ARTICLE IN BRIEF
A case-control study of Guillain-Barré syndrome and Zika virus in French Polynesia suggests the best evidence yet of an association between the two conditions, but experts say the link is suggestive but not definitive.
In a case-control study based in French Polynesia, researchers provided the strongest evidence yet linking the fast-growing outbreak of Zika virus infections to a concurrent, albeit much smaller, outbreak of Guillain-Barré syndrome (GBS) that occurred between October 2013 and April 2014.
Published in the February 29 online edition of The Lancet, the study found that 41 of 42 patients diagnosed with GBS in Tahiti had anti-Zika virus immunoglobulin M (IgM) or IgG, and all had neutralizing antibodies against Zika virus compared with 54 (56 percent) of 98 people in an age-matched control group (p<0.0001) admitted to the hospital with a non-febrile illness.
Neuroepidemiologists and neuroinfectious specialists who were not involved with the study praised the research, and said it added the best evidence to date of a potential link between GBS and Zika. But they agreed that it nonetheless lacked decisive proof of whether and how the virus might cause GBS. Indeed, some questioned whether the diagnosis of GBS in these cases was appropriate.
“It's not classical Guillain-Barré syndrome. I'm not 100 percent sure what it actually is,” Avindra Nath, MD, FAAN, chief of the section of infections of the nervous system and clinical director of the National Institute of Neurological Disorders and Stroke, told Neurology Today in a telephone interview from Liberia, where he was following up cases of Ebola infection that occurred last year. “They didn't find antibodies to the myelin lipids, which one would normally see in GBS,” said Dr. Nath, who was not involved with the current study. “They found it affected the axon, not the myelin. So these differences tell me it's an acute motor axonal neuropathy. It could still be caused by the Zika virus, but it may have nothing to do with the immune system.”
As researchers around the world scrambled to answer such questions, the director-general of the World Health Organization (WHO) emphasized in a news conference on March 8 that public health responses cannot await definite proof of mechanisms. “We can expect to see more cases and further geographical spread,” said Margaret Chan, MD.
Mosquito-borne transmission of the Zika virus, she noted, had been detected in 31 countries and territories in Latin America and the Caribbean [at press time], including three territories of the United States: Puerto Rico, American Samoa, and the US Virgin Islands.
Nine of those countries have reported an increased incidence of GBS or confirmation of Zika infection among GBS cases, she said, including Colombia and Venezuela.
“GBS has been detected in children and adolescents but is more common in older adults and slightly more common in men,” Dr. Chan said. “The anticipated need for expanded intensive care adds a further burden on health systems. Reports and investigations from several countries strongly suggest that sexual transmission of the virus is more common than previously assumed. All of this news is alarming.”
STUDY DETAILS AND COMMENTARY
Frédéric Ghawché, MD, a neurologist at the Centre Hospitalier de Polynésie Française in Papeete, Tahiti, and colleagues reported in The Lancet study that 93 percent (39) of the patients with GBS had Zika virus IgM, and that 88 percent (37) had experienced a transient illness in a median of six days before the onset of neurological symptoms. That suggested a recent Zika infection, the team of two dozen researchers from France, Tahiti, and Scotland concluded.
Electrophysiological findings were compatible with the acute motor axonal neuropathy (AMAN) type of GBS, the team reported, and had an unusually rapid evolution of disease, with a median duration of the installation and plateau phases of six and four days, respectively.
No patients died, but 29 percent (12 people) required respiratory assistance.
The researchers detected anti-glycolipid antibody activity in 31 percent (13) of the patients, and notably against glycolipid GA1 in eight (19 percent) patients by ELISA and 19 (46 percent) by glycoarray at admission.
As Dr. Nath noted, the researchers found that the typical AMAN-associated anti-ganglioside antibodies were rarely present. Past dengue virus history did not differ significantly between patients with GBS and those in control groups.
Despite the uncertainties, Dr. Nath called the study the most convincing evidence to date linking Zika infection to a GBS-like condition.
