Integrating lung microbiome, amino acid metabolism, and host immune response in elderly patients for severe lower respiratory Infections diagnosis: a multi-omics study

BACKGROUND: Lower respiratory infections (LRIs) cause significant morbidity and mortality in elderly individuals, but the mechanisms driving severe deterioration remain unclear. METHODS: This prospective study enrolled 105 patients aged ≥60 with suspected LRIs between October 2024 and April 2025. Bronchoalveolar lavage fluid (BALF) was analyzed using 16S rRNA sequencing, metagenomics, untargeted metabolomics, and cytokine profiling. Multi-omics data were integrated into a tripartite network, and severity-associated signatures were identified via PLS-DA, logistic regression, and ROC analysis. RESULTS: The cohort included 40 severe (sLRIs) and 65 mild (mLRIs) cases. sLRIs exhibited reduced microbial diversity, shifting from commensal genera to opportunistic pathogens (Klebsiella, Corynebacterium, Elizabethkingia), with Klebsiella pneumoniae as a major bacterial hub. Metabolomics revealed 180 differential metabolites. Phenylalanine and beta-Alanine metabolism emerged as key severity-associated pathways. sLRIs showed accumulation of pro-inflammatory metabolites L-phenylalanine and phenylpyruvic acid. L-3-phenyllactic acid (PLA) served as the central metabolic hub. Cytokine profiling revealed local hyperinflammation (elevated IL-1β, IL-6, IL-8, TNF-α, IFN-γ), with IL-6 as central hubs. Multivariate analysis identified PLA and IL-8 as independently associated with severe status. Combined metabolic-immune signatures achieved high diagnostic accuracy (AUC: 0.858-0.882). CONCLUSIONS: sLRIs in elderly patients are characterized by microbial dysbiosis, opportunistic pathogen enrichment, and remodeled Phenylalanine and beta-Alanine metabolism that correlates with hyperinflammation. BALF PLA and IL-8 represent promising metabolic-immune biomarkers for severity stratification.