Immunoassay of specific IgE: low level assays require measurement of allergen specific assay background.

The allergen-specific backgrounds of a paper disk radioimmunoassay (RIA) system and a cellulose sponge fluorescent enzyme immunoassay (FEIA) system were evaluated using three inhalants, timothy, short ragweed, and cat, with reagents obtained from the same manufacturer. Radioimmunoassay was performed with Phadebas RAST reagents by the modified RAST method, and FEIA by the Pharmacia CAP System. Horse serum, 5% HSA, assay diluent, and serum pools from nonallergic and allergic patients, were assayed. The lower limit of detection (LLD) was defined using both the Z distribution (as is conventional) and the t distribution. The solid phases, analytes, and assays differed (p less than .001) in background results. For RIA, background was lowest for timothy and highest for cat; for FEIA, background was lowest for cat and highest for short ragweed. For RIA, background assessed with the allergic serum pool was higher than the other analytes; for FEIA, responses of the five analytes did not differ. For timothy and short ragweed, background of RIA was lower than FEIA. For FEIA, the highest LLD calculated using the Z distribution was 11 SD lower than the manufacturer's recommended quantitative cutoff; for RIA, the highest LLD calculated was 2.5 SD higher than the recommended analytic cutoff. The analytic false positive rate for RIA may differ between allergic and nonallergic patient populations. Laboratories reporting results near either assay's background should set LLD based on assay of allergen-specific negative controls in each assay run.