Influence of total serum IgE levels on the in vitro detection of beta-lactams-specific IgE antibodies

Abstract

Background: Allergic reactions to beta-lactams are a frequent cause of adverse drug reactions; the diagnosis is based on history, clinical examination, skin testing (prick and intradermal) and demonstration of serum-specific IgE antibodies (Abs).

Objective: We compared the diagnostic performance of the Phadia CAP system for the detection of IgE to beta-lactams carried out using the new test with cut-off limits of 0.10 kUA/L and the old test with cut-off limits of 0.35 kUA/L for positive results; subsequently, we analysed the effect of total serum IgE values and of atopic phenotype on the diagnostic performance of the tests.

Methods: The study comprised a total of 34 patients with a history of immediate adverse reactions to beta-lactams, which were confirmed by positive skin testing, and 115 control subjects with tolerance to beta-lactams over the last year. The Phadia CAP System was used for the determination of serum total and specific IgE Abs towards penicilloyl G (c1), penicilloyl V (c2), ampicilloyl (c5) and amoxicilloyl (c6). The overall diagnostic performance was assessed as a diagnostic odds ratio (DOR).

Results: The new test showed a higher sensitivity (85% vs. 44%) than the old test and a lower specificity (54% vs. 80%) but the overall diagnostic performance was poor (DOR 6.78 vs. 3.16, P = 0.333) in both tests. The total IgE value influences the DOR of both tests; DOR was better for values under 200 kU/L [DOR = 66; 95% confidence interval (CI): 11.3-384.1] or 500 kU/L (DOR = 45.7; 95% CI: 5.3-394.4) for the new and old tests, respectively.

Conclusions: The reduction in the positive cut off value has not significantly improved the overall diagnostic performance of the beta-lactams-specific IgE assay. Because of the influence of serum total IgE on the detection of beta-lactam-specific IgE Abs, the combination of both tests is mandatory in the in vitro diagnostic approach of beta-lactam allergy.