心内科住院患者高血压急症的危险因素分析及预测模型的建立  

Nomogram model for predicting hypertensive emergency of the inpatients in department of cardiology based on the basis of risk factor analysis

作　　者：冯六六[1] 白艳艳[1] 史骏[1] 刘天华[1] 刘新兵[1] 单蓉 Feng Liu-liu;Bai Yan-yan;Shi Jun;Liu Tian-hua;Liu Xin-bing;Shan Rong(Department of Cardiology,Shidong Hospital of Shanghai Yangpu District,Shanghai 200438,China)

机构地区：[1]上海市杨浦区市东医院心内科,上海200438

出　　处：《中国急救医学》2020年第11期1072-1076,共5页Chinese Journal of Critical Care Medicine

基　　金：上海市卫生和计划生育委员会科研课题计划(201540298);上海市医学重点专科建设计划(ZK2015B23)。

摘　　要：目的对心内科住院高血压急症患者进行危险因素分析的基础上建立预测高血压急症发生的数学模型,为其临床防治提供合理的建议.方法采用前瞻性队列研究方案,以2017年3月至2019年3月在本院住院治疗的480例高血压患者为研究对象,根据有无发生高血压急症分为急症组(n=85)和非急症组(n=395),详细统计患者临床资料和诊断情况,采用多因素Logistic回归法寻找导致高血压急症的危险因素,并采用R软件建立预测高血压急症发生的数学模型.结果单因素分析结果显示,与非急症组比较,急症组糖尿病合并率和饮酒率高,空腹血糖、低密度脂蛋白胆固醇(LDL-C)、血清AngⅡ、SAS和SDS评分高(P<0.05);甘油三脂(TG)和血清脂联素水平低,Morisky、HPLP-ⅡR、生活方式自评分及家庭关系评分低(P<0.05).Logistic回归分析结果显示,糖尿病、饮酒、血清AngⅡ水平、SAS评分、SDS评分为发生高血压急症的独立危险因素(β>0,OR>1,P<0.05),血清脂联素、Morisky、HPLP-ⅡR、生活方式自评分为可预防高血压急症的保护性因素(β<0,OR<1,P<0.05).Nomogram预测模型证实随着血清AngⅡ水平、SAS和SDS评分的升高,血清脂联素、Morisky、HPLP-ⅡR、生活方式自评分的降低,高血压急症的发生风险依次升高,其中HPLP-ⅡR和生活方式自评分的作用相对突出,每减少10个单位,高血压急症风险值分别增加9.2和8.0分.结论本研究建立的高血压急症预测模型能较准确地预测高血压急症的发生,对其早期防治具有一定的指导意义.Objective To establish a mathematical prediction model for hypertension emergency of hospitalized patients in the department of cardiology on the basis of risk factor analysis,so as to provide reasonable suggestions for its clinical prevention and treatment.Methods Totally 480 cases hospitalized patients from March 2017 to March 2019 with high blood pressure were selected in a prospective cohort study programs of our hospital.According to the presence of hypertension emergency,the patients were divided into 2 groups:85 cases in emergency group and 395 cases in the control group.The detailed clinical data and clinical diagnosis results of patients were recorded,and risk factors were analyzed by multivariate Logistic regression,then a Nomogram model was established by R software to predict the occurrence of hypertension emergencies.Results Single factor analysis showed the diabetic rate and drinking rate were higher in the emergency group,the fasting blood glucose,LDL-C,serum Ang Ⅱ,SAS and SDS scores were also higher.The TG and serum adiponectin level were lower in the emergency group,the Morisky,HPLP-Ⅱ R and the way of life scores and family relationship scores were also lower.The difference was statistically significant between two groups(P <0.05).Logistic regression analysis results showed diabetes,alcohol consumption,serum Ang Ⅱ level,SAS scores and SDS scores were independent risk factors of hypertension emergency(β> 0,OR> 1,P <0.05),serum adiponectin,Morisky,HPLP-Ⅱ R,the way of life score were protective factors for hypertension emergency(β <0,OR <1,P <0.05).Nomogram model confirmed the risk of hypertension emergencies was higher as serum Ang Ⅱ levels,SAS score,SDS score increased,and serum adiponectin,Morisky,HPLP-Ⅱ R,the way of life score decreased.HPLP-Ⅱ R and the way of life were the most important factors,and the risk value of hypertension emergencies increased by 9.2 scores and 8.0 scores when 10 units decreased.Conclusion The Nomogram model for hypertension emergencies established in th

关 键 词：心内科 高血压急症 危险因素 Nomogram模型 

分 类 号：R544.1[医药卫生—心血管疾病]