As the summer of 2014 gave way to fall, Kevin Messacar, a pediatrician at Children’s Hospital Colorado, started seeing a wave of children with inexplicable paralysis. All of them shared the same story. One day, they had a cold. The next, they couldn’t move an arm or a leg. In some children, the paralysis was relatively mild, but others had to be supported with ventilators and feeding tubes after they stopped being able to breathe or swallow on their own.
The condition looked remarkably like polio—the viral disease that is on the verge of being eradicated worldwide. But none of the kids tested positive for poliovirus. Instead, their condition was given a new name: acute flaccid myelitis, or AFM. That year, 120 people, mostly young children, developed the condition across 34 states. The cases peaked in September and then rapidly tailed off.
“We didn’t know if it would go away,” Messacar says. “Unfortunately, it came back.”
After just a few dozen new cases the following year, AFM returned in force in 2016, afflicting 149 more people. The next year: another lull. And in 2018: another spike, with 62 confirmed cases so far and at least 93 more under investigation. Parents have described their children collapsing mid-run like “marionette dolls,” or going to bed with a fever and waking up paralyzed from the neck down.
This third wave confirms what many doctors had feared: AFM wasn’t a one-off, but likely a new biennial normal. It’s still rare, affecting just one in 1 million people, but that’s little comfort for the roughly 400 children who’ve been affected, many of whom are looking at lifelong disability or paralysis. “It’s exceptionally frustrating to see it again this year, when we know how much people’s lives are overturned,” says Priya Duggal from the Johns Hopkins Bloomberg School of Public Health. “We don’t really know much more than we knew in 2014—but we’re trying.”
AFM is a new term, but not a new syndrome. Its package of symptoms can be caused by a wide range of factors, including, as the Centers for Disease Control and Prevention notes, poliovirus, West Nile virus, environmental toxins, and genetic disorders. The question isn’t what causes AFM per se, but what is specifically behind the biennial spikes that have appeared since 2014. (There was a small peak in 2012, too, before the condition came to national attention.)
That has proven to be a tough problem to crack. In this era, it seems that scientists could easily grab tissue samples, sequence the genes of everything in them, and pinpoint some consistent microbial culprit. But that hasn’t happened—so far, no single germ has shown up in every case. Despite all the tools of modern science, new diseases, especially rare ones, can be very hard to understand.
AFM is uncommon enough that a hospital might get just a handful of cases in a given year, if any. Many centers must join forces to pull off a rigorous study—and that’s logistically complicated. The condition is also geographically unpredictable. Some places had cases in 2014 but none this year, and vice versa.
More importantly, it’s too risky to take biopsies of the actual affected tissues—the nerves of the brain and spine. Instead, doctors have mostly drawn and analyzed samples of spinal fluid, and there’s no guarantee that whatever causes AFM is actually there.
So far, most of the signs point toward a virus as the cause, and specifically some kind of enterovirus. Unlike influenza, which circulates in the winter, enteroviruses are infections of the autumn, which is when AFM cases peak. They mostly infect young children, and the average AFM patient is 4 years old. Enteroviruses need a large enough population of susceptible hosts in which to circulate, so many lie low after waves of infection and crop up in cycles of two or three years—just as AFM does. And although many enteroviruses circulate widely but have little effect, they have a track record of occasionally infecting the spinal cord and causing paralytic illnesses.
“It’s not too far of a jump” to suspect them, says Roberta DeBiasi, an infectious-disease chief at Children’s National Health System.
One particular enterovirus, known as EV-D68, has emerged as the lead suspect. First discovered in 1962, it seemed rare and unexceptional. But in 2014, it caused a huge surge of respiratory illness throughout the United States. That year, “our hospital was the busiest it’s ever been,” Messacar recalls. “The floors were packed, and we hit capacity.” And when paralyzed children started showing up at the same time, he put two and two together. He and others noted that in 2014 and 2016, EV-D68 was the most commonly identified virus in people with AFM.
