Clinical risk factors and predictive value of thin-section CT for bronchial mucus plugs in pediatric mycoplasma pneumoniae pneumonia

Abstract

Background

Bronchial mucus plugs in children with Mycoplasma pneumoniae pneumonia (MPP) are often associated with severe clinical presentations, but diagnosing this condition currently depends largely on invasive bronchoscopy. This study aimed to identify clinical risk factors for mucus plug formation and to evaluate the predictive value of chest thin-section CT imaging features in diagnosing bronchial mucus plugs.

Methods

This retrospective study included 186 children (aged 3–15 years) diagnosed with MPP at the Department of Pediatrics, the Hospital of 82nd Group Army PLA, between May 2023 to June 2024. All patients underwent chest thin-section CT and bronchoscopy. Based on bronchoscopic findings, participants were divided into a mucus plug group (n = 54, 29%) and a non-mucus plug group (n = 132, 71%). Clinical characteristics, inflammatory markers (blood counts, C-reactive protein [CRP], lactate dehydrogenase [LDH]), and CT imaging features (bronchial obstruction, bronchiolitis, lung consolidation) were compared. Univariate and multivariate logistic regression analyses were performed to identify risk factors. Patients were further classified into radiological mucus plug (n = 59) and non-mucus plug (n = 127) groups based on CT-detected bronchial obstruction. Consistency between CT and bronchoscopy results was analyzed. Receiver operating characteristic (ROC) curves assessed the diagnostic performance of CT features and D-dimer.

Results

Univariate analysis revealed significant differences between the two groups in age, lymphocyte count, lymphocyte percentage, C-reactive protein (CRP), D-dimer, peak body temperature, bronchial obstruction, bronchiolitis, and lung consolidation (all P < 0.05). Multivariate analysis identified elevated D-dimer (P = 0.050, OR = 2.048, 95% CI: 1.000–4.195) as an independent risk factor, and bronchial obstruction on CT(P < 0.001, OR = 31.339, 95% CI: 9.922–98.985) as a strong independent predictor for the presence of bronchial mucus plugs. ROC analysis: Bronchial obstruction: AUC = 0.838 (sensitivity: 79.6%, specificity: 87.9%, PPV: 72.9%, NPV: 91.3%). D-dimer: AUC = 0.692 (sensitivity: 56.6%, specificity: 81.7%). lung Consolidation: AUC = 0.643 (sensitivity: 57.4%, specificity: 73.8%).

Conclusion

Elevated D-dimer is an independent risk factor, while bronchial obstruction on thin-section CT is a reliable predictive imaging marker for bronchial mucus plugs in pediatric MPP. Bronchial obstruction on CT offers high diagnostic accuracy, especially with its strong specificity and negative predictive value, supporting its role as a non-invasive tool to guide clinical decisions.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12879-026-13513-1.

Background

Mycoplasma pneumoniae pneumonia (MPP) is a major cause of community-acquired pneumonia in children, accounting for over 40% of cases, with a high incidence [1]. Although MPP is often self-limiting and most children recover with active anti-infective therapy, some cases progress to refractory MPP (RMPP) [2]. Studies indicate that over one-third of RMPP cases involve airway mucus plug formation [3]. Once mucus plugs develop, the disease course lengthens and the risk of secondary infections, such as bacterial, adenoviral, or pertussis, increases [4]. Bronchoscopy has gradually become a critical tool for diagnosing and treating severe MPP (SMPP) and RMPP. It allows direct confirmation of mucus plugs and their removal via alveolar lavage, forceps extraction, or brushing. This approach reduces local inflammation, relieves clinical symptoms, and lowers the risk of long-term complications. However, bronchoscopy is invasive, carries risks, and is costly. Accurate early assessment of mucus plugs could help avoid unnecessary bronchoscopy, reduce patient discomfort, ease family financial burdens [5], and support timely referral from hospitals lacking pediatric bronchoscopy services, thus preventing treatment delays [4]. Although research on high-risk factors for bronchial mucus plugs is increasing, there is still no simple, reliable, non-invasive predictive method. This study aims to explore risk factors for bronchial mucus plug formation and evaluate the consistency between chest thin-section CT imaging features (e.g., bronchial obstruction [see Fig. 1], bronchiolitis, atelectasis) and bronchoscopic findings, thereby establishing a convenient, reliable, non-invasive predictive method to guide clinical decision-making.

