Exploring nutritional indicators of cardiovascular mortality risk in elderly hypertensive patients: The long-term...

Cardiovascular mortality is the leading cause of death among elderly hypertensive patients. However, the reference indicators for nutritional management in this population remain a subject of debate. The aim of this study is to explore and compare the predictive value of three commonly used nutritional assessment indicators for cardiovascular mortality in elderly hypertensive patients.

This study included 3611 elderly hypertensive patients aged 60 and above from seven cycles of NHANES (2005-2018). The population was categorized into two groups (malnourished vs. non-malnourished) using reference cutoff values for three nutritional assessment indicators: PNI, GNRI, and CONUT score. Multivariate Cox regression and competing risk analysis were employed to compare the predictive abilities of these three indicators for cardiovascular mortality risk. Subgroup analyses were also conducted to explore whether kidney dysfunction, cardiovascular disease, or gender interacted with the three nutritional indicators. Additionally, restricted cubic splines (RCS) curves were used to explore the dose-response relationship. Decision curve analysis was applied to assess the clinical value of these three indicators in predicting cardiovascular mortality risk. Time-dependent receiver operating characteristic (ROC) curves were used to calculate the area under the curve (AUC) for each indicator's prediction of cardiovascular mortality risk at different follow-up times. Furthermore, Net Reclassification Improvement (NRI) and Integrated Discrimination Improvement (IDI) were calculated based on multivariate Cox regression models to compare the predictive ability of these models over different follow-up durations. Malnourished patients diagnosed by PNI had a 2.70 times higher risk of cardiovascular death compared to non-malnourished patients (HR: 3.70, 95 % CI: 2.54-5.38), representing the highest cardiovascular mortality risk among the three groups. Patients diagnosed with malnutrition using GNRI and CONUT score had cardiovascular mortality risks increased by 1.39 times (HR: 2.39, 95 % CI: 1.58-3.63) and 0.84 times (HR: 1.84, 95 % CI: 1.33-2.55), respectively. In the multivariate competing risks model, the results were similar to those from the Cox regression analysis. The non-restricted cubic spline plot demonstrates an L-shaped association between GNRI and PNI with cardiovascular mortality, while the COUNT score shows an inverse L-shaped association. In addition, both the Time-ROC curve's AUC and NRI support that PNI's predictive advantage for cardiovascular mortality risk gradually increases with longer follow-up time.

PNI has superior predictive value for cardiovascular mortality risk compared to GNRI and COUNT score, especially for long-term prognosis.