Psychostimulants in Children and Adolescents
The most extensively evaluated agent with regard to safety in the treatment of
ADHD is methylphenidate. Nevertheless, all ADHD psychostimulants share similar
pharmacological properties, and the described side effects are quite similar for the
different molecules.
The main side effects observed are a decrease in appetite, stomachache, nausea,
headache, insomnia, and nervousness (Efron et al. 1997; Greenhill et al. 2001).
They are typically observed at the initiation of treatment by a psychostimulant
agent, and a dose-effect relationship has been demonstrated (Efron et al. 1997;
Greenhill et al. 2001). Side effects may lead to treatment discontinuation in less
than 5 % of school-aged children treated with methylphenidate. However, this per-
centage varies widely and is estimated at approximately 9 % of preschoolers and
18 % of children with pervasive developmental disorders (Efron et al. 1997;
Greenhill et al. 2001; Wigal et al. 2006).
Three major safety issues associated with the use of psychostimulants in children
and adolescents will be discussed in this chapter as they are specific to the pediatric
population and have been the subject of safety warnings issued by several regula-
tory agencies.
Cardiovascular Effects
Effects on Blood Pressure and Heart Rate

ADHD psychostimulants are sympathomimetic agents which increase noradrener-
gic and dopaminergic transmission. An increase of the heart rate and the blood
pressure may be considered as one of their intrinsic pharmacological properties
(Volkow et al. 2003). Indeed, clinical examination shortly after administration of
methylphenidate has shown a slight, but statistically significant, increase in blood
pressure and heart rate (increase of 2–6 bpm for the heart rate, 2–4 mmHg for the
systolic blood pressure, and 1–3 mmHg for the diastolic blood pressure) (Cortese et
al. 2013; Hammerness et al. 2011; Vitiello 2008).
In the Multimodal Treatment of Attention Deficit Hyperactivity Disorder (MTA)
study, 579 children aged 7–9 years were randomly assigned to one of the four treat-
ments: medication management, behavioral therapy, combination of medication
management and behavioral therapy, or usual community care, for a total of
14 months (Vitiello et al. 2012). No treatment effect on either systolic or diastolic
blood pressure could be detected. A moderate increase in the mean heart rate at the
end of the trial was observed in children treated with stimulants (84.2 bpm
[SD = 12.4] in the medication management group and 84.6 bpm [SD = 12.0] in med-
ication management and behavioral therapy) compared to those treated with behav-
ioral therapy alone or usual community treatment (79.1 [SD = 12.9] and 78.9 bpm
[SD = 12.9], respectively). Conversely, a large open-label study including 2,968
children treated for up to 15 weeks with amphetamine salts showed moderate
increases in both heart rate and blood pressure which tended to persist but were
considered not clinically relevant (Donner et al. 2007). Furthermore, currently
available data do not support the potential association between the use of stimulant
medications and risk of electrocardiographic modifications (Awudu and Besag
2014).
Beyond 2 years of treatment, limited evidence on cardiovascular toxicity is avail-
able. The MTA study included a naturalistic follow-up of participants at 3, 6, 8 and
10 years and did not show any increased risk of hypertension or tachycardia. Yet,
subjects with highest cumulative stimulant exposure during the trial presented
higher heart rates 8 years after treatment discontinuation regardless of current medi-
cation use (Vitiello et al. 2012). This raises the question of a potential impact of
stimulant treatment on long-term blood pressures and heart rates in previously
treated patients. Indeed, the risk for cardiovascular disease increases with increas-
ing values of blood pressure even though no specified cutoff has been defined
(Vitiello 2008). Currently, there is no evidence supporting the presence of an
increased risk for hypertension or cardiovascular disease in adults who were medi-
cated as children. Yet, this issue is still poorly investigated especially with regard to
the fact that stimulant treatments are usually prescribed from early childhood to
adulthood. It is of note that self-reported cardiac events in adulthood were not asso-
ciated with past ADHD medications in the 2004–2005 National Epidemiologic
Survey on Alcohol and Related Conditions (NESARC) (Peyre et al. 2014).
Sudden and Severe Cardiovascular Events

In 2006, the FDA has issued several safety warnings after 18 sudden cardiac deaths
in children treated with psychostimulant medications between 1992 and 2005.
However, six of these patients presented underlying cardiac abnormalities (Bange et
al. 2014). Ever since, the potential risk association between stimulant treatment and
sudden cardiac events (sudden cardiac death, acute myocardial infarction, and stroke)
in children has been a major focus. In a US population-based cohort study of children
and young adults, current users of ADHD medications were not found to be at
increased risk of severe cardiovascular events when compared to nonusers and for-
mer users (Cooper et al. 2011; Gould et al. 2009). Similarly, a review of five popula-
tion-based studies based on insurance claims data showed a moderate association of
ADHD stimulant medications and occurrence of mild cardiovascular events such as
tachycardia. However, no association with more serious events such as sudden car-
diac death, stroke, or myocardial infarction was found (Winterstein et al. 2012). To
date, there is no reliable evidence to support risk association between ADHD stimu-
lant drugs and severe cardiovascular events in children and adolescents.
Thus, the European ADHD Guidelines Group recommends a thorough clinical
interview to determine any potential cardiovascular risk factor before starting stim-
ulant medication (Cortese et al. 2013). Baseline heart rate and blood pressure should
be recorded and monitored every 3–6 months during therapy. Electrocardiographic
and cardiac ultrasound examinations are not to be performed systematically except
for individuals with personal or family cardiovascular risk factors.
Growth and Metabolic Effects
The possibility that long-term stimulant treatment of children may impact growth
velocity is now well documented. From 6 months to 3.5 years of treatment, an
analysis of 18 cohort studies of psychostimulants prescriptions found evidence of
significant height and weight deficits in treated populations (Faraone et al. 2008).
The effect on weight typically emerges in the first few months of treatment and
stabilizes afterward, but the effect on height takes at least 1 year to become clini-
cally detectable. The deficit in height is estimated to be approximately of 1 cm per
year during the first 1–3 years of therapy in children treated with methylphenidate
daily doses above 20 mg (Cortese et al. 2013; Faraone et al. 2008; Poulton 2005).
Nevertheless, weight and height rebounds have been reported after treatment dis-
continuation, and a full compensation of the initial weight and height drug-induced
loss is possible 2 years after the end of stimulant therapy (Cortese et al. 2013;
Faraone et al. 2008). No differences on growth suppression were found between
users of methylphenidate and those of other psychostimulant agents (Faraone et al.
2008).
The results of the Multimodal Treatment of Attention Deficit Hyperactivity
Disorder (MTA) study suggest that the effects of stimulants on growth and weight
are dose dependent. In the four treatment groups, medication management, behav-
ioral therapy, combination of medication management and behavioral therapy, and
routine community care, mean height increase during study period (14 months of
treatment and 10 month follow-up) was, respectively, 4.25 cm, 6.19 cm, 4.85 cm,
and 5.68 cm. The estimated loss of height was 1.23 cm/year in the medication man-
agement group. With regard to weight, the respective gain observed was 1.64 kg,
4.53 kg, 2.52 kg, and 3.13 kg. Again, estimated loss of weight was 2.48 kg/year in
the medication management group (MTA Cooperative Group National Institute of
Mental Health Multimodal Treatment Study of ADHD follow-up: 24 month out-
comes of treatment strategies for attention deficit hyperactivity disorder 2004).
During the naturalistic follow-up of the MTA population, growth suppression was
especially evident during the first 2 years of stimulant treatment (MTA Cooperative
Group National Institute of Mental Health Multimodal Treatment Study of ADHD
follow-up: 24 month outcomes of treatment strategies for attention deficit hyperac-
tivity disorder 2004; Biederman et al. 2010).
In a large cohort of adults with ADHD using data from the National Epidemiologic
Survey on Alcohol and Related Conditions (NESARC), ADHD participants who
received stimulants (n = 216) had a mean height of 68.29 in. (SD = 0.34), those who
never received stimulants (n = 591) had a mean height of 67.90 in. (SD = 0.23), and
participants without ADHD (n = 33,846) had a mean height of 66.93 in. (SD = 0.05).
Following statistical adjustments, no significant difference in adult height was found
between groups (Peyre et al. 2013).
In conclusion, current evidence supports the association between the use of stim-
ulants and height and weight gain delays. Impact on growth seems to be dose depen-
dent and reversible after discontinuation of the stimulant agent. These delays do not
seem to affect final height, but further data from children with long-term stimulant
medication and use during puberty are needed to clarify this issue. Also, the under-
lying physiopathological mechanisms remain to be defined. A first hypothesis is
related to the decrease of appetite caused by stimulant medications. According to
another hypothesis, hypothalamic dopamine changes induced by stimulants would
be associated with modifications of the pituitary function and subsequently growth.
This last hypothesis may further be supported by the fact that dopamine antagonists
increase weight and appear to accelerate height growth. More recently, a transient
decrease in insulin-dependent growth factor after 4 months of stimulant treatment
was reported in a small number of children, but no changes were evident after 8 or
14 months of follow-up. Despite these hypotheses, the mechanism implicated in the
height growth delay is still unknown (Vitiello 2008).
Giving these elements, the European ADHD Guidelines Group and several
national regulatory agencies recommend monitoring of appetite, weight, height, and
body mass index every 6 months in children and adolescents under stimulant ther-
apy (Graham et al. 2011). Some clinicians also recommend the administration of
medication after meal, use of high-calories snacks or late evening meals, and dis-
continuation of treatment during weekends or holidays, but evidence on the impact
of these measures is extremely limited. Moreover, the reduction of drug dose and
switching to alternative drug class or formulation have been proposed. Addressing
patients to a pediatric endocrinologist or a growth specialist is advised when height
and weight values are below critical thresholds.
Psychiatric and Neurological Effects
The association between the use of stimulant medications and stimulant misuse or
substance use disorder has been extensively evaluated. On one hand, most of studies
did not find an increased risk in stimulants misuse in treated ADHD children and
adolescents. Although the misuse of stimulants is estimated to be between 5 % and
9 % in high school and up to 35 % in college-aged populations (Goldstein 2013;
Wilens et al. 2008), ADHD patients were not found to be more at risk of stimulant
misuse than non-ADHD subjects. Several factors related to the risk of stimulant
misuse have been determined such as the presence of a conduct disorder, use of
immediate release psychostimulant agents, and male gender. Therefore, it is advised
to closely monitor the use of psychostimulant in patients with current or previous
substance abuse history and to preferably use extended release formulations (Cortese
et al. 2013; Wilens and Morrison 2011). On the other hand, ADHD children and
adolescents are more susceptible to present substance use or substance use disorder
(SUD) than non-ADHD individuals (Molina et al. 2013). Several studies have eval-
uated whether the use of stimulants increases further this existing risk, and results
are quite controversial. A meta-analysis of six studies suggests that psychostimulant
therapy in childhood might be associated with a reduction in the risk of alcohol or
drug use disorders. The ADHD-treated population of children was 1.9 (1.1–3.6) less
at risk of abuse compared to nontreated ADHD children (Wilens et al. 2003). A
more recent meta-analysis however suggested that the treatment of ADHD with
stimulant medication neither protects nor increases the risk for later substance use
disorder (Humphreys et al. 2013). All these results underline the need to carefully
monitor children and adolescents with ADHD especially given their increased risk
for use and abuse of multiple substances that may not improve with stimulant medi-
cation (Molina et al. 2013).
The occurrence of suicidal thoughts and behaviors during stimulant treatment
has also been studied. A register-based longitudinal study did not find any evidence
of an association between the use of drug treatments for ADHD and the risk of sui-
cidal behavior (Chen et al. 2014). However, it is highly recommended to screen
patients for suicidal thoughts and behavior before starting stimulant medication and
continue appropriate monitoring during treatment. Due to the high rate of psychiat-
ric comorbidities in ADHD, focus on the management of disorders underlying sui-
cidal thoughts or behaviors is also advised. The European ADHD Guidelines Group
does not consider the presence of suicidal thoughts or behaviors as an absolute
contraindication for ADHD drugs; however a dose reduction or discontinuation
may be warranted (Graham et al. 2011).
Occurrence of other psychiatric symptoms such as psychotic symptoms has been
found in approximately 1.5 % of ADHD subjects treated with stimulants in a review
of post-marketing surveillance reports and clinical trials (Mosholder et al. 2009). It
is recommended to reduce therapeutic doses of ADHD drugs or to consider discon-
tinuation of ADHD treatment in case of occurrence of such symptoms.
Further, in patients with tic disorders, the use of ADHD stimulant agents may
worsen clinical symptoms. However, a Cochrane Collaboration meta-analysis con-
cluded that psychostimulants do not overall worsen tics but may do so in individual
cases (Pringsheim and Steeves 2011). To date, there is no evidence associating treat-
ment by stimulant to the onset of tic disorders (Roessner et al. 2006).
Finally, there is no evidence of an increased risk of seizure episodes in psycho-
stimulant-treated children with well-controlled epilepsy (Koneski and Casella 2010;
Koneski et al. 2011). Nevertheless, the use of methylphenidate is not recommended
in case of seizures, and a close collaboration between pediatric neurologists and
psychiatrists is recommended in children with epilepsy.
13.2.2 Antidepressants
13.2.2.1 Antidepressant Classes and Indications
The most important classes of antidepressants prescribed in children and adoles-

cents are the following: (1) selective serotonin reuptake inhibitors (SSRIs) (citalo-
pram, fluoxetine, fluvoxamine, paroxetine, sertraline, escitalopram), (2)
serotonin-norepinephrine reuptake inhibitors (SNRIs) (venlafaxine, milnacipran,
duloxetine), (3) tricyclic antidepressants (TCAs) (imipramine, amitriptyline, nor-
triptyline), (4) noradrenergic and specific serotoninergic antidepressant (NaSSA)
(mirtazapine), and (5) monoamine oxidase inhibitors (MAOIs) (phenelzine,
selegiline). Other classes of antidepressants are also marketed for use in adults but
marginally used in children such as melatonergic antidepressants (agomelatonin),
norepinephrine reuptake inhibitors (reboxetine), or norepinephrine-dopamine reup-
take inhibitors (bupropion). In children, SSRIs are the most commonly prescribed
agents, representing approximately 70–80 % of total antidepressants’ prescriptions
(Hoffmann et al. 2014; Ma et al. 2005; Wijlaars et al. 2012).
Antidepressants are prescribed in three main indications in children and
adolescents:
1. Depression: In most countries, fluoxetine is the only antidepressant approved by
national agencies to treat depression in children over the age of 8 years. In some
countries such as the USA, escitalopram is also labeled for use in children aged
over 12 years.
2. Obsessive-compulsive disorder: fluvoxamine and sertraline are authorized in
several European countries and in the USA. In addition, fluoxetine and clomip-
ramine are also authorized in this indication in the USA.
3. Child enuresis: A tricyclic antidepressant, imipramine, is authorized in many
countries for childhood enuresis from the age of 6 years.
The use of antidepressants in children and adolescents has been evaluated in
several other indications. The efficacy of SSRIs associated with a cognitive-
behavioral therapy (CBT) in anxiety disorders has been extensively studied.
SRRIs have also been evaluated in ADHD, but data remain scarce (Mohatt et al.
2014; Park et al. 2014; Piacentini et al. 2014; Strawn et al. 2014). The use of
TCAs has been evaluated in depression, ADHD, anxiety disorders, and chronic
pain (Ghanizadeh 2013; Hazell and Mirzaie 2013; Kachko et al. 2014; Patten et al.
2012) and specifically that of amitriptyline for the treatment of chronic migraine
and functional abdominal pain in children and adolescents (Kaminski et al. 2011;
Powers et al. 2013). However, no further regulatory approvals have been issued
mainly because of the limited sample size and the low methodological quality of
these studies.
13.2.2.2 Safety of Antidepressants in Children and adolescents
Suicidal Behaviors and Use of Antidepressants
Since 2003, concerns have been raised about the possible risk of increased suicidal
behaviors in children and adolescents with depression treated with antidepressant
agents. In 2004, the FDA and many other national regulatory agencies issued warn-
ings concerning the risk of suicidal behavior associated with the use of antidepres-
sants in children and adolescents. Indeed, a meta-analysis conducted by the FDA
showed an increase in the risk of suicidal behaviors and ideations among young
patients treated with antidepressants for psychiatric indications as compared with
placebo-treated patients (OR = 2.22 [1.40–3.40]) (Friedman and Leon 2007).
Another meta-analysis conducted by the Cochrane Collaboration also found an
increased risk of suicide attempts and suicidal ideations of (OR = 1.80 [1.19–2.72])
in depressed children and adolescents treated with SSRIs versus placebo (Hetrick et
al. 2007).
Regulatory warnings do not distinguish between the different antidepressant
agents. Nevertheless, drug characteristics in terms of receptor selectivity and elimi-
nation half-life are different. In this respect, a systematic review of published and
unpublished data suggests that fluoxetine presents a favorable risk-benefit profile,
paroxetine and sertraline present an equivocal or weak risk-benefit profile, and ven-
lafaxine and citalopram present an unfavorable risk-benefit profile in childhood
depression (Whittington et al. 2004).
Furthermore, among 102,647 US depressed children and young adults aged
10–24 years who initiated therapy with antidepressants, similar rates of deliberate
self-harm were observed with either SSRI or SNRI agents (Miller et al. 2014a).
However, the rate of deliberate self-harm among patients who initiated therapy with
antidepressants at high therapeutic doses was approximately twice as high as among
matched patients who initiated modal therapeutic doses (hazard ratio (HR) = 2.2
[1.6–3.0]) (Miller et al. 2014b). A 2012 Cochrane meta-analysis found evidence of
an increased risk of suicide-related outcome in children and adolescents treated
with antidepressants compared to placebo (RR 1.58 [1.02–2.45]). Nevertheless,
there was no evidence that the magnitude of the effect (compared with placebo) was
modified by individual antidepressant class (Hetrick et al. 2012). It was also found
that initiating antidepressants at higher therapeutic doses increased the risk of delib-
erate self-harm in children and adolescents. Finally, some studies demonstrated a
similar risk of suicidal acts for SSRIs and TCAs (RR = 0.92 [0.43–2.00])
(Schneeweiss et al. 2010).
Following regulatory warnings about antidepressants’ use, the rate of depression
diagnoses and prescription of antidepressant drugs decreased in many countries
(Gibbons et al. 2007; Kurdyak et al. 2007; Murray et al. 2005). Thus, the risk of
“doing nothing” progressively became a concern and a matter of debate (Friedman
2014; Stone 2014). In fact, some epidemiological studies have shown that the sui-
cide rate in adolescents was inversely associated with the rate of antidepressants’
prescriptions (Friedman 2014; Gibbons et al. 2006; Stone 2014).
Other Psychiatric Side Effects
Other psychiatric side effects related to the use of antidepressants in children and
adolescents may be grouped into (1) mania spectrum (mania, hypomania, elevated
mood), (2) depression spectrum (aggravation of depression, irritability anger, hyper-
sensitivity), (3) agitation spectrum (agitation, akathisia, restlessness, nervousness,
hyperactivity), and (4) anxiety symptoms (Emslie et al. 2006; Gordon and Melvin
2013).
The related manic switching in vulnerable adolescents treated with antidepres-
sants is estimated to be 5.4 % (Martin et al. 2004). Randomized trials have sug-
gested that the risk is less than 2 % in the short and medium term after drug initiation
(Cheung et al. 2005). Simple mania spectrum symptoms are globally more preva-
lent than hypomania. Retrospective studies have found that fluoxetine can cause
irritability and hypomania-like symptoms and sertraline “behavioral activation”
(Gordon and Melvin 2013). In this context, it is advised to monitor patients closely
for emergent suicidality, hostility, agitation, and mania.
Neurological Side Effects
An increased risk of seizure has been reported with the use of clomipramine during
premarketing evaluation. This risk was related to either the dose of clomipramine or
the duration of the treatment or both. Seizures have also been associated with the
use SSRIs in children and adolescents (Cheung et al. 2005). Therefore, caution is
advised when clomipramine or SSRIs are prescribed to patients with a history of
seizures or other factors that may predispose to seizures.
Metabolic and Endocrine Side Effects
Currently, there are no data on endocrine adverse effects and the use of tricyclic
agents. However, concerns have risen about the endocrine side effects of SSRI
exposure in children and adolescents although studies remain scarce. Preclinical
studies found nonreversible testicular degeneration with histologic abnormalities in
young rats exposed to high levels of SSRIs (Monteiro Filho et al. 2014; Schmidt et
al. 1988). However, these studies were performed with very high exposures (20-fold
higher than clinical pediatric exposure) in animals. No data exist on fertility of
adults who had been treated in their childhood with antidepressants (Prozac Product
monograph; Schmidt et al. 1988).
Growth delay has also been cited as a potential SSRI side effect. Four cases of
children aged from 9 to 13 years treated with SSRIs for obsessive-compulsive dis-
order and Tourette syndrome had experienced growth delay with decreased growth
hormone (GH) levels, without any other anomaly on the gonadotropic axis. Authors
suggested that suppression of GH secretion may occur during therapy with SSRIs
and emphasize the need for further larger studies (Weintrob et al. 2002). An effect
on adult size has not been established.
In a retrospective cohort study including 11,970 children and adolescents under
antidepressant therapy, the presence of obesity/weight gain, type 2 diabetes melli-
tus, and dyslipidemia was more common in those under SSRIs compared to a ran-
dom sample of 4,500 children nontreated with psychotropic medications (OR = 1.49;
1.37; 1.44) (Jerrell 2010).
Effects of SSRIs on prolactin levels have been reported in animals and adult
patients (Muench and Hamer 2010; Peterson 2001), but in children and adolescents,
only one study has associated use of SSRIs and hyperprolactinemia in children and
adolescents (Jerrell 2010). In this study, reproductive and/or sexual adverse events
were not associated with the use of SSRIs.
Thus, despite information included in the summary of product characteristics of

