Duloxetine for treating painful neuropathy or chronic pain (Review)

Transcription

1 Duloxetine for treating painful neuropathy or chronic pain (Review) Lunn MPT, Hughes RAC, Wiffen PJ This is a reprint of a Cochrane review, prepared and maintained by The Cochrane Collaboration and published in The Cochrane Library 2010, Issue 3

2 T A B L E O F C O N T E N T S HEADER ABSTRACT PLAIN LANGUAGE SUMMARY SUMMARY OF FINDINGS FOR THE MAIN COMPARISON BACKGROUND OBJECTIVES METHODS RESULTS Figure Figure Figure Figure Figure Figure Figure Figure Figure DISCUSSION AUTHORS CONCLUSIONS ACKNOWLEDGEMENTS REFERENCES CHARACTERISTICS OF STUDIES DATA AND ANALYSES Analysis 1.1. Comparison 1 Duloxetine versus placebo in the treatment of painful neuropathy, Outcome 1 Number of patients with >50% improvement of pain at 12 weeks or less Analysis 1.2. Comparison 1 Duloxetine versus placebo in the treatment of painful neuropathy, Outcome 2 Mean improvement in pain at 12 weeks or less Analysis 1.3. Comparison 1 Duloxetine versus placebo in the treatment of painful neuropathy, Outcome 3 Number of patients with >30% improvement in pain at 12 weeks or less Analysis 1.4. Comparison 1 Duloxetine versus placebo in the treatment of painful neuropathy, Outcome 4 Mean improvement in SF-36 Physical subscore at 12 weeks or less Analysis 1.5. Comparison 1 Duloxetine versus placebo in the treatment of painful neuropathy, Outcome 5 Mean improvement in SF-36 Mental Subscore at 12 weeks or less Analysis 1.6. Comparison 1 Duloxetine versus placebo in the treatment of painful neuropathy, Outcome 6 Mean improvement in SF-36 Bodily Pain Subscore at 12 weeks or less Analysis 1.7. Comparison 1 Duloxetine versus placebo in the treatment of painful neuropathy, Outcome 7 Mean improvement in Patient Reported Global Impression of Change at 12 weeks or less Analysis 1.8. Comparison 1 Duloxetine versus placebo in the treatment of painful neuropathy, Outcome 8 Mean improvement in BPI severity - average pain at 12 weeks or less Analysis 1.9. Comparison 1 Duloxetine versus placebo in the treatment of painful neuropathy, Outcome 9 Mean improvement in pain at rest (night pain) at 12 weeks or less Analysis 2.1. Comparison 2 Duloxetine versus placebo in the treatment of fibromyalgia, Outcome 1 Number of patients with = or >50% improvement of pain at = or <12 weeks Analysis 2.2. Comparison 2 Duloxetine versus placebo in the treatment of fibromyalgia, Outcome 2 Number of patients with = or >50% improvement of pain at >12 weeks Analysis 2.3. Comparison 2 Duloxetine versus placebo in the treatment of fibromyalgia, Outcome 3 Number of patients with = or >30% improvement of pain at = or <12 weeks Analysis 2.4. Comparison 2 Duloxetine versus placebo in the treatment of fibromyalgia, Outcome 4 Mean improvement in the SF36 mental component summary subscore Analysis 2.5. Comparison 2 Duloxetine versus placebo in the treatment of fibromyalgia, Outcome 5 Mean improvement in the SF36 physical component summary subscore i

3 Analysis 2.6. Comparison 2 Duloxetine versus placebo in the treatment of fibromyalgia, Outcome 6 Mean improvement in the SF36 Bodily Pain subscore Analysis 2.7. Comparison 2 Duloxetine versus placebo in the treatment of fibromyalgia, Outcome 7 Mean improvement in the patient reported global impression of change at 12 weeks or less Analysis 3.1. Comparison 3 Duloxetine versus placebo: adverse events during first 12 weeks of treatment for painful neuropathy or fibromyalgia, Outcome 1 Proportion of patients with any adverse event Analysis 3.2. Comparison 3 Duloxetine versus placebo: adverse events during first 12 weeks of treatment for painful neuropathy or fibromyalgia, Outcome 2 Nausea Analysis 3.3. Comparison 3 Duloxetine versus placebo: adverse events during first 12 weeks of treatment for painful neuropathy or fibromyalgia, Outcome 3 Somnolence Analysis 3.4. Comparison 3 Duloxetine versus placebo: adverse events during first 12 weeks of treatment for painful neuropathy or fibromyalgia, Outcome 4 Dry mouth Analysis 3.5. Comparison 3 Duloxetine versus placebo: adverse events during first 12 weeks of treatment for painful neuropathy or fibromyalgia, Outcome 5 Dizziness Analysis 3.6. Comparison 3 Duloxetine versus placebo: adverse events during first 12 weeks of treatment for painful neuropathy or fibromyalgia, Outcome 6 Adverse event leading to cessation Analysis 3.7. Comparison 3 Duloxetine versus placebo: adverse events during first 12 weeks of treatment for painful neuropathy or fibromyalgia, Outcome 7 Serious adverse event APPENDICES FEEDBACK WHAT S NEW HISTORY CONTRIBUTIONS OF AUTHORS DECLARATIONS OF INTEREST SOURCES OF SUPPORT DIFFERENCES BETWEEN PROTOCOL AND REVIEW INDEX TERMS ii

