. 2
( 8)


2A2.3: The cardiotomy reservoir if required
(a) The reservoir can be used for any surgery where intracardiac clot is suspected, where
it is anticipated that large quantities of blood will be used or where the use of auto
transfusion is anticipated
(b) The reservoir and its packaging is checked as above and inserted into the appropriate
(c) Remove any venting cap and using the 3/8² cardiotomy return, connect the
cardiotomy to the oxygenator, ensuring that this return line cannot be kinked or
(d) Connect the sucker lines and recirculation lines to the cardiotomy reservoir

2A2.4: The cardioplegia system if required
(a) Remove packaging and check its integrity and sterility
(b) The circuitry is checked for faults (cracked connections, kinked tubing, etc.)
(c) Assemble circuit according to manufacturer™s instructions
(d) Ensure all connections (oxygenator, recirculation lines, etc.) are secure and correct
(e) Water lines are connected to the cardioplegia administration set heat exchanger.
Water is circulated to ensure that it is free from leaks

2A2.5: The centrifugal pump if required
(a) Remove packaging and check its integrity and sterility
(b) The relevant flow and drive connectors should be connected to the console
(c) The battery charger should be examined to determine whether or not there is
sufficient battery backup
(d) The perfusionist should check that the relevant hand-crank mechanism is available in
case of power failure
(e) The drive motor heads must be examined for dirt, as this may impair the function of
the device, including the possibility of disengagement

2A2.6: Arterial line filters if required
(a) Check the filter for sterility, any damage or debris
(b) If the filter is to be cut into the arterial line this should be carried out using the
appropriate sterile technique
(c) Ensure the filter holder is positioned to prevent the stretching or kinking of lines
(d) Position the filter securely in the holder
An air bubble trap would be primed in a similar fashion.

2A2.7: Cell saver if required
(a) Remove outer packaging and check its integrity and sterility
(b) Open the collection reservoir portion of the set and secure firmly in holder
(c) Connect the vacuum source to the reservoir
(d) The washing portion of the set should only be opened when either enough blood has
been collected to salvage or the perfusionist is confident that enough blood will be
collected to salvage
Chapter 2: Circuit setup and safety checks

(e) The washing portion of the set should be assembled neatly
(f) All ports and connections should be checked, closed and tightened where

2A2.8: Prebypass filters if used
If the circuit contains a prebypass filter there are a number of points the perfusionist must
• The prebypass filter should be removed before priming the circuit with blood
• The prebypass filter should be removed if the low pressure suction is required before
the lines have been divided
A ½² — 3/8² connector should be readily available to replace the prebypass filter if

2A.3: In-line blood chemistry/gas analyzer (e.g., CDI 500) setup
and calibration
2A3.1: Setup of CDI 500 arterial sensor shunt
(a) Turn off monitor and after the monitor has self-tested select the required configura-
tion of the sensor shunt
(b) Select calibration
(c) Verify the K* calibration value on the sensor packaging
(d) Check that the calibrator™s cable is connected to the monitor
(e) Remove blue cap from the base of the sensor shunt and attach to one of the calibra-
tor™s ports
(f) Loosen the blue cap on the top of the sensor shunt
(g) Initiate calibration by pressing √ twice on the monitor
(h) Calibration lasts 10 minutes
(i) After calibration tighten large luer cap and remove gas filter

2A3.2: Setup of CDI 500 Venous Line Sensor
(a) Remove venous sensor from packaging and cut into venous line
(b) After the monitor has been switched on and has self-tested the venous probe can be
connected to the venous sensor

2A.4: Priming the system
The perfusionist should ensure, if possible, that the following patient details are available
from the anesthetic and surgical staff, to provide a basis on which to decide the priming
• Height and weight
• Renal status
• Hb/HCT
• Heart size
• Fluid status
Chapter 2: Circuit setup and safety checks

2A4.1: Standard prime
(a) 1 l Hartmann™s solution is checked
(b) Preservative-free heparin should be injected into the liter bag of Hartmann™s solution
(dose per liter of prime as per institutional protocol) and labeled
(c) The Hartmann™s solution is run into the system via a giving set or rapid prime line.
It is important that this heparinized prime runs through the length of the circuit
(i.e., all filters are exposed to this heparinized prime). The prime is delivered via a
cardiotomy port (if a cardiotomy is in use)
(d) The reservoir should be inspected for obvious bubbles and tapped to remove them
(e) Sufficient prime should be added to the system to maintain a dynamic priming
(f) It is most important at this stage that the oxygenator manufacturer™s instructions are
carefully adhered to
(g) Turn off the CO2 flush
(h) A gravity feed prime is undertaken, with de-bubbling taking place in a logical
fashion, beginning with the oxygenator reservoir and progressing to the arterial line
and so on
(i) The “sash” should be clamped off, the arterial pump switched on and the prime
(j) The pressure line may now be connected, via an air-free isolator to the line pressure
gauge and pressure transducer
(k) The re-circulation lines are securely clamped, and the “sash” primed
(l) It is important to remember that air is easily dragged across the membrane of
hollow fiber oxygenators, so the following precautions should be taken to
avoid this:
• the venous line should be partially occluded so as to offer a resistance, and
therefore maintain a positive pressure as the prime is re-circulating
• the pump should be switched off slowly to avoid the momentum effect
(see below)
(m) When the circuit appears to be clear of bubbles, the re-circulation rate should now
be increased to around 5 l/minute, to remove any bubbles from within the oxygenator
membrane with the venous line partially clamped maintaining a post-membrane
pressure of around 80 mmHg. Before the “sash” is divided, a final check must be
made by both perfusionist and surgeon for the presence of bubbles. Before stopping
the re-circulation, the pump should be turned down slowly, reducing the chances of
the inertia effect of a sudden reduction in flow that would cause air to be dragged
across the membrane

2A4.2: Priming cardioplegia if required
(a) The type, temperature and concentration of blood cardioplegia should be
determined from the surgeon in advance. This information should be held
in the hospital™s database (e.g., proportion 4:1, 2:1, etc., the need for any
“hot shots,” etc.)
(b) Bags of Ringer™s solution should be carefully prepared. The vials of cardioplegia
should be carefully checked before injection. The bags must be labeled clearly as soon
as this has been done
Chapter 2: Circuit setup and safety checks

(c) The cardioplegia circuit is primed with Hartmann™s solution or Ringer™s solution,
checking that all air has been purged
(d) During priming, care must be taken that the main prime does not become
contaminated with cardioplegia
(e) The cardioplegia pump boots are placed in the raceway and appropriately sized
collets fitted (if applicable), or a check is made to ensure that the ratio is correctly
programed into the pump console
(f) The occlusion of the pump is then set as with the arterial pump (see later)

2A4.3: Priming the arterial line filter if required
(a) Place clamps either side of the arterial filter before the oxygenator is
gravity primed
(b) Once the circuit is primed, stop the pump, slowly release lower clamp and
allow prime to flow retrogradely through the filter via the bypass line,
expelling air through the purge line. The retrograde flow is provided by the
prime in the “sash”
(c) Release the top clamp, start the pump
(d) Invert the filter and de-air as normal
(e) Clamp the arterial filter bypass loop

2A4.4: Priming centrifugal pump if required
Centrifugal pumps differ from roller pumps in several important respects:
• They are non-occlusive devices
• They are constant energy devices
(a) A length of 3/8² PVC tubing is connected to the outlet of the venous reservoir and
clamped. A length of 3/8² PVC tubing is also connected to the oxygenator inlet
(b) The outlet of the membrane compartment is connected to the circuit as with a
roller pump
(c) If a “BioPump” bi-directional flow probe is required it should be inserted into the
arterial line, at least 6² away from the nearest connector
(d) The oxygenator venous reservoir is primed with heparinized Hartmann™s solution,
as described in the routine procedure
(e) The centrifugal pump is cut in as required ensuring sterile technique using a sterile
(f) The clamp on the inlet tube is then slowly released, allowing the prime to slowly fill
the head. The outlet port of the head (which is tangential to the body of the head)
is held uppermost. The head is thus filled with the priming solution, and as much
air as possible is purged
(g) The oxygenator is gravity primed as above
(h) The head should again be examined for bubbles and if found should be manipulated
out of the inlet port back into the venous reservoir
(i) When the outlet of the centrifugal head is clamped any air will collect at the center
of the casing (low mass). If the pump is then switched off the collected air will travel
vertically into the inlet tube. As before, this air can be manipulated back into the
venous reservoir
Chapter 2: Circuit setup and safety checks

2A4.5: Calibrating the flow probes
With the circuit fully primed:
(a) The motor drive is switched off
(b) Clamps are positioned some 6² on either side of the probe
(c) Calibrate the flow probe as directed by manufacturer™s instructions

2A.5: Setting occlusions
2A5.1: Occlusion of the arterial pump if a roller pump is used:
(a) Clamp the arterial line and any re-circulating lines and close the sampling ports
(b) The pump is carefully turned until the pressure on the gauge is around 300 mmHg
and the rate of fall of pressure can be observed
(c) Tighten the occlusion until there is no fall of pressure in this high-pressure range
(this ensures that there are no other leaks in the circuit and that all clamps are
(d) Adjust the occlusion until the fall off of pressure over the lower 260“280 mmHg
range takes approximately 10 seconds
(e) Both rollers must be treated individually. Should the occlusion between rollers be
obviously unequal, the pump should be changed

2A5.2: Occlusion of the suction pumps
(a) A clamp is placed on the negative side of the sucker boot and the pump is turned
until the boot collapses with the vacuum created
(b) The occlusion should now be “backed off ” until the vacuum is cleared
(c) The occlusion setting is then again increased, until the vacuum is just drawn and held
(d) In order to check the direction of rotation of the sucker/vent roller pumps, a small
quantity of heparinized saline or other appropriate fluid should be used by the scrub
nurse to check the suction

Appendix 2B: Electronic safety devices
(Adapted, with permission, from London Perfusion Science Protocols.)