SSRIs excluding the possibility of a pubertal delay, it is recommended to monitor
growth and pubertal development (size, weight, Tanner level) during and after treat-
ment with an SSRI agent. If a growth or pubertal delay is observed, it is strongly
advised to refer to a pediatric endocrinologist.
Serotonin Syndrome
Several case reports describe the occurrence of serotonin syndrome in children and
adolescents treated with antidepressant agents. Serotonin syndrome is associated
with hyperthermia, hypertension, headache, flushing, shaking, nausea, anxiety, agi-
tation, confusion, hallucination, or insomnia and is probably related to an excess of
serotonin. Nevertheless, this adverse event seems to be extremely rare (Ghanizadeh
2013; Boyer and Shannon 2005).
Risk of Cardiac Arrhythmias with Fluoxetine
In 2013, the FDA issued a communication to notify patients and health-care pro-
viders about the potential risk of cardiac arrhythmias in patients using fluoxetine.
This was related to the reporting of post-marketing cases of QT interval prolon-
gation and ventricular arrhythmias. Therefore, the use of fluoxetine is not recom-
mended in patients with underlying cardiovascular conditions, hypokalemia/
hypomagnesemia and those under medications that might predispose to prolon-
gation of the QT interval and occurrence of torsade de pointes. In addition, fluox-
etine should be used cautiously in patients with recent myocardial infarction,
uncompensated heart failure, congenital QT syndrome, drug overdose risk fac-
tors such as hepatic impairment, and those presenting a CYP2D6 poor metabo-
lizer status and/or concurrent use of CYP2D6 inhibitors or other highly bound
drugs.
Other Side Effects
The TADS study was conducted by the National Institute of Mental Health (NIMH)
to evaluate the efficacy and safety of antidepressants in depressed children and ado-
lescents. Patients were randomized in four treatment groups: fluoxetine, cognitive-
behavioral therapy, combination treatment (fluoxetine and cognitive-behavioral
therapy), and placebo for 12 weeks (Emslie et al. 2006). Sedation, insomnia, vomit-
ing, and upper abdominal pain were reported in less than 5 % of subjects treated
with fluoxetine, and headache was the most frequent side effect appearing in 11.9 %
of the patients in this treatment group. Moreover, in a randomized controlled trial
(RCT) evaluating the efficacy and safety of sertraline in depressed children and
adolescents, the most common side effects were fatigue, insomnia, restlessness,
headache, gastric distress, sore throat, and yawning (Melvin et al. 2006; Walkup et
al. 2008). In both trials, the probability of treatment discontinuation due to adverse
effects was relatively small (5–10 %) (Cheung et al. 2005) and declined over time
(Emslie et al. 2006).
With regard to tricyclic agents, a Cochrane Collaboration meta-analysis found a
higher incidence of vertigo (RR = 2.76 [1.73–4.43]), orthostatic hypotension
(RR = 4.86 [1.69–13.97]), tremor (RR = 5.43 [1.64–17.98]), and dry mouth
(RR = 3.35 [1.98–5.64]) in children and adolescents treated with TCAs compared to
those treated with placebo (Hazell and Mirzaie 2013).
13.2.3 Antipsychotics
13.2.3.1 Use of Antipsychotics in Children and Adolescents
Antipsychotic medications are clearly effective in treating several psychiatric con-

ditions in children and adolescents. However, as in adults, prescription of antipsy-
chotics involves a difficult balance between the need to relieve mental disease
symptoms and the risk of drug-induced toxicity (Masi and Liboni 2011).
Antipsychotics are classified according to the timeline of their development,
their pharmacology, and their safety profiles. First-generation or “typical” antipsy-
chotics (FGAs) act therapeutically by antagonizing dopamine D2 receptors. Their
safety profile, in particular neurological toxicity, depends primarily on their relative
potency in binding to dopamine D2 receptors. Second-generation or “atypical” anti-
psychotics (SGAs), launched in the 1990s, bind to other neuroreceptors than only
dopamine D2, and a newer antipsychotic, aripiprazole, has been developed which is
a partial agonist of dopamine D2 and serotonin 5-HT5 receptors. “Atypical” anti-
psychotics have been initially perceived as more effective and safer than “typical”
antipsychotics.
Hence, over the past three decades, SGAs have dominated prescribing prefer-
ences worldwide (Gardner et al. 2005; Muench and Hamer 2010; Verdoux et al.
2010). Indeed, the increase in antipsychotics’ use that has been consistently
reported in children and adolescents over the years is mainly attributed to an
increment in the prescription of SGAs (Cooper et al. 2006; Kalverdijk et al. 2008;
Olfson et al. 2006; Rani et al. 2008; Ronsley et al. 2013). Still, a wide variability
across countries has been observed with atypical antipsychotic subclass repre-
senting 66 % of antipsychotic use in youth in the USA and 48 % in the Netherlands
but only 5 % in Germany (data from year 2000) (Zito et al. 2008). The duration of
treatment with these agents has also been increasing (Kalverdijk et al. 2008; Rani
et al. 2008).
Currently, several FGAs and SGAs have a marketing authorization for use in the
pediatric population with certain differences between the USA and Europe
(Table 13.1). However, the use of antipsychotics in children and adolescents is
principally “off label,” that is for indications that are not approved by regulatory
Table 13.1 Approved indications of commonly used antipsychotics for pediatric patients
according to regulatory agency
FDA EMA
approval approval
patient age patient age
Antipsychotic drug Indication (years) (years)
FGAs Haloperidola Schizophrenia – ≥3b
Behavioral disorders (hyperactivity, – ≥3b
aggression)
Gilles de la Tourette syndrome – ≥3b
Chlorpromazine Childhood schizophrenia 1–12 ≥1b
Bipolar disorder (mania) 1–12 –
Autism – ≥1b
SGAs Clozapine Schizophrenia in patients – >16
unresponsive or intolerant to other
antipsychotics
Risperidone Schizophrenia 13–17 ≥15
Bipolar I disorder 10–17 ≥13
Irritability associated with autistic 5–16 –
disorder
Persistent aggression in conduct – ≥5
disorderc
Olanzapine Schizophrenia 13–17 –
Bipolar I disorder 13–17 –
Quetiapine Schizophrenia 13–17 –
Bipolar I disorder 10–17 –
Aripiprazole Schizophrenia 13–17 ≥15
Bipolar I disorderd 10–17 ≥13
Irritability associated with autistic 6–17 –
disorder
Dashes mean that the product has no pediatric approval in the specific indication. Ziprasidone has
no approval for pediatric use in the USA or the EU
a
Haloperidol solution for injection is the only form available in the USA, but in European coun-
tries, both injectable (5 mg/ml) and oral (2 mg/ml) solutions are marketed. Only oral haloperidol
is indicated in children.
b
Approved age ranges provided for haloperidol and chlorpromazine are based on current UK mar-
keting authorizations. There are no centralized EMA authorizations for these molecules in chil-
dren.
c
Short-term symptomatic treatment (up to 6 weeks) of persistent aggression in conduct disorder in
children from the age of 5 years and adolescents with subaverage intellectual functioning or mental
retardation diagnosed according to DSM-IV criteria, in whom the severity of aggressive or other
disruptive behaviors requires pharmacological treatment
d
As monotherapy or as an adjunct to lithium or valproate
authorities or used in children who are younger than the approved age ranges. “Off-
label” indications include mainly attention deficit hyperactivity disorder (ADHD),
conduct and behavioral disturbances (aggression, self-injury, disruptive behavior,
etc.), and mood disorders (Penfold et al. 2013; Baribeau and Anagnostou 2014).
13.2.3.2 Safety of Antipsychotics in Children and Adolescents
Because of the paucity of comparative studies in children, there is currently little

evidence to support the superiority of SGAs over FGAs or a difference between
SGAs for the treatment of children and adolescents with psychosis or behavioral
disorders (Kennedy et al. 2007; Kumra et al. 2008; Sikich 2008; Sikich et al. 2008;
Loy et al. 2012; Kumar et al. 2013). Conversely, the safety profile of antipsychotics
differs substantially. Atypical antipsychotics are associated with a lower risk of neu-
rological adverse reactions than first-generation drugs, but they are clearly associ-
ated with a higher risk of weight gain and metabolic abnormalities in both adults
and children (Caccia 2013; Maher et al. 2011; Seida et al. 2012). Nevertheless and
despite initial expectations, observed variability appears to be greater among spe-
cific agents than between the first- and second-generation antipsychotic classes
(Masi and Liboni 2011; Muench and Hamer 2010; Vitiello et al. 2009; Fraguas et al.
2011).
Current knowledge on the most prevalent adverse reactions to antipsychotics in
children and adolescents is provided in the following sections. Since the use of typi-
cal antipsychotics is actually limited in children and adolescents, safety issues are
exposed mainly with regard to second-generation compounds. The strength of asso-
ciation between a given agent and a specific adverse event is given as compared to
placebo since very few head-to-head comparisons of antipsychotics have been con-
ducted in children and adolescents. Occurrence of neuroleptic malignant syndrome,
a rare but potentially life-threatening adverse reaction, is not detailed in this section
(see Chap. 10).
Neurological Toxicity
Sedation and somnolence are common with all antipsychotic medications, and they
are usually dose dependent. Rates tend to be higher with FGAs, clozapine, ziprasi-
done, olanzapine, and risperidone and lower with aripiprazole and quetiapine
(Cohen et al. 2012).
Another major neurological adverse reaction is the occurrence of extrapyramidal
symptoms (EPS) which comprises drug-induced parkinsonism, akathisia, acute
dystonia, and tardive dyskinesia. Children and adolescents appear to be more sensi-
tive to EPS than adults especially when they present mental retardation or CNS
damage or are drug-naïve patients. EPS are more common with typical antipsychot-
ics like haloperidol, but newer antipsychotics are not totally free of such reactions.
Treatment with risperidone, olanzapine, and aripiprazole are related to an elevated
risk of EPS, especially at high doses. In fact, the incidence of EPS in schizophrenic
patients treated with aripiprazole monotherapy compared to placebo treatment is
much higher in pediatric patients compared to adults (Correll 2008). The risk of
EPS appears to be low with clozapine and quetiapine. Data on ziprasidone are too
scarce to draw reliable conclusions. The majority of extrapyramidal symptoms are
reversible after discontinuation of the offending agent with the exceptions of tardive
dystonia and tardive dyskinesia that may be difficult to treat and ultimately become
permanent in some patients. Moreover, EPS often generate major psychological
distress and poor compliance with therapy.
The majority of antipsychotics can lower the seizure threshold and should be
used with caution in patients who have a history of seizures and in those with
organic brain disorders. Convulsions are more common under treatment by certain
FGAs and clozapine, especially at high doses (Masi and Liboni 2011).
Finally, for aripiprazole, a safety warning was issued on the potential risk of
increased suicidal ideation and suicide in children, adolescents, and young adults
with major depressive disorder. The possibility of a suicide attempt is inherent in
psychotic illness, and close supervision of high-risk patients should accompany
drug therapy. Nevertheless, no suicides occurred in any of the pediatric trials.
Metabolic Disorders
All antipsychotics may be associated with weight gain and an increase of body
mass index (BMI) though such effects are probably the most significant adverse
reactions of atypical agents (Muench and Hamer 2010). Excessive weight gain
should not be disregarded in treated children and adolescents because it may con-
tribute to significant morbidity and mortality in adulthood. Being overweight is a
major determinant of a general metabolic disorder, the metabolic syndrome (obe-
sity, hypertriglyceridemia, low high-density lipoprotein cholesterol levels, hyper-
tension, and hyperglycemia) associated with atherosclerosis, coronary artery
disease, and colorectal cancer in adults (Correll et al. 2006). Also, in long-term use,
weight gain has been associated with liver enzyme abnormalities and fatty infiltra-
tion (Masi and Liboni 2011). In addition, it contributes to poor medication adher-
ence, social withdrawal, and low self-esteem which may be sources of significant
psychological morbidity. Mean weight gain during therapy appears to be important
with olanzapine, clozapine, and risperidone; moderate with quetiapine; and low
with aripiprazole and ziprasidone. Despite great interindividual variability in
weight gain, possibly related to genetic predisposition, dietary recommendations
and counseling (lifestyle, exercise) should be provided at the initiation of antipsy-
chotic therapy for all patients. Treatment with metformin to stabilize weight has
been proposed, but data are limited in children and adolescents (Klein et al. 2006;
Shin et al. 2009).
Increase in the blood levels of glucose, triglycerides, and cholesterol may also be
attributable to antipsychotic agents. Glycemic abnormalities may vary from hyper-
glycemia due to mild insulin resistance to new-onset diabetes and worsening of
glycemic control in patients with preexisting diabetes mellitus (Pringsheim et al.
2011a). Incidence is not well established in children and adolescents under treat-
ment, but the risk appears to be high with olanzapine and risperidone, moderate with
quetiapine and aripiprazole, and low with ziprasidone and the first-generation agent
haloperidol. Moreover, treatment with olanzapine, clozapine, and quetiapine has
been associated with increased blood levels of triglycerides and cholesterol, whereas
the risk is moderate with aripiprazole and low with risperidone, ziprasidone, and
haloperidol. Adolescents treated with olanzapine have an increased potential for

weight gain and hyperlipidemia compared to adult patients. Similar to weight gain,
drug-induced metabolic changes may persist over time and become clinically mean-
ingful only after prolonged use of antipsychotic medication. In the absence of long-
term follow-up studies, cardiometabolic adverse effects are probably underestimated
in children and adolescents.
Endocrine Disorders
Hyperprolactinemia is a direct consequence of the antagonism of dopamine D2

receptors by both first- and second-generation antipsychotic agents. In most cases,
hyperprolactinemia is dose dependent, tends to normalize with time, and completely
resolves after cessation of antipsychotic treatment. Clinical signs of hyperprolac-
tinemia include amenorrhea and other menstrual cycle disorders, galactorrhea, hir-
sutism, and sexual disturbances (decreased libido, erectile difficulties, etc.).
Nevertheless, hyperprolactinemia may persist throughout treatment and remain
totally asymptomatic or cause clinical signs that are difficult to express for a child
or an adolescent (e.g., sexual dysfunction). Additional long-term consequences of
prolonged hyperprolactinemia and subsequent hypogonadism include pubertal
delay, growth arrest, and osteoporosis (Masi and Liboni 2011; Vitiello et al. 2009).
Risperidone and olanzapine tend to favor meaningful increase in prolactin secre-
tion, whereas the effect of quetiapine and ziprasidone is moderate and aripiprazole
may even lower levels of prolactin.
Hematological Toxicity
Antipsychotic agents may cause neutropenia mainly in patients presenting low

baseline blood counts or using cytotoxic concomitant therapy. However, a risk of
life-threatening agranulocytosis has been reported in patients treated with clozap-
ine. Neutrophil blood counts generally normalize after discontinuation of clozapine
(Masi and Liboni 2011).
Cardiovascular Toxicity
Orthostatic hypotension and tachycardia have been described with the use of anti-
psychotic medications. These effects are less common in children and adolescents
than in elderly patients and in most cases are clinically irrelevant (Masi and Liboni
2011). Cardiovascular effects are more frequent with certain FGAs, such as chlor-
promazine and clozapine, though they have also been observed with the use of que-
tiapine and risperidone especially with rapid uptitration of dose. In fact, quetiapine
has been associated with hypertension in children, whereas this event was never
observed in treated adults. Moreover, myocarditis occurring at the beginning of
treatment has been reported with clozapine (Ronaldson et al. 2010).
A potentially serious adverse event related with antipsychotics’ use is prolon-

gation of ventricular repolarization that may lead to ventricular arrhythmia (e.g.,
torsades de pointes) and ultimately to sudden cardiac death (SCD). The incidence
of SCD in adults treated with antipsychotics is twice that of the general popula-
tion (Muench and Hamer 2010). In children, only one such case has been described
in a child treated with ziprasidone. However, subclinical QTc prolongation has
been reported in children under ziprasidone (Caccia 2013; Correll 2008).
Subsequently, cautious use of all agents is recommended in children especially
those with existing cardiac disease or family history of QT prolongation (Germanò
et al. 2014).
Overall, children and adolescents tend to be more sensitive to the toxicity of

antipsychotics than adult patients, especially when polypharmacotherapy is used.
They experience more frequently weight gain, metabolic abnormalities (hypergly-
cemia, dyslipidemia), hyperprolactinemia, and neurological adverse reactions
(sedation/somnolence, extrapyramidal symptoms, seizures). Conversely, they
appear to be less susceptible to cardiovascular adverse reactions.
To date, safety data derive almost exclusively from cohort studies and meta-
analyses of randomized controlled trials (Caccia 2013; Cohen et al. 2012; Pringsheim
et al. 2011a). Yet, these studies were limited in follow-up time and sample size and
therefore cannot provide reliable conclusions especially with regard to insulin/gly-
cemic control, metabolic disorders, and occurrence of rare and distant adverse
events. Moreover, the impact of antipsychotic medications on the physical, mental,
and neurological development of children and adolescents, a safety issue specific to
this population, has not been evaluated yet. In recent years, several networks have
been created to routinely assess the efficacy and safety of antipsychotics in children:
The Canadian Alliance for Monitoring Effectiveness and Safety of Antipsychotics
in Children (CAMESA) in Canada (http://camesaguideline.org/), the Pediatric
Atypical Antipsychotic monitoring Safety (PAMS) Study in the UK (Rani et al.
2009), and the SafEty of NeurolepTics in Infancy and Adolescence (SENTIA) in
Spain (http://SENTIA.es). Analyses of large population health databases and patient
registries through collaborative projects are actually required to clarify antipsychot-
ics’ safety profile and to define their distal risk-benefit ratio in children and adoles-
cents (Rani et al. 2011).
13.2.3.3 Safety Monitoring of Children and Adolescents Treated with

Antipsychotics
Antipsychotic medications present highly variable safety profiles. This may compli-
cate prescribing and patient management; however it offers many therapeutic alter-
natives and, thus, the possibility to match patients with the most appropriate
medication. Given the length of use of antipsychotics and the impact of patients’
family history or lifestyle on the choice of the antipsychotic agent, thorough evalu-
ation of risks for each distinct patient seems appropriate.
Currently, clozapine and olanzapine are not considered to be first-line antipsy-

chotic treatments in children and adolescents because of their unfavorable safety
profile. In particular, clozapine is reserved to patients who are resistant to other
agents. Risperidone is actually the most commonly prescribed (Patten et al. 2012)
and one of the most extensively evaluated antipsychotic agent in the pediatric popu-
lation (Pringsheim et al. 2011a). Nevertheless, its safety profile is far from ideal as
risperidone is frequently associated with rapid weight gain, metabolic abnormali-
ties, significant somnolence and sedation, and the occurrence of extrapyramidal
neurological symptoms. More recently developed agents such as aripiprazole, que-
tiapine, and ziprasidone offer interesting alternatives; however their efficacy and
safety need to be further evaluated in children and adolescents.
Regardless of the drug used, children and adolescents that receive antipsychotic
drugs should be proactively monitored for side effects. Physical examination
(weight, height, BMI, waist circumference, blood pressure, and neurological exami-
nation for EPS) and certain standard laboratory tests (fasting plasma glucose, total
cholesterol, LDL and HDL cholesterol, triglycerides, aspartate and alanine amino-
transferase, prolactin) should be systematically performed at least in the first month
after treatment initiation. More specific laboratory tests should be carried out
according to the specific risk of treating agents: insulin levels in olanzapine treat-
ment, thyroid-stimulating hormone (TSH) in quetiapine treatment, and neutrophil
counts in clozapine treatment. Guidance on the electrocardiogram monitoring of
treated patients has also been developed (Blair et al. 2004).
Evidence-based recommendations for monitoring of children and adolescents
under atypical antipsychotics have been developed by the CAMESA guideline
group in Canada. These recommendations are extremely useful in practice; however
they reflect the paucity of evidence-based knowledge in the area and the need for
further targeted research. In addition, they are limited to the first year of treatment
because of the absence of long-term prospective studies. Safety of antipsychotics is
a major focus especially in the young; however monitoring rates are still very low
and studies of pharmacologic and behavioral interventions are extremely limited
(Maayan and Correll 2011).
13.2.4 Other Psychotropic Medications
Anxiolytics and sedatives have no marketing authorization for use in child and ado-
lescent psychiatry except for certain drugs (e.g., diazepam, hydroxyzine) in certain
countries and under exceptional clinical conditions. Specific safety data on children
and adolescents are extremely limited, although the use in children can be very
prevalent in some countries (Murray et al. 2004; Pringsheim et al. 2011b; Zito et al.
2008).
Mood regulators such as lithium and anticonvulsants, carbamazepine and val-
proic acid, are also used beyond their marketing authorizations in pediatric psy-
chiatric indications. The safety of lithium in children and adolescents with bipolar
I disorder has been evaluated in clinical trials of limited sample size (Geller et al.
2012), and adverse events are similar to those observed in adult patients (e.g.,
thyroid dysfunction). Toxicity of anticonvulsants used mainly for neurological
indications has been extensively evaluated and is beyond the scope of this
chapter.
13.3 Conclusion
The majority of psychotropic drugs provide only symptomatic management of

clinical symptoms in lifelong psychiatric disorders. Children and adolescents
are therefore exposed to these agents in a chronic manner or at least for a pro-
longed period of time. In fact, drug utilization studies have demonstrated not
only that the incidence of the use of certain psychotropic drugs such as antipsy-
chotics is increasing in children and adolescents but also that the duration of the
therapy (Patten et al. 2012). In addition, young children are more likely to be
receiving multiple psychotropic medications to control psychiatric comorbidi-
ties and improve social functioning. Currently, 60–70 % of pharmacological
prescriptions in pediatric psychiatry are considered “off label” because they
concern age ranges, doses, and indications that are not approved by regulatory
authorities. For example, antipsychotic use in children and adolescents targets
mainly nonpsychotic disorders. All these trends in psychotropic prescribing
have significant implications for drug safety and patient monitoring during treat-
ment (Steinhausen 2014). Furthermore, there is a significant rise of prescriptions
of psychotropic drugs by family physicians and pediatricians (Ronsley et al.
2013; Meng et al. 2014) that may lack professional experience and resources to
monitor children for potential side effects. Physicians who do not have sufficient
knowledge and resources to carefully follow patients and assess drug safety
should refrain from prescribing these medications and refer the patient to a
specialist.
The substantial increase of psychotropic use in children and adolescents has not
been accompanied by a similar increase in research-based evidence about the effi-
cacy and safety of these medications. This increase in psychotropic drug use calls
for long-term efficacy and, most importantly, safety studies in large samples in
order to account for the dynamic processes of growth and brain development in the
young patients. Further research should also attempt to elucidate the physiopatho-
logical mechanisms of psychotropic-induced toxicity and the potential value of per-
sonalized approaches based on genetics and neurobiology. Additional safety data
are urgently needed to clarify the risk-benefit ratio of psychotropic medications in
children and adolescents and to adequately guide medical decision-making. In the
meantime, psychotropic prescribing in children and adolescents should involve a
cautious balance between patients’ therapeutic needs and a proactive monitoring of
clinical efficacy and drug-induced toxicity.
References
Awudu GA, Besag FM (2014) Cardiovascular effects of methylphenidate, amphetamines and

