The Causal Relationship Between Circulating Metabolites and the Risk of Atopic Dermatitis : A Two-Sample Mendelian Randomization Study

Background: Previous research has shown that metabolites (especially lipid-related metabolites) have a significant influence in the development of atopic dermatitis (AD). However, there is no evidence of a causal connection between metabolites and AD risk. The specific mechanisms require further elucidation. Our study employed a two-sample Mendelian randomization (TSMR) strategy to investigate how metabolite traits affect AD. Methods: Utilizing publicly accessible GWAS data, we conducted TSMR studies to investigate the relationship between 233 metabolites traits (213 lipid-related traits and 20 no lipid-related traits) and AD. Our TSMR study primarily employed the Inversevariance weighted method and four ancillary methods to analyze causation. Sensitivity analysis was performed to guarantee the TSMR results were trustworthy. Reverse MR analysis was used for investigating reverse causality. Results: After analyzing GWAS datasets for metabolites and AD, 13 metabolites were identified as positive. The MR analysis result indicates that total cholesterol in very small VLDL, cholesterol esters in very small VLDL, free cholesterol in IDL, concentration of medium LDL particles, concentration of large LDL particle, concentration of chylomicrons and extremely large VLDL particles, triglyceride levels in chylomicrons and extremely large VLDL, total lipid levels in chylomicrons and extremely large VLD, phospholipid levels in chylomicrons and extremely large VLDL, phospholipids in medium LDL, phospholipids in large LDL, phospholipids in small LDL, ratio of 18:2 linoleic acid to total fatty acids exhibited negative effects on AD. Reverse MR result analysis found that ratio of 18:2 linoleic acid to total fatty acids in serum was decreased in patients with AD. Sensitivity analyses ensure the stability of our results. Conclusion: These findings highlight a definite correlation between metabolite and AD, demonstrating the significant role of 13 lipidrelated metabolite traits. Our results significantly reduced the influence of unavoidable confounders and reverse causality. Our findings may set the framework for prospective therapeutic approaches and call for further investigation to validate them.