Central VA-ECMO #4: ECMO Weaning and Decannulation

1. Concept of Weaning from Central VA-ECMO

Weaning from central venoarterial extracorporeal membrane oxygenation (VA-ECMO) is the controlled transfer of circulatory responsibility from the extracorporeal circuit back to the native heart and lungs. In postcardiotomy pediatric and congenital heart surgery, successful separation requires more than visible improvement in ventricular contraction. The native circulation must generate adequate forward flow, tolerate physiologic preload and afterload, maintain acceptable filling pressures, support gas exchange, and preserve systemic oxygen delivery without unsustainable pharmacologic support.

There is currently no validated pediatric postcardiotomy protocol that defines universal flow targets, echocardiographic thresholds, trial-off duration, vent-management strategy, or decannulation criteria. Available evidence supports a multiparametric approach, with left ventricular ejection fraction (LVEF), left ventricular outflow tract velocity-time integral (LVOT VTI), right ventricular performance, hemodynamic tolerance, and biochemical recovery being the most consistently associated variables [1].

Accordingly, ECMO weaning should be considered a dynamic cardiovascular stress test rather than simply a reduction in pump flow.

2. Physiologic Transition from ECMO Flow to Native Cardiac Output

During high-flow VA-ECMO support, venous drainage diverts a substantial portion of systemic venous return away from the native heart. Native ventricular filling and stroke volume may therefore remain relatively low even when myocardial contractility has begun to recover.

As ECMO flow is progressively reduced:

1. Less systemic venous return is diverted into the circuit.
2. Right ventricular preload increases.
3. Native pulmonary blood flow increases.
4. Pulmonary venous return increases left-heart preload.
5. The systemic ventricle must generate progressively greater stroke volume.
6. Native cardiac output assumes an increasing proportion of systemic perfusion.

This process simultaneously tests both ventricles. The right ventricle must accept increased venous return and deliver blood through the pulmonary circulation, while the systemic ventricle must accommodate increased preload and generate adequate forward output against systemic afterload.

The conceptual endpoint is not simply “low ECMO flow,” but a circulation in which native cardiac output can replace extracorporeal flow without hemodynamic or metabolic deterioration.

3. Determine Readiness Before Reducing Support

Active weaning should begin only when the underlying cause of cardiac failure appears sufficiently reversible and the patient has reached reasonable physiologic stability.

Assessment should address:

• Recovery from myocardial stunning or ischemia-reperfusion injury
• Adequate rhythm and atrioventricular synchrony
• Corrected acidosis and major electrolyte disturbances
• Appropriate coronary perfusion
• Adequate preload without major volume overload
• Acceptable systemic and pulmonary vascular resistance
• Adequate lung recruitment and gas exchange
• Absence of an important residual structural lesion

In congenital heart surgery, the anatomy of the reconstructed circulation is particularly important. Ventricular performance cannot be interpreted independently of residual outflow obstruction, coronary insufficiency, pulmonary artery obstruction, atrioventricular valve regurgitation, residual intracardiac shunting, or abnormal loading conditions.

The relevant question is therefore not simply whether ventricular contractility has improved, but whether the entire reconstructed circulation can function when extracorporeal support is withdrawn.

4. Echocardiography: Assess Forward Flow, Not Contractility Alone

Echocardiography is central to the weaning assessment. The most useful evaluation combines ventricular systolic function with direct evidence of effective forward flow.

In pediatric VA-ECMO, a prospective multicenter study of 63 patients demonstrated that successful weaning was associated at minimum ECMO flow with higher ejection fraction, higher VTI, less mitral regurgitation, and lower central venous pressure. EF >41% and VTI >7.9 cm independently predicted successful weaning in that cohort [2]. These values are clinically useful reference points but should not be interpreted as universal pediatric decannulation thresholds.

Preliminary prospective pediatric data similarly demonstrate that patients successfully separated from VA-ECMO have higher LVEF, higher VTI, and better right ventricular function during weaning assessment [3].