atomoxetine in the treatment of attention-deficit hyperactivity disorder: an update. Drug Saf
37(9):661–676. doi:10.1007/s40264-014-0201-8
Bange F, Le Heuzey MF, Acquaviva E, Delorme R, Mouren MC (2014) Cardiovascular risks and
management during attention deficit hyperactivity disorder treatment with methylphenidate.
Arch Pediatr 21(1):108–112. doi:10.1016/j.arcped.2013.11.001
Baribeau DA, Anagnostou E (2014) An update on medication management of behavioral disorders
in autism. Curr Psychiatry Rep 16(3):437. doi:10.1007/s11920-014-0437-0
Biederman J, Spencer TJ, Monuteaux MC, Faraone SV (2010) A naturalistic 10-year prospective
study of height and weight in children with attention-deficit hyperactivity disorder grown up:
sex and treatment effects. J Pediatr 157(4):635–640, 640 e631. doi:10.1016/j.jpeds.
2010.04.025
Blair J, Taggart B, Martin A (2004) Electrocardiographic safety profile and monitoring guidelines
in pediatric psychopharmacology. J Neural Transm 111(7):791–815. doi:10.1007/s00702-004-
0153-8
Boyer EW, Shannon M (2005) The serotonin syndrome. N Engl J Med 352(11):1112–1120.
doi:10.1056/NEJMra041867
Caccia S (2013) Safety and pharmacokinetics of atypical antipsychotics in children and adoles-
cents. Paediatr Drugs 15(3):217–233. doi:10.1007/s40272-013-0024-6
Chen Q, Sjolander A, Runeson B, D’Onofrio BM, Lichtenstein P, Larsson H (2014) Drug treat-
ment for attention-deficit/hyperactivity disorder and suicidal behaviour: register based study.
BMJ 348:g3769. doi:10.1136/bmj.g3769
Cheung AH, Emslie GJ, Mayes TL (2005) Review of the efficacy and safety of antidepressants in
youthdepression.JChildPsycholPsychiatry46(7):735–754.doi:10.1111/j.1469-7610.2005.01467.x
Cohen D, Bonnot O, Bodeau N, Consoli A, Laurent C (2012) Adverse effects of second-generation
antipsychotics in children and adolescents: a Bayesian meta-analysis. J Clin Psychopharmacol
32(3):309–316. doi:10.1097/JCP.0b013e3182549259
Cooper WO, Arbogast PG, Ding H, Hickson GB, Fuchs DC, Ray WA (2006) Trends in prescribing
of antipsychotic medications for US children. Ambul Pediatr: Off J Ambul Pediatr Assoc
6(2):79–83. doi:10.1016/j.ambp.2005.11.002
Cooper WO, Habel LA, Sox CM, Chan KA, Arbogast PG, Cheetham TC, Murray KT, Quinn VP,
Stein CM, Callahan ST, Fireman BH, Fish FA, Kirshner HS, O’Duffy A, Connell FA, Ray WA
(2011) ADHD drugs and serious cardiovascular events in children and young adults. N Engl J
Med 365(20):1896–1904. doi:10.1056/NEJMoa1110212
Correll CU (2008) Antipsychotic use in children and adolescents: minimizing adverse effects to
maximize outcomes. J Am Acad Child Adolesc Psychiatry 47(1):9–20. doi:10.1097/
chi.0b013e31815b5cb1
Correll CU, Frederickson AM, Kane JM, Manu P (2006) Metabolic syndrome and the risk of coro-
nary heart disease in 367 patients treated with second-generation antipsychotic drugs. J Clin
Cortese S, Holtmann M, Banaschewski T, Buitelaar J, Coghill D, Danckaerts M, Dittmann RW,
Graham J, Taylor E, Sergeant J (2013) Practitioner review: current best practice in the manage-
ment of adverse events during treatment with ADHD medications in children and adolescents.
J Child Psychol Psychiatry 54(3):227–246. doi:10.1111/jcpp.12036
Donner R, Michaels MA, Ambrosini PJ (2007) Cardiovascular effects of mixed amphetamine salts
extended release in the treatment of school-aged children with attention-deficit/hyperactivity
disorder. Biol Psychiatry 61(5):706–712. doi:10.1016/j.biopsych.2006.05.002
Efron D, Jarman F, Barker M (1997) Side effects of methylphenidate and dexamphetamine in
children with attention deficit hyperactivity disorder: a double-blind, crossover trial. Pediatrics
100(4):662–666
Emslie G, Kratochvil C, Vitiello B, Silva S, Mayes T, McNulty S, Weller E, Waslick B, Casat C,

Walkup J, Pathak S, Rohde P, Posner K, March J (2006) Treatment for Adolescents with
Depression Study (TADS): safety results. J Am Acad Child Adolesc Psychiatry 45(12):1440–
1455. doi:10.1097/01.chi.0000240840.63737.1d
Faraone SV, Biederman J, Morley CP, Spencer TJ (2008) Effect of stimulants on height and
weight: a review of the literature. J Am Acad Child Adolesc Psychiatry 47(9):994–1009.
doi:10.1097/CHI.ObO13e31817eOea7
Fraguas D, Correll CU, Merchan-Naranjo J, Rapado-Castro M, Parellada M, Moreno C, Arango C
(2011) Efficacy and safety of second-generation antipsychotics in children and adolescents
with psychotic and bipolar spectrum disorders: comprehensive review of prospective head-to-
head and placebo-controlled comparisons. Eur Neuropsychopharmacol: J Eur Coll
Neuropsychopharmacol 21(8):621–645. doi:10.1016/j.euroneuro.2010.07.002
Friedman RA (2014) Antidepressants’ black-box warning – 10 years later. N Engl J Med
371(18):1666–1668. doi:10.1056/NEJMp1408480
Friedman RA, Leon AC (2007) Expanding the black box – depression, antidepressants, and the
risk of suicide. N Engl J Med 356(23):2343–2346. doi:10.1056/NEJMp078015
Gardner DM, Baldessarini RJ, Waraich P (2005) Modern antipsychotic drugs: a critical overview.
CMAJ: Can Med Assoc J = J Assoc Med Can 172(13):1703–1711. doi:10.1503/cmaj.1041064
Geller B, Luby JL, Joshi P, Wagner KD, Emslie G, Walkup JT, Axelson DA, Bolhofner K, Robb
A, Wolf DV, Riddle MA, Birmaher B, Nusrat N, Ryan ND, Vitiello B, Tillman R, Lavori P
(2012) A randomized controlled trial of risperidone, lithium, or divalproex sodium for initial
treatment of bipolar I disorder, manic or mixed phase, in children and adolescents. Arch Gen
Psychiatry 69(5):515–528. doi:10.1001/archgenpsychiatry.2011.1508
Germanò E, Italiano D, Lamberti M, Guerriero L, Privitera C, D’Amico G, Siracusano R, Ingrassia
M, Spina E, Calabrò MP, Gagliano A (2014) ECG parameters in children and adolescents
treated with aripiprazole and risperidone. Prog Neuropsychopharmacol Biol Psychiatry
51(0):23–27, doi:http://dx.doi.org/10.1016/j.pnpbp.2013.10.020
Ghanizadeh A (2013) A systematic review of the efficacy and safety of desipramine for treating
ADHD. Curr Drug Saf 8(3):169–174
Gibbons RD, Hur K, Bhaumik DK, Mann JJ (2006) The relationship between antidepressant pre-
scription rates and rate of early adolescent suicide. Am J Psychiatry 163(11):1898–1904.
doi:10.1176/ajp.2006.163.11.1898
Gibbons RD, Brown CH, Hur K, Marcus SM, Bhaumik DK, Erkens JA, Herings RM, Mann JJ
(2007) Early evidence on the effects of regulators’ suicidality warnings on SSRI prescriptions
and suicide in children and adolescents. Am J Psychiatry 164(9):1356–1363. doi:10.1176/appi.
ajp.2007.07030454
Goldstein BI (2013) Do stimulants prevent substance use and misuse among youth with attention-
deficit/hyperactivity disorder? The answer is still maybe. J Am Acad Child Adolesc Psychiatry
52(3):225–227. doi:10.1016/j.jaac.2012.12.016
Gordon M, Melvin G (2013) Selective serotonin re-uptake inhibitors – a review of the side effects
in adolescents. Aust Fam Physician 42(9):620–623
Gould MS, Walsh BT, Munfakh JL, Kleinman M, Duan N, Olfson M, Greenhill L, Cooper T
(2009) Sudden death and use of stimulant medications in youths. Am J Psychiatry 166(9):
992–1001. doi:10.1176/appi.ajp.2009.09040472
Graham J, Banaschewski T, Buitelaar J, Coghill D, Danckaerts M, Dittmann RW, Dopfner M,
Hamilton R, Hollis C, Holtmann M, Hulpke-Wette M, Lecendreux M, Rosenthal E,
Rothenberger A, Santosh P, Sergeant J, Simonoff E, Sonuga-Barke E, Wong IC, Zuddas A,
Steinhausen HC, Taylor E (2011) European guidelines on managing adverse effects of medica-
tion for ADHD. Eur Child Adolesc Psychiatry 20(1):17–37. doi:10.1007/s00787-010-0140-6
Greenhill LL, Swanson JM, Vitiello B, Davies M, Clevenger W, Wu M, Arnold LE, Abikoff HB,
Bukstein OG, Conners CK, Elliott GR, Hechtman L, Hinshaw SP, Hoza B, Jensen PS, Kraemer
HC, March JS, Newcorn JH, Severe JB, Wells K, Wigal T (2001) Impairment and deportment
responses to different methylphenidate doses in children with ADHD: the MTA titration
trial. J Am Acad Child Adolesc Psychiatry 40(2):180–187. doi:10.1097/00004583-

200102000-00012
Hammerness PG, Perrin JM, Shelley-Abrahamson R, Wilens TE (2011) Cardiovascular risk of
stimulant treatment in pediatric attention-deficit/hyperactivity disorder: update and clinical rec-
ommendations. J Am Acad Child Adolesc Psychiatry 50(10):978–990. doi:10.1016/j.
jaac.2011.07.018
Harrison JN, Cluxton-Keller F, Gross D (2012) Antipsychotic medication prescribing trends in
children and adolescents. J Pediatr Health Care 26(2):139–145. doi:10.1016/j.pedhc.2011.10.009
Hazell P, Mirzaie M (2013) Tricyclic drugs for depression in children and adolescents. Cochrane
Database Syst Rev (6):CD002317. doi:10.1002/14651858.CD002317.pub2
Hetrick S, Merry S, McKenzie J, Sindahl P, Proctor M (2007) Selective serotonin reuptake inhibi-
tors (SSRIs) for depressive disorders in children and adolescents. Cochrane Database Syst Rev
(3):CD004851. doi:10.1002/14651858.CD004851.pub2
Hetrick SE, McKenzie JE, Cox GR, Simmons MB, Merry SN (2012) Newer generation antide-
pressants for depressive disorders in children and adolescents. Cochrane Database Syst Rev
(11):CD004851. doi:10.1002/14651858.CD004851.pub3
Hoffmann F, Glaeske G, Bachmann CJ (2014) Trends in antidepressant prescriptions for children
and adolescents in Germany from 2005 to 2012. Pharmacoepidemiol Drug Saf 23(12):1268–
1272. doi:10.1002/pds.3649
Humphreys KL, Eng T, Lee SS (2013) Stimulant medication and substance use outcomes: a meta-
analysis. JAMA Psychiatry 70(7):740–749. doi:10.1001/jamapsychiatry.2013.1273
Jerrell JM (2010) Neuroendocrine-related adverse events associated with antidepressant treatment
in children and adolescents. CNS Neurosci Ther 16(2):83–90. doi:10.1111/j.1755-5949.
2009.00106.x
Kachko L, Ben Ami S, Lieberman A, Shor R, Tzeitlin E, Efrat R (2014) Neuropathic pain other
than CRPS in children and adolescents: incidence, referral, clinical characteristics, manage-
ment, and clinical outcomes. Paediatr Anaesth 24(6):608–613. doi:10.1111/pan.12375
Kalverdijk LJ, Tobi H, van den Berg PB, Buiskool J, Wagenaar L, Minderaa RB, de Jong-van den
Berg LT (2008) Use of antipsychotic drugs among Dutch youths between 1997 and 2005.
Psychiatr Serv 59(5):554–560. doi:10.1176/appi.ps.59.5.554
Kaminski A, Kamper A, Thaler K, Chapman A, Gartlehner G (2011) Antidepressants for the treat-
ment of abdominal pain-related functional gastrointestinal disorders in children and adoles-
cents. Cochrane Database Syst Rev (7):CD008013. doi:10.1002/14651858.CD008013.pub2
Kennedy E, Kumar A, Datta SS (2007) Antipsychotic medication for childhood-onset schizophre-
nia. Cochrane Database Syst Rev (3):CD004027. doi:10.1002/14651858.CD004027.pub2
Klein DJ, Cottingham EM, Sorter M, Barton BA, Morrison JA (2006) A randomized, double-
blind, placebo-controlled trial of metformin treatment of weight gain associated with initiation
of atypical antipsychotic therapy in children and adolescents. Am J Psychiatry 163(12):
2072–2079. doi:10.1176/ajp.2006.163.12.2072
Koelch M, Prestel A, Singer H, Keller F, Fegert JM, Schlack R, Hoelling H, Knopf H (2009)
Psychotropic medication in children and adolescents in Germany: prevalence, indications, and
psychopathological patterns. J Child Adolesc Psychopharmacol 19(6):765–770. doi:10.1089/
cap.2009.0018
Koneski JA, Casella EB (2010) Attention deficit and hyperactivity disorder in people with epilepsy:
diagnosis and implications to the treatment. Arq Neuropsiquiatr 68(1):107–114, doi:S0004-
282X2010000100023 [pii]
Koneski JA, Casella EB, Agertt F, Ferreira MG (2011) Efficacy and safety of methylphenidate in
treating ADHD symptoms in children and adolescents with uncontrolled seizures: a Brazilian
sample study and literature review. Epilepsy Behav 21(3):228–232. doi:10.1016/j.
yebeh.2011.02.029
Kumar A, Datta SS, Wright SD, Furtado VA, Russell PS (2013) Atypical antipsychotics for psy-
chosis in adolescents. Cochrane Database Syst Rev (10):CD009582. doi:10.1002/14651858.
CD009582.pub2
Kumra S, Kranzler H, Gerbino-Rosen G, Kester HM, De Thomas C, Kafantaris V, Correll CU,

Kane JM (2008) Clozapine and “high-dose” olanzapine in refractory early-onset schizophre-
nia: a 12-week randomized and double-blind comparison. Biol Psychiatry 63(5):524–529.
doi:10.1016/j.biopsych.2007.04.043
Kurdyak PA, Juurlink DN, Mamdani MM (2007) The effect of antidepressant warnings on pre-
scribing trends in Ontario, Canada. Am J Public Health 97(4):750–754. doi:10.2105/
AJPH.2006.087262
Loy JH, Merry SN, Hetrick SE, Stasiak K (2012) Atypical antipsychotics for disruptive behaviour
disorders in children and youths. Cochrane Database Syst Rev (9):CD008559.
doi:10.1002/14651858.CD008559.pub2
Ma J, Lee KV, Stafford RS (2005) Depression treatment during outpatient visits by U.S. children
and adolescents. J Adolesc Health 37(6):434–442. doi:10.1016/j.jadohealth.2005.07.012
Maayan L, Correll CU (2011) Weight gain and metabolic risks associated with antipsychotic medi-
cations in children and adolescents. J Child Adolesc Psychopharmacol 21(6):517–535.
doi:10.1089/cap.2011.0015
Maher AR, Maglione M, Bagley S, Suttorp M, Hu JH, Ewing B, Wang Z, Timmer M, Sultzer D,
Shekelle PG (2011) Efficacy and comparative effectiveness of atypical antipsychotic medica-
tions for off-label uses in adults: a systematic review and meta-analysis. JAMA 306(12):
1359–1369. doi:10.1001/jama.2011.1360
Martin A, Young C, Leckman JF, Mukonoweshuro C, Rosenheck R, Leslie D (2004) Age effects
on antidepressant-induced manic conversion. Arch Pediatr Adolesc Med 158(8):773–780.
doi:10.1001/archpedi.158.8.773
Masi G, Liboni F (2011) Management of schizophrenia in children and adolescents: focus on
pharmacotherapy. Drugs 71(2):179–208. doi:10.2165/11585350-000000000-00000
Melvin GA, Tonge BJ, King NJ, Heyne D, Gordon MS, Klimkeit E (2006) A comparison of
cognitive-behavioral therapy, sertraline, and their combination for adolescent depression. J Am
Acad Child Adolesc Psychiatry 45(10):1151–1161. doi:10.1097/01.chi.0000233157.21925.71
Meng X, D’Arcy C, Tempier R (2014) Long-term trend in pediatric antidepressant use, 1983–
2007: a population-based study. Can J Psychiatry Rev Can Psychiatr 59(2):89–97
Miller M, Pate V, Swanson SA, Azrael D, White A, Sturmer T (2014a) Antidepressant class, age,
and the risk of deliberate self-harm: a propensity score matched cohort study of SSRI and SNRI
users in the USA. CNS Drugs 28(1):79–88. doi:10.1007/s40263-013-0120-8
Miller M, Swanson SA, Azrael D, Pate V, Sturmer T (2014b) Antidepressant dose, age, and the
risk of deliberate self-harm. JAMA Intern Med 174(6):899–909. doi:10.1001/jamainternmed.
2014.1053
Mohatt J, Bennett SM, Walkup JT (2014) Treatment of separation, generalized, and social anxiety
disorders in youths. Am J Psychiatry 171(7):741–748. doi:10.1176/appi.ajp.2014.13101337
Molina BS, Hinshaw SP, Eugene Arnold L, Swanson JM, Pelham WE, Hechtman L, Hoza B,
Epstein JN, Wigal T, Abikoff HB, Greenhill LL, Jensen PS, Wells KC, Vitiello B, Gibbons RD,
Howard A, Houck PR, Hur K, Lu B, Marcus S (2013) Adolescent substance use in the multi-
modal treatment study of attention-deficit/hyperactivity disorder (ADHD) (MTA) as a function
of childhood ADHD, random assignment to childhood treatments, and subsequent medication.
J Am Acad Child Adolesc Psychiatry 52(3):250–263. doi:10.1016/j.jaac.2012.12.014
Monteiro Filho WO, de Torres SM, Amorim MJ, Andrade AJ, de Morais RN, Tenorio BM, da
Silva Junior VA (2014) Fluoxetine induces changes in the testicle and testosterone in adult
male rats exposed via placenta and lactation. Syst Biol Reprod Med 60(5):274–281. doi:10.31
09/19396368.2014.933984
Mosholder AD, Gelperin K, Hammad TA, Phelan K, Johann-Liang R (2009) Hallucinations and
other psychotic symptoms associated with the use of attention-deficit/hyperactivity disorder
drugs in children. Pediatrics 123(2):611–616. doi:10.1542/peds.2008-0185
MTA Cooperative Group (2004) National institute of mental health multimodal treatment study of
ADHD follow-up: 24-month outcomes of treatment strategies for attention-deficit/hyperactivity
disorder. Pediatrics 113(4):754–761
Muench J, Hamer AM (2010) Adverse effects of antipsychotic medications. Am Fam Physician

