A Practical Approach to Cardiac Anesthesia (Practical Approach Series) 5th Ed.

19 Coagulation Management During and After Cardiopulmonary Bypass

Linda Shore-Lesserson, S. Nini Malayaman, Jay C. Horrow, and Glenn P. Gravlee


 1. Previous concepts of independent intrinsic and extrinsic plasma coagulation pathways, a cell-free final common pathway, and platelet clotting have given way to an integrated concept of cell-based coagulation occurring on the platelet surface as depicted in Figure 19.2.

 2. In the absence of a history suggesting a preoperative bleeding disorder (e.g., von Willebrand’s Disease, warfarin therapy), routine hemostatic screening for patients undergoing cardiac surgery is not cost-effective at predicting excessive perioperative bleeding.

 3. Unfractionated heparin (UFH) is a hydrophilic macromolecular glycosaminoglycan of varying chain lengths which anticoagulates principally by potentiating antithrombin III (AT III)-induced inactivation of factors IIa (thrombin) and Xa.

 4. Intravenously administered heparin typically peaks within 1 min, redistributes only slightly, and has a dose-related half-life that reaches approximately 2 hrs in the large doses used for cardiopulmonary bypass (CPB). Heparin bolus administration decreases systemic vascular resistance by 10% to 20%.

 5. The usual initial heparin dose is 300 to 350 units/kg to achieve an Activated Clotting Time (ACT) exceeding 400 s for safe initiation and maintenance of CPB. Heparin 5,000 to 10,000 units should be added to CPB priming solution.

 6. Heparin is most often monitored using ACT, but ACT precision is suboptimal at the heparin concentrations required for CPB, and it is also subject to prolongation from hemodilution and hypothermia. As a result, some practitioners choose to monitor and maintain a target whole blood heparin concentration (typically 3 to 4 units/mL) in addition to exceeding a target ACT during CPB.

 7. Resistance to heparin-induced anticoagulation has a variety of causes, but therapy with either AT III concentrates or fresh frozen plasma (FFP) nearly always resolves it regardless of its cause.

 8. Heparin-induced thrombocytopenia (HIT) produces a severe procoagulant state that may occur after 5 or more continuous days of heparin administration. Diagnosis requires the combination of an appropriate clinical context and a complex variety of laboratory tests.

 9. Documented presence of HIT in a patient who must urgently undergo cardiac surgery requiring CPB is possibly best managed by using bivalirudin for anticoagulation.

10. Protamine neutralizes heparin stoichiometrically as a result of a strong cation (protamine)-to-anion (heparin) interaction.

11. A variety of protamine dosing methods are used clinically. Most often 60 to 80 mg of protamine per 100 units of heparin administered prior to CPB (including heparin placed in the CPB priming solution if residual CPB volume is to be reinfused unaltered by hemoconcentration or cell washing) suffices to neutralize heparin. Excessive protamine administration impairs blood clotting.

12. In order to avoid hypotension, protamine should be administered slowly, ideally by a continuous intravenous infusion over 5 to 10 min.

13. Protamine can induce severe anaphylactic or anaphylactoid reactions.

14. Post-CPB clotting abnormalities are best managed using a systematic approach. Post-CPB bleeding algorithms reduce bleeding and transfusion.

IN ESSENCE, CPB CREATES a blood “detour” to permit surgery on the heart. This detour must route the blood through an artificial heart and lung while maintaining its fluidity. Historically, fluidity represented the final frontier in the development of cardiac surgery because effective mechanisms for blood gas exchange and for propelling the blood had been established more than a decade before surmounting the fluidity challenge. The challenge was to find a therapeutic approach that would inhibit blood’s natural propensity to clot when it contacts foreign surfaces. Since the restoration of normal coagulation was desirable at the end of the surgical procedure, this clotting inhibition needed to be reversible, like turning a spigot on and off. The long-awaited solution was monumental: anticoagulation with heparin followed by neutralization with protamine. This fundamental approach to establishing and reversing blood fluidity remains unchanged after almost 50 yrs, although much fine-tuning has occurred. This chapter reviews anticoagulation and the restoration of coagulation in the patients undergoing CPB.

