Anticonvulsants for fibromyalgia

  • Review
  • Intervention

Authors


Abstract

Background

Fibromyalgia (FM) is a clinically well-defined chronic condition of unknown aetiology characterised by chronic widespread pain that often co-exists with sleep problems and fatigue. People often report high disability levels and poor health-related quality of life (HRQoL). Drug therapy focuses on reducing key symptoms and disability, and improving HRQoL. Anticonvulsants (antiepileptic drugs) are drugs frequently used for the treatment of chronic pain syndromes.

Objectives

To assess the benefits and harms of anticonvulsants for treating FM symptoms.

Search methods

We searched the Cochrane Central Register of Controlled Trials (CENTRAL) (Issue 8, 2013), MEDLINE (1966 to August 2013), PsycINFO (1966 to August 2013), SCOPUS (1980 to August 2013) and the reference lists of reviewed articles for published studies and www.clinicaltrials.gov (to August 2013) for unpublished trials.

Selection criteria

We selected randomised controlled trials of any formulation of anticonvulsants used for the treatment of people with FM of any age.

Data collection and analysis

Two review authors independently extracted the data of all included studies and assessed the risks of bias of the studies. We resolved discrepancies by discussion.

Main results

We included eight studies: five with pregabalin and one study each with gabapentin, lacosamide and levetiracetam. A total of 2480 people were included into anticonvulsants groups and 1099 people in placebo groups. The median therapy phase of the studies was 13 weeks. The amount and quality of evidence were insufficient to draw definite conclusions on the efficacy and safety of gabapentin, lacosamide and levetiracetam in FM. The amount and quality of evidence was sufficient to draw definite conclusions on the efficacy and safety of pregabalin in FM. Therefore, we focused on our interpretation of the evidence for pregabalin due to our greater certainty about its effects and its greater relevance to clinical practice. All pregabalin studies had a low risk of bias. Reporting a 50% or greater reduction in pain was more frequent with pregabalin use than with a placebo (risk ratio (RR) 1.59; 95% confidence interval (CI) 1.33 to 1.90; number needed to treat for an additional beneficial outcome (NNTB) 12; 95% CI 9 to 21). The number of people who reported being 'much' or 'very much' improved was higher with pregabalin than with placebo (RR 1.38; 95% CI 1.23 to 1.55; NNTB 9; 95% CI 7 to 15). Pregabalin did not substantially reduce fatigue (SMD -0.17; 95% CI -0.25 to -0.09; 2.7% absolute improvement on a 1 to 50 scale) compared with placebo. Pregabalin had a small benefit over placebo in reducing sleep problems by 6.2% fewer points on a scale of 0 to 100 (standardised mean difference (SMD) -0.35; 95% CI -0.43 to -0.27). The dropout rate due to adverse events was higher with pregabalin use than with placebo use (RR 1.68; 95% CI 1.36 to 2.07; number needed to treat for an additional harmful outcome (NNTH) 13; 95% CI 9 to 23). There was no significant difference in serious adverse events between pregabalin and placebo use (RR 1.03; 95% CI 0.71 to 1.49). Dizziness was reported as an adverse event more frequently with pregabalin use than with placebo use (RR 3.77; 95% CI 3.06 to 4.63; NNTH 4; 95% CI 3 to 5).

Authors' conclusions

The anticonvulsant, pregabalin, demonstrated a small benefit over placebo in reducing pain and sleep problems. Pregabalin use was shown not to substantially reduce fatigue compared with placebo. Study dropout rates due to adverse events were higher with pregabalin use compared with placebo. Dizziness was a particularly frequent adverse event seen with pregabalin use. At the time of writing this review, pregabalin is the only anticonvulsant drug approved for treating FM in the US and in 25 other non-European countries. However, pregabalin has not been approved for treating FM in Europe. The amount and quality of evidence were insufficient to draw definite conclusions on the efficacy and safety of gabapentin, lacosamide and levetiracetam in FM.

Plain language summary

Anticonvulsants for fibromyalgia

Researchers in The Cochrane Collaboration conducted a review of research about the effects of anticonvulsants (antiepileptic drugs) on fibromyalgia (FM). After searching for all relevant studies, they found eight studies with up to 3579 people. The anticonvulsants that they studied were gabapentin, lacosamide, levetiracetam and pregabalin. They found reliable conclusions for pregabalin only.

Key results: pregabalin compared with fake medication after an average of 13 weeks

- Pregabalin slightly reduces pain and sleep problems.
- Pregabalin slightly reduces fatigue.
- Pregabalin increases medication side effects, in particular dizziness.
- Pregabalin does not differ from fake medication in serious side effects.

Possible side effects may include drowsiness, weight gain, swelling in the legs, blurred vision and co-ordination problems. Rare complications may include angio-oedema (swelling that happens just below the surface of the skin, most often around the lips and eyes), allergies and diseases of the immune system, worsening of heart failure, impairment of a person's ability to drive or operate machinery, and abuse or dependency on pregabalin. Serious side effects, such as suicidal thoughts, are very rare.

What is FM and what are anticonvulsants?

People with FM suffer from chronic widespread pain, sleep problems and fatigue. There is no cure for FM at present. Treatments aim at relieving the symptoms and improving quality of life.

Anticonvulsants are medications which help people with epileptic seizures. Neurons are cells that send messages to the brain. For example some neurons send messages about light, sound or touch to the brain. If something goes wrong and the neurons send too many messages quickly, then people may feel pain more strongly. Anticonvulsants can help slow the neurons down in the spinal cord and the brain. Some anticonvulsants can reduce pain and sleep problems in chronic pain due to nerve damage and can stabilize mood in people with anxiety and depressive disorders. People with fibromyalgia can have more brain activity and stronger reactions to the stimulus from their senses, so anticonvulsants might help with that. Pregabalin is approved for use in FM patients in most countries of North and South America and Asia, but not in Europe.

Best estimate of what happens to people with FM when they take pregabalin

Pain:

- 9 more people out of 100 who took pregabalin reached 50% or greater pain reduction than people who took fake medication (9% absolute improvement).

- 23 out of 100 people had their pain reduced by 50% or greater when they took pregabalin compared with 14 out of 100 people when they took a fake medication.

Global improvement:

- 11 more people out of 100 who took pregabalin improved 'much' and 'very much' globally than people who took fake medication (12% absolute improvement).

- 39 out of 100 people improved 'much' and 'very much' globally when they took pregabalin compared with 28 out of 100 people when they took a fake medication.

Fatigue (measured on a 1 to 50 scale):

- People who took pregabalin reported a slight reduction of fatigue (2.7% absolute improvement) than people who took a fake medication.

- People who took pregabalin rated their fatigue as 34.6 compared with 35.6 in people who took a fake medication.

Sleep problems (measured on a 0 to 100 scale):

- People who took pregabalin reported a reduction of sleep problems (6.2% absolute improvement) than people who took a fake medication.

- People who took pregabalin rated their sleep problems to be 54.9 compared with 58.6 in people who took fake medication

Stopping medication due to side effects:

- 7 more people out of 100 stopped pregabalin compared with fake medication because of side effects (8% absolute deterioration).

- 18 people out of 100 who took anticonvulsants stopped medication due to side effects compared with 11 people out of 100 who took a fake medication.

Serious side effects:

- There were no differences between pregabalin (5.2%) and fake medication (4.1%) in the number of serious adverse events.

Dizziness:

- 24 more people out of 100 reported dizziness as a side effect of pregabalin (28% absolute deterioration).

- 35 people out of 100 who took pregabalin reported dizziness as a side effect compared with 11 people out of 100 who took a fake medication.

Summary of findings(Explanation)

Summary of findings for the main comparison. Pregabalin versus placebo at end of treatment for fibromyalgia
  1. 1 Frequency of serious adverse events not reported by one study.

    ** Arnold 2008: n = 190 participants; MAF (NRS 1-50).

    *** Arnold 2008: n = 190 participants; MOS Overall Sleep Problems Index (NRS 0-100).

Pregabalin versus placebo at final treatment for fibromyalgia
Patient or population: People with fibromyalgia
Settings: Research centres
Intervention: Pregabalin versus placebo at final treatment
OutcomesIllustrative comparative risks* (95% CI)Relative effect
(95% CI)
No of participants
(studies)
Quality of the evidence
(GRADE)
Comments
Assumed riskCorresponding risk
Control Anticonvulsants versus placebo at final treatment
50% pain reduction 137 per 1000 217 per 1000
(182 to 260)
RR 1.59
(1.33 to 1.90)
3256
(5 studies)
⊕⊕⊕⊕
high

Absolute risk difference (fewer pain) 8% (95% CI 6% to 11%)

Relative per cent improvement 59% (95% CI 33% to 90%)

NNTB 12 (95% CI 9 to 21)

Patient Global Impression of Change of 'much' or 'very much' improved 279 per 1000 385 per 1000
(344 to 432)
RR 1.38
(1.23 to 1.55)
3183
(5 studies)
⊕⊕⊕⊕
high

Absolute risk difference (more global impression of 'much' and 'very much' improved) 12% (95% CI 4% to 20%)

Relative per cent improvement 38% (95% CI 23% to 55%)

NNTB 9 (95% CI 7 to 15)

Fatigue (1-50 scale)

Higher scores indicate higher fatigue levels

MAF baseline fatigue score control group

35.6 (standard deviation 8.0) **

The mean fatigue in the intervention groups was
0.17 standard deviations lower
(0.25 to 0.09 lower)
 3195
(5 studies)
⊕⊕⊕⊕
high

SMD -0.17 (-0.25 to -0.09)

2.7% (95% CI 1.4% to 4.0%) fewer points on the fatigue scale (absolute improvement)

3.8% (95% CI 2.0% to 5.6%) relative improvement

NNTB 13 (95% CI 9 to 25)

Sleep problems (0-100 scale). Higher scores indicate more sleep problems

MOS baseline overall sleep problem index control group

58.5 (17.8) ***

The mean sleep problems in the intervention groups was
0.35 standard deviations lower
(0.43 to 0.27 lower)
 3139
(5 studies)
⊕⊕⊕⊕
high

SMD -0.35 (-0.43 to -0.27)

6.2% (95% CI 4.8%

to 7.7%) fewer points on the sleep problem scale (absolute improvement)

10.6% (95% CI 82.% to 13.1%) relative improvement

NNTB 7 (95% CI 5 to 8)

Withdrawal due to adverse events 110 per 1000 185 per 1000
(150 to 229)
RR 1.68
(1.36 to 2.07)
3259
(5 studies)
⊕⊕⊕⊕
high

Absolute risk difference (more withdrawal due to adverse events) 8% (95% CI 5% to 12%)

Relative per cent worsening 68% (95% CI 36% to 107%)

NNTH 13 (95% CI 9 to 23)

Serious adverse events 41 per 1000 42 per 1000
(29 to 61)
RR 1.03
(0.71 to 1.49)
2729
(4 studies)
⊕⊕⊕⊝
moderate 1

Absolute risk difference 0 (95% CI -1 to 1)

Relative per cent change 0 (95% CI -1 to 1)

Not statistically significant

Dizziness reported to be an adverse event 93 per 1000 350 per 1000
(284 to 429)
RR 3.77
(3.06 to 4.63)
3257
(5 studies)
⊕⊕⊕⊕
high

Absolute risk difference (more dizziness) 28% (95% CI 24% to 32%)

Relative per cent worsening 277% (95% CI 206% to 363%)

NNTH 4 (95% CI 3 to 5)

*The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).

CI: confidence interval; MAF: Multidimensional Assessment of Fatigue; MOS: Medical Outcomes Study; NNTB: number needed to treat for an additional beneficial outcome; NNTH: number needed to treat for an additional harmful outcome; NRS: numeric rating scale; RR: risk ratio; SMD: standardised mean difference.

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.

Background

Fibromyalgia (FM) is a common disorder occurring in all populations across the world. It has been estimated that five million adults have FM in the US (Lawrence 2008). The estimated overall prevalence of FM in the general population in European countries ranges from 2.1% (Wolfe 2013a) to 2.9% (Branco 2010). People often report high disability levels and poor quality of life, and make extensive use of medical care (Marschall 2011; Winkelmann 2011; Wolfe 1997). People with FM use a wide range of different types of drugs and non-pharmacological therapies (Bennett 2007; Bernardy 2013; Birse 2012; Corrigan 2012; Deare 2013; Derry 2012; Eccleston 2012; Häuser 2012a; Häuser 2013; Hearn 2012; Moore 2012a; Seidel 2013; Tort 2012; Wiffen 2013a; Wiffen 2013b).

Description of the condition

The key symptoms of FM are chronic widespread pain associated with sleep problems, cognitive dysfunction and physical fatigue (Häuser 2008; Wolfe 2010). Lacking a specific laboratory test, diagnosis is established by the patient's report of the key symptoms and a physician's exclusion of other illnesses that would account for the key symptoms (Eich 2012a; Wolfe 2010). For clinical diagnosis, the American College of Rheumatology (ACR) 1990 classification criteria (Wolfe 1990), the ACR 2010 preliminary diagnostic criteria (Wolfe 2010), and the Fibromyalgia Survey Diagnostic Criteria (Wolfe 2011a), can be used.

A biopsychosocial model of factors predisposing, triggering and perpetuating FM symptoms has been suggested (Sommer 2012). Polymorphisms of genes in the serotoninergic, dopaminergic and catecholaminergic systems (Buskila 2009), and physical and sexual abuse in childhood and adolescence (Häuser 2011a), predispose to FM. Social disadvantage, psychological and physical stress, and somatic diseases such as rheumatoid arthritis trigger the development of FM symptoms (Buskila 2009; Wolfe 2011). A small fibre pathology was detected in people with FM (Üçeyler 2013). Depressive disorders have a negative impact on the clinical outcome of FM (Lange 2010). Several factors, such as alterations of central pain pathways, hyporeactivity of the hypothalamus-pituitary-adrenal axis, increased systemic pro-inflammatory and reduced anti-inflammatory cytokine profiles, and disturbances in the dopaminergic and serotonergic systems have been associated with FM pathophysiology. However, none of these factors has been demonstrated to have a causal relationship to FM or to be specific for FM (Sommer 2012). The current pathophysiology concept views FM as the result of alterations in central processing of sensory input and its cognitive-emotional appraisal, along with aberrations in the endogenous inhibition of pain (Buskila 2009). Exposure to physical or psychosocial stressors as outlined above may contribute to dysfunctional pain processing (Bradley 2009; Sommer 2012).

Since specific treatment aimed at altering the pathogenesis is not possible, drug therapy focusing on symptom reduction is widely employed.

Description of the intervention

Central pain conditions, in which persistent pain occurs in the absence of objective anatomic abnormality, are presumed to respond best to centrally active neuromodulating agents such as antidepressants and anticonvulsants (Phillips 2011). Anticonvulsants reduce the release of several neurotransmitters and peptides including glutamate, noradrenaline (norepinephrine) and substance P (Dooley 2007). The reduced neurotransmitter release is presumed to account for the analgesic, anticonvulsant and anxiolytic effects of anticonvulsants.

How the intervention might work

Augmented central nervous system processing of stimuli might be one pathophysiological mechanism underlying FM (Gracely 2011). Pregabalin and gabapentin are structural analogues of the neurotransmitter gamma-aminobutyric acid (GABA). These second-generation anticonvulsants are alpha-2-delta ligands that bind to, and modulate, voltage-gated calcium channels. By reducing calcium influx at nerve terminals, both drugs diminish the release of several neurotransmitters, including glutamate, noradrenaline and substance P. This mechanism is assumed to be the basis for the analgesic, anticonvulsant and anxiolytic actions of the drugs (Dooley 2007). Both drugs differ in their pharmacokinetic and pharmacodynamic characteristics (Bockbrader 2010). Lacosamide, another anticonvulsant, is described as a functionalised amino acid molecule that selectively enhances the slow inactivation of voltage-gated sodium channels and interacts with the collapsin-response mediator protein-2. Voltage-gated sodium channels play an important role in the excitability of nociceptors (Beydoun 2009). The mode of action of the anticonvulsant, levetiracetam, is not fully elucidated, but it has been found to target high-voltage, N-type calcium channels, as well as the synaptic vesicle protein 2A (Ulloa 2009).

Why it is important to do this review

Previous Cochrane reviews have studied the efficacy and safety of gabapentin (Moore 2011), pregabalin (Moore 2009) and lacosamide (Hearn 2012) in chronic neuropathic pain and FM in adults. These reviews did not report on outcomes relevant for FM such as fatigue, sleep problems and psychological symptoms. The approach of combining different neuropathic pain syndromes and FM is continued in the protocol on antiepileptic drugs for neuropathic pain and FM (Wiffen 2013a).

There is a transatlantic difference in the approval of the anticonvulsant pregabalin as a treatment of FM by regulatory authorities (Briley 2010). Pregabalin has been approved by the US Food and Drug Administration (FDA) and by the Japanese Pharmaceutical and Medical Devices Agency but not by the European Medical Agency (EMA) for the management of FM. The FDA stated that the sponsors of pregabalin had provided adequate evidence of the benefits to support the indication for the management of FM (FDA 2007). The EMA denied clinically relevant effects for pregabalin in European patients. The adverse effects profile was considered to outweigh the benefits, given the lack of robust evidence of efficacy (European Medicines Agency 2009). Because of the divergent appraisals of pregabalin by the FDA and EMA, we saw the need to evaluate the benefits and harms of anticonvulsants according to more recently established methodological standards of pain medicine (Moore 2010b), to assist people with FM and doctors to share decision making on pharmacological treatment options.

Objectives

To assess the benefits and harms of anticonvulsant use in the treatment of FM.

Methods

Criteria for considering studies for this review

Types of studies

We included randomised controlled trials (RCTs), double-blind, with anticonvulsants and with a study duration of eight weeks and longer. We required a parallel design. We excluded RCTs with a cross-over or an enriched enrolment withdrawal design. We required full journal publication, with the exception of online clinical trial result summaries of otherwise unpublished clinical trials and abstracts with data for analysis. We excluded studies that were non-randomised, studies of experimental pain, case reports or series and clinical observations.

Types of participants

Participants of any age having a clinical diagnosis of FM by any published standardised criteria (Häuser 2010a; Wolfe 1990; Wolfe 2010; Wolfe 2011a).

Types of interventions

We included trials with anticonvulsants at any dose, administered by any route to reduce FM symptoms compared with placebo or another active drug administered by any route.

We allowed co-interventions, such as physical therapy or other drugs different from those being assessed in the trial.

We considered the following anticonvulsants used in pain therapy for this review (Attal 2010): carbamazepine, clonazepam, felbamate, gabapentin, hydantoin, levetiracetam, lacosamide, lamotrigine, oxcarbazepine, pregabalin, tiagabine, topiramate, valproic acid and vigabatrin.

Types of outcome measures

We followed some suggestions of the OMERACT (Outcome Measures in Rheumatology) Fibromyalgia Working Group on the key domains of FM (Mease 2009); the Initiative of Methods, Measurement and Pain Assessment in Clinical trials (IMMPACT) (Dworkin 2009); and of best practice in the reporting of systematic reviews in chronic pain (Moore 2010a), for selecting outcome measures. Furthermore, we selected a specific adverse event, namely self reported dizziness, as a measure of tolerability: dizziness had been reported to be the most frequent side effect of pregabalin (Straube 2010), and to be a major symptom of 'fibro fog' (Centers for Disease Control and Prevention 2011).

