H5N1 asymptomatic infections, redux redux

It seems like every other story that comes out about H5N1 contradicts the previous one. I've blogged previously about some reasons to think that the diagnosed cases of H5N1 are only the tip of the iceberg (see here, here, and here, for instance). Though there I present some evidence to suggest that we may be missing asymptomatic or mild influenza cases, other stories have come to the opposite conclusion. For example, a recent news story from Cambodia reports that no mild or asymptomatic cases of H5N1 infection were found:

Researchers who tested 351 Cambodian villagers after they had extensive contact with avian influenza-infected poultry in 2005 found that none had antibodies to the H5N1 virus, suggesting that it doesn't easily spread to humans and that mild cases are rare.

"Our findings suggest that asymptomatic and mild H5N1 virus infections had not occurred in the population we investigated," says the report, published by Emerging Infectious Diseases. The study was done by an international team of scientists, with Sirenda Vong of the Institute Pasteur in Phnom Penh, Cambodia, as first author.

However, a new report from South Korea suggests that there have been several asymptomatic cases there:

Five more South Koreans were infected with the H5N1 bird flu virus about three years ago but none of them developed any serious illnesses, officials said on Friday after recently completed testing on old samples.

South Korea, which did not have comprehensive testing at the time, sent samples of 318 poultry industry workers taken during an outbreak in late 2003 and early 2004 to the U.S. Centers for Disease Control and Prevention in 2005 for further examination.

Of those workers, four South Koreans were infected, the CDC has said. The government said in February the four did not develop major illnesses.

The results prompted South Korean health officials to send samples to the CDC from another 2,109 people and of these, five were also infected, the health agency said on Friday.

There could be a number of reasons for the discrepancy, including differences in viral strain, effective exposure, viral load in the poultry they were exposed to, etc. So, there's still a huge amount of unknowns here. What's clear, though, is that we simply don't know enough yet to be able to say with any certainty that as a general rule, H5N1 is or isn't causing mild or asymptomatic infections in exposed individuals--and that certainty highlights a big, and critically important, black box.