To analyze the causal relationships between basal metabolic rate(BMR) and frailty index (FI) based on bidirectional two-sample mendelian randomization(MR).

The data of BMR (n=454 874) and FI (n=175 226) were extracted from the published genome-wide association studies (GWAS). By setting conditions with P<5×10^-8 as the screening criterion, the linkage disequilibrium coefficient set to 0.001 and the width of the linkage disequilibrium region of 10 000 kb, and single nucleotide polymorphisms (SNP) were screened out as the final instrumental variables. Five methods including MR-Egger regression, weighted median, Inverse variance weighted (IVW), simple mode and weighted mode were used for MR analysis, and IVW was used as the main analysis method, and β and 95%CI were used to demonstrate the causal relationship between BMR and FI. MR-Egger intercept and MR-PRESSO analysis were used to test the pleiotropy, and the heterogeneity was analyzed by Cochran Q test of MR-Egger. Leave-one-out test were performed to analyze the sensitivity. Finally, reverse MR analysis was used to verify the robustness of the results.

A total of 787 SNPs associated with BMR and 9 SNPs associated with FI were screened. The MR analysis showed a causal relationship between BMR and FI(Weighted median, β=0.105, 95%CI: 0.063-0.146, P<0.001; IVW, β=0.110, 95%CI: 0.808-0.140, P<0.001; weighted mode, β=0.145, 95%CI: 0.047-0.244, P=0.004). The screened SNPs did not have horizontal pleiotropy because the the MR-Egger intercept was 0.000 2 (P=0.374), Therefore, the MR was an effective method for causal inference in this study. The Cochran Q-test results showed Q=137.053, P=0.188, indicating no heterogeneity among the SNPs included in the MR analysis. And no outlier SNPs were found by MR-PRESSO analysis. The sensitivity analysis based on the leave-one-out method showed that the individual SNP did not affect the robustness of the MR analysis results. No causal relationship between FI and BMR was found in the reverse MR analysis(P>0.05).

Genetically predicted high BMR level is significantly associated with elevated FI, and there is no association in reverse direction, which may provide fresh ideas for proposing feasible interventions to manage frailty.