But it’s not in every patient. So far, the CDC has only found the virus in the spinal fluid of a single child, and in fewer than half of the stool samples or nasal swabs it tested. “I am frustrated that, despite all of our efforts, we haven’t been able to identify the cause of this mystery illness,” said Nancy Messonnier, who directs the CDC’s National Center for Immunization and Respiratory Diseases, in a recent press briefing. The agency notes on its website that “the cause of most of the AFM cases remains unknown.”
Messacar thinks the case for EV-D68 is stronger than the CDC is admitting. Certainly, caution is commendable; the wrong viruses have been blamed for perplexing illnesses before. But “I don’t think AFM is as much of an unknown as it’s portrayed,” he says. He is frustrated with its billing as a “mystery illness.”
Enteroviruses, he says, are not like typical nerve-infecting germs. They can move through nerves directly, so they don’t always show up in spinal fluid. And EV-D68 isn’t even like typical enteroviruses. Unlike other members of its family, it is quickly destroyed in the gut, and doesn’t show up in stool. It mostly thrives at the back of the nose—a place that few doctors thought to examine when AFM first showed up. Why look inside the respiratory tract of a child with a neurological disease?
Even when doctors did take nasal swabs, their odds of finding EV-D68 were low. In many neurological infections, the worst symptoms aren’t caused by the virus itself, but by the body’s disproportionate immune response. That response can continue even after the virus has been cleared, which means that patients often test negative for whatever first triggered their illness. All the researchers I spoke to think AFM likely behaves in this way, especially since there can be a seven-day gap between the condition’s initial coldlike symptoms and the severe paralytic ones. By the time parents seek medical help, their children could be suffering from their body’s misplaced attempts to fight an enemy that’s no longer there. “The expectation that you’ll find a pathogen in every case is unrealistic because you’re already behind the clock,” Messacar says.
As a work-around, researchers could take the complicated step of analyzing the bodily fluids of AFM patients for immune cells or antibodies that specifically recognize EV-D68. Their existence would at least suggest that the virus was once present. But even without such confirmation, there are other lines of incriminating evidence.
After the EV-D68 epidemic of 2014, a few hospitals, including Messacar’s, started actively searching for the virus in nasal swabs taken from patients with generic cold symptoms. Their surveillance showed that the virus disappeared in 2015 and returned a year later, coinciding with the second AFM wave. It vanished again in 2017 and returned this summer. When the latest AFM wave hit, Children’s Hospital Colorado actually saw it coming.
There’s also compelling evidence from laboratory studies. Last year, Alison Hixon at the University of Colorado School of Medicine showed that EV-D68 strains from the 2014 outbreak can paralyze mice by infecting and killing the movement-controlling neurons in their spine. When Hixon isolated the virus from those neurons and injected it into another group of mice, those rodents also became paralyzed. That fulfills all the traditional criteria for causality. It falls short of a slam-dunk case only because the experiments were done in mice.
Another enterovirus, EV-71, has also been implicated in AFM. It’s endemic to East Asia, where it infrequently causes a similar polio-like illness with the same two-to-three-year periodicity. Messacar’s team detected it in Colorado this spring, and they’ve found it in 11 patients with AFM. Several enteroviruses could be behind the AFM cases.
But even if that’s true, it doesn’t explain why the disease suddenly became a national problem in 2014. Conspiratorial corners of the internet were quick to suggest that immigrants had imported a mystery virus—a ludicrous hypothesis, since EV-D68 was first identified in California five decades ago.
It’s possible, though, that the virus has changed since then. In one experiment, strains from 1962 didn’t paralyze mice in the same way that those from 2014 did. This would hardly be the first time that long-known viruses suddenly became more dangerous. Zika virus, for example, was thought to be innocuous when it was discovered in the 1940s, but only recently acquired a mutation that seemingly allows it to cause severe neurological problems.
