Cardiac & Resuscitation, Exclusives, Top Story

Will Mobile ECPR Change How EMS Treats Cardiac Arrest?

“He not busy being born, is busy dying.”—Bob Dylan (1965)

No one is busier dying than a patient in refractory ventricular fibrillation (RVF).These patients have exasperated EMS personnel for decades, and they, sadly, usually die. The treatment traditionally has been to administer ACLS for 30–45 minutes, and the lucky few who get a return of spontaneous circulation (ROSC) get admitted.

The survival rate using this approach has been around 8%.1 But, with the advent of adult, extracorporeal membrane oxygenation (ECMO), the approach to these patients may be changing.


The use of ECMO in arrested patients has been referred to as extracorporeal cardiopulmonary resuscitation (ECPR). ECPR, coupled with the use of the LUCAS mechanical compression device, the ResQPOD impedance threshold device (ITD), and percutaneous cardiac intervention (PCI), has been used routinely for the past 2 1/2 years by Demetris Yannopoulos, MD, and his colleagues at the University of Minnesota.

EMS services in the Twin Cities, including the two largest providers, the St. Paul Fire Department and North Memorial Medical Transportation, which serve a combined population of 570,000, have been providing Yannopoulos with ECPR patients. The results of each resuscitation have been studied with results from the first 72 cases treated with this approach published last year in the Journal of the American College of Cardiology (JACC).1

Patient Selection

In order for a patient to be eligible for ECPR at the University of Minnesota, the initial recorded rhythm must be ventricular fibrillation (v fib) or ventricular tachycardia (v tach). In addition, the patient must be between the ages of 18 and 75, have no DNR orders or dementia, must be able to fit in the LUCAS, and must have an estimated transport time of less than 30 minutes.

Also, the estimated time from the 9-1-1 call to the coronary catheterization lab (CCL) must be under 90 minutes.

The patient isn’t eligible for ECPR if the family or caregiver declines, if there are any contraindications to mechanical CPR, if the patient has a known terminal illness, or is pregnant.

Table 1: Inclusion and exclusion criteria for University of Minnesota ECPR protocol

If these conditions are met, after three defibrillations and 300 mg of amiodarone, the patient is transported to the university’s CCL with ongoing automated CPR. The goal is to keep the scene time to a minimum (i.e., 10–12 minutes), much like a trauma scene, to get the patient to a center of resuscitation excellence as soon as possible.

In addition to the above inclusion criteria, upon arrival, the patient must have an EtCO2 greater than 10 mmHg, a PaO2 greater than 50 mmHg, an SaO2 greater than 85%, and a lactate of less than 18.

It should be noted that the arrest does not have to be witnessed in this protocol. The reason for this is that Yannopoulos and colleagues feel that v fib/v tach acts as a surrogate clock, i.e., v fib/v tach is usually short-lived, typically lasting 10 minutes or less.

So, if the patient is in a shockable rhythm when EMS arrives, the down time is probably under 10 minutes.

In addition, the patient doesn’t have to have signs of neurologic activity, as is required in a field ECPR trial currently underway in Paris.2

JEMS: How Physicians Perform Prehospital ECMO on the Streets of Paris

Besides Paris, there are a couple of other ECPR trials that should be mentioned. Although the CHEER Trial (which includes mechanical CPR, hypothermia, ECMO and early reperfusion) only had 11 patients with out-of-hospital cardiac arrest (OHCA) who received ECPR, five (45%) survived.3

The SAVE-J Trial (i.e., study of ACLS for v fib with extracorporeal circulation) was performed in Japan.4 This was a much larger study of v fib/v tach arrest. The authors reported on a total of 454 patients, of whom 260 received ECPR. Both CHEER and SAVE-J had much different designs than the Yannopoulos project, so comparing them is difficult. For example, in SAVE-J the patients were brought to 46 different hospitals and the trial employed a convenience randomization scheme (i.e., patients received ECPR based on whether they presented to an ECPR-capable center).

A thorough discussion of all of the trials in the literature is beyond the scope of this article. But suffice it to say, by and large the data and their results support the use of ECPR in OHCA.

Yannopoulos utilized a protocol change by the services which participated and used only one hospital and, of course, did not randomize. The patient was brought directly to the CCL and, if the patient met the inclusion/exclusion criteria, was placed on ECMO—using the Cardiohelp System by Maquet—while the LUCAS mechanical chest compression device was performing CPR (i.e., chest compressions weren’t interrupted). (See Figure 1.)

Figure 1: CardioHelp system manufactured by Maquet

The Cardiohelp System utilizes the arterovenous (AV) approach. The femoral artery and vein are punctured using a percutaneous, ultrasound-guided technique. Following that, wires are inserted through the needles and into the femoral artery and vein. Following placement of the wires, the vessels are dilated with graduated dilators.

Catheters are then placed over the wires in the artery and vein, and then checked with fluoroscopy. Generally, the vein gets a 27 French catheter and the artery, a 15 French catheter.

