Death at the Corners

Late in the morning of Friday, May 14, a 19-year-old man was rushed to the emergency room of the Indian Medical Center in Gallup, New Mexico. The young man, a Navajo, was traveling into town with his family when he began struggling for breath in the back of the car. The family veered off the road at Thoreau, 30 miles east of Gallup, to call for help from a convenience store. An ambulance crew performed cardiopulmonary resuscitation on the way to the hospital, but neither they nor the emergency room physicians were able to revive him. A chest X-ray showed why: his lungs were full of fluid. The man’s sudden death was all the more shocking because--other than having some flulike symptoms the previous few days--he had obviously been healthy and athletic; doctors learned later that he was a cross-country track star.  Unlike many states, New Mexico requires unexplained deaths to be reported to a central registry, the Office of the Medical Investigator, in Albuquerque. All autopsies are performed in its labs. That Friday, emergency room physicians called in Richard Malone, the deputy medical investigator based in Gallup. Malone, who has worked for the OMI for 13 years, knows the community well.  What struck Malone immediately was that he had encountered a similar case at the same hospital only a month earlier. A 30-year-old woman with a seemingly mild, flulike illness had been rushed to the emergency room, suddenly unable to breathe. She, too, died rapidly, her lungs sodden. Like the young man, she was a member of the Navajo Nation living in the Four Corners region, where the boundaries of New Mexico, Arizona, Colorado, and Utah intersect. She lived outside Gallup, while the young man was a resident of the vast 25,000-square-mile reservation northwest of the city. Had the two victims not died at the Indian Medical Center in Gallup, which is run by the federally funded Indian Health Service, Malone might not have heard about them. The Navajo Nation is sovereign and self-governing and does not have to report deaths on Indian land to the OMI..  For the woman, who died in April, the cause of death was tentatively listed as adult respiratory distress syndrome--a term that simply labeled her illness without explaining it. Respiratory distress can result from many conditions, including heart failure, massive infection, and shock from a severe injury--none of which the victim had. Patricia McFeeley, the forensic pathologist at OMI who performed the autopsy, noted that the woman’s lungs weighed more than twice what they should have. They were filled not with pus, as would have occurred in pneumonia, but rather with the clear, yellowish fluid that makes up the plasma portion of blood. McFeeley ordered tests for every known infection that might have caused such a condition. All came back negative. Weeks after the autopsy the case continued to trouble her, and she went over it again with other pathologists. We were still stumped, she says.  Malone, too, had been troubled by the case. It bothered me every day, he says. I spent hours in conversation with her husband and family, trying to figure it out. Now, a month later, a second person had died in the same way. In the emergency room he approached the young man’s stricken parents, hoping for some clue to the illness.  What they revealed gave him chills. Their son had died on the way to his fiancée’s funeral, they explained. She had died five days earlier, with symptoms exactly like his. The couple had lived on the reservation with their infant son. But because the man’s fiancée had died on the reservation, the OMI had not been notified. That’s when I realized we had a crisis, Malone says.  The young woman’s funeral was to start 15 minutes later. Malone excused himself to call McFeeley. As he expected, she thought autopsies imperative for both the young man and his fiancée.  The difficult task of asking permission not just to perform two autopsies but to postpone a burial fell to Malone. In Navajo culture, great importance is placed on privacy and religious tradition. During the customary four-day mourning period after a death, the deceased is not even spoken of. Autopsies are discouraged, to say the least. Malone is Anglo, as descendants of English-speakers are called in the Southwest, but he’s a Four Corners native, and Navajo culture is not alien to him. He is a discreet, soft-spoken man himself, and it pained him to intrude on the grieving families. But against that Malone had to weigh the threat of a medical emergency. What if the couple had died of something contagious, or had been exposed to a toxin? Countless friends and relatives, including the baby, might be in danger. Autopsies, he explained to the families, were urgently needed. The families, he recalls, were very concerned themselves and most gracious in giving permission.  That Friday afternoon in Albuquerque, McFeeley prepared for a weekend of lab work. She also called the state health department in Santa Fe to report a possible outbreak in New Mexico of an unknown and deadly respiratory illness.  