(Respiratory Medicine) Gregory A. Schmidt - Extracorporeal Membrane Oxygenation For Adults-Humana (2022)
(Respiratory Medicine) Gregory A. Schmidt - Extracorporeal Membrane Oxygenation For Adults-Humana (2022)
Extracorporeal
Membrane
Oxygenation
for Adults
Second Edition
Respiratory Medicine
Series Editors
Sharon I. S. Rounds, Brown University
Providence, RI, USA
Anne Dixon, University of Vermont, Larner College of Medicine
Burlington, VT, USA
Lynn M. Schnapp, University of Wisconsin - Madison
Madison, WI, USA
Respiratory Medicine offers clinical and research-oriented resources for
pulmonologists and other practitioners and researchers interested in respiratory
care. Spanning a broad range of clinical and research issues in respiratory medicine,
the series covers such topics as COPD, asthma and allergy, pulmonary problems in
pregnancy, molecular basis of lung disease, sleep disordered breathing, and others.
The series editors are Sharon Rounds, MD, Professor of Medicine and of
Pathology and Laboratory Medicine at the Alpert Medical School at Brown
University, Anne Dixon, MD, Professor of Medicine and Director of the Division of
Pulmonary and Critical Care at Robert Larner, MD College of Medicine at the
University of Vermont, and Lynn M. Schnapp, MD, George R. And Elaine Love
Professor and Chair of Medicine at the University of Wisconsin-Madison School of
Medicine and Public Health.
Gregory A. Schmidt
Editor
Extracorporeal Membrane
Oxygenation for Adults
Second Edition
Editor
Gregory A. Schmidt
Division of Pulmonary Diseases, Critical Care, and Occupational Medicine
Department of Internal Medicine, University of Iowa
Iowa City, IA, USA
© The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature
Switzerland AG 2016, 2022
This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether
the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of
illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and
transmission or information storage and retrieval, electronic adaptation, computer software, or by similar
or dissimilar methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication
does not imply, even in the absence of a specific statement, that such names are exempt from the relevant
protective laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in this book
are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the
editors give a warranty, expressed or implied, with respect to the material contained herein or for any
errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional
claims in published maps and institutional affiliations.
This Humana imprint is published by the registered company Springer Nature Switzerland AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
To the Core ECMO Specialists at the
University of Iowa. Through an incredibly
challenging 2 years, you have continued to
lead, guide, educate, negotiate, muster
resources, and otherwise advocate for
excellent ECMO care. Our patients are
fortunate to have you. I’m grateful to work
alongside you.
Preface to the 2nd Edition
vii
viii Preface to the 2nd Edition
This book has been written for practicing physicians, nurses, perfusion special-
ists, therapists, and critical care trainees who are considering whether to refer their
patients for ECMO, debating whether to offer ECMO capability to their patients, or
are already providing ECMO but seek a practical reference to best practices and
updated information. It could never have been completed without the inspiration
from my colleagues at Iowa who strive daily to save the sickest patients; the trainees
whose curiosity makes us all want to know more; my contributors who are at the
forefront of a truly challenging field; and our publisher at Springer-Link who pushed
for this important revision. Finally, I recognize all those who do the hard work: the
ECMO specialists, bedside nurses, and therapists who dedicate their lives to the
critically ill. This is an exciting time, ripe with change and new opportunities.
1
Physiology of Extracorporeal Life Support (ECLS)���������������������������� 1
Matthew J. Brain, Warwick W. Butt, and Graeme MacLaren
2
Circuits, Membranes, and Pumps���������������������������������������������������������� 63
Bradley H. Rosen
3 Modes of ECMO�������������������������������������������������������������������������������������� 81
Jonathan Eaton, Christopher Trosclair, and L. Keith Scott
4 Vascular Access���������������������������������������������������������������������������������������� 97
Steven A. Conrad
5
Hypoxemic Respiratory Failure: Evidence,
Indications, and Exclusions�������������������������������������������������������������������� 115
Kathleen E. Melville, Cara Agerstrand, Daniel Brodie,
and Darryl Abrams
6
Ventilator Management During ECLS�������������������������������������������������� 125
Antonio Pesenti, Giacomo Bellani, Giacomo Grasselli,
and Tommaso Mauri
7
Managing the Systemic Circulation: Volume
Status and RV Function�������������������������������������������������������������������������� 147
Sundar Krishnan and Gregory A. Schmidt
8 Antithrombotic Therapy for ECMO������������������������������������������������������ 159
Usha S. Perepu
9 Membrane Dysfunction �������������������������������������������������������������������������� 173
B. D. Warren, M. J. Sobieszczyk, and P. E. Mason
10 ECCO2R in Obstructive Diseases: Evidence,
Indications, and Exclusions�������������������������������������������������������������������� 187
Lorenzo Del Sorbo, V. Marco Ranieri, and Vito Fanelli
ix
x Contents
11
ECMO as a Bridge to Lung Transplantation���������������������������������������� 205
Christian Kuehn and Ruslan Natanov
12
Daily Management of Patients on VV ECMO�������������������������������������� 217
Charles Rappaport and Kristina Rappaport
13 Crises During ECLS�������������������������������������������������������������������������������� 229
Purnema Madahar, Dana A. Mullin, Meaghan Flatley,
Darryl Abrams, Phillipe H. Lemaitre, Daniel Brodie,
and Cara Agerstrand
14 Mobilization During ECLS �������������������������������������������������������������������� 253
Gregory A. Schmidt
15 ECMO Weaning and Decannulation������������������������������������������������������ 265
Sharon L. McCartney and Sundar Krishnan
16
Venoarterial ECMO in Respiratory Failure������������������������������������������ 277
Avery Tung and Tae H. Song
17
Ethical Challenges in Extracorporeal
Membrane Oxygenation�������������������������������������������������������������������������� 293
Elizabeth Sonntag and Meera Pahuja
Index�������������������������������������������������������������������������������������������������������������������� 303
Contributors
xi
xii Contributors
Introduction
M. J. Brain (*)
Department of Medicine, Launceston General Hospital, Launceston, TAS, Australia
e-mail: [email protected]
W. W. Butt
Intensive Care, Royal Childrens Hospital, University of Melbourne Department of
Paediatrics, Melbourne, VIC, Australia
e-mail: [email protected]
G. MacLaren
Cardiothoracic ICU, National University Hospital, Singapore, Singapore
VA-ECMO
VV-ECMO
VPA-ECMO LVAD
Fig. 1.1 Schematic of ECMO configurations, circles represent pumps and diamonds represent
oxygenators. VA-ECMO: veno-arterial extracorporeal membrane oxygenation demonstrating
cavo-aortic flow. VV-ECMO: veno-venous cannulation demonstrating cavo-atrial flow from the
inferior vena cava to the right atrium via the oxygenator and pump. VV-ECMO may also require a
second cannula taking blood from the superior vena cava or dual lumen cannulas that access blood
from the inferior and superior vena cava, while returning blood to the right atrium. VPA-ECMO:
veno-pulmonary artery cannulation may be configured as atrial to pulmonary artery flow with or
without oxygenation support. LVAD: left ventricular assist device (usually implanted) taking left
ventricular blood and returning it to the proximal aorta. RVAD: a right ventricular assist device is
not shown but may be implanted or external and can be configured identically to VPA-ECMO
without an oxygenator, or may directly drain the right ventricle as per the LVAD. Extracorporeal
carbon dioxide removal (ECCO2-R) is commonly performed with a VV-ECMO configuration, usu-
ally with a single dual lumen catheter. Intravascular membrane oxygenators have also been devel-
oped [1] but are not currently in clinical use
respectively. This chapter will focus on VV-ECMO and only cover other configura-
tions when relevant.
A basic ECMO circuit consists of a blood pump and oxygenator connected by
conduits (Fig. 1.1). Other components may be added to this basic configuration,
most commonly renal replacement therapy connected in parallel via side ports.
Minimising the number of connections and maintaining simplicity is important for
safety, infection control and troubleshooting.
Each configuration creates a unique interaction with the cardiorespiratory sys-
tem. Sound understanding of the physiology and limitations of each mode is
required to prescribe, manage and wean this support and recognise evolving com-
plications of the therapy. Although designed primarily to replace cardiorespiratory
function, the interaction of ECLS with several other physiologic systems must be
1 Physiology of Extracorporeal Life Support (ECLS) 3
considered. For example, most patients who require ECLS will have sustained a
major insult such as severe sepsis, trauma or surgery, or have suffered from progres-
sive cardiac or pulmonary disease. A systemic inflammatory response syndrome
(SIRS) and coagulopathies may arise from the underlying pathology, as a reaction
to the non-biological material of the ECLS circuit, or from the secondary develop-
ment of sepsis due to the invasive nature and immunosuppressive consequences of
treatment. The varying metabolic responses to critical illness and chronic pathology
have direct implications for oxygenation and CO2 production, as well as nutritional
supplementation to facilitate later weaning.
At this time, ELCS is a bridge to either recovery of a failing cardiorespiratory
system or a destination therapy such as an LVAD or organ transplantation. In rapidly
progressing pathologies ECLS is sometimes a bridge to decision, buying time to
assess the disease prognosis and implications of any destination therapies [2].
In order to comprehensively understand ECLS and its effects on human physiol-
ogy, it is necessary to first review cellular metabolism and oxygen transport.
Cellular Metabolism
Glucose (and other simple carbohydrates) enter cells down a concentration gradient
through glucose transporters that allow tissue-specific behaviour such as preferen-
tial basal uptake by the brain, concentration-dependent uptake by the liver,
concentration-
sensing by the insulin-secreting pancreatic ß-cells and insulin-
dependent uptake in skeletal muscle and fat [3].
4 M. J. Brain et al.
The respiratory quotient (RQ) describes the ratio of the amount of carbon dioxide
(VCO2 ) produced per unit time to the amount of oxygen consumed (VO2 ).
VCO2
RQ = (1.1)
VO2
The respiratory quotient is dependent on the sources of fuel being used. For glu-
cose metabolism, the six-carbon atoms result in production of six molecules of car-
bon dioxide while consuming six molecules of oxygen; it thus has a respiratory
quotient of one. The oxidation of fatty acids (lipolysis) and some amino acids pro-
duce less CO2 per O2 (by not producing NADH in the pyruvate to acetyl-CoA reac-
tion, Fig. 1.2) and hence have respiratory quotients of less than one.
1 Physiology of Extracorporeal Life Support (ECLS) 5
2-
CH2 OPO3
OH
OH OH
OH (+) Glucagon
ATP ADP Glucose-6-P
Glucose Glycogen
(6 Carbons)
ATP (+) Insulin
ADP (-) Epinephrine
Fructose-6-P
Fructose-1,6-bis-P
2-
O3 PO
HO
Dihydroxyacetone Glyceraldehyde-3-P
Phosphate
(3 Carbons) 2-
(3 Carbons) O OPO3
NAD+
NADH 2,3-Diphosphoglycerate
1,3-Bisphosphoglycerate Rapoport-Leubering
Glycerol ADP Shunt(Erythrocytes)
ATP
Triglycerides 3-Phosphoglycerate
Phosphoenolpyruvate
Fatty Acid Synthesis Amino Acids
(+) Insulin ADP (+) Glucagon
(-) Acetyl-CoA ATP (+) Acetyl-CoA
After transfer to Liver
Acetyl-CoA Pyruvate
Ketone Bodies Lactate
(2 Carbons) NADH NAD+ (3 Carbons)
+ CO O OH O OH
2 NADH NAD+
H3C O H3C OH
(4 Carbons) Oxaloacetate Citrate (6 Carbons) 3-5 ADP
O2 2 NAD+
Tricarboxylic Acid
3NAD+ + FAD Oxidative Phosphorylation
Cycle
3NADH + FADH + 2CO HO 2 NADH
2 2 2
3-5 ATP
Fig. 1.2 Key intermediates in intracellular metabolism: After entering cells, glucose is phosphory-
lated (-P) and can then be incorporated into glycogen, enter synthetic reactions (not shown), or be
metabolised to two three-carbon pyruvate molecules (glycolysis). The conversion of pyruvate to
acetyl-CoA, the subsequent tricarboxylic acid (TCA) cycle and oxidative phosphorylation only
occur in mitochondria and depend on oxygen to restore nicotinamide adenine dinucleotide to its
oxidised form (NAD+) for continued cycling. The number of ATP generated depends on the source
of reduction power; a single mitochondrial NADH produces 2.5 ATP; however, electrons from
cytosolic NADH must be transferred to mitochondrial FADH2 which yields only 1.5 ATP each [4].
Amino acids can enter or be synthesised at several points. Acetyl-CoA is a key junction molecule
providing the TCA cycle with two carbon acetyl groups, not only from glycolysis but also from
fatty acids and some amino acids. In glucose excess, acetyl-CoA is the starting point for fatty acid
synthesis and, in the starvation state, of ketone body production when insufficient oxaloacetate
exists for acetyl groups to enter the TCA cycle. Production of ketone bodies occurs predominantly
in the liver from fatty acid breakdown and constitutes a glucose-sparing fuel for the brain and
heart. Humoral promoters and inhibitors of reactions are shown
6 M. J. Brain et al.
In contrast to oxidation, each acetyl-CoA molecule utilised for fatty acid synthe-
sis (lipogenesis) results in production of a molecule of CO2 (from pyruvate to
acetyl-CoA, Fig. 1.2) without increasing mitochondrial NADH. As the rate of oxy-
gen consumption depends on the mitochondrial concentration of NADH, lipogene-
sis results in CO2 production which exceeds oxygen consumption. Some oxygen
consumption still occurs as the synthetic reaction also consumes ATP; however, the
RQ will be greater than 1. Examples of respiratory quotients based on theoretical
stoichiometry include [5, 6]:
C6 H12 O6 + 6O 2
Glucose Oxidation : RQ = 1
→ 6CO 2 + H 2 O
C57 H110 O6 + 80.25 O 2
Lipolysis of glycerol triestearate : RQ = 0.667
→ 57 CO 2 + 55.5 H 2 O
NC2 H 5 O 2 + 2.25 O 2
Amino Acid ( Glycine ) Oxidation : RQ = 0.88
→ 2 CO 2 + 2.5 H 2 O
4 C6 H12 O 6 + O 2
Lipogenesis from Glucose∗ : RQ = 8
→ C16 H 32 O 2 + 8 CO 2 + 8 H 2 O
13.83 C6 H12 O6 + 5 O 2
Lipogenesis∗ : RQ = 5.6
→ C55 H104 O6 + 28 CO 2 + 31 H 2 O
Note that the RQ of lipogenesis is dependent on the fatty acid being produced
*
Key hormones coordinate the response of energy metabolism to stress. Insulin indi-
cates the fed state promoting hepatic glucose uptake, glycogen and amino acid syn-
thesis and diversion of acetyl-CoA to free-fatty acid production while in the
peripheral tissues stimulating myocyte synthesis of contractile elements and adipo-
cytes to triglyceride deposition.
1 Physiology of Extracorporeal Life Support (ECLS) 7
Erythrocyte Metabolism
Being a specialised organ for oxygen transport, erythrocytes are nearly 90% haemo-
globin by weight, with very few other organelles. Nevertheless, they require an
ongoing energy source for maintenance of membrane integrity, cytoskeleton struc-
ture, intracellular electrolyte and osmotic equilibrium and to maintain the ferric
moieties of haemoglobin in a reduced state (Fe2+).
Erythrocytes lack mitochondria and do not store glycogen and thus depend on
anaerobic glycolysis of plasma glucose to lactate for ATP production. Glycolysis in
erythrocytes is also utilised for reactions that do not produce ATP, such as reducing
power to correct oxidised haemoglobin (methaemoglobin carrying a Fe3+ iron atom
that cannot carry oxygen), glutathione production (protecting the cell membrane
against oxidative damage) and the production of 2,3-diphosphoglycerate (2,3-DPG)
that modulates the affinity of haemoglobin for oxygen [9].
8 M. J. Brain et al.
δ ε
O2 τ CO2
C
ε
Blood
Flow
Heated Water
Fig. 1.3 Schematic detail of hollow fibre oxygenator construction demonstrating extra-capillary
flow of blood around the gas-carrying hollow fibres. Cross current flow exists between gas and
blood. Heated water tubules are also demonstrated. The diffusion path for gas exchange is shown
(top-right) consisting of the porous membrane and the boundary layer of adsorbed proteins.
Parameters of effective diffusivity from Eq. 1.8 are demonstrated with ε being the porosity – the
area of membrane occupied by gas, τ the tortuosity an index of effective path length for gas to
traverse the membrane (a path length is shown but in reality will be unknown), and δ the constric-
tivity – the resistance to gas passage [11]
microporous membranes theoretically allow contact between plasma and the sweep
gas; however, more recent materials such as poly-4-methyl-1-pentene utilise closed
fibres and are thus considered true membranes [12]. Many systems utilise the lumen
of the hollow fibres for fresh gas flow, with blood flowing between fibres; this is
termed extra-capillary flow. Overall characteristics such as total surface area for gas
exchange, resistance to flow and trauma to cellular blood components are deter-
mined by factors such as membrane material, fibre diameter and length, fibre den-
sity and the velocity of the blood [11].
Heat loss over the extracorporeal circuit from blood to the environment can be
substantial, and heat exchangers are commonly incorporated into oxygenator
10 M. J. Brain et al.
design. Figure 1.3 demonstrates one such design where the microporous membrane
fibres are laid perpendicularly to impermeable capillaries that circulate heated
water, allowing a large surface area for heat transfer to blood.
Unlike most solutes dissolved in body fluids that are quantified in moles, gas con-
centrations are reported in units of pressure. The universal gas law describes the
relationship between the partial pressure of an ideal gas and its container, with ideal
gas molecules best summarised as having minimal mass and absent intermolecular
attraction:
P = nRT (1.2)
V
K Forward
O 2 ( gas ) O 2 ( dissolved ) (1.3)
K Reverse
The rate constant KForward in Eq. 1.4a describes the proportion of oxygen gas that
dissolves per unit time, while KReverse describes the proportion of dissolved oxygen
that leaves the solution to the gas phase. When the system described in Eq. 1.4a is
at thermodynamic equilibrium, the concentrations in the gas and liquid phases are
stable, and the constants may then be combined, resulting in Eq. 1.4b, also known
as Henry’s law. SC is the Bunsen solubility coefficient resulting from
SC = KForward/KReverse and is gas and solvent specific. SC is affected by other dissolved
1 Physiology of Extracorporeal Life Support (ECLS) 11
solutes and falls with increasing temperatures (i.e. KForward becomes smaller and
KReverse larger).
Due to difficulties in measuring the molar concentration of oxygen compared to
the ease of measuring a volume of 100% oxygen at standard conditions (STPD:
0 °C, 760 mmHg, dry gas), it is customary to report oxygen content in ml/dL. Under
these conditions, oxygen approximates an ideal gas such that 6.02 × 1023 gas mol-
ecules (i.e. one mole) occupies 22.414 litres at 0 °C. Quantification of human oxy-
gen consumption is performed using STPD rather than BTPS (body temperature
and pressure, saturated: Eq. 1.6) [14] as water vapour in the latter partially con-
denses with increasing pressure. This results in a significant deviation from an ideal
gas and invalidates the relationship between the number of molecules and volume
defined in Eq. 1.2.
It should be noted that this does not include oxygen combined with haemoglobin
and CO2 in reaction with water as bicarbonate. The significantly higher plasma con-
centration of dissolved carbon dioxide (Eq. 1.5) resulting from its greater solubility
allows for more rapid elimination by gas exchange membranes when compared to
oxygen under the same flow conditions.
While in the gaseous phase, summation of each individual gases partial pressure
will equal the total ambient pressure that the gas mixture exerts on its container,
allowing each individual gas to be reported as a fraction of the total. For example,
the partial pressure of oxygen in inhaled 37 °C air that is fully saturated with water
vapour at 1 atmosphere (i.e. BTPS) is:
This summative requirement is only met when the solution is exposed to a gas
phase. As solubility coefficients vary between gases, the number of moles of
12 M. J. Brain et al.
dissolved gas in a given quantity of solution that is in contact with a gas phase has
no such equivalent summation.
As at equilibrium, the partial pressure of a gas is proportional to the concentra-
tion in solution, it may be used as a substitute for concentration even when no gas
phase is present, as is the case for body fluids. In this case it represents the partial
pressure that would be required of a gas phase to maintain the existing concentra-
tion of dissolved molecules in solution.
If a dissolved gas is consumed by chemical reactions in solution (e.g. aerobic
metabolism of oxygen), the equilibrium partial pressure required of a gas phase
falls. Upon exposure to a gas phase with a higher partial pressure, such as in the
lungs or an oxygenator, gas molecules will dissolve, increasing the solution concen-
tration until equilibrium is again reached. This is the primary advantage of express-
ing concentrations of dissolved gases in body fluids as partial pressures – apart from
measurement practicalities, it allows easy quantification of the concentration gradi-
ent from the site of gas exposure to the site of usage. Its inconvenience comes when
considering stoichiometric relationships such as the respiratory quotient.
The equations and constants introduced thus far describe a steady-state where a
fixed quantity of gas is in equilibrium with a solution and assumes instantaneous
reactions occurring in stationary homogenous mediums. However, both in the
human body and in the oxygenators used for ECMO, an exchange membrane is
always interposed between the gas phase and the body fluids it dissolves in. Even in
the case of porous membranes, a direct gas/blood interface is usually prevented by
formation of a biofilm comprised of adsorbed blood proteins that adhere after a
short period of operation.
Membrane properties impose additional time constraints over which gas
exchange can occur and requires consideration of the mass transport of molecules
into the body as flux (J), defined as the passage of a quantity of solute per unit time.
The membrane flux of solute down a concentration gradient is described by
Fick’s law of diffusion [16]:
∆C
J = − D. A (1.7)
∆l
This expression says the flux (J, mmol.s−1) of a solute over the thickness of a
membrane (Δl, cm) is proportional to the diffusivity coefficient, D, the concentra-
tion gradient across the membrane (ΔC, mmol.mL−1 or mmol.cm−3) and the area of
the diffusion front (A, cm2). The minus sign is mathematically required to describe
flux from a high concentration to a low concentration [16]. This universal statement
of mass transport is applicable not only to the extracorporeal oxygenator but also to
gas transport from plasma to interstitial fluid and into cells.
1 Physiology of Extracorporeal Life Support (ECLS) 13
Dδ
DEff = ε (1.8)
τ
D is the diffusion coefficient described above for the gas liquid interface within
the pores, and ε is the porosity – the fraction of the interface occupied by gas. τ is
the tortuosity – a geometric description accounting for the increased length of diffu-
sion within the membrane. δ is constrictivity, which describes resistance to mole-
cules traversing pores due to their size relative to the pore diameter (Fig. 1.3)
[17, 18].
Under real conditions the value determined for the effective diffusion coefficient
is inseparable from the properties of any biofilm of adsorbed proteins or stationary
layer of blood [11, 17]. Furthermore the greater part of oxygen traversing the mem-
brane immediately undergoes a chemical reaction with haemoglobin until the latter
is saturated. This chemical reaction sustains the concentration gradient and is
described as an enhancement factor. After incorporating haemoglobin, the DEff for
oxygen becomes not only dependent on the membrane characteristics discussed but
also a function of haematocrit (%Hct). Equation 1.9 demonstrates an example of a
term for the effective diffusivity of a membrane exposed to a turbulent bovine
bloodstream [19, 20]:
By combining Eq. 1.4b (Henry’s law) for solute concentration and Fick’s law
(Eq. 1.7), an equation for diffusive membrane flux where the driving force is
expressed as a partial pressure gradient can be derived where k is the permeability
constant – the product of the solubility coefficient (SC, mmol.L−1.mmHg−1) already
14 M. J. Brain et al.
defined and effective diffusivity (DEff, cm2.s−1) having the units mol.cm−1.s−1.
mmHg−1 [21]:
Driving Force
Mass transfer per unit area =
Resistance to Transport
J (
PO2 ( Gas ) − PO2 ( Plasma ) )
( mmol.s −1
.cm −2 )
A
=
RTotal
(1.11)
∆l
where RTotal =
−kO2
From Eq. 1.11 it is apparent that maintaining the partial pressure gradient
between the gas phase and the blood phase is important in maximising oxygen flux.
Maintenance of the local pressure gradient at any point along a hollow fibre is influ-
enced by three design factors and one operational factor: the time the blood is
exposed to the membrane, relative flow direction, local turbulence and
haematocrit.
For a static liquid below a static gas, the rate of diffusion will decrease exponen-
tially until equilibrium when the conditions of Eq. 1.4b are met, that is, the rate of
gas dissolving into the liquid is equal to the rate of molecules leaving the liquid
(Fig. 1.4 left panel). It is clear from this figure that too short an exposure time will
result in submaximal oxygenation.
1 Physiology of Extracorporeal Life Support (ECLS) 15
14
Partial Pressure vs Time for a static gas/liquid
O2 Content vs Membrane Exposure Time
12
O2 Content mL/dL
8
Oxygen Content (ml/dL) for Hb
10 g/dL pH 7.44 Temp 37.5° & BE 0
6
Dissolved O2 in solution
4
0
0 0.5 1 1.5 2 2.5 3 3.5 4
Time
Time (seconds)
Fig. 1.4 Partial pressure vs. time for a gas dissolving in a liquid: Left panel depicts partial pres-
sures approaching equilibrium for a static solution below a gas phase. The slope of the curve (i.e.
the rate at which equilibrium is approached) is proportional to the concentration difference. Right
panel describes the oxygen content of serum containing red blood cells vs. time exposed to 100%
oxygen across a membrane. Time has been adjusted to approximate transit times in current oxy-
genators. (Plotted from equations of Katoh and Nickalls [19, 22])
Figure 1.4a depicts a stationary blood and gas phase; however a similar pattern
exists for two phases moving in the same direction (co-current flow), and thus at low
flow rates, equilibrium will occur, and diffusive flux will cease. Inspection of the
figure makes it apparent that replacing gas partly depleted of oxygen (where CO2
also contributes to the total partial pressure) will maintain the concentration gradi-
ent. Utilising countercurrent flow where blood and gas flow in opposite directions
decreases the maximum concentration gradient at the blood inlet end of a hollow
fibre but increases the gradient at the outlet, thereby maintaining a gradient over the
entire fibre and allowing flux to continue along the membrane relatively indepen-
dent of flow rates. Cross current flow is also utilised (demonstrated in Fig. 1.3)
resulting in differing gradients across the bloodstream.
16 M. J. Brain et al.
The total resistance to mass solute movement in Eq. 1.11, RTotal, is the sum of com-
ponent resistances, which can be divided into gas phase resistance (RG), membrane
resistance RM and blood side resistance, RB: RTotal = RG + RM + RB. Of these RG is
negligible and the factors influencing RM have been discussed. The most variable
and often most prominent component is RB due to the formation of a stationary film
on the blood side of the membrane (Fig. 1.3 boundary layer).
It is not practical for manufacturers to specify a membrane surface area (A, Eq. 1.11)
that encompasses the complex microscopic geometry of oxygenator tubules and
pores, the latter being incorporated into DEff as discussed [17]. Furthermore, it can-
not be assumed that any oxygenator design utilises the entire membrane area evenly.
Thus, while flux per unit area (J/A) is a useful description of isolated membrane
performance, oxygenators are better characterised by their total flux incorporating
area into the equation for resistance:
(P − PO2 ( Plasma ) )
( mL.min ) J
−1 O2 ( Gas )
=
RTotal
Oxygenator
(1.12)
∆l
where RTotal =
− kO2 . A
Although lacking the precision of Eq. 1.11 in defining properties of the mem-
brane, Eq. 1.12 can be incorporated into monitoring gas exchange efficiency of an
individual oxygenator over time and will be discussed after oxygen carriage is con-
sidered. The value of JOxygenator is usually reported in oxygenator product specifica-
tion sheets at varying blood flows.
1 Physiology of Extracorporeal Life Support (ECLS) 17
water flux
K UF =
transmembrane pressure
(1.13)
QUF
=
PBlood − PGas
QUF (mL.min-1) is termed the ultrafiltration rate and describes the appearance of
fluid within the hollow fibres, while the terms for pressure (PBlood and Pgas) describe
the heights of a fluid column relative to atmospheric pressure in each compartment.
Many of the factors already described for resistance to diffusion will be contained
in the ultrafiltration coefficient and will not be discussed further. The ultrafiltrate
represents a homogenous fluid that will contain dissolved solutes from plasma pro-
portional to the size of membrane pores, which are typically smaller than 1 micron.
A major drawback of early microporous membranes was significant plasma
leakage, as the open porous structure allowed water flux. This has been significantly
alleviated by newer closed-fibre membranes; however, some water flux still occurs
under normal operating conditions. The water evaporates in the fresh gas flow and
leads to an insensible water loss proportional to the fresh gas flow and may reach
significance when supporting low bodyweight patients [12].
Oxygenation has been described as a cascade of partial pressure gradients from the
external environment into the arterial blood and from capillaries into the extracel-
lular space, before finally diffusing across cell membranes into mitochondria.
The quantity of oxygen entering the body at steady state equals its consumption.
If no consumption occurs, no arterial-venous oxygen difference will exist, and the
circulating blood will be in equilibrium with the oxygen in the gas phase as per
Eqs. 1.3 and 1.10, abolishing further mass transfer. Oxygen will continue to diffuse
across the oxygenator; however, the movement will be in both directions at an
equal rate.
18 M. J. Brain et al.
Oxygen Transport
Oxygen Carriage
0.8
15
0.7
0.6
0.5
10
0.4
0.3
SaO2 at pH 7.44 Temp 37.5° & BE 0
5
0.2 SaO2 at pH 7.32 Temp 37.5° & BE 0
Fig. 1.5 The oxygen haemoglobin dissociation curve as calculated by the Thomas modification of
the Kelman Eq. 1.9. Also shown is the oxygen content for haemoglobin concentration of 10 g/dL
after applying Eq. 1.14. The effect of pH is demonstrated
The total oxygen content can then be multiplied by blood flow (cardiac output)
to give oxygen delivery, DO2 which has the units of flux: mL.min−1. Note the scaling
factor of 10 to convert the units of oxygen content to ml.L−1.
⋅
DO2 = QB × Ca O2 × 10 (1.15)
20 M. J. Brain et al.
The Bohr effect describes alterations in haemoglobin oxygen affinity due to carbon
dioxide and hydrogen ion concentrations. Carbon dioxide binds to amino acids in
the outer chains of haemoglobin to form carbaminohaemoglobin, stabilising the
molecule with the ferric elements in deeper crypts and facilitating release of oxygen
from haemoglobin. Similarly, increasing temperature and increasing hydrogen ion
concentrations stabilise haemoglobin in the deoxygenated state. Similar to oxygen,
carbon dioxide binding is reversible, and in compartments with a low CO2 concen-
tration, the effect is reversed, promoting oxygen uptake (the Haldane effect). This is
of importance when considering oxygenation in systems designed primarily for
CO2 removal and is discussed below.
The glycolysis product, 2,3-DPG also decreases haemoglobin affinity for oxy-
gen. 2,3-DPG binds to deoxygenated haemoglobin, lowering the apparent affinity
for oxygen by altering the electrostatic bonds that maintain the quaternary configu-
ration [10]. This has the most significance in stored blood where after 2 weeks,
2,3-DPG levels become negligible. After transfusion, DPG levels do not return to
normal until nearly 48 h [24]. In the absence of 2,3-DPG, the affinity for oxygen is
increased resulting in relatively lower oxygen release in tissues for the same pO2.
This phenomenon is only likely to be a factor in the most severe oxygenation prob-
lems, and the benefit of increased oxygen binding sites from transfusion probably
outweighs the transiently lower delivery (see Figs. 1.6 and 1.7: Venous oxygen satu-
ration vs. haemoglobin and cardiac output in a patient receiving VV-ECMO. VO2 is
held constant [25]).
Deoxygenated
Portion of
Venous Return
Fig. 1.6 Schematic of a VV-ECMO circuit. A mixture of fully oxygenated venous and deoxygen-
ated venous blood enters the right heart and perfuses the pulmonary circulation before the left heart
distributes it systemically (hence end-tidal CO2 will be not reflect true mixed venous pCO2).
Varying with the cannula position and venous return rate, some amount of recirculation is very
common, reducing the amount of oxygen delivered. Depending on recirculation, it may not be
possible to sample true mixed venous blood
1 Physiology of Extracorporeal Life Support (ECLS) 21
100% 100%
350.00 350.00
80% 80%
300.00 300.00
Hb Saturation
Hb Saturation
VO2 ml/min
VO2 ml/min
60% 60%
250.00 250.00
40% 40%
200.00 200.00
20% 20%
0% 150.00 0% 150.00
2.00 3.00 4.00 5.00 6.00 7.00 2.00 3.00 4.00 5.00 6.00 7.00
VV-ECMO Saturations vs Cardiac Output. VV-ECMO Flow 4 VV-ECMO Saturations vs Cardiac Output. VV-ECMO Flow 4
L/min, Required VO2 250 ml/min and Hb 10 g/dL L/min, Required VO2 250 ml/min and Hb 7 g/dL
100% 100%
350.00 350.00
80% 80%
300.00 300.00
Hb Saturation
Hb Saturation
VO2 ml/min
VO2 ml/min
60% 60%
250.00 250.00
40% 40%
200.00 200.00
20% 20%
0% 150.00 0% 150.00
2.00 3.00 4.00 5.00 6.00 7.00 2.00 3.00 4.00 5.00 6.00 7.00
Fig. 1.7 Arterial (SaO2) and central mixed-venous oxygen saturations (SvO2) vs. cardiac output in
a patient fully supported by VV-ECMO. Graph titles show physiologic and flow conditions along
with haemoglobin. See text for discussion [25]
any partial pressure of oxygen and allow for shifts of the curve due to temperature,
[H+] and carbon dioxide.
For convenience, the oxygen-haemoglobin dissociation curve is frequently
described by the partial pressure at which 50% of haemoglobin is saturated. A nor-
mal p50 for arterial blood is 26.3 mmHg. Values higher than this describe a ‘right-
shifted’ curve, that is haemoglobin affinity for oxygen is less. Two curves of
haemoglobin saturation are demonstrated in Fig. 1.5, the only difference being the
hydrogen ion content reflecting the higher carbon dioxide concentration in venous
blood. At high oxygen partial pressures consistent with arterial blood, the difference
in haemoglobin saturation between curves with differing p50 is minimal. In con-
trast, there is a significant difference in the saturation of Hb at a partial pressure of
40 mmHg commonly found in venous blood.
The implications of the oxygen Hb dissociation curve become clearer when oxy-
gen content is also plotted on the same chart (Fig. 1.5). At haemoglobin of 10 g.
dL−1, the oxygen content at a partial pressure of 100 mmHg is 13 mL.dL−1 and mini-
mally affected by the arteriovenous pH difference. However, at an oxygen partial
pressure of 40 mmHg, a 1 mL.dL−1 difference becomes apparent between the two
content curves, being 10 mL.dL−1 at pH of 7.44 and 9 mL.dL−1 at pH of 7.32. In the
tissues where oxygen is utilised in metabolism, this ‘right shifting’ of the curve as
the products of metabolism acidify capillary blood serves to bolster the partial pres-
sure gradient for oxygen diffusion from capillary to cell.
Unfortunately, ‘right-shifted’ oxygen haemoglobin dissociation curves, while
advantageous for unloading oxygen in acidotic tissues, may be counterproductive at
sites of oxygen uptake if abnormally low alveolar oxygen partial pressures exist as
in the adult respiratory distress syndrome (ARDS). Inspection of Fig. 1.5 reveals
that if oxygen uptake were to occur at a partial pressure of 60 mmHg, a right shifted
curve (pH 7.32) will carry 0.5 mL.dL−1 less oxygen. This gap widens if uptake
occurs at even lower partial pressures.
mean O2 tensions of the unmixed samples massively overestimates the final partial
pressures after mixing.
Understanding this concept is important as it highlights a physical limitation on
systemic oxygenation: utilising a saturable oxygen carrier (haemoglobin) makes
oxygen delivery flow limited. Even if supranormal oxygen tensions are achieved via
an extracorporeal circuit, an inadequate ratio of circuit flow to cardiac output results
in suboptimal oxygen delivery.
Various access configurations are utilised (Fig. 1.1); however, one of the most
common techniques is to cannulate the common femoral vein with a cannula which
has side and end fenestrations that allow blood to be drained from multiple points.
So called ‘multistage’ cannula where the side holes extend over 20–25 cm can be
placed via the inferior vena cava, across the right atrium and into the superior vena
cava to maximise harvest of venous return. If the access cannula only has openings
near the tip, then the tip is typically placed 5–10 cm below the cavo-atrial junction.
The return cannula has a single terminal orifice which terminates in the right atrium.
If higher extracorporeal circuit flows are required, the inferior vena cava may col-
lapse around the multistage cannula, intermittently restricting flow and causing the
external circuit to ‘shudder’. If higher circuit flows are necessary to achieve ade-
quate oxygenation and this negative access pressure cannot be resolved by giving
fluid, a second access cannula may need to be placed in the superior vena cava via
the internal jugular vein.
Figure 1.6 demonstrates a basic VV-ECMO circuit configuration. It can be appre-
ciated that the ECMO circuit is in parallel to venous return and thus a mixture of
oxygenated blood from the ECMO circuit and deoxygenated blood from the venous
return will enter the right heart. To appreciate the implications of this parallel circuit,
consider a young adult with severe acute respiratory distress syndrome fully sup-
ported by VV-ECMO with both femoral and internal jugular access (Fig. 1.6). Chest
x-ray demonstrates bilateral ‘white-out’, and it will be assumed the lungs are not
contributing to systemic oxygenation.
If the patient has the following parameters: cardiac output of 4 L/min, ECMO
flow of 4 L/min, a mixed central venous haemoglobin oxygen saturation of 52% and
an arterial saturation of 99%. What will be the effect on his saturations if his cardiac
output were to rise to 7 L.min−1 or were to fall to 2.5 L.min−1?
Apart from the non-contribution of the lungs, for this analysis other assumptions
include an oxygen-haemoglobin dissociation curve with a normal p50 and good
cannula position with minimal recirculation between the access and return lumens
in the inferior vena cava. To fully develop this model and illustrate several key
points, two other parameters are required: the haemoglobin concentration and the
patient’s total oxygen consumption (VO2). Initially, this hypothetical patient has a
haemoglobin concentration of 10 g/dL and is at steady state consuming 250 mL/min
of oxygen with no markers of tissue hypoxia. Thus, there are four independent vari-
ables: the cardiac output, the ECMO flow rate, the haemoglobin concentration and
the target VO2.
While providing circulating oxygenated blood is the goal of ECMO, it must be
highlighted that the VO2 required to avoid anaerobic metabolism is a parameter that
can only be achieved if the amount of oxygen delivered matches consumption. If
this required VO2 exceeds tissue delivery, VO2 is then supply-limited, initially
resulting in higher oxygen extraction with mixed-venous saturation decreasing first,
followed by clinical and biochemical markers of hypoxia such as confusion, oligu-
ria and rising lactate as anaerobic metabolism ensues.
Figure 1.7a demonstrates that under these conditions, increasing the cardiac out-
put from 4 L.min−1 to 7 L.min−1 will cause a drop in the arterial saturations from
1 Physiology of Extracorporeal Life Support (ECLS) 25
99% to 84%. In contrast dropping the cardiac output to 3 L.min−1 won’t affect the
arterial saturation but will cause the central mixed-venous saturation to fall from
52% to 25% as increased tissue extraction occurs.
The equations required to generate this model are the content and delivery equa-
tions (Eqs. 1.14 and 1.15) from which the principle of conservation of mass is
applied to determine the oxygen content at each of the following points: the ECMO
return cannula, the venous return and the pulmonary artery (Fig. 1.6). In VV-ECMO
the pulmonary artery is not the correct sampling site for mixed-venous saturation,
instead the pre-oxygenator blood tubing is the closest approximation.
The methods described in Table 1.1 for mixing bloodstreams are used to deter-
mine the pulmonary artery oxygen content and saturations. In the absence of lung
function, this represents systemic arterial oxygen delivery as well. Modelling the
solution requires an iterative approach to determine the highest achievable VO2 (if
the required VO2 cannot be met) by altering the tissue oxygen extraction – this
determines the venous oxygen flux to the right heart. Example values from Fig. 1.7a
are shown in Table 1.2.
To explain the fall in arterial saturations with increasing cardiac output, it should
be appreciated that the required VO2 did not change. What did change was the
total venous return, which increased by 75% with the increased cardiac output.
The ECMO flow remained at 4 L.min−1, so the ‘shunt’ bypassing the ECMO oxy-
genator via the great veins increased resulting in the fraction of deoxygenated
venous blood in the right ventricle increasing from 3% to 43%. The shunt is cal-
culated as (cardiac output – oxygenated blood flow)/cardiac output. The oxygen-
ated blood flow equals the cardiac output if the cardiac output is less than ECMO
blood flow.
In summary, the resulting arterial saturation is analogous to the example of mix-
ing bloodstreams of different content, and the same effect can be achieved if cardiac
output is left constant and ECMO flow is decreased (Fig. 1.7b). In practice, increases
in cardiac output are likely to be accompanied by increased VO2, and this will result
in a fall of mixed-venous saturation (Fig. 1.7c).
the right atrium, flowing in a retrograde fashion back down the proximal IVC while
no systemic oxygenation occurs. Similarly at low cardiac outputs, there will be
atrio-caval recirculation; the haemoglobin however is saturated with oxygen on the
first pass and will be unable to carry more. Thus when ECMO flow exceeds cardiac
output, the saturation is cardiac output limited. To supply the bodies required VO2
at low cardiac outputs, oxygen extraction has to increase, and the mixed central
venous oxygen content and venous saturation fall.
At high cardiac outputs and hence high venous return, recirculation will be sig-
nificantly lower as the oxygenated blood will be pushed through the right atrium
into the right ventricle. Figure 1.7c demonstrates falling mixed-venous saturations
with increasing VO2 requirements at higher cardiac outputs. This is caused by a
combination of falling arterial oxygen content by the reducing ECMO oxygenated
proportion of the venous return and increased peripheral extraction to achieve the
higher VO2 requirement.
The parameters in Fig. 1.7d are identical to Fig. 1.7a except for a lower haemoglo-
bin. The lower oxygen-carrying capacity means that at any cardiac output, the oxy-
gen extraction must be greater and hence the venous saturation will be lower. At
lower cardiac outputs, the combined effect is enough that no further oxygen extrac-
tion can occur and the achieved VO2 falls lower than the target VO2 – under these
conditions signs of tissue hypoxia will occur. Thus in the setting of low cardiac
output and borderline oxygenation, increasing the haemoglobin by transfusion may
alleviate hypoxia, while consideration is given to augmenting circulatory support.
Modelling VV-ECMO
The above analysis can be combined to display informative mixed central venous
and arterial saturations for any cardiac output/haemoglobin concentration (Figs. 1.8
and 1.9).
The surface plot for arterial oxygen saturations (Fig. 1.8) indicates that bolster-
ing haemoglobin to improve arterial oxygen saturations will only be of significant
benefit at higher cardiac outputs and even then cannot fully compensate for the
shunt past the ECMO circuit.
In contrast, Fig. 1.9 demonstrates venous saturations always increase with more
oxygen-carrying capacity. The left side of the venous saturation surface plot further
emphasises that oxygen delivery is dependent on cardiac output and falls when
output is less than ECMO flow rates.
28 M. J. Brain et al.
Fig. 1.8 Arterial oxygen saturation vs. haemoglobin and cardiac output in a patient receiving
VV-ECMO. VO2 is held constant [25]
Fig. 1.9 Venous oxygen saturation vs. haemoglobin and cardiac output in a patient receiving
VV-ECMO. VO2 is held constant [25]
1 Physiology of Extracorporeal Life Support (ECLS) 29
Fig. 1.10 Recirculation due to tricuspid closure. The left hand image demonstrates recirculation
in the atria after tricuspid valve closure at the beginning of ventricular systole (right ventricle at
top, atrium containing colour flow that is impacting the tricuspid valve). The right hand image
shows the return cannula in the superior vena cava (bicaval cannula) with blood flowing across the
tricuspid valve into the right ventricle. (Image M. Brain by permission of The Alfred Intensive
Care Unit, The Alfred Hospital, Melbourne)
30 M. J. Brain et al.
In the ECLS setting, the primary concern is optimising oxygen delivery within
the limits of the available system. Equations 1.16 and 1.17 describe oxygen content
and delivery. From these equations and Fig. 1.8, it can be appreciated that increasing
the haemoglobin concentration accommodates lower arterial saturations while aug-
menting mixed-venous saturation by increasing oxygen content and thus delivery.
This is important when other limitations such as ECMO flow rates or problems with
recirculation prevent achievement of higher saturation readings.
The pulmonary arterial mixed-venous oxygen saturation (or in VV-ECMO the pre-
oxygenator saturation) is an important measure of oxygen uptake. By utilising
blood gas analysis sampled from one of these sites, the central venous oxygen con-
tent (CvO2) can be calculated, again using Eq. 1.14. This concept has led to classic
descriptions of supply-dependent and independent oxygen uptake, and it is informa-
tive when interpreting mixed-venous saturations to calculate the content difference
with an approximation of cardiac output.
A critical point to recall when interpreting mixed-venous oxygen saturations is
that it does not represent the lowest peripheral saturation (see mixing blood of dif-
fering content above). Different organs (and in cases distinct parts of organs) will
have different metabolic activity and hence oxygen extraction. Those organs with a
high extraction ratio may still become supply-limited and thus hypoxic even when
the total organism DO2 exceeds VO2. What is actually important is the partial pres-
sure of oxygen in the capillary bed furthest from an arteriole in the relevant organ,
as this defines the lowest local oxygen gradient from blood to cells. Thus clinical
markers of hypoxia such as neurological status and renal function, along with lac-
tate trends, become as important as mixed CvO2. In the absence of practical ways to
measure either capillary oxygen tension or organ oxygen uptake, the aim should be
to target a venous oxygen content that supports higher oxygen extraction by some
organs, and this can be achieved by supplying more oxygen, increasing cardiac
output or increasing the haemoglobin.
The solubility of carbon dioxide in plasma is described by Henry’s law (Eq. 1.4b)
with a value for SC at 37 °C of 0.0308 mmol.L−1.mmHg−1 [15]. At a pCO2 of
40 mmHg, this equates to 1.232 mmol.L−1 or 2.7 mL.dL−1 of dissolved carbon diox-
ide. The solubility decreases to 2.88 × 10−2 mmol.L−1.mmHg−1 at 40 °C.
Like oxygen, only a small proportion of carbon dioxide is transported as dis-
solved gas, the remainder being in chemical equilibrium with the gas phase through
reactions with water for a total blood content of nearly 500 mL.L−1. In blood, these
reactions largely occur in erythrocytes where the enzyme carbonic anhydrase (CA)
is abundant [28]; however, CA is distributed in many other tissues and in pulmonary
and renal capillaries [15].
The following reactions describe the hydration of dissolved carbon dioxide with
arrows, indicating where the majority of reactants are at equilibrium. Equilibrium
concentrations are independent of the rate at which equilibrium is achieved. Though
not a hydration reaction, the reversible reaction of dissolved CO2 with amino acids
on haemoglobin is also listed here as part of CO2 storage:
CO 2 ( gas ) + H 2 O
CO 2 ( dissolved ) + H 2 O a. Dissolving in water
C. A .
CO 2 ( dissolved ) + H 2 O H 2 CO3 b. Formation of carbonic acid (1.16)
C. A .
C. A .
CO 2 ( dissolved ) + OH − HCO3− c. Reaction with hydroxyl ion
C. A .
+ −
H 2 CO3 H + HCO 3 d. Dissociation to bicarbonate
CO 2 ( dissolved ) + R − NH 2 R − NH − COO − + Η + e. Formation of carbamino groups
Being lipid soluble carbon dioxide can be considered to diffuse through all mem-
branes with local production determining regions of higher concentrations, while
areas of gas exchange have the lowest concentration. This is in contrast to HCO3−
which cannot easily cross membranes unless being exchanged for another anion
(Fig. 1.11).
Of the molecules in the Henderson-Hasselbalch equation (Eq. 1.17a), CO2 is the
only independent variable with the amount in the body being determined by the
1
Occasionally the term αPCO2 is used in this equation to describe the total concentration of CO2
and carbonic acid however the concentration of the latter is orders of magnitude smaller than the
former. Thus [CO2] can be calculated accurately from Henry’s Law without alteration of the solu-
bility constant (Eq. 1.4b).
1 Physiology of Extracorporeal Life Support (ECLS) 33
CO2
ADP
ADP
Energy Glucose
Expenditure
ATP
ATP
O2
CO2 CO 2 CO 2
+ + +
H 2O H2O H 2O
C.A.
Lactate Lactate
- (I)
O O
- (I)
+ H+ O O + H+
H 3C OH H 3C OH
Lactic Acid Lactic Acid
Fig. 1.11 Dissolved carbon dioxide diffuses across all tissue planes and dissociates in each com-
partment to HCO3− with the ratio of CO2 to HCO3− depending on the [H+]
balance of production and flux out of the body [13]. The amount of hydrogen ions
and bicarbonate in any compartment depend not only on the reactions of carbon
dioxide with water but also on the concentrations of other strong and weak electro-
lytes with the final balance being determined by the need to maintain electrical
neutrality and the ionisation constant of water [13, 34]. The simplest summary of
this complex interaction is that bicarbonate behaves as an electrical ‘spacer’ in the
following equation for electrical neutrality:
34 M. J. Brain et al.
30 35
Oxygenated Blood [CO2]Total (mmol.L-1)
AV Difference (mmol.L-1)
Deoxygenated Blood [CO2]Total (mmol.L-1)
30
25
20
[CO2]Total (mmol.L-1 whole blood)
15
Erythrocyte HCO3- 15
(mmol)
10
10
PCO2 (mmHg)
Fig. 1.12 The CO2 dissociation curve for whole blood at 37° and haematocrit of 45%. Upper lines
represent dissociation of total CO2 in mmol.L−1 in deoxygenated and oxygenated whole blood; the
line AV Difference demonstrates increased carriage by carbamino groups as the oxygen saturation
falls. To convert mmol.L−1 to ml.dL−1 multiply by 2.226. Shaded lower areas demonstrate the
amount of CO2 (mmol) in erythrocyte water (325 mL per L blood at pCO2 40 mmHg) and plasma
water (509 mL per L blood at pCO2 40 mmHg). The total water content of 1 L of whole blood is
836 mL. Carbamino CO2 represents the maximum amount of CO2 that fully deoxygenated haemo-
globin can carry. Summation of the amount of CO2 in mmol in shaded areas at any pCO2 divided
by whole blood water (836 mL) yields the concentration of CO2. (After [29–33])
Na + + K + + 2 × Ca 2 + + 2 × Mg 2 + + z +
= Cl− + HCO3− + 1.8 × PO 4 2 − + 0.28 × Alb − + lactate − + x −
(1.18)
CO 2
1 Physiology of Extracorporeal Life Support (ECLS) 35
This occurs because of the large supply of dissolved CO2 throughout the body
and the ready reversibility of its reaction with water. In contrast, concentrations of
the other charged species in Eq. 1.182 are all tightly regulated by homeostatic pro-
cesses and cannot be allowed to rapidly change (Fig. 1.11). Thus if lactic acid pro-
duction were to rise, the immediate effect is a shift of [HCO3−] to CO2 to maintain
electrical neutrality. A secondary increase in minute ventilation facilitates clearance
of the elevated CO2, and over time renal chloride loss increases to restore balance.
These mechanisms facilitate understanding the alkalotic effects of hypoalbumin-
emia and dilutional acidosis from relative hyperchloraemia commonly encountered
in critically ill patients. The reader is referred to other sources for a more detailed
discussion [13, 34].
Carbonic Anhydrase
Figure 1.12 demonstrates the total carbon dioxide content of whole blood at increas-
ing partial pressures. Total CO2 (often confusingly termed total bicarbonate) is
obtained by measuring the volume of CO2 gas produced after the addition of a
strong acid to a blood sample which shifts Eq. 1.17a fully to the left [36]. In contrast
the bicarbonate reported on an arterial blood gas sample is calculated from the mea-
sured pH and pCO2 using Eq. 1.17c and is typically 2–4 mmol.L−1 lower than total
CO2. This difference is due to carbon dioxide carriage on amino acids and particu-
larly carbamino formation on haemoglobin (Eq. 1.16e).
Inspection of the AV difference in Fig. 1.12 at a haematocrit of 45% demon-
strates the pCO2 increases from 40 mmHg to 47 mmHg as tissue metabolism
consumes oxygen. In this example the total CO2 rises correspondingly from
2
Being electrical neutrality, the units are available charge (milli-equivalents per litre mEq.L−1)
hence the concentration of double valent ions is multiplied by 2 to account for their charge density.
The multiplier of 1.8 for inorganic phosphate and 0.28 for albumin are approximations as the
charge density for these weak acids varies slightly with pH. [z+] and [x−] refer to other unmeasured
exogenous or endogenous cations and anions.
36 M. J. Brain et al.
19.6 mmol.L−1 (43.5 mL.dL−1) to 22.4 mmol.L−1 (49.8 mL.dL−1) the difference
being due to tissue CO2 production and is proportional to the respiratory quotient
[29]. The fraction of CO2 carried as carbamino groups increases in this process as
oxygen unloading increases the affinity of haemoglobin for carbon dioxide (haemo-
globin is more basic when deoxygenated, shifting Eq. 1.17b toward bicarbonate
[15]). It should be emphasised that whole blood bicarbonate is carried by both intra-
cellular and extracellular water [30].
Carbon dioxide presents a classic clearance problem. At a steady VO2 and fixed RQ,
a constant amount of CO2 will be produced. If the plasma partial pressure is also
stable, then the amount eliminated must equal production and also be constant.
Consider two oxygenators with identical membrane characteristics but differing
surface areas with the second membrane half the area of the first. Fresh gas and
ECMO flows, CO2 production and elimination are all constant, with a membrane
flux of 250 mL.min−1 in each system. The resistance to diffusive transport (RTotal) is
identical for the two membranes so that from Eq. 1.11:
∆l
(
J × RTotal = Area × PCO2 ( Plasma ) − PCO2 ( Gas ) ) where RTotal =
−kO2
= Area1 × ∆P1 Oxygenator 1 (1.19)
Area 2
= × ( x × ∆P2 ) Oxygenator 2
x
The numerals denote the two oxygenators, and x is a scaling factor for membrane
area. Halving the area for available transport (making x = 2) will double the required
transmembrane pressure gradient to achieve the same CO2 flux.
Due to the lower solubility of oxygen, a much larger membrane area is required
for oxygenation than is necessary for carbon dioxide removal. The corollary is stan-
dard oxygenators that are capable of excessive CO2 removal resulting in hypocap-
nia. Adding CO2 to the sweep gas or reducing the sweep gas flow so that the
convection of carbon dioxide away from the membrane is slowed may alleviate this.
ECCO2R refers to the support of hypercapnic respiratory failure, usually in the set-
ting of acutely decompensated pulmonary disease. Modern systems utilise a scaled
down VV-ECMO circuit allowing significantly smaller vascular access catheters.
1 Physiology of Extracorporeal Life Support (ECLS) 37
While achieving clearance of carbon dioxide, the limited membrane area and lower
blood flow significantly reduce any oxygenation effect from these systems.
Lower blood flows in ECCO2R are made possible by the higher solubility of carbon
dioxide and its more linear dissociation curve across the physiologic range com-
pared to the sigmoid curve for oxygen saturation of haemoglobin (Fig. 1.13).
Complete saturation of haemoglobin at a concentration of 10 g.dL−1 with oxygen
results in a maximum oxygen carriage of approximately 13 mL.dL−1. If returning
venous blood has a saturation of 70%, its oxygen content is 9.3 ml.dL−1 (Eq. 1.14
and Fig. 1.5) allowing only 37 mL of oxygen to be added per litre. Thus an adult
consuming oxygen at 300 mL.min−1 with a mixed-venous saturation of 70% would
require an ECMO flow and cardiac output over 7 L.min−1.
In contrast to the saturable uptake of oxygen, nearly all the carbon dioxide in
venous blood could be removed with sufficient exposure to the gas exchange mem-
brane and a constant supply of fresh gas to maintain the trans-membrane gradient.
From the dissociation curve for carbon dioxide (Fig. 1.12), a mixed-venous pCO2 of
60 mmHg equates to 52 ml.dL−1. Hence from 500 mL of blood at this pCO2, 250 mL
of CO2 could be removed by an efficient gas exchange system allowing an extracor-
poreal circuit running at 500 ml.min−1 to clear the minutely CO2 load of adult
patients.
60
25
50
20
40
Content in Blood ml/dL
30
Oxygen Content for Hb 10 g/dL pH 7.44 Temp
37.5° & BE 0
5
10
0 0
0 10 20 30 40 50 60 70 80 90 100
Partial Pressure (mmHg)
Fig. 1.13 Comparison of oxygen and carbon dioxide content in blood vs. partial pressure
Fig. 1.14 Acute right ventricular (RV) enlargement with secondary impairment of left ventricular
(LV) filling in hypoxia. Note the intrusion of the basal septum into the LV in the long axis view and
the flattened septum and enlarged RV in the short axis view. (Image M. Brain by permission of The
Alfred Intensive Care Unit, The Alfred Hospital, Melbourne)
40 M. J. Brain et al.
If alveoli are open to fresh gas flow, then ongoing oxygen consumption by the body
maintains a pressure gradient between oxygenated alveoli and venous blood in pul-
monary capillaries allowing diffusion to occur. This passive removal of oxygen
from the alveolus permits convection of further fresh gas into the alveolus without
tidal ventilation occurring – so-called passive oxygenation; however the lack of
tidal ventilation results in poor pulmonary CO2 clearance. Maintaining sufficient
mixing of oxygen with expired CO2 in the ventilated alveoli maintains this process
and can be achieved with low tidal volumes if sufficient pulmonary surface area
exists. Though supporting oxygenation, this high PEEP, low tidal volume ventila-
tion strategy leads to significant hypercapnia as convection of carbon dioxide out of
the lung is reduced.
Obstructive small airways disease imposes a separate (but often co-existent)
issue. Here, prolonged expiratory times are required to avoid excessive pulmonary
hyperinflation and barotrauma, and this strategy again decreases alveolar minute
ventilation and carbon dioxide clearance.
