A multi-centre randomised controlled trial of the efficacy and mechanism of action of gabapentin for the management of chronic pelvic pain in women

HYPOTHESIS: (i) Treatment with gabapentin alleviates pain, and improves physical and emotional functioning, in women with chronic pelvic pain (CPP) in the absence of any pelvic pathology; and (ii) Brain activity of women with CPP associated with an absence of obvious pelvic pathology can be altered by treatment with gabapentin and can predict response to treatment. AIM: To determine the efficacy and mechanism of action of gabapentin in the management of CPP in women. DESIGN: Double blind placebo controlled randomised multi-centre clinical trial informed by our pilot study (BMJ Open 2012 2[3] pii: e001297). POPULATION: Women with CPP in whom laparoscopy reveals no obvious pelvic pathology which could be responsible for pain. INTERVENTIONS: Gabapentin or placebo. PRIMARY OUTCOME: Dual measures of worst and average pelvic pain scores after 12 weeks on maximum tolerated dose (MTD), assessed by a numerical rating scale (NRS) over last 4 weeks of treatment. SECONDARY OUTCOMES: (1) Physical/emotional functioning (2) Central pain processing. ASSESSMENTS: NRS, SF-12 quality of life, Brief Pain Inventory, Chalder Fatigue Scale, General Health Questionnaire, Work and Productivity Activity Impairment, Pain catastrophising, and Sexual Activity Questionnaires. All outcomes will be completed at baseline and after 12 weeks on MTD. A subset of patients (recruited in Edinburgh) will undergo fMRI to assess brain activity pre- and during trial treatment. SAMPLE SIZE AND ANALYSIS: Data from our pilot has shown pain scores to be broadly normally distributed (SD 2-2.5). If the SD is at the lower end of these estimates, 86 patients in each group would be required to have 90% power (p=0.05) to detect a difference of 1 point. If the SD is at the higher end, we could detect the same size of difference with 80% power (p=0.05) with 100 patients in each group. To account for any increase in the risk of type I error that may be associated with having co-primary outcome measures and for an expected 20% loss to follow-up, we will randomise 300 patients. The analysis will primarily be intention-to-treat. We will use a linear regression model to estimate differences in NRS between groups after 12 weeks on MTD, including baseline score as a covariate. Data from the other continuous measures will be analysed in a similar fashion as the primary measure. Likert responses (pt satisfaction) will be analysed using standard methods (tests for trend, absolute and relative risks). Tests for interaction will be performed for subgroup variables by including interaction parameters in the linear models. Sensitivity analyses will be undertaken to test the robustness of conclusions. Based on previous studies and statistical techniques used to analyse the data (www.fmrib.ox.ac.uk/fsl), we estimate that we need to recruit 50 women (25 per group) for the fMRI substudy to identify significant differences using mixed effect analyses. ECONOMIC BENEFITS: CPP is the subject of 20% of gynaecological consults and causes a 45% reduction in work productivity. The determination of the true efficacy of gabapentin will have major socioeconomic benefits. PROJECT TIMETABLE: Before grant is activated, ethics and clinical trial authorisation will be obtained. After 6 months to set-up all centres, recruitment of 300 women will take place over 24 months (8 centres, each recruiting 1-2 women per month). Follow-up will be complete by month 36. Analysis/write-up will occur in final 6 months.