目的 急性呼吸衰竭是慢性阻塞性肺疾病（chronic obstructive pulmonary disease,COPD）患者预后恶化的重要事件,本研究拟基于套索逻辑（least absolute shrinkage and selection operator-Logistic,LASSO-Logistic）回归模型分析COPD患者诱发呼吸衰竭的影响因素。方法 本研究为单中心回顾性队列研究,连续收集2019年1月1日—2021年12月31日期间在本院住院的COPD患者360例。依据首次入院时间,将2019年1月—2021年6月入院者分为建模队列（n=280）,2021年7月—2021年12月入院者分为验证队列（n=80）。采集30项基线指标,包括人口学资料、实验室指标、动脉血气分析及肺功能参数。建模流程为：先进行单因素分析筛选候选变量,再采用LASSO-Logistic回归（10折交叉验证确定最优λ）进行进一步变量筛选,最终将入选变量纳入多因素Logistic回归构建预测模型。通过受试者工作特征（receiver operating characteristic,ROC）曲线、一致性指数（concordance index,C-index）和校准曲线评价模型区分度与拟合度,并采用Bootstrap（重复抽样1 000次）进行验证。结果 建模队列共纳入280例COPD患者,其中89例在1年内发生急性呼吸衰竭。验证队列包含80例患者。两队列在年龄、性别、BMI、GOLD分级以及主要实验室与肺功能指标方面差异均无统计学意义（P>0.05）,具有良好可比性。LASSO回归在30项基线指标中初筛得到10个候选变量,多因素Logistic回归最终确定年龄、GOLD分级、FVC、cTnI和NT-proBNP为独立预测因子。模型伪R2=0.28,似然比检验P<0.01。建模队列曲线下面积（area under the curve,AUC）为0.861,验证队列AUC=0.847（95%CI：0.809~0.905）,敏感性0.704,特异性0.782。模型C-index与AUC一致,校准曲线显示预测概率与实际结局具有良好一致性;Bootstrap（1 000次）验证提示模型稳定、泛化性能良好。结论 本研究构建的LASSO-Logistic回归预测模型,可有效识别COPD患者发生急性呼吸衰竭的风险。该模型简便、可重复性强,具有良好的预测性能和临床应用价值。
Abstract
Objective Acute respiratory failure (ARF) is a major adverse event affecting the prognosis of patients with chronic obstructive pulmonary disease (COPD). This study aimed to identify risk factors for ARF in COPD patients using a LASSO-Logistic regression model. Methods This single-center retrospective cohort study included 360 hospitalized COPD patients from January 1, 2019 to December 31, 2021. Based on admission time, patients were divided into a derivation cohort (n=280) and a validation cohort (n=80). Thirty baseline variables—covering demographic features, laboratory tests, arterial blood gas parameters, and pulmonary function indices—were collected. Candidate predictors were first screened by univariate analysis, followed by LASSO-Logistic regression with 10-fold cross-validation, and subsequently incorporated into multivariable logistic regression to construct the final prediction model. Model discrimination and calibration were evaluated using ROC curves, the C-index, and calibration plots, with internal validation performed using 1, 000-time bootstrap resampling. Results Among the 280 patients in the derivation cohort, 89 developed ARF within one year. Baseline characteristics were comparable between the derivation and validation cohorts (all P>0.05). LASSO regression initially identified 10 potential predictors, and multivariable Logistic regression ultimately determined age, GOLD stage, FVC, cTnI, and NT-proBNP as independent predictors. The model yielded a pseudo R2 of 0.28(likelihood ratio test P<0.01). The AUC was 0.861 in the derivation cohort and 0.847(95% CI: 0.809-0.905) in the validation cohort, with a sensitivity of 0.704 and specificity of 0.782. The C-index was consistent with the AUC, and calibration curves demonstrated good agreement between predicted and observed outcomes. Bootstrap validation confirmed the stability and robustness of the model. Conclusion The LASSO-Logistic regression model developed in this study can effectively predict the risk of ARF in COPD patients. The model is simple, reproducible, and demonstrates good predictive performance, indicating promising clinical utility for early risk stratification.
