A retrospective cohort study evaluating hospitalized pediatric patients with Mycoplasma pneumoniae pneumonia demonstrated that targeted next-generation sequencing of bronchoalveolar lavage fluid reveals distinct clinical phenotypes based on bacterial DNA load, co-infection profiles, and antimicrobial resistance mutations. Investigators analyzing seven hundred ninety-one pediatric cases admitted to the Maternal and Child Health Hospital of Guangxi Zhuang Autonomous Region between August 2023 and July 2024 established that quantitative bacterial burden and co-detected respiratory pathogens define fundamentally divergent manifestations of illness. While high pathogen burdens traditionally align with expectations of aggressive disease, the sequencing data revealed that high Mycoplasma pneumoniae DNA loads correlated with persistent fever and localized lung consolidation typical of classic monoinfection. In sharp contrast, lower Mycoplasma pneumoniae loads served as an informative indicator for a complex, co-infection-driven syndrome that disproportionately affected younger children presenting with heightened respiratory distress, wheezing, dyspnea, elevated systemic inflammatory markers, and higher rates of severe pneumonia.
Mycoplasma pneumoniae represents a principal etiological agent of community-acquired pneumonia in pediatric medicine, responsible for ten to forty percent of overall cases in most global settings and reaching thirty to fifty percent during recognized epidemics. Although infections frequently present as self-limiting lower respiratory tract disease, clinicians encounter marked clinical heterogeneity spanning mild bronchitis to life-threatening necrotizing pneumonia or extrapulmonary systemic manifestations. Conventional diagnostic pathways rely on serological assays, which frequently suffer from delayed seroconversion and non-specific cross-reactivity, or quantitative polymerase chain reaction tests that typically lack comprehensive co-pathogen panels and antimicrobial resistance mutation profiling. Furthermore, macrolide resistance rates among Mycoplasma pneumoniae exceed seventy to ninety percent across East Asia, contrasting with rates below ten to fifteen percent in North American and European cohorts, which poses distinct management challenges when clinical workflows fail to resolve co-pathogens and resistance patterns simultaneously.
To overcome diagnostic constraints and explain observed clinical heterogeneity through comprehensive microbial profiling, the study team utilized bronchoalveolar lavage fluid collected during diagnostic bronchoscopy performed within seventy-two hours of hospital admission. Sampling lower respiratory secretions directly from the alveolar space minimizes upper respiratory tract contamination from the oropharyngeal microbiome, providing a representative microbiological evaluation of lower airway pathology. Nucleic acid extracted from standardized bronchoalveolar lavage samples underwent targeted next-generation sequencing using a specialized multiplex polymerase chain reaction library panel designed to identify one hundred ninety-eight respiratory pathogens, including eighty bacteria, seventy-nine viruses, thirty-two fungi, and seven additional microbes. The platform simultaneously interrogated over three hundred antimicrobial resistance gene sites, allowing researchers to detect specific point mutations in domain five of the twenty-three S ribosomal RNA gene that confer macrolide resistance, specifically the A2063G and A2064G transition mutations, within a turnaround window of twelve hours.
The study protocol applied predefined microbiological and clinical criteria to ensure reliable patient stratification and quantitative precision across the cohort. Pediatric patients aged one month to sixteen years clinically diagnosed with Mycoplasma pneumoniae lower respiratory infection according to consensus diagnostic criteria were included when sequencing returned fifty or more pathogen reads and an estimated DNA concentration of at least one thousand copies per milliliter, calibrated against spike-in internal controls. Children with underlying primary immunodeficiencies, chronic lung disorders, asthma, structural heart disease, or incomplete hospital records were excluded from the analysis. Cohort stratification categorized cases into high Mycoplasma pneumoniae load, defined as DNA concentrations exceeding one million copies per milliliter, and low Mycoplasma pneumoniae load, spanning between one thousand and one million copies per milliliter. Cases with fewer than one thousand copies per milliliter were excluded due to established limits of quantitative precision.