In the meantime, Bruce Tempest, chief of medicine at the Indian Medical Center, began making phone calls of his own. Tempest had hurried to the emergency room when the young man was brought in; he has been with the Indian Health Service for 25 years, and his advice is often sought in puzzling cases. He already knew not only about the mysterious death of the young man’s fiancée but also about the case in April, and about yet another case, in Arizona the previous November, that a colleague had called him about. Now when Tempest phoned Arizona for more information, he learned that still another person had recently suffered a similar fate. Within a short time, five young people from the Four Corners had died, all apparently from devastating lung infections.  Tempest couldn’t help wondering if all five had died of the same illness. He immediately worried about plague, which is endemic in the Southwest and which often involves the lungs. Desert rodents are infested with fleas that carry plague bacteria, and cases occur nearly every spring and summer. The trouble with that hypothesis was that neither the Arizona victims nor the first New Mexico victim had tested positive for plague. By late Friday night McFeeley had crossed plague off the list for the young Navajo couple too--along with influenza, Legionnaires’ disease, and anthrax. On Monday, when Tempest learned of McFeeley’s findings, he called the state health department. With plague and several other diseases eliminated, they were no closer to explaining any of the deaths--but the case had strengthened for suspecting that all five deaths had a common cause. A day later, on Tuesday, May 18, state health officials notified the federal Centers for Disease Control and Prevention (CDC), in Atlanta.  Working with the Indian Health Service and the Navajo Nation’s own health department, state health officials began a systematic investigation. They interviewed the families of patients who had died, visited their homes, and grilled doctors who had cared for them. On May 24 the New Mexico Department of Health sent a Dear Doctor letter to all physicians in the state, describing a cluster of unusual illnesses with high mortality (by then there had been seven victims, six of whom had died). All had begun with a flulike illness and ended with respiratory distress. The disease appeared to be infectious, the letter noted, and was probably viral.  On Thursday, May 27, the story broke in New Mexico. The Albuquerque Journal, a statewide newspaper, ran with the headline MYSTERY FLU KILLS 6 IN TRIBAL AREA. Over the next few weeks a dozen more cases turned up, most but not all among Navajos. The death rate remained high, even in those who made it to the hospital faster than the first victims had. One patient who sat up in bed in the morning, talking and eating breakfast, was on a respirator by the afternoon and was dead that night.  The national media jumped on the story. Reporters and film crews descended on the Navajos, photographing funerals, printing victims’ names, trying to question their bereaved families. Outsiders charged into the lives of Navajo families who had just lost their loved ones, paying little heed to the traditional four-day mourning period, says Duane Beyal, press secretary for Peterson Zah, president of the Navajo Nation. Local residents began posting NO MEDIA signs on reservation roads; resentment ran so high that some people refused to cooperate with medical investigators.  Worst of all, news reports branded the illness a Navajo disease. Before it became clear that the disease was not spread casually from person to person, fear of contagion loomed large--an excuse, some believed, to revive latent racism. A group of Navajo schoolchildren saw their trip to California canceled by their hosts. We had cases where Navajo people wouldn’t be served in a restaurant, says Beyal, or their meals were served on paper plates, or the waitress wore rubber gloves. According to the Albuquerque Journal, a car with New Jersey license plates was seen driving across the reservation--all its passengers wearing surgical masks. (Some Navajos meanwhile wondered if tourists had brought the disease to the reservation. There was even speculation that the Army might have stored biological weapons in its munitions depot at Fort Wingate, 20 miles south of the reservation, and that a mysterious germ might have escaped from its bunkers.) Within a month, however, it would be quite clear that this was a disease confined to neither Navajos nor the Four Corners. Anglo and Hispanic victims began turning up--in eastern New Mexico, nowhere near the Four Corners area, and in eastern Texas. Needless to say, by then the popular press had already moved on, leaving Navajos to deal with the stigma it had created.  By late May there were further alarming developments. Two relatives of the deceased Navajo couple were also hospitalized, desperately ill with respiratory distress. (They had lived in the same house as the couple; both subsequently recovered.) On May 22 a medical technician who had assisted McFeeley in the autopsy room came down with a sudden unexplained fever and muscle aches. The case was unnerving because autopsies expose health workers to unusually intimate contact with potentially contaminated tissues. Though the technician recovered from his mysterious, never diagnosed illness, the medical community was spooked.  