Why We Do What We Do Critical Care Edition: Is there an Echo? What is all This I Keep Hearing About ECMO?

By: Moses Washington, MD
central  ecmo cannula pic

Figure 1: Central ECMO cannulation, image courtesy of google images, N Engl J Med 2011; 365:1905-1914

Extracorporeal Membrane Oxygenation (ECMO) is a procedure that over the past several years has seen a tremendous resurgence in its use in adults. ECMO is effectively a type of mechanical cardiopulmonary bypass that temporarily (days to weeks) supports the cardiovascular and/or respiratory system in severe illnesses. It has often been used as a last ditch effort in treating refractory illnesses such as ARDS (most commonly), peri-transplant, cardiogenic shock, and post-cardiac arrest. The original technology has been in existence since early 1970s, but very small trials showed its poor outcome resulted in its abandonment in adult population. ECMO only recently saw a resurgence beyond the cardio-thoracic operating rooms in adults during and since the 2009 H1N1 influenza pandemic. This review will cover how it works, indications, contraindications, trials showing outcomes, common complications, and its relevance to Emergency Medicine Critical Care.

The two main methods to implement ECMO are veno-venous (VV) and veno-arterial (VA). The goals of ECMO are to either improve oxygenation, CO2 removal, or allow cardiac unloading or resting the lung.The ECMO circuit consists of cannulas, a pump (which coordinates flow), heat exchanger, and an oxygenator. ECMO cannulas can be inserted centrally (via great vessels) or peripherally (femoral vessels). In VV ECMO blood is drained from the venous system goes to the membrane oxygenator and pumped back to the venous system in series via a circulatory pump and heat exchanger (1). This is typically achieved by cannulating both femoral veins. VV ECMO is reserved for only respiratory disease. VA ECMO in contrary supports failing cardiac and respiratory function or just cardiac failure. In VA ECMO blood is drained from the venous system and pumped into the arterial system. It effectively improves tissue oxygenation, circulatory flow (most importantly cerebral perfusion) and coronary artery perfusion by allowing the heart to rest and unloading the RV (1). This can be performed centrally or peripherally. Central access requires thoracotomy and is only performed by cardio-thoracic surgeons since one cannula is inserted to the right atrium and another into the arch of the aorta. Peripheral VA ECMO requires one to cannulate the femoral vein and artery. The second approach is very much akin to inserting central lines but just requiring larger dilators and cannulas (3). A very simplistic way of understanding how to improve oxygenation and C02 removal in ECMO is mainly all about the blood flow. Essentially, the longer the blood is contact with the membrane oxygenator the better the oxygenation. Since C02 is more rapidly diffuse just increasing the blood flow can improve C02 removal. Since using the circulatory pump can cause clotting consequences low dose heparin and heparin coated cannulas are used (3).

peripheral ECMO pic


Figure 2: Peripheral VA ECMO cannulation, image courtesy of google images, scancrit.com/2012/07/27/emergency-ecmo/

The most common indications for ECMO in the acute setting are for refractory respiratory failure secondary to ARDS, cardiogenic shock and in peri-arrest. By far the most data showing benefit of ECMO is in ARDS. The idea behind ECMO improving mortality in ARDS effectively is allowing the lungs to rest. This typically has been considered after maximizing low tidal ventilation via ARDSnet protocol, proning, attempting high flow oscillatory ventilation and inhaled nitrous oxide. For VA ECMO indications are refractory cardiogenic shock heart failure, or cardiac arrest. In order to be considered for this costly and labor intensive procedure the disease must have a presumed reversible cause like cardiac ischemia, cardiotoxicity, or viral myocarditis. Contraindications typically include age greater than 65 or 75, metastatic cancer, multi-organ failure, severe irreversible brain damage, and contraindications to anticoagulation (3).

