Prognostic Nutritional Index as a Novel Biomarker for Predicting Prognosis in Sepsis‐Associated Encephalopathy: A Multicenter Retrospective Cohort Study

Background Sepsis‐associated encephalopathy (SAE) has a high mortality rate with limited prognostic biomarkers. We investigated the relationship between the Prognostic Nutritional Index (PNI) and SAE outcomes. Methods This multicenter cohort study (2008–2019) enrolled 3202 SAE patients. The primary outcome was 28‐day all‐cause mortality. Multivariable‐adjusted analyses (logistic regression, propensity score matching, and inverse probability weighting) assessed PNI’s prognostic value, supplemented by generalized additive models (GAMs), Kaplan–Meier, and ROC analyses. External validation was performed. Results PNI independently predicted 28‐day mortality (adjusted OR: 0.85; 95% CI: 0.77–0.93). The GAM identified PNI = 34 as the optimal prognostic threshold. Patients with PNI < 34 had higher 28‐day mortality than those with PNI ≥ 34 in both original and validation cohorts ( p < 0.001). ROC analysis demonstrated strong discrimination in the original cohort (AUC = 0.879; sensitivity = 0.878; specificity = 0.880) and the validation cohort (AUC = 0.724). Higher PNI correlated with better neurological function (Glasgow Coma Scale, p < 0.001). Conclusions This multicenter study establishes the PNI as an independent predictor of 28‐day mortality in patients with SAE. We identified that SAE patients with PNI < 34 exhibited significantly higher 28‐day mortality rates and worse neurological function.