Mendelian Randomization Study Does Not Support a Bidirectional Link between Atherosclerosis and Venous Thromboembolism.

Abstract

Some observational studies suggested that atherosclerosis increased the risk of venous thromboembolism (VTE), and vice versa. However, the results were conflicting, and the causal relationship is yet to be established. Therefore, we applied Mendelian randomization (MR) analyses to assess the bidirectional causality between coronary heart disease (CHD) and VTE, deep venous thrombosis (DVT), and pulmonary embolism (PE). A total of 184,305 individuals with CHD were included from the CARDIoGRAMplusC4D Consortium. Information on VTE, DVT, and PE were obtained from the FinnGen biobank. Genetic instruments for CHD and VTE were constructed using 37 and 12 single-nucleotide polymorphisms, respectively. Inverse-variance weighted meta-analysis under a random-effect model was used as the preliminary estimate. Five complementary MR methods were also used, including weighted median, MR-Egger, multivariable MR (adjusted for the body mass index), simple mode, and weighted mode methods. The genetically instrumented VTE (odds ratio [OR]: 1.05; 95% confidence interval [CI]: 1.00-1.11; P=0.06), DVT (OR: 1.03; 95% CI: 0.99-1.08; P=0.19), or PE (OR: 1.07; 95% CI: 0.98-1.16; P=0.11) showed no causal relationships with CHD. There was also no clear evidence showing the causal effects of CHD on VTE (OR: 1.00; 95% CI: 0.82-1.22; P=0.98), DVT (OR: 1.00; 95% CI: 0.79-1.27; P=0.97), or PE (OR: 0.98; 95% CI: 0.82-1.18; P=0.87). No pleiotropic bias was found in the MR analyses. As heterogeneity was significant, a random model was used to minimize the effect of heterogeneity. No causal associations existed between CHD and VTE. Arterial and venous thromboses may represent separate entities.