Investigating the lasting effects of SARS-CoV-2 infection and the lung microbiota: No persistent microbial alterations in recovered COVID-19 patients with persistent radiological or respiratory abnormalities.

SARS-CoV-2 was a global pandemic where infected individuals experienced mild or severe disease. Unfortunately, some patients who experienced severe disease also had lasting abnormalities. The lung microbiome of thirty-eight adult COVID-19 patients with persistent respiratory symptoms and/or radiological abnormalities were analysed. The aim was to investigate if the lasting radiological abnormalities reported in this cohort were associated with an altered airway. Thirty-six bronchoalveolar lavage fluid samples from patients underwent 16S rRNA gene amplicon sequencing and compared to twenty-eight non-fibrotic control samples from a previously published study. COVID-19 patients had statistically significant greater number of genera but at uneven abundances, though not statistically significant, compared to non-fibrotic controls. Permanova suggested that COVID-19 can influence the lung microbiome composition after accounting for multivariate dispersion. Further analysis showed differences in the relative abundances of Actinomyces, Neisseria, Haemophilus, Rothia and Gemella. Indicator species analysis showed that a COVID-19 lung microbiome profile could be driven in part by differences in Fusobacterium, Actinomyces, Catonella, Oribacterium and Mycobacterium. Associations with clinical parameters were lacking apart from CT lung opacification which revealed a significant negative association with number of genera. Differential abundance analysis with MaAsLin2 pointed towards Porphyromonas as a potential explaining genus though this was not significant after post-hoc corrections. DESeq2 revealed enriched oral taxa in the BAL samples suggesting potential oral-translocation reflective of a disease state. Our findings suggest that individuals with persistent radiological abnormalities following SARS-CoV-2 infection have experience subtle shifts in their microbiome profile, but these are not strongly associated with clinical phenotypes and therefore unlikely of significance.