&NA;Morbidity and treatment ofClostridium difficilecolitis (CDC) continue to be controversial. Some claim minimum morbidity, which may be a function of differences in patient population and/or bacterial virulence.METHODS:To evaluate the effect of CDC in the critically ill, we retrospectively reviewed the records of 59 intensive care unit patients with CDC who were diagnosed by fecal toxin assays or clinical evidence of pseudomembranous colitis from January 1991 to October 1994. Symptoms, signs, antibiotic regimens, diagnostic tests, Acute Physiology and Chronic Health Evaluation II (APACHE II) scores, morbidity, and mortality were analyzed, and results of surgical treatment were compared with the literature.RESULTS:Mean age was 66.4 (17‐95) years, with a male to female ratio of 1.8:1. First treatment was metronidazole by mouth in 15 patients (25.4 percent), vancomycin by mouth in 30 patients (50.8 percent), sequential by mouth vancomycin/metronidazole in 3 patients (5.1 percent), and intravenous metronidazole in 5 patients (8.5 percent). Six patients had no medical therapy before surgery or discharge. Ten patients (17 percent) had recurrence and 12 (20.3 percent) required surgery for progressive toxicity or peritonitis. Of three patients who were initially treated by diverting stomas, one died and two required total colectomy (TAC). Two underwent partial resection (1 that was nearly a total colectomy), and seven others had a TAC. Surgical patients had worse mean APACHE II scores at diagnosis (24.4vs.199;P<0.001). Thirty‐day mortality in surgical patients was 41.7vs.14.7 percent in medical patients(P<0.5).CONLUSION:Twenty percent of critically ill patients with CDC required operation. TAC and diversion appeared to be more effective surgical treatments than diversion alone.