Spillover

by David Quammen

  Field joined an international team, led by a senior man from the CDC, which had convened from Atlanta and elsewhere to help the Malaysian professionals deal with the crisis. Their first task was to halt the immediate risk to people. “At that time, the human case rate was escalating,” Field told me later, during one of our talks in Brisbane. “Something like fifty new cases a week. So there was huge pressure—social, political—to stop the source of infection.” To do that, he added, the team had to understand the virus and learn how it behaved in pigs.

  They began at what he called “hot farms,” where the infection was still burning its way through resident pigs. You could tell a hot farm by ear; it was Field whom I quoted above, describing the “one-mile barking cough.” He and the rest of the team wanted sick pigs from which to collect samples, hoping those samples might yield a virus matching the one Paul Chua had isolated from his pig farmer. “And that’s what happened,” Field said. They dispatched samples to the Australian Animal Health Laboratory, in Geelong, where colleagues isolated a virus that matched Paul Chua’s. Final proof of that match came from AbuBakar’s team in Kuala Lumpur. All this confirmed pigs as an amplifier host of the same Nipah virus that was killing humans. But it said nothing about where Nipah might ultimately reside.

  The Malaysian government in the meantime had ordered a mass culling—that is, the extermination of every pig, infected or uninfected, on every farm that the outbreak had touched. Some of those piggeries had been abandoned by their operators, panicky and bewildered, even before the discovery of the new virus. People in certain areas even fled their homes; Sungai Nipah became a ghost town. By the end of the outbreak, at least 283 humans had been infected and 109 had died, for a case fatality rate of almost 40 percent. Nobody wanted to eat pork, or to handle it, or to buy it. Pigs were left starving in their pens. Some broke out to roam the roadways like feral dogs, foraging for food. Malaysia at that time contained 2.35 million pigs, half of them from Nipah-affected farms, so this could have become an almost medieval problem, like a scene from the Black Death: herds of infected pigs stampeding ravenously through empty villages. A phalanx of cullers, including soldiers from the army as well as police and veterinary officers, moved into the countryside wearing protective suits, gloves, masks, and goggles. Their assigned task was to shoot, bury, or otherwise dispose of more than a million animals, and to do it quickly, without splashing virus everywhere. Despite all precautions, at least half a dozen soldiers did get infected. Hume Field noted: “There’s no easy way to kill a million pigs.”

  Later in conversation he corrected himself: It was in fact 1.1 million pigs. The difference might seem like just a rounding error, he told me, but if you ever had to kill an “extra” hundred thousand pigs and dispose of their bodies in bulldozed pits, you’d remember the difference as significant.

  Field and the international team, racing ahead of the cullers, also visited farms that had been, but no longer were, hot—where the infection had come and gone. What they found at those sites, by drawing blood from surviving pigs and testing for antibodies, was that the virus seemed to be extraordinarily contagious, at least among swine, even when it wasn’t extraordinarily virulent. The prevalence of antibodies in the animals on recovered farms was typically between 80 and 100 percent. So pigs were far more hospitable and tolerant amplifiers than the poor horses, in Australia, who came down with Hendra. If Nipah virus hadn’t been zoonotic, capable of leaping into humans and killing them, Field told me, it might have passed as no more than “a blip on the productivity output” of Malaysian pig farming overall. “That’s an intriguing thought,” he added.

  I wasn’t sure, and neglected at that moment to ask, what intrigued him so about this alternate-universe version of Nipah. One possibility is that Field had in mind other potential zoonoses that are simmering, unrecognized, presently harmless to humans, among domesticated animals. How many such bugs may be working their way through large-scale livestock operations around the globe? How many RNA viruses may be achieving high rates of evolution (because they replicate quickly, they mutate often, their populations are big, and the herds are big too) in our factory farms? What are the odds, given such numbers, of a mutation that facilitates spillover? How many other Nipahs are slouching toward Bethlehem to be born?