81(5):617–622
Murray ML, de Vries CS, Wong IC (2004) A drug utilisation study of antidepressants in children
and adolescents using the General Practice Research Database. Arch Dis Child 89(12):
1098–1102. doi:10.1136/adc.2004.064956
Murray ML, Thompson M, Santosh PJ, Wong IC (2005) Effects of the committee on safety of
medicines advice on antidepressant prescribing to children and adolescents in the UK. Drug
Saf 28(12):1151–1157, doi:28129 [pii]
Olfson M, Blanco C, Liu L, Moreno C, Laje G (2006) National trends in the outpatient treatment
of children and adolescents with antipsychotic drugs. Arch Gen Psychiatry 63(6):679–685.
doi:10.1001/archpsyc.63.6.679
Park P, Caballero J, Omidian H (2014) Use of serotonin norepinephrine reuptake inhibitors in the
treatment of attention-deficit hyperactivity disorder in pediatrics. Ann Pharmacother 48(1):
86–92. doi:10.1177/1060028013506561
Patten SB, Waheed W, Bresee L (2012) A review of pharmacoepidemiologic studies of antipsy-
chotic use in children and adolescents. Can J Psychiatry Rev Can Psychiatr 57(12):717–721
Penfold RB, Stewart C, Hunkeler EM, Madden JM, Cummings JR, Owen-Smith AA, Rossom RC,
Lu CY, Lynch FL, Waitzfelder BE, Coleman KJ, Ahmedani BK, Beck AL, Zeber JE, Simon GE
(2013) Use of antipsychotic medications in pediatric populations: what do the data say? Curr
Psychiatry Rep 15(12):426. doi:10.1007/s11920-013-0426-8
Peterson MC (2001) Reversible galactorrhea and prolactin elevation related to fluoxetine use.
Mayo Clin Proc 76(2):215–216. doi:10.1016/S0025-6196(11)63132-7
Peyre H, Hoertel N, Cortese S, Acquaviva E, Limosin F, Delorme R (2013) Long-term effects of
ADHD medication on adult height: results from the NESARC. J Clin Psychiatry 74(11):
1123–1124. doi:10.4088/JCP.13l08580
Peyre H, Hoertel N, Hatteea H, Limosin F, Dubuc C, Delorme R (2014) Adulthood self-reported
cardiovascular risk and ADHD medications: results from the 2004–2005 National
Epidemiologic Survey on Alcohol and Related Conditions. J Clin Psychiatry 75(2):181–182.
doi:10.4088/JCP.13l08736
Piacentini J, Bennett S, Compton SN, Kendall PC, Birmaher B, Albano AM, March J, Sherrill J,
Sakolsky D, Ginsburg G, Rynn M, Bergman RL, Gosch E, Waslick B, Iyengar S, McCracken J,
Walkup J (2014) 24- and 36-week outcomes for the Child/Adolescent Anxiety Multimodal Study
(CAMS). J Am Acad Child Adolesc Psychiatry 53(3):297–310. doi:10.1016/j.jaac.2013.11.010
Poulton A (2005) Growth on stimulant medication; clarifying the confusion: a review. Arch Dis
Child 90(8):801–806. doi:10.1136/adc.2004.056952
Powers SW, Kashikar-Zuck SM, Allen JR, LeCates SL, Slater SK, Zafar M, Kabbouche MA,
O’Brien HL, Shenk CE, Rausch JR, Hershey AD (2013) Cognitive behavioral therapy plus
amitriptyline for chronic migraine in children and adolescents: a randomized clinical trial.
JAMA 310(24):2622–2630. doi:10.1001/jama.2013.282533
Pringsheim T, Steeves T (2011) Pharmacological treatment for Attention Deficit Hyperactivity
Disorder (ADHD) in children with comorbid tic disorders. Cochrane Database Syst Rev
(4):CD007990. doi:10.1002/14651858.CD007990.pub2
Pringsheim T, Lam D, Ching H, Patten S (2011a) Metabolic and neurological complications of
second-generation antipsychotic use in children: a systematic review and meta-analysis of ran-
domized controlled trials. Drug Saf 34(8):651–668. doi:10.2165/11592020-000000000-00000
Pringsheim T, Lam D, Patten SB (2011b) The pharmacoepidemiology of antipsychotic medica-
tions for Canadian children and adolescents: 2005–2009. J Child Adolesc Psychopharmacol
21(6):537–543. doi:10.1089/cap.2010.0145
Prozac Product monograph, Lilly research Laboratories, Eli Lily and Compagny; http://www.lilly.
ca/en/pdf/product-monograph/27_prozac-pm_20sep2013.pdf
Rani F, Murray ML, Byrne PJ, Wong IC (2008) Epidemiologic features of antipsychotic prescrib-
ing to children and adolescents in primary care in the United Kingdom. Pediatrics 121(5):
1002–1009. doi:10.1542/peds.2007–2008
Rani FA, Byrne PJ, Murray ML, Carter P, Wong IC (2009) Paediatric atypical antipsychotic moni-
toring safety (PAMS) study: pilot study in children and adolescents in secondary- and tertiary-
care settings. Drug Saf 32(4):325–333. doi:10.2165/00002018-200932040-00006
Rani FA, Byrne P, Cranswick N, Murray ML, Wong IC (2011) Mortality in children and adoles-
cents prescribed antipsychotic medication: a retrospective cohort study using the UK general
practice research database. Drug Saf 34(9):773–781. doi:10.2165/11591120-000000000-
00000
Roessner V, Robatzek M, Knapp G, Banaschewski T, Rothenberger A (2006) First-onset tics in
patients with attention-deficit-hyperactivity disorder: impact of stimulants. Dev Med Child
Neurol 48(7):616–621. doi:10.1017/S0012162206001290
Ronaldson KJ, Taylor AJ, Fitzgerald PB, Topliss DJ, Elsik M, McNeil JJ (2010) Diagnostic
characteristics of clozapine-induced myocarditis identified by an analysis of 38 cases and 47
controls. J Clin Psychiatry 71(8):976–981. doi:10.4088/JCP.09m05024yel
Ronsley R, Scott D, Warburton WP, Hamdi RD, Louie DC, Davidson J, Panagiotopoulos C (2013)
A population-based study of antipsychotic prescription trends in children and adolescents in
British Columbia, from 1996 to 2011. Can J Psychiatry Rev Can Psychiatr 58(6):361–369
Schmidt MJ, Fuller RW, Wong DT (1988) Fluoxetine, a highly selective serotonin reuptake
inhibitor: a review of preclinical studies. Br J Psychiatry Suppl 3:40–46
Schneeweiss S, Patrick AR, Solomon DH, Dormuth CR, Miller M, Mehta J, Lee JC, Wang PS
(2010) Comparative safety of antidepressant agents for children and adolescents regarding
suicidal acts. Pediatrics 125(5):876–888. doi:10.1542/peds.2009-2317
Seida JC, Schouten JR, Boylan K, Newton AS, Mousavi SS, Beaith A, Vandermeer B, Dryden
DM, Carrey N (2012) Antipsychotics for children and young adults: a comparative effective-
ness review. Pediatrics 129(3):e771–e784. doi:10.1542/peds.2011-2158
Shin L, Bregman H, Breeze JL, Noyes N, Frazier JA (2009) Metformin for weight control in
pediatric patients on atypical antipsychotic medication. J Child Adolesc Psychopharmacol
19(3):275–279. doi:10.1089/cap.2008.094
Sikich L (2008) Efficacy of atypical antipsychotics in early-onset schizophrenia and other
psychotic disorders. J Clin Psychiatry 69(Suppl 4):21–25
Sikich L, Frazier JA, McClellan J, Findling RL, Vitiello B, Ritz L, Ambler D, Puglia M, Maloney
AE, Michael E, De Jong S, Slifka K, Noyes N, Hlastala S, Pierson L, McNamara NK, Delporto-
Bedoya D, Anderson R, Hamer RM, Lieberman JA (2008) Double-blind comparison of first-
and second-generation antipsychotics in early-onset schizophrenia and schizoaffective
disorder: findings from the treatment of early-onset schizophrenia spectrum disorders (TEOSS)
study. Am J Psychiatry 165(11):1420–1431. doi:10.1176/appi.ajp.2008.08050756
Steinhausen HC (2014) Recent international trends in psychotropic medication prescriptions for
children and adolescents. Eur Child Adolesc Psychiatry. doi:10.1007/s00787-014-0631-y
Steinhausen HC, Bisgaard C (2014) Nationwide time trends in dispensed prescriptions of psycho-
tropic medication for children and adolescents in Denmark. Acta Psychiatr Scand 129(3):
221–231. doi:10.1111/acps.12155
Stone MB (2014) The FDA warning on antidepressants and suicidality – why the controversy? N
Engl J Med 371(18):1668–1671. doi:10.1056/NEJMp1411138
Strawn JR, Welge JA, Wehry AM, Keeshin B, Rynn MA (2014) Efficacy and tolerability of anti-
depressants in pediatric anxiety disorders: a systematic review and meta-analysis. Depress
Anxiety. doi:10.1002/da.22329
Thomas CP, Conrad P, Casler R, Goodman E (2006) Trends in the use of psychotropic medications
among adolescents, 1994 to 2001. Psychiatr Serv 57(1):63–69. doi:10.1176/appi.ps.57.1.63
Verdoux H, Tournier M, Begaud B (2010) Antipsychotic prescribing trends: a review of pharmaco-
epidemiological studies. Acta Psychiatr Scand 121(1):4–10. doi:10.1111/j.1600-0447.2009.
01425.x
Vitiello B (2008) Understanding the risk of using medications for attention deficit hyperactivity
disorder with respect to physical growth and cardiovascular function. Child Adolesc Psychiatr
Clin N Am 17(2):459–474, xi. doi:10.1016/j.chc.2007.11.010
Vitiello B, Correll C, van Zwieten-Boot B, Zuddas A, Parellada M, Arango C (2009) Antipsychotics

in children and adolescents: increasing use, evidence for efficacy and safety concerns. Eur
Neuropsychopharmacol: J Eur Coll Neuropsychopharmacol 19(9):629–635. doi:10.1016/j.
euroneuro.2009.04.008
Vitiello B, Elliott GR, Swanson JM, Arnold LE, Hechtman L, Abikoff H, Molina BS, Wells K,
Wigal T, Jensen PS, Greenhill LL, Kaltman JR, Severe JB, Odbert C, Hur K, Gibbons R (2012)
Blood pressure and heart rate over 10 years in the multimodal treatment study of children with
ADHD. Am J Psychiatry 169(2):167–177. doi:10.1176/appi.ajp.2011.10111705
Volkow ND, Wang GJ, Fowler JS, Molina PE, Logan J, Gatley SJ, Gifford A, Ding YS, Wong C,
Pappas NR, Zhu W, Swanson JM (2003) Cardiovascular effects of methylphenidate in humans
are associated with increases of dopamine in brain and of epinephrine in plasma.
Psychopharmacology (Berl) 166(3):264–270. doi:10.1007/s00213-002-1340-7
Walkup JT, Albano AM, Piacentini J, Birmaher B, Compton SN, Sherrill JT, Ginsburg GS, Rynn
MA, McCracken J, Waslick B, Iyengar S, March JS, Kendall PC (2008) Cognitive behavioral
therapy, sertraline, or a combination in childhood anxiety. N Engl J Med 359(26):2753–2766.
doi:10.1056/NEJMoa0804633
Weintrob N, Cohen D, Klipper-Aurbach Y, Zadik Z, Dickerman Z (2002) Decreased growth during
therapy with selective serotonin reuptake inhibitors. Arch Pediatr Adolesc Med 156(7):
696–701
Whittington CJ, Kendall T, Fonagy P, Cottrell D, Cotgrove A, Boddington E (2004) Selective
serotonin reuptake inhibitors in childhood depression: systematic review of published versus
unpublished data. Lancet 363(9418):1341–1345. doi:10.1016/S0140-6736(04)16043-1
Wigal T, Greenhill L, Chuang S, McGough J, Vitiello B, Skrobala A, Swanson J, Wigal S, Abikoff
H, Kollins S, McCracken J, Riddle M, Posner K, Ghuman J, Davies M, Thorp B, Stehli A
(2006) Safety and tolerability of methylphenidate in preschool children with ADHD. J Am
Acad Child Adolesc Psychiatry 45(11):1294–1303. doi:10.1097/01.chi.0000235082.63156.27
Wijlaars LP, Nazareth I, Petersen I (2012) Trends in depression and antidepressant prescribing in
children and adolescents: a cohort study in The Health Improvement Network (THIN). PLoS
One 7(3):e33181. doi:10.1371/journal.pone.0033181
Wilens TE, Morrison NR (2011) The intersection of attention-deficit/hyperactivity disorder and
substance abuse. Curr Opin Psychiatry 24(4):280–285. doi:10.1097/YCO.0b013e328345c956
Wilens TE, Faraone SV, Biederman J, Gunawardene S (2003) Does stimulant therapy of attention-
deficit/hyperactivity disorder beget later substance abuse? A meta-analytic review of the litera-
ture. Pediatrics 111(1):179–185
Wilens TE, Adler LA, Adams J, Sgambati S, Rotrosen J, Sawtelle R, Utzinger L, Fusillo S (2008)
Misuse and diversion of stimulants prescribed for ADHD: a systematic review of the literature.
J Am Acad Child Adolesc Psychiatry 47(1):21–31. doi:10.1097/chi.0b013e31815a56f1
Winterfeld U, Le Heuzey MF, Acquaviva E, Mouren MC, Brion F, Bourdon O (2009) Off-label use
of psychotropic medications in pediatric wards: a prospective study. Arch Pediatr 16(9):
1252–1260. doi:10.1016/j.arcped.2009.06.012
Winterstein AG, Gerhard T, Kubilis P, Saidi A, Linden S, Crystal S, Zito J, Shuster JJ, Olfson M
(2012) Cardiovascular safety of central nervous system stimulants in children and adolescents:
population based cohort study. BMJ 345:e4627. doi:10.1136/bmj.e4627
Zito JM, Safer DJ, de Jong-van den Berg LT, Janhsen K, Fegert JM, Gardner JF, Glaeske G, Valluri
SC (2008) A three-country comparison of psychotropic medication prevalence in youth. Child
Adolesc Psychiatry Ment Health 2(1):26. doi:10.1186/1753-2000-2-26
Chapter 14
Safety of Psychotropic Drugs in the Elderly
Kate L. Lapane, Anne Hume, Christine Ulbricht, and Giovanni Gambassi
Abstract The past decades have witnessed an unprecedented phenomenon of

global aging with a dramatic rise in the number of individuals above the age of
65 years. Drug therapy is the primary approach to managing chronic disease in
older adults. It comes as no surprise that increases in drug therapy have accompa-
nied the rise in the older adult population with chronic conditions. The safety
profile of drugs in older adults may differ substantially from that in younger indi-
viduals. Older adults, especially more frail individuals, may have significantly
altered pharmacokinetics such as decreased renal function and pharmacodynamics
which increase the likelihood of adverse drug effects. Inappropriate drug selection
can lead to complications in drug therapy. Complications in drug therapy may
appear as adverse drug events. In this chapter, we review particular safety con-
cerns with major psychiatric medications used in older adults. We specifically
focus on antipsychotics, antidepressants, benzodiazepines, and anti-dementia
agents.
Keywords Drug safety • Elderly • Antipsychotics • Antidepressants •

Benzodiazepines • Anti-dementia agents
K.L. Lapane (*) • C. Ulbricht

Department of Quantitative Health Sciences, Division of Epidemiology of Chronic Diseases
and Vulnerable Populations, University of Massachusetts Medical School, Worcester,
MA, USA
e-mail: Kate.Lapane@umassmed.edu; christine.ulbricht@umassmed.edu
A. Hume
College of Pharmacy, Rhode Island University, Kingston, RI, USA
Department of Quantitative Health Sciences, University of Massachusetts Medical School,
Worcester, MA, USA
e-mail: alhume@uri.edu
G. Gambassi
Department of Medical Sciences, Università Cattolica del Sacro Cuore, Rome, Italy
e-mail: giovanni.gambassi@rm.unicatt.it

DOI 10.1007/978-3-319-24741-0_14
286 K.L. Lapane et al.
14.1 Introduction
In many countries, the age distribution has slowly changed. In the United States,
13 % of the population is now aged ≥65 years (Census Bureau 2010). This propor-
tion will be 20 % by 2030, and 5 % will be aged 80 years of age or older (OECD
2009). Growth will occur in all racial and ethnic groups (Federal Interagency Forum
on Aging-Related Statistics 2004). This will pose challenges as the per capita
health-care expenditure is five times higher for those at least 65 years of age com-
pared those younger than 65 years (Lubitz et al. 2001). With increased longevity
comes a greater prevalence of multiple chronic diseases including psychiatric con-
ditions (Anderson 2007).
Drug therapy is the primary approach to managing chronic disease in older
adults. It comes as no surprise that increases in drug therapy have accompanied the
rise in the older adult population with chronic conditions. Over 85 % of adults at
least 65 years of age regularly use prescription drugs (Ihara et al. 2002), and 55 %
of older adults take three or more drugs on a regular basis (Kaiser Public Opinion
Update 2000). Older adults consume 31 % of all prescribed drugs (Baum et al.
1987), the average number being between 2 and 6 (Stewart and Cooper 1994).
Polypharmacy may be the new paradigm for quality drug therapy. However, the
number of drugs per se is not the primary factor for quality, as appropriateness for
the specific individual, life expectancy, burden of concomitant conditions, and per-
sonal preferences is most important.
The safety profile of drugs in older adults may differ substantially from that in
younger individuals. Older adults, especially more frail individuals, may have sig-
nificantly altered pharmacokinetics such as decreased renal function and pharmaco-
dynamics which increase the likelihood of adverse drug effects (Sera and McPherson
2012). Inappropriate drug selection (Stuck et al. 1994) can lead to complications in
drug therapy. Complications in drug therapy may appear as an adverse drug event.
The incidence of adverse drug events has been estimated to be 27.4 % in community-
dwelling older adults (Gandhi et al. 2003). The economic costs of preventable
adverse drug events are staggering (Field et al. 2005). Gurwitz et al. demonstrated
an incidence of adverse drug events of 50.1 per 1,000 person years among a popula-
tion of elderly Medicare beneficiaries visiting an outpatient physician practice
(Gurwitz et al. 2003). Older adults are two to three times more likely to experience
an adverse drug event than patients 20–30 years old. Medication-related problems
occur in 1.5–44 % of inpatients and over 30 % of outpatients (Monette et al. 1995).
Pharmacologic management is the top condition in need of quality of care improve-
ment initiatives in vulnerable older adults (Sloss et al. 2000). The Institute of
Medicine (IOM) “To Err is Human” report reinforces the need for improving drug
therapy in a population taking multiple medications and having comorbidities
(Institute of Medicine 2000).
In this chapter, we review particular safety concerns of major psychiatric medi-
cations used in older adults. We specifically focus on antipsychotics, antidepres-
sants, benzodiazepines, and anti-dementia agents.
14 Safety of Psychotropic Drugs in the Elderly 287
In older adults, increases in the use of antidepressants have been observed

(Newman and Schopflocher 2008; Crystal et al. 2003). Benzodiazepines have either
remained stable or declined but are still commonly prescribed to older adults.
Although decreases in antipsychotic use in older adults have been noted (Gallini
et al. 2014), antipsychotic use among older adults is common.
14.2 Antipsychotics
The off-label use of conventional and atypical antipsychotics to treat the behavioral
and psychological symptoms in dementia remains common despite their limited
efficacy. Despite their perceived better safety, some atypical antipsychotics may still
possess anticholinergic and hypotensive effects, as well as extrapyramidal effects
including tardive dyskinesias, although at a lower rate than conventional
antipsychotics.
In April 2005, the US Food and Drug Administration (FDA) issued a public
health advisory that the use of atypical antipsychotics such as risperidone, olanzap-
ine, quetiapine, and aripiprazole to treat older adults with dementia was associated
with an increased risk for death compared with placebo. In a pooled analysis of
trials involving atypical antipsychotics, the mortality rate was 60–70 % higher than
with placebo in 15 of the 17 trials. Evidence providing support for these warnings
has also raised further safety concerns about conventional antipsychotics (Scneider
et al. 2005; Wang et al. 2005; Kryzhanovskaya et al. 2006). In June 2008, the FDA
stated that the conventional antipsychotics share a similar or even higher risk of
increased mortality with the atypical antipsychotics. The FDA concluded that anti-
psychotics should not be used for the treatment of dementia-related psychosis.
Despite these warnings, a recent review suggested that antipsychotics should be
“used with caution only when non-pharmacologic approaches have failed to ade-
quately control behavioral and psychological symptoms in dementia” (Trifirò et al.
2009).
Antipsychotics appear to confer a greater mortality risk in older adults than any
other psychotropic medication. A retrospective case-control study of 90,786 demen-
tia patients within the Veterans Health Administration from 1998 to 2009 found that
newly prescribed antipsychotic users had an increased mortality risk when com-
pared to both matched nonusers and antidepressant users (Maust et al. 2015). This
risk ranged from 2.0 % for those receiving quetiapine to 3.8 % for those receiving
haloperidol when comparing antipsychotic users to nonusers. When comparing
antipsychotic users to antidepressant users, the mortality risk ranged from 3.2 % for
those receiving quetiapine to 12.3 % for those receiving haloperidol. Additionally,
a dose-response increase in mortality risk was observed for atypical antipsychotics
(olanzapine, quetiapine, and risperidone).
The potential causes of death associated with antipsychotic use merit consider-
ation of the potential for drug-drug and drug-disease interactions. Several plausible
mechanisms can be proposed, including cerebrovascular events, arrhythmias and

sudden cardiac death, venous thromboembolism and pulmonary embolism, and
aspiration pneumonia. Metabolic effects of antipsychotics may be a longer-term
safety concern. Antipsychotics may contribute to events that are not initially recog-
nized as the first step in a sequence that promotes premature death, such as falls
leading to hip fractures. Apart from mortality, other serious safety issues have been
raised regarding the use of antipsychotics including cerebrovascular events, cardiac
events, peripheral vascular disease, metabolic disease, infections, and falls/frac-
tures. Hip fracture, stroke, myocardial infarction, and ventricular arrhythmias par-
tially explain the mortality difference between first- and second-generation
antipsychotics (Jackson et al. 2014).
A pooled analysis documented a threefold increased risk of stroke and tran-
sient ischemic attacks for risperidone and olanzapine compared with placebo
(De Deyn et al. 2005). Following the warning on stroke and antipsychotic use,
observational studies compared the risk of stroke between atypical and conven-
tional antipsychotics, most in the general elderly population and in patients with
dementia. The studies suggest no increased risk of stroke with atypical com-
pared to conventional antipsychotics but could not explore the dose and duration
effect and risks of individual compounds or rule out confounding by indication
because of the strong interrelationship between ischemia, dementia, and subse-
quent strokes. Stroke may be related to the first weeks of treatment (Sacchetti
et al. 2010).
With respect to cardiac events, since the 1960s, sudden cardiac death has been
reported with conventional antipsychotic use (Straus et al. 2004) because of their
ability to prolong the QTc interval which may result in torsade de pointes and other
ventricular arrhythmias (Reilly et al. 2000). Observational studies have confirmed
the signals from spontaneous reports, and suggested conventional antipsychotics are
associated with an increased risk of sudden cardiac death. A recent systematic
review of the association between antipsychotics and myocardial infarction was
inconclusive owing to the heterogeneity of the studies, but the largest study revealed
no association (Brauer et al. 2011).
A relationship between antipsychotics and venous thromboembolism (VTE)
was first suggested five decades ago. Reviews of the available data for aripiprazole,
clozapine, and olanzapine have led to warnings about VTE being added to their
Summaries of Product Characteristics (MPA Report 2009). The studies on VTE
and antipsychotics mostly focus on schizophrenia or young patients and have
methodological issues such as small sample sizes and inadequate control of con-
founding. Inconsistencies in findings have been noted, but an increased risk of
VTE is likely (Zhang et al. 2011). Little data are available on the peripheral vascu-
lar effects of antipsychotics in dementia, which is highly relevant given the large
potential for interacting comedication on serotonin receptors and platelet
function.
In patients with either schizophrenia or bipolar disorder, atypical antipsychotics
such as olanzapine, quetiapine, and clozapine are associated with metabolic abnor-
malities including weight gain, lipid disturbances, and altered glucose homeostasis,
known risk factors for cardiovascular events (Newcomer 2005). Whether older
adults with behavioral and psychological symptoms in dementia receiving antipsy-
chotics develop similar disturbances is controversial as food intake is reduced in
these patients. Only a few small studies have been published so far on this associa-
tion. A study of 36 residents showed treatment with low-dose atypical antipsychot-
ics did not lead to weight gain or increased risk of developing type 2 diabetes or
lipid metabolism abnormalities (Rondanelli et al. 2006). In contrast, CATIE-AD
reported weight gain but no effect on glucose, total cholesterol, or triglyceride levels
during the use of olanzapine, quetiapine, and risperidone, and the risk increased
over time. Post-hoc analyses of other studies with olanzapine and risperidone were
consistent with the CATIE-AD results. A recently published Canadian study found
that among older patients with diabetes, the initiation of treatment with antipsy-
chotic drugs was associated with an increased risk of hospitalization for hypergly-
cemia (Lipscombe et al. 2009). The risk was high during the initial course of
treatment and was increased with the use of all antipsychotic agents. Among nurs-
ing home residents with dementia, conventional antipsychotics particularly short-
term therapy, but not atypical antipsychotic use, increased risk of diabetes onset
(Jalbert et al. 2011).
Infections, primarily pneumonia, have been listed as one of the most prevalent
causes of death among demented older adults using antipsychotics both in clinical
trials and observational studies. Although one study reported a threefold increased
risk with atypical antipsychotics and a 1.6-fold increase with conventional antipsy-
chotics compared with nonuse (Knol et al. 2008), others found a slightly higher
rate of fatal pneumonia during conventional antipsychotic use relative to atypical
antipsychotic use. The overall risk of antipsychotic use was not increased com-
pared to nonuse in a cohort of elderly persons (Setoguchi et al. 2008). Trifirò
showed that the use of either atypical or typical antipsychotics in older patients is
associated in a dose-dependent fashion with the development of community-
acquired pneumonia (Trifirò et al. 2010). More work is needed to understand this
effect (Trifirò 2011).
Information regarding adverse effects of antipsychotic treatment in older adults
with schizophrenia is scant. While older adults with schizophrenia are thought to
have a greater sensitivity to treatment-related adverse effects (Masand 2000), the
overall incidence of adverse events was low (Lasser et al. 2004).
14.3 Antidepressants
In general, antidepressants are equally effective in the treatment of depression