4 [Intervention Review] Duloxetine for treating painful neuropathy or chronic pain Michael PT Lunn 1, Richard AC Hughes 2, Philip J Wiffen 3 1 Department of Neurology, National Hospital for Neurology and Neurosurgery, London, UK. 2 MRC Centre for Neuromuscular Disease, National Hospital for Neurology and Neurosurgery, London, UK. 3 UK Cochrane Centre, Oxford, UK Contact address: Michael PT Lunn, Department of Neurology, National Hospital for Neurology and Neurosurgery, Queen Square, London, WC1N 3BG, UK. michael.lunn@uclh.nhs.uk. Editorial group: Cochrane Neuromuscular Disease Group. Publication status and date: Edited (no change to conclusions), comment added to review, published in Issue 3, Review content assessed as up-to-date: 1 March Citation: Lunn MPT, Hughes RAC, Wiffen PJ. Duloxetine for treating painful neuropathy or chronic pain. Cochrane Database of Systematic Reviews 2009, Issue 4. Art. No.: CD DOI: / CD pub2. Background A B S T R A C T Duloxetine is a balanced serotonin and noradrenaline reuptake inhibitor licensed for the treatment of major depressive disorders, urinary stress incontinence and the management of neuropathic pain associated with diabetic peripheral neuropathy. A number of trials have been conducted to investigate the use of duloxetine in neuropathic and nociceptive painful conditions. Objectives To assess the benefits and harms of duloxetine for treating painful neuropathy and different types of chronic pain. Search strategy We searched The Cochrane Neuromuscular Group Specialized Register (10 March 2009), The Cochrane Central Register of Controlled Trials (The Cochrane Library Issue 3, 2009), MEDLINE (January 1966 to March 2009), EMBASE (January 1980 to March 2009), and to March 2009 and the reference lists of identified publications for trials of duloxetine used for the treatment of painful peripheral neuropathy or chronic pain. Selection criteria We selected all randomised or quasi-randomised trials of any formulation of duloxetine, used for the treatment of painful peripheral neuropathy or chronic pain in adult participants. Data collection and analysis Two authors extracted data independently onto a specially designed proforma and cross checked them. Main results Six trials were identified including 2220 participants. Three studies included participants with painful diabetic neuropathy and three treated participants with fibromyalgia. Duloxetine at 60 mg daily is effective in treating painful diabetic peripheral neuropathy in the short-term to 12 weeks with a risk ratio (RR) for 50% pain reduction at 12 weeks of 1.65 (95% confidence interval (CI) 1.34 to 2.03), number needed to treat (NNT) 6 (95% CI 5 to 10). Duloxetine at 60 mg daily is also effective in fibromyalgia over 12 weeks (RR 50% reduction in pain 1.57, 95% CI 1.20 to 2.06; NNT 8, 95% CI 5 to 17) and 28 weeks (RR 1.58, 95% CI 1.10 to 2.27). Adverse events were common in both treatment and placebo arms but more common in the treatment arm with a dose dependent effect. Most side effects were minor, but 16% of participants stopped the drug due to side effects. Serious adverse events were rare. 1

5 Authors conclusions There is moderately strong evidence that duloxetine 60 mg and 120 mg daily are efficacious for treating pain in diabetic peripheral neuropathy and fibromyalgia but 20 mg daily is not. Minor side effects are common at therapeutic doses but serious side effects are rare. Direct comparisons of duloxetine with other antidepressants and with other drugs already shown to be efficacious in neuropathic pain would be appropriate and should include unbiased economic analyses. P L A I N L A N G U A G E S U M M A R Y Duloxetine for neuropathic pain Duloxetine is a drug used to treat depression and incontinence and it can be also be useful in treating certain types of pain. Pain can be produced spontaneously from damaged parts of the nervous system used to transmit pain information to the brain (neuropathic pain). When this damage is to nerves outside the spinal cord it is called a neuropathy. Pain is also produced when the nerves sense damage to another tissue (for example a pin being pressed into the skin) and this is called nociceptive pain. Some pain is of unclear origin, as no damage to the nervous system or to the tissues to which the nerves connect can be identified. This sort of pain happens, for example, in fibromyalgia. Duloxetine is also used to treat depression and problems with bladder function but the objective of this review was to consider all the evidence from double-blind randomised controlled trials relating to the use of duloxetine to treat painful neuropathy and chronic pains of all sorts. We looked at all the published scientific literature and identified six drug trials involving a total of 2220 participants that were of sufficient quality and reliability to include. Three tested the effect of duloxetine on painful diabetic neuropathy and three on the pain of fibromyalgia. The usual dose of duloxetine is 60 mg. At this dose, there was moderately strong evidence that duloxetine reduced pain in both painful diabetic peripheral neuropathy and fibromyalgia. In diabetic peripheral neuropathic pain the relative rate of 50% improvement with duloxetine 60 mg per day was just over one and a half times more than with placebo. This equates to needing to treat 6 people with diabetic peripheral neuropathic pain with duloxetine to achieve a 50% response in one person. The effect on fibromyalgia was similar. A dose of 20 mg was not effective and a higher dose of 120 mg was no more effective than 60 mg. Most people taking duloxetine will have at least one side effect. These are mostly minor and are commonly feeling sick, being too awake or too sleepy, developing a headache, having a dry mouth or becoming constipated or dizzy. About one in six people stop duloxetine because of side effects but serious problems caused by duloxetine are very rare. Duloxetine may be about as good at reducing these sorts of pain as some of the other antidepressant drugs on the market but the evidence supporting this comparison is not strong. We have concluded that duloxetine is useful for treating pain caused by fibromyalgia and diabetic neuropathy and it seems to be about as effective as other similar drugs already on the market. No direct comparison has been performed between duloxetine and any of these other drugs. As yet it is not clear whether the use of duloxetine is cost effective when compared to the other drugs already in use. 2

