Asnani2019 PDF

Cochrane
Library
Cochrane Database of Systematic Reviews

Interventions for treating neuropathic pain in people with sickle cell
disease (Review)
Asnani MR, Francis DK, Brandow AM, Hammond Gabbadon CEO, Ali A
Asnani MR, Francis DK, Brandow AM, Hammond Gabbadon CEO, Ali A.

Interventions for treating neuropathic pain in people with sickle cell disease.
Cochrane Database of Systematic Reviews 2019, Issue 7. Art. No.: CD012943.
DOI: 10.1002/14651858.CD012943.pub2.
www.cochranelibrary.com

Interventions for treating neuropathic pain in people with sickle cell disease (Review)
Copyright © 2019 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
Cochrane Trusted evidence.
Informed decisions.

Library Better health. Cochrane Database of Systematic Reviews
TABLE OF CONTENTS
HEADER......................................................................................................................................................................................................... 1
ABSTRACT..................................................................................................................................................................................................... 1
PLAIN LANGUAGE SUMMARY....................................................................................................................................................................... 2
SUMMARY OF FINDINGS.............................................................................................................................................................................. 4
BACKGROUND.............................................................................................................................................................................................. 6
OBJECTIVES.................................................................................................................................................................................................. 7
METHODS..................................................................................................................................................................................................... 7
Figure 1.................................................................................................................................................................................................. 9
RESULTS........................................................................................................................................................................................................ 11
Figure 2.................................................................................................................................................................................................. 12
DISCUSSION.................................................................................................................................................................................................. 13
AUTHORS' CONCLUSIONS........................................................................................................................................................................... 14
ACKNOWLEDGEMENTS................................................................................................................................................................................ 14
REFERENCES................................................................................................................................................................................................ 15
CHARACTERISTICS OF STUDIES.................................................................................................................................................................. 19
DATA AND ANALYSES.................................................................................................................................................................................... 21
Analysis 1.1. Comparison 1 Pharmacological intervention (pregabalin) versus placebo, Outcome 1 Self-reported pain relief - S- 22
LANSS.....................................................................................................................................................................................................
Analysis 1.2. Comparison 1 Pharmacological intervention (pregabalin) versus placebo, Outcome 2 Self-reported pain relief - 22
NPSI........................................................................................................................................................................................................
Analysis 1.3. Comparison 1 Pharmacological intervention (pregabalin) versus placebo, Outcome 3 Quality of life (at 3 months).... 22
Analysis 1.4. Comparison 1 Pharmacological intervention (pregabalin) versus placebo, Outcome 4 Treatment-related adverse 23
events (at 3 months).............................................................................................................................................................................
ADDITIONAL TABLES.................................................................................................................................................................................... 23
APPENDICES................................................................................................................................................................................................. 24
CONTRIBUTIONS OF AUTHORS................................................................................................................................................................... 24
DECLARATIONS OF INTEREST..................................................................................................................................................................... 25
SOURCES OF SUPPORT............................................................................................................................................................................... 25
Interventions for treating neuropathic pain in people with sickle cell disease (Review) i
Informed decisions.

[Intervention Review]
Interventions for treating neuropathic pain in people with sickle cell

disease
Monika R Asnani1, Damian K Francis2, Amanda M Brandow3, Christine EO Hammond Gabbadon1, Amza Ali4
1Sickle Cell Unit, Caribbean Institute for Health Research, The University of the West Indies, Kingston 7, Jamaica. 2Epidemiology and
Research Unit, Tropical Medicine Research Institute, The Caribbean Branch of the United States Cochrane Centre, Mona, Jamaica. 3Section
of Hematology/Oncology, Department of Pediatrics, Medical College of Wisconsin, Milwaukee, USA. 4Faculty of Medical Sciences, University
of the West Indies, Kingston, Jamaica
Contact address: Monika R Asnani, Sickle Cell Unit, Caribbean Institute for Health Research, The University of the West Indies, 7 Ring Road,
Mona Campus, Kingston 7, Jamaica. monika.parshadasnani@uwimona.edu.jm.
Editorial group: Cochrane Cystic Fibrosis and Genetic Disorders Group

Publication status and date: New, published in Issue 7, 2019.
Citation: Asnani MR, Francis DK, Brandow AM, Hammond Gabbadon CEO, Ali A. Interventions for treating neuropathic
pain in people with sickle cell disease. Cochrane Database of Systematic Reviews 2019, Issue 7. Art. No.: CD012943. DOI:
10.1002/14651858.CD012943.pub2.
ABSTRACT
Background
Pain is the hallmark of sickle cell disease (SCD) and it can be severe, frequent and unpredictable. Although nociceptive pain is more com-
mon, at times, people with SCD may have neuropathic pain. The latter can occur due to peripheral or central nerve injury. This review is
focused on identifying treatment of only painful sensory neuropathy in people with SCD.
Objectives
To determine the effectiveness and safety of any pharmacological or non-pharmacological therapies for treating neuropathic pain in peo-
ple with SCD.
Search methods
We searched the Cochrane Cystic Fibrosis and Genetic Disorders Group's Haemoglobinopathies Trials Register, compiled from electronic
database searches and handsearching of journals and conference abstract books. We also searched trial registries, the reference lists of
relevant articles and reviews and contacted experts in the field.
Date of last search: 31 January 2019.
Selection criteria
Randomised controlled trials (RCTs) (parallel or cross-over in design), quasi-RCTs of pharmacological or non-pharmacological therapies
for treating neuropathic pain in people with SCD compared to placebo or another intervention in any category (i.e. pharmacological or
non-pharmacological).
Data collection and analysis

Two review authors independently assessed all trials identified by the searches and extracted relevant data. Two authors independently
assessed the risk of bias in the selected trials using the Cochrane risk of bias tool. Two review authors independently rated the quality of
the evidence for each outcome using the GRADE guidelines.
Interventions for treating neuropathic pain in people with sickle cell disease (Review) 1
Informed decisions.

Main results
One RCT of 22 participants with SCD, conducted in the USA was included in this review. Participants were randomly assigned to either
pregabalin (n = 11) or placebo (n = 11). Oral pregabalin was administered at an initial dose of 75 mg twice daily. The drug was titrated at
increments of 75 mg to a maximum of 600 mg daily or decreased by 75 mg per day if necessary, based on clinical presentation and pain
level. Neuropathic pain was assessed using self-reports on the Leeds Assessment of Neuropathic Symptoms and Signs (S-LANNS) scale and
the Neuropathic Pain Symptom Inventory (NPSI), where higher scores were indicative of more pain. Outcomes included self-reported pain,
quality of life and withdrawal due to adverse effects measured at baseline and monthly for three months post-intervention. The overall
risk of bias was low with a high risk of bias due to attrition.
In relation to this reviews primary outcomes, for self-reported neuropathic pain relief, given the paucity of data, we are very uncertain
whether there is a difference between the pregabalin and placebo groups at the end of three months as measured by the S-LANSS scale,
mean difference (MD) -2.00 (95% confidence interval (CI) -9.18 to 5.18), or the NPSI scale, MD -11.10 (95% CI -33.97 to 11.77) (very low-
quality evidence). There was no report of 'Patient Global Impression of Change' in the included trial.
Although the mean quality of life scores (Short Form-36) at three months showed small increases in seven of the eight domains post-
intervention in the pregabalin group as compared to the placebo group, this was very low-quality evidence and we are very uncertain
whether pregabalin increases quality of life. Neither of our pre-defined outcomes of 'time to improvement of symptoms' or 'changes in
sleep quality', were measured in the included trial.
While treatment-related adverse effects appeared higher in pregabalin group than the placebo group at three months, this was very low-
quality evidence and we are very uncertain whether there is a difference, RR 1.33 (95% CI 0.39 to 4.62) (very low-quality evidence). There
was one withdrawal for adverse effects in the pregabalin group while three people withdrew or dropped out from the placebo group due
to adverse effects and complications and hospitalisation related to SCD.
Authors' conclusions
The included trial provided very low-quality evidence. Self-reported pain relief was greater in the pregabalin group compared to the place-
bo control group but only using the S-LANSS scale and we are very unsure whether there is a difference. While the pregabalin group tended
to have improved quality of life over the duration of the trial, this was very low-quality evidence and we are uncertain whether there is
a difference. Adverse effects and withdrawals were similar across the treatment and placebo control group in trial. There are both insuf-
ficient trials addressing this review question and insufficient outcomes addressed in the single included RCT. Therefore, there is still a
significant gap in evidence on interventions for neuropathic pain in people with SCD.
PLAIN LANGUAGE SUMMARY
Treatments for neuropathic pain in persons with sickle cell disease
Review question
We reviewed the evidence about the effect and safety of drug, or other therapies, for treating pain due to nerve damage (neuropathic pain)
in people with sickle cell disease (SCD).
Background
Pain is the most common complication of SCD. Some of this pain may be due to damage to nerves and this pain needs to be treated
differently from the usual sickle cell pain. The usual medications such as ibuprofen or morphine may not be effective in managing this nerve
pain. This trial aimed to examine how safe and effective medications or other alternative treatments are for nerve pain in people with SCD.
Search date
The evidence is current to: 31 January 2019.
Study characteristics
After a detailed search of scientific literature, we identified one trial that was eligible for inclusion. In this trial people with SCD who were
diagnosed to have neuropathic pain were randomly put into groups to take either a drug named pregabalin or a placebo (no active med-
ication) treatment.
Key results
The trial was conducted in the USA and included 22 participants with SCD, with 11 people in the pregabalin group and 11 in the placebo
drug group. Assessments were measured at baseline and monthly for three months.
Self-reported neuropathic pain relief and quality of life scores (Short Form-36) were no different between the pregabalin and placebo
groups. The outcomes of time to improvement of symptoms and changes in sleep quality were not measured in the included trial. Few
unwanted effects were noted and the numbers of these were not different between participants who were given pregabalin versus those
given placebo.
Informed decisions.

