Characterizing the prognostic utility of isolating yeast in surgical site infections and subsequent use of antifungal therapies

Abstract

Background: Whether the isolation of yeast from surgical site infections (SSIs) affects outcomes is unclear. We hypothesized that for SSI, yeasts are nonpathogenic and that neither the isolation of yeast nor the use of antifungal agents affects outcomes. Methods: Incisional SSIs from general surgery patients at a single institution between 1997 and 2017 with positive cultures were analyzed, categorized by the presence of yeast. Demographics and in-hospital mortality were compared by Student's t-test and Chi-square analysis. Independent predictors of isolation of yeast and in-hospital mortality were determined by multivariate logistic regression analysis (MV). Results: In total, 977 infections with positive cultures were identified: 190 (19.4%) with yeast and 787 (80.6%) without. By univariate analysis, cultures positive for yeast were associated with a higher severity of illness/APACHE II score (15.6 ± 0.5 versus 11.6 ± 0.2; P < 0.0001), diagnosis in the ICU (83/190, 43.6% versus 154/787, 19.5%; P < 0.0001), age (55.2 ± 1.1 versus 52.4 ± 0.5; P = 0.015), and female sex (105/190, 55.3% versus 362/787, 46.0%; P = 0.027). Independent predictors of isolation of yeast included sex, APACHE II Score, and diagnosis after discharge from index procedure, and for mortality, age, APACHE II, diagnosis after discharge, and receiving antifungal treatment. Isolation of yeast was not associated with mortality (P = 0.12). For fungal SSI, antifungal treatment was not associated with decreased mortality. Conclusion: Isolation of yeast from incisional SSI is not associated with mortality, and the use of antifungal agent is associated with higher mortality. Routine fungal cultures of SSI are not warranted.