PRO: Cardiac surgery without use of CPB does not require 'full' dose heparinization
Bruce D. Spiess, MD, FAHA
Professor and Vice Chair
Department of Anesthesiology
Director, VCURES Shock Center
Virginia Commonwealth Univ. Med. Center
Cardiopulmonary bypass (CPB) is over 50 years old. Off-pump heart
surgery is older. The last 5 years have seen a growth and development
of off-pump CABG that is exciting and revolutionary. Debate continues
as to outcome data from CPB CABG versus off-pump CABG.1-3 Blood
utilization is decreased in off-pump CABG and there could be some data
to suggest that neurologic outcome is different.3-5 However, data on
perioperative MI rate and graft patency are as yet either unstudied or
unable to demonstrate differences. Kaplan-Meier curves for survival
are also not yet available and randomized controlled trials are either
just now being conducted or what is available is retrospective with
largely uncomparable groups.
Unfractionated heparin (UFH) has gone hand in hand with the
development of CPB and for those 50 plus years. UFH has been the only
viable anticoagulant for use. The "best" heparin dose or activated
clotting time (ACT) for conduct of CPB is still in debate. What is
the standard of practice in one institution would be considered a
violation of protocol in another. This little discussion will
therefore not settle the question of safety for any given dose of UFH
in off-pump CABG when for 50 years we cannot agree on the science of
what is "best" for on pump CABG. That being said, there are some
points to be made that might provide assurance for the practitioner
that something less than "full dose" heparin is acceptable for
UFH is an imperfect or bad drug that actually contributes to or
sets up many of the coagulopathies and inflammatory side effects of
CPB.6-10 To date, we have never been able to understand the extent of
that statement because CPB could not be conducted with any other
agent. UFH anticoagulation function is dependent upon the
concentration and activity of antithrombin-III (AT-III). AT-III is a
circulating zymogen that when bound to UFH changes its binding
coefficient for plasma thrombin by 10-100 fold. The triple complex of
AT-III, UFH and thrombin then deactivates plasma thrombin.6
Unfortunately, for the line of CPB thinking, thrombin almost never
exists in plasma. Contemporary understanding of coagulation holds that
thrombin is generated on the surface of endothelial cells (from tissue
factor release), on the surface of platelets and on leukocytes.11 It
is the thrombin burst present in the neo-tissue of a platelet plug
that leads to local critical thrombin concentration, allowing fibrin
generation and stabilization of a clot. It therefore is the reactivity
of platelets to endothelial cells that secondarily leads to thrombin
burst. UFH does not attach to thrombin held at the active sites on
cells and UFH does not inactivate clot bound thrombin. Once a clot has
begun to form, the ACT has been triggered or the aPTT has signaled
clot generation, 96% of thrombin generation is yet to still occur, and
UFH does not block all of that thrombin activity. So, UFH is
critically ineffective at eliminating thrombin's action in the body.
Only with supra-pharmacologic dosages of UFH can one even begin to see
UFH attaching to clot bound thrombin. Interestingly, even though UFH
does not inactivate clot bound thrombin it actually forms a linkage
from the "backside" of thrombin to fibrin making the clot bound
thrombin held tighter in the clot matrix.
UFH drives down AT-III concentrations and AT-III is a natural
buffer to complement, bradykinin, and all of the serine protease
cascade proteins. For some patients AT-III levels on CPB rival those
found in severe septic shock, eclampsia and DIC.6 UFH partially
activates platelets and actually makes platelets stickier.7-11 Even in
very low concentrations it leads to the externalization of platelet
ligands: GP-Ib, Gp-IIb/IIIa and PF-4.7-10,12,13 The platelet does not
dump its granules with this level of stimulation but it is now
vulnerable to attack and in patients with already hyperactive
platelets (patients with vascular/cardiovascular disease, diabetes
etc) the use of UFH may further potentiate the ability of platelets to
find areas of dysfunctional or thrombotic vasculature. UFH attacks
the surface of endothelial cells, liberates some heparan and releases
tissue factor pathway inhibitor, held within the glycocalyx. UFH
changes the selectin responses, sticking, rolling and diapedesis of
leukocytes. It appears from some studies that sticking and rolling of
leukocytes may be decreased, especially at very high concentrations
but the diapedesis is increased.14 The sum totals of all these "side
effects" of UFH are unknown but there is significant suggestion that
the use of UFH alone increases the risk for thrombosis.
