Reevaluating the causal link between liver fat and hyperglycaemia

Given the proposed role of liver fat in diabetes, this study examined its potential causal relationship with glycaemic traits.

A two-sample Mendelian randomisation (MR) analysis using genome-wide association study (GWAS) data from 37 358 individuals was conducted to evaluate the causal effect of liver fat on fasting glucose level, HbA1c level, postprandial glucose level, fasting insulin level, and homeostatic model assessment of insulin resistance (HOMA-IR). A meta-analysis of randomised controlled trials (RCTs) examining drugs that target hepatic steatosis and their effects on glycaemic traits was subsequently performed.

MR analysis revealed no causal effect of liver fat content on the following glycaemic traits: fasting glucose level (β = -0.006; p = 0.827), HbA1c level (β = -0.009; p = 0.366), postprandial glucose level (β = -0.063; p = 0.257), fasting insulin level (β = 0.010; p = 0.691), and HOMA-IR (β = 0.001; p = 0.965). In 13 RCTs of liver fat-targeted drugs (n = 2482), no significant differences were observed in the changes in fasting blood glucose (SMD: -0.09, 95% CI: -0.18 to 0.01) or HbA1c levels (SMD: -0.02, 95% CI: -0.12 to 0.07) between drug-treated patients and controls. Meta-regression analysis revealed no statistically significant linear relationship between liver fat reduction and changes in fasting blood glucose or HbA1c levels.

The absence of a causal relationship between hepatic steatosis and glycaemic traits suggests that hyperglycaemia in metabolic dysfunction-associated steatotic liver disease may involve additional mechanisms beyond liver fat accumulation.