Of birds and men
A deadly virus is brewing in Asia. Could this be the next killer pandemic?
Jean Taylor is trying to figure out how to live with a killer, one that could wipe out tens of thousands of people in the state of Maryland in just a few weeks. The epidemiologist is asking where sick people will be taken if the hospitals are overwhelmed; when schools and businesses should be ordered closed to reduce the spread of infection; and who will still be standing to deliver food to supermarkets, to pump gas, to bury the dead.
Across the country, federal, state, and local officials like Taylor are trying to figure out how the United States would cope with a killer flu, one that would be very different from the usual influenza that strikes each winter. A global epidemic, or pandemic, would be caused by a new, lethal flu virus, one to which people would have no immunity. The new flu would spread around the world within weeks and could infect one third of all people, killing 1 to 5 percent of them. That's what happened in 1918, when the Spanish flu killed 25 million in six months; some historians place the total killed at 100 million. Since then, two other far less lethal flu pandemics, in 1957 and 1968, have swept the globe. "Nature is the worst terrorist you can imagine," says Harvey Fineberg, president of the Institute of Medicine and former dean of the Harvard School of Public Health.
Flu viruses are notoriously good at mutating into new strains, and sooner or later one will morph into a mass killer. No one can say when that will happen; scientists say it could be this year, or 20 years from now. But in recent months normally sanguine health officials have been making increasingly dire predictions of a nightmarish 1918-style assault, one that could kill up to 2.2 million people in the United States. "We at WHO believe that the world is now in the gravest possible danger of a pandemic," says Shigeru Omi, Western Pacific regional leader for the World Health Organization. Those words echo Julie Gerberding, director of the federal Centers for Disease Control and
Prevention, who called the current situation "a very high threat."
The doctors are spooked by the continuing outbreaks of a new strain of avian influenza that has sickened at least 69 people and killed 46 in Vietnam, Thailand, and Cambodia in the past 16 months. One more death, of a 26-year-old man in Cambodia, was reported last week, and Vietnamese authorities are investigating a village where a 5-year-old fell ill with the disease in mid-March, shortly after his 13-year-old sister died. This new virus, called H5N1, is from the same family as the killer 1918 strain. Flu typically kills the old and weak, but with the new flu, most of the victims have been healthy young people. The 1918 flu killed the young and healthy, too.
The doctors are alarmed not because of the number of people that "bird flu" has killed but because the H5N1 virus displays an ominous adaptability and persistence. About 70 percent of those infected so far have died. Since 1997, when the new virus first showed up in chickens and killed six people in Hong Kong, it has spread to birds in eight countries in the region despite repeated efforts to halt it by slaughtering millions of chickens. "The virus has gotten even more widespread," says Klaus Stohr, head of influenza for WHO.
Of course, the world is not the way it was in 1918. This time around, humankind has weapons that can be deployed against a killer flu, rather than having to just wait and endure. We know how to track a virus's spread and work across borders to fight it--a technique that proved hugely useful in halting the spread of SARS in 2003. We know how to make antiviral medications and flu vaccines that could radically reduce the risk of serious illness and death. Last week, the National Institutes of Health launched tests of an experimental vaccine for H5N1 flu. New manufacturing techniques such as reverse genetics could also speed up production (box, Page 46). But if a killer virus strikes anytime soon, none of those 21st-century defenses will be ready. "We're clearly not adequately prepared," says Andrew Pavia, chairman of the pandemic task force for the Infectious Diseases Society of America. "About the most hopeful sign is that we have a lot of good and well-placed people who are beginning to get concerned."
All this casts harsh light on long-standing weaknesses in the nation's medical defense system, which include an unreliable vaccine supply and insufficient hospital surge capacity. Many hospital emergency rooms are already hard pressed to treat the influx of patients from a normal flu season, let alone a contagion that could send as many as 10 million people to the hospital. And the country's vaccine production capability is woefully inadequate, with only two manufacturers in the market and little financial incentive for other companies to enter. When contamination shut down one company's plant last year, obliterating half the nation's flu vaccine supply, the United States' entire influenza immunization program was thrown into chaos.
