NEW YORK (GenomeWeb) – A rare syndrome of limb weakness and paralysis is currently striking children between the ages of two and eight years old in the US. It is the third outbreak in five years, and disease detectives at local hospitals, state health labs, and labs at the Centers for Disease Control and Prevention are using all available diagnostics technologies and approaches to try and find a common link among the cases of acute flaccid myelitis (AFM). Yet, the precise cause of the polio-like illness remains a mystery.
The CDC has been investigating 273 cases so far this year with 106 cases in 29 states currently confirmed.
During a call with the media last week, the Director of the National Center for Immunization and Respiratory Diseases at CDC, Nancy Messonnier, emphasized that the agency's investigators are essentially leaving all options open — looking into infectious and non-infectious etiologies as well as things like auto-immune mediated processes — and pursuing many different hypotheses simultaneously.
"We're not ruling anything out at this point," Messonnier said.
The agency is also reaching out to other experts by partnering with a network of neurologists to assist with confirming cases and by establishing the AFM Task Force, a group of academic researchers and other specialists from multiple disciplines. Messonnier said the CDC convened the task force in part "to be able to have the right set of expertise to think broadly across all the hypotheses" and make sure the agency is doing the appropriate testing.
Establishing the case definition has been an important part of the investigation, according to Janell Routh, an investigator at CDC.
"We know that the symptoms of AFM are symptoms that we see with other illnesses," such as Guillame-Barre syndrome or transverse myelytis, Routh said. "To differentiate AFM from other illnesses, our case definition looks specifically at the MRIs of the spinal cord ... the lesions are really specific for the grey matter in the cord, which is different than other illnesses," she said. In the spinal cord, the grey matter tends to contain neuronal cell bodies, while white matter tends to be made up of myelinated tracts.
In addition to the neurological symptoms, about 90 percent of patients with limb weakness have had a fever and mild upper respiratory illness prior to the sudden paralysis, according to Messonnier. The paralysis appears to be permanent in about half of the cases so far.
Yet the AFM syndrome is still quite rare — by comparison, there were more than 21,000 paralytic cases of poliomyelitis during its peak in the US in 1952. Polio also causes acute flaccid myelitis, but it is due to a virus that is shed in stool, with peaks in summer months when children play in contaminated water, while AFM seems to be spiking along with fall and winter cold seasons. Nevertheless, the CDC is testing patient samples for polio, just to be on the safe side. There have been no cases so far.
Interestingly, AFM also appears to occur with a periodicity of two years. It first emerged in 2014 and peaked again in 2016.
Mary Anne Jackson, a pediatric infectious diseases specialist at Children's Mercy Kansas City (CMKC), said in an interview that this periodicity is likely to be real and not a function of small sample size or missed cases. "Parechovirus has the same every-other-year cycle about it, so that's not an unusual feature of a virus," Jackson said.
In terms of other clues, an unusually intense outbreak of respiratory infections caused by EV-D68 during the initial 2014 AFM outbreak led to the hypothesis that EV-D68 could be the causative agent.
Steve Oberste, the Polio and Picornavirus Laboratory Branch chief at CDC, said in an interview that the CDC continues to test patient samples for enteroviruses, in part by using a PCR-based assay that was previously developed during the 2014 EV-D68 outbreak, as well as a newer version of that test.
Indeed, according to a Morbidity and Mortality Weekly Report issued last week, enterovirus and rhinovirus testing is done at CDC for all patients, when possible. According to MMWR, testing has been done on a total of 125 clinical specimens from 71 patients this year including 21 CSF, 59 upper respiratory, and 45 stool samples or rectal swab specimens.
Specimens from 38 of the patients were positive by EV/RV real-time reverse-transcription PCR.
These included 11 positive tests for EV-A71, 14 for EV-D68, and 13 for other viruses. But the positives were also primarily from nonsterile sites, so it is more challenging to demonstrate these are related to the neurological syndrome. From sterile cerebrospinal fluid specimens, only two patients were positive; one for EV-A71 and the other for EV-D68.
One working theory is that, if the syndrome is caused by an infectious agent, the agent is hard to detect because it is cleared by the time of the sudden limb weakness.
This is supported by the fact that one of the two cases with positive CFS test this year was a rare case in an adult who happened to be on immune-suppressive medication, while the other was in a child who had very rapid progression of paralysis, according to Messonnier.
However, the volume of CSF that can be obtained from kids is also very small and limits the testing that can be done.
At CDC, Steve Oberste and Janell Routh are part of a team tasked with testing patient samples sent on from local hospitals and state labs. The agency is also "the lab of last resort," Routh said, so the volume of CSF, which is in limited supply to begin with in pediatric patients, available for it to test has been low.
"However, it is the most specific sample of course, because if you find something — a virus or another agent — in a sterile site specimen, then you can be pretty confident that is what was causing the illness, particularly if it is a CNS disease and a CSF specimen," Oberste said. On the other hand, a throat swab or stool sample, especially from a pediatric patient, could contain pathogens that may be there simply by chance and not causing any infection.
