Paradoxic pandemic: The inexorable spread of hand, foot and mouth disease
As researchers describe a new way to make vaccines to fight diseases like foot and mouth disease in animals and polio in humans, we look at a related human viral infection called hand, foot and mouth disease. Recent outbreaks in Cambodia and Vietnam raise the prospect of what is often considered to be an innocuous childhood infection spreading further, causing increasingly severe illness, even deaths, and straining health systems around the world.
Much concern is rightly focused on influenza and the risk of pandemics, but there are other emerging and re-emerging infections that demand our attention. In this post, I look at hand, foot and mouth disease from various perspectives, taking in epidemiology, clinical medicine, structural biology, virology and public health, to try and understand what threat it presents, what we are doing to counter it, and what difference an effective vaccine would make.
It starts with a high temperature, though that’s no surprise – children under five are always getting fevers. But then the spots appear: a rash on her hands and feet and around her mouth. And then you notice her finger twitching.
That trivial twitch means she already has grade 2 hand, foot and mouth disease, a viral infection that is no problem for most children to ride out but which in a few cases leads to severe neurological illness and, rarely, death. The difficulty is that there is no way to know how the infection is going to progress in any individual.
Faced with this uncertainty, countries where severe hand, foot and mouth disease is endemic end up admitting hundreds of thousands of children to hospital each year, just in case, putting their health systems under great stress.
As the number of endemic countries rises, as the virus spreads slowly but surely through populations and across borders, is this often harmless childhood illness becoming a serious international problem? What action are we taking to defend ourselves against the possibility of a ‘creeping pandemic’?
The view from Vietnam
Hand, foot and mouth disease first reached Vietnam in 2005, mostly affecting babies and young children. By 2008, some 10,000 cases of hand, foot and mouth disease were reported each year in a country that has around 7.5 million children under five.
Then, in 2011, something changed.
“It was still early in the year but the number of cases was already above 10,000,” explains Dr Rogier van Doorn, head of Emerging Viral Infections at Oxford University’s Clinical Research Unit at the Hospital for Tropical Diseases in Ho Chi Minh City. “Sitting down with our partners from hospitals across the city, it was clear that this otherwise mild illness had become something far more serious.”
Hand, foot and mouth disease was originally described in Canada, the UK and New Zealand in the 1950s: it was a mild infection caused by a Coxsackie virus. In the 1970s, the disease was reported in Japan, only now it was caused by enterovirus 71 (EV71). Coxsackie viruses and EV71 are members of a large set of human enteroviruses – they live in our guts but can infect the nervous system as well. In all, 11 enteroviruses can cause hand, foot and mouth disease; others cause encephalitis and polio, but not all enteroviruses cause clinical symptoms. Related viruses cause diseases in animals, including foot and mouth disease.
By the late 1990s, hand, foot and mouth disease had reached Malaysia and the story took an ominous turn. Children whose illness was caused by EV71 were now more likely to suffer severe consequences – some even died from the disease. What had initially seemed to be a mild childhood infection was now a potential killer.
In Vietnam, then, faced with an emerging disease that might be on the verge of breaking into an epidemic, van Doorn and his team needed to know what they were dealing with. Working with Dr Truong Huu Khanh and other clinicians at Children’s Hospital 1, they started collecting information about more than 3500 children hospitalised with hand, foot and mouth disease between mid-September and the end of November 2011. They found that, of the children with severe hand, foot and mouth disease, three-quarters were infected with EV71.
The view from the hospital
The clinical stages of hand, foot and mouth disease were defined in Taiwan in 1999. Grade 1 is uncomplicated fever along with spots on the hands, feet and mouth; grade 2 shows early indications that the nervous system has been affected, usually a twitch in the fingers. Grade 2 has subcategories depending on whether the twitch has been observed by the child’s carer (2a) or by a medical professional (2b). The latter is taken to indicate severe disease and requires a stay in hospital but most children with grade 2a are also admitted, as are some with grade 1 disease depending on the doctor’s judgement.
Grade 3 disease means more serious symptoms: central functions of the nervous system are affected, giving rise to signs such as high blood pressure, increased heart rate and a persistent fever that does not respond to treatment. Grade 4 is the most serious level of disease, in which the lungs are compromised and the threat of death is at its highest.
