三种新型胰岛素抵抗指标水平与中年人群发生冠心病的相关性

Correlation between levels of three novel insulin resistance indicators and the occurrence of coronary heart disease in middle-aged population

摘要:

目的]探讨胰岛素抵抗代谢评分(METS-IR)、估计葡萄糖处置率(eGDR)、甘油三酯葡萄糖指数结合体重指数(TyG-BMI)等三种新型胰岛素抵抗(IR)指标水平与中年人群发生冠心病的相关性。 [方法]选取2015年1月—2022年12月在柳州市人民医院心血管内科住院行冠状动脉造影的533例中年人群为研究对象。根据冠状动脉造影结果将研究对象分为冠心病组(241例)和对照组(292例)。收集两组一般资料和实验室检查指标,根据相关公式计算METS-IR、eGDR、TyG-BMI。采用双变量Spearman相关、多因素Logistic回归、限制性立方样条图分析不同IR指标水平与中年人群发生冠心病的相关性。采用亚组分析和交互作用检验主要观察结果在不同人群中的差异。采用受试者工作特征(ROC)曲线评估不同IR指标对中年人群发生冠心病的预测效能。 [结果]冠心病组METS-IR、TyG-BMI高于对照组,eGDR低于对照组(P＜0.05)。Spearman相关分析结果显示,METS-IR、TyG-BMI与冠心病呈正相关关系(r＝0.312、0.267,均P＜0.001),eGDR与冠心病呈负相关关系(r＝－0.275,P＜0.001)。多因素Logistic回归分析结果显示,无论作为连续变量还是分类变量,METS-IR、eGDR、TyG-BMI水平均是中年人群发生冠心病的独立影响因素(P均＜0.001)。与最低四分位数组相比,METS-IR位于第2、3、4四分位数组的人群发生冠心病的风险分别增加2.286倍(OR:3.286,95%CI:1.823～5.924)、2.061倍(OR:3.061,95%CI:1.670～5.609)、4.646倍(OR:5.646,95%CI:3.018～10.563)；eGDR位于第4四分位数组的人群发生冠心病的风险降低79.2%(OR:0.208,95%CI:0.114～0.379)；TyG-BMI位于第2、3、4四分位数组的人群发生冠心病的风险分别增加1.579倍(OR:2.579,95%CI:1.438～4.625)、2.155倍(OR:3.155,95%CI:1.761～5.655)、3.142倍(OR:4.142,95%CI:2.317～7.407)。限制性立方样条图显示,在调整混杂因素影响后,METS-IR、TyG-BMI水平与冠心病呈非线性正相关关系(均P总体＜0.001,P非线性＝0.002、0.043),eGDR水平与冠心病呈非线性负相关关系(P总体＜0.001,P非线性＝0.010)。亚组分析显示,不同IR指标与冠心病的关联在超重/肥胖、糖尿病、高血压以外的人群中基本一致。ROC曲线分析显示,METS-IR、eGDR、TyG-BMI水平联合临床因素预测中年人群发生冠心病的曲线下面积分别为0.746(95%CI:0.704～0.788,P＜0.001)、0.750(95%CI:0.709～0.792,P＜0.001)和0.742(95%CI:0.700～0.784,P＜0.001)。 [结论]METS-IR、eGDR和TyG-BMI水平与中年人群发生冠心病独立相关,三者对冠心病均具有一定的预测价值,均可作为预测冠心病较为理想的指标。

Abstract:

Aim To investigate the correlation between levels of three novel insulin resistance (IR) indicators including metabolic score for insulin resistance (METS-IR), estimated glucose disposal rate (eGDR), triglyceride-glucose index combined with body mass index (TyG-BMI), and the occurrence of coronary heart disease(CHD) in middle-aged population. Methods A total of 533 middle-aged individuals who were hospitalized in the Department of Cardiology of Liuzhou People's Hospital from January 2015 to December 2022 and underwent coronary angiography were enrolled as the research subjects. They were divided into CHD group (241 cases) and control group (292 cases) according to the results of coronary angiography. The general data and laboratory examination indicators of the subjects were collected, and METS-IR, eGDR, TyG-BMI were calculated using relevant formulas. Bivariate Spearman correlation, multivariate Logistic regression, and restricted cubic spline plots were used to analyze the correlation between levels of various IR indicators and the occurrence of CHD in middle-aged population. Subgroup analysis and interaction tests were conducted to examine differences in the main results among different people. Receiver operating characteristic (ROC) curve was used to evaluate the value of various IR indicators in predicting the occurrence of CHD in middle-aged population. ResultsMETS-IR, TyG-BMI were higher in CHD group than those in control group, while eGDR was lower than that in control group (P＜0.05). The Spearman correlation analysis results showed positive correlations between METS-IR, TyG-BMI and CHD (r＝0.312 and 0.267, respectively; all P＜0.001), and negative correlation between eGDR and CHD in middle-aged population (r＝－0.275, P＜0.001). Multivariate Logistic regression analysis showed that whether as continuous or categorical variable, METS-IR, eGDR, TyG-BMI were independent influencing factors for CHD in middle-aged population (P＜0.001). Compared with the lowest quartile group, the risk of CHD in 2nd, 3rd, and 4th quartile groups of METS-IR increased by 2.286-fold (OR:3.6,5%CI:1.823~5.924), 2.061-fold (OR:3.1,5%CI:1.670~5.609), and 4.646-fold (OR:5.6,5%CI:3.018~10.563) respectively; the risk of CHD in 4th quartile group of eGDR decreased 79.2% (OR:0.8,5%CI:0.114~0.379); the risk of CHD in 2nd, 3rd, and 4th quartile groups of TyG-BMI increased by 1.579-fold (OR:2.9,5%CI:1.438~4.625), 2.155-fold (OR:3.5,5%CI:1.761~5.655), and 3.142-fold (OR:4.2,5%CI:2.317~7.407) respectively. The restricted cubic spline plots revealed that, after adjusting for confounding factors, METS-IR, TyG-BMI levels were positively nonlinearly correlated with CHD (P for overall＜0.001, P for nonlinearity＝0.002 and 0.043, respectively), while eGDR level was negatively nonlinearly correlated with CHD (P for overall＜0.001, P for nonlinearity＝0.010). Subgroup analysis showed that the association between various IR indicators and CHD were basically consistent in population other than those with overweight/obesity, diabetes and hypertension. ROC curve analysis showed that the areas under the curve (AUC) of METS-IR, eGDR, TyG-BMI combined with clinical factors in predicting CHD were 0.746 (95%CI:0.704～0.788, P＜0.001), 0.750(95%CI:0.709～0.792, P＜0.001) and 0.742 (95%CI:0.700～0.784, P＜0.001) respectively. Conclusion The levels of METS-IR, eGDR, and TyG-BMI are independently associated with the occurrence of CHD in middle-aged population. All three indicators have certain predictive value for CHD and can serve as relatively ideal indicators for predicting CHD.