This needle/wire/catheter is known as the Seldinger technique, which is used very commonly in medicine for placing central lines and in other procedures.

On paper, the technique seems straightforward, but that’s deceptive. It requires significant skill and experience, especially during cardiac arrest because the chest compressions never stop.

Over the 2 1/2 years that Yannopoulos and his colleagues have been performing the technique, they’ve become extremely proficient at the procedure. In the JACC article, the average time from arrival at scene to ECMO initiation was 6.1 minutes, although it’s often accomplished in less than 5 minutes.1

Of course, the goal of ECPR is to not only have the patient survive, but to survive neurologically intact. In the JACC article, of the 62 patients who met the inclusion criteria, 47 (76%) were admitted. Twenty six of the 62 (42%) who received ECPR survived to discharge with a Cerebral Performance Category (CPC) score of 1 or 2.1

Table 2 is a schematic of the case series included in the survival results. All patients who survived to discharge were alive at three months and had a CPC of 1. Table 3 contains the definitions of the CPC categories.

Table 2: Schematic of the University of Minnesota case series including the survival results1 (All patients survived to discharge were alive at three months and had a CPC of 1.)
Table 3: Table 3: Definitions of cerebral performance categories (CPCs)

Good prognosis predictors included: an EtCO2 greater than 10, a lactate less than 18, ROSC during the resuscitation and, most importantly, a time from 9-1-1 call to ECMO of 50 minutes or less.

Utilizing the historical control survival of 8% with the previous standard of care mentioned above, there were more than 5 times more survivors utilizing the ECPR protocol for v fib/v tach.

Also, it’s clear from this work, as well as other (OHCA) studies, that the vast majority, greater than 95%, of patients who are discharged, are neurologically intact.5

Despite the high survival rate, Chart 1 shows that those patients who underwent ECPR were very ill.Most (87%) had significant coronary artery disease (CAD) (i.e., occlusion of 70% or more); 30% had single-vessel CAD ; 26% had two-vessel CAD; and 44% had three-vessel CAD. The most commonly diseased artery was the left anterior descending, followed by the left circumflex and the right coronary.

Chart 1: Patient characteristics of those who underwent ECPR

These findings support the opinion that RVF should be considered an ST-elevation myocardial infarction (STEMI) equivalent and needs to be treated as such (i.e., PCI as soon as possible).

In addition to having significant CAD, these patients frequently had “stunned heart syndrome,” with many (45%) requiring left ventricular support with an intra-aortic balloon pump. ECMO was required for an average of three days.

It’s therefore important that clinicians understand the extent of left ventricular compromise, or this approach won’t be successful. Post-procedural left ventricular support is critical to survival.

There were other interesting results in the series. Cardiac arrest was the first evidence of CAD in 91% of the patients, and no patients had any ischemic symptoms in the two weeks prior to their arrests. The average age of the survivors was 59.

Since publication of the article in JACC, the series has continued and now has a total of approximately 180 patients. The results are holding at a 40–45% neurologically intact discharge rate.

Emerging from the data are some important predictors of survival; EtCO2 > 10 mmHg, serum lactate < 18, and any episodes of ROSC during the resuscitation are all positive signs predictive of survival.

But the most important factor impacting functional survival is the care rendered in the field, starting with time from the 9-1-1 call is received to the initiation of ECPR. All patients are treated with high-quality CPR and, within a very short time period, the ITD and LUCAS mechanical compression device.

Chart 2 show that the sweet spot for ECPR appears to be in the 40–50 minute range. In the 50–60 minute time frame there’s a fair chance for survival, but with increasing deaths.More rapid recognition of cardiac arrest by dispatch, coupled with decreased scene time protocol (e.g., initiating transport after one shock, with all other cares occurring while en route), will help to improve survival to an even greater extent.

Chart 2: Effect of 9-1-1 call to ECPR initiation on arrival

ECMO Response Teams

Yannopoulos and his team at the University of Minnesota are focusing on research that will develop a self-sustaining, community resource model that reduces the time from 9-1-1 to ECPR. The approach will be multifaceted. It will include rapid ECMO response teams who will be intensely trained and have extensive experience in ECPR. These teams will be on call to respond emergently to any hospital that has a patient who may benefit from ECPR.

A component of the research will define what personnel will optimally make up these teams. To begin with, the physician team member will probably be a cardiac interventionalist, critical care specialist or an emergency medicine physician. This physician will be assisted by a perfusionist, and a critical care paramedic and nurse—or two critical care paramedics.

The team will respond to a hospital in a chase car(s) and ECMO will be initiated at the hospital. That hospital will have the option of performing a PCI, or transferring the patient to another PCI-capable center. If the hospital performs the PCI, it may elect to keep the patient or transfer them to an ECMO center. In any case, the ECMO team would help manage the patient for the first few hours after PCI.

