Eight Aspergillus antibody detection assays--three indirect haemagglutination assays (IHA-LD, IHA-Roche, IHA-Fumouze), three enzyme immunoassays (EIA-IgG, EIA-IgM, EIA-IgA, DDV) and two complement fixation tests (CF-metabolic and CF-somatic, Virion--and one latex agglutination test (LAT) for Aspergillus galactomannan antigen detection (Sanofi Pasteur) were evaluated in 14 patients with proven invasive aspergillosis (a total of 47 serum samples and one cerebrospinal fluid sample) and in 68 selected control individuals (one selected serum sample each). For the antibody tests, sensitivity ranged from 14% to 36% and specificity from 72% to 99%. The antigen detection test had a sensitivity of 36% and a specificity of 100%. Currently commercially available antibody detection assays for the serodiagnosis of invasive aspergillosis are inadequate. The antigen detection test appears to be highly specific, but lacks sufficient sensitivity.

The Pastorex Aspergius antigen test for detection of AspergiUus galactomannan antigen in the sera of patients with invasive aspergillosis is used in many clinical laboratories. A serum sample contaminated with Penicllium chrysogenum gave a strongly positive reaction (1:128) which was heat stable, was not eliminated by pronase treatment, and was not detected by a normal rabbit globulin control. This observation was shown to be due to cross-reactions of the monoclonal antibody EB-A2 used by the kit with several airborne fungi likely to contaminate serum samples, including Penicilium chrysogenum, Cladosporium herbarum, Acremonwium species, Alternaria alternata, Fusarium oxysporum, WangieUla dermatidis, and Rhodotorula rubra.

No abstract