“For me, it’s not really about the viruses,” Duggal, from Johns Hopkins, says. “I’m really trying to figure out what causes the paralysis.” Much like poliovirus, which paralyzed just 1 percent of those it infected, it’s likely that AFM is caused by widely circulating viruses that only lead to problems for a small, susceptible minority. That’s why you don’t hear news reports of entire schools coming down with AFM. The disease doesn’t even sweep through entire families: Whenever an affected child has had a sibling, Duggal says, that other child has always been healthy. In one dramatic case, scientists isolated genetically identical strains of EV-D68 from two California siblings, one of whom had AFM and the other of whom had the sniffles.
Duggal is now sequencing the genes of people from 60 affected families to see if the ones with AFM have any unique mutations. Other factors might be relevant, too. Gut bacteria, for example, can affect an animal’s susceptibility to polio. And since humidity and temperature affect the global circulation of enteroviruses, Carlos Pardo-Villamizar from the Johns Hopkins School of Medicine wonders if the world’s changing climate is influencing the new trends in AFM.
To an extent, every disease behaves like this. No germ sickens every person it infects. Instead, the outcomes of encounters between pathogens and hosts almost always depend on their respective genes and other factors like climate, diet, the microbiome, and more.
The point is: Diseases are complicated. If anything, scientists have been lucky to study many viruses—flu, Ebola, and smallpox, to name a few—that are potent enough for their consequences to be clear, regardless of other variables. But there’ll be many instances in which the threads of cause and effect are harder to untangle. Epstein-Barr virus, for example, infects half of Americans before they get to middle school, and 90 percent of them by adulthood. It usually does nothing, sometimes leads to mono, and very infrequently causes cancers.
AFM is similar, and its emergence provides a new opportunity for researchers to confront age-old questions. How do you prove what causes a disease? When does your evidence become strong enough? And while you’re collecting that evidence, what do you do for the people who are affected?
In 2014 and 2016, there were only one or two cases of full recovery,” Duggal says. This year, for whatever reason, the children in Colorado infected with EV-71 are recovering more quickly. But at least one child has died, and some are facing long-term disabilities. Every patient immediately gets intensive physical therapy. If that fails, doctors have tried antibody infusions to reduce inflammation, but it’s unclear if these do any good. “There’s no proven efficacy, but also little risk,” DeBiasi, the infectious-disease chief, says. Once you get to other therapies like plasmapheresis, a process that filters blood, “you’re starting to go up the risk equation. It’s not a cookbook approach.”*
There’s no clear line on prevention, either. With uncertainty lingering around viral causes, the CDC’s advice is generic: “It’s always important to practice disease prevention steps, such as staying up-to-date on vaccines, washing your hands, and protecting yourself from mosquito bites.” Basically: Do the stuff that prevents other diseases.
Messacar would like it to be bolder. By all means, he says, be open to changing evidence and continue looking at other possible causes, but in the meantime, proceed as if EV-D68 is the actual culprit. That means two things. First, begin developing vaccines, a process that could take years. “It may seem early to start thinking about that, but if we don’t do the groundwork and AFM comes back in a bigger way, we’re going to be years behind,” he says.
Second, Messacar says, public-health workers should actively search for the virus in the same way they do for the flu. The CDC does have a surveillance program for enteroviruses, but it’s a passive system that relies on clinicians sending in samples. A more active program, of the kind that only a few hospitals do, would specifically test for EV-D68 in the nose of any child who gets admitted with respiratory problems.
A month ago, after the third wave of AFM had started, Pardo-Villamizar began convening a nationwide group of colleagues from hospitals that were seeing cases. Their goal is to share as much information as possible on how best to study, diagnose, and treat the illness. “We always depend on the CDC, but they’re designed to establish surveillance for diseases, not management and diagnosis,” he says. “We, as clinicians and scientists, should be doing that.”
For Messacar, the most important step is to take the disease seriously. “Right now, it’s very uncommon compared to polio in the 1950s, which caused tens of thousands of cases a year,” he says. “But I don’t want to downplay this as a rare disease, because of the long-term consequences. It’s jumping up the list of public-health priorities, and it deserves increased funding and attention.” A fourth wave is likely to hit in 2020. He wants the country to be ready.
* This article previously implied that plasmapheresis is a low-risk therapy. We regret the error.