Synthetic sling or artificial urinary sphincter for men with urodynamic stress incontinence after prostate surgery: the MASTER non-inferiority RCT

Background: Stress urinary incontinence is common in men after prostate surgery and can be difficult to improve. Implantation of an artificial urinary sphincter is the most common surgical procedure for persistent stress urinary incontinence, but it requires specialist surgical skills, and revisions may be necessary. In addition, the sphincter is relatively expensive and its operation requires adequate patient dexterity. New surgical approaches include the male synthetic sling, which is emerging as a possible alternative. However, robust comparable data, derived from randomised controlled trials, on the relative safety and efficacy of the male synthetic sling and the artificial urinary sphincter are lacking. Objective: We aimed to compare the clinical effectiveness and cost-effectiveness of the male synthetic sling with those of the artificial urinary sphincter surgery in men with persistent stress urinary incontinence after prostate surgery. Design: This was a multicentre, non-inferiority randomised controlled trial, with a parallel non-randomised cohort and embedded qualitative component. Randomised controlled trial allocation was carried out by remote web-based randomisation (1: 1), minimised on previous prostate surgery (radical prostatectomy or transurethral resection of the prostate), radiotherapy (or not, in relation to prostate surgery) and centre. Surgeons and participants were not blind to the treatment received. Non-randomised cohort allocation was participant and/or surgeon preference. Setting: The trial was set in 28 UK urological centres in the NHS. Participants: Participants were men with urodynamic stress incontinence after prostate surgery for whom surgery was deemed appropriate. Exclusion criteria included previous sling or artificial urinary sphincter surgery, unresolved bladder neck contracture or urethral stricture after prostate surgery, and an inability to give informed consent or complete trial documentation. Interventions: We compared male synthetic sling with artificial urinary sphincter. Main outcome measures: The clinical primary outcome measure was men's reports of continence (assessed from questions 3 and 4 of the International Consultation on Incontinence Questionnaire -Urinary Incontinence Short Form) at 12 months post randomisation (with a non-inferiority margin of 15%). The primary economic outcome was cost-effectiveness (assessed as the incremental cost per quality-adjusted life-year at 24 months post randomisation). Results: In total, 380 men were included in the randomised controlled trial (n = 190 in each group), and 99 out of 100 men were included in the non-randomised cohort. In terms of continence, the male sling was non-inferior to the artificial urinary sphincter (intention-to-treat estimated absolute risk difference -0.034, 95% confidence interval -0.117 to 0.048; non-inferiority p = 0.003), indicating a lower success rate in those randomised to receive a sling, but with a confidence interval excluding the non-inferiority margin of -15%. In both groups, treatment resulted in a reduction in incontinence symptoms (as measured by the International Consultation on Incontinence Questionnaire-Urinary Incontinence Short Form). Between baseline and 12 months' follow-up, the International Consultation on Incontinence Questionnaire-Urinary Incontinence Short Form score fell from 16.1 to 8.7 in the male sling group and from 16.4 to 7.5 in the artificial urinary sphincter group (mean difference for the time point at 12 months 1.30, 95% confidence interval 0.11 to 2.49; p = 0.032). The number of serious adverse events was small (male sling group, n = 8; artificial urinary sphincter group, n = 15; one man in the artificial urinary sphincter group experienced three serious adverse events). Quality-of-life scores improved and satisfaction was high in both groups. Secondary outcomes that showed statistically significant differences favoured the artificial urinary sphincter over the male sling. Outcomes of the non-randomised cohort were similar. The male sling cost less than the artificial sphincter but was associated with a smaller quality-adjusted life-year gain. The incremental cost-effectiveness ratio for male slings compared with an artificial urinary sphincter suggests that there is a cost saving of 425,870 pound for each quality-adjusted life-year lost. The probability that slings would be cost-effective at a 30,000 pound willingness-to-pay threshold for a quality-adjusted life-year was 99%. Limitations: Follow-up beyond 24 months is not available. More specific surgical/device-related pain outcomes were not included. Conclusions: Continence rates improved from baseline, with the male sling non-inferior to the artificial urinary sphincter. Symptoms and quality of life significantly improved in both groups. Men were generally satisfied with both procedures. Overall, secondary and post hoc analyses favoured the artificial urinary sphincter over the male sling. Future work: Participant reports of any further surgery, satisfaction and quality of life at 5-year follow-up will inform longer-term outcomes. Administration of an additional pain questionnaire would provide further information on pain levels after both surgeries. Trial registration: This trial is registered as ISRCTN49212975.