6 S U M M A R Y O F F I N D I N G S F O R T H E M A I N C O M P A R I S O N [Explanation] Duloxetine for treating painful neuropathy or chronic pain Patient or population: patients with treating painful neuropathy or chronic pain Settings: Intervention: Duloxetine Outcomes Illustrative comparative risks*(95% CI) Relative effect (95% CI) Greater than 50% improvement of diabetic peripheral neuropathic pain- Duloxetine 60 mg daily 11pointLikertscore 1 (follow-up: 12 weeks) Assumed risk Control 288per Corresponding risk Duloxetine 475per1000 (386to585) RR 1.65 (1.34 to 2.03) No of Participants (studies) 655 (3) Quality of the evidence (GRADE) moderate 3 Comments Number needed to treat (NNT) for 50% or more improvement in pain with duloxetine60mgdaily= 6(CI5to10) 4 Improvement in diabetic peripheral neuropathic pain -Duloxetine 60mg daily 11-point Likert Scale. Scalefrom:0to10. (follow-up: 12 weeks) The mean improvement in diabetic peripheral neuropathic pain - duloxetine 60mgdailyinthecontrol groups was on Likert pain scale The mean Improvement in diabetic peripheral neuropathic pain - Duloxetine 60 mg daily in the intervention groups was 1.04 lower (1.37 to 0.71 lower) 618 (3) moderate 5 Greater than 30% improvement in diabetic peripheral neuropathic pain -Duloxetine 60mg daily 11-pointLikertScale 6 (follow-up: 12 weeks) 429per per1000 (545to785) RR 1.53 (1.27 to 1.83) 442 (2) moderate NNTfor30%ormoreimprovement in pain with duloxetine60mgdaily= 5(CI3to8) 3

7 Greater than 50% improvement of fibromyalgia pain -Duloxetine 60 mg daily 11-point Likert scale (follow-up: 12 weeks) Adverse event leading to cessation - Duloxetine 60mgdaily 233per per per1000 (280to480) 139per1000 (100to193) RR 1.57 (1.2 to 2.06) RR 1.67 (1.2 to 2.32) 528 (2) 1215 (5) moderate 7 moderate NNTfor50%ormoreimprovement in fibromyalgia pain with duloxetine 60mgdaily=8(CI5to 17) Number needed to harm (NNH) for cessation of duloxetine treatment at duloxetine60mgdaily= 17(CI12to50) *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; RR: Risk ratio; GRADE Working Group grades of evidance 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. Lowquality:Furtherresearchisverylikelytohaveanimportantimpactonourconfidenceintheestimateofeffectandislikelytochangetheestimate. Very low quality: We are very uncertain about the estimate. 1 Titled24hourpainscoreor24hourpainseverity 2 Assumedrisksgeneratedfromplacebopopulation 3 Three trials only all company sponsored and performed but all trials pre-registered on clinical trials.gov have been published. No publication bias detected. 4 CI=confidenceinterval 5 Riskofbiasrelevantin2of3studies.Qualityofevidencegradedmoderatebecauseofhighdropoutandcompanysponsorshipofall studies. 6 Titled24hourpainscoreor24hourpainseverity 7 Riskofbiasrelevantin2of3studies.Qualityofevidencegradedmoderatebecauseofhighdropoutandcompanysponsorshipofall studies. 4

8 B A C K G R O U N D Pain is common in peripheral nerve diseases such as the peripheral neuropathy associated with diabetes mellitus. Painful neuropathy is a particular example of neuropathic pain. Neuropathic pain is pain arising as a direct consequence of a lesion or disease affecting the somatosensory system (Treede 2008). Neuropathic pain is different from conventional or nociceptive pain. Nociceptive pain arises from the activation of primary pain receptors in response to injury or inflammation. Painful neuropathies are diseases of the peripheral nerves which cause neuropathic pain. Chronic pain has been classified as pain exceeding three months duration (Nagda 2004). Chronic pain is a major health problem affecting one in five people in Europe (Breivik 2006). A recent community based study in North West England estimated the prevalence of chronic painful peripheral neuropathy in people without diabetes as 4.9% (Daousi 2004).The prevalence in people with diabetes in the same community was 16.2%. In a large community study in the United Kingdom, the annual incidence of neuropathic pain was calculated as at least 84 per 100,000 by adding the incidence of the four commonest and most disabling causes (diabetic neuropathy, trigeminal neuralgia, postherpetic neuralgia and phantom limb pain) (Hall 2006). The pain of painful peripheral neuropathy can be diverse and distressing. Descriptions include burning, cold, electric shocks, lancinating, tight or aching. Other spontaneous and evoked positive sensory symptoms include painful numbness, tingling or paraesthesiae. Stimuli which are not usually painful may be perceived as painful, a phenomenon called allodynia. Chronic pain can have serious complex adverse psychological and social effects. Duloxetine is one of a newer type of antidepressant drug. It is a relatively balanced dual reuptake inhibitor of serotonin and noradrenaline (Schuessler 2006). Theoretically these actions should make it a good pain modulating agent (Bymaster 2001; Bymaster 2005). Serotonin modulates both pro-nociceptive and anti-nociceptive descending effects on central pain pathways from the brainstem. Noradrenaline has a predominantly anti-nociceptive effect. Balance between facilitation and depression of pain pathways is important for normal function. Drugs that inhibit the reuptake of serotonin and noradrenaline potentiate monoamine neurotransmission in the descending inhibitory spinal pathways and so reduce nociceptive afferent transmission in the ascending spinal pain pathways. Potentiation of both serotonin and noradrenaline is required to produce effective analgesia. The action of drugs such as duloxetine is independent of their effects on depression (Perahia 2006). Onset of benefit occurs within days, earlier and at lower doses than in depression. Furthermore, they have similar effects on pain in depressed and non-depressed people. Common side effects include nausea, headache, dry mouth, insomnia, constipation, dizziness, fatigue, somnolence, hyperhydrosis and diarrhoea (Gahimer 2007). These are mainly classified as mild to moderate and anecdotally appear less prevalent than the side effects with tricyclic antidepressants. Duloxetine is licensed in the United States, European Union and United Kingdom for the treatment of major depressive disorder (Nose 2007), urinary stress incontinence (Mariappan 2005) and for the management of neuropathic pain associated with diabetic peripheral neuropathy. We do not know of a published systematic review of duloxetine for any pain condition and it was not included in a previous Cochrane review of antidepressants for neuropathic pain (Saarto 2007). This review aims to fill the gap. Painful neuropathy is the principal focus of this review because duloxetine has been chiefly tried for this indication. However, previous Cochrane reviews of interventions for pain have covered all forms of either neuropathic pain or acute and chronic pain. For conformity with these other reviews, we will include all forms of chronic pain but not acute pain, for which duloxetine has not been proposed as a treatment. O B J E C T I V E S To assess the benefits and harms of duloxetine for treating painful neuropathy and different types of chronic pain. M E T H O D S Criteria for considering studies for this review Types of studies For the detection of benefits, we included only double-blind randomised trials of duloxetine for treating painful neuropathy or chronic pain conditions. Duloxetine was to have been administered for a minimum of eight weeks. Types of participants We included participants with any form of painful peripheral neuropathy or chronic pain. Types of interventions We included all formulations and doses of duloxetine in comparison with placebo or other controls. We reported comparisons with placebo and with other controls separately. 5

