BackgroundChronic pelvic pain affects 2–24% of women worldwide, and evidence for medical treatments is limited. Gabapentin is effective in treating some chronic pain conditions, but its effect on central pain processing is unknown.ObjectivesTo test the hypothesis that gabapentin can reduce pain and improve physical and emotional functioning in women with chronic pelvic pain. We investigated the mechanism of action of gabapentin in a subset of women.DesignA randomised, double-blind, placebo-controlled, multicentre trial with a brain imaging substudy.SettingThis trial took place in 39 UK hospitals.ParticipantsA target of 300 women with a history of chronic pelvic pain in whom a laparoscopy revealed no obvious pelvic pathology.InterventionWomen were randomised to receive 300 mg of gabapentin (which was escalated to a maximum of 2700 mg daily) or a matched placebo over a 4-week dose-escalation period, followed by 12 weeks on optimal dose. A mechanistic substudy was also undertaken, in which a subset of participants had a functional magnetic resonance imaging scan of their brain before and following 16 weeks of treatment.Main outcome measuresThe dual primary measure of the worst and average pelvic pain scores was assessed weekly by a numerical rating scale (0–10) in weeks 13–16 post randomisation. The secondary outcomes were patient-reported questionnaires, assessed physical functioning, fatigue, psychological health, sexual activity, work and productivity, and pain catastrophising. Health-care resource use, analgesic use and adverse events were also collected. The main outcome measure for the mechanistic study was brain activity at rest and in response to noxious stimuli.ResultsIn the main trial, 306 participants were randomised. The mean worst pain score was 7.1 (standard deviation 2.6) in the gabapentin group and 7.4 (standard deviation 2.2) in the placebo group (adjusted mean difference –0.20, 97.5% confidence interval –0.81 to 0.42;p = 0.47). The mean average pain score was 4.3 (standard deviation 2.3) in the gabapentin group and 4.5 (standard deviation 2.2) in the placebo group (adjusted mean difference –0.18, 97.5% confidence interval –0.71 to 0.35;p = 0.45). No significant between-group differences were observed for any secondary outcome. A higher proportion of women experienced a serious adverse event in the gabapentin group than in the placebo group (10/153 vs. 3/153;p = 0.04). Dizziness, drowsiness and visual disturbances were more common in the gabapentin group than in the placebo group. In the mechanistic study, 45 participants had a baseline functional magnetic resonance imaging scan of their brain, with 25 participants returning for a scan at the end of treatment. Gabapentin significantly decreased evoked activity in the anterior cingulate cortex and cuneus. Change in anterior cingulate cortex activity after treatment related to improvement on the pain interference scale, and baseline activation of this region predicted response to treatment.ConclusionsGabapentin did not reduce pain and did not improve other outcomes compared with placebo over 16 weeks. Serious adverse effects were significantly higher in the gabapentin group than in the placebo group. Gabapentin reduces evoked activity in the anterior cingulate cortex, with changes of activity in this region tracking reported pain, and baseline activity predicting response to treatment.LimitationsPrimary outcome data were unavailable in 62 and 60 women for the average and worst numerical rating scale pain scores, respectively. A sensitivity analysis using imputation methods did not change the result.Future workClinical trials to investigate other pharmacological interventions (monotherapy vs. combination therapy), physiotherapy and cognitive–behavioural therapy to treat women with chronic pelvic pain are needed.Trial registrationCurrent Controlled Trials ISRCTN77451762 and EudraCT 2014-005035-13.FundingThis project was funded by the Efficacy and Mechanism Evaluation (EME) programme, a Medical Research Council and National Institute for Health Research (NIHR) partnership. This will be published in full inEfficacy and Mechanism Evaluation; Vol. 7, No. 7. See the NIHR Journals Library website for further project information.