Following the 2009 H1N1 pandemic there was an increase use of ECMO. One study most often quoted was a retrospective case series which documented their experience with ECMO. The study was conducted across 15 ICU’s and ECMO centers in New Zealand and Australia. Out of the 252 patients referred to ICU’s for the influenza pandemic 68 received ECMO.  At the time the data was reported there were a few still remaining in the ICU, those treated with ECMO only had 21% mortality compared to typically 30% which has been seen in the literature (4). This led to the authors saying that the results of their study should help clinicians in healthcare planning during the pandemic. Another study and the largest study was the Efficacy And Economic Assessment of Conventional Ventilatory Support versus Extracorporeal Membrane Oxygenation for Severe Adult Respiratory Failure (CESAR) trial. The CESAR trial was a multi center study conducted in the UK of 180 patients randomized in a 1:1 ratio comparing patients who received conventional mechanical ventilation to those who would be considered to receive ECMO for respiratory failure. Interestingly not only did this study assess for survivability to discharge but also assessed neurologic disability after 6 months of discharge.  At first glance the results seemed very astounding in which 63% (ECMO group) compared to 47% of conventional management group did not  die or have severe disability after 6 months. The study comes under a lot of scrutiny since 68/90 patients randomized to receive ECMO and of those sixty-eight, forty-one survived without severe disability. The remainder were sent to a ECMO center but received conventional management. The study did not achieve its primary objective in effectively comparing ECMO vs conventional management in a 1:1 ratio. This has led to many criticizing that just being referred to an ECMO center in which the staff is better equipped to taking care of the sickest patients may be a confounder in this study. Nevertheless when analyzing the data for face value it can’t be ignored that those who received VV ECMO did have a better outcome (5). Overall, many experts feel that VV ECMO has promise with anecdotal evidence in treating respiratory failure but to date there are not concrete studies that has shown it should be the mainstay treatment in refractory severe ARDS.

VA ECMO and its implication in being a salvation mechanism in refractory cardiogenic failure and specifically cardiac arrest is where most ED physicians are intrigued in its utility. To date only early CPR and defibrillation has been shown to be beneficial in cardiac arrest. The use of VA ECMO in post-cardiac arrest has been referred to as extended cardiac life support (ECLS) or ECPR. Even the AHA is recommending considering ECLS in patients who are presumed to have a reversible cause of cardiac arrest. This is based on some key studies, one study in Taiwan by Chen et al, which looked back at 172 patients over a 3 year period who underwent witnessed in-hospital cardiac arrest secondary to a reversible cause. Out of the studied patients, 113 fit the conventional CPR group to 59 who received VA ECMO after receiving standard ACLS for 10 minutes. They did an intention to treat analysis in which the primary endpoint was survival to hospital discharge. Their results showed that 20% patients showed neurologically intact survival to hospital discharge in the ECMO group. The ECMO group also saw an earlier first rhythm of Ventricular tachycardia or fibrillation although not statistically significant (6). Those in the ECMO group also had ROSC faster than the conventional CPR group. Guen et al, evaluated 51 patients over 32 months, who received ECMO by a cardio-thoracic team who suffered cardiac arrest. All initial rhythms were included for analysis. This study only saw two survivors at 28 days, which reflects the known poor results from most out of hospital cardiac arrest studies. In contrast to the last study median time of onset to ECLS was 75 minutes. This led the authors to conclude using ECMO in the patient population suffering from OHCA should be restricted to selective patients (7). The experience of Dr. Shinar and Dr. Bellezzo, ED doctors, has shown that not only can VA ECMO be life saving but can also be done by trained ED doctors. They documented their experience over a 1 year period. They described a 3 stage approach to initiating VA ECMO in the ED. Stage 1 consists activating the ECLS nurse and ECLS circuit, then cannulating the femoral vessels while CPR is ongoing with angioplasty catheters. Stage 2 proceeds with replacing those catheters with larger cannulas with progressive dilators up to 23 French for the ECMO cannulas and priming and preparing the ECLS circuit. Finally stage 3 is hooking up the ECLS circuit to the cannulas and establishing flow. If ROSC was obtained before completion of stage 3 the protocol stops and standard ACLS care resumes. Eight of twelve patients completed all stages and five out of those eight survived with good neurological outcome to hospital discharge. Interestingly one of those survivors was a patient whose cause of cardiac arrest was an aortic dissection, which typically is a contraindication to initiating ECMO (8).  Another small study (26 patients) was a retrospective observational study at University of Pennsylvania by Johnson also analyzed ECLS for OHCA between July 2007-April 2014. The primary outcome was survival to discharge. This study was interesting in that Emergency medicine team (with senior resident initiating venous access) in concert with CT surgeons following a preset protocol initiated ecmo in post-cardiac arrest patients. The included patients were standard in comparison to other studies and resulted in 4/26 patients surviving to hospital discharge with 75% of those patients being neurologically intact. There was also a significant amount of complications (69%) bleeding and ischemic events (9). These studies show that there is promise that ECMO may be implemented more and taught in the future as truly advanced cardiac life support.

