Short- and long-term impact of nutrition in a large cohort of patients undergoing transcatheter aortic valve implantation: prognostic value of GNRI

Abstract

Abstract Background Nutritional status, as measured by Geriatric Nutritional Risk Index (GNRI), an easy-to-use tool combining both clinical and laboratory parameters, was shown to predict short-term survival in patients with aortic stenosis undergoing transcatheter aortic valve implantation (TAVI) by our group and others, and this has been extensively confirmed later on. However, data on the long-term impact of nutritional indexes on TAVI outcomes are still lacking. Purpose We aimed at confirming our previous short-term findings in a large population and to investigate the prognostic impact of GNRI in the long-term. Methods All consecutive patients who underwent TAVI between Oct. 2015 and Dec. 2021 at our centre were selected. Baseline nutritional parameters were consistently recorded. An up-to-6-year prospective follow-up was performed. Patients with missing albumin values or with no follow-up available were excluded. Taking into account the old age of patients usually being treated with TAVI, the geriatric version of NRI (GNRI for age>65) was used, and the very few patients younger than 65 years were excluded. GNRI was calculated as described by Bouillanne et al using the following formula: Geriatric (G) NRI = (1.489)×Albumin (g/L) + [(41.7×(present weight/ideal weight)]. The primary endpoint was long-term all-cause mortality. Patients were divided into: GNRI>98 (no risk of malnutrition) and ≤98 (risk of malnutrition). Kaplan-Meier curves were plotted and impact on prognosis was calculated using Cox regression analyses. Results A total of 1084 patients, median (IQR) age 82.2 (78.5–85.5) years, 44.4% female, were included. Median (IQR) BMI was 26.6 (24.2–30.0) kg/m2, corresponding to a slightly overweight population. Median (IQR) GNRI at baseline was 113.7 (106–120.2). 8.6% of patients showed risk of malnutrition (GNRI ≤98). Kaplan-Meier curves showed a significant impact on long-term mortality in patients with risk of malnutrition compared to those with no malnutrition risk, with curves starting to diverge in the first weeks after the procedure. The effect of nutritional status measured by GNRI increased with time. A significant association was shown in univariate analyses (HR 2.3 [95% CI 1.7–3.1], p<0.001) and also in multivariate analyses after adjusting for potential confounding factors (including age, sex and Euroscore II, among others) (HR 1.7 [95% CI 1.2–2.5], p<0.002). According to area under curve for ROC curves, GNRI showed a better predictive value than the widely used BMI. Conclusion This study confirmed the negative impact of the malnutrition risk status, as measured by GNRI, on short-term mortality after TAVI, and showed an even more pronounced effect in the long-term. GNRI could be easily recorded on admission, before TAVI procedure, and considered for heart team decisions. Further studies may be warranted to identify potential mechanisms for nutritional effects and to investigate the potential use of nutrition-improving measures in selected patients. Funding Acknowledgement Type of funding sources: None.