Factors of severity at admission during an epidemic of dengue 1 in New Caledonia (South Pacific) in 2003.

We report a retrospective study of an epidemic of dengue in New-Caledonia (South Pacific) in 2003 among adult inpatients. The aim was to establish clinical and biological criteria for the severity of the infection at the time of admission. During 7 months, all inpatients older than 15 y having a laboratory-confirmed diagnosis of dengue fever (IgM or PCR) were included (n=170). Two groups were defined: severe cases (death and/or transfer to intensive care unit, n=24) and benign cases (n=146). Data were analysed using Epi-Info software. Univariate analysis showed that shock, haemorrhage and neurological complications were significantly more frequent in serious cases, respectively 37.5% vs 0.7%, 62.5% vs 32.2%, 25% vs 9.6% (p<0.05). Relevant biological criteria were: creatinine > 140 micromol/l (OR 12 (95% CI 3.93-37.44)), free bilirubin > 18 micromol/l (OR 12.69 ( 95% CI 2.88-59.5)), amylase > 220 UI/l (OR 27.34 (95% CI 4.57-210)) and platelets < 45,000/mm3 (OR 4.35 (95% CI 1.43-14.2)) with p<0.005 (VPP = 100% for association of 3 criteria). We suggest this combination of 4 biological criteria inclines to severity.

Prospective study of the duration and magnitude of viraemia in children hospitalised during the 1996-1997 dengue-2 outbreak in French Polynesia.

The magnitude and duration of viraemia in children admitted to the hospital with dengue was studied during a dengue 2 outbreak in French Polynesia in 1996-1997. Forty-nine patients from whom at least 3 plasma samples were available were included in the study. Based on analysis of IgG-ELISA and haemagglutination inhibition assay, 21 of these were primary and 28 were secondary infections. According to World Health Organization criteria, 42 were dengue fever and 7 were dengue haemorrhagic fever. Virus was detectable by reverse transcription-PCR in all patients for at least the first 3 days of the onset of fever, but was never detected after the 6th day (mean duration = 4.4 days). Plasma virus titers ranged from 1.7-5.6 Log(10) TCID(50)/ml. A significant difference was not observed in the magnitude and duration of viraemia in patients with primary versus secondary infections. The severity of the illness, however, was correlated with both criteria.

Dengue: an evaluation of dengue severity in French Polynesia based on an analysis of 403 laboratory-confirmed cases.

We conducted a retrospective study of 403 laboratory-confirmed dengue cases hospitalized in Tahiti between August 1989 and March 1997. According to standard WHO criteria, 337 of these cases were dengue fever (DF) and 64 were dengue haemorrhagic fever (DHF). Of the 10 fatal cases, 6 were DF and 4 were DHF. As an alternative, we used a correspondence analysis procedure to define dengue severity based on basic clinical and biological criteria for which we assigned a severity score, and then selected the 50 most severe cases from this analysis. Of the latter, 17 patients had been classified as DF and 33 as DHF by the WHO criteria. From this analysis, haemorrhages and decreased platelets counts associated with hepatic disorders are the main criteria associated with the severe dengue cases. Thus in our study population, the WHO classification does not account for the overall severity of dengue; hepatic failure should be considered as a specific severe form of dengue since plasma leakage, which is the pathophysiological hallmark of DHF, is only one of the pathogenic mechanisms leading to severity.

Variation in oral susceptibility to dengue type 2 virus of populations of Aedes aegypti from the islands of Tahiti and Moorea, French Polynesia.

Twenty three samples of Aedes aegypti populations from the islands of Tahiti and Moorea (French Polynesia) were tested for their oral susceptibility to dengue type 2 virus. The high infection rates obtained suggest that the artificial feeding protocol used was more efficient than those previously described. Statistical analysis of the results allowed us to define two distinct geographic areas on Tahiti with respect to the susceptibility of Ae. aegypti: the east coast, with homogeneous infection rates, and the west coast, with heterogeneous infection rates. No geographic differences could be demonstrated on Moorea. The possible mechanisms of this phenomenon are discussed in connection with recent findings on the variability of susceptibility of Ae. aegypti to insecticides.

Changing clinical and biological manifestations of dengue during the dengue-2 epidemic in French Polynesia in 1996/97--description and analysis in a prospective study.

