Abstract
OBJECTIVE: To investigate fasting C-peptide to high-density lipoprotein cholesterol ratio (FHR) as a predictor for non-alcoholic fatty liver disease (NAFLD) in Chinese adults with type 2 diabetes mellitus (T2DM). METHODS: This study enrolled 718 participants with T2DM from Shenzhen People's Hospital, China. Participants were stratified by FCP/HDL-C ratio (FHR) quartiles. Multiple linear regression assessed the association between FHR and NAFLD. A generalized additive model (GAM) tested for nonlinearity. Subgroup analyses evaluated result robustness. The area under the curve (AUC) evaluated the performance of the FHR model for NAFLD occurrence. RESULTS: After adjusting for relevant variables, FHR was positively correlated with NAFLD (OR = 1.30, 95% CI (1.15, 1.48)). FHR demonstrated a nonlinear association with NAFLD, characterized by a threshold value of 1.23. The effect sizes and confidence intervals on the left and right sides of the inflection point were 3.07 (1.51, 6.24) and 1.20 (1.05, 1.37), respectively. Subgroup analysis showed a stronger correlation could be detected in patients with systolic blood pressure (SBP) <140 mmHg, alanine transaminase (ALT) > 40U/L, fasting blood glucose (FBG) ⩽7 mmol/L, urinary albumin to creatinine ratio (UACR) ⩽30mg/g, triglyceride (TG) ⩽1.7 mmol/L and the patients with drinking history. The FHR ratio model exhibited better discriminative ability in NAFLD (AUC = 0.697) compared to individual FCP (AUC = 0.649) or HDL-C (AUC = 0.635) alone. CONCLUSION: The association between FHR and NAFLD was nonlinear, with a positive relationship observed when FHR exceeded the threshold of 1.23.