Title Challenges of VA-ECMO Weaning in Low-Flow, Low-Gradient Severe Aortic Stenosis Post-STEMI and ECPR
Introduction
Low-flow, low-gradient (LF-LG) severe aortic stenosis (AS) in post–ST-elevation myocardial infarction (STEMI) cardiogenic shock supported with veno-arterial extracorporeal membrane oxygenation (VA-ECMO) presents complex diagnostic and therapeutic challenges. Retrograde ECMO flow alters cardiac loading conditions and transvalvular flow, complicating accurate assessment of valvular disease.
Clinical Case
A 52-year-old woman with tobacco use disorder, poorly controlled type 2 diabetes, and hyperlipidemia presented with inferolateral STEMI complicated by ventricular fibrillation cardiac arrest. Despite immediate advanced cardiac life support for 40 minutes, return of spontaneous circulation was not achieved. Extracorporeal cardiopulmonary resuscitation and emergent bifemoral VA-ECMO were initiated within the arrest period.
Diagnostic Evaluation: Initial transesophageal echocardiography revealed a severely calcified aortic valve with absent leaflet opening and near cardiac standstill; vitals showed profound shock (BP 54/44 mmHg, SpO₂ 76%). After volume and vasopressor optimization, repeat imaging showed severely reduced LV systolic function with inferolateral akinesis. Coronary angiography demonstrated 100% right coronary artery occlusion, revascularized with a drug-eluting stent. Pulmonary artery catheterization showed preserved cardiac output/index without pulmonary hypertension. On hospital day 2, the patient failed VA-ECMO weaning trials, with flow reduction below 1 L/min consistently causing tachycardia, hypotension, elevated pulmonary pressures, hypoxemia, and transient ST-elevation. TTE showed LVEF 20%, severe RV dysfunction, and low-flow, low-gradient severe aortic stenosis (peak velocity 2.2 m/s, AVA 0.67 cm², dimensionless index 0.23). To distinguish true severe from pseudosevere AS, a low-dose dobutamine stress echocardiogram was performed on hospital day 3 with ECMO flow reduced to 1 L/min: stroke volume increased >20% with rising peak velocity (2.2 to 3.08 m/s) and persistently reduced AVA (0.71 cm²), confirming true severe AS. Pulmonary artery diastolic pressure rose despite improved LV function, indicating AS as the primary barrier to weaning. CT calcium scoring confirmed a bicuspid aortic valve (annular area 342 mm², calcium score 984).
Treatment: Given prohibitive surgical risk from cardiogenic shock, recent STEMI, and ECMO dependence, a multidisciplinary team proceeded with TAVI on hospital day 5 using a 23 mm balloon-expandable valve, deployed without pre-dilation to avoid acute aortic regurgitation under high afterload.
Follow-up: Post-deployment gradient decreased to 2 mmHg with a pulmonary capillary wedge pressure of 11 mmHg. The patient completed a retrograde flow trial the next day and was decannulated uneventfully. She was discharged neurologically intact to rehabilitation on hospital day 13.
Discussion
Our case highlights two challenges: accurately grading AS severity on VA-ECMO and determining ECMO weaning in the presence of severe AS. Gradients, AVA, and Doppler-derived SV are flow-dependent, and ECMO-induced afterload with reduced native flow often underestimates severity. Calcium scoring, less validated in bicuspid valves, still supported severe AS. Dobutamine stress echocardiography showed limited flow reserve with AVA < 1 cm², confirming true severe AS. Weaning failure removed two of the three "slow, full, constricted" principles required in severe AS. Pulmonary artery diastolic and right atrial pressures helped distinguish AS from biventricular failure, guiding TAVI before successful decannulation. No clear guidelines exist for grading LF-LG AS severity or guiding ECMO weaning in this setting.