A systematic review and network meta-analysis comparing treatments for faecal incontinence

Although numerous treatments exist for fecal incontinence (FI), no consensus exists on the best treatment strategy. The aim was to review the literature and to compare the clinical outcomes and effectiveness of treatments available for FI.

A systematic literature review was performed, from inception to May 2018, of the following databases: MEDLINE, EMBASE, Science Citation Index Expanded, Cochrane Library. The search terms used were "faecal incontinence" and "treatment". Only randomized controlled trials (RCTs) comparing treatments for FI were considered. A Bayesian network meta-analysis was performed using the Markov chain Monte Carlo method.

Forty-seven RCTs were included comparing 37 treatments and reporting on 3748 participants. No treatment ranked best or worst with high probability for any outcome of interest. No significant difference was identified between treatments for frequency of FI per week, or in changing the resting pressure, maximum resting pressure, squeeze pressure, and maximum squeeze pressure. Radiofrequency resulted in more adverse events compared to placebo. Sacral nerve stimulation (SNS) and zinc-aluminium improved the fecal incontinence quality of life questionnaire (FIQL) lifestyle, coping, and embarrassment domains compared to placebo. Transcutaneous posterior tibial nerve stimulation (TPTNS) improved the FIQL embarrassment domain compared to placebo. Autologous myoblasts and zinc-aluminium improved the FIQL depression domain compared to placebo. SNS, artificial bowel sphincter (ABS), and zinc-aluminium significantly improved incontinence scores compared to placebo. Injection of non-animal stabilized hyaluronic acid/dextranomer (NASHA/Dx) resulted in more patients with ≥50% reduction in FI episodes compared to placebo.

SNS, ABS, TPTNS, NASHA/Dx, zinc-aluminium, and autologous myoblasts resulted in isolated improvements in specific outcomes of interest. No difference was identified in incontinence episodes, no treatment ranked best persistently or persistently improved outcomes, and many included treatments did not significantly benefit patients compared to placebo. Large multicentre RCTs with long-term follow-up and standardized inclusion criteria and outcome measures are needed.