Abstract: Objective To investigate the correlation between the ratio of fasting blood glucose to high density lipoprotein cholesterol (GHR) and metabolically associated fatty liver disease (MAFLD). Methods A retrospective study was conducted on individuals who underwent physical examinations at our hospital from January 2021 to December 2023. Multivariate logistic regression was used to analyze the association between GHR and MAFLD. The restricted cubic spline (RCS) was employed to explore the non-linear relationship, and the predictive efficacy of GHR was evaluated through the receiver operating characteristic curve (ROC). Results A total of 17 472 subjects were included, among which 6 403 had MAFLD and 11 069 did not. After stepwise adjustment for various confounding factors, GHR was still significantly positively correlated with MAFLD (P<0.001); for each unit increase in GHR, the risk of MAFLD increased by 14.9%. Compared with the lowest GHR quartile Q1 group, the risks of MAFLD in Q2, Q3, and Q4 groups were 1.507, 2.045, and 2.721 times that of the Q1 group, respectively; the risk of MAFLD increased with the increase in GHR quartiles (P_trend<0.001). RCS analysis indicated a nonlinear association between GHR and MAFLD (P<0.001), with an inverted L-shaped curve and a turning point at 5.739. ROC curve analysis showed that GHR was significantly better than HDL-C or FBG alone in predicting MAFLD, with an area under the curve (AUC) of 0.769. Conclusion GHR is an independent risk factor for MAFLD and may serve as a promising biomarker for predicting MAFLD.

Key words: Metabolic associated fatty liver disease, Fasting blood glucose, High density lipoprotein cholesterol