Weaning from VA ECMO

Author: James Day
Reviewers: Aidan Burrell, Chris Nickson

Everything ECMO 008

A 54-year-old man developed acute pulmonary oedema and cardiogenic shock after an anterior STEMI. He was cannulated and started on VA ECMO after revascularisation in the cath lab. He is now five days post cannulation and was extubated 2 days ago.

Q1. How can you determine if he is ready to be weaned from VA ECMO?

Check for the following:

  1. Aetiology of cardiac failure is compatible with myocardial recovery
  2. Requires minimal ECMO support (flows ≤2-3L/min)
  3. Pulsatile arterial waveform present for >24h
  4. MAP> 60 mmHg in the absence of “high-dose” inopressors
  5. Major metabolic disturbances have resolved
  6. Lung function is not severely impaired

Q2. How do you perform a weaning trial?

This is what is required:

  • About 1 hour of time
  • Echocardiographer (either TTE or TOE depending on views)
  • IV heparin bolus (if not anticoagulated or coagulopathic)
  • A weaning proforma
  • Support person (to manage patient and ECMO console)

If the patient does not have a coagulopathy then a small bolus of IV heparin (e.g. 2,500 units; though the dose used varies depending on individual circumstances) should be given to prevent clotting at low circuit blood flows. A baseline transthoracic echo should be performed. One person will need to reduce the flows and watch the patient whilst the other concentrates on the echo.

Many variables can be measured, but the following variables should be measured using echocardiography at a minimum:

  • Aortic VTI
  • TDSa mitral annulus
  • LVEF

The ECMO flow is dropped in 0.5l/min increments for five minutes at a time. The flows are reduced to 1-1.5L/min but no less. The above variables are measured at each increment.

The following haemodynamic variables are observed and recorded at each increment as well:

  • MAP
  • HR
  • CVP
  • PAP (if pulmonary arterial catheter is present)
  • SpO2

The level of vasopressor and inotropic support are also recorded at each increment.

Q3. What is a successful weaning study?

As ECMO flows are dropped, there is increased venous return to the right heart, which when the heart is adequately recovered, will translate into an increase in left ventricular performance. This should offset any deleterious effects of a reduction in ECMO flows. A successful weaning study therefore will have

  • No significant deterioration in the haemodynamic variables
  • Improvements in cardiac function as assessed by echocardiography

Q4. Are there specific values that predict weaning success?

Several studies have looked at specific values1-3. The measurements below are predictive of success on decannulation:

  • Aortic VTI (>10cm)
  • TDSa mitral annulus (>6cms-1)
  • LVEF (>20-25%)

In practice no single number is entirely predictive, rather, a combination of factors which include echocardiographic values, hemodynamic and patient factors are all taken into account in the decision.

At  an ECMO circuit blood flow of 1 L/min the MAP drops to 40 mmHg and the CVP rises to 20 mmHg. The oxygen saturations also drop to 80%. As a result you decide to abandon the weaning study.

Q5. What should you do now?

Turn the ECMO circuit blood flow setting back up to where it was prior to the weaning study and reduce the ventilator support back to where it was pre-study.

Wait 48 hours before attempting another weaning study.

Two days later, another weaning study is performed. This time there is no instability in the haemodynamic variables and the echocardiogram shows an improvement in left ventricle performance on lowering the ECMO blood flows.

Q6. What should you do now?

The patient should be decannulated as soon as is feasible. Femoral arterial cannulae (whether inserted percutaneously or open) and femoral venous cannulae inserted via surgical cut down approach are usually removed in the operating theatre by the vascular or cardiothoracic surgical teams.

In the meantime, the ECMO blood flow can be reduced to ~2.5 L/min pending this. The ventilator should be optimised and the patient should be observed for any signs of differential hypoxia due to secondary respiratory dysfunction given that cardiac function has recovered (see Everything ECMO 007). A low dose inotrope may be restarted prior to decannulation (if not on any) in case cardiac output is still inadequate post decannulation. If there is significant pulmonary dysfunction the patient may need to be transitioned from VA ECMO to VV ECMO.

Q7. What if the patient had failed the weaning study again?

As the number of days on VA ECMO increases the risk of complications increases. It may be that the underlying pathology leading to cardiac failure may not recover and the patient may need to be considered for longer term mechanical support such as a ventricular assist device (VAD).

If the patient is not a candidate for a VAD then the decision may be to withdraw care on ECMO or decannulate with the understanding that the patient would not be a candidate for further ECMO if deteriorates despite medical optimization.

These decisions typically involve in-depth discussions between the intensive care, cardiology and allied health teams and the patient/surrogate to reach a consensus. The timing of the decision to withdraw ECMO support depends on many factors, and further weaning studies may be attempted if further optimisation is possible or there is still some potential for cardiac recovery.


The Alfred ICU approach to weaning VA ECMO was discussed on the EDECMO podcast in an interview with Deirdre Murphy (Former Deputy Director at the Alfred ICU)4.


References and Links

  1. Aissaoui N, El-Banayosy A, Combes A. How to wean a patient from veno-arterial extracorporeal membrane oxygenation. Intensive Care Med (2015) 41:902-905. [pubmed]
  2. Aissaoui N, Luyt CE, Leprince P, Trouillet JL, Leger 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 [pubmed]
  3. Pappalardo F et al. Timing and strategy for weaning from venoarterial ECMO are complex issues. Journal of Cardiothoracic and Vascular Anesthesia (2015) 29:906-911. [pubmed]
  4. EDECMO 24 – Weaning VA-ECMO, with Deirdre Murphy [Internet]. ED ECMO. 2017 [cited 16 March 2017]. Available from: http://edecmo.org/edecmo-24-weaning-va-ecmo-with-deirdre-murphy/

5 comments to “Weaning from VA ECMO”
  1. Pingback: Weaning from VA ECMO – Global Intensive Care

  2. Pingback: Everything ECMO! | Edwin M. Thames

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.