Classical dengue fever is characterised by fever, headache, retro-orbital pain, myalgia/arthralgia, skin rash, and leucopenia, usually with no deaths occurring. In contrast, DHF is clinically characterised by haemorrhagic diathesis and vascular leakage (manifesting as a rise in haematocrit), ultimately resulting in hypovolaemic shock. The pathogenesis of DHF is not fully understood, but it has been well documented that secondary dengue infection is a major risk factor of the disease (2, 3). As a consequence, and maybe also as a result of genetic factors, travellers rarely develop DHF (4). A high percentage of dengue infections in travellers may be asymptomatic (5, 6).

It has, however, been recognised that travellers have an important role in introducing more virulent dengue strains into endemic areas where usually only mild disease occurs (7), or into non-endemic areas where the mosquito vector is common (8). The recent introduction of Aedes albopictus to Europe, notably Italy, France, and Albania, may serve as a warning of things to come (9).

The European Network on Imported Infectious Disease Surveillance (TropNetEurop, http://www.tropnet.net/) was founded in 1999 to detect emerging infections of potential regional, national, or global impact at their point of entry into the European population. The network currently consists of 46 collaborating centres in 15 European countries. Annually, the collaborating centres give approximately 220 000 consultations prior to travel, and treat 63 000 patients post-travel (10, 11).

Within this network, the numbers of reported dengue cases increased from 64 in 1999 to 218 in 2002. The increase has been attributed primarily to the increased number of reporting sites of the network rather than to an increased incidence of dengue infections in travellers. The median age in our population was 32 years (range 1-69 years). Males were slightly more frequently affected with a male to female ratio of 1.2:1. The median duration of travel during which patients acquired the dengue infection decreased from 38 days in 1999 to 23 days in 2002.

Forty five per cent of dengue cases were acquired by patients who had travelled in countries in South East Asia, 19% were imported from South and Central America, 16% from the Indian subcontinent, 12% from the Caribbean, and 8% from Africa. This distribution mainly reflects worldwide dengue activity, as well as the popularity of these countries as tourist destinations. Thailand, Vietnam, and Indonesia are not only highly endemic areas for dengue viruses; they are also countries with an expanding tourism sector. Thailand alone was the place of acquisition for 134 cases (28%) of all travel associated dengue infections in our network over the past four years. In 2002 there was a proportional rise in the number of infections acquired in South East Asia compared with 2001. Reporting increased most notably from Thailand, Indonesia, and Brazil, while India and the Philippines saw a decline in numbers. There was a clear seasonal pattern in the reporting of dengue with peaks during the travel seasons in summer and winter.

Most patients were European-born travellers (94%), but we noted that immigrants born outside Europe and foreign visitors had a 4.3 times higher risk of developing DHF when compared with those born in Europe. The reporting system does not include information on the countries of residence of non-European travellers, but many are likely to be from dengue-endemic countries. Overall, 13 cases of dengue DHF have been reported (2.7% of all dengue infections). The travel history of these patients showed that infections were acquired in all the world's endemic regions: seven in South East Asia, two in Central America, two in South America, one on the Indian subcontinent, and one in Central Africa. The median travel duration in patients with DHF was 29 days and was not significantly longer compared with patients who developed classical dengue fever (26 days, p-value 0.38). No deaths occurred among the reported dengue patients (DF and DHF).

Three fatal cases of dengue in Europe have been reported to the network from sources outside the immediate membership. The first case occurred in 1997 in the United Kingdom (UK) and was not published. The other two patients died of DHF in 2002; one was a Bangladeshi immigrant living in the UK who returned from visiting Bangladesh (12), the other was a young Finnish journalist travelling in South East Asia (publication in progress).

Overall, in our network we saw an increase in the total number of reported dengue cases, an increase in cases with more severe disease (DHF), and an increase in cases with unusual clinical findings (several cases presented with liver involvement and two cases with affected optic nerves(13)). As long as no effective vaccine is available, dengue viruses will present an increasingly serious threat to European travellers and an even greater threat to those living in the countries to which they travel.