Causal relationships between metabolic‐associated fatty liver disease and iron status: Two‐sample Mendelian randomization

Abstract Background & Aims Dysregulated iron homeostasis plays an important role in the hepatic manifestation of metabolic‐associated fatty liver disease (MAFLD). We investigated the causal effects of five iron metabolism markers, regular iron supplementation and MAFLD risk. Methods Genetic summary statistics were obtained from open genome‐wide association study databases. Two‐sample bidirectional Mendelian randomization analysis was performed to estimate the causal effect between iron status and MAFLD, including Mendelian randomization inverse‐variance weighted, weighted median methods and Mendelian randomization‐Egger regression. The Mendelian randomization‐PRESSO outlier test, Cochran's Q test and Mendelian randomization‐Egger regression were used to assess outliers, heterogeneity and pleiotropy respectively. Results Mendelian randomization inverse‐variance weighted results showed that the genetically predicted per standard deviation increase in liver iron (Data set 2: odds ratio 1.193, 95% confidence interval [CI] 1.074–1.326, p = .001) was associated with an increased MAFLD risk, consistent with the weighted median estimates and Mendelian randomization–Egger regression, although Data set 1 was not significant. Mendelian randomization inverse‐variance weighted analysis showed that genetically predicted MAFLD was significantly associated with increased serum ferritin levels in both datasets (Dataset 1: β = .038, 95% CI = .014 to .062, p = .002; Dataset 2: β = .081, 95% CI = .025 to .136, p = .004), and a similar result was observed with the weighted median methods for Dataset 2 instead of Mendelian randomization‐Egger regression. Conclusions This study uncovered genetically predicted causal associations between iron metabolism status and MAFLD. These findings underscore the need for improved guidelines for managing MAFLD risk by emphasizing hepatic iron levels as a risk factor and ferritin levels as a prognostic factor.