Many studies demonstrate administration of timely and active antibiotics correlates to better clinical outcomes. However, some clinicians will continue broad empiric therapy until blood cultures are officially finalized. This practice may stem from the concern for pathogens potentially growing after several days of incubation. Lack of timely de-escalation can be costly, cause unnecessary antibiotic exposure, and increase selective pressure for antibiotic resistance.
This study sought to determine the timeframe in which the majority of clinically significant pathogens are identified on blood cultures. Secondary objectives were to examine 1) the effect of time-to-positivity (TTP) on time-to-de-escalation of antibiotics (DABX); 2) the impact of source of infection and antibiotic administration prior to blood culture collection on TTP.
This was a single center, retrospective chart review of randomly selected positive bacterial blood cultures from adults hospitalized between 9/1/2013 – 8/31/2014. Culture bottles containing < 10 cc of blood or duplicate isolates recovered during the same infection time period were excluded. Kaplan-Meier time-to-event plot was created for positive blood cultures for clinically significant pathogens and contaminants. Two-factor Anova or Fisher’s Exact test were used for categorical variables.
793 blood samples were screened; 310 samples met inclusion criteria. 33 (11%) cultures were polymicrobial and 243 (67%) isolates were considered clinically significant pathogens. The median TTP for clinically significant pathogens and contaminants was 18 and 25 hours, respectively (p < 0.003). By 48 hours, 95% of clinically significant pathogens were detected. There was no significant association between TTP and DABX. TTP was not affected by source of infection or administration of antibiotics prior to blood sample collection.
TTP is an important part of antimicrobial therapy decision-making. Aside from a few slower growing bacteria, few clinically significant pathogens are detected beyond 48 hours. De-escalation of broad empiric antibacterial therapy should be considered before laboratory finalization of blood cultures.