Fig. 1.

Schematic Diagram of Bronchial Obstruction as a Diagnostic Imaging Feature. (I) Normal anatomy: A₁ and B₁ represent a normal bronchial bifurcation, with a₁ and b₁ at the same level on the contralateral side. (II) Bronchial narrowing: B₂ is significantly narrowed compared to A₂, and is also markedly smaller compared to b₂. (III) Complete obstruction (type 1 – same bifurcation): B₃ is completely occluded compared to A₃. (IV) Complete obstruction (type 2 – contralateral same level): At the same level, branch A₄ is occluded while a₄ remains patent

Methods

Study population

This retrospective study enrolled hospitalized pediatric patients at the Hospital of 82nd Group Army PLA from May 2023 to June 2024. Inclusion criteria were as follows: (1) availability of thin-section chest CT scans (either performed at our hospital or accessible via QR code from external cloud-based scans with clear image quality); (2) complete clinical data; (3) confirmed diagnosis of Mycoplasma pneumoniae pneumonia (MPP); and (4) hospitalized children who underwent bronchoscopy diagnosis and treatment. Exclusion criteria included: (1) history of bronchial asthma; (2) concomitant bacterial, adenoviral, or other infections; (3) endocrine disorders, malnutrition, or immunodeficiency; and (4) incomplete clinical data. The study was approved by the hospital’s ethics committee (Approval No. 2024009).

Diagnostic criteria for MPP [4]

The diagnosis of Mycoplasma pneumoniae pneumonia (MPP) was based on clinical symptoms, signs, and imaging findings, combined with one or both of the following criteria:

1. Single serum MP antibody titer ≥ 1:160 (particle agglutination method) or a fourfold or greater increase in MP antibody titer between paired serum samples during the disease course.

2. Positive detection of MP-DNA, RNA, or MP via next-generation sequencing (NGS) in throat swabs, sputum, or bronchoalveolar lavage fluid (BALF).

Bronchoscopy and lavage procedures

Bronchoscopy for MPP patients followed the Chinese Guidelines for Pediatric Flexible Bronchoscopy [6]. Bronchoscopy was performed after chest thin-section CT, with a median interval of 1 d (interquartile range: 1–2 d; range, 1–5 d). Written informed consent was obtained from guardians before the procedure. Bronchoscopy was performed according to strict clinical indications: (1) imaging findings showing high-density consolidation covering more than two-thirds of one lung lobe; (2) persistent wheezing or hypoxemia, with imaging evidence of diffuse bronchiolitis in a single lung or involving ≥ 4/5 of both lungs, with or without atelectasis or consolidation; (3) suspected mixed infection during the disease course with difficulty in obtaining sputum, requiring bronchoalveolar lavage for etiological examination; or (4) disease duration ≥ 2–3 weeks with poor radiographic absorption, raising suspicion of bronchiolitis obliterans/constrictive bronchiolitis. Patients fasted for 6–8 h before the procedure; those who could not tolerate fasting received intravenous 10% glucose solution. Preoperative nasal irrigation was performed. A combined “anesthesia during advancement” approach with conscious sedation was used. Operators reviewed CT images beforehand to identify infected regions. The healthy lung was examined first, followed by the affected side. Warm normal saline was instilled (1 mL/kg per lavage, ≤ 20 mL per lavage, total ≤ 5–10 mL/kg), followed by suction at 100–200 mmHg to collect BALF. If mucus plugs were identified, repeated lavage was performed; if unsuccessful, plugs were removed using cytology brushes or forceps.All procedures were performed by experienced pediatric respiratory endoscopists. No severe complications, such as pneumothorax, pulmonary hemorrhage, or laryngospasm, occurred. Minor complications were observed in 9 cases (4.8%), including transient low-grade fever (< 38.5 °C) in 3 cases (1.6%), hoarseness in 4 cases (2.2%), and minor nasal mucosal bleeding (requiring no intervention) in 2 cases (1.1%). All minor complications resolved spontaneously without specific treatment.