2B.1: Level sensors
• The level sensor should be positioned at around the 400 ml mark on the reservoir
• If the option is available, level sensors should be set to slow the pump down before
stopping it
• Level sensors should not be overridden unless it is absolutely necessary

2B.2: Bubble detectors
• Perfusionists must use a gas bubble detector placed in the circuit: it is usual practice to
have the bubble detector on the arterial outlet of the circuit
Chapter 2: Circuit setup and safety checks

2B.3: Pressure alarms
• Most modern heart-lung machines have integrated electronic alarms for limits of
pressure during a case
• These limits should be checked and correctly set to appropriate parameters before each

2B.4: Temperature alarms
• Most modern heart-lung machines have integrated electronic alarms for limits of
temperature during a case
• Arterial blood, venous blood and cardioplegia temperature alarms should be checked
and correctly set to appropriate parameters before each case
• Where available the water temperature alarm limits should also be checked and set

2B.5: Gas alarms
• Most modern gas blenders have alarms for gas failure
• These alarms can be checked when the gas lines are connected to the hospital gas
• Connecting the lines then disconnecting them individually should trigger the alarm

2B.6: Electrical failure alarm
• Most modern heart-lung machines have an integrated alarm that sounds when the
mains power supply fails and UPS is activated. If this occurs, all unnecessary equipment
should be turned off to conserve the battery.

• Mejak BL, Stammers A, Rauch E, et al. A
Suggested Further Reading retrospective study on perfusion incidents
• American Society of Extra-Corporeal and safety devices. Perfusion 2000; 15:
Technology, Herndon, VA. www.amsect.org 51“61.
• Gravlee GP, Davis RF, Stammers AH, • Recommendations for Standards of
Ungerleider RM. Cardiopulmonary Monitoring during Cardiopulmonary
Bypass Principles and Practice. Bypass. Published by the: Society of Clinical
3rd edition, 2008, Lippincott Perfusion Scientists of Great Britain &
Williams & Wilkins. Ireland, Association of Cardiothoracic
Anaesthetists, Society of Cardiothoracic
• Jenkins OF, Morris R, Simpson JM.
Surgeons in Great Britain & Ireland. July
Australian perfusion incident survey.
Perfusion 1997; 12: 279“288.
• Wheeldon DR. Safety during
• Kay PH, Munsch CM. Techniques in
cardiopulmonary bypass. London: Franklin
Extracorporeal Circulation. 4th edition,
Scientific Projects, 1986; 7: 57“65.
2004. London: Arnold.

Priming solutions for

cardiopulmonary bypass circuits
3 George Hallward and Roger Hall

The cardiopulmonary bypass (CPB) circuit must be primed with a fluid solution, so that
adequate flow rates can be rapidly achieved on initiation of CPB without risk of air embo-
lism. The optimum composition of the CPB priming solution is still a matter for debate.
Currently used “primes” have evolved from historical concepts of using a solution with
similar electrolyte content and osmolarity to the intravascular and interstitial compartments,
providing a fluid that when mixed with blood is capable of maintaining oxygen delivery,
carbon dioxide removal and physiological homeostasis.

Prime volume
The volume of prime required is either based on a standard empirically derived volume
greater than a minimum safe priming volume, or may be guided by the patient™s weight or
body surface area. In practice, the minimum volume required is that which fills both venous
and arterial limbs of the circuit and maintains an adequate reserve volume in the venous
reservoir to ensure that air is not entrained into the arterial side of the circuit during initia-
tion of CPB. This volume is determined by both the caliber and length of tubing connecting
the patient to the CPB machine and by the design, and therefore capacity, of the venous res-
ervoir and oxygenator. Reduction of the prime volume may thus be achieved by modification
of the circuit.
The initial hematocrit (HCT) achieved after initiation of CPB is determined by the volume
of the prime in relation to the patient™s pre-CPB HCT. In adults, priming volumes are com-
monly in the range of 1400“1800 ml, typically representing 30“35% of the patient™s blood
volume. In children, especially infants and neonates, even the minimum priming volume is
often far greater than their blood volume, making the use of non-blood-containing primes

Acceptable hemodilution
Initiation of CPB inevitably leads to hemodilution by the priming fluid. Some degree of
hemodilution is beneficial as blood viscosity is reduced, improving microcirculatory flow.
Most centers aim for an HCT of less than 30% during CPB; however, there is no consen-
sus regarding optimal HCT. HCT is the main determinant of the oxygen-carrying capacity
of blood. Theoretically, minimum acceptable HCT should meet the oxygen delivery (DO2)
required to match systemic O2 consumption (VO2). However, DO2 is influenced by pump flow
rate and systemic temperature and VO2 also alters proportionately with temperature. There is
thus wide variation in practice with regard to the minimum, safe, acceptable HCT. Values as
low as 14% have been advocated by some, whilst others have suggested using venous oxygen
saturation (SvO2) rather than a specific HCT value as transfusion trigger. Experience with
36 Cardiopulmonary Bypass, ed. S. Ghosh, F. Falter and D. J. Cook. Published by Cambridge University Press.
© Cambridge University Press 2009.
Chapter 3: Priming solutions for CPB circuits

Jehovah™s Witness patients who refuse blood transfusions show that cardiac surgery and CPB
with low HCTs is not only possible, but is also relatively safe.
Factors affecting the HCT during CPB include:
• patient size;
• preoperative hemoglobin concentration/HCT;
• pre-CPB blood loss;
• pre-CPB fluid administration;
• CPB prime volume; and
• urine output.
One method of reducing the degree of hemodilution, without using “Bank” blood, is to
use autologous blood to partially prime the CPB circuit. This method replaces part of the
CPB prime volume with the patient™s own blood thus reducing the degree of hemodilution.
Autologous priming can be achieved by either antegrade or retrograde routes. Antegrade
priming utilizes partial filling of the venous reservoir with the patient™s own blood from
the venous limb of the CPB circuit on initiation of CPB, but before institution of CPB flow
through the oxygenator and arterial limb of the circuit. Retrograde priming utilizes retro-
grade filling of the venous reservoir via the arterial limb of the CPB circuit, just prior to the
initiation of CPB, displacing the crystalloid prime volume in the arterial line tubing, filter
and oxygenator and so partially filling the reservoir with the patient™s blood. Both meth-
ods reduce the volume of crystalloid in the prime by replacing it with 400“500 ml of the
patient™s blood. Safe autologous priming relies on good teamwork between perfusionist,
anesthetist and surgeon to select appropriate patients and to ensure hemodynamic stabil-
ity, usually with the help of vasopressors, during the period of partial exsanguination of
the patient.
In general, acceptance of a degree of hemodilution during CPB, the use of autologous
priming, collection and processing of shed mediastinal blood and the return of residual pump
blood at the end of CPB can all lead to a decrease in allogenic blood transfusions with their
consequent risks and uncertain risk/benefit profile.