9 Types of outcome measures Primary outcomes The primary outcome was short-term (up to 12 weeks) improvement of pain compared with baseline using validated scales of pain intensity or pain relief. Both visual analogue and categorical scales were accepted. Where pain relief was expressed as none, minor, moderate, major or complete, only moderate, major or complete were to be considered as improvement. Where pain was measured with a continuous scale, improvement was to be taken to be an improvement of 50% or more from baseline on that scale. If results were only reported as improvement on a continuous scale, we planned to try to obtain results from the authors to provide this dichotomous analysis. Electronic searches We searched the Specialized Registers of the Cochrane Neuromuscular Disease Group and the Cochrane PAPAS Group for randomised trials using the following search terms: (duloxetine OR Cymbalta) AND (painful neuropathy OR postherpetic neuralgia OR trigeminal neuralgia OR (peripheral neuropathy AND pain)). We adapted the strategy to search the Cochrane Central Register of Controlled Trials (The Cochrane Library Issue 3, 2009), MEDLINE (January 1950 to March 2009) and EMBASE (January 1980 to March 2009). We searched the National Institute for Health ( and World Health Organisation databases for current on-going registered trials. The search strategies for MEDLINE, EMBASE and CENTRAL are listed in Appendix 1, Appendix 2 and Appendix 3 Secondary outcomes (1) Long-term (more than 12 weeks) improvement of pain compared with baseline was analysed as for the primary outcome. (2) Improvement in short-term (up to 12 weeks) and long-term (more than 12 weeks) pain of at least 30% compared with baseline using validated scales of pain intensity and or pain relief, analysed as for the primary outcome. (3) Improvement in any validated Quality of Life Score of more than 30% compared to the baseline. (4) As the outcome measures for the assessment of pain were likely to be diverse and the majority use standard subjective scales for pain intensity or pain relief or both, further results were to be analysed according to the third to sixth types in a hierarchy modified from Wiffen et al. (Wiffen 2005). The full hierarchy of outcome measures is as follows: 1. Patient reported pain relief of 50% or greater. 2. Patient reported pain relief of 30% or greater. 3. Patient reported global impression of clinical change. 4. Pain on movement. 5. Pain on rest. 6. Any other pain related measure. (5) Adverse events during treatment were reported. We analysed categories of: all adverse events, severe or serious adverse events that lead to hospitalisation or death, and adverse events leading to cessation of treatment. The percentage of pain improvement considered clinically important for dichotomous outcomes was chosen as 30% and 50% in line with the recommendations made by the Initiative on Methods, Measurement, and Pain Assessment in Clinical Trials (IMM- PACT) (Dworkin 2008). Improvement in pain intensity of 30% or more is considered moderately important and 50% or more, substantial improvement. Search methods for identification of studies Searching other resources We also wrote to Eli Lilly who make duloxetine and to pain experts asking for information about other or ongoing trials. Data collection and analysis Selection of studies Two review authors (ML and RACH) scrutinised all the titles and abstracts revealed by the searches and determined which fulfilled the selection criteria. We resolved disagreement by discussion. It was not necessary to engage the third review author (PW) as disagreements were resolved. Data extraction and management Two review authors (ML and RACH) extracted data independently onto a specially designed data extraction form. We would have resolved disagreements by discussion if necessary with the third review author (PW) but this was not necessary. One author (ML) entered data into Review Manager (RevMan) and a second author (RACH) checked them since the double data entry facility is no longer available in the RevMan software. Assessment of risk of bias of included studies We used the methods of the Cochrane Collaboration to assess risk of bias as set out in Chapter 8 of the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2008) rather than those stipulated in the protocol for this review which predated the new methods. 6