Thin-section chest CT protocol

Chest thin-section CT scans were performed using a United Imaging uCT760 64-slice spiral CT with the following parameters: collimation width 40 mm, pitch 1.0875, rotation time 0.5 s, scan time 2.8 s, minimum slice thickness 0.625 mm, craniocaudal scanning direction, and field of view 200 mm. Two experienced associate chief radiologists retrospectively reviewed the images, remaining blinded to clinical data and patient grouping. They generated standard reports to assess for bronchial obstruction; any discrepancies were resolved by consensus. The key imaging features assessed were bronchial obstruction, pulmonary consolidation, and bronchiolitis. Bronchial obstruction was defined as either: (1) absence of a bronchus compared to the contralateral bronchus at the same level; or (2) loss of one branch at a bronchial bifurcation while the other branch remains visible (as illustrated in Fig. 1). Patients were classified into ‘imaging-based mucus plug’ and ‘imaging-based non-mucus plug’ groups based on the presence of bronchial obstruction. These groups were then compared with bronchoscopic findings (the gold standard) to evaluate diagnostic consistency.

Clinical data collection

Demographics and clinical parameters included sex, age, and peak fever temperature. Laboratory and imaging data comprised complete blood count, CRP, D-dimer, LDH, ALT, and thin-section CT features (bronchial obstruction, pulmonary consolidation, obliterative bronchiolitis). For patients with multiple test results, the first values after admission were used.

Statistical analysis

Data were analyzed using SPSS 26.0. Categorical variables were presented as n (%), and group comparisons were performed using the chi-square test. For continuous data, normally distributed variables were expressed as mean ± standard deviation (X ± s) and compared using the independent samples t-test; non-normally distributed variables were expressed as median (interquartile range) [M (P25, P75)] and compared between groups using the Mann–Whitney U test.

Univariate analysis was first performed to identify associations between clinically relevant characteristics and mucus plug status. Variables that were statistically significant or clinically meaningful (with a two-sided P-value threshold of 0.2) were entered into a multivariate logistic regression model to calculate adjusted odds ratios. Multicollinearity diagnostics were conducted, and collinear variables were excluded. Binary logistic regression analysis was used to identify factors affecting test result consistency. Receiver operating characteristic (ROC) curve analysis was performed to evaluate the diagnostic performance of D-dimer and CT imaging features for mucus plugs. A P value ≤ 0.05 was considered statistically significant.

Results

General characteristics

During the study period, 538 children were hospitalized and diagnosed with MPP at our hospital; 216 of them underwent bronchoscopy. After excluding 15 cases with secondary bacterial or adenoviral infections and 15 cases with off-site CT scans whose cloud-based thin-section images could not be retrieved, 186 patients were included in the final analysis. Among them, 100 were male (53.8%) and 86 were female (46.2%). Based on bronchoscopic findings, 54 patients (29%) were assigned to the mucus plug group and 132 to the non-mucus plug group. In the mucus plug group, 35 cases (18.8% of the total cohort) required removal of mucus plugs using cytology brushes or forceps. Based on the presence of bronchial obstruction on thin-section chest CT, patients were categorized into an imaging-based mucus plug group (59 cases, 31.7%) and imaging-based non-mucus plug group (127 cases, 68.3%). The baseline demographic and clinical characteristics of the two groups are shown in Table 1. Subgroup analysis based on the time interval between CT and bronchoscopy (≤ 2 d vs. ≥3 d) showed that longer intervals were associated with decreased sensitivity (83.3% to 66.7%) and negative predictive value (93.4% to 81.0%), while specificity and positive predictive value remained stable. Detailed data are presented in Supplementary Table S1.

