Abstract
BACKGROUND: Observational studies examining sarcopenia-related traits and fall risk remain controversial. Herein, we conducted meta-analyses of cohort studies triangulated with Mendelian randomization (MR) analyses to examine the potential causality between sarcopenia-related traits and risk of falls in older adults. METHODS: Literature search across PubMed, Embase, and Cochrane Library was performed from inception to February 2023 to identify cohort studies examining sarcopenia-related traits (including hand strength, appendicular lean mass, and walking speed) and falls. We assessed the association between these traits and fall risk using random-effects models to calculate pooled odds ratios (OR) and 95% confidence intervals (CIs). MR analyses were conducted using summary statistics derived from the UK Biobank consortium for sarcopenia-related traits and FinnGen consortium for falls. The inverse-variance weighted method was used as primary analysis. RESULTS: Our meta-analysis included 34 cohort studies. The combined analysis of sarcopenia-related traits revealed a 33% reduced fall risk with each unit increase in walking speed (OR 0.67, 95% CI 0.54-0.84) and a 2% decrease with each unit increase in hand strength (OR 0.98, 95% CI 0.97-0.99). However, appendicular lean mass had no significant effect on falls. In the MR analyses, only walking speed was causally associated with falls (OR 0.64, 95% CI 0.48-0.84). Hand strength and appendicular lean mass showed no statistically significant causal effect on falls. CONCLUSION: Evidence from meta-analysis and MR suggests a strong association between walking speed and fall risk in older adults. However, the relationship between hand strength, appendicular lean mass, and falls has not yet been established.