Abstract
Background
ECMO recipients who develop bloodstream infections (BSI) meeting CLABSI criteria are publically-reported in inter-facility comparisons and contribute to potential penalties from CMS. We aimed to determine the incidence of BSI, specifically CLABSI, following receipt of ECMO at one of the largest ECMO centers in the US.
Methods
Adults who received ECMO at Duke University Hospital from 1/1/2014–12/31/2016 were included in the study. Cases were patients who acquired BSI during the ECMO exposure period, defined as 2 days after cannulation through 7 days after decannulation. Electronic medical records of case patients were reviewed and data were abstracted using a standardized template. To calculate CLABSI incidence rates (IR), we assumed that all patients on ECMO had 1 or more central venous catheters (CVC) for the duration of ECMO.
Results
426 patients received 3532 days of ECMO during the 3-year study period. 29 (6.8%) patients acquired BSI (IR 8.2 /1000 ECMO days (ED)) after a median ECMO duration of 7 (range 2, 39) days. Of these, 13 met criteria for primary CLABSI (IR 3.7/1000 ED), whereas 9 had a single blood culture (BC) positive for a common commensal organism and 7 had BSI secondary to pneumonia. Although ECMO patients only represented 8% of CVC days during the study period, they accounted for 22% of reported CLABSIs for the medical (MICU) and cardiothoracic surgical ICUs (CTICU). 10 (77%) CLABSI patients had femoral sites of ECMO cannulation and/or CVC insertion and 12 (92%) were receiving broad-spectrum antibiotics at the time of CLABSI. Patients with CLABSI and secondary BSI predominantly had infections due to gram-negative rods (Fig 1). The organism recovered from the BC was susceptible to the antibiotics received in 14 (48%) of patients with positive BC.
Conclusion
The rate of CLABSI was more than 3 times higher in our cohort of ECMO recipients than nationally reported rates for academic MICUs and CTICUs. Research is needed to understand the preventability of these infections, as traditional prevention measures such as avoidance of the femoral site and prophylactic ABX may not be applicable or effective in this high-risk population. Furthermore, NHSN should update its CLABSI definition and/or risk adjustment to account for hospitals with high ECMO utilization.
Disclosures
M. Daneshmand, Maquet: Grant Investigator, Grant recipient; D. Sexton, Centers for Disease Control and Prevention: Grant Investigator, Grant recipient; Centers for Disease Control and Prevention Foundation: Grant Investigator, Grant recipient; UpToDate: Collaborator, Royalty Recipient