Characteristics and influencing factors of bronchoalveolar lavage fluid cellular patterns in children with Mycoplasma pneumoniae pneumonia: a cross-sectional study

Introduction

Mycoplasma pneumoniae (MP) is the most important pathogen of community-acquired pneumonia worldwide, particularly among children and adolescents (1), accounting for 20% to 40% of community-acquired pneumonia cases during epidemic periods (2). During the coronavirus disease 2019 (COVID-19) pandemic, its incidence declined significantly due to the implementation of non-pharmaceutical interventions (NPIs). However, a large-scale global surveillance study showed that the MP detection rates reached an unprecedented high by the end of 2023 (3). Although the vast majority of infections are mild and self-limiting (4), some patients develop severe disease with complications such as atelectasis, pleural effusion and pulmonary embolism, or even long-term sequelae like bronchiolitis obliterans, posing a severe threat to children’s health.

At present, the clinical assessment of Mycoplasma pneumoniae pneumonia (MPP) in children relies largely on blood inflammatory markers [such as white blood cell count (WBC), C-reactive protein (CRP), procalcitonin (PCT), lactate dehydrogenase (LDH), etc.] and imaging findings. However, blood inflammatory markers primarily reflect the systemic inflammatory status and may not accurately represent changes in the local immune microenvironment of the lung (5). Bronchoscopy allows for direct visualization of pathological changes of the airway mucosa, while analysis of BALF cellular profiles can offer a unique window into the local inflammatory response in the lungs. In healthy children and adults, BALF cellular profiles are predominantly characterized by macrophages (approximately 85–95%), with small proportions of lymphocytes (5–15%) and neutrophils (≤3%), eosinophils (≤1%) (6-8). In contrast, in children with MPP, BALF cellular profiles show marked alterations that reflect the nature and intensity of pulmonary inflammation.

Accumulating evidence indicates that BALF cellular patterns in children with MPP are characterized by neutrophil predominance, although these findings have been observed primarily in patients with severe MPP (9,10). However, the correlation between these cellular patterns and clinical features has not been well explained. MP-DNA load has been shown to correlate with various clinical and laboratory parameters (11,12). Specifically, several studies have reported that the high MP-DNA load group exhibits elevated levels of inflammatory cytokines and a higher proportion of BALF neutrophils, and that high MP-DNA load is associated with enhanced airway inflammation (5,13). Moreover, high MP loads and persistent long-term MP-DNA in lower airway were associated with severity of pediatric MPP (11). Nevertheless, the specific role of MP-DNA load in shaping BALF cellular patterns—particularly neutrophil predominance—and the correlation between common blood inflammatory markers and BALF cellular patterns remain unclear. In addition, an analysis of serum and BALF cytokine profiles found that different age groups exhibited distinct immune response patterns, and that most serum and BALF cytokine levels were not significantly correlated, suggesting that age is an important confounding factor and that local and systemic immune responses are relatively independent (14). Although that study was based on BALF cytokines rather than cellular profiles, its findings on age-related immune patterns differences and local-systemic dissociation provide important background for the present study.

Despite these important advances, several critical knowledge gaps persist. First, the majority of existing studies have been limited by relatively small sample sizes (typically <200 participants), which compromises statistical power. Moreover, the lack of standardized cytological and classification criteria limits cross-study comparability and generalizability. Second, most investigations have focused on severe or refractory MPP, whereas large-scale studies encompassing the full spectrum of disease severity remain scarce. Third, previous studies have not adequately adjusted for key confounders, it remains unclear whether MP-DNA load independently associates with BALF neutrophilic inflammatory response. Fourth, whether common blood inflammatory markers can reflect local pulmonary inflammation as represented by BALF cellular pattern, has not been systematically evaluated using adequate statistical power and multivariable adjustment.

To address these knowledge gaps, we designed a large-scale, single-center, retrospective cross-sectional study with rigorous methodological standards. The innovations of this study are threefold: (I) large sample with consecutive enrollment: we consecutively enrolled 238 children with confirmed MPP, providing sufficient statistical power for multivariable analysis; (II) standardized cytological examination and quality control based on the Chinese expert consensus (15); (III) comprehensive confounding control: we incorporated local MP-DNA load, common blood inflammatory markers, disease course (time from symptom onset to bronchoscopy), and pre-bronchoscopy systemic corticosteroid use into multivariable regression models to identify independent associated factors of BALF cellular pattern(neutrophil predominance).

Therefore, this study aimed to: (I) characterize BALF cellular patterns in a large, consecutively enrolled pediatric MPP cohort; (II) investigate the independent influencing factors of BALF neutrophil predominance after adjusting for key confounders; (III) evaluate whether common blood inflammatory markers correlate with BALF cellular profiles; and (IV) explore the association between BALF lymphocyte proportion and pleural effusion as a secondary exploratory analysis. The findings may have potential clinical relevance for assessing pulmonary immune status and inflammatory phenotype in children with MPP. We present this article in accordance with the STROBE reporting checklist (available at https://tp.amegroups.com/article/view/10.21037/tp-2026-0194/rc).

