Randomized controlled trial shows biofeedback to be superior to pelvic floor exercises for fecal incontinence

Abstract

Purpose: This study aimed to compare manometric biofeedback with pelvic floor exercises for the treatment of fecal incontinence in a randomized controlled trial controlling for nonspecific treatment effects.

Methods: After excluding patients who were adequately treated with medication, education, and behavioral strategies (21%), 108 patients (83 females; average age, 59.6 years) underwent either pelvic floor exercises alone (n = 63) or manometric biofeedback plus pelvic floor exercises (n = 45). Patients in both groups were taught behavioral strategies to avoid incontinence.

Results: At three-month follow-up, biofeedback patients had significantly greater reductions on the Fecal Incontinence Severity Index (P = 0.01) and fewer days with fecal incontinence (P = 0.083). Biofeedback training increased anal canal squeeze pressure more than pelvic floor exercises did (P = 0.014) and with less abdominal tension during squeeze (P = 0.001). Three months after training 76% of patients treated with biofeedback vs. 41% patients treated with pelvic floor exercises (chi-squared = 12.5, P < 0.001) reported adequate relief. Before treatment, the groups did not differ on demographic, physiologic, or psychologic variables, symptom severity, duration of illness, quality-of-life impact, or expectation of benefit. At 12-month follow-up, biofeedback patients continued to show significantly greater reduction in Fecal Incontinence Severity Index scores (F = 4.83, P = 0.03), and more patients continued to report adequate relief (chi-squared = 3.64, P = 0.056).

Conclusion: This investigation provides definitive support for the efficacy of biofeedback. Biofeedback training resulted in greater reductions in fecal incontinence severity and days with fecal incontinence. Biofeedback was also more effective than pelvic floor exercises alone in producing adequate relief of fecal incontinence symptoms in patients for whom conservative medical management had failed.

Figure 1. Schematic of study flowchart

Figure 2. Consort guidelines flow-chart

Figure 3

Primary Outcome Measure using a Chi Squared analysis comparing the proportion of subjects reporting adequate relief at 3-month follow-up in the biofeedback group compared to the diazepam and to the placebo groups.

Figure 4

Pelvic floor electromyography values (microvolts) during simulated attempts to defecate (baseline electromyography = white, 3-month follow-up electromyography = black).