ECPR AND

Resuscitative
ECMO
A Detailed Look at
Emergent Extracorporeal
Life Support

EDITORS

ZACHARY SHINAR AND JENELLE BADULAK

Editors: Zachary Shinar and Jenelle Badulak
Manuscript Editors: Cindy Cook
ELSO Advisory Committee: Peter Rycus, Tom Brogan, and Christine Stead
Book Correspondence: ShinarELSOECPR@gmail.com

©2021 Extracorporeal Life Support Organization

CHAPTER AUTHORS

Matthias Arlt MD
Department of Anesthesiology,
University Hospital Regensburg,
Regensburg, Germany.
matthias.arlt@klinik.uni-regensburg.de

Jason A. Bartos MD PhD

Cardiovascular Division
University of Minnesota
Minneapolis, Minnesota
jabartos@umn.edu

Joseph Bellezzo MD
Department of Emergency Medicine
Sharp Memorial Hospital
San Diego, California
emergency.md@gmail.com

Jan Belohlavek MD PhD

Clinic of Cardiology and Angiology,
Charles University and General University Hospital,
Prague, Czech Republic
jan.belohlavek@vfn.cz

Lars Mikael Broman MD PhD

Associate Professor of Anesthesia and Intensive Care
Karolinska University Hospital
Stockholm, Sweden
lars.broman@karolinska.se

 iii
Elliott Cohen MD
Critical Care Medicine/ECMO Medical Director
Centennial Medical Center
Nashville, Tennessee
ecohenus@gmail.com

Julia Coull MD
Department of Intensive Care and Hyperbaric Medicine
The Alfred Hospital
Melbourne, Australia
juliacoull15@gmail.com

Arne Diehl MD
Department of Intensive Care and Hyperbaric Medicine
The Alfred Hospital
Melbourne, Australia
a.diehl@alfred.org.au

Björn Frenckner MD PhD

Professor of Pediatric Surgery/ECMO Center Karolinska
Karolinska University Hospital
Stockholm, Sweden
bjorn.frenckner@sll.se

Laura W. Geer RN BSN CCRN

MCS/LVAD Coordinator
University of Utah Health
Salt Lake City, Utah
laura.geer@hsc.utah.edu

Brian Grunau MD MHSc

Department of Emergency Medicine/Center for Health Evaluation and
Outcome Sciences,
University of British Columbia;
Vancouver, BC Canada
brian.grunau2@vch.ca

iv ECPR and Resuscitative ECMO

Amy E. Hackmann MD FACS
Department of Cardiovascular and Thoracic Surgery
Parkland Hospital/ University of Texas Southwestern Medical Center
Dallas, Texas
amy.hackmann@utsouthwestern.edu

Alice Hutin MD PhD

SAMU de Paris
Hôpital Universitaire Necker Enfants Malades
Paris, France
alhutin@aol.com

Lionel Lamhaut MD PhD

SAMU de Paris
CHU Necker
Paris, France
lionel@lamhaut.fr

James H Lantry III MD Maj, MC, USAFR

Medical Critical Care Services Physician/ Adjunct Assistant Professor of
Emergency and Critical Care Medicine
INOVA Fairfax Hospital/University of Maryland Medical School
Baltimore, Maryland
jlantrymd@gmail.com

Magnus Larsson MD PhD

Consultant Paediatric Surgery
Karolinska University Hospital
Stockholm, Sweden
magnusnennelarsson@gmail.com

Jon Marinaro MD FCCM

UNM Center for Adult Critical Care/UNM ECMO Program
Department of Emergency Medicine/Department of Surgery
Albuquerque, New Mexico
jmarinaro@salud.unm.edu

Chapter Authors v
Akshay Mungur MD
SAMU de Paris
Hôpital Universitaire Necker Enfants Malades
Paris, France
akshay.mungur@aphp.fr

Phillip Mason MD
Medical Director, Adult ECMO Program
Brooke Army Medical Center
Fort Sam Houston, TX
phillipmason@yahoo.com

Joshua C. Reynolds MD MS
Department of Emergency Medicine
Michigan State University
Grand Rapids, Michigan
reyno406@msu.edu

Alexander (Sacha) Richardson MD FCICM

Intensive Care
The Alfred Hospital
Melbourne, Australia
s.richardson@doctors.org.uk

Rory Spiegel MD
Departments of Emergency Medicine and Critical Care
Medstar Washington Hospital Center
Washington DC
rspiegs@gmail.com

Isaac Tawil MD FCCM

Critical Care and Emergency Medicine
University of New Mexico School of Medicine
Albuquerque, New Mexico
itawil@salud.unm.edu

vi ECPR and Resuscitative ECMO

Joseph E. Tonna MD MS FCCM FACEP FAAEM
Division of Emergency Medicine/Division of Cardiothoracic Surgery
University of Utah Health
Salt Lake City, Utah
joseph.tonna@hsc.utah.edu

Scott D. Weingart MD FCCM

Division of Emergency Critical Care
Stony Brook Medicine
Stony Brook, New York
emcrit@gmail.com

Sage Whitmore MD
Critical Care Medicine
Centennial Medical Center
Nashville, Tennessee
whitmoresp@gmail.com

Chapter Authors vii

CONTENTS

Preface xv
Foreword xviii

1. Starting an ECMO Program1

• Foundational Steps 3
• Four Organizational Steps to Create a R-ECMO Program 6
› Step 1: Organizing Stakeholder Teams 7
› Step 2: Design Key Systems and Protocols 11
› Step 3: Team Training 16
› Step 4: Putting it All Together 16

2. Optimizing the Pre-ECMO Resuscitation23

• Time-Dependent Considerations 25
• Chest Compressions 27
• Airway Management and Gas Exchange 31
• Pharmacotherapy 34
• Defibrillation 37
• Evaluating for Reversible Etiologies of Cardiac Arrest 39
• Conclusions 40

3. ECPR Candidacy Assessment and Outcomes47

• Clinical Trial Data for ECPR Compared to Traditional CPR 48

• Age Limits 49
• Bystander CPR 50
• Witnessed Arrest and No Flow Duration 51
• Low-Flow Duration 53

 ix
• Initial Rhythm 55
• Location of Arrest 56
• Presumed Etiology 57
• Past Medical History 59
• pH and Lactate 60
• Intermittent ROSC and Persistent VF 60
• Signs of Life During CPR 61
• ETCO2 and PO2 62
• Institutional Volumes 63
• Summary 63

4. Running the ECPR Code73

• Personnel 75
• Resuscitation Room Setup 77
• Code Choreography (Logistics of ECPR) 78
• Three Stages of ECPR 80
• Summary 83

5. ECMO Cannulation for ECPR89

• Cannulation Strategies 92
• Cannula Selection 95
• Guidewires 98
• ECMO Cannulation Procedure 101
• Distal Perfusion Cannulas 105
• Special Circumstances and Procedural Pearls 107

6. Hybrid Cutdown Technique for

ECPR Implementation115

• ECPR Team 116

• Procedure Description 116
• Results 120
• Benefits and Complications of Cutdown Technique 121
• Summary 122

x
7. Physiology of VA ECMO for Cardiac Arrest and
Cardiac Support125

• Physiologic Principles 125

• VA ECMO 133
› ECMO Pump 134
› Membrane Lung 136
› Clinical Example 136
› Dual Circulations 138
› Coagulation 141

8. ECMO Initiation147

• Circuit Preparation 147

• Wet-to-Wet Connection 147
• Going on Pump 151
• Troubleshooting 152
• Securement 156

9. ECPR Management of the First Four Hours159

• Key Management Priorities 160

• Establishing ECMO Support & Achieving Hemodynamic Stability
› General Vitals Signs Monitoring 161
› Arterial Line Placement & Blood Pressure Monitoring 161
› Central Venous Access & Central Venous Pressure (CVP)
Monitoring 161
› ECMO Circuit Monitoring: Pressures, Delta P, Central Venous O2
Saturation, Temperature 162
› Optimize Hemodynamics 163
› Optimize Gas Exchange 164
› Initial Diagnostics 165
• Treat Underlying Pathology 168
• Excluding Traumatic CPR Complications 169
• Preventing & Recognising Early ECMO Complications 171
• Caring for the Post-ECPR Patient 175

xi
• Determining Patient Disposition 176
• Debriefing with the ECMO Team 178

10. Circuit Catastrophes183

• The Complete Circuit Check 184

• Air Entrainment in the ECMO Circuit 187
• Accidental Decannulation 188
• Circuit Rupture 189
• Pump Failure 189
• Catastrophic Circuit Clot 190
• Basic Emergency Management Protocol: Emergent vs. Controlled
Separation from ECMO 190

11. Transport of ECMO Patients193

• Transport Team 194

• Redundant Equipment for Transport 199
• Transport after ECPR 199
• Complications and Risks 202
• Inter-hospital Transports 203

12. Prehospital ECPR for Out of Hospital

Cardiac Arrest207

• Rationale Behind Prehospital ECPR 207

• Capabilities 209
• Prehospital ECPR for Equality of Care 209
• Coordination with Prehospital Emergency Medical Services 209
• Prehospital ECPR Team 210
• Logistics of Program Set-up 211
• Patient Selection 213
• Transportation Considerations 214
• Worldwide Prehospital ECPR 215

xii
13. ECMO and Shock States219

• Cardiogenic Shock 221

• Rhythm Instability 223
• Massive Pulmonary Embolism 223
• Initial Medical Management of the Patient in Shock 223
• Other Mechanical Circulatory Support Devices 226
• Differences for V-A ECMO Compared to ECPR 227
• Post Initiation Management 229
• Outcome Data 230

14. ECMO in Trauma235

• Refractory Cardiopulmonary and Cardiocirculatory Deterioration

after Severe Trauma 236
› The Role of Extracorporeal Membrane Oxygenation (ECMO) in
Trauma 236
› Indications for ECMO Support in Severe Trauma 236
› Limitations for ECMO in Severe Trauma 237
› ECMO Cannulation and Circuits in Trauma 237
› Effectiveness of ECMO Support in Trauma 238
• Patient Management on ECMO in Trauma 238
› Hemorrhagic Shock and ECMO 238
› Intracranial Hemorrhage on ECMO 239
› Airway Injuries 239
› Burn Injuries 240
› Thoracic Injuries 240
› Advanced Liver Injuries 241
• Pediatric Trauma and ECMO 241
• Evacuation-Medevac and ECMO 241
• Emergency Preservation and Resuscitation 242

15. Crash V-V ECMO247

• Indications for Crash V-V ECMO 248

› Rapidly Progressing Acute Respiratory Distress Syndrome 248
› Refractory Asthma 248

xiii
 › Airway Obstruction 249
› Anaphylaxis 250
• Cannulation and Configuration 250
› Single vs Dual Site 250
› Cannula Selection 252
› Procedural Modifications 252
• Physiology 253
› Extracorporeal Gas Exchange During V-V ECMO 253
› V-V ECMO Impact on Hemodynamics 253
› Recirculation 254
› Arterial Blood Oxygen Content 254
• Initial Management 255
› ECMO Pump Flow 255
› ECMO Sweep Gas Flow 256
› Ventilator Settings 256
› Anticoagulation 257

Index 263

xiv
PREFACE

Over the last decade, ECMO has revolutionized the world of resusci-
tation. From ARDS to cardiac arrest, ECMO is treating an incredibly wide
spectrum of critically ill patients. It is now being utilized on six continents
with cases exponentially increasing. The power of ECMO is being realized
more each year.
Through the ingenious and collaborative efforts of ECMO teams
world-wide, we have refined resuscitative ECMO into an algorithmic pro-
cess, capturing novel solutions and techniques that continue to push the
boundaries of patient care. This book represents the collective efforts of
these pioneers. We focus on several aspects of resuscitative ECMO includ-
ing ECPR, crash V-V ECMO, V-A ECMO for cardiogenic shock, and V-A
ECMO for trauma. We additionally address the various locations where
these efforts may occur, including the emergency department, intensive
care unit, catheterization lab, and the prehospital setting.
We are proud and privileged to edit these writings and create a book
that can be used as a bedside reference during a resuscitation as well as
for academic study.

Zack and Jenelle

 xv
FOREWORD

There are few times in the history of cardiovascular medicine that

an idea opens the door to substantial innovation, leading to better out-
comes and a second chance in life.
In 1953, Dr Gibbons at Jefferson University Hospital performed open
heart surgery and in the same year, Dr. C. Walton Lillehei, at the Univer-
sity of Minnesota, started operating in young children with congenital heart
defects. His method connected the child with their parents’ circulation
thereby supporting them during open heart surgery. Twenty-eight out of 45
children survived. The immediate need to develop circulatory support, with
the ability to artificially oxygenate the circulating blood, was met with inno-
vation by DeWall, who produced the first bubble oxygenator. Subsequently,
Kay and Cross introduced the first mass production oxygenator. This series
of monumental scientific steps allowed the field of modern cardiac surgery
to flourish and save millions of lives worldwide.
A few years later the first layered Teflon based extracorporeal mem-
brane oxygenator (ECMO) was developed, but it was not until 1971
when the first adult trauma patient survived using ECMO. Despite the
initial negative adult clinical trial, ECMO support for pediatric patients
increased and evidence suggested that its application saved lives. In
1989, Dr. Bartlett founded the Extracorporeal Life Support Organization
(ELSO). This organization helped disseminate ECMO knowledge, its
applications, indications and logistics. It also facilitated broad access of
highly specialized knowledge to the medical establishment in the U.S. and
the rest of the world.
Over the last 10 years, a significant resurgence in the application of
this novel technology in the battle against cardiac arrest and resuscita-
tion has been observed in multiple countries. The application of ECMO
during resuscitation is called extracorporeal cardiopulmonary resuscita-
tion or ECPR. ECPR has shown, in multiple cohorts of patients and one
NIH funded randomized trial (THE ARREST TRIAL), increased survival

 xvii
to hospital discharge for patients with almost no chance of surviving with
standard ACLS. Despite these promising and exciting data, broad dissem-
ination of ECPR programs remains extremely challenging.
Cardiac arrest is the most time sensitive medical emergency in mod-
ern medicine, similar to exsanguinating hemorrhagic shock. Therefore,
ECPR initiation success and early application determine the size of its
effect. In addition, the number of patients and the location of the arrests,
in combination with the heterogeneity of the EMS response protocols and
times, offer additional organizational challenges for aspiring programs.
In this book, world experts in the field of ECPR have partnered to-
gether with ELSO to provide the first comprehensive primer of resusci-
tative ECMO. The expertise of the contributors offers a safe blueprint
and guide for the whole spectrum of program setup. This ranges from
physicians’ training to safe cannulation techniques. Patient selection,
decision trees, pit crew configurations, detailed percutaneous and surgi-
cal cut down technique descriptions, physiology details, and post ECMO
management algorithms are all included in the text. In addition, the chap-
ters include prehospital efforts of how to engage EMS, and potential addi-
tional indications for ECMO in trauma and cardiogenic shock.
We are facing a second historical moment in cardiovascular medicine
when, this time, a developed technology is taking on a grave disease with
the potential to save countless lives.
This book offers detailed knowledge and concentrated expertise for
anyone interested in starting an ECPR program in their community.
This high-quality effort will be referenced for years to come.

Demetris Yannopoulos MD
Professor of Medicine and Emergency Medicine
Medical Director, Center for Resuscitation Medicine,
MN Mobile Resuscitation Consortium
University of Minnesota Medical School

xviii ECPR and Resuscitative ECMO

CHAPTER 1
Implementation of a
Resuscitative ECMO Program

Jon Marinaro and Isaac Tawil

Introduction
Resuscitative extracorporeal membrane oxygenation (R-ECMO) is
the rapid implementation of mechanical cardiopulmonary support in
acutely deteriorating patients. This may include emergent deployment of
venovenous (V-V) or, more commonly, venoarterial (V-A) Extracorporeal
Membrane Oxygenation (ECMO). A classic example of R-ECMO is extra-
corporeal cardiopulmonary resuscitation (ECPR), and the terms are often
used interchangeably. ECPR represents the most emergent and often the
most challenging application of ECMO. The emergent nature adds com-
plexity to patient selection, team deployment, multidisciplinary collabo-

Foundational Term
ɋ Resuscitative ECMO – Resuscitative extracorporeal membrane
oxygenation (R-ECMO) is the rapid implementation of mechan-
ical cardiopulmonary support in acutely deteriorating patients.
This may include emergent deployment of venovenous or
venoarterial ECMO. A classic example of resuscitative ECMO is
extracorporeal cardiopulmonary resuscitation (ECPR), which is
ECMO initiation during active chest compressions.

 1
ration, and technical aspects of cannulation. R-ECMO demands the most
of your ECMO systems of organization and requires multidisciplinary
and multi-professional collaboration. Well-organized systems are used to
implement mechanical support in the setting of precipitous cardiogenic
shock and acute pulmonary failure. For example, you might utilize your
R-ECMO system to respond to the cardiac catheterization lab (CCL) for
cardiogenic shock, the Emergency Department for massive pulmonary
embolism (PE) management, or intensive care unit (ICU) for V-V ECMO
in refractory asthma as well as ECPR wherever it occurs.
Out of hospital cardiac arrest (OHCA) affects nearly 350,000 people
a year in the United States, yet only 8.4% survive with a good neurologic
outcome.1,2 Implementation of a R-ECMO program has the potential to
greatly improve this dismal survival rate while still maintaining a good
neurologic outcome.3 Over the last decade, ECPR cases have risen sub-
stantially; in 2019 there were more than 8,000 cases with intact neuro-
logic survival of 29%.4 The American Heart Association 2019 Update
states “ECPR may be considered for selected patients as rescue therapy
when conventional CPR efforts are failing in settings in which it can be
expeditiously implemented and supported by skilled providers” (Class
2b; Level of Evidence C-LD).5 A R-ECMO initiative at your institution is
an exciting yet daunting step forward for any hospital. This process can
challenge every system in your organization and potentially expose some
basic deficiencies not previously appreciated.
Every institution has unique limitations that need to be individually
addressed to create an R-ECMO program. Small private hospitals with no
ECMO programs as well as large academic facilities with a wealth of phy-
sicians wishing to own a piece of the project each presents challenges that
can, with time, be assets used to benefit your R-ECMO program. Some
centers have mechanical cardiac support programs with cardiac surgeons
who will want to be involved in ECMO and other programs may be driven
by interested emergency medicine and critical care physicians. Both mod-
els are sufficient to support a R-ECMO program. Importantly, without a
durable inhouse mechanical cardiac support program, an ECMO trans-
port system will need to be developed to potentially extricate patients to
a higher level of care.
As you consider starting a R-ECMO program there are two foun-
dational steps that must occur before moving into the logistical devel-
opment of a program. First and foremost, the plan needs to gain the

2 ECPR and Resuscitative ECMO

approval of hospital leadership both for costs and potential benefits. If
institutional leadership is unsupportive, you cannot successfully mobi-
lize this effort. Next, you must assure that the institution has a multidis-
ciplinary and multi-professional team interested in making this project
come to fruition. This cannot be a solo ED initiative or ICU initiative, as
it mandates support from many disciplines. If broad support cannot be
garnered, strong consideration should be made to stop at this step. Once
these preliminary steps are successfully navigated, then proceeding to
team organization, developing ECMO systems and protocols, and finally
creating the initial and ongoing team training process is mandatory before
the program can activate. Keeping a written plan to demonstrate that you
are progressing in a logical manner is critical to move the process forward
and to show all stakeholders of your progress.

Foundational Steps
Foundational Step 1 - Cost/Benefit Assessment
Given the significant efforts needed for a successful R-ECMO endeavor, it
behooves one to first identify the needs and costs at your institution and
within your community.
As a resuscitation leader at your institution, you must articulate
the benefits of creating a R-ECMO program to hospital leadership and
other involved services. Be informed of your institution’s cardiac arrest
data and explore other potential applications of R-ECMO. For instance,
R-ECMO may be deployed for massive pulmonary embolus patients and
cardiogenic shock patients. Explore options to utilize V-V ECMO as part
of your R-ECMO program to assist in emergent airway management,
acute trauma patients with severe pulmonary compromise, or refractory
asthma patients.
The potential utility of an ECMO program should be weighed against
the costs and resources required. For institutions with established ECMO
programs, simple repurposing of equipment and personnel makes start-
ing a R-ECMO program much easier. Institutions without a program will
find startup to be commensurately higher. Consider your current or po-
tential physician, RN, and ECMO specialist staffing and whether starting
a program will require additional positions. Those costs may significantly
drive your budget for years to come and not as a one-time expense.

Chapter 1  3
 As a R-ECMO leader you should have a strong understanding of the
ECMO diagnosis related group (DRG), institutional payer mix, proper
coding and billing, costs of equipment, supplies, physicians, and ECMO
specialists. The substantial upfront and ongoing costs of ECMO may dis-
suade the hospital financial officers from supporting the R-ECMO program
if a cogent argument cannot be made.
Proper documentation of the care given is one key component to
appropriate reimbursement and may offset some of the costs associated
with running the program. For example, the daily critical care manage-
ment and ECMO management are two separate notes that can be written
each day. Assure that ECMO cannulation notes, cannula manipulation
notes after the initial 24 hours, and peripheral decannulation documen-
tation are done routinely, as those are all billable procedures (Figure 3).6
Understanding the tangibles of billing and expenses for starting a
R-ECMO program are important but it is also necessary to understand
the concept of quality-​adjusted life year (QALY) and the relationship
between actual dollars and QALY. In an excellent review, Bharmal et al.
discuss the cost effectiveness of ECPR. Utilizing the more recently ac-
cepted $150,000-per-QALY as the threshold for cost effectiveness, the au-
thors determined that the use of ECPR costs $56,156 per QALY saved.7
This dollar per QALY is similar to dialysis ($72,476/QALY), kidney trans-
plantation ($39,939-$80,486/QALY), or orthotopic heart transplantation
((<$100,000/QALY). The authors further note that these QALY data are
not only financially relevant but may increase acceptance of ECPR as a
logical, lifesaving modality to offer.
Bharmal and colleagues also reference the importance of consider-
ing all predictors of improved survival with patient selection to reach
this cost/benefit result. Programs that stretch their indications and go for
the difficult, yet unlikely save, are less likely to attain adequate dollar per
QALY results.
Organizing a R-ECMO program can provide many potential down-
stream benefits that are rarely captured in standard cost/benefit analyses
but should be discussed with hospital leadership. At many institutions, the
addition of ECPR can provide a springboard to improve roles-based ACLS
delivery and meticulous emergency vascular access throughout the insti-
tution.8 It may also serve to advance critical care transport systems with
the increased skills to retrieve potential ECMO patients, perform remote
ECMO cannulations, and deliver ECMO patients to regional transplant

4 ECPR and Resuscitative ECMO

and durable ventricular assist device (VAD) centers. The scope of nursing
expertise expands through greater utilization of mechanical cardiopulmo-
nary support and advanced therapeutic modalities. Furthermore, this cut-
ting edge resuscitation creates institutional pride throughout emergency
and critical care services and beyond.
Once you have established the need for R-ECMO at your institution
with consensus among your hospital leadership and financial stakehold-
ers, you must assure that a team of practitioners is committed to this en-
deavor.

Foundational Step 2 – Assure Interest in a Multi-Professional

and Multidisciplinary R-ECMO Program
Collaboration among medical specialties and across the allied health
professions is fundamental to designing and implementing a suc-
cessful R-ECMO system and deserves special emphasis. A focus on a
multi-professional and multidisciplinary approach will facilitate the
unique challenges each center will encounter. To demonstrate the en-
tirety of the medical professionals that need to be on the same team, let
us walk through a typical OHCA ECPR case.
A sudden cardiac arrest is recognized, EMS is activated and minimiza-
tion of no-flow time with prompt bystander CPR is required. A 911 oper-
ator trained to screen for potential ECPR candidacy identifies the patient
and relays this to the responding paramedics. EMS providers provide op-
timal, rapid, initial ACLS and change from a “stay and play” mentality to a
brief ACLS algorithm utilizing laryngeal mask airway, mechanical CPR de-
vice, and assurance that there are no ECPR exclusion criteria. This is fol-
lowed by a “scoop and run” transport plan to get the patient to an R-ECMO
program in time to have the shortest low-flow period. The R-ECMO team
along with the ED is activated via a standardized R-ECMO alert. This is
followed by a well-choreographed “code” in the ED that utilizes roles
based ACLS, a secondary ECPR criteria screen, and preparation for ECMO
cannulation. The R-ECMO team performs intra-arrest cannulation un-
der transesophageal echocardiography guidance provided by a properly
trained EM or R-ECMO team member. Immediate post-cannulation chal-
lenges involve rapid CT scan and interpretation with direct radiologist in-
volvement. Next, the patient is transported to the CCL and interventional
cardiology for stent placement with potential left heart decompression.

Chapter 1  5
Lower extremity reperfusion cannulae placement should be considered
by the interventional cardiologists or vascular surgeons. Daily intensive
care challenges are managed with nursing, respiratory therapists, radiol-
ogy, pharmacy, clinical dieticians, physical therapists, and ancillary staff.
Potential complication management includes operative care with general
surgery, infection management with infectious disease specialists, and
blood product usage from the transfusion medicine team. Finally, cardi-
ologists help support the patient through cardiac recovery, critical care
physicians help wean the patient of mechanical support with the aid of
cardiology, and finally, operative decannulation with vascular surgery. All
of the above groups must participate in the creation of the R-ECMO pro-
gram. Despite the different institutional cultures, R-ECMO requires a bold
and massive multidisciplinary team endeavor for success and optimal out-
comes. If these two foundational steps can be secured it is time to proceed
to organization of your R-ECMO program.

Four Organizational Steps to Create a

R-ECMO Program
Below is a sample guide to setting up a R-ECMO program, used at our
institution (Figure 1).9 The steps are divided into broad categories of:

1. Organizing Stakeholder Team

2. Design Key Systems and Protocols
3. Training
4. Implementation and Evaluation

In practice, many of the items in each category occur concurrently, but

addressing these chronologically proved beneficial at our institution. For
example, embarking on systems development without input and buy-in
from key stakeholders will cause problems and similarly some degree of
system development must come before team training and rehearsal.
For a successful healthcare endeavor to have longevity, it must be
flexible and constantly evolving to improve upon the various components.
Thus, a successful program should expect to revisit this stepwise progres-
sion of systems planning as future challenges or growth opportunities
arise.

6 ECPR and Resuscitative ECMO

 Design Key
Organizing Stakeholder Implementation
Systems & Training
Team and Evaluation
Protocols

Identify stakeholders ACLS delivery system Small group stakeholder Ongoing simulation
lectures training
Generate stakeholder ARDS care system Joint stakeholder lecture Continue data
input and simulation event collections

Delineate goals R-ECMO team call and Monitor system fidelity

response system
Protocol development Cannulation strategy Monitor stakeholder
experiences

Evaluate potential case Patient transport Make system

load & resource needs, adjustments
costs & revenues

6 – 10 months 3, 6, 12 months
10 – 12 months before 1 month before
before after
implementation implementation
implementation implementation

Figure 1. Guide to R-ECMO creation.

Step 1: Organizing Stakeholder Teams

Formation of an initial resuscitation committee with all relevant stake-
holders is important and should occur long before a go-live date. Phy-
sician stakeholders include emergency medicine, intensive care, surgical
specialties (cardiothoracic, vascular, general), cardiology, anesthesiology,
radiology, and palliative care. Close multi-professional collaboration be-
tween physicians and nursing, ECMO specialists, pharmacists, and bed-
side therapists is also a prerequisite for a successful R-ECMO team. Early
on, the team must clearly establish the R-ECMO mission and timeline.
The opportunities for this mission to solve previous multidisciplinary
challenges should not be overlooked. At the University of New Mexico,
establishing a R-ECMO system sparked and facilitated many advances
including our resuscitative transesophageal echocardiography (TEE)
program, improved roles based ACLS delivery, and a system of greater
collaboration between ICU and ED nursing.

Chapter 1  7
 As you embark on early meetings with stakeholder groups, try to an-
ticipate what their particular goals, challenges, and concerns might be.
Take time to meet with local EMS leaders. The R-ECMO center should
understand and anticipate the potential political challenges with trans-
porting and potentially diverting appropriate candidates to the R-ECMO
center. An early step of our program was to invite EMS leaders to educa-
tional offerings on the in-hospital R-ECMO program we were performing
before we planned ECPR for OHCA. This got EMS leadership enthusias-
tically engaged in the process and prompted them to work with 911 dis-
patch and purchase automated CPR devices so they could perform CPR
during transport of potential ECMO patients.
Anticipating the concerns from cardiology of CCL cases with high risk
of severe anoxic injury, we met to discuss how we could improve our re-
sponse for cardiogenic shock cases in the CCL. After successfully imple-
menting R-ECMO to their patients, they were more excited to do these
cases from the ED. Similarly, any potential turf battles over cannulation
were solved with early and open discussions, a willingness to train to-
gether, and the intensivists’ willingness to bear the brunt of the cannula-
tion call.

Cannulation Team
An early decision for your R-ECMO program is the size and composition
of your cannulation team. R-ECMO cannulation is fraught with poten-
tial complications that will be minimized with repetition and volume of
cannulation experience. Teams that include dozens of cannulators will
ensure no single physician has significant exposure to the process and ex-
pertise in troubleshooting problems. Conversely, a team needs to be large
enough to ensure adequate service coverage and prevent over extending
a smaller group. Safe cannulation has been demonstrated among various
specialists10-12; therefore, the specialties of the cannulation team is of less
importance than the training and quality assurance system designed to
optimize outcomes.
Training of cannulators may occur either at national courses or at
the home institution by mentoring from previously trained physicians i.e.
cardiac/vascular surgeons. Establishing and nurturing relationships with
institutional specialists ensures backup during the early stages of the pro-
gram. Creation of a standardized cannulation privilege set within the hos-
pital guarantees that those who wish to cannulate have met the training

8 ECPR and Resuscitative ECMO

requirements and have appropriate knowledge of the processes and pit-
falls. At our institution, we organized a training program for percutane-
ous cannulations led by our vascular surgeon and ultimately credentialed
a multidisciplinary group of cannulators. This was a stepwise progression
that began with establishing a level of expertise in basic vascular ultra-
sound (consistently identifying the common femora artery and vein).
Cannulators were then taught wire manipulation techniques in elective
vascular surgery cases, followed by supervised non-emergency cannula-
tions and ultimately R-ECMO cases. This formalized group training in-
stilled confidence among surgeon cannulators in their non-surgeon team
members. Our critical care cannulators provide all V-A and V-V cannula-
tions in the institution, facilitating a sufficient caseload for skill mainte-
nance. Frequently, trainees wish to participate in ECMO cannulations; at
our institution we allow fellows and residents to assist in non-ECPR can-
nulations such as V-V ECMO patients, but any unstable patient remains
the purview of faculty only.

Management Team
Frequently, large numbers of physicians wish to provide cannulation ser-
vices and initial resuscitative management; however, it is important to
cultivate a strong ECMO management team for the post-cannulation pe-
riod. These physicians may or may not be cannulation team members.
Cannulation takes minutes while ECMO management lasts days to weeks.
We strongly recommend that physicians tasked with care of the ECMO
patient, at a minimum, attend an ECMO management course as offered
by the Extracorporeal Life Support Organization (ELSO) and many other
organizations. Continuing education in this rapidly evolving field is key to
optimize ECMO care and establishing a routine institutional case review
conference helps achieve this.
All ECMO programs require a qualified ECMO specialist support
team. Several models exist for obtaining this expertise at the bedside: 1)
utilizing outsourced ECMO specialists from a private perfusionist com-
pany, 2) expanding internally trained specialists from the cardiac surgery
perfusionist service to provide bedside ECMO care, or 3) expanding the
scope and training of bedside ICU nurses, ED nurses, and/or respiratory
therapists to include ECMO specialization. Each model has benefits and
drawbacks often impacted by local pressures. Utilizing internal ECMO
specialists from the pool of intensive care nurses and respiratory ther-

Chapter 1  9
apy staffing, can lead to staff shortages in the ICU. Conversely, a private
company specialist team may be limited contractually. A sudden influx
of ECMO patients, as was seen with COVID-19 pandemic, may exhaust
their ability to provide coverage and overextend the institutional agree-
ment. They may limit the extent of coverage they provide without further
contract negotiations. In addition, private teams may be less nimble to
add services such as offsite retrieval ECMO or participation in regional
educational offerings without further cost. The model ultimately chosen
should result in a close partnership in the ECMO process and day-to-day
management. A common thread among the various groups is the leader-
ship of a dedicated ECMO coordinator who oversees the ECMO specialist
group, functions as a bedside clinical resource, educator, and QA leader.

Patient Selection
Defining exactly which patients to treat is a fundamental task of the ini-
tial stakeholder group. Developing inclusion/exclusion criteria that makes
sense for your institutional capabilities is critical. An expanded view of
these criteria is given in Chapter 2 but several brief statements here can
help guide your initial decisions to target patient populations most likely
to experience improved survival with a new R-ECMO program. There are
approximately 290,000 in-hospital cardiac arrests each year in the U.S.
With a mean age of 66, this population will provide some of your best can-
didates for a R-ECMO program.14 Ideal criteria include witnessed arrest,
early application of basic life support, an initial shockable rhythm (19% of
IHCA patients), and known past medical history, to exclude patients with
end-stage liver, lung or kidney disease, or metastatic malignancy. Thus,
there is a unique opportunity to rigidly apply ECPR criteria for patients
who arrest in the hospital and potentially allow for early success of your
new program. In addition, duration of CPR prior to ECMO is prognostic
for survival.13,14 Patients who arrest in the hospital have the best potential
to be cannulated for ECMO. As IHCA patient arrests are often monitored
and receive rapid institution of quality CPR, they fulfill the established
goal of very short no-flow (no CPR) and low-flow (CPR only) time prior
to V-A ECMO. Low-flow time less than 60 minutes has been shown to
have a significantly improved survival rate.14,15 A specific cause of hemody-
namic instability that may prove beneficial for a new program is massive
pulmonary embolism which has a mortality rate as high as 65% and two
thirds of such deaths occur within 2 hours of symptom onset.15 We have

10 ECPR and Resuscitative ECMO

found when R-ECMO precedes cardiac arrest an important survival ben-
efit occurs. If cardiac arrest occurs but R-ECMO can be applied immedi-
ately, the mortality is still greatly improved.16

Caseload Evaluation
As R-ECMO patients impact multiple clinical service lines and units, pre-
dicting and preparing for the increased caseload is an important early ex-
ercise. Once inclusion/exclusion criteria have been chosen, you should
be able to approximate the potential target caseload from previous years’
cardiac arrest data from the ED, the CCL, and total number of massive
PE patients. To optimize resource utilization and staffing, consider many
variables including off-hours resource availability, physician training, and
ECMO console availability depending on seasonal surges, pediatric needs
(if resources are shared), or post-cardiotomy ECMO volume at your in-
stitution. Local market saturation may also affect these calculations. If
multiple hospitals in a small city offer R-ECMO this may dilute potential
total volume of cases (when indications are strictly followed). With lower
volume the overall complication rate may increase. Importantly, caseload
estimates facilitate approximation of costs including quantity of ECMO
consoles and disposables. Understanding these numbers and how they af-
fect each stakeholder is crucial to the success of your program.

Step 2: Design Key Systems and Protocols

As with many complex systems, a standardized R-ECMO approach and
team consistency is paramount to minimizing complications and opti-
mizing outcomes. For example, whether cardiac surgeons or emergency
physicians are implementing resuscitative ECMO, there should be an
accepted approach to patient selection and percutaneous ECMO cannu-
lation technique. The approach should strive to incorporate elements con-
sidered “best practice” without being overly rigid in areas of uncertainty.
The approach should also be tailored to your unique practice setting and
resources. As such, we additionally recommend a standardized approach
to R-ECMO team alerts and response, the ACLS delivery, post-cannulation
patient transport, left heart decompression, extremity reperfusion, and
other R-ECMO processes.17

Chapter 1  11
R-ECMO Team Response
Activation of the R-ECMO team for the patient in extremis can be in con-
junction with your hospital’s ‘shock team’ or an entirely new process may
need to be implemented. At our institution we created the H.E.L.P. New
Mexico team (Heart Embolism Lung Program) that serves as the initial
call for all physicians within the state, including those in-house, providing
immediate access to the R-ECMO physician and rapid deployment team.
Calls include all cardiac arrest alerts, ST-elevation myocardial infarction
(STEMI) alerts, cardiogenic shock patients, all submassive and massive
pulmonary embolism patients, and acute lung failure patients. These pro-
cess improvements may shorten the time to R-ECMO activation, which
can improve outcomes.13,14

ACLS Delivery
The importance of optimal ACLS delivery to achieve neurologically in-
tact survivors cannot be overstated. Thus, an efficient “roles based” ACLS
delivery system should be considered a prerequisite to ECPR. While the
various divisions of labor differ depending on local practice settings and
resources, a commonality should be a resuscitation leader tasked only
with coordination of the team and not involved in any particular proce-
dure.19 There are several well-described roles based ACLS approaches.19,20

Patient Transport
Patient transportation on ECMO is challenging and high risk best man-
aged using a standardized approach. A checklist assuring ECMO support
remains uncompromised throughout transport is advised. Best results oc-
cur when you standardize where equipment is stored, how patients are
transferred to and from their beds, what personnel is present, and what
routes and elevators to use throughout the building. In addition, use of
a portable transport ECMO stand (Figure 2) can limit risks when rolling
both the patient and an ECMO circuit.

Complication Mitigation
Benchmarks for complications in adult ECPR are found in the ELSO da-
tabase and should be considered as the initial starting numbers for any
program. For example, the ELSO database lists a cannulae site bleeding of
16.7% with 30% survival; CNS bleeding occurring in 2.8% of patients with

12 ECPR and Resuscitative ECMO

 UNM Transport Rack

Figure 2. UNM ECMO transport rack to assist with rapid transport both
in hospital and out of hospital transfers. The rack fits on the end of the
patient’s bed, comfortably resting between the legs. Shown unloaded and
then loaded with all ECMO equipment secured.

10% survival; and limb ischemia rate of 4.1% with 25% survival.4 Deviation
from these rates should trigger a quality reassessment to identify areas for
improvement. Any single ECMO case can have complications due to anat-
omy or disease related variables. Systemic changes to your process over a
single problem case are cautioned against, though consistent complication
trends should be met with rapid, appropriate adjustments. A cumbersome
administrative process to affect system change will not be beneficial if it
delays correction of an obvious issue. Complications are not always as ob-
vious as a death due to an error in cannulation. It can be as innocuous as
how the dressings on the cannulae are changed, leading to possible blood
stream infections, or whether dual oxygenators should be used to increase
flow on V-V patients with high cardiac output.

Chapter 1  13
 A standardized approach to initial and ongoing anticoagulation for
various R-ECMO patients helps manage the balance between thrombotic
and hemorrhagic complications. R-ECMO anticoagulation challenges in-
clude massive pulmonary embolism patients, those undergoing cardiac
catheterization who may be on dual antiplatelet medications, and those
with CPR-induced trauma or complications such as cannula site bleed-
ing. Enlisting surgeons who are willing to operate on fully anticoagulated
ECMO patients is strongly recommended. Our institution routinely per-
forms full body CT scans on all patients who received ECPR in the ED
due to the high frequency of injuries associated with prolonged CPR.21
In addition, a rapid CT scan of the head, chest, abdomen, and pelvis may
narrow the differential for the cardiac arrest, and if cerebral hemorrhage,
cerebral edema, massive pulmonary embolism, or an aortic dissection is
discovered, these will impact the care plans.22 There are few R-ECMO pro-
cesses that do not benefit from standardization.

Implementing Coronary Angiography

It is important to achieve consensus with interventional cardiology on
patient selection and timing of angiography and PCI in ECPR and cardio-
genic shock patients. In addition, reassuring cardiology that the R-ECMO
service is available for their unstable PCI cases creates a two-way street
of support that increases cooperation between services. Assuring a full
body CT scan occurs prior to angiography may limit trips to the CCL if the
cause of arrest is noncardiac. When iatrogenic or idiopathic hemorrhage
is discovered, this may guide the cardiologist. Active bleeding may limit
their approach and reduce their use of dual antiplatelet therapy.

Quality Assurance System

R-ECMO service has three main aspects requiring regular review: 1) ad-
herence to prearranged indications, 2) complications during cannulation
and ICU management, and 3) outcomes. Indications should be adhered
to 100% of the time. A young program needs to learn to say “No” fre-
quently. Tremendous amounts of pressure to place marginal patients on
ECMO is common. The R-ECMO team needs to thoroughly understand
the indications and exclusions and be aware that every early case will be
highly scrutinized. Since significant amounts of unknowns occur in ev-
ery cardiac arrest patient, known contraindications should provide a hard
stop for cannulation. Indications for R-ECMO should not be limited to

14 ECPR and Resuscitative ECMO

those in cardiac arrest as expansion of R-ECMO to those in extremis but
not have arrested, may potentially improve outcomes.23
Quality outcome data for R-ECMO programs often require time to
attain. Therefore survival, although important, should not be a primary
determinant of program success or failure. With average ECPR survival
being 29%, even programs with exemplary processes may still have several
“successful cannulations” that result in death before a survivor occurs. For
this reason, adherence to indications and lack of complications should be
the main initial program metrics. The R-ECMO team must reassure hospi-
tal leadership, nursing, and non-ECMO physicians that with low compli-
cations and wise patient selection, the 29% goal is attainable.4

Procedures: Cannulation, Extremity Reperfusion, Left Heart

Decompression, and Unanticipated Operations
Cannulation needs to be safely and skillfully performed. The practitioners
who cannulate must be well trained, calm under pressure, and able to
rapidly troubleshoot vascular access issues, wire mishaps, cannulation
problems, and circuit issues. There should be very little on-the-job train-
ing for ECPR. The most experienced cannulating physician should be the
only one performing intra-arrest cannulations. Training for cannulations
should be done in non-arrest situations that have significantly more lee-
way for mishaps or the delays inherent with teaching.
Creation of standardized ECMO equipment carts underpins quality
care. Our carts are laid out with wires and other disposables in a fash-
ion that mimics the process of cannulation from start to finish. Processes
to keep the carts fully stocked and strategically placed in the Emergency
Department, ICU, and CCL to ease geographic constraints are as import-
ant as the contents themselves. In addition, we have premade University of
New Mexico ECMO packs (Medline Industries Inc.) that have all disposable
ECMO cannulation supplies except wires and the cannulae. These kits are
inexpensive and allow us to have a standard setup each and every time.
While it remains a source of debate which V-A ECMO patients require
lower extremity reperfusion, some patients do need a reperfusion can-
nula. The monitoring of extremity perfusion via a standardized decision
tree and technique for reperfusion cannula placement is critical, as limb
ischemia is fraught with morbidity and mortality.4 Similarly, institutional
triggers and management approach for left heart decompression should
be settled in the planning stages not during an actual case. Lastly, gen-

Chapter 1  15
eral surgeons need to be aware and comfortable with operating on ECMO
patients. CPR has a published rate of liver lacerations from 0.6-2.1% and
other potential injuries may require immediate laparotomy.24, 25 With the
duration of CPR potentially reaching 60 minutes in an ECPR case the au-
thors postulate that even greater injuries may result. If these potential is-
sues have not been discussed with anesthesiology, and general and trauma
surgery you may have a potential survivor experience rapid demise.

STEP 3: Team Training

Once your program achieves consensus on key components of the opera-
tions and creates initial designs of these various systems, it is time to train
the multidisciplinary team. Your educational initiative should include lec-
tures and reference materials or video to introduce ECMO systems to vari-
ous groups. This is no small task as you will be targeting every stakeholder
group previously mentioned. Broadening this initiative to larger groups,
you will receive significant feedback to take into consideration as you con-
tinuously improve upon your initial design. Lectures alone are inadequate;
therefore, continued in-situ simulation should be a primary component of
R-ECMO system training.26,27 Simulation of every component of the pro-
cess should occur, including ACLS, ECMO cannulation, patient imaging,
transport, etc.
Bedside rounds provide an important educational opportunity. Given
the complexity of R-ECMO patients and the challenges of day-to-day man-
agement, it is important to maximize the learning potential for the entire
team. One strategy adopted from the Alfred Hospital in Melbourne Aus-
tralia, is the addition of separate daily bedside ECMO rounds. This per-
mits the exchange of ideas among faculty for some of the more complex
issues. It also reviews some of the fundamental management concepts
for learners at various stages of expertise. Additionally, routine morbidity
and mortality reviews (with set criteria for whom to evaluate) is key to
ongoing education assuring a quality control.

Step 4: Putting it All Together

Thus far, organizing a program and creating a multidisciplinary team has
required tremendous efforts spanning many months. The logistical plan
that was created initially will show completion of each of those main

16 ECPR and Resuscitative ECMO

 1. Day -1 ECMO initiation day
a. Cannulation procedure note (provided by physician)
i. CPT codes
1. 33952-ECMO insertion peripheral, Percutaneous
2. 33954-ECMO insertion peripheral, Open
b. Initiation note (provided by physician)
i. Need to put “ECMO Initiation Note” in the subject line
(bills more than daily management note)
ii. CPT codes
1. 33946-ECMO Initiation, V-V
2. 33947-ECMO Initiation, V-A
2. Daily documentation (day 2 and beyond - provided by
physician)
a. ECMO daily management: put circuit recommendations
here, e.g. sweep, flow, anticoagulation, cannulas, pressure
limits, hemolysis labs, pre/post membrane gases
i. CPT codes
1. 33948-ECMO Daily Management, V-V
2. 33949-ECMO Daily Management, V-A
b. Inpatient consult visit (ECMO consult team separate
from ICU team) OR critical care note (ECMO services
provided by ICU team): put any non-circuit recommenda-
tions here, e.g. ventilator management, diuresis, sedation,
tracheostomy, etc.
i. CPT codes: 99251,99252, 99253,99254,99255—In-patient
Consult Visit
ii. 99291, +99292—Critical Care
c. Cannula repositioning procedure note: (if performed-
must be significant reposition)
i. CPT codes:
1. 33958-ECMO reposition peripheral, Percutaneous
2. 33962-ECMO reposition peripheral, Open

Figure 3. ECMO Initiation and Billing Tips (>6 years old).

Chapter 1  17
 3. Decannulation day
a. Decannulation procedure note (provided by physician)
i. CPT codes:
1. 33966-ECMO removal peripheral, Percutaneous
2. 33984-ECMO removal peripheral, Open

Figure 3. (continued)

tasks: hospital leadership has provided support for the R-ECMO program.
The multidisciplinary team meets regularly, and each group understands
their role on the team. Following the process flowchart (Figure 1) will
ensure an organized stakeholder team that has well designed key systems
and protocols and adequate training of all involved with prehospital pa-
tient care, in-hospital patient care, cannulation, nursing, ongoing ECMO
care, and transport of these patients. Many details are beyond the scope
of this chapter but must not go overlooked. Such specifics include which
ECMO devices and cannulae to purchase, designing ECMO order-sets on
your electronic medical record system, training your coders to identify
all the critical care and ECMO billing6 (Figure 3) nuances among others.
Finally, it is time to announce the go-live date. Consider a hospital
wide launch of the R-ECMO program with an educational event that gar-
ners excitement, describes the purpose of R-ECMO, R-ECMO indications,
and the potential benefits to the entire hospital population. While waiting
for a qualified patient, the cannulation team should review the indications
and exclusionary criteria and the standardized cannulation process. The
R-ECMO team should assure that all potential activations are closely eval-
uated. Their attendance at all cardiac arrests and other potential resusci-
tations gives the team the greatest opportunity to activate the system. If
not a candidate, to assist the team caring for the arrest with intra-arrest
arterial and venous access. This provides the chance to practice skills that
are difficult to attain.
The success of that first patient is not a result of the 5-15 minutes
it takes to cannulate but rather from the fruits of 10-12 months of labor
and preparation. As a result of the organizing committee’s tremendous

18 ECPR and Resuscitative ECMO

efforts your R-ECMO program will be primed for success and ready to
provide the highest level of resuscitation to the sickest patients in your
institution.

References:
1. Benjamin, E.J., et al., Heart Disease and Stroke Statistics—2018 Up-
date: A Report From the American Heart Association. Circulation, 2018.
137(12): p. e67-e492.
2. Yam N, McMullan DM. Extracorporeal cardiopulmonary resuscita-
tion. Ann Transl Med. 2017;5(4):72.
3. Yannopoulos, D., et al., Minnesota Resuscitation Consortium’s Ad-
vanced Perfusion and Reperfusion Cardiac Life Support Strategy for
Out-of-Hospital Refractory Ventricular Fibrillation. Journal of the
American Heart Association, 2016. 5(6): p. e003732.
4. ELSO, E.L.S.O., International Summary. 2019, ELSO.
5. Panchal, A.R., et al., 2019 American Heart Association Focused Update
on Advanced Cardiovascular Life Support: Use of Advanced Airways,
Vasopressors, and Extracorporeal Cardiopulmonary Resuscitation During
Cardiac Arrest: An Update to the American Heart Association Guidelines
for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care.
Circulation, 2019. 140(24): p. e881-e894.
6. Abbott, Abbott Coding Guide, in Extracorporeal Membrane Oxygenation
(ECMO). 2020: One St. Jude Medical Dr., St. Paul MN 55117, USA.
7. Bharmal, M.I., et al., Cost-utility of extracorporeal cardiopulmonary
resuscitation in patients with cardiac arrest. Resuscitation, 2019. 136:
p. 126-130.
8. Shinar, Z., ACLS improvements with development of an ECPR program.
2014.
9. Albright, D., Assistant Professor, R.-E.O.F. Chart, Editor. 2020.
10. Conrad, S.A., et al., Percutaneous cannulation for extracorporeal mem-
brane oxygenation by intensivists: a retrospective single-institution case
series. Read Online: Critical Care Medicine| Society of Critical Care
Medicine, 2015. 43(5): p. 1010-1015.
11. Bellezzo, J.M., et al., Emergency physician-initiated extracorporeal car-
diopulmonary resuscitation. Resuscitation, 2012. 83(8): p. 966-970.

Chapter 1  19
12. Burrell, A.J., et al., Percutaneous cannulation in predominantly venoar-
terial extracorporeal membrane oxygenation by intensivists. Critical care
medicine, 2015. 43(12): p. e595.
13. Andersen, L.W., et al., In-Hospital Cardiac Arrest: A Review. JAMA,
2019. 321(12): p. 1200-1210.
14. Bartos, J.A., et al., Improved Survival With Extracorporeal Cardio-
pulmonary Resuscitation Despite Progressive Metabolic Derangement
Associated With Prolonged Resuscitation. Circulation, 2020. 141(11):
p. 877-886.
15. Bĕlohlávek, J., V. Dytrych, and A. Linhart, Pulmonary embolism, part
I: Epidemiology, risk factors and risk stratification, pathophysiology,
clinical presentation, diagnosis and nonthrombotic pulmonary embolism.
Experimental and clinical cardiology, 2013. 18(2): p. 129-138.
16. Guliani, S., et al., Veno-Arterial Extracorporeal Membrane Oxygenation
Is An Effective Management Strategy For Massive Pulmonary Embolism
Patients. Journal of Vascular Surgery: Venous and Lymphatic Disor-
ders, 2020.
17. Osofsky, R.B., et al., Cannula-Associated Limb Ischemia Severity Score
Predicts Need for Distal Perfusion Catheter Placement in Venoarterial
Extracorporeal Membrane Oxygenation Patients. Journal of Vascular
Surgery, 2020. 72(1): p. e129-e130.
18. Link, M.S., et al., Part 7: Adult Advanced Cardiovascular Life Support.
Circulation, 2015. 132(18_suppl_2): p. S444-S464.
19. Nixon, P., Alfred ICU ECPR Role Cards. 2016: intensiveblog.com.
20. Shinar, Z., J. Bellezzo, and S. Weingart, EDECMO Episode 12- The
Nurse-Based ECMO Program at Sharp Memorial Hospital with Suzanne
Chill-out RN, BSN, in EDECMO, Z. Sinar, Editor. 2014.
21. Zotzmann V, Rilinger J, Lang CN, Duerschmied D, Benk C, Bode C,
Wengenmayer T, Staudacher DL. Early full-body computed tomog-
raphy in patients after extracorporeal cardiopulmonary resuscitation
(eCPR). Resuscitation. 2020 Jan 1;146:149-154.
22. Ryu, J.-A., et al., The association of findings on brain computed tomogra-
phy with neurologic outcomes following extracorporeal cardiopulmonary
resuscitation. Critical care (London, England), 2017. 21(1): p. 15-15.
23. Guerrero-Miranda, C.Y. and S.A. Hall, Cardiac catheterization and
percutaneous intervention procedures on extracorporeal membrane
oxygenation support. Annals of Cardiothoracic Surgery, 2018. 8(1): p.
123-128.

20 ECPR and Resuscitative ECMO

24. Buschmann, C.T. and M. Tsokos, Frequent and rare complications
of resuscitation attempts. Intensive Care Medicine, 2009. 35(3): p.
397-404.
25. Das Gupta, J., et al., Decompressive laparotomy for a patient on
VA-ECMO for massive pulmonary embolism that suffered traumatic liver
laceration after mechanical CPR. Journal of Surgical Case Reports,
2018. 2018(10).
26. Whitmore, S.P., et al., Simulation training enables emergency medicine
providers to rapidly and safely initiate extracorporeal cardiopulmonary
resuscitation (ECPR) in a simulated cardiac arrest scenario. Resuscita-
tion, 2019. 138: p. 68-73.
27. Su, L., et al., Implementation of an Extracorporeal Cardiopulmonary
Resuscitation Simulation Program Reduces Extracorporeal Cardiopul-
monary Resuscitation Times in Real Patients*. Pediatric Critical Care
Medicine, 2014. 15(9).

Chapter 1  21
CHAPTER 2
Optimizing the Pre-ECMO
Resuscitation

Joshua C. Reynolds

Introduction
Successful rescue of refractory cardiac arrest with extracorporeal
cardiopulmonary resuscitation (ECPR) is predicated on minimizing the
degree of ischemic insult sustained during the interval between collapse
and extracorporeal restoration of circulation. To this end, conventional

Foundational Terms
ɋ High quality cardiopulmonary resuscitation (CPR) - CPR
involving adequate rate of chest compressions (100-120/min),
adequate depth of chest compressions (5 cm), and adequate
recoil of the chest (complete removal of pressure from the
chest). This also includes minimizing interruptions in chest com-
pressions, early defibrillation with appropriate pad placement,
and airway management which may include intubation, laryngeal
mask airway (LMA), or simple non-rebreather mask
ɋ No flow – Period of time when an arresting patient did not
receive bystander CPR
ɋ Low flow – Period of time that an arresting patient received
chest compressions

 23
Figure 1: A primary goal of conventional resuscitation prior to ECPR is to
mitigate ischemic insult during the ‘low-flow’ period, the elapsed interval
between onset of chest compressions and spontaneous or extracorporeal
return of circulation, such that neurologic and systemic recover is possible
on ECMO. Brief periods of ischemic insult are typically better tolerated
than more prolonged periods of ischemic insult.

CPR is not a definitive therapy; it rarely produces sufficient cardiac out-

put and gas exchange to sustain normal metabolic activity.1-2 Rather, con-
ventional CPR is a mitigating bridge to restoration of a perfusing heart
rhythm, restoration of sufficient cardiac output and/or systemic vascular
resistance to support circulation, or, in this particular case, extracorpo-
real restoration of circulation. The primary goal of conventional CPR is
prolonging the window of tolerable ischemic insult, until one of these
other endpoints is achieved and/or the etiology of cardiac arrest is cor-
rected (Figure 1). Thus, for an ECPR candidate, the aim of conventional
CPR is to provide artificial circulation that approximates normal physiol-
ogy, thereby minimizing ischemic insult and maximizing the likelihood of
neurologic, cardiovascular, and metabolic recovery on ECMO (extracor-
poreal membrane oxygenation).

24 ECPR and Resuscitative ECMO

 The ischemic insult from cardiac arrest is categorized as the ‘no-flow’
period, which describes the interval between collapse and initiation of
chest compressions, and the ‘low-flow’ period, which describes the inter-
val between initiation of chest compressions and return of spontaneous
(or extracorporeal) circulation (ROSC). Unless the cardiac arrest is wit-
nessed in a healthcare setting, the primary means to minimize ischemic in-
jury during ‘no-flow’ are layperson recognition of cardiac arrest, initiation
of bystander CPR, and utilization of public access defibrillation.3 These
critical public health interventions are beyond the scope of this particu-
lar text, but selection criteria for ECPR candidates frequently account for
some estimate of ‘no-flow’ (Chapter 3). This chapter will address means
to minimize ischemic insult during ‘low-flow’, primarily by optimizing
cardiac output, delivering energy substrate to tissues, and promoting gas
exchange. It will outline the best available evidence to guide ideal resusci-
tation of cardiac arrest, but also note when adjustments or supplements to
these practices are prudent in preparation for ECPR.