Priming solutions
There are many different recipes for priming solutions using crystalloid, colloid or blood as
primary constituents. Historically, blood was used to prime the CPB circuit in an attempt to
preserve a high hematocrit; early in the evolution of CPB this was thought to be an important
determinant for successful outcome. It later became clear, however, that use of allogenic blood
in the prime may have worsened, rather than improved, outcomes. In 1962, Cooley and
coworkers showed improved outcome by adding 5% dextrose to the prime instead of just
blood. Five percent dextrose later fell out of favor for two reasons: firstly, the realization that
metabolism of glucose leads to a hypotonic solution; and secondly, fears about hyperglycemia
worsening neurological outcome. In part, accumulation of knowledge about the deleterious
effects of blood primes and acceptance that a lower hematocrit is compatible with good out-
comes has led to acceptance of crystalloids as priming solutions. The introduction of hypo-
thermic bypass in the 1960s, the inability of blood banks to support cardiac surgery with large
amounts of whole blood and the prevalence of blood-borne infections were also important in
the shift to “clear” primes. In general, an ideal priming solution should have the same tonicity,
electrolyte composition and pH as that of plasma. Of these ideal properties the most impor-
tant is that of “tonicity,” in order to avoid red cell lysis and the fluid shifts from the extracellular
Chapter 3: Priming solutions for CPB circuits

to the intracellular compartment that occur with hypotonic solutions. Fluid shifts may occur
in any organ or tissue, but the organs most vulnerable to fluid accumulation are the brain and
lungs. Intracellular fluid gain causes cerebral or pulmonary edema and impairs organ func-
tion. It is important to appreciate that fluids which are nominally isotonic but which have
glucose as a major constituent, e.g., 5% dextrose or dextrose/saline, become very hypotonic
when the glucose is metabolized. For this reason, glucose-containing solutions should not be
a major constituent of a prime and only those fluids with a near physiological sodium concen-
tration should be used.
Suitable solutions used include lactated Ringer™s (Hartmann™s), Ringer™s, normal saline,
Plasma-Lyte and Normosol (see Tables 3.1 and 3.2). All of these solutions have similar sodium
concentrations (130“150 mmol/l) and may contain physiological concentrations of potas-
sium (Hartmann™s, Plasma-Lyte). There are some differences in anion composition, but all
have chloride as a major anionic constituent, the balance in Hartmann™s or Plasma-Lyte being
made up with lactate or acetate, respectively. Both lactate and acetate are ultimately metabo-
lized to bicarbonate in the liver, thus producing a near ideal physiological solution. Hart-
mann™s solution is the most commonly used crystalloid in priming fluids in the UK, although
there is variation in practice amongst different units. Normosol-A and Plasma-Lyte are bal-
anced solutions more commonly used in the USA.
The priming solution has been implicated as one of the potential causes of the disturbance
of pH associated with development of metabolic acidosis on initiation of CPB. This acidosis
is probably caused by hyperchloremia and is more likely to occur with normal saline, which
has a higher chloride load than the more “physiological” solutions. Other possible reasons for
this include an increase in unmeasured anions such as acetate and gluconate. This metabolic

Table 3.1. Composition of commonly used priming fluids

Na+ K+ Cl’ Ca2+ Mg2+ HCO3’ pH Other mosmol/l
Dextrose 5% 0 0 0 0 0 0 4.2 Glucose 279
50 g/l
Saline 0.9% 154 0 154 0 0 0 5.0 “ 308
Hartmann™s 131 5.0 111 2.0 0 29 6.5 “ 280
Plasmalyte A 140 5.0 98 0 3 27 7.4 “ 294
29 (glu-
Normasol R 140 5.0 98 0 3 27 7.4 “ 294
29 (glu-
Bicarbonate 150 0 0 0 0 150 7.0 “ 300
Gelofusine 154 0.4 120 0.4 0 0 7.1“7.7 Gelatine 274
40 g/l
Starch 154 0 154 0 0 0 4.5“5.5 Starch 308
Human 100“160 <2 100“160 0 0 <0.1 7.1 Albumin 300
Albumin 4.5 citrate 40“50 g/l

Chapter 3: Priming solutions for CPB circuits

Table 3.2. Commonly used additives

Heparin 1000“2500 U/l of prime to ensure adequate anticoagulation
Bicarbonate 25 mmol/l of prime as buffer when unbalanced priming solutions are used
Mannitol Osmotic diuretic and free radical scavenger
Calcium Needed if citrated blood is added to the prime to prevent chelation of calcium
Steroids To attenuate systemic inflammatory response to CPB (evidence weak)

acidosis is a benign phenomenon and probably accounts for much of the base deficit observed
while on bypass.
Colloid solutions, including 4.5% albumin, gelatins, e.g., gelofusine, dextrans and
starches, e.g., hydroxyethyl starch, have been advocated for use in the CPB prime on
account of their potential to counteract the decrease in colloid oncotic pressure associ-
ated with hemodilution of albumin and other circulating plasma proteins during CPB.
This reduction in colloid oncotic pressure causes movement of water out of the intravas-
cular space and into the interstitial and intracellular spaces, contributing to postoperative
edema and subsequent organ dysfunction. Thus, using colloids, with their high molecular
weight, to maintain oncotic pressure and therefore reduce fluid shifts seems an attractive
strategy. The drawback to this hypothesis is that whilst, in theory, colloid solutions ought
to remain in the intravascular space, in practice the “tight junctions,” which render the
endothelial lining impermeable to large molecules, become more permeable on activa-
tion of the systemic inflammatory response associated with CPB. This may paradoxically
increase the amount of extravasated fluid, as the high-molecular-weight constituents of
colloid solutions become trapped in the interstitial fluid, potentially adding to edema by
drawing more free fluid into the interstitium. Furthermore, some of the constituents of
colloids have undesirable properties: dextrans interfere with coagulation, starches may
remain in the body for years, with unknown long-term consequences and albumin solu-
tions are in scarce supply and pose infection hazards. Cost and availability are also an issue
with colloid solutions.
The use of colloid-based primes has not been shown to significantly influence clinical out-
comes such as the duration of ventilatory support and length of intensive care unit (ICU) or
hospital stay. None of the types of colloids has been shown to have significant advantages over
another. Albumin may have a beneficial effect as a constituent of the prime: it is thought to
coat the extracorporeal circuit, making it appear less “foreign” to the body™s immune mecha-
nisms and so to ameliorate the inflammatory response.
The lack of measurable benefit, potential risks and the significant cost penalty incurred
in comparison to crystalloid fluids have resulted in colloids no longer being widely used as a
priming fluid in adult CPB.
The use of mannitol as a colloidal fluid added to the CPB prime is perhaps the one excep-
tion to the above discussion. Mannitol is a common constituent of primes, but the indica-
tion for its use is for its properties as a potent osmotic diuretic, rather than to simply raise
the oncotic pressure of the prime. Maintenance of urine output both during CPB and in the
immediate postoperative period is desirable to enhance elimination from the body of the fluid
load presented by prebypass iv fluids, the priming fluid volume and cardioplegia solution. It
has also been postulated that mannitol may help to preserve renal function and reduce the
incidence of post-CPB renal dysfunction, although the evidence for this is extremely weak. In
Chapter 3: Priming solutions for CPB circuits

addition, mannitol is a free radical scavenger and it is appealing to think that the free radicals
produced during periods of hypoperfusion, ischemia and reperfusion might be “mopped up”
during bypass, thus reducing end-organ damage. However, this concept remains unproven in
any clinically relevant way.

Experimental oxygen-carrying solutions
The idea of using oxygen-carrying solutions as blood substitutes may be an attractive means
of maintaining oxygen delivery. They would address the expense, limited supply and disease
transmission associated with blood transfusion. Both hemoglobin-based substitutes and per-
fluorocarbons have been researched in the context of use in the CPB priming fluid, but none
have yet proven to be both safe and efficacious as alternatives for oxygen carriage. Despite
several decades of research no molecule seems close to being marketed as a viable alternative
to red cells in the clinical arena and it remains to be seen whether there is any future for the
use of these oxygen-carrying solutions during CPB.

solution in cardiopulmonary bypass affect
Suggested Further Reading outcome? A prospective randomized study.
• Bunn F, Alderson P, Hawkins V. Colloid J Thorac Cardiovasc Surg 1989; 98(5 Pt1):
solutions for fluid resuscitation. Cochrane 751“6.
Database Syst Rev 2003; Art No
• Paone G, Silverman N. The paradox of on
bypass transfusion thresholds in blood
• Cooley DA, Beall AC, Grondin P. Open conservation. Circulation 1997; 96(suppl II):
heart operations with disposable II-205“8.
oxygenators, 5% dextrose prime, and
• Rawn JD. Blood transfusion in cardiac
normothermia. Surgery 1962; 52:713“19.
surgery: a silent epidemic revisited.
• Fang WC, Helm RE, Krieger KH, et al. Circulation 2007; 116(22): 2523“4.
Impact of minimum haematocrit during
• Riegger L, Voepel-Lewis T, Kulik T, et al.
cardiopulmonary bypass on mortality in
Albumin versus crystalloid prime solution
patients undergoing coronary artery surgery.
for cardiopulmonary bypass in young
Circulation 1997; 96(suppl II): II-194“99.
children. Crit Care Med 2002; 30(12):
• Harris EA, Seelye ER, Barratt-Boyes BG. 2649“54.
Respiratory and acid-base changes during
• Rosengart TK, DeBois WJ, Helm RE.
CPB in man. Br J Anaesth 1970; 42: 912“21.
Retrograde autologous priming (RAP) for
• Hoeft A, Korb H, Mehlhorn U, et al. cardiopulmonary bypass: a safe and
Priming of cardiopulmonary bypass with effective means of decreasing hemodilution
human albumin or ringer lactate: effect on and transfusion requirements. J Thorac
colloid osmotic pressure and extravascular Cardiovasc Surg 1998; 115(2): 426“38.
lung water. Br J Anaesth 1991; 66:73“80.
• Rosengart TK, Helm RE, DeBois WJ. Open
• Klein HG, Spahn DR, Carson JL. Red blood heart operations without transfusion using a
cell transfusion in clinical practice. Lancet multimodality blood conservation strategy in
2007; 370(9585):,415“26. 50 Jehovah™s Witness patients: implications
for a “bloodless” surgical technique. J Am
• Lilley A. The selection of priming fluids for
Coll Surg 1997; 184: 618“29.
cardiopulmonary bypass in the UK and
Ireland. Perfusion 2002; 17:315“319. • Russell JA, Navickis RJ, Wilkes MM.
Albumin versus crystalloid for pump
• Liskaser FJ, Bellomo R, Hayhoe M, et al. Role
priming in cardiac surgery: a meta-analysis
of pump prime in etiology and pathogenesis
of controlled trials. J Cardiothorac Vasc
of cardiopulmonary bypass “ associated
Anesth 2004; 18(4): 429“37.
acidosis. Anesthesiology 2000; 93: 1170“3.
• Serious Hazards of Transfusion, http://www.
• Marelli D, Paul A, Samson R, et al. Does
the addition of albumin to the prime

