Abstract: Objective To explore the association between triglyceride-glucose-body mass index (TyG-BMI) and nonalcoholic fatty liver disease (NAFLD) in lean population and its predictive value. Methods 2553 healthy subjects from the Second Affiliated Hospital of Xi′an Jiaotong University from June 2020 to May 2021 were randomly included and divided into Q1-Q4 groups according to the TyG-BMI quartiles. The correlation between TyG-BMI and lean NAFLD was analyzed with restricted cubic spline (RCS), and its predictive value was explored by logistic regression and receiver operating characteristic (ROC) curve. Results There were 209 patients with lean NAFLD (8.2%), and the prevalence of NAFLD in Q1-Q4 groups were 0.3%, 1.1%, 7.1% and 24.3%, respectively (P<0.001). RCS showed that the risk of NAFLD in lean population upraised significantly with the increase of TyG-BMI (P<0.001), and there was a nonlinear relationship between them (P for nonlinear < 0.001). Logistic regression showed that after adjusting confounding factors, increased TyG-BMI was significantly associated with the increased risk of lean NAFLD (odds ratio [OR] 1.076, 95% CI: 1.066~1.086), and the risk of lean NAFLD in Q3 (177.85, 194.28) and Q4 (194.28, 269.07) groups was still 5.545 times (95% CI: 1.137~27.039) and 9.853 times (95% CI: 1.788~54.279) than Q1 (115.92, 162.43) group. ROC curve showed that TyG-BMI had a good predictive value for lean NAFLD, with area under the curve of 0.858, critical value of 187.95, the sensitivity and specificity were 89.5% and 70.8%, respectively, which were better than TyG, BMI and waist circumference (WC). Conclusion The TyG-BMI is significantly associated with the risk of NAFLD in lean population, and has a good predictive value, which can be used for early screening and diagnosis of lean NAFLD.

Key words: Nonalcoholic fatty liver disease, Lean population, Triglyceride-glucose-body mass index, Insulin resistance