Time-Dependent Considerations
Overarching the transition from conventional CPR to ECPR is a
time-dependent shift in the risk/benefit ratio of continuing conventional
CPR or initiating ECPR. Clinical features that render patients attractive
candidates for ECPR (e.g. younger age, fewer comorbidities, shockable
cardiac rhythm, witnessed collapse, bystander CPR) are also associated
with higher likelihood of ROSC and favorable clinical outcome after con-
ventional CPR.4-6 Ideally, ECPR should be deployed in a manner that maxi-
mizes potential benefit to patients while minimizing unnecessary exposure
to complications and costs. Distracting from the emphasis on high-quality
conventional CPR is potentially harmful and some ECPR candidates, by
virtue of their favorable clinical features, will achieve ROSC within min-
utes of conventional resuscitation.7 However, traditional resuscitation
most often fails and the likelihood of favorable clinical outcome steadily
declines with elapsed duration of resuscitation.4,5 Population-based obser-
vational studies suggest that for cardiac arrest patients at-large, the like-
lihood of functional recovery at hospital discharge reaches an inflection
point approximately 20 minutes after onset of professional CPR (Figure
2). It asymptotically plateaus between 10-30 minutes, depending on the
presence or absence of particular features (i.e. shockable rhythm, wit-

Chapter 2 25
 A 30%
B 30%

25%
Probaility of mRS 0-3 25%

Probaility of mRS 0-3

20% 20%

15% Shockable 15%

Witnessed
10% 10%

5% 5%
Non-shockable Unwitnessed

0% 0%
0 10 20 30 40 50 0 10 20 30 40 50
CPR Duration (minutes) CPR Duration (minutes)
Shockable (n) 2,522 1,772 993 439 122 Witnessed (n) 5,317 4,007 2,373 967 227
Non-shockable (n) 8,846 7,423 4,718 1,635 321 Unwitnessed (n) 6,051 5,118 3,338 1,110 216

30%
C 30%
D
25% Q1: 0 - 6.9 minutes
25%
Q2: 7 - 8.9 minutes

Probaility of mRS 0-3

Probaility of mRS 0-3

Q3: 9 - 10.9 minutes

20% 20%
Q4: ≥ 11 minutes

15% 15%

10% 10%
Bystander CPR

5% 5%
No bystander CPR
0% 0%
0 10 20 30 40 50 0 10 20 30 40 50
CPR Duration (minutes) CPR Duration (minutes)
Bystander CPR (n) 4,342 3,495 2,046 719 127 Q1 (n) 2,844 2,300 1,409 505 111
No bystander CPR (n) 7,026 5,700 3,665 1,355 316 Q2 (n) 2,862 2,376 1,455 548 113
Q3 (n) 2,820 2,311 1,378 539 117
Q4 (n) 2,842 2,208 1,269 482 102

Figure 2: The probability of survival to hospital discharge with acceptable

functional recovery (modified Rankin scale 0-3) steadily diminishes with
elapsing duration of resuscitation. The rate and inflection point of decline
varies by the presence or absence of specific case features, such as initial
shockable cardiac rhythm (A), witnessed cardiac arrest (B), bystander
CPR (C), and ‘no-flow’ interval (elapsed time from 911 dispatch to onset
of professional resuscitation) (D), Gray lines denote the 95% confidence
around each point estimate (black lines). Reproduced from Reynolds JC,
et al. Circulation. 2016;134(25):2084-94.

nessed collapse, bystander CPR, and ‘no-flow’ interval).4,5,8 For ECPR can-
didates, who inherently display more favorable features, 10-20 minutes of
professional resuscitation (closer to 20 minutes for shockable rhythms;
closer to 10 minutes for nonshockable rhythms) appears to strike a bal-
ance between maximizing clinical outcomes with conventional resusci-
tation and operating within the time-constrained therapeutic window of
ECPR.9,10

26 ECPR and Resuscitative ECMO

Chest Compressions
Chest Compression Parameters
Chest compressions are the crux of successfully mitigating ischemic
injury during ‘low-flow’ prior to ECPR. Clinical data guide optimal pa-
rameters for chest compressions in most adults. Patients with outliers of
habitus or thoracic dimensions may require adjustments to these param-
eters. Typically, with hands positioned over the lower sternum, compres-
sions should be performed with a rate of 100-120 per minute, to a depth
of at least 2 inches (5 cm), and with approximately equal time divided
between compression and recoil (i.e. duty cycle). It is paramount to avoid
leaning on the chest to allow for complete chest wall recoil between com-
pressions.12 Moreover, interruptions in chest compression should be min-
imized such that compression fraction (proportion of total resuscitation
time in which chest compressions are performed) is at least 80%.12 To this
end, it is imperative to minimize periprocedural pauses in chest compres-
sions.13,14 Chest compression pauses dissipate the accumulated arteriove-
nous pressure gradient built with each successive compression, such that
each pause necessitates an additional bout of compressions to rebuild the
same gradient. This phenomenon is readily observed with coronary per-
fusion pressure (CPP) in preclinical models of cardiac arrest (Figure 3).

Particular Considerations for ECPR Candidates

Several mechanical devices are commercially available to deliver highly
reliable and uninterrupted chest compressions, utilizing either sternal or
circumferential thoracic compression/decompression.15,16 Although no
convincing evidence suggests superior clinical outcomes with these de-
vices in the cardiac arrest population at-large, they reliably deliver chest
compressions. As such, mechanical chest compression devices seem to be
most useful in settings with limited personnel, challenging human or en-
vironmental factors. Additionally, they seem to generate less torso move-
ment compared to manual chest compressions, potentially facilitating
vascular access for ECPR. Given the complex orchestration of personnel
and equipment required to institute ECPR, it is prudent to utilize a me-
chanical chest compression device for both logistical simplification and
assurance of reliable chest compressions.

Chapter 2 27
28
Advanced Specific Considerations for ECPR
Evidence-Based Practice Candidates
Chest ■ Position hands over the lower sternum ■ Target diastolic blood pressure 30-40 mmHg
Compressions ■ Rate 100-120 per minute ■ Ultrasound may guide hand placement to avoid
■ Depth at least 2 inches iatrogenic left ventricular outflow tract obstruction
■ Do not lean and allow complete chest recoil ■ Continue chest compressions during the initial
■ Chest compression fraction at least 80% seconds of ECPR until blood flow exceeds 2.0 L/min
■ If total blood flow drops below 2.0 L/min after
ECMO initiation, resume chest compressions while
troubleshooting the circuit
Airway ■ Advanced airway strategy does not yield superior ■ Select the advanced airway they can insert most
Management and clinical outcomes over bag-mask ventilation quickly and with the highest probability of success on
Gas Exchange ■ Choose most experienced clinician for advanced the first attempt
airway ■ Mechanical ventilation parameters during chest
■ Use the highest available oxygen concentration compressions:
during CPR • Volume assist-control
■ Titrate supplemental oxygen to normoxia after ROSC • Tidal volume: 6 mL/kg
or ECPR • Peak flow: 30 L/min

ECPR and Resuscitative ECMO

• Maximum peak pressure: 100 mmHg
• Respiratory rate: 10
• FiO2 100%
• PEEP: 0
 Advanced Specific Considerations for ECPR
Evidence-Based Practice Candidates
Pharmacotherapy ■ Periodic bolus administration of 1 mg epinephrine ■ Excessive α-adrenergic stimulation risks post-arrest
increases the likelihood of ROSC, short-term survival, microvascular insufficiency and the ‘no-reflow’
and longer-term survival; the effects on neurologic phenomenon
outcome beyond hospital discharge are uncertain ■ Excessive β-adrenergic stimulation risks post-arrest
■ Early epinephrine likely more effective myocardial dysfunction
■ Epinephrine may be more effective in patients with a ■ Titrate epinephrine dosing to diastolic blood
nonshockable cardiac rhythm pressure 30-40 mmHg
■ Antidysrhythmics increase the likelihood of ROSC but
the effects on longer-term outcome are uncertain
■ Use metabolic therapies and antidotes in a targeted
fashion when specific etiologies of cardiac arrest are
known or strongly suspected

Chapter 2
Defibrillation ■ Minimize the total duration of the peri-shock pause ■ Repeated unsuccessful shocks exacerbate myocardial
■ ‘Coarse’ appearing ventricular fibrillation typically stunning after extracorporeal restoration of
indicates a greater degree of metabolic substrate circulation
within the myocardium and has a higher probability ■ If ECPR is imminent, defer additional defibrillation
of successful defibrillation than ‘fine’ appearing attempts of fine-appearing VF until extracorporeal
ventricular fibrillation restoration of circulation
■ Optimize extracorporeal circulatory parameters
(e.g. flow, mean arterial pressure, sweep) before
re-attempting defibrillation

29
Figure 3: Five successive chest compressions generate an arterio-venous
pressure gradient. In this case, coronary perfusion pressure is the pressure
gradient between the aorta and right atrium. During individual compres-
sions, pressure increases in both the aorta and right atrium, whereas during
relaxation, differential pressure persists in the aorta. Thus, myocardial
blood flow is closely related to coronary perfusion during the relaxation
phase of each chest compression. Note that coronary perfusion during
relaxation increases over the initial 3-4 compressions, but then declines
within only a few seconds after compressions are paused. Reproduced
from Callaway CW, et al. Textbook of Critical care 7th edition. Philadelphia,
PA: Elsevier, 2017.

The arterial sheath placed during the initial phases of ECPR (Chapter
4) can be leveraged to provide a real-time metric to titrate chest compres-
sion parameters and other hemodynamic interventions. As noted, CPP is
the gradient between aortic and right atrial pressure. Observational studies
in humans note no instances of ROSC when CPP failed to rise above 15
mmHg, but it was measured fairly late in the course of resuscitation after
out-of-hospital cardiac arrest.17 Large animal models of cardiac arrest mea-
suring CPP much earlier during resuscitation suggest that CPP closer to 25

30 ECPR and Resuscitative ECMO

mmHg is associated not only with ROSC, but also with improved cerebral
perfusion pressure and brain oxygen tension.18-20 Assuming a normal central
venous pressure of 5-10 mmHg, targeting diastolic blood pressure of 25-30
mmHg should approximate CPP of 20 mmHg. If central venous pressure
rises during the ‘low flow’ state of cardiac arrest, then diastolic blood pres-
sure of 35-40 mmHg would be a more conservative target to approximate a
CPP of 20 mmHg.
In tandem with verification of vascular cannula placement, use of
transesophageal ultrasound can guide optimal hand/mechanical device
placement during chest compressions. Given anatomic heterogeneity in
the population, the left ventricular outflow tract (LVOT) occasionally
falls in the region of compression when hand placement is guided by
standard anatomic landmarks. Compressing the LVOT is often counter-
productive to maximizing cardiac output generated by CPR. Preclinical
data suggest that ultrasound-guided compressions directly over the left
ventricle generate higher CPP,21 and human case series report feasibility
of transesophageal echocardiographic-guided hand placement away from
the left ventricular outflow tract.22,23
Little direct evidence guides the optimal timing to cease chest com-
pressions after extracorporeal restoration of circulation. One practical
approach is to continue external chest compressions during the initial
seconds of ECPR until total blood flow approaches 2.5–3 L/min. Prema-
ture cessation before sufficient extracorporeal recovery of blood flow
risks interruptions to coronary and cerebral perfusion. Belated cessation
hinders blood flow via obstructive interference with the ECMO pump.
Total blood flow for a patient on ECMO comprises native cardiac output
and extracorporeal flow. In turn, total flow and systemic vascular resis-
tance dictate mean arterial pressure. Chapters 7 and 9 further elucidate
these concepts and delineate hemodynamic goals during the initial hours
of ECMO.

Airway Management and Gas Exchange

Airway Management
The optimal strategy of airway management during cardiac arrest remains
subject to debate. The secure and definitive nature of tracheal intubation
is potentially offset by its oft-requisite interruption in chest compressions,

Chapter 2 31
which can extend for unacceptable intervals.14 In contrast, bag-valve mask
ventilation requires no procedural pause in chest compressions per se but
must be interposed with chest compressions and risks both aspiration and
an ineffective seal. Supraglottic airways theoretically require less time to
insert compared to tracheal intubation, provide a measure of security,
and, depending on the particular type, offer some protection against aspi-
ration. Several large, randomized trials conducted in different prehospital
systems of care with a range of intubation success rates provide the best
evidence to date on which strategy is optimal.24 Even in settings with high
tracheal intubation success, an initial strategy of advanced airway inser-
tion (either supraglottic airway or tracheal intubation) was not superior
to bag-mask ventilation during CPR for improving clinical outcomes. Of
note, no randomized data compare an initial strategy of bag-mask ven-
tilation to supraglottic airway insertion. Both bag-mask ventilation and
tracheal intubation are defined skills that require specific training and
clinical experience to achieve and maintain proficiency. Bag-mask ventila-
tion may be adequate during initial resuscitation, provided it is objectively
successful. The rationale for transitioning from bag-mask ventilation to
an advanced airway is highly dependent on clinical context, including
patient anatomy, aspiration risk, and oxygenation/ventilation success. If
an advanced airway is required, providers should select a strategy (e.g.
tracheal intubation or supraglottic airway) based on clinical context and
their particular skill set.

Oxygenation and Ventilation

Both ‘no-flow’ and ‘low-flow’ periods of cardiac arrest are characterized
by central venous hypoxemia and insufficient oxygen delivery to tissues,
which manifests as post-resuscitation oxygen debt. Though limited data
guide the optimal fraction of inspired oxygen during CPR, it is biologi-
cally reasonable to use the highest available oxygen concentration. After
spontaneous or extracorporeal restoration of circulation, supplemental
oxygen should be titrated to normoxia in lieu of continued hyperoxia.25
Accounting for anatomic dead space, chest compressions alone are
unlikely to generate significant ventilation in humans.2 The need for gas
exchange must be balanced with the inevitable interruption in chest com-
pressions needed for delivery of synchronized breaths. An advanced air-
way does allow for gas exchange to occur asynchronously in parallel with

32 ECPR and Resuscitative ECMO

chest compressions. The historical practice of delivering ventilations at
a ratio of 15 compressions per two ventilation is associated with lower
likelihood of survival compared to 30 compressions per two ventilations,
but continuous compressions with positive pressure ventilation do not
yield superior survival or neurologic outcome over and above a ratio of 30
compressions per two ventilations.26

Monitoring
Waveform capnography is a valuable tool to confirm correct advanced
airway placement, monitor airway patency, and monitor adequacy of cir-
culation. During cardiac arrest, end-tidal CO2 (EtCO2) is proportionate
to cardiac output and pulmonary blood flow.27 EtCO2 levels may be very
low (<10 mmHg) at onset of resuscitation; whereas, artificial circulation
should cause EtCO2 values to increase and these levels may be used as
feedback to improve or modify chest compressions. EtCO2 values greater
than 20 mmHg are associated with successful resuscitation, where values
persistently less than 10 mmHg after at least 20 minutes of resuscitation
are associated with failure of resuscitation.28-30 However, certain resusci-
tation drugs disrupt the association between capnography readings and
pulmonary blood flow. Epinephrine reduces EtCO2 levels and sodium bi-
carbonate produces a transient elevation of EtCO2 levels.

Particular Considerations for ECPR Candidates

The expected clinical course for ECPR candidates, the complex orches-
tration of cannulation, and the desire for an unbiased measure of EtCO2
necessitate some advanced airway maneuver beyond bag-mask ventila-
tion. However, this does not automatically mandate tracheal intubation.
Though there are no randomized data specifically among ECPR candi-
dates, several decades of operating room experience and indirect evidence
from traditional resuscitation of cardiac arrest suggest that a temporary
supraglottic airway is perfectly sufficient to prevent aspiration and pro-
vide meaningful gas exchange.31,32 The typically shorter interval required
for insertion compared to tracheal intubation makes this an attractive
option. Once ECPR is initiated, the supraglottic airway may then be ex-
changed for an endotracheal tube in a controlled fashion. Alternatively, if
a supraglottic airway is failing to prevent aspiration or provide meaningful

Chapter 2 33
gas exchange during traditional CPR, tracheal intubation is required while
simultaneously preparing for ECPR. In either case, advanced airway ma-
neuvers should be performed in such a fashion to maximize probability
of success with the first attempt while minimizing interruptions in chest
compressions. This may necessitate particular models of supraglottic air-
ways or video laryngoscopy in lieu of direct laryngoscopy.
Depending on clinical context, it may be desirable to introduce me-
chanical ventilation into the resuscitation of ECPR candidates. Given the
sophisticated organization of personnel and equipment required to insti-
tute ECPR, simplifying logistics and assuring reliable ventilation param-
eters are desirable.33 Suggested initial ventilator settings are described in
Table 1. Of note, a high peak pressure limit is required due to simultane-
ous external chest compressions.

Pharmacotherapy
Vasopressors
Epinephrine increases systemic vascular resistance via α-adrenergic re-
ceptor agonism to augment the arteriovenous pressure gradient generated
by external chest compressions. However, β-adrenergic effects also in-
crease myocardial oxygen consumption, ectopic ventricular arrhythmias,
hypoxemia from pulmonary arteriovenous shunting, and post-arrest myo-
cardial dysfunction. Furthermore, excessive α-adrenergic effect impairs
microcirculation, which potentially offsets any benefit from improving
macrocirculation. To this end, epinephrine consistently improves return
of circulation and short-term survival, but its effects on neurologic recov-
ery are less certain.34 In massive, population-based, matched cohort stud-
ies, use of prehospital epinephrine is associated with higher probability of
ROSC, but lower probability of 1-month survival and favorable neurologic
outcome, which suggests that when epinephrine is required to restore car-
diac activity, severe brain injury has already occurred or that it contrib-
utes to brain injury.35 Randomized trials of epinephrine in out-of-hospital
cardiac arrest consistently demonstrate superior ROSC, survival to hos-
pital admission, and survival to hospital discharge compared to placebo.34
The larger trial noted epinephrine yielded more survivors with poor neu-
rologic outcome at hospital discharge but overall improved 3-month sur-
vival. The net increase in survivors comprised subjects with both favorable

34 ECPR and Resuscitative ECMO

and unfavorable neurologic outcome without a clear statistical difference
between the two groups. Of note, the neurologic status of comatose car-
diac arrest survivors can improve for up to 6 months.36
Multiple observational studies and a secondary analyses of ran-
domized trial data suggest that epinephrine is more effective when ad-
ministered earlier during resuscitation. This time-dependent effect is
consistently observed for ROSC and other short-term outcomes, but in-
consistently observed for longer-term survival and neurologic outcome.34
Secondary analyses of randomized data suggest that epinephrine is more
effective in patients with non-shockable cardiac rhythms compared to
shockable cardiac rhythms. This interaction is more pronounced for
ROSC and survival, and less pronounced for neurologic outcome.34
Given the consistent improvements in ROSC and survival, the un-
certain neurologic effects, and the potential for time-dependent and
rhythm-dependent effects, epinephrine remains in most cardiac arrest
algorithms.37 However, optimal dosing and timing of epinephrine remains
an active area of investigation. Cardiac arrest is the sole disease in which
vasopressors are administered in a rote, repeated fashion without active
titration to particular physiologic parameter. Since the cumulative dose
of epinephrine is colinear with the duration of resuscitation, scant data
inform the maximum tolerated cumulative dose of epinephrine in OHCA
without imparting adverse adrenergic effects. It is worth noting that 1 mg
of epinephrine administered every 3-5 minutes approximates 3-5 mcg/kg/
min for a 70 kg adult, whereas typical epinephrine dosing for shock ranges
from 0.1-1.0 mcg/kg/min. Titrated epinephrine infusion is an emerging
paradigm described in preclinical models and cases of in-hospital cardiac
arrest with invasive hemodynamic monitoring already in place.38,39 This
approach may balance positive and negative adrenergic effects of epi-
nephrine by reducing the cumulative and time-averaged dose, but it lacks
rigorous testing in of out-of-hospital cardiac arrest.
Despite its pharmacologic ability to increase systematic vascular re-
sistance without excessive α-adrenergic stimulation, randomized data
indicate no clear benefit of vasopressin in addition to or in lieu of epi-
nephrine. Vasopressin is deemphasized in many cardiac arrest treatment
algorithms for the sake of simplicity.34

Chapter 2 35
Antidysrhythmics
The physiologic rationale for administering antidysrhythmic drugs during
cardiac arrest is extrapolated from cardiac electrophysiology. Amiodarone
increases action potential duration and the refractory period in myocytes
while slowing atrioventricular conduction.40 Lidocaine prolongs the myo-
cyte refractory period, decreases ventricular automaticity, and suppresses
ventricular ectopy via local anesthetic action as a sodium channel inhibitor.
It suppresses activity of depolarized arrhythmogenic tissue by depressing
action potential phase 0 but does not interfere with normal myocardium.40
Randomized data of antidysrhythmic drugs during cardiac arrest
demonstrate superior ROSC with amiodarone or lidocaine compared
to placebo. Neither resulted in superior longer-term survival or neuro-
logic outcome. A priori subgroup analysis of randomized data suggests
a time-dependent effect of active drug (amiodarone or lidocaine) com-
pared to placebo.41 Given the potential to promote ROSC, the possibility
of time-dependent effects, and the absence of evidence demonstrating
overt harm, it is reasonable to administer either amiodarone or lidocaine
to patients after unsuccessful initial defibrillation attempts.

Metabolic Therapies
Empiric treatments of metabolic disturbances (e.g. sodium bicarbonate
or other buffers, dextrose, calcium, and magnesium) lack supporting data
in the broad cardiac arrest population at-large.42-48 However, it is still ap-
propriate to treat known or suspected metabolic abnormalities with spe-
cific metabolic agents and/or antidotes.

Particular Considerations for ECPR Candidates

ECPR candidates require a delicate balance of epinephrine. Expedited ad-
ministration promotes the arteriovenous pressure gradient generated by
chest compressions, but excessive vasoconstriction from α-adrenergic
stimulation risks technical challenges to vascular cannulation, restriction
of total blood flow after ECMO initiation, and post-arrest microvascular
insufficiency (i.e. no-reflow phenomenon). Likewise, expedited admin-
istration (especially in patients with non-shockable rhythms) promotes
chronotropy and inotropy, but excessive β-adrenergic stimulation risks

36 ECPR and Resuscitative ECMO

post-arrest myocardial dysfunction. As noted previously, the central arte-
rial catheter placed during phased institution of ECPR (Chapter 4) can
be leveraged to provide a real-time metric to guide hemodynamic inter-
ventions. Depending on the estimate of central venous pressure, titrated
epinephrine infusion in concert with fine-tuning of chest compression pa-
rameters should target diastolic blood pressure 30-40 mmHg to approxi-
mate CPP 20 mmHg.

Defibrillation
Transthoracic defibrillation converts ventricular dysrhythmias to an orga-
nized cardiac rhythm via multiple mechanisms, including depolarization
of the myocardium, cancellation of aberrant wavefronts, and/or prolon-
gation of the refractory period.49 Since excessive exposure to electrical
energy damages the myocardium, rescue shocks should be delivered at
the lowest effective energy while minimizing the number of unsuccessful
shocks.50 Successful defibrillation hinges on adequate delivery of electri-
cal current to the myocardium. Transthoracic impedance is opposition
of the thorax to the electrical circuit created when current is applied to
patients during defibrillation. Greater body weight, larger body surface
area, larger thoracic diameter, and smaller pad/paddle surface area are all
associated with higher transthoracic impedance.
Defibrillation attempts typically necessitate an unavoidable brief
pause in compressions. The total duration of pause both before defibrilla-
tion (pre-shock pause) and after defibrillation (post-shock pause) consti-
tutes the peri-shock pause. The duration of peri-shock pause is inversely
associated with both defibrillation success and clinical outcomes, likely
reflecting the underlying decline in CPP associated with pauses in chest
compressions.13 To minimize peri-shock pause, the defibrillator should be
charged during chest compressions, chest compressions should pause for
no more than a few seconds during shock delivery, chest compressions
should resume immediately after shock delivery, and subsequent assess-
ment of cardiac rhythm and presence of a palpable pulse should be delayed
by several minutes.51
The ventricular fibrillation (VF) waveform, in particular, is a dy-
namic, quantitative entity characterized by amplitude, frequency, and
nonlinear organization. Immediately after onset of cardiac arrest, VF
waveforms typically have a large amplitude, high frequency, highly nonlin-

Chapter 2 37
Figure 4: The ventricular fibrillation waveform in the upper rhythm strip
has a ‘coarse’ appearance with a higher amplitude, higher frequency, and
greater degree of nonlinear organization. Histochemical analysis of the
myocardium would typically demonstrate a larger reservoir of high energy
phosphates and other cellular energy substrate. Defibrillation of this wave-
form typically has a higher probability of success. The ventricular fibrillation
waveform in the lower rhythm strip has a ‘fine’ appearance with a lower
amplitude, lower frequency, and lower degree of nonlinear organization.
Histochemical analysis of the myocardium would typically demonstrate
a smaller reservoir of high energy phosphate and other cellular energy
substrate. Defibrillation of this waveform typically has a lower probability of
success.

ear appearance that is visually coarse. Untreated, they gradually degrade

into small amplitude, low frequency, linear appearance that is visually
fine (Figure 4). This quantitative deterioration reflects consumption
of high-energy phosphates and metabolic substrate in the myocardium.
Chest compressions that generate sufficient coronary perfusion pressure
recirculate metabolic substrate and quantitively restore degraded VF to
a more robust appearance. In preclinical models and retrospective clini-
cal analyses, the probability of defibrillation success is directly associated
with ‘coarseness’ of the VF waveform.52,53 No prospective human studies
have specifically tested whether altering treatment by predicting defibril-
lation success improves either defibrillation success or clinical outcomes.

38 ECPR and Resuscitative ECMO

Particular Considerations for ECPR Candidates
Given the objective of minimizing ischemic injury, shortening peri-shock
pause is especially critical for ECPR candidates. This requires orchestra-
tion of defibrillator charging and merely the briefest of pauses in chest
compression to deliver a rescue shock. Given the inevitable deterioration
of the VF waveform, as resuscitation progresses after initial unsuccessful
defibrillation attempts and ECPR is imminent, it is reasonable to defer
additional defibrillation attempts until extracorporeal restoration of met-
abolic substrate in the myocardium. Moreover, clinicians may elect to de-
fer subsequent rhythm checks altogether to maximize chest compression
fraction until extracorporeal restoration of circulation.

Evaluating for Reversible Etiologies of

Cardiac Arrest
Diagnosing a reversible etiology of cardiac arrest permits both specific
treatments to correct it and more nuanced consideration of whether to
utilize ECPR. Such etiologies include primary cardiac events (either isch-
emia or primary dysrhythmia), correctable causes of obstructive shock
(i.e. pneumothorax, pulmonary embolism, and cardiac tamponade), elec-
trolyte disturbance, poisoning, trauma/hemorrhage, and hypothermia.
ECPR can bridge most of these conditions to definitive therapy or recov-
ery. Historical elements, physical exam findings, and bedside radiographic
or sonographic findings raise or lower suspicion for various etiologies of
cardiac arrest in a Bayesian fashion. Clinicians should pay particular at-
tention to historical elements such as the events leading up to cardiac
arrest, comorbidities, medication lists, and recent interactions with the
healthcare system. Although the precise sensitivities and specificities are
not certain, bedside ultrasound is increasingly used to screen for various
etiologies of cardiac arrest. The typical framework of sonographic assess-
ment comprises serial evaluation for specific etiologies of hypovolemic,
cardiogenic, and obstructive shock.54

Chapter 2 39
Conclusions
Optimizing the pre-ECMO resuscitation of ECPR candidates demands
orchestration of personnel and resources. The overarching goals of re-
suscitation are to minimize ischemic injury, search for a correctable etiol-
ogy of cardiac arrest, and facilitate transition to rescue ECPR after 10-20
minutes of professional resuscitation. The therapies employed to achieve
these goals should be guided by the best available evidence with prudent
modifications for ECPR candidates. In lieu of rote, algorithmic interven-
tions, pre-ECMO resuscitation of ECPR candidates typically necessitates
invasive hemodynamic monitoring, a critical appraisal of clinical context,
dynamic adaptation to the course of resuscitation, and a more nuanced
approach to correcting physiologic derangements.

References
1. Safar P, Brown TC, Holtey WJ, Wilder RJ. Ventilation and circulation
with closed-chest cardiac massage in man. JAMA 1961;176:574–576.
2. Deakin CD, O’Neill JF, Tabor T. Does compression-only cardiopul-
monary resuscitation generate adequate passive ventilation during
cardiac arrest? Resuscitation 2007;75:53–59.
3. Brady WJ, Mattu A, Slovis CM. Lay Responder Care for an Adult
with Out-of-Hospital Cardiac Arrest. N Engl J Med. 2019 Dec
5;381(23):2242-2251.
4. Reynolds JC, Grunau BE, Rittenberger JC, Sawyer KN, Kurz MC,
Callaway CW. The association between duration of resuscitation and
favorable outcome after out-of-hospital cardiac arrest: implications
for prolonging or terminating resuscitation. Circulation. 2016 Dec
20;134(25):2084-94.
5. Reynolds JC, Frisch A, Rittenberger JC, Calllaway CW. Duration of
resuscitation efforts and functional outcome after out-of-hospital
cardiac arrest: when should we change to novel therapies? Circula-
tion. 2013;128(23):2488-94.
6. Grunau BE, Reynolds JC, Scheuermeyer FX, et al. Comparing the
prognosis of those with initial shockable and non-shockable rhythms
with increasing durations of CPR: Informing minimum durations of
resuscitation. Resuscitation. 2016;101:50-6.

40 ECPR and Resuscitative ECMO

 7. Brouwer TF, Walker RG, Chapman FW, Koster RW. Association
between chest compression interruptions and clinical outcomes of
ventricular fibrillation out-of-hospital cardiac arrest. Circulation.
2015;132(11):1030-7.
8. Nagao K, Nonogi H, Yonemoto N, et al. Duration of prehospital
resuscitation efforts after out-of-hospital cardiac arrest. Circulation.
2016; 133:1386–1396.
9. Reynolds JC, Grunau BE, Elmer J, et al. Prevalence, natural history,
and time-dependent outcomes of a multi-center North American co-
hort of out-of-hospital cardiac arrest extracorporeal CPR candidates.
Resuscitation. 2017;117:24–31.
10. Grunau B, Reynolds J, Scheuermeyer F, et al. Relationship between
time-to-ROSC and survival in out-of-hospital cardiac arrest ECPR
candidates: when is the best time to consider transport to hospital?
Prehosp Emerg Care. 2016; 20:615–622.
11. Ewy GA. The mechanism of blood flow during chest compressions
for cardiac arrest is probably influenced by the patient’s chest con-
figuration. Acute Medicine and Surgery. 2018;5(3):236-40.
12. Kleinman ME, Brennan EE, Goldberger ZD, et al. Part 5: Adult Basic
Life Support and Cardiopulmonary Resuscitation Quality: 2015
American Heart Association Guidelines Update for Cardiopulmo-
nary Resuscitaiton and Emergency Cardiovascular Care. Circulation.
2015 Nov 3;132(18 Suppl 2):S414-35.
13. Cheskes S, Schmicker RH, Christenson J, et al. Perishock pause: an
independent predictor of survival from out-of-hospital shockable
cardiac arrest. Circulation. 2011;124(1):58-66.
14. Wang HE, Simeone SJ, Weaver MD, Callaway CW. Interruptions in
cardiopulmonary resuscitation from paramedic endotracheal intuba-
tion. Ann Emerg Med. 2009 Nov;54(5):645-652.e1.
15. Brooks SC, Hassan N, Bigham BL, Morrison LJ. Mechanical versus
manual chest compressions for cardiac arrest. Cochrane Database
Syst Rev 2014;2:CD007260.
16. Perkins GD, Lall R, Quinn T, et al. PARAMEDIC trial collaborators.
Mechanical versus manual chest compression for out-of-hospital
cardiac arrest (PARAMEDIC): a pragmatic, cluster randomised con-
trolled trial. Lancet. 2015;385:947–955.

Chapter 2 41
17. Paradis NA, Martin GB, Rivers EP, et al. Coronary perfusion pressure
and the return of spontaneous circulation in human cardiopulmo-
nary resuscitation. JAMA. 1990 Feb 23;263(8):1106-13.
18. Reynolds JC, Salcido DD, Menegazzi JJ. Coronary perfusion pressure
and return of spontaneous circulation after prolonged cardiac arrest.
Prehosp Emerg Care. 2010 Jan-Mar;14(1):78-84.
19. Morgan RW, French B, Kilbaugh TJ, et al. A quantitative comparison
of physiologic indicators of cardiopulmonary resuscitation quality:
diastolic blood pressure versus end-tidal carbon dioxide. Resuscita-
tion. 2016;104:6-11.
20. Sutton RM, Friess SH, Bhalala U, et al. Hemodynamic directed CPR
improves short-term survival from asphyxia-associated cardiac ar-
rest. Resuscitation. 2013 May;84(5):696-701.
21. Anderson KL, Gastaneda MG, Boudreau SM, Cox K, Bebarta VS.
Ultrasound guided chest compressions over the left ventricle during
cardiopulmonary resuscitation increases coronary perfusion pres-
sure and return of spontaneous circulation in a swine model of
traumatic cardiac arrest. Circulation. 2014;130:A15853.
22. Teran F, Dean AJ, Centeno C, et al. Evaluation of out-of-hospital
cardiac arrest using transesophageal echocardiography in the emer-
gency department. Resuscitation. 2019 Apr;137:140-147.
23. Zanatta M, Benato P, Cianci V. Ultrasound guided chest com-
pressions during cardiopulmonary resuscitation. Resuscitation.
2015;87:e13-4.
24. Granfeldt A, Avis SR, Nicholson TC, et al. Advanced airway manage-
ment during adult cardiac arrest: A systematic review. Resuscitation.
2019;139:133-143.
25. Holmberg MJ, Ticholson T, Nolan JP, et al. Oxygenation and
ventilation targets after cardiac arrest: a systematic review and
meta-analysis. Resuscitation. 2020;152:107-15.
26. Nichol G, Leroux B, Wang H, et al. Trial of continuous or interrupted
chest compressions during CPR. N Engl J Med. 2015;373:2203-14.
27. Weil MH, Bisera J, Trevino RP, Rackow EC. Cardiac output and
end-tidal carbon dioxide. Crit Care Med. 1985;13:907–909.
28. Salen P, O’Connor R, Sierzenski P, et al. Can cardiac sonography and
capnography be used independently and in combination to predict
resuscitation outcomes? Acad Emerg Med. 2001;8:610–615.

42 ECPR and Resuscitative ECMO

29. Callaham M, Barton C. Prediction of outcome from cardiopulmo-
nary resuscitation from end-tidal carbon dioxide concentration. Crit
Care Med. 1990;18:358–362.
30. Levine RL, Wayne MA, Miller CC. End-tidal carbon dioxide
and outcome of out-of-hospital cardiac arrest. New Engl J Med.
1997;337:301–306.
31. Wang HE, Schmicker RH, Daya MR, et al. Effect of a Strategy of
Initial Laryngeal Tube Insertion vs Endotracheal Intubation on
72-Hour Survival in Adults With Out-of-Hospital Cardiac Arrest: A
Randomized Clinical Trial. JAMA. 2018 Aug 28;320(8):769-778.
32. Benger JR, Kirby K, Black S, et al. Effect of a Strategy of a Supraglot-
tic Airway Device vs Tracheal Intubation During Out-of-Hospital
Cardiac Arrest on Functional Outcome: The AIRWAYS-2 Random-
ized Clinical Trial. JAMA. 2018 Aug 28;320(8):779-791.
33. Weingart S. Podcast 125 – The New Intra-Arrest (Cardiac Arrest
Management). EmCrit. https://emcrit.org/emcrit/new-intra-arrest/
Accessed April 1, 2019.
34. Holmberg MJ, Issa MS, Moskowitz A, et al. Vasopressors during
adult cardiac arrest: A systematic review and meta-analysis. Resusci-
tation. 2019 Jun;139:106-121.
35. Hagihara A, Hasegawa M, Abe T, Nagata T, Wakata Y, Miyazaki
S. Prehospital epinephrine use and survival among patients with
out-of-hospital cardiac arrest. JAMA. 2012;307:1161–1168.
36. Tong JT, Eyngorn I, Mlynash M, Albers GW, Hirsch KG. Functional
neurologic outcomes change over the first 6 months after cardiac
arrest. Crit Care Med. 2016;44:e1202-7.
37. Panchal AR, Berg KM, Hirsch KG, et al. 2019 American heart Asso-
ciation Focused Update on Advanced Cardiovascular Life Support:
Use of Advanced Airways, Vasopressors, and Extracorporeal Car-
diopulmonary Resuscitation During Cardiac Arrest: AN Update to
the American Heart Association Guidelines for Cardiopulmonary
Resuscitation and Emergency Cardiovascular Care. 2019 American
Heart Association Focused Update on Advanced Cardiovascular Life-
Support: Use Circulation. 2019 Dec 10;140(24):e881-e894.
38. Friess SH, Sutton RM, French B, et al. Hemodynamic directed CPR
improves cerebral perfusion pressure and brain tissue oxygenation.
Resuscitation. 2014;85:1298–303.

Chapter 2 43
39. Sutton RM, Friess SH, Naim MY, et al. Patient-centric blood
pressure-targeted cardiopulmonary resuscitation improves survival
from cardiac arrest. Am J Respir Crit Care Med. 2014;190:1255–1262.
40. Soar J, Nolan JP, Böttiger BW, et al. European Resuscitation Concil
Guidelines for Resuscitation 2015: Section 3. Adult Advanced Life
Support. Resuscitation. 2015 Oct;95:100-47.
41. Ali MU, Fitzpatrick-Lewis D, Kenny M, et al. Effectiveness of anti-
arrhythmic drugs for shockable cardiac arrest: A systematic review.
Resuscitation. 2018 Nov;132:63-72.
42. Bar-Joseph G, Abramson NS, Kelsey SF, et al. Improved resuscita-
tion outcome in emergency medical systems with increased usage
of sodium bicarbonate during cardiopulmonary resuscitation. Acta
Anaesthesiol Scand. 2005;49:6–15.
43. Vukmir RB, Katz L; Sodium Bicarbonate Study Group. Sodium bicar-
bonate improves outcome in prolonged prehospital cardiac arrest.
Am J Emerg Med. 2006;24:156–161.
44. Mader TJ, Smithline HA, Gibson P. Aminophylline in undiffer-
entiated out-of-hospital asystolic cardiac arrest. Resuscitation.
1999;41:39–45.
45. Mader TJ, Smithline HA, Durkin L, Scriver G. A randomized con-
trolled trial of intravenous aminophylline for atropine-resistant
out-of-hospital asystolic cardiac arrest. Acad Emerg Med.
2003;10:192–197.
46. Longstreth WT Jr, Copass MK, Dennis LK, Rauch-Matthews ME,
Stark MS, Cobb LA. Intravenous glucose after out-of-hospital car-
diopulmonary arrest: a community-based randomized trial. Neurol-
ogy. 1993;43:2534–2541.
47. Fatovich DM, Prentice DA, Dobb GJ. Magnesium in cardiac arrest
(the MAGIC trial). Resuscitation. 1997;35:237–241.
48. Thel MC, Armstrong AL, McNulty SE, Califf RM, O’Connor CM.
Randomized trial of magnesium in in-hospital cardiac arrest: Duke
Internal Medicine Housestaff. Lancet. 1997;350:1272–1276.
49. White RD. New concepts in transthoracic defibrillation. Emerg Med
Clin N Am. 2002;20:785–807.
50. Walcott GP, Killingsworth CR, Ideker RE. Do clinically relevant
transthoracic defibrillation energies cause myocardial damage and
dysfunction? Resuscitation. 2003;59:59–70.

44 ECPR and Resuscitative ECMO

51. Cheskes S, Common MR, Byers PA, Zhan C, Morrison LJ. Compres-
sions during defibrillator charging shortens shock pause duration
and improves chest compression fraction during shockable out of
hospital cardiac arrest. Resuscitation. 2014 Aug;85(8):1007-11.
52. Callaway CW, Menegazzi JJ. Waveform analysis of ventricular
fibrillation to predict defibrillation. Curr Opin Crit Care. 2005
Jun;11(3):192-9.
53. Lightfoot CB, Nremt-P, Callaway CW, et al. Dynamic nature
of electrocardiographic waveform predicts rescue shock out-
come in porcine ventricular fibrillation. Ann Emerg Med. 2003
Aug;42(2):230-41.
54. Atkinson P, Bowra J, Milne J, et al. International Federation for
Emergency Medicine Consensus Statement: Sonography in hypoten-
sion and cardiac arrest (SHoC): An international consensus on the
use of point of care ultrasound for undifferentiated hypotension and
during cardiac arrest. CJEM. 2017 Nov;19(6):459-470.

Chapter 2 45
CHAPTER 3
ECPR Candidacy Assessment
and Outcomes

Brian Grunau • Jan Belohlavek

Introduction
Extracorporeal cardiopulmonary resuscitation (ECPR) for refrac-
tory cardiac arrest is a resource intensive intervention and may be deemed
untenable if the proportion of favourable outcomes is low.1,2 For this rea-
son, careful patient selection is critical. Examples of protocols for patient
selection criteria from select centers are displayed in Figures 1, 2, 3 and
5. For many out-of-hospital cardiac arrest patients, few details about the
patient’s past medical history and arrest etiology are known. As such, eli-
gibility often rests upon the Utstein characteristics and patient physiolog-
ical parameters.3
The current literature is suboptimal for a distinct and unified ECPR
selection criterion. First, ideal patient characteristics are derived from

Foundational Terms
ɋ ROSC – Return of spontaneous circulation – cardiac arrest pa-
tient who transitioned to having a palpable pulse
ɋ IHCA – In-hospital cardiac arrest
ɋ OHCA – out-of-hospital cardiac arrest

 47
observational published data of those treated with ECPR. This limits our
ability to ascertain survivability beyond these highly selected cohorts. Sec-
ond, selection for ECPR treatment by clinicians is limited by bias, which
limits external validity. Beyond the scope of reported inclusion criteria,
clinicians likely apply an additional gestalt in their decision to perform
ECPR;4,5 thus, if an inclusion criteria is objectively and comprehensively
applied in another setting results may differ. Finally, many studies have
been ambiguous in regard to differentiating cases initiated during active
chest compressions versus those commenced after ROSC. The latter rep-
resents a better prognosis and should not be classified as “ECPR.”6
While the literature can assist with defining those most likely to ben-
efit from ECPR, there remain cases with unfavourable characteristics who
survive,7 demonstrating our incomplete knowledge of this disease and its
outcomes. However, ECPR programs are resource intensive and depend
on support from clinical teams and hospital leadership. Varying views on
the proportion of survivors that classify a program as “successful” impacts
the constituents of an ECPR eligibility criterion, depending on whether
the aim is to be more restrictive or inclusive.
With these limitations in mind, this section reviews the individual
components for ECPR eligibility with corresponding evidence and out-
comes.

Clinical Trial Data for ECPR Compared to

Traditional CPR
Overall research supports benefit of ECPR compared to traditional re-
suscitation. Multiple case series have shown better outcomes with ECPR
than predicted population studies.4,17,70 Brussels, Taiwan, and San Di-
ego have published propensity matched analyses that suggest benefit to
ECPR over traditional measures.69,71,75 One notable exception was the lack
of benefit seen in OHCA in Paris.76 The SAVE-J trial showed a statisti-
cally significant improvement in one month survival for patients going
to ECPR centers (12.3% vs. 1.5% p<0.001) compared to those who did
not.68 The ARREST trial was the first randomized control trial in ECPR.
This study showed 6 of 14 patients in the ECPR cohort surviving to dis-
charge with CPC of all survivors at 1 or 2 at 3 months. No patient in the
standard ACLS group survived to 4 months. The study was stopped early

48 ECPR and Resuscitative ECMO

 Appendix C: The Prague Hyperinvasive Protocol Criteria

Belohlavek et al. Journal of Translational Medicine 2012; 10:16

Used with Permission

Figure 1: The Prague Hyperinvasive Approach to OHCA.

because of low probability for standard ACLS to show benefit. This study
showed that in a well-organized system ECPR in VF patients was better
than standard ACLS. The Prague OHCA Hyperinvasive trial randomized
OHCAs in the prehospital setting to either standard care or a bundle ther-
apy including transport to hospital under mechanical CPR and in-hospital
ECPR initiation. It showed results favoring ECMO for 30-day neurologi-
cal outcome, 180-day mortality, and patients resuscitated for more than
45 minutes. (Figure 1).72 Unfortunately, clinical trials remain limited to
a single inclusion criterion, and thus are only able to inform of benefits
to that specific patient group rather than providing evidence for optimal
eligibility criteria. However, the ARREST trial and Hyperinvasive trial give
significant credence to the efficacy of ECPR over standard ACLS.

Age Limits
Age is commonly incorporated into ECPR criteria, often applying a maxi-
mum of 60-75 years old. While age is associated with overall cardiac arrest
outcomes,8 supporting data in ECPR is conflicting. One European study
examining 423 ECPR-treated patients found young age was not associ-

Chapter 3 49
ated with favourable neurological outcomes at hospital discharge.9 Other
studies have reported a similar lack of association between age and ECPR
outcomes.10-14 Age limits within the 60-75 year range may not differenti-
ate predicted survival beyond comorbidities and other Utstein elements;
however, a ceiling age may still be reasonable. A study of patients who
received ECPR for an out-of-hospital cardiac arrest (OHCA) showed no
favourable 1-month neurological outcomes in the 22 patients aged ≥75
years.11
Elderly patients may have less resilience with prolonged periods of
CPR. A study reported an 18% favourable neurological outcome in those
>75 years old, but no positive outcomes in those with low-flow intervals
(measured from the initiation of CPR until extracorporeal membrane
oxygenation [ECMO] support) >60 minutes.15 Further complicating the
age-outcome relationship are differences of cardiac arrest etiology be-
tween age groups. Pathologies that affect younger cardiac arrest victims
(e.g. aortic dissection, occult drug overdose) may be less amenable to
ECPR treatment; whereas, etiologies more prevalent in older adults (e.g.
acute coronary syndrome) may represent better ECPR candidates. One
study found that the peak survival rate occurred at the age of 60, and a
bimodal peak of survival with favourable neurological outcome at ages 40
and 65.11 Overall, age appears to be a poor predictor of ECPR-treatment
outcomes; however, an age limit of 75 appears to be a reasonable ceiling
for ECPR eligibility, especially for prolonged low-flow durations typically
seen with OHCAs.

Bystander CPR
The correlation between bystander CPR and ECPR-treated patient out-
comes is mixed. Some studies have reported an association with sur-
vival,9,12,16 while others have not.14,17-20 One such study found that among
cases without bystander CPR 3.8% of patients had favourable neurological
outcomes; however, in another cohort it was reported at 13%.9,19 While the
outcomes of those without bystander CPR appear to be lower than com-
parators, methods to further risk stratify these cases may be appropriate.
One ECPR program that did not exclude patients based on bystander CPR
status reported that 2/12 (17%) cases without bystander CPR survived to
hospital discharge.18 According to existing data, cases with bystander CPR

50 ECPR and Resuscitative ECMO

may have an increased probability of favourable outcomes; however, data
remain insufficient to use it as an exclusion criteria.

Witnessed Arrest and No Flow Duration

Publications of ECPR-treated cases include few unwitnessed arrests
treated with ECPR as they are often deemed ineligible. However, within
available data, witnessed arrest is consistently associated with better
outcomes.16,17 Several cohorts have reported no survivors among unwit-
nessed cases.21,22 In one study of 111 ED and OHCA ECPR cases, only 1
of 16 (6.2%) patients with an unwitnessed arrest survived to discharge.20
Conversely, in another study of 156 OHCA, 3 of 21 (14%) patients with
unwitnessed arrests who received ECPR survived.19 These data suggest
that there may be cases of unwitnessed arrest with very short no-flow in-
tervals who may benefit from ECPR.
Initial cardiac rhythm may serve to risk stratify those who likely had a
preceding short no-flow interval. One study of patients treated with con-
ventional resuscitation reported that among patients with initial shock-
able rhythms, there was 94% certainty that the preceding no-flow interval
was <10 minutes.23 Thus, initial rhythm may supersede witnessed status.
A program that included only patients with initial shockable rhythms re-
ported (Figure 2) 33% (4/12) survival in their unwitnessed subgroup.18
Several ECPR programs have incorporated no-flow intervals into the
inclusion criteria (Figure 3)24-26; however, evidence supporting distinct
cutoffs is lacking. Establishing reliable estimates of no-flow durations is
always difficult. One study examined the relationship between no-flow
duration and outcomes among ECPR-​eligible cases treated with conven-
tional resuscitation showed no favourable neurological outcomes with
no-flow >10 minutes.27
In summary, witnessed arrests have a better prognosis than unwit-
nessed arrest; however, unwitnessed cases with initial shockable rhythms
may remain reasonable candidates. For witnessed or unwitnessed cases
with known no-flow intervals, an evidence-based threshold for ineligibil-
ity is lacking; however, 10 minutes may be a reasonable ceiling to exclude
eligibility.

Chapter 3 51
52 ECPR and Resuscitative ECMO
Figure 2. (opposite page) The University of Minnesota Refractory VF/
VT OHCA protocol. The criteria and decision-making process from emer-
gency medical services (EMS) evaluation through cardiac catheterization
laboratory (CCL) assessment and treatment are shown. Timely patient
delivery to the CCL, evidence of adequate cardiopulmonary resuscitation
(CPR)-generated perfusion, hemodynamic stabilization on CCL arrival,
and reperfusion therapy were the 4 pillars of the protocol. ABG = arterial
blood gas; ACLS = advanced cardiac life support; CICU = cardiac intensive
care unit; DC = direct current; ECLS = extracorporeal life support; ETCO2 =
end-tidal carbon dioxide; Hg = mercury; IABP = intra-aortic balloon pump;
ITD = impedance threshold device; IV/IO = intravenous/intraosseous; LUCAS
= Lund University Cardiac Arrest System; O2 Sat = fraction of oxygen sat-
urated hemoglobin; OHCA = out of hospital cardiac arrest; PaO2 = arterial
partial pressure of oxygen; PCI = percutaneous coronary intervention; PRN
= pro re nata or as needed; ROSC = return of spontaneous circulation; VF =
ventricular fibrillation; VT = ventricular tachycardia. Used with permission.

Low-Flow Duration
Numerous studies have identified a clear relationship between outcomes
and the low-flow (CPR-to-ECMO support) interval.12,13,28-32 One study of
133 ECPR-treated OHCA and IHCAs reported survival of 67%, 29%, 10%,
and 6% with low-flow durations <20, 20-45, 45-60, and 60-135 minutes
(Figure 4).29 Another study of 79 ECPR-treated OHCAs (with good neu-
rological outcomes in 14%) reported that 40 minutes was the optimal cut
point to discriminate between good and poor neurological outcomes. At
40 minutes the likelihood of survival fell from approximately 30% to 15%,
with the largest low-flow interval in a case with a good neurological be-
ing approximately 70 minutes.12 A similar study reported that 60 minutes
was the best cut-off, with survival 48% above and 12% below this inter-
val.33 Similarly, in one study of 52 ECPR-treated OHCAs, no survivor had
a low flow duration ≥66 min.4 Taken together, the likelihood of positive
hospital-discharge outcomes are highly correlated with low-flow duration
and are rare when it exceeds 65-70 minutes. As time is the paramount
parameter influencing ECPR outcomes, modified by ECPR programme
logistics, vigorous efforts should be implemented to shorten the low-flow
interval.34

Chapter 3 53
 PATIENT IDENTIFICATION
ST. PAUL'S HOSPITAL ECPR Protocol
WRITE PATIENT NAME HERE AND LOOK UP
HOSPITAL-BASED RECORDS
ECPR ACTIVATION CHECKLIST

ALS Crew # (circle): 261-A1 (Cambie) 245-A1 (King Ed) 246-A1 (Burnaby) 256-A1 (NVan)
Paramedic Dispatch time (“Time 0”) __________ Must be ≤ 50 minutes*
Estimated arrival time at SPH ___________ ª Use arrest time If EMS-witnessed
ª If periods of ROSC, minus from total time

*If you need to phone the EMS crew back for more information, call BCAS dispatch at 604-708-7500
[For HYPOTHERMIA-related arrests these guidelines do not apply, please call CV Surgeon to discuss]

PREHOSPITAL INCLUSION (all need to be met)

Patient Characteristics
£ Age ≤ 65
£ No major co-morbidities (including CHF, COPD/significant lung disease, dialysis, liver failure,
malignancy, alcoholism) or pre-existing major neurological deficits
£ Not too large for Lucas Device
Arrest Characteristics
£ No history or evidence of recent recreational drug use
£ Non-traumatic
£ Witnessed arrest (seen or heard)
£ One of the following:
• Initial shockable rhythm
• Signs of Life with CPR: Moving or Gasping or Pupils ≤ 5
£ Cause of arrest is either:
• No obvious cause (ie. presumed cardiac)
• Overdose of cardiac toxins (Β-blocker, CC-blocker, Psych meds, digoxin)
£ ETCO2 > 10 mmHg
£ EMS Dispatch Time (~911 Call) à SPH < 50 min
(Use arrest time if EMS-witnessed; If periods with ROSC, subtract from total time)
If all checked: Activate ECPR
Protocol

EMERGENCY DEPARTMENT-LEVEL EXCLUSION

(any one item makes a patient ineligible)
£ Hospital record review finds a major co-morbidity (listed above) or major neurological deficits
£ ETCO2 < 10
£ Lactate > 18 OR PaO2 < 50 (from ABG sample obtained during cannulation; do not delay ECMO
initiation for results)
£ If initial non-shockable rhythm, signs of life with CPR no longer present

If none checked: Initiate ECPR

GIVE THIS SHEET TO THE RECORDING NURSE IN THE RESUSC BAY Page 1 of 1
*This document will become part of the patient’s chart*

Figure 3. Vancouver ECPR Protocol for OHCA

54 ECPR and Resuscitative ECMO

 eCPR Survival
80
***

60

40
%

20

0
6-20 20-45 45-60 60-135
CPR duration (min)
n 14 33 43 43

Figure 4. Survival among 133 ECPR-treated in- and out-of-hospital refrac-

tory cardiac arrest cases, stratified by the duration of professional CPR
(***p=0.001). Wengenmayer et al. Critical Care (2017) 21:157. Used with
permission.

Initial Rhythm
The association of initial rhythm with successful conventional cardiac
arrest resuscitations is well established. The ability for defibrillation to
achieve ROSC likely plays a major role in this association. In theory, initial
rhythm in refractory cardiac arrest cases treated with ECPR should not
have predictive power, as none have ROSC. The degree of neurologic injury
should not be dependent on rhythm but rather blood flow. However, stud-
ies consistently show differential outcomes based on initial rhythm.5,13,14
One study of 68 ECPR-treated OHCAs (overall mean low-flow interval
84 min) reported 32% and 0% survival in cases with initial shockable and
non-shockable rhythms, respectively.35 A large European study (n=423)
reported results of 24% vs. 11%, respectively.9

Chapter 3 55
 The survival difference between these types of rhythms may be me-
diated by differences in resilience to low-flow durations. One inpatient
study (190 patients with median low-flow interval 30 min) examined the
inter-play between initial cardiac rhythm, low- flow intervals, and out-
comes.31 Authors reported favourable neurological outcomes classified by
initial cardiac rhythms: initial shockable (49%), PEA (34%), and asystole
(12%). Interestingly, increasing no-flow duration had a stronger detri-
mental impact on the outcomes of those with PEA compared to those
with initial shockable rhythms. They reported that low-flow intervals of
22 minutes for PEA and 46 minutes for shockable rhythms were the best
cut-points to discriminate good vs. poor neurological outcomes. This
particular study examined IHCAs and achieved very short low-flow du-
rations overall—likely the reason for highly successful outcomes within
the shockable and non-shockable cases alike. However, the comparably
poor outcomes of PEA cases with low-flow intervals >22 minutes may
help explain the dismal outcomes of non-shockable rhythms cases among
OHCA cohorts where low-flow intervals are typically prolonged.35 Taken
together, non-shockable cases may be reasonable candidates to consider
for ECPR provided low-flow intervals are short, although these cases will
represent a small proportion of overall candidates. In contrast, cases with
initial shockable rhythms may result in positive outcomes, even with lon-
ger low-flow intervals.

Location of Arrest
Several studies have compared OHCA vs. IHCA’s. One systematic review
reported that OHCA was associated with worse survival, but not neuro-
logical outcomes.16 A large study (n=423) reported better outcomes with
OHCA cases.9 Another study of ECPR comparing 230 OHCAs and IHCAs
reported that despite OHCAs having a significantly longer duration from
collapse to ECMO (68 vs. 44 minutes), the proportion of favourable neu-
rological outcomes were similar (26 vs. 25%, respectively).36 OHCAs have
been demonstrated to have more favourable baseline characteristics than
IHCAs.36-39 These cases are more likely to have been previously well but
struck by a sudden catastrophic event resulting in hemodynamic collapse
(such as an arrhythmia), which are more amendable to a short success-
ful ECMO course. In comparison IHCA patients are usually ill and rep-

56 ECPR and Resuscitative ECMO

resent candidates with compromised physiology. Although OHCAs may
have more favourable characteristics for ECPR, the challenge is to achieve
rapid ECMO support with a low-flow interval compatible with survival.

Presumed Etiology
Acute Coronary Syndrome
In some studies an ischemic coronary cause of the cardiac arrest has been
identified as a predictor of positive outcomes.21,31 This may explain the
association of initial shockable rhythm with improved outcomes, which
is usually absent in etiologies not amendable to ECPR (intracranial hae-
morrhage, aortic dissection, or massive bleeding).35 One study of 74
ECPR-treated OHCAs showed 35% favourable neurological outcomes at
ICU discharge in those with coronary etiologies versus 20% in those with-
out.40 Interestingly, among cases with non-shockable cardiac rhythms,
67% still had a CAD-etiology (versus 76% among those with shockable
initial rhythms). Another investigation found approximately the same
proportion of shockable and nonshockable cases with a ACS-etiology.35
Thus differentiating etiology of arrest by rhythm is far from clear.
One large retrospective registry study (n=252 OHCAs) reported a
frequent occurrence of myocardial infarction in refractory cardiac arrest
treated with ECPR (63% in OHCA, 55% in IHCA).39 The increase in in-
cidence of acute coronary syndromes with age,41 may help explain why
the best outcomes are in those aged 60, rather than younger patients.11
Further a past history of cardiovascular disease has been shown to be in-
dependently associated with improved ECPR outcomes.9,24 Conversely,
other studies show similar proportions of ischemic coronary etiologies
between those with positive and negative outcomes.35

Pulmonary Embolism
One study examining cases with known or suspected PE reported survival
with favourable neurological outcome in 4 of 19 ECPR-treated cases (me-
dian low-flow duration 77 minutes [IQR 39 -98]) compared with 0 of 20
matched historical controls treated primarily with conventional resuscita-
tion.42 In another case series, 8/9 ECPR cases with PE survived (six with
OHCA, all had PEA initial rhythm). ECPR was initiated based on PE sus-

Chapter 3 57
picion from echocardiography images.43 Another study of “high risk” PE
cases reported 2/18 survivors among those who had ECPR initiated during
cardiac arrest.44 PE may thus represent an appropriate indication for
ECPR, even with prolonged low-flow periods; however, identifying these
cases from others with non-shockable rhythms may prove challenging.

