摘要
目的分析老年慢性阻塞性肺疾病急性加重期(AECOPD)合并脑梗患者死亡相关危险因素,建立个体化预测该类患者死亡风险的列线图模型。方法采用回顾性研究方法,分析2013年1月至2018年6月在我院住院治疗的老年AECOPD合并脑梗患者(722例)的临床资料,将最终研究对象随机分为建模组(70%)与验证组(30%),建模组数据用于建立二元多因素logistic模型,用R软件绘制列线图,分析患者死亡危险因素。验证组数据用于验证模型。采用ROC曲线和校准图验证列线图模型的预测效能及准确度。结果性别(OR=2.25,95%CI:0.19~1.55)、年龄(OR=1.06,95%CI:1.01~1.12)、 COPD病程(OR=1.08,95%CI:1.04~1.13)、 GOLD分级(OR=2.91,95%CI:1.95~4.45)、 NIHSS评分分级(OR=2.24,95%CI:1.25~4.16)、肺心病(OR=2.84,95%CI:1.35~5.98)、呼衰(OR=5.26,95%CI:2.24~12.63)、自发性气胸(OR=5.23,95%CI:2.09~13.16)与肺部感染(OR=2.81,95%CI:1.52~5.38)是患者死亡的独立危险因素(P<0.05)。建模组模型ROC曲线的AUC为0.907,验证组ROC曲线的AUC为0.847,两组ROC曲线比较差异无统计学意义(D=1.447,P=0.149)。校准图显示,两组预测模型校准曲线与标准曲线均接近,差异有统计学意义(均P>0.10)。结论通过性别、年龄、COPD病程、GOLD分级、NIHSS评分分级、肺心病、呼衰、自发性气胸、肺部感染独立危险因素,建立的预测老年AECOPD合并脑梗患者死亡风险的列线图模型,具有良好的诊断效能和准确度,对甄别该类患者中的高死亡风险人群,制定干预决策有指导意义。
Objective To analyze the risk factors associated with death in patients with acute exacerbation of chronic obstructive pulmonary disease(AECOPD)complicated by cerebral infarction,and to establish a nomogram model for individualized prediction of the risk to death in the patients.Methods The clinical data of elderly patients with AECOPD complicated with cerebral infarction(722 cases)who were hospitalized in our hospital from January 2013 to June 2018 were analyzed by retrospective study.The subjects were randomly divided into modeling groups(70%)and validation group(30%),the modeling group data was used to establish a binary multivariate logistic model,and the nomogram was drawn by R software to analyze the risk factors for death.The validation group data is used to validate the model.The ROC curve and calibration plot are used to verify the predictive performance and accuracy of the nomogram model.Results Gender(OR = 2.25,95% CI:0.19 ~ 1.55),age(OR = 1.06,95% CI:1.01 ~ 1.12),COPD duration(OR = 1.08,95% CI:1.04 ~ 1.13),GOLD classification(OR = 2.91,95% CI:1.95 ~ 4.45),NIHSS score rating(OR = 2.24,95% CI:1.25 ~ 4.16),pulmonary heart disease(OR = 2.84,95% CI:1.35 ~ 5.98),respiratory failure(OR = 5.26,95%CI:2.24 ~ 12.63),spontaneous pneumothorax (OR = 5.23,95% CI:2.09 ~ 13.16)and pulmonary infection(OR = 2.81,95% CI:1.52 ~ 5.38)were independent risks factors of death(P < 0.05).The AUC of the ROC curve of the model group model was 0.907,and the which of the validation group was 0.847.There was no significant difference between the two groups in the ROC curve(D = 1.447,P = 0.149).The calibration test showed that the calibration plots of the two sets of predictive models were close to the standard curve and were statistically significant(P = 0.841,and 0.214,respectively > 0.10).Conclusion A nomogram model for predicting the risk to death in elderly patients with AECOPD and cerebral infarction is established underlying the factors including gender,age,COPD duration,GOLD grade,NIHSS score,pulmonary heart disease,respirato
作者
王文娟
黄志东
李贵宇
罗艳洁
WANG Wenjuan;HUANG Zhidong;LI Guiyu;LUO Yanjie(Department of Geriatrics,Nanxishan Hospital of Guangxi Zhuang Autonomous Region(Guangxi Zhuang Autono? mous Region Second People′s Hospital),Guilin 541002,China)
出处
《实用医学杂志》
CAS
北大核心
2019年第13期2128-2132,共5页
The Journal of Practical Medicine
关键词
老年
慢性肺阻塞性肺疾病
并发症
死亡风险
列线图
elderly
chronic obstructive pulmonary disease
complication
death risk
nomogram