Table 1.

Baseline demographic and clinical characteristics of patients with and without mucus plugs (Comparative analysis)

Characteristics	With mucus plug (n = 54)	Without mucus plug (n = 132)	χ² Value/t Value	P value
Age (years), Mean ± SD	9.2 ± 2.5	8.2 ± 2.7	t = 2.107	0.035
Sex, n (%)			χ² = 0.098	0.754
Male	30 (55.6%)	70 (53.0%)
Female	24 (44.4%)	62 (47.0%)
White Blood Cell (×10⁹/L), Mean ± SD	7.2 ± 2.0	7.8 ± 2.5	t = 1.337	0.181
Neutrophil Count (×10⁹/L), Mean ± SD	4.7 ± 1.9	4.7 ± 2.0	t = 0.099	0.921
Neutrophil Percentage (%), Mean ± SD	62.8 ± 14.1	59.0 ± 11.2	t = 2.528	0.011
Lymphocyte Count (×10⁹/L), Mean ± SD	1.7 ± 0.7	2.3 ± 1.0	t = 3.920	< 0.001
Lymphocyte Percentage (%), Mean ± SD	26.1 ± 11.4	30.1 ± 9.5	t = 2.962	0.003
CRP (mg/L), Mean ± SD	32.5 ± 29.8	20.1 ± 24.6	t = 3.138	0.002
LDH (U/L), Mean ± SD	258.0 ± 90.9	229.3 ± 53.0	t = 1.157	0.247

Multivariate logistic regression analysis using stepwise selection was performed (Table 2). Variables with statistically significant between-group differences (P < 0.05) were selected as independent variables. In addition, white blood cell count (P = 0.181, between 0.05 and 0.2) was included based on a P < 0.2 threshold due to its clinical relevance. Presence of mucus plugs was used as the dependent variable in the multivariate model.

Table 2.

Multivariate logistic regression analysis of predictors associated with mucus plug formation

Characteristics	P value	OR value	95% CI lower limit	95% CI upper limit
Age	0.109	1.186	0.963	1.462
White Blood Cell Count	0.318	1.136	0.884	1.461
Lymphocyte Count	0.228	0.599	0.260	1.378
CRP	0.599	0.994	0.970	1.018
D-dimer	0.050	2.048	1.000	4.195
Peak Body Temperature	0.145	1.978	0.791	4.950
Bronchial Obstruction	< 0.001	31.339	9.922	98.985
Bronchiolitis	0.709	1.419	0.226	8.917
Lung Consolidation	0.931	1.095	0.141	8.479

Multicollinearity diagnostics were then performed. Two variables showed elevated variance inflation factor (VIF) values: lymphocyte percentage (VIF = 15.30, > 10), indicating severe multicollinearity; and neutrophil percentage (VIF = 9.82, close to 10), suggesting moderate multicollinearity. Given their clinical significance and strong correlations with lymphocyte count and neutrophil count, respectively, these two variables were excluded. After exclusion, the VIF values of the remaining variables ranged from 1.29 to 3.76, indicating no significant multicollinearity.

The final variables included age, white blood cell count, lymphocyte count, CRP, D-dimer, peak body temperature, bronchial obstruction (yes/no), bronchiolitis (yes/no), and pulmonary consolidation (yes/no). Variables with P ≤ 0.05 in the multivariate model were identified as independent influencing factors.

The stepwise multivariate logistic regression analysis showed that D-dimer (P = 0.050) and bronchial obstruction(P < 0.001) were statistically significant independent predictors. Each 1 mg/L increase in D-dimer was associated with a 104.8% higher risk of mucus plug formation; patients with bronchial obstruction had a 31.339-fold higher likelihood of having mucus plugs than those without. However, this very high odds ratio should be interpreted with caution, as it partly reflects the strong diagnostic concordance between CT-detected bronchial obstruction and bronchoscopic findings (Table 3) rather than an independent etiological association.