(Mottram et al. 2006). The major categories of antidepressants and individual drugs
differ in their side effect and drug interaction profiles. Although for younger patients
evidence exists that suggests the use of antidepressants may increase the risk of
suicidality and suicidal ideation, among older adults with depression, the risk is
reduced. Selective serotonin reuptake inhibitors (SSRIs) are the most commonly
prescribed antidepressant class including in frail older adults. Their widespread use
has been based primarily on their lack of traditional tricyclic antidepressant (TCA)
side effects, although SSRIs commonly produce adverse gastrointestinal and sexual
symptoms and less frequently central nervous system effects including insomnia,
anxiety, and tremors (Grimsley and Jann 1992).
Both tertiary and secondary amine TCAs have been used in older adults for
many years. TCA side effects commonly include dry mouth, blurred vision, urinary
retention, and constipation due to their anticholinergic and antihistaminic activity.
Amoxapine has been associated with a higher risk of extrapyramidal side effects
including akathisia and tardive dyskinesia due to its metabolism to loxapine, a
neuroleptic, and maprotiline with seizures (Rosenstein 1993). Frail older adults
remain at the greatest risk of anticholinergic effects, cardiovascular effects, and
effects on appetite. The anticholinergic effects of older TCAs have included the
loss of accommodation with blurring of vision, dry mouth, constipation, urinary
retention, tachycardia, confusion, and delirium (Cole and Bodkin 1990). These
effects range from what might be considered minor in nature (i.e., blurring of
vision and dry mouth), to moderate (i.e., urinary retention), to potentially very seri-
ous reactions (i.e., delirium). Anticholinergic side effects minor in nature might
potentially have important effects on the older adult’s quality of life. As an exam-
ple, while tolerance to some anticholinergic effects is known to occur, the loss of
accommodation generally does not improve over time and would limit the indi-
vidual’s ability to perform the simple pleasurable act of reading a book. The pres-
ence of concomitant diseases such as prostatic hypertrophy or diabetes mellitus, for
example, in older adults may increase the risk that clinically important urinary
retention will occur resulting in detrimental effects on the individual’s continence
and overall functional status. Most importantly, confusion and delirium may
develop from the use of strongly anticholinergic drugs such as amitriptyline in
individuals with preexisting cognitive impairment. Potentially, the resulting prob-
lems with memory, concentration, and behavioral disturbances may be overlooked
as part of the patient’s primary illness or inappropriately treated with a neuroleptic
such as haloperidol or with physical restraints. Trazodone has sedating properties
(Nierenberg 1994).
Cardiovascular side effects of TCAs must also be considered. Orthostatic
hypotension and cardiac conduction defects are the most common TCA cardio-
vascular side effects. Orthostatic hypotension is due to the blockade of alpha-1
adrenergic receptors. When orthostatic hypotension is symptomatic and results in
syncope, older adults are at increased risk of falls and fractures. Although symp-
tomatic orthostatic hypotension can be minimized by the use of small doses of
secondary amines and adequate ambulation and hydration, the risk of this side
effect remains. TCAs may induce heart block in individuals with preexisting con-
duction delays.
TCAs may increase appetite and weight, potentially due to their effects on hista-
minergic systems. Although viewed as a negative outcome in younger populations,
weight gain may have beneficial effects in older adults if their nutritional status is
poor or marginal due to decreased appetite.
SSRIs other than paroxetine generally lack the anticholinergic properties associ-
ated with TCAs. With respect to effects on appetite and weight, SSRIs have been
associated with either no change or a decrease in weight at least in the short term
(Kinney-Parker 1988). The effects of SSRIs on the cardiovascular system are con-
troversial with citalopram associated with QT prolongation. In one clinical trial for
treating agitation in patients with probable Alzheimer’s disease, citalopram titrated
from 10 to 30 mg daily was associated with significant improvements in agitation
but also with an increase in QTc interval when compared to placebo (Porsteinsson
et al. 2014). Worsened cognition was also seen in this trial. Current recommenda-
tions are to limit the dosage of citalopram to no more than 20 mg daily in people
over 60 years of age. In older adults, multiple risk factors for torsade de pointes are
frequently present including hypokalemia, hypomagnesemia, and bradyarrhyth-
mias. SSRIs have been associated with an increased risk of bleeding episodes espe-
cially among older adults. Concomitant use of antiplatelet or anticoagulant drugs
for atrial fibrillation or myocardial infarction in older adults likely increases the
underlying risk (Jiang et al. 2015). Although hyponatremia has been reported to
occur in an estimated 10 % of older adults treated with antidepressants (Mannesse
2013), the risk may be greater than SSRIs even considering their widespread usage.
In addition, serotonin syndrome presenting as restlessness, anxiety, agitation, and
confusion in older adults has been associated with the use of SSRIs and SNRIs. This
side effect has been reported to occur primarily when the antidepressant was used in
combination with other drugs which have an effect on serotonin such as buspirone,
tramadol, and dextromethorphan.
Venlafaxine and bupropion generally have fewer anticholinergic effects, while
duloxetine possesses this property and may result in urinary retention. Increases in
blood pressure have been associated with these antidepressants (Augustin et al. 1997).
In addition to differences in side effects among major antidepressant classes, the
presence or absence of drug-drug interactions has become an increasingly important
issue in selecting among the available drugs. Traditionally TCAs have had relatively
few drug-drug interactions of clinical significance. Newer drugs such as the SSRI,
however, have been implicated in many potentially serious interactions which may
be important in older populations due to the number of prescribed medications and
the variety of interactions which have been reported. Monoamine oxidase (MAO)
inhibitor use has been limited for several reasons, most important of which is the
presence of many potentially serious drug-drug and drug-food interactions which
potentially can result in palpitations, severe headache, and hypertensive crisis.
Depression increases the risk of falls (Quach et al. 2013). Antidepressants have
the potential to impair gait, balance, and blood pressure regulation, although an
inconsistent association between antidepressants and falls has been noted
(Hartikainen et al. 2007). Serotonin norepinephrine reuptake inhibitors (SNRIs)
may also increase the risk of falls on the same order that SSRIs and TCAs do
(Gribbini et al. 2011). A recent finding that antidepressants increase the risk of out-
door falls, but not indoor falls (Quach et al. 2013), seems counter to studies docu-
menting the link between antidepressants and falls in nursing homes (Thapa et al.
1998). Excess fractures associated with both SSRIs and TCAs have been confirmed
(Rosenstein 1993; Rabenda et al. 2013). The estimated contribution of antidepres-

sant use to the population rate of hip fractures varies between 3 and 7 % (Prieto-
Alhambra et al. 2014).
Falls and fractures have been associated with the use of many antidepressants.
People who have a history of falls or fractures should avoid SSRIs and SNRIs unless
safer alternatives are not available due to SSRIs’ ability to produce ataxia and syn-
cope and impair psychomotor functioning (American Geriatrics Society 2012).
TCAs should also be avoided for people with a history of falls or fractures because
of the same reasons as SSRIs.
14.4 Benzodiazepines and BZDP Receptor Agonists
Benzodiazepines continue to be overprescribed in older adults, with recent data

showing that 8.7 % of community-dwelling Americans between 65 and 80 years of
age have been prescribed the drugs despite their known risks (Olfson et al. 2015). A
quarter of the drugs were long acting and almost one-third of benzodiazepine use in
this age group was for more than 4 months (Olfson et al. 2015). Sedation and
impaired driving from the use of benzodiazepines are familiar to clinicians as well
as withdrawal seizures from acutely stopping short-acting benzodiazepines. Hip
fractures continue to be associated with the use of benzodiazepines including the
short-acting drugs when used in higher dosages, concomitantly with other drugs,
and shortly after initiating therapy (Zint et al. 2010). Although cognitive impairment
has also been associated with these drugs, a recent case-control study from the
Quebec Health Insurance Program database evaluated the association between past
benzodiazepine use in independently living older adults and the subsequent risk of
developing Alzheimer’s disease (Billioti de Gage et al. 2014). A dose-effect rela-
tionship was demonstrated with higher dosages associated with a greater risk of
Alzheimer’s disease, as well as with longer-acting benzodiazepines relative to
shorter-acting agents (Billioti deGage et al. 2014).
In addition, benzodiazepine receptor agonists such as zolpidem, zopiclone, and
zaleplon are commonly prescribed for acute and chronic insomnia in older adults.
Initially these drugs were perceived to lack many of the adverse effects attributed to
benzodiazepines. Since their initial introduction, however, multiple observational
studies have reported adverse effects such as falls and hip fractures in older adults
similar to those as associated with benzodiazepines (Diem et al. 2014; Bakken et al.
2014). More recently, these drugs particularly zolpidem have been associated with
anterograde amnesia the day after administration, although this is likely less common
than with the short-acting benzodiazepine triazolam. Similar to benzodiazepines,
reports of impaired driving have been reported especially with the use of zolpidem by
women (Farkas et al. 2013). In a small crossover study of 16 healthy adults with a
mean age of 59.4 years, the use of zolpidem resulted in poorer psychomotor and driv-
ing performance as well as memory recall compared with sustained-release melato-
nin (Otmani et al. 2008).
14.5 Anti-dementia Medications
Two broad categories of drugs, the cholinesterase inhibitors and memantine, are
commonly used to treat cognitive impairment, Alzheimer’s disease, and other
dementias. For the cholinesterase inhibitors including donepezil, rivastigmine, and
galantamine, the primary safety concern has been on their potential gastrointestinal
effects, especially anorexia, nausea, vomiting, and diarrhea. In a meta-analysis of
clinical trials of cholinesterase inhibitors, the excess rates of dropouts due to
adverse events were low (donepezil 2 %, rivastigmine 9 %, galantamine 14 %) and
gastrointestinal in nature (Lanctôt et al. 2003). The recent introduction of donepezil
23 mg tablet has an increased risk of gastrointestinal symptoms including weight
loss and vomiting over the 10 mg tablet yet only modest additional benefits on rat-
ing scales (Farlow et al. 2010). For the frail patient with Alzheimer’s disease, this
may result in additional challenges to maintain an adequate weight. In addition,
cholinesterase inhibitors may make worsen urinary incontinence in older adults,
resulting in the subsequent prescribing of urinary antimuscarinic drugs such as tolt-
erodine and oxybutynin which in turn may further increase confusion. More
recently, in a Canadian population-based cohort, the use of cholinesterase inhibitors
has been linked to an increased risk of bradycardia and syncope with subsequent
falling and fractures in community-dwelling older adults (Gill et al. 2009). Although
the precise mechanism for developing bradycardia is unclear, the drugs may
enhance vagal tone (Gill et al. 2009).
Memantine is well tolerated with few adverse events (Lanctôt et al. 2003; Farlow
et al. 2010). In a recent multicountry study including 46,737 Medicare beneficiaries
and 29,496 Danish participants, cholinesterase inhibitor users had similar risks of
heart failure and myocardial infarction, but memantine was associated with
increased risks of cardiac events in the Danish sample. In both countries, memantine
users had greater mortality rates most likely due to channeling of the sickest patients
to memantine therapy (Gill et al. 2009).
14.6 Conclusions
The safe use of psychiatric drugs presents special concerns in older adults.
Interactions due to comorbid conditions and concomitant drugs, as well as pharma-
cokinetic and pharmacodynamics changes, increase the risk of adverse effects with
psychiatric drugs. Seemingly simple side effects such as blurred vision may have
significant negative effects on quality of life in the older adult. More importantly,
despite limited efficacy, the use of antipsychotics in older adults is associated with
increased risks of death, adverse cardiac outcomes, and pneumonia. Falls and frac-
tures are commonly linked to many psychiatric medications. The risks of treatment
must be considered carefully in older adults.
References
American Geriatrics Society (2012) Beers Criteria Update Expert Panel. American Geriatrics
Society updated Beers Criteria for potentially inappropriate medication use in older adults. J
Am Geriatr Soc 60(4):616–631
Anderson G (2007) Chronic conditions: making the case for ongoing care. Johns
Hopkins University, Baltimore. http://www.fightchronicdisease.com/news/pfcd/documents/
ChronicCareChartbook_FINAL.pdf
Augustin BG, Cold JA, Jann MW (1997) Venlafaxine and nefazodone, two pharmacologically
distinct antidepressants. Pharmacotherapy 17:511–530
Bakken MS, Engeland A, Engesaeter LB, Ranhoff AH, Hunskaar S, Ruths S (2014) Risk of hip
fracture among older people using anxiolytic and hypnotic drugs: a nationwide prospective
cohort study. Eur J Clin Pharmacol 70:873–880
Baum C, Kennedy DL, Knapp DE et al (1987) Drug utilization in the US – 1986. Department of
Health and Human Services, Food and Drug Administration, Rockville
Billioti deGage S, Moride Y, Ducruet T, Kurth T, Verdoux H, Tournier M et al (2014) Benzodiazepine
use and risk of Alzheimer’s disease: case-control study. BMJ 349:g5205
Brauer R, Douglas I, Smeeth (2011) The association between antipsychotic agents and the risk of
myocardial infarction: a systematic review. Br J Clin Pharmacol 72:871–878
Census Bureau (2010) Census. http://www.census.gov/prod/cen2010/briefs/c2010br-03.pdf
Cole JO, Bodkin JA (1990) Antidepressant side effects. J Clin Psychiatry 51:21–26
Crystal S, Sambamoorthi U, Walkup JT, Akincigil A (2003) Diagnosis and treatment of depression
in the elderly medicare population: predictors, disparities, and trends. J Am Geriatr Soc
51:1718–1728
De Deyn P, Katz I, Brodaty H, Lyons B, Greenspan A, Burns A (2005) Management of agitation,
aggression, and psychosis associated with dementia: a pooled analysis including three random-
ized, placebo-controlled double-blind trials in nursing home residents treated with risperidone.
Clin Neurol Neurosurg 107:497–508
Diem SJ, Ewing SK, Stone KL, Ancoli-Israel S, Redline S, Ensrud KE (2014) The Osteoporotic
Fractures in Men (MrOS) Study Group. Use of non-benzodiazepine sedative hypnotics and risk
of falls in older men. J Gerontol Geriatr Res 3:158
Farkas RH, Unger EF, Temple R (2013) Zolpidem and driving impairment – identifying persons at
risk. N Engl J Med 369:689–691
Farlow MR, Salloway S, Tariot PN, Yardley J, Moline ML, Wang Q et al (2010) Effectiveness and
tolerability of high dose (23 mg/d) versus standard-dose (10 mg/d) donepezil in moderate to
severe Alzheimer’s disease: a 24-week, randomized double-blind study. Clin Ther
32:1234–1251
Federal Interagency Forum on Aging-Related Statistics (2004) Older Americans 2004: key indica-
tors of well-being. Federal interagency forum on aging-related statistics. US Government
Printing Office, Washington, DC. Available at: http://www.agingstats.gov/chartbook2004/
OA_2004.pdf
Field TS, Gilman BH, Subramanian S, Fuller JC, Bates DW, Gurwitz JH (2005) The costs associ-
ated with adverse drug events among older adults in the ambulatory setting. Med Care
43:1171–1176
Gallini A, Andrieu S, Donohue JM, Oumouhou N, Lapeyre-Mestre M, Gardette V (2014) Trends
in use of antipsychotics in elderly patients with dementia: impact of national safety warnings.
Eur Neuropsychopharmacol 24:95–104
Gandhi TK, Weingart SN, Borus J, Seger AC, Peterson J, Burdick E et al (2003) Adverse drug
events in ambulatory care. N Eng J Med 348:1556–1564
Gill SS, Anderson GM, Fischer HD, Bell CM, Li P, Normand SL, Rochon PA (2009) Syncope and
its consequences in patients with dementia receiving cholinesterase inhibitors: a population-
based cohort study. Arch Intern Med 169:867–873
Gribbin J, Hubbard R, Gladman J, Smith C, Lewis S (2011) Serotonin-norepinephrine reuptake

inhibitor antidepressants and the risk of falls in older people: case-control and case-series
analysis of a large UK primary care database. Drugs Aging 28:895–902
Grimsley SR, Jann MW (1992) Paroxetine, sertraline and fluvoxamine: new selective serotonin
reuptake inhibitors. Clin Pharmacol 11:930–957
Gurwitz JH, Field TS, Harrold LR, Rothschild J, Debellis K, Seger AC et al (2003) Incidence and
preventability of adverse drug events among older persons in the ambulatory setting. JAMA
289:1107–1116
Hartikainen S, Lönnroos E, Louhivuori K (2007) Medication as a risk factor for falls: critical
systematic review. J Gerontol A Biol Sci Med Sci 62:1172–1181
Ihara E, Summer L, Shirey L (2002) Prescription drugs: a vital component of health care. In:
Challenges for the 21st century: chronic and disabling conditions. Center on an Aging Society,
Georgetown University, Washington, DC, pp 1–5. Available at: http://ihcrp.georgetown.edu/
agingsociety/pdfs/rxdrugs.pdf
Institute of Medicine (2000) In: Kohn L, Corrigan J, Donaldson M (eds) To err is human: building
a safer health care system. National Academy Press, Washington, DC
Jackson JW, Schneeweiss S, VanderWeele TJ, Blacker D (2014) Quantifying the role of adverse
events in the mortality difference between first and second-generation antipsychotics in older
adults: systematic review and meta-synthesis. PLoS ONE 9:e105376
Jalbert JJ, Daiello LA, Eaton CB, Miller SC, Lapane KL (2011) Antipsychotic use and the risk of
diabetes in nursing home residents with dementia. Am J Geriatr Pharmacother 9:153–163
Jiang HY, Chen HZ, Hu XJ, Yu ZH, Yang W, Deng M, Zhang YH, Ruan B (2015) Use of selective
serotonin reuptake inhibitors and risk of upper gastrointestinal bleeding: a systematic review
and meta-analysis. Clin Gastroenterol Hepatol 13:42–50.e3
Kaiser Public Opinion Update (2000) The public and prescription drugs. The Kaiser Family
Foundation, Washington, DC. Available at: http://www.kff.org/rxdrugs/loader.cfm?url/
commonspot/security/getfile.cfm&PageID=13541
Kinney-Parker JL (1988) Fluoxetine and weight: something lost and something gained? Clin
Pharmacol 8:727–733
Knol W, van Marum R, Jansen P, Souverein P, Schobben A, Egberts A (2008) Antipsychotic drug
use and risk of pneumonia in elderly people. J Am Geriatr Soc Sci Med 56:661–666
Kryzhanovskaya LA, Jeste DV, Young CA, Polzer JP, Roddy TE, Jansen JF et al (2006) A review
of treatment emergent adverse events during olanzapine clinical trials in elderly patients with
dementia. J Clin Psychiatry 67:933–945
Lanctôt KL, Herrmann N, Yau KK et al (2003) Efficacy and safety of cholinesterase inhibitors in
Alzheimer’s disease: a meta-analysis. CMAJ: Can Med Assoc J 169:557–564
Lasser R, Bossie CA, Zhu Y et al (2004) Efficacy and safety of long-acting risperidone in elderly
patients with schizophrenia and schizoaffective disorder. Int J Geriatr Psychiatry 19:
898–905
Lipscombe L, Lévesque L, Gruneir A, Fischer H, Juurlink D, Gill S et al (2009) Antipsychotic
drugs and hyperglycemia in older patients with diabetes. Arch Intern Med 169:1282–1289
Lubitz J, Greenberg LG, Gorina Y, Wartzman L, Gibosn D (2001) Three decades of health care use
by the elderly 1965–1998. Health Aff 20:19–32
Mannesse CK (2013) Characteristics, prevalence, risk factors, and underlying mechanism of hypo-
natremia in elderly patients treated with antidepressants: a cross-sectional study. Maturitas
76:357–363
Masand PS (2000) Side effects of antipsychotics in the elderly. J Clin Psychiatry 61(S 8):43–49;
discussion 50–51
Maust DT, Kim HM, Seyfried LS, Chiang C, Kavanagh J, Schneider LS, Kales HC (2015)
Antipsychotics, other psychotropics, and the risk of death in patients with dementia: number
needed to harm. JAMA Psychiatry 2015;72(5):438–445
Monette J, Gurwitz JH, Avorn J (1995) Epidemiology of adverse drug events in the nursing home
setting. Drugs Aging 7:203–211
Mottram PG, Wilson K, Strobl JJ (2006) Antidepressants for depressed elderly. Cochrane Database
Syst Rev (1):CD003491
Newcomer J (2005) Second-generation (atypical) antipsychotics and metabolic effects: a compre-
hensive literature review. CNS Drugs 19(S 1):1–93
Newman SC, Schopflocher D (2008) Trends in antidepressant prescriptions among the elderly in
Alberta during 1997 to 2004. Can J Psychiatry 53:704–707
Nierenberg AA (1994) Trazodone for antidepressant-associated insomnia. Am J Psychiatry
151:1069–1072
OECD Fact Book (2009) OECD Paris. http://oberon.sourceoecd.org/vl=865128/cl=23/nw=1/rpsv/
factbook2009/index.htm
Olfson M, King M, Schoenbaum M (2015) Benzodiazepine use in the United States. JAMA
Otmani S, Demazieres A, Staner C, Jacob N, Nir T, Zisapel N, Staner L (2008) Effects of prolonged-
release melatonin, zolpidem, and their combination on psychomotor functions, memory recall,
and driving skills in healthy meddle aged and elderly volunteers. Hum Psychopharmacol
23:693–705
Porsteinsson AP, Drye LT, Pollock BG, Devanand DP, Frangakis C, Ismail Z, Marano C, Meinert
CL, Mintzer JE, Munro CA, Pelton G, Rabins PV, Rosenberg PB, Schneider LS, Shade DM,
Weintraub D, Yesavage J, Lyketsos CG, CitAD Research Group (2014) Effect of citalopram on
agitation in Alzheimer disease: the CitAD randomized clinical trial. JAMA 311(7):682–691
Prieto-Alhambra D, Petri H, Goldenberg JS, Khong TP, Klungel OH, Robinson NJ, de Vries F
(2014) Excess risk of hip fractures attributable to the use of antidepressants in five European
countries and the USA. Osteoporos Int 25:847–855
Quach L, Yang FM, Berry SD, Newton E, Jones RN, Burr JA, Lipsitz LA (2013) Depression, anti-
depressants, and falls among community-dwelling elderly people: the MOBILIZE Boston
study. J Gerontol A Biol Sci Med Sci 68:1575–1581
Rabenda V, Nicolet D, Beaudart C, Bruyère O, Reginster JY (2013) Relationship between use of
antidepressants and risk of fractures: a meta-analysis. Osteoporos Int 24:121–137
Reilly J, Ayis S, Ferrier I, Jones S, Thomas S (2000) QTc interval abnormalities and psychotropic
drug therapy in psychiatric patients. Lancet 355:1048–1052
REPORT MPA (2009) The risk of venous thromboembolism associated with antipsychotics. http://
w w w. m h r a . g o v. u k / S a f e t y i n f o r m a t i o n / G e n e r a l s a f e t y i n f o r m a t i o n a n d a d v i c e /
Productspecificinformationandadvice/Antipsychoticdrugs/index.htm
Rondanelli M, Sarra S, Antoniello N, Mansi S, Govoni S, Falfo F et al (2006) No effect of atypical
antipsychotic drugs on weight gain and risk of developing type II diabetes or lipid abnormali-
ties among nursing home elderly patients with Alzheimer’s disease. Minerva Med
97:147–151
Rosenstein DL (1993) Seizures associated with antidepressants: a review. J Clin Psychiatry
54:289–299
Sacchetti E, Turrina C, Valsecchi P (2010) Cerebrovascular accidents in elderly people treated with
antipsychotic drugs: a systematic review. Drug Saf 33:273–288
Schneider LS, Dagerman KS, Insel P (2005) Risk of death with atypical antipsychotic drug treat-
ment for dementia: meta-analysis of randomized placebo-controlled trials. JAMA
294:1934–1943
Sera LC, McPherson ML (2012) Pharmacokinetics and pharmacodynamics changes associated
with aging and implications for drug therapy. Clin Geriatr Med 28:273–286
Setoguchi S, Wang PS, Alan Brookhart M, Canning CF, Kaci L, Schneeweiss S (2008) Potential
causes of higher mortality in elderly users of conventional and atypical antipsychotic medica-
tions. J Am Geriatr Soc 56:1644–1650
Sloss EM, Solomon DH, Shekelle PG et al (2000) Selecting target conditions for quality of care
improvement in vulnerable older adults. J Am Geriatr Soc 48:363–369
Stewart RB, Cooper JW (1994) Polypharmacy in the aged. Practical solutions. Drugs Aging
4:449–461
Straus S, Bleumink G, Dieleman J, van der Lei J, ‘t Jong G, Kingma J et al (2004) Antipsychotics
and the risk of sudden cardiac death. Arch Int Med 164:1293–1297
Stuck A, Beers MH, Steiner A, Aronow HU, Rubenstein LZ, Beck JC (1994) Inappropriate
medication use in community-residing older persons. Arch Intern Med 154:2195–2200
Thapa PB, Gideon P, Cost TW, Milam AB, Ray WA (1998) Antidepressants and the risk of falls
among nursing home residents. N Engl J Med 339:875–882
Trifirò G (2011) Antipsychotic drug use and community-acquired pneumonia. Curr Infect Dis Rep
13:262–268
Trifirò G, Spina E, Gambassi G (2009) Use of antipsychotics in elderly patients with dementia: do
atypical and conventional agents have a similar safety profile? Pharmacol Res 59:1–12
Trifirò G, Gambassi G, Sen EF, Caputi AP, Bagnardi V, Brea J, Sturkenboom MC (2010)
Association of community-acquired pneumonia with antipsychotic drug use in elderly patients:
a nested case-control study. Ann Intern Med 152:418–425
Wang PS, Schneeweiss S, Avorn J, Fischer MA, Mogun H, Solomon DH et al (2005) Risk of death
in elderly users of conventional vs. atypical antipsychotic medications. N Engl J Med
353:2335–2341
Zhang R, Dong L, Shao F, Tan X, Ying K (2011) Antipsychotics and venous thromboembolism
risk: a meta-analysis. Pharmacopsychiatry 44:183–188
Zint K, Haefeli WE, Glynn RJ, Mogun H, Avorn J, Sturmer T (2010) Impact of drug interactions,
dosage, and duration of therapy on the risk of hip fracture associated with benzodiazepine use
in older adults. Pharmacoepidemiol Drug Saf 19:1248–1255
Chapter 15
Safety of Psychotropic Drugs in Pregnancy
and Breastfeeding
Olav Spigset and Hedvig Nordeng
Abstract Pharmacovigilance studies are vital to our understanding of the safety of

