Abstract
This cohort study investigated the association between neutrophil percentage-to-albumin ratio (NPAR) and mortality risks in U.S. adults with diabetes or prediabetes using National Health and Nutrition Examination Survey (NHANES) data (1999-2018). Among 101,316 screened participants, 47,477 eligible individuals (7,171 diabetes; 15,175 prediabetes) were analyzed after exclusions for age < 20 years, pregnancy, and missing data. NPAR was calculated as neutrophil percentage divided by albumin (g/dL). Mortality outcomes were ascertained via linkage to the National Death Index through 2019. Weighted multivariable Cox models adjusted for sociodemographic, metabolic, and clinical covariates revealed a J-shaped relationship between NPAR and mortality. Over a median follow-up of 99.5 months (IQR: 58-142), the highest NPAR quartile (Q4) demonstrated significantly elevated all-cause mortality (HR 1.585, 95% CI 1.434-1.753) and cardiovascular mortality (HR 1.642, 95% CI 1.370-1.969) compared to Q1. Restricted cubic spline analysis identified an inflection point at NPAR = 1.399. Below this threshold, NPAR associated only with cardiovascular mortality (HR 1.871, 95% CI 1.249-2.804), while values above 1.399 markedly increased risks for both all-cause (HR 3.241, 95% CI 2.343-4.483) and cardiovascular mortality (HR 2.062, 95% CI 1.213-3.503). Subgroup analyses confirmed consistency across age, gender, BMI, and comorbidity strata. These findings position NPAR as a novel integrative biomarker reflecting inflammatory-nutritional interplay, with critical prognostic value for mortality risk stratification in dysglycemic populations.