Therapeutic Monitoring of Antiepileptic Drugs For Epilepsy (Review)

Therapeutic monitoring of antiepileptic drugs for epilepsy
(Review)
Tomson T, Dahl ML, Kimland E
This is a reprint of a Cochrane review, prepared and maintained by The Cochrane Collaboration and published in The Cochrane Library
2010, Issue 5
http://www.thecochranelibrary.com
Therapeutic monitoring of antiepileptic drugs for epilepsy (Review)

Copyright © 2010 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
TABLE OF CONTENTS
HEADER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
ABSTRACT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
PLAIN LANGUAGE SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
BACKGROUND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
OBJECTIVES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
METHODS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
RESULTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
DISCUSSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
AUTHORS’ CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
ACKNOWLEDGEMENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
CHARACTERISTICS OF STUDIES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
DATA AND ANALYSES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
Analysis 1.1. Comparison 1 TDM versus control, Outcome 1 12-month remission. . . . . . . . . . . . . 9
Analysis 1.2. Comparison 1 TDM versus control, Outcome 2 Seizure free last 12-month follow up. . . . . . . 10
Analysis 1.3. Comparison 1 TDM versus control, Outcome 3 Adverse effects. . . . . . . . . . . . . . . 10
Analysis 1.4. Comparison 1 TDM versus control, Outcome 4 Completed 2-year follow up. . . . . . . . . . 11
APPENDICES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
WHAT’S NEW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
HISTORY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
CONTRIBUTIONS OF AUTHORS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
DECLARATIONS OF INTEREST . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
SOURCES OF SUPPORT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
INDEX TERMS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
Therapeutic monitoring of antiepileptic drugs for epilepsy (Review) i

[Intervention Review]
Torbjörn Tomson1 , Marja-Liisa Dahl2 , Elin Kimland3

1 Department of Neurology, Karolinska University Hospital, Stockholm, Sweden. 2 Department of Clinical Pharmacology, Akademiska
sjukhuset, 75185 Uppsala, Sweden. 3 Adverse Drug Monitoring Centre, Clinical Pharmacology L7:05, Karolinska University Hospital
- Solna, Stockholm, Sweden
Contact address: Torbjörn Tomson, Department of Neurology, Karolinska University Hospital, Stockholm, S-171 76, Sweden.
torbjorn.tomson@karolinska.se.
Editorial group: Cochrane Epilepsy Group.