Under present circumstances, it will take at least six months from the start of a pandemic to get vaccine to the public. Global production is now approximately 900 million doses a year, enough for only about 15 percent of the world's population. Most of that is produced in Europe, and no one expects vaccine to be traveling across borders if things go bad. On March 17, the federal government asked vaccine companies to submit proposals to increase domestic production. The sole U.S.-based plant, in Swiftwater, Pa., can produce about 60 million doses per year. Many of the first doses produced would probably go to healthcare workers and emergency-service providers. "We have to get the message out loud and clear that vaccine will not save us," says Michael Osterholm, an infectious-disease specialist who heads the Center for Infectious Disease Research and Policy at the University of Minnesota. "We will have very little of it, and it will get here too late."
Meds and money. Federal officials say in the absence of a vaccine, antiviral medications like Tamiflu (oseltamivir), which can reduce the severity of flu symptoms and is effective against H5N1, would be the best option. Last week, the Infectious Diseases Society of America said the country should stockpile enough of the drug to treat 50 percent of the U.S. population, or about 150 million doses. Right now, the country has just 2 million doses stockpiled. Britain, by contrast, has ordered up 14.6 million doses, enough for 25 percent of its population (box, Page 48).
These sorts of advance preparations cost serious money--an estimated $100 million to build a new vaccine plant, for instance. Yet the call to arms comes at a time when perpetually lean local health departments have exhausted a few years' worth of federal bioterrorism funding that came their way after 9/11. State budgets are hurting, and the Bush administration proposes cutting funding for the CDC, which is leading much of the nation's antipandemic efforts, by $500 million.
Given that funds are scarce and much remains to be done, it may seem odd to hear health officials say that the United States should invest in the health of Asian chickens. The United States has already banned imports of poultry from countries with H5N1 outbreaks, to keep people from contracting the virus by eating infected birds, which may have happened in Asia. Earlier this month, inspectors in Cleveland and Columbus, Ohio, seized crates of boneless chicken feet that had been smuggled in from Thailand labeled as jellyfish. The shipments were part of 9,000 pounds of chicken feet that were distributed in 11 states.
Scientists also believe that if the rampant levels of H5N1 infection in Asian poultry could be reduced, it would also reduce the odds of the virus infecting people or mutating into a more lethal form. Flu viruses and other scourges often arise in animals, then infect humans. SARS, which killed 800 people in 2003, jumped from civet cats in live markets in southern China. H5N1 has already proved its versatility, infecting ducks, tigers, leopards, and humans. "We can talk about vaccine development, strength of surveillance, stocking of Tamiflu, and all that, but in the end the reduction of pandemic risk will be decided by the number of chickens infected in Asia," says WHO's Stohr.
Public-health officials in many countries are also closely watching the number of new human bird flu cases reported in Asia, hoping to get an early warning when a contagious mutation starts spreading. Although the conventional wisdom is that a pandemic can't be stopped, Thailand is starting to explore the notion of trying to contain an H5N1 pandemic, according to Scott Dowell, a CDC epidemiologist who heads the center's international office in Thailand. Last September, after a Thai woman contracted H5N1 from her daughter in the first documented case of human-to-human transmission, the Thai government launched a door-to-door search for new cases with almost 1 million volunteers. No other cases were found. But many researchers feel that Vietnam has been less than forthcoming about its H5N1 cases, and there has been an almost total news blackout in Laos and Cambodia, making it impossible to combat the spread of H5N1 there. The situation is reminiscent of SARS, when the Chinese government suppressed early reports of outbreaks until the virus had escaped its borders. H5N1 also might be getting harder to track. In at least one case, a child without respiratory symptoms who died from what appeared to be encephalitis was later found to be infected with H5N1. And recently two family members of H5N1 victims in Vietnam tested positive for the virus, despite the fact that they hadn't fallen ill.
