Is there a risk from eating food contaminated with bird flu virus?

Lurking on the margin of discussions about bird flu is the question whether oral transmission is a likely, or even possible, route of infection. The question has both medical and economic importance. Fear of eating infected poultry is one of the driving forces behind drops in poultry sales in countries where the disease has appeared. So there has been an intensive effort by the poultry industry, with help from government and public health leaders, to inform people this is not a danger:

Colin Blakemore, chief executive of the UK Medical Research Council, says the public need not worry. "There is no evidence of transmission to people by eating cooked eggs or chicken," he said on BBC radio last week, adding that the only food risk he could see was from "drinking swans' blood".

Blakemore's sound bite came a day after Britain's first case of H5N1 in a wild bird was confirmed - a dead swan found floating in a harbour in Cellardyke, Scotland. And it echoes a slew of recent reassurances by governments worldwide and by the World Health Organization (WHO), all conscious of damaging public confidence and the poultry industry. (Declan Butler in Nature)

Many scientists privately, and now publicly, are saying, "Not so fast." This, again from Declan Butler's Nature article:

But many flu scientists are concerned that, although the risks are low compared with those associated with contact with diseased birds, there is not enough evidence to say that the virus cannot be transmitted by eating infected poultry. "Oral transmission is an open question," says Masato Tashiro, a virologist at the National Institute of Infectious Diseases in Tokyo. "Direct evidence of oral infection is lacking, but so too is proof against."

Butler's piece reports the results of a risk assessment by the European Food Safety Authority (EFSA), published in its journal last week (.pdf here; hat tip reader Solitaire). Reading the EFSA monograph is interesting, although if you read it, don't forget to read between the lines as well as on the lines. Here's a short version, minus their rosy interpretation.

• We know the High Pathogenic Avian Influenza viruses, including H5N1, are present in high titers (high viral loads) in many edible tissues of a chicken, duck or turkey
  
  
• There is circumstantial but reliable evidence that other animals, especially cats (large and small), pigs and small carnivores like the stone marten may be infected through its food
  
  
• We don't know the route of transmission in humans, although most people expect it is respiratory and possibly oropharyngeal (throat)
  
  
• Some clinical descriptions of human cases include diarrhea and H5N1 RNA has been found in the intestines of two patients and via a rectal swab in another
  
  
• No epidemiology has been done that sheds light on the risks of oral exposure
  
  
• The gastrointestinal tract as a portal of entry cannot be excluded at this time

High viral titers can be present before poultry is obviously sick. H5N1 has been isolated from frozen Japanese imported duck meat and was able to cause disease when inoculated into experimental animals, demonstrating, as EFSA points out, that edible duck products could be infected even when the animals are not sick at slaughter. Some Highly Pathogenic viruses have been isolated from the surfaces, the whites and the yolk of eggs. The surface isolates might be fecal contamination, but the interior of the egg would have been contaminated from the systemic infection of the bird prior to laying. The feces and respiratory secretions of infected birds also contain high viral titers that might cross-contaminate surfaces and other ingredients during food preparation.

Vietnamese cases have also shown gastrointestinal symptoms and two cases following ingestion of raw duck blood are well known, so the question of infection by the oral route would still seem very much a possibility. To infect upon ingestion, the virus would have to find a cell with suitable receptors. Unfortunately we know little about the viral receptors on any cells except those in the respiratory tract (and those imperfectly), so whether there are ready targets for infection anywhere from the mouth to the anus is unknown. The mouth, throat, parts of the upper respiratory tract shared with the alimentary tract, and the esophagus all are traversed before the virus reaches the acid environment of the stomach. And whether the virus can withstand this environment is still a matter of debate:

Acid stability of AI [Avian Influenza] viruses appears to be a multifactorial issue in which not only the virus strain but also the virus quantity, the embedding medium, the pH value and the duration of exposure will play a role. Whether influenza viruses will be inactivated by exposure to the low pH of the gastric secretions after passing through the stomach is thus unpredictable and cannot be guaranteed. (EFSA Monograph, p. 22)

The EFSA document goes to some lengths to explain why it is less likely a human could be infected via the lower intestinal tract (after the stomach). Whether or not this is true, the question whether the lower tract is a portal of entry is only part of the primary question of whether ingestion is a risk. The entire GI tract starting at the mouth needs to be considered. The EFSA document provides little data in this regard.

As Butler points out in his new article in Nature, EFSA takes a decidedly "glass is half full" view of the available evidence, scanty as it is. EFSA's advice is that poultry products are safe to eat and have "not been implicated in the transmission of the H5N1 avian influenza virus to humans".

We've said it before when confronted with statements that have meager support and we'll say it again: Maybe.

Finally, we note the EFSA monograph assumes, as do others opining on this issue, that most human cases are from birds:

Almost all cases of AI virus infection in humans have been caused by highly pathogenic H5 or H7 viruses that were directly transmitted from infected birds to humans. While such human infections generally result from direct and intensive contact with infected or diseased poultry, other routes of infection such as consumption of edible tissues from infected avians or contact with contaminated water have been suggested as possible sources of infection." (EFSA Monograph, p. 10)

The kinds of bird handling implicated were slaughtering, defeathering (plucking), butchering and preparation of infected birds for consumption. History of contact was usually a week or so before clinical infection. Exposure to chicken droppings in areas where free range animals are kept and become ill has not been shown, but is a presumed risk factor. It is assumed that contact with infectious particles through the respiratory tract or conjunctivae (eyes) is the portal of entry in these poultry exposure cases.

On the other hand, the quality of medical history from many cases is poor. Often if infected birds have been seen anywhere in the neighborhood the case is assumed to be from contact with infected birds. But the EFSA monograph cites a personal communication from Peter Horby, WHO's epidemiologist in Vietnam, that one third of the Vietnamese cases have no history of direct exposure to poultry. In other countries, like Indonesia, many of the cases are so poorly investigated little reliance can be put on vague histories of poultry contact. Thus EFSA, along with many others (including WHO), would seem to have overstated this.

There's a lot of conventional wisdom about bird flu that will probably be shown to be wrong. Unfortunately no one is wise enough to know which.