Grip strength, a simple, reliable and non-invasive measure of overall muscle strength, is often considered as an indicator of frailty [1], [2], a clinical syndrome characterized by a decrease in the physiological reserves of an individual, and an important predictor of physical disability [3]. Longitudinal studies have suggested that low grip strength is an important predictor of cardiovascular disease (CVD) [4], [5], [6], [7]. However, many of these studies do not fully address bias due to unmeasured confounding, limiting their ability to estimate the causal association between grip strength and CVD.

Mendelian randomization (MR) studies address unmeasured confounding by using genetic variants as instrumental variables [8], [9]. Two-sample MR studies leverage published data from large biobanks and consortia to conduct MR using summary measures of the associations between the genetic variants and each of the risk factors and outcomes of interest [10]. Two-sample MR studies have found an association between grip strength and coronary artery disease [11], [12], [13], myocardial infraction [12], [13], and atrial fibrillation [11], [13]. However, all of these studies have been conducted using the same data sources (the UK Biobank or the CHARGE consortium for the genetic associations for grip strength [12], [13], the CARDIoGRAMplusC4D Consortium for myocardial infarction and coronary artery disease and the AFGen Consortium for atrial fibrillation [11], [13]), although they did not use the same number of genetic variants as instruments for grip strength. Because they rely on estimates that are not age- or sex- specific, these studies also tacitly assume that the effect of grip strength on CVD does not vary with sex or age [10]. The validity of this assumption is not guaranteed, as sex differences with regards to CVD events [14] are well documented, and evidence suggests that the influence of low grip strength on cardiovascular risk differs according to both sex [15], [16], [17] and age [18].

We conducted a sex- and age-stratified one-sample MR study of grip strength and carotid intima-media thickness (cIMT). cIMT is an objective non-invasive measure known to be a marker of CVD risk [19] and more specifically an important surrogate measure of vascular atherosclerosis [20], [21], [22]. Since atherosclerosis is the underlying cause of most CVD events, our focus on cIMT is clinically advantageous as intimal thickening is believed to precede CVD events [23], [24], [25].