EM image of a cluster of rotavirus in diarrhoeal faeces/copyright - Ruth Bishop

Rotavirus accounts for more infant deaths from diarrhoea than any other agent, yet the future of rotavirus vaccines remains uncertain because of an enigmatic link between the virus and a rare bowel complication that has hampered attempts so far to introduce the vaccines worldwide.

The decision to forge ahead may depend more on the price we decide to pay for protection, rather than on vaccine efficacy, warns the microbiologist who led the team that discovered rotavirus in 1973 in Melbourne children admitted to hospital with acute gastroenteritis.

"Ultimately we have to put a price on babies, and this will involve negotiations on the costs of vaccination," said Ruth Bishop, research fellow of the Murdoch Children's Research Institute, Royal Children's Hospital, Melbourne, and director of the WHO Collaborating Center for Reserach on Human Rotaviruses.

In developing countries, 70 children die every hour from severe diarrhoeal diseases, against 50 per year in the US, and rotavirus is the single most important pathogen involved.

Yet concerns over its association with a rare complication known as intussusception, a form of bowel blockage, is preventing the first effective vaccine from being introduced to developing countries, says Bishop.

In 1999, the new RRV-TV vaccine, based on a live rhesus monkey-derived rotavirus, was withdrawn after the US authorities discovered a rise in the number of children with intussusception, including one death. Since then, says Bishop, there has been "quite bitter dispute" over whether or not the vaccine should have gone through further trials.

But whether or not the intussusception was a problem peculiar to the rhesus rotavirus strain, or will also occur with other strains used in future vaccines, remains unknown. "We're facing an enigma: is interssusception going to be a consequence of any live oral rotavirus vaccine? Is it indeed going to be a consequence of any live oral vaccine, not necessarily just rotavirus?" Bishop asked.

"We need now to know the risk factors for intussusception in developing countries and whether or not natural rotavirus can be associated," Bishop continued.

The alternative option may be to focus on injectable DNA-based vaccines, which may avoid complications but may not induce sufficient immunity at the site of virus infection: the intestinal luminal surface, for which there is now at least one candidate vaccine.

Two new live vaccines are also currently in clinical trials - WC3, made by Merck, consisting of an artificially created bovine/human hybrid strain, and another produced by Glaxo Smithkine.

"I'm on record in 1974, in response to a journalist question, 'how long will it take to get a vaccine', for saying 'five years', and here we are, nearly 30 years later, trying to get a vaccine that is safe and effective," Bishop said.

The RRV-TV vaccine contained each of four main serotypes, which carry antigenic differences, although they also have some antigens in common. Now, a fifth serotype, originally discovered in Japan in the 1980s, is growing in prevalence worldwide, which will also need to be covered by a successful vaccine.