About 10 per cent of all children who come to Children’s Hospital 1 with hand, foot and mouth disease are admitted; most of them are under three years old. The average length of stay in hospital is three days, which is obviously an unsettling experience for babies and toddlers (and their parents), even though 96 per cent of them never progress beyond grade 2 disease.
But it is also a problem for the health system in Vietnam. Thousands of children are admitted for treatment and observation because there is no way of knowing who is likely to recover without any complications and who really does need to be in hospital. It puts huge strain on a heath care system already dealing with annual epidemics of dengue and other infections.
Even in hospital, however, there is no specific treatment for hand, foot and mouth disease, no antiviral drug. There are Ministry of Health management guidelines based on expert opinion to guide treatment based on the grade of illness, but there is no hard data on how effective this approach is.
Neurological symptoms, from the early twitch to more severe complications, are alleviated with Phenobarbital, a sedative which may help stop the convulsions but does nothing to treat the disease itself. If the illness gets worse, the patient receives intravenous immunoglobulin, which is thought to work by either suppressing the virus directly or by damping down the child’s immune response to it. But this is very expensive, has a lot of side-effects and no one really knows how useful it is. At grade 3 and 4, when the child’s life is seriously at risk, intensive care becomes necessary.
Despite the presence of the potentially more threatening EV71, van Doorn and his colleagues reported that just six (0.16 per cent) of the children in their study died from hand, foot and mouth disease. Six too many, of course, but this is the paradox of EV71: it causes mild illness in the majority of cases and yet it puts a massive burden on hospitals because we cannot tell who is at high risk of severe illness. Grading may guide treatment options but it cannot accurately predict the risk of progression, which would indicate who can be safely discharged and who needs to stay in hospital.
The view from the virus
Our understanding of how EV71 causes disease is still at an early stage, but reports of its structure have started to give us a better idea of how it operates – and how we might try to stop it.
Like all viruses, EV71 consists of a small packet of genetic information held in a protein shell. Its mission is to deliver its genome to a living cell, so that the infected cell will unwittingly start churning out the proteins and genetic code required to make new virus particles. These particles then assemble themselves before going to find new cells to infect, working their way through the host’s body and out into the environment to find new bodies.
The reason we feel ill when we have a viral infection can be either because our cells are overworked manufacturing viruses and have fewer resources available for us, or because our immune system tries to fight the infection but causes collateral damage such as inflammation.
Enteroviruses belong to the picornaviridae family. Their genomes are stored as chains of RNA inside an icosahedral outer shell. This 20-sided polyhedron is made up of 60 copies each of just four proteins. The precise nature of these proteins determines the structure of the viral shell, with variations on both inner and outer surfaces affecting its ability to invade cells and evade the immune system.
The structure of EV71 was published in March 2012. A paper in Science described some of its distinct features and how it differs from other enteroviruses, while a few days later a different research group revealed more about how EV71 works in Nature Structural Molecular Biology. Both studies used X-ray crystallography to explore the viral structure, taking advantage of synchrotrons such as Diamond Light Source that produce very high energy X-rays, allowing incredible resolution almost down to the level of individual atoms.
Eight months after its structure was published, a glass sculpture of EV71 was unveiled in the dining room of the Royal Society in London at an event celebrating 10 years of Diamond Light Source. Despite the grandeur of the room, it was a rather underwhelming part of the ceremonies: the UK’s science minister made a joke about politicians knowing more about foot-in-mouth disease than hand, foot and mouth disease, and then revealed the sculpture with little context or explanation.
From a distance, it looked like something a child would make by cutting out paper shapes. Closer to the display case, however, the intricate viral genome could be seen, poised in the centre of its icosahedral shell. Each of the 20 corners of the shell had a five-pointed star sitting on it and each of its 20 faces had a curious squiggle.
The stars and squiggles are structures formed by the proteins that make up the shell. The squiggles in particular are key to understanding the virus. They represent ‘canyons’ on its surface where the virus is able to bind receptors on host cells. Binding changes the structure of the shell, creating tensions and stresses that twist the proteins, opening up perforations in the surface and stretching the shell, expanding it, almost as if the virus were taking a breath ready to blow its genome into its target cell.