In August 1996 dengue-2 virus was detected in French Polynesia for the first time since 1976. A prospective study was conducted from November 1996 to April 1997. Each time one of 7 physicians suspected dengue, the patient was enrolled and epidemiological, clinical and biological data were recorded. Dengue diagnosis was confirmed by virus isolation and IgM detection. The aims of this study were to find clinical and biological predictive factors constituting a specific profile of dengue (DF) and dengue haemorrhagic fever (DHF/DSS) and to assess the possibility of diagnosing dengue at primary health care level using clinical criteria and basic laboratory parameters. Of 298 clinically suspect cases, 196 (66%) were confirmed as dengue. The association of macular rash, pruritus, low platelet count and leukopenia was statistically predictive of dengue but not clinically, since these four signs occur in many other viral infections. As the prevalence of clinical and biological manifestations varied over time in our study, a specific profile useful for dengue diagnosis cannot be defined. With six cases of DHF, the morbidity of this dengue-2 outbreak was very low despite the sequential infection scheme DEN-3/DEN-2. The clinical expression of dengue could depend on a specific virus strain circulating in a specific population in a particular place, with varying virulence over time.

Possible dengue sequential infection: dengue spread in a neighbourhood during the 1996/97 dengue-2 epidemic in French Polynesia.

A DEN-2 epidemic occurred in French Polynesia from August 1996 to April 1997 after 7 years of DEN-3 circulation. The susceptible population constituted all expatriates and Polynesians under 21. In August 1996, two successive DEN-2 cases occurred in Teroma, a Tahitian neighbourhood close to the international airport of Tahiti. A serological prospective study of persons < 21 years living in Teroma was conducted. The study population was bled in September 1996, October 1996 and June 1997. Analysis of dengue spread in Teroma confirmed that dengue transmission occurs primarily in the house, thus vector control campaigns should incorporate focal insecticide spraying and systematic daily use of insecticide in houses. The evolution in time of the disease demonstrated that among a susceptible population, prevalence and incidence rates are related to the time of exposure, and consequently to age. Comparison of dengue incidence or dengue prevalence between populations therefore requires adjusted age rates. Most studies did not adjust for age, leading to the conclusion that DHF is more frequent during secondary than during primary dengue infection. Prospective studies taking into account the time of dengue exposure are necessary to confirm the sequential infection hypothesis.

[The specific epidemiological surveillance of dengue: the method and its importance since the dengue-2 epidemic in French Polynesia in 1996]. / Surveillance épidémiologiqe specifique de la dengue: méthode et intérêt lors de l'épidémie de dengue 2 en Polynésie française en 1996.

Dengue fever is present in tropical and subtropical regions and its geographical extension and the simultaneous increase of its mortality are worrisome. In endemic or epidemic countries, the aim of dengue-specific epidemiological surveillance is to confirm as soon as possible the circulation of a new viral dengue serotype, i.e. the beginning of an epidemic. The efficiency of the control strategy is improved by an earlier epidemic alert. In French Polynesia, dengue-3 virus circulated since 1989 at low level and, in May 1996, a specific epidemiological surveillance was undertaken because of the threat of a dengue-4 epidemic. From each suspected dengue case reported by 18 Polynesian physicians located in the Société Islands, a blood sample was taken for virological assay and clinical data were reported. Between May and November 1996, the virology unit of the Institut Malardé isolated 21 viruses (2 dengue-3 and 19 dengue-2) from 302 suspected cases. The dengue-specific epidemiological surveillance confirmed that dengue-2 virus was circulating and reduced the time of the epidemiological alert by 2 or 3 months compared to previous epidemics. Taking into account the day of illness, a logistic regression undertaken on the clinical data showed that the absence of cough was the only predictive sign of dengue diagnosis. The performance of this dengue-specific epidemiological surveillance system led us to consider its implementation in all concerned countries. A collaboration with international reference laboratories could be a solution for the developing countries.

Implication of macrophage inflammatory protein-1alpha in the inhibition of human haematopoietic progenitor growth by dengue virus.