LVEF

LVEF provides a readily available measure of global left ventricular systolic performance. However, EF is load-dependent and may change substantially as ECMO flow is reduced. An apparently acceptable EF during high ECMO support does not prove that the ventricle will tolerate restored preload and systemic workload.

LVOT VTI

LVOT VTI provides a practical surrogate for forward stroke distance. This is particularly important during ECMO weaning because visible myocardial contraction without effective forward ejection may overestimate functional recovery.

In pediatric patients, the association between higher VTI and successful outcome supports assessing forward flow together with ventricular contraction, rather than relying on EF alone [2,3].

5. Evidence from Adult and Mixed VA-ECMO Populations

Adult and mixed-population studies provide additional physiologic guidance, but their absolute thresholds should not be directly transferred to children.

In an 85-patient study, successful VA-ECMO weaning was associated with an LVEF of approximately 40% compared with 30% in unsuccessful patients, LVOT VTI of approximately 15 versus 11 cm, more consistent aortic-valve opening, and better right ventricular function. An LVEF >33.4% predicted successful separation with 93% sensitivity and 72% specificity [4].

Another prospective study evaluated patients during ECMO flow reduction and found that those tolerating minimum support with LVEF >20–25%, aortic VTI ≥10 cm, and lateral mitral-annular systolic velocity ≥6 cm/s were successfully weaned [5].

These values are useful as physiologic benchmarks but are not pediatric postcardiotomy criteria. Body size, heart rate, ventricular geometry, congenital anatomy, and loading conditions differ substantially between adult cardiogenic shock and pediatric congenital heart disease.

6. Dynamic Change May Be More Important Than a Single Threshold

A major principle emerging from the available literature is that the response to flow reduction may provide more information than measurements obtained at one static flow condition.

During ECMO reduction, a recovering heart should demonstrate increasing native forward output as preload returns. This may appear as:

• Increasing arterial pulsatility
• Increasing LVOT VTI
• Consistent semilunar valve opening
• Preserved or improving ventricular systolic function
• Stable or improving blood pressure
• Appropriate ventricular filling without progressive distension

Dynamic tissue-Doppler measurements may also reveal myocardial reserve. Improvement in lateral mitral annular e′ velocity and tricuspid annular S′ velocity during an ECMO flow study was associated with successful weaning in a multicenter cohort [6].

The important physiologic question is therefore: Does cardiac performance improve appropriately when the extracorporeal circuit returns preload and workload to the native heart?

7. Right Ventricular Function Must Be Preserved

VA-ECMO weaning is inherently a biventricular challenge.

As venous drainage is reduced, right ventricular preload increases substantially. The right ventricle must deliver this additional volume through the pulmonary circulation before the left ventricle can receive adequate preload and generate systemic output.

Pediatric data show that better right ventricular function accompanies successful weaning [2,3]. Adult studies similarly demonstrate an association between normal right ventricular performance and successful separation [4].

Therefore, assessment should include qualitative RV systolic function together with appropriate quantitative indices when technically feasible. Increasing central venous pressure, RV dilation, worsening septal shift, or deteriorating RV function during flow reduction should prompt concern even when LVEF appears acceptable.

In patients with pulmonary hypertension, elevated pulmonary vascular resistance, or postoperative pulmonary artery abnormalities, the right ventricle may become the principal limitation to ECMO separation.

8. Hemodynamic and Biochemical Assessment

Echocardiography should never be interpreted in isolation.

Successful weaning should be accompanied by stable systemic perfusion while ECMO flow is progressively reduced. Appropriate findings include preserved arterial pressure, increasing native pulse pressure, stable filling pressures, adequate urine output, and absence of progressive metabolic acidosis.

In the pediatric multicenter study, lower central venous pressure accompanied successful weaning, whereas greater mitral regurgitation was associated with an unfavorable course [2].

Across broader VA-ECMO populations, lower lactate and lower CK-MB at admission have been associated with successful weaning, although these markers are supportive rather than definitive [1]. Trends in lactate are particularly useful for determining whether systemic oxygen delivery remains adequate as extracorporeal support is withdrawn.

