Surgical patients with pseudomembranous colitis: factors affecting prognosis.

BACKGROUND

Although several studies have identified the factors that contribute to the development of antibiotic-associated colitis (AAC), little data are available in regard to those factors that may affect the prognosis of patients with the disease. Therefore we conducted a retrospective analysis of 201 surgical patients with AAC to identify risk factors predictive of increased morbidity or mortality.

METHODS

We conducted a review of the charts of 201 surgical patients hospitalized between Jan. 1, 1991, and June 30, 1993, in whom AAC developed. AAC was defined as the presence of diarrhea associated with a positive latex agglutination or toxin assay for Clostridium difficile. An additional 52 procedure-matched charts of patients admitted to a surgical service during the same period were also reviewed and constituted the control group. We analyzed the contribution of 21 variables to prognosis in both groups.

RESULTS

There was no difference between the two groups in the preoperative length of stay, the number of antibiotics per patient and the number of antibiotic-days, number of patients receiving preoperative bowel preparation, total parenteral nutrition, and overall mortality rate. Patients in the control group were at increased risk of death if they had a history of preexisting renal dysfunction, prolonged preoperative hospital stay, and a poor nutritional status. AAC developed 10.0 +/- 13.8 days after operation in the study group. All patients were receiving multiple antibiotics at the time of diagnosis (3.6 +/- 7.5 antibiotic), with a mean of 14.3 +/- 20.7 antibiotic-days. The overall mortality rate in the study group was 8%. In five patients (2%) toxic megacolon developed; four deaths occurred among these patients (80% mortality rate). A 25% mortality rate was directly attributable to complications of AAC. Six variables were identified as predictive of increased mortality rate: steroids, laxatives, length of preoperative stay, postoperative interval before the onset of symptoms, use of total parenteral nutrition, and abdominal distention.

CONCLUSIONS

AAC carries a significant mortality rate in surgical patients; therefore the diagnosis of AAC should be aggressively pursued and patients with the disease should be promptly treated. Patients receiving steroids, total parenteral nutrition, and multiple antibiotics in whom signs and symptoms of AAC develop late in their postoperative course, and patients with abdominal distention and marked leukocytosis, are at increased risk of dying of AAC and should be aggressively treated.