I. Physiology of Coagulation

   A. Mechanisms of hemostasis

      1. Plasma coagulation. Figure 19.1 depicts the plasma coagulation pathway. Blood contact with foreign surfaces classically was thought to activate the intrinsic pathway, whereas vascular injury or disruption was thought to activate the extrinsic pathway. These definitions seem counterintuitive because vascular disruption should be intrinsic and foreign bodies extrinsic, but logic has held little sway in coagulation pathway nomenclature. Thankfully, distinctions between the intrinsic and extrinsic pathways have become less important because both the activants and the pathways overlap (e.g., connection between VIIa and IXa). The coagulation factor numbering system was defined in order of discovery rather than use, which explains the illogic numerical progression through the pathways [1].

Figure 19.1 Schematic representation of the hemostatic system depicting the vascular, platelet, and coagulation components. F, factor; HMWK, high-molecular weight kininogen; vWF, von Willebrand’s factor; Ca++; ionized calcium; VIII:C, factor VIII coagulation component; TxA2, thromboxane A2; ADP, adenosine diphosphate.

        a. Intrinsic pathway. Contact activation involves binding of factor XII to negatively charged surfaces, which leads to the common pathway through factors XI, IX, platelet factor 3, cofactor VIII, and calcium. Kallikrein is also formed in this reaction and serves as a positive feedback mechanism and as an initiator of fibrinolysis (a negative feedback mechanism) and inflammation. For cardiac surgery, this pathway’s clinical importance lies more in the access it provides for heparin monitoring and neutralization than in its role in normal hemostasis.

        b. Extrinsic pathway. Tissue factor (TF) initiates the extrinsic pathway, which proceeds to activate factor IX and rapidly stimulates the common pathway with the aid of factor VII and calcium.

        c. Common pathway. Beginning with the assisted activation of factor X, this pathway proceeds to convert prothrombin (factor II) to thrombin and fibrinogen (factor I) to fibrin monomer, which initiates the actual substance of the clot. Fibrin monomer then cross-links to form a more stable clot with the aid of calcium and factor XIII. Rather than thinking of the common pathway as the result of activation of two independent paths, the concept of cell-based coagulation has been adapted to better explain the hemostatic mechanisms that occur simultaneously in the body (Fig. 19.2).When tissue injury occurs, TF is expressed on the surface of TF-bearing cells. Presentation of TF to its ligand factor VII causes the activation of factors IX and X on the TF-bearing cell. This is called “Initiation.” The activation of factor X causes thrombin formation which then incites further protease activation. These reactions occur on the phospholipid surface of the platelet and are called “Amplification.” The final stage of clot formation is known as “Propagation.”

Figure 19.2 Cell-based model of hemostasis. Cellular hemostasis is thought to occur in three stages: Initiation, amplification, and propagation. The initiation stage (1) takes place on TF-bearing cells (cells such as monocytes that can bind TF and present it to a ligand), which come into play when endothelial injury occurs and TF is exposed. The initiation stage is characterized by presentation of TF to its ligand factor VII and the subsequent activation of factors IX and X on the TF-bearing cell. The activation of factor X to Xa causes thrombin production and activation. Once generated, thrombin feeds back to activate factors VIII, V, and platelets. The amplification stage (2) then occurs on the surface of the activated platelet, which exposes surface phospholipids that act as receptors for the activated factors VIIIa and IXa. The platelet surface allows for further thrombin formation and hence the amplification of coagulation. Continued generation and activation of thrombin causes further positive feedback mechanisms (3) to occur that ultimately ensure the formation of a stable clot, including cleavage of fibrinogen to fibrin, release and activation of factor XIII for fibrin cross-linkage, and the release of a thrombin-activatable fibrinolysis inhibitor.

