Hong Kong Flu Still Poses Pandemic Threat

Hong Kong Flu Still Poses Pandemic Threat


JAMA: The Journal of the American Medical Association Vol. 288 No. 19, November 20, 2002


Brian Vastag

WashingtonIn December 1997, news watchers were confronted with sobering images of government workers in Hong Kong slaughtering 1.4 million chickens in a massive effort to prevent avian influenza from spreading through the densely populated city and, potentially, across the globe.

Earlier that year, what was later identified as a strain of influenza previously known to infect birds, but not people, had sent 18 patients to Hong Kong hospitals. The first case led to the death of a 3-year-old boy, followed by the deaths of five other patients.



Chickenson their feet and on the rackfor sale at a shop in Hong Kong. In June 2001 and May 2002, a total of 4 million chickens were destroyed there to stanch the potential spread of influenza. (Photo credit: Corbis)

Somehow the virus had jumped species. Working with the US Centers for Disease Control and Prevention (CDC), Hong Kong health officials began implementing a control plan designed to obviate such drastic measures in the future.

It didn't. In June 2001, and again in May 2002, a total of 4 million chickens in Hong Kong were destroyed to stanch potential pandemics. Both cullings were precipitated by detection of influenza virus in Hong Kong poultry, said John Siu-Lun Tam, PhD, a professor of microbiology at the Chinese University of Hong Kong, during a scientific session at the meeting of the World Medical Association held here in October.

And although angst over the specter posed by bioterrorism dominated the meeting, a CDC official warned that the threat of an influenza pandemic is still "very real." Such a worldwide outbreak would dwarf the influenza epidemics that sweep through the United States every winter, said Nancy Cox, PhD, chief of the influenza branch of the National Center for Infectious Diseases at the CDC.

While a typical flu season claims about 20 000 US lives annually, a pandemic could provoke 207 000 deaths, 734 000 hospitalizations, and 42 million outpatient visits, estimated Cox and CDC colleagues Martin Meltzer, PhD, and Keiji Fukuda, MD (Emerg Infect Dis. 1999;5:659-671).

"It would be extremely overwhelming," said Cox.




EMERGING STRAINS



The worst influenza pandemic of the 20th century (in 1918) spread so quickly that it claimed 20 million lives worldwide. The other two influenza pandemics of the past century, in 1957 and 1968, led to many fewer deaths, but still killed 70 000 and 34 000, respectively, in the United States alone. And while the winter flu preys mainly on the elderly and others with weakened immune systems, pandemics lead to a higher proportion of deaths among children and young adults, said Cox. Most of those killed in 1918 were working-age adults.

Methods to predict the where and when of a pandemic have not advanced appreciably since the 1918 pandemic. Cox warned that new strains can emerge suddenly and unpredictablytwo factors that differentiate a pandemic (extensive emergence of a virulent new strain) from an epidemic (localized circulation of a known strain).

The 1997 Hong Kong outbreak caused such intense concern not only because the strain was new, but also because it apparently jumped directly from poultry to people. In the past, influenza watchers thought that to infect people, any new avian strain would first have to spread through herds of pigs. Susceptible to both bird and human influenza, the animals act as incubators of virulent strains, said Cox. As the porcine immune system fights infection, avian and human viruses shuffle genes in a sort of natural genetic engineering that can breed deadlier varieties.

But since people in Hong Kong prefer poultry to pork, pigs were not suspected in the 1997 "bird flu" cases. Instead, epidemiological investigations isolated domestic birds as the common link. The index case, a 3-year-old boy, often played with pet ducks. Twelve of the other 17 patients had handled poultry in the week prior to becoming ill. A recently published study confirmed the potential for chicken-to-human transmission, finding antibodies to the 1997 strain in 10% of 1523 Hong Kong poultry workers (J Infect Dis. 2002;185:1005-1010).

Conditions are ripe for nonoccupational exposure as well, said Tam. "There is a strong emphasis on freshness," he said, and customers like to pick up and inspect live chickens at the 930 poultry markets in the city. "They touch the neck and blow on the other end to see how good it is," he said, holding up and turning his hands as if looking at the bird's rear. Prior to 1997, Tam said, conditions at the markets were unsanitary, with feathers and feces abundant.

