Abstract
Nocardiosis is an underrecognized clinical entity in South Africa, for which interspecies epidemiological and clinical differences are poorly understood. The taxonomical state of flux and the lack of a simple antimicrobial susceptibility testing method are partly responsible. Definitive identification is molecularly based, which further complicates the study of this ubiquitous organism, as this methodology is beyond the scope of most routine diagnostic laboratories. The Etest methodology has been proposed as an alternative to the reference broth microdilution method, although there have been a limited number of comparative studies. We profiled 51 clinical isolates of aerobic actinomycetes, including 39 Nocardia species, using sequence-based (16S rRNA) identification. Broth microdilution and Etests were done concurrently on all isolates. The overall level of categorical and essential agreement for broth microdilution and Etest for the Nocardia isolates ranged from 67.5 to 100% and 46.2 to 81.6%, respectively. Very major errors were seen with amikacin, amoxicillin-clavulanate, ciprofloxacin, clarithromycin, and imipenem. For Nocardia species, uniform susceptibility to co-trimoxazole, amikacin, and linezolid was demonstrated, with a 48.8% susceptibility rate to imipenem. Nocardia farcinica (20.5%) and Nocardia cyriacigeorgica (15.4%) were the most commonly identified species among the 82% of isolates identified to species level using 16S rRNA sequences. Furthermore, drug susceptibility patterns demonstrated limited concordance with species identification. Our results suggest that, in a routine diagnostic setting, the Etest is not an acceptable alternative to the reference method of broth microdilution for antimicrobial susceptibility testing. Given the diversity and limited understanding of this group of organisms, further widespread evaluation of clinical isolates, from both clinical and diagnostic perspectives, is warranted.