Chinese Hepatolgy ›› 2026, Vol. 31 ›› Issue (5): 697-703.

• Metabolic Associated Fatty Liver Disease • Previous Articles     Next Articles

The value of the ZJU index, TyG index, and their derivative index in screening and predicting the severity of metabolic dysfunction-associated fatty liver disease

LIU Xiao-tong, ZHANG Bao-jing, WANG Man, WU Feng-xia, SONG Chen-yuan, WANG Shu-jun   

  1. Department of Gastroenterology, Zhengzhou First People′s Hospital, Zhengzhou 450000, China
  • Received:2025-09-21 Published:2026-07-10
  • Contact: WANG Shu-jun, Email: wangshujun80@163.com

Abstract: Objective To explore the predictive value of the Zhejiang University index (ZJU index), triglyceride-glucose index (TyG index), and their derived indices for metabolic associated fatty liver disease (MAFLD) and its severity. Methods A total of 522 patients in the health examination center and gastroenterology department of Zhengzhou First People′s Hospital from November 2023 to April 2025 were enrolled. According to the diagnosis of MAFLD by liver transient elastography, the participants were divided into MAFLD group (295 cases) and non-MAFLD group (227 cases). According to the controlled attenuation parameters detected by transient elastography, the MAFLD group was divided into mild group, moderate group, and severe group. Logistic regression and ROC curve analysis were conducted to explore the correlation and predictive value of the ZJU index, TyG index, and TyG-BMI index with the occurrence and severity of MAFLD. Results The MAFLD group exhibited significantly higher metabolic indices than the non-MAFLD group (ZJU: 39.2 ± 5.5 vs. 31.8 ± 3.7; TyG: 9.1 ± 0.6 vs. 8.3 ± 0.5; TyG-BMI: 243.2 ± 37.3 vs. 186.6 ± 28.9; all P<0.05). Furthermore, ZJU, TyG, and TyG-BMI progressively increased across mild, moderate, and severe MAFLD subgroups, and the differences were statistically significant (P< 0.05). Multivariate logistic regression identified elevated ZJU (OR=1.058), TyG (OR=20.815), and TyG-BMI (OR=1.485) as independent risk factors for MAFLD occurrence (all P<0.05), and for moderate-to-severe MAFLD development (OR=1.517, 11.702, and 1.061; all P<0.05). For predicting MAFLD, AUC values were: TyG, 0.839 (95% CI 0.805~0.873; cut-off 8.72); ZJU, 0.876 (95% CI 0.847~0.905; cut-off 34.41); TyG-BMI, 0.891 (95% CI 0.864~0.919; cut-off 212.84). Corresponding AUCs for moderate-to-severe MAFLD prediction were: TyG, 0.832 (95% CI 0.796~0.867; cut-off 8.79); ZJU, 0.916 (95% CI 0.893~0.939; cut-off 36.19); TyG-BMI, 0.916 (95% CI 0.892~0.940; cut-off 224.47). Conclusion The TyG, ZJU and TyG-BMI are all independent risk factors for MAFLD and moderate to severe MAFLD, and have predictive value for MAFLD and moderate to severe MAFLD, among which TyG-BMI has a higher predictive value for MAFLD. The ZJU and TyG-BMI have higher predictive value for moderate to severe MAFLD.

Key words: Metabolically associated fatty liver disease, Zhejiang University index, Triglyceride glucose index, Triglyceride-glucose-body mass index, Transient elastography