Anticoagulation, coagulopathies,

blood transfusion and conservation
4 Liza Enriquez and Linda Shore-Lesserson

Anticoagulation is required for any form of extracorporeal circulation to prevent activa-
tion of the coagulation system by contact between blood and artificial, non-biological
Cardiopulmonary bypass circuits comprise of a large surface area of mainly plastic mate-
rial, which if left to come into contact with blood without appropriate anticoagulation, would
result in formation of clots within the circuit in a matter of minutes. In order to safely con-
duct CPB for the duration required for surgical procedures, or to maintain patients on extra-
corporeal support, anticoagulation must be adequate to prevent the development of even
“minor” clots. Inadequate anticoagulation can in its most serious form lead to death and in
lesser forms lead to impairment of organ function, usually manifest as neurological or renal
dysfunction. Furthermore, any clots within the CPB system can trigger the development of
disseminated intravascular coagulation (DIC), which results in the rapid consumption of
clotting factors and failure of the body™s coagulation system.
Heparin is the most commonly used anticoagulant in the context of CPB. This chapter
describes the coagulation pathway, the pharmacology of heparin, monitoring of anticoagu-
lation status, problems associated with heparin usage, alternatives to heparin, the reversal
of anticoagulation following termination of CPB and the prevention and management of

The coagulation cascade
Coagulation occurs by interaction of a series of proteins that are activated and propagated
by a variety of stimuli, including contact with foreign surfaces, contact with receptors on the
surfaces of platelets and by factors produced by the systemic inflammatory response, all of
which are pertinent in the context of CPB.
Most of the proteins required for the cascade are produced by the liver as inactive pre-
cursors which are then modified into clotting factors. The implication of the term “cascade”
is that a small stimulus results in a reaction which may be amplified to produce a significant
There are two routes for activation of the coagulation system. The intrinsic pathway is
activated by contact with collagen from damaged blood vessels or any negatively charged
surface. Platelet activation is normally involved. The extrinsic pathway is activated by contact
with tissue factor from the surface of extravascular cells. Both routes end in a final common
pathway “ the proteolytic activation of thrombin and the cleaving of fibrinogen to form a
fibrin clot. The intrinsic pathway is the predominant route, with the extrinsic pathway acting
synergistically (see Figure 4.1).

Cardiopulmonary Bypass, ed. S. Ghosh, F. Falter and D. J. Cook. Published by Cambridge University Press.
© Cambridge University Press 2009.
Chapter 4: Anticoagulation

Figure 4.1. Overview of extrinsic and intrinsic clotting pathways.

Pharmacological strategies for anticoagulation during CPB
Unfractionated heparin (UFH) remains the standard anticoagulant for CBP for several rea-
sons. It is relatively safe, easy to use, has a fast onset of action and is measurable, titratable and
reversible. It is also cost-effective.

Native heparin is a polymer with a molecular weight ranging from 3 to 40 kDa, although
the average molecular weight of most commercial heparin preparations is in the range of
12“15 kDa. Heparin is a member of the glycosaminoglycan family of carbohydrates (which
includes the closely related molecule heparan sulfate) and consists of a variably sulfated
repeating disaccharide unit that is negatively charged at physiological pH. Heparin is nor-
mally released by mast cells and basophils in the body and is commercially derived from
bovine lung or porcine intestinal mucosa.

Mechanism of anticoagulant action
Heparin contains a specific pentasaccharide sulfation sequence that binds to the enzyme
inhibitor antithrombin III (AT-III) causing a conformational change that results in increasing
Chapter 4: Anticoagulation

AT-III™s activity. The activated AT-III then inactivates thrombin and other proteases involved
in blood clotting. These factors include IIa (thrombin), Xa, IXa, XIa and XIIa. It is most active
against thrombin and Xa. The rate of inactivation of these proteases by AT-III can increase by
up to 1000-fold due to the binding of heparin. In addition, heparin increases the activity of
heparin cofactor II, which also inhibits thrombin.
Heparin™s onset is immediate and has a half-life of approximately 2.5 hours at doses of
300“400 USP units (U)/kg. It is provided in units, with 1 U, according to the US Pharmaco-
poeia, maintaining fluidity of 1 ml of citrated sheep plasma for 1 hour after recalcification.

Dosing of heparin can vary among institutions. The most common initial dose for CPB
is 300“400 USP U/kg. Some centers base the initial dose on a bedside ex vivo heparin
dose-response titration. Many institutions add heparin to the CPB priming solution at
approximately the same concentration as that of the patient™s bloodstream or as a fixed dose.
Supplemental heparin doses are guided by monitoring of anticoagulation using the activated
clotting time (ACT) or heparin concentration monitoring.

The ACT is a funtional assay of heparin anticoagulation and is the most widely employed
test. Most institutions use a level between 400 and 480 seconds as an acceptable ACT level at
which to conduct CPB. Hypothermia, hemodilution, platelet function abnormalities and low
fibrinogen are some of the factors that can prolong ACT, even in the setting of incomplete
heparinization. ACT monitoring will be discussed in further detail under the section “Point-
of-care testing.”

Heparin resistance
Heparin resistance is defined as failure to raise the ACT to expected levels despite an ade-
quate dose and plasma concentration of heparin. Clinical conditions involving congenital or
acquired AT-III deficiency are associated with heparin resistance. Hemodilution during CPB
can decrease AT-III levels, though usually this does not result in heparin resistance because it
is also associated with dilution of procoagulant factors. Prior treatment with heparin causes
depletion or dysfunction of AT-III and this is the most likely reason that cardiac surgery
patients will present with heparin resistance. Another cause of heparin resistance is the pres-
ence of large quantities of heparin-binding protein in the circulation, which binds to and
inactivates heparin.
Administering additional heparin boluses of up to 600“800 USP U/kg may be necessary
to obtain an ACT level sufficient for the conduct of CPB. Definitive treatment is aimed at
increasing levels of AT-III. This can be done by administering fresh frozen plasma (FFP),
which contains antithrombin; however, exposure to transfusion-borne infectious diseases is
a risk. Supplemental AT-III concentrate is another alternative and provides greater protection
against disease transmission than FFP. AT-III is also available in recombinant formulations,
which have been used to treat congenital deficiency.

Heparin-induced thrombocytopenia (HIT)
Heparin-induced thrombocytopenia (HIT) develops in 5% of patients receiving heparin
and is categorized into two subtypes. The first type is generally mild and involves a transient
Chapter 4: Anticoagulation

decrease in platelet count. These patients can safely receive heparin for cardiac surgery. The
second type occurs later in heparin therapy (5“14 days after administration) and is a more
severe, immune-mediated decrease in the platelet count. Antibodies against the complex of
platelet factor 4 (PF4) and heparin bind to platelets, activate the platelets and cause the result-
ant platelet count to drop precipitously. In the setting of endothelial injury, this enhance-
ment in platelet activation predisposes to the formation of platelet clots (white clots) and
Heparin-induced thrombocytopenia is a clinicopathological syndrome and requires both
clinical evidence (thrombocytopenia or thrombosis) and laboratory findings to confirm the
diagnosis. Laboratory diagnosis can be made in two ways: functional assay or antibody-based
assay. Functional tests detect heparin-dependent platelet activation in the presence of the
patient™s sera and UFH. The serotonin release assay (SRA) is considered the gold standard:
when an affected patient™s serum is exposed to heparin, an exaggerated reaction occurs and
serotonin is released from dense granules. Using C-14-labeled serotonin the concentration
released is then measurable. Other functional tests include the heparin-induced platelet acti-
vation assay (HIPAA) and the platelet-rich plasma (PRP) aggregation assay, which measure
hyper-aggregability in response to heparin. Enzyme-linked immunological assays measure
IgG, IgM or IgA antibodies that bind to the PF4/heparin complex.
The Seventh American College of Chest Physicians (ACCP) Conference on Antithrom-
botic and Thrombolytic Therapy resulted in the publication of evidence-based guidelines.
Recommendations were made for patients undergoing cardiac surgery with previous HIT, as
well as those with acute or subacute HIT. Grade 1 recommendations are strong and indicate
that a high level of evidence suggests that the benefits of a particular intervention outweigh
the risks, burden and costs. Grade 2 recommendations suggest that individual patients™ or
physicians™ values may lead to different choices. Management of these patients can be sum-
marized as follows: Patients with a history of HIT who are antibody negative and require
cardiac surgery can receive unfractionated heparin. For patients with acute HIT who require
cardiac surgery, the guideline developers recommend delaying surgery, if possible, until HIT
antibodies are negative or using alternative anticoagulant approaches such as bivalrudin or
hirudin. Combinations of unfractionated heparin and antiplatelet agents such as epoproste-
nol or tirofiban are also recommended.