Faced with an outbreak that seemed to be spreading, and without a clue as to what was causing it, state health officials sought help. That Memorial Day weekend, a task force of some 40 people--CDC officials, infectious-disease and toxicology experts, doctors who had treated local victims--convened in Albuquerque. The meeting was held at OMI’s headquarters at the University of New Mexico campus. Among those present was Toby Merlin, a pathologist from Albuquerque’s Lovelace Medical Center who specializes in identifying infectious agents.  It was a casual-looking group, he recalls. Only the visitors from the CDC wore ties; the rest showed up in jeans and shorts. But there was no mistaking the meeting’s sense of urgency. In a marathon all-day session, participants pooled information and searched for common denominators. First, recalls Merlin, we reviewed the clinical presentation of people who’d come down with this illness: high counts of white blood cells, low counts of platelets [the tiny cell fragments that help clot blood], fever, cough, and so on. We considered where patients lived, where they might have acquired the disease, any connection people had to each other. Did they share well water or a dwelling? Had they been to dinner together? Then we discussed possible causes, ranking them by how likely they seemed.  The most striking finding in all the cases was the patients’ fluid-filled lungs. Possible causes of lung disease were written on sheets of paper and taped to the walls. There were two major categories, Merlin says, toxins and infectious agents. The list of toxins included chemicals such as the herbicide paraquat; phosgene, a poison gas used during World War I; and heavy metals. But the pathologists had already looked for these substances in the victims and hadn’t found them. Possible infectious agents included everything from influenza viruses to the bacteria that cause plague and other animal-borne diseases such as brucellosis and Q fever-- some frankly unlikely on the basis of the clinical picture and pathology, not to mention previous tests, according to Merlin. In fact, none really matched the new disease. The one possibility that kept coming up, Merlin says, was a new agent.  That Saturday night, when the day’s sessions were over, Merlin went to his lab to examine all the available autopsy slides. At the next day’s meetings he gave his impression. Intracellular pathogens--organisms that actually enter the host cells--were at the top of my list, he recalls. These organisms include viruses, bacteria known as mycoplasma, and microorganisms called chlamydia. The tip-off was the presence of large, atypical white blood cells, or lymphocytes, in the lungs, lymph nodes, and spleen. The immune system makes these lymphocytes specifically to engulf and destroy cell-invading pathogens.  Meanwhile, in a nearby building on campus, Stephen Young’s virology lab was working around the clock, testing blood, tissue, and secretions swabbed from patients’ noses and throats. Although McFeeley had found no sign of influenza in the young Navajo couple, an aberrant form of influenza remained a prime suspect. Earlier in the spring, at the tail end of the flu season, a new strain of influenza A had turned up in New Mexico, causing an unusual number of pneumonias. My lab was called in to check the possibility, explains Young. When you’re confronted with an unknown disease you don’t automatically look for a new agent. You presume it’s something you know about and have diagnostic tools for. So we looked for the normal repertoire of respiratory disease viruses--first influenza, then adenoviruses, then cytomegaloviruses, and so on. But we were coming up with nothing.  As the weekend wore on, Young became edgy. I began to realize this might be something unusual, he says. His staff wore gloves, and they handled specimens inside a biological safety cabinet--an enclosure with continuously flowing air filtered to trap infectious particles. But when the agent is unknown, Young says, you have no idea what its capacity is for infection or transmission. Taking no chances, he ordered his staff to put on surgical masks, double gloves, and waterproof suits. He also began preparing specimens to ship to the CDC..  If there is a Marine Corps of infectious-disease fighters, the CDC’s Special Pathogens Branch is it. The viruses we work with, says the lab’s chief, C. J. Peters, are the ones that can take a perfectly healthy person and kill him in a matter of days. The most dangerous are those that spread through the air and for which there are no vaccines. Such agents must be handled at a P4 facility, in what amounts to room-size safety cabinets, by researchers wearing space suits and respirators. Before leaving, lab workers take a disinfectant shower with the space suit on, and then a soap-and-water shower without it.  