  Maybe the Next Big One will emerge from a Malaysian piggery, travel to Singapore in exported sows, and then from Singapore to the world (riding airplanes, as SARS did) in the lungs of a tourist or a flight attendant who ate a lunch of mu shu pork at one of those trendy, overpriced cafes along the waterfront near the Raffles Hotel. Forget about palm civets, for a moment, and consider mass-production animal husbandry. It’s almost impossible to screen your pigs, cows, chickens, ducks, sheep, and goats for a virus of any sort until you’ve identified that virus (or at least a close relative), and we have only begun trying. The larger meaning of Nipah, in accord with Hume Field’s “intriguing thought,” is that tomorrow’s pandemic zoonosis may be no more than “a blip on the productivity output” of some livestock industry today.

  Nipah has other meanings too, not quite so large but also intriguing. One takes us back to the subject of bats.

  70

  After three weeks in Malaysia, Hume Field split away from the pig investigation and, along with a Malaysian veterinarian named Mohd Yob Johara and a few other colleagues, began searching for the origin of the virus. That was why he had been asked to join the international response team, after all—because of his experience in tracing a closely related virus, Hendra, to its reservoirs.

  Drawing on the Hendra parallel, Field’s little group now focused mainly on bats, of which Malaysia contains a high diversity, including thirteen species of fruit-eating bats and about sixty species of small insectivorous bats. Two of the native fruit bats are flying foxes, big animals with broad wingspans, belonging to the same genus, Pteropus, as the Hendra reservoirs in Australia. The small bats were caught using mist nets erected near their feeding and roosting sites. For flying foxes, the team used a more opportunistic method. Bat hunting is legal in most parts of Malaysia, so Field and Johara accompanied sport hunters into the woods and, with the hunters’ indulgence, took samples from bagged animals. Some hunters were shooting wild boar, so the researchers snipped bits from boar carcasses also, to test whether the virus had gotten from domestic pigs into wild ones. Another group from the international team, around the same time, sampled domestic dogs, rats, house shrews, chickens, ducks, and pigeons. Both groups wanted answers to the same urgent question: Where was this virus lurking in the bigger world beyond the piggeries?

  The wild boar, the rats, the shrews, and the birds all tested negative—no signs of Nipah nor of antibodies against it. Some of the dogs tested positive for antibodies, probably because they had been living closely with sick pigs or eating dead ones. The dogs didn’t seem to be spreading the virus much, neither from one canine to another nor to humans (though some evidence suggests that dog-to-human transmission did happen occasionally). Most of the bats tested negative, except for a few species, two of which stood apart from all others, showing significant prevalence of Nipah antibodies within their populations. Those two were the variable flying fox (Pteropus hypomelanus) and the large flying fox (Pteropus vampyrus). This wasn’t surprising, given the other similarities between Nipah and Hendra. But it didn’t constitute final proof of the bats as reservoirs. Antibodies merely suggested exposure, which could mean one thing or another, and the samples taken by Field and Johara didn’t yield any live virus.

  That task remained for Paul Chua, back in Malaysia following his mission to Fort Collins and Atlanta. Later in 1999, after the furor, after the 1.1 million pigs had been killed and the outbreak among humans stifled, Chua and his own team visited one of the flying fox colonies and tried a new technique. Instead of shooting bats and dissecting out tissues, they spread big plastic sheets beneath the roosting sites and collected a few precious drops of bat urine. Beneath the feeding sites, too, they collected samp
les—in the form of masticated fruit. Some of the fruit was mango; there was also a local delectable known as jambu air (in English, water apple). The water apple is an unprepossessing little thing, bell-shaped, usually pinkish or red, sweet and succulent enough to quench the thirst of children. Culturing those samples sedulously, Chua’s group grew three isolates of Nipah virus, two from urine and one from a gobbet of water apple. The virus closely matched strains found in Nipah-sickened humans. This proved that flying foxes are reservoirs of Nipah virus, capable of spilling it into pigs that spill it into people.

  But more. Chua’s work established a plausible scenario for spillover. How did the virus go from bats to pigs? All it required was a mango or water apple tree, laden with ripe fruit, overhanging a pigsty. An infected bat feeds on a water apple, discarding the pulp (as bats do), which is besmeared with virus; the pulp drops down among the pigs; one pig snarfs it up and gets a good dose of virus; the virus replicates in that pig and passes to others; soon the whole herd is infected and human handlers begin to fall sick. It wasn’t a far-fetched scenario. Amid the diversified agriculture of Malaysia at the time, wherein marketable fruit could supplement revenue from livestock, there were more than a few pigsties with mango, water apple, and other fruit trees growing nearby. Nipah virus may have been falling in sweet little packets. What pig could resist?

