Background Severe fungal pneumonia with sepsis carries high mortality. Early and accurate prognosis is essential for improving outcomes. We developed and validated a nomogram based on immune-inflammation-nutrition indicators to predict 28-day mortality in these patients. Methods We conducted a retrospective, exploratory cohort study analyzing 486 sepsis patients with severe fungal pneumonia, randomly splitting them into training (n=365) and validation (n=121) sets. Using LASSO regression and multivariate logistic regression, we identified independent predictors from clinical, laboratory, and immune-inflammation-nutrition data. We built a nomogram and evaluated its discrimination, calibration, and clinical utility with ROC curves, calibration plots, and decision curve analysis (DCA). Results Eight independent predictors entered the nomogram: respiratory failure (RF), chronic obstructive pulmonary disease (COPD), prothrombin time (PT), glucose, blood urea nitrogen (BUN), white blood cell count (WBC), albumin-to-alkaline phosphatase ratio (AAPR), and lactate-albumin ratio (LAR). The nomogram achieved AUCs of 0.884 (training) and 0.834 (validation), outperforming the SOFA score (0.740 and 0.691). Furthermore, temporal validation using an independent later cohort achieved an AUC of 0.875. Calibration curves showed good agreement between predicted and observed outcomes. DCA confirmed clinical utility across a wide range of threshold probabilities. Conclusion We developed and internally validated a practical nomogram that integrates clinical variables with immune-inflammation-nutrition indicators to predict 28-day mortality in sepsis patients with severe fungal pneumonia. This tool may help clinicians with early risk stratification and individualized treatment decisions.
Du et al. (Tue,) studied this question.