Logistic模型预测冠状动脉狭窄程度  

Prediction of coronary artery stenosis with logistics model

作　　者：张琪[1] 刘凌翀[2] 李桂杰[2] 宋歌声 史浩[2] 王滨[1] 

机构地区：[1]滨州医学院临床医学院医学影像学系,烟台264003 [2]山东省千佛山医院影像科

出　　处：《中国医院统计》2017年第2期105-107,110,共4页Chinese Journal of Hospital Statistics

摘　　要：目的临床工作中需要一种非侵入性、非创伤性和辐射性的诊断意见对冠状动脉硬化心脏病、冠脉狭窄进行诊断。本研究探讨冠状动脉狭窄与血压、血脂和血糖的关系,并结合年龄和性别建立冠脉狭窄logistics预测模型,以探索无创性诊断冠脉狭窄的可行性。方法本文收集了接受冠状动脉血管造影患者894例,冠脉DSA检查结果分为狭窄组和无狭窄组。两组均测定安静时血压、空腹血糖、甘油三酯、血清总胆固醇、高密度脂蛋白胆固醇和低密度脂蛋白胆固醇等生化指标。利用t检验分析两组之间各指标的差异是否具有统计学意义,并将其拟合logistic模型建立预测模型。结果两组年龄、性别、空腹血糖、血清总胆固醇和高密度脂蛋白胆固醇差异具有统计学意义(P<0.05),狭窄组中高血压患者及男性患者比例高于无狭窄组,无狭窄组中女性患者比例高于男性患者。两组间甘油三酯和低密度脂蛋白胆固醇差异无统计学意义(P>0.05)。将年龄、性别、血压、空腹血糖、血清总胆固醇和高密度脂蛋白胆固醇拟合logistics模型后,AUC为0.751。高密度脂蛋白对冠心病为保护性指标;老年男性、高血压、总胆固醇作为冠心病的危险因素,是老年人防治冠心痛的重点内容;降低胆固醇和血糖、对防治老年人,特别是男性人群的冠心病有积极的意义。结论本文建立的logistics模型在预测冠脉狭窄情况上有一定作用,可以在一定程度上实现冠脉狭窄的无创性诊断。Objective The clinics needs a non-invasive, non traumatic diagnosis of coronary heart disease and eorocnary artery stenosis. To evaluate the relationship between coronary artery stenosis and blood pressure, blood fat and blood sugar, and combine them with age and sex to establish logistics forecast model. To explore the feasibility of noninvasive diagnosis of coronary artery stenosis. Methods We collected a total of 894 patients who accepted coronary artery angiography examination in our hospital, with an average age of 58, including 545 male and 347 female cases. DSA results were used to classify stenosis group （with more than one branch of coronary artery stenosis, 744 cases） and no stenosis group （without coronary artery stenosis, 150 cases ）. Blood pressure, fasting blood glucose, triglyeeride, total cholesterol, high density lipoprotein cholesterol and low density lipoprotein cholesterol were measured in both groups. T test was used to analyze the differences of each index between both groups, and the logistics model was fitted to built a prediction model with these indexes. Results Age, gender, fasting blood glucose, serum total cholesterol and high density fat cholesterol differences had statistical significance （P〈0.05） between two groups. The proportion of male patients and patients with hypertension in stenosis group was higher than that in no stenosis group ; the proportion of women in no stenosis group was higher than that of men. There was no significant difference in triglyceride and low density lipoprotein cholesterol between both groups （P〈0.05）. The age, sex, blood pressure, fasting blood glucose, total cholesterol and high density lipoprotein cholesterol were fitted to the logistics model, and the AUC was 0. 751. High density lipo- protein in coronary heart disease （ CHD ） was a kind of protection material. Elderly male, hypertension, total cholesterol were risk factors for coronary heart disease. Conclusion The logistics model has certain significance in the prediction o

关 键 词：DSA 冠状动脉狭窄 高血压 血脂 血糖 logistics预测模型 

分 类 号：R181.2[医药卫生—流行病学]