Co-infection was formally identified when bronchoalveolar lavage sequencing revealed one or more additional co-pathogens reaching or exceeding the threshold of one thousand copies per milliliter, whereas monoinfection required the absence of any secondary pathogen at or above that quantitative boundary. Across the entire cohort of seven hundred ninety-one children, targeted sequencing classified three hundred nine patients as having monoinfection and four hundred eighty-two patients as experiencing co-infection, yielding an overall co-infection prevalence of sixty point nine percent. When stratified by bacterial burden, six hundred fifty-two children exhibited high Mycoplasma pneumoniae DNA loads, while one hundred thirty-nine fell into the low-load category. Antimicrobial resistance testing identified macrolide-resistant Mycoplasma pneumoniae strains in four hundred sixty-one children, representing fifty-eight point three percent of the cohort, while three hundred thirty harbored macrolide-susceptible organisms, establishing a substantial foundation for independent multidimensional analyses across clinical, demographic, and laboratory parameters.
Direct comparisons between children with monoinfection and those harboring co-infections highlighted profound clinical distinctions. Patients with confirmed co-infections were significantly younger than their monoinfected counterparts, exhibiting a median age of forty-eight months compared to seventy-two months among those with monoinfection. Hospitalization was significantly prolonged in the co-infection group, showing a median duration of seven days versus six days in the monoinfection cohort. Symptomatically, co-infected pediatric patients displayed a higher incidence of wheezing, occurring in thirteen point three percent of cases compared to seven point eight percent among monoinfections, but experienced lower frequencies of general fever and persistent fever lasting five or more days. Furthermore, systemic laboratory markers of inflammation, including overall white blood cell counts, circulating platelet counts, and serum creatine kinase-MB levels, were significantly elevated among co-infected patients, while their peripheral neutrophil percentages were comparatively reduced.
Evaluating patients according to targeted sequencing DNA load exposed an unexpected epidemiological pattern regarding disease severity. Children categorized into the low-load group were notably younger, with a median age of thirty-six months compared to sixty-nine months in the high-load group, and demonstrated an overwhelming ninety-seven point eight percent rate of co-infection, compared to fifty-three point one percent among high-load patients. Clinically, the low Mycoplasma pneumoniae load group exhibited a greater incidence of wheezing at twenty-three point seven percent versus eight point four percent, and dyspnea at fourteen point four percent versus five point four percent. Severe pneumonia was documented in twenty-one point six percent of low-load patients compared to thirteen point three percent of children with high bacterial burdens, accompanied by longer hospital stays and elevated levels of white blood cells, platelets, alanine aminotransferase, and creatine kinase-MB, despite reduced rates of persistent fever.
Radiographic assessments and multivariable regression modeling confirmed that bacterial load and co-infection status govern divergent pathological mechanisms. While lobar consolidation represented the predominant radiographic feature across the cohort, it was significantly more frequent among high-load infections, occurring in eighty-three point one percent of cases compared to sixty-nine point eight percent in low-load cases, which conversely displayed higher frequencies of pleural effusion, inflammatory exudates, and increased lung markings. Multivariable logistic regression adjusting for age and other clinical confounders established that a high Mycoplasma pneumoniae DNA load independently doubled the odds of developing persistent fever, yielding an odds ratio of two point thirteen, while simultaneously conferring an independent protective effect against severe pneumonia, with an odds ratio of zero point fifty-six. Linear regression models showed that a high bacterial load independently correlated with a hospital stay shortened by one point twenty-nine days, as well as significantly lower overall white blood cell and platelet counts.
The pathophysiological divergence uncovered by the sequencing data indicates two separate mechanisms governing pediatric respiratory deterioration. In high-load presentations, dense bacterial proliferation acts as a direct driver of robust inflammatory responses, inducing continuous pyrogen release, prominent febrile episodes, and localized lobar tissue consolidation characteristic of primary bacterial invasion. Conversely, in low-load presentations, severe illness is not driven by ongoing Mycoplasma proliferation but by a multifaceted co-infection syndrome in younger pediatric hosts possessing immature immune systems. In these vulnerable children, secondary bacterial and viral co-pathogens, coupled with post-inflammatory dysregulated immune cascades, precipitate widespread airway reactivity, wheezing, and pronounced systemic inflammation. Subgroup evaluations restricted exclusively to the four hundred eighty-two co-infected patients corroborated this dynamic, demonstrating that low Mycoplasma loads maintained a higher incidence of severe pneumonia at twenty-one point three percent versus thirteen point three percent and significantly higher circulating leukocyte counts.