            Activated factor X also initiates platelet-surface clotting activity as depicted in Figure 19.2. This platelet-surface activity greatly speeds the overall formation of clot and should be considered as a vital component of the clotting cascade.

        d. Thrombin is the most important enzyme in the pathway because (in addition to activating fibrinogen) it

           (1) Provides positive feedback by activating cofactors V and VIII

           (2) Accelerates cross-linking of fibrinogen by activating factor XIII

           (3) Strongly stimulates platelet adhesion and aggregation

           (4) Facilitates clot resorption by releasing tissue plasminogen activator (tPA) from endothelial cells

           (5) Activates protein C, which provides negative feedback by inactivating factors Va and VIIIa

      2. Platelet activation. As shown in Figure 19.1, a variety of stimuli initiate platelet activation, and thrombin is an especially potent one. This sets off a cascade of events that initiates platelet adhesion to endogenous or extracorporeal surfaces, followed by platelet aggregation and formation of a primary platelet plug. Fibrin clots and platelet plugs form simultaneously and mesh together, yielding a product more tenacious and difficult to dissolve than either alone. Whereas formerly plasma-based and platelet-based clotting were often thought to represent independent pathways, more recent accounts show that the plasma coagulation pathway evolves mainly on the platelet surface, such that the plasma and platelet clotting processes are more interdependent than independent.

        a. Von Willebrand factor (vWF) is an important ligand for platelet adhesion at low shear stress, and for aggregation at higher shear rates. Fibrinogen is the major ligand for platelet aggregation.

        b. Products released from platelet storage granules (adenosine diphosphate [ADP], epinephrine, calcium, thromboxane A2, factor V, and vWF) serve to perpetuate platelet activation and the plasma coagulation cascade.

      3. Nature’s system of checks and balances demands counterbalancing forces to discourage runaway clot formation and to dissolve clots. These counterbalances include the following:

        a. Proteins C and S, which inactivate factors Va and VIIIa

        b. ATIII, antithrombin, or AT, which inhibits thrombin as well as factors XIa, IXa, XIIa, and Xa

        c. TF inhibitor, which inhibits the initiation of the extrinsic pathway

        d. tPA, which is released from endothelium and converts plasminogen to plasmin, which in turn breaks down fibrin. Plasminogen activator inhibitor 1 in turn inhibits tPA to prevent uncontrolled fibrinolysis.

   B. Tests of hemostatic function

      1. Table 19.1 lists commonly used tests of hemostatic function [1]. These tests may be used to detect hemostatic abnormalities preoperatively or after CPB. With the exception of the ACT, typically they are not used during CPB except under extenuating circumstances, because most of them will be abnormal as a result of hemodilution, anticoagulation, and sometimes hypothermia.


        a. Most studies suggest that routine preoperative hemostatic screening is not helpful in predicting patients who will bleed excessively during surgery. If the patient’s clinical history (e.g., nosebleeds; prolonged bleeding with small cuts, dental work, or surgery; easy bruising; strong family history of pathologic bleeding) suggests the need for hemostatic screening, selective use of these and other tests is appropriate. Similarly, when the patient is taking medications that alter hemostatic function, specific hemostatic function tests may be indicated. Examples include the following:

Table 19.1 Common clinical tests of hemostatic function

           (1) Heparin: Activated partial thromboplastin time (aPTT) or ACT

           (2) Low-molecular-weight heparin (LMWH), including the pentasaccharide fondaparinux: No test or anti-Xa plasma activity

           (3) Warfarin: Prothrombin time (PT) and/or international normalized ratio (INR)

           (4) Platelet inhibitors including aspirin: No testing, bleeding time, or specific platelet function tests. Preliminary data suggests that specific platelet function tests in patients taking thienopyridine agents may correlate with postoperative bleeding risk.

II. Heparin anticoagulation

   A. Heparin pharmacology [2]

Structure. As drugs go, UFH might be described as impure. Heparin resides physiologically in mast cells, and it is commercially derived most often from the lungs of cattle (bovine lung heparin) or the intestines of pigs (porcine mucosal heparin). Commercial preparations used for CPB typically include a range of molecular weights from 3,000 to 40,000 Da, with a mean molecular weight of approximately 15,000 Da. Each molecule is a heavily sulfated glycosaminoglycan polymer, so heparin is a strong biologic acid that is negatively charged at physiologic pH.

     a. Porcine mucosal and bovine lung heparin both are satisfactory for CPB and both have been widely used.