Publication status and date: New search for studies and content updated (no change to conclusions), published in Issue 5, 2010.
Review content assessed as up-to-date: 5 February 2010.
Citation: Tomson T, Dahl ML, Kimland E. Therapeutic monitoring of antiepileptic drugs for epilepsy. Cochrane Database of Systematic
Reviews 2007, Issue 2. Art. No.: CD002216. DOI: 10.1002/14651858.CD002216.pub2.
ABSTRACT
Background
This is an updated version of the original Cochrane review published in Issue 1, 2007.
The aim of drug treatment for epilepsy is to prevent seizures without causing adverse effects. To achieve this, drug dosages need to be
individualised. Measuring antiepileptic drug levels in body fluids (therapeutic drug monitoring) is frequently used to optimise drug
dosage for individual patients.
Objectives
To review the evidence regarding the effects of therapeutic drug monitoring upon outcomes in epilepsy.
Search methods
We searched the Cochrane Epilepsy Group Specialised Register (February 2010), the Cochrane Central Register of Controlled Trials
(CENTRAL) (The Cochrane Library 2010, Issue 1) and MEDLINE (1950 to January week 4, 2010). No language restrictions were
imposed. We checked the reference lists of retrieved articles for additional reports of relevant studies.
Selection criteria
Randomised controlled trials comparing the outcomes of antiepileptic drug monotherapy guided by therapeutic drug monitoring with
drug treatment without the aid of therapeutic drug monitoring.
Data collection and analysis
We based this review on published aggregate data. The main outcomes measured were the proportions of patients achieving a 12-month
remission from seizures, reporting adverse effects, and being withdrawn from the treatment they had been randomised to receive.
Main results
Only one study met the inclusion criteria for the review. In this open study, 180 patients with newly-diagnosed, untreated epilepsy were
randomised to treatment with the antiepileptic drug selected by their physician either with or without therapeutic drug serum level
monitoring as an aid to dosage adjustments. The antiepileptic drugs used were carbamazepine, valproate, phenytoin, phenobarbital
and primidone. A 12-month remission from seizures was achieved by 60% of the patients randomised to therapeutic drug monitoring
Therapeutic monitoring of antiepileptic drugs for epilepsy (Review) 1
(intervention group) and by 61% in the control group. A total of 56% in the intervention group and 58% in the control group were
seizure free during the last 12 months of follow up. Adverse effects were reported by 48% in the intervention group and 47% of the
control group patients. Of those randomised to therapeutic drug monitoring, 62% completed the two-year follow up compared with
67% of the control group.
Authors’ conclusions
We found no clear evidence to support routine antiepileptic drug serum concentration measurement with the aim of reaching predefined
target ranges for the optimisation of treatment of patients with newly-diagnosed epilepsy with antiepileptic drug monotherapy. However,
this does not exclude the possible usefulness of therapeutic drug monitoring of specific antiepileptic drugs during polytherapy, in special
situations or in selected patients, although evidence is lacking.
PLAIN LANGUAGE SUMMARY
No evidence to support routine therapeutic monitoring of antiepileptic drugs in the treatment of epilepsy.
No evidence was found to indicate that the routine measurement of serum drug concentrations to inform drug dose adjustments is
superior to drug dose adjustments made on clinical grounds alone in newly-diagnosed epilepsy patients treated with a single drug:
carbamazepine, valproate, phenytoin, phenobarbital or primidone. One under-powered study was found, and this review does not
exclude the possibility that therapeutic drug monitoring might be useful in patients with newly-diagnosed epilepsy, nor does it exclude
the possible usefulness of monitoring in special situations or in selected patients.
BACKGROUND
may be a wide variation in response to a drug given in a standard
This review is an update of a previously published review in The dose. Some patients may suffer from poor efficacy whereas others
Cochrane Database of Systematic Reviews (Issue 1, 2007) on may experience toxic effects unless the dosage is individualised.
’Therapeutic monitoring of antiepileptic drugs for epilepsy’.
Epilepsy is a disorder characterised by spontaneously occurring Phenytoin was the drug of choice for most seizure types when drug
recurrent epileptic seizures. It has been estimated that more than level measurements were introduced in the 1960s and it was found
40 million people in the world have epilepsy; the main treatment to exhibit particularly complex dose-dependent pharmacokinet-
is with antiepileptic drugs. The aim of treatment is to prevent ics. As a consequence, it is extremely difficult to predict the effect
seizures without causing side effects. The treatment of epilepsy of a dose change of this drug. Studies carried out in the 1960s
needs to be individualised with respect to both choice of drugs and 1970s demonstrated a correlation between the concentration
and drug dosage. The latter became apparent during the 1960s of phenytoin in serum and its therapeutic and toxic effects, thus
when methods for measuring serum concentrations of antiepilep- measuring the serum concentration of phenytoin was soon estab-
tic drugs were developed (Buchtal 1960). The serum concentra- lished as a guide to individualised dosing (Kutt 1968; Kutt 1974;
tion of an antiepileptic drug varies markedly between patients Lund 1974). Since then, drug level monitoring (therapeutic drug
given the same dosage; the reason for this is that people differ in monitoring) has been established as a routine aid to optimising
their ability to absorb, distribute, metabolise and excrete drugs. treatment with other antiepileptic drugs. The goal of therapeu-
These processes are summarised in the term ’pharmacokinetics’. tic drug monitoring is to optimise a patient’s clinical outcome by
The rate at which these pharmacokinetic processes proceed may be managing the medication regimen, assisted by the measurement
influenced by factors such as the formulation of the drug, concur- of drug concentrations. In general, therapeutic drug monitoring
rent disease, concomitant medication and genetic variables of the is considered to be of potential value when there is a need for in-
individual patient. Thus, many factors contribute to differences dividualised dosing owing to marked inter-individual differences
in pharmacokinetics and to the individual variability in the serum in drug response when such differences are accounted for by vari-
concentration of a drug. As a consequence of this variability, there ations in pharmacokinetics, and when it is difficult to monitor
drug treatment by direct observation of the therapeutic response Criteria for considering studies for this review
and adverse effects, as is sometimes the case in epilepsy. Although
most antiepileptic drugs do not share the problems of dose-depen-
dent kinetics that phenytoin has, many display pronounced inter-
Types of studies
individual variability in pharmacokinetics that suggest the need
for individualised dosing. Furthermore, the serum concentration Randomised controlled trials comparing the outcomes of
of many antiepileptic drugs can be affected by interactions with antiepileptic drug therapy guided by therapeutic drug monitoring
other drugs. Therapeutic drug monitoring may facilitate the iden- with drug treatment without the aid of drug monitoring. Both ad-
tification of such interactions. equately and quasi-randomised, and blinded and unblinded trials
were to be included.
The concept of therapeutic drug monitoring rests on the assump-
tion that drug concentration correlates better with clinical effects
than dose. For some antiepileptic drugs, target ranges of serum
concentration have been determined. These are drug concentra- Types of participants
tions known to be associated with a high probability of seizure People with epilepsy who were receiving treatment with antiepilep-
control and low risk of toxicity. However, comparatively few stud- tic drugs as monotherapy. The review included patients of all ages
ies have been designed specifically to explore the relationship be- with different seizure types and the use of all established antiepilep-
tween serum concentrations and effects of antiepileptic drugs, and tic drugs to prevent seizures. Two separate cohorts of patients were
the documentation in this respect for many of the drugs is scarce to be analysed:
(Tomson 2000). Provided there is a distinct concentration-effect (1) patients with newly-diagnosed epilepsy starting treatment;
relationship within the individual, therapeutic drug monitoring (2) patients with established epilepsy on continuous treatment
may be justifiable to control for changes due to drug interactions. with antiepileptic drugs.
This is also possible in the absence of a defined target drug con- Children (under 16 years of age) and adults were to be analysed
centration range. It is, however, likely that the value of therapeutic separately, if sufficient data were available.
drug monitoring will vary with the different antiepileptic drugs,
depending on their pharmacological properties.
The therapeutic drug monitoring service may vary in its meth- Types of interventions
ods. Drug concentrations can be measured in specimens other The use of therapeutic drug monitoring to optimise antiepileptic
than serum, such as saliva, while analytical methods with varying drug therapy versus drug therapy without guidance by therapeutic
specificities may also be used. Moreover, there are different ways drug monitoring.
in which the results of the analysis are presented to the treating It is recognised that antiepileptic drug levels may be analysed in
physician. This may be just the crude drug concentration or, in many ways. Drug concentrations may be measured in different
some settings, it may be part of a more comprehensive pharma- specimens, such as serum, plasma or saliva. Total as well as un-
cokinetic service with suggestions for dose adjustments. bound serum concentrations may be analysed, and various ana-
lytical methods can be used. Moreover, the results of the analy-
Given the heterogeneity in the concept of therapeutic drug mon-
sis can be presented to the treating physician in different ways:
itoring and in the pharmacological characteristics of antiepileptic
as the actual drug concentration, as the drug level together with
drugs, it is not surprising that the use of therapeutic drug moni-
a suggested target range, or together with an interpretation with
toring varies markedly and that we lack consensus concerning the
suggestions for dose adjustments as part of a more comprehensive
value of its application in epilepsy (Chadwick 1987). The focus of
pharmacokinetic service.
this review is, therefore, on studies assessing the extent to which
In this review any measurement of antiepileptic drug concentra-
therapeutic drug monitoring contributes towards the greater ef-
tion that was made in order to assist the treating physician in his
fectiveness of antiepileptic drug treatment.
or her therapeutic decision making was to be considered.
OBJECTIVES Types of outcome measures