We derived the outcomes of efficacy (benefits) from the key domains as defined by a consensus among experts and people with FM (Mease 2009). We assessed harms by the safety and tolerability of anticonvulsants. We defined safety as the frequency of serious adverse events independent of any determination of study-relatedness by the investigators. We defined tolerability by the frequency of study withdrawals due to adverse event and by the frequency of dizziness reported as adverse event.

Major outcomes
  1. Self reported 50% or greater pain reduction: we used the following preference a. electronic diaries, paper diaries; b. 24-hour recall pain, weekly recall pain (visual analogue scale (VAS)); c. paper VAS, paper numeric 11-point ordinal numeric rating scale (NRS); d. combined pain measures.

  2. Patient-perceived improvement (Patient Global Impression of Change (PGIC)): number of participants who reported to be 'much' or 'very much' improved.

  3. Self reported fatigue: we used the following preference: validated combined scales (e.g. Multidimensional Fatigue Inventory (MFI), Fatigue Severity Scale (FSS), other validated scales (single-item scales (e.g. Fibromyalgia Impact Questionnaire (FIQ) fatigue VAS).

  4. Self reported sleep problems: we used the following preference: validated combined scales (e.g. Medical Outcomes Study (MOS) Sleep Problem Index, other validated combined scales), single-item assessment (e.g. FIQ sleep VAS, other single-item scales).

  5. Safety: serious adverse events.

  6. Tolerability:

    1. withdrawals due to adverse event;

    2. specific adverse event: dizziness.

Minor outcomes
  1. Self reported pain intensity.

  2. Self reported 30% or greater pain reduction.

  3. Self reported health-related quality of life (HRQoL) measured by the total score of the FIQ.

  4. Self reported depression: we used the following preference: validated combined scales (Hospital Anxiety and Depression Scale (HADS), Beck Depression Inventory (BDI), other validated scales), single-item assessment (e.g. FIQ subscale for depression, other single-item scales).

  5. Self reported anxiety: we used the following preference: validated combined scales (HADS, BAI Beck Anxiety Inventory, State Trait Anxiety Inventory (STAI), other validated scales), single-item scale (FIQ anxiety VAS, other single-item scales).

  6. Self reported disability (impairment of physical function): we used the following preference: validated combined scale (Short-Form Health Survey (SF-36) Physical Functioning, FIQ physical impairment, Brief Pain Inventory (BPI) interference from pain, other validated combined scales), single-item scale.

Search methods for identification of studies

Electronic searches

We performed an electronic search in the Cochrane Central Register of Controlled Trials (CENTRAL) (Issue 8, 2013), MEDLINE accessed through PubMed (1966 to August 2013) and SCOPUS accessed through Ovid (1980 to August 2013). The search strategy used for MEDLINE is detailed in Appendix 1. Furthermore, we searched www.clinicaltrials.gov (website of the US National Institute of Health) to August 2013 for unpublished trials. Our search included all languages.

Searching other resources

We searched bibliographies from reviewed articles and we retrieved relevant articles. We contacted content experts to assist in locating other relevant or unpublished studies.

Data collection and analysis

Selection of studies

Two review authors (WH, NÜ) independently scrutinised all the titles and abstracts revealed by the searches and determined which fulfilled the selection criteria. A third review author (CS) verified that the selection had been properly realised.

Data extraction and management

Two review authors (NÜ, WH) extracted data independently into a specially designed data extraction form. We resolved disagreements by discussion if necessary with the third review author (CS). One review author (WH) entered data into Review Manager 5 (RevMan 2012), and a second author (NÜ) checked them. We resolved discrepancies by discussion.

Assessment of risk of bias in included studies

Two review authors (NÜ, WH) independently extracted the following information from reports of included studies into standardised forms: methods (setting, design, duration, follow-up and statistical analysis of study), participants (mean age, gender, race, inclusion and exclusion criteria), intervention (dosages of anticonvulsants, rescue and co-medication allowed), type of outcome measures, type of safety assessment and sponsoring. Two review authors (NÜ, WH) independently assessed the risk of bias of each included trial. We resolved disagreements by consensus and, if needed, referral to a third review author (CS). For each included study, we assessed risk of bias against key criteria: random sequence generation (selection bias); allocation concealment (selection bias); incomplete outcome data (attrition bias); selective outcome reporting (reporting bias); blinding of participants and personnel (performance bias); blinding of outcome assessment (detection bias) in accordance with methods recommended by The Cochrane Collaboration (Higgins 2011). We judged each of these criteria explicitly using: 'low risk' of bias, 'high risk' of bias and 'unclear risk' of bias (either lack of information or uncertainty over the potential for bias).

Measures of treatment effect

The effect measures of choice were risk ratio (RR) for dichotomous data and standardised mean difference (SMD) (when different scales were used to measure the same outcomes) for continuous data. We expressed uncertainty using 95% confidence intervals (CIs). We used Cohen's categories to evaluate the magnitude of the effect size, calculated by SMD, with Hedges' g of 0.2 = small, 0.5 = medium and 0.8 = large (Cohen 1988). We labelled g < 0.2 to be a 'not substantial' effect size.

Unit of analysis issues

In the case of multiple anticonvulsant dosage arms compared with one placebo group, we adjusted the number of participants in the placebo group according to the number of participants in the different anticonvulsant-dosage arms for continuous outcomes.

Dealing with missing data

If both intention-to-treat (ITT) and per protocol data were reported, we used ITT data for analysis. The ITT population consisted of participants who were randomised, took the assigned study medication and provided at least one post-baseline assessment (modified ITT). Wherever possible, we assigned zero improvement to missing participants. However, most studies in chronic pain report results, including responder results, using last observation carried forward. This has been questioned as being potentially biased with outcomes of withdrawal being important outcomes that make last observation carried forward unreliable. Last observation carried forward can lead to overestimation of efficacy, particularly in situations where adverse event withdrawal rates differ between active and control groups. At this time, it is unclear what strategy can actually be used to deal with missing data inside studies (Moore 2012b). We examined and clearly reported imputation strategies.

Where means or standard deviations (SD) were missing, we attempted to obtain these data through contacting trial authors. Where SDs were not available from trial authors, we calculated them from t-values, CIs or standard errors, where reported in articles (Higgins 2011). Where missing SDs could not be calculated from t values, CIs or standard errors, we calculated them from SDs of studies with the same drug and the same outcome scale by a validated imputation method (Furukawa 2006). Where 30% and 50% pain reduction rates were not reported and not provided on request, we calculated them from means and SDs by a validated imputation method (Furukawa 2005).

Assessment of heterogeneity

We used the I2 statistic for heterogeneity. I2 statistic values less than 25% indicated low, 25% to 50% indicated moderate and 50% or greater indicated substantial heterogeneity (Higgins 2003).

Assessment of reporting biases

We addressed publication bias by visual inspection of funnel plots and tests for funnel plot asymmetry (Begg 1994; Egger 1997), when there were at least 10 studies included in the meta-analysis (Higgins 2011).

We addressed outcome reporting bias by checking if the means and SDs of all primary and secondary outcomes as outlined in the methods section of the studies published had been reported or had been provided on request.

Data synthesis

We undertook each meta-analysis using a random-effects model in Review Manager 5 (RevMan 2012).

Subgroup analysis and investigation of heterogeneity

We performed subgroup comparisons of different dosages of pregabalin since differences in the efficacy and adverse events of different dosages of pregabalin had been reported (Straube 2010). We performed subgroup comparisons of different types of anticonvulsants. We post-hoc decided not to perform the planned subgroup analysis of studies with American, European, and Japanese participants to test for potential transatlantic differences between the efficacy and adverse events of anticonvulsants for the following reasons. There was no study with only European participants available. We decided to report the results of the comparisons of European and Non-European participants as presented in the publication and provided on request (Pauer 2011), and by the EMA report (European Medicines Agency 2009). The comparison of American and Japanese participants would have been biased by the different dosage regimens in the studies with fixed dosages in the US studies and flexible dosage in the Japanese study (Ohta 2012).

Sensitivity analysis

The protocol "Antidepressants and centrally active agents for fibromyalgia syndrome" intended to perform sensitivity analyses (different statistical models applied, presence of temporal differences, diagnostic criteria used in the trial, according to the presence/absence of any mental disorder, according to the presence/absence of any concomitant systemic disease) (Nishishinya 2006).

'Summary of findings' table

The quality of evidence was rated by the GRADE approach with GRADEprofiler (Guyatt 2011).

The numbers needed to treat for an additional beneficial (NNTB) or harmful (NNTH) outcome for continuous variables (fatigue, sleep problems) in the 'Summary of findings' table were calculated using the Wells calculator software available at the Cochrane Musculoskeletal Group editorial office, which estimates the proportion of participants who will benefit from treatment from SMDs. The estimation of responders is nearly independent from the minimally important difference (MID) (Norman 2001). We used a minimal clinically important difference of 15% for calculation.

Results

Description of studies

See Characteristics of excluded studies and Characteristics of included studies tables for full description of studies.

Results of the search

We identified 710 studies. We excluded 700 references, as they did not meet inclusion criteria related with the interventions evaluated in this review. We did not find head-to-head comparisons of anticonvulsants or of anticonvulsants with other drugs used for FM treatment. We identified 10 studies potentially related with anticonvulsants and obtained the full text for each of them. We excluded two studies after full-text review (Crofford 2008; Roth 2012) (see Figure 1).

Figure 1.

Study flow diagram.

Included studies

We included five studies with pregabalin (Arnold 2008; Crofford 2005; Mease 2008; Ohta 2012; Pauer 2011), and one study each with gabapentin (Arnold 2007), lacosamide (UCB 2006), and levetiracetam (Rowbotham 2012) into analysis.

Study characteristics

Two studies were initiated by investigators and were financially supported by grants from the National Institute of Health (Arnold 2007; Rowbotham 2012). The gabapentin study also received support from the manufacturer (Arnold 2007). The remaining studies were initiated and financed by the manufacturer of pregabalin (Arnold 2008; Crofford 2005; Mease 2008; Ohta 2012; Pauer 2011) and lacosamide (UCB 2006). Conflicts of interest of the study investigators were reported by all except three studies (Pauer 2011; Rowbotham 2012; UCB 2006). All studies were conducted in research centres. The median number of study centres was 44 (range 1 to 84). One study did not report the number of study centres (UCB 2006). Of the five studies with pregabalin, three studies were conducted in the US (Arnold 2008; Crofford 2005; Mease 2008), one study in Japan (Ohta 2012), and one study included countries all over the world except for the US (Europe, Middle America, Australia, Canada, India and Korea) (Pauer 2011). The studies with gabapentin (Arnold 2007), lacosamide (UCB 2006), and levetiracetam (Rowbotham 2012), were conducted in the US. All studies had a parallel design. The median therapy phase of the studies was 13 (range 8 to 15) weeks.

Participant characteristics

The lacosamide study did not report details of inclusion and exclusion criteria and participant characteristics except for age and gender (UCB 2006). Race was not reported by the levetiracetam study (Rowbotham 2012). All studies included people over 18 years old. All studies excluded people with somatic diseases including inflammatory rheumatic diseases. All studies excluded people with unstable or severe mental disorders or psychological conditions. Details regarding the exclusion of people with moderate and severe depressive or anxiety disorders were not reported by any study. All studies excluded people taking centrally acting drugs such as antidepressants. The median of screened participants who were randomised was 64 (range 56 to 76). A total of 2480 participants were included into true drug and of 1099 participants in placebo groups. Middle-aged white women prevailed in all studies: the median of the mean age was 49 (range 47 to 50) years. The median of the percentage of women was 93% (range 89% to 96%). The median of the percentage of white women was 91% (range 0% to 97%).

Interventions

Four pregabalin studies tested three fixed dosage arms against placebo. Three studies tested 300, 450, and 600 mg/day against placebo (Arnold 2008; Mease 2008; Pauer 2011), and one study tested 150, 300, and 450 mg/day against placebo (Crofford 2005). One study each used a flexible dosage of pregabalin (300 or 450 mg/day) (Ohta 2012), of gabapentin (between 1200 and 2400 mg/day) (Arnold 2007), and of levetiracetam (up to 3000 mg/day) (Rowbotham 2012), according to efficacy and tolerability. The study with lacosamide used a fixed dosage of 400 mg/day (UCB 2006). The rescue medication was paracetamol (acetaminophen) or over-the-counter non-steroidal anti-inflammatory drugs (dosage not reported) in two trials (Arnold 2007; Ohta 2012), paracetamol less than 4 g/day and aspirin less than 325 mg/day in two trials (Arnold 2008; Crofford 2005), and paracetamol up to 4 g/day (Mease 2008). Three studies did not report details of rescue medication (Pauer 2011; Rowbotham 2012; UCB 2006).

Outcomes

Pain was assessed in one study by the BPI 24 mean pain score (Arnold 2007), in one study by the pain score of the McGill Pain Questionnaire (Crofford 2005), and in the other studies by a 0 to 10 or 0 to 100 NRS daily pain scale. Fatigue was assessed by the sum score of Multidimensional Assessment of Fatigue Questionnaire (MAF) in three studies (Arnold 2008; Crofford 2005; Mease 2008), in the remaining studies by the single-item fatigue score of the FIQ. Sleep disturbances were assessed by a single-item sleep interference scale in three studies (Arnold 2007; Rowbotham 2012; UCB 2006), and by the MOS Sleep Problems Index in the remaining studies.Depression and anxiety were assessed in all studies by the HADS except for one study that used the Montgomery Depression Rating Scale (MDRS) for depression and the FIQ single-item scale for anxiety (Arnold 2007), and one study that used the FIQ single-item scales (Rowbotham 2012). Disability was assessed by the BPI mean interference scale in two studies (Arnold 2007; UCB 2006), by the FIQ disability scale in one study (Rowbotham 2012), and by the physical functioning scale of the SF-36 in the remaining studies. HRQoL was assessed in all studies by the FIQ total score. No study assessed sexual function or cognitive disturbances. Tender point pain threshold measurements were only conducted in one study with gabapentin (Arnold 2007).

Excluded studies

We excluded one study with three dosage arms of pregabalin (300, 450 and 600 mg/day), 1051 participants and 26 weeks' double-blind duration and an enriched enrolment withdrawal design (primary endpoint: loss of therapeutic response, defined as < 30% pain reduction) from meta-analysis because the outcomes were different from the ones of the studies with a parallel design (Crofford 2008). We excluded one study with cross-over design and flexible dosages of pregabalin (300 or 450 mg/day) and 115 participants because of short (four weeks) study duration and because of cross-over design (Roth 2012).

Risk of bias in included studies

We compiled a 'Risk of bias' table (See Figure 2 and Figure 3 for 'Risk of bias' summary and graph). Most risks of bias were low. The only criteria indicating high risk of bias were incomplete outcome data and selective non-reporting by some studies. Detailed information regarding risk of bias assessments of every study are given in the Characteristics of included studies table.

Figure 2.

Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.

Figure 3.

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

Allocation

Random sequence generation and allocation concealment were adequate in all studies except for one that did not report details (UCB 2006).

Blinding

Blinding of participants, personnel and outcome assessors were adequate in all studies.

Incomplete outcome data

Three studies reported observed cases analyses (Arnold 2007; Rowbotham 2012; UCB 2006). All other studies used ITT analysis. These studies imputed missing data by last observation carried forward method.

Selective reporting

Visual inspection of funnel plots was not indicative for a publication bias. The studies with pregabalin had been registered for the application of an approval for FM management by regulatory agencies. We had searched these databases. Therefore, we do not assume a non-publication of RCTs with pregabalin in FM.

Arnold 2007 did not report anxiety and fatigue outcomes. Crofford 2005 did not report HRQoL outcomes and serious adverse events. Mease 2008 did not report the details of disability outcomes. Rowbotham 2012 did not report fatigue, anxiety, depression and disability outcomes. UCB 2006 did not report details of adverse events.

Other potential sources of bias

All studies excluded people with unstable mental disorders or psychological conditions. The external validity of the pregabalin studies is difficult to assess, because the exclusion criteria for mental disorders were not well defined and a standardised psychiatric interview was not conducted by any of the studies reviewed. A pooled analysis of three studies with pregabalin (Arnold 2008; Crofford 2005; Mease 2008), reported that baseline HADS scores indicated moderate-to-severe anxiety in 38% of participants and moderate-to-severe depressive symptoms in 27% of participants (Arnold 2010a). These HADS scores indicate that some participants with anxiety or depressive disorders, or both, were included in the trials.

Effects of interventions

See: Summary of findings for the main comparison Pregabalin versus placebo at end of treatment for fibromyalgia

Gabapentin versus placebo

50% or greater pain reduction: one study with 150 participants was entered into an analysis of a 50% or greater pain reduction. A total of 32 of 75 (42.7%) participants with gabapentin and 20 of 75 (26.7%) participants with placebo reported a 50% or greater pain reduction. The RR of a 50% or greater pain reduction by anticonvulsants versus placebo was statistically significant (RR 1.60; 95% CI 1.01 to 2.53; P value = 0.04) (Analysis 1.2).

Sleep problems: one study with 119 participants was entered into an analysis of the effects of gabapentin on reduction of sleep problems. The overall effect on sleep problems was statistically significant (SMD -0.71; 95% CI -1.08 to -0.24; P value < 0.001). Based on Cohen's categories, the effect of gabapentin versus placebo on fatigue was small (Analysis 1.3).

Withdrawal due to adverse events: one study with 150 participants was entered into an analysis of withdrawals due to adverse events. A total of 12 of 75 (16.0%) participants dropped out due to adverse events with gabapentin and 7 of 75 (9.3%) participants with placebo. The RR of dropping out due to adverse events was not statistically significant (RR 1.71; 95% CI 0.71 to 4.11; P value = 0.23) (Analysis 1.5).

Dizziness: one study with 150 participants was entered into an analysis of dizziness attributed to be an adverse event of gabapentin. A total of 38 of 75 (50.7%) participants with gabapentin and 23 of 75 (30.7%) participants with placebo reported dizziness that was attributed to therapy. The RR of dizziness attributed to gabapentin versus placebo was statistically significant (RR 2.71; 95% CI 1.21 to 6.07; P value = 0.02) (Analysis 1.6).

Pain: one study with 119 participants was entered into an analysis of the effects of gabapentin on pain reduction. The overall effect on pain was statistically significant (SMD -0.49; 95% CI -0.86 to -0.13; P value = 0.008) (Analysis 1.1).

30% or greater pain reduction: one study with 150 participants were entered into an analysis of a 30% or greater pain reduction. A total of 38 of 75 (50.7%) participants with gabapentin and 23 of 75 (30.7%) participants with placebo reported a 30% or greater pain reduction. The RR of a 30% or greater pain reduction by gabapentin versus placebo was statistically significant (RR 1.65; 95% CI 1.10 to 2.48; P value = 0.02) (Analysis 1.7).

