Balanced Biventricular Assist Versus Extracorporeal Membrane Oxygenation in Cardiac Arrest

Abstract

Mechanical assist devices in refractory cardiac arrest are increasingly employed. We compared the hemodynamics and organ perfusion during cardiac arrest with either veno-arterial extracorporeal membrane oxygenation (ECMO) or biventricular assisted circulation combining left- and right-sided impeller devices (BiPella) in an acute experimental setting. Twenty pigs were randomized in two equal groups receiving circulatory support either by ECMO or by BiPella during 40 minutes of ventricular fibrillation (VF) followed by three attempts of cardioversion, and if successful, 60 minute observation with spontaneous, unsupported circulation. Hemodynamic variables were continuously recorded. Tissue perfusion was evaluated by fluorescent microsphere injections. Cardiac function was visualized by intracardiac echocardiography. During VF device output, carotid flow, kidney perfusion, mean aortic pressure (AOPmean), and mean left ventricular pressure (LVPmean) were all significantly higher in the ECMO group, and serum-lactate values were lower compared with the BiPella group. No difference in myocardial or cerebral perfusion was observed between groups. In 15 animals with sustained cardiac function for 60 minutes after return of spontaneous circulation, left ventricular subendocardial blood flow rate averaged 0.59 ± 0.05 ml/min/gm during VF compared with 0.31 ± 0.07 ml/min/gm in five animals with circulatory collapse (p = 0.005). Corresponding values for the midmyocardium was 0.91 ± 0.06 vs. 0.65 ± 0.15 ml/min/gm (p = 0.085). Both BiPella and ECMO could sustain vital organ function. ECMO provided a more optimal systemic circulatory support related to near physiologic output. Myocardial tissue perfusion and sustained cardiac function were related to coronary perfusion pressure during VF, irrespective of mode of circulatory support.

Effects of Add-On Left Ventricular Assist Device to Extracorporeal Membrane Oxygenation During Refractory Cardiac Arrest in a Porcine Model

“…Coronary perfusion pressure, calculated as the difference between mean aortic- and mean LV pressure during VF, was previously found to be a predictor of ROSC and postresuscitation cardiac function, with a critical pressure difference of 10–15 mm Hg. 14,15 This study seems to provide further support and indicate the benefit of additional CPP increase. Consequently, it may be appropriate to monitor CPP clinically during ECPR, for instance, as part of shock and CA studies.…”

“…Impella CP performance was maximized limited by suction to keep LV pressure as low as possible, although more aggressively in group 2 due to the intent of replacing reduced ECMO flow. 14,15…”

“…Impella CP performance was maximized limited by suction to keep LV pressure as low as possible, although more aggressively in group 2 due to the intent of replacing reduced ECMO flow. 14,15 Endpoints Main group comparisons were as follows: (1) Hemodynamics: Device output, mean aortic-and left ventricular pressure (mAP, mLVP), and CPP (= mAP-mLVP). ( 2) Flow and tissue blood-flow rate (perfusion): Mean left descending artery-and pulmonary flow (mLADf, mPAf) and assessment of tissue perfusion using microsphere quantification.…”

“…14 Further, an increase in LV pressure and subsequent reduction of coronary perfusion pressure (CPP) was related to poor post-ROSC cardiac function. 15…”

“…lww.com/ASAIO/A689). 14,16 Paralytic agents were not administered except for 10 mg of vecuronium intravenously directly before attempted cardioversion. All hemodynamic and transittime flow variables were continuously recorded and digitized (ACQ-7700, Data Sciences International, St. Paul, MN).…”