대한대장항문학회

Purpose: The aim of this study was to identify the prognostic factors associated with poor outcome of biofeedback therapy. Methods: One hundred thirty-seven (137) constipated patients with pelvic outlet obstruction (median age 49 years) had more than one biofeedback session after defecography. Follow-up data (mean follow-up: 14 months; range: 2∼37 months) were obtained in 114 patients. Any differences in demographics, clinical symptoms, and parameters of an anorectal physiological study were evaluated between the success group and the failure group. Results: At follow-up, 80 (70 percent) patients felt improvement in symptoms, but 34 (30 percent) patients did not. Pre-biofeedback presence of symptoms of difficult defecation predict poor outcome (88 vs. 69 percent for failure vs. success, P＜0.05). The positive and the negative predictive values of difficult defecation for poor outcome were 35 percent and 86 percent, respectively. A negative mean pressure change on pre-biofeedback anal manometry was related to a poor outcome (65 vs. 26 percent for failure vs. success, P＜0.001). The positive and the negative predictive values of negative mean pressure change for poor outcome were 51 percent and 83 percent, respectively. A negative electrical current change on pre-biofeedback anal electromyography was related to a poor outcome (23 vs. 9 percent for failure vs. success, P＜0.05). The positive and the negative predictive values of negative electrical-current change for poor outcome were 53 percent and 74 percent, respectively. Conclusions: Difficult defecation, negative mean pressure change in pre-biofeedback anal manometry, and negative electrical current change in pre-biofeedback anal electromyography were predictors associated with poor outcome of biofeedback therapy for constipated patients with pelvic outlet obstruction. J Korean Soc Coloproctol 2008;24:313-321