Health-related quality of life (HRQoL): one study with 119 participants was entered into an analysis of the effects of gabapentin on the total score of the FIQ. The overall effect on HRQoL was statistically significant (SMD -0.66; 95% CI -1.03 to -0.29; P value < 0.001). Based on Cohen's categories, the effect of gabapentin versus placebo on HRQoL was not substantial (Analysis 1.4).

Depression: one study with 119 participants was entered into an analysis of the antidepressive effects of gabapentin. The overall effect on depression was statistically significant (SMD -0.52; 95% CI -0.89 to -0.16; P value = 0.0005). Based on Cohen's categories, the effect of gabapentin versus placebo on depression was moderate (Analysis 1.8).

Disability: one study with 119 participants was entered into an analysis of the effects of gabapentin on disability reduction. The overall effect on disability was statistically significant (SMD -0.94; 95% CI -1.32 to -0.56; P value < 0.0001). Based on Cohen's categories, the effect of gabapentin versus placebo on depression was large (Analysis 1.9).

Lacosamide versus placebo

One study was analysed. The number of participants with reported or analysable results ranged from 121 to 159. The imputation method could not be applied because baseline values were not reported. There was no significant difference on effect sizes and RRs for pain reduction (SMD -0.25; 95% CI -0.56 to 0.07) (Analysis 2.1), fatigue reduction (SMD -0.07; 95% CI -0.43 to 0.28) (Analysis 2.2), sleep problems (SMD -0.19; 95% CI -0.51 to 0.12) (Analysis 2.3), reductions of limitations of HRQoL (SMD -0.15; 95% CI -0.47 to 0.16) (Analysis 2.4), serious adverse events (RR 0.15; 95% CI 0.01 to 2.82) (Analysis 2.5), anxiety (SMD 0.00; 95% CI -0.34 to 0.34) (Analysis 2.7), depression reduction (SMD 0.11; 95% CI -0.23 to 0.45) (Analysis 2.8), disability reduction (RR -0.19; 95% CI -0.51 to 0.12) (Analysis 2.9), or PGIC 'much' or 'very much' improved (RR 1.32; 95% CI 0.85 to 2.04) (Analysis 2.10). However, the RR of dizziness attributed to lacosamide versus placebo was statistically significant (RR 2.34; 95% CI 1.08 to 5.06; P value = 0.03) (Analysis 2.6).

Levetiracetam versus placebo

One study was analysed. The number of participants with reported or analysable results ranged from 50 to 66. The pain response rates had to be calculated by an imputation method because the data had not been reported and were not provided on request. There was no significant difference on effect sizes or RRs of mean pain reduction (SMD -0.21; 95% CI -0.77 to 0.36) (Analysis 3.1), 50% pain reduction (RR 1.52; 95% CI 0.43 to 5.34) (Analysis 3.2), reduction of sleep problems (SMD -0.16; 95% CI -0.72 to 0.40) (Analysis 3.3), withdrawals due to adverse events (RR 0.87; 95% CI 0.21 to 3.56) (Analysis 3.5), and 30% pain reduction (RR 1.23; 95% CI 0.65 to 2.33) (Analysis 3.7). However, the RR of dizziness attributed to levetiracetam versus placebo was statistically significant (RR 1.30; (95% CI 1.02 to 1.66; P value = 0.04) (Analysis 3.6).

Pregabalin versus placebo

50% or greater pain reduction: five studies with 3256 participants were entered into an analysis of a 50% or greater pain reduction. A total of 514 of 2319 (22.2%) participants with pregabalin and 128 of 937 (13.7%) participants with placebo reported a 50% or greater pain reduction. The RR of a 50% or greater pain reduction by anticonvulsants versus placebo was significant (RR 1.59; 95% CI 1.33 to 1.90; P value < 0.0001). The NNTB by drug over placebo was 12 (95% CI 9 to 21) (Figure 4).

Figure 4.

Forest plot of comparison: 4 Pregabalin versus placebo at end of treatment, outcome: 4.2 50% pain reduction.

Patient global impression of change (PGIC): five studies with 3183 participants were entered into an analysis of the effects of anticonvulsants on PGIC 'much' or 'very much' improved. A total of 892 of 2265 (39.4%) participants with pregabalin and 256 of 918 (27.9%) participants with placebo reported to be 'much' or 'very much' improved. The RR of a PGIC 'much' or 'very much' improved by pregabalin versus placebo was statistically significant (RR 1.38; 95% CI 1.23 to 1.55; P value < 0.0001). The NNTB to achieve a benefit by pregabalin over placebo was 9 (95% CI 7 to 13) (see Analysis 4.13).

Fatigue: five studies with 3195 participants were entered into an analysis of the effects of pregabalin on fatigue reduction. The overall effect on fatigue was statistically significant (SMD -0.17; 95% CI -0.25 to -0.09; P value < 0.001). Based on Cohen's categories, the effect of pregabalin versus placebo on fatigue was not substantial (see Figure 5).

Figure 5.

Forest plot of comparison: 4 Pregabalin versus placebo at end of treatment, outcome: 4.3 Fatigue.

Sleep problems: five studies with 3193 participants were entered into an analysis of the effects of pregabalin on reduction of sleep problems. The overall effect on sleep problems was statistically significant (SMD -0.35; 95% CI -0.43 to -0.27; P value < 0.001). Based on Cohen's categories, the effect of anticonvulsants versus placebo on fatigue was small (see Figure 6).

Figure 6.

Forest plot of comparison: 4 Pregabalin versus placebo at end of treatment, outcome: 4.4 Sleep problems.

Withdrawal due to adverse events: five studies with 3259 participants were entered into an analysis of withdrawals due to adverse events. A total of 449 of 2317 (19.4%) participants dropped out due to adverse events with pregabalin and 104 of 942 (11.0%) participants with placebo. The RR of dropping out due to adverse events was statistically significant (RR 1.68; 95% CI 1.36 to 2.07; P value < 0.001). The NNTH by pregabalin over placebo was 12 (95% CI 9 to 17) (see Analysis 4.6).

Serious adverse events: four studies with 2729 participants were entered into an analysis of withdrawals due to serious adverse events. In 100 of 1921 (5.2%) participants with pregabalin and in 33 of 808 (4.1%) participants with placebo a serious adverse event was recorded. The RR of dropping out due to serious adverse events was not statistically significant (RR 1.03; 95% CI 0.71 to 1.49; P value = 0.99) (Analysis 4.7).

Dizziness: five studies with 3257 participants were entered into an analysis of dizziness attributed to be an adverse event of drug therapy. A total of 883 of 2319 (38.1%) participants with pregabalin and 87 of 938 (9.3%) participants with placebo reported dizziness that was attributed to drug therapy. The RR of dizziness attributed to pregabalin versus placebo was statistically significant (RR 3.77; 95% CI 3.06 to 4.63; P value < 0.001). The NNTH by pregabalin over placebo was 4 (95% CI 3 to 5) (see Analysis 4.8).

Mean pain reduction: five studies with 3252 participants were entered into an analysis of the effects of pregabalin on pain reduction. The overall effect on pain was statistically significant (SMD -0.28; 95% CI -0.35 to -0.20; P value < 0.001). According to Cohen's categories, the effect of pregabalin compared with placebo on pain was small (Analysis 4.1).

30% or greater pain reduction: five studies with 3259 participants were entered into an analysis of a 30% or greater pain reduction. A total of 928 of 2319 (40.0%) participants with pregabalin and 274 of 940 (29.1%) participants with placebo reported a 30% or greater pain reduction. The RR of a 30% or greater pain reduction by pregabalin versus placebo was significant (RR 1.37; 95% CI 1.22 to 1.53; P value < 0.0001). The NNTB by pregabalin over placebo was 9 (96% CI 7 to 13) (Analysis 4.9).

Health-related quality of life (HRQoL): four studies with 2724 participants were entered into an analysis of the effects of pregabalin on the total score of the FIQ. The overall effect on HRQoL was statistically significant (SMD -0.17; 95% CI -0.26 to -0.09; P value < 0.001). Based on Cohen's categories, the effect of pregabalin versus placebo on HRQoL was not substantial (Analysis 4.5).

Depression: five studies with 3212 participants were entered into an analysis of the antidepressive effects of pregabalin. The overall effect on depression was statistically significant (SMD -0.09; 95% CI -0.16 to -0.01; P value < 0.001). Based on Cohen's categories, the effect of pregabalin versus placebo on depression was not substantial (Analysis 4.11).

Anxiety: five studies with 3214 participants were entered into an analysis of the effects of pregabalin on anxiety reduction. The overall effect on anxiety was statistically significant (SMD -0.12; 95% CI -0.20 to -0.04; P value < 0.001). Based on Cohen's categories, the effect of pregabalin versus placebo on anxiety was not substantial (Analysis 4.10).

Disability: five studies with 3145 participants were entered into an analysis of the effects of pregabalin on disability reduction. The overall effect on disability was not statistically significant (SMD -0.01; 95% CI -0.11 to 0.09; P value = 0.82) (Analysis 4.12).

Subgroup analysis

Different dosages of pregabalin: there were no statistically significant overall subgroup differences in the reduction of mean pain (P value = 0.64), 30% or greater pain reduction (P value = 0.79), 50% or greater pain reduction (P value = 0.59), fatigue (P value = 0.49), sleep problems (P value = 0.29), limitations of disease-related HRQoL (P value = 0.75), anxiety (P value = 0.65), depression (P value = 0.96), PGIC (P value = 0.84), dropout due to adverse events (P value = 0.10), and serious adverse events (P value = 0.83). There was a statistically significant overall subgroup difference in the dropout rate due to adverse events (P value = 0.04), which was highest in the 600 mg/day arm. There was a statistically significant overall subgroup difference in the reduction of disability (P value = 0.03). The effect size of pregabalin flexible (SMD -0.29; 95% CI -0.47 to -0.12) on disability was statistically significant and small, the one for pregabalin 600 mg/day arm included zero and the ones for other pregabalin dosages arms were statistically significant but not substantial.

Studies with and without European participants: Pauer 2011 reported that the placebo-adjusted results demonstrated greater improvement in PGIC and sleep endpoints in European participants (n = 379) and greater improvement in pain and FIQ total score in non-European participants (n = 357). Discontinuation rates due to adverse events were more frequent in European participants (23.0%) than in non-European participants (16.0%). The EMA presented in its refusal assessment report for pregabalin the effect sizes of pregabalin 300, 450 and 600 mg/day on pain in European participants (European Medicines Agency 2009). The 95% CIs of all three dosages included zero indicating a non-significant superiority of pregabalin over placebo.

We did not conduct the intended subgroup analyses with gender and pain because individual participant data were not available.

Sensitivity analyses

The intended sensitivity analyses (different statistical models applied, presence of temporal differences, diagnostic criteria used in the trial, according to the presence/absence of any mental disorder, according to the presence or absence of any concomitant systemic disease) were not conducted, because the studies did not differ in these characteristics.

Discussion

Summary of main results

There is high-quality evidence that pregabalin has a small benefit compared with placebo in reducing pain and sleep problems. There is high-quality evidence that the effects of pregabalin compared with placebo in reducing fatigue, depression, anxiety and limitations of HRQoL are not substantial. There is high-quality evidence that there is no significant difference between pregabalin and placebo in the reduction of disability. There is high-quality evidence that the dropout rate due to adverse events with pregabalin is significantly higher than with placebo. There is moderate-quality evidence that the frequency of serious adverse events is similar between pregabalin and placebo. The amount and quality of evidence are insufficient to draw definite conclusions on the efficacy and safety of gabapentin, lacosamide and levetiracetam in FM. It would be inappropriate to suggest a class effect of anticonvulsants on FM symptoms.

Overall completeness and applicability of evidence

We are confident that we have identified all available evidence on pregabalin because all of the trials had been registered as part of the application for regulatory approval for FM management. We cannot rule out the possibility that negative study results with other anticonvulsants had not been published or have been missed by our search strategy.

The applicability (external validity) of evidence is strongly limited for the following reasons:

1. The studies were performed in research centres and not in routine clinical care. It is known that the efficacy of drug therapies is higher in the context of RCTs than in routine clinical care (Routman 2010).

2. The exclusion criteria were strict. Participants were not allowed to take some defined concomitant medications for their FM symptoms. This excluded a large number of people who were unwilling or unable to come off medications such as other antidepressants and anticonvulsants. For this reason, participant selection in the RCTs was biased towards recruiting participants better able to manage their symptoms without medication than seen in the community. Participants with other medical disorders, such as inflammatory rheumatic diseases, were also excluded. The study results also cannot be applied to people with FM complicated by other medical disorders. The study results cannot be applied to people with FM and concomitant mental disease, because no subgroup analysis of people with depressive or anxiety disorders had been presented by the manufacturer.

3. The majority of the participants were middle-aged women, making it difficult to apply the results to the general population. The manufacturer did not perform subgroup analyses of participants over 65 years of age or of male participants. The results of anticonvulsants cannot be applied to children and adolescents with FM because they had also been excluded.

Quality of the evidence

All of the reviewed studies except for one (Rowbotham 2012) had been sponsored by pharmaceutical companies. The quality of evidence of this review is based on the data presented in peer reviewed journals and some additional details which were provided on request by the pharmaceutical companies or principal investigators. However, not all data requested were provided. We had no access to individual participant data. A selective non-reporting of serious adverse events by one study is possible (Crofford 2005). Despite these limitations, the overall quality of evidence was high for most primary outcomes in the majority of studies (see Summary of findings for the main comparison).

Potential biases in the review process

Absence of a publication bias can never be confirmed.

Efficacy outcomes were analysed using last observation carried forward to impute for missing data. This procedure may lead to an overestimation of efficacy (Moore 2012b). The use of the baseline observation carried forward method has been recommended for analysis (Moore 2010b).

In addition, the 30% and 50% pain reduction rates of one study were not reported and not provided on request (Rowbotham 2012). We calculated these values using an established imputation method (Furukawa 2005).

The influence of allowed co-interventions (e.g. rescue medication) on positive effects and adverse events was unclear because type and dosage of co-interventions were not clearly reported or controlled for.

Agreements and disagreements with other studies or reviews

Our results on the efficacy, tolerability and safety of pregabalin are in line with those of one systematic review (Straube 2010), with individual participant data of four studies that we included in our review (Arnold 2008; Crofford 2005; Mease 2008; Pauer 2011). Significant benefit of pregabalin over placebo was seen for mean pain and sleep scores and the proportion of participants achieving at least 50% pain relief. A minority of participants had substantial benefit with pregabalin, and more had moderate benefit. Many participants had only marginal or no benefit (Moore 2009; Straube 2010). Straube 2010 reported a benefit of pregabalin over placebo on most of the individual domains of the SF-36. However, the authors found no significant effect on disability (physical role limitation) or any positive effect on physical functioning. In line with Tzellos 2010 and Straube 2010, we found no significant differences in the efficacy of pregabalin 300, 450, and 600 mg/day. The proportions of participants with somnolence, dizziness and more than 7% weight gain were significantly greater with pregabalin than with placebo (Straube 2010). The tolerability of pregabalin by people with FM was limited, because 18% of participants with anticonvulsants dropped out due to adverse events. In accordance with Straube 2010, we found a dose-response relationship of adverse events (dizziness, dropout rate). There was no difference with regard to serious adverse events between pregabalin and placebo (Straube 2010). However, it should be noted that rare serious adverse events with pregabalin, such as heart failure exacerbation (Page 2008), psychosis (Olaizola 2006), suicidal tendencies (Mutschler 2011), and abuse and dependency (Gahr 2013), had been reported outside the context of RCTs.

The EERW (Enriched Enrollment Randomized Withdrawal) trial of Crofford 2008, which could not be combined with the studies with a parallel design for meta-analysis, confirmed the efficacy of pregabalin in people with FM. The trial used a six-week open titration to allow participants to obtain a substantial benefit (at least 50% pain relief and 'much' or 'very much' improved on PGIC) and have tolerable adverse events. Participants who achieved this outcome then entered a randomised, double-blind, 26-week withdrawal phase in which either the established dose or placebo was used. During the open-label phase, 54% had at least 50% pain relief on a VAS over baseline and a PGIC of 'much' or 'very much' improved and so were able to enter the randomised double-blind phase. For pregabalin, 90 of 279 (32%) participants experienced loss of therapeutic response over 26 weeks, compared with 174 of 287 (61%) participants with placebo. The NNTH to prevent one person experiencing a loss of treatment response for FM was 4 (95% CI 3 to 5). A total of 37 of 279 (13.3%) participants dropped out due to adverse events with pregabalin and 20 of 287 (7.0%) participants with placebo (NNTH 16; 95% CI 9 to 75).

Our results on gabapentin are in line with a Cochrane review on gabapentin in neuropathic pain syndromes that included one study in FM (Arnold 2007). The amount of evidence for gabapentin in FM was low, excluding any confidence on its efficacy (Moore 2011). Our results that lacosamide is not effective in FM is in line with a Cochrane review on lacosamide in neuropathic pain syndromes (Hearn 2012), which included one study in FM (UCB 2006). We found no other studies or reviews on levetiracetam in FM. Of note, gabapentin, lacosamide and levetiracetam were not licensed by any regulatory agency for FM or mental disorders.

Considering the current differences in regulatory approval regarding the use of pregabalin in FM in the US and Japan versus Europe, it seems relevant to comment if our data support either of these positions. According to EMA analysis, pregabalin did not significantly reduce mean pain in European participants (European Medicines Agency 2009). Moreover, European participants dropped out more frequently due to adverse events (23%) than US (16%) (European Medicines Agency 2009) and Japanese participants (14%) (Ohta 2012). Individual participant data were provided by Pfizer to EMA, but not to us. Therefore, EMA, but not us, could perform analysis of European and American participants. Therefore, our review cannot provide support for any of these regulatory positions.

The data on the efficacy of anticonvulsants were gathered in the context of RCTs in which people with FM received more frequent and detailed attention by the number of study visits and control examinations than they receive in routine clinical care. The context of an RCT might explain the remarkable number of participants that reported a substantial pain reduction (15%) and an overall 'much' or 'very much' improvement in PGIC (28%) with placebo. We found comparable placebo response rates in studies with duloxetine, milnacipran and sodium oxybate (Häuser 2012b). The effectiveness of drugs in clinical practice can be lower. Of 3123 adults with FM who participated in an 11-year longitudinal study of the National Data Bank of Rheumatic Diseases, up to 39% were treated with centrally acting agents (duloxetine, gabapentin, milnacipran and pregabalin). For participants treated with these drugs, pain scores were reduced significantly following the start of these drugs. The mean pain score decreased from 6.2 to 6.0 (relative improvement of 2.8%). There was no significant improvement in fatigue or functional status (Wolfe 2013b).

The pooled data from three extension studies for up to one year suggested that the adverse event and safety profile of patients with FM treated with open-label pregabalin (75 to 300 mg twice daily) for up to one year were stable and were consistent with those of RCTs. 12.4% of the participants discontinued pregabalin because of adverse events within one year. The most commonly reported treatment-emergent adverse events with open-label pregabalin treatment were dizziness, somnolence, headache, peripheral oedema, and increased weight (Arnold 2012). The data of RCTs and of extension studies do not answer the question for how long pregabalin - if effective and tolerated - should be taken by the patient. Longitudinal studies demonstrated that FM symptoms persisted in nearly all participants over more than 10 years (Eich 2012a). No data are available on the long-term (> one year) effectiveness, tolerability and safety of pregabalin in FM.

