Background: Resin-containing blood culture bottles (RBB) are used to increase the isolation of microorganisms by binding antimicrobials in sampled blood. Since RBB implementation in April 2018, our infection preventionists noted an increase in positive blood cultures on routine surveillance. Objective: To describe the change in bacterial isolation post-RBB implementation. Methods: All positive blood culture sets drawn in adult inpatient units or the emergency room between October 2017 and September 2018 and their associated organisms were obtained from the hospital laboratory database. Then, regardless of central-line placement or “present on admission” designation, the 2019 NHSN surveillance definitions for laboratory-confirmed bloodstream infection (LCBI-1 and LCBI-2) were applied to categorize all positive cultures as “common commensals” (CCs) or pathogens. A univariate analysis was performed using the Mantel-Haenszel χ2 test (OpenEpi version 3.01). Results: Although the number of monthly blood cultures drawn remained effectively stable before and after implementation (pre-RBB median, 3,512.5; post-RBB median, 3,626), the rate ratio of positive cultures increased by 1.36 times: pre-RBB median, 127 sets per month and post-RBB median, 172.5 sets per month (χ2 = 5.785; P = .008). The rate ratio of pathogen-containing cultures increased by 1.40 times (pre-RBB median, 98 sets per month and post-RBB median, 137.5 sets per month; χ² = 5.615; P = .009) with only a 1.24 increase in CCs (pre-RBB median, 29 and post-RBB median, 36; χ² = 0.553; P = .229) (Fig. 1). Conclusions: After RBB implementation, the monthly incidence rate of pathogen-containing sets increased. Additionally, the increase in these sets as well as of overall positive blood cultures was statistically significant. Current literature on RBBs does not suggest preferential increased isolation of pathogens. Further study is needed to determine whether our findings are related to blood-culturing practices or the RBBs themselves.

Funding: None

Disclosures: None