"We know that hospitals are conducting broad searches as well using those same samples within their own hospital laboratories or send-out labs, so [the CDC is] only getting a fraction of the sample that was collected, because they are doing a whole battery of testing on the patient at those hospitals," Routh said.
Indeed, Jackson said her institution does things like PCR and viral culturing, and also has the capability to do metagenomic analyses. "Depending what the clinical scenario is, the workup might be more extensive and not limited just to viruses, so we will do bacterial cultures and serology, things like that," she said.
In addition, the BioFire FilmArray Respiratory Panel (RP) is designed to detect enterovirus species A – D, including enterovirus D68, and that panel is being used by some hospitals like CMKC on respiratory samples, to generate clues in AFM cases.
"When we had the big EV-D68 outbreak in 2014, we saw severe respiratory disease with intractable broncospasms in a lot of children with asthma," Jackson explained. However, she noted that there does not seem to be the same tremendous uptick in EV-D68 respiratory infections this year to coincide with the AFM outbreak, although this may be due to acquired immunity reigning in the severity of symptoms, she said.
Wade Stevenson, SVP of Global Marketing at BioFire, said in an email that the respiratory panel does not speciate or identify specific human rhinoviruses (HRVs) or enteroviruses, but it can be used to rapidly identify cases of rhinovirus/enterovirus and route them for confirmatory testing for EV-D68.
The firm is also developing and testing its publicly accessible BioFire Syndromic Trends software which aggregates de-identified data from participating sites. The software is not part of the firm's FDA-cleared product and should not be used as an aid in the diagnosis of individual patients, Stevenson said.
Jackson's CMKC is a member of the Syndromic Trends Research Group, Stevenson added. "They are part of a cohort that is testing an algorithm designed to differentiate Enterovirus D68 among positive HRV/EV samples detected by the RP version 1.7 panel," Stevenson said.
Jackson said that overall her hospital is identifying cases and "making sure we have CSF and MRI imaging, and utilizing clinical testing." In terms of EV-D68, the lab there is looking at nasopharyngeal swabs, stool, and CSF. "We'll also look for other agents that have been associated with polio-like cases in the past, so, we'll look for other enteroviruses ... and we'll identify those other viruses," she said.
"The CDC always leads the way, but I think pediatric centers who have been involved in this kind of viral research are understanding that there may be some role for those centers to get together and to look at this," Jackson said.
Although there have been recent reports showing that EV-D68 can cause limb paralysis in animal models and that it meets six of the eight Hill Criteria — a set of criteria used to demonstrate causality in infectious diseases — suggesting it is the cause of AFM, the CDC is not ready to name this specific virus as the only cause.
"Let's say it turns out to be an enterovirus; it doesn't necessarily all have to be the same one," Oberste said. "There are over 100 enteroviruses, virtually all of which are known to cause neurologic illness at some level, and, probably at some frequency, paralysis," he said.
The CDC had done metagenomics on some of the original samples. "One of the issues is that metagenomics takes a relatively large sample volume compared to real-time PCR," Oberste said. "There was nothing that jumped out, so we didn't feel like that was a very fruitful avenue," he said. However, the agency is now going back to look at some of the most recent samples with metagenomic technologies, "just because some of the methods have improved a little bit over time and we have better bioinformatics methods," he said.
The agency has also tried a lot of different, and even decades-old, approaches, Oberste said, such as bacterial and viral culture, serology, and experiments with suckling mice.
Routh further noted that CDC researchers are also trying to connect the dots with epidemiological methods as well.
"We're doing both in-depth chart extraction on the clinical information from patients, to try to get at those exposures prior to weakness onset, and we also have done a few in-depth hypothesis-generating interviews with AFM patients and their families, asking a whole host of exposure questions to see what might come up and if there is anything in common," she said, adding, "At this point we still haven't found any common links between patients."
Pinning down an etiology would be a relief to doctors and parents, and could potentially lead to prevention or treatment strategies. If AFM is caused by a virus, for example, a vaccine could be developed.
But Jackson also pointed out that there are illnesses, such as Kawasaki disease, for which no known cause has ever been discovered, despite decades of investigation, and that this is part of the frustrating reality of pediatrics.
AFM is now being tracked in all states, according a statement issued by the Council of State and Territorial Epidemiologists last week. Additionally, "Adding AFM to the list of nationally notifiable conditions is an option that CSTE will continue to consider," the group said. The CDC has also invested in outreach to alert physicians about the illness, and the agency aims to improve its tracking and follow up on AFM patients over the long term.
Oberste emphasized that CDC investigators will continue to hammer away at the mystery.
"It's been quite a challenge, and we keep thinking, what else could we do that's different, what else could we look for, or could we look in a different way? — honestly, there aren't a lot of brilliant new ideas coming forward at this time," he said. "It is kind of frustrating, but that is kind of how this business works sometimes, unfortunately — you don't always get the nice clean answers," he added.