If pregabalin is considered for the treatment of FM, potential benefits and harms should be frankly discussed between the physician and the patient. The warnings (angio-oedema, hypersensitivity reactions, impairment of a person's ability to drive or operate machinery) should be discussed (Häuser 2010a). A marked increase of reports on abuse and dependence of pregabalin has been noted (Gahr 2013). Pregabalin should be individually tapered according to effectiveness and tolerability. The recommended dose is 150 mg/day twice daily. Pregabalin should be started with 25 to 50 mg at night and at the weekend because of potential neuropsychiatric side effects. Titration in weekly increments of 25 to 50 mg based on tolerability and therapeutic response is recommended (Boomershine 2010).

Therapy of FM with drugs only, such as pregabalin alone, should be discouraged since current best practices in FM guidelines recommend using the combination of pharmacological therapy with aerobic exercise and psychological therapies (Eich 2012b; Fitzcharles 2013). Although the combination of pregabalin with non-pharmacological treatment options had not been tested in RCTs, combination therapy should be considered. This is especially true for such symptoms that are only marginally responsive to pregabalin, but can be effectively treated with other therapies such as aerobic exercise for fatigue (Häuser 2010b), and cognitive behavioural therapies for depression (Bernardy 2010).

Authors' conclusions

Implications for practice

Of all the anticonvulsants considered in this review, the amount and quality of evidence is only sufficient to draw definite conclusions on the efficacy and safety of pregabalin in FM. The potential benefit of pregabalin to reduce pain is counterbalanced by its potential harms: 217 of 1000 participants reported a substantial (at least 50%) pain reduction at end of treatment with pregabalin (137 per 1000 with placebo). A total of 185 per 1000 discontinued medication due to adverse events with pregabalin (110 per 1000 with placebo). The mean improvement of sleep problems by pregabalin compared with placebo was 10.6%. The incremental benefit over placebo to reduce fatigue, limitations of quality of life, depression and anxiety was not substantial from a statistical point of view. Pregabalin did not reduce disability.

Of all the anticonvulsants considered in this review, only pregabalin is approved for FM. However, its approval is not universal, with pregabalin being approved for use in most countries of America and Asia, but not in Europe. Pregabalin is approved for the treatment of general anxiety disorder in some European countries.

Lastly, it should be noted that the positive effects of pregabalin on FM symptoms has been proven only for six months in placebo-controlled trials (Crofford 2008). It is reasonable to consider a trial of omission with responders after six months. Pregabalin should be discontinued gradually.

Implications for research

The main research directions that would help:

Available studies with anticonvulsants in FM:

1. A re-analysis of the data using baseline observation carried forward and responder analysis where discontinuation is classified as non-response would allow a determination of the true efficacy of pregabalin in FM. These analyses should be performed by a research team not affiliated with the manufacturers of pregabalin.

Future studies with anticonvulsants in FM:

1. Pregabalin provides a substantial benefit along with mild or no side effects only to a minority of participants. Clinical effectiveness trials with anticonvulsants are necessary; they should provide information on the proportion of participants with a moderate or substantial response, with tolerable adverse events and with willingness to continue therapy (Moore 2010b).

2. The significance of pregabalin compared with established therapies, such as aerobic exercise, amitriptyline and cognitive behavioural therapies, in the management of FM still needs to be determined.

3. Clinical studies in FM should be conducted with patient samples representative for clinical practice including participants with co-morbid anxiety and depressive disorders and inflammatory rheumatic diseases. Moreover, studies should include participants with different disease durations and different severity levels all over the world. Subgroup analyses for males and elderly people should be conducted.

4. The assessment strategies of adverse events in randomised controlled trials should be standardised by regulatory agencies. Potential effects of co-interventions on outcomes should be controlled for.

5. The definition of subgroups (e.g. FM with and without general anxiety disorder) and the development of more tailored therapies are major tasks of future research.

Acknowledgements

We thank Dr. Petra Klose, Internal Medicine V (Integrative Medicine), Kliniken Essen-Mitte, Essen, Germany, for her assistance in the search of literature.

Data and analyses

Download statistical data

Comparison 1. Gabapentin flexible 1200 to 2400 mg/day versus placebo at end of treatment
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Pain1119Std. Mean Difference (IV, Random, 95% CI)-0.49 [-0.86, -0.13]
2 50% pain reduction1150Risk Ratio (IV, Random, 95% CI)1.6 [1.01, 2.53]
3 Sleep problems1119Std. Mean Difference (IV, Random, 95% CI)-0.71 [-1.08, -0.34]
4 Health-related quality of life1119Std. Mean Difference (IV, Random, 95% CI)-0.66 [-1.03, -0.29]
5 Withdrawal due to adverse events1150Risk Ratio (IV, Random, 95% CI)1.71 [0.71, 4.11]
6 Dizziness1150Risk Ratio (IV, Random, 95% CI)2.71 [1.21, 6.07]
7 30% pain reduction1150Risk Ratio (M-H, Random, 95% CI)1.65 [1.10, 2.48]
8 Depression1119Std. Mean Difference (IV, Random, 95% CI)-0.52 [-0.89, -0.16]
9 Disability1119Std. Mean Difference (IV, Random, 95% CI)-0.94 [-1.32, -0.56]
Analysis 1.1.

Comparison 1 Gabapentin flexible 1200 to 2400 mg/day versus placebo at end of treatment, Outcome 1 Pain.

Analysis 1.2.

Comparison 1 Gabapentin flexible 1200 to 2400 mg/day versus placebo at end of treatment, Outcome 2 50% pain reduction.

Analysis 1.3.

Comparison 1 Gabapentin flexible 1200 to 2400 mg/day versus placebo at end of treatment, Outcome 3 Sleep problems.

Analysis 1.4.

Comparison 1 Gabapentin flexible 1200 to 2400 mg/day versus placebo at end of treatment, Outcome 4 Health-related quality of life.

Analysis 1.5.

Comparison 1 Gabapentin flexible 1200 to 2400 mg/day versus placebo at end of treatment, Outcome 5 Withdrawal due to adverse events.

Analysis 1.6.

Comparison 1 Gabapentin flexible 1200 to 2400 mg/day versus placebo at end of treatment, Outcome 6 Dizziness.

Analysis 1.7.

Comparison 1 Gabapentin flexible 1200 to 2400 mg/day versus placebo at end of treatment, Outcome 7 30% pain reduction.

Analysis 1.8.

Comparison 1 Gabapentin flexible 1200 to 2400 mg/day versus placebo at end of treatment, Outcome 8 Depression.

Analysis 1.9.

Comparison 1 Gabapentin flexible 1200 to 2400 mg/day versus placebo at end of treatment, Outcome 9 Disability.

Comparison 2. Lacosamide 400 mg/day versus placebo at end of treatment
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Pain1158Std. Mean Difference (IV, Random, 95% CI)-0.25 [-0.56, 0.07]
2 Fatigue1121Std. Mean Difference (IV, Random, 95% CI)-0.07 [-0.43, 0.28]
3 Sleep problems1158Std. Mean Difference (IV, Random, 95% CI)-0.19 [-0.51, 0.12]
4 Health-related quality of life1157Std. Mean Difference (IV, Random, 95% CI)-0.15 [-0.47, 0.16]
5 Serious adverse events1159Risk Ratio (IV, Random, 95% CI)0.15 [0.01, 2.82]
6 Dizziness1159Risk Ratio (IV, Random, 95% CI)2.34 [1.08, 5.06]
7 Anxiety1133Std. Mean Difference (IV, Random, 95% CI)0.0 [-0.34, 0.34]
8 Depression1134Std. Mean Difference (IV, Random, 95% CI)0.11 [-0.23, 0.45]
9 Disability1158Std. Mean Difference (IV, Random, 95% CI)-0.19 [-0.51, 0.12]
10 Patient Global Impression of Change 'much' or 'very much' improved1134Risk Ratio (IV, Random, 95% CI)1.32 [0.85, 2.04]
Analysis 2.1.

Comparison 2 Lacosamide 400 mg/day versus placebo at end of treatment, Outcome 1 Pain.

Analysis 2.2.

Comparison 2 Lacosamide 400 mg/day versus placebo at end of treatment, Outcome 2 Fatigue.

Analysis 2.3.

Comparison 2 Lacosamide 400 mg/day versus placebo at end of treatment, Outcome 3 Sleep problems.

Analysis 2.4.

Comparison 2 Lacosamide 400 mg/day versus placebo at end of treatment, Outcome 4 Health-related quality of life.

Analysis 2.5.

Comparison 2 Lacosamide 400 mg/day versus placebo at end of treatment, Outcome 5 Serious adverse events.

Analysis 2.6.

Comparison 2 Lacosamide 400 mg/day versus placebo at end of treatment, Outcome 6 Dizziness.

Analysis 2.7.

Comparison 2 Lacosamide 400 mg/day versus placebo at end of treatment, Outcome 7 Anxiety.

Analysis 2.8.

Comparison 2 Lacosamide 400 mg/day versus placebo at end of treatment, Outcome 8 Depression.

Analysis 2.9.

Comparison 2 Lacosamide 400 mg/day versus placebo at end of treatment, Outcome 9 Disability.

Analysis 2.10.

Comparison 2 Lacosamide 400 mg/day versus placebo at end of treatment, Outcome 10 Patient Global Impression of Change 'much' or 'very much' improved.

Comparison 3. Levetiracetam up to 3000 mg/day versus placebo at end of treatment
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Pain150Std. Mean Difference (IV, Random, 95% CI)-0.21 [-0.77, 0.36]
2 50% pain reduction166Risk Ratio (IV, Random, 95% CI)1.52 [0.43, 5.34]
3 Sleep problems150Std. Mean Difference (IV, Random, 95% CI)-0.16 [-0.72, 0.40]
4 Health-related quality of life150Std. Mean Difference (IV, Random, 95% CI)0.14 [-0.42, 0.70]
5 Withdrawal due to adverse events166Risk Ratio (IV, Random, 95% CI)0.87 [0.21, 3.56]
6 Dizziness166Risk Ratio (IV, Random, 95% CI)1.3 [1.02, 1.66]
7 30% pain reduction166Risk Ratio (IV, Random, 95% CI)1.23 [0.65, 2.33]
Analysis 3.1.

Comparison 3 Levetiracetam up to 3000 mg/day versus placebo at end of treatment, Outcome 1 Pain.

Analysis 3.2.

Comparison 3 Levetiracetam up to 3000 mg/day versus placebo at end of treatment, Outcome 2 50% pain reduction.

Analysis 3.3.

Comparison 3 Levetiracetam up to 3000 mg/day versus placebo at end of treatment, Outcome 3 Sleep problems.

Analysis 3.4.

Comparison 3 Levetiracetam up to 3000 mg/day versus placebo at end of treatment, Outcome 4 Health-related quality of life.

Analysis 3.5.

Comparison 3 Levetiracetam up to 3000 mg/day versus placebo at end of treatment, Outcome 5 Withdrawal due to adverse events.

Analysis 3.6.

Comparison 3 Levetiracetam up to 3000 mg/day versus placebo at end of treatment, Outcome 6 Dizziness.

Analysis 3.7.

Comparison 3 Levetiracetam up to 3000 mg/day versus placebo at end of treatment, Outcome 7 30% pain reduction.

Comparison 4. Pregabalin versus placebo at end of treatment
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Pain53252Std. Mean Difference (IV, Random, 95% CI)-0.28 [-0.35, -0.20]
1.1 Pregabalin 150 mg/d1175Std. Mean Difference (IV, Random, 95% CI)-0.10 [-0.45, 0.24]
1.2 Pregabalin 300 mg/d4913Std. Mean Difference (IV, Random, 95% CI)-0.24 [-0.39, -0.09]
1.3 Pregabalin 450 mg/d4915Std. Mean Difference (IV, Random, 95% CI)-0.36 [-0.51, -0.20]
1.4 Pregabalin 600 mg/d3751Std. Mean Difference (IV, Random, 95% CI)-0.30 [-0.51, -0.09]
1.5 Pregabalin flexible 300 or 450 mg/d1498Std. Mean Difference (IV, Random, 95% CI)-0.24 [-0.41, -0.06]
2 50% pain reduction53256Risk Ratio (IV, Random, 95% CI)1.59 [1.33, 1.90]
2.1 Pregabalin 150 mg/d1175Risk Ratio (IV, Random, 95% CI)0.92 [0.39, 2.19]
2.2 Pregabalin 300 mg/d4915Risk Ratio (IV, Random, 95% CI)1.45 [1.03, 2.05]
2.3 Pregabalin 450 mg/d4916Risk Ratio (IV, Random, 95% CI)1.75 [1.23, 2.49]
2.4 Pregabalin 600 mg/d3752Risk Ratio (IV, Random, 95% CI)1.51 [1.04, 2.20]
2.5 Pregabalin flexible 300 mg/d or 450 mg/d1498Risk Ratio (IV, Random, 95% CI)1.88 [1.26, 2.83]
3 Fatigue53195Std. Mean Difference (IV, Random, 95% CI)-0.17 [-0.25, -0.09]
3.1 Pregabalin 150 mg/d1165Std. Mean Difference (IV, Random, 95% CI)-0.20 [-0.56, 0.15]
3.2 Pregabalin 300 mg/d4892Std. Mean Difference (IV, Random, 95% CI)-0.16 [-0.31, -0.01]
3.3 Pregabalin 450 mg/d4897Std. Mean Difference (IV, Random, 95% CI)-0.15 [-0.30, 0.01]
3.4 Pregabalin 600 mg/d3743Std. Mean Difference (IV, Random, 95% CI)-0.08 [-0.25, 0.09]
3.5 Pregabalin flexible 300 or 450 mg/d1498Std. Mean Difference (IV, Random, 95% CI)-0.31 [-0.49, -0.14]
4 Sleep problems53193Std. Mean Difference (IV, Random, 95% CI)-0.35 [-0.43, -0.27]
4.1 Pregabalin 150 mg/d1163Std. Mean Difference (IV, Random, 95% CI)-0.44 [-0.80, -0.08]
4.2 Pregabalin 300 mg/d4897Std. Mean Difference (IV, Random, 95% CI)-0.29 [-0.44, -0.14]
4.3 Pregabalin 450 mg/d4893Std. Mean Difference (IV, Random, 95% CI)-0.45 [-0.63, -0.27]
4.4 Pregabalin 600 mg/d3744Std. Mean Difference (IV, Random, 95% CI)-0.40 [-0.57, -0.23]
4.5 Pregabalin flexible 300 or 450 mg/d1496Std. Mean Difference (IV, Random, 95% CI)-0.20 [-0.37, -0.02]
5 Health-related quality of life42724Std. Mean Difference (IV, Random, 95% CI)-0.17 [-0.26, -0.09]
5.1 Pregabalin 300 mg/d3738Std. Mean Difference (IV, Random, 95% CI)-0.12 [-0.28, 0.05]
5.2 Pregabalin 450 mg/d3737Std. Mean Difference (IV, Random, 95% CI)-0.24 [-0.41, -0.07]
5.3 Pregabalin 600 mg/d3751Std. Mean Difference (IV, Random, 95% CI)-0.14 [-0.31, 0.02]
5.4 Pregabalin flexible 300 mg/d or 450 mg/d1498Std. Mean Difference (IV, Random, 95% CI)-0.20 [-0.37, -0.02]
6 Withdrawal due to adverse events53259Risk Ratio (IV, Random, 95% CI)1.68 [1.36, 2.07]
6.1 Pregabalin 150 mg/d1175Risk Ratio (IV, Random, 95% CI)1.19 [0.35, 4.08]
6.2 Pregabalin 300 mg/d4917Risk Ratio (IV, Random, 95% CI)1.54 [1.02, 2.34]
6.3 Pregabalin 450 mg/d4917Risk Ratio (IV, Random, 95% CI)2.02 [1.32, 3.09]
6.4 Pregabalin 600 mg/d3752Risk Ratio (IV, Random, 95% CI)2.53 [1.65, 3.86]
6.5 Pregabalin flexible 300 mg/d or 450 mg/d1498Risk Ratio (IV, Random, 95% CI)1.01 [0.65, 1.57]
7 Serious adverse events42729Risk Ratio (IV, Random, 95% CI)1.03 [0.71, 1.49]
7.1 Pregabalin 300 mg/d3738Risk Ratio (IV, Random, 95% CI)0.96 [0.50, 1.86]
7.2 Pregabalin 450 mg/d3741Risk Ratio (IV, Random, 95% CI)1.03 [0.52, 2.03]
7.3 Pregabalin 600 mg/d3752Risk Ratio (IV, Random, 95% CI)1.01 [0.55, 1.87]
7.4 Pregabalin flexible 300 mg/d or 450 mg/d1498Risk Ratio (IV, Random, 95% CI)2.98 [0.31, 28.42]
8 Dizziness53257Risk Ratio (IV, Random, 95% CI)3.77 [3.06, 4.63]
8.1 Pregabalin 150 mg/d1175Risk Ratio (IV, Random, 95% CI)2.44 [0.91, 6.54]
8.2 Pregabalin 300 mg/d4916Risk Ratio (IV, Random, 95% CI)3.11 [2.09, 4.65]
8.3 Pregabalin 450 mg/d4917Risk Ratio (IV, Random, 95% CI)3.95 [2.68, 5.82]
8.4 Pregabalin 600 mg/d3751Risk Ratio (IV, Random, 95% CI)4.00 [2.65, 6.03]
8.5 Pregabalin flexible 300 mg/d or 450 mg/d1498Risk Ratio (IV, Random, 95% CI)4.89 [2.89, 8.28]
9 30% pain reduction53259Risk Ratio (IV, Random, 95% CI)1.37 [1.22, 1.53]
9.1 Pregabalin 150 mg/d1176Risk Ratio (IV, Random, 95% CI)1.05 [0.62, 1.77]
9.2 Pregabalin 300 mg/d4916Risk Ratio (IV, Random, 95% CI)1.35 [1.08, 1.68]
9.3 Pregabalin 450 mg/d4917Risk Ratio (IV, Random, 95% CI)1.49 [1.20, 1.85]
9.4 Pregabalin 600 mg/d3752Risk Ratio (IV, Random, 95% CI)1.37 [1.07, 1.76]
9.5 Pregabalin flexible 300 mg/d or 450 mg/d1498Risk Ratio (IV, Random, 95% CI)1.32 [1.04, 1.68]
10 Anxiety53214Std. Mean Difference (IV, Random, 95% CI)-0.12 [-0.20, -0.04]
10.1 Pregabalin 150 mg/d1166Std. Mean Difference (IV, Random, 95% CI)0.07 [-0.28, 0.42]
10.2 Pregabalin 300 mg/d4903Std. Mean Difference (IV, Random, 95% CI)-0.05 [-0.20, 0.10]
10.3 Pregabalin 450 mg/d4900Std. Mean Difference (IV, Random, 95% CI)-0.15 [-0.30, 0.00]
10.4 Pregabalin 600 mg/d3749Std. Mean Difference (IV, Random, 95% CI)-0.16 [-0.32, 0.01]
10.5 Pregabalin flexible 300 mg/d or 450 mg/d1496Std. Mean Difference (IV, Random, 95% CI)-0.17 [-0.35, 0.01]
11 Depression53212Std. Mean Difference (IV, Random, 95% CI)-0.09 [-0.16, -0.01]
11.1 Pregabalin 150 mg/d1165Std. Mean Difference (IV, Random, 95% CI)-0.10 [-0.45, 0.25]
11.2 Pregabalin 300 mg/d4902Std. Mean Difference (IV, Random, 95% CI)-0.05 [-0.20, 0.10]
11.3 Pregabalin 450 mg/d4900Std. Mean Difference (IV, Random, 95% CI)-0.13 [-0.28, 0.02]
11.4 Pregabalin 600 mg/d3749Std. Mean Difference (IV, Random, 95% CI)-0.06 [-0.23, 0.10]
11.5 Pregabalin flexible 300 mg/d or 450 mg/d1496Std. Mean Difference (IV, Random, 95% CI)-0.09 [-0.27, 0.08]
12 Disability53145Std. Mean Difference (IV, Random, 95% CI)-0.01 [-0.11, 0.09]
12.1 Pregabalin 150 mg/d1166Std. Mean Difference (IV, Random, 95% CI)0.11 [-0.24, 0.47]
12.2 Pregabalin 300 mg/d4879Std. Mean Difference (IV, Random, 95% CI)0.04 [-0.11, 0.20]
12.3 Pregabalin 450 mg/d4888Std. Mean Difference (IV, Random, 95% CI)0.00 [-0.17, 0.18]
12.4 Pregabalin 600 mg/d3714Std. Mean Difference (IV, Random, 95% CI)0.03 [-0.14, 0.21]
12.5 Pregabalin flexible 300 mg/d or 450 mg/d1498Std. Mean Difference (IV, Random, 95% CI)-0.29 [-0.47, -0.12]
13 Patient Global Impression of Change 'much' or 'very much' improved53183Risk Ratio (IV, Random, 95% CI)1.38 [1.23, 1.55]
13.1 Pregabalin 150 mg/d1175Risk Ratio (IV, Random, 95% CI)1.30 [0.74, 2.29]
13.2 Pregabalin 300 mg/d4886Risk Ratio (IV, Random, 95% CI)1.25 [0.99, 1.57]
13.3 Pregabalin 450 mg/d4892Risk Ratio (IV, Random, 95% CI)1.54 [1.15, 2.06]
13.4 Pregabalin 600 mg/d3734Risk Ratio (IV, Random, 95% CI)1.37 [1.06, 1.76]
13.5 Pregabalin flexible 300 mg/d or 450 mg/d1496Risk Ratio (IV, Random, 95% CI)1.44 [1.11, 1.87]
Analysis 4.1.

