Академический Документы
Профессиональный Документы
Культура Документы
Extracorporeal
Life Support for
Adults
Respiratory Medicine
Series Editor :
Sharon I.S. Rounds
vii
viii Preface
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 eld; and our publisher at Springer-Link who pushed
for this important book. Finally, I recognize all those who do the hard work: the
nurses, perfusionists, and therapists who dedicate their lives to the critically ill. This
is an exciting time, ripe with change and opportunity. We seek a path forward for the
benet of all our patients.
ix
x Contents
Darryl Abrams, MD Division of Pulmonary, Allergy and Critical Care, New York-
Presbyterian Hospital/Columbia University Medical Center, New York, NY, USA
Cara L. Agerstrand, MD Division of Pulmonary, Allergy, and Critical Care
Medicine, Department of Medicine, Columbia University College of Physicians
and Surgeons/New York-Presbyterian Hospital, New York, NY, USA
Matthew Bacchetta, MD, MBA, MA Division of Thoracic Surgery, New York-
Presbyterian Hospital/Columbia University Medical Center, New York, NY, USA
Cherry Ballard-Croft, PhD Division of Cardiothoracic Surgery, Department of
Surgery, University of Kentucky College of Medicine, Lexington, KY, USA
Giacomo Bellani, MD, PhD Department of Health Sciences, University of
Milano-Bicocca, Monza, Italy
Department of Anesthesia and Critical Care, San Gerardo Hospital and Milano-
Bicocca University, Monza, Italy
Matthew J. Brain, MBBS (Hons), FRACP, FCICM, DDU School of Public
Health and Preventive Medicine, Monash University, Malvern East, VIC, Australia
The Alfred Intensive Care Unit, Melbourne, VIC, Australia
Department of Medicine, Launceston General Hospital, Launceston, TAS, Australia
Nicolas Brchot, MD, PhD Service de Ranimation Mdicale, Hospital
PitiSalptrire, Paris, France
Daniel Brodie, MD Division of Pulmonary, Allergy and Critical Care, New York-
Presbyterian Hospital/Columbia University Medical Center, New York, NY, USA
Warwick W. Butt, FRACP, FCICM ICU RCH, Department of Paediatrics UoM,
Clinical Sciences Theme MCRI, Royal Childrens Hospital, Melbourne, VIC, Australia
Paediatric Intensive Care Unit, Parkville, VIC, Australia
xi
xii Contributors
Introduction
Fig. 1.1 Schematic of ECMO configurations, circles represent pumps, 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 (ECCO2R) is commonly performed with a VV-ECMO configuration,
usually with a single dual-lumen catheter. Intravascular membrane oxygenators have also been
developed [1] but are not currently in clinical use
1 Physiology ofExtracorporeal Life Support (ECLS) 3
required to prescribe, manage and wean this support and recognise evolving
complications of the therapy. Although designed primarily to replace cardiorespira-
tory function, the interaction of ECLS with several other physiologic systems must
be 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. The systemic inflammatory response syndrome
(SIRS) may arise from the underlying disease or as a reaction to the non-biological
material of the ECLS circuit. The metabolic response to critical illness has direct
implications for oxygenation and CO2 removal, as well as nutritional supplementation
to facilitate later weaning.
In order to comprehensively understand ECLS and its effects on human physiology,
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 for tissue-specific behaviour such as prefer-
ential basal uptake by the brain, concentration-dependent uptake by the liver,
concentration-sensing by the insulin-secreting pancreatic -cells and insulin-
dependent uptake in skeletal muscle and fat [2].
Intracellular glucose is rapidly phosphorylated in the cytosol by hexokinases,
after which it becomes the primary substrate for energy production or biosynthetic
reactions including glycogen storage (Fig.1.2). Utilisable intracellular energy is
Fig. 1.2 Key intermediates in intracellular metabolism: After entering cells, glucose is phosphor-
ylated (-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 tricarboxylic acid (TCA) cycle and oxidative phosphorylation only occur in mito-
chondria 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 [3]. Different amino
acids can enter or be synthesised from the pathway 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, ketone body production when insufficient oxaloac-
etate exists for acetyl groups to enter the TCA cycle. Ketone bodies are produced predominantly
in the liver from fatty acid breakdown and constitute a glucose-sparing fuel for the brain and heart.
Humoral promoters and inhibitors of reactions are shown
1 Physiology ofExtracorporeal Life Support (ECLS) 5
stored in the phosphate bonds of adenosine triphosphate (ATP) and it is the breaking
of chemical bonds within glucose that powers ATP regeneration from adenosine
diphosphate (ADP) and inorganic phosphate (Pi).
Glycolysis describes the fracturing of the six-carbon glucose molecule into
two three-carbon pyruvate molecules with the net generation of two ATP molecules.
For glycolysis to continue, oxidative power (NAD+ concentration) must be continu-
ally restored. Under anaerobic conditions, this occurs by conversion of pyruvate to
lactate. Under aerobic conditions, pyruvate loses a carbon dioxide molecule to yield
acetyl coenzyme-A.This two-carbon acetyl group can be incorporated into fatty
acids for storage or can enter the tricarboxylic acid (TCA) cycle to complete the
chemical breakdown of glucose to CO2. This yields 38 ATP molecules, significantly
more than glycolysis, but generates NADH in such quantities that a powerful elec-
tron acceptor is required for efficient restoration of NAD+ so that the cycle can
continue. This electron acceptor is oxygen.
Oxidative phosphorylation describes the process of restoring NAD+ to perpetuate
the TCA cycle. Although oxygen is utilised as an electron acceptor in many enzyme
systems, its highest consumption is in this process. Oxidative phosphorylation
occurs in the inner mitochondrial matrix and it is to this intracellular destination that
oxygen must diffuse in sufficient quantities to sustain ATP generation for normal
cellular processes.
When oxygen is not available in sufficient quantities, ATP generation from ADP
can only continue in the cytosol by glycolysis. This process is inefficient, as not
only is less ATP produced but the resulting lactic acid is not as readily cleared from
the tissues or body as carbon dioxide. Lactic acid is thus a marker of glycolysis
activity in a hypoxic environment and usually indicates inadequate tissue perfusion
or global hypoxemia of the organism.
The respiratory quotient (RQ) describes the ratio of the amount of carbon dioxide
( VCO 2 ) produced per unit time to the amount of oxygen consumed ( VO 2 ).
VCO2
RQ = (1.1)
VO
2
The respiratory quotient depends on the sources of fuel being used. For glucose
metabolism, the six carbon atoms result in production of six molecules of carbon
dioxide while consuming six molecules of oxygen; it thus has a respiratory quotient
of 1. The reactions for oxidation of some amino acids and fatty acids (lipolysis)
produce less CO2 (by not including the pyruvate to acetyl-CoA reaction, Fig.1.2)
and hence have respiratory quotients of less than 1.
In contrast, each acetyl-CoA molecule utilised for fatty acid synthesis (lipogenesis)
results in production of a molecule of CO2 (from pyruvate to acetyl-CoA, Fig.1.2)
6 M.J. Brain et al.
Key hormones coordinate the response to nutrition supply and stress. Insulin marks
the fed state, promoting hepatic glucose uptake, glycogen and amino acid synthesis
and conversion of acetyl-CoA to free-fatty acid production while in the peripheral
tissues stimulating myocyte synthesis of contractile elements and adipocyte
triglyceride deposition.
Glucagon is secreted by pancreatic -cells in response to low blood glucose
levels and promotes glycogen breakdown and conversion of amino acids
1 Physiology ofExtracorporeal Life Support (ECLS) 7
(from muscle breakdown), lactate and glycerol. Glycerol results from adipocyte
triglyceride metabolism and the released free fatty acids are converted to ketone
bodies by the liver for use as a secondary fuel source when glucose is scarce.
The catecholamines epinephrine and norepinephrine are released in response to
physiologic stress. By increasing intracellular cyclic AMP, they promote glycoge-
nolysis in muscles and catabolism of protein to release amino acids. In the liver,
epinephrine promotes gluconeogenesis, glycogenolysis and inhibits glycolysis.
These responses result in the hyperglycaemia that characterises the stress state
and is exacerbated by exogenous administration of catecholamines and glucose.
The adverse effects of hyperglycaemia include osmotic diuresis, fat deposition in
the liver and impaired immune function.
The metabolic profile of patients receiving ECLS is typical of the stressed state
but the differences between this and the starvation state are important. In starvation
there is an overall decrease in energy expenditure with maximal use of triglycerides
and ketoacids promoting conservation of muscle bulk. The brain, heart and renal
cortex adapt to utilising ketoacids for significant proportions of their metabolic
requirements. In contrast, the chronic stressed state is characterised by increased
resting energy expenditure, accelerated catabolism of lean body massprimarily
amino acids from muscle catabolism [3]and the immunosuppressive effects of
hyperglycaemia and persistently elevated humoral mediators, including catechol-
amines and cortisol.
In those requiring ECLS, particularly those needing prolonged periods of heavy
sedation, the combination of muscle catabolism, disuse atrophy, critical illness
myopathy and myopathy associated with muscle relaxants can result in profound
weakness. The respiratory musculature is not spared from this process, with the
result being prolonged weaning, a requirement for tracheostomy and the risk of
secondary infection.
Erythrocyte Metabolism
Being a specialised organ for oxygen transport, erythrocytes are nearly 90%
haemoglobin-by-weight, with very few other organelles. Nevertheless, they require
an ongoing energy source to maintain membrane integrity, cytoskeleton structure,
intracellular electrolyte and osmotic equilibrium and to keep the iron 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. However,
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 [8].
8 M.J. Brain et al.
Mitochondria can couple ATP production to NADH oxidation only if sufficient oxy-
gen exists in the environment of cells. Similarly, carbon dioxide diffuses from the
mitochondria, through intracellular membranes, and away from the cell. The flux of
oxygen into the environment of cells and the reverse movement of carbon dioxide
can be divided into two components:
1 . Diffusion of gas molecules into and between liquid phases
2. The carriage of oxygen and carbon dioxide in blood
When considering pulmonary gas exchange a third component must be considered:
the convective transport of the gas to the alveolar epithelium. However, exposure of
the extracorporeal membrane to fresh gas flow is somewhat simpler in ECLS and
will be considered later in the context of carbon dioxide transport.
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. (1.9) are demonstrated with being the porositythe
area of membrane occupied by gas, 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 the constric-
tivitythe resistance to passage
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 [10]. Most systems direct fresh gas
through the lumen of the hollow fibres, while blood flows between the tubules
(termed extra-capillary flow). The reverse configuration is also sometimes utilised,
however, and overall characteristics such as total surface area for gas exchange,
resistance to flow and trauma to formed blood components will be determined by
factors such as membrane material, fibre diameter and length, fibre density and the
velocity of the blood [11].
10 M.J. Brain et al.
Heat loss over the extracorporeal circuit into the environment can be substantial
and heat exchangers are commonly incorporated into the oxygenator. Figure1.3
demonstrates one such design where the microporous membrane fibres are laid
perpendicularly to impermeable capillaries that circulate heated water, allowing
heat to be regulated.
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 intermolecular
attraction:
P = nRT (1.2)
V
The universal gas equation: P=the partial pressure, n=number of molecules of gas
measured in moles, T is temperature in degrees Kelvin and V is volume of the con-
tainer in litres. R is the ideal gas constant which in SI units is 8.314JK1mol1 or
in conventional units: 62.36mmHgK1mol1
At a constant temperature, Eq. (1.2) simplifies to P n / V . Concentration is
defined as moles/unit volume, i.e. n/V; hence pressure is proportional to gas concen-
tration, i.e. the greater the number of gaseous molecules in a given volume, the
more force those gas molecules will exert on the walls of the container. The physical
reaction of dissolving in solution is also proportional to the partial pressure of the
gas above the solution, so that for oxygen dissolution [12]:
K Forward
O2 ( gas )
O2 ( dissolved ) (1.3)
K Reverse
[O2 ](gas) K Forward = [O2 ](dissolved ) K Reverse (1.4)
[O2 ](dissolved ) = SC [O2 ](gas) (1.5)
The rate constant KForward in Eq. (1.4) describes the proportion of oxygen gas that
dissolves per unit time, while KReverse describes the proportion of the dissolved con-
centration that leaves the solution to the gas phase. When the system described
in Eq. (1.4) is at thermodynamic equilibrium, the concentrations in the gas and
liquid phases are stable and the constants may then be combined, resulting in
Eq.(1.5), also known as Henrys Law. SC is the Bunsen solubility coefficient where
SC=KForward/KReverse and is gas- and solvent-specific. SC is affected by other dissolved
1 Physiology ofExtracorporeal Life Support (ECLS) 11
solutes and falls with increasing temperature (i.e. KForward becomes smaller and
KReverse larger).
Due to difficulties in measuring the molar concentration of oxygen compared to
measuring a volume of 100% oxygen at standard conditions (STPD: 0C, 760mmHg,
dry gas), it is customary to report oxygen content in mLdL1. Under these conditions,
oxygen approximates an ideal gas such that 6.021023 gas molecules (i.e. 1mol)
occupy 22.414L at 0C.Quantification of human oxygen consumption is performed
using STPD rather than BTPS (body temperature and pressure, saturated: Eq. (1.7))
[13] as water vapour in the latter partially condenses with increasing pressure. This
vapour results in a significant deviation from an ideal gas and invalidates the relation-
ship between the number of molecules and volume defined in Eq. (1.2).
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 materials, a gasblood interface is usually prevented by forma-
tion of a biofilm comprising adsorbed blood proteins that forms after a short period
of operation. These factors impose a time constraint in 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
Ficks Law of Diffusion [15]:
DC
J = -D A (1.8)
Dl
This expression says the flux (J, mmols1) of a solute over the thickness of a mem-
brane (l, cm) is proportional to the diffusivity coefficient, D, the concentration
gradient across the membrane (C, mmolmL1 or mmolcm3) 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 [15]. This universal statement of
mass transport is applicable not only to the extracorporeal oxygenator, but also of
oxygen moving from plasma to interstitial fluid and into cells.
The diffusivity coefficient, D, is expressed as area over time (cm2s1) and describes
a unique constant of the gas, barrier, and solution under steady state conditions.
Higher numbers represent greater diffusibility with coefficients in gases being
1 Physiology ofExtracorporeal Life Support (ECLS) 13
DEff = ( 2.13 - 0.0092 %Hct ) 10 -5 ( cm 2 -1
s ) (1.10)
By combining Eq. (1.5) (Henrys Law) for solute concentration and Ficks
Law (Eq.1.8) an equation for diffusive membrane flux can be derived where k
is the permeability constantthe product of the solubility coefficient (SC,
mmolL1mmHg1) already defined and effective diffusivity (DEff, cm2s1) having
the units molcm1s1mmHg1 [20]:
PO2 ( Gas) - PO2 ( Plasma )
J = - k O2 A (1.11)
Dl
14 M.J. Brain et al.
J (
PO2 ( Gas) - PO2 ( Plasma ) )
( mmols -1
cm -2 ) A
=
RTotal
(1.12)
Dl
where RTotal =
- kO 2
From Eq. (1.12) it is apparent that maintaining the partial pressure gradient between
the gas phase and the blood phase is important to maximise oxygen flux. Maintenance
of the local pressure gradient at any point along a hollow fibre is influenced by three
design factors and one operational factor: the time the blood is exposed to the mem-
brane, relative flow direction, local turbulence and haematocrit.
For a static liquid below a static gas, the rate of diffusion will decrease exponentially
until equilibrium when the conditions of Eq. (1.5) are met, i.e. the rate of gas dis-
solving into the liquid is equal to the rate of molecules leaving the liquid (Fig.1.4a).
It is clear from this figure that too short an exposure time will result in submaximal
oxygenation.
The more complicated relationship when haemoglobin is present is displayed
in Fig.1.4b. Here the oxygen content of the blood displays a plateau due to the
oxygenhaemoglobin dissociation curve (discussed below).
In ECMO, the time of blood exposure to the membrane is proportional to the
length of membrane/hollow fibre traversed and inversely proportional to the blood
flow rate. As the length of microtubules adds to resistance to blood flow, determin-
ing the optimal length for oxygen flux over the physiological range of blood flows
likely to be encountered is an important design parameter. With current hollow fibre
oxygenators this length is around 4cm for adequate oxygenation, with shorter
lengths needed for carbon dioxide removal (see ECCO2R below).
1 Physiology ofExtracorporeal Life Support (ECLS) 15
a b
14.00
Partial Pressure vs Time for a static gas/liquid O2 Content vs Membrane Exposure Time
12.00
O2 Content mL/dL
8.00
Oxygen Content (ml/dL) for Hb 10 g/dL pH
6.00 7.44 Temp 37.5 & BE 0
Dissolved O2 in solution
4.00
2.00
0.00
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: (a) Depicts partial pressures
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. (b) 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 oxygenators. Data
derived from Katoh [18]
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 blood stream.
turbulent flow vortices more effectively purges tubule contents than laminar flow
(discussed below) and brings erythrocytes into closer proximity to gas exchange
membranes, increasing the local concentration gradient.
The total resistance to mass solute movement in Eq. (1.12), RTotal, is the sum of
component resistances, which can be divided into gas phase resistance (RG), mem-
brane resistance RM, and blood side resistance, RB: RTotal=RG+RM+RB. Of these RG
is negligible and the factors influencing RM have been extensively discussed. The most
variable 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.
(1.12)) that encompasses the complex microscopic geometry of a membrane and its
pores, the latter being incorporated into DEff as discussed [16]. Furthermore, it can-
not be guaranteed that any oxygenator design utilises the entire membrane area
evenly. Thus, while flux per unit area (J/A) is a useful description of isolated mem-
brane performance, oxygenators are better characterised by their total flux by incor-
porating area into the equation for resistance:
(P - PO2 ( Plasma ) )
( mL min ) J
-1 O2 ( Gas )
Oxygenator =
RTotal
(1.13)
Dl
where RTotal =
- kO 2 A
Although lacking the precision of Eq. (1.12) in defining properties of the membrane,
Eq. (1.13) can be incorporated into monitoring gas exchange efficiency of an indi-
vidual oxygenator over time and will be discussed after oxygen carriage is consid-
ered. The value of JOxygenator is usually reported in oxygenator product specification
sheets at varying blood flows.
water flux
K UF =
transmembrane pressure
(1.14)
QF
=
PBlood - PGas
QF (mLmin ) is termed the ultrafiltration rate and describes the appearance of fluid
1
within the gas containing hollow fibres, while the terms for pressure (PBlood and Pgas)
describe the heights of a fluid column relative to atmospheric pressure in each com-
partment. Many of the factors already described for resistance to diffusion will be
contained in the ultrafiltration coefficient and will not be discussed further. The ultra-
filtrate represents a homogenous fluid that will contain dissolved solutes from plasma
proportional to the size of membrane pores, which are typically smaller than 1m.
A major drawback of early microporous membranes was significant plasma leak-
age, as the open porous structure allowed significant ultrafiltration. This has been
significantly alleviated by newer closed-fibre membranes; however, some water
flux still occurs under normal operating conditions. The water then 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 [10].
Oxygen Transport
Oxygen Carriage
80%
70% 15.00
Blood Oxygen Content ml/dL
Hemoglobin Saturation %
60%
50%
10.00
40%
0% 0.00
0 20 40 60 80 100 120
Partial Pressure Oxygen (mmHg)
Fig. 1.5 The oxygenhaemoglobin dissociation curve as calculated by the Thomas modification
of the Kelman Eq. (1.9). Also shown is the oxygen content for haemoglobin concentration of
10gdL1 after applying Eq. (1.15)
1 Physiology ofExtracorporeal Life Support (ECLS) 19
(100% saturation) the maximum oxygen carrying capacity is 1.39mL per gram of
haemoglobin in adults [13] or 1.312mL per gram of foetal haemoglobin [14]. Thus,
the total oxygen content of adult arterial blood (CaO2) can be described by:
(
CaO2 = Hb SaO2 1.39 + 0.0031 PaO2 mLdL-1 ) (1.15)
The total oxygen content can then be multiplied by blood flow (cardiac output) to
give oxygen delivery, DO 2 which has the units of flux: mLmin1. Note the scaling
factor of 10 to convert the units of oxygen content to mLL1.
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 ferrous elements in deeper crypts and facilitating release of oxy-
gen from haemoglobin. Similarly, increasing temperature and increasing hydrogen
ion concentrations stabilise haemoglobin in the deoxygenated state. Similar to oxy-
gen, carbon dioxide binding is reversible and in compartments with a low CO2 con-
centration 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 will be discussed later in the chapter.
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 [9]. This has the most significance in transfused blood where, after 2 weeks
of storage, 2,3-DPG levels become negligible. After transfusion of this blood DPG
levels do not return to normal until nearly 48h [22]. In the absence of 2,3-DPG the
affinity for oxygen is increased resulting in less oxygen delivery at the same periph-
eral PO2. As will be seen, this is only likely to be a factor in the most severe oxygen-
ation problems and the benefit of increased oxygen binding sites from transfusion
will outweigh the transiently lower delivery (see Figs.1.8 and 1.9).
shift the curve to the right (i.e. to a lower affinity state) are local tissue metabolism
and hence local oxygen consumption (Fig.1.5).
Several equations exist to model the normal oxygenhaemoglobin dissociation
curve. One of the most informative is the Thomas modification [23] of the Kelman
[24] equation and its inverse [25], which calculate the haemoglobin saturation for
any partial pressure of oxygen and allow for shifts of the curve due to temperature,
[H+] and carbon dioxide.
For convenience, the oxygenhaemoglobin dissociation curve is frequently
described by the partial pressure at which 50% of haemoglobin is saturated. A nor-
mal P50 for arterial blood is 26.3mmHg. Values higher than this describe a right-
shifted curve, i.e. a haemoglobin with lower affinity for oxygen. Two curves of
haemoglobin saturation are demonstrated in Fig.1.7, 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 differ-
ence in haemoglobin saturation is minimal. In contrast, there is a significant differ-
ence in the saturation of Hb at a partial pressure of 40mmHg commonly found in
venous blood.
The implications of the oxygen Hb dissociation curve become clearer when
oxygen content is also plotted on the same chart (Fig.1.5). At haemoglobin of
10gdL1 the oxygen content at a partial pressure of 100mmHg is 13mLdL1 and
minimally affected by the arteriovenous pH difference. However, at an oxygen
partial pressure of 40mmHg, a 1mLdL1 difference becomes apparent between
the two content curves, being 10mLdL1 at pH of 7.44 and 9mLdL1 at pH of 7.32.
In the tissues a drop in content can only occur if the oxygen is utilised by metabo-
lism, so this right-shifting of the curve as the products of metabolism acidify
capillary blood serves to bolster the partial pressure gradient for diffusion from
capillary to cell.
It should be noted that right-shifted oxygenhaemoglobin dissociation curves,
while advantageous for unloading oxygen in acidotic tissues, may be counterpro-
ductive at sites of oxygen uptake if abnormally low alveolar partial pressures exist.
Inspection of Fig.1.5 reveals that if oxygen uptake were to occur at a partial pressure
of 60mmHg, a right shifted curve (pH7.32) will carry 0.5mLdL1 less oxygen.
This gap widens if uptake occurs at even lower partial pressures.
As demonstrated in Table1.1, the average of the venous and arterial blood satu-
rations approximates the complete solution suitably when the flows are similar, but
overestimates saturations when the volumetric flow differs. The approximation
worsens if the venous oxygen tension is reduced further. In contrast, taking the
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 may
result in suboptimal oxygen delivery.
Having covered oxygen carriage and mixing blood with differing oxygen content,
veno-venous (VV) extracorporeal membrane oxygenation can now be considered.
As blood is accessed and returned to the venous system/right atrium (Fig.1.6),
22 M.J. Brain et al.
Fig. 1.6 Schematic of a VV-ECMO circuit. It will be assumed that both femoral and internal jugu-
lar access exists so all venous return is modeled with one vessel. A mixture of fully oxygenated and
venous blood enters the right heart and perfuses the pulmonary circulation before the left heart
distributes it systemically (hence end-tidal CO2 will not reflect mixed venous PCO2). Varying with
the cannula position and venous return (cardiac output), some amount of recirculation is very com-
mon, reducing the amount of oxygen delivered. Depending on recirculation, it may not be possible
to sample true mixed venous blood
systemic oxygen delivery remains dependent on cardiac output, making this system
simpler to quantitatively analyse.
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
along the inferior vena cava. This access cannula typically ends around 10cm below
the cavo-atrial junction, while the return cannula has a single terminal orifice in the
right atrium. If higher extracorporeal circuit flows are attempted, the inferior vena
cava may collapse around the multistage cannula, intermittently obstructing flow
and causing the external circuit to shudder, with consequent hypoxia. If these
higher circuit flows are necessary to achieve adequate oxygenation and this negative
access pressure cannot be resolved by giving fluid, a second access cannula may
need to be placed in the internal jugular vein.
Figure 1.6 demonstrates a basic circuit. It can be appreciated 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 supported 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.
1 Physiology ofExtracorporeal Life Support (ECLS) 23
The patient has the following parameters: cardiac output of 4Lmin1, ECMO
flow of 4Lmin1, 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 7Lmin1 or fall to 2.5Lmin1?
Apart from the non-contribution of the lungs, other assumptions include an
oxyhaemoglobin 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 param-
eters are required: the haemoglobin concentration and the patients total oxygen con-
sumption (VO2). Initially, this hypothetical patient has a haemoglobin concentration
of 10gdL1 and is at steady state consuming 250mLmin1 of oxygen with no mark-
ers of tissue hypoxia. Thus, there are four independent variables: the cardiac output,
the ECMO flow rate, the haemoglobin concentration and the target VO2.
While providing circulating oxygen 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 said to be 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, oliguria and ris-
ing lactate as anaerobic metabolism ensues.
Figure1.7a suggests that under these conditions, increasing the cardiac output
from 4 to 7Lmin1 will cause a drop in the arterial saturations from 99 to 84%.
In contrast dropping the cardiac output to 3Lmin1 will not affect the arterial satu-
ration but will cause the central mixed venous saturation to fall from 52 to 25%.
The equations required to generate this model are the content and delivery equa-
tions (Eqs.1.15 and 1.16) 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). It should be noted that
in VV-ECMO the pulmonary artery is not the correct sampling site for mixed venous
saturation. Instead, the pre-oxygenator blood is the best approximation (but depend-
ing on the vessels cannulated may not include superior vena cava flow).
The methods for mixing blood streams (Table1.1) are used to determine the pulmo-
nary artery oxygen content and saturations. In the absence of any lung function this is
modeled as the systemic arterial oxygen delivery. 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 extractionthis determines the venous oxygen
flux to the right heart. Example values from Fig.1.7a are shown in Table1.2.
To explain the fall in arterial saturations with increasing cardiac output it should be
appreciated that in this case the required VO2 did not change. What did change was
the total venous return, which increased by 75% with the increased cardiac output.
24 M.J. Brain et al.
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 ECMO flow conditions and haemoglo-
bin. See text for discussion. (a) VV-ECMO blood flow 5L.min1, body oxygen consumption 250ml.
min1 and hemoglobin 10g.dl1. (b) VV-ECMO blood flow 3L.min1, body oxygen consumption
250ml.min1 and hemoglobin 10g.dl1. (c) VV-ECMO blood flow 4L.min1, body oxygen consump-
tion increasing with cardiac output, hemoglobin 10g.dl1. (d) VV-ECMO blood flow 4L.min1, body
oxygen consumption 250ml.min1 and hemoglobin 7g.dl1
The ECMO flow remained at 4Lmin1 so the ECMO shunt, i.e. the fraction of
deoxygenated venous blood in the right ventricle increased from 3 to 43%. The
shunt is calculated as (Cardiac outputOxygenated Blood flow)/Cardiac output
where oxygenated blood flow equals the cardiac output if the cardiac output is less
than ECMO blood flow. The oxygenated blood volumetric flow will equal ECMO
set flow minus any recirculation when cardiac output is greater than ECMO flow.
In summary, the resulting arterial saturation is analogous to the example of mixing
blood streams 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 reality, increases
in cardiac output are likely to be accompanied by an increased VO2 and in this
setting the mixed venous saturation will fall (Fig.1.7c).
Table 1.2 Solutions for Fig.1.7a
Flux of
fully
Oxygenated oxygenated Cardiac Fraction of
blood flow blood output/ Venous O2 venous
ECMO (factors delivered venous flux return (deoxygenated) Total O2 flux CaO2-
set rate recirculation) by ECMO return to heart CvO2 SvO2 blood leaving leaving RV CaO2 SaO2 CvO2 VO2
(Lmin1) (Lmin1) (mLmin1) (Lmin1) (mLmin1) (mLL1) (%) RV (%) (mLmin1) (mLL1) (%) (mLL1) (mLmin1)
4.0 1.5 201.0 1.5 0.0 0.0 0 0 201.0 134.0 100 134.0 201.0
1 Physiology ofExtracorporeal Life Support (ECLS)
4.0 2.5 335.0 2.5 85.0 34.0 25 0 335.0 134.0 100 100.0 250.0
4.0 3.9 522.6 4.0 279.6 69.9 52 3 529.6 132.4 99 62.5 250.0
4.0 4.0 536.0 7.0 500.5 71.5 53 43 750.5 107.2 80 35.7 250.0
Calculations performed with haemoglobin of 10gdL1. Oxygenated blood flow refers to the volumetric flow of oxygenated blood into the right heart. For
calculation purposes oxygen content is expressed in mLL1 rather than mLdL1
25
26 M.J. Brain et al.
To appreciate why the mixed venous saturations fall to the left of Fig.1.7a (where VO2
was held constant) when the cardiac output falls, recirculation between the access
and return cannula must be considered. Recirculation is an additional variable to the
independent variables identified above.
Several factors influence recirculation; however, this model incorporates only
cardiac output: if the heart stops, there will be no venous return and recirculation
might approach 100% (if the great veins do not collapse), i.e. blood will enter the
ECMO circuit from the IVC and return via the right atrium, flowing in a retrograde
fashion back down the IVC, resulting in no systemic oxygenation (amongst other
adverse effects!). Similarly at low cardiac outputs, the blood not ejected will recir-
culate; however, it becomes saturated with oxygen on the first pass and will be
unable to carry more. This leads to an important point: when the ECMO flow falls
below cardiac output, the amount of oxygenated blood ejected by the right heart is
cardiac output-limited. Conversely, at high cardiac outputs and hence high venous
return, recirculation will be minimal as the oxygenated blood will be washed
through the right atrium into the right ventricle.
Under conditions where the cardiac output is less than the ECMO flow, the blood
entering the arterial circulation is fully saturated with oxygen. However, to supply
the required VO2, oxygen extraction from the low output has to increase, and hence,
mixed central venous oxygen content and venous saturation fall.
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 ECMO fraction of cardiac output mecha-
nism described above AND increased 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, so these graphs illustrate the effect of haemorrhage and transfusion. The lower
oxygen-carrying capacity means that at any cardiac output, the oxygen 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 extraction can occur
and the achieved VO2 falls lower than the target VO2under 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 circulatory support.
1 Physiology ofExtracorporeal Life Support (ECLS) 27
Modelling VV-ECMO
VV-ECMO Arterial Saturations at ECMO Flow 4 L/min and VO2 250 ml/min
100.00%
90.00%
80.00%
90.00%-100.00%
Arterial Sat
70.00% 80.00%-90.00%
70.00%-80.00%
60.00%
60.00%-70.00%
50.00% 50.00%-60.00%
40.00%-50.00%
40.00%
1.50
2.50
3.50
4.50
5.50
6.50 Hb 13
Hb 12
Cardiac Output L/min Hb 11
7.50 Hb 10
Hb 9
Hb 8
Hb 7 /dL
8.50 Hb 6 bin g
oglo
Hb 5 Hem
Fig. 1.8 Arterial oxygen saturation vs. haemoglobin and cardiac output in a patient receiving
VV-ECMO.VO2 is held constant
28 M.J. Brain et al.
VV-ECMO Venous Saturations at ECMO Flow 4 L/min and VO2 250 ml/min
100%
90%-100%
90%
80%-90%
80%
70%-80%
70%
60%-70%
60% 50%-60%
Venous Sat
50% 40%-50%
40% 30%-40%
30% 20%-30%
20% 10%-20%
10% 0%-10%
Hb 11
0%
dL
1.50 2.00 2.50 Hb 8 g/
3.00 3.50 4.00
4.50 5.00 5.50 b in
6.00 6.50 7.00 Hb 5 lo
7.50 8.00 8.50 og
Cardiac Output L/min e m
H
Fig. 1.9 Central mixed-venous oxygen saturation vs. haemoglobin and cardiac output in a patient
receiving VV-ECMO.VO2 is held constant
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
In patients with severely reduced lung function, high cardiac output or poor
VV-ECMO circuit flows may not be able to oxygenate an adequate amount of the
cardiac output to achieve normal arterial oxygen saturations. This may be
improved by adding a second inflow cannula, thus allowing increased circuit flows
without venous collapse. However, if this is not technically possible then targeting
a lower arterial saturation may be necessary.
CO2 ( dissolved ) + H 2 O
+ -
H + HCO3 a. Summary Equation
K
[CO2 ] = HCO3- b. Equilibrium Reaction
H
+
(1.18)
HCO3-
pH = pK + log c. Logarithmic Form
[CO2 ]
Where [CO2 ] = 0.0308 PCO2 at 37C. ( Henry s Law )
K now describes the compound equilibrium of multiple reactions and is derived
experimentally. [H2O] is notably excluded from Eq. (1.18b) because it is incorpo-
rated into K. This is permissible because its concentration is proportionally so
much greater than the other molecules that any consumption of H2O in chemical
reactions produces a negligible effect on its total concentration.1
Although less quoted than the logarithmic form (the HendersonHasselbalch
equation), Eq. (1.18b) is most useful in understanding carbon dioxide in body
fluids. For plasma with a [H+] of 3.98108molL1 (pH of 7.4) at 37C, the
apparent K is 8.13107EqL1mmHg1 (pK 6.09) and thus K/[H+] is around
20, i.e. the bicarbonate concentration is 20 times the dissolved carbon dioxide
concentration.
Over the physiologic range K/[H+] varies with temperature and electrolyte bal-
ance. However, it is always greater than 1 and so the total CO2 contained in blood is
largely stored as bicarbonate ions in plasma and erythrocyte water (Fig.1.12).
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 Eq. (1.18a), CO2 is the only independent variable with the
amount in the body being determined by the balance of production and flux out of
the body [12]. The [H+] depends not only on the reactions of carbon dioxide with
water but also on the concentrations of other strong and weak electrolytes with the
final balance being determined by the need to maintain electrical neutrality and the
1
Occasionally the term PCO2 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 Henrys Law without alteration of the
solubility constant (Eq.1.5).
1 Physiology ofExtracorporeal Life Support (ECLS) 33
Fig. 1.11 Dissolved carbon dioxide diffuses across all tissue planes and dissociates in each com-
partment to HCO3 with the ratio of HCO3 to CO2 depending on the [H+]
ionisation constant of water [12, 27]. The simplest summary of this complex
interaction is that bicarbonate behaves as an electrical spacer in the following
equation for electrical neutrality2:
2
Being electrical neutrality, the units are available charge (milli-equivalents per litre mEqL1);
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.
34 M.J. Brain et al.
Na + + K + + 2 Ca 2 + + 2 Mg + + z +
Carbonic Anhydrase
Figure1.12 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 produced after the addition of a strong
acid to a blood sample which shifts Eq. (1.18a) fully to the left [29]. In contrast the
bicarbonate reported on an arterial blood gas sample is calculated from the measured
pH and PCO2 using Eq. (1.18c) and is typically 24mmolL1 lower than total CO2.
1 Physiology ofExtracorporeal Life Support (ECLS) 35
30 35
Oxygenated Blood [CO2]Total (mmol.L-1)
AV Difference (mmol.L-1)
15
Erythrocyte HCO3- 15
(mmol)
10
10
5 Plasma HCO3-(mmol) 5
Dissolved CO2-(mmol)
0 0
10 20 30 40 50 60 70 80
Pco2 (mmHg)
Fig. 1.12 The CO2 dissociation curve for whole blood at 37C and haematocrit of 45%. Upper
lines represent dissociation of total CO2 in mmolL1 in deoxygenated and oxygenated whole
blood; the line AV Difference demonstrates increased carriage by carbamino groups as the oxygen
saturation falls. To convert mmolL1 to mLdL1 multiply by 2.226. Shaded lower areas demon-
strate the amount of CO2 (mmol) in erythrocyte water (325mL per L blood at PCO2 40mmHg) and
plasma water (509mL per L blood at PCO2 40mmHg). Dissolved CO2 is represented across all
water in 1L whole blood (836mL). Carbamino CO2 demonstrates the maximum amount (mmol)
that fully deoxygenated haemoglobin can carry. Summation of the amount of CO2 in mmol (shaded
areas) at any PCO2 divided by whole blood water (836mL) yields the concentration of CO2 per
litre. After [3034]
This difference is due to carbon dioxide carriage on amino acids and particularly
carbamino formation on haemoglobin (Eq.1.17e).
Inspection of Fig.1.12 is the dissociation curve for CO2 at a haematocrit of 45%
and demonstrates the PCO2 increases from 40 to 47mmHg as tissue metabolism
consumes oxygen. In this example the total CO2 rises correspondingly from
19.6mmolL1 (43.5mLdL1) to 22.4mmolL1 (49.8mLdL1) the difference
being due to tissue CO2 production and is proportional to the respiratory quotient
[30]. The proportion of CO2 carried as carbamino groups increases in this process
as oxygen unloading increases the affinity of haemoglobin for carbon dioxide (hae-
moglobin is more basic when deoxygenated, shifting Eq. (1.18b) toward bicarbon-
ate [14]). It should be emphasised that the blood bicarbonate is carried by blood
water which consists of intracellular and extracellular water [31].
36 M.J. Brain et al.
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 250mLmin1 in each system. The resistance to diffusive transport (RTotal) is
identical for the two membranes so that from Eq. (1.12):
(
J RTotal = Area PCO2 ( Plasma ) - PCO2 ( Gas) ) where RTotal =
Dl
- k O2
= Area1 D P1 Oxygenator 1 (1.20)
Area 2
= ( x D 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 (x=2) will double the required trans-
membrane 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. This often results in
hypocapnia and is avoided by reducing the partial pressure gradient of CO2 by either
adding carbon dioxide to the sweep gas or by reducing the sweep gas flow so that
the convection of carbon dioxide away from the membrane is slowed.
ECCO2R refers to the support of hypercapnic respiratory failure, usually in the setting
of acutely decompensated pulmonary disease. Modern systems utilise a scaled
down VV-ECMO circuit utilising significantly smaller vascular access catheters.
While clearing carbon dioxide, the limited membrane area and lower blood flow
significantly reduce the oxygenation that can be achieved from these systems to
clinically negligible effects.
Lower blood flows are made possible by the higher solubility of carbon dioxide and
its more linear dissociation curve across the physiologic range compared to the
sigmoid curve for oxygen saturation of haemoglobin (Fig.1.13). Complete satura-
tion of haemoglobin at a concentration of 10gdL1 with oxygen results in a
1 Physiology ofExtracorporeal Life Support (ECLS) 37
60.00
25.00
50.00
20.00
15.00
30.00
Oxygen Content for Hb 10g/dL
pH 7.44 Temp 37.5 & BE 0
5.00
10.00
0.00 0.00
0 20 40 60 80 100
Partial Pressure Oxygen (mmHg)
The example above assumes no recirculation, however, this is an increased risk with
the integrated vascular access devices used for ECCO2R.Significant recirculation of
blood that had already been completely cleared of CO2 will limit overall removal, and in
38 M.J. Brain et al.
this setting increasing the ECMO flow rate may not influence the membranes expo-
sure to carbon dioxide-containing blood. Optimal placement of vascular access
catheters in vessels with sufficient blood flow past the catheter tip is required to
minimise this phenomenon.
Fig. 1.14 Acute right ventricular (RV) enlargement with left ventricular (LV) compression 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
Oxygenation utilising the low tidal volumes outlined above is further augmented by
maximising the oxygen partial pressure gradient between perfused alveoli and
pulmonary capillary blood to increase diffusive flux. As discussed, each gas in a
40 M.J. Brain et al.
Table 1.3 The proportional Atmospheric pressure (mmHg) 760 760 760
effect on alveolar oxygen
Water vapour (mmHg) 47 47 47
tension of CO2 removal at
increasing FIO2 Calculated inspired oxygen partial pressure (Eq.1.21)
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 ECCO2R
PaCO2 (mmHg) 40 40 40
PAO2 (mmHg) 106 390 536
Proportional effect of CO2 removal on PAO2
% Change 125 18 12
1 Physiology ofExtracorporeal Life Support (ECLS) 41
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.
These pumps consist of a turbine design that accelerates blood in the line of flow,
typically operating at even higher speeds (up to 9000rpm). Currently they are only
implemented in ventricular assist devices and, comparable 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 ECLS circuits this dis-
sociation between pump speed and flow has resulted in Doppler monitoring of the
velocity of blood in the tubing. In implanted devices (ventricular assist devices), it
is not currently possible to provide a long term accurate flow measurement system
so flow is calculated from the rotation speed. Accurate haematocrit measurements
are important in determining the actual flow, as will be discussed. However, in prac-
tice, exact determination of flow may be less important than clinical evidence of
adequate forward output (perfusion, exercise tolerance) and ventricular unloading
(pulmonary congestion), while echocardiography can demonstrate ventricular
collapse (Fig.1.20).
1 Physiology ofExtracorporeal Life Support (ECLS) 43
The basic hydraulic principles outlined below describe the flow of a uniform incom-
pressible fluid in solid tubing. The conduits utilised in ECLS can be considered
solid tubes; however, blood can only approximate a uniform fluid due to the pres-
ence of suspended cells and chemical interactions of suspended proteins with the
conduit walls.
Two fluid characteristics are relevant to flow: density and viscosity. As a descrip-
tion of mass per unit volume, density () particularly affects the momentum of mov-
ing fluid and resistance to acceleration and for blood is slightly greater than water at
1.06gmL1 at 37C.
Viscoelasticity
Fig. 1.15 Relative Viscosity of Blood compared to water vs. haematocrit and shear rate. Viscosity
rises exponentially with increasing haematocrit but falls with increasing shear rates, as aggregating
forces between red cells become less prominent (back panel). Viscosity data plotted from equa-
tions in [37] and shear rates of vessels from [38, 39]
Fig. 1.16 Graphic description of the Bernoulli equation for laminar flow in a rigid conduit from a
high pressure compartment to a low pressure compartment. 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) across the system is constant
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 limit 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, i.e. 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 ECLS cannulas in vessels using colour flow Doppler techniques.
Figure1.16 also displays the total internal energy in the system, i.e. sum of poten-
tial and kinetic energy. This is nearly constant, falling only 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 and energy of momentum raises the pressure (potential energy) in the
receiving container. This 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,
constrained pressure and any kinetic energy:
46 M.J. Brain et al.
r velocity2
r gh + + P = Constant where g = 9.807ms2
2
h = fluid column height ( m )
r velocity2 (1.23)
+ P = Constant For neglible gravity
2
Kinetic Potential
+ = Constant
Energy Energy
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 pres-
sure. For tubes of varying cross section the velocity will increase as diameter
decreases, leading to drops in pressure; however, the total internal energy will still
remain constant such that:
r V12 r V22
+ P1 = + P2
2 2 (1.24)
r (V22 - V12 )
DP =
2
Inspection of Fig.1.16 and the former consideration of viscoelasticity reveals why
the Bernoulli equation only strictly applies to an incompressible Newtonian fluid in
a solid conduitenergy lost to viscous forces or stored as elastic potential energy is
not included in Eq. (1.24). Nevertheless the concepts of conservation of energy from
potential (pressure) to kinetic (velocity) provide the foundation for understanding
conduit flow and Eq. (1.24) is used clinically in Doppler calculations.
Resistance toFlow
Equation (1.22) implies that any resistance simply alters the ratio between the pres-
sure 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, the 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 pressure
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 stationary blood film in contact with
the conduit walls from which biofilms form. Laminar flow is the least traumatic on
1 Physiology ofExtracorporeal Life Support (ECLS) 47
formed blood elements and the most energy efficient, being directly proportional to
the pressure gradient and inversely proportional to resistance defined as:
8 length h
Resistance to laminar flow =
p radius4
(
mmHgL-1 min -1 ) (1.25)
Combining Eqs. (1.22) and (1.25) and converting radius to diameter, d, yields
Poiseuilles equation:
( P - POutlet ) p d 4
Q = Inlet (1.26)
128 length h
The effect of viscosity () 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, which can be limited by vessel size and tech-
nical 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 [42] where, for a conduit of diameter d and a fluid with a mean velocity v :
r dv
Re = (1.27)
h
48 M.J. Brain et al.
At Re numbers less than 100, pure laminar flow is apparent; from 100 to 1000 lami-
nar flow occurs but an increasingly wide boundary layer appears against the station-
ary conduit; above 1000 there is a transition to turbulent flow streams and eddy
currents; Re numbers >10,000 are purely turbulent flow [42]. The linking of Re to
velocity is critical in determining the maximal flow through extracorporeal circuits
and will be discussed below. The distance blood must flow before laminar flow can
be established is also related to Re and is known as the entrance length:
Re0.03tube diameter (Fig. 1.16).
A Reynolds number can also be calculated for the resistance (drag) that a particu-
late experiences in a viscous fluid. For erythrocytes suspended in plasma the cell
diameter (dRBC) and a coefficient, , for cell deformity Replasma is defined as [37]:
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, mLs1) through any two points of the extracorporeal circuit must be
constant and flow can be described as the product of blood velocity, V (cms1) and
cross sectional area, CSA (cm2):
Q1 = Q2
(1.29)
CSA1 V1 = CSA 2 V2
The cross section of the conduit is calculated as radius2 and an ECLS flow of
5Lmin1 in 1cm diameter tubing yields a velocity of 1.06ms1. Oxygenator cross-
sectional areas vary but an approximation can be made by dividing the oxygenator
priming volume by the oxygenator width in cm, yielding a volume per cm. This can
be divided by the volume in 1cm 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 1.3cms1 or around 100 times slower than the conduit velocity, 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
1 Physiology ofExtracorporeal Life Support (ECLS) 49
Fig. 1.17 Schematic of a simplified ECLS circuit with a centrifugal vortex pump, a membrane
oxygenator and conduits between the patient access cannulae. Systems consisting of centrifugal
vortex pumps (as opposed to roller pumps) are typically valveless, allowing 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
hollow fibre oxygenators, the blood must spread across the opening face of the
tubules. More clot deposition over time will gradually decrease the number of paths
that blood can take.
r v12 r v22
r gh1 + + P1 + hpump + hRBC = r gh2 + + P2 + hfriction + hRBC (1.30)
2 2
Fig. 1.18 Energy exchange across the extracorporeal circuit; after [43]. Refer to Fig.1.16 for
discussion of energy transfer from pressure to velocity. The total internal energy of the fluid is
depicted without inclusion of heat generated from viscous friction. Thus the total energy falls at all
points, most prominently in the low flow region of the oxygenator, being added only by the pump.
