Construction of an early prediction model for exacerbation of pneumonia caused by Mycoplasma pneumoniae in children: An incremental value study based on dynamic changes in C-reactive protein and IgG subclasses

Background/Aim. Exacerbation of Mycoplasma pneumoniae pneumonia (MPP) in children commonly manifests as severe or refractory disease. The aim of this study was to evaluate the predictive value of the dynamic percentage change in C-reactive protein - CRP (ΔCRP%) and immunoglobulin (Ig) G1 subclass for the exacerbation of MPP in children, and to determine whether their incorporation into the basic prediction model improves the prediction of disease exacerbation. Methods. This retrospective study included hospitalized pediatric patients diagnosed with MPP between 2020 and 2023, who were enrolled and stratified according to the occurrence of MPP exacerbation. Baseline clinical characteristics, laboratory parameters, serial CRP measurements (at 0 and 72 hrs), and IgG1-IgG4 subclass levels were collected. According to the predicted probabilities generated by the enhanced model, patients were categorized into three risk groups: low-risk (< 0.20), medium-risk (0.20-0.35), and high-risk (≥ 0.35). Results. Among the 512 pediatric patients, 110 (21.5%) experienced MPP exacerbation. Multivariate analysis identified fever duration, bilateral lung involvement, decreased lymphocyte percentage, elevated lactate dehydrogenase, increased D-dimer, higher ΔCRP% (per 10 percentage-point increase), and low IgG1 levels as independent predictors (p < 0.05). The enhanced model demonstrated superior discriminative performance, with an area under the receiver operating characteristic (ROC) curve (AUC) of 0.873, significantly higher than that of the basic model (0.792; p = 0.004). Risk stratification analysis showed a progressive increase in the observed exacerbation rate across the low-, medium-, and high-risk groups (approximately 10%, 30%, and 50%, respectively). Conclusion. Both ΔCRP% and low IgG1 levels are independent predictors of MPP exacerbation in children. The incorporation of these markers significantly improves the model’s discrimination, calibration, and clinical utility for predicting the risk of disease progression.