In a population-based cohort of 5,084 patients, baseline gut microbiota composition was associated with the risk of future COVID-19 hospitalization or death, with higher relative abundances of butyrate-producing bacteria linked to a lower risk, according to lead study author Robert F.J. Kullberg, MD, of Amsterdam University Medical Center, and colleagues.
Fecal samples collected between May 2011 and November 2015 in the Healthy Life in an Urban Setting (HELIUS) study were analyzed using 16S rRNA gene sequencing (V4 region) and ANOVA (Bray-Curtis) to assess overall community differences. During the first 2 years of the COVID-19 pandemic, 73 patients developed severe disease, defined as hospitalization or mortality. Baseline microbiota composition differed between those who later developed severe COVID-19 and those who did not. Cox proportional-hazards models then quantified associations between individual genera or features and severe outcomes.
Greater abundances of the anaerobic gut symbiont Enterocloster were linked to an increased risk, while Oscillospirales species—many of which produce butyrate—were linked to reduced risk of developing severe COVID-19. An abundance of butyrate-producing bacteria showed a protective association both when modeled as a continuous variable and when comparing the highest vs the lowest tertile.
These associations remained statistically significant after adjustment for age, sex, ethnicity, body mass index, time from inclusion to January 1, 2020, and comorbidities, including hypertension, diabetes, cancer, and cardiovascular, pulmonary, and gastrointestinal disease. Findings were consistent in analyses controlling for antibiotic use, vaccination status, and time frame. Three sensitivity analyses yielded similar results: adding recent antibiotic exposure (within 3 months of sampling) as a confounder, restricting outcomes to 2020, and censoring at the first SARS-CoV-2 vaccination.
Among 934 patients with confirmed SARS-CoV-2 infection during follow-up, microbiota composition differed between those with and without severe COVID-19, and butyrate-producer abundance was lower in severe cases.
Patients who developed severe disease were older (median 57 vs 52 years), had higher body mass index (median 31 vs 27), and more comorbidities than those without severe disease. Common diseases included hypertension (49% vs 26%), diabetes (31% vs 11%), cardiovascular disease (23% vs 10%), and pulmonary disease (26% vs 10%). Only 3% of the severe group were of Dutch ethnicity vs 30% among nonsevere cases; 37% were South-Asian Surinamese vs 16%, 15% were Turkish vs 9%,3% were Ghanaian or identified as other vs 10%, while relatively 12% of both severe and nonsevere groups were Moroccan.
Preclinical studies suggest that butyrate enhances epithelial barrier integrity and modulates systemic and pulmonary immune responses—for example, via G protein-coupled receptor signaling and histone deacetylase inhibition—supporting more effective CD8+ T cell antiviral activity and reduced airway neutrophilia. Murine studies showed Faecalibacterium prausnitzii and other Oscillospiraceae improved pneumonia outcomes.
The researchers emphasize that while infection acquisition is largely exposure-driven, baseline gut microbiota may relate primarily to severity given infection, not to infection risk per se.
"Combined with the preclinical evidence, our findings suggest that microbiome-directed therapies aimed at increasing intestinal abundances of butyrate-producers might limit the risk of future severe viral pneumonia," the researchers wrote. "Butyrate-producing gut microbiota may represent a novel therapeutic target for the prevention of viral pneumonia."
A key limitation is the long interval between stool collection and pandemic outcomes; while adult microbiota can be relatively stable over years, changes over time could attenuate associations. Also, the relative abundance of butyrate-producers is a proxy measure—actual butyrate production varies by strain and environmental context, which were not directly measured.
Funding included an Amsterdam University Medical Center PhD scholarship, a Niels Stensen Fellowship, the Dutch Heart Foundation, ZonMw/NWO, the European Union FP-7 program, and the European Fund for the Integration of non-EU immigrants. One author reported being a co-founder and member of the Scientific Advisory Board of Caelus Pharmaceuticals and Advanced Microbiota Therapeutics. Another author reported grants from ZonMw/NWO and the EU outside the submitted work.