Determination of immunity to measles virus in young adults: comparative evaluation of a commercial enzyme immunoassay and the hemagglutination inhibition techniques

Abstract

Background: Determination of the immune status against measles in young adults requires careful evaluation of the laboratory methods because of waning immunity. The hemagglutination inhibition (HI) test and enzyme-linked immunosorbent assay (ELISA) may lack the sensitivity required to detect very low levels of antibodies. In addition, the correlation between ELISA-IgG assays and the degree of protection from measles is not well defined. Objectives: (a) Evaluation of a commonly used measles ELISA-IgG test kit in comparison with the hemagglutination inhibition (HI) test which corresponds strongly to virus neutralization; (b) determination of false negative rates of the ELISA-IgG and the HI tests; (c) evaluation of the ELISA-IgG test kit as a quantitative assay. Study design: One hundred and eighty serum samples collected from 60 vaccinated young adults immediately before vaccination and 14 and 28 days postvaccination, were tested comparatively by HI and by a commercial ELISA-IgG kit. For evaluation of false negative rates, postvaccination sera of a cohort of 48 vaccinees with negative HI or ELISA-IgG prevaccination sera were tested for IgM. Sixty-three of the samples were also titrated by the ELISA-IgG kit using serial dilutions, for comparison with HI titers. Results: Using the HI test as a reference method, the ELISA-IgG kit was found to have overall accuracy of 81%, sensitivity of 80% and specificity of 84%. The false negative and the false positive rates were 20% and 16%, respectively. In contrast, when we used postvaccination IgM test to distinguish between true and false prevaccination negatives in both the HI and ELISA-IgG tests, we found that the false negative rates were 75.6% by ELISA and 72.5% by HI, and the false positive rates were 2.4% and 0%, respectively. Serum titers determined by the ELISA-IgG test were generally 5–10-fold higher than the corresponding HI titers, but without a consistent correlation. Conclusions: Both the ELISA-IgG and the HI tests frequently failed to detect residual immunity. The two tests also did not correlate well with each other suggesting that different antigenic determinants of the virus are involved in each assay and therefore the HI test should not be used as a reference method for evaluation of the sensitivity of ELISA IgG kits.