SINCE THE PUBLICATION OF MY ARTICLE “FEARING THE FLU” (FALL 2005), the avian virus has expanded its range among domestic birds, from Southeast and Central Asia into Asia Minor, the Middle East, Africa and Europe. Among humans, Indonesia and Egypt recently have suffered intense outbreaks. But Vietnam, once an epicenter of infection, has been free of human cases for at least six months. Unfortunately, reports indicate that despite vaccination, culling and surveillance programs, the virus has re-emerged among the people (killing two) and poultry in Thailand, also thought to be free of the virus.

It has become clear as well that migratory birds don’t effectively transmit the disease to domestic birds or continue to circulate it among themselves. There is no evidence in the wild of any bird species that remains asymptomatic and still sheds virus. And of the thousands of migratory birds that have been tested worldwide, none has been shown to be a carrier. Rather, the illicit poultry trade seems responsible for almost all the spread of the disease among domestic birds.

Why is it apparently so difficult for humans to contract and transmit H5N1? One possible explanation is that normal human flu virus tends to bind onto host cells with a particular molecular receptor, which ends with a sialic acid of type 2,6, as opposed to a “birdlike” receptor ending with sialic acid type 2,3. Humans typically have cells in their respiratory tracts containing 2,3 receptors, but those cells tend to grow deep in lung tissue, perhaps too deep to be coughed up.

In one instance, in Indonesia, there is molecular evidence that the virus did spread from person to person in one family (seven people died). What’s more, recent investigation has established that numerous mutations were observed in the Indonesian cluster, which probably involved two generations of spread. Though the changes represent increased adaptation to human hosts, these strains vanished with the deaths of their hosts. Recent research by the Centers for Disease Control and Prevention using a ferret model (the best approximation of human disease) demonstrates that recombining highly pathogenic H5N1 virus with common human strains does not readily produce a lethal, transmissible strain.

Finally, the degree to which people have contracted asymptomatic infections remains ambiguous. Evidence of infection without sickness has been seen in Turkey and elsewhere. But a survey of blood serum done in Cambodia indicates a very low prevalence of the disease in the bloodstreams of the surveyed population. If the virus has produced a significant number of asymptomatic infections, that would imply a much lower death rate (there have been 241 confirmed human cases and 141 deaths worldwide), and many more immune people, than it now appears. Conversely, if there are few mild infections, it would mean that fewer people have had the opportunity to build up immunity through exposure. That may seem an unsettling fact unless one considers, by the same reasoning, that if few people are getting the disease (whether in severe or mild form), it may, indeed, be very difficult to contract.