Before starting work on the Four Corners samples, Peters reviewed the possibilities with his staff. A hallmark of this disease was the sudden flooding of patients’ lungs. The phenomenon, known as capillary-leak syndrome, resembles hemorrhage, but only the plasma of the blood oozes through the capillary walls. The fluid fills the air sacs of the lungs, blocking the uptake of oxygen and in effect suffocating the patient.  That made viruses that cause internal bleeding, so-called hemorrhagic fevers, theoretical candidates. The list included notorious, sure-death organisms like Africa’s Marburg, Ebola, and Lassa viruses, as well as Asia’s hantaviruses. On the other hand, no hemorrhagic fevers were known to be native to North America, and none of the patients had traveled overseas or been around foreign visitors. Nobody really felt that it was one of the organisms they were familiar with, Peters says. With the breadth of experience at CDC, that rings an alarm bell. You worry that it’s something new. Altogether, 25 diseases were tested for at the CDC’s laboratories.  Because viruses are vanishingly small, scarcely more than wisps of genetic material, detecting them in human tissues amounts to an incredible needle-and-haystack problem. It’s easier and faster to look for antibodies, telltale proteins churned out by the body in response to invading viruses. In addition, antibody testing provides a way to screen for the virus’s identity. The basic approach is to take samples of patients’ blood or tissue and expose them to bits of proteins, called antigens, belonging to various known organisms. If a person has been exposed to one of those organisms before--or perhaps to a related microbe-- that person’s antibodies will react to the antigen in a detectable way.  By June 4, investigators in the Special Pathogens Branch knew what they were dealing with. In all their tests, only one type of viral antigen had caused an antibody reaction. Furthermore, when they checked blood taken at intervals from the same patient, antibody levels to the suspect virus rose with time--the classic sign of a recent infection (it takes several weeks for antibodies to peak) rather than one that’s long since past. But the evidence that clinched it was a snippet of the virus’s genes. Researchers recovered it from a patient’s lungs by means of the polymerase chain reaction (PCR), a technique that probes for and amplifies minute amounts of genetic material. This snippet shared certain characteristic features with the pathogen under suspicion. And yet their gene sequences weren’t an exact match. That’s when investigators realized they had the new agent that kept coming to the top of their list. It appeared to be a previously unknown species of hantavirus.  Even though the CDC had looked for a hantavirus, finding one still came as a surprise. It was a long shot, says Peters. Hantaviruses are a well-known source of sickness in Asia and parts of Europe. They are carried by rodents, and each species of virus usually inhabits one species of rodent, without harming it. But while hantaviral infections were known to occur in North American rodents, no hantavirus had ever been proved to cause human disease in this country. Besides, all the known hantaviruses caused bleeding in the kidneys, not flooding of the lungs. Frankly, says Peters, the CDC tested for hantaviruses simply because they were the only hemorrhagic fever viruses known to cause the high white-cell counts seen in patients in the Southwest. And yet there was a certain irony to the finding: in 1992 hantaviruses were flagged by the Institute of Medicine, an advisory branch of the National Academy of Sciences, as potential emerging viruses--organisms that might spread from animals into the human population in this country and start wreaking havoc. But the institute’s report had in mind the Asian hantaviruses, not a native virus like this one. We never dreamed of a scenario like this, says Rockefeller University virologist Stephen Morse, one of the report’s authors. It would have seemed too close to fiction.  Talk to any scientist who has ever studied hantaviruses, and one name that’s sure to come up is Karl M. Johnson. Johnson, 64, now retired and living in Montana, is an infectious-disease specialist who spent most of his career abroad, chasing hemorrhagic fever viruses. He and two colleagues, Ho Wang Lee and Pyund Woo Lee, isolated the first hantavirus in Korea in 1976.  The disease he studied then was a severe form of what’s now known as hemorrhagic fever with renal syndrome. Some 200,000 cases of this serious illness occur each year in Asia, about half of them in China. Outbreaks coincide with harvest times, when farm workers disrupt rodent burrows and risk inhaling the virus in aerosolized rodent urine as they stir up clouds of dust. The disease starts with flulike symptoms and signs of small hemorrhages under the skin; internal hemorrhaging follows, and the victim goes into shock. Finally the kidneys falter. Death rates vary from 5 to 15 percent. Early research on this illness, also called songo, was carried out in the 1940s by the Japanese army in one of the more sordid episodes of medical history: for their work the Japanese investigators infected volunteers--reportedly Chinese prisoners of war. The disease first came to the attention of U.S. doctors during the Korean War, when 3,000 United Nations soldiers came down with the infection.  