The development and evaluation of nine non-conventional lipid parameters for metabolic dysfunction-associated fatty liver disease in Chinese medical health examination adults: a single-centre retrospective study

Lian Song¹Lirong Zhang¹Yinhui Hang¹Zijie Yang^1,2Jing Yang^1,2Dongqing Wang^1*

• ¹Affiliated Hospital of Jiangsu University, Zhenjiang, China
• ²Jiangsu University, Zhenjiang, China

Objective: Metabolic dysfunction-associated fatty liver disease (MAFLD) represents a prevalent chronic hepatic condition globally, characterized by hepatic steatosis concurrent with at least one cardiometabolic risk factor, such as overweight/obesity, type 2 diabetes mellitus (T2DM), or metabolic dysregulation. This study aimed to evaluate the associations between nine non-conventional lipid parameters—BMI, NHHR, AIP, RC, GHR, CHG, LCI, TyG, TyG-BMI—and MAFLD, and to compare their predictive performance for MAFLD screening. Methods: This study utilized the electronic medical record at Wuhan Union Hospital between January 2020 and November 2021, and multi-model adjustment weighted logistic regression analysis was applied to investigate the association of the nine parameters with MAFLD. Receiver operating characteristic (ROC) curves were analyzed to assess the screening ability of the nine parameters. Furthermore, the association between the most predictive parameter and MAFLD was investigated with RCS analysis, and differences in risk across populations were explored with subgroup analyses. Results: A total of 1,592 participants were included in the final analysis, among whom 937 (58.86%) were diagnosed with MAFLD. Multivariable logistic regression identified NHHR, BMI, AIP, RC, GHR, LCI, TyG, and TyG-BMI as independent risk factors for MAFLD, with TyG-BMI demonstrating the strongest association (OR = 3.7, 95% CI: 3.05–4.48). The area under the ROC curve (AUC) for TyG-BMI was 0.81, and its predictive performance was significantly superior to that of the other parameters (all P ﹤ 0.001 by DeLong’s test). RCS analysis revealed a nonlinear relationship between TyG-BMI and MAFLD (P for nonlinearity﹤0.001), with an identified inflection point at a TyG-BMI value of 222.426. Additionally, MAFLD patients in the highest TyG-BMI tertile exhibited a significantly increased risk of atherosclerotic cardiovascular disease (ASCVD) compared to those in the lowest tertile (OR = 2.55, 95% CI: 1.337–4.91) after adjustment for confounders. Conclusion: The evaluated non-conventional lipid parameters, particularly TyG-BMI, are useful indicators for MAFLD identification. TyG-BMI demonstrated the strongest predictive ability for MAFLD and was independently associated with ASCVD risk in affected individuals. Elevated TyG-BMI may therefore serve as a clinically accessible marker for identifying individuals at high risk of MAFLD and for stratifying cardiovascular risk in patients with established MAFLD.