In both settings, allowing ‘permissive’ hypercapnia is a well-documented strat-
egy; however, in some patients the resulting arterial pCO2 is so high that significant
acidosis ensues. Institution of ECCO2-R in these settings improves carbon dioxide
clearance and may alleviate the hypercapnic acidosis. In other settings such as head
trauma, evidence suggests that maintaining normal pCO2 is optimal, and this may
not be achievable with coexistent lung injury without strategies such as ECCO2-R.
Oxygenation utilising high PEEP low tidal volume strategies can be augmented by
maximising the oxygen partial pressure gradient between perfused alveoli and pul-
monary capillary blood to drive diffusive flux. As discussed, each gas in a container
contributes to the total partial pressure, which in an alveolus is usually atmospheric
pressure saturated with water vapour (PIO2, Eq. 1.6). The partial pressure of carbon
dioxide in perfused alveoli will be close to the mixed-venous partial pressure, and
the alveolar gas equation describes the resulting alveolar oxygen tension (PAO2):
The alveolar gas equation and its variations. PA denotes alveolar partial pressure,
and Pa denotes arterial partial pressure. Expressions a. and b. require knowledge of
the respiratory quotient and assume that other soluble gases in the alveolus (mostly
nitrogen) have reached equilibrium between the alveolus and plasma. In critical
care environments where end-tidal CO2 monitoring is available, Eq. 1.20c is the
most precise expression of alveolar gas and makes no assumptions about the respi-
ratory quotient [15]:
1 Physiology of Extracorporeal Life Support (ECLS) 41
Pa CO 1 − R.Q.
PAO = PIO − 2
+ Pa × FIO × a.
2 2
R.Q. CO2 2
R.Q.
Pa CO (1.20)
PAO ≈ PIO − 2
b.
2 2
R.Q.
PIO − PEO
PAO = PIO − Pa CO × 2 2
c.
2 2
2
PEO
2
It is clear from inspecting Eq. 1.20b that the alveolar PAO2 and thus the pressure
gradient for gas exchange fall with increasing PaCO2. Table 1.3 demonstrates this
effect at different FiO2 levels. At a low PiO2, removal of CO2 with ECCO2-R signifi-
cantly increases the partial pressure gradient for oxygen diffusion from the alveolus
into plasma. However, the effect rapidly decreases with increasing PIO2 to the point
of being negligible.
In summary then any improvement in systemic oxygenation with ECCO2R is
most likely to be the result of facilitating ventilation strategies that optimise recruit-
ment and surface area for oxygenation by allowing alternative strategies for CO2
removal.
Table 1.3 The proportional effect on alveolar oxygen tension of CO2 removal at increasing FiO2
Atmospheric pressure (mmHg) 760 760 760
Water vapour (mmHg) 47 47 47
Calculated inspired oxygen partial pressure
PiO2 (mmHg) 150 428 570
RQ 0.85 0.85 0.85
Alveolar oxygen tension
PaCO2 (mmHg) 90 90 90
PAO2 (mmHg) 47 331 477
Alveolar oxygen tension with ECCO2-R
PaCO2 (mmHg) 40 40 40
PAO2 (mmHg) 106 390 536
Proportional effect of CO2 removal on alveolar PAO2
%change 125% 18% 12%
42 M. J. Brain et al.
(venous to venous or venous to arterial) that determine whether the mode of support
is primarily oxygenation and/or circulatory as well (Fig. 1.1).
This section will consider the path of blood in plastic conduits initially with only
a pump and then with an oxygenator interposed. Several of the parameters that
describe the extracorporeal blood path also apply to flow within the systemic circu-
lation, and interactions with the native circulation will be considered where appro-
priate. Various coatings may be applied to the circuitry to enhance biocompatibility,
reducing the inflammatory response to the circuit and the activation of the clotting
cascade in blood traversing the circuit.
Two principles, conservation of energy and conservation of mass, describe the
flow of blood through the extracorporeal circuit. The circuit flow will be considered
initially as conduits with a pump such as is found in a ventricular assist configura-
tions (LVAD in Fig. 1.1) before also discussing the oxygenator.
Two types of pump are currently used for ECLS, positive displacement pumps
(roller pumps) and velocity pumps, with the latter becoming more common. LVAD
devices generating pulsatile flow are no longer in use in adults and will not be
discussed.
Velocity Pumps
Velocity pumps increase the kinetic energy (velocity) of the flowing fluid by high
speed rotation of an impeller rotor. This kinetic energy is converted to potential
energy (pressure) by the containment of the flow stream.
1 Physiology of Extracorporeal Life Support (ECLS) 43
These pumps consist of a turbine design that accelerates blood in the line of flow,
typically operating at even higher speeds (up to 9000 rpm). Currently they are only
implemented in ventricular assist devices and, compared to centrifugal devices, are
sensitive to loading conditions.
The physics behind the sensitivity of centrifugal and axial flow devices to loading
conditions will be discussed after consideration of flow. In ECMO axial flow pumps,
the relationship between pump speed and blood flow is load dependent and has thus
resulted in Doppler monitoring of the velocity of blood in the tubing. In implanted
devices (ventricular assist devices), it is not currently possible to accurately provide
a durable long-term flow measurement system, so flow is approximated from inter-
preting variations in rotation speed. Accurate haematocrit measurements are impor-
tant in this estimation as will be discussed. In practice, exact determination of flow
may be less important than clinical evidence of adequate forward output (perfusion,
exercise tolerance) and ventricular unloading (pulmonary congestion), while echo-
cardiography can measure flow and demonstrate ventricular collapse.
The basic hydraulic principles outlined below describe the flow of a uniform incom-
pressible fluid in solid tubing. The conduits utilised in ECMO can be considered
solid tubes; however, blood only approximates a uniform fluid due to the presence
of suspended cells and chemical interactions of suspended proteins with the con-
duit walls.
44 M. J. Brain et al.
Two fluid characteristics are relevant to flow, density and viscosity. As a descrip-
tion of mass per unit volume, density (ρ) particularly affects the momentum of mov-
ing fluid and resistance to acceleration. For blood, density is slightly greater than
water at 1.06 g.mL−1 at 37 °C.
Viscoelasticity
14
12
10
Relative Viscosity
0 60%
25 50%
a
75
Cav
150 40%
Aor t ing
300
Aor t ing
a
t
d
a
cen
ato
end
a
700
20% m
Des
900 ae
Asc
1100 H
1300 10%
s
illarie
1500
Shea
s
r Rate
riole
(s -1)
Cap
Ar te
Fig. 1.15 Relative Viscosity of Blood compared to water vs. hematocrit and shear rate. Viscosity
rises exponentially with increasing hematocrit but falls with increasing shear rates, as aggregating
forces between red cells become less prominent (back panel). Viscosity data plotted from equa-
tions in [44] and shear rates of vessels from [45, 46]
1 Physiology of Extracorporeal Life Support (ECLS) 45
1. Implanted ventricular assist devices where blood flow is estimated from the rotor
rpm – the software of these devices often requires a current haematocrit.
2. In ECMO when interpreting a falling haematocrit. This may mask other increases
in circuit resistance such as clotting within the oxygenator and similarly must be
considered when pressure gradients across the oxygenator increase after transfu-
sion (discussed under ‘the oxygenator as a resistor’).
Figure 1.15 also displays decreasing viscosity with increasing shear rate. Shear
rate is a measure of the rate at which adjacent fluid streamlines move with respect
to each other (see laminar flow below). At low blood velocities, aggregates compris-
ing erythrocytes and plasma proteins contribute to adhesive forces. As blood veloc-
ity increases, aggregates become smaller as erythrocytes move with respect to each
other. At high velocities, laminar layers of erythrocytes and plasma move with sig-
nificantly less intracellular adhesive forces [45]. Vessel size contributes to the shear
rate in combination with velocity. Arterioles and capillaries having the highest shear
rates and the great veins the lowest (Fig. 1.15) [46, 47]. Centralisation of erythro-
cytes also occurs in small capillaries, decreasing the effective haematocrit contribut-
ing to lower relative viscosities [46, 48].
To accelerate blood, force is required to overcome the viscous adhesive forces
between erythrocyte aggregates and layers of plasma, along with drag forces cre-
ated by red cells moving at different rates to the surrounding plasma [44]. These
components of force will be lost as heat (viscous friction). Some of the accelerating
force will also deform erythrocytes and in that respect imparts potential energy that
is released as the erythrocytes relax back to their normal shape (elasticity).
⋅ PInlet − POutlet
Q= (1.21)
R
This is a statement of energy transfer: The region of high pressure has potential
energy, and the conduit allows this to be converted to kinetic energy by accelerating
a mass of fluid toward the low pressure region.
The resistances describe several properties of the fluid and conduit that limits the
rate of energy transfer. However, before discussing these resistances, it is important
to note this equation does not describe conditions within the conduit but only the
mass transfer between the two regions of pressures. Most importantly, most of the
flow acceleration, that is conversion of potential to kinetic energy, occurs before the
conduit orifice, and there is minimal pressure change within the conduit itself
(Fig. 1.16). Such flow convergence and divergence can be seen at the inlet and out-
lets of ECMO cannulas in vessels using colour flow Doppler techniques.
46 M. J. Brain et al.
Static Static
Fluid Fluid
Entrance
Length
Flow Flow
Acceleration & Deceleration &
Convergence Divergence
Fig. 1.16 Graphic description of the Bernoulli equation for laminar flow in a rigid level conduit
from a high pressure compartment to a low pressure compartment for an ideal fluid. Although the
driving pressure gradient for flow is between the two compartments, pressure (potential energy) is
actually lower in the region of flow than in either compartment due to conversion to kinetic energy.
This lower pressure also exists in the regions of flow acceleration and deceleration where potential
and kinetic energy interchange; however, the total energy (potential + kinetic) at each point along
the system is constant
Figure 1.16 also displays the total internal energy in the system, that is sum of
potential and kinetic energy. This is depicted as constant though in reality falls
slightly across the system due to heat loss from viscous friction. Within the conduit
the velocity (for laminar flow) is also constant and does not vary from one section
to another without cross section variation. The fluid in the tube acquires momentum
in the zone of flow acceleration. This momentum (kinetic energy) raises the pres-
sure (potential energy) in the receiving container until no further pressure gradient
exists and flow ceases.
This system is described by the Bernoulli equation stating that the internal energy
of a fluid is the sum of potential energy due to its column height, gravity, con-
strained pressure and any kinetic energy [49]:
The constant is the total energy of the system, and this equation simply says that
at each point along the tube conservation of energy will dictate the velocity and
pressure. For tubes of varying cross section, flow must be constant; thus the velocity
will increase as diameter decreases, leading to drops in pressure; however, the total
internal energy will still remain constant such that for any two points along the tube
the sum of momentum and pressure must balance [50]:
ρ × V12 ρ × V22
+ P1 = + P2
2 2 (1.23)
∆P = 1 ρ (V22 − V12 )
2
Resistance to Flow
Equation 1.21 implies that resistance simply alters the ratio between the pressure
gradient and blood flow. However, due to viscous friction, any constrictions to flow
create small turbulent vortices in their wake that create vibration and increased
shear velocities at the vessel wall, this energy being lost as heat and sound. For a
uniform diameter flow path with a fixed resistor, these factors tend to result in a loss
of potential energy (i.e. a pressure drop) across the resistor in addition to any pres-
sure drop from altering velocity.
Laminar flow describes a parabolic speed profile (Fig. 1.16) with a maximum blood
velocity in the centre of the cylinder and a near stationery blood film in contact with
the conduit walls at which biofilms form. Laminar flow is the least traumatic on
formed blood elements and the most energy efficient, being directly proportional to
the pressure gradient and inversely proportional to resistance defined as:
8 × length ×η
Resistance to laminar flow =
π × radius 4
( mmHg.L−1.min −1 ) (1.24)
Combining Eqs. 1.21 and 1.24 and converting radius to diameter, d, yields
Poiseuille’s equation:
48 M. J. Brain et al.
⋅ ( PInlet − POutlet ) × π × d 4
Q= (1.25)
128 × length ×η
The effect of viscosity (η) has been discussed, and while length of the tubing is
important, it is far outweighed by the radius to the fourth power, whereby the flow
rate can be doubled by 20% increase in tube diameter. This becomes important in
selecting vascular access catheter size for a desired flow rate, which can be limited
by vessel size and technical factors relating to insertion.
Turbulent Flow
Reynolds Number
The most important parameter in determining the character and average rate of flow
in a conduit is the ratio of fluid momentum to viscous forces known as Reynolds
number [50] where for a conduit of diameter d and a fluid with a mean velocity v :
ρ ×d ×v
Re = (1.26)
η
At Re numbers less than 100, pure laminar flow is apparent; from 100 to 1000
laminar flow occurs, but an increasingly wide boundary layer appears against the
stationary conduit; above 1000 there is a transition to turbulent flow streams and
eddy currents; Re numbers >10,000 are purely turbulent flow [50]. The linking of Re
to velocity is critical in determining the maximal flow through extracorporeal cir-
cuits and will be discussed below. The distance blood that must flow before laminar
flow can be established is also related to Re and is known as the entrance length: Re
× 0.03 × tube diameter (Fig. 1.16).
1 Physiology of Extracorporeal Life Support (ECLS) 49
A Reynolds number can also be calculated for the resistance (drag) that a particu-
late experiences locally in a viscous fluid. For erythrocytes suspended in plasma, the
cell diameter (dRBC) and a coefficient, ϕ, for cell deformity Replasma is defined as [44]:
The membrane oxygenator, with its complex network of hollow fibres designed to
maximise the surface area available for oxygenation, also significantly increases the
cross-sectional area of the blood path (Fig. 1.17). By conservation of energy and
mass, flow (Q, ml.s−1) through any two points of the extracorporeal circuit must be
constant, and flow can be described as the product of blood velocity, V (cm.s−1), and
cross-sectional area, CSA (cm2):
Q1 = Q2
(1.28)
CSA1 × V1 = CSA 2 × V2
The cross section of the conduit is calculated as π.radius2 and an ECMO flow of
5 L.min−1 in 1 cm diameter tubing yields a velocity of 1.06 m.s−1. Oxygenator cross-
sectional areas vary, but an approximation can be made by dividing the oxygenator
priming volume by the oxygenator width in centimetre, yielding a volume per cm.
This can be divided by the volume in 1 cm of conduit yielding a ratio of the cross-
sectional areas. A value of 80 would imply the axial velocity of blood within the
oxygenator is around 1.3 cm.s−1 or around 100 times slower than the conduit veloc-
ity, allowing significantly more time for gas exchange.
Oxygenators are characterised by the surface area of their membrane, but there
is no guarantee that the blood flow distributes evenly over this area. In particular, for
hollow fibre oxygenators, the blood must spread across the opening face of the
tubules. Clot deposition over time will gradually decrease the number of paths that
blood can take.
50 M. J. Brain et al.
Fig. 1.17 Schematic of a simplified ECMO circuit with a centrifugal vortex pump, a membrane
oxygenator and conduits between the patient access cannula. Systems consisting of centrifugal
vortex pumps (as opposed to roller pumps) are typically valveless and can flow in either direction
if the pump is off. The pressure drop across the oxygenator is depicted as the differing heights of
the blood columns. Cross-sectional area of the conduit and oxygenator (red areas) is shown
Figure 1.18 depicts total energy exchange across an extracorporeal circuit running
at steady state. As in Fig. 1.16, the total energy of blood is made up of potential
energy (pressure) and kinetic energy (velocity); however, energy loss to the system
from viscous friction (most notably across the oxygenator) and addition of energy
by the pump are now depicted. As we are now using blood, a full statement for
stable energy transfer must now replace the Bernoulli equation where h denotes
energy added or lost to the system [49, 50]:
ρ × v12 ρ × v22
ρ gh1 + + P1 + hpump + hRBC = ρ gh2 + + P2 + hfriction + hRBC (1.29)
2 2
Here the internal energy of the fluid at the intake comprises the fluid column
height, intake velocity and intake pressure and elastic energy stored in the erythro-
cyte. Energy is added by the pump (hpump), and for forward flow to occur, the sum of
pump energy and intake internal energy must be greater than the sum of energy lost
as viscous friction and the internal energy (fluid pressure and momentum) of the
Fig. 1.18 Energy exchange across the extracorporeal circuit; after [51]. Refer to Fig. 1.16 for
discussion of energy transfer from pressure to velocity. The total internal energy of the fluid is
depicted with sequential losses of heat generated from viscous friction at constriction points. Thus
the total energy falls at all points other than the pump, most prominently in the low flow region of
the oxygenator. Narrow spacing between laminar flow lines represents high velocity at low pressure
52 M. J. Brain et al.
receiving vessel. Gravity and the fluid column height are now important, as the
suspended red cells have mass and will sink to the bottom of a stationary container.
It should be noted that ‘lost’ energy is contained as heat [50].
At start-up, the high velocity vanes of the rotating axial flow pump add kinetic
energy to the column of blood. This kinetic energy progressively accelerates the
blood column until a Reynolds number is reached where the amount of turbulent
flow is sufficient for energy losses from viscous friction to balance the left and right
side of Eq. 1.29 and a constant velocity results. This does not mean turbulent flow
will exist throughout the system; rather the energy loss from small areas of turbu-
lence will summate to oppose the applied energy of the rotor.
As the conduits before and after the pump are of equal cross-sectional area and
all blood must come via the pump, the pre- and post-pump velocity is constant (con-
servation of mass). Instead, the applied kinetic energy is converted to pressure
against the pump housing, against the post-pump conduit walls and against any
resistance (the oxygenator) or pressure load downstream (Fig. 1.18).
The power (watts) required of the pump is related to the required force exerted
on the blood (torque) and the rotation speed (revolutions per second, converted from
rpm by dividing by 60).
rpm
P = torque × 2π × (1.30)
60
This effect of occlusion increasing the rpm in velocity pumps is in contrast to the
effect on a volume displacement (roller) pump. If a roller pump continues against an
occlusion, then its continual volume displacement will steadily increase the pres-
sure in the distal conduit until mechanical failure occurs, that is the work the roller
pump performs on the blood increases with occlusion.
Flow Regurgitation
It is apparent from Eq. 1.29 that a high opposing blood velocity or pressure at the
pump outflow opposes forward flow and the energy imparted by the pump must be
sufficient to overcome this. This is relevant in pulsatile circulations where transient
peak pressures in the receiving vessel may exceed the energy imparted by the pump,
causing not only flow deceleration but occasionally flow reversal. In this situation,
the viscous resistance to circuit flow is now being overcome by the driving pressure
beyond the pump, and the energy imparted by the rotors will again result in viscous
friction at the blades from extremely turbulent flow (Eqs. 1.26 and 1.27).
Cavitation
to three things: adequacy of overall adequacy cardiac output, right ventricular func-
tion and interaction with pulmonary haemodynamics, and adequacy of venous
access (above).
Fig. 1.19 Images of thrombus within a pump head and oxygenator before and after removal of
circuit from patient and washing out blood. (Image M. Brain. Launceston General Hospital)
56 M. J. Brain et al.
Surface Activation
XIIa
High Molecular Weight Kininogen
Prekallikrein
(+)
Tissue Factor
XI XIa VIIa
Ca2+
Hep:AT (-)
IX IXa
Degranulation
(+)
VIII VIIIa
Phospholipids
Ca2+ (+)
V
X Xa Va Platelets Aggregation
Hep:AT (-)
LMWH
Argatroban
Clot
Phospholipids (-)
(+)
Ca 2+
Ca2+
XIII
Prothrombin (II) Thrombin (IIa) XIIIa
Ca2+
Argatroban Hep:AT (-)
(-)
Fig. 1.20 The coagulation cascade and inhibition by common anticoagulants (red). Black arrows
depict major steps where zymogens are converted to active clotting factors, and major potentiators
are listed above the steps in black. The positive feedback loops amplifying thrombin production are
depicted in grey. Both fibrinogen and fibrin molecules can bind with the GPIIb/IIIa receptors on
two adjacent platelets allowing aggregation. Hep:AT denotes the major inhibitory actions of
unfractionated heparin when bound with antithrombin. LMWH low molecular weight heparin
on the pathogenesis of fibrosis and reparative properties of the lung in both normal
aging, acute inflammatory and chronic pathologies.
Beginning at 4–5 weeks gestation as two primary lung buds of endodermal epi-
thelium within a mesoderm and vascular plexus [62], later the tips of the elongating
branches differentiate to alveolar type I and II cells. Development of alveoli in
humans is maximal in the first 2–3 years of life but continues into adolescence
[63–65]. ARDS is initially characterised by damage and increased permeability of
the pulmonary endothelium and alveolar epithelium resulting in increased fluid,
proteins, neutrophils and red blood cells first in the lung interstitium and then the
alveolar spaces. This is followed by fibrin deposition and alveolar type II cell hyper-
plasia which later differentiate into the gas exchanging type I cells that line normal
alveoli [66].
Understanding of the interaction between underlying stroma, alveolar type I pro-
genitor cells, collagen deposition from fibroblasts, and inflammatory cells in restor-
ing functional alveoli for gas exchange is still evolving. Distinct regions of the lung
contain populations of epithelial cells that function as adult progenitor cells, with
58 M. J. Brain et al.
Conclusions
References
1. Kim G-B, Kim S-J, Kim M-H, Hong C-U, Kang H-S. Development of a hollow fiber mem-
brane module for using implantable artificial lung. J Membr Sci. 2009;326(1):130–6.
2. MacLaren G, Brain MJ, Butt WW. ECMO in acute and chronic adult respiratory failure: recent
trends and future directions. Minerva Anestesiol. 2013;79(9):1059–65.
3. Scheepers A, Joost H-G, Schürmann A. The glucose transporter families SGLT and
GLUT: molecular basis of normal and aberrant function. JPEN J Parenter Enteral Nutr.
2004;28(5):364–71.
4. Stryer L. Biochemistry. 4th ed. WH Freeman and Company; 1995.
5. Beals M, Gross L, Harrell S. Efficiency of ATP production [Internet]. Metabolism for energy
and the respiratory quotient. 1999 [cited 2012 Aug 15]. Available from: http://www.tiem.utk.
edu/~gross/bioed/webmodules/respiratoryquotient.html.
6. Elia M, Livesey G. Theory and validity of indirect calorimetry during net lipid synthesis. Am
J Clin Nutr. 1988;47(4):591–607.
7. Wolfe RR, Allsop JR, Burke JF. Glucose metabolism in man: responses to intravenous glucose
infusion. Metabolism. 1979;28(3):210–20.
8. Hellerstein MK, et al. De novo lipogenesis in humans: metabolic and regulatory aspects. Eur J
Clin Nutr. 1999;53:53–65.
9. Mulquiney PJ, Kuchel PW. Control analysis of 2,3-bisphosphoglycerate metabolism [Internet].
Biochem J. 1999 [cited 2012 Oct 11];342:597–604. Available from: http://www.biochemj.org/
bj/342/bj3420597.htm
10. Mulquiney PJ, Bubb WA, Kuchel PW. In situ kinetic characterization of 2,3-bisphosphoglycerate
synthase/phosphatase [Internet]. Biochem. J. 1999 [cited 2012 Oct 11];342:567–580. Available
from: http://www.biochemj.org/bj/342/bj3420567.htm.
11. Drummond M, Braile DM, Lima-Oliveira AP, Camim AS, Oyama RS, Sandoval
GH. Technological evolution of membrane oxygenators. Braz J Cardiovasc Surg.
2005;20(4):432–7.
12. Lawson DS, Holt D. Insensible water loss from the Jostra Quadrox D oxygenator: an in vitro
study. Perfusion. 2007;22(6):407–10.
13. Peter S. Strong ions plus carbon dioxide. In: Kellum JA, Elbers PW, editors. Stewart’s Textbook
of Acid-Base. 2nd Edition. Lulu Enterprises; 2009. p. 111–32.
1 Physiology of Extracorporeal Life Support (ECLS) 59
14. Anaesthsia Correspondence Web Site [Internet]. [cited 2012 Sep 17]. Available from: http://
www.anaesthesiacorrespondence.net/Correspond3.asp?articleid=3598&archive=1.
15. Lumb AB. Nunn’s applied respiratory physiology. 5th ed. Oxford: Butterworth-Heinemann.
16. Sargent JA. Principles and biophysics of dialysis. In: Jacobs C, editor. Replacement of renal
function by dialysis. Springer; 1996. p. 34–145.
17. Cussler EL. Diffusion: mass transfer in fluid systems. Cambridge University Press; 1997. 606 p.
18. Grathwohl P. Diffusion in natural porous media: contaminant transport, sorption/desorption
and dissolution kinetics. Kluwer Academic Publishers; 1998. 230 p.
19. Katoh S, Yoshida F. Rates of absorption of oxygen into blood under turbulent conditions.
Chem Eng J. 1972;3:276–85.
20. Matsuda N, Sakai K. Blood flow and oxygen transfer rate of an outside blood flow membrane
oxygenator. J Membr Sci. 2000;170:153–8.
21. Weibel ER. The pathway for oxygen: structure and function in the mammalian respiratory
system. Harvard University Press; 1984. 448 p.
22. Nickalls R. Inverse solutions of the Severinghaus and Thomas equations which allow Po2 to
be derived directly from So2 [Internet]. [cited 2012 Sep 3]. Available from: www.nickalls.org/
dick/papers/anes/severinghaus.pdf.
23. Epstein FH, Hsia CCW. Respiratory function of hemoglobin. N Engl J Med.
1998;338(4):239–48.
24. Heaton A, Keegan T, Holme S. In vivo regeneration of red cell 2, 3-diphosphoglycerate
following transfusion of DPG-depleted AS-1, AS-3 and CPDA-1 red cells. Br J Haematol.
1989;71(1):131–6.
25. Brain MJ. VV-ECMO oxygen saturation tool [Internet]. [cited 2014 Nov 4]. Available from:
http://www.irenabyss.com.au/Equipment/VVECMO/.
26. Thomas LJ. Algorithms for selected blood acid-base and blood gas calculations. J Appl
Physiol. 1972;33(1):154–8.
27. Kelman GR. Digital computer subroutine for the conversion of oxygen tension into saturation.
J Appl Physiol. 1966;21(4):1375–6.
28. Arthurs GJ, Sudhakar M. Carbon dioxide transport. Contin Educ Anaesth Crit Care Pain.
2005;5(6):207–10.
29. Mochizuki M. The CO∼ 2 dissociation curve at steady state in vivo. Yamagata Med
J. 2004;22(1):25–8.
30. Beilin LJ, Knight GJ, Munro-Faure AD, Anderson J. The sodium, potassium, and water con-
tents of red blood cells of healthy human adults. J Clin Investig. 1966;45(11):1817.
31. Mochizuki M. Analysis of bicarbonate concentration in human blood plasma at steady state
in vivo. Yamagata Med J. 2004;22(1):09–24.
32. Mochizuki M, Takiwaki H, Kagawa T, Tazawa H. Derivation of theoretical equations of
the CO2 dissociation curve and the carbamate fraction in the Haldane effect. Jpn J Physiol.
1983;33(4):579–99.
33. Tazawa H, Mochizuki M, Tamura M, Kagawa T. Quantitative analyses of the CO2 disso-
ciation curve of oxygenated blood and the Haldane effect in human blood. Jpn J Physiol.
1983;33(4):601–18.
34. Fencl V, Jabor A, Kazda A, Figge J. Diagnosis of metabolic acid-base disturbances in critically
ill patients. Am J Respir Crit Care Med. 2000;162(6):2246–51.
35. Gutknecht J, Bisson MA, Tosteson FC. Diffusion of carbon dioxide through lipid bilayer
membranes. Effects of carbonic anhydrase, bicarbonate, and unstirred layers. J Gen Physiol.
1977;69(6):779.
36. Centor R. Serum total carbon dioxide - clinical methods - NCBI bookshelf. In: Walker H,
Hall W, Hurst J, editors. Clinical methods: the history, physical, and laboratory examinations
[Internet]. Butterworths; 1990 [cited 2012 Sep 22]. Available from: http://www.ncbi.nlm.nih.
gov/books/NBK308/.
37. Baker A, Richardson D, Craig G. Extracorporeal carbon dioxide removal (ECCO2 R) in respi-
ratory failure: an overview, and where next? 2C04 3A13. [cited 2012 Sep 22]; Available from:
http://journal.ics.ac.uk/pdf/1303232.pdf.
60 M. J. Brain et al.
38. ARDS Network. Ventilation with lower tidal volumes as compared with traditional tidal
volumes for acute lung injury and the acute respiratory distress syndrome. N Engl J Med.
2000;342(18):1301–8.
39. Guérin C, Reignier J, Richard J-C, Beuret P, Gacouin A, Boulain T, et al. Prone positioning in
severe acute respiratory distress syndrome. N Engl J Med. 2013;368(23):2159–68.
40. Abroug F, Ouanes-Besbes L, Dachraoui F, Ouanes I, Brochard L. An updated study-level
meta-analysis of randomised controlled trials on proning in ARDS and acute lung injury. Crit
Care. 2011;15(1):R6.
41. Perier F, Tuffet S, Maraffi T, Alcala G, Victor M, Haudebourg A-F, et al. Effect of positive end-
expiratory pressure and proning on ventilation and perfusion in COVID-19 acute respiratory
distress syndrome. Am J Respir Crit Care Med. 2020;202(12):1713–7.
42. Messerole E, Peine P, Wittkopp S, Marini JJ, Albert RK. The pragmatics of prone positioning.
Am J Respir Crit Care Med. 2002;165(10):1359–63.
43. Cove M. Ultra-protective pulmonary ventilation supported by low flow extracorporeal carbon
dioxide removal (ECCO2R) and prone positioning for ARDS; a pilot study. ClinicalTrials.gov
Identifier: NCT02252094 [Internet]. Home - ClinicalTrials.gov. 2014 [cited 2021 July 23].
Available from: https://clinicaltrials.gov/ct2/home.
44. Jung J, Lyczkowski RW, Panchal CB, Hassanein A. Multiphase hemodynamic simulation of
pulsatile flow in a coronary artery. J Biomech. 2006;39(11):2064–73.
45. Lipowsky H. Microvascular rheology and hemodynamics. Microcirculation. 2005;12(1):5–15.
46. Papaioannou TG, Stefanadis C, et al. Vascular wall shear stress: basic principles and methods.
Hell J Cardiol. 2005;46(1):9–15.
47. Lipowsky HH, Kovalcheck S, Zweifach BW. The distribution of blood rheological parameters
in the microvasculature of cat mesentery. Circ Res. 1978;43(5):738–49.
48. Ercan M, Koksal C. The relationship between shear rate and vessel diameter. Anesth Analg.
2003;96(1):307–8.
49. Walker J. Fundamentals of physics. 9th ed., extended ed. Hoboken: Wiley; 2011. 1248 p.
50. Bahrami M. Introduction to fluid mechanics [Internet]. Simon Fraser University: Lecture
Notes for Introduction to Fluid Mechanics: ENSC283. 2011 [cited 2012 Oct 13]. Available
from: http://www.sfu.ca/~mbahrami/ENSC%20283/Notes/.
51. Bloomfield LA. How everything works: making physics out of the ordinary. Wiley; 2007. 736 p.
52. Zochios V, Parhar K, Tunnicliffe W, Roscoe A, Gao F. The right ventricle in ARDS. Chest.
2017;152(1):181–93.
53. Hoffman KR, Burrell AJC, Diehl A, Butt W. Elevated carboxyhaemoglobin as a novel indicator
for extracorporeal membrane haemolysis and oxygenator exchange. Crit Care. 2021;25(1):159.
54. Burns J, Hurtado-Doce A, Lees N. Carboxyhemoglobin associated with hemolysis as a marker
of impending oxygenator failure in VA ECMO. Crit Care Med. 2015;43:38.
55. Omar HR, Mirsaeidi M, Socias S, Sprenker C, Caldeira C, Camporesi EM, et al. Plasma free
hemoglobin is an independent predictor of mortality among patients on extracorporeal mem-
brane oxygenation support. Frati G, editor. PLoS ONE. 2015;10(4):e0124034.
56. Ghitescu I, Copotoiu SM, Toma RS, Ghitescu VT, Copotoiu R. Mean filter life span in con-
tinuous veno-venous hemofiltration for septic patients. Jurnalul Roman de Anestezie Terapie
Intensiva/Rom J Anaesth Intensive Care. 2009;16(1):17–22.
57. Brain M, Winson E, Roodenburg O, McNeil J. Non anti-coagulant factors associated with filter
life in continuous renal replacement therapy (CRRT): a systematic review and meta-analysis.
BMC Nephrol. 2017;18:69.
58. Jackson SP. The growing complexity of platelet aggregation. Blood. 2007;109(12):5087–95.
59. GlaxoSmithKline. Argatroban injection prescribing information [Internet]. 2008 [cited
2021 July 23]. Available from: https://www.accessdata.fda.gov/drugsatfda_docs/
label/2008/020883s014lbl.pdf.
60. Ruby K, Salvatore L, Lomachinsky R, Ornstein D, Khan J. Hemolysis interference in the anti-
Xa heparin activity assay - ISTH congress abstracts. Res Pract Thromb Haemost [Internet].
1 Physiology of Extracorporeal Life Support (ECLS) 61
2021 [cited 2021 July 23];5 (Suppl 1). Available from: https://abstracts.isth.org/abstract/
hemolysis-interference-in-the-anti-xa-heparin-activity-assay/.
61. Khan J, Chandler WL. Interference in the anti-Xa heparin activity assay due to hemolysis and
icterus during pediatric extracorporeal life support. Artif Organs. 2019;43(9):880–7.
62. Hogan BLM, Barkauskas CE, Chapman HA, Epstein JA, Jain R, Hsia CCW, et al. Repair and
regeneration of the respiratory system: complexity, plasticity, and mechanisms of lung stem
cell function. Cell Stem Cell. 2014;15(2):123–38.
63. Dunnill MS. Postnatal growth of the lung. Thorax. 1962;17(4):329–33.
64. Narayanan M, Owers-Bradley J, Beardsmore CS, Mada M, Ball I, Garipov R, et al.
Alveolarization continues during childhood and adolescence: new evidence from helium-3
magnetic resonance. Am J Respir Crit Care Med. 2012;185(2):186–91.
65. Thurlbeck WM. Postnatal human lung growth. Thorax. 1982;37(8):564–71.
66. Matthay MA, Zemans RL, Zimmerman GA, Arabi YM, Beitler JR, Mercat A, et al. Acute
respiratory distress syndrome. Nat Rev Dis Primers. 2019;5(1):18.
67. Mora AL, Rojas M. Adult stem cells for chronic lung diseases: B-MSCs aging and lung repair.
Respirology. 2013;18(7):1041–6.
Chapter 2
Circuits, Membranes, and Pumps
Bradley H. Rosen
Introduction
Circuit Anatomy
Circuit designs all attempt to balance efficacy, safety, convenience, and simplicity.
There is no one-size-fits-all solution, however, since varied patients, circumstances,
and clinician preferences may necessitate that safety override simplicity or that
B. H. Rosen (*)
Division of Pulmonary, Critical Care, and Occupational Medicine, Department of Internal
Medicine, Roy J. and Lucille A. Carver College of Medicine, University of Iowa Hospitals
and Clinics, Iowa City, IA, USA
e-mail: [email protected]
portability trump efficacy. For example, inserting multiple stopcocks into the circuit
can allow easy access for renal replacement therapy: one circuit carries out gas
exchange and dialysis. This is convenient, but each additional connector represents
an opportunity for failure (leak, thrombosis, air entrainment, rupture). In an indi-
vidual circumstance, whether to conduct renal replacement using the ECLS circuit
may depend on the ease of obtaining alternate venous access, the expected duration
of renal failure, or the ECLS physician’s experience and preference with regard to
circuit complexity. Thus, circuits range from rather complex designs incorporating
many safety and monitoring functions (see Fig. 2.1) to minimalistic, simpler layouts
that lack the various bells and whistles (Fig. 2.2).
Fig. 2.1 Schematic of a complex circuit design that implements all optional features, including
separate non-integrated blood analyzers, a manifold with access sites (closed unless being
accessed), and renal replacement circuit integration options. The dotted line indicates where the
manifold may be inserted between the pump and the oxygenator, if there is sufficient tubing to
allow for this
Fig. 2.2 Similar to the previous Fig. 2.1 but simplified with only necessary components: pump,
oxygenator, and flow probe/bubble sensor. The blood analyzers are internalized within the pump
and oxygenator. Renal replacement circuit is included here along the manifold, which derives post-
oxygenator and returns pre-pump. Some centers do not include a manifold, simplifying the circuit
even further
2 Circuits, Membranes, and Pumps 65
Circuit Priming
Priming refers to the process by which the gas (ambient air present at manufacture)
is replaced with a physiologically compatible fluid. For ECLS in adults, the circuit
is primed with crystalloid fluids, such as normal saline, Ringer’s lactate, or propri-
etary mixed electrolyte solutions (e.g., Plasma-Lyte® or Normosol®). Purchased
circuits come attached to a large, empty priming bag. The bag is filled with sterile
crystalloid, clamps are opened to the venous and arterial limbs, and the priming bag
is raised to allow gravity to move the fluid into the circuit components while air
moves to the priming bag. The volume necessary to prime a given circuit depends
on the priming volume of each component (oxygenator, heat exchanger blood
phase, pump, manifold, and tubing) and directly relates to the degree of hemodilu-
tion that follows. For pediatric and neonatal ECLS, hemodilution is prevented by
priming the circuit with blood, but this is not necessary for adults where the typical
priming volume averages 500–1000 mL for a complex circuit design and for a sim-
ple one as low as 300 mL. In fully primed condition, a circuit can be stored for a
period of at least 30 days, although each institution has its own policies regarding
shelf life. Priming with a colloid may shorten the shelf life of a primed circuit,
another reason many programs choose a crystalloid prime. A simplified ECLS cir-
cuit can be fully primed in less than 10 min due to the microporous nature of the
membranes in use. Once the circuit is primed, the heat exchanger can be turned on
to raise the temperature to 37 °C before connecting the patient, as long as time per-
mits. Cardiopulmonary bypass circuits are often primed first with carbon dioxide
(to displace oxygen and nitrogen), hastening the subsequent fluid priming process,
but this is not generally done for ECLS circuits.
ECLS circuits can appear intimidating, especially when one realizes that 5 L of
blood can rush through it each minute. A systematic approach to the intricacies of
the circuit and its components keeps the clinician from becoming overwhelmed, so
we begin with a brief, general tour. Figure 2.1 represents a comprehensive sche-
matic of a circuit, whereas Fig. 2.2 shows a greatly simplified design with few
extraneous components. In each instance, we describe the circuit beginning with the
drainage (outflow) cannula, proceeding through the gas-exchanging membrane, and
ending back at the patient through the return (inflow) cannula. Dual-lumen cannulas
allow blood to exit and enter at the same site but; for illustration purposes, we’ve
separated these in the figures.
Starting with the drainage cannula at its exit from the patient (internal jugular or
femoral vein, or right atrium), the distal end is connected to large-diameter conduct-
ing tubing. This is an important point for inadvertent disconnection, especially
immediately following the initiation of ECLS if the tie bands were not securely
2 Circuits, Membranes, and Pumps 67
fastened. In addition, like other areas of the circuit where there is turbulence or
stasis, this is a common site for thrombus to form. Careful examination of this con-
nection is an essential part of the regular circuit check (see Chap. 12). The conduct-
ing tubing should be kept relatively short in order to reduce resistance to blood flow,
surface area of contact with blood, and the priming volume. The conducting tubing
leads to a centrifugal pump before entering the membrane oxygenator. As there is
considerable heat loss as the blood traverses the circuit, a heat exchanger is neces-
sary to rewarm the blood to body temperature (this may be incorporated into the
oxygenator and hidden from direct view). If added separately, the heat exchanger is
placed proximal to the oxygenator. The oxygenator also receives the sweep gas
(usually medical oxygen), being joined to wall oxygen through a blender or to an
E-cylinder through a flow meter. The newly arterialized blood completes its extra-
corporeal course through the return cannula, delivering oxygenated and warmed
blood to the vascular system.
While a bridge between the outflow and inflow cannulas is often used as a shunt
to conduct weaning trials during veno-arterial (VA) ECLS, such a bridge is not
needed for veno-venous (VV) support since weaning can be conducted by reducing
or eliminating gas flow to the membrane while leaving circuit flow to the patient
undisturbed.
In order to monitor circuit function and prevent complications, devices to mea-
sure pressure and flow and to detect bubbles are included, and information is relayed
to a console (see Fig. 2.3 for an example). Typically, fluid pressure is measured on
the venous side of the pump (“PINLET”), providing information about how much suc-
tion is required to draw the needed circuit blood flow. Two additional pressure trans-
ducers (“PPRE” and “PPOST”) flank the oxygenator so that the membrane pressure
drop (ΔP) can be calculated. In addition, PPOST displays the pressure that drives flow
back to the patient. Circuit blood flow is monitored using an ultrasonic flow probe,
since centrifugal pumps do not guarantee a fixed relationship between revolutions
per minute and volume displaced, as was true for roller pumps. The flow probe may
be integrated within the pump or added as an aftermarket device. Ultrasound probes
are also used to identify bubbles, so some circuit designs utilize the same sensor for
both flow measurement and bubble detection. Bubbles distal to the oxygenator can
produce systemic embolism and are especially dangerous in VA modes but less so
in VV as the patient’s lungs will provide protection from air embolism.
Spectrophotometric sensors allow real-time measurement of such values as PO2,
PCO2, pH, inlet saturation (SPREO2), outlet saturation (SPOSTO2), and hemoglobin
concentration, among others. These sensors must be calibrated periodically by com-
paring the displayed value against a blood sample analyzed simultaneously using
conventional laboratory methods. The console receives data from various devices
along the circuit, displaying pump speed, flow, pressures, temperature, and other
physiological information. The console also may display alarm notifications and
allows the user to adjust the pump, heat exchanger, and other functions. The console
generally is integrated with the power supply and battery.
Ports are included in the circuit so that blood can be sampled and agents can
be infused. These are often collected in a manifold consisting of a series of Luer
68 B. H. Rosen
Fig. 2.3 ECLS console showing relevant pressures within the circuit (PINLET, PRE, POST, same as in
Figs. 2.1 and 2.2) along with pump speed (rpm) and flow (L/min) along the top of the screen, and
SPREO2 in the lower right of the screen with a battery charge indicator in the corner. The dial and
buttons along the right of the panel are used to adjust parameters, in addition to the touch screen
itself. At the top of the figure is a colored bar that can be used for rapid visualization of the
pump speed
lock connections with a three-way stopcock controlling flow to each. The mani-
fold derives from a region either after the oxygenator or between the centrifugal
pump and the oxygenator (a “safe zone” of interruption) and re-infuses proximal
to the pump so that small amounts of entrained air can be eliminated by the oxy-
genator. With more compact systems, the pump and oxygenator are often adja-
cent without tubing or connectors between the devices, and the manifold will
generally originate post-oxygenator and return pre-pump. Various infusions of
medications and anticoagulant agents may be connected to the circuit through
these ports, as sufficient flow can be drawn from the circuit so as to combine
renal replacement therapy (RRT) and gas exchange simultaneously, avoiding the
need for invasive vascular access solely for dialysis. Integration of RRT is often
through the manifold, as shown in Figs. 2.1 and 2.2, although there are variations
that may involve keeping the system completely pre-pump or interface directly
from ports on the oxygenator itself. Any ports along the circuit tubing itself tend
2 Circuits, Membranes, and Pumps 69
to begin as pigtails that end in a stopcock, since breakage of this can be clamped,
while integrated Luer locks that break cannot be managed without a circuit
exchange.
Single- and double-lumen cannulas are described more fully in Chap. 4. Cannulas
tend to be wire-reinforced to limit kinking and occlusion. They are attached to the
circuit tubing by means of adaptors, and these connections are secured by tie bands.
Tubing is clear, medical grade polyvinylchloride (PVC) allowing the clinician to
recognize blood color (as a clue to circuit function and recirculation) and to identify
fibrin, clots, and gas bubbles. Tubing can be clamped and, when changing out a
circuit due to oxygenator failure for instance, cut and reconnected to reinstitute
circuit flow.
Centrifugal Pumps
There are two types of pumps that have been employed in ECLS circuits: roller/
occlusive and centrifugal pumps. Practitioners of modern ECLS have settled on the
centrifugal pump design as the safer of the two, and this is the type that will be dis-
cussed here.
These devices increasingly employ a magnetically driven impeller within a spiral
housing. While many impellers are completely levitated by magnets, some have a
spindle to which fibrin can deposit, and this can contribute to hemolysis. The impel-
ler imparts mechanical energy to the blood, raising velocity and pressure as it moves
from the center of the vortex to the periphery. The housing constrains and directs the
blood flow toward the circumference where it exits the pump. This principle differs
entirely from that employed in roller head pumps, producing several advantages and
compromises. Where roller head pump flow is independent of afterload (to the point
of causing tubing rupture!), a centrifugal pump is unable to overcome excessive
afterload. Instead, flow will drop as afterload increases, despite an unchanged pump
rotation speed. In a similar vein, a centrifugal device is unlikely to cause cavitation
at the outflow cannula, as it is unable to generate sufficiently negative pressure. Of
course, this means that hypovolemia tends to threaten the adequacy of circuit flow
when a centrifugal pump is used. This lack of absolute relationship between pump
revolutions and blood flow necessitates a flow meter. An increasing discrepancy
between pump speed and measured flow is a strong signal of trouble with regard to
function of the circuit.
70 B. H. Rosen
In the case of power loss, many of these pumps can be hand-cranked. Internal
batteries can provide up to a couple hours of support without an external power
source, depending on the system and pump speeds needed. Additional discussion of
ECLS crises can be found in Chap. 13.
Advantages of centrifugal pumps over roller pumps include reduced tubing
length, safety benefits due to absent issues of spallation (i.e., liberation of micro-
scopic particles of tubing into the circulation), microemboli, raceway rupture, and
greater flexibility in circuit design because the pump does not need to be at the base
of the unit. Disadvantages are fewer, including direct priming volume required for
the pump (although tubing length is reduced) and blood flow that depends on pre-
load and afterload (requiring a blood flow probe). Centrifugal pumps were initially
reported to produce unacceptable rates of hemolysis due to heat generation, but
engineering improvements have solved this problem [5]. Designs have evolved to
reduce hemolysis and the risk of gaseous microembolism [6, 7], changing the land-
scape of ECLS circuits and leading to worldwide adoption of the technology [8, 9].
Membrane Oxygenators
The ideal membrane lung would be highly permeable to relevant gases (O2 and
CO2) while resisting fluid transudation from the blood to the gas phase (termed
“plasma leak”). Blood should flow through the device with minimal resistance,
allowing high flows with little pressure and without trauma to blood elements.
Surfaces exposed to blood would only minimally activate the host coagulation and
immune systems. These properties would be complemented by reliability, durabil-
ity, and a small priming volume. The human lung juxtaposes blood and gas over a
tremendous surface area, with incredibly thin diffusion distances, yet maintains a
clear separation between blood and gas phases. Scientists and engineers have strug-
gled to mimic these attributes: the history of ECLS is a remarkable story of inspira-
tion, invention, and persistence.
Both PMP and polypropylene hollow-fiber membranes employ large numbers of
fine capillary tubes to carry the sweep gas while being bathed by flowing blood.
This extra-capillary blood flows countercurrent to the direction of gas movement,
increasing the efficiency of gas exchange. It is important to realize the stark contrast
in surface area when comparing an oxygenator with the human lung it attempts to
replace: Most PMP devices provide at most two square meters of gas exchange
surface, while the lung exposes upward of 70 square meters to blood flow.
Additionally, in the best membrane lungs, oxygen must diffuse 150 μm from sweep
gas to blood, whereas the comparable distance within the human lung is a mere
0.5 μm. These disadvantages are offset by increasing the effective dwell time of the
blood within the artificial lung. In addition, so-called “secondary flows,” which
describe the mixing of blood around the gas-fluid interface due to purposefully cre-
ated turbulence, further enhance gas transfer. Blood cells are regularly being brought
into close approximation with the gas-filled capillaries, effectively reducing
2 Circuits, Membranes, and Pumps 71
400
O2 Transfer (mL/min)
300
200
100
0
0 2 4 6 8
Blood Flow (L/min)
diffusing distance [10, 11]. This layout allows for up to a two and a half-fold reduc-
tion in surface area necessary for gas exchange [12]. Typical gas-exchanging capac-
ities for membrane lungs are shown in Figs. 2.4 (oxygen) and 2.5 (carbon dioxide).
Providing sufficient oxygen transfer to meet the entire metabolic demand (roughly
250 mL O2/min) requires blood flow of roughly 4 L/min. In contrast, carbon dioxide
transfer is relatively advantaged so that much less blood flow is required, especially
at very high sweep gas flow rates. For example, nearly all of the metabolically pro-
duced carbon dioxide can be eliminated with only 1 L/min of blood flow, especially
at high gas flows (see Chap. 10) [13].
The most common external appearance of PMP oxygenators is an extruded
square evenly balanced on one corner (see Fig. 2.6). At the lower corner, blood
enters the device from the pump under pressure denoted as “pre-membrane pres-
sure” (PPRE); typical pressures are in the range of 225–275 mmHg, certainly less
than 400 mmHg; (see Table 2.1). The blood ascends to the opposite corner at the
highest elevation of the device and then flows down to the exit connector directly
opposite the inlet, by which point it is oxygenated and carbon dioxide has been
removed. At that point, the pressure (“post-membrane pressure,” PPOST) will be less
than PPRE. The difference between these is called the “ΔP”, representing the resis-
tive pressure drop across the membrane at the current flow. With the current genera-
tion of PMP devices, ΔP should range from the teens to low thirties at typical circuit
blood flow rates (see also Chap. 9).
With regard to the pressures and the information that may be gleaned from their
trends, an elevated PPRE has variable implications depending on whether PPOST is
also elevated. In the case where both are elevated (ΔP is preserved), the circuit
should be examined, not the oxygenator: the distal tubing and circuit may be
obstructed by kinks or thrombosis. Similar findings are seen transiently when
patients cough, Valsalva, or are suctioned. When PPRE rises along with an increase in
ΔP, resistance within the membrane is excessive, raising concerns for thrombosis,
heparin-induced thrombocytopenia, or accumulation of fibrin or cellular elements
72 B. H. Rosen
CO2 Transfer
500
CO2 2:1
400 CO2 1:1
CO2 Transfer (mL/min)
CO2 0.5:1
300
200
100
0
0 2 4 6 8
Blood Flow (L/min)
Fig. 2.5 Carbon dioxide transfer in mL/min as functions of both blood flow in L/min and sweep
gas flow
on the membrane as viewed through the transparent sides of the device. An increas-
ing ΔP along with evidence of impaired gas exchange, which would show as
reduced PO2 and elevated PCO2 from a post-oxygenator blood sample, at a given
flow and sweep portends failure of the membrane (see Chap. 9).
Another major advance in modern PMP oxygenators is their low resistance to
blood flow, producing several important advantages. First, this property ushered in
the era of lower pressure, afterload-sensitive centrifugal pumps, affording the safety
features discussed above. Secondly, lower pressures translate to safer, longer-lived
circuits, conferring additional safeguards against catastrophic rupture or circuit fail-
ure. Additionally, such low resistance to blood flow permits novel applications of
ECLS such as pumpless arteriovenous extracorporeal CO2 removal (AV ECCO2R)
that rely solely on the difference between arterial and venous blood pressures to
drive flow [14] (see Chap. 3).
In the modern era of ECLS, the centerpiece of the circuit is a PMP oxygenator,
and these devices have eclipsed prior generations of artificial lung. Nevertheless,
other oxygenator designs are seen occasionally and include both silicone mem-
branes first developed by Kolobow [15] (most recently marketed as the Medtronic
1–4500-A2) as well as the polypropylene microporous hollow-fiber membrane.
Silicone membrane oxygenators employed sheets enclosing a plastic polymer
screen and wrapped around a polycarbonate core. They remain the only gas
exchange device that is FDA-approved for long-term use (defined as use for more
than 6 h). However, they require priming with considerable volume (665 mL each
[16]), exhibit a large pressure drop across the membrane that limits the use of cen-
trifugal pumps, and are relatively inefficient in gas exchange so that at least two
large surface area units are needed per patient. Polypropylene hollow-fiber mem-
branes are highly efficient with respect to gas exchange but tend to develop plasma
leak (described further below). They also present a low resistance to blood flow,
need only a small priming volume, and remain in use for cardiopulmonary bypass
where they provide excellent short-term support.
Rated Flow: Blood exiting the membrane is normally fully oxygenated, typically
with a PO2 in excess of 300 mmHg. As blood flow is increased, greater demands are
placed on the capacity for gas diffusion across the hollow fiber barrier. In part, this
74 B. H. Rosen
relates to the simple volume of oxygen that must diffuse as more deoxygenated
blood is pushed through the membrane but also to the increasing blood velocity
(thus reduced dwell time) produced at higher flows. At sufficiently high flows, the
membrane fails to fully saturate the blood. The flow threshold for full oxygenation
is termed the “rated flow.”
Plasma Leak: Hollow fiber membranes should be sufficiently permeable to allow
rapid gas diffusion while remaining impermeable to fluid movement. Plasma leak is
the phenomenon whereby plasma phospholipids leak from the circulating whole
blood to the gas compartment of the oxygenator and then serve to propagate further
plasma leakage in a positive feedback cycle [17]. Small to moderate amounts of
clear fluid may normally traverse the membrane, particularly with lower sweep gas
flows, and a drainage port is provided for egress of this condensation. Excessive
condensation will reduce oxygenator efficiency, which will lower post-oxygenator
PO2, and can be corrected when the clinician increases the sweep (>10 L/min) and
jostles the membrane by knocking on the housing to “cough” or “burp” the mem-
brane. Greater volumes of heme-tinged or proteinaceous-appearing fluid emanating
from the drainage port signal a failing membrane, severely impairing the efficiency
of gas exchange, accompanied by a rise in ΔP, and eventually requiring exchange
of the device. Clinically significant plasma leak can be confirmed by analysis of
liquid from the gas outlet for proteins [18]. Additional implications are a significant
loss of proteins and immunoglobulins, potentially significant from immunologic
and nutritional standpoints. PMP membranes are much less susceptible to plasma
leak than prior artificial lungs, yet they are not fully immune to this complica-
tion [19].
Acute kidney injury (AKI) is highly prevalent in those undergoing ECLS, affecting
up to 70% of patients in some case series [20, 21]. The resulting fluid overload is a
frequent indication for renal replacement therapy (RRT) as it will impede native
lung function and hinder efforts at animation of the patient during ECLS. The initial
management of non-oliguric renal failure can be accomplished by loop diuretics—
continuous infusions are preferred to avoid rapid changes in volume status—but
more precise management of volume status and metabolic derangements can be
achieved with continuous renal replacement therapy (CRRT). In addition to volume
management, CRRT may be used to avoid both profound metabolic acidosis, which
may increase respiratory drive such that it defeats the goal of ECLS to reduce pul-
monary injury, and reduced renal clearance of medications or uremia that can con-
tribute to acute encephalopathy. Accordingly, the decision regarding timing of
CRRT initiation should be individualized as with any critically ill patient and is
discussed in Chap. 12.
If already in place at the time of ECLS, standard dialysis catheters may be used
for CRRT as per usual protocols; however, given that many patients develop AKI
2 Circuits, Membranes, and Pumps 75
during their ECLS run, integration of CRRT into the circuit is common. There is
anecdotal evidence that use of the ECLS circuit for CRRT access can extend the
life of dialysis membranes and by reducing patient lines can facilitate ambulation
while on ECLS support. Disadvantages of integrating RRT systems into the ECLS
circuit include the need for additional connections and thus sites of air entrain-
ment, blood loss, and infection, along with risk of losing both circuits if one is
compromised. While there are numerous ways to integrate the CRRT, the return
should enter before the oxygenator to allow the membrane to eliminate any
entrained air.
The most common method of CRRT integration (see Fig. 2.1) is to have the
intake enter the CRRT post-pump, either from the arterial side post-membrane or
from between the pump and oxygenator (dotted line in Fig. 2.1). While it is possible
to have the positive pressure from the ECLS pump drive flow through the dialysis
filter on its own, such passive configurations are uncommon as the clinician loses
independent control of dialysis when circuit flows are adjusted for respiratory sup-
port. Instead, it is more common that standard CRRT systems are implemented in a
parallel circuit that will introduce a small shunt (generally <10% of flow runs
through CRRT), although pressure alarms frequently need to be adjusted to allow
for a much greater positive pressure than expected in normal CRRT runs.
Additionally, the outflow from the CRRT circuit will increase the PINLET, sometimes
enough that the pressure becomes positive. So long as the clinical staff are aware of
the small shunt fraction and adjust pressure alarms, this tends to be well-tolerated
by the other circuit components.
Anticoagulation used for ECLS is sufficient for CRRT; although if specific rea-
sons preclude anticoagulant use (e.g., peri-procedural hemostasis or pulmonary-
renal syndromes, etc.), citrate can be run within the dialysis circuit alone. The
greater flows and degree of anticoagulation used in ECLS tend to allow for pro-
longed CRRT circuit longevity, with some centers being able to run up to 7–10 days
between changing dialysis filters.
While on ECLS, the patient’s blood is exposed to the environment of the outside
world, risking substantial heat loss and even hypothermia. Left unchecked, this
could produce coagulopathy, circulatory failure, and other organ dysfunction. The
heat exchanger consists of an external electric heater-cooler which pumps
temperature-controlled water through tubing to the heat transfer unit. The heat
transfer unit separates the water phase from the circulating blood by a temperature-
conductive material comprised of stainless steel, aluminum, or polypropylene. This
material is often coated with polymers or other surface coatings to abrogate blood
component activation that might stimulate inflammation or clotting. The blood and
water phases are directed in a countercurrent fashion to maximize efficiency of heat
transfer. While the trend is toward smaller units, larger heat exchangers are more
effective in regulating temperature but require larger priming volumes that contrib-
ute to hemodilution.
Heat exchangers may be integrated into the membrane lung or added as a stand-
alone device. After blood is fully saturated with oxygen, warming carries the poten-
tial to release microbubbles. Thus, the heat exchanger is often placed proximal to
the oxygenator, which can then trap any oxygen released.
When circuit blood flow is sufficiently high, the heat exchanger can be effective
in stabilizing body temperature. Accordingly, the clinician should be aware that
fever can be masked and should thus pay close attention to other indicators of infec-
tion. This function of the ECLS circuit can be used to prevent hyperthermia in
patients following cardiac arrest who are undergoing targeted temperature
2 Circuits, Membranes, and Pumps 77
management. However, with smaller devices and lower flows, the efficiency can be
reduced considerably, such that patients can not only mount a fever, but even raise
the water bath above the set temperature.