10 Measures of treatment effect The effect measures of choice were the risk ratio (RR) for dichotomous data and the weighted mean difference (WMD) or standardised mean difference (SMD) for continuous data. Uncertainty was expressed with 95% confidence intervals (CIs). We also expressed the most important results as numbers needed to treat (NNT) and numbers need to harm (NNH) where appropriate. Cross-over trials If cross-over trials had been included we would have analysed the results using the estimated differences in effects and their standard errors with the generic inverse variance (GIV) facility in RevMan. For results using dichotomous outcomes this would have been more difficult, but would have been be done if the results could have been be converted to odds ratios (OR) on the log scale and the standard errors calculated. If necessary we would also have analysed the results following the methods of Elbourne 2002 (Elbourne 2002) with the assistance of a statistician. Sensitivity analysis We intended to conduct the following sensitivity analyses: trials which did and did not have perfect scores for risk of bias, trials with more or less than 20% drop out or loss to follow-up rate, trials which were and were not led by the company which produced the drug, and trials with more and less than 100 participants. Adverse events Since randomised trials may not capture all important adverse events, we reported in the Discussion information about adverse events collected from other Cochrane reviews of duloxetine and from the review of non-randomised studies in the most recent edition of Meyler s Side Effects of Drugs (Aronson 2006). Economic issues We considered costs in the Discussion. R E S U L T S Data synthesis and assessment of heterogeneity We undertook each meta-analysis using a fixed-effect model in the RevMan software. We used the I 2 statistic for heterogeneity and if its value had been greater than 50% we would have inspected the trials, forest and L Abbé plots for differences between trials which might have explained the heterogeneity. If no explanation could be found we would have repeated the analysis with a random-effects model. Assessment of reporting biases If there had been sufficient trials, we would have inspected funnel plots for asymmetry which might have been due to publication bias. We were aware that funnel plots and statistical tests based on them are not reliable indicators of publication bias and any interpretations made on the basis of them would have been made with great caution. Subgroup analysis We reported separately results for painful diabetic neuropathy, fibromyalgia and chronic pain and would have reported separately other specific causes of neuropathic pain: trigeminal neuralgia, postherpetic neuralgia and central ( thalamic ) pain. In addition to reporting together the results of all forms of painful neuropathy, we would also have reported separately the results for the following different diagnostic subgroups: diabetic neuropathy, HIV neuropathy, idiopathic painful neuropathy. Description of studies See: Characteristics of included studies; Characteristics of excluded studies; Characteristics of studies awaiting classification; Characteristics of ongoing studies. We searched the specialised registers of the Neuromuscular Disease Group and the Pain, Palliative and Supportive Care (PAPAS) Group for randomised trials using the search terms above. We searched CENTRAL (The Cochrane Library 2009, Issue 1), MED- LINE (from January 1950 to March 2009) and EMBASE (from January 1980 to March 2009) using the search strategies listed in Appendix 1 and Appendix 2. We identified 130 references to possible trials (MEDLINE 12, EMBASE 75, CENTRAL 19, Neuromuscular Disease Group and the Pain, Palliative and Supportive Care (PAPAS) Group Register and Library 22 and handsearches 2). Following exclusion of duplicates and studies clearly not relevant, two authors checked 37 titles identifying 14 randomised controlled trials or possibly randomised controlled trials and two further trials awaiting classification (one abstract (Skljarevski 2008) and one trial in Spanish awaiting translation (Canovas 2007). Of these, eight were excluded (Goldstein 2004; Brannan 2005; Raskin 2005a; Raskin 2006a; Raskin 2006b; Russell 2006; Wernicke 2006b; Wu 2006). Two of the excluded texts were reports of other studies which we have included (Raskin 2005a; Russell 2006). Five trials did not meet the pre-defined criteria for inclusion in this review. Two were open studies without blinding or a control group (Raskin 2006b; Wu 2006). The Raskin 2006a trial investigated duloxetine versus routine care for diabetic peripheral neuropathic pain but this 7

11 trial, although randomised, was not blinded; it is mentioned in the Discussion of this review. The Wernicke 2006b study was a 52-week extension of the Goldstein 2005 randomised trial with a similar design but without blinding. The Brannan 2005 trial measured outcomes at seven weeks, not the eight weeks stipulated in our protocol. The trial Goldstein 2004 tested duloxetine for eight weeks for depression, but included pain scales as secondary outcome measures. However, it was not clear what sort of pain the participants had (for example musculoskeletal, neuropathic, headache) and the levels of pain at baseline were low compared to the included trials. This trial was therefore excluded. Six trials, including 2220 participants, met the predefined inclusion criteria for this review (Arnold 2004; Arnold 2005; Goldstein 2005; Raskin 2005; Russell 2008; Wernicke 2006). The characteristics of these studies are described in the Characteristics of included studies table. Three studies, including 1139 participants, compared duloxetine with placebo for three months for painful diabetic peripheral neuropathy (Goldstein 2005; Raskin 2005; Wernicke 2006). Three studies, including 1081 participants, tested duloxetine for fibromyalgia, two for 12 weeks (Arnold 2004; Arnold 2005) and one for six months (Russell 2008). The three studies of diabetic peripheral neuropathy were very similar in design and were all conducted by the same drug company. Participants all had to be 18 years old or older, they had to have had a length dependent painful peripheral neuropathy for at least six months that scored > 3 on the Michigan Neuropathy Screeening Instrument (MNSI) and they had a minimum average 24-hour pain score of 4 on an 11-point Likert scale. The age, sex, pain severity and duration of pain at entry for the participants were similar in the treatment groups in each trial and between trials. All three trials treated participants with duloxetine in capsule form at doses of 60 mg once or twice per day or identical placebo, with the addition of a 20 mg once daily dose in the trial of Goldstein (Goldstein 2005). Treatment was for 12 weeks with a one week taper in two trials (Raskin 2005; Wernicke 2006). All three trials took place in healthcare and research centre settings. The three studies of fibromyalgia included participants aged 18 or older who fulfilled American College of Rheumatology criteria for fibromyalgia and had significant pain at entry ( 4 on the pain intensity item of the Fibromyalgia Impact Questionnaire (Arnold 2004) or Brief Pain Inventory (Arnold 2005; Russell 2008)). They were also conducted by the same drug company that performed the diabetic neuropathy studies. One study included only women (Arnold 2005) and the other two included almost only women, despite being open, reflecting the epidemiology of this condition. In the two studies where ages were given (Arnold 2004; Russell 2008) the participants were approximately 10 years younger than in the diabetic peripheral neuropathy trials. Participants had similar levels of pain at entry in each trial. Only Arnold 2005 stated the duration of pain at entry (>12 weeks). All three trials were blinded. Participants were treated with duloxetine in capsules or identical appearing placebos in identical dosage schedules in outpatient research facilities for 12 weeks, Arnold 2004 and Arnold 2005, or for 28 weeks (Russell 2008). The Russell 2008 trial included a two week titration phase and a two week taper. Two studies await classification. One is written in Spanish and awaits translation (Canovas 2007). Duloxetine was compared to amitriptyline for the treatment of neuropathic pain due to a number of different causes. One possible study, published in abstract form only (Skljarevski 2008), compared duloxetine to placebo in the treatment of chronic low back pain over 13 weeks. Full publication is awaited. Risk of bias in included studies The six included studies had a low risk of bias on most attributes (see Figure 1 and Characteristics of included studies). All the studies except for Raskin 2005 had dropout rates more than 20%, which was deemed to increase the overall risk of bias from low to moderate. All were sponsored and performed by the same company. 8

