Performance evaluation of AlfredAST rapid susceptibility testing directly from positive blood cultures in the routine laboratory workflow.

The aim of this study was to evaluate the performance of the new automated system AlfredAST which is based on light scattering technology for rapid susceptibility testing directly from positive blood cultures as well as its applicability in the routine laboratory workflow. We evaluated 176 significant episodes of bacteremia due to 92 Gram-negative and 84 Gram-positive bacteria. The antimicrobial agents tested were ceftriaxone, ciprofloxacin, gentamicin, meropenem, piperacillin-tazobactam, and colistin for Gram negatives and cefoxitin, vancomycin, linezolid, and daptomycin for Gram positives. Concordance assessment was performed in comparison with our routine method, Vitek2 (bioMérieux). Discrepancies were resolved with MICRONAUT-S (Merlin) or E-test (bioMérieux). Out of 690 susceptibility determinations, 94.05% showed categorical agreement (CA) with the routine method and this percentage increased to 94.49 after discrepancy analysis. There were 1.45% very major errors, 3.33% major errors, and 1.16% minor errors (decreased to 1.45, 3.04, and 1.01 after discrepancy analysis). The CA for most of the antibiotics was above 90% except for daptomycin for Gram positives (87.30%) and ceftriaxone for Gram negatives (88.23%). The concordance was slightly better for Gram negative than for Gram-positive bacteria (94.30 versus 93.70%, respectively). The total turnaround time for a complete AlfredAST result was 6-6.5h. The evaluated method gave rapid and reliable results in a few hours, versus 48h for the conventional one. Implementing this technology in routine workflow allows clinicians to optimize the treatment on the same day of blood culture positivity with potential positive clinical benefits and impact on antibiotic stewardship.