When Johnson came along, researchers had been trying unsuccessfully to isolate the virus from human patients. Johnson’s hunch was that the immune system mounted such a massive antibody response to the infection that it mopped up the viral particles and hid them from investigators’ tools. He also knew that rodents were thought to be the carriers. So instead of isolating the virus from humans, his team isolated it from its rodent carrier by mixing mouse tissues with antibody-laden serum taken from recovering patients. The patients’ antibodies, which were marked with fluorescent compounds, stuck to the virus, thus revealing its presence. The virus was named Hantaan, for a river in the region of Korea that the rodent specimens came from. Its subsequently discovered relatives became known as hantaviruses.  Hantaviruses are extremely tricky to grow. It took researchers another three years of cell culture work to produce a sample of Hantaan virus that would show up on an electron micrograph. From its structure it was immediately obvious that the virus belonged to the largest known family of animal viruses, called Bunyaviridae. These animal viruses have a quirk that may account for the family’s tremendous size--more than 200 species. Their genetic material consists of single-stranded RNA, rather than the usual double-stranded DNA, and their genome is divided into three distinct segments. With that genome you can get more recombinants than usual, Johnson says. That is, if two different hantaviruses infect the same cell, they can swap RNA and produce up to six new genetic types. It’s been shown in the lab that you can do this, Johnson says. There’s no reason to assume that nature isn’t doing it also.  In the early 1980s more hantaviruses came to light--notably the Seoul virus, a close genetic relative of Hantaan, which causes a milder form of hemorrhagic fever with renal syndrome. Seoul is carried by the Norway rat, which has stowed away on ships and established itself all over the world, taking the virus with it. Indeed, Seoul is found in cities in the United States--in the port city of Baltimore nearly half the Norway rats are infected--so the lack of hemorrhagic fever here is a mystery. (The first three cases were picked up only this year.) However, some researchers think the virus also contributes to chronic kidney disease and high blood pressure among inner-city dwellers exposed to rats. In the Journal of Infectious Diseases last March, virologist Gregory Gurri Glass at Johns Hopkins showed that inner-city residents undergoing kidney dialysis were five times more likely than the normal population to show signs of previous Seoul virus infection.  In Scandinavia in 1979 a mild form of hemorrhagic fever was traced to the Puumala virus, which is carried by bank voles. And as early as 1982 researchers from the National Institutes of Health began finding evidence of a hantavirus in U.S. rodents--though it did not appear to cause human illness. Viral antibodies were found in meadow voles from Alaska to Virginia to California, recalls Richard Yanagihara, one of the NIH team, which was based in Frederick, Maryland. Since meadow voles also lived around Frederick, the NIH researchers looked for a handy source of the virus in their backyard. Sure enough, they found it in voles on Prospect Hill, a nearby estate owned by their Nobel Prize-winning boss, Carlton Gajdusek, and they succeeded in isolating the virus by early 1983. Like the voles, the Prospect Hill virus appears to be native to North America. Genetically it resembles Puumala more than Hantaan or Seoul.  Ten years later, with the discovery of a previously unknown--and this time lethal--hantavirus in the United States, Johnson suddenly found himself called out of retirement. In mid-June he was flown to New Mexico to talk to researchers and meet with the victims’ doctors. In all his years of studying hantaviruses, he had never seen anything quite like it. If Hantaan is a hurricane, Johnson concluded after seeing how desperately ill the New Mexico patients were, this disease is a tornado.  Biologically, observes Johnson, the new disease is quite similar to its hantaviral brethren; capillary-leak syndrome underlies them all. In this new one, he says, it just looks as if the holes in the capillaries are smaller, so no red blood cells are escaping. You don’t see the hemorrhagic signs. But plasma protein and fluid leak out, so it’s similar to bleeding. Yet when it comes to clinical symptoms, the Four Corners disease is quite different. Somehow, he says, there’s a preferential attack by the virus on the capillaries of the lung.  