Narrow spacing between laminar flow lines represents high velocity at low pressure
1 Physiology ofExtracorporeal Life Support (ECLS) 51
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 erythrocyte.
Energy is added by the pump (pump header, 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 (pressure and velocity) of the
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 [42].
At startup, the high velocity veins of a 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. (1.30). This does not mean turbulent flow will exist throughout the
system; rather the energy lost from small areas of turbulence will summate to
oppose the applied energy of the rotor resulting in a steady terminal velocity.
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
(conservation of momentum). Instead, the kinetic energy is converted to pressure
against the pump housing, against the 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 to
rpm by dividing by 60).
rpm
P = torque 2p (1.31)
60
Torque represents force multiplied by angular distance and has SI units of Newton
metre (Nm); however, it is more understandable to clinicians as joules per radian.
It should be noted in Eq. (1.31) that this is the force required to move the rotors to
both propel the fluid column and overcome viscous friction at the rotor blades.
applied energy and a new steady rotor velocity results. In other words, by Eq. (1.31),
if the same power is applied to the rotors then rpm will increase, conversely if the
rotors are set to maintain a continuous rpm, power consumption must fall.
This effect of occlusion lowering the work done by 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 pressure in the distal conduit until mechanical failure occurs, i.e. the work the
roller pump performs on the blood increases with occlusion.
Flow Regurgitation
It should be apparent from Eq. (1.30) that a high opposing velocity or pressure in the
outflow region opposes forward flow and the energy imparted by the pump must
overcome this. This is of importance 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 be wholly spent
overcoming viscous friction at the blades from extremely turbulent flow.
As introduced above, turbulent flow refers not only to eddies of the blood stream
but also to local turbulence of plasma around erythrocytes (Eq.1.27) and the shear
stress on erythrocyte cell membranes in this setting will be high, often resulting in
haemolysis.
Cavitation
As introduced above, turbulent flow describes rapidly varying random inverse fluc-
tuations of pressure and velocity that sum to the total internal energy of the fluid
[42]. The amplitude of these fluctuations increases with the internal energy of the
fluid and, as discussed, high local velocities may induce erythrocyte haemolysis
from shear stress. A second mechanism of haemolysis is known as cavitation.
If the peak negative pressure generated in regions of turbulent flow falls below
the vapour pressure of dissolved gases in blood (usually at the tips of the rotor
blades) then bubbles will form and then rapidly collapse as the fluctuating pressure
increases again. Cavitation describes the implosion of these small bubbles in blood;
as each bubble collapses, a shockwave occurs as the walls of the cavity collapse and
enough force may be generated to rupture cell membranes.
The positive and negative pressure fluctuations in turbulent flow occur around
the average pressure in the region of flow, whereas the vapour pressure is an abso-
lute property of the fluid at a particular temperature. Therefore, cavitation is more
prominent if low pressures exist at the inlet of the circuit and can be avoided by
maintaining central venous pressure, keeping the pump head below the level of the
access cannula and avoiding excessive pump speed.
1 Physiology ofExtracorporeal Life Support (ECLS) 53
The interaction of power consumption with axial velocity just described for velocity
pumps is of particular relevance to the operation and monitoring of VADs. The loca-
tion of the device between left ventricle and aorta (Fig.1.1) exposes it to pulsatile
flow (if enough ventricular function exists) while the short path length and lack of
an oxygenator results in minimal dampening of this pressure fluctuation. The effects
of a varying pressure gradient can be considered in an analogous fashion to the
discussion of occlusion (Fig.1.19).
In a normal heart, the aortic valve opens in ventricular systole and the slightly
higher pressure in the left ventricle results in antegrade flow and transfer of pressure
energy to the aortic blood stream and elastic aortic wall. In diastole, the pressure
falls in the relaxing ventricle and the pressure of blood in the aorta closes the aortic
valve, maintaining a pressure gradient between the ventricular cavity and aorta. The
stored potential energy of the aortic blood drives flow through the peripheral circu-
lation after ventricular contraction has ceased for as long as aortic pressure is greater
than right atrial pressure (Fig.1.19, lower left panel).
In the LVAD supported circulation, the total energy of blood in the LVAD outlet
exceeds that in the aorta, and in the zone of flow deceleration (Fig.1.16 and Eq.
(1.23)) the kinetic energy of the two streams becomes equal with the additional
energy augmenting aortic pressure. Any pressure increase in the left ventricle
(whether enough to open the aortic valve or not) will lower the pressure gradient
between the ventricle and aorta, requiring less force to be exerted on the blood by
the rotors. By Eq. (1.31), this results in a higher flow and motor rpm for the same
motor power output or a reduction in motor power to maintain the same rpm
(Fig.1.19). These devices are thus described as afterload-sensitive.
In reality, pulsatile fluctuations in the arterial trace (accentuated if the aortic valve
starts opening), motor rpm and motor power occur, with the proportions dependent
on the LVAD controller software, pump speed and conditions. LVAD computer-
controllers monitor the waveform of speed oscillations to provide numerical indica-
tors of the ventricleLVAD interaction. The averaged peak-trough speed difference is
known as the pulsatility index. Monitoring this value over time gives a guide to
changing ventricular conditions, with high values suggesting significant LV contrac-
tility (which may mean the LVAD flow can be increased to further unload the ven-
tricle) and falling values indicating loss of contractility (that acutely may mean the
LV is collapsing and beginning to occlude the LVAD inflow). Similarly, the shape of
the waveform (rate of speed change) compared to a sine wave (over pumping index)
can be monitored and a numeric index produced. As they depend on multiple factors
including LV function, RV function, aortic pressure, aortic valve opening and heart
rate, and may be confounded by dynamic factors such as thrombus formation and
position [44], such parameters are only useful when monitored for changes over
time, with alterations triggering other tests such as echocardiography.
54 M.J. Brain et al.
AoVClose
PAo AoVOpen
Systole LVAD POut
Diastole LVAD POut
Pressure
Systole P
Systole LVAD PIn
PLV
Diastole P
MVClose
MVOpen Diastole LVAD PIn
PLA
Time Time
Normal Circulation LVAD Supported Circulation
Fig. 1.19 Afterload sensitivity of a left ventricular assist device. Top panel depicts an axial flow
LVAD device flowing from the Left Ventricle (LV) to the Aorta (Ao). Middle panel (not to scale)
depicts combined total energy of the blood stream as it progresses through the LVAD and the
contribution of potential energy (pressure) and kinetic energy (velocity) in systole (black lines)
and diastole (red). Viscous friction results in net energy loss at all points except the pump where
external energy is added. Left lower panel depicts the pressure vs. time graph for aorta, left ven-
tricle and left atrium (LA) for a normal circulation; mitral (MV) and aortic (AoV) valve state is
indicated. Right lower panel depicts pressure vs. time graph for an LVAD supported circulation.
The aortic valve will not open unless the LV pressure exceeds the aortic pressure for part of car-
diac cycle. Even without valve opening, aortic pressure fluctuations result from instantaneous
changes in pressure between the LV (LVAD PIn) and Aorta (LVAD POut) that occur through the
cardiac cycle from any cardiac activity
1 Physiology ofExtracorporeal Life Support (ECLS) 55
VA-ECMO
Central VA-ECMO usually involves right atrial cannulation with blood returning to
the proximal aorta, while peripheral VA-ECMO drains the great veins or right
atrium and returns to the descending aorta (Fig.1.1). The same principles determine
circuit flow as for VAD configurations; however, a pulsatile pressure gradient across
the extracorporeal circuit will not occur unless enough ventricular pressure is gener-
ated to open the aortic valve. Furthermore, the pressure difference between the aorta
and ventricle in systole and diastole will be proportionally greater than that between
the aorta and right atrium due to the lower diastolic pressure in the ventricle. Hence,
VA-ECMO will not produce a pulsatile flow unless enough ventricular function
exists to open the aortic valve.
Flow Reversal
The absence of valves in the blood path of many ECLS configurations allows blood
to flow in reverse when the pump is not running. This is particularly prominent in
VA-ECMO configurations and may be damaging, as such flow creates a relatively
low resistance systemic shunt from the arterial limb to the venous limb and in the
failing circulation will rapidly worsen organ perfusion. Flow reversal is avoided by
clamping the ECLS circuit at any time where low or no flow is anticipated, includ-
ing at initiation of support, utilising the characteristic of velocity pumps to continue
rotation when occluded. Before unclamping, the rotor speed is increased to a level
that should oppose flow reversal and then increased further as required (see occluded
flow above).
Differential Cyanosis
In peripheral VA-ECMO, flow in the upper aorta is retrograde from the point of
maximal pressure generation at the site of the ECMO outlet. As cardiac function
improves and the aortic valve begins opening, the moving column of blood ejected
by the ventricle will meet the moving column from the extracorporeal circuit. At
this point the velocity of the two streams will become equal (and low) and a swirling
region of high pressure will exist. The level of this region of relative stasis will
depend on the energy contained in the fluid columns.
If the pathology that required peripheral VA-ECMO also resulted in poor oxy-
genation by the lungs then the ventricular component of the aortic pressure stream
will contain hypoxic blood and the great vessels supplying the head and neck will
be perfused with this. The first vessels originating from the aorta are the coronary
arteries followed by the brachiocephalic trunk supplying the right arm and right
carotid. It is thus common practice to monitor saturations in the right arm to detect
this phenomenon.
56 M.J. Brain et al.
Valvular Incompetence
At the intake of the circuit depicted in Fig.1.18, pressure rapidly falls as the kinetic
energy of the blood increases. Low pressure vortices of turbulent flow will exist
around the zone of flow acceleration and these forces will tend to draw the compli-
ant venous vessel or heart chamber wall toward the intake if not opposed by the
distending pressure from the inflow of further blood. In ECLS circuits this may
manifest as transient circuit shuddering, increased haemolysis of blood and low
flow alarms. In VAD, ventricular arrhythmias may be precipitated, numeric indices
of pulsatility will decrease (ventricular contraction is ineffective due to lack of pre-
load) and poor peripheral perfusion may result (Fig.1.20).
Fig. 1.20 Left ventricular suck down due to hypovolemia and excessive VAD flows, all imaged in
diastole. The top left panel shows a long axis view with the VAD inflow cannula to the left of the
ventricular cavity. The anterior leaflet of the mitral valve is mid-cavity; however, the posterior
leaflet and left atrium are compressed with a narrow slit accommodating ventricular inflow. Top
right is a short axis view of the same valve demonstrating a small distorted mitral valve and an
unusual concavity to the ventricular septum. The bottom panels are the result of decreasing pump
speed and giving fluid; the left ventricular cavity has a more normal rounded appearance and the
left atrium and mitral inflow are no longer collapsed. Image M.Brain by permission of The Alfred
Intensive Care Unit, The Alfred Hospital, Melbourne
58 M.J. Brain et al.
Thrombus Formation
Conclusions
ECLS is a means of removing carbon dioxide and enhancing global oxygen delivery,
whether by increasing the oxygen content of blood or by augmenting flow. Although
seemingly complex, it can be comprehensively understood when framed in a
physiological and biomechanical context. The practical use of the technology will
be explored in subsequent chapters.
1 Physiology ofExtracorporeal Life Support (ECLS) 59
References
1. Kim G-B, Kim S-J, Kim M-H, Hong C-U, Kang H-S.Development of a hollow fiber membrane
module for using implantable artificial lung. J Membr Sci. 2009;326(1):1306.
2. Scheepers A, Joost H-G, Schrmann A.The glucose transporter families SGLT and GLUT:
molecular basis of normal and aberrant function. JPEN J Parenter Enteral Nutr.
2004;28(5):36471.
3. Stryer L.Biochemistry [Internet]. W.H.Freeman; 1995. Available from: http://books.google.
com.au/books?id=5PdpAAAAMAAJ
4. Beals M, Gross L, Harrell S.Efficiency of ATP production 4th Edition. Metabolism for energy
and the respiratory quotient. 1999 [cited 2012 Aug 15]. Available from: http://www.tiem.utk.
edu/~gross/bioed/webmodules/respiratoryquotient.html
5. Elia M, Livesey G.Theory and validity of indirect calorimetry during net lipid synthesis.
Am J Clin Nutr. 1988;47(4):591607.
6. Wolfe RR, Allsop JR, Burke JF.Glucose metabolism in man: responses to intravenous glucose
infusion. Metabolism. 1979;28(3):21020.
7. Hellerstein MK etal. De novo lipogenesis in humans: metabolic and regulatory aspects. Eur J
Clin Nutr. 1999;53:5365.
8. Mulquiney PJ, Kuchel PW.Control analysis of 2,3-bisphosphoglycerate metabolism [Internet].
Biochem J.1999 [cited 2012 Oct 10];342:597604. Available from: http://www.biochemj.org/
bj/342/bj3420597.htm
9. Mulquiney PJ, Bubb WA, Kuchel PW.In situ kinetic characterization of 2,3-bisphosphoglycerate
synthase/phosphatase [Internet]. Biochem J.1999 [cited 2012 Oct 10];342:56780. Available
from: http://www.biochemj.org/bj/342/bj3420567.htm
10. Lawson DS, Holt D.Insensible water loss from the Jostra Quadrox D oxygenator: an invitro
study. Perfusion. 2007;22(6):40710.
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):4327.
12. Peter S.Strong ions plus carbon dioxide. In: Kellum JA, Elbers PW, editors. Stewarts text-
book of acid-base. 2nd ed. London, UK: Lulu Enterprises; 2009. p.11132.
13. Anaesthsia Correspondence Web Site [Internet]. [cited 2012 Sep 17]. Available from: http://
www.anaesthesiacorrespondence.net/Correspond3.asp?articleid=3598&archive=1
14. Lumb AB.Nunns applied respiratory physiology. 5th ed. Oxford: Butterworth-Heinemann; 2000.
15. Sargent, John A, Gotch, Frank.Principles and biophysics of dialysis. Replacement of Renal
function by Dialysis. Dordrecht: Springer; 1996. p.34145.
16. Cussler EL.Diffusion: mass transfer in fluid systems. Cambridge: Cambridge University
Press; 1997.
17. Grathwohl P.Diffusion in natural porous media: contaminant transport, sorption/desorption
and dissolution kinetics. Boston, MA: Kluwer Academic Publishers; 1998.
18. Katoh S, Yoshida F.Rates of absorption of oxygen into blood under turbulent conditions.
Chem Eng J.1972;3:27685.
19. Matsuda N, Sakai K.Blood flow and oxygen transfer rate of an outside blood flow membrane
oxygenator. J Membr Sci. 2000;170:1538.
20. Weibel ER.The pathway for oxygen: structure and function in the mammalian respiratory
system. Cambridge, MA: Harvard University Press; 1984.
21. Epstein FH, Hsia CCW.Respiratory function of hemoglobin. N Engl J Med. 1998;
338(4):23948.
22. 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):1316.
23. Thomas LJ.Algorithms for selected blood acid-base and blood gas calculations. J Appl Physiol.
1972;33(1):1548.
60 M.J. Brain et al.
24. Kelman GR.Digital computer subroutine for the conversion of oxygen tension into saturation.
J Appl Physiol. 1966;21(4):13756.
25. 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
26. Arthurs GJ, Sudhakar M.Carbon dioxide transport. Contin Educ Anaesth Crit Care Pain.
2005;5(6):20710.
27. 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):224651.
28. Gutknecht J, Bisson MA, Tosteson FC.Diffusion of carbon dioxide through lipid bilayer mem-
branes. Effects of carbonic anhydrase, bicarbonate, and unstirred layers. J Gen Physiol.
1977;69(6):779.
29. Centor R.Serum total carbon dioxideclinical methodsNCBI 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/
30. Mochizuki M.The CO2 dissociation curve at steady state invivo. Yamagata Med J. 2004;
22(1):258.
31. Beilin LJ, Knight GJ, Munro-Faure AD, Anderson J.The sodium, potassium, and water contents
of red blood cells of healthy human adults. J Clin Invest. 1966;45(11):1817.
32. Mochizuki M.Analysis of bicarbonate concentration in human blood plasma at steady state
invivo. Yamagata Med J.2004;22(1):0924.
33. 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):57999.
34. Tazawa H, Mochizuki M, Tamura M, Kagawa T.Quantitative analyses of the CO2 dissociation
curve of oxygenated blood and the Haldane effect in human blood. Jpn J Physiol.
1983;33(4):60118.
35. Baker A, Richardson D, Craig G.Extracorporeal carbon dioxide removal (ECCO2R) in
respiratory failure: an overview, and where next? Journal of the Intensive Care Society
2012;13(3):232237
36. 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):13018.
37. Jung J, Lyczkowski RW, Panchal CB, Hassanein A.Multiphase hemodynamic simulation of
pulsatile flow in a coronary artery. J Biomech. 2006;39(11):206473.
38. Papaioannou TG, Stefanadis C. Vascular wall shear stress: basic principles and methods.
Hellenic J Cardiol. 2005;46(1):915.
39. Lipowsky H.Microvascular rheology and hemodynamics. Microcirculation. 2005;12(1):515.
40. Lipowsky HH, Kovalcheck S, Zweifach BW.The distribution of blood rheological parameters
in the microvasculature of cat mesentery. Circ Res. 1978;43(5):73849.
41. Ercan M, Koksal C.The relationship between shear rate and vessel diameter. Anesth Analg.
2003;96(1):3078.
42. Bahrami M.Introduction to fluid mechanics: ENSC283 2011 [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/
43. Bloomfield LA.How everything works: making physics out of the ordinary. Hoboken, NJ:
Wiley; 2007.
44. Salamonsen RF, Lim E, Gaddum N, AlOmari A-HH, Gregory SD, Stevens M, etal. Theoretical
foundations of a starling-like controller for rotary blood pumps. Artif Organs. 2012;
36(9):78796.
Chapter 2
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, but only at the expense of generating excessively high
inspiratory airway pressures. ECMO in this setting facilitates lung-protective venti-
lation and minimizes ventilator-associated lung injury. Despite the increasing use of
ECMO in ARDS, high-level evidence supporting its benefit for this indication is
lacking. The first successful use of ECMO for severe post-traumatic hypoxemic
respiratory failure was in 1971 by Dr. J.D. Hill and colleagues. The patient received
venoarterial extracorporeal support with a Bramson membrane lung for 75 h and
survived [3]. This success prompted others to attempt the institution of ECMO sup-
port for severe hypoxemic respiratory failure, with reports of approximately 150
cases performed from the original Hill report through 1974, though mortality was
8590 % [46]. In 1979, Warren Zapol and colleagues published the results of a
multicenter, randomized, controlled trial of mechanical ventilation supplemented
with venoarterial ECMO versus conventional mechanical ventilation alone, accord-
ing to the standard of care at that time, as a therapy for severe acute hypoxemic
respiratory failure [7]. Ninety subjects were randomized to conventional mechani-
cal ventilation alone or mechanical ventilation and ECMO, with slow and fast entry
criteria utilized for enrollment based on a combination of severity and duration of
hypoxemia, resulting in a maximum PaO2 to FiO2 ratio of 83. Subjects with pro-
longed mechanical ventilation (greater than 21 days), evidence of left atrial hyper-
tension (pulmonary capillary wedge pressure greater than 25), and chronic or
irreversible diseases were excluded. The trial demonstrated no significant difference
in survival (9.5 % in the ECMO group, 8.3 % in the control group, Table 2.1), with
high rates of infectious and bleeding complications. The low rate of survival in the
Table 2.1 Demographics and outcomes of prospective randomized trials of ECLS for ARDS
ECMO Non-ECMO
Study Year No. of pts. PaO2:FIO2 Modality survival (%) survival (%)
Zapol et al. [7] 1979 90 <83 ECCO2R 9.5a 8.3
Morris et al. [12] 1994 40 63 ECCO2R 33a 42
Peek et al. [33] 2009 180 75 ECMO 63b, c 47
ECLS extracorporeal life support, ECMO extracorporeal membrane oxygenation, ARDS acute
respiratory distress syndrome, ECCO2R extracorporeal carbon dioxide removal, PaO2 partial pres-
sure of arterial oxygen, FIO2 fraction of inspired oxygen
a
No statistically significant difference in survival between groups
b
22/90 patients (24 %) within ECMO referral group did not receive ECMO
c
Relative risk of death or severe disability 0.69 (95 % CI 0.050.97, p = 0.03) compared to non-
ECMO group
2 Hypoxemic Respiratory Failure: Evidence, Indications, and Exclusions 63
control group also highlights the high severity of illness in the study population and
the different standard of care for mechanical ventilation in ARDS at that time.
In the late 1970s and into the 1980s, Gattinoni and colleagues in Italy, among
others, began experimenting with the use of extracorporeal support in animals, and
then in humans with severe forms of ARDS. The concept involved using venove-
nous extracorporeal support at low blood flow rates principally to remove CO2, a
technique known as extracorporeal CO2 removal or ECCO2R. By removing CO2
directly from the blood, they could ventilate the subjects at lower airway pressures
and respiratory rateswhat they termed low-frequency positive-pressure ventila-
tion (LFPPV) while providing passive oxygenation through the endotracheal tube
[811]. The goal was to provide what Gattinoni referred to as pulmonary rest, the
precursor to low-volume, low-pressure ventilation strategies [11]. Results of a pro-
spective, uncontrolled study of 43 subjects treated with a LFPPV-ECCO2R strategy
for severe respiratory failure (average PaO2 to FIO2 ratio of 67) demonstrated a
survival of 49 %, despite an anticipated survival of less than 10 % based on similar
inclusion and exclusion criteria to those used in the trial by Zapol [9].
Prompted by a significantly higher survival than prior studies of subjects with
severe ARDS, a lack of randomization or control group in the Gattinoni study, and
the hypothesis that ventilator-associated lung injury plays a critical role in mortality
in ARDS, Morris and colleagues at the University of Utah evaluated the role of
LFPPV-ECCO2R in a prospective, randomized, controlled trial of subjects with
severe ARDS [12]. Forty subjects were randomized between 1987 and 1991 to
receive either conventional mechanical ventilation or a strategy of pressure-control
inverse ratio ventilation (PCIRV), followed by LFPPV-ECCO2R if PCIRV was
unable to maintain adequate PaO2 or pH. Entry criteria were similar to those used
by Zapol and Gattinoni. Oxygenation was markedly impaired, with an average PaO2
to FIO2 ratio of 63. The results demonstrated no difference in mortality between
groups, with a survival of 42 % in the control group and 33 % in the PCIRV plus
LFPPV-ECCO2R group. Hemorrhage with high rates of transfusion led to ECCO2R
discontinuation in seven (37 %) of the subjects in the experimental group, and four
subjects (21 %) developed clot within the extracorporeal circuit. Although the
authors concluded there was no role for ECCO2R as a therapy for ARDS, there were
several notable limitations of this trial. The intervention group received two differ-
ent experimental treatment modalities (inverse ratio ventilation and ECCO2R),
worldwide experience with extracorporeal techniques was limited prior to study
initiation, and the intervention arm was exposed to notably high airway pressures
despite extracorporeal support. Additionally, survival among those in the control
group was higher than in the previous studies, highlighting the differences in patient
characteristics, therapeutic interventions, or protocols for respiratory care between
this study and previously published data. Furthermore, while all of these studies
used uniform entry criteria that facilitate comparisons between trials, these criteria
may not reflect best practice for the initiation of extracorporeal support.
Starting in the 1980s, there were multiple non-randomized, observational studies
evaluating ECLS for severe ARDS, several of which are listed in Table 2.2 [9,
1322]. The studies listed are expectedly heterogeneous in their demographics,
64 D. Abrams et al.
Table 2.2 Demographics and outcomes of observational ECMO trials for acute respiratory failure
Study Year No. of pts. PaO2:FIO2 Indication Modality Survival (%)
Gattinoni et al. [9] 1986 43 67 ARDS ECCO2R 49
Wagner et al. [13] 1990 76 ARDS ECCO2R 50
Brunet et al. [14] 1993 23 84 ARDS ECCO2R 52
Manert et al. [15] 1996 21 54 ARDS ECMO 81
Kolla et al. [16] 1997 100 56 Severe ARFa ECMO 54
Peek et al. [17] 1997 50 65 Severe ARF ECMO 66
Lewandowski 1997 49 67 ARDS ECMO 55
et al. [18]
Linden et al. [19] 2000 17 46 ARDS ECMO 76
Bartlett et al. [20] 2000 86 55 ARDS ECMO 61
Davies et al. [21] 2009 68 56 ARDS/H1N1 ECMO 75
Freed et al. [25] 2010 6 61 ARDS/H1N1 ECMO 67
Roch et al. [24] 2010 9 52 ARDS/H1N1 ECMO 56
Schmid et al. [22] 2011 176 77 ARDS ECMO 56
ECMO extracorporeal membrane oxygenation, PaO2 partial pressure of arterial oxygen, FIO2 frac-
tion of inspired oxygen, ARDS acute respiratory distress syndrome, ECCO2R extracorporeal car-
bon dioxide removal, H1N1 denotes the 2009 novel influenza A(H1N1) virus, ARF acute respiratory
failure
a
Defined as shunt fraction >30 %, compliance <0.5 mL/cm H2O/kg; life-threatening anatomic
airway obstruction, refractory status asthmaticus, or uncorrectable hypercapnia with pH < 7.0 and
end-inspiratory pressure >45 mmHg
however they all included subjects with marked impairment in oxygenation that
would meet the most recent definition of severe ARDS [1].
Among these studies is the Australia-New Zealand experience with ECMO, pub-
lished by Davies and colleagues, during the influenza A(H1N1) outbreak in 2009
[21]. Sixty-eight subjects were treated with ECMO for ARDS in 15 specialist ICUs
that provided ECMO support during the study period. All cases were confirmed or
strongly suspected to be a result of influenza A(H1N1), with a high severity of ill-
ness as demonstrated by ventilatory parameters: median PaO2 to FIO2 ratio of 56,
median positive end-expiratory pressure (PEEP) of 18 cm H2O, median nadir pH of
7.2, and median highest PaCO2 of 69. Survival in this cohort of subjects was 75 %
[23]. Similar outcomes were demonstrated in other centers during the influenza
pandemic, though with smaller sample sizes (Table 2.2) [9, 1322, 24, 25].
Contemporaneous with the report by Davies et al., a study with similar demo-
graphics and outcomes of subjects with influenza A(H1N1) who were managed
without the use of ECMO was reported by Miller et al. in Utah [26]. Among the 47
subjects admitted to the ICU with a confirmed diagnosis of influenza, 30 (64 %) met
criteria for ARDS, with a median PaO2 to FIO2 ratio of 61 and median PEEP of 22.
Multisystem organ failure was common (87 %). None of the subjects with ARDS
received so-called rescue therapies such as inhaled nitric oxide, prone position-
ing, inhaled epoprostenol, or high-frequency oscillatory ventilation, although neu-
romuscular blocking agents were used in 47 % of subjects. Despite a high severity
2 Hypoxemic Respiratory Failure: Evidence, Indications, and Exclusions 65
respiratory failure and a Lung Injury Score (a.k.a. Murray Score, a composite score
based on PaO2 to FIO2 ratio, PEEP, respiratory system compliance, and radiographic
findings [34]) of 3.0 or uncompensated hypercapnia (pH < 7.2) despite optimal
conventional management were randomly assigned to receive ongoing conven-
tional mechanical ventilation at designated treatment centers or be transferred to a
single ECMO center at Glenfield Hospital in Leicester for consideration of treat-
ment with venovenous ECMO. Subjects were excluded from the trial if they had
contraindications to anticoagulation or if they had been on high pressure (peak
inspiratory airway pressure >30 cm H2O) or high FIO2 (>0.8) for greater than 7
days. Hemodynamically stable subjects randomized to the ECMO referral arm were
initially managed on transfer to Glenfield with a standardized management protocol
that included a pressure-restricted ventilation strategy, diuresis to dry weight, trans-
fusion to a hematocrit of 40 %, prone positioning and full nutrition. Those who were
hemodynamically unstable 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 subjects who received ECMO were managed with a lung-
protective ventilation strategy. 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 of a lung-protective ventilation
strategy in the conventional management group (a low-volume, low-pressure strat-
egy was advised) and perhaps because many of these subjects were difficult to ven-
tilate, only 70 % of those subjects were managed with such a strategy at any time
during the study. Subjects were well-matched between groups and the majority
were randomized early in ARDS, with 62 % of ECMO subjects and 66 % of con-
ventionally managed subjects having been on high pressure or high FIO2 for 48 h or
less, with an average of 28 h in each group. Average PaO2 to FIO2 was approxi-
mately 75 in both groups. The primary outcomedeath or severe disability by
6 months after randomizationoccurred in 37 % of the subjects referred for ECMO,
as compared with 53 % of those in the conventional management group, relative
risk 0.69 (95 % confidence interval 0.050.97, p = 0.03).
The results from CESAR, as well as the study by Noah and colleagues, may
reasonably support a strategy of transferring patients with severe ARDS to a center
capable of performing ECMO as part of a standardized management protocol [35].
However, this trial was not a randomized trial of ECMO as compared with
standard-of-care mechanical ventilation. The higher survival in the ECMO-referral
group may 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.
To date, there is no prospective randomized trial comparing modern-day ECMO
technology and techniques to standard-of-care mechanical ventilation in
ARDS. Given that the current body of literature on ECMO in ARDS has been used
to justify or dispute the efficacy of ECMO in cases of severe ARDS, there appears to
be clinical equipoise to perform another prospective, randomized clinical trial in
which subjects randomized to the ECMO arm are guaranteed to receive ECMO and
those randomized to the mechanical ventilation arm are managed with a standardized
2 Hypoxemic Respiratory Failure: Evidence, Indications, and Exclusions 67
ventilation protocol [36, 37]. Centers that already offer ECMO and believe in its util-
ity in ARDS may find it difficult to withhold ECMO from subjects who are failing
mechanical ventilation alone, and have a need to permit crossover within the study,
which may bias results toward the null hypothesis when analyzed by intention to
treat. Nonetheless, a carefully designed clinical trial that adequately matches for
baseline demographics and potential confounders and adheres to treatment protocols
would provide useful additional evidence to help settle the ongoing debate about the
role of ECMO in severe ARDS.
Indications
There is no single set of accepted criteria for the initiation of ECMO in ARDS, and
the threshold for initiation of ECMO varies considerably across studies and guide-
lines. The decision to initiate ECMO would ideally be based on a riskbenefit analy-
sis that incorporates the risk of mortality with or without extracorporeal life support
while factoring in the risk of complications as a result of its use. Unfortunately,
precise data to inform this sort of decision-making do not exist. As riskbenefit
analysis improves, thresholds for ECMO initiation will likely change to reflect risk
benefit tradeoffs compared with standard-of-care mechanical ventilation.
The early randomized trials used two sets of criteria to assess the need for
ECMO: PaO2 less than 50 mmHg for 2 h at FIO2 1.0 and PEEP greater 5 cm H2O
(fast-entry criteria) or PaO2 less than 50 mmHg for greater than 12 h at FIO2 greater
than 0.6, PEEP greater than 5 cm H2O, and shunt fraction (Qs/Qt) greater than 0.3
despite 48 h of maximal medical therapy (slow-entry criteria) [7, 12]. While these
criteria take into account the risks of prolonged hypoxemia and oxygen toxicity,
there is no consideration of hypercapnia with resulting acidemia or plateau airway
pressures as factors that may influence the decision to initiate ECMO. Furthermore,
the specific time and oxygenation cutoffs used for entry criteria have not been inde-
pendently validated or shown to correlate with mortality. Among the best data avail-
able to estimate the prognosis of patients with ARDS without ECMO support comes
from the work of the ARDS Definition Task Force [1], which redefined ARDS based
on the degree of hypoxemia as a predictor of mortality. Lower PaO2 to FIO2 ratio
cutoffs correlate with increased mortality (45 % when PaO2 to FIO2 is less than 100,
vs. 2732 % in mild to moderate ARDS) and longer duration of mechanical ventila-
tion (9 days in severe ARDS vs. 57 days in mild to moderate ARDS). Based on
these data, patients with lower ratios of PaO2 to FIO2 would seem to benefit the most
from aggressive interventions. However, the appropriate cut off for initiating ECMO
has yet to be determined. Given the consensus decision by the ARDSnet trial inves-
tigators to set a goal PaO2 of 5580 mmHg [29], ultimately representing a range of
values within which the amount of PEEP or FIO2 is not deescalated because the
patient is presumed to be near the steep portion of the oxyhemoglobin dissociation
curve, one may consider a PaO2 to FIO2 ratio of 80 as a reasonable cutoff for con-
sideration of ECMO (Table 2.3) [2]. This threshold is similar to the PaO2 to FIO2
68 D. Abrams et al.
ratio that was used as part of the slow entry criteria in the early ECMO trials [7, 12],
and at the upper limit of many observational studies [1522, 24, 25]. The PaO2 to
FIO2 ratio is one component of the Lung Injury Score which uses a combination of
physiologic and radiographic characteristics to quantify the extent of lung injury in
ARDS [34]. However, because of its non-physiologic component and a lack of data
to support its ability to predict survival among patients with the most severe forms
of ARDS, it may be less useful as a criterion for initiation of ECMO.
The amount of PEEP that is appropriate to achieve adequate oxygenation in
severe ARDS prior to initiation of ECMO has not been well established and varies
between patients. There was no difference in survival in several studies of high ver-
sus low PEEP strategy in ARDS [3840]. Nonetheless, many of the studies involv-
ing ECMO for ARDS have documented high levels of PEEP to improve oxygenation
prior to the initiation of ECMO. It is reasonable to attempt to achieve levels of PEEP
up to 1520 cm H2O, in conjunction with increases in FIO2, to reach an adequate
level of oxygenation prior to the initiation of ECMO, assuming those levels of PEEP
do not significantly compromise the patients hemodynamic status.
In addition to oxygenation parameters, the ARDSnet protocol specifies a thresh-
old of pH less than 7.15 (usually a result of uncompensated hypercapnia in the set-
ting of poor lung compliance) as the point at which plateau airway pressure targets
may be exceeded in order to increase minute ventilation and correct the acidemia. A
pH less than 7.15 may therefore be a reasonable threshold to initiate ECMO in
attempting to avoid exceeding plateau airway pressure limits that may worsen
ventilator-associated lung injury. Along similar lines, a third indication for the ini-
2 Hypoxemic Respiratory Failure: Evidence, Indications, and Exclusions 69
Exclusions
There are few absolute contraindications for ECMO in severe ARDS. Any condi-
tion that precludes the use of anticoagulation is typically considered an absolute
contraindication because of the need for systemic anticoagulant therapy to main-
tain the integrity of the circuit. However, in patients with severe bleeding, antico-
agulation may be withheld for significant periods of time with apparent safety
[42]. Given the advances in technology, the inability to anticoagulate should per-
haps be considered only a relative contraindication to receiving ECMO. Thrombotic
thrombocytopenic purpura (TTP) is another disorder that may be problematic
because of the high risk of thrombosis (within the patient or the circuit) if platelet
transfusion is needed in the setting of life-threatening hemorrhage and thrombocy-
topenia. While this may not be an absolute contraindication, the likelihood of sur-
vival without ECMO would have to be sufficiently low to consider accepting such
a high risk.
The theoretical benefit of ECMO for severe cases of ARDS largely relates to
lung protection: a decreased need for high-pressure ventilation or toxic fractions of
inspired oxygen. Yet, this potential for benefit may be limited if the patient has
already been exposed to high airway pressures and inspired oxygen levels such that
damage is irreversible. Many clinicians think that patients who have been receiving
high-pressure ventilation with plateau pressure exceeding 30 cm H2O for greater
than 7 days are less likely to benefit from ECMO [4345]. Similarly, prolonged
exposure to high fractions of inspired oxygen, which may induce lung inflamma-
tion, could nullify any beneficial effect of ECMO support, though this remains an
area of controversy [46, 47]. Earlier initiation of ECMO, perhaps for these or other
reasons, has been associated with better outcomes in some, but not all, observational
studies [18, 45, 48, 49]. Other relative contraindications include limitations in vas-
cular access that would preclude cannula placement and any conditions in which
ECMO would be unlikely to alter the patients overall prognosis, including but not
limited to advanced malignancy or severe and irreversible brain injury. Finally,
ECMO, when considered as a bridge to lung transplantation, should only be offered
if the patient is actually a candidate for transplantation.
70 D. Abrams et al.
References
1. Ranieri VM, Rubenfeld GD, Thompson BT, Ferguson ND, Caldwell E, Fan E, et al. Acute
respiratory distress syndrome: the Berlin Definition. JAMA. 2012;307(23):252633.
2. Brodie D, Bacchetta M. Extracorporeal membrane oxygenation for ARDS in adults. N Engl J
Med. 2011;365(20):190514.
3. Hill JD, OBrien 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):62934.
4. Gille JP. [Respiratory support by extracorporeal circulation with a membrane artificial lung
(authors transl)]. Bull Physiopathol Respir (Nancy). 1974;10(3):373410.
5. Bartlett RH, Gazzaniga AB, Fong SW, Burns NE. Prolonged extracorporeal cardiopulmonary
support in man. J Thorac Cardiovasc Surg. 1974;68(6):91832.
6. Hill JD, Ratliff JL, Fallat RH, Tucker HJ, Lamy M, Dietrich HP, et al. Prognostic factors in the
treatment of acute respiratory insufficiency with long-term extracorporeal oxygenation.
J Thorac Cardiovasc Surg. 1974;68(6):90517.
7. 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):21936.
8. Gattinoni L, Pesenti A, Caspani ML, Pelizzola A, Mascheroni D, Marcolin R, et al. The role of
total static lung compliance in the management of severe ARDS unresponsive to conventional
treatment. Intensive Care Med. 1984;10(3):1216.
9. Gattinoni L, Pesenti A, Mascheroni D, Marcolin R, Fumagalli R, Rossi F, et al. Low-frequency
positive-pressure ventilation with extracorporeal CO2 removal in severe acute respiratory fail-
ure. JAMA. 1986;256(7):8816.
10. Gattinoni L, Kolobow T, Agostoni A, Damia G, Pelizzola A, Rossi GP, et al. Clinical applica-
tion of low frequency positive pressure ventilation with extracorporeal CO2 removal (LFPPV-
ECCO2R) in treatment of adult respiratory distress syndrome (ARDS). Int J Artif Organs.
1979;2(6):2823.
11. 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):
1835.
12. 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):
295305.
13. Wagner PK, Knoch M, Sangmeister C, Muller E, Lennartz H, Rothmund M. Extracorporeal
gas exchange in adult respiratory distress syndrome: associated morbidity and its surgical
treatment. Br J Surg. 1990;77(12):13958.
14. Brunet F, Belghith M, Mira JP, Lanore JJ, Vaxelaire JF, Dall'ava Santucci J, et al. Extracorporeal
carbon dioxide removal and low-frequency positive-pressure ventilation. Improvement in arte-
rial oxygenation with reduction of risk of pulmonary barotrauma in patients with adult respira-
tory distress syndrome. Chest. 1993;104(3):88998.
15. Manert W, Haller M, Briegel J, Hummel T, Kilger E, Polasek J, et al. Venovenous extracorpo-
real membrane oxygenation (ECMO) with a heparin-lock bypass system. An effective addition
in the treatment of acute respiratory failure (ARDS). Anaesthesist. 1996;45(5):43748.
16. Kolla S, Awad SS, Rich PB, Schreiner RJ, Hirschl RB, Bartlett RH. Extracorporeal life support
for 100 adult patients with severe respiratory failure. Ann Surg. 1997;226(4):54464. discus-
sion 565-6.
17. Peek GJ, Moore HM, Moore N, Sosnowski AW, Firmin RK. Extracorporeal membrane oxy-
genation for adult respiratory failure. Chest. 1997;112(3):75964.
2 Hypoxemic Respiratory Failure: Evidence, Indications, and Exclusions 71
18. Lewandowski K, Rossaint R, Pappert D, Gerlach H, Slama KJ, Weidemann H, et al. High
survival rate in 122 ARDS patients managed according to a clinical algorithm including extra-
corporeal membrane oxygenation. Intensive Care Med. 1997;23(8):81935.
19. Linden V, Palmer K, Reinhard J, Westman R, Ehren H, Granholm T, et al. High survival in
adult patients with acute respiratory distress syndrome treated by extracorporeal membrane
oxygenation, minimal sedation, and pressure supported ventilation. Intensive Care Med.
2000;26(11):16307.
20. Bartlett RH, Roloff DW, Custer JR, Younger JG, Hirschl RB. Extracorporeal life support: the
University of Michigan experience. JAMA. 2000;283(7):9048.
21. 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(17):188895.
22. Schmid C, Philipp A, Hilker M, Rupprecht L, Arlt M, Keyser A, et al. Venovenous extracorpo-
real membrane oxygenation for acute lung failure in adults. J Heart Lung Transplant.
2012;31(1):915.
23. Davies A, Jones D, Gattas D. Extracorporeal membrane oxygenation for ARDS due to 2009
influenza A(H1N1)reply. JAMA. 2010;303(10):9412.
24. Roch A, Lepaul-Ercole R, Grisoli D, Bessereau J, Brissy O, Castanier M, et al. Extracorporeal
membrane oxygenation for severe influenza A (H1N1) acute respiratory distress syndrome: a
prospective observational comparative study. Intensive Care Med. 2010;36(11):1899905.
25. Freed DH, Henzler D, White CW, Fowler R, Zarychanski R, Hutchison J, et al. Extracorporeal
lung support for patients who had severe respiratory failure secondary to influenza A (H1N1)
2009 infection in Canada. Can J Anaesth. 2010;57(3):2407.
26. Miller 3rd RR, 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):7528.
27. 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):165968.
28. Matthay MA, Ware LB, Zimmerman GA. The acute respiratory distress syndrome. J Clin
Invest. 2012;122(8):273140.
29. Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung
injury and the acute respiratory distress syndrome. The Acute Respiratory Distress Syndrome
Network. N Engl J Med. 2000;342(18):13018.
30. Matthay MA, Brower RG, Carson S, Douglas IS, Eisner M, Hite D, et al. Randomized,
placebo-controlled clinical trial of an aerosolized beta(2)-agonist for treatment of acute lung
injury. Am J Respir Crit Care Med. 2011;184(5):5618.
31. Rice TW, Wheeler AP, Thompson BT, deBoisblanc BP, Steingrub J, Rock P. Enteral omega-3
fatty acid, gamma-linolenic acid, and antioxidant supplementation in acute lung injury. JAMA.
2011;306(14):157481.
32. Wiedemann HP, Wheeler AP, Bernard GR, Thompson BT, Hayden D, deBoisblanc B, et al.
Comparison of two fluid-management strategies in acute lung injury. N Engl J Med.
2006;354(24):256475.
33. 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):135163.
34. Murray JF, Matthay MA, Luce JM, Flick MR. An expanded definition of the adult respiratory
distress syndrome. Am Rev Respir Dis. 1988;138(3):7203.
35. Zwischenberger JB, Lynch JE. Will CESAR answer the adult ECMO debate? Lancet.
2009;374(9698):13078.
36. Dalton HJ, MacLaren G. Extracorporeal membrane oxygenation in pandemic flu: insufficient
evidence or worth the effort? Crit Care Med. 2010;38(6):14845.
72 D. Abrams et al.
37. Pellegrino VA, Davies AR. CESAR: deliverance or just the beginning? Crit Care Resusc.
2010;12(2):757.
38. Brower RG, Lanken PN, MacIntyre N, Matthay MA, Morris A, Ancukiewicz M, et al. Higher
versus lower positive end-expiratory pressures in patients with the acute respiratory distress
syndrome. N Engl J Med. 2004;351(4):32736.
39. Meade MO, Cook DJ, Guyatt GH, Slutsky AS, Arabi YM, Cooper DJ, et al. Ventilation strat-
egy using low tidal volumes, recruitment maneuvers, and high positive end-expiratory pres-
sure for acute lung injury and acute respiratory distress syndrome: a randomized controlled
trial. JAMA. 2008;299(6):63745.
40. Mercat A, Richard JC, Vielle B, Jaber S, Osman D, Diehl JL, et al. Positive end-expiratory
pressure setting in adults with acute lung injury and acute respiratory distress syndrome: a
randomized controlled trial. JAMA. 2008;299(6):64655.
41. International consensus conferences in intensive care medicine: ventilator-associated lung
injury in ARDS. This official conference report was cosponsored by the American Thoracic
Society, The European Society of Intensive Care Medicine, and The Societe de Reanimation
de Langue Francaise, and was approved by the ATS Board of Directors, July 1999. Am J
Respir Crit Care Med. 1999;160(6):211824.
42. Muellenbach RM, Kredel M, Kunze E, Kranke P, Kuestermann J, Brack A, et al. Prolonged
heparin-free extracorporeal membrane oxygenation in multiple injured acute respiratory
distress syndrome patients with traumatic brain injury. J Trauma Acute Care Surg. 2012;
72(5):14447.
43. Rouby JJ, Brochard L. Tidal recruitment and overinflation in acute respiratory distress syn-
drome: yin and yang. Am J Respir Crit Care Med. 2007;175(2):1046.
44. Pugin J, Verghese G, Widmer MC, Matthay MA. The alveolar space is the site of intense
inflammatory and profibrotic reactions in the early phase of acute respiratory distress syn-
drome. Crit Care Med. 1999;27(2):30412.
45. Pranikoff T, Hirschl RB, Steimle CN, Anderson HL, 3rd, Bartlett RH. Mortality is directly
related to the duration of mechanical ventilation before the initiation of extracorporeal life
support for severe respiratory failure. Crit Care Med. 1997;25(1):2832.
46. Jackson RM. Pulmonary oxygen toxicity. Chest. 1985;88(6):9005.
47. Davis WB, Rennard SI, Bitterman PB, Crystal RG. Pulmonary oxygen toxicity. Early revers-
ible changes in human alveolar structures induced by hyperoxia. N Engl J Med. 1983;
309(15):87883.
48. Beiderlinden M, Eikermann M, Boes T, Breitfeld C, Peters J. Treatment of severe acute respi-
ratory distress syndrome: role of extracorporeal gas exchange. Intensive Care Med.
2006;32(10):162731.
49. Mols G, Loop T, Geiger K, Farthmann E, Benzing A. Extracorporeal membrane oxygenation:
a ten-year experience. Am J Surg. 2000;180(2):14454.
