The causal associations of non-alcoholic fatty liver disease with blood pressure and the mediating effects of cardiometabolic risk factors: A Mendelian randomization study

Abstract

Background and aimsNon-alcoholic fatty liver disease (NAFLD) is prevalent in hypertensive people, but the causal effect remains unclear. We employed Mendelian randomization (MR) approach to assess the causality between NAFLD and different blood pressure (BP) parameters. Method and resultsInstrumental variables for genetically predicted NAFLD, including chronically elevated serum alanine aminotransferase levels (cALT) and imaging and biopsy-confirmed NAFLD, were obtained from a genome-wide association study (N = 164,197). Multiple MR methods were implemented, including Inverse variance weighted, MR-Egger, Maximum likelihood, Weighted median, Simple median, Penalised weighted median, MR-RAPS, and cML-MA. Outliers were detected using MR-PRESSO, and pleiotropy was assessed using MR-Egger intercept and Phenoscanner. Heterogeneity was quantified using Cochran's Q and Rucker's Q′ tests. Potential shared risk factors were analyzed to reveal the mediating effect. A higher genetic predisposition to cALT was causally associated with an increased risk of elevated BP levels, resulting in 0.65 mmHg (95% CI, 0.42–0.87), 0.38 mmHg (95% CI, 0.25–0.50) and 0.33 mmHg (95% CI, 0.22–0.44) higher for systolic BP, diastolic BP and pulse pressure, respectively. When more stringent criteria were used, imaging and biopsy-confirmed NAFLD showed a 1.12 mmHg (95% CI, 0.94–1.30) increase in SBP and a 0.55 mmHg (95% CI, 0.39–0.70) increase in DBP. Risk factor and mediation analyses suggested type 2 diabetes and fasting insulin levels might mediate the causal relationship between NAFLD and BP. ConclusionThe two-sample MR analyses showed robust causal effects of genetically predicted NAFLD on 3 different BP indices. The shared genetic profile between NAFLD and BP may suggest important therapeutic targets and early interventions for cardiometabolic risk factors.