One of the more perplexing problems in the management of disease in patients with intra-abdominal infections is whether to provide empirical anti-infective therapy for the wide range of microorganisms that are commonly encountered in such patients. This debate is usually phrased in terms of the pathogenicity of specific microorganisms. A retrospective study1 of patients with postoperative peritonitis found that empirical antibiotic therapy that was not active against all of the bacterial species subsequently shown to be present resulted in higher failure rates, which were defined as either mortality or need for reoperation. This conclusion was supported by retrospective analyses2 of other infections. However, the pathogenicity of Candida species has been debated, particularly when Candida has been identified in community-acquired infections. The study by Dupont and colleagues, which looked at correlates of mortality by using multiple logistic regression, found that results of direct (microscopic) examination of peritoneal fluid that was positive for Candida were an independent predictor of failure in both community-acquired and postoperative infections. It is likely that the presence of Candida on direct examination is a rough measure of organism density. Because of a high incidence of antifungal therapy in this population, Dupont et al were unable to address the role of empirical antifungal therapy. The risks of retrospective studies, including missing data, the impact of unrecognized treatment practices, and publication bias, prevent firm conclusions. Nonetheless, the results of this study add to the burden of evidence that Candida is indeed an invasive pathogen and support the use of direct microscopic examination of peritoneal fluid as a diagnostic adjunct in this critically ill patient population. Whether such testing will alter treatment failure rates remains to be determined.
Solomkin JS. Predictive Factors of Mortality Due to Polymicrobial Peritonitis With Candida Isolation in Peritoneal Fluid in Critically Ill Patients—Invited Critique. Arch Surg. 2002;137(12):1347. doi:10.1001/archsurg.137.12.1347