Methods

Study design and participants

This was a single-center, cross-sectional study conducted at the Department of Respiratory, Tianjin Children’s Hospital. Tianjin Children’s Hospital is a tertiary comprehensive children’s hospital in northern China, serving both urban and rural populations. All exposure and outcome variables were measured during the acute phase of hospitalization within a narrow time window (blood tests within 24 hours of admission, BALF obtained during bronchoscopy within 72 hours of admission), and no longitudinal follow-up.

We applied consecutive sampling. All pediatric patients hospitalized with MPP who underwent bronchoscopy with bronchoalveolar lavage (BAL) between November 2023 and April 2024 were screened for eligibility.

Inclusion criteria

• Pediatric patients hospitalized in our department between November 2023 and April 2024, diagnosed with MPP and met the criteria for bronchoscopy.
• Age range: ≥1 month to ≤16 years.
• MPP was diagnosed according to the “Guidelines for the Diagnosis and Treatment of Mycoplasma pneumoniae Pneumonia in Children (2023 Edition)” issued by the National Health Commission of the People’s Republic of China (16). Confirmation of MP infection required a positive MP-DNA test in bronchoalveolar lavage fluid (BALF).
• Underwent bronchoscopy with BAL based on established clinical indications as outlined in the relevant guidelines (6,16).

Exclusion criteria

• Patients with severe underlying medical conditions, including but not limited to immunodeficiency disorders, severe neuromuscular diseases, congenital heart disease, or major respiratory tract malformations.
• Patients with chronic inflammatory airway diseases (e.g., asthma, recurrent respiratory infections, bronchiectasis).
• Patients with co-infections involving other identified pathogens.
• Patients with incomplete clinical records (missing key variables including blood tests, BALF cytology, or MP-DNA load).

Participant flow

During the study period, a total of 306 consecutive patients were screened. Of these, 68 were excluded: 32 had co-infections (6 with bacterial co-infection, 26 with viral co-infection), 28 had incomplete records, 8 had chronic inflammatory airway diseases (asthma, recurrent respiratory infections). The remaining 238 patients were included in the final analysis.

Ethical statement

This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments, and received approval from the Institutional Ethics Committee of Tianjin Children’s Hospital (approval No. 2022-LXKY-004).

Informed consent

Due to its retrospective design, the requirement for informed consent was waived by the Institutional Ethics Committee of Tianjin Children’s Hospital. All patient data were anonymized and de-identified prior to analysis. No identifiable personal information (e.g., names, medical record numbers, or addresses) was included in the dataset.

Clinical data collection and variables

All data were retrospectively extracted from electronic medical records using a standardized data collection form. Clinical characteristics were assessed at admission, blood samples were collected within 24 hours of admission, and BALF was obtained during bronchoscopy performed within 72 hours of admission. All variables were measured using standardized protocols across all participants. Two research assistants independently reviewed the electronic medical records, and discrepancies were resolved by a third senior investigator.

Demographic and clinical characteristics

The following variables were collected: age (years), gender, peak fever temperature (℃), duration of fever (days), time from symptom onset to bronchoscopy (days), presence of atelectasis (yes/no), presence of pleural effusion (yes/no), length of hospital stay (days), and use of systemic corticosteroids prior to bronchoscopy (yes/no).

Blood laboratory indicators

Blood samples were analyzed in the Clinical Laboratory and the Biochemistry Laboratory of Tianjin Children’s Hospital. The indicators were as follows: WBC, neutrophil count (NE), lymphocyte count, platelet count (PLT), which were tested by a Mindray BC-7500cs automated hematology analyzer; CRP, PCT, LDH, ferritin (FER) using a Roche COBAS-8000 modular analyzer; D-dimer (D-D), fibrinogen (FIB) were measured using a fully automated coagulation analyzer; and serum cytokines (7 items): interleukin-2 (IL-2), interleukin-4 (IL-4), interleukin-6 (IL-6), tumor necrosis factor-α (TNF-α), interleukin-17A (IL-17A), interleukin-10 (IL-10), interferon-γ (IFN-γ) were quantified via flow cytometry-based multiplex assays with the BD CBA Kit (BD Biosciences, Franklin Lakes, New Jersey, USA).

BALF samples

BALF samples were obtained during the first bronchoscopy procedure. MP-DNA load (copies/mL) was quantified using a commercial qPCR kit (DaAn Gene Co., Ltd., Guangzhou, China) following the manufacturer’s instructions.

Bronchoscopy procedure

Bronchoscopy was performed according to the indications, standard operational protocols and relevant guidelines (6,16). The patient was placed in the supine position, and the bronchoscope was inserted through one nasal cavity. Each segmental bronchus was examined sequentially step-by-step, and the lesion site (the most severely affected area on chest imaging) was lavaged with 0.9% normal saline. The BALF was aspirated under negative pressure into a sterile specimen container and immediately sent for microbiological and cytological examinations.