      1. Action. A specific pentasaccharide sequence that binds to ATIII is present on approximately 30% of heparin molecules. This binding potentiates the action of ATIII more than 1,000-fold, thereby allowing heparin to inhibit thrombin and factor Xa most importantly, but also factors IXa, XIa, and XIIa.

        a. Inhibition of thrombin requires simultaneous binding of heparin to both ATIII and thrombin, whereas inhibition of factor Xa requires only that heparin binds to ATIII. The former reaction limits thrombin inhibition to longer saccharide chains (18 or more saccharide units); hence, shorter chains can selectively inhibit Xa. This is the primary principle underlying therapy with LMWH and with the “ultimate” LMWH fondaparinux, which contains just the critical pentasaccharide sequence needed for binding ATIII, hence it induces virtually exclusive Xa inhibition.

           (1) Because thrombin inhibition appears pivotal for CPB anticoagulation and also because LMWH and heparinoids have a long half-life and are poorly neutralized by protamine, LMWH (including fondaparinux) is inadvisable as a CPB anticoagulant. Fondaparinux lasts even longer than traditional LMWHs and is even less neutralized by protamine.

        b. Heparin binds and activates cofactor II, a non–ATIII-dependent thrombin inhibitor. This may explain why heparin-induced anticoagulation can be effective even in the presence of marked ATIII deficiency, although the primary mechanism of anticoagulant action is ATIII inhibition.

      2. Potency. Heparin potency is tested by measuring the anticoagulation effect in animal plasma. The United States Pharmacopoeia (USP) defines 1 unit of activity as the amount of heparin that maintains the fluidity of 1 mL of citrated sheep plasma for 1 hr after recalcification.

        a. Heparin dosing is best recorded in USP units, because commercial preparations vary in the number of USP units per milligram. The most common concentration is 100 units/mg (1,000 units/mL) [3].


      3. Pharmacokinetics. After central venous administration, heparin’s effect peaks within 1 min, and there is a small rapid redistribution that most often is clinically insignificant [4,5].

        a. Heparin’s large molecular size and its polarity restrict its distribution mainly to the intravascular space and endothelial cells.

        b. The onset of CPB increases circulating blood volume by approximately 1,000 to 1,500 mL; hence, plasma heparin concentration drops proportionately with the onset of CPB unless heparin is added to the CPB priming solution.

        c. Heparin is eliminated by the kidneys or by metabolism in the reticuloendothelial system.

        d. Elimination half-life has been determined only by bioassay, that is, by the time course of clotting time prolongation. By this standard, heparin’s elimination time is dose dependent [6]. At lower doses, such as 100 to 150 USP units/kg, elimination half-time is approximately 1 hr. At CPB doses of 300 to 400 USP units/kg, elimination half-time is 2 or more hours; hence, clinically significant anticoagulation might persist for 4 to 6 hrs in the absence of neutralization by protamine. Hypothermia and probably CPB itself prolong elimination.

      4. Side effects. Heparin’s actions on the hemostatic system extend beyond its primary anticoagulant mechanism to include activation of tPA, platelet activation, and enhancement of TF pathway inhibitor.

        a. Lipoprotein lipase activation influences plasma lipid concentrations, which indirectly affects the plasma concentrations of lipid-soluble drugs.

        b. Heparin boluses decrease systemic vascular resistance. Typically this effect is small (10% to 20%), but rarely it can be more impressive and may merit treatment with a vasopressor or calcium chloride.

        c. Anaphylaxis rarely occurs.

        d. HIT is covered elsewhere in this chapter.


   B. Dosing and monitoring

      1. Dosing. The most common initial dose for CPB is 300 to 350 USP U/kg. Some centers choose an initial dose of 400 U/kg or base the initial dose on a bedside ex vivo heparin dose-response titration.

        a. Since heparin distributes primarily into the plasma compartment, increasing the dose with increasing body weight assumes that plasma volume increases in direct proportion to body weight. This is not the case, because fat does not increase blood volume in proportion to weight. Consequently, there is seldom reason to exceed an initial dose of 35,000 to 40,000 units, even in patients weighing more than 100 kg, as lean body mass tends to peak at 90 kg for females and 110 kg for males.

        b. Heparin dosing for coronary revascularization procedures performed without CPB is controversial. Published doses range from 100 to 300 U/kg, but most centers use 100 to 150 units/kg and set minimum acceptable ACT values at 200 to 300 s.

        c. The CPB priming solution should contain heparin at approximately the same concentration as that of the patient’s bloodstream at the onset of CPB. Since this most often would be 3 to 4 U/mL, a priming volume of 1,500 mL should contain at least 5,000 units of heparin. CPB priming solutions commonly contain 5,000 to 10,000 units of heparin.

        d. Supplemental heparin doses typically are guided by monitoring of anticoagulation.

