Impact of Nutritional Status on Mortality in Older Patients Hospitalized for Acute Heart Failure

Backgrounds/Objectives: Advances in prevention and medical care in the field of cardiology have led to an increase in the number of older patients with heart failure. In this population, assessment of nutritional status is particularly important. However, the prognostic impact of severity-based nutritional assessment at admission remains unclear. We conducted a study to elucidate the impact of malnutrition severity at admission on the prognosis of older patients hospitalized for acute heart failure (AHF). Methods: This study investigated the relationship between the Geriatric Nutritional Risk Index (GNRI) at admission and prognosis in 214 older patients aged ≥65 years who were hospitalized for AHF (mean age, 85 ± 8 years; male, 49%) between 2019 and 2023. GNRI was assessed by dividing patients into four groups: GNRI > 98 as normal (n = 64), 92 ≤ GNRI < 98 as mild risk (n = 54), 82 ≤ GNRI < 92 as moderate risk (n = 66), and GNRI < 82 as severe risk (n = 30). The discriminative performance of GNRI for 1-year all-cause mortality was compared with that of the Controlling Nutritional Status (CONUT) score and the Prognostic Nutritional Index (PNI). Results: During a median follow-up of 356 days, 76 deaths were observed. Worse GNRI categories were associated with older age, underweight, frailty, and anemia. Multivariable Cox proportional hazards models revealed that moderate GNRI risk (hazard ratio (HR), 2.69; 95% confidence interval (CI), 1.34-5.40) and severe GNRI risk (HR, 9.75; 95% CI, 4.30-22.10) were associated with higher all-cause mortality when compared with normal GNRI, along with age (HR per 1-year increase, 1.07; 95% CI, 1.03-1.11). Sensitivity analysis using GNRI as a continuous variable demonstrated similar results; GNRI was inversely associated with all-cause mortality (HR per 1 GNRI increase, 0.92; 95% CI, 0.90-0.95). In a subgroup analysis of age ≥85 years, the inverse association between GNRI and all-cause mortality was consistent. For 1-year all-cause mortality, GNRI showed moderate discrimination (area under the curve (AUC), 0.71; 95% CI, 0.63-0.80). Although the AUC of GNRI was not significantly different from that of the CONUT score or the PNI, GNRI demonstrated significantly better risk reclassification (net reclassification improvement, 0.47 vs. CONUT, p = 0.05; 0.43 vs. PNI, p = 0.02). Conclusions: In older patients with AHF including the oldest-old, nutritional status assessed by the GNRI at admission was predictive of prognosis. The importance of evaluating nutritional status at admission in clinical settings is reaffirmed.