PURPOSE: Pelvic-floor rehabilitation does not provide the same degree of relief in all fecal incontinent patients. We aimed at studying prospectively the ability of tests to predict the outcome of pelvic-floor rehabilitation in patients with fecal incontinence. MATERIALS AND METHODS: Two hundred fifty consecutive patients (228 women) underwent medical history and a standardized series of tests, including physical examination, anal manometry, pudendal nerve latency testing, anal sensitivity testing, rectal capacity measurement, defecography, endoanal sonography, and endoanal magnetic resonance imaging. Subsequently, patients were referred for pelvic-floor rehabilitation. Outcome of pelvic-floor rehabilitation was quantified by the Vaizey incontinence score. Linear regression analyses were used to identify candidate predictors and to construct a multivariable prediction model for the posttreatment Vaizey score. RESULTS: After pelvic-floor rehabilitation, the mean baseline Vaizey score (18, SD+/-3) was reduced with 3.2 points (p<0.001). In addition to the baseline Vaizey score, three elements from medical history were significantly associated with the posttreatment Vaizey score (presence of passive incontinence, thin stool consistency, primary repair of a rupture after vaginal delivery at childbed) (R2, 0.18). The predictive value was significantly but marginally improved by adding the following test results: perineal and/or perianal scar tissue (physical examination), and maximal squeeze pressure (anal manometry; R2, 0.20; p=0.05). CONCLUSION: Additional tests have a limited role in predicting success of pelvic-floor rehabilitation in patients with fecal incontinence.