In 2014 there has been very intriguing studies that have developed that has shown good results involving ECMO for post-cardiac arrest. Like most disease entities often one therapy is not a cure all and post-cardiac arrest specifically likely needs a multi-procedural approach in order to achieve better mortality and neurological outcomes. The first study was by the SAVE-J group out of Japan and led by Dr. Sakamoto. His group in a prospective observational trial compared ECPR vs. conventional ACLS for OHCA over 4 years with the initial rhythm having to be VF/VT. Not only did they use ECPR but also therapeutic hypothermia, and intra-aortic ballon pump. The primary outcome was good neurological outcome at 1 month and 6 months following discharge.  Out of 454 patients, 159 (20 hospitals) patients in the non-ECPR group and 234 (26 hospitals in ECPR group) were analyzed in an intention to treat and protocol to treat fashion.  The ECPR  group had a favorable neurological outcome (CPC 1 or 2) in 12.3% (32/260) at 1 month and 11.2% at 6 months, while the non-ECPR group had 1.5% at 1 month and 2.6% at 6 months (10). Overall the results may look poor but one must take into account that in Japan the EMS paramedics and EMTs are not allowed to pronounce death in the field and are required to bring all cardiac arrests to the hospital in comparison to here in the states. The CHEER trial led by Stub in a pilot study analyzed 26 patients with all rhythms included who received mechanical CPR, pre-hospital initiated therapeutic hypothermia, ECPR and PCI in Melbourne Australia. This trial had a median rapid time (56 minutes) to initiate ECMO and had astounding results with 14/26 or 54% having a CPC score of 1 following hospital discharge.  Although small, this study has shown the most promising data including ECPR as treatment modality with favorable neurological outcomes in post-cardiac arrest care (11).

A procedure this invasive obviously has complications. The most common complication encountered is bleeding which includes not only bleeding due to the cannulation process but also intracranial and GI bleeding. Other common complications are mechanical failure of ECMO circuit and limb ischemia. Limb ischemia is often combated by inserting another cannula to facilitate arterial flow to the lower limbs.


Figure 3: Acute Limb Ischemia, courtesy of google images, www.slideshare.net/AbinoDavid/acute-limb-ischemia-14815482 

This technique may become a standard procedure which begins in the Emergency department and taught as advanced cardiac life support. It is feasible to initiate an ECMO protocol in a center which has coordinated effort between the members of the Emergency Department, Vascular or Cardio-thoracic surgeons, and Critical Care Units. With a select patient population with severe respiratory failure and refractory cardiac arrest more lives may be saved with ECMO with excellent neurological outcomes. It is of this author’s opinion like with many other interventions earlier implementation of ECMO in the peri-arrest period or respiratory failure may yield better outcomes on a larger scale.


Gattoni L, et al. Clinical Review: Extracorporeal membrane oxygenation. Critical Care 2011, 15: 243.

Chauhan S and Subin S. Extracorporeal membrane oxygenation, an anesthesiologist’s perspective: Physiology and principles Part 1. Annals of Cardiac Anesthesia 2011. Vol 14, No.3. pp 218-229

Marasco S, et al. Review of ECMO (Extra Corporeal Membrane Oxygenation) Support in Critically Ill Adult Patients. Heart, Lung and Circulation 2008.17S: S41-S47.

The Australia and New Zealand Extracorporeal Membrane Oxygenation (ANZ ECMO) Influenza Investigators. Extracorporeal Membrane Oxygenation for 2009 Influenza A (H1N1) Acute Respiratory Distress Syndrome. JAMA 2009, Vol 302, No 17. pp.1888-1895

Peek G, et al. Efficacy and economic assessment of conventional ventilatory support versus extracorporeal membrane oxygenation for severe adult respiratory failure (CESAR): a multi-centre randomized controlled trial. Lancet 2009, vol 374. pp.1351-1363

Chen YS, 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. Lancet 2008, Vol 372. pp 554-561

Guen M, et al. Extracorporeal life support following out-of-hospital refractory cardiac arrest. Critical Care 2011, 15:R 29.

Bellezzo J, et al. Emergency physician-initiated extracorporeal cardiopulmonary resuscitation. Resuscitation 2012, 83. pp 966-970.

Johnson NJ, et al. Extracorporeal life support as rescue strategy for out-of-hospital and emergency department cardiac arrest. Resuscitation 2014, 85. pp 1527-1532.

10. Sakamoto T, et al. Extracorporeal cardiopulmonary resuscitation versus conventional cardiopulmonary resuscitation in adults with out-of-hospital cardiac arrest: A prospective observational study. Resuscitation 2014, 85. pp 762-768.

11. Stub D, et al. Refractory cardiac arrest treated with mechanical CPR, hypothermia, ECMO and early reperfusion (the CHEER trial). Resuscitation 2015, 86. pp 88-94.

Leave a Reply