Table 3.

These groups were then compared with the bronchoscopy-based classification

Radiological group	Bronchoscopy group		Total
	Mucus plug group	Non-mucus plug group
Radiological Mucus Plug Group	43	16	59
Radiological Non-Mucus Plug Group	11	116	127
Total	54	132	186

Patients were divided into imaging mucus plug and imaging non-mucus plug groups based on the presence or absence of bronchial obstruction on thin-section CT. These groups were then compared with bronchoscopic findings (Table 3). Representative CT images and corresponding bronchoscopic views are shown in Fig. 2.

Fig. 2.

Corresponding Images of Thin-Section CT and Bronchoscopy. (I) Thin-section CT showing patent airways (A₁, B₁, a₁, b₁), with the corresponding bronchoscopic image (III) revealing a clear lumen without mucus plugs. (II) Thin-section CT showing occlusion of bronchus A₂, while B₂, a₂, and b₂ remain patent. The corresponding bronchoscopic image (IV) reveals a mucus plug within bronchus E_A2

To further evaluate the predictive efficacy of other imaging features and D-dimer levels for mucus plug formation, receiver operating characteristic (ROC) curve analysis was performed. The results demonstrated that bronchial obstruction, D-dimer levels, and pulmonary consolidation were all effective indicators for predicting the presence of mucus plugs. The optimal D-dimer cut-off value was 0.880 mg/L. Among all evaluated parameters, bronchial obstruction showed the highest predictive efficacy. Refer to Fig. 3; Table 4 for detailed illustrations and data.

Fig. 3.

ROC curves for predicting mucus plugs in mycoplasma pneumoniae pneumonia

Table 4.

ROC curve data for D-dimer, bronchiolitis, lung consolidation, and bronchial obstruction

Variable	AUC	95% CI	P value	Optimal cutoff value	Sensitivity	Specificity
D-dimer	0.692	0.581–0.756	0.000	0.88	0.566	0.817
Bronchiolitis	0.362	0.279–0.445	0.063	-	0.148	0.576
Lung Consolidation	0.643	0.562–0.725	0.002	-	0.574	0.738
Bronchial Obstruction	0.838	0.767–0.908	0.000	-	0.796	0.879

For bronchial obstruction on thin-section CT in predicting mucus plugs, the AUC was 0.838, with a sensitivity of 79.6%, specificity of 87.9%, positive predictive value (PPV) of 72.9%, and negative predictive value (NPV) of 91.3%(as derived from Table 3). For D-dimer, the AUC was 0.692, with a sensitivity of 56.6% and specificity of 81.7%. For lung consolidation, the AUC was 0.643, with a sensitivity of 57.4% and specificity of 73.8%. Among these variables, bronchial obstruction demonstrated the highest AUC value, sensitivity, and specificity.

Discussion

Mycoplasma pneumoniae pneumonia (MPP) is a leading cause of community-acquired pneumonia in children in China, particularly in those over five years of age. Although mild cases typically have a favorable prognosis, refractory, severe, and critical cases may result in complications such as bronchiectasis, bronchiolitis obliterans, and organizing pneumonia [4]. Mucus plugs are a major factor that complicates the treatment of MPP [3]. After invading the respiratory tract, Mycoplasma pneumoniae induces epithelial cell necrosis and shedding, damages cilia, causes airway wall edema, and triggers infiltration of lymphocytes and/or macrophages in the submucosa, appearing as a “sleeve-like” change. Neutrophils may also infiltrate, and the airway lumen may contain inflammatory cells similar to alveolar cells, along with mucus. The mucus is initially transparent but becomes more viscous as the disease advances, eventually forming mucus plugs and, in some cases, plastic bronchitis. Studies have shown that over 30% of refractory Mycoplasma pneumoniae pneumonia (RMPP) cases are complicated by bronchial mucus plugs [3]. Without prompt treatment, impaired bronchial ventilation may lead to irreversible damage. With the continuous development and refinement of bronchoscopy, its application in RMPP treatment allows for direct visualization of the lesion site, removal of blockages, clearance of pathogens and inflammatory secretions, reduction of inflammatory factors, and improvement of bronchial obstruction symptoms, ultimately achieving therapeutic goals [7–9]. However, bronchoscopy is invasive, and broadening its indications without careful selection may cause harm to children. Moreover, it is relatively costly. Therefore, careful evaluation and weighing of benefits and risks before performing bronchoscopy, along with seeking evidence of mucus plugs, have become increasingly important.

