Evaluation of the ATB 32 C system for identification of clinical yeast isolates

OBJECTIVE: To compare the ATB 32 C system for routine identification of clinical yeast isolates in a clinical microbiology laboratory with identification carried out by conventional methods in a mycology reference laboratory. METHODS: A total of 113 strains initially isolated at our hospital and identified in the reference laboratory were returned in duplicate, under separate code numbers, to the microbiology laboratory where the ATB 32 C system was used for identification by: 1) visual assessment of turbidity at 72 h with use of identification table; 2) visual assessment at 72 h with use of ATB 32 C analytical profile index; and 3) automatic readings with the ATB reader at 48 h and 72 h with results of growth assessments transmitted to a computer and interpreted by the ATB 32 C software. RESULTS: Visual assessment plus identification table and visual assessment plus profile index provided correct identification in 98% and 91% of strains, respectively. Visual assessment was, however, sometimes difficult and required more experience than is usually available in a routine clinical microbiology laboratory. Automatic readings with computer identification plus supplementary tests correctly identified 87% and 86% after 48 h and 72 h, respectively. CONCLUSIONS: The ATB 32 C system with automatic readings and computer identification is a satisfactory system for identification of clinical yeast isolates in a routine clinical microbiology laboratory.