In conclusion, the effect of pregabalin on SCD neuropathic pain was no different to placebo. We are unable to make firm conclusions
regarding our objectives on the basis of a single small trial which only addressed three of our seven prespecified outcomes. Larger, well-
conducted trials of different treatments for neuropathic pain in people with SCD need to be carried out.
Quality of the evidence

The evidence in this review was assessed as very-low quality. The single trial in this review lost more than 15% of participants to follow-up
over three months. We downgraded the level of evidence due to lack of clarity in how the participants were assigned to the pregabalin or
the placebo group and due to the small number of events and participants in the trial.
SUMMARY OF FINDINGS

Summary of findings for the main comparison. Summary of findings - Pharmacological intervention (pregabalin) versus placebo for treating
Library
Cochrane
neuropathic pain in people with sickle cell disease
Pregabalin compared to placebo for treating neuropathic pain in people with sickle cell disease
Patient or population: adults and children with sickle cell disease experiencing neuropathic pain
Better health.
Informed decisions.
Trusted evidence.
Settings: outpatient sickle cell disease clinic
Intervention: pregabalin
Comparison: placebo
Outcomes Illustrative comparative risks* Relative effect No of Partici- Quality of the evi- Comments
(95% CI) (95% CI) pants dence
(studies) (GRADE)
Assumed risk Corresponding
risk
Placebo Pregabalin
Self-reported pain relief - NPSI: The mean The mean NPSI 16 ⊕⊕⊕⊝ No significant difference
ranges from 0 to 100. Higher scores NPSI Score score in the pre- between pregabalin and
are indicative of neuropathic pain. in the place- gabalin group was (1 study) very low1,2,3 placebo groups in self-re-
bo group was 11.10 points lower ported pain events.
Follow-up: 3 months 32.6 points (95% CI -33.97 low-
er to 11.77 higher)
Self-reported pain relief - S- The mean S- The mean S-LANSS 16 ⊕⊕⊕⊝ No significant difference
LANSS: Possible total scores range LANSS score score in the pre- between pregabalin and
from 0 to 24, and a score of 12 or in the placebo gabalin group was (1 study) very low1,2,3 placebo groups in self-re-

higher is indicative of neuropathic group was 7.9 2.00 points lower ported pain events.
pain. points (95% CI -9.18 lower
to 5.18 higher)
Follow-up: 3 months
PGIC See comment See comment N/A N/A N/A This outcome was not mea-
sured.
Follow-up: 3 months
Time to improvement of symp- See comment See comment N/A N/A N/A This outcome was not mea-
toms sured.
4

Follow-up: 3 months
Change in QoL See comment See comment NA 16 ⊕⊕⊕⊝ There was no significant dif-
Library
Cochrane
very low1,2,3 ference in any of the 8 QoL
Follow-up: 3 months (1 study) subscales between the pre-
gabalin and placebo groups.
Changes in sleep quality as mea- See comment See comment N/A N/A N/A This outcome was not mea-
sured by sleep questionnaires sured.
Better health.
Informed decisions.
Trusted evidence.
Follow-up: 3 months
Treatment-related adverse ef- 273 per 1000 363 per 1000 RR 1.33 (95% CI 22 ⊕⊕⊕⊝
fects 0.39 lower to to
Follow-up: 3 months (106 to 1261) 4.62 higher) (1 study) very low1,2,3
The basis for the assumed risk is the event rate in the control group unless otherwise stated in the comments and footnotes. The corresponding risk (and its 95% CI) is
based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).
CI: confidence interval; NA: not applicable; NPSI: Neuropathic Pain Symptoms Inventory; PGIC: Patient Global Impression of Change; QoL: quality of life; S-LANSS: Leeds
Assessment of Neuropathic Symptoms
and Signs
GRADE Working Group grades of evidence

High quality: further research is very unlikely to change our confidence in the estimate of effect.
Moderate quality: further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Low quality: further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Very low quality: we are very uncertain about the estimate.
1. Downgraded once due to key risk of bias domain random sequence generation which might influence primary outcomes.
2. Downgraded once due to imprecision: small number of events/participants and large CI around the relative effect.
3. Downgraded once due to indirectness: majority female and all over 18.


5

Informed decisions.

BACKGROUND crawling sensations or pins and needles, which itself may or may
not be unpleasant (dysaesthetic). When pain is present, it may be
Description of the condition burning or lancinating in quality, typically implying either differen-
tial small-fibre involvement (burning) or larger-fibre involvement
Sickle cell disease (SCD) refers to a group of genetic disorders of
(lancinating pain). In most situations, it is indeed neuropathic pain
haemoglobin which occur when the sickle gene is inherited either
that is the most common and most disabling feature of peripher-
homozygously or heterozygously with another abnormal haemo-
al neuropathic involvement. The International Association for the
globin gene. This monogenic disorder results when the 17th nu-
Study of Pain defines neuropathic pain as "pain caused by a lesion
cleotide in the β-globin chain changes from thymine to adenine re-
or disease of the somatosensory system" (Jensen 2011). They fur-
sulting in the 6th amino acid becoming valine instead of glutamic
ther define that a lesion or disease of the central somatosensory
acid (Rees 2010). In the deoxygenated state, this mutation leads to
system (e.g. thalamic origin) may lead to 'central neuropathic pain'
polymerisation of sickle haemoglobin and subsequent red cell sick-
and of the peripheral nervous system to 'peripheral neuropathic
ling. The disorder is multisystemic and presents with many acute
pain'. The former is comparatively rare and typically both forms
complications, as well as progressive organ damage. Overall sur-
result from the same pathophysiological processes. Neuropathic
vival is reduced (Wierenga 2001), although this continues to im-
pain can be persistent and long-lasting as once nerve injury has oc-
prove. Despite optimising available treatment, life expectancy is
curred it may be irreversible.
still almost 20 years shorter than in the general population (Ser-
jeant 2013). In SCD neuropathic pain can occur due to peripheral or central
nerve injury, and it has been hypothesised to occur from vaso-oc-
Although the disease originated in sub-Saharan Africa and the In-
clusion of vasa nervorum of nerves (Ballas 2012). Emerging neu-
dian subcontinent, the disease is now global (Stuart 2004) and has
robiological data suggest that SCD pain has a complex aetiology
been declared a public health burden by the World Health Organi-
that includes neuropathic components (Brandow 2015a). This is ev-
zation (WHO) (WHO 2009). Estimates show that approximately 5%
idenced by the pain descriptors that individuals use (i.e. burning,
of the world’s population carries a haemoglobinopathy gene (Piel
lancinating, sticking, numbness, tingling) and accumulating neuro-
2013a) and the prevalence of the disease at birth is expected to rise
biological evidence in both the sickle cell mouse model and peo-
as a result of improving survival and increasing migration to high-
ple with SCD, that SCD pain has features of peripheral and central
er-income countries (Piel 2013b).
nervous system sensitisation resulting in pain hypersensitivity and
Pain is the hallmark of SCD and it can be severe, frequent and un- neuropathic pain (Ballas 2007; Ballas 2012; Brandow 2017). Fur-
predictable (Ballas 2012). The acute painful crisis is the most fre- thermore, in SCD symptoms of neuropathy may be present without
quent reason for healthcare visits by both children and adults with any associated 'neuropathic pain' and assessments should be able
SCD (Ballas 2005). Acute painful crises increase in frequency as peo- to differentiate between the two (Ballas 2013). Screening question-
ple get older and more frequent crises have been associated with naires reveal that 30% to 40% of people with SCD report neuropath-
an increased mortality rate. The vaso-occlusion seen in SCD leads ic pain and the presence of neuropathic pain is associated with old-
to a cycle of hypoxia, ischaemia and tissue injury which causes the er age and is more common in females (Brandow 2014; Brandow
activation of nociceptive pathways, stimulating peripheral affer- 2015b).
ent pathways and resulting in a perception of pain (Ballas 2012).
It is important to make the distinction between nociceptive and
Recurrent and irreversible tissue damage can lead to bone infarc-
neuropathic pain as their pathophysiological mechanisms differ
tion, avascular necrosis (hip or shoulder) and arthritis, resulting in
and hence therapies need to be tailored appropriately. This review
chronic pain; other complications such as leg ulceration and chron-
is focused on identifying treatment of only painful sensory neu-
ic osteomyelitis may also result in chronic pain (Ballas 2011a). How-
ropathy in people with SCD.
ever, most people with SCD who suffer from chronic pain do not
have an obvious anatomic source. As the understanding of the di- Description of the intervention
verse pathophysiological mechanisms behind underlying SCD pain
evolves, it is becoming apparent that chronic pain is more frequent Neuropathic pain from any cause, including non-SCD causes, can
than previously thought (Smith 2010); and may occur in up to 29% be quite difficult to treat and is not likely to be amenable to any one
to 40% of adolescents and adults with SCD, which profoundly af- intervention (Finnerup 2015; Kalso 2013).
fects health-related quality of life (QoL). Both the acute and chron-
ic pain suffered by people with SCD are associated with significant Opioids are the backbone of treatment for both acute and chronic
psychological complications such as depression and anxiety and SCD pain (Ballas 2007); however, their effectiveness is not well stud-
reduced overall health-related QoL (McClish 2005). ied and opioids can be associated with numerous adverse events,
including opioid-induced hypersensitivity. Furthermore, the effec-
Although nociceptive pain is more common, at times, people with tiveness of these drugs for treating neuropathic pain in SCD is un-
SCD may have neuropathic pain. In general, neuropathy is the pres- clear. The National Institute for Health and Care Excellence (NICE)
ence of nerve damage which may either be symptomatic or asymp- guidelines recommend amitriptyline, duloxetine, gabapentin or
tomatic, i.e. neuropathy may be present without the presence of pregabalin as initial treatments for neuropathic pain (National
pain and may only be detected by careful clinical examination or 2013).
neurophysiological testing. When symptomatic, however, symp-
toms may be sensory or motor or both. Motor symptoms may be Hydroxyurea is a ribonucleotide reductase inhibitor and is one of
negative, e.g. weakness or atrophy, or positive, e.g. twitching (fas- the two drugs approved for treating SCD. A Cochrane Review has
ciculations). Sensory involvement is not necessarily painful; indi- shown benefits for prevention of pain and other severe complica-
viduals may report negative sensory symptoms, e.g. numbness, but tions of SCD (Nevitt 2017) as has further research in both adults
may also report positive sensory symptoms, e.g. tingling, itching,
Informed decisions.