Advances in technology have led to the widespread clinical use of chest thin-section CT, particularly for diagnosing pediatric pneumonia. To identify bronchi on chest CT images, evaluation in the lung window is necessary, relying on the CT’s ability to delineate the bronchial wall. When spatial resolution is sufficient to distinguish the bronchial wall, the bronchus can be observed. In both axial and longitudinal views, it appears as a sharply defined, high-density “wall” surrounding a gas-density “lumen.” However, the spatial resolution of conventional CT is limited, making it difficult to visualize the walls of smaller bronchi. In adult pulmonary anatomy [10], the average diameter of segmental bronchi (levels 2–4) is 5–8 mm, with a wall thickness of 1.5 mm; the diameter of airways at levels 6–8 is 1.5–3 mm, with a wall thickness of about 0.3 mm; and the diameter of airways at levels 11–13 is 0.7–1 mm, with a wall thickness of 0.1–0.15 mm. Conventional CT, with a slice thickness of 5 mm to 1 cm, cannot clearly visualize bronchi below level 4, nor can it continuously observe changes in pulmonary imaging [11]. Moreover, children’s airways are narrower than those of adults, and their bronchial walls are relatively thinner, making conventional CT less suitable for observing small airway changes in children. In contrast, thin-section CT, with a smaller slice thickness of about 1–2.5 mm, can visualize bronchi at levels 5–6 and above [12]. In Mycoplasma pneumoniae pneumonia, bronchial wall thickening or peribronchial inflammatory exudates create a sharp contrast with the airway lumen, making it easier to observe bronchial pathways and small airway conditions [13]. This study found that bronchial obstruction has significant clinical value in suggesting the presence of mucus plugs. As defined in Fig. 1, bronchial obstruction was characterized as either: (1) complete obstruction at the same bifurcation (type 1), where one branch is occluded while the other remains visible, as shown in Fig. 1-III (B3 vs. A3); or (2) absence of a bronchus compared to the contralateral bronchus at the same level (type 2), as shown in Fig. 1-IV (A4 vs. a4). The results of this study confirm that chest thin-section CT findings, particularly bronchial obstruction as a diagnostic imaging feature, can serve as a reliable predictive imaging marker for the presence of bronchial mucus plugs. The findings show that bronchial obstruction, which is a characteristic thin-section CT finding, has high concordance with bronchoscopy results for predicting mucus plugs, achieving a sensitivity of 79.6%, specificity of 87.9%, PPV of 72.9%, and NPV of 91.3%. This imaging feature can serve as a valuable non-invasive diagnostic supplement for mucus plugs in MPP. Nevertheless, the very high odds ratio for bronchial obstruction observed in the multivariate analysis (OR = 31.339) should be interpreted cautiously. This high OR primarily reflects the diagnostic concordance between CT and bronchoscopy rather than an independent etiological predictive ability. The clinical utility of this CT feature lies mainly in its high specificity and negative predictive value for identifying mucus plugs, supporting its role as a non-invasive diagnostic tool rather than a conventional risk factor.