  71

  Malaysia acted firmly, tightening its agricultural regulations, closing some farms, getting the pigsties out from under the fruit trees, and producing a blitz of cautionary public education. Watch out for Nipah! Watch out for asthmatic swine! Still, eliminating all threat of this virus wasn’t so simple. Two years later, it re-emerged in Bangladesh, Malaysia’s regional neighbor, a Muslim country containing very few pigs.

  Bangladesh is at special risk from infectious disease outbreaks for several reasons, most obvious of which is the density of its population. Within its fifty-seven thousand square miles of territory it contains almost 150 million people, making it the most densely populated country in the world (apart from tiny city-states such as Singapore and Malta). Its generally low elevation (barely thirty feet above sea level in most areas) and its regular cycles of flooding (because of monsoonal rains and high rivers) exacerbate the problem of waterborne diseases such as cholera and bacterial diarrhea, which kill tens of thousands of Bangladeshis (especially children) each year. Although the numbers for Nipah are much smaller and the mechanism very different, the emergence of this virus in Bangladesh and the fact that (as you’ll see) it can sometimes be transmitted human-to-human have caused researchers and health officials to take the situation very seriously. Any infectious disease that achieves highly efficient airborne transmission might rampage through greater Dhaka (with its 17 million people), the other major cities, and the continuous crowded sprawl of villages to devastating result. And such a vast epidemic in Bangladesh, besides killing Bangladeshis, would also give the virus in question abundant opportunity to adapt still better to human hosts.

  The first Nipah outbreak in Bangladesh, during April and May 2001, occurred in a place called Chandpur, a village of six hundred souls in the southern lowlands. Thirteen people got sick, nine of them died, blood samples confirmed the presence of Nipah, and then the problem seemed to go away. People die all too frequently in Bangladesh, from one cause or another, and this cluster didn’t provoke any panic or rigorous investigation. From where had the virus come? Unknown. If bats again were the reservoir, what had caused the spillover? Unknown. Was there an amplifier host? Unknown. Pigs, anyway, weren’t implicated.

  When considered in retrospect by a team of epidemiologists, several years later, the Chandpur cases seemed to share only two risk factors worth mentioning. Some of the victims had lived with or cared for other victims, suggesting the possibility of person-to-person transmission, which was new. And more than a few of them had had contact with a sick cow. A cow? The epidemiologists’ published report, conscientious, exact, groping for leads, mentioned that animal several times. If the virus thrives in Malaysian pigs, couldn’t it flourish in a Bangladeshi cow? Maybe. The cow’s role remains undetermined.

  In January 2003 another outbreak began, up in Naogaon District, about a hundred miles north of Chandpur. Again febrile illnesses, befuddlement, encephalitis, hospitalizations, a high death rate; and no good explanation of how the virus arrived. One suggestive fact was that a herd of pigs had passed through the area, presumably attended by nomadic drovers, and some of the Nipah encephalitis patients had been exposed to them. Aha. Reports didn’t suggest that the pigs were sneezing and wheezing and stumbling and dying, as in Malaysia, but they may have been infected and infectious nonetheless. Disease scientists in Bangladesh were still puzzling over outbreaks one and two when the third began, in January 2004. It struck a couple of villages within Rajbari District, just west of the Padma River (an outlet finger of the Ganges), across from Dhaka. Again the case numbers were small, only a dozen; but ten of the dozen died. One other pattern in the data seemed curious: Most of these victims were children—boys, below the age of fifteen.

  Another squad of epidemiologists arrived, including an American named Joel M. Montgomery, on a postgraduate training fellowship with the CDC. They came with their clipboards and questionnaires and phlebotomy tools, as epidemiologists do, hoping to make sense of what had happened. They did a case-control study, meaning that they tried to identify the source of the outbreak, and its spread, by identifying behavioral differences between those who had gotten sick and those who hadn’t. What were the risky activities that made one a candidate for infection?

