The comparative accuracy of 8 commercial rapid immunochromatographic assays for the diagnosis of acute dengue virus infection

Abstract

Background. The serological diagnosis of acute dengue virus infection relies on the detection of dengue-specific immunoglobulin M (IgM) antibodies. Immunochromatographic tests are rapid diagnostic tests (RDTs) that can be performed at the bedside, but they have not been fully validated for diagnosis of dengue infection. Methods. More than 20 RDTs for diagnosis of acute dengue infection are commercially available. Of these, 8 were selected for evaluation of performance by use of characterized dengue and nondengue serum specimens, and results were compared with those of a previously published dengue IgM/IgG enzyme-linked immunosorbent assay in conjunction with dengue virus serotyping by reverse-transcriptase polymerase chain reaction. Results. Assay sensitivities were low, ranging from 6.4% (95% confidence interval [CI], 4.0%-9.7%) to 65.3% (95% CI, 59.9%-70.5%), and specificities ranged from 69.1% (95% CI, 61.4%-76.0%) to 100% (95% CI, 97.8%-100%). Of the 8 tests, only 2 had sensitivities of >50%, the level considered to be clinically useful, and, of these, 1 had relatively low specificity (69.1%). Samples collected early in the infection were less likely to test positive than those collected later. A thermal stability study demonstrated a loss in performance of some RDTs when they were stored at a high ambient temperature for 3 months. Conclusions. Users of RDTs for dengue should be aware that many of these tests have a diagnostic accuracy that falls well below the manufacturers' claims. If an acute specimen yields a negative result, a convalescent serum sample should be tested to confirm the result. No RDT adequately differentiated primary and secondary dengue infections, and the tests should not be used for this purpose. © 2006 by the Infectious Diseases Society of America. All rights reserved.