BALF cytological examination

First, mucus was removed from the BALF sample. A volume of 0.5–1.0 mL of the preprocessed BALF was placed in a cytocentrifuge filter and centrifuged at 800 rpm for 3 minutes using a centrifuge (Liqun Medical Technology Co., Ltd., Shandong, China). The slides were then removed, air-dried, stained with hematoxylin and eosin, and mounted.

Cell sorting and counting were conducted under an optical microscope at a magnification of ×40 by two independent technicians blinded to clinical data. A minimum of 400 cells were counted per slide. Cell classifications were expressed as percentages [e.g., neutrophil percentage = (neutrophil count / total nucleated cells) × 100%]. All procedures were performed in accordance with the Consensus of Chinese Experts on Morphological Examination of Bronchoalveolar Lavage Fluid Cells [2023] (15). Disagreements (>5% difference) were resolved by a third senior technician. Cell classification was based on standardized morphological criteria under light microscopy (×40):

• Neutrophils: polymorphonuclear cells with multi-lobed nuclei and neutral-staining cytoplasm;
• Lymphocytes: mononuclear cells with round, dark-staining nuclei and scant cytoplasm;
• Macrophages: large mononuclear cells with abundant cytoplasm and irregular or kidney-shaped nuclei;
• Eosinophils: polymorphonuclear cells with bi-lobed nuclei and eosinophilic cytoplasmic granules.

Outcome and exposure definitions

The primary outcome was the percentage of neutrophils in BALF, expressed as a continuous variable (%). The primary exposure variable was MP-DNA load (copies/mL) in BALF. Due to the skewed distribution of raw values, MP-DNA load was log₁₀-transformed for parametric analyses (log₁₀ MP-DNA load).

Statistical analysis

Statistical analysis was performed using SPSS software (version 20.0). Two-sided P values <0.05 were considered statistically significant.

Descriptive analysis

Continuous variables were tested for normality using the Shapiro-Wilk test. Normally distributed data were expressed as mean ± standard deviation (SD); non-normally distributed data were presented as median with interquartile range (IQR) [M (P₂₅, P₇₅)]. Categorical variables as numbers and percentages [n (%)].

Univariable analysis

Given the non-normal distribution of most variables, correlations between BALF cell proportions (neutrophils, lymphocytes, macrophages) and clinical/laboratory variables were assessed using Spearman’s rank correlation coefficient. For subgroup comparisons, the Kruskal-Wallis H test (followed by Mann-Whitney U with Bonferroni correction) was used to compare BALF cell proportions across MP-DNA load groups: low-load (10³–10⁴ copies/mL), medium-load (10⁵–10⁶ copies/mL), and high-load (≥10⁷ copies/mL). The Mann-Whitney U test was used to compare patients with and without pleural effusion.

Multivariable analysis

To identify factors independently associated with BALF neutrophil percentage (the primary outcome), a multiple linear regression model was constructed. The dependent variable was BALF neutrophils percentage (continuous). The independent variables included:

• Primary independent variable: BALF MP-DNA load (log₁₀-transformed to normalize its distribution);
• Covariates selected a priori based on univariable significance (P<0.05), previous literature (5,9,17) and clinical plausibility: age (years), gender, time from symptom onset to bronchoscopy (days), the levels of CRP, IFN-γ, and use of systemic corticosteroids prior to bronchoscopy (yes/no). IL-2 and TNF-α were not included in the final model because of severe collinearity [variance inflation factor (VIF) >10].

Multicollinearity was assessed using the variance inflation factor, with VIF >5 indicating significant collinearity. Residual plots were examined to verify the assumptions of linearity and homoscedasticity. Results were reported as unstandardized coefficients (B), standardized β coefficients, 95% confidence intervals (CIs) for B, and P values.

Secondary exploratory analysis

To explore the potential association between BALF lymphocyte percentage and pleural effusion, a secondary exploratory analysis was performed. Patients were divided into pleural effusion (n=38) and non-pleural effusion (n=200) groups. Univariable comparisons were made using the Mann-Whitney U test. Multivariable logistic regression was not feasible due to the small number of pleural effusion cases (n=38). Therefore, only univariable comparisons (Mann-Whitney U test) are presented for this exploratory analysis.

Sensitivity analysis

To assess the robustness of the primary findings after excluding collinear variables, the sensitivity analyses were performed by alternately replacing CRP or IFN-γ with IL-2 or TNF-α in the multivariable regression model.

Missing data

No missing data were present for the primary variables of interest.

Results

Participant characteristics and BALF cellular patterns

A total of 306 hospitalized children were screened, and 238 patients were included in the final analysis, consisting of 117 males and 121 females, with a male-to-female ratio of 0.97. The mean age was 7.79±2.76 years. The mean duration of fever was 7.14±2.44 days, with a mean peak temperature of 39.45±0.64 ℃, and the mean time from symptom onset to bronchoscopy was 8.48±3.23 days. The length of hospital stay was 7.01±2.16 days. Atelectasis was present in 118 patients (49.58%), while pleural effusion was observed in 38 patients (15.97%). Systemic corticosteroids were administered before bronchoscopy in 194 patients (81.5%).

