Causal association between serum bilirubin and ischemic stroke: Multivariable Mendelian Randomization

BACKGROUNDPast studies have mainly focused on total bilirubin levels, and have not clearly distinguished between direct and indirect bilirubin, a subgroup of bilirubin. In this study, the differences between these subgroups were clearly examined, and the causal association with ischemic stroke was examined in more detail. METHODSUtilizing Two sample Multivariable Mendelian randomization (MVMR) analyses, summary data for bilirubin were extracted from the KCPS-II (Korean Cancer Prevention Study-II; n=159,844) and the KoGES (Korean Genome and Epidemiology Study; n=72,299), while ischemic stroke data were derived from the BBJ (Bio Bank Japan; n=201,800). RESULTSThe crude two-sample MR analysis revealed a significant negative association between total bilirubin and ischemic stroke in KoGES data (odds ratio [OR], 0.86; 95% confidence interval [CI], 0.75-0.98). Subsequent bivariable MR analyses, controlling for lipid profile, also showed significant results. In KCPS-II data, direct bilirubin showed significance in both crude (0.65, 0.43-0.97) and bivariable analyses, while indirect bilirubin demonstrated significant associations in MVMR analyses (0.76, 0.59-0.98), emphasizing its role in mitigating the risk of ischemic stroke. CONCLUSIONSOur study establishes a causal association between genetically determined levels of serum bilirubin (total, direct, and indirect) and a reduced risk of ischemic stroke in an Asian population. Notably, the protective effect was predominantly associated with indirect bilirubin. The findings highlight the significance of considering bilirubin subgroup in understanding the mechanisms underlying endogenous antioxidants and its impact on ischemic stroke.