Possibility exists also to directly drain left heart chambers. Fumagalli et al. [50] reported draining blood from the LV, through a percutaneously placed transaortic cannula, pumping directly into the femoral artery, with a normalization of left heart filling pressures, and resolution of pulmonary edema, as a bridge to heart transplantation.

Several experimental animal models reported a significant reduction of LV preload, in peripherally inserted ECMO, comparing pre- and post-transaortic cannula insertion in one study [51] and a significant reduction in LV total energy and work in a second publication. The LV energetic charge was significantly increased by a combination of transaortic cannula and peripheral ECMO. A third study compared four different conditions: baseline, during isolated ECMO, ECMO with transaortic venting cannula, and a combination to the previous two with IABP, showing that venting reduced LV energy and work, compared with other techniques alone [52].

The Impella is a minimal intra-aortic impeller blood pump, a form of minimally invasive LV assist device (LVAD) that can be used to support the cardiogenic shock patient in the short and medium term. It can be positioned in the LV, via open chest, but also advanced through the femoral artery over a guidewire until reaching its definitive position. It is designed as a support device, but several works report its use as an adjunctive mechanical support for ECMO patients, in which LV decompression is indicated. Its main effects can be summarized as follows: it directly unloads the LV and it reduces myocardial workload and oxygen consumption while increasing cardiac output and coronary and end-organ perfusion. Chaparro and colleagues [53] reported for the first time the use of combined Impella and ECMO for biventricular and respiratory failure, as a bridge to recovery, in a myocarditis patient. Beuthered et al. [54] also reported the case of a 34-year-old woman with fulminant myocarditis needing ECMO support and subsequently an Impella device to decompress the LV, for acute pulmonary edema. Although reporting effective unloading of the LV, the paper also underlines the occurrence of Impella pump failure, to underline technical challenges in managing complex mechanical assistance. Koeckert et al. [55] reported successful use of Impella device for ECMO complicated by pulmonary edema, with weaning for myocardial recovery. Vlaesselars [56] also reported similar combination in a pediatric patient with congenital cardiomyopathy. Interestingly, the procedure was echocardiography guided.

There is no established gold standard to evaluate optimal LV decompression.

In literature, the most frequently assessed parameters include echocardiographic inspection of the heart chamber size, aortic valve opening, and Doppler evaluation of flow velocities. Most often, published literature reports effective LV decompression clinical resolution of symptoms, such as disappearance of pulmonary edema and hemoptysis and preload pressure reduction. This represents the clinical goal that is desirable to achieve. We will not further discuss hemodynamic monitoring but refer the reader to the specific section.

Nonetheless, it is worthwhile to mention the possibility of TEE evaluation of the proximal part of left coronary arteries and estimation of blood flow velocity in the left anterior descending (LAD) coronary artery, by means of pulsed Doppler. The left main is visualized as an echo-free space, by placing the transducer just above the aortic leaflets, with adjustments needed to fully visualize the length of the vessel. Once the Y-shaped bifurcation between LAD and circumflex artery is visualized, coronary blood flow velocity can be evaluated by pulsed Doppler. There are no reports, to our knowledge, of this technique applied to ECMO patients. Although coronary arteries might be more challenging to visualize due to chamber decompression, still an attempt to visualize non-pulsatile or pulsatile (especially in combination with IABP or native valve opening) flow on the left main coronary artery may be useful to assess perfusion of a recovering heart.

Although the measurement of cardiac biomarkers may seem appealing to evaluate LV rest and unloading as well as to prognosticate outcome, Luyt et al. [57] reported that serial measurement of N-terminal fragment of the B-type natriuretic peptide and troponin I-C and midregional fragment of the proatrial natriuretic peptide, proadrenomedullin, and copeptin have no role as prognostic markers in refractory cardiogenic shock patients rescued by ECMO.

In conclusion, a combination of clinical and radiographic resolution, hemodynamic parameters, and standard echocardiography is up to the present days the standard to evaluate effective LV rest and decompression.