【摘要】 目的 探究分析重症肺炎患者气管插管后低血压（PIH）的独立影响因素，构建PIH预测模型。方法 选择2023年1月—2024年1月江苏省人民医院接受治疗的重症肺炎患者资料，共500例。将其分为训练集（n=350）和验证集（n=150）。训练集按照是否发生PIH分为PIH组（n=166）与非PIH组（n=183）。收集比较两组患者资料。多因素Logistic回归分析筛选出独立影响因素。构建预测模型并评估其临床实用性和校准度。验证集患者作为外部验证组进行验证。结果 单因素分析结果中，PIH组患者年龄、APACHE II评分均高于非PIH组，BMI、SBP、DBP均低于非PIH组，而慢性阻塞性肺疾病（COPD）合并症占比高于非PIH组（t=6.563、9.332、10.226、8.347、6.236，χ~2=11.590，均P<0.05）。多因素Logistic回归分析显示，年龄、APACHE II评分、BM、SBP、DBP、COPD均为重症肺炎患者发生PIH的独立影响因素（P<0.05）。ROC曲线下最大面积（AUC）为0.884,95%CI:0.846～0.916。外部验证的验证集ROC曲线中，AUC值为0.914,95%CI:0.858～0.954。预测模型的决策曲线与校准曲线提示模型具有临床实用性和良好校准度。结论 本研究构建的重症肺炎患者PIH预测模型的预测效能较好，具有一定的预测价值。更多还原

【Abstract】 ObjectiveTo explore and analyze the independent influencing factors of post-intubation hypotension(PIH)in patients with severe pneumonia after tracheal intubation, and to construct a PIH prediction model.MethodsThe data of patients with severe pneumonia treated in Jiangsu Province People’s Hospital from January 2023 to January 2024were selected, with a total of 500 cases. They were divided into a training set(n=350) and a validation set(n=150). The training set was divided into a PIH group(n = 166) and a non-PIH group(n=183) according to whether PIH occurred.The data of the two groups of patients were collected and compared. Multivariate Logistic regression analysis was used to screen out the independent influencing factors. The prediction model was constructed and its clinical practicability and calibration degree were evaluated. The patients in the validation set were used as an external validation group for verification.ResultsIn the results of univariate analysis, the age and APACHE II score of patients in the PIH group were higher than those in the non-PIH group, and the BMI, SBP and DBP were lower than those in the non-PIH group,while the proportion of chronic obstructive pulmonary disease( COPD) complications was higher than that in the nonPIH group( t = 6.563, 9.332, 10.226, 8.347, 6.236, χ~2= 11.590, P < 0.05). Multivariate Logistic regression analysis showed that age, APACHE II score, BMI, SBP, DBP, and COPD were all independent influencing factors for the occurrence of PIH in patients with severe pneumonia(P < 0.05). The maximum area under the ROC curve(AUC) was 0.884,95% CI: 0.846-0.916. In the ROC curve of the validation set for external validation, the AUC value was 0.914, 95%CI: 0.858-0.954. The decision curve and calibration curve of the prediction model suggested that the model had clinical practicability and good calibration degree. Conclusion The predictive model for PIH in patients with severe pneumonia constructed in this study demonstrated good predictive efficacy and has significant predictive value, making it a practical tool for healthcare professionals to identify high-risk patients for PIH following intubation.更多还原