儿童游客高原肺水肿预测模型的构建与验证  

Development and validation of a predictive model for high altitude pulmonary edema in pediatric tourists

作　　者：刘铃钶 郭生春 蔡延萍 欧广源 鲍的羊子 李守元 潘波艇 LIU Lingke;GUO Shengchun;CAI Yanping;OU Guangyuan;BAO Deyangzi;LI Shouyuan;PAN Boting(Department of Pediatrics,the Affiliated Hospital of Shaoxing University,Shaoxing 312000,China;不详)

机构地区：[1]绍兴文理学院附属医院儿科,312000 [2]大柴旦行委人民医院内儿科 [3]大柴旦行委人民医院放射科

出　　处：《浙江医学》2025年第5期506-511,共6页Zhejiang Medical Journal

基　　金：浙江省医学会临床医学科研专项资金资助项目(2023ZYC-A183)。

摘　　要：目的构建并验证儿童游客高原肺水肿(HAPE)预测模型。方法回顾性分析2023年6月至2024年6月大柴旦行委人民医院收治的133例因发热、咳嗽、呼吸困难或低氧血症等临床表现行肺部影像学检查的患儿,其中诊断为HAPE 65例(HAPE组),未患HAPE 68例(非HAPE组)。比较HAPE组和非HAPE组患儿临床病理资料,采用logistic回归分析发生HAPE的影响因素,基于这些影响因素构建预测HAPE风险的列线图。通过ROC曲线、校准曲线和决策曲线分析预测模型预测儿童游客发生HAPE的效能。结果儿童游客发生HAPE的影响因素包括症状出现天数、外周血氧饱和度、体质指数和发热。ROC曲线显示该模型预测HAPE的AUC为0.869,显示了较好的预测能力。校准曲线表明预测概率与观察概率之间有良好的一致性(Hosmer-Lemeshow检验P=0.512)。决策曲线显示在0~0.5的阈值概率范围内该模型显示出显著的净收益。结论使用影响因素构建的HAPE预测模型可较好预测儿童游客发生HAPE风险,有助于HAPE的早期识别和干预。Objective To develop and validate the predictive model for high altitude pulmonary edema(HAPE)in pediatric tourists.Methods A retrospective analysis was conducted on 133 pediatric patients undergoing pulmonary imaging examination due to fever,cough,dyspnea or hypoxemia who were admitted to Dachaidan People's Hospital from June 2023 to June 2024.Among them,65 patients were diagnosed with HAPE(HAPE group),while 68 did not(no HAPE group).The clinicopathological data of HAPE group and no HAPE group were compared.Logistic regression analysis was performed to identify significant predictors for HAPE.The performance of the predictive nomogram constructed based on the above identified predictors was evaluated using ROC curves,calibration curve,and decision curve analysis.Results The main predictors of HAPE of pediatric tourists included the number of days since symptom occurrence,peripheral blood oxygen saturation,body mass index,and fever.The ROC curve demonstrated strong discriminative ability,with an AUC of 0.869.Calibration analysis showed good agreement between predicted and observed probabilities(Hosmer-Lemeshow test,P=0.512).Decision curve analysis highlighted the clinical utility of the model,particularly in the 0 to 0.5 probability threshold range,where significant net benefit was observed.Conclusion The HAPE prediction model constructed using influencing factors can better predict the risk of HAPE occurrence in pediatric tourists,which is helpful for early identification and intervention of HAPE.

关 键 词：高原肺水肿 儿童游客 预测模型 

分 类 号：R59[医药卫生—内科学]