Key points are not available for this paper at this time.
Introduction Pseudomonas aeruginosa (Pae) is a major cause of bacterial pneumonia in older adults, yet the composition of the lower respiratory tract (LRT) microbiota in patients with P. aeruginosa pneumonia (PAP), and its potential prognostic relevance, remain poorly defined. Methods Here, we performed 16S rRNA gene sequencing of bronchoalveolar lavage fluid collected from older adults with PAP at hospital admission and discharge, yielding 42 paired samples. We characterized longitudinal microbiota dynamics, identified candidate microbial biomarkers using LEfSe and random forest analyses, inferred functional profiles with Tax4Fun2, and compared microbiota-associated changes following meropenem-based therapy (n = 12) or levofloxacin-based therapy (n = 9). Results PAP was associated with marked shifts in LRT microbial community structure, with post-treatment samples showing reduced Pseudomonas abundance and increased relative abundances of Rothia, Streptococcus and Porphyromonas ( p 0.05). Random forest modeling identified P. aeruginosa, Streptococcus pneumoniae , and Prevotella melaninogenica as potential diagnostic biomarkers (AUC = 0.93). Functional prediction further suggested a reduction in biofilm-formation-related pathways after treatment ( p 0.01). Notably, Pseudomonas abundance correlated positively with white blood cell count and C-reactive protein, but inversely with lymphocyte and platelet counts, supporting the potential value of these clinical indices in prognostic assessment. In this cohort, meropenem-based therapy, compared with levofloxacin -based therapy, was associated with enrichment of several potentially pathogenic anaerobic genera, including Porphyromonas, Campylobacter, Neisseria and Prevotella . Discussion Collectively, these findings define the LRT microbiota landscape in elderly patients with PAP, nominate candidate microbial biomarkers, and indicate that antibiotic regimen is associated with distinct post-treatment microbial community structures, with potential implications for therapeutic decision-making.
Jiang et al. (Wed,) studied this question.