Associations of Geriatric Nutritional Risk Index and Prognostic Nutritional Index With All-Cause and Cardiovascular Mortality in DKD: NHANES 1999-2018.

BACKGROUND: Malnutrition and immune status significantly influence the prognosis of diabetic kidney disease (DKD). However, the correlations between two novel nutritional-immune indices-geriatric nutritional risk index (GNRI) and prognostic nutritional index (PNI)-and the risk of all-cause mortality (ACM) and cardiovascular mortality (CVM) in patients with DKD remain unclear. Our investigation was, therefore, designed to explore these associations. METHODS: This study analyzed data from 2038 DKD individuals in the National Health and Nutrition Examination Survey (NHANES) (1999-2018). Multivariate weighted Cox regression, Kaplan-Meier survival curves (K-M curves), subgroup analysis, and interaction analysis were employed to clarify the associations of GNRI and PNI with ACM and CVM. We also applied restricted cubic splines (RCS) and threshold analysis to evaluate potential nonlinear correlations and inflection points. Subsequently, time-dependent receiver operating characteristic (ROC) curves were utilized to explore and compare the prognostic performance of GNRI and PNI for survival outcomes over time. RESULTS: Throughout the median follow-up period of 68 months, 1006 (49.36%) of 2038 patients with DKD died, including 374 (37.18%) cardiovascular disease (CVD) deaths. Multivariate weighted Cox regression analysis and K-M curves illustrated that high levels of GNRI and PNI are independent protective factors against ACM and CVM in DKD patients. RCS analysis and threshold analysis revealed a nonlinear correlation of PNI with ACM and CVM, and the inflection points were identified at 51.24 and 49.63, respectively. What's more, ROC curves demonstrated that GNRI and PNI exhibit significant predictive potential in both the short and long terms, with similar predictive performance. CONCLUSIONS: As novel nutritional-immune markers, GNRI and PNI could independently forecast the ACM and CVM in DKD patients, aiding in risk stratification and timely intervention for this disease.