Last October, a young girl with severe and unusual liver failure was admitted to a hospital in Birmingham, Alabama. Her symptoms were typical: skin and eyes yellow with jaundice, markers of liver damage off the charts. But she tested negative for all the usual suspects behind liver disease. Her only positive test was, surprisingly, for adenovirus—a common virus best known for causing mild colds, pink eye, or stomach flu. In rare cases, it’s linked to hepatitis, or inflammation of the liver, in immunocompromised patients. But this girl had been healthy.
Then it happened again. A second kid came in, about the same age, with all the same symptoms, and again positive for adenovirus. “One patient is a fluke; two is a pattern,” says Markus Buchfellner, a pediatric infectious-diseases doctor at the University of Alabama at Birmingham (UAB). Two quickly became three and then four. Alarmed, the hospital’s doctors alerted local health authorities and the CDC, whose investigation ultimately found nine such cases of unusual hepatitis in children in Alabama. Two needed liver transplants.
Buchfellner originally thought that whatever was happening was local to Alabama. But this spring, investigators in the UK began independently puzzling over their own mysterious uptick in hepatitis among kids. They have since identified more than 150 such cases in the UK This prompted the CDC to cast a wider net, bringing the number of suspected cases across the US to 109. Fifteen of the kids have needed liver transplants, and five have died. Worldwide, probable cases now total 348 spread across 20 countries.
The early evidence continues to point to a link with adenovirus—an unexpected correlation that is too strong to dismiss and not strong enough to close the case. Seventy percent of the probable cases globally have tested positive for adenovirus, according to the World Health Organization. But although biopsies have been conducted in a small fraction of those cases, they have failed to find adenovirus in the kids’ livers. At the same time, we definitely know that a different virus infected a massive number of kids recently: SARS-CoV-2, of course. Yet the correlation here is even less clear; only 18 percent of the probable cases tested positive for COVID.
Adenovirus and coronavirus aren’t necessarily mutually exclusive explanations. The leading hypotheses now suggest an interaction between adenovirus and the pandemic—either because social distancing changed the patterns of adenovirus immunity, allowing for more severe or simply more adenovirus infections, or because of previous infection or co-infection with the coronavirus triggers an unusual response to adenovirus. Alternatively, did the adenovirus itself change recently, evolving to more readily damage the liver?
Severe liver failure in kids is very rare, says Helena Gutierrez, the medical director for pediatric liver transplants at UAB and Children’s of Alabama. But when it does happen, the significant proportion of cases even in normal times remains entirely mysterious. No identifiable cause is ever found in almost half of kids with liver failure so severe that they might need a transplant. Ultimately, understanding the recent pattern of unexplained liver-failure cases in kids may shed light on previously mysterious cases that were once too infrequent to attract much attention.
But why is there an increase right now? The only culprit that can be conclusively ruled out is COVID vaccines, because kids under 5, who make up the bulk of the hepatitis cases, cannot yet be vaccinated. In the weeks ahead, experts will be looking at three key pieces of data to parse the remaining hypotheses.
The first and perhaps most obvious set of data to gather is: Have these kids had COVID before? The overwhelming majority of the kids with hepatitis tested negative for the coronavirus, but investigators are now collecting antibody data to see if any of them had COVID in the past. “I don’t think it’s directly related to the virus itself,” says Buchfellner, but perhaps a COVID infection could have predisposed a kid to liver failure once something else—say, an adenovirus infection—came along. And although multisystem inflammatory syndrome, or MIS-C, following coronavirus infection can affect the liver, the hepatitis patients did not exhibit the other hallmark signs of that condition, such as high inflammatory markers and heart damage.
When the COVID antibody data do come out, a lot of the kids will be positive—simply because a lot of kids in general have had COVID recently. Experts will want to go one step further to determine whether the coronavirus is really playing a role. If so, they’d expect that kids with hepatitis are more likely to have COVID antibodies than a control group of kids who did not have hepatitis.