BALF cellular patterns analysis showed marked neutrophil predominance, with a median proportion of 80.0% (IQR, 71.8%, 84.0%). Lymphocyte and macrophage proportions were substantially lower, with medians of 3.0% (IQR, 2.0%, 5.0%) and 13.5% (IQR, 11.0%, 19.0%), respectively. Eosinophils were rarely observed (median 0%, IQR, 0%, 0 %). The median MP-DNA load in BALF was 10^6.68 (10^6.14, 10^7.33) copies/mL. The details are shown in Table 1.

Univariable correlations between BALF cellular patterns and clinical/laboratory variables

Spearman correlations analysis was performed to assess associations between BALF cellular patterns (percentage of neutrophils, lymphocytes, macrophages) and clinical/laboratory variables, as shown in Figure 1.

Figure 1 Correlation between clinical features/laboratory variables and BALF cell classification percentages. The bar represents the range of parameter correlations, ranging from –0.4 (blue) to 0.4 (red). The greater the absolute value, the stronger the correlation. *, P<0.05; **, P<0.01; ***, P<0.001. BALF, bronchoalveolar lavage fluid; CRP, C-reactive protein; DNA, deoxyribonucleic acid; FER, ferritin; FIB, fibrinogen; IFN-γ, interferon-γ; IL-2, interleukin-2; IL-4, interleukin-4; IL-6, interleukin-6; IL-10, interleukin-10; IL-17A, interleukin-17A; LDH, lactate dehydrogenase; MP, Mycoplasma pneumoniae; PCT, procalcitonin; Tmax, temperature maximum; TNF-α, tumor necrosis factor-α; WBC, white blood cell.

BALF cell proportions vs. blood inflammatory markers

No significant correlations were found between BALF neutrophil, lymphocyte, or macrophage proportions and blood white blood cell, neutrophil, lymphocyte counts, or levels of PCT, LDH, FER and D-dimer. However, a weak positive correlation was observed between CRP level and the BALF neutrophil proportion (Spearman’s rs =0.144, P=0.03), and a weak negative correlation between CRP and BALF lymphocyte proportion (Spearman’s rs =−0.145, P=0.03), and no significant correlation between CRP level and the BALF macrophage proportion.

BALF cell proportions vs. serum cytokines

IL-2 and TNF-α levels showed negative correlations with BALF neutrophil proportion (IL-2: Spearman’s rs =−0.187, P=0.004; TNF-α: rs =−0.168, P=0.01), and positive correlations with BALF macrophage proportion (IL-2: rs =0.181, P=0.005; TNF-α: rs =0.141, P=0.03). IFN-γ showed negative correlations with BALF neutrophil proportion (Spearman’s rs =−0.132, P=0.041). No correlations were found with IL-4, IL-6, IL-10, or IL-17A.

BALF cell proportions vs. MP-DNA load

The MP-DNA load (log₁₀-transformed) in BALF showed a significant positive correlation with BALF neutrophil proportion (Spearman’s rs =0.268, P<0.001) and a significant negative correlation with BALF macrophage proportion (rs =−0.309, P<0.001). The correlation with BALF lymphocyte proportion did not reach statistical significance (rs =−0.125, P=0.055).

BALF cell proportions vs. clinical manifestations

No significant correlations were found with age, gender, peak fever temperature, length of hospital stays, or the presence of atelectasis, use of systemic corticosteroids prior to bronchoscopy. However, the presence of pleural effusion was negatively correlated with the BALF lymphocyte proportion (Spearman’s rs =−0.136, P=0.04)

Subgroup comparisons by MP-DNA load

Based on BALF MP-DNA load, patients were categorized into three groups: low-load group (10³–10⁴ copies/mL, n=20), medium-load group (10⁵–10⁶ copies/mL, n=115), and high-load group (≥10⁷ copies/mL, n=103). The Kruskal-Wallis H test, showed significant overall differences among the three groups for neutrophil, lymphocyte, and macrophage proportions (P< 0.05). Post-hoc pairwise comparisons (Mann-Whitney U with Bonferroni correction) revealed (Figure 2):

• Neutrophil proportion: the high-load group had significantly higher neutrophil proportion than the medium-load group (Z=−3.924, P<0.001), while no significant difference was observed between low- and medium-load groups;
• Lymphocyte proportion: the high-load group had significantly lower lymphocyte proportion than the medium-load group (Z=−3.081, P=0.002);
• Macrophage proportion: the high-load group had significantly lower macrophage proportion than the medium-load group (Z=−4.415, P<0.001).

Figure 2 Comparison of neutrophil, lymphocyte, and macrophage proportions in BALF among different MP-DNA load groups. NS, not significant; **, P<0.01; ***, P<0.001. BALF, bronchoalveolar lavage fluid; DNA, deoxyribonucleic acid; MP, Mycoplasma pneumoniae.

