Diagnostic performance of Food and Drug Administration-cleared serologic assays for natural rubber latex-specific IgE antibody

Background: In the absence of Food and Drug Administration-approved natural rubber latex skin testing reagents, latex-specific IgE antibody immunoassays are used in the diagnosis of latex allergy. Comparative diagnostic performance of these tests has not been definitively determined. Objective: We sought to study the predictive value of available Food and Drug Administration (510K)-cleared latex-specific IgE antibody immunoassays in the diagnosis of latex allergy. Methods: Subjects (n = 312) were classified as having a positive (n = 117) or a negative (n = 195) latex allergy history (Hx) or having a positive (n = 131) or a negative (n = 181) puncture skin test (PST) response (Greer reagent). The 14 subjects with a negative Hx and a positive PST response had negative responses to glove provocation testing and thus were considered sensitized but asymptomatic. Sera from 22 subjects were split to evaluate intra-assay variation. All 334 coded sera were analyzed for latex-specific IgE antibodies in the Diagnostic Products Corporation microplate AlaSTAT, Hycor HY-TEC EIA System, and Pharmacia-UpJohn CAP System. Variance and diagnostic performance parameters of each test were computed with 95% confidence intervals in relation to the subjects' Hx and PST status. Results: Intra-assay concordance of split sera results was 96.0% for all 3 assays, with coefficients of variation of less than 25% and between-assay coefficients of variation of less than 21%. The diagnostic performance of the CAP and AlaSTAT assays were equivalent in comparison with PST results: sensitivity, CAP 76.3% and ALASTAT 73.3% and specificity, CAP 96.7% and AlaSTAT 97.2% (P = NS). The HY-TEC assay was more sensitive (91.6%) and less specific (73.3%) than the CAP and AlaSTAT assays (P < .001). From 9% to 25% of the sera were discordant, being positive in at least 1, but not all 3, assays. Conclusion: The CAP and AlaSTAT assays produce 24% and 27% of false-negative results, respectively, whereas the HY-TEC produces 27% of false-positive results when compared with the PST.