儿童先天性心脏病手术后重症肺炎的风险预测研究

Risk prediction for severe pneumonia following congenital heart disease surgery in children

目的建立并验证儿童先天性心脏病体外循环手术后发生重症肺炎的风险预测模型。方法将2014年10月至2017年12月于重庆医科大学附属儿童医院胸心外科确诊为先心病并接受体外循环下手术治疗的患者纳入建模组(n=1415),将2017年12月至2018年10月接受体外循环手术的患者纳入验证组(n=393)。采用单因素和多因素Logistic回归分析筛选体外循环术后发生重症肺炎的独立危险因素。根据回归系数建立相应的列线图。用受试者工作特征曲线(ROC)下的面积(AUC)及校准曲线来评价模型。结果单因素和多因素Logistic回归分析发现:年龄、体质量、术前住院时间、先天性心脏手术风险调整分级(RACHS-1评分)、体外循环时间是影响体外循环手术后重症肺炎的5个独立危险因素。根据5个独立危险因素,建立个体化预测模型。模型评价结果显示:建模组和验证组的AUC分别为0.943(95%CI,0.929-0.954)和0.925(95%CI,0.894-0.949),校准试验的P值分别为0.592和0.256。结论本研究建立的个体化预测模型可较好地应用于评估先天性心脏病患儿体外循环术后发生重症肺炎的风险。

Objective To establish and validate a risk prediction model for severe pneumonia following cardiopulmonary bypass(CPB)in children with congenital heart disease.Methods All the children diagnosed with congenital heart disease and treated with CPB surgery in the Department of Thoracic Surgery,Children's hospital of Chongqing Medical University from October 2014 to December 2017 were assigned into the modeling group(n=1415),while those treated from December 2017 to October 2018 were enrolled into the validation group(n=393).The independent risk factors for severe pneumonia after surgery were screened using univariate and multivariate Logistic regression analysis,and the corresponding nomogram prediction model was constructed according to the regression coefficients.The receiver operating characteristic(ROC)curve and calibration curve were used respectively to evaluate the discriminant validity and calibration of the prediction model.Results The following 5 indicators were revealed to be the independent risk factors for severe pneumonia after CPB surgery:age,weight,preoperative hospital stay,score of Risk Adjustment in Congenital Heart Surgery-1(RACHS-1)and time of cardiopulmonary bypass.Based on them,an individualized prediction model was established.The area under the curve(AUC)of the prediction model was 0.943(95%CI:0.929-0.954)and 0.925(95%CI:0.894-0.949)for the modeling and validation groups,indicating good discriminant validity.The P value of the calibration test was 0.592 and 0.256 for the 2 groups.Conclusion Our established individualized prediction model can be well used to evaluate the risk of severe pneumonia in children with congenital heart disease after CPB.