甲襞微循环联合甘油三酯葡萄糖乘积指数对糖尿病合并冠心病的预测价值

Predictive Value of Nailfold Microcirculation Combined with Triglyceride Glucose Product Index in Diabetes Mellitus Combined with Coronary Heart Disease

摘要: 目的：评估甲襞微循环联合甘油三酯葡萄糖乘积指数(TyG指数)对糖尿病合并冠心病患者的预测价值。方法：选取2022年9月至2023年9月于烟台毓璜顶医院住院且符合条件的55名糖尿病患者为研究对象。回顾性收集患者一般资料、冠脉造影检查资料、实验室相关指标等数据。采用甲襞微循环检测仪收集所有患者甲襞毛细血管的检查数据，采用多因素Logistic回归分析TyG指数和甲襞微循环指标与糖尿病患者合并冠心病的关系，绘制ROC曲线，评估各指标对糖尿病患者合并冠心病的预测价值。结果：糖尿病合并冠心病的患者更可能是男性、高血压、糖化血红蛋白较高、空腹血糖较高、肌钙蛋白较高和左室射血分数较低的患者。糖尿病合并冠心病患者的甲襞毛细血管的变化主要表现为管袢长度更短、交叉管袢数更多，冠脉病变严重程度与交叉管袢数成正相关。将差异性分析中，P < 0.1的变量纳入模型中，根据模型结果显示，交叉管袢数、TyG指数、性别、TyG指数均为糖尿病患者合并冠心病的独立预测因素，其中TyG指数的预测价值最优。模型的ROC曲线结果显示，AUC为0.960 (0.915, 1.000)，灵敏度为0.878，特异度为0.929，正确指数为0.807。临床预测模型显示该模型在5%~100%的阈概率区间内均表现出更高的临床净获益，且校准图中显示拟合优度较好。结论：糖尿病并发冠心病的患者外周微循环改变更明显，交叉管袢数、TyG指数、性别、高血压均为糖尿病患者并发冠心病的独立预测因素，四者联合对糖尿病合并冠心病患者有较好预测价值。

Abstract: Objective: To evaluate the predictive value of nailfold microcirculation combined with the triglyceride-glucose product index (TyG index) in patients with diabetes mellitus complicated by coronary heart disease. Methods: Fifty-five diabetic patients who were hospitalized in Yantai Yuhuangding Hospital from September 2022 to September 2023 and were eligible for the study were selected. General data, coronary angiography data, and laboratory-related indexes were retrospectively collected. Nailfold microcirculation detector was used to collect all the patients’ nailfold capillary examination data, and the relationship between TyG index and nailfold microcirculation indexes and diabetic patients’ coronary artery disease was analyzed by multifactorial Logistic regression, and ROC curves were plotted to evaluate the predictive value of the indexes on diabetic patients’ coronary artery disease. Results: Patients with diabetes mellitus combined with coronary artery disease are more likely to be male, hypertensive, and have higher glycosylated hemoglobin, higher fasting glucose, higher troponin, and lower left ventricular ejection fraction. Changes in the nailfold capillaries in patients with diabetes mellitus combined with coronary artery disease were mainly characterized by shorter length of collaterals and higher number of crossed collaterals, and the severity of coronary artery disease was positively correlated with the number of crossed collaterals. Variables with P < 0.1 in the analysis of variance were included in the model, and according to the results of the model, the number of cross-tubular collaterals, TyG index, gender, and TyG index were all independent predictors of diabetic patients with comorbid coronary artery disease, with the TyG index having the best predictive value. The results of the ROC curve of the model showed an AUC of 0.960 (0.915, 1.000), a sensitivity of 0.878, a specificity of 0.929, and a correctness index of 0.807. The clinical prediction model showed that the model demonstrated a higher net clinical benefit across the threshold probability intervals from 5% to 100%, and a better goodness-of-fit was shown in the calibration plot. Conclusions: Peripheral microcirculatory changes are more pronounced in patients with diabetes mellitus complicated with coronary artery disease, and the number of cross-tubular collaterals, TyG index, gender, and hypertension are all independent predictors of diabetes mellitus complicated with coronary artery disease, and the combination of all four of them has a better predictive value for patients with diabetes mellitus complicated with coronary artery disease.