Multivariable analysis: factors independently associated with BALF neutrophil proportion

To identify factors independently associated with BALF neutrophil proportion (the primary outcome), a multiple linear regression model was constructed. As shown in Table 2, BALF MP-DNA load remained significantly and independently associated with BALF neutrophil proportion after adjusting for all covariates (B=5.435; 95% CI: 3.073–7.796; P<0.001). IFN-γ showed a borderline negative association (B=−0.070; 95% CI: −0.145 to 0.005; P=0.07), none of the other variables reached statistical significance. No significant multicollinearity was detected (all VIF <1.2).

The regression model was statistically significant (F=4.097, P<0.001) and explained 11.1% of the variance in BALF neutrophil proportion (R²=0.111, adjusted R²=0.084). The Durbin-Watson statistics were 1.936, indicating independent residuals. Visual examination of residual plots revealed no apparent heteroscedasticity or nonlinearity, indicating that the assumptions of linearity and homoscedasticity were met. These results confirm that local MP-DNA load is an independent correlate of neutrophilic predominance in BALF of children with MPP.

Sensitivity analysis

In four separate multivariable models where CRP or IFN-γ was replaced by IL-2 or TNF-α, MP-DNA load remained significantly associated with BALF neutrophil proportion (all P<0.001), with regression coefficients similar to those in the primary model. These results indicate that the exclusion of collinear variables did not materially affect the primary association.

Exploratory analysis: pleural effusion and BALF lymphocyte proportion

Given the univariable association between BALF lymphocyte proportion and pleural effusion (rs =−0.136, P=0.04), we performed a secondary exploratory analysis.

Univariable comparison (Table 3)

Patients with pleural effusion (n=38) had significantly lower BALF lymphocyte proportion compared to those without effusion (n=200) [3.00% (2.00–5.00%) vs. 4.00% (2.75–6.25%); Z=−2.093, P=0.04]. They also had significantly higher levels of CRP, PCT, LDH, ferritin, D-dimer, IL-6, and IL-10 (all P<0.05). No significant difference was observed in BALF MP-DNA load between the two groups (P>0.05).

Multivariable logistic regression

Due to the limited number of pleural effusion cases (n=38), the sample size did not meet the “10 events per variable” criterion for reliable multivariable logistic regression. Thus, only univariable results are presented for this exploratory analysis.

We attempted to construct a parsimonious model including BALF lymphocyte proportion, CRP, and age. The analysis showed that BALF lymphocyte proportion was not independently associated with pleural effusion [odds ratio (OR) =0.971; 95% CI: 0.913–1.031; P=0.34]. Of note, CRP paradoxically showed an OR <1 (OR =0.981; 95% CI: 0.970–0.993; P=0.001), suggesting model instability likely due to the small sample size.

Discussion

This large-scale, single-center, cross-sectional study systematically analyzed the characteristics and influencing factors of BALF cellular patterns in 238 children with MPP. The main findings were as follows: (I) BALF cellular patterns were characterized by marked neutrophil predominance, with a median proportion of 80.0%; (II) the BALF pattern of neutrophil predominance was significantly and independently positively correlated with local MP-DNA load, but showed no correlation with common blood inflammatory markers; (III) subgroup analysis revealed that the BALF neutrophil proportion was significantly higher in patients with an MP-DNA load ≥10⁷ copies/mL, which might be indicative of a threshold effect; and (IV) exploratory analysis showed that pleural effusion was associated with a lower BALF lymphocyte proportion in univariable analysis; however, due to the limited number of cases (n=38), multivariable regression did not confirm an independent association.

Neutrophils, the primary phagocytic, serve as the first line of innate immune defense against respiratory infections. They eliminate pathogens through mechanisms such as phagocytosis, the release of neutrophil extracellular traps (NETs), the generation of reactive oxygen species, and degranulation (18). Alveolar epithelial cells and macrophages recognize MP lipoproteins through Toll-like receptors (primarily TLR2 and TLR6) after mycoplasma infection. Following TLR2 activation, the myeloid differentiation primary response 88 (MyD88) is recruited, ultimately activating the transcription factor NF-κB signaling pathway (19). This cascade leads to the production of large amounts of neutrophil chemokines (e.g., IL-8, TNF-α), which recruit neutrophils to the site of infection (17,20). Overactivated neutrophils and excessive cytokine response result in a toxic inflammatory reaction that causes extensive macrophage necroptosis and varying degrees of functional impairment (21). However, the clearance of Mycoplasma pneumoniae (MP) relies not on neutrophils but on activated macrophages (22,23), which can phagocytose MP and present antigenic information, thereby participating in both innate and adaptive immunity. Consistent with this pathological process, the present study found that BALF cellular patterns in children with MPP were characterized by marked neutrophil predominance (median 80.0%), with markedly reduced lymphocyte (median 3.0%) and macrophage (median 13.5%) proportions, which is consistent with previous reports (5,9).

