Review finds little evidence of airborne spread of flu

Editor's note: This story was revised Mar 14 to correct a misquote, introduced in editing, that was attributed to Michael T. Osterholm and to include qualifying details that were omitted from the earlier version.

Mar 13, 2007 (CIDRAP News) – A rigorous review of research into influenza transmission concludes that the virus is primarily transmitted at close quarters, a finding that downplays the likelihood of airborne transmission, particularly in healthcare settings, and could have implications for pandemic planning.

In the April 1 issue of Lancet Infectious Diseases, published online ahead of print, Gabrielle Brankston and colleagues of Toronto's University Health Network describe their evaluation of 32 experimental and observational studies that they sifted out of 2,012 citations addressing flu transmission in the English-language literature.

They come to two conclusions. First, that despite more than 70 years' research, much of what is believed to be known about flu transmission is based on poorly structured studies or faulty interpretations of results.

And second, that the preponderance of the robust evidence supports flu being acquired only within a short distance of an infected person—making it much more likely that the virus is transmitted by large respiratory droplets that fall out of the air within several feet, rather than by fine aerosols that can travel long distances and hang in the air for extended periods of time.

The authors found evidence that flu "may be transmitted by the airborne route under certain experimental conditions," the report says. Overall, however, the evidence suggests that airborne transmission, "as traditionally defined, is unlikely to be of significance in most clinical settings."

"Our final conclusion was that we cannot say flu is transmitted over long distances, and the only way flu can transmit over long distances would be through the airborne route," Dr. Michael Gardam, an epidemiologist and the study's senior author, said in an interview. "It really seems to transmit via close contact, which is traditionally understood to be droplet or direct or indirect contact. We cannot rule out airborne transmission, but we see no evidence of airborne transmission—and so we think it is unlikely to be a major player in the overall epidemiology."

The team's finding is the latest entry, though likely not the last, in a vigorous ongoing debate over flu transmission. The issue is vital for pandemic planning because assessment of transmission risk drives decisions about prevention—in particular, whether to offer health care workers surgical masks, which catch droplets, or respirators, which block airborne particles and may intercept flu viruses.

The masks-versus-respirators issue is one of the most sensitive in pandemic planning. N-95 respirators are hard to wear correctly, especially for long periods of time, and respirator manufacturers have conceded that global demand in a pandemic would outstrip their production capacity. Acknowledging those difficulties, the US, Canadian, and British pandemic flu plans all initially recommended surgical masks for health care workers in almost all situations.

But in Canada—where the airborne disease SARS (severe acute respiratory syndrome) shut down the city of Toronto for several weeks in 2003—the Ontario Nurses Association has threatened work stoppages if its members do not receive N-95 respirators during a pandemic. And in the United States, the Centers for Disease Control and Prevention expanded its guidance on respirator use last October under pressure from the healthcare industry, and now says using a respirator is "prudent" for any healthcare worker in contact with a known or potentially infected patient.

In September, an article in Emerging Infectious Diseases charged that health care workers would be inadequately protected by masks and should receive respirators. In December, a research report in the American Journal of Industrial Medicine suggested that masks protect nearly as well against aerosols as respirators do—a finding that was quickly challenged by others, including experts working for mask and respirator manufacturers (see link to previous story below). And in February, the Institute of Medicine held a daylong workshop on pandemic protection for healthcare workers, though it will not report its recommendations until September.

Dr. Eric Toner, senior associate at the University of Pittsburgh Medical Center's Center for Biosecurity and author of several articles on flu protection during a pandemic, applauded the Toronto group's review of flu-transmission research for making explicit how little solid evidence exists.

"A lot of our assumptions turn out to be based on urban legends, or extrapolations made from incomplete information," he said. ""It lays out that so much of what we thought we know, we really don't know."

But because the review makes clear how much remains to be proved, Toner said, flu protection measures should remain conservative, especially for healthcare workers.

"My thinking is, people who are going to have close contact with pandemic-flu patients should wear the greatest degree of protection available, which is in most circumstances an N-95 (respirator)," he said. "And since there is a limited supply of N-95s, they ought to be saved for those people at highest risk of exposure. In the absence of clear answers, we ought to err on the side of caution."

Michael T. Osterholm, PhD, MPH, director of the University of Minnesota Center for Infectious Disease Research and Policy, publisher of CIDRAP News, expressed skepticism about the authors' conclusion that airborne transmission is unlikely to be a significant factor in spreading flu. He said that view doesn't fit well with the seasonal flu's typical pattern of erupting nearly simultaneously in widely separated places.

"From my 30 years of experience I've always been impressed with how seasonal flu functions on a community basis in a way that suggests it could be an aerosol-transmitted disease," he said. "We can see outbreaks in areas that are separated by hundreds of miles; that's much more consistent with an aerosol-transmitted pathogen—though that doesn't prove aerosol transmission. Pandemics have marched around the world in weeks to months in eras before we had jet planes. That smacks much more of an agent that's transmitted by the aerosol route."

Gardam, the senior author, said the team hopes the paper will stimulate two initiatives: greater attention to antiviral prophylaxis for pandemic flu, and fresh attempts to conduct robust flu research.

"The question of respiratory-protection policy has gone far beyond science at this point," he said. "But I hope people will be willing to say that we have to do really well-done studies so that the results can inform future decisions."

Brankston G, Gitterman L, Hirji Z, et al. Transmission of influenza A in human beings. Lancet Infect Dis 2007 Apr 1 (early online publication Feb 19)

See also:

Feb 28, 2007, CIDRAP News story "Questions raised about study on masks as aerosol barrier"

Oct 18, 2006, CIDRAP News story "HHS backs respirator use in caring for pandemic flu patients"

Gift Opportunity

Ebola and Emerging Infectious Disease Fund

Your support is critical to ensure CIDRAP's capacity to respond. Your gift in any amount is deeply appreciated.

Newsletter Sign-up

Get news & practices.

Sign up now»

OUR UNDERWRITERS

Unrestricted financial support provided by

Bentson Foundation 3M United Health Foundation Gilead Become an underwriter»