(Charache 1997) and children (Heeney 2010; McGann 2011; Ware Non-pharmacologic interventions include:
1999).
• psychological treatments such as cognitive or behavioural ther-
In July 2017, L-Glutamine was approved for preventing acute apies which attempt to assess and modify the thoughts and be-
complications in people with SCD. Along with pharmacologi- haviours associated with pain (Morley 1999);
cal agents (such as opioids and non-steroidal anti-inflammatory • physiotherapy and exercise, which may improve mood and in-
agents (NSAIDs)) non-pharmacological interventions, such as cog- crease pain tolerance (Anshel 1994);
nitive behavioural therapy, herbal treatments, relaxation therapy, • TENS (transcutaneous electrical nerve stimulation); which stim-
and acupuncture have also shown benefit in managing acute SCD ulates nerve endings and utilises central and peripheral mecha-
pain (Ballas 2007; Ballas 2011b; Lutz 2015). nisms to decrease hyperalgesia (Vance 2014).
In this review, we will assess the effectiveness and safety of
Why it is important to do this review
all pharmacological and non-pharmacological interventions that
have been used to treat neuropathic pain in SCD. Pain in SCD can be very difficult to manage and often does not re-
spond to either opioids or other commonly-used classes of anal-
How the intervention might work gesics. As mechanisms that underlie neuropathic pain differ from
other kinds of pain present in SCD (i.e. Ischaemic, inflammatory re-
Different classes of pharmacological agents intended for use in
sulting in nociceptive pain), therapy that is directed specifically at
managing pain have different mechanisms of action and so it is im-
neuropathic pain may be required for a subset of patients. Neuro-
portant to assess their effectiveness in managing neuropathic pain
pathic pain has been shown to be difficult to treat in conditions oth-
in SCD – both individually and as part of combination therapy.
er than SCD. Thus, it is important to systematically review the exist-
The consensus statement of the Canadian Pain Society provides ing evidence to assess the effectiveness of any therapies, whether
one useful approach to summarising the pharmacological and non- pharmacological or non-pharmacological, for treating neuropathic
pharmacological management of neuropathic pain by mode of ac- pain in people with SCD.
tion (Moulin 2007).
OBJECTIVES
Pharmacological interventions include, but are not limited to:
To determine the effectiveness and safety of any pharmacological
• tricyclic antidepressants, which block the hyperalgesic effect of or non-pharmacological therapies for treating neuropathic pain in
n-methyl-d-aspartate (NMDA) agonists; people with SCD.
• anticonvulsants such as gabapentin and pregabalin, which bind
to calcium channels in the dorsal horn resulting in inhibition of METHODS
excitatory neurotransmitters;
Criteria for considering studies for this review
• venlafaxine and duloxetine, which act as serotonin noradrena-
line re-uptake inhibitors (SNRIs); Types of studies
• topical lidocaine, which is a sodium channel blocker and has Randomised, controlled trials (RCTs) and quasi-RCTs. Cross-over
minimal adverse effects; trials were considered.
• opioids with the analgesic effect of binding to mu- and kap-
pa-opioid receptors and blocking the release of substance P Types of participants
neurotransmitters; People with any type of SCD (all ages, both genders, all genotypes)
• cannabinoids with anti-inflammatory and central analgesic who described experiencing pain symptoms associated with neu-
properties; ropathic pain (that is, pain described as burning or lancinating),
• invasive techniques, which may be used after failure of standard or where trial investigators described participants as having 'neu-
treatments and include intravenous lidocaine and opioid spinal ropathic pain'. We excluded trials with individuals who may have
infusions via implantable pumps as well as spinal cord stimula- had other co-morbidities that could potentially lead to neuropathic
tion; pain, e.g. diabetes mellitus and cancer-related neuropathy.
• NSAIDS, which act by inhibiting the production of
As described above, the diagnosis of neuropathic pain could
prostaglandins which cause inflammation at sites of tissue dam-
be based on clinician opinion, based on pain classification as-
age.
sessments, such as the Leeds Assessment of Neuropathic Symp-
As neuropathic pain is expected to be of long-term nature and be- toms and Signs (LANSS) scale or the PainDETECT questionnaire
cause pharmacological agents may have various adverse events, (Brandow 2014), or use of classical neuropathic pain descriptors
many individuals seek non-pharmacological treatments. Studies such as lancinating, tingling or burning. The named scales are neu-
in SCD have shown some limited benefits of non-pharmacological ropathic pain screening measures that have not been specifically
treatments (Lutz 2015). Some of these treatments are also asso- validated in people with SCD.
ciated with the lowering of anxiety and depression and improve-
Types of interventions
ments in coping skills (Edwards 2010; Hildenbrand 2014). While
these studies have looked at the effects in SCD pain in general, it We made provisions to include trials where any pharmacological
remains to be seen if there is benefit of their use for treating neuro- or non-pharmacological treatment is compared to either placebo
pathic pain in people with SCD. or to another intervention in any category (i.e. pharmacological
or non-pharmacological). Treatment duration should have been at
Informed decisions.

least two weeks. All doses, routes and trial settings were to be in- searching the abstract books of five major conferences: the Euro-
cluded. pean Haematology Association conference; the American Society
of Hematology conference; the British Society for Haematology An-
Types of outcome measures nual Scientific Meeting; the Caribbean Public Health Agency Annual
Primary outcomes Scientific Meeting (formerly the Caribbean Health Research Coun-
cil Meeting); and the National Sickle Cell Disease Program Annual
1. Self-reported pain relief of: Meeting. For full details of all searching activities for the register,
a. 30% or greater (moderate); please see the relevant section of the Cochrane Cystic Fibrosis and
b. 50% or greater (substantial). Genetic Disorders Group's website.
2. Patient Global Impression of Change (PGIC):
Date of last search: 31 January 2019.
a. much or very much improved (moderate);
b. very much improved (substantial). We also searched the following databases, trials registries and re-
sources:
We have used the definitions from the the Initiative on Methods,
Measurement, and Pain Assessment in Clinical Trials (IMMPACT) for • Caribbean Public Health Agency (CARPHA) EvIDeNCe Portal
moderate and substantial benefit in chronic pain studies (Dworkin (http://carphaevidenceportal.bvsalud.org/);
2008). These are recommendations for assessing pain outcomes • US National Institutes of Health Ongoing Trials Register Clinical-
since pain responses are not usually normally distributed. trials.gov (www.clinicaltrials.gov);
Secondary outcomes • World Health Organization International Clinical Trials Registry
Platform (WHO ICTRP) (http://apps.who.int/trialsearch/);
1. Time to improvement of symptoms
• EU Clinical Trials Register (www.clinicaltrialsregister.eu/).
2. Change in psychosocial outcomes (e.g. QoL, activities of daily
living, depression or anxiety (or both), or coping (all assessed us- Please refer to Appendix 1 for full search strategies. All additional
ing any well-validated measures or expert diagnosis)) searches were conducted in January 2019.
3. Changes in sleep quality as measured by sleep questionnaires
Searching other resources
4. Treatment-related adverse effects
a. number of people experiencing serious adverse effects (any We screened the references of included trials and any relevant sys-
significant event, e.g. leading to hospitalisation, death, or sig- tematic reviews identified for further references to relevant trials.
nificant disability) We also contacted noted experts in the field conducting relevant
b. number of people experiencing other adverse effects trials.
c. number of withdrawals due to any adverse events (including
Data collection and analysis
serious adverse events)
5. Withdrawal from treatment due to lack of efficacy Selection of studies
We compiled the database search results using Endnote and dis-
Search methods for identification of studies
carded duplicates of same record. We then exported data from the
We conducted a comprehensive search for all relevant published search database to Microsoft Excel to facilitate screening.
and unpublished trials without restrictions on language, year or
publication status. Two review authors (CG and AB) independently assessed each tri-
al identified by the search for eligibility by reading its title and ab-
Electronic searches stract. A third author (MA) provided consensus as needed. We dis-
carded trials that clearly did not meet inclusion criteria, and ob-
The review authors identified relevant trials from the Cystic Fibrosis tained full-text copies of the remaining trials. We attempted to link
and Genetic Disorders Group's Haemoglobinopathies Trials Regis- multiple reports of the same trials. We did not anonymise trials
ter using the terms: (sickle cell OR (haemoglobinopathies AND gen- before assessment. Two authors (MA, CG) independently read full
eral)) AND pain. reports and selected relevant trials meeting eligibility criteria. We
The Haemoglobinopathies Trials Register is compiled from elec- presented a PRISMA flow chart in the full review (Figure 1) (Moher
tronic searches of the Cochrane Central Register of Controlled Tri- 2009), as well as a list of excluded trials for those trials not excluded
als (CENTRAL) (updated each new issue of the Cochrane Library) at the title and abstract screening phase that a reader may plausi-
and weekly searches of MEDLINE. Unpublished work is identified by bly expect to see in the included studies section of the review (Hig-
gins 2011a).

Informed decisions.

Figure 1. Flow Diagram

Data extraction and management 2. performance bias – blinding of participants and personnel;
Two review authors (MA, CG) independently extracted data using 3. detection bias – blinding of outcome assessment;
the data extraction template from the Cochrane Cystic Fibrosis and 4. attrition bias - incomplete outcome data;
Genetic Disorders Group; we resolved disagreements by discussion 5. reporting bias; and
with a third author (DF). 6. other bias.
We extracted data on age, sex, genotype of participants and self- Two review authors (MA, DF) collated judgements and rationale for
reported change in pain as assessed by multiple pain scales. We al- each bias domain in a tabular form. We categorised risk of bias
so collected information about the number of participants in each judgements according to 'low', 'high', or 'unclear' risk according to
trial, drug and dosing regimens, trial design (placebo or active con- the Cochrane criteria (Higgins 2011b).
trol), trial duration and follow-up, withdrawals and adverse events.
We entered the extracted data directly into the Review Manager Measures of treatment effect
software (RevMan 2014).
We performed statistical analyses using Review Manager 5
We searched for multiple reports of the same trial to include as a (RevMan) (RevMan 2014).
single unit of interest in the review. We collected characteristics of
Continuous data
the included trials and populated a table of 'Characteristics of in-
cluded studies' in the full review. We reported on the primary out- For primary and secondary outcomes (e.g. self-reported pain score,
come of change in pain (pain reduction) at time periods greater QoL) that were reported on a continuous scale, we calculated the
than two weeks as available in the included trial. estimate of treatment effect as the mean difference (MD) with 95%
CIs. We used means and standard deviations (SD). For future up-
Assessment of risk of bias in included studies dates, if additional trials are identified where continuous scores
Two authors used the Cochrane risk of bias tool (MA and DF) to as- measuring the same outcome but in a variety of ways (e.g. differ-
sess risk of bias in six domains (Higgins 2011b): ent scales to measure anxiety) are pooled, we will use the standard-
ised mean difference (SMD) with 95% CIs (Deeks 2011). We assessed
1. selection bias – sequence generation and allocation conceal- whether data were skewed as per chapter 9 of the Cochrane Hand-
ment; book for Systematic Reviews of Interventions (Deeks 2011).
Informed decisions.

