Objective To identify clinical risk factors and characterize associated features of necrotizing pneumonia (NP) in children with severe pneumonia. Methods This retrospective case–control study included children with severe pneumonia hospitalized at the Children's Hospital of Soochow University between July 2022 and June 2025. Children who met the diagnostic criteria for NP were classified into the NP group, while a 1:2 control group (NNP) was randomly selected from eligible patients without NP during the same period. Clinical characteristics were compared between the two groups, followed by risk factor analysis and construction of ROC curves. Results A total of 33 children with NP were included. Compared with the NNP group, significant differences were observed in duration of fever, hemoptysis, wheezing, chest pain, duration of corticosteroid use after admission, alanine aminotransferase, lactate dehydrogenase, albumin, white blood cell count (WBC), neutrophil percentage, platelet count, C-reactive protein, fibrin degradation products (FDPs), fibrinogen, D-dimer, anticoagulant therapy, pulmonary consolidation, pleural effusion, mucus plugs, and the number of bronchoscopic procedures (all P 0.05). After variable selection using LASSO regression under the λ 1se criterion, duration of fever, wheezing, chest pain, WBC, FDPs, and the number of bronchoscopic procedures remained associated with NP. Further multivariable logistic regression and ROC curve analyses demonstrated that chest pain, elevated WBC and FDPs were independent factors associated with NP. The combined model of these three variables showed good predictive performance for NP, with an AUC of 0.941 (95% CI: 0.891–0.990), and both sensitivity and specificity of 87.88%. Conclusion Chest pain, elevated WBC, and increased FDPs were independently associated with NP, suggesting that these factors may help identify children at higher risk and inform clinical assessment.
Che et al. (Wed,) studied this question.