To assess if the use of therapeutic drug monitoring improves the (1) Proportion of patients achieving a 12-month remission from
outcome of drug treatment for epilepsy in terms of improved seizures.
seizure control and reduced adverse drug effects. (2) Proportion of patients reporting adverse effects considered by
the investigator to be drug related during the observation period.
(3) Proportion of patients withdrawn from the treatment to which
METHODS they had been randomised.

(4) Proportion of patients achieving at least a 50% reduction in (a) Patient characteristics: age; type of epilepsy; aetiology of
number of seizures during the period of observation. This out- epilepsy.
come measure is not applicable to cohorts with newly-diagnosed (b) Therapeutic drug monitoring method: measurement of bound
epilepsy. or unbound drug concentrations; nature of specimens analysed
The review did not consider surrogate outcomes such as number (for example, saliva versus plasma); relation of timing of taking
of patients achieving serum concentrations of antiepileptic drugs sample for analysis and dosing; whether the therapeutic drug mon-
within a specific target range. itoring results given to treating physicians included interpretation
of results and suggestions for dose adjustments versus provision
of crude drug levels only; whether the laboratory analysing drug
levels was taking part in a quality control program or not.
Search methods for identification of studies (4) Statistical heterogeneity was to be assessed using the chi-
The original search methods for earlier versions of this review have squared test for heterogeneity and I2 statistic. Provided no signif-
been archived in Appendix 1. icant heterogeneity was found, results were to be summarised in a
For the most recent update of this review, we searched as follows: fixed-effect meta-analysis.
(a) Cochrane Epilepsy Group Specialised Register (6 February (5) The primary analysis was by intention to treat, including all pa-
2010); tients randomised to treatment guided by therapeutic drug mon-
(b) CENTRAL (The Cochrane Central Register of Controlled itoring or treatment without drug monitoring, whether the ran-
Trials (The Cochrane Library Issue 1, 2010) using the search strat- domised patients completed the evaluation period or not.
egy outlined in Appendix 2; (6) Dichotomous data was to be presented as odds ratios and
(c) MEDLINE (Ovid, 1950 to January week 4, 2010) using the relative risks, and analysed using both fixed-effect and random-
search strategy outlined in Appendix 3. effects models.
No language restrictions were imposed. We checked the reference (7) Sensitivity analyses were to be made including all studies and
lists of retrieved articles for additional reports of relevant studies. also only those using adequate methods of randomisation.
Data collection and analysis RESULTS

Selection of studies
Two review authors (TT and MD) discarded irrelevant citations
based on the titles of publications and their abstracts. If there was
Description of studies
any suggestion that an article could possibly be relevant it was See: Characteristics of included studies; Characteristics of excluded
retrieved for further assessment. studies.
The same two review authors assessed independently the method- Only one prospective randomised study was identified that met
ological quality of each trial. They recorded details of method of the main inclusion criteria of this review (Jannuzzi 2000). The
randomisation, concealment of randomisation and use of inten- focus of that study was on the clinical impact of therapeutic drug
tion-to-treat analysis. Disagreements were resolved by discussion. monitoring in patients with newly-diagnosed epilepsy. One hun-
Data extraction dred and eighty patients, aged 6 to 65 years, with untreated partial
Descriptive characteristics and study data were extracted by the or idiopathic generalised non-absence epilepsy, and who had a his-
same two review authors using a standard form. tory of at least two seizures in the previous month, were enrolled
Data analysis plan in this unblinded, prospective parallel-group study. The ability of
(1) The two cohorts, patients with newly-diagnosed epilepsy start- prospective participants to comply with the study protocol was
ing treatment and patients with established epilepsy on continu- evaluated and informed consent obtained from the patients or
ous treatment with antiepileptic drugs, were analysed separately; their guardians. All participants who met the eligibility criteria
a separate analysis for children and adults was also undertaken. were randomised through a central office into two groups using a
(2) The primary analysis was to include therapeutic drug moni- stratification procedure aimed at ensuring a balanced distribution
toring of all drugs. However, it was envisaged that the impact of of different types of epilepsy between groups. All participants were
drug monitoring may vary with different drugs, depending at least prescribed the antiepileptic drug selected as the most appropriate
in part on the pharmacokinetic and pharmacodynamic properties by their physician, including carbamazepine, valproate, pheny-
of the drugs. Subgroup analyses were, therefore, undertaken sep- toin, phenobarbital or primidone. In one group, dosage was ad-
arately for the different drugs. justed to achieve a steady-state serum antiepileptic drug concen-
(3) Clinical heterogeneity between trials was to be assessed by tration within a target range during a period of three months or
comparing the following: less, whereas, in the other groups, dosage was adjusted on purely