Complications related to the heat exchanger are few. Leakage of water into the
blood phase will result in hemolysis, so this possibility should be included in the
differential diagnosis whenever an ECLS patient has unexplained hemolysis.
Finally, the water bath of the heat exchanger can serve as a reservoir for microbes.
Water should be exchanged on a regular basis, and some recommend surveillance
cultures to detect potentially pathogenic organisms.
Conclusions
References
1. Khoshbin E, Dux AEW, Killer H, Sosnowski AW, Firmin RK, Peek GJ. A comparison of
radiographic signs of pulmonary inflammation during ECMO between silicon and poly-
methyl pentene oxygenators. Perfusion. 2007;22:15–22.
2. Peek GJ, Firmin RK. The inflammatory and coagulative response to prolonged extracorporeal
membrane oxygenation. ASAIO J. 1999;45(4):250–63.
3. Lehle K, Philipp A, Gleich O, et al. Efficiency in extracorporeal membrane oxygenation-
cellular deposits on polymethylpentene membranes increase resistance to blood flow
and reduce gas exchange capacity. ASAIO J. 2008;54(6):612–7. https://doi.org/10.1097/
MAT.0b013e318186a807.
4. Wildschut ED, Ahsman MJ, Allegaert K, Mathot RAA, Tibboel D. Determinants of drug
absorption in different ECMO circuits. Intensive Care Med. 2010;36:2109–16.
5. Lawson DS, Ing R, Cheifetz IM, et al. Hemolytic characteristics of three commercially
available centrifugal blood pumps. Pediatr Crit Care Med. 2005;6(5):573–7. https://doi.
org/10.1097/01.pcc.0000163282.63992.13.
6. Burnside J, Gomez D, Preston TJ, Olshove VF, Phillips A. In-vitro quantification of gas-
eous microemboli in two extracorporeal life support circuits. J Extra Corpor Technol.
2011;43:123–9.
7. Yee S, Qiu F, Su X, et al. Evaluation of HL-20 roller pump and Rotaflow centrifugal pump
on perfusion quality and gaseous microemboli delivery. Artif Organs. 2010;34(11):937–43.
https://doi.org/10.1111/j.1525-1594.2010.01079.x.
78 B. H. Rosen
8. Lawson DS, Lawson AF, Walczak R, et al. North American neonatal extracorporeal mem-
brane oxygenation (ECMO) devices and team roles: 2008 survey results of Extracorporeal Life
Support Organization (ELSO) centers. J Extra Corpor Technol. 2008;40(3):166–74.
9. Lawson DS, Walczak R, Lawson AF, et al. North American neonatal extracorporeal membrane
oxygenation (ECMO) devices: 2002 survey results. J Extra Corpor Technol. 2004;36(1):16–21.
10. Drinker PA, Bartlett RH, Rishon MB, Noyes BS. Augmentation of membrane gas transfer by
induced secondary flows. Surgery. 1969;66(4):775–81.
11. Diller TE, Mikic BB, Drinker PA. Shear-induced augmentation of oxygen transfer in blood. J
Biomech Eng. 1980;102:67–72.
12. Gaylor JDS. Membrane oxygenators: current developments in design and application. J
Biomed Eng. 1988;10:541–7.
13. Mueller T, Lubnow M, Philipp A, et al. Extracorporeal pumpless interventional lung assist in
clinical practice: determinants of efficacy. Eur Respir J. 2009;33:551–9.
14. Johnson P, Frohlich S, Westbrook A. Use of extracorporeal membrane lung assist device
(Novalung) in H1N1 patients. J Card Surg. 2011;26(4):449–52. https://doi.org/10.1111/j.1540-
8191.2011.01261.x.
15. Kolobow T, Bowman RL. Construction and evaluation of an alveolar membrane artificial
heart-lung. ASAIO J. 1963;9:238–43.
16. Khoshbin E, Roberts N, Harvey C, et al. Poly-methyl pentene oxygenators have improved
gas exchange capability and reduced transfusion requirements in adult extracorpo-
real membrane oxygenation. ASAIO J. 2005;51(3):281–7. https://doi.org/10.1097/01.
mat.0000159741.33681.f1.
17. Montoya JP, Shanley CJ, Merz SI, Bartlett RH. Plasma leakage through microporous mem-
branes: role of phospholipids. ASAIO J. 1992;38(M399-M405):M399–405.
18. Eash HJ, Jones HM, Hattler BG, Federspiel WJ. Evaluation of plasma resistant hollow fiber
membranes for artificial lungs. ASAIO J. 2004;50(5):491–7. https://doi.org/10.1097/01.
mat.0000138078.04558.fe.
19. Puis L, Ampe L, Hertleer R. Case report: plasma leakage in a polymethylpentene oxy-
genator during extracorporeal life support. Perfusion. 2009;24(1):51–2. https://doi.
org/10.1177/0267659109106294.
20. Antonucci E, Lamanna I, Fagnoul D, Vincent JL, De Backer D, Silvio TF. The impact of renal
failure and renal replacement therapy on outcome during extracorporeal membrane oxygen-
ation therapy. Artif Organs. 2016;40(8):746–54. https://doi.org/10.1111/aor.12695.
21. Ostermann M, Connor M Jr, Kashani K. Continuous renal replacement therapy during extracor-
poreal membrane oxygenation: why, when and how? Curr Opin Crit Care. 2018;24(6):493–503.
https://doi.org/10.1097/MCC.0000000000000559.
22. Fosse E, Moen O, Johnson E, et al. Reduced complement and granulocyte activation with
heparin-coated cardiopulmonary bypass. Ann Thorac Surg. 1994;58:472–7.
23. Jansen PGM, Velthius H, Huybregts RAJM, et al. Reduced complement activation and
improved postoperative performance after CPB with heparin-coated circuits. J Thorac
Cardiovasc Surg. 1995;110:829–34.
24. Pekna M, Hagman L, Halden E, Nilsson UR, Nilsson B, Thelin S. Complement activa-
tion during cardiopulmonary bypass: effects of immobilized heparin. Ann Thorac Surg.
1994;58:421–4.
25. Gravlee GP. Heparin-coated cardiopulmonary bypass circuits. J Cardiothorac Vasc Anesth.
1994;8(2):213–22.
26. Gu YJ, van Oeveren W, Akkerman C, Boonstra PW, Huyzen RJ, Wildevuur CRH. Heparin-
coated circuits reduce the inflammatory response to cardiopulmonary bypass. Ann Thorac
Surg. 1993;55:917–22.
27. Ranucci M, Ballotta A, Kandil H, et al. Bivalrudin-based versus conventional heparin anticoag-
ulation for postcardiotomy extracorporeal membrane oxygenation. Crit Care. 2011;15:R275.
2 Circuits, Membranes, and Pumps 79
28. Silvetti S. Do we need heparin coating for extracorporeal membrane oxygenation? New
concepts and controversial positions about coating surfaces of extracorporeal circuits. Artif
Organs. 2015;39:176–9.
29. Cuker A. Clinical and laboratory diagnosis of heparin-induced thrombocytopenia: an inte-
grated approach. Semin Thromb Hemost. 2014;40:106–14.
30. Pollak U, Yacobobich J, Tamary H, Dagan O, Manor-Shulman O. Heparin-induced thrombo-
cytopenia and extracorporeal membrane oxygenation: a case report and review of the litera-
ture. J Extra Corpor Technol. 2011;43(1):5–12.
31. Beiderlinden M, Treschan T, Gorlinger K, Peters J. Argatroban in extracorporeal membrane
oxygenation. Artif Organs. 2007;31(6):461–5.
32. Pappalardo F, Maj G, Scandroglio A, Sampietro F, Zangrillo A, Koster A. Bioline heparin-
coated ECMO with bivalirudin anticoagulation in a patient with acute heparin-induced throm-
bocytopenia: the immune reaction appeared to continue unabated. Perfusion. 2009;24:135–7.
Chapter 3
Modes of ECMO
Chapter 3 Modes
The ability to understand and manipulate extracorporeal therapies has become piv-
otal in practicing critical care medicine. Extracorporeal life support (ECLS) is a
collective term for therapies supporting a range of causes of cardiac or pulmonary
failure, emphasizing oxygenation, carbon dioxide removal, circulatory support, or a
combination. Extracorporeal membrane oxygenation (ECMO) is the provision of
oxygen and carbon dioxide exchange using an extracorporeal circuit. These terms
explicitly exclude cardiopulmonary bypass for surgical procedures. ECLS has rap-
idly evolved from a rescue therapy to an early interventional therapeutic modality.
Much of this change can be attributed to our familiarity with the technology, its
technological advances, and experience gained during the influenza and COVID-19
pandemics.
This text emphasizes adult ECMO, especially veno-venous ECMO (VV ECMO)
related to support for the acute respiratory distress syndrome (ARDS), since this has
been the major area of growth in the field. Nevertheless, practitioners must recog-
nize when veno-arterial (VA) ECMO is appropriate, as well as the role for
Nomenclature
Table 3.1 This table shows the correct method to document ECLS cannulation configuration that
should be adopted by all ECMO centers to standardize reporting
Abbreviation
Access (primary, uppercase; secondary, A or a Systemic artery
lowercase)
V Systemic vein
P Pulmonary artery
Cannula entry site c Carotid artery
f Femoral vessel
j Jugular vein
s Subclavian vessel
3 Modes of ECMO 83
Table 3.2 Selecting the mode: veno-venous (VV), veno-arterial (VA), or arteriovenous (AV)
pumpless
Table 3.3 Venovenous (VV) and venoarterial (VA) ECMO—advantages and disadvantages
Factor VV VA
Cannulas One may suffice Two are needed
Gas exchange efficiency Lower Higher
PaO2 Lower Higher
Recirculation Moderate None
Circulatory support None Full is possible
RV loading Small effect Reduced
LV loading No effect Increased
Systemic pulsatility Normal Reduced
Risk of systemic embolism Low Higher
Circuit pressure Lower Higher
RV right ventricle, LV left ventricle, PaO2 partial pressure of oxygen
large vein. The rest of this chapter describes the circuit design for each mode, their
physiological impact, factors leading to mode selection for individual patients, and
the advantages of each mode (Tables 3.2 and 3.3).
The VV Circuit
Veno-venous ECMO is used primarily for gas exchange in patients with isolated,
refractory respiratory failure. Because blood is drained as well as returned into the
venous system, adequate native cardiac function is essential since this form of
ECMO provides no direct circulatory support [2].
84 J. Eaton et al.
Fig. 3.2 Illustration of the standard ECMO configuration for the single cannula, dual-lumen
setup. Blood is withdrawn through a distal (inferior vena cava) and proximal (superior vena cava)
port and returned through a port positioned near and directed toward the tricuspid valve.
Configuration is (dl) Vj-V ECMO
possible is preferred. A general rule of thumb is that the vessel diameter (in mil-
limeters; judged by bedside ultrasound) should be at least one-third the cannula
size (in French). This can be estimated prior to cannulation by bedside vascular
ultrasound, which allows an additional opportunity to verify that there is no
major thrombus or stenosis present that would complicate cannulation at that
site. Cannulation is more technically challenging and riskier than for the two-
site approach. Placement of the guidewire into the IVC should be verified and
the cannula visualized continuously (transthoracic or transesophageal echocar-
diography or fluoroscopy) during insertion to avoid catastrophic vascular or car-
diac injuries, requiring additional personnel, space, and equipment. See also
Chap. 4.
some second-order effects that could unload the right heart. For example, to the
extent that the membrane supplants the need for mechanical ventilation, lower tidal
volumes and PEEP are usually used, and these changed ventilator settings could
unload the RV. Similarly, once the pulmonary circulation is perfused with blood
having a higher PO2 and lower PCO2, vascular resistance falls acutely. In the CESAR
Trial of ECMO for severe ARDS, all patients were managed with the VV mode, and
no patient had to transition to a VA circuit during the trial. In contrast to the VV
mode, VA ECMO can fully support the circulation, so is more appropriate for
patients with massive pulmonary thromboembolism, ARDS complicated by severe
cor pulmonale, extra-corporeal cardiopulmonary resuscitation (eCPR), or following
cardiac surgery (Table 3.2). In hemodynamically unstable patients for whom ECLS
is being contemplated, it is important to delineate the basis for circulatory failure
(using echocardiographic imaging, for example) as this may influence the choice of
VV or VA ECMO.
Advantages of VV ECMO
Several advantages are conferred because the VV mode returns blood to the central
veins, rather than the arteries (Table 3.3). These include a reduced risk of systemic
embolism, including catastrophic cerebral or coronary embolism; no injury to sys-
temic arteries that could cause hemorrhage or distal ischemia; lower circuit pres-
sures with consequently lower chance of catastrophic circuit failure; and the
potential for a single cannula to suffice, possibly facilitating mobilization (see Chap.
14). VV ECMO also maintains the pulsatility of the systemic circulation, which is
lost to some degree during VA support, as described below. Finally, decannulation
can be performed at the bedside with no need for surgical repair or ligation of
vessels.
Veno-arterial ECMO
Veno-arterial (VA) ECMO can provide both cardiac and respiratory support [4]. For
adults with respiratory failure but preserved cardiac function, VV-ECMO is prefer-
able since it avoids the risks associated with large bore-arterial cannulation.
The VA Circuit
With VA ECLS, blood is withdrawn from the venous circulation either by direct
surgical cannulation of the RA or by a cannula placed in a vein with the tip posi-
tioned in the RA, SVC, or IVC. Blood is returned through the carotid artery in
88 J. Eaton et al.
Fig. 3.3 Illustration of VA cannulation for neonates and younger children. The outflow cannula is
inserted through the internal jugular vein with return through the common carotid artery.
Configuration is Vj-Ac ECMO
neonates and infants (Fig. 3.3) and the descending aorta (via the femoral artery)
in older children and adults (Fig. 3.4). Where to place the return cannula depends
partly on the blood flow needed. If a need for full support is anticipated (e.g.,
100–125 cc/kg/min in a neonate), the excessive resistance of peripheral arteries
may not allow sufficient flow. Since blood flow is largely dependent on cannula
diameter, the largest possible cannula is usually inserted. Vessel size is judged by
ultrasound to select a cannula of adequate size to provide flow without obstructing
distal arterial flow beyond the cannula. If the arterial cannula compromises or
obstructs nutrient blood flow, as is occasionally the case when a femoral cannula
threatens perfusion to the leg, a small catheter can be aimed distally to provide
additional blood flow (Fig. 3.5). Central cannulation of the right atrium and the
ascending aorta allows larger cannulas to be used, providing higher flows and
3 Modes of ECMO 89
The gas exchange effects of VA ECMO are similar to those of VV ECMO, but there
are important differences. First, PaO2 values are typically higher on the VA mode
because most of the blood has passed through the ECMO circuit, with less provided
by the (failing) native circulation. In contrast, with the VV configuration much
venous (desaturated) blood is not captured by the drainage cannula(s) and passes
90 J. Eaton et al.
Fig. 3.5 Illustration of a hybrid system that started as a VA (Vf-Af) configuration and was con-
verted to veno-venoarterial (VVA) because of upper body hypoxemia. This configuration (Vf-Vj-
Af) ECMO also arises following initial VV support when circulatory support becomes necessary,
requiring the addition of a veno-arterial component
through the diseased lungs, producing venous admixture and causing lower PaO2
values. Second, cannula position in VA ECMO precludes the possibility of recircu-
lation, so gas exchange is more efficient. Finally, there is often a gradient of arterial
PaO2 related to the competition from fully oxygenated blood moving retrograde
from the descending aorta and de-oxygenated blood coming from the native cardiac
output and moving antegrade through the aorta.
The higher the native cardiac output and the more diseased the lungs, the lower
will be the proximal aortic oxygen partial pressure and the more distally this effect
will be seen. Commonly, the right arm PaO2 (and similarly the pulse oxygen satura-
tion) is lower than that in the left arm or in the legs. Since the coronary and carotid
arteries are supplied from the proximal aorta, this peculiarity of varying arterial
oxygen values may further threaten the circulation or neurological status.
While the gas exchange differences between VV and VA ECMO are modest,
circulatory physiology is completely different. VV ECMO has essentially no
3 Modes of ECMO 91
circulatory effect, yet the basic function of VA ECMO is to provide circulatory sup-
port, returning oxygenated blood to the arterial circulation at physiological perfu-
sion pressures. Since blood is drawn from the vena cava or RA, VA ECMO unloads
the right ventricle, and this mode has been used for patients with massive pulmo-
nary embolism, for example. However, the left ventricle does not benefit similarly.
Even though the LV receives less blood from the lungs, a failing LV may only eject
little stroke volume, especially once extracorporeal support raises blood pressure to
normal levels. Related to the low native cardiac output, aortic root flow may be slug-
gish with areas of stasis. This has the potential to promote thrombosis, raising the
risk of stroke or other manifestation of systemic embolization. Moreover, blood
from Thebesian veins and other sources drains into the LV, so this LV distension can
lead to complications such as pulmonary edema and RV overload. Because of this,
efforts are necessary to augment LV systolic function, monitor LA and LV size, and
possibly decompress the LV. Such measures should be included in the VA ECLS
care pathway. One method to reduce LV distention is to vent the LA. With central
cannulation, the technique is surgically straightforward. However, venting the LA is
more challenging when patients are cannulated peripherally. Options include trans-
cutaneous septoplasty, insertion of a vent cannula through a mini thoracotomy, or
placement of an intra-aortic balloon pump (IABP). Concerns with the use of an
IABP is that it could obstruct ECLS flow returning through the descending aorta or
flow from the femoral arterial cannula could compete with the inflated state of the
IABP [3]. However, several studies have shown a benefit of the IABP in off-loading
the LV and possibly enhancing cerebral and coronary blood flow.
Another difference between VV and VA ECMO relates to systemic arterial pul-
satility. Since pump flow is non-pulsatile, the systemic blood pressure tends also to
be lacking in variability, although there is often some component related to the
native circulation. Even when mean blood pressures are in the normal range, the
systolic pressure on VA ECMO tends to be significantly lower (and diastolic higher)
compared to normal physiology. Concerns have been expressed that non-pulsatile
systemic blood pressures could lead to renal dysfunction, but this has not borne out
in practice. Mean blood pressure relates to total circuit flow, native cardiac output,
and systemic vascular resistance. The adequacy of perfusion can be judged by
assessing mean blood pressure, central venous oxyhemoglobin saturation, lactic
acid levels, and end-organ function, but metrics related to pulse contour analysis
(including many modern minimally invasive cardiac output technologies) will not
be valid. Still, the return of a pulsatile waveform on the peripheral arterial trans-
ducer can be useful as a gauge of cardiac recovery.
Advantages of VA ECMO
In comparison with VV ECMO, the VA mode leads to more efficient gas exchange
and no possibility of recirculation. Systemic PO2 is typically much higher than in
VV configuration, although this is sometimes not true of the cerebral, coronary, and
92 J. Eaton et al.
right upper extremity circulations as described above. Most importantly, the hemo-
dynamic effects discussed previously allow full support of the circulation, including
unloading of the RV, whereas VV does not. Thus, “eCPR” is feasible even in patients
suffering full cardiac arrest.
VVA-ECMO
In this configuration, a venous outflow cannula leads to the pump and artificial
membrane, after which blood is returned both to the right atrium (or vena cava) and
to the systemic arteries (Fig. 3.5). Often this arises when a patient is started on VA
ECMO but has persistent upper extremity, cardiac, or cerebral hypoxia. This usually
evolves in the face of worsening lung function where the blood entering the LA
becomes progressively hypoxemic. Placement of a second venous cannula (inflow,
connected by a “Y” to the arterial inflow) in the IVC, SVC, or RA allows a stream
of fully oxygenated blood to traverse the lungs, raising the saturation of blood pass-
ing through the left heart and into the proximal aorta. A downside lies in the fact that
flow is diverted from the arterial system, possibly reducing oxygen delivery.
Increasing the circuit flow rate may compensate for this but may be limited by can-
nula size or the adequacy of venous drainage. Furthermore, after a venous inflow
cannula is spliced into the arterial limb of the circuit, much of the flow may prefer-
entially go toward the venous system, being of lower pressure. This may negatively
impact mean arterial pressure support but may be remedied by applying a partially
occlusive clamp to the venous branch of the inflow limb of the circuit (Fig. 3.6).
Alternatively, this hybrid configuration may be reached when a patient on VV
ECMO develops cardiac dysfunction, for example, due to progressive cor pulmo-
nale. Now the addition of an arterial inflow cannula unloads the RV and directly
supports the circulation. This configuration is often accompanied by significant ino-
tropic support of the circulation.
Other hybrid systems have been described, such as combining VV ECMO with
IABP support or performing an atrial septoplasty to unload the RV in a patient with
severe veno-occlusive disease. There are few data on these hybrid modes and their
impact on ECMO duration, complications, or survival. In one small series, ARDS
3 Modes of ECMO 93
membrane specialized for CO2 removal; (2) arterio-venous pumpless system (AV
ECCO2R); and (3) adapting a continuous renal replacement therapy (CRRT) circuit
to integrate a CO2 removal membrane into the circuit. We will discuss the limits and
merits of each system in more detail.
VV ECCO2R
Gattinoni and colleagues were one of the first groups to describe effective carbon
dioxide removal using an extracorporeal device [5]. This approach uses a pump to
drain blood from the venous system, pass it through a gas exchanger, and return it
to the venous system. Since CO2 diffuses through the membrane more readily than
oxygen, much lower circuit blood flow is required, meaning smaller cannula size,
narrower tubing, and lower pump speeds. In theory, this should reduce hemolysis
and reduce complications of cannula insertion, while providing sufficient support to
reduce ventilator settings [6]. The same goals can be achieved through much larger
ECMO circuits, but without some of the potential advantages.
VV ECCO2R can be used for patients with predominantly obstructive physiol-
ogy, as in COPD exacerbations or status asthmaticus (to prevent intubation or to
liberate more rapidly), as discussed in Chap. 10. Recently, VV ECCO2R has been
used to facilitate “ultra-lung-protective” ventilation for patients with severe ARDS,
allowing tidal volumes of only 2–4 mL/kg.
VV ECCO2R has been used successfully in dealing with COVID-19-related
hypercapnic respiratory failure. In one case series, 29 patients with COVID-19 were
treated with an extracorporeal CO2 removal device with adequate control of CO2
and pH levels [7]. Complications were mostly cannulation issues, and there was no
significant bleeding, hemolysis, or device failure. Others have reported similar
experience [8]. One such VV-ECCO2R device that is currently undergoing FDA
review is shown in Fig. 3.7.
AV ECCO2R
CVVRRT
Patients on renal replacement therapy who develop respiratory failure could be can-
didates for CVVRRT. This approach inserts a gas exchange membrane in the return
limb of the RRT circuit. By and large these systems can process a blood flow up to
450 mL/min, but most are limited to 200–300 mL/min due to cannula and circuit
pressure limitations [13]. The use of this mode is in its infancy and has numerous
limitations. Further investigation is needed before this is adopted widely.
References
1. Conrad SA, Broman LM, Taccone FS, Lorusso R, Malfertheiner MV, Pappalardo F, et al. The
extracorporeal life support organization maastricht treaty for nomenclature in extracorporeal
life support. a position paper of the extracorporeal life support organization. Am J Respir Crit
Care Med. 2018;198(4):447–51 https://doi.org/10.1164/rccm.201710-2130CP.
2. Brodie D, Bacchetta M. Extracorporeal membrane oxygenation for ARDS in adults. N Engl J
Med. 2011;365(20):1905–14.
4. Schroeter T, Vollroth M, Höbartner M, Dhein S, Sahlisch S, Borger MA, Mohr
FW. Extracorporeal Membrane Oxygenation and Intra-Aortic Ballon Pump – an appropriate
combination or useless battle of materials? Thorac Cardiovasc Surg. 2013;61(S 01):OP19.
96 J. Eaton et al.
4. Nguyen DQ, Kulick DM, Bolman RM III, Dunitz JM, Hertz MI, Park SJ. Temporary
ECMO support following lung and heart-lung transplantation. J Heart Lung Transplant.
2000;19(3):313–6.
5. Low-Frequency Positive Pressure Ventilation with Extracorpor...: Anesthesia & Analgesia.
Accessed May 20, 2014. http://journals.lww.com/anesthesia-analgesia/Fulltext/1978/07000/
Low_Frequency_Positive_Pressure_Ventilation_with.18.aspx.
6. Fanelli V, Costamagna A, Terragni PP, Ranieri VM. Low-flow ECMO and CO2 removal.
In: Sangalli F, Patroniti N, Pesenti A, editors. ECMO-extracorporeal life support in adults.
Springer Milan; 2014. p. 303–15.
7. Akkanti B, Jagpal S, Darwish R, Romero RS, Scott LK, Dinh K, Hussain S, Radbel J, Saad
MA, Enfield KB, Conrad SA. Physiologic improvement in respiratory acidosis using extra-
corporeal Co2 removal with hemolung respiratory assist system in the management of severe
respiratory failure from coronavirus disease 2019. Crit Care Explor. 2021;3(3):e0372.
8. Boparai S, Keith Scott L, Conrad S, Motayar N. Extra-corporeal carbon dioxide removal in
COVID-19 ARDS. Chest. 2020;158(4):A1029.
9. Brunston RL Jr, Tao W, Bidani A, Alpard SK, Traber DL, Zwischenberger JB. Prolonged
hemodynamic stability during arteriovenous carbon dioxide removal for severe respiratory
failure. J Thorac Cardiovasc Surg. 1997;114(6):1107–14.
10. Arteriovenous Extracorporeal Carbon Dioxide Removal A Mathem... : ASAIO
J. Accessed May 16, 2014. http://journals.lww.com/asaiojournal/Fulltext/1998/07000/
Arteriovenous_Extracorporeal_Carbon_Dioxide.7.aspx.
11. Conrad SA, et al. Total extracorporeal arteriovenous carbon dioxide removal in acute respira-
tory failure: a phase I clinical study. I. Intensive Care Med. 2001;27:1340–51.
12. Mauri T, Zanella A, Pesenti A. Extracorporeal gas exchange: present and future. In: Jean-Louis
VP, editor. Annual update in intensive care and emergency medicine 2013. Berlin Heidelberg:
Springer; 2013. p. 609–19.
13. Grant AA, Hart VJ, Lineen EB, et al. Rescue therapy for hypercapnia due to high
PEEP mechanical ventilation in patients with ARDS and renal failure. Artif Organs.
2019;43:599–604.
Chapter 4
Vascular Access
Steven A. Conrad
Introduction
S. A. Conrad (*)
Departments of Medicine, Emergency Medicine, Pediatrics and Surgery, Louisiana State
University Health Sciences Center, Shreveport, LA, USA
Ochsner LSU Health Academic Medical Center, Shreveport, LA, USA
LSU Health Shreveport, Shreveport, LA, USA
e-mail: [email protected]
Single-Lumen Design
Cannulas with a single lumen are required for venoarterial and arteriovenous vascu-
lar access and are an optional approach for venovenous access. Two fundamental
designs are manufactured, intended for either drainage or return (often referred to as
venous and arterial cannulas, respectively). The drainage cannula design is charac-
terized by a greater length (up to approximately 50 cm), greater available diameter
(up to 29 Fr), and a longer distal segment with multiple side holes to facilitate drain-
age. The greater length allows insertion into more central veins such as the superior
vena cava (SVC) or inferior vena cava (IVC) (Fig. 4.1a). Drainage cannulas are
designed with different placement and configurations of side holes. The return can-
nula design is characterized by a shorter length and a shorter distal segment with a
limited number of side holes, since deeper insertion is not required, and flow is not
dependent on side holes as in the drainage design (Fig. 4.1b). An excessive number
of side holes can increase the risk of hemolysis in return cannulas.
Recently introduced expandable, wire-reinforced cannulas that incorporate a dis-
tal segment of wall free wire mesh (Smartcannula LLC, Switzerland) are available
in some markets. These cannulas expand to a larger diameter within the vessel to
minimize flow resistance, and the distal mesh maintains vessel patency for improved
drainage [3].
4 Vascular Access 99
Dual-Lumen Design
Cannulas incorporating two lumens are a more recent design that have facilitated
the application of venovenous support for respiratory failure. Although designed for
percutaneous insertion, they can be placed surgically as well. Two fundamental
designs are available that have features to support different needs.
The bicaval design requires insertion via the internal jugular vein with the cannula
traversing the right atrium and the tip positioned in the IVC (Avalon Elite, Maquet
Cardiopulmonary, Rastatt, Germany and Crescent, Medtronic, Bloomington, MN,
USA) [4, 5] (Fig. 4.2). The drainage lumen extends the length of the cannula with
two drainage ports, one in each the IVC and SVC. The return lumen is shorter, ter-
minating in the right atrium with the return port directed toward the tricuspid valve.
This bicaval design effectively separates upper and lower body venous drainage and
results in lower recirculation with more effective systemic oxygen delivery.
Recirculation fractions on the order of 2–3% are typical, unlike 20–30% recircula-
tion fractions characteristic of two-site single-lumen cannulation [4]. These cannulas
are available in sizes suitable for venovenous support ranging from neonates to adults.
The second design is similar to a hemodialysis catheter with a single proximal
drainage port and a distal reinfusion port. Two variations of this design exist. A
longer and larger diameter version (Protek Duo, LivaNova PLC) is intended for
insertion via the right internal jugular vein into the main pulmonary artery, with the
proximal drainage port in the right atrium and the return port in the pulmonary
artery. It is intended for high-flow venovenous support, with the added advantage
that it can support right heart function. Recirculation is minimal to absent since the
pulmonary valve prevents oxygenated blood from reaching the drainage port. A
shorter and smaller diameter version that can be inserted into the internal jugular or
femoral vein (as for hemodialysis) is available in 15.5 Fr diameter and intended for
low flow (500–800 mL/min) extracorporeal circuits used for extracorporeal carbon
dioxide removal (ECCO2R) [6, 7] (see Chap. 10). The cannula flow must not exceed
100 S. A. Conrad
the insertion vessel flow, or recirculation will limit effective blood flow. Placement
in the SVC via an internal jugular vein or in an iliac vein via the femoral vein usu-
ally assures adequate vessel blood flow.
Blood flow through vascular cannulas is driven by the difference between the pres-
sure at the hub of the cannula and the intravascular pressure at the tip of the cannula.
Although an ECMO cannula is cylindrical in shape, in which the relationship
between flow and pressure is expected to be linear in the presence of laminar flow,
the relationship is only linear at the low end of the flow range where laminar flow
predominates. The laminar flow relationship between flow ( Q ) and pressure gradi-
ent (ΔP) can be described by the Hagen-Poiseuille equation:
∆P ⋅ r 4
Q =
µ⋅L
where r is the radius, L is the length, and μ is the blood viscosity. Maximizing can-
nula blood flow involves the insertion of the largest diameter cannula that can be
safely inserted and keeping the length as short as possible.
4 Vascular Access 101
Fig. 4.3 Representative pressure-flow relationships for various size single-lumen return cannulas.
The graph depicts the nonlinear relationship between flow and pressure due to a combination of
laminar, transitional, and turbulent flow resulting from the complex geometry of the cannula
The actual pressure-flow relationship over the full range of blood flow is nonlin-
ear, most likely due to the development of transitional and turbulent flow in the
lumen and at the side holes as flow increases. Turbulence increases the pressure
gradient required to maintain a given level of flow. This nonlinear relationship is
demonstrated in a typical pressure-flow diagram shown in Fig. 4.3. Note that manu-
facturers provide pressure-flow data for their cannulas using water as the fluid, so
that the actual pressures required to generate flow will be higher with blood.
Patient Preparation
During surgical cannulation the vessel is exposed, and cannula size selection can be
made visually at the time of cannulation. Determination of cannula diameter prior
to percutaneous cannulation, however, requires imaging to minimize complications.
Without vessel sizing the use of too large a cannula can result in venous obstruction,
102 S. A. Conrad
Fr = D( mm ) ⋅ 3
For noncircular vessels the French size is approximately equal to the circumfer-
ence in millimeters. Both diameter and circumference measurements are easily per-
formed with bedside ultrasound machines. The chosen cannula should be 10–20%
smaller than the measured vessel to help assure successful placement and prevent
complete obstruction of blood flow.
Infection Control
Insertion Technique
Three techniques for cannula insertion are commonplace. Historically, all cannula-
tions were performed by surgeons using an open surgical technique. While some
vessels still require an open approach, surgical cannulation in most cases has been
replaced with percutaneous cannulation and performed by surgeons, intensivists,
interventionalists, and emergency physicians.
Percutaneous
Percutaneous cannulation has been used successfully for both venovenous and
peripheral venoarterial support [11]. It is the preferred technique since it is associ-
ated with a lower incidence of cannulation site bleeding and infection [12, 13]. It
can also be nonobstructive, allowing blood flow around the cannula after placement,
avoiding ischemic and congestive complications. It can be used for any peripheral
arterial access as well as venous access, although experience with carotid cannula-
tion is limited.
The same Seldinger technique used for smaller vascular access catheters is used
for ECLS cannulation, but with multiple dilators of no more than 4 Fr increments in
size (typically 8, 12, 16, 20, 24, 28, and 30–32 Fr), with the largest size approxi-
mately equal to the size of the cannula to be inserted. Vessel size is determined prior
to insertion with ultrasound, and an appropriately sized cannula is chosen. If the
cannula is being placed into the thoracic cavity, adequate sedation and a neuromus-
cular blocker are administered to prevent inspiratory effort and air embolism. Under
aseptic conditions and following infiltration of a local anesthetic, the vessel is iden-
tified with ultrasound, and an approach is chosen to avoid injury to neighboring
vessels, since vessels may overlie each other. The access needle is inserted using
ultrasound to guide it through the center of the vessel, and a 0.35″ or 0.38″ guide-
wire is advanced. Fluoroscopy is invaluable for preventing guidewire misadventures
during advancement and recommended for the bicaval dual-lumen cannula to assure
placement of the wire directly across the atrium.
104 S. A. Conrad
Semi-Open
Open Surgical
The preferred technique for cervical cannulation when carotid or subclavian arterial
access is required, or other peripheral artery following failure of percutaneous can-
nulation, is the open surgical technique. Following sedation, neuromuscular block-
ade, and skin preparation, an incision is made perpendicular to the axis of the vessel
with dissection carried down to expose the vessels. The cannula can be sized by
visual comparison with the vessel diameter. The vessels are freed from surrounding
tissue, and ligatures are placed proximal and distal to control bleeding. An arteriot-
omy (and venotomy) is made, and the cannula with its blunt-tipped loading dilator
is inserted into the vessel while loosening the proximal ligature to admit the can-
nula. Following insertion to the proper depth, the ligatures are secured, typically
with pledgets to prevent vessel injury, as vessel repair may be performed after
decannulation. The subcutaneous tissue and skin are closed, taking care to securely
close the skin around the cannula. The above description is generic, and variations
are numerous, subject to the surgeon’s preferences and experience.
If open cannulation is performed on the common femoral artery, which has no
collateral circulation, then a smaller antegrade cannula is placed distal to the pri-
mary cannula in the superficial femoral artery to maintain distal perfusion. A 6–8 Fr
cannula is sufficient and attached to the circuit as for percutaneous cannulation.
An alternative approach to insertion of the cannula into the artery is to suture a
vascular graft to the artery (end to side), with placement of the cannula into the
graft. The artery then remains fully open, and distal ischemia is avoided, and a
larger cannula can be used. This approach is often chosen when long-duration sup-
port is anticipated, such as bridge to transplant.
Cannulation Configuration
The foremost decision regarding vascular access is the mode of support (respiratory,
cardiac, or hybrid) which dictates the configuration of cannula placement.
Extracorporeal life support for both respiratory and cardiac failure was historically
performed using a venoarterial (VA) configuration. While still preferred for cardiac
failure, other configurations have been developed that are more suitable for other
types of support. As these configurations are presented in Chap. 2, the following
will focus on cannula sizes and vascular access associated with each configuration.
The venoarterial configuration drains blood from the central venous circulation and
returns it to the arterial circulation. The cervical approach is used in neonates,
infants, and small children, since the femoral vessels are small prior to the age of
106 S. A. Conrad
walking. Cervical vessels have collateral circulation from the contralateral side,
allowing ligation of the carotid artery and internal jugular vein. The arterial return
cannula, typically 8–10 Fr, is placed into the common carotid artery and advanced
into the proximal innominate artery. The venous drainage cannula, typically 10–14
Fr, is placed into the internal jugular vein and advanced into the right atrium. This
configuration supplies oxygenated blood to the proximal aorta, but coronary and
right upper extremity blood may be poorly saturated if pulmonary dysfunction is
present, and the left ventricle is ejecting (see also Chap. 16).
Femoral cannulation can be used for venoarterial cardiac support in adults and
older children. A long drainage cannula, typically 27–29 Fr, is inserted into the
common femoral vein and advanced to the right atrium, and a short return cannula,
typically 17–19 Fr, is placed into the common femoral artery. Alternatively, a short
drainage cannula, typically 23–24 Fr, can be placed into the right atrium via the
internal jugular in place of the long femoral venous cannula. These approaches are
suitable if the native lungs can provide adequate saturation of blood since, in the
presence of cardiac ejection, the upper half of the body is supplied by the native
heart and lungs and the lower half by the extracorporeal circuit.
Venovenous cannulation was introduced later than venoarterial and is suitable for
respiratory failure with adequate cardiovascular function. It provides oxygenated
blood into the venous system while relying on the native heart for oxygen delivery,
making oxygenated blood available to all tissues, including the myocardium.
Percutaneous cannulation for venovenous support has replaced surgical cannulation
as the preferred approach.
The venovenous configuration was introduced to extracorporeal support using
two single-lumen cannulas, one placed into the proximal IVC via the femora vein
for drainage, typically 27–29 Fr, and the second return cannula, typically 23–24 Fr,
placed into the SVC via the right or left internal jugular veins for return. This direc-
tion of flow (femoral-jugular) has been shown to result in less recirculation than the
jugular-femoral direction [16, 17]. An alternative configuration that can result in
higher flows and even lower recirculation is the placement of three cannulas, one in
the SVC via an internal jugular and one in the proximal IVC via a femoral vein, with
the return cannula placed into the right atrium via the femoral or jugular approach
[18]. This separation of venous return reduces recirculation to levels lower than pos-
sible with the two-cannula approach.
A significant advance in venovenous support was the introduction of the dual-
lumen, bicaval venovenous cannula for respiratory support (see dual-lumen design,
above). Developed initially for neonates [19], cannulas are also available for pediat-
ric and adult patients. This cannula design allows a single vascular access via the
right internal jugular vein. The cannula is placed by accessing the right internal
jugular, advancing the guidewire across the atrium into the IVC under fluoroscopic
imaging, then inserting the cannula with the distal drainage port in the proximal
4 Vascular Access 107
IVC, return port in the mid-right atrium, and second drainage port in the SVC. The
catheter can be placed via the left internal jugular in selected patients, but other
access points are not supported.
Venovenous support can target carbon dioxide removal (extracorporeal carbon diox-
ide removal, VV ECCO2R) to support lung-protective ventilation in patients for
whom oxygenation can be adequately provided through mechanical ventilation (see
Chaps. 6 and 10). Since ECCO2R only requires 0.5–1 L/min blood flow, cannulation
108 S. A. Conrad
for VV ECCO2R can be achieved by two smaller single-lumen cannulas (e.g., 14–16
Fr) placed into the femoral and jugular veins, or a dual-lumen cannula (15–16 Fr)
placed into either a femoral or jugular vein, similar to a hemodialysis catheter.
Although much more invasive, direct cannulation of the right atrium and aortic root
through a median sternotomy remains an important approach to vascular access for
extracorporeal support. The most common use is support following failure to wean
from cardiopulmonary bypass (CPB), in which the cardiopulmonary bypass circuit
is replaced with the ECLS circuit, connecting directly to the right atrial drain can-
nula and the aortic root [20]. Typically, the sternum is left open and draped with an
occlusive dressing. CBP arterial and venous cannulas are large and support more
flow than can be achieved using peripheral access.
The transthoracic approach is associated with more bleeding and greater infec-
tion risk, so it is generally used for patients expected to recover cardiac function
quickly. If prolonged support is required, the patient may be transitioned to periph-
eral cannulation or to a ventricular assist device. This approach has also been used
to provide high-flow support in patients with severe sepsis [21].
Decannulation
When extracorporeal support is no longer required and weaning trials have been
successful (see Chap. 15), the patient is prepared for decannulation by discontinu-
ing anticoagulation to allow coagulation studies to return to near normal. In the case
4 Vascular Access 109
Complications of Cannulation
Vascular Injury
Injury to the target or adjacent vessel during cannulation can result in inability to
achieve vascular access, hemorrhage into areas such as the retroperitoneal space,
transection of a vessel, exsanguination, and death. Immediate attempts at surgical
exploration to complete cannulation or to repair injured vessels may be attempted
but may not be successful. The risk is higher with percutaneous cannulation since
the vessels are not visible. The use of ultrasound before and during percutaneous
cannulation can mitigate these risks by allowing for selection of appropriate cannula
size, identification of adjacent vessels, selection of an approach to avoid overlying
vessels, and guidance of vessel puncture in the center to ensure proper entry into the
vessel and reduce risk of injury.
110 S. A. Conrad
Inadequate Flow
The inability to achieve the expected circuit flow can result in the inability to achieve
adequate cardiac support during venoarterial ECMO, or persistent hypoxemia with
inability to achieve lung-protective settings during venovenous support, thereby
decreasing the chance of survival. Three conditions that commonly lead to inade-
quate flow are placement of a cannula that is smaller than required, improper place-
ment resulting in impaired venous drainage, and hypovolemia.
Choice of cannula size should be driven by the flow needed for adequate sup-
port, typically 50–80 mL/kg/min for an adult and higher for pediatric patients.
Drainage can almost always be achieved using a single, appropriately sized
drainage cannula. Uncommonly and in special circumstances, two drainage can-
nulas maybe required. The reinfusion cannula is typically smaller than the
drainage cannula since flow is driven with a much greater pressure gradient
generated by the blood pump and is rarely the cause of inadequate flow. Improper
placement can be identified by radiography or echocardiography and subse-
quently be corrected.
Hypovolemia is the most common transient cause of inadequate flow. It can
result in “chattering” or “chugging” of the venous line in which the vascular struc-
tures cyclically collapse around the cannula resulting in intermittent flow. The nega-
tive pressure generated at the pump during this sudden cessation of flow contributes
to hemolysis and must be corrected. Initially reducing the pump speed until volume
expansion with colloid or blood can be achieved, then restoration of flow, resolves
the problem.
Limb Ischemia
Recirculation
Recirculation during venovenous support occurs when some of the reinfused blood
is aspirated directly into the venous drainage cannula rather than being delivered to
the patient’s circulation, thereby reducing the effective extracorporeal flow. It is
unavoidable with the use of single-lumen cannulas, where recirculation fraction can
reach 30% or higher. Recirculation manifests as a decrease in oxygen delivery and
drop in systemic arterial saturation, with an increase in drainage saturation along
with a decrease in the saturation difference between the drainage and return limbs.
Increases in recirculation can occur with displacement of the cannulas and may
require radiography to detect. It also increases with increasing flow such that higher
flows may reduce oxygen delivery. In single-lumen cannulation, placement of the
drainage cannula in the proximal IVC rather than in the right atrium and placement
of the return cannula in the SVC above the cavo-atrial junction results in the lowest
recirculation [22].
Recirculation is less extensive with dual-lumen cannulas. The bicaval design
is associated with the lowest degree of recirculation, often under 3%. The veno-
pulmonary design also has minimal to absent recirculation. Recirculation is
absent in venoarterial cannulation since the circuit blood is returned into the arte-
rial system.
112 S. A. Conrad
Infectious Complications
Infection of the cannula insertion sites is a challenging problem since the patient
may be totally dependent on extracorporeal support, and recannulation may be risky
or impossible. Prevention is by assuring good skin asepsis at the time of cannulation
as well as throughout the duration of extracorporeal support. If infection does
develop and appears to be localized to the skin, then the use of appropriate antibiot-
ics may be successful in eradicating the infection.
If bacteremia develops, then consideration should be given to replacing the extra-
corporeal circuit after an initial treatment period with antibiotics, since seeding of
the large surface area circuit can result in persistent bacteremia. If the cannula site
infection is not responsive to antibiotic therapy alone, then recannulation at an alter-
native site may be required.
References
1. Conrad SA, Broman LM, Taccone FS, Lorusso R, Malfertheiner MV, Pappalardo F, et al. The
extracorporeal life support organization Maastricht treaty for nomenclature in extracorporeal
life support. A position paper of the extracorporeal life support organization. Am J Respir Crit
Care Med. 2018;198(4):447–51.
2. Broman LM, Taccone FS, Lorusso R, Malfertheiner MV, Pappalardo F, Di Nardo M, et al. The
ELSO Maastricht treaty for ECLS nomenclature: abbreviations for cannulation configuration
in extracorporeal life support - a position paper of the extracorporeal life support organization.
Crit Care. 2019;23(1):36.
3. von Segesser LK, Berdajs D, Abdel-Sayed S, Ferrari E, Halbe M, Wilhelm M, et al. New,
optimized, dual-lumen cannula for veno-venous ECMO. Perfusion. 2018;33(1_suppl):18–23.
4. Wang D, Zhou X, Liu X, Sidor B, Lynch J, Zwischenberger JB. Wang-Zwische double
lumen cannula-toward a percutaneous and ambulatory paracorporeal artificial lung. ASAIO
J. 2008;54(6):606–11.
5. Bermudez CA, Rocha RV, Sappington PL, Toyoda Y, Murray HN, Boujoukos AJ. Initial expe-
rience with single cannulation for venovenous extracorporeal oxygenation in adults. Ann
Thorac Surg. 2010;90(3):991–5.
6. Batchinsky AI, Jordan BS, Regn D, Necsoiu C, Federspiel WJ, Morris MJ, et al. Respiratory
dialysis: reduction in dependence on mechanical ventilation by venovenous extracorporeal
CO2 removal. Crit Care Med. 2011;39(6):1382–7.
7. Morimont P, Batchinsky A, Lambermont B. Update on the role of extracorporeal CO(2)
removal as an adjunct to mechanical ventilation in ARDS. Crit Care. 2015;19:117.
8. Iserson KV. J.-F.-B. Charriere: the man behind the “French” gauge. J Emerg Med.
1987;5(6):545–8.
9. Bizzarro MJ, Conrad SA, Kaufman DA, Rycus P, Extracorporeal Life Support Organization
Task Force on Infections EMO. Infections acquired during extracorporeal membrane oxygen-
ation in neonates, children, and adults. Pediatr Crit Care Med. 2011;12(3):277–81.
10. Extracorporeal Life Support Organization Task Force on Infections. Infection control and extra-
corporeal life support. 2010. Available from: http://elso.org/downloads/resources/committees/
infectious-disease-and-antibiotic/Infection-Control-and-Extracorporeal-Life-Support.pdf.
11. Pranikoff T, Hirschl RB, Remenapp R, Swaniker F, Bartlett RH. Venovenous extracorporeal
life support via percutaneous cannulation in 94 patients. Chest. 1999;115(3):818–22.
4 Vascular Access 113
12. Conrad SA, Grier LR, Scott LK, Green R, Jordan M. Percutaneous cannulation for extracorpo-
real membrane oxygenation by intensivists: a retrospective single-institution case series. Crit
Care Med. 2015;43(5):1010–5.
13. Burrell AJ, Pellegrino VA, Sheldrake J, Pilcher DV. Percutaneous cannulation in predomi-
nantly venoarterial extracorporeal membrane oxygenation by intensivists. Crit Care Med.
2015;43(12):e595.
14. Kim DJ, Cho YJ, Park SH, Lim C, Park KH, Jheon S, et al. Near-infrared spectroscopy moni-
toring for early detection of limb ischemia in patients on veno-arterial extracorporeal mem-
brane oxygenation. ASAIO J. 2017;63(5):613–7.
15. Breeding J, Hamp T, Grealy R, Nair P, Iyer A, Kawanishi Y. Effects of extracorporeal membrane
oxygenation pump flow, backflow cannulae, mean arterial blood pressure, and pulse pressure
on Doppler-derived flow velocities of the lower limbs in patients on peripheral veno-arterial
extracorporeal membrane oxygenation: a pilot study. Aust Crit Care. 2019;32(3):206–12.
16. Lee JH, Won JY, Han JU, Son HS, Jung JS. Differences in recirculation: differences according
to different methods of cannulation in veno-venous extracorporeal membrane oxygenation.
Perfusion. 2018;33(1 Suppl):1420143.
17. Conrad SA, Wang D. Evaluation of recirculation during venovenous extracorporeal membrane
oxygenation using computational fluid dynamics incorporating fluid-structure interaction.
ASAIO J. 2021;67(8):943–53.
18. Ichiba S, Peek GJ, Sosnowski AW, Brennan KJ, Firmin RK. Modifying a venovenous
extracorporeal membrane oxygenation circuit to reduce recirculation. Ann Thorac Surg.
2000;69(1):298–9.
19. Anderson HL 3rd, Otsu T, Chapman RA, Barlett RH. Venovenous extracorporeal life support
in neonates using a double lumen catheter. ASAIO Trans. 1989;35(3):650–3.
20. Field ML, Al-Alao B, Mediratta N, Sosnowski A. Open and closed chest extrathoracic cannu-
lation for cardiopulmonary bypass and extracorporeal life support: methods, indications, and
outcomes. Postgrad Med J. 2006;82(967):323–31.
21. Maclaren G, Butt W, Best D, Donath S, Taylor A. Extracorporeal membrane oxygenation
for refractory septic shock in children: one institution’s experience. Pediatr Crit Care Med.
2007;8(5):447–51.
22. Conrad SA. Computational simulation of recirculation during venovenous extracorporeal
membrane oxygenation. Perfusion. 2018;33(1 Suppl):142–3.
Chapter 5
Hypoxemic Respiratory Failure: Evidence,
Indications, and Exclusions
Abbreviations
Evidence
ECMO is increasingly being used for patients with the acute respiratory distress
syndrome (ARDS), particularly in cases of severe ARDS in which life-threatening
hypoxemia or hypercapnia persists despite maximal conventional mechanical ven-
tilatory support [1, 2]. In addition, ECMO is used in some patients in whom life-
threatening gas exchange abnormalities are sufficiently improved with the use of
positive-pressure ventilation in order to sustain life, yet only at the expense of gen-
erating excessively high inspiratory airway pressures. ECMO in this setting facili-
tates lung-protective ventilation and minimizes ventilator-induced lung injury. For
decades, there was an increase in the use of ECMO in ARDS despite high level
evidence supporting its benefit. In the last several years, however, there has been
compelling data from a large randomized controlled trial as well as subsequent
analyses that have helped better inform the use of ECMO for this indication [3–7].
The first successful use of ECMO for ARDS dates back to the early 1970s [8]. In
the following decades, there was enthusiasm for the use of this modality, but small
case series had notably high mortality, and two randomized controlled trials, one
with ECMO and the other with extracorporeal carbon dioxide removal, did not
show benefit [9, 10]. The early technology being used at the time, combined with
the inexperience of the centers, led to serious complications and seemed to offset
any potential benefit that the patients may have realized.
Interest in ECMO for ARDS was reignited around 2009 due in part to increased
use in patients with respiratory failure during the influenza A(H1N1) pandemic, as
well as the publication of a large randomized controlled trial comparing ECMO to
conventional management [11]. An observational study of 68 patients with con-
firmed or suspected H1N1-associated ARDS treated with ECMO showed a survival
of 75% [12, 13]. However, a study with a similar patient population managed with-
out the use of ECMO demonstrated nearly identical outcomes, which raised ques-
tions about whether ECMO provides any survival advantage over conventional
medical management in this population [14]. In an attempt to reconcile this issue, a
comparison of outcomes between ECMO-referred and non-ECMO-referred sub-
jects with confirmed or suspected H1N1-associated ARDS was performed [15].
After matching for appropriate variables, in this intention-to-treat analysis, ECMO-
referred subjects consistently had a mortality approximately half that of the non-
ECMO-referred subjects (24% vs. 47% by propensity score matching, RR 0.51,
95% CI 0.31–0.84, p = 0.008).
Survival rates of subjects receiving extracorporeal support in these observational
studies were higher than those reported for subjects with and without extracorporeal
support in the earlier randomized trials. However, there are inherent flaws in com-
paring non-randomized studies, and changes in clinical management over time con-
found the comparison of survival rates from different eras. Likewise, ECMO
technology had evolved significantly since the early randomized trials, with more
efficient membranes for gas exchange, the advent of centrifugal pumps,
heparin-coated circuits that tolerate lower levels of anticoagulation resulting in
5 Hypoxemic Respiratory Failure: Evidence, Indications, and Exclusions 117
lower bleeding risk, and cannulae that permit single-vessel access with minimal
recirculation.
In an attempt to estimate the effect of ECMO in ARDS using mostly modern
ECMO technology and coinciding with increasing usage within the critical care
community in the setting of the 2009 pandemic, the Conventional Ventilation or
ECMO for Severe Adult Respiratory Failure (CESAR) trial was performed [11]. In
this prospective, randomized, controlled trial, 180 subjects, age 18–65, with severe
but potentially reversible respiratory failure and a Lung Injury Score (a.k.a. Murray
Score, a composite score based on the ratio of partial pressure of oxygen in arterial
blood to the fraction of inspired oxygen (PaO2:FIO2), positive end-expiratory pres-
sure (PEEP), respiratory system compliance, and radiographic findings) of ≥3.0 or
uncompensated hypercapnia (pH <7.2) despite “optimal conventional management”
were randomly assigned to receive ongoing conventional mechanical ventilation at
designated treatment centers or be transferred to a single ECMO center for consid-
eration of treatment with venovenous ECMO. Hemodynamically stable subjects
randomized to the ECMO referral arm were initially managed on transfer with a
standardized management protocol that included a pressure-restricted ventilation
strategy, diuresis, and prone positioning. Those who were hemodynamically unsta-
ble or failed to respond to this strategy within 12 h were placed on ECMO. Only
76% of the subjects referred for ECMO actually received ECMO; however all sub-
jects who received ECMO were managed with a lung-protective ventilation strat-
egy. In total, 93% of subjects in the ECMO referral arm received treatment with a
low-volume, low-pressure strategy at some point in their care. By comparison,
because there was no mandate for a lung-protective ventilation strategy in the con-
ventional management group and perhaps because many of these subjects were dif-
ficult to ventilate, only 70% of those subjects were managed with such a strategy at
any time during the study. The primary outcome—death or severe disability at
6 months after randomization—occurred in 37% of the subjects referred for ECMO,
as compared with 53% of those in the conventional management group, with a rela-
tive risk of 0.69 (95% confidence interval 0.05–0.97, p = 0.03).
The results from CESAR suggested that it is reasonable to transfer patients with
severe ARDS to a center capable of performing ECMO as part of a standardized
management protocol. However, this trial was not a randomized trial of ECMO as
compared with standard-of-care mechanical ventilation. The higher survival in the
ECMO-referred group could also be accounted for by differences in the care
between study groups, most importantly, the discrepancy in the use of a low-volume,
low-pressure mechanical ventilation strategy.
In 2018, the ECMO to Rescue Lung Injury in Severe ARDS (EOLIA) trial was
published in an attempt to reconcile questions left unanswered by CESAR [3]. In
this international, prospective, randomized controlled trial, 259 patients with very
severe ARDS were randomized to immediately receive venovenous ECMO or to
continue conventional treatment. Eligible patients met the American-European
Consensus Conference definition of ARDS, had been receiving invasive mechanical
ventilation for fewer than 7 days, and met certain disease-severity criteria, all while
118 K. E. Melville et al.
adhering to a low tidal volume strategy and receiving an FiO2 of greater or equal to
0.8 with a PEEP of at least 10 cm H2O. The specific disease severity criteria included
a PaO2:FiO2 of less than 50 mmHg for more than 3 h, a PaO2:FiO2 of less than
80 mmHg for more than 6 h, or an arterial blood pH of less than 7.25 with a partial
pressure of arterial carbon dioxide (PaCO2) of at least 60 mmHg while maintaining
a high respiratory rate of 35 breaths per minute, in an attempt to achieve a plateau
pressure below 33 cm H2O (see Table 5.1).
The trial was stopped early for projected futility in meeting the primary end-
point, a prespecified 20% absolute mortality reduction in the group randomized to
receive ECMO.
There was, however, a large, but not statistically significant, difference in mortal-
ity at 60 days. Forty-four of 124 (35%) of patients in the ECMO group and 57 of
125 (46%) in the control group died, yielding a relative risk of death of 0.76 (95%
CI 0.55–1.04, P = 0.09).
Crossover to ECMO for patients with severe refractory hypoxemia was permit-
ted under the trial design based on strict criteria. A total of 35 patients (28%) in the
control group crossed over and were cannulated to venovenous ECMO. These
patients, on average, had more severe ARDS, and worse multiorgan dysfunction,
including cardiovascular failure, compared to the cohort overall. The 60-day mor-
tality in this subset was 57% (20 of 35 patients), but a reasonable argument can be
made that significantly more of them would have died without receiving
ECMO. Therefore, in the intention-to-treat analysis, the mortality rates in the
control group were likely lower than they would have been without the allowance of
crossover. Similarly, an important secondary endpoint of “treatment failure,” which
was defined as death in the ECMO group and death or crossover to ECMO in the
control group, was significantly different between the two groups (35% in the
ECMO group and 58% in the treatment group).
In contrast to CESAR and other earlier trials, the patients in the control group
were held to strict standards of lung protective mechanical ventilation, consistent
with current standards of care. Ninety percent of patients in the control arm were
placed in the prone position, and 56% of patients in the intervention arm were
placed prone prior to ECMO initiation (an additional 10% were placed prone during
ECMO). Ventilator parameters, including tidal volumes, plateau airway pressures,
and driving pressures, were able to be decreased to a greater extent in the group
treated with ECMO, with post hoc analysis suggesting that the greatest benefit was
seen in those patients entering the trial under the third criteria related to poorly
compliant lungs. It is therefore reasonable to assume that much of the benefit
observed is due, in part, to minimization of ventilator-induced lung injury (VILI).
Importantly, overall differences in complications and adverse events were not
statistically significant between the two groups. However, there were more bleeding
events necessitating blood transfusion as well as cases of severe thrombocytopenia
in the ECMO group. There were fewer cases of ischemic stroke in the ECMO group
compared to the control group.