      2. Monitoring. Until the late 1970s, heparin dosing was guided by experiential practices and varied a great deal from hospital to hospital. Using ACT, a variation on the Lee–White clotting time, Bull et al. [7] identified rather staggering variations in the approach to heparin dosing and in both the initial anticoagulant response and the time course of anticoagulation in response to a fixed dose of heparin.This landmark work rapidly led to the realization that the anticoagulant response to heparin should be monitored, although a few centers continue to dose heparin empirically.

      3. Approaches to anticoagulation monitoring for CPB. The ACT is the most widely used test, although some centers monitor blood heparin concentration as well.

        a. ACT uses an activant such as celite or kaolin to activate clotting, then measures the clotting time in a test tube. Heparin prolongs ACT with a roughly linear dose-response pattern (Fig. 19.3). Normal ACT depends upon such factors as the specific activant and device, prewarming (vs. room temperature) of test tubes, and operator technique, but generally falls between 110 and 140 s.

Figure 19.3 Graph of a heparin (and protamine) dosing algorithm. In the graph, the control ACT is shown as point A and the ACT resulting from an initial heparin bolus of 200 units/kg is shown in point B. The line connecting A and B then is extrapolated and a desired ACT is selected. Point C represents the intersection between this line and a target ACT of 400 s, theoretically requiring an additional heparin dose represented by the difference between points C and B on the horizontal axis (arrow C). Similarly, to achieve an ACT of 480 s (higher horizontal dotted line intersecting the ACT versus heparin dose line at point D), one would administer the additional heparin dose represented by arrow D. To estimate heparin concentration and calculate protamine dose at the time of heparin neutralization, the most recently measured ACT value is plotted on the dose-response line (point E in the example). The whole blood heparin concentration present theoretically is represented by the difference between point E and point A on the horizontal axis (arrow E). The protamine dose required to neutralize the remaining heparin then may be calculated. Protamine 1.0 mg/kg is administered for every 100 units/kg of heparin present. (Modified from Bull BS, Huse WM, Brauer FS, et al. Heparin therapy during extracorporeal circulation: II. The use of a dose-response curve to individualize heparin and protamine dosage. J Thorac Cardiovasc Surg. 1975;69:686; and Gravlee GP. Anticoagulation for cardiopulmonary bypass. In: Gravlee GP, Davis RF, Kurusz M, et al., eds. Cardiopulmonary Bypass: Principles and Practice, 2nd ed. Philadelphia, PA: Lippincott Williams & Wilkins; 2000:435–472, with permission.)

            Although originally described as a manual test, most centers use one of the two automated approaches to ACT (International Technidyne, Edison, NJ, U.S.A., or Medtronic HemoTec, Fridley, MN, U.S.A.). These two automated approaches yield slightly different values both at baseline and with anticoagulation because of different activators and endpoint detection techniques.

            ACT is prolonged by hypothermia and hemodilution; hence, conditions often imposed by CPB alter the ACT-heparin dose-response relationship [8]. Some see this as risking underanticoagulation, although hemodilution and hypothermia legitimately enhance anticoagulation. Overreliance on hypothermic enhancement of ACT prolongation risks underanticoagulation upon rewarming. Also, at temperatures below 25°C, ACT prolongation becomes so profound that alternative tests such as whole blood heparin concentration measurement may be advisable.

            The target ACT level for CPB is controversial. There are studies supporting the safety of ACT values as low as 300 s, inside and outside of the context of heparin-coated surfaces, yet most centers accept only values exceeding 400 to 480 s. In addition, different devices and tests yield different dose-response relationships for heparin concentration versus ACT, as well as different sensitivities to hypothermia and hemodilution.


           (1) As a clotting test, ACT is somewhat crude and may vary as much as 10% on repeated testing at heparin concentrations used for CPB [9], so it seems reasonable to build in a safety margin by accepting 400 s as a minimum safe threshold for sustained CPB.

