by revere, cross-posted from Effect Measure

I’m just getting around to reading the Brief Report by Blachere et al., “Measurement of airborne influenza virus in a hospital emergency department” (Clinical Infectious Diseases 2009:48:483-440) but it’s quite interesting. We’ve noted fairly often here that we still don’t know for sure what the main modes of transmission of influenza are, something that surprises many people. We “know” that flu can be passed from person to person via the respiratory secretions from runny noses, coughs and sneezes but we often don’t think more deeply about this. We know that viral material can remain viable on inanimate surfaces like doorknobs and arm rests for long periods (maybe days), but we don’t know how often this kind of exposure results in actual infection. As for the virus passing through the air between people, we don’t know if this is through the rather large particles easily visible with coughs and sneezes, particles which are quite heavy and settle out quickly within a few feet or most of the source and aren’t breathed deep into the lungs; or much via the much smaller aerosols that can remain suspended in the air for long periods (perhaps days) and penetrate easily into the depths of the lungs. You can see immediately how the size of the droplets might make a difference. If you go into a hospital emergency room during flu season, are you only likely to get infected if you sit next to an actively shedding flu patient in the waiting room or is the air of the waiting room full of floating flu virus? The paper by Blachere et al. set out to measure the sizes of floating aerosols containing viral material in the air of a hospital emergency department at the height of the 2008 flu season (February).


The study design was straightforward but carrying it you was technically demanding. Stationary aerosol samplers were set up on tripods at various places and at various heights for six afternoons in the emergency department of the West Virginia University Hospital (Morgantown, West Virginia is also the location of an important laboratory of the National Institute of Occupational Safety and Health, NIOSH, experts in measuring contaminants in workplace air). Seven ER docs also wore personal aerosol samplers for 3 – 4 hours (each was tested for flu first to ensure they weren’t the source of any detected viral material). The aerosol samplers were able to distinguish the sizes of the particles: > 4 microns, 1 – 4 microns, and < 1 micron. Particles greater than 4 microns are still respirable if they are also less than 10 microns or so, while they other two sizes are both respirable and will remain suspended in the air for long periods. The collected material in the three size ranges then underwent PCR to see if there was genetic material indicative of flu virus (the primer was from the well conserved M1 matrix protein gene of the flu virus).

Usable data was available from four of the six sampling days and on three of the four genetic material from flu virus was detected. Almost half (49%) were from the two stages of sampling for sizes less than 4 microns, i.e., definitely in the respirable and suspended particle range. Some of the particles larger than 4 microns would also be in that category, but the data in the paper does not permit me to say how much. We know that many of the particles released in sneezes, coughs, talking and normal respiration are in the 4 – 1- micron range, however. None of the samplers in two examination rooms had positive results, but there were positive samples from the Waiting, Reception and triage Rooms, while three of the four ER docs had positive results in their personal samplers.

This is the best information we have to date about influenza A (there wasn’t enough influenza B to study) and where it is in the environment of a health care facility during flu season. But there is a great deal still to learn. Finding viral genetic material does not completely settle the issue. We don’t know if the viral genetic material detected was part of a replicable virus or not, which is critically important. And while information on humidity, temperature and barometric pressure were collected, there were not enough samples to be able to assess the effect of the environment on viral prevalence in the air. This paper shows that the difficult task of measuring and sizing viral aerosols in a health care environment is possible, in itself a major advance.

This is another small piece of a very big puzzle. We don’t have to have every last piece in place but there are large areas where we can see hardly any of the main features, and transmission was one of them. Science is usually slow, but with sufficient resources, it is steady.