Accordingly, the direction of outflow arterial flow and distance from coronaries is also of importance [28]. Our results add to the controversy about the effect of IABP-generated pulsatility both on macro- and microcirculation; in small non-randomized clinical apply for it studies, Jung [36,37] and den Uil [38] have reported on a favorable effect of IABP on microcirculation in cardiogenic shock, but Mustermann [39] showed paradoxically improved microvascular flow after withdrawing the IABP. Microcirculation changes definitely play a key role in critical states [40-42] and the role of pulsatility and different support combinations (that is, central vs. peripheral, subclavian vs. femoral) on microcirculatory changes remains to be further studied.

An important observation of our study is that VA ECMO significantly increases coronary perfusion pressure over time, mainly in the FF configuration. CoPP actually reached the baseline values by the end of our protocol, that is, approximately two hours after the ECMO onset (Figures (Figures33 and and4,4, Additional file 4 and Table Table3).3). When pooling data from all animals, the CoPP increased progressively from 15.0 mmHg at the end of the cardiac arrest to 34.0 mmHg five minutes after starting ECMO, rose to 53.0 mmHg at the ECMO switch and continued to rise to 68.0 mmHg before CPR, not different from baseline. The increases in CoPP were more pronounced during the FF ECMO period of the protocol, which was responsible more than FS ECMO for an overall CoPP increase.

In contrast to FF ECMO, where baseline to before CPR values reached the same values, in the FS ECMO starting cohort, the CoPP at the end of the protocol still remained significantly lower compared to baseline (45.5 vs. 78.5 mmHg, respectively, P = 0.041). To our knowledge, this phenomenon of the CoPP increase over time during VA ECMO treated cardiac arrest has not been described before. This finding is important, because CoPP is a key prerequisite for ROSC in prolonged cardiac arrest [43,44]. We can only speculate about the pathophysiological mechanism behind the low CoPP during ECMO start and gradual CoPP increase with time on ECMO; however, vasoplegia with low peripheral resistance post low flow cardiac arrest followed by improved vasoreactivity during further ECMO reperfusion offers a reasonable explanation.

We intentionally did not use vasopressors in this phase of the experiment in order to observe the ?natural” course of ECMO reperfusion. In human care, we are used to keeping the perfusion pressure (that is, MAP) within optimal range with norepinephrine to assure adequately both cerebral flow and perfusion pressure.The adequacy of CoPP was reflected by a high resuscitability of our animals (see Additional file 5), despite a rather Cilengitide strenuous and prolonged protocol.