Changes in Biochemical Parameters and Their Correlation with Disease Severity in Patients with Metabolic Associated Fatty Liver Disease (MAFLD): A Cross-sectional Study.

INTRODUCTION: Prior studies often lack a comprehensive comparison of biochemical profiles between patients with Metabolic-Associated Fatty Liver Disease (MAFLD) and healthy individuals. This study aimed to explore changes in biochemical parameters in MAFLD and their correlation with fatty liver severity after controlling for major confounders. METHODS: Participants diagnosed with MAFLD at Beijing Jishuitan Hospital (August 2021-October 2024) and healthy controls were enrolled. A 1:1 Propensity Score Matching (PSM) was applied to balance gender, age, and Body Mass Index (BMI). Multivariate logistic regression (adjusted for comorbidities) and ROC analysis (including the TyG index) were performed to identify independent risk factors and evaluate diagnostic performance. MAFLD patients were stratified by ultrasound into mild, moderate, and severe groups. RESULTS: After PSM (n = 1,138 pairs), demographics were well-balanced (p > 0.05). MAFLD patients exhibited significantly higher levels of TG, TC, apoA, FPG, FINS, ALT, AST, ALP, and γ- GT (all p < 0.05 or p < 0.001) compared to matched controls. Subgroup analysis showed that diabetes and hypertension significantly exacerbated biochemical abnormalities. HbA1c (p = 0.037), ALT (p < 0.001), AST (p = 0.007), and γ-GT (p = 0.022) increased progressively with fatty liver severity. ROC analysis revealed that a Combined Model (ALT + γ-GT + BMI) yielded the highest diagnostic efficacy (AUC = 0.824, 95% CI: 0.795-0.853), followed by ALT (AUC = 0.725) and the TyG index (AUC = 0.710). DISCUSSION: These findings underscore the critical role of integrating liver enzymes with metabolic indices for early MAFLD detection, suggesting that routine monitoring of ALT, γ-GT, and BMI can significantly enhance screening efficiency in primary care settings with limited ultrasound resources. CONCLUSION: Biochemical parameters, particularly ALT, AST, γ-GT, and HbA1c, are closely related to MAFLD severity. While individual markers have limited diagnostic power, the Combined Model provides superior accuracy for early screening. Regular monitoring of these core indicators facilitates early diagnosis and intervention.