UFH causes heparin-PF-4 antibodies to be formed and heparin induced
thrombocytopenia (HITT) is a major problem.15 Type I-HIT occurs with
an acute but modest drop in platelet count, probably due to the
expression of platelet ligands. Type II-HITT, caused by immunoglobulin
activation leads to full blown HITT in 1-3% of CPB patients but
antibodies can be found in 30-50% of CPB patients by day 2-4 after
CPB. The time when early graft thrombosis is most likely is 2-4 days
after CABG. Interestingly, the patients with the highest HITT-II
antibodies after CPB have the highest rate of graft thrombosis in a
single prospective study.16 We simply have accepted UFH as part of
CPB, and these events of thrombosis are just now being
investigated. The big fear has historically been bleeding, not
Off-pump CABG could be more analogous to percutaneous coronary
intervention (PCI) than to CPB. In PCI the levels of thrombin as well
as the "platelet stickiness" have been shown to correlate with early
graft thrombosis.17, 18 In on-pump CABG, the biggest surge of thrombin
comes after protamine administration and persists for a number of
hours.19 It is actually a larger surge than what occurs during the
pump run itself. In off-pump CABG it might be appropriate to ask,
should we give protamine at all and therefore risk the surge of
thrombin? That could be a more appropriate question than whether
approximately 30-50% more UFH provides the "sweet spot" for
The ACT utilized for PCI ranges from about 250-350 seconds. By
selection of their disease, patients for off-pump CABG are more
pro-thrombotic than the general population. The risk, to be feared,
after off-pump CABG is early coronary graft thrombosis, not a major
bleeding diathesis. Little prospective data exists in a randomized
fashion examining this issue either from on pump or off-pump
CABG. From PCI, we know that UFH anticoagulation combined with a
platelet ligand inhibitor (abciximab or others) was far superior to
simply UFH alone in terms of early and even later graft closure. From
the McSPI database, recent work has shown that early platelet
inhibition with aspirin decreased major myocardial events after
on-pump CABG.20 Are these anti-platelet agents in part countering the
pro-thrombotic or partial activation of the platelets by UFH itself?
There was nothing to suggest that more UFH improves outcome.
UFH is an imperfect drug. It requires protamine, another very
imperfect drug to reverse its action. The SCA audience does not need
more instruction on the evils of protamine. UFH-protamine complexes
particularly are bad and can cause a 10- 90% acute drop in platelet
count. If you increase UFH dose, then the protamine dose will need to
be increased. If a particular amount of a bad drug (albeit time
tested) is required for surgery there is no reason to make the
philosophical jump that more of that bad drug would be somehow better.
There is simply nothing scientifically sound about either what
constitutes "full dose" UFH, nor whether that dose actually decreases
the risk of graft dysfunction.
Today some other fascinating agents are coming on the scene- the
direct thrombin inhibitors. These agents block thrombin directly at
both of its active sites and actually attack it on cell surfaces as
well as in clot bound forms. These agents do not need the other bad
drug-protamine. Argatroban, lepirudin and bivalirudin have been used
in HITT-II.15 Bivalirudin has been extensively tested in PCI, and
therein has been shown to provide improved outcome in patients as
compared to UFH alone or UFH plus abciximab.21 A recent study of 100
off-pump CABG patients in New Zealand used bivalirudin versus UFH with
protamine reversal.22 The beauty of the New Zealand study was that the
patients underwent re-catheterization at 3 months after their off-pump
CABG. Those who had bivalirudin, and avoided UFH had a significantly
fewer graft flow reductions. So, the questions and directions for the
future, are not, is more UFH better but, rather is there a way to
replace or modify our use of UFH completely.
