Abstract: Background：The severity of mycoplasma pneumoniae pneumonia (MPP) is related to the formation of mucus plugs (MUP) in the airway, but the current MUP detection mainly depends on bronchoscopy. Objective：To establish a nomogram predictive model based on clinical features and plasma cytokines for airway MUP formation in MPP children. Design：Case-control study. Methods：Children who were diagnosed as MPP and underwent bronchoscopy were classified into MUP and non-MUP groups. Clinical features, inflammatory markers and cytokines in both plasma and BALF were collected. Univariate analysis was performed to identify statistically significant clinical features, inflammatory markers, and cytokine risk factors between the two groups. Spearman correlation analysis was conducted to assess the correlation between inflammatory markers, plasma cytokines and BALF cytokines and to exclude collinearity issues. A receiver operating characteristic (ROC) curve was generated to assess the predictive performance of a multivariable logistic regressionbased predictive model using the R package of RMS. Main outcome measures：Predictive efficacy of the nomogram model for predicting the probability of airway mucus plug formation in children with MPP. Results：A total of 263 children with MPP were included in the analysis, including 134 males and 129 females, with an average age of (7.0 ± 0.2) years. There were 82 (31.2%) cases in the MUP group and 181 cases in the non-MUP group. Univariate analysis showed significant differences between the two groups in age, maximum temperature before bronchoscopy, decreased breath sounds upon auscultation at admission, chest CT showing pulmonary consolidation, imaging suggesting pleural effusion, extrapulmonary complications, proportion of severe pneumonia, neutrophil percentage, CRP, PCT, D-dimer, LDH, IgA, lymphocyte percentage, and ALT. Cytokine analysis revealed significant differences between the two groups in levels of IL-5, IL-6, IL-8, and IFN-γ in plasma, as well as in levels of IL-1β, IL-5, IL-8, IL-10, IFN-γ, and TNF-α in BALF. Spearman correlation analysis showed there was no collinearity in neutrophil percentage, CRP, PCT, LDH, D-dimer, IgA, plasma IFN-γ, plasma IL-6, plasma IL-5, and plasma IL-8; there was a certain correlation between blood inflammatory markers and BALF cytokines; plasma cytokines (IFN-γ, IL-6, IL-5, and IL-8) were positively correlated with their corresponding BALF cytokines. The nomogram predictive model for the formation of airway mucus plugs in MPP patients, based on age, pleural effusion, D-dimer, and plasma IFN-γ levels, had an AUC of 0.817 (95% CI: 0.747-0.889) with the sensitivity of 79.0% and specificity of 69.1%. Conclusion：The nomogram predictive model based on age, pleural effusion, D-dimer, and plasma IFN-γ levels had a good predictive performance for the formation of airway mucus plugs in MPP patients.

Key words: Mycoplasma pneumoniae pneumonia, Mucus plug, Children, Predictive model