The geriatric nutrition risk index is longitudinally associated with incident Sarcopenia: evidence from a 5-year prospective cohort

Previous studies investigating the association between the geriatric nutrition risk index (GNRI) and sarcopenia either lacked longitudinal evidence or narrowly focused on specific populations.

We aimed to reveal longitudinal associations of GNRI with sarcopenia risk in community-dwelling Chinese. We also investigated interaction effects of potential factors on such associations.

We included participants aged ≥ 50 years with sufficient data from the WCHAT study who did not have sarcopenia at baseline and completed sarcopenia assessment during follow-up. GNRI was calculated according to the formula based on serum albumin, height and weight. Sarcopenia was diagnosed according to the 2019 AWGS consensus. Longitudinal associations between GNRI and sarcopenia were estimated by logistic regression with GNRI as either a continuous or categorical variable by tertiles, using generalized estimating equations (GEE) as sensitivity analyses. Subgroup analyses by potential covariates were conducted to detect interaction effects.

A total of 1907 participants without baseline sarcopenia were finally included, of whom 327 (17.1%) developed incident sarcopenia during 5-year follow-up. After controlling for confounders, sarcopenia risk decreased with each one standard deviation increase in GNRI (OR_adjusted=0.36, 95% CI 0.31–0.43), and it also decreased successively from the lowest (< 111.2) through middle (111.2-117.7) to the highest (≥ 117.8) tertile of the GNRI level (P for trend < 0.001). Similar results were yielded by GEE. Such associations generally remained robust across subgroups with distinct characteristics, while significant differences were observed between different age groups (≥ 65 vs. <65 years) (interaction P-value < 0.05).

GNRI is longitudinally associated with sarcopenia risk with possibly age-specific differences in association magnitude, which holds implications for policymakers to conduct population-based risk assessment.