Two years ago, doctors in cities in the northeast of Brazil began to see the first startling cases of what today is called congenital Zika syndrome (CZS) – babies born with a range of severe birth defects, including small, misshapen heads (microcephaly). The doctors suspected – and laboratory work soon proved – that the damage was caused by a Zika virus infection contracted while the mothers were pregnant.
In a matter of weeks, Brazil declared the phenomenon a national health emergency – and, as international alarm at what was unfolding grew, the World Health Organization declared it a global one less than three months later. The rest of Brazil, then the rest of the Americas, braced for disaster, as a newly vicious strain of the Zika virus was carried by mosquitoes inexorably north.
The disaster didn't come – to the immense relief of pregnant women and governments across the region. Babies in other places have since been born with CZS: in Colombia, in Puerto Rico and the continental United States, and in Southeast Asia. But the phenomenon is relatively rare – in most places, it seems, about 7 per cent of pregnant women who contract Zika will have affected babies. Yet, in northeastern Brazil, back in 2015, the rates of birth defects appear to be manyfold higher, and a much greater proportion of the affected babies have more severe clinical presentations such as microcephaly. The difference is stark. And it only happened once: There was another wave of Zika, a year later, but there was no second spike of affected babies.
And while epidemiologists and public-health experts in Brazil are glad the CZS crisis was not replicated on the same devastating scale – the fact is that, two years later, no one knows why not – and no one can explain what happened in 2015.
This is the kind of baffling public-health puzzle the AIDS virus represented in the early 1980s. There have been some significant scientific gains on Zika: There is a vaccine for the virus already in trials and scientists have mapped out much of the path of how Zika destroys fetal neural development. But when it comes to the mystery of northeastern Brazil, there are mostly just big questions, and a range of best guesses, with researchers facing immense obstacles in trying to prove one theory or another. The challenge has been compounded by the political and economic crises that continue to grip Brazil: It has shrunk research budgets and paralyzed ministries and departments responsible for trying to unravel this mystery.
The first big question: Is there really a mystery at all?
Among epidemiologists, there is debate about even what to debate. There are some who argue we don't yet know for sure that anything unique happened in the northeast of Brazil. The reason, essentially, is this. There were 1,373 known babies born with known CZS in the northeast during the first wave in 2015. But to know what percentage that represents, of the total number of babies born to pregnant women who were infected with Zika, epidemiologists would have to know how many of those women there were. And they don't.
Zika, in 2015, was not a "notifiable disease" in Brazil: Doctors did not have to report cases to health authorities. Zika, when all this started, was not believed to be a big deal. More than half the people who get infected don't realize they have it, and most never get tested. Diagnostics, back then, were poor. Women who got Zika while pregnant in the first wave may have had no symptoms; may not have been sick enough to see a doctor or get a test; may have reported their symptoms, but no one considered Zika.
So while researchers know there were, comparatively, many babies born with CZS in the northeast – it is possible the denominator is also comparatively high: maybe a huge number of pregnant women were infected, proportionally many more than anywhere else.
Because of the alarm raised in Brazil, surveillance systems were put in place in other countries – thus, there is data on how many pregnant women got Zika in Colombia and elsewhere. But there is no comparable number here.
"We can't really tell if what happened here was replicated or not in other areas of Brazil or Latin America because we don't really know how many women were exposed," said Ernesto Marques, an infectious-disease researcher with the University of Pittsburgh who is conducting Zika research in the northeastern Brazilian city of Recife. "And each country is defining and confirming cases differently, which also creates biases."
Laura Rodrigues, a Brazilian professor of infectious-disease epidemiology at the London School of Hygiene and Tropical Medicine, believes northeastern Brazil may have had a fast-moving outbreak of a particularly injurious strain of virus that swept through a Zika-naive population, something much more severe than anything seen elsewhere in the Americas.
"My personal gut feeling is that we had large Zika epidemics, with lots of mosquitoes and most people getting infected, in northeast Brazil, and in Cape Verde and French Polynesia" – where, retrospectively, scientists found rates of CZS comparable to northeastern Brazil's, in outbreaks that occurred before the connection with Zika was known. "And in other places transmission was much slower, with fewer mosquitos, more air conditioning, and so on, so it will take years for the same proportion of people to get infected, and microcephalic babies will not peak but are spread over years," she said.