One major unresolved question about this--and indeed all hantaviral infections--is whether the illness is caused by the virus alone or by the overwhelming way the patient’s immune system reacts to it. My own nickel’s bet is that it’s immunological, says Johnson. In other hantavirus infections there’s a torrent of antibodies, a massive immune storm. I think the explosive nature of this new disease, with all those crazy lymphocytes appearing in the blood, is compatible with that notion, too. If immune overkill is the problem, that might explain why very young children seem to be spared. (Patients have ranged in age from 12 to 70, with most in their twenties and thirties.) Children’s immune systems, Johnson speculates, may not be mature enough to muster the all-out, self- destructive attacks that their elders can launch.  Once U.S. researchers knew they were dealing with a hantavirus, they began looking for its rodent carrier. On June 7 teams from the Indian Health Service, the Navajo Nation health department, and the CDC began trapping rodents in and around the homes of Four Corners residents, especially in the homes of patients who had died. They set 100 traps--10 inside, 90 outdoors--at each site for two nights. Despite the risk, a collective decision was made not to put on protective gear during the trapping. We didn’t want to go in wearing respirators, scaring the hell out of everybody, recalls John Sarisky, an environmental-disease specialist for the IHS. To process the animals for CDC tests, however, the teams donned gowns, goggles, gloves, respirators, and shoe covers. Over a two-month period, some 1,900 rodents were dissected and nearly 10,000 vials of blood and organ samples were packed in dry ice and sent off to the CDC for analysis. By June 18, after little more than a week of work, the identity of the virus’s host was becoming clear. Twelve of the first 42 rodents examined had hantavirus, and all 12 were of the same species: the common deer mouse, Peromyscus maniculatus.  It’s hard to believe such an adorable little animal could cause so much trouble, says Robert Parmenter, a biologist at the University of New Mexico who monitors local rodent populations. With tawny fur, big eyes, long, dark whiskers, and an inquisitive snout, Peromyscus is indeed your basic storybook mouse. Unfortunately, by late June it was clear that some 30 percent of the tested deer mice harbored hantavirus. (In some parts of the country they are also a host for the corkscrew-shaped bacterium that causes Lyme disease. In September deer mice tested for Lyme bacteria near San Clemente, some 50 miles from the urban sprawl of L.A., were instead found to carry hantavirus.)  In June, as soon as a rodent carrier was suspected, New Mexico health officials began advising residents to mouseproof their homes. Though deer mice live in rural areas, they seem to regard the average human house, with its sloppy food-storage habits, as a big mouse house, says Parmenter. Of all Peromyscus mice, they are the most likely to enter human dwellings--it’s uptown, it’s condominiums for them. For mice already inside, health officials recommended trapping, then dousing the corpses with virus-killing disinfectant.  Deer mice have inhabited the Southwest for thousands of years; indeed, with the exception of the Southeast, they live all over the United States. Their relationship with their hantavirus is, biologically, ideal. Viruses can’t replicate unless they’re inside the cells of another organism; if they harm the host, they endanger their own chance of growing and spreading. Hantaviruses don’t seem to bother their rodent hosts. Infected, Peromyscus goes blithely about its mouse business, shedding copious amounts of virus in its urine and saliva, which in turn infect other mice. In fact, mouse and virus appear so well adapted to each other that they probably evolved together. The hantavirus may have existed quietly in North America for thousands of years--only causing trouble when humans blundered into the mouse-virus cycle. From the virus’s point of view, however, humans are a dead end; infected humans don’t shed virus in their urine and pass it on to anyone else. They probably become infected accidentally, because they’re unlucky enough to have a receptor--a bit of protein on the surface of some cells, perhaps those lining the tiny blood vessels in the lungs--that the virus can latch onto.  Ongoing genetic studies seem to support the idea that the Four Corners virus has existed in indigenous rodents for a long time. So far, the CDC’s analysis suggests that this strain is closer to Prospect Hill and Puumala than it is to the Asian viruses. That would certainly rule out that it’s a recent Asian import. And Brian Hjelle, a virologist who heads a sequencing effort at the University of New Mexico, thinks the virus could even predate Puumala and Prospect Hill, pushing back its origin even further. It may have branched off from an ancestral virus before Puumala and Prospect Hill diverged, he says. (The apparent age of these American viruses and their wide range have led Hjelle and Peters to discount the rumor that they escaped from Army research labs. If hantaviruses were recently introduced from a lab, comments Peters, you’d expect a local pocket of infection, not a broad geographic range. They wouldn’t have spread that fast. )  But if deer mice and hantaviruses have been around for so long, why has disease broken out only now? In over 35 years of looking for ‘new’ viruses, says Johnson, I’ve always found that it’s an old virus, and Johnny-come-lately man suddenly recognizes it. For that to happen, you need a cluster in one time and one place. Until then, doctors use arcane names, like idiopathic adult respiratory distress syndrome, to label sporadic cases they don’t understand.  