Measuring Contractility on Mechanical Circulatory Support: If It Cannot Be Done, Does It Matter?

To the Editor: It is with some concern that I read the case report entitled “Ejection Fraction May Not Reflect Contractility: Example in Veno-Arterial Extracorporeal Membrane Oxygenation for Heart Failure” by Morimont et al.1 I do agree with the authors; ejection fraction and cardiac contractility are not at all equivalent, often times do not parallel changes in the other and neither are valid surrogates for unsupported physiology while on circulatory support.2–4 The concept of assessing load-independent function while on mechanical support has been explored over the past several decades both clinically and experimentally.4,5 The assessment of contractility requires that ventricular loading conditions vary and that the ventricular response be assessed to derive the load-independent relationship instantly, within the first few beats, to be valid.6,7 The central assumption in the authors case report is that end-systolic elastance (Ees, slope of the end-systolic pressure-volume relationship (ESPVR) is valid when derived while varying extracorporeal membrane oxygenation (ECMO) support (ventricular volume?) and then the independent response of the ventricle is reliably measured at its end-systolic point by noting the dichrotic notch from the femoral arterial line and the ventricular volume measured by echocardiography. These assumptions are not trivial for sure, but the more important issue remains the conditions during which the studies were performed. The ESPVR is not linear at all volumes; moreover, it is only valid during acute (instantaneous) alterations in load and, thus, has not been modeled at a series of steady state conditions the authors report.6,7 The authors varied ventricular volume quite indirectly, and sometimes unintentionally, by changing veno-arterial ECMO flow rates and then purport to identify an independent and physiologically coupled change in end-systolic pressure. Where the ventricular volume starts and, thus, where on the curvilinear line of Ees the ventricle is assessed are important.3 Most notably, the arterial blood pressure and the end-systolic pressure are not free to vary completely independently in response to the cardiac volume (preload) because the ECMO circuit is supporting some flow and, thus, provides some pressure support to the arterial system. Ironically, the effect of the ECMO support is ever greater during the ramp-up in flow or at higher flow rates, and thus, the end-systolic point (dichrotic notch) tends to remain fixed through large volume changes in the ventricle.3,5 At a minimum, readers and Morimont et al. should very cautiously interpret the Ees derived so indirectly from the femoral arterial pressure (wave amplification), echocardiograms, and under varied inotropic support, but we all should disregard values obtained under conditions that quite directly do not allow for independent ventricular load assessment. ECMO, indirectly, and ventricular assist devices, more directly, are deliberate and potent ventricular–arterial uncouplers—they reduce stroke volume and end-diastolic volume (EDV) while supporting blood pressure, thereby, artificially, instantly, and dramatically increasing Ea (ventricular load). You cannot accurately assess ventricular–arterial uncoupling while directly uncoupling! The heart [likely] attempts to instantly recouple by increasing contractility in the face of such a dramatic rise in Ea but under acute, exhaustive heart failure conditions may fail to sufficiently overcome the afterload (Ea) and, thus, further fails to eject, precipitating ever more cardiac distention—a condition that clinically is encountered all too often. Regardless, we cannot hope, nor should we trust that Ees under conditions of mechanical circulatory support can ever be a valid measure of ventricular contractility; therefore, Ees will always fail to correctly inform the management of our patients while on circulatory support. Patrick I. McConnellCardiothoracic SurgeryNationwide Children’s HospitalHeart CenterColumbus, Ohio