Introduction: Between 1985 and 1996, 190 patients underwent a low anterior rectal resection with coloanal anastomosis for adenocarcinoma of the lower one-third of the rectum.
Methods: This article reports on 31 (17 males) of these patients with a very low localization of the tumor (distal tumor margin 1.3 +/- 0.9 cm above the dentate line). If the function of the sphincter was acceptable and we could exclude tumor infiltration into the sphincter through endosonography, we relocated the resection plane distally into the intersphincteric region to attain an acceptable margin of safety. In all of these cases, it was impossible for us to perform the usual surgical procedure of a mechanical anastomosis by means of a circular stapler. After intersphincteric rectal resection, the anastomosis was handsewn, using interrupted sutures from the perineal approach, 2.5 to 3 cm above the anal verge, implementing Parks' retractor. A protective stoma was performed in all cases. All data were documented prospectively.
Complications: Postoperative mortality was 0 percent. Postoperatively, none of the patients showed an indication for relaparotomy. The leakage rate was 48 percent. Only 16 percent later needed additional surgery for anastomotic strictures or for rectovaginal fistulas. Long-term observations showed that the anastomosis healed well in 27 patients (87.1 percent). Four patients (12.9 percent) decided to have a terminal colostomy performed (anastomotic stricture, 3 patients; anorectal incontinence, 1 patient).
Follow-up: During the follow-up period of 6.8 +/- 3.7 years, six patients (19.4 percent) developed a tumor progression (9.7 percent local recurrences and 12.9 percent distant spread). The five-year survival rate was 79 percent (Dukes A, 100 percent (n = 18); Dukes B, 67 percent (n = 4); and Dukes C, 44 percent (n = 9)). Continence: One-third of patients developed anorectal incontinence for liquid (29.6 percent) or solid stool (3.7 percent). Average stool frequency was 3.3 times per day. Resting pressure decreased significantly by 29 percent (preoperative, 105 +/- 37 cm H2O and postoperative, 75 +/- 19 cm H2O; P < 0.05), whereas squeeze pressure did not change.
Conclusion: In selected patients with tumors close to the dentate line, an intersphincteric resection of the rectum may help to avoid an abdominoperineal excision of the rectum with a terminal stoma, without any curtailment of oncologic standards. A protective stoma for three months is advantageous.