Chapter 3
Cardiogenic Shock: Evidence, Indications,
and Exclusions
Introduction
VA-ECMO drains the blood from the right atrium via a femoral venous or internal
jugular venous cannula or, in patients with an open chest, directly from the right
atrium [5, 10, 11]. The blood is pumped through a membrane oxygenator allowing
oxygen to be added and carbon dioxide to be removed. After passing through the
oxygenator, blood is then actively pumped into the arterial system either via a can-
nula placed in a peripheral artery, usually femoral or subclavian (closed chest), or
directly into the aorta (open chest). VA-ECMO is typically a high blood ow extra-
corporeal circuit that can pump up to 7 L/min and provide full or partial cardiopul-
monary support. VA-ECMO is a closed system, which differs from standard
cardiopulmonary bypass used in the operating room, which is an open system with
a bloodair interface.
There are no randomized controlled trials comparing ECMO with other mechanical
support systems in myocardial infarction-associated cardiogenic shock; however,
several nonrandomized studies suggest a survival advantage from the early use of
ECMO in such circumstances [8, 18]. A nonrandomized, prospective, observational
study compared patients with ST-segment elevation myocardial infarction-related
profound cardiogenic shock undergoing PCI with (n = 46) and without (n = 25)
ECMO support. Those receiving ECMO had a signicantly lower 30-day mortality
than those who did not receive ECMO (39.1 % vs. 72 %, p = 0.008) [8]. Interpretation
of this data is limited by the fact that each cohort was enrolled over a different period
(19932002 for the non-ECMO cohort versus 20022009 for the ECMO cohort),
potentially leading to discrepancies in both medical and interventional management
between groups, especially given that coronary stents were unavailable at the study
center prior to 1998. Additionally, higher rates of TIMI (thrombolysis in myocardial
infarction) grade 3 ow were achieved in the ECMO group, which may have been a
consequence of improved hemodynamic stability in the catheterization lab; due to
improved PCI technique over time; or other potential explanations. Ultimately, a
prospective randomized controlled trial is needed to determine the true benet, if
any, of VA-ECMO in myocardial infarction-associated cardiogenic shock.
Several observational studies comparing conventional CPR and ECPR have been
reported [2024]. In a prospective, non-randomized, observational study of patients
with witnessed in-hospital cardiac arrests of cardiac origin lasting longer than 10
min, propensity analysis matching 46 subjects who received conventional CPR to
46 subjects who received ECPR demonstrated a signicantly higher survival at dis-
charge in the ECPR group (32.6 % vs. 17.4 %; HR 0.51, 95 % CI 0.350.74,
p < 0.0001) and at 1 year (HR 0.53; 95 % CI 0.330.83, p = 0.006) [20]. There was a
trend toward improved neurological outcomes at both time points, however these
results did not reach statistical signicance. Multivariable analysis revealed that an
initial rhythm of ventricular brillation or ventricular tachycardia and use of ECPR
were positively associated with survival to discharge, whereas longer duration of
CPR was negatively associated with survival. Importantly, survival fell to only 17 %
if more than 60 min elapsed from collapse to initiation of ECPR [20]. A more recent
analysis of 406 subjects suffering in-hospital cardiac arrest (of which only 120 sub-
jects underwent matched propensity analysis) demonstrated a signicantly higher
rate of 2-year survival with minimal neurological impairment among those treated
with ECPR compared with conventional CPR (20.0 % vs. 5.0 %; HR = 0.53, 95 %
CI = 0.360.80, p = 0.002). Independent predictors of good neurological outcome
included age less than 65 years (HR = 0.46; 95 % CI = 0.260.81, p = 0.008), CPR
duration of less than 35 min (HR = 0.37, 95 % CI = 0.180.76, p = 0.007), and subse-
quent cardiovascular intervention (HR = 0.36, 95 % CI = 0.180.68, p = 0.002) [24].
Results of ECPR for out-of-hospital cardiac arrest are poorer [18, 22, 2527].
Japanese series suggest that good outcomes can be obtained in 1520 % of the
patients, with provision of ECMO support in less than 60 min [18, 22, 25, 26]. To
attain this objective, out-of-hospital cardiac arrest patients should be scooped and
run as rapidly as possible to the nearest ECMO center [28].
to wean from ECMO [10]. Of the 26 patients (65 %) with long-term follow-up,
health-related quality of life physical domain scores were lower than age- and
gender-matched control subjects. However, they compared favorably to subjects
from other cohorts who were either bridged with ventricular assist devices to heart
transplantation or were long-term survivors of ARDS [31, 32].
ECMO Postcardiotomy
ECMO has been used to rewarm victims of accidental deep hypothermia (<28 C).
Unlike other rewarming techniques, it restores organ perfusion immediately in
patients with inadequate circulation. In a series of 46 patients with accidental hypo-
thermia, 32 had rewarming with ECMO bypass, resulting in 15 long-term survivors
[57]. In most of these patients, deep hypothermia developed after mountaineering
accidents or suicide attempts. The mean interval from discovery of the patient to
rewarming with cardiopulmonary bypass was 141 50 min. At follow-up there were
no hypothermia-related sequelae that impaired quality of life in survivors.
Neurologic and neuropsychological decits observed in the early period after
rewarming had fully or almost completely disappeared. One patient had cerebellar
atrophy on magnetic resonance imaging with mild clinical signs, a condition that
may have been caused by hypothermia. This landmark study demonstrated that
VA-ECMO can rescue and rewarm young, otherwise healthy people who had acci-
dental deep hypothermia with no or minimal cerebral impairment, even following
prolonged circulatory arrest.
Successful rescue therapy with ECMO has been described in several cases of life-
threatening pulmonary embolism [58, 59], even in patients with persistent cardiac
arrest [60]. Whether adjunctive treatments such as catheter-guided thrombectomy
or surgical embolectomy should be performed in ECMO-treated patients to hasten
recovery is still debated [59, 60], since complete lysis of pulmonary artery clot has
been reported after a few days on ECMO and heparin treatment [58, 61, 62].
The use of mechanical circulatory support remains controversial for refractory sep-
tic shock. ECMO was shown to be effective in this situation as a salvage therapy in
children [6365]. Alternatively, poor results were reported for adult patients who
received ECMO in the context of sepsis-associated refractory vasoplegia and pre-
served cardiac output [66]. However, profound myocardial dysfunction can also
occur during bacterial septic shock. In a recent series of 14 patients who received
VA-ECMO for refractory cardiovascular failure following bacterial septic shock
[67], 12 (86 %) could be weaned and 10 patients (71 %) were long-term survivors
who recovered normal left ventricular function, and reported a highly preserved
quality of life. Larger studies are needed to determine whether the benet of ECMO
outweighs the risk, especially in cases where septic shock is complicated by marked
disturbances in coagulation.
80 N. Brchot and A. Combes
ECMO-Associated Complications
The benets of ECMO must be weighed against the risks inherent in its use. Because
the ECMO system injects oxygenated blood into the descending aorta in counter-
current fashion, it increases left ventricular (LV) afterload and, in 1520 % of
patients, is associated with severe hydrostatic pulmonary edema. The latter is even
more frequent in patients with little or no residual LV ejection on ECMO. In this
context, hydrostatic pulmonary edema is further aggravated by mitral regurgitation
induced by LV dilation and increased LV end-diastolic pressure, resulting from poor
LV unloading [68]. Restoring pulsatility and decreasing cardiac afterload with an
intra-aortic balloon pump in patients receiving peripheral VA-ECMO was associ-
ated with lower LVED dimensions and pulmonary artery pressures both in animals
[69] and humans [70].
Hemorrhagic complications are among the most frequent adverse events [5]. The
use of lower levels of anticoagulation limits bleeding risks, with some centers tar-
geting activated partial thromboplastin times as low as 4060 s. There are, however,
no universally accepted anticoagulation protocols, and anticoagulation may need to
be adjusted to the specic needs of the individual patient. Thromboses, either within
the circuit or related to the indwelling portions of the cannulae, may result in stroke.
Infectious complications have been reported to varying degrees, with some
reports indicating longer durations of mechanical ventilation, ECMO support, and
hospital stays associated with infection [71]. Limb ischemia and compartment syn-
drome are of concern in VA-ECMO. Insertion of a distal reperfusion cannula into
the supercial femoral artery should be considered to perfuse blood to the extrem-
ity. Other complications that have been associated with ECMO include hemolysis,
thrombocytopenia, acquired von Willebrand syndrome, disseminated intravascular
coagulopathy, and air embolism [72]. In a recent meta-analysis incorporating 1866
patients from 20 studies of ECMO for cardiogenic shock or cardiac arrest between
2000 and 2012, complication rates (95 % CI) were reported as follows: lower
extremity ischemia, 16.9 % (12.522.6); fasciotomy or compartment syndrome,
10.3 % (7.314.5); lower extremity amputation, 4.7 % (2.39.3); stroke, 5.9 %
(4.28.3); major or signicant bleeding, 40.8 % (26.856.6); re-thoracotomy for
postcardiotomy bleeding or tamponade, 41.9 % (24.361.8); and signicant infec-
tion, 30.4 % (19.544.0) [73]. Cardiac or major vascular perforation is a rare but
potentially lethal complication of cannulation, the frequency of which depends on
the type of cannula used. Ultimately, complication rates will vary according to insti-
tutional experience and patient selection.
predictors of ICU death included ECMO implantation under CPR, severe liver or
renal failure, and female sex, while myocarditis is associated with better survival
[5]. To improve outcomes, patients with rapidly progressing cardiac dysfunction
should be urgently transferred to experienced centers, where cardiologists, heart
surgeons, and intensivists can implement all the complex medical and surgical inter-
ventions required by these critically ill patients, and can also rapidly diagnose and
treat ECMO-related complications.
Health-related quality of life, evaluated in the 28 long-term survivors of the latter
study [5], revealed persistent problems with work or other daily activities because of
physical health and frequent interferences with normal social activities, while mental
health and vitality were deemed satisfactory. Higher SF-36 scores were also obtained
for patients with longer follow-up, suggesting time-dependent improvement. Lastly,
patients who had peripheral VA-ECMO reported persistent problems at the cannula-
insertion site (lymphocele, late wound-healing, symptoms related to crural nerve
injury such as skin numbness or paresthesia), which tended to regress over time.
Conclusion
References
7. Idelchik GM, Simpson L, Civitello AB, Loyalka P, Gregoric ID, et al. Use of the percutaneous
left ventricular assist device in patients with severe refractory cardiogenic shock as a bridge to
long-term left ventricular assist device implantation. J Heart Lung Transplant.
2008;27:10611.
8. Sheu JJ, Tsai TH, Lee FY, Fang HY, Sun CK, et al. Early extracorporeal membrane oxygenator-
assisted primary percutaneous coronary intervention improved 30-day clinical outcomes in
patients with ST-segment elevation myocardial infarction complicated with profound cardio-
genic shock. Crit Care Med. 2010;38:18107.
9. Kim H, Lim SH, Hong J, Hong YS, Lee CJ, et al. Efcacy of veno-arterial extracorporeal
membrane oxygenation in acute myocardial infarction with cardiogenic shock. Resuscitation.
2012;83:9715.
10. Mirabel M, Luyt CE, Leprince P, Trouillet JL, Leger P, et al. Outcomes, long-term quality of
life, and psychologic assessment of fulminant myocarditis patients rescued by mechanical
circulatory support. Crit Care Med. 2011;39:102935.
11. Abrams D, Combes A, Brodie D. Whats new in extracorporeal membrane oxygenation for
cardiac failure and cardiac arrest in adults? Intensive Care Med. 2014;40:60912.
12. Cornish J, Gerstmann D, Begnaud M, Null D, Ackerman N. Inight use of extracorporeal
membrane oxygenation for severe neonatal respiratory failure. Perfusion. 1986;1:281287.
13. Beurtheret S, Mordant P, Paoletti X, Marijon E, Celermajer DS, et al. Emergency circulatory
support in refractory cardiogenic shock patients in remote institutions: a pilot study (the
cardiac-RESCUE program). Eur Heart J. 2013;34:11220.
14. Foley DS, Pranikoff T, Younger JG, Swaniker F, Hemmila MR, et al. A review of 100 patients
transported on extracorporeal life support. ASAIO J. 2002;48:6129.
15. Forrest P, Ratchford J, Burns B, Herkes R, Jackson A, et al. Retrieval of critically ill adults
using extracorporeal membrane oxygenation: an Australian experience. Intensive Care Med.
2011;37:82430.
16. Javidfar J, Brodie D, Takayama H, Mongero L, Zwischenberger J, et al. Safe transport of criti-
cally ill adult patients on extracorporeal membrane oxygenation support to a regional extracor-
poreal membrane oxygenation center. ASAIO J. 2011;57:4215.
17. Linden V, Palmer K, Reinhard J, Westman R, Ehren H, et al. Inter-hospital transportation of
patients with severe acute respiratory failure on extracorporeal membrane oxygenation
national and international experience. Intensive Care Med. 2001;27:16438.
18. Sakamoto S, Taniguchi N, Nakajima S, Takahashi A. Extracorporeal life support for cardio-
genic shock or cardiac arrest due to acute coronary syndrome. Ann Thorac Surg.
2012;94:17.
19. Cave DM, Gazmuri RJ, Otto CW, Nadkarni VM, Cheng A, et al. Part 7: CPR techniques and
devices: 2010 American Heart Association Guidelines for Cardiopulmonary Resuscitation and
Emergency Cardiovascular Care. Circulation. 2010;122:S7208.
20. Chen YS, Lin JW, Yu HY, Ko WJ, Jerng JS, et al. Cardiopulmonary resuscitation with assisted
extracorporeal life-support versus conventional cardiopulmonary resuscitation in adults with
in-hospital cardiac arrest: an observational study and propensity analysis. Lancet.
2008;372:55461.
21. Chen YS, Yu HY, Huang SC, Lin JW, Chi NH, et al. Extracorporeal membrane oxygenation
support can extend the duration of cardiopulmonary resuscitation. Crit Care Med.
2008;36:252935.
22. Maekawa K, Tanno K, Hase M, Mori K, Asai Y. Extracorporeal cardiopulmonary resuscitation
for patients with out-of-hospital cardiac arrest of cardiac origin: a propensity-matched study
and predictor analysis. Crit Care Med. 2013;41:118696.
23. Lin JW, Wang MJ, Yu HY, Wang CH, Chang WT, et al. Comparing the survival between extra-
corporeal rescue and conventional resuscitation in adult in-hospital cardiac arrests: propensity
analysis of three-year data. Resuscitation. 2010;81:796803.
24. Shin TG, Choi JH, Jo IJ, Sim MS, Song HG, et al. Extracorporeal cardiopulmonary resuscita-
tion in patients with inhospital cardiac arrest: a comparison with conventional cardiopulmo-
nary resuscitation. Crit Care Med. 2011;39:17.
3 Cardiogenic Shock: Evidence, Indications, and Exclusions 83
25. Kagawa E, Dote K, Kato M, Sasaki S, Nakano Y, et al. Should we emergently revascularize
occluded coronaries for cardiac arrest?: rapid-response extracorporeal membrane oxygenation
and intra-arrest percutaneous coronary intervention. Circulation. 2012;126:160513.
26. Kagawa E, Inoue I, Kawagoe T, Ishihara M, Shimatani Y, et al. Assessment of outcomes and
differences between in- and out-of-hospital cardiac arrest patients treated with cardiopulmo-
nary resuscitation using extracorporeal life support. Resuscitation. 2010;81:96873.
27. Le Guen M, Nicolas-Robin A, Carreira S, Raux M, Leprince P, et al. Extracorporeal life sup-
port following out-of-hospital refractory cardiac arrest. Crit Care. 2011;15:R29.
28. Reynolds JC, Frisch A, Rittenberger JC, Callaway CW. Duration of resuscitation efforts and
functional outcome after out-of-hospital cardiac arrest: when should we change to novel thera-
pies? Circulation. 2013;128:248894.
29. Pages ON, Aubert S, Combes A, Luyt CE, Pavie A, et al. Paracorporeal pulsatile biventricular
assist device versus extracorporal membrane oxygenation-extracorporal life support in adult
fulminant myocarditis. J Thorac Cardiovasc Surg. 2009;137:1947.
30. Asaumi Y, Yasuda S, Morii I, Kakuchi H, Otsuka Y, et al. Favourable clinical outcome in
patients with cardiogenic shock due to fulminant myocarditis supported by percutaneous
extracorporeal membrane oxygenation. Eur Heart J. 2005;26:218592.
31. George RS, Yacoub MH, Bowles CT, Hipkin M, Rogers P, et al. Quality of life after removal
of left ventricular assist device for myocardial recovery. J Heart Lung Transplant. 2008;27:
16572.
32. Heyland DK, Groll D, Caeser M. Survivors of acute respiratory distress syndrome: relation-
ship between pulmonary dysfunction and long-term health-related quality of life. Crit Care
Med. 2005;33:154956.
33. Baud FJ, Megarbane B, Deye N, Leprince P. Clinical review: aggressive management and
extracorporeal support for drug-induced cardiotoxicity. Crit Care. 2007;11:207.
34. Masson R, Colas V, Parienti JJ, Lehoux P, Massetti M, et al. A comparison of survival with and
without extracorporeal life support treatment for severe poisoning due to drug intoxication.
Resuscitation. 2012;83:14137.
35. Elsharkawy HA, Li L, Esa WA, Sessler DI, Bashour CA. Outcome in patients who require
venoarterial extracorporeal membrane oxygenation support after cardiac surgery. J Cardiothorac
Vasc Anesth. 2010;24:94651.
36. Kumar TK, Zurakowski D, Dalton H, Talwar S, Allard-Picou A, et al. Extracorporeal mem-
brane oxygenation in postcardiotomy patients: factors inuencing outcome. J Thorac
Cardiovasc Surg. 2010;140:3306. e332.
37. Rastan AJ, Dege A, Mohr M, Doll N, Falk V, et al. Early and late outcomes of 517 consecutive
adult patients treated with extracorporeal membrane oxygenation for refractory postcardiot-
omy cardiogenic shock. J Thorac Cardiovasc Surg. 2010;139:30211. 311 e301.
38. Slottosch I, Liakopoulos O, Kuhn E, Deppe AC, Scherner M, et al. Outcomes after peripheral
extracorporeal membrane oxygenation therapy for postcardiotomy cardiogenic shock: a
single-center experience. J Surg Res. 2013;181:e4755.
39. Wang J, Han J, Jia Y, Zeng W, Shi J, et al. Early and intermediate results of rescue extracorpo-
real membrane oxygenation in adult cardiogenic shock. Ann Thorac Surg. 2009;88:
1897903.
40. Wang JG, Han J, Jia YX, Zeng W, Hou XT, et al. Outcome of veno-arterial extracorporeal
membrane oxygenation for patients undergoing valvular surgery. PLoS One. 2013;8:e63924.
41. Arpesella G, Loforte A, Mikus E, Mikus PM. Extracorporeal membrane oxygenation for pri-
mary allograft failure. Transplant Proc. 2008;40:35967.
42. Chen YS, Ko WJ, Chi NH, Wu IH, Huang SC, et al. Risk factor screening scale to optimize
treatment for potential heart transplant candidates under extracorporeal membrane oxygen-
ation. Am J Transplant. 2004;4:181825.
43. Chou NK, Luo JM, Chi NH, Wu IH, Huang SC, et al. Extracorporeal membrane oxygenation
and Thoratec pneumatic ventricular assist devices as double bridge to heart transplantation.
Transplant Proc. 2012;44:87880.
84 N. Brchot and A. Combes
44. Chung JC, Tsai PR, Chou NK, Chi NH, Wang SS, et al. Extracorporeal membrane oxygenation
bridge to adult heart transplantation. Clin Transplant. 2010;24:37580.
45. Pagani FD, Lynch W, Swaniker F, Dyke DB, Bartlett R, et al. Extracorporeal life support to left
ventricular assist device bridge to heart transplant: a strategy to optimize survival and resource
utilization. Circulation. 1999;100(19 Suppl):II20610.
46. McMurray JJ, Adamopoulos S, Anker SD, Auricchio A, Bohm M, et al. ESC guidelines for the
diagnosis and treatment of acute and chronic heart failure 2012: The Task Force for the
Diagnosis and Treatment of Acute and Chronic Heart Failure 2012 of the European Society of
Cardiology. Developed in collaboration with the Heart Failure Association (HFA) of the
ESC. Eur Heart J. 2012;33:1787847.
47. Takayama H, Truby L, Koekort M, Uriel N, Colombo P, et al. Clinical outcome of mechanical
circulatory support for refractory cardiogenic shock in the current era. J Heart Lung Transplant.
2013;32:10611.
48. DAlessandro C, Aubert S, Golmard JL, Praschker BL, Luyt CE, et al. Extra-corporeal mem-
brane oxygenation temporary support for early graft failure after cardiac transplantation. Eur J
Cardiothorac Surg. 2010;37:3439.
49. DAlessandro C, Golmard JL, Barreda E, Laali M, Makris R, et al. Predictive risk factors for
primary graft failure requiring temporary extra-corporeal membrane oxygenation support after
cardiac transplantation in adults. Eur J Cardiothorac Surg. 2011;40:9629.
50. Marasco SF, Vale M, Pellegrino V, Preovolos A, Leet A, et al. Extracorporeal membrane oxy-
genation in primary graft failure after heart transplantation. Ann Thorac Surg. 2010;90:
15416.
51. Lebreton G, Nicolescu M, Leger P, Leprince P. Implantation of left ventricular support under
extracorporeal membrane oxygenation. Eur J Cardiothorac Surg. 2011;40:e1657.
52. Scherer M, Moritz A, Martens S. The use of extracorporeal membrane oxygenation in patients
with therapy refractory cardiogenic shock as a bridge to implantable left ventricular assist
device and perioperative right heart support. J Artif Organs. 2009;12:1605.
53. Scherer M, Sirat AS, Moritz A, Martens S. Extracorporeal membrane oxygenation as periop-
erative right ventricular support in patients with biventricular failure undergoing left ventricu-
lar assist device implantation. Eur J Cardiothorac Surg. 2011;39:93944. discussion 944.
54. Haneya A, Philipp A, Diez C, Metterlein T, Puehler T, et al. Successful use of temporary right
ventricular support to avoid implantation of biventricular long-term assist device: a transcuta-
neous approach. ASAIO J. 2011;57:2747.
55. Haneya A, Philipp A, Puehler T, Rupprecht L, Kobuch R, et al. Temporary percutaneous right
ventricular support using a centrifugal pump in patients with postoperative acute refractory
right ventricular failure after left ventricular assist device implantation. Eur J Cardiothorac
Surg. 2012;41:21923.
56. Takayama H, Naka Y, Kodali SK, Vincent JA, Addonizio LJ, et al. A novel approach to percu-
taneous right-ventricular mechanical support. Eur J Cardiothorac Surg. 2012;41:4236.
57. Walpoth BH, Walpoth-Aslan BN, Mattle HP, Radanov BP, Schroth G, et al. Outcome of survi-
vors of accidental deep hypothermia and circulatory arrest treated with extracorporeal blood
warming. N Engl J Med. 1997;337:15005.
58. Omar HR, Miller J, Mangar D, Camporesi EM. Experience with extracorporeal membrane
oxygenation in massive and submassive pulmonary embolism in a tertiary care center. Am J
Emerg Med. 2013;31:16167.
59. Deehring R, Kiss AB, Garrett A, Hillier AG. Extracorporeal membrane oxygenation as a
bridge to surgical embolectomy in acute fulminant pulmonary embolism. Am J Emerg Med.
2006;24:87980.
60. Hsieh PC, Wang SS, Ko WJ, Han YY, Chu SH. Successful resuscitation of acute massive pul-
monary embolism with extracorporeal membrane oxygenation and open embolectomy. Ann
Thorac Surg. 2001;72:2667.
61. Maggio P, Hemmila M, Haft J, Bartlett R. Extracorporeal life support for massive pulmonary
embolism. J Trauma. 2007;62:5706.
3 Cardiogenic Shock: Evidence, Indications, and Exclusions 85
62. Davies MJ, Arsiwala SS, Moore HM, Kerr S, Sosnowski AW, et al. Extracorporeal membrane
oxygenation for the treatment of massive pulmonary embolism. Ann Thorac Surg. 1995;
60:18013.
63. Bartlett RH. Extracorporeal support for septic shock. Pediatr Crit Care Med. 2007;8:4989.
64. Maclaren G, Butt W, Best D, Donath S, Taylor A. Extracorporeal membrane oxygenation for
refractory septic shock in children: one institutions experience. Pediatr Crit Care Med.
2007;8:44751.
65. MacLaren G, Butt W, Best D, Donath S. Central extracorporeal membrane oxygenation for
refractory pediatric septic shock. Pediatr Crit Care Med. 2011;12:1336.
66. Huang CT, Tsai YJ, Tsai PR, Ko WJ. Extracorporeal membrane oxygenation resuscitation in
adult patients with refractory septic shock. J Thorac Cardiovasc Surg. 2013;146(5):10416.
67. Brechot N, Luyt CE, Schmidt M, Leprince P, Trouillet JL, et al. Venoarterial extracorporeal
membrane oxygenation support for refractory cardiovascular dysfunction during severe bacte-
rial septic shock. Crit Care Med. 2013;41:161626.
68. Hanna BD. Left atrial decompression: is there a standard during extracorporeal support of the
failing heart? Crit Care Med. 2006;34:26889.
69. Sauren LD, Reesink KD, Selder JL, Beghi C, van der Veen FH, et al. The acute effect of intra-
aortic balloon counterpulsation during extracorporeal life support: an experimental study. Artif
Organs. 2007;31:318.
70. Petroni T, Harrois A, Amour J, Lebreton G, Brechot N, et al. Intra-aortic balloon pump effects
on macrocirculation and microcirculation in cardiogenic shock patients supported by venoar-
terial extracorporeal membrane oxygenation. Crit Care Med. 2014;42(9):207582.
71. Schmidt M, Brechot N, Hariri S, Guiguet M, Luyt CE, et al. Nosocomial infections in adult
cardiogenic shock patients supported by venoarterial extracorporeal membrane oxygenation.
Clin Infect Dis. 2012;55:163341.
72. Heilmann C, Geisen U, Beyersdorf F, Nakamura L, Benk C, et al. Acquired von Willebrand
syndrome in patients with extracorporeal life support (ECLS). Intensive Care Med.
2012;38:628.
73. Cheng R, Hachamovitch R, Kittleson M, Patel J, Arabia F, et al. Complications of extracorpo-
real membrane oxygenation for treatment of cardiogenic shock and cardiac arrest: a meta-
analysis of 1,866 adult patients. Ann Thorac Surg. 2014;97:6106.
74. Paden ML, Conrad SA, Rycus PT, Thiagarajan RR. Extracorporeal Life Support Organization
Registry Report 2012. ASAIO J. 2013;59:20210.
Chapter 4
ECCO2R inObstructive Diseases:
Evidence, Indications, andExclusions
Introduction
In the last decade, there has been an impressive technological and clinical evolution
of extracorporeal life support strategies [14], as demonstrated by the progressively
increasing number of scientific publications on this topic (Fig.4.1). To date, a vari-
ety of modalities of extracorporeal life support have been developed, characterized
by different configurations, different 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) (Fig.4.1) [6, 7], due
to the relative ease and efficiency in blood CO2 clearance granted by extracorporeal
gas exchangers as compared to oxygen delivery. CO2 is transported in blood as
bicarbonate ion or bound to hemoglobin, characterized by a steep linear dissocia-
tion kinetic, and high solubility, features which facilitate rapid diffusion across the
extracorporeal membrane. Therefore, ventilation (sweep gas flow), rather than per-
fusion (extracorporeal blood flow rate), of the artificial lung and its total surface
area are the main determinants of CO2 clearance from the blood. Since whole body
CO2 production averages 200250mL/min, blood flow as low as 0.51L/min
L. Del Sorbo, MD
Dipartimento di Anestesiologia e Rianimazione, Azienda Ospedaliera Citt della
Salute e della Scienza di Torino, Universit di Torino, Torino, Italy
Inter-departmental Division of Critical Care Medicine, University Health Network,
University of Toronto, Toronto, ON, Canada
V.M. Ranieri, MD (*)
Dipartimento di Anestesiologia e Medicina degli Stati Critici, Ospedale S.Giovanni
Battista-Molinette, Universit di Torino, Corso Dogliotti 14, 10126 Torino, Italy
Dipartimento di Anestesiologia e Rianimazione, Azienda Ospedaliera Citt della
Salute e della Scienza di Torino, Universit di Torino, Torino, Italy
e-mail: marco.ranieri@unito.it
800 20
18
n. publications on ECCO2R
n. publications on ECMO 700
16
600
14
500
12
400 10
300 8
6
200
4
100
2
0 0
99
00
01
02
03
04
05
06
07
08
09
10
11
12
13
19
20
20
20
20
20
20
20
20
20
20
20
20
20
20
Year
Fig. 4.1 Number of articles per year published in Pubmed (accessed on January 2014) on extracor-
poreal membrane oxygenation (ECMO) and on extracorporeal carbon dioxide removal (ECCO2R)
through the artificial lung theoretically removes the entire metabolic production,
without however providing significant oxygenation [811].
These features have been translated into clinical strategies that are remarkably
less invasive as compared to full extracorporeal membrane oxygenation, easier to
manage, and as efficient in CO2 clearance [6]. Several ECCO2R devices have been
approved in Europe and are available for clinical use (PALP from Maquet: http://
www.maquet.com; iLA activve from Novalung, http://www.novalung.com;
HemolungRAS from Alung, http://www.alung.com; Decap from Hemodec,
http://www.hemodec.com; and Abylcap from Bellco: www.bellco.net).
In most of these systems blood is drawn from a central vein by a draining can-
nula, driven with a centrifugal or a roller pump through the extracorporeal lung
where CO2 diffuses into the sweep gas, and returned into the venous circulation
(Fig.4.2a). Central venous access with a double-lumen cannula is required by most
of the ECCO2R systems. It is nonetheless possible to set up an ECCO2R circuit so
that two central veins are accessed, one for drainage and the other for reinfusion
(Fig.4.2b). Reducing the dimensions and number of cannulas as well as the dose of
anticoagulation, while maintaining or increasing the efficiency of CO2 removal,
have been key factors to reduce the invasiveness of these systems [6].
An alternative configuration for ECCO2R involves inserting an arteriovenous
(AV) bypass, usually by connecting the femoral artery to the contralateral femoral
vein, equipped with an artificial gas exchanger (Fig.4.2c). In this system, blood
flows through the artificial lung based on the AV pressure gradient, not on a pump,
thus requiring significantly less machinery. Although this approach is efficient, it is
more invasive and prone to related complications [6].
Rapid technological progress and effective marketing have seen ECCO2R applied
to several different patient populations. In patients with acute respiratory distress
syndrome (ARDS) [2, 12], ECCO2R has been used to limit the hypercapnic acidosis
4ECCO2R inObstructive Diseases: Evidence, Indications, andExclusions 89
Fig. 4.2 Possible configurations of extracorporeal carbon dioxide removal (ECCO2R) circuits.
(a)Minimally invasive venovenous ECCO2R circuit with a single venous vascular access through
a double-lumen cannula that can be typically inserted in the internal jugular vein or in the femoral
vein. (b) Venovenous 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 membrane
providing extracorporeal gas exchange in the circuit connecting the femoral artery with the
contralateral femoral vein
resulting from very low tidal volumes during super-protective mechanical ventila-
tion. Zimmermann and colleagues demonstrated that pumpless AV ECCO2R cleared
sufficient CO2 from 51 patients with ARDS as to enable protective mechanical ven-
tilation with tidal volumes lower than 6cc/kg of predicted body weight. It is worth
noting that 6% of these patients suffered from ECCO2R-related complications, such
as ischemia of the limb due to arterial cannulation [13]. In a randomized control trial
in patients with ARDS, low tidal volume ventilation (6mL/kg) was compared with
super-protective mechanical ventilation (3mL/kg) combined with pumpless
AV-ECCO2R.Although the study was not powered to find any difference in terms of
mortality or ventilator-free days, a reduced inflammatory response was detected in
the group treated with ECCO2R [14]. Another study in severe ARDS patients used a
minimally invasive venovenous ECCO2R device to reduce tidal volume from
6.30.2 to 4.20.2mL/kg of predicted body weight [15]: pulmonary inflammatory
mediators were reduced significantly, as was lung hyperinflation on CT.
90 L. Del Sorbo and V.M. Ranieri
At the same time that ECCO2R is being used more widely in ARDS, the evolu-
tion toward progressively less invasive devices has prompted investigations into a
broader spectrum of disease [1, 7]. In particular, the idea of applying ECCO2R in a
variety of patients with respiratory acidosis secondary to acute exacerbations of
obstructive lung disease has gained great interest [1, 7, 16, 17]. Exacerbations of
chronic obstructive pulmonary disease (COPD) and asthma are characterized by
hypercapnia, without necessarily compromising oxygenation [18, 19]. This patho-
physiological feature represents the optimal setting for ECCO2R, potentially pro-
viding an alternative, not just an adjunct, to mechanical ventilation as a way to
support respiration [17] until the lung recovers [2022]. This chapter focuses on the
main evidence for the innovative use of ECCO2R in patients with obstructive lung
diseases, along with clinical indications and exclusion criteria.
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
efficacy. However, numerous clinical reports demonstrate the efficiency of this
technologically advanced strategy, which is based on a strong pathophysiological
rationale [17, 23]. The main feature of exacerbations of obstructive diseases is
acute worsening of expiratory flow-limitation caused by increased resistance in
small airways [19, 24]. Increased resistive load is associated with a longer expira-
tory time constant (the product of airway resistance and lung compliance), which
defines the time for exponential fall in lung volume during passive exhalation [25].
The direct consequence of a longer time constant is the development of dynamic
alveolar hyperinflation, 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 (intrinsic-PEEPFig.4.3) [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.4.3) [24, 2628]. Since alveolar ventilation is
inversely proportional to PaCO2 (according to the equation PaCO 2 = V CO / V ,
2 A
where VCO 2 is CO2 production and VA is alveolar ventilation), acute exacerba-
tions of obstructive lung disease 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 also reduces the work
4ECCO2R inObstructive Diseases: Evidence, Indications, andExclusions 91
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 hemodynamic decom-
pensation [19, 3234].
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
a COPD exacerbation
Worsening of
1. Airflow resistance
and
2. Expiratory flow-limitation
pulmonary artery
Neuro-mechanical pressure
uncoupling RV preload
with dyspnea
LV afterload
Rapid shallow
Muscle fatigue breathing
Ventilatory pump failure 1. low VT
2. high RR
Hypercapnia and
respiratory acidosis
b COPD exacerbation
Worsening of
1. Airflow resistance
and
2. Expiratory flow-limitation
Rapid shallow
Muscle fatigue breathing
Ventilatory pump failure 1. low VT
2. high RR
Hypercapnia and
respiratory acidosis
ECCO2R
RV preload
CDYN Neuro-mechanical
uncoupling LV afterload
Fig. 4.3(continued)
Indications
COPD is one of the most significant social and economic burdens in Western coun-
tries, being the fourth-leading cause of death and projected to become the third-
leading cause by 2020 [35, 36]. Its natural history is characterized by progressive
deterioration of lung function, punctuated by a variable number of acute exacerba-
tions [37, 38]. COPD exacerbations are associated with considerable morbidity and
mortality, which ranges between 8% and 26% for patients admitted to the intensive
care unit [3941].
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 [4244]. Firstly, NIV has become the standard of
care for exacerbations refractory to medical treatment [31, 4345]. 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 breath-
ing and helping to reverse muscle fatigue. Together these effects are instrumental in
decreasing the need for endotracheal intubation and improving survival [4648].
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
[49]. Moreover, MV is associated with a number of undesired side effects [32, 34,
50] which may produce morbidity or mortality [51]. 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,
94 L. Del Sorbo and V.M. Ranieri
* 35
120 7,5
Respirations / min
30
100 7,4 *
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 9-12 17-20 NIV 0-4 9-12 17-20 NIV 0-4 9-12 17-20
Time intervals
Fig. 4.4 Changes over time of partial pressure of arterial carbon dioxide (PaCO2), pH, and respira-
tory 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 ref. [58]
The same pumpless AV ECCO2R device was studied in a larger case series of 21
patients with acute hypercapnic ventilatory failure not responding to NIV [58]. By
reducing PaCO2 and increasing pH, ECCO2R lowered respiratory rate and prevented
endotracheal intubation in 90% of the patients (Fig.4.4). The ECCO2R-treated
patients were compared to a matched group of controls undergoing MV after a
failed attempt on NIV.Significantly fewer patients in the ECCO2R group required
tracheostomy, although no difference was found in ICU length of stay or mortality.
ECCO2R patients suffered nine bleeding complications, and one case each of femo-
ral artery pseudoaneurysm and heparin-induced thrombocytopenia.
Similar promising results have been reported by other groups using less invasive
ECCO2R devices characterized by venovenous configurations, using a double-lumen
cannula similar in size to a dialysis catheter. In one particularly illustrative case a
72-year-old patient with COPD failing NIV was treated with VV-ECCO2R, while
additional physiological data were collected [23]. Extracorporeal CO2 clearance was
calculated by measuring the CO2 concentration in the sweep gas, native lung CO2
clearance was estimated assuming a respiratory quotient equal to one, and intra-tidal
esophageal pressure swings were recorded as a surrogate for the work of breathing.
The investigators showed the progressive transition of gas exchange from the natu-
ral to the artificial lung. Moreover, they documented the ECCO2R-induced rapid
reduction of esophageal pressure variation and respiratory rate, proving a beneficial
effect on the work of breathing. Avoiding endotracheal intubation and MV, the
patient was able to speak, drink, and eat, despite the severity of the exacerbation.
96 L. Del Sorbo and V.M. Ranieri
Acute severe asthma is a major health burden, as its prevalence is growing worldwide
with increasing admissions to hospital and intensive care unit [19, 62]. The in-hospital
mortality rate due to acute exacerbations 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 ICU for intubation and MV [19, 63].
The treatment of acute severe asthma consists of measures to reverse airflow
obstruction. Ventilatory assistance is provided to patients refractory to medical treat-
ment and developing hypoxemia or hypoventilation [19, 62]. However, while hyper-
capnia is observed in about 10% of asthma patients in the acute setting, only a small
percentage of these require invasive MV [64]. 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 [65, 66].
4ECCO2R inObstructive Diseases: Evidence, Indications, andExclusions 97
Fig. 4.5 Trends in ventilator parameters demonstrating the timeframe of improvement in respira-
tory mechanics of two patients with acute severe asthma refractory to conventional therapy and
invasive mechanical ventilation and therefore treated with extracorporeal carbon dioxide removal
(ECCO2R), as described by Brenner etal. [74]. Reproduced with permission from ref. [74]
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
treatment particularly complex. Since ECCO2R has become a minimally invasive
treatment, well-tolerated, and easy to manage, requiring similar resources as renal
dialysis, every patient with adequate indications should be considered 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
100 L. Del Sorbo and V.M. Ranieri
technology, there are still a number of clinically relevant, 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 potential vascular trauma during the
insertion of the circuit cannula, and bleeding due to the need for therapeutic antico-
agulation to avoid thrombosis of the circuit. Therefore, anatomical abnormalities or
vascular diseases preventing the correct insertion of the ECCO2R cannula, pre-
existing risk of bleeding, contraindications to the administration of anticoagulants,
or actual hemorrhagic diseases, represent major conditions that may preclude the
application of ECCO2R (Table4.1).
Moreover, additional exclusion criteria are applied in clinical trials investigating
the efficacy of ECCO2R in improving the recovery from respiratory failure [54, 58,
59]. 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 (Table4.1). An exception would
be a patient with end-stage obstructive lung disease in whom ECCO2R might serve
as a bridge to lung transplant [75].
References
37. Burge S, Wedzicha JA.COPD exacerbations: definitions and classifications. Eur Respir J
Suppl. 2003;41:46s53.
38. Hurst JR, Vestbo J, Anzueto A, etal. Susceptibility to exacerbation in chronic obstructive
pulmonary disease. N Engl J Med. 2010;363:112838.
39. Patil SP, Krishnan JA, Lechtzin N, Diette GB.In-hospital mortality following acute exacerba-
tions of chronic obstructive pulmonary disease. Arch Intern Med. 2003;163:11806.
40. Donaldson GC, Wedzicha JA.COPD exacerbations. 1: epidemiology. Thorax. 2006;61:
1648.
41. 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;60:92531.
42. Quon BS, Gan WQ, Sin DD.Contemporary management of acute exacerbations of COPD:
asystematic review and metaanalysis. Chest. 2008;133:75666.
43. 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
1:1232.
44. 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;163:125676.
45. International Consensus Conferences in Intensive Care Medicine: noninvasive positive pres-
sure ventilation in acute respiratory failure. Am J Respir Crit Care Med. 2001;163:28391.
46. Brochard L, Mancebo J, Wysocki M, etal. Noninvasive ventilation for acute exacerbations of
chronic obstructive pulmonary disease. N Engl J Med. 1995;333:81722.
47. 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;CD004104.
48. Scala R, Naldi M, Archinucci I, Coniglio G, Nava S.Noninvasive positive pressure ventilation
in patients with acute exacerbations of COPD and varying levels of consciousness. Chest.
2005;128:165766.
49. Chandra D, Stamm JA, Taylor B, etal. Outcomes of noninvasive ventilation for acute exacer-
bations of chronic obstructive pulmonary disease in the United States, 1998-2008. Am J Respir
Crit Care Med. 2012;185:1529.
50. Rello J, Ollendorf DA, Oster G, etal. Epidemiology and outcomes of ventilator-associated
pneumonia in a large US database. Chest. 2002;122:211521.
51. 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;128:51824.
52. Cardenas Jr VJ, Lynch JE, Ates R, Miller L, Zwischenberger JB.Venovenous carbon dioxide
removal in chronic obstructive pulmonary disease: experience in one patient. ASAIO
J.2009;55:4202.
53. 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;72:5713.
54. Abrams DC, Brenner K, Burkart KM, etal. Pilot study of extracorporeal carbon dioxide
removal to facilitate extubation and ambulation in exacerbations of chronic obstructive pulmo-
nary disease. Ann Am Thorac Soc. 2013;10:30714.
55. Bonin F, Sommerwerck U, Lund LW, Teschler H.Avoidance of intubation during acute exacer-
bation of chronic obstructive pulmonary disease for a lung transplant candidate using extracor-
poreal carbon dioxide removal with the Hemolung. J Thorac Cardiovasc Surg. 2013;145:e434.
56. 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:
398405.
57. Brederlau J, Wurmb T, Wilczek S, etal. Extracorporeal lung assist might avoid invasive ven-
tilation in exacerbation of COPD.Eur Respir J.2012;40:7835.
4ECCO2R inObstructive Diseases: Evidence, Indications, andExclusions 103
58. Kluge S, Braune SA, Engel M, etal. Avoiding invasive mechanical ventilation by extracorporeal
carbon dioxide removal in patients failing noninvasive ventilation. Intensive Care Med. 2012;38:
16329.
59. Burki NK, Mani RK, Herth FJ, etal. A novel extracorporeal CO(2) removal system: results of
a pilot study of hypercapnic respiratory failure in patients with COPD.Chest. 2013;143:
67886.
60. Garcia JP, Kon ZN, Evans C, etal. Ambulatory veno-venous extracorporeal membrane oxy-
genation: innovation and pitfalls. J Thorac Cardiovasc Surg. 2011;142:75561.
61. 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;30:9714.
62. Phipps P, Garrard CS.The pulmonary physician in critical care. 12: acute severe asthma in the
intensive care unit. Thorax. 2003;58:818.
63. Krishnan V, Diette GB, Rand CS, etal. Mortality in patients hospitalized for asthma exacerba-
tions in the United States. Am J Respir Crit Care Med. 2006;174:6338.
64. McFadden Jr ER, Lyons HA.Arterial-blood gas tension in asthma. N Engl J Med. 1968;278:
102732.
65. Braman SS, Kaemmerlen JT.Intensive care of status asthmaticus. A 10-year experience.
JAMA. 1990;264:3668.
66. 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;119:1159.
67. 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;48:1113.
68. Conrad SA, Green R, Scott LK.Near-fatal pediatric asthma managed with pumpless arteriove-
nous carbon dioxide removal. Crit Care Med. 2007;35:26249.
69. 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;35:9458.
70. Shapiro MB, Kleaveland AC, Bartlett RH.Extracorporeal life support for status asthmaticus.
Chest. 1993;103:16514.
71. 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.
72. 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;55:4752.
73. MacDonnell KF, Moon HS, Sekar TS, Ahluwalia MP.Extracorporeal membrane oxygenator
support in a case of severe status asthmaticus. Ann Thorac Surg. 1981;31:1715.
74. Brenner K, Abrams DC, Agerstrand CL, Brodie D.Extracorporeal carbon dioxide removal for
refractory status asthmaticus: experience in distinct exacerbation phenotypes. Perfusion.
2014;29:268.
75. Del Sorbo L, Boffini M, Rinaldi M, Ranieri VM.Bridging to lung transplantation by extracor-
poreal support. Minerva Anestesiol. 2012;78:24350.
Chapter 5
ECLS as a Bridge to Lung Transplantation
Christian Kuehn
Introduction
Extracorporeal life support (ECLS) has for many years been an established treatment
procedure for patients suffering from acute cardiac, respiratory, or cardiorespira-
tory failure. The procedure can essentially be deployed as a bridge to recovery or
bridge to transplantation where, depending on the underlying disease, a heart,
lung, or combined heartlung transplant is being considered. There has recently
been rapid technical development, and this is nowadays revolutionizing ECLS
treatment before, during, and after lung transplantations. In this chapter, we dis-
cuss technical developments in ECLS treatment and their application to the eld of
lung transplantation.
Historical Aspects
Since the 1980s, lung transplantation has evolved from an experimental technique
to an established option for treatment of severe lung diseases and diseases of the
pulmonary circulation. Technical advances in isolated lung transplantation were
driven especially by Cooper and colleagues in Toronto. In this early era of lung
transplantation, ECLS was viewed as a contraindication to transplantation because
mechanical support was marked by myriad complications and poor outcomes.
Especially troublesome were bleeding, hemolysis, infection, and technical compli-
cations with the hardware then available. Nevertheless, as the number of transplan-
tations rose and transplantation techniques developed further, in 1991 our center
C. Kuehn, MD (*)
Department of Cardiac, Thoracic, Transplantation and Vascular Surgery, Privatdozent Dr.
med., Hannover Medical School, Carl-Neuberg-Str. 1, Hannover 30625, Germany
e-mail: kuehn.christian@mh-hannover.de
reported the rst use of ECLS as a bridging strategy for re-transplantation in patients
suffering severe early graft failure [1].
