Colectomy for fulminant Clostridium difficile colitis: predictors of mortality

The purpose of this study was to define clinical and radiographic variables associated with postoperative mortality after urgent colectomy for fulminant Clostridium difficile colitis. Data were obtained regarding patients undergoing colectomy for fulminant C. difficile colitis at two institutions (1997-2005). Univariate analysis of factors predicting 30-day mortality was performed using chi2 and Student's t tests. Multivariable logistic regression was done to include all variables whose P value was < 0.20. Clinical variables analyzed included: age, gender, recent operation, comorbidities, preoperative multisystem organ failure, vasopressors, symptom duration, time to surgery, serum albumin, change in serum albumin, serum creatinine, white blood cell count, and extent of colectomy. Computed tomography variables included: ascites, megacolon, and extent of colitis. Thirty-five patients (mean age 70 years, 46% male) underwent urgent colectomy for C. difficile colitis. The 30-day mortality rate was 45.7 per cent (16/35). The only clinical variable associated with mortality was preoperative multisystem organ failure (nonsurvivors 9/16 vs survivors: 4/19; P = 0.037). None of the three patients undergoing partial colectomy survived, although the difference in survival versus those undergoing subtotal colectomy was not significant. Patients with fulminant C. difficile colitis undergoing colectomy have a high mortality rate. Preoperative presence of multisystem organ failure was independently predictive of mortality.