Performance and potential clinical impact of Alfred60AST (Alifax®) for direct antimicrobial susceptibility testing on positive blood culture bottles.

Rapid pathogen identification (ID) and antimicrobial susceptibility testing (AST) of bacteria-causing bloodstream infections can improve patients' outcome. In this study, we evaluated the performance of Alfred60AST (Alifax) which provides AST directly on positive blood culture (BC) bottles by light scattering. In a selected group of patients with a clinical suspicion of severe sepsis or at risk for infections with multiresistant organisms, we compared Alfred60AST AST results with traditional AST results (Vitek2 (bioMérieux) or disk diffusion). Discrepancy analysis was performed by Etest (bioMérieux) or broth microdilution. In total, 222 samples were evaluated. On 595 susceptibility determinations, 93.4% showed categorical agreement (CA) with the standard method. Eighty-one percent of isolates showed a 100% categorical agreement (CA) which increased to 84.3% after discrepancy analysis. There were 8 very major discrepancies (VMD), 18 major discrepancies (MD), and 13 minor discrepancies (MiD). Most discrepant results were observed for piperacillin-tazobactam (15.6%) and clindamycin (18.9%). Analysis time was 6-6.5 h for a complete Alfred60AST AST result. In addition, we evaluated the behavior of clinicians in adjusting antibiotic therapy according to the routine AST results. In 37% of all patients, antibiotic therapy was altered after reporting of AST result and adjustment was more frequent for Gram-negative than for Gram-positive isolates. With some improvements, Alfred60AST provides accurate and rapid preliminary AST results for organisms causing bloodstream infections and may have at least a potential clinical benefit in about one-third of patients with severe sepsis, by delivering faster results compared with conventional methods.