Left-ventricular unloading with Impella during refractory cardiac arrest treated with veno-arterial extracorporeal membrane oxygenation - A systematic review and meta-analysis

Abstract Background Extracorporeal cardio-pulmonary resuscitation (ECPR) is the implementation of veno-arterial extracorporeal membrane oxygenation (VA-ECMO) during refractory cardiac arrest. The role of left-ventricular (LV) unloading with Impella in addition to VA-ECMO ("ECMELLA") during ECPR is unlear. Purpose This is the first systematic review and meta-analysis to characterise patients receiving LV unloading and to compare in-hospital mortality between ECMELLA and VA-ECMO treatment during ECPR. Methods A systematic review and meta-analysis was performed according to the PRISMA checklist on clinical studies with adult patients with refractory cardiac arrest treated with ECMELLA or VA-ECMO as part of ECPR. The meta-analysis was performed by using a random-effects model. Risk of bias was assessed with Cochrane ROBINS-I tool, heterogeneity with Cochrane Q test and publication bias with funnel plots. Results Altogether, 13 study records from 32 tertiary care centers with a total of 1,037 adult patients with ECPR were included, out of which 785 patients (75.7%) received VA-ECMO and 252 (24.3%) ECMELLA support. Compared to VA-ECMO, the ECMELLA group was comprised of more male patients (71.4% vs. 57.3%), more patients with initial shockable ECG rhythms (55.2% vs. 40.3%), lower pre-ECPR pH (7.07 vs. 7.11) and higher lactate levels (10.9 vs. 10.1 mmol/L), as well as higher rates of acute myocardial infarction (75.0% vs. 41.9%) and percutaneous coronary interventions (71.8% vs. 35.4%) (Table). Age, rate of out-of-hospital cardiac arrest and low-flow times were similar between both treatment groups. Compared to ECMELLA, VA-ECMO alone was more frequently used in pulmonary embolism (6.4% vs. 0.4%) and miscellaneous causes of cardiac arrest (19.0% vs. 4.4%). ECMELLA support was associated with a 47% reduced mortality risk [95% Confidence Interval 0.31-0.93] compared to VA-ECMO alone: 123 (48.8%) vs. 539 (68.7%) patients (P=0.026). Patients with ECMELLA had a longer hospital length of stay (18.7 days vs. 10.6 days). Results remained robust in sensitivity analyses after exclusion of smaller studies and gray literature. Conclusions ECMELLA support was predominantly utilised in patients with acute myocardial infarction and VA-ECMO for pulmonary embolism and miscellaneous causes of cardiac arrest across institutions. Indications and frequency of ECMELLA use varied strongly between institutions. ECMELLA support during ECPR was associated with improved survival across studies and might be considered in well selected patients with refractory cardiac arrest. Further scientific evidence is urgently required to elaborate standardised guidelines for the use of LV unloading during ECPR.