Adult data further demonstrate that low post-test LVEF, inadequate blood pressure response, and prolonged ECMO support are associated with weaning failure [7]. These adult thresholds should not be applied numerically to pediatric patients, but the physiologic message is relevant: failure to maintain pressure and ventricular performance under reduced support indicates inadequate circulatory reserve.

9. Support the Recovering Heart and Lungs

Successful separation does not require complete withdrawal of vasoactive support.

Catecholamines and other vasoactive agents may be used as necessary to maintain appropriate contractility and vascular tone. The objective is a stable circulation supported by reasonable pharmacologic therapy, rather than an artificially high blood pressure generated by escalating vasoactive doses.

Rhythm is equally important. Sinus rhythm and atrioventricular synchrony may substantially influence output in postoperative congenital heart patients with impaired ventricular compliance. Temporary pacing should therefore be optimized when clinically necessary.

Adequate pulmonary function must also be established before complete separation. As ECMO gas exchange decreases, the native lungs must provide sufficient oxygenation and carbon dioxide clearance. Ventilation should achieve adequate recruitment while avoiding excessive intrathoracic pressure that impairs venous return or increases RV afterload.

Arterial blood gases should be reassessed as support is reduced.

10. Systemic Afterload During Weaning

Systemic vascular resistance affects both the ECMO circuit and the recovering systemic ventricle.

With centrifugal pumping, systemic hypertension increases resistance to ECMO arterial return and can reduce circuit flow. At the same time, excessive systemic afterload increases ventricular wall stress and may reduce native stroke volume.

Therefore, the goal during weaning is not maximal systemic blood pressure, but adequate perfusion pressure with an appropriate ventricular afterload.

Conversely, excessive vasodilation can produce inadequate coronary and systemic perfusion. Vasoactive therapy should therefore be titrated dynamically as the balance between ECMO flow and native cardiac output changes.

11. Left Atrial Vent Management

An LA or left-heart vent may be required during high-flow central VA-ECMO to prevent left-heart distension and pulmonary venous hypertension.

During assessment for myocardial recovery, continued aggressive venting can reduce left ventricular preload sufficiently to obscure the ventricle's ability to generate meaningful native output. When clinically appropriate, reducing or temporarily stopping the vent allows pulmonary venous return to fill the left heart and provides a more physiologic test of ventricular performance.

The response should be monitored for adequate aortic-valve opening and forward ejection without progressive LV distension, rising filling pressures, or pulmonary congestion.

Importantly, the available evidence does not establish a validated pediatric central-ECMO protocol for the timing of LA vent discontinuation. Vent management should therefore be individualized according to ventricular loading, pulmonary congestion, and the purpose of the weaning trial rather than based on a fixed threshold.

12. Progressive Flow Reduction and Final Separation

ECMO flow should be reduced in controlled stages while repeatedly reassessing the patient.

At progressively lower support, favorable findings include:

• Increasing native arterial pulsatility
• Adequate LVOT VTI
• Consistent semilunar valve opening
• Stable biventricular function
• Acceptable atrial and central venous pressures
• Stable systemic blood pressure
• Adequate oxygenation and ventilation
• Stable or improving lactate and acid-base status
• No escalation to unsustainable vasoactive support

No single variable is sufficient for the decision to decannulate. The strongest current evidence supports combining ventricular function, forward-flow measurements, right-heart assessment, and hemodynamic tolerance [1,2].

The literature does not establish an optimal pediatric minimum-flow target or standardized duration for a trial off support. Adult protocols using absolute flows such as 1–2 L/min cannot be directly translated to infants and small children because they are not indexed to body size.

13. Decannulation and Failure to Separate

Central VA-ECMO decannulation should proceed only after the native circulation demonstrates sustained stability at minimal or absent extracorporeal support.