There are no studies that definitively can tell us what is best or
correct. For on-pump CABG there are no large prospective studies that
show an ACT of 450 versus 600 or higher are better for graft
thrombosis. Studies examining using the automated "Hepcon" system
versus routine ACT for off-pump surgery, and not surprisingly, the
automated system called for more UFH and protamine. In 8 patients
nothing could be concluded about what is "best".23 Others have trialed
whether to reverse the UFH with whole, half or no protamine.24 As
stated above, we know that the largest surge in thrombin production is
immediately after protamine. However, if we give more UFH and also do
not give protamine will we see off-pump CABG bleeding increase to the
levels seen for on-pump cases? The study with whole, half and no
protamine was inconclusive except to say that the group with low and
no protamine had fewer protamine side effects- no big surprise. One
study of 76 off-pump patients per group did examine whether those
anticoagulated with 300 IU/kg or 150 IU/kg UFH and ultimately an ACT
of >400 or >300 had more thrombosis events.25 There were more
revised grafts in the patients with a lower ACT and UFH dosage but the
assessment was subjective by the surgeons at the time of operation and
not blinded. There are no large series wherein patients have been
re-catheterized at 3 or 6 months to assess graft patency with blinded
examiners scoring the perfusion. The science has not been performed,
hence this academic debate. Is that study worth doing simply to decide
whether a bit more of UFH is good? Perhaps it is better to think
outside the box, understand the biology of the systems involved and
ask the critical question, is not 50 years of a bad drug long enough?
We should further develop testing that examines hypercoagulability
such as the thromboelastograph (which has been shown to correlate to
early thrombosis risks), the Hemodyne platelet contractile force or
others, so that we know which patients are at risk. Anti-platelet
agents during and early after surgery, perhaps even in conjunction
with aprotinin added to UFH. Fear of graft thrombosis with aprotinin
exists but where is the science? There is science to suggest
decreased thrombin generation and serine protease activity with its
use. If it were combined with a short acting anti-platelet agent, the
three actions in combination could be useful. Perhaps we should very
actively work to investigate and replace UFH itself. Off-pump CABG is
just the perfect place to replace UFH and that is being investigated
at this time. We have been arguing for close to 50 years about what
is the right ACT for CPB and we are no closer to any
agreement. Probably few patients in the United States would volunteer
for the properly organized prospective randomized study of off-pump
CABG and UFH dosing to solve the issue. But changing a dose of a bad
drug, UFH, by 30-50%, is probably not a secret key to the biologic
events that lead to graft thrombosis.
- Khan NE, et al. A randomized comparison of off-pump and on-pump multivessel coronary artery bypass surgery. N Engl J Med. 2004;350:21-8.
- Reston JT, Tregear SJ, Turkelson CM. Meta-analysi of short-term and mid-term outcomes following off-pump coronary artery bypass grafting. Ann Thorac Surg 2003;76:1510-5.
- Magee MJ, Coombs LP, Peterson ED, Mack MJ. Patient selection and current practice strategy for off-pump coronary artery bypass surgery. Circulation 2003;108:Suppl I:II9-4.
- Nuttall GA, et al. A comparison of bleeding and transfusion in patients who undergo coronary artery bypass grafting via sternotomy with and without cardiopulmonary bypass. J Cardiothorac Vasc Anesth 2003:17:447-51.
- Buceriues J, Gummert JF, Walther T, Doll N, Falk V, Schmitt DV, Mohr FW. Predictors of prolonged ICU stay after on-pump versus off-pump coronary artery bypass grafting. Intensive Care Med 2004:30:88-95.
- Spiess BD. Heparin: beyond an anticoagulant. In Spiess BD (ed). The Relationship Between Coagulation, Inflammation and Endothelium-A Pyramid Towards Outcome. SCA monograph, Lippincott, Williams and Wilkins. Baltimore, Maryland, 2000, pgs 169-190.
- Schneider DJ, Trachy PB, MannKG, Sobel BE. Differential effects of anticoagulants on the activation of platelets ex vivo. Circulation 1997:96:2877-2883.