Myocarditis and Other Primary Cardiac Causes

ECPR-treated cases with myocarditis may have a better prognosis than
comparators45; however, establishing this etiology during a cardiac arrest
is difficult. One study showed 4 out of 12 ECPR patients with myocarditis
survived.46 Another ECPR ELSO-database study reported a 61% survival
in myocarditis, 43% in cardiomyopathy, 18% in sepsis, and 27% in myo-
cardial infarction.47 Preexisting arrhythmia was a significant univariate
predictor of favourable neurological outcomes in a propensity matched
study.4

Drug Overdose
Drug overdose of cardiotoxic agents is included in some ECPR eligibility
criteria.48 Reports have indicated success for agents such as calcium chan-
nel blockers, beta blockers, or cardiac glycosides that produced electrical
instability.49-51 One study reported survival in 3/10 cases treated with ECPR
for toxicological causes.52 Overall there is little data to base eligibility on
specific toxins. The incidence of V-A ECMO use for toxicological indica-
tions appears to be low.53,54 Overdoses that produce respiratory depres-
sion and subsequent arrest probably represent unfavourable candidates
from ECPR; however, evidence supporting this statement is unavailable.
Cardiac arrest cases with sudden primary cardiac etiologies usually have
a normoxemic status preceding the cardiac arrest which helps maintain
cerebral oxygenation during CPR; in contrast, cardiac arrest due to respi-
ratory depression are due to hypoxemia, and thus cerebral death has likely
already begun at the onset of the cardiac arrest, increasing the chances of
a poor neurological outcome.

58 ECPR and Resuscitative ECMO

Hypothermia and Drowning
In comparison to normothermic cardiac arrest cases, patient with hypo-
thermia may have extended resilience with low flow durations as long as six
hours.55 Overall outcomes among non-asphyxia related cases are positive,
especially if potassium is found in the normal range (survival >50%).56,57
One systematic review, including 237 hypothermic arrest cases, reported
31% favourable neurological outcomes. Male sex, asphyxia-related mech-
anism, greater age, higher potassium level, longer CPR duration, and
higher temperature were associated with worse outcomes.58 However
asphyxia-related OHCA may still benefit from ECPR. One study reported
24% survival among cases with hypothermia and asphyxia resulting from
water submission or avalanche burial.55 Another study of 77 ECPR-treated
drowning-related OHCA (36% had concomitant hypothermia) reported
a 23% survival.59 Thus, accidental hypothermia represents an excellent
indication for ECPR. Although less promising drowning or burial cases
should not be an absolute exclusion criterion, especially if the cardiac ar-
rest likely occurred prior to hypoxemia. ECMO-facilitated rewarming has
been recommended for temperatures <32˚C60; however, robust evidence
supporting this temperature threshold is unavailable.

Past Medical History

Patients with serious comorbidities or neurological deficits are typically
excluded from ECPR eligibility, and thus the potential benefit of ECPR in
these patient types is unclear. A past history of cardiovascular disease has
been shown to be associated with improved ECPR outcomes, while no as-
sociation has been detected with diabetes, chronic obstructive pulmonary
disease (COPD), asthma, chronic renal disease, cirrhosis, cancer, hyper-
tension, smoking, or immunocompromise.9,24,45,61 One study reported dys-
lipidemia to be associated with nonsurvival.38 Another large investigation
examined a step-wise approach to creating optimal exclusion criteria.
The authors found minimal difference in favorable outcomes (21.5% to
22.2%) when terminal illness, cancer, end stage liver disease, and preex-
isting severe cognitive impairment were excluded.9

Chapter 3 59
pH and Lactate
Systematic reviews have found that higher pH and lower lactate were both
associated with better outcomes of ECPR-treated cases.13,14,16 This is not
surprising, however thresholds for defining “non-survivable” metabolic
parameters are needed. There have been reports of patients surviving
with a lactate value of 19 and a pH value of 6.68.62 One study examining
52 ECPR-treated OHCAs found a 94% mortality when lactate was ≥13.4
One ECPR program restricts eligibility to those with intra-arrest lactate
values ≤18; while evidence supporting this threshold is lacking, the results
of the program are excellent.18 The authors’ own experience is that initial
pH and lactates may not be used as strict criteria for stopping ECPR im-
plementation as survivors with unmeasurably low pH on admission have
been observed. Thus, pH and lactate values are correlated with outcomes;
robust cut-off values to classify non-survival have not been identified;
however, overall positive outcomes have been demonstrated with a clini-
cal threshold of a lactate of ≤18.

Intermittent ROSC and Persistent VF

Non-sustained episodes of ROSC prior to ECPR initiation have been found
to be independently predictive of outcomes at hospital discharge.5,11,20
While “persistent” VF has been used for ECPR eligibility assessment,24
there are no data to indicate whether these individuals are better ECPR
candidates than those with initial rhythms of VF. Linz and colleagues
found no association between the number of defibrillations and survival
(median 2 defibrillations [IQR 0-5] for survivors and 2 [IQR 0-4] for
nonsurvivors) in 423 ECPR cases.9 Thus, intermittent episodes of ROSC
may identify appropriate ECPR candidates, superseding other prognostic
factors (such as unwitnessed arrest or no bystander CPR). In contrast,
refractory VF does not appear to confer superior prognosis compared to
simply an initial VF rhythm.

60 ECPR and Resuscitative ECMO

 RCA after 20 min of CPR with AED

- Signs of life during CPR at the time of decision

- Or Hypothermia (T<32
)
- Or Intoxication or general anesthesia
NO Yes Consider
ECPR*
NO No major comorbidity
Yes
Witness CA and immediate CPR (no flow<5 min )
Yes
NO
ASYSTOLIE Initial Rhythms
- VF-VT
- PEA

NO EtCO2 at ECPR decision >10 mmHg

Yes

NO ECPR in less than 60 min (acceptable 100 min)

- Age < 55 YO
NO - No Flow <30 min Yes
+ Hospital arrival <120 min after CA
Classical ALS Consider ECPR
Yes
Consider NHBD * ECPR ASAP but no maximal delay

Appendix B: Indications for Extracorporeal Cardiopulmonary resuscitation in Paris.

AED: Automatic External Defibrillator

ALS: Advance Life Support
CA: Cardiac Arrest
CPR: Cardio-pulmonary resuscitation
ECPR: Extracorporeal Cardio-pulmonary resuscitation
NHBD: Non Heart Beating Donation
PEA: Pules Electric Activity
TV: Tachycardia Ventricular
VF: Ventricular Fibrillation

Lamhaut et al. Resuscitation. 2017;117:109-117.

Figure 5. The Service d’Aide Médicale Urgente (SAMU) of Paris ECPR pro-
tocol. Used with permission.

Signs of Life During CPR

“Signs of life” during CPR may be helpful to identify those with neu-
rological viability who may be appropriate ECPR candidates. One
before-and-after study compared 156 ECPR-treated OHCAs between two
time periods.24 The ECPR inclusion criteria included typical components;
however, cases with “signs of life” (defined as “breathing efforts, gasp,
movements, pupils different from mydriasis”) were considered for ECPR

Chapter 3 61
regardless of other Utstein characteristics (Figure 5). In the second pe-
riod, 29/42 (78%) of cases had “signs of life,” suggesting that these cases
may have been prioritized over others. “Signs of life” during CPR was the
most potent independent predictor of good neurological outcome at hos-
pital discharge. No patient without “signs of life” survived.
One study examined clinical findings at hospital arrival among 52
OHCAs treated with ECPR (15% had favourable neurological outcomes
at hospital discharge).4 No patients had reactive pupils at hospital arrival.
Spontaneous breathing was present in 75% of those with favourable out-
comes compared to 11% with poor outcomes (p<0.05). Among 26 cases
with pupil diameter greater than 6, none survived. An additional inves-
tigation of 85 patients calculated the association of multiple data points
(including Utstein characteristics, pH, lactate, time intervals, and phys-
ical exam) with neurological recovery, and reported that only GCS >3 at
hospital arrival, pupil diameter at hospital arrival, and no-flow interval
were independently associated with outcomes.63 With regards to pupil-
lary findings at hospital arrival, 21% of those with favourable neurologic
outcomes had a pupillary reflex at arrival vs. 5.6% of those with poor out-
comes (p=0.049). The mean pupil diameter was 4.1 (SD 1.1) and 5.2 (SD
1.2; p=0.0023), respectively, making this a difficult distinguishing finding
due to the small difference and interrater variability.64,65
Movement, breathing, and reactive pupils during CPR may provide
excellent insight into those with intact neurological systems, but these
findings do not provide sufficient criteria to deem candidates ineligible.
These findings may identify appropriate ECPR candidates who otherwise
have less favourable phenotypes.

ETCO2 and PO2

One ECPR program used ETCO2 and PO2 values in eligibility assess-
ment.18 Although outcomes of this program are positive, the outcomes
of patients with values outside of used thresholds are unclear. Further
data are required to clarify the utility of these intra-arrest parameters in
candidacy assessment.

62 ECPR and Resuscitative ECMO

Institutional Volumes
Creating institutional ECPR eligibility criteria is complex. When estab-
lishing a program, one may want to restrict candidacy to those most likely
to survive, in order to maintain support and enthusiasm (see Chapter 1).
However, this will limit experience and skill development among provid-
ers within the prehospital, emergency department, and critical care areas.
Previous data has shown that volume of ECMO experience is correlated
with outcomes, which also appears to apply for ECPR.66,67 Thus, institu-
tional criteria may balance the requisite volume for institutional compe-
tency with the desire for an optimal proportion of positive outcomes.

Summary
Predefined, straightforward, and reproducible inclusion criteria are crit-
ical for uniform and rapid deployment of an ECPR protocol (example
criteria available in Figures 1,2,3, and 5). Available evidence supports pref-
erentially treating those under the age of 75 with either initial shockable
rhythms, signs of life during CPR, or intermittent ROSC. Whereas cases
with initial shockable rhythms have survived despite prolonged low-flow
intervals (i.e. up to and beyond 60 minutes), those with non-shockable
initial rhythms appear to have diminished resilience for prolonged peri-
ods of CPR. Unwitnessed arrest and no bystander CPR may be appropri-
ate exclusion criteria, unless evidence suggests that the no-flow duration
was short. Comorbidities and suspicion of etiology may be further meth-
ods of honing candidacy assessment.

References
1. Dennis M, Zmudzki F, Burns B, et al. Cost effectiveness and qual-
ity of life analysis of extracorporeal cardiopulmonary resuscitation
(ECPR) for refractory cardiac arrest. Resuscitation. 2019;139:49-56.
2. Le Guen M, Nicolas-Robin A, Carreira S, et al. Extracorporeal life
support following out-of-hospital refractory cardiac arrest. Crit
Care. 2011;15(1):R29.
3. Perkins GD, Jacobs IG, Nadkarni VM, et al. Cardiac Arrest and
Cardiopulmonary Resuscitation Outcome Reports: Update of the

Chapter 3 63
 Utstein Resuscitation Registry Templates for Out-of-Hospital
Cardiac Arrest: A Statement for Healthcare Professionals From a
Task Force of the International Liaison Committee. Resuscitation.
2015;96:328-340.
4. Maekawa K, Tanno K, Hase M, Mori K, Asai Y. Extracorporeal car-
diopulmonary resuscitation for patients with out-of-hospital cardiac
arrest of cardiac origin: a propensity-matched study and predictor
analysis. Crit Care Med. 2013;41(5):1186-1196.
5. Bougouin W, Dumas F, Lamhaut L, et al. Extracorporeal cardiopul-
monary resuscitation in out-of-hospital cardiac arrest: a registry
study. Eur Heart J. October 2019.
6. Conrad SA, Broman LM, Taccone FS, et al. The Extracorporeal
Life Support Organization Maastricht Treaty for Nomenclature
in Extracorporeal Life Support. A Position Paper of the Extracor-
poreal Life Support Organization. Am J Respir Crit Care Med.
2018;198(4):447-451.
7. Leick J, Liebetrau C, Szardien S, et al. Door-to-implantation
time of extracorporeal life support systems predicts mortality
in patients with out-of-hospital cardiac arrest. Clin Res Cardiol.
2013;102(9):661-669.
8. Hiemstra B, Bergman R, Absalom AR, et al. Long-term outcome of
elderly out-of-hospital cardiac arrest survivors as compared with
their younger counterparts and the general population. Ther Adv
Cardiovasc Dis. September 2018:1753944718792420.
9. Lunz D, Calabrò L, Belliato M, et al. Extracorporeal membrane
oxygenation for refractory cardiac arrest: a retrospective multicenter
study. Intensive Care Med. February 2020.
10. Haas NL, Coute RA, Hsu CH, Cranford JA, Neumar RW. Descriptive
analysis of extracorporeal cardiopulmonary resuscitation following
out-of-hospital cardiac arrest-An ELSO registry study. Resuscitation.
2017;119:56-62.
11. Goto T, Morita S, Kitamura T, et al. Impact of extracorporeal cardio-
pulmonary resuscitation on outcomes of elderly patients who had
out-of-hospital cardiac arrests: a single-centre retrospective analysis.
BMJ Open. 2018;8(5):e019811.
12. Yukawa T, Kashiura M, Sugiyama K, Tanabe T, Hamabe Y. Neurologi-
cal outcomes and duration from cardiac arrest to the initiation of ex-
tracorporeal membrane oxygenation in patients with out-of-hospital

64 ECPR and Resuscitative ECMO

 cardiac arrest: a retrospective study. Scand J Trauma Resusc Emerg
Med. 2017;25(1):95.
13. D’Arrigo S, Cacciola S, Dennis M, et al. Predictors of favourable out-
come after in-hospital cardiac arrest treated with extracorporeal car-
diopulmonary resuscitation: A systematic review and meta-analysis.
Resuscitation. 2017;121:62-70.
14. Debaty G, Babaz V, Durand M, et al. Prognostic factors for ex-
tracorporeal cardiopulmonary resuscitation recipients following
out-of-hospital refractory cardiac arrest. A systematic review and
meta-analysis. Resuscitation. 2017;112:1-10.
15. Yu H-Y, Wang C-H, Chi N-H, et al. Effect of interplay between
age and low-flow duration on neurologic outcomes of extra-
corporeal cardiopulmonary resuscitation. Intensive Care Med.
2019;45(1):44-54.
16. Wang J, Ma Q, Zhang H, Liu S, Zheng Y. Predictors of survival and
neurologic outcome for adults with extracorporeal cardiopulmonary
resuscitation: A systemic review and meta-analysis. Med (United
States). 2018;97(48).
17. Kim S, Jung J, Park J, Park J, Hong Y, Lee S. An optimal transition
time to extracorporeal cardiopulmonary resuscitation for predicting
good neurological outcome in patients with out-of-hospital cardiac
arrest: a propensity-matched study. Crit Care. 2014;18(5):535.
18. Yannopoulos D, Bartos JA, Raveendran G, et al. Coronary Artery Dis-
ease in Patients With Out-of-Hospital Refractory Ventricular Fibril-
lation Cardiac Arrest. J Am Coll Cardiol. 2017;70(9):1109-1117.
19. Otani T, Sawano H, Hayashi Y. Optimal extracorporeal cardiopul-
monary resuscitation inclusion criteria for favorable neurological
outcomes: a single-center retrospective analysis. Acute Med Surg.
2020;7(1):1-7.
20. Lee SW, Han KS, Park JS, Lee JS, Kim SJ. Prognostic indicators of
survival and survival prediction model following extracorporeal
cardiopulmonary resuscitation in patients with sudden refractory
cardiac arrest. Ann Intensive Care. 2017;7(1).
21. Spangenberg T, Meincke F, Brooks S, et al. “Shock and Go?” extra-
corporeal cardio-pulmonary resuscitation in the golden-hour of
ROSC. Catheter Cardiovasc Interv. 2016;88(5):691-696.

Chapter 3 65
22. Ha TS, Yang JH, Cho YH, et al. Clinical outcomes after rescue extra-
corporeal cardiopulmonary resuscitation for out-of-hospital cardiac
arrest. Emerg Med J. 2017;34(2):107-111.
23. Tanguay-Rioux X, Grunau B, Neumar R, Tallon J, Boone R, Chris-
tenson J. Is initial rhythm in OHCA a predictor of preceding no flow
time? Implications for bystander response and ECPR candidacy
evaluation. Resuscitation. 2018;128:88-92.
24. Lamhaut L, Hutin A, Puymirat E, et al. A Pre-Hospital Extracorpo-
real Cardio Pulmonary Resuscitation (ECPR) strategy for treatment
of refractory out hospital cardiac arrest: An observational study and
propensity analysis. Resuscitation. 2017;117:109-117.
25. Stub D, Bernard S, Pellegrino V, et al. Refractory cardiac arrest
treated with mechanical CPR, hypothermia, ECMO and early reper-
fusion (the CHEER trial). Resuscitation. 2015;86:88-94.
26. Grunau B, Carrier S, Bashir J, et al. A comprehensive regional
clinical and educational ECPR protocol decreases time to ECMO
in patients with refractory out-of-hospital cardiac arrest. CJEM.
2017;19(6):424-433.
27. Grunau B, Guy A, Kawano T, et al. The Association Between
No-Flow Interval and Neurological Outcomes in Out-Of-Hospital
Cardiac Arrest: Implications for Rescuer Response, Initiating
Resuscitation, and ECPR Candidacy Evaluation. Resuscitation.
2019;140:A235.
28. Otani T, Sawano H, Natsukawa T, et al. Low-flow time is associated
with a favorable neurological outcome in out-of-hospital cardiac
arrest patients resuscitated with extracorporeal cardiopulmonary
resuscitation. J Crit Care. 2018;48:15-20.
29. Wengenmayer T, Rombach S, Ramshorn F, et al. Influence of
low-flow time on survival after extracorporeal cardiopulmonary
resuscitation (eCPR). Crit Care. 2017;21(1):157.
30. Yu H-Y, Wang C-H, Chi N-H, et al. Effect of interplay between
age and low-flow duration on neurologic outcomes of extra-
corporeal cardiopulmonary resuscitation. Intensive Care Med.
2019;45(1):44-54.
31. Ko R-E, Ryu J-A, Cho YH, et al. The differential neurologic prognosis
of low-flow time according to the initial rhythm in patients who un-
dergo extracorporeal cardiopulmonary resuscitation. Resuscitation.
2020;148(4):121-127.

66 ECPR and Resuscitative ECMO

32. Komeyama S, Takagi K, Tsuboi H, et al. The Early Initiation of Ex-
tracorporeal Life Support May Improve the Neurological Outcome
in Adults with Cardiac Arrest due to Cardiac Events. Intern Med.
2019;58(10):1391-1397.
33. Chen Y-S, Chao A, Yu H-Y, et al. Analysis and results of prolonged
resuscitation in cardiac arrest patients rescued by extracorporeal
membrane oxygenation. J Am Coll Cardiol. 2003;41(2):197-203.
34. Grunau B, Carrier S, Bashir J, et al. A comprehensive regional
clinical and educational ECPR protocol decreases time to ECMO in
patients with refractory out-of-hospital cardiac arrest. Can J Emerg
Med. 2017;19(6).
35. Pozzi M, Koffel C, Armoiry X, et al. Extracorporeal life support for
refractory out-of-hospital cardiac arrest: Should we still fight for? A
single-centre, 5-year experience. Int J Cardiol. 2016;204:70-76.
36. Wang C-H, Chou N-K, Becker LB, et al. Improved outcome of
extracorporeal cardiopulmonary resuscitation for out-of-hospital
cardiac arrest - A comparison with that for extracorporeal rescue for
in-hospital cardiac arrest. Resuscitation. 2014;85(9):1219-1224.
37. Avalli L, Maggioni E, Formica F, et al. Favourable survival of
in-hospital compared to out-of-hospital refractory cardiac arrest pa-
tients treated with extracorporeal membrane oxygenation: an Italian
tertiary care centre experience. Resuscitation. 2012;83(5):579-583.
38. Kagawa E, Inoue I, Kawagoe T, et al. Assessment of outcomes and
differences between in- and out-of-hospital cardiac arrest patients
treated with cardiopulmonary resuscitation using extracorporeal life
support. Resuscitation. 2010;81(8):968-973.
39. Duerschmied D, Zotzmann V, Rieder M, et al. Myocardial infarc-
tion type 1 is frequent in refractory out-of-hospital cardiac arrest
(OHCA) treated with extracorporeal cardiopulmonary resuscitation
(ECPR). Sci Rep. 2020;10(1):8423.
40. Lamhaut L, Tea V, Raphalen J-H, et al. Coronary lesions in refractory
out of hospital cardiac arrest (OHCA) treated by extra corporeal
pulmonary resuscitation (ECPR). Resuscitation. 2018;126:154-159.
41. Rosengren A, Wallentin L, Simoons M, et al. Age, clinical presenta-
tion, and outcome of acute coronary syndromes in the Euroheart
acute coronary syndrome survey. Eur Heart J. 2006;27(7):789-795.
42. Mandigers L, Scholten E, Rietdijk WJR, et al. Survival and neurolog-
ical outcome with extracorporeal cardiopulmonary resuscitation for

Chapter 3 67
 refractory cardiac arrest caused by massive pulmonary embolism:
A two center observational study. Resuscitation. 2019;136(October
2018):8-13.
43. Miyazaki K, Hikone M, Kuwahara Y, Ishida T, Sugiyama K, Ham-
abe Y. Extracorporeal CPR for massive pulmonary embolism
in a “hybrid 2136 emergency department”. Am J Emerg Med.
2019;37(12):2132-2135.
44. Meneveau N, Guillon B, Planquette B, et al. Outcomes after extra-
corporeal membrane oxygenation for the treatment of high-risk
pulmonary embolism: A multicentre series of 52 cases. Eur Heart J.
2018;39(47):4196-4204.
45. Beurtheret S, Mordant P, Paoletti X, et al. Emergency circulatory
support in refractory cardiogenic shock patients in remote institu-
tions: a pilot study (the cardiac-RESCUE program). Eur Heart J.
2013;34(2):112-120.
46. Lorusso R, Centofanti P, Gelsomino S, et al. Venoarterial Extracor-
poreal Membrane Oxygenation for Acute Fulminant Myocarditis in
Adult Patients: A 5-Year Multi-Institutional Experience. Ann Thorac
Surg. 2016;101(3):919-926.
47. Conrad SJ, Bridges BC, Kalra Y, Pietsch JB, Smith AH. Extracorpo-
real Cardiopulmonary Resuscitation Among Patients with Structur-
ally Normal Hearts. ASAIO J. 2017;63(6):781-786.
48. Lamhaut L, Lecomte L. A Comparative Study Between a Pre-hospital
and an In-hospital Circulatory Support Strategy (ECMO) in Refrac-
tory Cardiac Arrest. ClinicalTrials.gov. 2017.
49. Kolcz J, Pietrzyk J, Januszewska K, Procelewska M, Mroczek T,
Malec E. Extracorporeal life support in severe propranolol and ver-
apamil intoxication. J Intensive Care Med. 22(6):381-385.
50. Holzer M, Sterz F, Schoerkhuber W, et al. Successful resuscitation of
a verapamil-intoxicated patient with percutaneous cardiopulmonary
bypass. Crit Care Med. 1999;27(12):2818-2823.
51. Farag M, Badowski D, Koschny R, Skopp G, Brcic A, Szabo GB. Ex-
tracorporeal life support and digoxin-specific Fab fragments for suc-
cessful management of Taxus baccata intoxication with low output
and ventricular arrhythmia. Am J Emerg Med. 2017;35(12):1987.
e3-1987.e7.

68 ECPR and Resuscitative ECMO

52. Masson R, Colas V, Parienti J-J, et al. A comparison of survival with
and without extracorporeal life support treatment for severe poison-
ing due to drug intoxication. Resuscitation. 2012;83(11):1413-1417.
53. Wang G, Levitan R, Wiegand T, Lowry J, RF S, Yin S. Extracorporeal
Membrane Oxygenation (ECMO) for Severe Toxicological Expo-
sures: Review of the Toxicology Investigators Consortium (ToxIC).
J MEd Toxicol. 2016;12(1):95-99.
54. Cole JB, Olives TD, Ulici A, et al. Extracorporeal Membrane Oxy-
genation for Poisonings Reported to U.S. Poison Centers from 2000
to 2018: An Analysis of the National Poison Data System. Crit Care
Med. 2020;May 12, 20:1-9.
55. Hilmo J, Naesheim T, Gilbert M. “ Nobody is dead until warm and
dead ”: Prolonged resuscitation is warranted in arrested hypother-
mic victims also in remote areas – A retrospective study from north-
ern Norway. Resuscitation. 2014;85(9):1204-1211.
56. Pasquier M, Rousson V, Darocha T, et al. Hypothermia outcome
prediction after extracorporeal life support for hypothermic cardiac
arrest patients: An external validation of the HOPE score. Resuscita-
tion. 2019;139(January):321-328.
57. Saczkowski RS, Brown DJA, Abu-Laban RB, Fradet G, Schulze CJ,
Kuzak ND. Prediction and risk stratification of survival in accidental
hypothermia requiring extracorporeal life support: An individual pa-
tient data meta-analysis. Resuscitation. 2018;127(February):51-57.
58. Pasquier M, Hugli O, Paal P, et al. Hypothermia outcome prediction
after extracorporeal life support for hypothermic cardiac arrest pa-
tients: The HOPE score. Resuscitation. 2018;126:58-64.
59. Burke CR, Chan T, Brogan T V., et al. Extracorporeal life support for
victims of drowning. Resuscitation. 2016;104:19-23.
60. Brown DJA, Brugger H, Boyd J, Paal P. Accidental hypothermia. N
Engl J Med. 2012;367(20):1930-1938.
61. Ryu JA, Cho YH, Sung K, Choi SH, Yang JH, Choi JH, Lee DS, Yang
JH. Predictors of neurological outcomes after successful extracor-
poreal cardiopulmonary resuscitation. BMC Anesthesiol. 2015 Mar
8;15:26.
62. Miyazaki K, Hikone M, Kuwahara Y, Ishida T, Sugiyama K, Ham-
abe Y. Extracorporeal CPR for massive pulmonary embolism
in a “hybrid 2136 emergency department”. Am J Emerg Med.
2019;37(12):2132-2135.

Chapter 3 69
63. Murakami N, Kokubu N, Nagano N, et al. Prognostic Impact of
No-Flow Time on 30-Day Neurological Outcomes in Patients With
Out-of-Hospital Cardiac Arrest Who Received Extracorporeal Car-
diopulmonary Resuscitation. Circ J. June 2020.
64. Olson DM, Stutzman S, Saju C, Wilson M, Zhao W, Aiyagari V.
Interrater Reliability of Pupillary Assessments. Neurocrit Care.
2016;24(2):251-257.
65. Couret D, Boumaza D, Grisotto C, et al. Reliability of standard
pupillometry practice in neurocritical care: an observational,
double-blinded study. Crit Care. 2016;20(1):99.
66. Belohlavek J, Smid O, Franek O, Kolouch P. Hyperinvasive approach
prolongs the time window for favorable outcomes in refractory
out-of-hospital cardiac arrest: A preliminary analysis of the “Prague
OHCA Study.” Resuscitation. 2016;106(Suppl. 1):e18.
67. Bartos JA, Carlson K, Carlson C, et al. Surviving refractory
out-of-hospital ventricular fibrillation cardiac arrest: Critical care
and extracorporeal membrane oxygenation management. Resuscita-
tion. 2018;132:47-55.
68. Sakamoto T, Morimura N, Nagao K, et al. Extracorporeal cardiopul-
monary resuscitation versus conventional cardiopulmonary resus-
citation in adults with out-of-hospital cardiac arrest: A prospective
observational study. Resuscitation. February 2014:1-7.
69. Chen Y-S, Lin J-W, Yu H-Y, 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;372(9638):554-561.
70. Fagnoul D, Taccone FS, Belhaj A, et al. Extracorporeal life support
associated with hypothermia and normoxemia in refractory cardiac
arrest. Resuscitation. 2013;84(11):1519-1524.
71. Shinar Z, Plantmason L, Reynolds J, Dembitsky W, Bellezzo J, Ho
C, Glaser D, Adamson R. Emergency Physician-Initiated Resusci-
tative Extracorporeal Membrane Oxygenation. J Emerg Med. 2019
Jun;56(6):666-673.
72. Belohlavek J, Kucera K, Jarkovsky J, et al. Hyperinvasive approach
to out-of hospital cardiac arrest using mechanical chest com-
pression device, prehospital intraarrest cooling, extracorporeal
life support and early invasive assessment compared to standard

70 ECPR and Resuscitative ECMO

 of care. A randomized parallel groups comparative study pro-
posal. “Prague OHCA study”. J Transl Med. 2012 Aug 10;10:163.
10.1186/1479-5876-10-163.
73. Yannopoulos D, Bartos J, Raveendran G, Walser E, Connett J,
Murray TA, Collins G, Zhang L, Kalra R, Kosmopoulos M, John R,
Shaffer A, Frascone RJ, Wesley K, Conterato M, Biros M, Tolar J,
Aufderheide TP. Advanced reperfusion strategies for patients with
out-of-hospital cardiac arrest and refractory ventricular fibrillation
(ARREST): a phase 2, single centre, open-label, randomised con-
trolled trial. Lancet. 2020 Nov 12:S0140-6736(20)32338-2.
74. Bol ME, Suverein MM, Lorusso R, et al. Early initiation of extra-
corporeal life support in refractory out-of-hospital cardiac ar-
rest: Design and rationale of the INCEPTION trial. Am Heart J.
2019;210:58-68.
75. Patricio D, Peluso L, Brasseur A, Lheureux O, Belliato M, Vincent JL,
Creteur J, Taccone FS. Comparison of extracorporeal and conven-
tional cardiopulmonary resuscitation: a retrospective propensity
score matched study. Crit Care. 2019 Jan 28;23(1):27.
76. Bougouin W, Dumas F, Lamhaut L, Marijon E, Carli P, Combes
A, Pirracchio R, Aissaoui N, Karam N, Deye N, Sideris G, Began-
ton F, Jost D, Cariou A, Jouven X; Sudden Death Expertise Center
investigators. Extracorporeal cardiopulmonary resuscitation in
out-of-hospital cardiac arrest: a registry study. Eur Heart J. 2020
Jun 1;41(21):1961-1971.

Chapter 3 71
CHAPTER 4
Running the ECPR Code

Joseph Bellezzo

Introduction
Traditionally, the logistics of performing cardiopulmonary resusci-
tation (CPR) were directed at optimizing performance of advanced
cardiac life support (ACLS) protocols. While the focus of conventional
CPR via ACLS is the establishment of return of spontaneous circulation
(ROSC), the end goal of extracorporeal cardiopulmonary resuscitation
(ECPR) is rapid establishment of cardiopulmonary bypass. This is typ-
ically performed via the femoral vasculature and can contribute to sus-
tained ROSC. Thus, the positioning of equipment and personnel within
the resuscitation room, the choreography of the ECPR code, and the focus
of efforts necessarily differ from the traditional code.

Foundational Terms
ɋ Nurse Code Team Leader – key leader of nurse lead code. They
are responsible for coordinating the resuscitation, pulse checks,
chest compression quality, and algorithmic medication orders
ɋ Wire Assistant – health care assistant who aids in wire control,
dilator selection/loading, and racking the wire to prevent kinking
ɋ ECMO specialist—a perfusionist, nurse, respiratory therapist,
or physician who is in charge of priming and managing the ECMO
circuit

 73
 ACLS Sub-Team
ɋ Nurse Code Team Leader
• Nurse-led code: ensure high quality CPR
• Keep ACLS code time
• Documentation
ɋ Physician Code Team Leader
• Advanced airway management
• Supervision of ACLS
• Determine ECPR eligibility
ɋ Respiratory Therapist
• Assist Physician Code Team Leader with airway
• Ventilator management
ɋ Chest Compressor
• Perform human chest compressions
• Assist with placement of mechanical chest compression
device
• Manage mechanical chest compression device
ɋ Med/Electric (M/E) Nurse
• Ensure adequate IV access
• Administration of ACLS medications
• Defibrillation
ɋ Pharmacist
• Prepare medications

ECPR Sub-Team
ɋ Cannulator 1/Cannulator 2 (Optional)
• Cannulation of femoral vessels
ɋ Wire Assistant (W/A)
• Assist with cannulation
ɋ ECMO Specialist
• Pump and circuit maintenance
• Priming and deairing the circuit
• Troubleshooting
ɋ Assistant /Runner
• Provide extra equipment

74 ECPR and Resuscitative ECMO

 Since most practitioners are right-handed, the typical orientation is to
focus cannulation on the right femoral vessels. The goal is to create the op-
timal environment to allow quick, efficient, and successful cannulation of
the femoral vasculature, while traditional CPR continues. All other aspects
of the ECPR resuscitation revolve around the person performing the can-
nulation. For these reasons the components of ECPR include specific per-
sonnel, appropriate resuscitation-room setup, the code choreography of ECPR,
and the specific steps, or stages, involved in establishing extracorporeal cir-
culatory support. Each of these components are discussed in detail.
There are two primary means by which an arresting patient presents
to the healthcare system: 1) pre-hospital emergency medical services
(EMS) provides advanced notification of the pending arrival of a patient
in cardiopulmonary arrest; or 2) the patient arrests in the hospital. In the
first case, advanced notice by EMS provides the ECPR team the opportu-
nity to prepare for their arrival and, with some exceptions, the resuscita-
tion is performed in the emergency department (ED). In the second case,
the code is either run in the location that the arrest occurs, or the patient
is moved to a designated ECPR suite within the hospital.

Personnel
ECPR often occurs in tertiary care institutions and is performed by an
ECMO team consisting of health care professionals from the ED, inten-
sive care, cardiology, and/or surgery, in addition to specialized nurses, re-
spiratory therapists, and/or perfusionists.
The ECPR resuscitation team should be divided into two separate
and independently functioning sub-teams: 1) the ACLS sub-team, which
performs traditional ACLS and, 2) the ECPR sub-team, which is respon-
sible for cannulation and establishment of extracorporeal support. Hav-
ing members of either sub-team turn attention to the actions of the other
team is strongly discouraged and may result in decreased effectiveness of
both sub-teams. The two sub-teams are typically divided, assigned roles,
and positioned in the resuscitation room as shown in Figure 1.
Some ECMO centers have found it helpful to establish laminated role
cards which outline positioning of each team member, as well as their
roles and responsibilities during ECPR. These role cards help each team

Chapter 4 75
Figure 1: ECMO: ECMO operator; W/A: wire assistant, U/S: ultrasound
machine; CPR: person assigned to perform hands-on chest compressions;
MD: physician code team leader; RT: respiratory therapist; Vent: ventilator;
M/E nurse: nurse assigned to deliver medications and defibrillation; cow:
computer on wheels; NCTL: Nurse Code Team Leader; Pharm: Pharmacist;
Art: arterial catheter; Vein: Venous catheter.

member focus on their specific responsibilities while avoiding tasks as-

signed to others.
Some centers do not have the luxury of sufficient human resources
to divert for two separate teams. Recognizing the potential pitfalls and
limitations will smaller teams is paramount.

76 ECPR and Resuscitative ECMO

Resuscitation Room Setup
The location where ECPR is performed within the hospital varies depend-
ing on each ECMO center and whether the event occurred as an OHCA
or IHCA. As discussed in Chapter 1, successful ECPR centers have histor-
ically cannulated in the intensive care unit, cardiac catheterization labo-
ratory or operating room by intensivists, cardiologists or cardiothoracic
surgeons, respectively. Over the past decade, successful ECPR by emer-
gency physicians has evolved.8,9 ECPR for OHCA may be performed in al-
most any location within the hospital depending on hospital standards,
provider expertise, equipment availability, and staff experience. Once the
site of ECPR is established, proper room preparation is necessary.
Appropriate setup of the resuscitation room prior to the arrival of a
patient who has suffered OHCA arrest enables a controlled and stepwise
approach to transferring the patient from the ambulance gurney to the hos-
pital gurney. This allows the ECPR sub-team to maintain sterile technique,
facilitates performance of high-quality CPR by the ACLS sub-team, result-
ing in a focused and controlled cannulation process. Figure 1 illustrates
one model for positioning of individual team members and equipment.
Resuscitation areas often have significant size constraints. During ac-
tive ECPR moving equipment into and out of the resuscitation room is
not logistically possible. Providers should anticipate the need for neces-
sary equipment, bring in only the equipment mandatory for the resusci-
tation and remove all unnecessary items from the room. For example, a
simple item like a step stool, needed by the provider performing manual
compressions, should be staged near the rear of the room prior to place-
ment of other large equipment.
For the following discussion, we will assume that the patient has suf-
fered an OHCA and EMS has notified the resuscitation team of the im-
pending arrival of the patient by ambulance with personnel placement
depicted in Figure 1. Outside the confines of the resuscitation room are
the extracorporeal membrane oxygenation (ECMO) machine operators,
a pharmacist when available, and extra personnel to provide general assis-
tance. The person performing cannulation is positioned adjacent to an ul-
trasound machine used for visualization of the femoral vessels to support
percutaneous femoral vascular access during CPR. The cannulator(s)
should be gowned in personal protective equipment that meets their in-
stitutional standards and prepare for sterile vascular access, ideally prior

Chapter 4 77
to patient arrival. The ED gurney is positioned as close to the cannulator
as possible in order to provide ample space for the incoming ambulance
gurney to enter the room on the contralateral side.
Anticipating use of the 3-stage approach to ECPR, which is described
in detail below, two durable mayo stands (or other table or trolley) are
positioned at the foot of the hospital gurney and serve two purposes: 1)
to support the vascular angiocatheters that will be initially placed during
Stage 1 of ECPR, and 2) to provide a durable surface for sterile drapes that
are placed during the ECMO cannulation process of Stage 2, as well as to
support the cannula/circuit combination that is connected during Stage 3.

Code Choreography (Logistics of ECPR)

Preparing the resuscitation room as outlined above serves two main pur-
poses: 1) it offers the cannulator the optimal position for successful can-
nulation, and 2) it allows the ACLS team to continue ACLS, including high
quality chest compressions until extracorporeal support is established.
As the ambulance gurney enters the resuscitation room adjacent to
the hospital gurney, all unnecessary EMS transport equipment (cardiac
monitor, ECG leads, belt buckles/straps, etc.) are removed from the pa-
tient. Human chest compressions are continued by EMS personnel un-
til the patient is moved to the hospital gurney. Using either a traditional
backboard, a slider board, or other transfer device, the patient is quickly
moved from the ambulance gurney to the hospital gurney. The ACLS
sub-team member assigned to perform hands-on chest compressions
should immediately resume chest compressions, minimizing no compres-
sion time. The ambulance gurney is removed from the room.
The Physician Code Team Leader is responsible for advanced airway
management, code oversight, and determining whether the patient meets
institutional ECPR inclusion criteria. If the patient does not yet have IV ac-
cess, consider immediate placement of intraosseous access until definitive
IV access can be established. If a separate cannulator is available, the Phy-
sician Code Team Leader should resist the urge to become involved with
the cannulation process. Their attention, when diverted by a procedural
focus, can delay essential therapies and create less sound decision making.
Simultaneously, an assigned team member should immediately disrobe
the patient. Clothing may need to be cut off. The groin region(s) should be

78 ECPR and Resuscitative ECMO

sterilely prepped and draped for immediate percutaneous femoral vessel
access. Use of long sterile drapes that cover the inguinal region all the way
past the feet onto the mayo stands abutting the bed are ideal.
If not already placed, the ACLS sub-team should consider transition
to a mechanical chest compression device, when available, and placement
should occur during a scheduled ACLS rhythm check. While efficacy stud-
ies on mechanical chest compression devices have been thus far incon-
clusive,7 the use of such devices is thought to be beneficial to the ECPR
process.8 Two mechanisms account for this perceived benefit: 1) patients
who are resuscitated with ECPR have longer low-flow downtimes than
traditional CPR cases, and mechanical chest compression devices allow
for consistent, high quality of CPR over a prolonged period of time, and
2) mechanical chest compression devices transmit less physical force into
the inguinal region, where cannulation is occurring, than hands-on chest
compressions. Given the multiple simultaneous tasks that are required
during an ECPR code, dedicating one provider to oversee the quality of
the traditional resuscitation is imperative. A dedicated nurse code team
leader has been used successfully for this role in numerous hospitals. This
provider is responsible for dictating the commencement, cessation, and
quality of chest compressions, epinephrine timing, and defibrillation.
All other components of ACLS should continue under the supervision
of the Nurse and Physician Code Team Leaders of the ACLS sub-team.
Given the often-prolonged resuscitation efforts that occurred pre-hospital,
ACLS failure is commonly assumed and extracorporeal support is consid-
ered vital. Therefore, interventions should be focused on optimizing chest
compressions and facilitating successful cannulation for rapid establish-
ment of ECMO. See Chapter 2 for additional details regarding optimizing
the pre-ECMO resuscitation.
If the resuscitation team has a mechanical chest compression device
available consider transferring the patient, from EMS gurney to hospital
gurney, prior to entry into the resuscitation room. The ambulance bay, a
hallway, or other physical location may provide a more opportune place
to perform this maneuver than in the resuscitation room where there are
space constraints, noise pollution, and extensive personnel present.
The ECPR sub-team then begins ECMO initiation using a 3-Stage ap-
proach.

Chapter 4 79
The Three Stages of CPR
“A Journey of a thousand miles begins with a single
step” -Laozi, Daoist philosopher

The Japanese philosophy of Kaizen, often embraced in the business

world, focuses on deliberate, continuous, integrated steps toward achiev-
ing large-goal success. The concept of ‘baby steps’ is embraced as a means
of breaking down complex problems into smaller, achievable tasks. To ask
a resuscitationist to “run the ECPR code” may seem like a daunting task
en masse. The Kaizen approach is applied here to break the ECPR process
into smaller, achievable ‘baby steps,’ which are herein referred to as the
3-Stages of ECPR.
Time is of the essence and information is imperfect when an arrested
patient arrives to the ED with CPR in progress. When patients present to
the ED in the peri-arrest phases of cardiovascular collapse, decisive action
by the treating physician can make the difference between life and death.
Critical information necessary to determine whether ECPR is an appro-
priate intervention is often initially unavailable or becomes available only
in a piecemeal fashion. At the same time, successful neurologic outcomes
after arrest are inversely associated with the time it takes to reestablish
brain perfusion (either intrinsic perfusion with ROSC or extracorporeal
perfusion with ECMO).9,10 Minimizing the time from patient arrival to
successful cannulation is critical to a neurologically intact recovery.6
The 3-Stage approach to ECPR was developed to address these issues
and is shown in Figure 2.

ECPR Stage 1
Stage 1 of ECPR involves percutaneous placement of femoral arterial and
venous angiocatheters. The choice of which commercially available cathe-
ters to use is institution-specific but typically involves use of a small-bore
micropuncture needle and Seldinger-guided sheath access to the femoral
artery and larger bore sheath placement into the femoral vein. Both are
ideally performed by ultrasound guidance. These catheters serve as con-
duits for placement of the ECMO wires during Stage 2; however, they also
serve as immediate vascular access resources. Intra-arrest arterial trans-
duction is a valuable resource and should be utilized whenever possible in

80 ECPR and Resuscitative ECMO

Figure 2: ECMO: Extracorporeal Life Support; ROSC: Return of Sponta-
neous Circulation.

the management of the cardiac arrest patient. Transduction allows for a

good metric of CPR quality with higher mean arterial pressure indicating
higher quality chest compressions. Additionally, as the arrest proceeds,
compliance of the chest changes making the ideal vector of compression
dynamic. Broken ribs can alter the ideal compression direction and the ar-
terial line can give clues to these changes. The arterial line also allows for
the diagnosis of pseudo-PEA which may benefit from vasopressors over
chest compressions. Finally, the arterial line helps predict a re-arrest in
patients who have had ROSC after the short half-life of epinephrine has
elapsed. Successful placement of catheters in both the femoral artery and
femoral vein mark the completion of Stage 1 of ECPR. If ROSC has not
been established, and the patient meets established inclusion criteria, the
cannulator moves on to Stage 2 of ECPR.

ECPR Stage 2
Stage 2 of ECPR is defined by exchange of the original femoral catheters
placed in Stage 1 with larger ECMO cannulas. While chest compres-
sions (either hands-on or mechanical) continue, no further attempts at
defibrillation are recommended at this point. Given the time needed to
reach this point of the resuscitation, it is assumed that the patient has

Chapter 4 81
 A B

Figure 3. Image of resuscitation cart (A) and inventory list (B).

not responded to prior attempts at defibrillation and further attempts

are futile. Further, defibrillation attempts interrupt the time-critical pro-
cess of ECPR cannulation. Details of the cannulation process (catheter
selection, wires used, and specific methods for femoral vascular access)
are discussed in depth in Chapters 5 and 6. But in general terms, a guide
wire is placed through the conduit catheters in the femoral vessels and
serial dilation is performed using commercially available vascular dilators.
ECMO cannulas are then placed over the guidewire, in the respective fem-
oral blood vessels. If the system is not pre-primed, then priming occurs
simultaneously with the dilation process. Anticoagulation (i.e. heparin)
should be considered. Successful placement of ECMO cannulas and prim-
ing of the ECMO circuit completes Stage 2 of ECPR. If ROSC has not been
achieved and the patient still meets inclusion criteria, the team moves to
Stage 3 of ECPR.

82 ECPR and Resuscitative ECMO

The ECMO Cart
Stage 1 of ECPR, in its simplest description, is the simultaneous perfor-
mance of ACLS by the ACLS sub-team and placement of any commer-
cially available vascular catheters in the arterial and venous systems by
the ECPR sub-team. Stage 1 is a frequent event while Stages 2 and beyond
occur less frequently. Therefore, items needed for Stage 1 are kept in a
separate cart from items needed for later stages of ECPR since turnover
of items used in Stage 1 is more frequent.
Figure 3 shows a resuscitation cart with an itemized list of its con-
tents, which contains items used for Stage 1.
Figure 4 shows a typical ECMO cart with an itemized list of its con-
tents. While versions of a functional ECMO cart abound, the typical cart
should contain all items needed to transition from Stage 1 (placement of
vascular catheters in the venous and arterial systems) to Stage 2 (dilation
and placement of ECMO cannulas) and beyond.

ECPR Stage 3
Stage 3 of ECPR involves connecting the ECMO cannulas to the extracor-
poreal circuit using the ‘wet-to-wet’ technique, described in Chapter 8,
and initiation of extracorporeal pump blood flow. At this point, the ACLS
sub-team should discontinue chest compressions.
Optimizing pump settings and parameters during the initiation phase
(circuit blood flow rate, sweep gas flow rate, and initial circuit main-
tenance) are discussed in Chapter 8. A blender can be used to set the
fraction of delivered oxygen percent (FdO2). Management of the patient
during the first hour (blood pressure, tissue perfusion, ventilator man-
agement, and patient maintenance) is discussed in Chapter 9 while cir-
cuit catastrophes and troubleshooting are discussed in Chapter 10.

Summary
Running a successful ECPR code depends largely on proper planning and
logistics. When considering the use of ECPR in the resuscitative interven-
tion arsenal, the ECPR team should establish protocols and standards that
have proven effective in optimizing procedural efficiency.

Chapter 4 83
Figure 4a. Image of ECMO cart.

84 ECPR and Resuscitative ECMO

 ECMO Equipment Check List Month/Year:
Contents of ECMO and/or Open Sternotomy Bins must be verified by RN when replaced by CSR, then
sealed with tape.
Date
Items AM PM Items AM PM
SICU Suture Box 10‐ Cap Bouffant Large (3274)
Sterile Bowl 1 Box‐ Mask Fluidgard (15310)
Normal Saline 500cc 2‐ Towels Sterile 4 pack
ECMO Cutdown Tray **sterilization 5‐ Gown Sterile Large (95111)
through SPD** 2‐20F Aortic cannula & 2‐ 28F DLP Venous
Ioban Drapes 60x45in cannula‐ taped to the top of the “Open
1‐ DuraPrep Sternotomy CPS Pack”
12‐ Alcohol Prep Pads Open Sternotomy Tray w Wire Cutters
6‐ 18 G Needles (305196) Stryker Head Light
6‐ Medication Labels (N‐200) 4‐ Sorin Dilator Kits (Vascular Access Kit)
4‐ Syringe 10cc 2‐ Amplatz Super Stiff J‐tip 0.035” 180cm
4‐ Syringe 60cc Fixed Core Straight 0.038” 145cm
3‐ Stopcock 3way (AG7986) Fixed Core J wire 0.035” 145cm
2” Foam Tape Guidewire 0.018” 300cm (1003282)
10‐ Dead End Caps (B6006) 2‐ Maquet Arterial Catheter 15 Fr
Tube Straight Connector (3510S) 2‐ Maquet Arterial Catheter 17 Fr
Disposable Scalpel 2‐ Maquet Arterial Catheter 19 Fr
2‐ M/M Arterial Connector Tubing 2 ea Avalon 27F / 31F
4‐ Suture Silk 0 SH 30in (GS834) 2‐ Cook Micropuncture Sets
4‐ Suture Silk 0 KS 30in (SS624) 1‐ Site Rite Needle Kit
2‐ Thin Wall Perc Needles 5 Fr sheath with wire
2‐ Protexis Glove Sz 6.5 Latex Free 6 Fr sheath with wire
2‐ Protexis Glove Sz 7 Latex Free Super Arrow‐Flex Perc Sheath 7F Introducer
4‐ Protexis Glove Sz 7.5 Latex Free Super Arrow‐Flex Perc Sheath 8F Introducer
2‐ Protexis Glove Sz 8 Latex Free Coons Taper Dilator 2‐20F, 2‐18F,
4‐ Sponge 4X4 10s (6939) 1‐16F, 2‐14F/20 cm (GO3947)
3‐ Lap Sponges 2‐ Venous Catheter 21 Fr
2‐ Cath Grip Anchors 2‐ Venous Catheter 23 Fr
Betadine / Chlorhexadine Bottle 2‐ Venous Catheter 25 Fr
2‐ Prep Chloraprep 26ml (260815) 1‐ ECMO Circuit (plus one set up)
“ECMO MD Instruments” box Avalon BiCaval 27x2 & 31x2
**sterilization through SPD** ECMO Cart in Knuckle Room
Heparin Box (locked) 1‐ Tube of Flow Probe Paste
4‐ Plasma‐Lyte 1000ml (L7070) 2‐ Full Oxygen Tanks (2 on cart)
Drape Angio Femoral (89704) 4‐ Tubing Clamps w/ Guard on each ECMO
2‐ Drape Steri 43X59in (9072) Machine
ECMO straight tubing 1‐ Non‐Disposable Oxygenator Clamps
Instructions/Perfusion Records/Data Sheets
RotoFlow Hand crank on each machine
Check Battery Charge
Initials Indicating Supplies are Complete

Figure 4b. ECMO inventory list.

Chapter 4 85
 Recognizing that the end goal of ECPR is not necessarily ROSC, which
has already proven unsuccessful by traditional resuscitation measures, the
aim in ECPR, then, is the establishment of extracorporeal circulation to
perfuse the brain and vital organs, hopefully providing the time necessary
for clinicians to determine the cause of the arrest and correct it. Success-
ful salvage therefore depends on rapid cannulation of the femoral vessels,
followed by initiation of cardiopulmonary bypass, which often improves
the likelihood of neurological-intact survival after cardiac arrest.
Selection of appropriate ECPR personnel, equipment, and training is
the first step in a good ECPR plan. Proper room setup, ECPR code cho-
reography, and cannulation technique are important components of the
ECPR process. Cannulation is the key procedure in the ECPR process and
can be challenging and time consuming. In order to allow the cannulation
process to be performed in parallel to, and not in sequence with the ACLS
process, the 3-stage approach to cannulation was established to allow both
processes to happen simultaneously. The end goal in running the ECPR
code is the rapid successful initiation of ECMO in the patient who has
suffered cardiopulmonary arrest and has not achieved ROSC in response
to ACLS efforts.

References
1. Berdowski J, Berg RA, Tijssen JG, Koster RW. Global incidences of
out-of-hospital cardiac arrest and survival rate: systemic review of 67
prospective studies. Resuscitation. 2010; 81 (11): 1479-1487
2. Nolan JP, Lyon RM, Sasson C, et al. Advances in the hospital man-
agement of patients following an out of hospital cardiac arrest.
Heart. 2012;367(20): 1201-1206
3. Wampler DA, Collet L, Manifold CA, Velasquez C, McMullan JUT.
Cardiac arrest survival is rare without prehospital return of sponta-
neous circulation. Prehosp Emerg Care. 2012; 16:451-455
4. Brooks SC, Anderson ML, Bruder E, et al. Alternative techniques and
ancillary devices for cardiopulmonary resuscitation: 2015 American
Heart Association guidelines update for cardiopulmonary resuscita-
tion and emergency cardiovascular care, part 6. Circulation. 2015;
132(18 suppl. 2):S436-S443

86 ECPR and Resuscitative ECMO

 5. Kim SJ, Kim HJ, Lee HY, Ahn HS, Lee SW. Comparing extracorpo-
real cardiopulmonary resuscitation with conventional cardiopulmo-
nary resuscitation: a meta-analysis. Resuscitation. 2016 Feb 2. Pii:
S0300-9572(16)00045-9
6. Leick J, Liebetrau C, Szardien S, et al. Door-to-implantation time of
extracorporeal life support systems predicts mortality in patients
with out-of-hospital cardiac arrest. Clinical research in cardiology:
official journal of the German Cardiac Society 2013;102:661-669
7. Robertsson S, Lindgren E, Smekal D, et al. Mechanical chest com-
pression and simultaneous defibrillation vs conventional cardio-
pulmonary resuscitation in out-of-hospital cardiac arrest: the LINC
randomized trial. JAMA. 2014; 311(1):53-61
8. Gates S, Quinn T, Deakin CD, Blair L, Couper K, Perkins GD. Me-
chanical chest compressions for out of hospital cardiac arrest: Sys-
tematic review and meta-analysis. Resuscitation. 2015 Sep;94:91-7.
9. Willms DC, Atkins PJ, Dembitsky WP, et al. Analysis of clinical
trends in a program of emergent ECLS for cardiovascular collapse.
Asaio J. 1997;43:65-68. (Retrospective observational; 81 patients)
10. Bunch TJ, White RD, Gersh BJ, et al. Long-term outcomes of
out-of-hospital cardiac arrest after successful early defibrillation. N
Engl J Med. 2003;348:2626-2633.
11. Bellezzo JM, Shinar Z, Davis DP, Jaski BE, Chillcott S, Sta-
hovich M, Walker C, Baradarian S, Dembitsky W. Emergency
physician-initiated extracorporeal cardiopulmonary resuscitation.
Resuscitation. 2012 Aug;83(8):966-70.
1 2. Shinar Z, Plantmason L, Reynolds J, Dembitsky W, Bellezzo J, Ho
C, Glaser D, Adamson R. Emergency Physician-Initiated Resusci-
tative Extracorporeal Membrane Oxygenation. J Emerg Med. 2019
Jun;56(6):666-673.

Chapter 4 87
CHAPTER 5
ECMO Cannulation for ECPR

Jason A. Bartos

Introduction
Initiation of extracorporeal membrane oxygenation (ECMO) in the
setting of extracorporeal cardiopulmonary resuscitation (ECPR) requires
rapid arterial and venous access, insertion of cannulas, and connection
to the circuit and pump/membrane lung. In principle, the procedure is

Foundational Terms
ɋ Drainage (venous) cannula – this cannula typically enters the
femoral vein in ECPR
ɋ Return (arterial) cannula –this cannula typically enters the
femoral artery in ECPR
ɋ Distal perfusion catheter – A catheter that is inserted such
that the catheter tip points distally in the femoral artery. This
allows for perfusion to the leg that is obstructed by the femoral
artery cannula of V-A ECMO
ɋ Guidewires – long wires designed to navigate vessels and pro-
vide a track over which catheters and cannulas are placed. These
can be similar to wires used in typical central line kits or much
stiffer.
ɋ Dilators – short tapered catheters passed over a wire to dilate
the tissue and vessel entry such that larger cannulas can enter.
They are removed prior to insertion of the desired cannula

 89
 Steps of Cannulation
1. Clean the groin site with topical antiseptic
2. Ultrasound location of vessels
3. Micropuncture or 18 gauge needle access to femoral artery/vein
4. Insert guidewires (≥145 cm) and create skin nicks.
5. Dilate vessels with serial dilators (commonly 1-2 dilators for
artery and 2-3 dilators for vein)
6. Insert ECMO cannulas – (commonly 15-17 Fr artery, 21-25 Fr
vein).
7. Connect the ECMO cannulas to circuit using wet-to-wet tech-
nique to exclude air
8. Unclamp cannulas and circuit to initiate flow

straight forward and similar to other procedures routinely performed

in the emergency department, intensive care unit, operating room, and
catheterization laboratory every day. However, the difficulty of ECPR is
amplified by the need for large cannulas, potential for unique anatomical
challenges, ongoing cardiopulmonary resuscitation (CPR) causing con-
stant patient motion, and the realization that every minute that passes
decreases the likelihood of survival. Importantly, CPR should continue
throughout ECMO cannulation to maintain perfusion. This chapter ex-
amines the critical procedural decisions and steps of ECMO cannulation
and the tools that can improve the speed and safety of the procedure.

Need for Speed

Standard advanced cardiac life support (ACLS) provides 15-25% of nor-
mal cardiac output.1-3 Therefore, a substantial mismatch between meta-
bolic demand and supply continues throughout resuscitation resulting in
progressive acidemia.4 Transport to an ECMO center or delivery of ECMO
equipment to the patient’s bedside typically results in mean durations of
CPR of 45-90 min prior to initiation of extracorporeal perfusion.4-7 Mul-
tiple patient cohorts have shown a 4-25% decrease in survival with ev-
ery 10 minutes delay in ECMO beyond 20-30 minutes.4, 8 Therefore, all

90 ECPR and Resuscitative ECMO

aspects of the ECPR system including patient extraction and transport,
patient preparation, and cannulation must be optimized to minimize the
time-to-ECMO.

Need for Safety

Vascular complications at the cannulation site are the most common
complications associated with ECPR. The combination of anticoagulation
and insertion of large cannulas can cause vascular injuries ranging from
mild to life threatening.9 The rate of vascular complications reaches as
high as 40-50% in the setting of ECPR due to the added difficulty of per-
forming the procedure during ongoing CPR.6, 10, 11 However, highly trained
and experienced teams can reduce the risk to 5% or less.12, 13 Bleeding,
infection, and limb ischemia can complicate recovery and reduce survival.
In addition, air embolism, traumatic organ injury, or even unintended
decannulation due to improper cannula positioning can be devastating.
Therefore, development of a cannulation strategy requires consideration
of patient-specific risks and strategies to mitigate these risks.