The mechanisms were investigated of haematopoietic progenitor growth inhibition, observed after in vitro infection of cord blood mononuclear cells (CBMNC) by a clinical isolate of dengue 3 (29-56DSS). The level of virus replication was not different when CBMNC were inoculated with 29-56DSS compared with a prototype strain of dengue 3 (H-87) which had no inhibitory effect. An inhibitory effect was also observed when cell-free and heat-inactivated supernatants from 29-56DSS cultures, but not from H-87 cultures, were added to cultures of normal CBMNC, suggesting an indirect mechanism via the release of soluble suppressive factor(s). Macrophage inflammatory protein-1alpha (MIP-1alpha) was detected at a significantly higher level in 29-56DSS cultures than in controls. Blocking experiments with anti-MIP-1alpha antibody demonstrated that the inhibitory effect was related at least partly to high MIP-1alpha levels. To our knowledge, this is the first report suggesting an indirect effect of dengue infection on haematopoiesis mediated by a suppressive cytokine.

Plasma levels of tumour necrosis factor alpha and transforming growth factor beta-1 in children with dengue 2 virus infection in French Polynesia.

The pathogenesis of dengue haemorrhagic fever (DHF) is not well understood. In the absence of predictive clinical or biological criteria, the management of DHF patients remains difficult. The role played by cytokines in the occurrence of DHF has been suggested by several authors. In this study, we determined the plasma levels of tumour necrosis factor alpha (TNF alpha) and transforming growth factor beta-1 (TGF beta-1) in 52 children with laboratory-confirmed dengue virus infection admitted to hospital during the recent dengue 2 outbreak in French Polynesia. Thirty-three children were classified as having dengue fever (DF) and 19 as DHF. The plasma of both DF and DHF patients contained similar levels of TNF alpha. By contrast, plasma obtained from children with DHF had significantly higher levels of TGF beta-1 than plasma from children with DF, especially from days 1 to 3 after the onset of fever.

Dengue virus inhibits human hematopoietic progenitor growth in vitro.

Dengue disease, whether it be classical dengue fever (DF), dengue hemorrhagic fever (DHF), or dengue shock syndrome (DSS), is frequently associated with hematologic disorders. The underlying cause of these abnormalities is unknown. To determine if an inhibitory effect on human hematopoietic progenitor growth can be observed, normal cord blood mononuclear cells were exposed to low-passaged clinical isolates from DF, DHF, and DSS patients and to the prototype strain of dengue-3 virus (H-87). In primary methylcellulose cultures, there was no inhibition of colony formation. After an initial 8-day liquid culture, inhibition was observed with the isolates, but strain H-87 had no effect. Furthermore, isolates from patients with DSS showed a more potent inhibitory effect. These data represent the first documented study of in vitro impaired progenitor cell growth by dengue virus and suggest that this inhibition could be dependent upon the isolate tested.

Multicentre evaluation of dengue IgM dot enzyme immunoassay.

BACKGROUND: The traditional methods used in the diagnosis of dengue infection do not lend themselves to field application. As such, clinical specimens have to be sent to a central laboratory for processing which invariably leads to delay. This affects patient management and disease control. The development of the dengue IgM dot enzyme immunoassay has opened up the possibility of carrying out the test in peripheral health settings. OBJECTIVES: This multicentre study was conducted to evaluate a new, commercial nitrocellulose membrane based IgM capture enzyme immunoassay. STUDY DESIGN: The sensitivity and specificity of the test were compared with in-house dengue IgM enzyme-linked immunoassays routinely performed by each of the selected centres. Known positive and negative dengue specimens, as well as specimens from non-dengue cases, were included in the evaluation. RESULTS: Based on 402 specimens tested by the six centres, the sensitivity was 92.1% and specificity 88.1%, with an overall agreement of 92.8% when compared with IgM EIA assays performed on microplates. CONCLUSIONS: The results suggest that this commercial kit has a role to play in the diagnosis of dengue infection, especially in peripheral health settings.

Rapid and sensitive streptavidin-biotin amplified fluorogenic enzyme-linked immunosorbent-assay for direct detection and identification of dengue viral antigens in serum.