Alternatives to unfractionated heparin
Low-molecular-weight heparin (LMWH)
Intravenously administered LMWH has a half-life at least twice as long as that of UFH and
possibly several times as long for some LMWH compounds. Problems during CPB arise from
the fact that protamine neutralization only reverses the factor IIa inhibition and leaves the
predominant factor Xa inhibition intact. LMWH therapy also complicates heparin monitor-
ing because activated partial thromboplastin time (APT) (and presumably ACT) is much less
sensitive to Xa inhibition and will not accurately measure the full anticoagulant effect. Factor
Xa inhibition can be measured, but not with a simple bedside test. LMWHs are not recom-
mended for use in HIT patients

Danaparoid is a low-molecular-weight heparinoid with a long half-life (18“24 hours). It is a
polysulfated glycosaminoglycan composed of heparan sulfate (84%), dermatan sulfate (12%)
Chapter 4: Anticoagulation

and chondroitin sulfate (4%). There is a 30% cross-reactivity with heparin antibodies, which
precludes its use in HIT patients. Monitoring is via anti-Xa levels and currently there is no
antidote. It has been studied in CPB and has not been proven to be safe because of excess
bleeding and thrombosis. Danaparoid is no longer available in the USA.

Ancrod (viprinex) is a defibrinogenating agent extracted from Malayan pit viper venom.
Fibrinogen levels must be <500 mg/l prior to instituting CPB, which requires more than 12
hours after administering Ancrod to be achieved. Other disadvantages include no antidote,
lack of monitoring and bleeding complications. Ancrod is not recommended for use in HIT

Direct thrombin inhibitors (DTIs)
These directly inhibit the procoagulant and prothrombotic actions of thrombin and do not
require a cofactor. Their advantage is that they do not interact with or produce heparin-
dependent antibodies. The main differences between the two types of thrombin inhibitors are
listed in Table 4.1
• Lepirudin “ This is a recombinant analogue of the anticoagulant hirudin produced in
leech saliva. It has a short half-life of 80 minutes and is monitored via activated partial
thromboplastin time (aPTT) or ACT and has no antidote. It can, however, be
eliminated by hemofiltration. Lepirudin is metabolized by the kidney requiring dose
adjustments in patients with renal insufficiency. The advantage is that it lacks
cross-reactivity with heparin but antihirudin antibodies develop in as many as 60%
of patients. Current evidence suggests that these antihirudin antibodies do not
interfere with the anticoagulant activity of hirudin and their significance is unknown.
• Argatroban “ This is a synthetic molecule derived from L-arginine and is widely used
in patients with HIT who require percutaneous coronary intervention. Its half-life is
45“55 minutes, it lacks cross-reactivity with heparin antibodies and is monitored via
the aPTT or ACT. There is no antidote. Argatroban is metabolized in the liver requiring
dose adjustments in patients with moderate liver disease. Argatroban has not yet been
approved for use in CPB. It is not available in the UK.
• Bivalirudin “ This is a synthetic peptide based on the structure of hirudin. Its advantage
is its short half-life of 25 minutes. It is monitored via the aPTT, ACT or ecarin clotting
time, if available. The dose for CPB is a 1 mg/kg bolus followed by a 2.5 mg/kg/hour
infusion. Bivalirudin is metabolized by proteolytic enzymes present in the blood and by
the kidney. Only minor dose adjustments are necessary for patients with renal insuffi-
ciency. Multicenter trials have demonstrated it is not inferior to heparin when used in
CPB. Currently, bivalirudin is widely used in cardiac catheterization laboratories as the
anticoagulant for percutaneous coronary intervention, even in patients without HIT.

“Reversal” of anticoagulation
Heparin neutralization
Protamine is a naturally occurring polypeptide with multiple cationic sites, a “polycation”
that binds and inactivates heparin.
Chapter 4: Anticoagulation

Table 4.1. Key differences between direct thrombin inhibitors (DTIs) and indirect thrombin inhibitors

Heparin DTIs
Mode of action Indirect Direct
Cofactor needed Yes “ AT-III No
Inhibits clot-bound thrombin No Yes
Activates platelets Yes No
Antigenicity Yes No “ bivalirudin; Yes “ hirudin
Antidote drug Yes “ protamine No

Several protamine dosing techniques have been utilized. The recommended dose range of
protamine for heparin reversal is 1“1.3 mg protamine per 100 U of heparin. Other approaches
include calculating the protamine dose based on the heparin dose“response curve generated
by some automated systems such as the Hepcon (Medtronic Inc). Protamine must be admin-
istered slowly in order to prevent adverse hemodynamic effects such as hypotension.
Protamine reactions have been classified into three types. A Type I reaction may result
from rapid administration resulting in decreases in both systemic and pulmonary arterial
pressures, decreased preload and hypotension. The Type II reaction is immunological and
is categorized as IIA anaphylaxis, IIB anaphylactoid and IIC non-cardiogenic pulmonary
edema. Type III reactions are caused by heparin/protamine ionic complexes that can adhere
in the pulmonary circulation and cause pulmonary vasoconstriction. This results in cata-
strophic pulmonary hypertension and resultant right heart failure.
Adequacy of neutralization should be assessed by repeating ACT 3“5 minutes after

Alternatives to protamine
This synthetic polycation can be administered to patients who are allergic to protamine without
adverse effects. However, when administered rapidly, hexadimethrine mimics the response to
rapid administration of protamine because it forms complexes with heparin. Systemic hypo-
tension, decreased systemic vascular resistance (SVR) and pulmonary vasoconstriction are
among the adverse reactions seen. Following reports of renal toxicity, hexadimethrine was
withdrawn from clinical use in the USA.

Platelet factor 4 (PF4)
Platelets contain PF4, a potent antiheparin compound, on their surface, which utilizes lysine
residues at it C-termini to neutralize heparin, rather than the electrostatic binding that occurs
with protamine.
It is hypothesized that the cause of heparin-induced thrombocytopenia is an immuno-
logical reaction to the PF4/heparin complex.

Methylene blue
This chemical dye binds electrostatically to heparin in a similar fashion to protamine. Large
doses do not effectively restore the ACT to normal. An inhibitor of nitric oxide synthetase,
Chapter 4: Anticoagulation

methylene blue increases pulmonary and systemic vascular resistance at higher doses,
making its use quite hazardous.

Omit neutralization
Due to drug elimination, heparin will dissipate spontaneously with time with consequent
decline in anticoagulation. This option may result in an increase in transfusion requirements,
hemodynamic instability and consumptive coagulopathy as a result of hemorrhage and

Heparinase, an enzyme produced by the gram-negative Flavobacterium, hydrolyzes the
heparin molecule into smaller inactive fragments. Some of these small fragments do possess
the potential for some anti-Xa activity, thus the utility of heparinase in reversing heparin after
CPB is limited.

