Abstract
BACKGROUND: In critically ill patients, Candida spp. can often be identified in pulmonary samples. The impact of prompt antifungal therapy in these patients is unknown. METHODS: In this retrospective study, 500 adult patients with pulmonary Candida spp. colonization admitted to the intensive care unit (ICU) between 2010 and 2012 were included. The patients were analyzed according to whether or not they received antifungal therapy, which was administered at the discretion of the attending physician. Logistic regression analysis was performed to investigate the impact of antifungal therapy on hospital mortality and new onset of ventilator-associated pneumonia. In a stepwise backward elimination, the impact of age, cancer as an underlying disease, Simplified Acute Physiology Score (SAPS) II, and Sequential Organ Failure Assessment (SOFA) score were considered. RESULTS: After excluding 178 patients with multifocal Candida spp., isolated pulmonary Candida spp. colonization was found in 322 patients (cohort 1). Pre-existing pneumonia was found in 147/322 patients. Out of the remaining 175 patients (cohort 2), 44 patients received any antifungal therapy, and 131 were defined as the control group. Patients who received antifungal therapy had higher hospital mortality (50 vs. 30 %, p = 0.02) and pneumonia rates (47.7 vs. 16.8 %; p < 0.001) than those who did not. In Cox regression analysis, antifungal therapy was not independently associated with favorable outcome (mortality: odds ratio 0.854 (95 % CI 0.467-1.561); new pneumonia: 1.048 (0.536-2.046)), but SAPS II and SOFA score were significantly (p < 0.05) independent covariates for worse outcome. CONCLUSIONS: In critically ill patients with pulmonary Candida spp. colonization, antifungal therapy may not have an impact on the incidence of new pneumonia or in-hospital mortality after adjustment for confounders.