Bloodstream infections in mechanical circulatory support device recipients in the International Society of Heart and Lung Transplantation Mechanically Assisted Circulation Support Registry: Epidemiology, risk factors, and mortality

Background: We used multicenter international data from the International Society of Heart and Lung Transplantation Mechanically Assisted Circulation Support (IMACS) registry to determine bloodstream infection (BSI) event rate, independent risk factors, and association with mortality.

Methods: Included were patients registered in IMACS from January 2013 through December 2015, assessed BSI event rate of mechanical circulatory support (MCS) and non-MCS-related BSIs, and conducted univariate and multivariate analyses between BSI with baseline characteristics and mortality.

Results: We documented 1,606 BSIs in 1,231 of 10,171 MCS recipients (12%), with an event rate of 2.43 BSIs/100 patient-months within 3 months after implant (early onset) and 1.03 BSIs/100 patient-months after 3 months (late onset). Of these episodes, 1,378 (85.8%) were non- MCS-related BSI. Increasing body mass index and bilirubin were independent correlates of MCS-related BSI. Independent correlates of non-MCS-related BSI included older age, higher body mass index, previous cardiac surgery, baseline chronic renal disease and dialysis, pre-implant frailty, presence of biventricular assist device, total artificial heart or right ventricular assist device, and Interagency Registry for Mechanically Assisted Circulatory Support category 1. Survival after 3 months after implant of patients who developed early-onset BSI was 56.9% at 24 months vs 77.4% in patients without early-onset BSI (p < 0.001). Early-onset BSI was an independent correlate of mortality at 3 months after implantation (hazard ratio, 2.56; 95% confidence interval, 2.09-3.15; p < 0.001).

Conclusions: Early-onset BSI was associated with significantly increased 24-month mortality. More than 85% of these BSIs were not device related. There is an opportunity for infection prevention practices to decrease the BSI event rate, which may affect 24-month survival. These data can also serve as benchmarking for individual institutions.