Monitoring anticoagulation status in the operating room
Point-of-care testing (POC)
Point-of-care testing devices allow the monitoring of hemostasis at “the bedside” rather
than sending specimens to a central laboratory facility. These instruments rapidly assess
coagulation and/or platelet function to aid in providing appropriate targeted therapy. As
a result there is a reduction in blood loss and transfusion, fewer complications and cost
The ACT is an automated variation of the Lee“White clotting time and is the most com-
monly used test to measure heparin anticoagulation. It uses an activator such as celite or
kaolin to activate clotting, then measures the clotting time in a test tube or cartridge. Normal
baseline ACT levels, without any heparin in the blood, should be between 80 and 140 seconds.
For CPB, prolongation of the ACT to greater than 400 or 480 seconds is considered adequate,
though this is highly debated. For off-pump coronary artery bypass operations (OPCAB),
“partial heparinization” may be used in some centers whereby an ACT greater than 300 sec-
onds is targeted. The Hemochron (International Technidyne Corp, Edison, NJ, USA) and
the HemoTec ACT (Medtronic HemoTec, Parker, CO, USA) are two automated ACT devices
used in the operating room.
During CPB the sensitivity of the ACT to heparin is altered by hemodilution and hypo-
thermia. As a result ACT measurements do not correlate with heparin concentration or with
antifactor Xa activity. The Hepcon HMS® analyzer (Medtronic Inc, Minneapolis, MN, USA)
uses protamine titration assays to determine the blood heparin level. This device can also
provide a dose“response curve for an individual patient and indicate how much heparin to
administer in order to reach a specific targeted ACT before going onto CPB. In addition, it can
be utilized for protamine dosing after CPB.
Other tests used less commonly to monitor heparin effectiveness during CPB are
High Dose Thrombin Time (HiTT) (International Technidyne Inc, Edison, NJ, USA)
and Heparin Management Test (HMT) Cascade Analyzer (Helena, Beaumont, TX, USA).
HiTT measures the conversion of fibrinogen to fibrin by thrombin and, unlike ACT,
HiTT is not affected by hemodilution, hypothermia or aprotinin. The Cascade® coagulation
Chapter 4: Anticoagulation

analyzer can measure prothrombin time (PT), aPTT and HiTT levels in whole blood at
the point of care.

Tests of platelet function
Thromboelastography (TEG)
Thromboelastography measures the viscoelastic properties of blood as it is induced to clot
under a low shear environment resembling sluggish venous flow. The patterns of change in
shear“elasticity enable the determination of the kinetics of clot formation and clot growth
and provide information about clot strength and stability. The strength and stability of the
clot provides information about the ability of the clot to cause hemostasis effectively, while the
kinetics determine the adequacy of quantitative factors available for clot formation.
There are four major parameters to the TEG tracing, which measure different stages of clot
development: R, K, alpha angle and MA (maximal amplitude). In addition, clot lysis indices
are measured as the amplitude at 30 and 60 minutes after MA (LY30 and LY60) (see Figure
4.2). Normal values vary depending on the type of activator used.
• R value “ This is a measure of clotting time from the start of the bioassay to the
initial fibrin formation. R time can be prolonged by coagulation factor deficiencies,
anticoagulation, severe thrombocytopenia and hypofibrinogenemia. R time can be
shortened in hypercoagulability states.
• K value “ This represents clot kinetics, measuring the speed to reach a specific level of
clot strength. It is the time from beginning of clot formation (the end of R time) until
the amplitude reaches 20 mm. K time can be prolonged by coagulation factor deficien-
cies, hypofibrinogenemia, thrombocytopenia and thrombocytopathy. It is shortened in
hypercoagulable states.
• Alpha angle “ This is the angle between the line in the middle of the TEG tracing and
the line tangential to the developing “body” of the TEG tracing. The alpha angle
represents the acceleration (kinetics) of fibrin build up and cross-linking (clot
strengthening). It is increased in hypercoagulable states and decreased in
thrombocytopenia and hypofibringenemia.
• MA “ This is the maximum amplitude reflecting the ultimate strength of the clot,
which depends on platelet number and function and platelet interactions with
fibrin. It is increased in hypercoagulable states and decreased in thrombocytopenia,
thrombocytopathy and hypofibrinogenemia.
• Lysis “ Indices measured by the TEG include LY30 and LY60. These are the percentage
ratio of the amplitude at 30 and 60 minutes after MA to the MA itself. LY30 or LY60 are
increased in states of fibrinolysis.

Other POC tests of platelet function
Sonoclot is an alternative test to TEG of the viscoelastic properties of blood. Sonoclot uses an
ultrasonic vibrational method to stimulate clot formation.
The newest group of POC platelet function tests were specifically designed to measure
agonist-induced platelet-mediated hemostasis. These monitoring systems include the Veri-
fyNow (Accumetrics, San Diego, CA, USA), the Clot Signature Analyzer (CSA, Xylum, Scars-
dale, NY, USA), the Platelet Function Analyzer, PFA-100 (Dade Behring, Miami, FL, USA)
and Plateletworks (Helena Laboratories, Beaumont, TX, USA) (see Table 4.2). The CSA is not
currently FDA approved.
Chapter 4: Anticoagulation

Figure 4.2. Normal TEG trace (refer to text for details of abbreviations).

Table 4.2. POC devices to assess platelet function

Instrument Mechanism/agonist Clinical utility
Thromboelastograph® Viscoelastic/thrombin, ADP, Post-CPB, liver transplant, pediatrics,
arachidonic acid (AA) obstetrics, drug efficacy
Sonoclot® Viscoelastic/thrombin Post-CPB, liver transplant
PlateletWorks® Platelet count ratio/ADP, AA, Post-CPB, drug therapy
PFA-100® In vitro bleeding time/ADP, von Willebrand™s disease, congenital
epinephrine disorder, aspirin therapy,
VerifyNow® Agglutination/thrombin receptor Drug therapy
agonist peptide (TRAP), AA, ADP
Clot Signature Analyzer® Shear-induced in vitro bleeding Post-CPB, drug effects
Whole blood aggregometry Electrical impedance/many Post-CPB

Coagulation disorders after CPB
Persistent bleeding after CPB is multifactorial. It is usually associated with long bypass times
(>2 hours) as a result of which platelet dysfunction, hemodilution, protein activation/con-
sumption and fibrinolysis occur. Prompt diagnostic and therapeutic action is necessary to
avoid impaired hemodynamics due to hemorrhage.

Platelet abnormalities
Thrombocytopenia can occur after CPB due to dilution of blood volume with the extracor-
poreal circuit volume and platelet consumption or sequestration. Platelet function impair-
ment is considered to be the main hemostatic defect during CPB. Platelet dysfunction
occurs from contact with the extracorporeal surfaces, hypothermia, down-regulation of
receptors and by exposure to heparin and protamine. In addition, patients on antithrombotic
Chapter 4: Anticoagulation

medications preoperatively can have platelet dysfunction that becomes significantly exag-
gerated after CPB. Many patients taking aspirin or other platelet-inhibiting drugs regularly
cannot discontinue therapy within 7 days of surgery and unfortunately no antidote can
correct the platelet defect. These patients have a very difficult bleeding diathesis that often
requires multiple transfusions and/or pro-coagulant factor therapy.

Systemic inflammatory response syndrome
Contact of blood with the CPB circuit results in the systemic inflammatory response syndrome
(SIRS), which is characterized by the activation of the kallikrein“bradykinin system, com-
plement, coagulation pathways and fibrinolysis. SIRS may cause disseminated intravascular
coagulation (DIC) by aggravating consumption of coagulation factors (see Chapter 11).

Heparin rebound
This phenomenon may be observed after apparent adequate reversal of heparinization and
may be explained by a redistribution either of protamine to peripheral compartments or of
peripherally bound heparin to the central compartment. Treatment is with small incremental
doses of protamine.

Hemostasis is impaired by hypothermia in many ways including: sequestration of platelets,
transient platelet dysfunction, activation of a specific heparin-like inhibitor of Xa, slowing of
the enzymatic reactions involved in the coagulation cascade and accentuation of fibrinolysis.

The CPB circuit contains a large surface of thrombogenic material and, despite clinically
adequate doses and blood concentrations of heparin, activation of coagulation pathways is
accompanied by persistent fibrinolytic activity causing consumption of coagulation factors.

Other causes
Hemodilution, liberal use of cardiotomy suction and prolonged CPB all further aggravate

Prevention of bleeding
Antifibrinolytic agents
The synthetic antifibrinolytic agents µ-aminocaparoic acid (EACA) and tranexamic

acid (TA) bind to lysine binding sites in both plasminogen and plasmin and produce a
structural change. This prevents the conversion of plasminogen to plasmin and also
prevents the activation of plasmin. Minimization of plasmin activity inhibits fibrin
degradation, decreases the formation of fibrin degradation products (FDPs) and
decreases lysis of existing clots. It is these FDPs that inhibit platelet function, so the
lysine analogue anti-fibrinolytic agents also have an indirect effect in preserving platelet
Chapter 4: Anticoagulation

