Analysis of available diagnostic tests for latex sensitization in an at-risk population

Abstract

Background: Lack of a Food and Drug Administration (FDA)-approved skin testing reagent for latex allergy in the United States requires reliance on patient history and serologic assays for diagnosis.

Objective: To determine the diagnostic sensitivity, specificity, and predictive values of an FDA-cleared antilatex IgE serology test and an enzyme-linked immunosorbent assay (ELISA) with various sources of latex protein antigens in an at-risk but unselected population of health care workers.

Methods: Health care workers underwent duplicate latex and serologic testing for latex specific IgE with the CAP assay and ELISA from June 1, 1998, through December 31, 2002. Logistic regression with receiver operating characteristic curve analysis determined the values, resulting in 98% and 99% specificity for the CAP assay and ELISA, respectively.

Results: Results of paired skin and serologic tests were available for 792 participants. Forty duplicate skin test results (5%) were positive. For the CAP assay, sensitivity was 35%; specificity, 98%; positive predictive value, 48.3%; and negative predictive value, 96.6%. ELISA demonstrated similar results. Multivariable logistic regression yielding a 98% or 99% specificity for the various ELISAs demonstrated that the adjusted odds of a positive skin test result significantly increased with positive CAP assay and ELISA results using a powdered glove extract.

Conclusions: The performance of the FDA-cleared antilatex IgE serologic test for latex allergy has much lower sensitivity than previously reported. This finding confirms that this serologic test should be used only for patients with a history of latex allergy and not for screening the population with a low prevalence of latex sensitization.