Epidemics of dengue haemorrhagic fever in Thailand occur in a predictable three-year cycle that spreads outwards from the country’s capital city, according to a team of Thai and US researchers. Being able to anticipate epidemics in this way could guide the allocation of public-health resources needed to contain future outbreaks.

The researchers analysed 850,000 infections occurring in more than 70 of Thailand's provinces between 1983 and 1997. They discovered a three-year cycle during which the disease, after originating in Bangkok, spreads outwards at a speed of on average 148 kilometres per month until it reaches the country's outlying provinces.

"We had not expected to find that nearly the whole country is affected by the wave," says one of the researchers, Donald Burke of the Johns Hopkins Bloomberg School of Public Health in Baltimore, United States. The study was published in this week's Nature.

Some areas near Thailand's border, however, did not fit the pattern of infection. The researchers speculate that infection in these provinces was influenced by other urban areas in South East Asia.

Dengue fever, a mosquito-borne virus, infects 50 to 100 million people each year. Of these, up to half a million infections occur as the severe, life-threatening form of the disease, known as dengue haemorrhagic fever.

Large, unanticipated epidemics of the haemorrhagic form often overwhelm healthcare systems. Burke says that their results suggest that high priority should be placed on surveillance systems in urban areas of Southeast Asia.

Michael Nathan, a scientist working on communicable diseases at the World Health Organisation, says that his organisation is trying to encourage governments to deal with the disease in a preventative rather than a reactive way. "Anything that helps to predict a major epidemic is important," he says.

Similar waves, moving in a predictable pattern over time and geography, have been observed in animal ecological systems. However this is the first time that such a pattern has been observed in a human vector-borne disease such as dengue.

Burke's research team used a statistical technique recently developed by the US space agency NASA. "We were surprised that a technique devised for analysing waves in water and sound can actually be applied to epidemiology," says Burke. "And that it can tease out both the temporal and spatial waves present in the [epidemiological] data."

A second study published in the same issue of Nature sheds light on how the dengue virus infects host cells. A virus enters the cell of a host by fusing its membrane coat with that of the cell. However, scientists have had limited information about the protein that inserts into the target membrane to enable fusion.

US researchers have now described the crystal structure of this protein after membrane fusion. This enabled them to model how the protein interacts with the target membrane and to propose a mechanism of fusion. They believe the research may help in developing drugs that stop such viruses entering host cells.

Link to research paper in Nature: Travelling waves in the occurrence of dengue haemorrhagic fever in Thailand

Link to research paper in Nature: Structure of the dengue virus envelope protein after membrane fusion

Reference: Nature 427, 344 (2004) / Nature 427, 313 (2004)