Using a standard frequentist approach to statistical analysis, the EOLIA trial was
negative, having not demonstrated a statistically significant difference in mortality
between the ECMO and control groups. However, given the difference in mortality
between groups, the data was reevaluated using a Bayesian analysis to add to study
interpretation [5]. In clinical trials evaluated using a traditional frequentist analysis,
the study is considered positive if the evidence is enough to confidently reject the
null hypothesis. In contrast, the Bayesian method allows a probability to be assigned
to the hypothesis, incorporating prior beliefs and prior data into the study interpreta-
tion. A range of subjective reference priors, from “strongly skeptical” to “strongly
enthusiastic” were used in modeling, along with data-derived prior distributions,
which were based on previous relevant studies of ECMO. After incorporation of
these “prior” probabilities, the study reported posterior probabilities of a mortality
benefit of ECMO. They found that the probability of a relative risk of less than one
(e.g., probability of a mortality benefit of early ECMO) was 88–99% across the
range of prior assumptions. For an absolute risk reduction (ARR) of mortality of 2%
or more, the posterior probabilities ranged from 78% to 98%, whereas for an ARR
of at least 20% (i.e., the prespecified anticipated mortality reduction), the posterior
probabilities were much lower, ranging from 0% to 2%. This elegant reanalysis
provides a wider context for interpreting the data from the EOLIA trial that may be
more clinically intuitive than the frequentist approach. It illustrates the high proba-
bility that ECMO lowers mortality in this context, and as the accompanying edito-
rial stated, the key questions left to be answered are “by how much does ECMO
work, in whom, and at what cost?” [16].
120 K. E. Melville et al.
Given the difficulty in enrolling patients for EOLIA, attributable, in part, to the
lack of clinical equipoise, which has only grown more among those performing
ECMO since the results of EOLIA, there will likely not be another large random-
ized trial of ECMO for ARDS. Therefore, following the publication of EOLIA,
composite studies have attempted to combine and better interpret all the existing
data we do have to quantify the overall benefit of ECMO.
A meta-analysis by Munshi et al. included five studies, with a total of 773
patients, and was the first to incorporate data from the two modern randomized
controlled trials of ECMO in severe acute respiratory failure (CESAR and EOLIA)
[6]. The other studies included were observational in nature with matching tech-
niques. Using the pooled data from the randomized controlled trials (429 patients),
the primary outcome of 60-day mortality was significantly lower for those receiving
ECMO (RR 0.73; 95% CI 0.58–0.92) than for those that did not. When data from all
five of the studies were analyzed together, there was also a significantly lower
30-day mortality in the ECMO group (RR 0.69; 95% CI 0.5–0.95). The meta-
analysis did not pool adverse events given inconsistency in the reporting across
studies.
As discussed previously, over the time that these trials have been performed,
ventilator management has also improved in an attempt to limit VILI, and strategies
using low tidal volumes, low airway pressures, high PEEP, and prone positioning
have become standard to varying extents. Given that many of these studies, and
other randomized controlled trials of therapies in ARDS, do not directly compare all
possible modalities commonly used in the treatment of ARDS, network meta-
analyses are useful to quantify the relative association of different management
strategies on mortality. Sud et al. compared the effects of low tidal volume ventila-
tion, prone positioning, high PEEP, and venovenous ECMO, among others, on in-
hospital mortality in ARDS [7]. Analyzing 34 randomized controlled trials, with
9085 patients, they found that prone positioning combined with low tidal volume
ventilation yielded the best outcomes (RR 0.74, 95% CI 0.6–0.92), but with veno-
venous ECMO also rating highly (RR 0.78, 95% CI 0.58–1.05) when compared to
low tidal volume ventilation. The data used in the latter comparison comes from
EOLIA and, therefore, includes only those patients with very severe ARDS [3].
Data from the 429 patients in EOLIA and CESAR were then combined in an
individual patient data meta-analysis, looking for differences in 90-day mortality
between conventional treatment and VV ECMO [4]. This revealed a relative risk of
death of 0.75 in the ECMO group, (95% CI 0.6–0.94, p = 0.013) as 36% of those in
the ECMO group died, compared with 48% in the control group. The relative risk
of treatment failure (defined as death in the ECMO group and death or crossover to
ECMO in the control group) was 0.65 (95% CI 0.52–0.8). Additionally, those ran-
domized to ECMO had more ventilator-free days, more days out of the intensive
care unit, and more days without vasopressors, renal replacement therapy, and neu-
rologic failure than those in the control group.
Taken together, there is compelling evidence to suggest that early initiation of
venovenous ECMO for severe ARDS leads to reduction in mortality compared with
conventional management. The use of “ultra-lung-protective” ventilation with
5 Hypoxemic Respiratory Failure: Evidence, Indications, and Exclusions 121
ECMO and the minimization of VILI likely drive a considerable proportion of the
improved outcomes seen with ECMO [17].
The role of ECMO in ARDS related to emerging infectious diseases is evolving.
As discussed previously, a relative surge in very severe ARDS case volume during
the influenza A(H1N1) pandemic in 2009 greatly increased the use and interest in
ECMO for respiratory failure. ECMO was also used for severe cases of ARDS dur-
ing the outbreak of Middle East Respiratory Syndrome (MERS) in 2012, but that
data is more limited, especially given comparatively fewer cases. ECMO centers
began using ECMO for severe ARDS due to coronavirus disease 2019 (COVID-19)
near the onset of the pandemic in 2020. In 2021, a systematic review and meta-
analysis was published, using observational studies of ECMO in adults with
COVID-19 ARDS over the first year of the pandemic [18]. Twenty-two studies,
with a total of 1896 patients, were included, and almost all cases (98.6%) used
venovenous ECMO. The primary outcome of interest was in-hospital mortality, and
secondary outcomes included duration of ECMO and mechanical ventilation, abil-
ity to wean from ECMO, and complications during therapy. In the studies that
reported pre-initiation PaO2:FiO2 ratios (1344 patients), the mean was 67.8. Prior to
ECMO initiation, incidence of prone positioning was 85.3%, and 96.3% of the
patients had received neuromuscular blocking agents. The in-hospital mortality of
those patients receiving venovenous ECMO was 35.7% (95% CI 30.7–40.7). Meta-
regression found that both age and ECMO duration were associated with increased
mortality, whereas increasing BMI seemed protective. In the full cohort, mean dura-
tion of ECMO (18 studies, 1844 patients) was 15.1 days, mean intensive care unit
length of stay was 29 days, and liberation from ECMO therapy was accomplished
in 67.6% of the cases. Two of the studies included in this analysis compared mortal-
ity rates between patients receiving ECMO with those receiving conventional ther-
apy with mechanical ventilation. Mortality rates of those placed on ECMO were
46.2% and 57.1% compared with 47.8% and 63.2%, respectively. As experience
increases for this indication, appropriate patient selection criteria may also be
refined. The use of ECMO in a potentially strained and resource-limited environ-
ment, such as that encountered during a pandemic, also raises ethical and logistical
issues that need to be navigated thoughtfully [19–23].
Indications
Exclusions
Conclusion
The use of venovenous ECMO as a therapy for severe ARDS is now supported by
high-level evidence [3–7]. Increased experience and improvements in technology,
coupled with appropriate patient selection and a better understanding of the mecha-
nism of benefit, have helped to improve outcomes.
References
1. Brodie D, Slutsky AS, Combes A. Extracorporeal life support for adults with respiratory fail-
ure and related indications: a review. JAMA. 2019;322(6):557–68.
2. Combes A, Schmidt M, Hodgson CL, Fan E, Ferguson ND, Fraser JF, et al. Extracorporeal
life support for adults with acute respiratory distress syndrome. Intensive Care Med.
2020;46(12):2464–76.
3. Combes A, Hajage D, Capellier G, Demoule A, Lavoue S, Guervilly C, et al. Extracorporeal
membrane oxygenation for severe acute respiratory distress syndrome. N Engl J Med.
2018;378(21):1965–75.
4. Combes A, Peek GJ, Hajage D, Hardy P, Abrams D, Schmidt M, et al. ECMO for severe
ARDS: systematic review and individual patient data meta-analysis. Intensive Care Med.
2020;46(11):2048–57.
5. Goligher EC, Tomlinson G, Hajage D, Wijeysundera DN, Fan E, Juni P, et al. Extracorporeal
membrane oxygenation for severe acute respiratory distress syndrome and posterior prob-
ability of mortality benefit in a post hoc Bayesian analysis of a randomized clinical trial.
JAMA. 2018;320(21):2251–9.
6. Munshi L, Walkey A, Goligher E, Pham T, Uleryk EM, Fan E. Venovenous extracorporeal
membrane oxygenation for acute respiratory distress syndrome: a systematic review and meta-
analysis. Lancet Respir Med. 2019;7(2):163–72.
7. Sud S, Friedrich JO, Adhikari NKJ, Fan E, Ferguson ND, Guyatt G, et al. Comparative effec-
tiveness of protective ventilation strategies for moderate and severe acute respiratory distress
syndrome. A network meta-analysis. Am J Respir Crit Care Med. 2021;203(11):1366–77.
8. Hill JD, O’Brien TG, Murray JJ, Dontigny L, Bramson ML, Osborn JJ, et al. Prolonged extra-
corporeal oxygenation for acute post-traumatic respiratory failure (shock-lung syndrome). Use
of the Bramson membrane lung. N Engl J Med. 1972;286(12):629–34.
9. Zapol WM, Snider MT, Hill JD, Fallat RJ, Bartlett RH, Edmunds LH, et al. Extracorporeal
membrane oxygenation in severe acute respiratory failure. A randomized prospective study.
JAMA. 1979;242(20):2193–6.
10. Morris AH, Wallace CJ, Menlove RL, Clemmer TP, Orme JF Jr, Weaver LK, et al. Randomized
clinical trial of pressure-controlled inverse ratio ventilation and extracorporeal CO2 removal
for adult respiratory distress syndrome. Am J Respir Crit Care Med. 1994;149(2 Pt 1):295–305.
11. Peek GJ, Mugford M, Tiruvoipati R, Wilson A, Allen E, Thalanany MM, et al. Efficacy and
economic assessment of conventional ventilatory support versus extracorporeal membrane
oxygenation for severe adult respiratory failure (CESAR): a multicentre randomised con-
trolled trial. Lancet. 2009;374(9698):1351–63.
12. Davies A, Jones D, Bailey M, Beca J, Bellomo R, Blackwell N, et al. Extracorporeal
membrane oxygenation for 2009 influenza a(H1N1) acute respiratory distress syndrome.
JAMA. 2009;302(17):1888–95.
13. Davies A, Jones D, Gattas D. Extracorporeal membrane oxygenation for ARDS due to 2009
influenza a(H1N1)—reply. JAMA. 2010;303(10):941–2.
124 K. E. Melville et al.
14. Miller RR 3rd, Markewitz BA, Rolfs RT, Brown SM, Dascomb KK, Grissom CK, et al.
Clinical findings and demographic factors associated with ICU admission in Utah due to novel
2009 influenza A(H1N1) infection. Chest. 2010;137(4):752–8.
15. Noah MA, Peek GJ, Finney SJ, Griffiths MJ, Harrison DA, Grieve R, et al. Referral to an
extracorporeal membrane oxygenation center and mortality among patients with severe 2009
influenza A(H1N1). JAMA. 2011;306(15):1659–68.
16. Lewis RJ, Angus DC. Time for clinicians to embrace their inner Bayesian?: reanalysis of results
of a clinical trial of extracorporeal membrane oxygenation. JAMA. 2018;320(21):2208–10.
17. Abrams D, Schmidt M, Pham T, Beitler JR, Fan E, Goligher EC, et al. Mechanical ventilation
for ARDS during extracorporeal life support: research and practice. Am J Respir Crit Care
Med. 2019;201(5):514–25.
18. Ramanathan K, Shekar K, Ling RR, Barbaro RP, Wong SN, Tan CS, et al. Extracorporeal
membrane oxygenation for COVID-19: a systematic review and meta-analysis. Crit Care.
2021;25(1):211.
19. Supady A, Badulak J, Evans L, Curtis JR, Brodie D. Should we ration extracorporeal mem-
brane oxygenation during the COVID-19 pandemic? Lancet Respir Med. 2021;9(4):326–8.
20. Supady A, Brodie D, Curtis JR. Ten things to consider when implementing rationing guide-
lines during a pandemic. Intensive Care Med. 2021;47(5):605–8.
21. Supady A, Curtis JR, Abrams D, Lorusso R, Bein T, Boldt J, et al. Allocating scarce inten-
sive care resources during the COVID-19 pandemic: practical challenges to theoretical frame-
works. Lancet Respir Med. 2021;9(4):430–4.
22. Abrams D, Lorusso R, Vincent JL, Brodie D. ECMO during the COVID-19 pandemic: when
is it unjustified? Crit Care. 2020;24(1):507.
23. Badulak J, Antonini MV, Stead CM, Shekerdemian L, Raman L, Paden ML, et al. Extracorporeal
membrane oxygenation for COVID-19: updated 2021 guidelines from the extracorporeal life
support organization. ASAIO J. 2021;67(5):485–95.
24. Abrams D, Ferguson ND, Brochard L, Fan E, Mercat A, Combes A, et al. ECMO for ARDS:
from salvage to standard of care? Lancet Respir Med. 2019;7(2):108–10.
Chapter 6
Ventilator Management During ECLS
“The wisdom of old men. They do not grow wise. They grow careful.”
From Farewell to arms
by Ernest Hemingway
Introduction
In this chapter we will discuss the ventilatory management during ECLS. We will
focus almost exclusively on acute hypoxemic respiratory failure, but the principles
here exposed are by and large applicable to hypercapnic respiratory failure too.
Conversely, we will sometimes introduce concepts that, though derived from the
experience of ECLS in chronic respiratory failure patients, we presume applicable
to the acute patient as well.
The reader will forgive us for taking first a historical approach. Extracorporeal
membrane oxygenation (ECMO) got to the stage in clinical medicine when in 1974
the National Institutes of Health (NIH) funded the first prospective controlled ran-
domized clinical study on ARDS [1]. It aimed at comparing the outcome of severe
acute respiratory failure patients treated by veno-arterial (VA) ECMO with the out-
come of patients treated conventionally by mechanical ventilation. The results of
the study were discouraging, showing the same high mortality (exceeding 90%) in
both arms [1]. In discussing their results, Zapol and colleagues identified, among
others, two possible reasons for the study failure: (a) the VA mode, that could have
increased the incidence of pulmonary vascular thrombosis, and (b) the decrease in
tidal volume and respiratory rate applied in the ECMO group which might have
caused a worsening of respiratory compliance. However, following the institution of
ECLS, tidal volume was decreased only to a little extent, as it went from approxi-
mately 800 mL to an average of approximately 600 mL. Moreover, this decrease
was associated with a constantly high plateau pressure, in a range between 40 and
50 cmH2O [2]. At that time, indeed, the recommended tidal volume was 10–15 mL/
kg or more, as it was one of the major determinants of oxygenation [3, 4]. Taking an
opposite position, in 1981, Ted Kolobow wrote a provocative editorial: “Why did
ECMO fail?” [5] in which he suggested that “severely diseased lungs have a chance
to heal only if the environment remains conducive to the healing of the lungs. This
environment does not consist of high airway pressures, high tidal volumes, high
PEEP, high FiO2, or a severe pulmonary hypoperfusion with severe and lethal lung
tissue alkalosis” [5]. He had developed in his laboratory at the NIH a prototype that
became the first commercially available membrane lung, and he was one of the most
experienced scientists in this field [6]. He pointed out that providing viable blood
gases by extracorporeal means was not enough to foster lung healing. The ventila-
tory treatment should be reset to take into account that the native lung management
was no longer dictated by the need to maximize gas exchange. High airway pres-
sures and high tidal volume, a must to “optimize” gas exchange through the native
lung, had to be seen as the cause of what later became known as ventilator induced
lung injury (VILI) [7].
The goal of ECMO was redefined from “buying time for the lung to heal” [8] to
“rest the lung” [9]. In the same years, Gattinoni et al. introduced in clinical use a
novel ECLS technique called LFPPV ECCO2R (low-frequency positive pressure
ventilation with extracorporeal CO2 removal) [10]. The stage was set to allow a new
approach to the ventilator management of the patients undergoing ECLS.
ECLS evolved from pure hypoxia rescue therapy to a strategy that, by exploiting
the physiological advantages provided by extracorporeal gas exchange, fosters lung
healing and VILI prevention while avoiding systemic biotrauma and multiple organ
failure [9].
Thus, over the last 30 years, the focus of ECLS treatment of ARDS shifted
toward optimal ventilation management during ECLS. Still, data on how to best
ventilate ARDS patients undergoing ECLS are scanty and mainly come from obser-
vational or physiological studies.
Oxygenation
ECLS was born as ECMO [11], and oxygenation is still the major focus of the tech-
nique. Oxygen transfer (to the extracorporeal blood) for any given membrane lung
will depend on blood flow, on hemoglobin (Hb) concentration, and on input Hb
oxygen saturation. For any given (fixed) intrapulmonary shunt (proportion of the
cardiac output that crosses the lung and leaves it with an unchanged oxygen content
equal to the mixed venous one), veno-venous ECMO will increase the arterial oxy-
gen content according to the change it is able to operate in the mixed venous oxygen
content. If a patient has a 50% intrapulmonary shunt fraction and a mixed venous
oxygen saturation (SvO2) of 50%, then the arterial oxygen saturation (SaO2) will
roughly be 75% (given an arteriovenous difference of 25%). If by oxygenating a
substantial (say 50%) proportion of venous blood ECMO can raise SvO2 to 75%,
then SaO2 will now be the average between 75% and 100%, that is, approximately
87.5% (approximation is due to not taking into account the dissolved O2). It becomes
obvious that the higher the ratio (extracorporeal blood flow/cardiac output), the
higher the effect on the mixed venous and therefore on arterial oxygenation (this is
one of the reasons to advocate β-blocking in certain VV bypasses) [12].
Real life is much more complicated than that, mainly because of the negative
effect of bypass recirculation (which increases with increasing blood flow) and
because intrapulmonary shunt is not fixed, but tends to increase substantially with
the increases in SvO2, in part because of the release of any residual hypoxic
vasoconstriction.
Our ability to provide oxygen to the patient is moreover limited by the fact that
we can add only a limited amount of oxygen to blood, due the concentration of
hemoglobin and to the fact that patients normally have, and probably require, a
mixed venous oxygen saturation around 70%.
A basic principle, described fully in Chap. 1, is that oxygenation is mainly
dependent on blood flow and requires extracorporeal flows very similar to the nor-
mal cardiac output.
Carbon Dioxide
In contrast, CO2 removal requires much lower blood flows [13]. This is due to the
very high carbon dioxide content of blood (mostly carried as bicarbonate ion). We
can expect the carbon dioxide content of normal venous blood to be approximately
600 mL/L and some 50% more in compensated chronic hypercapnia. This means
that one liter of blood contains two to three times the total CO2 produced by the
entire body metabolism in a minute. Therefore, we can assume that total CO2
removal (that is the removal of the minute CO2 production) can be achieved at a
blood flow in the range of 0.5 L/min, as opposed to the 5 L/min required to provide
the 250 mL/min physiological oxygen consumption.
128 A. Pesenti et al.
0
0 0.5 1
R
In the second half of the 1970s, the pioneering work of Kolobow and Gattinoni
provided the basic pathophysiological foundations to suggest a possible clinical role
for ECCO2R.
This extraordinary pair of scientist-inventor and scientist-clinician realized that the
membrane oxygenator was indeed a membrane lung, exchanging both oxygen and
CO2. In a little more than a year, they described first a membrane lung optimized for
CO2 removal (the so-called CDML: the carbon dioxide membrane lung) [14], and
then they showed that by removing CO2 at incremental rates, it was possible to control
the ventilation of an awake sheep [15], down to complete apnea [16], and developed a
mode of ventilation that would maintain lung volume while using very low respiratory
rates [17]. They clarified that in apneic oxygenation alveolar PO2 is a function of
alveolar PN2, in turn at equilibrium with the PN2 of the gas ventilating the membrane
lung [16]. Another observation, particularly important in the spontaneously breathing
subject, defines the role of the ratio between the amount of CO2 eliminated and the
amount of oxygen consumed by the natural lung (respiratory quotient, RQ) in deter-
mining alveolar PO2. They stressed the need of increasing FiO2 to maintain a constant
FaO2 when the alveolar gas equation is faced with RQs much lower than 1, as is the
case when the amount of CO2 removed by the natural lung is decreased by the amount
removed by the membrane lung (Fig. 6.1). When no CO2 is exchanged through the
natural lung (apneic oxygenation), then FaO2 is maintained constant only by FiO2 1,
and the role of the membrane lung PN2 becomes predominant [18].
While ECLS has been mainly considered as a rescue treatment for refractory hypox-
emia [11], the considerable amount of CO2 removed by the artificial lung allows
decreasing dramatically the load mechanical ventilation imposes on the natural lung
6 Ventilator Management During ECLS 129
Table 6.1 Guidelines proposed by different ECMO centers to guide ventilation settings in the
early ECMO phase
Tidal volume RR PEEP
Guidelines Ventilation mode (mL/kg) (1/min) (cmH2O) FiO2
ELSO [20] PCV To reach Ppeak of 20 4–5 10 Not
cmH2O reported
CESAR trial PCV To reach Ppeak of 10 10–15 0.3
[21] 20–25 cmH2O
Karolinska PCV or PSV To reach Ppeak of Not 5–10 0.4
[22] 20–25 cmH2O reported
EOLIA trial PCV or VCV assist To reach Pplat <20 15–30 ≥10 0.3–0.6
[23] control cmH2O
RR respiratory rate, PEEP positive end-expiratory pressure, FiO2 inspired O2 fraction, PCV pres-
sure controlled ventilation, PSV pressure support ventilation, VCV volume controlled ventilation,
Pplat airway plateau pressure
Three major goals are important in supporting the severely hypoxemic ARDS
patient: rescue from hypoxia, lung recruitment (open lung strategy), and VILI pre-
vention (lung rest). During conventional mechanical ventilation, a high price is paid
in terms of airway pressures (PEEP, plateau, and mean airway pressure) to maintain
viable arterial oxygenation, while sufficient alveolar ventilation has to be main-
tained to eliminate CO2. Veno-venous ECMO is capable of providing the entire
oxygen consumption, even when the native lung oxygen transfer is nil (100% intra-
pulmonary shunt); under these conditions, though, blood flow must be maximized,
and relatively low arterial oxygen saturation levels are accepted, sometimes around
80%. While these levels of oxygenation might not be better than the one the patient
could achieve before bypass, the good side of ECMO is that oxygenation can be
maintained while the lung is being managed under “protective” ventilatory settings.
However, a compromise is often recommended (Table 6.1), mostly because some
uncertainties exist as to whether a recruited lung will really heal faster than a col-
lapsed one. Airway pressures are set at compromise values, with the aim of main-
taining a reasonable recruitment status and favoring the oxygenation function of the
native lung, hence maximizing arterial oxygen saturation. A rather different
130 A. Pesenti et al.
Table 6.2 Main ventilator settings from three different published case series: consensus during
the early ECMO phase is lacking
Tidal volume RR PEEP
(mL/kg) (1/min) (cmH2O) FiO2
Brogan et al. [27] Before ECLS Not Reported 20 [15;25] 12 [10;17] 1 [1;1]
24 h of ECLS Not Reported 10 [10;15] 10[8;14] 0.5 [0.4;0.5]
Zimmermann et al. [28] Before ECLS 6.6 [5.3;7.2] 25 [22;27] 17 [14;20] 1 [0.8;1]
24 h of ECLS 4.4 [3.4;5.4] 21 [18;26] 17 [14;20] 0.7 [0.6;0.9]
Patroniti et al. [29] Before ECLS 6.2 [4.7;7.7] 28 [20;33] 16 [14;19] 1 [1;1]
24 h of ECLS 4.6 [3.0;6.3] 10 [8;12] 16 [14;19] 0.6 [0.4;0.8]
RR respiratory rate, PEEP positive end-expiratory pressure, FiO2 inspired O2 fraction, ECLS extra-
corporeal lung support
Prone Positioning
HFOV combines a bias flow and an oscillating piston to deliver tidal volumes lower
than the anatomical dead space (around 1–2 mL/kg) at very high respiratory fre-
quency (usually 3–6 Hz in adults) [51]. The rapid oscillations of gas are delivered
above and below a constant mean airway pressure (Paw), usually set 5 cmH2O
higher than the level achieved during conventional ventilation. Oxygenation during
HFOV depends on the set level of Paw and FiO2, while CO2 removal depends on the
pressure amplitude and frequency of oscillation [51]. Compared to conventional
ventilation, the delivery of small tidal volumes should limit alveolar overdistension,
and the use of higher Paw should promote alveolar recruitment while avoiding
cyclical alveolar collapse, thus improving gas exchange and maintaining the goal of
lung protection.
HFOV is extensively used in neonates with respiratory distress syndrome [52],
while the experience in adult ARDS patients is more limited and more
controversial.
Recently, the results of two large, multicenter, randomized trials comparing
HFOV to conventional ventilation as a first-line ventilator strategy in adult ARDS
patients have been published. The Oscillation for Acute Respiratory Distress
134 A. Pesenti et al.
Syndrome Treated Early (OSCILLATE) trial was stopped after inclusion of 548 of
a planned 1200 patients because the mortality of patients treated with HFOV was
significantly higher than the control group (47% vs 35%); in addition, patients
assigned to HFOV received higher doses of sedatives, neuromuscular blocking
agents, and vasoactive drugs [53]. The Oscillation for ARDS (OSCAR) trial, which
included a total of 795 patients, failed to show any difference in 30-day mortality
between HFOV and conventional ventilation (41.7% vs 41.1%) [54]. Despite some
important methodological differences, the results of these trials argue against a
widespread application of HFOV in ARDS patients. These two studies however
enrolled patients in whom oxygenation was far from the level that would have quali-
fied them for an oxygenation rescue maneuver, and it is doubtful that we should
disqualify at this time HFOV as a possible oxygenation rescue procedure.
Literature on the application of HFOV during ECMO are scarce and limited to
few case reports. Banach et al. described the case of a 35-year-old man with lobar
pneumonia undergoing ECMO for refractory hypoxemia who required discontinu-
ation of the extracorporeal support because of hemorrhagic complications: ECMO
was then substituted with HFOV coupled with pumpless arteriovenous extracorpo-
real lung assist (PECLA) for CO2 removal [55].
Finally, Muellenbach et al. showed in an animal model [56] and in a human case
of post-traumatic ARDS [57] that the combination of an arteriovenous extracorpo-
real lung assist (AV-ECLA) with HFOV allowed the use of oscillatory frequencies
higher than those usually applied (up to 10–15 Hz), thus minimizing the risk of
barotrauma and volutrauma.
Management of Pneumothoraces
Following the phase of initial ECMO application and when the patient’s conditions
are more stable, mechanical ventilation is switched from controlled to assisted
mode. The aim is then a progressive decrease in support, ECLS discontinuation,
and, finally, patient’s extubation. In particular, early switch to protective-assisted
MV may improve respiratory muscle function and gas exchange, decrease the need
6 Ventilator Management During ECLS 135
for sedation, and aid weaning from ventilator [58]. These changes, though dictated
mainly by the individual patient evolution (some patients stay days on bypass, but a
few stay months), are however sought for at different times by different groups. In
recent years, important technological advances contributed to significantly improve
the safety of ECLS circuits’ use. This led many investigators to a change in strategy,
sometimes liberating the patient first from the ventilator and later from bypass. It has
then become possible to report patients being extubated while still on bypass (awake
ECMO) [59]. We have not adopted this approach but feel safe to recommend initiat-
ing weaning (i.e., assisted breathing rather than controlled) as soon as possible while
assuring patients’ safety and comfort. While most groups delay the application of
assisted ventilation until the native lung performance improves substantially, others
almost immediately switch the ventilator to pressure support mode [60]. Namely,
Karolinska group reported their experience in 17 adult ARDS patients treated by
ECLS between 1995 and 1999. They discontinued muscle relaxation and set MV to
PSV right after starting ECLS. Sedation targets were raised, and within a few hours,
patients were only mildly sedated. They report that patients remained awake during
the day, able to interact with staff and family, and participate in quiet activities like
watching TV. Spontaneous assisted breathing was enhanced and maintained, when
needed, by adding 5% CO2 to the gas ventilating the membrane lung. Assisted
mechanical ventilation mode was pressure support ventilation (PSV) for all patients;
no data are reported to indicate how PSV was titrated during ECLS. We know that
tidal volume was reduced from a range of 450–947 mL before ECLS start to
≈122–662 mL during extracorporeal treatment. Nonetheless, patients’ mortality
(24%) was surprisingly low given their baseline severity. Thus, from this study we
can conclude that in severe ARDS patients ECLS treatment coupled with minimal
sedation and PSV with low tidal volumes may be associated with high survival.
Our group previously showed that PSV may be difficult to implement in ARDS
patients with very low respiratory system compliance (Crs), likely because peak
inspiratory flow is reached rapidly and the flow-based expiratory phase of PSV
starts while patient is still inspiring (premature expiratory cycling) [61]. Thus, these
patients are at high risk of patient-ventilator asynchrony. Asynchrony is a serious
threat, as higher asynchrony is associated with iatrogenic injury and delayed wean-
ing from the ventilator. To this end, we compared PSV with neurally adjusted ven-
tilator assist (NAVA) [62] in a group of severe ARDS patients with Crs ranging
between 7 and 31 mL/cmH2O undergoing veno-venous ECMO and low tidal vol-
ume (i.e., 3–4 mL/kg) PSV. We could show that, in this selected group of patients,
the low Crs values coupled to short inspiratory time led to premature expiratory
cycling and high patient-ventilator asynchrony. NAVA was associated with improved
patient-ventilator interaction compared to PSV, the more so in patients with the low-
est Crs values. More recently, we also evaluated the respiratory pattern of severe
ARDS patients switched to assisted ventilation as per clinical decision when
increasing gas flows were applied (i.e., at increasing rate of extracorporeal CO2
removal) [63]. The respiratory drive and effort were effectively controlled at higher
gas flow rates, suggesting that after switch to assisted ventilation, careful titration of
the extracorporeal CO2 removal rate may be critical for success.
136 A. Pesenti et al.
Other modes of assisted ventilation have been applied during ECMO. Besides
CPAP, we believe we should mention airway pressure release ventilation since this
particular mode of ventilation, alternating two pressure levels at fixed inspiratory/
expiratory time, allows spontaneous breathing at all times during the respiratory
cycle, hence avoiding respiratory muscle atrophy and favoring optimal distribution
of the inspired tidal volume.
Institution of ECMO to avoid intubation represents the most advanced frontier for
assisted/spontaneous breathing during ECLS. Reports on this approach in bridge to
lung transplant patients (a few of whom have been affected by severe ARDS) are so
many that it may be considered as standard of care. For example, Fuehner et al.
recently published a retrospective analysis of 26 patients treated with ECLS before
intubation at the onset of end-stage respiratory failure awaiting lung transplant [64].
Of these, 16 patients did not require intubation and were transplanted after a median
time of 11 days on ECMO support. Two of these patients died from septic multior-
gan failure after transplantation. Another patient died from lung cancer 60 days after
transplantation. The remaining 13 patients were discharged from the hospital
20–87 days after transplantation, and all of these patients remained alive during the
follow-up period (7–39 months). These authors also selected a group of patients
who underwent intubation and MV and no ECMO while waiting transplant: of
them, those who survived until hospital discharge required more ventilator days,
spent a longer time in the ICU, and were discharged later from hospital [64]. The
overall survival of the intention-to-treat populations at 6 months after transplant was
62% in the awake ECMO group and 35% in the MV group (P = 0.05). Considering
only patients who reached transplantation, the 6-month post-transplant survival
rates were 80% in the awake ECMO group and 50% in the MV group, respectively
(P = 0.02). Thus, ECMO should be considered before intubation in patients with
end-stage respiratory failure awaiting lung transplant but only in highly specialized
centers with prepared and dedicated staff.
Is awake ECMO approach feasible in severe ARDS patients? Dr. Hoeper and
colleagues reported the results of a single-center, uncontrolled pilot trial designed to
assess the feasibility of veno-venous ECMO in awake, non-intubated, spontane-
ously breathing patients with ARDS [65]. They enrolled six patients, four of whom
were immunocompromised. Three patients were weaned from ECMO without
being intubated and were subsequently discharged from hospital alive, while the
other three required intubation and invasive MV (and two of them died in hospital).
The authors concluded that an “awake ECMO” strategy appears feasible in selected
patients with ARDS and deserves further evaluation as a potential alternative to
intubation and mechanical ventilation. More recently, another clinical study [66]
showed that a strategy of early extubation during ECMO in severe ARDS was asso-
ciated with 50% success (i.e., ECMO weaning without re-intubation). Interestingly,
6 Ventilator Management During ECLS 137
the patients successfully treated by awake ECMO had higher lung weight measured
by CT scan. Subsequent physiological study confirmed the correlation between dif-
ficult control of respiratory drive by ECMO and higher lung weight [67].
Avoiding intubation particularly in specific situations, like ARDS in immuno-
compromised patients, might become a major goal of ECMO. Careful selection of
candidates is necessary to individualize the bypass technique to be applied and the
optimal ventilatory management. The field is very promising, but systematic clini-
cal experience is still lacking.
Ventilatory
PS CPAP T-tube CPAP
Mode (+22 cm H2O)(3 cm H2O) (3 cm H2O)
FIO2 100
(%)
0
100
PaO2 80
(mmHg)
60
PaCO2 40
(mmHg)
20
PECOR 0
(%) 0 1 2 3 4 5 6 7 PECOR days
Fig. 6.2 Data from the first partial extracorporeal low-flow partial CO2 removal (PECOR) clinical
case: a patient with high flow bilateral airleaks, after institution of PECOR, could promptly be
switched from high-level PSV to CPAP and finally extubated with complete resolution of the air
leak. In this case cannulation was percutaneous, blood flow ranged between 0.4 and 0.6 l/min, and
CO2 removal between 22% and 40% and the total patient VCO2. (Modified from Ref. [74])
As already stated, Kolobow and Gattinoni showed how ECCO2R could control
spontaneous breathing and the need for mechanical ventilation in sheep [15, 18].
When we faced the first ARDS patients recovering from severe ARDS on ECLS, we
could show that ARDS patients undergoing CPAP ventilation while on bypass could
decrease their ventilation according to the amount of CO2 removed by the mem-
brane lung [75]. The same effect has been recently observed in ARDS patients being
ventilated using NAVA while on ECMO [76].
These data however, while supporting the possibility of targeting ventilatory
needs and respiratory drive in ARDS patients, should be interpreted with caution.
Experience shows that in the early acute interstitial edema phase, ARDS patients
have a very high respiratory drive and are severely dyspneic and tachypneic. In the
early phase of severe ARDS, blood gases represent a minor portion of the respira-
tory drive. In the clinical setting of ECMO in severe ARDS patients, it is not uncom-
mon to verify that even when PaCO2 is brought down to very low levels
(25–28 mmHg), the respiratory rate is still very high, and the inspiratory effort may
still generate very deep negative intrathoracic pressures and very high transpulmo-
nary pressure, carrying a high risk of barotrauma and VILI in spite of ECCO2R
(Fig. 6.3). These observations contrast with what has been reported in acutely
decompensated COPD patients, in whom respiratory drive (and the need for intuba-
tion) can be effectively controlled by ECCO2R [77–78].
Which is then the possible role of ECCO2R in ARDS patients? Most important at
this time is the ultra-protective ventilator approach proposed by Terragni et al. in
2009 [79]. This group of investigators showed that 25–30% of ARDS patients,
140 A. Pesenti et al.
Fig. 6.3 Airway pressure (Paw), esophageal pressure (Pes), and electrical activity of the dia-
phragm (EAdi) tracings from one representative patient undergoing ECMO and pressure support
ventilation: ECMO sweep gas flow (GF) was gradually decreased from 4 L/min (100%) to 2 L/min
to 0 L/min (0%). Note the increasing esophageal pressure swings at GF 0%. (Bellani, Grasselli,
Mauri, Pesenti: unpublished data)
Conclusions
The concept of VILI and the need for protective ventilation have gained widespread
acceptance, having been clearly validated in the clinical field. The practical applica-
tion of these concepts in the patient undergoing ECLS, though based on solid patho-
physiological grounds, lacks evidence to allow the drawing of guidelines or
recommendations. Clinical experience suggests two alternative approaches, possi-
bly applicable at different times in the same patient: One looks for optimized lung
recruitment by a higher PEEP level, and the other privileges lung rest in spite of the
risk of substantial lung collapse. Both approaches require a third essential element:
time and with it the patience necessary for the lung to recover.
References
1. Zapol WM, Snider MT, Hill JD, Fallat RJ, Bartlett RH, Edmunds LH, Morris AH, Peirce
EC 2nd, Thomas AN, Proctor HJ, Drinker PA, Pratt PC, Bagniewski A, Miller RG Jr.
Extracorporeal membrane oxygenation in severe acute respiratory failure. A randomized pro-
spective study. JAMA. 1979;242(20):2193–6.
2. National Heart, Lung, and Blood Institute: Division of Lung Disease. Section four: hemo-
dynamics, gas exchange and mechanics. In: Extracorporeal support for respiratory insuffi-
ciency. A collaborative study in response to RFP-NHLI-73-20. Ed. US Department of Health,
Education and Welfare. Washington, DC, 1979.
3. Pontoppidan H, Geffin B, Lowenstein E. Acute respiratory failure in the adult. 3. N Engl J
Med. 1972;287(16):799–806.
4. Suter PM, Fairley HB, Isenberg MD. Effect of tidal volume and positive end-expiratory pres-
sure on compliance during mechanical ventilation. Chest. 1978;73(2):158–62.
5. Kolobow T, Solca M, Gattinoni L, Pesenti A. Adult respiratory distress syndrome (ARDS):
why did ECMO fail? Int J Artif Organs. 1981;4(2):58–9.
6. Kolobow T, Bowman RL. Construction and evaluation of an alveolar membrane artificial
heart-lung. Trans Am Soc Artif Intern Organs. 1963;9:238–43.
7. Dreyfuss D, Saumon G. Ventilator-induced lung injury: lessons from experimental studies. Am
J Respir Crit Care Med. 1998;157(1):294–323.
8. Zapol WM, Kitz RJ. Buying time with artificial lungs. N Engl J Med. 1972;286(12):657–8.
9. Gattinoni L, Agostoni A, Pesenti A, Pelizzola A, Rossi GP, Langer M, Vesconi S, Uziel L, Fox
U, Longoni F, Kolobow T, Damia G. Treatment of acute respiratory failure with low-frequency
positive-pressure ventilation and extracorporeal removal of CO2. Lancet. 1980;2(8189):292–4.
10. Gattinoni L, Kolobow T, Agostoni A, Damia G, Pelizzola A, Rossi GP, Langer M, Solca M,
Citterio R, Pesenti A, Fox U, Uziel L. Clinical application of low frequency positive pressure
ventilation with extracorporeal CO2 removal (LFPPV-ECCO2R) in treatment of adult respira-
tory distress syndrome (ARDS). Int J Artif Organs. 1979;2(6):282–3.
142 A. Pesenti et al.
11. Hill JD, O'Brien TG, Murray JJ, Dontigny L, Bramson ML, Osborn JJ, Gerbode F. Prolonged
extracorporeal oxygenation for acute post-traumatic respiratory failure (shock-lung syn-
drome). Use of the Bramson membrane lung. N Engl J Med. 1972;286(12):629–34.
12. Guarracino F, Zangrillo A, Ruggeri L, Pieri M, Calabrò MG, Landoni G, Stefani M, Doroni L,
Pappalardo F. β-Blockers to optimize peripheral oxygenation during extracorporeal membrane
oxygenation: a case series. J Cardiothorac Vasc Anesth. 2012;26(1):58–63.
13. Gattinoni L, Pesenti A, Kolobow T, Damia G. A new look at therapy of the adult respiratory
distress syndrome: motionless lungs. Int Anesthesiol Clin. 1983;21(2):97–117.
14. Kolobow T, Gattinoni L, Tomlinson T, White D, Pierce J, Iapichino G. The carbon dioxide
membrane lung (CDML): a new concept. Trans Am Soc Artif Intern Organs. 1977;23:17–21.
15. Kolobow T, Gattinoni L, Tomlinson TA, Pierce JE. Control of breathing using an extracorpo-
real membrane lung. Anesthesiology. 1977;46(2):138–41.
16. Kolobow T, Gattinoni L, Tomlinson T, Pierce JE. An alternative to breathing. J Thorac
Cardiovasc Surg. 1978;75(2):261–6.
17. Gattinoni L, Kolobow T, Tomlinson T, Iapichino G, Samaja M, White D, Pierce J. Low-
frequency positive pressure ventilation with extracorporeal carbon dioxide removal (LFPPV-
ECCO2R): an experimental study. Anesth Analg. 1978;57(4):470–7.
18. Gattinoni L, Kolobow T, Tomlinson T, White D, Pierce J. Control of intermittent posi-
tive pressure breathing (IPPB) by extracorporeal removal of carbon dioxide. Br J Anaesth.
1978;50(8):753–8.
19. Brodie D, Bacchetta M. Extracorporeal membrane oxygenation for ARDS in adults. N Engl J
Med. 2011;365(20):1905–14.
20. ELSO Patient Specific guidelines. Available at http://www.elsonet.org/index.
php?option=com_phocadownload&view=category&id=4&Itemid=627.
21. Peek GJ, Mugford M, Tiruvoipati R, Wilson A, Allen E, Thalanany MM, Hibbert CL, Truesdale
A, Clemens F, Cooper N, Firmin RK, Elbourne D, CESAR trial collaboration. Efficacy and
economic assessment of conventional ventilator support versus extracorporeal membrane oxy-
genation for severe adult respiratory failure (CESAR): a multicentre randomised controlled
trial. Lancet. 2009;374(9698):1351–63.
22. Holzgraefe B, Broomé M, Kalzén H, Konrad D, Palmér K, Frenckner B. Extracorporeal
membrane oxygenation for pandemic H1N1 2009 respiratory failure. Minerva Anestesiol.
2010;76(12):1043–51.
23. Combes A, Hajage D, Capellier G, Demoule A, Lavoué S, Guervilly C, Da Silva D, Zafrani L,
Tirot P, Veber B, Maury E, Levy B, Cohen Y, Richard C, Kalfon P, Bouadma L, Mehdaoui H,
Beduneau G, Lebreton G, Brochard L, Ferguson ND, Fan E, Slutsky AS, Brodie D, Mercat A,
EOLIA Trial Group, REVA, and ECMONet. Extracorporeal membrane oxygenation for severe
acute respiratory distress syndrome. N Engl J Med. 2018;378(21):1965–75.
24. Mauri T, Bellani G, Foti G, Grasselli G, Pesenti A. Successful use of neurally adjusted ven-
tilatory assist in a patient with extremely low respiratory system compliance undergoing
ECMO. Intensive Care Med. 2011;37(1):166–7.
25. Bein T, Weber-Carstens S, Goldmann A, Müller T, Staudinger T, Brederlau J, Muellenbach R,
Dembinski R, Graf BM, Wewalka M, Philipp A, Wernecke KD, Lubnow M, Slutsky AS. Lower
tidal volume strategy (≈3 ml/kg) combined with extracorporeal CO2 removal versus ‘conven-
tional’ protective ventilation (6 ml/kg) in severe ARDS: the prospective randomized Xtravent-
study. Intensive Care Med. 2013;39(5):847–56.
26. Pham T, Combes A, Roze H, Chevret S, Mercat A, Roch A, Mourvillier B, Ara-Somohano C,
Bastien O, Zogheib E, et al. Extracorporeal membrane oxygenation for pandemic influenza
A(H1N1)-induced acute respiratory distress syndrome: a cohort study and propensity matched
analysis. Am J Respir Crit Care Med. 2013;187(3):276–85.
27. Brogan TV, Thiagarajan RR, Rycus PT, Bartlett RH, Bratton SL. Extracorporeal membrane
oxygenation in adults with severe respiratory failure: a multi-center database. Intensive Care
Med. 2009;35(12):2105–14.
6 Ventilator Management During ECLS 143
28. Zimmermann M, Bein T, Arlt M, Philipp A, Rupprecht L, Mueller T, Lubnow M, Graf BM,
Schlitt HJ. Pumpless extracorporeal interventional lung assist in patients with acute respiratory
distress syndrome: a prospective pilot study. Crit Care. 2009;13(1):R10.
29. Patroniti N, Zangrillo A, Pappalardo F, Peris A, Cianchi G, Braschi A, Iotti GA, Arcadipane
A, Panarello G, Ranieri VM, Terragni P, Antonelli M, Gattinoni L, Oleari F, Pesenti A. The
Italian ECMO network experience during the 2009 influenza A(H1N1) pandemic: preparation
for severe respiratory emergency outbreaks. Intensive Care Med. 2011;37(9):1447–57.
30. Pipeling MR, Fan E. Therapies for refractory hypoxemia in acute respiratory distress syn-
drome. JAMA. 2010;304(22):2521–7.
31. Patroniti N, Isgrò S, Zanella A. Clinical management of severely hypoxemic patients. Curr
Opin Crit Care. 2011;17(1):50–6.
32. Ullrich R, Lorber C, Röder G, Urak G, Faryniak B, Sladen RN, Germann P. Controlled air-
way pressure therapy, nitric oxide inhalation, prone position, and extracorporeal membrane
oxygenation (ECMO) as components of an integrated approach to ARDS. Anesthesiology.
1999;91(6):1577–86.
33. Bryan AC. Conference on the scientific basis of respiratory therapy. Pulmonary physiotherapy
in the pediatric age group. Comments of a devil’s advocate. Am Rev Respir Dis. 1974;110(6
Pt 2):143–4.
34. Guérin C, Reignier J, Richard JC, Beuret P, Gacouin A, Boulain T, Mercier E, Badet M, Mercat
A, Baudin O, Clavel M, Chatellier D, Jaber S, Rosselli S, Mancebo J, Sirodot M, Hilbert G,
Bengler C, Richecoeur J, Gainnier M, Bayle F, Bourdin G, Leray V, Girard R, Baboi L, Ayzac
L, PROSEVA Study Group. Prone positioning in severe acute respiratory distress syndrome. N
Engl J Med. 2013;368(23):2159–68.
35. Hemmila MR, Napolitano LM. Severe respiratory failure: advanced treatment options. Crit
Care Med. 2006;34(9 Suppl):S278–90.
36. Pelosi P, Brazzi L, Gattinoni L. Prone position in acute respiratory distress syndrome. Eur
Respir J. 2002;20(4):1017–28.
37. Galiatsou E, Kostanti E, Svarna E, Kitsakos A, Koulouras V, Efremidis SC, Nakos G. Prone
position augments recruitment and prevents alveolar overinflation in acute lung injury. Am J
Respir Crit Care Med. 2006;174(2):187–97.
38. Sud S, Friedrich JO, Taccone P, Polli F, Adhikari NK, Latini R, Pesenti A, Guérin C, Mancebo
J, Curley MA, Fernandez R, Chan MC, Beuret P, Voggenreiter G, Sud M, Tognoni G, Gattinoni
L. Prone ventilation reduces mortality in patients with acute respiratory failure and severe
hypoxemia: systematic review and meta-analysis. Intensive Care Med. 2010;36(4):585–99.
39. Abroug F, Ouanes-Besbes L, Dachraoui F, Ouanes I, Brochard L. An updated study-level
meta-analysis of randomized controlled trials on proning in ARDS and acute lung injury. Crit
Care. 2011;15(1):R6.
40. Hemmila MR, Rowe SA, Boules TN, Miskulin J, McGillicuddy JW, Schuerer DJ, Haft JW,
Swaniker F, Arbabi S, Hirschl RB, Bartlett RH. Extracorporeal life support for severe acute
respiratory distress syndrome in adults. Ann Surg. 2004;240(4):595–607.
41. Haefner SM, Bratton SL, Annich GM, Bartlett RH, Custer JR. Complications of intermittent
prone positioning in pediatric patients receiving extracorporeal membrane oxygenation for
respiratory failure. Chest. 2003;123(5):1589–94.
42. Goettler CE, Pryor JP, Hoey BA, Phillips JK, Balas MC, Shapiro MB. Prone positioning does
not affect cannula function during extracorporeal membrane oxygenation or continuous renal
replacement therapy. Crit Care. 2002;6(5):452–5.
43. Litmathe J, Sucker C, Easo J, Wigger L, Dapunt O. Prone and ECMO – a contradiction per se?
Perfusion. 2012;27(1):78–82.
44. Griffiths MJ, Evans TW. Inhaled nitric oxide therapy in adults. N Engl J Med.
2005;353(25):2683–95.
45. Rossaint R, Falke KJ, López F, Slama K, Pison U, Zapol WM. Inhaled nitric oxide for the adult
respiratory distress syndrome. N Engl J Med. 1993;328(6):399–405.
144 A. Pesenti et al.
46. Adhikari NK, Burns KE, Friedrich JO, Granton JT, Cook DJ, Meade MO. Effect of nitric
oxide on oxygenation and mortality in acute lung injury: systematic review and meta-analysis.
BMJ. 2007;334(7597):779.
47. Haraldsson A, Kieler-Jensen N, Nathorst-Westfelt U, Bergh CH, Ricksten SE. Comparison of
inhaled nitric oxide and inhaled aerosolized prostacyclin in the evaluation of heart transplant
candidates with elevated pulmonary vascular resistance. Chest. 1998;114(3):780–6.
48. Walmrath D, Schneider T, Schermuly R, Olschewski H, Grimminger F, Seeger W. Direct com-
parison of inhaled nitric oxide and aerosolized prostacyclin in acute respiratory distress syn-
drome. Am J Respir Crit Care Med. 1996;153(3):991–6.
49. Walmrath D, Schneider T, Pilch J, Grimminger F, Seeger W. Aerosolised prostacyclin in adult
respiratory distress syndrome. Lancet. 1993;342(8877):961–2.
50. van Heerden PV, Barden A, Michalopoulos N, Bulsara MK, Roberts BL. Dose-response to
inhaled aerosolized prostacyclin for hypoxemia due to ARDS. Chest. 2000;117(3):819–27.
51. Hager DN. High-frequency oscillatory ventilation in adults with acute respiratory distress syn-
drome. Curr Opin Anaesthesiol. 2012;25(1):17–23.
52. Courtney SE, Durand DJ, Asselin JM, Hudak ML, Aschner JL, Shoemaker CT, Neonatal
Ventilation Study Group. High-frequency oscillatory ventilation versus conventional mechani-
cal ventilation for very-low-birth-weight infants. N Engl J Med. 2002;347(9):643–52.
53. Ferguson ND, Cook DJ, Guyatt GH, Mehta S, Hand L, Austin P, Zhou Q, Matte A, Walter
SD, Lamontagne F, Granton JT, Arabi YM, Arroliga AC, Stewart TE, Slutsky AS, Meade MO,
OSCILLATE Trial Investigators; Canadian Critical Care Trials Group. High-frequency oscil-
lation in early acute respiratory distress syndrome. N Engl J Med. 2013;368(9):795–805.
54. Young D, Lamb SE, Shah S, MacKenzie I, Tunnicliffe W, Lall R, Rowan K, Cuthbertson BH,
OSCAR Study Group. High-frequency oscillation for acute respiratory distress syndrome. N
Engl J Med. 2013;368(9):806–13.
55. Banach M, Soukup J, Bucher M, Andres J. High frequency oscillation, extracorporeal mem-
brane oxygenation and pumpless arteriovenous lung assist in the management of severe
ARDS. Anestezjol Intens Ter. 2010;42(4):201–5.
56. Muellenbach RM, Kuestermann J, Kredel M, Johannes A, Wolfsteiner U, Schuster F, Wunder
C, Kranke P, Roewer N, Brederlau J. Arteriovenous extracorporeal lung assist allows for
maximization of oscillatory frequencies: a large-animal model of respiratory distress. BMC
Anesthesiol. 2008;8:7.
57. Muellenbach RM, Wunder C, Nuechter DC, Smul T, Trautner H, Kredel M, Roewer N,
Brederlau J. Early treatment with arteriovenous extracorporeal lung assist and high-frequency
oscillatory ventilation in a case of severe acute respiratory distress syndrome. Acta Anaesthesiol
Scand. 2007;51(6):766–9.
58. Putensen C, Muders T, Varelmann D, Wrigge H. The impact of spontaneous breathing during
mechanical ventilation. Curr Opin Crit Care. 2006;12(1):13–8.
59. Xia J, Gu S, Li M, Liu D, Huang X, Yi L, Wu L, et al. Spontaneous breathing in patients with
severe acute respiratory distress syndrome receiving prolonged extracorporeal membrane oxy-
genation. BMC Pulm Med. 2019;19(1):237.
60. Lindén V, Palmér K, Reinhard J, Westman R, Ehrén H, Granholm T, Frenckner B. High sur-
vival in adult patients with acute respiratory distress syndrome treated by extracorporeal mem-
brane oxygenation, minimal sedation, and pressure supported ventilation. Intensive Care Med.
2000;26(11):1630–7.
61. Cereda M, Foti G, Marcora B, Gili M, Giacomini M, Sparacino ME, Pesenti A. Pressure sup-
port ventilation in patients with acute lung injury. Crit Care Med. 2000;28(5):1269–75.
62. Mauri T, Bellani G, Grasselli G, Confalonieri A, Rona R, Patroniti N, Pesenti A. Patient-
ventilator interaction in ARDS patients with extremely low compliance undergoing ECMO: a
novel approach based on diaphragm electrical activity. Intensive Care Med. 2013;39(2):282–91.
63. Mauri T, Grasselli G, Suriano G, Eronia N, Spadaro S, Turrini C, Patroniti N, Bellani G,
Pesenti A. Control of respiratory drive and effort in extracorporeal membrane oxygen-
ation patients recovering from severe acute respiratory distress syndrome. Anesthesiology.
2016;125(1):159–67.
6 Ventilator Management During ECLS 145
64. Fuehner T, Kuehn C, Hadem J, Wiesner O, Gottlieb J, Tudorache I, Olsson KM, Greer M,
Sommer W, Welte T, Haverich A, Hoeper MM, Warnecke G. Extracorporeal membrane oxy-
genation in awake patients as bridge to lung transplantation. Am J Respir Crit Care Med.
2012;185(7):763–8.
65. Hoeper MM, Wiesner O, Hadem J, Wahl O, Suhling H, Duesberg C, Sommer W, Warnecke
G, Greer M, Boenisch O, Busch M, Kielstein JT, Schneider A, Haverich A, Welte T, Kühn
C. Extracorporeal membrane oxygenation instead of invasive mechanical ventilation in
patients with acute respiratory distress syndrome. Intensive Care Med. 2013;39(11):2056–7.
66. Crotti S, Bottino N, Ruggeri GM, Spinelli E, Tubiolo D, Lissoni A, Protti A, Gattinoni
L. Spontaneous breathing during extracorporeal membrane oxygenation in acute respiratory
failure. Anesthesiology. 2017;126(4):678–87.
67. Spinelli E, Mauri T, Lissoni A, Crotti S, Langer T, Albanese M, Volta CA, Fornari C, Tagliabue
P, Grasselli G, Pesenti A. Spontaneous Breathing Patterns During Maximum Extracorporeal
CO2 Removal in Subjects With Early Severe ARDS. Respir Care. 2020;65(7):911–9.
68. Akoumianaki E, Vaporidi K, Georgopoulos D. The Injurious Effects of Elevated or
Nonelevated Respiratory Rate during Mechanical Ventilation. Am J Respir Crit Care Med.
2019;199:149–57.
69. Goligher EC, Jonkman AH, Dianti J, Vaporidi K, Beitler JR, Patel BK, Yoshida T, Jaber S,
Dres M, Mauri T, Bellani G, Demoule A, Brochard L, Heunks L. Clinical strategies for imple-
menting lung and diaphragm-protective ventilation: avoiding insufficient and excessive effort.
Intensive Care Med. 2020;46(12):2314–26. https://doi.org/10.1007/s00134-020-06288-9.
70. Telias I, Junhasavasdikul D, Rittayamai N, Piquilloud L, Chen L, Ferguson ND, Goligher EC,
Brochard L. Airway Occlusion Pressure As an Estimate of Respiratory Drive and Inspiratory
Effort during Assisted Ventilation. Am J Respir Crit Care Med. 2020;201(9):1086–98.
71. Bertoni M, Telias I, Urner M, Long M, Del Sorbo L, Fan E, Sinderby C, Beck J, Liu L, Qiu H,
Wong J, Slutsky AS, Ferguson ND, Brochard LJ, Goligher EC. A novel non-invasive method
to detect excessively high respiratory effort and dynamic transpulmonary driving pressure dur-
ing mechanical ventilation. Crit Care. 2019;23(1):346.
72. Foti G, Cereda M, Banfi G, Pelosi P, Fumagalli R, Pesenti A. End-inspiratory airway occlusion:
a method to assess the pressure developed by inspiratory muscles in patients with acute lung
injury undergoing pressure support. Am J Respir Crit Care Med. 1997;156(4 Pt 1):1210–6.
73. Gattinoni L, Kolobow T, Damia G, Agostoni A, Pesenti A. Extracorporeal carbon dioxide
removal (ECCO2R): a new form of respiratory assistance. Int J Artif Organs. 1979;2(4):183–5.
74. Pesenti A, Rossi GP, Pelosi P, Brazzi L, Gattinoni L. Percutaneous extracorporeal CO2 removal
in a patient with bullous emphysema with recurrent bilateral pneumothoraces and respiratory
failure. Anesthesiology. 1990;72(3):571–3.
75. Marcolin R, Mascheroni D, Pesenti A, Bombino M, Gattinoni L. Ventilatory impact of partial
extracorporeal CO2 removal (PECOR) in ARF patients. ASAIO Trans. 1986;32(1):508–10.
76. Karagiannidis C, Lubnow M, Philipp A, Riegger GA, Schmid C, Pfeifer M, Mueller
T. Autoregulation of ventilation with neurally adjusted ventilatory assist on extracorporeal
lung support. Intensive Care Med. 2010;36(12):2038–44.
