Comparison of the Value of Four Objective Nutritional Indices in Assessing the Long-Term Prognosis of Elderly Patients with Heart Failure with Preserved Ejection Fraction.

Abstract

Background: The long-term prognosis of heart failure with preserved ejection fraction (HFpEF) is influenced by malnutrition. Currently, there's a deficit in objective and comprehensive nutritional assessment methods to evaluate the nutritional status and predicting the long-term outcomes of HFpEF patients.

Methods: Our retrospective study included two hundred and eighteen elderly HFpEF patients admitted to the cardiovascular ward at the Air Force Medical Centre from January 2016 to December 2021. Based on follow-up outcomes, patients were categorized into all-cause death (99 cases) and Survival (119 cases) groups. We compared general data, laboratory results, and nutritional indexes between groups. Differences in subgroups based on Triglyceride-Total Cholesterol-Body Weight Index (TCBI), Geriatric Nutritional Risk Index (GNRI), Prognostic Nutritional Index (PNI), and Controlled Nutrition Score (CONUT) were analyzed using Kaplan-Meier survival curves and log-rank test. COX regression was used to identify all-cause mortality risk factors, and the predictive accuracy of the four nutritional indices was assessed using receiver operating characteristic (ROC) curves and Delong test analysis.

Results: A total of 101 (45.41%) HFpEF patients experienced all-cause mortality during 59.02 $\pm$ 1.79 months of follow-up. The all-cause mortality group exhibited lower GNRI and PNI levels, and higher CONUT levels than the Survival group (p $<$ 0.05). Kaplan-Meier analysis revealed lower cumulative survival in the low GNRI ( $\le$ 96.50) and low PNI ( $\le$ 43.75) groups, but higher in the low CONUT ( $\le$ 2) group, compared to their respective medium and high-value groups. Multifactorial COX regression identified low PNI ( $\le$ 43.75) as an independent all-cause mortality risk factor in elderly HFpEF patients. ROC and Delong's test indicated PNI (area under the curve [AUC] = 0.698, 95% confidence interval [CI] 0.629-0.768) as a more effective predictor of all-cause mortality than TCBI (AUC = 0.533, 95% CI 0.456-0.610) and CONUT (AUC = 0.621, 95% CI 0.547-0.695; p $<$ 0.05). However, there was no significant difference compared to GNRI (AUC = 0.663, 95% CI 0.590-0.735; p $>$ 0.05).

Conclusions: In elderly HFpEF patients a PNI $\le$ 43.75 was identified as an independent risk factor for all-cause mortality. Moreover, PNI demonstrates superior prognostic performance in predicting all-cause mortality in elderly patients with HFpEF when compared to TCBI, GNRI, and COUNT.