Protein Nutritional Status and Frailty: A Mendelian Randomization Study

Abstract

ABSTRACT Background Observational studies have suggested that better protein nutritional status may contribute to prevention of frailty. Objective We sought to examine this hypothesis using a Mendelian randomization (MR) analysis. Methods We conducted a two-sample MR study using GWAS summary statistics data of the UK Biobank. We applied genetically predicted serum albumin as a primary exposure measure and serum total protein as a secondary exposure measure. The outcome measure was the Rockwood frailty index (FI) based on 49 deficits from 356,432 individuals (53.3% of them were women, with a mean ± SD age of 56.7 ± 8.0 y. The association between serum protein measures and FI was mainly analyzed by use of the inverse variance weighted method. Results A genetically predicted serum albumin concentration was not statistically significantly associated with FI in the full sample. However, in women, we observed a preventive association between genetically predicted serum albumin and FI (β = −0.172 per g/L; 95% CI: −0.336, −0.007; P = 0.041). In the full sample, genetically predicted serum total protein was inversely associated with FI (β: −0.153 per g/L; 95% CI: −0.251, −0.056; P = 0.002). In both women and men, higher serum total protein was significantly inversely associated with FI; regression coefficients were −0.148 per g/L (95% CI: −0.287, −0.009; P = 0.037) for women, −0.154 per g/L (95% CI: −0.290, −0.018; P = 0.027) for men. Conclusions The present MR study implies that better protein nutritional status modestly contributes to reducing the risk of frailty.