This report provoked discussion and controversy. After all, deterioration to the
point of mechanical ventilation or necessity for ECLS signaled the end of transplant
candidacy in most lung transplantation centers. At the time, this was a sensible
approach in light of the existing complications, but also the inability to sustain
patients for more than a few days using ECLS. The rst report in 1992 of successful
ECLS bridging to lung transplantation for post-traumatic ARDS remained, in the
end, only a case report that led to further debate [2]. Thus, from this early era of lung
transplantation, one nds no systematic data and only a few case reports about the
use of ECLS.
Over the ensuing decade, lung transplantation became ever more established and
the number of transplants continued to rise. As transplantation became a realistic
option for patients with end-stage lung disease, the number of patients on the wait-
ing list climbed. Indeed, listed patients rose out of proportion to the available and
suitable organs, so that increasing numbers of patients decompensated while await-
ing a donor. Mechanical ventilation was still viewed as a signicant risk factor for
mortality after lung transplantation [3].
Even after successful intubation many patients had life-threatening hypercapnia and
acidosis despite all attempts to ventilate adequately. Further technical developments in
ECLS and the introduction of pumpless support systems offered a novel option: extra-
corporeal carbon dioxide removal (ECCO2R; Chap. 4) as a bridge to lung transplanta-
tion [4]. These patients could be successfully bridged with a pumpless (arteriovenous)
device, with a notable 1-year survival rate after transplantation of 80 %. The median
duration of ECLS support was over 2 weeks but, because pumpless AV ECLS allows
only low blood ow rates, it is suitable only for CO2 elimination.
For patients with greatly impaired oxygen exchange, adequate saturation of the
blood cannot be attained with low-ow, pumpless circuits (see Chap. 1). By adding
a pump and taking advantage of newer, low-resistance membranes, sufcient blood
ow could be captured to make venovenous ECLS practical for sustaining both CO2
removal and oxygenation [5]. Numerous technical innovations led to a clear reduc-
tion of complications and lowered the threshold for using ECLS early, rather than
only as a last resort. This cast an old tool in a new light: ECLS could be utilized to
bridge even the sickest patients to transplant.
The single, most important technological change was probably the introduction
of coated, polymethylpentene, hollow-ber membranes in the oxygenator, affording
excellent gas transfer [6]. Another crucial advance was to rene the system surface
coating to enhance blood compatibility. Together with powerful centrifugal pumps
engineered to minimize trauma to blood elements, systems could run in a stable
manner for a long duration, some up to 30 days. Based on these technical improve-
ments and the increasing routine around the use of ECLS, the Hanover group pub-
lished the rst series of patients bridged to transplant under awake-ECLS: using no
sedation or mechanical ventilation [7]. This success led to more and more lung
transplant centers adopting ECLS and awake-ECLS as an option for carrying
patients successfully to transplant.
5 ECLS as a Bridge to Lung Transplantation 107
Support Variations
For lung-transplant bridging, three principal ECLS methods are considered. The rst
is to create an arteriovenous shunt from one femoral artery to the (contralateral)
femoral vein, using the systemic blood pressure to drive ow across the gas-
exchanging membrane. This pumpless conguration is referred to as interventional
lung assist (iLA). The second option is a venovenous (VV) ECLS system, which
typically drains from the inferior vena cava and supplies the right atrium. A third
approach is venoarterial (VA) ECLS, whereby blood drains from the right atrium
and feeds into a large artery, such as the femoral artery, subclavian artery, or aorta.
Each of these modes is further described in Chap. 6. In individual cases the implantation
of additional cannulas is considered, depending on the experience of the center.
When choosing a cannulating strategy, the therapy goal (successful bridging to
transplant) should always be kept in mind.
Venovenous ECLS
This classical form of support for respiratory insufciency can take place with two
different variants of cannulation. Using double-lumen cannulas (mostly inserted in the
jugular vein), patients with hypercapnic lung failure can be bridged very well. Smaller
diameter, dual-lumen cannulas afford only modest ow rates, often not sufcient for
the simultaneous saturation of the blood with oxygen. Nevertheless CO2 can be elimi-
nated very well with low ow rates, as with iLA. Another advantage of dual-lumen,
internal jugular cannulas is the fact that only one cannulation is required. With a
single cannulation of the jugular vein (mostly on the right side), patients retain free-
dom of movement at the legs and hips so as to participate in physical therapy and
mobilization. Since the dual-lumen cannula allows sedation, intubation, and
mechanical ventilation to be avoided, many patients can be kept awake and ambula-
tory while awaiting transplant, providing a real advantage compared to mechanical
ventilation. This approach is particularly applicable to patients with cystic brosis.
108 C. Kuehn
Venoarterial ECLS
For patients who have isolated lung failure (without concurrent circulatory failure),
the left ventricle continues to eject de-oxygenated blood despite VA-ECLS. Inguinal
cannulation would thus produce an oxygen watershed in the aorta [8], threatening
the oxygen supply to coronary arteries and the upper part of the body, and making
VA-ECLS a less than ideal choice. For patients deteriorating from pulmonary
hypertension, however, gas exchange usually remains acceptable. This advantage
(compared to patients with gas exchange failure) makes peripheral VA-ECLS a
good option for bridging these patients to transplantation. As these patients slide
into circulatory insufciency, VA-ECLS through the femoral artery unloads the
right heart and stabilizes them hemodynamically. As a rule, the efciency of gas
exchange is maintained so that there is no danger of hypoxemia in the area of the
coronary vessels or supra-aortic branches.
This form of support is also established under local anesthesia in the awake
patient. To avoid the danger of distal, arterial leg ischemia, we routinely insert a
small catheter oriented to assure perfusion of the distal leg with oxygenated blood.
We try to implant the cannulas in the groin on only one side: this allows maximal
freedom of movement on the contralateral side, enabling mobilization of the patient.
The most difcult question to answer is how to support a patient with combined gas
exchange failure and right heart dysfunction. Because of the right heart problem,
VV-ECLS is not sufcient. Yet VA-ECLS by itself is not suitable due to the
5 ECLS as a Bridge to Lung Transplantation 109
watershed and danger of hypoxemia in the upper half of the body. One could consider
central ECLS cannulation, but this is clearly more invasive, so that this means rarely
succeeds with local anesthesia in waking patients. In these fragile patients, even a
brief period of anesthesia and the risks of intubation are best avoided.
In these cases we usually begin with VV support. Should this not be sufcient
because of right heart problems, an arterial limb (with distal leg perfusion) is estab-
lished also under local anesthesia, converting to a veno-arterial-venous (VAV)
mode. In the end, blood drains from the caudal atrium and returns through a
Y-connector to both the arterial and venous systems, supporting both gas exchange
and the circulation (see Chap. 6). In these cases, a throttle screw apportions ow
through the arterial and venous limbs; a second ow probe is recommended to regu-
late ow according to demand. This form of support is enormously fragile and
prone to malfunction, demanding a sufciently high expertise of the user, especially
in the context of bridging to lung transplantation.
to eat and drink independently, interact and communicate, and participate in physical
therapy. Mobilization is somewhat restricted, often limiting participation in therapy
to what can be achieved in the bed or beside chair. On the other hand, some patients
can be ambulated successfully, even beyond the connes of the ICU. It is easy to
imagine that such an approach leads to better physical and mental preparation for
the rigors of recovery from transplantation. Meanwhile, all other measures usu-
ally taken to prepare patients for transplant can be conducted in the awake patient
on ECLS.
Often, this ideal scenario is not realized. The patient may deteriorate unexpectedly,
requiring urgent intubation. Alternatively, the awake-ECLS plan may simply fail,
leading to unplanned mechanical ventilation. If a patient has to be intubated acutely,
steps should be taken to facilitate weaning immediately once ECLS is instituted and
gas exchange and right heart function stabilize. Should complete weaning and extu-
bation be impossible, treatment should aim to withdraw sedatives and get the patient
breathing spontaneously. A thorough neurological assessment, a prerequisite for
transplantation, is only possible once the patient is awake and breathing.
Once a transplant candidate in placed on ECLS as a BTT, the outcome is deter-
mined largely by whether new complications set in before an acceptable donor organ
is procured. However, it is important to resist the pressure to accept marginal organs
just because the patient is on ECLS. To generate a good outcome it is important to
accept good organs for these sick patients [9]. Should a good donor organ be accepted,
the operation is usually done over a minimally invasive access through anterolateral
thoracotomy. At the end of the transplantation the decision for ECLS decannulation
lies with the surgeon. Only in patients with pulmonary hypertension who had
VA-ECLS implanted before transplantation is ECLS usually kept for left heart protec-
tion. Should decannulation be attempted, heparin anticoagulation is antagonized,
sometimes supplemented with platelets or coagulation factors. Through these
measures the risk of bleeding in the early postoperative phase can be minimized.
A very challenging scenario arises when complications intervene before the
patient can be transplanted. If absolute contraindications to transplantation develop,
ECLS becomes a bridge to nowhere. In this case, if there is no prospect for recov-
ery short of transplant, ECLS should be terminally withdrawn (see Chap. 13).
Additional end-organ failures (kidney, liver), fulminant infections, and inexorably
rising need for vasoactive infusions are seen by the authors as precluding successful
transplantation. In short: beware of catastrophic transplantations. Nevertheless, the
door to transplantation is not always closed denitively. At times, it may be appropriate
to remove the patient from the active transplant list, pending clinical improvement,
after which the topic of transplantation can be raised anew.
At the same time, one must never lose sight of the goal of ECLS as a BTT: transplant
must have a realistic chance for success. For occasional patients, ECLS may serve
as a bridge to recovery: that is, treatment of an identied precipitant for cardiopul-
monary failure may allow recovery and extubation, after which the patient can be
transplanted from a clinically stable situation. However, once the treating transplant
surgeon, intensivist, and pulmonologist no longer believe the goal of transplant is
realistic, an appropriate conversation should be conducted with the patient or his
surrogates to change the goals of treatment to palliation.
5 ECLS as a Bridge to Lung Transplantation 111
Pulmonary Failure
The vast majority of ECLS procedures are deployed in patients who are intubated
and mechanically ventilated. However, it is well-documented that 1-year survival
rates after lung transplantations are less than 50 % in patients requiring mechanical
112 C. Kuehn
Intraoperative VA-ECLS
Postoperative ECLS
Selected patients may benet from maintaining VA-ECLS support for a few days
after lung transplantation. For example, these could be patients who show acute
signs of ischemiareperfusion damage at the end of the operation. In the case of
5 ECLS as a Bridge to Lung Transplantation 113
Over the last few years, intensive research has been conducted on the develop-
ment of normothermic perfusion systems for donor lungs. With Ex-Vivo Lung
Perfusion (EVLP), developed in Sweden in 2007, the donor lung is connected to
ventilator and a system of tubes, similar to extracorporeal circulation [19]. With a
normothermic system, it is possible to evaluate marginal donor organs and assess
organ function using parameters such as oxygen uptake, lung elasticity, and perfu-
sion and ventilation pressures, to determine their suitability for the transplantation
process. Moreover, the cold ischemic time can be reduced through the use of this
type of normothermic organ perfusion, thus preventing harm to the donor organ
from a lack of oxygen and nutrients. The further development of the EVLP can be
seen in the standardization of integrated components in a portable casing, the
Organ Care System (OCS). The use of the portable OCS system has been described
in an initial pilot trial conducted by our Hanover group in cooperation with
colleagues from Madrid [20]. This system was used on 12 patients where the
114 C. Kuehn
donor organs, following initial cold perfusion, were connected to the OCS and
transported with warm perfusion and ventilation to the transplant clinic for successful
transplantation.
Conclusions
References
1. Jurmann 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):947. discussion 98.
2. Demertzis S, Haverich A, Ziemer G, Nerlich M, Mller KM, Wagner TO, et al. Successful
lung transplantation for posttraumatic adult respiratory distress syndrome after extracorporeal
membrane oxygenation support. J Heart Lung Transplant. 1992;11(5):10057.
3. Smits JM, Mertens BJ, Van Houwelingen HC, Haverich A, Persijn GG, Laufer G. Predictors
of lung transplant survival in EuroTransplant. Am J Transplant. 2003;3:14006.
4. Fischer S, Simon AR, Welte T, Hoeper MM, Meyer A, Tessmann R, et al. Bridge to lung trans-
plantation with the novel pumpless interventional lung assist device NovaLung. J Thorac
Cardiovasc Surg. 2006 Mar;131(3):71923.
5. Fischer S, Hoeper MM, Tomaszek S, Simon A, Gottlieb J, Welte T, et al. Bridge to lung trans-
plantation with the extracorporeal membrane ventilator Novalung in the veno-venous mode:
the initial Hannover experience. ASAIO J. 2007 Mar-Apr; 53(2):16870.
6. 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.
2006 Sep-Oct; 52(5):50912.
7. Olsson KM, Simon A, Strueber M, Hadem J, Wiesner O, Gottlieb J, et al. Extracorporeal
membrane oxygenation in nonintubated patients as bridge to lung transplantation. Am J
Transplant. 2010 Sep;10(9):21738. Epub 2010 Jul 15.
8. Hoeper MM, Tudorache I, Khn C, Marsch G, Hartung D, Wiesner O, et al. Extracorporeal
membrane oxygenation watershed. Circulation. 2014;130(10):8645.
9. Sommer W, Khn C, Tudorache I, Avsar M, Gottlieb J, Boethig D, et al. Extended criteria
donor lungs and clinical outcome: results of an alternative allocation algorithm. J Heart Lung
Transplant. 2013;32(11):106572.
5 ECLS as a Bridge to Lung Transplantation 115
10. Gregoric ID, Chandra D, Myers TJ, Scheinin SA, Loyalka P, Kar B. Extracorporeal membrane
oxygenation as a bridge to emergency heart-lung transplantation in a patient with idiopathic
pulmonary arterial hypertension. J Heart Lung Transplant. 2008;27:4668 [EBM IV].
11. Strueber M, Hoeper MM, Fischer S, Cypel M, Warnecke G, Gottlieb J, et al. Bridge to thoracic
organ transplantation in patients with pulmonary arterial hypertension using a pumpless lung
assist device. Am J Transplant. 2009 Apr;9(4):8537.
12. de Perrot M, Granton JT, McRae K, Cypel M, Pierre A, Waddell TK, et al. Impact of extracor-
poreal life support on outcome in patients with idiopathic pulmonary arterial hypertension
awaiting lung transplantation. J Heart Lung Transplant. 2011;30(9):9971002.
13. 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
Jan-Feb;54(1):310.
14. Wang D, Zhou X, Lick SD, Liu X, Qian K, Zwischenberger JB. An ambulatory pulmonary and
right heart assist device (OxyRVAD) in an ovine survival model. J Heart Lung Transplant.
2007;26:9749.
15. 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):7638.
16. 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):15106.
17. Bittner HB, Binner C, Lehmann S, Kuntze T, Rastan A, Mohr FW. Replacing cardiopulmonary
bypass with extracorporeal membrane oxygenation in lung transplantation operations. Eur J
Cardiothorac Surg. 2007;31:4627.
18. Tudorache I, Sommer W, Khn C, Wiesner O, Hadem J, Fhner T, et al. Lung transplantation
for severe pulmonary hypertension-awake extracorporeal membrane oxygenation for postop-
erative left ventricular remodelling. Transplantation. 2015;99(2):4518.
19. Steen S, Ingemansson R, Eriksson L, Pierre L, Algotsson L, Wierup P, et al. First human trans-
plantation of a nonacceptable donor lung after reconditioning ex vivo. Ann Thorac Surg.
2007;83(6):21914.
20. Warnecke G, Moradiellos J, Tudorache I, Khn 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):18518.
21. Thiara AP, Hyland V, Norum H, Aasmundstad TA, Karlsen HM, Fiane AE, et al. Extracorporeal
membrane oxygenation support for 59 days without changing the ECMO circuit: a case of
Legionella pneumonia. Perfusion. 2009 Jan; 24(1):457.
22. Strueber M. Extracorporeal support as a bridge to lung transplantation. Curr Opin Crit Care.
2010;16:6973.
23. Rosenzweig EB, Brodie D, Abrams DC, Agerstrand CL, Bacchetta M. Extracorporeal mem-
brane oxygenation as a novel bridging strategy for acute right heart failure in group 1 pulmo-
nary arterial hypertension. ASAIO J. 2014;60(1):12933.
24. Turner DA, Cheifetz IM, Rehder KJ, Williford WL, Bonadonna D, Banuelos SJ, et al. Active
rehabilitation and physical therapy during extracorporeal membrane oxygenation while awaiting
lung transplantation: a practical approach. Crit Care Med. 2011;39(12):25938.
Chapter 6
Modes of ECLS
Introduction
Extracorporeal life support (ECLS) requires that blood ow from the body, through
a gas-exchanging membrane, and back again to the body. Multiple options have
been devised both for blood removal and blood return, each with advantages,
disadvantages, and physiological consequences. Most adults with acute respiratory
distress syndrome (ARDS) are managed using venovenous (VV) ECLS, meaning
that blood is withdrawn from large central veins or the right atrium, and returned
also to the venous side of the circulation. In venoarterial (VA) ECLS, blood is
returned instead to the aorta (or other large artery). Occasionally a hybrid mode is
used in which blood is removed from a central vein, passed through the membrane,
and returned both to the central venous and arterial circulations, termed VAV
ECLS. Finally, blood can move from artery to vein, using the patients native blood
pressure to drive ow through a low-resistance membrane (AV ECLS or AVCO2R).
The rest of this chapter describes the circuit design for each mode, their physiologi-
cal impact, factors leading to mode selection for individual patients, and the
advantages of each mode (Tables 6.1 and 6.2).
Table 6.2 Venovenous (VV) and venoarterial (VA) ECLSadvantages and disadvantages
Factor VV VA
Cannulas One may sufce Two are needed
Gas exchange efciency Lower Higher
PaO2 Lower Higher
Recirculation Moderate None
Circulatory support None Full is possible
RV loading No 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
Venovenous ECLS
The VV Circuit
Cannulation for VV ECLS consists of inserting large-bore drainage cannulas in the infe-
rior (IVC) and superior (SVC) vena cavae, usually via the femoral and internal jugular
veins (Fig. 6.1). Dual-lumen cannulas have channels for both removing and returning
blood, and require only a single insertion site (Fig. 6.2). These are placed so that outow
6 Modes of ECLS 119
ports are located in both the SVC and IVC (traversing the right atrium; RA), while the
inow port resides within the RA and is directed toward the tricuspid valve [1]. With both
of these circuit congurations, blood ows from the central veins through external tubing
to pump, membrane, warmer, and other circuit components (see Chap. 8), then is returned
back to the central veins or RA. Since blood is removed from and returned to the central
venous circulation in the VV mode, some blood returned and fully oxygenated may be
immediately withdrawn again in a process called recirculation. Recirculation creates
a partially closed circulatory loop that decreases the efciency of oxygenation [2]. The
degree of recirculation depends strongly on cannula location, but can be inuenced by
the intravascular volume state, circuit blood ow rates, and patient positioning.
Single-site, dual-lumen cannulation for VV ECLS has some theoretical benets, such
as less recirculation when compared with dual-site cannulation [3]. Downsides include
increased cost, larger cannula diameter, higher risk, and need for continuous visualiza-
tion during insertion (Chap. 7). Dual-lumen cannulas are more expensive to purchase
than two single-lumen cannulas, but these costs are expected to decrease as technology
advances and more vendors enter the marketplace. Dual-lumen cannulas for adult VV
ECLS range in size from 23 to 31 French. Achievable circuit ow depends strongly on
120 L.K. Scott and B. Schmidt
cannula diameter, so the largest size possible is preferred. A general rule of thumb is that
the vessel diameter (in millimeters; judged by bedside ultrasound) should be at least one-
third the cannula size (in French). Cannulation is more technically challenging and risk-
ier than for the two-site approach. Placement of the guidewire down into the IVC should
be veried and the cannula visualized continuously during insertion to avoid catastrophic
vascular or cardiac injuries, requiring additional personnel, space, and equipment.
The physiology of VV ECLS is fully described in Chap. 1, but several points deserve
emphasis here. First, since oxygenated blood returns to the venous circulation, it
mixes with venous blood that was not captured by the outow cannula. The oxygen
saturation (SO2) of the resulting mixture is often in the mid-80s, even though the oxy-
gen partial pressure (PO2) of post-membrane blood may be 400 mmHg. Since this
mixture passes through failed lungs, the consequence is often a level of systemic PO2
6 Modes of ECLS 121
that is lower than many clinicians are comfortable with and, as discussed below, lower
than is typically achieved during VA ECLS. However, as long as total cardiac output
is adequate, saturations in the 80s (and even high 70s) are usually well-tolerated. Most
importantly, SaO2 is raised by increasing the fraction of native cardiac output captured
by the outow cannulas and directed to the membrane. Thus full support of oxygen-
ation typically requires circuit blood ow rates of 35 L/min, values that sometimes
produce hemolysis, recirculation, or circuit chattering (when the great veins collapse
around the cannula). If the goal of ECLS is largely carbon dioxide removal (ECCO2R;
see Chap. 4), much lower ow rates are possible [4] and smaller cannulas can be used.
ECCO2R using a VV circuit may be effective for patients with conditions largely
characterized by hypercapnia, such as chronic obstructive pulmonary disease (COPD)
exacerbations [5] or status asthmaticus [6], or to support an ultra-lung-protective
mechanical ventilation strategy for patients with severe ARDS [7, 8].
A second physiological consequence of VV ECLS is that it has virtually no
circulatory impact, a result of the fact that equal ows of blood are simultaneously
withdrawn and returned [9]. Because of this, right and left heart function, as well as
the pulmonary circulation, must be reasonably intact for VV ECLS to be an appro-
priate choice. Thus this mode is used primarily for patients with isolated, refractory
respiratory failure. Roughly a quarter of patients with severe ARDS have right
ventricular (RV) dysfunction [10] related to the pulmonary vascular effects of lung
injury, hypoxia, thrombosis, and mechanical ventilation. When cor pulmonale is
severe, VA ECLS may be more appropriate. On the other hand, the VV mode has
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 is likely to fall. In the
CESAR Trial of ECLS for severe ARDS, all patients were managed with VV ECLS
and no patient had to transition to a VA circuit during the trial [11]. In contrast to the
VV mode, VA ECLS 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 6.1). 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 inuence the choice of
VV or VA ECLS.
Advantages of VV ECLS
Several advantages are conferred because the VV mode returns blood to the central
veins, rather than the arteries (Table 6.2). These include a reduced risk of systemic
embolism, including catastrophic cerebral or coronary embolism; no injury to
systemic arteries that could cause hemorrhage or distal ischemia; lower circuit pres-
sures with consequently lower chance of catastrophic circuit failure; and the
122 L.K. Scott and B. Schmidt
potential for a single cannula to sufce, possibly facilitating mobilization (see Chap.
12). VV ECLS 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.
Venoarterial ECLS
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 positioned
in the RA, SVC, or IVC. Blood is returned through the carotid artery in neonates
and infants (Fig. 6.3) and the descending aorta (via the femoral artery) in older
children and adults (Fig. 6.4). Where to place the return cannula depends partly on
the blood ow needed. If a need for full support is anticipated (for example,
100125 cc/kg/min in a neonate), the excessive resistance of peripheral arteries may
not allow sufcient ow [12]. Central cannulation of the right atrium and the
ascending aorta allows larger cannulas to be used, providing higher ows and
reliable coronary and cerebral perfusion, but at a cost of problematic bleeding from
the surgical wound and sternum, and the risk and expense of two surgeries [13]. If
the arterial cannula compromises or obstructs nutrient blood ow, 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 ow (Fig. 6.5) [14].
Venoarterial (VA) ECLS can support both respiration and circulation [15]. For
adults with respiratory failure but preserved cardiac function, VV ECLS is pref-
erable, whereas VA ECLS is appropriate for those with circulatory failure
(Table 6.1). With regards to oxygenation and carbon dioxide elimination, the
gas exchange effects of VA ECLS are similar to those of VV ECLS, 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 pro-
vided by the (failing) native circulation. In contrast, with the VV conguration
much venous (desaturated) blood is not captured by the drainage cannula(s) and
passes through the diseased lungs, producing venous admixture and causing
lower PaO2 values. Second, cannula position in VA ECLS precludes the possi-
bility of recirculation, so gas exchange is more efcient. 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.
6 Modes of ECLS 123
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 saturation) 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 peculiar-
ity of varying arterial oxygen values may further threaten the circulation or
neurological status.
While the gas exchange differences between VV and VA ECLS are modest,
circulatory physiology is completely different. VV ECLS has essentially no
circulatory effect, yet the basic function of VA ECLS is to provide circulatory
support, returning oxygenated blood to the arterial circulation at physiological
perfusion pressures [16]. Since blood is drawn from the vena cava or RA, VA
ECLS unloads the right ventricle, and this mode has been used for patients with
massive pulmonary embolism, for example. However, the left ventricle does not
benet similarly. Even though the LV receives less blood from the lungs [17],
a failing LV may only eject little stroke volume, especially once extracorporeal
124 L.K. Scott and B. Schmidt
support raises blood pressure to normal levels. Related to the low native cardiac
output, aortic root ow may be sluggish 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 decom-
press the LV. Such measures should be included in the VA ECLS care pathway
[18]. 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 transcutaneous septoplasty, insertion of a vent cannula through a mini
thoracotomy [19], or placement of an intra-aortic balloon pump (IABP). One
concern with the use of an IABP is that it could obstruct ECLS ow returning
through the descending aorta [20]. However, several studies have shown a ben-
et of the IABP in off-loading the LV [21] and possibly enhancing cerebral [22]
and coronary blood ow [23].
Another difference between VV and VA ECLS relates to systemic arterial
pulsatility. Since pump ow 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 ECLS tends to be signicantly 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 [24]. Mean blood pressure relates to total cir-
cuit ow, 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 met-
rics related to pulse contour analysis (including many modern minimally inva-
sive cardiac output technologies) will not be valid. Still, the return of a pulsatile
waveform on the peripheral arterial transducer can be useful as a gauge of
cardiac recovery.
Advantages of VA ECLS
In comparison with VV ECLS, the VA mode leads to more efcient gas exchange
and no possibility of recirculation. Systemic PO2 is typically much higher than in
VV conguration, although this is sometimes not true of the cerebral, coronary, and
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 [25].
126 L.K. Scott and B. Schmidt
The traditional ECLS congurations each have limitations. For example, VA ECLS
may not provide adequate cardiac and cerebral oxygenation, particularly with
peripheral cannulation. Likewise, if a patient on VV ECLS suffers declining heart
function, some form of hemodynamic support may be needed. Hybrid systems aim
to address these challenges. Over the past few years many different congurations
have been reported. We focus on the systems most studied.
In this conguration, a venous outow cannula leads to the pump and articial
membrane, after which blood is returned both to the right atrium (or vena cava) and
to the systemic arteries (Fig. 6.6). Often this arises when a patient is started on
VA ECLS 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
(inow, connected by a Y to the arterial inow) in the IVC, SVC, or RA allows
a stream of fully oxygenated blood to traverse the lungs, raising the saturation of
blood passing through the left heart and into the proximal aorta [26]. A downside
lies in the fact that ow is diverted from the arterial system, possibly reducing
oxygen delivery. Increasing the circuit ow rate may compensate for this but may
be limited by cannula size or the adequacy of venous drainage. Furthermore, after
a venous inow cannula is spliced into the arterial limb of the circuit, much of the
ow may preferentially 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 inow limb of the
circuit (Fig. 6.7).
For some patients, especially those with severe airow obstruction, or to facilitate
lung protection in those with ARDS, extracorporeal therapy focuses on carbon
dioxide removal rather than oxygenation (see Chap. 4). This is termed Arterio-
Venous CO2 Removal (AVCO2R). Mathematical simulation shows that it is possible
to achieve total CO2 removal by directing only 1015 % of cardiac output to the
articial membrane as long as the sweep gas to blood ow ratio is 5 or greater [30].
These modest blood ow requirements can be met with correspondingly smaller
cannulas. Further, modern hollow-ber membranes present such a small resistance
to blood ow that the patients own arterial to venous blood pressure gradient is
sufcient to produce the needed circuit ow. The complete circuit can be very sim-
ple. A 1214Fr arterial catheter is inserted percutaneously using the Seldinger tech-
nique and the vein is cannulated using a larger cannula to assure low resistance and
to maximize blood ow. Oxygen is used as the sweep gas. In such a pumpless sys-
tem, circuit ow is determined by the size of the arterial cannula (the highest resis-
tance component of the circuit), and the pressure gradient between the arterial and
venous systems. As with more complex extracorporeal circuits, carbon dioxide
removal is related to sweep gas and circuit blood ow. In one clinical trial, blood
ow rates above 500 mL/min were seen, producing a mean CO2 removal of 112 mL/
min [31]. This system has shown efcacy in life-threatening asthma, COPD exacer-
bations, and ARDS [32, 33]. Pediatric use has been described, although cannula size
is more likely to be limiting than in adults.
We cared for an intubated 8-year-old male with asthma and severe dynamic
hyperination and marked hypercapnia (pH 6.8; PaCO2 184 mmHg). After sizing
the femoral vessels with ultrasound, a 12F arterial and a 14F venous catheter were
inserted. Using the conguration shown in Fig. 6.8, circuit ow was 750 mL/min.
The evolution of pH and PaCO2 over the rst hours on AV pumpless ECLS is shown
in Fig. 6.9. AVCO2R support allowed us to rapidly reduce the ventilator settings,
improve dynamic hyperination, and avoid barotrauma.
6 Modes of ECLS 129
Fig. 6.8 The conguration of the AVCO2R circuit. A simple, pumpless system using arterial and
venous cannulas attached to a low-resistance, hollow-ber gas exchange membrane
References
1. Thiagarajan RR, Brogan TV, Scheurer MA, Laussen PC, Rycus PT, Bratton SL. Extracorporeal
membrane oxygenation to support cardiopulmonary resuscitation in adults. Ann Thorac Surg.
2009;87(3):77885.
2. Combes A, Bacchetta M, Brodie D, Mller T, Pellegrino V. Extracorporeal membrane oxygen-
ation for respiratory failure in adults. Curr Opin Crit Care. 2012;18(1):99104.
3. Javidfar J, Wang D, Zwischenberger JB, Costa J, Mongero L, Sonett J, et al. Insertion of bica-
val dual lumen extracorporeal membrane oxygenation catheter with image guidance. ASAIO
J. 2011;57(3):2035.
4. Gattinoni L, Kolobow T, Tomlinson T, Iapichino G, Samaja M, White D, Pierce J. Low-
frequency positive pressure ventilation with extracorporeal carbon dioxide removal (LVPPV-
ECCO2R): an experimental study. Anesth Analg. 1978. http://journals.lww.com/
anesthesia-analgesia/Fulltext/1978/07000/Low_Frequency_Positive_Pressure_Ventilation_
with.18.aspx. Accessed 20 May 2014.
5. Abrams DC, Brenner K, Burkart KM, Agerstrand CL, Thomashow BM, Bacchetta M, et al.
Pilot study of extracorporeal carbon dioxide removal to facilitate extubation and ambulation in
130 L.K. Scott and B. Schmidt
23. Pierrakos CN, Bonios MJ, Drakos SG, Charitos EI, Tsolakis EJ, Ntalianis A, et al. Mechanical
assistance by intra-aortic balloon pump counterpulsation during reperfusion increases coro-
nary blood ow and mitigates the no-reow phenomenon: an experimental study. Artif Organs.
2011;35:86774. doi:10.1111/j.1525-1594.2011.01241.x.
24. Mohammadzadeh A, Jafari N, Hasanpour M, Sahandifar S, Ghafari M, Alaei V. Effects of
pulsatile perfusion during cardiopulmonary bypass on biochemical markers and kidney func-
tion in patients undergoing cardiac surgeries. Am J Cardiovasc Dis. 2013;3(3):15862.
25. Stub D, Bernard S, Pellegrino V, Smith K, Walker T, Sheldrake J, Hockings L, Shaw J, Duffy
SJ, Burrell A, Cameron P, Smit DV, Kaye DM. Refractory cardiac arrest treated with mechani-
cal CPR, hypothermia, ECMO and early reperfusion (the CHEER trial). Resuscitation. 2014.
pii: S0300-9572(14)00751-5. doi:10.1016/j.resuscitation.2014.09.010. [Epub ahead of print].
26. Moisan M, Lafargue M, Calderon J, Oses P, Ouattara A. Pulmonary alveolar proteinosis
requiring hybrid extracorporeal life support, and complicated by acute necrotizing pneumo-
nia. Ann Fr Anesth Reanim. 2013;32(4):e715.
27. Nakamura H, Yamaguchi H, Amano A, Nakao T. Venovenous extracorporeal membrane
oxygenation is effective against post-cardiotomy acute respiratory failure in adults. Gen
Thorac Cardiovasc Surg. 2013;61(7):4028.
28. Hoopes CW, Gurley JC, Zwischenberger JB, Diaz-Guzman E. Mechanical support for pulmo-
nary veno-occlusive disease: combined atrial septostomy and venovenous extracorporeal
membrane oxygenation. Semin Thorac Cardiovasc Surg. 2012;24(3):2324.
29. Sthr F, Emmert MY, Lachat ML, Stocker R, Maggiorini M, Falk V, et al. Extracorporeal
membrane oxygenation for acute respiratory distress syndrome: is the conguration mode an
important predictor for the outcome? Interact Cardiovasc Thorac Surg. 2011;12(5):67680.
30. Conrad SA, Brown EG, Grier LR, Baier J, Blount J, Heminc T, Zwischenberger JB, Bidani
A. Arteriovenous extracorporeal carbon dioxide removal: a mathematical model and experi-
mental evaluation. ASAIO J. 1998. http://journals.lww.com/asaiojournal/Fulltext/1998/07000/
Arteriovenous_Extracorporeal_Carbon_Dioxide.7.aspx. Accessed 16 May 2014.
31. Conrad SA, Zwischenberger JB, Grier LR, Alpard SK, Bidani A. Total extracorporeal arterio-
venous carbon dioxide removal in acute respiratory failure: a phase I clinical study. Intensive
Care Med. 2001;27:134051.
32. Jr B, Robert L, Tao W, Bidani A, Alpard SK, Traber DL, et al. Prolonged hemodynamic
stability during arteriovenous carbon dioxide removal for severe respiratory failure. J Thorac
Cardiovasc Surg. 1997;114(6):110714.
33. Mauri T, Zanella A, Pesenti A. Extracorporeal gas exchange: present and future. In: Vincent
J-L, editor. Annual update in intensive care and emergency medicine 2013. Berlin Heidelberg:
Springer; 2013. p. 60919.
Chapter 7
Vascular Access for ECLS
Steven A. Conrad
Introduction
Access to the central circulation to provide and maintain blood ow necessary for
adequate gas exchange is one of the most essential aspects for successful extracor-
poreal support. Inadequate extracorporeal ow can lead to failure to deliver suf-
cient support and limit any potential benet of ECLS. Cannulae and cannula
insertion techniques are quite variable, and the choice will depend on the goals and
mode of support, the size of the patient, size of the vessels, as well as institutional
and logistical concerns.
against the loading dilator and is tapered to facilitate insertion through tissue,
whereas this feature is optional in surgical cannulas.
Wire-reinforced cannulas contain a layer of metal spiral-wound wire embedded
in the wall of the cannula. This reinforcement allows the cannula to ex without
kinking, resist attening from external compression, and prevent collapse when neg-
ative pressures are applied to the lumen. Since these complications can result in loss
of extracorporeal support, reinforced cannulas are generally the preferred design.
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 drainage or for reinfusion (referred to as
venous and arterial cannulas, respectively). The venous design is characterized by a
greater length (up to approximately 50 cm), greater available diameter (up to 28 Fr),
and a longer distal segment with multiple side holes to facilitate drainage. The
length allows insertion into more central veins such as the superior (SVC) or infe-
rior vena cava (IVC). The arterial 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 ow is not dependent on side holes as is the venous design. An
excess number of side holes can increase the risk of hemolysis in arterial cannulas.
Recently introduced expandable, wire-reinforced cannulas that incorporate a distal
segment of wall-free wire mesh (Smartcanula LLC, Switzerland) are available in some
markets. These cannulas expand to a larger diameter within the vessel to minimize
ow resistance, and the distal mesh maintains vessel patency for improved drainage.
Dual-Lumen Design
Cannulae that incorporate two lumens with two drainage and a single infusion port
are a more recent design that has greatly facilitated the application of venovenous
support for respiratory failure. Although designed for percutaneous insertion, they
can be placed surgically as well. Three fundamental designs are available that have
features to support different needs.
The cavo-atrial design [1, 2] (OriGen, OriGen Biomedical) (Fig. 7.1) is inserted
via the internal jugular vein with the tip positioned in the low right atrium near the
IVC ostium. It has two drainage ports, one distal at the inferior atrium and one
proximately in the superior vena cava, with the reinfusion port in the mid right
atrium directed at the tricuspid valve. The proximity of the distal lumen to the rein-
fusion port allows some recirculation, but effective blood ow remains adequate.
Placement is somewhat easier than the bicaval design since the IVC does not have
to be accessed.
7 Vascular Access for ECLS 135
Fig. 7.1 A dual-lumen venous cannula, designed for drainage from the SVC and inferior right
atrium with reinfusion into the mid right atrium. (Reprinted with permission from OriGen
Biomedical)
The bicaval design (Avalon Elite, Maquet) requires insertion via the internal
jugular vein with the cannula traversing the right atrium and the tip positioned in the
IVC [3, 4] (Fig. 7.2). The drainage lumen extends the length of the cannula with
drainage ports in both the IVC and SVC. The reinfusion lumen is shorter, terminat-
ing in the right atrium with the reinfusion port directed toward the tricuspid valve.
This bicaval drainage design effectively separates the upper and lower venous sys-
tems and results in low recirculation with more effective blood ow.
The third design is similar to a hemodialysis catheter with a single proximal
drainage lumen and a distal reinfusion lumen [5]. This catheter is intended for low-
ow extracorporeal circuits used for carbon dioxide removal (ECCO2R; Chap. 4). If
the cannula ow exceeds the insertion vessel ow, recirculation will limit effective
ow, but placement in the internal jugular or femoral and iliac veins usually assures
adequate blood ow.
Fig. 7.3 Representative pressure-ow relationships for various size single-lumen cannulae. The
graph depicts the nonlinear relationship between ow and pressure due to a combination of lami-
nar and disturbed/turbulent ow resulting from the complex geometry of the catheter
The HagenPoisseulle equation for laminar ow, although not directly applicable
to cannula ow, does illustrate the major determinants of blood ow ( Q ):
DP r 4
Q =
m L
7 Vascular Access for ECLS 137
where P is the pressure difference, r and L are the cannula radius and length,
respectively, and is blood viscosity. Maximizing blood ow involves use of the
largest diameter cannulas that can be safely inserted, and keeping the length as short
as possible. The nonlinearity of actual pressure-ow curves most likely comes from
the tip of the cannula, which includes side holes and a tapered tip, causing a depar-
ture from purely laminar ow.
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. Without vessel sizing the
use of too large a cannula can result in venous obstruction, failure to cannulate, or
other complications such as vessel laceration or transection. Too small a cannula
can result in suboptimal blood ow and ineffective support.
Bedside ultrasound with a vascular transducer can provide high quality images
of the cervical and femoral vessels. Vessel size can be obtained by using the built-in
measurement tools and converting to the French gauge system described by Joseph-
Frdric-Benot Charrire [6] used for sizing cannulas. In the case of vessels with a
circular shape, conversion of vessel diameter in mm to French size is accomplished
with the following simple formula:
Fr = D(mm ) 3
The chosen cannula size should be slightly smaller than the measured vessel to help
assure successful placement and prevent complete obstruction of blood ow.
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, intervention-
alists, intensivists, and emergency physicians.
Percutaneous
Percutaneous cannulation has been used successfully for venovenous support [9],
and is preferred since it is associated with a lower incidence of cannulation-site
bleeding and infection. It also is non-obstructive, allowing blood ow around the
cannula. It can be used for arterial (other than carotid) as well as venous access.
The same Seldinger technique used for smaller vascular access catheters is used
for ECLS cannulae, but with multiple dilators of no more than 4 Fr difference in size
(typically 12, 16, 20, 24, 28, and 3032 Fr) with the largest size approximately
equal to the size of the cannula to be inserted. Prior to insertion, vessel size is deter-
mined with ultrasound and an appropriate size cannula is chosen. Under adequate
sedation a neuromuscular blocker is administered to prevent respiratory effort.
Under aseptic conditions and following inltration of a local anesthetic, the vessel
is identied with ultrasound and an approach is chosen to avoid injury to neighbor-
ing vessels, since vessels may overlie each other. The access needle is inserted using
ultrasound to guide it into the center of the vessel, and a .035 to .038 guidewire 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 across the atrium.
Following placement of the guidewire, a skin incision is made just large enough
to admit the cannula. The tract is then dilated sequentially to the target size. The
cannula is placed over its tapered loading dilator, and advanced into position. Using
a tubing clamp to control back-bleeding, the guidewire and dilator are removed, and
7 Vascular Access for ECLS 139
the cannula ushed with heparinized saline (2 units/mL) to maintain patency until
attached to the ECLS circuit and extracorporeal circulation has begun. The cannula
is sutured to prevent decannulation, taking care to avoid crimping the cannula or
providing a pivot point for cannula kinking.
Percutaneous cannulation of the femoral artery may result in inadequate distal
perfusion and the development lower limb ischemia. This can be managed by per-
cutaneous placement of a retrograde arterial cannula (68 Fr), or surgical cannula-
tion of the posterior tibial artery to assure adequate perfusion of the limb.
Semi-open
Open Surgical
The preferred technique for cervical cannulation when carotid arterial access is
required is the open surgical technique (Fig. 7.4). Following sedation, neuromus-
cular blockade, and skin preparation, an incision is made perpendicular to the axis
of the vessel, and carried down to expose the cervical vessels. The cannula can be
sized by visual comparison with vessel size. The vessels are freed from surround-
ing tissue, and ligatures are placed proximal and distal to control bleeding. An
arteriotomy (or 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 cannula. Following insertion to the proper depth, the ligatures are secured,
typically with pledgets to prevent vessel injury, as vessel repair may be performed
at decannulation. The subcutaneous tissue and skin are closed around the cannu-
lae, taking care to securely close the skin around the cannulae. The above descrip-
tion is generic, and variations are numerous, subject to the surgeons preferences
and experience.
If open cannulation is performed on the femoral artery which, unlike the carotid,
has no collateral circulation, then a smaller retrograde perfusion catheter is placed
140 S.A. Conrad
Fig. 7.4 Technique of surgical cannulation of the cervical vessels for extracorporeal life support.
The technique for the femoral vessels is similar. (Used with permission from [11])
Cannulation Configurations
The foremost decision regarding vascular access is the mode of support, which dic-
tates the cannulation conguration. Extracorporeal life support for both respiratory
and cardiac failure was historically performed using only a venoarterial (VA)
7 Vascular Access for ECLS 141
conguration. While still the preferred conguration for cardiac failure, other con-
gurations have been developed that are more suitable for other types of support.
Venoarterial
The venoarterial conguration drains blood from the central venous circulation and
returns it to the arterial circulation. The cervical approach is used in neonates and
infants, in whom femoral vessels are small, and since they have adequate collateral
circulation of the cerebral vessels following ligation of the carotid artery. The arte-
rial cannula is placed into the carotid artery and advanced to the proximal innomi-
nate artery. The venous cannula is placed into the internal jugular vein and advanced
into the right atrium. This conguration supplies oxygenated blood to the proximal
aorta, but coronary and right upper extremity blood may be poorly saturated if pul-
monary failure is present and the left ventricle is ejecting (Chap. 6).
Venoarterial cannulation may be performed using the femoral vessels. This con-
guration is 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
Venovenous cannulation was introduced later than venoarterial, and is suitable for
respiratory failure with adequate native cardiovascular function (Chap. 6). It provides
oxygenated blood into the venous system and uses the native heart for oxygen deliv-
ery, making oxygenated blood available to all tissues, including the myocardium.
Cannulae for venovenous support may be placed percutaneously or surgically.
The venovenous conguration was introduced to extracorporeal support using
two single-lumen cannulae, one placed into the femoral vein and advanced to the
intrahepatic inferior vena cava for drainage, and the second placed into the internal
jugular vein and advanced to the superior cavo-atrial junction. An alternative con-
guration to gain better ow and reduce recirculation was to introduce three can-
nulae, two for drainage placed at the superior cavo-atrial junction and distal IVC
respectively, and one for return placed near the inferior cavo-atrial junction.
A major advance in venovenous support was the introduction of the dual-lumen
venovenous cannula. Developed initially for neonates and infants, cannula are now
available for adult and pediatric patients. These cannulae have a single shaft, incor-
porating a drainage lumen with ports in the SVC and IVC (or low right atrium) and
a reinfusion lumen with a port in the mid-right atrium. Recirculation rates with
these cannulae are lower than with the single-lumen congurations, and are negli-
gible with the bicaval design.
142 S.A. Conrad
Veno-arterio-venous
Low-Flow Venovenous
Arteriovenous
Transthoracic
Although much more invasive, direct cannulation of the right atrium and aortic root
through a sternotomy remains an important approach to vascular access. The most
common use is support of post-cardiotomy failure to wean from cardiopulmonary
7 Vascular Access for ECLS 143
bypass (CPB), in which the cardiopulmonary circuit is replaced with the ECLS
circuit. Typically the sternum is left open and draped. The CPB cannulae are large
and support more ow than can be achieved using peripheral access.
The transthoracic approach is associated with more bleeding and infection risk,
so is generally used for patients expected to recover quickly. If prolonged support is
required, the patient may be transitioned to peripheral cannulation, or to a ventricu-
lar assist device. This approach has also been used to provide high-ow support in
patients with severe sepsis [10].
Decannulation
When extracorporeal support is no longer required, the patient is removed from sup-
port by clamping the circuit near the cannulas and removing the circuit.
Anticoagulation is discontinued prior to surgical decannulation or arterial percuta-
neous decannulation, and is held shortly before percutaneous venous decannulation.
The cannulae are ushed to prevent thrombus formation.
Percutaneous venous cannulae are removed by rst placing a purse string suture
in the incision, withdrawing the cannula, and securing the suture. Percutaneous arte-
rial cannulae are removed by withdrawing the cannula and applying pressure until
hemostasis, with care not to fully compress the artery. Arterial puncture closure
devices may be used if appropriately sized. Venous cannulae placed by the semi-
open technique are removed as if placed percutaneously. Short-term anticoagulation
or anti-platelet therapy is used to help prevent venous thrombus formation.
