Study design

A schematic view of the study and the three MR assumptions are shown in Fig. 1. The genetic variants associated with various hormones and hormone-mediated diseases modulated by the pituitary-thyroid/adrenal/gonadal axis were used as genetic IVs to examine the causal associations between those IVs and the risk of developing VTE, DVT and PE. Three key assumptions need to be met to conduct an MR study. First, the selected genetic variants should be strongly correlated with exposure. Second, the genetic variants must be independent of potential confounding factors. Finally, the influence of genetic variants on outcomes should only be mediated through exposure [18]. Since the expression of sex hormones has sex-specific differences, sex stratification analysis for the effect of genetically predicted estradiol and testosterone on the risk of developing VTE was further conducted. Written informed consent was obtained from the participants for all the included GWASs, and the original studies were all approved by the appropriate ethics committees. As we only extracted data from the GWASs database for analysis, no additional ethical review was required for this study.

Schematic view and three assumptions of this Mendelian Randomization study. (By Figdraw) SNPs, single-nucleotide polymorphisms; VTE, venous thromboembolism; DVT, deep vein thrombosis; PE, pulmonary embolism