This document provides the framework to support provision of ECMO during an escalating pandemic. It should be used to inform local policies and procedures which should be current and reviewed regularly.
Extracorporeal membrane oxygenation (ECMO) is an advanced form of life support – targeted at the heart and lungs. It may be indicated in cases of acute severe cardiac or pulmonary failure that is both potentially reversible and unresponsive to conventional management. Usually delivered in an intensive care unit (ICU), there are two main types of ECMO – venovenous (V-V) and veno-arterial (V-A).
Both types provide respiratory support, but only V-A ECMO provides haemodynamic support. NSW adult patients who require ECMO can currently receive treatment at eight of the 38 tertiary adult ICUs in metropolitan Sydney as well as John Hunter Hospital. Infants and children who require ECMO receive this at the paediatric intensive care units at Children's Hospital, Westmead and Sydney Children's Hospital, Randwick. NSW has a well-established adult ECMO retrieval service which has been operational for more than 10 years, supported by clinicians on a dedicated roster from Royal Prince Alfred and St Vincent's Hospitals. The Kids ECMO Referral Service (KERS) is currently being implemented for paediatric and neonatal ECMO referrals and retrievals in NSW.
Acute respiratory failure is the predominant feature of severe COVID-19. Mortality due to severe COVID-19 associated acute respiratory distress syndrome (ARDS) is similar to severe ARDS from non-COVID-19 causes.1 As such, V-V ECMO may have a role in the management of these critically ill patients.
In a pandemic, the use of ECMO needs to take into consideration whether ECMO is the best treatment option based on the normal ECMO criterion, the patient's ability to benefit from this modality and current system capacity. In NSW, the decision to initiate ECMO follows a multidisciplinary discussion led by an ECMO intensivist. In neonates and children all ECMO discussions will be conducted through KERS after referral to newborn and paediatric emergency transport service (NETS).
[back to top]
The World Health Organization currently recommends that patients with ARDS who have refractory hypoxaemia despite optimised conventional management and are in settings with access to expertise in ECMO, should be referred for consideration of ECMO. Significant international (and some smaller Australian) experience exists in the use of V-V ECMO for COVID-19. An international cohort study published by the Lancet in October 2020 looked at data from the Extracorporeal Life Support Organisation (ELSO). The study looked at 1,035 patients from 213 hospitals in 36 countries with confirmed COVID-19 who had ECMO support. At 90 days post ECMO initiation there was a 36% mortality rate whilst 30% were able to be discharged from hospital. A multicentre French study of 302 patients with a median age of 52 revealed 46% of patients were alive at 90 days with a mortality rate of 54%.2
A similar survival rate for COVID-19 and non- COVID-19 respiratory failure patients on V-V ECMO has been demonstrated.3 The duration on V-V ECMO for COVID-19 patients appears to be six or more days longer than non-COVID-19 V-V ECMO patients. Hospital length of stay is also very long.4 Similarly good survival has been demonstrated in other health systems.5
The overwhelming majority of COVID-19 cases in children do not require hospitalisation and respiratory failure is rare. Nevertheless, ECMO has been used to manage a small number of cases with severe SARS‐CoV‐2‐associated paediatric ARDS in immunocompetent and immunosuppressed older children and adolescents.6 7 Children who present with paediatric inflammatory multisystem syndrome temporally related to SARS‐CoV‐2 (PIMS-TS), are at risk of acutely developing refractory shock with cardiac and coronary dysfunction and may benefit from VA- ECMO in extreme cases.8
[back to top]
It is important to optimise conventional therapy prior to consideration of ECMO. This includes but is not limited to immunomodulating drugs, neuromuscular blockade, judicious fluid management, optimised lung protective ventilation, appropriate positive end expiratory pressure (PEEP) and most importantly prone ventilation. Hence, ECMO is not recommended as first- line therapy.
ECMO is contraindicated in patients with preexisting conditions which are incompatible with recovery (e.g. severe cardiac or baseline respiratory disease or severe neurological injury, end stage malignancy) with the exception of bridge to transplant for a few patients.
V-V ECMO may be utilised for patients with COVID-19 and severe respiratory failure with expected outcomes comparable to patients supported with V-V ECMO pre-pandemic.4
V-A ECMO may be utilised for patients with COVID-19 and severe cardiac failure; however, the experience is more limited, and outcomes are worse than respiratory ECMO.4
The usual contraindications to all ECMO apply to the COVID-19 patient group as they do to non-COVID-19 patients.
As ECMO is a highly specialised service requiring higher ratios of skilled staff in initiation, maintenance and post treatment phases, its use during a pandemic should be carefully considered:
Note, If a TIER 2 centre does not have the resources or capacity to provide ongoing care for the ECMO patient such that a transfer to a TIER 1 centre is anticipated, early discussion with the TIER 1 centre, ideally prior to initiation of ECMO, should occur.
The chart indicates eligibility and expected outcome for V-V ECMO according to patient age and risk score (1-5) which is the diagnostic group 1, 2 or 3 plus presence of acute (+1) and/or chronic modifiers (+1).
Green - Good expected outcome
Yellow - Uncertain expected outcome
Red - Poor expected outcome
Black - Negligible benefit
Higher likelihood of survival and long-term benefit: should receive ECMO if patients fulfil severity criteria and adequate resources are available. Refer early and prepare for ECMO initiation.
Uncertain outcome benefit following the application off ECMO: application of ECMO in this group should only occur following detailed patient assessment and discussion with a senior panel of ECMO consultants. Application of ECMO will depend on specific patient and disease factors and ECMO capacity.
Low likelihood of patient benefit with the application of ECMO: application of ECMO where these processes are present should be avoided. Application of ECMO on compassionate grounds MUST involve a senior panel consensus.
Increase score by one for the presence of chronic or acute modifiers.
If both acute and chronic modifiers are present increase score by two.
* Severe acute restrictive lung disease with relatively clear CXR (early) is suggestive of cryptogenic organising pneumonia (bronchiolitis obliterans with organising pneumonia) and biopsy should be performed prior to instituting ECMO to distinguish from obliterative bronchiolitis.
ACI intensive Care
Consulted with Senior Clinicians from St Vincent’s and Royal Prince Alfred Hospitals that provide ECMO Retrieval service. Consulted with Paediatric Intensive care clinicians.
Dr Nigel Lyons, Deputy Secretary, Health System Strategy and Planning, NSW Ministry of Health
To assist clinicians with ECMO in COVID-19 pandemic