Each of the four serotypes of dengue viruses is responsible for a spectrum of illnesses that range from nonspecific febrile syndrome with good prognosis to dengue haemorrhagic fever or dengue shock syndrome. Definite diagnosis of dengue is provided by the detection of virus in acute-phase sera of patients. Virus isolation can be accomplished with mosquito cell lines or mosquito inoculations. However, these methods are time consuming and labour intensive. The reverse-transcriptase polymerase chain reaction (RT-PCR) provides a potential means of rapid diagnosis but requires specialised facilities and equipment and is expensive. Therefore a rapid, simple, sensitive, and economical method for direct detection of viral antigens in viraemic sera is needed for clinical and epidemiological investigations. An amplified fluorogenic enzyme-linked immunosorbent assay (F-ELISA) is described for the detection and identification of dengue-3 viruses in serum specimens. This assay utilizes biotinylated mouse IgG antibody directed against dengue antigens captured by anti-dengue monoclonal antibody coated onto polystyrene microplate wells. It takes advantage of the high affinity of biotin for the multivalent binding sites of streptavidin-labelled beta-galactosidase, and combines the amplification effect of biotin-streptavidin interaction with the high sensitivity of fluorogenic detection methods. Following optimisation of the procedure by reducing non-specific binding of proteins and enhancing the specific binding of antigens, F-ELISA was tested on 259 sera submitted routinely to our laboratory for confirmation of dengue diagnosis. The sensitivity of the F-ELISA was 90%, the specificity was 99% and the agreement rate was 98% between F-ELISA and virus isolation results.

Molecular epidemiology of dengue-1 and dengue-4 viruses.

Genetic variation between geographically and temporally distinct isolates of dengue-1 (DEN-1) and dengue-4 (DEN-4) viruses was investigated. The nucleotide sequences of a fragment of the envelope protein gene encoding amino acids 28 to 87 of 35 DEN-1 isolates and 28 DEN-4 isolates were determined. Maximum nucleotide sequence variation was 6.9% and 4.9% for DEN-1 and DEN-4 viruses, respectively. Taking a divergence of 6% between the nucleotide sequences as the cut-off value, three genotype groups were defined for DEN-1 viruses, whereas only one was observed for DEN-4 viruses. Molecular analysis of isolates from the South Pacific permits the classification of the recent strains of DEN-1 (1988-1989 epidemics) into a genotype distinct from the genotype which comprises earlier strains. This observation suggests that the recent epidemics were due to the introduction of a new genotype rather than to the re-emergence of the earlier strain.

Correlation between detection of plasminogen cross-reactive antibodies and hemorrhage in dengue virus infection.

Although dengue fever (DF) is usually self-limited, some patients experience severe and prolonged illness characterized by capillary leakage, which may progress to hypovolemic shock (dengue hemorrhagic fever/dengue shock syndrome; DHF/DSS) with hemorrhage of unknown etiology. Development of antibodies potentially cross-reactive to plasminogen has been reported in a high percentage of Thai patients with DF and DHF/DSS. Correlation between detection of plasminogen cross-reactive antibodies and hemorrhage was evaluated in 88 Tahitian children with dengue virus type 3 infection who presented with (n = 59) or without (n = 29) hemorrhage. Plasminogen cross-reactive antibodies were found in acute and convalescent sera of 33 and 11 children, respectively (56% vs. 31%, P < .05), and closely paralleled antibodies to the cross-reactive site in dengue virus E protein. Antibodies were more frequent in children with secondary than primary infections (60% vs. 32%, P < .05). Plasminogen cross-reactive antibodies did not correlate with occurrence of DHF/DSS or thrombocytopenia. These results are consistent with the possibility that plasminogen cross-reactive antibodies play a role in the etiology of hemorrhage in dengue virus infection.

Immunotypes of Chlamydia trachomatis isolated from genital tract specimens in Tahiti.

The distribution of Chlamydia trachomatis immunotypes (serovars) in Tahiti was studied by immunotyping of local isolates using monoclonal antibodies in the micro-immunofluorescence test. From 115 isolates obtained from the genital tracts of patients attending a sexually transmitted diseases clinic and other gynecological consultations, eight immunotypes were identified: E (51.3%), F (16.5%), G (13%), H (5.2%), J (6.9%), D (3.5%), K (1.7%) and I (0.9%). In addition, an isolate with mixed immunotypes, EJ (0.9%), was observed. This distribution was compared with those in different geographical areas.

Molecular epidemiology of dengue 3 viruses and genetic relatedness among dengue 3 strains isolated from patients with mild or severe form of dengue fever in French Polynesia.

