📚 Guideline basis
ASE guidelines for echocardiography in the management of patients on mechanical circulatory support; EACVI/ELSO position papers on imaging in extracorporeal life support; ASE 2025 diastolic update (LVAD-specific filling pressure indicators); Aissaoui N, et al. on echocardiographic VA-ECMO weaning (Intensive Care Med 2011).
The governing principle
On mechanical support, the echocardiogram no longer measures systemic cardiac output. LVOT VTI measures native ejection only; the circuit supplies the rest. Every parameter in this book that assumes the left ventricle is the sole source of systemic flow must be re-derived:
Total systemic flow = circuit flow + (SV_native × HR)A patient on VA-ECMO at 4 L/min with a closed aortic valve has an LVOT VTI approaching zero and a perfectly adequate circulation. Reporting "no forward flow" in that patient is a category error.
Pre-cannulation assessment
The echocardiogram before cannulation is a screening study for contraindications, and it is time-critical.
Finding | Consequence |
Significant aortic regurgitation | Contraindicates VA-ECMO (retrograde flow drives regurgitation, causing catastrophic LV distension) and Impella (device sits across the valve); relative contraindication to IABP |
LV or LA thrombus | Contraindicates Impella; embolic risk |
Aortic dissection | Contraindicates VA-ECMO and IABP |
Ventricular septal defect | Impella increases the left-to-right shunt |
Severe RV failure | An isolated LV device will not suffice; consider VA-ECMO or RV support |
Tamponade | Treat the tamponade rather than cannulating |
Severe peripheral vascular disease | Influences cannulation site |
Mechanical aortic valve | Contraindicates Impella |
Patent foramen ovale | Right-to-left shunting risk during VV-ECMO and with RV support |
Excluding aortic regurgitation is the single most important pre-cannulation echocardiographic responsibility. It takes seconds and prevents a catastrophic complication.
Cannula and device positioning
Device | Target position | Echocardiographic confirmation |
VV-ECMO, two-cannula | Drainage cannula in the IVC near the cavo-atrial junction; return cannula in the RA directed at the tricuspid valve, or in the SVC | Subcostal and ME bicaval views; assess recirculation by colour Doppler between drainage and return ports |
VV-ECMO, dual-lumen | Return port directed at the tricuspid valve; drainage ports in SVC and IVC | ME bicaval and RV inflow views; colour Doppler confirms the return jet crossing the tricuspid valve |
VA-ECMO, peripheral | Venous drainage cannula tip in the RA or IVC; arterial cannula in the femoral artery, flow retrograde in the descending aorta | Subcostal/bicaval for the venous cannula; descending aorta on TEE for the arterial jet |
Impella | Inlet in the LV ~3.5 cm below the aortic valve; outlet in the ascending aorta | ME long axis or deep transgastric; measure inlet-to-annulus distance; confirm the outlet above the valve |
IABP | Tip distal to the left subclavian artery origin, above the coeliac axis | Suprasternal notch on TTE; descending aorta and arch on TEE |
Durable LVAD inflow cannula | Aligned with the mitral inflow, directed toward the mitral valve, free of septal or free-wall obstruction | Apical views; colour and spectral Doppler of the inflow cannula |
Cannula malposition presents as inadequate flow, high circuit pressures, "chatter" or suck-down events, or recirculation — and TEE resolves all four faster than any other modality.
VA-ECMO: the physiology and its echocardiographic signatures
Peripheral VA-ECMO provides flow but increases LV afterload by delivering retrograde arterial flow against which the native ventricle must eject. In a severely failing ventricle this produces LV distension.
LV distension — recognise it early
Sign | Detail |
Aortic valve fails to open | Assessed by M-mode across the aortic valve; count opening frequency (every beat, every second beat, never) |
Increasing LV end-diastolic dimension | Serial measurement in PLAX |
Spontaneous echo contrast or thrombus in the LV or aortic root | Stasis; a marker of both distension and thromboembolic risk |
Worsening pulmonary oedema and pink frothy secretions | The clinical correlate |
Rising LA size, worsening mitral regurgitation | Backward transmission |
Distension causes pulmonary oedema, LV thrombus, and prevents myocardial recovery by maintaining high wall stress. Management: reduce ECMO flow if tolerated, add inotrope to promote ejection, reduce afterload, or vent the LV mechanically (Impella, atrial septostomy, surgical vent).
Differential hypoxaemia (Harlequin / north–south syndrome)
In peripheral VA-ECMO with recovering native cardiac function and poor native lung gas exchange, poorly oxygenated blood ejected by the native heart perfuses the aortic root, coronaries and cerebral vessels, while well-oxygenated retrograde circuit blood perfuses the lower body. Right radial arterial blood gas and right-hand pulse oximetry detect it; echocardiography contributes by demonstrating recovering native ejection with aortic valve opening. Management: improve native lung ventilation, or convert to veno-arterial-venous configuration.
Weaning VA-ECMO
Staged flow reduction with echocardiographic assessment at each step. The most-cited echocardiographic predictors of successful weaning, from a prospective cohort at minimal flow, are:
Parameter | Threshold associated with successful weaning |
Aortic VTI | ≥ 10 cm |
LVEF | > 20–25% |
Lateral mitral annular tissue Doppler s′ | ≥ 6 cm/s |
These are cohort-derived predictors, not guideline thresholds, and they were established at low flow rather than off support. Also assess during the trial: aortic valve opening on every beat, RV function (RV failure is a common reason for failed weaning), absence of significant MR, and the absence of rising filling pressures.