77. Burki NK, Mani RK, Herth FJ, Schmidt W, Teschler H, Bonin F, Becker H, Randerath WJ,
Stieglitz S, Hagmeyer L, Priegnitz C, Pfeifer M, Blaas SH, Putensen C, Theuerkauf N, Quintel
M, Moerer O. A novel extracorporeal CO2 removal system: results of a pilot study of hyper-
capnic respiratory failure in patients with COPD. Chest. 2013;143(3):678–86.
78. Cardenas VJ Jr, Lynch JE, Ates R, Miller L, Zwischenberger JB. Venovenous carbon diox-
ide removal in chronic obstructive pulmonary disease: experience in one patient. ASAIO
J. 2009;55(4):420–2.
79. Terragni PP, Del Sorbo L, Mascia L, Urbino R, Martin EL, Birocco A, Faggiano C, Quintel M,
Gattinoni L, Ranieri VM. Tidal volume lower than 6 ml/kg enhances lung protection: role of
extracorporeal carbon dioxide removal. Anesthesiology. 2009;111(4):826–35.
80. Terragni PP, Rosboch G, Tealdi A, Corno E, Menaldo E, Davini O, Gandini G, Herrmann P,
Mascia L, Quintel M, Slutsky AS, Gattinoni L, Ranieri VM. Tidal hyperinflation during low
146 A. Pesenti et al.
tidal volume ventilation in acute respiratory distress syndrome. Am J Respir Crit Care Med.
2007;175(2):160–6.
81. Bein T, Osborn E, Hofmann HS, Zimmermann M, Philipp A, Schlitt HJ, Graf BM. Successful
treatment of a severely injured soldier from Afghanistan with pumpless extracorporeal lung
assist and neurally adjusted ventilatory support. Int J Emerg Med. 2010;3(3):177–9.
82. McNamee JJ, Gillies MA, Barrett NA, Perkins GD, Tunnicliffe W, Young D, et al. Effect of
lower tidal volume ventilation facilitated by extracorporeal carbon dioxide removal vs stan-
dard care ventilation on 90-day mortality in patients with acute hypoxemic respiratory failure:
the REST randomized clinical trial. JAMA. 2021;326(11):1013–23.
83. Kluge S, Braune SA, Engel M, Nierhaus A, Frings D, Ebelt H, Uhrig A, Metschke M,
Wegscheider K, Suttorp N, Rousseau S. Avoiding invasive mechanical ventilation by extracor-
poreal carbon dioxide removal in patients failing noninvasive ventilation. Intensive Care Med.
2012;38(10):1632–9.
Chapter 7
Managing the Systemic Circulation:
Volume Status and RV Function
In patients with acute, severe respiratory failure, the pulmonary parenchymal dis-
ease process is often associated with severe pulmonary arterial hypertension and
right ventricular (RV) dysfunction, also called acute cor pulmonale. Intensive care
therapies such as fluid management, mechanical ventilation, and vasoactive infu-
sions can also exacerbate RV dysfunction and shock. In such patients, monitoring
and management of hemodynamic parameters can then be further complicated
when extracorporeal support is needed. It is hence essential for the intensivist to
develop familiarity with the unique considerations of fluid management and RV
dysfunction in patients on extracorporeal membrane oxygenation (ECMO) [1].
In this chapter, we will review the validity of various hemodynamic parameters
in assessing volume status and RV function in patients with severe respiratory fail-
ure on veno-venous extracorporeal membrane oxygenation (V-V ECMO) support.
We then discuss the nuances of fluid management in these patients. The next section
describes the challenges with monitoring RV function and cardiac output in patients
on extracorporeal support and the principles for RV support in this patient popula-
tion. Throughout the chapter, we lay a strong emphasis on the role of point-of-care
echocardiography in guiding bedside management. The chapter focuses primarily
on the care of patients on V-V ECMO. Where relevant, some additional points are
made regarding veno-arterial extracorporeal membrane oxygenation (V-A ECMO)
patients.
S. Krishnan (*)
Department of Anesthesiology, Duke University School of Medicine, Duke University
Medical Center, Durham, NC, USA
e-mail: [email protected]
G. A. Schmidt
Division of Pulmonary Diseases, Critical Care, and Occupational Medicine, Department of
Internal Medicine, University of Iowa, Iowa City, IA, USA
e-mail: [email protected]
Volume Status
As with all critically ill patients, adequate intravascular volume is required for
maintaining native cardiac output for organ perfusion. Maintaining adequate intra-
vascular volume is essential for one additional reason in patients on extracorporeal
support. Hypovolemia would lead to reduced extracorporeal support and crucially
in patients on V-V ECMO, reduced extracorporeal gas exchange, and hence, hypox-
emia. At the same time, positive fluid balance has been associated with worse out-
comes in critical illness. Most patients at this stage of disease would benefit from
fluid removal for relief of pulmonary and systemic edema.
Hence, the goal for volume status is dual – an adequate preload is required to
optimize cardiac output as well as ECMO flow, while also avoiding fluid overload.
Often, the goal of maintaining ECMO flow can be the driver for volume
resuscitation.
Monitoring
Circuit Negative Pressure and Cannula Chatter An early sign of reduced venous
return is a worsening negative pressure on the venous drainage cannula, as moni-
tored by the ECMO circuit. At typical flows (2–4 lpm) with typical sized cannulae
in normal sized adults, negative pressure readings lower than −80 to −100 mm Hg
usually reflect hypovolemia. In patients with severe hypovolemia, the ECMO drain-
age cannulae will often start vibrating (colloquially called “chatter”) as the pump
attempts to drain volume and the inferior vena cava (IVC) collapses around the can-
nula. As with negative pressure values, chatter is dependent on the pump settings
and is more common when high ECMO flows are attempted in conditions with
reduced venous return. At lower pump settings, the cannula chatter might be less
obvious.
CVP In critically ill patients, it has been well established that static parameters of
cardiac filling, including central venous pressure and pulmonary wedge pressure,
are inadequate markers of preload. In patients with severe ARDS on mechanical
support, despite intravascular hypovolemia, CVP can be elevated due to RV dys-
function or elevated pleural pressure. Additionally, the position of the central venous
catheter vis-a-vis the ECMO cannulae can further impact the CVP value in either
direction.
c ontour analysis or bioreactance are invalid during ECMO and should not be used
for this purpose. In patients on V-A ECMO, the reduction of native cardiac pulsatil-
ity in the arterial waveform further invalidates PPV.
Passive Leg Raising On the other hand, passive leg raising (PLR) does not rely on
changes in pleural pressure and has been shown useful during V-V ECMO, with an
increment in PLR-induced stroke volume greater than 10% (using transthoracic
echocardiography) predicting fluid response. PLR-related changes in end-tidal CO2
are likely to be erroneous as most CO2 is removed via the extracorporeal mem-
brane. Prone ventilation precludes using PLR to predict fluid responsiveness. An
alternative is to infuse a mini-fluid bolus (100 cc crystalloid) over 1 min, using pulse
contour-based methods to judge the effect [2].
POCUS The presence of a venous cannula can make IVC measurements difficult
to interpret. However, collapse of the IVC around the ECMO cannula on POCUS
examination suggests hypovolemia. Additionally, the very low tidal volumes that
are often employed in ventilatory management of V-V ECMO patients would
decrease the accuracy of respiratory variation in IVC diameter as a marker of fluid
responsiveness. In patients who are breathing spontaneously, high respiratory drive
and greatly reduced lung compliance may facilitate inspiratory collapse of the IVC,
mimicking hypovolemia.
Management
Table 7.1 Markers for volume overload and hypovolemia in patients on ECMO
Markers that suggest volume overload Markers that suggest hypovolemia
Elevated CVPa High negative pressures on the ECMO inflowb
Severe RV dilation (RVEDD/LVEDD >1.0) Chattering in the ECMO tubingb
Interventricular septal shift, D-shaped LV IVC collapse around ECMO cannula
Systolic blunting of hepatic venous flow Increasing vasopressor requirement
Pulsatile portal venous flow (>50%)
Systolic blunting of renal venous flow
Increased renal arterial resistive index
a
In addition to the other, well-recognized caveats related to the usefulness of CVP as a marker of
fluid responsiveness in critically ill patients, CVP can also be falsely elevated or reduced in ECMO
patients, depending on cannula position
b
High negative pressures on ECMO inflow and ECMO cannula chatter can also occur due to can-
nula malpositioning or tamponade, if the cannula is unable to drain venous return
Diuresis Once adequate flow has been established, volume removal should be initi-
ated, with the goal of relieving pulmonary and systemic edema. As intravascular
volume is decreased, signs of low preload might begin to appear. Potential responses
to cannula chatter and inadequate ECMO circuit flow include volume challenge,
reducing circuit flow, or calming inspiratory effort by raising sweep gas flow or
through pharmacologic means. It is important to note, however, that conditions such
as cannula malposition or tamponade might limit venous return to the specific site
of drainage from the cannula, despite there being adequate intravascular volume.
Echocardiography would play an important role in ruling out these confounders for
cannula chatter and high negative pressures. The presence of clots within the venous
drainage cannula might also lead to these extrinsic signs of hypovolemia and can be
more difficult to rule out.
Monitoring
blood flow from the ECMO cannula. On the other hand, all echocardiographic
parameters of RV function become less accurate in patients on V-A ECMO because
cardiac filling is unloaded by the pump.
Arterial Line-Based Cardiac Output Monitoring Minimally invasive, noncali-
brated hemodynamic monitors that utilize the arterial waveform are likely not
affected by V-V ECMO [8]. However, the quality of the arterial waveform and
inherent inconsistencies in this technology might make the data unreliable. In
patients on V-A ECMO, the addition of continuous pump flow to the reduced LV
pulsatility invalidates analysis of arterial waveform-based cardiac output assessment.
Monitors that rely on indicator dilution (thermal or lithium) would suffer indica-
tor loss into the extracorporeal circuit. The amount of flow in the ECMO circuit and
native cardiac output would impact the validity of this assessment. In patients on
low flow V-V ECMO for hypercarbic respiratory failure, transpulmonary thermodi-
lution cardiac output measurements are similar to those obtained by pulse contour
analysis [9].
Thermodilution Cardiac Output While pulmonary artery catheters (PACs) have
fallen out of favor in most critical care settings, they are still recommended in the
management of patients with severe cardiogenic shock. PACs offer important infor-
mation about left- and right-sided filling pressures, cardiac output, and the ability to
follow trends. However, these values should be interpreted with caution in patients
with severe pulmonary disease and on extracorporeal circulatory support. Various
7 Managing the Systemic Circulation: Volume Status and RV Function 153
SvO2 Venous oxygen saturation measured in the pulmonary artery (SvO2) or the
superior vena cava (ScvO2) is often used as a surrogate marker of cardiac output. In
patients on V-V ECMO, ScvO2 might or might not be reflective of the adequacy of
tissue perfusion, depending on the site of blood sampling and the position of the
reinfusion cannula. Similarly, Fick cardiac output measurements using ScvO2 will
be inaccurate in patients on V-V ECMO. Further downstream, oxygenated blood
returning from the V-V ECMO circuit mixes with the native cardiac output, making
the SvO2 an invalid measure of tissue oxygenation.
In patients on peripheral cannulation for V-A ECMO, pulmonary arterial flow
would consist mostly of blood from the superior vena cava and coronary sinus,
while blood from the inferior vena cava is drawn into the ECMO circuit. Measuring
SvO2 on the pre-membrane side of the oxygenator provides another method of eval-
uating tissue perfusion.
Lactate Serum lactate concentrations remain a reliable parameter for organ perfu-
sion in patients on ECMO. Elevated lactate levels correlate with mortality outcomes
in patients with ARDS with respiratory failure. Monitoring lactate trends can further
assist in evaluating the efficacy of circulatory support [11].
Management
Both intravascular volume depletion and volume overload are detrimental to organ
perfusion. Specifically, volume overload increases RV wall tension, with reduced
coronary perfusion pressure as well as reduced left ventricular diastolic filling due
to septal shift toward the left. This could lead to a worsening spiral of reduced car-
diac output, organ hypoperfusion, and venous congestion.
Vasoactives Vasoactive management for RV failure follows the same principles as
prior to initiation of V-V ECMO. Norepinephrine improves systemic pressure and
RV perfusion. At very high doses, norepinephrine and other alpha-agonists might
worsen pulmonary hypertension. On the other hand, vasopressin has been shown to
have minimal effects on pulmonary vascular resistance. Inotropic support with epi-
nephrine or dopamine might often still be required. Inodilators like milrinone and
dobutamine should be used cautiously because of the potential decrease in systemic
afterload. Inhaled vasodilators provide the benefit of decrease in pulmonary vascu-
lar resistance, without affecting systemic pressures. These inhaled therapies likely
still remain physiologically effective while the patient is on ECMO, despite low
tidal volume ventilation.
Table 7.3 Clinical features that suggest the use of VV vs VA ECMO in patients with
respiratory failure
VV ECMO is appropriate VA ECMO is appropriate
None or mild requirement for inotropic High dose requirement for inotropic support
support
Echocardiographic measures do not show Echocardiographic measures show severe RV
severe RV dysfunction dysfunction
Continued evidence of improving tissue Continued deterioration of circulatory parameters
perfusion while on VV ECMO and organ function, despite VV ECMO support
Septic shock with adequate cardiac Bi-ventricular failure or severe LV failure
output
alone. Table 7.3 lists criteria that suggest appropriateness of V-V or V-A ECMO in
patients with respiratory failure.
V-V ECMO as Initial Support In a review of 717 ARDS patients with shock in the
ELSO registry, only 18% of patients were started on V-A ECMO [15]. V-V ECMO
was associated with less gastrointestinal bleeding and hemolysis, but overall rates
of bleeding, stroke, and renal failure were similar. Survival to discharge was 58%
for V-V ECMO in contrast to 43% for V-A ECMO (p = 0.002). Multivariable regres-
sion analysis revealed V-V ECMO to be an independent predictor of survival to
discharge relative to V-A ECMO. Further, the conversion from V-V to V-A ECMO
in the study was only 4%.
Conclusion
References
13. Bunge JJH, Caliskan K, Gommers D, Reis MD. Right ventricular dysfunction during acute
respiratory distress syndrome and veno-venous extracorporeal membrane oxygenation. J
Thorac Dis. 2018;10(Suppl 5):S674–82.
14. Reis Miranda D, van Thiel R, Brodie D, Bakker J. Right ventricular unloading after ini-
tiation of venovenous extracorporeal membrane oxygenation. Am J Respir Crit Care Med.
2015;191(3):346–8.
15. Kon ZN, Bittle GJ, Pasrija C, et al. Venovenous versus venoarterial extracorporeal membrane
oxygenation for adult patients with acute respiratory distress syndrome requiring precannula-
tion hemodynamic support: a review of the ELSO registry. Ann Thorac Surg. 2017;104:645–9.
16. Cain MT, Smith NJ, Barash M, Simpson P, Durham LA 3rd, Makker H, Roberts C, Falcucci
O, Wang D, Walker R, Ahmed G, Brown SA, Nanchal RS, Joyce DL. Extracorporeal mem-
brane oxygenation with right ventricular assist device for COVID-19 ARDS. J Surg Res.
2021;264:81–9.
Chapter 8
Antithrombotic Therapy for ECMO
Usha S. Perepu
Introduction
Extra corporeal membrane oxygenation (ECMO) has been shown to improve sur-
vival in patients with acute respiratory failure when conventional ventilatory strate-
gies are inadequate [1]. Blood flow through these artificial devices disrupts normal
hemostasis and promotes coagulation by activation of the coagulation pathway.
Coagulation is induced by contact of the blood with the surface of the circuit and
membranes of the ECMO device. Therefore, anticoagulation is necessary to prevent
thrombus formation and maintain device function. However, bleeding risk is high
due to consumption of coagulation factors along with platelet dysfunction. Hence a
precarious balance between thrombosis and bleeding exists. Studies have shown
that thrombosis of the circuit and systemic thromboembolism is around 53% and is
partly dependent on the anticoagulation regimen used [4]. Severe hemorrhage of
around 40% with intracranial bleeding in 16–21% has been reported. It is also
known that hemostatic complications are associated with increased mortality.
U. S. Perepu (*)
Department of Internal Medicine, University of Iowa, Iowa City, IA, USA
e-mail: [email protected]
a ECMO Circuit
Hy
ce
pe
rfa
• Procoagulant hydrophobic • Free hemoglobin from hemolysis
rc
su
oag
artificial surface VV-ECMO • Circulating microparticles
lial
• Early fibrinogen deposition
ula
the
• Lack of endothelium
bil
do
y it
En
b Patient
Hy
ce
pe
rfa
bil
do
ity
En
Fig. 8.1 The prothrombotic changes described by Virchow’s triad in respect to (a) the ECMO
circuit and (b) patient factors (Adapted from Doyle and Hunt [22])
hypercoagulability. The etiology of the hemostatic changes can be divided into two
categories: (1) interactions between the blood and the extracorporeal circuit and (2)
patient-related factors (Fig. 8.1).
The contact system activation does not play a significant role in cell-based coagula-
tion; this is thought to be the primary mode of thrombin generation in blood flow
associated with extracorporeal circulation. The contact or intrinsic system is
8 Antithrombotic Therapy for ECMO 161
ECMO Biomaterial En
Endothelium
with absorbed proteins
C3
C3a
C3 Anaphylatoxins
C3b
C
C33 C5a
Contact Activation
b
C3bBb C3b
C3b3
C 3b3bBb
3b3bBb
bB
C3b3bBb
fXIIf fXII C5
C5
+ C6,
C C7,
C6 C C8,
C7 C CC9
Kallikrien
SC5b-9 or C5b-9 (MAC)
fXIIa Prekallikrein
HMWK Complement Activation
Bradykinin
fXII fXIa
fXIa
Ia Coagulation Cascade
fIX
X fiXa
fiX
fiXa
Xaa
Neutrophil Extravasation
ffX
X fXa
Fibrinogen hro
ombin
bin
iin
Thrombinn Prot
Pro
Pr
Prrot
o
otthrom
hrom
ro
ombin
ombin
Prothrombinbin
Fibrin
Integrans
Platelet Degranulation
IL-1β
TNF-α
Pro-inflammatory cytokines
Neutrophil Activation
Endothelial Activation
Platelet Activation
Fig. 8.2 The inflammatory response to extracorporeal membrane oxygenation (ECMO). During
ECMO, the complement and contact systems are activated as a result of blood-biomaterial interac-
tion. The alternative complement pathway (AP) is primarily responsible for producing the ana-
phylatoxins C3a and C5a and the membrane attack complex (MAC). This occurs as the result of
increased hydrolysis of C3 on the biomaterial surface. The contact system is responsible for pro-
ducing activated factor XII (FXIIa), which induces the intrinsic coagulation pathway, leading to
thrombin formation. Products produced by each of these systems promote the production of proin-
flammatory cytokines and have direct effects on leukocytes, platelets, and the vascular endothe-
lium. In particular, neutrophils are activated, leading to increased neutrophilic infiltration of tissue
and eventual organ damage (adapted from Millar et al. [36])
XI to XIa which in turn converts factor IX to IXa thus leading to activation of factor
X and thrombin generation. Although tissue factor-induced coagulation is thought
to be less prominent in extracorporeal circulation due to the absence of endothelial
injury, disseminated intravascular coagulation due to sepsis leads to tissue factor
expression by monocytes and macrophages which results in thrombin generation.
Also, TNF alpha and IL-6 have been described to induce endothelial cell expression
of soluble tissue factor.
Platelets not only play a role in hemostasis but are also a key mediator of inflam-
mation during ECMO by releasing proinflammatory cytokines, adhesion molecules,
and growth factors (Fig. 8.2). Complement activation, thrombin generation, and
adhesion of fibrinogen to the circuit surface all result in platelet activation. On the
other hand, continuous activation of the coagulation system can result in consump-
tion of coagulation factors and platelets leading to a bleeding due to thrombocyto-
penia. Severe thrombocytopenia of <50,000 cells/mm3 is seen in 22% of patients on
ECMO making anticoagulation challenging and increasing transfusion require-
ment [30].
Fibrinolysis is an essential next step in the hemostatic regulation once endothe-
lial injury is resolved. Activation of fibrinolysis is initiated by release of tissue plas-
minogen activator which activates plasminogen to plasmin causing fibrin breakdown.
Hyperfibrinolysis is excessive fibrin breakdown which can be associated with very
high D dimers resulting in bleeding. This can be commonly seen in patients on
ECMO [29].
Extracorporeal circulation results in red blood cell breakdown leading to
increased levels of free circulating hemoglobin (Hb) [22]. Free Hb is known to
deplete nitric oxide by binding which in turn leads to vasoconstriction and increas-
ing platelet activation thereby creating a prothrombotic state. Free plasma hemoglo-
bin levels >50 mg/dL after 24 h have been shown to be an independent predictor of
mortality in ECMO [27]. In addition, erythrocyte-derived microparticles resulting
from hemolysis or from transfused blood cells generate thrombin via contact path-
way [28].
The complement system is activated by three pathways: the classical pathway (CP),
the alternate pathway (AP), and the lectin pathway (LP), all of which are triggered
by distinct mechanisms with a common goal of C3 activation leading to the forma-
tion of the lytic membrane attack complex (MAC). The AP is activated in patients
on ECMO because of contact of the blood with the foreign material of the circuit.
Due to the absence of regulatory proteins to suppress the complement system by the
artificial surface, there is propagation of the complement cascade leading to exces-
sive inflammatory response and capillary leak syndrome. Studies have shown that
there is rapid activation of complement and formation of MAC within the first hour
of initiation of ECMO [20]. These levels tend to normalize within 2 days.
8 Antithrombotic Therapy for ECMO 163
Anticoagulation
Various modalities have been adopted to counteract the prothrombotic state with
ECMO. Nowadays almost all ECMO circuits are coated with heparin to reduce
activation of the coagulation pathway. Anticoagulants and antiplatelet agents have
been considered to overcome the risk for thrombosis. However, there is no consen-
sus on the ideal anticoagulation strategy [1, 4]. Large surveys showed that 96–100%
of centers reported using unfractionated heparin (UFH) as their main anticoagulant
of choice [3]. This is likely as UFH is the anticoagulant with the most experience, is
inexpensive, has short half-life with protamine as a reversal agent, and has relatively
low side effect profile. UFH is a complex glycosaminoglycan that binds to anti-
thrombin (AT); once bound, the UFH-AT complex inactivates coagulation factors
such as thrombin and factor Xa. It inhibits thrombin after it is formed but does not
prevent thrombin generation nor does it inhibit thrombin already bound to fibrin.
The onset of action is instantaneous after an intravenous infusion with a half-life of
around 45 min. UFH is metabolized in the reticuloendothelial system and the liver
and excreted in the urine. No dose modifications are needed for renal impairment
making it ideal for patients on ECMO with multiorgan failure. ELSO recommends
starting a 50–100 unit/kg bolus dose of UFH at the time of cannulation followed by
the initiation of a UFH infusion at 7.5–20 units/kg/h to achieve an ACT goal of
180–200s although there is significant institutional variability [2].
Argatroban and bivalirudin are synthetic parenteral short-acting direct thrombin
inhibitors (DTIs) which bind to both circulating and clot-bound thrombin inhibiting
thrombin-mediated cleavage of fibrinogen to fibrin; activation of coagulation factor
V, VIII, and XIII; and platelet aggregation. They have a more predictable
164 U. S. Perepu
Blood transfusion protocols are not evidence-based but rather based on clinical
experience and transfusion thresholds which vary largely between centers.
Platelet transfusions are considered to aim for a platelet count around
>100,000 cells/mm3. A lower threshold may be adequate in adults at low risk for
bleeding. Fresh frozen plasma may be considered if the prothrombin time is
166 U. S. Perepu
Monitoring of Anticoagulation
Specific Situations
Conclusion
Patients requiring ECMO are critically ill with various underlying conditions which
impact the coagulation and immune system. Balancing bleeding and thrombosis
remains a challenge. Robust guidelines with the help of quality research are needed
for the management of anticoagulation, treatment of bleeding, and improvement of
survival.
References
1. Esper SA, Welsby IJ, Subramaniam KJ, et al. Adult extracorporeal membrane oxygen-
ation: an international survey of transfusion and anticoagulation techniques. Vox Sang.
2017;112(5):443–52.
2. ELSO anticoagulation guidelines 2014.
3. Bembea MM, Annich G, Rycus P, et al. Variability in anticoagulation management of patients
on extracorporeal membrane oxygenation: an international survey. Pediatr Crit Care Med.
2013;14:e77–84.
4. Sklar MC, Sy E, Lequier L, Fan E, Kanji HD. Anticoagulation practices during venovenous
extracorporeal membrane oxygenation for respiratory failure. A systematic review. Ann Am
Thorac Soc. 2016;13(12):2242–50. https://doi.org/10.1513/AnnalsATS.201605-364SR. PMID:
27690525.
5. Krueger K, Schmutz A, Zieger B, Kalbhenn J. Venovenous extracorporeal membrane oxygen-
ation with prophylactic subcutaneous anticoagulation only: an observational study in more
than 60 patients. Artif Organs. 2017;41(2):186–92.
6. Carter KT, Kutcher ME, Shake JG, Panos AL, Cochran RP, Creswell LL, et al. Heparin-sparing
anticoagulation strategies are viable options for patients on veno-venous ECMO. J Surg Res.
2019;243:399–409. https://doi.org/10.1016/j.jss.2019.05.050. PMID: 31277018.
7. Lv X, Deng M, Wang L, Dong Y, Chen L, Dai X. Low vs standardized dose anticoagula-
tion regimens for extracorporeal membrane oxygenation: a meta-analysis. PLoS One.
2021;16(4):e0249854. https://doi.org/10.1371/journal.pone.0249854. eCollection 2021.
8. Study using citrate to replace heparin in babies requiring extracorporeal membrane oxygen-
ation (ECMO)
9. Shum HP, Kwan AM, Chan KC, Yan WW. The use of regional citrate anticoagulation con-
tinuous venovenous hemofiltration in extracorporeal membrane oxygenation. ASAIO
J. 2014;60(4):413–8.
10. Kaseer H, Soto-Arenall M, Sanghavi D, Moss J, Ratzlaff R, Pham S, Guru P. Heparin vs
bivalirudin anticoagulation for extracorporeal membrane oxygenation. J Card Surg.
2020;35(4):779–86. https://doi.org/10.1111/jocs.14458. PMID: 32048330.
11. Pieri M, Agracheva N, Bonaveglio E, Greco T, De Bonis M, Covello RD, Zangrillo A,
Pappalardo F. Bivalirudin versus heparin as an anticoagulant during extracorporeal membrane
oxygenation: a case-control study. J Cardiothorac Vasc Anesth. 2013;27(1):30–4.
12. Ranucci M, Ballotta A, Kandil H, et al. Bivalirudin-based versus conventional hepa-
rin anticoagulation for postcardiotomy extracorporeal membrane oxygenation. Crit Care.
2011;15(6):R275.
13. Bohman JK, Yi L, Seelhammer T. Observational case-control comparison of bivalirudin versus
heparin anticoagulation for adult extracorporeal membrane oxygenation (ECMO). Am J Res
Crit Care Med. 2018;197:A5114.
8 Antithrombotic Therapy for ECMO 171
31. Duffy MJ, Mullan BA, Craig TR, Shyamsundar M, MacSweeney RE, Thompson G, Stevenson
M, McAuley DF. Impaired endothelium-dependent vasodilatation is a novel predictor of mor-
tality in intensive care. Crit Care Med. 2011;39(4):629–35.
32. McILwain RB, Timpa JG, Kurundkar AR, Holt DW, Kelly DR, Hartman YE, Neel ML,
Karnatak RK, Schelonka RL, Anantharamaiah GM, et al. Plasma concentrations of inflamma-
tory cytokines rise rapidly during ECMO-related SIRS due to the release of preformed stores
in the intestine. Lab Investig. 2010;90(1):128–39.
33. Reynolds MM, Annich GM. The artificial endothelium. Organogenesis. 2011;7(1):42–9.
https://doi.org/10.4161/org.7.1.14029. PMID: 21289481; PMCID: PMC3082033.
34. Phillips RC, Shahi N, Leopold D, Levek C, Shirek G, Hilton S, Hyslop R, Gien J, Kinsella JP,
Buckvold S, Liechty KW, Kim JS, Marwan AI. Thromboelastography-guided management
of coagulopathy in neonates with congenital diaphragmatic hernia supported by extracorpo-
real membrane oxygenation. Pediatr Surg Int. 2020;36(9):1027–33. https://doi.org/10.1007/
s00383-020-04694-0. PMID: 32607833.
35. Panigada M, Iapichino G E, Brioni M, et al. Thromboelastography-based anticoagulation man-
agement during extracorporeal membrane oxygenation: a safety and feasibility pilot study.
Ann. Intensive Care. 2018;8:7. https://doi.org/10.1186/s13613-017-0352-8.
36. Millar JE, Fanning JP, McDonald CI, McAuley DF, Fraser JF. The inflammatory response to
extracorporeal membrane oxygenation (ECMO): a review of the pathophysiology. Crit Care.
2016;20(1):387. https://doi.org/10.1186/s13054-016-1570-4. PMID: 27890016; PMCID:
PMC5125043
Chapter 9
Membrane Dysfunction
Introduction
B. D. Warren (*)
Department of Internal Medicine, Brooke Army Medical Center, San Antonio, TX, USA
M. J. Sobieszczyk
Department of Pulmonary and Critical Care Medicine, Brooke Army Medical Center,
San Antonio, TX, USA
e-mail: [email protected]
P. E. Mason
Department of Surgery, Brooke Army Medical Center, San Antonio, TX, USA
e-mail: [email protected]
Hematologic Profile
The normal physiologic balance between thrombosis and fibrinolysis, largely regu-
lated at the level of the endothelium, is exceedingly complex. Stability of the coagu-
lation system in patients requiring ECMO support is further complicated by the
disease state as well as the prothrombotic nature of continuous blood to nonbiologic
surface interaction within the extracorporeal circuit [12, 17]. Details of the complex
cross talk between the coagulation system, complementation activation, and sys-
temic inflammatory response as well as the influence of flow dynamics, pressure
changes, and sheer stress within the ECMO circuit have been expertly reviewed
9 Membrane Dysfunction 175
elsewhere [5, 12, 17, 18]. Despite technological advances in pump and ML design,
combined with systemic anticoagulation and local antithrombotic surface modifica-
tions within the circuit, there remains considerable hematologic dysfunction.
Additionally, there is marked variability between ECMO centers regarding the
choice of anticoagulant (heparin vs direct thrombin inhibitor), use of biopassive
(phosphorylcholine)- vs biomimetic (heparin)-coated circuits, as well as methodol-
ogy for monitoring hematologic changes [11, 19–21] (see also Chap. 8). Frequent
assessment of the hematologic profile is integral to the evaluation of ML dysfunc-
tion and possible preemptive replacement as discussed below. While various hema-
tologic markers have been evaluated, our institution utilizes a battery of INR,
activated partial thromboplastin time (aPTT), anti-Xa, D-dimer, fibrinogen, lactate
dehydrogenase (LDH), plasma-free hemoglobin (PfHb), hemolysis index, platelet
count, and thromboelastography (TEG®) in our integrated assessment of ML
performance.
∅PML / QB (9.2)
where V O2,ML = O2 transfer across the ML (ml/min), QB = blood flow rate (L/min),
and CxO2 = O2 content of (pre-/post-ML) blood (ml/dL). The oxygen content of
pre-/post-membrane (CPRE,O2 and CPOST,O2) blood is defined by Eq. (9.4) [26].
where FSO2 is the sweep gas inlet oxygen fraction, PATM is atmospheric pressure,
PH2O is water vapor pressure, PPOSTCO2 is the partial pressure of CO2 in the post-
membrane blood (mmHg), and RQ is the respiratory quotient (VO2/VCO2) [13,
14, 29].
dividing the PPOSTO2 by FSO2 as this has been shown to be inversely proportional to
QS/QB [13, 30, 31]. However, a progressive increase in the ML shunt fraction as well
as sustained ratio >30% has been shown to correlate with overall ML aging and gas
transfer efficiency [32].
ML Dysfunction
Optimal timing as well as specific criteria for an ECMO circuit change is a nonstan-
dardized practice between various centers. While multiple technical or patient-
specific factors can contribute to the decision, one essential element is the evaluation
of ML efficiency. Prompt recognition of ML dysfunction allows elective replace-
ment, as described below, and reduces the likelihood of emergent exchange with its
inherent risks. However, premature replacement of an adequately functioning cir-
cuit also generates risk [16]. Membrane lung exchange may be indicated by one or
more of the following factors: (a) circuit-associated hematologic abnormalities, (b)
progressive blood flow obstruction, or (c) inadequate gas exchange (Fig. 9.1).
Coagulopathy Hemolysis
↑ aPTT
↑ anti-Xa ↑ LDH
1 ↓ Plt
D-Dimer > 25-30 mg/dL
↑PfHb
Hemolysis Index
Fibrinogen < 200 mg/dL Hemoglobinuria
Hematologic ↑ TEG® R –Time
Profile
Fig. 9.1 ML dysfunction is recognized through changes in hematologic profile, pressure across
the ML, or impaired gas exchange. aPTT activated partial thromboplastin time, anti-Xa anti-factor
Xa, Plt platelet count, TEG R-time thromboelastography reaction time to initial thrombin forma-
tion, LDH lactate dehydrogenase, PfHb plasma-free hemoglobin, ΔPML pressure difference across
the ML, PPREML inlet pressure, PPOSTML outlet pressure, RML resistance across the ML, QB circuit
blood flow, V O2, ML oxygen transfer across the ML, CPOSTO2 outlet oxygen content, CPREO2 inlet
oxygen content, PPOSTO2 outlet oxygen partial pressure, FSO2 oxygen fraction in the sweep gas, V
CO2, ML carbon dioxide transfer across the membrane, QG sweep gas flow, FOCO2 gas outlet fraction
of carbon dioxide, PPOSTCO2 outlet partial pressure of carbon dioxide, PPRECO2 inlet partial pres-
sure of carbon dioxide
178 B. D. Warren et al.
Hematologic Abnormalities
Persistent hypoxemia in the presence of a post-oxygenator PO2 < 200 mmHg should
prompt calculation of V O2,ML [30] (see Eqs. (9.3) and (9.4) above).
Before calculating V O2,ML, conditions must be optimized to maximize oxygen
transfer with potential targets including circuit blood flow, blood oxygen content
entering the circuit (CPREO2), and the sweep gas oxygen concentration [16]. Elevated
CPREO2 due to recirculation or impaired tissue extraction will limit oxygen transfer.
Oxygen fraction of the sweep gas (FSO2) provides the gradient for transfer such that
values below 1.0 can limit maximal transfer. If V O2,ML is <100–150 ml/min after
optimization in a patient with sustained hypoxemia, ML exchange is indicated [16,
23, 26].
ECMO facilitates CO2 clearance by generating a CO2 pressure gradient between the
transiting blood and the gas phase within the ML. Carbon dioxide removal ( V
CO2,ML) is proportional to the sweep gas flow, largely independent of the blood flow
rate through the circuit [65], and adversely affected by regions of increased clot
burden that allow for ventilation without adequate perfusion (ML dead space) [26].
The calculation for V CO2,ML is shown in Eq. (9.8) [13, 66].
where QG is sweep gas flow, FOCO2 is sweep gas outlet fraction of CO2, and 1000
is mL/L.
9 Membrane Dysfunction 181
Exchanging the ML
References
1. Brodie D, Slutsky AS, Combes A. Extracorporeal life support for adults with respiratory fail-
ure and related indications: a review. JAMA. 2019;322(6):557–68.
2. Davies A, Jones D, Bailey M, Beca J, Bellomo R, Blackwell N, et al. Extracorporeal mem-
brane oxygenation for 2009 influenza A (H1N1) acute respiratory distress syndrome.
JAMA. 2009;302:1888–95.
3. Extracorporeal life support registry report. Available online: https://www.elso.org/Registry/
Statistics/InternationalSummary.aspx. Accessed on 15 Aug 2021.
4. Aubron C, Cheng AC, Pilcher D, Leong T, Magrin G, Cooper DJ, et al. Factors associated with
outcomes of patients on extracorporeal membrane oxygenation support: a 5-year cohort study.
Crit Care. 2013;17:R73.
5. Mulder MM, Fawzy I, Lancé MD. ECMO and anticoagulation: a comprehensive review. Neth
J Crit Care. 2018;26:6–13.
182 B. D. Warren et al.
6. Rastan AJ, Lachmann N, Walther T, Doll N, Gradistanac T, Gommert JF, et al. Autopsy find-
ings in patients on postcardiotomy extracorporeal membrane oxygenation (ECMO). Int J Artif
Organs. 2006;29:1121–31.
7. Betit P. Technical advances in the field of ecmo. Respir Care. 2018;63(9):1162–73. https://doi.
org/10.4187/respcare.06320.
8. Ündar A, Wang S, Palanzo DA. Impact of Polymethylpentene OXYGENATORS on out-
comes of all extracorporeal life support patients in the United States. Artif Organs.
2013;37(12):1080–1. https://doi.org/10.1111/aor.12242.
9. Besser MW, Klein AA. The coagulopathy of cardiopulmonary bypass. Crit Rev Clin Lab Sci.
2010;47(5–6):197–212. https://doi.org/10.3109/10408363.2010.549291.
10. Gorbet MB, Sefton MV. Biomaterial-associated thrombosis: roles of coagulation factors,
complement, platelets and leukocytes. Biomaterials. 2004;25(26):5681–703. https://doi.
org/10.1016/j.biomaterials.2004.01.023.
11. Oliver WC. Anticoagulation and coagulation management for ECMO. Semin Cardiothorac
Vasc Anesth. 2009;13(3):154–75. https://doi.org/10.1177/1089253209347384.
12. Millar JE, Fanning JP, McDonald CI, McAuley DF, Fraser JF. The inflammatory response to
extracorporeal membrane oxygenation (ecmo): a review of the pathophysiology. Crit Care.
2016;20(1):1–10. https://doi.org/10.1186/s13054-016-1570-4.
13. Epis F, Belliato M. Oxygenator performance and artificial-native lung interaction. J Thorac
Dis. 2018;10(S5):S596–605. https://doi.org/10.21037/jtd.2017.10.05.
14. Ficial B, Vasques F, Zhang J, Whebell S, Slattery M, Lamas T, Daly K, Agnew N, Camporota
L. Physiological basis of extracorporeal membrane oxygenation AND extracorporeal carbon
dioxide removal in respiratory failure. Membranes. 2021;11(3):225. https://doi.org/10.3390/
membranes11030225.
15. Lehle K, Philipp A, Müller T, Schettler F, Bein T, Schmid C, Lubnow M. Flow dynamics
of different ADULT Ecmo systems: a clinical evaluation. Artif Organs. 2013;38(5):391–8.
https://doi.org/10.1111/aor.12180.
16. Zakhary B, Vercaemst L, Mason P, Antonini MV, Lorusso R, Brodie D. How I approach mem-
brane lung dysfunction in patients receiving ECMO. Crit Care. 2020;24(1):1–4. https://doi.
org/10.1186/s13054-020-03388-2.
17. Doyle AJ, Hunt BJ. Current understanding of how extracorporeal membrane oxygenators acti-
vate haemostasis and other blood components. Front Med. 2018;5:352. https://doi.org/10.3389/
fmed.2018.00352.
18. Pan KC, McKenzie DP, Pellegrino V, Murphy D, Butt W. The meaning of a high PLASMA
free haemoglobin: retrospective review of the prevalence of haemolysis and circuit throm-
bosis in an ADULT ECMO Centre over 5 years. Perfusion. 2015;31(3):223–31. https://doi.
org/10.1177/0267659115595282.
19. Chlebowski MM, Baltagi S, Carlson M, Levy JH, Spinella PC. Clinical controversies in antico-
agulation monitoring and antithrombin supplementation for ecmo. Crit Care. 2020;24(1):1–2.
https://doi.org/10.1186/s13054-020-2726-9.
20. Despotis GJ, Avidan MS, Hogue CW. Mechanisms and attenuation of hemostatic activation
during extracorporeal circulation. Ann Thorac Surg. 2001;72:S1821–31.
21. Esper SA, Levy JH, Waters JH, Welsby IJ. Extracorporeal membrane oxygenation in the adult:
a review of anticoagulation monitoring and transfusion. Anesth Analg. 2014;118:731–43.
22. Lehle K, Philipp A, Gleich O, Holzamer A, Müller T, Bein T, Schmid C. Efficiency in
extracorporeal MEMBRANE oxygenation—cellular deposits on polymethylpentene
MEMBRANES increase resistance to blood flow and reduce gas exchange capacity. ASAIO
J. 2008;54(6):612–7. https://doi.org/10.1097/mat.0b013e318186a807.
23. 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;40:1870–7. https://doi.org/10.1007/s00134-014-3489-z.
24. Park M, Costa EL, Maciel AT, Silva DP, Friedrich N, Barbosa EV, Hirota AS, Schettino G,
Azevedo LC. Determinants of oxygen and carbon DIOXIDE transfer during extracorporeal
9 Membrane Dysfunction 183
therapy: laboratory monitoring of unfractionated heparin therapy. Arch Pathol Lab Med.
1998;122(9):782–98. PMID: 9740136.
40. Liveris A, Bello RA, Friedmann P, Duffy MA, Manwani D, Killinger JS, Rodriquez D,
Weinstein S. Anti-factor Xa assay is a superior correlate of heparin dose than activated
partial thromboplastin time or activated clotting time in pediatric extracorporeal mem-
brane oxygenation. Pediatr Crit Care Med. 2014;15(2):e72–9. https://doi.org/10.1097/
PCC.0000000000000028. PMID: 24335992.
41. Alban S. Pharmacological strategies for inhibition of thrombin activity. Curr Pharm Des.
2008;14(12):1152–75. https://doi.org/10.2174/138161208784246135. PMID: 18473863.
42. Krueger K, Schmutz A, Zieger B, Kalbhenn J. Venovenous extracorporeal membrane oxygen-
ation with prophylactic subcutaneous anticoagulation only: an observational study in more
than 60 patients. Artif Organs. 2017;41(2):186–92. https://doi.org/10.1111/aor.12737. PMID:
27256966.
43. Di Nisio M, Middeldorp S, Büller HR. Direct thrombin inhibitors. N Engl J Med.
2005;353(10):1028–40. https://doi.org/10.1056/NEJMra044440. Erratum in: N Engl J Med.
2005 Dec 29;353(26):2827. PMID: 16148288.
44. Colman E, Yin EB, Laine G, Chatterjee S, Saatee S, Herlihy JP, Reyes MA, Bracey
AW. Evaluation of a heparin monitoring protocol for extracorporeal membrane oxygenation
and review of the literature. J Thorac Dis. 2019;11(8):3325–35. https://doi.org/10.21037/
jtd.2019.08.44. PMID: 31559035; PMCID: PMC6753426.
45. Delmas C, Jacquemin A, Vardon-Bounes F, Georges B, Guerrero F, Hernandez N, Marcheix
B, Seguin T, Minville V, Conil JM, Silva S. Anticoagulation monitoring under ECMO support:
a comparative study between the activated coagulation time and the anti-Xa activity assay. J
Intensive Care Med. 2020;35(7):679–86. https://doi.org/10.1177/0885066618776937. PMID:
29768983.
46. Kostousov V, Nguyen K, Hundalani SG, Teruya J. The influence of free hemoglobin and bili-
rubin on heparin monitoring by activated partial thromboplastin time and anti-Xa assay. Arch
Pathol Lab Med. 2014;138(11):1503–6. https://doi.org/10.5858/arpa.2013-0572-OA. PMID:
25357112.
47. Alexander DC, Butt WW, Best JD, Donath SM, Monagle PT, Shekerdemian LS. Correlation
of thromboelastography with standard tests of anticoagulation in paediatric patients receiv-
ing extracorporeal life support. Thromb Res. 2010;125(5):387–92. https://doi.org/10.1016/j.
thromres.2009.07.001. PMID: 19674773.
48. Panigada M, E Iapichino G, Brioni M, Panarello G, Protti A, Grasselli G, Occhipinti G,
Novembrino C, Consonni D, Arcadipane A, Gattinoni L. Thromboelastography-based anti-
coagulation management during extracorporeal membrane oxygenation: a safety and feasibil-
ity pilot study. Ann Intensive Care. 2018;8(1):7. https://doi.org/10.1186/s13613-017-0352-8.
PMID: 29340875; PMCID: PMC5770349.
49. Rother RP, Bell L, Hillmen P, Gladwin MT. The clinical sequelae of intravascular hemo-
lysis and extracellular plasma hemoglobin: a novel mechanism of human disease.
JAMA. 2005;293(13):1653–62. https://doi.org/10.1001/jama.293.13.1653. PMID: 15811985.
50. Boggio LN, Simpson ML. Fibrinogen replacement underutilization in ECMO. Blood.
2020;136(Supplement 1):4–5. https://doi.org/10.1182/blood-2020-142812.
51. Dornia C, Philipp A, Bauer S, Stroszczynski C, Schreyer AG, Müller T, Koehl GE, Lehle
K. D-dimers are a predictor of clot volume inside membrane oxygenators during extracor-
poreal membrane oxygenation. Artif Organs. 2015;39(9):782–7. https://doi.org/10.1111/
aor.12460. PMID: 25845704.
52. Lubnow M, Philipp A, Dornia C, Schroll S, Bein T, Creutzenberg M, Diez C, Schmid C, Pfeifer
M, Riegger G, Müller T, Lehle K. D-dimers as an early marker for oxygenator exchange in
extracorporeal membrane oxygenation. J Crit Care. 2014;29(3):473. https://doi.org/10.1016/j.
jcrc.2013.12.008.
53. Kraft F, Schmidt C, Van Aken H, Zarbock A. Inflammatory response and extracorporeal cir-
culation. Best Pract Res Clin Anaesthesiol. 2015;29(2):113–23. https://doi.org/10.1016/j.
bpa.2015.03.001. PMID: 26060024.
9 Membrane Dysfunction 185
54. Whiteheart SW. Platelet granules: surprise packages. Blood. 2011;118(5):1190–1. https://doi.
org/10.1182/blood-2011-06-359836. PMID: 21816838.
55. Kalbhenn J, Schmidt R, Nakamura L, Schelling J, Rosenfelder S, Zieger B. Early diagno-
sis of acquired von Willebrand Syndrome (AVWS) is elementary for clinical practice in
patients treated with ECMO therapy. J Atheroscler Thromb. 2015;22(3):265–71. https://doi.
org/10.5551/jat.27268. PMID: 25186021.
56. Abrams D, Baldwin MR, Champion M, Agerstrand C, Eisenberger A, Bacchetta M, Brodie
D. Thrombocytopenia and extracorporeal membrane oxygenation in adults with acute respira-
tory failure: a cohort study. Intensive Care Med. 2016;42(5):844–52. https://doi.org/10.1007/
s00134-016-4312-9. PMID: 27007099; PMCID: PMC5407307.
57. Fisser C, Winkler M, Malfertheiner MV, Philipp A, Foltan M, Lunz D, Zeman F, Maier LS,
Lubnow M, Müller T. Argatroban versus heparin in patients without heparin-induced throm-
bocytopenia during venovenous extracorporeal membrane oxygenation: a propensity-score
matched study. Crit Care. 2021;25(1):160. https://doi.org/10.1186/s13054-021-03581-x.
58. Dufour N, Radjou A, Thuong M. Hemolysis and plasma free hemoglobin during extracorpo-
real membrane oxygenation support: from clinical implications to laboratory details. ASAIO
J. 2020;66(3):239–46. https://doi.org/10.1097/MAT.0000000000000974. PMID: 30985331.
59. Zangrillo A, Landoni G, Biondi-Zoccai G, Greco M, Greco T, Frati G, Patroniti N, Antonelli
M, Pesenti A, Pappalardo F. A meta-analysis of complications and mortality of extracorporeal
membrane oxygenation. Crit Care Resusc. 2013;15(3):172–8. PMID: 23944202.
60. Horton AM, Butt W. Pump-induced haemolysis: is the constrained vortex pump
better or worse than the roller pump? Perfusion. 1992;7(2):103–8. https://doi.
org/10.1177/026765919200700204.
61. Kawahito S, Maeda T, Motomura T, Ishitoya H, Takano T, Nonaka K, Linneweber J, Ichikawa
S, Kawamura M, Hanazaki K, Glueck J, Nosé Y. Hemolytic characteristics of oxygenators
during clinical extracorporeal membrane oxygenation. ASAIO J. 2002;48(6):636–9. https://
doi.org/10.1097/00002480-200211000-00010. PMID: 12455774.
62. Jeney V, Balla J, Yachie A, Varga Z, Vercellotti GM, Eaton JW, Balla G. Pro-oxidant and cyto-
toxic effects of circulating heme. Blood. 2002;100(3):879–87. https://doi.org/10.1182/blood.
v100.3.879. PMID: 12130498.
63. Schaer DJ, Buehler PW, Alayash AI, Belcher JD, Vercellotti GM. Hemolysis and free hemo-
globin revisited: exploring hemoglobin and hemin scavengers as a novel class of therapeu-
tic proteins. Blood. 2013;121(8):1276–84. https://doi.org/10.1182/blood-2012-11-451229.
PMID: 23264591; PMCID: PMC3578950.
64. Marques-Garcia F. Methods for hemolysis interference study in laboratory medicine – a criti-
cal review. EJIFCC. 2020;31(1):85–97. PMID: 32256292; PMCID: PMC7109502.
65. Kolobow T, Gattinoni L, Tomlinson TA, Pierce JE. Control of breathing using an
extracorporeal membrane lung. Anesthesiology. 1977;46(2):138–41. https://doi.
org/10.1097/00000542-197702000-00012. PMID: 13683.
66. Zanella A, Mangili P, Giani M, Redaelli S, Scaravilli V, Castagna L, Sosio S, Pirrone F,
Albertini M, Patroniti N, Pesenti A. Extracorporeal carbon dioxide removal through
ventilation of acidified dialysate: an experimental study. J Heart Lung Transplant.
2014;33(5):536–41. https://doi.org/10.1016/j.healun.2013.12.006. PMID: 24439968.
Chapter 10
ECCO2R in Obstructive Diseases:
Evidence, Indications, and Exclusions
Introduction
In the last decade, there has been an impressive technological and clinical evolution
of extracorporeal life support (ECLS) strategies [1–4]. To date, a variety of modali-
ties of ECLS have been developed, characterized by different configurations, differ-
ent vascular accesses, different rates of blood flow, and different clinical goals [5].
In particular, remarkable interest has been focused on extracorporeal carbon dioxide
removal (ECCO2R) [6, 7], due to the relative ease and efficiency in blood CO2 clear-
ance granted by extracorporeal gas exchangers as compared to oxygen delivery.
CO2 present in blood, mainly dissociated into bicarbonates and hydrogen ions and
bound to hemoglobin, is characterized by a steep, linear dissociation kinetic and
high solubility, which provide an elevated diffusion rate across the membrane lung
(ML). Therefore, ventilation (sweep gas flow) rather than perfusion (extracorporeal
blood flow rate) of the ML and its total surface area is the main determinant of the
CO2 clearance from the blood. Since CO2 is, on average, metabolically produced at
the rate of 200–250 ml/min, ECCO2R systems running blood flow through the ML
even as low as 0.5–1 L/min may theoretically remove the entire CO2 production
(without providing significant oxygenation) [8–11].
These features have been exploited by the industry and technologically trans-
lated in clinical strategies of ECLS providing ECCO2R that are remarkably less
invasive as compared to the full extracorporeal membrane oxygenation (ECMO),
easier to manage, and as efficient in CO2 clearance [6]. Several ECCO2R devices
have been recently approved, at least in Europe, and are available for clinical use.
In most of these systems, the blood is drawn from a central vein by a draining
cannula, driven with a centrifugal or roller pump through the ML, where CO2 is
removed by diffusion sustained by a sweep gas flow, and returned into the venous
circulation usually through a second lumen of the draining cannula (Fig. 10.1a).
Hence, central venous access with a dual-lumen cannula is commonly required by
most ECCO2R systems. It is nonetheless possible to set an ECCO2R circuit so that
two central veins, one for drainage and the other for reinfusion, are accessed
(Fig. 10.1b). However, reducing the dimensions and number of central venous
a Veno-venous ECCO2-R
with one endovascular dual-lumen catheter
L/min
Sweep O2
gas flow
Gas
exchanger
Blood
flow
CO2
Pump Double
lumen
cannula
0.3-1 L/min
Console
Fig. 10.1 Possible configurations of extracorporeal carbon dioxide removal (ECCO2R) circuits.
(a) Minimally invasive veno-venous ECCO2R circuit with a single venous vascular access through
a dual-lumen cannula that can be typically inserted in the internal jugular vein or in the femoral
vein. (b) Veno-venous ECCO2R circuit with two separate venous vascular accesses, with the flow
usually running from femoral vein to the jugular vein or from femoral vein to femoral vein. (c)
Pumpless arteriovenous ECCO2R device with the typical placement of the ML connecting the
femoral artery with the contralateral femoral vein
10 ECCO2R in Obstructive Diseases: Evidence, Indications, and Exclusions 189
Veno-venous ECCO2-R
b with double vascular access
L/min
Sweep O2
gas flow
Gas
exchanger
Blood
flow
CO2
Pump Double
lumen
cannula
0.3-1 L/min
Console
Arterio-venous ECCO2-R
c
Blood
flow
Gas
exchanger
L/min Sweep
gas flow
O2 CO2
Evidence
Although the use of ECCO2R has been described in several case series of patients
with obstructive lung disease, we lack randomized trials to clearly demonstrate effi-
cacy. However, numerous clinical reports demonstrate the efficiency of this techno-
logically advanced strategy, which is based on a strong pathophysiological rationale
[17, 23]. The main feature of exacerbations of obstructive diseases is acute worsen-
ing of expiratory flow-limitation caused by increased resistance in small airways
[19, 24]. Increased resistive load is associated with a longer expiratory time con-
stant (the product of airway resistance and lung compliance), which defines the time
for exponential fall in lung volume during passive exhalation [25]. The direct con-
sequence of a longer time constant is the development of dynamic alveolar hyperin-
flation, which is also generally associated with an increase in the end-expiratory
elastic recoil of the respiratory system, also called intrinsic positive end-expiratory
pressure (Fig. 10.2) [25].
Dynamic alveolar hyperinflation and intrinsic-PEEP have detrimental effects on
the respiratory and cardiovascular system, remarkably increasing the work of
breathing, reducing venous return, and compromising cardiac output and lung per-
fusion [19, 26]. The persistence of these pathophysiological features results in
respiratory muscle fatigue and rapid shallow breathing, with the consequent reduc-
tion of alveolar ventilation (Fig. 10.2) [24, 26–28]. Since alveolar ventilation is
inversely proportional to PaCO2 (according to the equation PaCO2= V CO2/ V A,
where V CO2 is the CO2 production and V A is alveolar ventilation), acute exacerba-
tions of obstructive lung diseases produce respiratory acidosis. The rise in PaCO2,
in turn, increases ventilatory demand, which may worsen alveolar hyperinflation
and muscle fatigue, creating a vicious loop [18, 29, 30].
In this situation, noninvasive ventilation (NIV) can reduce the work of breathing
and increase alveolar ventilation, but its efficiency may not be sufficient especially
in the most severe cases [31]. Invasive mechanical ventilation (MV) also reduces the
work of breathing and is effective in correcting gas exchange but may produce
severe side effects. Moreover, with improper settings, invasive ventilation may
worsen alveolar hyperinflation, causing barotrauma, lung rupture, and hemody-
namic decompensation [19, 32–34].
ECCO2R, on the other hand, represents an innovative and physiologically attrac-
tive option, at least in theory. In contrast to NIV and invasive ventilation, whose goal
is to reduce PaCO2 by mechanically augmenting alveolar ventilation in already
hyperinflated lungs, ECCO2R reduces the alveolar ventilation required to eliminate
the CO2 production. Reduced tidal volume and respiratory rate extends expiratory
time, suiting better the high expiratory time constant characteristic of diseases with
expiratory flow limitation. By these physiological mechanisms ECCO2R can inter-
rupt the vicious circle of dynamic hyperinflation, thereby reestablishing a more
favorable balance between the work of breathing and the respiratory load in patients
with acute obstructive lung diseases exacerbations [17, 23].
192 L. Del Sorbo et al.
COPD exacerbation
a
worsening of
1. Airflow resistance
and
2. Expiratory flow limitation
threshold loading of
CDYN VD/VT
inspiratory muscles
pulmonary
Neuro-mechanical artery pressure
uncoupling
with dyspnea RV preload
LV afterload
Hypercapnia
and
respiratory acidosis
ventilatory drive
and VA demand
Fig. 10.2 (a) Schematic representation of the pathophysiological mechanisms underlying exacer-
bations of chronic obstructive pulmonary disease (COPD). (b) Schematic representation of the
pathophysiological rationale of ECCO2R application in COPD. PEEPi intrinsic positive end expi-
ratory pressure, CDYN dynamic compliance, VD dead space, VT tidal volume, VD/VT fraction of dead
space, RV right ventricle, LV left ventricle, RR respiratory rate, VA alveolar ventilation
10 ECCO2R in Obstructive Diseases: Evidence, Indications, and Exclusions 193
COPD exacerbation
b
worsening of
1. Airflow resistance
and
2. Expiratory flow limitation
Hypercapnia and
respiratory acidosis
ECCO2R
RV preload
Neuro-mechanical
CDYN uncoupling
LV afterload
Indications
COPD is one of the most significant social and economic burdens in western coun-
tries, being the fourth leading cause of death [35–37]. Its natural history is charac-
terized by progressive deterioration of lung function, punctuated by a variable
number of acute exacerbations [38, 39]. COPD exacerbations are associated with
considerable morbidity and mortality, which ranges between 8% and 26% for
patients admitted to the intensive care unit [40–42].
During acute exacerbations of COPD, patients develop alveolar hyperinflation,
which in the most severe cases may lead to muscle fatigue and respiratory acidosis,
creating a vicious loop refractory to medical treatment [18, 29, 30]. Advanced strat-
egies are available to break this cycle, improve respiratory function, and gain time
for treatment of the exacerbation [43–45].
Firstly, NIV has become the standard of care for exacerbations refractory to med-
ical treatment [31, 44–46]. NIV effectively supports the respiratory muscles,
increases alveolar ventilation, reduces PaCO2, and improves oxygenation. Moreover,
externally applied PEEP counteracts intrinsic-PEEP, reducing the inspiratory
threshold load, further lowering the work of breathing and helping to reverse muscle
fatigue. Together these effects are instrumental in decreasing the need for endotra-
cheal intubation and improving survival [47–49].