Based on the above findings, a practical algorithm is proposed to guide the use of bronchoscopy in children with refractory, severe, or critical Mycoplasma pneumoniae pneumonia (MPP): (1) Perform chest thin-section CT.(2) If CT shows bronchial obstruction, the probability of mucus plugs is high (PPV 72.9%). Bronchoscopy is recommended for confirmation and removal.(3) If CT does not show obstruction, the negative predictive value is 91.3%. In such cases, a normal D-dimer level (< 0.88 mg/L, the optimal cut-off derived from our ROC analysis, see Table 4) further reduces the likelihood of mucus plugs, and bronchoscopy may be deferred with close clinical follow-up.(4) If CT is negative but D-dimer is elevated (> 0.88 mg/L), clinical judgment should be exercised, as the risk of false-negative CT exists (sensitivity 79.6%).This algorithm requires prospective validation before widespread implementation.

It should be noted that thin-section CT is reconstructed from the same raw data as conventional chest CT and does not require additional scanning; therefore, the radiation dose is essentially the same as that of a routine chest CT. For critically ill children, the clinical benefits of accurately identifying bronchial mucus plugs and guiding appropriate bronchoscopy use clearly outweigh the potential radiation risks. Nevertheless, CT still involves radiation exposure and should not be used excessively or indiscriminately in any pediatric population.

Despite the high consistency between chest CT imaging features and bronchoscopy results in this study, false positives and false negatives still occurred. Bronchial mucus plugs consist of inflammatory exudates, necrotic epithelial cell debris, and abnormally secreted mucoproteins. Epithelial damage impairs ciliary function, reducing airway clearance [14]. These materials have a higher density than gas and become visible when they accumulate in the airways. Therefore, once mucus plugs form, they should be detectable on thin-section CT. In some patients, thin-section CT suggested small airway bronchial obstruction indicative of mucus plugs, but bronchoscopy did not confirm them on the affected side. Possible reasons for false positives include: (1) pulmonary and airway congestion and edema, airway compression, and the presence of non-mucus plug materials such as sputum or flocculent, cord-like substances filling the lumen; (2) mucus plugs that do not completely fill the airway but adhere to the bronchial wall, making them difficult to distinguish on thin-section CT. Some CT scans showed segmental airway obstruction, suggesting this possibility, and such mucus plugs were removed by suction without the use of forceps or brushes; (3) imaging errors: the thin-section CT used in this study was reconstructed and not high-resolution CT, which may still result in imaging errors. In other cases, thin-section CT did not show bronchial obstruction or other signs of mucus plugs, but bronchoscopy revealed mucus plugs. Common reasons for false negatives may include a delay between CT and bronchoscopy, during which the patient’s condition may have changed. In this study, bronchoscopy was performed 1–5 d after thin-section CT. Two typical cases showed no signs of mucus plugs on CT, but bronchoscopy revealed mucus plugs in areas not indicated by CT. The interval between CT and bronchoscopy in these cases was 4 and 5 d, respectively. Considering the clinical history, it is likely that the patient’s condition changed during this period, leading to discrepancies between CT imaging and bronchoscopy results.

Multiple factors influence the formation of bronchial mucus plugs. Wu Huifang et al. reported that D-dimer can predict mucus plugs in MPP [15], a finding consistent with our results. In our study, D-dimer had a sensitivity of 56.6% and specificity of 81.7%. The optimal cut-off value was 0.88 mg/L, and elevated D-dimer levels showed a strong association with mucus plug formation, emerging as an independent factor. D-dimer is a specific structure in fibrin degradation products [16], and its clinical significance lies in its elevation indicating a hypercoagulable state and secondary hyperfibrinolysis. Studies have shown that D-dimer can serve as an indicator of severe infection and is also a marker for assessing the severity of community-acquired pneumonia [17, 18]. Elevated D-dimer levels are associated with community-acquired pneumonia, sepsis, and other diseases, and are positively correlated with disease severity [19]. In this study, D-dimer levels were significantly higher in the MPP group with mucus plugs, indicating a severe systemic inflammatory response. A large number of inflammatory mediators and immune complexes cause immune damage to the body, destroying respiratory epithelial cell membranes [20], leading to epithelial cell necrosis and shedding, and the formation of mucus plugs that obstruct the airways. Han Yu et al. [21] also pointed out that MPP induces local hypoxia, ischemia, and acidosis, damaging vascular endothelial cells, activating the coagulation system, and releasing procoagulant substances, resulting in a hypercoagulable state and elevated D-dimer levels. Additionally, the aggregation and degradation of inflammatory cells release a large number of inflammatory substances, leading to the formation of highly viscous materials that promote mucus plug formation [22].

