Factors predicting the response to biofeedback-assisted pelvic floor muscle training for urinary incontinence.

OBJECTIVE To evaluate the efficacy of biofeedback-assisted pelvic floor muscle training (PFMT) for urinary incontinence and to determine the patient characteristics predictive of success. METHODS Clinical and physiologic data of 86 patients with urinary incontinence who had received biofeedback-assisted PFMT were analyzed retrospectively. The clinical response was determined to be a success (requiring no more therapy) or a failure (requiring surgery or other medical therapy) at 3 months after completion of treatment. RESULTS The success group included 57% of the enrolled patients. In the univariate analysis, the following factors had P values of less than 0.20: alcohol consumption, detrusor overactivity, type of urinary incontinence, and the change in the average amplitude of tonic contraction before treatment and after the 8th session of biofeedback-assisted PFMT. In the multiple logistic regression analysis, the only independent predictive factor of a successful response was a significant change in the average tonic contraction before treatment and after the 8th session (odds ratio, 1.661; 95% confidence interval, 1.015- 2.721). CONCLUSION After biofeedback-assisted PFMT, 57% of patients with urinary incontinence required no further therapy. Increased pelvic floor muscle activity after the 8th session of PFMT predicted a successful response to treatment.