Skip to content

Supplemental Oxygen Alters the Airway Microbiome in Cystic Fibrosis

Key findings

  • This study examined whether supplemental oxygen might alter the airway microbiome in people living with cystic fibrosis (CF)
  • 11 patients each provided an expectorated sputum sample that underwent culture under 21%, 50%, and 100% oxygen
  • Hyperoxia decreased the absolute abundance of some facultative anaerobes, including Rothia mucilaginosa, Streptococcus species and Actinomyces oris, and surprisingly some aerobes such as Aspergillus fumigatus
  • There was no significant effect of hyperoxia on the abundance of typical CF pathogens such as Pseudomonas aeruginosa and Staphylococcus aureus
  • Still, it remains to be studied whether the hyperoxia that depletes less oxygen-tolerant species also eventually accelerates the growth of pathogens, thereby increasing the subsequent risk of CF exacerbations or disease progression

About 10% of people with cystic fibrosis (CF) are treated with supplemental oxygen, which is historically considered a benign medical intervention. However, observational studies and clinical trials have shown that targeting high patient oxygen levels is associated with an increased risk of serious bacterial infections, including bacteremia and ventilator-associated pneumonia, and higher mortality in patients with critical illness.

Proposed mechanisms for the contributions of oxygen to disease progression have focused on the generation of reactive oxygen species and injury to host pulmonary cells.

Now, however, Massachusetts General Hospital researchers have provided evidence that supplemental oxygen may lead to dysbiosis in the airway microbiome of people living with CF. Peggy S. Lai, MD, MPH, a physician–scientist in the Division of Pulmonary and Critical Care Medicine at Mass General, Jacob Vieira, BS, a data analyst in the Division, and colleagues published their findings in mSystems.


The 11 study participants were prospectively recruited through the Adult Cystic Fibrosis Center at Mass General between November 2019 and March 2020. Their average age was 29, six were on CF transmembrane conductance regulator modulators, six received antimicrobials, and one was on supplemental oxygen.

During routine outpatient clinical visits, each patient provided a spontaneously expectorated sputum sample that underwent culture under 21%, 50%, and 100% oxygen.


Hyperoxia significantly reduced absolute microbial load and the alpha diversity of species. However, its effects varied, as it:

  • Decreased the absolute abundance of some facultative anaerobes, such as Rothia mucilaginosa, Streptococcus species, and Actinomyces oris
  • It had no significant effect on the abundance of typical CF pathogens such as Pseudomonas aeruginosa and Staphylococcus aureus

In fact, S. aureus appeared to be the most oxygen-tolerant microbe studied. The long-term clinical implications are unclear—colonization by S. aureus later in CF disease has not been associated with altered lung function. However, hyperoxia may eventually enable accelerated growth of S. aureus to fill in for depleted species, increasing the risk of pulmonary infection.

Guidance for the Clinic

The decrease in alpha diversity of the airway microbiome due to supplemental oxygen might be analogous to the effect of administering antimicrobials. After prescribing supplemental oxygen, clinicians should consider monitoring sputum microbial communities for the emergence of harmful pathogens. Oxygen is prescribed to a quarter of all hospitalized patients and these findings may have broader relevance beyond CF. Whether oxygen supplementation is a risk factor for hospital-acquired or ventilator-associated pneumonia deserves further study.

Learn about the Adult Cystic Fibrosis Center

Refer a patient to the Division of Pulmonary and Critical Care Medicine


Research at Massachusetts General Hospital identifies a link between endothelial S1PR1 and the development of pulmonary fibrosis.


Molly Wolf, MD, and Peggy S. Lai, MD, MPH, of the Division of Pulmonary and Critical Care Medicine, explain that common methods of quantifying microbial exposure have inherent limitations, which probably explain the contradictory evidence on whether microbial exposure harms lung health.