Surgically placed cannulae are removed with an open technique. The skin inci-
sion is re-opened, temporary ligatures placed, and the existing ligatures removed.
The vessel is either repaired or ligated, and the incision closed.
Complications of Cannulation
Recirculation
Recirculation occurs when reinfused blood is aspirated into the drainage cannula,
reducing the effective extracorporeal ow. It is unavoidable with the use of single-
lumen cannulae. Recirculation manifests as a decrease in delivered oxygen and drop
in systemic arterial saturation. Increases in recirculation can occur with displace-
ment of the cannula, and may require radiography to detect. It also increases with
increasing ow, such that high ows may actually reduce delivered oxygen.
Recirculation is less extensive with dual-lumen cannulae. The bicaval design is
associated with the lowest degree of recirculation, often under 3 %. The cavo-atrial
cannula has higher recirculation than the bicaval, but less than the use of two-site
single-lumen cannulation.
144 S.A. Conrad
Limb Ischemia
Ischemia of the lower limb is one of the major risks associated with cannulation of
the femoral artery. It can usually be managed by placement of a retrograde cannula
either just distal to the cannulation site or in the posterior tibial artery, to provide at
least 150200 mL of blood ow per minute. An alternative percutaneous strategy is
to place two smaller arterial cannulae, one in each femoral, together providing the
total ow of a single larger cannula.
If limb ischemia is not detected in time, sufcient muscle necrosis can occur that
may require fasciotomy or even amputation. Careful clinical examination, Doppler
monitoring of distal pulses, and plethysmographic assessment with pulse oximetry
can help identify this condition early. Many will routinely place a retrograde can-
nula at the time of cannulation to minimize this risk.
Vascular Injury
Injury to the target or adjacent vessel during cannulation can result in inability to
achieve vascular access, transection of a vessel, hemorrhage into areas such as the
retroperitoneal space, exsanguination, and death. Immediate attempts at surgical
repair and completion of cannulation are required, but may not be successful. The
risk is higher with percutaneous cannulation since the vessels are not visible. The
use of ultrasound can mitigate these risks, by allowing for appropriate cannula size,
identication of adjacent vessels, selection of an approach, and guidance of the
puncture to ensure proper entry into the vessel.
Inadequate Flow
The inability to achieve the expected ow can result in the inability to achieve
adequate cardiac support (venoarterial) or persistent hypoxemia and inability to
achieve lung protective settings (venovenous), decreasing the chance of survival.
7 Vascular Access for ECLS 145
Infectious Complications
Infection of the cannulation insertion site is a challenging problem, since the patient
may be totally dependent on extracorporeal support and recannulation may be risky
or impossible. Prevention by good skin asepsis at the time of insertion and during
extracorporeal support is important to minimizing this risk. If infection does develop
and appears to be localized, then use of appropriate antibiotics may be successful.
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 may be
required.
References
1. Andrews AF, Zwischenberger JB, Cilley RE, Drake KL. Venovenous extracorporeal mem-
brane oxygenation (ECMO) using a double-lumen cannula. Artif Organs. 1987;11(3):2658.
2. Anderson 3rd HL, Otsu T, Chapman RA, Barlett RH. Venovenous extracorporeal life support
in neonates using a double lumen catheter. ASAIO Trans. 1989;35(3):6503.
3. Wang D, Zhou X, Liu X, Sidor B, Lynch J, Zwischenberger JB. Wang-Zwische double lumen
cannula-toward a percutaneous and ambulatory paracorporeal articial lung. ASAIO
J. 2008;54(6):60611.
4. 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):9915.
5. 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):13827.
146 S.A. Conrad
6. Iserson KV. J.-F.-B. Charriere: the man behind the French gauge. J Emerg Med.
1987;5(6):5458.
7. 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):27781.
8. Extracorporeal Life Support Organization Task Force on Infections. Infection control and
extracorporeal life support 2010. http://elso.org/downloads/resources/committees/infectious-
disease-and-antibiotic/Infection-Control-and-Extracorporeal-Life-Support.pdf.
9. Pranikoff T, Hirschl RB, Remenapp R, Swaniker F, Bartlett RH. Venovenous extracorporeal
life support via percutaneous cannulation in 94 patients. Chest. 1999;115(3):81822.
10. Maclaren G, Butt W, Best D, Donath S, Taylor A. Extracorporeal membrane oxygenation for
refractory septic shock in children: one institutions experience. Pediatr Crit Care Med.
2007;8(5):44751.
11. Field ML, Al-Alao B, Mediratta N, Sosnowski A. Open and closed chest extrathoracic can-
nulation for cardiopulmonary bypass and extracorporeal life support: methods, indications,
and outcomes. Postgrad Med J. 2006;82(967):32331.
Chapter 8
Circuits, Membranes, and Pumps
Bradley H. Rosen
Introduction
Circuit Anatomy
Circuit designs all attempt to balance efcacy, safety, convenience, and simplicity.
There is no one-size-ts-all solution, however, since varied patients, circumstances,
and clinician preferences may necessitate that safety override simplicity or that por-
tability trump efcacy. For example, inserting multiple stopcocks into the circuit
Fig. 8.1 Schematic of a complex circuit design that implements all optional features, including a
compliance chamber, separate non-integrated blood analyzers, a manifold with access sites (closed
unless being accessed), and a bridge
Fig. 8.2 Similar to the previous gure, but simplied with only necessary components: pump,
oxygenator, and ow probe/bubble sensor. The blood analyzers are internalized within the pump
and oxygenator
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 physicians experience and preference with regard to
circuit complexity. Thus, circuits range from rather complex designs incorporating
many safety and monitoring functions (see Fig. 8.1) to minimalistic, simpler layouts
that lack the various bells and whistles (Fig. 8.2).
In designing a circuit, simplicity is one of the paramount concerns. While com-
ponents and connectors can be cut into a circuit after purchase, each modication
produces a weak point susceptible to rupture or brin accumulation due to turbu-
lence. Any such alteration should be performed while the circuit is dry (prior to
priming, see below), and with regard for sterility. Additionally, each Luer lock is a
site of potential air entrainment, blood leak, or microbial contamination. The major-
ity of connections and access ports are located on the venous side of the circuit,
between the pump and the oxygenator. This is intentional: the lack of connectors on
the arterial side reduces the potential for accidental exsanguination, while the simi-
lar lack of connectors proximal to the pump inlet limits the risk of air entrainment
and gas embolism.
8 Circuits, Membranes, and Pumps 149
All circuits should involve as little tubing as possible, while allowing adequate
spacing of components and facilitating mobilization of the patient. Greater tubing
lengths incur more resistance to ow (proportional to length according to Poiseuilles
law), necessitating higher circuit pressures and leading to more damage to blood
elements.
Circuit length and complexity also relate to the degree to which blood is exposed
to plastic surfaces and this interaction elicits an inammatory response. It is believed
that induced inammation may further compromise lung function, leading to fur-
ther gas exchange deterioration. It has been hypothesized that heparin coating of
polymethylpentene (PMP) oxygenators serves to reduce this response [1]. An added
consequence of circuit-induced inammation is excessive brinogen production
leading to increased brin deposition on circuit surfaces. Further, this inammation
promotes platelet adherence, elevating the risk of thrombosis which impairs oxy-
genator function [2]. Cellular deposition along the membrane surface (on the blood
side) correlates with a rising resistive pressure across the device [3]. Some brin
deposition within the oxygenator and circuit is unavoidable (apparent rst on the
venous side), but excessive deposition is deleterious to circuit function.
There are several additional implications of circuit length. The greater the surface
area, the more that medications commonly used in the care of critically ill patients
are subject to adsorption. Antibiotics (meropenem, cefazolin, and vancomycin),
sedatives (midazolam), and analgesics (morphine, fentanyl, and acetaminophen) are
all meaningfully adsorbed, to a degree related to the lipophilic nature of the drug.
Antibiotics are only moderately affected (6585 % recovered after 180 min), but
midazolam and fentanyl are severely adsorbed with less than 1 % recovered [4].
Even if an agent is not adsorbed, the extracorporeal circuit expands the volume of
distribution of any pharmaceutical due to the volume of blood within the circuit itself
(up to 1 L). Tubing length will also contribute signicantly to the volume required to
prime the circuit, producing hemodilution. Finally, tubing surface area also relates to
the degree of heat loss. This can be substantial, such that ECLS circuits must incor-
porate a means for temperature control (see Heat Exchanger below).
Circuit Priming
Priming refers to the process by which the gas (ambient air present at manufacture)
is replaced with a physiologically compatible uid. For ECLS in adults, the circuit
is primed with crystalloid uids, such as normal saline, Ringers lactate, or propri-
etary mixed electrolyte solutions (e.g., Plasma-Lyte or Normosol). Purchased cir-
cuits come attached to a large, empty priming bag. The bag is lled with sterile
crystalloid, clamps are opened to the venous and arterial limbs, and the priming bag
is raised to allow gravity to move the uid 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, bridge, manifold, and tubing) and directly relates to the degree of
150 B.H. Rosen
hemodilution that follows. For pediatric and neonatal ECLS, hemodilution is pre-
vented by priming the circuit with blood, but this is not necessary for adults where
the typical priming volume averages 7501000 mL for a complex circuit design and
for a simple 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 regard-
ing shelf life. Priming with a colloid may shorten the shelf life of a primed circuit,
another reason many programs choose a crystalloid prime. A simplied 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 rst with carbon dioxide
(to displace oxygen and nitrogen), hastening the subsequent uid 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 rushes 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 8.1 represents a comprehensive schematic of
a circuit, whereas Fig. 8.2 shows a greatly simplied design with few extraneous
components. In each instance, we describe the circuit beginning with the outow
(venous) cannula, proceeding through the gas-exchanging membrane, and ending
back at the patient through the inow cannula, which may enter an artery (venoarte-
rial or VA ECLS) or vein (venovenous or VV ECLS). Sometimes used for venove-
nous ECLS, dual-lumen cannulas allow blood to exit and enter at the same site but,
for illustration purposes, we have separated these in the gures.
Starting with the outow 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
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. 10). The conduct-
ing tubing should be kept relatively short in order to reduce resistance to blood ow,
surface area of contact with blood, and the priming volume. The conducting tubing
leads to a centrifugal pump (in some designs a compliance chamber or bladder pre-
cedes the pump as in Fig. 8.1) 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 oxygen), being joined to wall oxygen or an E-cylinder through a ow
8 Circuits, Membranes, and Pumps 151
meter. The newly arterialized blood completes its extracorporeal course through the
inow cannula, delivering oxygenated and warmed blood to the vascular system.
Outow and inow cannulas may be bridged directly by a length of large-diameter
tubing (the bridge; see Fig. 8.1) connected through two high-ow stopcocks.
When opened, the bridge offers a shunt to keep blood owing within the circuit
while clamps are used to isolate the patient. In so doing, the clinician can judge
whether the patient has recovered sufciently to sustain respiration and circulation
without ECLS (see Chap. 13). This is important during VA ECLS, but a bridge is not
needed for VV support since weaning can be conducted by reducing or eliminating
gas ow to the membrane while leaving circuit ow to the patient undisturbed.
During most ECLS operation the bridge remains closed and, because any blood
within is stagnant, blood is generally displaced by saline when the bridge is closed.
In order to monitor circuit function and to prevent complications, devices to mea-
sure pressure and ow and to detect bubbles are included, and information is relayed
to a console. Typically, pressure is measured on the venous side of the pump (P1),
providing information about how much suction is required to draw the needed circuit
blood ow. Two additional pressure transducers (P2 and P3) ank the oxygenator
so that its resistance can be estimated based on the drop in pressure across the
membrane(delta-P). In addition, P3 displays the pressure that drives ow back to
the patient. Circuit blood ow is monitored using an ultrasonic ow probe, since
centrifugal pumps do not guarantee a xed relationship between revolutions per min-
ute and volume displaced, as was true for roller pumps. The ow probe may be inte-
grated 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 ow
measurement and bubble detection. Bubbles distal to the oxygenator can produce
systemic embolism and are especially dangerous in VA modes.
Spectrophotometric sensors allow real-time measurement of such values as PO2,
PCO2, pH, SaO2, SvO2, and hemoglobin concentration, among others. These sensors
must be calibrated periodically by comparing the displayed value against a blood sam-
ple analyzed simultaneously using conventional laboratory methods. The console
receives data from various devices along the circuit, displaying pump speed, ow, pres-
sures, temperature, and other physiological information. The console also may display
alarm notications 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 lock
connections with a three-way stopcock controlling ow to each. The manifold
derives from the region 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 oxygenator. Various infusions of medica-
tions and anticoagulant agents may be connected to the circuit through these ports,
but more often, other vascular access is utilized (see Chaps. 7 and 10). Sufcient
ow 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.
152 B.H. Rosen
Single- and double-lumen cannulas are described more fully in Chap. 7. 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 rec-
ognize blood color (as a clue to circuit function and recirculation) and to identify
brin, 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 ow.
Compliance Chamber
This device was previously employed when roller head pumps were more common,
as a safety device to dampen any excessive negative pressure generated by the
pump, rather than causing cavitation or hemolysis in the patient. Essentially an
external venous reservoir, this device may also provide information regarding rela-
tive hypovolemia. Collapsing of the compliance chamber would suggest to the clini-
cian that ow through the circuit be slowed or additional uid volume be
administered. With broad use of centrifugal pumps, compliance chambers are gen-
erally felt to be an unnecessary complexity.
The simplest design is a silicone bladder with inlet and outlet ports, placed
between the outow cannula and the pump (Fig. 8.1). A pressure transducer can be
used to signal an alarm or to slow or shut off the pump if negative pressure reaches
a degree that could result in cavitation within the venous system. Various designs
have different port sizes, priming volumes, and orientations. One device is a verti-
cally oriented inline reservoir consisting of a compliance balloon housed within a
PVC chamber. This obviates placing the device on the ground and the lengths of
tubing to the bladder and from it. The vertical orientation may also allow for a more
constant ow assisted by gravity, reducing likelihood of settling blood and throm-
bosis, as well as entrainment of gaseous bubbles at the top of the chamber.
Additionally, this device can be heparin-coated and is FDA-approved for use in
ECLS. It is offered in a standard size with 20 mL priming volume and -in. ports,
as well as a larger version with 115 mL priming volume and 3/8-in. ports [5].
Pumps
There are two types of pumps employed in ECLS circuits: roller/occlusive and cen-
trifugal pumps. Practitioners of modern ECLS have settled on the centrifugal pump
design as the safer of the two.
8 Circuits, Membranes, and Pumps 153
Roller Head (Occlusive) Pumps: The pump itself has two roller heads within an
enclosure and one head contacts (and variably occludes) the tubing at all times.
These roller heads must be properly adjusted to accurately account for blood ow
[6]. They are placed 180 to each other, to pull blood from the venous limb and
simultaneously push it forward along the circuit path. The length of tubing within
the enclosure is termed the raceway, and that section is advanced or withdrawn
periodically to avoid foci of excessive wear on the tubing. This necessitates a signi-
cantly longer circuit when a roller pump is employed, and accordingly a larger prim-
ing volume. Another limitation that roller pumps impose on circuit design is that
they must form the base of the circuit, requiring more extensive lengths of tubing.
Outside of restricting circuit design, roller head pumps have intrinsic limitations
arising from the fact that the device operates on the principle of positive uid dis-
placement. This allows it to generate excessive positive pressures if, for example,
there is a kink in the outow tubing, or very negative suction (due to intravascular
hypovolemia). It was this attribute that mandated compliance chambers when roller
head pumps were more commonly used. A high positive pressure was required to
drive blood through older silicone rubber membrane lungs, but newer, low-resistance
membranes make this unnecessary.
Flow is calculated based upon the length and diameter of the tubing within the
raceway, combined with the number of pump revolutions. Higher ow (more revo-
lutions) leads to increased wear on tubing and can cause the liberation of micro-
scopic particles of tubing due to material fatigue, termed spallation, and resulting
microembolism to the patient [7]. Specic types of tubing are recommended for use
within the raceway due to their resistance to wear compared with other tubing [8].
In case of emergency events, the pump may be operated manually by a hand-
crank and newer models are equipped with an internal battery backup of limited
duration. Advantages of roller pumps include lower cost; lower direct pump prime
volume (although more tubing is necessary); and afterload-independent ow.
Disadvantages include the need for longer tubing; location at the base of the circuit
(in current designs manufactured); spallation and microembolism; and challenges
in properly setting the occlusion.
Centrifugal pumps: These devices increasingly employ a magnetically driven
impeller within a spiral housing. The impeller 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 ow toward the circumfer-
ence 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 ow is independent of afterload (to the point of causing tubing rupture!),
a centrifugal pump is unable to overcome excessive afterload. Instead, ow 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 outow cannula, as it
is unable to generate sufciently negative pressure. Of course, this means that hypo-
volemia tends to threaten the adequacy of circuit ow when a centrifugal pump is
used. This lack of absolute relationship between pump revolutions and blood ow
necessitates a ow meter. An increasing discrepancy between pump speed and mea-
sured ow is a strong signal of trouble with regard to function of the circuit.
154 B.H. Rosen
Should there be a complete loss of power, these pumps can also be hand-cranked.
Internal batteries provide up to 90 min of support without an external power source.
Advantages of centrifugal pumps include reduced tubing length; safety benets
due to absent issues of spallation, microemboli, and raceway rupture; and greater ex-
ibility in circuit design because the pump does not need to be at the base of the unit.
Disadvantages include direct priming volume required for the pump (although
tubing length is reduced) and blood ow that depends on preload and afterload
(requiring a blood ow probe). Centrifugal pumps were initially reported to produce
unacceptable rates of hemolysis due to heat generation, but engineering improve-
ments have solved this problem [9]. Designs have evolved to reduce hemolysis and
the risk of gaseous microembolism [10, 11], changing the landscape of ECLS cir-
cuits and leading to worldwide adoption of the technology [12, 13].
Membrane Oxygenators
The ideal membrane lung would be highly permeable to relevant gases (O2 and
CO2) while resisting uid transudation from the blood to the gas phase (termed
plasma leak). Blood should ow through the device with minimal resistance,
allowing high ows 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 (see Chap. 14).
Both PMP and polypropylene hollow-ber membranes employ large numbers of
ne capillary tubes to carry the sweep gas while being bathed by owing blood.
This extra-capillary blood ows countercurrent to the direction of gas movement,
increasing the efciency 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 2 m2 of gas exchange surface while the
lung exposes upwards of 70 m2 to blood ow. Additionally, in the best membrane
lungs, oxygen must diffuse 150 m from sweep gas to blood, whereas the compa-
rable distance within the lung is a mere 0.5 m. These disadvantages are offset by
increasing the effective dwell time of the blood within the articial lung. In addi-
tion, so called secondary ows, which describe the mixing of blood around the
gasuid interface due to purposefully created turbulence, further enhances gas
transfer. Blood cells are regularly being brought into close approximation with the
gas-lled capillaries, effectively reducing diffusing distance [14, 15]. This layout
allows for up to a two and a half-fold reduction in surface area necessary for gas
exchange [16]. Typical gas-exchanging capacities for membrane lungs are shown in
Figs. 8.3 (oxygen) and 8.4 (carbon dioxide). Providing sufcient oxygen transfer to
8 Circuits, Membranes, and Pumps 155
Fig. 8.4 Carbon dioxide transfer in mL/min as functions of both blood ow in L/min and sweep
gas ow
meet the entire metabolic demand (roughly 250 mL O2/min) requires blood ow of
roughly 4 L/min. In contrast, carbon dioxide transfer is relatively advantaged; so
that much less blood ow is required, especially at very high sweep gas ow rates.
For example, nearly all of the metabolically produced carbon dioxide can be elimi-
nated with only 1 L/min of blood ow, especially at high gas ows [17].
The most common external appearance of PMP oxygenators is an extruded
square evenly balanced on one corner (see Fig. 8.5). At the lower corner, blood
enters the device from the pump under pressure denoted as pre-membrane or
inlet pressure (typical pressures are in the range of 225275 mmHg, certainly less
than 400 mmHg; see Table 8.1). The blood ascends to the opposite corner at the
highest elevation of the device and then ows 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 or outlet pressure) will be
156 B.H. Rosen
Table 8.1 Circuit pressures Location in the circuit Normal operating pressure
Proximal to the pump: P1 100 to 200 mmHg
Oxygenator inlet: P2 225275 mmHg
Oxygenator outlet: P3 190260 mmHg
Delta-P: (P2 minusP3) 1035 mmHg
less than the inlet pressure. The difference between these is called the delta pres-
sure (or delta-P), representing the resistive pressure drop across the membrane at
the current ow. With the current generation of PMP devices, delta-P should range
from the teens to low 30s at typical circuit blood ow rates.
With regard to the pressures and the information that may be gleaned from their
trends, an elevated oxygenator inlet pressure (P2) has variable implications depend-
ing on whether the outlet pressure (P3) is also elevated. In the case where both are
elevated (delta-P is preserved), the circuit should be examined, not the oxygenator:
the distal tubing and circuit may be obstructed by kinks or thrombosis. Similar nd-
ings are seen transiently when patients cough, Valsalva, or are suctioned. When
inlet pressure rises along with an increase in delta-P, resistance within the mem-
brane is excessive, raising concerns for thrombosis, heparin-induced thrombocyto-
penia, or accumulation of brin or cellular elements on the membrane. Ultimately,
this portends failure of the membrane.
Another major advance in modern PMP oxygenators is their low resistance to
blood ow, producing several important advantages. First, this property ushered in
the era of lower pressure, afterload-sensitive centrifugal pumps, affording the safety
8 Circuits, Membranes, and Pumps 157
Rated Flow: Blood exiting the membrane is normally fully oxygenated, typically
with a PO2 in excess of 300 mmHg. As blood ow is increased, greater demands are
placed on the capacity for gas diffusion across the hollow ber barrier. In part, this
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 ows. At sufciently high ows, the
membrane fails to fully saturate the blood. The ow threshold for full oxygenation
is termed the rated ow.
Plasma Leak: Hollow ber membranes should be sufciently permeable to allow
rapid gas diffusion, while remaining impermeable to uid movement. Plasma leak
is the phenomenon whereby plasma phospholipids leak from the circulating whole
blood to the gas compartment of the oxygenator, then serve to propagate further
plasma leakage in a positive feedback cycle [21]. Small amounts of uid will nor-
mally traverse the membrane and a drainage port is provided for egress. Greater
volumes signal a failing membrane, severely impairing the efciency of gas
exchange, accompanied by a rise in delta-P, and eventually requiring exchange of
the device. Important plasma leak can be conrmed by analysis of liquid from the
gas outlet for proteins [22]. Additional implications are a signicant loss of proteins
158 B.H. Rosen
While on ECLS, the patients blood is exposed to the environment of the outside
world, risking substantial heat loss. Left unchecked, this could produce coagulopa-
thy, circulatory failure, and other organ dysfunction. The heat exchanger consists of
8 Circuits, Membranes, and Pumps 159
Acknowledgment The author thanks the following content consultants: Jennifer Crumley,
M.S.N.; Tom Rath, C.C.P.; and Elizabeth Moore, M.S.N.
References
1. Khoshbin E, Dux AEW, Killer H, Sosnowski AW, Firmin RK, Peek GJ. A comparison of radio-
graphic signs of pulmonary inammation during ECMO between silicon and poly-methyl pen-
tene oxygenators. Perfusion. 2007;22:1522.
2. Peek GJ, Firmin RK. The inammatory and coagulative response to prolonged extracorporeal
membrane oxygenation. ASAIO J. 1999;45:25063.
3. Lehle K, Philipp A, Gleich O, Holzamer A, Muller T, Bein T, et al. Efciency in extracorporeal
membrane oxygenation-cellular deposits on polymethylpentene membranes increase resis-
tance to blood ow and reduce gas exchange capacity. ASAIO J. 2008;54:6127.
4. Wildschut ED, Ahsman MJ, Allegaert K, Mathot RAA, Tibboel D. Determinants of drug
absorption in different ECMO circuits. Intensive Care Med. 2010;36:210916.
5. Tamari Y, Lee-Sensiba K, King S, Hall MH. An improved bladder for pump control during
ECMO procedures. J Extra Corpor Technol. 1999;31:8490.
6. Snyder EJ, Harb HM, Cullen JA, McElwee DL. Setting roller pump occlusion with the
Transonic HT109 owmeter. ASAIO J. 1997;43:604.
7. Peek GJ, Thompson A, Killer HM, Firmin RK. Spallation performance of extracorporeal
membrane oxygenation tubing. Perfusion. 2000;15:45766.
160 B.H. Rosen
8. Peek GJ, Wong K, Morrison C, Killer HM, Firmin RK. Tubing failure during prolonged roller
pump use: a laboratory study. Perfusion. 1999;14:44352.
9. Lawson DS, Ing R, Cheifetz IM, Walczak R, Craig D, Schulman S, et al. Hemolytic character-
istics of three commercially available centrifugal blood pumps. Pediatr Crit Care Med.
2005;6:5737.
10. Burnside J, Gomez D, Preston TJ, Olshove VF, Phillips A. In-vitro quantication of gaseous
microemboli in two extracorporeal life support circuits. J Extra Corpor Technol.
2011;43:1239.
11. Yee S, Qiu F, Su X, Rider A, Kunselman AR, Guan Y, et al. Evaluation of HL-20 roller pump
and Rotaow centrifugal pump on perfusion quality and gaseous microemboli delivery. Artif
Organs. 2010;34:93743.
12. Lawson DS, Lawson AF, Walczak R, McRobb C, McDermott P, Shearer IR, et al. North
American neonatal extracorporeal membrane oxygenation (ECMO) devices and team roles:
2008 survey results of Extracorporeal Life Support Organization (ELSO) centers. J Extra
Corpor Technol. 2008;40:16674.
13. Lawson DS, Walczak R, Lawson AF, Shearer IR, Ing R, Schulman S, et al. North American
neonatal extracorporeal membrane oxygenation (ECMO) devices: 2002 survey results. J Extra
Corpor Technol. 2004;36:1621.
14. Drinker PA, Bartlett RH, Rishon MB, Noyes BS. Augmentation of membrane gas transfer by
induced secondary ows. Surgery. 1969;66:77581.
15. Diller TE, Mikic BB, Drinker PA. Shear-induced augmentation of oxygen transfer in blood. J
Biomech Eng. 1980;102:6772.
16. Gaylor JDS. Membrane oxygenators: current developments in design and application. J
Biomed Eng. 1988;10:5417.
17. Mueller T, Lubnow M, Philipp A, Bein T, Jeron A, Luchner A, et al. Extracorporeal pumpless
interventional lung assist in clinical practice: determinants of efcacy. Eur Respir
J. 2009;33:5519.
18. Johnson P, Frohlich S, Westbrook A. Use of extracorporeal membrane lung assist device
(Novalung) in H1N1 patients. J Card Surg. 2011;26:44952.
19. Kolobow T, Bowman RL. Construction and evaluation of an alveolar membrane articial
heart-lung. Trans Am Soc Artif Intern Organs. 1963;9:23843.
20. Khoshbin E, Roberts N, Harvey C, Machin D, Killer H, Peek GJ, et al. Poly-methyl pentene
oxygenators have improved gas exchange capability and reduced transfusion requirements in
adult extracorporeal membrane oxygenation. ASAIO J. 2005;51:2817.
21. Montoya JP, Shanley CJ, Merz SI, Bartlett RH. Plasma leakage through microporous mem-
branes: role of phospholipids. ASAIO J. 1992;38:M399405.
22. Eash HJ, Jones HM, Hattler BG, Federspiel WJ. Evaluation of plasma resistant hollow ber
membranes for articial lungs. ASAIO J. 2004;50:4917.
23. Puis L, Ampe L, Hertleer R. Case report: plasma leakage in a polymethylpentene oxygenator
during extracorporeal life support. Perfusion. 2009;24:512.
24. Fosse E, Moen O, Johnson E, Semb G, Brockmeier V, Mollnes TE, et al. Reduced complement
and granulocyte activation with heparin-coated cardiopulmonary bypass. Ann Thorac Surg.
1994;58:4727.
25. Jansen PGM, Velthius H, Huybregts RAJM, Paulus R, Bulder ER, van der Spoel HI, et al.
Reduced complement activation and improved postoperative performance after CPB with
heparin-coated circuits. J Thorac Cardiovasc Surg. 1995;110:82934.
26. Pekna M, Hagman L, Halden E, Nilsson UR, Nilsson B, Thelin S. Complement activation dur-
ing cardiopulmonary bypass: effects of immobilized heparin. Ann Thorac Surg.
1994;58:4214.
27. Gravlee GP. Heparin-coated cardiopulmonary bypass circuits. J Cardiothorac Vasc Anesth.
1994;8:21322.
28. Gu YJ, van Oeveren W, Akkerman C, Boonstra PW, Huyzen RJ, Wildevuur CRH. Heparin-
coated circuits reduce the inammatory response to cardiopulmonary bypass. Ann Thorac
Surg. 1993;55:91722.
8 Circuits, Membranes, and Pumps 161
The wisdom of old men. They do not grow wise. They grow
careful.
From Farewell to arms by Ernest Hemingway
Introduction
severely diseased lungs have a chance to heal only if the environment remains con-
ducive 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 hypo-
perfusion with severe and lethal lung tissue alkalosis [5]. He had developed in his
laboratory at the NIH a prototype that became the rst commercially available mem-
brane lung, and he was one of the most experienced scientists in this eld [6]. He
pointed out that providing viable blood gases by extracorporeal means was not
enough to foster lung healing. The approach to ventilation should be reevaluated to
take into account that gas exchange via the native lung was no longer essential. High
airway pressures and high tidal volume, a must to optimize gas exchange through
the native lung, were postulated to cause what later became known as VILI [7].
The goal of ECMO was redened from buying time for the lung to heal [8] to
rest the lung [9]. In the same years, Gattinoni and colleagues introduced a novel
ECLS technique called low-frequency positive-pressure ventilation with extracor-
poreal carbon dioxide removal (LFPPV ECCO2R) [10]. The stage was set for a new
approach to the ventilator management of ECLS patients. Over the next 30 years,
ECLS evolved from pure hypoxemia rescue 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]. Still, how best to ventilate ARDS patients undergoing ECLS is based on
scanty data, derived mainly from observational 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 circuit blood ow, hemoglobin concentration (Hb), and oxyhemo-
globin saturation of the blood entering the membrane lung. For any given intrapul-
monary shunt (proportion of the cardiac output that crosses the lung and leaves it
with an unchanged oxygen content equal to the mixed venous one), VV ECMO will
increase the arterial oxygen content in proportion to how much it can raise the
mixed venous oxygen content. For example, if a patient has a mixed venous oxygen
saturation (SvO2) of 50 % and intrapulmonary shunt fraction of 50 %, then the arte-
rial oxygen saturation (SaO2) will roughly be 75 % (given an arteriovenous differ-
ence of 25 %). If ECMO can oxygenate a substantial (say 50 %) proportion of
venous blood, SvO2 rises to 75 % and SaO2 will be the average of 75 % and 100 %
9 Ventilator Management During ECLS 165
(approx 87.5 %; not taking into account dissolved O2). It becomes obvious that the
higher the ratio (extracorporeal blood ow/cardiac output), the higher the effect on
the mixed venous and therefore the arterial oxygen saturation (this is one of the
reasons to advocate -blockers for some patients undergoing VV ECMO) [12].
Real life is much more complicated than that, mainly because of the negative
effect of bypass recirculation (which increases by increasing blood ow), and
because intrapulmonary shunt is not xed, but rather tends to increase along with
SvO2 as hypoxic vasoconstriction is relieved.
The ability to provide oxygen through ECMO is limited by the fact that venous
blood already contains much oxygen (patients normally have, and probably require,
a mixed venous oxygen saturation around 70 %) and there is only so much hemo-
globin available to bind oxygen. This basic principle [13] shows why oxygenation
depends mainly on circuit blood ow and why this ow must be very similar to the
normal cardiac output.
Carbon Dioxide
In contrast, CO2 removal requires much lower blood ow [13]. This is due to the
very high carbon dioxide content of blood (mostly carried as bicarbonate ion). The
carbon dioxide content of normal venous blood is approximately 600 mL/L (50 %
more in compensated chronic hypercapnia), meaning that 1 L of blood contains two
to three times the entire bodys metabolic CO2 production. Total CO2 removal (i.e.,
the removal of the minute CO2 production) can be achieved at a blood ow in the
range of 0.5 L/min, as opposed to the 5 L/min required to provide the 250 mL/min
of oxygen consumed physiologically.
In the second half of the 1970s, the pioneering work of Kolobow and Gattinoni
provided the pathophysiological foundations to suggest a clinical role for
ECCO2R. This extraordinary pair of scientist-inventor and scientist-clinician real-
ized that the membrane oxygenator was indeed a membrane lung, exchanging both
oxygen and CO2. In a little more than a year they (a) described a membrane lung
optimized for CO2 removal (the so-called CDML: the carbon dioxide membrane
lung) [14]; (b) showed that it was possible to control the ventilation of an awake
sheep by removing CO2 at incremental rates [15], even to complete apnea [16]; and
(c) developed a mode of ventilation that would maintain lung volume while using
very low respiratory rates [17]. They claried that, during apneic oxygenation, alve-
olar PO2 is a function of alveolar PN2, in turn at equilibrium with the PN2 of the gas
ventilating the membrane lung [16]. Another observation dened the role of the
ratio of CO2 eliminated and oxygen consumed by the natural lung (respiratory quo-
tient, RQ) in determining alveolar PO2, of particular importance in the spontane-
ously breathing subject. This stresses the need to raise FiO2 to maintain the alveolar
fraction of oxygen (FaO2) when the alveolar gas equation is faced with RQs much
lower than 1 (as is the case when CO2 is also removed by the membrane lung).
When no CO2 is exchanged through the natural lung (apneic oxygenation), FaO2 can
be maintained constant only when the FiO2 is 1 [18].
166 A. Pesenti et al.
While ECLS has been mainly considered as a rescue treatment for refractory hypox-
emia [11], the considerable amount of CO2 removed by the articial lung allows
dramatically reducing the load imposed on the natural lung by mechanical ventila-
tion and implementing a truly protective ventilatory strategy (low pressure, tidal
volume, and respiratory rate) [19]. These reductions will lower mean airway pres-
sure, potentially causing alveolar derecruitment. If necessary, this can be prevented
by an appropriate increase in positive end-expiratory pressure (PEEP). While these
principles seem evident, no consensus or formal evidence exists on how to set the
ventilator during the early oxygenation rescue phase of ECMO treatment (Table
9.1) [2022].
Three major goals are important in supporting the severely hypoxemic ARDS
patient: (a) rescue from hypoxia; (b) lung recruitment (open lung strategy); and (c)
VILI prevention (lung rest). During conventional mechanical ventilation a high
price is paid in terms of airway pressures (PEEP, plateau, and mean airway pres-
sure) to maintain arterial oxygenation, while alveolar ventilation must be sufcient
to eliminate CO2. VV-ECMO is capable of providing the entire oxygen consump-
tion, even when native lung oxygen transfer is nil (100 % intrapulmonary shunt). To
achieve this, circuit blood ow must be maximized and relatively low arterial oxy-
gen saturation levels are to be accepted, sometimes around 80 %. While this level of
oxygenation might not be better than before bypass, at least it affords viable oxy-
genation in concert with lung protection. However, a compromise is often recom-
mended (Table 9.1), mostly because some uncertainties exist as to whether a
recruited lung will heal faster than a collapsed one. Airway pressures are set at
Table 9.1 Guidelines proposed by different ECMO centers to guide ventilation settings in the
early ECMO phase
Ventilation Tidal volume RR (per PEEP
Guidelines mode (mL/kg) min) (cmH2O) FiO2
ELSO (ref. 20) PCV To reach Peak 45 10 Not reported
of 20 cmH2O
CESAR trial (ref. 21) PCV To reach Peak 10 1015 0.3
of 2025 cmH2O
Karolinska (ref. 22) PCV or PSV To reach Peak of Not 510 0.4
2025 cmH2O reported
RR respiratory rate, PEEP positive end-expiratory pressure, FiO2 inspired O2 fraction, PCV
pressure-controlled ventilation, PSV pressure support ventilation
9 Ventilator Management During ECLS 167
Table 9.2 Consensus on how to set ventilation during the early ECMO phase is lacking: this table
reports main ventilator settings from three different published case series
Tidal volume RR (per PEEP
(mL/kg) min) (cmH2O) FiO2
Brogan et al. Before ECLS Not reported 20 [15;25] 12 [10;17] 1 [1;1]
(ref. 26) 24 h of ECLS Not reported 10 [10;15] 10 [8;14] 0.5 [0.4;0.5]
Zimmermann Before ECLS 6.6 [5.3;7.2] 25 [22;27] 17 [14;20] 1 [0.8;1]
et al. (ref. 27) 24 h of ECLS 4.4 [3.4;5.4] 21 [18;26] 17 [14;20] 0.7 [0.6;0.9]
Patroniti Before ECLS 6.2 [4.7;7.7] 28 [20;33] 16 [14;19] 1 [1;1]
et al. (ref. 28) 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
extracorporeal lung support
hypoxemic patients with severe right ventricular dysfunction or failure. Severe pul-
monary hypoperfusion may however favor thrombosis of the pulmonary circulation,
as originally suggested by Ratliff [30]. Hyperventilation of hypoperfused regions is
a hallmark of ARDS [31], and VA bypass might contribute to worsen it, leading to
severe tissue alkalosis and hemorrhagic edema [32]. In summary, protective ventila-
tion is all the more important during VA ECMO.
Prone Positioning
HFOV combines a bias ow and an oscillating piston to deliver tidal volumes lower
than the anatomical dead space (around 12 mL/kg) at very high respiratory fre-
quency (usually 36 Hz in adults) [54]. The rapid oscillations of gas are delivered
9 Ventilator Management During ECLS 171
above and below the mean airway pressure (Paw), usually set 5 cmH2O higher than
the mean observed during conventional ventilation. Oxygenation depends on the set
level of Paw and FiO2, while CO2 removal is inuenced by the pressure amplitude
and frequency of oscillation [54]. Compared to conventional ventilation, the small
tidal volumes should limit alveolar overdistension while the higher Paw promotes
alveolar recruitment and limits cyclical alveolar collapse, thus improving gas
exchange and lung protection.
HFOV is used extensively in neonates with respiratory distress syndrome [55],
while the experience in adult ARDS patients is more limited and controversial.
Recently, the results of two large, multicenter, randomized trials comparing HFOV
to conventional ventilation in adult ARDS patients have been published. The
Oscillation for Acute Respiratory Distress Syndrome Treated Early (OSCILLATE)
trial was stopped after inclusion of 548 of a planned 1200 patients because the mor-
tality of patients treated with HFOV was signicantly 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 [56]. 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 %) [57]. Despite some important methodological differences, the results of
these trials argue against widespread application of HFOV in ARDS patients.
Subjects enrolled, however, were less hypoxemic than would typically qualify for
rescue measures, so these results leave open the possibility that HFOV could be
benecial.
Few case reports describe the application of HFOV during ECMO. Banach and
coauthors reported the case of a 35-year-old man with lobar pneumonia and
refractory hypoxemia who required discontinuation of ECMO because of hemor-
rhagic complications: ECMO was then replaced by HFOV coupled with pumpless
arteriovenous extracorporeal lung assist (PECLA) for CO2 removal [58].
Muellenbach and colleagues showed in both an animal model [59] and in a patient
with post-traumatic ARDS [60] that the combination of an arteriovenous extracor-
poreal lung assist with HFOV allowed the use of oscillatory frequencies higher
than those usually applied (up to 1015 Hz), thus minimizing the risk of baro- and
volutrauma.
Management of Pneumothoraces
Finally, we discuss ventilation during ECMO for patients with established baro-
trauma (pneumothorax, pneumomediastinum, subcutaneous emphysema). In such
cases we have learned [23], like others, that the best way to manage pneumothora-
ces is to decrease ventilation further, lower airway pressure as much as possible, and
wait for the gas to be reabsorbed. Chest drainage becomes mandatory only when
pneumothorax causes hemodynamic impairment.
172 A. Pesenti et al.
Following the phase of initial ECMO application and when the patients condition is
more stable, mechanical ventilation is switched from control to assisted mode. The
aim is then a progressive decrease in support, ECLS discontinuation and, nally,
extubation. The pace toward extubation, though dictated mainly by the individual
patient evolution (some patients stay days on bypass, but a few stay months), is
approached differently in various centers. An early switch to protective, assisted MV
may improve respiratory muscle function and gas exchange, decrease the need for
sedation, and aid weaning from ventilator [61]. In recent years, important techno-
logical advances contributed to signicantly improve the safety of ECLS. This led
many investigators to a change in strategy, moving to wean the patient rst from the
ventilator, and even extubate, then later from bypass [62]. Our own approach is to
start the weaning process (i.e. assisted breathing rather than controlled) as soon as
the patients safety and comfort allow. While most groups delay assisted ventilation
until the native lung improves substantially, others almost immediately switch the
ventilator to pressure support mode [63]. For example, the Karolinska group reported
their experience in 17 adult ARDS patients treated by ECLS between 1995 and
1999. They discontinued muscle relaxation and set the ventilator to pressure support
right after starting ECLS. Sedation was lightened and, soon, patients were awake
during the day, able to interact with staff and family, and participating in quiet activi-
ties like watching TV. Spontaneous assisted breathing was enhanced, when needed,
by adding 5 % CO2 to the sweep gas. Pressure support ventilation was used for all
patients, but no data are reported to indicate how pressures were titrated. Tidal vol-
ume fell from a range of 450947 mL before ECLS to 122662 mL during extracor-
poreal treatment. Mortality was surprisingly low (24 %) given the baseline severity.
From this study we conclude that in severe ARDS patients ECLS coupled with mini-
mal sedation, PSV, and low tidal volumes may be associated with high survival.
Our group previously showed that PSV may be difcult to implement in ARDS
patients with very low respiratory system compliance (Crs), likely because inspira-
tory ow falls rapidly and the ow-triggered expiratory phase of PSV starts while
the patient is still inspiring (premature expiratory cycling) [64]. Thus, these patients
are at high risk of patientventilator asynchrony. Asynchrony is a serious threat, as
higher asynchrony is associated with iatrogenic injury and delayed weaning from
the ventilator. To this end, we compared PSV with neurally adjusted ventilator assist
(NAVA) [65] in a group of severe ARDS patients with Crs ranging between 7 and 31
mL/cmH2O undergoing venovenous ECLS. In this selected group of patients, low
tidal volume PSV (34 mL/kg) led to premature expiratory cycling and high patient
ventilator asynchrony. In contrast, NAVA improved patientventilator interaction,
the more so in patients with the lowest Crs values [62].
Other modes of assisted ventilation have been applied during ECMO. For example,
airway pressure release ventilation (APRV) has been used [66], a mode of ventilation
that alternates two pressure levels at xed inspiratory and expiratory times. APRV
allows spontaneous breathing throughout the respiratory cycle, potentially avoiding
respiratory muscle atrophy and increasing ventilation to dependent lung zones.
9 Ventilator Management During ECLS 173
Institution of ECMO to avoid intubation represents the most advanced frontier for
assisted, spontaneous breathing during ECLS. Reports on this approach in patients
awaiting lung transplant (a few of whom were affected by severe ARDS) are so
numerous that it may be considered standard of care. Most notably, Fuehner and
coauthors published a retrospective analysis of 26 patients treated with ECLS at the
onset of end-stage respiratory failure [67]. Of these, 16 patients did not require sub-
sequent intubation and were transplanted after a median of 11 days on ECMO. Two
died from septic multiorgan failure after transplantation. Another patient died 60
days after transplantation from lung cancer. The remaining 13 patients were dis-
charged from the hospital 2087 days after transplantation and all of these patients
remained alive during the follow-up period (739 months). In comparison, an his-
torical control group underwent MV without ECMO while waiting transplant: those
who survived until hospital discharge required longer times on the ventilator, in the
ICU, and in hospital [67]. Overall survival was 62 % in the awake ECMO group and
35 % in the MV group (intention-to-treat populations at 6 months; 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
(p = 0.02). Thus, ECMO should be considered before intubation in patients with
end-stage respiratory failure waiting for lung transplant, but only in highly special-
ized centers with prepared and dedicated staff.
Is an awake ECMO approach feasible in severe ARDS patients? Hoeper and
colleagues reported the results of a single-center, uncontrolled, pilot trial designed
to assess the feasibility of venovenous ECMO in awake, non-intubated, spontane-
ously breathing patients with ARDS [68]. Six patients were enrolled, four of whom
were immunocompromised. Three were ultimately weaned from ECMO without
being intubated and discharged from hospital alive. On the other hand, the other
three required invasive MV, and two died in hospital. The authors concluded that
awake ECMO appears feasible in selected patients with ARDS and deserves fur-
ther evaluation.
Avoiding intubation is particularly attractive for specic situations, such as in
immunocompromised patients with ARDS, so that ECMO could be considered a
primary approach rather than only a rescue tactic. This concept is very promising,
but systematic clinical experience is still lacking.
No universally accepted rules exist as to when disconnect a patient from ECMO. The
risk inherent with extracorporeal support has to be evaluated against the risk of the
patient deteriorating without such support. For example, new bleeding may dictate
earlier than optimal disconnection in order to stop anticoagulation. On the other
hand, a new infection may challenge the marginal patients physiological reserve,
174 A. Pesenti et al.
making discontinuation of ECMO unduly risky. We consider very useful the guide-
line proposed by Palmer: a test of at least 4 h is conducted with the sweep gas ow
shut off; the ventilator in PSV mode; peak inspiratory pressure not more than 25
cmH2O; mean airway pressure between 10 and 15 cmH2O; FiO2 0.5 or less; and
PEEP between 5 and 10 cmH2O. If the patient passes the test, we proceed with
decannulation under local anesthesia in combination with mild sedation [63].
Fig. 9.1 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 ow (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)
recovering from severe ARDS, we could show that they decreased their ventilation
(on CPAP) in proportion to the amount of CO2 removed by the membrane lung [71].
The same effect has been observed in ARDS patients ventilated with NAVA while
on ECMO [72].