clinical grounds aimed at achieving optimal seizure control over relevant and the two-year follow up was adequate. The sample size
the shortest reasonable period. calculations were based on the assumption of a 50% seizure free-
Exclusion criteria were: pregnancy; history of drug or alcohol dom rate in the control group and in the intervention group a 25%
abuse; previous treatment with any antiepileptic drug; presence of absolute improvement corresponding to a 75% seizure freedom
any known progressive disease or a diagnosis of benign rolandic rate. A sample size of 146 patients was thus considered adequate
epilepsy; absence epilepsy; and epileptic encephalopathy. for a power of 80% at the P < 0.05 level. The investigators never-
The target ranges used were 10 to 20 µg/ml (40 to 80 µM) for theless enrolled 180 patients. The reasons for this are not declared.
phenytoin, 15 to 40 µg/ml (64 to 172 µM) for phenobarbital, 4 Although the expected seizure response rate in the control group
to 11 µg/ml (17 to 46 µM) for carbamazepine, and 40 to 100 was close to the one observed in the study, the assumed effect
µg/ml (280 to 700 µM) for valproate. For primidone treated pa- of therapeutic drug monitoring on the primary efficacy outcome
tients, only metabolically derived phenobarbital was used for ther- variable could be considered as unrealistically high. The limited
apeutic monitoring purposes. Blood samples for determination of number of patients may thus have contributed to the failure to
antiepileptic drug levels were collected in both groups in a sim- demonstrate an effect of the intervention in the present study. Ad-
ilar way but the results were not made available to the treating ditionally, although the potential value of drug level monitoring
physician in one of the groups. If no satisfactory response was is likely to vary between drugs, comparisons were made between
achieved after 6 to 12 months of treatment, and the physician the two groups regardless of type of treatment. Most patients in
thought that knowledge of serum drug concentration was nec- both the intervention group (52/93; 56%) and the control group
essary to continue treatment, patients randomised to the control (48/87; 55%) were treated with carbamazepine, while the number
group could be crossed over to the intervention group. However, receiving other antiepileptic drugs was much lower. A substantial
only one patient crossed over, after one month. Hence, this is not proportion (64/180; 36%) of the enrolled patients exited the study
likely to have had a significant effect on the overall results of the prematurely. Of these, 43 were lost to follow up. However, the
study. Patients were followed up for 24 months or until a change proportion of completers was similar in the two groups, and the
in treatment strategy was clinically indicated. The primary effi- analyses in the study of Januzzi et al used the intention-to-treat
cacy endpoint was the proportion of patients achieving complete population.
seizure remission during the previous 12 months of follow up and
the analyses based on the intention-to-treat population.
One additional prospective randomised study assessing the ef- Effects of interventions
fect of monitoring plasma anticonvulsant levels was identified
The proportion of patients who achieved a 12-month remission
(Fröscher 1981). In this, 127 outpatients with epilepsy were ran-
from seizures was 60% (56/93) (95% CI 50 to 70) in the group
domly assigned to two groups. In one, the treating physician was
with treatment supported by drug monitoring data, which was not
not informed of the results of plasma concentration determina-
significantly different from the 61% (53/87) (95% CI 50 to 71) in
tions, whereas, in the other, plasma levels were reported and the
the control group, relative risk (RR) 0.99 (95% CI 0.78 to 1.25).
treating physician was asked to try to keep plasma concentrations
As for the primary endpoint of the study, seizure freedom rates
within a therapeutic range. Each patient was followed for one
during the last 12 months of follow up, the results were also very
year. This study included patients with chronic epilepsy and those
close: 56% (52/93) (95% CI 45 to 66) in the intervention group
who had been on treatment for a considerable time before enrol-
and 58% (50/87) (95% CI 48 to 68) in the control group (not
ment. However, this study was excluded from the review since the
statistically significant), RR 0.97 (95% CI 0.75 to 1.26). Adverse
majority of patients were treated with a combination of different
effects were reported in 45 of 93 patients (48%, 95% CI 38 to
antiepileptic drugs (on average 2.5 per patient). In all, only 14
59) in the drug level monitoring group compared with 41 of the
patients taking part in this study received monotherapy.
87 in the control group (47%, 95% CI 36 to 58) (not statistically
significant), RR 0.84 (95% CI 0.64 to 1.11). The rates of the most
frequently reported types of side effects (somnolence, dizziness,
Risk of bias in included studies headache, irritability and fatigue) were also very similar in the
The only study that met our criteria was an open, prospective ran- two groups. Among the 93 patients in the intervention group,
domised study with a parallel-group design. The randomisation 58 (62%, 95% CI 52 to 72) completed the two-year follow up
procedure was adequate and the two treatment groups comparable compared with 58/87 (67%, 95% CI 56 to 76) in the control
with respect to age, seizure frequency and type of epilepsy. The group, RR 0.94 (95% CI 0.75 to 1.16). Of the 64 patients who
study design was open, but there would have been considerable exited the study prematurely, 43 were lost to follow up, 10 exited
practical problems associated with blinding procedures, and this is due to insufficient efficacy and/or adverse events, and 11 for other
not likely to have had a major impact on the results. The selected reasons.
primary efficacy endpoint, the proportion of participants achiev- Exclusion of all patients who did not complete the two-year follow
ing complete seizure remission during the previous 12 months, is up from the analysis of the primary endpoint of the study, seizure