function. Dosing of these agents is highly variable and is dependent on institution and
country. Typical regimens are given below:
· EACA 100“150 mg/kg bolus, followed by infusion at 10“15 mg/kg/hour, or 4“6 g
bolus, followed by infusion at 1 g/hour; and
· TA: 10“50 mg/kg bolus, followed by infusion at 1“15 mg/kg/hour, or 5 g bolus,
followed by repeat boluses to total of 15 g.
• Aprotinin, a serine protease inhibitor isolated from bovine lung that inhibits several
enzymatic activators of coagulation including plasmin and kallikrein. Its action on
kallikrein leads to the inhibition of the formation of factor XIIa. As a result, both
the intrinsic pathway of coagulation and fibrinolysis are inhibited. Its action on
plasmin independently slows fibrinolysis. Dosing “ Full Dose Regimen (Hammer-
smith Protocol): a test dose of 10 000 kallikrein-inhibiting units (KIU) should be
administered followed 10 minutes later by 2 million KIU as a bolus, 2 million KIU
added to the pump prime and 500 000 KIU/hour as an infusion. As of November
2007, worldwide marketing of aprotinin has been suspended after publication of
observational studies and the randomized Canadian BART (“Blood Conservation
using Antifibrinolytics: A Randomized Trial”) trial. The BART trial was an independ-
ent randomized control trial conducted in high-risk cardiac patients that was halted
after data revealed reduced bleeding as well as an increase in all-cause mortality in
patients receiving aprotinin compared to those receiving either aminocaproic acid
or tranexemic acid.
• Heparin and protamine dosing: ACT should return to baseline following administra-
tion of protamine; additional doses of protamine (25“50 mg) may be necessary.
Reheparinization (heparin rebound) after apparent adequate reversal may be explained
by a redistribution either of protamine to peripheral compartments or of peripherally
bound heparin to the central compartment.
• Desmopressin is an analogue of vasopressin that releases von Willebrand factor (VWF)
from normal endothelial cells and is used in the treatment of hemophilia. Factor VIII
coagulant activity increases 2- to 20-fold in addition to an increase in factor XII levels.
Desmopressin has been beneficial to subgroups of patients, such as cirrhotic and
uremic patients, undergoing cardiac surgery. It affords no hemostatic benefit to patients
taking aspirin prior to cardiac surgery and is not recommended as a prophylactic
hemostatic agent for patients undergoing elective cardiac surgery. It may also be used
to augment the function of exogenous transfused platelets.
• Use of non-pharmacological strategies:
· The use of heparin-bonded CPB circuits makes the extracorporeal circuit more
biocompatible thus effectively reducing the proinflammatory aspects of CPB.
· Some clinicians advocate the use of a reduced heparin dose in conjunction with
heparin-bonded circuits to decrease postoperative blood loss and transfusion

Management of the bleeding patient
Determining the cause of bleeding quickly is vital to expedite treatment of the bleeding
patient. Surgical causes of bleeding generally present with generous chest tube drainage
early after operation. Non-surgical causes of bleeding usually manifest as a generalized
Chapter 4: Anticoagulation

Hypothermia (<35°C) accentuates hemostatic defects and should be corrected. The
administration of platelets and coagulation factors should generally be guided by additional
coagulation studies, but empirical therapy may be necessary when such tests are not readily
available, or following massive transfusion.
If oozing continues despite adequate surgical hemostasis and the ACT is normal or the
heparin“protamine titration assay shows no residual heparin, thrombocytopenia or platelet
dysfunction is most likely. Both defects are recognized complications of CPB. Platelet
transfusion may be necessary and should be given to maintain the platelet count above
100 000/μl. Significant depletion of coagulation factors, particularly factors V and VIII, during
CPB is less commonly responsible for bleeding but should be treated with fresh frozen plasma;
both the prothrombin time and partial thromboplastin time are usually prolonged in such
instances. Hypofibrinogenemia (fibrinogen level <100 mg/dl or a prolonged thrombin time
without residual heparin) should be treated with cryoprecipitate. Desmopressin (DDAVP),
0.3 μg/kg (intravenously slowly over 20 minutes), can increase the activity of factors VIII and
XII and the von Willebrand factor by releasing them from the vascular endothelium. DDAVP
may be effective in reversing qualitative platelet defects in some patients, but is not recom-
mended for routine use.
Accelerated fibrinolysis may occasionally be encountered following CPB and should be
treated with µ-aminocaproic acid (5 g followed by 1 g/hour) or tranexamic acid (10 mg/kg),
if not already being given; the diagnosis should be confirmed by elevated fibrin degradation
products (>32 mg/ml), or evidence of clot lysis on thromboelastography.
Recombinant factor VIIa (rFVIIa, NovoSeven) is a vitamin K-dependent glycoprotein
that promotes hemostasis by activating the extrinsic pathway of the coagulation cascade.
Tissue factor-bearing cells present tissue factor to rFVIIa. This complex can activate factor
X to factor Xa, as well as factor IX to IXa. Factor Xa, in complex with other factors, then
converts prothrombin to thrombin, which leads to the formation of a hemostatic plug by
converting fibrinogen to fibrin and thereby inducing local hemostasis. This process can also
occur on the surface of activated platelets. rFVIIa has been approved for use in hemophili-
acs who are resistant to factor VIII concentrates. Numerous reports have been published
in cardiac surgery as an “off-label” treatment option in patients with probable or identifi-
able coagulation defects or as a rescue therapy in hemorrhagic patients refractory to other

Transfusion and the use of algorithms
Point-of-care testing in conjunction with transfusion algorithms can reduce both transfu-
sion requirements and blood loss. Many transfusion algorithms have been published and
demonstrate a successful reduction in bleeding and transfusion requirements in high-risk
cardiac surgical patients. Most of these transfusion algorithms utilize the thromboelasto-
gram, others use point of care PT and international normalized ratio (INR) testing, and
others use tests of platelet function and number. Any combination of tests that examines
the presumed defects incurred during CPB will accomplish the same goal “ the reduction
in the transfusion of blood products by more rational and specific guidance of that transfu-
sion therapy.
Hemostasis and bleeding in conjunction with cardiac surgery is a multifactorial problem.
The defects that occur are dynamic in origin and knowledge of physiological responses to
CPB continue to evolve.
Chapter 4: Anticoagulation

4th ed. Philadelphia: Lippincott Williams
Suggested Further Reading and Wilkins; 2008: 494“515.
• Despotis GJ, Joist JH, Goodnough LT. • Spiess BD, Gillies BSA, Chandler W, Verrier
Monitoring of hemostasis in cardiac E. Changes in transfusion therapy and
surgical patients: impact of point-of-care reexploration rate after institution of a
testing on blood loss and transfusion blood management program in cardiac
outcomes. Clinl Chem 1997; 43: surgical patients. J Cardiothorac Vasc Anesth
1684“96. 1995; 9: 168“73.
• Raivio P, Suojaranta-Ylinen R, Kuitunen • Spiess BD, Horrow JC, Kaplan JA.
AH. Recombinant factor VIIa in the Transfusion medicine and coagulation
treatment of postoperative hemorrhage disorders. In Kaplan JA, ed. Kaplan™s Cardiac
after cardiac surgery. Ann Thorac Surg 2005; Anesthesia, 5th ed. Philadelphia: Elsevier
80: 66“71. Saunders; 2006: 937“84.
• Rochon AG, Shore-Lesserson L. • Spiess BD, Tuman KJ, McCarthy RJ,
Coagulation monitoring. Anesthesiol Clin DeLaria GA, Schillo R, Ivankovich AD.
North Am 2006; 24(4): 839“56. Thrombelastography as an indicator of
post-cardiopulmonary bypass
• Romagnoli S, Bevilacqua S, Gelsomino S,
coagulopathies. J Clin Monit 1987; 3: 25“30.
et al. Small-dose recombinant activated
factor VII (NovoSeven) in cardiac surgery. • The Seventh American College of Chest
Anesth Analg 2006; 102: 1320“6. Physicians (ACCP). Conference on
Antithrombotic and Thrombolytic Therapy:
• Shore-Lesserson L. Coagulation
evidence-based guidelines. Chest 2004; 126:
monitoring. In Kaplan JA, ed. Kaplan™s
Cardiac Anesthesia, 5th ed. Philadelphia:
Elsevier Saunders; 2006: 557“82. • Warkentin TE, Greinacher A. Heparin-
• Shore-Lesserson L, Horrow JC, Gravlee GP. induced thrombocytopenia: recognition,
Coagulation management during and after treatment, and prevention: The 7th ACCP
cardiopulmonary bypass. In Kaplan JA, Conference on Antithrombotic &
Reich DL, Lake CL, Konstadt SN, eds. Thrombolytic Therapy. Chest 2004; 126:
A Practical Approach to Cardiac Anesthesia, 311S“337S.

Conduct of cardiopulmonary

5 Betsy Evans, Helen Dunningham and John Wallwork

The pump is your friend!
Caves, 1976

Cardiopulmonary bypass is an incredible facility when used correctly by the team of surgeon,
anesthetist and perfusionist. A comprehensive understanding of the physiology of CPB is
essential for optimum benefit, together with knowledge of the risks, limitations and poten-
tial adverse effects if used incorrectly. The management of CPB involves a multi-disciplinary
approach with coordinated actions and precise communication being crucial for a safe, effec-
tive outcome.
Before each case the conduct of CPB should be planned. All members of the team need to
be aware of the intended method for cannulation, the systemic and myocardial temperatures
required during surgery, the technique of myocardial protection to be used, whether deep
hypothermic circulatory arrest (DHCA) will be required and the most appropriate sites for
monitoring during CPB.
Prior to assembly of the CPB circuit, patient demographic data and information relating
to physiological and pathological status are required to enable selection of equipment tailored
to the patient™s needs.

