OBJECTIVES Recently, rapid phenotypic antimicrobial susceptibility testing (AST) based on microscopic imaging analysis has been developed. The aim of this study was to determine whether implementation of antimicrobial stewardship programmes (ASP) based on rapid phenotypic AST can increase the proportion of patients with haematological malignancies who receive optimal targeted antibiotics during early periods of bacteraemia.

METHODS This randomized controlled trial enrolled patients with haematological malignancies and at least one positive blood culture. Patients were randomly assigned 1:1 to conventional (n = 60) or rapid phenotypic (n = 56) AST. The primary outcome was the proportion of patients receiving optimal targeted antibiotics 72 hr after blood collection for culture.

RESULTS The percentage receiving optimal targeted antibiotics at 72 hr was significantly higher in the rapid phenotypic AST group (45/56, 80.4%) than in conventional AST group (34/60, 56.7%) (relative risk (RR) 1.42, 95% confidence interval (CI) 1.09-1.83). The percentage receiving unnecessary broad-spectrum antibiotics at 72 hr was significantly lower (7/26, 12.5% vs 18/60, 30.0%; RR 0.42, 95% CI 0.19-0.92) and the mean time to optimal targeted antibiotic treatment was significantly shorter (38.1, standard deviation (SD) 38.2 vs 72.8, SD 93.0 hr; p < 0.001) in the rapid phenotypic AST group. The mean time from blood collection to the AST result was significantly shorter in the rapid phenotypic AST group (48.3, SD 17.6 vs 83.1, SD 22.2 hr).

DISCUSSION ASP based on rapid phenotypic AST can rapidly optimize antibiotic treatment for bacteraemia in patients with haematological malignancy. Rapid phenotypic AST can improve antimicrobial stewardship in immunocompromised patients.