  Of course, young boys in Bangladesh, like young boys anywhere, engage in lots of risky activities, many of which could result in cracked skulls, broken arms, drowning, snakebite, getting arrested, or being hit by a train. But which kinds of risky behavior could give you Nipah? Montgomery and his colleagues ticked through some possibilities: fishing, hunting, touching dead animals, playing cricket, playing soccer, playing hide-and-seek, picking fruit off the ground and eating it. Among that list, as data accumulated, “touching dead animals” looked like it might be important; several of the sickened children, a week earlier, had helped bury some dead chickens and ducks. Evidently the kids had been acting out funeral rites with deceased poultry. Then again, more than a few uninfected village children had also touched the dead animals. The ducks and chickens turned out to be a false lead. See how tricky it is to do epidemiology in a Bangladesh village? None of those innocent childhood pastimes I’ve mentioned, from duck burial to cricket, was significantly more associated with the infected boys (whether recovered or dead) than with their healthy peers. But one was: climbing trees.

  Climbing trees? That was puzzling. Although the Montgomery group documented a strong correlation, their results didn’t explain why tree climbing might expose young Bangladeshis to Nipah infection. They could only make a calculated guess: It put the boys closer to bats.

  Three months afterward, in April 2004, health officials in Bangladesh learned of still another outbreak. Faridpur District, just adjacent to Rajbari along the Padma River’s right bank, was the latest site. Faridpur and Rajbari, reachable only by slow ferry, are where the urban clamor of greater Dhaka, groping upward in concrete and rebar, gives way to the silty, deltaic lowlands of southern Bangladesh. Rice paddies line the road. Palms and banana trees grow like weeds in a vacant lot. Among thirty-six patients in Faridpur, twenty-seven died. And the pattern of social connectedness among cases suggested another concern, which had also arisen regarding the Chandpur outbreak: that some people had caught the infection from other people. A team of investigators noted that such person-to-person transmission “increases the risk for wider spread of this highly lethal pathogen. In an impoverished, densely populated country such as Bangladesh, a lethal virus could rapidly spread before effective interventions are implemented.” Judicious language, by which they meant: It could go like wildfire in dry grass.

  Then came still another Bangladesh outbre
ak, the fifth within four years, this time in Tangail District, about sixty miles northwest of Dhaka. Twelve cases, eleven deaths, all during January 2005. Now it began to seem that Bangladesh was uniquely, persistently tormented by this killer disease, recurring in the early months of each year. Malaysia had had no further outbreaks. India, just north of the northwestern Bangladesh border, had had one. Elsewhere in the world, Nipah was unknown. Again a team went out from Dhaka and did a case-control study, looking for the cause of the spillover. The team leader was Stephen Luby, an American physician and epidemiologist from the CDC, seconded to Dhaka as a program director within the International Center for Diarrheal Disease Research, Bangladesh (fastidiously initialed as the ICDDR,B but commonly known as the Cholera Hospital), where he worked closely with his Bangladeshi counterpart from the Ministry of Health, Mahmudur Rahman.

  Luby’s group, like Montgomery’s earlier, questioned people about potentially risky activities—things done by patients who sickened and died, or sickened and recovered, that might not have been done by neighbors who remained healthy. For the fatalities, they got their answers from surviving relatives or friends. Had the person climbed a tree? Some had, most hadn’t, both among the patients and the healthy controls. Had the person touched a pig? No, nobody in Tangail was in the habit of touching pigs. Touched a fruit bat? No, nobody. Touched a duck? Yes, but so what, lots of people did that. Touched a sick chicken? Eaten a guava? Eaten a banana? Eaten an animal that was ill at the time of slaughter? Eaten a star fruit? Touched someone who was feverish, confused, and who later died?

  The questions themselves are like pen strokes on a sketch of Bangladeshi village life. But none of those questions—not even the one about tree climbing, this time—yielded any statistically significant distinction between those who had gotten sick and those who hadn’t. Only one question asked by the Luby team did: Have you drunk any raw date-palm sap recently?