“The researchers did take considerable efforts to ensure that the serological reactions they observed were directed toward Zika and not just cross-reacting with dengue,” he said. “The results are pretty impressive from an epidemiologic as well as laboratory standpoint. Having said that, it's not quite a slam dunk. It's one study. One would like to see the results replicated.”
A commentary accompanying the paper in The Lancet noted the same unusual presentation of the cases in Tahiti as Dr. Nath observed. “The researchers did not find the expected pattern of antibodies, nor did they find evidence for molecular mimicry between Zika virus antigens and the anti-glycolipid antibodies that might induce an autoimmune response,” wrote David W. Smith, MD, a pathologist at the University of Western Australia, and John Mackenzie, PhD, a professor of tropical diseases at Curtin University in Australia.
“Flavivirus antibodies are widely cross-reactive across the species,” the commentary stated, “and there are also cross-species immune recall phenomena that can lead to spurious early antibody responses when the person has had another flavivirus infection in the past. In fact, only one of the 42 cases showed the standard criterion of neutralization titres to Zika virus that are four-fold or higher than the titre to the dengue viruses.”
Even so, the commentary concluded that it is “very likely” that most of the patients had recently been infected with Zika. “Suffice it to say Zika virus can be added to our list of viruses that can cause Guillain-Barré syndrome,” Dr. Smith and Dr. Mackenzie wrote.
But Daniel M. Pastula, MD, MHS, a neurologist and medical epidemiologist at the University of Colorado Anschutz Medical Campus in Denver, who also was not involved with the study, questioned the diagnosis of GBS in these cases. He noted that the study failed to use recently developed criteria for diagnosing GBS by the Brighton Collaboration, an independent global vaccine safety safety research network for health care professionals.
“The Brighton Collaboration criteria is now used across epidemiologic studies for GBS,” Dr. Pastula said. “This paper would have been more robust if they had used the criteria.”
All 42 of the GBS patients in the Tahiti case-control study received the standard therapy for the disorder, intravenous immunoglobulin, and one also underwent plasmapheresis. The median duration of hospital stay was 11 days (with a range of seven to 20) for all patients, and 51 days (with a range of 16 to 70) for the 16 patients who were admitted into intensive care. Three months after discharge, 24 of the patients (57 percent) were able to walk without assistance.
Those outcomes, while within historical norms and reassuring to the majority of patients, nevertheless demonstrate that the illness is not self-limiting for all patients, said Dr. Pastula.
“Previous data suggests about 85 percent of people are walking independently at one year,” he said. “But others may not fully recover. At one year, full recovery of motor strength occurs in about 60 to 70 percent of patients. Severe motor problems persist in about 15 percent.”
Because it remains unclear whether the neurologic effects observed in Tahiti are due to an immune reaction to the virus or a direct injury by the virus itself, Dr. Nath questioned whether immune therapies make sense.
“Currently the recommendation is to treat with immune therapies,” he said. “I'm not sure that will do anything. We need to do clinical trials. We may be doing the wrong thing.”
Compared to immune-mediated disorders like classical GBS, Dr. Nath added, “Usually neuropathies don't recover as well.”
Even so, immune therapies would be unlikely to have any ill effect on the AMAN subtype of GBS, said Kenneth L. Tyler, MD, FAAN, professor and chair of neurology at the University of Colorado Anschutz Medical Campus in Denver.
“We have no reason to believe that IVIG, which we normally use to treat Guillain-Barré, would have any harmful effect in these cases,” Dr. Tyler said. “Whether they would have a beneficial effect, I don't know. But I'm not aware of any data that would suggest that the usual therapies would be harmful. The outcomes for the patients in Tahiti were not bad based on this preliminary data.”
WHAT TO EXPECT
“In 2015, Brazil most definitely had a Zika outbreak and an accompanying increase in GBS cases,” said James J. Sejvar, MD, a neuroepidemiologist in the Division of High-Consequence Pathogens and Pathology at the CDC's National Center for Emerging and Zoonotic Diseases in Atlanta.
