Mining human epidemiological data can only produce correlations. Mendelian randomization is a way to add data on genetic variants known to affect disease status into the mix so as to add some support for causation. The result isn’t a determination of causation, but in the best case is supportive of that conclusion. Physical frailty is well known to correlate with neurodegeneration, and a range of reasonable hypotheses exist to explain why this is the case. Both emerge from chronic inflammation and other underlying dysfunctions of aging, for example. Or frailty involves significant dysfunction in the cardiovascular system, which in turn negatively affects the aging brain. In the absence of ways to eliminate the chronic inflammation of aging or reverse vascular aging, it is hard to prove any of this. Which doesn’t matter! Regardless, the right course is to try to build approaches to reverse the damage and dysfunction of aging.
Physical frailty is associated with a higher risk of developing dementia, but it remains unclear whether this relationship is causal. This prospective cohort study was based on UK Biobank participants without dementia at enrollment (between 2006 and 2010). Physical frailty was defined by 5 criteria (weight loss, exhaustion, physical inactivity, slow walking speed, and low grip strength). Incident dementia was tracked through linked hospital admission records and death registries, using the International Classification of Diseases, Tenth Revision (ICD-10) codes. Cox proportional hazard regression models and bidirectional Mendelian randomization (MR) analyses were used to evaluate the causal association of physical frailty with incident dementia.
Among 489,573 participants (mean age 57.03 years, 54.4% female), 8,900 dementia cases were documented over a median follow-up of 13.58 years. Compared with nonfrail individuals, the risk of dementia was 50% higher in those with prefrailty (hazard ratio [HR]:1.50) and 182% higher in those with frailty (HR: 2.82). Participants with frailty and high genetic risk had the highest risk of dementia compared with those with low genetic risk and nonfrailty (HR: 3.87 for high polygenic risk score; HR: 8.45 for APOE-ε4 carriers). The forward MR analysis indicated a potential causal relationship between physical frailty and dementia (odds ratio [OR]:1.79) while the reverse MR suggested a null causal association (OR: 1.00). Structural equation modeling points to genetic background and neurologic and immunometabolic function as potential underlying mechanisms linking physical frailty to dementia.
