论文部分内容阅读
目的:探讨Essen卒中风险评分量表(Essen stroke risk score，ESRS)联合与中性粒细胞/淋巴细胞比值(neutrophil lymphocyte ratio，NLR)对冠心病的诊断价值。方法:收集2015年7月1日至2019年6月30日在安徽医科大学第二附属医院老年心血管内科因疑似冠心病住院并行冠状动脉造影检查的患者为研究对象，根据冠脉造影结果分为冠心病组(血管狭窄≥50%)与对照组(血管狭窄<50%)。收集两组患者的临床、实验室和冠脉造影资料，计算NLR，评价ESRS、冠脉狭窄程度积分(Gensini法)。采用logistic回归模型及受试者工作特征曲线(receiver operator characteristic curve，ROC曲线)进行分析，评价ESRS联合NLR对冠心病的诊断价值。结果:研究期间共入组患者325例，其中冠心病组219例，对照组106例。冠心病组年龄、高血压、糖尿病、外周动脉疾病、ESRS、NLR、空腹血糖水平均显著高于对照组(均n P<0.05)。Spearman相关分析显示ESRS(n r＝0.515，n P<0.001)及NLR(n r＝0.250，n P<0.001)与冠脉Gensini积分呈正相关关系。通过logistic回归分析，建立了logistic回归模型1为logit(P1)＝-2.072＋0.566×ESRS＋0.299×NLR＋0.173×空腹血糖(诊断准确率为71.7%)与logistic回归模型2为logit(P2)＝-1.169＋0.594×ESRS＋0.302×NLR(诊断准确率为70.8%)。2个模型诊断准确率比较，差异无统计学意义(n P＝0.499)，最终选择logistic回归模型2作为联合诊断模型。ESRS、NLR及logistic回归模型2的ROC曲线下面积分别为0.713、0.634及0.736，差异均有统计学意义(均n P<0.05)，ESRS的诊断临界值为2，NLR的诊断临界值为2.74，logistic回归模型2联合诊断的95%n CI(0.684，0.783)，敏感度60.27%，特异度为78.30%，该模型显著优于单一指标ESRS(n P＝0.047)及NLR(n P<0.001)的诊断效能。n 结论:ESRS联合NLR构建的诊断模型显著优于单一指标的诊断效能，可用于冠心病的预测。“,”Objective:To evaluate the diagnostic value of Essen stroke risk score(ESRS) combined with neutrophil lymphocyte ratio(NLR) in coronary heart disease(CHD).Methods:From July 1, 2015 to June 30, 2019, patients who were hospitalized in the Second Affiliated Hospital of Anhui Medical University for suspected CHD and underwent coronary angiography were selected as the study objects.According to the results of coronary angiography, these patients were divided into CHD group (stenosis rate ≥ 50%) and control group(stenosis rate<50%). The clinical, laboratory and angiographic data of the two groups were collected, including NLR, ESRS and coronary narrow degree integral(Gensini integral method). Logistic regression model and receiver operator characteristic(ROC) were used to evaluate the diagnostic value of ESRS combined with NLR in CHD.Results:During the study period, 325 patients were enrolled, including 219 CHD patients and 106 controls.The age, hypertension, diabetes mellitus, peripheral artery disease, ESRS, NLR and fasting blood glucose levels in the CHD group were significantly higher than those in the control group(all n P<0.05). Spearman correlation analysis showed that ESRS(n r=0.515, n P<0.001) and NLR(n r=0.250, n P<0.001) were positively correlated with coronary Gensini score.Two logistic regression models were established, where model 1 was logit(P1)=-2.072+ 0.566×ESRS+ 0.299×NLR+ 0.173×fasting blood glucose(the diagnostic accuracy rate was 71.7%) and model 2 was logit(P2)=-1.169+ 0.594×ESRS+ 0.302×NLR(the diagnostic accuracy rate was 70.8%). There was no significant difference in the diagnostic accuracy between the two models(n P=0.499). Finally, logistic regression model 2 was selected as the joint diagnostic model.The area under curve of ESRS, NLR and logistic regression model 2 was 0.713, 0.634 and 0.736, respectively, and the difference was statistically significant(all n P<0.05). The diagnostic threshold of ESRS was 2, and the diagnostic threshold of NLR was 2.74.The 95% CI of the joint diagnostic model 2 was (0.684, 0.783), which showed a sensitivity of 60.27% and a specificity of 78.30%.This model was superior to the diagnostic efficacy of ESRS(n P=0.047) and NLR(n P<0.001).n Conclusion:The joint diagnostic model of ESRS combined with NLR is superior to the single index in the diagnosis of CHD, which may thus be used to predict the occurrence of CHD.