Development and validation of a risk score for predicting invasive fungal infectious in an intensive care unit.

Delay in the initiation of appropriate antifungal therapy is associated with substantial morbidity and mortality. The aim of this study was to derive a risk score system for the development of invasive fungal infections in an intensive care unit (ICU). We retrospectively evaluated 1812 patients who stayed in the ICU for > or = 4 days, used univariate and multivariable logistic regression to identify potential risk factors associated with invasive fungal infections (IFI), and created a risk score system. Seven variables were identified as important predictors of ICU-IFI (diabetes mellitus, gastrointestinal surgery, hematological malignancies, mechanical ventilation > or = 2 days, central venous catheter, total parenteral nutrition, broad-spectrum antibiotic use > or = 4 days). The area under the receiver operating characteristic curve was 0.807 and was similar in the validation set. The percentages of patients with ICU-IFI in the low, intermediate, and high risk groups were 5.2%, 31.6%, and 63.2% in the derivation cohort and 4.2%, 30.1%, and 66.7% in the validation cohort, respectively. A new risk score was developed to predict ICU-IFI and was validated in an independent cohort. The new risk score may help clinicians identify patients who are at high risk of developing ICU-IFI and increase their odds of survival.