But, she added, "a gut feeling is not science" and they will have to wait for data to be sure.
Retrospective work is now under way to try to calculate exactly how many pregnant women got infected back in 2015. Yale University epidemiology professor Albert Ko and colleagues are tracing women in the northeast who were pregnant during that first huge outbreak to try to calculate how many were exposed to Zika, based on their antibodies.
But it is slow and difficult work. Complicating matters is the fact that once the alarm was raised, there was overreporting of affected babies.
Lack of a denominator notwithstanding, Brazil's Ministry of Health and many prominent researchers are proceeding on the conclusion that there was an aberration in the northeast, even if they don't know how great of an aberration it was. And that leads to the question of what caused it.
The theories: Does previous arbovirus exposure protect fetuses from CZS?
This is one key line of research being pursued here: the hypothesis that women who had previously been exposed to another arbovirus, one transmitted by mosquitos or ticks, usually dengue fever, or vaccinated for another, yellow fever – both of which have long been endemic in Brazil – had some degree of protection from Zika or from its most virulent effects. "We know that antibodies produced by dengue appear in laboratory models to offer protection," Dr. Ko said.
Luciano Pamplona, an expert on arboviruses at the Federal University of Ceara in the northeast, has published work showing that the areas where the highest rates of CZS occurred in 2015 were also the areas with the lowest rates of yellow-fever vaccination coverage.
"We can't say that a woman who has the vaccination is protected [from CZS], but it has very strong evidence and biological plausibility," he said.
Dr. Marques in Recife notes this hypothesis fits with the fact that the highest rates of babies born with CZS were in younger women – those who had had less time to be exposed to dengue or to get the yellow-fever vaccine in one of the periodic emergency campaigns the Brazilian government does when there is an outbreak, roughly every seven years.
Does previous exposure actually make it worse?
Paolo Zanotto, a virologist at the University of Sao Paulo, is investigating whether previous infections with arboviruses such as dengue meant women had antibodies that served to make Zika more virulent – that actually assisted it in invading cells and replicating.
He points out that historically there is much more dengue in the north of Brazil than the south; he believes that as many as 80 per cent of people in the northeast carry those dengue antibodies. He knows his hypothesis directly contradicts that of many of his fellow researchers. "This is a good indication of how messy the thing is – it's a kaleidoscope that moves around depending how you shift your hand," he said.
Maybe it's both
Epidemiologist Joao Bosco Siqueira, who conducts research at the Federal University of Goias in central Brazil, has a theory that both those conflicting theories are, in a way, true. The twist may lie in the way dengue hits people: "If you have infections close to one another – if you get dengue in one year and again in the next, the second infection is not a severe disease. But if the interval between infections is longer, the risk for a more severe dengue case is higher." The dengue and Zika viruses are extremely similar, he notes – so he wonders if the key factor is how long it's been since the pregnant woman had dengue when she contracts Zika. Maybe a recent dengue infection protects her, but one she had 10 years ago makes Zika more virulent?
The northeast of Brazil hasn't had a large outbreak of dengue in the past five years, but the central and much of the southeastern regions have – and those are the areas in which CZS rates are the same as everywhere else in the world.
"Or maybe I'm completely wrong – maybe the problem is not recent infection but an infection a long time ago," Dr. Siqueira said with a laugh that was somewhat despairing. "We lack the data, still."
Could it be cows?
Ion de Andrade, an expert on arboviruses and pediatrics with Potiguar University in the northeast, and his colleagues suspect a different culprit: livestock. They mapped the Zika cases in the worst-affected areas to show that the far more cases of CZS occurred in less-populated small cities and rural areas than in major centres. They call it a "surprising rurality" (arboviruses typically flourish in densely populated urban areas, where mosquitoes are most plentiful) and note a correlation with the denser presence of livestock. They hypothesize that a woman's previous or simultaneous exposure to another, cow- or pig-borne pathogen, such as bovine viral diarrhea virus, may have caused the CZS.
Is it something else all together?
