Identification and antimicrobial susceptibility testing of Gram-positive and Gram-negative bacteria from positive...

Rapid identification and antimicrobial susceptibility testing remain a crucial step for early efficient therapy of bloodstream infections. Traditional methods require turnaround times of at least 2 days, while rapid procedures are often associated with extended hands-on time. The Accelerate Pheno™ System provides microbial identification results within 90 min and susceptibility data in approximately 7 h directly from positive blood cultures with only few minutes of hands-on time. The aim of this study was, therefore, to evaluate the performance of the Accelerate Pheno™ System in identification and antimicrobial susceptibility testing of both Gram-positive and Gram-negative bacteria directly from clinical blood culture samples. We analyzed 108 and 67 blood culture bottles using the Accelerate PhenoTest™ BC kit with software version v1.0 and the FDA-cleared version v1.2, respectively. Reliable identification was achieved for Enterobacteriaceae, staphylococci, and enterococci, with 76/80 (95%), 42/46 (91%), and 10/11 (91%) correct identifications. Limitations were observed in the identification of streptococci, including Streptococcus pneumoniae and Streptococcus pyogenes, and coagulase-negative staphylococci. Antimicrobial susceptibility results for Enterobacteriaceae, for amikacin, ertapenem, ciprofloxacin, gentamicin, meropenem, and piperacillin-tazobactam ranged between 86 and 100% categorical agreement. Using v1.2, results for ceftazidime showed 100% concordance with the reference method. For staphylococci, the overall performance reached 92% using v1.2. Qualitative tests for detection of methicillin or macrolide-lincosamide-streptogramin B (MLSB) resistance caused major and very major errors for isolates. Overall, the present data show that the Accelerate Pheno™ system can, in combination with Gram stain, be used as a rapid complementation to standard microbial diagnosis of bloodstream infections.