Dichotomous data are not missing at random (e.g. withdrawal due to adverse events or
serious adverse events) we will use the imputation of the mean ap-
For dichotomous data (e.g. treatment-related adverse events), we
proach to replace missing values. We do not expect data missing at
calculated risk ratios (RR) with 95% confidence intervals (CI) using
random, such as in the instance of loss of record with pain assess-
number of events in the intervention group and the control group.
ment, to be related to the actual pain outcome. In these instances
Time-to-event data we will analyse available data without any adjustment for missing
values (Higgins 2008).
Should future update of this review include trials with time-to-
event data (e.g. time-to-improvement of symptoms), we will use Assessment of heterogeneity
survival analysis to calculate hazard ratios of the intervention effect
We did not assess heterogeneity as we only included a single tri-
on such outcomes as described in the Cochrane Handbook for Sys-
al in the analysis of the review. We intend to assess heterogeneity
tematic Reviews of Interventions (Deeks 2011).
in future updates of this review using the criteria below, as recom-
Unit of analysis issues mended in the Cochrane Handbook for Systematic Reviews of Inter-
ventions (Deeks 2011).
For future updates of this review, if we include additional trials,
making a meta-analysis possible, we will give consideration to the We will visually inspect heterogeneity using the combined data pre-
following unit of analysis issues. sented in the forest plots, and by considering the I2 statistic (Hig-
gins 2003). The Chi2 values and a P value less than 0.10 will con-
Unit of analysis for included trials will be at the participant level stitute statistical heterogeneity as recommended in the Cochrane
where there is no evidence of clustering by pain outcome or trial Handbook for Systematic Reviews of Interventions (Deeks 2011). The
design in the case of cluster RCTs. I2 statistic reflects the likelihood that variation of results across tri-
als are due to heterogeneity rather than by chance, and we plan to
Clustering may be introduced by included trials that report mul- interpret this using the following simple classification:
tiple pain events per participant. Where this clustering effect is
present, we will use the number of trial clusters and an estimate of • 0% to 40%: might not be important;
the intraclass correlation co-efficient to inflate the standard errors • 30% to 60%: may represent moderate heterogeneity;
(SEs) associated with each clustered trial (Cooper 2016).
• 50% to 90%: may represent substantial heterogeneity;
Cluster-randomised trials • 75% to 100%: considerable heterogeneity.
For any trials included in the future, if clustered randomisation has In addition to the I2 value, we will observe (a) the magnitude and di-
been used, evidence that trial investigators have presented their re- rection of effects; and (b) the strength of evidence for heterogene-
sults after appropriately controlling for clustering effects (e.g. using ity (e.g. P value from the Chi2 test, or a CI for I2).
techniques such as robust SEs or multilevel modelling) will be ex-
amined. If it is unclear whether a cluster-randomised trial has used Assessment of reporting biases
appropriate statistical techniques for clustering, we will contact tri-
We addressed selective outcome reporting by comparing out-
al investigators for further information. Where appropriate statisti-
comes described in the methods of the published report with those
cal techniques are not used, we will request individual participant
reported in the results.
data and re-analyse these using techniques to control for cluster-
ing. Following this, we will undertake a meta-analysis of effect sizes We did not consider a funnel plot for assessing publication bias as
and SEs in RevMan using the generic inverse method (Deeks 2011). the requisite number of included trials (at least 10) was not met
(Sterne 2011).
Studies with multiple treatment groups
We aimed to summarise separately pharmacological and non- Data synthesis
pharmacological treatment groups. Within the pharmacological A meta-analysis was not possible for this review as only one trial
treatment group, we planned to combine the different medications met the pre-planned inclusion criteria. We entered relevant data on
if the mechanisms of action were similar. reported outcomes into RevMan to calculate effect estimates sup-
ported by a description of trial findings. For future updates of the
Cross-over trials
review, we plan to add to the analysis using a fixed-effect model
If we had included trials, we would have analysed data from the first for meta-analysis where there is no evidence of substantial hetero-
period as carry over of the treatment effect would have been likely geneity between trials. In the event of substantial heterogeneity be-
to affect the comparative efficacy of the intervention. If needed, we tween trials, we will use a random-effects model.
will contact trial authors for first-period data.
Two review authors (DF and MA) independently rated the quality
Dealing with missing data of the evidence for each outcome using the guidelines provided in
chapter 12 of the Cochrane Handbook for Systematic Reviews of In-
Data reported by the single included RCT were complete. We de-
terventions (Schünemann 2011).
scribe attrition in the Characteristics of included studies table.
Subgroup analysis and investigation of heterogeneity
In future versions of the review, we will request any additional data
that may be required from the trial authors if data presented in pub- We did not conduct any subgroup analysis but consider the follow-
lished paper are insufficient. Where possible, we will undertake an ing subgroups to be important for future updates of this review:
intention-to-treat (ITT) analysis. Where we assume missing values
Informed decisions.

1. age (children under 18 years versus adults 18 years and older); Design and sample size and setting
2. drug class (for pharmacological interventions). The single included trial was a RCT with two arms having 22 partic-
ipants randomly assigned to either pregabalin (n = 11) or placebo
Sensitivity analysis
(n = 11) (Schlaeger 2017). The trial was conducted in the USA with
In this current version of the review, we were unable to conduct a participants recruited from an outpatient SCD clinic.
meta-analysis or sensitivity analysis with only one included trial.
Future update of this review will consider sensitivity analysis on the Participants
basis of: There were 22 participants included; the majority of participants
were female (n = 16) and of African descent with mean age of 33.1
1. imputation of missing values using the mean;
years ranging from 18 to 82 years.
2. fixed-effect versus random-effects models where there are likely
unimportant heterogeneity (0% to 40%) in both models; Disease status for all participants was as follows: homozygous S
3. trial design, where the effect of blinded versus unblinded trials (SS) genotype= 15; heterozygous SC genotype = 6; sickle-beta tha-
on outcomes are assessed. lassaemia genotype (SBThal) = 1. In the placebo group there were
six participants with SS and five with SC; in the pregabalin group
Summary of findings table there were nine participants with SS, one with SC and one with
We generated a summary of findings tabled with the following out- SBThal.
comes.
Intervention
1. Self-reported pain relief The intervention was pregabalin given orally with an initial dose of
2. PGIC 75 mg twice daily. The pregabalin was titrated at increments of 75
3. Time to improvement of symptoms mg to a maximum of 600 mg daily or decreased by 75 mg per day if
4. Change in QoL necessary based on clinical presentation and pain level.
5. Changes in sleep quality as measured by sleep questionnaires Outcomes
The single included trial reported on one of this reviews two prima-
We planned to present a summary of findings table for each com- ry outcomes, self-reported pain. Of the five secondary outcomes,
parison made (e.g. pharmacological and non-pharmacological in- only QoL and withdrawal due to adverse effects were measured and
terventions in future reviews) but are currently only able to present reported by the trial.
a table on a pharmacological intervention.
The trial did not report on PGIC in pain, time to improvement of
The summary of findings table includes a comment and brief de- pain, changes in sleep quality and withdrawal due to lack of effi-
scription of the GRADE approach as used to assess the quality of cacy. The outcomes were assessed at baseline and monthly up to
the body of evidence (Schünemann 2011). We took guidance in de- three months.
veloping the summary of findings table from the GRADE Handbook
Excluded studies
(GRADE 2013).
See Characteristics of excluded studies.
RESULTS
A total of 13 trials were excluded as the population was not being
Description of studies defined as having neuropathic pain or neuropathic pain descriptors
such as burning or lancinating pain (Co 1979; Ezenwa 2016; Ezenwa
See 'Characteristics of included studies' and 'Characteristics of ex-
2018; Gil 1996; Hollins 2012; Horst 2016; Howard 2018; Matthie 2018;
cluded studies'.
Myers 1999; Osunkwo 2012; Wallen 2014; Wang 1988).
Results of the search
Risk of bias in included studies
The search of the Group's trials register identified 182 references of
which 31 (representing 14 trials) were selected for full text assess- A graphical display of the risk of bias assessment is presented in an
ment after initial screening. Only one full text was assessed as meet- a figure (Figure 2). The single trial had issues with incomplete out-
ing the criteria to be included in this review (Figure 1). come data (more than 20% attrition rate) and 'unclear' description
of random sequence generation which affected the overall risk of
Included studies bias. The other domains were assessed as being at low risk of bias.
See Characteristics of included studies.

Informed decisions.

Figure 2. Risk of bias summary: review authors' judgements about each risk of bias item for each included study.

Allocation Overall, the trial was assessed as being at low risk of bias for blind-
ing of participants, personnel and outcome assessors.
Sequence generation
The single included RCT was assessed as being at unclear risk of Incomplete outcome data
bias for allocation bias. Insufficient information was provided on There was 27% loss to follow-up, however, dropouts were similar
how the allocation was carried out. in both groups and appropriately described. The trial authors did
not report an intention-to-treat analysis and therefore, we assessed
Allocation concealment
this trial as being at high risk of bias for incomplete outcome data.
There was a low risk of bias for this domain as only the external
pharmacist was aware of the allocation code. Selective reporting
We did not have access to the trial protocol and were therefore un-
Blinding
able to assess reporting bias. The risk of bias for this domain is un-
The included trial was described as a "double-blind pilot study". clear.
The individuals responsible for participant care were blinded as
the allocation was centralised and the initial dose of the medica- Other potential sources of bias
tion was similar for both groups. No information was given on the The trial was not perceived as having any other potential source of
physical appearance of the medication given to the pregabalin and bias.
placebo control groups. The participants also had no knowledge of
which group they were assigned to. Outcome assessors were also Effects of interventions
unaware of which group received the intervention versus the place-
bo as this was only known by the pharmacist responsible for se- See: Summary of findings for the main comparison Summary of
quence generation. findings - Pharmacological intervention (pregabalin) versus place-
bo for treating neuropathic pain in people with sickle cell disease
Informed decisions.

The Schlaeger trial (22 participants) assessed the management of b. Activities of daily living, depression or anxiety (or both), or coping
sickle cell pain using pregabalin versus placebo over three months There was no assessment or report on any other psychosocial out-
(Schlaeger 2017). Self-reported pain relief was assessed using the come in the included trial (Schlaeger 2017).
S-LANNS and the NPSI which are both validated tools for examining
neuropathic pain. 3. Changes in sleep quality as measured by sleep questionnaire
Pregabalin versus placebo This outcome was not assessed in the included trial (Schlaeger
2017).
Primary outcomes
1. Pain relief
a. Number of people experiencing serious adverse effects
The included trial reported on pain relief at baseline, one, two and
three months (Analysis 1.1; Analysis 1.2). Notably, the single includ- There were two reported cases of hospitalisations due to sickle cell
ed study (22 participants) by Schlaeger did not report on the pain pain or complications of the disease with these participants being
outcome as defined in the current review protocol. They instead re- subsequently dropped from the placebo arm of the trial (Schlaeger
ported a decline in neuropathic pain in the pregabalin group with 2017). In addition, two other participants withdrew from the trial
18% of participants reporting a S-LANSS score of over 12 (indica- due to adverse events (one in the placebo group and one in the pre-
tive of neuropathic pain) at baseline as compared to 13% at three gabalin group) at the three-month visit.
months. Conversely, neuropathic pain was reported by 27% of con-
trols at baseline and 38% at the end of trial, suggesting an increase b. Number of people experiencing other adverse effects
in neuropathic pain in the placebo group. When compared to place- Overall, by three months, there were seven reported cases of ad-
bo, there were no differences observed at any time point; at three verse effects. In the pregabalin group there were four cases of
months the mean pain score was two points lower, MD -2.00 (95% CI adverse events, two due to sleepiness; two due to dizziness and
-9.18 lower to 5.18 higher) (very low-quality evidence) (Analysis 1.1) drowsiness, with one of the two also complaining of dry mouth);
for participants in the pregabalin group. Similarly, as assessed by and one had stomach cramps after ingesting the trial medication
the NPSI, participants' self-reported pain showed no difference be- (but when taken after meals the participant felt better). In the
tween groups, although there was a decline in pain score from base- placebo control group, three participants reported adverse events;
line to end of the trial for both the placebo and pregabalin group. one due to sleepiness; one due to light-headedness; and one due to
The MD of 11 points between the pregabalin and placebo group was sleepiness and a mild tingling sensation in the fifth digit of the right
not statistically significant, MD -11.10 (95% CI -33.97 lower to 11.77 hand (Schlaeger 2017). We found that the risk of adverse events,
higher) (very low-quality evidence) (Analysis 1.1). although higher in the pregabalin group, was not significantly dif-
ferent from the placebo group, RR 1.33 (95% CI 0.39 lower to 4.62
2. Patient Global Impression of Change (PGIC) higher) (very low-quality evidence) (Analysis 1.4).
There was no report of PGIC in the single RCT included in this review c. Number of withdrawals due to any adverse events (including
(Schlaeger 2017). serious adverse events)
Secondary outcome Four participants left the trial, two for complications related to SCD
and two for adverse events related to treatment or placebo (Sch-
1. Time to improvement of symptoms
laeger 2017).
There was no report given on this outcome in the included trial
(Schlaeger 2017). 5. Withdrawal from treatment due to lack of efficacy
There was no report of withdrawal due to a lack of efficacy (Sch-