The nucleotide sequences of a short fragment of the envelope protein gene encoding amino acids 25 to 89 of 27 dengue 3 viruses were determined by direct sequencing of PCR-amplified products, and the viruses were compared regarding their time of isolation and geographic distribution. Four distinct genotypic groups were discerned at 6% divergence between nucleotide sequences. The first group contained isolates from the South Pacific (1988 to 1992), Singapore (1973) and Indonesia (1973 to 1991). The second group comprised viruses from Asia (1956 to 1989) including the reference strain H-87. The third was composed of one isolate from Thailand (1971), and the fourth included the early strains from French Polynesia (1964 to 1969) and from Puerto Rico (1963). Furthermore, the difference between early and recent strains from the South Pacific was as high as 12.3%. This observation suggests that the recent epidemics in the South Pacific were probably the consequence of the spread of a new variant that emerged from New Caledonia. However, relatedness between nucleotide sequence and disease severity, or between strains from epidemics with mild disease (New Caledonia) and strains from epidemics with severe disease (French Polynesia) could not be demonstrated.

Ultra-rapid, simple, sensitive, and economical silica method for extraction of dengue viral RNA from clinical specimens and mosquitoes by reverse transcriptase-polymerase chain reaction.

A rapid, simple and efficient single-tube procedure is described for the isolation of dengue virus RNA from small amount of serum (10 microliters) followed by a reverse transcriptase-polymerase chain reaction (RT-PCR). Recovery of RNA is based on the lysing and nuclease-inactivating properties of guanidinium thiocyanate in the presence of silica. The silica RT-PCR can be completed within 5 hours starting from RNA extraction to agarose gel electrophoresis. All of the 63 dengue-3 culture-positive sera were RT-PCR-positive (virus titres: < 10(2) to 11(10.69.). Of 33 culture-negative acute sera from serologically confirmed dengue fever patients collected during dengue-3 epidemic, 4 were RT-PCR-positive. RT-PCR was also positive in 29 of 30 dengue-1 culture-positive sera (virus titres range: < 10(2) to 10(8.69). Dengue-1 virus was also detected in field-caught Aedes aegypti mosquitoes by silica RT-PCR.

Seroepidemiological survey of HTLV-I infection in French Polynesia, Cook Islands and Fiji.

Different population groups of French Polynesia, Cook Islands and Fiji were screened for Human T-Lymphotropic Virus type I (HTLV-I) antibodies. Among 1487 individuals sampled in French Polynesia, twelve were considered Western Blot (WB) indeterminate and one was considered WB-positive for HTLV-I infection. This positive subject originated from France and was a blood donor. Out of 196 Polynesians of the Cook Islands, one was WB-indeterminate. Among populations sampled in Fiji, one of 222 Melanesians was found WB-indeterminate and one of 211 Indians was WB-indeterminate.

Serum levels of tumor necrosis factor-alpha (TNF-alpha), interleukin-6 (IL-6), and interleukin-1 beta (IL-1 beta) in dengue-infected patients.

Sensitive immunoenzymatic assays were used to study the levels of tumor necrosis factor-alpha (TNF-alpha), interleukin-6 (IL-6), and IL-1 beta in sera from dengue-infected patients obtained during the 1989-1990 outbreak of dengue-3 in Tahiti, French Polynesia. The patients, both children (n = 47) and adults (n = 18), were clinically classified as having dengue hemorrhagic fever (DHF) and graded according to the severity of illness (grade I = fever, grade II = fever with spontaneous hemorrhagic manifestations, grade III = circulatory failure, grade IV = deep shock). The serum samples were obtained from day 1 to day 10 after the onset of the disease. High levels of TNF-alpha were observed in dengue-infected children of all severity grades. The highest values of TNF-alpha were found before day 6 after the onset of the infection, these values decreased from day 6 to day 10. The highest values were observed in sera from grade III and IV patients. High values of IL-6 were observed in serum samples of grade I and II patients on day 1, which decreased on day 4, and by day 5 were similar to those obtained from 25 control children. In grade III and IV patients, the highest values of IL-6 were observed from day 3 to day 5 after the onset of infection; after day 5, these values were very low.(ABSTRACT TRUNCATED AT 250 WORDS)