Neutrophil-to-albumin ratio and all-cause mortality in critically ill patients with cerebral infarction: a retrospective cohort study using the MIMIC-IV database

BACKGROUND: The neutrophil percentage-to-albumin ratio (NPAR) has been linked to outcomes in cardiovascular disease, yet its prognostic value in cerebral infarction (CI) remains unknown. This study investigates the association between NPAR and all-cause mortality in critically ill CI patients. METHODS: Data from 2338 CI patients (MIMIC-IV database) were stratified into NPAR tertiles (T1: < 20.975, T2: 20.975-26.888, and T3: ≥ 26.888). Primary endpoints included ICU, 30-, 90-, and 365-day mortality. Survival analysis used Kaplan-Meier curves and multivariable Cox regression. Restricted cubic splines (RCS) tested nonlinear associations, and subgroup analyses assessed consistency across clinical subgroups. RESULTS: Mortality rates were 12.45% (ICU), 25.15% (30 days), 33.40% (90 days), and 43.54% (365 days). Higher NPAR was associated with worse survival (all log-rank p < 0.001). Adjusted Cox models showed T3 vs. T1 increased 30-day, 90-day, and 365-day mortality risk. RCS revealed nonlinear relationships (P-nonlinearity < 0.01), with an inflection point at NPAR = 23.6. Subgroup analyses showed consistent associations across sex, comorbidities, and severity scores, except age, which modified ICU mortality risk. CONCLUSIONS: Higher NPAR levels are independently associated with increased short- and long-term all-cause mortality in critically ill CI patients. NPAR may serve as a practical and accessible prognostic marker for risk stratification in the ICU.