2. Change in psychosocial outcomes
laeger 2017).
a. QoL
DISCUSSION
The included trial assessed QoL using the SF-36 instrument and
reported data at baseline, one, two and three months (Table 1) Summary of main results
(Analysis 1.1; Analysis 1.2). Higher SF-36 scores indicate higher
QoL and less disability. Overall, the placebo group and the prega- The authors of this review screened 182 references (abstract and ti-
balin group reported similar SF-36 mean scores at baseline and tles) of which just one was trial (22 participants) was included (Sch-
the three time points of the trial. The MD in SF-36 specific domain laeger 2017).
mean scores varied between the pregabalin and placebo groups.
Whereas for the treatment group, the mean SF-36 domain scores Regarding self-reported neuropathic pain relief, given the paucity
increased with time for seven of the eight domains (the only excep- of data, we are very uncertain whether there is a difference between
tion was for 'physical function' domain where the mean score fell), the pregabalin and placebo groups at the end of three months as
the scores decreased for four domains (physical health-role limita- measured by both the S-LANSS and the NPSI scales (very low-qual-
tion, emotional health-role limitations, emotional well-being, and ity evidence). There was no report of 'Patient Global Impression of
social functioning) of the eight studied domains for the placebo Change' in the included trial. We are also very uncertain whether
group. At three months, none of the MDs, however, were statistical- there is a difference in various QoL domains over the duration of the
ly significant (very low-quality evidence) (Analysis 1.3). trial, between the pregabalin and placebo groups. Similar numbers
of treatment-related adverse effects and withdrawals were seen
across both groups in trial. No other outcomes were reported on in
this trial.
Informed decisions.

Overall completeness and applicability of evidence ture. There is therefore some limited potential that some trials were
missed.
The evidence from this synthesis, while consistent with our review
question, is inadequate to inform practice and has limited gener- Agreements and disagreements with other studies or
alisability. The single trial in this review focused on an adult popu- reviews
lation (18 years and older) addressing a single pharmacological in-
tervention versus placebo for SCD-related neuropathic pain (Sch- There are very few studies addressing the efficacy of interventions
laeger 2017). The trial had a fairly small number of participants and for neuropathic pain in persons with SCD and we found no existing
was of short duration, and therefore the evidence it provides is in- review with which to compare our findings.
sufficient to form robust conclusions as to which is the most effec-
tive intervention for treating neuropathic pain in people with SCD. AUTHORS' CONCLUSIONS
Hence, there remains a significant gap in the evidence on interven-
tions for treating neuropathic pain in this population. Implications for practice
There is insufficient rigorous evidence on which to base concrete
Quality of the evidence conclusions about interventions for neuropathic pain in people
The evidence from this review has been assessed as very-low quali- with sickle cell disease (SCD). In light of this, current interventions
ty. The review includes only one pilot trial with 22 participants (with used by clinicians must balance potential benefits against any pos-
over 15% loss to follow-up or dropouts over three months) (Sch- sible adverse effects of treatment in this population. Additionally,
laeger 2017). This recent trial adhered for the most part to the Con- more randomised control trials (RCTs) evaluating the effectiveness
solidated Standards of Reporting Trials (CONSORT) (Moher 2010). of different interventions for neuropathic pain in SCD are urgent-
The evidence was at 'high' risk of bias for incomplete outcome data ly needed. Our review has highlighted a major gap in the evidence
and 'unclear' risk of bias for random sequence generation. All other pool and highlights the need for caution and the conscientious use
domains were assessed as being at 'low' risk of bias. As a result of of currently available interventions for treating neuropathic pain in
the risk of bias assessment, the level of evidence was downgraded people with SCD. The single RCT of pregabalin versus placebo for
as appropriate randomisation might impact the trial findings (Sum- treating neuropathic pain in people with SCD presented very low-
mary of findings for the main comparison). The level of evidence quality evidence and we are unsure whether there is a treatment
was further downgraded due to the small number of events and few difference. This pilot trial was not sufficiently powered to identify
participants in the trial, which was additionally affected by high lev- a significant effect of this intervention in people with SCD and it re-
els of attrition. The trial was also downgraded once due to indirect- mains to be seen if the trend to improvement shown in this trial can
ness, with the majority being female and all over 18 years of age. be duplicated or strengthened by more robust studies in the future.
Therefore, we were unable to make robust conclusions regarding
our objectives on the basis of a single included trial which only ad- Implications for research
dressed three of the seven pre-specified outcomes. This systematic review has identified a major gap in the research
area of treatments for managing neuropathic pain in people with
Potential biases in the review process SCD. Well-conducted RCTs are needed to follow on from this pilot
Standard systematic review methodology and guidance from the trial and to assess the effectiveness and possible adverse effects of
Cochrane Handbook for Systematic Reviews of Interventions was the different interventions for treating neuropathic pain in people
employed to reduce bias in the review process (Higgins 2011c). This with SCD.
included the use of at least two review authors in every aspect of
identifying potential trials, deciding on inclusion and exclusion of ACKNOWLEDGEMENTS
trials, extracting data, and conducting analyses. However, a few po-
Institutional support was provided by the Caribbean Institute for
tential sources of bias may remain.
Health Research (CAIHR) and the Cochrane Caribbean.
Publication bias
This project is supported by the National Institute for Health Re-
A comprehensive search of both published and unpublished search, via Cochrane Infrastructure funding to the Cochrane Cystic
sources was undertaken to find relevant trials. However, we did not Fibrosis and Genetic Disorders Group. The views and opinions ex-
make contact with drug companies for data on RCTs which might pressed therein are those of the authors and do not necessarily re-
not show up readily in our search of the published and grey litera- flect those of the Systematic Reviews Programme, NIHR, NHS or the
Department of Health.
Informed decisions.

REFERENCES

References to studies included in this review cell disease. Journal of Pain and Symptom Management
2012;43(6):1082-93. [CFGD Register: SC250]
Schlaeger 2017 {published data only}
Schlaeger JM, Molokie RE, Yao Y, Suarez ML, Golembiewski J, Horst 2016 {published data only}
Wilkie DJ, et al. Management of sickle cell pain using pregabalin: Horst J, Frei-Jones M, Deych E, Shannon W, Kharasch ED.
a pilot study. Pain Management Nursing 2017;18(6):391-400. Pharmacokinetics and analgesic effects of methadone in
[CFGD Register: SC317] children and adults with sickle cell disease. Pediatric Blood &
Cancer 2016;63(12):2123-30. [CFGD Register: SC311]

References to studies excluded from this review Howard 2018 {published data only}
Co 1979 {published data only} Howard J, Inusa B, Liossi C, Jacob E, Murphy PB, Hart N, et
Co LL, Schmitz TH, Havdala H, Reyes A, Westerman MP. al. Prevention of morbidity in sickle cell disease--qualitative
Acupuncture: an evaluation in the painful crises of sickle cell outcomes, pain and quality of life in a randomised cross-
anaemia. Pain 1979;7(2):181-5. [CFGD Register: SC24] over pilot trial of overnight supplementary oxygen and auto-
adjusting continuous positive airways pressure (poms2a): study
Ezenwa 2016 {published data only} protocol for a randomised controlled trial. Trials 2015;16:376.
[CFGD Register: SC328d]
Ezenwa M, Yao Y, Molokie R, Engeland C, Wang Z, Suarez M, et al.
Feasibility of an internet-enabled tablet-based guided imagery Howard J, Inusa B, Liossi C, Rees DC, Thein SL, Cheng E, et
for stress and pain reduction intervention in adults with sickle al. Prevention of morbidity in sickle cell disease (POMS 2): a
cell disease. Journal of Pain 2015;16(4 Suppl 1):S101. [CFGD pilot study of nocturnal respiratory support shows that auto-
Register: SC307a] adjusting positive airways pressure is safe and is preferred
to oxygen therapy. Blood 2015;126(23):993. [CFGD Register:
* Ezenwa MO, Yao Y, Engeland CG, Molokie RE, Wang ZJ,
SC328c]
Suarez ML, et al. A randomized controlled pilot study feasibility
of a tablet-based guided audio-visual relaxation intervention * Howard J, Slee AE, Skene S, Inusa B, Kawadler J, Downes M,
for reducing stress and pain in adults with sickle cell disease. et al. Overnight auto-adjusting continuous airway pressure
Journal of Advanced Nursing 2016;72(6):1452-63. [CFGD +standard care compared with standard care alone in the
Register: SC307b] prevention of morbidity in sickle cell disease phase II (POMS2b):
study protocol for a randomised controlled trial. Trials
Ezenwa 2018 {published data only}
2018;19(1). [CFGD Register: SC328a]
Ezenwa M, Yao Y, Nguyen M, Mandernach M, Hunter C, Yoon S,
et al. The effects of relaxation intervention on pain, stress, and Inusa B, Howard J, Chakravorty S, Pelidis M, Thein SL,
autonomic responses among adults with sickle cell pain in Kirkham FJ. Prevention of morbidity in sickle cell disease phase
the outpatient setting. Journal of Pain 2018;19(3 Supplement II (improvement of pain and quality of life in children with
1):S18-9. [CFGD Register: SC337] sickle cell disease with auto-adjusting continuous positive
airways pressure: phase II) (poms 2B pediatric cohort). Blood
Gil 1996 {published data only} 2017;130(Supplement 1):3537. [CFGD Register: SC328b]
* Gil KM, Carson JW, Sedway JA, Porter LS, Schaeffer JJ,
Orringer E. Follow-up of coping skills training in adults with Inusa B, Stotesbury H, Koelbel M, Kawadler J, Howard J,
sickle cell disease: analysis of daily pain and coping practice Rees D, et al. Prevention of morbidity in sickle cell disease
diaries. Health Psychology 2000;19(1):85-90. [CFGD Register: phase II (POMS 2B paediatric cohort): improvement of pain
SC39b] and quality of life in children with sickle cell disease with auto-
adjusting continuous positive airways pressure. British Journal
Gil KM, Wilson JJ, Abrams MA, Orringer E, Clark WC, Janal MN. of Haematology 2018;181(Supplement 1):115-6. [CFGD Regiter:
Coping skills training in adults with sickle cell disease: post- SC328h]
intervention and 3-month follow-up results. The National Sickle
Cell Disease Program Annual Meeting; 1996 March. 1996:08. Kawadler J, Liossi C, Inusa B, Clark C, Rees D, Pelidis M, et al.
[CFGD Register: SC39c] Reduced cortical thickness in patients with Sickle Cell Disease
and a high pain burden: baseline results from the Prevention of
Gil KM, Wilson JJ, Edens JL, Webster DA, Abrams MA, Orringer E, Morbidity in Sickle Cell Anaemia (POMS2b) trial. British Journal
et al. Effects of cognitive coping skills training on coping of Haematology 2017;176(Suppl 1):18. [CFGD Register: SC328e]
strategies and experimental pain sensitivity in African
American adults with sickle cell disease. Health Psychology Kolbel M, Stotesbury H, Kawadler J, Howard J, Inusa B, Rees D,
1996;15(1):3-10. [CFGD Register: SC39a] et al. Does auto-adjusting positive airway pressure (APAP) aid
memory and learning in children with sickle cell disease and
Hollins 2012 {published data only} sleep disordered-breathing?. British Journal of Haematology
Hollins M, Stonerock GL, Kisaalita NR, Jones S, Orringer E, 2018;181(Suppl 1):113. [CFGD Register: SC328a]
Gil KM. Detecting the emergence of chronic pain in sickle
Slee A, Stotesbury H, Kawadler J, Koelbel M, Chakravorty S,
Pelidis M, et al. Prevention of morbidity in sickle cell disease
Informed decisions.

