JMIR Publications

ABSTRACT

Background:

Over half of people over 60 experience cognitive impairment, with limited treatment options and irreversible pathology, making it crucial to identify risk factors. Studies have examined the association between sarcopenia and cognitive impairment; however, the evidence is inconclusive and cannot be used to make causal inferences.

Objective:

By triangulating data from cohort study and Mendelian randomization (MR) analysis, this study aims to appraise the causal association of sarcopenia with cognitive impairment.

Methods:

Using UK Biobank data, we first examined the associations of sarcopenia and its indices (appendicular lean mass [ALM], handgrip strength, and gait speed) with cognitive function (fluid intelligence and prospective memory) by employing mixed-effects regression models. Then, we explored the causal associations of genetically-predicted sarcopenic indices with cognitive function through two-sample MR and further investigated whether such associations were mediated by physical inactivity, depression, anxiety, falls, frailty, and vitamin D use.

Results:

A total of 34,457 participant were included in the observational analysis, with a mean age of 56.4 (7.6) years and 51.1% of them being female. The cohort study revealed that sarcopenia was significantly associated with cognitive impairment, which was evidenced by reduced fluid intelligence scores (β=-0.865, 95%CI: -1.589, -0.142, P=0.019). Each of the sarcopenic indices also exhibited significant associations with either fluid intelligence or prospective memory (all P values <0.05). MR analyses yielded compelling evidence of positive associations between the genetically-predicted increases in ALM (β=0.094, 95%CI: 0.071, 0.117, P<0.001), handgrip strength (β=0.180, 95%CI: 0.075, 0.285, P<0.001) and gait speed (β=0.775, 95%CI: 0.528, 0.285, P<0.001) and improved cognitive function. The effects of ALM and handgrip strength on cognitive function were partially mediated by genetically-predicted physical activity, with indirect effects of 0.008 (95%CI: 0.002, 0.015) for ALM and 0.019 (95%CI: 0.001, 0.047) for handgrip strength.

Conclusions:

Our study suggests that sarcopenia is a potential causal risk factor for cognitive impairment, with physical activity acting as a modifiable mediator in this relationship.