Cannulation Strategies
Venoarterial (V-A) ECMO requires insertion of a venous cannula that
drains blood from the upper and lower body and an arterial cannula that
reinfuses oxygenated blood into the arterial system. Multiple options
exist for cannulation of V-A ECMO. Considerations that may affect the
choice of cannulation site include the ability to access and prepare the
cannulation site, expertise of the cannulating team, and location and state
of the patient. The methods of cannulation include percutaneous, surgical
cutdown, or hybrid approaches. The use of these methods depends on the
expertise of the cannulating team and available imaging. ECPR typically
relies on a limited number of strategies due to distinct advantages in the
setting of ongoing CPR; however, knowledge of all available options pro-
vides bailout strategies and advantages in special circumstances. Table 1
describes the cannulation locations and potential advantages and disad-
vantages.

Chapter 5 91
Table 1: Cannulation Locations
Location Advantages Disadvantages/Risks
Femoral ■ Rapid cannulation ■ Peripheral arterial disease may
■ Easy site preparation impede cannulation
■ Minimized body motion ■ Lower extremity ischemia
during CPR due to ■ North-South Syndrome
distance from chest
Axillary ■ Reduces risk of North- ■ Increased difficulty due to
South syndrome proximity to CPR
■ Patient may mobilize on ■ Slower cannulation if surgical
ECMO graft is required
■ Control of cannulation site
bleeding is more difficult
■ Upper extremity ischemia or
hyperperfusion
Common ■ Reduces risk of North- ■ Increased stroke risk
Carotid South syndrome ■ Increased difficulty due to
■ Patient may mobilize on proximity to CPR
ECMO
Central ■ Rapid cannulation if ■ Requires sternotomy or
Cannulation sternotomy is already thoracotomy
performed ■ Bleeding risk increased
■ High flows due to use of ■ Cannot be performed during
large and short cannulas closed chest compressions
■ No North-South ■ Decannulation requires
Syndrome surgery
■ May be performed
simultaneous to open
chest cardiac massage

Cannulation Location
Choice of cannulation location takes into account the safety of large bore
vascular access at the site and its impact on the ongoing CPR. Detailed
medical history may be unavailable at the time of ECPR, leaving the can-
nulating team to perform a rapid point-of-care assessment. Skin assess-
ment for scars indicating prior procedures, assessment for risk of severe
peripheral vascular disease such as prior amputations or dialysis access,
and point of care vascular ultrasound can provide important information
prior to beginning the procedure. These findings may lead to use of the
contralateral artery or different equipment to traverse the artery.

92 ECPR and Resuscitative ECMO

 A B C

Figure 1: Common cannulation strategies. A) Femoral cannulation with

arterial cannula in the common femoral artery and venous cannula in the
femoral vein extending to the heart. Contralateral cannulation is shown
here although ipsilateral is also common. B) Axillary cannulation with arte-
rial cannula placed in the axillary artery and venous cannula in the internal
jugular vein. The venous cannula may also be placed in the subclavian vein.
C) Central cannulation with arterial cannula in the ascending aorta and
venous cannula in the right atrium. The venous cannula may also be placed
in the superior or inferior vena cava.

Cannulation locations are classified as peripheral or central. The pe-

ripheral approach includes the femoral (Figure 1A), axillary (Figure 1B),
and carotid arteries, though carotid access is primarily used only in chil-
dren. The arterial and venous cannulas are often inserted ipsilaterally,
though various combinations may be used. Increase speed of insertion
and low complication rates have driven our institution to utilize ipsilat-
eral over contralateral femoral cannulation for ECPR. Cannulation of the
axillary artery may be performed by direct insertion of the arterial can-
nula into the artery or, more commonly, with surgical placement of an
end-to-side Dacron graft on the axillary artery and insertion of the cannula
into the graft. The latter approach limits the risk of upper extremity isch-
emia but increases the risk of upper extremity hyperperfusion. The venous
cannula may be placed in the axillary, subclavian, or internal jugular veins.

Chapter 5 93
 Central cannulation involves direct cannulation of the great vessels
or heart chambers (Figure 1C). In this conformation, the venous cannula
is typically in the right atrium or one of the vena cavas while the arterial
cannula is inserted into the ascending aorta. By perfusing the ascending
aorta, central cannulation eliminates the risk of North-South Syndrome,
which results when antegrade blood flow from the heart competes with
retrograde blood flow from V-A ECMO.14 Central cannulation requires a
sternotomy or thoracotomy, limiting its use in most cardiac arrests with
the exception of cases where a sternotomy has recently been performed.

Femoral Cannulation
Peripheral cannulation using the femoral artery and vein is the most
common approach used in ECPR. Rapid cannulation is possible with-
out interfering with ongoing CPR. The distance from the chest allows
rapid preparation and use of the site while also minimizing the motion
caused by chest compressions. In addition, the femoral artery and vein are
amenable to ultrasound-guided access and both the right and left vessels
can be reached from the same cannulator position minimizing complexity
if bilateral cannulation is needed. Peripheral vascular disease, thrombus,
or prior surgical procedures can make cannulation in the femoral arteries
difficult or impossible in some cases; therefore, alternatives must be con-
sidered.
Femoral cannulation may be performed using a percutaneous, surgi-
cal cutdown, or hybrid approach. Percutaneous insertion uses Seldinger
technique: needle access of the vessel followed by wire insertion and use
of sequential dilators to achieve insertion of the large cannula over the
wire.6, 10, 13, 15 Imaging guidance is commonly used. With the prevalence
and portability of ultrasound, ultrasound-guided vascular access is most
common. Fluoroscopy can also be used. The percutaneous approach offers
more rapid cannulation with some studies reporting cannulation in 6-8
minutes.12, 13, 16 Surgical cutdown is also commonly used though cannula-
tion frequently requires 20-30 minutes.17 Similar rates of limb ischemia are
observed compared to percutaneous methods, but the increased vascular
control may offer advantages in patients with severely diseased vessels.11,
17, 18
Lastly, a hybrid approach has been developed using a cutdown to vi-
sualize the femoral vessels. A percutaneous approach is then used with
insertion through the skin distally and Seldinger technique to enter the

94 ECPR and Resuscitative ECMO

vessels under direct visualization.19 The cutdown is closed after cannula-
tion is complete. This technique requires 20 minutes for completion.19

Central Cannulation
In the setting of ECPR, central cannulation is reserved for cases where a
sternotomy is present or recent. For patients with a sternotomy who suf-
fer a cardiac arrest in the operating room before cardiopulmonary bypass
is initiated or after bypass is removed, central ECMO is a rapidly deploy-
able option for initiation of support. Effective perfusion can be achieved
via open chest cardiac massage performed by one surgeon while ECMO
cannulation is performed by a second. In patients with recent sternotomy,
defined as within 10 days of the cardiac arrest, the Society of Thoracic
Surgeons recommends emergent resternotomy within five minutes of loss
of pulses.20 In these cases, central cannulation for ECMO may be con-
sidered given the reopened chest; however, peripheral cannulation may
be preferred depending on circumstances such as location of the cardiac
arrest and available resources.

Cannula Selection
The choice of arterial and venous cannulas requires a compromise be-
tween maximizing flow and restrictions on cannula size related to patient
factors. Maximizing flow requires consideration of fluid dynamics as de-
scribed by Poiseuille’s Law:

Q = ∆P x (π / 8) x (1 / η) x (r4 / L)

Q is the flow rate, ∆P is the pressure drop across the cannula, η is the
viscosity, r is the radius of the cannula, and L is the length of the cannula.
This equation assumes an incompressible Newtonian fluid with laminar
flow. While blood is not considered Newtonian and flow through a can-
nula may be turbulent, the relative effect of cannula properties on flow
remains consistent. As observed in Poiseuille’s Law, the most important
regulator of flow through a cannula is the radius of the cannula (to the
fourth power) due to its exponential effect. Therefore, a larger diame-
ter cannula will allow substantially increased flow. Length of the cannula
is also important but to a lesser extent; flow is increased with use of a

Chapter 5 95
shorter cannula. Flow properties and cannula characteristics can be found
in published studies or manufacturer data.14, 21

Arterial Return Cannulas

Selection of arterial cannulas is primarily restricted by the relatively small
size of the common femoral artery compared to the venous system. Small
stature, use of high vasopressor doses, and peripheral vascular disease
may further reduce the functional luminal size of the artery. Oversized
cannulas may cause vascular injury during insertion leading to bleeding
and potentially limb ischemia. Cannulas may also occlude the femoral
vessel resulting in limb ischemia even without vascular injury. Therefore,
a distal perfusion cannula should be considered to ensure perfusion of
the lower extremity. Multiple options for these cannulas exist including
direct common femoral cannulation and retrograde posterior tibial can-
nulation. Other considerations include the need for surgical decannula-
tion associated with larger cannulas. These factors must all be considered
when choosing appropriate arterial cannulas (Table 2).
Arterial cannulas are typically 15-19 Fr (Figure 2). Therefore, the
pressure drop is substantial in accordance with Poiseuille’s Law. To par-
tially accommodate for this, their length is minimized. They are typically
15-30cm long with end-holes terminating in the common iliac artery or
abdominal aorta. Arterial cannulas of 15-17 Fr are commonly used in the
femoral artery, resulting in 4-5 L of blood flow per minute which is typ-
ically considered sufficient in ECPR patients.13, 19 Arterial cannulas are
wire-reinforced to resist kinking in tortuous vessels. They also have a side
port with a Luer lock to allow connection to a distal perfusion cannula.

Venous Drainage Cannulas

Venous cannulas must simultaneously drain blood from the upper and
lower body. Therefore, if a single cannula is used, it must be long enough
to reach across the right atrium. Two venous cannulas may be employed
to drain the upper and lower body. This extends the cannulation time,
adds another access site with potential vascular complications, and in-
creases the complexity of the circuit. Therefore, this is not routinely done
in the setting of ECPR. Veins are more compliant and less likely to have
vascular disease allowing use of venous cannulas with larger diameter.

96 ECPR and Resuscitative ECMO

Table 2: Characteristics of Cannulas Used in ECPR
Cannula
Diameter Length Special Features
Type
Arterial ■ 15-19 Fr ■ 15-30 cm ■ Wire-reinforced
■ Luer lock side port for the distal
perfusion cannula
■ Length tailored to patient
Venous ■ 21-25 Fr ■ 55-60 cm ■ Wire-reinforced
■ Extended side holes to
simultaneously drain the SVC,
right atrium, and IVC
Distal ■ 6-9 Fr ■ 10-23 cm ■ Some are wire-reinforced
Perfusion ■ Length tailored to patient
■ Some with large side-arm tubing
to reduce resistance

A B

Figure 2: The proper positioning and relative size of

the commonly used V-A ECMO cannulas. A) The ve-
nous cannula, placed in the right femoral vein, is shown
on CT scan across the right atrium with the end-hole in
the superior vena cava. The extended side holes can be C
seen down to the IVC. B) The arterial cannula, placed in
the right femoral artery, is shown on CT scan extend-
ing to the distal abdominal aorta. C) Commonly used
arterial cannulas are 15-17 Fr in diameter while the commonly used venous
cannulas are 21-25 Fr in diameter. One millimeter is equivalent to 3 Fr. The
bar demonstrates the scale to 1 cm.

Chapter 5 97
The increased diameter reduces the pressure drop across the cannula
compensating for the necessary length (Table 2).
Venous cannulas used for ECPR are typically 21-25 Fr and 55-60cm
long extending from the femoral vein across the right atrium and into the
superior vena cava when placed correctly (Figure 2). They have end-holes
and side holes extending 20-30 cm from the tip to drain blood across the
right atrium and inferior vena cava. They are also wire-reinforced to resist
kinking. There are generally no side ports to reduce the risk of air aspira-
tion into the circuit.

Guidewires
The guidewire is a critical component of percutaneous ECMO cannula-
tion using the Seldinger technique. Upon needle access into the artery or
vein, a guidewire must be inserted for all subsequent dilations and can-
nula insertion. Dilators and ECMO cannulas are engineered to pass over
0.035" guidewires. Larger wires will not pass through the cannulas and
smaller wires will provide suboptimal support and potentially cause ves-
sel injuries when passing the dilators and cannulas. The access needles
are also designed for a 0.035" guidewire. Guidewires of length 145-260
cm are available, though 145 cm guidewires are most widely used as they
support all needs of ECMO insertion without leaving a long length of wire
to manage outside the patient.
There are three key properties of any guidewire that can be selected to
optimize its function in a specific circumstance (Table 3): the tip shape,
wire stiffness, and hydrophilicity of the wire coating. While most patients
can be cannulated with a standard J-tipped super-stiff wire, vascular com-
plexity such as severe peripheral arterial disease or severe tortuosity may
require use of alternative wires.
The shape of the wire tip affects the direction the wire travels and
the safety during wire advancement. A J-tip encourages the wire to re-
main in a large vessel, as the J-tip tends to deflect from smaller branches.
Therefore, J-tip wires will tend to pass in the direction desired for ECMO
cannula insertion. The J-tip also reduces the potential for trauma during
wire advancement, as the J-tip is unlikely to pierce through a vessel wall,
capillary bed, or thin-walled right atrium. A straight wire is useful if pas-
sage through narrow vessels is needed, as may be the case in the setting
of severe peripheral arterial disease. Straight wires also allow the cannula-

98 ECPR and Resuscitative ECMO

Table 3: Guidewire Characteristics

Standard Tortuosity or Vascular Disease

Tip J-tip Straight
Stiffness Super stiff Floppy
Hydrophilicity Neutral or Hydrophilic
hydrophobic

tor to make modifications to the wire such as the addition of a 45-degree

bend on the wire tip. This may be helpful if the patient’s unique anatomy
directs the J-tip wire down an undesirable branch. Use of this technique
is rare and requires imaging to assist in directing the wire. In most cases,
a J-tip wire will be successful.
Wire stiffness is critical as it supports the advancement of the can-
nula. The ideal wire for ECMO insertion would have the maximal stiffness
as it would remain straight, resist kinking, and provide a stiff rail for inser-
tion of the ECMO cannula. However, stiff wires are unable to bend around
tortuous vessels or severe peripheral vascular disease. Therefore, these
situations may require a floppy wire to traverse the vessel. If a balance
cannot be achieved with a single wire, two wires may be necessary. This is
performed with use of a floppy wire to traverse the vessel, insertion of a
long sheath or catheter over the wire beyond the tortuosity, and exchange
of the wire for the preferred stiff option. Of note, most stiff wires used for
ECMO insertion have softer J-tips to prevent vascular trauma that would
arise with a stiff tip. In this configuration, the stiff rail can provide support
for cannula insertion while wire advancement remains safe.
The third critical characteristic of guidewires is the hydrophilic na-
ture of the coating on the wire. Hydrophilic wires can assist in crossing
tight vessel stenoses and very severe tortuosity. However, hydrophilicity
also confers a higher risk of vessel dissection and perforation. Therefore,
hydrophilic wires are generally floppy requiring exchange for stiffer wires
prior to cannula insertion.

Chapter 5 99
Table 4: Imaging Modalities and Use During ECMO
Cannulation
Surface
Ultrasound:
Procedural Transesophageal Vascular and
Component Fluoroscopy Echocardiography Transthoracic
Planning ■ May be used NA ■ Ideal method
Vascular to isolate for choosing
Access the femoral location of
head to insertion,
determine assessing
optimal vascular disease,
entry and guiding
location needle insertion
■ Vessel ■ Vessel pulsatility
calcification does not
can be distinguish
observed artery from vein
during CPR
Confirmation ■ Confirm ■ Confirm wire ■ Confirm
of Vascular wire in aorta across right atrium wire in IVC
Access and IVC into superior vena (subxiphoid)
■ Navigate cava. ■ Confirm
wires ■ Confirm wire in wire in aorta
through aorta (subxiphoid)
difficult ■ Difficult
anatomy to achieve
transthoracic
echo images
with ongoing
CPR
Confirmation ■ Confirm ■ Confirm venous ■ Confirm venous
of Cannula optimal cannula position cannula entering
Position arterial across the right right atrium
and venous atrium into the (unable to
cannula superior vena cava visualize tip in
position superior vena
cava)
■ Difficult
to achieve
transthoracic
echo images
with ongoing
CPR

100 ECPR and Resuscitative ECMO

ECMO Cannulation Procedure
Percutaneous femoral cannulation is the most common strategy in the set-
ting of ECPR for the reasons described above. Therefore, this description
pertains specifically to percutaneous femoral cannulation. Only minimal
adjustments are necessary to apply this procedure to other cannulation
sites. Cutdowns are described in detail in Chapter 6. Specific details of the
cannulation procedure may be institutionally dependent, but the general
order of operations and precautions taken are widely applicable. Imaging
plays an important role in ECMO cannulation, though the available imag-
ing modalities vary between institutions (Table 4).
When approaching any ECPR patient, cannulation speed and safety
need to be considered. Procedural strategies, primary and alternative,
should be established and the team should be trained before any patient
arrives. The equipment needed for all cannulation strategies should be
readily available. Clear roles should be defined prior to patient arrival in-
cluding those assigned to maintaining high quality CPR and resuscitation,
and those assigned to ECMO cannulation.22 In general, ACLS has been
underway for at least 15-20 minutes prior to arrival of a cannulation team.
Therefore, any necessary advanced airway management and vascular ac-
cess efforts should have already been completed. Once ECMO cannula-
tion begins, the team focuses on only two critical tasks: 1) continuing
high quality ACLS and 2) ECMO cannulation. All other tests and thera-
pies should wait until these tasks are complete. This reduces distractions
and interruptions, which is critical to ensure rapid and safe cannulation.

Cannulation Team and Responsibilities

The ECMO cannulation team includes the cannulator, sterile assistant
(a.k.a. wire assistant), and nonsterile assistant. The medical background
of the assistants is institution-dependent, as physicians, nurses, para-
medics, emergency department technicians, and catheterization labora-
tory technicians may fill those roles. Additional staff will be needed to
move the patient to the desired cannulation bed or table. The nonsterile
assistant will be responsible for handing off equipment such as wires, sy-
ringes, and cannulas to the sterile assistant. They will also be responsible
for managing the ECMO machine and handing off the circuit tubing when
appropriate. The sterile assistant prepares the equipment, manages the

Chapter 5 101
wires, exchanges dilators, and eventually places the cannulas on the wires.
Wire management includes ensuring that they do not contact nonsterile
surfaces and pinning the wires to establish the rail for the cannulator to
insert dilators or cannulas. The sterile assistant will also assist the can-
nulator when connecting cannulas to the circuit. The cannulator will be
responsible for achieving vascular access, inserting the wires, managing
the available imaging guidance, inserting the cannulas, and connecting
the cannulas to the circuit.

Preparing the Equipment and Cannulation Site

The procedure begins with cleaning the cannulation site, typically bilat-
eral groins, with topical antiseptic with or without shaving. The cannu-
lator then applies the large femoral drape over the patient establishing
a sterile field. The nonsterile assistant passes equipment to the sterile
assistant and ensures that the ECMO machine is primed, activated, and
any bubbles are removed from the circuit. The sterile assistant prepares
the needles, wires, dilators, and cannulas. The cannulator uses ultrasound
guidance with a sterile sleeve to visualize the femoral vessels (Figure 3).
The ideal insertion location is in the common femoral artery (CFA) at the
level of the femoral heads. The cannulator should identify the superficial
femoral artery (SFA) and deep femoral artery (DFA). The area just prox-
imal to the SFA-DFA bifurcation, where the CFA is visible, will be the cor-
rect insertion location. Adjustments in the cannulation site or equipment
used can also be made at this time if severe peripheral arterial disease is
present.

ECMO Cannulation Procedure

The cannulator inserts an 18-gauge access needle with a syringe applied
for continuous negative pressure into the vein using ultrasound guidance.
When the flash is obtained, the syringe is removed, and a stiff J-tipped
guidewire is inserted through the needle into the vessel and advanced to
the chest. The same is then done for the artery. Unfortunately, vessel pul-
satility and blood oxygenation are often unreliable for identification of
arterial and venous access, as both vessels may be pulsatile during CPR
and the blood from both vessels may be poorly oxygenated. If available,
fluoroscopy may be used to ensure that a wire is seen in the inferior vena

102 ECPR and Resuscitative ECMO

 A B

SFA CFA

DFA
FV FV

Figure 3: Vascular ultrasound of the femoral vessels prior to cannulation.

Ultrasound guidance during ECMO cannulation helps to identify the correct
insertion point as well as signs of potential risks of cannulation. A) Femoral
vessels distal to ideal cannulation site. The superficial femoral artery (SFA),
deep femoral artery (DFA) and femoral vein (FV) are observed. There is
also calcium noted in the DFA at the 7 o’clock position suggestive of some
peripheral vascular disease. B) Just proximal to the image in A in the same
patient, this demonstrates the common femoral artery (CFA) and femoral
vein (FV). This would be an acceptable region for cannula insertion. There is
only mild calcific vascular disease suggesting minimal risk.

cava (IVC) and aorta. Alternatively, transesophageal echocardiography

can be used to observe the venous wire traversing the right atrium into
the superior vena cava while the arterial wire can be observed in the aorta.
Transthoracic echocardiography is very challenging due to the need for
continuous chest compressions.
Once proper wire placement has been confirmed, a bolus of anticoag-
ulation is administered, typically 50-100 U/kg of heparin. A scalpel is used
to make an appropriately sized nick in the skin over the needle. Making
the nick over the wire is avoided, as damaging the wire may make advance-
ment of dilators and cannulas difficult. The needles are then removed
and the dilators advanced sequentially while holding pressure over the

Chapter 5 103
access site to limit blood loss. Typically, only 1-2 dilators are necessary
to achieve the appropriate dilation. When a dilator is exchanged on one
wire, the dilator is left in the vessel on the other wire to limit bleeding.
The sterile assistant is responsible for swapping the dilators on the wire,
while also pinning the wire to form the rail necessary for rapid advance-
ment of the dilators. Racking the wire, a technique including alternating
slight advancement and retraction of the wire, can also be used to prevent
kinking. After use of the dilator that is equal to, or slightly smaller than
the desired cannula size, the cannula is placed on the wire by the sterile
assistant and advanced to its appropriate location by the cannulator. The
venous cannula is typically inserted first with the goal to insert it over the
wire across the right atrium into the superior vena cava. This position is
best confirmed by imaging if available. The depth of cannula insertion
will depend on the patient’s stature but is confirmed by imaging or prior
measurement. Once the position is confirmed, the dilator and wire are
removed and a tubing clamp is secured on the cannula to limit blood loss.
The arterial cannula is then inserted over the wire. The entire narrow por-
tion of the cannula will be inserted in all people independent of stature.
The dilator and wire are removed, and a tubing clamp is placed on the
cannula to limit bleeding. If fluoroscopy and TEE are not available, pro-
cedural adjustments can enhance cannulation safety. First, the intended
depth of the venous cannula can be estimated by laying the cannula on the
patient and measuring from the groin to the 3rd intercostal space. Sec-
ond, the guidewire should be inserted at least to the point that the back
of the wire is at the patient’s feet. If uncertainty arises because of the pa-
tient’s stature, a clamp can be placed on the wire to mark the length of the
cannula. It can then be advanced to the clamp to ensure that the cannula
will not extend beyond the wire once inserted. Lastly, the cannula should
be inserted to within 10 cm of the intended final depth, at which time the
cannula obturator should be pinned in place while the cannula is inserted
to its correct final position. While imaging is preferred to optimize safety,
these processes can improve safety of blind insertion.
Once both cannulas are in place, the ECMO circuit tubing is con-
nected to them using the wet-to-wet technique (see Chapter 8). Circuit
flow is initiated and the cannulator waits to ensure that venous blood is
removed through the venous cannula and oxygenated blood is returned
through the arterial cannula. When correct flow is confirmed, the cannu-
las are secured in place and the resuscitation continues.

104 ECPR and Resuscitative ECMO

Additional Post-ECMO Procedures
While ECMO flow is critical, ongoing resuscitation is necessary to sta-
bilize the patient. Multiple procedural steps may be considered: 1) in-
sertion of venous access in the contralateral groin for central medication
administration, 2) insertion of arterial access in the contralateral femoral
artery for arterial pressure monitoring, 3) initiation of therapeutic hypo-
thermia by surface, catheter-based, or ECMO circuit cooling methods, 4)
insertion of a right radial arterial line for blood gas monitoring, and 5)
evaluation and treatment of underlying etiologies including possible cor-
onary angiography and revascularization.12, 16, 23 The timing of these inter-
ventions may vary, but ongoing resuscitation is critical to ensure optimal
patient survival.

Distal Perfusion Cannulas

Limb ischemia is a common complication of peripheral ECMO due to the
space occupying arterial cannula. Distal perfusion cannulas prevent limb
ischemia by shunting a small portion of the oxygenated ECMO blood flow
to the limb. This is accomplished by inserting a sheath into the SFA di-
rected distally (Figure 4). The sheath is connected to the Luer lock port
on the arterial cannula. Patient selection for distal perfusion cannulas is
difficult given that the vascular tone changes throughout the course of
treatment as vasopressor doses and post-arrest vasoplegia change. There-
fore, some teams advocate placement of distal perfusion cannulas in all
patients with femoral V-A ECMO.11, 12 Others use perfusion assessment
with Doppler ultrasound and capillary refill to estimate risk for limb isch-
emia. Our program at the University of Minnesota inserts distal perfusion
cannulas for all patients resulting in a reduced rate of limb ischemia.12,
13
Placement becomes increasingly difficult over time as residual flow
through the SFA decreases. Therefore, it is often preferred to place the
distal perfusion cannula immediately after ECMO flow is initiated and the
patient has stabilized.
Multiple considerations arise when addressing distal perfusion. First,
slow flow through the sheath may produce thrombosis, so the sheath size
should be maximized for the SFA. Typically, 8-9 Fr sheaths are the largest
used for this purpose. Second, legs bend and flex during patient trans-
fers even if the patient is heavily sedated. Therefore, there is substantial

Chapter 5 105
 V

D
A

Figure 4: Typical ECMO cannulation in the right groin. The arterial cannula
A) venous cannula V), and distally-directed distal perfusion cannula D) can
be observed. The distal perfusion cannula is connected to the Luer lock
port of the arterial cannula via tubing.

risk of kinking the sheath leading to turbulent flow which also can induce
thrombosis. Wire-reinforced sheaths can be used to limit kinking. Third,
distal perfusion cannula length should be tailored to the patient to avoid
displacement. Sheaths range in length from 10 to 23 cm. Larger patients
may require longer sheaths.

Distal Perfusion Cannula Insertion

The SFA is smaller and deeper than the CFA making insertion more diffi-
cult. In addition, the resistance provided by the arterial cannula and the
continuous flow of ECMO reduce or eliminate any pulsatility. Ultrasound
guidance is critical. Given the difficulty, vessel injury may occur from
multiple needle sticks and the anticoagulation used for V-A ECMO may
cause continued bleeding. Therefore, a micropuncture (21 gauge) needle

106 ECPR and Resuscitative ECMO

is recommended for access. Once the needle and micropuncture wire are
in place, the micropuncture kit is used to upsize the access site to a 0.035"
wire for sheath insertion. With severe peripheral vascular disease access
may be difficult, requiring fluoroscopy and angiography. Transfer to a cath-
eterization laboratory may be preferred for distal perfusion cannula place-
ment if this is suspected.

Special Circumstances and Procedural Pearls

For ECPR patients, little past medical history may be available and the
need for rapid initiation of ECMO precludes efforts to gather more in-
formation. Therefore, the cannulating team may encounter surprises and
must be able to adapt to ensure safe initiation.

Vascular Occlusion
The most commonly encountered problem is severe peripheral arterial
disease. If the femoral artery is severely calcified on ultrasound, con-
sider switching to the contralateral artery. If disease or tortuosity pre-
vents passage of the standard super stiff wire, a floppy, hydrophilic, or
straight-tipped wire may be helpful. When these are used, they should
be exchanged for a stiff wire prior to cannula insertion. If concern for
bilateral occlusion of the femoral or iliac arteries arises, assess for signs
of arterial bypass including scars. Vascular bypass grafts can be visualized
with ultrasound. If they are present, arterial ECMO cannulas can be in-
serted into a bypass graft if necessary (Figure 5). A micropuncture (21
gauge) needle may be used for access to avoid complications from multi-
ple punctures, and a 15 Fr cannula may be used to minimize disruption of
the graft. Visualization of the wire with imaging guidance is important to
ensure that the wire advances appropriately. The procedure is otherwise
unchanged.
Venous occlusion is less common; however, IVC filters may impede
venous cannula insertion. Two options exist in this situation. Some IVC
filters can be traversed with a wire and venous cannula such that ECMO
cannulation can proceed normally (Figure 6). If the venous cannula can-
not pass through the filter, a bicaval venous cannulation strategy may be
employed with insertion of a venous cannula through the femoral vein

Chapter 5 107
Figure 5: Insertion of an arterial ECMO cannula into an aorto-​femoral graft.
The patient had prior aorto-bifem bypass surgery. A 15 Fr arterial cannula
was successfully inserted into the right sided aortofemoral bypass graft
during ongoing CPR. The calcified native vessel can be seen in the inset
between the arterial cannula in the graft and the venous cannula in the fem-
oral vein. She recovered well and had surgical decannulation four days later.

into the abdominal inferior vena cava and a short venous cannula in the
internal jugular vein (Figure 7).

Aortic Dissection
Aortic dissection is considered a contraindication for V-A ECMO,24 al-
though this is a topic of ongoing debate.25, 26 As such, identification of aortic
dissection by ultrasound may lead to termination of ECMO cannulation.
However, if ECMO cannulation is attempted, it is critical to ensure that
the ECMO cannula is placed in the true lumen. This can be assessed by
observing the wire as it advances through the aorta. If it passes easily to the
coronary cusp, it is likely in the true lumen. Fluoroscopy and angiography
may be necessary to confirm placement.

Severe Aortic Insufficiency

Peripheral ECMO pumps blood retrograde from the abdominal aorta.
When LV ejection is minimal, blood flows back to the aortic root and cor-
onary arteries. If significant aortic insufficiency is present, blood contin-

108 ECPR and Resuscitative ECMO

 A

Figure 6: Venous cannula traversing through an IVC filter observed in the

abdominal inferior vena cava. The patient was cannulated with VA ECMO
for ECPR. The arterial cannula can be seen in the right iliac artery in the
coronal slice (a). A wire was able to traverse the IVC filter and the venous
cannula followed allowing for standard cannulation. Figure 7: (above right)
Bicaval drainage in the setting of V-A ECMO. Placement of venous cannulas
in the internal jugular vein and femoral vein for separate drainage of the
upper body and lower body venous return can be useful if an IVC filter pre-
vents advancement of a venous cannula from the femoral vein. The arterial
cannula is placed in the usual position through the femoral artery.

Chapter 5 109
 Figure 8: Cannulation of a patient
with situs inversus with V-A ECMO
for ECPR. This patient had no known
past medical history but was found to
have situs inversus during emergent
V-A ECMO cannulation. He was can-
nulated from the left femoral artery
and vein. The venous cannula can be
observed coursing up and transiting
the hepatic IVC left of midline. The
arterial cannula can be seen transit-
ing in the left iliac artery in the coro-
nal slice (left). He recovered well and
was decannulated five days later.

ues through the aortic valve causing severely increased left ventricular
end-diastolic pressure and left ventricular distention. This typically re-
sults in worsening mitral valve regurgitation, and retrograde blood flow
into the left atrium and the pulmonary veins resulting in severe pulmo-
nary edema. When severe aortic insufficiency produces severe fluid con-
gestion in the lungs, hemodynamic compromise, and inability to support
the patient. Unfortunately, no therapies have been proven to alleviate
this retrograde flow. LV venting strategies have been attempted without
demonstrated success. This is an area for future study; however, severe
aortic insufficiency remains a largely fatal condition when coupled with
cardiac arrest and V-A ECMO.

110 ECPR and Resuscitative ECMO

Large Inguinal Hernia
The presence of a large inguinal hernia can impede access to the femoral
vessels. However, the hernia itself is not a contraindication to femoral
ECMO cannulation. The contralateral groin may be preferred. However,
the artery and vein can be accessed around the hernia with ultrasound
guidance.

Situs Inversus
Situs inversus is an uncommon congenital condition in which the internal
organs are arranged in a mirror image of normal anatomy. It may occur
alone or in association with other abnormalities. If the arterial and venous
wires are in parallel vascular spaces, but they are located on opposite sides
of the body from the intended location, consider whether situs inversus
could be present. Confirmation is by observing the curling of the arterial
wire leftward over the aortic arch and the advancement of the venous wire
straight into the internal jugular vein on the left side of the body. Once
the artery and vein are verified, the cannulation procedure is unchanged
(Figure 8).

References
1. Lurie KG, Mulligan KA, McKnite S, Detloff B, Lindstrom P, Lindner
KH. Optimizing standard cardiopulmonary resuscitation with an
inspiratory impedance threshold valve. Chest. 1998;113:1084-1090
2. Duggal C, Weil MH, Gazmuri RJ, Tang W, Sun S, O’Connell F, Ali M.
Regional blood flow during closed-chest cardiac resuscitation in rats.
Journal of applied physiology. 1993;74:147-152
3. Lurie K, Voelckel W, Plaisance P, Zielinski T, McKnite S, Kor D, Sugi-
yama A, Sukhum P. Use of an inspiratory impedance threshold valve
during cardiopulmonary resuscitation: A progress report. Resuscita-
tion. 2000;44:219-230
4. Bartos JA, Grunau B, Carlson C, Duval S, Ripeckyj A, Kalra R,
Raveendran G, John R, Conterato M, Frascone RJ, Trembley A,
Aufderheide TP, Yannopoulos D. Improved survival with extracor-
poreal cardiopulmonary resuscitation despite progressive metabolic

Chapter 5 111
 derangement associated with prolonged resuscitation. Circulation.
2020;141:877-886
5. Bougouin W, Dumas F, Lamhaut L, Marijon E, Carli P, Combes A,
Pirracchio R, Aissaoui N, Karam N, Deye N, Sideris G, Beganton F,
Jost D, Cariou A, Jouven X, Sudden Death Expertise Center i. Extra-
corporeal cardiopulmonary resuscitation in out-of-hospital cardiac
arrest: A registry study. European heart journal. 2019
6. Stub D, Bernard S, Pellegrino V, Smith K, Walker T, Sheldrake J,
Hockings L, Shaw J, Duffy SJ, Burrell A, Cameron P, Smit de V, Kaye
DM. Refractory cardiac arrest treated with mechanical cpr, hypo-
thermia, ecmo and early reperfusion (the cheer trial). Resuscitation.
2015;86:88-94
7. Wang CH, Chou NK, Becker LB, Lin JW, Yu HY, Chi NH, Hunag
SC, Ko WJ, Wang SS, Tseng LJ, Lin MH, Wu IH, Ma MH, Chen YS.
Improved outcome of extracorporeal cardiopulmonary resuscita-
tion for out-of-hospital cardiac arrest—a comparison with that for
extracorporeal rescue for in-hospital cardiac arrest. Resuscitation.
2014;85:1219-1224
8. Wengenmayer T, Rombach S, Ramshorn F, Biever P, Bode C, Duer-
schmied D, Staudacher DL. Influence of low-flow time on survival
after extracorporeal cardiopulmonary resuscitation (ecpr). Critical
care. 2017;21:157
9. Rupprecht L, Lunz D, Philipp A, Lubnow M, Schmid C. Pit-
falls in percutaneous ecmo cannulation. Heart, lung and vessels.
2015;7:320-326
10. Johnson NJ, Acker M, Hsu CH, Desai N, Vallabhajosyula P, Lazar S,
Horak J, Wald J, McCarthy F, Rame E, Gray K, Perman SM, Becker L,
Cowie D, Grossestreuer A, Smith T, Gaieski DF. Extracorporeal life
support as rescue strategy for out-of-hospital and emergency depart-
ment cardiac arrest. Resuscitation. 2014;85:1527-1532
11. Kreibich M, Benk C, Leitner S, Beyersdorf F, Morlock J, Scherer C,
Rylski B, Trummer G. Local and lower limb complications during
and after femoral cannulation for extracorporeal life support. The
Thoracic and cardiovascular surgeon. 2019;67:176-182
12. Bartos JA, Carlson K, Carlson C, Raveendran G, John R, Aufderheide
TP, Yannopoulos D. Surviving refractory out-of-hospital ventricular
fibrillation cardiac arrest: Critical care and extracorporeal mem-
brane oxygenation management. Resuscitation. 2018;132:47-55

112 ECPR and Resuscitative ECMO

13. Yannopoulos D, Bartos JA, Martin C, Raveendran G, Missov E,
Conterato M, Frascone RJ, Trembley A, Sipprell K, John R, George
S, Carlson K, Brunsvold ME, Garcia S, Aufderheide TP. Minnesota
resuscitation consortium’s advanced perfusion and reperfusion
cardiac life support strategy for out-of-hospital refractory ventricular
fibrillation. Journal of the American Heart Association. 2016;5:e003732
14. Pavlushkov E, Berman M, Valchanov K. Cannulation techniques
for extracorporeal life support. Annals of translational medicine.
2017;5:70
15. Bellezzo JM, Shinar Z, Davis DP, Jaski BE, Chillcott S, Sta-
hovich M, Walker C, Baradarian S, Dembitsky W. Emergency
physician-initiated extracorporeal cardiopulmonary resuscitation.
Resuscitation. 2012;83:966-970
16. Yannopoulos D, Bartos JA, Raveendran G, Conterato M, Frascone
RJ, Trembley A, John R, Connett J, Benditt DG, Lurie KG, Wil-
son RF, Aufderheide TP. Coronary artery disease in patients with
out-of-hospital refractory ventricular fibrillation cardiac arrest.
Journal of the American College of Cardiology. 2017;70:1109-1117
17. Le Guen M, Nicolas-Robin A, Carreira S, Raux M, Leprince P, Riou
B, Langeron O. Extracorporeal life support following out-of-hospital
refractory cardiac arrest. Critical care. 2011;15:R29
18. Chen YS, Lin JW, Yu HY, Ko WJ, Jerng JS, Chang WT, Chen WJ,
Huang SC, Chi NH, Wang CH, Chen LC, Tsai PR, Wang SS, Hwang
JJ, Lin FY. Cardiopulmonary resuscitation with assisted extracorpo-
real life-support versus conventional cardiopulmonary resuscitation
in adults with in-hospital cardiac arrest: An observational study and
propensity analysis. Lancet. 2008;372:554-561
19. Lamhaut L, Hutin A, Puymirat E, Jouan J, Raphalen JH, Jouffroy R,
Jaffry M, Dagron C, An K, Dumas F, Marijon E, Bougouin W, Tourtier
JP, Baud F, Jouven X, Danchin N, Spaulding C, Carli P. 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
20. Society of Thoracic Surgeons Task Force on Resuscitation After
Cardiac S. The society of thoracic surgeons expert consensus for the
resuscitation of patients who arrest after cardiac surgery. The Annals
of thoracic surgery. 2017;103:1005-1020

Chapter 5 113
21. Wang S, Force M, Kunselman AR, Palanzo D, Brehm C, Undar A. He-
modynamic evaluation of avalon elite bi-caval dual lumen cannulas
and femoral arterial cannulas. Artificial organs. 2019;43:41-53
22. Hutin A, Abu-Habsa M, Burns B, Bernard S, Bellezzo J, Shinar
Z, Torres EC, Gueugniaud PY, Carli P, Lamhaut L. Early ecpr for
out-of-hospital cardiac arrest: Best practice in 2018. Resuscitation.
2018;130:44-48
23. Yannopoulos D, Bartos JA, Aufderheide TP, Callaway CW, Deo
R, Garcia S, Halperin HR, Kern KB, Kudenchuk PJ, Neumar RW,
Raveendran G, American Heart Association Emergency Cardiovascu-
lar Care C. The evolving role of the cardiac catheterization labo-
ratory in the management of patients with out-of-hospital cardiac
arrest: A scientific statement from the american heart association.
Circulation. 2019;139:e530-e552
24. King CS, Roy A, Ryan L, Singh R. Cardiac support: Emphasis on
venoarterial ecmo. Critical care clinics. 2017;33:777-794
25. Shinar Z. Contraindicated? - Aortic dissection and ECPR. Resuscita-
tion. 2020 Nov;156:268-269.
26. Ohbe H, Ogura T, Matsui H, Yasunaga H. Extracorporeal cardiopul-
monary resuscitation for acute aortic dissection during cardiac ar-
rest: A nationwide retrospective observational study. Resuscitation.
2020 Nov;156:237-243.

114 ECPR and Resuscitative ECMO

CHAPTER 6
Hybrid Cutdown Technique for
ECPR Implementation

Alice Hutin • Lionel Lamhaut

Introduction
The hybrid cutdown technique for ECPR is an alternative method to
the percutaneous approach to cannula insertion. We have used this ap-
proach since 2011 for prehospital extracorporeal cardiopulmonary resus-
citation (ECPR) cannulation in Paris. We chose this after several failed
attempts to implement ECPR using a percutaneous approach. This tech-
nique has been specifically customized to be used by non-surgeons and
has been successful in the austere environments of prehospital cannula-
tion.1,10 Our success with this technique has made it our preferred choice
for both in-hospital and prehospital environments.

Foundational Term
ɋ Hybrid femoral cutdown- vascular access procedure to locate
the femoral vessels via incision just below the inguinal ligament.
Visualization of a percutaneously inserted needle into the fem-
oral vessel allows for confirmation of wire insertion and secure-
ment of cannula with surrounding skin

 115
ECPR Team
Our ECPR team relies on 3 people: a physician, an anesthetic nurse, and a
paramedic. This configuration is the same for ECPR responses within and
outside of the hospital. It joins the resuscitation team already treating the
patient. In the Paris pre-hospital setting, an initial mobile intensive care
ambulance team that consists of a paramedic, an anesthetic nurse, and
a physician is deployed to the scene of cardiac arrests prior to the ECPR
team.
The ECPR physician is the proceduralist carrying out the cutdown.
In our team, the ECPR physician is either an emergency physician or an
intensivist. They are assisted during the procedure by a separate provider
who can be the physician from the initial ambulance or any other person
able to hold retractors.

Procedure Description
Sterile Field, Tools and Other Equipment
The ECPR physician and assistant dress in sterile gowns. All necessary
equipment is handed to them by the extracorporeal membrane oxygen-
ation (ECMO) paramedic. As shown in Figure 1, our sterile kit (panel
A) is similar to an “angio kit”: it includes sterile gowns for the 2 opera-
tors (panel B), large sterile field (panel C), a scalpel, sterile pads, suture
thread (panel D). The operators are separately handed the sterile toolbox
shown in panel E (with retractors, scissors, surgical clamps, and forceps).
Extra sterile tools such as vascular clamps are available in case of vascular
complications. The J-tip guidewires (100 and 150cm) and cannulas (ar-
terial cannula 15, 17 or 19 Fr, venous 21 or 23 Fr) are handed on demand
to the ECPR physician.

Cutdown Anatomy
Unlike the vertical cutdown done by surgeons,2 our 5 cm cutdown is made
horizontally and located a few centimeters below the inguinal ligament.
At this location, movement due to automated chest compressions is less
disruptive. While not common practice in our program, vessel location
can be checked by ultrasound before the incision.

116 ECPR and Resuscitative ECMO

 A

B E

C D

Figure 1: Pre-Hospital Sterile Cannulation Pack. (Sterile gowns (B), surgi-

cal field (C), tools (E) and other equipment (D) used for ECPR implemen-
tation).

Dissection
After the incision, the helper retracts the skin on both sides and blunt
dissection of the subcutaneous tissue is made using the fingers. This tech-
nique is used to limit the risk of vascular injury and hemorrhage. Scissors
are only used to cut the aponeurosis and open the vessel sheath. The ves-
sels are then clearly exposed and identified. The artery appears white with
vaso-vasorum on its vessel wall. It feels much thicker than the vein which
is darker, thin-walled, and collapsible.

Tunneled Guidewire Insertion

Once the vessels have been clearly identified and exposed, the assistant
removes the retractor below the incision. The physician will then ask for
the first guidewire set to insert the first cannula. Our experience is that

Chapter 6 117
 A B
Arterial Cannula

Figure 2: Return Cannula with depiction of step-off at cannula/introducer

interface.

accessing the vein first is advantageous as it is more collapsible and lies

just behind the artery. Accessing the artery first can complicate the sec-
ondary access to the vein.
The hollow needle is inserted through the skin just distal to the inci-
sion. Under strict visual control, the needle is passed into the vessel. Once
the needle is in the vessel with an adequate blood return, the guidewire
can be inserted in the needle. It is crucial never to let go of the needle at
this point as it can easily go through the back wall of the vessel.

Cannula Insertion
Once the guidewire is in place, the needle can be removed. As most of
the subcutaneous tissue has been dissected, very little to no dilation is
necessary and wire kinking is rare. A larger incision can be made to the
skin around the wire. The cannula then is threaded over the guidewire,
through the skin, and into the vessel. Some resistance can come from the
insertion of the tip of the cannula through the vessel wall (Figure 2).

Variations within Our Team

The different physicians in our team do not all have the same approach.
Most start with the venous cannula; however, some start with the arterial
cannula. One of our cannulators also has the habit of placing both guide-
wires followed by the cannulas.

118 ECPR and Resuscitative ECMO

 A B
Cutdown

Arterial cannula

Venous cannula

Figure 3: Hybrid Cutdown Pictorial. Access and return cannulas are in-
serted percutaneously yet direct visualization of vessels is maintained.

Wet-to-Wet Connection and Pump Activation

Once both cannulas have been inserted, the operator is given the sterile
circuit tubing in order to connect it to the cannulas, avoiding any bubbles.
At this point, the clamps on the circuit and the cannulas can be removed
and the pump is activated. Automated cardiac compressions can be in-
terrupted, and flow is gradually increased. The arteriovenous difference
(bright red and dark red) should immediately appear between the two can-
nulas. Figure 3 illustrates the cutdown and insertion of the cannula before
(panel A) and after (panel B) wet-to-wet connection and pump activation.

Distal Perfusion Catheter

In order to avoid distal limb ischemia, a single lumen catheter is placed in
the superficial femoral artery just distal to the entrance point of the arte-
rial cannula. The catheter is placed under visual control using the same
Seldinger tunneling technique as for the cannulas (the needle is inserted
through the skin proximal to the cutdown). The needle penetrates the
femoral artery just distal to the insertion site of the cannula (Figure 4).

Chapter 6 119
Figure 4: Arterial (Return) Cannula is inserted cranial while reperfusion
catheter is inserted caudal.

The catheter is then connected to the arterial cannula in order to supply

oxygenated blood to the leg.

Securing Cannulas
Securing each cannula is imperative before the patient is moved. In the ab-
sence of bleeding, the cutdown can be sutured. A compressive bandage is
applied across the groin. The circuit is also secured to the leg in order to
avoid accidental decannulation during patient extraction.

Results
As recently published, mean ECPR implementation time using this
technique is 21.3 minutes.1 This technique has proven to be efficient in
both prehospital and in-hospital implementation, and failure rate is low
(7.6%). Chhor et al report a median time to ECPR implementation of 19
min with an ultrasound guided percutaneous approach, and a 14.5% fail-
ure rate.3 Kashiura et al found 8% of patients cannulated percutaneously
with ultrasound required switch to surgical approach. In this study per-
cutaneous cannulation failure rate dropped with cannulation using both
ultrasound and fluoroscopy.4 Voicu et al found similar results with ultra-
sound guided puncture followed by fluoroscopy.5

120 ECPR and Resuscitative ECMO

Benefits and Complications of Hybrid Cutdown
Technique
Benefits
The hybrid cutdown technique has four main advantages. First, it can be
used without the need for imaging. This is especially advantageous for
pre-hospital locations. Second, direct visual access to the vessels allows
clear identification of the vein from the artery. Anatomic variations occur
frequently and often the femoral vessels lie directly on top of one another.
Third, this technique increases the likelihood of a single needle puncture
for each vessel, decreasing the risk of bleeding. In the event of bleeding, it
can be immediately visualized and managed. Finally, some patients (e.g.
those with obesity or atherosclerosis) are not amenable to the percutane-
ous cannulation. Having a second technique that the provider is skilled at
performing is critical in these cases.

Complications
Globally, complications due to this technique and failure to implement
ECPR are the same as for surgical6,7 and percutaneous cannulation5,8,9
techniques. Hemorrhage is clearly a dreaded complication. The use of
blunt dissection limits the occurrence of vascular injury. However, bleed-
ing and vessel tearing can occur during cannula insertion. Applying com-
pression often suffices to stem the bleed. If not, vascular clamping can be
necessary. In this case, cannulation of the other leg might become neces-
sary. Although the bleeding might have stopped, relapse can occur once
the pump is activated. Even in the absence of massive bleeding, blood
transfusion is often required once in the hospital due to coagulopathy re-
lated to heparinization of the circuit and dilution of coagulation factors.

Cutdown Training
Our team has been using this technique for almost a decade. The two initial
cannulators have trained the rest of the team. Over the years, a more formal
training has been set up to train physicians from around the world. The
first steps of the cutdown are taught on a gel pelvis model. Location and
steps are taught during these training sessions. In order to visualize the

Chapter 6 121
vessels themselves, training is then done on cadavers. Ideally, perfused ca-
davers make the cannulation process even more realistic. This step has the
advantage of locating the vessels and seeing/feeling the difference between
the vein and the artery. Real time simulation is a crucial step to avoid pos-
sible complications. Indeed, at this point the cannulator sees how and why
cannulation can fail. For example, the trainee recognizes the importance
of a ready guidewire for the speed and success of a one stick insertion.
Once the future cannulators have gone through these training steps,
they assist the cannulator for a certain number of procedures. According to
our experience, a new cannulator must assist at least 5 procedures and do
5 proctored procedures before being able to cannulate alone.

Other Applications
The cutdown technique was initially developed for ECPR implementation
for refractory OHCA. Since then, our team has been trained to use the
cutdown for resuscitative endovascular balloon occlusion of the aorta
(REBOA) in traumatic cardiac arrest.
As previously mentioned, we exclusively use the cutdown technique
for ECPR cannulation. In the very rare event of ECMO implementation in
the prehospital setting in the absence of cardiac arrest (i.e. post cardiac
arrest refractory cardiogenic shock with impossible patient transporta-
tion), a percutaneous approach could be used.

Summary
We exclusively employ this hybrid cutdown technique for in-hospital and
prehospital ECPR implementation. Over the years this technique has
shown to be safe and efficient for ECPR implementation by non-surgeons.

References
1. Lamhaut L, Hutin A, Dagron C, Baud F, An K, Carli P. A new hybrid
technique for extracorporeal cardiopulmonary resuscitation for use
by nonsurgeons. Emergencias. 2021 Abr;33(2):156-157.

122 ECPR and Resuscitative ECMO

 2. Banfi C, Pozzi M, Brunner M-E, Rigamonti F, Murith N, Mugnai
D, et al. Veno-arterial extracorporeal membrane oxygenation: an
overview of different cannulation techniques. J Thorac Dis. sept
2016;8(9):E875‑85.
3. Chhor V, Follin A, Joachim J, Champigneulle B, Chatelon J, Favé G,
et al. Risk factors of percutaneous cannulation failure by intensivists
for veno-arterial extracorporeal life support for refractory cardiac
arrest. Intensive Care Med. 2017;43(11):1742‑4.
4. Kashiura M, Sugiyama K, Tanabe T, Akashi A, Hamabe Y. Effect of
ultrasonography and fluoroscopic guidance on the incidence of com-
plications of cannulation in extracorporeal cardiopulmonary resusci-
tation in out-of-hospital cardiac arrest: a retrospective observational
study. BMC Anesthesiol. 06 2017;17(1):4.
5. Voicu S, Henry P, Malissin I, Jean-Guillaume D, Koumoulidis A,
Magkoutis N, et al. Improving cannulation time for extracorporeal
life support in refractory cardiac arrest of presumed cardiac cause
- Comparison of two percutaneous cannulation techniques in the
catheterization laboratory in a center without on-site cardiovascular
surgery. Resuscitation. janv 2018;122:69‑75.
6. Le Guen M, Nicolas-Robin A, Carreira S, Raux M, Leprince P, Riou B,
et al. Extracorporeal life support following out-of-hospital refractory
cardiac arrest. Crit Care Lond Engl. 2011;15(1):R29.
7. Rousse N, Robin E, Juthier F, Hysi I, Banfi C, Al Ibrahim M, et al.
Extracorporeal Life Support in Out-of-Hospital Refractory Cardiac
Arrest. Artif Organs. sept 2016;40(9):904‑9.
8. Champigneulle B, Bellenfant-Zegdi F, Follin A, Lebard C, Guinvarch
A, Thomas F, et al. Extracorporeal life support (ECLS) for refractory
cardiac arrest after drowning: an 11-year experience. Resuscitation.
2015;88:126‑31.
9. Stub D, Bernard S, Pellegrino V, Smith K, Walker T, Sheldrake J, et
al. Refractory cardiac arrest treated with mechanical CPR, hypother-
mia, ECMO and early reperfusion (the CHEER trial). Resuscitation.
janv 2015;86:88‑94.
10. Hutin A, Ricard-Hibon A, Briole N, Dupin A, Dagron C, Raphalen JH,
et al. First description of a Helicopter-borne ECPR Team for remote
refractory out of hospital cardiac arrest. Prehospital Emerg Care Off J
Natl Assoc EMS Physicians Natl Assoc State EMS Dir. 4 déc 2020;1‑6.

Chapter 6 123
CHAPTER 7
Physiology of V-A ECMO for
Cardiac Arrest and Cardiac
Support

Sage P Whitmore • Elliott S Cohen

Introduction
This chapter focuses on the physiologic principles of resuscitative
extracorporeal membrane oxygenation (ECMO), with a focus on extra-
corporeal cardiopulmonary resuscitation (ECPR) and cardiac support,
including cardiorespiratory physiology and patient-circuit interactions.

Physiologic Principles
Oxygen Delivery and Consumption
Homeostasis of the human organism depends on the delivery of oxygen to
tissues to sustain cellular respiration. Shock, multiorgan failure, and death
occur due to failure to oxygenate and circulate blood. Venoarterial (V-A)
ECMO provides both of these functions. Oxygen delivery (DO2) occurs as
oxygenated arterial blood makes its way through tissue capillary networks,
where O2 dissociates from hemoglobin and is used for aerobic cellular res-
piration by the mitochondria. Oxygen consumption (VO2) by cells varies
depending on metabolic rate, stress, temperature, and other factors. In
health, DO2 is maintained at approximately five times VO2 (DO2:VO2 =

 125
 Foundational Terms
ɋ DO2 – Oxygen delivery – amount of oxygen that is delivered to
body tissues
ɋ VO2—Oxygen consumption - amount of oxygen consumed by
the tissues
ɋ SvO2—mixed venous oxygen saturation (in the pulmonary artery)
ɋ Pump head—disposable piece connected to centrifugal pump
motor that spins and pumps blood
ɋ Membrane lung—artificial lung that oxygenates blood and
removes CO2
ɋ Sweep gas—mixture of air and oxygen used for gas exchange in
the membrane lung
ɋ FdO2—fraction of oxygen in the sweep gas set by a blender
ɋ PINLET—negative pressure of blood entering the pump head
ɋ SPREO2—saturation of blood being drained in venous drainage
cannula
ɋ PPRE—positive pressure of blood after the pump and before the
membrane lung
ɋ PPOST—positive pressure after the membrane lung
ɋ SPOSTO2—saturation of blood after the membrane lung
ɋ Delta P—difference between PPRE and PPOST, reflective of resis-
tance across the membrane lung
ɋ Differential hypoxemia or “North-South Syndrome”— hy-
poxemia in the proximal “north” aorta due to native lung dys-
function and adequate oxygenation of the distal “south” aorta by
the membrane lung
ɋ Left ventricular (LV) venting – an intervention to decompress
the left ventricle (reducing ventricular volume and end-diastolic
pressure). This prevents pulmonary edema, myocardial isch-
emia, and cardiac clot
ɋ “Chugging/chatter”—excessively negative pressure in the ve-
nous drainage cannula leading to intermittent suction events of
the cannula onto the vena cava causing the tubing to shake

126 ECPR and Resuscitative ECMO

Figure 1a: Normal balance of oxygen delivery and consumption. A) arterial
blood is ejected from the heart at a typical resting cardiac output (CO).
Arterial blood gas values are shown. B) oxygen extraction occurs at the tis-
sue level; example oxygen delivery (DO2) and oxygen consumption (VO2)
calculations are shown (*assuming a hemoglobin of 15g/dL). The tissue
oxygen extraction ratio (ER) is 20%, and the cells produce a small amount
of carbon dioxide (CO2), lactate, and other metabolites (H+). C) mixed
venous blood returns to the right heart with a lower oxygen saturation
(SvO2) and higher CO2 level reflective of metabolism at the tissue level.
D) expired end-tidal CO2 (ETCO2) correlates with pulmonary blood flow,
which in turn reflects adequate cardiac output.

5:1); therefore, the tissue oxygen extraction ratio (ER) is about 20-25%
(Figure 1a). Partially deoxygenated blood returns to the heart through
the venous system. Clinically, we use oxygen saturations, rather than true
oxygen content, to monitor this relationship. Ideally, fully oxygenated ar-
terial blood is ejected from the left ventricle (SaO2 = 100%) and returns
to the right ventricle at about 75% saturation (SvO2 = 75%). If DO2 falls
relative to VO2, whether due to cardiogenic shock, hypovolemia, hemor-
rhage, hypoxemia, or other pathology, ER increases and SvO2 falls (Figure
1b). SvO2 also falls when tissue VO2 rises relative to DO2 (e.g. sepsis, shiv-
ering, hyperthermia, etc.). Conversely, a high SvO2 is potentially ominous
and can be seen when oxygenated blood is shunted past tissue capillary

Chapter 7 127
Figure 1b: Example of poor oxygen delivery. A) an example of cardiogenic
shock with low cardiac output (CO) and cardiogenic pulmonary edema. B)
the arterial blood gas reflects an example of acute hypoxemic respiratory
failure with mild respiratory alkalosis. C) as a result of poor oxygen deliv-
ery (DO2), the extraction ratio (ER) is higher at a stable rate of oxygen
consumption (VO2). D) mixed venous blood returns to the right heart
with a severely decreased oxygen saturation (SvO2) and much higher CO2
level, reflecting tissue level perfusion abnormalities seen in shock with low
cardiac output. E) expired end-tidal CO2 (ETCO2) is low despite high PCO2
in pulmonary arterial blood, indicated lower pulmonary blood flow and
increased physiologic deadspace.

networks (e.g. microcirculatory thrombi) or oxygen cannot be utilized by

tissues (e.g. mitochondrial dysfunction or cell death). Below a DO2:VO2
of 2:1, anaerobic metabolism ensues and serum lactate rises. Low SvO2 is
a more sensitive indicator of inadequate DO2 than elevated lactate, which
represents a late finding.

