Varying ECMO Flow Effect on Pulmonary Capillary Wedge Pressure

Extracorporeal membrane oxygenation (ECMO) is a common therapy for patients with cardiogenic shock refractory to conventional medical therapies. In adults, venoarterial ECMO (VA ECMO) typically involves a drainage cannula with retrograde flow up the femoral artery, providing oxygenated blood and hemodynamic support. In traditional models, the downside to increased ECMO flow is increased left ventricular (LV) afterload which, in a failing ventricle, may lead to increased LV end-diastolic pressure (LVEDP) and elevated pulmonary capillary wedge pressure (PCWP). The Extracorporeal Life Support Organization guidelines suggest using the least amount of flow needed to avoid this effect as well as pulmonary edema.¹

The single-center PAPO-Flow Study by Saura et al evaluated the effects of varying flow on PCWP in consecutively enrolled adult patients in a French medical ICU.² Eligible patients were within 48 hours of their femoral-femoral cannulation, deemed stable by the treating physician to tolerate flow variations, and had a pulmonary artery (PA) catheter. Echocardiography-based hemodynamic assessments and PA catheter data were obtained at 2 L/min of flow. ECMO flows increased by 0.5-L/min increments lasting 20 minutes, with hemodynamic evaluations occurring at each step. At the end of the study (4 L/min), patients had a repeat echocardiogram.

Of the 80 patients enrolled, 68% had Society for Cardiovascular Angiography and Interventions shock stage D (deteriorating), 25% had had a previous cardiac arrest, and 11% received extracorporeal cardiopulmonary resuscitation. Baseline values showed that elevated PCWP was associated with elevated central venous pressure (CVP), preserved right ventricular (RV) function, LV dilation, and significant mitral regurgitation. At the final flow titration (4 L/min), PCWP decreased by at least 20% in 36% of patients, remained stable in 58% of patients, and increased by at least 20% in 6% of patients. Those whose PCWP decreased were more likely to have had lower baseline CVP, better biventricular function, lower severity profile, and lower mortality.

The investigators sought to evaluate the impact of varying ECMO flows on PCWP; however, PCWP is only a surrogate for LVEDP, and its binary value does not clinically equate to pulmonary edema. A study limitation was that it did not evaluate patients’ baseline hemodynamics and whether VA ECMO itself was the main driver in varying PCWP or other hemodynamic variables. Additionally, shock, which is a dynamic condition, was studied at only one point in time, and treating clinicians had to deem the patient stable enough (22% of patients were excluded). 

The results attempt to challenge the previous hemodynamic paradigm via the interplay of PCWP and RV/LV hemodynamics. Increased ECMO flow leads to increased LV afterload and mean arterial pressure, which was re-demonstrated in this study. Increased flow also increases right-sided drainage and therefore decreases RV preload. 

It is important to understand the balance of RV preload with LV afterload with respect to flow variation and a patient’s own hemodynamic balance. Though it may not immediately impact clinical practice, it is intriguing as a basis for further studies related to weaning, LV assist device bridging, and LV venting needs.


References

1. Otake M, Morita H, Sato K, Saku K. Impact of venoarterial extracorporeal membrane oxygenation on hemodynamics and cardiac mechanics: insights from pressure-volume loop analysis. Int J Heart Fail. 2025 Jul 7;7(3):125-138.
2. Saura O, Hékimian G, Del Marmol G, et al; PAPO-Flow Study Investigators. Effect of ECMO flow variations on pulmonary capillary wedge pressure in patients with cardiogenic shock. J Am Coll Cardiol. 2025 Sep 16;86(11):768-778.