WHO: "Prepare Now for Influenza Pandemic"

Avian Influenza: Assessing the Pandemic Threat was released in pre-publication form in January to elicit comments from the scientific and public health community.The report summarizes a 14-month saga of outbreaks of H5N1 in birds and humans, and analyzes developments in the virus and its ecology that strongly suggest a new flu pandemic could break out in the near future.

Starting in late 2003, a series of major outbreaks of lethal avian influenza have caused severe economic losses in a number of Asian countries. Of greater concern, the virus has also infected humans, with what appears to be a high mortality rate. As of early March 2005, 69 human cases of H5N1 had been officially confirmed, with 46 deaths.

According to the report, the H5N1 virus appears to have established itself as endemic in parts of Asia,with "a permanent ecological niche in poultry." Human cases continue to emerge, and "the virus may be evolving in ways that increasing favor the start of a pandemic," the report says.

Studies show that H5N1 has become both more pathogenic in poultry and hardier than in the past, with an ability to survive several days longer. It has also expanded its host range to include cat species and other mammals: last October, the virus sickened 147 captive tigers in Thailand that had been fed infected chicken carcasses.

Also alarming is the recent detection of highly pathogenic H5N1 in dead migratory birds. "Wild waterfowl are the natural reservoir of all influenza A viruses and have historically carried low-pathogenic viruses," the report notes. This suggests "the role of migratory waterfowl in the evolution and maintenance of highly pathogenic H5N1 may be changing."

At the same time, domestic ducks, which generally fall ill when infected with H5N1, have been found recently to be secreting large amounts of lethal virus without showing any symptoms of disease. This suggests that healthy ducks "play a role in maintaining transmission by silently seeding outbreaks in other poultry." It may also explain why some recent human cases cannot be linked to contact with diseased poultry.

During 2004, a number of large bird flu outbreaks on commercial poultry farms were successfully contained through culling and vaccination.

Health workers collect culled chickens in Hong Kong after an outbreak of avian influenza. H5N1 has affected millions of birds in nine Asian countries. But the virus has failed so far to pass easily between humans, the crucial development for which pandemic watchers remain on alert. Photo by Bobby Yip © Reuters/CORBIS

A greater concern now, says the report, is with outbreaks in rural areas of Asia, where most families keep free-ranging ducks and chickens. Outbreaks on these small family farms "may escape detection, are difficult to control, and increase the likelihood of human exposures, which may occur when children play in areas shared by poultry or when families slaughter or prepare birds for consumption."

H5N1's potential to ignite a pandemic depends, however, on its acquiring the ability to pass easily between humans. There are two ways this could occur. H5N1 could infect someone who also is infected with a human form of influenza A, and the two viruses could exchange genes (reassortment). Or, the virus could adapt in an evolutionary fashion during subsequent human infections, acquiring the ability to transmit itself efficiently from one person to another (adaptive mutation). The deadly 1918 Spanish flu pandemic is believed to have been caused by a virus that mutated in this way.

The report notes that opportunities for either of these events to occur continue to increase and appear more likely.

The WHO report also notes that a virus that gradually acquires an improved ability to pass between humans would be harder to detect through surveillance than a fully transmissible pandemic virus that emerged from a reassortment event. "The resulting explosion of cases would be difficult for any surveillance system to miss."

Past pandemics may hold some important lessons about how to prepare for and cope with any new pandemic.The report highlights several conclusions:

• Pandemics are as unpredictable as the viruses that cause them. Different pandemics have shown very different levels of mortality, severity of illness, and patterns of spread.
• All pandemics produce a rapid initial surge of cases, which then increase exponentially. This means "a sudden sharp increase in the need for medical care will always occur."
• The overall impact of a pandemic depends on its ability to cause severe illness in nontraditional age groups, particularly young adults. Milder pandemics have tended to target the elderly and the very young, those most affected by ordinary influenza.
• Pandemics tend to unfold in waves. Age groups and areas not affected initially are likely to be more vulnerable during a second wave, which is often more severe.
• Good surveillance is key to early detection of the onset of a pandemic. It makes it possible to alert heath services, isolate and characterize the virus, and make it available to vaccine makers.
• Surveillance capacity in Asia is particularly important, as this is where most pandemics have originated.
• Quarantine and travel restrictions have proved ineffective in stopping the international spread of pandemics. But within countries, banning public gatherings and closing schools are potentially effective measures.
• Stopping the pandemic's spread is extremely difficult, but slowing it can help health services cope by having fewer people ill at one time.
• In the past, vaccines have arrived too late and in too little quantities to have an impact. Nevertheless, the report emphasizes the importance of advance work to prepare the way for vaccine production once a pandemic virus emerges.
• Countries with a domestic manufacturing capacity will be the first to receive vaccines.
• The "best-case scenario" is a pandemic that has its main impact on the very old, the very young and the chronically ill, as these can be more easily targeted with vaccines. Still, health systems must anticipate a high demand for medical care.

The report notes that vaccine development and production will by necessity be primarily the responsibility of wealthier countries. However, many experts believe an effective vaccine against a pandemic version of H5N1 will have to be developed using "reverse genetics," a process that would result in a "genetically modified organism," raising biosafety issues that could delay production in Europe.

The report says it is "impossible to predict with any accuracy” the number of deaths that would result from a new pandemic. "Best case scenarios," modeled on the relatively mild 1968 pandemic, predict excess deaths ranging from 2 million to 7.4 million. However, "other estimates that factor in a more virulent virus, similar to that responsible for the deadly 1918 pandemic, estimate much higher numbers of deaths."