While supporting the possibility of targeting ventilatory needs and respiratory
drive in ARDS patients, these data should be interpreted with caution. Experience
shows that in the early, acute interstitial edema phase, ARDS patients have a very
high respiratory drive characterized by severe dyspnea and tachypnea. Blood gas
values have only a minor impact on respiratory drive. In ARDS patients on ECMO,
respiratory rate remains high and inspiratory effort may still generate very negative
intrathoracic pressures (and high transpulmonary pressures risking barotrauma and
VILI) despite low values for PaCO2 (2528 mmHg; Fig. 9.1). These observations
contrast with what has been reported in acutely decompensated COPD patients, in
whom respiratory drive (and the need for intubation) can be effectively controlled
by ECCO2R [73, 74].
One possible role for ECCO2R in ARDS patients, then, revolves around the
ultra-protective ventilator approach proposed by Terragni and colleagues [75].
These investigators showed that 2530 % of ARDS patients showed evidence of
severe hyperination on CT scan, even while ventilated according to the NIH lung-
protective recommendations (Fig. 9.2) [76]. They reasoned that VILI risk could be
reduced by lowering tidal volume from 6 to 4 mL/kg, while maintaining nearly
normal arterial pH and PCO2 through low-ow (250400 mL/min)
ECCO2R. Although their study was not designed to provide outcome data, it showed
the feasibility of such an approach. Two case reports suggest the possibility of using
ultra-protective ventilation (down to 2 mL/kg) facilitated by extracorporeal gas
exchange ([23, 77], case report 2). Finally, the Xtravent study reported on using
tidal volumes of 3 mL/kg, supported by ECCO2R, in a group of severe ARDS
patients [24]. Although the study was stopped due to slow enrolment, post-hoc anal-
ysis revealed signicantly earlier weaning from MV in the ultra-protective arm.
We can foresee the use of minimally invasive, low-ow ECCO2R devices in the
early phase of ARDS, either to avoid or delay intubation, possibly by extending the
176 A. Pesenti et al.
Fig. 9.2 Lung CT scan from one representative patient showing signs of diffuse alveolar over-
distension (red areas) despite being ventilated according to the NIH recommendations to achieve
lung protective ventilation (modied from ref. 76)
range of application of NIV or CPAP. Even though muscle paralysis and controlled
mechanical ventilation are certain to maintain a role in ARDS management, judi-
cious application of ECCO2R may lead to earlier assisted breathing, weaning, and
extubation. The possibility of less sedation, greater capacity to interact and com-
municate, and briefer intubation (or none at all) is certainly appealing, and could
reduce the incidence of ventilator-associated pneumonia and ICU-acquired weak-
ness. The most promising applications of low-ow ECCO2R appear to be related
however to the concept of ultra-protective ventilation in ARDS [76] and to the treat-
ment of COPD patients [78].
Conclusions
The concept of VILI and the need for protective ventilation has gained, following
the early experimental work of Kolobow and Dreyfuss, widespread acceptance hav-
ing been clearly validated in the clinical eld.
The practical application of these concepts in the patient undergoing ECLS,
though based on solid pathophysiological grounds, lacks however enough scientic
evidence to allow the drawing of guidelines or recommendations. Clinical experi-
ence suggests two alternative approaches, possibly applicable at different times in
the same patient: both recommend protective ventilation with tidal volumes lower
than 6 mL/kg, one looks for optimized lung recruitment by a higher PEEP level, 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 neces-
sary for the lung to recover.
9 Ventilator Management During ECLS 177
References
1. 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):21936.
2. National Heart, Lung, and Blood Institute: Division of Lung Disease. Section four: hemody-
namics, gas exchange and mechanics. In: US Department of Health, Education and Welfare,
editor. Extracorporeal support for respiratory insufciency. A collaborative study in response
to RFP-NHLI-73-20. Washington, DC: US Department of Health, Education and Welfare;
1979.
3. Pontoppidan H, Gefn B, Lowenstein E. Acute respiratory failure in the adult. 3. N Engl J
Med. 1972;287(16):799806.
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):15862.
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):589.
6. Kolobow T, Bowman RL. Construction and evaluation of an alveolar membrane articial
heart-lung. Trans Am Soc Artif Intern Organs. 1963;9:23843.
7. Dreyfuss D, Saumon G. Ventilator-induced lung injury: lessons from experimental studies.
Am J Respir Crit Care Med. 1998;157(1):294323.
8. Zapol WM, Kitz RJ. Buying time with articial lungs. N Engl J Med. 1972;286(12):6578.
9. Gattinoni L, Agostoni A, Pesenti A, Pelizzola A, Rossi GP, Langer M, et al. Treatment of acute
respiratory failure with low-frequency positive-pressure ventilation and extracorporeal
removal of CO2. Lancet. 1980;2(8189):2924.
10. Gattinoni L, Kolobow T, Agostoni A, Damia G, Pelizzola A, Rossi GP, et al. Clinical applica-
tion of low frequency positive pressure ventilation with extracorporeal CO2 removal (LFPPV-
ECCO2R) in treatment of adult respiratory distress syndrome (ARDS). Int J Artif Organs.
1979;2(6):2823.
11. Hill JD, OBrien 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):62934.
12. Guarracino F, Zangrillo A, Ruggeri L, Pieri M, Calabr MG, Landoni G, et al. -Blockers to
optimize peripheral oxygenation during extracorporeal membrane oxygenation: a case series.
J Cardiothorac Vasc Anesth. 2012;26(1):5863.
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):97117.
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;23:1721.
15. Kolobow T, Gattinoni L, Tomlinson TA, Pierce JE. Control of breathing using an extracorpo-
real membrane lung. Anesthesiology. 1977;46(2):13841.
16. Kolobow T, Gattinoni L, Tomlinson T, Pierce JE. An alternative to breathing. J Thorac
Cardiovasc Surg. 1978;75(2):2616.
17. Gattinoni L, Kolobow T, Tomlinson T, Iapichino G, Samaja M, White D, et al. Low-frequency
positive pressure ventilation with extracorporeal carbon dioxide removal (LFPPV-ECCO2R):
an experimental study. Anesth Analg. 1978;57(4):4707.
18. Gattinoni L, Kolobow T, Tomlinson T, White D, Pierce J. Control of intermittent positive pres-
sure breathing (IPPB) by extracorporeal removal of carbon dioxide. Br J Anaesth.
1978;50(8):7538.
19. Brodie D, Bacchetta M. Extracorporeal membrane oxygenation for ARDS in adults. N Engl J
Med. 2011;365(20):190514.
20. ELSO Patient Specic guidelines. http://www.elsonet.org/index.php?option=com_phocadow
nload&view=category&id=4&Itemid=627.
178 A. Pesenti et al.
21. Peek GJ, Mugford M, Tiruvoipati R, Wilson A, Allen E, Thalanany MM, et al. Efcacy 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):135163.
22. Holzgraefe B, Broom M, Kalzn H, Konrad D, Palmr K, Frenckner B. Extracorporeal mem-
brane oxygenation for pandemic H1N1 2009 respiratory failure. Minerva Anestesiol.
2010;76(12):104351.
23. Mauri T, Bellani G, Foti G, Grasselli G, Pesenti A. Successful use of neutrally adjusted ventila-
tory assist in a patient with extremely low respiratory system compliance undergoing
ECMO. Intensive Care Med. 2011;37(1):1667.
24. Bein T, Weber-Carstens S, Goldmann A, Mller T, Staudinger T, Brederlau J, et al. Lower tidal
volume strategy (3 ml/kg) combined with extracorporeal CO2 removal versus conventional
protective ventilation (6 ml/kg) in severe ARDS: the prospective randomized Xtravent-study.
Intensive Care Med. 2013;39(5):84756.
25. Pham T, Combes A, Roze H, Chevret S, Mercat A, Roch A, et al. Extracorporeal membrane
oxygenation for pandemic inuenza A(H1N1)-induced acute respiratory distress syndrome: a
cohort study and propensity matched analysis. Am J Respir Crit Care Med.
2013;187(3):27685.
26. 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):210514.
27. Zimmermann M, Bein T, Arlt M, Philipp A, Rupprecht L, Mueller T, et al. Pumpless extracor-
poreal interventional lung assist in patients with acute respiratory distress syndrome: a pro-
spective pilot study. Crit Care. 2009;13(1):R10.
28. Patroniti N, Zangrillo A, Pappalardo F, Peris A, Cianchi G, Braschi A, et al. The Italian ECMO
network experience during the 2009 inuenza A(H1N1) pandemic: preparation for severe
respiratory emergency outbreaks. Intensive Care Med. 2011;37(9):144757.
29. Repess X, Charron C, Vieillard-Baron A. Right ventricular failure in acute lung injury and
acute respiratory distress syndrome. Minerva Anestesiol. 2012;78(8):9418.
30. Ratliff JL, Hill JD, Fallat RJ, Parrot J, Tucker HJ. Complications associated with membrane
lung support by venoarterial perfusion. Ann Thorac Surg. 1975;19(5):5379.
31. Nuckton TJ, Alonso JA, Kallet RH, Daniel BM, Pittet JF, Eisner MD, et al. Pulmonary dead-
space fraction as a risk factor for death in the acute respiratory distress syndrome. N Engl J
Med. 2002;346(17):12816.
32. Kolobow T, Spragg RG, Pierce JE. Massive pulmonary infarction during total cardiopulmo-
nary bypass in unanesthetized spontaneously breathing lambs. Int J Artif Organs.
1981;4(2):7681.
33. Pipeling MR, Fan E. Therapies for refractory hypoxemia in acute respiratory distress syn-
drome. JAMA. 2010;304(22):25217.
34. Patroniti N, Isgr S, Zanella A. Clinical management of severely hypoxemic patients. Curr
Opin Crit Care. 2011;17(1):506.
35. Ullrich R, Lorber C, Rder G, Urak G, Faryniak B, Sladen RN, et al. Controlled airway pres-
sure therapy, nitric oxide inhalation, prone position, and extracorporeal membrane oxygen-
ation (ECMO) as components of an integrated approach to ARDS. Anesthesiology.
1999;91(6):157786.
36. Bryan AC. Conference on the scientic basis of respiratory therapy. Pulmonary physiotherapy
in the pediatric age group. Comments of a devils advocate. Am Rev Respir Dis. 1974;110(6
Pt 2):1434.
37. Gurin C, Reignier J, Richard JC, Beuret P, Gacouin A, Boulain T, et al. Prone positioning in
severe acute respiratory distress syndrome. N Engl J Med. 2013;368(23):215968.
38. Hemmila MR, Napolitano LM. Severe respiratory failure: advanced treatment options. Crit
Care Med. 2006;34(9 Suppl):S27890.
39. Pelosi P, Brazzi L, Gattinoni L. Prone position in acute respiratory distress syndrome. Eur
Respir J. 2002;20(4):101728.
9 Ventilator Management During ECLS 179
40. Galiatsou E, Kostanti E, Svarna E, Kitsakos A, Koulouras V, Efremidis SC, et al. Prone posi-
tion augments recruitment and prevents alveolar overination in acute lung injury. Am J Respir
Crit Care Med. 2006;174(2):18797.
41. Sud S, Friedrich JO, Taccone P, Polli F, Adhikari NK, Latini R, et al. Prone ventilation reduces
mortality in patients with acute respiratory failure and severe hypoxemia: systematic review
and meta-analysis. Intensive Care Med. 2010;36(4):58599.
42. 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.
43. Hemmila MR, Rowe SA, Boules TN, Miskulin J, McGillicuddy JW, Schuerer DJ, et al.
Extracorporeal life support for severe acute respiratory distress syndrome in adults. Ann Surg.
2004;240(4):595607.
44. 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):158994.
45. 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):4525.
46. Litmathe J, Sucker C, Easo J, Wigger L, Dapunt O. Prone and ECMOa contradiction per se?
Perfusion. 2012;27(1):7882.
47. Grifths MJ, Evans TW. Inhaled nitric oxide therapy in adults. N Engl J Med.
2005;353(25):268395.
48. Rossaint R, Falke KJ, Lpez F, Slama K, Pison U, Zapol WM. Inhaled nitric oxide for the adult
respiratory distress syndrome. N Engl J Med. 1993;328(6):399405.
49. 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.
50. 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):7806.
51. 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):9916.
52. Walmrath D, Schneider T, Pilch J, Grimminger F, Seeger W. Aerosolised prostacyclin in adult
respiratory distress syndrome. Lancet. 1993;342(8877):9612.
53. 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):81927.
54. Hager DN. High-frequency oscillatory ventilation in adults with acute respiratory distress syn-
drome. Curr Opin Anaesthesiol. 2012;25(1):1723.
55. Courtney SE, Durand DJ, Asselin JM, Hudak ML, Aschner JL, Shoemaker CT, et al. High-
frequency oscillatory ventilation versus conventional mechanical ventilation for very-low-
birth-weight infants. N Engl J Med. 2002;347(9):64352.
56. Ferguson ND, Cook DJ, Guyatt GH, Mehta S, Hand L, Austin P, et al. High-frequency oscilla-
tion in early acute respiratory distress syndrome. N Engl J Med. 2013;368(9):795805.
57. Young D, Lamb SE, Shah S, MacKenzie I, Tunnicliffe W, Lall R, et al. High-frequency oscil-
lation for acute respiratory distress syndrome. N Engl J Med. 2013;368(9):80613.
58. 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):2015.
59. Muellenbach RM, Kuestermann J, Kredel M, Johannes A, Wolfsteiner U, Schuster F, et al.
Arteriovenous extracorporeal lung assist allows for maximization of oscillatory frequencies: a
large-animal model of respiratory distress. BMC Anesthesiol. 2008;8:7.
60. Muellenbach RM, Wunder C, Nuechter DC, Smul T, Trautner H, Kredel M, et al. Early treat-
ment with arteriovenous extracorporeal lung assist and high-frequency oscillatory ventilation
180 A. Pesenti et al.
in a case of severe acute respiratory distress syndrome. Acta Anaesthesiol Scand. 2007;
51(6):7669.
61. Putensen C, Muders T, Varelmann D, Wrigge H. The impact of spontaneous breathing during
mechanical ventilation. Curr Opin Crit Care. 2006;12(1):138.
62. Mauri T, Bellani G, Grasselli G, Confalonieri A, Rona R, Patroniti N, et al. 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):28291.
63. Lindn V, Palmr K, Reinhard J, Westman R, Ehrn H, Granholm T, et al. High survival in
adult patients with acute respiratory distress syndrome treated by extracorporeal membrane
oxygenation, minimal sedation, and pressure supported ventilation. Intensive Care Med.
2000;26(11):16307.
64. Cereda M, Foti G, Marcora B, Gili M, Giacomini M, Sparacino ME, et al. Pressure support
ventilation in patients with acute lung injury. Crit Care Med. 2000;28(5):126975.
65. Sinderby C. Neurally adjusted ventilatory assist (NAVA). Minerva Anestesiol.
2002;68(5):37880.
66. Combes A. Extracorporeal membrane oxygenation (ECMO) for severe acute respiratory dis-
tress syndrome (ARDS). The EOLIA (ECMO to rescue Lung Injury in severe ARDS) trial: a
multicenter, international, randomized, controlled open trial. Ranimation. 2011;20:4961.
67. 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):7638.
68. Hoeper MM, Wiesner O, Hadem J, Wahl O, Suhling H, Duesberg C, et al. Extracorporeal
membrane oxygenation instead of invasive mechanical ventilation in patients with acute respi-
ratory distress syndrome. Intensive Care Med. 2013;39(11):20567.
69. 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):1835.
70. 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):5713.
71. 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):50810.
72. Karagiannidis C, Lubnow M, Philipp A, Riegger GA, Schmid C, Pfeifer M, et al. Autoregulation
of ventilation with neurally adjusted ventilatory assist on extracorporeal lung support. Intensive
Care Med. 2010;36(12):203844.
73. Burki NK, Mani RK, Herth FJ, Schmidt W, Teschler H, Bonin F, et al. A novel extracorporeal
CO(2) removal system: results of a pilot study of hypercapnic respiratory failure in patients
with COPD. Chest. 2013;143(3):67886.
74. Cardenas Jr VJ, Lynch JE, Ates R, Miller L, Zwischenberger JB. Venovenous carbon dioxide
removal in chronic obstructive pulmonary disease: experience in one patient. ASAIO
J. 2009;55(4):4202.
75. Terragni PP, Del Sorbo L, Mascia L, Urbino R, Martin EL, Birocco A, et al. Tidal volume
lower than 6 ml/kg enhances lung protection: role of extracorporeal carbon dioxide removal.
Anesthesiology. 2009;111(4):82635.
76. Terragni PP, Rosboch G, Tealdi A, Corno E, Menaldo E, Davini O, et al. Tidal hyperination
during low tidal volume ventilation in acute respiratory distress syndrome. Am J Respir Crit
Care Med. 2007;175(2):1606.
77. Bein T, Osborn E, Hofmann HS, Zimmermann M, Philipp A, Schlitt HJ, et al. Successful treat-
ment of a severely injured soldier from Afghanistan with pumpless extracorporeal lung assist
and neurally adjusted ventilatory support. Int J Emerg Med. 2010;3(3):1779.
78. Kluge S, Braune SA, Engel M, Nierhaus A, Frings D, Ebelt H, et al. Avoiding invasive
mechanical ventilation by extracorporeal carbon dioxide removal in patients failing noninva-
sive ventilation. Intensive Care Med. 2012;38(10):16329.
Chapter 10
Daily Care on ECLS
Giles J. Peek
Introduction
Daily management of the patient on ECLS requires a careful assessment of the cir-
cuit, the patient, and the interaction between them. Robust communication between
ECLS specialists, intensivists, surgeons, nurses, respiratory therapists, and others is
essential to providing safe and effective care. Once the patient has been established
on ECLS their care passes through three phases of stabilization, stability, and nal
improvement before they can be weaned off ECLS and decannulated. Accordingly,
this chapter will address daily care in four sections: (a) housekeeping of the cir-
cuit; (b) initial stabilization of gas exchange and cardiovascular homeostasis; (c)
meticulous progress toward regaining organ function and limiting complications
during a phase of relative stability; and, nally, (d) judgments regarding when to
withdraw ECLS as the patient improves.
Circuit Housekeeping
The ECLS specialist examines the circuit from end-to-end at the beginning of each
shift and rechecks it hourly to ensure its safe operation and detect problems before
they become circuit-threatening. A checklist should be used to ensure that nothing
is omitted. We begin by examining the exit site of the drainage cannula (or double-
lumen cannula) for signs of infection, bleeding, or visible recirculation, while also
assuring that the sutures are secure. A transparent lm dressing is helpful in
allowing the cannulation site to be easily viewed. After 2 or 3 weeks the skin under
these dressings can become sweaty and macerated so we often switch to a breath-
able dressing. We swab the cannulation site two or three times per week for micro-
biological surveillance. If the cannulation site is breaking down (which can happen
after several months on ECLS), we may apply an antiseptic ointment (chloram-
phenicol or betadine) sparingly.
Next the connector joining the cannula to the venous line is checked: this is espe-
cially important during the rst circuit check following cannulation in case steps
were overlooked in error. The connection on the cannula side is bonded in the fac-
tory and does not require tie strapping. The connector should also be examined for
cracks and clots with a bright light. This light is then used to follow the circuit past
the bridge (if present), along to the compliance chamber (if used), and at the inlet
pressure measurement port (if used). Each connector should be checked for secu-
rity, tie straps, cracks, and clots. Luer connectors should have their bungs or pigtails
checked for integrity and the pigtails themselves should be examined for clots. Any
pigtails that are not being used should either be ushed with heparin-containing
saline or removed altogether (a simpler circuit is a safer circuit). If a bridge is used
it should be examined carefully for clots and the tubing for cracks; constant clamp-
ing and unclamping can lead to tubing failure. It is becoming increasingly common
to omit the bridge to remove a source of clot formation. Having either a stopcock or
removable bridge is a good compromise between a streamlined circuit with few
connectors and the security of being able to use a bridge if necessary.
If a venous compliance chamber is being used, its function should be tested by
clamping the inlet tubing and watching the bladder collapse. It should be examined
for clots and the housing examined for cracks. Air can be added or withdrawn to
adjust the level of pressure in the chamber so that a dimple is seen in the compliance
tubing. If the inlet pressure is being measured from the chamber, the transducer
should be zeroed (the transducer tubing and system should be air-lled for this pur-
pose). If a compliance chamber is not being used and the inlet pressure is being
measured directly from the circuit, the (saline-lled) transducer system should be
zeroed with caution to prevent air being entrained into the circuit. Because of this
risk, it is increasingly common to omit measurement of inlet pressure.
The tubing is then followed to the pump which is checked with the light for clots.
The pump should be listened to for abnormal sounds and felt for excessive heat
generation or vibration. The console should be surveyed according to the manufac-
turers instructions. The alarms, alarm limits, pressure measurements (pre- and
post-membrane), pump speed, ow, battery, and power supply should be veried. If
inlet pressure is not being measured, the relationship of pump speed and circuit ow
should be checked to ensure that excessive rpms are not being used. This relation-
ship varies from pump to pump so that each ECLS center should create a nomogram
of what ow to expect at a given rpm setting. For the 3/8 Thoratec Centrimag the
blood ow in mL/min should exceed the rpm; thus a blood ow of 4.5 L/min with a
pump speed of 3250 rpm indicates unobstructed venous outow. A lower ratio of
ow to rpm signals higher afterload, as when the membrane is failing or during
venoarterial (VA) ECLS.
10 Daily Care on ECLS 183
The tubing is then followed to the oxygenator. If pigtails are used to access the
circulation, they are best placed between pump and oxygenator where inadvertant
opening forces blood out rather than sucks air in (hypovolemia being much easier to
correct than a circuit full of air). If a continuous venovenous hemoltration (CVVH)
machine is connected to the ECLS circuit at this point, the inlet and outlet pressure
limits for the CVVH access must be adjusted to account for the positive pressure in
the ECLS circuit.
The membrane lung (oxygenator) should be examined for clots and cracks, and
all taps, bungs, and connections examined for cracks and checked for tightness. The
pressure gradient across the membrane should be measured and compared to prior
values; with polymethylpentene membranes this should be on the order of 1050
mmHg, depending on ow, with higher values suggesting clot build up and incipient
oxygenator failure. Membrane function should be checked by noting the color of
blood exiting the membrane, which should be bright red. The PO2 of a blood sample
taken from the outlet pigtail of the oxygenator will be on the order of greater than
30 kPa (225 mmHg), assuming 100 % oxygen as the sweep gas. Falling post-
membrane PO2 values may signal a failing oxygenator, but its ability to support gas
exchange is not impaired until the PO2 drops below about 15 kPa (112 mmHg) and
the saturation begins to fall. The decision to change out the oxygenator is based on
the rate of falling post-membrane PO2 and the trajectory of the patients improve-
ment. If the patient is likely to come off ECLS in the next 48 h and membrane func-
tion is deteriorating slowly, it would be better to push on with the current oxygenator
(since replacement has risks, costs, and may induce additional systemic inamma-
tion, delaying lung recovery). Membrane life can be prolonged by increasing the
degree of anticoagulation and maintaining high circuit blood ow (at least 2 L/min).
However, if the patient will not recover soon, it is better to replace a failing oxygen-
ator electively rather than put the patient at risk of a sudden deterioration and emer-
gency replacement (see Chap. 11).
The sweep gas ow and the FIO2 selector on the blender should be veried.
Condensation can build up in the gas phase of the oxygenator and lead to impairment
of gas exchange. This can be seen as clear water dripping from the gas exhaust port
and will be reected in a poor post-oxygenator PO2, but a normal pressure gradient.
To prevent uid accumulation, the membrane should be ushed with high ow
sweep gas for a few seconds every shift, or more often if uid is clearly building up.
In addition, the membrane can be burped by occluding the gas exhaust port with a
nger for a few seconds and allowing the gas pressure to build up, then releasing.
Care should be exercised to prevent excessive pressure: some centers use a safety
valve on the sweep gas inlet to prevent this. Burping should not be attempted if using
micro-porous polypropylene oxygenators, because these membranes develop plasma
leakage characterized by bubbling of amber-colored plasma from the gas exhaust.
Water heater function is judged by noting whether the tubing is warm, the pump
turning, the reservoir still contains water, and the patients temperature is stable.
Reservoir water can become colonized with pseudomonas: we send samples for
microbiological culture periodically.
184 G.J. Peek
Circuit housekeeping concludes with careful appraisal of the blood tubing from
oxygenator past the bridge (if used) to the patient. If single-lumen cannulae are
being used, the return cannula is checked in the same way as the drainage cannula
was; if a double-lumen cannula is being used the assessment is nished at the
connector.
ECLS ow is initiated and built up slowly over a few minutes, until further increases
in pump speed do not produce more ow. The speed is then reduced by 1020 % to
prevent excessive negative inlet pressure. If adequate ow (around 60 mL/kg/min in
adults) cannot be achieved readily, the patient may be hypovolemic or cannula posi-
tion may be suboptimal. During initiation of ECLS the patient may become hypo-
tensive due to sudden changes in concentrations of potassium, calcium, or
magnesium. The risk of this seems less if the circuit is primed using a balanced
electrolyte solution. If hypotension occurs administration of calcium will often
prove effective.
During the early minutes we keep the sweep gas ow low (around 2 L/min) in
order to prevent a sudden reduction in PaCO2 resulting in cerebral vasoconstriction.
Once circuit blood ow has been optimized, we set the sweep gas ow at a similar
value (in L/min) and start reducing the ventilation as described in Chap. 9 in order
to reduce the risk of ventilator-induced lung injury (VILI) [2]. Our own practice is
to use pressure-controlled ventilation with the inspiratory pressure initially below
30 cm H2O, but reduced over the next several hours to 10 cm H2O above PEEP [3].
We generally apply 10 or 15 cm H2O PEEP and a respiratory rate of 10/min while
the patient is paralyzed, then convert to pressure support of 10 cm H2O over PEEP
after paralysis is discontinued. As ventilator settings are reduced, sweep gas ow is
adjusted to achieve the target PaCO2 (usually 46 kPa; 3045 mmHg) based on arte-
rial blood gas results. For patients with severe hypercapnia, rapid reduction in
PaCO2 may not be desired and a relatively normal arterial pH is a more appropriate
target for the rst 1224 h as the metabolic compensation resolves.
The patients systemic oxygenation depends largely on how much of the cardiac
output can be directed to the ECLS circuit, as discussed in Chap. 1. Once circuit
ow is adequate, it should be possible to discontinue other oxygenation rescue ther-
apies, such as inhaled prostacyclin or nitric oxide. Ventilator FIO2 can be reduced as
long as circuit ow and membrane function are adequate, with a goal SpO2 gener-
ally above 80 %. We aim to reduce the FiO2 to 30 % over the rst 1224 h. If this is
not possible, one of two problems is likely:
1. The ECLS circuit is inefcient because of recirculation, poor cannula position,
or inadequate size of the venous cannula.
2. The patient has a very high cardiac output and the ECLS circuit ow is capturing
an insufcient proportion of this resulting in an effective right to left shunt (since
the lungs are not working well).
10 Daily Care on ECLS 185
This phase of the illness is far from the passive period one might expect, the aim
being to actively move the patient in the direction of recovery from underlying ill-
ness, iatrogenic VILI, microcirculatory dysfunction, and ICU-acquired weakness.
During this phase we hope that capillary leak resolves, edema improves, and that the
catabolism of injury is replaced by repair and healing. Lung protection may allow
radiographic clearing and improvements in lung mechanics. A coherent plan of
management is crucial, using a systematic approach and institutionally agreed pro-
tocols. We aim to use as many of the standard ICU bundles as possible to ensure that
normal intensive care management proceeds, but with some specic provisos dis-
cussed in more detail below.
186 G.J. Peek
Lung Management
After stabilization, the patient should be on low pressure (or volume) settings with
a machine rate of 10 breaths/min but triggering the ventilator (usually on pressure
support of 10 cm H2O). The tidal volume may be as low as 0.5 mL/kg and the
patient may be quite tachypneic (2030 breaths/min or even higher) as a conse-
quence of the disease process. Raising the sweep gas ow to enhance CO2 removal
has little apparent impact on dyspnea in some patients. At this stage, the intensivist
is confronted with several options for continued ventilator support: extubation [5],
tracheostomy, or continued conventional ventilation through the endotracheal tube.
Patients with little or no sputum, particularly those with ARDS rather than
pneumonia, and with good neurological function, may be best extubated. This
approach prioritizes preventing ventilator-associated pneumonia by removing the
endotracheal tube; avoiding sedatives; and facilitating communication and mobili-
zation, as described in Chap. 12. Potential downsides include the uncertain conse-
quences of derecruitment on lung healing; the potential impact of spontaneous
breathing on progression of VILI [6]; and the risk of aspiration. Most patients,
however, end up undergoing a tracheostomy. This can usually be done percutane-
ously at the bedside with bronchoscopic control, but surgical tracheostomy is also
used. In either case it is essential that the operator is both experienced and an inte-
gral part of the ECLS team, since bleeding complications are best prevented. If
tracheostomy site bleeding occurs, surgical exploration may be needed. An alterna-
tive approach is to re-intubate while packing the stoma for 48 h. For some patients,
we choose to continue ventilation through a conventional endotracheal tube.
Younger patients who have only been ventilated for 1 or 2 days prior to ECLS may
recover sufciently quickly to avoid a tracheostomy. We also leave the endotra-
cheal tube in patients with severe air leak syndrome and surgical emphysema as a
means to apply high-frequency oscillatory ventilation at a low pressure (mean air-
way pressure of 13 cm H2O and no attempt to recruit), hoping to allow the lung to
heal while minimizing ventilator alarms.
Whichever approach to airway management is chosen, patients are positioned
semi-upright to reduce the risk of ventilator-associated pneumonia and facilitate
ventilation and recruitment of the dependent lung. We employ regular physiother-
apy with bagging, suction, lavage, nebulized medications, and bronchoscopy as
needed to clear secretions. We occasionally manage ECLS patients in the prone
position to augment lung protection; unload a compromised right ventricle [7]; or
aid recruitment when CT demonstrates dependent consolidation. It is our practice to
prone patients for around 18 h in each 24, usually turning prone in the afternoon and
un-proning after morning rounds [8]. Of course, sedating a patient who is awake so
that they can be proned is not very sensible, as an awake, upright patient, breathing
spontaneously and with a good cough is preferred.
Pleural effusions and pleural air should be drained to expand the lung. As with
tracheostomy, experienced operators who are an integral part of the ECLS team
should be chosen to insert chest tubes. It is essential that the pleural space is kept
10 Daily Care on ECLS 187
empty because the lung will not recover if it is surrounded by infected uid or incar-
cerated by brosis. It is misguided to refrain from placing a chest tube because of
fear of bleeding; the result will often be a thoracotomy if the lung becomes trapped
in a collapsed state. In the event of persistent failure of the lung to expand it is likely
that there is either a necrotic segment with persistent bronchopleural stula, or a
collection which will not drain. In either event a CT chest with contrast is helpful in
determining the best treatment. This may be a targeted drain or an extensive formal
exploration of the chest with decortication and resection of the necrotic portion of
the lung [9]. Finally if a lung biopsy is needed on ECLS, it must be done surgically,
not transbronchially, for obvious reasons.
This may seem an eclectic mix but they are all intimately related. When ECLS is
initiated, most patients are paralyzed and receiving infusions of an opiate and a
sedative. Many are on multiple drugs as they are difcult to sedate. We rst stop
the paralytic and, once it has worn off, simplify and reduce the sedative regime. As
with other mechanically ventilated patients, we use a protocol to aid daily interrup-
tion of sedation and allow us to assess neurologic function prior to restarting the
infusions at a reduced dose [10]. It is often necessary to maintain the opiate infusion
to keep the patient comfortable or even increase it as the sedative is withdrawn.
Having the patient comfortable but awake allows a closer eye to be kept on neuro-
logical function as the patient can move and obey commands. Moreover, this is also
important in maintaining movement of joints and recovery of muscles. Physical
therapy, splinting, and passive and active range-of-motion exercises may prevent
foot drop, exion deformities, and other forms of neuromuscular debility. Ultimately,
it is possible to have patients undertaking exercise while on ECLS and even mobi-
lizing out of bed (Chap. 12).
Delirium is common, and usual ICU approaches to prevention are recommended,
including the ABCDE steps (Assess and manage pain; Both daily sedative interrup-
tion and spontaneous breathing trials; Choice of analgesia and sedatives; Delirium
monitoring; and Early mobility) [11]. Some patients may become agitated at night
despite having been normal during the daytime; treatable causes of delirium should
be excluded before resorting to pharmacologic approaches. We strive to modulate
room lighting and staff activity to help re-establish circadian rhythm, and a window
with a view of the sky is invaluable, especially since these patients may be critically
ill for many weeks.
Many patients complain of being cold on ECLS and develop uncontrollable shiv-
ering despite having a core temperature of 37 C. A warm air blanket can provide
considerable symptomatic relief. Of course patients on ECLS are at risk of cerebral
complications both of their disease and of anticoagulation, so we have a low thresh-
old for obtaining a head CT in patients with new or persisting neurological ndings.
188 G.J. Peek
Heart
One of the most important heart-related tasks is to ensure that the non-cardiogenic
pulmonary edema of the patients ARDS is not, in fact, cardiogenic. Endocarditis or
acute valve failure can result in severe respiratory failure and a radiographic appear-
ance identical to ARDS. Obtaining a high-quality echocardiogram (transthoracic or
transoesophageal) is a high priority early in the management of patients on
ECLS. H1N1 inuenza causes a myocarditis in a small proportion of patients, mani-
festing as severe bradycardia but amenable to temporary trans-venous pacing and
lasting for 12 weeks.
We aim to provide full nutrition via an enteral route but, if this is not possible, total
parenteral nutrition (TPN) is used. Normal nasogastric feeding regimes can be used
but great care should be exercised when inserting the nasogastric tube as even a
small abrasion in the throat during insertion has proved fatal. When inserting lines
for TPN, we recommend avoiding the subclavian route unless one is extraordinarily
competent in line insertion, ultrasound guidance, and control of bleeding from the
subclavian vein and artery. Peripherally inserted central lines are ideal for TPN dur-
ing ECLS as it is rarely needed for more than a week or 2. If access for TPN really
is impossible to obtain, the ECLS circuit can be used. Feeding is the best prophy-
laxis against gastric ulceration, but local care bundles to prevent stress ulceration
should be used. Fully awake extubated patents can eat as normal but may benet
from continued enteral feeding to ensure sufcient intake.
Kidneys
Patients will often already be in acute renal failure when referred for ECLS, or may
go on to develop it while on ECLS. These patients should receive CVVH: this can
be done via a standard double-lumen CVVH catheter, allowing the same protocol to
be used in both ECLS and non-ECLS patients. Additional anticoagulation is not
required if heparin is being given to the patient to prevent clotting in the ECLS cir-
cuit. The subclavian route should be avoided as bleeding is very challenging in this
area. If access is impossible the circuit can be used with wide-bore pigtails between
the pump and the oxygenator (to prevent air embolus). In this case, the access pres-
sures on the CVVH machine will need to be recongured to account for the positive
line pressure which is much greater than the default settings on the machine are
used to. In addition, the tap on the red inlet line to the CVVH machine may need to
be turned slightly to reduce the inlet pressure (as this pressure is usually negative
10 Daily Care on ECLS 189
envelope. If lactic acidosis, tachycardia, and reduced tissue oxygen saturation (by
near-infrared spectroscopy) remain unresponsive to treatment, it is reasonable to
transfuse, even when the hemoglobin concentration is above 8 g/dL.
To control bleeding on ECLS, especially when surgery is required, antibrinolyt-
ics such as aprotinin, tranexamic acid, and epsilon-aminocaproic acid can be useful
adjuncts. We transfuse fresh frozen plasma, cryoprecipitate, and platelets aiming to
maintain values of INR <1.5, brinogen >200 mg/dL, and platelet count >150/mm3
when patients are bleeding but these values can be relaxed somewhat when patients
are not bleeding.
One hopes that clear signs of improvement are seen after 9 days or so of
ECLS. Recovery may be signaled through better gas exchange (for oxygen or CO2),
radiographic clearing, improving respiratory mechanics (increased tidal volumes
for any given pressure; or falling pressures for any given tidal volume), or general
hemodynamic stability (see Chap. 13). Even if there is no clinical improvement
after 9 days, excellent outcomes are still possible. Many patients simply take longer
to recover. The chest radiograph often lags behind other signs of improvement and
so should not be relied on in isolation. If a month passes with no apparent healing,
a stringent search for non-reversible causes should be undertaken such as a full body
CT scan, bone marrow aspiration, repeat echocardiography (and estimation of PA
pressure), and possibly a lung biopsy. If all of these investigations are unrevealing
and there are no other extenuating factors, further patience is recommended.
If the patient is ready to trial off ECLS, the lung mechanics will have improved to
the point where the tidal volume is around the 4 mL/kg on rest settings and gas
exchange is clearly improving. In order to test whether the patient is ready to come
off ECLS a trial is conducted as described in Chap. 13, generally by returning to
normal lung-protective ventilator settings, then turning the sweep gas ow off. When
this is well-tolerated, the patient can be decannulated. For patients extubated while
on ECLS, the criteria for passing a trial off support are even simpler: they are identi-
cal to those for judging a spontaneous breathing trial [14]. If there is uncertainty it is
perfectly reasonable to prolong the trial off for several more hours or even overnight
as patients often slowly improve as they re-establish spontaneous breathing.
Conclusion
References
1. Butt W, Heard M, Peek GJ. Clinical management of the extracorporeal membrane oxygenation
circuit. Pediatr Crit Care Med. 2013;14(5 Suppl 1):S139.
2. Brower RG, Matthay MA, Morris A, et al. 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;342:13018.
3. Peek GJ, Mugford M, Tiruvoipati R, et al. Efcacy and economic assessment of conventional
ventilatory support versus extracorporeal membrane oxygenation for severe adult respiratory
failure (CESAR): a multicentre randomised controlled trial. Lancet. 2009;374:135163.
4. Schmidt GA. Cardiopulmonary interactions in acute lung injury. Curr Opin Crit Care.
2013;19:516.
5. Hoeper MM, Wiesner O, Hadem J, et al. Extracorporeal membrane oxygenation instead of
invasive mechanical ventilation in patients with acute respiratory distress syndrome. Intensive
Care Med. 2013;39:20567.
6. Yoshida T, Torsani V, Gomes S, et al. Spontaneous effort causes occult pendelluft during
mechanical ventilation. Am J Respir Crit Care Med. 2013;188:14207.
7. Jozwiak M, Teboul J-L, Anguel N, et al. Benecial hemodynamic effects of prone positioning
in patients with acute respiratory distress syndrome. Am J Respir Crit Care Med.
2013;188:142833.
8. Gurin C, Reignier J, Richard J-C, et al. Prone positioning in severe acute respiratory distress
syndrome. N Engl J Med. 2013;368:215968.
9. Joshi V, Harvey C, Nakas A, et al. The need for thoracic surgery in adult patients receiving
extracorporeal membrane oxygenation: a 16-year experience. Perfusion. 2013;28:32832.
10. Kress JP, Pohlman AS, OConnor MF, Hall JB. Daily interruption of sedative infusions in criti-
cally ill patients undergoing mechanical ventilation. N Engl J Med. 2000;342:14717.
11. Pandharipande P, Banerjee A, McGrane S, et al. Liberation and animation for ventilated ICU
patients: the ABCDE bundle for the back-end of critical care. Crit Care. 2010;14:1579.
12. Wiedemann HP, Wheeler AP, Bernard GB, et al. Comparison of two uid-management strate-
gies in acute lung injury. N Engl J Med. 2006;354:256475.
13. Hbert PC, Wells G, Blajchman MA, et al. A multicenter, randomized, controlled clinical trial
of transfusion requirements in critical care. N Engl J Med. 1999;340:40917.
14. Ely EW, Baker AM, Dunagan DP, et al. Effect on the duration of mechanical ventilation of
identifying patients capable of breathing spontaneously. N Engl J Med. 1996;335:18649.
Chapter 11
Crises During ECLS
Introduction
The use and success of extracorporeal life support (ECLS) has historically been
limited by frequent and potentially severe complications. Increased experience
among providers and improvements in circuit technology have reduced, but not
eliminated, these complications in the modern era of ECLS. Accurate estimates of
current complication rates are difcult to quantify. Published rates and those found
in registries typically reect the use of both older and newer technology and prac-
tice, tending to overestimate the current risks. The Extracorporeal Life Support
Organization (ELSO), which currently compiles the most extensive database on
complications related to ECLS, includes ECLS cases with both modern and out-
dated technology, as well as centers with greater and lesser degrees of experience
with ECLS using a variety of ECLS congurations. Determining the expected rate
of complications in any given circumstance should be attempted with caution.
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.
Circuit Crises
Circuit Air
Any loss of circuit integrity may result in the entrainment of air into the circuit. Air
in the circuit is an emergency situation that must be addressed immediately; how-
ever, the risk to the patient is highest when air is present on the post-oxygenator side
of the circuit since the oxygenator itself may serve as an air trap. In venovenous
ECLS, circuit air risks venous air embolism. In venoarterial ECLS or venovenous
ECLS in the presence of an atrial septal defect or patent foramen ovale, there is a
risk of arterial air embolism.
Pre-oxygenator air can originate from peripheral or central venous lines that are
inadvertently left open or accessed for intravenous infusions. Because circuit pressure
is negative between the outow cannula and the pump, cracked stopcocks or tubing, or
loose connections can entrain air, as can partial dislodgement of the drainage cannula
resulting in exposure of a side port. Air can arise within the oxygenator itself from
improper priming, oxygenator membrane rupture, or occlusion of the gas exhaust out-
let. Post-oxygenator air is seen if air passes through the oxygenator from the pre-ox
ygenator side or is directly entrained through defects in the circuit after the
oxygenator.
Another mechanism through which air can enter the circuit is known as cavita-
tion, which occurs when a markedly negative pressure within the venous drainage
cannula and tubing draws a large amount of gas out of solution. Extreme changes in
pressure induce hydrodynamic phenomena, including bubble nucleation, growth,
and collapse. This process can be minimized by avoiding extremes in negative pres-
sure on the venous side of the circuit. Cavitation can also occur in the setting of a
high sweep gas ow rate and a low blood ow rate.
In order to minimize introduction of air into the circuit, all medication and blood
products should be given directly to the patient through separate venous access. If
the circuit must be used, the pre-oxygenator limb should be accessed so that the
oxygenator can trap any air inadvertently introduced. A bubble detector may also be
placed on the post-oxygenator tubing, which can be set to stop the pump immedi-
ately if air is detected.
Small amounts of air on the venous (outow) side can often be removed without
ceasing ECLS by carefully withdrawing the air into a syringe; however, 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 the circuit and
isolating it from the patient. When air appears on the post-oxygenator side, the out-
11 Crises During ECLS 195
Fig. 11.1 Appropriate clamp position when isolating patient from circuit. During an ECLS crises in
which the integrity of the circuit is at risk or in question the patient may be separated from the circuit
by clamping the venous and arterial limbs. Reprinted with permission from CollectedMed.com
If air reaches the patient, management is mostly supportive, although venous air
may be aspirated under some circumstances. (Table 11.1 highlights diagnosis and
management of circuit-related crises.)
Thrombosis
The interaction between blood and the ECLS circuit may alter both procoagulant
and anticoagulant pathways, potentially resulting in thrombosis in the circuit or the
patient. Thrombosis in the ECLS circuit appears as dark red clot or white strands of
198 C.L. Agerstrand et al.
Fig. 11.2 Picture of blood and brin clot in oxygenator. Arrows depict blood and brin clot forma-
tion on the oxygenator. Reprinted with permission from CollectedMed.com
Oxygenator Failure
Polymethylpentene hollow ber oxygenators are more efcient and durable than
older polypropylene or silicone oxygenators. Despite improvements in durability,
oxygenator effectiveness is critical to the successful management of ECLS and its
function should be assessed frequently. A decreasing post-oxygenator PO2, an
increasing trans-membrane pressure gradient, or a need to steadily increase sweep
gas ow rate to manage PaCO2 may be indications of a failing oxygenator.
Oxygenator thrombosis is a common cause of an increase in the trans-membrane
pressure gradient and oxygenator malfunction. This phenomenon can be assessed
by visual inspection of the oxygenator during a circuit check. If the pressure gradi-
ent increases over time, a signicant amount of thrombus may be developing.
Raising the level of anticoagulation may stabilize the membrane but, otherwise, it
will need replacement.
Membrane leakage, with blood passing across the membrane and into the gas
exhaust line, is another potential complication resulting in oxygenator failure. If
blood is found in the gas exhaust line, the oxygenator should be replaced. However,
because oxygenator replacement is not without risk, close monitoring alone may be
considered if the leakage is felt to be minimal. The gas exhaust outlet must remain
vented to the atmosphere; otherwise, signicant increase in oxygenator gas pressure
could place the patient at risk for air embolism across the compromised membrane.
Oxygenator replacement requires that the patient temporarily be removed from
ECLS. Experienced teams can change out an oxygenator in approximately 30 s.
However, this takes careful coordination and planning. Patients should be placed on
emergency ventilator settings and the pump turned off or to a very low ow rate.
Vasopressors and inotropic agents should be immediately available to manage any
hemodynamic compromise. The circuit is clamped near the inow and outow
ports of the oxygenator, which is then removed and replaced. Gas and heater lines
are connected to the new oxygenator and ECLS support is resumed.
Pump Failure
Pump failure is uncommon but has been reported [9]. If pump failure occurs for any
reason, a patient who is dependent on ECLS should be placed on emergency ventila-
tor settings. During VA ECLS it is imperative to clamp the circuit to prevent reversal
of ow from the patients arterial system through the ECLS circuit into the venous
system (a large arteriovenous shunt), leading to cardiopulmonary collapse. A hand
crank, which should be kept alongside ECLS circuits, is used to generate blood ow
while the source of the problem is identied and corrected. If disruption of the elec-
trical power supply is the cause of pump failure, then battery power or a backup
power supply should be obtained. If the ECLS circuit was inadvertently powered
off, it should be restarted immediately with an assessment of the alarm functions.
200 C.L. Agerstrand et al.
If the circuit has adequate power supply and the pump is rotating, but there is no
ow, 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 conrmed by thor-
ough visual inspection, assessing for cracks or thromboses. If the pump head is
intact, it may be placed back onto the motor and the pump restarted. In the event
there is still no ow, the motor should be replaced.
Tubing rupture or cannula fracture constitutes an emergency and can result in exsan-
guination or entrainment of air into the circuit and patient. Risk of exsanguination
is greatest if the rupture or fracture occurs post-pump, where the circuit is under
positive pressure. The risk of air entrainment is greatest when tubing rupture or can-
nula fracture occurs pre-pump, under negative pressure conditions. For post-pump
rupture or fracture, the circuit should be immediately clamped off, which temporar-
ily discontinues ECLS and requires placing the patient on emergency ventilator
settings, while the defective tubing or cannula is isolated and replaced. For small
cracks in the tubing or cannula that occur pre-pump, it may be possible to continue
ECLS support while repairing the problem if the patient is on VV-ECMO and there
is no atrial septal defect or other potential right-to-left circulatory shunt.
