急性A型主动脉夹层术后肺炎风险预测模型构建与评价  

Construction and assessment of a risk predictive model for postoperative pneumonia in patients with acute type A aortic dissection

作　　者：谢茜 严丹阳 付翔杰 黄凌瑾[3] 李宁[1,2] 姚润 XIE Xi;YAN Dan-yang;FU Xiang-jie;HUANG Ling-jin;LI Ning;YAO Run(Department of Blood Transfusion,Transfusion Research Center,Xiangya Hospital,Central South University,Changsha 410008,China;National Clinical Research Center for Geriatric Disorders,Changsha 410008,China;Department of Cardiovascular Surgery,Xiangya Hospital,Central South University,Changsha 410008,China)

机构地区：[1]中南大学湘雅医院输血科临床输血研究中心,湖南长沙410008 [2]国家老年疾病临床医学研究中心(湘雅医院),湖南长沙410008 [3]中南大学湘雅医院心脏大血管外科,湖南长沙410008

出　　处：《中国感染控制杂志》2023年第7期788-795,共8页Chinese Journal of Infection Control

基　　金：湖南省自然科学基金项目(2022JJ40840、2022JJ70076);湖南省卫生健康委员会科研项目(202218013370);中南大学湘雅医院青年基金项目(2021Q13)。

摘　　要：目的 探讨急性A型主动脉夹层(AAAD)患者术后肺炎的危险因素,构建AAAD患者术后肺炎的预测模型。方法 回顾性选取2014年1月—2020年6月于中南大学湘雅医院进行AAAD紧急手术的患者为研究对象。通过全子集回归筛选预测因子,以及约登指数确定连续型资料的截断值,基于二分类logistic回归构建AAAD患者术后肺炎预测模型。采用受试者工作特征曲线下面积(AUC)、校准曲线和决策曲线评价模型的性能。结果 共纳入210例AAAD患者,53例(25.24%)发生术后肺炎。采用全子集回归共筛选出6个预测因子并根据约登指数确定截断值:男性(OR=2.21,95%CI:0.88~5.54)、慢性肺部疾病(OR=2.53,95%CI:1.12~5.74)、血小板分布宽度17.5%(OR=3.27,95%CI:1.57~6.78)、手术时长9 h(OR=2.76,95%CI:1.25~6.06)、机械通气时长99 h(OR=3.87,95%CI:1.63~9.18)和红细胞输注9单位(OR=1.69,95%CI:0.80~3.60)。构建的AAAD术后肺炎风险预测模型AUC为0.789(95%CI:0.718~0.860),最佳风险阈值为0.21。校准曲线和Hosmer-Lemeshow检验(P=0.48)显示该模型具有良好的校准度。决策曲线显示在术后肺炎预测风险为0~69%的患者使用本模型能够获益。结论 本研究构建的模型可以较好地预测AAAD患者发生术后肺炎的风险,临床可对高风险患者实施早期干预。Objective To explore the risk factors for postoperative pneumonia in patients with acute type A aortic dissection(AAAD)and construct a predictive model for postoperative pneumonia in AAAD patients.Methods Patients who underwent emergency surgery for AAAD at Xiangya Hospital of Central South University from January 2014 to June 2020 were retrospectively selected for the study.Predictive factors were screened by all-subsets regression,and the cut-off values for continuous data were determined by the Youden index.A binary logistic regression model was constructed to predict postoperative pneumonia in AAAD patients.The efficacy of the model was assessed by the area under the receiver operating characteristic curve(AUC),calibration curve,and decision curve.Results A total of 210 AAAD patients were included,53(25.24%)of whom developed postoperative pneumonia.Six predictive factors were screened through all-subsets regression,and the cut-off values were determined by the Youden index:male patients(OR=2.21,95%CI:0.88-5.54),chronic pulmonary disease(OR=2.53,95%CI:1.12-5.74),platelet distribution width>17.5%(OR=3.27,95%CI:1.57-6.78),surgical duration>9 hours(OR=2.76,95%CI:1.25-6.06),mechanical ventilation duration>99 hours(OR=3.87,95%CI:1.63-9.18),and red blood cell transfusion>9 units(OR=1.69,95%CI:0.80-3.60).The AUC of the constructed predictive model for postoperative pneumonia was 0.789(95%CI:0.718-0.860),and the optimal risk threshold was 0.21.The calibration curve and Hosmer-Lemeshow test(P=0.48)demonstrated good calibration of the mo-del.The decision curve showed that patients with a predicted postoperative pneumonia risk of 0-69%could benefit from using this model.Conclusion The predictive model constructed in this study can effectively predict the risk of postoperative pneumonia in AAAD patients.Early intervention can be implemented for high-risk patients in clinical practice.

关 键 词：急性A型主动脉夹层 手术后肺炎 预测模型 医院感染 

分 类 号：R181.32[医药卫生—流行病学]