- Xiao Z, Theroux P. Platelet activation with unfractionated heparin at therapeutic concentrations and comparisons with a low-molecular weight heparin and with a direct thrombin inhibitor. Circulation 1998;97:251-6.
- Visentin GP, Ford SE, Scott JP, Aster RH. Antibodies from patients with heparin-induced thrombocytopenia/thrombosis are specific for platelet factor 4 complexed with heparin or bound to endothelial cells. J Clin Invest 1994;93:81-8.
- Diamond MS, Alon R, Parkos CA, Quinn MT, Springer TA. Heparin is an adhesive ligand for the leukocyte integrin MAC-1 (CD11b/CD-18). J Cell Biol 1995;130:1473-1482.
- Mann KG, Butenas S, Brummel K. The dynamics of thrombin formation. Arterioscler Thromb Vasc Biol. 2003;23:17-25.
- Mascelli MA et al. Therapeutic heparin concentrations augment platelet reactivity: Implications for the pharmacologic assessment of the glycoprotein IIb/IIIa antagonist abciximab. Am Heart J 2000;139;696-703.
- Furman MI, et al. Leukocyte-platelet aggregation, platelet surface P-selectin, and platelet surface glycoprotein IIa after percutaneous coronary intervention: Effects of dalteparin or unfractionated heparin in combination with abciximab. Am Heart J 2001;142:790-8.
- Xie A, thorlacius H, Raud J, Hedqvist P, Lindbom L. Inhibitory effect of locally administered heparin on leukocyte rolling and chemoattractant-induced adhesion in rat mesenteric venules in vivo. Brit J Pharmacol 1997;122:906-910.
- Warkentin TE, Greinacher A. Heparin-induced thrombocytopenia and cardiac surgery. Ann Thorac Surg 2003;76:638-48.
- Slaughter T, Bennett-Guerrero E, Su Z, et al. Anti-heparin/PF4 antibodies detected prior to cardiac surgery identify patients at high risk for adverse perioperative outcomes.(abstract) Anesth Analg 2002;93:SCA28
- Kabbani SS, et al. Platelet reactivity characterized prospectively a determinant of outcome 90 days after percutaneous coronary intervention. Circulation 2001;104:181-186.
- Krishnaswami A. et al. Patients with coronary artery disease who present with chest pain have significantly elevated platelet contractile force and clot elastic modulus. Thromb Haemost 2002;88: 739-44.
- Chandler WL, Velan T. Estimating the rate of thrombin and fibrin generation in vivo during cardiopulmonary bypass. Blood 2003;101:4355-62.
- Mangano DT, Multicenter Study of Perioperative Ischemia Research Group. Aspirin and mortality from coronary bypass surgery. N Engl J Med. 2002;347:1309-17.
- Lincoff AM, Bittl JA, Harrington RA et al. Bivalirudin and provisional glycoprotein Iib/Iia blockade compared with heparin and planned glycoprotein Iib/Iia blockade during precutaneous coronary intervention. REPLACE-2 randomized trial. JAMA 2003;289:853-63.
- Merry AF, Raudkivid P, White HD et al. anticoagulation with bivalirudin (a direct thrombin inhibitor) vs heparin. A randomized trial in OPCAB graft surgery (abstract). Anesth Analg 2002;93:SCA33.
- Baykut D, Weichelt K, Wehrle J, Zerkowski HR, Bernet F. The value of heparin concentration monitoring in off-pump coronary bypass surgery. Eur J Med Res 2003;8:161-4.
- Gatti G, Pugliese P. Heparin reversal in off-pump coronary bypass surgery: complete, partial, or no reversal? Cardiovasc Surg 2002;10:245-50.
- Donias HW, D'Ancona G, Pande RU, Schimpf D, Kawaguchi AT, Karamanoukian HL. Heparin dose, transfusion rates, and intraoperative graft patency in minimally invasive direct coronary artery bypass. Heart Surg Forum 2003;6:176-80.