The crystallographers described these movements as “tectonic”. Their work is a beautiful display of problem-solving: taking structural information and converting it into a dynamic understanding of movement and mechanisms. You can imagine the surface of the virus rippling as 7000 connected protein atoms reposition themselves. It is this microscopic upheaval that uncoats the viral particle and injects the viral genome into a child’s cells.
The view from the lab
In February 2013, almost everyone in Vietnam is preparing for the lunar new year celebrations. Daily life has ground to a halt. On the end of the telephone line, van Doorn has time to discuss EV71, its effects on the healthcare system and what threat, if any, it represents to the rest of the world.
In the end, 100,000 cases of hand, foot and mouth disease were reported in the southern part of Vietnam in 2011, and there will have been many more that went unreported. The outbreak peaked in September, just as van Doorn and his colleagues started collecting the data for their study.
But the outbreak has continued above the normal rate throughout 2012 and has spread to central and northern parts of the country, with another rise in the autumn bringing the annual total of reported infections up to 100,000 again. It is still infecting several times as many children as in previous years and no one knows how many are affected in the community or district hospitals throughout Asia.
In Japan and other countries where EV71-associated hand, foot and mouth disease has been endemic for many years, the infection has settled into a three-year cycle of peaks and troughs, and van Doorn expects the situation in Vietnam to follow suit – eventually.
“The virus infects young children,” he explains, “so it may be that every three years or so, there is a new cohort of babies who are vulnerable to infection and that is large enough to sustain an outbreak. I suspect the number of cases in Vietnam will continue to be high and unsteady before it settles into a similar cyclical pattern.”
In the meantime, what can be done to reduce the burden of EV71 infections?
A better understanding of clinical factors that indicate how the disease will progress in an individual child would help, as well as better diagnostic tests, especially if they could determine which one among every 20 hospitalised children will develop severe disease.
“A start would be to know which children have EV71, but it is not straightforward,” says van Doorn. “It’s not like dengue virus, which circulates in the blood and is relatively easy to detect. EV71 is in the skin and the brain so blood samples don’t help – you have to take stool samples or use respiratory secretions. These do not yield the best data, so it’s not a viable solution at the moment.”
Clinical trials of the available treatment options are crucial to know if they are actually helping patients. At the moment, these treatments are given with the best of intentions but without good evidence. As for new treatments, our increased knowledge about the structure of EV71 raises the prospect of designing drugs to target its critical functions, such as binding to cells and uncoating to release its genome. However, drug development is a long process – there will not be an antiviral therapy available in the near future.
Vaccines are a better bet, and researchers can use crystallographic data to inform their development too. “Vaccines have been successful in bringing down the morbidity and mortality of diseases caused by other picornaviridae like hepatitis A and polio,” says van Doorn. “It is vital that vaccines for EV71 and hand, foot and mouth disease are developed and assessed as well.”
In January 2013, a Chinese group reported results from an early clinical trial of vaccines for EV71 and identified a candidate for the next phase of testing. Other vaccines are being worked on in Singapore, Taiwan and Malaysia. And the foot and mouth disease vaccine described in PLoS Pathogens last night raises the prospect of constructing an artificial EV71 shell without any infectious genetic material in it. This could generate an immune response to protect people from the real virus.
Several questions remain about EV71 and more basic research is still required. For example, van Doorn describes a three-year prospective study working with researchers from three large hospitals in Ho Chi Minh City to assess people’s exposure to EV71 and related viruses across Vietnam. It will tell them much about the prevalence of EV71 in the population, how it spreads geographically and how fast, and whether there is anything distinct about the virus or the immune response in people who get severely ill.
In this study, the team are also taking samples from very severely ill patients and working with the Wellcome Trust Sanger Institute in the UK to sequence their genomes. “We think that a couple of hundred patients will be enough to reveal host genetic associations with severe disease, and if there is anything that makes some patients more vulnerable.”
They also want to know more about how the virus is transmitted from person to person. “We think everyone basically gets infected at an early age and is then protected from disease for the rest of their life,” van Doorn says. “The virus replicates in the gut, and children can shed it for a long time – the shedding time can be as long as several weeks after illness – which suggests that even though the patient’s immune system has fought off the infection, the virus persists.” What they don’t know is whether the virus can reinfect people later in life, replicating in their gut without making them sick, but being shed into their environment.