freedom rates during the last 12 months of follow up, resulted in therapeutic drug monitoring have, therefore, not been addressed
a RR of 1.04 (95% CI 0.91 to 1.19). in the included study. The risk of chronic toxicity and the man-
Serum drug levels agement of late treatment failure are examples of situations where
The mean serum concentrations of carbamazepine and valproate drug level monitoring might play a role that could not be assessed
at nearly all time points were within the predefined target ranges in the present study. A further special feature of the included study
for patients in both groups; there were no significant differences is the key role of the target ranges of antiepileptic drug serum
between the groups in this respect. Serum phenobarbital levels concentrations. In the intervention group, dosage was adjusted
below the target range were twice as common in the control group to achieve serum concentrations within predefined target ranges.
as in the intervention group, but the numbers of participants were The outcome thus partly depended on the validity of the selected
low. After the last dosage adjustment, general serum drug levels target ranges. Their relevance for patients with newly-diagnosed
outside the target range were found in 6% of those participants epilepsy has in fact been questioned (Tomson 2000) and thera-
in the drug monitored group compared with 22% of the control peutic monitoring is, therefore, often employed without strict ad-
participants (P < 0.01). Of the patients who continued to have herence to target ranges but rather using the individual patient’s
seizures during the last 12 months of follow up, 1 out of 10 in the optimal drug concentration as a reference for future therapeutic
therapeutic drug monitoring group and 4 out of 11 in the control decisions. Such use of drug-level monitoring has not been evalu-
group had serum drug levels below the target range. ated in the included study. Neither can it be said that therapeutic
drug monitoring would not prove more useful as an aid for physi-
cians who are less experienced in the treatment of epilepsy and
thus less skilled in individualising drug dosage on clinical grounds
alone.
DISCUSSION
This review has demonstrated the lack of relevant randomised
studies assessing the impact of therapeutic drug monitoring in ef-
forts to optimise the drug treatment of newly-diagnosed epilepsy. AUTHORS’ CONCLUSIONS
The only identified study meeting our inclusion criteria provides
no evidence for the effectiveness of the routine use of therapeutic Implications for practice
monitoring of antiepileptic drugs during the first two years after
The result of this review, with only one study meeting our inclu-
the initiation of monotherapy treatment for patients with newly-
sion criteria, provides no clear evidence to support the routine use
diagnosed epilepsy. This does not exclude the possibility that ther-
of therapeutic drug monitoring for the optimisation of treatment
apeutic monitoring of antiepileptic drugs may be useful in individ-
with antiepileptic drugs of patients with newly-diagnosed epilepsy.
ual patients, in other stages of epilepsy, in patients on polytherapy
However, benefits of therapeutic drug monitoring in newly-diag-
or in other specific clinical situations. Nor do the results of this
nosed patients cannot be excluded, neither can its usefulness in
included study exclude some impact of therapeutic drug moni-
special situations or in selected patients, although strict evidence
toring on the outcome of the treatment of patients with newly-
is lacking.
diagnosed epilepsy.
The study failed to demonstrate a difference between the two treat- Implications for research
ment groups, with and without monitoring, but was powered to
detect only an improvement in response rate from 50% to 75%. This review has demonstrated the shortage of data from ran-
More realistic, less pronounced effects may still be of clinical rel- domised controlled studies concerning the effectiveness of using
evance. Such effects are not excluded, although they may seem therapeutic drug monitoring as an aid to dosage adjustment in
unlikely considering the very similar rates of seizure remission and antiepileptic drug monotherapy for newly-diagnosed epilepsy. Fu-
side effects in the two treatment groups. Furthermore, although ture studies should be appropriately powered, ideally analysing in-
it is likely that the value of therapeutic drug monitoring varies dividual antiepileptic drugs, and could consider not using strict
between drugs, owing to their individual pharmacokinetic prop- adherence to poorly defined target ranges of drug concentrations
erties, all antiepileptic drugs were analysed together in the identi- as part of the intervention. Although the practical and ethical dif-
fied study. Phenytoin, the antiepileptic drug for which therapeu- ficulties would be considerable, the extension of follow up beyond
tic drug monitoring is potentially most important, was used by two years would be of interest.
a small fraction of the patients and was unevenly distributed be-
tween the two groups. Additionally, although a two-year follow up
is unusually long for an intervention study in epilepsy, it should
be recognised that treatment continues for considerably longer
for the vast majority of patients. Potential long-term benefits of ACKNOWLEDGEMENTS

Mrs Jeannette Grünstein, Department of Clinical Pharmacology,
Karolinska University Hospital-Huddinge, for assistance with col-
lecting the articles.
REFERENCES
References to studies included in this review Kutt 1968
Kutt H, McDowell F. Management of epilepsy with
Jannuzzi 2000 {published data only} diphenylhydantoin sodium. JAMA 1968;203:969–72.
Jannuzzi G, Cian P, Fattore C, Gatti G, Bartoli A, Monaco
F, et al.A multicenter randomized controlled trial on the
clinical impact of therapeutic drug monitoring in patients Kutt 1974
with newly diagnosed epilepsy. Epilepsia 2000;41(2): Kutt H, Penry JK. Usefulness of blood levels of antiepileptic
222–30. drugs. Archives of Neurology 1974;31:283–8.
References to studies excluded from this review

Lefebvre 2009
Fröscher 1981 {published data only} Lefebvre C, Manheimer E, Glanville J. Chapter 6: Searching
Fröscher W, Eichelbaum M, Gugler R, Hildenbrand G, for studies. In: Higgins JPT, Green S (editors). Cochrane
Penin H. A prospective randomised trial on the effect of Handbook for Systematic Reviews of Interventions
monitoring plasma anticonvulsant levels in epilepsy. Journal Version 5.0.2 (updated September 2009). The Cochrane
of Neurology 1981;224:193–201. Collaboration, 2009. Available from www.cochrane-
handbook.org.
Additional references
Buchtal 1960 Lund 1974
Buchtal F, Svensmark O. Aspects of pharmacology of Lund L. Anticonvulsant effect of diphenylhydantoin relative
phenytoin (Dilantin) and phenobarbital relevant to their to plasma levels. Archives of Neurology 1974;31(5):289–94.
dosage in the treatment of epilepsy. Epilepsia 1960;1:
373–84. Tomson 2000
Chadwick 1987 Tomson T, Johannessen SI. Therapeutic monitoring of
Chadwick DW. Overuse of monitoring of blood the new antiepileptic drugs. European Journal of Clinical
concentrations of antiepileptic drugs. British Medical Pharmacology 2000;55:697–705.
Journal 1987;294:723–4. ∗
Indicates the major publication for the study

