Ye Lan, Yuan Xinyu, Li Yuntao
Department of Respiratory and Critical Care Medicine, The Fourth Affiliated Hospital of Soochow University (Suzhou Dushu Lake Hospital) Suzhou 215000, Jiangsu, China.
Am J Transl Res. 2025 Jul 15;17(7):5485-5492. doi: 10.62347/KAAK3231. eCollection 2025.
To identify factors associated with failed weaning from mechanical ventilation in elderly patients with chronic obstructive pulmonary disease (COPD) and type II respiratory failure.
This retrospective study included 210 patients treated at the Fourth Affiliated Hospital of Soochow University from April 2021 to April 2024. Patients were divided into a modeling group (n = 147) and a validation group (n = 63) in a 7:3 ratio. Univariate and multivariate logistic regression analyses were performed to determine risk factors for weaning failure. A risk prediction model was developed based on the multivariate results using the glm function and visualized as a nomogram with the rms package. The model's predictive performance was evaluated using receiver operating characteristic (ROC) curves.
Multivariate analysis identified elevated N-terminal pro-brain natriuretic peptide (NT-proBNP), low 25-hydroxyvitamin D [25(OH)D], high rapid shallow breathing index, longer COPD disease duration, and higher Acute Physiology and Chronic Health Evaluation II (APACHE II) scores as independent risk factors (all P < 0.05). The area under the ROC curve (AUC) for predicting weaning failure was 0.802 in the modeling group and 0.824 in the validation group, indicating good predictive accuracy.
NT-proBNP, 25(OH)D, rapid shallow breathing index, COPD duration, and APACHE II score are key predictors of mechanical ventilation weaning failure in elderly COPD patients with type II respiratory failure. The developed model demonstrates robust predictive value and may aid clinical decision-making.
确定慢性阻塞性肺疾病（COPD）合并Ⅱ型呼吸衰竭老年患者机械通气撤机失败的相关因素。
本回顾性研究纳入了2021年4月至2024年4月在苏州大学附属第四医院接受治疗的210例患者。患者按7:3的比例分为建模组（n = 147）和验证组（n = 63）。进行单因素和多因素逻辑回归分析以确定撤机失败的危险因素。基于多因素结果使用glm函数建立风险预测模型，并使用rms包将其可视化为列线图。使用受试者工作特征（ROC）曲线评估模型的预测性能。
多因素分析确定N末端脑钠肽前体（NT-proBNP）升高、25-羟基维生素D[25(OH)D]水平低、快速浅呼吸指数高、COPD病程长以及急性生理与慢性健康状况评分系统II（APACHE II）评分高为独立危险因素（均P < 0.05）。建模组预测撤机失败的ROC曲线下面积（AUC）为0.802，验证组为0.824，表明预测准确性良好。
NT-proBNP、25(OH)D、快速浅呼吸指数、COPD病程和APACHE II评分是COPD合并Ⅱ型呼吸衰竭老年患者机械通气撤机失败的关键预测因素。所建立的模型具有强大的预测价值，可能有助于临床决策。
