High serum level of fibroblast growth factor 21 is an independent predictor of non-alcoholic fatty liver disease: A 3-year prospective study in China

Abstract

Fibroblast growth factor 21 (FGF21), a hormone predominantly secreted by the liver, has been shown to be positively associated with the severity of non-alcoholic fatty liver disease (NAFLD) in cross-sectional studies. We investigated the prospective association of FGF21 with NAFLD development in a 3-year prospective study involving a population-based cohort comprising 808 Chinese subjects. Serum FGF21 levels at baseline and follow-up were measured using an enzyme-linked immunosorbent assay. Independent predictors of NAFLD development were identified using multiple logistic regressions. The predicting accuracy of the models was evaluated using area under the receiver-operating characteristic (ROC) curves (AUCs). In subjects who had progressed to NAFLD, the baseline FGF21 concentration (319.12 pg/ml [172.65, 518.78]) was significantly higher than that in subjects who did not develop NAFLD (199.10 pg/ml [123.56, 322.80]) (p <0.001). At follow-up, significant increase of FGF21 level was observed in those subjects who developed NAFLD (p <0.05). Baseline FGF21 was an independent predictor of NAFLD (OR: 7.102 [95% CI 2.488-20.270]; p <0.001), together with body mass index (BMI) (OR: 1.489 [95% CI 1.310-1.691]; p <0.001). The ROC-AUC was 0.816 (95% CI 0.766-0.867) for the FGF21 Model, which was calculated with FGF21 and BMI. FGF21 Model <0.13 can be used to rule out (sensitivity=85.71%, negative likelihood ratio=0.23) and ≥0.30 can be rule in (specificity=86.34%, positive likelihood ratio=3.66) ultrasonography-diagnosed NAFLD after 3 years. High serum FGF21 concentration was an independent predictor of NAFLD in humans. The FGF21 Model and its cut-offs may be useful for early diagnosis and intervention of NAFLD.