Hemodynamics and Tissue Perfusion

DO2 depends upon cardiac output (CO) and arterial oxygen content (CaO2);
CaO2 is determined by hemoglobin concentration (Hgb), and arterial hemo-
globin saturation (SaO2) (Figure 2). Cardiac output is equal to heart rate

128 ECPR and Resuscitative ECMO

Figure 2: Determinants of oxygen delivery and mean arterial pressure. V-A
ECMO improves arterial pressure and oxygen delivery by partially replacing
cardiac output and improving oxygenation. V-V ECMO improves oxygen-
ation but does not directly augment hemodynamics. V-A, venoarterial; V-V,
venovenous; MAP, mean arterial pressure; DO2, oxygen delivery; CVP, cen-
tral venous pressure; SVR, systemic vascular resistance; CO, cardiac output;
CaO2, oxygen content of arterial blood; HR, heart rate; SV, stroke volume;
Hgb, hemoglobin concentration; SaO2; arterial oxygen saturation.

(HR) x stroke volume (SV); SV in turn is affected by preload, afterload,

and contractility. Briefly, preload describes the stretching of myocardium
during diastole, which up to a certain point increases CO as described by
the Frank-Starling curve. Clinically, venous return (VR) is the source of car-
diac preload. VR depends upon a pressure gradient driving blood flow from
large capacity veins into the receiving cardiac chambers (large peripheral
veins to the right heart; pulmonary veins to the left heart) (Figure 3).1 VR,
and thus CO, will drop if hypovolemia or loss of vascular tone (e.g. sep-
sis, anaphylaxis, etc.) occurs, or if intrathoracic or intracardiac pressures
rise (e.g. positive pressure ventilation, air trapping, tension pneumotho-
rax, cardiac tamponade, massive pulmonary embolism, etc). Afterload
describes the wall stress on myocardium during systole. High afterload
negatively impacts CO. Clinically, afterload correlates with the pressure
and resistance downstream of the respective ventricle. Pulmonary hyper-
tension or a large pulmonary embolism increases RV afterload, as does
positive-pressure ventilation due to increasing pulmonary vascular resis-
tance. Arterial hypertension or aortic stenosis increase LV afterload, as

Chapter 7 129
Figure 3: Determinants of cardiac performance. A) the source of venous
return is blood flowing from post-capillary venules and large capacitance
veins towards the vena cave. This venous blood is pressurized (mean
systemic pressure) when blood volume and venous tone are adequate. B)
venous blood returns to the right atrium as long as mean systemic pressure
is higher than central venous pressure (CVP, equal to right atrial pressure
[RAP]). The right ventricle (RV) ejects blood into the pulmonary artery
(PA), working against afterload in the form of pulmonary vascular resis-
tance (PVR). C) PVR increases during hypoxemic pulmonary vasoconstric-
tion, large pulmonary embolism, high intrathoracic pressures, and other
scenarios. D) blood returns to the left heart via the pulmonary veins (PV)
to the left atrium (LA), through the mitral valve to the left ventricle (LV),
where it is ejected into the aorta proportionate to LV contractility. E) LV
ejection is impeded by afterload in the form of aortic stenosis, high sys-
temic vascular resistance (SVR) or high systemic arterial blood pressure.

does vigorous spontaneous breathing with negative intrathoracic pressure.

Note that spontaneous versus positive pressure ventilation have opposite
effects on the RV versus the LV.2,3 Finally, the intrinsic contractility of the
myocardium depends on many factors, including coronary perfusion, ox-
ygen saturation, energy substrate, intracellular pH, calcium, and others.
In addition to DO2, downstream organs depend upon a perfusing blood
pressure. Organ perfusion pressure is the difference between systemic

130 ECPR and Resuscitative ECMO

Figure 4: Determinants of vascular tone. Vasodilation occurs when vas-
cular smooth muscle relaxes in response to a number of different signals
(left). The term vasoplegia is used when pathologic loss of systemic
vascular resistance and arterial hypotension occur. Vasoconstriction occurs
due to smooth muscle contraction in response to the agents listed (right),
which can be used therapeutically at the bedside (right). Methylene blue
and cyanocobalamine cause vasoconstriction by scavenging endothelial
nitric oxide. α1, adrenergic receptor; V1, vascular vasopressin receptor;
AGIIr, angiotensin II receptor.

blood pressure and the hydrostatic back-pressure inside the organ. Mean
arterial pressure (MAP) is the product of CO x systemic vascular resistance
(SVR) plus a small contribution from central venous pressure (CVP) (Fig-
ure 2). The loss of SVR, or “vasoplegia,” can occur in multiple conditions
such as septic shock, anaphylaxis, systemic inflammation, use of antihy-
pertensive medications, and more. Vasopressors, nitric oxide scavengers,
and other adjuncts are used to try to increase vascular tone (Figure 4). 4,5
While ECMO can replace CO and augment SaO2, it cannot improve low
SVR. ECMO does not help patients whose primary problem is vasoplegic shock.

Chapter 7 131
Figure 5: Schematic of peripheral femoral V-A ECMO. A) venous blood is
drained from the superior vena cava, right atrium, and inferior vena cava
into a venous cannula with multiple side holes. B) the pump head gener-
ates negative pressure pulling blood into the inlet where the pump inlet
pressure (PINLET) and pre-membrane O2 saturation (SPREO2) are measured,
then pumps blood towards the membrane lung. C) pressurized “postpump/
pre-membrane” blood enters the oxygenator where the pre-membrane
pressure (PPRE) is measured. Diffusion of oxygen and carbon dioxide in the
membrane lung results in arterialization of blood. D) “post-membrane”
blood then exits the membrane lung where the post-membrane pressure
(PPOST) and post-membrane O2 saturation (SPOSTO2) are measured. The
different between PPRE and PPOST is called the “delta P” and reflects the resis-
tance in the membrane lung. E) arterialized blood returns to the body via
the shorter arterial cannula and is pumped retrograde up the aorta towards
the heart, perfusing the branch vessels. F) in the absence of native cardiac
activity, ECMO blood reaches the aortic arch, cerebral vessels, and coro-
nary arteries. G) bronchial and thebesian venous blood returns to the left
ventricle, leading to eventual overdistension and stasis in the left heart.

132 ECPR and Resuscitative ECMO

V-A ECMO
Peripheral V-A ECMO is the most common configuration used urgently
for cardiovascular collapse, because it can be rapidly deployed in nearly
any patient location, by surgeons and non-surgeons alike. In the operat-
ing room with additional time and resources, a cardiothoracic surgeon
may instead centrally cannulate the right atrium and aortic root for V-A
ECMO, altering some oxygenation physiology but otherwise providing
similar hemodynamic support. For the purposes of illustration, we will
use a typical example of peripheral V-A ECMO cannulation via the com-
mon femoral vein and common femoral artery (Figure 5).
A long 21 or 25 Fr drainage cannula is inserted in the common femo-
ral vein with the tip ideally just inside the superior vena cava (SVC) with
side holes draining blood from the RA and much of the inferior vena cava
(IVC) as well. The two most important factors determining how well
blood drains into the circuit are patient preload (volume status and filling
pressure in the large veins and right atrium) and venous cannula resistance
(based mostly on diameter, as well as number of side holes). The pressure
inside the venous cannula and venous limb of the circuit (PINLET) is deeply
negative (approximately -50 to -100 mmHg); air bubbles can rush in and
de-prime the circuit if any disconnection occurs. If excessive negative
pressure is exerted on the blood here, gas can also bubble out of solution
(cavitation) and de-prime the circuit. Blood is drawn into the pump inlet,
pressurized by centrifugal force, and pushed through the outlet. Pressur-
ized blood (“post-pump/pre-membrane”) travels through tubing into the
membrane lung, through which fresh gas flows and oxygen/carbon dioxide
are exchanged. Blood exits the membrane lung (“post-membrane”) and
is pushed through a shorter arterial cannula, 15-20 Fr depending on the
size of the patient, inserted in the common femoral artery. Because of the
narrow diameter of the arterial cannula, post-pump pressure (PPOST) in
the circuit can be anywhere from 150-300 mmHg or more, but this pres-
sure is dispersed quickly at the tip of the cannula. Any inadvertent circuit
disconnection post-pump can result in rapid exsanguination of the pa-
tient. Near the hub of the arterial cannula is a luer side-port which can be
connected via ¼" tubing to a reperfusion catheter placed antegrade into
the superficial femoral artery or retrograde into the posterior tibial artery.
This prevents limb ischemia, as even a modestly sized femoral arterial
cannula can completely obstruct distal blood flow.

Chapter 7 133
Figure 6: Centrifugal ECMO pump. Venous drainage from the patient en-
ters the pump inlet along the top of the pump head. Pressure is generated
by centrifugal force as the rotors spin and push the blood to the periphery
and out the outlet towards the oxygenator. The pump head is powered by
the motor magnetically to reduce heat and friction. Pump speed (RPM)
is set on the ECMO console. (Centrimag pump, Abbot Cardiovascular,
Chicago, IL).

ECMO Pump
Modern centrifugal ECMO pumps use disposable pump heads that are
magnetically driven by the motor, minimizing heat and friction. Because
there is no physical connection between pump and motor, the centrifu-
gal pump is preload-dependent and afterload-sensitive (Figure 6). Pump
flow is dependent upon adequate venous drainage, adequate RPM, and
low resistance; just as native cardiac output is dependent upon adequate
venous return, normal HR and SV, and appropriate afterload, respectively.
Hypovolemic patients have low right sided filling pressures. Consequently
drainage (preload) through the venous cannula will be inadequate. Flows
will drop regardless of pump RPM; in fact, turning up the RPM in a low
preload situation may cause worsening suction and turbulence along the
venous cannula. This causes the ECMO tubing to physically shake and
bounce (“chugging” or “chatter”). The treatment is volume resuscitation.

134 ECPR and Resuscitative ECMO

Figure 7: ECMO membrane oxygenator. Venous, deoxygenated blood
enters from the right side into the oxygenator inlet. Sweep gas (blue
arrow, green tube) brings oxygen into a vast network of tiny hollow poly-
methylpentene fibers. The RBCs travel a path among and between these
fibers, across which O2 and CO2 diffuse. Warm water from a separate heat
exchanger (orange arrows) travels through an internal heating element to
maintain blood temperature. (Quadrox oxygenator, Getinge Group, Gote-
borg, Sweden).

Venous drainage may also be threatened if the patient is coughing or bear-

ing down, or if pressures outside the heart and vena cava rise (e.g. tension
pneumothorax, cardiac tamponade, abdominal compartment syndrome,
etc.). On the other side of the circuit, pump afterload increases if the
membrane lung has substantial thrombosis, the arterial cannula is kinked,
or the patient is severely hypertensive. High afterload causes ECMO flow
to drop. Conversely, a clue to the development of vasoplegia, possibly
septic shock, is that ECMO flow slowly rises despite a constant RPM be-
cause afterload is dropping. While troubleshooting the circuit is covered
in other chapters, generally the first maneuver to address dropping ECMO
flow is to give the patient volume while systematically addressing all possi-
ble causes, rather than reflexively increasing pump RPM.

Chapter 7 135
Membrane Lung
The ECMO membrane lung is an artificial membrane lung made of tiny
hollow polymethylpentene (PMP) fibers with an immense total surface
area (Figure 7). Blood flows among and between this network of fibers
while fresh gas flows through the fibers; oxygen and carbon dioxide dif-
fuse across the PMP much like the alveolar capillary interface in the native
lung. Increasing the gas flow (called “sweep” gas) is analogous to increas-
ing the minute ventilation on a ventilator, which increases CO2 elim-
ination. The fraction of delivered oxygen percent (FdO2) of the sweep
gas can be controlled using a blender. Modern membrane lungs are so
efficient that the PO2 of ECMO return blood is usually >500 mmHg at a
blender FdO2 of 1.0.
The primary purpose of V-A ECMO is to replace native cardiac output
in order to sustain DO2 and MAP in a patient at high risk of death. V-A
ECMO also decongests the right heart, replaces native ventilation, and
supports oxygenation. The most important ECMO variable is pump flow,
which must be adequate to support the patient and rest the heart, usually
above 50 mL/kg/minute. Three big concepts make peripheral V-A ECMO
easier to understand: 1) during periods where no perceptible native car-
diac output is generated (e.g. cardiac arrest, non-perfusing arrhythmia,
profound hypokinesis, etc.) V-A ECMO replaces heart and lung function
entirely, generating a MAP and providing DO2 to all end organs; 2) if
cardiac performance improves, native blood is ejected from the LV and
competes with ECMO blood flowing retrograde up the aorta, combining
to generate a MAP and DO2; and 3) native blood ejected from the LV is
oxygenated by the native lungs, not the ECMO circuit.

Clinical Example
Let’s look at an example of the physiologic changes occurring in a pa-
tient cannulated during cardiac arrest. Upon initiation of V-A ECMO,
the pump RPM are gradually increased to achieve ~4 L/min of flow, and
chest compressions are stopped. The sweep gas is set at 2 L/min, FdO2
1.0. The patient likely has no organized cardiac activity or contractility.
ECMO flow completely replaces native CO, generating a MAP and per-
fusing end organs, including generating coronary perfusion pressure and
cerebral perfusion pressure. An arterial pressure tracing will show a flat

136 ECPR and Resuscitative ECMO

waveform MAP that increases with higher ECMO flow. An arterial blood
gas (ABG) drawn from any location (right radial, left radial, femoral, or
post-oxygenator ECMO circuit) will show the same data—a rising pH,
clearing PCO2, and a PO2 of >300 mmHg. Blender FdO2 may be weaned
down to avoid cerebral hyperoxia. Sweep is gradually increased with care
not to hyperventilate the patient. At this moment, the mechanical venti-
lator is essentially irrelevant, and settings should be minimized to avoid
lung injury. The CVP, PA pressures, and end tidal CO2 (ETCO2, reflective
of pulmonary blood flow and native cardiac output) will all be lower as
blood is drained from the right heart and diverted through the ECMO
circuit. Echocardiography would show that the aortic valve does not open.
As expected, SvO2 would gradually rise and is still an excellent marker of
overall adequacy of DO2; there may be a variable increase in serum lactate
due to a washout of stagnant peripheral blood and/or liver and kidney
injury. Exogenous epinephrine and other vasoactive drugs given during
the resuscitation may result in a transient surge of hypertension after ini-
tiating ECMO; however, this may be followed by a drop in MAP due to
a systemic inflammatory response induced by ECMO itself, and a fluid
bolus and initiation of vasopressors may be necessary.
Within several minutes, the characteristic post-arrest acidemia, hy-
percarbia, and hyperkalemia should clear, and the heart may now respond
to cardioversion or defibrillation, or it may spontaneously convert into an
organized rhythm. If the heart is strong enough to generate a stroke vol-
ume, pulsatility will now be seen on the arterial tracing, and echocardiog-
raphy will show corresponding aortic valve opening. Aortic valve opening
and LV ejection are critically important to prevent gradual overdistension
of the LV, which can perpetuate myocardial ischemia and cause worsen-
ing pulmonary edema, even pulmonary hemorrhage, if not addressed. 6
Theoretically the aortic valve, root, and left ventricular outflow tract can
also form clot if left stagnant. If PA pressures are monitored, a gradually
rising PA diastolic and pulmonary capillary wedge pressure would suggest
progressive LV distension.7 About 10-20 mmHg of pulse pressure is suffi-
cient to ensure that the LV is ejecting adequately. Peripheral V-A ECMO,
while restoring tissue perfusion, does not unload the LV well and actu-
ally worsens LV afterload8; with high flows and a rising MAP, native LV
pulsatility may vanish. Maintaining a perfusing rhythm via defibrillation,
cardioversion, and antiarrhythmics, followed by the use of inotropes, will
help to restore pulsatility. If pulsatility cannot be restored after the initial

Chapter 7 137
Figure 8: Peripheral femoral V-A ECMO with return of native cardiac
function and normal native pulmonary function. A) two parallel circulations
occur on V-A ECMO with native cardiac output. Venous blood is partially
drained through the venous cannula, while the remainder is ejected by
the right ventricle into the pulmonary circulation. B) blood returns to the
left heart and is ejected by the left ventricle into the aortic arch, where it
perfuses the coronary arteries and cerebral vessels. C) antegrade native
ejected blood meets retrograde ECMO return blood in a “mixing cloud” in
the aorta.

resuscitation with modest doses of inotropes, an LV venting strategy must

be employed in consultation with interventional cardiology and/or car-
diac surgery. LV venting is discussed elsewhere, but percutaneous devices
such as intra-aortic balloon counterpulsation, a microaxial Impella pump
(Abiomed, Danvers, MA), or TandemHeart device (LivaNova, London,
UK), or open surgical venting, are highly effective.7

Dual Circulations
Once native LV ejection returns (or LV venting is established), the pa-
tient will have established two competing parallel circulations—native
blood flowing through the right ventricle, lungs, left ventricle, and ante-

138 ECPR and Resuscitative ECMO

Figure 9: Peripheral femoral V-A ECMO with return of native cardiac func-
tion and impaired native pulmonary function. A) in the case of a large V/Q
mismatch in the native lung (e.g. aspiration, edema, hemorrhage, pneumo-
nia, acute respiratory distress syndrome, etc.), blood returning to the left
heart is poorly oxygenated. B) hypoxemic blood is ejected into the aortic
arch, leading to coronary ischemia and cerebral hypoxia. A right upper
extremity arterial blood gas is critical to recognize this phenomenon. C)
returning ECMO blood moving retrograde up the aorta continues to supply
the lower body with oxygen. This is termed “North-South” syndrome. In-
creasing ECMO flow (and reducing antegrade LV venting, if applicable) will
drain the right heart and reduce native antegrade blood ejection, pushing
the mixing cloud towards the aortic valve to mitigate this upper body hy-
poxemia. At the same time, optimizing mechanical ventilation will improve
the oxygen saturation of native blood ejected by the left ventricle.

grade into the aortic root, and ECMO blood flowing through the circuit
and retrograde up the aorta towards the arch (Figure 8). The two circu-
lations meet in a “mixing point” in the aorta; the location of this mixing
point depends on ECMO flow versus native cardiac output. Blood ejected
from the LV has been oxygenated only by the native lungs. If the patient
has suffered significant pulmonary edema, aspiration, lung contusion, or
other physiologic shunt, the native blood in the aortic root may be hypox-

Chapter 7 139
emic (Figure 9). An ABG sampled from the right upper extremity may
show a startlingly lower PaO2 than one sampled at the left radial or fem-
oral artery, which is diagnostic of “North-South Syndrome” or “Differen-
tial Hypoxemia.” The right upper extremity is the preferred location to
sample an ABG and measure pulse oximetry, because this most accurately
reflects oxygenation of the brain and coronary arteries.
Regional cerebral and coronary hypoxemia due to North-South syn-
drome must be actively managed. First, take steps to optimize pulmonary
function by titrating mechanical ventilation to recruit the lungs and con-
sidering diuresis, antibiotics, and/or therapeutic bronchoscopy to treat
any underlying lung problem to improve the oxygen saturation of blood
ejected from the LV. If unsuccessful, the patient may require conversion
to venoarteriaovenous (V-AV) ECMO (with a return cannula added in the
right internal jugular vein) or an axillary/subclavian/innominate return
cannula or central cannulation. If cerebral hypoxemia is severe tempo-
rizing measures can be pursued to move the mixing point more proximal
towards the aortic valve. These measures include maximizing ECMO flow
to minimize native blood flow through the heart and lungs, and limiting
inotrope doses and antegrade LV vent flow (i.e. Impella flow) to the min-
imum necessary to prevent LV distension. These maneuvers reduce hy-
poxemic blood ejection into the aortic root.
After ECMO initiation, detailed monitoring ensures patient safety
and guide management.9 An arterial line is mandatory, ideally in the right
upper extremity, for MAP, pulse pressure, and blood gas monitoring. Ad-
equate flow on V-A ECMO is at least 50 mL/kg/min of ideal body weight,
which is monitored on a flow probe attached to the ECMO circuit. He-
moglobin is generally kept in the 8-10 g/dL range. The adequacy of DO2 is
assessed by serial monitoring of SvO2 either by pulmonary artery catheter
or sampling the post-pump/pre-oxygenator venous line of the ECMO cir-
cuit, as well as trending lactate levels. The optimal MAP remains unde-
fined, but 65-75 mmHg is reasonable, possibly higher if there is evidence
of malperfusion (capillary refill, skin temperature, urine output, mental
status, and lactate levels), understanding that this may lead to further
LV dysfunction. Near-infrared spectroscopy (NIRS) applied to the bilat-
eral frontal skull may assist in trending cerebral oxygenation and alert
the provider to a sudden drop in brain perfusion, although the only evi-
dence basis is extrapolated from cardiac surgery. NIRS can also be used
to monitor for ischemic limb complications distal to the arterial cannula.

140 ECPR and Resuscitative ECMO

Essential laboratory values should be checked often post-cannulation, in-
cluding blood counts, basic chemistries, coagulation profile, lactate, SvO2,
and arterial blood gas at a minimum. Table 1 summarizes key monitoring
variables and relationships.
Over the next several hours, the effects of ECMO on cerebral perfu-
sion, renal function, coagulation, and systemic inflammation also require
close attention. Neurologic complications are common with ECMO sup-
port, including hypoxic ischemic brain injury, ischemia-reperfusion injury
post arrest, acute ischemic stroke, embolic stroke, cerebral air embolism,
and intracranial hemorrhage.10 Management of ECMO flow and MAP di-
rectly affects cerebral perfusion pressure, cerebral blood flow, and brain
oxygen delivery, particularly if cerebral autoregulation has been disrupted
by brain injury. Clinical exam is difficult in these patients; therefore, mul-
timodal neurologic monitoring including continuous electroencepha-
lography (EEG) and brain near-infrared spectroscopy (NIRS) has been
advocated to guide the hemodynamic optimization of cerebral blood
flow on ECMO.11 Also note that the plasma concentrations of fentanyl,
propofol, dexmedetomidine, and midazolam are significantly reduced on
ECMO; increased dosing, or alternatives such as hydromorphone and ket-
amine, may be required.10
Acute kidney injury (AKI) occurs commonly, and renal replacement
may be required in almost two-thirds of cases.12 It is unclear if ECMO it-
self specifically contributes to AKI via mechanisms such as non-pulsatile
blood flow, inflammation, ischemia-reperfusion injury, or hemolysis, or if
it is merely associated with severe disease.13 Anecdotally, urine output and
serum creatinine do improve if ECMO restores perfusion and oxygenation
early in AKI. It is prudent to maintain a normal or high-normal MAP and
involve a nephrologist early if oliguria, hyperkalemia, or metabolic acido-
sis are concerning in the hours following cannulation. The appearance of
new hyperkalemia after cannulation should prompt an evaluation for limb
ischemia and/or compartment syndrome. The unexplained appearance of
dark pink urine should prompt an evaluation for hemolysis.

Coagulation
ECMO induces myriad alterations in the coagulation system related
to blood interaction with foreign material.14 Fibrinogen, von Wille-
brand factor (vWF), and platelets adhere to and are activated by these

Chapter 7 141
 Variable (Goal) Significance Comments

142
ECMO flow >50 mL/kg/ Major determinant of DO2 Bolus fluids if unable to reach flow goals initially
min or 2.4 L/min * BSA Titrate flow based on needs of native heart using
(m2) or ~3 L/min for hemodynamic and echocardiographic data
normal sized adult
MAP (65-75 mm Hg) Major determinant of organ perfusion and cerebral blood flow. Vasopressors and a rapidly titratable
Hypertension may increase risk of hemorrhage antihypertensive drip should be ready at the
bedside
CVP (trend) High CVP may indicate inadequate drainage of the right heart, No recommendation for absolute numbers
pneumothorax, cardiac tamponade, high airway pressures, or
abdominal compartment syndrome
Low CVP may correlate with hypovolemia or vasoplegia, and the
risk of inadequate venous drainage to the ECMO pump
PA pressures (trend) Rising PA diastolic pressure may indicate LV overdistension and No recommendation for absolute numbers
risk of pulmonary edema
Low PA pressures and low PA pulsatility indicate a well-drained
right heart and/or poor RV function
SvO2 (>60%) Indicates adequacy of DO2/VO2 relationship Persistently low SvO2 should prompt optimization
of DO2 parameters, and evaluation for shivering,

ECPR and Resuscitative ECMO

fever, agitation, seizure, or high work of breathing
Lactate (trend) Marker of restoration of adequate DO2 May see an initial rise in lactate due to tissue
wash-out; may also be affected by liver and kidney
injury
Pulse Pressure Pulsatility indicates native LV ejection Inotropes may be necessary to maintain pulsatility.
(10-20 mm Hg) Lack of pulsatility indicates need for LV venting
strategy
 Variable (Goal) Significance Comments
LV vent flows (~1 L/min) Ensures LV decompression to avoid myocardial ischemia, Available on the console of the venting system, or a
intracardiac thrombosis, and pulmonary edema dedicated flow probe
Lower extremity Ensures perfusion of the extremities distal to an arterial cannula Dedicated flow probe(s) on the circuit tubing
reperfusion flow (200- or sheath connected to reperfusion catheter(s)
500 mL/min)
Brain and lower Dropping (regional cerebral oxygenation) rSO2 may indicate No recommendation for absolute numbers
extremity NIRS (trend) poor perfusion of the monitored tissue
SaO2 (95-98%) Major determinant of DO2 Blood gas PaO2 is physiologically irrelevant
pH (7.35-7.45) Cardiac performance and hemodynamics may be sensitive to Use sweep gas and sodium bicarbonate to correct
acidosis ongoing respiratory and metabolic acidosis,
respectively
PaCO2 (35-45 mm Hg) Contributes to acidosis; also contributes to cerebral Do not rapidly correct or overcorrect CO2
vasoconstriction/vasodilation
Hgb (8-10 g/dL) Major determinant of DO2 Hgb goal may be raised if persistently low DO2

Chapter 7
Minor determinant of proper clot formation
Platelets (unclear goal) Thrombocytopenia may be a risk for hemorrhage Be aware of possible HITT
aPTT, ACT, and/or Maintain therapeutic anticoagulation levels Goals and protocols will be institution specific
Anti-Xa level

Table 1: Monitoring the V-A ECMO patient. DO2, oxygen delivery; MAP, mean arterial blood pressure; SBP, systolic blood pres-
sure; CVP, central venous pressure; PA, pulmonary artery; LV, left ventricle; RV, right ventricle; SvO2, mixed venous oxygen
saturation; VO2, tissue oxygen consumption; NIRS, near-infrared spectroscopy; rSO2, regional tissue oxygen saturation; SaO2,
arterial oxygen saturation; PaO2, partial pressure of oxygen in arterial blood; PaCO2, partial pressure of carbon dioxide in arte-
rial blood; Hgb, hemoglobin concentration; HITT, heparin-induced thrombocytopenia with thrombosis; aPTT, activated partial

143
thromboplastin time; ACT, activated clotting time.
non-endothelial surfaces, leading to clot formation via the extrinsic path-
way involving tissue factor, factor VII, fibrinogen conversion to fibrin, and
thrombin deposition. Thrombocytopenia and hypofibrinogenemia, as well
as low factor VIII and vWF ensue. Concomitantly, fibrinolytic pathways
are also activated, leading to elevated D-dimer levels. The overall picture
is akin to a low-level disseminated intravascular coagulopathy (DIC),
with excessive bleeding and clotting both possible. Anticoagulation with
unfractionated heparin (UFH) aborts thrombin deposition by catalyzing
antithrombin (ATIII) and inhibiting factor X and fibrinogen. Not only
does UFH prevent thrombosis within the circuit, but by curtailing the on-
going coagulation and fibrinolytic cascades, UFH likely prevents excessive
bleeding as well. Anticoagulation (in the absence of strong contraindica-
tions) and comprehensive coagulation monitoring are essential.
The ECMO provider must also be aware of heparin induced thrombo-
cytopenia with thrombosis (HITT), a catastrophic clotting disorder due
to heparin induced platelet antibodies which activate platelets upon ex-
posure to heparin.15 Although it remains relatively uncommon and most
often presents with thrombocytopenia several days after heparinization,
it could develop rapidly upon heparinization for ECMO cannulation if the
patient has been exposed to heparin recently (e.g. cardiac catheterization,
cardiac surgery, prophylaxis, etc.). HITT can cause rampant widespread
thrombosis, threatening life, limbs, and the circuit. The treatment is elimi-
nation of all heparin exposure (including in IV flushes) and administration
of direct thrombin inhibitors such as bivalirudin or argatroban. The circuit
may need to be changed out rapidly if the tubing or oxygenator clot off.
Along with activation of platelets and the coagulation cascade, blood
exposure to foreign ECMO circuit surfaces also activates the inflammatory
response, including the complement system, kallikrein and bradykinin,
neutrophil activation, release of pro-inflammatory cytokines, and endo-
thelial activation.16 There is potential to develop a systemic inflammatory
response syndrome (SIRS) similar to sepsis, including capillary leak, third
spacing of plasma fluid, and vasoplegia early after cannulation. This may
require additional fluids, vasopressors, or even stress dose steroids, al-
though evidence-based recommendations for the latter cannot be made.
In summary, the decision to cannulate and the subsequent manage-
ment of V-A ECMO requires a detailed understanding of the patient’s
pre-ECMO physiology. Cannulation and establishing flow are just the
beginning. The interactions between the ECMO circuit and the patient’s

144 ECPR and Resuscitative ECMO

body are complex, requiring active management. Comprehensive, sys-
tematic patient and circuit monitoring are needed to ensure safety, de-
liver adequate organ support, and detect complications.

References
1. Funk DJ, Jacobsohn E, Kumar A. Role of the venous return in critical
illness and shock—Part I: physiology. Crit Care Med. 2013;41:255-262
2. Piazza G, Goldhaber SZ. The acutely decompensated right
ventricle: pathways for diagnosis and management. Chest.
2005;128(3):1836-1852
3. Guyatt GH. Positive pressure ventilation as a mechanism of reduction of
left ventricular afterload. J Canad Med Assoc. 1982;126:1310-1312
4. Omar S, Zedan A, Nugent k. Cardiac vasoplegia syndrome: pathophysi-
ology, risk factors, treatment. Am J Med Sci. 2015;349(1):80-88
5. Sharawy N. Vasoplegia in septic shock: do we really fight the right enemy?
J Crit Care. 2014;29:83-87
6. Leopald R, Florchinger B, Schopka, S, et al. Cardiac decompression
on extracorporeal life support: a review and discussion of the literature.
ASAIO. 2013;59(6):547-553
7. Donker DW, Brodie D, Henriques JPS, et al. Left ventricular un-
loading during veno-arterial ECMO: a simulation study. ASAIO.
2019;65(2):11-20
8. Thiele H, Ohman EM, de Waha-Thiele S, et al. Management of car-
diogenic shock complicating myocardial infarction: an update 2019. Eur
Heart J. 2019;40(32):2671-2683
9. Bhatia M, Katz J. Contemporary comprehensive monitoring of
veno-arterial extracorporeal membrane oxygenation patients. Can J
Card. 2020;36:291-299
10. Cho SM, Farrokh S, Whitman G, et al. Neurocritical care for extracorpo-
real membrane oxygenation patients. Crit Care Med. 2019;47:1773-1781
11. Zanatta P, Bosco E, Forti A, et al. Neurologic monitoring during
ECMO. In: Sangalli F, Patroniti N, Pesenti A, eds. ECMO- Extracorpo-
real Life Support in Adults. 1st ed. Milan, Italy: Springer; 2014:389-399
12. Kielstein JT, Heiden AM, Beutel G, et al. Renal function and survival
in 200 patients undergoing ECMO therapy. Neph Dialysis Transplant.
2013;28(1):86-90

Chapter 7 145
13. Kilburn DJ, Shekar K, Fraser JF. The complex relationship of extra-
corporeal membrane oxygenation and acute kidney injury: causation or
association? Biomed Research International. 2016:1-14
14. Mulder MMG, Fawzy I, Lance MD. ECMO and anticoagulation: a com-
prehensive review. Neth J Crit Care. 2017;26(1):6-13
15. Bain J, Flannery AH, Flynn J, Dager W. Heparin induced thrombocyto-
penia with mechanical circulatory support devices: review of the litera-
ture and management considerations. 2017. J Thromb Thrombolysis.
2017;44:76-87
16. Millar JE, Fanning JP, McDonald CI, et al. The inflammatory response
to extracorporeal membrane oxygenation (ECMO): a review of the
pathophysiology. Crit Care. 2016;20(387):1-10
CHAPTER 8
ECMO Initiation

Rory Spiegel • Scott Weingart

Introduction
Initiation of extracorporeal membrane oxygenation (ECMO) in-
volves taking a fully cannulated patient with a primed ECMO circuit to full
ECMO flow. In this chapter, we will explore how best to achieve this goal.

Circuit Preparation
During the cannulation process, a large sterile field should be estab-
lished. This field is imperative for the initiation process and should be
assessed for integrity after cannulation. On this sterile field, initiation is

Foundational Terms
ɋ Wet-to-wet connection—the process of deairing and connect-
ing circuit tubing to cannulas
ɋ Venous drainage insufficiency—blood flow limitation due to
inadequate pump preload
ɋ Arterial return obstruction—blood flow limitation due to
excessive pump afterload
ɋ Cavitation – high negative pressures cause air to form within a
given liquid

 147
 Initiation of ECMO
ɋ Connect cannulas to circuit via wet-to-wet connection
ɋ Pre-initiation “B A STAR” checklist (Figure 2)
ɋ Initiate flow
• Remove clamps (operator removes arterial return tubing
clamp last)
ɋ Optimizing ECMO support
• Increase RPMs until 3-4 L/min achieved
• If flows remain low consider:
ɦ Extravascular cannula
ɦ Venous drainage insufficiency (very negative venous drain-
age pressure, PINLET)
› Hypovolemia
› Bleeding
› Tension physiology in the chest or abdomen
› Pericardial tamponade
› Tube kinking
› Insufficiently large drain cannula
ɦ Arterial return obstruction (very high arterial return
pressure, PPOST)
› High vasopressor doses
› Tube kinking
› Insufficiently large return cannula
ɦ Large membrane lung clot (high delta P)
• If no flow consider:
ɦ Clamp still on tubing/cannula
ɦ Arterial bubble detector stopped pump
ɦ Pump airlock
ɋ Secure cannulas/tubing

commenced by cutting the primed circuit drainage (venous) and return

(arterial) limbs from the priming configuration.1 Cutting requires either
bandage scissors or sterile trauma shears as the circuit tubing resists the
intentions of daintier scissors. Choose the cut points to maximize cir-
cuit length. The more considerate manufacturers mark “cut here” indi-

148 ECPR and Resuscitative ECMO

cators on the tubing. Some circuits are pre-cut. With these, the tubing is
removed from the connection piece prior to initiation.
The drainage limb (typically marked with a blue stripe) must be in
line with the venous drainage cannula, and the return limb (typically
marked with a red stripe) must be in line with the arterial return can-
nula. There are numerous examples of return tubing being inadvertently
connected to the arterial cannula. This problem is difficult to correct, es-
pecially when the tubing has been fully hubbed on the cannulae. Double
checking that the appropriate cannulas are connected is prudent. A bolus
of heparin (50-100 units/kg) should have already been administered to
patients without contraindications to anticoagulation.2

Wet-to-Wet Connection
To minimize stagnant blood sitting in the cannula placed first, flush the
cannula with saline using a bulb or piston catheter syringe. If this has not
been done and there has been an extended period after the first cannula’s
placement, it is worthwhile to back-bleed the cannula by briefly releasing
the clamp. Beware of the trajectory of this evacuated blood prior to clamp
release. Also, remember any clot in the arterial cannula will travel imme-
diately into the patient’s arterial circulation. It is difficult to assess the
true incidence of stroke or coronary embolism at cannulation, but consid-
eration of back-bleeding the arterial cannula just prior to initiation may
be warranted.
When hooking up the circuit lines to the cannulae, all air must be
eliminated, this is referred to as the wet-to-wet connection. While this
can be successfully accomplished by doing the actual hook-up in a large
basin of saline (a wet-to-wet connection), more commonly it is accom-
plished by a continuous stream of saline from a syringe or pitcher. One
person is responsible for connecting the correct circuit limb to the end of
the cannula, while a second person pushes/pours saline over the connec-
tion point.

Chapter 8 149
 A C

B D

Figure 1: Wet-to-wet connection hand placement. a) Wet-to-wet connec-

tion of bottom edge, b) Wet-to-wet complete connection, c) Wet-to-wet
connection attached to first notch, d) Wet-to-wet connection attached to
second notch.

Steps of the Wet-to-Wet Connection

ɋ Clamps should be positioned 2-3" from the ends of the cannula/tubing
ɋ Hold the tubing and cannula in each hand with the ends emerging
from the thenar sides
ɋ Squeeze tip of circuit tubing between your thumb and second finger to
change the circular shape to an oval
ɋ Have your partner fill up the tubing and cannula ends until no air
remains
ɋ Insert the inferior edge of the hard plastic of the cannula into the
circuit tubing and hold them at 45⁰ to each other (Figure 1a)
ɋ Have your partner pour a continuous stream of saline into the juncture
ɋ Rotate both wrists to allow the full circumference of the hard-plastic
edge to enter the circuit tubing (Figure 1b)

150 ECPR and Resuscitative ECMO

ɋ Advance the tubing one notch up on the plastic connector (Figure
1c)
ɋ Check for air bubbles—if there any bubbles, separate the two and
start again
ɋ If no bubbles are seen, use a cranking motion to advance the circuit
tubing fully on to the connector
ɋ If bubbles are found after full advancement, the tubing will usually
require a longitudinal scalpel slit in order to be separated

Full video is available at reanimateconference.com/underwaterseal

Going on Pump
Prior to initiating mechanical support, it is important to complete a stan-
dardized checklist to ensure all steps necessary for safe initiation of ECMO
have been met. See Figure 2 for an example of an ECMO initiation check-
list. Figure 3 outlines the steps of initiating pump blood flow.
Initial pump speed should be set to neutral flow (usually 1500-2000
RPM on most pumps, though it may be much higher depending on man-
ufacturer).2 This prevents blood from the arterial system, which is un-
der positive pressure, from flowing out of the patient and reversing flow
through the centrifugal pump once clamps are removed.
Similar to pump flow, start sweep gas flow low and slowly, increase
as needed after serial blood gases. This prevents rapid PCO2 normaliza-
tion. The eventual goal should be to achieve a PaCO2 of 40-50 mm Hg.
Start sweep gas flow rate at 2 L/min and FdO2 at 100%. Ideally this should
achieve a systemic oxygen saturation of greater than 95%. FdO2 may be
titrated later, to achieve desired systemic oxygen tension. 2
If a heat exchanger is present on the ECMO machine, the circuit’s
temperature control should be set to achieve the targeted temperature
used when managing post-cardiac arrest patients. This typically ranges
between 33 C and 36 C. 2
Once the clinicians have ensured air bubbles are not present, all
clamps on the field should be removed while the ECMO specialist’s
clamps on the tubing close to the pump remain in place. Once speed is
increased, the venous drainage and then the arterial return tubing clamp
can slowly be removed from to initiate flow. The pump RPMs should be
ramped up over 10-15 seconds. A safety measure at this point is to have

Chapter 8 151
 ECMO Initiation Checklist “B-A-STAR.”
ɋ B: Final bubble check
ɋ A: Anticoagulation
ɋ S: Sweep gas flow set (2 L/min, FdO2 100%); O2 line hook-up
confirmed
ɋ T: Temperature set. If heat exchanger present
ɋ A: Arterial return cannula’s sideport cap is securely placed
ɋ R: RPMs set at neutral (1500-2000 on most pumps)

Figure 2

the cannulator with an open clamp positioned over the return tubing on
the field. 2,3 If a defective circuit set-up leads to the entrainment of air, the
return tubing can be clamped before it gets to the patient. Additionally, a
hand should be kept on the cannulas to prevent dislodgment until proper
securement.
As pump flow is ramped up, the color of the blood in the drainage and
return limbs of the circuit should be inspected to confirm proper cannula
placement. If the drainage limb is properly located in the venous circu-
lation, then the blood being withdrawn should be relatively dark. Once
the blood passes through the membrane lung, it should appear bright red
in the return limb. This dark/bright contrast is a critical step of ensuring
proper ECMO configuration.

Troubleshooting
Loss of Color Differentiation
If both circuit limbs remain dark, there is likely a problem with the mem-
brane lung or sweep gas. Clinicians should recheck whether their oxygen
source is correctly installed, gas is flowing appropriately, and FdO2 is set
at 100%. If both limbs have bright red blood, the first step is to ensure
the recirculation bridge on the circuit, if present, is closed.2,4 If so, both
cannulae are likely placed in the same side of the circulation (arterial/

152 ECPR and Resuscitative ECMO

arterial or venous/venous). An additional possibility of bright/bright cir-
cuit visualization is severe malfunction of oxygen extraction due to the
underlying disease process. Another cause of bright/bright loss of color
differentiation can be arteriovenous ECMO (draining from artery, return-
ing to vein).

Air Entrainment
Air bubbles on the arterial return limb of the circuit cannot be tolerated
due to risk of air embolism.2,4 Once eliminated, the positive pressure in
the post-pump circuit precludes entrainment of air. Conversely, due to the
negative pressure in the drainage limb, the venous side of the circuit can
be a continual source of air entrainment. Typically, this indicates a leak
in the system. Air bubbles on the drainage (venous) side of the circuit
may not lead to patient harm, as they may be removed by the membrane
lung. This should not be relied upon. If they are not addressed, they may
eventually reach the arterial side of the circuit causing potentially cata-
strophic outcomes or lead to larger amounts of air resulting in airlock of
the pump head and cessation of blood flow. (See Chapter 10 for details of
air entrainment)
Clinicians should closely inspect the venous drainage limb of the cir-
cuit for signs of a leak or locations where air can be introduced into the
circuit.4 Sources of air bubbles include:
ɋ Any point of connection or 3-way connector
ɋ Some cannulae (typically intended for arterial insertion) have a
built-in side-port which may have become unsecured and open to air
ɋ Damage to the circuit itself typically when securing the cannulae
ɋ The membrane lung, typically when the gas flow far exceeds blood
flow, has the potential for introducing air bubbles
ɋ Excessively negative venous pressures with chatter can lead to cav-
itation where gas is pulled out of solution resulting in air embolism
within the circuit.
ɋ Infusion pumps may inject small bubbles (generally not clinically
significant)

Chapter 8 153
 A D

B E

Figure 3. Steps of clamp removal to initiate ECMO blood flow. 3a) Re-
move clamps on sterile field. 3b) Location of clamps near pump controlled
by ECMO operator. 3c) Operator remove venous tubing clamp. 3d) Oper-
ator removes arterial tubing clamp and blood starts to drain from patient
filling venous tubing. 3e) Blood continues to move through circuit until
entire circuit is filled with blood.

Inadequate Blood Flow

For most adults, blood flow rates of 3-4 L/min are typically adequate to
maintain perfusion to the vital organs.2 This is predicated upon native
heart function eventually providing cardiac output augmentation and
the reduced metabolic needs due to targeted temperature management.

154 ECPR and Resuscitative ECMO

Pump speeds should be increased until target flows are achieved. If, de-
spite increasing pump speeds, target flow rates are not achieved, clini-
cians should evaluate for potential causes of low flow states in patients
on V-A ECMO.
In cases where the pump has persistently low or decreasing flow de-
spite increasing pump speeds, clinicians should consider the following
etiologies as the cause of the low-flow state4:
ɋ Extravascular Cannula
ɋ Venous drainage insufficiency (very negative venous drainage pres-
sure, PINLET)
• Hypovolemia
• Bleeding
• Tension physiology in the chest or abdomen
• Pericardial tamponade
• Tube kinking
• Insufficiently large drain cannula
ɋ Arterial return obstruction (very high arterial return pressure, PPOST)
• High vasopressor doses
• Tube kinking
• Insufficiently large return cannula
ɋ Large membrane lung clot (high delta P)

In cases where no blood flow is occurring, consider the following po-

tential causes:
ɋ Clamp still on tubing/cannula
ɋ Arterial bubble detector stopped pump
ɋ Pump airlock

Once proper cannula placement has been confirmed, the appropriate

color blood is visualized in drainage and return circuit limbs, and target
flow rates have been achieved, manual chest compressions can be halted
at this time. Importantly, flow rates are not an adequate surrogate for a
perfusion pressure. An arterial line should be placed at this time if one is
not already present. If the patient remains in ventricular fibrillation or
pulseless ventricular tachycardia, defibrillation may be attempted after an
adequate MAP has been for at least 3 minutes. If unsuccessful, further
defibrillation attempts may be futile until PCI.

Chapter 8 155
Securement
Cannulae can be secured using either sutures or commercial adhesive
devices. Sutures are commonly used to secure cannulae in place. Typi-
cally, at least four securement points with thick suture are used down the
length of the visible cannula and circuit tubing.5
An alternative method to secure the cannula is to use an adhesive se-
curement device. Designed to hold the cannula in place without further
skin punctures, the theoretical advantage is to decrease sources of bleeding
and infection. 6 These are commercially made devices specific for large can-
nula securement and are generally more costly than the traditional suture
approach.7 Most centers using this method use at least 2 devices per can-
nula.5
No high-quality data exist suggesting one method is superior. What-
ever approach is used, it should remain consistent throughout each in-
stitution. In order to prevent cannula dislodgement, a checklist should
be designed and implemented to ensure proper securement of cannulae
prior to moving the patients.

Conclusion
Transitioning a patient in cardiac arrest onto an ECMO circuit is a compli-
cated process that requires an experienced team well versed in the intri-
cacies of the procedure. To ensure consistency with this tenuous stage in
ECPR it is important to have a standardized protocol which is known and
rehearsed by all team members participating in the resuscitation.

References
1. Makdisi G, Wang IW. Extra Corporeal Membrane Oxygenation
(ECMO) review of a lifesaving technology. J Thorac Dis. 2015;
7(7):E166-76.
2. ELSO Guidelines for Cardiopulmonary Extracorporeal Life Support
Extracorporeal Life Support Organization , Version 1.4 August 2017
Ann Arbor, MI, USA.
3. Shekar K, Mullany DV, Thomson B, Ziegenfuss M, Platts DG, Fraser
JF. Extracorporeal life support devices and strategies for manage-

156 ECPR and Resuscitative ECMO

 ment of acute cardiorespiratory failure in adult patients: a compre-
hensive review. Crit Care. 2014; 18(3):219.
4. Sidebotham D. Troubleshooting adult ECMO. J Extra Corpor Tech-
nol. 2011; 43(1):P27-32.
5. Bull T, Corley A, Lye I, Spooner AJ, Fraser JF. Cannula and circuit
management in peripheral extracorporeal membrane oxygen-
ation: An international survey of 45 countries PLoS ONE. 2019;
14(12):e0227248.
6. Allou N, Lo Pinto H, Persichini R, et al. Cannula-Related Infec-
tion in Patients Supported by Peripheral ECMO. ASAIO J. 2019;
65(2):180-186.
7. Inwood S. An exploration of the past, present and future of catheter
securement. British J Nursing. 2014; 23(Sup8):S26-S27.

Chapter 8 157
CHAPTER 9
ECPR Management of the First
Four Hours

Sacha Richardson • Julia Coull • Arne Diehl

Introduction
Venoarterial (V-A) extracorporeal membrane oxygenation (ECMO)
is a powerful therapy with complex interactions between support me-
chanics and patient physiology. The support needs to be tailored for each
patient to achieve adequate organ perfusion and gas exchange whilst pro-
viding the best chance for native cardiopulmonary recovery. This may re-
quire interventions in the immediate period after support is established,
whilst also identifying the underlying etiology of the arrest. Extracorporeal
cardiopulmonary resuscitation (ECPR) patients are particularly prone to
a range of complications from extended CPR times. These complications
should be actively excluded or, if identified, treated in a timely manner.

Foundational Terms
ɋ ScvO2 – reflects oxygen delivery vs. consumption. Low values
reflect inadequate oxygen delivery (<60%) and should prompt
strategies to increase blood flow (ECMO support or native heart)
ɋ Targeted temperature management – postcardiac arrest
patients may benefit from prevention of fevers and mild hypo-
thermia. ECMO is capable of creating a fixed temperature for the
patient with goals at 33-36˚C

 159
 This chapter outlines the management principles for the first 4 hours
to aid with the dual goals of stabilising the patient prior to transfer to
an ECMO credentialed ICU in addition to identifying and treating early
complications. Sadly, a proportion of patients will become unsupportable
during this period. Despite correct deployment of V-A ECMO, palliative
therapies may need to be instituted.

Establishing ECMO Support & Achieving

Hemodynamic Stability
Hemodynamic stability is a broad concept and in the instance of ECPR
encompasses factors relevant to both the patient and ECMO circuit. Both
systems must be envisioned as complementary to one another; concen-
trating all efforts on either the patient’s native circulation or the ECMO
circuit may lead to undesirable outcomes. Patient variables to optimise
include heart rate, rhythm, preload, left ventricle (LV) contractility, and
afterload. ECMO circuit hemodynamic variables include blood flow rate,
pump preload, and afterload. Balancing these factors should result in con-
sistent perfusion to the tissues and provide conditions for native cardiac
recovery. Although full cardiac recovery may not be an achievable target
in the four hours following ECPR, partial recovery of function should be
actively sought.

160 ECPR and Resuscitative ECMO

General Vital Signs Monitoring
The minimum requirements for general patient monitoring include con-
tinuous EKG, end-tidal CO2, peripheral oxygen saturations (SpO2) and
invasive blood pressure measurement. Until an arterial line is placed, non-
invasive measurement using a Doppler and a manual blood pressure cuff
may assist with initial vasopressor dosing. Peripheral oxygen saturation
monitors may be inaccurate in post ECPR patients due to a non-pulsatile
circulation or poor peripheral perfusion.

Arterial Line Placement and Blood Pressure Monitoring

Typically following ECPR, there is minimal or absent arterial pulsatility.
As such, an arterial line should be sited with ultrasound guidance as tac-
tile or anatomical landmark methods are less likely to be successful.
Right upper limb blood gas sampling gives the best indicator of native
PaO2 which may represent the cerebral oxygen delivery, depending on the
mixing point in the aorta between native circulation and V-A ECMO blood
flow. Therefore, we recommend a right upper limb arterial catheter (ra-
dial or brachial). See Chapter 7 for description of dual circulations and
differential hypoxemia.

Central Venous Access and Central Venous Pressure (CVP)

Monitoring
Central venous access is typically required for administration of multiple
infusions including vasoactive medications in ECPR patients. Central line
placement in ECMO patients has a unique set of potential pitfalls that
need to be considered and managed to minimise risk. An experienced op-
erator should perform these procedures using ultrasound guidance. The
internal jugular or subclavian vein are the preferred insertion locations as
the femoral vessels are utilised for ECMO cannulation.
Guidewire aspiration into the ECMO circuit is a rare, but catastrophic
complication. The location of the ECMO venous drainage cannula and
negative circuit pressure contribute to this risk. A high drainage cannula
tip and increased ECMO pump speed have been associated with guide-
wire aspiration.1 Techniques to minimise this risk include constant firm
guidewire handling throughout the insertion procedure, minimisation of

Chapter 9 161
guidewire insertion depth and reduction of ECMO pump speed.
Air embolism risk increases during central venous line access due to
the negative pressure generated within the drainage cannula overcoming
the resistance of one-way valves on the central line or sheath allowing
entrainment of ambient air. Reduction of the ECMO pump speed and mi-
nimisation of open atmospheric connections can reduce this risk.

ECMO Circuit Monitoring: Pressures, Delta P, Central Venous

O2 Saturation, Temperature
Pressures/Delta P
Most commercially available V-A ECMO circuits allow for monitoring of
pressures at various points within the circuit. The pressure drop across
the membrane defines the delta pressure and is proportional to resis-
tance within the oxygenator. Increasing delta P may indicate progressive
clot buildup, requiring consideration of a circuit change. These pressures
should be frequently monitored and documented to assess trends. Man-
ufacturers of the oxygenator define the maximum delta P permissible,
and a maximum change per litre/min blood flow (e.g maximum 10 mmHg
Delta P for each litre/min blood flow with an absolute maximum delta P
of 50 mmHg)

Blood Flow and ScvO2

Optimal blood flow on V-A ECMO has yet to be defined. Total circu-
lation is made up of V-A ECMO blood flow plus native cardiac output,
which may be absent. Increasing V-A ECMO blood flow increases aortic
pressure and thus, LV afterload, potentially reducing any native contrac-
tility. Therefore, we recommend support should be adjusted to achieve
adequate organ perfusion and yet reduce aortic pressure to allow for any
potential native LV ejection. Serial measurement of lactate and central ve-
nous oxygen saturation (ScvO2), mixed venous oxygen saturation (SvO2)
or pre-membrane oxygen saturation (SPREO2) may be used to guide ade-
quacy of support. Typical initial blood flows are 2.5-3.5 L/min obtained
with approximately 3000 RPM pump head speeds. See Chapter 7 for V-A
ECMO physiology.
Check circuit blood flow stability. Unstable (falling) circuit blood
flow should prompt a search for a cause of loss of circulating volume or in-

162 ECPR and Resuscitative ECMO

creased resistance in the circuit (see Chapter 8). In the context of ECPR,
intra-abdominal, including retroperitoneal, and thoracic sources of hem-
orrhage are possible. Also, cardiac tamponade (associated with prolonged
chest compression or trauma) or ECMO driven left ventricular distension
with pulmonary congestion may be the cause.

Temperature
The core temperature can be measured from the drainage blood in the
ECMO circuit or by alternative means of core temperature measurements
including pharyngeal or urinary catheter tip temperature probes (see
temperature management below).

Optimize Hemodynamics
Vasopressors and Inotropes
Once V-A ECMO support is established, adequate blood flow to the organs
should be assured, even if there is absent native cardiac output initially.
This can be achieved with an alpha agonist infusion such as norepineph-
rine.
An optimal MAP target following cardiac arrest or ECPR has not been
identified.2,3 We recommend titrating vasopressors (occasionally vasodi-
lators) to a MAP target ≥60 mmHg for organ perfusion pressure and <80
mmHg to minimize risk of LV distension.
Inotropes should be used to achieve some flow through native circu-
lation due to the risk of thrombosis from stasis in the pulmonary circula-
tion and cardiac chambers (e.g. target pulsatility on the invasive arterial
trace of 10-15mmHg). Ensuring LV ejection reduces the risk of compli-
cations such as LV distension, valvular regurgitation, pulmonary edema,
and intracardiac thrombus formation. No inotrope has been proven to
be superior in the post arrest setting. We recommend an inotrope based
on physician familiarity. Inodilators such as milrinone and dobutamine
might be avoided if there is post arrest hypotension and systemic inflam-
matory response syndrome (SIRS).
There may be an exacerbation of LV distension and pulmonary conges-
tion when the LV is less responsive to inotropy (for example after a large
anterior myocardial infarction) due to the asymmetric effect of increasing
RV output without a corresponding increase in LV output by the inotrope.

Chapter 9 163
In such cases, a reduction in inotropy may be required and other tech-
niques may need to be utilized to improve LV ejection.

Blood Flow Targets

Optimal blood flow on V-A ECMO has yet to be defined. Total circulation
consists of V-A ECMO blood flow plus native cardiac output, which may
be absent. Increasing V-A ECMO blood flow increases aortic pressure also
increasing LV afterload, potentially reducing any weak native contractil-
ity. We recommend that support should be adjusted to achieve adequate
organ perfusion whilst manipulating aortic pressure to encourage native
cardiac ejection.
Typical initial ECMO blood flows of 3.0-3.6 L/min are recorded in
the ELSO dataset.37 To achieve adequate organ perfusion, total cardiac
output (ECMO blood flow + native CO) should be equivalent to a cardiac
index of >2.0 L/m2/min, which equates to a blood flow of 3.6 L/min for a
70 kg, 175 cm patient. When cardiac output is absent, the ECMO blood
flow should target the full requirement. As the native circulation recovers,
ECMO blood flow may be weaned.
Central venous oxygen saturation, venous drainage O2 saturation
(SPREO2), pulse pressure (with corresponding end tidal CO2), echocardi-
ography parameters, and end organ perfusion parameters such as lactate
clearance and urine output, may help guide pump flow requirements.

Optimize Gas Exchange

Consider mechanical ventilation with PEEP ≥10 cmH₂O for reducing LV
afterload and preventing or treating pulmonary oedema.
Due to the reduced pulmonary blood flow, reduced minute ventilation
requirements are used to achieve a low normal end tidal CO2 (etCO2).
Normal minute ventilation may result in significant respiratory alkalosis
within the pulmonary circulation. Absent etCO2 during mechanical venti-
lation may indicate lack of native circulation and hence absent pulmonary
blood flow. This should trigger a search for therapies to achieve resto-
ration of some native circulation due to the risks of blood stasis.
Observational data suggest that higher PaO2 may be harmful in hy-
poxic ischaemic encephalopathy patients following cardiac arrest.26–29
Avoidance of hyperoxia can be achieved through careful blending of
ECMO fresh gas flow with an air and oxygen mix, as well as reducing the

164 ECPR and Resuscitative ECMO

FiO2 on the mechanical ventilator. We recommend targeting patient arte-
rial oxygen saturation of 92-97%.
Studies evaluating CO₂ targets after resuscitation from cardiac ar-
rest are conflicting, but a predominance of them demonstrates increas-
ing mortality with initial hypocarbia.30–34. A rapid reduction in PaCO2 is
potentially linked to neurologic complications in V-A ECMO and gradual
normalisation is therefore suggested30; however, there is insufficient data
to give specific PaCO2 targets in ECPR populations. A reasonable target
appears to be 40 mmHg. Changes to sweep gas flow (1-2 l/min changes at
a time with ABG assessment of effect) and changes to mechanical ventila-
tor minute ventilation can be used to control patient PaCO2.

