Objectives: The aim of this study was to identify the risk factors associated with early mortality after postcardiotomy venoarterial extracorporeal membrane oxygenation. Methods: This is an analysis of the postcardiotomy extracorporeal membrane oxygenation registry, a retrospective multicenter cohort study including 781 patients aged more than 18 years who required venoarterial extracorporeal membrane oxygenation for cardiopulmonary failure after cardiac surgery from 2010 to 2018 at 19 cardiac surgery centers.

Objectives: Refractory postcardiotomy cardiogenic shock is associated with a high mortality, and venoarterial extracorporeal membrane oxygenation can offer acute cardiopulmonary life support. The aim of this study was to identify pre-venoarterial extracorporeal membrane oxygenation risk factors of 90-day mortality. Methods: We retrospectively analyzed 105 consecutive patients supported with venoarterial extracorporeal membrane oxygenation due to refractory postcardiotomy cardiogenic shock. The association between preimplant variables and all-cause mortality at 90 days was analyzed with univariable and multivariable logistic regression. Results: Main surgical subgroups were single noncoronary artery bypass grafting (29%), isolated coronary artery bypass grafting (20%), and 2 and 3 concomitant surgical procedures (31% and 20%, respectively). The median age of patients was 62 years (interquartile range, 52-68 years), and 76% were men. Cardiopulmonary resuscitation was performed in 30% of patients before venoarterial extracorporeal membrane oxygenation initiation. The median duration of venoarterial extracorporeal membrane oxygenation was 7 days (interquartile range, 3-14). The 90-day overall mortality was 57%, and in-hospital mortality was 56%. Forty-seven percent of patients died on venoarterial extracorporeal membrane oxygenation, 51% of patients were successfully weaned, 1% of patients were bridged to heart transplantation, and 1% of patients were bridged to left ventricular assist device. Multivariable logistic regression analysis identified arterial lactate (odds ratio per unit, 1.22; 95% confidence interval, 1.07-14.0; P ¼ .004) and ischemic heart disease (odds ratio, 7.87; 95% confidence interval, 2.55-24.3; P<.001) as independent risk factors of 90-day mortality. Conclusions: In patients with postcardiotomy cardiogenic shock, ischemic heart disease and level of arterial lactate before venoarterial extracorporeal membrane oxygenation initiation were identified as independent pre-venoarterial extracorporeal membrane oxygenation risk factors of 90-day mortality. These risk factors are easily available for pre-venoarterial extracorporeal membrane oxygenation risk prediction and may improve patient selection for this resourceintensive therapy.

Background: There is limited experience of using the MANTA plug-based vascular closure device for percutaneous arterial closure of the femoral artery after venoarterial extracorporeal membrane oxygenation. Objectives: To study femoral artery complications and need for subsequent vascular interventions after percutaneous decannulation of venoarterial extracorporeal membrane oxygenation (VA ECMO) using the MANTA plug-based vascular closure device. Methods: We studied 34 consecutive patients who underwent percutaneous decannulation of VA ECMO using the MANTA device. Primary outcomes were conversion to surgical cutdown of the groin at decannulation (immediate) or later. Secondary outcomes were type of vascular complication necessitating conversion to surgical cutdown of the groin. Results: Six (17.7%) patients had to undergo immediate (n = 3) or late (n = 3) conversion to surgical cutdown of the groin. Of these, three were owing to occlusion of the common femoral artery resulting in insufficient distal perfusion and three owing to bleeding or pseudoaneurysm. The mechanism of failure was complete intravascular deployment of the MANTA device in three patients, incomplete MANTA sealing of the arteriotomy in one patient, MANTA-unrelated thrombotic occlusion in one patient, and unknown in one patient. Surgical cut-down was typically performed with concomitant catheter thrombectomy with or without patch reconstruction of the artery. Conclusion: Percutaneous decannulation of VA ECMO using the MANTA VCD was feasible but a substantial number of patients needed to be converted to unplanned surgical repair, owing to either closure site-located stenosis/occlusion or bleeding. If suboptimal MANTA positioning is suspected, a low threshold for conversion to surgical cutdown of the groin is recommended.

No abstract