Introduction

Mycoplasma pneumoniae (MP) is one of the major pathogens responsible for community-acquired pneumonia (CAP) in children. Although Mycoplasma pneumoniae pneumonia (MPP) is generally self-limiting, the incidence of refractory MPP (RMPP) and severe cases has been increasing in recent years, accompanied by more complications and a heavier disease burden1,2. MP infection can trigger a strong immune response and mucosal injury, leading to excessive mucus secretion, impaired ciliary function, and airway obstruction3. Radiologically, pulmonary consolidation is a common manifestation of MPP. However, complications within the airway—such as plastic bronchitis, mucus plug obstruction, or severe mucosal damage—are often difficult to detect by chest X-ray or CT alone, making bronchoscopy a valuable diagnostic tool in such cases4.

Bronchoalveolar lavage (BAL), as an adjunctive treatment for MPP, can effectively remove airway secretions and mucus plugs, promote resolution of inflammation and lung re-expansion, and reduce complications4. However, as an invasive and high-cost procedure, BAL is not suitable for all MPP patients with pulmonary consolidation. At present, there is a lack of objective criteria to determine which patients truly require BAL intervention.

This study aimed to retrospectively analyze the clinical, laboratory, and radiological data (including quantitative pulmonary consolidation volume) of 323 children with MPP and pulmonary consolidation (BAL group: 163 cases; non-BAL group: 160 cases) in order to identify independent risk factors for BAL intervention and to develop and validate a predictive model. The model is expected to enhance clinical decision-making accuracy, promote individualized management, minimize unnecessary procedures, and optimize the allocation of healthcare resources.

Methods

Study population

This was a retrospective cohort study involving 323 children with MPP and radiologically confirmed pulmonary consolidation, admitted to the Department of Pediatrics at the Second Affiliated Hospital of Guangxi Medical University between September 2020 and October 2023. Eligible patients ranged in age from 1 month to 15 years and met the following criteria: (1) diagnosed with Mycoplasma pneumoniae pneumonia (MPP) according to the 2023 Chinese guideline (Expert consensus on the diagnosis and treatment of Mycoplasma pneumoniae pneumonia in children)5. Specifically, in addition to clinical symptoms and signs, all patients had a confirmed MP infection evidenced by at least one of the following laboratory criteria: (i) a positive nucleic acid amplification test (NAAT) for MP-DNA or MP-RNA, or (ii) a single serum particle agglutination test titer of anti-MP antibody ≥ 1:160, or a demonstrated fourfold or greater increase in antibody titer during the course of the illness; (2) pulmonary consolidation confirmed by chest CT; (3) BAL evaluated or performed based on the 2018 Chinese guidelines for pediatric flexible bronchoscopy6. Exclusion criteria included: (1) severe underlying diseases; (2) allergy to macrolides; (3) contraindications for bronchoscopy; (4) withdrawal from treatment or lack of consent; and (5) incomplete clinical data.

Data collection

Participants were divided into the BAL group (n = 163) and the non-BAL group (n = 160) based on whether they underwent BAL. Data extracted from electronic medical records included: (1) demographic and clinical characteristics (age, sex, hospital stay, fever peak/duration, lung auscultation findings, corticosteroid therapy); (2) laboratory tests: White blood cell (WBC), Neutrophil Percentage (NEU%), Lymphocyte Percentage (LYM%), High-sensitivity C-reactive protein (hsCRP), Procalcitonin (PCT), Lactic dehydrogenase (LDH), Creatine kinase isomer-MB (CK-MB), D-Dimer (D-D), Fibrinogen (FIB), Erythrocyte Sedimentation Rate (ESR), Alanine Aminotransferase (ALT), Immunoglobulin A (IgA), Immunoglobulin M (IgM), Immunoglobulin G (IgG); (3) imaging features (atelectasis, pleural effusion, consolidation extent). Pulmonary consolidation volume was quantified using Vitrea® Advanced Visualization software (Canon Medical Systems).

Statistical analysis

The normality of continuous variables was assessed using the Shapiro-Wilk test. Data conforming to a normal distribution were expressed as mean ± standard deviation (\({ {\bar {\rm x}}}\) ± s), and comparisons between groups were performed using the independent samples t-test. Non-normally distributed data were presented as median and interquartile range [M (P25–P75)], and the Mann–Whitney U test (rank-sum test) was used for group comparisons. Categorical variables were expressed as frequencies and percentages, and comparisons between groups were conducted using the chi-square test or Fisher’s exact test, as appropriate. Variables with statistically significant differences in univariate analysis were included in a multivariate logistic regression model using a backward stepwise selection method to identify independent predictors of BAL intervention. A nomogram was then constructed based on the final logistic regression model. The predictive performance of the model was assessed using the area under the receiver operating characteristic (ROC) curve (AUC) and calibration plot. Statistical analyses were performed using SPSS version 26.0 and R software version 4.3.2. A two-tailed P value ≤ 0.05 was considered statistically significant.

Results

Clinical characteristics

There were no significant differences in sex between the BAL and non-BAL groups (P > 0.05). The BAL group had a significantly longer hospital stay [7 (6,8) vs. 5 (4,7) days, P < 0.001]. BAL cases were more frequent in summer (38.7%), whereas non-BAL cases peaked in winter (51.3%) (P < 0.001). The BAL group also had higher fever peaks, longer fever duration, and longer macrolide use (all P < 0.05), but lower glucocorticoid use and wheezing incidence (P < 0.05), as shown in Table 1.

Table 1 Comparison of clinical characteristics between the non-BAL and BAL groups.
Full size table

Laboratory findings

The BAL group showed significantly higher neutrophil percentage, hsCRP, and D-dimer levels, and lower WBC and CK-MB levels compared to the non-BAL group (P < 0.05), as shown in Table 2.