12 Figure 1. Methodological quality summary: review authors judgements about each methodological quality item for each included study. The issue of incomplete outcome data in the Arnold 2004 study was unclear from the publication; 36% of the placebo group and 44% of the duloxetine group discontinued the study. From information provided by the authors, all participants with at least one follow-up measurement from baseline were included in the analysis, with the last observation carried forward. Four other studies were judged to be at low risk of bias. Sequence generation methods for randomisation were not described in Arnold 2005 and the methods of allocation concealment were also unclear. However, since allocation concealment was adequate in Arnold 2004 it was judged that these attributes were likely to be satisfactory, subsequently confirmed by the author. As in Arnold 2004, there were large but matched numbers of dropouts (41 (35%) from the duloxetine 60 mg daily group, 45 (39%) from the duloxetine 60 mg twice daily group, and 52 (43%) from the placebo group). A partial intention to treat analysis was carried out but the results may be biassed because of the dropouts. Similarly, dropout rates of 24% for placebo treated participants to 29% for those on high dose duloxetine treatment, of which significantly more were due to side effects in the duloxetine 60 mg twice daily group (P = < 0.001), are a potential source of bias in Goldstein 9

13 2005. Potential sources of bias were found in (Russell 2008) (allocation concealment) and (Wernicke 2006) (selective reporting of outcomes). These were not considered significant sources of bias since the same groups had previously performed trials with adequate allocation concealment and reporting. One study had a low risk of bias (Raskin 2005). Effects of interventions See: Summary of findings for the main comparison The effects of the interventions were analysed separately for painful peripheral neuropathy and fibromyalgia and meta-analysis combining the trials for the two conditions was not performed. There were no trials of duloxetine for other causes of neuropathic pain. There were no cross-over trials. None of the randomised trials included health economic data. Painful peripheral neuropathy Primary outcome: short-term (up to 12 weeks) improvement of pain compared with baseline Three trials in painful diabetic neuropathy reported data on the primary outcome measure of 50% improvement of pain compared with baseline at less than 12 weeks (Goldstein 2005; Raskin 2005; Wernicke 2006). Participants were treated with duloxetine 20 mg, 60 mg or 120 mg per day. Combining data from all doses from the three trials together (1102 participants), the RR of 50% improvement with any dose was 1.63 (95% CI 1.35 to 1.97) greater than with placebo (see Figure 2, Analysis 1.1). The RR of improvement was significantly greater than placebo for the 60 mg daily and 120 mg daily doses but not the 20 mg daily dose, for which it was 1.43 (95% CI 0.98 to 2.09). The RR of improvement with 120 mg duloxetine per day (1.66, 95% CI 1.35 to 2.04) was not significantly greater than that with 60 mg per day (1.65, 95% CI 1.34 to 2.03). 10

14 Figure 2. Duloxetine versus placebo in the treatment of painful neuropathy: Number of patients with >50% improvement of pain at <12 weeks. The mean improvement in pain at less than 12 weeks on an 11- point Likert scale was significantly greater than placebo with the 60 mg (-1.04 (95% CI to -0.71)) and 120 mg (-1.16 (95% CI to -0.83)) doses of duloxetine than with placebo, but not with the 20 mg dose (see Figure 3, Analysis 1.2). 11

15 Figure 3. Duloxetine versus placebo in the treatment of pain: Mean improvement in pain at 12 weeks. Secondary outcomes None of the included trials of painful diabetic neuropathy included outcomes at more than 12 weeks. Two trials (Raskin 2005; Wernicke 2006) included data on >30% improvement of pain at 12 weeks or less. The results were similar to those for at least 50% improvement. Relative rates of improvement were significantly greater than placebo with duloxetine for the 60 mg dose (1.53, 95% CI 1.27 to 1.83), the 120 mg dose (1.55, 95% CI 1.30 to 1.86) and for both doses combined (1.54, 95% CI 1.30 to 1.82)(see Figure 4, Analysis 1.3). Data for this outcome for the 20 mg dose were not available. 12

