| 摘要: |
| 目的 探讨甘油三酯葡萄糖指数（TyG）及其衍生指标TyG-BMI、TyG-丙氨酸转氨酶（ALT）与瘦型代谢相关脂肪性肝病（MAFLD）患病风险的相关性。方法 选择2023年7－12月在我院健康管理中心进行年度健康体检并诊断为瘦型MAFLD的207例患者及100名瘦型健康对照为研究对象，比较两组血脂、血糖、肝功能、TyG、TyG-BMI、TyG-ALT等，采用单因素和多因素logistic回归模型分析影响瘦型MAFLD的因素。将所有研究对象按照TyG及其衍生指标四分位数分为4个亚组（Q1～Q4），观察各亚组瘦型MAFLD的患病率。绘制TyG、TyG-BMI、TyG-ALT预测瘦型MAFLD的ROC曲线，评估各指标的预测效能。结果 纳入的8 764例体检者中MAFLD患者2 350例（26.8%），其中瘦型207例（8.8%，207/2 350）。与瘦型健康对照组相比，瘦型MAFLD组患者的年龄大、男性多见、BMI高，空腹血糖、总胆固醇、甘油三酯、低密度脂蛋白胆固醇、ALT、天冬氨酸转氨酶、γ-谷氨酰转肽酶、碱性磷酸酶、总胆红素、TyG、TyG-BMI与TyG-ALT均升高，高密度脂蛋白胆固醇降低（均P＜0.01）。logistic回归分析显示，年龄、男性、ALT水平升高是瘦型MAFLD患病的独立危险因素。TyG的Q4亚组瘦型MAFLD的患病率为34.3%（71/207），高于Q1亚组（10.6%，22/207）及Q2亚组（24.2%，50/207），差异均有统计学意义（均P＜0.05）；TyG-BMI的Q4亚组和TyG-ALT的Q4亚组瘦型MAFLD的患病率均高于其Q1、Q2、Q3亚组［35.3%（73/207） vs 8.2%（17/207）、24.6%（51/207）、31.9%（66/207），33.8%（70/207） vs 14.0%（29/207）、23.2%（48/207）、29.0%（60/207）］，差异均有统计学意义（均P＜0.05）。TyG-BMI预测瘦型MAFLD的AUC值为0.869 0（95%CI0.825 5～0.912 6，P＜0.001），高于TyG［AUC＝0.818 8（95%CI0.768 0～0.869 6，P＜0.001）］和TyG-ALT［AUC＝0.772 5（95%CI0.718 7～0.826 2，P＜0.001）］。结论 TyG、TyG-BMI与TyG-ALT均与瘦型MAFLD相关，三者对瘦型MAFLD具有一定的预测价值；TyG及其衍生指标计算简易且价格低廉，可用于对瘦型MAFLD的临床初步评估。 |
| 关键词: 代谢相关脂肪性肝病 瘦型 甘油三酯葡萄糖指数 体重指数 丙氨酸转氨酶 |
| DOI：10.16781/j.CN31-2187/R.20240019 |
| 投稿时间：2024-01-08修订日期：2024-02-27 |
| 基金项目:上海市青年科技英才扬帆计划项目（21YF1458800），军队护理创新与培育专项计划（2021HL029）． |
|
| Predictive value of triglyceride-glucose index and its derivatives for lean metabolic associated fatty liver disease |
| DONG Xu1△,GE Yu2△,WANG Chaoqun3,XU Mingxiao1,CHEN Yi1,CHEN Lin4* |
(1. Department of Infectious Diseases, The First Affiliated Hospital of Naval Medical University (Second Military Medical University), Shanghai 200433, China;
2. Department of Ultrasound, The First Affiliated Hospital of Naval Medical University (Second Military Medical University), Shanghai 200433, China;
3. Department of Endocrinology, The First Affiliated Hospital of Naval Medical University (Second Military Medical University), Shanghai 200433, China;
4. Health Management Center, The First Affiliated Hospital of Naval Medical University (Second Military Medical University), Shanghai 200433, China
△Co-first authors.
* Corresponding author) |
| Abstract: |
| Objective To explore the correlations between triglyceride glucose index (TyG) and its derivatives TyG-body mass index (BMI) and TyG-alanine transaminase (ALT) with the risk of lean metabolic associated fatty liver disease (MAFLD). Methods A total of 207 patients diagnosed with lean MAFLD and 100 lean healthy controls who received annual health examination in Health Management Center of our hospital from Jul. to Dec. 2023 were enrolled. Plasma lipids, blood glucose, liver function, TyG, TyG-BMI and TyG-ALT were compared between the 2 groups. The influencing factors of lean MAFLD were analyzed by univariate and multivariate logistic regression models. All subjects were divided into 4 subgroups (Q1-Q4) according to the quartile of TyG and its derivatives, and the prevalence of lean MAFLD in each subgroup was observed. The receiver operating characteristic (ROC) curves of TyG, TyG-BMI and TyG-ALT for lean MAFLD were plotted to evaluate the prediction efficiency. Results Of the 8 764 health examination cases included, 2 350 (26.8%) had MAFLD, of which 207 were lean MAFLD (8.8%, 207/2 350). Compared with the lean healthy controls, the patients in the lean MAFLD group were older, with more male and high BMI, and their fasting blood glucose, total cholesterol, triglyceride, low density lipoprotein-cholesterol, ALT, aspartate transaminase, γ-glutamyl transpeptidase, alkaline phosphatase, total bilirubin, TyG, TyG-BMI and TyG-ALT were significantly increased, while high density lipoprotein-cholesterol was significantly decreased (all P＜0.01). Logistic regression analysis showed that age, male, and elevated ALT level were independent risk factors for lean MAFLD. The prevalence of lean MAFLD in the Q4 subgroup of TyG was significantly higher than that in the Q1 and Q2 subgroups (34.3% ［71/207］ vs 10.6% ［22/207］ and 24.2% ［50/207］, both P＜0.05). The prevalence rates of lean MAFLD in the Q4 subgroup of TyG-BMI and the Q4 subgroup of TyG-ALT were significantly higher than those in the corresponding Q1, Q2, and Q3 subgroups (35.3% ［73/207］ vs 8.2% ［17/207］, 24.6% ［51/207］, and 31.9% ［66/207］; 33.8% ［70/207］ vs 14.0% ［29/207］, 23.2% ［48/207］, and 29.0% ［60/207］; all P＜0.05). The area under curve (AUC) of TyG-BMI in predicting lean MAFLD was 0.869 0 (95% confidence interval ［CI］ 0.825 5-0.912 6, P＜0.001), which was higher than that of TyG (AUC＝0.818 8 ［95% CI 0.768 0-0.869 6, P＜0.001］) and TyG-ALT (AUC＝0.772 5 ［95% CI 0.718 7-0.826 2, P＜0.001］). Conclusion TyG, TyG-BMI, and TyG-ALT are associated with lean MAFLD, and have predictive value for lean MAFLD. TyG and its derivatives are easy to calculate and cheap, and can be used for preliminary clinical assessment of lean MAFLD. |
| Key words: metabolic associated fatty liver disease lean type triglyceride-glucose index body mass index alanine transaminase |
