During the period of 1980 to 1986, 49 total abdominal colectomies were performed at the authors' institution for lower gastrointestinal bleeding. The overall mortality was 13 of 49 (27%). In the elective/urgent group, mortality was one of 14 (7%); in the emergency with ileostomy group it was two of two (100%); and in the emergency with anastomosis group it was ten of 33 (30%). Morbidity and mortality in this latter group were affected by age [mortality three of 15 (21%) for age less than 70 vs seven of 18 (37%) for age greater than or equal to 70] and the number of units of blood needed preoperatively and intraoperatively. There was no effect on outcome from type of anastomosis (stapled vs hand-sewn), choice of antibiotics, degree of underlying illness, or day of operation following admission. Thirteen patients had less than 10 units of blood transfused; one (7%) died and there was one complication. Conversely, 20 patients had ten or more blood transfusions and nine (45%) died (P = 0.05 vs former group). This latter group also had 16 major complications, including five anastomotic leaks, three intra-abdominal abscesses, and three myocardial infarctions. Total abdominal colectomy when done as an emergency for colonic hemorrhage is associated with excessive morbidity and mortality rates. An important factor contributing to morbidity and mortality that may be controllable is the amount of blood loss. Should total abdominal colectomy be the surgeon's operation of choice, it should be performed prior to 10 units of blood being needed. If this limit is passed, strong consideration should be given to performance of an ileostomy.