           (2) Aprotinin, which is now used rarely in clinical practice, complicates the use of ACT monitoring, as a result of marked prolongation of celite ACTs in the presence of heparin and aprotinin. This may represent enhanced anticoagulation to some degree, but in the presence of aprotinin it is probably wise to titrate heparin to a celite ACT level exceeding 750 s or to use a kaolin ACT instead. Kaolin ACT minimum levels do not have to be adjusted in the presence of aprotinin.

        b. Whole blood heparin concentrations can be measured during CPB. The most commonly used technique is automated protamine titration (Medtronic HemoTec, Fridley, MN). Advocates of this monitoring technique argue that CPB-induced distortion of the ACT-heparin dose-response relationship mandates maintenance of the heparin concentration originally needed before CPB to achieve the target ACT level [10]. Heparin dosing based upon concentration alone substantially increases the amount of heparin given during CPB, which enhances suppression of thrombin formation. This benefit may accrue at the expense of heparin rebound, if not monitored, and more profound platelet activation that may aggravate and prolong platelet dysfunction after CPB. The distortion of the ACT-heparin sensitivity relationship can be partly overcome using plasma-modified ACT testing [11] or by maximal activation of the ACT test sample, as is done in a Thromboelastograph (TEG) modification of the ACT [12]. Whole blood point-of-care measurement of heparin concentration can also be performed with the HepTest POC-Hi (Americana Diagnostica, Stamford, CT). This test correlates well with heparin concentrations during CPB and more closely approximates the trend in plasma anti-Xa levels than does the standard ACT [13].

           (1) Whole blood heparin concentrations of 3.0 to 4.0 U/mL most often are sufficient for CPB. Plasma heparin concentrations (typically anti-Xa concentrations) are higher because circulating heparin resides in the plasma compartment.

           (2) Patients can vary widely in their sensitivity to heparin-induced anticoagulation; therefore, isolated use of heparin concentration could lead to dangerous underanticoagulation. If this technique is chosen, simultaneous use of ACT or another clotting time is strongly advised.

           (3) Heparin concentration monitoring may be a useful adjunct to ACT monitoring both in the presence of aprotinin and during periods of hypothermia below 25°C.

           (4) Heparin concentration monitoring may be advantageous in selecting a protamine dose, because the dose will be chosen in relation to approximate actual blood heparin concentration. The weakness of this technique is its dependence on a calculated blood volume determination at a time when blood volume may vary substantially.

        c. Other monitors of anticoagulation. Neither ACT nor heparin concentration is perfect, so other tests have been evaluated or are under investigation. The aPTT and traditional thrombin time are typically so sensitive to heparin that those tests are not useful at the heparin concentrations needed for CPB. The high-dose thrombin time (HiTT, International Technidyne) offers a linear dose-response in the usual heparin concentration range used for CPB. The heparin management test (HMT, Helena, Beaumont, TX) offers another platform for ACT monitoring and can be used to monitor heparin at high (cardiac surgery) and low (vascular surgery) concentration ranges. Clotting times have also been successfully monitored using viscoelastic tests—the Sonoclot (SkACT, Sienco, Arvada, CO) [14] and the TEG.

   C. Heparin resistance. Heparin resistance is loosely defined as the need for greater than expected heparin doses to achieve the target ACT for CPB. As noted earlier, ACT prolongation in response to heparin varies greatly. A number of factors that may decrease the ACT response to heparin are listed in Table 19.2 [2]. ATIII deficiency is cited most often for heparin resistance, but overall the correlation between ATIII concentrations and anticoagulant response to a bolus of heparin has been weak and inconsistent, perhaps because heparin resistance often is multifactorial.

   Clinical approach. Most often heparin resistance can be managed by simply giving more heparin. It is reasonable to consider administering supplemental ATIII if more than 600 USP units/kg has been required to reach the target ACT level. In some circumstances (e.g., urgent need to initiate CPB, minimal ACT prolongation after 300 U/kg), a lower intervention threshold seems reasonable. Similarly, if maintaining the target ACT during CPB requires administering more than 100 USP units/kg per 30 min of CPB, supplemental ATIII seems reasonable.


        a. ATIII can be provided in the form of FFP, liquid plasma, and ATIII concentrates (human or recombinant).

           (1) FFP or liquid plasma dose typically is 2 to 4 units in adults.