Initial Diagnostics
Echocardiography
Early echocardiography to assess biventricular and valvular function is
helpful in predicting complications of ECMO support. The presence of
greater than mild aortic regurgitation dramatically increases the likeli-
hood of catastrophic LV distension and pulmonary oedema. Detection of
regional wall motion abnormalities increases the likelihood of ongoing
coronary ischaemia. Structural heart disease may also be identified such
as hypertrophic obstructive cardiomyopathy (HOCM) or dilated cardio-
myopathies. Calculation of native stroke volume and cardiac output can
be useful in guiding the degree of ECMO support required.

EKG
A 12-lead EKG should be performed urgently following establishment of
support. This may identify ongoing cardiac ischaemia. Early cardiology
review and transfer to the catheterisation laboratory for urgent percuta-
neous coronary intervention (PCI) may be warranted.

Rhythm Disturbance
Rhythm disturbances can be the primary cause of the cardiac arrest or
may be a complication during the first few hours of mechanical life sup-
port. Restoration of an organised rhythm is important to enhance the na-
tive circulation and provide conditions for left ventricular (LV) ejection.
Observation of the rhythm is advisable in the initial few minutes after
establishing mechanical cardiac support. Restoration of coronary perfu-

Chapter 9 165
 Key Management Priorities
The First Four Hours
ɋ Establish patient and circuit monitoring
ɋ Optimize hemodynamics- right upper extremity arterial line,
ECMO blood flow, vasopressors, blood, fluids
ɋ Optimize gas exchange: titrate sweep to arterial blood gas, ven-
tilation strategies
ɋ Complete initial diagnostics: EKG, echocardiogram, CT, ultra-
sound
ɋ Identify and manage underlying pathology
• Cardiac causes
• Non—cardiac causes
ɋ Exclude CPR complications
ɋ Prevent & recognize early ECMO complications:
• LV distention
• Bleeding & anticoagulation
• Distal limb perfusion
• Differential hypoxemia
ɋ Determine patient disposition
ɋ Recognize the unsupportable patient
ɋ Debrief with the ECMO team

sion pressure with V-A ECMO flow, along with an improving acid base
status, may increase the success of cardioversion attempts. Malignant or
agonal rhythms seen after long CPR times can spontaneously improve af-
ter several minutes of V-A ECMO support.
If ongoing coronary ischemia or other reversible factors are identified,
multiple attempts at cardioversion should be avoided until these factors
can be addressed.

Tachyarrhythmias
Increased coronary perfusion may allow spontaneous conversion to an
organised rhythm. Persistent ventricular tachyarrhythmias warrant an at-
tempt at cardioversion along with pharmacological measures to control
the rate.

166 ECPR and Resuscitative ECMO

Bradyarrhythmias
Slow organised rhythms may be related to ischaemia or occur in relation
to the underlying disease process: drug overdoses (e.g. beta blockers, cal-
cium channel blockers, digoxin), hyperkalemia, or hypothermia. Gener-
ally treating the underlying pathology whilst the patient is supported on
ECMO results in restoration of a native circulation.
In the case of accidental hypothermia, we recommend following the
ILCOR guidelines4 regarding delaying cardioversion until the patient
is warmed. This can be readily achieved with the heat exchanger in the
ECMO oxygenator. During prolonged arrest times, body core tempera-
ture may decline to levels where ventricular fibrillation can be refractory
(<34ºC), requiring rewarming before successful cardioversion is likely.
Bradyarrhythmias may also occur secondary to electrolyte abnor-
malities, particularly hyper- and hypokalemia, or sympathomimetic tox-
idromes.

Chest Radiograph
Check drainage cannula and central line position and assess pulmonary
parenchyma/pleural spaces.

Abdominal Ultrasound
May be used to identify traumatic abdominal pathologies caused by pro-
longed CPR efforts (e.g. FAST protocol).

Computed Tomography
CT imaging can be helpful in ECPR cases. If the cause of the cardiac arrest
is unclear, or if there are signs of significant internal haemorrhage, CT can
take place immediately after the cannulation, otherwise after coronary an-
giography +/- PCI.
If the cause of the arrest is unclear, consider:
ɋ CT brain
ɋ CT pulmonary angiography (timing of contrast administration and
ECMO blood flow may need to be adjusted to improve image quality
in the pulmonary circulation)
ɋ CT abdomen/pelvis

Chapter 9 167
 If the cause for cardiac arrest is identified in the cath lab, consider
imaging:
ɋ CT brain

With falling ECMO blood flows or drainage insufficiency (drainage

line chatter), these additional scans may be useful:
ɋ CT abdomen/pelvis
ɋ CT chest

The rationale for routine CT imaging is to help identify causes of the

cardiac arrest as well as early identification of catastrophic brain injuries
and solid organ bleeding from prolonged mechanical chest compressions.
If the patient has drainage insufficiency and an abdominal ultrasound
scan is suggestive of free fluid, arterial phase CT scan of abdomen and
pelvis should be considered to rule out injury to liver and spleen. This
often occurs early and may present with falling ECMO blood flow (for
a set RPM) or drainage insufficiency. Liver and splenic lacerations are
commonly amenable to embolization with interventional radiology. One
study showed immediate CT scan after ECPR revealed cause of arrest in
17%, pneumo- or hemothorax in 23%, and futile findings in 19% of the
patients.41

Treat Underlying Pathology

Cardiac Causes
Cardiac causes predominate in refractory cardiac arrest. Observational
studies suggest improved outcomes in cardiac arrest patients treated with
PCI.5–7 ECPR series targeting cardiac arrest believed to be due to acute cor-
onary etiologies that utilized protocolised catheterization post cannulation
are associated with increased survival.8,9 Emergent coronary angiography
for ECPR patients without an obvious alternate noncardiac cause should be
considered, independent of age or presenting rhythm.
The presence of significant coronary pathology is high in refractory
cardiac arrests, estimated >75% of cases, with multivessel disease being
common.10 Other pathologies that may be related to cardiac ischaemia:

168 ECPR and Resuscitative ECMO

ɋ Coronary vasospasm/ dissection
ɋ Aortic dissection
ɋ Primary dysrhythmias with or without structural or congenital heart
disease
ɋ Cardiac toxicity - sympathomimetic toxidrome
ɋ Congenital coronary anomalies

Noncardiac Causes
Noncardiac causes are commonly indistinguishable from cardiac causes
during the arrest period and decision making for ECPR. Accidental hy-
pothermia arrests may represent an exception as they are identifiable by
a relevant exposure and an intra-arrest core temperature measurement.
Further reversible causes such as hypovolemia, hypo- or hyperkalemia
may be deduced from intra-arrest investigations.
There may be features in the history suggesting a particular disease
process; however, the history is rarely diagnostic.
Important and noncardiac causes include:
ɋ Pulmonary embolism
ɋ Intracranial haemorrhage (e.g. subarachnoid hemorrhage with dys-
rhythmia)
ɋ Accidental hypothermia
ɋ Abdominal sepsis or haemorrhage (non-traumatic)
ɋ Drowning
ɋ Toxicology

Excluding Traumatic CPR Complications

ECPR is a multifaceted, complex, and time-sensitive intervention that
correspondingly has a high complication rate. Complications may be
divided into those that arise from prolonged CPR and those related to
ECMO cannula insertion. Bleeding is the most common complication en-
countered and ranges from small volume bleeding at the ECMO cannula
site to life threatening abdominal solid organ injury haemorrhage or car-
diac tamponade.
Injuries resulting from prolonged CPR are common and well de-
scribed with up to 11% of patients incurring multiple injuries and 3%

Chapter 9 169
with major injuries.11 The most common traumatic injuries associated
with chest compressions include rib and sternal fractures, pneumothorax
and visceral organ damage.12 Active measures should be taken to exclude
life-threatening complications in patients post ECPR.
Chest wall injuries are the most common CPR-related injuries. Rib
fractures occur in approximately 31% of patients, with flail segments found
in almost 2% of these cases.11 The rate of sternal fractures is also high at
more than 15%.11 When compared to standard manual CPR, a higher in-
cidence of sternal fracture has been reported with both suction-cup and
piston-based active compression decompression (ACD) CPR devices.11
Other chest injuries include mediastinal hematoma (10.2%), pneumo-
thorax (2.5%), hemothorax (2.1%), and pulmonary contusion (41%).11,13
Cardiac and vascular injuries occur less frequently. Pericardial injury
and hemopericardium occur in 8.9% and 7.5% respectively. Cardiac con-
tusion, papillary muscle rupture, prosthetic valve dehiscence, myocardial
rupture, and cardiac tamponade are infrequent.
Abdominal visceral complications have a reported prevalence of up
to 30.8% in postcardiac arrest patients.13 Laceration and rupture of the
spleen has been documented in multiple case reports and liver injuries oc-
curred in 0.6-3% of patients.13 Splenic and hepatic injuries could result in
life-threatening haemorrhage, especially post thrombolysis or antiplatelet
therapy. A high index of suspicion should be maintained in the patient with
unexplained cardiovascular instability post ECPR.
Plain radiographs are likely inadequate to diagnose chest wall injuries;
with rib and sternal fractures being underreported on antero-posterior
chest radiographs. In one study 33% of sternal fractures documented at
autopsy were not diagnosed on chest radiographs.14 Computed tomog-
raphy (CT)probably represents the most proficient imaging modality
to detect both causation for arrest and potential CPR traumatic compli-
cations.15 CT angiography can be particularly helpful in locating arterial
bleeding sites and assessing suitability for angiographic intervention.
Ultrasound in the form of echocardiography, lung ultrasound and
Focused Assessment with Sonography for Trauma (F.A.S.T.) can also be
useful in the identification of postcardiac arrest traumatic injuries. FAST
has been used to identify free abdominal fluid in a patient with abdominal
distension and instability post CPR. The patient then proceeded to CT
abdomen to diagnose liver laceration.16

170 ECPR and Resuscitative ECMO

Preventing and Recognising Early ECMO
Complications
LV Distension and Pulmonary Oedema
ECPR patients are considered to be at higher risk for LV distension,17
particularly those with poor native contractility (narrow or absent pulse
pressure).
When left ventricular inflow exceeds outflow, LV distension occurs.
As distention progresses, LV end diastolic pressure (LVEDP) rises caus-
ing pulmonary oedema and ultimately catastrophic pooling of circulating
volume within the lung compartment. Following ECPR the patient may
have different RV and LV function, particularly in the setting of LV infarc-
tion or significant aortic valve regurgitation with poor LV function. Aortic
valve regurgitation may be pre-existing and unknown or occur in associa-
tion with an acute pathology such as Type-A aortic dissection.
Typical medical management would include high PEEP (often >15
mmHg) to reverse pulmonary oedema, reducing RV outflow, as well as sup-
porting LV outflow by decreasing the LV transmural pressure gradient. Ino-
tropic agents should be used if LV function can be recruited, otherwise they
may prove counterproductive, (e.g, worsen infarct size or arrhythmias).
If medical management fails, LV venting may be required. Techniques
are varied and include direct surgical LV transapical drainage or transaor-
tic catheters (Impella®).18,19 Indirect methods may also be used such as
decompression of the left atrium via an atrial septostomy,20 intra-aortic
balloon pump (IABP) or reducing left sided inflow via transseptal can-
nulae, surgical drainage of the right upper pulmonary vein21 or a drainage
catheter floated (or radiologically guided) into the pulmonary circula-
tion.22,23
There is a broad range of approaches to LV distension and its manage-
ment in V-A ECMO and institutional practice varies. Patient selection bias
impedes the generalisability of any local practice. As such, if a LV venting
capability technique is to be chosen at an institution, it should be after
discussion with relevant stakeholders (cardiothoracic surgeons, interven-
tional radiology, intensivists, ECMO physicians) taking into consideration
technical abilities, cost, availability, and skill maintenance. However, the
early deployment of medical measures may obviate the need for invasive
measures.

Chapter 9 171
Anticoagulation/Bleeding Complications
Starting Anticoagulation
Systemic anticoagulation should be provided for all patients on ECMO5,6
in the absence of bleeding with no anticipated or recent surgery. Antico-
agulation practices are based on patient bleeding risk profiles and local
guidelines. ECPR patients in particular are at high risk of bleeding with
the potential to develop coagulopathies such as disseminated intravas-
cular coagulation (DIC) following prolonged arrest times. They are also
often administered anti-platelet agents in the catheterisation laboratory.

Heparin and Alternative Anticoagulants

Heparin is the most commonly used anticoagulant in ECMO patients. An-
ticoagulation targets are adjusted according to local protocols using ACT,
APTT, or anti-Xa levels. If bleeding is an ongoing concern, a low dose of
500 U/hr without measurable anticoagulant effect or no anticoagulation at
all may be tolerated for several days.24
Systemic anticoagulation may also be maintained with parenteral di-
rect thrombin inhibitors (lepirudin, bivalirudin, and argatroban) depend-
ing on local preference.
Heparin induced thrombocytopenia (HIT) on ECMO is rare. How-
ever, thrombocytopenia with ECMO is common and therefore confound-
ing one of the criteria to assess the probability of HIT. Heparin should
be ceased and platelets should not be transfused if HIT is clinically sus-
pected. Anticoagulant coated circuit components contain heparin and so
it is recommended to switch to a non-heparin bonded circuit (ex: phos-
phorylcholine) if feasible. However, ECMO patients who have confirmed
HIT have been successfully managed with anticoagulant coated circuits.25

Threshold for Blood Transfusion

A specific threshold for RBC transfusion in ECMO has not been identified7
and clinicians may choose to follow local guidelines. Controversy around
guidelines suggesting 70 g/L (7g/dL) in critical care and gastrointestinal
bleed patients versus 100 g/L (10 g/dL) in acute coronary patients ver-
sus traditional practice at the 120 g/L (12 g/dL) have prevented clear rec-
ommendations.42-43 A transfusion trigger between 70-120 g/L (7-12 g/dL)
would be reasonable. Patients should have a current crossmatch.

172 ECPR and Resuscitative ECMO

Platelets
Thrombocytopenia in the setting of ECMO is common yet needs to be
carefully evaluated. Bleeding or intravascular consumption in particu-
lar needs to be excluded and HIT should be considered. A safe platelet
threshold is unknown with institutions utilizing thresholds ranging from
20,000 to 80,000 103/ul depending on the cause and the bleeding risk pro-
file though limited data exists to support this practice.

The Bleeding ECMO Patient

The general approach to the bleeding ECMO patient, whether postpro-
cedural or spontaneous, include replacement of all clotting element defi-
ciencies. This includes:
ɋ Giving cryoprecipitate to target fibrinogen >1.5 g/L (>150 mg/dL)
ɋ Giving platelets to target institutional goals (often at least 50,000
103/ul) or normal MA (maximum amplitude) on TEG
ɋ Giving fresh frozen plasma to target INR <1.3 or normal R-time on TEG

These targets will depend on severity and location of bleeding. Ongo-

ing severe bleeding despite correction of any deficiency should be man-
aged in consultation with the ECMO consultant, haematologist, and the
relevant surgical specialists.
Options to consider include:
ɋ Surgery or Interventional Radiology
ɋ Tranexamic acid
ɋ Factor VIIa
ɋ Prothrombin protein complexes
ɋ Protamine: may be used if unreversed heparin is definitively impli-
cated in bleeding (TCT and reptilase tests may support a clinical
diagnosis). Protamine may promote circuit thrombosis in anticoag-
ulant coated (heparin coated) components and should be used with
utmost caution. Where protamine is to be administered a pre-primed
circuit and capability for emergent circuit change must be arranged.
Unless bleeding is temporally related to heparin administration, pro-
tamine should not be used.
ɋ Palliation

Chapter 9 173
 The usual focus on haemorrhage control should be treated as in any
other massive transfusion scenario. In addition, the heater of the ECMO
circuit may effectively be used to maintain normothermia.
Whilst drainage insufficiency (chatter) may be an early indicator of
bleeding and intravascular volume loss, its absence does not rule out sig-
nificant haemorrhage. The ECMO blood flows may need to be adjusted to
avoid ongoing drainage insufficiency during blood volume resuscitation.

Inadequate Limb Perfusion

Observational series suggest that ultrasound-guided placement of a distal
(antegrade) perfusion cannula, sometimes called distal limb perfusion
cannula, in the ipsilateral superficial femoral artery is associated with re-
duced critical limb ischemia.3 Typically, it is perfused off the side port of
the arterial cannula. Accordingly, we recommend a distal perfusion can-
nula should be sited on the side of arterial cannulation, ideally within 4
hours of cannulation, though this should not delay urgent interventions.
Observations assessing appearance, warmth, capillary refill time and
presence of distal pulses of the limb along with assessment of blood flow
within the distal perfusion cannula should be completed hourly. The
dorsalis pedis and posterior tibial pulses may not be palpable, especially
with a weak or absent native circulation, and therefore doppler pulse as-
sessment may be required. Flow within the distal perfusion cannula can
be demonstrated in two ways. Firstly, by visual observation of turbulent
blood flow at the connection hub using a light for trans-illumination or
using a doppler device to obtain a signal from within the distal perfusion
cannula connection tubing.
Peripheral tissue oxygenation can be monitored with near infrared
spectroscopy (NIRS), however it should not replace regular clinical as-
sessment.
If there is delay in placing the distal perfusion cannula or there is sus-
picion of reduced blood flow through the cannula, additional assessments
should be made to confirm the adequacy of the perfusion to the leg and
exclude ischaemic injury. The lower limbs should be assessed for signs of
compartment syndrome, including compartmental tension, sensorimotor
function, and detection of ischaemic contractures. These are late signs
and efforts should be made to identify compromised perfusion before
this. An arterial vascular ultrasound assessment may help quantitatively

174 ECPR and Resuscitative ECMO

measure flow in the arterial system of the limb. Clinical vascular assess-
ment in these patients is often hampered by shock, high dose vasoactive
medications, and poor cardiac pulsatility.

Development of Differential Oxygenation with Hypoxemia

As native cardiac function returns there will be an increase in blood flow
passing through the pulmonary circulation. If the patient has concurrent
lung pathology such as aspiration pneumonitis or ongoing pulmonary oe-
dema, pulmonary shunt will lead to hypoxia in the arterial circulation.
Optimisation of the patient’s respiratory status may be required—ad-
justment to mechanical ventilator settings, bronchoscopy, inhaled nitric
oxide and proning may be beneficial. Increasing V-A ECMO flow may
capture a greater proportion of blood returning to the right heart and
therefore reduce pulmonary blood flow. Occasionally reconfiguration of
the support to V-AV is required to provide some respiratory support. Ad-
ditional options include reconfiguration to central cannulation or reloca-
tion of the femoral arterial return cannula to the right axillary, subclavian,
or innominate artery.

Caring for the Post ECPR Patient

Temperature Management
Targeted temperature management (TTM) can be precisely achieved
using the heat exchanger on the ECMO oxygenator. Based on protocols
with best outcomes and consensus,35,36 we advise a temperature between
33ºC and 36ºC for 24 hours, then gradually rewarming to 37ºC. The target
temperature is largely defined by local protocols and recommendations.
Severe bleeding is more common in ECMO patients than in patients re-
suscitated from conventional cardiac arrest. If bleeding is difficult to con-
trol, a target temperature of 36ºC may be preferable.
The blood flow in the extracorporeal component of the ECMO circuit
cools the patient whilst a heat exchanger device is commonly used to raise
the ECMO blood and thus patient temperature. Occasionally the patient’s
native heat production (shivering/sepsis/SIRS) may be greater than the
heat dissipation of the extracorporeal tubing. When this occurs, turning
the heater device off and initiating topical active cooling measures may be

Chapter 9 175
needed to achieve the desired target temperature.
For those patients with body core temperature below 33ºC there is
currently no data to suggest a particular rewarming rate. Commonly used
rates are around 3ºC/hour; however, complications (e.g. dysrhythmias or
bleeding) may require faster rewarming.

Post-arrest Systems Review

The care of a post ECPR patient is complex. Often they arrive at an emer-
gency department with little warning or history, undergo ECPR, and are
then moved for imaging and intervention. After the initial emergency
management is complete, it is important to undertake a comprehensive
systems review to ensure that all aspects of their care are being addressed.
The accompanying graphic can be used to facilitate this systems review,
as well as troubleshoot encountered problems and formulate a specific
management plan.

Determining Patient Disposition

Transfer to an ECMO Expertise Centre
The best outcomes for ECMO supported patients occur at centres that
have both expertise in ECMO management and perform >30 ECMO runs
per year.38 As such, centres that perform ECMO infrequently should have
an established referral pathway to an ICU with relevant expertise who
can provide ongoing care. Some patients will achieve a full neurological
recovery but will not achieve sufficient native circulation to separate from
ECMO. In this case their only hope of survival is transition to a durable
implanted ventricular assist device (VAD) or cardiac transplantation.
Early discussion with the referral hospital should be encouraged to
provide guidance and support until retrieval can be undertaken.

The Unsupportable Patient

In some patients, despite appropriate deployment of V-A ECMO, an
organ-perfusing circulation cannot be sustained. This may become appar-
ent within the first 4 hours. Common reasons to cease support include cat-

176 ECPR and Resuscitative ECMO

Chapter 9 177
astrophic LV distension and pulmonary oedema, uncontrollable bleeding
from solid organ injury or major vessel, or identification of an unsurvivable
brain injury. Subsequently, progressive multiorgan failure, severe hypoxic
ischemic encephalopathy, or brain death may become apparent. Our insti-
tution has had survivors without organized heart function with LV venting
for many hours. Some programs have instituted termination guidelines if
native heart function does not return within the first 2-3 hours. ECPR V-A
ECMO support times are recorded in the ELSO Registry are typically 2-3
days. This is similar for both patients who sustain native cardiac recovery
as well as for those in whom a palliative decannulation occurs.37
There may be institutional variation in guidelines regarding termina-
tion of support - these should be predefined as much as possible to aid the
decision making of the clinician.

Debriefing with the ECMO Team

Debriefing the team after a resuscitation may result in several worthy out-
comes. Its purpose is to facilitate discussion around individual and team
thought processes as well as encourage valuable reflection.40 Debriefing
has been demonstrated to improve team effectiveness and in the case of
resuscitation, reduce the time to first compression, improve the rate of
return of spontaneous circulation, and improve neurological outcomes.40

Summary
For a patient in refractory cardiac arrest, timely mechanical circulatory
support is the crucial first step in resuscitation. However, it is only when
this step is paired with a strong focus on ECMO care and treating the un-
derlying pathology in the hours following, that a path may be opened to
survival from refractory cardiac arrest.

References
1. Picard L, Cherait C, Constant O, et al. Central venous catheter
placement during extracorporeal membrane oxygenation therapy.
Anaesth Crit Care Pain Med. 2018;37(3):269-270.

178 ECPR and Resuscitative ECMO

 2. Jakkula Pekka, Ameloot Koen, De Deyne Cathy, et al. Abstract 31:
Low versus High Blood Pressure Targets After Out-Of-Hospital Car-
diac Arrest. Circulation. 2019;140(Suppl_2):A31-A31.
3. Jakkula P, Pettilä V, Skrifvars MB, et al. Targeting low-normal
or high-normal mean arterial pressure after cardiac arrest
and resuscitation: a randomised pilot trial. Intensive Care Med.
2018;44(12):2091-2101.
4. 2019 International Consensus on Cardiopulmonary Resuscitation
and Emergency Cardiovascular Care Science With Treatment Rec-
ommendations - Resuscitation. Accessed June 19, 2020. https://www.
resuscitationjournal.com/article/S0300-9572(19)30665-3/fulltext
5. Dumas F, Bougouin W, Geri G, et al. Emergency Percutaneous
Coronary Intervention in Post-Cardiac Arrest Patients Without
ST-Segment Elevation Pattern: Insights From the PROCAT II Regis-
try. JACC Cardiovasc Interv. 2016;9(10):1011-1018.
6. Reynolds JC, Callaway CW, El Khoudary SR, Moore CG, Alvarez RJ,
Rittenberger JC. Coronary angiography predicts improved outcome
following cardiac arrest: propensity-adjusted analysis. J Intensive
Care Med. 2009;24(3):179-186.
7. Spaulding CM, Joly LM, Rosenberg A, et al. Immediate coronary an-
giography in survivors of out-of-hospital cardiac arrest. N Engl J Med.
1997;336(23):1629-1633.
8. 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.
9. Tonna JE, Selzman CH, Mallin MP, et al. Development and Im-
plementation of a Comprehensive, Multidisciplinary Emergency
Department Extracorporeal Membrane Oxygenation Program. Ann
Emerg Med. 2017;70(1):32-40.
10. Yannopoulos D, Bartos JA, Aufderheide TP, et al. The Evolving Role
of the Cardiac Catheterization Laboratory in the Management of
Patients With Out-of-Hospital Cardiac Arrest: A Scientific Statement
From the American Heart Association. Circulation. 2019;139(12).
11. Miller AC, Rosati SF, Suffredini AF, Schrump DS. A systematic review
and pooled analysis of CPR-associated cardiovascular and thoracic
injuries. Resuscitation. 2014;85(6):724-731.
12. Poole K, Couper K, Smyth MA, Yeung J, Perkins GD. Mechanical
CPR: Who? When? How? Crit Care. 2018;22(1):140.

Chapter 9 179
13. Ram P, Menezes RG, Sirinvaravong N, et al. Breaking your
heart—A review on CPR-related injuries. Am J Emerg Med.
2018;36(5):838-842.
14. Lederer W, Mair D, Rabl W, Baubin M. Frequency of rib and sternum
fractures associated with out-of-hospital cardiopulmonary resusci-
tation is underestimated by conventional chest X-ray. Resuscitation.
2004;60(2):157-162.
15. Dunham GM, Perez-Girbes A, Bolster F, Sheehan K, Linnau KF. Use
of whole body CT to detect patterns of CPR-related injuries after
sudden cardiac arrest. Eur Radiol. 2018;28(10):4122-4127.
16. Nashiki H, Miyate Y, Terui Y, Otani M. Focused assessment with so-
nography for trauma (FAST) identifies liver injury following cardio-
pulmonary resuscitation. BMJ Case Rep. 2017;2017.
17. Truby LK, Takeda K, Mauro C, et al. Incidence and Implications
of Left Ventricular Distention During Venoarterial Extracorporeal
Membrane Oxygenation Support: ASAIO J. 2017;63(3):257-265.
18. Tepper S, Masood MF, Baltazar Garcia M, et al. Left Ventricular
Unloading by Impella Device Versus Surgical Vent During Extracor-
poreal Life Support. Ann Thorac Surg. 2017;104(3):861-867.
19. Schrage B, Burkhoff D, Rübsamen N, et al. Unloading of the
Left Ventricle During Venoarterial Extracorporeal Membrane
Oxygenation Therapy in Cardiogenic Shock. JACC Heart Fail.
2018;6(12):1035-1043.
20. Alhussein M, Osten M, Horlick E, et al. Percutaneous left atrial de-
compression in adults with refractory cardiogenic shock supported
with veno-arterial extracorporeal membrane oxygenation. J Card
Surg. 2017;32(6):396-401.
21. Keenan JE, Schechter MA, Bonadonna DK, et al. Early Experience
with a Novel Cannulation Strategy for Left Ventricular Decompres-
sion during Nonpostcardiotomy Venoarterial ECMO. ASAIO J Am
Soc Artif Intern Organs 1992. 2016;62(3):e30-34.
22. Donker DW, Brodie D, Henriques JPS, Broomé M. Left ventricular
unloading during veno-arterial ECMO: a review of percutaneous and
surgical unloading interventions. Perfusion. 2019;34(2):98-105.
23. Donker DW, Brodie D, Henriques JPS, Broomé M. Left Ventricular
Unloading During Veno-Arterial ECMO: A Simulation Study. Asaio J.
2019;65(1):11-20.

180 ECPR and Resuscitative ECMO

24. Wood KL, Ayers B, Gosev I, et al. Venoarterial-Extracorporeal Mem-
brane Oxygenation Without Routine Systemic Anticoagulation De-
creases Adverse Events. Ann Thorac Surg. 2020;109(5):1458-1466.
25. Natt B, Hypes C, Basken R, Malo J, Kazui T, Mosier J. Suspected
Heparin-Induced Thrombocytopenia in Patients Receiving Ex-
tracorporeal Membrane Oxygenation. J Extra Corpor Technol.
2017;49(1):54-58.
26. Youn CS, Park KN, Kim SH, et al. The Cumulative Partial Pressure
of Arterial Oxygen Is Associated With Neurological Outcomes After
Cardiac Arrest Treated With Targeted Temperature Management.
Crit Care Med. 2018;46(4):e279-e285.
27. Damiani E, Adrario E, Girardis M, et al. Arterial hyperoxia and mor-
tality in critically ill patients: a systematic review and meta-analysis.
Crit Care. 2014;18(6).
28. Munshi L, Kiss A, Cypel M, Keshavjee S, Ferguson ND, Fan E. Oxy-
gen Thresholds and Mortality During Extracorporeal Life Support in
Adult Patients. Crit Care Med. 2017;45(12):1997-2005.
29. Group TI-RI and the A and NZICSCT. Conservative Oxygen Therapy
during Mechanical Ventilation in the ICU. N Engl J Med. Published
online October 14, 2019.
30. Diehl A, Burrell AJC, Udy AA, et al. Association Between Arte-
rial Carbon Dioxide Tension and Clinical Outcomes in Venoar-
terial Extracorporeal Membrane Oxygenation*. Crit Care Med.
2020;48(7):977-984.
31. Lee BK, Jeung KW, Lee HY, et al. Association between mean arterial
blood gas tension and outcome in cardiac arrest patients treated
with therapeutic hypothermia. Am J Emerg Med. 2014;32(1):55-60.
32. Wang C-H, Huang C-H, Chang W-T, et al. Association between early
arterial blood gas tensions and neurological outcome in adult pa-
tients following in-hospital cardiac arrest. Resuscitation. 2015;89:1-7.
33. Tolins ML, Henning DJ, Gaieski DF, Grossestreuer AV, Jaworski A,
Johnson NJ. Initial arterial carbon dioxide tension is associated with
neurological outcome after resuscitation from cardiac arrest. Resusci-
tation. 2017;114:53-58.
34. Schneider AG, Eastwood GM, Bellomo R, et al. Arterial carbon diox-
ide tension and outcome in patients admitted to the intensive care
unit after cardiac arrest. Resuscitation. 2013;84(7):927-934.

Chapter 9 181
35. Nielsen N, Wetterslev J, Cronberg T, et al. Targeted Temperature
Management at 33°C versus 36°C after Cardiac Arrest. N Engl J Med.
2013;369(23):2197-2206.
36. Rolfes C, Muellenbach RM, Lepper PM, et al. Targeted temperature
management in patients undergoing extracorporeal life support after
out-of-hospital cardiac arrest: an EURO-ELSO 2018 annual confer-
ence survey. Perfusion. 2019;34(8):714-716.
37. Richardson A (Sacha) C, Schmidt M, Bailey M, Pellegrino VA, Rycus
PT, Pilcher DV. ECMO Cardio-Pulmonary Resuscitation (ECPR),
trends in survival from an international multicentre cohort study
over 12-years. Resuscitation. 2017;112:34-40.
38. Barbaro RP, Odetola FO, Kidwell KM, et al. Association of
Hospital-Level Volume of Extracorporeal Membrane Oxygenation
Cases and Mortality. Analysis of the Extracorporeal Life Support Or-
ganization Registry. Am J Respir Crit Care Med. 2015;191(8):894-901.
39. Bartos JA, Carlson K, Carlson C, et al. Surviving refractory
out-of-hospital ventricular fibrillation cardiac arrest: Critical care
and extracorporeal membrane oxygenation management. Resuscita-
tion. 2018;132:47-55.
40. Kessler DO, Cheng A, Mullan PC. Debriefing in the emergency
department after clinical events: a practical guide. Ann Emerg Med.
2015;65(6):690-698.
41. Zotzmann V, Rilinger J, Lang CN, et al. Early full-body computed
tomography in patients after extracorporeal cardiopulmonary resus-
citation (eCPR). Resuscitation. 2020;146:149-154.
42. Holst LB, Haase N, Wetterslev J, et al. Lower versus higher he-
moglobin threshold for transfusion in septic shock. N Engl J Med.
2014;371(15):1381-1391.
43. Villanueva C, Colomo A, Bosch A, et al. Transfusion strategies for
acute upper gastrointestinal bleeding [published correction ap-
pears in N Engl J Med. 2013 Jun 13;368(24):2341]. N Engl J Med.
2013;368(1):11-21.

182 ECPR and Resuscitative ECMO

CHAPTER 10
Circuit Catastrophes

Laura Geer • Joseph Tonna

Introduction
The extracorporeal membrane oxygenation (ECMO) specialist needs
to be keenly aware of five main circuit catastrophes that may require im-
mediate action, up to and including separation from ECMO: 1) cannula
dislodgement, 2) air in circuit, 3) catastrophic clot, 4) circuit rupture,
and 5) pump failure. A standardized circuit check is a reliable way to
quickly diagnose these catastrophes.

Foundational Terms
ɋ RPMs – revolutions per minute reflect the pump speed which
subsequently creates negative pressure on the access limb and
positive pressure on the return limb. Excessive pressures can
lead to cavitation, chatter, and/or hemolysis
ɋ Circuit check—standardized physical exam of the ECMO circuit
ɋ Air entrainment—when air gets sucked into the ECMO circuit,
risking catastrophic depriming of the pump head or air embolism
to the patient
ɋ Circuit change—process of coming off ECMO, cutting tubing
and reconnecting to a new primed circuit. Can be used if there is
a mechanical circuit catastrophe that cannot be temporized

 183
 The cursory circuit check involves running the length of the circuit
from the drainage cannula through the pump, oxygenator, tubing, and
back into the return cannula. Assessment for air, clot, cannula dislodge-
ment, pump failure, blood color (bright vs. dark), and circuit discon-
nection should occur within a few seconds through simple observation.
Failure to diagnose a circuit problem can lead to catastrophe including
death. Alternatively, rapid diagnosis can save the patient from complica-
tions. This chapter explains how to complete a physical assessment of an
ECMO circuit, how to respond to the five ECMO circuit catastrophes, and
ways to minimize complications in your program.

The Complete Circuit Check

Prevention of circuit catastrophes is paramount to the success of any
ECMO program. This includes predicting fail points, methodical atten-
tion to detail, and seamless continuity of care between off-going and
on-coming providers. The circuit check should occur at ECMO initiation,
during a circuit emergency, and at the beginning of every shift (Figure 1).
Upon initial entry into a patient’s room, the ECMO specialist observes
for color change between the drainage (dark red) and return (bright red)
cannulas confirming appropriate blood oxygenation. If both cannulas are
dark red, the specialist should immediately inspect the membrane lung,
or oxygenator, and sweep gas tubing connections. Inadvertent sweep gas
shutoff may have occurred. If continuous pressure monitoring equipment
is utilized, an elevated differential pressure, or delta P, can be indicative of
clot formation inside the membrane lung. In addition, acquiring a pre and
post-membrane oxygen blood gas measurement can be helpful in deter-
mining the efficiency of membrane lung gas exchange.
The ECMO specialist should next observe the drainage cannula exit
site, checking for any signs of infection, inspecting the dressing appear-
ance, and assessing suture tissue integrity. We recommend noting depth
of cannula insertion, as cannulas can become malpositioned with patient
movement. Using a flashlight, the ECMO specialist can inspect for white
fibrin strands and darker clots along the drainage cannula which com-
monly develop over time.
Each connection, stopcock, and pigtail should be checked for integrity.
Stopcocks should be hand tightened (never tightened with clamps) and in

184 ECPR and Resuscitative ECMO

 ECMO Circuit Check
ɋ ECMO pump
1. Power/electricity on
2. Revolutions per minute (RPM) – is the pump running and at
the correct speed?
3. Flow – does the pump RPM create an adequate ECMO blood
flow?
4. Pressures – does the pump produce a pressure adequate for
flow and are the pressures (if measured) obtained reasonable?
ɋ Sweep gas
1. Flow – sweep gas flow correctly adjusted?
2. Pressure – is there pressure in the gas line to the oxygenator?
3. Is the tubing plugged into the wall or gas a tank? What is the
amount of gas in the tank?
ɋ Heater on – power/electricity. Important in small patients. The
tubing should be lukewarm.
ɋ Tubing & membrane lung
1. Look – color is an indicator for oxygenation. Bright red blood
post-membrane and dark red blood pre-membrane indicates
oxygenation of the blood. Any clots? Any damage to tubing or
connectors?
2. Feel – tubing should be lukewarm.
3. Cannulation site/s: Bleeding? Integrity of distal perfusion line?

(The table modified after Broman 2019)

Figure 1

the “off” position. If a bridge is present in the circuit, it should be free of

blood and filled with a crystalloid fluid to avoid clotting.
If an inlet pressure monitor located before the motor is in use, the
pressure should be noted. “Chatter” or “chugging” of the venous tubing
indicates a problem with the venous drainage and should be identified.
The motor should be positioned level with the patient and in a location
with good ventilation to prevent overheating. The outflow of the pump-

Chapter 10 185

 Circuit Catastrophes
ɋ Air in the Circuit – clamp tubing, stop pump, push/pull air into
side-port syringe, disconnect tubing, circuit change as necessary
ɋ Accidental Decannulation – clamp circuit, stop pump,
re-cannulate, utilize bridge if available
ɋ Circuit Rupture – clamp circuit, stop pump, stop bleeding
(pressure, clamping), replace damaged area or complete circuit
exchange
ɋ Pump Failure – clamp circuit, keep back-up available, possible
use of hand crank, exchange pump
ɋ Circuit Clot – clamp circuit, stop pump, exchange affected com-
ponents, adequate anticoagulation and flow is key to prevention

head should be positioned upwards to promote movement of any trapped

air out of the pump.
Next, a thorough inspection of the membrane lung should be com-
pleted. The sweep gas tubing should be positioned so that it is easily ob-
served and cannot be stepped on or kinked. A ball-drop test should be
completed to ensure there is gas flow through the gas regulator by briefly
pinching the sweep gas tubing and observing the regulator ball drop. Using
a flashlight, inspect for fibrin or clot formations inside the membrane lung,
which commonly form in the corners and on the inflow side of the mem-
brane. The membrane lung should be positioned level with the patient to
decrease the risk of air entrainment. Condensation will develop inside the
microtubules of the membrane lung due to the sudden cooling of gas after
it is no longer in contact with warm blood, which can lead to impaired gas
exchange. This condensation can be flushed out by increasing the gas flow
rate for a few seconds, which is called “sighing” the membrane lung. A sign
should be placed over the sweep gas wall regulator to prevent accidental
adjustment by ancillary staff. If pressure monitors are in use, the pre, post,
and delta P pressures of the membrane lung should be noted.
After inspection of the membrane lung, the specialist should assess
for correct placement of the flow probe and, if utilized, bubble detector.
The integrity of tie-bands, stopcocks, and pigtails are assessed and any
clot formations along the arterial return tubing are noted. All cannulas

186 ECPR and Resuscitative ECMO

should be secured to the patient and to the bed using anchoring devices
and a specialized ECMO tubing holder. Care should be taken to ensure
cannulas are not within grasp of the patient and are visible during reposi-
tioning maneuvers to prevent accidental dislodgement.
On the ECMO console the alarm settings should be verified and the
operational mode confirmed. The specialist needs to take note of the dis-
played flow and revolutions per minute (RPM). On the ECMO cart, two
tubing clamps must be present along with readily available backup equip-
ment (secondary motor vs. hand crank). The ECMO cart brakes should
be locked with the console screen visible from the doorway and doors
open to ensure alarms can be easily heard. In addition, keeping a backup
pre-primed ECMO circuit for emergent ECMO placement or to use in
the event of a catastrophic circuit failure on an existing patient should be
considered in any ECMO program.

Air Entrainment into the ECMO Circuit

Air can enter the ECMO circuit through excessive negative pressure or
through poorly secured connections (i.e. stopcocks, pigtails) and cannula
dislodgement. Small amounts of circuit air prior to the membrane can be
trapped by an in-circuit bladder, or the membrane lung. If air has entered
the membrane lung, the de-airing port can be utilized to passively vent the
air out. If air is observed in the tubing, a syringe can be used to remove
small amounts of air at leurlock access points along the circuit or on the
membrane lung.
If a large amount of air is seen in the circuit, or even small amounts
within the arterial section, immediate action must occur to prevent air
embolism to the patient. Large amounts of entrained air can also result in
pump head airlock. This is due to de-priming of the pump resulting in ces-
sation of blood flow. The ECMO specialist should follow the Basic Emer-
gency Management Protocol (described below) to cease ECMO flow, which
will also stop the movement of air towards the patient. Lowering the head
of the patient’s bed may prevent air that did reach the patient from en-
tering the cerebral circulation if on V-A ECMO, or on V-V ECMO via a
patent foramen ovale. Once ECMO flow has been stopped, the ECMO
circuit must be de-aired. Many membrane lungs have a connection for an
emergency priming line or stopcock access ports where an ECMO Spe-

Chapter 10 187

 Back-Up Emergency Equipment
Examples:
ɋ Sterile tubing clamps
ɋ Shears
ɋ IV fluid
ɋ Tie-gun and tie-straps
ɋ Appropriately sized connectors and tubing
ɋ Sterile gloves & gowns
ɋ Facemasks/shields, bouffants
ɋ ChloraPrep sticks
ɋ Secondary ECMO console or hand crank
ɋ Secondary motor
ɋ Secondary flow probe

Table 1

cialist can attach a 60 cc syringe, filled with crystalloid, to push air out
of the circuit into another syringe attached to a different access port. If
on V-A ECMO, de-airing could involve disconnecting the drainage tub-
ing to allow native cardiac activity to passively fill the circuit, displacing
the entrained air. If the initial circuit is able to be de-aired, ECMO flow
should be re-established while the ECMO specialist should observe there
is no additional air entering the system. If the ECMO team is unable to
adequately de-air the tubing, or if the pump was not stopped fast enough
resulting in foaming of the blood when it mixes with air (which is difficult
to remove), exchanging to a new primed circuit should be considered. An
example of emergency backup equipment that should be readily available
for a circuit or component exchange is listed in Table 1.

Accidental Decannulation
Accidental decannulation of an ECMO cannula is usually preventable
with thorough circuit checks and frequent tubing assessments. A cannula
can quickly become dislodged after placement but prior to being secured,

188 ECPR and Resuscitative ECMO

during patient transport, or by the patient themselves. When reposition-
ing patients, cannulas should be well visualized and not covered by linens.
Cannula securement is key in preventing decannulation. If a cannula be-
comes dislodged the first action is to immediately clamp the return and
drainage tubing and stop ECMO flow using the Emergent Separation from
ECMO approach described below. Accidental venous drainage decannu-
lation will result in air entrainment into the circuit and accidental arte-
rial return decannulation will result in hemorrhage. After ECMO flow has
been stopped, pressure needs to be applied at the exit site and the antico-
agulant infusion held. If a bridge is present, this can be utilized to main-
tain circuit patency. A new appropriately sized cannula is retrieved and
preparation for replacement of the cannula begins. Cutdown skills may be
beneficial in this scenario (see Chapter 6).

Circuit Rupture
Patient transport, ambulation, or repositioning are high risk events for cir-
cuit damage. If a small component, such as a stopcock becomes damaged,
temporizing measures such as clamping pigtails or covering holes with oc-
clusive tape, bone wax, or even a sterile gloved finger can be attempted to
stop further air entrainment (pre-pump side) or hemorrhage (post-pump
side). If temporizing measures are successful, the broken component can
be changed using one or two clamps on either side of the component. This
Controlled Separation from ECMO process is outlined below. If temporizing
measures are unsuccessful, Emergent Separation from ECMO must be per-
formed to enable emergent replacement of the damaged circuit compo-
nent (or entire circuit).

Pump Failure
Pump failure is a situation that requires immediate action because the
patient is no longer receiving ECMO support. This can occur if the elec-
trical and battery power runs out, or the motor ceases to move blood effi-
ciently due to decoupling of the magnets in the motor and pump head. In
these cases, flow ceases. In a V-A ECMO pump failure, blood flow will re-
verse through the centrifugal pump causing a massive left to right shunt.
The circuit must be clamped using the Emergent Separation from ECMO

Chapter 10 189

method and flow reinitiated using a hand crank or a backup system. In
the case of pump head disengagement, or decoupling, the pump can be-
come noisy with excessive vibrations. If this occurs, ECMO flow should
be stopped using the Emergent Separation from ECMO method, the pump
head reseated, and flow reinitiated.

Catastrophic Circuit Clot

During the life of a circuit, white fibrin or dark clots will commonly form
in areas of stagnant or turbulent blood flow. To limit the development of
these clots, a systemic anticoagulant such as heparin is commonly used
with dynamic therapeutic monitoring using activated partial thrombo-
plastin time, rotational thromboelastography (ROTEM) or thromboelas-
tography (TEG), Anti-Xa, and/or activated clotting time (ACT). Regular
laboratory testing or pressure monitoring is recommended to assess cir-
cuit compromise due to clot formation.
The ultimate decision to exchange a circuit component should be
made when clot or fibrin results in compromised circuit flow. This can
occur, for example, if a clot is ingested into the pump head or when mem-
brane lung thrombosis limits flow and causes inefficient gas exchange.
When a clot is large enough to cause complete or significant flow obstruc-
tion, a large portion or the entire ECMO circuit may need to be replaced.
A catastrophic circuit clot can occur during ECMO initiation if the pa-
tient is not given adequate anticoagulation, or if protamine is given while
using a heparin coated circuit. The ECMO team should utilize the Basic
Emergency Management Protocol to cease blood flow and replace the clot-
ted component.

Basic Emergency Management Protocol:

Emergent vs. Controlled Separation from ECMO
In the event an ECMO specialist realizes that an emergent separation from
ECMO is needed, a CLAMP, STOP, RESCUE approach should be imple-
mented immediately as described below (Figure 2). This approach could
be required with any of the previously described circuit catastrophes when
temporizing measures are not sufficient.

190 ECPR and Resuscitative ECMO

 “CLAMP, STOP, RESCUE” Emergent
Separation from ECMO Checklist.
ɋ CLAMP the arterial and venous cannulas
ɋ STOP the ECMO flow by decreasing RPMs to zero
ɋ RESCUE the patient with increased hemodynamic support and
ventilator settings
ɋ Correct problem (may require circuit change)
ɋ Reinitiate ECMO flow (See Chapter 8 Figure 2 – “B A STAR”)

Figure 2

When a component of the ECMO circuit is beginning to fail, but is

still providing passable support to the patient (i.e. a membrane lung with
a post-membrane PaO2 <100, or a circuit rupture that has been tempo-
rized with bone wax), the ECMO team can implement a controlled sepa-
ration from support as described below.

1. Prepare a sterile field with any components to be exchanged

2. Perform a time-out to clarify provider roles during the procedure
3. Adjust ventilator settings and have vasoactive medications avail-
able to support the patient’s gas exchange and hemodynamics
4. Decrease ECMO flows to approximately 1.5 liters
5. If no bridge present, clamp the patient’s return then drainage cannu-
las. If bridge present, see Table 2 on utilization with centrifugal pump
6. Decrease RPMs to zero
7. Correct existing circuit problem
8. Restart ECMO by increasing RPMs to approximately 1500
9. Unclamp drainage then the return cannulas
10. Return ECMO blood flow to adequate hemodynamic support

Chapter 10 191

 Stopping ECMO flow with Circuit Bridge
ɋ Clamp tubing post oxygenator and shunt.
ɋ Clamp return and drainage cannulas above bridge.
ɋ Open bridge flow and remove post oxygenator clamp.
ɋ Allow blood flow through bridge

Resuming ECMO flow with Circuit Bridge

ɋ Clamp tubing post oxygenator and shunt.
ɋ Close blood flow through the bridge
ɋ Increase RPMs to 1500
ɋ Unclamp drainage and return clamps
ɋ Remove post oxygenator clamp
ɋ Clear bridge

Table 2

References:
1. Brogan TV, Lequier L, Lorusso R, MacLaren G, Peek G, eds. Extra-
corporeal Life Support: the ELSO Red Book. 5th ed. Ann Arbor, MI:
Extracorporeal Life Support Organization; 2017.
2. Short BL, Williams L, eds. ECMO Specialist Training Manual. 3rd ed.
Ann Arbor, MI: Extracorporeal Life Support Organization; 2010.

192 ECPR and Resuscitative ECMO

CHAPTER 11
Transport of ECMO Patients

Björn Frenckner • Lars Mikael Broman

Introduction
The literature on transports of patients on extracorporeal mem-
brane oxygenation (ECMO) is abundant. It almost exclusively focuses on
interhospital transports as opposed to intrahospital transports, i.e. trans-
ferring the patient between different units in the same hospital.1 Only a
few publications have discussed intrahospital transports.2–5
Furthermore, most publications describe primary transports, i.e. the
transport team cannulates the patient at the referring hospital and then
brings the patient to the home hospital’s ECMO unit.
Patients subjected to extracorporeal support for refractory cardiac
arrest (extracorporeal cardiopulmonary resuscitation, ECPR) are rarely
cannulated in the intensive care unit (ICU). Instead this is mostly per-
formed in the emergency department, angiography lab or wherever in
the hospital the cardiac arrest occurred. The patient will therefore have
to be transferred to the ICU, where ECMO treatment will continue. The
distance of the intrahospital transport varies between different hospitals
and the route may include long corridors, elevators, and outdoor spaces
between hospital buildings.
The patient just cannulated for ECMO due to refractory cardiac arrest
is likely totally dependent on the ECMO flow, and the transport must be
organized accordingly. In other words, an accidental arrest of the ECMO
blood flow must be avoided or managed within seconds. Patient safety is
the first priority.

 193
 Transport of ECMO Patients
ɋ Redundancy of equipment – console, oxygenator, tubing clamps,
Fogarty catheters, sterile scissors, clamps, and tubing connectors
ɋ Initiate movement from starting location
• 3 -5 People arranged in formation of figure 1
• Ensure competencies of team sufficient to care for patient if
transport unexpectantly delayed
• ECMO circuit check – See Chapter 10 Figure 1
ɋ En Route
• Watch protrusions from side walls
• Constant slack of tubing – not too much and not too little, i.e.
distance from machine/ancillary equipment to patient
• Watch ECMO flows, watch color of returning blood, BP,
rhythm, and ventilation
ɋ Park patient in new location
• ECMO circuit check – See Chapter 10 Figure 1
• Ensure adequate turnover to new team

Transport Team
The local organization varies significantly between different countries,
different regions, and different hospitals. The composition of the mobile
ECMO team will therefore be different between various programs. Both
the number of staff needed and the composition of different professions
have to be considered.
The number of personnel needed is dictated by the number of items
needing transport. This encompasses the bed along with all the accessory
equipment. At the Karolinska Insitute, five staff members are generally
involved. Two are at the front end of the bed, pulling the bed and pushing
the ventilator on a separate cart ahead of them. Two providers, generally
the ECMO physician specialist, i.e. the team-leader, and the ECMO spe-
cialist will be at the back of the bed with one hand each on the corner of
the bed and one the ECMO cart (Figure 1a and Figure 2a-c). The latter
two providers hold both the bed and the ECMO cart. The distance kept
between the cart and the bed ensures tubing slackness with some flac-

194 ECPR and Resuscitative ECMO

 a.
A b.
B
Trolley w. ECMO Transport Trolley w. ECMO Transport
emergency emergency
Assistant Assistant
equipment equipment

Ventilator Ventilator
ECMO Transport
Assistant ECMO Transport
Assistant
ECMO Specialist ECMO
Nurse/Perfusionist

Bed
Bed

Direction of
movement

ECMO Specialist
ECMO Specialist ECMO
Physician
Nurse/Perfusionist Machine ECMO Specialist Physician

Figure 1. Schematic
Figure 1. Schematic drawing of the ECMO transport. a) In this setting
drawing of the ECMO transport.
a. In this setting the ECMO machine is on a separate cart being behind the patient. It is of utmost importance to keep a fairly constant distance between the bed and the ECMO machine
thewill ECMO machine is on a separate cart being behind the patient. It is of
because of the tubing. The both staff members at the end of the bed are therefore holding one hand each on the bed and the other hand on the ECMO cart. The emergency equipment
always follow the patient during transport.
b. In this setting there is an integrated ECMO machine (Cardiohelp®) in the bed between the legs of the patient. The emergency equipment will always follow the patient during transport.
utmost importance to keep a fairly constant distance between the bed and
the ECMO machine because of the tubing. Both staff members at the end
of the bed are therefore holding one hand each on the bed and the other
hand on the ECMO cart. The emergency equipment will always follow the
patient during transport. b) In this setting there is an integrated ECMO
machine (Cardiohelp®) in the bed between the legs of the patient. The
emergency equipment will always follow the patient during transport.

cidity. This allows for safe movement and rapid adjustment in distance if
needed. Tubing that is too tight risks accidental decannulation. Too much
slack risks intravenous (i.v.) lines becoming pinched or torn off by moving
parts. The provider on the tubing side of the ECMO machine is generally
the ECMO physician specialist. He or she keeps constant control of the
tubing and all additional i.v. lines, etc. Any command from the two staff
pulling the ECMO cart has to be immediately obeyed by the team leader.
The fifth person for the transport carries or pushes rescue and other
equipment, opens doors, pushes elevator buttons etc.
In other settings the ECMO machine can be mounted together with
the bed similar to docking systems used for ventilators, or put in the bed
(Figure 1b and 3a-b). This reduces the risk of separation between the pa-
tient and the ECMO machine during transport and the number of people
required for the transport.

Chapter 11 195

 A C

Figure 2: Intrahospital transport at Karolinska, Sweden. The ventilator

leads followed by the bed, generally with feet first and finally the ECMO
cart. a) Two staff pulling the foot end of the bed and pushing the ventila-
tor ahead of them, b) Two other staff at the rear end of the bed. Each of
them has one hand on the bed and one on the ECMO cart ensuring that
the distance between bed and ECMO cart is not too much. The person to
the right takes extra responsibility for the tubing. c) Patient transport seen
from behind.

Secondly, the competencies needed for the transport should be con-

sidered. The transport team should be able to independently care for the
patient and manage any urgent complication that may occur. Potential
complications of every step along the way must be considered and con-
tingency plans formulated. An elevator may cease to function, a plane may
get delayed, or an ambulance may get a flat tire. These examples of real
situations demonstrate that planning is imperative for patient safety.
At Karolinska the competencies needed for intrahospital transport
are covered by the ECMO specialist (an experienced ECMO ICU nurse)
and by the ECMO physician (experienced ECMO intensivist). The other
three staff members needed for the transport can be other health care

196 ECPR and Resuscitative ECMO

 A

Figure 3: Intrahospital transport at Regensburg, Germany. a) Three staff

moving the patient in hospital corridor, b) The patient has arrived to
the ICU. The ECMO machine is visible in the bed between the legs of the
patient. Pictures generously provided by Professor Thomas Müller and the
rest of the Regensburg team.

Chapter 11 197

Table 1. Redundant equipment for transport
Equipment on transport Comment
Emergency pharmaceutics Incl. needles and syringes for
preparation, etc.
Spare ECMO console and drive unit Keep console on charge between
missions

Spare oxygenator Oxygenator and pump head could be

connected with tubing as one unit, i.e.
Spare pump head Rescue kit, including adaptors for rapid
saline priming
Oxygen bottle 3 L (+)
Saline 2 L, for saline priming
Packed red blood cells
Fresh frozen plasma Thawed
Sterile 3/8-inch Luered-lock connectors S
10 cm 3-way stop-cocks S
50 mL syringes S. Luered-lock
Scissors S
Clamps for ECMO circuit S
Fogarty catheter S. 3 mL syringe and 20 mL saline (S) for
inflation, check for correct length
100 mL syringe S. For aspiration in case of clot
Adapter 3/8-inch to Luered-lock
Spare electrical cable extension Multiple-plug
Spare gas hose extension Oxygen and air
Portable suction device
Tubes for suctioning
Orotracheal tube Proper size(s)
Laryngoscope Blade of proper size, check battery
before
Magill’s forceps
Guide for tube e.g. Bougie
Balloon for ventilation

Abbreviations: S, Sterile

198 ECPR and Resuscitative ECMO

professionals who understand the goals and dangers of an ECMO trans-
port. In other organizations there might be need for perfusionists, respi-
ratory therapists, etc. depending on individual education and experience.

Redundant Equipment for Transport

The equipment that must accompany the patient depends on the antici-
pated length of the transport. Except for very short transports (e.g. within
the ward), it is recommended that rescue equipment for both ECMO and
ventilation circuits be readily available. This includes a spare ECMO con-
sole, a spare oxygenator, tubing clamps, Fogarty catheters (for cannula
thromboses), sterile instruments, sterile spare connectors, blood prod-
ucts, and drugs (Table 1).

Transport after ECPR

In an ECPR situation, transport to the destined ICU should not be ini-
tiated until the patient is stabilized on ECMO. Securing the cannulae,
tubing, endotracheal (ET) tube, and intravenous lines is imperative. The
patient is then moved from the place of resuscitation (emergency room
table, angiography table, etc.) to the ICU bed, in which the patient will be
transported. Infusions are transferred to the transport pumps (if separate
transport pumps are required). Furthermore, the patient is switched over
to a transport ventilator. Pulse oximetry, blood pressure cuffs and rhythm
monitoring are transferred to the portable device (Table 2).
Before disconnecting the electricity and gas from the wall, the team
leader initiates a timeout which includes a short introduction of the pa-
tient and the current situation. Explanation of the purpose of ECMO and
a clear description to all providers of “red flags,” such as a known bleed-
ing site, are key. A Situation-Background-Assessment-Recommendation
(SBAR) format is advisable for such briefings.6,7 A checklist is recom-
mended to ensure that no equipment is left behind (Table. 2). First, after
checking that ECMO machine has a fully charged battery, the wall elec-
tricity is disconnected. The ECMO pump is then powered by the portable
battery. Once the gas has been switched to a portable tank, it is recom-
mended to ensure proper function of the ECMO circuit again.

