Sly Saint
Senior Member (Voting Rights)
Abstract
Background: Acute COVID-19 alters the airway microbiota. Whether these changes persist post-COVID-19 and contribute to Long COVID (LC) is unknown. We aimed to describe the microbiota based on LC severity and subtype; and identify links with inflammation to find potential therapeutic targets.Methods: Adults were prospectively recruited to the longitudinal UK Post-hospitalisation COVID-19 study (PHOSP-COVID). Patients were grouped by previously described ‘severity clusters’ (of ongoing health impairments) and symptom subgroups. Oral and lower airway microbiota (at six-months post-discharge) were characterised through 16S rRNA sequencing, and microbiota diversity was assessed. Co-occurrence network analysis identified influential taxa. Associations between influential taxa and inflammatory proteins from plasma were assessed through Spearman’s rank correlations.
Findings: 771 participants had their microbiota assessed. Of participants with oral rinses and ‘severity cluster’ assignment (mean age: 57·5), 278/722 (38·5%) were women. The very severe cluster was associated with microbiota dysbiosis post-COVID-19. Persistent neurocognitive symptoms were associated with oral microbiota dysbiosis, while gastrointestinal symptoms were associated with lower airway microbiota dysbiosis. Microbiota influential taxa shifted to a higher prevalence of lactic acid bacteria (LAB) associated with worse health outcomes. LAB abundance was associated with elevated inflammatory proteins.
Interpretation: The association of microbial dysbiosis with neurocognitive and gastrointestinal symptoms suggests potential oral–brain and gut–lung crosstalk in LC. Shifts in influential microbiota taxa, and their link to inflammation based on worse health outcomes, suggest a potential LAB role change which may be associated with the host immune response.