Association of the total cholesterol–high-density lipoprotein–glucose index with metabolic-associated steatotic liver disease: a 5-year retrospective cohort study

Objective Current research on the relationship between the Total Cholesterol, High-Density Lipoprotein, and Glucose (CHG) index and the risk of Metabolic-Associated Steatotic Liver Disease (MASLD) is still limited. This study aims to explore the relationship between them and their predictive value in Chinese adults. Methods A retrospective cohort analysis enrolled 6,274 individuals who received health assessments at Shenzhen Kuichong People’s Hospital from 2018 to 2023. Cox proportional hazards regression analysis was implemented to evaluate associations between CHG and MASLD risk. Cox proportional hazards regression model with restricted cubic spline functions was utilized to assess potential non-linear associations. Additionally, ROC analysis evaluated CHG’s predictive ability for MASLD. Results Multivariable Cox regression showed that for each 0.1 unit increase in CHG, the HR for MASLD risk was 1.055 (95% CI: 1.030, 1.080). Non-linear association between them was identified, with inflection point at CHG = 5.42. Left of this inflection point, HR for CHG (per 0.1-unit)-MASLD association was 1.088 (95% CI: 1.055–1.122); right of this inflection point, the HR was 0.969 (95% CI: 0.912–1.029). Additionally, ROC curve analysis demonstrated that the AUC of CHG for predicting the 5-year risk of MASLD reached 0.678, which exceeded those of any single component indicator. When compared with TyG and FLI, CHG exhibited comparable predictive performance with no statistically significant differences. Time-dependent ROC analysis showed that within the 2.0–5-year period, the AUC of CHG remained between 0.656 and 0.678. Conclusion This study indicates that increased CHG is independently positively associated with the risk of MASLD and exhibits a non-linear association. Reducing CHG values below 5.42 and implementing further reduction measures may significantly decrease the risk of MASLD. CHG exhibits certain predictive value for MASLD risk and has the potential to serve as an early identification marker for high-risk populations, providing a novel perspective for clinical prevention strategies of MASLD.