Comparison 4 Pregabalin versus placebo at end of treatment, Outcome 1 Pain.

Analysis 4.2.

Comparison 4 Pregabalin versus placebo at end of treatment, Outcome 2 50% pain reduction.

Analysis 4.3.

Comparison 4 Pregabalin versus placebo at end of treatment, Outcome 3 Fatigue.

Analysis 4.4.

Comparison 4 Pregabalin versus placebo at end of treatment, Outcome 4 Sleep problems.

Analysis 4.5.

Comparison 4 Pregabalin versus placebo at end of treatment, Outcome 5 Health-related quality of life.

Analysis 4.6.

Comparison 4 Pregabalin versus placebo at end of treatment, Outcome 6 Withdrawal due to adverse events.

Analysis 4.7.

Comparison 4 Pregabalin versus placebo at end of treatment, Outcome 7 Serious adverse events.

Analysis 4.8.

Comparison 4 Pregabalin versus placebo at end of treatment, Outcome 8 Dizziness.

Analysis 4.9.

Comparison 4 Pregabalin versus placebo at end of treatment, Outcome 9 30% pain reduction.

Analysis 4.10.

Comparison 4 Pregabalin versus placebo at end of treatment, Outcome 10 Anxiety.

Analysis 4.11.

Comparison 4 Pregabalin versus placebo at end of treatment, Outcome 11 Depression.

Analysis 4.12.

Comparison 4 Pregabalin versus placebo at end of treatment, Outcome 12 Disability.

Analysis 4.13.

Comparison 4 Pregabalin versus placebo at end of treatment, Outcome 13 Patient Global Impression of Change 'much' or 'very much' improved.

Appendices

Appendix 1. MEDLINE search strategy

SearchQueryItems found
#1Search anticonvulsants128,656
#2Search anticonvulsant*[tiab] OR anti convulsant*[tiab] OR anti-convulsant*[tiab] OR antiepileptic*[tiab] OR anti epileptic*[tiab] OR anti-epileptic*[tiab]34,452
#3Search carbamazepine13,816
#4Search Carbamazepine[tiab] or Amizepin[tiab] or Amizepine[tiab] or Atretol[tiab] or Biston[tiab] or Calepsin[tiab] or Carbamazepin[tiab] or Carbategral[tiab] or Carbatrol[tiab] or Convuline[tiab] or Epimax[tiab] or Epitol[tiab] or Equetro[tiab] or Finlepsin[tiab] or Lexin[tiab] or Mazepine[tiab] or Neurotol[tiab] or Neurotop[tiab] or Servimazepin[tiab] or Sirtal[tiab] or Tegral[tiab] or Tegretal[tiab] or Tegretol [tiab]or Tegrital[tiab] or Telesmin[tiab] or Teril[tiab] or Timonil[tiab]11,543
#5Search Oxcarbazepine[tiab] or Apydan[tiab] or Oxocarbamazepine[tiab] or Oxocarbazepine[tiab] or Timox[tiab] or Trileptal[tiab]1297
#6Search Felbamat[tiab] or Felbatol[tiab] or Taloxa[tiab]15
#7Search Gabapentin[tiab] or Neurontin[tiab] or Neurotonin[tiab]3835
#8Search Levetiracetam[tiab] or Keppra[tiab]1845
#9Search Pregabalin[tiab] or Lyrica[tiab]1502
#10Search phenytoin16,173
#11Search Phenytoin[tiab] or Alepsin[tiab] or Aleviatin[tiab] or Antilepsin[tiab] or Antisacer[tiab] or Cansoin[tiab] or Citrullamon[tiab] or Comital[tiab] or Danten[tiab] or Dantoin[tiab] or Denyl[tiab] or Difetoin[tiab] or Differenin[tiab] or Difhydan[tiab] or Di Hydan[tiab] or Dihydan[tiab] or Dilantin[tiab] or Dintoin[tiab] or Dintoina[tiab] or Diphantoin[tiab] or Diphantoine[tiab] or Diphantoin[tiab] or Diphedal[tiab] or Diphedan[tiab] or Diphenin[tiab] or Diphenine[tiab] or Diphentoin[tiab] or Diphenylan[tiab] or Diphenytoin[tiab] or Ekko[tiab] or Epanutin[tiab] or Epelin[tiab] or Epilantin[tiab] or Eptal[tiab] or Eptoin[tiab] or Fenantoin[tiab] or Fenitoin[tiab] or Fenytoin[tiab] or Fenytoine[tiab] or Hidantal[tiab] or Hydantin[tiab] or Hydantinal [tiab]or Hydantoinal[tiab] or Idantoin[tiab] or Lepitoin[tiab] or Minetoin[tiab] or Neosidantoina[tiab] or Phenhydan[tiab] or Phenhydane[tiab] or Phenybin[tiab] or Phenydan[tiab] or Phenydantin[tiab] or Phenytoine[tiab] or Phenytoin[tiab] or Phenytoinum[tiab] or Phenytonium[tiab] or Sanepil[tiab] or Sodantoin[tiab] or Sodanton[tiab] or Solantoin[tiab] or Solantyl[tiab] or Tacosal[tiab] or Zentropil[tiab]10,425
#12Search lacosamide[tiab] or erlosamide[tiab] or vimpat[tiab]308
#13Search Lamotrigine[tiab] or Labileno[tiab] or Lamictal[tiab]3651
#14Search valproic acid11,669
#15Search Sodium[tiab] valproate[tiab] or Alpha Propylvalerate[tiab] or Alpha Propylvaleric Acid[tiab] or Apilepsin[tiab] or Convulex[tiab] or Depacon[tiab] or Depakene[tiab] or Depakin[tiab] or Depakine[tiab] or Deprakine[tiab] or Dipropylacetate[tiab] or Dipropylacetatic Acid[tiab] or Dipropyl Acetic Acid[tiab] or Dipropylacetic Acid[tiab] or Diprosin[tiab] or Epilim[tiab] or Ergenyl[tiab] or Everiden[tiab] or Goilim[tiab] or Labazene[tiab] or Leptilan[tiab] or Leptilanil[tiab] or Mylproin[tiab] or Myproic Acid[tiab] or Orfiril[tiab] or Orlept[tiab] or Propymal[tiab] or Valerin[tiab] or Valparin[tiab] or Valpro[tiab] or Valproate[tiab] or Valproate Sodium[tiab] or Vupral[tiab]7647
#16Search Tiagabin[tiab] or Gabitril[tiab]29
#17Search Topiramate[tiab] or Epitomax[tiab] or Topamax[tiab] or Topimax[tiab]3056
#18Search Clonazepam[tiab] or Clonex[tiab] or Klonopin[tiab] or Paxam[tiab] or Kriadex[tiab] or Ravotril[tiab] or Rivatril[tiab] or Rivotril[tiab]2933
#19Search ((((((((((((((((((#2) OR #3) OR #4) OR #5) OR #6) OR #7) OR #8) OR #9) OR #10) OR #11) OR #12) OR #13) OR #14) OR #15) OR #16) OR #17) OR #18) OR #1)144,982
#20Search Fibromialgia OR Fibrositis OR FMS12,163
#21Search fibromyalgi*7538
#22Search (#20 or #21)12,180
#23Search randomized controlled trial[pt] OR controlled clinical trial[pt] OR randomized[tiab] OR placebo[tiab] OR drug therapy[sh] OR randomly[tiab] OR trial[tiab] OR groups[tiab]3,215,866
#24Search animals[mh] NOT humans[mh]3,811,701
#25Search (#23 not #24)2,759,538
#26Search (#19 AND #22 AND #25)224

Appendix 2. CENTRAL search strategy

ID Search Hits
#1(anticonvulsants):ti,ab,kw or (Anti convulsant$ or anticonvulsant$ or anti-convulsant$ or anti epileptic$ or antiepileptic$ or anti-epileptic$):ti,ab,kw or (Carbamazepine or Amizepin or Amizepine or Atretol or Biston or Calepsin or Carbamazepin or Carbategral or Carbatrol or Convuline):ti,ab,kw or (Epimax or Epitol or Equetro or Finlepsin or Lexin or Mazepine or Neurotol or Neurotop or Servimazepin or Sirtal or Tegral or Tegretal or Tegretol or Tegrital or Telesmin or Teril or Timonil):ti,ab,kw in Trials3563
#2MeSH descriptor Anticonvulsants explode all trees1919
#3MeSH descriptor Carbamazepine explode all trees699
#4(Oxcarbazepine or Apydan or Oxocarbamazepine or Oxocarbazepine or Timox or Trileptal):ti,ab,kw or (Clonazepam or Clonex or Klonopin or Paxam or Kriadex or Ravotril or Rivatril or Rivotril):ti,ab,kw or (Felbamat or Felbatol or Taloxa):ti,ab,kw or (Gabapentin or Neurontin or Neurotonin):ti,ab,kw or (Levetiracetam or Keppra):ti,ab,kw in Trials1323
#5(Pregabalin or Lyrica):ti,ab,kw or (Phenytoin or Alepsin or Aleviatin or Antilepsin or Antisacer):ti,ab,kw or (Cansoin or Citrullamon or Comital or Danten or Dantoin or Denyl or Difetoin or Differenin or Difhydan ):ti,ab,kw or (Di Hydan or Dihydan or Dilantin or Dintoin or Dintoina or Diphantoin or Diphantoine or Diphantoin or Diphedal or Diphedan or Diphenin or Diphenine or Diphentoin):ti,ab,kw or (Diphenylan or Diphenytoin or Ekko or Epanutin or Epelin or Epilantin or Eptal or Eptoin or Fenantoin or Fenitoin or Fenytoin or Fenytoine):ti,ab,kw in Trials1258
#6(Hidantal or Hydantin or Hydantinal or Hydantoinal or Idantoin or Lepitoin or Minetoin or Neosidantoina ):ti,ab,kw or (Phenhydan or Phenhydane or Phenybin or Phenydan or Phenydantin or Phenytoine or Phenytoin ):ti,ab,kw or (Phenytoinum or Phenytonium or Sanepil or Sodantoin or Sodanton or Solantoin or Solantyl or Tacosal or Zentropil):ti,ab,kw or (lacosamide or erlosamide or vimpat):ti,ab,kw or (Lamotrigine or Labileno or Lamictal):ti,ab,kw in Trials1477
#7MeSH descriptor Phenytoin explode all trees500
#8(Valproic acid):ti,ab,kw or (Sodium valproate or Alpha Propylvalerate or Alpha Propylvaleric Acid or Apilepsin or Convulex or Depacon or Depakene or Depakin or Depakine or Deprakine ):ti,ab,kw or (Dipropylacetate or Dipropylacetatic Acid or Dipropyl Acetic Acid or Dipropylacetic Acid or Diprosin or Epilim or Ergenyl or Everiden ):ti,ab,kw or (Goilim or Labazene or Leptilan or Leptilanil or Mylproin or Myproic Acid or Orfiril or Orlept ):ti,ab,kw or (Propymal or Valerin or Valparin or Valpro or Valproate or Valproate Sodium or Vupral):ti,ab,kw in Trials1154
#9 MeSH descriptor Valproic Acid explode all trees648
#10(Tiagabin or Gabitril):ti,ab,kw or (Topiramate or Epitomax or Topamax or Topimax):ti,ab,kw or (Vigabatrin or Sabril):ti,ab,kw in Trials780
#11(fibromyalgia):ti,ab,kw or (fibromyalgi$):ti,ab,kw or (fibrositis):ti,ab,kw or (fms):ti,ab,kw in Trials987
#12MeSH descriptor Fibromyalgia explode all trees565
#13(randomized controlled trial):pt or (controlled tiral):pt or (random allocation):ti,ab,kw or (double blind):ti,ab,kw or (single blind):ti,ab,kw in Trials381,667
#14(clinical trial):pt and (pacebo$):ti,ab,kw and (random):ti,ab,kw and (clinical trial):ti,ab,kw and (randomized controlled trial or controlled trial):ti,ab,kw in Trials1
#15(placebo$):ti,ab,kw or (random$):ti,ab,kw or (clinical trial):ti,ab,kw or (clinical trial):pt or (randomized controlled trial or controlled trial):ti,ab,kw455,068
#16(animal):ti,ab,kw in Trials12,551
#17(human):ti,ab,kw in Trials462,550
#18(#1 OR #2 OR #3 OR #4 OR #5 OR #6 OR #7 OR #8 OR #9 OR #10)6177
#19(#11 OR #12)987
#20(#13 OR #14 OR #15)517,597
#21(#17 AND NOT #16)450,301
#22(#18 AND #19 AND #20 AND #21)32

