UK Pre Hospital ECPR -The Sub30 Study

Background

We are having a bit of a flurry of Extra Corporeal Membrane Oxygenation (ECMO) on St Emlyn’s at the moment! Following the podcasts from Alice Hutin (SAMU, France) on ECPR and Chris Bishop (C4TS, UK) on trauma ECMO, is the publication of Sub30: a feasibility study of a pre-hospital extracorporeal membrane oxygenation in patients with refractory out-of-hospital cardiac arrest in London. ECMO in cardiac arrest has huge promise. By draining blood from a large vein, removing CO2 and adding oxygen, and then returning it under pressure to a large artery, ECPR has the potential to reperfuse the heart and importantly the brain during cardiac arrest. This can prevent hypoxic injury, and buy time for further investigations and therapies. It’s a technique with some conflicting evidence – see the ARREST trial, PRAGUE-OHCA trial, INCEPTION trial (we’ve written about that on St Emlyn’s) and, probably most importantly, the CHEER3 study which we’ll discuss more in this article (it’s a similar pre hospital ECPR feasibility study from Melbourne). What studies do agree on, is the importance of time between cardiac arrest and starting ECPR, on successful outcome. Currently within an hour is the most widely accepted target. The sub30 team set themselves the ambitious aim of achieving ECPR flow within 30 minutes of cardiac arrest. This is no mean feat!

Sub30 is a first in the UK study – we’ve never had pre hospital ECPR in the UK before. As such, it is potentially transformative for the development of future services here. I’m delighted to see it completed and published, and give huge credit to the team who successfully enabled a system able to facilitate this therapy. As always, please read the full paper for yourself.

Abstract

Aim

Sub30 study is an open-label, prospective, single-arm feasibility study with the primary objective of assessing the logistics, feasibility, and safety of ECPR delivery in a pre-hospital setting for refractory out-of-hospital cardiac arrest patients in London, United Kingdom.

Methods

Open-label, single-arm, feasibility, prospective study.

Results

Forty-three eligible patients were identified by London Ambulance Service over 27 trial recruitment days during a 13-month study period resulting in the despatch of the pre-hospital ECPR team to 18 patients. Five patients met full criteria and were cannulated for ECPR. All patients were male with a median age of 61 years and received ECPR full flows at a mean of 47 min (range 37–59 min) from initial collapse after a median travel time to scene of 14 min (range 3–20 min). No patient met the primary outcome measure of being established on pre-hospital ECPR within 30 min of the call to the emergency services. Out of 5 patients, 3 patients had care withdrawn and 2 survived to hospital discharge (both CPC score 3 and modified Rankin Score (mRS) score 4 and 5 respectively).

Conclusions

Whilst our study did not meet primary outcome of achieving full ECPR flow within 30-minute of collapse, it demonstrated safe, timely and effective delivery of ECPR with comparable survival rates by pre-hospital teams in a large metropolitan city and this has potential to improve outcomes in refractory out-of-hospital cardiac arrest patients.
Singer B, Win Hla TT, Abu-Habsa M, Davies G, Wrigley F, Faulkner M, Finney SJ. Sub30: Feasibility study of a pre-hospital extracorporeal membrane oxygenation (ECMO) in patients with refractory out-of-hospital cardiac arrest in London, United Kingdom. Resuscitation. 2024 Dec 5:110455. doi: 10.1016/j.resuscitation.2024.110455. Epub ahead of print. PMID: 39645021.

What Kind of Study is This?

It’s a single arm feasibility study. It’s not designed or powered to detect benefit, but is being done to identify if the technique is possible.

Tell Me About the Patients

Identifying patients potentially suitable for ECPR is complex. The trial protocol describes the initial screening by an advanced paramedic practitioner (APP) in the London Ambulance Service (LAS) control room. They identified 43 patients during the 27 study shifts, to which the team was dispatched. The team were stood down from 25 of these (58%) as further information revealed that the patient would not be included. They arrived at 18 patients who were screened against the candidacy criteria: witnessed out-of-hospital cardiac arrest, presumed cardiac aetiology, chest compressions within 3 minutes.

6 patients met the candidacy criteria. One gained ROSC, and 5 were ultimately cannulated and placed onto ECPR. We’ll focus on those 5. Note that the study design was for 6 patients, but it stopped at 5 as recruitment had to be paused between Feb 2020 and May 2022, due to the COVID pandemic.

Demographics: All male, median age 55.4 years (39-64)
Cardiac arrest variables: All witnessed, all in a public place.
Aetiology: Myocardial infarction/ischaemia
Bystander CPR: Yes, for all
Initial Rhythm: 4 x shockable, 1 x PEA

What Were the Measured Outcomes in This Study?

Primary: How many patients could be established onto ECPR within 30 minutes of cardiac arrest

Secondary: Organ dysfunction, survival to discharge, neurological outcome at discharge and 90 days, and incidence of ECPR related complications.

What Are the Main Results?