Table 2 Comparison of laboratory parameters between the Non-BAL and BAL groups.
Full size table

Radiological characteristics

Atelectasis (9.82% vs. 1.25%), pleural effusion (9.20% vs. 3.75%), proportion of lung consolidation (1.36% vs. 0.28%) and Consolidation volume (9620.00 mm3 vs. 2035.90 mm3) were significantly higher in the BAL group (P < 0.05). Single-lobe consolidation was more common in the non-BAL group (P < 0.05), as shown in Table 3.

Table 3 Comparison of radiological features between the Non-BAL and BAL groups.
Full size table

Predictive model development

Variables with statistically significant differences in univariate analysis were included in a multivariate logistic regression model using a backward stepwise method. The analysis identified pulmonary consolidation proportion (OR = 2.10), age (OR = 1.01), length of hospital stay (OR = 1.61), and atelectasis (OR = 6.49) as independent risk factors for undergoing BAL. In contrast, corticosteroid use was found to be a protective factor (OR = 0.25), as shown in Table 4.

Table 4 Multivariate logistic regression analysis.
Full size table

Based on the identified independent predictors, a nomogram model was constructed to estimate the probability of undergoing BAL (Fig. 1). ROC curve analysis demonstrated good discriminatory power, with an area under the curve (AUC) of 0.87 (95% CI: 0.83–0.91) (Fig. 2). The calibration plot indicated satisfactory agreement between predicted and observed probabilities (Fig. 3). For instance, in a child with MPP and pulmonary consolidation—aged 4 years, hospitalized for 6 days, not treated with corticosteroids, with a consolidation proportion of 2%, and no atelectasis—the corresponding scores for each variable were as follows: 24 points for hospitalization duration, 8 points for age, 24 points for absence of corticosteroid use, and 0 points for absence of atelectasis, yielding a total of 56 points. The predicted probability of requiring BAL was approximately 75%.

Fig. 1
Fig. 1The alternative text for this image may have been generated using AI.
Full size image

Nomogram prediction model for undergoing BAL in pediatric patients.

Fig. 2
Fig. 2The alternative text for this image may have been generated using AI.
Full size image

ROC curve of the nomogram prediction model.

Fig. 3
Fig. 3The alternative text for this image may have been generated using AI.
Full size image

Calibration curve of the nomogram prediction model.

Discussion

The incidence of MPP has increased among pediatric CAP cases, with a rising proportion of RMPP among those with pulmonary consolidation7. MP infection can disrupt immune homeostasis and provoke a robust cellular immune response, damaging ciliary function and impeding mucus clearance—factors that contribute to airway obstruction, consolidation, and atelectasis8. Early identification and appropriate intervention are critical to preventing disease progression and serious complications.

BAL has become increasingly utilized in the diagnosis and management of pediatric MPP due to its ability to dilute and remove viscous secretions and necrotic tissue, visualize airway pathology, and collect representative pathogen samples4,9,10.

In this study, the median age of children in the BAL group was higher than that in the non-BAL group, which may be attributed to the more mature immune systems in older children, making them more prone to excessive inflammatory responses to Mycoplasma pneumoniae (MP) infection. In addition, increasing age has been associated with a higher risk of macrolide resistance11,12,13. Correspondingly, children in the BAL group exhibited higher peak temperatures, longer durations of fever and hospitalization, and significantly prolonged courses of macrolide and tetracycline antibiotic therapy. These findings suggest that persistent high fever reflects a stronger inflammatory response, and a prolonged disease course may further trigger complications such as plastic bronchitis and atelectasis, thereby increasing disease severity14.

Interestingly, the non-BAL group had higher rates of glucocorticoid use and wheezing, possibly reflecting a younger age and higher prevalence of wheezing disorders or allergic tendencies in this subgroup15. Glucocorticoids suppress pro-inflammatory cytokines (e.g., IL-1β, IL-2, IL-6, MIP-1β, IL-8) and may reduce the need for BAL in patients with milder disease or those with an allergy-related inflammatory phenotype16,17.

Inflammatory markers such as neutrophil percentage, D-dimer, and hsCRP were elevated in the BAL group, indicating more severe systemic inflammation, which markers have been previously associated with RMPP11,13,16,18,19. Although other parameters like LDH, ALT, and immunoglobulins have been proposed as severity indicators, they did not reach significance in this study, possibly due to sample or site limitations.

Bronchoscopic findings in the BAL group commonly included mucosal hyperemia, edema, follicular hyperplasia, mucus plugs, and bronchial narrowing—85% had mucus plugs and 38% had follicular changes—consistent with characteristic airway pathology of MPP4,20,21. These changes are linked to MP’s P1 protein-mediated adherence and subsequent epithelial damage and immune infiltration.

CT imaging further revealed that BAL group patients had higher rates of atelectasis and pleural effusion. Quantitative consolidation analysis using Vitrea® confirmed a significantly higher proportion of lung consolidation in the BAL group, aligning with clinical and inflammatory severity. Prior studies have successfully used CT-derived parameters in predictive models for RMPP22,23, supporting their utility in clinical assessment.

Several studies have developed models to predict the need for BAL in MPP, incorporating variables such as D-dimer, ALT, fever duration, CRP, and consolidation volume12,23,24. In this study, age, proportion of lung consolidation, hospital stay, and atelectasis were identified as risk factors, while glucocorticoid use was protective. The resulting nomogram had strong predictive power (AUC 0.86; sensitivity 0.83; specificity 0.74), and offers an intuitive tool for bedside application.

Conclusions

In a nutshell, BAL plays a crucial diagnostic and therapeutic role in children with MPP and pulmonary consolidation. The predictive model developed in this study offers a practical reference for identifying candidates for BAL, facilitating timely and individualized treatment.