The development and application of a risk prediction model for cardiac arrest in ICU patients with mechanical ventilation (ICU机械通气患者心脏骤停风险预测模型的构建及应用研究)

Objective To investigate the risk factors of cardiac arrest and to establish the logistic regression model. Methods Totally 238 patients with mechanical ventilation in a tertiary hospital were selected from September, 2019 to January, 2020. Multiple indicators of cardiac arrest group (n=72) and non-cardiac arrest group (n=166) were compared. Logistic regression was used to establish a risk prediction model, and the model power and its accuracy were evaluated by receiver operating characteristic. Results The study finally included arrhythmia, hypoxemia, acid-base imbalance, septic shock, history of cardiac arrest, multiple organ dysfunction syndrome to construct the risk prediction model. This model’s area under curve of receiver operating characteristic was 0. 909. When the best predicted probability was 0. 343, the sensitivity and specificity of the logistic model were 0. 819 and 0. 855, respectively. Conclusion The risk prediction model has satisfactory prediction effects and can be used to predict the risk of cardiac arrest in patients with mechanical ventilation, providing the reference for management and preventative treatment for high-risk cardiac arrest patients. (目的 探讨影响心脏骤停风险的因素, 并建立Logistic回归预测模型。方法 选取2019年9月—2020年1月医院收治的行机械通气的238例患者作为研究对象, 将心脏骤停组(n=72)和非心脏骤停组(n=166)的各项因素进行对比, 通过Logistic回归分析建立风险预测模型, 并通过ROC曲线检验模型的预测效果。结果 本研究共纳入心律失常、低氧血症、酸碱失衡、感染性休克、心脏骤停病史、多器官功能障碍综合症6个影响因素构建风险预测模型。本研究的ROC下面积为0. 909, 当最佳界值为0. 343时, 灵敏度为0. 819, 特异度为0. 855。结论 本模型预测效果良好, 适用于临床实践, 在机械通气患者中应用此模型可为临床医护人员及时采取预防性管理措施提供参考。)