1166. Infection Rate in Venous–Venous Extracorporeal Membrane Oxygenation (VV-ECMO)

Abstract Background The Extracorporeal Life Support Organization (ELSO) Infectious Disease Taskforce has identified infection rates as an area requiring further study. We aimed to evaluate our patients on venous–venous extracorporeal membrane oxygenation (VV ECMO) for infection rates, organisms, and specific factors that may alter infection risk. Methods Retrospective data were collected from September 2012 to 2015 for adult patients on VV ECMO. Demographic information as well as type and location of cannulation, antibiotics prior to and during cannulation, and presence of bacteremia were obtained. Counts and percentages are reported for categorical variables. Results 47 patients between 18 and 82 years of age with a median of age of 58 years were included. 76.6% (N = 36) were male. The average number of hospital days until cannulation was 7.8 days. 400 ECMO days were captured with a range of time on ECMO of 1–41 days. 46.8% (N = 22) of patients were alive at the time of discharge. 70.2% (N = 32) of cannulations were in the OR (Figure 1). 55% (N = 26) of patients had a Right Ventricular Assist Device (RVAD) with in-line oxygenator. 66% (N = 31) of all patients were on antibiotics prior to cannulation for other indications while 53% (N = 25) of all patients received perioperative antibiotics (Figure 2). 59% (N = 28) of all patients had blood cultures while on ECMO, with only 8% (N = 4) of all patients returning positive. Of the positive cultures, 2 were true bacteremia resulting in an infection rate of 5 per 1,000 ECMO days (Figure 3). One patient grew Escherichia coli while the other grew Escherichia coli and Streptococcus pneumoniae. Neither patient received antibiotics perioperatively or for other indications (Figure 4). Due to the low number of infected patients, further statistical analysis could not be performed. Conclusion We report a low rate of bloodstream infection amongst our VV-ECMO patients. Further investigation to ascertain trends in infection rate with cannulation location/type as well as association with antibiotic is currently being pursued. Other avenues of research will include a comparison of infection rates between VV and Venous Arterial (VA) ECMO. Information from these investigations will help aid the development of guidelines to minimize the rate of nosocomial infections in ECMO patients. Disclosures All authors: No reported disclosures.