Constructing A Risk Warning Model for Severe Mycoplasma Pneumoniae Pneumonia Children Based on Clinical Multi Parameters
10.13241/j.cnki.pmb.2025.03.015
- VernacularTitle:基于临床多参数构建重症肺炎支原体肺炎患儿的风险预警模型
- Author:
Wan-ting MO
1
;
Ping-ming GAO
;
Rui-ping WAN
;
Hui-wen XIAN
;
Dan-xia LIN
Author Information
1. 南方医科大学第二临床医学院 广东 广州 510515;佛山市妇幼保健院儿科 广东佛山 528000
- Publication Type:Journal Article
- Keywords:
Severe;
Mycoplasma pneumoniae pneumonia;
General data;
Laboratory indicator;
Imaging indicator;
Risk warning model
- From:
Progress in Modern Biomedicine
2025;25(3):511-518
- CountryChina
- Language:Chinese
-
Abstract:
Objective:To construct a risk warning model for severe mycoplasma pneumoniae pneumonia(SMPP)children based on clinical data,laboratory indicators and imaging indicators.Methods:162 Mycoplasma pneumoniae pneumonia(MPP)children who were admitted in Foshan Women and Children Hospital from January 2021 to December 2023 were selected,64 SMPP children were included in severe group,the remaining 98 children were included in mild group.The general data,laboratory indicators and imaging indicators of the children were collected.The influencing factors for the occurrence of SMPP were analyzed by univariate and multivariate logistic regression models,and a risk warning model for the occurrence of SMPP children was constructed based on multivariate logistic regression model.The predictive value of the risk warning model for the occurrence of SMPP were analyzed by receiver operating characteristic(ROC)curve.Results:The proportion of 3 years old ≤ age＜6 years old,course of disease,body temperature,fever course,C-reactive protein(CRP),erythrocyte sedimentation rate(ESR),lactate dehydrogenase(LDH),cyanosis of lips,positive triconcave sign,pleural effusion,lesion site was the lower lobe,abnormal electrocardiogram and extrapulmonary manifestations in severe group were significantly higher than those in mild group(P＜0.05),there were no significant differences in gender,white blood cell count(WBC),neutrophil ratio and procalcitonin(PCT)between the two groups(P＞0.05).Multivariate logistic regression analysis model showed that,3 years old ≤age＜6 years old,high body temperature,long fever course,CRP elevated,ESR elevated,LDH elevated,cyanosis of lips,positive triconcave sign,pleural effusion,lesion site was the lower lobe,abnormal electrocardiogram and extrapulmonary manifestations were risk factors for the occurrence of SMPP(P＜0.05).ROC curve analysis showed that,the area under the curve(AUC)of the risk warning model was 0.829,the sensitivity was 84.82％,and the specificity was 78.15％,the actual prediction curve of the risk warning model was in good agreement with the prediction curve,the decision curve showed that,the threshold probability range of the model was 4.61％～88.14％.Conclusion:The risk warning model based on clinical multi parameters such as general data,laboratory indicators and imaging indicators has certain predictive value for the occurrence of SMPP.