Appendix 3. PsycInfo search strategy

  Searches Results
1(Anti convulsant$ or anticonvulsant$ or anti-convulsant$ or anti epileptic$ or antiepileptic$ or anti-epileptic$).ab. or (Anti convulsant$ or anticonvulsant$ or anti-convulsant$ or anti epileptic$ or antiepileptic$ or anti-epileptic$).ti. or (Anti convulsant$ or anticonvulsant$ or anti-convulsant$ or anti epileptic$ or antiepileptic$ or anti-epileptic$).id. or anticonvulsants.ab. or anticonvulsants.ti. or anticonvulsants.id. or anticonvulsants.sh. or (Anti convulsant$ or anticonvulsant$ or anti-convulsant$ or anti epileptic$ or antiepileptic$ or anti-epileptic$).sh.7859
2anticonvulsants.ab. or anticonvulsants.ti. or anticonvulsants.id. or anticonvulsants.sh.1603
3(Carbamazepine or Amizepin or Amizepine or Atretol or Biston or Calepsin or Carbamazepin or Carbategral or Carbatrol or Convuline or Epimax or Epitol or Equetro or Finlepsin or Lexin or Mazepine).ab. or (Carbamazepine or Amizepin or Amizepine or Atretol or Biston or Calepsin or Carbamazepin or Carbategral or Carbatrol or Convuline or Epimax or Epitol or Equetro or Finlepsin or Lexin or Mazepine).ti. or (Carbamazepine or Amizepin or Amizepine or Atretol or Biston or Calepsin or Carbamazepin or Carbategral or Carbatrol or Convuline or Epimax or Epitol or Equetro or Finlepsin or Lexin or Mazepine).id. or (Carbamazepine or Amizepin or Amizepine or Atretol or Biston or Calepsin or Carbamazepin or Carbategral or Carbatrol or Convuline or Epimax or Epitol or Equetro or Finlepsin or Lexin or Mazepine).sh.2925
4(Neurotol or Neurotop or Servimazepin or Sirtal or Tegral or Tegretal or Tegretol or Tegrital or Telesmin or Teril or Timonil).ab. or (Neurotol or Neurotop or Servimazepin or Sirtal or Tegral or Tegretal or Tegretol or Tegrital or Telesmin or Teril or Timonil).ti. or (Neurotol or Neurotop or Servimazepin or Sirtal or Tegral or Tegretal or Tegretol or Tegrital or Telesmin or Teril or Timonil).id. or (Neurotol or Neurotop or Servimazepin or Sirtal or Tegral or Tegretal or Tegretol or Tegrital or Telesmin or Teril or Timonil).sh.51
5(Clonazepam or Clonex or Klonopin or Paxam or Kriadex or Ravotril or Rivatril or Rivotril).ab. or (Clonazepam or Clonex or Klonopin or Paxam or Kriadex or Ravotril or Rivatril or Rivotril).ti. or (Clonazepam or Clonex or Klonopin or Paxam or Kriadex or Ravotril or Rivatril or Rivotril).id. or (Clonazepam or Clonex or Klonopin or Paxam or Kriadex or Ravotril or Rivatril or Rivotril).sh.914
6(Oxcarbazepine or Apydan or Oxocarbamazepine or Oxocarbazepine or Timox or Trileptal).ab. or (Oxcarbazepine or Apydan or Oxocarbamazepine or Oxocarbazepine or Timox or Trileptal).ti. or (Oxcarbazepine or Apydan or Oxocarbamazepine or Oxocarbazepine or Timox or Trileptal).id. or (Oxcarbazepine or Apydan or Oxocarbamazepine or Oxocarbazepine or Timox or Trileptal).sh.436
7(Felbamat or Felbatol or Taloxa).ab. or (Felbamat or Felbatol or Taloxa).ti. or (Felbamat or Felbatol or Taloxa).id. or (Felbamat or Felbatol or Taloxa).sh.2
8(Gabapentin or Neurontin or Neurotonin).ab. or (Gabapentin or Neurontin or Neurotonin).ti. or (Gabapentin or Neurontin or Neurotonin).id. or (Gabapentin or Neurontin or Neurotonin).sh.996
9(Levetiracetam or Keppra).ab. or (Levetiracetam or Keppra).ti. or (Levetiracetam or Keppra).id. or (Levetiracetam or Keppra).sh.642
10(Pregabalin or Lyrica).ab. or (Pregabalin or Lyrica).ti. or (Pregabalin or Lyrica).id. or (Pregabalin or Lyrica).sh.459
11(Phenytoin or Alepsin or Aleviatin or Antilepsin or Antisacer or Cansoin or Citrullamon or Comital or Danten or Dantoin or Denyl or Difetoin or Differenin or Difhydan).ab. or (Phenytoin or Alepsin or Aleviatin or Antilepsin or Antisacer or Cansoin or Citrullamon or Comital or Danten or Dantoin or Denyl or Difetoin or Differenin or Difhydan).ti. or (Phenytoin or Alepsin or Aleviatin or Antilepsin or Antisacer or Cansoin or Citrullamon or Comital or Danten or Dantoin or Denyl or Difetoin or Differenin or Difhydan).id. or (Phenytoin or Alepsin or Aleviatin or Antilepsin or Antisacer or Cansoin or Citrullamon or Comital or Danten or Dantoin or Denyl or Difetoin or Differenin or Difhydan).sh.882
12(Di Hydan or Dihydan or Dilantin or Dintoin or Dintoina or Diphantoin or Diphantoine or Diphantoin or Diphedal or Diphedan or Diphenin or Diphenine or Diphentoin).ab. or (Di Hydan or Dihydan or Dilantin or Dintoin or Dintoina or Diphantoin or Diphantoine or Diphantoin or Diphedal or Diphedan or Diphenin or Diphenine or Diphentoin).ti. or (Di Hydan or Dihydan or Dilantin or Dintoin or Dintoina or Diphantoin or Diphantoine or Diphantoin or Diphedal or Diphedan or Diphenin or Diphenine or Diphentoin).id. or (Di Hydan or Dihydan or Dilantin or Dintoin or Dintoina or Diphantoin or Diphantoine or Diphantoin or Diphedal or Diphedan or Diphenin or Diphenine or Diphentoin).sh.82
13(Diphenylan or Diphenytoin or Ekko or Epanutin or Epelin or Epilantin or Eptal or Eptoin or Fenantoin or Fenitoin or Fenytoin or Fenytoine).ab. or (Diphenylan or Diphenytoin or Ekko or Epanutin or Epelin or Epilantin or Eptal or Eptoin or Fenantoin or Fenitoin or Fenytoin or Fenytoine).ti. or (Diphenylan or Diphenytoin or Ekko or Epanutin or Epelin or Epilantin or Eptal or Eptoin or Fenantoin or Fenitoin or Fenytoin or Fenytoine).id. or (Diphenylan or Diphenytoin or Ekko or Epanutin or Epelin or Epilantin or Eptal or Eptoin or Fenantoin or Fenitoin or Fenytoin or Fenytoine).sh.1
14(Hidantal or Hydantin or Hydantinal or Hydantoinal or Idantoin or Lepitoin or Minetoin or Neosidantoina).ab. or (Hidantal or Hydantin or Hydantinal or Hydantoinal or Idantoin or Lepitoin or Minetoin or Neosidantoina).ti. or (Hidantal or Hydantin or Hydantinal or Hydantoinal or Idantoin or Lepitoin or Minetoin or Neosidantoina).id. or (Hidantal or Hydantin or Hydantinal or Hydantoinal or Idantoin or Lepitoin or Minetoin or Neosidantoina).sh.0
15(Phenhydan or Phenhydane or Phenybin or Phenydan or Phenydantin or Phenytoine or Phenytoin).ab. or (Phenhydan or Phenhydane or Phenybin or Phenydan or Phenydantin or Phenytoine or Phenytoin).ti. or (Phenhydan or Phenhydane or Phenybin or Phenydan or Phenydantin or Phenytoine or Phenytoin).id. or (Phenhydan or Phenhydane or Phenybin or Phenydan or Phenydantin or Phenytoine or Phenytoin).sh.883
16(Phenytoinum or Phenytonium or Sanepil or Sodantoin or Sodanton or Solantoin or Solantyl or Tacosal or Zentropil).ab. or (Phenytoinum or Phenytonium or Sanepil or Sodantoin or Sodanton or Solantoin or Solantyl or Tacosal or Zentropil).ti. or (Phenytoinum or Phenytonium or Sanepil or Sodantoin or Sodanton or Solantoin or Solantyl or Tacosal or Zentropil).id. or (Phenytoinum or Phenytonium or Sanepil or Sodantoin or Sodanton or Solantoin or Solantyl or Tacosal or Zentropil).sh.1
17(lacosamide or erlosamide or vimpat).ab. or (lacosamide or erlosamide or vimpat).ti. or (lacosamide or erlosamide or vimpat).id. or (lacosamide or erlosamide or vimpat).sh.88
18(Lamotrigine or Labileno or Lamictal).ab. or (Lamotrigine or Labileno or Lamictal).ti. or (Lamotrigine or Labileno or Lamictal).id. or (Lamotrigine or Labileno or Lamictal).sh.1472
19Valproic acid.ab. or Valproic acid.ti. or Valproic acid.id. or Valproic acid.sh.1856
20(Sodium valproate or Alpha Propylvalerate or Alpha Propylvaleric Acid or Apilepsin or Convulex or Depacon or Depakene or Depakin or Depakine or Deprakine).ab. or (Sodium valproate or Alpha Propylvalerate or Alpha Propylvaleric Acid or Apilepsin or Convulex or Depacon or Depakene or Depakin or Depakine or Deprakine).ti. or (Sodium valproate or Alpha Propylvalerate or Alpha Propylvaleric Acid or Apilepsin or Convulex or Depacon or Depakene or Depakin or Depakine or Deprakine).id. or (Sodium valproate or Alpha Propylvalerate or Alpha Propylvaleric Acid or Apilepsin or Convulex or Depacon or Depakene or Depakin or Depakine or Deprakine).sh.523
21(Dipropylacetate or Dipropylacetatic Acid or Dipropyl Acetic Acid or Dipropylacetic Acid or Diprosin or Epilim or Ergenyl or Everiden).ab. or (Dipropylacetate or Dipropylacetatic Acid or Dipropyl Acetic Acid or Dipropylacetic Acid or Diprosin or Epilim or Ergenyl or Everiden).ti. or (Dipropylacetate or Dipropylacetatic Acid or Dipropyl Acetic Acid or Dipropylacetic Acid or Diprosin or Epilim or Ergenyl or Everiden).id. or (Dipropylacetate or Dipropylacetatic Acid or Dipropyl Acetic Acid or Dipropylacetic Acid or Diprosin or Epilim or Ergenyl or Everiden).sh.34
22(Goilim or Labazene or Leptilan or Leptilanil or Mylproin or Myproic Acid or Orfiril or Orlept).ab. or (Goilim or Labazene or Leptilan or Leptilanil or Mylproin or Myproic Acid or Orfiril or Orlept).ti. or (Goilim or Labazene or Leptilan or Leptilanil or Mylproin or Myproic Acid or Orfiril or Orlept).id. or (Goilim or Labazene or Leptilan or Leptilanil or Mylproin or Myproic Acid or Orfiril or Orlept).sh.0
23(Propymal or Valerin or Valparin or Valpro or Valproate or Valproate Sodium or Vupral).ab. or (Propymal or Valerin or Valparin or Valpro or Valproate or Valproate Sodium or Vupral).ti. or (Propymal or Valerin or Valparin or Valpro or Valproate or Valproate Sodium or Vupral).id. or (Propymal or Valerin or Valparin or Valpro or Valproate or Valproate Sodium or Vupral).sh.2326
24(Tiagabin or Gabitril).ab. or (Tiagabin or Gabitril).ti. or (Tiagabin or Gabitril).id. or (Tiagabin or Gabitril).sh.4
25(Topiramate or Epitomax or Topamax or Topimax).ab. or (Topiramate or Epitomax or Topamax or Topimax).ti. or (Topiramate or Epitomax or Topamax or Topimax).id. or (Topiramate or Epitomax or Topamax or Topimax).sh.1216
26(Vigabatrin or Sabril).ab. or (Vigabatrin or Sabril).ti. or (Vigabatrin or Sabril).id. or (Vigabatrin or Sabril).sh.245
27fibromyalgia.ab. or fibromyalgia.ti. or fibromyalgia.id. or fibromyalgia.sh.2097
28fibrositis.ab. or fibrositis.ti. or fibrositis.id. or fibrositis.sh.39
29fms.ab. or fms.ti. or fms.id. or fms.sh.599
30randomized controlled trial.ab. or randomized controlled trial.ti. or randomized controlled trial.id. or randomized controlled trial.pt.7360
31controlled clinical trial.ab. or controlled clinical trial.ti. or controlled clinical trial.id. or controlled clinical trial.pt.909
32double blind.ab. or double blind.ti. or single blind.ab. or single blind.ti.18,096
33placebo.ab. or placebo.ti. or random$.ab. or random$.ti.139,111
341 or 2 or 3 or 4 or 5 or 6 or 7 or 8 or 9 or 10 or 11 or 12 or 13 or 14 or 15 or 16 or 17 or 18 or 19 or 20 or 21 or 22 or 23 or 24 or 25 or 2614,709
3527 or 28 or 292366
3630 or 31 or 32 or 33142,181
3734 and 35 and 3626

Appendix 4. Scopus search strategy

1

(TITLE-ABS-KEY(anticonvulsants/) OR TITLE-ABS-KEY(anti convulsant$ OR anticonvulsant$ OR anti-convulsant$ OR anti epileptic$ OR antiepileptic$ OR anti-epileptic$))

59.068

2

(TITLE-ABS-KEY(anticonvulsants) OR TITLE-ABS-KEY(anti convulsant$ OR anticonvulsant$ OR anti-convulsant$ OR anti epileptic$ OR antiepileptic$ OR anti-epileptic$))

59.068

3

(TITLE-ABS-KEY(carbamazepine) OR TITLE-ABS-KEY(carbamazepine OR amizepin OR amizepine OR atretol OR biston OR calepsin OR carbamazepin OR carbategral OR carbatrol OR convuline OR epimax) OR TITLE-ABS-KEY(epitol OR equetro OR finlepsin OR lexin OR mazepine OR neurotol OR neurotop OR servimazepin OR sirtal OR tegral OR tegretal OR tegretol OR tegrital OR telesmin OR teril OR timonil))

51.288

4

(TITLE-ABS-KEY(oxcarbazepine OR apydan OR oxocarbamazepine OR oxocarbazepine OR timox OR trileptal) OR TITLE-ABS-KEY(clonazepam OR clonex OR klonopin OR paxam OR kriadex OR ravotril OR rivatril OR rivotril))

24.960

5

(TITLE-ABS-KEY(felbamat OR felbatol OR taloxa) OR TITLE-ABS-KEY(gabapentin OR neurontin OR neurotonin) OR TITLE-ABS-KEY(levetiracetam OR keppra) OR TITLE-ABS-KEY(pregabalin OR lyrica))

23.281

6

(TITLE-ABS-KEY(phenytoin OR alepsin OR aleviatin OR antilepsin OR antisacer OR cansoin OR citrullamon OR comital OR danten OR dantoin OR denyl OR difetoin OR differenin OR difhydan OR di hydan OR dihydan OR dilantin OR dintoin OR dintoina OR diphantoin OR diphantoine) OR TITLE-ABS-KEY(fenantoin OR fenitoin OR fenytoin OR fenytoine OR hidantal OR hydantin OR hydantinal OR hydantoinal OR idantoin OR lepitoin OR minetoin OR neosidantoina OR phenhydan OR phenhydane OR phenybin OR phenydan OR phenydantin OR phenytoine OR phenytoin OR phenyto))

54.875

7

(TITLE-ABS-KEY(lacosamide OR erlosamide OR vimpat) OR TITLE-ABS-KEY(lamotrigine OR labileno OR lamictal))

16.366

8

(TITLE-ABS-KEY(valproic acid) OR TITLE-ABS-KEY(sodium valproate OR alpha propylvalerate OR alpha propylvaleric acid OR apilepsin OR convulex OR depacon OR depakene OR depakin OR depakine OR deprakine OR dipropylacetate OR dipropylacetatic acid OR dipropyl acetic acid OR dipropylacetic acid OR diprosin) OR TITLE-ABS-KEY(epilim OR ergenyl OR everiden OR goilim OR labazene OR leptilan OR leptilanil OR mylproin OR myproic acid OR orfiril OR orlept OR propymal OR valerin OR valparin OR valpro OR valproate OR valproate sodium OR vupral))

43.266

9

(TITLE-ABS-KEY(tiagabin OR gabitril) OR TITLE-ABS-KEY(topiramate OR epitomax OR topamax OR topimax) OR TITLE-ABS-KEY(vigabatrin OR sabril))

17.274

10

(TITLE-ABS-KEY(fibromyalgia) OR TITLE-ABS-KEY(fibrositis) OR TITLE-ABS-KEY(fms))

22.441

11

(TITLE-ABS-KEY(randomized controlled trial) OR TITLE-ABS-KEY(controlled trial) OR TITLE-ABS-KEY(placebo) OR TITLE-ABS-KEY(single blind) OR TITLE-ABS-KEY(double blind))

901.308

12

#1 OR #2 OR #3 OR #4 OR #5 OR #6 OR #7 OR #8 OR #9

((TITLE-ABS-KEY(anticonvulsants/) OR TITLE-ABS-KEY(anti convulsant$ OR anticonvulsant$ OR anti-convulsant$ OR anti epileptic$ OR antiepileptic$ OR anti-epileptic$))) OR ((TITLE-ABS-KEY(anticonvulsants) OR TITLE-ABS-KEY(anti convulsant$ OR anticonvulsant$ OR anti-convulsant$ OR anti epileptic$ OR antiepileptic$ OR anti-epileptic$))) OR ((TITLE-ABS-KEY(carbamazepine) OR TITLE-ABS-KEY(carbamazepine OR amizepin OR amizepine OR atretol OR biston OR calepsin OR carbamazepin OR carbategral OR carbatrol OR convuline OR epimax) OR TITLE-ABS-KEY(epitol OR equetro OR finlepsin OR lexin OR mazepine OR neurotol OR neurotop OR servimazepin OR sirtal OR tegral OR tegretal OR tegretol OR tegrital OR telesmin OR teril OR timonil))) OR ((TITLE-ABS-KEY(oxcarbazepine OR apydan OR oxocarbamazepine OR oxocarbazepine OR timox OR trileptal) OR TITLE-ABS-KEY(clonazepam OR clonex OR klonopin OR paxam OR kriadex OR ravotril OR rivatril OR rivotril))) OR ((TITLE-ABS-KEY(felbamat OR felbatol OR taloxa) OR TITLE-ABS-KEY(gabapentin OR neurontin OR neurotonin) OR TITLE-ABS-KEY(levetiracetam OR keppra) OR TITLE-ABS-KEY(pregabalin OR lyrica))) OR ((TITLE-ABS-KEY(phenytoin OR alepsin OR aleviatin OR antilepsin OR antisacer OR cansoin OR citrullamon OR comital OR danten OR dantoin OR denyl OR difetoin OR differenin OR difhydan OR di hydan OR dihydan OR dilantin OR dintoin OR dintoina OR diphantoin OR diphantoine) OR TITLE-ABS-KEY(fenantoin OR fenitoin OR fenytoin OR fenytoine OR hidantal OR hydantin OR hydantinal OR hydantoinal OR idantoin OR lepitoin OR minetoin OR neosidantoina OR phenhydan OR phenhydane OR phenybin OR phenydan OR phenydantin OR phenytoine OR phenytoin OR phenyto))) OR ((TITLE-ABS-KEY(lacosamide OR erlosamide OR vimpat) OR TITLE-ABS-KEY(lamotrigine OR labileno OR lamictal))) OR ((TITLE-ABS-KEY(valproic acid) OR TITLE-ABS-KEY(sodium valproate OR alpha propylvalerate OR alpha propylvaleric acid OR apilepsin OR convulex OR depacon OR depakene OR depakin OR depakine OR deprakine OR dipropylacetate OR dipropylacetatic acid OR dipropyl acetic acid OR dipropylacetic acid OR diprosin) OR TITLE-ABS-KEY(epilim OR ergenyl OR everiden OR goilim OR labazene OR leptilan OR leptilanil OR mylproin OR myproic acid OR orfiril OR orlept OR propymal OR valerin OR valparin OR valpro OR valproate OR valproate sodium OR vupral))) OR ((TITLE-ABS-KEY(tiagabin OR gabitril) OR TITLE-ABS-KEY(topiramate OR epitomax OR topamax OR topimax) OR TITLE-ABS-KEY(vigabatrin OR sabril)))

164.432

13

(#1 OR #2 OR #3 OR #4 OR #5 OR #6 OR #7 OR #8 OR #9) AND #10 AND #11

((TITLE-ABS-KEY(fibromyalgia) OR TITLE-ABS-KEY(fibrositis) OR TITLE-ABS-KEY(fms))) AND ((TITLE-ABS-KEY(randomized controlled trial) OR TITLE-ABS-KEY(controlled trial) OR TITLE-ABS-KEY(placebo) OR TITLE-ABS-KEY(single blind) OR TITLE-ABS-KEY(double blind))) AND (((TITLE-ABS-KEY(anticonvulsants/) OR TITLE-ABS-KEY(anti convulsant$ OR anticonvulsant$ OR anti-convulsant$ OR anti epileptic$ OR antiepileptic$ OR anti-epileptic$))) OR ((TITLE-ABS-KEY(anticonvulsants) OR TITLE-ABS-KEY(anti convulsant$ OR anticonvulsant$ OR anti-convulsant$ OR anti epileptic$ OR antiepileptic$ OR anti-epileptic$))) OR ((TITLE-ABS-KEY(carbamazepine) OR TITLE-ABS-KEY(carbamazepine OR amizepin OR amizepine OR atretol OR biston OR calepsin OR carbamazepin OR carbategral OR carbatrol OR convuline OR epimax) OR TITLE-ABS-KEY(epitol OR equetro OR finlepsin OR lexin OR mazepine OR neurotol OR neurotop OR servimazepin OR sirtal OR tegral OR tegretal OR tegretol OR tegrital OR telesmin OR teril OR timonil))) OR ((TITLE-ABS-KEY(oxcarbazepine OR apydan OR oxocarbamazepine OR oxocarbazepine OR timox OR trileptal) OR TITLE-ABS-KEY(clonazepam OR clonex OR klonopin OR paxam OR kriadex OR ravotril OR rivatril OR rivotril))) OR ((TITLE-ABS-KEY(felbamat OR felbatol OR taloxa) OR TITLE-ABS-KEY(gabapentin OR neurontin OR neurotonin) OR TITLE-ABS-KEY(levetiracetam OR keppra) OR TITLE-ABS-KEY(pregabalin OR lyrica))) OR ((TITLE-ABS-KEY(phenytoin OR alepsin OR aleviatin OR antilepsin OR antisacer OR cansoin OR citrullamon OR comital OR danten OR dantoin OR denyl OR difetoin OR differenin OR difhydan OR di hydan OR dihydan OR dilantin OR dintoin OR dintoina OR diphantoin OR diphantoine) OR TITLE-ABS-KEY(fenantoin OR fenitoin OR fenytoin OR fenytoine OR hidantal OR hydantin OR hydantinal OR hydantoinal OR idantoin OR lepitoin OR minetoin OR neosidantoina OR phenhydan OR phenhydane OR phenybin OR phenydan OR phenydantin OR phenytoine OR phenytoin OR phenyto))) OR ((TITLE-ABS-KEY(lacosamide OR erlosamide OR vimpat) OR TITLE-ABS-KEY(lamotrigine OR labileno OR lamictal))) OR ((TITLE-ABS-KEY(valproic acid) OR TITLE-ABS-KEY(sodium valproate OR alpha propylvalerate OR alpha propylvaleric acid OR apilepsin OR convulex OR depacon OR depakene OR depakin OR depakine OR deprakine OR dipropylacetate OR dipropylacetatic acid OR dipropyl acetic acid OR dipropylacetic acid OR diprosin) OR TITLE-ABS-KEY(epilim OR ergenyl OR everiden OR goilim OR labazene OR leptilan OR leptilanil OR mylproin OR myproic acid OR orfiril OR orlept OR propymal OR valerin OR valparin OR valpro OR valproate OR valproate sodium OR vupral))) OR ((TITLE-ABS-KEY(tiagabin OR gabitril) OR TITLE-ABS-KEY(topiramate OR epitomax OR topamax OR topimax) OR TITLE-ABS-KEY(vigabatrin OR sabril))))

375

Appendix 5. clinicaltrials.gov search strategy

anticonvulsant and fibromyalgia: 33

fibromyalgia and antiepileptic drugs: 33

What's new

Last assessed as up-to-date: 30 August 2013.

