Informed LASSO机器学习方法在心上型完全性肺静脉异位引流术后生存分析中的应用  

Informed LASSO machine learning method in postoperative survival analysis of supracardiac total anomalous pulmonary venous connection

作　　者：刘晓冰[1] 刘付蓉 陈泽宇 徐光正 邱海龙 季尔超 李晓华[1] 温树生[1] 刘涛 庄建[1] LIU Xiaobing;LIU Furong;CHEN Zeyu;XU Guangzheng;QIU Hailong;JI Erchao;LI Xiaohua;WEN Shusheng;LIU Tao;ZHUANG Jian(Department of Cardiovascular Surgery,Guangdong Provincial People's Hospital,Guangdong Academy of Medical Sciences,Guangdong Cardiovascular Institute,Guangzhou,510080,P.R.China;Department of Biostatistics School,Public Health Brown University,Rhode Island,02912,USA;School of Social Science,Tsinghua University,Beijing,100084,P.R.China)

机构地区：[1]广东省人民医院(广东省医学科学院)广东省心血管病研究所心外科,广州510080 [2]布朗大学生物统计系公共卫生学院,美国罗德岛州02912 [3]清华大学社会科学学院,北京100084

出　　处：《中国胸心血管外科临床杂志》2022年第7期848-853,共6页Chinese Journal of Clinical Thoracic and Cardiovascular Surgery

基　　金：广东省科技计划项目(2019B020230003);广东省中医药局科研课题(20181004;20211004);广东省登峰计划项目(DFJH201802);广东省医学科学技术研究基金(A2020029)。

摘　　要：目的研究心上型完全性肺静脉异位引流(total anomalous pulmonary venous connection,TAPVC)病例特征,分析术后死亡事件的相关危险因素,探究Informed LASSO机器学习方法解决罕见先天性心脏病研究中的小样本量问题。方法回顾性分析2009—2019年于广东省人民医院接受手术修复的241例心上型TAPVC患者的临床资料,其中男179例、女62例,中位手术年龄71(33,232)d。首先使用单变量Cox等比分析模型对患者的临床数据进行初步筛选,保留P≤0.05的特征变量。为解决小样本量的限制,我们提出Informed LASSO方法进行多因素Cox风险分析。结果单变量分析中,有统计学意义(P<0.001)的临床变量包括:手术体重≤2.5 kg(HR=16.00)、主肺动脉内径(HR=0.78)、体外循环时间(HR=1.21)、主动脉阻断时间(HR=1.28)和术后呼吸机辅助时间(HR=1.13/d)。Informed LASSO多变量分析表明,术后死亡独立风险因素包括体外循环时间(a HR=1.308/30 min)、年龄(a HR=0.898)、术后呼吸机辅助时间(a HR=1.023/d)、体重≤2.5 kg(aHR=2.545)、右侧垂直静脉回流(a HR=1.977)、术前肺静脉梗阻(a HR=1.633)和急诊手术(a HR=1.383)。结论Informed LASSO利用既往研究结果可提高小样本量的模型分析功效,体外循环时间、手术年龄、术后呼吸机辅助时间、体重、右侧垂直静脉回流、术前肺静脉梗阻和急诊手术为心上型TAPVC术后死亡风险因素。Objective To characterize surgical outcomes of supra-cardiac total anomalous pulmonary venous connection(TAPVC),investigate risk factors for postoperative death,and explore informed LASSO machine learning methods to solve"small sample size problem"in research of rare congenital heart diseases.Methods A retrospective analysis of 241 patients with supra-cardiac TAPVC who underwent surgical repair in Guangdong Provincial People’s Hospital from 2009 to 2019 was conducted,including 179 males and 62 females with a median surgical age of 71(33,232)d.Detailed clinical data of the postoperative death-related factors were extracted.Univariable Cox proportional hazard models were used to initially screen potential risk factors for postoperative death.Factors with P≤0.05 were retained.To solve the limitation of small sample size and the"P>n"problem,we proposed a novel LASSO method for conducting multivariable Cox regression analysis that was capable of bringing in findings of related studies to improve analysis power and to reduce false-negative findings.Results Univariable Cox analyses showed several potential clinical risk factors,among which highly significant factors(P<0.001)included:surgical weight≤2.5 kg(HR=16.00),main pulmonary artery diameter(HR=0.78),prolonged cardiopulmonary bypass time(HR=1.21),aortic block time(HR=1.28),and postoperative ventilator-assisted time(HR=1.13/d).LASSO multivariable analysis revealed that independent risk factors for postoperative death included cardiopulmonary bypass time(a HR=1.308/30 min),age(a HR=0.898),postoperative ventilator-assisted time(a HR=1.023/d),weight≤2.5 kg(a HR=2.545),right vertical venous return(a HR=1.977),preoperative pulmonary venous obstruction(a HR=1.633)and emergency surgery(a HR=1.383).Conclusion Our proposed informed LASSO method can use previous studies’results to improve the power of analysis and effectively solve the"P>n"and small sample size limitation.Cardiopulmonary bypass time,surgical age,postoperative ventilator-assisted time,weight,right ver

关 键 词：完全性肺静脉异位引流 风险因素 机器学习 LASSO 

分 类 号：R654.3[医药卫生—外科学] TP181[医药卫生—临床医学]