Body Height and Incident Risk of Venous Thromboembolism: A Cosibling Design.

Abstract

Background: Body height has been associated with an increased risk of venous thromboembolism (VTE), but the association can be confounded with shared familial factors (genetic/environmental). A cosibling design is useful for deeper understanding about the relationship between VTE and height.

Methods and results: From Swedish national registry databases, we used a corelative design with full siblings alongside a general Swedish population sample. A cohort of male conscripts (n=1 610 870), born in 1951 to 1992 without previous VTE, was followed from enlistment (1969-2010) until 2012. Another cohort of first-time pregnant women (n=1 093 342) from the medical birth register, without previous VTE, was followed from first pregnancy (1982-2012) until 2012. Using the Multi-Generation Register, we identified all full-sibling pairs discordant for height. This cosibling design allowed for adjustment for familial factors (genetic/environmental). Compared with the tallest women (>185 cm) and men (>190 cm), there was a graded decreased risk by lower height for both men and women. The risk was lowest in women and men with the shortest stature (<155 and <160 cm, respectively): hazard ratios=0.31 (95% confidence interval, 0.22-0.42) and 0.35 (95% confidence interval, 0.22-0.55), respectively. There was a graded association also in the cosibling design comparing siblings with varying degree of discordance for height (reference was the taller sibling): ≥10 cm difference between brothers hazard ratios=0.69 (95% confidence interval, 0.61-0.78) and sisters hazard ratios=0.65 (95% confidence interval, 0.52-0.80), respectively.

Conclusions: Height is an independent predictor of VTE. The use of sibling pairs reduces the likelihood that familial confounding explains the results. The findings are important for the understanding of the pathogenesis of VTE.