Anastomosis to the rectum: operative experience

Four hundred sixty-six consecutive procedures involving anastomosis to the rectum were performed between March 1969 and December 1982. Three hundred ninety-six (85 percent) were stapled anastomoses and 70 (15 percent) were hand-sutured anastomoses. The stapled anastomoses were constructed using the GIA or EEA instrument, some of the latter utilizing a pull-through technique. The hand-sutured anastomoses were constructed in the pelvic space, or externally as a staged pull-through procedure. A diverting stoma was constructed in all 14 staged pull-through procedures, in 47 of 56 (84 percent) conventional hand-sutured anastomoses, and in 38 of 396 (10 percent) stapled anastomoses. While the majority of very low anastomoses (0 to 5 cm from the dentate line) were stapled, 13 conventional hand-sutured anastomoses and all 14 of the staged pull-through procedures were constructed at this level. One patient (0.2 percent) died as the result of an anastomotic complication. Twelve patients (2.5 percent) had anastomotic complications requiring reoperation. The reoperation rate for stapled anastomoses was six of 396 (1.5 percent). For hand-sutured anastomoses, the reoperation rate was six of 70 (8.6 percent). The results show that, for anastomosis to the rectum, stapling instruments are at least as good as hand-suturing. Both stapling techniques and hand-suturing techniques provide the surgeon the capacity to construct safely very low anastomoses. A temporary, diverting stoma is required much less frequently with stapled than with hand-sutured anastomoses. The need for a permanent colostomy should be determined by the stage and level of disease, the systemic health of the patient, and the patient's anatomy, rather than by the selection of anastomotic technique.