BALF cellular composition in pediatric LRTI: from pathogen-specific patterns to clinical correlates

Objective Lower respiratory tract infections (LRTI) remain a leading cause of pediatric hospitalization, yet distinguishing infection from non-infectious airway conditions and identifying the causative pathogen remains a clinical challenge. While bronchoscopy with bronchoalveolar lavage provides direct lower airway sampling, the clinical value of routine bronchoalveolar lavage fluid (BALF) cellular analysis beyond its conventional diagnostic use remains poorly defined. Methods This retrospective study included hospitalized children with LRTI who underwent bronchoscopy with BALF cellular analysis and targeted next-generation sequencing, as well as children with foreign-body (FB) aspiration as a noninfectious reference group. First, BALF cellular profiles were compared across three pathogen-defined LRTI subgroups: typical bacterial, viral, and Mycoplasma pneumoniae (MP). Second, multivariable linear regression analyses were performed to identify clinical, laboratory, radiologic, and bronchoscopic correlates of BALF cellular composition. Finally, secondary logistic regression analyses were performed to assess the potential clinical utility of BALF cellular composition for identifying LRTI using FB as a reference group and for discriminating typical bacterial LRTI from nontypical bacterial LRTI. Results Among 488 children with LRTI, BALF cellular composition differed across pathogen-defined groups, with MP infection showing the most pronounced neutrophil-skewed profile. Higher pathogen burden was independently associated with a more neutrophil-skewed BALF profile, whereas mucus plugs were positively associated with ln(Neu/Mph) in pooled interaction models but showed heterogeneous directions in pathogen-specific models. Exploratory analyses suggested limited species-level variation within the bacterial and viral categories. In secondary analyses, BALF cellular composition showed good discrimination between LRTI and noninfectious group, but provided no significant incremental discrimination for typical bacterial versus nontypical bacterial LRTI when added to clinical variables. Conclusion BALF cellular composition showed distinct pathogen-associated patterns in pediatric LRTIs, with MP infection exhibiting the most pronounced neutrophil-skewed profile. Higher pathogen burden was associated with a more neutrophil-skewed profile, while mucus-plug associations were sensitive to model specification. BALF cellular composition discriminated LRTI from noninfectious airway conditions but added no significant discrimination for typical bacterial versus nontypical bacterial LRTI beyond clinical variables.