VV-ECMO
The circuit does not support the circulation, so echocardiography retains its normal meaning for cardiac output — with two additions:
Recirculation occurs when oxygenated return blood is immediately drained back into the circuit. Suspect it when circuit oxygenation is high but patient oxygenation is poor. Colour Doppler in the ME bicaval view demonstrates return flow directed toward the drainage cannula rather than across the tricuspid valve. Management: reposition the cannula, reduce flow, or increase intravascular volume.
RV assessment during weaning matters because the RV is the ventricle at risk in the hypoxaemic, hypercapnic, high-driving-pressure patient. Reducing sweep gas during a weaning trial raises PaCO₂ and PVR; a rising RV:LV ratio and new septal dyskinesia during the trial indicate that the RV, not the lung, is the limiting factor (Chapters 9, 32).
Impella and IABP
Device | Echocardiographic monitoring |
Impella | Inlet position (repositioning is frequent — check daily and after any patient movement); aortic valve for new regurgitation caused by the device; haemolysis prompts a check for inlet malposition or suction against the septum; LV size as a marker of adequate unloading |
IABP | Tip position; augmentation is visible as a diastolic flow signal; new or worsening AR is an indication to remove the device |
Durable LVAD
Assessment | Findings |
Ramp study | Stepwise pump speed increases with measurement of LV end-diastolic dimension, aortic valve opening frequency, and septal position at each speed; identifies the optimal speed |
Optimal unloading | LV decompressed, septum midline, aortic valve opening intermittently (typically every third beat or so, per institutional protocol), mitral regurgitation reduced |
Suction event | Sudden LV cavity collapse with the inflow cannula abutting the septum or free wall; often with ventricular arrhythmia; reduce speed and give volume |
Under-unloading | Dilated LV, aortic valve opening every beat, septum shifted right, persistent MR |
Over-unloading | Small LV, septum shifted left, aortic valve never opening, new or worsening tricuspid regurgitation from RV geometric distortion |
RV failure post-implant | A leading cause of early mortality; RV dilatation, falling TAPSE, rising CVP; the septum shifted left by over-unloading worsens it |
Filling pressures | ASE 2025 lists LVAD-specific indicators of elevated LV filling pressure: E/A > 2, RAP > 10 mmHg, PASP > 40 mmHg, average E/e′ > 14 or septal ≥ 15, LAVi > 33 mL/m², and interatrial septal position |
Interatrial septal position provides a direct LAP–RAP comparison: neutral implies LAP = RAP; bulging right implies LAP exceeds RAP by roughly 5 mmHg; bulging left implies LAP is roughly 5 mmHg lower.
ICU-Specific Limitations
Confounder | Effect | Response |
Non-pulsatile or reduced-pulsatility flow | VTI, gradients and Doppler-derived pressures lose their usual meaning | Interpret against circuit parameters, not against normal ranges |
Acoustic shadowing from cannulae and devices | Obscures adjacent structures | Multiple windows; TEE |
Post-sternotomy, open chest | TTE frequently unobtainable | TEE is the default modality in this population |
Anticoagulation and thrombus | New masses on cannulae are common | Confirm in two planes; distinguish from artefact |
Severe TR | Confounds RV and volume assessment | Interpret with hepatic vein Doppler |
🛑 Critical pitfall: Reporting "severely impaired LV function, minimal forward flow" in a patient on VA-ECMO without stating the circuit flow. The finding may be entirely expected and the report will be read as deterioration.
🛑 Critical pitfall: Failing to exclude aortic regurgitation before VA-ECMO or Impella. Retrograde flow into an incompetent aortic valve produces immediate, catastrophic LV distension.
🛑 Critical pitfall: Missing progressive LV distension because the aortic valve was not specifically examined. M-mode across the aortic valve, documenting opening frequency, should be part of every VA-ECMO study.
- 💡 Clinical pearl: Document aortic valve opening frequency on every VA-ECMO and LVAD study. It is the most informative single observation about the balance between native ejection and device unloading.
- 💡 Clinical pearl: Interatrial septal position is a free, continuously available comparison of left and right atrial pressures, and it is particularly valuable in LVAD patients where conventional filling pressure indices are unreliable.
- 💡 Clinical pearl: When VV-ECMO weaning fails, scan the right ventricle before concluding the lung is not ready. Rising PaCO₂ during a sweep-gas reduction trial can precipitate RV failure that looks like respiratory failure.
References
- Stainback RF, Estep JD, Agler DA, et al. Echocardiography in the management of patients with left ventricular assist devices: recommendations from the American Society of Echocardiography. J Am Soc Echocardiogr 2015;28:853–909.
- Aissaoui N, Luyt CE, Leprince P, et al. Predictors of successful extracorporeal membrane oxygenation (ECMO) weaning after assistance for refractory cardiogenic shock. Intensive Care Med 2011;37:1738–45.
- Nagueh SF, Sanborn DY, Oh JK, et al. ASE 2025 diastolic function update (LVAD section). J Am Soc Echocardiogr 2025;38:537–69.
- Douflé G, Roscoe A, Billia F, Fan E. Echocardiography for adult patients supported with extracorporeal membrane oxygenation. Crit Care 2015;19:326.
- Hahn RT, Abraham T, Adams MS, et al. Guidelines for performing a comprehensive transesophageal echocardiographic examination. J Am Soc Echocardiogr 2013;26:921–64.