           (2) For ATIII concentrates, an initial dose of 500–1,000 units is usually sufficient.

      1. Prediction of heparin resistance. Medtronic HemoTec’s HMS Plus Hemostasis Management System and International Technidyne’s Hemochron RxDx system each provide an in vitro opportunity to titrate the patient’s whole blood to predetermined heparin concentrations and hence to predict the initial heparin dose required to achieve a particular target ACT. Some centers use one of these devices to predict heparin resistance prior to heparin administration. This approach allows “customization” of initial heparin dosing as well as advance preparation for anticipated heparin resistance, for example, ordering ATIII concentrate or FFP.

   D. Heparin-induced thrombocytopenia (HIT)

     1. Benign Thrombocytopenia from heparin’s proaggregatory effect on platelets develops in 5% to 28% of patients.

        a. This mild decrease in platelet count typically occurs within 1 to 2 days of heparin administration without thrombosis or immune response. This response was formerly called HIT type I, and it is not considered pathologic. However, thrombocytopenia occurring within 1 to 2 days of heparin administration can be pathologic immune-mediated HIT if the patient’s plasma retains heparin antibodies from a previous exposure. This latter situation is most likely to occur in a patient who has experienced clinical HIT within the past several weeks.

        b. HIT is a severe condition that most often occurs after 5 continuous days of heparin administration (average onset time, 9 days), and is immune mediated. Antibody binding to the complex formed between heparin and platelet factor 4 (PF4) is responsible for the syndrome. Antibody binding to platelets and endothelial cells causes platelet activation, endothelial injury, and complement activation. The platelet clots formed are referred to, in lay terms, as “white clots.” This syndrome is highly morbid and can be fatal as a result of thromboembolic phenomena.


           (1) Among patients who develop HIT II, the incidence of thrombosis approximates 20%, the mortality of which can be as high as 40%.

           (2) Diagnosis: thrombocytopenia (usually defined as platelet count <100,000/uL, but this is complicated for several days by post-CPB hemodilution), demonstration of heparin/PF4 antibody plus (ideally) documentation of heparin-induced aggregation of platelets.

             (a) Heparin-induced serotonin release assay: a functional test, often considered the gold standard.

             (b) Heparin-induced platelet activation assay (HIPAA): a functional test, may be nonspecific.

Table 19.2 Potential causes of heparin resistance

             (c) Enzyme-linked immunosorbent assay (ELISA) specific for the heparin/PF4 complex or for PF4 alone: patients with a positive antibody test do not always develop thrombosis. Antibodies to the heparin/PF4 complex are associated with adverse outcomes after cardiac surgery [15]. Whether the antibody is causative of outcome or merely a marker for a more critically ill patient has not yet been determined. Antibodies associated with HIT often become undetectable 50 to 85 days after discontinuing heparin. After 100 days of heparin discontinuation, one can be confident that short-term re-exposure to heparin (as for CPB) will not result in antibody formation [16]. Continued heparin re-exposure is not recommended even though the clinical syndrome does not always recur upon reexposure to the drug. Sometimes the syndrome resolves despite continued heparin therapy. HIT can be associated with heparin resistance and thus should be part of the differential diagnosis.

           (3) Treatments for HIT and alternative anticoagulant sources

             (a) In HIT, changing tissue source of heparin, LMWH, heparinoids, and ancrod have been used but are no longer recommended. Cross-reactivity with other heparins does exist (see Section E).

             (b) Plasmapheresis can be used to remove the antibody, but may be insufficient therapy by itself. Heparin plus platelet inhibitors can be combined to reduce the aggregability of platelets. Hirudin, bivalirudin, and argatroban are direct thrombin inhibitors.

   E. Alternatives to unfractionated heparin

      1. LMWH (shorter-chain heparin molecules, including fondaparinux). 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 in 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 monitoring because aPTT (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 directly in plasma and with a plasma-modified whole blood test, but there is no simple point-of-care test available. LMWHs are not recommended for use in HIT patients because of cross-reactivity of the antibody, although some reports suggest that fondaparinux may be safe because it is too small to bind antibodies.

