By Jonathan Leung, Consultant EM.
Extracorporeal membrane oxygenation [ECMO] -CPR has been viewed as the ‘Magic bullet’ for the future of resuscitation. It seems logical that if the heart and lung cease to work, hopefully temporarily, we should fit a machine to replace it function until recovery. The theory does make sense. However, the million dollar question is does it truly save life with a meaningful neurological outcome?
We will attempt to answer this question with the current limited evidence available.
ECMO is a modified form of cardiopulmonary bypass, which aim to provide oxygen delivery to tissue and organ. It is a closed circuit which contain the draining cannula, centrifugal pump, gas source, oxygenator and the return cannula. The present of the heat exchanger is optional.
There is no venous reservoir as compare to cardiopulmonary bypass [an open circuit]. The pump will provide a continuous laminar blood flow as compare to the physiological pulsatile flow. Hence, the monitoring of BP would be represented by the mean arterial pressure [MAP] in the content of VA-ECMO.
There are many different types of ECMO methods depends on the patient’s conditions.
The type of ECMO depends on 3 factors:-
- Patient’s cannulation sites
- Number of cannula
- ECMO flow
To simplify matters, there are mainly 2 main types of ECMO:-
1) Venous -Venous ECMO [VV-ECMO]
- Indication – Severe respiratory failure i.e. ARDS, chronic lung disease pending for lung transplant
- Cannulation sites:
- Drainage cannula – Any central vein i.e femoral vein – aim to sit in the inferior vena cava
- Return cannula – Internal jugular vein and aim to sit near the right atrium
- ECMO flow rate – 2 litres/min and titrate up to 4 litres/min
- Good cardiac function is essential
ED ECMO have a fantastic rundown of the whole procedures- http://edecmo.org/logistics/vv-ecmo/
To simplify matters, there are mainly 2 main types of ECMO:-
2) Venous- arterial ECMO [VA ECMO]
- Cardiac arrest with a potential reversible cause i.e. thromboembolic disease, toxicological causes, severe hypothermia
- Bridging to heart transplant
- Cannulation sites:
- Drainage cannula: Femoral vein assess and cannula tip at right atrium
- Return cannula – Femoral artery [Peripheral VA-ECMO(standard rescue cannulation)] and proximal aortic arch [Central VA ECMO]
- Independent of cardiac function in the initial rescue phase, but important consideration during recovery phase and ECMO weaning
- This provide a retrograde laminar flow to the arterial system
What is ECMO integrated CPR?
Extracorporeal cardiorespiratory resuscitation (eCPR) has been described in the last two decade and the concept have gain a huge amount of popularity over the last few years.
CPR have always been described as a ‘team sport’. eCPR have bring the team sport to a new level. This should be the future model of how CPR should be run.
1.1) Ongoing mechanical CPR
1.3) Placement of Transoesophageal echocardiogram – intra-arrest echocardiography
1.4) Femoral vein cannulation with central line
1.5) Femoral arterial line cannulation – can provide intra-arrest arterial pressure monitoring
Central access to artery and vein allow the possibility of changing over to ECMO cannula if ECMO perfusion is needed.
The ECMO cannula can then connected with ECMO machine of choice and start the pump. Mechanical CPR should then be stopped as tissue perfusion should be restored. The time to perfusion restoration should be around 20-30 minutes.
Please see ED ECMO website for further detail – http://edecmo.org/logistics/ecpr/
In the field of eCPR, there are a large among of single-armed observational trial i.e. CHEER trial and full of case report of survivors. The survival rate with good neurological outcome can be as high as 54%. On the surface, this seems to be an amazing intervention. However, there are potential for survival and selection bias.