Chapter 11 199

Table 2. Steps to be carried out before leaving primary site of
cannulation
Steps (no priority order) Comments
Transfer infusion to infusion pump/ Check concentrations and flow rates.
syringe pumps Check all infusion lines that they have
safe route
Check i.v. access for emergency drugs Two lines are preferred
Transfer patient over on transport Check: oxygen and air bottles, batteries.
ventilator Check settings and alarms before
connecting to transport ventilator.
Check inspiratory and expiratory tidal
volumes after transfer to transport
ventilator.
Transfer non-invasive monitoring ECG, pulse oximetry, etc.
Transfer invasive monitoring Blood pressure re-calibrate
Assess blood gas
Check anticoagulation if applicable (e.g. Time to start continuous
ACT) anticoagulation regimen?
Check sedation Agitated patient pulling tubes and
catheters etc. has to be avoided
Check ECMO tubing and i.v. lines
connected to the circuit.
Perform ECMO Circuit check

Abbreviations: ECG, electrocardiogram; i.v., intravenous

In our program, we developed the so called ECMO Circuit

Check checklist for this (see Chapter 10, Figure 1). This specialized
device-patient interface checklist is used every time a situation occurs
that can destabilize the patient. Examples include moving from one bed
to another, moving from ambulance to an aircraft, equipment changes
(emergency or not), when a patient is “un-plugged” (switched to batter-
ies and portable oxygen), at arrival after “plug-in” at new location, and for
rapid assessment and stabilization of any alarm or event out of the ordi-
nary.8 The ECMO Circuit Check is also used as part of the circuit and pa-
tient survey that the ECMO specialist performs at the beginning of every

200 ECPR and Resuscitative ECMO

Table 3. Check list before leaving place of cannulation. The
following steps should be performed before the team starts
moving the patient.
Item Comments
All staff for transport bedside? Everybody knows their individual task
during transport
Time out SBAR – Red flags important. Everybody
listens, everybody has a voice
Disconnect patient from electricity
and gas wall outlets and connect to
transport resources
Perform ECMO Circuit check

shift. It may be specifically adjusted to any mechanical patient support,

e.g. ventilator, intra-aortic balloon pump (IABP) and ECMO.5 When exe-
cuting the ECMO Circuit Check it is important for the whole team to lis-
ten to the staff member performing the check so that everybody gets the
same timely information. This entire procedure only takes a few minutes.
Finally, once everything has been disconnected and checklists com-
pleted (Table 3), the team starts moving the patient. With proper prepa-
ration, the risk for unexpected complications is minimized. In most cases
the route of transport is well known for the transport team. If this is not
the case, the route should be mapped by the team before transporting the
patient. A special situation may occur if the transport also includes small
elevators. In these cases, it is advisable that the team practice this as a
simulation in advance. When reaching the final destination, the reverse
procedure compared to leaving the resuscitation room is performed. The
electricity and gas are again plugged into the wall, monitoring devices are
changed for stationary, transfusions are transferred to other pumps if
needed, and the patient is connected to the ICU ventilator. As a final step,
the ECMO Circuit Check is repeated.

Chapter 11 201

Complications and Risks
Few complications during intrahospital transports have been described
in the literature. Underreporting bias cannot be excluded. Bosch-Alcaraz
et al.4 described 3 complications in 16 transports including clotting of the
distal reperfusion cannula, pump stop caused by malfunction of the bat-
tery, and elevator breakdown without any consequences for the patient.
In another report, no major mechanical complications such as circuit
rupture or cannula dislodgement were noted among 57 pediatric intra-
hospital transports.3 However, 3 patients experienced insufficient venous
return requiring volume boluses. Furthermore, in 103 intrahospital trans-
ports for CT examinations at the Karolinska Institutet, no serious patient
complications or equipment failure occurred.2. Complications during
interhospital transports have been described in more detail. In a study
of over 900 interhospital transports, complications were noted in 28 %.9
They were divided into four different risk categories. Risk category 1 was
defined as complications that must be addressed within seconds and risk
category 2 within minutes. Risk categories 3 and 4 included complications
with no immediate risk for morbidity or mortality. Risk category 1 com-
plications occurred in 2% of the transports. Miscommunication at pump
transfer and pericardial tamponade were the most common category 1
complications, but air in the circuit, battery failure, and clotting of the
circuit or oxygenator also occurred. Risk category 2 complications were
the most common type accounting for 18 % of all transports. Among these
were loss of tidal volume, i.e. an undesired significant decrease of tidal
volumes, hypovolemia, bleeding, and circulatory instability. Most com-
plications occurring during interhospital transports can also occur during
intrahospital transports and the team must be prepared to address these
immediately. Risk factors reported for category 1 and 2 adverse events
during interhospital transports are miscommunication, time on trans-
port, V-A ECMO, and transfer between different transport vehicles.9,10
A well-trained, organized, and experienced team will minimize the
risk for and consequences of complications. Simulation of intrahospital
transports can be of significant value when training the team.

202 ECPR and Resuscitative ECMO

Interhospital Transports
Some health care systems are organized in the “Hub and Spoke” model.
In this model, several smaller centers (spoke) cooperate with a larger
referral ECMO center (hub).11,12 When ECMO treatment is indicated, a
transport team from the hub dispatches to the smaller center, cannulates
the patient and transports the patient on ECMO to the hub center. In
emergency cases, the spoke center can perform the cannulation and start
ECMO treatment. The patient is then retrieved to the hub center by their
transport team. This is called a secondary interhospital ECMO transport
meaning that the transport team is retrieving a patient already on ECMO.1
Interhospital ECMO transports have been performed since the 1970s and
the literature is now abundant.13,14
In secondary interhospital transports, timing is less of a concern. The
patient is already on ECMO and the transport can be accomplished during
daylight hours. Patient safety is the main issue. When the transport team
arrives, they evaluate the patient and receive a complete report from the
transferring providers. In most instances, the ECMO machine can be
changed to the receiving hospitals device.15
The basic principle for exchanging a device is the same regardless.
If the exchange involves the same brand/model device, then flow probes
and pump head are moved over to the transport team’s device directly.
During this procedure, a clamp is used to cease flow within the circuit.
For integrated pump and membrane lung circuits, the clamp is placed on
the return tubing. For nonintegrated circuits, the clamp should be placed
between the pump and membrane lung. As soon as the pump has reached
speed sufficient to prevent backflow, the clamp is released. The ECMO
Circuit Check should be repeated. Cases of neglecting to turn the sweep
gas on or heat exchanger on can have fatal consequences.
In cases where different ECMO machines are used, a pump exchange
is required. Both the drainage line from and the infusion line to the pa-
tient have to be cut and connected to the new ECMO machine. This is
a critical procedure requiring a short pump stop. This halt in perfusion
should not exceed 30–40 seconds. The change is preferably done by two
teams consisting of two people working simultaneously, one team for
each line. Necessary procedural steps include sterilization of the tubing
segments to be cut, preparation of the new tubing with a single connector
at each tube end, a short time out, and a procedure walk-through. It is very

Chapter 11 203

important to check that the drainage will be connected to the drainage
and return to return.

Pump Exchange
Initiating the procedure involves a team leader countdown in order to
maximize timing of the two teams. First, the circuit is clamped near the
pump (see above) by a separate staff member. Two sterile clamps are put
on the tubing segments, the tubing is cut, and patient connected to the
new machine. Air is evacuated by dripping saline over the connector and
tubing while these are connected, the “wet-to-wet connection” (Chapter
8). A less common alternative is the air may be evacuated via a syringe
connected to a Luer-locked three-way stopcock. The air is aspirated af-
ter the tubing clamp on the patient side has been released. This second
method requires one staff at each tubing limb. Communication is essen-
tial. After all air has been evacuated, flow is reestablished by speeding
up the pump followed by release of the clamp. This is executed by the
ECMO specialist at the pump. If the pump change is not done smoothly,
time without perfusion may seriously injure the patient. Changing ECMO
pumps is a procedure that the transport team should practice regularly
and repeatedly in a wet lab simulation. Only confident and skilled provid-
ers should be performing these critical exchanges.
Just as in intrahospital transport, the patient is then moved to a
transport stretcher. Only necessary infusions are kept and transferred to
transport devices. The ventilator is also switched to the transport venti-
lator. Before disconnecting the patent from the stationary electricity and
gas, a timeout is recommended. This is initiated by the transport team
leader and ensures that the transport team is well informed. Check lists
are imperative. After electricity and gas has been transferred to trans-
port resources, the ECMO Circuit Check is executed again. The patient is
moved to the transport vehicle. After loading into the vehicle, it is highly
recommended to use both power and gas supplies from the transport ve-
hicle, if available. Portable resources should be saved pending an emer-
gency or unforeseen delay. After connecting to the vehicle, the ECMO
Circuit Check is repeated. Depending upon distance, weather conditions,
and availability, different transport modes (road ambulance, helicopter,
or fixed wing aircraft) may be used.1 Fixed wing aircraft transport will
involve one extra pair of loading/unloading procedures, as the patient is

204 ECPR and Resuscitative ECMO

transported by road ambulance to the airport. After having reached the
new ICU, the reverse procedures are performed. Critically, use of the
ECMO Circuit Check every time the equipment is disconnected from or
connected to wall electricity and gas, reduces the risk of serious mistakes.

References
1. ELSO. Guidelines for ECMO Transport. https://www.elso.org/
Portals/0/Files/ELSO GUIDELINES FOR ECMO TRANSPORT_
May2015.pdf. Published 2015.
2. Gullberg Lidegran M, Gordon Murkes L, Andersson Lindholm J,
Frenckner B. Optimizing Contrast-Enhanced Thoracoabdominal
CT in Patients During Extracorporeal Membrane Oxygenation. Acad
Radiol. February 2020.
3. Prodhan P, Fiser RT, Cenac S, et al. Intrahospital transport of
children on extracorporeal membrane oxygenation: indications,
process, interventions, and effectiveness. Pediatr Crit Care Med.
2010;11(2):227-233.
4. Bosch-Alcaraz A, Alcolea-Monge S, Dominguez-Delso MC,
Santaolalla-Bertolin M, Segura-Matute S. Complications during
intra-hospital transport of pediatric patient on extracorporeal mem-
brane oxygenation. Med intensiva. 2019;43(8):507-508.
5. Broman M. Complication of ECMO during transport. Patient Safety
Network, WebM&M Cases & Commentaries, Agency for Health
Research and Quality (AHRQ), U.S. Department of Health and
Humans Services, Washington D.C., USA. https://psnet.ahrq.gov/
web-mm/complications-ecmo-during-transport. Published 2020.
6. Randmaa M, Martensson G, Leo Swenne C, Engstrom M. SBAR
improves communication and safety climate and decreases incident
reports due to communication errors in an anaesthetic clinic: a pro-
spective intervention study. BMJ Open. 2014;4(1):e004268.
7. Raiten JM, Lane-Fall M, Gutsche JT, et al. Transition of Care in the
Cardiothoracic Intensive Care Unit: A Review of Handoffs in Periop-
erative Cardiothoracic and Vascular Practice. J Cardiothorac Vasc
Anesth. 2015;29(4):1089-1095.

Chapter 11 205

 8. Broman LM. Interhospital Transport on Extracorporeal Membrane
Oxygenation of Neonates-Perspective for the Future. Front Pediatr.
2019;7:329.
9. Fletcher-Sandersjöö A, Frenckner B, Broman M. A Single-Center
Experience of 900 Interhospital Transports on Extracorporeal Mem-
brane Oxygenation. Ann Thorac Surg. 2019;107(1).
10. Ericsson A, Frenckner B, Broman LM. Adverse Events during
Inter-Hospital Transports on Extracorporeal Membrane Oxygen-
ation. Prehospital Emerg Care. 2017;21(4).
11. Broman LM. Inter-hospital transports on extracorporeal mem-
brane oxygenation in different health-care systems. J Thorac Dis.
2017;9(9):3425-3429.
12. Combes A, Brodie D, Bartlett R, et al. Position paper for the orga-
nization of extracorporeal membrane oxygenation programs for
acute respiratory failure in adult patients. Am J Respir Crit Care Med.
2014;190(5):488-496.
13. Bartlett RH, Gazzaniga AB, Fong SW, Jefferies MR, Roohk H V,
Haiduc N. Extracorporeal membrane oxygenator support for cardio-
pulmonary failure. Experience in 28 cases. J Thorac Cardiovasc Surg.
1977;73(3):375-386.
14. Frenckner B, McKamie W, Fiser RT. Transport of the Patient Sup-
ported with ECMO. In: Brogan T V, Lequier L, MacLaren G, Peek
G, eds. Extracorporeal Life Support: The ELSO Red Book. 5th ed.
Ann Arbor, Mi, USA: Extracorporeal Life Support Organization;
2017:599-612.
15. Broman LM, Dirnberger DR, Malfertheiner M V, et al. International
Survey on Extracorporeal Membrane Oxygenation Transport. ASAIO
J. 2020;66(2):214-225.

206 ECPR and Resuscitative ECMO

CHAPTER 12
Prehospital ECPR for Out of
Hospital Cardiac Arrest

Lionel Lamhaut, Akshay Mungur, Alice Hutin

Rationale Behind Prehospital ECPR

In the Paris region, the average out of hospital cardiac arrest (OHCA)
survival rate is 4.5% with about 40,000 sudden deaths per year.1 Sur-
vival rates remain low despite improvements of basic life support both
in-hospital and pre-hospital. The objective of cardiopulmonary resuscita-
tion is not only to obtain return of spontaneous circulation (ROSC) but
also to supply adequate organ perfusion, especially to the brain. The sur-
vival rate is closely correlated with the preservation of cerebral circulation
via cardiac massage. Even high quality cardiac massage results in a low
flow (LF) state that negatively affects survival rate as duration increases.2
Extracorporeal cardiopulmonary resuscitation (ECPR) provides ad-
equate organ perfusion while allowing etiological treatment. This is con-
firmed by various studies in which neurological prognosis of patients is

Foundational Terms
ɋ Mobile ICU Team –ambulance service of Paris that includes an
emergency physician and medic. They are the initial responders
to out of hospital cardiac arrests

 207
 Paris ECPR Work Flow
ɋ Cardiac Arrest call to Paris central
ɋ Dispatch of non-ECMO ambulance to scene
ɋ Dispatch of ECMO ambulance to scene if criteria met
ɋ Non-ECMO ambulance assesses patient and affirms or rejects
ECMO inclusion
ɋ ECMO ambulance arrives
ɋ Police/Firefighters secure area and set up visual seclusion from
bypassers
ɋ ECMO team prepares patient in sterile fashion
ɋ Cannulation/Initiation on scene
ɋ Patient transported with ECMO team in non-​ECMO ambulance
to hospital with police escort. Cardiac catheterization may occur
at different hospital en route to accepting hospital

better if they have had ECPR vs. conventional CPR.3 The earlier extracor-
poreal perfusion with ECMO can be initiated, the better the prognosis.4
Studies show that implementation of ECPR should be done within
60 minutes of collapse. One solution is a “load and go” strategy where
patients get transported to the closest ECPR center. However, outside the
hospital, there are horizontal and vertical constraints, i.e., traffic, high
rise buildings, difficult patient extraction, etc. Despite the establishment
of “load and go” strategies, teams in Vienna5 and Nancy6 have shown that
it remains difficult to even reach the hospital within 60 minutes. An al-
ternative approach has been established with success in Paris where the
ECPR team is transported to the patient. If a pre-hospital ECPR system
is well run, the patient’s time in a LF state decreases and survival can
potentially increase. Prehospital ECPR can circumvent patient extraction
difficulties and transportation delays common in many cities worldwide.7
At the SAMU de Paris, our data suggests that adopting an aggressive
strategy of dispatching the ECPR team outside the hospital as soon as
the cardiac arrest is recognized significantly increases survival. This is
dependent on the presence of a specialized team capable of judging the
usefulness of the technique at the scene of the cardiac arrest.8 Our pro-

208 ECPR and Resuscitative ECMO

gram reduces the LF time, respecting the “Golden hour” of ECPR, which
contributes to improved survival.

Capabilities
A highly trained team is critical to the success of any pre-hospital ECPR
team. Our data show that concerns about non-surgeons performing can-
nulation and initiation can be alleviated with quality training. Indeed, the
failure rate of ECPR implementation in our program is the same (7.2%)
prehospital as it is in the hospital.9 The crucial point is to have access to an
efficient emergency medical system with an immediately available ECPR
team.
For all these reasons, we believe prehospital ECPR to be the strategy
of choice in refractory OHCA.

Prehospital ECPR for Equality of Care

Inhospital ECPR for OHCA remains a treatment available only in certain
hospitals. These “ECPR centers” are mostly in urban areas. Despite a “load
and go” strategy with all the limits mentioned previously, distance in itself
can be an obstacle, especially in rural areas. Some patients cannot benefit
from this technique because of their remote locations. The objective of
prehospital ECPR is also to give access to patients with refractory OHCA
in rural areas far from ECPR centers. We have recently demonstrated that
our ECPR team can be transported to remote regions of the Ile de France
with an encouraging survival rate, although the LF is not as optimal as the
ideal 60 minute cutoff (reference in press).

Coordination with Prehospital Emergency

Medical Services
The prehospital ECPR team is made of only 3 people with very specific
roles. In Paris, the first medical team to arrive is the mobile ICU “resusci-
tation team.” This involves a paramedic, a nurse, and an emergency phy-
sician. As the mobile ICU team is activated recruitment of the ECPR team

Chapter 12 209

is made if certain inclusion criteria are met (Chapter 3 Figure 5). These
two teams coordinate efforts until resuscitation at the scene has finished
and the patient is transferred to the intensive care unit.

Prehospital ECPR Team

The ECPR team consists of:
ɋ An intensivist or emergency physician who has had specific training
for ECPR implementation.
ɋ A certified anesthetic nurse specially trained in the priming of the
circuit and in the management of patients requiring intensive care.
ɋ A specially trained paramedic capable of rapidly navigating big city
traffic and assisting in on scene ECMO resuscitations.

When the ECPR team is dispatched, the team leaves the base in a ded-
icated ECPR vehicle in which all the necessary equipment is loaded. Long
distance dispatches involve transport via a helicopter.

Initiation
The ECMO Team is triggered as soon as a compatible cardiac arrest is sus-
pected. Once on the premises, the initial EMS team (mobile ICU) relays
information to the ECPR physician. The ECPR physician then judges the
appropriateness of ECPR implementation by applying the strict indica-
tions. Ensuring that proper traditional resuscitation continues during and
after implementation of ECPR is critical. Both teams quality check this
process to ensure high quality CPR continues.
If inclusion criteria are met, the ECPR physician commences cannula-
tion. As described in Chapter 6, the ECPR physician uses a hybrid cutdown
technique. The mobile ICU physician acts as an assistant in the process.
Once ECPR has been started the ECPR physician continues the intensive
care of the patient until arrival at the hospital. This includes initiation of
vasopressors and systematic sedation, temperature control, gas blending,
and circuit troubleshooting. In case of major bleeding, blood transfusion
can also be given. All of these capabilities are organized within our ECPR
ambulance in a portable format. The capabilities are transferred to the mo-
bile ICU where the patient will be transported to the accepting hospital.

210 ECPR and Resuscitative ECMO

 The ECPR nurse is responsible for priming the circuit. With the help
of the onsite team, the nurse anticipates drug administration and prepares
vasoactive and sedative medications necessary for immediate post ECPR
care. At this moment, point of care blood analyses are also obtained: an
arterial blood gas is obtained with blood from the arterial cannula inser-
tion. These data assist the mobile ICU team with monitoring of the pa-
tient and transfer to the hospital.
The paramedic of the team is not only the driver of the ECPR team
but also assists the “cannulators” during ECPR. As both physicians are
dressed in sterile gowns, the paramedic is responsible for the supply of all
the necessary equipment. Beyond implementation, he is also responsible
for all the logistics and helps anticipate patient extraction, with the help
of the mobile ICU ambulance driver and BLS teams on site.

Logistics of Program Setup

Multidisciplinary Program
Prehospital ECPR requires several agencies with perfect knowledge of the
role played by all participants. The dispatch doctor and assistant are re-
sponsible for sending the BLS and ALS teams as well as the early dispatch
of the ECPR team. In the situation of a call for a witnessed cardiac arrest,
all teams are sent simultaneously. Dispatcher assisted CPR is additionally
directed by these personnel. Indeed, ECPR is indicated only if CPR is im-
mediately delivered to limit the no flow as much as possible.
On the scene, while the BLS and ALS teams deliver high quality CPR,
the help of the police can also be sought in order to secure the scene and
facilitate patient extraction, especially in public areas.
During the implementation procedure itself, the cardiologist in the
nearest ECPR center is alerted by the physician in the dispatch center in
order to anticipate and organize diagnostic exams such as coronary an-
giography. The intensive care unit is also alerted of an incoming ECPR
patient.
In order to complete the process within the “Golden Hour” of resus-
citation, a coordinated effort involving many knowledgeable providers is
necessary. Anticipating complications and roadblocks to any given patient
scenario is necessary.

Chapter 12 211

Protocol Based Program
A strict and universally accepted protocol must be used by all participants
to ensure the best synergy. The dispatch assistant must quickly diagnose
a cardiac arrest call in order to initiate bystander CPR and trigger rescue
teams. The ECPR team is systematically dispatched at the same time as
rescue teams for all witnessed cardiac arrests that meet inclusion crite-
ria. In the event of a cardiac arrest occurring after the mobile ICU team
arrives (i.e., initial dispatch for chest pain), the mobile ICU teams know
to request the ECPR as early as possible. Mobile ICU teams also know to
cancel the ECPR team if exclusions exist.
Beyond the dispatching protocols, cardiac arrest therapeutics and
timing of ECPR initiation are also controlled. Epinephrine doses are lim-
ited to 5 mg. ECPR initiation is withheld until twenty minutes of tradi-
tional CPR has failed.
As mentioned previously, post ECPR care is also protocolized. Pa-
tients are empirically treated with vasoactive drugs (dobutamine and nor-
epinephrine) once the ECPR pump is activated. Sedation is immediately
started when ECPR implementation is decided. This written protocol is
available for mobile ICU teams and treatment preparation can be antici-
pated while waiting for the ECPR team.
After implementation, a radial arterial catheter is inserted in order
to measure invasive blood pressure. In Paris we have decided to use the
left radial artery in order to leave the right radial artery for coronary an-
giography. Other centers have used the right radial artery as its takeoff
from the aorta is closer to coronary vessels and thus easier to monitor
differential oxygenation. Vasoactive drug dosage is adapted according to
mean arterial pressure, as is ECPR blood flow. Intravenous fluid loading is
also commonly required especially in case of low venous pressures. Blood
transfusion is not systematic in the field but can be available in case of
massive bleeding during implementation.

Barriers to Starting a Prehospital Program

One of the main obstacles to the implementation of this technique is the
prior existence of a regulation call center (whether it be medical or not).
France has a medical dispatch center responding to the emergency medi-
cal phone number 15. This call center is managed by experienced doctors

212 ECPR and Resuscitative ECMO

(emergency physicians or intensivists). They can quickly judge whether
ECPR is indicated. In addition, France has a mobile medical intensive care
unit that provides onsite high-quality CPR. This provides the foundational
framework necessary if indications for ECPR are met. This whole chain of
survival is therefore essential but requires medical workers at early stages.
Once the regulation and dispatch aspects are dealt with, another ob-
stacle to the implementation of this technique is the availability of doctors
capable of performing this technique. These doctors are assigned specif-
ically to this position. This results in large human resource constraints.
The immediate availability of emergency physicians (as opposed to sur-
geons for example) can, therefore, increase access to ECPR. In Paris, one
of the 5 ECPR physicians is specifically assigned to the EPCR team every
day. In the absence of ECPR calls, this physician can, if necessary, help in
the dispatch center, work in the ICU, and be sent to a multiple casualty
event if necessary. Beyond ECPR implementation, we have incidentally
noticed that the presence of the ECPR team on scene of a cardiac arrest
can also be of use in the event that the patient gains ROSC without ECMO.
Finally, the last obstacle is the economics. ECPR is an expensive tech-
nique that induces long and expensive ICU care. However, the progress
made in patient selection limits the implementation of this technique in
patients with no hope for intact neurologic survival. Indeed, we must keep
in mind that the objective of ECPR is to treat patients who have an optimal
chain of survival (limiting no flow and low flow) and for whom a revers-
ible cause is suspected. Despite this aim, some patients develop brain death
and can become organ donors. Although this is a very delicate matter, it is
important to keep in mind that ECPR can also benefit other patients. This
offsets some of the negative financial aspects of ECPR.

Patient Selection
In order to ensure early arrival on the scene, the ECPR team is dispatched
at the same time as the emergency medical services. This strategy leads
to a high number of cancellations of the ECPR team (for reasons such as
ROSC, clear absence of indication for ECPR found by mobile ICU team,
or even absence of cardiac arrest altogether).
Inclusion criteria:

Chapter 12 213

ɋ <70 years
ɋ No flow <5min
ɋ Refractory VF
ɋ Signs of life per CPR (irrespective of the cardiac rhythm)
ɋ EtCO2 >10 mmHg
ɋ Non-traumatic cardiac arrest
ɋ Hypothermia/Drug intoxications

Beyond these written criteria, it is important to verify that the pa-

tient was treated with an optimal chain of survival: immediate bystander
CPR, followed by basic and advanced life support. It may be difficult to
ascertain precise timing (especially for no flow) in the prehospital setting
which is why signs of life are a promising criterion as they reflect the prob-
ability of high-quality CPR.

Transportation Considerations
The technique of ECPR in the prehospital setting is a peculiar and im-
pressive technique, especially in the eyes of the general public. Securing
the ground is therefore essential to protect the patient and all providers.
Thus, law enforcement must block all potential access and limit the num-
ber of unnecessary bystanders.
The patient must be evacuated quickly. The vertical and horizontal
limitations of large cities can make patient extraction difficult. However,
the implementation of ECPR, which restores adequate blood flow, makes
it possible to carry out the evacuation more serenely. Anticipation of the
evacuation, whether by fire ladder, elevator, ambulance, helicopter, etc.,
is essential to arrive at the hospital as early as possible to treat the under-
lying cause of cardiac arrest.
Femoral venoarterial (V-A) access is the best choice for pre-hospital
ECPR cannulation, allowing for the continuation of CPR. Cannula inser-
tion can be performed percutaneously or with cutdown to visualize the
femoral vessels. In the prehospital setting, cannulators adapt themselves
to the scene which can be more or less friendly. Limited space can restrict
the number of people who have close access to the patient.
Our ECPR team employs only the femoral cutdown technique. We
teach it to future members of our team for ECPR cannulation. One under-

214 ECPR and Resuscitative ECMO

appreciated key aspect of a successful cutdown cannulation is adequate
lighting. Hence, a headlamp for the cannulators is crucial.
We initially used the percutaneous approach for prehospital cannula-
tion in our first prehospital ECPR cases. However, several failed attempts
lead to the use of the hybrid technique. The prehospital scene is not al-
ways the best setting for the use of ultrasound; however, this technique
can be used by trained operators if available.

Worldwide Prehospital ECPR

Prehospital ECPR has been available in the city of Paris since 2011 and
is now routine care. Since then, several cities in France have started pre-
hospital ECPR using the same type of program (Lille, Lyon, Perpignan).
Many physicians were trained over a 2-day course set up by the SAMU
de Paris ECPR team. Other cities around the world such as Albuquerque,
Minneapolis, and London, have also initiated prehospital ECPR programs.
In order to be efficient, it is important for each program to organize pre-
hospital ECPR according to preexisting local emergency care networks.

Albuquerque
In 2019, the emergency physicians from the University of New Mexico
Hospital set up a prehospital ECPR program with the help of Albuquerque
Fire Rescue. The program relies on early recognition of witnessed cardiac
arrest followed by rapid dispatch of all participants. ECPR implementa-
tion takes place in a dedicated vehicle converted into a prehospital inten-
sive care unit. Cannulation is done using point-of-care ultrasound. Once
the pump is activated the patient is stabilized and taken to the hospital
for further care.

Minneapolis
In the same manner, Dr. Yannopoulos at the University of Minnesota has
been working on the development of a mobile ECMO unit for the deploy-
ment of ECPR in the prehospital setting. ECPR cannulation takes place
in the closest Emergency Department and soon in a dedicated ambulance
using ultrasound or fluoroscopy.

Chapter 12 215

London
The city of London has also developed a protocol (SUB30, NCT03700125)
to evaluate the possibility of establishing ECPR flow within 30 minutes of
collapse. Such a protocol relies on a close collaboration between dispatch
and response services (London Ambulance Service, London Air Ambu-
lance), prehospital physicians from the London Air Ambulance, and the
hospital facility (Barts Health).

Conclusion
The possibility of prehospital ECPR implementation is growing around
the world. This might not be the best solution for all cities or patients,
but in situations where the patient cannot be taken to the hospital for
in-hospital ECPR implementation within a reasonable amount of time
(maximum 60 min lowflow), ECPR should be taken to the patient. This
type of program must definitely be planned, protocolized and adapted ac-
cording to local resources.

References
1. Bougouin W, Lamhaut L, Marijon E, Jost D, Dumas F, Deye N,
et al. Characteristics and prognosis of sudden cardiac death in
Greater Paris: population-based approach from the Paris Sudden
Death Expertise Center (Paris-SDEC). Intensive Care Med. juin
2014;40(6):846‑54.
2. Reynolds JC, Frisch A, Rittenberger JC, Callaway CW. Duration of
resuscitation efforts and functional outcome after out-of-hospital
cardiac arrest: when should we change to novel therapies? Circula-
tion. 3 déc 2013;128(23):2488‑94.
3. Chen Y-S, Lin J-W, Yu H-Y, Ko W-J, Jerng J-S, Chang W-T, et al. Car-
diopulmonary resuscitation with assisted extracorporeal life-support
versus conventional cardiopulmonary resuscitation in adults with
in-hospital cardiac arrest: an observational study and propensity
analysis. Lancet. 16 août 2008;372(9638):554‑61.
4. Wengenmayer T, Rombach S, Ramshorn F, Biever P, Bode C, Duer-
schmied D, et al. Influence of low-flow time on survival after extra-

216 ECPR and Resuscitative ECMO

 corporeal cardiopulmonary resuscitation (eCPR). Crit Care Lond
Engl. 22 juin 2017;21(1):157.
5. Poppe M, Weiser C, Holzer M, Sulzgruber P, Datler P, Keferböck M,
et al. The incidence of « load&go » out-of-hospital cardiac arrest
candidates for emergency department utilization of emergency
extracorporeal life support: A one-year review. Resuscitation. juin
2015;91:131‑6.
6. Chouihed T, Kimmoun A, Lauvray A, Laithier F-X, Jaeger D, Le-
moine S, et al. Improving Patient Selection for Refractory Out of
Hospital Cardiac Arrest Treated with Extracorporeal Life Support.
Shock Augusta Ga. janv 2018;49(1):24‑8.
7. Lamhaut L, Jouffroy R, Soldan M, Phillipe P, Deluze T, Jaffry M, et al.
Safety and feasibility of prehospital extra corporeal life support im-
plementation by non-surgeons for out-of-hospital refractory cardiac
arrest. Resuscitation. nov 2013;84(11):1525‑9.
8. Lamhaut L, Hutin A, Puymirat E, Jouan J, Raphalen J-H, Jouffroy R,
et al. A Pre-Hospital Extracorporeal Cardio Pulmonary Resuscitation
(ECPR) strategy for treatment of refractory out hospital cardiac ar-
rest: An observational study and propensity analysis. Resuscitation.
août 2017;117:109‑17.
9. Lamhaut L, Hutin A, Dagron C, Baud F, An K, Carli P. A new hybrid
technique for ECPR-implementation by non-surgeons. Emergencias.
2021 Abr;33(2):156-157.

Chapter 12 217

CHAPTER 13
ECMO and Shock States

Amy E. Hackmann

Introduction
Given the high mortality associated with cardiac arrest, the survival
benefit of extracorporeal cardiopulmonary resuscitation (ECPR) heavily
outweighs the risk. Thus, the decision to perform ECPR in the right can-
didate is relatively easy compared to the non-arresting patient. Complica-
tions are high in the cardiac arrest setting due to the difficulty of placing
cannulas during active chest compressions combined with the inherent
trauma involved with cardiopulmonary resuscitation (CPR). As total ar-
rest time increases, the chance of patient salvage becomes exceedingly
low and the risk/benefit analysis increasingly favors extracorporeal mem-
brane oxygenation (ECMO). In cardiogenic shock, the decision to initi-
ate ECMO is more challenging. The complications of cannula insertion
decrease owing to the more controlled environment, but the potential for
harm increases with any invasive procedure. Thus, the overall benefit may

Foundational Term
ɋ Percutaneous Microaxial Ventricular Assist Device – Impella
–provides augmentation of left ventricular output without oxy-
genation by pulling blood from the left ventricle into the ascend-
ing aorta. A right ventricle support device exists, as well.

 219
 Stage A ■ No signs of symptoms of cardiogenic shock
– “At Risk” ■ Normal Laboratory Values
■ Normal blood pressure and cardiac output
■ 3% in-hospital mortality
Stage B ■ Hypotension (SBP <90mmHg) or tachycardia (HR >100 bpm)
– “Beginning” ■ No evidence of hypoperfusion
■ No to minimal renal impairment
■ Normal Lactate
■ Normal cardiac output
■ 7.1% in-hospital mortality
Stage C ■ Hypoperfusion requiring interventions such as inotropes
– “Classic” or vasopressors
■ Cool, dusky or mottled extremities
■ Elevated lactate, elevated transaminases, or creatinine
twice baseline
■ Hypotension, tachycardia, or cardiac index below 2.2
despite pharmacologic therapies
■ 12.4% in-hospital mortality
Stage D ■ Findings of Stage C for more than 30 minutes, however,
– “Deteriorating” requiring the addition of multiple inotropes/vasopressors
or mechanical support devices to maintain perfusion
■ 40.4% in-hospital mortality
Stage E ■ Cardiac Arrest
– “Extremis” ■ Refractory ventricular arrhythmias
■ Hypotension despite maximal support
■ On VA ECMO support
■ 67% in-hospital mortality

Table 1: Society for Cardiovascular Angiography & Interventions (SCAI)

stages of cardiogenic shock.5,12

decrease. An informed risk/benefit assessment of ECMO, other mechani-

cal support devices, and traditional management becomes key.
The first objective of the practitioner is verifying the etiology of the
shock. Patients with distributive shock from sepsis, for example, have far
less to gain and possibly experience significant harm from ECMO initi-
ation compared to those patients with ECMO responsive shock states.
Specifically, the practitioner should be looking for three main ECMO re-
sponsive etiologies: cardiogenic shock, rhythm instability, and massive
pulmonary embolism. Cardiogenic shock may include acute coronary

220 ECPR and Resuscitative ECMO

syndrome, myocarditis, cardiomyopathies, or sepsis with myocardial dys-
function. All of these types of shock lead to a low cardiac output state
that can be rescued with venoarterial (V-A) ECMO. This is in contrast to
vasodilatory shock where resuscitative V-A ECMO is not beneficial.
Diagnosis of specific shock states is critical to determining the po-
tential benefit of ECMO initiation. Among the many tools to rapidly aid
in diagnosis, the echocardiogram is most crucial. Shock states can mimic
each other, and inaccurate determination can be to the detriment of the
patient. Patient status may not afford the luxury of a formal echo prior to
decision making. In such cases, a bedside ultrasound by the right practi-
tioner may suffice. Studies have shown that bedside echocardiogram can
yield accurate assessments of low ejection fractions in non-cardiology
practitioners.1 To solidify the diagnosis, a measure of cardiac output is
extremely helpful. The most easily accessible measurement is the cen-
tral venous oxygen saturation (ScvO2) from an upper body central line.
Once the diagnosis is made, a risk/benefit assessment for each of the three
ECMO responsive shock states can be made more accurately.

Cardiogenic Shock
Cardiogenic shock represents the most common cause of ECMO respon-
sive shock states. The outcomes of extracorporeal life support for cardio-
genic shock differ significantly, both in complication and survival rates
compared to ECMO initiated during cardiac arrest.2 Therefore, the timing
of support in the hemodynamically decompensating patient becomes crit-
ical.3 This decision may be influenced by a number of factors including the
underlying cause of heart failure, degree of end-organ dysfunction, avail-
ability of advanced therapies, program size, and operator experience.4

Stages of Cardiogenic Shock

The Society for Cardiovascuar Angiograpy and Interventions (SCAI)
multidisciplinary consensus document stratifies cardiogenic shock by
physical exam, laboratory values, and hemodynamic measures into stages
A through E (Table 1).5 Validation studies demonstrate worsening out-
comes in patients with more profound shock, thus providing objective
indications to escalate to mechanical circulatory support when other

Chapter 13 221

therapies are not achieving clinical improvement. Stage A represents pa-
tients at risk for cardiogenic shock, most commonly after an acute myo-
cardial infarction. Stage B includes early shock, with mild hypotension or
tachycardia and normal laboratory studies. Mechanical support is most
commonly indicated in stages C, D, and E. Stage C represents classic car-
diogenic shock and may have physical exam signs such as cold, clammy
extremities, altered mental status, or oliguria. Other findings may be el-
evated lactate, elevated creatinine, or elevated transaminases. Hemody-
namics criteria include cardiac index below 2.2 L/min/m2 with elevated
filling pressures, hypotension, and tachycardia. If inotropes and vasopres-
sors fail to result in improvement within thirty minutes in the setting of
stage C shock, the patient can then be classified as stage D. Patients who
continue to deteriorate in stage D may present in extremis, stage E, with
near arrest or cardiac arrest states. Stage A and B cardiogenic shock are as-
sociated with low hospital mortality, less than 10%, although it is import-
ant to recognize that it is not without some risk of patient death.6 Stage C
carries 12% risk of in-hospital mortality, while stage D portends 40% risk
of patient death and stage E, 67%. Earlier recognition and reversal of the
shock state is critical to improving survival.
Patients admitted to the emergency department in cardiogenic shock
commonly present with acute myocardial infarction (MI) or decompen-
sated chronic heart failure; however, more unusual presentations require
a high degree of suspicion from the treating physician. A flu-like illness
may indicate viral myocarditis, especially in a young, otherwise healthy
individual. Overdoses may include cardiac suppressant medications, such
as beta blockers or calcium channel blockers. Peripartum cardiomyopathy
may present weeks after delivery with cardiogenic shock. Mechanical com-
plications of MI may occur in patients presenting subacutely, including
papillary muscle rupture leading to acute, severe mitral regurgitation, ven-
tricular septal defect, or ventricular free wall rupture. Assessment may be
more challenging in patients with acute decompensation of chronic heart
failure especially if their baseline heart function is unknown. The decision
to urgently place a patient on ECMO for cardiogenic shock may also de-
pend on the likelihood of recovery from the condition, or the ability of an
intervention to rapidly change the status of the patient.7

222 ECPR and Resuscitative ECMO

Rhythm Instability
Patients who have frequently recurring non-perfusing rhythms such as
pulseless ventricular tachycardia (VT) or ventricular fibrillation (VF),
but continue to achieve rapid return of spontaneous circulation, can be
stabilized on V-A ECMO thus preventing frequent severe hypoperfusion
of end organs with each arrythmia. Examples include toxic ingestions
such as sodium channel blockers or QT-interval prolonging agents. This
is in contrast to the classic ECPR patient who is cannulated for refractory
cardiac arrest. Additional examples include patients with underlying car-
diomyopathy who are in cardiogenic shock due to a difficult to control
rhythm such as rapid atrial fibrillation or ventricular tachycardia or pa-
tients with frequent or malignant ventricular arrhythmias at risk of de-
grading to cardiac arrest.

Massive Pulmonary Embolism

Any patient presenting with massive or submassive pulmonary embolism
(PE) is at risk for acute right ventricular (RV) failure. The most com-
mon scenario is cardiac arrest during or shortly after intubation. In se-
lect cases, use of inotropes prior to induction may mitigate hemodynamic
collapse. In all cases, avoidance of hypotension, hypoxemia, and hyper-
carbia is paramount. Overall, failure to recognize severe right ventricular
dysfunction is a driver of poor outcomes in this population. A number
of hemodynamic calculations such as pulmonary artery pulsatility index
(PAPi), or CVP:wedge ratio, may guide assessment of RV function. High
CVP with unexpectedly low PA pressures, in particular narrow PA pulse
pressure, is worrisome. On echo, RV dilatation, reduced tricuspid annular
plane systolic excursion (TAPSE), and severe tricuspid valve regurgita-
tion signal RV failure.

Initial Medical Management of the Patient in

Shock
While earlier initiation of support improves outcomes, the amount of time
needed for end-organ damage to become irreversible is unknown. Some

Chapter 13 223

 V-A ECMO for Shock
ɋ Maximize medical therapies
ɋ Decision to Initiate V-A ECMO
• Consider in cardiogenic shock SCAI stages C, D and E, electri-
cal instability/arrythmias, and massive PE
• Must assess risk of current situation (ED, ICU) vs. benefit of
transport to fluoroscopy suite (IR, Cath lab)
• Experience of cannulators, experience of nurses, available
equipment, and time delays associated with transport should
be considered
ɋ Initiation
• Take time to assess femoral vessel size (ultrasound) and pre-
vious surgical interventions (femoral, popliteal scars)
• Full sterile technique with trained wire assistant
• Can often place distal perfusion catheter prior to arterial
ECMO cannula
• “Slow is smooth, smooth is fast”
ɦ Take time for success of “first stick” femoral artery wire
insertion
ɦ Confirm proper vessel cannulation via ultrasound or x-ray
wire confirmation
ɦ Ensure smooth wire and dilator insertion to avoid wire
kinking, backwall damage to the vessel, or retrograde
perforation of a cannula
ɦ Do not overdilate or over-insert cannula
• Consider “awake” cannulation in the spontaneously breathing
patient at risk of cardiac arrest on induction
ɋ Post initiation
• Assess left ventricular distention with echocardiogram
• Increase contractility with inotropes and decrease afterload
by decreasing vasopressors as able, preventing hypertension
and/or avoiding excessive ECMO flow
• Mechanical venting of the LV if medical management is inad-
equate

224 ECPR and Resuscitative ECMO

authors have suggested that as little as four hours of profound shock may
lead to irreversible cellular damage.8 When reviewing the ELSO reports of
outcomes from cardiac support and ECPR, nearly 20% of patients wean
from ECMO support but die prior to discharge.2 A large portion of those
patients experience multisystem organ failure, which may be unrecov-
erable or may disqualify the patient from advanced heart failure thera-
pies. Timely restoration of cardiac output may mitigate this catastrophic
end-organ damage.9 After ECMO initiation, assessment should be made
to whether the patient will have bridge to recovery versus bridge to dura-
ble left ventricular assist device (LVAD) or cardiac transplantation.
Common diagnostic studies can aid in the identification of cardio-
genic shock, and may guide treatment options, as well. In addition to
ischemic changes on electrocardiography, heart rate and rhythm are im-
portant factors in shock management. Treatment of advanced heart block
or profound bradycardia are simple methods to improve cardiac output,
while tachycardia above the 130s may indicate compensation for very low
stroke volume. Additionally, restoration of sinus rhythm from atrial fibril-
lation may improve cardiac output by 20%. Cardiogenic shock occurs in
upwards of 10% of ST elevation myocardial infarctions but may also be
seen in non-ST elevation myocardial infarction (NSTEMI) and from right
coronary infarctions.10
Invasive monitoring should be performed in patients presenting with
cardiogenic shock. An arterial line allows for frequent blood gas analy-
sis as well as continuous blood pressure measurement. Central venous
access is useful for measurement of ScvO2 which is key in determining
cardiac output in the absence of a pulmonary artery catheter (PAC). If
available, a PAC allows targeted therapy, potentially earlier recognition of
the need for mechanical circulatory support, and assessment of right, left
or biventricular failure.5
The initial treatment of any patient with signs and symptoms of car-
diogenic shock includes intravenous inotropes. Dobutamine should be
the first line treatment for low cardiac output.11 Norepinephrine is often
started slightly before or along with dobutamine if hypotension is present
in order to augment aortic root pressure and thus ensure adequate coro-
nary artery perfusion pressure.12 Second line agents include epinephrine
or vasopressin. Patients who have transient or no response to incremental
dose increases of these medications are in stage C to D cardiogenic shock
and at high risk for ongoing hemodynamic collapse and death.

Chapter 13 225

 All shock states should be maximally medically managed in concert
with the decision to initiate ECMO. Management includes vasoactive
drugs, antiarrhythmics with rhythm instability, and thrombolytics in mas-
sive pulmonary embolism. The one significant management difference in
an ECMO center would be initiating mechanical support prior to intuba-
tion. Intubation may cause instability and potential arrest which can be
alleviated by ECMO initiation prior to or in lieu of mechanical ventilation.
Cannulation in the awake patient has inherent challenges. Mild sedation
may prevent an arrest that renders cannulation much more difficult. Even
insertion of wires in the femoral vessels prior to intubation may be pru-
dent in the patient in extremis.

Other Mechanical Circulatory Support Devices

There is a spectrum of mechanical support devices available for patients
in cardiogenic shock. The most widely available is the intra-aortic balloon
pump (IABP), which can be placed by a broad range of operators and does
not require specialized imaging for the procedure. Only minimal enhance-
ment of cardiac output occurs after placement of a balloon pump, gener-
ally 0.5-1 liter per minute.10 Specific diseases, such as coronary ischemia
and acute mitral regurgitation may benefit more; however, the SHOCK
trial failed to show a benefit of balloon pump prior to revascularization for
acute myocardial infarction.13 The primary indication for placement of an
intra-aortic balloon pump in refractory cardiogenic shock is no availability
of any other mechanical support options.
The most commonly used mechanical support device in cardiogenic
shock is the percutaneous microaxial left ventricular assist device, such as
Impella®.10 Safe placement of this pump generally requires fluoroscopy to
allow crossing of the aortic valve with a catheter and guidewire as well as
delivery of the pump through the arterial vasculature. Relevant contrain-
dications include severe peripheral vascular disease, severe aortic steno-
sis, aortic dissection, thrombus in the left ventricle, and ventricular septal
defect. The percutaneously placed devices may improve cardiac output up
to 3.5-5 liters per minute. Other devices capable of high levels of support,
such as the TandemHeart® (TandemLife®), are less able to be rapidly
deployed in the emergency setting. Both of these pumps provide single
ventricle support, which may be inadequate in a patient with biventricular

226 ECPR and Resuscitative ECMO

 Echocardiogram, MAP,
ScvO2

Severe ventricular Moderate ventricular

dysfunction dysfunction
SCAI “D” or “E” SCAI “C” or “D”

1○ LV Bi-V or RV
dysfunction dysfunction

No hypoxemia:
Hypoxemia: VA Impella® CP or
Impella® 5.0-5.5
VA ECMO ECMO IABP
or TandemHeart®

Figure 1: Algorithm for selecting temporary mechanical circulatory sup-

port in a patient with cardiogenic shock. Mean arterial pressure (MAP),
central venous oxygen saturation (ScvO2), Society for Cardiovascular Angi-
ography and Intervention (SCAI), left ventricle (LV), right ventricle (RV),
biventricular (Bi-V), intra-aortic balloon pump (IABP).

failure. Data regarding the benefits of these pumps is limited, and most
studies show results similar to the IABP.14 A microaxial RV assist device is
available at select medical centers as well. Limitations of this device are
similar to the left sided pump. A dual lumen RV cannula using an ECMO
pump is another potential option for RV support (ProtekDuo®, Tandem-
Life®). Decisions about which device to choose is difficult. Figure 1 gives
a potential algorithmic approach to this decision making.

V-A ECMO
Major benefits of ECMO compared to other temporary support options
are near-complete replacement of native heart function, the ability to
enhance oxygenation, rapid deployment, no need for specialized imag-
ing equipment, and portability.15 ECMO is the only single device which
provides biventricular support. Contraindications to ECMO vary by insti-

Chapter 13 227

tution and are mostly relative in nature. These include chronic multiple
organ dysfunction, advanced age, bleeding or clotting disorders, severe
peripheral vascular disease, aortic insufficiency, or limited life expectancy
from noncardiac medical problems.16

Management Differences for V-A ECMO Compared to ECPR

The decision to initiate ECMO in the shock state must involve an accu-
rate assessment of the cannulator’s skill combined with an assessment of
the patient’s risk of morbidity/mortality. Literature evaluating the risk of
cannulation varies but roughly 15% of patients with femoral cannulation
develop limb ischemia and 1% have retroperitoneal bleeding.17-20 These
studies are predominantly in experienced centers with cardiovascular sur-
geons performing the cannulation. Mortality is significantly affected by
vascular complication with one study showing more than a 20% decrease
in survival to discharge in those patients with a reported vascular compli-
cation.17
After the decision to initiate ECMO is made, the practitioner must
decide where to cannulate the patient. Patient status, practitioner expe-
rience and hospital resources should all factor into this decision. Many
times the safest place for cannulation initiation is the patient’s current lo-
cation.21 Transport risks adverse outcomes secondary to decreased mon-
itoring, decreased space constraints within elevators and hallways, and
potential for arrest while in transit. Advantages of transport may include
fluoroscopy or human resources not available in the current location.
Patients who have the luxury of time should have increased attention
to the details of cannulation. The steps of cannulation in this setting are
the same, although impeccable technique should be used to decrease the
risk of complications. Pre-scanning the intended targets for cannulation
can help guide cannula size choice and ensure patency of the vessels. Also,
placement of a distal perfusion catheter can be done prior to arterial can-
nulation to allow immediate reperfusion of the lower leg.
The decision to initiate ECMO in a rapidly decompensating patient
can be challenging; however, the benefits of avoiding profound shock with
multisystem organ failure or cardiac arrest are significant.22 Patients in
Stage C, D, or E cardiogenic shock, who do not respond to increases in
inotropes or vasopressors should be considered for mechanical circula-
tory support.6

228 ECPR and Resuscitative ECMO

 Any patient who has experienced a cardiac arrest is at risk for recur-
rence as the myocardium is stunned by the insult which led to the arrest
plus the effects of chest compressions. Even with return of spontaneous
circulation during a cardiac arrest, there should be a low threshold for
initiation of ECMO to maintain stability.21 Outcomes from standard CPR
are poor even in hospitalized patients. Patients with frequent or malig-
nant ventricular arrhythmias are at risk of degrading to cardiac arrest and
therefore, should be considered for mechanical support early in their care.
Inotropes in these patients may worsen arrhythmias, and antiarrhythmics
tend to worsen cardiac function.
In patients with RV failure, especially from pulmonary embolism, the
need for intubation can lead to cardiac arrest.21 In such patients, prepara-
tion of the groin vessels by placement of arterial and venous access prior
to induction can decrease the potential for disaster. Select centers may
place these patients on ECMO while awake to avoid the derangements
associated with intubation. Overall, patients with primary RV failure,
biventricular failure, or severe respiratory and cardiac failure should be
supported with ECMO, as no other mechanical support option to treat
these conditions is readily available in most centers.15 Also, hemodynam-
ically unstable patients requiring transfer to another facility may benefit
from initiation of ECMO in the emergency department prior to transfer,
if this can be done with a low risk of complications by an experienced
operator.

Post Initiation Management

After initiating ECMO, secure the cannulas with sutures or securement
devices per local practice, as these patients will be transported to differ-
ent locations and transferred to another bed in the hospital. Dislodge-
ment of a cannula is disastrous and potentially fatal.
After stabilization, the patient should undergo diagnostic studies and
therapies to reverse the cause of cardiogenic shock, when possible. This
may require radiology, interventional cardiology, or cardiothoracic sur-
gery.
Assessment of LV distention should be made as soon as possible with
echocardiogram. Interventions in the emergency department are limited
to the use of inotropes to promote LV contractility. See Chapters 7 and 9
for an in-depth explanation of LV distention.

Chapter 13 229

Study Design n Data Source Etiology of shock Survival
Lorusso Cohort 57 13 centers Myocarditis Discharge
2016 Italy & UK 71.9%
Aso Cohort 4,658 > 1k centers Ischemic 42.2%, ADHF 34.8%, valvular 13.7%, Discharge
2016 Japan myocarditis 4%, cardiomyopathy 4.1% 26.4%
Truby Cohort 179 Single center Post-cardiotomy 39%, MI 26%, 30-day
2015 New York City primary graft failure 10%, ADHF 13% 44.7%
Biancari Meta- 2,986 31 studies Post-cardiotomy Discharge
2018 analysis International 36.1%
Meneveau Case 52 9 centers Pulmonary embolism 30-day
2018 series France (75% were post cardiac arrest) 39%

Table 2: Select observational studies showing survival rates for various

forms of shock supported with V-A ECMO.

Outcome Data
Clinical data is limited to observational studies secondary to lack of clin-
ical equipoise adding to the challenge of deciding which patients would
benefit from V-A ECMO. In the ELSO Registry, 44% of patients receiving
ECMO for cardiac support are discharged from the hospital alive.2 Several
case series suggest improved survival in post-MI cardiogenic shock when
V-A ECMO is used in addition to revascularization.23 Table 2 highlights se-
lect observational studies exploring survival for various etiologies of shock
supported on V-A ECMO24-28. Complications are common, including lower
extremity ischemia or amputation, stroke, major bleeding, and infection.29

Conclusion
Initiation of V-A ECMO should be considered in a hemodynamically de-
compensating patient with cardiogenic shock, rhythm instability, or mas-
sive PE when appropriate personnel and equipment are available to avoid
delays leading to worse outcomes.

230 ECPR and Resuscitative ECMO

References
1. Unlüer EE, Karagöz A, Akoğlu H, Bayata S. Visual estimation of bed-
side echocardiographic ejection fraction by emergency physicians.
West J Emerg Med. 2014;15(2):221-226.
2. ELSO Registry International Summary, July 2020. https://www.elso.
org/Registry/Statistics/InternationalSummary.aspx. Updated July
2020. Accessed August 4, 2020.
3. Amado J, Gago P, Santos W, Mimoso J, de Jesus I. Cardiogenic
shock: Inotropes and vasopressors. Rev Port Cardiol. 2016; 355(12)
L 681-695.
4. Tongers J, Sieweke JT, Kühn C, et al. Early escalation of mechanical
circulatory support stabilizes and potentially rescues patients in
refractory cardiogenic shock. Circ Heart Fail. 2020;13(3):1-11.
5. Baran DA, Grines CL, Bailey S, et al. SCAI clinical expert consensus
statement on the classification of cardiogenic shock. Catheter Car-
diovasc Interv. 2019; 94:29-37.
6. Jentzer JC, van Diepen S, Barsness GW, et al. Cardiogenic shock
classification to predict mortality in the cardiac intensive care unit. J
Am Coll Cardiol. 2019; 74(17): 2117-28.
7. Rao P, Khalpey Z, Smith R, Burhoff D, Kociol RD. Venoarterial extra-
corporeal membrane oxygenation for cardiogenic shock and cardiac
arrest: cardinal considerations for initiation and management. Circ
Heart Fail. 2018; 11(9):1-17.
8. Squara P, Hollenberg S, Payen D. Reconsidering vasopressor for car-
diogenic shock. Chest. 2019; 156(2):392-401.
9. O’Brien C, Beaubien-Souligny W, Amsallem M, Denault A, Haddad
F. Cardiogenic shock: reflection at the crossroad between perfu-
sion, tissue hypoxia, and mitochondrial function. Can J Cardiol.
2109;36:184-196.
10. Vahdatpour C, Collins D, Goldberg S. Cardiogenic Shock. J Am Heart
Assoc. 2019;8:e011991.
11. Polyzogopoulou E, Arfaras-Melainis A, Bistola V, Parissis J. Ino-
tropic agents in cardiogenic shock. Curr Opin Crit Care. 2020;
26(4):403-410.
12. Levy B, Klein T, Kimmoun A. Vasopressor use in cardiogenic shock.
Curr Opin Crit Care. 2020; 26(4): 411-416.