The primary outcome was not met, as none of the patients were established onto ECPR within 30 minutes. That said, the mean time was 47.4 minutes, which is the fastest yet published in the literature for pre hospital ECPR (the next nearest is CHEER3: 50 minutes), and well within the widely accepted target of 60 minutes. A huge strength of sub30 was that the team reliably arrived with the patient very quickly; a mean of 10.8 minutes from 999 to ECPR team with patient. This shows that the deployment model has promise for future services. The mean time from team arrival to ECPR was 32.2 minutes.

Comparing that with CHEER3 shows some differences. In CHEER3 the team took substantially longer to reach the patient, a mean of 27 minutes. In contrast, their mean time from decision to initiate ECPR to establishment of ECPR was 16 min (range 11–26 min). This isn’t quite the same thing as sub30, as their ‘cannulation time’ also includes the assessment and decision making phase. Nonetheless the CHEER3 patients were on ECPR in about the same number of minutes as the sub30 patients, despite about 20 minutes longer to reach them. Their team composition was different. In sub30 there were 2 advanced PHEM clinicians for conventional resuscitation, and 2 experienced ECMO clinicians for ECPR – I assume one cannulating and the second setting up the circuit and console. In CHEER3 the ECPR team was 3 clinicians: 2 cannulating physicians and an ECMO paramedic to prime the circuit and prepare the console. This is something to consider. The sub30 paper comments that the demands of pre hospital ECPR fully occupied all team members. Perhaps something to learn from these 2 papers is that the procedure is complex enough that it needs 3 people dedicated to the ECPR. Indeed the most well known example, the SAMU ECPR team, deploy with a team of 3: physician (cannulation), paramedic (assist with cannulation), nurse (circuit & console).

What happened to the patients?

2 things are really important to remember when answering this question. Firstly, neurologically intact survival from OHCA in the UK remains poor (the recent PARAMEDIC-3 trial was 2.7 and 2.8% respectively). The second is that this was not a primary outcome, and sub30 was not designed to detect a survival benefit. However, it was one of the secondary outcomes, and the publication generously includes a lot of patient specific information.

All 5 of the patients were supported on ECMO to hospital, and all received coronary angiography and revascularisation. Life sustaining treatment was withdrawn from 3 of them due to “severity of critical illness” which I interpret as unsupportable multiple organ failure, rather than hypoxic brain injury. The remaining 2 patients both survived to ICU discharge, and remained alive at 90 days, but both had severe neurological injury with cerebral performance category (CPC) scores of 3 (severe disability) and 4 (coma or vegetative state).

Although no neurologically intact survivors is sad, with only 5 patients, it’s not hugely surprising. The best published ECPR survival rates are around 45% CPC 1-2 so at best you could expect 2 from sub30s patients, and this result could feasibly be due to chance. The authors acknowledge this in the paper.

I think what is most interesting about these results is the lack of the usual dichotomisation. Usually in ECPR trials there are 2 clear groups: early deaths who are either brain dead or rapidly unsupportable and die of multiple organ failure, and the neurologically intact survivors. It’s rare to see ECPR patients surviving with severe neurological injury in the literature, and I wonder if this is explained by differences in intensive care and neuroprognostication, rather than ECPR itself.

Should We Change Practice Based on This Study?

Sub30 has shown that, within Greater London, it was possible to establish candidate OHCA patients on ECPR, within a timeframe which generates neurological survival in other studies. For the UK this is massive progress – there are many teams with the enthusiasm and the willingness to do this, and sub30 shows it is possible. Often the challenges lie in NHS commissioning and onward care of the patients on ECMO, so it is fantastic to see this overcome.

The authors have suggested some changes for the future, including improving early recognition of arrest and delivery of bystander CPR and community AED use. They acknowledged that a team of 4 is probably unsustainable, and I would add that perhaps the skill mix could be optimised to 3 clinicians all with ECMO skills. This could be both more sustainable and decrease the time from arrival to being established on ECPR. They also make an interesting comment about varying incidence and location of OHCA, and how studying these things alongside travel mode, route and time could further decrease the arrest to ECPR time.

Summary

It’s great to see that pre hospital ECPR is possible within the UK, and within internationally accepted timeframes. Indeed, the authors acknowledged that perhaps sub30 was ambitious, and sub60 would have been more reasonable! If that had been the case, they’d have achieved their primary outcome with 100% success.

We should learn a lot more about the OHCA ECPR field in coming months. SAMU, who run a 24/7 ECPR service in Paris, will be publishing a review of their first 10 years. Also ongoing is the PreCare feasibility trial (Sydney), and the On-Scene trial (Netherlands). On-Scene in particular is hugely impressive, as they have achieved delivery of ECPR to their entire country through their HEMS system, and have recruited nearly 200 patients, over 100 of whom have received ECPR.

If On-Scene shows benefit, then the question of ECPR for OHCA moves from should we to how can we? A question which I think sub30 goes some way to helping us answer. Most importantly perhaps, it gives the evidence that UK OHCA ECPR is possible, and fuel for the fire to keep pushing to improve OHCA survival.