medications in pregnancy, but great care must be taken in the analysis and interpre-
tation of observational data to avoid problems with confounding and bias. Data on
drug excretion in breast milk and possible effects of the breastfed infant often stem
from case reports or small case series, making the generalization of the results a
challenge. This chapter reviews the safety of psychotropic drugs in pregnant and
breastfeeding women and discusses methodological issues that have to be dealt with
in the interpretation of published data.
Keywords Pregnancy • Teratology • Fetal safety • Postpartum depression •

Breastfeeding • Breast milk
15.1 Introduction
The most common mental disorders during pregnancy and in the postpartum period
are depression, bipolar disorder, and anxiety disorders. For women with mild symp-
toms, non-pharmacological treatment is often sufficient, but in patients with moder-
ate to severe symptoms, drug treatment is frequently used.
O. Spigset (*)
Department of Clinical Pharmacology, St. Olav University Hospital, Trondheim, Norway
Department of Laboratory Medicine, Children’s and Women’s Health,
Norwegian University of Science and Technology, Trondheim, Norway
e-mail: olav.spigset@legemidler.no
H. Nordeng
Pharmacoepidemiology and Drug Safety Research Group, School of Pharmacy,
PharmaTox Strategic Research Initiative, Faculty of Mathematics and Natural Sciences,
University of Oslo, Oslo, Norway
Division of Mental Health, Norwegian Institute of Public Health, Oslo, Norway
e-mail: h.m.e.nordeng@farmasi.uio.no

DOI 10.1007/978-3-319-24741-0_15
300 O. Spigset and H. Nordeng
15.2 Drug Safety in Pregnancy
In the general population, knowledge of medication efficacy and safety is primarily

based on prospective and randomized clinical trials. Pregnant women, however, are
routinely excluded from these studies due to uncertainties about the effects on fetal
development. Available information on medication safety in human pregnancies is
therefore most commonly based on case reports or case series initially and then,
several years after the drug has reached the market, on pharmacoepidemiological
studies. Case reports are most commonly limited by retrospective reporting (“report-
ing bias”) and lack of information about important potential confounding factors,
making causality difficult or impossible to determine. However, history has shown
that many known teratogens have been identified through case reports (e.g., thalido-
mide, isotretinoin, mycophenolate) and that such reports therefore may be valuable,
especially when the particular outcome is infrequent in the unexposed population
and the risk is clearly increased.
Pharmacoepidemiological studies in pregnancy typically have case-control or
cohort design and are based on data from adverse drug reaction surveillance sys-
tems, pregnancy registries, clinical networks (e.g., teratology information services),
or large health-care databases (e.g., a prescription registry linked to a medical birth
registry). Such studies are vital to our understanding of the safety of medications in
pregnancy, but as they are based on observational data, great care must be taken in
the analysis and interpretation of their results. Major methodological challenges are
related to sample size/power, errors in measurement of medication use, bias in the
form of, e.g., selection bias and information bias (misclassification of exposure or
outcome), and confounding. Potential sources of confounding in studies with psy-
chotropic drugs are sociodemographic status, alcohol intake, smoking, nutritional
status including vitamin intake, and concomitant drug therapy.
A major challenge in pregnancy studies is confounding by indication, i.e., to
distinguish between a fetal effect caused by the drug under study and by the mater-
nal disorder being treated. In fact, in many cases the underlying (and untreated) ill-
ness might pose a greater risk to the unborn child than the medication. Including a
disease comparison group (women with the same disease and ideally also the same
disease severity, but not treated with the drug), or comparing drug use across differ-
ent indications, offers advantages over studies comparing exposed cases to healthy
controls only.
In all pharmacoepidemiological studies, sources of errors or bias should be
acknowledged and discussed and preferably quantified by performing sensitivity
analysis of estimates under an array of assumptions about possible bias directions
and magnitudes.
Many fetal adverse events (e.g., malformations, neonatal persistent pulmonary
hypertension) occur spontaneously at low rates, and most human teratogens increase
such risk modestly. For example, 800 exposed cases and 800 unexposed controls
are required to detect a twofold increased risk for malformations with 80 % statisti-
cal power and an alpha value of 0.05, assuming a baseline prevalence rate of 3 %
(Strom 2012). When studying specific malformations, the absolute risk is
15 Safety of Psychotropic Drugs in Pregnancy and Breastfeeding 301
Table 15.1 Overview of the most commonly used medications for treatment of mental disorders
according to their safety during pregnancy
First trimester Second trimester Third trimester
exposure exposure exposure
Antidepressantsa May be used. Avoid May be used May be used. Transient
paroxetine due to a perinatal complications
possible increased risk of may occur in infants
cardiac defects exposed up to delivery
Antipsychotics May be used May be used May be used. Transient

dyskinesias may occur in
neonates exposed up to
delivery, particularly for
first generation drugs
Lithium Avoid if possible. Small Avoid if possible May be used

increased risk of cardiac
defects (in particular
Ebstein’s anomaly)
Carbamazepine Should be avoided. Avoid if possible Avoid if possible

Increased risk of
malformations
Valproic acid Contraindicated. Contraindicated. Risk of Contraindicated. Risk of

Considerably increased long-term cognitive long-term cognitive
risk of malformations impairment impairment
Lamotrigine May be used May be used May be used

Anxiolytics and May be used sporadically May be used sporadically May be used
hypnotics sporadically. Respiratory
distress may occur in
infants exposed up to
delivery
Light gray: Generally considered safe, but uncertainties may exist, e.g., for some of the drugs
within a group or related to the total amount of data available
Medium gray: Increased risk of harmful effects cannot be excluded
Dark gray: The fetal risk exceeds the therapeutic advantage for the mother in the treatment of
mental disorders in pregnancy
a
For other antidepressants than selective serotonin reuptake inhibitors, see text
considerably lower (i.e., the background risk of any cardiac defects is about 1 %; of
a neural tube defect about 1:1000), and several thousand exposed cases are often
needed to detect a true increase in fetal risk.
15.2.1 Antidepressants
Antidepressants, and in particular selective serotonin reuptake inhibitors (SSRIs),

are probably the most studied and debated drug group with respect to safety in preg-
nancy. There has been a marked increase in the use of antidepressants in pregnancy
over the last 10 years, with the most recent prevalence data ranging from 3 % to 4 %
in Europe and in Australia to more than 8 % in North America (Nordeng 2016).
The majority of studies that have evaluated the safety of antidepressants in early
pregnancy do not indicate an enhanced risk of major congenital malformations after
exposure to SSRIs in general (Koren and Nordeng 2012). A meta-analysis including
studies published on SSRIs up to 2010, however, suggested an increased risk of
cardiac defects (especially septal defects) with first-trimester exposure to parox-
etine. For cardiac defects, the odds ratio (OR) was 1.46, with a 95 % confidence
interval (CI) of 1.17–1.82 (Wurst et al. 2010).
In one of our studies based on data from more than 60,000 pregnant women and
their children in the Norwegian Mother and Child Cohort Study, we found that
exposure to SSRIs during the first trimester was not associated with an increased
risk of overall congenital malformations (adjusted OR 1.22; 95 % CI 0.81–1.84) or
cardiovascular malformations (adjusted OR 1.51; 95 % CI 0.67–3.43) (Nordeng
et al. 2012). Exposure to antidepressants in general was not associated with low
birth weight (adjusted OR 0.62; 95 % CI 0.33–1.16). In the crude analysis, exposure
to antidepressants was associated with preterm birth (OR 1.46; 95 % CI 1.04–2.04);
however, after adjustment for depression and sociodemographic characteristics, this
increased risk disappeared (adjusted OR 1.21; 95 % CI 0.87–1.69). Moreover, hav-
ing symptoms of depression in pregnancy was in itself associated with preterm birth
(adjusted OR 1.13; 95 % CI 1.03–1.25). These findings are reassuring and illustrate
the importance of adjustment for the underlying illness in pharmacoepidemiologi-
cal studies. Without adjustment, antidepressant use would wrongly have been
accused for being associated with preterm birth (Nordeng et al. 2012).
Studies have shown conflicting results as to whether SSRIs increase the risk of
other outcomes in the newborn, such as low birth weight and perinatal complica-
tions; however most of these studies have not been able to distinguish whether this
is due to antidepressant treatment or is caused by the underlying mental disorder. A
systematic review of the literature showed that depression in itself increased the
risks for low birth weight by 49 %, for intrauterine growth restriction by 45 %, and
for preterm delivery by 39 % (Grote et al. 2010).
In a literature review on third trimester safety of antidepressants, we found that
neonatal adverse findings were quite common among infants exposed to SSRIs
close to delivery (2–3 out of 10 exposed infants) (Nordeng and Spigset 2005).
Respiratory distress, irritability, and suckling difficulties were most frequent and
resolved spontaneously in most cases within 2 weeks postpartum (Nordeng and
Spigset 2005; Moses-Kolko et al. 2005). However, such findings are relatively com-
mon also among unexposed newborns. Although several studies have found a statis-
tical significant association between prenatal exposure to SSRIs and persistent
pulmonary hypertension of the newborn (OR: 2.5; 95 % CI 1.3–4.7), the absolute
risk of this serious complication remains low (3–12 of 1000 exposed infants)
(Grigoriadis et al. 2013).
As serotonin plays a critical role in hemostasis, concern has been raised with
respect to perinatal bleeding complications. We identified no overall increased risk
of vaginal bleeding during pregnancy or after delivery among women exposed to

antidepressants. Interestingly, women not medicated with antidepressants but with
depressive symptoms had a moderately increased risk of vaginal bleeding in early
and mid pregnancy (Lupattelli et al. 2014).
Recently, concerns about long-term neurodevelopmental outcomes after prena-
tal antidepressant exposure have been raised. Animal studies have found exposure
to elevated serotonin levels to be associated with abnormal development of the
central nervous system (Schaefer et al. 2015). Results from human neurodevelop-
mental studies, however, have been inconsistent: some link antidepressants to
adverse neurodevelopmental outcomes like ADHD, autism spectrum disorders, and
motor and cognitive dysfunction, whereas others do not (Gentile and Galbally
2011). A recent review of the literature found no clear evidence of negative effects
by SSRIs on early neurodevelopment, but studies on later neurodevelopment
showed conflicting results with respect to untoward effects on psychomotor, cogni-
tive and language development, temperament, and behavioral problems (Hermansen
and Melinder 2015). General recommendations for SSRI use in pregnancy are sum-
marized in (Table 15.1).
Tricyclic antidepressants (TCA) and serotonin/noradrenaline reuptake inhibitors
(SNRIs) are less well studied than SSRIs, but may be used in women who do not
respond to SSRIs. Most data exists for amitriptyline, clomipramine, imipramine,
nortriptyline, and venlafaxine (Schaefer et al. 2015). Data do not indicate any
increased risk of major malformations, but most studies have been too small or have
other methodological limitations to draw definite conclusions. An increased risk of
poor neonatal adaptation has been reported after use of venlafaxine during preg-
nancy, as for other serotonergic antidepressants.
15.2.2 Antipsychotics
Antipsychotics are prescribed to 0.2–0.3 % of all pregnant women (Petersen et al.

2014), mainly for bipolar disorder and schizophrenia. First-generation antipsycho-
tics are known to affect the menstrual cycle and may also reduce fertility secondary
to the group’s prolactin-increasing effects. Second-generation antipsychotics, on
the other hand, have metabolic side effects that may increase the risk of excessive
maternal weight gain during pregnancy, gestational diabetes, and preeclampsia.
None of the published studies indicate an increased risk of teratogenic effects,
though less data are available for the newest second-generation antipsychotics
(Galbally et al. 2014).
Concern has been regarding the use of antipsychotics in the third trimester and
transient dyskinesias in the newborn. These findings, however, have been mostly
related to the use of first-generation antipsychotics. Most guidelines nevertheless
recommend continuing antipsychotic treatment until delivery to avoid the risk of
relapse and to monitor the newborn for possible neuromuscular adverse effects and
feeding problems. General recommendations for the use of antipsychotics in preg-

nancy are summarized in (Table 15.1).
15.2.3 Mood Stabilizers
Lithium has traditionally been considered a teratogenic drug due to an association

with cardiac malformations, in particular Ebstein’s anomaly. This risk, however,
may have been overestimated, and today it is generally considered that the absolute
risk of Ebstein’s anomaly after first-trimester exposure to lithium is low (0.5–1:1000,
as compared to 1: 20,000 in the general population). Lithium should be reserved for
cases where drugs with a more favorable fetal risk profile are unsuitable. Women
using lithium in pregnancy could be offered an early ultrasound examination to
exclude possible fetal cardiac defects (Schaefer et al. 2015). The newborn should be
monitored for signs of neuromuscular toxicity first week postpartum.
Carbamazepine, valproic acid, and lamotrigine are antiepileptic drugs frequently
used in patients with bipolar disorders. Carbamazepine and valproic acid should be
avoided in pregnant women with psychiatric illness due to an increased risk of mal-
formations and adverse neurodevelopmental effect in the offspring. Both carba-
mazepine and valproic acid interfere with folate metabolism, which may be one of
the mechanisms involved. First-trimester exposure to carbamazepine has been asso-
ciated with a twofold to fivefold increased risk of neural tube defects. An increased
risk of other malformations, e.g., cleft palate, cardiac malformations, and hypospa-
dias, has also been reported. A meta-analysis including more than 60,000 women
with epilepsy using carbamazepine as monotherapy reported an overall malforma-
tion rate of 5.7 % (95 % CI 3.7–7.7 %) (Meador et al. 2008), which is significantly
higher than the expected baseline rate of 3–4 %. There is plausible evidence of a
dose-response effect with increased risk with doses above 1000 mg/day (Tomson
et al. 2011).
Valproic acid is one of the most teratogenic antiepileptic drugs and should not be
prescribed to women in reproductive age. The meta-analysis mentioned above
found a malformation rate of 17.6 % (95 % CI 5.3–30.0 %) among women on val-
proic acid monotherapy (Meador et al. 2008). The most common malformations
include neural tube defects, cardiac defects, and defects of the extremities and uro-
genital tract, with a dose-dependent increased risk at least for spina bifida and hypo-
spadia (Vajda and Eadie 2005). It has also been suggested that a fetal valproic acid
syndrome, including specific anomalies of the face and fingers, may exist (Kozma
2001). Finally, exposure to valproic acid is linked to impaired cognitive function
with lower IQ scores, especially after exposure to doses above 800–1000 mg/day
(Bromley et al. 2014).
A principal limitation of the literature on the safety of carbamazepine and val-
proic acid in pregnancy is that most studies have been conducted among women
with epilepsy where the underlying disease in itself is associated with an increased
fetal risk. Moreover, the doses are generally higher than those used in psychiatric
disorders. However, as the risks related to use of these drugs are clearly significant,
it is not justifiable to be less restrictive in the recommendations in bipolar disorder
than in epilepsy, although the risk possibly is lower.
Knowledge on the safety of lamotrigine in pregnancy stems from thousands of
pregnancies showing no increased risk of malformations or long-term adverse
effects in prenatally exposed children. Lamotrigine could therefore be considered
the antiepileptic drug of choice for pregnant women with bipolar disorders (Schaefer
et al. 2015). General recommendations for the use of mood stabilizers in pregnancy
are summarized in (Table 15.1).
15.2.4 Anxiolytics and Hypnotics
Among benzodiazepines and benzodiazepine-like hypnotics (zopiclone, zolpidem,

and zaleplon), most data on safety during pregnancy exists for diazepam and zolpi-
dem. One meta-analysis and a recent literature review concluded that there is little
evidence for major malformations after use of benzodiazepines during pregnancy
(Dolovich et al. 1998; Bellantuono et al. 2013). A study including more than 500
early pregnancy exposures to zolpidem found no overall increased risk of major
malformations (Wikner and Källen 2011). There were four cases of non-atresial
intestinal malformations (0.8 cases expected), but the authors suggest that this find-
ing could be spurious due to multiple testing (Wikner and Källen 2011).
Prolonged use of benzodiazepines, particularly in high doses and close to deliv-
ery, has been associated with neonatal pharmacological effects and withdrawal
symptoms, including poor suckling, floppy infant syndrome, and respiratory dis-
tress. Moreover, after prolonged use of benzodiazepines throughout pregnancy, con-
cern has been raised about unfavorable long-term neurodevelopmental effects.
However, although some data indicate that a small number of children may have a
delayed development during the first year or so, they have generally developed nor-
mally by 4 years of age (McElhatton 1994).
Several important limitations exist for studies on the safety of benzodiazepines
in pregnancy. Most studies did not have information about duration of therapy or
indication for use, they did not differentiate between the various benzodiazepines,
and in most studies there was a high degree of co-medication with other psychotro-
pic drugs. Moreover, recall bias has been a major problem in the studies employing
a case-control design (Dolovich et al. 1998). Women using benzodiazepines also
drink more alcohol, are more often smokers, and have lower socioeconomic status
than women not using benzodiazepines in pregnancy, factors that clearly could con-
found the results.
According to current knowledge, benzodiazepines and benzodiazepine-like hyp-
notics could be prescribed for short-term use in the lowest possible doses during
pregnancy. Special precautions should be taken close to delivery. General recom-
mendations for the use of anxiolytics and hypnotics in pregnancy are summarized
in (Table 15.1).
15.3 Drug Safety During Breastfeeding
Human milk represents the ideal source of nutrients for small infants and provides
superior immunological and antioxidant protection to milk substitutes (Newton
2004). As the infant should not unnecessarily be denied the benefits of breast milk,
women are strongly encouraged to breastfeed whenever possible (World Health
Organization 2003; American Academy of Pediatrics 2012). The obvious dilemma
when treating a breastfeeding mother with a psychotropic drug is weighing the
potential risk to the infant due to drug exposure through the milk against the disad-
vantage of not receiving breast milk. Another alternative, stopping or not commenc-
ing maternal drug treatment, might be even more harmful, taking into account the
risk for the mother and thereby indirectly also for the infant if the mother is not
receiving adequate treatment for her mental disorder (Cornisha et al. 2005).
Specific questions to be answered when deciding how to handle drug treatment
in a woman with a mental disorder postpartum include: What are the risks for the
mother and the infant if the maternal disease is not adequately treated? How strong
is the mother’s desire to breastfeed? What are the disadvantages for the infant of not
receiving breast milk? What are the risks for the infant of being exposed to the
medication through breast milk? Is there any evidence to suggest that some specific
drugs within a therapeutic group are more favorable than others to use related to
infant risk? If necessary, could any practical strategies be used to reduce drug expo-
sure to the infant? And finally, if there is a (often small or even just theoretical) risk
of adverse effects in the infant due to drug exposure and breastfeeding nevertheless
is allowed, should the infant be monitored in any way?
In order to provide precise answers to these questions, knowledge about to which
extent the various psychotropic drugs are excreted into breast milk and about the
infant’s age-related capacity to metabolize these drugs is a prerequisite. Moreover,
knowledge about the theoretical infant dose and plasma concentrations of the drugs
that could be expected in the infant and whether any adverse effects have been
reported is also required. Unfortunately, available information about milk or infant
plasma concentrations and effects of psychotropic drugs in breastfed infants is
almost exclusively based upon case reports and small case series rather than upon
prospective studies with unexposed control groups. For many drugs, information is
extremely sparse.
15.3.1 Pharmacokinetic Considerations
Passage of psychotropic drugs between maternal plasma and breast milk is based
upon principles of passive diffusion and will therefore follow a gradient from high
to low concentration of free (unbound) drug. The timing of breastfeeding in relation
to the time of maternal ingestion of the drug will thus be an important determinant
for the concentration of the drug in the milk. When the concentration reaches its
peak level in the maternal plasma, usually within a couple of hours after intake, it
will, after a short delay, also reach its highest level in breast milk. Thereafter, as the
maternal plasma drug concentration gradually declines, the concentration in breast

