Key points are not available for this paper at this time.
Kawasaki Disease (KD) is an acute pediatric vasculitis and the leading cause of acquired heart disease in children, yet its etiology involving lipid metabolism remains unclear. This study aimed to investigate the potential causal associations between plasma lipid species, measured in a pediatric cohort, and the risk of KD using a two-sample Mendelian randomization (MR) approach. We utilized summary statistics from a genome-wide association study (GWAS) of 227 plasma lipid species in 1,149 children and adolescents (median age 11.2 years). Summary statistics for KD were obtained from a meta-analysis of two GWAS datasets comprising 479 cases and 490,361 controls of European ancestry. We employed inverse-variance weighted (IVW) MR as the primary analysis, complemented by nine robust univariable MR methods (UVMR), five multivariable MR (MVMR) methods (MVMR-IVW, -Egger, -PRESSO, -Lasso, -Median), and MR-Bayesian Model Averaging (MR-BMA) to assess causality and account for pleiotropy and lipid correlations. The primary IVW analysis suggested nine lipid species associated with KD risk. Robust UVMR analyses provided corroborating evidence, notably showing consistent negative associations for Lysophosphatidylcholine (0:0/20:0) and Triglyceride (52:5) across multiple methods. Subsequent MVMR consistently identified positive causal associations between Phosphatidylcholine (40:8), Phosphatidylinositol (36:4), and Sphingomyelin (d36:3) and increased risk of KD. Furthermore, MR-BMA identified Triglyceride (56:5) as the lipid species with the strongest evidence for a protective effect (highest marginal inclusion probability = 0.275). Leveraging genetic data specific to a pediatric population, this MR study provides novel evidence supporting potential causal roles for specific plasma lipid species in the pathogenesis of Kawasaki disease. These findings offer new insights into the molecular mechanisms underlying Kawasaki disease and highlight potential pathways for future investigation and therapeutic targeting.
Zhen et al. (Tue,) studied this question.