“We haven't heard of any increased reporting of either Zika or GBS in Brazil this year yet. If Zika is going to recur there, we're only now getting into the period where we should start to see it. The rainy season has now ended; that's when the mosquito larvae start to hatch. We would expect the peak of Zika cases to be around March, April, May.”
The CDC, he noted, already has a team there prepared to run a case-control study of GBS cases should they recur.
GBS clusters are already clustering, he said, in Colombia, El Salvador, Venezuela, and Honduras.
In a report on its website, the WHO noted that Colombia typically registers about five cases of GBS per week, but saw 86 GBS in a recent five-week period, three times higher than average. During the month of January, Venezuela recorded 252 GBS cases with a spatiotemporal association to Zika virus.
J. David Beckham, MD, an associate professor of medicine in the divisions of infectious diseases, neurology, and immunology & microbiology at University of Colorado-Denver, told Neurology Today that he is working with the CDC to analyze serum samples from patients in Colombia.
“They'll be sending us blood from these patients who are acutely infected so we can look at T-cell responses and antibody responses,” he said.
As for the United States, Zika transmission has already begun, said Dr. Sejvar. “There's definitely Zika transmission in Puerto Rico,” he said. “We've set up active surveillance for GBS in Puerto Rico in anticipation that they may experience an outbreak of GBS.”
For the continental United States, it is likely only a matter of months until Zika infections are seen in Florida and the Gulf states, said Dr. Nath. [See “Predictions of Zika Virus Risk Estimated for 50 US Cities.”]
“We're very worried about the southern parts of the United States,” he said. “The CDC is already monitoring the mosquito population in Florida and elsewhere.”
Whatever the number of Zika infections might turn out to be in the United States, any associated cases of GBS are certain to be far fewer, according to epidemiologic evidence. A study published in 2013 in Lancet Neurology calculated a GBS rate of 0.24 per 1,000 Zika virus infections, or about one case for every 4,000 infections.
“Be watchful and be observant for unexpected clusters of Guillain-Barré syndrome, either in its conventional form or, based on this paper, in its ANAM form,” Dr. Tyler said. “If you begin to see a higher than expected incidence of this disease, which in most settings is sporadic, then that would be of concern.”
PROJECTIONS OF ZIKA RISK IN 50 US CITIES
Weather, travel, and poverty may be among the factors that could potentially contribute to Zika virus outbreaks in US cities during peak summer months, according to an analysis based on several computer simulation models published in PLOS Currents Outbreaks on March 16.
Using meteorologically-driven models for 2005-2015 to simulate the potential seasonal abundance of the Aedes aegypti mosquito vector in 50 US cities, researchers at the National Center for Atmospheric Research projected that conditions for the mosquito populations would be largely unsuitable in most cities in the US during winter months (December-March), except in southern Florida and south Texas, where relatively warm weather could sustain low-to-moderate potential mosquito populations. But they projected that as the weather warms, conditions could be ripe for Aedes aegypti mosquito populations along the East Coast, across the southern tier of the country, and as far west as Phoenix and Los Angeles.
The researchers analyzed and factored into their analysis past travel patterns from countries and territories with Zika outbreaks, and projected that cities in southern Florida and impoverished areas in southern Texas could be particularly vulnerable to local viral transmission. Poverty could play a role in greater exposure to mosquitoes and the virus, the researchers wrote, explaining that poorer populations might have less access to air conditioning, house screens, as well as clean water.
While the simulation models offer a realistic assessment of where mosquito vector may be found, they do not fully account for all the areas that may be impacted by imported Zika cases, said Kenneth L. Tyler, MD, FAAN, professor and chair of neurology at the University of Colorado Anschutz Medical Campus in Denver. And, he noted, the models do not factor in the potential for sexual transmission of the virus, which researchers are still trying to understand.