Chapter 13 231

13. van Diepen S, Katz JN, Albert NM, et al. Contemporary Management
of Cardiogenic Shock: a scientific statement from the American
Heart Association. Circulation. 2017;136:e232-268.
14. Schrage B, Ibrahim K, Loehn T, et al. Impella support for acute myo-
cardial infarction complicated by cardiogenic shock: Matched-Pair
IABP-SHOCK II trial 30-day mortality analysis. Circulation. 2019;
139(10):1249-58.
15. Guglin M, Zucer MJ, Bazan VM, et al. Venoarterial ECMO for adults.
J Am Coll Cardiol. 2019; 73(6):698-716.
16. ELSO Guidelines for Adult Cardiac Failure, Version 1.3. https://
www.elso.org/Resources/Guidelines.aspx. Updated December 2013.
Accessed August 4, 2020.
17. Wong JK, Melvin AL, Joshi DJ, Lee CY, Archibald WJ, Angona
RE, Tchantchaleishvili V, Massey HT, Hicks GL, Knight PA.
Cannulation-Related Complications on Veno-Arterial Extracorporeal
Membrane Oxygenation: Prevalence and Effect on Mortality. Artif
Organs. 2017 Sep;41(9):827-834.
18. Yang, F., Hou, D., Wang, J. et al. Vascular complications in adult post-
cardiotomy cardiogenic shock patients receiving venoarterial extra-
corporeal membrane oxygenation. Ann. Intensive Care 8, 72 (2018).
19. Augusto R, Passos Silva M, Campos J, Coelho A, Coelho N, Semião
AC, Brandão D, Canedo A. Arterial Vascular Complications in
Peripheral Venoarterial Extracorporeal Membrane Oxygenation Sup-
port. Rev Port Cir Cardiotorac Vasc. 2019 Jan-Mar;26(1):45-50.
20. Bonicolini E, Martucci G, Simons J, Raffa GM, Spina C, Coco VL,
Arcadipane A, Pilato M, Lorusso R. Limb ischemia in peripheral
veno-arterial extracorporeal membrane oxygenation: a narrative re-
view of incidence, prevention, monitoring, and treatment. Crit Care.
2019 Jul 30;23(1):266.
21. de Chambrun MP, Bréchot N, Combes A. The place of extracorpo-
real life support in cardiogenic shock. Curr Opin Crit Care. 2020;
26(4):424-431.
22. Kapur NK, Davila CD. Timing, timing, timing: the emerging concept
of the ‘door to support’ time for cardiogenic shock. Eur Heart J.
2017;38:3532-3534.
23. Mehta H, Eisen HJ, Cleveland JC. Indications and complications for
VA-ECMO for cardiac failure. Am Col Cardiol Latest in Cardiology.
https://www.acc.org/latest-in-cardiology/articles/2015/07/14/09/27/

232 ECPR and Resuscitative ECMO

 indications-and-complications-for-va-ecmo-for-cardiac-failure. Pub-
lished July 14, 2015. Accessed August 30, 2020.
24. Lorusso R, Centofanti P, Gelsomino S, Barili F, Di Mauro M, Orlando
P, Botta L, Milazzo F, Actis Dato G, Casabona R, Casali G, Musumeci
F, De Bonis M, Zangrillo A, Alfieri O, Pellegrini C, Mazzola S, Coletti
G, Vizzardi E, Bianco R, Gerosa G, Massetti M, Caldaroni F, Pilato
E, Pacini D, Di Bartolomeo R, Marinelli G, Sponga S, Livi U, Mauro
R, Mariscalco G, Beghi C, Miceli A, Glauber M, Pappalardo F, Russo
CF; GIROC Investigators. Venoarterial Extracorporeal Membrane
Oxygenation for Acute Fulminant Myocarditis in Adult Patients:
A 5-Year Multi-Institutional Experience. Ann Thorac Surg. 2016
Mar;101(3):919-26.
25. Aso S, Matsui H, Fushimi K, Yasunaga H. In-hospital mortality and
successful weaning from venoarterial extracorporeal membrane
oxygenation: analysis of 5,263 patients using a national inpatient
database in Japan. Crit Care. 2016 Apr 5;20:80.
26. Truby L, Mundy L, Kalesan B, Kirtane A, Colombo PC, Takeda K,
Fukuhara S, Naka Y, Takayama H. Contemporary Outcomes of
Venoarterial Extracorporeal Membrane Oxygenation for Refractory
Cardiogenic Shock at a Large Tertiary Care Center. ASAIO J. 2015
Jul-Aug;61(4):403-9.
27. Biancari F, Perrotti A, Dalén M, Guerrieri M, Fiore A, Reichart D,
Dell’Aquila AM, Gatti G, Ala-Kokko T, Kinnunen EM, Tauriainen T,
Chocron S, Airaksinen JKE, Ruggieri VG, Brascia D. Meta-Analysis
of the Outcome After Postcardiotomy Venoarterial Extracorporeal
Membrane Oxygenation in Adult Patients. J Cardiothorac Vasc
Anesth. 2018 Jun;32(3):1175-1182.
28. Meneveau N, Guillon B, Planquette B, Piton G, Kimmoun A,
Gaide-Chevronnay L, Aissaoui N, Neuschwander A, Zogheib E,
Dupont H, Pili-Floury S, Ecarnot F, Schiele F, Deye N, de Prost N,
Favory R, Girard P, Cristinar M, Ferré A, Meyer G, Capellier G, San-
chez O. Outcomes after extracorporeal membrane oxygenation for
the treatment of high-risk pulmonary embolism: a multicentre series
of 52 cases. Eur Heart J. 2018 Dec 14;39(47):4196-4204.
29. Persico N, Bourenne J, Roch A. Veno-arterial extracorporeal mem-
brane oxygenation for acute myocardial infarction-associated cardio-
genic shock: can we predict survival before decision of implantation?
J Thor Dis. 2016;8(9):2331-3.

Chapter 13 233

CHAPTER 14
ECMO in Trauma

Magnus Larsson • Matthias Arlt

Introduction
Trauma is a leading cause of death in people under the age of 40
years1. The most common reasons for immediate death at the scene are
traumatic brain injury, high spinal lesions, and heart/ great vessel injury.
After arrival at the hospital, hemorrhage and coagulopathy are the main
reasons for death.2 In the treatment of trauma, time is critical. To save
time and lives after trauma, prehospital teams that quickly transport in-
jured patients to a well-organized trauma center with staff adhering to the
principles of Advanced Trauma Life Support (ATLS)3 are essential.
Modern trauma care focuses on the principles of damage control
resuscitation4 and damage control surgery5 to restore and maintain the
patient’s cardiocirculatory and cardiopulmonary function until definitive
surgical trauma care6 can be performed.

Foundational Terms
ɋ Vicious Cycle of Hemorrhagic Shock – includes bleeding, coag-
ulopathy, dilution of factors secondary to instilled crystalloid,
acidosis, and hypothermia. These last two further contribute to
coagulation deficits. ECMO is capable of rapidly correcting these
deficits.

 235
Refractory Cardiopulmonary and
Cardiocirculatory Deterioration after Severe
Trauma
Severe trauma and injudicious fluid resuscitation including massive
blood transfusion can lead to severe acute respiratory distress syndrome
(ARDS).7. Cases refractory to conventional ventilator support therapy
may benefit from venovenous (V-V) ECMO. Intractable bleeding requir-
ing complex source control surgery can lead to refractory severe acidosis,
hypothermia, coagulopathy, and death due to circulatory failure. These
patients can be supported with venoarterial (V-A) ECMO to reestablish
adequate hemodynamics and homeostasis.

The Role of Extracorporeal Membrane Oxygenation (ECMO)

in Trauma
ECMO was first used in trauma by Dr. Donald Hill and his team in
1971.8 The miniaturization of the ECMO circuit and development of
heparin-coated circuits has allowed for its application in austere environ-
ments, including alongside the chain of rescue of severe trauma patients
with refractory cardiopulmonary failure and hemorrhagic shock. V-A
ECMO offers unique beneficial advantages for the trauma patient. Ani-
mal studies have shown that V-A ECMO can normalize temperature, pH,
lactate, and coagulation.9 It also increases arterial pressure and reduces
central venous pressure, decreasing bleeding from large veins.10

Indications for ECMO Support in Severe Trauma

Careful patient selection is important. No universally agreed upon indi-
cations for the application of ECMO in trauma exist. However, a system-
atic review of ECMO in trauma patients found that “the use of ECMO
in trauma when needed may provide survival benefits that significantly
[outweigh] the feared risk of bleeding”.11 ECMO is a rescue strategy
for patients with cardiopulmonary failure unresponsive to conventional
treatment.12,13,14
The most common indications for V-V ECMO in trauma include hy-
poxemic states such as ARDS refractory to advanced ventilator support.

236 ECPR and Resuscitative ECMO

V-V ECMO can also be used for airway injuries resulting in inability to
effectively ventilate. Indications for V-A ECMO support include hemor-
rhagic cardiac arrest or persisting shock despite blood resuscitation and
vasopressor support.

Limitations for ECMO in Severe Trauma

In the gravely ill patient with hemorrhagic shock, there are no absolute
contraindications to ECMO. Given the potential complications associated
with ECMO support and the requisite diversion of resources, a rapid risk/
benefit assessment of its use is prudent. Femoral vessel injury may make
cannulation more difficult and should be included in this risk/benefit as-
sessment. Additionally, careful consideration should be paid to identify
cases of futility such as major vessel injury identified on computed tomog-
raphy (CT) scan or surgical observation. However, severe bleeding is not
a contraindication for using ECMO as support can be initiated without
administration of an anticoagulant (e.g., heparin).

ECMO Cannulation and Circuits in Trauma

Modern ECMO systems are coated with heparin or non-heparin poly-
mers that enhance biocompatibility of the plastic surface of the circuit,
thus reducing thrombosis risk. This coating enables delay of systemic
heparin initiation until bleeding has stopped and the coagulation status
normalizes for at least for 48-72 hours.13 Furthermore, trauma patients
with hemorrhagic shock often experience coagulopathy, and thus are not
in need of anticoagulation. The risk from bleeding related mortality in
ECMO-trauma patients is probably overestimated. Mortality related to
ECMO treatment in trauma has been reported to be less than 15%.15.
Cannulation of the polytrauma patient is optimally performed simul-
taneously with damage control surgery, ideally by a physician indepen-
dent of the trauma-team. Spine protection during cannulation of the neck
is crucial. Cannulation of the common femoral artery and veins either
percutaneously or via “cut down” is the best choice. ECMO cannulation
will thus not obstruct the simultaneously ongoing surgery in the chest,
abdomen, or pelvis.
As previously stated, V-V ECMO can be utilized in the setting of severe
pulmonary/airway injury or posttraumatic ARDS; whereas, V-A ECMO

Chapter 14 237

should be considered in trauma patients in bleeding shock and those
with cardiac failure. The venous drainage cannula must be long enough to
reach the center of the right atrium. In extensive bleeding from both sides
of the diaphragm, it is better to drain both from vessels in the upper and
lower body. To do this, a second venous drainage cannula can be inserted
via the internal jugular vein (dual cannula drainage VV-A ECMO). The
arterial return cannula is introduced through the common femoral artery,
distant from potentially large bleeding organs. A distal perfusion cannula
on the ipsilateral leg should be considered to avoid limb ischemia. Our
institution considers this a mandatory intervention.

Effectiveness of ECMO Support in Trauma

Trauma patients are often young and previously healthy; thus, they are
ideal ECMO patients. ECMO may be the only option to keep the patient
alive until full surgical repair and restoration of homeostasis have been
achieved. An overall 70% survival rate from ECMO in trauma has been
reported with survival to hospital discharge in 61% (63% respiratory, 50%
cardiac, 25% ECPR).15

Patient Management on ECMO in Trauma

Hemorrhagic Shock and ECMO
ECMO should be considered as early as possible if the patient fulfills the
inclusion criteria of ECMO support. Ideally, ECMO should be initiated
prior to the onset of the hypothermia, acidosis, and coagulopathy triad
and in parallel to the other definitive trauma treatment. The ECMO can-
nulae are optimal for infusion of large amounts of blood products and
can be integrated into massive transfusion systems. The heat exchanger
integrated in the circuit can rapidly restore the normothermia. The mem-
brane lung improves oxygenation, normalizes CO2 and eventually pH.
Damage control surgery must be performed simultaneously while
ECMO support is initiated. This is crucial to improve the patient’s circu-
latory function, gas exchange, and homeostasis but also to maintain the
blood volume for circulation in the ECMO circuit.

238 ECPR and Resuscitative ECMO

 Traumatic Injuries Supportable with ECMO
ɋ V-A ECMO:
• Exsanguination: allows large bore access and rapid infusion of
blood products via ECMO circuit
• Severe cardiac, great vessel or pulmonary trauma: bypass
during thoracotomy and repair
• Advanced liver injuries: bypass to allow Heaney maneuver
ɋ V-V ECMO:
• Airway trauma & inability to intubate or cric
• Severe bronchopleural fistula
• Inhalational burns and ARDS

Intracranial Hemorrhage on ECMO

Although reports are few and studies are lacking in this patient group,
some centers have reported favorable outcomes for a limited number of
ECMO patients with coexisting intracranial hemorrhage.16
Ventriculostomy for intracranial pressure (ICP) monitoring as well
as craniotomy with evacuation of hematomas have been performed
during ECMO.17 These procedures are risky but possible when performed
cautiously and with cessation of anticoagulation. Infusion of recombinant
activated factor VII and prothrombin complex may be necessary. Survival
ranges of 60-93% have been reported; neurosurgery may be performed
during ECMO without increased bleeding and should be performed when
clinically indicated.18

Airway Injuries
Trauma to the upper airways due to gunshot wounds, blast injuries, rup-
ture of the trachea, or airway obstruction resulting in inability to intubate
or ventilate can be treated with V-V or V-A ECMO support until surgical
exposure and treatment can be performed. ECMO has been employed on
newborns suffering with ruptured trachea due to vacuum-extraction,19 as
well as adults with distal tracheal injuries just above the sternal notch,
making acute cricothyrotomy impossible.20 Additionally, central airway

Chapter 14 239

injuries to the bronchi or severe injury requiring pneumonectomy can be
temporized with V-V ECMO.31-34

Burn Injuries
Severe burn injuries are commonly complicated by ARDS, affecting ap-
proximately 40% of mechanically ventilated burn patients. It is caused by
smoke inhalation, ventilator-associated pneumonia, fluid resuscitation,
and inflammation. Respiratory failure is responsible for a mortality rate
of 40% in this group. ECMO has not been used commonly on burn pa-
tients, but excellent survival has been reported (87.5%).21 So far ECMO
has mainly been used in pediatric burn patients. Early consultation of the
ECMO team when burn patients develop declining pulmonary function is
important for appropriate patient selection.

Thoracic Injuries
Thoracic trauma is one of the leading reasons for death and accounts for
25-50% of all traumatic injuries. The majority (90%) of thoracic injuries
can be treated conservatively or with tube thoracostomy. When emer-
gency thoracotomy is needed, the survival rate is below 10%.22 Before
ECMO is considered, other immediate lifesaving treatments should be
considered, e.g., tube thoracostomy for evacuation of tension pneumo-
thorax or hemothorax, or pericardiocentesis for tamponade.
Emergency thoracotomy can be combined with immediate cannula-
tion for V-A or VV-A ECMO. This secures systemic perfusion and satura-
tion of the brain, heart, and other vital organs. However, the aorta should
NOT be cross-clamped if the arterial return cannula is in the femoral ar-
tery. Parallel to the cannulation, the trauma surgeon can open the tho-
racic cavity, evacuate cardiac tamponade, control the bleeding heart or
great vessels, or clamp the pulmonary hilus when air embolism or severe
pulmonary hemorrhage is present. ECMO buys time for surgery and also
reduces the blood flow through the bypassed heart and lungs. The ven-
tilator can be stopped with the endotracheal tube clamped in order to
tamponade the bleeding pulmonary tree and avoid air leakage.
Literature on survival in thoracic trauma patients supported with
ECMO is sparse but several case series show dramatic survival stories as-

240 ECPR and Resuscitative ECMO

sociated with such injuries. A case series from Korea showed that all five
of their trauma ECPR cases who experienced auricle or pericardial rup-
ture survived to discharge.30

Advanced Liver Injuries

In cases of advanced liver injuries (grade IV and V) the Pringle’s maneu-
ver and perihepatic packing may not be enough to arrest bleeding. The
only option may be to clamp both the suprahepatic and infrahepatic in-
ferior vena cava (Heaney maneuver). This will eventually lead to cardiac
arrest because of the decrease in cardiac preload. Under these circum-
stances V-V ECMO can bypass the liver, maintain cardiac preload, and
preserve systemic perfusion. This provides time to control arterial and ve-
nous bleeding either with open surgery or via angiographic coiling. There
have been a few cases described of acute liver transplantation on bypass
following trauma.23

Pediatric Trauma and ECMO

Studies of ECMO in severe pediatric trauma are scarce. A systematic
review from 2020 concluded that ECMO use should be considered in
severely injured children. A survival rate of 50-100% was similar to out-
comes reported in the adult population. Neither head injuries nor intraab-
dominal injuries served as contraindications. Both post-trauma ARDS and
acute polytrauma with cardiovascular collapse were reported. In the larg-
est study, 47% of the children were cannulated at adult ECMO centers,
suggesting that these centers should also consider pediatric cannulation
in severe trauma.24

Evacuation-Medevac and ECMO

Transportation of trauma patients on ECMO can be a safe way to evacuate
a patient with pulmonary and/or cardiac failure. It is associated with very
few complications and has been done both in civil and military settings.25
Transportation of soldiers with severe respiratory failure on ECMO from
Afghanistan showed a survival rate of 90%.26

Chapter 14 241

 Advantages of ECMO in severe trauma:
1. Provides cardiopulmonary support and a bridge to damage
control surgery
2. Increases the mean arterial pressure and reduces the central
venous pressure
3. Normalizes body temperature
4. Provides oxygenation and normalizes the acid-base balance
5. Can bypass injured vital organs such as the heart, lungs or liver
6. Offers the largest possible infusion lines for blood products
including rapid reversal of coagulopathy

Limitations of ECMO in severe trauma:

1. Blood volume must be large enough for circulation in the ECMO
circuit
2. Can cause iatrogenic bleeding during cannulation

Emergency Preservation and Resuscitation

A unique consideration in traumatic arrest is the idea of utilizing ECMO
to cool patients to very cold temperatures (10-15ºC) in hopes of slowing
metabolism such that the time needed to repair these injuries would not
create neurologic devastation or organ dysfunction. Work is still active on
this with one center attempting a clinical trial.27-29

Conclusion
ECMO is not contraindicated in the severely bleeding trauma patient. In
fact, ECMO might provide the only chance to save the life of a patient
with devastating injuries.
ECMO is a common resource at level 1 trauma centers. It provides
safe and effective hemodynamic support and gas exchange in trauma pa-
tients when conventional therapies do not suffice. Even patients in hem-
orrhagic shock can be treated using heparin-free ECMO. When indicated,
ECMO support should be established alongside damage control surgery

242 ECPR and Resuscitative ECMO

as early as possible. ECMO can normalize body temperature, optimize
acid-base balance, and assist in cessation of the vicious circle of coagulop-
athy. Currently, ECMO in trauma is in its infancy and universally agreed
upon indications are lacking. Trauma surgery has made immense progress
in recent decades. To reduce the mortality in severely injured patients,
ECMO plays an important role.

References
1. Haagsma JA, Graetz N, Bolliger I, Naghavi M, et al. The global bur-
den of injury: incidence, mortality, disability-adjusted life years and
time trends from the Global Burden of Disease study 2013. Inj Prev.
2016 Feb;22(1):3-18.
2. Probst C, Zelle BA, Sittaro NA, Lohse R, Krettek C, Pape HC. Late
death after multiple severe trauma: when does it occur and what are
the causes? J Trauma. 2009 Apr;66(4):1212-7.
3. Advanced Trauma Life Support for Doctors, Edn. 10th. (American
College of Surgeons Committe on Trauma, Chicago; 2018)
4. Chad G. Ball Damage control resuscitation: history, theory and tech-
nique; Can J Surg. 2014 Feb; 57(1): 55–60
5. Schreiber, M.A. Damage control Syrgery. Critical care clinics 20,
101-118 (2004)
6. IATSIC Manual of Definitive Surgical Trauma Care, Edn. 3:d. (CRC
Press Taylor and Francis Group, 2011).
7. Matthay, MA. Acute respiratory distress syndrome. Nature Reviews
Disease Primers volume 5, Article number: 18 (2019)
8. Hill, J.D. et al. Prolonged extracorporeal oxygenation for acute
post-traumatic respiratory failure (shock-lung syndrome). The New
England journal of medicine 286, 629-634 (1972).
9. Larsson, M et al. Extracorporeal Membrane Oxygenation Improves
Coagulopathy in an Experimental Traumatic Hemorrhagic Model.
Eur J Trauma Emerg Surg. 2017 Oct;43(5):701-709
10. Larsson, M. Experimental extracorporeal membrane oxygenation
reduces central venous pressure: an adjunct to control of venous
hemorrhage? Perfusion, 25 (4) (2010), pp. 217-223

Chapter 14 243

11. Bedeir, K, Kelly, E. Extracorporeal Life Support in Trauma: Worth
the Risks? A Systematic Review of Published Series. J Trauma Acute
Care Surg. 2017 Feb;82(2):400-406
12. Arlt, M. Extracorporeal Membrane Oxygenation in Severe Trauma
Patients With Bleeding Shock. Resuscitation. 2010 Jul;81(7):804-9
13. Bonacchi, M. Extracorporeal life support in polytraumatized pa-
tients; Int J Surg. 33 (2016) 213-217
14. Tisherman, SA. Development of the Emergency Preservation and
Resuscitation for Cardiac Arrest From Trauma Clinical Trial. J
Trauma Acute Care Surg. 2017 Nov;83(5):803-809
15. Swol, J, Zonies D. Indications and Outcomes of Extracorporeal Life
Support in Trauma Patients. J Trauma Acute Care Surg. 2018 Jun;
84(6):831-837
16. Muellenbach, R. Prolonged Heparin-Free Extracorporeal Membrane
Oxygenation in Multiple Injured Acute Respiratory Distress Syn-
drome Patients With Traumatic Brain Injury. J Trauma Acute Care
Surg . 2012 May;72(5):1444-7
17. Friesenecker, B. E. Craniotomy during ECMO in a severely trauma-
tized patient. Acta Neurochir. (2005) 147: 993-996
18. Mohney, N. Successful use of Extracorporeal Membrane Oxygen-
ation for respiratory failure after cranial surgery. World Neurosur-
gery. (2018) 120: 426-429
19. Browaldh N, Li Y, Falk L, Ekborn A, Westrup B, et al. (2016) Laryn-
geal Injury in Neonate and Temporary Extracorporeal Membrane
Oxygenation. Ann Otolaryngol Rhinol. 3(12): 1145.
20. Aprile V. Extracorporeal membrane oxygenation in traumatic
tracheal injuries: a bold life-saving option.J Thorac Dis. 2019 Jul;
11(7): 2660–2663
21. Scott Eldredge, R. Effectiveness of ECMO for burn-related acute
respiratory distress syndrome. J Burns (2019) 45: 317-321
22. Hunt PA, Greaves I, Owens WA. Emergency thoracotomy in thoracic
trauma-a review. Injury. 2006 Jan;37(1):1-19.
23. Ahmed, N. Management of liver trauma in adults. J Emerg Trauma
Shock. 2011 Jan-Mar; 4(1): 114–119
24. Puzio, T. Extracorporeal life support in pediatric trauma: a system-
atic review. Trauma Surg Acute Care Open. 2019; 4(1)

244 ECPR and Resuscitative ECMO

25. Bein, T. Transportable extracorporeal lung support for rescue of
severe respiratory failure in combat casualties J Trauma Acute Care
Srg. 2012;73(6):1450-1456
26. McDonald Johnston A, Alderman JE. Thoracic Injury in Patients
Injured by Explosions on the Battlefield and in Terrorist Incidents.
Chest. 2020 Apr;157(4):888-897.
27. Tisherman S, Safar P, Radovsky A, Peitzman A, Sterz F, Kuboyama K.
Therapeutic deep hypothermic circulatory arrest in dogs: a resusci-
tation modality for hemorrhagic shock with “irreparable” injury. J
Trauma. 1990;30(7):836-847.
28. Tisherman S. Salvage techniques in traumatic cardiac arrest: thora-
cotomy, extracorporeal life support, and therapeutic hypothermia.
Curr Opin Crit Care. 2013;19(6):594-598.
29. Kutcher M, Forsythe R, Tisherman S. Emergency preservation and
resuscitation for cardiac arrest from trauma. Int J Surg. 2016;33(Pt
B):209-212.
30. Huh U, Song S, Chung S, et al. Is Extracorporeal Cardiopulmonary
Resuscitation Practical in Severe Chest Trauma? : A Systematic
Review in Single Center of Developing Country. J Trauma Acute Care
Surg. August 2017.
31. Martucci G, Panarello G, Bertani A, Occhipinti G, Pintaudi S, Arcadi-
pane A. Veno-venous ECMO in ARDS after post-traumatic pneumo-
nectomy. Intensive Care Med. 2013 Dec;39(12):2235-6.
32. Wang FY, Fang B, Yu ZH, Shao JS, Wen WB, Zhou LX. Severe tho-
racic trauma caused left pneumonectomy complicated by right trau-
matic wet lung, reversed by extracorporeal membrane oxygenation
support-a case report. BMC Pulm Med. 2019 Feb 6;19(1):30.
33. Schmoekel NH, O’Connor JV, Scalea TM. Nonoperative Damage
Control: The Use of Extracorporeal Membrane Oxygenation in Trau-
matic Bronchial Avulsion as a Bridge to Definitive Operation. Ann
Thorac Surg. 2016 Jun;101(6):2384-6.
34. Halonen-Watras J, O’Connor J, Scalea T. Traumatic pneumo-
nectomy: a viable option for patients in extremis. Am Surg. 2011
Apr;77(4):493-7.

Chapter 14 245

CHAPTER 15
Crash V-V ECMO

Phillip Mason • James Lantry

Venoarterial (V-A) extracorporeal membrane oxygenation (ECMO)

provides both direct cardiovascular and respiratory support and is the
configuration most often associated with crashing patients. Venovenous
(V-V) ECMO, which provides direct support only to respiratory function,
is more often initiated in a semi-urgent or even elective fashion. However,
V-V ECMO can be started emergently when a primary airway or respira-
tory problem leads to an established or impending shock state. Choosing

Foundational Terms
ɋ Recirculation – Phenomenon unique to V-V ECMO where oxy-
genated blood returned to the body from the ECMO machine is
immediately drawn back into the drainage cannula. This results
from having the drainage and access cannulas too close to one
another and can greatly reduce the efficacy of ECMO.
ɋ Crash V-V ECMO—emergent implantation of V-V ECMO when
a primary airway or respiratory problems leads to an impending
shock state
ɋ Rest ventilator settings—low pressure, low volume, low
respiratory rate, low FiO2 ventilator settings to avoid ventilator
induced lung injury for ARDS while on V-V ECMO
ɋ Rescue ventilator settings—increased ventilator settings used
during a circuit emergency to support gas exchange through the
native lung until ECMO flow can be reestablished

 247
V-V ECMO over V-A ECMO in cases where respiratory failure predom-
inates eliminates several risks including lower extremity ischemia, dif-
ferential hypoxemia, end-organ embolic events, and increased afterload
which may cause left ventricular distention.
In this chapter, we discuss conditions that may be appropriate for
crash V-V ECMO, relevant V-V ECMO physiology, technique modifica-
tions specific to this scenario, and initial management strategies.

Indications for Crash V-V ECMO

Rapidly Progressing Acute Respiratory Distress Syndrome
The most common indication for V-V ECMO is severe acute respiratory
distress syndrome (ARDS).1 Thresholds for ECMO vary but three criteria
are commonly used: ratio of PaO2 to FiO2 (P:F ratio) of <80 mm Hg for
six hours; P:F ratio <50 mm Hg for three hours; or pH <7.25 with PaCO2
>60 mm Hg for six hours.2,3 ARDS often evolves slowly and relative he-
modynamic stability is maintained permitting a deliberate, even elective
ECMO initiation. If physiologic derangements progress rapidly or are far
advanced when ECMO is considered, then crash V-V ECMO is required.
Hypoxemia may lead to a shock state in the setting of normal cardiac
function as a result of decreased oxygen content of arterial blood and
hypercapnia can exacerbate hypotension through systemic vasodilation.
However, most cases of severe ARDS requiring crash V-V ECMO result
from the combined effects of hypoxemia, hypercarbia, and high-pressure
ventilation on pulmonary vascular resistance and venous return, leading
to decreased cardiac output. In this scenario, despite significant hemo-
dynamic compromise, V-V ECMO remains the preferred initial configu-
ration. Persistent hypoxemia after drowning events represents a unique
scenario where the extracorporeal circuit provides the added benefit of
highly efficient rewarming.4,5

Refractory Asthma
In most cases of near-fatal asthma (NFA), lung protective mechanical
ventilation and aggressive medical therapy improve gas exchange and
reverse respiratory acidosis, though mortality still approaches 10% in
patients requiring mechanical ventilation.6,7 When severe respiratory

248 ECPR and Resuscitative ECMO

 Crash V-V ECMO Procedure – Dual
Catheter Strategy
ɋ Sterile preparation of femoral vein and right internal jugular vein
(or bilateral femoral vein) sites
ɋ Seldinger technique insertion of long 150 cm wire into right
internal jugular vein and right femoral vein (or bilateral femoral
veins)
ɋ Perform confirmatory radiograph or ultrasound/TEE that shows
jugular wires in the vena cava (not coiled in the heart)
ɋ Dilate vessels and insert 25 Fr venous drainage cannula into fem-
oral vein and 21 Fr short “arterial” return cannula into internal
jugular (or 21 Fr long “arterial” return cannula into the opposite
femoral vein)
ɋ Ensure red/reinfusion limb of ECMO tubing connects to internal
jugular cannula (or return cannula in femoral vein)
ɋ Increase RPMs to initial flow of 3.5-5 L/min
ɋ Set initial sweep gas at 2 L/min flow. Adjust off an initial ABG

acidosis persists despite optimal ventilator management, or when dy-

namic hyperinflation causes barotrauma or hemodynamic compromise,
V-V ECMO may be indicated. While there are no formal guidelines, the
authors recommend considering ECMO for severe asthma exacerbations
with pH <7.25 or plateau pressure (PPLAT) >35-40 cm H2O despite optimal
ventilator management and adjunctive therapies. Utilizing V-V ECMO in
NFA rapidly resolves respiratory acidosis, high ventilator pressure, and
hyperinflation, and has been associated with >80% survival to discharge.8

Airway Obstruction
V-V ECMO can provide gas exchange sufficient to sustain life in the com-
plete absence of native lung function, or during apnea, and has been uti-
lized with near uniform success in cases of upper airway obstruction or
difficult airway.9-12 Most reports describe elective initiation prior to at-
tempts at airway management or intervention. There are no reported
cases of use in failed rapid sequence intubation. Success in an unantic-

Chapter 15 249

ipated “can’t intubate/can’t ventilate” scenario, even with highly skilled
personnel and extensive preparation, would be unlikely. Cases in which
partial ventilation can be maintained after failed intubation may be more
amenable to crash V-V ECMO. Success is highly likely if V-V ECMO is
initiated prior to neuromuscular blockade or attempts at airway manage-
ment.

Anaphylaxis
Anaphylaxis causes multiple derangements including bronchospasm, up-
per airway edema, and systemic vasodilation. Most reports in anaphylaxis
describe the use of V-A ECMO, even ECPR.13-16 However, if upper airway
edema or bronchospasm are the predominant processes, V-V ECMO is
appropriate and its successful use has been described.17

Cannulation and Configuration

When the term “crash” is applied to any procedure it implies urgency and
a need to complete the procedure quickly. However, it is not meant to en-
courage shortcuts or elimination of steps which can lead to complications.
Patients who require crash procedures are also the least able to tolerate
complications. Skilled practitioners know that speed comes from training,
experience, planning, and deliberate execution, not from cutting corners.
Key steps, such as ultrasound guidance, wire confirmation with imaging,
proper skin preparation and draping, and serial dilation should not be
eliminated in the interest of saving time. The axiom “slow is smooth and
smooth is fast” applies to crash V-V ECMO cannulation. As with difficult
airways, crash V-V ECMO cannulation calls for the most experienced op-
erator available and is not suited to novice operators or junior level train-
ees.

Single vs. Dual Site

Dual site cannulation, with the drainage cannula placed through a femoral
vein and the reinfusion cannula placed through either the internal jugu-
lar or contralateral femoral vein, is the preferred strategy for crash V-V
ECMO. The procedure is technically easier than single site cannulation. It

250 ECPR and Resuscitative ECMO

Figure 1: (left) When using the femoral-femoral V-V ECMO configuration
the drainage cannula, which can be identified by its larger diameter and
multiple drainage holes (arrows), should be positioned lower than the
reinfusion cannula to avoid recirculation. Figure 2: (right) Wire inserted
into the internal jugular vein tracking through the heart before entering the
inferior vena cava (IVC). Even a wire that initially passes directly into the
IVC can migrate into the heart during the cannulation procedure. Failure to
visualize the wire course throughout the procedure may lead to injury to
the heart during dual lumen, bi-caval cannulation.

does not require advanced imaging and is readily accomplished at the bed-
side. The choice between femoro-jugular and femoro-femoral configura-
tions is driven mostly by operator preference, although femoro-femoral
cannulation may be simpler under emergency circumstances when access
to the head of the bed is restricted. Both strategies require proper cannula
positioning to avoid recirculation (Figure 1).
Dual lumen cannulation allows V-V ECMO support through a single
cannula in the internal jugular vein. The primary advantage over dual site
cannulation is the ability to ambulate, though ambulation with femoral
cannulas in place is becoming more common and is considered routine in
some centers. Dual lumen cannulation is more time consuming and tech-
nically more difficult than two site cannulation. Realtime imaging with
fluoroscopy or transesophageal echo is strongly recommended to contin-

Chapter 15 251

uously monitor guidewire placement and avoid injury to the heart (Figure
2).18,19 While it may be considered emergently in patients with anatomical
features that preclude use of the femoral veins, dual lumen cannulation
should generally be reserved for elective scenarios.

Cannula Selection
Cannula size selection should not be modified to decrease the number of
dilations and shorten the procedure. The trivial gains achieved will likely
be offset by inadequate support or other complications such as hemo-
lysis. A reasonable selection for most adult patients is a 25 Fr drainage
cannula placed through the femoral vein and a 21 Fr reinfusion cannula
placed through the internal jugular or contralateral femoral vein. Patients
with normal oxygenation needing only CO2 clearance can be successfully
treated with smaller cannulas and lower flow; however, this should be re-
served for patients with isolated obstructive lung disease. Patients with
ARDS and predominantly CO2 retention may later develop severe hypox-
emia that cannot be adequately managed with a low flow circuit. Utilizing
a large venous cannula also provides adequate flow if the patient subse-
quently develops hemodynamic compromise and requires conversion to a
hybrid circuit that provides V-A and V-V ECMO support simultaneously.

Procedural Modifications
While the steps to safe ECMO cannulation are the same there are some
procedural modifications that can be considered in emergent cases. Pa-
tients undergoing crash V-V ECMO cannulation often have hemodynam-
ical compromise and are at risk of cardiac arrest during the procedure.
Transient hypotension is also common at the initiation of V-V ECMO
flow and may result in cardiac arrest in unstable patients. Accessing the
common femoral artery as the first step in the cannulation procedure will
expedite ECPR or rapid conversion to V-A ECMO if indicated. If arterial
cannulation is not required, the access can be converted to a femoral ar-
terial line or wire only access can be maintained during the procedure and
removed if it is not needed.
Utilizing a stiff guidewire increases the risk of vascular trauma but
may be justified in emergent cases, particularly when anatomical factors
such as large body habitus or scarring from prior surgeries are present.

252 ECPR and Resuscitative ECMO

 Using a minimal skin incision, or no incision at all, is appropriate
in elective ECMO cannulations. The authors recommend making a skin
incision just large enough to accommodate the cannula in emergent
procedures. Confirmation of wire and cannula location can be done via
fluoroscopy, transesophageal echocardiography, plain film x-ray, and
transthoracic ultrasound.

Physiology
The fundamental difference between V-V and V-A ECMO is the location
of the reinfusion cannula. Femoral artery reinfusion during peripheral V-A
ECMO provides respiratory and direct cardiovascular support but at the
price of increased left ventricular afterload, and the potential for arterial
embolic events, lower extremity ischemia, and upper body hypoxemia.
Reinfusing blood into the venous circulation limits the direct impact of
V-V ECMO to respiratory support but the rate of complications is lower
than with V-A ECMO.

Extracorporeal Gas Exchange During V-V ECMO

Three adjustable parameters determine gas transfer in the ECMO circuit
membrane lung (ML): oxygen concentration in the sweep gas, commonly
referred to as FDO2, blood flow, and sweep gas flow.20 In adult V-V ECMO
cases the FDO2 is generally left at 1.0, though some centers adjust it in
the weaning phase. Circuit blood flow, which is controlled indirectly by
adjusting pump RPM, is the primary determinant of oxygen delivery by
the ML.20,21 In general, the more oxygenated blood that flows through the
ML, the greater the oxygen delivery to the patient. The sweep gas flow rate
determines CO2 clearance in the ML and should be adjusted to achieve
target PaCO2. 21,22

V-V ECMO Impact on Hemodynamics

V-V ECMO has no direct impact on hemodynamics. The circuit is not
connected to the arterial circulation, blood removed from the venous cir-
culation by the drainage cannula is replaced by the same volume of ox-
ygenated blood returned through the reinfusion cannula. Therefore, VV

Chapter 15 253

ECMO neither directly offloads the right ventricle nor overloads it. That
is not to say the indirect hemodynamic impacts of V-V ECMO are insignif-
icant. Increased cardiac index and mean arterial pressure and decreased
pulmonary artery pressure occur rapidly after initiating V-V ECMO in
severe ARDS, likely attributable to improvements in PaCO2 and oxygen-
ation.20,23 V-V ECMO is the preferred configuration in vasopressor depen-
dent patients with a primary respiratory diagnosis and absence of severe
ventricular dysfunction.24-26 Ongoing reassessment and frequent echocar-
diography will identify patients who require conversion to V-A ECMO,
though this is uncommon.

Recirculation
Recirculation, a phenomenon unique to V-V ECMO, describes oxygen-
ated blood from the reinfusion cannula being immediately drawn into the
drainage cannula rather than entering the pulmonary and then systemic
circulation. Recirculated blood does not contribute to systemic oxygen
delivery and reduces the efficiency of ECMO support. Recirculation re-
sults from a complex interplay of cannula design, cannula position, blood
flow rate, cardiac output, and other patient factors.27 Recirculation is al-
ways present but often remains clinically insignificant. However, exten-
sive recirculation can cause inadequate support and must be addressed.
Quantifying recirculation at the bedside is difficult but it can be diagnosed
by decreased arterial oxygen saturation with increasing pre-membrane
lung saturation in the ECMO circuit. This is potentially recognized by
bright red blood in the venous drainage limb of the ECMO circuit. Adjust-
ing cannula position, decreasing pump flow, and assessing the patient for
decreased cardiac output are the mainstays of correcting recirculation.

Arterial Blood Oxygen Content

Arterial blood oxygen content during V-V ECMO is a function of the
ECMO ML and the native pulmonary circulation, which may have vary-
ing degrees of residual function. If the effective ECMO circuit flow (to-
tal flow minus recirculation fraction) equals the cardiac output, all blood
will be oxygenated in the ML before entering the right ventricle and the
systemic saturation will be normal even with no native lung function. Car-
diac output often exceeds ECMO flow leaving some amount of venous

254 ECPR and Resuscitative ECMO

blood to enter the pulmonary circulation without first passing through
the ML, thus relying entirely on residual lung function for oxygenation.
In the absence of gas exchange in the lung, the ratio of effective ECMO
flow to cardiac output (QECMO/QCO) is the primary determinant of arterial
oxygen content, with a value of 60% roughly correlating with systemic
SaO2 of 90%.20,28

Initial Management
ECMO Pump Flow
Users set the pump RPM and the resultant flow is determined by a host of
factors that influence resistance throughout the circuit including cannula
size and position, individual pump characteristics, and patient volume
status. In centrifugal pumps typical of modern ECMO systems the rela-
tionship between pump RPM and flow is nonlinear.29 Experienced pump
operators will be facile in navigating this situation.
For hypoxemic patients or those requiring full support due to upper
airway obstruction, blood flow should be increased until arterial oxygen
saturation normalizes, or the maximum sustainable flow is reached. Max-
imum sustainable flow has been exceeded when flows become erratic or
there is physical movement of the venous limb of the circuit, commonly
called “chatter” or “chugging,” which reflects collapse of the vena cava
around the drainage cannula. When venous pressures are measured,
highly negative values may also be indicative though absolute thresholds
are lacking. When these indicators are absent, the maximum achievable
flow can be estimated as the point where incremental increases in RPM
yield little or no increase in blood flow.30 Fluid administration may allow
additional flow in some cases but should be used judiciously given the
adverse effects of volume overload. Blood flow of 3.5–5 LPM is typical
in adults with severe ARDS, though flows over 6 LPM are achievable and
desirable in some cases.
Patients needing primarily CO2 clearance can be supported with lower
flows, often in the range of 2-3 LPM.

Chapter 15 255

ECMO Sweep Gas Flow
The oxygen concentration of the sweep gas, FDO2, should remain at 1.0
for all emergent V-V ECMO cases. Sweep gas flow rate, the primary de-
terminant of CO2 elimination, is generally set in a 2:1 ratio with blood
flow (i.e. 4 LPM blood flow:2 LPM gas flow) when ECMO is initiated. In
patients with compensated respiratory acidosis sweep gas flow should be
initiated at a lower rate to avoid rapid correction of CO2 with resultant
alkalosis which has been associated with adverse outcomes.31,32 Optimal
rate of CO2 correction remains unknown. In all cases a blood gas should
be obtained within 10 minutes of ECMO initiation to assess response to
therapy and inform sweep gas adjustments.

Ventilator Settings
During V-V ECMO initiated for ARDS the circuit should be the primary
source of gas exchange while ventilator settings are adjusted to avoid ven-
tilator induced lung injury, a strategy commonly referred to as “lung rest.”
Ideal lung rest ventilator settings are the subject of debate but generally
focus on management of several key parameters. Limiting plateau pressure
(PPLAT) to 30 cm H2O and tidal volume (VT) to 6 cc/kg of ideal body weight
(IBW), and further decreasing as low as 4 cc/kg as needed, is a widely ac-
cepted standard of care.33 When V-V ECMO is initiated, these upper limits
still apply though the extracorporeal gas exchange permits much lower
values which are believed to optimize lung protection. Positive end expi-
ratory pressure (PEEP) set too low predisposes to the harmful effects of
tidal recruitment/de-recruitment termed “atelectrauma,” while excessive
PEEP can overdistend alveoli and contribute to volutrauma. The effect of
PEEP on pulmonary vascular resistance is biphasic with adverse impacts
at either extreme with an optimal level in the mid-range. Unfortunately,
the specific PEEP value that best balances these factors differs for every
scenario and patient it is difficult to measure these parameters at the bed-
side. The best available evidence suggests outcomes are best with PEEP
levels of 10–15 cm H2O in patients with severe ARDS.34 Driving pressure
(∆P), the difference between PPLAT and PEEP, is emerging as the value
most associated with lung injury and should be managed aggressively.35,36
Limiting ∆P to 15 cm H2O is a reasonable target initially, although values
of 10 or less are often utilized.

256 ECPR and Resuscitative ECMO

 In cases of NFA, ECMO can provide near complete gas exchange al-
lowing the ventilator strategy to focus exclusively on eliminating dynamic
hyperinflation. VT should initially be 6–8 cc/kg IBW and RR at 6–8 breaths
per minute. Monitoring PPLAT will provide the best insight into dynamic
hyperinflation and both VT and RR can be further reduced as needed.
If there is compromise to ECMO flow, then ventilator should be re-
verted to pre-ECMO settings (tidal volume, PEEP, FiO2, respiratory rate)
or bag ventilation until circuit flow can be restored.

Anticoagulation
A heparin bolus is usually given during ECMO cannulation and continuous
anticoagulation should generally be started within 2-4 hours. However,
withholding anticoagulation in patients at high risk of bleeding is well de-
scribed and becoming more common.37,38 Heparin is the most widely used
anticoagulant though bivalirudin and argatroban can also be used. In the
absence of a specific institutional protocol, targeting an aPTT of 45-65
seconds is appropriate. Detailed anticoagulation monitoring strategies
and specific targets vary widely and are beyond the scope of this chapter.

Conclusion
While V-A ECMO is the most common configuration in crashing patients,
V-V ECMO is the preferred modality when an underlying respiratory or
airway condition is the primary pathophysiologic process. ARDS is the
most common indication for crash V-V ECMO with NFA, airway obstruc-
tion, and anaphylaxis accounting for the remainder of cases. Dual site
cannulation with a 25 Fr drainage and 21 Fr reinfusion cannulas is the
preferred modality. Minor modifications to the cannulation procedure
are appropriate but key steps should not be eliminated in the interest of
shortening the procedure. Once ECMO is established, rapid blood gases
will improve and is usually accompanied by hemodynamic improvement.
Pump blood flow is the primary determinant of oxygen delivery and the
sweep gas should be adjusted to achieve the desired PaCO2. Rest ventila-
tor settings are established to avoid ventilator induced lung injury and fa-
cilitate lung healing. Anticoagulation with heparin or an alternative agent

Chapter 15 257

is typically employed during ECMO but can be eliminated in the setting
of ongoing bleeding.

References
1. Extracorporeal Life Support Organization. ECLS Registry Report
- International Summary https://www.elso.org/Registry/Statistics/In-
ternationalSummary.aspx. Published 2020. Accessed May 14, 2020.
2. Combes A, Hajage D, Capellier G, et al. Extracorporeal Membrane
Oxygenation for Severe Acute Respiratory Distress Syndrome. NEJM
2018;378(21):1965-1975.
3. Brodie D, Slutsky AS, Combes A. Extracorporeal Life Support for
Adults With Respiratory Failure and Related Indications: A Review.
JAMA. 2019;322(6):557-568.
4. Kim KI, Lee WY, Kim HS, Jeong JH, Ko HH. Extracorporeal mem-
brane oxygenation in near-drowning patients with cardiac or pul-
monary failure. Scandinavian journal of trauma, resuscitation and
emergency medicine. 2014;22:77.
5. Burke CR, Chan T, Brogan TV, et al. Extracorporeal life support for
victims of drowning. Resuscitation. 2016;104:19-23.
6. Krishnan V, Diette GB, Rand CS, et al. Mortality in patients hospital-
ized for asthma exacerbations in the United States. Am J Respir Crit
Care Med. 2006;174(6):633-638.
7. Kaur BP, Lahewala S, Arora S, et al. Asthma: Hospitalization Trends
and Predictors of In-Hospital Mortality and Hospitalization Costs in
the USA (2001-2010). Int Arch Allergy Immunol. 2015;168(2):71-78.
8. Yeo HJ, Kim D, Jeon D, Kim YS, Rycus P, Cho WH. Extracorporeal
membrane oxygenation for life-threatening asthma refractory to
mechanical ventilation: analysis of the Extracorporeal Life Support
Organization registry. Crit Care. 2017;21(1):297.
9. Malpas G, Hung O, Gilchrist A, et al. The use of extracorporeal
membrane oxygenation in the anticipated difficult airway: a case
report and systematic review. Can J Anaesth. 2018;65(6):685-697.
10. Yunoki K, Miyawaki I, Yamazaki K, Mima H. Extracorporeal
Membrane Oxygenation-Assisted Airway Management for Dif-
ficult Airways. Journal of cardiothoracic and vascular anesthesia.
2018;32(6):2721-2725.

258 ECPR and Resuscitative ECMO

11. Hong Y, Jo KW, Lyu J, et al. Use of venovenous extracorporeal
membrane oxygenation in central airway obstruction to facilitate
interventions leading to definitive airway security. J Crit Care.
2013;28(5):669-674.
12. Holliday T, Jackson A. Emergency use of extracorporeal membrane
oxygenation for a foreign body obstructing the airway. Critical care
and resuscitation : journal of the Australasian Academy of Critical Care
Medicine. 2010;12(4):273-275.
13. Zhang ZP, Su X, Liu CW. Cardiac arrest with anaphylactic shock: a
successful resuscitation using extracorporeal membrane oxygen-
ation. Am J Emerg Med. 2015;33(1):130 e133-134.
14. Weiss GM, Fandrick AD, Sidebotham D. Successful rescue of an
adult with refractory anaphylactic shock and abdominal compart-
ment syndrome with venoarterial extracorporeal membrane oxygen-
ation and bedside laparotomy. Seminars in cardiothoracic and vascular
anesthesia. 2015;19(1):66-70.
15. Wang ML, Chang CT, Huang HH, Yeh YC, Lee TS, Hung KY.
Chlorhexidine-related refractory anaphylactic shock: a case success-
fully resuscitated with extracorporeal membrane oxygenation. J Clin
Anesth. 2016;34:654-657.
16. Carelli M, Seco M, Forrest P, Wilson MK, Vallely MP, Ramponi F.
Extracorporeal membrane oxygenation support in refractory periop-
erative anaphylactic shock to rocuronium: a report of two cases.
Perfusion. 2019;34(8):717-720.
17. Shinar Z, Bellezzo J, Weingart S. EDECMO Podcast #31
– Anaphylaxis & Epi-Pens. Are we ready for VV-ECMO
in the Emergency Department? [Internet]. https://
edecmo.org/edecmo-31-anaphylaxis-epi-pens-read
y-vv-ecmo-emergency-department/: EDECMO; 2016 10/11/2016
[cited 5/19/2020].
18. Teman NR, Haft JW, Napolitano LM. Optimal endovascular methods
for placement of bicaval dual-lumen cannulae for venovenous extra-
corporeal membrane oxygenation. ASAIO J. 2013;59(4):442-447.
19. Ngai CW, Ng PY, Sin WC. Bicaval dual lumen cannula in adult
veno-venous extracorporeal membrane oxygenation-clinical pearls
for safe cannulation. J Thorac Dis. 2018;10(Suppl 5):S624-S628.

Chapter 15 259

20. Schmidt M, Tachon G, Devilliers C, et al. Blood oxygenation and
decarboxylation determinants during venovenous ECMO for respira-
tory failure in adults. Intensive Care Med. 2013;39(5):838-846.
21. Lehle K, Philipp A, Hiller KA, et al. Efficiency of gas transfer in
venovenous extracorporeal membrane oxygenation: analysis of
317 cases with four different ECMO systems. Intensive Care Med.
2014;40(12):1870-1877.
22. Sun L, Kaesler A, Fernando P, Thompson AJ, Toomasian JM,
Bartlett RH. CO2 clearance by membrane lungs. Perfusion.
2018;33(4):249-253.
23. Reis Miranda D, van Thiel R, Brodie D, Bakker J. Right ventricular
unloading after initiation of venovenous extracorporeal membrane
oxygenation. Am J Respir Crit Care Med. 2015;191(3):346-348.
24. Lazzeri C, Bonizzoli M, Cianchi G, et al. Right ventricular
dysfunction and Pre Implantation Vasopressors in Refrac-
tory ARDS Supported by VV-ECMO. Heart, lung & circulation.
2018;27(12):1483-1488.
25. Gutsche JT, Mikkelsen ME, McCarthy FH, et al. Veno-Venous Extra-
corporeal Life Support in Hemodynamically Unstable Patients With
ARDS. Anesthesia and analgesia. 2017;124(3):846-848.
26. Kon ZN, Bittle GJ, Pasrija C, et al. Venovenous Versus Venoarterial
Extracorporeal Membrane Oxygenation for Adult Patients With
Acute Respiratory Distress Syndrome Requiring Precannulation
Hemodynamic Support: A Review of the ELSO Registry. The Annals
of thoracic surgery. 2017;104(2):645-649.
27. Abrams D, Bacchetta M, Brodie D. Recirculation in venovenous ex-
tracorporeal membrane oxygenation. ASAIO J. 2015;61(2):115-121.
28. Messai E, Bouguerra A, Harmelin G, Di Lascio G, Cianchi G, Bonac-
chi M. A new formula for determining arterial oxygen saturation
during venovenous extracorporeal oxygenation. Intensive Care Med.
2013;39(2):327-334.
29. Park M, Mendes PV, Hirota AS, dos Santos EV, Costa EL, Azevedo
LC. Blood flow/pump rotation ratio as an artificial lung performance
monitoring tool during extracorporeal respiratory support using
centrifugal pumps. Rev Bras Ter Intensiva. 2015;27(2):178-184.
30. Zakhary B, Vercaemst L, Mason P, Lorusso R, Brodie D. How I man-
age drainage insufficiency on extracorporeal membrane oxygenation.
Crit Care. 2020;24(1):151.

260 ECPR and Resuscitative ECMO

31. Cavayas YA, Munshi L, Del Sorbo L, Fan E. The Early Change in
PaCO2 After Extracorporeal Membrane Oxygenation Initiation is
Associated with Neurological Complications. Am J Respir Crit Care
Med. 2020.
32. Bembea MM, Lee R, Masten D, et al. Magnitude of arterial carbon di-
oxide change at initiation of extracorporeal membrane oxygenation
support is associated with survival. The Journal of extra-corporeal
technology. 2013;45(1):26-32.
33. ARDSNet. Ventilation with lower tidal volumes as compared
with traditional tidal volumes for acute lung injury and the
acute respiratory distress syndrome. The Acute Respiratory Dis-
tress Syndrome Network. The New England journal of medicine.
2000;342(18):1301-1308.
34. Schmidt M, Stewart C, Bailey M, et al. Mechanical ventilation man-
agement during extracorporeal membrane oxygenation for acute
respiratory distress syndrome: a retrospective international multi-
center study. Crit Care Med. 2015;43(3):654-664.
35. Amato MB, Meade MO, Slutsky AS, et al. Driving Pressure and Sur-
vival in the Acute Respiratory Distress Syndrome. The New England
journal of medicine. 2015;372(8):747-755.
36. Serpa Neto A, Schmidt M, Azevedo LC, et al. Associations between
ventilator settings during extracorporeal membrane oxygenation
for refractory hypoxemia and outcome in patients with acute
respiratory distress syndrome: a pooled individual patient data
analysis : Mechanical ventilation during ECMO. Intensive Care Med.
2016;42(11):1672-1684.
37. Kurihara C, Walter JM, Karim A, Thakkar S, Saine M, Odell DD,
Kim S, Tomic R, Wunderink RG, Budinger GRS, Bharat A. Fea-
sibility of Venovenous Extracorporeal Membrane Oxygenation
Without Systemic Anticoagulation. Ann Thorac Surg. 2020
Oct;110(4):1209-1215.
38. Muellenbach RM, Kredel M, Kunze E, et al. Prolonged heparin-free
extracorporeal membrane oxygenation in multiple injured acute
respiratory distress syndrome patients with traumatic brain injury. J
Trauma Acute Care Surg. 2012;72(5):1444-1447.

Chapter 15 261

INDEX

Access Cannulas – 89, 96 Chest Compressions – 23-25, 27

Acute Coronary Syndrome – 50, 57 Chronic Obstructive Pulmonary Disease
Acute Kidney Injury – 141 – 59
Acute Respiratory Distress Syndrome Code Choreography – 75, 78
– 248 Complications – 91, 121, 169-174, 202
Advanced Cardiac Life Support – 5, 12, Computed Tomography – 167-168
73-79 Coronary Perfusion Pressure (CPP) –
Age – 49-50, 57 30
Air Embolism – 153, 162, 187 CPT codes – 17
Airway Management – 28, 31 Cutdown – 115-120
Airway Obstruction – 249 Decannulation (Accidental) – 188
Anaphylaxis – 250 Defibrillation – 29, 37-39
Anticoagulation – 82, 103, 143, 172 Delta P – 126, 148, 162
Antidysrhythmics – 29 Diabetes – 59
Aortic Dissection – 108 Diagnosis Related Group (DRG) – 4
Aortic insufficiency – 108-110 Differential Hypoxemia – 126, 139-140
ARREST Trial – 48 Dilators – 89, 90, 94
Arterial cannula (return) – 89, 91, 93, Distal Perfusion Cannulas – 89, 119
96, 97 Drainage (Venous) Cannulas – 96, 133,
Arterial Line – 81, 105, 140, 155, 161 247
Asthma – 248 Drowning – 59
Asystole – 56 Drug Overdose – 58
Billing – 4 Echocardiography – 103, 137, 165
Bleeding – 144, 172 Emergency Medical Services (EMS) –
Bubbles – 119, 151,153 8,52-53, 78
Bystander CPR – 25-26, 50 Emergency Preservation and
Cannulation – 90-104 Resuscitation – 242
Central – 95 Epinephrine – 29,34
Percutaneous – 101 End-tidal CO2 – 53,127, 128
Surgical Cutdown – 115-120 Equipment – 13, 77, 82,85, 116
Venovenous – 250 Extraction Ratio – 127-128
Cannulator – 77-78, 101 Flow (Blood) – 31, 154, 162, 164
Cardiac arrest – 47 Fluoroscopy – 94,100
Cardiac Output – 127-129 Guidewires – 89, 98,161
Cardiogenic Shock – 221 Hemoglobin – 128,140
Carts (Supply) – 15, 82, 84, 194-196 Hemorrhagic Shock – 238
Central Venous Oxygen Saturation Hyperinvasive Trial – 49
(ScvO2) – 162 Hypothermia (exposure) – 59, 235

 263
Hypothermia (Therapeutic) – 105, 159, Pulmonary Embolism – 10,57, 223
175 Pulsatility – 137, 142
Impella – 138, 140, 171, 226-227 Pulseless Electrical Activity (PEA) – 56
In-Hospital Cardiac Arrest – 10, 53, 56 Quality adjusted life years (QALY) – 4
Inclusion Criteria – 49, 52, 54, 63 Rhythm (Initial) – 55
Inguinal Hernia – 111 Recirculation – 247,251, 254
Initiation Checklist – 148, 152 Respiratory Therapists – 9, 75
Intra-Aortic Balloon Pump – 171, Return Cannula – 89, 91, 93, 96, 97
226,227 Return of Spontaneous Circulation
Intracranial Hemorrhage – 239 (ROSC) – 25, 47, 60
Lactate – 60 Revolutions Per Minute (RPM) – 136,
Left Ventricular Distension – 137, 148, 151-152, 162, 183, 185
163,171 SAVE-J Trial – 48
Limb ischemia – 94, 105, 174, 228 Securement – 156
Low Flow period – 23-25, 53 Selection criteria – 49, 52, 54, 63
Mean Arterial Pressure (MAP) – 131, Signs of Life – 54, 61, 62
136, 142, 163 Situs Inversus – 110-111,
Mechanical chest compression devices – Sterile Field – 116,147,154
27, 74, 79 Stroke Volume – 129
Membrane Lung – 126, 133, 136 SvO2 – 142, 162
Myocarditis – 58, 230 Sweep Gas – 126, 135, 151-152, 185,
Near-infrared spectroscopy (NIRS) – 249, 256
140, 143 Systemic Vascular Resistance – 131
No Flow period – 23,26, 51 Targeted Temperature Management – 175
North-South Syndrome – 126, 139-140 Temperature (body) – 152, 236, 242
Out-of-Hospital Cardiac Arrest – 56 Transfusion – 172
Oxygen Consumption (DO2) – 125, Ultrasound – 28,31, 76, 94, 103
127-129, 142-143 Utstein criteria – 62
Oxygen Delivery (VO2) – 125, 127-128 Vascular Complications – 91
Oxygen Saturation – 127, 140 Vasopressors – 18, 34, 91, 163
Positive end expiratory pressure – 164, Venous Drainage Cannulas – 96, 133,
171, 256 247
PO2 – 62 Ventricular fibrillation – 37-38, 55
Poiseuille’s Law – 95 Viscous Cycle of Trauma – 235
pH – 60 Von Willebrand Factor – 141
Physician Code Team Leader – 74, 76,78 Wet-to-Wet Connection – 147, 149
Pulmonary (O)Edema – 128, 171 Witnessed arrest – 51

264 ECPR and Resuscitative ECMO