AECOPD并发肺血栓栓塞症的危险因素分析

Analysis of Risk Factors for Pulmonary Thromboembolism Complicated with AECOPD

目的本研究旨在探讨慢性阻塞性肺疾病急性加重(AECOPD)并发肺血栓栓塞症的独立危险因素,并通过深入分析这些因素,构建数学模型以评估预测效能。方法选取2019年1月—2023年3月内蒙古自治区人民医院治疗的AECOPD患者的385份详细病历数据进行分析。研究对象分为无肺血栓栓塞症组和有肺血栓栓塞症组。运用单变量和多元变量分析策略来识别AECOPD并发肺血栓栓塞症的决定性因素。结果在参与研究的385例患者中,有53例被诊断为合并肺血栓栓塞症。单变量以及多元逻辑回归分析确认,巴氏指数、卧床时长、下肢静脉血栓、右心功能障碍、动脉血氧分压、纤维蛋白原和C反应蛋白为AECOPD并发肺血栓栓塞症的独立风险因素,差异均有统计学意义(P<0.05)。通过使用包含独立影响因素及P值的逻辑回归模型,构建预测AECOPD并发肺血栓栓塞症的ROC曲线,其曲线下面积分别呈现出0.707、0.660、0.575、0.654、0.719、0.712和0.669的区分能力,其中P值一项预测表现尤为出色,达0.905。结论AECOPD患者的肺血栓栓塞症与巴氏指数等因素紧密相关。所建立的预测模型在特异性和灵敏性方面展现出显著的预测能力。

Objective This study aims to explore the independent risk factors for pulmonary thromboembolism in acute exacerbation of chronic obstructive pulmonary disease(AECOPD),and to construct a mathematical model to evaluate predictive efficacy through in-depth analysis of these factors.Method 385 detailed medical records of AECOPD patients treated at Inner Mongolia Autonomous Region People's Hospital from January 2019 to March 2023 were selected for analysis.The research subjects were divided into a group without pulmonary thromboembolism and a group with pulmonary thromboembolism.Using univariate and multivariate analysis strategies to identify the determining factors of pulmonary thromboembolism in AECOPD.Result Among the 385 patients involved in the study,53 were diagnosed with concomitant pulmonary thromboembolism.Univariate and multiple logistic regression analysis confi rmed that Pap's index,length of bed rest,lower limb venous thrombosis,right heart dysfunction,arterial oxygen pressure,fi brinogen,and C-reactive protein were independent risk factors for AECOPD complicated with pulmonary thromboembolism(P<0.05).By using a logistic regression model that includes independent infl uencing factors and P-values,a ROC curve was constructed to predict the occurrence of pulmonary thromboembolism in AECOPD.The areas under the curve showed discriminative abilities of 0.707,0.660,0.575,0.654,0.719,0.712,and 0.669,respectively.The P-value prediction was particularly outstanding,reaching 0.905.Conclusion Pulmonary thromboembolism in AECOPD patients is closely related to factors such as Pap's index.The established prediction model exhibits signifi cant predictive ability in terms of specifi city and sensitivity.