Furthermore, univariable analysis in this study revealed that BALF MP-DNA load was significantly positively correlated with the BALF neutrophil proportion. After adjusting for potential confounders including age, gender, time from symptom onset to bronchoscopy, CRP, IFN-γ, and pre-bronchoscopy systemic corticosteroid use in multivariable linear regression model, MP-DNA load remained an independent correlative factor of BALF neutrophil proportion. Of note, corticosteroid use (with a preoperative utilization rate of 81.5%) did not show an independent association in the multivariable analysis (P=0.16). A possible explanation is that corticosteroid administration was primarily based on clinical disease severity, which was closely associated with both MP-DNA load and neutrophil proportion. Therefore, glucocorticoids may serve as an intermediate variable rather than an independent confounder. The regression model explained 11.1% of the variance in BALF neutrophil proportion (R²=0.111). Although this is modest, it is clinically acceptable given that pulmonary inflammation is influenced by multiple factors (e.g., pathogen virulence, host genetic background, immune status). Importantly, the primary objective of this analysis was to identify independent correlations of neutrophilic inflammation rather than to build a predictive model. The statistical significance of the model (P<0.001) and the robust association of MP-DNA load (B=5.435, P<0.001) support the independent correlation between local pathogen burden and BALF cellular pattern of neutrophilic predominance.

Previous studies have also indicated that patients with high MP loads in BALF exhibit more significant neutrophil infiltration than children with lower loads (5), Furthermore, higher MP loads correlate with more severe localized airway inflammation, elevated inflammatory factor levels (13), more severe disease, and longer treatment durations (24,25). In the present study, further subgroup analysis revealed that only the high-load group (≥10⁷ copies/mL) had a significantly higher neutrophil proportion than the medium- and low-group, while no significant difference between the low- and medium-load groups, suggesting a potential threshold effect—namely, only when the local pathogen burden exceeds a critical cut-off value does it markedly enhance neutrophilic airway inflammation. This finding might serve as a potential quantitative reference for identifying children at high risk of excessive inflammation. Moreover, the independent association observed could imply that patients with a high pathogen load might benefit from more intensive anti-inflammatory therapy, although causal inference is limited by the cross-sectional design.

Another important finding of this study was that the BALF neutrophil proportion showed no significant correlations with common blood inflammatory markers or with most serum cytokines in univariable analysis, with only weak correlations observed for CRP, IL-2, TNF-α and IFN-γ. However, none of these associations remained statistically significant after multivariable adjustment. This dissociation between local pulmonary neutrophilic inflammation and systemic inflammatory responses was observed in our study. Similar trends have been observed in previous studies (5,14), indicating that local and systemic immune responses are relatively independent. This dissociation may be explained by two potential mechanisms. First, local inflammation is primarily triggered by direct MP stimulation of epithelial cells and alveolar macrophages, whereas systemic inflammation depends on cytokine “spillover” and secondary immune activation; these processes may be asynchronous in timing and intensity. Additionally, like the digestive tract, the respiratory tract is in direct contact with the external environment and is frequently exposed to pathogens, necessitating a swifter local immune response (26). Second, the BALF neutrophil proportion reflects local recruitment and activation, whereas blood inflammatory markers are more susceptible to various interfering factors. This compartmentalization underscores the importance of direct BALF sampling via bronchoscopy to accurately assess the pulmonary immune microenvironment—serological testing alone cannot capture local immune information.

Pleural effusion is a common complication of severe MPP in children. Multiple studies have demonstrated that inflammatory markers including CRP, LDH, PCT, and D-dimer are significantly elevated in MPP children with pleural effusion (27,28). In the present study, the incidence of pleural effusion was 16.0% (38/238). Univariable analysis showed the pleural effusion group had a significantly lower BALF lymphocyte proportion compared to the non-effusion group, as well as significantly higher levels of blood inflammatory markers, which is consistent with previous reports (9,27). However, due to the limited number of pleural effusion cases (n=38), the sample size did not meet the “10 events per variable” criterion for reliable multivariable logistic regression analysis. We attempted to construct a parsimonious model including BALF lymphocyte proportion, CRP, and age. The analysis showed that the lymphocyte proportion was not independently associated with pleural effusion, while CRP paradoxically showed a protective effect (OR =0.981), suggesting model instability likely due to the small sample size. Therefore, the observed association should be interpreted with caution. Future studies with larger sample sizes or multicenter designs are needed to further validate these findings.

The present study has several limitations that should be carefully considered when interpreting the results. (I) The cross-sectional design lacks temporal inference and cannot establish causality. All variables were measured at a single time point during the acute phase, making it impossible to determine the temporal sequence between MP-DNA load and neutrophil predominance. Theoretically, a high MP-DNA load may trigger neutrophil recruitment, whereas massive neutrophil infiltration may in turn promote MP proliferation and persistent colonization. Therefore, the “independent association” reported in this study should not be misinterpreted as a causal relationship. Furthermore, the dynamic changes in BALF cellular patterns and their relationship with treatment response and disease outcomes cannot be assessed, these patterns may merely be correlates of disease severity, sampling timing, prior therapy, or host response rather than independent clinical markers. (II) Selection bias and limited representativeness. BALF sampling is an invasive procedure and only indicated for children with specific clinical indications. Thus, the findings may not be generalizable to children with mild or outpatient MPP. (III) Potential bias in BALF cellular classification. Although we performed standardized classification according to established guidelines and implemented blinding with independent duplicate counting, morphological classification of cells remains somewhat subjective. (IV) Residual confounding. Although we adjusted for major confounders (age, gender, disease course, CRP, IFN-γ, and corticosteroid use), unmeasured confounders such as MP strain virulence, host genetic background and nutritional status may still exist. (V) Single-center design and the need for external validation. This retrospective single-center study may be influenced by local MP strain characteristics and clinical practice patterns. Future multicenter prospective cohort studies are needed for external validation.

