Rationale:

• The dengue virus is present in the blood only during the acute febrile stage. Virus isolation by culture or PCR-based tests is most appropriate at this time.
• Because it takes time for IgM antibodies to develop, serum samples obtained in the first 5 days after the onset of illness may not contain detectable dengue-specific IgM antibodies.
• Rapid diagnostic assays test for the presence of dengue-specific IgM antibodies. Commercially available tests are simple to use and interpret, but they tend to have a high rate of false-positive results.
• Early and convalescent serum samples are required for serologic confirmation of the diagnosis.

Evidence:

• According to a review, virus isolation rates are up to 36% with the Aedes albopictus (C6-36) mosquito cell line and up to 80% using direct mosquito inoculation (9).
• WHO consensus guidelines state that five serologic techniques are available to diagnose dengue infections: IgM-capture ELISA, indirect IgG ELISA, hemagglutination inhibition, complement fixation, and the neutralization test (55).
• According to WHO guidelines, unequivocal serologic confirmation depends on a significant (four-fold or greater) increase in specific antibodies between acute-phase and convalescent-phase serum samples (55).
• In a systematic review and meta-analysis of the diagnostic accuracy of rapid immunochromatographic assays for the detection of dengue virus IgM antibodies during acute infection, 302 potentially suitable studies were identified, and 11 of these were selected for inclusion in the meta-analysis. Ranges of individual study results were as follows: sensitivity, 0.45 to 1.0; specificity, 0.57 to 1.0; diagnostic odds ratio, 4.5 to 1287; positive likelihood ratio, 2.3 to 59; and negative likelihood ratio, 0.01 to 0.56 (67).
• Dengue IgG and IgM antibodies were assessed in a group of prevaccinated healthy travelers using indirect IgG ELISA and IgM-capture ELISA kits. The IgM test was highly specific (negative result in all healthy vaccinees). The IgG test result was positive in 11% to 17% and 15% to 44% of healthy travelers vaccinated against Japanese encephalitis and yellow fever, respectively. The investigators concluded that the specificity of the IgG ELISA in prevaccinated travelers is much lower than that in unvaccinated populations (68).
• Dengue virus NS1 antigen was detected in 199 of 213 acute serum samples from patients with laboratory-confirmed acute dengue virus infection but in none of 354 healthy blood donors' serum specimens. This finding suggests that a commercial dengue NS1 antigen-capture ELISA may be superior to virus isolation and RT-PCR for laboratory diagnosis of acute dengue infections when based on the use of a single serum sample. The dengue NS1 antigen-capture ELISA yielded an overall sensitivity of 93.4% and a specificity of 100% (69). Additional studies are needed to confirm these findings.
• A study used a nested RT-PCR assay to detect concurrent infection with three dengue virus serotypes (DENV-1, DENV-2, and DENV-3) and two serotypes (DENV-1/DENV-2 and DENV-2/DENV-4) in three serum specimens from children hospitalized during the 2000-2001 Thai dengue epidemic. The data suggest that nested RT-PCR is an effective method of detecting concurrent dengue virus infections (70).
• Acute- and convalescent-phase samples were collected from hospitalized children with suspected dengue in Nicaragua from September 2003 to February 2004, and specimens were collected in a community setting before and after the 2003-2004 dengue season. Detection of anti-dengue IgM, IgA, and IgG in serum, filter-paper blood spots, and saliva was compared to gold standard detection in serum samples. The highest sensitivity and specificity for dengue diagnosis were obtained by measuring IgM or IgA in serum or filter-paper blood spots. Intermediate and poor results were obtained in saliva for IgM and IgA, respectively. Detection of IgG alone in serum, filter-paper blood spots, or saliva yielded the best results for diagnosis of dengue virus infection (71).
• A study compared the performance of two commercial tests for the detection of dengue NS1 protein during the clinical phase of dengue virus infection, an immunochromatographic test for rapid detection of the NS1 antigen and a two-step sandwich-format microplate ELISA, with a one-step sandwich-format microplate ELISA. A total of 222 samples from patients with acute infection caused by one of the four dengue serotypes detected by RT-PCR and/or virus isolation and 48 acute-phase serum samples from patients not infected with dengue virus were studied. The sensitivity of the one-step sandwich-format microplate ELISA test on acute serum samples was 87.4%, and the sensitivity of the immunochromatographic test was 81.5% after 15 minutes and 82.4% after 30 minutes. Both tests had a specificity of 100%. The two-step sandwich-format microplate ELISA test had a sensitivity of 60.4% and a specificity of 97.9%. These results support the use of diagnostic tools based on NS1 antigen detection for the diagnosis of acute dengue virus infection (72).

Comments:

• A number of commercial serologic test kits for anti-dengue IgM and IgG antibodies are available, some producing results within 15 minutes. Note, however, that the accuracy of most rapid dengue test kits has yet to be properly validated.
• Dengue virus can be recovered from the blood 2 days before the onset of disease and for 5 to 6 days after.
• The antigen battery for most serologic tests should include all four dengue serotypes, another flavivirus (e.g., Japanese encephalitis virus), a nonflavivirus (e.g., chikungunya virus), and an uninfected tissue control antigen. The interpretation of serologic test results may be complicated by the occurrence of cross-reactive antibodies to antigenic determinants shared by other members of the flavivirus family (e.g., Japanese encephalitis).
• Dengue-specific RT-PCR has a sensitivity greater than 90% in the first few days of illness, which decreases rapidly to less than 10% at 7 days after the onset of illness. Using whole blood appears to be better than using plasma (73).
• Rheumatoid factor may lead to a false-positive IgM-capture ELISA result for dengue (74).