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 ber 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, at times, in up to 3040 % of cases [9]. The etiology and severity of
ECLS-related bleeding varies from minor oozing at surgical sites to fatal intracere-
bral hemorrhage; however the risk of severe hemorrhage appears to have decreased
over time (Table 11.2). 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
moderate degree of anemia [2628]. Similar to our practice with other critically ill
patients, our protocol 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 donorrecipient cross-matching.
Thrombocytopenia also is a common complication of ECLS, with platelet activa-
tion and consumption potentially resulting in a notable amount of thrombocytopenia
over the rst 45 days of ECLS support [2932]. 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 undergoing
an invasive procedure. An exception to this approach may be in patients with severe
pulmonary hypertension, in whom the risk of platelet sequestration in the pulmo-
nary vasculature, which can precipitate pulmonary hypertensive crises, may out-
weigh the benet 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
biocompatible and less prone to thrombosis, low levels of systemic anticoagulation
may be administered without a signicant increase in the rate of circuit thrombosis,
therefore minimizing the likelihood of exacerbating any bleeding. In the setting of
life-threatening hemorrhage, anticoagulation may be withheld for undetermined
periods of time without clinically signicant thrombosis, although anticoagulation
should be resumed as soon as possible [33]. Fresh frozen plasma can be adminis-
tered as needed. Fibrinogen levels may be maintained above 100 mg/dL with the
transfusion of cryoprecipitate.
If bleeding from surgical or cannulation sites occurs, reinforcement of the site
with additional sutures, application of a pressure dressing, or the use of electrocau-
tery may reduce bleeding. In cases of pulmonary hemorrhage, bronchoscopy may be
used to localize the source of bleeding to potentially guide denitive interventions.
Uncontrollable, life-threatening bleeding has been treated successfully with anti-
brinolytics such as aminocaproic acid (AMICAR) and tranexamic acid [3436].
Recombinant factor VII, which complexes with tissue factor to increase the production
of thrombin and enhance clot formation, has also been used to manage hemorrhage
during ECLS [3740]. However, with any of these agents, the risk of oxygenator and
circuit thromboses, including fatal clots, may be increased [41, 42]. These agents
should only be considered in life-threatening situations when other measures have fail-
ure. Prior to using of any of these therapies, consultation with a hematologist is recom-
mended. (Table 11.3 highlights diagnosis and management of patient-related crises.)
Hemolysis
[4348]. Despite these advances, hemolysis can still occur from extremes in nega-
tive pressure in the venous drainage cannula and tubing, typically when pressures
exceed 100 mmHg (and perhaps even from pressures simply exceeding
50 mmHg), resulting in cavitation and red blood cell destruction. Hemolysis can
occur at sites of turbulent, non-laminar blood ow in the circuit or patient, includ-
ing at points where tubing may be kinked or intentionally split, or where vascular
access points or hemoltration devices are connected to the circuit. In situ throm-
boses can also cause non-laminar ow and result in hemolysis, as can overheated
water baths.
The components of any ECLS circuit depend on institutional preference, how-
ever, a simplied circuit without a bridge may reduce the rate of hemolysis.
Likewise, externally applied, non-invasive monitors can reduce the necessity for
access points, further minimizing the amount of turbulent ow. The use of a sim-
plied circuit consisting of only the key components, including centrifugal pump,
membrane oxygenator, pre- and post-oxygenator pressure monitors, and heat
exchanger, is preferred. As compared with older or more complex ECLS circuits,
signicant hemolysis does not typically occur with the use of a simplied circuit
design.
If severe hemolysis is identied that cannot be attributed to kinks in cannula
tubing or a systemic process in the patient, and cannot be corrected by reducing
the temperature of the water bath, then the circuit or pump head may need to be
replaced. With any occurrence, a source of non-circuit-related hemolysis should
be excluded clinically.
206 C.L. Agerstrand et al.
Refractory Hypoxemia
Shock
Management of shock states during venovenous ECLS is treated the same as shock
without ECLS because venovenous congurations do not provide circulatory sup-
port. If cardiogenic shock develops in the setting of venovenous ECLS, conversion
to venoarterial ECLS, which provides circulatory support, should be considered,
although the upper body oxygenation issues with femoral arterial reinfusion men-
tioned above, must be kept in mind. Inotropic medications may also be considered.
If a high-output shock state develops on ECLS of any conguration, the extracor-
poreal blood ow rate will account for a lower percentage of the patients cardiac
output. Consequently, the proportion of blood being oxygenated by the circuit will
be lower, which may result in worsened hypoxemia. While the use of ECLS in high
output shock has not been well studied, reports have demonstrated the successful
use of venoarterial ECLS for adult patients with septic shock in order to augment
cardiopulmonary function in the setting of vasodilation, reduced cardiac output, and
high metabolic demand. Judicious patient selection is paramount to maximize the
benet of ECLS in the setting of shock [5256].
Conclusion
References
24. 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(6):167480. discussion 1680.
25. 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. 2012;147(1):
4955.
26. Simon T, Snyder EL, Solheim BG. Rossis principles of transfusion medicine. 4th ed. West
Sussex, UK: Wiley-Blackwell; 2009.
27. Roback JD, Neuman RB, Quyyumi A, Suliff R. Insufcient nitric oxide bioavailability: a
hypothesis to explain adverse effects of red blood cell transfusion. Transfusion. 2011;51(4):
85966.
28. Carson JL, Grossman BJ, Kleinman S, Tinmouth AT, Marques MB, Fung MK, et al. Red blood
cell transfusion: a clinical practice guideline from the AABB. Ann Intern Med. 2012;157(1):
4958.
29. Plotz FB, van Oeveren W, Bartlett RH, Wildevuur CR. Blood activation during neonatal extra-
corporeal life support. J Thorac Cardiovasc Surg. 1993;105(5):82332.
30. Anderson III HL, Cilley RE, Zwischenberger JB, Bartlett RH. Thrombocytopenia in neonates
after extracorporeal membrane oxygenation. ASAIO Trans. 1986;32(1):5347.
31.Robinson TM, Kickler TS, Walker LK, Ness P, Bell W. Effect of extracorporeal membrane
oxygenation on platelets in newborns. Crit Care Med. 1993;21(7):102934.
32. Stallion A, Cofer BR, Rafferty JA, Ziegler MM, Ryckman FC. The signicant relationship
between platelet count and haemorrhagic complications on ECMO. Perfusion. 1994;9(4):
2659.
33. Lamb KM, Cowan SW, Evans N, Pitcher H, Moritz T, Lazar M, et al. Successful management
of bleeding complications in patients supported with extracorporeal membrane oxygenation
with primary respiratory failure. Perfusion. 2013;28(2):12531.
34. Downard CD, Betit P, Chang RW, Garza JJ, Arnold JH, Wilson JM. Impact of AMICAR on
hemorrhagic complications of ECMO: a ten-year review. J Pediatr Surg. 2003;38(8):12126.
35. 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(4):53640. discussion 5401.
36. van der Staak FH, de Haan AF, 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(4):5949.
37. Velik-Salchner C, Sergi C, Fries D, Moser P, Streif W, Kolbitsch C. Use of recombinant factor
VIIa (novoseven) in combination with other coagulation products led to a thrombotic occlu-
sion of the truncus brachiocephalicus in a neonate supported by extracorporal membrane oxy-
genation. Anesth Analg. 2005;101(3):924.
38. Wittenstein B, Ng C, Ravin H, Goldman A. Recombinant factor VII for severe bleeding during
extracorporeal membrane oxygenation following open heart surgery. Pediatr Crit Care Med.
2005;6(4):4736.
39. Dominguez TE, Mitchell M, Friess SH, Huh JW, Manno CS, Ravishankar C. Use of recombi-
nant factor VIIa for refractory hemorrhage during extracorporeal membrane oxygenation.
Pediatr Crit Care Med. 2005;6(3):34851.
40. Repesse X, Au SM, Brechot N, Trouillet JL, Leprince P, Chastre J. Recombinant factor VIIa
for uncontrollable bleeding in patients with extracorporeal membrane oxygenation: report on
15 cases and literature review. Crit Care. 2013;17(2):R55.
41. 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(4):8524.
42. 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(5):16435.
210 C.L. Agerstrand et al.
43. Morgan IS, Codispoti M, Sanger K, Mankad PS. Superiority of centrifugal pump over roller
pump in paediatric cardiac surgery: prospective randomised trial. Eur J Cardiothorac Surg.
1998;13(5):52632.
44. Klein M, Dauben HP, Schulte HD, Gams E. Centrifugal pumping during routine open heart
surgery improves clinical outcome. Artif Organs. 1998;22(4):32636.
45. Parolari A, Alamanni F, Naliato M, Spirito R, Franze V, Pompilio G, et al. Adult cardiac sur-
gery outcomes: role of the pump type. Eur J Cardiothorac Surg. 2000;18(5):57582.
46. Khoshbin E, Roberts N, Harvey C, Machin D, Killer H, Peek GJ, et al. Poly-methyl pentene
oxygenators have improved gas exchange capability and reduced transfusion requirements in
adult extracorporeal membrane oxygenation. ASAIO J. 2005;51(3):2817.
47. Dogal NM, Mathis RK, Lin J, Qiu F, Kunselman A, Undar A. Evaluation of three hollow-ber
membrane oxygenators without integrated arterial lters for neonatal cardiopulmonary bypass.
Perfusion. 2012;27(2):13240.
48. 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 ber and at sheet membrane oxygenators. Artif Organs. 2000;24(1):438.
49. Mongero L, Beck J. On bypass: advanced perfusion techniques. New York: Springer; 2008.
50. Javidfar J, Brodie D, Costa J, LaVelle M, Miller J, Newmark A, et al. Subclavian artery can-
nulation for venoarterial extracorporeal membrane oxygenation. ASAIO J.
2012;58(5):4948.
51. Wang D, Zhou X, Liu X, Sidor B, Lynch J, Zwischenberger JB. Wang-Zwische double lumen
cannula-toward a percutaneous and ambulatory paracorporeal articial lung. ASAIO J.
2008;54(6):60611.
52. Vohra HA, Adamson L, Weeden DF, Haw MP. Use of extracorporeal membrane oxygenation
in the management of septic shock with severe cardiac dysfunction after Ravitch procedure.
Ann Thorac Surg. 2009;87(1):e45.
53. 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(5):70710.
54. Maclaren G, Butt W, Best D, Donath S, Taylor A. Extracorporeal membrane oxygenation for
refractory septic shock in children: one institutions experience. Pediatr Crit Care Med.
2007;8(5):44751.
55. Huang CT, Tsai YJ, Tsai PR, Ko WJ. Extracorporeal membrane oxygenation resuscitation in
adult patients with refractory septic shock. J Thorac Cardiovasc Surg. 2013;146(5):10416.
56. Brechot N, Luyt CE, Schmidt M, Leprince P, Trouillet JL, Leger P, et al. Venoarterial extracor-
poreal membrane oxygenation support for refractory cardiovascular dysfunction during severe
bacterial septic shock. Crit Care Med. 2013;41(7):161626.
Chapter 12
Mobilization During ECLS
Gregory A. Schmidt
Introduction
Technological Advances
Patients with severe ARDS can generally be treated with the VV, rather than veno-
arterial (VA), mode. For example, in the Conventional Ventilation or ECMO for
Severe Acute Respiratory Failure (CESAR) trial, all ECLS patients were treated
with the VV mode and none required conversion to VA ECLS [3] (and R. Firmin,
as early as the rst day of critical illness; and (2) weakness may persist for years.
These ndings suggest that early approaches to treatment and prevention might
yield the most gains.
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 up-
regulated. Also like in peripheral muscle, active contraction (that is, active breath-
ing) can effectively modify the degree of catabolism, helping to maintain contractile
function [12]. This has potentially important implications for the role of ECLS. By
freeing the diseased lungs from the burden of gas exchange, extracorporeal support
could allow less ventilation (even extubation, see Chap. 9) and, therefore, less seda-
tion, 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 inuences requires
changing our view of the critically ill from fragile beings in need of (over-) protec-
tion 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 ECLS.
This early experience has led many centers, not only to consider and trial ambu-
latory ECLS, but to actively develop and nurture specialized teams to aid mobiliza-
tion. Following the creation of a mobilization program aimed at ECLS patients, 35
of 100 patients with refractory respiratory or cardiac failure could be actively
rehabilitated [24]. Since bridge-to-recovery patients tend to be more ill and suffer-
ing 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 ECLS through a dual-lumen cannula in
the internal jugular vein, four patients were on VA ECLS (internal jugular vein to
subclavian artery) and eight had femoral access. Of the 16 bridge-to-recovery
patients, 14 survived, most going directly home following discharge. It should be
emphasized that these results followed an explicit plan to create and train a special-
ized team including physical and occupational therapists, perfusionists, critical care
RNs, critical care nurse practitioners, respiratory therapists, and ECLS intensivists
and surgeons.
It is no small task to safely mobilize patients during ECLS. Barriers include severity
of illness, resource and stafng limitations, equipment needs, risks of catastrophic
complications, and cultural barriers. There is broad agreement that mobilization is
safe for critically ill patients, including those with endotracheal tubes requiring
mechanical ventilation [25], but still very limited condence that we can safely do
this with the very sickest. Before mobilization, patients should be screened for con-
traindications, such as those described in the trial by Schweickert and colleagues as
modied for ECLS in Table 12.1 [26]. Many ECLS patients will have neurological
failure as a consequence of the underlying illness or its treatment, precluding extu-
bation, ambulation, or any meaningful therapy. Severe circulatory failure may also
be limiting, although many patients in shock and on vasoactive infusion can sit,
stand, and even walk [1]. In others, mobilization threatens 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-specic 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 reassuring. For example, in the small series cited above [17, 18] serious com-
plications were not seen. In the larger series of bridge-to-transplant and bridge-to-
recovery, there were no complications related to physical therapy [24]. Even in
the broader experience gained in generally critically ill patients, physical therapy
and mobilization have been remarkably complication-free [14, 27]. 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 peri-
ods of patientventilator dyssynchrony. Falls to knees without injury did not
216 G.A. Schmidt
Table 12.1 Safety screen: Mean arterial blood pressure <65 mmHg
contraindications to
Heart rate <40 or >130/min
mobilization
Respiratory rate >40/min
Critical hypoxemia despite ECMO (SaO2 <78 %)
Increased intracranial pressure
Active hemorrhage
Active myocardial ischemia
Undergoing a procedure
Agitation requiring increased sedative in the last 30 min
Delirium precluding meaningful cooperation
Insecure device, especially the ECLS cannula
Malfunction of cannula or circuit with movement
Modied from [26]
occur in two studies [1, 14], but occurred rarely in another [13]. Accidental extu-
bations 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 arrhythmia) and only 4 of 1110 required minimal additional treatment
or cost [27].
A nal barrier relates to each ICUs 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 stafng to include full-time
physical and occupational therapists succeeded in improving delirium, functional
mobility, and ICU length of stay [28]. 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 stafng is an effective means to change practice [29]. 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 benets are subtle and delayed, leading
some clinicians to urge rst do no harm. When an adverse event complicates
mobilization (and there is little doubt that an eventual patient will suffer a complica-
tion), there is a risk that an emotional, short-term-oriented response could threaten
an entire mobility program. The ECLS team must emphasize the data supporting
mobilization; acknowledge that rare patients may be harmed; and be prepared to
defend the mobility team if their role is questioned. In the event of a complication,
we encourage a just culture approach, rather than nger-pointing.
12 Mobilization During ECLS 217
Once a patient passes the safety screen described above, the mobility team should
be assembled. Safely ushering an ECLS 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: four caregivers at a mini-
mum and often several more (Fig. 12.1). 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. 8. 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 ECLS 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 [30]. 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 rm 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 ow, 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 ow should be
disconnected from wall oxygen, using a transport cylinder. Sweep gas ow 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 wheel-
chair 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 [31]. 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 distress, as
well as the ability to measure blood pressure. Any deterioration, including those
listed in Table 12.2, should prompt an immediate halt to mobilization, reassessment
of safety, and a decision whether to terminate the session. Returning the patient to
12 Mobilization During ECLS 219
the ICU bed requires careful planning and may benet 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 [32]. Electrical stimulation of muscles has been hypothesized to
improve strength and limit wasting, but efcacy remains unproved [33]. As the
acceptance of ICU mobilization increases, equipment is becoming widely available
to aid mobilization of critically ill patients and make it safer.
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
ECLS, 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, Nigos C, Pawlik AJ, Esbrook CL, et al. Early
physical and occupational therapy in mechanically ventilated, critically ill patients: a random-
ized controlled trial. Lancet. 2009;373:187482.
2. Engel HJ, Needham DM, Morris PE, Gropper MA. ICU early mobilization: from recommen-
dation to implementation at three medical centers. Crit Care Med. 2013;41:S6980.
3. Peek GJ, Mugford M, Tiruvoipati R, Wilson A, Allen E, Thalanany MM, et al. Efcacy 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;374:135163.
4. Kress JP, Pohlman AS, OConnor MF, Hall JB. Daily interruption of sedative infusions in criti-
cally ill patients undergoing mechanical ventilation. N Engl J Med. 2000;342:14717.
5. Riker RR, Shehabi Y, Bokesch PM, Ceraso D, Wisemandle W, Koura F, et al.
Dexmedetomidine vs. midazolam for sedation of critically ill patients: a randomized trial.
JAMA. 2009;301:48999.
6. Kress JP, Gehlbach B, Lacy M, Pliskin N, Pohlman AS, Hall JB. The long-term psychological
effects of daily sedative interruption on critically ill patients. Am J Respir Crit Care Med.
2003;168:145761.
7. Shehabi Y, Bellomo R, Reade MC, Bailey M, Bass F, Howe B, et al. Early intensive care seda-
tion predicts long-term mortality in ventilated critically ill patients. Am J Respir Crit Care
Med. 2012;186:72431.
8. Girard TD, Kress JP, Fuchs BD, Thomason JW, Schweickert WD, Pun BT, et al. Efcacy and
safety of a paired sedation and ventilator weaning protocol for mechanically ventilated patients
in intensive care (Awakening and Breathing Controlled trial): a randomised controlled trial.
Lancet. 2008;371:12634.
9. Fan E, Cheek F, Chlan L, Gosselink R, Hart N, Herridge MS, et al. An ofcial American
Thoracic Society clinical practice guideline: the diagnosis of intensive care unit-acquired
weakness in adults. Am J Respir Crit Care Med. 2014;190:143746.
220 G.A. Schmidt
29. Garzon-Serrano J, Ryan C, Waak K, Hirshberg R, Tully S, Bittner EA, et al. Early mobilization
in critically ill patients: patients mobilization level depends on health care providers profes-
sion. PM R. 2011;3:30713.
30. Yeung JH, Ong GJ, Davies RP, Gao F, Perkins GD. Factors affecting team leadership skill
and their relationship with quality of cardiopulmonary resuscitation. Crit Care Med.
2012;40:261721.
31. Hickmann CE, Roeseler J, Castanares-Zapatero D, Herrera EI, Mongodin A, Laterre PF.
Energy expenditure in the critically ill performing early physical therapy. Intensive Care Med.
2014;40:54855.
32. Burtin C, Clerckx B, Robbeets C, Ferdinande P, Langer D, Troosters T, et al. Early exercise in
critically ill patients enhances short-term functional recovery. Crit Care Med. 2009;37:
2499505.
33. Kho ME, Truong AD, Zanni JM, Ciesla ND, Brower RG, Palmer JB, et al. Neuromuscular
electrical stimulation in mechanically ventilated patients: a randomized, sham-controlled pilot
trial with blinded outcome assessment. J Crit Care. 2015;30:329.
Chapter 13
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,
oxygenator sweep gas, and ECMO ow during the weaning process; and cardiopul-
monary assessment during weaning. The process of weaning and trialing-off extra-
corporeal support is somewhat different for venovenous (VV) and venoarterial
(VA) ECMO, because of the difference in physiological support provided by the
two modalities. Further, we will discuss the procedure of decannulation and the
follow-up of patients after decannulation. Finally, we review the termination of
ECMO in patients where further therapy is deemed unrecoverable.
Signs of Readiness
Therapy directed toward improving pulmonary function starts soon after ARDS
patients are placed on ECMO support. Modalities that might help hasten recovery
in most patients include diuresis, antibiotic therapy, bronchodilators, and bronchos-
copy and airway evacuation. Ventilator management involves limiting FiO2 to 30 %,
maintaining PEEP to preserve recruitment, and limiting tidal volumes with low lev-
els of support.
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 typi-
cally used during extracorporeal support, an increase in PaO2 or a decrease in
PaCO2 might be noticed. Increased levels of CO2 in capnometry might also provide
a clue to enhanced alveolar gas exchange. Improved pulmonary compliance will
lead to increased tidal volumes on unchanged pressure-control ventilator settings.
Usually, respiratory system compliance >20 cc/cm H2O suggests adequate recovery
to proceed with weaning. An improvement in pulmonary aeration on chest radio-
graph often accompanies improving alveolar recruitment. Signs of improving lung
function are listed in Table 13.1.
Once signicant aeration is achieved, as assessed by improving tidal volumes
(>100150 cc) on rest settings and improvements in chest radiograph, the Cilley test
[1] should be performed daily. The test involves increasing the FiO2 to 100 %, with
no other changes in ventilator settings. A positive test is marked by a rapid increase
in oxygen saturation within a couple of minutes, and is a marker for readiness to
advance ventilator settings.
In ARDS patients on VA-ECMO support for cardiac failure in addition to respira-
tory failure, cardiac recovery would usually follow with improved pulmonary func-
tion, especially if the cardiac failure is related to acute right ventricular dysfunction.
Improved gas exchange, decreased ventilatory airway pressures (on volume-preset
modes), and decreased pulmonary vascular resistance will lead to enhanced right
ventricular function. As ventricular function improves and native cardiac output
carries blood forward, an increase in pulse pressure becomes evident.
ARDS patients with improving pulmonary function can be weaned and liberated
from extracorporeal support. On occasion, the decision to wean off extracorporeal
support might be hastened by bleeding complications related to the anticoagulation
necessary to allow blood ow through the extracorporeal circuit. Weaning and lib-
eration from ECMO is managed with continuous cardiopulmonary assessments,
while changes are made in ventilator settings, ECMO ow, and oxygenator sweep
gas ow (Fig. 13.1). Weaning involves decreasing extracorporeal support and evalu-
ating cardiopulmonary recovery without ECMO. Trialing-off involves temporary
discontinuation of extracorporeal support to prove that cardiopulmonary function
has recovered to the point that it can be supported without ECMO. Finally, libera-
tion 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 function, it is possible to trial off ECMO [2].
Ventilator Management
When the decision is made to advance from rest settings, recruitment maneuvers
might be required before lung-protective ventilator settings are employed. Further
weaning is only possible when the patient is able to achieve adequate tidal volumes
(approximately 6 mL/kg) with acceptable airway pressure (plateau airway pressure
<30 cm H2O). Adjusting sedation to allow spontaneous breathing assists in improv-
ing lung recruitment and weaning. Adequate minute ventilation will have to be
established before patients can be weaned or trialed off extracorporeal support.
As ECMO support for gas exchange is decreased, the patients respiratory rate
will likely increase. The magnitude of increase in work of breathing during weaning
is an indication of the amount of extracorporeal support the patient was receiving in
addition to their native lung function. A high respiratory rate (>30/min), low tidal
volumes, or visible distress are signs that the patient is not ready for trialing off
ECMO. Ideally, the patient should be supportable with minimal ventilator settings,
with the ability to increase ventilatory support if needed, once extracorporeal sup-
port is removed.
In some patients, it might be possible to decrease sedation to the point of com-
plete wakefulness while maintaining adequate gas exchange on extracorporeal
support. At this point, the endotracheal tube remains the only major indication for
sedation, and extubation can be considered before weaning ECMO ow.
226 S. Krishnan and G.A. Schmidt
Recruitment maneuvers
Are
No respiratory
mechanics
acceptable?
Yes
Maintain sweep gas FiO2 at 100%. Decrease sweep gas flow in steps to 1 lpm
Are
No hemodynamic
and respiratory
parameters
stable?
Yes
Fig. 13.1 Algorithm for weaning ECLS in the ventilated ARDS patient on VV-ECMO
might lead to discontinuing ECMO sooner than planned. On the other hand, for
patients who require signicant sedation to tolerate the endotracheal tube, extuba-
tion with continued ECMO support might assist in mobilization and enhanced
recovery. The institutions ability to provide prolonged ECMO support must be
taken into account in patients expected to require respiratory support for a long
duration. In addition, the condition of the ECMO circuit is a paramount consider-
ation before withdrawing ventilator support. Finally, the intensive care unit should
be able to provide emergency ventilatory support in case of ECMO failure in an
extubated patient.
Once adequate tidal volumes are possible on lung-protective ventilator settings, the
sweep gas ow is reduced in steps to 1 L/min and sweep FiO2 is maintained at 100
%. Alternatively, it is also acceptable to decrease sweep gas ow to 2 L/min then
decrease sweep FiO2 to maintain SaO2 >95 %. When gas exchange is stable at these
settings, the patient is trialed off ECMO by turning off the sweep gas altogether and
capping off the oxygenator. Gas exchange, work of breathing, and hemodynamic
status should be assessed during this trial. There needs to be no interruption in blood
ow in the extracorporeal circuit during a trial off VV-ECMO. The return tubing
from the oxygenator would soon change in color from oxygenated blood to deoxy-
genated blood, similar to the tubing draining blood from the patient. If the patient
remains stable over the next hour or 2, they have successfully passed their trial off
ECMO and are ready for decannulation. If the patient shows signs of inadequate
native pulmonary function, the sweep gas ow is simply turned back on and extra-
corporeal support is resumed. Ventilatory settings and sedation can be managed to
the patients comfort, until the next trial-off period.
In patients on VA-ECMO support, oxygenator sweep gas ow is only titrated to
PaCO2, and not decreased for purposes of weaning and assessing the degree of car-
diopulmonary recovery. Decreasing oxygenator sweep gas ow would lead to inad-
equate oxygenation of blood owing through the VA-ECMO circuit, causing
right-to-left shunting.
Circuit blood ow does not need to be changed while attempting weaning from
VV-ECMO. However, as the patients intravascular volume is decreased with diure-
sis, it might be difcult to maintain a high rate of blood ow through the extracor-
poreal circuit. As the patients pulmonary gas exchange improves, it becomes
possible to decrease circuit blood ow. A minimum ow rate of 5001000 mL/min
needs to be maintained to prevent stasis of blood in the cannulae, circuit, pump, and
oxygenator.
228 S. Krishnan and G.A. Schmidt
Anticoagulation Management
Cardiopulmonary Assessment
As patients are weaned off extracorporeal support, cardiac and pulmonary workload
increases. Cardiopulmonary function should be closely monitored during this pro-
cess. As alluded to earlier in this chapter, respiratory rate, tidal volume, gas
exchange, and respiratory distress are vital clues to pulmonary function during
weaning and trialing off from ECMO, much as one would judge the success of a
spontaneous breathing trial. It is important to note that in patients on VA-ECMO
support, ashing introduces oxygenated blood into the patient. Hence gas exchange
assessments should be made prior to ashing the cannulae.
In patients on VA-ECMO, cardiac function assessments are made as ECMO ow
is decreased. On echocardiography, the heart is empty while the patient is on full
13 ECMO Weaning and Decannulation 229
Decannulation
Following Decannulation
Since patients have demonstrated adequate gas exchange (and hemodynamic stabil-
ity) 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. Procedures that were withheld because of the anti-
coagulation necessary for extracorporeal support (e.g., tracheostomy, chest tube
placement, invasive line placement) can now be performed.
230 S. Krishnan and G.A. Schmidt
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 (uid overload, thromboembolism), performing appropriate diagnos-
tic studies, and using the ventilator to support respiratory function (or, for the extu-
bated patient, non-invasive ventilation or re-intubation). In the 2448 h after
decannulation, patients also need to be assessed for venous thrombosis at the can-
nula site.
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
illness, such as nosocomial pneumonia, septic shock, or acute renal failure turn a
desperate situation into an unrecoverable downward spiral (Table 13.2). 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 [3, 4], perhaps more so in ECLS patients [5, 6]. Difcult 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 [6]. Laying out goals for
patients who are being bridged to recovery is more challenging, because recovery
can be protracted, hard to predict accurately, and the course may be punctuated with
periodic downturns that make survival seem unlikely. At the time that consent for
ECLS is obtained, it is appropriate to raise the possibility that it will fail to achieve
its goals and may have to be withdrawn. During the treatment course, new compli-
cations or additional organ failures should serve as opportunities to revisit the goals,
judging anew whether they are still attainable. Even without such events, 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) [7]
can be helpful for patients and families.
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. For a patient with decisional capacity who does not
accept the recommendation to withdraw ECLS, this choice should be respected,
even when it seems a poor use of resources [6].
Occasionally, the balance of benets and burdens shifts unfavorably as the
patient or family succumbs to the stress of chronic critical illness. Some patients
will request withdrawal of a life-sustaining device even while the clinicians judge
that the overall picture is improving, a situation that carries a high emotional burden
for both patient and caregivers [8]. Ethicists generally accept the moral equivalence
of withdrawing and withholding life-sustaining treatments for the critically ill, such
as dialysis and mechanical ventilation. Whether similar reasoning applies to circula-
tory support devices is less clear, especially when these treatments are highly effec-
tive in sustaining life for many weeks or months. For example, a substantial minority
of health care providers and many patients consider deactivating implantable car-
diac devices such as pacemakers or total articial hearts to be euthanasia or
physician-assisted suicide [9]. Since ECLS is never instituted as destination therapy,
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.
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 [10]. Patient
comfort should be assured, using analgesics and sedatives as needed, and anticipat-
ing that withdrawal of ECLS may provoke or heighten dyspnea. We then turn off the
sweep gas, preventing further gas exchange in the membrane and, assuming ECLS
was necessary, leading to death. For those on VA ECLS, circuit blood ow can be
reduced as vasoactive drug infusions are stopped.
References
1. ELSO Adult Respiratory Failure Supplement to the ELSO General Guidelines. December
2013. https://www.elso.org/Portals/0/IGD/Archive/FileManager/989d4d4d14cusersshyerdoc
umentselsoguidelinesforadultrespiratoryfailure1.3.pdf. Accessed Jan 2015.
2. Lynch WR. Weaning, trialing and futility. In: ECMO: extracorporeal cardiopulmonary support
in critical care. 4th ed. ELSO; 2012.
3. Luce JM. A history of resolving conicts over end-of-life care in intensive care units in the
United States. Crit Care Med. 2010;38:16239.
4. Siegel MD. End-of-life decision making in the ICU. Clin Chest Med. 2009;30:18194.
5. 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;59:64250.
6. Abrams DC, Prager K, Blinderman CD, Burkart KM, Brodie D. Ethical dilemmas encountered
with the use of extracorporeal membrane oxygenation in adults. Chest. 2014;145:87682.
7. Schenker Y, Tiver GA, Hong SY, White DB. Discussion of treatment trials in intensive care. J
Crit Care. 2013;28:8629.
8. Bruce CR, Allen NG, Fahy BN, Gordon HL, Suarez EE, Bruckner BA. Challenges in deacti-
vating a total articial heart for a patient with capacity. Chest. 2014;145:62531.
9. Rady MY, Verheijde JL. Ethical challenges with deactivation of durable mechanical circula-
tory support at the end of life: left ventricular assist devices and total articial hearts. J
Intensive Care Med. 2014;29:312.
10. Cook D, Rocker G. Dying with dignity in the intensive care unit. N Engl J Med.
2014;370:250614.
Chapter 14
The Story of ECLS: History and Future
Introduction
Cardiopulmonary bypass (CPB) for cardiac surgery was rst introduced in the 1950s
[1]. In the 1970s, CPB use at the bedside for pulmonary support began and was termed
extracorporeal membrane oxygenation (ECMO) [2]. Later, in the 1980s, applications
for extracorporeal technology expanded, and the term extracorporeal life support
(ECLS) was coined by researchers at University of Michigan as an umbrella term for
all forms of extracorporeal technologies. These technologies include oxygenation,
CO2 removal, and hemodynamic support, as well as supportive techniques for the
kidneys and liver. Extracorporeal membrane carbon dioxide removal (ECCO2R) is the
term used for the selective removal of CO2 [3]. Despite the other terms, ECMO is most
often used to refer to all forms of extracorporeal support if gas exchange is involved.
ECMO is a rapidly evolving eld that was developed for critically ill patients
with respiratory failure more than 40 years ago. ECMO currently serves as a form
of life support for critically ill patients when traditional supportive care is no longer
effective. Recently, it has also become a bridge to transplantation and a way to man-
age acute shock. Other patient groups with diverse diseases, such as H1N1, are also
receiving ECMO as part of their treatment program for acute respiratory distress
syndrome (ARDS). While technological advances and better management of criti-
cally ill patients allow the option of ventilator support in some patients, ECMO is
still the main option for temporary cardiopulmonary support in numerous patient
populations [4]. In general, ECMO should be considered if the disease or trauma
meets the following criteria: (1) severe, (2) acute, and (3) potentially reversible [5].
The development of cardiopulmonary bypass (CPB) laid the groundwork for the
establishment of ECMO as a viable treatment option (Table 14.1) [6]. With CPB,
the blood circulation and gas exchange (CO2 removal and oxygenation) are tempo-
rarily supported by an external mechanical device, allowing surgeons to operate on
a nonbeating, bloodless heart. The medical community soon realized the necessity
of a bypass machine to support lung and heart function during more complex sur-
geries, such as congenital heart defects.
CPB was rst devised for mechanical cardiopulmonary support in the operating
room by John Gibbon (Fig. 14.1). During his time in Boston in the early 1930s,
Gibbon monitored a patient that became hypotensive and short of breath following
a cholecystectomy. Gibbon thought that the patient might recover if her venous
blood was oxygenated by an external source. Although the patient did not survive,
Gibbon persisted. In 1934, he constructed a rudimentary heartlung machine capa-
ble of fully supporting circulation in a feline model for 30 min.
On September 2, 1952, the rst successful intracardiac procedure with direct
vision was performed by Dr. F. John Lewis and Dr. C. Walton Lillehei at the
University of Minnesota [7]. Dr. Lewis repaired an atrial septal defect (ASD) in a
5-year-old patient via full-body hypothermia and in-ow stasis. The previous year,
Dr. Clarence Dennis performed two unsuccessful intracardiac procedures at the
University of Minnesota. These were the rst attempts at total CPB using a heart
lung device developed by Dr. Dennis and his colleagues [8]. According to Dr.
Lillehei, these attempts of Dr. Dennis failed because of the high perfusion rates
required by the heartlung device [9]. Other complications included excessive
bleeding which obscured the visual eld, incorrect preoperative diagnoses, and air
emboli from inadvertent evacuation of total arterial reservoir volume.
Gibbons heartlung machine (Fig. 14.2) was used successfully on May 6, 1953,
at the Jefferson Hospital in Philadelphia during the operation of an 18-year-old
woman with an ASD [10]. Interestingly, Gibbon once stated, After we nally got
ready, it was ridiculously easy to repair the ASD [8]. Yet, the operation was not
quite that simple. At one point during the procedure, there was reduced extracorpo-
real blood ow, loss of circulating volume, and thrombosis in the CPB circuit,
resulting in prolonged hypotension, hypoxia, and acidosis [10]. Despite these com-
plications, the patient survived. Dr. Gibbon subsequently applied his CPB pump
with a screen/lm oxygenator four more times, all of which were failures [10]. Once
again, these failures related directly to incorrect preoperative diagnoses and intra-
cardiac bleeding which visually obscured the surgical eld. These pioneering expe-
riences highlighted the need for further research to develop both the CPB procedure,
a safer CPB machine with diminished thrombosis risk, and improved diagnostics.
Nevertheless, the events of May 6, 1953, laid the foundation for the future of extra-
corporeal membrane oxygenators.
14 The Story of ECLS: History and Future 235
In the early 1950s, there was considerable pessimism regarding the feasibility
of CPB. Of the reported cases from 1952 to 1954 that utilized total CPB, only 1 out
of 18 cases (5.5 %) had been successful. Many felt the magnitude of the proce-
dures performed and the multiple comorbidities of patients requiring open heart
surgery made recovery of the surgically repaired heart impossible. Thus, the idea
of long-term cardiopulmonary support emerged. Clinicians realized that if long-
term cardiopulmonary support could be achieved, cardiac repair outcomes could
be greatly improved, and other previously untreatable cardiorespiratory diseases
could be overcome.
14 The Story of ECLS: History and Future 237
Along with the reported failures of early CPB machines was the discovery of the
azygous ow concept (low ow concept) [1, 9]. Lillehei and his colleagues utilized
canine experimentation models to prove that the accepted cardiac output ow rate
for CPB of 100165 mL/kg/min was much higher than actually necessary. Canines
were surgically oriented such that only the azygous vein would return blood to the
heart with ow rates of only 814 mL/kg/min. Internal organs of these animals were
well protected during these trials. From this work, Lillehei concluded that success-
ful perfusion could be achieved with 2025 mL/kg/min [9].
Reduced perfusion rates enabled clinicians to further rene the practice of using
autogenous lungs [11]. The use of nonhuman donor lungs was pioneered by Dr.
William Mustard who made use of monkey lungs in the CPB device that he and his
team applied unsuccessfully ve times from 1951 to 1953 [12]. Swift development
of pulmonary edema often followed any perturbations of the many venous catheters
required in the surgical eld during such procedures. In order to preclude such com-
plications, Lillehei and colleagues initially found during experimental trials that they
could use secondary canine subjects as cross-circulatory vehicles during intracardiac
procedures on their primary subjects. Other studies showed that cross-circulation
could provide adequate support based on the azygous (low) ow concept [1, 9].
The application of cross-circulation to human cases allowed the idea of direct-
vision intracardiac procedures to gain acceptance within the medical community.
Until then, only hypothermia with inow stasis had positive outcomes. The cross-
circulation method, which involved the use of a volunteer as a living oxygenator,
was used exclusively and extensively from its inception in March 1954 for all but the
simplest cases, which still utilized either early bubble oxygenators or autogenous
lung support. From March 26, 1954 until July 9, 1955, Dr. C. Walton Lillehei and
coworkers utilized cross-circulation with human lung donors to achieve blood oxy-
genation in 45 infants and children undergoing operations for cardiac anomalies.
The human donors lungs served as the extracorporeal oxygenator, allowing Lillehei
to operate on a large range of patients. The following abnormalities were repaired:
ventricular septal defect (VSD), patent ductus arteriosus (PDA) (with pulmonary
hypertension), tetralogy of Fallot, atrioventricular communis, isolated infundibular
pulmonary stenosis, pulmonary stenosis, ASD, and anomalous pulmonary venous
return. Remarkably, 28 patients survived Lilleheis cutting-edge surgery. Throughout
these trials, perfusion rates of 2530 mL/kg/min adequately supported the patients
[1317]. The widespread acceptance that gross congenital heart defects could be
effectively repaired when provided with adequate extracorporeal support spurred
many clinicians and researchers to further develop existing technology.
Dr. Gibbon demonstrated that a metal lm oxygenator was effective [1, 18]. This
device was composed of a metal cylinder which rotated while partially submerged
in a blood reservoir. This allowed blood to lm over the rotating cylinder, exposing
238 J.A. Morris et al.
the blood to an oxygen-rich atmosphere in which the cylinder and blood reservoir
were enclosed. Gibbons device was then modied for use in humans. The upgrade
that allowed for larger perfusion volumes consisted of a vertically-oriented sequence
of stationary metal discs over which blood was allowed to ow downwards. This
set-up created a large surface area of blood lm which was directly exposed to oxy-
gen in an enclosed atmosphere. This device was successfully applied in the rst
total body perfusion via external oxygenator pump during an intracardiac repair
[19]. Gibbon successfully applied this version of his pump only once.
Dr. J.W. Kirklin and his colleagues at the Mayo Clinic in Rochester, Minnesota
obtained a Gibbon pump apparatus and modied it to meet their updated specica-
tions [17]. In March 1955, Kirklin and his colleagues used their Mayo-Gibbon
pump oxygenator in a series of landmark intracardiac procedures (245 patients by
September 1958) and were so successful that the device was commercialized [15,
16, 20, 21]. In contrast, Lillehei and others reported that the number of teams who
could effectively deploy the Mayo-Gibbon pump oxygenator was extremely limited
[20, 21]. This was due to the pumps complexity, cost-prohibitive construction,
complicated maintenance/sterilization routines, and the requirement for very expe-
rienced technicians for even basic operation [17].
To achieve clinical acceptance, a new device must be manufactured, distributed,
sterilized, and deployed cheaply. In May 1953, the DeWall-Lillehei bubble oxygen-
ator was the rst device that satised these requirements [22]. Lillehei and DeWall
based their device on the earlier designs of Clarke, Golan, and Gupta which had been
introduced on a smaller scale beginning in 1950 [23, 24]. The original DeWall-
Lillehei bubble oxygenator functioned as follows: (1) blood and oxygen are sepa-
rately introduced into a vertical oxygenating chamber; (2) the blood and oxygen mix
as they ow upwards and enter a defoaming chamber where the surface tension in the
bubbles present in the blood is released; (3) the blood is then pumped down into a
reservoir submersed in a liquid medium which acts as the heat exchanger; and (4)
excess bubbles are lost in an upward diffusion as the blood is pumped down before
exiting the reservoir and reentering the body [22, 23]. A Sigma motor pump was used
with the DeWall-Lillehei bubble oxygenator. This pump was composed of several
actuating rods/ngers and was so named for the undulating, sinusoidal S-shape
taken on by the rods while they massaged the circuit tubing to induce blood ow [23].
The safety, reliability, and effectiveness of the DeWall-Lillehei oxygenator
were shown in numerous reports leading to rapid acceptance [22, 23, 25]. Dr.
Lillehei states (and others have conrmed) that the DeWall-Lillehei bubble oxy-
genator achieved instant success because it was efcient, inexpensive, heat
sterilizable, easy to assemble, had no moving parts, and was disposable [26].
Improved versions of the DeWall-Lillehei bubble oxygenator consisted of a single,
unitized plastic sheet which came as pre-sterilized, ready-to-assemble, and com-
pletely disposable kits [27, 28]. At the University of Minnesota Hospital between
1953 and 1966, a total of 2715 cardiotomy procedures were performed with 2581
of these cases utilizing bubble oxygenators [29]. In 1966, DeWall and colleagues
introduced the Temptrol (Bentley) disposable blood oxygenator which was an inte-
grated device that combined oxygenator and heat exchanger into one small device.
14 The Story of ECLS: History and Future 239
Early membrane oxygenators had proven gas exchange ability. However, these
devices suffered from limitations of existing materials: (1) poor gas permeability,
(2) weak structural integrity, (3) difculty manufacturing, (4) blood-surface incom-
patibility, (5) membrane acting as an additional barrier to diffusive exchange, and
(6) difculty achieving evenly distributed blood ow. Researchers are still strug-
gling with these obstacles today.
240 J.A. Morris et al.
In 1944, Kolff and Berk were testing their hemodialysis machine design when
they observed gas exchange in blood via passive diffusion across the large surface
area cellophane tubing [39]. By 1956, this observation led to creation of a mem-
brane gas exchanger composed of polyethylene [34]. Encouraged by the 1944 nd-
ings of Kolff and Berk, Clowes and colleagues were experimenting with materials
such as ethyl cellulose and polyethylene in the hopes of creating a membrane gas
exchanger [40, 41]. In 1956, they used Teon for their rst clinically applicable
gas exchanger [42].
Early membrane oxygenators were less traumatic to blood than direct-contact
oxygenators, but also had limitations. The early materials used to create lung mem-
branes were hydrophilic, and this greatly exacerbated plasma leakage with extended
use. Therefore, designers began to incorporate hydrophobic materials, such as
Teon, to prevent excessive plasma leakage [42, 43]. Unfortunately, the solubility
of CO2 in hydrophobic materials is much less than in hydrophilic materials. The
term membrane lung was then used to emphasize the primary goal of membrane
oxygenator designers which was efcient O2 and CO2 exchange similar to that of
anatomical lungs, not just the exchange of O2 alone.
In 1957, Kammermeyer reported that O2 and CO2 are 40 times more permeable
in silicone rubber (dimethylpolysiloxane polymer) than in Teon and that CO2
permeability is 8 times that of O2 in silicone rubber [43]. Designers quickly began
experimenting with silicone rubber to nd a solution to its low structural capacity
and tendency for forming pinholes. In 1959, Thomas developed a method to support
a thin, silicone rubber membrane on a mesh of nylon ber using a dip-coating
method. Two membranes were used in conjunction (layered) to alleviate the prob-
lem of plasma leakage through the pinhole defects. That same year, Dr. Nora Burns
was able to produce adequately thin, pinhole-free, and structurally sound silicone
rubber under the tutelage of Professor Dennis Melrose at Hammersmith Hospital,
London [44]. This material was so revolutionary and effective that its modern equiv-
alent is still employed today in long-term use devices that are used both experimen-
tally and clinically.
Very shortly after Burns innovation, many prominent researchers began to
employ silicone rubber membranes in their devices, either supported by nylon or
integral silicate material [45]. In 1960, E. Converse Pierce II introduced an updated
version of Clowes device [46]. Kolobow and Bowman published their work in
1963 on what they termed an alveolar membrane articial heartlung [47]. The
term alveolar membrane refers specically to the ability of silicone rubber
membranes to effectively replicate natural alveolar tissue. Bramson et al. produced
a silicone rubber membrane lung (SRML) oxygenator with an integrated heat-
exchanger in 1965 [48]. Land et al. submitted their experience utilizing membrane
materials in their oxygenator-dialyzer in 1967 [49]. That same year, Day et al. pub-
lished their investigations of cellophane and silicone rubber membranes [50].
These investigations, and many others, demonstrated that SRMLs were not
achieving expected rates of gas exchange. Oxygen is capable of being exchanged at
a rate of 12,100 cc O2/m2/min through a 0.001 in. thick silicone rubber membrane
and 300 cc O2/m2/min through a 0.004 in. thick membrane (standard thickness for
14 The Story of ECLS: History and Future 241
the thin membranes introduced by Burns). CO2 is exchanged at a much higher rate.
