|
| 支撑喉镜下显微手术患者围术期呼吸系统不良事件的动态列线图模型 |
| Dynamic nomogram model of perioperative respiratory adverse events in patients undergoing microsurgery under suspension laryngoscope |
| |
| DOI：10.12089/jca.2025.05.008 |
| 中文关键词: 支撑喉镜下显微手术 围术期呼吸系统不良事件 危险因素 预测模型 列线图 |
| 英文关键词: Suspension laryngoscopic microsurgery Perioperative respiratory adverse events Risk factors Predictive model Nomogram |
| 基金项目: |
|
| 摘要点击次数: 304 |
| 全文下载次数: 72 |
| 中文摘要: |
目的：探讨影响支撑喉镜下显微手术患者发生围术期呼吸系统不良事件（PRAE）的影响因素并构建预测模型。
方法：回顾性分析2021年7月至2022年12月行全麻支撑喉镜下显微手术的成年患者391例，按照7∶3采用随机抽样法分为训练集（n=273）和验证集（n=118）。依据训练集的数据，采用LASSO回归方法探究发生PRAE的影响因素，建立列线图模型，使用验证集数据对列线图模型进行验证。通过绘制校准曲线、受试者工作特征曲线下面积（AUROC）评估其性能表现。
结果：训练集中有103例（37.7%）发生PRAE，验证集中有42例（35.6%）发生PRAE。经LASSO回归分析筛选出5个非零变量：年龄、BMI、急慢性呼吸系统疾病史、吸烟史、术中气道压峰值，将这些变量纳入多因素Logistic回归分析后显示，年龄、BMI、急慢性呼吸系统疾病史、吸烟史、术中气道压峰值是支撑喉镜下显微手术患者发生PRAE的危险因素。根据多因素Logistic回归分析结果构建支撑喉镜下显微手术患者发生PRAE的列线图模型，训练集与验证集模型曲线下面积（AUC）分别为0.881（95%CI 0.817～0.945）和0.832（95%CI 0.779～0.885）。校准曲线显示，训练集与验证集的预测曲线与标准曲线拟合良好。
结论：基于年龄、BMI、急慢性呼吸系统疾病史、吸烟史、术中气道压峰值等预测因素构建的列线图模型可以较好预测PRAE的发生。 |
| 英文摘要: |
Objective: To investigate the factors influencing perioperative respiratory adverse events (PRAE) in patients undergoing microsurgery under suspension laryngoscope and to construct a predictive nomogram model.
Methods: A total of 391 adult patients who underwent microsurgery under general anesthesia with a self-retaining laryngoscope from July 2021 to December 2022 were retrospectively analyzed. Patients were randomly assigned to a training set (n = 273) and a validation set (n = 118) using a 7∶3 ratio. Using data from the training set, we employed LASSO regression to identify factors associated with PRAE. A nomogram model was then developed based on these factors and validated using data from the validation set. Model performance was assessed using calibration curves and the area under the receiver operating characteristic curve (AUROC).
Results: The incidence of PRAE was 103 patients (37.7%) in the training set and 42 patients (35.6%) in the validation set. LASSO regression identified five variables associated with PRAE: age, BMI, history of acute and chronic respiratory diseases, smoking history, and peak airway pressure during surgery. These variables were included in a multivariate logistic regression analysis, which confirmed them as risk factors for PRAE. The nomogram model was constructed based on these factors. The AUROC for the training set was 0.881 (95% CI 0.817 to 0.945), and for the validation set, it was 0.832 (95% CI 0.779 to 0.885). Calibration curves demonstrated good agreement between predicted and actual PRAE probabilities in both sets.
Conclusion: A nomogram model based on age, BMI, history of acute and chronic respiratory diseases, smoking history, and peak intraoperative airway pressure can effectively predict the occurrence of PRAE in patients undergoing microsurgery under self-retaining laryngoscope. |
|
查看全文
查看/发表评论 下载PDF阅读器 |
| 关闭 |