To deep drive into the evidence, we have conducted a systematic review of all the current literature in order to answer the following 3 parts question:-
‘Can [Extra-corporeal membrane oxygenation] application during refractory CPR for adult [cardiac arrest] patient result in neurological intact survival benefit compare with [conventional CPR]?’
|Population||Adult who have suffered Cardiorespiratory arrest and underwent CPR|
|Intervention||Extracorporeal membrane oxygenation aided CPR|
|Outcome||Neurologically intact survival|
Literature inclusion and exclusion criteria:
|Inclusion criteria||Exclusion criteria|
|Age > 18 years old who suffer from cardiorespiratory arrest||Studies include Intra-aortic ballon pump, Extracorporeal Carbon dioxide removal device|
|Studies involving comparison between conventional CPR and ECMO- CPR||Studies involving post cardiotomy unable to wean from Cardiopulmonary bypass|
|Studies involving observational & experimental study||Studies involving acute respiratory distress syndrome / respiratory arrest|
|Limited to human study||Systematic review and meta-analysis|
|Limited to english literature||Case report, Case series, Single arm observational studies|
- 8 studies were identified
- 5/8 studies on In-Hospital Cardiac Arrest and 3/8 on Out Of Hospital Cardiac Arrest
- There are no randomised controlled trial at present time on this topic.
- All of the studies selected are high quality observational studies in attempt to address the question
In hospital cardiac arrest
Chen et al. Cardiopulmonary resuscitation with assisted extracorporeal life-support versus conventional cardiopulmonary resuscitation in adults with in-hospital cardiac arrest: an observational study and propensity analysis. The Lancet. 2008; 372(9638):554–561.
- Prospective single centre observational studies on IHCA with presumed cardiac causes
- Propensity score matching in an attempt to minimise the effect of potential bias and confounder.
- Demonstrated a NNT of 6
- Failed to demonstrate a statistically significant neurologic intact survival benefit at discharge and 1 year.
- This study have subsequently raised the question of whether E-CPR would lead to greater chance of survival with poorer neurological outcome. Nonetheless, the study was not powered to demonstrate neurological outcome benefit as the matched cohort population is was too small.
Shin et al. Two-year survival and neurological outcome of in-hospital cardiac arrest patients rescued by extracorporeal cardiopulmonary resuscitation. International Journal of Cardiology. 2013; 168(4):3424–3430
- South Korean single tertiary centre retrospective observational cohort study
- Demonstrated a significant neurologic intact (CPC-1-2 or MGOS >/=4) survival benefit at discharge, 6 months,1 year and 2 years with a NNT of 6 throughout all outcomes.
- Limitation:- Retrospective observational study
- The conventional CPR group have a very low survival rate, which raised the question about the unaddressed CPR quality during the study.
Out of Hospital Cardiac arrest
- SAVE-J study group:-
- Multi-centre prospective observational study design.
- A larger study cohort was yielded with minimal background characteristic differences. (E-CPR [N=260] and C-CPR [N=194])
- Demonstrate a significantly better neurologic intact survival at 1 month (12.3% [E-CPR], 1.5% [C-CPR], p< 0.0001) and 6 months (11.2% [E-CPR], 2.6% [C-CPR], p= 0.001)
- NNT of 12 at 6 months
Very low survival rate in the conventional CPR group.
- Long term neurologic intact survival benefit have not yet been addressed for OOHCA.
What are the factors associated with better or worse survival benefit?
Is ECMO-CPR a cost effective option?
- The SAVE-J study group have quoted the cost per one Quality adjust life years(QALYs) as $81,575 (£51,810) for ECPR group and $67,040 (£42,579) for CPR group.
- Nonetheless, E-CPR application has become standard of care in Japan with a relatively loose indication as compared to worldwide practice, this can certainly increase the cost per QALYs in Japan. Hence, this figure lacks external validity.
Validity of literature
- The literature is at risk for selection bias
- Most of the studies included are conducted in highly specialised tertiary academic cardiothoracic centre in Asia – Limit external validity.
- OOHCA – an average time from arrest to ECMO implementation is quoted around 50-60 minutes. This highly efficient delivery of ECMO required a very mature multilevel co-ordination from emergency medical service [EMS] to hospital, this factor should certainly be consider when attempted to applied such evidence.
The evidence of eCPR is not perfect and at best CEBM level 2a- of evidence. However, the current evidences has an association of potential neurological intact survival benefit.
However, patient selection is extremely crucial to survival benefit yield. The intervention is not without risk and can potentially raise complex ethical issue i.e.withdrawal decision of ECMO and organ transplant. Hence, eCPR should be consider on an individual basis and relative should be well informed about the application.
As eCPR seems to be promising, the future of resuscitation should include centralisation of cardiac arrest care.