Comparison of the Accelerate PhenoTM System with the conventional methods for the identification and antimicrobial susceptibility testing of positive blood cultures

Background: Rapid identification (ID) and antimicrobial susceptibility testing (AST) of microorganisms from blood specimens are essential to assure appropriate patient management. The purpose of this preliminary study is to evaluate the performance of the innovative Accelerate PhenoTM System (APS, Accelerate Diagnostics) for ID and AST directly from positive blood cultures (BCs) in comparison to conventional methods (CM). Materials/methods: Forty positive BCs were submitted to both APS and CM for ID (QuickFISH - AdvanDx- and MALDI-TOF MS –performed by Bruker Daltonics or bioMérieux instruments- from liquid and/or solid culture) and AST (Phoenix system -Becton Dickinson- for Gram negative, Vitek 2 – bioMérieux- for Gram positive, and, when necessary, to E-Test). Results: Four of the 40 samples were interrupted by APS before providing ID. For the remaining 36 samples, limited to the microbial targets included in the assay, all APS ID results were in agreement with those of CM except 4 (sensitivity 94.7%; specificity 99.5%). APS additionally revealed Pseudomonas aeruginosa besides Staphylococcus aureus in 2 cases, missed P. aeruginosa in one mixed infection and misidentified one S. aureus as coagulase negative staphylococci. Among the 36 BCs examined AST was not performed, according to the protocol in two samples containing Streptococcus spp., due to abortion after ID in two samples containing both S. aureus and P. aeruginosa and due to growth control failure in 4 samples (3 S. aureus and 1 Klebsiella spp.). Among the 238 agent/microorganism combinations associated with correct ID, 17 minor errors (7.2%) and 9 major errors (4.8%) were observed. Moreover, APS incorrectly classified a MSSA as MRSA. The APS ID results were obtained in 1.5 h (versus 35 min by QuickFISH, 1.5 h by MALDI-TOF MS directly from BCs and 18-24 h by MALDI-TOF from solid culture) and the AST results were available in 8h, on average 34.1 h before CM. Conclusions: Although the limited number of the samples tested in this preliminary study does not allow an accurate assessment of the APS, further analysis should be performed to achieve this purpose. At present, the reduction in time-to-result represents one of the main advantage of this technology.