ABSTRACT
Inadequate therapy in bloodstream infections is suggested to be associated with higher mortality. We evaluated the reduction in inappropriate antibiotic therapy using rapid identification and antibiotic susceptibility testing (FAST) compared to standard of care (SOC) testing in patients with bloodstream infections. The FAST method used polymerase chain reaction (PCR) for identification and to detect growth in the presence or absence of antibiotics after only 6 h. For SOC testing, the BD Phoenix system was used. Patients with blood cultures growing Staphylococcus, Streptococcus or Enterococcus species or Gram-negative rods were randomised for FAST or SOC tests. A total of 129 patients were randomised for FAST and 121 patients for the SOC group. At the time SOC results became available, 78 patients in the FAST group could have been switched to more appropriate therapy. Although FAST results were highly accurate (agreement with SOC was 94%), they were only implemented in a minority (16) of patients. However, significantly fewer patients in the FAST group used inappropriate therapy at the time of SOC results (p = 0.025). The time to results in the FAST group was reduced by 15.6 h (p < 0.001). In the patients switched after FAST, this was done after a mean of 42.3 h compared to 61.4 h in those switched after SOC tests (p < 0.001). In bacteraemic patients, FAST resulted in significantly more patients using appropriate antibiotic therapy at the time SOC results were available and 15.6 h earlier than SOC tests. However, the implementation of FAST results was not optimal and no benefit on clinical outcome was shown.
