In the early months of the COVID-19 pandemic, nearly 75% of hospitalized COVID patients received antibiotics on admission, primarily because of limited treatment options and concerns about bacterial coinfections.
One of those antibiotics was azithromycin, a macrolide antibiotic commonly used for respiratory infections. Use of azithromycin was driven in part by a study, now retracted, that suggested it could improve outcomes in COVID patients when used in combination with the antimalaria drug hydroxychloroquine. Although subsequent trials would find the combination had no benefit for COVID patients, widespread azithromycin use continued for several months.
Early in the pandemic azithromycin was “routinely used despite the absence of data supporting a clinical benefit,” according to Michael Pulia, MD, PhD, an emergency physician and associate professor at the University of Wisconsin-Madison School of Medicine and Public Health.
The use of an antibiotic to treat a viral infection reflected the desperation of “having so many ill patients and just wanting to give them something,” said Charles Lanegelier, MD, PhD, an infectious disease physician and professor at the University of California San Francisco.
“People were scrambling for anything,” Langelier said.
Now, in a new study this week in Nature Microbiology, Langelier and a team of US researchers show that, in addition to having no clinical benefit for COVID patients, use of azithromycin was associated with potentially harmful changes in the upper respiratory microbiome and increased expression of antibiotic-resistance genes after only one day of exposure.
“I think it really tells us that inappropriate or unnecessary use of azithromycin has direct and fairly rapid biological consequences in terms of driving antimicrobial resistance,” Langelier told CIDRAP News.
‘Measurable biological consequences’ after only brief exposure
For the prospective observational study, the researchers analyzed nasal swabs of 1,164 adults hospitalized for COVID-19 who were enrolled in Immunophenotyping Assessment in COVID-19 Cohort (IMPACC), a longitudinal study conducted at more than 20 US hospitals from May 2020 to March 2021. The goal of IMPACC, which was funded in part by the National Institute of Allergy and Infectious Diseases, was to assess the clinical and immunologic manifestations of COVID-19 in hospitalized patients.
Of the patients, 31.4% were treated empirically with azithromycin and other antibiotics, 40.7% received no antibiotics, and 27.8% received antibiotics other than azithromycin. Empiric azithromycin use was most common among patients with the highest COVID severity.
Using RNA from the nasal swabs, Langelier and his colleagues analyzed the upper respiratory microbiome, resistome, and system immune response in the patients, comparing the findings from those treated with azithromycin with those who received no antibiotics or other antibiotics.
I think it really tells us that inappropriate or unnecessary use of azithromycin has direct and fairly rapid biological consequences in terms of driving antimicrobial resistance.
What they found was that patients who received azithromycin had a different mix of bacteria in their upper airway, with a reduced presence of some harmless bacteria and an increased presence of pathogenic species, such as Staphylococcus and Klebsiella. More concerning, they found the patients exposed to azithromycin had more “detectably expressed,” or active, macrolide resistance genes than the other groups of patients had.
“That was probably the most striking finding in the study,” Langelier said. In addition, the researchers were surprised to find that observed effect on macrolide resistance genes occurred after only one day of treatment and continued for at least a week after treatment was stopped.
“Things did not go back to baseline and recover after a week,” Langelier said. “It really suggests that even a small amount of exposure has measurable biological consequences.”
Less surprising was the finding that azithromycin was not associated with any changes in host inflammatory gene expression in the blood or airway, which means there was no effect on the inflammatory response in COVID patients who received the antibiotic.
Wider impacts
What concerns Langelier even more is the potential impact of widespread use of azithromycin for other respiratory infections, especially when those infections are caused by viruses. He noted that, as a parent of a three-year-old, he’s become even more aware of the widespread use of antibiotics for mild respiratory infections in children and adults. Research estimates that at least 30% of antibiotics prescribed in US outpatient settings are inappropriate or unnecessary.
“I think it really does emphasize that there are a lot of harms that can come from unnecessarily prescribing this drug in patients who don’t need it,” he said. “And azithromycin has many important use cases for treating a number of [bacterial] infections, so we really need to preserve the drug for those purposes.”
The researchers also note in the study that azithromycin is used to prevent bacterial infections in people with chronic obstructive pulmonary disease and cystic fibrosis, lung transplant patients, and HIV.
“Further work is needed to understand the impact of azithromycin prophylaxis on the upper and lower respiratory microbiome and resistome in these patient populations,” they wrote.
Pulia, who wasn’t involved in the study but has done research on antibiotic use in COVID patients, said the findings are timely given the recent recommendation by the American Thoracic Society to treat all hospitalized pneumonia patients with antibiotics even if a viral pathogen has been detected. Azithromycin is a first-line treatment for community-acquired pneumonia.
“Disruption of the upper airway microbiome and selective pressure for resistance genes represent important examples of the potential deleterious effects of routine antibiotic prescribing for patients hospitalized with viral respiratory infections,” Pulia said in an email.
Langelier said similar studies are planned for other antibiotic classes and other patient populations.
