Home > library > article reviews

Article Reviews

Efficacy and economic assessment of conventional ventilatory support versus extracorporeal membrane oxygenation for severe adult respiratory failure (CESAR): a multicentre randomised controlled trial

by Neil Orford - Biography

29th Oct, 2009

ECMO in severe respiratory failure
Reviewer’s summary
The CESAR study, published in the Lancet is an important paper in the critical care literature. The finding of improved survival without severe disability at 6-months in critically ill patients with severe adult respiratory distress syndrome treated at an ECMO capable centre is arguably proof enough to raise ECMO to standard therapy for severe respiratory failure not responding to conventional therapy.
The ANZ ECMO Investigator paper published at the same time in JAMA sets the ideal platform to argue the time has come to build the ECMO capability in 1st world tertiary ICU’s. As the southern hemisphere takes stock of the H1N1 surge of severe respiratory failure requiring ventilatory support, it appears that the message is there will be more young adults on ventilators and receiving ECMO in the short-term future.
There are a number of points of debate. Of course, the usual questions about CESAR’s methodology (trial of referral to an ECMO centre, rather than received ECMO, non-standardised conventional ventilation arm etc) exist, and have been addressed in the papers and accompanying editorials. Although the increased length of stay and cost of treatment for ECMO referred patients was discussed by the CESAR investigators, it is arguably the economic and organisational issues around the provision of an ECMO service to a whole population that are most daunting.
Of the screened patients 103 were excluded because there was no ECMO bed available, and 5 patients died before they could be put on ECMO. With 180 patients randomised, this suggests there is a long way to go before ECMO is available to all. With 68 from the conventional arm requiring ECMO, and another possible 30 from the 108 It is possible that many of these had other exclusion criteria, with approximately 60% of screened patients excluded for reasons other than lack of bed availability.
The other major organisational issue is where should ECMO be administered? The CESAR structure was 1 ECMO capable centre with all the experience and skills. In contrast the ANZ ECMO Investigators reported 15 ECMO centres in 2 countries, of which 4 are heart-lung transplant centres. The 2 models seem to be;

The increase in use of ECMO that will occur due to CESAR and H1N1, combined with improved equipment and reduced complications, mean ECMO can be delivered at cardiac surgical centres with adequate case load, multi-site review process to ensure safety.
Evidence for efficacy is confined to quarternary centre, and use should be confined to these, with investment in providing service capability that includes referral, retrieval, and the ability to cannulate and commence at the referring site.

The Papers
Title: Efficacy and economic assessment of conventional ventilatory support versus extracorporeal membrane oxygenation for severe adult respiratory failure (CESAR): a multicentre randomised controlled trial
Authors: G Peek, M Mugford, R Tiruvoipati, et al, for the CESAR trial collaboration
Journal: Lancet 2009;374:1351-63
Background:

ARDS mortality is 34-58%, survivors have significant morbidity.
Conventional ventilation leads to VILI
ECMO alternative that allows lung rest and recovery, prevents VILI,
2 prior ECMO RCT’s with negative result, but “old” and not relevant
Established in neonates, adult role controversial. Increase used with anecdotal success, improved technology and reduced complications

Methodology

Enrolled 3 types of centres.
Inclusion
- Exclusion
  - PIP>30 cmH2O or FiO2>0.8 for >7 days
  - Intracranial bleeding or CI to heparin
- Enrolment
  - Originating hospital intensivist contacted Glenfield to confirm eligibility and bed availability
  - Randomised 1:1
- Treatments
  - Conventional
    - IPPV (Vt 4-8 ml/kg, pplat <30cmH2O)
    - HFOV allowed
  - Consideration for ECMO
    - Transfer Glenfield
    - IPPV pplat <30cmH2O,SpO2>90, prone. If not respond 12 hrs then ECMO
    - V-v, servo controlled roller pumps, poly-methyl pentene oxygenators. Ventilated PCV “lung-rest settings” PIP 20-25, PEEP 10-15, RR 10, FiO2 0.3.
    - Continue until lung recovery or failure
- Outcomes
  - Primary: Survival without severe disability at 6 months
  - Secondary
    - Mortality
    - Cost analysis
    - Duration ventilation
    - HFOV, NO, prone, steroids
    - Duration ventilation, ICU, hospital
    - ECMO - method, duration, flows
    - Health-status 6 months - daily life, QOL, resp, cognitive
- Statistical analysis
  - Power based on 70% mortality in conventional management group
  - Assumed 10% risk severe disability in survivors, 2-sided µ 0.05, ß 0.2, require 120 each group to detect 25% reduction in primary outcome (73% to 55%)
- Results
  - Death and disability by IIT decreased with ECMO referral; ECMO 63% vs IPPV 47% survive without serious disability at 6 months (RR 0.69, 0.05-0.97,0.03)
  - Survival only; Non-significant trend towards ECMO (RR 0.73, 95% CI 0.52-1.03, p=0.07)
  - Length of stay for survivors and non-survivors greater with ECMO
  - Health care costs greater for ECMO
  - Mean QALY’s higher with ECMO, cost per QALY ECMO £ 19,252
- Discussion
  - Analysis not of ECMO, but transfer to ECMO hospital, 25 % did not receive ECMO, 5 died before reaching ECMO hospital
  - Interesting points
    - Majority pneumonia, less than 2 organ failures, over half ventilated <48 hrs, hypoxia (97%)
    - For non-ECMO, a lot of prone and steroids
    - For both, lots of CRRT, not much NO, HFOV
    - ECMO facility not available for 103 patients
  - Why improved outcomes
  - Sustain life long enough to allow recovery?
  - ?Rest the lung and prevent VILI
- Limitations
  - Non-standardised conventional care
  - ECMO at 1 centre, non-ECMO at many
  - Lack of blinding
  - 5 patients died before transfer or in transit, no model for cannulation at referral site.
  - A study of referral to ECMO capable centre, not ECMO

Title: Extracorporeal membrane oxygenation for 2009 Influenza (H1N1) ARDS

Authors: ANZ ECMO Influenza Investigators

Journal: JAMA, Oct 19, 2009

Summary: A descriptive analysis of all adult and paediatric (neonates excluded) patients treated with ECMO 1 June-31 Aug 2009 in Aust, NZ ICU’s, during the 2009 H1N1 epidemic. In 200 ICU’s, 15 supplied ECMO for 61 (32%) of 194 H1N1/Influenza A ventilated patients. Patients young, pregnant, obese, severe respiratory failure, require prolonged support. Mortality 21%.