Background: Cardiogenic shock (CS) is a condition of acute severe heart failure, resulting in hypotension, reduced tissue and endorgan-perfusion and prognostic relevant consecutive multiorgan failure. Mechanical support devices such as Impella and venous-arterial extracorporeal membrane oxygenation (vaECMO) are available to stabilize hemodynamics and optimize organ perfusion in this time crucial situation. However, the use and selection of the appropriate device, especially in situations where biventricular support (ECMELLA) is required, are controversially discussed in the recent literature. In a clinical case series, we measured the hemodynamics and vascular resistance of patients in cardiogenic shock under counteracting high running rates of the Impella and vaECMO to obtain indications of the best possible support and thus mostly optimal use of the devices in cardiogenic shock.

Methods: 10 patients with CS complicating myocardial infarction treated with ECMELLA were included in our case series. VA-ECMO and Impella performance were inversely modified at a predefined variation (±0.7 l/min). Hemodynamic measurements using a Swan-Ganz catheter and measurements of vascular resistance and the renal resistive index (RRI) as a marker for renal organ perfusion were carried out for different times during support. They took place in the initial hemodynamic stable baseline situation (Impella   2.08l/min±0.34, vaECMO 2.6l/min±0.3),  30 minutes after ECMO-flow upregulation and Impella-flow down regulation (Setup 1) and again 30 minutes after ECMO-flow down regulation with simultaneous Impella-flow upregulation (Setup 2).

Results: Increasing VA-ECMO output from Baseline (BL) 2.6±0.3 to 3.3±0.28 l/min and decreasing Impella® output from BL 2.08±0.34 to 1.35±0.31 (Setup 1) lead to a significant decrease of cardiac output (CO) from 5.13±1.3 to 3.86±1.1 l/min (p= 0.023) and of cardiac index (CI) form 2.7±0.76 to 2.81±0.7 (p=0.04). Systemic vascular resistance (SVR) rose significantly from BL 1235±382 to 1618±337 s*cm^-5 (p=0.02). RRI also increased significantly from 0.606±0.07 to 0.652±0.05 (p=0.04).  On the other hand, decreasing VA-ECMO output from BL 2.6±0.3 to 1.88±0.27 and enhancing Impella® support from BL 2.08±0.34 to 2.79±0.42 (Setup 2), lead to an increase of CO (5.12±1.3 to 5.43±1.14 l/min; p= 0.57) and CI (2.7±0.76 to 2.81±0.7; p=0.4) and to a reduction of SVR (1235±382 to 1086±306 s*cm^-5 (p=0.2).  RRI was significantly reduced from 0.606±0.07 to 0.578±0.06 (p=0.04). Also the comparison of Setup 1 (ECMO>Impella) and Setup 2 (Impella>ECMO) revealed significant changes. CO and CI elevated from 3.86±1.1 to 5.43±1.14 l/min (p=0.006) and from 2.7±0.76 to 2.81±0.7 (p=0.04) respectively whereas SVR decreased from 1618±337 to 1086±306 s*cm⁵ (p=0.002) and RRI decreased from 0.662±0.05 to 0.578±0.06 (p=0.004). The mean arterial pressure, heart rate, doses of catecholamines, fluid management and ventilator settings remained without significant changes during the measurements.

Conclusion: The collected data indicate that adequate selection of the appropriate mechanical support device depended to the clinical situation of the patient plays a key role. When biventricular myocardial unloading is considered, a higher Impella flow than vaECMO flow should be strived in order to generate best possible hemodynamic support, optimize end-organ perfusion, and reduce device-associated complications such as hemolysis and bleeding.