In an effort to better predict when H5N1 could explode, scientists at the CDC in Atlanta are working in a high-level biohazard lab, mixing H5N1 genes with those of common flu viruses--a risky experiment, since that's exactly what could create a killer bug in nature. "The way to find out if you've really got a threat and Mother Nature just hasn't rolled the dice enough times is, you construct different gene combinations and you test their properties," says Nancy Cox, chief of the CDC's influenza branch. "You can learn a great deal about what might happen." Preliminary results won't be available for at least six months. "Science is often not as quick as we would like it to be," Cox says.
Indeed. The United States has been working on pandemic preparedness in various forms for a decade, but in the past year efforts have accelerated--49 states have drafted pandemic flu response plans, up from 29 a year ago. Although final plans are due to CDC by fall, most of the states are just now grappling with essential questions, such as who would get vaccines first. Healthcare workers and public-safety officers will probably top the list, says Richard Raymond, chief medical officer for Nebraska Health and Human Services. "That's a different mind-set than people are used to, and it's going to be a little bit controversial."
The Department of Health and Human Services unveiled a draft national pandemic plan last August; many reviewers said the feds need to be far more explicit, providing state and local governments with priority lists for vaccine distribution and other guidance. HHS is convening panels to revise the plan, with the aim of finishing sometime this summer. "This is one of those rare times when states are saying we really do need some direction and guidance from the feds," says George Hardy, executive director of the Association of State and Territorial Health Officials.
Public-health officials also have to maintain a delicate balance between sounding the alarm and overreacting. None of them have forgotten the swine flu debacle of 1976, when more than 40 million Americans, including President Gerald Ford, were inoculated against the swine flu virus in a campaign to head off a pandemic that never materialized.
Officials are also grappling with how to handle the economic and social disruption that pandemics cause. Pandemics last much longer than a hurricane or other natural disaster and typically hit in waves, with a first wave of infections followed by a second wave some three to 12 months later. If children fall ill, parents will have to stay home from work to take care of them, and business will suffer. Ditto if schools are closed to reduce the risk of infection. "We're talking about reducing morbidity and mortality, and maintaining social order," says Matt Cartter, epidemiology program coordinator for the Connecticut Department of Health. "There will have to be a balance. What are you going to do? Are you going to vaccinate sanitary workers so you still have garbage pickup?"
Although the 2003 SARS outbreak is estimated to have cost the global economy at least $30 billion, most businesses have yet to consider the cost of a flu pandemic, both in terms of employee absenteeism and disruptions of the global economy. The CDC estimates the economic impact of a pandemic in the United States at between $71 billion and $167 billion, but those numbers don't include disruptions to commerce and society. "We've never suffered an event of such magnitude that it shuts down the global economy," says infectious-disease specialist Osterholm. "In 1918 we were much more self-sufficient."
SPREADING THE BIRD FLU
H5N1 avian flu virus has turned up in humans who contracted it from chickens or ducks. But many believe it will be only a matter of time before it spreads rapidly from person to person. There are four scenarios in which H5N1 could evolve to become more dangerous.
Scenario A
1 The host passes H5N1 to an intermediate host--a pig, for example.
2 A person passes a human strain of influenza A to the same intermediate host.
3 When both viruses infect the pig's cell, the genes from the H5N1 mix with genes from the human flu and create a new strain.
4 The new strain can spread from the intermediate host to humans.
[drawing labels]
Host (duck)
Human host
H5N1 strain
Human strain
Intermediate host (pig)
New strain
Scenario B
H5N1 infects an intermediate host. There, it undergoes small genetic changes that make it more efficient at infecting mammals. This strain then infects humans.
[drawing label]
Host
Scenario C
Without undergoing genetic change, H5N1 directly infects humans. Small mutations accumulate that let it spread quickly from human to human.