CHARACTERISTICS OF STUDIES
Characteristics of included studies [ordered by study ID]
Jannuzzi 2000
Methods Randomised, open, parallel study over a 24-month study period
Participants Patients of both sexes, aged 6 to 65 years with partial or generalised epilepsy
The median number of seizures was 3 during 4 months before intervention
Interventions Monotherapy with carbamazepine, phenytoin, valproate, phenobarbital or primidone guided by thera-
peutic drug monitoring versus without support of drug level monitoring
Outcomes Proportion achieving seizure remission in the last 12 months

Proportion remaining seizure free since initiation of treatment
Proportion with side effects at any time
Notes Under powered

Not possible to evaluate individual drugs
Risk of bias
Item Authors’ judgement Description
Allocation concealment? Yes A - Adequate
Characteristics of excluded studies [ordered by study ID]
Study Reason for exclusion
Fröscher 1981 Patients treated with antiepileptic drugs in polytherapy

DATA AND ANALYSES
Comparison 1. TDM versus control
No. of No. of
Outcome or subgroup title studies participants Statistical method Effect size
1 12-month remission 1 180 Risk Ratio (M-H, Fixed, 95% CI) 0.99 [0.78, 1.25]
2 Seizure free last 12-month follow 1 180 Risk Ratio (M-H, Fixed, 95% CI) 0.97 [0.75, 1.26]
up
3 Adverse effects 1 180 Risk Ratio (M-H, Fixed, 95% CI) 0.84 [0.64, 1.11]
4 Completed 2-year follow up 1 180 Risk Ratio (M-H, Fixed, 95% CI) 0.94 [0.75, 1.16]
Analysis 1.1. Comparison 1 TDM versus control, Outcome 1 12-month remission.
Review: Therapeutic monitoring of antiepileptic drugs for epilepsy
Comparison: 1 TDM versus control
Outcome: 1 12-month remission
Study or subgroup TDM Control Risk Ratio Weight Risk Ratio

n/N n/N M-H,Fixed,95% CI M-H,Fixed,95% CI
Jannuzzi 2000 56/93 53/87 100.0 % 0.99 [ 0.78, 1.25 ]
Total (95% CI) 93 87 100.0 % 0.99 [ 0.78, 1.25 ]

Total events: 56 (TDM), 53 (Control)
Heterogeneity: not applicable
Test for overall effect: Z = 0.10 (P = 0.92)
0.1 0.2 0.5 1 2 5 10

Favours control Favours TDM

Analysis 1.2. Comparison 1 TDM versus control, Outcome 2 Seizure free last 12-month follow up.
Outcome: 2 Seizure free last 12-month follow up

Jannuzzi 2000 52/93 50/87 100.0 % 0.97 [ 0.75, 1.26 ]
Total (95% CI) 93 87 100.0 % 0.97 [ 0.75, 1.26 ]

0.1 0.2 0.5 1 2 5 10

Analysis 1.3. Comparison 1 TDM versus control, Outcome 3 Adverse effects.
Outcome: 3 Adverse effects

Jannuzzi 2000 45/93 50/87 100.0 % 0.84 [ 0.64, 1.11 ]
Total (95% CI) 93 87 100.0 % 0.84 [ 0.64, 1.11 ]

0.1 0.2 0.5 1 2 5 10

Favours TDM Favours control

Analysis 1.4. Comparison 1 TDM versus control, Outcome 4 Completed 2-year follow up.
Outcome: 4 Completed 2-year follow up

Jannuzzi 2000 58/93 58/87 100.0 % 0.94 [ 0.75, 1.16 ]
Total (95% CI) 93 87 100.0 % 0.94 [ 0.75, 1.16 ]