16 Figure 4. Duloxetine versus placebo in the treatment of pain: Number of patients with >30% improvement in pain at <12 weeks. Trials that included quality of life information used the SF-36. We included data on the relevant physical, mental and bodily pain subsections of the SF-36. In painful diabetic neuropathy, the effect of 20 mg duloxetine was not significant on any of the selected SF- 36 subscores at up to 12 weeks (Raskin 2005; Wernicke 2006). The WMD of improvement on the physical summary component was significantly greater with the 60 mg dose (2.51, 95% CI 1.00 to 4.01) and 120 mg dose (2.80, 95% CI 1.04 to 4.55) (see Analysis 1.4). The WMD on the mental summary component was significantly greater only with the 120 mg dose (2.23, 95% CI 0.69 to 3.77) (see Analysis 1.5). The WMD on the bodily pain subscale showed significantly more improvement than placebo with the 60 mg dose (5.58, 95% CI 1.74 to 9.42) and even more with the 120 mg dose (8.19, 95% CI 4.33 to 12.05) but not with the 20 mg dose (see Analysis 1.6). Three studies reported the patient global impression of change (PGIC) and pain at rest (night pain) (Goldstein 2005; Raskin 2005; Wernicke 2006) and two reported the bodily pain index (BPI) (Raskin 2005; Wernicke 2006). Mean improvements only were reported. The WMD for each outcome was not significant for the 20 mg dose but was significant and similar in magnitude for the 60 mg and 120 mg doses (see Figure 5, Analysis 1.7). However, a clinically meaningful difference in the PGIC is suggested as 1 point (Dworkin 2008) and hence the change associated with 60 mg duloxetine (-0.59, 95% CI to -0.41) may not be clinically significant. The RR for the BPI (see Analysis 1.8; Figure 6) is statistically significantly reduced by (95% CI to -0.57) but again this borders on a change considered clinically significant (Dworkin 2008). 13

17 Figure 5. Duloxetine versus placebo in the treatment of pain: Patient reported global impression of change. Figure 6. Duloxetine versus placebo in the treatment of pain: BPI severity - average pain. Heterogeneity Significant heterogeneity was not present except for the SF-36 physical component summary, PGIC, bodily pain index and pain at rest. Heterogeneity was present in the subgroup analyses and also in the All Doses where applicable. These analyses were performed with a random-effects model. Fibromyalgia 14

18 Primary outcome Three trials reported data corresponding to the primary outcome for this review. Two scales were used, the Fibromyalgia Impact Questionnaire pain score (Burckhardt 1991) in Arnold 2004 and the Brief Pain Inventory average pain severity score in Arnold 2005 and Russell The 20 mg dose of duloxetine in the Russell 2008 study did not show significant differences in any of the reported measures. Short-term (up to 12 weeks) > 50% improvement of pain compared with baseline was reported for all three studies. The RR of improvement was significantly greater with duloxetine 60 mg (1.57, 95% CI 1.20 to 2.06) and with 120 mg daily (1.73, 95% CI 1.36 to 2.19) than with placebo (see Analysis 2.1). The RR of improvement compared with placebo for all doses in all three trials, with a total of 1072 participants, was 1.71 (95% CI 1.37 to 2.13). Secondary outcomes Only a single study looked at long-term outcome at more than 12 weeks (Russell 2008). These investigators documented outcomes at 28 weeks. Improvement of pain > 50% compared with baseline at 28 weeks was similar between the 60 mg and 120 mg doses and when all participants were combined, the RR for improvement was 1.62 (95% CI 1.25 to 2.11) (see Analysis 2.2). The RR of > 30% improvement at 12 weeks or less was significantly greater than placebo with the duloxetine 60 mg dose (RR 1.52, 95% CI 1.24 to 1.85) and 120 mg dose (1.52, 95% CI 1.24 to 1.86) but not with the 20 mg dose (see Figure 7, Analysis 2.3). The RR for all doses combined was 1.50 (95% CI 1.25 to 1.80). Figure 7. Duloxetine versus placebo in the treatment of fibromyalgia: >30% improvement <12 weeks. 15

19 All three studies (Arnold 2004; Arnold 2005; Russell 2008) documented the physical and mental component summary scores and bodily pain subscores of the SF-36. For the mental component summary score, the only significant result was at the 120 mg dose (WMD 3.09, 95% CI 1.47 to 4.70) more improvement on duloxetine than placebo (see Analysis 2.4). Similarly for the physical component summary score, the only significant result was also at the 120 mg dose (WMD 1.80, 95% CI 0.50 to 3.10) more improvement on duloxetine (see Analysis 2.5). For the bodily pain subscale the RR of improvement from two studies (Arnold 2004; Arnold 2005) was significantly greater for duloxetine than placebo at both the 60 mg dose (RR 8.20, 95% CI 3.20 to 13.20) and the 120 mg dose (RR 8.42, 95% CI 4.80 to 12.03) (Figure 8, see Analysis 2.6). Figure 8. Duloxetine versus placebo in the treatment of fibromyalgia: SF36 Bodily Pain. Two studies reported the patient global impression of improvement (Arnold 2005; Russell 2008), which was significantly in favour of duloxetine at the 20 mg dose (WMD -0.54, 95% CI to -0.12), 60 mg dose (WMD -0.45, 95% CI to ) and 120 mg dose (WMD -0.54, 95% CI to -0.26) (see Analysis 2.7). Heterogeneity Significant heterogeneity was found in some analyses (for example Analyses 1.4, 1.8, 1.9, 2.4 and 3.1). Heterogeneity was present in the subgroup analyses and also in the All Doses where applicable. The reason for this was not evident from the trial design and hence a random-effects model was selected for these outcomes where appropriate (see Methods above and Data and analyses). Adverse events Adverse events were analysed across all included studies. Adverse events were very common in all of the trials in both experimental and placebo groups. The rate of any adverse event was high in both the treatment and placebo arms of all studies, with 470 adverse events being reported in 706 control participants and 1398 adverse events occurring in 1435 participants in the combined treatment arms. Because the numbers of adverse events exceeded patient numbers (i.e. >1 adverse event per patient) at doses of 60 mg daily, the RR was not estimable for some studies. At a duloxetine dose of 20 mg the rate of any adverse event was not significantly different from the placebo rate (RR 1.35, 95% CI 1.00 to 1.82) (Goldstein 2005). Adverse events were significantly more common in the 60 mg and 120 mg duloxetine groups. With the usual clinical 60 mg daily dose of duloxetine, the RR for any treatment emergent adverse event was 1.21 (95% CI 1.12 to 1.30) and at 120 mg daily it was 1.19 (95% CI 1.09 to 1.30) (Analysis 3.1). 16

