Smith et al. [26] have shown in young lambs, by using labeled albumin microspheres injected into carotid perfusion cannula, that relative coronary blood flow was significantly less than in the control group. Kato et al. [27] demonstrated references in puppies, that coronary arterial flow in VA ECMO with carotid arterial cannula decreased even proportionally to ECMO flow increase despite no significant changes in the mean or diastolic pressures in the ascending aorta. And most interestingly, Kamimura et al. [28], in neonatal dogs, proved the difference in coronary blood flow distributed from the proximal arterial cannula (1 cm above aortic valve) being significantly higher than that from the distal cannula (ostium of the brachiocephalic trunk).

He concludes that proximal arterial cannula appears necessary to provide sufficient oxygenated blood to the coronary circulation during V-A ECMO. However, animals in these three experimental protocols were not in cardiac arrest. We speculate that not only distance, but also the angle under which the outflow arterial cannula in FS approach is located, may play the role, causing prograde flow in aortic arch, thus underperfusing coronary arteries. These considerations are of critical importance in cardiogenic shock with persistent lung failure treated by peripheral ECMO to avoid coronary perfusion by poorly oxygenated blood ejected from the left ventricle.It has also been demonstrated that added pulsatility to continuous flow mechanical support improves organ perfusion in terms of blood flow, flow velocity in coronary artery [29,30], energy equivalent pressure and surplus hemodynamic energy, though not having influenced mean carotid pressure [31,32], and even improves a renal perfusion [33].

However, in most of these experiments, a central cannulation approach was used for both inflow and outflow cannulae. This surgical approach is usually not used in cardiac arrest patients [17,18]. We, therefore, investigated whether pulsatility represented by IABP might play a similar beneficial role also in peripheral configuration of FF vs. FS ECMO in the cardiac arrest model. Our results show that in this situation, IABP may have an unfavorable effect on macrocirculation in the FF configuration. We speculate that an intermittent aortic occlusion by the IABP balloon may diminish blood flow available for the aortic root (and thus for coronary arteries), more than for the aortic arch (and thus for the carotids).

These results are in accordance with a report by Sauren [34], where addition of IABP in severely hypotensive ECMO-treated animals was not beneficial; however, the animals were not in cardiac Carfilzomib arrest. On the other hand, the subclavian approach seems to be suitable for IABP assistance because the outflow cannula is located proximal to the IABP balloon. Drakos et al.