关键词
慢性阻塞性肺病 /
急性呼吸衰竭 /
预测模型 /
套索回归 /
逻辑回归
Key words
COPD /
acute respiratory failure /
predictive model /
LASSO regression /
Logistic regression
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] SANTUS P, RADOVANOVIC D, SAAD M, et al.Acute dyspnea in the emergency department: a clinical review[J]. Intern Emerg Med, 2023, 18(5): 1491-1507.
[2] SHAH N M, KALTSAKAS G.Respiratory complications of obesity: from early changes to respiratory failure[J]. Breathe, 2023, 19(1): 220263.
[3] BARTZIOKAS K, PAPAPORFYRIOU A, HILLAS G, et al.Global initiative for chronic obstructive lung disease (GOLD) recommendations: strengths and concerns for future needs[J]. Postgrad Med, 2023, 135(4): 327-333.
[4] ZHAO X, LIU G, LIU D, et al.Clinical and economic burden of anxiety/depression among older adult COPD patients: evidence from the COPD-AD China Registry study[J]. Front Psychiatry, 2024, 14: 1221767.
[5] 刘敏, 高笑宇, 杨海丽, 等. AECOPD合并Ⅱ型呼吸衰竭患者发生营养风险的影响因素分析及其预测模型构建[J]. 现代生物医学进展, 2024, 24(10): 1883-1887.
[6] 付玉芬, 牟婷, 何翔, 等. 慢性阻塞性肺疾病频繁急性加重的危险因素分析和预测模型构建[J]. 中国医学科学院学报, 2024, 46(4): 519-527.
[7] 朱雪琴, 滕镕, 丁薇, 等. 冠心病伴心力衰竭老年患者合并肺部感染风险预测模型的构建[J]. 中国临床保健杂志, 2023, 26(2): 247-250.
[8] 童朝晖. 中国老年慢性阻塞性肺疾病临床诊治实践指南[J]. 中华结核和呼吸杂志, 2020, 43(2): 100-119.
[9] FUJISHIMA S.Guideline-based management of acute respiratory failure and acute respiratory distress syndrome[J]. J Intensive Care, 2023, 11(1): 10.
[10] 梁振宇, 王凤燕, 陈子正, 等.2023年GOLD 慢性阻塞性肺疾病诊断、管理及预防全球策略更新要点解读[J]. 中国全科医学, 2023, 26(11): 1287-1295.
[11] PONIKOWSKI P, VOORS AA, ANKER SD, et al.2016 ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure[J]. Rev Esp Cardiol, 2016, 69(12): 1167.
[12] PAVORD ID, JONES PW, BURGEL PR, et al. Exacerbations of COPD[J]. Int J Chron Obstruct Pulmon Dis, 2016, 11 Spec Iss (Spec Iss): 21-30.
[13] GBD 2019 Respiratory Disease Collaborators. Global burden of respiratory diseases and risk factors, 1990-2019: an analysis for the Global Burden of Disease Study 2019BD 2019 Respiratory Disease Collaborators. Global burden of respiratory diseases and risk factors, 1990-2019: an analysis for the Global Burden of Disease Study 2019[J]. Lancet Respir Med, 2022, 10(6): 585-619.
[14] SINGH D, AGUSTI A, ANZUETO A, et al.Global Strategy for the Diagnosis, Management, and Prevention of Chronic Obstructive Lung Disease: the GOLD science committee report 2023[J]. Eur Respir J, 2023, 61(1): 2201938.
[15] BELLANI G, LAFFEY J G, PHAM T, et al.Epidemiology, patterns of care, and mortality for patients with acute respiratory distress syndrome in intensive care units in 50 countries[J]. JAMA, 2016, 315(8): 788-800.
[16] WU Z, MCGOOGAN J M.Characteristics of and important lessons from the coronavirus disease 2019(COVID-19) outbreak in China: summary of a report of 72 314 cases from the Chinese center for disease control and prevention[J]. JAMA, 2020, 323(13): 1239-1242.
[17] TOPHAM M, HARPER S, STEVENSON M, et al.Predicting mortality in COPD patients using routinely collected data[J]. Thorax, 2022, 77(8): 757-765.
基金
中国中医科学院望京医院院级科研课题“清金化浊方善慢阻肺急性加重患者血液高凝状态的研究”（WJYY2020-20）
PDF(2808 KB)
