Possible Cluster of Human Bird-Flu Infections Expands in Missouri

Seven people in contact with a patient hospitalized with bird flu also developed symptoms, the C.D.C. reported. Some are undergoing further tests.

A possible cluster of bird-flu infections in Missouri has grown to include eight people, in what may be the first examples of person-to-person transmission in the United States, the Centers for Disease Control and Prevention reported on Friday.

If confirmed, the cases in Missouri could indicate that the virus may have acquired the ability to infect people more easily. Worldwide, clusters of bird flu among people are extremely rare. Most cases have resulted from close contact with infected birds.

Health officials in Missouri initially identified a patient with bird flu who was hospitalized last month with unusual symptoms. The patient may have infected one household member and six health care workers, all of whom developed symptoms, according to the C.D.C.

Investigators have not yet confirmed whether any of those seven individuals were infected with the virus, called H5N1, leaving open the possibility that they had Covid or some other illness with flulike symptoms.

Still, the news alarmed experts.

“We should be very concerned at this point,” said Dr. James Lawler, co-director of the University of Nebraska’s Global Center for Health Security.

“Nobody should be hitting the panic button yet, but we should really be devoting a lot of resources into figuring out what’s going on.”

If officials confirm H5N1 infection in the household member or any of the health care workers, “it means the virus is inching closer and closer to what would be a real pandemic virus,” Dr. Lawler said. “That is when Pandora’s box is open.”

There have been 13 other confirmed cases of bird flu in people since the outbreak in dairy cattle was announced in March. Those cases were all acquired from interactions with infected dairy cattle or poultry.

Missouri has not reported any infected herds. The hospitalized patient was tested as part of routine flu surveillance. The infection worried experts, because the individual had no known contact with cattle or birds.

There have been no H5N1-related deaths reported so far in Americans, and most cases have been relatively mild, suggesting that the virus does not have the high mortality rate of up to 50 percent seen in Asia.

But the small number of confirmed cases so far makes it difficult to estimate its virulence, said Caitlin Rivers, an epidemiologist at the Johns Hopkins Center for Health Security.

“If we’re thinking about pandemic influenza, even in a worst-case scenario only a tiny fraction of cases would be severe enough,” she said. “At the scale that we would expect pandemic influenza, you quickly get to numbers that overwhelm.”

In previous pandemics, health care workers have often been the sentinel cases, Dr. Rivers said: “Across outbreaks, health care workers are often an early sign that something has gone wrong.”

“If any of those six cases are confirmed, then that would definitely increase my level of alarm,” she added.

In a news briefing two weeks ago, federal officials referred to the hospitalized patient as a “one-off” and said no close contacts had become sick.

But the next day, the C.D.C. disclosed that someone who lived with the patient had developed gastrointestinal symptoms, which can sometimes accompany flu.

Two health care workers who had contact with the patient also developed symptoms, C.D.C. officials said at the time. One of those workers turned up negative for flu but may have been tested too late to pick up signs of infection.

On Friday, the C.D.C. added four more health care workers to the list of possible cases. Three of the workers developed symptoms after the hospital instituted precautions against respiratory infection, but they were not tested for active infection at the time.

Officials are now analyzing results of antibody tests that would reveal whether these individuals were exposed to the virus.

No one in the possible cluster has been identified, nor has the hospital at which the initial patient was treated. The investigation is led by Missouri health officials with remote assistance from the C.D.C.

Experts criticized this slow trickle of information and the lack of important epidemiological details, including the timing of symptoms in the initial patient and close contacts.

“Epidemiologists can get a lot of information out of dates and timing and the nature of the contact,” Dr. Rivers said.

The tone of official communications has also been unduly optimistic, Dr. Lawler said.

“The messaging that has come out has been maybe a little more confident and a little more placating than it probably should be,” he said. “It does not help to build back the credibility of the public health sector.”

Ideally, the first case would have led to a full investigation, but Missouri health officials did not get confirmation of H5N1 until after the patient was discharged.

Even now, a thorough recounting of the investigation and timeline of infections would help clarify how the virus might have spread, experts said.

Because the initially identified patient had no known exposure to infected animals, he or she may have acquired the virus from another person who was infected.

H5N1 has infected a staggering array of birds and animals, including mammals, over the past few years, but it is has not proved capable of spreading easily among people.

C.D.C. officials were able to sequence only a portion of the virus isolated from the hospitalized patient. The analysis suggested that the virus had not diverged from the version circulating in dairy cattle, nor had it acquired mutations that would allow it to spread readily from person to person.

But every new infection gives the virus opportunities to acquire those mutations and evolve into a form that could set off the next pandemic.

With the advent of respiratory season, cases of seasonal flu, Covid and respiratory syncytial virus may muddy the picture further, making it difficult to identify H5N1 cases.

“It’s not even a needle in a haystack, it’s kind of like a needle in a stack of needles,” Dr. Lawler said.

In the wake of the backlash against public health measures during the coronavirus pandemic, officials have been reluctant to compel people to get tested. But there are other tools that help officials gauge the extent of the outbreak.

Wastewater analyses have shown H5N1 spikes in several states that are not linked to infected animals or birds, although the peaks are usually transient.

If there were large numbers of undetected infections, “we would see a really consistent wastewater signal,” said Sam Scarpino, director of A.I. and life sciences at Northeastern University.

Still, health officials should test wastewater in areas where they suspect the virus might be circulating, and investigate the source of any signals.

“We need to get this thing out of the dairy cow populations, because if it continues to circulate in these populations, it’s just a matter of time before we’re dealing with the big outbreak,” Dr. Scarpino said.

“We have more than enough evidence now to know that we really need to take decisive action.”