Evaluation of EUCAST rapid antimicrobial susceptibility testing (RAST) directly from blood culture bottles

This study was performed to evaluate EUCAST blood culture (BC) rapid antimicrobial susceptibility testing (RAST) for Gram-negative organisms. Clinical BCs positive with Gram-negative-bacilli between April 23, 2019 and June 21, 2019, and 41 spiked isolates were tested by RAST against ceftazidime, piperacillin-tazobactam, meropenem, ciprofloxacin, gentamicin, and amikacin and compared against the routine method, Vitek 2. Vitek turnaround time was also audited. A total of 194 isolates were included, of which 68 Escherichia coli, 47 Klebsiella pneumoniae, and 33 Pseudomonas aeruginosa were further analyzed. For those with interpretable results, using CLSI M52 defined cutoffs for equivalence in antimicrobial susceptibility testing (VME ≤ 1.5%, ME ≤ 3.0%, CA ≥ 90%), RAST was considered equivalent to Vitek 2 for all reading times for the following drug-bug combinations: E. coli (meropenem, gentamicin, and amikacin); K. pneumoniae (ceftazidime, ciprofloxacin, and gentamicin); and P. aeruginosa (gentamicin). The highest error rates were MEs in piperacillin-tazobactam for E. coli at 4 h (5/9, 55.6%) and P. aeruginosa at 6 h (1/1, 100%). The number of MEs/VMEs for other drug-bug combinations and readings times ranged from 0 to 2 isolates (maximum of 7.7%). Median turnaround time to availability of Vitek results was 19.58 h for E. coli (IQR: 17.38-21.38 h), 19.43 h for K. pneumoniae (IQR: 17.72-20.67 h), and 23.3 h for P. aeruginosa (IQR: 21.32-24.35 h). Introduction of RAST shortens turnaround time to results. With the exception of piperacillin-tazobactam, the absolute number of errors was low. A larger-scale evaluation is required to confirm the suitability of implementing this testing method in local laboratories.