DateEventDescription
23 October 2012Amended

The protocol 'Antidepressants and centrally active agents for fibromyalgia syndrome' published in 2006 (Nishishinya 2006) has been split into several systematic reviews that will be/have been published as:

- Anticonvulsants for fibromyalgia
- Monoamine oxidase inhibitors (MAOIs) for fibromyalgia syndrome
- Non-steroidal anti-inflammatory drugs (NSAIDs), analgesics and opioid agents for fibromyalgia
- Sedatives and hypnotic agents for fibromyalgia
- Selective serotonin reuptake inhibitors (SSRIs) for fibromyalgia
- Serotonin and noradrenaline reuptake inhibitors (SNRIs) for fibromyalgia syndrome
- Tricyclic agents for fibromyalgia

Contributions of authors

NÜ: systematic review study selection, data extraction, data analysis, interpretation of findings; review final proof.

CS: data extraction, data analysis, interpretation of findings; review final proof.

BW: protocol review and revision; interpretation of findings; review final proof.

WH: topic conception and methodological aspects; protocol revision; systematic review study selection, data extraction, data analysis, interpretation of findings and author of review.

Declarations of interest

NÜ received consultancy honoraria from Grünenthal GmbH and travel grants from Pfizer, Astellas, Grünenthal GmbH and CSL Behring. CS received honoraria for educational lectures from Eli-Lilly and Pfizer. BW received a consulting fee from Jazz Pharmaceuticals and was a site investigator for a milnacipran trial in FM. WH received honoraria for educational lectures from Eli-Lilly, Janssen-Cilag, Mundipharma and Pfizer and a travel grant from Eli-Lilly. He serves on the advisory board of Daiichi Sankyo.

Differences between protocol and review

The protocol 'Antidepressants and centrally active agents for fibromyalgia syndrome' (Nishishinya 2006), has been split into several systematic reviews. A random-effects model was used for all analyses irrespective of the amount of heterogeneity. The intended subgroup analyses with gender and pain were not conducted because individual participant data were not available. The intended sensitivity analyses (different statistical models applied, presence of temporal differences, diagnostic criteria used in the trial, according to the presence/absence of any mental disorder, according to the presence/absence of any concomitant systemic disease) were not conducted because the studies did not differ in these characteristics. The GRADE approach was used for the grading of evidence.

Characteristics of studies

Characteristics of included studies [ordered by study ID]

Arnold 2007

Methods

Study setting: Multicentre study with 3 outpatient research centres in the USA

Study design: Parallel

Duration therapy: 12 weeks

Follow-up: Not performed

Analysis: ITT; LOCF; for the primary analysis of continuous variables collected at more than 2 time points, a longitudinal analysis that compared the rate of change of the outcome during the treatment period between groups was used. The difference in rate of change was estimated by random regression methods, as described elsewhere. A model for the mean of the outcome variable that included terms for treatment, time, treatment-by-time interaction and centre was used. Time was modelled as a continuous variable. To account for the correlation of observations among participants, the SAS procedure MIXED (SAS Institute, Cary, NC) with the best fitting of the following covariance structures was used: unstructured, first-order heterogeneous autoregressive and first-order autoregressive. The longitudinal analyses used all available observations from all time points from all participants who completed a baseline evaluation. As a secondary analysis, changes from baseline to end point (the LOCF method) were analysed using an analysis of variance model, with a term for centre

Participants

Participants: 150 (90% women, 97% white, mean age 48 years)

Inclusion criteria: ACR 1990 criteria; age ≥ 18 years; participants were required to score 4 on the mean pain severity item of the BPI

Exclusion criteria: Pain from traumatic injury or structural or regional rheumatic disease; rheumatoid arthritis, inflammatory arthritis or autoimmune disease; unstable medical or psychiatric illness; lifetime history of psychosis, hypomania or mania, epilepsy or dementia; substance abuse in the last 6 months; serious risk of suicide; pregnancy or breastfeeding; unacceptable contraception in those of childbearing potential; participants who, in the opinion of the investigator, were treatment refractory; prior treatment with gabapentin or pregabalin; and treatment with an investigational drug within 30 days of screening. Concomitant medication exclusions consisted of medications or herbal agents with CNS effects, with the exception of episodic use of sedating antihistamines (antidepressants required a 14-day washout period prior to beginning study medication except for fluoxetine, which required a 30-day washout period); analgesics, with the exception of paracetamol (acetaminophen) or over-the-counter NSAIDs; and unconventional or alternative therapies

Interventions

Active drug: Gabapentin flexible 1200-2400 mg/d (75 participants): 300 mg once a day at bedtime for 1 week, 300 mg twice a day for 1 week, 300 mg twice a day and 600 mg once a day at bedtime for 2 weeks, 600 mg 3 times a day for 2 weeks, and 600 mg twice a day and 1200 mg once a day at bedtime (2400 mg/d) for the remainder of the study beginning at week 6. If a participant could not tolerate 2400 mg/d, the dosage was reduced to a minimum of 1200 mg/d, administered 3 times a day. The study medication dose was stable for at least the last 4 weeks of the therapy phase. During the tapering phase, the dosage was decreased by 300 mg/d until discontinuation

Placebo: 75 participants

Rescue or allowed medication, or both: Paracetamol (acetaminophen) or over-the-counter NSAIDs (dosage not reported)

Outcomes

Pain: BPI mean pain severity (NRS 0-10). LOCF analysis reported; 50% pain reduction rates not reported and calculated by imputation method

Fatigue: FIQ (VAS 0-10): Not reported

Sleep: BPI sleep interference (NRS 0-10). Observed cases reported

Depression: Montgomery Asberg Depression Rating Scale score (NRS 0-60). Observed cases reported

Anxiety: FIQ (VAS 0-10): Not reported

Disability: BPI interference from pain (NRS 0-10). Observed cases reported

Quality of life: FIQ total score (0-80). Observed cases reported

Patient-perceived improvement: PGIC: Data extracted from figure

AEs: At the randomisation visit, and at each subsequent visit until the end of the therapy phase, AEs were reviewed (no details reported)

Notes

Safety: Gabapentin-treated participants reported dizziness, sedation, lightheadedness and weight gain significantly more frequently than did placebo-treated participants. Most treatment-emergent AEs were mild to moderate in severity. There were no clinically important findings in the laboratory results, physical examinations, or ECGs

Funding sources and any declaration of interest of primary investigators: Supported by NIH grant N01-AR-2-2264 from the National Institute of Arthritis and Musculoskeletal and Skin Diseases (Dr. Arnold, principal investigator). Dr. Arnold received consulting fees from Eli Lilly (more than USD10,000) and from Pfizer, Cypress Bioscience, Wyeth Pharmaceuticals, Sanofi-Aventis, Böehringer Ingelheim, Sepracor, Forest Laboratories, Allergan, and Vivus (< USD10,000 each). She also has received research support from Eli Lilly, Pfizer, Cypress Bioscience, Wyeth Pharmaceuticals, Sanofi-Aventis and Böehringer Ingelheim. Dr. Keck received consulting fees (< USD10,000) from, or is a member of the scientific advisory boards of, Abbott, AstraZeneca Pharmaceuticals, Bristol-Myers Squibb, GlaxoSmithKline, Eli Lilly and Pfizer. He is a principal or co-investigator on research studies sponsored by Abbott, the American Diabetes Association, AstraZeneca Pharmaceuticals, Bristol-Myers Squibb, GlaxoSmithKline, Eli Lilly, Janssen Pharmaceutica, the National Institute of Mental Health, the National Institute of Drug Abuse, Pfizer, the Stanley Medical Research Institute and UCB

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskComputer-generated random sequence stratified by major depression status (details provided on request)
Allocation concealment (selection bias)Low riskCentral independent unit (details reported on request)
Blinding of participants and personnel (performance bias)
All outcomes
Low riskDouble blind (number and appearance of placebo capsules similar)
Blinding of outcome assessment (detection bias)
All outcomes
Low riskIndependent data imputation and statistical analysis (details reported on request)
Incomplete outcome data (attrition bias)
All outcomes
High riskObserved cases analysis
Selective reporting (reporting bias)High riskOutcomes anxiety and fatigue not reported

Arnold 2008

Methods

Study setting: Multicentre study with 84 outpatient research centres in the USA

Study design: Parallel

Duration therapy: 14 weeks

Follow-up: Not performed

Analysis: ITT; LOCF; analysis of covariance with treatment centre and baseline scores as covariates

Participants

Participants: 745 (95% women, 91% white, mean age 50 years)

Inclusion criteria: ACR 1990 criteria; age ≥ 18 years; pain score of at least 40 on a 100 VAS

Exclusion criteria: Inflammatory rheumatic diseases; active infections; untreated endocrine disorder; severe painful disorder (e.g. painful diabetic peripheral neuropathy, postherpetic pain) that might confound the assessment of pain due to FM; unstable medical or psychiatric disorder (e.g. serious hepatic, respiratory, neurological, haematological or immunological illness, unstable cardiovascular disease; or any other severe acute or chronic medical or psychiatric condition or laboratory abnormalities, including creatinine clearance < 50 mL/min; history of illicit alcohol or drug abuse within the past 2 years; pending workers's compensation, current receipt of disability; past or pending litigation for monetary compensation related to FM; other concomitant medication for FM (antidepressants, anticonvulsants) as well as agents used to treat pain or insomnia

Interventions

Active drug: Pregabalin 300 mg/d (183 participants), pregabalin 450 mg/d (190 participants), pregabalin 600 mg/d (188 participants) twice/d, 2 weeks' dose escalation

Placebo: 184 participants

Rescue or allowed medication: Paracetamol (acetaminophen) < 4 g/d, aspirin < 325 mg/d

Outcomes

Pain: Daily diary mean pain (NRS 0-10)

Fatigue: MAF (NRS 1-50)

Sleep: MOS Sleep Problems Index (NRS 0-100)

Depression: HADS (NRS 0-21)

Anxiety: HADS (NRS 0-21)

Disability: SF-36, physical functioning (NRS 50-0)

Quality of life: FIQ total score (0-80)

Patient-perceived improvement: PGIC (1-7)

AEs: Observed or spontaneously reported AEs; laboratory results, physical examinations, ECG

Notes

Safety: Of the 745 participants who received study medication, 614 (82%) had at least 1 AE and 473 (53%) had at least 1 AE that was judged to be related with treatment. The incidence of all-causality AEs was higher with pregabalin (72%) than with placebo (38%). The severity of the majority of treatment-related AEs was mild to moderate (89% pregabalin vs. 98% placebo)

Dizziness and somnolence were the most common AEs in pregabalin groups leading to discontinuation of treatment. There were no clinically important findings in the laboratory results, physical examinations or ECGs

Funding sources and any declaration of interest of primary investigators: Dr. Arnold received consulting fees from Eli Lilly, Pfizer, Cypress Biosciences, Wyeth Pharmaceuticals, Sanofi-Aventis, Böehringer Ingelheim, Allergan, Forest Laboratories and Vivus. Dr. Arnold received research grant support from Eli Lilly, Pfizer, Cypress Biosciences, Wyeth Pharmaceuticals, Sanofi-Aventis, Böehringer Ingelheim, Allergan and Forest Laboratories. Dr. Arnold is on the Speakers Bureau for Eli Lilly and Pfizer. Dr. Russell has consulted for conducted research studies for Pfizer, Autoimmune Technologies, LLC, LKB World, Jazz Pharmaceuticals, Gruenethal and Allergan. He is on speaker's panel for Merck, Ortho-McNeil and Pfizer. Erdal Diri received research grants from Hoffmann-LaRoche, Pfizer, Pain Therapeutics, Proctor & Gamble Pharmaceutics and Corona. He is a speaker and consultant for Pfizer, Amgen, Centecor and Abbott. Rachel Duan, James Youg, Susan Martin, Jeannette Barreite and George Haig are employees of Pfizer and own Pfizer stocks. Uma Sharma is a consultant for Pfizer, Wyeth, Eisal, Analgesic Research and Amgen. Editorial support was provided by Yilmarie Yanchik, an employee of Pfizer. Statistical support was provided by Ed Whalen, an employee of Pfizer

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskComputer-generated random sequence stratified by major depression status (details provided on request)
Allocation concealment (selection bias)Low riskCentral independent unit (details reported on request)
Blinding of participants and personnel (performance bias)
All outcomes
Low riskDouble blind (number and appearance of placebo capsules similar)
Blinding of outcome assessment (detection bias)
All outcomes
Low riskIndependent data imputation and statistical analysis (details reported on request)
Incomplete outcome data (attrition bias)
All outcomes
Low riskImputation using LOCF and ITT analysis for efficacy data
Selective reporting (reporting bias)Low riskAll outcomes reported

Crofford 2005

Methods

Study setting: Multicentre study with 40 outpatient research centres in the USA

Study design: Parallel

Duration therapy: 8 weeks

Follow-up: Not performed

Analysis: ITT; LOCF; analysis of covariance, with treatment and centre as the main effects and the baseline value as the covariate

Participants

Participants: 529 (92% women, 93% white, mean age 49 years)

Inclusion criteria: ACR 1990 criteria; age ≥ 18 years; score of 40 mm on the 100-mm VAS of the SF-MPQ

Exclusion criteria:

Inflammatory rheumatic disease or other severe painful disorders that might confound assessment of FM pain. People were also excluded if they had clinically significant or unstable medical or psychological conditions that, in the opinion of the investigator, would compromise participation in the study. Participants with a calculated creatinine clearance rate of 60 mL/min (Cockroft-Gault equation) were specifically excluded. Those who had failed to respond to previous treatment with gabapentin at dosages of 1200 mg/d for pain associated with FM were excluded. Women who were not postmenopausal were tested to confirm that they were not pregnant or breastfeeding during the study, and all women of childbearing potential were advised to use contraception reliably. Participants who were receiving disability, applying for disability, or engaged in litigation related to FM were excluded

Interventions

Active drug: Pregabalin 150 mg/d (132 participants), pregabalin 300 mg/d (134 participants), pregabalin 450 mg/d (132 participants) 3 times/d

Placebo: 131 participants

Rescue or allowed medication: Paracetamol (acetaminophen) < 4 g/d, aspirin < 325 mg/d

Outcomes

Pain: SF-MPQ (VAS 0-100)

Fatigue: MAF (NRS 1-50)

Sleep: MOS Sleep Problems Index (NRS 0-100)

Depression: HADS (NRS 0-21): details provided on request

Anxiety: HADS (NRS 0-21); details provided on request

Disability: SF-36, physical functioning (NRS 50-0)

Quality of life: Not assessed

Patient-perceived improvement: PGIC: not reported

AEs: All spontaneously reported or observed treatment-emergent AEs were recorded at each clinic visit, along with the dates on which they began and ended. The sponsor classified AEs using Coding Symbols for a Thesaurus of Adverse Reaction Terms, 4th Edition (COSTART IV) . A 12-lead ECG was recorded at screening and end point, and clinical laboratory tests (haematology, urinalysis and chemistry) were performed at screening, week 3, and end point. Serious AE not reported

Notes

Safety: Most AEs were mild or moderate. Dizziness and somnolence were the 2 most frequently reported AEs and tended to be dose-related across pregabalin groups. Non-CNS AEs that were more frequent in the pregabalin groups included weight gain and peripheral oedema. There were no clinically important findings in the analyses of haematology, blood chemistry or urinalysis. Similarly, there were no clinically significant findings in the visual function, physical, or neurological examinations or on the ECGs

Funding sources and any declaration of interest of primary investigators: Supported by Pfizer Global Research and Development, Ann
Arbor, Michigan. Dr. Crofford has received consulting fees of < USD10,000 from Cypress Bioscience, Eli Lilly & Co, Orphan Pharmaceuticals, Pfizer and Wyeth. Dr. Rowbotham has received consulting fees of < USD10,000 from Eli Lilly & Co and Xenoport, owns stock in Xenoport and Neuromolecular and was a co-investigator on a study of gabapentin funded by Pfizer. Dr. Mease has received consulting fees of < USD10,000 from Pfizer, Cypress Bioscience, Eli Lilly & Co and Pierre Fabre and owns stock in Cypress Bioscience. Dr. Dworkin has served on the advisory board or as a consultant for fees of < USD10,000 for Abbott Laboratories, Alpharma, AstraZeneca Pharmaceuticals, Bristol-Myers Squibb, Elan Pharmaceuticals, Eli Lilly & Co, GlaxoSmithKline, Johnson & Johnson, Merck KGaA, NeurogesX Inc., Ortho-McNeil Pharmaceutical and UCB Pharma; has received consulting fees of > USD10,000 from Pfizer, Allergan, Novartis, Epicept and Endo; and owns stock in NeurogesX, Inc. Ms Corbin, Mr. Young, Ms LaMoreaux, Ms Martin and Dr. Sharma own stock in Pfizer

