Assessing the Causal Relationships Between Lipid Species and Stroke by Using Mendelian Randomization.

Abstract

Circulating lipids and changes in lipid profiles have long been associated with the development of stroke but causal relationships remain unclear.In this study, we aimed to assess the causal relationships between lipid species and multiple stroke phenotypes to inform stroke prevention and treatment strategies. We conducted a two-sample Mendelian randomization analysis using data from genome-wide association studies. The primary method for causal assessment was inverse variance weighting (IVW), complemented by the MR-Egger, weighted median, and weighted mode methods. Sensitivity analyses, based on MR-Egger, MR-PRESSO, and Cochran's Q statistics, were also applied to reinforce the results. In total, potential causality was observed for 133 pairs of lipids with stroke types(P < 0.05). After multiple testing correction (PFDR < 0.05), potential causal associations remained for 10 pairs of lipids, including specific sterol esters and phosphatidylcholines, with various stroke subtypes. The findings demonstrate the significant role that genetically determined lipid profiles may play in the pathogenesis of stroke. Further research is needed to establish whether these biomarkers can be used for stroke prevention or treatment.