Since then, the city's health department has mandated new hygiene and safety standards. Ducks and geese, which serve as the natural reservoir for the 15 known strains of influenza A, are now shipped and slaughtered separately from chickensto interrupt the first link in the pandemic chain, said Tam.

New surveillance measures at the Chinese border, where 85% of the poultry consumed in Hong Kong originates, have also taken effect. Each arriving flock is quarantined, tested, and held for 2 days. Flocks with one or more sick birds are rejected while clean flocks move to a central wholesale warehouse, where they wait 2 days more.

"Four days is the time it takes a chicken to die" from influenza, said Tam. If a bird is still flapping at the retail market, it is probably not infected.

On the farms where the rest of Hong Kong's chickens come from, any sign of infectionsick birds, positive antibody testsleads to a quick slaughter and ring vaccination of birds at surrounding sites.

Despite these measures, which have gradually been phased in over the past 5 years, the health department again resorted to massivealthough not massively publicizedcullings in 2001 and 2002. Questioned by a meeting attendee about this, Tam said that the prevalence of the virus type detected in 1997 is too widespread to eliminate completely. "We see this as a problem for the region, not just Hong Kong," he said. "We tell people that we cannot eliminate [the virus], that we are trying to reduce the source to prevent human cases."




INCREASED SURVEILLANCE



Fortunately, the 1997 virus apparently never acquired the ability to spread from person to person, although a shred of evidence from a study of household contacts of patients suggests that such transmission is possible. Researchers found antibodies against the bird flu strain in a household member who had no contact with poultry (J Infect Dis. 1999;180:1763-1770). Concern that a mutation fostering airborne human transmissibility could arise, making the final link in the pandemic chain, has the Hong Kong health department, the CDC, and the World Health Organization constantly searching for methods to improve surveillance, detection, vaccination, and treatment.

The region that includes southern China and Hong Konga historical, although altogether not well understood, source of new influenza variantsis high on WHO's hot list. With WHO's support, the city's surveillance system, widely credited for halting the 1997 outbreak, has been expanded and now includes 111 sentinel physicians.

In 1999, the sentinels and other local hospitals and clinics sent 17 351 specimens to a central laboratory for influenza testing. The health department collects and disseminates influenza information, including reports on extant subtypes, on a weekly basis.




BLIP ON SURVEILLANCE RADAR



In March 1999, the system flagged another new influenza strain found in samples from two sick children, although the source was never identified, said Cox. A cohort study found no evidence of person-to-person transmission, but an epidemiological investigation revealed that 23% of one group of poultry workers tested positive for exposure (Emerg Infect Dis. 2002;8:154-159). The strain was found in Asian poultry and pigs, but has not reappeared in Hong Kong's clinics. This odd blip on the surveillance radar is a testament to influenza's unpredictability.

That blip also prompted the CDC to develop vaccine for the strain; Cox said that three candidate vaccines are approaching human trials. But even while the development of a vaccine targeting this novel strain is moving forward, vaccine production remains the "Achilles' heel of pandemic preparation," said Cox. With current technology, manufacturers need 8 months to produce 250 million doses of trivalent influenza vaccine, which typically comprises the three strains that experts believe (based on surveillance of strains circulating worldwide) constitute the most likely threat for the following winter.

Cox guaranteed that in the event of a pandemic, which could spread globally in a few months, "there won't be enough [vaccine] to go around." She called for the development of "a library of candidate vaccines on the shelf, ready to go," as well as expedited research into advanced production methods, such as cell-based cultures. In an age of sequenced genomes, influenza vaccine is still produced relatively crudely, inside incubated chicken eggs.

Until that method is replaced, then, Cox suggested an answer to the public health community's version of the hoary chicken-and-egg riddle: stockpiles of egg vaccine first, before another outbreak of chicken virus.

http://jama.ama-assn.org/issues/v288n19/ffull/jmn1120-1.html

Tweet