OBJECTIVE Transanal excision of rectal villous adenomas is a widely used surgical technique, because it is a one-step procedure, requiring no sophisticated instrumentation, and allowing complete histologic analysis of the excised tumor. Therefore, it ranks alongside radical surgery and palliative destructive procedures, but its results are highly variable in the published series. This discrepancy may be explained by the variable completeness of tumor excision because of potential dissection difficulties. Because intraoperative exposure may be a major limiting factor, one of us (JF) has developed a tractable cutaneomucous flap procedure to lower the rectal tumor to the anal verge, where control of the dissection line is easier. This retrospective review of consecutive patients operated on during ten-year period reports long-term results after transanal excision for large rectal villous adenomas with the tractable flap technique. METHODS From 1978 to 1988, 207 consecutive patients (100 males), mean age 68 (range, 24-90) years, were operated on for an apparently benign villous rectal adenoma. Twenty-one patients (10 percent) were referred after failure of previous treatments: 11 endoscopic, 8 surgical, 1 laser, 1 radiotherapy. Mean distance of lower tumor edge from anal margin was 5.6 (range, 0-13) cm and was <10 cm in 82 percent. RESULTS Three patients (1.5 percent), including one with a Tis carcinoma, underwent a secondary treatment for immediate gross failure of resection: one further local excision and two palliative laser destructions. Immediate postoperative course was uneventful for 96 percent; there was one death from perineal gangrenous infection, four cases of hemorrhage, and three urinary retentions. Subsequently one case of transient fecal incontinence and 11 medically managed stenoses were noted. Mean size of resected tumor was 5.4 (range, 1-17) cm. Deep excision margins concerned the rectal muscular layers in 199 patients (96 percent) and perirectal fat in 8 (4 percent). Specimen margins were negative for cancer in 175 (85 percent) and positive or unknown in 32 cases. Histologic evaluation demonstrated in situ cancer in 28 (14 percent) and invasive carcinoma in 9 (4 percent). In three patients (1 percent), two abdominoperineal resections were immediately performed (one T2 with a mucinous contingent, one T3) and one adjuvant radiotherapy (one undifferentiated T2). Four patients (2 percent) did not return for postoperative evaluation. For the remaining 198 patients, mean follow-up was 74 +/- 34 (median, 75; range, 1-168) months. Forty-four died from unrelated causes. Recurrence occurred in seven (3.6 percent) and was malignant in two, who subsequently died. Specific recurrence-free probability was 99.5 percent at one year, 96 percent at five years, and 95 percent at ten years. A lesion size >6 cm (10 vs. 1 percent for smaller tumors) and the presence of an invasive carcinoma (20 vs. 3 percent without invasive carcinoma) were significantly associated with an increased probability of recurrence at five years. CONCLUSIONS Providing that adequate intraoperative exposure is obtained and advanced malignant tumors receive immediate secondary treatment, transanal resection of clinically benign, large rectal villous adenomas is safe and effective. It is an alternative to rectal resection, which exposes the patient to potentially adverse effects, and also to destructive procedures, which preclude any histologic evaluation of the tumor.