A second option is invasive mechanical ventilation (MV), used when NIV fails
or altered consciousness precludes effective NIV [32, 34]. Failure of NIV occurs in
a significant percentage of cases (30%) and is associated with increased mortality
[50]. Moreover, MV is associated with a number of undesired side effects [32, 34,
51], which may produce morbidity or mortality [52]. There is good reason to believe
that avoiding endotracheal intubation and MV in patients with severe COPD exac-
erbation would substantially improve outcome.
ECCO2R is a promising, innovative strategy to improve the respiratory function
of patients with severe COPD exacerbation and obviate the need for MV [16, 17].
ECCO2R effectively lowers the alveolar ventilation required to eliminate CO2
(Fig. 10.3), thereby reestablishing a compensated balance between metabolic
10 ECCO2R in Obstructive Diseases: Evidence, Indications, and Exclusions 195
140 7,6 40
* 35
120 7,5
30
100 7,4 *
Respirations / min
mmHg
pH
25
80 7,3
* 20
60 7,2
15
40 7,1 10
20 7,0 5
ICU BL 5-8 13-16 21-24 ICU BL 5-8 13-16 21-24 ICU BL 5-8 13-16 21-24
NIV 0-4 0-12 17-20 NIV 0-4 9-12 17-20 NIV 0-4 9-12 17-20
Time intervals
Fig. 10.3 Changes over time of partial pressure of arterial carbon dioxide (PaCO2), pH, and respi-
ratory rate from admission in the intensive care unit (ICU), during noninvasive ventilation (NIV),
at baseline (BL), and after initiation of extracorporeal carbon dioxide removal (ECCO2R) in
patients with hypercapnic respiratory failure treated with the pumpless arteriovenous ECCO2R
device with the goal of avoiding endotracheal intubation. The data demonstrate the significant
decrease in respiratory rate secondary to the reduction of PaCO2 facilitated by the ECCO2R device.
Reproduced with permission from reference [81]
production and the impaired respiratory system. Several reports verify this hypoth-
esis, showing the successful use of ECCO2R in chronic obstructive respiratory dis-
eases [16, 17, 53, 54].
One of the first reports on the application of ECCO2R to support respiratory
function of a COPD patient was published in 1990 by Pesenti and colleagues [55].
A 19-year-old woman with bullous emphysema, recurrent pneumothoraxes, bilat-
eral air leaks, and lung infection had failed attempts at weaning after 28 days of
MV. A CT scan revealed the presence of pneumomediastinum, subcutaneous
emphysema, and hyperinflation of multiple lung bullae, likely sustained by pro-
longed MV and the high minute volumes required for CO2 clearance. This vicious
cycle was broken by low-blood-flow (0.4–0.6 L/min) venovenous ECCO2R, which
allowed the patient to transition to a spontaneous ventilatory mode and then to full
liberation from MV. A subsequent CT showed a decrease in the size of bullae with
expansion of the healthier lung parenchyma, suggesting that ECCO2R relieved
dynamic hyperinflation.
Since this first description, a number of observational studies have been reported
on the clinical application of ECCO2R in patients with COPD exacerbations to
facilitate early liberation from MV or to avoid MV after NIV failure [53, 56].
Devices characterized by venovenous configurations with a dual-lumen cannula
similar in size to a dialysis catheter are used in most of these clinical applications.
Abrams and colleagues [57] reported five older patients (age 73 ± 8.7 years) with
acute COPD exacerbation who failed NIV, requiring MV. After an average time of
16.5 ± 5.9 h of MV, ECCO2R was initiated. In all five patients, ECCO2R facilitated
extubation within 24 h of treatment (median duration of MV post ECCO2R = 4 h,
196 L. Del Sorbo et al.
range 1.5–21.5 h). The ECCO2R flow required to accomplish the goal of extubation
ranged between 1 and 1.7 L/min resulting in pH of 7.34–7.48. Once extubated,
patients were intensively rehabilitated while on ECCO2R, with a mean time to
ambulation of 29.4 ± 12.6 h after ECCO2R.
In a larger case series [58], ECCO2R was used to treat 30 patients who failed NIV
but refused endotracheal intubation. The mortality rate of this group of patients was
significantly decreased compared to that of 30 historical controls who received MV.
In another particularly interesting case report, Diehl and colleagues reported two
intubated patients with COPD exacerbation in whom ECCO2R lowered PaCO2 and
work of breathing [59]. Interestingly, during ECCO2R, the total CO2 production
decreased likely due to the reduced work of breathing, suggesting a direct and indi-
rect effect of ECCO2R on CO2 clearance and metabolic homeostasis.
It is worth highlighting that, in these case reports, ECCO2R was applied as an
adjunct to mechanical ventilation, subjecting patients to the potential complications
of two invasive strategies of respiratory support. Excitingly, a new approach pro-
poses that ECCO2R may be effective in preventing endotracheal intubation alto-
gether in patients failing NIV. This innovative strategy could represent an important
step toward ECCO2R as alternative to MV, with the important result of sidestepping
dynamic hyperinflation, barotrauma, hemodynamic impairment, ventilator-
associated pneumonia, immobility, critical illness polyneuropathy, and difficult
weaning secondary to ventilator-induced diaphragmatic dysfunction [20, 21, 60].
Avoiding these adverse effects could improve prognosis, especially considering the
severity of underlying lung disease and frequent presence of comorbidities [52, 61].
Along this line, in a multicenter study 25 patients supported with ECCO2R due
to the high risk of NIV failure after COPD exacerbation were compared with 21
historical controls, identified with a rigorous matching method (GenMatch), who
received only NIV [62]. ECCO2R applied in addition to NIV for an average of
29 ± 5 h with blood flow of 255 ± 78 ml/min significantly reduced the risk of intuba-
tion (hazard ratio 0.27, 95% CI, 0.07–0.98). Moreover, in the cohort of patients
treated with ECCO2R, hospital mortality was significantly lower compared with
matched controls (8% vs 33%, p = 0.035). ECCO2R-related complications, includ-
ing bleeding, clots in the circuit, pump malfunction, and membrane lung failure,
were reported in 13 patients (52%).
Another multicenter study [63] showed that, in 25 patients with COPD exacerba-
tion failing NIV and treated with ECCO2R, intubation was avoided in 56%. There
were 11 patients who received MV despite ECCO2R due to either progressive
hypoxemia or worsening hypoventilation. Duration of mechanical ventilation in
these patients was significantly lower (8 vs. 14 days, p = 0.02) than in 25 matched
controls selected by age, SAPS II, and pH at the time of NIV failure. However, ICU
and hospital length of stay, as well as mortality, did not differ between the two
groups. ECCO2R-related major adverse events were observed in 11 patients (44%).
The effects of ECCO2R have been investigated also in stable COPD patients with
chronic hypercapnia despite domiciliary NIV [64]. Ten COPD patients enrolled in
the home care NIV program for at least 6 months with PaCO2 >50 mmHg and pH
>7.35 were treated with a 24 h trial of ECCO2R. The treatment was prematurely
10 ECCO2R in Obstructive Diseases: Evidence, Indications, and Exclusions 197
Severe acute asthma is a major health burden, as its prevalence is growing world-
wide with increasing admissions to hospital and intensive care unit [19, 67]. The
in-hospital mortality rate due to acute exacerbation of asthma in the United States is
0.5%, yet this accounts for about one third of all asthma deaths. Moreover, mortality
reaches 8% for those requiring admission to the intensive care unit for intubation
and MV [19, 68].
The treatment of severe acute exacerbation of asthma consists of measures to
reverse the airflow obstruction. Ventilatory assistance is provided to patients refrac-
tory to medical treatment and developing hypoxemia or hypoventilation [19, 67].
However, while hypercapnia is observed in about 10% of asthma patients in the
acute setting, only a small percentage of these require invasive MV [69]. Interestingly,
from the combined data in two reports of 2655 asthma admissions over 20 years,
only 135 patients were diagnosed with life-threatening episodes, and a mere 48
were intubated [70, 71].
The accepted indication for MV is progressive hypercapnia with impending
hemodynamic decompensation, secondary to severe lung hyperinflation. The goal
of MV is to provide adequate gas exchange while waiting for airflow obstruction to
respond to bronchodilator therapy. However, MV may aggravate alveolar hyperin-
flation, thereby causing worsening hypercapnia, barotrauma, pneumothorax, and
further hemodynamic deterioration, with attendant risk of morbidity or mortality
[19, 33, 67].
198 L. Del Sorbo et al.
failure is profound, ECCO2R may not be sufficient and full extracorporeal mem-
brane oxygenation (ECMO) may provide a more adequate means of support
[75–77].
Extracorporeal membrane oxygenation was first applied in near-fatal asthma in
1981, followed by a progressively increasing number of reports [78]. In the interna-
tional extracorporeal life support organization (ELSO) registry [77], asthma was the
basis for ECMO in 24 out of 1257 adult patients between 1986 and 2006. Before
ECMO, the average pH was 7.17 ± 0.16, PaCO2 119.7 ± 58.1 mmHg, and PaO2/
FiO2 ratio 244 ± 180, despite invasive MV. Extracorporeal support was provided for
111.9 ± 71.2 h and 83.3% of patients survived. Complications were described in 19
of the 24 asthmatic subjects (79.2%). Between 1986 and 2007, 64 pediatric asthmat-
ics were treated with ECMO for life-threatening asthma, and 60 of those survived
(94%) [76]. Prior to initiation of ECMO, median pH was 6.96 (range 6.78–7.28),
median PaCO2 was 123 mmHg (range 70–237 mmHg), and PaO2 was 126 mmHg
(range 59–636 mmHg), despite MV. The median time of ECMO support was 94 h,
during which a remarkable number of cardiovascular, hemorrhagic, and mechanical
complications were reported. These results show that in both pediatric and adult
patients treated with ECMO for near-fatal asthma, hypercapnia, rather than hypox-
emia, was the dominant gas exchange problem, suggesting that less invasive
ECCO2R might have sufficed, perhaps with fewer complications.
In this vein, several case reports describe using pumpless AV-ECCO2R in adult
and pediatric patients with acute severe asthma [72–74]. More recently, a minimally
invasive venovenous ECCO2R device was used as an adjunct to MV in two adults
with asthma refractory to conventional therapies [79]. The first had a pH of 6.94,
PaCO2 of 147 mmHg, and PaO2 of 416 mmHg despite optimal MV. ECCO2R was
started with a blood flow at 1.3 L/min and sweep gas flow of 2.3 L/min, markedly
improving the respiratory acidosis to a pH of 7.24, PaCO2 of 56 mmHg, and PaO2
of 310 mmHg. After 24 h of ECCO2R support, the patient was liberated from MV
and progressively weaned from the extracorporeal device, before being successfully
discharged. The second patient not only had severe respiratory acidosis despite MV
but required vasopressors for hemodynamic support. ECCO2R contributed to nor-
malization of pH and PaCO2, while respiratory rate was lowered from 8 to 4 breaths/
min and tidal volume from 6 to 4 mL/kg. In one day, intrinsic PEEP resolved, shock
was reversed, and neuromuscular blocking agents were discontinued.
In summary, minimally invasive ECCO2R appears very promising for patients
with near-fatal asthma refractory to conventional treatment, but systematic evalua-
tion is needed to prove its efficacy and determine the true risks.
Exclusions
It is important to emphasize that the current lack of large, systematic clinical trials
providing definitive evidence of the efficacy of ECCO2R in improving clinically
significant outcomes makes the decision of including or excluding patients from the
200 L. Del Sorbo et al.
treatment with ECCO2R particularly complex. Since ECCO2R has become a mini-
mally invasive treatment, well-tolerated, and easy to manage, requiring similar
resources as renal dialysis, every patient with adequate indications should be con-
sidered as a potential candidate for its application, without absolute exclusions [4].
The balance between risks and benefits should be evaluated in each clinical case
before exposing a patient to this innovative respiratory support strategy. Despite
remarkable improvements in technology, there are still a number of clinically rele-
vant, unwanted side effects related to the application of ECCO2R, which may limit
its clinical consideration in patients with specific conditions [4, 6, 16]. The most
frequent complications caused by ECCO2R are mechanical, secondary to the poten-
tial vascular trauma during the intravascular insertion of the circuit cannula, and
bleeding due to the need of therapeutic anticoagulation to avoid thrombosis of the
circuit. Therefore, anatomical abnormalities or vascular diseases preventing the cor-
rect insertion of the ECCO2R cannula, preexisting risk of bleeding, contraindica-
tions to the administration of anticoagulants, or actual hemorrhagic diseases,
represent major conditions that may preclude the application of ECCO2R
(Table 10.1).
Moreover, additional exclusion criteria are applied in clinical trials investigating
the efficacy ECCO2R in improving the recovery from respiratory failure [57, 80,
81]. In this setting, conditions that limit recovery, such as terminal diseases, severe
malnutrition or cachexia, severely debilitated state, advanced malignancy, and
immunocompromised condition, are exclusionary (Table 10.1). An exception would
be a patient with end-stage, obstructive lung disease in whom ECCO2R might serve
as a bridge to lung transplant [82].
References
32. Reddy RM, Guntupalli KK. Review of ventilatory techniques to optimize mechanical venti-
lation in acute exacerbation of chronic obstructive pulmonary disease. Int J Chron Obstruct
Pulmon Dis. 2007;2:441–52.
33. Kao CC, Jain S, Guntupalli KK, Bandi V. Mechanical ventilation for asthma: a 10-year experi-
ence. J Asthma. 2008;45:552–6.
34. Ward NS, Dushay KM. Clinical concise review: mechanical ventilation of patients with
chronic obstructive pulmonary disease. Crit Care Med. 2008;36:1614–9.
35. Calverley PM, Walker P. Chronic obstructive pulmonary disease. Lancet. 2003;362:1053–61.
36. Chapman KR, Mannino DM, Soriano JB, Vermeire PA, Buist AS, Thun MJ, Connell C, Jemal
A, Lee TA, Miravitlles M, Aldington S, Beasley R. Epidemiology and costs of chronic obstruc-
tive pulmonary disease. Eur Respir J. 2006;27:188–207.
37. Mathers CD, Loncar D. Projections of global mortality and burden of disease from 2002 to
2030. PLoS Med. 2006;3:e442.
38. Burge S, Wedzicha JA. COPD exacerbations: definitions and classifications. Eur Respir J
Suppl. 2003;41:46s–53s.
39. Hurst JR, Vestbo J, Anzueto A, Locantore N, Mullerova H, Tal-Singer R, Miller B, Lomas
DA, Agusti A, Macnee W, Calverley P, Rennard S, Wouters EF, Wedzicha JA. Susceptibility to
exacerbation in chronic obstructive pulmonary disease. N Engl J Med. 2010;363:1128–38.
40. Patil SP, Krishnan JA, Lechtzin N, Diette GB. In-hospital mortality following acute exacerba-
tions of chronic obstructive pulmonary disease. Arch Intern Med. 2003;163:1180–6.
41. Donaldson GC, Wedzicha JA. COPD exacerbations 1: epidemiology. Thorax. 2006;61:164–8.
42. Soler-Cataluna JJ, Martinez-Garcia MA, Roman Sanchez P, Salcedo E, Navarro M, Ochando
R. Severe acute exacerbations and mortality in patients with chronic obstructive pulmonary
disease. Thorax. 2005;60:925–31.
43. Quon BS, Gan WQ, Sin DD. Contemporary management of acute exacerbations of COPD: a
systematic review and meta-analysis. Chest. 2008;133:756–66.
44. Chronic Obstructive Pulmonary Disease. National clinical guideline on management of chronic
obstructive pulmonary disease in adults in primary and secondary care. Thorax. 2004;59(Suppl
1):1–232.
45. Pauwels RA, Buist AS, Calverley PM, Jenkins CR, Hurd SS. Global strategy for the diagnosis,
management, and prevention of chronic obstructive pulmonary disease. NHLBI/WHO Global
Initiative for Chronic Obstructive Lung Disease (GOLD) Workshop summary. Am J Respir
Crit Care Med. 2001;163:1256–76.
46. Organized jointly by the American Thoracic Society, the European Respiratory Society, the
European Society of Intensive Care Medicine, and the Société de Réanimation de Langue
Française, and approved by ATS Board of Directors, December 2000. International Consensus
Conferences in Intensive Care Medicine: noninvasive positive pressure ventilation in acute
Respiratory failure. Am J Respir Crit Care Med. 2001;163:283–91.
47. Brochard L, Mancebo J, Wysocki M, Lofaso F, Conti G, Rauss A, Simonneau G, Benito S,
Gasparetto A, Lemaire F, et al. Noninvasive ventilation for acute exacerbations of chronic
obstructive pulmonary disease. N Engl J Med. 1995;333:817–22.
48. Ram FS, Picot J, Lightowler J, Wedzicha JA. Non-invasive positive pressure ventilation for
treatment of respiratory failure due to exacerbations of chronic obstructive pulmonary disease.
Cochrane Database Syst Rev. 2004;(1):CD004104.
49. Scala R, Naldi M, Archinucci I, Coniglio G, Nava S. Noninvasive positive pressure ventila-
tion in patients with acute exacerbations of COPD and varying levels of consciousness. Chest.
2005;128:1657–66.
50. Chandra D, Stamm JA, Taylor B, Ramos RM, Satterwhite L, Krishnan JA, Mannino D,
Sciurba FC, Holguin F. Outcomes of noninvasive ventilation for acute exacerbations of chronic
obstructive pulmonary disease in the United States, 1998–2008. Am J Respir Crit Care Med.
2012;185:152–9.
51. Rello J, Ollendorf DA, Oster G, Vera-Llonch M, Bellm L, Redman R, Kollef MH, Group
VAPOSA. Epidemiology and outcomes of ventilator-associated pneumonia in a large US data-
base. Chest. 2002;122:2115–21.
10 ECCO2R in Obstructive Diseases: Evidence, Indications, and Exclusions 203
52. Ai-Ping C, Lee KH, Lim TK. In-hospital and 5-year mortality of patients treated in the ICU for
acute exacerbation of COPD: a retrospective study. Chest. 2005;128:518–24.
53. Sklar MC, Beloncle F, Katsios CM, Brochard L, Friedrich JO. Extracorporeal carbon dioxide
removal in patients with chronic obstructive pulmonary disease: a systematic review. Intensive
Care Med. 2015;41:1752–62.
54. d’Andrea A, Banfi C, Bendjelid K, Giraud R. The use of extracorporeal carbon dioxide
removal in acute chronic obstructive pulmonary disease exacerbation: a narrative review. Can
J Anaesth. 2020;67:462–74.
55. Pesenti A, Rossi GP, Pelosi P, Brazzi L, Gattinoni L. Percutaneous extracorporeal CO2 removal
in a patient with bullous emphysema with recurrent bilateral pneumothoraces and respiratory
failure. Anesthesiology. 1990;72:571–3.
56. Taccone FS, Malfertheiner MV, Ferrari F, Di Nardo M, Swol J, Broman LM, Vercaemst L,
Barrett N, Pappalardo F, Belohlavek J, Mueller T, Lorusso R, Belliato M, et al. Extracorporeal
CO2 removal in critically ill patients: a systematic review. Minerva Anestesiol. 2017;83:762–72.
57. Abrams DC, Brenner K, Burkart KM, Agerstrand CL, Thomashow BM, Bacchetta M, Brodie
D. Pilot study of extracorporeal carbon dioxide removal to facilitate extubation and ambu-
lation in exacerbations of chronic obstructive pulmonary disease. Ann Am Thorac Soc.
2013;10:307–14.
58. Morelli A, D’Egidio A, Orecchioni A, Alessandri F, Mascia L, Ranieri VM. Extracorporeal
CO2 removal in hypercapnic patients who fail noninvasive ventilation and refuse endotracheal
intubation: a case series. Intensive Care Med Exp. 2015;3:824.
59. Diehl JL, Piquilloud L, Richard JM, Mancebo J, Mercat A. Effects of extracorporeal carbon
dioxide removal on work of breathing in patients with chronic obstructive pulmonary disease.
Intensive Care Med. 2016;42:951–2.
60. Bonin F, Sommerwerck U, Lund LW, Teschler H. Avoidance of intubation during acute
exacerbation of chronic obstructive pulmonary disease for a lung transplant candidate using
extracorporeal carbon dioxide removal with the Hemolung. J Thorac Cardiovasc Surg.
2013;145:e43–4.
61. Menzies R, Gibbons W, Goldberg P. Determinants of weaning and survival among patients
with COPD who require mechanical ventilation for acute respiratory failure. Chest.
1989;95:398–405.
62. Del Sorbo L, Pisani L, Filippini C, Fanelli V, Fasano L, Terragni P, Dell’Amore A, Urbino R,
Mascia L, Evangelista A, Antro C, D’Amato R, Sucre MJ, Simonetti U, Persico P, Nava S,
Ranieri VM. Extracorporeal Co2 removal in hypercapnic patients at risk of noninvasive venti-
lation failure: a matched cohort study with historical control. Crit Care Med. 2015;43:120–7.
63. Braune S, Sieweke A, Brettner F, Staudinger T, Joannidis M, Verbrugge S, Frings D, Nierhaus
A, Wegscheider K, Kluge S. The feasibility and safety of extracorporeal carbon dioxide
removal to avoid intubation in patients with COPD unresponsive to noninvasive ventilation
for acute hypercapnic respiratory failure (ECLAIR study): multicentre case-control study.
Intensive Care Med. 2016;42:1437–44.
64. Pisani L, Nava S, Desiderio E, Polverino M, Tonetti T, Ranieri VM. Extracorporeal CO2
removal (ECCO2R) in patients with stable COPD with chronic hypercapnia: a proof-of-
concept study. Thorax. 2020;75:897–900.
65. Garcia JP, Kon ZN, Evans C, Wu Z, Iacono AT, McCormick B, Griffith BP. Ambulatory veno-
venous extracorporeal membrane oxygenation: innovation and pitfalls. J Thorac Cardiovasc
Surg. 2011;142:755–61.
66. Moazami N, Sun B, Feldman D. Stable patients on left ventricular assist device support have
a disproportionate advantage: time to re-evaluate the current UNOS policy. J Heart Lung
Transplant. 2011;30:971–4.
67. Phipps P, Garrard CS. The pulmonary physician in critical care. 12: Acute severe asthma in the
intensive care unit. Thorax. 2003;58:81–8.
68. Krishnan V, Diette GB, Rand CS, Bilderback AL, Merriman B, Hansel NN, Krishnan
JA. Mortality in patients hospitalized for asthma exacerbations in the United States. Am J
Respir Crit Care Med. 2006;174:633–8.
204 L. Del Sorbo et al.
69. McFadden ER Jr, Lyons HA. Arterial-blood gas tension in asthma. N Engl J Med.
1968;278:1027–32.
70. Braman SS, Kaemmerlen JT. Intensive care of status asthmaticus. A 10-year experience.
JAMA. 1990;264:366–8.
71. Trawick DR, Holm C, Wirth J. Influence of gender on rates of hospitalization, hospital course,
and hypercapnea in high-risk patients admitted for asthma : a 10-year retrospective study at
Yale-New Haven Hospital. Chest. 2001;119:115–9.
72. Jung C, Lauten A, Pfeifer R, Bahrmann P, Figulla HR, Ferrari M. Pumpless extracorporeal
lung assist for the treatment of severe, refractory status asthmaticus. J Asthma. 2011;48:111–3.
73. Conrad SA, Green R, Scott LK. Near-fatal pediatric asthma managed with pumpless arteriove-
nous carbon dioxide removal. Crit Care Med. 2007;35:2624–9.
74. Elliot SC, Paramasivam K, Oram J, Bodenham AR, Howell SJ, Mallick A. Pumpless extracor-
poreal carbon dioxide removal for life-threatening asthma. Crit Care Med. 2007;35:945–8.
75. Shapiro MB, Kleaveland AC, Bartlett RH. Extracorporeal life support for status asthmaticus.
Chest. 1993;103:1651–4.
76. Hebbar KB, Petrillo-Albarano T, Coto-Puckett W, Heard M, Rycus PT, Fortenberry
JD. Experience with use of extracorporeal life support for severe refractory status asthmaticus
in children. Crit Care. 2009;13:R29.
77. Mikkelsen ME, Woo YJ, Sager JS, Fuchs BD, Christie JD. Outcomes using extracorporeal life
support for adult respiratory failure due to status asthmaticus. ASAIO J. 2009;55:47–52.
78. MacDonnell KF, Moon HS, Sekar TS, Ahluwalia MP. Extracorporeal membrane oxygenator
support in a case of severe status asthmaticus. Ann Thorac Surg. 1981;31:171–5.
79. Brenner K, Abrams DC, Agerstrand CL, Brodie D. Extracorporeal carbon dioxide removal
for refractory status asthmaticus: experience in distinct exacerbation phenotypes. Perfusion.
2014;29:26–8.
80. Burki NK, Mani RK, Herth FJ, Schmidt W, Teschler H, Bonin F, Becker H, Randerath WJ,
Stieglitz S, Hagmeyer L, Priegnitz C, Pfeifer M, Blaas SH, Putensen C, Theuerkauf N, Quintel
M, Moerer O. A novel extracorporeal CO(2) removal system: results of a pilot study of hyper-
capnic respiratory failure in patients with COPD. Chest. 2013;143:678–86.
81. Kluge S, Braune SA, Engel M, Nierhaus A, Frings D, Ebelt H, Uhrig A, Metschke M,
Wegscheider K, Suttorp N, Rousseau S. Avoiding invasive mechanical ventilation by extracor-
poreal carbon dioxide removal in patients failing noninvasive ventilation. Intensive Care Med.
2012;38:1632–9.
82. Del Sorbo L, Boffini M, Rinaldi M, Ranieri VM. Bridging to lung transplantation by extracor-
poreal support. Minerva Anestesiol. 2012;78:243–50.
Chapter 11
ECMO as a Bridge to Lung
Transplantation
Introduction
Historical Aspects
Over the last four decades, lung transplantation has evolved from an experimental
technique to an established option for patients with terminal pulmonary disease and
disease of the pulmonary circulation. Technical advances in isolated lung transplan-
tation were driven especially by Cooper and colleagues in Toronto. In the early era
of lung transplantation, ECMO was seen as a contraindication for transplantation as
mechanical support was marked by complications and poor outcome. In particular,
bleeding, hemolysis, infection, and technical complications marked the postopera-
tive course in the early days of mechanical support in lung transplantation patients.
Modes of Support
In general, patients on the transplant waiting list with a deteriorating clinical status
may be eligible for ECLS. In patients with deteriorating pulmonary function, the
choice of support is either the pumpless interventional lung assist (iLA) or the more
common VV-ECMO. In the setting of the iLA, systemic pressure drives blood from
the femoral artery via a membrane oxygenator to the femoral vein. In the VV-ECMO
setting, oxygen-poor blood is drawn from the femoral vein and, after oxygenation,
returned in either the jugular or the contralateral femoral vein. In VV-ECMO, blood
is actively pumped using a centrifugal pump. In case of hemodynamic instability, or
patients with strained right ventricular function, the use of venoarterial (VA-)
ECMO should be considered. In this type of support, venous blood is drawn from
the right heart via the inferior vena cava, and oxygenated blood is returned in either
the femoral or subclavian artery. In rare cases, the arterial cannula can be placed in
the ascending aorta. In extreme cases where both cardiac and pulmonary failures are
present, the choice for a combined veno-venoarterial (VVA-) ECMO may be con-
sidered. This constellation however is reserved for extremely ill patients and requires
a high level of ECMO expertise.
iLA has proven useful in patients awaiting lung transplantation with refractory
hypercapnia [5]. Although significant reduction of carbon dioxide has been reported
by several authors, to date no randomized clinical trials comparing iLA and
VV-ECMO are provided in the literature. Several limitations of this support system
have been discussed. Due to the lack of a pump, the flow rate is limited to 2.5 L/min
which allows for adequate removal of CO2 but is not sufficient for oxygenation.
Furthermore, cannulation of the femoral artery can occlude the vessel leading to
distal leg ischemia. In the worst case, amputation of the limb can follow [6, 7]. Due
to the significant limitations, this system has mainly been replaced by VV-ECMO.
Veno-venous ECMO
In the case of VV-ECMO, support is mainly indicated for patients with severe respi-
ratory failure. Depending on the type of respiratory failure, either the classical
VV-ECMO with two cannulas or a dual-lumen, single cannula VV-ECMO is pre-
ferred. In patients with predominantly hypercapnic respiratory failure, relatively
low flow rates may be sufficient to provide normalization of the CO2 concentrations.
To achieve this, a dual-lumen cannula can be inserted into the jugular vein (usually
on the right side). Echocardiographic or fluoroscopic guidance for optimal
208 C. Kuehn and R. Natanov
positioning is necessary for this type of cannulation (see Chap. 4). Although the
flow rate is usually too low to achieve oxygenation, it is effective in the removal of
excess carbon dioxide. This usually leads to the normalization of respiratory acido-
sis and may reduce the need for vasoactive agents. Furthermore, as a dual-lumen
cannula may be inserted under local anesthesia, sedation and intubation of the
patient may be avoided. Physical therapy and patient mobilization are very much
possible under this type of ECLS (see Chap. 14). A significant limitation is the rela-
tively low blood flow rate (usually up to 3.5 L/min), therefore limiting the rate of
blood oxygenation. Patients with severe, end-stage respiratory disease due to com-
bined hypoxic and hypercapnic respiratory failure may not be sufficiently treated
with a single cannula only. For these patients, the classic VV-ECMO with cannula-
tion of both the jugular and femoral vein may be a preferred option. In this setting,
high blood flow (up to 6 L/min) may be achieved, providing near-complete replace-
ment of native lung function. Classical, two-cannula VV-ECMO may also be initi-
ated in the awake patient, provided the patient is cooperative. Large diameter
cannulas are placed in the jugular and femoral vein (both usually on the right side).
Blood is drained from the inferior vena cava, oxygenated, and returned to the right
atrium. Optimal positioning of the cannulas provides maximal blood flow without
significant recirculation of oxygenated blood. VV-ECMO for BTT may be suitable
in patients with end-stage pulmonary fibrosis, sarcoidosis, or bronchiolitis obliter-
ans. As COVID-19 has become a serious source of pulmonary disease, our center
has recently bridged several patients to transplantation with severe COVID-19-
associated ARDS. To prevent loss of muscle mass and deterioration of the general
patient’s condition, early implantation of VV-ECMO is advised. However, to avoid
the situation of a bridge to nowhere, the chances for listing for lung transplantation
must be critically evaluated before initiation of VV-ECMO (see Chap. 17).
Venoarterial ECMO
hemodynamics. Cannulas are typically implanted in the right groin using ultrasound
guidance. Local anesthesia may be used to avoid the need for sedation and intuba-
tion. In the ideal setting, VA-ECMO is implanted electively. Intraoperative continu-
ation of VA-ECMO may be necessary after successful BTT.
In our clinic, examination and evaluation of patients for transplantation are done by
a multidisciplinary team. Patient education concerning the waiting period, technical
aspects of the procedure, and long-term prognosis are pivotal in the preoperative
patient preparation. Patients are furthermore educated in the possibility of deteriora-
tion while on the waiting list and the available invasive therapeutic options. As each
patient is a unique individual, there is no simple algorithm to address these possi-
bilities. In the evaluation of the possibility of ECLS as BTT, patient age and comor-
bidities are significant variables. Patients with single organ failure are usually
considered as potential candidates for ECLS as BTT. Usually, there are two sce-
narios where ECLS can be considered as a potential BTT for lung transplantation.
In the first type of patient, there is an isolated gas exchange failure, thus prompting
VV-ECMO. The second type of patient presents with right-sided heart failure and is
suitable for VA-ECMO. In ideal circumstances, patients and physicians recognize
deterioration and end-stage failure. In this case, actions can be taken before the need
210 C. Kuehn and R. Natanov
Pulmonary Failure
Increased strain on the right ventricle due to high pulmonary pressures may lead to
right-sided decompensation. Clinical evaluation of the patient involves echocar-
diography, hemodynamic monitoring, and clinical chemistry (see Chap. 7).
Echocardiographic signs include enlargement of the right ventricle and flattening of
the interventricular septum, the so-called D-sign. Invasive measurements show high
central venous pressure, low central venous saturation, and low cardiac output.
Acute kidney failure and acute liver dysfunction are common end-organ failures in
right heart failure. In these patients, implantation of VA-ECMO usually stabilizes
the hemodynamic situation immediately and restores end-organ function. Continued
use of VA-ECMO during and after transplantation has been associated with an
improved postoperative outcome when compared to patients with mechanical ven-
tilation alone. Midterm results show a benefit for intra- and postoperative use of
VA-ECMO in terms of reduced mortality and improvement of left ventricular func-
tion 1 year after transplantation [11]. This effect was also seen in patients trans-
planted due to pulmonary fibrosis combined with pulmonary hypertension [12]. In
contrast to the early practice of VA-ECMO, where the arterial cannula was placed
directly into the pulmonary artery, peripheral cannulation in the femoral artery and
vein is currently the standard. When done percutaneously, easy removal at the bed-
side avoids the need for a second operation and thus provides the possibility for
early postoperative weaning from ventilation.
In most cases, listed patients are at the transplantation center at the time of decom-
pensation so that ECMO may be initiated electively. Sometimes, however, patients
are in an external center or are not yet listed at the time of decompensation. In these
patients, interdisciplinary evaluation should be done early, preferably prior to intu-
bation. When eligible for transplantation, patients must be evaluated for the type of
ECLS support and a retrieval team is sent to the destination. After careful evaluation
of the patient, cannulation is done in typical manner. Interhospital patient transport
on ECLS is quite possible, even in awake patients [13]. Transport can be by air or
land and enables the possibility of listing once the patient has arrived at the trans-
plantation center. In patients with pulmonary hypertension, it is preferable to arrange
transport while the patient is awake.
212 C. Kuehn and R. Natanov
As previously stated, classical therapy options for patients with end-stage pulmo-
nary failure involved sedation and mechanical ventilation. Follow-up of these
patients however showed an extremely poor outcome; survival rates 1 year after
transplantation were less than 50%. With such poor prognosis, the question arises
whether ELCS has a place. Due to the scarcity of available donor organs, all efforts
should be taken to maximize postoperative success rates in high-risk patients. Early
patient stabilization with ECLS could provide the tool needed for this. In our experi-
ence, we have learned that sedation, intubation, and mechanical ventilation are
associated with extremely poor postoperative survival. Therefore, our practice has
evolved toward avoiding mechanical ventilation. Should the patient already be on
mechanical ventilation, then weaning should be prioritized enabling them to eat,
drink, exercise, and communicate with medical staff. Our strategy to keep patients
awake on ECLS has been shown to be beneficial when compared to patients on
mechanical ventilation.
Intraoperative VA-ECMO
Postoperative ELCS
To maximize organ donor acceptance, novel organ perfusion systems have been
developed in recent years. In ex vivo lung perfusion, donor lungs are connected to a
system of tubes and perfused to evaluate marginal organs. Parameters such as oxy-
gen uptake, lung elasticity, perfusion, and ventilation pressures determine the suit-
ability for transplantation. Organs that seem to be marginal at first glance may still
be accepted once an adequate function is confirmed using the organ perfusion sys-
tem. Furthermore, due to continuous perfusion of the organ with a normothermic
solution containing nutrients, cold ischemic time can be significantly reduced and
donor organ damage may be ameliorated. To date, extensive research in the normo-
thermic perfusion system has not shown a benefit for medium-term organ function
[15]. Further development of this system into a portable system, the organ care
system (OCS), has led to a pilot study between the Hannover group in cooperation
with colleagues from Madrid [16]. In this study, 12 donor lungs were transplanted
after initial cold perfusion followed by warm perfusion and ventilation using the
OCS. Follow-up data has not shown a benefit in short-term survival [17]. However,
further research must be done to determine whether the use of the OCS has a benefit
in organ function in the long term.
214 C. Kuehn and R. Natanov
Conclusions
References
1. Jurmanm MJ, Haverich A, Demertzis S, Schaefers HJ, Wagner TO, Borst HG. Extracorporeal
membrane oxygenation as a bridge to lung transplantation. Eur J Cardiothorac Surg.
1991;5(2):94–7.
2. Fischer S, Simon AR, Welte T, Hoeper MM, Meyer A, Tessmann R, et al. Bridge to lung
transplantation with the novel pumpless interventional lung assist device NovaLung. J Thorac
Cardiovasc Surg. 2006;131(3):719–23.
3. Smits JMA, Mertens BJA, Van Houwelingen HC, Haverich A, Persijn GG, Laufer G. Predictors
of lung transplant survival in eurotransplant. Am J Transplant. 2003;3(11):1400–6.
4. Agati S, Ciccarello G, Fachile N, Scappatura RM, Grasso D, Salvo D, et al. DIDECMO: a new
polymethylpentene oxygenator for pediatric extracorporeal membrane oxygenation. ASAIO
J. 1992;52(5):509–12.
5. Fischer S, Hoeper MM, Bein T, Simon AR, Gottlieb J, Wisser W, et al. Interventional lung
assist: a new concept of protective ventilation in bridge to lung transplantation. ASAIO
J. 2008;54(1):3–10.
6. Kinaschuk K, Bozso SJ, Halloran K, Kapasi A, Jackson K, Nagendran J. Mechanical circula-
tory support as a bridge to lung transplantation: a single Canadian institution review. Can
Respir J. 2017;2017:5947978.
7. Secretariat MA. Extracorporeal lung support technologies – bridge to recovery and bridge to
lung transplantation in adult patients. Ont Health Technol Assess Ser. 2010;10:1–47.
8. Hoeper MM, Tudorache I, Kühn C, Marsch G, Hartung D, Wiesner O, et al. Extracorporeal
membrane oxygenation watershed. Circulation. 2014;130:864–5.
9. Fuehner T, Kuehn C, Hadem J, Wiesner O, Gottlieb J, Tudorache I, et al. Extracorporeal mem-
brane oxygenation in awake patients as bridge to lung transplantation. Am J Respir Crit Care
Med. 2012;185(7):763–8.
10. Ius F, Natanov R, Salman J, Kuehn C, Sommer W, Avsar M, et al. Extracorporeal membrane
oxygenation as a bridge to lung transplantation may not impact overall mortality risk after
transplantation: results from a 7-year single-centre experience. Eur J Cardiothorac Surg.
2018;54(2):334–40.
11. Salman J, Ius F, Sommer W, Siemeni T, Kuehn C, Avsar M, et al. Mid-term results of bilateral
lung transplant with postoperatively extended intraoperative extracorporeal membrane oxy-
genation for severe pulmonary hypertension. Eur J Cardiothorac Surg. 2017;52(1):163–70.
11 ECMO as a Bridge to Lung Transplantation 215
12. Salman J, Bernhard BA, Ius F, Poyanmehr R, Sommer W, Aburahma K, et al. Intraoperative
extracorporeal circulatory support in lung transplantation for pulmonary fibrosis. Ann Thorac
Surg. 2021;111(4):1316–24.
13. Fleissner F, Mogaldea A, Martens A, Natanov R, Rümke S, Salman J, et al. ECLS supported
transport of ICU patients: does out-of -house implantation impact survival? J Cardiothorac
Surg. 2021;16(1):1–6.
14. Ius F, Kuehn C, Tudorache I, Sommer W, Avsar M, Boethig D, et al. Lung transplantation on
cardiopulmonary support: venoarterial extracorporeal membrane oxygenation outperformed
cardiopulmonary bypass. J Thorac Cardiovasc Surg. 2012;144(6):1510–6.
15. Nilsson T, Wallinder A, Henriksen I, Nilsson JC, Ricksten SE, Møller-Sørensen H, et al. Lung
transplantation after ex vivo lung perfusion in two Scandinavian centres. Eur J Cardiothorac
Surg. 2019;55(4):766–72.
16. Warnecke G, Moradiellos J, Tudorache I, Kühn C, Avsar M, Wiegmann B, et al. Normothermic
perfusion of donor lungs for preservation and assessment with the organ care system lung
before bilateral transplantation: a pilot study of 12 patients. Lancet. 2012;380(9856):1851–8.
17. Warnecke G, Van Raemdonck D, Smith MA, Massard G, Kukreja J, Rea F, et al. Normothermic
ex-vivo preservation with the portable Organ Care System Lung device for bilateral lung trans-
plantation (INSPIRE): a randomised, open-label, non-inferiority, phase 3 study. Lancet Respir
Med. 2018;6(5):357–67.
Chapter 12
Daily Management of Patients on VV
ECMO
Daily management consists of judging gas exchange, reviewing organ system func-
tion, identifying and preventing complications, and ensuring communication
between the ECMO managing LIP, ECMO specialists, bedside nurses, respiratory
and physical therapists, and the patient or surrogate decision-makers (see Table 12.1).
Gas Exchange
The fundamental purpose of ECMO is to accomplish gas exchange when the native
lung cannot. The factors underlying adequate oxygen and carbon dioxide exchange
are described in Chap. 1 and these should be reviewed systematically on (at least) a
daily basis.
Oxygenation Assuming that the patient’s native cardiovascular system is intact, an
acceptable arterial oxygen saturation (SaO2) assures adequate systemic oxygen-
ation. Arterial saturation is measured using pulse oximetry (SpO2) or arterial blood
gas analysis. In patients on VV ECMO, values are often lower than typically toler-
ated in mechanically ventilated patients. For example, an SpO2 > 75% may be toler-
C. Rappaport (*)
Division of Pulmonary, Critical Care and Occupational Medicine, University of Iowa Hospital
and Clinics, Iowa City, IA, USA
e-mail: [email protected]
K. Rappaport
University of Iowa Hospital and Clinics, Iowa City, IA, USA
e-mail: [email protected]
flow compromise follows PEEP reduction or lowered sweep gas flow (and accord-
ingly greater inspiratory efforts), attention to the ventilator and patient’s work of
breathing is necessary. Abolishing inspiratory effort by raising sweep gas flow or
infusing narcotics may be simpler and more effective than giving repeated fluid
boluses.
The other common basis for poor circuit blood flow is cannula malposition. One
common example is migration of the distal aspect of a dual-lumen internal jugular
(dl Vj-V) into the hepatic vein. This is readily identified with POCUS, using a sub-
xiphoid or transhepatic window, and aided by color Doppler flow imaging.
Inadvertent withdrawal of a cannula, as can occur during turning, mobilization, pro-
ning, or transport of the EMCO patient may also be to blame and can be detected
with POCUS or radiographic imaging. The daily survey for patients with dl Vj-V
cannulation includes using color Doppler to identify the return jet [directed toward
the tricuspid valve (TV)]. The operator should ensure that the jet has not rotated
away from the TV, migrated distally (e.g., hepatic vein), or shifted proximally (in
the superior vena cava). For those with Vf-Vf cannulation, color Doppler can iden-
tify the return jet near or in the right atrium, rather than distally (superior vena cava)
or proximally (toward the hepatic vein). If POCUS windows do not allow adequate
visualization, chest (for Vj) or abdominal (for Vf) plain radiographs can detect gross
malposition but not the orientation of the return jet.
Membrane lung function: With a fully functional oxygenator, the outlet PO2 is
usually >250 mmHg and outlet saturation (SPOSTO2) is 1.0. Additional key indicators
of membrane lung function include the pre-oxygenator-to-post-oxygenator PCO2
difference (normally >10 mmHg), which signals clearance of CO2, and the mem-
brane pressure drop, normally <50 mmHg (see also Chap. 9). Most often deteriora-
tion in membrane lung function develops over many hours, even days, but
occasionally can evolve over a much shorter period, producing a crisis (see
Chap. 13).
Recirculation: Recirculation occurs when highly oxygenated blood, having just
returned to the body, is withdrawn again via the drainage cannula, reducing the
efficiency of ECMO support. At times, recirculation can be so dramatic as to pro-
duce critical systemic hypoxemia. More often, the ECMO specialist notices brief
220 C. Rappaport and K. Rappaport
appropriate. If the plan, however, is to support the patient as a bridge to lung heal-
ing, lowering sweep gas flow is counterproductive. A better approach is to ignore
the PaCO2 (and pH), leave the sweep gas flow at a level that accommodates a low
work of breathing, and wait until native lung function has improved. Some patients
exhibit high work of breathing despite raising the sweep gas flow, even once
PaCO2 is dramatically lowered. Some conditions to include in the differential
diagnosis of such patients are agitation, anxiety, severe pain, developing shock,
fever, or systemic inflammation. Without identifying and treating these underlying
conditions, high respiratory drive may persist despite increased extracorporeal
CO2 removal [4].
As the patient’s trajectory stabilizes, greater respiratory work may be acceptable,
or even desired, in accord with the principles of weaning described in Chap. 15.
Spontaneous breathing while on VV ECMO has been shown to reduce ventilation-
perfusion (V/Q) mismatch, preserve respiratory muscle and diaphragm tone, and
preserve functional residual capacity [5]. The managing physician should assess
daily the status of native lung function and the role for continuing ECMO support.
If native lung function is improving, then support can be shifted from ECMO to the
patient and ventilator. Reduced sweep gas flow should be monitored by clinical
exam, ventilator graphics, and P0.1 as described above. If the inevitable increase in
respiratory effort is well-tolerated, this suggests that further reductions in sweep gas
flow may be achieved.
Ancillary Therapies
Antithrombotic Therapy
The managing physician should review the coagulation profile, target measures for
antithrombotic therapy, and any evidence of bleeding that might prompt a lower
antithrombotic goal or cessation of anticoagulation altogether. The antithrombotic
regimen (heparin, bivalirudin, argatroban) and the measures of antithrombotic effect
(aPTT, thromboelastogram, anti-factor Xa levels, activated clotting time) are center-
specific as further described in Chap. 8. The anticoagulation titration algorithm
should be reviewed with the ECMO specialist on daily rounds to ensure the level of
anticoagulation is appropriate. Signs of active, external bleeding should prompt the
team to either pause anticoagulation or switch to a lower-dose algorithm. Internal
bleeding is harder to appreciate, but a significant hemorrhage can cause a reduction
in ECMO circuit flow, reduced central venous saturation, lactic acidosis, fall in
hemoglobin concentration, or signs of shock.
Low platelet counts often prompt concern for heparin-induced thrombocytope-
nia (HIT). However, thrombocytopenia is quite common in patients on ECMO,
whereas proven HIT is quite uncommon, with a reported incidence of <1% [6]. If
HIT is diagnosed or strongly suspected, heparin must be stopped and treatment with
argatroban or bivalirudin initiated. Furthermore, many circuits are heparin-coated,
222 C. Rappaport and K. Rappaport
thus a circuit change is necessary to eliminate all contact with heparin. Given the
nontrivial implications of treating for HIT, testing should be done only if the clinical
suspicion is quite high (see Chap. 8).
patients has also been shown to reduce muscle wasting [21]. These findings likely
apply also to ECMO patients. Mobilization is feasible and safe when conducted in
experienced centers, even for patients with femoral cannulas (see Chap. 14). Given
the complexity of lines and devices, a multidisciplinary team of nurse, respiratory
therapist, ECMO specialist, physical therapist, occupational therapist, and intensiv-
ist is advisable [22].
Hemodynamics
Neurological Function
New neurological events are common on VV ECMO and portend a poor prognosis.
Intracranial bleeding occurs in approximately 5% of patients [29]. Multiple mecha-
nisms are likely in play, but AKI and rapid reductions in PaCO2 after ECMO initia-
tion (with subsequent alterations in cerebral blood flow) have been independently
associated with intracranial bleeding [30]. Interestingly, intracranial bleeding has
not been associated with anticoagulation profile, age, or comorbid conditions. The
daily survey should include a detailed neurological assessment, preferably after
sedative interruption.
12 Daily Management of Patients on VV ECMO 225
Goals of Therapy
References
1. The National Heart, Lung, and Blood Institute Acute Respiratory Distress Syndrome (ARDS)
Clinical Trials Network. Comparison of two fluid-management strategies in acute lung injury.
N Engl J Med. 2006;354:2564–75.
2. Mauri T, Grasselli G, Suriano G, Eronia N, et al. Control of respiratory drive and effort in
extracorporeal membrane oxygenation patients recovering from severe acute respiratory dis-
tress syndrome. Anesthesiology. 2016;125:159–67.
3. Telias I, Junhasavasdikul D, Rittayamai N, Piquilloud L, et al. Airway occlusion pressure as
an estimate of respiratory drive and inspiratory effort during assisted ventilation. Am J Respir
Crit Care Med. 2020;201(9):1086–98.
4. Crotti S, Bottino N, Ruggeri GM, Spinelli E, et al. Spontaneous breathing during extracorpo-
real membrane oxygenation in acute respiratory failure. Anesthesiology. 2017;126:678–87.
10.1097.
5. Langer T, Vecchi V, Belenkiy SM, Cannon JW, et al. Extracorporeal gas exchange and spon-
taneous breathing for the treatment of acute respiratory distress syndrome: an alternative to
mechanical ventilation? Crit Care Med. 2014;42(3):e211–20.
6. Natt B, Hypes C, Basken R, Malo J, et al. Suspected heparin-induced thrombocytope-
nia in patients receiving extracorporeal membrane oxygenation. J Extra Corpor Technol.
2017;49:54–8.
7. Chen Y, Tsai F, Fang J, Yang C. Acute kidney injury in adults receiving extracorporeal mem-
brane oxygenation. J Formos Med Assoc. 2014;113:778–85.
8. Kielstein J, Heiden A, Beutel G, Gottlieb J, et al. Renal function and survival in 200 patients
undergoing ECMO therapy. Nephrol Dial Transplant. 2013;28:86–90.
9. Timing of Initiation of Renal-Replacement Therapy in Acute Kidney Injury The STARRT-AKI
Investigators for the Canadian Critical Care Trials Group, the Australian and New Zealand
Intensive Care Society Clinical Trials Group, the United Kingdom Critical Care Research
Group, the Canadian Nephrology Trials Network, and the Irish Critical Care Trials Group. N
Engl J Med. 2020;383:240–51.
10. Cove ME, MacLaren G, Brodie D, Kellum J. Optimising the timing of renal replacement
therapy in acute kidney injury. Crit Care. 2021;25:184.
11. Askenazi DJ, Selewski DT, Paden ML, et al. Renal replacement therapy in critically ill patients
receiving extracorporeal membrane oxygenation. Clin J Am Soc Nephrol. 2012;7(8):1328–36.
12. Dyk M. The use of CRRT in ECMO patients the Egyptian. J. Crit. Care Med. 2018;6(3):95–100.
13. Guérin C, Reignier J, Richard JC, Beuret P, et al. PROSEVA Study Group. Prone positioning
in severe acute respiratory distress syndrome. N Engl J Med. 2013;368(23):2159–68.
14. Jozwiak M, Teboul JL, Anguel N, Persichini R, et al. Beneficial hemodynamic effects of prone
positioning in patients with acute respiratory distress syndrome. Am J Respir Crit Care Med.
2013;188(12):1428–33.
15. Albert RK, Keniston A, Baboi L, Ayzac L, et al. Prone position-induced improvement in gas
exchange does not predict improved survival in the acute respiratory distress syndrome. Am J
Respir Crit Care Med. 2014;189:494–6.
16. Franchineau G, Bréchot N, Hekimian G, et al. Prone positioning monitored by electrical
impedance tomography in patients with severe acute respiratory distress syndrome on veno-
venous ECMO. Ann Intensive Care. 2020;10(1):12.
17. Guervilly C, Prud'homme E, Pauly V, Bourenne J. Prone positioning and extracorporeal mem-
brane oxygenation for severe acute respiratory distress syndrome: time for a randomized trial?
Intensive Care Med. 2019;45(7):1040–2.
18. Giani M, Martucci G, Madotto F, Belliato M, et al. Prone positioning during venovenous extra-
corporeal membrane oxygenation in acute respiratory distress syndrome: a multicenter cohort
study and propensity-matched analysis. Ann Am Thorac Soc. 2021;18:495–50.
12 Daily Management of Patients on VV ECMO 227
19. Schweickert WD, Pohlman MC, Pohlman AS, Nigos C, et al. Early physical and occupational
therapy in mechanically ventilated, critically ill patients: a randomised controlled trial. Lancet.
2009;373(9678):1874–82.
20. Parker A, Sricharoenchai T, Needham DM. Early rehabilitation in the intensive care
unit: preventing physical and mental health impairments. Curr Phys Med Rehabil Rep.
2013;1(4):307–14.
21. Puthucheary ZA, Rawal J, McPhail M, Connolly B, et al. Acute skeletal muscle wasting in
critical illness. HEJAMA. 2013;310(15):1591–600.
22. Haji JY, Mehra S, Doraiswamy P. Awake ECMO and mobilizing patients on ECMO. Indian J
Thorac Cardiovasc Surg. 2021;37:309–18.
23. Ellouze O, Lamirel J, Perrot J, Missaoui A, et al. Extubation of patients undergoing extracor-
poreal life support. A retrospective study. Perfusion. 2019;34(1):50–7.
24. Justyna S, Shekar K, Protti A, Tukacs M, et al. Extubate before venovenous extracorporeal
membranous oxygenation decannulation or decannulate while remaining on the ventilator?
The EuroELSO 2019 weaning survey. ASAIO J. 2021;67(4):e86–9.
25. Srivastava MC, Ramani GV, Garcia JP, Griffith BP, et al. Veno-venous extracorporeal mem-
brane oxygenation bridging to pharmacotherapy in pulmonary arterial hypertensive crisis. J
Heart Lung Transplant. 2010;29(7):811–3.
26. Grant C, Richards JB, Frakes M, Cohen J, et al. ECMO and right ventricular failure: review of
the literature. J Intensive Care Med. 2021;36(3):352–60.
27. Patel K, Baran D, Nazir H, Anandarangam T. Effect of V-V ECMO flow rates on IVC diameter
variation on trans-thoracic echocardiography. Crit Care Med. 2014;42(12):1406–7.
28. Guinot PG, Zogheib E, Detave M, Moubarak M, et al. Passive leg raising can predict fluid
responsiveness in patients placed on venovenous extracorporeal membrane oxygenation. Crit
Care. 2011;15(5):R216.
29. Luyt CE, Bréchot N, Demondion P, Jovanovic T, et al. Brain injury during venovenous extra-
corporeal membrane oxygenation. Intensive Care Med. 2016;42(5):897–907.
30. 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;201(12):1525–35.
Chapter 13
Crises During ECLS
Introduction
The use and success of extracorporeal life support (ECLS) has historically been
limited by frequent and potentially severe complications. Advancements in circuit
technology and increased experience with ECLS management have reduced, but
certainly not eliminated, these complications in the modern era of ECLS. Accurate
estimates of current complication rates are difficult to enumerate as published rates,
and those found in registries, typically reflect the use of both older and newer tech-
nology and reporting of complications may be underestimated. The Extracorporeal
Life Support Organization (ELSO) registry, which compiles the most extensive
database on complications related to ECLS, includes ECLS cases with both modern
and outdated technology, as well as centers with greater and lesser degrees of expe-
rience with ECLS using a variety of ECLS configurations. Determining the expected
rate of complications in any given circumstance should be attempted with caution,
although the registry allows for some degree of risk adjustment.
Crises in ECLS can be partitioned into those primarily originating in the circuit
and those originating in the patient. Patient outcomes are inextricably linked to the
timely recognition and management of both. Recurring provider training is essential
to maintain these skills and facilitate appropriate emergency responses.
Circuit Crises
During any ECLS crisis requiring circuit clamping, all limbs of the circuit should be
clamped in the most proximal position to the patient, in order to isolate the patient
from the circuit (Fig. 13.1). Clamps should be placed directly on the cannula and are
best positioned between the stainless steel wires and the hard connector on the mal-
leable tubing. Depending on the configuration of the ECLS circuit, there should
always be at least one tubing clamp per cannula present at the bedside and easily
accessible during an emergency. In cases of massive hemorrhage, the tubing line
associated with the source of the bleed should be clamped first. If it is unclear where
the source of bleeding is occurring, then the reinfusion cannulae should be clamped
first, followed by the drainage cannulae, and lastly the distal perfusion cannula, if
present. Clamping of the ECLS circuit requires simultaneous placement of the
patient on emergency ventilator settings or maximal supplemental oxygen or nonin-
vasive ventilatory support in non-intubated patients. Vasopressors and inotropes
should be readily available for hemodynamic support. Notify the ECMO specialist
and/or perfusionist immediately to help troubleshoot and assist with management.
Circuit Air
Any loss of circuit integrity may result in the entrainment of air into the circuit.
Circuit air is an emergency situation that must be addressed immediately; however,
since the oxygenator may serve as an air trap, risk to the patient is highest when air
is present on the post-oxygenator side of the circuit. In venovenous ECLS, circuit
air risks propagation of a venous air embolism. In venoarterial ECLS, or venove-
nous ECLS in the presence of an intracardiac or intrapulmonary shunt, there is a risk
of arterial air embolism.
Pre-oxygenator air can originate from peripheral or central venous lines that are
damaged, inadvertently left open, or accessed for intravenous infusions. Because
circuit pressure is negative between the drainage cannula and the pump, cracked
stopcocks, connectors, tubing, or loose connections can entrain air, as can partial
dislodgement of the drainage cannula, which results in exposure of a side port. Air
can arise within the oxygenator itself from improper priming, oxygenator mem-
brane rupture, or occlusion of the gas exhaust outlet. Post-oxygenator air is seen if
air passes through the oxygenator from the pre-oxygenator side or is directly
entrained through defects in the circuit after the oxygenator.
Another mechanism through which air may enter the circuit is known as cavita-
tion, which occurs when a high negative pressure within the drainage cannula or
tubing draws a large amount of gas out of solution. Extreme changes in pressure
induce hydrodynamic phenomena, including bubble nucleation, growth, and col-
lapse. This process can be minimized by avoiding extremes in negative pressure on
the venous side of the circuit. Cavitation can also occur in the setting of a high
sweep gas flow rate and a low blood flow rate.
In order to minimize introduction of air into the circuit, medication and blood
products should be administered to the patient through separate venous access, and
independent vascular access should be placed for continuous renal replacement
therapy, if possible. If the circuit must be used, the pre-oxygenator port/limb should
be accessed so that the oxygenator can trap any air inadvertently introduced (see
also Chap. 2). A bubble detector may also be placed on the post-oxygenator tubing,
which can be set to stop the pump immediately if air is detected.