Brazil's Ministry of Health is encouraging investigation of the co-factor hypotheses. "It is becoming increasingly clear that something beyond Zika virus occurred in a part of the northeast region and was one of the causative factors of microcephaly," said Fatima Marinho, co-ordinator of epidemiological analysis and information at the ministry. "We have always been betting on an environmental co-factor, like water, which could carry various contaminants, even biological ones." She wants to see more work on Dr. Andrade's bovine-virus theory and more on other pathogens that could be carried in water.
Ernesto Marques is focused on the fact that the overwhelming number of babies born with CZS in the first wave of Zika infection in the northeast have poor, young mothers who used the public-health system. That, he says, could be a clue to environmental factors.
Investigators are also considering genetics – is there something in their genetic makeup that made a community of women in the northeast particularly vulnerable?
What's the role of abortion?
There is also the possibility that some other non-disease factor is responsible for the fact that other places were spared the experience of the northeast. For example, maybe the huge push by governments to control the vector – mosquitoes – and keep women from getting Zika had a large effect.
And then there's the difficult-to-track impact of abortion. The procedure is illegal in Brazil, but many women do terminate pregnancies – but, of course, there is no data to track a criminalized procedure. However, in northeastern Brazil, women who had babies with CZS in the first wave were too late in their pregnancies to take any action once the alarm about microcephaly was raised.
In Colombia, however, abortion is legal in cases of fetal abnormality; it is legal in the United States and parts of Mexico, countries that might have shown elevated rates of babies with CZS. Even where legal, however, it is a stigmatized procedure, which means women aren't reporting pregnancies ended because of suspected or confirmed fetal abnormalities from Zika.
The slow search for answers
Brazil declared its Zika emergency over this past May.
The WHO's advice to people who live in Zika-affected countries continues to be to try to avoid infection, from mosquitoes or sex; a travel warning for pregnant women from non-affected countries remains in effect.
Dr. Pamplona, who believes there may be a co-relation between CZS and low yellow-fever vaccine coverage, wants to pursue the research – by using not just maps but data from individual women who had Zika but may or may not have had affected babies – and the vaccine. However, he is struggling to find funding for a bigger study. "It's very difficult to get resources," he said. There was new money for Zika research right after the crisis began, but funds were subsequently redirected to work on chikungunya, another arbovirus, when the country was hit with a vicious outbreak this past year, he said.
Dr. Zanotto, the Sao Paulo virologist, says his lab is still working on Zika, but their work is affected by successive political scandals that has effectively paralyzed government over much of the past three years. "The political layer [at funding ministries] is in turmoil and it's very disruptive for science. … Things are moving, but not as fast as one would want."
Dr. Siqueira, the dengue expert, is continuing his research – although he is also uncertain where the funds will come from – and, he says, there is another problem: who to study. "We haven't seen a wave in areas that were previously unaffected – fortunately, it didn't happen – but once you don't have new cases it becomes much, much harder to perform research.
"You need to look back at retrospective data, the cases lack lab surveillance – this is a truly huge challenge at this point. Now, we're trying to do research with fewer pregnant women infected during pregnancy – and the number of malformations is much lower than we have seen in the northeast – and we truly can't reproduce all the natural scenarios of the northeast because countries and regions are different and different backgrounds."
But all of these questions need answers urgently, even if the disaster in the northeast hasn't been replicated, Dr. Ko said. "The virus is obviously circulating in Africa and Asia, causing outbreaks; it may be a large proportion of women of childbearing age who are susceptible and we need to get a handle on what the risk is."
A look at some potential co-factors
These maps illustrate the mystery of congenital Zika syndrome in Brazil.
The first map shows confirmed cases of
microcephaly (the most visible sign of CZS) from the first, most severe outbreak of Zika, from late 2015 to mid-2016.
The second map shows where the most
Zika virus cases were recorded, while the third shows the concentration of cases of dengue fever in that period.
The fourth map shows where yellow-fever vaccine coverage is highest. Scientists are investigating whether yellow-fever vaccination, or dengue-fever exposure, could be the elusive co-factor that caused the rate of CZS to be so much higher in the northeast of Brazil.
Maps by Trish McAlaster and Elisângela Mendonça
Source: Brazilian Ministry of Health