phase 2 (poms 2B pediatric): improvement of cognition in Cell Disease Program; 2007 Sep 17-22; Washington DC, USA.
children with sickle cell disease with auto-adjusting continuous 2007:261. [CFGD Register: SC205a]
positive airways pressure: a single-blind, randomized,
controlled phase II trial. British Journal of Haematology * Wallen GR, Middleton KR, Ames N, Brooks AT, Handel D.
2018;181(Supplement 1):31. [CFGD Register: SC328i] Randomized trial of hypnosis as a pain and symptom
management strategy in adults with sickle cell disease.
Slee AE, Kawadler JM, Koelbel M, Stotesbury H, Rees DC, Integrative Medicine Insights 2014;9:25-33. [CFGD Register:
Pelidis M, et al. Prevention of morbidity in sickle cell disease SC205b]
phase II (improvement of cognition in children with sickle
cell disease with auto-adjusting continuous positive airways Wallen GR, Velummylum S, Middleton KR, Todaro A, Ames N,
pressure: phase II) (poms 2B paediatric cohort). Blood Miller-Davis C, et al. Hypnosis as a potential adjunct mind-body
2017;130(Supplement 1). [CFGD Register: SC328j] intervention in adults with sickle cell disease. American Journal
of Hematology 2012;87(7):E43. [CFGD Register: SC205c]
Stotesbury H, Kawadler J, Inusa B, Pelidis M, Chakravorty S,
Rees D, et al. Neurocognitive deficits in Sickle Cell Disease: links Wang 1988 {published data only}
with nocturnal oxygen desaturation in adolescents, but not * Wang WC, George SL, Wilimas JA. Transcutaneous electrical
children. British Journal of Haematology 2017;176(Suppl 1):89. nerve stimulation treatment of sickle cell pain crises. Acta
[CFGD Register: SC328f] Haematologica 1988;80(2):99-102. [CFGD Register: SC90a]
Matthie 2018 {published data only} Wang WC, Parker LJ, George SL, Harber JR, Presbury GJ,
Matthie N, Jenerette C, Griest Pell TJ, Lane D, Fu P, Wilimas JA. Transcutaneous electrical nerve stimulation (TENS)
Margevicius S, et al. The Effects of a Single Electronic Music treatment of sickle cell painful crises. Blood 1985;5. [Abstract
Improvisation Session on the Pain of Adults with Sickle Cell no.: 154; CFGD Register: SC90b]
Disease: a Mixed Methods Pilot Study. Journal of Music Therapy

2018;55(2):156-85. [CFGD Register: SC344]
Additional references
Myers 1999 {published data only} Anshel 1994
Myers CD, Robinson ME, Guthrie TH, Lamp SP, Lottenberg R. Anshel MH, Russell KG. Effect of aerobic and strength training
Adjunctive approaches for sickle cell chronic pain. Alternative on pain tolerance, pain appraisal and mood of unfit males
Health Practitioner 1999;5(3):203-12. [CFGD Register: SC151] as a function of pain location. Journal of Sports Sciences
1994;12(6):535-47.
Osunkwo 2012 {published data only}
Osunkwo I, Dampier C, Seydafkan S, Cherry K, Ghosh S, Ballas 2005
Eckman J, et al. Randomized double blind study evaluating Ballas SK, Lusardi M. Hospital readmission for adult acute
response to high dose vitamin D therapy in pediatric sickle sickle cell painful episodes: frequency, etiology, and prognostic
cell disease. American Journal of Hematology 2011;86(10):E19. significance. American Journal of Hematology 2005;79(1):17-25.
[Abstract no.: 028; CFGD Register: SC231b]
Ballas 2007
Osunkwo I, Ziegler T, Alvarez J, George J, Cherry K, Rhodes J,
et al. A randomized double blind, placebo controlled study Ballas SK. Current issues in sickle cell pain and its management.
of vitamin D to ameliorate sickle cell chronic pain. Journal of http://asheducationbook.hematologylibrary.org/
Pain 2012;13(4 Suppl 1):S73. [Abstract no.: 388; CFGD Register: content/2007/1/97.full (accessed 29 January 2017):97-105. [DOI:
SC231c] 10.1182/asheducation-2007.1.97]
* Osunkwo I, Ziegler TR, Alvarez J, McCracken C, Cherry K, Ballas 2011a

Osunkwo CE, et al. High dose vitamin D therapy for chronic Ballas SK. Defining the phenotypes of sickle cell disease.
pain in children and adolescents with sickle cell disease: results Hemoglobin 2011;35(5-6):511-9.
of a randomized double blind pilot study. British Journal of
Haematology 2012;159(2):211-5. [CFGD Register: SC231a] Ballas 2011b
Ballas SK. Update on pain management in sickle cell disease.
Palermo 2018 {published data only} Hemoglobin 2011;35(5-6):520-9.
Palermo TM, Dudeney J, Santanelli JP, Carletti A, Zempsky WT.
Feasibility and acceptability of internet-delivered cognitive Ballas 2012
behavioral therapy for chronic pain in adolescents with sickle Ballas SK, Gupta K, Adams-Graves P. Sickle cell pain: a critical
cell disease and their parents. Journal of Pediatric Hematology/ reappraisal. Blood 2012;120(18):3647-56.
Oncology 2018;40(2):122-7. [CFGD Register: SC325]
Ballas 2013
Wallen 2014 {published data only} Ballas SK, Darbari DS. Neuropathy, neuropathic pain, and sickle
Wallen GR, Handel D, Mendoza J, Chestnut W, Nichols J, Yates J, cell disease. American Journal of Hematology 2013;88(11):927-9.
et al. A randomized pilot study exploring hypnosis as a pain
and symptom management strategy in patients with sickle cell
disease. 35th Anniversary Convention of the National Sickle
Informed decisions.

Brandow 2014 _design/client/handbook/handbook.html (accessed 20 June

Brandow AM, Farley RA, Panepinto JA. Neuropathic pain in 2017).
patients with sickle cell disease. Pediatric Blood & Cancer
Heeney 2010
2014;61(3):512-7.
Heeney MM, Ware RE. Hydroxyurea for children with sickle
Brandow 2015a cell disease. Hematololgy/Oncology Clinics of North America
Brandow AM, Farley RA, Panepinto JA. Early insights into the 2010;24(1):199-214.
neurobiology of pain in sickle cell disease: A systematic review
Higgins 2011a
of the literature. Pediatric Blood & Cancer 2015;62(9):1501-11.
Higgins JPT, Deeks JJ, editor(s). Chapter 7: Selecting studies
Brandow 2015b and collecting data. In: Higgins JPT, Green S, editor(s). Cochrane
Brandow AM, Farley RA, Dasgupta M, Hoffmann RG, Handbook for Systematic Reviews of Interventions Version
Panepinto JA. The use of neuropathic pain drugs in children 5.1.0 (updated March 2011). The Cochrane Collaboration, 2011.
with sickle cell disease is associated with older age, female Available from handbook.cochrane.org.
sex, and longer length of hospital stay. Journal of Pediatric
Higgins 2011b
Hematology/Oncology 2015;37(1):10-5.
Higgins JPT, Altman DG, Sterne JAC, editor(s) on behalf of the
Brandow 2017 Cochrane Statistical Methods Group and the Cochrane Bias
Brandow AM, Zappia KJ, Stucky CL. Sickle cell disease: a natural Methods Group. Chapter 8: Assessing risk of bias in included
model of acute and chronic pain. Pain 2017;158 Suppl 1:S79-84. studies. In: Higgins JPT, Green S, editor(s). Cochrane Handbook
for Systematic Reviews of Interventions. Version 5.1.0 (updated
Charache 1997 March 2011). The Cochrane Collaboration, 2011. Available from
Charache S. Mechanism of action of hydroxyurea in the handbook.cochrane.org.
management of sickle cell anemia in adults. Seminars in
Higgins 2011c
Hematology 1997;34(3 Suppl 3):15-21.
Higgins JPT, Green S, editor(s). Cochrane Handbook for
Cooper 2016 Systematic Reviews of Interventions Version 5.1.0 (updated
Cooper TE, Hambleton IR, Ballas SK, Wiffen PJ. Pharmacological March 2011). The Cochrane Collaboration, 2011. Available from
interventions for painful sickle cell vaso-occlusive crises in handbook.cochrane.org.
adults. Cochrane Database of Systematic Reviews 2016, Issue 5.
Hildenbrand 2014
[DOI: 10.1002/14651858.CD012187]
Hildenbrand AK, Nicholls EG, Daly BP, Marsac ML, Tarazi R,
Deeks 2011 Deepti R. Psychosocial and pharmacological management
Deeks JJ, Higgins JPT, Altman DG on behalf of the Cochrane of pain in pediatric sickle cell disease. Postgraduate Medicine
Statistical Methods Group. Chapter 9: Analysing data 2014;126(2):123-33.
and undertaking meta-analysis. In: Higgins JPT, Green S,
Jensen 2011
editor(s). Cochrane Handbook for Systematic Reviews of
Interventions Version 5.1.0 (updated March 2011). The Cochrane Jensen TS, Baron R, Haanpaa M, Kalso E, Loeser JD,
Collaboration, 2011. Available from handbook.cochrane.org. Rice AS, et al. A new definition of neuropathic pain. Pain
2011;152(10):2204-5.
Dworkin 2008
Kalso 2013
Dworkin RH, Turk DC, Wyrwich KW, Beaton D, Cleeland CS,
Farrar JT, et al. Interpreting the clinical importance of Kalso E, Aldington DJ, Moore RA. Drugs for neuropathic pain.
treatment outcomes in chronic pain clinical trials: IMMPACT BMJ 2013;347:f7339.
recommendations. Journal of Pain 2008;9(2):105-21.
Lutz 2015
Edwards 2010 Lutz B, Meiler SE, Bekker A, Tao YX. Updated mechanisms
Edwards LY, Edwards CL. Psychosocial treatments in pain of sickle cell disease-associated chronic pain. Translational
management of sickle cell disease. Journal of the National Perioperative and Pain Medicine 2015;2:8-17.
Medical Association 2010;102(11):1084-94.
McClish 2005
Finnerup 2015 McClish DK, Penberthy LT, Bovbjerg VE, Roberts JD, Aisiku IP,
Finnerup, NB, Attal N, Haroutounian S, McNicol E, Baron R, Levenson JL, et al. Health related quality of life in sickle
Dworkin RH, et al. Pharmacotherapy for neuropathic pain in cell patients: the PiSCES project. Health and Quality of Life
adults: a systematic review and meta-analysis. Lancet Neurology Outcomes 2005;3:50.
2015;14(2):162-73.
McGann 2011
GRADE 2013 McGann PT, Ware RE. Hydroxyurea for sickle cell anemia: what
Schünemann H, Brożek J, Guyatt G, Oxman A, editor(s). GRADE have we learned and what questions still remain?. Current
Handbook. http://gdt.guidelinedevelopment.org/central_prod/ Opinion in Hematology 2011;18(3):158-65.
Informed decisions.