0.1 0.2 0.5 1 2 5 10

APPENDICES
Appendix 1. Search methods for original version of this review

We searched the Cochrane Epilepsy Group Specialised Register (4 September 2006). This register contains reports of trials identified
from regular searches of the Cochrane Central Register of Controlled Trials (CENTRAL) and of MEDLINE. Relevant reports are also
identified by handsearching selected journals and conference proceedings.
In addition, we searched the Cochrane Central Register of Controlled Trials (The Cochrane Library Issue 4, 2005), MEDLINE (January
1966 to April 2005) and EMBASE (1974 to May 2005). Details of the search strategy used are given below. No language restrictions
were imposed. We checked the reference lists of retrieved articles for additional reports of relevant studies.
The following search strategy was used for MEDLINE and was modified for other databases.
1 RANDOMISED-CONTROLLED TRIAL in Publication Type
2 CONTROLLED-CLINICAL TRIAL in Publication Type
3 RANDOMISED-CONTROLLED-TRIALS / ALL
4 RANDOM-ALLOCATION / ALL
5 DOUBLE-BLIND-METHOD / ALL
6 SINGLE-BLIND-METHOD / ALL
7 #1 or #2 or #3 or #4 or #5 or #6
8 TG=ANIMAL not (TG=HUMAN and TG=ANIMAL)
9 #7 not #8
10 CLINICAL TRIAL in PT
11 Explode CLINICAL-TRIALS / ALL
12 CLIN* and TRIAL* in TI
13 CLIN* and TRIAL* in AB
14 (SINGL* or DOUBL* or TREBL* or TRIPL*) and (BLIND* or MASK*) in TI
15 (SINGL* or DOUBL* or TREBL* or TRIPL*) and (BLIND* or MASK*) in AB
16 PLACEBOS / ALL
17 PLACEBO*in TI
18 PLACEBO* in AB
19 RANDOM* in TI
20 RANDOM* in AB
21 Explode RESEARCH-DESIGN / ALL
22 #10 or #11or #12 or #13 or #14 or #15 or #16 or #17 or #18 or #19 or #20 or #21
23 #22 not #8
24 #23 or #9
25 TDM
26 DRUG-MONITORING (MeSH)
27 (THERAP* or DRUG*) and MONITOR*
28 (SERUM or PLASMA) and MONITOR*
29 (SALIVA and MONITOR*)
30 #25 or #26 or #27 or #28 or #29
31 DRUG-ADMINISTRATION-SCHEDULE (MeSH)
32 DOSE-RESPONSE-RELATIONSHIP-DRUG (MeSH)
33 PHARMACOKINETICS (MeSH)
34 (DRUG and ANALYSIS)
35 (SERUM or PLASMA) and DRUG-CONCENTRATION
36 (THERAPEUTIC and RANGE)
37 EPILEPSY (MeSH)
38 EPILEP*
39 ANTICONVULSANTS (MeSH)
40 ANTICONVULSANT*
41 ANTIEPILEP*
42 SEIZURE (MeSH)
43 SEIZURE*
44 CONVULSION*
45 PHENOBARBITAL (MeSH)
46 PHENOBARBITA*
47 PHENYTOIN (MeSH)
48 PHENYTOIN
49 CARBAMAZEPINE (MeSH)
50 CARBAMAZEPIN*
51 TOPIRAMATE
52 GABAPENTIN
53 FELBAMATE
54 OXCARBAZEPIN*
55 TIAGABINE
56 ZONISAMIDE
57 # 30 or #31 or #32 or #33 or #34 or #35 or #36
58 #37 or #38 or #39 or #40 or #41 or #42 or #43 or #44 or #45 or #46 or #47 or #48 or #49 or #50 or #51 or #52 or #53 or #54 or
#55 or #56
59 #57 and #58
60 #59 and #24

Appendix 2. CENTRAL search strategy
#1 (epilep* or seizure* or convulsion*)
#2 MeSH descriptor Epilepsy explode all trees
#3 MeSH descriptor Seizures explode all trees
#4 anticonvulsant*
#5 MeSH descriptor Anticonvulsants explode all trees
#6 antiepilep*
#7 phenytoin
#8 valpro*
#9 carbamazepine
#10 phenobarbit*
#11 MeSH descriptor Phenobarbital explode all trees
#12 MeSH descriptor Phenytoin explode all trees
#13 MeSH descriptor Carbamazepine explode all trees
#14 ethosuximide
#15 primidone
#16 topiramate
#17 gabapentin
#18 felbamate
#19 oxcarbazepin*
#20 tiagabine
#21 zonisamide
#22 levetiracetam
#23 lacosamide
#24 pregabalin
#25 rufinamide
#26 TDM
#27 MeSH descriptor Drug Monitoring explode all trees
#28 (therap* or drug*) NEAR/2 (monitor*)
#29 (serum or plasma) NEAR/2 (monitor*)
#30 saliva* NEAR/2 monitor*
#31 MeSH descriptor Drug Administration Schedule explode all trees
#32 MeSH descriptor Dose-Response Relationship, Drug explode all trees
#33 pharmacokinetics
#34 drug NEAR/2 analysis
#35 (serum or plasma) NEAR/2 (drug NEXT concentration)
#36 therapeutic NEXT range
#37 MeSH descriptor Pharmacokinetics explode all trees
#38 (#1 OR #2 OR #3)
#39 (#4 OR #5 OR #6 OR #7 OR #8 OR #9 OR #10 OR #11 OR #12 OR #13 OR #14 OR #15 OR #16 OR #17 OR #18 OR #
19 OR #20 OR #21 OR #22 OR #23 OR #24 OR #25)
#40 (#26 OR #27 OR #28 OR #29 OR #30 OR #31 OR #32 OR #33 OR #34 OR #35 OR# 36 OR #37)
#41 (#38 AND #39 AND #40)

Appendix 3. MEDLINE search strategy
This strategy is based on the Cochrane Highly Sensitive Search Strategy for identifying randomized trials published in Lefebvre 2009.
1. randomized controlled trial.pt.
2. controlled clinical trial.pt.
3. randomized.ab.
4. placebo.ab.
5. clinical trials as topic.sh.
6. randomly.ab.
7. trial.ti.
8. 1 or 2 or 3 or 4 or 5 or 6 or 7
9. exp animals/ not humans.sh.
10. 8 not 9
11. exp Epilepsy/
12. Seizures/
13. (epilep$ or seizure$ or convuls$).tw.
14. 11 or 12 or 13
15. exp Anticonvulsants/
16. anticonvulsant$.tw.
17. antiepilep$.tw.
18. phenytoin.tw.
19. valpro$.tw.
20. carbamazepine.tw.
21. ethosuximide.tw.
22. phenobarbit$.tw.
23. exp Phenobarbital/
24. primidone.tw.
25. exp Phenytoin/
26. exp Carbamazepine/
27. topiramate.tw.
28. gabapentin.tw.
29. felbamate.tw.
30. oxcarbazepin$.tw.
31. tiagabine.tw.
32. zonisamide.tw.
33. levetiracetam.tw.
34. lacosamide.tw.
35. pregabalin.tw.
36. rufinamide.tw.
37. or/15-36
38. TDM.tw.
39. exp Drug Monitoring/
40. ((therap$ or drug$) adj2 monitor$).tw.
41. ((serum or plasma) adj2 monitor$).tw.
42. (saliva$ adj2 monitor$).tw.
43. or/38-42
44. exp Drug Administration Schedule/
45. exp Dose-Response Relationship, Drug/
46. exp Pharmacokinetics/
47. (drug adj2 analysis).tw.
48. ((serum or plasma) adj2 (drug adj concentration)).tw.
49. (therapeutic adj range).tw.
50. or/43-49