Future research should focus on: (I) longitudinal studies to clarify the dynamic changes in BALF cellular patterns; (II) mechanistic investigations into the specific pathways of neutrophil activation under high MP load; and (III) verify the generalizability of the present findings through multicenter prospective studies; (IV) the exploration of personalized treatment strategies based on local inflammatory phenotypes.

Conclusions

Overall, this study found that BALF cellular patterns are characterized by neutrophil predominance, which is independently associated with local MP-DNA load but not with most common blood inflammatory markers. Accordingly, attention should be paid to identifying distinct inflammatory phenotypes based on BALF cellular patterns, particularly in patients with a high MP-DNA load. These findings may provide further insights into the local pulmonary immune microenvironment in children with MPP and may offer new approaches for clinical assessment and potentially treatment decision-making. Given the cross-sectional design of this study, causal relationships cannot be established, and the above findings require validation in prospective cohort studies.

Acknowledgments

The authors would like to thank the patients and participating investigators and staff associated with the clinical studies discussed here.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://tp.amegroups.com/article/view/10.21037/tp-2026-0194/rc

Data Sharing Statement: Available at https://tp.amegroups.com/article/view/10.21037/tp-2026-0194/dss

Peer Review File: Available at https://tp.amegroups.com/article/view/10.21037/tp-2026-0194/prf

Funding: This work was supported by the Special Project of Key Discipline from Tianjin Health Commission (No. TJWJ2022XK038), the Tianjin Key Medical Discipline Construction Project (No. TJYXZDXK-3-016B), and the Science and Technology Projects of Xizang Autonomous Region, China (No. XZ202502ZY0008).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tp.amegroups.com/article/view/10.21037/tp-2026-0194/coif). The authors have no conflicts of interest to declare.