Moher 2009 RevMan 2014 [Computer program]

Moher D, Liberati A, Tetzlaff J, Altman DG, PRISMA Group. The Nordic Cochrane Centre, The Cochrane Collaboration.
Preferred reporting items for systematic reviews and Review Manager (RevMan). Version 5.3. Copenhagen: The
meta-analyses: the PRISMA statement. PLoS Medicine Nordic Cochrane Centre, The Cochrane Collaboration, 2014.
2009;6(7):e1000097.
Schünemann 2011
Moher 2010 Schünemann HJ, Oxman AD, Vist GE, Higgins JPT, Deeks JJ,
Moher, David, Hopewell, Sally, Schulz, Kenneth F, et al. Glasziou P, et al on behalf of the Cochrane Applicability and
CONSORT 2010 Explanation and Elaboration: updated Recommendations Methods Group and the Cochrane Statistical
guidelines for reporting parallel group randomised trials. BMJ Methods Group. Chapter 12: Interpreting results and drawing
Vol. 340:c869. [DOI: https://doi.org/10.1136/bmj.c869] conclusions. In: Higgins JPT, Green S, editor(s). Cochrane
Handbook for Systematic Reviews of Interventions Version
Morley 1999 5.1.0 (updated March 2011). The Cochrane Collaboration, 2011.
Morley SJ, Eccleston C, Williams A. Systematic review and meta- Available from handbook.cochrane.org.
analysis of randomised controlled trials of cognitive behaviour
therapy and behaviour therapy for chronic pain in adults, Serjeant 2013
excluding headache. Pain 1999;80(1-2):1-13. Serjeant GR. The natural history of sickle cell disease. Cold
Spring Harbour Perspectives in Medicine 2013;3:a011783.
Moulin 2007
Moulin DE, Clark AJ, Gilron I, Ware MA, Watson CP, Sessle BJ, et Smith 2010
al. Pharmacological management of chronic neuropathic pain Smith WR, Scherer M. Sickle-cell pain: advances in
—consensus statement and guidelines from the Canadian Pain epidemiology and etiology. Hematology. American Society of
Society. Pain Research & Management 2007;12(1):13-21. [DOI: Hematology. Education Program 2010;2010:409-15.
10.1155/2007/730785.]
Sterne 2011
National 2013 Sterne JAC, Egger M, Moher D, editor(s) on behalf of the
National Institute for Health and Care Excellence. NICE. Cochrane Bias Methods Group. Chapter 10: Addressing
National Institute for Health and Care Excellence. Neuropathic reporting biases. In: Higgins JPT, Green S, editor(s). Cochrane
pain – pharmacological management. The pharmacological Handbook for Systematic Reviews of Interventions Version
management of neuropathic pain in adults in non-specialist 5.1.0 (updated March 2011). The Cochrane Collaboration, 2011.
settings. https://www.nice.org.uk/guidance/cg173/evidence/ Available from handbook.cochrane.org.
neuropathic-pain-pharmacological-management-full-
guideline-191621341 (accessed 20 June 2017). Stuart 2004
Stuart MJ, Nagel RL. Sickle-cell disease. Lancet
Nevitt 2017 2004;364(9442):1343-60.
Nevitt SJ, Jones AP, Howard J. Hydroxyurea (hydroxycarbamide)
for sickle cell disease. Cochrane Database of Systematic Reviews Vance 2014
2017, Issue 4. [DOI: 10.1002/14651858.CD002202.pub2] Vance CG, Dailey Dl, Rakel BA, Sluka KA. Using TENS for
pain control: the state of the evidence. Pain Management
Piel 2013a 2014;4(3):197-209.
Piel FB, Patil AP, Howes RE, Nyangiri OA, Gething PW, Dewi M, et
al. Global epidemiology of sickle haemoglobin in neonates: a Ware 1999
contemporary geostatistical model-based map and population Ware RE, Zimmerman SA, Schultz WH. Hydroxyurea as
estimates. Lancet 2013;381(9861):142-51. an alternative to blood transfusions for the prevention of
recurrent stroke in children with sickle cell disease. Blood
Piel 2013b 1999;94(9):3022-6.
Piel FB, Hay SI, Gupta S, Weatherall DJ, Williams TN. Global
burden of sickle cell anaemia in children under five, 2010-2050: WHO 2009
modelling based on demographics, excess mortality, and World Health Organization. Recognition of sickle cell anaemia
interventions. PLoS Med 2013;10:e1001484. as a public health problem. Sixty-Third Session. A/RES/63/237.
Geneva, 2009:1-2.
Rees 2010
Rees DC, Williams TN, Gladwin MT. Sickle-cell disease. Lancet Wierenga 2001
2010;376(9757):2018-31. Wierenga KJ, Hambleton IR, Lewis NA. Survival estimates for
patients with homozygous sickle-cell disease in Jamaica: a
clinic-based population study. Lancet 2001;357(9257):680-3.

* Indicates the major publication for the study

Informed decisions.

CHARACTERISTICS OF STUDIES
Characteristics of included studies [ordered by study ID]

Schlaeger 2017
Methods Pilot RCT; parallel design
Participants 22 participants
Sex: male: 6 (placebo: 2, pregabalin: 4); female: 16 (placebo: 9; pregabalin: 7)
Age: mean (SD) all participants: 33.1 (9.9) years (placebo: 34.1 (9.5) years; pregabalin: 32.0 (10.7) years.
Range 18 - 82 years
Disease status all participants: homozygous S (SS) genotype= 15; heterozygous SC genotype = 6; sick-
le-beta thalassaemia genotype (SBThal) = 1 (placebo: SS: 6; SC: 5; SBThal: 0; pregabalin: SS: 9; SC: 1;
SBThal: 1)
Pregabalin group:
at 1 month (n = 9) (1 withdrew without taking pregabalin, 1 withdrew for side effects)

at 2 months (n = 8) (1 was dropped-combative behavior)
at 3 months (n = 8)
Placebo group:
at 1 month (n = 11)
at 2 months (n = 10) (withdrew for side effects)
at 3 months (n = 8) (1 dropped-SCD symptoms/hospitalization, 1 missing due to hospitalization for sick-
le-cell pain)
Interventions Treatment group: pregabalin, dose and frequency of administration: initial dose: 75 mg twice daily; in-
creased by 75 mg increments, to a maximum of 600 mg daily, or decreased the dose by 75 mg per day if
necessary based on clinical presentation and pain level
Control group: placebo
Duration of treatment: 3 months
Length of follow-up: not stated
Outcomes Self-reported pain (NPSI, S-LANNS); QoL (QOL SF-36), adverse events, withdrawals due to adverse
events
Time points of measurements: baseline, 1 month, 2 months, 3 months
Notes Eligibility criteria

(a) diagnosed with SCD, receiving ongoing care at the clinic
(b) aged 18 - 82 years old
(c) a history of SCD pain that was not well controlled (pain now score ≥ 4 on a 0 - 10 scale at registra-
tion)
Risk of bias
Bias Authors' judgement Support for judgement
Random sequence genera- Unclear risk The University of Illinois Hospital and Health Sciences System experimental
tion (selection bias) pharmacist randomly assigned participants'. No explanation of how the par-
ticipants were randomly assigned by the pharmacist
Informed decisions.

Schlaeger 2017 (Continued)
Allocation concealment Low risk Pharmacist was the only person with knowledge of group assignment
(selection bias)
Blinding of participants Low risk Described as a double-blind pilot study

and personnel (perfor-
mance bias) The prescribing physician was blind to the study drug treatment (pregabalin
All outcomes or placebo), the physician was not blind to the dose (in mg), which allowed all
participants' doses to be titrated
Initial dose of 75 mg twice daily and increments of 75 mg was similar for both
groups
Blinding of outcome as- Low risk Pharmacist was the only person with knowledge of group assignment
sessment (detection bias)
All outcomes
Incomplete outcome data High risk 27% loss to follow-up

(attrition bias)
All outcomes Numbers completing the 3-month trial: pregabalin 8/11; placebo 8/11 (for fur-
ther information, see 'Interventions' in above table)
Selective reporting (re- Unclear risk Unable to assessed as protocol could not be accessed
porting bias)
Other bias Low risk Unlikely from the publication
NPSI: Neuropathic Pain Symptom Inventory

QoL: quality of life
RCT: randomised controlled trial
SCD: sickle cell disease
S-LANNS: Leeds Assessment of Neuropathic Symptoms and Signs scale

Characteristics of excluded studies [ordered by study ID]

Study Reason for exclusion
Co 1979 SCD population without neuropathic pain
Ezenwa 2016 SCD population without neuropathic pain
Ezenwa 2018 SCD population without neuropathic pain
Gil 1996 SCD population without neuropathic pain
Hollins 2012 SCD population without neuropathic pain
Horst 2016 SCD population without neuropathic pain
Howard 2018 Neuropathic pain not measured
Matthie 2018 SCD population without neuropathic pain
Myers 1999 SCD population without neuropathic pain
Osunkwo 2012 SCD population without neuropathic pain
Informed decisions.

Study Reason for exclusion
Palermo 2018 Neuropathic pain not measured
Wallen 2014 SCD population without neuropathic pain
Wang 1988 SCD population without neuropathic pain
SCD: sickle cel disease

DATA AND ANALYSES

Comparison 1. Pharmacological intervention (pregabalin) versus placebo
Outcome or subgroup title No. of No. of Statistical method Effect size

studies partici-
pants
1 Self-reported pain relief - S-LANSS 1 Mean Difference (IV, Fixed, 95% CI) Totals not selected
1.1 at 1 month 1 Mean Difference (IV, Fixed, 95% CI) 0.0 [0.0, 0.0]
1.2 at 2 months 1 Mean Difference (IV, Fixed, 95% CI) 0.0 [0.0, 0.0]
2 Self-reported pain relief - NPSI 1 Mean Difference (IV, Fixed, 95% CI) Totals not selected
2.1 at 1 month 1 Mean Difference (IV, Fixed, 95% CI) 0.0 [0.0, 0.0]
3 Quality of life (at 3 months) 1 Mean Difference (IV, Fixed, 95% CI) Totals not selected
3.1 Physical function 1 Mean Difference (IV, Fixed, 95% CI) 0.0 [0.0, 0.0]
3.2 Role limitation as a result of physical 1 Mean Difference (IV, Fixed, 95% CI) 0.0 [0.0, 0.0]
health
3.3 Role limitation as a result of emotion- 1 Mean Difference (IV, Fixed, 95% CI) 0.0 [0.0, 0.0]
al problems
3.4 Energy or fatigue 1 Mean Difference (IV, Fixed, 95% CI) 0.0 [0.0, 0.0]
3.5 Emotional well-being 1 Mean Difference (IV, Fixed, 95% CI) 0.0 [0.0, 0.0]
3.6 Social functioning 1 Mean Difference (IV, Fixed, 95% CI) 0.0 [0.0, 0.0]
3.7 Pain 1 Mean Difference (IV, Fixed, 95% CI) 0.0 [0.0, 0.0]
3.8 General health 1 Mean Difference (IV, Fixed, 95% CI) 0.0 [0.0, 0.0]
Informed decisions.

