2113. Evaluation of Empiric Antifungal Therapy in Critically Ill Patients with Liver Disease, Sepsis, and No Evidence of Active Fungal Infection

BACKGROUND: While candidemia is uncommon in the immunocompetent, critically ill population, it is associated with longer lengths of stay (LOS), higher cost, and higher mortality. In critically ill patients with liver disease and sepsis of unknown origin, antifungals (AF) are commonly used empirically. Recent studies suggest that this practice may not improve clinical outcomes but had little representation of patients with liver disease. This study aims to evaluate clinical outcomes of critically ill patients with liver disease, sepsis, and no evidence of active fungal infection who received empiric AF vs. those who did not. METHODS: This was a single-center, retrospective review of adults with liver disease and sepsis, identified by ICD-10 codes, who were discharged from the intensive care unit (ICU) between October 1, 2015 and December 31, 2018. Patients with neutropenia, marrow or organ transplant, HIV infection, systemic immunosuppressants, or fungal infection at sepsis onset were excluded. The primary outcome was inpatient mortality. Secondary outcomes included ICU LOS, total LOS, and development of fungal bloodstream infection (BSI) > 48 hours after sepsis onset. Fisher’s exact and Wilcoxon rank-sum tests were used to compare baseline characteristics. Multivariable logistic regression models were used to compare outcomes. Model covariates were variables with P-values < 0.2 in univariate analysis. RESULTS: A total of 119 patients were included with 92 receiving empiric AF (micafungin or fluconazole) and 27 receiving no AF. Patients receiving empiric AF were more likely to have hepatic disease upon admission and less likely to have a bacterial infection. Both groups were similar in intubation and vasopressor requirements, febrile episodes, and Candida score. Unadjusted inpatient mortality for empiric vs. no AF was 70.4% vs. 70.7%. Unadjusted ICU LOS, total LOS, and development of a fungal BSI were 10 vs. 11 days, 19 vs. 19 days, and 63.0% vs. 2.2% (P < 0.001). In multivariable models, there was no difference in inpatient mortality between groups (OR 1.20, 95% CI 0.77–1.63). CONCLUSION: In critically ill patients with liver disease, sepsis, and no evidence of active fungal infection, receipt of empiric antifungal therapy did not improve inpatient mortality, ICU LOS, or total LOS but did reduce fungal BSI. DISCLOSURES: All authors: No reported disclosures.