支气管肺炎患儿反复喘息的影响因素分析及预测模型构建

Analysis of influencing factors of recurrent wheezing in children with bronchopneumonia and construction of prediction model

目的探究支气管肺炎患儿反复喘息的临床影响因素并构建风险预测模型。方法155例支气管肺炎患儿据出院后1年内反复喘息发生情况分为反复喘息组47例和非反复喘息组108例。比较2组一般临床资料,包括住院时间,总病程,湿疹史,个人过敏史(食物过敏史、药物过敏史、变应性鼻炎史、变应性皮炎史等),生活环境,感冒次数,变应原暴露,实验室指标[T淋巴细胞亚群、嗜酸性粒细胞(EOS)、25-羟基维生素D3、总免疫球蛋白E(IgE)]。采用多因素Logistic回归分析反复喘息的影响因素并构建风险预测模型,Homser-Lemeshow检验拟合优度,绘制受试者工作特征(ROC)曲线评估模型的预测效能。结果与非反复喘息组比较,反复喘息组的个人过敏史、EOS阳性(>9%)、维生素D不足(<30μg/L)及总IgE阳性(>60 IU/mL)比例高(P<0.05)。多因素Logistic回归分析显示,有个人过敏史、EOS阳性、维生素D不足、总IgE阳性是支气管肺炎患儿反复喘息的独立危险因素(P<0.05)。预测模型Homser-Lemeshow检验有较好的拟合优度(χ^(2)=3.055,P=0.692),模型预测患儿反复喘息的曲线下面积为0.816,敏感度、特异度分别为70.21%、80.56%。结论基于个人过敏史、EOS阳性、维生素D不足、总IgE阳性影响因素构建的预测模型对患儿反复喘息具有良好的预测效能。

Objective To explore clinical influencing factors of recurrent wheezing in children with bronchopneumonia,and construct a risk prediction model.Methods A total of 155 children with bronchopneumonia were divided into the recurrent wheezing group(47 cases)and the non-recurrent wheezing group(108 cases)according to the occurrence of recurrent wheezing within 1 year after discharge.General clinical data,including the length of hospital stay,total course of disease,history of eczema,history of other personal allergies(food allergy history,drug allergy history,allergic rhinitis history and allergic dermatitis history),family history of allergies,living environment,frequency of colds and exposure to allergens,and laboratory indexes[T lymphocyte subsets,eosinophils(EOS),25-hydroxy vitamin D3 and total immunoglobulin E(IgE)]were compared between the 2 groups.Multivariate Logistic regression analysis was conducted to identify influencing factors of recurrent wheezing,and a risk prediction model was constructed.Homser-Lemeshow tested the goodness of fit of the model.The predictive performance of the model was evaluated using the receiver operating characteristic(ROC)curve.Results Compared with the non-recurrent wheezing group,the proportions of children with history of other personal allergies,EOS positivity(>9%),vitamin D insufficiency(<30μg/L)and total IgE positivity(>60 IU/mL)were higher in the recurrent wheezing group(P<0.05).Multivariate Logistic regression analysis showed that history of personal allergies,EOS positive,vitamin D insufficiency and total IgE positive were independent risk factors for recurrent wheezing in children with bronchopneumonia(P<0.05).Homser-Lemeshow showed that prediction model had goodness of fit(χ^(2)=3.055,P=0.692).The model predicted that the area under curve for recurrent wheezing was 0.816,sensitivity and specificity of the model for predicting recurrent wheezing were 70.21%and 80.56%,respectively.Conclusion The prediction model constructed based on history of personal allergies,EOS positive,vitamin D insufficiency and total IgE positive has good predictive performance for recurrent wheezing in children.