Prognostic Value of Four Preimplantation Malnutrition Estimation Tools in Predicting Heart Failure Hospitalization of the Older Diabetic Patients with Right Ventricular Pacing

Abstract

Objectives: The prognostic value of preimplantation nutritional status is not yet known for older diabetic patients that received right ventricular pacing (RVP). The study aimed to investigate the clinical value of the four malnutrition screening tools for the prediction of heart failure hospitalization (HFH) in older diabetic patients that received RVP.

Design: Retrospective observational cohort study.

Setting and participants: This study was conducted between January 2017 and January 2018 at the Fuwai Hospital, Beijing, China, and included older (age ≥ 65 years) diabetic patients that received RVP for the first time Measurements: The Prognostic Nutritional Index (PNI), Geriatric Nutritional Risk Index (GNRI), Naples Prognostic Score (NPS), and the Controlling Nutritional Status (CONUT) score were used to estimate the preimplantation nutritional status of the patients. Univariate and multivariate Cox proportional hazard regression analyses were performed to investigate the association between preimplantation malnutrition and HFH.

Results: Overall, 231 older diabetic patients receiving RVP were included. The median follow-up period after RVP was 53 months. HFH was reported for 19.9% of the included patients. Our results showed preimplantation malnutrition for 18.2%, 15.2%, 86.6% and 66.2% of the included patients based on the PNI, GNRI, NPS, and CONUT score, respectively. The cumulative rate of HFH during follow-up period was significantly higher for patients in the preimplantation malnutrition group based on the PNI (log-rank = 13.0, P = 0.001), GNRI (log-rank = 8.5, P = 0.01), and NPS (log-rank = 15.7, P < 0.001) compared to the normal nutrition group, but was not statistically significant for those in the preimplantation malnutrition group based on the CONUT score (log-rank = 2.7, P = 0.3). As continuous variables, all the nutritional indices showed significant correlation with HFH (all P < 0.05). However, multivariate analysis showed that only GNRI was independently associated with HFH (HR = 0.97, 95% CI: 0.937-0.997, P = 0.032). As categorical variables, PNI, GNRI, and NPS showed significant correlation with HFH. After adjustment of confounding factors, moderate-to-severe degree of malnutrition was an independent predictor of HFH based on the PNI (HR = 4.66, 95% CI: 1.03-21.00, P = 0.045) and GNRI (HR = 3.02, 95% CI: 1.02-9.00, P = 0.047).

Conclusion: Preimplantation malnutrition was highly prevalent in older diabetic patients that received RVP. The malnutrition prediction tools, PNI and GNRI, showed significant prognostic value in accurately predicting HFH in older diabetic patients with RVP.

Keywords: Nutritional index; diabetes; older; right ventricular pacing.