Small amounts of air on the venous side of the circuit can often be removed
without interrupting blood flow by carefully withdrawing the air into a syringe. If a
large quantity of air is present or the patient is at particularly high risk of systemic
air embolization, ECLS should be suspended temporarily by clamping all of the
232 P. Madahar et al.
drainage and reinfusion limbs at the most proximal position to the patient, in order
to isolate the circuit from the patient. When air appears on the post-oxygenator side
of the circuit, the drainage and reinfusion lines should be clamped immediately,
thereby halting blood flow (Fig. 13.1). The patient should be placed in the
Trendelenburg position so that any embolization of air is directed toward the lower
extremities and emergency circuit clamping measures should be followed [1, 2].
To remove air from the circuit tubing, attach a syringe to an access port and posi-
tion the tubing that contains the air below the circuit access point. If the pump head
is affected, remove the pump head from its casing, hold it below the circuit access
point, and tap it until the air flows into the circuit tubing. The air will rise to the most
superior position, which should be toward the access point. Aspirate with the syringe
until all the air is removed. With a clamped circuit, removing volume creates even
more negative circuit pressure if fluid is not administered through a second access
point. When large volumes are withdrawn, excessive negative pressure risks hemo-
lysis and further entrainment of air through cavitation. If air is present within the
oxygenator itself, the oxygenator is removed from its holder and tapped gently until
air rises toward a port on its upper portion where it can be aspirated. If a significant
amount of air is present within the oxygenator, the circuit may have to be reprimed.
A cardiotomy reservoir spliced into the circuit can allow the circuit to be reprimed
more efficiently. In the event the circuit cannot be de-aired quickly and easily, the
oxygenator, or possibly the entire circuit, may need to be replaced.
If air reaches the patient, management is mostly supportive, although venous air
may be aspirated under some circumstances [1]. (Table 13.1 highlights the diagno-
sis and management of circuit-related crises.)
Thrombosis
The interaction between blood and the ECLS circuit may affect both procoagulant
and anticoagulant pathways, potentially resulting in thrombosis occurring in the
circuit or the patient. Thrombosis in the ECLS circuit appears as dark red clot or
white strands of fibrin (Fig. 13.2). Inadequate anticoagulation, disseminated intra-
vascular coagulation, heparin-induced thrombocytopenia with thrombosis (HITT),
antithrombin III deficiency, and other hypercoagulable states may further increase
the risk of circuit thrombosis, especially at sites of low or turbulent flow. A simpli-
fied circuit with only the essential components may reduce this risk.
Whereas older ECLS circuits required high levels of anticoagulation to prevent
thrombosis, newer biocompatible or heparin-coated circuit components can be
managed with lower levels of anticoagulation [3, 4]. Although there are no univer-
sally accepted anticoagulation guidelines in ECLS, the administration of heparin
with a goal activated partial thromboplastin time 1.5 times the normal reference
range or an anti-Xa level of 0.1–0.3 IU/mL has been suggested as sufficient to main-
tain circuit patency in many patients [5–8] (see also Chap. 8).
13 Crises During ECLS 233
Oxygenator Failure
Modern polymethylpentene hollow fiber oxygenators are more efficient and durable
than older polypropylene or silicone oxygenators, though over time their effective-
ness may wane and oxygen transfer across the membrane should be assessed if there
is concern for an impairment in oxygenator function. A decreasing post-oxygenator
partial pressure of oxygen (PO2), an increasing transmembrane pressure gradient,
blood flow obstruction, development of circuit-related coagulopathy or hemolysis,
or a need to steadily increase sweep gas flow rate to manage the partial pressure of
arterial carbon dioxide (PaCO2) may be indications of a failing oxygenator [17] (see
also Chap. 9).
Thrombosis within the oxygenator is a common cause of an increase in the trans-
membrane pressure gradient and impaired oxygenator function. This phenomenon
can be partially assessed by visual inspection of the oxygenator during a circuit
check, though visualization of clot is limited to only the most superficial layers of
13 Crises During ECLS 237
the oxygenator membrane and may not be possible for the inlet side of the oxygen-
ator in some pump–oxygenator combinations. Therefore, monitoring the pressure
drop across the oxygenator (delta pressure) is essential. If the pressure gradient
increases for a given blood flow over time, a significant amount of thrombus may be
developing within the oxygenator. Increasing the level of anticoagulation may sta-
bilize these occurrences, though the oxygenator may ultimately need replacement.
Membrane leakage, with blood passing across the membrane and into the hollow
fibers, as seen in the gas exhaust line, is another cause of oxygenator failure. If
blood is found in the gas exhaust line, the oxygenator must be replaced. The gas
exhaust outlet must remain vented to the atmosphere; otherwise, significant increase
in oxygenator gas pressure could place the patient at risk for air embolism across the
compromised membrane. Oxygenator replacement requires that the patient tempo-
rarily be removed from ECLS. Experienced teams can change an oxygenator in less
than 30 seconds. However, this takes careful coordination and planning. Simulation
training is recommended for such scenarios. Emergency circuit clamping measures
should be followed and the pump turned off or to a very low flow rate. Vasopressors
and inotropic agents should be immediately available to manage any hemodynamic
compromise. The circuit is clamped near the inflow and outflow ports of the oxy-
genator, which is then removed and replaced; in the case of integrated circuits, the
full disposable pump and oxygenator will need to be exchanged. Gas and heater
lines are connected to the new oxygenator and ECLS support is resumed.
Pump Failure
Pump failure is a known but uncommon complication of ELCS [18]. If pump fail-
ure occurs for any reason, emergency circuit clamping measures should be fol-
lowed immediately. During VA ECLS, it is imperative to clamp the circuit to
prevent reversal of blood flow from the patient’s arterial system, under the pressure
of native antegrade blood flow through the ECLS circuit into the venous system
(leading to creation of a large arteriovenous shunt) that can lead to cardiopulmo-
nary collapse. A hand crank, which should be kept alongside ECLS circuits, is used
to generate blood flow while the source of the problem is identified and corrected.
If disruption of the electrical power supply is the cause of pump failure, then bat-
tery power or a backup power supply should be obtained. If the ECLS circuit was
inadvertently powered off, it should be restarted immediately and alarm functional-
ity assessed.
If the circuit has adequate power supply and the pump is rotating, but there is no
flow, there may be inadequate occlusion in the case of a roller pump or inadequate
preload or excess afterload with a centrifugal pump in the circuit. Centrifugal pumps
may decouple from the motor, in which case the pump should be stopped and
removed from its seat on the motor. Pump integrity should be confirmed by thor-
ough visual inspection, assessing for cracks or thromboses. If the pump head is
238 P. Madahar et al.
intact, it may be placed back onto the motor and the pump restarted. In the event
there is still no flow, the motor should be replaced.
Heat Exchanger
ECLS circuits contain a heat exchanger connected to a water source that enables
temperature regulation of the patient. Before the patient is placed on ECLS, the heat
exchanger should be tested by connecting to the water source to ensure that there is
no water leakage between the heat exchanger and the membrane fiber bundle within
the oxygenator. If a water leak develops while the patient is on ECLS, the oxygen-
ator must be changed immediately, as hemolysis, blood volume changes, or infec-
tion may ensue from water transfer to the patient or blood loss into the water bath.
Inadvertent Decannulation
Patient Crises
Bleeding is the most common reported complication of ECLS support and has been
described in nearly 25% of cases [20, 21]. The etiology and severity of ECLS-
related bleeding varies from minor oozing at surgical sites to fatal intracerebral
hemorrhage; however, the risk of severe hemorrhage appears to have decreased in
recent years. Bleeding and the associated high transfusion requirements necessary
to maintain an acceptable level of hemoglobin were nearly ubiquitous in early
ECLS experience, with reports of intracerebral hemorrhage in greater than 50% of
patients and transfusion requirements of greater than 6 units of packed red blood
cells (pRBCs) per day in select series [22, 23].
Modern ECLS circuitry and evolving approaches to patient management have
reduced both the risk of hemorrhage and the need for transfusion of pRBCs in
patients receiving extracorporeal support. Two modern randomized controlled trials
of ECLS in severe acute respiratory distress syndrome (ARDS), known as the
Efficacy and economic assessment of conventional ventilator support versus extra-
corporeal membrane oxygenation for severe adult respiratory failure (CESAR) trial
and the ECMO to Rescue Lung Injury in Severe ARDS (EOLIA) trial, reported on
serious bleeding events. In the CESAR trial, only one death related to hemorrhage
was reported. In the EOLIA trial, three subjects (2% of those in the intervention
arm) were reported to have massive hemorrhage [6, 24]. However, in the EOLIA
trial, there was a significant difference in bleeding events leading to transfusion in
those who received ECLS (48%) compared to those who did not receive ECLS
(28%) [6].
Bleeding during ECLS may be difficult to manage due to both the potential coag-
ulopathy induced by the ECLS circuit and the need to provide systemic anticoagula-
tion to prevent thrombus formation within the circuit [24]. Modern circuits are more
240 P. Madahar et al.
biocompatible and require a lesser degree of anticoagulation than older ECLS tech-
nology, which may also reduce the risk and severity of bleeding.
Contemporary trends in critical care have reduced the hemoglobin threshold at
which transfusions are thought to be necessary, including in patients supported with
ECLS [6, 25–30]. Therefore, while transfusions remain common during ECLS sup-
port, and severe or hemodynamically unstable bleeding requires transfusion of
pRBCs, transfusion for the sake of maintaining a normal hemoglobin level may not
be necessary [31, 32].
The optimal hemoglobin for patients receiving ECLS is not known and practice
varies widely by center. Abundant evidence has demonstrated that liberal transfu-
sion strategies targeting higher hemoglobin levels result in increased morbidity
and mortality in critically ill medical and surgical patients when compared with
more conservative approaches [27–29, 33–35]. This includes higher rates of
transfusion-related acute lung injury, worsened ARDS, volume overload, infec-
tions, sepsis, poor wound healing, and intensive care unit, hospital, and 1-year
mortality [33, 34, 36–38]. Transfusion-associated morbidity extends to patients
receiving even small amounts of blood [33, 39]. Adverse events associated with
transfusions increase in a dose-dependent manner and provide evidence suggest-
ing that even as little as 1–2 units of blood may increase morbidity and mortality
considerably [40].
Currently, there is no randomized trial comparing liberal and conservative trans-
fusion protocols in patients receiving ECLS. Nevertheless, several studies have
shown similar, if not improved, outcomes in patients receiving ECLS who were
managed with a conservative transfusion approach compared to utilizing a liberal
transfusion threshold [26, 34, 41]. Given the potentially harmful consequences
associated with the transfusion of pRBCs and the questionable ability of transfused
blood to actually improve oxygen delivery to tissues, it may be beneficial to lower
the transfusion threshold in patients receiving ECLS and tolerate a moderate degree
of anemia [42–44]. Similar to our practice with other critically ill patients, our pro-
tocol is to transfuse non-bleeding patients without active cardiac disease when the
hemoglobin is below 7.0 g/dL, although this threshold may be even lower in patients
awaiting transplantation, where blood product transfusions may affect donor–recip-
ient cross-matching.
Thrombocytopenia is also a common complication of ECLS, with platelet activa-
tion and consumption potentially resulting in a notable amount of thrombocytope-
nia over the first 4–5 days of ECLS support [45–48]. As with hemoglobin, the
optimal platelet level in patients receiving ECLS is not known. One approach is to
maintain a count greater than 20,000 platelets per milliliter in non-bleeding patients
and greater than 50,000 platelets per milliliter in bleeding patients or those undergo-
ing an invasive procedure. One exception to this approach may be in patients with
severe pulmonary hypertension, in whom the risk of platelet sequestration in the
pulmonary vasculature, which can precipitate pulmonary hypertensive crises, may
outweigh the benefit of transfusion in the absence of severe bleeding.
It may be possible to continue anticoagulation in the setting of mild to moderate
bleeding. With modern extracorporeal circuit components that are more
13 Crises During ECLS 241
Hemolysis
Refractory Hypoxemia
which reinfused, oxygenated blood is returned to the circuit without having passed
through the patient, is common in the traditional two-site venovenous ECLS con-
figuration, especially when the drainage and reinfusion ports are in close proximity
to one another or the pump flows are too high and the venous cannula drains a dis-
proportionate amount of reinfused blood [82, 83]. Recirculation should be sus-
pected when a high pre-oxygenator SO2 is observed. However, clinical judgment
must be used to distinguish whether the high pre-oxygenator SO2 is due to recircula-
tion or whether it is a consequence of either pulmonary recovery, poor tissue oxygen
extraction in the setting of sepsis, or decreased cardiac output without correlating
change in blood flow rate in venovenous ECLS or venoarterial ECLS with central
configurations (right atrium to aorta or right atrium to left atrium). Repositioning of
the cannulae or use of a bicaval dual-lumen cannula may reduce recirculation [84–
86]. Oxygenator function may be assessed by monitoring the pre- and post-
oxygenator blood gases, the pressure gradient across the oxygenator, the color
differential of the circulating blood in the drainage and re-infusion limbs by visual
inspection, and the SO2 of the reinfusion limb. Provided that the blood flow is less
than the rated flow of the oxygenator, the post-oxygenator oxygen saturation should
be greater than 99% in a functioning oxygenator.
Shock
Limb Ischemia
of the total arterial reinfusion flow. If unable to monitor flow, pedal pulses should be
assessed frequently, along with visual inspection of the DPC tubing to look for
separation of blood as a sign of reduced flow. Thrombosis of the cannula results in
leg ischemia, and therefore, continuous NIRS oxygen saturation monitoring, as well
as routine Doppler signal monitoring, is important. Management of thrombosed
DPC necessitates immediate removal of the thrombus via sterile aspiration or
removal of the entire DPC. If the clot propagates and is not adequately removed
with catheter removal, therapeutic anticoagulation may be needed. Thrombosis of
the SFA necessitates reconfiguration if a DPC cannot be replaced.
Conclusion
Conflict of Interests Drs. Abrams and Agerstrand are members of the steering committee of the
Extracorporeal Life Support Organization (ELSO).
Dr. Brodie receives research support from ALung Technologies. He has been on the medical
advisory boards for Abiomed, Xenios, Medtronic, Cellenkos, and Hemovent. He is the president-
elect of the Extracorporeal Life Support Organization (ELSO).
All other authors have no conflicts of interest to disclose.
References
1. Chan KM, Wan WTP, Ling L, So JMC, Wong CHL, Tam SBS. Management of circuit air in
extracorporeal membrane oxygenation: a single center experience. ASAIO J. 2021. https://doi.
org/10.1097/mat.0000000000001494.
2. Yan S, Lou S, Zhao Y, Liu G, Ji B. Air in extracorporeal membrane oxygenation: can never be
overemphasized. Perfusion. 2021;36:97–9.
3. Kopp R, Mottaghy K, Kirschfink M. Mechanism of complement activation during extracor-
poreal blood-biomaterial interaction: effects of heparin coated and uncoated surfaces. ASAIO
J. 2002;48:598–605.
4. Deptula J, Glogowski K, Merrigan K, Hanson K, Felix D, Hammel J, Duncan K. Evaluation
of biocompatible cardiopulmonary bypass circuit use during pediatric open heart surgery. J
Extra-corporeal Technol. 2006;38:22–6.
5. Bembea MM, Annich G, Rycus P, Oldenburg G, Berkowitz I, Pronovost P. Variability in anti-
coagulation management of patients on extracorporeal membrane oxygenation. Pediatr Crit
Care Me. 2013;14:e77–84.
6. Combes A, Hajage D, Capellier G, et al. Extracorporeal membrane oxygenation for severe
acute respiratory distress syndrome. N Engl J Med. 2018;378:1965–75.
248 P. Madahar et al.
27. Walsh TS, Boyd JA, Watson D, et al. Restrictive versus liberal transfusion strategies for older
mechanically ventilated critically ill patients. Crit Care Med. 2013;41:2354–63.
28. Vincent JL, Baron J-F, Reinhart K, Gattinoni L, Thijs L, Webb A, Meier-Hellmann A, Nollet
G, Peres-Bota D, Investigators for the A. Anemia and blood transfusion in critically ill patients.
JAMA. 2002;288:1499–507.
29. Hébert PC, Fergusson DA. Red blood cell transfusions in critically ill patients.
JAMA. 2002;288:1525–6.
30. Hébert PC, Wells G, Blajchman MA, Marshall J, Martin C, Pagliarello G, Tweeddale M,
Schweitzer I, Yetisir E. A multicenter, randomized, controlled clinical trial of transfusion
requirements in critical care. N Engl J Med. 1999;340:409–17.
31. Martucci G, Grasselli G, Tanaka K, Tuzzolino F, Panarello G, Schmidt M, Bellani G,
Arcadipane A. Hemoglobin trigger and approach to red blood cell transfusions during veno-
venous extracorporeal membrane oxygenation: the international TRAIN-ECMO survey.
Perfusion. 2019;34:39–48.
32. Butch SH, Knafl P, Oberman HA, Bartlett RH. Blood utilization in adult patients undergoing
extracorporeal membrane oxygenated therapy. Transfusion. 1996;36:61–3.
33. Qin CX, Yesantharao LV, Merkel KR, Goswami DK, Garcia AV, Whitman GJR, Frank SM,
Bembea MM. Blood utilization and clinical outcomes in extracorporeal membrane oxygen-
ation patients. Anesth Analg. 2020;131:901–8.
34. Martucci G, Panarello G, Occhipinti G, et al. Anticoagulation and transfusions management in
veno-venous extracorporeal membrane oxygenation for acute respiratory distress syndrome:
assessment of factors associated with transfusion requirements and mortality. J Intensive Care
Med. 2017;34:630–9.
35. Corwin HL, Gettinger A, Pearl RG, Fink MP, Levy MM, Abraham E, MacIntyre NR, Shabot
MM, Duh M-S, Shapiro MJ. The CRIT Study; anemia and blood transfusion in the critically
ill—current clinical practice in the United States. Crit Care Med. 2004;32:39–52.
36. Marik PE, Corwin HL. Acute lung injury following blood transfusion; Expanding the defini-
tion. Crit Care Med. 2008;36:3080–4.
37. Koch CG, Li L, Sessler DI, Figueroa P, Hoeltge GA, Mihaljevic T, Blackstone EH. Duration
of red-cell storage and complications after cardiac surgery. N Engl J Med. 2008;358:1229–39.
38. Shorr AF, Duh M-S, Kelly KM, Kollef MH, Group CS. Red blood cell transfusion and
ventilator-associated pneumonia; a potential link? Crit Care Med. 2004;32:666–74.
39. Ferraris VA, Davenport DL, Saha SP, Bernard A, Austin PC, Zwischenberger JB. Intraoperative
transfusion of small amounts of blood heralds worse postoperative outcome in patients having
noncardiac thoracic operations. Ann Thorac Surg. 2011;91:1674–80.
40. Ferraris VA, Davenport DL, Saha SP, Austin PC, Zwischenberger JB. Surgical outcomes
and transfusion of minimal amounts of blood in the operating room. Arch Surg-Chicago.
2012;147:49–55.
41. Abbasciano RG, Yusuff H, Vlaar A, Lai F, Murphy GJ. Blood transfusion threshold in patients
receiving Extra Corporeal Membrane Oxygenation (ECMO) support for cardiac and respira-
tory failure – a systematic review and meta-analysis. J Cardiothor Vasc An. 2020;35:1192–202.
42. Rossi’s Principles of Transfusion Medicine, 4th ed.
43. Roback JD, Neuman RB, Quyyumi A, Sutliff R. Insufficient nitric oxide bioavailabil-
ity: a hypothesis to explain adverse effects of red blood cell transfusion. Transfusion.
2011;51:859–66.
44. Carson JL, Grossman BJ, Kleinman S, et al. Red blood cell transfusion: a clinical practice
guideline from the AABB*. Ann Intern Med. 2012;157:49.
45. Anderson HL, Cilley RE, Zwischenberger JB, Bartlett RH. Thrombocytopenia in neonates
after extracorporeal membrane oxygenation. ASAIO J. 1986;32:534–7.
46. Robinson TM, Kickler TS, Walker LK, Ness P, Bell W. Effect of extracorporeal membrane
oxygenation on platelets in newborns. Crit Care Med. 1993;21:1029–34.
47. Stallion A, Cofer BR, Rafferty JA, Ziegler MM, Ryckman FC. The significant relationship
between platelet count and haemorrhagic complications on ECMO. Perfusion. 1994;9:265–9.
250 P. Madahar et al.
48. Jiritano F, Serraino GF, Ten Cate H, Fina D, Matteucci M, Mastroroberto P, Lorusso R. Platelets
and extra-corporeal membrane oxygenation in adult patients: a systematic review and meta-
analysis. Intensive Care Med. 2020;46:1154–69.
49. Lamb K, Cowan S, Evans N, Pitcher H, Moritz T, Lazar M, Hirose H, Cavarocchi N. Successful
management of bleeding complications in patients supported with extracorporeal membrane
oxygenation with primary respiratory failure. Perfusion. 2013;28:125–31.
50. von Stumm M, Subbotina I, Biermann D, Gottschalk U, Mueller G, Kozlik-Feldmann R,
Reichenspurner H, Riso A, Sachweh JS. Impact of delayed systemic heparinization on post-
operative bleeding and thromboembolism during post-cardiotomy extracorporeal membrane
oxygenation in neonates. Perfusion. 2020;35:626–32.
51. Wilson JM, Bower LK, Fackler JC, Beals DA, Bergus BO, Kevy SV. Aminocaproic acid
decreases the incidence of intracranial hemorrhage and other hemorrhagic complications of
ECMO. J Pediatr Surg. 1993;28:536–41.
52. van der Staak FHJ, de Haan AFJ, Geven WB, Festen C. Surgical repair of congenital diaphrag-
matic hernia during extracorporeal membrane oxygenation: hemorrhagic complications and
the effect of tranexamic acid. J Pediatr Surg. 1997;32:594–9.
53. Downard CD, Betit P, Chang RW, Garza JJ, Arnold JH, Wilson JM. Impact of Amicar on hem-
orrhagic complications of ECMO: a ten-year review. J Pediatr Surg. 2003;38:1212–6.
54. Buckley LF, Reardon DP, Camp PC, Weinhouse GL, Silver DA, Couper GS, Connors
JM. Aminocaproic acid for the management of bleeding in patients on extracorporeal mem-
brane oxygenation: four adult case reports and a review of the literature. Hear Lung J Acute
Critical Care. 2016;45:232–6.
55. Lotz C, Streiber N, Roewer N, Lepper PM, Muellenbach RM, Kredel M. Therapeutic inter-
ventions and risk factors of bleeding during extracorporeal membrane oxygenation. ASAIO
J. 2017;63:624–30.
56. Coleman M, Davis J, Maher KO, Deshpande SR. Clinical and hematological outcomes of ami-
nocaproic acid use during pediatric cardiac ECMO. J Extra-corporeal Technol. 2021;53:40–5.
57. Dominguez TE, Mitchell M, Friess SH, Huh JW, Manno CS, Ravishankar C, Gaynor JW,
Tabbutt S. Use of recombinant factor VIIa for refractory hemorrhage during extracorporeal
membrane oxygenation&ast. Pediatr Crit Care Me. 2005;6:348–51.
58. Wittenstein B, Ng C, Ravn H, Goldman A. Recombinant factor VII for severe bleeding during
extracorporeal membrane oxygenation following open heart surgery. Pediatr Crit Care Me.
2005;6:473–6.
59. Repessé X, Au SM, Bréchot N, Trouillet J-L, Leprince P, Chastre J, Combes A, Luyt
C-E. Recombinant factor VIIa for uncontrollable bleeding in patients with extracorporeal
membrane oxygenation: report on 15 cases and literature review. Crit Care. 2013;17:R55.
60. Anselmi A, Guinet P, Ruggieri VG, et al. Safety of recombinant factor VIIa in patients under
extracorporeal membrane oxygenation. Eur J Cardio Thorac. 2016;49:78–84.
61. Bui JD, Despotis GD, Trulock EP, Patterson GA, Goodnough LT. Fatal thrombosis after
administration of activated prothrombin complex concentrates in a patient supported by extra-
corporeal membrane oxygenation who had received activated recombinant factor VII. J Thorac
Cardiovasc Surg. 2002;124:852–4.
62. Syburra T, Lachat M, Genoni M, Wilhelm MJ. Fatal outcome of recombinant factor VIIa
in heart transplantation with extracorporeal membrane oxygenation. Ann Thorac Surg.
2010;89:1643–5.
63. Morgan IS, Codispoti M, Sanger K, Mankad PS. Superiority of centrifugal pump over roller
pump in paediatric cardiac surgery: prospective randomised trial. Eur J Cardio Thorac.
1998;13:526–32.
64. Klein M, Dauben H, Schulte HD, Gam E. Centrifugal pumping during routine open heart
surgery improves clinical outcome. Artif Organs. 1998;22:326–36.
65. Parolari A, Alamanni F, Naliato M, Spirito R, Franzè V, Pompilio G, Agrifoglio M, Biglioli
P. Adult cardiac surgery outcomes: role of the pump type. Eur J Cardio Thorac. 2000;18:575–82.
13 Crises During ECLS 251
66. Khoshbin E, Roberts N, Harvey C, Machin D, Killer H, Peek GJ, Sosnowski AW, Firmin
RK. Poly-methyl pentene oxygenators have improved gas exchange capability and
reduced transfusion requirements in adult extracorporeal membrane oxygenation. ASAIO
J. 2005;51:281–7.
67. Dogal N, Mathis R, Lin J, Qiu F, Kunselman A, Ündar A. Evaluation of three hollow-fiber
membrane oxygenators without integrated arterial filters for neonatal cardiopulmonary bypass.
Perfusion. 2012;27:132–40.
68. Gu YJ, Boonstra PW, Graaff R, Rijnsburger AA, Mungroop H, van Oeveren W. Pressure
drop, shear stress, and activation of leukocytes during cardiopulmonary bypass: a comparison
between hollow fiber and flat sheet membrane oxygenators. Artif Organs. 2000;24:43–8.
69. Lehle K, Philipp A, Zeman F, Lunz D, Lubnow M, Wendel H-P, Göbölös L, Schmid C, Müller
T. Technical-induced hemolysis in patients with respiratory failure supported with veno-
venous ECMO – prevalence and risk factors. Plos One. 2015;10:e0143527.
70. Dufour N, Radjou A, Thuong M. Hemolysis and plasma free hemoglobin during extracorpo-
real membrane oxygenation support. Asaio J. 2019. Publish Ahead of Print:NA.
71. Appelt H, Philipp A, Mueller T, Foltan M, Lubnow M, Lunz D, Zeman F, Lehle K. Factors asso-
ciated with hemolysis during extracorporeal membrane oxygenation (ECMO)—Comparison
of VA- versus VV ECMO. Plos One. 2020;15:e0227793.
72. Nagler B, Hermann A, Robak O, et al. Incidence and etiology of system exchanges in
patients receiving extracorporeal membrane oxygenation. Asaio J. 2021. Publish Ahead of
Print:776–84.
73. Goligher EC, Tomlinson G, Hajage D, Wijeysundera DN, Fan E, Jüni P, Brodie D, Slutsky
AS, Combes A. Extracorporeal membrane oxygenation for severe acute respiratory distress
syndrome and posterior probability of mortality benefit in a post hoc bayesian analysis of a
randomized clinical trial. JAMA. 2018;320:2251.
74. Zakhary B, Vercaemst L, Mason P, Lorusso R, Brodie D. How I manage drainage insufficiency
on extracorporeal membrane oxygenation. Crit Care. 2020;24:151.
75. Broman LM, Wittberg LP, Westlund CJ, et al. Pressure and flow properties of cannulae for extra-
corporeal membrane oxygenation II: drainage (venous) cannulae. Perfusion. 2019;34:65–73.
76. On bypass, advanced perfusion techniques. 2008. https://doi.org/10.1007/978-1-59745-305-9.
77. Combes A, Price S, Slutsky AS, Brodie D. Temporary circulatory support for cardiogenic
shock. Lancet. 2020;396:199–212.
78. Javidfar J, Brodie D, Costa J, Miller J, Jurrado J, LaVelle M, Newmark A, Takayama H, Sonett
JR, Bacchetta M. Subclavian artery cannulation for venoarterial extracorporeal membrane
oxygenation. ASAIO J. 2012;58:494–8.
79. Biscotti M, Bacchetta M. The “Sport Model”: extracorporeal membrane oxygenation using the
subclavian artery. Ann Thorac Surg. 2014;98:1487–9.
80. Chicotka S, Rosenzweig EB, Brodie D, Bacchetta M. The “Central Sport Model”. ASAIO
J. 2017;63:e39–44.
81. Ranney DN, Benrashid E, Meza JM, et al. Central cannulation as a viable alternative to periph-
eral cannulation in extracorporeal membrane oxygenation. Semin Thorac Cardiovasc Surg.
2017;29:188–95.
82. Abrams D, Bacchetta M, Brodie D. Recirculation in venovenous extracorporeal membrane
oxygenation. ASAIO J. 2015;61:115–21.
83. Palmér O, Palmér K, Hultman J, Broman M. Cannula design and recirculation during venove-
nous extracorporeal membrane oxygenation. ASAIO J. 2016;62:737–42.
84. Frenckner B, Broman M, Broomé M. Position of draining venous cannula in extracorporeal
membrane oxygenation for respiratory and respiratory/circulatory support in adult patients.
Crit Care. 2018;22:163.
85. Javidfar J, Brodie D, Wang D, Ibrahimiye AN, Yang J, Zwischenberger JB, Sonett J, Bacchetta
M. Use of bicaval dual-lumen catheter for adult venovenous extracorporeal membrane oxy-
genation. Ann Thorac Surg. 2011;91:1763–9.
252 P. Madahar et al.
86. Tipograf Y, Gannon WD, Foley NM, Hozain A, Ukita R, Warhoover M, McMaster W, Nesbitt
JC, Shah AS, Bacchetta M. A dual-lumen bicaval cannula for venovenous extracorporeal
membrane oxygenation. Ann Thorac Surg. 2020;109:1047–53.
87. Brasseur A, Scolletta S, Lorusso R, Taccone FS. Hybrid extracorporeal membrane oxygen-
ation. J Thorac Dis. 2018;10:S707–15.
88. Coco VL, Swol J, Piero MED, Massimi G, Chiarini G, Broman LM, Lorusso R. Dynamic
extracorporeal life support: a novel management modality in temporary cardio-circulatory
assistance. Artif Organs. 2021;45:427–34.
89. Bréchot N, Hajage D, Kimmoun A, et al. Venoarterial extracorporeal membrane oxygenation
to rescue sepsis-induced cardiogenic shock: a retrospective, multicentre, international cohort
study. Lancet. 2020;396:545–52.
90. Ling RR, Ramanathan K, Poon WH, Tan CS, Brechot N, Brodie D, Combes A, MacLaren
G. Venoarterial extracorporeal membrane oxygenation as mechanical circulatory support in
adult septic shock: a systematic review and meta-analysis with individual participant data
meta-regression analysis. Crit Care. 2021;25:246.
91. Myers LC, Lee C, Thompson BT, Cudemus G, Raz Y, Roy N. Outcomes of adult patients with
septic shock undergoing extracorporeal membrane oxygenation therapy. Ann Thorac Surg.
2020;110:871–7.
92. Huang C-T, Tsai Y-J, Tsai P-R, Ko W-J. Extracorporeal membrane oxygenation resuscitation
in adult patients with refractory septic shock. J Thorac Cardiovasc Surg. 2013;146:1041–6.
93. MacLaren G, Pellegrino V, Butt W, Preovolos A, Salamonsen R. Successful use of ECMO in
adults with life-threatening infections. Anaesth Intensive Care. 2004;32:707–10.
94. Elmously A, Bobka T, Khin S, et al. Distal perfusion cannulation and limb complica-
tions in venoarterial extracorporeal membrane oxygenation. J Extra-corporeal Technol.
2018;50:155–60.
95. Juo Y, Skancke M, Sanaiha Y, Mantha A, Jimenez JC, Benharash P. Efficacy of distal perfu-
sion cannulae in preventing limb ischemia during extracorporeal membrane oxygenation: a
systematic review and meta-analysis. Artif Organs. 2017;41:E263–73.
96. Mohite PN, Fatullayev J, Maunz O, et al. Distal limb perfusion in peripheral VA-ECMO. Artif
Organs. 2014;38:940–4.
Chapter 14
Mobilization During ECLS
Gregory A. Schmidt
Until recently, patients managed on ECMO were deeply sedated, often paralyzed,
always mechanically ventilated, and moved as gingerly as possible. Now, techno-
logical innovations in gas-exchanging membranes, pumps, and cannulas, combined
with the recognition that life-sustaining gas exchange can be achieved through the
veno-venous (VV) mode, have spurred enthusiasm to mobilize ECMO patients, as
we do other critically ill patients [1, 2]. These changes have also forced us to recon-
sider the role of ECMO in severe ARDS. Is it merely to rescue the desperately ill
once mechanical ventilation, paralysis, prone positioning, and inhaled vasodilators
have failed? Or, instead, can it serve as a means to avoid noxious interventions and
their sequelae, earning an earlier and more prominent place in the intensivist’s
armamentarium for the treatment of the acute respiratory distress syndrome (ARDS)?
Technological Advances
Patients with severe ARDS can generally be treated with the VV, rather than the
veno-arterial (VA), mode. For example, in the Conventional Ventilation or ECMO
for Severe Acute Respiratory Failure (CESAR) trial, all patients were treated with
the VV mode and none required conversion to VA ECMO [3]. VV ECMO offers
three major advantages. First, returning blood from the gas exchange membrane to
the veins, rather than systemic arteries, requires much lower pressures and corre-
spondingly simpler, safer circuits (see Chap. 2). Second, since only venous access
is required, mobilization-related risks of large vessel arterial cannulation
G. A. Schmidt (*)
Division of Pulmonary Diseases, Critical Care, and Occupational Medicine, Department of
Internal Medicine, University of Iowa, Iowa City, IA, USA
e-mail: [email protected]
(hemorrhage, limb ischemia, dislodgement) are avoided. Finally, by its very nature,
VA ECMO requires at least two cannulas, adding to the complexity of movement,
whereas VV support typically needs only a single cannula (see Chap. 4).
ECMO devices, too, have evolved in a direction to make mobilization more
feasible. Modern polymethylpentene (PMP) gas-exchanging membranes present
a much lower resistance to blood flow than earlier technologies (see Chap. 2).
This has two key advantages: (a) higher circuit blood flow is possible, capturing a
greater fraction of the native cardiac output, making it easier to attain acceptable
oxygenation, as described in Chap. 1 and (b) the lower resistance requires less
pressure to drive blood flow, producing safer circuits and smaller, more reliable
pumps. Safer circuits and pumps demand less in the way of monitors and other
accoutrements, reducing the clutter that formerly surrounded ECMO patients. It is
also hard to overstate the importance of dual-lumen cannulas. Only one access
site is needed, usually freeing the groin and making it easier to have patients sit,
stand, and even walk. Even as recently as the CESAR trial, all patients had two
(some three) cannulas, including one in the femoral vein. The presence of only
one critical cannula also reduces the burden on the mobilization team. As described
below, when we mobilize ECMO patients, we dedicate one member of the team to
focus entirely on the cannula(s), since the patient’s life depends on its reliable
function.
Like skeletal muscles, the diaphragm suffers atrophy and contractile dysfunction
during critical illness and mechanical ventilation. This occurs acutely, worsens pro-
gressively, and is associated with prolonged ventilation and risk of death. Muscle
protein synthesis is inhibited and multiple pathways of self-destruction are upregu-
lated. Also like in peripheral muscle, active contraction (that is, active breathing)
can effectively modify the degree of catabolism, helping to maintain contractile
function [12]. This has potentially important implications for the role of ECMO. By
freeing the diseased lungs from the burden of gas exchange, extracorporeal support
could allow less ventilation (even extubation; see Chap. 12) and, therefore, less
sedation, delirium, immobility, and VIDD.
Probably the greatest causal factor for ICU-AW is muscular silence, a hypothesis
that may explain the contributory role of sedatives, neuromuscular blocking drugs,
controlled mechanical ventilation, and bed rest. Reducing these influences requires
changing our view of the critically ill from fragile beings in need of (over)protection
to subjects at risk of complication if not liberated from noxious treatments. Of
course, some patients are incredibly delicate and unable to sustain any provocation,
but their number is probably fewer than we commonly believe. As discussed below,
there is accumulating evidence that patients can be safely liberated while on ECMO.
This early experience has led many centers, not only to consider and trial ambu-
latory ECMO, but to actively develop and nurture specialized teams to aid mobiliza-
tion. Following the creation of a mobilization program aimed at ECMO patients, 35
of 100 patients with refractory respiratory or cardiac failure could be actively reha-
bilitated [29]. Since bridge-to-recovery patients tend to be more ill and suffering
from more organ failures than bridge-to-transplant patients, it is interesting that 16
of the 35 patients undergoing physical therapy were not transplant candidates.
Further, while most were treated with VV ECMO through a dual-lumen cannula in
the internal jugular vein, 4 patients were on VA ECMO (internal jugular vein to
subclavian artery) and 8 had femoral access. Of the 16 bridge-to-recovery patients,
14 survived, most going directly home following discharge. It should be empha-
sized that these results followed an explicit plan to create and train a specialized
team including physical and occupational therapists, perfusionists, critical care
RNs, critical care nurse practitioners, respiratory therapists, and ECMO intensivists
and surgeons.
Increasingly, patients with arterial cannulas for VA ECMO or having femoral
cannulas for VV ECMO are considered excellent candidates for mobilization.
Indeed, in some trials the great majority of patients underwent active mobilization
while cannulated via the femoral route [18, 21]. Rare instances of accidental decan-
nulation [18], cannula site bleeding [18], and iliopsoas hematoma [30] are reported.
Nevertheless, overall, the evidence supports the safety of ambulation with femoral
cannulation, even for VA ECMO [31, 32].
It is no small task to safely mobilize patients during ECMO. Barriers include sever-
ity of illness, resource and staffing limitations, obesity, equipment needs, risks of
catastrophic complications, and cultural barriers [20]. There is broad agreement that
mobilization is safe for critically ill patients, including those with endotracheal
tubes requiring mechanical ventilation [33], but still very limited confidence that we
can safely do this with the very sickest. Before mobilization, patients should be
screened for contraindications, such as those described in the trial by Schweickert
and colleagues as modified for ECMO in Table 14.1 [34]. Many ECMO patients
will have neurological failure as a consequence of the underlying illness or its treat-
ment, precluding extubation, ambulation, or any meaningful therapy. Severe circu-
latory failure may also be limiting, although many patients in shock and on
vasoactive infusion can sit, stand, and even walk [1]. In others, mobilization threat-
ens circuit function because movement changes cannula position, causes catheter
kinking, or provokes transient caval collapse. Morbid obesity, extensive surgical
wounds or dressings, hardware, and other patient-specific features may limit
mobility.
Fear of complications when ambulating is understandable since circuit failure, or
even a simple stumble, can be life-threatening. Accumulating data, however, are
258 G. A. Schmidt
reassuring. For example, in the small series cited above [17, 18], serious complica-
tions were not seen. In the larger series of bridge-to-transplant and bridge-to-
recovery, there were no complications related to physical therapy [24]. The
physiological stress related to mobilization appears to be quite modest [35]. Even in
the broader experience gained in generally critically ill patients, physical therapy
and mobilization have been remarkably complication-free [14, 36]. In a prospec-
tive, randomized trial of early, aggressive physical therapy, only 4% of sessions
were terminated for safety concerns [1]. The most commonly reported were periods
of patient-ventilator dyssynchrony. Falls to knees without injury did not occur in
two studies [1, 14], but occurred rarely in another [13]. Accidental extubations were
not seen in these early studies. In a large, prospective evaluation of safety in a single
center, fewer than 1% of sessions had a potential safety event (most often arrhyth-
mia) and only 4 of 1110 required minimal additional treatment or cost [27].
A final barrier relates to each ICU’s culture of liberation in general and mobiliza-
tion in particular. For example, physical rehabilitation is built on a foundation of
sedative interruption, regular spontaneous breathing trials, delirium prevention, and
a multidisciplinary, team-based approach to patient care. A quality improvement
project aimed at reducing heavy sedation and bolstering staffing to include full-time
physical and occupational therapists succeeded in improving delirium, functional
mobility, and ICU length of stay [37]. Leadership is also essential. When imple-
menting a new mobility culture, champions should anticipate some resistance from
clinicians unconvinced of safety. At times, such resistance arises from physicians,
primary care clinicians, bedside nurses, and even family members. Physical thera-
pists appear ready to mobilize patients to higher levels than nurses, suggesting that
added therapist staffing is an effective means to change practice [38]. Time invested
in identifying supporters is valuable, but skeptics may respond to education or
active listening regarding their concerns.
The risks of early mobilization, although rare, are immediate, obvious, and
potentially catastrophic. In contrast, the benefits are subtle and delayed, leading
14 Mobilization During ECLS 259
Once a patient passes the safety screen described above, the mobility team should
be assembled. Safely ushering an ECMO patient out of bed, and especially out of
the ICU, requires monitoring, attention to the cannula, and a perfusionist to roll and
attend to the circuit, in addition to the physical therapist: three caregivers at a mini-
mum and often several more (Fig. 14.1). It can be expensive to provide such an
experienced team on a regular basis, especially since physical therapy resources are
currently in high demand in the ICU. Specialized equipment can make this easier,
but adds an additional layer of expense. The potential for effective therapy depends
on circuit design that is conducive to transport, as described in Chap. 2. Centers
beginning a program of mobilization are advised to consult with or visit experi-
enced ICUs.
The team should be experienced in the care and mobilization of the critically ill.
The complexity of mobilizing the ECMO patient, combined with the ever-present
risk of circuit disruption, demands a coordinated effort. The therapist should be the
single, clear team leader to state each step of the plan, communicate with the patient,
and judge whether to continue or abort the session, akin to the role of team leader
during cardiopulmonary resuscitation [39]. When the team leader clearly states
what will happen next, the perfusionist can anticipate how to maneuver and monitor
the circuit; the assistant can stabilize the cannula and guide its movement along with
the patient; and the patient can muster his resources to make a good effort. Because
patients are often anxious, a calm but firm approach is essential. Responsibilities of
the nurse include vital sign monitoring, observation for distress, and attention to the
stability of tubes and lines. The perfusionist attends to the cannula and circuit,
including adequacy of circuit flow, and anticipates and communicates to the team
leader any threat to its continued function. Additional staff may be needed to deal
with a mechanical ventilator (transport ventilator), provide hand-bag ventilation,
assist especially debilitated or obese patients, carry backup supplies, push wheel-
chairs and carts, or provide an additional level of monitoring for very fragile patients.
Before embarking on ambulation, necessary devices, such as endotracheal tubes,
should be examined for integrity and securement. Any inessential devices, cathe-
ters, or infusions should be discontinued temporarily. Sweep gas flow should be
disconnected from the wall oxygen, using a transport cylinder. Sweep gas flow rates
higher than maintenance may be needed to adjust for the increased metabolic
demand of exercise, and this should be factored in when calculating whether the
cylinder will support the entire session. Extra oxygen and a spare battery provide an
added level of safety. When leaving the bedside, it is useful to bring along a
wheelchair in case the patient tires or needs additional assistance.
Physical activity in the critically ill raises total body oxygen consumption mean-
ingfully and probably even more than in healthy volunteers performing a similar
degree of work [40]. This added burden could threaten a limited cardiopulmonary
reserve, so that careful monitoring is essential for safety. At a minimum, this
includes continuous assessment of rhythm, heart rate, oximetry, and subjective dis-
tress, as well as the ability to measure blood pressure. Any deterioration, including
those listed in Table 14.2, should prompt an immediate halt to mobilization, reas-
sessment of safety, and a decision whether to terminate the session. Returning the
patient to the ICU bed requires careful planning and may benefit from a checklist so
that essential steps (e.g., reconnecting the sweep gas to wall oxygen) are not omitted.
Cycle ergometry, with passive or increasingly active motion, increases muscle
force and functional status at the time of ICU discharge, providing another option
for strengthening [41]. Electrical stimulation of muscles has been hypothesized to
improve strength and limit wasting, but efficacy remains unproved [42]. As the
14 Mobilization During ECLS 261
Conclusion
Critically ill patients are at very high risk for ICU-AW and VIDD, potentially
increasing morbidity, complications, and length of stay. Physical therapy is clearly
effective in raising functional capacity and appears remarkably safe. Even during
ECMO, mobilization is possible with an experienced, dedicated team, bringing into
focus a new vision: the awake, extubated, ambulating patient.
References
1. Schweickert WD, Pohlman MC, Pohlman AS, et al. Early physical and occupational ther-
apy in mechanically ventilated, critically ill patients: a randomized controlled trial. Lancet.
2009;373:1874–82.
2. Engel HJ, Needham DM, Morris PE, et al. ICU early mobilization: from recommendation to
implementation at three medical centers. Crit Care Med. 2013;41:S69–80.
3. Peek GJ, Mugford M, Tiruvoipati R, et al. Efficacy and economic assessment of conventional
ventilatory support versus extracorporeal membrane oxygenation for severe adult respiratory
failure (CESAR): a multicentre randomised controlled trial. Lancet. 2009;374:1351–63.
4. Kress JP, Pohlman AS, O’Connor MF, et al. Daily interruption of sedative infusions in criti-
cally ill patients undergoing mechanical ventilation. N Engl J Med. 2000;342:1471–7.
5. Riker RR, Shehabi Y, Bokesch PM, et al. Dexmedetomidine vs midazolam for sedation of criti-
cally ill patients: a randomized trial. JAMA. 2009;301:489–99.
6. Kress JP, Gehlbach B, Lacy M, et al. The long-term psychological effects of daily sedative
interruption on critically ill patients. Am J Respir Crit Care Med. 2003;168:1457–61.
262 G. A. Schmidt
7. Shehabi Y, Bellomo R, Reade MC, et al. Early intensive care sedation predicts long-term mor-
tality in ventilated critically ill patients. Am J Respir Crit Care Med. 2012;186:724–31.
8. Girard TD, Kress JP, Fuchs BD, et al. Efficacy and safety of a paired sedation and ventilator
weaning protocol for mechanically ventilated patients in intensive care (awakening and breath-
ing controlled trial): a randomised controlled trial. Lancet. 2008;371:126–34.
9. Fan E, Cheek F, Chlan L, et al. An official American Thoracic Society clinical practice guide-
line: the diagnosis of intensive care unit-acquired weakness in adults. Am J Respir Crit Care
Med. 2014;190:1437–46.
10. Demoule A, Jung B, Prodanovic H, et al. Diaphragm dysfunction on admission to the intensive
care unit. Prevalence, risk factors, and prognostic impact – a prospective study. Am J Respir
Crit Care Med. 2013;188:213–9.
11. Wollersheim T, Woehlecke J, Krebs M, et al. Dynamics of myosin degradation in intensive care
unit-acquired weakness during severe critical illness. Intensive Care Med. 2014;40:528–38.
12. Sassoon CSH, Zhu E, Caiozzo VJ. Assist-control mechanical ventilation attenuates ventilator-
induced diaphragm dysfunction. Am J Respir Crit Care Med. 2004;170:626–32.
13. Bailey P, Thomsen GE, Spuhler VJ, et al. Early activity is feasible and safe in respiratory fail-
ure patients. Crit Care Med. 2007;35:139–45.
14. Morris PE, Goad A, Thompson C, et al. Early intensive care unit mobility therapy in the treat-
ment of acute respiratory failure. Crit Care Med. 2008;36:2238–43.
15. Balas MC, Vasilevskis EE, Olsen KM, et al. Effectiveness and safety of the awakening and
breathing coordination, delirium monitoring/management, and early exercise/mobility bundle.
Crit Care Med. 2014;42:1024–36.
16. Thiagarajan RR, Teele SA, Teele KP, et al. Physical therapy and rehabilitation issues for
patients supported with extracorporeal membrane oxygenation. J Pediatr Rehabil Med.
2012;5:47–52.
17. Bonizzoli M, Lazzeri C, Drago A, Boninsegni LT, Donati M, Di Valvasone S, et al. Effects of a
physiotherapic program in patients on venovenous extracorporeal membrane oxygenation: an
8-year single-center experience. Minerva Anestesiol. 2019;85:989–94.
18. Braune S, Bojes P, Mecklenburg A, Angriman F, Soeffker G, Warnke K, et al. Feasibility,
safety, and resource utilization of active mobilization of patients on extracorporeal life support:
a prospective observational study. Ann Intensive Care. 2020;10:161.
19. The ECMO-PT Study Investigators and International ECMO Network. Early mobilization
during extracorporeal membrane oxygenation was safe and feasible: a pilot randomized con-
trolled trial. Intensive Care Med. 2020;46:1057–9.
20. Marhong JD, DeBacker J, Viau-Lapointe J, Munshi L, Del Sorbo L, Burry L, et al. Sedation
and mobilization during venovenous extracorporeal membrane oxygenation for acute respira-
tory failure: an international survey. Crit Care Med. 2017;45:1893–9.
21. Wells CL, Forrester J, Vogel J, Rector R, Tabatabai A, Herr D. Safety and feasibility of early
physical therapy for patients on extracorporeal membrane oxygenator: University of Maryland
Medical Center experience. Crit Care Med. 2018;46:53–9.
22. Garcia JP, Iacono A, Kon ZN, et al. Ambulatory extracorporeal membrane oxygenation: a new
approach for bridge-to-lung transplantation. J Thorac Cardiovasc Surg. 2010;139:e137–9.
23. Garcia JP, Kon ZN, Evans C, et al. Ambulatory veno-venous extracorporeal membrane oxy-
genation: innovation and pitfalls. J Thorac Cardiovasc Surg. 2011;142:755–61.
24. Turner DA, Cheifetz IM, Rehder KJ, et al. Active rehabilitation and physical therapy dur-
ing extracorporeal membrane oxygenation while awaiting lung transplantation: a practical
approach. Crit Care Med. 2011;39:2593–8.
25. Hayes D Jr, Kukreja J, Tobias JD, et al. Ambulatory venovenous extracorporeal respiratory
support as a bridge for cystic fibrosis patients to emergent lung transplantation. J Cyst Fibros.
2012;11:40–5.
26. Rehder KJ, Turner DA, Hartwig MG, et al. Active rehabilitation during extracorporeal mem-
brane oxygenation as a bridge to lung transplantation. Respir Care. 2013;58:1291–8.
14 Mobilization During ECLS 263
27. Olsson KM, Simon A, Strueber M, et al. Extracorporeal membrane oxygenation in nonintu-
bated patients as bridge to lung transplantation. Am J Transplant. 2010;10:2173–8.
28. Hoopes CW, Kukreja J, Golden J, et al. Extracorporeal membrane oxygenation as a bridge to
pulmonary transplantation. J Thorac Cardiovasc Surg. 2013;145:862–8.
29. Abrams D, Javidfar J, Farrand E, et al. Early mobilization of patients receiving extracorporeal
membrane oxygenation: a retrospective cohort study. Crit Care. 2014;18:R38.
30. Taniguchi H, Ikeda T, Takeuchi I, Ichiba S. Iliopsoas hematoma in patients undergoing veno-
venous ECMO. Am J Crit Care. 2021;30:55–63.
31. Pasrija C, Mackowick KM, Raithel M, Tran D, Boulos FM, Deatrick KB, et al. Ambulation
with femoral arterial cannulation can be safely performed on venoarterial extracorporeal mem-
brane oxygenation. Ann Thorac Surg. 2019;107:1389–94.
32. Da Cunha FD, Marcolino MAZ, Macagnan FE, Della Mea Plentz R, Kessler A. Safety and
potential benefits of physical therapy in adult patients on extracorporeal membrane oxygen-
ation support: a systematic review. Rev Bras Ter Intensiva. 2019;31:227–39.
33. Hodgson CL, Stiller K, Needham DM, et al. Expert consensus and recommendations on
safety criteria for active mobilization of mechanically ventilated critically ill adults. Crit Care.
2014;18:658. https://doi.org/10.1186/s13054-014-0658-y.
34. Pohlman MC, Schweickert WD, Pohlman AS, et al. Feasibility of physical and occupational
therapy beginning from initiation of mechanical ventilation. Crit Care Med. 2010;38:2089–94.
35. Hayes K, Holland AE, Pellegrino VA, Young M, Paul E, Hodgson CL. Early rehabilitation dur-
ing extracorporeal membrane oxygenation has minimal impact on physiological parameters: a
pilot randomized controlled trial. Aust Crit Care. 2021;34:217–25.
36. Sricharoenchai T, Parker AM, Zanni JM, et al. Safety of physical therapy interventions in criti-
cally ill patients: a single-center prospective evaluation of 1110 intensive care unit admissions.
J Crit Care. 2014;29:395–400.
37. Needham DM, Korupolu R, Zanni JM, et al. Early physical medicine and rehabilitation for
patients with acute respiratory failure: a quality improvement project. Arch Phys Med Rehabil.
2010;91:536–42.
38. Garzon-Serrano J, Ryan C, Waak K, et al. Early mobilization in critically ill patients:
patients’ mobilization level depends on health care provider’s profession. Phys Med Rehabil.
2011;3:307–13.
39. Yeung JH, Ong GJ, Davies RP, et al. Factors affecting team leadership skill and their relation-
ship with quality of cardiopulmonary resuscitation. Crit Care Med. 2012;40:2617–21.
40. Hickmann CE, Roeseler J, Castanares-Zapatero D, et al. Energy expenditure in the critically ill
performing early physical therapy. Intensive Care Med. 2014;40:548–55.
41. Burtin C, Clerckx B, Robbeets C, et al. Early exercise in critically ill patients enhances short-
term functional recovery. Crit Care Med. 2009;37:2499–505.
42. Kho ME, Truong AD, Zanni JM, et al. Neuromuscular electrical stimulation in mechanically
ventilated patients: a randomized, sham-controlled pilot trial with blinded outcome assess-
ment. J Crit Care. 2015;30:32–9.
Chapter 15
ECMO Weaning and Decannulation
Introduction
As native lung function improves in patients with acute respiratory distress syn-
drome (ARDS) on extracorporeal life support, therapy is directed toward weaning
from extracorporeal membrane oxygenation (ECMO) and decannulation. The speed
with which patients can be weaned off support depends on the improvement in their
cardiopulmonary function and needs to be titrated to individual response. In this
chapter, we will discuss the signs of readiness; management of the ventilator, oxy-
genator sweep gas, and ECMO flow during the weaning process; and cardiopulmo-
nary assessment during weaning. Further, we will discuss the procedure of
decannulation and the follow-up of patients after decannulation. The process of
weaning and trialing off extracorporeal support is radically different for venovenous
(VV) and venoarterial (VA) ECMO, because of the difference in physiological sup-
port provided by the two modalities. Hence, we will present a short, separate section
on weaning from VA ECMO. Finally, we review the termination of ECMO in
patients where further therapy is deemed unrecoverable.
Signs of Readiness
Therapy directed toward improving pulmonary function should start prior to patients
being placed on ECMO and continue after ECMO support has been initiated.
Therapies should be targeted to the individual pathology that led to the need for
respiratory support, but in general may include diuresis, antibiotic therapy, broncho-
dilators, and bronchoscopy with airway clearance. Ventilator management while on
ECMO involves limiting FiO2 to the lowest possible (21–50%), maintaining PEEP
to preserve recruitment, and limiting tidal volumes to 4–8 mL/kg.
For ventilated ARDS patients supported with mechanical ventilation, improve-
ment in pulmonary function can be assessed initially while the patient is still on
full ECMO support. Even with ventilator settings unchanged from “rest” settings
typically used during extracorporeal support, an increase in PaO2 or a decrease in
sweep flow to maintain normocarbia might occur, indicating improving native
lung function. Additionally, improved pulmonary compliance will lead to
increased tidal volumes on unchanged pressure-control ventilator settings.
Usually, respiratory system compliance >20 mL/cm H2O suggests adequate recov-
ery to proceed with weaning. An improvement in pulmonary aeration on chest
radiograph can accompany improving alveolar recruitment, but at times the radio-
graph may lag behind clinical improvement. Lung ultrasound can be used to eval-
uate the subpleural areas of the lung. Signs of improving lung function are listed
in Table 15.1.
Once significant aeration is achieved, as assessed by improving tidal volumes
(>150 mL) on rest settings and improvements in chest radiograph, the Cilley test
should be performed daily. The patient should be hemodynamically stable and
comfortable on mechanical ventilation. To perform the Cilley test, also called the
oxygen challenge test, the FiO2 is increased to 100%, with no other changes in
ventilator settings. If the patient has improved lung aeration, there will be a rapid
increase in oxygen saturation within a couple of minutes and an arterial blood gas
should show a PaO2 of >225 mmHg at 15 min after the increase in FiO2 [1]. This
ARDS patients with improving pulmonary function can be weaned and liberated
from extracorporeal support. When the lowest ECMO flow to provide adequate sup-
port at low ventilator settings is used, weaning from VV ECMO becomes a more
gradual process. Additionally, with improving lung function, sweep gas flow rate
will decrease to maintain normocarbia, indicating improved alveolar gas exchange.
As such, weaning of VV ECMO should be a routine progression of ECMO manage-
ment in the setting of improving pulmonary function. On occasion, the decision to
wean off extracorporeal support might be hastened by bleeding complications
related to anticoagulation. With moderate ventilatory support, a trial off during VV
ECMO includes assessing native lung function while discontinuing all sweep gas
flow to the ECMO circuit. Importantly, blood flow continues through the VV circuit,
without extracorporeal gas exchange. Finally, liberation from ECMO involves
decannulation and discontinuation of anticoagulation. It is generally accepted that
when extracorporeal support is contributing to less than 30% of native organ func-
tion, it is possible to trial off ECMO [3, 4]. An algorithm for VV ECMO decannula-
tion used at the authors’ institution is presented in Fig. 15.1.
Ventilator Management
Recruitment Maneuvers
Ventilator settings:
FiO2 <50% TV 6 ml/kg,
Pplat<30 mm Hg, RR <25
YES
much as one would judge the success of a spontaneous breathing trial. Additionally,
P0.1 (negative pressure generated during the first 100 msec after the initiation of an
inspiratory effort against a closed respiratory circuit) will increase in patients with
increased work of breathing during the ECMO wean and trial off. While values
>2–3 cm H2O have been shown to be predictive of failed extubation, there is cur-
rently no validated P0.1 cutoff to assess readiness for liberation from ECMO
support.
15 ECMO Weaning and Decannulation 269
There are multiple ways to wean ECMO and different institutions have different
preferences. Various methods will be described subsequently.
