慢性阻塞性肺疾病急性加重期患者出院后30d内再入院的影响因素分析及Nomogram模型构建  

Influencing Factors and Nomogram Model for Readmission in AECOPD Patients within 30 Days after Discharge

作　　者：周寅川 荣蓉[1] 黄祎丹 李业彪[2] ZHOU Yinchuan;RONG Rong;HUANG Yidan;LI Yebiao(Department of Respiratory Medicine,Houjie Hospital Affiliated to Guangdong Medical University,Dongguan 523900,China;Department of Respiratory Medicine,Guangdong Provincial Agricultural Reclamation Central Hospital,Zhanjiang 524002,China)

机构地区：[1]广东医学院附属厚街医院呼吸内科,广东省东莞市523900 [2]广东省农垦中心医院呼吸内科,广东省湛江市524002

出　　处：《实用心脑肺血管病杂志》2020年第8期57-63,共7页Practical Journal of Cardiac Cerebral Pneumal and Vascular Disease

基　　金：湛江市科技计划项目(170507201681089)。

摘　　要：背景慢性阻塞性肺疾病(COPD)是一种以气道不完全可逆性气流受限为特征的肺部慢性疾病,与肺部、体循环免疫反应有关。慢性阻塞性肺疾病急性加重期(AECOPD)患者出院30 d内有继发性加重风险,再入院治疗可能性大,是COPD患者住院、死亡的主要原因。目的探讨AECOPD患者出院30 d内再入院的影响因素并构建Nomogram模型,以为临床诊疗提供参考。方法选取广东医学院附属厚街医院2017年2月—2019年2月收治的286例AECOPD患者作为试验组,根据患者出院30 d内是否再入院治疗分为再入院亚组(n=128)及未再入院亚组(n=286);采用相同的纳入与排除标准选取广东省农垦中心医院2017年1月—2019年1月收治的128例AECOPD患者作为外部验证组。收集患者入院24 h内的临床资料及出院30内再入院情况。绘制受试者工作特征曲线(ROC曲线)以评估相关变量对AECOPD患者出院30 d内再入院的最佳截断值;采用多因素Logistic回归分析AECOPD患者出院30 d内再入院的危险因素,采用R版3.5.2中的"rms"软件包构建预测AECOPD患者出院30 d内再入院的Nomogram模型并进行外部验证,计算一致性指数(CI)。结果再入院亚组患者年龄大于非再入院亚组,有吸烟史者占比、COPD评估测试(CAT)评分≥30分者占比、冠心病发生率、糖尿病发生率、红细胞分布宽度(RDW)、白细胞计数(WBC)、红细胞沉降率(ESR)、C反应蛋白(CRP)、中性粒细胞与淋巴细胞比值(NLR)、血小板与淋巴细胞比值(PLR)高于非再入院亚组,血红蛋白(Hb)、淋巴细胞与单核细胞比值(LMR)、白蛋白与球蛋白比值(AGR)低于非再入院亚组(P<0.05)。ROC曲线结果显示,年龄、Hb、RDW、WBC、ESR、CRP、NLR、PLR、LMR、AGR预测AECOPD患者出院30 d内再入院的最佳截断值分别为69岁、12 g/L、19.7%、9×10~9/L、19.9 mm/1 h、2.2 mg/dl、3.5、146.9、2.7、1.4。多因素Logistic回归分析结果显示,CAT评分≥30分、糖尿病、CRP≥2.2Background Chronic obstructive pulmonary disease(COPD)is a chronic lung disease characterized by incomplete reversible airway airflow limitation,which is associated with pulmonary and systemic circulatory immune responses.Acute exacerbation of COPD(AECOPD),a major cause of hospitalization and death in COPD patients,is associated with a risk of secondary exacerbation within 30 days of discharge and subsequent readmission.Objective To investigate the influencing factors of post-discharge 30-day readmission in AECOPD patients to establish a Nomogram model for predicting readmission risk,providing a reference for clinical diagnosis and treatment of this disease.Methods Eligible AECOPD patients included 286 who were recruited from Houjie Hospital Affiliated to Guangdong Medical University from February 2017 to February 2019(experimental group),and 128 from Guangdong Provincial Agricultural Reclamation Central Hospital from January 2017 to January 2019(validation group).Clinical data of within 24 hours of admission and post-discharge 30-day readmission were collected.By post-discharge 30-day readmission prevalence,experimental group patients were divided into readmission subgroup(n=128)and non-readmission subgroup(n=286).ROC curve analysis was performed to evaluate the optimal cutoff value of related variables for predicting post-discharge 30-day readmission.The Nomogram model was constructed using readmission risk factors identified by multivariate Logistic regression with rms package of R(version 3.5.2),and was used to predict the post-discharge 30-day readmission in validation group for verification,and consistency index(CI)was calculated.Results The readmission subgroup had greater average age,percentages of having a smoking history,and CAT score≥30,prevalence of coronary heart disease and diabetes,average RDW,WBC,ESR,CRP,NLR,and PLR,but lower average Hb,LMR and AGR than non-readmission subgroup(P<0.05).ROC curve analyses revealed that the optimal cutoff values for age,Hb,RDW,WBC,ESR,CRP,NLR,PLR,LMR and AGR in pr

关 键 词：慢性阻塞性肺疾病 加重期 炎症 预后 Nomogram模型 

分 类 号：R563.9[医药卫生—呼吸系统]