20 The most common adverse events were nausea (Analysis 3.2), dry mouth (Analysis 3.4), dizziness (Analysis 3.5), somnolence (Analysis 3.3), fatigue, insomnia, constipation, decreased appetite, sweating and rhinitis. All have a dose dependency, with a greater frequency of side effects at 120 mg daily than 60 mg daily. Adverse events leading to cessation of treatment were significantly more frequent in the duloxetine 60 mg (RR 1.67, 95% CI 1.20 to 2.32) and 120 mg groups (RR 2.30, 95% CI 1.74 to 3.05) than in the placebo group (Figure 9, see Analysis 3.6). Figure 9. Adverse events leading to cessation of treatment. 17

21 Serious adverse events were rare, with 19 serious adverse events in 1250 participants randomised to duloxetine and 14 in 606 randomised to placebo. The was no significant difference in severe adverse effects between duloxetine and placebo (RR 0.64, 95% CI 0.33 to 1.25). D I S C U S S I O N This systematic review identified six double-blind randomised trials investigating the effects of duloxetine in chronic painful conditions that met the criteria for inclusion laid down in the published protocol. With the exception of a high proportion of dropouts in five of the six trials, they were considered by us to have a low risk of bias. They were sponsored and published by researchers from one pharmaceutical company producing the drug but this fact on its own was not judged to be a significant source of bias. They were well performed and the company was very open about providing any further data required. One randomised double-blind trial (Brannan 2005) was excluded on the basis that the duration of follow-up was only seven weeks, shorter than the eight weeks specified in the published protocol; this trial is discussed below. Two studies await further information before inclusion (see above). Duloxetine has only been tested in randomised controlled trials treating two sorts of pain: painful diabetic neuropathy and fibromyalgia. These are very different in presentation and pathogenesis and as a result have been analysed separately. The meta-analyses showed similar beneficial effects in both. This might but does not necessarily imply that duloxetine exerts its pain relieving effect by similar mechanisms in these two very different conditions. Other systematic reviews have analysed these two conditions separately and together and reported similar benefit (Sultan 2008). We judged that until the pathogenesis is better understood that they should be analysed separately. Some authors have suggested that the role of duloxetine in pain management is in treating depression, a kind of implied role in pain management (Guay 2006). The trials included in this review randomised participants with and without depression and at baseline all trials were balanced for the presence or absence of depression. There were broadly consistent improvements in pain scores in all six trials included in the two meta-analyses. In five (Arnold 2004; Arnold 2005; Goldstein 2005; Russell 2008; Wernicke 2006) some indices of depression improved significantly, so that treatment of depression might play a part in the treatment of pain, even when participants were excluded from entering the trial if they had pre-existing major depressive disorder. However in Raskin 2005, depression scales did not change despite similar pain scale improvements but patients with pre-existing depression were specifically excluded from entering the trial. Russell et al. (Russell 2008) used two regression models to separate the direct effect of duloxetine on pain and the indirect effect on the treatment of depression in the treatment of pain. They calculated that between 21% and 38% of the pain treatment effect was due to treatment of depressive symptoms. Fava 2004 estimated a 50% influence. For painful diabetic neuropathy, the RR of > 50% reduction in pain at 8 to 12 weeks was 1.63 (95% CI 1.35 to 1.97) greater with duloxetine. For a standard dosage of 60 mg duloxetine daily, this corresponds to a number needed to treat (NNT) of 6 (95% CI 5 to 10) (Summary of findings for the main comparison). For > 30% improvement of pain, the RR was 1.53 (95% CI 1.27 to 1.83), corresponding to a NNT of 5 (95% CI 3 to 8) (Summary of findings for the main comparison). These NNTs are of similar magnitude but not as good as the NNT for tricyclic antidepressants for the achievement of moderate pain relief (NNT 3.6, 95% CI 3 to 4.5) (Saarto 2007; Kajdasz 2007) and for amitriptyline for the achievement of global improvement (NNT 3.2, 95% CI 2.6 to 4.2) (Sultan 2008). There were improvements in the other prespecified secondary outcome measures at 60 mg and 120 mg daily doses of duloxetine, with slightly but not significantly greater improvements with the 120 mg dose. The PGIC and BPI secondary outcomes are statistically significant, but only border on magnitudes of change that are currently considered to be clinically significant (Dworkin 2008). No randomised controlled trials of the effect of duloxetine on painful diabetic neuropathy for periods longer than 12 weeks were discovered. The magnitude of improvement in pain at 12 weeks in participants with fibromyalgia was similar to that seen in participants with diabetic peripheral neuropathy. There was no clear differentiation of the effects of duloxetine in fibromyalgia or diabetic peripheral neuropathy and so even if fibromyalgia is a different sort of pain (Dadabhoy 2006), it seems to respond to duloxetine in a similar way. However, the absolute risk reduction was less than that for diabetic peripheral neuropathic pain and the corresponding NNT for 50% or more pain relief at a duloxetine dose of 60 mg daily was 8 (95% CI 5 to 17) (Summary of findings for the main comparison). A number of studies were not included in the meta-analysis and are included here for completeness. Brannan 2005 performed a double-blind randomised controlled trial of duloxetine in major depressive disorders, in which they also measured the effect on pain. The outcomes were measured at seven weeks and hence it 18