RÉSUMÉ
BACKGROUND: Observational studies have associated nonalcoholic fatty liver disease (NAFLD) with adverse pregnancy events, but findings show heterogeneity, leaving the causal direction and mediating pathways unclear. We aimed to investigate the causal relation between NAFLD and various pregnancy events, and to elucidate the underlying mediating pathways while determining the proportion of this correlation that is mediated through these pathways. METHODS: A genome-wide association study involving over 6 million participants employing Mendelian randomization (MR) and mediation analysis was performed. The study used genetically predicted NAFLD as exposures and cardiometabolic traits as mediators, with various adverse pregnancy events as outcomes. The main analysis was performed using the inverse variance weighted (IVW) approach, while sensitivity analyses included the weighted median, weighted mode, MR-Egger, and MR-PRESSO methods. Mediation analyses were performed using a two-step MR framework. RESULTS: In this MR cohort study, NAFLD was found to be strongly associated with elevated risks of GDM (P = 0.019 for the discovery dataset, P < 0.001 for the discovery dataset) and HDPs, including any HDP (P < 0.001 for the both datasets), gestational hypertension (P = 0.007 for the discovery dataset, P < 0.001 for the discovery dataset), and pre-eclampsia or eclampsia (P = 0.040 for the discovery dataset, P < 0.001 for the discovery dataset). However, no significant associations were found with hemorrhage in early pregnancy, postpartum hemorrhage, preterm birth, or offspring birthweight for both datasets. Cardiometabolic traits played a significant mediating role in these associations, rather than solely acting as confounding factors. CONCLUSIONS: This study provided evidence supporting a correlation between NAFLD and a higher risk of adverse pregnancy events and introduces some new insights. These findings may inform preventions and interventions for remediating adverse pregnancy outcomes attributable to NAFLD.