Histoplasma capsulatum antigen detection: comparison of the performance characteristics of a new inhibition immunoassay to those of an established antibody sandwich immunoassay.

Dr. Beatriz L. Gomez and colleagues described a new inhibition immunoassay for the diagnosis of histoplasmosis (1). My laboratory developed and extensively evaluated an antibody sandwich immunoassay for this purpose. It is important to contrast the two methods, as their performance characteristics differ. The characteristics of the assay developed in my laboratory are high sensitivity in disseminated histoplasmosis but less sensitivity in pulmonary and chronic pulmonary histoplasmosis, with greater sensitivity in urine, cross-reactions with Blastomyces dermatitidis, Paracoccidioides brasiliensis, and Penicillium marneffei, but not with Aspergillus spp., Cryptococcus spp., and Mycobacterium tuberculosis (2). The report by Gomez et al. demonstrated cross-reactivity in their inhibition assay with Aspergillus spp. (20.00%), Cryptococcus spp. (10.00%), and M. tuberculosis (22.22%) (1). Results obtained by using the inhibition immunoassay reported by Gomez et al. differ from ours in several respects. Detection in serum is more sensitive than in urine, and significant cross-reactivity is noted with Aspergillus spp., Cryptococcus spp., and M. tuberculosis. Furthermore, 2.27% of the normal human sera tested false positive for Histoplasma capsulatum. Also, the sensitivity in cases of acute pulmonary histoplasmosis (88.88%) was greater than in cases of disseminated histoplasmosis (non-AIDS, 62.50%; AIDS, 72.72%). In our experience the sensitivity is higher for disseminated disease (urine 92%, serum 82%) than for acute pulmonary disease (44%) (3). In conclusion, the detection characteristics of the inhibition immunoassay differ considerably from those of the antibody sandwich immunoassay, raising concern about the accuracy and usefulness of the inhibition immunoassay.