Introduction:
Several observational studies have shown that taller body height is associated with greater risk of venous thromboembolism (VTE), but it is not known whether the association is causal. We used instrumental variable analysis (Mendelian randomization) to explore the causal relationship between height and VTE using a genetic risk score (GRS) for height as the instrument.
Hypothesis:
There is a causal relationship between taller standing height and greater risk of VTE, as demonstrated by a Mendelian randomization approach.
Methods:
We created a weighted GRS for height in white men and women in the Longitudinal Investigation of Thromboembolism Etiology [consisting of two longitudinal cohort studies: Atherosclerosis Risk in Communities (ARIC) and Cardiovascular Health Study (CHS)] using 668 single nucleotide polymorphisms from a recently published meta-analysis. Incident hospitalized VTE events were identified and verified by physician review of medical records. We estimated the association and causal risk differences (RD) and 95% confidence intervals (CI) for VTE incidence per standard deviation (SD) increment in height (9.4 cm). The association models were adjusted for age, sex, waist circumference, and study site.
Results:
There were 9,137 ARIC and 3,163 CHS participants at risk for VTE at baseline and with genetic data, and they experienced 367 (ARIC) and 105 (CHS) incident VTE events over a median 22.7 and 11.8 years of follow-up, respectively. Baseline age ranges were 45-64 and 65-98 years and mean heights were 169 and 165 cm in ARIC and CHS, respectively. The GRS was a strong instrument for height (R
2
=0.08 in ARIC and R
2
=0.07 in CHS) and had little to no correlation with other measured VTE risk factors (all R
2
≤0.01). In ARIC, taller height was associated with greater risk of VTE [association VTE RD: 1.0% per SD in height (95% CI: 0.3 to 1.6%)]. The causal RD had the same magnitude as the association RD, but did not quite reach statistical significance [causal VTE RD per SD in height: 1.1% (95% CI: -0.3 to 2.5%)]. Predicted risks of VTE at the 10th and 90th percentiles of height (157 and 181 cm) were 2.6% and 5.4%, respectively, representing more than a doubling of risk. There was no association between height and VTE risk in CHS [association VTE RD per SD in height: 0.1% (95% CI: -1.0 to 1.2%); causal VTE RD per SD in height: -0.3% (95% CI: -2.5 to 1.9%)].
Conclusion:
Taller height was associated with greater VTE risk with some supporting causal evidence in middle-aged adults from ARIC, but there was no relation between height and VTE in older adults from CHS. Future studies should further explore the causal relation between height and VTE among different age groups.
Interaction of a genetic risk score with physical activity, physical inactivity, and body mass index in relation to venous thromboembolism risk