Outcome or subgroup title No. of No. of Statistical method Effect size

studies partici-
pants
4 Treatment-related adverse events (at 3 1 Risk Ratio (M-H, Fixed, 95% CI) Totals not selected
months)

Analysis 1.1. Comparison 1 Pharmacological intervention (pregabalin)
versus placebo, Outcome 1 Self-reported pain relief - S-LANSS.
Study or subgroup Pregabalin Placebo Mean Difference Mean Difference
N Mean(SD) N Mean(SD) Fixed, 95% CI Fixed, 95% CI
1.1.1 at 1 month
Schlaeger 2017 9 6.2 (6.8) 11 9.8 (6.1) -3.6[-9.32,2.12]

1.1.2 at 2 months
Schlaeger 2017 8 5.9 (4.1) 10 10.8 (8.4) -4.9[-10.83,1.03]

1.1.3 at 3 months
Schlaeger 2017 8 5.9 (3.9) 8 7.9 (9.6) -2[-9.18,5.18]
Favours pregabalin -100 -50 0 50 100 Favours placebo

versus placebo, Outcome 2 Self-reported pain relief - NPSI.
1.2.1 at 1 month
Schlaeger 2017 9 18.9 (13.8) 11 32.8 (21.5) -13.9[-29.48,1.68]

1.2.2 at 2 months
Schlaeger 2017 8 16 (14.8) 10 31.5 (25.3) -15.5[-34.24,3.24]

1.2.3 at 3 months
Schlaeger 2017 8 21.5 (14.2) 8 32.6 (29.8) -11.1[-33.97,11.77]

Analysis 1.3. Comparison 1 Pharmacological intervention
(pregabalin) versus placebo, Outcome 3 Quality of life (at 3 months).
1.3.1 Physical function
Schlaeger 2017 8 60 (17.9) 8 67 (18.6) -7[-24.89,10.89]

1.3.2 Role limitation as a result of physical health
Informed decisions.


Schlaeger 2017 8 43.8 (47.7) 8 37.5 (46.3) 6.3[-39.76,52.36]

1.3.3 Role limitation as a result of emotional problems
Schlaeger 2017 8 79.2 (39.6) 8 58.3 (49.6) 20.9[-23.08,64.88]

1.3.4 Energy or fatigue
Schlaeger 2017 8 44.4 (20.9) 8 49.4 (30.9) -5[-30.85,20.85]

1.3.5 Emotional well-being
Schlaeger 2017 8 75.5 (13.6) 8 60 (32.4) 15.5[-8.85,39.85]

1.3.6 Social functioning
Schlaeger 2017 8 68.8 (17.7) 8 59.4 (37.1) 9.4[-19.08,37.88]

1.3.7 Pain
Schlaeger 2017 8 60.2 (31.9) 8 51.6 (21.8) 8.6[-18.17,35.37]

1.3.8 General health
Schlaeger 2017 8 36.1 (21.1) 8 45.8 (23.4) -9.7[-31.53,12.13]

versus placebo, Outcome 4 Treatment-related adverse events (at 3 months).
Study or subgroup Pregabalin Placebo Risk Ratio Risk Ratio
n/N n/N M-H, Fixed, 95% CI M-H, Fixed, 95% CI
Schlaeger 2017 4/11 3/11 1.33[0.39,4.62]
Favours pregabalin 0.01 0.1 1 10 100 Favours placebo

ADDITIONAL TABLES

Table 1. Change in psychosocial outcomes: SF-36 data (at 1 and 2 months)
Quality of life Pregabalin Control group*
group* (mean (SD))
(mean (SD))
Physical function At 1 month 67.8 (19.1) 59.5 (25.0)
At 2 months 56.3 (22.0) 65.5 (22.4)
Role imitation as a results of physical health At 1 month 48.7 (39.4) 47.7 (48.0)
At 2 months 56.3 (47.7) 50.0 (45.6)
Role imitation as a results of emotional problems At 1 month 92.1 (22.5) 72.7 (46.7)
Informed decisions.

Table 1. Change in psychosocial outcomes: SF-36 data (at 1 and 2 months) (Continued)

At 2 months 95.8 (11.8) 56.7 (44.6)
Energy/fatigue At 1 month 49.4 (17.8) 42.3 (27.5)
At 2 months 38.9 (22.6) 47.0 (28.8)
Emotional well-being At 1 month 73.3 (20.3) 72.7 (18.4)
At 2 months 81.0 (8.2) 64.0 (27.5)
Social functioning At 1 month 62.5 (23.4) 63.6 (28.2)
At 2 months 64.1 (25.4) 60.0 (28.7)
Pain At 1 month 57.0 (27.2) 57.3 (20.3)
At 2 months 56.6 (31.9) 50.0 (18.9)
General health At 1 month 38.7 (18.3) 44.2 (22.7)
At 2 months 37.8 (21.8) 45.7 (23.7)
At 1 month: pregabalin group = 9 participants; control group = 11 participants

At 2 months: pregabalin group = 8 participants; control group = 10 participants

APPENDICES
Appendix 1. Electronic searching

Database/ Resource Strategy
CARPHA EvIDeNCe Portal Search subject: sickle AND pain*
Limit your search by language: No limits (default option “documents written in English”
unticked)
ClinicalTrials.gov Other terms: pain

Study Type: Interventional Studies
Condition/ Disease: sickle
WHO ICTRP sickle AND pain*
EU Clinical Trials Register sickle AND pain*

CONTRIBUTIONS OF AUTHORS

Task Author(s) responsible
Informed decisions.

Protocol stage: draft the protocol Monika Asnani (MA) Damian Francis (DF) Amanda
Brandow (AB) Christine Hammond Gabbadon (CG)
Amza Ali (AA)
Review stage: select which trials to include (2 + 1 arbiter) CG, AB, Arbiter (MA)
Review stage: extract data from trials (2 people) AB, CG, AA
Review stage: enter data into RevMan DF, MA
Review stage: carry out the analysis DF, MA
Review stage: interpret the analysis DF, MA, AB, CG, AA
Review stage: draft the final review DF, MA, AB, CG, AA
Update stage: update the review MA, DF, AB, CG, AA

Review Guarantor: MA
DECLARATIONS OF INTEREST
Monika R Asnani, Damian K Francis, Christine EO Hammond Gabbadon, Sheik A Ali: none known.
Amanda M Brandow: is employed as an academic physician by the Medical College of Wisconsin. She receives funding from the National
Institutes of Health/National Heart Lung and Blood Institute to support her research in sickle cell disease.
SOURCES OF SUPPORT
Internal sources
• No sources of support supplied
External sources
• National Institute for Health Research, UK.
This systematic review was supported by the National Institute for Health Research, via Cochrane Infrastructure funding to the Cochrane
Cystic Fibrosis and Genetic Disorders Group.

Asnani2019 PDF

Uploaded by

Copyright:

Available Formats

Asnani2019 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

Asnani2019 PDF

Uploaded by

Copyright:

Available Formats

Asnani MR, Francis DK, Brandow AM, Hammond Gabbadon CEO, Ali A.

Interventions for treating neuropathic pain in people with sickle cell

Editorial group: Cochrane Cystic Fibrosis and Genetic Disorders Group

Date of last search: 31 January 2019.

Data collection and analysis

Treatments for neuropathic pain in persons with sickle cell disease

Quality of the evidence

Cochrane Database of Systematic Reviews

GRADE Working Group grades of evidence

Cochrane Database of Systematic Reviews

Figure 1. Flow Diagram

There was no report of withdrawal due to a lack of efficacy (Sch-

* Osunkwo I, Ziegler TR, Alvarez J, McCracken C, Cherry K, Ballas 2011a

Brandow 2014 _design/client/handbook/handbook.html (accessed 20 June

Moher 2009 RevMan 2014 [Computer program]

Characteristics of included studies [ordered by study ID]

Sex: male: 6 (placebo: 2, pregabalin: 4); female: 16 (placebo: 9; pregabalin: 7)

at 1 month (n = 9) (1 withdrew without taking pregabalin, 1 withdrew for side effects)

Control group: placebo

Duration of treatment: 3 months

Length of follow-up: not stated

Time points of measurements: baseline, 1 month, 2 months, 3 months

Notes Eligibility criteria

Bias Authors' judgement Support for judgement

Blinding of participants Low risk Described as a double-blind pilot study

Incomplete outcome data High risk 27% loss to follow-up

Other bias Low risk Unlikely from the publication

NPSI: Neuropathic Pain Symptom Inventory

Co 1979 SCD population without neuropathic pain

Ezenwa 2016 SCD population without neuropathic pain

Ezenwa 2018 SCD population without neuropathic pain

Gil 1996 SCD population without neuropathic pain

Hollins 2012 SCD population without neuropathic pain

Horst 2016 SCD population without neuropathic pain

Howard 2018 Neuropathic pain not measured

Matthie 2018 SCD population without neuropathic pain

Myers 1999 SCD population without neuropathic pain

Osunkwo 2012 SCD population without neuropathic pain

Study Reason for exclusion

Palermo 2018 Neuropathic pain not measured

Wallen 2014 SCD population without neuropathic pain

Wang 1988 SCD population without neuropathic pain

SCD: sickle cel disease

Outcome or subgroup title No. of No. of Statistical method Effect size

Outcome or subgroup title No. of No. of Statistical method Effect size

Favours pregabalin -100 -50 0 50 100 Favours placebo

Favours pregabalin -100 -50 0 50 100 Favours placebo

Favours pregabalin -100 -50 0 50 100 Favours placebo

Study or subgroup Pregabalin Placebo Mean Difference Mean Difference

Favours pregabalin -100 -50 0 50 100 Favours placebo

Favours pregabalin 0.01 0.1 1 10 100 Favours placebo

Physical function At 1 month 67.8 (19.1) 59.5 (25.0)

At 2 months 56.3 (22.0) 65.5 (22.4)

At 2 months 56.3 (47.7) 50.0 (45.6)

Table 1. Change in psychosocial outcomes: SF-36 data (at 1 and 2 months) (Continued)

Energy/fatigue At 1 month 49.4 (17.8) 42.3 (27.5)

At 2 months 38.9 (22.6) 47.0 (28.8)

Emotional well-being At 1 month 73.3 (20.3) 72.7 (18.4)

At 2 months 81.0 (8.2) 64.0 (27.5)