摘要
目的探究慢阻肺急性加重患者继发肺部真菌感染的危险因素,应用列线图构建并验证用于辅助临床预测慢阻肺急性加重患者发生肺部真菌感染概率的可视化工具。方法采用回顾性队列研究的方法,收集2021年1月—2021年12月期间,于成都医学院第一附属医院呼吸与危重医学科住院治疗的慢阻肺急性加重患者作为训练集。并收集2020年1月—2020年12月期间的慢阻肺急性加重患者作为验证集。通过单因素、Lasso回归分析及多因素Logistic分析到独立危险因素。用独立危险因素构建列线图预测模型,并通过受试者工作特征(Receiver operating characteristic,ROC)曲线下面积(The area under ROC curve,AUC)、校准曲线和决策曲线分析(decision curve analysis,DCA)对列线图进行多方面评估。结果使用糖皮质激素、抗生素联合使用、抗生素使用时间及低蛋白血症是慢阻肺急性加重患者继发肺部真菌感染的独立危险因素(均P<0.05)。所构建预测模型的训练集和验证集的AUC值分别为0.915[95%CI:0.891~0.940]和0.830[95%CI:0.790~0.871]。训练集和验证集中的校准曲线显示,列线图预测的真菌感染概率与实际情况相符。而DCA曲线结果表明该列线图预测模型具有较好的临床实用性。结论糖皮质激素、多种抗生素联合使用、抗生素使用时间过长及低蛋白血症是慢阻肺急性加重患者继发肺部真菌感染的独立危险因素。本研究构建的慢阻肺急性加重患者继发肺部真菌感染的临床预测模型,具有较强的预测价值及实用性。
Objective To investigate the risk factors for secondary pulmonary fungal infection in patients with acute exacerbation of chronic obstructive pulmonary disease(AECOPD).And a visual tool using nomogram was developed and validated to assist in the clinical prediction of the probability of pulmonary fungal infection occurrence in AECOPD patients.Methods A retrospective cohort study method was used to collect AECOPD patients hospitalized in the Department of Respiratory,The First Affiliated Hospital of Chengdu Medical College from January 2021 to December 2021 as a training set.And AECOPD patients between January 2020 and December 2020 were collected as a validation set.Independent risk factors were determined through univariate,Lasso regression analyses.and multivariable logistic,A nomogram prediction model was constructed with these independent risk factors,and the nomogram was evaluated by receiver operating characteristic area under the curve(AUC),calibration curve,and decision curve analysis(DCA).Results The use of glucocorticoid,combined use of antibiotics,duration of antibiotic use and hypoalbuminemia were independent risk factors for secondary pulmonary fungal infection in AECOPD patients(all P<0.05).The training set and validation set of the constructed prediction model had an AUC value of 0.915[95%CI:0.891-0.940]and 0.830[95%CI:0.790-0.871],respectively.The calibration curve showed that the predicted probability was in good agreement with the actual observed probability of pulmonary fungal infection in AECOPD patients.The corresponding decision curve analysis(DCA)indicated the nomogram had relatively ideal clinical utility.Conclusions The result showed that the use of glucocorticoid,combined use of antibiotics,prolonged antibiotic therapy and hypoalbuminemia was independent risk factors for pulmonary fungal infection in AECOPD patients.The clinical prediction model for secondary pulmonary fungal infection in AECOPD patients constructed in this study has strong predictive power and clinical practicability.
作者
林希
黄娜
LIN Xi;HUANG Na(Department of Respiratory Medicine,The First Affiliated Hospital of Chengdu Medical College,Chengdu,Sichuan 610500,P.R.China)
出处
《中国呼吸与危重监护杂志》
CAS
CSCD
2024年第2期77-85,共9页
Chinese Journal of Respiratory and Critical Care Medicine
基金
国家临床重点专科培养项目(CYFY2018GLPHX02)
关键词
慢性阻塞性肺疾病急性加重期
肺部真菌感染
列线图
预测模型
Acute exacerbation of chronic obstructive pulmonary disease
pulmonary fungal infection
nomogram
prediction model
作者简介
通信作者:黄娜,Email:717308813@qq.com