[drawing label]
Host
HOW H5N1 KILLS
How the virus kills people is not completely understood, but lack of exposure to H5N1 means that most people have no antibodies to fight it. Unchecked, it could spread with devastating effects on the body.
Encephalitis
Brain swells, possibly inducing coma and death. (1 known case)
Typical symptoms
high fevers
severe muscle and joint pain
respiratory infections
Secondary infection
Flu weakens lungs, leaving damaged lining. Secondary bacterial infections can set in after the flu, causing pneumonia.
Immune system
The body mounts a strong reaction to H5N1 and overreleases cytokine, or chemical messengers, which tell immune cells to go to the lungs to fight the infection. When too many immune cells clog lung tissue, this "cytokine storm" damages the lungs, and the person suffocates.
Pneumonia
Infection or inflammation causes fluid buildup and impairs lung function. Infection spreading through the bloodstream can lead to death.
TRACKING THE VIRUS
So far, all known cases of H5N1 have been diagnosed in Asia. In the past year there have been 69 confirmed cases of the H5N1 virus; 46 of those infected, or 70 percent, died. Scientists are keeping a close watch on its spread, hoping to detect the first signs of a pandemic, or global flu outbreak.
[map of Asia]
All past and present H5N1 outbreaks in poultry: China; Hong Kong; South Korea; Japan; Laos; Malaysia; Indonesia
H5N1 detected in humans and poultry: Thailand; Cambodia; Vietnam
Human carriers?
Two known cases were diagnosed with no symptoms, suggesting that humans could become carriers of H5N1.
SCENARIO D: PANDEMIC
Scientists fear H5N1 will morph into a new virus with the deadliness of avian flu and the transmissibility of a human strain, allowing the new virus to spread quickly among people. How it happens: In a person simultaneously infected with H5N1 and a human flu strain, the two viruses can mix their genes, or "reassort," producing a new strain with the most dangerous traits of both.
[drawing labels]
H5N1
Human strain
New superstrain
DEVELOPING A VACCINE
H5N1's surface bristles with two types of protein spikes. One type, hemagglutinin (HA), helps the virus stick to a cell to cause infection. The other, neuraminidase (NA), enables newly formed viruses to leave a host cell.
A successful vaccine will cause the body to make antibodies against these two types of protein spikes on the surface of the virus.
Then the antibodies attach to the protein spikes of the invading H5N1 virus, disabling its ability to infect.
Antibodies created by a person's exposure to other flu strains will not protect against H5N1.
[drawing labels]
H5N1
HA protein spike
NA protein spike
Antibody
THE FLU OVER TIME
1918
The Spanish flu pandemic kills more than 500,000 people in the U.S. and as many as 50 million worldwide.
1957-58
The Asian flu pandemic kills 70,000 in the U.S.
1968-69
The Hong Kong flu pandemic takes about 34,000 lives in the United States
1997
For the first time, an influenza virus spreads directly from birds to people at poultry markets in Hong Kong: 18 are hospitalized, 6 die.
2003
In Hong Kong, 2 cases, 1 fatal. The sudden death of 19,000 chickens on one farm in South Korea leads experts to suspect an epidemic.
2004-March 2005
Cambodia: 1 case, fatal; Thailand: 17 cases, 12 fatal; Vietnam: 51 cases, 33 fatal. WHO says domesticated ducks could serve as a "silent reservoir" of H5N1.
Sources: CDC; NIH; Andrew T. Pavia, M.D.; University of Utah; World Health Organization; news reports
Written by Stephen Rountree & Nicole Schofer, timeline by Katy Kelly-- USN&WR
Graphic by Rod Little-- USN&WR
With Elizabeth Querna, Bill Bainbridge, Susan Brink and Nisha Ramachandran
This story appears in the April 4, 2005 print edition of U.S. News & World Report.