Fulminant Cdifficile colitis is an increasing surgical disease. Byrn et al performed a single-institution retrospective review of 73 C difficile colitis cases requiring colectomy from 1994 to 2005. The main research finding was identification of vasopressor requirement, mental status changes, and treatment length as independent mortality predictors.
In contrast to previous retrospective reviews of colectomy for C difficile colitis using only descriptive statistics,
Byrn et al identify associations of in-hospital mortality using multivariate regression. These findings contribute to our clinical understanding of this disease; however, we should recognize potential limitations.
First, the small sample size of 73 patients may limit the ability of Byrn et al to identify additional significant results. For example,
some nonsignificant mortality predictors may actually predict mortality but generate nonsignificant P values because of the small study cohort (ie, type II error). Second, Byrn et al identify mortality predictors generally associated with severe illness (eg, vasopressor requirement), which may not be as clinically useful for physicians caring for patients with C difficile colitis. It might be more helpful to identify factors predicting colitis severity or likelihood of medical treatment failure in order to prevent fulminant colitis progression. Earlier identification of patients with increased likelihood of fulminant colitis may allow earlier operative intervention and improve outcomes. Perhaps Byrn et al could evaluate factors predictive of colitis severity or medical treatment failure in the 5645 patients with C difficile colitis at their institution who did not require colectomy. It will be difficult to decrease mortality from fulminant colitis until such research identifies predictors of treatment failure or methods to promptly diagnose medically refractory disease.
McGory ML, Ko CY. Predictors of Mortality After Colectomy for Fulminant Clostridium difficile Colitis—Invited Critique. Arch Surg. 2008;143(2):155. doi:10.1001/archsurg.2007.47