In this weaning strategy, the patient’s oxygenation is trialed prior to their ability to
decarboxylate [1]. Before commencing the oxygenation challenge, the FiO2 on the
ventilator is increased to 60%, since 60% or less would be an adequate FiO2 on the
ventilator for decannulation. Additionally, ventilator settings are increased to non-
rest settings, but targeting tidal volumes of 6–8 mL/kg ideal body weight. After a
few minutes at these settings, the sweep gas FiO2 is decreased sequentially from
100% to 60%, 30%, and then 21% [1]. During the FiO2 wean, a peripheral satura-
tion of >88% should be maintained. If the patient’s saturation falls below this value,
the patient does not meet criteria for continuing the wean and the patient is placed
back on full ECMO support and rest settings on the ventilator can be resumed. If the
patient continues to meet weaning criteria and the FiO2 of the sweep gas is decreased
to 21%, the carboxy challenge can be performed. In the carboxy challenge, the
sweep gas flow rate is reduced by 30% every 5–10 min while measuring the patient’s
response (end-tidal CO2, CO2 by blood gases, in addition to physical exam findings
such as tachypnea) [1]. Once the sweep gas flow is turned off, the weaning test is
successful and a trial off ECMO can ensue. Blood gases can be drawn to continue
to ensure adequate oxygenation and CO2 removal. ECMO circuit flow is never
reduced and additional anticoagulation is not necessary during trial off ECMO. Trials
off ECMO can last between 4 and 24 h prior to decannulation.
ventilation. As the trial off ECMO continues, blood gases can be less frequent if the
oxygenation and ventilation have been sufficient. Signs of failure of trial off are
detailed in Table 15.2. Decannulation can follow if trials are successful for 2–24 h.
At some institutions, or in select patients, a trial off can proceed for a longer time
before decannulation is deemed appropriate. If the patient shows signs of inade-
quate native pulmonary function, the sweep gas flow is simply turned back on and
extracorporeal support is resumed. Ventilatory settings and sedation can be man-
aged to the patient’s comfort, until the next trial off period.
Circuit blood flow does not need to be changed while attempting weaning from VV
ECMO. However, as the patient’s intravascular volume is decreased with diuresis, it
might be difficult to maintain a high rate of blood flow through the extracorporeal
circuit. As the patient’s pulmonary gas exchange improves, their demands from
ECMO may lessen and it may be possible to decrease circuit blood flow. A mini-
mum flow rate of 500–1000 mL/min needs to be maintained to prevent stasis of
blood in the cannulae, circuit, pump, and oxygenator. Low blood flow rates through
the ECMO circuit should also prompt higher anticoagulation goals to prevent
thrombosis.
Sedation Management
While patients are on VV ECMO, sedation should be maintained at the lowest level
possible to maintain patient comfort and hemodynamic stability. In patients with a
tracheostomy, it may be possible to maintain complete wakefulness without seda-
tion during ECMO weaning or trial off ECMO, while intubated patients may require
some degree of sedation to maintain comfort. In patients on mechanical ventilation,
sometimes sedation may need to be increased during ECMO weaning or trial off to
prevent ventilator dyssynchrony. Each patient will have different needs, but the low-
est level of sedation to maintain patient comfort, hemodynamic stability, and venti-
lator synchrony should be used.
15 ECMO Weaning and Decannulation 271
Hemodynamic Management
Anticoagulation Management
Decannulation
Following Decannulation
Since patients have demonstrated adequate gas exchange (and hemodynamic sta-
bility) prior to stopping ECLS, decannulation is usually well-tolerated. Continued
recovery can be anticipated and attention turned to further healing, rehabilitation,
and prevention of complications. Additionally, if appropriate, further weaning and
liberation from mechanical ventilator may be appropriate. Procedures that were
withheld because of the anticoagulation necessary for extracorporeal support (e.g.,
tracheostomy, chest tube placement, invasive line placement) can now be
performed.
Some patients will have setbacks, for example, with worsening dyspnea. It is
helpful to return to basic principles, considering the differential diagnosis for the
deterioration (fluid overload, thromboembolism), performing appropriate diagnos-
tic studies, and using the ventilator to support respiratory function (or, for the extu-
bated patient, noninvasive ventilation or re-intubation). In the 24–48 h after
decannulation, patients also need to be assessed for venous thrombosis at the can-
nula site.
Weaning from VA ECMO is different from VV ECMO for three major reasons: first,
both the cardiac and pulmonary systems will be gradually removed from extracor-
poreal support; second, the ECMO flow is decreased but sweep gas flow is
unchanged; and third, ECMO support can never truly be interrupted for long periods
of time because of concerns for stasis.
Assessment of readiness for weaning and discontinuation should be performed
daily. Signs of improving organ function include improved pulsatility in the arterial
waveform, reduction in cardiac filling pressures, improved organ function, and an
improvement in radiographic and sonographic signs [7].
For weaning, ECMO flow is reduced slowly by 0.5–1 L/min every 4–6 h.
Ventilatory and hemodynamic support might have to be uptitrated as ECMO wean
progresses. At a flow of about 1 L/min, if the patient can maintain a mean arterial
pressure >65 mmHg, mixed venous oxygen saturation >65%, and an arterial satura-
tion > 90% with low ventilator settings and low vasoactive support, they might be
ready for a trial off from ECMO support. After adequate information has been
acquired, the patient is placed back on 1.5–2 LPM flow while awaiting
decannulation.
Often, the trial off from ECMO occurs in the operating room, with hemodynamic
and echocardiographic monitoring. With clamps on the cannulae, the patient is sep-
arated from the pump. A bridge between the two limbs of the circuit keeps blood
flowing across the extracorporeal circuit. If a prolonged trial is desired, occasional
flashing (release of clamps on the cannulae) can be used to keep clots from forming.
However, since this maneuver introduces oxygenated blood into the patient, any
further assessments of gas exchange should be made prior to flashing the cannulae.
Decannulation is performed in the operating room, to allow for good vascular
control of the arterial cannulation site. For patients placed on VA or VV ECMO as
a bridge to transplantation or durable mechanical support, decannulation occurs in
the operating room along with the surgical procedure.
Most patients treated with ECLS will be successfully weaned or bridged to trans-
plant. Others, however, simply never recover. The primary disease may fail to
resolve; contraindications to transplant arise; ECLS complications (especially intra-
cranial hemorrhage) change the trajectory of illness; or the consequences of critical
274 S. L. McCartney and S. Krishnan
illness, such as nosocomial pneumonia, septic shock, or acute renal failure turn a
desperate situation into an unrecoverable downward spiral (Table 15.3). It is impor-
tant to remember that some patients will recover despite many weeks, even months,
of ECLS, so decisions to terminate support should not be made casually.
Determining that continued critical care is futile is controversial and exceedingly
complex [8, 9], perhaps more so in ECLS patients [10, 11]. Difficult decisions are
made easier when there is clarity of goals even before ECLS is begun and when
these goals are revisited regularly. When initiating ECLS, its purpose should be
stated and these goals should be shared amongst the healthcare team and the patient
(or surrogates). If ECLS is begun as a bridge to transplant, it is imperative to assess
candidacy realistically and to be certain that there are no contraindications to trans-
plant that could strand the patient on a “bridge to nowhere” [11]. Laying out goals
for patients who are being bridged to recovery is more challenging, because recov-
ery can be protracted and hard to predict accurately, and the course may be punctu-
ated with periodic downturns that make survival seem unlikely. The ECLS team
should meet regularly with the patient to address status, progress toward the goals
(or regression), and prognosis. For those unlikely to survive, the concept of a time-
limited trial (in which explicit metrics and a clear schedule are enunciated) [12] can
be helpful for patients and families. While the ethics of prolonging or terminating
extracorporeal support are discussed in future chapters, we will discuss the practical
aspects of ECLS termination here.
When the caregivers and patient concur that continued ECLS cannot achieve its
goals and should be withdrawn, there is no ethical dilemma. In this circumstance,
we advocate changing the treatment goals to focus on comfort and withdrawing all
unnecessary ICU interventions. Since ECLS is never instituted as destination ther-
apy, most consider its termination to be similar to withdrawing other life-support
treatments. Nevertheless, since withdrawal generally leads to immediate death,
some discomfort is understandable, especially when caring for conscious patients.
15 ECMO Weaning and Decannulation 275
Once a decision is made to cease ECLS in a patient expected to die, this change
in plan should be communicated to primary caregivers, consultants, referring physi-
cians, ICU nurses, perfusionists, therapists, and other stakeholders. Involvement of
palliative care specialists and spiritual care services may be helpful [13]. Patient
comfort should be assured, using analgesics and sedatives as needed, and anticipat-
ing that withdrawal of VV ECLS may provoke or heighten dyspnea. We then turn
off the sweep gas, preventing further gas exchange in the membrane. Assuming VV
ECMO was life-sustaining, the termination of support will lead to death. For those
on VA ECLS, circuit blood flow can be reduced as vasoactive drug infusions are
stopped. The patient should be kept comfortable from worsening dyspnea or other
perceived discomfort.
Funding Statement Support was provided solely from institutional and/or departmental sources.
References
1. Vasques F, Romitti F, Gattinoni L, et al. How I wean patients from veno-venous extra-corporeal
membrane oxygenation. Crit Care. 2019;23:316.
2. Hartley EL, Sanderson B, Vasques F, Daly KJ, Lozinski M, Barrett NA, Camporota
L. Prediction of readiness to decannulation from venovenous extracorporeal membrane oxy-
genation. Perfusion. 2020;35(1_suppl):57–64.
3. Lynch WR. Weaning, trialing and futility. In: ECMO: extracorporeal cardiopulmonary sup-
port in critical care. 4th ed. ELSO; 2012.
4. ELSO guidelines for adult respiratory failure; August 2017.
5. Marhong JD, Munshi L, Detsky M, et al. Mechanical ventilation during extracorporeal life
support (ECLS): a systematic review. Intensive Care Med. 2015;41(6):994–1003.
6. Marhong JD, Telesnicki T, Munshi L, et al. Mechanical ventilation during extracorporeal
membrane oxygenation. An international survey. Ann Am Thorac Soc. 2014;11(6):956–61.
7. Fried JA, Masoumi A, Takeda K, Brodie D. How I approach weaning from venoarterial
ECMO. Crit Care. 2020;24(1):307.
8. Luce JM. A history of resolving conflicts over end-of-life care in intensive care units in the
United States. Crit Care Med. 2010;38:1623–9.
9. Siegel MD. End-of-life decision making in the ICU. Clin Chest Med. 2009;30:181–94.
10. Rosenberg AA, Haft JW, Bartlett R, Iwashyna TJ, Huang SK, Lynch WR, Napolitano
LM. Prolonged duration ECMO for ARDS: futility, native lung recovery, or transplantation?
ASAIO J. 2013;59:642–50.
11. Abrams DC, Prager K, Blinderman CD, Burkart KM, Brodie D. Ethical dilemmas encountered
with the use of extracorporeal membrane oxygenation in adults. Chest. 2014;145:876–82.
12. Schenker Y, Tiver GA, Hong SY, White DB. Discussion of treatment trials in intensive care. J
Crit Care. 2013;28:862–9.
13. Cook D, Rocker G. Dying with dignity in the intensive care unit. N Engl J Med.
2014;370:2506–14.
Chapter 16
Venoarterial ECMO in Respiratory Failure
Introduction
When used to treat respiratory failure, the primary benefit of extracorporeal circula-
tion (ECMO) is to facilitate gas exchange. With a membrane lung (ML) in the extra-
corporeal circuit, oxygen can be added, and carbon dioxide can be removed from
venous blood to preserve oxygenation and carbon dioxide homeostasis. In that way,
patients with damaged or diseased lungs incapable of adequate gas exchange can be
supported until their lungs recover.
Although venoarterial (VA) ECMO is more commonly used in adults overall [1],
venovenous (VV) ECMO is more common in patients with respiratory failure due
to ARDS [2]. However, VA ECMO may also be used to provide cardiovascular sup-
port to patients with both hemodynamic instability and respiratory failure. When
configured in this manner, the ECMO circuit returns oxygenated blood not to the
venous system but to the arterial circulation. This approach can widen the range of
patients with respiratory failure who can be supported by ECMO technology.
VA ECMO may be indicated in two broad categories of patients with respiratory
failure. The first involves disease states that, when combined with respiratory fail-
ure, may require VA support. Examples include cardiogenic shock due to failure of
the right or left ventricle (or both), septic/vasoplegic shock requiring augmentation
of cardiac output for blood pressure support, and other conditions that prevent ade-
quate cardiac output such as pulmonary emboli, myocarditis, intracardiac shunts, or
persistent arrhythmias.
A. Tung (*)
Department of Anesthesia and Critical Care, University of Chicago, Chicago, IL, USA
e-mail: [email protected]
T. H. Song
Section of Cardiac Surgery, Department of Surgery, University of Chicago, Chicago, IL, USA
e-mail: [email protected]
1. Concurrent LV dysfunction
(a) Ischemic heart disease
(b) Takotsubo syndrome
(c) Stunned myocardium
(d) Myocarditis
2. Severe RV dysfunction
(a) Acute cor pulmonale due to ARDS
(b) Pulmonary embolism
3. High cardiac output states
(a) Septic shock
4. Intractable arrhythmias
16 Venoarterial ECMO in Respiratory Failure 279
Fundamentals of VA ECMO
Anatomy
As the name suggests, VA ECMO differs from VV ECMO in locating the ECMO
return cannula in the arterial side instead of the venous side of the circulation.
Venous blood is drained from the right atrium or superior/inferior vena cava into the
pump, routed through an ML, and returned to the body via the arterial system.
Locations for the venous (drainage) cannula are described in Chaps. 3 and 4 and
include central, femoral, internal jugular, and (less commonly) subclavian sites. For
VA ECMO, the size and location of the return cannula is limited by the size of the
target artery and the ability to preserve flow distal to the artery after partial or total
occlusion due to the cannula. Common locations for the return cannula thus include
central, femoral, and axillary arteries. Although cannula locations are similar to
those of a standard cardiopulmonary bypass (CPB) circuit, usually a CPB circuit
involves passive drainage of venous blood into a reservoir which then feeds the
centrifugal pump. In contrast, VA ECMO has no reservoir, so adequacy of venous
drainage is crucial to arterial return.
Equipment
Cannulation techniques for the drainage cannula are similar to those used for VV
ECMO. Because VA ECMO is often used to support cardiac output and blood pres-
sure in hemodynamically unstable patients, adequacy of venous drainage is critical
to hemodynamic support and locating the tip of the venous cannula in the right
atrium and/or vena cavae is thus a necessary component of VA ECMO. Venous can-
nula sizes for VA ECMO in adults typically range from 21 to 27F.
As with VV ECMO, a centrifugal pump is used to draw blood from the venous
drainage line, actively pump it through a ML, and return it to the circulation. Unlike
VV ECMO however, VA ECMO creates a parallel circuit where blood bypasses the
heart and lungs and is pumped from the venous to arterial circulation. A ML is thus
mandatory for VA ECMO as deoxygenated venous blood would otherwise be
shunted into the arterial system and produce severe arterial oxygen desaturation.
The sweep gas therefore cannot be turned off in VA ECMO. With large, well-
positioned cannulas, it is possible to generate blood flows as high as 7 LPM, but 3–5
LPM flows are more typical.
Return cannulas are usually 15–21F in diameter. Although larger cannulas may
permit higher flow, they are also associated with a higher incidence of cannulation-
related adverse events such as bleeding and vascular injury. In addition, for a return
cannula that is inserted into the femoral artery, ischemia of the ipsilateral leg may
occur unless a distal perfusion catheter is placed. This catheter is often a 6- to 8-F
introducer placed distal to the insertion site that diverts blood flow from the return
280 A. Tung and T. H. Song
cannula to the artery distal to the cannula insertion site (usually the superficial fem-
oral artery). The incidence of lower extremity ischemia with femoral cannulation
historically ranges from 12% to 22% [4] and fasciotomy for compartment syndrome
or amputation occurs with a measurable frequency. Distal perfusion catheters reduce
the incidence of such complications [5].
As with VV ECMO, anticoagulation is required for VA ECMO to prevent throm-
bosis which may result in thromboembolic events, ML failure, or pump malfunc-
tion. Bleeding occurs in approximately 25% of all VA ECMO patients [6] and is
seen even in the absence of anticoagulation. Both heparin and direct thrombin inhib-
itors such as bivalirudin and argatroban have been used for systemic anticoagulation
[7]. Few data exist to identify an optimal anticoagulation strategy and it is likely that
no single approach is best (see Chap. 8).
Adequacy of anticoagulation may be monitored via a variety of strategies. For
patients on heparin, the 2014 ELSO guidelines suggest titrating to an activated clot-
ting time (ACT) = 180–220 s [8]. Anti-Xa monitoring has also been used with an
ELSO-suggested goal of 0.3–0.7 iU/mL [8]. Prothrombin time (PTT) ranges
between 50 and 80 s are also used in some centers. Divergence between measures
of coagulation is common, and clinical experience suggests that, after prolonged
ECMO runs, target Anti-Xa levels may be associated with supratherapeutic PTT
levels [9]. Although thromboelastography (TEG) may permit more precise titration
of the degree of anticoagulation, existing data suggest a poor correlation between
point-of-care TEG testing and bleeding and thrombotic complications [10]. As with
any therapeutic titration, evidence of bleeding or clotting should prompt a review of
anticoagulation management.
Monitoring
VA ECMO differs from VV ECMO in affecting cardiac output and end-organ perfu-
sion in addition to gas exchange. As a result, monitoring goals for VA ECMO
include adequacy of limb perfusion distal to the (usually femoral) return cannula,
function of the native heart and lungs, adequacy of upper body oxygenation (for
femoral cannulation), and effectiveness of overall cardiorespiratory support in addi-
tion to circuit, cannula, and ML function and integrity. In addition, weaning from
VA ECMO differs from weaning from VV ECMO because both hemodynamic and
respiratory support are being weaned.
Because gas exchange in patients on VA ECMO occurs via the ECMO circuit,
working in parallel with the patient’s native heart and lungs, monitoring adequacy
of oxygenation via the ECMO circuit may be needed to troubleshoot inadequate
systemic O2 saturation. Factors affecting the PO2 at the ML outlet include blender
16 Venoarterial ECMO in Respiratory Failure 281
concentration, circuit blood flow rate, PO2 at the oxygenator intake, and amount of
microthrombi in the ML. In general, if blood flow through the ML is below the rated
flow for the device and the inflow saturation is 70% or higher, the saturation at the
ML outlet should be >95% [11]. Because many factors affect post-membrane oxy-
genation, a reasonable clinical strategy for monitoring ML function is to obtain a
baseline post-ML PO2 5–10 min after initiation and monitor trends over time (see
Chap. 9).
Unlike VV ECMO, VA ECMO affects oxygen delivery not just by improving arte-
rial oxygen saturation but also by augmenting systemic cardiac output to deliver
oxygenated blood to peripheral organs. Because systemic blood flow in patients on
VA ECMO involves the output of both the native circulation and the ECMO system
itself, directly monitoring systemic cardiac output (and thus oxygen delivery)
requires monitoring both native cardiac output and ECMO flow. As a rough rule of
thumb, a normal blood pressure and systemic SvO2 (>70%) suggest adequate oxy-
gen delivery to peripheral organs.
Because overall oxygenation represents a combination of ECMO and native lung
function, diagnosing the cause of changes in systemic oxygenation can be challeng-
ing. Native lung function, ML function, or differences in flow delivery between the
ML and the native circulation may all play a role. Patients with respiratory failure
who require VA ECMO may also have septic or cardiogenic shock, and as a result,
the amount of native cardiac output passing through the lungs and heart (instead of
the ECMO device) may be higher or lower.
Estimating the cardiac output passing through the native heart and lungs is dif-
ficult. Arterial pulsatility is an imperfect monitor for left ventricle (LV) ejection but
suggests at least some intrinsic cardiac output. Transthoracic echocardiography
allows more accurate estimates of stroke volume by measuring the left ventricular
outflow tract velocity-time integral. Perhaps the most straightforward monitoring
approach is pulmonary artery (PA) catheterization, which permits direct measure-
ment of intracardiac flow.
The interaction between native and ECMO flows in a VA ECMO circuit can also
have consequences on cardiac function. In principle, increasing ECMO flow would
augment venous drainage from the right atrium and decrease the amount of blood
flow passing through the native heart/lungs. In practice, however, the increased LV
afterload due to increased ECMO pump output may also limit LV ejection and con-
tribute to LV distention. In patients with cardiogenic shock and LV dysfunction,
such distention may slow cardiac recovery [12]. In addition, LV distention can
increase left atrial pressure, potentially worsening pulmonary edema (and oxygen-
ation via the native lungs).
Several strategies for avoiding, detecting, and monitoring LV distention have
been suggested. These include echocardiography to assess for aortic valve opening,
LV size, and stasis/clot formation; chest radiography to detect worsening
282 A. Tung and T. H. Song
When lung failure is present and the VA ECMO return cannula is in the femoral
artery, deoxygenated blood from the native circulation may enter subclavian and
carotid arteries before mixing with oxygenated blood traveling retrograde from the
distal aorta. The extent to which desaturated blood from the native circulation enters
the cerebral vessels depends on the balance between ECMO and native flows. Due
to dynamic changes in the ratio of native to ECMO blood flow, the “watershed” area
may change as cardiac function improves.
If oxygenated blood from the femoral ECMO return cannula does not reach the
upper body, the syndrome is variously termed “differential hypoxia,” “North-
South,” or “Harlequin” syndrome. Relative or absolute cerebral hypoxemia may
result, so patients on VA ECMO with a femoral return cannula require monitoring
[16]. Pulse oximetry, arterial waveform, and blood gas monitoring at the right radial
location will allow continuous assessment of oxygen saturation at the right brachio-
cephalic artery and hence the right common carotid artery. A large right radial pulse
pressure suggests potentially significant native ejection (of deoxygenated blood in
patients with lung failure). Blood gas monitoring from the right radial artery will
enable more precise monitoring of oxygen saturation but can only be performed
intermittently.
Strategies to ameliorate the risk of upper body desaturation due to Harlequin
syndrome include optimization of native lung function and increasing ECMO flow
to push the watershed zone proximal to the aortic arch [15]. Increasing ECMO flow
may also worsen LV distention as described above, potentially worsening
16 Venoarterial ECMO in Respiratory Failure 283
Weaning
Unlike VV ECMO, which is initiated for hypoxia or hypercapnia and can be weaned
by reducing the sweep gas flow to zero, VA ECMO requires that the ML be kept on
to prevent returning deoxygenated blood to the arterial system. Weaning from VA
ECMO thus involves reducing the ECMO circuit flow while monitoring systemic
perfusion and pressure (see Chap. 15). During VA ECMO weaning, oxygen satura-
tion may not change, or not be helpful for guiding the weaning process. Because
most uses of VA ECMO are for cardiogenic shock with adequate lung function,
oxygenation is usually not as much of a concern. But when VA ECMO is used for
respiratory failure, a reduction in ECMO flow may produce desaturation. Since the
ECMO circuit cannot be completely stopped due to thrombosis concerns, VA
ECMO is generally weaned by reducing the flow to a lower limit (1–2 LPM) while
monitoring oxygen saturation and hemodynamic stability. Adequacy of native lung
function to support oxygenation unassisted should be verified before decannulation.
In most cases, by the time the patient is considered ready for decannulation, hemo-
dynamic instability has resolved, the native heart is ejecting, and ECMO flow is
contributing little oxygenated blood to the upper body where saturation is usually
monitored.
284 A. Tung and T. H. Song
Vasoplegic Shock
Since VV ECMO does not support the circulation, respiratory failure complicated
by severe LV or RV dysfunction may require VA ECMO. When considering VV
ECMO for patients with hemodynamic instability, the circulation should be assessed
comprehensively, generally including echocardiography. Indications for VA instead
of VV ECMO include hemodynamic instability despite vasoactive infusion, preex-
isting right or left heart failure, elevated lactate or low SvO2 despite maximal medi-
cal therapy, valvular dysfunction preventing adequate cardiac output, or intractable
arrhythmias.
LV Dysfunction The most common indication for VA ECMO is severe LV failure.
In such cases and especially after cardiac arrest, respiratory failure may follow car-
diogenic shock due to elevations in LV filling pressure and consequent pulmonary
edema. For those patients, cardiac dysfunction is the primary reason for VA ECMO
support. Alternatively, respiratory failure precedes hemodynamic failure, as when
stress (takotsubo) cardiomyopathy complicates ARDS or when severe hypoxemia
produces a stunned myocardium.
By supporting both lung and LV function, VA ECMO may restore myocardial
oxygen supply-demand balance in patients with ischemic cardiomyopathy or buy
time for recovery when LV dysfunction is reversible. Other causes of cardiogenic
shock that respond well to VA ECMO but are beyond the scope of this chapter
include fulminant myocarditis [24] and refractory arrhythmias [25]. As noted above,
increased left ventricular loading with high ECMO flow rates may hamper myocar-
dial recovery [26] so attention should be paid to the degree of LV distention during
VA ECMO.
RV Dysfunction Severe hypoxemic respiratory failure may itself induce a suffi-
cient increase in pulmonary vascular resistance to cause RV dysfunction or failure.
Clinical clues include elevated right atrial pressure, tricuspid regurgitation, or echo-
286 A. Tung and T. H. Song
cardiographic signs (see Chap. 7). A decision to initiate VV ECMO in such patients
initially can be reasonable as existing data suggests that right ventricular function
may improve with VV support alone [17, 27]. However, if RV dysfunction is pro-
found, VA ECMO may be a better choice. Similarly, if RV function deteriorates on
VV ECMO, vasopressor requirements escalate, or hypoxemia persists after initia-
tion, then converting to VVA ECMO may stabilize cardiorespiratory function while
the lung and heart recover. Such support may allow time to optimize mechanical
ventilation, restore metabolic homeostasis, and permit the right ventricle to recover.
Evidence to support VA ECMO for RV support in patients with ARDS remains
at the level of case reports [28, 29]. A small trial compared VV, VA, and VVA ECMO
and found no clear difference in mortality. However, the incidence of RV dysfunc-
tion in ARDS is as high as 25% [30] and associated with greater mortality [31]. It
remains unclear whether the increased complexity of dual return cannulas and arte-
rial cannulation offsets the physiological benefit of improved oxygenation and sys-
temic blood flow. More recently, a specialized, dual-lumen cannula that returns flow
directly into the pulmonary artery has been used to support both lung and RV func-
tion. However, VA and VVA ECMO remain viable options in certain situations.
Indications for adding an arterial return cannula to VV ECMO or converting to
VA ECMO for cardiogenic shock due to associated right heart failure include dete-
riorating hemodynamic status while on VV ECMO, increasing lactate levels despite
maximal vasopressor therapy, worsening or persistently severe hypoxemia, and
echocardiographic evidence of severe RV dysfunction (RV dilation, decreased RV
contractility, new or worsening tricuspid regurgitation).
Acute pulmonary embolism is another cause of severe pulmonary and cardiovas-
cular dysfunction which may respond to VA ECMO. Such patients are often hypox-
emic due to acute lung failure and in cardiogenic shock due to acute increases in RV
afterload. While initial treatment with thrombolytic therapy lowers mortality in
hemodynamically unstable patients, VA ECMO may improve survival in those who
remain in cardiorespiratory failure after thrombolytic therapy or in whom throm-
bolysis is contraindicated. One case series described 95% survival-to-discharge in
patients receiving VA ECMO as a primary management strategy for PE with cardio-
genic shock and an 80% survival in patients presenting with cardiac arrest [27].
These results likely represent an upper bound on survival as other case series report
worse outcomes [32, 33]. As with ECMO for sepsis, survival rates vary widely
between studies and a clear understanding of what factors predict successful out-
come remains elusive. As an example, an elevated lactate level on presentation has
been reported as a risk factor for poor outcome in some cohorts [33] but not others
[27]. Regardless of predictors, an organized ECMO protocol is likely needed to
optimize outcomes with VA ECMO for hemodynamically unstable patients with PE
as they may rapidly evolve to cardiac arrest.
In general, PE sufficiently severe to cause hypoxemia usually also causes right
ventricular dysfunction. As a result and because the clinical course of PE is often
uncertain, VA ECMO is more commonly chosen when extracorporeal support is
needed. Indications for choosing VA ECMO for patients with PE include hemody-
namic instability; ready availability of ECMO support; failure of thrombolytic
16 Venoarterial ECMO in Respiratory Failure 287
The last category of patients with hypoxemic respiratory failure who may benefit
from VA ECMO are those who remain persistently hypoxemic despite maximal VV
support. Since correction of hypoxemia using VV ECMO relies on capturing a large
fraction of cardiac output (see Chap. 1), this scenario is usually seen in high output
states. Hypoxia and hypercarbia can contribute to agitation in patients on ECMO,
impeding their ability to participate in physical therapy. ELSO guidelines define
adequate oxygen support as an arterial oxygen saturation > 80% and adequate CO2
removal as ability to remove the CO2 produced by consuming oxygen at a 3 cc/kg/
min rate [11].
Reasons for persistent hypoxemia or inadequate CO2 clearance during VV
ECMO include ML malfunction, excessive recirculation, and inadequate circuit
blood flow. Of these, the former is easily diagnosed and treated (Chap. 9). Managing
recirculation may not be as straightforward. Some degree of recirculation is often
present with VV ECMO. Factors that increase the likelihood of recirculation include
close positioning of the SVC and IVC cannulas, single (vs multistage)-lumen drain-
age cannulas [34], and increased ECMO flow rates [35]. Circuit blood flow may rise
with fluid loading, cannula repositioning, or exchange for larger cannulas, but when
cardiac output is high, it may simply be impossible to achieve flows sufficient for
adequate oxygenation.
Inadequate oxygen saturation despite optimal VV ECMO support is difficult to
treat. Even if hemoglobin levels are increased to allow adequate oxygen delivery, an
arterial oxygen saturation < 90% can lead to persistent agitation requiring heavy
sedation. Although ELSO guidelines define adequate oxygen support as an arterial
oxygen saturation > 80% [11], in clinical practice such patients almost always
require heavy sedation to control persistent agitation.
Conversion to VA ECMO with femoral cannulation invites a severe harlequin
syndrome as such patients often have elevated cardiac outputs. Central cannulation
is extremely invasive. A third potential solution is to insert an arterial return cannula
in the right axillary artery. Oxygenated blood from the ECMO circuit then enters the
right common carotid artery, ensuring that blood entering the cerebral vessels is
adequately oxygenated. Generally, 15- to 19-F cannulas are used for this purpose
and can be placed percutaneously or surgically with a graft. Small case series [36,
37] suggest that full VA ECMO support is feasible with this approach although
complications include arm swelling, hematoma, and infection [37]. With an ambu-
latory VA configuration (drainage usually from the internal jugular vein and return
288 A. Tung and T. H. Song
via axillary artery cannulation), a venous drainage limb is not necessary although
may be useful to prevent reperfusion injury to the lung.
Indications for adding an arterial return cannula to existing VV ECMO or con-
verting to ambulatory VA ECMO include persistent hypoxemia/hypercarbia, agita-
tion due to inadequate arterial oxygen saturation, poor response to cannula
repositioning or increasing ECMO flows, and adequacy of vascular access.
Conclusion
References
1. ECLS Registry Report, October 2021. Extracorporeal Life Support Organization (ELSO).
Available at: https://www.elso.org/Registry/Statistics/InternationalSummary.aspx. Accessed
11/4/2021.
2. Kon ZN, Bittle GJ, Pasrija C, Pham SM, Mazzeffi MA, Herr DL, Sanchez PG, Griffith
BP. Venovenous versus venoarterial extracorporeal membrane oxygenation for adult patients
with acute respiratory distress syndrome requiring precannulation hemodynamic support: a
review of the ELSO registry. Ann Thorac Surg. 2017;104:645–9.
3. Abrams D, Bacchetta M, Brodie D. Recirculation in venovenous extracorporeal membrane
oxygenation. ASAIO J. 2015;61:115–21.
4. Cheng R, Hachamovitch R, Kittleson M, Patel J, Arabia F, Moriguchi J, Esmailian F, Azarbal
B. Complications of extracorporeal membrane oxygenation for treatment of cardiogenic shock
and cardiac arrest: a meta-analysis of 1,866 adult patients. Ann Thorac Surg. 2014;97:610–6.
16 Venoarterial ECMO in Respiratory Failure 289
5. Juo YY, Skancke M, Sanaiha Y, Mantha A, Jimenez JC, Benharash P. Efficacy of distal perfu-
sion cannulae in preventing limb ischemia during extracorporeal membrane oxygenation: a
systematic review and meta-analysis. Artif Organs. 2017;41:E263–73.
6. Sy E, Sklar MC, Lequier L, Fan E, Kanji HD. Anticoagulation practices and the prevalence
of major bleeding, thromboembolic events, and mortality in venoarterial extracorporeal mem-
brane oxygenation: A systematic review and meta-analysis. J Crit Care. 2017;39:87–96.
7. Kaseer H, Soto-Arenall M, Sanghavi D, Moss J, Ratzlaff R, Pham S, Guru P. Heparin vs
bivalirudin anticoagulation for extracorporeal membrane oxygenation. J Card Surg.
2020;35(4):779–86.
8. ELSO Anticoagulation Guideline. Extracorporeal Life Support Organization (ELSO).
Available at: http://www.elso.org/Portals/0/Files/elsoanticoagulationguideline8-2014-table-
contents.pdf. Accessed 11/4/2021.
9. Adatya S, Sunny R, Fitzpatrick MJ, Colvin M, Thennapan T, John R, Dodge Zantek N, Pritzker
M, Eckman P, Uriel N. Coagulation factor abnormalities related to discordance between anti-
factor Xa and activated partial thromboplastin time in patients supported with continuous-flow
left ventricular assist devices. J Heart Lung Transplant. 2016;35:1311–20.
10. Jiritano F, Fina D, Lorusso R, Ten Cate H, Kowalewski M, Matteucci M, Serra R, Mastroroberto
P, Serraino GF. Systematic review and meta-analysis of the clinical effectiveness of point-of-
care testing for anticoagulation management during ECMO. J Clin Anesth. 2021;73:110330.
11. General guidelines for all ECLS cases, August 2017. Extracorporeal Life Support Organization
(ELSO). Available at: https://www.elso.org/Portals/0/ELSO%20Guidelines%20General%20
All%20ECLS%20Version%201_4.pdf. Accessed 11/4/2021.
12. Kurihara H, Kitamura M, Shibuya M, Tsuda Y, Endo M, Koyangi H. Effect of transaortic catheter
venting on left ventricular function during venoarterial bypass. ASAIO J. 1997;43:M838–41.
13. Rihal CS, Naidu SS, Givertz MM, Szeto WY, Burke JA, Kapur NK, Kern M, Garratt KN,
Goldstein JA, Dimas V, Tu T, Society for Cardiovascular Angiography and Interventions
(SCAI), Heart Failure Society of America (HFSA), Society of Thoracic Surgeons (STS),
American Heart Association (AHA), and American College of Cardiology (ACC). 2015
SCAI/ACC/HFSA/STS Clinical Expert Consensus Statement on the Use of Percutaneous
Mechanical Circulatory Support Devices in Cardiovascular Care: Endorsed by the American
Heart Association, the Cardiological Society of India, and Sociedad Latino Americana
de Cardiología Intervencionista; Affirmation of Value by the Canadian Association of
Interventional Cardiology-Association Canadienne de Cardiologie d’intervention. J Am Coll
Cardiol. 2015;65:2140–1.
14. Ranka S, Mastoris I, Kapur NK, Tedford RJ, Rali A, Acharya P, Weidling R, Goyal A, Sauer
AJ, Gupta B, Haglund N, Gupta K, Fang JC, Lindenfeld J, Shah Z. Right heart catheteriza-
tion in cardiogenic shock is associated with improved outcomes: insights from the nationwide
readmissions database. J Am Heart Assoc. 2021;10:e019843.
15. Rao P, Khalpey Z, Smith R, Burkhoff D, Kociol RD. Venoarterial extracorporeal membrane
oxygenation for cardiogenic shock and cardiac arrest. Circ Heart Fail. 2018;11:e004905.
16. Hoeper MM, Tudorache I, Kühn C, Marsch G, Hartung D, Wiesner O, Boenisch O, Haverich A,
Hinrichs J. Extracorporeal membrane oxygenation watershed. Circulation. 2014;130:864–5.
17. Gutsche JT, Mikkelsen ME, McCarthy FH, Miano TA, Vernick WJ, Ramakrishna H, Patel
PA, Augoustides Y, Szeto WY, Desai ND, Lane-Fall MB, Williams ML. Veno-venous extra-
corporeal life support in hemodynamically unstable patients with ARDS. Anesth Analg.
2017;124:846–8.
18. Buia A, Hopf HB, Herrmann E, Schmandra T, Hanisch E. Septic shock: ECMO beyond
ARDS? Introducing the Simon two-stage protocol when randomisation is considered unethi-
cal. Scand J Trauma Resusc Emerg Med. 2020;28:22.
19. Ro SK, Kim WK, Lim JY, Yoo JS, Hong SB, Kim JB. Extracorporeal life support for adults
with refractory septic shock. J Thorac Cardiovasc Surg. 2018;156:1104–9.
290 A. Tung and T. H. Song
20. Bréchot N, Luyt CE, Schmidt M, Leprince P, Trouillet JL, Léger P, Pavie A, Chastre J, Combes
A. Venoarterial extracorporeal membrane oxygenation support for refractory cardiovascular
dysfunction during severe bacterial septic shock. Crit Care Med. 2013;41:1616–26.
21. Falk L, Hultman J, Broman LM. Extracorporeal membrane oxygenation for septic shock. Crit
Care Med. 2019;47:1097–105.
22. Bréchot N, Hajage D, Kimmoun A, Demiselle J, Agerstrand C, Montero S, Schmidt M, Luyt
CE, Lebreton G, Hékimian G, Flecher E, Zogheib E, Levy B, Slutsky AS, Brodie D, Asfar P,
Combes A, International ECMO Network. Venoarterial extracorporeal membrane oxygenation
to rescue sepsis-induced cardiogenic shock: a retrospective, multicentre, international cohort
study. Lancet. 2020;396(10250):545–52.
23. Evans L, Rhodes A, Alhazzani W, Antonelli M, Coopersmith CM, French C, et.al. Surviving
sepsis campaign: international guidelines for management of sepsis and septic shock 2021.
Intensive Care Med 2021; 47:1181–1247.
24. Diddle JW, Almodovar MC, Rajagopal SK, Rycus PT, Thiagarajan RR. Extracorporeal
membrane oxygenation for the support of adults with acute myocarditis. Crit Care Med.
2015;43:1016–25.
25. Bhandary SP, Joseph N, Hofmann JP, Saranteas T, Papadimos TJ. Extracorporeal life support
for refractory ventricular tachycardia. Ann Transl Med. 2017;5:73.
26. Schrage B, Becher PM, Bernhardt A, Bezerra H, Blankenberg S, Brunner S, Colson P, Cudemus
Deseda G, Dabboura S, Eckner D, Eden M, Eitel I, Frank D, Frey N, Funamoto M, Goßling
A, Graf T, Hagl C, Kirchhof P, Kupka D, Landmesser U, Lipinski J, Lopes M, Majunke N,
Maniuc O, McGrath D, Möbius-Winkler S, Morrow DA, Mourad M, Noel C, Nordbeck P,
Orban M, Pappalardo F, Patel SM, Pauschinger M, Pazzanese V, Reichenspurner H, Sandri
M, Schulze PC, Schwinger RHG, Sinning JM, Aksoy A, Skurk C, Szczanowicz L, Thiele H,
Tietz F, Varshney A, Wechsler L, Westermann D. Left ventricular unloading is associated with
lower mortality in patients with cardiogenic shock treated with venoarterial extracorporeal
membrane oxygenation: results from an International, Multicenter Cohort study. Circulation.
2020;142:2095–106.
27. Pasrija C, Kon Z. A protocolized approach to veno-arterial extracorporeal membrane oxygen-
ation for massive pulmonary embolism. Resuscitation. 2018;126:e12. Epub 2018 Jan 31
28. Sauza-Sosa JC, Rojas-Gomez CA, Montes de Oca-Sandoval MA, Hernandez-Rendon E,
Lima-Linares R, Camarena-Alejo G. Acute cor pulmonale in a critically ill patient with
COVID-19 resolved with venoarteriovenous extracorporeal membrane oxygenation. CASE
(Phila). 2021;5:138–42.
29. Bergman ZR, Prathibha S, Bauman BD, Yannopoulos D, Brunsvold ME. Venoarteriovenous
ECMO in concomitant acute respiratory distress syndrome and cardiomyopathy associated
with COVID-19 infection. Case Rep Crit Care. 2021;2021:8848013.
30. Mekontso Dessap A, Boissier F, Charron C, Bégot E, Repessé X, Legras A, Brun-Buisson
C, Vignon P, Vieillard-Baron A. Acute cor pulmonale during protective ventilation for acute
respiratory distress syndrome: prevalence, predictors, and clinical impact. Intensive Care Med.
2016;42:862–70.
31. Boissier F, Katsahian S, Razazi K, Thille AW, Roche-Campo F, Leon R, Vivier E, Brochard L,
Vieillard-Baron A, Brun-Buisson C, Mekontso DA. Prevalence and prognosis of cor pulmo-
nale during protective ventilation for acute respiratory distress syndrome. Intensive Care Med.
2013;39:1725–33.
32. 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, Sanchez O. Outcomes
after extracorporeal membrane oxygenation for the treatment of high-risk pulmonary embo-
lism: a multicentre series of 52 cases. Eur Heart J. 2018;39:4196–204.
33. George B, Parazino M, Omar HR, Davis G, Guglin M, Gurley J, Smyth S. A retrospective
comparison of survivors and non-survivors of massive pulmonary embolism receiving veno-
arterial extracorporeal membrane oxygenation support. Resuscitation. 2018;122:1–5.
16 Venoarterial ECMO in Respiratory Failure 291
34. Palmér O, Palmér K, Hultman J, Broman M. Cannula design and recirculation during venove-
nous extracorporeal membrane oxygenation. ASAIO J. 2016;62:737–42.
35. van Heijst AF, van der Staak FH, de Haan AF, Liem KD, Festen C, Geven WB, van de Bor
M. Recirculation in double lumen catheter veno-venous extracorporeal membrane oxygen-
ation measured by an ultrasound dilution technique. ASAIO J. 2001;47:372–6.
36. Javidfar J, Brodie D, Costa J, Miller J, Jurrado J, LaVelle M, Newmark A, Takayama H, Sonett
JR, Bacchetta M. Subclavian artery cannulation for venoarterial extracorporeal membrane
oxygenation. ASAIO J. 2012;58:494–8.
37. Biscotti M, Bacchetta M. The “sport model”: extracorporeal membrane oxygenation using the
subclavian artery. Ann Thorac Surg. 2014;98:1487–9.
Chapter 17
Ethical Challenges in Extracorporeal
Membrane Oxygenation
Introduction
Table 17.1 This table defines and describes common ethical principles found throughout this
chapter [5, 6, 8, 21, 22]
Term Definition Comment
Beneficence Acting in the best interest of
the patient; doing good
Nonmaleficence Avoiding harm to the patient
Autonomy A patient’s right to choose A patient must have decision-making
what medical recommendations capacity to exercise their autonomy.
they will accept or reject. A patient cannot demand a medical
intervention that has not been offered, or is
inappropriate or non-indicated.
Justice Equal distribution of goods
amongst a society.
Fairness.
Shared When a physician and patient Proper shared decision-making involves a
decision-making work together to make health physician making recommendations based
care decisions. on clinical knowledge and evidence-based
medicine and a patient sharing their values
and preferences so that a decision can be
made together regarding what is in the
patient’s best interest.
This requires a physician to respect
diversity in patient values, attitudes,
cultures, religions, etc.
Surrogate When a physician and patient Typically, a surrogate decision-maker is
decision-making representative work together to asked to base decisions on the patient’s
make health care decisions for known values and preferences.
a patient who does not have
capacity.
Decisional A patient’s ability to Decisional capacity is determined by
capacity understand the medical provider judgment. In general, a provider
situation and reason through will evaluate whether a patient is thinking
available options in order to clearly, displaying consistency, and making
consent to or refuse a particular rational decisions.
medical treatment. Decisional capacity exists on a spectrum,
that is to say, a different level of capacity
may be required for a simple, low-risk
procedure (i.e., central line insertion), than
for a complex, higher-risk procedure (i.e.,
VV-ECMO).
Advance care A patient expressing their Advance care planning encompasses a wide
planning wishes for medical care should variety of conversations about medical care
they become seriously ill. which may range from a conversation with
a health care provider to a legal
documentation of wishes.
Substituted When a surrogate decision- The patient’s values and preferences may
judgement maker uses a patient’s values have been explicitly expressed; however,
and preferences rather than often they must be inferred by the surrogate
their own to make decisions based on known characteristics of the
about medical care. patient.
17 Ethical Challenges in Extracorporeal Membrane Oxygenation 295
Initiating ECMO
and preferences and making decisions based upon them. This is called substituted
judgement [6, 7]. It is important to take into consideration that decision-making is
often being done in a clinically emergent situation, where the decision-maker may
be under a great deal of fear or stress. Clinicians must be prepared to guide decision-
makers through the process of applying substituted judgement. For example, many
patients have not discussed with their loved ones what they would want in the many
scenarios that may come to light during an ICU stay, especially the idea of ECMO;
however, they likely have shared aspects of their life they find meaningful, valuable,
and important, such as mobility, cognitive function, spending time with family and
friends, work, and independence. Often a patient’s medical preferences can be
inferred based on the values they displayed or espoused in their day-to-day lives. A
point should be made to ensure that the quality of life that is expected with the treat-
ment plan will be acceptable to the patient. For example, being kept alive to be
discharged to a nursing home, fully reliant on others for care, may be acceptable to
one patient, but not to another. Because ECMO is often started in the acute setting,
often under a great deal of emotional stress, expectation-setting is extremely diffi-
cult. Most patients and surrogates are unfamiliar with ECMO and see it as a lifesav-
ing procedure. For this reason, once the patient is stabilized, it is extremely important
to continue conversations with the family to continue to build reasonable expecta-
tions, clarify the goals of ECMO, and discuss scenarios of recovery and
non-recovery.
Continuing ECMO
Often with ECMO initiation, knowledge of whether or not the patient will be a can-
didate for recovery or transplantation is not known until after cannulation and stabi-
lization [5]. By its inherent characteristics, ECMO is a time-limited therapy. That is
to say, if treatment goals are not met (transplant or recovery), the therapy becomes
a bridge to nowhere and, if ECMO is continued, will ultimately be met with com-
plications. This should be explained to family in detail as soon as possible. ELSO
recommends that the possibility of stopping for futility should be explained to the
family before ECMO is initiated. Furthermore, families should be prepared for the
goals of care to shift if treatment goals are not met, or in the face of complications
[1]. One important part of setting expectations with a complicated therapy like
ECMO is to meet families where they are, and to communicate big picture updates,
rather than getting lost in the details. This is best done by setting clear and specific
goals, like recovery of certain organ systems, wakefulness, and rehabilitation [1]. It
is also important to note that expectations may change throughout the course and
multiple meetings are prudent to reassess understanding and reset expectations.
There are many clinicians involved in the care of an ECMO patient, and providers
may change frequently, making consistent messaging difficult [1]. An effort should
be made to ensure that all team members, including providers, nurses, respiratory
therapists, and ECMO specialists, are on the same page about the plan of care. A
17 Ethical Challenges in Extracorporeal Membrane Oxygenation 297
Terminating ECMO
Morbidity and mortality rates increase the longer a patient is on ECMO [2]. Ongoing
ECMO use may be considered inappropriate if a patient is no longer a transplant
candidate, or if recovery is improbable or impossible. For example, if ECMO is
being used as a bridge to lung transplantation and the patient faces severe decon-
ditioning, multiorgan failure, or a difficult to eradicate infection, he may no longer
be a suitable candidate. Similarly, a patient who was originally placed on ECMO as
a bridge to recovery may suffer complications such as severe brain damage, multi-
organ failure, or permanent lung damage [2]. Given the limitations in outcome data,
it is sometimes very difficult to know whether sufficient time has been given to
assess the potential for meaningful recovery [3]. ELSO does give some recommen-
dations regarding stopping ECMO for futility. For example, ECMO “should be dis-
continued promptly if there’s no hope for a healthy survival.” Additional guidance
includes that 2 weeks of no lung function in a patient who is not a transplant candi-
date is often considered futile. They note, however, that there have been cases of
lung recovery after 50 days of ECMO. Therefore, discontinuation of ECMO for
futility should be undertaken in a case-by-case basis [2, 3].
As discussed in the section on initiating ECMO, the team should have done con-
siderable work up-front to prepare a patient and surrogates for the possibility of
terminating ECMO. In one study, decisions to withdraw ECMO were influenced by
patient comorbidities, patient wishes, and the etiology of respiratory failure (such as
pulmonary fibrosis). Unfortunately, these investigators found that survey respon-
dents reported low rates of engaging in shared decision-making in patients receiving
ECMO [11]. Understanding and applying the principles of shared decision-making,
surrogate decision-making, and substituted judgment are imperative to respecting a
patient’s wishes when they cannot speak for themselves and can also help lessen the
burden that families feel in making difficult end-of-life decisions. Treatment plans
should align not only with patients’ preferences and values but also with prognosis
and appropriateness of the care [10]. There is no ethical obligation to provide an
intervention for which there is no reasonable chance of achieving its aim. Therefore,
rather than a shared decision-making approach of when to discontinue ECMO, the
298 E. Sonntag and M. Pahuja
providers should use their clinical judgment to determine when treatment goals are
unlikely to be met and recommend discontinuation [1]. This means the provider will
need to feel comfortable making recommendations in the space of uncertainty as is
common in the application of ECMO.
The best-case scenario is that the family assents to discontinuation of ECMO at
the health care team’s recommendation. Of course, even when best practices are
used for patient selection and expectation-setting, withdrawal of any life-sustaining
therapy is often challenging for the patient, family, and providers. In many cases
surrogates may request to continue ECMO, even after the team has determined it is
no longer a bridge to recovery or transplant. In fact, in a single-center study, one
ECMO program described disagreement about continuation of ECMO as the most
common ethical issue faced in ECMO care [4].
This ethical dilemma occurs when autonomy and beneficence/nonmaleficence
are in conflict. Autonomy is the basic ethical principle that supports a patient’s right
to choose and this should be upheld whenever possible. Autonomy, however, is not
absolute. It gives a patient a right to choose between offered medical therapies and
to refuse therapies, but not to demand therapy the health care team feels is inappro-
priate, harmful, or futile [2]. Furthermore, respect for patient autonomy does not
require burdening a patient or family with the entire decision-making process.
However, a fully paternalistic approach is not appropriate either. Some centers have
patients or their surrogates sign documents stating that ECMO withdrawal decisions
will be in the hands of the providers. This may help set expectations and avoid dis-
putes at the time of ECMO discontinuation; however, it is not a perfect solution. It
is a provider’s duty to build trust and understand a patient’s values and preferences,
so that end-of-life decisions can be framed by both clinical realism and the patient’s
best interest. These types of decisions can take an emotional toll on surrogates [2].
Surrogate decision-makers for ICU patients have a PTSD rate of 33%. That rises to
82% when making end-of-life decisions [7].
Additionally, disagreements about continuation of ECMO do not always occur
between surrogates and health care providers, but also amongst members of the
team [4]. For example, respiratory therapists, ECMO technicians, and nurses may
have different perceptions of quality of life or suffering based on their time at the
bedside. It is incredibly important that throughout ECMO initiation, continuation,
and withdrawal, all members of the health care team are communicating and on the
same page. This not only helps to create consistent messaging to patients and sur-
rogates, but also helps avoid moral distress amongst team members.
If conflicting views cannot be resolved to allow ECMO withdrawal, fallback
approaches include “do not escalate” (for complications or subsequent organ fail-
ures) or awaiting ECMO circuit or membrane failure [2]. Although it has been
established that withdrawal and withholding of life-sustaining therapy are ethically
equivalent, they can often feel emotionally different: some clinicians equate with-
drawal of support as ending a life. For this reason, it may be easier for some to await
complications, rather than withdraw the circuit. Ethics consultation may be helpful
in such circumstances.
17 Ethical Challenges in Extracorporeal Membrane Oxygenation 299
Ethics Consultation
Health Disparities
ECMO has changed the landscape of how we care for patients with refractory respi-
ratory failure and brought with it new ethical challenges. We have detailed in this
chapter how a thorough understanding of medical ethics, health care disparities, and
communication skills may help a physician navigate the ethical dilemmas that arise
in ECMO care [2, 10]. This education and understanding of ethics will continue to
be imperative as the number of ECMO programs and treated patients increase over
time [20]. However, residency and fellowship programs who train in ECMO use
currently lack a formal medical ethics curriculum. Integrating the four basic prin-
ciples of beneficence, nonmaleficence, autonomy, and justice and promulgating cur-
ricula on informed consent, shared decision-making, surrogate decision-making,
and end-of-life care could help physicians prepare for the common ethical chal-
lenges they are sure to encounter in practice [21].
References
1. Stephens AL, Bruce CR. Setting expectations for ECMO: improving communication between
clinical teams and decision makers. Methodist Debakey Cardiovasc J. 2018;14:120.
2. Ramanathan K, Cove ME, Caleb MG, Teoh KLK, Maclaren G. Ethical dilemmas of adult
ECMO: emerging conceptual challenges. J Cardiothorac Vasc Anesth. 2015;29:229–33.
3. Extracorporeal Life Support Organization – ECMO and ECLS Extracorporeal Life Support
Organization (ELSO). In: Extracorporeal Life Support Organization (ELSO) World’s Largest
Registry of ECMO Runs and ECLS Centers. https://www.elso.org/. Accessed 21 Nov 2021.
4. Courtwright AM, Robinson EM, Feins K, Carr-Loveland J, Donahue V, Roy N, McCannon
J. Ethics committee consultation and extracorporeal membrane oxygenation. Ann Am Thorac
Soc. 2016;13:1553–8.
5. Wirpsa MJ, Carabini LM, Neely KJ, Kroll C, Wocial LD. Mitigating ethical conflict and moral
distress in the care of patients on ECMO: impact of an automatic ethics consultation protocol.
J Med Ethics. 2021; https://doi.org/10.1136/medethics-2020-106881.
6. Chen DT, Shepherd LL, Mohrmann ME. Substituted judgment. J Gen Intern Med. 2008;24:145.
7. Azoulay E, Pochard F, Kentish-Barnes N, et al. Risk of post-traumatic stress symptoms in fam-
ily members of intensive care unit patients. Am J Respir Crit Care Med. 2005;171:987–94.
302 E. Sonntag and M. Pahuja
8. Quill TE, Holloway R. Time-limited trials near the end of life. JAMA. 2011;306:1483.
9. Graffagnino JP, Avant LC, Calkins BC, Swetz KM. Home therapies in advanced heart failure:
inotropes and diuretics. Curr Heart Fail Rep. 2020;17:314–23.
10. Chang DW, Neville TH, Parrish J, et al. Evaluation of time-limited trials among critically
ill patients with advanced medical illnesses and reduction of nonbeneficial ICU treatments.
JAMA Intern Med. 2021;181:786.
11. Abrams D, Pham T, Burns KE, et al. Practice patterns and ethical considerations in the man-
agement of venovenous extracorporeal membrane oxygenation patients. Crit Care Med.
2019;47:1346–55.
12. Lopez L, Hart LH, Katz MH. Racial and ethnic health disparities related to covid-19.
JAMA. 2021;325:719.
13. Havranek EP, Mujahid MS, Barr DA, et al; American Heart Association Council on quality of
care and outcomes research, council on epidemiology and prevention, council on cardiovas-
cular and stroke nursing, council on lifestyle and cardiometabolic health, and stroke council.
Social determinants of risk and outcomes for cardiovascular disease: a scientific statement
from the American Heart Association Circulation 2015;132(9):873–898.
14. Chan T, Barrett CS, Tjoeng YL, Wilkes J, Bratton SL, Thiagarajan RR. Racial variations
in extracorporeal membrane oxygenation use following congenital heart surgery. J Thorac
Cardiovasc Surg. 2018;156:306–15.
15. Chan T, Di Gennaro J, Farris RW, Radman M, McMullan DM. Racial and ethnic variation in
pediatric cardiac extracorporeal life support survival. Crit Care Med. 2017;45:670–8.
16. Scheunemann LP, White DB. The ethics and reality of rationing in medicine. Chest.
2011;140:1625–32.
17. Emanuel EJ, Persad G, Upshur R, Thome B, Parker M, Glickman A, Zhang C, Boyle C, Smith
M, Phillips JP. Fair allocation of scarce medical resources in the time of covid-19. N Engl J
Med. 2020;382:2049–55.
18. Bartlett RH, Ogino MT, Brodie D, et al. Initial elso guidance document: ECMO for COVID-19
patients with severe cardiopulmonary failure. ASAIO J. 2020;66:472–4.
19. Miller WD, Peek ME, Parker WF. Scarce resource allocation scores threaten to exacerbate
racial disparities in health care. Chest. 2020;158:1332–4.
20. Enumah ZO, Carrese J, Choi CW. The ethics of extracorporeal membrane oxygenation: revis-
iting the principles of clinical bioethics. Ann Thorac Surg. 2021;112:61–6.
21. Sonntag E, Shah K, Katz J. Educating resident and fellow physicians on the ethics of mechani-
cal circulatory support. AMA J Ethics. 2019; https://doi.org/10.1001/amajethics.2019.407.
22. Jonsen AR, Siegler M, Winslade WJ, Mishra R. Clinical ethics: a practical approach to ethical
decisions in clinical medicine. New York: McGraw Hill; 2022.
Index
© The Editor(s) (if applicable) and The Author(s), under exclusive license to 303
Springer Nature Switzerland AG 2022
G. A. Schmidt (ed.), Extracorporeal Membrane Oxygenation for Adults,
Respiratory Medicine, https://doi.org/10.1007/978-3-031-05299-6
304 Index
S
Saturable oxygen carrier (haemoglobin), 23 V
Scarce resource allocation, 300, 301 Vasoactive drug requirements, 224
Sedation and mechanical ventilation, 206 Vasoactive management for RV failure,
Sedation management, 270 154
Seldinger technique, 103 Velocity pumps, 42, 43
Severe acute asthma, 197, 199 Veno-arterial (VA) configuration, 105–106
Severe adult respiratory distress Veno-arterial (VA) ECLS, circuit blood
syndrome, 259 flow, 275
310 Index