OBJECTIVE Large rectoceles have been associated with symptoms of impaired rectal evacuation, often leading to rectocele repair. However, these symptoms, or the anatomic abnormality, may be caused, at least in part, by a primary disturbance of rectoanal coordination. This study aimed to determine the efficacy of biofeedback therapy in such patients. METHODS Thirty-two female patients (median age, 52 years) complaining of impaired rectal evacuation and with a rectocele greater than 2 cm at proctography were evaluated by structured questionnaire before, immediately after treatment, and at follow-up. Physiologic and proctographic findings were related to outcome. RESULTS Immediate results were available in 32 patients and medium-term follow-up (median, 10; range, 2-30 months) in 25 patients. At follow-up 14 (56 percent) patients felt a little and 4 (16 percent) patients felt major improvement in symptoms, including 3 (12 percent) with complete symptom relief. Immediately after biofeedback there was a modest reduction in need to strain (from 72 to 50 percent), feeling of incomplete evacuation (from 78 to 59 percent), need to assist defecation digitally (from 84 to 63 percent), and need to use an evacuant (from 47 to 28 percent), and this was maintained at follow-up. Bowel frequency was significantly normalized at follow-up (P = 0.02). Pretreatment presence of symptoms of digitally assisting defecation, pelvic floor incoordination, and proctographic rectocele size and contrast trapping, did not predict outcome. CONCLUSIONS Behavioral therapy, including biofeedback, leads to major symptom relief in a minority, and partial symptom relief in a majority, of patients with a feeling of impaired defecation and the presence of a large rectocele. Residual symptoms are common. Biofeedback may be a reasonable first-line treatment for such patients.