Performance of indirect immunoglobulin M (IgM) serology tests and IgM capture assays for laboratory diagnosis of measles

As progress is made toward elimination of measles, the laboratory confirmat ion of measles becomes increasingly important. However, both false-positive and false-negative results can occur with the routinely used indirect meas les immunoglobulin M (IgM) serology tests. The measles IgM capture assay is considered to be more specific, and therefore, its use is indicated for co nfirmatory testing, but its relative performance has not been fully assesse d. Four commercial indirect measles IgM serology test kits (the Behring, Cl ark, Gull, and PanBio assays) and a commercial IgM capture assay (the Light Diagnostics assay) were evaluated for their abilities to detect measles vi rus-specific IgM antibody with a total of 308 serum samples from patients i nvolved in a measles outbreak and with confirmed cases of measles and 454 s amples from subjects without measles. The Centers for Disease Control and P revention (CDC) IgM capture assay was also used in a part of the evaluation . Among the indirect assays, the overall sensitivities ranged from 82.8% (C lark assay) to 88.6% (Behring assay) and specificity ranged from 86.6% (Pan Bio assay) to 99.6% (Gull assay). These rates were 92.2 and 86.6%, respecti vely, for the Light Diagnostics capture assay and 87.0 and 94.8%, respectiv ely, for the CDC capture assay. While the Light Diagnostics capture assay h ad the best detection rate (80%) with the acute-phase samples compared with those for the rest of the tests (CDC capture assay, 77%; Behring assay, 70 %; Gull assay, 69%; PanBio assay, 58%; and Clark assay, 57%), all tests sho wed a significantly improved sensitivity in the range of 92% (Clark and Pan Bio assays) to 97% (Light Diagnostics and CDC capture assays) with the conv alescent-phase samples, as expected. The best seropositivity rates (in the range of 92 to 100%) were observed with samples collected 6 to 14 days afte r the onset of symptoms. The Gull assay showed the highest positive predict ive value (99.6%), followed by the Behring assay (97.8%) and the CDC captur e assay (96.1%). Overall, the Gull and Behring assays were found to be as g ood as or better than the capture assays. In conclusion, laboratory diagnos is of measles based on IgM serology varies depending on the timing of speci men collection and the test used, and the case for the use of the IgM captu re assay as the confirmatory test appears to be uncertain.