Because central cannulation requires surgical removal of venous and arterial cannulas, the operative team must be prepared to re-establish support if the circulation deteriorates. After cannula removal, ventricular filling, systemic pressure, rhythm, gas exchange, and surgical hemostasis should be reassessed before definitive completion of the procedure.

The supplied evidence does not define pediatric-specific techniques for central cannula removal, recannulation thresholds, or postoperative management after decannulation.

If separation fails, the response should be mechanistic rather than simply increasing catecholamines. The team should determine whether the limiting factor is:

• Inadequate systemic ventricular recovery
• Right ventricular failure
• Excessive pulmonary or systemic vascular resistance
• Inadequate preload
• Pulmonary dysfunction
• Significant valve regurgitation
• Rhythm disturbance
• Residual structural disease

Failure of a weaning trial provides important diagnostic information. If the mechanism is reversible, further mechanical support may permit additional recovery. If a significant residual anatomic lesion is responsible, catheter-based or surgical correction may be required before another attempt.

14. Practical Principle

Central VA-ECMO weaning can be summarized as:

ECMO flow decreases → native venous return increases → biventricular preload increases → native forward flow must increase → the heart and lungs progressively assume full circulatory responsibility.

The most convincing evidence of readiness is therefore not a single EF value, but demonstrated biventricular reserve and increasing forward output during reduced ECMO support.

In pediatric patients, higher EF and LVOT VTI, preserved right ventricular function, lower CVP, and less significant mitral regurgitation are favorable findings [2,3]. Adult and mixed-population studies reinforce the value of LVEF, LVOT VTI, valve opening, tissue-Doppler response, and hemodynamic stability [4-7].

These parameters should guide a structured physiologic assessment, while recognizing that standardized pediatric postcardiotomy central VA-ECMO decannulation thresholds have not yet been established.

References

[1] Hsu HR, Sekhar P, Grover J, Tian DH, Downey C, Maudlin B, et al. Predictors of successful weaning from veno-arterial extracorporeal membrane oxygenation (V-A ECMO): a systematic review and meta-analysis. PLoS One. 2025;20(3):e0310289. (PLOS)

[2] Punn R, Falkensammer CB, Blinder JJ, Fifer CG, Thorsson T, Perens G, et al. Hemodynamic and echocardiographic predictors of mortality in pediatric patients on venoarterial extracorporeal membrane oxygenation: a multicenter investigation. J Am Soc Echocardiogr. 2023;36(2):233-241. (PubMed)

[3] Natali Roldan G, Sepulveda SE, Medina J, Blando J, Martín A, Pilan ML. Echocardiographic predictors of successful weaning in pediatric VA-ECMO: prospective study in a single national referral center for congenital heart disease. ASAIO J. 2025;71(Suppl 4):61. (Ovid)

[4] Alonso-Fernandez-Gatta M, Merchán-Gómez S, González-Cebrian M, Diego-Nieto A, González-Martín J, Toranzo-Nieto I, et al. Echocardiographic prediction of successful weaning from venoarterial extracorporeal membrane oxygenation. Am J Crit Care. 2022;31(6):483-493. (PubMed)

[5] Aissaoui N, Luyt CE, Leprince P, Trouillet JL, Léger P, Pavie A, Diebold B, Chastre J, Combes A. Predictors of successful extracorporeal membrane oxygenation (ECMO) weaning after assistance for refractory cardiogenic shock. Intensive Care Med. 2011;37:1738-1745. (Springer)

[6] Kim D, Jang WJ, Park TK, Cho YH, Choi JO, Jeon ES, Yang JH. Echocardiographic predictors of successful extracorporeal membrane oxygenation weaning after refractory cardiogenic shock. J Am Soc Echocardiogr. 2021;34(4):414-422.e4. (PubMed)

[7] Cusanno A, Aissaoui N, Minville V, Porterie J, Biendel C, Volle K, Crognier L, Conil JM, Delmas C. Predictors of weaning failure in case of VA ECMO implantation. Sci Rep. 2022;12:13842. (nature.com)