1013. If Blood Cultures Were Not Done Before Starting Antibiotics, Is It of Any Value to Obtain Them Later?

Abstract

Background. Obtaining blood cultures before starting antibiotics is one of the pillars of the Surviving Sepsis Campaign (SSC), and delay in obtaining blood cultures (BC) afer starting antibiotics is associated with increased mortality (Levy M 2015, Pruinelli L 2018), but we were unable to fnd data on the relationship between such a delay and a reduction in percentage of positive cultures. Methods. All adult patients (>18) admitted from the UFHealth Shands Emergency Department (ED) between August 2012 and December 2016 were included in the study (N = 30,743), excluding hospital-hospital transfers. BC were done with BacTec aerobic, anaerobic, and pediatric resin bottles, incubated for 5 days. We calculated the hourly rate of positive BC obtained before and afer the start of IV antibiotics by subtracting the time stamp in the electronic medical record (Epic) between the frst BC collection time and the start of the frst IV antibiotic dose. We considered S. aureus, all Gram-negative rods, β-hemolytic Streptococci and Enterococci as signifcant pathogens and coagulase negative Staphylococci, S. viridans, Propionibacterium sp., Micrococcus sp. and Bacillus sp. as contaminants hospital ransfers. Results. The percentage of BC with signifcant growth was unchanged during the frst hour afer starting IV antibiotics, but declined signifcantly in the period 1-12 hours afer IV antibiotics were started. The overall positivity rate before starting IV antibiotics was 1,646/20,867 (7.9%) of patients and declined to 112/3,490 (3.2%), P < 0.0001, in the 1-12 hour period aferwards, but did not decline to 0. Septic patients averaged 1,143/4,923 (23.2%) positive and declined to 65/728 (8.9%), P < 0.0001, while nonseptic patients averaged 503/15,944 (3.15%) positive before antibiotics and declined to 47/2,762 (1.7%) P < 0.0001, 1-12 hours afer. It should be pointed out that these are group averages from diferent patient groups at each hourly time, rather than individual patients with blood cultures drawn serially Conclusion. We conclude that IV antibiotics dramatically reduce the likelihood of getting a positive blood culture, but not during the frst hour of administration; however, the residual positivity rate remains high enough that blood cultures are still clinically worthwhile.