“While the public health response at the moment includes closing schools and so on, again based on little evidence, it may be important to include adults in that response as well, in order to prevent them getting infected without knowing and passing EV71 to other young children and babies.”
The world view
Outbreaks of hand, foot and mouth disease are known all over the world, but not as big or severe as in South-east Asia at the moment. Having previously worked mostly on influenza, including the 2009 flu pandemic, van Doorn sees obvious differences in the threat of these viruses. Although evidence is thin on the ground, he is willing to speculate about why this might be: “Perhaps hand, foot and mouth disease doesn’t spread in the same way because it mostly affects children under two and they just don’t travel as much.
“Another factor is that there are 11 enteroviruses that can cause hand, foot and mouth disease and infection with one of them may give a degree of protection against all the others. The different distributions of these viruses in different populations may underlie the geographic distribution and spread of the big outbreaks we see.”
And while changes in the structure of the flu virus essentially make it a different virus each season, similar changes in EV71 don’t alter its infectiousness: “There seems to be cross-protection between the different strains of EV71, so it’s not the human immune response alone that’s driving changes in the virus and the emergence of new genotypes.
“We need to understand much more about this virus, how it changes in time and space and what led to the epidemic we are still living through. We also need to understand the risk to neighbouring countries and further afield. We see no reason why EV71 should not spread outside South-east Asia to reach Europe, Africa and the Americas.”
While more lethal viruses such as coronavirus are being monitored carefully around the world, there are less obvious threats from infectious agents that generally cause mild illness or are thought to be under control. If EV71 continues to infect more countries, we could eventually be faced with a new global challenge. There would be relatively few deaths from hand, foot and mouth disease, but very large numbers of young children would present to hospitals, with huge implications for already over-worked healthcare systems.
At the same time, it is impossible to rule out a minor change in the EV71 genome turning it into a much deadlier disease. If we tolerate it for now, are we harbouring a sleeper virus that could one day switch into a pandemic infection?
The fact is, we do not know the answers to such questions. Researchers like van Doorn and Khanh in Vietnam, linked with the Sanger Institute and Diamond Light Source in the UK and colleagues in Japan, the United States, Malaysia and Singapore are helping us learn more about EV71 and develop potential tools to control it. We need to understand this emerging infection in order to address the problems it presents today and to be prepared for the possible problems it could present in the future.
It’s all part of ensuring that we choose the right response when a feverish child’s finger starts to twitch.
Note on funding:
Dr Rogier van Doorn’s work is supported through the Wellcome Trust’s Major Overseas Programme in Vietnam. The prospective study of viruses in Vietnam will form the basis of a series of PhD Fellowships for young Vietnamese scientists supported through the Programme, and of a Wellcome Trust Fellowship for Dr Tan, a Vietnamese postdoctoral scientist. This work was also supported by the Li Ka Shing Foundation.
Khanh TH, Sabanathan S, Thanh TT, Thoa le PK, Thuong TC, Hang Vt, Farrar J, Hien TT, Chau Nv, & van Doorn HR (2012). Enterovirus 71-associated hand, foot, and mouth disease, Southern Vietnam, 2011. Emerging infectious diseases, 18 (12), 2002-5 PMID: 23194699
Plevka P, Perera R, Cardosa J, Kuhn RJ, & Rossmann MG (2012). Crystal structure of human enterovirus 71. Science, 336 (6086) PMID: 22383808
Porta, C., et al. (2013). Rational Engineering of Recombinant Picornavirus Capsids to Produce Safe, Protective Vaccine Antigen PLoS Pathogens, 9 (3) DOI: 10.1371/journal.ppat.1003255
Wang X, et al. (2012). A sensor-adaptor mechanism for enterovirus uncoating from structures of EV71. Nature Structural & Molecular Biology, 19 (4), 424-9 PMID: 22388738
Zhu FC, et al. (2013). Immunogenicity and safety of an enterovirus 71 vaccine in healthy Chinese children and infants: a randomised, double-blind, placebo-controlled phase 2 clinical trial. Lancet PMID: 23352749