Analysis of antimicrobial susceptibility demonstrated that the detection of twenty-three S ribosomal RNA point mutations directly governed therapeutic trajectories without modifying baseline clinical presentations. However, macrolide resistance represented the single strongest independent predictor of therapeutic failure, conferring a more than three-fold increase in the odds of switching to second-line non-macrolide antimicrobials, with an odds ratio of three point thirteen. Children harboring resistant pathogens required a longer median course of macrolide therapy of seven days versus five days, with second-line antimicrobial switches occurring in sixteen point five percent of resistant cases compared to only five point eight percent of susceptible infections, confirming the diagnostic value of rapid genotypic resistance detection.
To evaluate whether methodological thresholds altered the observed clinical relationships, the researchers conducted sensitivity analyses adjusting the microbiological cut-off defining co-infection. Testing a stricter definition where Mycoplasma concentrations had to fall below ten times that of any co-detected organism reduced the co-infection rate to thirty-three point eight percent, while an inclusive definition incorporating any detected non-Mycoplasma organism increased the rate to seventy-four point three percent. Across both alternative definitions, multivariable models confirmed the stability of primary outcomes: the association between co-infection and severe pneumonia remained statistically non-significant, younger age remained consistently linked to co-infection status, and the protective effect of a high bacterial load against severe pneumonia remained stable, confirming that the analytical conclusions were not artifacts of the chosen quantitative cut-off.
The authors outlined important structural limitations inherent to their retrospective single-center design, emphasizing that bronchoalveolar lavage was performed exclusively when clinically indicated for hospitalized patients exhibiting severe symptoms, treatment failure, or suspected mixed infections. Consequently, the findings cannot be generalized to mild outpatient presentations or assumed to mirror microbial dynamics in upper respiratory specimens such as throat swabs or sputum. Additionally, the quantitative dynamic range of targeted sequencing panels may lack precision at extreme ends of pathogen concentration, and distinguishing colonization from active infection remains an interpretive challenge in clinical microbiology. Nonetheless, the investigation provides evidence that integrating quantitative DNA load, comprehensive co-infection profiling, and genotypic resistance testing from targeted sequencing establishes a practical framework to stratify pediatric pneumonia, anticipate clinical trajectories, and inform timely therapeutic decisions.
Subject of Research: Multidimensional targeted next-generation sequencing profiling of pediatric Mycoplasma pneumoniae pneumonia in bronchoalveolar lavage fluid
Article Title: Associations of Mycoplasma pneumoniae load, co-infections, and macrolide resistance with clinical-laboratory profiles in hospitalized pediatric pneumonia: a targeted next-generation sequencing study of bronchoalveolar lavage fluid
Article References: Liang, X., Wei, R., Liang, D., Huang, W., Huang, R., Lu, S., Lu, Y., Ma, H., Shi, Q., Li, D., Chen, D., Qin, B., Qi, X., Tang, Y., Mo, W., & Wei, Z. (2026). Associations of Mycoplasma pneumoniae load, co-infections, and macrolide resistance with clinical-laboratory profiles in hospitalized pediatric pneumonia: a targeted next-generation sequencing study of bronchoalveolar lavage fluid. Annals of Clinical Microbiology and Antimicrobials, 25(1), Article 30. https://doi.org/10.1186/s12941-026-00870-7
Image Credits: AI Generated
DOI: 10.1186/s12941-026-00870-7
Keywords: Mycoplasma pneumoniae, pediatric pneumonia, targeted next-generation sequencing, bronchoalveolar lavage fluid, macrolide resistance, co-infection, pathogen DNA load, clinical phenotypes, Associations, Mycoplasma, pneumoniae, load
News Source: Harold Sullivan. (November 9, 2026). Sequencing Pediatric Lung Fluid Links Microbial Load to Distinct Pneumonia Phenotypes. Scienmag.



