Department of General Surgery, Shanghai Tenth People's Hospital, Tongji University School of Medicine, Shanghai, China.
College of Traditional Chinese Medicine, Southern Medical University, Guangzhou, China.
Front Public Health. 2022 Nov 2;10:1049804. doi: 10.3389/fpubh.2022.1049804. eCollection 2022.
Few studies discussed the predictive ability of aspartate aminotransferase/alanine aminotransferase (AST/ALT, DeRitis) ratio for diabetes risk. The aim of this study was to characterize the role of AST/ALT ratio in the prediction of Chinese diabetes.
This retrospective cohort study analyzed a Chinese population comprising 87,883 participants without diabetes at baseline between 2010 and 2016. Cox proportional hazards regression was used to identify independent risk factors. Restricted cubic spline (RCS) was performed to investigate the non-linear correlation between AST/ALT ratio and diabetes risk.
During a median follow-up period of 3.01 years, 1,877 participants developed diabetes. Comparing the baseline characteristics, diabetes group exhibited lower AST/ALT ratio. The Kaplan-Meier curve showed that participants with low AST/ALT ratio had higher cumulative incidence, and Cox regression also demonstrated that the lower AST/ALT ratio, the higher diabetes risk (HR: 0.56, 95% CI: 0.37-0.85, = 0.006). The RCS model revealed a non-linear correlation between AST/ALT ratio and diabetes risk. In the condition of AST/ALT ratio ≤1.18, diabetes risk increased as it decreased (HR: 0.42, 95% CI: 0.19-0.91, = 0.028). In contrast, AST/ALT ratio did not independently affect diabetes when beyond 1.18.
AST/ALT ratio is a valuable predictor of diabetes. Diabetes risk increases rapidly in the condition of AST/ALT ratio ≤1.18.
很少有研究探讨天冬氨酸氨基转移酶/丙氨酸氨基转移酶（AST/ALT，DeRitis）比值对糖尿病风险的预测能力。本研究旨在探讨 AST/ALT 比值在中国人群糖尿病预测中的作用。
本回顾性队列研究分析了 2010 年至 2016 年间基线时无糖尿病的 87883 例中国人群。使用 Cox 比例风险回归模型确定独立的危险因素。采用限制性三次样条（RCS）分析 AST/ALT 比值与糖尿病风险之间的非线性关系。
在中位随访 3.01 年期间，1877 例参与者发生糖尿病。与基线特征相比，糖尿病组的 AST/ALT 比值较低。Kaplan-Meier 曲线显示 AST/ALT 比值较低的参与者累积发病率较高，Cox 回归也表明 AST/ALT 比值越低，糖尿病风险越高（HR：0.56，95%CI：0.37-0.85，P=0.006）。RCS 模型显示 AST/ALT 比值与糖尿病风险之间存在非线性关系。在 AST/ALT 比值≤1.18 的情况下，随着比值的降低，糖尿病风险增加（HR：0.42，95%CI：0.19-0.91，P=0.028）。而当 AST/ALT 比值超过 1.18 时，其对糖尿病无独立影响。
AST/ALT 比值是糖尿病的有价值预测指标。在 AST/ALT 比值≤1.18 的情况下，糖尿病风险迅速增加。
