Swedish scientists say that Tamiflu -- the frontline weapon in any bird-flu pandemic -- cannot be broken down by sewage systems and this could help the virus mutate dangerously into a drug-resistant strain.
Countries around the world are stockpiling Tamiflu in the belief it will help curb any future outbreak of H5N1 avian flu among humans.
Tamiflu, whose lab name is oseltamivir, is not a cure for flu but can ease its symptoms, thus aiding vulnerable patients such as the elderly, and reduce the time of illness, thus easing the burden on caregivers.
Scientists led by Jerker Fick, a chemist at Umea University, tested the survivability of the Tamiflu molecule in water drawn from three phases in a typical sewage system.
The first was raw sewage water; the second was water that had been filtered and treated with chemicals; the third was water from 'activated sludge,' in which microbes are used to digest waste material.
Tamiflu's active ingredient survived all three processes, which means that it is released in the waste water leaving the plant.
The finding is important because of the risk that Tamiflu, if overprescribed, could end up in the wild in concentrations high enough to let H5N1 adapt to this key drug, the authors say.
Flu viruses are common among waterfowl, especially dabbling ducks such as mallards which often forage for food near sewage outlets.
'The biggest threat is that resistance will become common among low pathogenic influenza viruses carried by wild ducks,' said co-author Bjoern Olsen, professor of infectious diseases at the University of Uppsala and University of Kalmar.
These avian viruses could then recombinate with ordinary human flu viruses, creating new strains that are resistant to Tamiflu, he said.
'Antiviral medicines such as Tamiflu must be used with care and only when the medical situation justifies it,' Olsen warned. 'Otherwise, there is a risk that they will be ineffective when most needed, such as during the next influenza pandemic.'
The study, published online on Wednesday by the open-access Public Library of Science (PLoS), pointed the finger at Japan.
It quoted figures from Swiss maker Roche, which estimated that in the 2004-5 influenza season, 16 million Japanese fell ill with flu, of whom six million received Tamiflu.
At such dosages, the amount of Tamiflu released into the Japanese environment is roughly equivalent to what is predicted in areas where the drug would be widely used in a pandemic.
Coincidentally, 'Japan also has a high rate of emerging resistance to Tamiflu,' the paper said. A 2004 study published in The Lancet found that among a small group of infected Japanese children, 18 percent had a mutated form of the virus that made these patients between 300 and 100,000 times more resistant to Tamiflu.