Abstract
The Hepatic Steatosis Index (HSI) is a well-established metric for evaluating hepatic fat accumulation. Although prior studies have examined the effect of HSI on diabetes, its association with prediabetes remains incompletely delineated. This study aims to elucidate the effect of HSI on prediabetes within a Chinese population. We conducted a multicenter cohort analysis involving 75,750 baseline normoglycemic adults from the Rich Healthcare Group in China. To assess the effect of HSI on prediabetes, this study employed Cox proportional hazards regression, restricted cubic spline models, subgroups, and sensitivity analyses. Multivariable-adjusted models identified HSI as an independent predictor of prediabetes (HR: 1.06, 95% CI: 1.04-1.07 for females, HR: 1.04, 95% CI: 1.03-1.05 for males). A nonlinear relationship between HSI and prediabetes risk was observed in both sexes, with an inflection point at 36.48 for females and 31.27 for males. Among females, a significant positive link with prediabetes emerged below the threshold of 36.48 (HR: 1.08, 95% CI: 1.06-1.10, P < 0.0001), while above this level the association was attenuated and failed to reach statistical significance (HR: 1.00, 95% CI: 0.97-1.02, P = 0.7579). In contrast, hazard ratios for males were 1.02 (95% CI: 1.00-1.04) below and 1.04 (95% CI: 1.03-1.05) above their respective inflection point. Our analysis revealed a non-linear positive correlation between HSI and prediabetes across both sexes. Precise assessment of HSI facilitates early identification of subjects at heightened risk, thereby enabling the implementation of targeted preventive strategies.