The going theory for the Four Corners outbreak is that wet weather in the last year produced bumper yields of the seeds, nuts, berries, and insects that rodents eat, creating a mouse population explosion. When food is plentiful, a female can give birth to four or five litters, averaging four pups each, in a year. Pups mature enough to mate in less than two months, explains Parmenter, so in a good year, it goes geometric pretty quickly. Rodent census data aren’t complete, he adds, but in some areas in New Mexico there may be over 6,000 deer mice per square mile. Crowding and competition, in turn, may have increased the proportion of animals carrying the virus, because they’re more likely to do things that spread infection: mark territories with urine or bite each other. As the countryside becomes overrun, people are more likely to encounter mice, and infected ones at that. In Europe and Asia, hantaviral disease has been linked to rises in rodent populations.  Still, common sense would lead one to think there have been deer- mouse population explosions in the past. Indeed, biologists know so for a fact. And they think people probably died from hantaviral infections in the past, but no one knew what killed them.  There are, I’m told, about 50,000 cases of adult respiratory distress syndrome annually that go under the rubric ‘unexplained,’ says Peters. The CDC is now studying autopsy reports of people who died of the syndrome, and is using PCR to probe for the virus in their tissues. Most will not turn out to be hantaviral, Peters predicts, but the number will be more than we expected.  As a case in point, hantaviral disease was retroactively confirmed this August in a 58-year-old bridge inspector who died in Louisiana in June. Probes found the virus in his lung tissue. Interestingly, though, gene sequencing indicated that his strain was different from the one in the Southwest. That fits in with the general biology of hantaviruses: Peromyscus maniculatus doesn’t live in the Southeast. A different rodent (perhaps the cotton mouse, P. gossypinus) must be carrying the strain there. Since each rodent species usually carries a different hantavirus, the strain of virus in the Southeast should be different as well.  Still, some pieces of the puzzle seem to be missing. Why aren’t more people sick? asks Peters. After all, 30 percent of the rodents tested around the Four Corners are infected. I think something is going on with individual rodents or individual people that makes a match for disease. Certain rodents may shed more virus than others, he speculates, or some people may be more genetically susceptible to the disease. The phenomenon has been noted with other hantaviruses. You can have lots of infected rodents and relatively few infected people, says Peters. There will be sporadic cases, but then there are some clusters that you don’t know how to explain.  The greatest worry for doctors in the Southwest is that in its early stages this disease is nearly impossible to diagnose. The fever and muscle aches make it look just like the flu. By the time the diagnosis becomes clear, the patient is so desperately ill that it may be too late to help. All doctors can offer is what they call supportive care: treating symptoms, putting the patient on a respirator. A quick diagnostic test is urgently needed to help physicians intervene before the disease rages out of control. In other hantaviral illnesses, whose onset is more distinctive, prompt treatment with the antiviral drug ribavirin may lessen the severity. Though it’s not clear that it helps, ribavirin is being tried in patients who have or are suspected of having the Four Corners virus. As for vaccine prospects, a hantavirus vaccine is now being tested in humans by an Army research team led by virologist Connie Schmaljohn. That vaccine is directed against Hantaan, but a Puumala vaccine could be ready for trials by late 1994. Since the Four Corners strain appears related to Puumala, the vaccine may prove useful in the Southwest, says Schmaljohn. But that’s purely speculative--and of little comfort to Four Corners residents and physicians this winter.  The difficulty of diagnosing this illness becomes especially troubling as the winter flu season gains momentum. Doctors may find themselves facing patients who are short of breath, and wondering which, if any, are heading for respiratory collapse. To complicate matters, certain rodents--like deer mice--tend to invade people’s houses when the weather turns cold, potentially increasing the risk of hantaviral infection. This could become a nightmare, says virologist Young. It’s like being tied to the train track and seeing the train coming.  