摘要
目的 探寻儿童肺炎支原体肺炎的证素特点,根据中医证素与临床数据构建风险预测模型,并对模型的预测效能进行评价。方法 回顾性选取2021年9月至2022年8月在山东中医药大学附属医院儿科住院的肺炎支原体肺炎患儿病历资料180例,根据是否诊断为大叶性肺炎分为大叶性肺炎组(80例)和非大叶性肺炎组(100例)。采用单因素分析与二元Logistic回归分析筛选预测因子,初步建立诊断预测模型,通过Hosmer-Lemeshow拟合度检验和受试者工作特征曲线(ROC曲线)评估模型,并绘制列线图。结果 二元Logistic回归分析结果显示,年龄、呕吐痰涎、乳酸脱氢酶(LDH)、中医证素湿和热是儿童肺炎支原体性感染导致大叶性肺炎的危险因素(P <0.05),基于上述预测因素构建Logistic回归模型P=1/[1+exp(-1.794X1+0.014X2+1.668X3+0.503X4+0.293X5-6.23)],并以列线图形式呈现。Hosmer-lemeshow检验结果显示模型拟合度优(χ~2=7.775,P=0456),ROC曲线下面积为0.834,诊断敏感度为92.5%,诊断特异性为63%。结论 初步构建包括年龄、呕吐痰涎症状、LDH以及中医证素热、湿共5个预测因素在内的儿童肺炎支原体肺炎感染大叶性肺炎的风险预测模型,其拟合度与准确性较好,可以在临床上作为诊断预测的参考。
Objective:To explore the characteristics of syndrome elements of mycoplasma pneumoniae pneumonia in children,construct a risk prediction model based on TCM syndrome elements and clinical data,and evaluate the prediction efficiency of the model.Methods:Medical records of 180 children with mycoplasma pneumoniae pneumonia who were hospitalized in the pediatrics department of the Affiliated Hospital of Shandong University of Chinese Medicine from September 2021 to August 2022 were retrospectively selected.The 180 children who met the inclusion criteria were selected as the study subjects,and were divided into the lobar pneumonia group(80 cases)and non-lobar pneumonia group(100 cases)according to whether they were diagnosed with lobar pneumonia.Single factor analysis and binary Logistic regression analysis were used to screen the predictors,and a preliminary diagnosis prediction model was established.The model was evaluated by Hosmer-Lemeshow fit test and receiver operating characteristic curve(ROC curve),and a nomogram was drawn.Results:Binary Logistic regression analysis showed that age,vomiting of phlegm-drool,LDH,TCM syndrome elements of dampness and heat were the risk factors for lobular pneumonia caused by mycoplasma pneumoniae infection in children(P<0.05).Based on the above predictive factors,the Logistic regression model P=1/[1+exp(-1.794X1+0.014X2+1.668X3+0.503X4+0.293X5-6.23)]was constructed and presented in the form of a column graph.Hosmer-Lemeshow test showed that the model had a good fit(χ~2=7.775,P=0456),the area under ROC curve was 0.834,the diagnostic sensitivity was 92.5%,and the diagnostic specificity was 63%.Conclusion:The risk prediction model of mycoplasma pneumoniae infection with lobar pneumonia in children was established,which included age,vomiting of phlegm-drool,lactate dehydrogenase,TCM syndrome elements of heat and dampness.The model has good fitting degree and accuracy,and can be used as a reference for diagnosis and prediction in clinic.
作者
张昊晨
潘月丽
周旭
Zhang Haochen;Pan Yueli;Zhou Xu(Shandong Traditional Chinese Medicine University,Shandong,Jinan 250014,China)
出处
《中国中医急症》
2024年第8期1322-1326,共5页
Journal of Emergency in Traditional Chinese Medicine
基金
国家自然科学基金项目(82104933)
山东省中医药科技青年项目(2020Q015)。
关键词
肺炎支原体肺炎
大叶性肺炎
中医证素
预测模型
列线图
Mycoplasmic pneumonia
Lobar pneumonia
TCM syndrome element
Prediction model
Nomogram
