Abstract
BACKGROUND: Fatty liver index (FLI) was recently established to predict non-alcoholic fatty liver disease (NAFLD) in general population, which is known to be associated with coronary artery atherosclerotic disease (CAD).This study aims to investigate whether FLI correlates with NAFLD and with newly diagnosed CAD in a special Chinese population who underwent coronary angiography. METHODS: Patients with CAD (n = 231) and without CAD (n = 482) as confirmed by coronary angiography were included. Among them, 574 patients underwent B-ultrosonography were divided into NAFLD group (n = 209) and non-NAFLD group (n = 365). Correlation between FLI and NAFLD was analyzed using pearson's correlation. The associations between FLI and NAFLD as well as CAD were assessed using logistic regression. The predictive accuracy of FLI for NAFLD was evaluated using receiver operating characteristics (ROC) curve analysis. RESULTS: FLI was significantly higher in NAFLD group (37.10 ± 1.95) than in non-NAFLD group (17.70 ± 1.04), P < 0.01. FLI correlated with NAFLD (r = 0.372, P < 0.001). The algorithm for FLI had a ROC-AUC of 0.721 (95% CI: 0.678-0.764) in the prediction of NAFLD. Logistic regression analysis showed that FLI was associated with NAFLD (adjusted OR = 1.038, 95% CI: 1.029-1.047, P < 0.01). The proportion of patients with CAD did not differ among the groups of FLI ≤ 30 (32.3%), 30-60 (31.0%), and ≥60 (35.3%). No significant association was found between FLI and CAD (adjusted OR = 0.992, 95% CI: 0.981-1.003 in men and OR = 0.987, 95% CI: 0.963-1.012 in women, P > 0.05). CONCLUSIONS: FLI showed good correlation with NAFLD in patients who underwent coronary angiography, but not with newly diagnosed CAD. This might be underestimated because some patients in non-CAD group may have other underlying cardiovascular diseases.