Arterial cannulation
The arterial cannula is usually the narrowest part of the CPB circuit with resultant high resist-
ance, pressure gradients, high velocity jets and turbulence. The effect of jets on the interior
wall of the aorta can lead to arterial dissection, embolization and flow disturbances in head
and neck vessels.
The “performance index” of an arterial cannula is the pressure gradient versus the outer
diameter at any given flow. The narrowest portion of the catheter that enters the aorta should
be as short as safely possible and the diameter should then gradually increase in size to mini-
mize the gradient. Pressure gradients greater than 100 mmHg can cause excessive hemolysis
and should be avoided. Many different types of cannulae are available and are discussed fur-
ther in Chapter 1.
“Straight” arterial cannulae are the most commonly used, with some having a flange to
allow secure fixation to the aorta with minimal tip within the vessel. The straight design allows
non-turbulent blood flow through the cannula, but results in a single jet of blood, which can
cause damage to the aortic wall. The straight nature of the cannula means that the flow direction
is reliant on the surgical placement. In addition to direct placement in the aorta these cannulae
can be used for peripheral (e.g. femoral or axillary) arterial cannulation or within a graft.
Right-angled cannulae have been designed to allow the blood flow jet to be directed
around the aortic arch, assuming correct placement. Right-angled “diffusion” cannulae, with
54 Cardiopulmonary Bypass, ed. S. Ghosh, F. Falter and D. J. Cook. Published by Cambridge University Press.
© Cambridge University Press 2009.
Chapter 5: Conduct of cardiopulmonary bypass

diffusion holes and a sealed end, may attenuate the damaging jet effect by changing the flow
characteristics into the aorta. However, concern has been expressed regarding increased
hemolysis due to the more turbulent flow effect through the cannula. These cannulae are not
suitable for femoral placement.

Connection to patient
Usually arterial inflow is directed into the ascending aorta. The advantages of this site are:
• ease;
• safety;
• single incision;
• size of cannula not usually limited by vessel diameter; and
• no risk of limb ischemia.
The site for cannulation in the ascending aorta is traditionally determined by intraopera-
tive palpation for calcific atherosclerotic plaques; however, newer techniques such as trans-
esophageal echocardiography (TOE) or epivascular ultrasonic scanning are now being used
to determine plaque-free areas for cannulation. If significant atherosclerosis is present such
that aortic cannulation and cross-clamping is deemed unsafe, because of the risk of stroke
due to dislodgement and embolization of atherosclerotic material, femoral arterial cannula-
tion should be considered. It must be noted that retrograde perfusion via femoral arterial
cannulation is not without risk of embolization of atheroma and in such instances subclavian
or innominate arterial cannulation may be preferable. In the event of a totally calcified “por-
celain aorta,” alternative strategies that minimize aortic handling such as OPCAB surgery or
the use of DHCA may be appropriate.
Prior to insertion of the aortic cannula, the chosen site is prepared with placement of
opposing purse-string sutures and clearance of the adventitial tissue within the boundaries
of these sutures. With the mean arterial pressure controlled at between 70 and 80 mmHg, to
avoid excessive bleeding or trauma to the aorta, particularly dissection, a full-thickness inci-
sion is made in the aortic wall through which the aortic cannula is passed. Only 1“2 cm of the
cannula tip is advanced and directed towards the arch to avoid inadvertent cannulation of the
head and neck vessels or dissection of the posterior wall of the aorta. The aortic cannula is
immediately de-aired by allowing blood to fill the tubing, which is then clamped and secured
with the purse-string sutures, prior to connecting to the arterial inflow circuitry of the CPB
machine. During connection to the circuit it is essential to ensure that no air is present at the
connection site. When the connection is complete the perfusionist will inform the surgeon of
the “swing” on the arterial pressure line and the pressure within the system to confirm correct
intraluminal placement of the cannula.

Complications of aortic root cannulation
If air is introduced into the aortic line during aortic cannulation, the line must be discon-
nected from the aortic cannula and the air aspirated prior to reconnection. If gross air
embolism is noted in the aortic line during established CPB, it may be possible for the
perfusionist to remove the air via recirculation lines in the CPB circuit with only a brief
interruption to pump flow. If gross systemic air embolism occurs, de-airing of the cer-
ebral circulation must be attempted. With the patient in the Trendelenberg position CPB is
terminated followed by removal of the arterial cannula from the aorta, leaving the purse-
string sutures loose. The arterial line is filled and then inserted into the SVC. Retrograde
Chapter 5: Conduct of cardiopulmonary bypass

perfusion to the cerebral circulation via the SVC using low flow rates (1“2 l/minute), at a
blood temperature of 20°C, enables de-airing of the cerebral circulation back to the aorta.
During this de-airing process the perfusionist should re-prime the circuit, followed by the
surgeon re-cannulating the ascending aorta and recommencement of CPB at 28°C until
surgery is completed. CPB should be discontinued at a core temperature of 35°C. The use of
relative hypothermia increases the solubility of gaseous emboli and may reduce the extent
of cerebral injury.
Further potential complications of aortic root cannulation are summarized in Table 5.1.

Peripheral arterial cannulation
Indications for the use of peripheral arterial cannulation not only include aortic aneurysm or
an aorta that is not suitable for cannulation due to calcification, but also to establish CPB in
anticipation of complications arising from redo-sternotomy. Increasingly, peripheral cannu-
lation is also being used to enable “limited access,” minimally invasive surgery. Femoral can-
nulation renders it necessary to use smaller size cannulae, with consequent higher pressure

Table 5.1. Complications of aortic root cannulation

Inability to introduce the cannula
• adventitia occluding the incision site
• inadequate incision size
• atheromatous plaque within aortic wall
Intramural placement
Embolization of atheromatous plaque
Air embolization on connection to the circuit
Persistent bleeding around cannula
Malposition of tip towards aortic valve or into arch vessels
Dissection of aorta
Kink in circuit
Inadequate size leading to high pressure and low flow generation
Aneurysm formation at site of cannulation at later stage

Table 5.2. Complications of peripheral cannulation

Trauma to vessel
Retrograde arterial dissection with retroperitoneal hemorrhage or extension of dissection to aortic root
Thrombosis or embolism
Limb ischemia (can be reduced by using an end to side polytetrafluoroethane (PTFE) graft sutured to the vessel)
Malperfusion of cerebral and systemic circulation as a result of cannulation of the false lumen of an aortic
Lymph fistula or lymphocele
Late vascular stenosis

Chapter 5: Conduct of cardiopulmonary bypass

gradients, jet effects and possibly lower flow rates; this may be improved by cannulation of
iliac arteries.
Axillary cannulation is usually employed in cases of ascending aortic dissection, to avoid
the risk of inadvertent retrograde perfusion via the false lumen of the dissection, which can
occur with femoral cannulation in these patients. The axillary artery is less likely than the
femoral artery to have atherosclerotic disease or dissection and also has a good collateral flow,
with less risk of limb ischemia. In addition to these benefits it provides antegrade flow, with
reduced risk of cerebral embolization. Direct arterial cannulation or indirect cannulation via
a side graft can be used to access the axillary artery; side graft placement and cannulation is
usually preferable to direct cannulation.
The potential complications of peripheral cannulation are summarized in Table 5.2.

Venous cannulation and drainage
Venous blood inflow to the CPB circuit is usually achieved by gravity drainage, using
the “siphon” effect, but earlier CPB circuits used suction to aid venous drainage; in pediatric
cases, drainage is still often aided by applying suction to the venous lines. Gravity siphoning
as the means of obtaining adequate drainage relies on:
(1) no air being present in the tubing between the patient and the pump, otherwise an
“air-lock” develops and drainage stops; and
(2) the venous reservoir being kept below the level of the patient™s thorax.
The degree of venous drainage is determined by the patient™s central venous pressure
(CVP), the difference in height between the patient and the top of the blood level in the venous
reservoir and resistance exerted by the circuit (cannulae, lines and connectors). The CVP is
influenced by intravascular volume and venous compliance; which is in turn influenced by
sympathetic tone. This is largely dependent on the extent of inflammatory response to CPB
and by drugs used perioperatively.
Excessive drainage may cause the veins to collapse around the cannulae with intermittent
reduction in venous drainage and the potential for generation of gaseous emboli in the circuit;
a phenomenon referred to as “cavitation.”

Types and sizes of venous cannulae (see Chapter 1)
The cannula tip is the narrowest component in the venous circuit and therefore the limiting
factor for venous drainage. The appropriate size is selected based on the flow characteristics of
the cannula (detailed in the manufacturer™s guidelines) and the required flow for the patient
based on cardiac index. One third of total flow is derived from SVC drainage and two-thirds
from IVC drainage.


. 2
( 8)