Body mass index and body composition in relation to 14 cardiovascular conditions in UK Biobank: a Mendelian randomization study

Aims The causal role of adiposity for several cardiovascular diseases (CVDs) is unclear. Our primary aim was to apply the Mendelian randomization design to investigate the associations of body mass index (BMI) with 13 CVDs and arterial hypertension. We also assessed the roles of fat mass and fat-free mass on the same outcomes. Methods and results Single-nucleotide polymorphisms associated with BMI and fat mass and fat-free mass indices were used as instru- and results mental variables to estimate the associations with the cardiovascular conditions among 367 703 UK Biobank participants. After correcting for multiple testing, genetically predicted BMI was significantly positively associated with eight outcomes, including and with decreasing magnitude of association: aortic valve stenosis, heart failure, deep vein thrombosis, arterial hypertension, peripheral artery disease, coronary artery disease, atrial fibrillation, and pulmonary embolism. The odds ratio (OR) per 1 kg/m(2) increase in BMI ranged from 1.06 [95% confidence interval (CI) 1.02-1.11; P=2.6 x 10(-3)] for pulmonary embolism to 1.13 (95% CI 1.05-1.21; P=1.2 x 10(-3)) for aortic valve stenosis. There was suggestive evidence of positive associations of genetically predicted fat mass index with nine outcomes (P < 0.05). The strongest magnitude of association was with aortic valve stenosis (OR per 1 kg/m(2) increase in fat mass index 1.46, 95% CI 1.13-1.88; P=3.9 x 10(-3)). There was suggestive evidence of inverse associations of fat-free mass index with atrial fibrillation, ischaemic stroke, and abdominal aortic aneurysm. Conclusion This study provides evidence that higher BMI and particularly fat mass index are associated with increased risk of aortic valve stenosis and most other cardiovascular conditions.