New Data on How Many Kids Got That Covid Mystery Illness

By Maryn McKenna

Other reports rolled in—a death in France, a cluster in Italy—and in the US, the CDC put together what had been observed of the unusual condition and asked physicians to report cases that matched to state health departments. Coming up with a case definition is a first step in conducting surveillance for what might be a new disease. With it, the CDC learned by mid-May of more than 100 US kids with the condition, three of whom had died.

On August 7, the agency updated those numbers. As of the end of July, 570 US kids had experienced that constellation of shock, heart problems, and gastrointestinal problems; had blood work that indicated inflammation was going on in their bodies; and were positive for Covid-19. Many of them were seriously ill: 364 needed to be treated in an ICU, and 10 died. In many of them, several organs—heart, lungs, kidneys and brain—were affected, which earned the disorder its agreed-on new “multisystem inflammatory syndrome in children” name. Almost one in five of the survivors was left with kidney damage or with weakened arteries and aneurysms for which they will need long-term drug treatment and monitoring.

Many of those cases came through a research network that Randolph supervises, the Overcoming Covid project, which collects case reports from 70 pediatric referral hospitals across the US and is collecting blood and respiratory samples from patients to get ready to launch clinical studies in children with acute Covid-19 and with MIS-C. The network published its first accounting in June, uncovering 186 child and teen patients in 26 states and confirming the novel syndrome was causing “serious and life-threatening illness,” including four deaths. “I think people are very cognizant of this now,” Randolph says. “They’re recognizing less severe cases as well, and defining that there is a spectrum of illness.”

It seems clear at this point that MIS-C is not Kawasaki: The children affected by the new syndrome are school-aged and teenagers, whereas Kawasaki occurs mostly in toddlers. And almost every child reported with MIS-C shows evidence of having had a novel coronavirus infection, whereas the underlying cause of Kawasaki is unknown. But the similarities between the two are enough to light up the science of both.

“If you frame the question as, ‘Are these two diseases the same?’ the answer to that is: clearly not,” says Jane Burns, a professor of pediatrics and director of the Kawasaki Disease Research Center at the University of California, San Diego. “If by the same you mean, ‘Are these immunologic responses in specifically genetically susceptible children, that happened as a result of an exposure that these children had, and then they make an unusual immune response to that exposure?’ Then yes, they're exactly the same.”

That could be a critical insight, because it’s actually not uncommon for children to have inflammatory responses to an infection, and for physicians to never figure out just what the cause was. Still, it could take a while for researchers to gather enough data to understand the immune processes that create MIS-C, even with the help of networks such as Randolph’s.

“Learning what a genetic predisposition might be requires tons of samples, because it’s unlikely to be due to a single rare mutation,” says Burns, who has seen 10 MIS-C patients in her center. As with Kawasaki disease, she said, “It’s likely to be a complex pathway” that allows for a range of responses, from very mild disease to the most serious versions that put children into ICUs.

Finding a genetic link will be challenging because, viewed simply by the numbers, MIS-C is rare, occurring in only about two children out of 100,000. Researchers worry, though, that it is not as unusual as those numbers make it seem, because the MIS-C case definition that causes a child to be counted requires proof of either infection or exposure—and as has been observed many times since the pandemic started, children can carry the virus without symptoms, so their cases are not consistently recognized or recorded.