Abstract. Large-scale epidemiological surveillance of dengue in the field and dengue patient management require simple methods for sample collection, storage, and transportation as well as effective diagnostic tools. We evaluated the kinetics of three biological markers of dengue infection-non-structural protein 1 (NS1) antigen, immunoglobulin M (IgM), and IgA-in sequential capillary blood samples collected from fingertips of confirmed dengue patients. The overall sensitivities and specificities of the tests were 96% and 100%, respectively, for NS1, 58.1% and 100%, respectively, for IgM, and 33% and 100%, respectively, for IgA. During the acute phase of the disease, NS1 was the best marker of dengue infection, with a sensitivity of 98.7%, whereas from day 5, all three markers exhibited relevant levels of sensitivity. This first descriptive study of the kinetics of biological markers of dengue in capillary blood samples confirms the usefulness of this biological compartment for dengue diagnosis and argues for its exploitation in community-level and remote settings.

* RT-PCR = reverse transcriptase-polymerase chain reaction. † Laboratory-confirmed dengue cases based on positive NS1 or/and IgM patients reported by the laboratories. DENGUE VIROLOGICAL SURVEILLANCE USING FILTER PAPERfor DENV-1 (336 of 352), 4.3% for DENV-2 (15 of 352), and 0.3% for DENV-4 (1 of 352) ( Table 3 ).The first epidemic reported in Saint Barthélemy began at the end of October 2008 (Week 43), within 3 weeks after the dengue epidemic alert was issued for Saint Martin ( Table 3 and Figure 2 ). The biological surveillance network reported 152 cases, of which 62 of them were analyzed by RT-PCR ( Table 3 ). This epidemic was similar to the one observed in neighboring Saint Martin in two characteristics: i) DENV-1 was the predominant serotype (91.9%; 57 of 62), and ii) the epidemic ended in February 2010 (Week 07). This dengue outbreak was followed by a 39-week long inter-epidemic period of sporadic transmission.In mid-November 2009, another dengue outbreak took place in Saint Barthélemy. Although the two epidemics had similar durations (17 weeks in 2008-2009 versus 16 weeks in 2009-2010), 2-fold more laboratory-confirmed dengue cases were reported for the latter Saint Barthélemy outbreak ( N = 379) than for the previous one ( N = 152). One hundred seventeen NS1-positive blood samples were subtyped by RT-PCR: 114 were DENV-1 (97.4%) and 3 were DENV-2 (2.6%). Although the DENV-1 serotype predominated during this period in Saint Barthélemy, the DENV-2 serotype was found to predominate among specimens provided from the same period in Saint Martin. DISCUSSIONDengue has become a major public health burden in tropical and sub-tropical regions of the world, partly because of the increased number of reported cases, particularly severe clinical forms, and the increased number of deaths attributable to this disease. With no specific treatments available for patient management and no vaccine yet available to protect vulnerable populations, prevention of this arboviral disease plays a crucial role in its control. Among the different approaches implemented, an active surveillance system was advocated in the 1980s notably to detect dengue outbreaks early and to monitor the introduction of new dengue subtypes. 6,[25][26][27] We reported here the epidemiological usefulness of collecting venous blood samples on filter paper to monitor the dengue serotypes in circulation in two French territories in the Caribbean region: the islands of Saint Martin and Saint Barthélemy. This prospective approach to the virological surveillance of dengue serotypes was evaluated during a 27-month period. It allows detecting dengue serotypes using RT-PCR in 90.1% of overall blood samples provided from NS1-positive dengue cases ( Table 1 ).For samples collected during the acute phase of the disease, the percentage in which dengue virus RNA was detected by RT-PCR analysis was 95.5%. First, these results show the relevance of using, as a first line, such highly specific diagnostic tools for acute dengue diagnosis in areas where techniques for den...

Abstract. Two recent cases of human infection with Tonate virus, one of which was a fatal case of encephalitis, have renewed interest in these viruses in French Guiana. The clinical aspects of confirmed and probable cases of infection with this virus indicate that it has pathogenic properties in humans similar to those of other viruses of the Venezuelan equine encephalitis complex. To determine the prevalence of antibodies to Tonate virus in the various ethnic groups and areas of French Guiana, 3,516 human sera were tested with a hemagglutination inhibition test. Of these, 11.9% were positive for the virus, but significant differences in seroprevalence were found by age, with an increase with age. After adjustment for age, significant differences were found between places of residence. The prevalence of antibody to Tonate virus was higher in savannah areas, especially in the Bas Maroni (odds ratio [OR] ϭ 22.2, 95% confidence interval [CI] ϭ 15.2-32.4) and Bas Oyapock areas (OR ϭ 13.4; 95% CI ϭ 9.8-18.4). The ethnic differences observed in this study were due mainly to differences in place of residence, except that whites were significantly less frequently infected than other ethnic groups. This study indicates that Tonate virus infection is highly prevalent in French Guiana, especially in savannah areas.

No abstract