Prognostic impact of nutrition measures in patients with heart failure varies with coexisting physical frailty.

Abstract

Malnutrition is prevalent among patients with heart failure (HF); however, the effects of coexisting malnutrition and frailty on prognosis are unknown. This study examines the impact of malnutrition and frailty on the prognosis of patients with HF. We examined 1617 patients with HF aged 65years or older (age: 78.6±7.4; 44% female) from a Japanese multicentre prospective cohort study. The nutritional status was evaluated using the Geriatric Nutritional Risk Index (GNRI), Controlling Nutritional Status (CONUT), and Mini Nutritional Assessment Short Form on discharge. Frailty was assessed using the criteria determined in a previous study on patients with HF. The prognostic impact of each nutrition measure on the risk of composite all-cause mortality and cardiac readmissions within 2years of hospital discharge was assessed using Kaplan-Meier survival curves and Cox proportional hazards model analysis for non-frail and frail groups. Over 2324.2 person-years of follow-up, 88 patients died and 448 patients experienced readmission due to HF. In the non-frail group, poor nutritional status assessed using the GNRI and CONUT was associated with an increased hazard ratio (HR) of composite outcomes in the crude model; however, adjustment for potential confounders diminished the association. In the frail group, all three nutritional indicators were associated with the cumulative incidence of the study outcome (log-rank test, P<0.05). In multivariate analysis, only the CONUT score was associated with an increased HR even after adjustment for confounders. The CONUT score predicted a poor prognosis in HF patients with coexisting physical frailty, highlighting the potential clinical benefit of nutritional assessment based on biochemical data for further risk stratification.