milk will decrease until the mother ingests the next dose. For drugs with a short
elimination half-life, such as zolpidem (t1/2 = 2–3 h), the risk of adverse effects in
the infant can thus be reduced by breastfeeding at times when the drug concentra-
tion in breast milk is at its lowest level. This can be achieved – as obvious for zolpi-
dem but also valid for non-hypnotic drugs with short elimination half-lives – by
taking the daily dose in the evening and avoiding breastfeeding during the night.
However, the vast majority of psychotropic drugs have long or very long elimina-
tion half-lives with quite stable concentrations in breast milk. For such drugs, avoid-
ing breastfeeding during the peak concentration period will only reduce the infant’s
drug intake to a small extent. Simulation of data from a study from our group of the
excretion of paroxetine (Öhman et al. 1999) showed that total infant exposure would
be reduced only by about 20–30 % by avoiding the peak phase in milk.
As psychotropic drugs are lipophilic, concentrations in breast milk increase and
decrease in parallel with the milk triglyceride content. In two previous studies from
our group, we have shown that the concentrations of paroxetine and aripiprazole
were considerably higher in hindmilk than in foremilk, related to the higher triglyc-
eride content in hindmilk (Öhman et al. 1999; Nordeng et al. 2014a). However, as
also the nutritional value of milk is linked to its triglyceride levels, efforts to avoid
additional drug exposure by discarding milk with high triglyceride levels, such as
hindmilk, cannot be recommended.
Because newborns, and in particular premature infants, have an immature liver and
kidney function, they eliminate drugs at a lower rate than older children and adults
(for an overview related to psychotropic drugs, see Spigset and Hägg 1998). This
applies both to enzymes belonging to the cytochrome P-450 (CYP) system (by which
most psychotropic drugs are metabolized) and to the family of glucuronidating (UGT)
enzymes (by which, e.g., oxazepam and lamotrigine are metabolized). For example,
the elimination half-life of diazepam (which is mostly metabolized by CYP2C19) is
approximately 80 h in premature infants, about 30 h in full-term newborns, and
10–20 h in infants after the newborn period. During the first months of life, hepatic
function gradually matures, and after about 3–4 months of age, the metabolic capacity
reaches adult levels. In contrast, renal function is not fully developed until the infant
is at least 6 months old. Therefore, drugs with a high degree of renal elimination, such
as lithium, are of particular concern related to accumulation in young infants.
In addition to the pharmacokinetic aspects discussed above, premature or seri-
ously ill infants will often have a lower tolerance to the pharmacological action of
drugs than healthy children, including possible unfavorable effects.
15.3.2 Calculation of Infant Dose
If the drug concentration in breast milk is known, the infant’s theoretical dose can
be estimated by multiplying this value with the volume of milk that the infant
ingests. In such calculations, milk intake in an infant who is fully breastfed is stan-
dardized to 150 ml/kg body weight per day. To assess the risk of adverse effects, the
estimated dose can then be related, e.g., to the recommended pediatric dosage for
that drug for individuals of the same age. For example, for lamotrigine a fully
breastfed infant is exposed to amounts corresponding to between 25 % and 50 % of
the therapeutic pediatric dosage for infants, indicating that there might be a risk of
pharmacological effects in the infant (Nordmo et al. 2009; Hale 2014).
It is even more common to calculate the infant’s weight-adjusted relative dose,
i.e., the dose the infant ingests per kilogram bodyweight in relation to the maternal
dose per kilogram bodyweight. Infant exposure is regarded as minimal when the
relative dose is below 2 %, small when the relative dose is 2–5 %, moderate when
the relative dose is 5–10 %, and high when the relative dose is above 10 %. With
relative doses above 10 %, it is generally considered that a risk of pharmacological
effects in the infant does exist (Hale 2014; Ito 2000). With lower relative doses than
10 %, breastfeeding is in principle assumed to be safe, although infrequent cases of
possible untoward infant affection have been reported also for drugs with relative
doses below (but close to) 10 %, such as citalopram and aripiprazole (Hale 2014;
Berle and Spigset 2011). On the other hand, breastfeeding is not necessarily always
contraindicated during maternal treatment with drugs for which the relative doses
are above 10 %, such as lithium and lamotrigine (see later).
15.3.3 Evidence of Adverse Effects
Most data on adverse effects in breastfed infants are derived from case reports
(Rubin et al. 2004). Such reports are clearly of interest for drugs for which no or
very little previous data exist, but it is often complicated or even impossible to deter-
mine whether there is a causal connection between reported infant symptoms and
drug exposure.
A literature review of 183 reports on the use of psychotropic drugs during breast-
feeding (Fortinguerra et al. 2009) found that infant adverse effects had been pub-
lished for all groups of psychotropic drugs (but not for all individual drugs!).
Notably, another review found that in about 80 % of the cases with adverse effects
after maternal drug use in general, the infant was younger than 2 months of age
(Anderson et al. 2003). This is accordance with the expected gradual maturation of
hepatic and renal function during the first months of life and clearly illustrates that
infant age is a critical factor to take into account when assessing the individual
infant risk.
Very few prospective and systematic studies have compared the occurrence of
symptoms possibly related to psychotropic drug exposure in breastfed infants of
mothers taking and not taking the medication under study. One of a few such studies
has been performed by our group (Berle et al. 2004). In that study, excretion of selec-
tive serotonin reuptake inhibitors (SSRIs) or venlafaxine to breast milk was investi-
gated in 25 mothers with 26 exposed infants. Ten common symptoms of SSRI
exposure (regurgitation/vomiting, irritability, tremor, suckling or feeding problems,
decreased or increased sleep, yawning, etc.) were rated by the mothers and compared
to a control group of 68 breastfed infants of the same age where the mother did not
use any medication. There were no differences between the groups neither regarding
any of the specific symptoms nor regarding the total symptom score (5.9 in the anti-
depressant group vs. 7.6 in the control group on a scale ranging from 0 to 30).
15.3.3.1 Antidepressants
Depression is a common illness among women in the postpartum period with an

incidence of 10–15 % (O’Hara and Schwain 1996). In addition to the potentially
harmful effects the depression may have on the mother, challenges exist related to
caretaking of the newborn infant. Chronic maternal depression in the first year
Table 15.2 Infant doses and plasma concentrations of newer antidepressants after exposure via
breast milk
Absolute infant plasma Relative infant plasma
Drug Relative infant dosea concentration concentrationb
Selective serotonin reuptake inhibitors
Citalopram 3–10 % Negligiblec Up to 10 %d
Escitalopram 3–6 % <5 ng/ml <4 %
Fluoxetinee <12 % Up to 100 ng/mlf Up to 80 %f
Fluvoxamine <2 % <LODg –
Paroxetine 0.5–3 % <LODg –
Sertraline 0.5–3 % <LODg –
Other newer antidepressants
Venlafaxineh 6–9 % Up to 40 ng/ml Up to 30 %
Duloxetine <1 % <LODg –
Reboxetine 1–3 % <5 ng/ml <2 %
Bupropioni 2% <LODg –
Mirtazapine 0.5–3 % 0.2 ng/ml <1 %
a
Infant daily dose per kg body weight expressed as a percentage of maternal daily dose per kg body
weight. A value below 10 % is generally considered negligible
b
Infant plasma concentration expressed either as a percentage of the measured maternal plasma
concentration or as a percentage of what could be considered a low therapeutic concentration in
adults
c
In most cases below the lower limits of detection for the analytical methods employed, which
were mostly in the range of 2–5 ng/ml. However, in a few cases, which also have been associated
with suspected adverse effects, concentrations up to 15 ng/ml have been found
d
In a few cases, which also have been associated with suspected adverse effects, concentrations up
to about 50 % of the therapeutic concentration range have been found
e
Sum of fluoxetine and the active metabolite norfluoxetine
f
In some cases, which also have been associated with suspected adverse effects, concentrations as
high as about 500 ng/ml, i.e., clearly within the therapeutic concentration range, have been found
g
Below the lower limits of detection for the analytical methods, which were mostly in the range of
1–5 ng/ml
h
Sum of venlafaxine and the active metabolite O-desmethylvenlafaxine
i
Including one or several of the active metabolites of bupropion
postpartum is associated with delayed psychomotor development in the child at

15 months and may also affect the cognitive and emotional development (Cornisha
et al. 2005; Poobalan et al. 2007).
Key data on infant exposure for antidepressants are presented in Table 15.2.
Regarding the choice of specific drug, it is sometimes recommended that sertraline
should be preferred over other SSRIs due to the low infant exposure to that drug
(Gentile 2007). It has also been recommended that when possible, fluoxetine and
citalopram should be avoided or used with caution due to the higher infant plasma
levels (Table 15.2) and the possible risk of adverse effects such as irritability, sleep
disturbances, colic, and poor suckling.
Nevertheless, irrespective of which SSRI if the mother has been treated with dur-
ing pregnancy, we suggest that the same drug could also be used in the postpartum
period (Berle and Spigset 2011; Berle et al. 2004). Based upon the current literature
including a prospective study from our group (Berle et al. 2004), we recommend
that when antidepressant treatment is indicated in the postpartum period, the woman
should generally not be advised to discontinue breastfeeding. Some reviews and
guidelines recommend infant monitoring, particularly if the infant is sick, is prema-
ture, or has a low body weight (Weissman et al. 2004; Lanza di Scalea and Wisner
2009). However, given the very low risk of any untoward effects, we consider there
is no general need for routine follow-up examinations.
The numbers of exposed infants vary significantly between drugs, with about
100 cases for fluoxetine, paroxetine, sertraline, and citalopram, but less than 25 for
the other newer antidepressants. Some degree of uncertainty inevitably exists for the
drugs with the lowest numbers of exposed infants, even when no adverse effects
have been reported. On this basis, drugs for which little data exist, such as fluvox-
amine, venlafaxine, duloxetine, reboxetine, bupropion, and mirtazapine, should not
be viewed as first-line therapies, but could be considered in special cases.
15.3.3.2 Antipsychotics
The risk of a psychotic reaction is higher in the postpartum period than anytime
during a woman’s life. In a study of female patients with bipolar disorder, 36 % had
had their first psychotic episode in the postpartum period (Hunt and Silverstone
1995). In another study of women who previously had been hospitalized with psy-
chiatric symptoms, the risk of postpartum psychosis was increased about 100-fold
(Nager et al. 2008). Thus, treatment with antipsychotics related to breastfeeding is
not an uncommon issue.
In general, sparse data exist for the excretion of second-generation antipsychot-
ics in breast milk, and even less data is available regarding possible effects in the
breastfed infants. In total, about 60 cases for olanzapine, 20–30 cases for quetiapine,
about 10 cases for risperidone and clozapine, 4 cases for aripiprazole, and 2 cases
for ziprasidone have been published to date (Nordeng et al. 2014a; Hale 2014;
Klinger et al. 2013). Key data on infant exposure for these drugs are presented in
Table 15.3.
Table 15.3 Infant doses and plasma concentrations of second-generation antipsychotics and
mood stabilizers after exposure via breast milk
Absolute infant plasma Relative infant plasma
Drug Relative infant dosea concentration concentrationb
Second-generation antipsychotics
Aripiprazole Up to 8.3 %c NAd NAd
Clozapine 1.2 % NAd NAd
Olanzapine 1–4 % <LODe,f –f
Quetiapine 0.1–0.4 % 1.4 ng/ml 6%
Risperidone 2–5 %g <LODe –
Ziprasidone 1.2 % <LODe –
Mood stabilizers
Lithium Up to 40–80 % Up to 0.7 mmol/l Up to ~100 %
Carbamazepine 1–8 % 0.5–1.0 μg/mlh 10–40 %
Lamotrigine Up to 20 % 0.8–1.6 μg/ml 20–50 %
Valproic acid 0.5–4 %i 0.5–2.0 μg/ml 4%
a
Infant daily dose per kg body weight expressed as a percentage of maternal daily dose per kg body
weight. A value below 10 % is generally considered negligible
b
Infant plasma concentration expressed either as a percentage of the measured maternal plasma
concentration or as a percentage of what could be considered a low therapeutic concentration in
adults
c
Sum of aripiprazole and the active metabolite dehydroaripiprazole
d
No data available
e
Below the lower limit of detection for the analytical methods employed, which were mostly in the
range of 1–5 ng/ml
f
In one infant at one specific point of time the concentration was 40 % of the maternal concentra-
tion
g
Sum of risperidone and the active metabolite 9-hydroxyrisperidone
h
A concentration of 4.5 μg/mL, which is within the therapeutic range, has been reported in a single
case
i
Maximum 7 % in a single case
In a case series of four breastfed infants exposed to clozapine, agranulocytosis

was reported in one and drowsiness in another of the infants (Dev and Krupp 1995).
Although the relative infant dose is low, these adverse effects could be readily
explained from the drug’s adverse effect profile. Due to the severity of these reac-
tions, it is generally recommended that women treated with clozapine should avoid
breastfeeding.
For olanzapine and quetiapine, some prospective data are available, although
only five and six subjects have been included, respectively (Gardiner et al. 2003;
Misri et al. 2006). Whether the infant dose seems to be even lower for quetiapine
than for olanzapine, it is remarkable that two of the six infants exposed to quetiapine
(Misri et al. 2006) scored slightly below the normal range for mental and psycho-
motor development (82/82 and 84/91, respectively; normal range 85–115 for both
subscales). However, as no quetiapine was found in breast milk in any of these
cases, the low scores may well have been unrelated to the drug exposure. No untow-
ard effects have been suspected for olanzapine. For risperidone, infant exposure is
in the same range as for olanzapine and quetiapine, and no untoward effects have
been reported. In a recent case study, we found that infant exposure to aripiprazole
was higher than previously reported and also higher than for olanzapine, quetiapine,
and risperidone (Nordeng et al. 2014a). Although no adverse effects were observed
in our case, others have reported that somnolence could occur (Hale 2014). Thus,
we recommend that infants exposed to aripiprazole should be monitored for poten-
tial adverse effects including drowsiness, poor feeding, and sleeping pattern
changes. For ziprasidone very little information exists; thus any specific advice can-
not be given.
In the counseling of mothers treated with second-generation antipsychotics in
the postpartum period, an individualized risk/benefit approach should be applied
regardless of which drug is preferred. As stated above, most evidence of non-risk
exists for olanzapine. However, with the exception of clozapine, no clear-cut evi-
dence of harmful effects exists for the other drugs either.
15.3.3.3 Mood Stabilizers
For lithium, there have been two reports of toxicity in breastfed infants exposed
via breast milk (for a review, see Spigset and Hägg 1998). In these infants, the
plasma levels of lithium were 0.6 and 0.7 mmol/l, respectively. The infants pre-
sented with symptoms such as tremor and involuntary movements, and the high
lithium concentrations were thought to be caused by decreased renal clearance due
to dehydration during a common cold. Based upon these reports, lithium is often
considered not compatible with breastfeeding. However, there are also studies sug-
gesting that lithium administration is not an absolute contraindication to breast-
feeding: Although the infant plasma levels regularly amount to 20–40 % of the
maternal plasma concentration (Spigset and Hägg 1998; Viguera et al. 2007), most
often no untoward effects have been observed. Therefore, if the mother has a
strong desire to breastfeed and the infant is healthy, breastfeeding could be allowed.
In these cases, however, the mother should be instructed to watch for symptoms of
lithium toxicity in the infant, e.g., neuromuscular affection and feeding difficul-
ties, and be particularly observant during situations where the infant could become
dehydrated (infections, vomiting, etc.). It might also be helpful to monitor infant
plasma concentrations of lithium, creatinine, and possibly also thyroxin and thy-
roid-stimulating hormone.
For valproic acid and carbamazepine, infant exposure is generally low
(Table 15.3). For carbamazepine there are a few reports on possible adverse effects
in breastfed infants, including sedation and poor suckling. However, these reports
are not clear-cut and cannot be viewed as conclusive evidence for a relationship
(for a review, see Fortinguerra et al. 2009; Hägg and Spigset 2000). Carbamazepine
is generally considered compatible with breastfeeding, but as the infant plasma
concentrations of carbamazepine in some cases have been relatively high, it could
be pertinent to observe the infant for drowsiness and poor suckling, particularly if
the infant is premature or the mother is using combination therapy (Hale 2014;
Fortinguerra et al. 2009). For valproic acid, infant plasma levels after exposure via
breast milk are considerably lower than the therapeutic plasma concentrations
obtained in the treatment of seizures in infants and children (for a review, see
Fortinguerra et al. 2009; Hägg and Spigset 2000). Thus, valproic acid is consid-
ered fully compatible with breastfeeding (Hale 2014; Fortinguerra et al. 2009).
However, due to its teratogenic potential if the breastfeeding mother should
become pregnant once again, it has been argued that valproic acid nevertheless is
best avoided also in the postpartum period. Notably, recent recommendations from
the European Medicines Agency (EMA) state that valproic acid should only be
prescribed to women of fertile age if other treatments are ineffective or not toler-
ated and if so that the women should be advised to use effective contraception
(EMA 2014).
Lamotrigine is excreted in breast milk to a relatively high extent, and there is a
consistent finding across studies in mothers with epilepsy that the infant plasma
concentrations amount to about 20 % of the maternal drug levels but in some cases
up to 50 % (Hale 2014; Newport et al. 2008), i.e., levels where pharmacological
effects cannot be excluded. Although more than 50 cases of exposed infants have
been reported in the literature, infant adverse reactions have been reported in a sin-
gle case only. In this case, severe apnea was seen in a 16-day-old infant (Nordmo
et al. 2009). Based upon these data, an individualized risk/benefit approach should
be applied, and, if breastfeeding is allowed, it would be advisable to monitor the
infant for adverse effects such as sedation and poor suckling. Moreover, if adverse
reactions are suspected, the infant plasma concentration of lamotrigine should be
measured.
15.3.3.4 Anxiolytics and Hypnotics
In general, benzodiazepines have long elimination half-lives, which are even longer
in neonates than in adults. For diazepam (plus the active metabolite desmethyldia-
zepam), the relative infant dose ranges from 3 % to 14 %. Infant plasma concentra-
tions above 10 % of the maternal levels are regularly seen, and a certain percentage
of exposed infants have been reported to have signs of central nervous system (CNS)
affection such as sedation, lethargy, and lack of response to stimuli (Spigset and
Hägg 1998; Hale 2014; Rubin et al. 2004; Kelly et al. 2012). Thus, diazepam in
more than single doses is not considered compatible with breastfeeding. Oxazepam
is excreted to a lower extent and is, although less studied, probably preferable to
diazepam. However, also oxazepam has been associated with infant CNS affection
and should, if allowed during breastfeeding, be used in the lowest possible dose and
for the shortest possible time. Moreover, the infant should in such cases be observed
for signs of CNS depression. If high doses and/or long-term treatment are required
in young infants, breastfeeding should preferably be stopped.
When treatment with hypnotics is indicated during breastfeeding, we suggest
that benzodiazepines with relatively long elimination half-lives, such as flunitraze-
pam and nitrazepam, should be avoided due to the risk of accumulation in the infant.
In contrast, zopiclone and zolpidem are more swiftly eliminated. Of these, zolpidem
is, in contrast to zolpidem, completely cleared from the milk after 10 h, and as the
mother should be asleep and not breastfeed during the first hours after intake, infant
exposure for zolpidem will be close to zero (Spigset and Hägg 1998). Although both
zopiclone and zolpidem are often considered compatible with breastfeeding, we
recommend zolpidem as the preferred choice.
15.4 General Clinical Recommendations
A maternal mental disorder should be appropriately treated during pregnancy and in

the postpartum period because untreated or inadequately treated disease will result
in maternal suffering and could lead to poor compliance with prenatal and postnatal
care, missed maternity checkups, and poor lifestyle (i.e., inappropriate nutrition,
smoking, use of alcohol and illicit drugs), being potentially harmful both to the
mother and the child.
The risk of relapse in pregnancy after discontinuation of psychotropic medica-
tion is high, ranging from 52 % among women with bipolar disorder (Viguera et al.
2000) to 68 % in women with major depression (Cohen et al. 2006). Psychotherapy
is recommended as the treatment for mild to moderate mental disorders during preg-
nancy and should also be used complementary to pharmacotherapy.
Many women with moderate to severe mental illness will have to continue
treatment with psychotropic drugs during pregnancy. The selection of a psychotro-
pic drug should be guided by the women’s prior treatment response bearing in
mind the individual drug’s possible adverse effects on her pregnancy. Informing
about the pros and cons of pharmacological treatment requires focus on risk com-
munication and documentation of decision processes. Doctors should be aware
that women tend to overestimate the risks associated with the use of psychotropic
drugs during pregnancy (Nordeng et al. 2010; Petersen et al. 2015) and that adher-
ence to prescribed psychotropic drugs often is low (Lupattelli et al. 2015). In a
study among women in 18 countries, we found that low adherence in pregnancy
was reported by 51 % of women using drugs for anxiety disorders during preg-
nancy, 47 % using drugs for depression, and 43 % using drugs for other mental
disorders. The belief that the benefit of pharmacotherapy outweighed the risks was
positively correlated with higher medication adherence (r = 0.282; p < 0.001)
(Lupattelli et al. 2015).
The metabolism of many psychotropic drugs increases during pregnancy. This
effect is best documented and possibly also most pronounced for lamotrigine but is
apparent also for antidepressants and antipsychotics. Also the renal clearance of
lithium is increased in pregnancy, causing lower plasma concentrations than in non-
pregnant women. Thus, for psychotropic drugs in general, therapeutic drug moni-
toring is advised and it should be expected that the dosage has to be gradually
increased during pregnancy. Monitoring should preferably start prior to pregnancy
and be performed regularly throughout pregnancy. For lamotrigine a frequency of
every to every other month has been recommended (Pennel et al. 2008); for other
drugs a frequency of at least every 2–3 months has been suggested, depending on
the individual drug concentration measured. At delivery, the dosage should be low-
ered to that given before pregnancy.
Some psychotropic drugs can be safely given to breastfeeding mothers, but for
other drugs, an individual risk-benefit assessment has to be performed. Such an
assessment should be based upon factors including infant age, the importance of
breastfeeding for the mother, and the availability of alternative drugs with a more
favorable risk profile. When breastfeeding needs to be individually adapted, extra
efforts from the doctor in terms of providing information to the mother and per-
forming follow-ups of the mother and infant are often required. Such a solution also
presupposes that the mother is able to comply with the advices given.
15.5 Future Directions for Research
15.5.1 Pregnancy
Future research on the safety of psychotropic drugs in pregnancy should focus on

