ABSTRACT
Extracorporeal membrane oxygenation (ECMO) as a bridge to left ventricular assist device (LVAD) implantation has shown promise in improving end-organ function and optimizing outcomes in some critically ill patients, but the practice remains controversial. Retrospective review of patients who received LVADs from May 2008 to September 2016 at a high-volume, tertiary care cardiovascular center was performed. Subjects were Interagency Registry for Mechanically Assisted Circulatory Support (INTERMACS) class 1 patients divided into ECMO bridge and non-ECMO bridge cohorts. Patient demographics, adverse events, and survival at immediate and 1 year postoperative time points were compared between groups. In total, 235 patients received a HeartMate II or HVAD during the study period. Among INTERMACS 1 patients, 18 were ECMO bridge and 17 were non-ECMO bridge. Age, gender and bridge-to-transplant proportions (50% vs. 53%) were similar between groups. The ECMO bridge group had lower hemoglobin (7.9 ± 1.1 vs. 10.2 ± 2.2; p < 0.01), platelet (101 [70] vs. 176 [115]; p < 0.05), and prealbumin levels (10.6 ± 4.3 vs. 17.3 ± 7.7; p < 0.01). Nearly half (n = 8; 44%) of the ECMO bridge patients required packed red blood cell transfusions before VAD and were more likely to be on an epinephrine drip (78% vs. 12%; p < 0.01). However, along with these adjunctive measures, the ECMO bridge did effectively improve hemodynamic profiles by the time of VAD implant resulting in lower central venous pressure (7.7 ± 2.5 vs. 10.4 ± 4.8; p < 0.01) and mean pulmonary arterial pressure (18 ± 9 vs. 32 ± 8; p < 0.01). It also allowed for restoration of end-organ function as noted by comparable creatinine (1.0 [1.2] vs. 1.4 [0.6]) and total bilirubin levels (1.6 ± 1 vs.1.5 ± 1.7) between the two groups. There was no difference in rates of adverse events. Survival at 30 days postoperative and at 1 year (77% vs. 88%; p = 0.6) was similar. This study demonstrates that ECMO bridge is a central component of a multifaceted strategy for stabilization of select patients with severe hemodynamic instability before LVAD implantation. Further studies to optimize patient selection should be further explored.