Regardless of the physical deposition of each devices membranes (sheets or chan-
nels), researchers were only able to achieve around 30 cc O2/m2/min [50]. The arti-
cial materials used were not themselves the source of this discrepancy. Rather, it
was the blood that was creating this gas exchange barrier. In 1960, Marx et al.
showed that blood diffusion into a lm is proportional to the square of the thick-
ness of the blood lm and the diffusion resistance of the boundary layer [51]. More
succinctly, gas exchanges easily into the surface of a blood volume but diffusivity of
gas through blood itself is much slower. In order to facilitate more efcient gas dif-
fusion, designers sought a way to narrow the blood channels within SRML oxygen-
ator devices and to create secondary ow pathways.
Bodell et al. provided an innovative solution to one half of this problem in 1963
with the rst capillary oxygenator which utilized tubes between 100 and 500 m in
diameter as channels to alleviate the problem of thick blood streams as a barrier to
blood diffusivity [52]. Two orientations were investigated. In the extracapillary ori-
entation, gas was directed into capillaries while blood was directed around capillar-
ies. In the intracapillary orientation, blood was directed into capillaries and gas was
directed around capillaries. By 1965, Wilson et al. determined that an intracapillary
orientation was more effective given the restrictions of available materials [53]. The
intracapillary hollow-ber membrane lung delivered a twofold advantage. Gas dif-
fusivity through constricted channels was akin to the one-cell thick capillaries of the
natural alveolus, and there was more exacting control over the volumes in the blood
and gas chambers.
Various methods were devised in an effort to solve the other half of the blood
diffusivity issue and induce secondary ows throughout SRML oxygenators. Two
teams, the Dorson and Peirce groups, sought to create secondary ows passively.
Dorsons group directed blood into a helical tube arrangement while Peirces
group forced blood around intrusive conical elements, regularly interspersed
0.040 in. apart, in an envelope-style membrane [54, 55]. These efforts were efca-
cious, but lacked the dramatic results reported by other research teams operating at
the same time.
Kolobow et al. demonstrated foresight when they found a solution to the problem
of secondary ow generation as early as 1963 [56]. This group investigated the use
of hypobaric oxygenation to increase membrane surface area which had the added
benet of actively inducing secondary blood ows. Application of intermittent suc-
tion allowed for cyclic deformation of the exible membrane material into the gas
chamber of the device, creating larger surface areas for diffusion and inducing sec-
ondary ows that dramatically increased oxygen diffusion rates. Kolobow et al.
reported oxygen diffusion rates in excess of 200 cc O2/m2/min [56]. The Kolobow
device was manufactured rst by SciMed, then AveCor, and nally MedTronic. It
became the only commercially available solid silicone rubber membrane lung
(SRML) device to be consistently utilized over several decades. At their peak,
SRMLs were so effective they were used in 90 % of all reported ECMO treatments
conducted worldwide.
242 J.A. Morris et al.
SRML equivalents [65]. PMP devices with infused heparin coatings signicantly
outperformed existing modalities to such a degree that some ECMO centers
switched to their use exclusively. Moreover, in 2005, Toomasian et al. at the
University of Michigan reported that PMP devices had better gas exchange, more
platelet preservation, and appreciably lower pressure gradients than commercially
available descendants of the Kolobow coil-type SRML [66]. These advantages
make PMP membranes ideal for extended ECMO/ECLS. The low resistance of
PMP oxygenators makes them ideally suited for use in tandem with centrifugal
pumps or even for use in pumpless arteriovenous ECMO protocols [65, 66].
The success of new nonmicroporous materials such as PMP has encouraged
innovation along similar lines in older nonmicroporous materials. Motomura et al.
from Fuji Systems Corporation in Yokohama, Japan, in conjunction with their asso-
ciates at Baylor College of Medicine, Texas, have developed a silicone rubber
hollow-ber oxygenator (SRHF) with ultra-thin membrane walls (35 m thick)
[67]. Their device, the Fuji-B.C.M. ECMO oxygenator module, outperforms tradi-
tional coil-type SRML oxygenators, both in terms of gas exchange and prevention
of hemolysis while still maintaining blood compatibility. Also, the plasma leakage
that characterizes microporous polypropylene membranes during long-term use is
absent in their device. This ultra-thin SRHF device provides an alternative to the
successful PMP devices [65, 67].
ECMO Configurations
VA ECMO
In the early days of ECMO, the VA mode was used almost exclusively since ECMO
developed from cardiopulmonary bypass. In VA ECMO, deoxygenated blood drains
from the central venous system most often accessing the right internal jugular vein,
244 J.A. Morris et al.
is pumped through the gas exchanger, and then returned to the patient through the
right common carotid artery into the aortic arch [69]. VA ECMO provides both
cardiac and respiratory support and is frequently used in patients who are hemody-
namically unstable, are post-cardiotomy, or have heart failure.
VV ECMO
in the care of critically ill patients, other than low tidal volume ventilation, ventilator
strategies have not improved long-term survival of ARDS patients. Following the
success of ECMO in the neonatal population (see below) and its expanded use
throughout the late 1970s and 1980s, physicians turned to ECMO once again in the
treatment of adult ARDS in the late 1980s and early 1990s [85]. Further support for
ECMO in adult patients came from: (1) the positive results of a European study that
analyzed the use of a pumpless extracorporeal lung assist device over the course of
a decade [90], (2) advancements in ECMO technology and management practices,
and (3) general improvements in critical care [89]. Together, these factors spurred
clinicians to reevaluate ECMO in adult ARDS patients in the late 1990s.
Initially, small studies and case reports predominated, with studies such as one by
Kolla et al. [91] that reported a 54 % survival (n = 100) in adult ARDS patients. More
conclusive evidence supporting ECMO as a management therapy in ARDS came in
2009, when a large multicenter randomized controlled trial, the CESAR
(Conventional Ventilatory Support versus Extracorporeal Membrane Oxygenation
for Severe Adult Respiratory Failure) trial, showed a 16 % increase in 6-month sur-
vival without severe disability for the ECMO group versus the conventional treat-
ment group (63 % vs. 47 %) [92]. However, this study was not without limitations.
All patients were treated by best therapy at the local hospital, or referred to a sin-
gle ECMO center and thus treated by the same staff [92]. Utilizing a single referral
center diminished variability in administering ECMO, but imposed transportation
morbidity and mortality and is not a realistic approach for countries with more than
one ECMO center [88]. Moreover, many ARDS trials have shown that survival with
a low tidal volume critical-care protocol is approximately 60 % [93, 94]. Critics
noted the modest benets seen in the CESAR trial may not support the inclusion of
ECMO in the standard critical care protocols for adult ARDS patients.
During the 2009 H1N1 pandemic, many hospitals successfully employed VV
ECMO as a rescue treatment for ARDS [9598]. In this VV ECMO experience,
68 % of the H1N1 patients survived following ECMO treatment [98]. Among these
patients, mortality risk was correlated with extrapulmonary organ function just prior
to receiving the VV ECMO treatment [99]. Moreover, this study highlighted the
feasibility and efcacy of using a national VV ECMO protocol at 14 different ter-
tiary centers [98]. A recent upsurge in H1N1 patients in winter of 20132014 has
seen a much broader application of VV ECMO treatment at an earlier stage of
ARDS. The initial results appear promising but outcomes data is not yet available.
In 1975, an illiterate Mexican teenager crossed the U.S. border to give birth to a
baby girl. On her way to Los Angeles from Baja, Mexico, the young mother went
into labor. After a long and difcult labor at Orange County Medical Center, the
newborn was hypoxic, hypotensive, and experiencing seizures. Her arterial pO2 was
14 The Story of ECLS: History and Future 247
12 [99]. At the time, Dr. Bartlett and his colleagues were working on developing
ECMO in adult patients. Until that night, ECMO had been perceived as
unsuccessfulapproximately 150 adult patients had been placed on ECMO, with
only a 1015 % survival rate [100]. Moreover, ECMO had never been employed
successfully in children. Yet, the baby, Esperanza, had run out of options and Dr.
Bartlett was called [99] (Fig. 14.3).
This rst neonatal case would change the face of ECMO forever. After the
infants mother consented, the cyanotic infant was placed on VA ECMO using
right jugular vein-carotid artery cannulation. At rst, Esperanzas lung function
was not recovering while on ECMO. Bartlett ligated the patent ductus arterio-
sus. Once Esperanzas pulmonary hypertension subsided, she was slowly taken
off life support. Now, almost 40 years later, Esperanza has grown into a mature
woman (Fig. 14.4). Had it not been for this opportunistic neonatal case, Dr. Bartlett
believes that their ECMO research would have been abandoned (Bartlett, unpub-
lished interview).
Following the success of ECMO in Esperanzas case, Bartlett published an
ECMO study on 45 neonates in 1982 [101]. In this study, the newborns selected for
ECMO treatment had already failed to recover on traditional therapies and were
labeled as moribund. Bartlett reported a greater than 50 % survival in these newborn
248 J.A. Morris et al.
Fig. 14.4 (a) Photograph of Esperanza while she was on ECMO in 1975. (b) Photograph of Dr.
Bartlett and Esperanza at age 34. Photos courtesy of Dr. Bartlett, with permission
PPHN treatments between 1982 and 1983 at two Harvard hospitals revealed that
11 of 13 (85 %) infants suffering from PPHN and treated with conventional ventila-
tor therapies had died [104]. These ndings suggested that ECMO could signi-
cantly increase survival rates when compared to conventional ventilator therapies in
specic patient populations. Consequently, Pearl ORourke conducted a two-phase
trial to study ECMO versus conventional medical therapy (CMT). In total, 39 new-
borns with severe PPH and respiratory failure that had an 85 % chance of dying
were randomly assigned to either ECMO or conventional medical treatment (CMT)
groups [104]. After the fourth death in either treatment group, randomization would
be suspended and future patients would receive the superior treatment during phase
II. Phase II would similarly be suspended if four deaths occurred or until it was
shown that the phase II treatment was signicantly better than the suspended treat-
ment protocol. During the rst phase, all nine infants survived the ECMO treatment
while only six out of ten infants survived CMT. After the fourth CMT death, phase
II began in which all patients received ECMO. In the second phase, 19 out of 20
infants treated with ECMO survived. Overall, 97 % (28/29) of the patients survived
ECMO treatment, while just 60 % (6/10) survived CMT. ORourke encountered
constant criticism from either those who thought the study should be a 50:50 trial or
those who thought the control group was unethical.
Between 1993 and 1995, a large-scale randomized trial of neonatal ECMO in
mature (at least 35 weeks gestation and birth weight of at least 2 kg) newborn infants
with severe respiratory failure was undertaken in the United Kingdom (UK) [105].
This study enrolled 185 patients from 55 hospitals in which patients were either
maintained at their original hospital for Conventional Medical Therapy (CMT) or
referred to one of ve specialist centers for ECMO treatment. Importantly, there were
numerous causes of respiratory failure among this patient population, allowing
observation of subgroup trends. Overall, 63 of 93 (68 %) ECMO patients survived,
while only 38 of 92 (41 %) CMT patients survived. Moreover, the difference in sur-
vival rates remained regardless of the primary diagnosis, disease severity, and referral
center. The results of this study indicated that ECMO produces a survival benet for
neonates suffering from a variety of potentially reversible respiratory failure [105].
A majority of newborn respiratory failure conditions are potentially reversible
[103]. Moreover, neonatal lungs might be more capable of recovery than older lungs
because they are typically placed on ECMO at a much earlier time point in their
disease progression. All of these factors play a role in the increased neonatal sur-
vival rates on ECMO when compared to pediatric and adult cases.
ECMO was rst successfully used in adult and neonatal ARDS patients during
the 1970s [86, 106]. In the 1980s and 1990s, as adult ECMO fell out of favor;
pediatric application rapidly evolved for severe acute cardiopulmonary failure.
250 J.A. Morris et al.
ECMO became standard for complex congenital cardiac surgery whenever the
patent failed to wean from CPB or experienced postoperative cardiac or respira-
tory failure. Across the United States, all of the major childrens hospitals have
incorporated ECMO capability into their pediatric heart, pediatric ICU, and neo-
natal programs [107].
Among the pediatric population, ECMO is used in the management of respira-
tory failure associated with ARDS, pneumonia, sepsis, burns, drowning, inhalation
injury, and other potentially reversible pulmonary diseases, trauma, or injury
[69, 108]. By January 2013, 16.3 % of all reported ECMO cases (8674 out of 53,190
total cases) were in the pediatric population (ELSO Registry, July 2012). Despite
the widespread application, no randomized controlled trial has been completed that
specically evaluates ECMO protocols in pediatric patients. Furthermore, pediatric
indications for ECMO tend to be much more varied and judgmental than in the
neonatal population. Retrospective studies and single-center reports indicate that
pediatric ECMO benets high-risk respiratory failure patients and decreases mor-
tality [103105, 109, 110]. In one study at the University of Michigan, data from 25
non-neonatal pediatric patients treated with ECLS for severe respiratory failure
(from 1991 to 1993) were analyzed [110]. Of the 25 selected patients, 22 survived
to discharge, which represents a survival rate of 88 %. In a larger, multicenter, ret-
rospective study, 331 pediatric patients admitted during 1991 from 32 hospitals
were examined [109]. This study revealed that ECMO was associated with improved
survival in acute respiratory failure patients. Moreover, once patients were matched
according to diagnosis and risk into pairs, the ECMO-treated patients had a signi-
cantly lower mortality rate than non-ECMO patients that received conventional
therapies (26.4 % vs. 47.2 %) [109]. Collectively, these studies of the early 1990s
generated more enthusiasm toward pediatric ECMO in respiratory failure patients.
Every year, numerous patients on the United Network of Organ Sharing (UNOS)
waitlist die before they can receive a lung transplant [111]. Moreover, the longer
patients spend on the waitlist, the higher their chance of death post-transplant due to
their deconditioning prior to transplantation. Since the lung allocation scoring
(LAS) system was implemented, waitlist times have decreased, but only for specic
diseases [80]. Individuals awaiting lung transplantation from diseases such as pul-
monary hypertension secondary to pulmonary brosis still have long waitlist times
[77]. While awaiting transplant, such patients often deteriorate requiring mechani-
cal ventilation [81]. Yet, mechanical ventilation is associated with an increased risk
of death in the rst 6 months following lung transplantation [112].
In the past 5 years, the outcomes of using VV ECMO as a bridge to lung trans-
plantation have vastly improved, especially in patients with a high LAS score
(Hoopes, 2013, unpublished interview). Patients bridged to a lung transplant with
ECMO had a survival advantage compared to patients with a LAS score greater than
14 The Story of ECLS: History and Future 251
Fig. 14.5 Example of ECMO ambulation: (a) extubated, bed-ridden patient on ECMO; (b) patient
on ECMO and able to sit up in bed and ambulate. Reproduced from: Tulman DB, Stawicki SPA,
Whitson BA, Gupta SC, Tripathi RS, Firstenberg MS, et al. Veno-venous ECMO: a synopsis of
nine key potential challenges, considerations, and controversies. BMC Anesthesiol 2014; 14, 65
(Open Access article)
252 J.A. Morris et al.
In a 2011 retrospective study by Garcia et al. [76], three patients were ambulatory
while on VV ECMO, attempting bedside exercises, while one patient was able to
walk on a treadmill. In 2012, ambulatory single cannula-venous VV ECMO was
used successfully in four cystic brosis patients awaiting bilateral lung transplanta-
tion [77]. Renements to the other ECMO circuit components including pumps,
circuit coatings, and decreased bulk, allow for long-term patient support with poten-
tial ambulation, until an organ becomes available for transplantation.
Clinical application of ambulatory ECMO as a bridge to lung transplant has rap-
idly gained acceptance. Several authors have reported series with 1020 patients
with approximately an 80 % survival after transplant. Hoopes et al. [79] reporting
on the USC/UK experience reported an 83 % overall survival with 25 of the last 27
alive to discharge following lung transplant. Most of the later patients had a percu-
taneous double-lumen catheter that allowed the patient out-of-bed, off the ventila-
tor, with ability to exercise on a treadmill. Ambulatory ECMO as a bridge is
supported by the recent medical literature regarding the avoidance of frailty in the
respiratory failure population.
Fig. 14.6 ELSO charter meeting in Ann Arbor, Michigan. Photo courtesy of ELSO, used with
permission
Fig. 14.7 Since 1990, the number of active ECLS centers has more than doubled. Beginning in
2003, the number of centers has steadily increased, increasing access to ECMO. Data collected
from the ELSO Registry January 2013
254 J.A. Morris et al.
Table 14.2 Overall patient outcomes from ELSO registry (January 2013)
Total Survived ECLS Survived to Discharge
Neonatal
Respiratory 26,205 22,145 85 % 19,559 75 %
Cardiac 4987 3058 61 % 2010 40 %
ECPR 851 540 63 % 331 39 %
Pediatric
Respiratory 5656 3692 65 % 3183 56 %
Cardiac 6225 4034 65 % 3054 49 %
ECPR 1745 948 54 % 708 41 %
Adult
Respiratory 3761 2400 64 % 2084 55 %
Cardiac 2884 1581 55 % 1132 39 %
ECPR 876 325 37 % 241 28 %
Overall total 53,190 38,723 73 % 32,302 61 %
potent tool for retro-analysis. As of January 2013, the ELSO Registry contained
53,190 ECMO cases. Of these cases, 32,043 were neonates, 13,626 were children,
and 7521 were adults (Table 14.2) (ELSO Registry, January 2013). The ECLS pro-
cedure was survived by 38,723 patients (73 %) and 32,302 patients (61 %) survived
to decannulation or transfer. These statistics are remarkable when compared to the
results of the earliest attempts to establish ECMO as a practical and safe technique.
Conclusions
Since the 1970s, the ECMO eld has evolved and expanded exponentially. Through
technology improvements and the knowledge gained from the collective member-
ship of ELSO, published reports, and shared experiences including the ELSO
Registry, ECMO has ourished and saved thousands of patients that were otherwise
moribund. Use of the ELSO Registry and careful analysis of ECMO studies will
remain central to maintaining positive progress in the eld. The potential survival
benets of early ECMO deployment in the emergency room setting may make
ECMO the standard of care for patients with any form of potentially reversible car-
diogenic shock (Bartlett, unpublished interview, 2013). Moreover, the development
of circuits that do not require continuous anticoagulation would not only revolution-
ize modern-day ECMO, but would vastly impact the medical eld in general. As
technology continues to develop and researchers push the boundaries of ECMO,
more lives will be saved, but at a cost of more capital equipment and critical care
expertise.
14 The Story of ECLS: History and Future 255
References
1. Gibbon Jr JH, Miller BJ, Dobell AR, Engell HC, Voigt GB. The closure of interventricular
septal defects in dogs during open cardiotomy with the maintenance of the cardiorespiratory
functions by a pump-oxygenator. J Thorac Surg. 1954;28(3):23540.
2. Gille JP, Bagniewski AM. Ten years of use of extracorporeal membrane oxygenation (ECMO)
in the treatment of acute respiratory insufciency (ARI). Trans Am Soc Artif Intern Organs.
1976;22:1029.
3. Gaffney AM, Wildhirt SM, Grifn MJ, Annich GM, Radomski MW. Extracorporeal life
support. BMJ. 2010;341:c5317.
4. Dalton HJ, Butt WW. Extracorporeal life support: an update of Rogers Textbook of Pediatric
Intensive Care. Pediatr Crit Care Med. 2012;13(4):46171.
5. Froelich C. Pediatric ECMO: old dog or new trick? Pediatric Grand Rounds
6. Gibbon Jr JH. Application of a mechanical heart and lung apparatus to cardiac surgery. Minn
Med. 1954;37(3):17185.
7. Cohen M, Lillehei CW. A quantitative study of the azygos factor during vena caval occlusion
in the dog. Surg Gynecol Obstet. 1954;98(2):22532.
8. Dennis C. A heart-lung machine for open-heart operations. How it came about. ASAIO Trans.
1989;35(4):76777.
9. Lillehei CW. Controlled cross circulation for direct-vision intracardiac surgery; correction of
ventricular septal defects, atrioventricularis communis, and tetralogy of Fallot. Postgrad
Med. 1955;17(5):38896.
10. Kurusz M. May 6, 1953: the untold story. ASAIO J. 2012;58(1):25.
11. Warden HE, Dewall RA, Read RC, Aust JB, Cohen M, Varco RL, et al. Total cardiac by-pass
utilizing continuous perfusion from reservoir of oxygenated blood. Proc Soc Exp Biol Med.
1955;90(1):24650.
12. Fazel S, Williams WG, Goldman BS. Thinned blood, monkey lungs, and the cold heart.
J Card Surg. 2004;19(3):2758.
13. Gibbon Jr JH. An oxygenator with large surface area to Volume ratio. J Lab Clin Med.
1939;24:11928.
14. Miller BJ, Gibbon JH, Fineburg C. An improved mechanical heart and lung apparatus; its use
during open cardiotomy in experimental animals. Med Clin North Am. 1953;1:160324.
15. Jones RE, Donald DE, Swan JC, Harshbarger HG, Kirklin JW, Wood EH. Apparatus of the
Gibbon type for mechanical bypass of the heart and lungs; preliminary report. Proc Staff
Meet Mayo Clin. 1955;30:10513.
16. Kirklin JW, Dushane JW, Patrick RT, Donald DE, Terry Jr HR, Burchell HB. et al. Intracardiac
surgery with the aid of a mechanical pump-oxygenator system (gibbon type): report of eight
cases. Proc Staff Meet Mayo Clin. 1955;30(10):2016.
17. Kirklin JW, Donald DE, Harshbarger HG, Hetzel PS, Patrick RT, Swan HJ, et al. Studies in
extracorporeal circulation. I. Applicability of Gibbon-type pumpoxygenator to human intra-
cardiac surgery: 40 cases. Ann Surg. 1956;144:28.
18. Dennis C. Perspective in review. One groups struggle with development of a pump-
oxygenator. Trans Am Soc Artif Intern Organs. 1985;31:111.
19. Stoney WS. Evolution of cardiopulmonary bypass. Circulation. 2009;119:284453.
20. Kirklin JW, Donald DE, Harshbarger HG, Hetzel PS, Patrick RT, Swan HJ. Studies in extra-
corporeal circulation. I. Applicability of Gibbon-type pump-oxygenator to human intracar-
diac surgery: 40 cases. Ann Surg. 1956;144(1):28.
21. Kirklin JW. Open-heart surgery at the Mayo Clinic. The 25th anniversary. Mayo Clin Proc.
1980;55(5):33941.
22. DeWall RA, Naja H, Roden T. A hard-shell temperature-controlling disposable blood
oxygenator. Use in clinical open-heart surgery. JAMA. 1966;197(13):10658.
23. Dewall RA, Gott VL, Lillehei CW, Read RC, Varco RL, Warden HE. Total body perfusion for
open cardiotomy utilizing the bubble oxygenator; physiologic responses in man. J Thorac
Surg. 1956;32(5):591603.
256 J.A. Morris et al.
24. Clarke Jr LC, Gollan F, Gupta VB. The oxygenation of blood by gas dispersion. Science.
1950;111(2874):857.
25. DeWall RA, Bentley DJ, Hirose M, Battung V, Naja H, Roden T. A temperature controlling
(omnithermic) disposable bubble oxygenator for total body perfusion. Dis Chest. 1966;49(2):
20711.
26. Lillehei CW. A personalized history of extracorporeal circulation. Trans Am Soc Artif Intern
Organs. 1982;28:516.
27. Gott VL, Dewall RA, Paneth M, Zuhdi MN, Weirich W, Varco RI, Lillehei CW. A selfcon-
tained disposable oxygenator of plastic sheet for intracardiac surgery. Thorax. 1957;12:19.
28. Gott VL, Sellers RD, Dewall RA, Varco RL, Lillehei CW. A disposable unitized plastic sheet
oxygenator for open heart surgery. Dis Chest. 1957;32(6):61525.
29. Naef AP. The story of thoracic surgery. Toronto: Hogrefe and Huber; 1990.
30. Cross ES, Berne RM, Hirose Y. Evaluation of a rotating disc type oxygenator. Proc Soc Exp
Biol Med. 1956;93:2104.
31. Osborn JJ, Bramson ML, Gerbode F. A rotating disc blood oxygenator and integral heat
exchanger of improved inherent efciency. J Thorac Cardiovasc Surg. 1960;39:42737.
32. Melrose DG, Bramson ML, Osborn JJ, Gerbode F. The membrane oxygenator; some aspects
of oxygen and carbon dioxide transport across polyethylene lm. Lancet. 1958;1:10501.
33. Cross FS, Berne RM, Hirose Y, Jones RD, Kay EB. Description and evaluation of a rotating
disc type reservoir-oxygenator. Surg Forum. 1957;7:2748.
34. Kolff WJ, Efer DB, Groves LK, Peereboom G, Moraca PP. Disposable membrane oxygen-
ator (heart-lung machine) and its use in experimental surgery. Cleve Clin Q. 1956;23:
6997.
35. Zuhdi N, McCollough B, Carey J, Krieger C, Greer A. Hypothermic perfusion for open-heart
surgical procedures. Report on the use of a heart-lung machine primed with ve per cent
dextrose in water inducing hemodilution. J Int Coll Surg. 1961;35:31926.
36. Zuhdi N, McCollough B, Carey J, Greer A. Double-belical reservoir heart-lung machine
designed for hypothermic perfusion; primed with 5 per cent glucose in water; inducing hemo-
dilution. Arch Surg. 1961;82:3205.
37. Dewall RA, Lillehei CW. Simplied total body perfusion. Reduced ows, moderate
hypothermia, and hemodilution. JAMA. 1962;179:4304.
38. Dewall RA, Lillehei RC, Sellers RD. Hemodilution perfusions for open-heart surgery. Use of
ve per cent dextrose in water for the priming volume. N Engl J Med. 1962;266:107884.
39. Kolff WJ, Baltzer R. The articial coil lung. Trans Am Soc Artif Intern Organs. 1955;
1:3942.
40. Clowes Jr GHA, Hopkins AL, Kolobow T. Oxygen diffusion through plastic lms. Trans Am
Soc Artif Intern Organs. 1955;1:234.
41. Clowes Jr GH, Hopkins AL, Neville WE. An articial lung dependent upon diffusion of
oxygen and carbon dioxide through plastic membranes. J Thorac Surg. 1956;32:6307.
42. Clowes Jr GH, Neville WE. Further development of a blood oxygenator dependent upon the
diffusion of gases through plastic membranes. Trans Am Soc Artif Intern Organs.
1957;3:528.
43. Kammermeyer K. Silicone rubber as a selective barrier. Ind Eng Chem. 1957;49:16856.
44. Burns N. Production of a silicone rubber lm for the membrane lung. Biomed Eng.
1969;4:3569.
45. Iwahashi H, Yuri K, Nose Y. Development of the oxygenator: past, present, and future. J Artif
Organs. 2004;7:11120.
46. Peirce 2nd EC. A modication of the Clowes membrane lung. J Thorac Cardiovasc Surg.
1960;39:43848.
47. Kolobow T, Bowman RL. Construction and evaluation of an alveolar membrane articial
heart-lung. Trans Am Soc Artif Intern Organs. 1963;9:23843.
48. Bramson ML, Osborn JJ, Main FB, et al. A new disposable membrane oxygenator with inte-
gral heat exchange. J Thorac Cardiovasc Surg. 1965;50:391400.
14 The Story of ECLS: History and Future 257
49. Land AJ, Dos SJ, Carlson RG, et al. A new membrane oxygenator-dialyzer. Surg Clin North
Am. 1967;47(6):146170.
50. Day SW, Crystal DK, Wagner CL, Kranz JM. Properties of synthetic membranes in extracor-
poreal circuits. Am J Surg. 1967;114(2):3149.
51. Marx TI, Snyder WE, St John AD, Moeller CE. Diffusion of oxygen into a lm of whole
blood. J Appl Physiol. 1960;15:11239.
52. Bodell BR, Head JM, Head LR, Formolo AJ, Head JR. A capillary membrane oxygenator.
J Thoracic Cardiovasc Surg. 1963;46:63950.
53. Wilson R, Shepley DJ, Llewellyn-Thomas E. A membrane oxygenator with a low priming
volume for extra-corporeal circuit. Can J Surg. 1965;8:30911.
54. Dorson Jr W, Baker E, Hull H. A shell and tube oxygenator. Trans Am Soc Artif Intern
Organs. 1968;14:2429.
55. Peirce 2nd EC, Dibelius NR. The membrane lung: studies with a new high permeability co-
polymer membrane. Trans Am Soc Artif Intern Organs. 1968;14:2206.
56. Kolobow T, Zapol W, Pierce JE, Keeley AF, Replogle RL, Haller A. Partial extracorporeal
gas exchange in alert newborn lambs with a membrane articial lung perfused via an A-V
shunt for periods up to 96 hours. Trans Am Soc Artif Intern Organs. 1968;14:32834.
57. Bartlett RH, Isherwood J, Moss RA, Olszewski WL, Polet H, Drinker PA. A toroidal ow
membrane oxygenator: four day partial bypass in dogs. Surg Forum. 1969;20:1523.
58. Bartlett RH, Drinker PA, Burns NE, et al. The toroidal membrane oxygenator: design, perfor-
mance, and prolonged bypass testing of a clinical model. Trans Am Soc Artif Intern Organs.
1972;18:36974.
59. McCaughan JS, Weeder R, Schuder JC, Blakemore WS. Evaluation of new nonwettable mac-
roporous membranes with high permeability coefcients for possible use in a membrane
oxygenator. J Thorac Cardiovasc Surg. 1960;40:57481.
60. Lim MW. The history of extracorporeal oxygenators. Anaesthesia. 2006;61(10):98495.
61. Mortensen JD. An intravenacaval blood gas exchange (IVCBGE) device. A preliminary
report. ASAIO Trans. 1987;33(3):5703.
62. Cox Jr CS, Zwischenberger JB, Traber DL, Traber LD, Haque AK, Herndon DN. Use of an
intravascular oxygenator/carbon dioxide removal device in an ovine smoke inhalation injury
model. ASAIO Trans. 1991;37(3):M4113.
63. Kaushik M, Wojewodzka-Zelezniakowicz M, Cruz DN, Ferrer-Nadal A, Teixeira C, Iglesias
E, et al. Extracorporeal carbon dioxide removal: the future of lung support lies in the history.
Blood Purif. 2012;34(2):94106. doi:10.1159/000341904. Epub 2012 Oct 24.
64. Zwischenberger JB, Tao W, Bidani A. Intravascular membrane oxygenator and carbon
dioxide removal devices: a review of performance and improvements. ASAIO J. 1999;
45(1):416.
65. Peek GJ, Killer HM, Reeves R, Sosnowski AW, Firmin RK. Early experience with a
polymethyl pentene oxygenator for adult extracorporeal life support. ASAIO J. 2002;48(5):
4802.
66. Toomasian JM, Schreiner RJ, Meyer DE, Schmidt ME, Hagan SE, Grifth GW, et al. A poly-
methylpentene ber gas exchanger for long-term extracorporeal life support. ASAIO J.
2005;51(4):3907.
67. Motomura T, Maeda T, Kawahito S, Takano T, Nonaka K, Linneweber J, et al. Development
of silicone rubber hollow ber membrane oxygenator for ECMO. Artif Organs. 2003;27(11):
10503.
68. Bartlett RH. The development of prolonged extracorporeal circulation. In: Zwischenberger
JB, Steinhorn RH, Bartlett RH, editors. ECMO: extracorporeal cardiopulmonary support in
critical care. Michigan, Ann Arbor: Extracorporeal Life Support Organization; Ann Arbor,
Michigan, 2000.
69. Crawford D, Harvey B. A review of extracorporeal membrane oxygenation in the UK. Nurs
Child Young People. 2012;24(8):1822.
258 J.A. Morris et al.
70. Anderson 3rd HL, Otsu T, Chapman RA, Barlett RH. Venovenous extracorporeal life support
in neonates using a double lumen catheter. ASAIO Trans. 1989;35(3):6503.
71. Otsu T, Merz SI, Hultquist KA, Attorri RJ, Anderson 3rd HL, Schefer DE, et al. Laboratory
evaluation of a double lumen catheter for venovenous neonatal ECMO. ASAIO Trans.
1989;35(3):64750.
72. Wang D, Zhou X, Liu X, Sidor B, Lynch J, Zwischenberger JB. Wang-Zwische double lumen
cannula-toward a percutaneous and ambulatory paracorporeal articial lung. ASAIO J.
2008;54(6):60611.
73. Rais-Bahrami K, Walton DM, Sell JE, Rivera O, Mikesell GT, Short BL. Improved
oxygenation with reduced recirculation during venovenous ECMO: comparison of two
catheters. Perfusion. 2002;17(6):4159.
74. Garcia JP, Iacono A, Kon ZN, Grifth BP. Ambulatory extracorporeal membrane oxygen-
ation: a new approach for bridge-to-lung transplantation. J Thorac Cardiovasc Surg. 2010;
139:e1379.
75. Reeb J, Falcoz PE, Santelmo N, Massard G. Double lumen bi-cava cannula for veno-venous
extracorporeal membrane oxygenation as bridge to lung transplantation in non-intubated
patient. Interact Cardiovasc Thorac Surg. 2012;14:1257.
76. Garcia JP, Kon ZN, Wu Z, Iacnon AT, McCormick B, Grifth BP. Ambulatory veno-venous
extracorporeal membrane oxygenation: innovation and pitfalls. J Thorac Cardiovasc Surg.
2011;142(4):75561.
77. Hayes Jr D, Kukreja J, Tobias JD, Ballard HO, Hoopes CW. Ambulatory venovenous
extracorporeal respiratory support as a bridge for cystic brosis patients to emergent kung
transplantation. J Cyst Fibros. 2012;11(1):405.
78. Bellezzo JM, Shinar Z, Davis DP, Jaski BE, Chillcott S, Stahovich M, et al. Emergency
physician-initiated extracorporeal cardiopulmonary resuscitation. Resuscitation. 2012;83(8):
96670.
79. Hoopes CW, Kukreja J, Golden J, Davenport DL, Diaz-Guzman E, Zwischenberger JB.
Extracorporeal membrane oxygenation as a bridge to pulmonary transplantation. J Thorac
Cardiovasc Surg. 2013;145(3):8627. discussion 8678.
80. Diaz-Guzman E, Hoopes CW, Zwischenberger JB. The evolution of extracorporeal life
support as a bridge to lung transplantation. ASAIO J. 2013;59(1):310.
81. Iglesias M, Jungebluth P, Petit C, Matute MP, Rovira I, Martnez E, et al. Extracorporeal lung
membrane provides better lung protection than conventional treatment for severe postpneu-
monectomy noncardiogenic acute respiratory distress syndrome. J Thorac Cardiovasc Surg.
2008;135:136271.
82. Meyer A, Struber M, Fischer S. Advances in extracorporeal ventilation. Anesthesiol Clin.
2008;26:38191.
83. Hill JD, OBrien 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;286:62934.
84. Zapol WM, Snider MT, Hill JD, Fallat RJ, Edmunds LH, Morris AH, et al. Extracorporeal
membrane oxygenation in severe acute respiratory failure. A randomized prospective study.
JAMA. 1979;242(20):21936.
85. Zwischenberger JB, Bartlett RH. Extracorporeal life support: an overview. In: Van Meurs K,
Lally KP, Peek G, Zwischenberger JB, editors. ECMO extracorporeal cardiopulmonary support
in critical care. 3rd ed. Ann Arbor: Extracorporeal Life Support Organization; 2005. p. 127.
86. Kanto WP, Shapiro MB. The development of prolonged extracorporeal circulation. In:
Zwishcenberger JB, Steinhron RH, Bartlett RH, editors. ECMO: Extracorporeal cardiopul-
monary support in critical care. 2nd ed. Extracorporeal Life Support Organization; 2000.
87. Morris AH, Wallace CJ, Menlove RL, Clemmer TP, Orme Jr JF, 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):
295305.
14 The Story of ECLS: History and Future 259
88. Zwischenberger JB, Lynch JE. Will CESAR answer the adult ECMO debate? Lancet.
2009;374:13078.
89. Sadahiro T, Oda S, Nakamura M, Watanabe E, Abe R, Nakada TA, et al. Trends in and per-
spectives on extracorporeal membrane oxygenation for severe adult respiratory failure. Gen
Thorac Cardiovasc Surg. 2012;60:192201.
90. Flrchinger B, Philipp A, Klose A, Hilker M, Kobuch R, Rupprecht L, et al. Pumpless extra-
corporeal lung assist: a 10-year institutional experience. Ann Thorac Surg. 2008;86(2):410
7. discussion 417.
91. Kolla S, Awad SS, Rich PB, Schreiner RJ, Hirsch RB, Bartlett RH. Extracorporeal life
support for 100 adult patients with severe respiratory failure. Ann Surg. 1997;226:54464.
discussion 6566.
92. Peek GJ, Mugford M, Tiruvoipati R, Wilson A, Allen E, Thalanany MM, et al. Efcacy and
economic assessment of conventional ventilator support versus extracorporeal membrane
oxygenation for severe adult respiratory failure (CESAR): a multicenter randomized
controlled trial. Lancet. 2009;374:135163.
93. Ketoconazole for early treatment of acute lung injury and acute respiratory distress
syndrome: a randomized controlled trial. The ARDS Network. JAMA. 2000;283(15):
19952002.
94. Eisner MD, Thompson T, Hudson LD, Luce JM, Hayden D, Schoenfeld D, et al. Efcacy of
low tidal volume ventilation in patients with different clinical risk factors for acute lung
injury and the acute respiratory distress syndrome. Am J Respir Crit Care Med. 2001;164(2):
2316.
95. Davies A, Jones D, Balley M, Beca J, Bellomo R, Blackwell N, et al. Extracorporeal mem-
brane oxygenation for 2009 Inuenza A (H1N1) acute respiratory distress syndrome. JAMA.
2009;302:188895.
96. Freed DH, Henzler D, White CW, Fowler R, Zarychanski R, Hutchison J, et al. Extracorporeal
lung support for patients who had severe respiratory failure secondary to inuenza A (H1N1)
2009 infection in Canada. Can J Anaesth. 2010;57:2407.
97. Roch A, Lepaul-Ercole R, Grisoli D, Bessereau J, Brissy O, Castanier M, et al. Extracorporeal
membrane oxygenation for severe inuenza A (H1N1) acute respiratory distress syndrome: a
prospective observational comparative study. Intensive Care Med. 2010;36:1899905.
98. Pappalardo F, Pieri M, Greco T, Patroniti N, Pesenti A, Arcadipane A, et al. Predicting mor-
tality risk in patients undergoing venovenous ECMO for ARDS due to inuenza A (H1N1)
pneumonia: the ECMO net score. Intensive Care Med. 2013;39:27581.
99. Bartlett RH. Esperanza: presidential address. Trans Am Soc Artif Intern Organs. 1985;31(1):
7236.
100. Mortensen JD. Extracorporeal membrane oxygenation for pulmonary assist in patients with
ART. Chest. 1975;67(1):2930.
101. Bartlett RH, Andrews AF, Toomasian JM, Haiduc NJ, Gazzaniga AB. Extracorporeal
membrane oxygenation for newborn respiratory failure: forty-ve cases. Surgery. 1982;92(2):
42533.
102. Ware JH. Investigating therapies of potentially great benet: ECMO. Statist Sci.
1989;4(4):298306.
103. Bartlett RH, Roloff DW, Cornell RG, Andrews AF, Dillon PW, Zwischenberger
JB. Extracorporeal circulation in neonatal respiratory failure: a prospective randomized
study. Pediatrics. 1985;76:47987.
104. ORourke PP, Crone RK, Vacanti JP, Ware JH, Lillehei CW, Parad RB, et al. Extracorporeal
membrane oxygenation and conventional medical therapy in neonates with persistent pulmo-
nary hypertension of the newborn: a prospective randomized study. Pediatrics. 1989;84:
95763.
105. UK Collaborative ECMO Trial Group. UK collaborative randomized trial of neonatal extra-
corporeal membrane oxygenation. Lancet. 1996;348(9020):7582.
106. Bartlett RH, Harken DE. Instrumentation for cardiopulmonary bypasspast, present, and
future. Med Instrum. 1976;10(2):11924.
260 J.A. Morris et al.
107. Bartlett RH. Extracorporeal life support: history and new directions. ASAIO J. 2005;
51(5):879.
108. Friedman DF, Montenegro LM. Extracorporeal membrane oxygenation and cardiopulmo-
nary bypass. In: Handbook of pediatric transfusion medicine; 2004. p. 1819.
109. Green TP, Timmons OD, Fackler JC, Moler FW, Thompson AE, Sweeney MF. The impact of
extracorporeal membrane oxygenation on survival in pediatric patients with acute respiratory
failure. Pediatric Critical Care Study Group. Crit Care Med. 1996;24(2):3239.
110. Moler FW, Custer JR, Bartlett RH, Palmisano JM, Akingbola O, Taylor RP, et al.
Extracorporeal life support for severe pediatric respiratory failure: an updated experience
19911993. J Pediatr. 1994;124(6):87580.
111. Kreider M, Hadjiliadis D, Kotloff RM. Candidate selection, timing of listing, and choice of
procedure for lung transplantation. Clin Chest Med. 2011;32(2):199211.
112. Singer JP, Blanc PD, Hoopes C, Golden JA, Koff JL, Leard LE, et al. The impact of pretrans-
plant mechanical ventilation on short- and long-term survival after lung transplantation. Am
J Transplant. 2011;11(10):2197204.
113. Diaz-Guzman E, Davenport DL, Zwischenberger JB, Hoopes CW. Lung function and ECMO
after lung transplantation. Ann Thorac Surg. 2012;94:68694.
114. Cypel M, Keshavjee S. Extracorporeal life support as a bridge to lung transplantation. Clin
Chest Med. 2011;32:24551.
Index
decannulation, 229230 H
disconnection, 173 Haemoglobin
and drug intoxication, 77 arterial oxygen saturations, 27
and ECPR, 7576 oxygen affinity, 19
flow management, 227 venous oxygen saturation, 28
flow regurgitation, 52 Haldane effect, 19
and heart transplantation, 7778 Heart transplantation, 7778
high airway pressures and high tidal Heat exchanger, 200
volumes, 164 Hemolysis, 202, 205
LVAD implantation, 78 Hemorrhage, 201, 202
and massive pulmonary embolism, 79 Henrys Law, 11
microporous devices, 242 Heparin-induced thrombocytopenia (HIT),
nonmicroporous devices, 242243 158, 189, 196198
occlusion of flow path, 5152 High-frequency oscillatory ventilation
outcomes after VA-ECMO, 8081 (HFOV), 170171
oxygenation, 164165 HIT. See Heparin-induced thrombocytopenia
patient and disease-specific issues (HIT)
configurations and cannulation Hydrogen ion concentration, 3234
strategies, 74 Hypercapnic acidosis, 39
retrieval units, 7475 Hypovolemia, 155
VA-ECMO indication, 74 Hypoxemia, 208209
PECOR, 174176
PGD, 252
postcardiotomy cardiogenic shock, 77 I
and profound hypothermia, 79 IABP. See Intra-aortic balloon pump (IABP)
refractory septic shock, 79 iLA. See Interventional lung assist (iLA)
termination of ECLS, 230231 Inadvertent decannulation, 200201
VA ECMO, 243244 Interventional lung assist (iLA), 107
VV ECMO, 244 Intra-aortic balloon pump (IABP), 125, 128
Extracorporeal support, cannulation
blood flow, 135137
dual-lumen L
bicaval design (Avalon Elite, Maquet), LAS system. See Lung allocation scoring
134, 135 (LAS) system
cavo-atrial design (OriGen, OriGen Left ventricular assist devices (LVAD), 2, 5354
Biomedical), 134 Left ventricular end diastolic pressure (LVEDP),
hemodialysis catheter, 135 5657
percutaneous cannula, 133134 Liberation, 218. See also Weaning and
single lumen design, 134 liberation, ECMO
wire-reinforced cannulas, 134 Limb ischemia, 144
Ex-vivo lung perfusion (EVLP), 113 Lung allocation scoring (LAS) system, 250
Lung transplantation
bleeding, hemolysis and infection, 105
F bridge to decision, 111
Flashing the cannulae, 228 BTT (see Bridge to transplant (BTT))
Frank-Starling mechanism, 56 centrifugal pumps, 106
complications, 110
diseases, 250
G DLC, 251
Glycolysis iLA, 107
ATP molecules, 5 intraoperative VA-ECLS, 112
glycogen storage, 34 intubation, 110
intracellular metabolism, 4 LAS system, 250251
TCA cycle, 5 mechanical ventilation, 106, 111112
266 Index
R
P Rapoport-Luebering shunt, 8
Partial extracorporeal CO2 removal (PECOR) Registered nurses (RNs), 245
barotrauma, 174, 175 Registered respiratory therapists
bilateral pleurectomy, 174 (RRTs), 245
clinical application, 174 Rehabilitation, 214, 216
ventilatory control, 174176 Renal replacement therapy (RRT), 151
Passive oxygenation, 39 Respiratory quotient (RQ), 56
Patent ductus arteriosus (PDA), 237 Respiratory system compliance (Crs), 172
Patient crises Reynolds number, 4748
bleeding and transfusions Right ventricular assist devices (RVAD), 2
adverse events, 202 RNs. See Registered nurses (RNs)
anticoagulation, 202, 203 RRT. See Renal replacement therapy
bronchoscopy, 203 (RRT)
diagnosis and management, 204205 RRTs. See Registered respiratory therapists
hemoglobin, 201203 (RRTs)
hemorrhagic complications, 201 RVAD. See Right ventricular assist devices
patient management, 202 (RVAD)
thrombocytopenia, 203
thrombosis, 203
hemolysis, 203, 205 S
refractory hypoxemia, 206207 Septic shock, 79
shock, 207 Shock, 207
Patient-ventilator interaction, 172 Silicone rubber membrane lungs (SRML)
PECOR. See Partial extracorporeal CO2 alveolar membrane, 240
removal (PECOR) extracapillary orientation, 241
Pediatric acute respiratory distress syndrome, gas exchange, 240241
249250 helical tube arrangement, 241
PEEP. See Positive end-expiratory pressure hemodialysis machine design, 240
(PEEP) heparin, 242
Persistent pulmonary hypertension of the hypobaric oxygenation, 241
newborn (PPHN), 248 incorporate hydrophobic materials, 240
Physical therapy, mobilization intracapillary orientation, 241
ambulation, 214, 215 limitations, 239
bridge-to-transplant, 215 O2 and CO2, 240
intubation, 213 SRML. See Silicone rubber membrane lungs
resources, ICU, 217 (SRML)
Positive end-expiratory pressure (PEEP) Systematic review, 99, 150
airway pressures, 166 Systemic inflammatory response syndrome
ECCO2R, 174 (SIRS), 3
268 Index