Breakpoints for susceptibility testing should not divide wild-type distributions of important target species.

The fluconazole MIC distributions for Candida glabrata from testing 34 different clinical isolates and performing 51 tests on a single isolate mirrored each other. Since what is perceived as biological variation in isolates without resistance mechanisms is mainly methodological variation, breakpoints which divide this distribution not only lack a sound biological basis but also result in poor reproducibility of susceptibility characterization. This makes 2, 4, 8, and possibly 16 microg/ml unsuitable breakpoints for C. glabrata and fluconazole.