Editor's Note: (This story originally published on May 31, 2022.)
(CNN) People who have a Covid-19 rebound after treatment with the antiviral drug Paxlovid can be contagious and may not know it because they might not have symptoms, researchers warn.
"People who experience rebound are at risk of transmitting to other people, even though they're outside what people accept as the usual window for being able to transmit," said Dr. Michael Charness of the Veterans Administration Medical Center in Boston.
Charness and his colleagues recently collaborated with a team of researchers at Columbia University to look into cases of Covid-19 that return after Paxlovid treatment. He said they've found at least two instances in which people have transmitted to others when their infection recurs.
In one case, a 67-year-old man infected a 6-month-old after a half-hour near the child.
The man was 12 days past his first positive Covid-19 test. He had taken a five-day course of Paxlovid and was feeling better. He didn't have any symptoms when he saw the baby, who was his grandson, but about eight hours later, he started to feel ill again.
The baby tested positive about three days later, as did both of his parents. Neither the baby nor its parents had any other close contacts before they got sick.
"It indicates that you can transmit during rebound even before you develop symptoms," Charness said. "And you know, we studied a small number of people. It's certainly conceivable that there are other people out there who don't have symptoms and still have a viral rebound."
In another instance, a 63-year-old man infected two family members during three days of relapse after Paxlovid.
Based on this research, the US Centers for Disease Control and Prevention issued new guidance last week for people experiencing Covid-19 rebound after Paxlovid.
The CDC said people who test positive again and whose symptoms come back after finishing their antiviral pills should restart their isolation period and isolate for five full days. The agency says people can end their isolation period after those five additional days as long as their fever has been gone for 24 hours without fever-reducing medication and they're feeling better. The agency also recommends that people wear a mask for 10 days after their symptoms come back.
The findings and guidance come as Paxlovid use has increased in the United States. According the White House, over the past two months, filled prescriptions for Paxlovid have climbed from about 27,000 a week to 182,000 a week.
The administration credits the increase to its test-to-treat program, which created one-stop hubs in grocery and drug stores where people could take a Covid-19 test and immediately receive and fill a prescription for antiviral medications. The antiviral drugs should be taken within the first few days of symptoms.
The drug works well. In clinical trials, Paxlovid reduced the odds that a person at risk of severe Covid-19 would need to be hospitalized by almost 90% compared with a placebo.
For that reason, the CDC says, early treatment with this medication is still recommended.
As helpful as it is, though, researchers say people should be aware the drug may not completely extinguish the infection.
Charness and his co-authors have now collected at least 10 such cases of Covid-19 recurrence after Paxlovid. Half of them have come from just two families, leading the researchers to conclude that such cases are not all that rare.
The research is shared as a preprint. It has not been scrutinized by outside researchers or published in a medical journal.
Genetic testing suggests that when people get a second round of Covid-19 after Paxlovid, it's not because they've been infected by a different strain of the virus. There's also no sign that the virus has changed or mutated to develop some kind of resistance to the drug.
So far, rebound cases have been mild. There haven't been any reports of severe disease during a Covid-19 relapse. Because of this, the CDC says, there's no reason to think that more treatment is needed.
Why this might be happening is still a mystery.
In his studies, Charness said, the researchers watched the amount of virus in a person's body -- called their viral load -- go down on Paxlovid treatment.
"People take Paxlovid, and what we know it does very well is, it blocks viral replication," he said. And so the levels of virus go down. But then in some people -- no one knows how many, because not enough people have been studied -- levels of the virus begin to climb again nine to 12 days after they first test positive, Charness said.
It's not entirely clear that taht rebound is linked to Paxlovid. In studies of more than 2,200 Covid-19 patients, Pfizer, the company that makes the drug, said there were a few patients who had their Covid-19 come bouncing back after a negative test, but they were in the group that took Paxlovid as well as in those who got the placebo, suggesting that Covid just reappears in some people, even without treatment.
Charness' team has done its own comparison study, however, and found something different. When researchers looked at 1,000 cases of Covid-19 diagnosed between December and March in players and support staff of the National Basketball Association who had not taken the drug, they didn't find any cases of Covid-19 returning. This study is still unpublished.
They say more research is needed to understand whether there could be any connection to the drug.
Charness said the fact that the infection can come back this way after treatment presents some questions. For one, would rebound be as common in people who started the drug later, maybe on day four or five after their first symptoms, after their immune systems have had longer to initially see the virus? Would a longer course of treatment -- maybe taking the drug for six or seven days, rather than five -- lower the risk that the virus would come back?
"No one knows," he said. "Somebody should be studying this."