Objectives: It is unclear whether liver enzymes or the interactions of various liver enzymes is a predictor of type 2 diabetes mellitus (T2DM), which is independent of fatty liver.
Methods: A total of 48,001 subjects participated in baseline examinations. Among the subjects, 33,355 were followed for an average of 2.2 years. Cox proportional hazard models were used to examine the adjusted associations of AST, GGT and ALT with T2DM.
Results: The cumulative incidence of T2DM was 8.05% to 9.02% for fatty liver and 2.25% to 4.10% for non-fatty liver, both showing statistically significant differences. Compared with the normal liver enzyme levels in the group with fatty liver, the adjusted incident hazard ratios in T2DM were: ALT 1.23 (95% CI 1.10 to 1.50); AST 1.30 (95% CI 1.07-1.59); and GGT 1.34 (95% CI 1.08 to 1.65). In addition, compared with the normal liver enzyme levels in the group with non-fatty liver, the adjusted incident hazard ratios in type 2 diabetes were: ALT 1.27 (95% CI 1.02 to 1.59); AST 1.33 (95% CI 1.02 to 1.59); and GGT 1.53 (95% CI 1.19 to 1.98). There are significant interactions of T2DM hazard ratios between GGT and ALT and between GGT and AST in addition to ALT and AST.
Conclusions: Our results suggest that the incidence of T2DM in the group with fatty liver is significantly higher than that in the normal population, and the rise of serum AST, GGT and ALT levels are risk factors independent of fatty liver for the development of T2DM after adjusting for confounding factors.
Keywords: association; cohort; cohorte; diabète de type 2; enzyme; enzymes; fatty liver; stéatose hépatique; type 2 diabetes.
Copyright © 2018 Diabetes Canada. Published by Elsevier Inc. All rights reserved.