This study has several limitations. First, as a retrospective, single-center study that only included severely ill children who underwent bronchoscopy, selection bias may have been introduced, and the external validity of our findings is therefore limited. Caution is warranted when generalizing the results to the broader population of children with Mycoplasma pneumoniae pneumonia, especially to those with mild disease who do not meet the clinical indications for bronchoscopy. Future prospective, multicenter studies including patients with varying degrees of disease severity are needed to further validate the generalizability of these predictive markers. Second, the interval between chest CT and bronchoscopy (median 1 d, range 1–5 d) means that we cannot completely exclude the possibility that dynamic changes in the disease course may have influenced the observed correlations. Third, although thin-section CT is becoming more widely available, its application in evaluating bronchial obstruction depends on advanced imaging infrastructure, which may restrict its immediate use in resource-limited settings.

Conclusion

This study highlights the significant value of chest thin-section CT in identifying bronchial mucus plugs in children with Mycoplasma pneumoniae pneumonia. With continued advances in CT technology and the growing use of cloud-based imaging platforms, clinicians can now access thin-section CT images more conveniently via hospital networks and mobile devices. By continuously improving their image interpretation skills and integrating common clinical indicators such as age, peak body temperature, D-dimer, and CRP, clinicians can more accurately predict the presence of mucus plugs. However, given the single-center design and the inclusion of only bronchoscopy-selected patients, the external validity of these findings requires further validation in multicenter studies. These findings are also applicable to primary healthcare settings, where they can enhance diagnostic and treatment services and reduce unnecessary bronchoscopy use.

Electronic Supplementary Material

Below is the link to the electronic supplementary material.

Abbreviations

MPP

Mycoplasma pneumoniae pneumonia

RMPP

Refractory Mycoplasma pneumoniae pneumonia

SMPP

Severe Mycoplasma pneumoniae pneumonia

CT

Computed Tomography

CRP

C-reactive protein

LDH

Lactate dehydrogenase

ALT

Alanine aminotransferase

D-dimer

D-dimer (fibrin degradation product)

ROC

Receiver operating characteristic

AUC

Area under the curve

BALF

Bronchoalveolar lavage fluid

NGS

Next-generation sequencing

Author contributions

Liming Li, Zhiliang Zhang and Fang Yuan made substantial contributions to the design of this work. Haiyan Chen, Yanjie Wang, and Yongjuan Wang contributed to the data collection and analysis. Chunhui Song, and Kun Wang interpreted the results to draft and revise the manuscript.

Funding

None.

Data availability

The data-sets analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

This study was conducted in strict adherence to the ethical principles of the Declaration of Helsinki. The study protocol was reviewed and approved by the Ethics Committee of The 82nd Army Group Hospital of the PLA (Ethics Approval Number: 2024009). All methods were performed in accordance with relevant guidelines and regulations.

Consent for publication

Not applicable.

Clarification of guardian consent

The consent provided by guardians on behalf of patients under the age of 16 was informed consent. Parents or legal guardians of all children under 16 years of age participating in this study were fully informed of the details of the study prior to enrollment and voluntarily signed written informed consent forms. All study participants (including their guardians) provided written informed consent for future research.

Competing interests

The authors declare no competing interests.