51. 10 and 14 and 37 and 50
WHAT’S NEW
Last assessed as up-to-date: 5 February 2010.
Date Event Description
6 February 2010 New search has been performed Searches updated 6th February 2010; no new trials identified
HISTORY
Protocol first published: Issue 4, 2006
Review first published: Issue 1, 2007
Date Event Description
10 September 2008 Amended Converted to new review format.
CONTRIBUTIONS OF AUTHORS
Dr Tomson was primarily responsible for all aspects of the protocol design. Dr Dahl commented on the draft stage of the protocol.
Dr Kimland performed the literature search. Dr Tomson and Dr Dahl selected the studies according to the protocol. All authors
participated in writing the review.
DECLARATIONS OF INTEREST
None known.
SOURCES OF SUPPORT

Internal sources
• No sources of support supplied
External sources
• The Swedish Council on Technology Assessment in Health Care, Sweden.
INDEX TERMS
Medical Subject Headings (MeSH)

Anticonvulsants [∗ administration & dosage; blood]; Carbamazepine [administration & dosage; blood]; Drug Monitoring; Epilepsy
[blood; ∗ drug therapy]; Phenobarbital [administration & dosage; blood]; Phenytoin [administration & dosage; blood]; Primidone
[administration & dosage; blood]; Valproic Acid [administration & dosage; blood]
MeSH check words

Humans


Therapeutic Monitoring of Antiepileptic Drugs For Epilepsy (Review)

Uploaded by

Copyright:

Available Formats

Therapeutic Monitoring of Antiepileptic Drugs For Epilepsy (Review)

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

Therapeutic Monitoring of Antiepileptic Drugs For Epilepsy (Review)

Uploaded by

Copyright:

Available Formats

Therapeutic monitoring of antiepileptic drugs for epilepsy

Tomson T, Dahl ML, Kimland E

Therapeutic monitoring of antiepileptic drugs for epilepsy (Review)

Therapeutic monitoring of antiepileptic drugs for epilepsy (Review) i

Therapeutic monitoring of antiepileptic drugs for epilepsy

Torbjörn Tomson1 , Marja-Liisa Dahl2 , Elin Kimland3

Editorial group: Cochrane Epilepsy Group.

PLAIN LANGUAGE SUMMARY

Therapeutic monitoring of antiepileptic drugs for epilepsy

OBJECTIVES Types of outcome measures

Therapeutic monitoring of antiepileptic drugs for epilepsy (Review) 3

Data collection and analysis RESULTS

Therapeutic monitoring of antiepileptic drugs for epilepsy (Review) 4

Therapeutic monitoring of antiepileptic drugs for epilepsy (Review) 5

Therapeutic monitoring of antiepileptic drugs for epilepsy (Review) 6

References to studies excluded from this review

Therapeutic monitoring of antiepileptic drugs for epilepsy (Review) 7

Characteristics of included studies [ordered by study ID]

Methods Randomised, open, parallel study over a 24-month study period

Outcomes Proportion achieving seizure remission in the last 12 months

Notes Under powered

Item Authors’ judgement Description

Allocation concealment? Yes A - Adequate

Characteristics of excluded studies [ordered by study ID]

Study Reason for exclusion

Fröscher 1981 Patients treated with antiepileptic drugs in polytherapy

Therapeutic monitoring of antiepileptic drugs for epilepsy (Review) 8

Comparison 1. TDM versus control

Analysis 1.1. Comparison 1 TDM versus control, Outcome 1 12-month remission.

Review: Therapeutic monitoring of antiepileptic drugs for epilepsy

Comparison: 1 TDM versus control

Outcome: 1 12-month remission

Study or subgroup TDM Control Risk Ratio Weight Risk Ratio

Total (95% CI) 93 87 100.0 % 0.99 [ 0.78, 1.25 ]

0.1 0.2 0.5 1 2 5 10

Therapeutic monitoring of antiepileptic drugs for epilepsy (Review) 9

Review: Therapeutic monitoring of antiepileptic drugs for epilepsy

Comparison: 1 TDM versus control

Outcome: 2 Seizure free last 12-month follow up

Study or subgroup TDM Control Risk Ratio Weight Risk Ratio

Total (95% CI) 93 87 100.0 % 0.97 [ 0.75, 1.26 ]

0.1 0.2 0.5 1 2 5 10

Analysis 1.3. Comparison 1 TDM versus control, Outcome 3 Adverse effects.

Review: Therapeutic monitoring of antiepileptic drugs for epilepsy

Comparison: 1 TDM versus control

Outcome: 3 Adverse effects

Study or subgroup TDM Control Risk Ratio Weight Risk Ratio

Total (95% CI) 93 87 100.0 % 0.84 [ 0.64, 1.11 ]

0.1 0.2 0.5 1 2 5 10

Therapeutic monitoring of antiepileptic drugs for epilepsy (Review) 10

Review: Therapeutic monitoring of antiepileptic drugs for epilepsy

Comparison: 1 TDM versus control

Outcome: 4 Completed 2-year follow up

Study or subgroup TDM Control Risk Ratio Weight Risk Ratio

Total (95% CI) 93 87 100.0 % 0.94 [ 0.75, 1.16 ]

0.1 0.2 0.5 1 2 5 10

Appendix 1. Search methods for original version of this review

Therapeutic monitoring of antiepileptic drugs for epilepsy (Review) 12

Therapeutic monitoring of antiepileptic drugs for epilepsy (Review) 13