From the Journals

Scant risk for SARS-CoV-2 from hospital air


Everywhere they look within hospitals, researchers find RNA from SARS-CoV-2 in the air. But viable viruses typically are found only close to patients, according to a review of published studies.

The finding supports recommendations to use surgical masks in most parts of the hospital, reserving respirators (such as N95 or FFP2) for aerosol-generating procedures on patients’ respiratory tracts, said Gabriel Birgand, PhD, an infectious disease researcher at Imperial College London.

“When the virus is spreading a lot in the community, it’s probably more likely for you to be contaminated in your friends’ areas or in your building than in your work area, where you are well equipped and compliant with all the measures,” he said in an interview. “So it’s pretty good news.”

The systematic review by Dr. Birgand and colleagues was published in JAMA Network Open.

Recommended precautions to protect health care workers from SARS-CoV-2 infections remain controversial. Most authorities believe droplets are the primary route of transmission, which would mean surgical masks may be sufficient protection. But some research has suggested transmission by aerosols as well, making N95 respirators seem necessary. There is even disagreement about the definitions of the words “aerosol” and “droplet.”

To better understand where traces of the virus can be found in the air in hospitals, Dr. Birgand and colleagues analyzed all the studies they could find on the subject in English.

They identified 24 articles with original data. All of the studies used reverse transcription–polymerase chain reaction (PCR) tests to identify SARS-CoV-2 RNA. In five studies, attempts were also made to culture viable viruses. Three studies assessed the particle size relative to RNA concentration or viral titer.

Of 893 air samples across the 24 studies, 52.7% were taken from areas close to patients, 26.5% were taken in clinical areas, 13.7% in staff areas, 4.7% in public areas, and 2.4% in toilets or bathrooms.

Among those studies that quantified RNA, the median interquartile range of concentrations varied from 1.0 x 103 copies/m3 in clinical areas to 9.7 x 103 copies/m3 in toilets or bathrooms.

One study found an RNA concentration of 2.0 x 103 copies for particle sizes >4 mcm and 1.3 x 103 copies/m3 for particle sizes ≤4 mcm, both in patients’ rooms.

Three studies included viral cultures; of those, two resulted in positive cultures, both in a non-ICU setting. In one study, 3 of 39 samples were positive, and in the other, 4 of 4 were positive. Viral cultures in toilets, clinical areas, staff areas, and public areas were negative.

One of these studies assessed viral concentration and found that the median interquartile range was 4.8 tissue culture infectious dose (TCID50)/m3 for particles <1 mcm, 4.27 TCID50/m3 for particles 1-4 mcm, and 1.82 TCID50/m3 for particles >4 mcm.

Although viable viruses weren’t found in staff areas, the presence of viral RNA in places such as dining rooms and meeting rooms raises a concern, Dr. Birgand said.

“All of these staff areas are probably playing an important role in contamination,” he said. “It’s pretty easy to see when you are dining, you are not wearing a face mask, and it’s associated with a strong risk when there is a strong dissemination of the virus in the community.”

Studies on contact tracing among health care workers have also identified meeting rooms and dining rooms as the second most common source of infection after community contact, he said.

In general, the findings of the review correspond to epidemiologic studies, said Angela Rasmussen, PhD, a virologist with the Georgetown University Center for Global Health Science and Security, Washington, who was not involved in the review. “Absent aerosol-generating procedures, health care workers are largely not getting infected when they take droplet precautions.”

One reason may be that patients shed the most infectious viruses a couple of days before and after symptoms begin. By the time they’re hospitalized, they’re less likely to be contagious but may continue to shed viral RNA.

“We don’t really know the basis for the persistence of RNA being produced long after people have been infected and have recovered from the acute infection,” she said, “but it has been observed quite frequently.”

Although the virus cannot remain viable for very long in the air, remnants may still be detected in the form of RNA, Dr. Rasmussen said. In addition, hospitals often do a good job of ventilation.

She pointed out that it can be difficult to cultivate viruses in air samples because of contaminants such as bacteria and fungi. “That’s one of the limitations of a study like this. You’re not really sure if it’s because there’s no viable virus there or because you just aren’t able to collect samples that would allow you to determine that.”

Dr. Birgand and colleagues acknowledged other limitations. The studies they reviewed used different approaches to sampling. Different procedures may have been underway in the rooms being sampled, and factors such as temperature and humidity could have affected the results. In addition, the studies used different cycle thresholds for PCR positivity.

A version of this article first appeared on

Recommended Reading

Doctors publish paper on COVID-19 protocol; Experts unconvinced
The Hospitalist
Strategies for tracking SARS-CoV-2 could help detect next pandemic
The Hospitalist
Latest rise in child COVID-19 cases is relatively small
The Hospitalist
COVID-19 anticoagulation trials ‘paused’ for futility, safety
The Hospitalist
Call to arms: vaccinating the health workforce of 21 million strong
The Hospitalist
Child abuse visits to EDs declined in 2020, but not admissions
The Hospitalist
Moderna’s COVID-19 vaccine deemed ‘highly effective,’ but further studies needed
The Hospitalist
Hospital volumes start to fall again, even as COVID-19 soars
The Hospitalist
COVID-19 mortality rates declined, but vary by hospital
The Hospitalist
New resilience center targets traumatized health care workers
The Hospitalist
   Comments ()