|
| 心脏外科手术患者术后谵妄危险因素分析及预测模型建立 |
| Risk factors and nomogram prediction model establishment for postoperative delirium in patients undergoing cardiac surgery |
| |
| DOI：10.12089/jca.2025.12.002 |
| 中文关键词: 心脏手术 术后谵妄 列线图 预测模型 |
| 英文关键词: Cardiac surgery Postoperative delirium Nomogram Prediction model |
| 基金项目:辽宁省重点技术资助项目(2020JH2/10300121) |
|
| 摘要点击次数: 1433 |
| 全文下载次数: 424 |
| 中文摘要: |
目的：探讨全身麻醉下行心脏外科手术患者发生术后谵妄（POD）的危险因素，并构建列线图预测模型。
方法：选择2021年5—12月择期行全身麻醉下心脏手术患者689例，男451例，女238例，年龄>18岁，ASAⅢ或Ⅳ级。收集患者围术期相关资料，根据术后7 d内是否发生POD将患者分为两组：POD组和非POD组。将采集的患者资料数据集按7∶3的比例随机分为训练集和验证集。采用LASSO回归和多因素Logistic回归分析训练集，筛选危险因素并构建列线图预测模型，绘制受试者工作特征（ROC）曲线并计算曲线下面积（AUC），评估模型的预测效能。采用校准曲线和决策曲线分析（DCA）验证模型的准确性和临床实用性。
结果：有149例（21.6%）患者术后7 d内发生POD。多因素Logistic回归分析显示，肺部疾病、糖尿病、心房颤动、简易智力状态检查量表（MMSE）评分偏低及术后输血是心脏手术后发生POD的独立危险因素（P<0.05）。基于上述危险因素构建列线图预测模型，训练集的AUC为0.723（95%CI 0.659～0.769），验证集的AUC为0.751（95%CI 0.698～0.801）。校准曲线显示，列线图模型预测曲线与实测曲线基本吻合。DCA曲线显示，列线图模型预测心脏手术患者发生POD有较好的临床效益。
结论：肺部疾病、糖尿病、心房颤动、MMSE评分偏低及术后输血是心脏外科手术患者发生POD的独立危险因素，基于此构建的列线图模型具有良好的预测效能及临床应用价值。 |
| 英文摘要: |
Objective: To investigate the risk factors for postoperative delirium (POD) in patients undergoing cardiac surgery under general anesthesia and to construct a nomogram prediction model.
Methods: A total of 689 patients, 451 males and 238 females, aged > 18 years, ASA physical status Ⅲ or Ⅳ, who underwent elective cardiac surgery under general anesthesia from May to December 2021 were enrolled. Perioperative data were collected, and the patients were divided into two groups based on POD occurrence within 7 days postoperatively: POD group and non-POD group. The dataset was randomly split into a training set and a validation set in a 7∶3 ratio. LASSO regression and multivariate logistic regression analyses were applied to the training set to identify risk factors and develop the nomogram model. Receiver operating characteristic (ROC) curves were plotted, and the area under the curve (AUC) was calculated to evaluate predictive performance. Calibration curves and decision curve analysis (DCA) were used for model validation.
Results: A total of 149 patients (21.6%) developed POD within 7 days postoperatively. Multivariate logistic regression identified pulmonary disease, diabetes mellitus, atrial fibrillation, decreased mini-mental state examination (MMSE) score, and postoperative blood transfusion as independent risk factors for POD after cardiac surgery (P < 0.05). The nomogram model constructed using these risk factors achieved an AUC of 0.723 (95% CI 0.659-0.769) in the training set and 0.751 (95% CI 0.698-0.801) in the validation set, respectively. The calibration curve showed that the predicted curve of the nomogram model was basically consistent with the measured curve. The DCA curve showed that this nomogram model had good clinical benefits in predicting the occurrence of POD during cardiac surgery.
Conclusion: Pulmonary disease, diabetes mellitus, atrial fibrillation, decreased MMSE score, and postoperative blood transfusion are independent risk factors for POD after cardiac surgery. The nomogram model constructed based on these factors exhibits strong predictive performance and clinical application value for assessing the risk of POD in patients undergoing cardiac surgery. |
|
查看全文
查看/发表评论 下载PDF阅读器 |
| 关闭 |