Ethical Statement:The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments, and received approval from the Institutional Ethics Committee of Tianjin Children’s Hospital (approval No. 2022-LXKY-004). It is a retrospective analysis, in which all patient information was reported in an anonymous manner. In accordance with the guidelines of the Institutional Ethics Committee of Tianjin Children’s Hospital, the requirement for informed consent was waived due to the retrospective design of this study.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Diaz MH, Hersh AL, Olson J, et al. Mycoplasma pneumoniae Infections in Hospitalized Children - United States, 2018-2024. MMWR Morb Mortal Wkly Rep 2025;74:394-400. [Crossref] [PubMed]
2. Waites KB, Xiao L, Liu Y, et al. Mycoplasma pneumoniae from the Respiratory Tract and Beyond. Clin Microbiol Rev 2017;30:747-809. [Crossref] [PubMed]
3. Meyer Sauteur PM, Beeton ML. European Society of Clinical Microbiology and Infectious Diseases (ESCMID) Study Group for Mycoplasma and Chlamydia Infections (ESGMAC), and the ESGMAC Mycoplasma pneumoniae Surveillance (MAPS) study group. Mycoplasma pneumoniae: delayed re-emergence after COVID-19 pandemic restrictions. Lancet Microbe 2024;5:e100-1. [Crossref] [PubMed]
4. Meyer Sauteur PM. Childhood community-acquired pneumonia. Eur J Pediatr 2024;183:1129-36. [Crossref] [PubMed]
5. Deng F, Cao H, Liang X, et al. Analysis of cytokine levels, cytological findings, and MP-DNA level in bronchoalveolar lavage fluid of children with Mycoplasma pneumoniae pneumonia. Immun Inflamm Dis 2023;11:e849. [Crossref] [PubMed]
6. Expert Group on pediatric respiratory endoscopy diagnosis and treatment, HHRDC. Guideline of pediatric flexible bronchoscopy in China (2018 version). Chin J Appl Clin Pediatr 2018;33:983-9.
7. Meyer KC, Raghu G, Baughman RP, et al. An official American Thoracic Society clinical practice guideline: the clinical utility of bronchoalveolar lavage cellular analysis in interstitial lung disease. Am J Respir Crit Care Med 2012;185:1004-14. [Crossref] [PubMed]
8. Riedler J, Grigg J, Stone C, et al. Bronchoalveolar lavage cellularity in healthy children. Am J Respir Crit Care Med 1995;152:163-8. [Crossref] [PubMed]
9. Ma K, Liu X, Wu XT, et al. The correlation between the level of cellular classification in bronchoalveolar lavage fluid of children with severe Mycoplasma pneumoniae and clinical characteristics. Eur J Pediatr 2024;183:4351-62. Erratum in: Eur J Pediatr 2024;183:4595.
10. Cai C, Hu PP, Lu M, et al. Correlation of cytokine and cytology levels in bronchoalveolar alveolar lavage fluid of children with severe Mycoplasma pneumoniae pneumonia. Chinese Journal of Applied Clinical Pediatrics 2020;35:1421-4.
11. Liu J, Zhao F, Lu J, et al. High Mycoplasma pneumoniae loads and persistent long-term Mycoplasma pneumoniae DNA in lower airway associated with severity of pediatric Mycoplasma pneumoniae pneumonia. BMC Infect Dis 2019;19:1045. [Crossref] [PubMed]
12. Wang W, Wang L, Yin Z, et al. Correlation of DNA load, genotyping, and clinical phenotype of Mycoplasma pneumoniae infection in children. Front Pediatr 2024;12:1369431. [Crossref] [PubMed]
13. Zhong H, Zeng Z, Gu H, et al. Effect of Macrolide Resistance and Mycoplasma pneumoniae DNA Load in Bronchoalveolar Lavage Fluid on Immune and Inflammatory Responses in Children with Mycoplasma pneumoniae Pneumonia. Indian Pediatr 2025;62:893-8. [Crossref] [PubMed]
14. Sun Y, Tong L, Lin M, et al. Age-Specific Cytokine Profiling in Children with Mycoplasma Pneumoniae Infections in Post-COVID-19 Era: A Retrospective Study. J Inflamm Res 2025;18:17731-46. [Crossref] [PubMed]
15. Expert Committee of J.EDU Medical Cell Platform Soliciting Contributions. Consensus of Chinese Experts on Morphological Examination of Bronchoalveolar Lavage Fluid Cells(2023). Journal of Modern Laboratory Medicine 2023;38:11-6.
16. National Health Commission of the People’s Republic of China. Guidelines for the diagnosis and treatment of Mycoplasma pneumoniae pneumonia in children(2023 edition). Inter J Epidemiol Infect Dis 2023;50:79-85.
17. Gu H, Zhu Y, Zhou Y, et al. LncRNA MALAT1 Affects Mycoplasma pneumoniae Pneumonia via NF-κB Regulation. Front Cell Dev Biol 2020;8:563693. [Crossref] [PubMed]
18. Mankan AK, Czajka-Francuz P, Prendes M, et al. Intracellular DNA sensing by neutrophils and amplification of the innate immune response. Front Immunol 2023;14:1208137. [Crossref] [PubMed]
19. Chen M, Deng H, Zhao Y, et al. Toll-Like Receptor 2 Modulates Pulmonary Inflammation and TNF-α Release Mediated by Mycoplasma pneumoniae. Front Cell Infect Microbiol 2022;12:824027. [Crossref] [PubMed]
20. Chen Z, Shao X, Dou X, et al. Role of the Mycoplasma pneumoniae/Interleukin-8/Neutrophil Axis in the Pathogenesis of Pneumonia. PLoS One 2016;11:e0146377. [Crossref] [PubMed]
21. Mei X, Wang J, Zhang C, et al. Apigenin suppresses mycoplasma-induced alveolar macrophages necroptosis via enhancing the methylation of TNF-α promoter by PPARγ-Uhrf1 axis. Phytomedicine 2023;108:154504. [Crossref] [PubMed]
22. Song Z, Han C, Luo G, et al. Yinqin Qingfei granules alleviate Mycoplasma pneumoniae pneumonia via inhibiting NLRP3 inflammasome-mediated macrophage pyroptosis. Front Pharmacol 2024;15:1437475. [Crossref] [PubMed]
23. Lai JF, Zindl CL, Duffy LB, et al. Critical role of macrophages and their activation via MyD88-NFκB signaling in lung innate immunity to Mycoplasma pneumoniae. PLoS One 2010;5:e14417. [Crossref] [PubMed]
24. Li W, Deng F. Clinical significance of measuring MP-DNA, C-reactive protein, and inflammatory cytokines in children with mycoplasma pneumoniae pneumonia. Am J Transl Res 2024;16:577-83. [Crossref] [PubMed]
25. Fang KN, Wang J, Ni JW. Correlation between Mycoplasma pneumoniae DNA replication level and disease severity in children with severe Mycoplasma pneumoniae pneumonia. Zhongguo Dang Dai Er Ke Za Zhi 2019;21:876-80. [Crossref] [PubMed]
26. Dzidic M, Boix-Amorós A, Selma-Royo M, et al. Gut Microbiota and Mucosal Immunity in the Neonate. Med Sci (Basel) 2018;6:56. [Crossref] [PubMed]
27. Li S, Zhou J, He J, et al. Clinical profiles of Mycoplasma pneumoniae pneumonia in children with different pleural effusion patterns: a retrospective study. BMC Infect Dis 2024;24:919. [Crossref] [PubMed]
28. Ji L, Lin L, Ye J, et al. Clinical Characteristics and Predictive Factors Analysis of Mycoplasma Pneumoniae Pneumonia Complicated with Pleural Effusion in Children. Iran J Allergy Asthma Immunol 2025;24:451-61.