The second key piece of data is about the adenovirus itself. Adenoviruses are very common, so could all the positive tests simply reflect incidental infections unrelated to liver failure? Here, too, investigators will want to see if kids hospitalized with hepatitis are more likely to test positive for adenovirus than those hospitalized for other reasons. If they are, the link to adenovirus becomes stronger. The UK is analyzing these exact data and is expected to have results in the next week.
Exactly how many kids test positive for adenovirus sounds like a simple statistic, but it can be messy early on, when investigators are dealing with mostly retrospective data. Different doctors in different hospitals might think to order different tests. UAB happened to test for adenovirus, but it’s so low on the list of hepatitis culprits that the test is not necessarily routine. And how tests are done can affect whether they come back positive, says Benjamin Lee, a pediatric infectious-diseases doctor at the University of Vermont. “Is the virus able to be detected in the blood at the time the patient presents for care? Are there other sites that need to be tested?” he asks. What about the nose and throat? Or stool? And indeed, UK investigators have had to make sense of a mélange of blood, stool, and respiratory samples, with varying positivity rates.
A third prong of the investigation will focus on the adenoviruses found in these samples. Sequencing their genomes can determine whether the recently acquired new mutations that can explain the link to liver failure. Adenovirus variants have popped up before, and this type of virus is especially apt at reshuffling its genome. Whole genome sequencing is in the works, although scientists in the UK originally had trouble getting enough virus out of early samples. And scientists don’t have a big database of old adenovirus samples of this kind to compare with the new ones. “We take that for granted out with SARS-CoV-2,” says James Platts-Mills, an infectious-diseases doctor at the University of Virginia. So the initial progress may be slow.
Partial sequencing of the viral genome, though, has already pinpointed one particular type of adenovirus that predominates in the hepatitis cases: adenovirus 41, also known as 41F. (There are more than 100 types of adenovirus. F refers to the species; the number reflects the order in which the types were discovered.) Adenovirus 41 infects the GI tract. Platts-Mills has studied adenovirus 41 in developing countries, where it is a leading cause of hospitalizations for diarrhea in children. It circulates in wealthy countries, too, but in the US it doesn’t cause enough trouble to justify active surveillance. Potentially, Platts-Mills says, the hepatitis cases are only the “tip of the iceberg” of a large number of undocumented mild adenovirus cases 41 cases. The invisible arise, if there is one, could be due to either new viral mutations or many young children getting infected at once, with COVID restrictions relaxing.
Still, it’s surprising to see adenovirus 41 specifically as a suspect in these hepatitis cases, adenovirus experts told me. Although adenovirus has been linked to severe liver failure, it’s not been adenovirus 41 but types 1, 2, 3, 5, and 7. Plus, these cases almost always happen in patients with suppressed immune systems. “In those immunocompromised kids, you could see it in the liver. When we made slides, you could see the viral particles,” says Kurt Schaberg, a pathologist at UC Davis who has studied adenovirus hepatitis. The dark centers of the infected liver cells become big and swollen. It’s all quite obvious. Biopsies didn’t find any of these patterns in the livers of the non-immunocompromised kids. If adenovirus plays a role, it is probably more indirect. Perhaps it somehow triggers the immune system to start attacking the liver, either by itself or in combination with another virus, toxin, or environmental factor. And this might continue even after the virus itself is cleared, so tests for adenovirus could turn up negative.
All of this means that figuring out the answer to these hepatitis cases in kids won’t be straightforward. “If we found virus in the liver, we would be done,” says Buchfellner, in Alabama. “The fact we can’t find that means it’s much harder to prove.” Instead of a single direct cause, investigators are probably looking for an indirect one or multiple indirect ones. In the weeks ahead, nailing down three key questions—whether these kids have also been infected with COVID, whether their adenovirus infections are incidental, and whether their viruses have mutated—will at least narrow down the list of plausible hypotheses.
Meanwhile, the nine kids in Alabama are all recovering. Whatever the cause, doctors stressed to me, the risk of severe hepatitis for healthy kids is still very, very small.