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskComputer-generated random sequence stratified by major depression status (details provided on request)
Allocation concealment (selection bias)Low riskCentral independent unit (details reported on request)
Blinding of participants and personnel (performance bias)
All outcomes
Low riskDouble blind (number and appearance of placebo capsules similar, details reported on request)
Blinding of outcome assessment (detection bias)
All outcomes
Low riskIndependent data imputation and statistical analysis (details reported on request)
Incomplete outcome data (attrition bias)
All outcomes
Low riskImputation using LOCF and ITT analysis for efficacy data
Selective reporting (reporting bias)High riskQuality of life scores and SAEs not reported and not provided on request

Mease 2008

Methods

Study setting: Multicentre study with 79 outpatient research centres in the USA

Study design: Parallel

Duration therapy: 13 weeks

Follow-up: Not performed

Analysis: ITT; LOCF; analysis of covariance with treatment centre and baseline scores as covariates

Participants

Participants: 558 (94% women, 91% white, mean age 49 years)

Inclusion criteria: ACR 1990 criteria; age ≥ 18 years; pain score of at least 4 on a 0-10 NRS

Exclusion criteria: Inflammatory or rheumatological disease; other severe pain disorders; clinically significant or unstable medical or psychological conditions; creatinine clearance < 60 mL/min, severe depression, receiving or applying for disability benefits

Interventions

Active drug: pregabalin 300 mg/d (185 participants), pregabalin 450 mg/d (183 participants), pregabalin 600 mg/d (190 participants) twice/d, 2 weeks' dose escalation

Placebo: 190 participants

Rescue or allowed medication: paracetamol (acetaminophen) < 4 g/d

Outcomes

Pain: Daily diary mean pain (NRS 0-10); 50% pain reduction rates provided on request

Fatigue: MAF (NRS 1-50), details provided on request

Sleep: MOS Sleep Problems Index (NRS 0-100)

Depression: HADS (NRS 0-21): details provided on request

Anxiety: HADS (NRS 0-21): details provided on request

Disability: SF-36, physical functioning (NRS 50-0): data provided on request

Quality of life: FIQ total score (0-100); standard deviations provided on request

Patient-perceived improvement: PGIC (1-7)

AEs: Volunteered by participants or observed by the clinician at every visit, laboratory results, physical examinations, ECG

Notes

Safety: At least 1 AE was reported by 89.2% of participants with pregabalin 300 mg/d, 91.8% of participants with pregabalin 450 mg/d, 93.7% of participants with pregabalin 600 mg/d and by 76.3% of participants with placebo

Dizziness, somnolence and weight gain were more common in pregabalin groups compared with placebo. There were no clinically important findings in the laboratory results, physical examinations, or ECGs

Funding sources and any declaration of interest of primary investigators: Supported by Pfizer Global Research and Development, Ann
Arbor, Michigan. Dr. Mease has received research grant support from Pfizer, Cypress Bioscience, Forest Laboratories, Eli Lilly, Allergan, Wyeth Pharmaceuticals, Jazz Pharmaceuticals and Fralex Therapeutics. Dr. Russell has received research grant support from Pfizer, Eli Lilly, Allergan, Böehringer Ingelheim, Cypress Bioscience, Jazz Pharmaceuticals, Grünenthal and Pierre Fabre. Dr. Arnold has received research grant support from Eli Lilly, Pfizer, Cypress Biosciences, Wyeth Pharmaceuticals, Sanofi-Aventis, Böehringer Ingelheim, Allergan and Forest Laboratories. Dr. Florian, Mr. Young and Ms Mareen are employees of Pfizer and own stock of Pfizer. Dr Sharma was an employee of Pfizer when the study was conducted and owns stocks of Pfizer

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskComputer-generated random sequence (details provided on request)
Allocation concealment (selection bias)Low riskCentral independent unit (details reported on request)
Blinding of participants and personnel (performance bias)
All outcomes
Low riskDouble blind (number and appearance of placebo capsules similar)
Blinding of outcome assessment (detection bias)
All outcomes
Low riskIndependent data imputation and statistical analysis (details reported on request)
Incomplete outcome data (attrition bias)
All outcomes
Low riskImputation using LOCF and ITT analysis for efficacy data
Selective reporting (reporting bias)High riskDisability not reported

Ohta 2012

Methods

Study setting: Multicentre study with 44 outpatient research centres in Japan

Study design: Parallel

Duration therapy: 14 weeks

Follow-up: Not performed

Analysis: ITT; LOCF; a mixed-effect model taking baseline value as covariate was used for the analysis, which included participants as the random effect and dose groups, points at time of evaluation, and interaction between a dose group and its point at time of evaluation as the fixed effects

Participants

Participants: 501 (89% women, 10% Japanese, mean age 47 years)

Inclusion criteria: ACR 1990 criteria; age ≥ 18 years; pain score of ≥ 40 mm on the 100-mm VAS at visit 2, and had assessed and documented their pain score on at least 4
of the past 7 days prior to visit 2 while recording a mean pain score of ≥ 4 on the 11-point NRS

Exclusion criteria: Decrease of ≥ 30% on their pain VAS during the placebo run-in period (at visit 2 compared with visit 1), in order to remove potential placebo-responders. Participants were also excluded if they were being treated for depression or if they were at risk of suicide or self harm in the opinion of the study investigator; inflammatory or rheumatological disease; other severe pain disorders; clinical significant or unstable medical or psychological conditions; history of malignancy; creatinine clearance < 60 mL/min, severe depression, receiving or applying for disability benefits

Interventions

Active drug: Pregabalin (251 participants): treatment was started at 150 mg/d, escalated to 300 mg/d 1 week later, and to 450 mg/d after another week. The dose was adjusted (increased or decreased) until visit 5 of the study, after which the maintenance dose was either 300 or 450 mg/d

Placebo: 250 participants

Rescue or allowed medication: Paracetamol (acetaminophen) or NSAIDs for additional pain relief, although for NSAIDs, the participant must have already been on a stable regimen for longer than 30 days

Outcomes

Pain: Daily diary mean pain (NRS 0-10)

Fatigue: FIQ Fatigue single scale (VAS 0-10)

Sleep: MOS Sleep Problems Index (NRS 100-0)

Depression: HADS (NRS 0-21)

Anxiety: HADS (NRS 0-21)

Disability: SF-36, physical functioning (NRS 50-0)

Quality of life: FIQ total score (0-100)

Patient-perceived improvement: PGIC (1-7)

AEs: volunteered by participants or observed by the clinician at every visit, laboratory results, physical examinations, ECG

Notes

Safety: The incidence of all-causality AEs was higher with pregabalin (occurring in 225 of 250 participants, 90.0%) than with placebo (175 of 248 participants, 70.6%). Similarly, the incidence of treatment-related AEs was higher with pregabalin (206 of 250 participants, 82.4%) than with placebo (128 of 248 participants, 51.6%). The most common AEs in this study were somnolence, dizziness, nasopharyngitis, increased weight and constipation with pregabalin treatment, and somnolence and nasopharyngitis with placebo.

A laboratory test result of increased creatine kinase was more frequent with pregabalin (7 of 250 participants, 2.8%) than with placebo (1 of 248 participants, 0.4%), although all cases were of mild severity. Increased weight was reported more frequently with pregabalin (39 of 250 participants, 15.6% (38 mild, 1 moderate)) than with placebo (9 of 248 participants, 3.6% (8 mild, 1 moderate)). There were no clinically significant changes in blood pressure or pulse rate in the pregabalin and placebo group.
There were 4 serious AEs in 4 (0.8%) participants who had received ≥ 1 dose of treatment; 1 participant was from the placebo group (abnormal liver function test result) and 3 were from the pregabalin group (breast cancer, viral gastroenteritis and musculoskeletal stiffness). Severe AEs were observed in 2 participants from the pregabalin group (breast cancer and loss of consciousness, each in 1 participant) and all other AEs were mild or moderate. As assessed by the study investigators, there was no causal relationship to the study drug for any serious or severe AEs.
Suicidal ideation (mild), as rated by C-SSRS, was noted in 2 participants in the pregabalin group. The investigator (a physician specialising in psychosomatic medicine) had noted suicidal ideation in each of these people prior to the start of the study. In each case, incidence was attributable to family environment and was judged to have no causal relationship to the study drug. The suicidal ideation in these cases was not judged by the study investigator to be a real desire, and treatment with the study drug was continued.

Funding sources and any declaration of interest of primary investigators: The study was funded by Pfizer Japan Inc. Medical writing support was provided by Joshua Fink PhD, of UBC Scientific Solutions, and funded by Pfizer Inc. HOhta, MO and MS are employees of Pfizer Japan, Inc. KN and HOka received a consultancy fee from Pfizer Japan, Inc. for their participation in this study. CU declares no competing interests. KN, HOka and CU were not compensated for their work on the manuscript.

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskComputer-generated random sequence (details provided on request)
Allocation concealment (selection bias)Low riskCentral independent unit (details reported on request)
Blinding of participants and personnel (performance bias)
All outcomes
Low riskDouble blind (number and appearance of placebo capsules similar)
Blinding of outcome assessment (detection bias)
All outcomes
Low riskIndependent data imputation and statistical analysis (details reported on request)
Incomplete outcome data (attrition bias)
All outcomes
Low riskImputation using LOCF and ITT analysis for efficacy data
Selective reporting (reporting bias)Low riskAll outcomes reported on request

Pauer 2011

Methods

Study setting: Multicentre study with 73 outpatient research centres in the USA, Middle America, South America, Western Europe, Asia, Australia and India

Study design: Parallel

Duration therapy: 14 weeks

Follow-up: Not performed

Analysis: ITT; LOCF; analysis of covariance with treatment centre, week, treatment by week interaction and baseline scores as covariates

Participants

Participants: 986 (91% women, 76% white, mean age 49 years)

Inclusion criteria: ACR 1990 criteria; age ≥ 18 years; pain score of at least 40 on a 100 VAS

Exclusion criteria: Participants who demonstrated a high placebo response (≥ 30% decrease on the VAS following the 1-week run-in period compared with screening). Provided on request: other severe pain conditions; any widespread inflammatory musculoskeletal disorder, active infections or untreated endocrine disorders; previous participation in a clinical trial with pregabalin, previous exposure to pregabalin or currently pregabalin for any condition; severe depression according to the judgement of the investigator; serious internal diseases or any other acute or chronic medical or psychiatric condition or laboratory abnormality that may increase the risk associated with trial participation; intake of any experimental drug within 30 days prior to screening; use of prohibited pain/sleep medication (including antidepressants, sedatives, hypnotics, NSAIDs, opioids, muscle relaxants) in the absence of appropriate washout period; pending disability claims or currently receiving monetary compensation pertinent to the patient's FM or co-morbid diseases

Interventions

Active drug: Pregabalin 300 mg/d (184 participants), pregabalin 450 mg/d (182 participants), pregabalin 600 mg/d (186 participants) twice/d, 2 weeks' dose escalation

Placebo: 184 participants

Rescue or allowed medication: No details reported

Outcomes

Pain: Daily diary mean pain (NRS 0-10)

Fatigue: MAF (NRS 1-50)

Sleep: MOS Sleep Problems Index (NRS 0-100)

Depression: HADS (NRS 0-21): data available on clinicaltrials.gov/ct2/show/NCT00333866

Anxiety: HADS (NRS 0-21): data available on clinicaltrials.gov/ct2/show/NCT00333866

Disability: SF-36, physical functioning (NRS 50-0): data available on clinicaltrials.gov/ct2/show/NCT00333866

Quality of life: FIQ total score (0-100)

Patient-perceived improvement: PGIC (1-7)

AEs: AE were recorded at each visit. Investigators rated the severity of each AE and its relationship to study drug. Clinical laboratory evaluations, physical examinations, abbreviated neurological examinations, and 12-lead ECGs were performed at regular intervals

Notes

Safety: Of the 736 participants receiving study medication, 626 (85%) experienced at least 1 AE. The occurrence of AE increased with dosage (73%, 85%, 90% and 92% for placebo, 300, 450 and 600 mg/d pregabalin-treated participants, respectively).The AEs most frequently reported by pregabalin-treated participants were dizziness, somnolence, weight gain, headache, peripheral oedema, fatigue and dry mouth. Most AE were rated by investigators as mild or moderate. 18 participants experienced SAE, 4 participants treated with placebo and 14 pregabalin. Only 1 SAE, an incidence of chest pain in 1 person in the 450 mg/d pregabalin group, was considered by the investigator to be related to treatment and the person was withdrawn from the study. 8 other participants experienced an SAE that led to withdrawal from the study and 1 participant experienced an SAE that led to a dose reduction

There were no clinically relevant differences in clinical laboratory evaluations, vital signs, physical examination or ECG findings

Funding sources and any declaration of interest of primary investigators: Supported by Pfizer. No declaration of interest of primary investigators included

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskComputer-generated random sequence (details provided on request)
Allocation concealment (selection bias)Low riskCentral independent unit (details reported on request)
Blinding of participants and personnel (performance bias)
All outcomes
Low riskDouble blind (number and appearance of placebo capsules similar)
Blinding of outcome assessment (detection bias)
All outcomes
Low riskIndependent data imputation and statistical analysis (details reported on request)
Incomplete outcome data (attrition bias)
All outcomes
Low riskImputation using LOCF and ITT analysis for efficacy data
Selective reporting (reporting bias)Low riskAll outcomes reported

Rowbotham 2012

Methods

Study setting: Single-centre study in the USA, investigator-initiated

Study design: Parallel

Duration therapy: 8 weeks

Follow-up: Not performed

Analysis: Completer analysis by mixed effect regressions

Participants

Participants: 60 participants (93% female, 79% white, mean age 49 years)

Inclusion criteria: ACR 1990 criteria; age ≥ 18 years; McGill Pain Questionnaire VAS of at least 40mm on 100-mm scale at screening and mean daily diary pain at least 4 out of 10 during the 7 days prior to study drug initiation

Exclusion criteria: Pregnant or lactating women, hypersensitivity to levetiracetam or prior treatment with it; clinically significant liver, kidney or haematological disorders, a Westergen erythrocyte sedimentation rate exceeding 40 mm/min, abnormal elevated antinuclear antibody (1:160) or rheumatoid factor (> 80 IU/mL) levels, another explanation for their pain, illicit drug or alcohol abuse within the last year, involvement in unsettled litigation pertaining to their FM (such as automobile accident, civil lawsuit or worker's compensation), and ongoing monetary compensation as a result of litigation or disability claims with the exception of US social security disability benefits; subjects who are considered unreliable as to medication compliance or adherence to scheduled appointments as determined by the investigators; subjects who have serious or unstable medical or psychological conditions that in the opinion of the investigator(s), would compromise the subject's participation in the study

Interventions

Active drug: 40 participants were titrated to a maximum 3000 mg/d over a 6-week period and tapered off study medication after 8 weeks of treatment

Placebo: 26 participants

Rescue or allowed medication: Continuation of stable doses of antidepressants and opiates allowed

Outcomes

Pain: Diary pain (NRS 0-10); 30% and 50% pain reduction not reported and calculated by imputation method

Fatigue: FIQ Fatigue single scale not reported

Sleep: Diary sleep interference (NRS 0-10)

Depression: FIQ Depression not reported

Anxiety: FIQ Anxiety not reported

Disability: FIQ Disability not reported

Quality of life: FIQ total score (0-100)

Patient-perceived improvement: PGIC (1-7)

AEs: Subject-reported AEs were classified into 1 of 39 categories and a subject was considered positive even if they reported an AE at only 1 visit after beginning treatment

Notes

Safety: There were no SAEs in either group. AEs such as thinking abnormal, restlessness, itching and movement disorder were significantly more frequent in the levetiracetam group

Funding sources and any declaration of interest of primary investigators: Supported by an investigator-initiated grant from UCB Pharma and NIH grant K24 NS02164. No declaration of interest of primary investigator included

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskComputer-generated random sequence (details provided on request)
Allocation concealment (selection bias)Low riskCentral independent unit (details reported on request)
Blinding of participants and personnel (performance bias)
All outcomes
Low riskDouble blind (number and appearance of placebo capsules similar)
Blinding of outcome assessment (detection bias)
All outcomes
Low riskIndependent data imputation and statistical analysis (details reported on request)
Incomplete outcome data (attrition bias)
All outcomes
High riskOnly observed cases data available
Selective reporting (reporting bias)High riskOutcomes FIQ subscales fatigue, anxiety, depression, disability not reported

UCB 2006

  1. a

    ACR: American College of Rheumatology; AE: adverse event; BPI: Brief Pain Inventory; C-SSRS: Columbia Suicide Severity Rating Scale; CNS: central nervous system; ECG: electrocardiogram; FIQ: Fibromyalgia Impact Questionnaire; FM: fibromyalgia; HADS: Hospital Anxiety and Depression Scale; ITT: intention to treat; LOCF: last observation carried forward; MAF: Multidimensional Assessment of Fatigue; MOS: Medical Outcomes Study; NIH: National Institutes of Health; NRS: numeric rating scale; NSAID: non-steroidal anti-inflammatory drug; PGIC: Patient Global Impression of Change; SAE: serious adverse events; SF-36: Short-Form Health Survey - 36 items; SF-MPQ: Short-Form McGill Pain Questionnaire; VAS: visual analogue scale.

Methods

Study setting: Multicentre study, number of outpatient research centres in the USA not reported

Study design: Parallel

Duration therapy: 12 weeks

Follow-up: Not performed

Analysis: ITT; LOCF; analysis of covariance with treatment centre, week, treatment by week interaction and baseline scores as covariates

Participants

Participants: 159 (93% women, race not reported, mean age 50 years)

Inclusion criteria: Not reported

Exclusion criteria: Not reported

Interventions

Active drug: Lacosamide 400 mg (81 participants), 4-week titration from 100 mg/d to 400 mg/d, increasing by 100 mg/d at weekly intervals; 8-week maintenance

Placebo: 78 participants

Rescue or allowed medication: No details reported

Outcomes

Pain: Daily diary mean pain (NRS 0-10)

Fatigue: Fatigue score of FIQ (VAS 0-10)

Sleep: Mean daily interference with sleep (NRS 0-10)

Depression: HADS (NRS 0-21)

Anxiety: HADS (NRS 0-21)

Disability: Mean daily interference with activity (NRS 0-10)

Quality of life: FIQ total score (0-100)

Patient-perceived improvement: PGIC (1-7)

AEs: No details reported

Notes

Safety: No deaths in either group. No further details reported

Funding sources and any declaration of interest of primary investigators: Supported by UCB. No declaration of interest of primary investigators included

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskNot reported
Allocation concealment (selection bias)Unclear riskNot reported
Blinding of participants and personnel (performance bias)
All outcomes
Low riskDouble blind (number and appearance of placebo capsules similar)
Blinding of outcome assessment (detection bias)
All outcomes
Low riskParticipant reported and participant blinded
Incomplete outcome data (attrition bias)
All outcomes
High riskImputation using LOCF for continuous data reported, but not used. ITT for PGIC, AEs and withdrawals
Selective reporting (reporting bias)High riskNo details of AEs reported

Characteristics of excluded studies [ordered by study ID]

StudyReason for exclusion
Crofford 2008Pregabalin: Study design (enriched enrolment randomised withdrawal design) over 26 weeks cannot be combined with parallel or cross-over designs for meta-analysis
Roth 2012Pregabalin: Pregabalin and placebo treatment 4 weeks each within a cross-over design

Ancillary