To avert such a crisis, researchers at the CDC and the University of New Mexico raced to develop a screening test. When they identified the new organism, Peters’s team at the CDC relied on antibody tests using antigens from other, known hantavirus species. But researchers suspected all along that those weren’t going to be sensitive enough to work for all patients with the new strain; antigens from the Four Corners virus were needed for a more precise test. To get them, researchers either had to culture vast quantities of virus or clone the virus’s genes and use them to produce the antigens. By late September, Hjelle and his New Mexico colleagues had gone the cloning route and mass-produced antigens for a prototype Four Corners test. His lab has been using the test, which gives results in 24 hours, since early October; a faster, commercial version of the test, as well as one from the CDC, was expected for December.  Better tests will help not only to diagnose and treat patients earlier but to track the disease’s true extent. By October, 60 cases of hantaviral disease had been reported--about half from the Four Corners states, the rest from California, Texas, Louisiana, Idaho, Nevada, the Dakotas, and Montana. Some 40 cases had been confirmed, and 25 of the patients had died. But the cases of the remaining patients--people who’d had all the classic symptoms of hantaviral respiratory distress--remained inconclusive.  One of them was Susan Thorpe, a 36-year-old Albuquerque woman. Thorpe, an Anglo, was hospitalized on May 10--before the outbreak was recognized--panting for air, her lungs filling with fluid. The week before, she’d been ill with what she took to be the flu. She survived her harrowing illness, but its cause remains a mystery. As tests specific for the Southwest strain become available, Thorpe and other survivors in limbo may find out what happened to them. If Thorpe tests positive, however, her case might sound an ominous note. She lives in one of Albuquerque’s better neighborhoods. She doesn’t garden and has never seen a mouse in her home. She hadn’t visited the Four Corners area before she got sick. She can’t imagine how she might have been exposed to rodent urine, except perhaps on one of the city’s softball fields. As of October, Thorpe remained on the CDC’s suspected victims list.  Puzzling cases like hers reflect some of the basic gaps in our understanding of hantaviruses. How much virus must a person inhale to become infected? Once inhaled, where does the virus go? Where does it replicate? When it comes to such key questions, Johnson says, we’re still in kindergarten. Peters ruefully agrees. He suspects that the Four Corners virus lands in the respiratory tract, multiplies there, then is spread throughout the body by the bloodstream. But we don’t know that, he adds.  In fact, says Peters, we don’t really know the answer to the most basic question of all: Why haven’t we noticed this disease before? As far as Peters is concerned, it’s too early to consider the issue settled. The emergence of AIDS in the 1980s starkly revealed our human vulnerability to new and completely unanticipated infectious diseases. Did it seem very likely, he asks, that a monkey virus existing for years in Africa would cross species and result in worldwide AIDS?  To find out where the Four Corners virus might be heading, Peters thinks researchers must determine where it came from. We can’t just say, ‘Gee, there was lots of rain and there were more rodents than usual, so we just noticed it this year.’ We’d be remiss in protecting society and predicting the virus’s future if we didn’t look into all the possibilities. Those possibilities include the chance that the virus has recently changed in some way that makes it more easily transmitted to human beings, or more easily spread among mice. It could even have jumped from one rodent species to another, bringing it closer to humans.  In the past such a jump would have been considered unlikely given the one-rodent, one-virus pattern classically observed with hantaviruses. But U.S researchers investigating a hantavirus outbreak near Belgrade recently discovered that the Puumala virus, normally transmitted by voles, was apparently being spread to humans by the common house mouse, Mus musculus. We don’t know what it means, says Peters, though he doubts that Puumala will become a permanent resident in house mice. Nevertheless the crossover is enough to keep the CDC on the lookout for hantaviruses turning up in new hosts.  Peters himself is inclined to discount this possibility. But he’d prefer proof to rule it out. Peromyscus maniculatus is distributed over nearly all the United States, he says. If the virus is spreading--either within that species or to others as well--it could become a national problem. I don’t want to be an alarmist, but we’d better find out what’s going on before we turn out the lights and go to bed. There’s a prowler in the house.