Deploying a Mobile ECMO Unit

Another important and interesting aspect of this research will be the development of a mobile ECMO unit. Like the historic concept of paramedic response units that are strategically placed within response regions, this Mobile ECMO Unit will be community-based and will probably be located at a hospital.

The most important reason for this new Mobile ECMO Unit is time; because ECMO has been found to be most successful for patients if it’s started within 30 minutes of the cardiac arrest incident occurring.

One design currently being considered involves a semi-truck tractor attached to a large ambulance box (similar to an expandable RV) equipped with full lights and sirens for emergency response. The box will have hydraulically activated RV-type sliders that, when deployed, will transform the ambulance into a fully functional CCL.

This Mobile ECMO Unit will be specifically designed for ECMO establishment, with all the necessary equipment in pre-assigned and inventoried locations. It will have X-ray and fluoroscopy capability, multiple redundant power sources, fully redundant telemetry/communication support, and camera systems capable of advanced telemedicine. It will carry all necessary drugs, fluids and blood, as well as have refrigeration.

In addition to the establishment of ECMO, this unit will be capable of performing a PCI should a cardiac interventionist be the team physician member. The vehicle will operate very much like mobile MRI or CT vehicles (i.e., stroke ambulances) do today, with the patient being brought to the unit and the team working inside.

There are several advantages to this approach. Having done much of their training in the vehicle, the team will be very familiar with its surroundings, and know exactly where everything is in the dedicated space. This also has the advantage of decreasing disruption of hospital and/or ED operations that can occur when ECMO is being established.

Although such a vehicle seems futuristic, there are examples of pediatric ECMO units that are large ambulances based on semi-tractors, which are capable of transporting ECMO patients but don’t establish ECMO. Two such vehicles caring for pediatric patients are currently operating in Minnesota. (See Figure 2.)

Figure 2: Example of a mobile pediatric ECMO unit currently in operation in Minnesota

An additional, important feature of a Mobile ECMO Unit is that it would have the capability of intercepting an ambulance that;’s en route from a rural or outer suburban location. The patient would be transferred from the referral ambulance to the ECMO unit, placed on ECMO, and possibly undergo PCI in the field; the rationale being that although ECMO preserves the brain, the heart continues to sustain damage as long as blockage of the coronary artery(s) persists. (Figure 3 shows the designs of the new mobile ECMO unit.)

Figure 3: Mobile ECMO unit design by LifeLine Emergency Vehicles

Unanswered Questions

Many questions remain to be answered: How will this approach be self-sustaining? How would the ECMO unit be licensed? How will procedures be billed? Who does the billing? How do the providers get paid? Could this be done by helicopter? It may someday be staffed on by nurses or medics and directed by a base physician.

Of course, one of the big questions is: What does the cost-benefit curve look like? That question is beyond the scope of this research. First, it’s our goal to see if this new community response model works; then it will be up to policymakers to decide whether it’s worth it.

Considering the multiple other current, and perhaps future, indications for ECMO, this vehicle could become very busy, indeed. (See Table 4.) One could even conceive of the practice of ECMO placement and management someday becoming a medical specialty or sub-specialty.

Table 4: Other possible indications for ECMO


Preliminary findings appear to show that intact survival from refractory v fib can be drastically improved by taking a much more aggressive approach than the current standard of care. If these preliminary findings hold, there may soon be a major paradigm shift in the treatment of out-of-hospital cardiac arrest.

For more on Resuscitation Advances, watch for and download The State of the Future of Resuscitation, a supplement to JEMS that’s being released in mid-February 2019.


1. Yannopoulos D, Bartos JA, Raveendran G, et al. Coronary artery disease in patients with out-of-hospital refractory ventricular fibrillation cardiac arrest. J Am Coll Cardiol. 2017;70(9):1109–1117. Read full-text article online (Open Access).

2. Lamhaut L, Hutin A, Puymirat E, et al. A Pre-Hospital Extracorporeal Cardio Pulmonary Resuscitation (ECPR) strategy for treatment of refractory out hospital cardiac arrest: An observational study and propensity analysis. Resuscitation. 2017;117:109–117. Read full-text article online (Subscription).

3. Stub D, Bernard S, Pellegrino V, et al. Refractory cardiac arrest treated with mechanical CPR, hypothermia, ECMO and early reperfusion (the CHEER trial). Resuscitation. 2015;86:88–94. Read full-text article online (Subscription).

4. Sakamoto T, Morimura N, Nagao K, et al. Extracorporeal cardiopulmonary resuscitation versus conventional cardiopulmonary resuscitation in adults with out-of-hospital cardiac arrest: A prospective observational study. Resuscitation. 2014;85(6):762–768. Read full-text article online (Subscription).

5. Aufderheide TP, Frascone RJ, Wayne MA, et al. Standard cardiopulmonary resuscitation versus active compression-decompression cardiopulmonary resuscitation with augmentation of negative intrathoracic pressure for out-of-hospital cardiac arrest: A randomised trial. Lancet. 2011;377(9762):301–311. Read full-text article online (via HHS Public Access).