Adverse pregnancy outcomes and coronary artery disease risk: A negative control Mendelian randomization study

Abstract

Background: Adverse pregnancy outcomes are predictive for future cardiovascular disease risk, but it is unclear whether they play a causal role. We conducted a Mendelian randomization study with males as a negative control population to estimate the associations between genetic liability to adverse pregnancy outcomes and risk of coronary artery disease. Methods: We extracted uncorrelated (R2<0.01) single-nucleotide polymorphisms strongly associated (p-value<5e-8) with miscarriage, gestational diabetes, hypertensive disorders of pregnancy, preeclampsia, placental abruption, poor fetal growth and preterm birth from relevant genome-wide association studies. Genetic associations with risk of coronary artery disease were extracted separately for females and males. The main analysis was the inverse-variance weighted analysis, while MR Egger, weighted median and weighted mode regression, bidirectional analyses, and the negative control population with males were sensitivity analyses to evaluate bias due to genetic pleiotropy. Results: The number of cases for the adverse pregnancy outcomes ranged from 691 (182,824 controls) for placental abruption to 49,996 (174,109 controls) for miscarriage, and there were 22,997 (310,499 controls) and 54,083 (240,453 controls) cases of coronary artery disease for females and males, respectively. We observed an association between genetic liability to hypertensive disorders of pregnancy and preeclampsia, and to some extent gestational diabetes, poor fetal growth and preterm birth, with an increased risk of coronary artery disease among females, which was supported by the MR Egger, weighted median and weighted mode regressions. However, in the negative control population of males, we observed largely the same associations as for females. Conclusions: The associations between adverse pregnancy outcomes and coronary artery disease risk were likely driven by confounding (e.g., shared genetic liability). Thus, our study does not support the hypothesis that adverse pregnancy outcomes are causal risk factors for cardiovascular diseases.