prospective, longitudinal studies with sufficient information on key confounding
factors including maternal mental health and psychiatric diagnosis, as well as ade-
quate follow-up times to assess long-term outcomes in children.
The possibility of record linkage between large health databases and birth cohorts
in Europe and North America has opened up for fast-track high-quality epidemio-
logical studies and will be important data sources for studies on drug safety in preg-
nancy in the future. International collaboration is also advancing and will not only
allow replication of previous findings but also pooling of data to investigate less
frequently used medications and rare pregnancy outcomes that none of the individ-
ual research institutions would have adequate study power to examine alone. This
possibility has already been explored among the five Nordic countries linking
national prescription databases and medical birth registries to study the risk of
stillbirth and infant mortality after prenatal exposure to SSRIs among more than 1.6
million singleton pregnancies (Stephansson et al. 2013).
As the size of the data sources increases, alternative study designs, like the
sibling-control design, can be applied to study medication safety in pregnancy
(Brandlistuen et al. 2013). The sibling-control design offers important advantages
over studies comparing unrelated individuals because siblings share familial envi-
ronment (e.g., maternal chronic disease) as well as on average 50 % of their
genetic setup, but they may differ in medication exposure during pregnancy. In
fact, it has been suggested that the sibling-control design may provide one of the
most effective approaches to control for family factors when large cohorts with a
sufficient number of discordant siblings are available (Susser et al. 2010). In addi-
tion, use of advanced techniques in biostatistics, including directed acyclic graphs
(DAGs), propensity score matching, and marginal structural models, will enable
us to control complex, time-varying exposures and confounding factors in drug

safety studies in pregnancy to obtain estimates of effects that are less biased than
traditional statistical methods, thereby giving us greater confidence in the validity
of the findings.
At the same time, it is becoming increasingly clear to scientists, clinicians, and the
public alike that the reproductive safety of medications cannot be assured without
studying long-term consequences for the child. In the near future, large birth cohorts
being followed into adolescence and even longer will enable such studies, thereby
filling an important knowledge gap on medication safety (Nordeng et al. 2014b).
Several birth cohorts are now also collecting biological samples of the mother
and/or child (e.g., blood, urine, hair). Enabling linkage between epidemiological
and biological data. Such data sources will be an extremely valuable data source in
the future. It opens up for pharmacogenetic analyses, biomarker analyses, and toxi-
cology analyses and even for new research fields like human pharmacoepigenetics.
Human studies have just started to emerge showing alterations in DNA methylation
patterns after in utero exposure to medications and linking these alterations to neu-
robehavioral disorders (Soubry et al. 2011; Non et al. 2014; van Mil et al. 2014).
These sparse human findings must be replicated, but if proven correct, they will
have profound consequences since it would imply that our current understanding of
pharmacology is an oversimplification.
15.5.2 Lactation
Obviously, a single published case report has a limited value in the guidance of
whether a woman treated with a drug could be allowed to breastfeed or not. However,
also single case reports can be valuable if performed and reported properly. Preferably,
repeated drug concentration measurements in milk during a full-dose interval should
be carried out, and also maternal and infant plasma drug concentrations are of inter-
est. Moreover, thorough and repeated clinical examinations of the infant should be
performed. Infant age is crucial and should be reported. If relevant, also active drug
metabolites and the maternal and infant CYP genotypes should be included.
Far too often, published reports of possible untoward reactions in breastfed
infants have severe methodological limitations. Most importantly, there is often not
possible to separate potential drug effects from the infant’s normal state or from
concurrent disease. In order to assess causality when an adverse effect is suspected
in an exposed infant, analysis of the infant plasma concentration of the drug is cen-
tral. Moreover, a period without breastfeeding, i.e., when the mother pumps and
discards milk, should be scheduled to observe whether the symptoms disappear. If
breastfeeding later is resumed and the symptoms return, this would provide an even
stronger indication of causality.
In order to further clarify the possible risks associated with drug exposure
through breast milk, prospective studies with age-matched control groups and with
systematic adverse effect ratings are clearly warranted. To date, only very few such
studies have been performed for psychotropic drugs. Although it is a challenge to
collect such large clinical materials, it is feasible if sufficient time, effort, and
patience are put into the task.
References
Anderson PO, Pochop SL, Manoguerra AS (2003) Adverse drug reactions in breastfed infants: less
than imagined. Clin Pediatr (Phila) 42(4):325–340
American Academy of Pediatrics (2012) Policy statement: breastfeeding and the use of human
milk. Pediatrics 129(3):e827–e841
Bellantuono C, Tofani S, Di Sciascio G, Santone G (2013) Benzodiazepine exposure in pregnancy
and risk of major malformations: a critical overview. Gen Hosp Psychiatry 35(1):3–8
Berle JO, Spigset O (2011) Antidepressant use during breastfeeding. Curr Womens Health Rev
7(1):28–34
Berle JO, Steen VM, Aamo TO, Breilid H, Zahlsen K, Spigset O (2004) Breastfeeding during
maternal antidepressant treatment with serotonin reuptake inhibitors: infant exposure, clinical
symptoms, and cytochrome p450 genotypes. J Clin Psychiatry 65(9):1228–1234
Brandlistuen RE, Yström E, Nulman I, Koren G, Nordeng H (2013) Prenatal paracetamol exposure
and child neurodevelopment: a sibling-controlled cohort study. Int J Epidemiol
42(6):1702–1713
Bromley R, Weston J, Adab N, Greenhalgh J, Sanniti A, McKay AJ et al (2014) Treatment for
epilepsy in pregnancy: neurodevelopmental outcomes in the child. Cochrane Database Syst
Rev 10:CD010236
Cohen LS, Altshuler LL, Harlow BL, Nonacs R, Newport DJ, Viguera AC et al (2006) Relapse of
major depression during pregnancy in women who maintain or discontinue antidepressant
treatment. JAMA 295(5):499–507
Cornisha AM, McMahona CA, Ungerera JA, Barnettb B, Kowalenkoc N, Tennant C (2005)
Postnatal depression and infant cognitive and motor development in the second postnatal year:
the impact of depression chronicity and infant gender. Infant Behav Dev 28:407–417
Dev VJ, Krupp P (1995) Adverse event profile and safety of clozapine. Rev Contemp Pharmacother
6:197–208
Dolovich LR, Addis A, Vaillancourt JM, Power JD, Koren G, Einarson TR (1998) Benzodiazepine
use in pregnancy and major malformations or oral cleft: meta-analysis of cohort and case-
control studies. BMJ 317(7162):839–843
EMA (2014) CMDh agrees to strengthen warnings on the use of valproate medicines in women
and girls. Available from http://www.ema.europa.eu/docs/en_GB/document_library/Press_
release/2014/11/WC500177638.pdf Cited 20.05.2015
Fortinguerra F, Clavenna A, Bonati M (2009) Psychotropic drug use during breastfeeding: a review
of the evidence. Pediatrics 124(4):e547–e556
Galbally M, Snellen M, Power J (2014) Antipsychotic drugs in pregnancy: a review of their mater-
nal and fetal effects. Ther Adv Drug Saf 5(2):100–109
Gardiner SJ, Kristensen JH, Begg EJ, Hackett LP, Wilson DA, Ilett KF et al (2003) Transfer of
olanzapine into breast milk, calculation of infant drug dose, and effect on breast-fed infants.
Am J Psychiatry 160(8):1428–1431
Gentile S (2007) Use of contemporary antidepressants during breastfeeding: a proposal for a spe-
cific safety index. Drug Saf 30(2):107–121
Gentile S, Galbally M (2011) Prenatal exposure to antidepressant medications and neurodevelop-
mental outcomes: a systematic review. J Affect Disord 128(1-2):1–9
Grigoriadis S, VonderPorten EH, Mamisashvili L, Eady A, Tomlinson G, Dennis CL et al (2013)
The effect of prenatal antidepressant exposure on neonatal adaptation: a systematic review and
meta-analysis. J Clin Psychiatry 74(4):e309–e320
Grote NK, Bridge JA, Gavin AR, Melville JL, Iyengar S, Katon WJ (2010) A meta-analysis of
depression during pregnancy and the risk of preterm birth, low birth weight, and intrauterine
growth restriction. Arch Gen Psychiatry 67(10):1012–1024
Hägg S, Spigset O (2000) Anticonvulsant use during lactation. Drug Saf 22(6):425–440
Hale TW (2014) Medications and mothers’ milk, 16th edn. Hale Publishing, Plano
Hermansen TK, Melinder A (2015) Prenatal SSRI exposure: effects on later child development.
Child Neuropsychol 21(5):543–569
Hunt N, Silverstone T (1995) Does puerperal illness distinguish a subgroup of bipolar patients? J
Affect Disord 34(2):101–107
Ito S (2000) Drug therapy for breast-feeding women. N Engl J Med 343(2):118–126
Kelly LE, Poon S, Madadi P, Koren G (2012) Neonatal benzodiazepines exposure during breast-
feeding. J Pediatr 161(3):448–451
Klinger G, Stahl B, Fusar-Poli P, Merlob P (2013) Antipsychotic drugs and breastfeeding. Pediatr
Endocrinol Rev 10(3):308–317
Koren G, Nordeng H (2012) Antidepressant use during pregnancy: the benefit-risk ratio. Am J
Kozma C (2001) Valproic acid embryopathy: report of two siblings with further expansion of the
phenotypic abnormalities and a review of the literature. Am J Med Genet 98(2):168–175
Lanza di Scalea T, Wisner KL (2009) Antidepressant medication use during breastfeeding. Clin
Lupattelli A, Spigset O, Koren G, Nordeng H (2014) Risk of vaginal bleeding and postpartum
hemorrhage after use of antidepressants in pregnancy: a study from the Norwegian Mother and
Child Cohort Study. J Clin Psychopharmacol 34(1):143–148
Lupattelli A, Spigset O, Björnsdóttir I, Hameen-Anttila K, Mårdby AC, Panchaud A et al (2015)
Patterns and determinants of low adherence to psychotropic medications during pregnancy – a
multinational study. Depress Anxiety 32(6):426–436
McElhatton PR (1994) The effects of benzodiazepine use during pregnancy and lactation. Reprod
Toxicol 8(6):461–475
Meador K, Reynolds MW, Crean S, Fahrbach K, Probst C (2008) Pregnancy outcomes in women
with epilepsy: a systematic review and meta-analysis of published pregnancy registries and
cohorts. Epilepsy Res 81(1):1–13
Misri S, Corral M, Wardrop AA, Kendrick K (2006) Quetiapine augmentation in lactation: a series
of case reports. J Clin Psychopharmacol 26(5):508–511
Moses-Kolko EL, Bogen D, Perel J, Bregar A, Uhl K, Levin B et al (2005) Neonatal signs after late
in utero exposure to serotonin reuptake inhibitors: literature review and implications for clini-
cal applications. JAMA 293(19):2372–2383
Nager A, Sundquist K, Ramirez-Leon V, Johansson LM (2008) Obstetric complications and post-
partum psychosis: a follow-up study of 1.1 million first-time mothers between 1975 and
2003 in Sweden. Acta Psychiatr Scand 117(1):12–19
Newport DJ, Pennell PB, Calamaras MR, Ritchie JC, Newman M, Knight B et al (2008) Lamotrigine
in breast milk and nursing infants: determination of exposure. Pediatrics 122(1):e223–e231
Newton ER (2004) Breastmilk: the gold standard. Clin Obstet Gynecol 47(3):632–642
Non AL, Binder AM, Kubzansky LD, Michels KB (2014) Genome-wide DNA methylation in
neonates exposed to maternal depression, anxiety, or SSRI medication during pregnancy.
Epigenetics 9(7):964–972
Nordeng H (2016) Drug utilisation studies in pregnancy. In: Elseviers M, Andersen M, Benko R
(eds.). Drug utilization research: methods and application. Wiley-Blackwell, Hoboken, NJ, USA
Nordeng H, Spigset O (2005) Treatment with selective serotonin reuptake inhibitors in the third
trimester of pregnancy: effects on the infant. Drug Saf 28(7):565–581
Nordeng H, Ystrom E, Einarson A (2010) Perception of risk regarding the use of medications and
other exposures during pregnancy. Eur J Clin Pharmacol 66(2):207–214
Nordeng H, van Gelder MM, Spigset O, Koren G, Einarson A, Eberhard-Gran M (2012) Pregnancy
outcome after exposure to antidepressants and the role of maternal depression: results from the
Norwegian Mother and Child Cohort Study. J Clin Psychopharmacol 32(2):186–194
Nordeng H, Gjerdalen G, Brede WR, Michelsen LS, Spigset O (2014a) Transfer of aripiprazole to
breast milk: a case report. J Clin Psychopharmacol 34(2):272–275
Nordeng H, Ystrøm E, Eberhard-Gran M (2014b) Medication safety in pregnancy – results from
the MoBa-study. Nor J Epidemiol 24(1-2):161–168
Nordmo E, Aronsen L, Wasland K, Smabrekke L, Vorren S (2009) Severe apnea in an infant

exposed to lamotrigine in breast milk. Ann Pharmacother 43(11):1893–1897
O’Hara MW, Schwain AM (1996) Rates and risk of postpartum depression – a meta-analysis. Rev
Öhman R, Hagg S, Carleborg L, Spigset O (1999) Excretion of paroxetine into breast milk. J Clin
Pennell PB, Peng L, Newport DJ, Ritchie JC, Koganti A, Holley DK et al (2008) Lamotrigine in
pregnancy: clearance, therapeutic drug monitoring, and seizure frequency. Neurology 70(22 Pt 2):
2130–2136
Petersen I, McCrea RL, Osborn DJ, Evans S, Pinfold V, Cowen PJ et al (2014) Discontinuation of
antipsychotic medication in pregnancy: a cohort study. Schizophr Res 159(1):218–225
Petersen I, McCrea R, Lupattelli A, Nordeng H (2015) Women’s perception of risks in pregnancy:
multinational population based internet survey. BMJ Open 5(6):e007390
Poobalan AS, Aucott LS, Ross L, Smith WC, Helms PJ, Williams JH (2007) Effects of treating
postnatal depression on mother-infant interaction and child development: systematic review.
Br J Psychiatry 191:378–386
Rubin ET, Lee A, Ito S (2004) When breastfeeding mothers need CNS-acting drugs. Can J Clin
Pharmacol 11:e257–e266
Schaefer C, Peters PWJ, Miller RK (2015) Drugs during pregnancy and lactation: treatment
options and risk assessment, 3rd edn. Elsevier, Amsterdam
Soubry A, Murphy S, Huang Z, Murtha A, Schildkraut J, Jirtle R et al (2011) The effects of depres-
sion and use of antidepressive medicines during pregnancy on the methylation status of the
IGF2 imprinted control regions in the offspring. Clin Epigenetics 3:2
Spigset O, Hägg S (1998) Excretion of psychotropic drugs into breast milk. CNS Drugs
9:111–134
Stephansson O, Kieler H, Haglund B, Artama M, Engeland A, Furu K et al (2013) Selective sero-
tonin reuptake inhibitors during pregnancy and risk of stillbirth and infant mortality. JAMA
309(1):48–54
Strom BL, Kimmel SE, Hennessy S (eds.) (2012) Pharmacoepidemiology, 5th edn. Wiley-
Blackwell, Hoboken, NJ, USA
Susser E, Eide MG, Begg M (2010) Invited commentary: the use of sibship studies to detect famil-
ial confounding. Am J Epidemiol 172(5):537–539
Tomson T, Battino D, Bonizzoni E, Craig J, Lindhout D, Sabers A et al (2011) Dose-dependent
risk of malformations with antiepileptic drugs: an analysis of data from the EURAP epilepsy
and pregnancy registry. Lancet Neurol 10(7):609–617
Vajda FJ, Eadie MJ (2005) Maternal valproate dosage and foetal malformations. Acta Neurol
Scand 112(3):137–143
van Mil NH, Steegers-Theunissen RP, Bouwland-Both MI, Verbiest MM, Rijlaarsdam J, Hofman
A et al (2014) DNA methylation profiles at birth and child ADHD symptoms. J Psychiatr Res
49:51–59
Viguera AC, Nonacs R, Cohen LS, Tondo L, Murray A, Baldessarini RJ (2000) Risk of recurrence
of bipolar disorder in pregnant and nonpregnant women after discontinuing lithium mainte-
nance. Am J Psychiatry 157(2):179–184
Viguera AC, Newport DJ, Ritchie J, Stowe Z, Whitfield T, Mogielnicki J et al (2007) Lithium in
breast milk and nursing infants: clinical implications. Am J Psychiatry 164(2):342–345
Weissman AM, Levy BT, Hartz AJ, Bentler S, Donohue M, Ellingrod VL et al (2004) Pooled
analysis of antidepressant levels in lactating mothers, breast milk, and nursing infants. Am J
Psychiatry 161(6):1066–1078
Wikner BN, Källen B (2011) Are hypnotic benzodiazepine receptor agonists teratogenic in
humans? J Clin Psychopharmacol 31(3):356–359
World Health Organization (WHO) (2003) Infant and young child feeding. World Health
Organization, Geneva
Wurst KE, Poole C, Ephross SA, Olshan AF (2010) First trimester paroxetine use and the preva-
lence of congenital, specifically cardiac, defects: a meta-analysis of epidemiological studies.
Birth Defects Res A Clin Mol Teratol 88(3):159–170

Pharmacovigilance in Psychiatry Edoardo PDF

Uploaded by

Copyright:

Available Formats

Pharmacovigilance in Psychiatry Edoardo PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Pharmacovigilance in Psychiatry Edoardo PDF

Uploaded by

Copyright:

Available Formats

What is the book about?

What is the book about?

What are some of the risks discussed regarding psychotropic drug use during pregnancy and breastfeeding?

What are some of the risks discussed regarding psychotropic drug use during pregnancy and breastfeeding?

Edoardo Spina

Gianluca Trifirò Editors

ISBN 978-3-319-24739-7 ISBN 978-3-319-24741-0 (eBook)

Library of Congress Control Number: 2015957257

Springer Cham Heidelberg New York Dordrecht London

Printed on acid-free paper

Adis is a brand of Springer

A number of large-size observational studies have been conducted in several

Messina, Italy Gianluca Trifirò

Part I General Aspects of Pharmacovigilance in Psychiatry

1 Pharmacovigilance in Psychiatry: An Introduction . . . . . . . . . . . . . . 3

Part II Pharmacovigilance of Specific Psychotropic Drug Classes

8 Safety and Tolerability of Antidepressants . . . . . . . . . . . . . . . . . . . . . 149

9 Safety and Tolerability of Antipsychotics . . . . . . . . . . . . . . . . . . . . . . 167

Part III Pharmacovigilance of Psychotropic Drugs in Special Populations

13 Safety of Psychotropic Drugs in Children and Adolescents. . . . . . . . 257

Edoardo Spina, MD, PhD Department of Clinical and Experimental Medicine,

Eric Acquaviva Department of Child and Adolescent Psychiatry, Assistance

Antonio Clavenna Department of Public Health, Laboratory for Mother and

Department of Pharmacotherapy, University of North Texas Health Science Center

Edoardo Spina, Gianluca Trifirò, and Achille Patrizio Caputi

Abstract The discipline of pharmacovigilance is of particular importance in the

Keywords Pharmacovigilance • Psychiatry • Psychotropic drugs • Adverse drug

The World Health Organization (WHO) defines pharmacovigilance as “science and

E. Spina (*) • G. Trifirò • A.P. Caputi

© Springer International Publishing Switzerland 2016 3

contribute to pharmacovigilance activities. Such activities are a critical aspect of

noradrenaline, was marketed in some European countries as an antidepressant and

The constant developments in pharmacovigilance and psychiatry led to the

Berson A, Descatoire V, Sutton A, Fau D, Maulny B, Vadrot N, Feldmann G, Berthon B, Tordjmann

Paola Maria Cutroneo and Giovanni Polimeni

Keywords Adverse drug reactions • Pharmacovigilance • Adverse events • Safety

P.M. Cutroneo (*) • G. Polimeni

© Springer International Publishing Switzerland 2016 9

2.1 Adverse Drug Reactions: Terminologies and Definitions

An adverse drug reaction (ADR) is described by the World Health Organization

2.2 Social and Economic Impact of Adverse Drug Reactions

emergency hospitalizations (95 % CI 55,531–143,724) occurred annually for

2.3 Pharmacovigilance for Evaluating Adverse Drug

A medicinal product is authorized by competent authorities if the benefit–risk bal-

2.3.1 Pharmacovigilance: Definitions and Objectives

2.3.2 Examples of Safety Regulatory Actions and Drug

increased risk of suicidality in children and adolescents. Moreover, epidemiologic

2.4 Classification of ADRs

Adverse drug reactions can be difficult and sometimes impossible to distinguish

2.4.1 Type A Adverse Event

Type A reactions are due to an exaggerated, but otherwise normal, pharmacologi-

2.4.2 Type B Adverse Event

• Type IV reactions are described as delayed or hypersensitivity reactions. These

Table 2.3 Classification of adverse drug reactions

Al Hamid A, Ghaleb M, Aljadhey H et al (2014) A systematic review of hospitalization resulting

Francesco Salvo, Annie Fourrier-Réglat, Nicholas Moore, Bernard Bégaud,

© Springer International Publishing Switzerland 2016 27

following sections, we will present the interests, pitfalls and complementarity of

Keywords Psychotropic drugs • Safety • Monitoring • Limitations •