As bird flu spreads, here’s what scientists are concerned about:

Nearly four months have passed since bird flu began spreading among dairy cows in the country, raising concerns that if left unchecked, the virus could eventually cause a pandemic.

The recent outbreak of human infections linked to a chicken farm in Colorado only underscores that the threat remains real.

Genetic sequences of the virus recovered from the sick poultry workers were highly similar to those circulating in dairy herds, suggesting that the cows somehow introduced the virus into the poultry flocks.

At one large poultry facility, workers were culling chickens under particularly dangerous conditions.

Health officials said they struggled to properly wear protective gear over their mouths, noses and eyes while handling thousands of sick birds in sweltering barns where industrial fans blew feathers and other material that could contain the virus into the air.

Given these circumstances, it’s not surprising that people are becoming infected with the virus, says Jennifer Nuzzo, an epidemiologist at Brown University.

“This is a gamble with people’s lives,” she said. “There’s no other way to describe it.”

State and federal health officials are still investigating the extent of the outbreak, but so far all who have tested positive have only experienced mild flu-like symptoms.

Nuzzo said the outbreak on the poultry farm was a reminder of the risks of having a reservoir of the virus in dairy herds, providing plenty of opportunities for the virus to jump between species and adapt to mammals.

“Every time an avian flu virus has an opportunity to infect humans, it’s like buying a lottery ticket that you don’t want to win,” said Troy Sutton, a virologist at Pennsylvania State University who studies avian flu transmission.

Based on the latest research, here’s what scientists are learning and worrying about as they study the virus.

Certain mutations may make the virus more dangerous

Luckily, there are no signs that we’ve drawn the dreaded lucky draw — at least not yet.

The virus, which travels through dairy cows, is primarily adapted to infect birds but not humans, but it clearly has some mutations that help it take hold in mammals, says Thomas Peacock, a virologist at Britain’s Pirbright Institute, who studies avian influenza.

“It would not be surprising if the same amount of virus was much more infectious than the pure avian virus identified in poultry last year,” Peacock said, adding that this could favour its potential to infect humans as well.

Scientists are still trying to understand more about what exactly these changes in the dairy cow H5N1 virus do, and some virus samples show that certain mutations enhance the virus’s ability to replicate inside mammalian cells, he says.

A major concern is whether the virus could evolve to exploit receptors of the type found in the human upper respiratory tract, which could allow bird flu to spread easily between humans through the air, similar to the seasonal influenza viruses that normally infect humans.

Peacock and other scientists are closely watching changes to the hemagglutinin protein on the surface of the virus, which is the epicenter of this evolution.

New research suggests the virus may be more susceptible to infecting mammals than other types of bird flu.

There are still many unanswered questions about how humans become infected with the virus.

One possibility raised by federal health officials is that milk getting into the eyes, for example, could explain some of the infections in dairy workers who only had conjunctivitis, and they have speculated that aerosolized milk could be another source of infection.

Peacock said the virus has yet to successfully establish itself in the human upper respiratory tract, but that replication is clearly occurring there in at least some people, with nasal swab tests testing positive for traces of viral genetic material.

Studies of avian influenza conducted before the dairy cattle epidemic showed that airborne transmission could occur among ferrets, which are used as a model for human infection, by artificially introducing just a few mutations.

Since the infection began in dairy cows, scientists have begun to analyze how the virus behaves in an effort to understand the threat it poses to humans.

A new study from the University of Wisconsin-Madison found that the virus can be transmitted through respiratory droplets in ferrets, but only with low efficiency.

Study author Amy Eisfeld said the lab had not seen such cases with other highly pathogenic avian influenza strains it had isolated from the wild and tested in ferrets.

“This virus has some worrying characteristics,” said Eisfeld, a virologist at the University of Wisconsin-Madison. “It’s important to monitor what’s going on here and limit infection in cattle and exposure to humans.”

No virus was detected in nasal swabs from the infected ferrets, but the blood had antibodies indicative of infection with the virus.

Sutton points out that just because lab ferrets are infected with the virus doesn’t necessarily mean it will be contagious in humans.

In another worrying finding from the study, the team found that the virus can bind to a type of receptor found in the upper respiratory tract in humans, suggesting that the virus has “features that may facilitate infection and transmission in mammals,” the authors wrote.

To find out, Eisfeld says they artificially produced these receptors, attached them to pieces of plastic, then added the virus to see if it would stick.

Because this wasn’t done in actual humans, the results should be interpreted with caution, she said: “We don’t want to sound the alarm or suggest that this could be transmitted between humans.”

Peacock, who was not involved in the study, says the results suggest the virus is “more infectious” for mammals than previous avian viruses. And while the virus doesn’t appear to be spreading among humans yet, he worries that could change. “Because this is an influenza virus, if you put pressure on it to do something, it will learn to do something.”

These findings are certainly disturbing, but Troy Sutton says they need to be put into context: Laboratory experiments inherently create “an environment conducive to infection” that may not reflect what’s happening outside the lab.

“It’s not exactly like a real-world human nose,” he says. “You’ve got mucus and bacteria and all sorts of other things that get in the way.”

The virus can spread among cattle through respiratory infection

Infected cows have been found to continue shedding high levels of the virus in their milk for days to weeks before eventually recovering.

This supports the hypothesis that the virus is mainly spread from cow to cow during the milking process and through shared equipment on dairy farms, rather than through the respiratory route.

However, some studies suggest that respiratory infection may also be a possibility.

In one recent study, a small number of non-milking cows were intentionally infected with aerosols containing avian influenza collected from milk. Only one of four consistently had detectable viral genetic material in nasal swabs, while the remaining cows only tested positive on a few days. Autopsies showed evidence of viral replication in the lungs, but none of the cows showed severe symptoms.

In contrast, lactating cows whose teats were intentionally infected quickly showed signs of disease and increased viral loads.

Taken together, these findings support the hypothesis that contact with infected milk is the primary source of infection, but also suggest that the respiratory route may still play a role, said Dr. Amy Baker, lead author of the preprint study and a research veterinarian at the National Animal Disease Center in Ames, Iowa.

“This doesn’t tell us if this is the primary route of infection on dairy farms, but it does show that we need to be prepared that at least respiratory infections and transmission is a possibility,” she said.

The risks of being complacent about preventing the spread of infection

To be clear, there is no evidence yet that people with the virus have transmitted it to others, so the CDC still considers the risk to the general public to be low. Additionally, a recent study of workers at two affected Michigan dairy farms found that analysis of blood samples did not reveal any antibodies that would suggest past infection.

All human infections so far appear to have occurred in “highly contaminated, high viral load environments”, which is reassuring as it means measures can be taken to minimise spread, Sutton said.

“If people start getting infected even though the viral load is low, that’s cause for concern,” he said.

Federal health officials maintain it is still possible to quell the outbreak in dairy cows.

Nuzzo is skeptical: He has yet to see a clear, actionable strategy to end the outbreak, with new cases appearing in dairy herds almost weekly.

As a scientist following the situation from outside the US, Peacock is at a loss to understand why months of government measures have not curbed the spread of a virus that could spark a true pandemic.

“My view is that if there had been even a moderate effort to stop this, it would have been stopped already,” he said.