Construction and validation of a predictive nomogram model for invasive fungal infections in sepsis patients with severe pneumonia in the ICU.

BACKGROUND

Invasive fungal infections (IFI) represent a serious complication in critically ill sepsis patients with severe pneumonia, contributing to increased mortality and prolonged hospitalization. Early prediction of IFI remains challenging due to the lack of specific clinical tools. This study aimed to develop and validate a predictive nomogram for IFI risk in this high-risk population.

METHODS

A total of 1,890 sepsis patients with severe pneumonia admitted to the ICU of the primary center were retrospectively enrolled from Shaanxi Provincial People's Hospital. Patients were randomly divided into a training set (n = 1,418) and an internal validation set (n = 472). Additionally, an independent external validation cohort of 378 patients from another tertiary hospital (Xi'an No. 3 Hospital) was collected. Least Absolute Shrinkage and Selection Operator (LASSO) regression and multivariate logistic regression were used to identify independent predictors. A nomogram was constructed and evaluated for discrimination, calibration, and clinical utility using receiver operating characteristic (ROC) curves, calibration plots, and decision curve analysis (DCA).

RESULTS

The incidence of IFI was 25.8% (488/1,890). Eight independent predictors were identified: respiratory rate (RR), diabetes, respiratory failure (RF), acute-on-chronic liver failure (ACFL), prothrombin activity (PTA), D-dimer, activated partial thromboplastin time (APTT), and lactate. The nomogram demonstrated excellent discrimination, with area under the curve (AUC) values of 0.859 (training), 0.830 (internal validation), and 0.872 (external validation), outperforming the Sequential Organ Failure Assessment (SOFA) score. Calibration and DCA confirmed its clinical applicability.

CONCLUSION

We developed and validated an easy-to-use nomogram that accurately predicts the risk of IFI in sepsis patients with severe pneumonia. This tool may assist clinicians in early identification and intervention for high-risk individuals.