      2. Heparinoids. Two glycosaminoglycans with heparin-like properties may be available for clinical use: Dermatan sulfate and danaparoid. They are referred to as heparinoids because they represent a class of synthetic or naturally occurring heparin analogs. Danaparoid is a natural composite of heparan sulfate (80%), dermatan sulfate (20%), and chondroitin sulfates. Case reports document that they have been used successfully in CPB when heparin use was contraindicated, but cross-reactivity and bleeding complications preclude their current use in HIT patients [17]. Since 2002, Danaparoid has not been available in the United States but is still available in some other countries.

      3. Hirudin [18]

        a. Thrombin inhibitor isolated from the salivary glands of the medicinal leech (Hirudo medicinalis).

        b. Independent of ATIII and inhibits clot-bound thrombin.

        c. Inhibits thrombin activation of protein C.

        d. Hirudin is a small molecule (molecular weight 7 kDa) that is eliminated by the kidney and is easily filtered at the end of CPB. Half-life is 40 min.

        e. r-Hirudin dose: 0.25 mg/kg bolus and an infusion to maintain the hirudin concentration at 2.5 μg/mL as determined by ecarin clotting time.

        f. Ecarin clotting time: not clinically available at this time [19].

        g. r-Hirudin-treated patients maintain platelet counts and hemoglobin levels and have few bleeding complications, if renal function is normal.

      4. Bivalirudin

        a. Synthetic polypeptide that directly inhibits thrombin by binding simultaneously to its active catalytic site and its substrate recognition site.


        b. Half-life is 24 min. Elimination is primarily by proteolysis and, to a smaller degree, renal elimination. Even though this half-life is shorter than heparin’s, the absence of a reversal agent and the profound degree of anticoagulation used for CPB may cause coagulopathy for 2 or more hours following CPB.

        c. Dose during interventional procedures: 0.75 mg/kg bolus followed by a 1.75 mg/kg/hour infusion, yielding a median ACT of 346 s.

        d. Dose for CPB: 1.0 mg/kg bolus followed by a 2.5 mg/kg/hr infusion. A recirculation limb and avoidance of circuit (and saphenous vein graft) stasis is necessary.

        e. Studies in interventional procedures suggest lower bleeding rates than UFH and similar efficacy.

      5. Argatroban. A direct thrombin inhibitor that is approved by the U.S. Food and Drug Administration (FDA) for anticoagulation in HIT patients but not yet approved for use in CPB. Argatroban undergoes hepatic metabolism with a half-life 39 to 51 min.

III. Neutralization of heparin

   A. Proof of concept. Protamine, commercially prepared from fish sperm, first found clinical utility in its combination with insulin to delay insulin’s absorption and prolong its effect. Combination of protamine with heparin, intended to achieve a similar prolongation of heparin’s effect, resulted instead in inactivation of heparin. Combining the strongly cationic protamine with strongly anionic heparin produces a stable complex devoid of anticoagulant activity.


     1. Heparin and protamine combine in proportion to weight. One milligram of protamine neutralizes 1 mg (typically 100 U) of heparin [20].

   B. Protamine dose. Since protamine appears to distribute within the circulatory system as heparin does, the protamine dose required to neutralize a given dose of heparin equals the number of milligrams of heparin remaining in the patient’s circulation at the time of neutralization. Thus, clinical protocols to determine the initial dose of protamine first estimate the blood heparin concentration and blood volume. Direct assay of heparin concentration is difficult and unnecessary. Indirect assay using protamine effect in vitro is more accurate than ratio-based estimates and is easily performed. Three different methods of choosing an initial protamine dose are commonly used:


      1. Empiric ratio. Most clinicians choose a dose of protamine based on the total number of units of heparin administered, giving between 0.6 and 1.3 mg of protamine for every 100 units of heparin administered. Clinical efficacy has been documented using a ratio as low as 0.6 mg protamine to 100 units of heparin administered. Initial doses using this ratio result in a mild-to-moderate protamine excess relative to heparin that ensures total neutralization and minimizes the likelihood of subsequent heparin rebound. Ratios exceeding 1 mg/100 units tend to be excessive.

          Example: A patient receives 25,000 units of heparin before CPB, with no subsequent heparin required, and 5,000 units in the bypass pump prime. A ratio of 1 mg:100 units yields a 300 mg protamine neutralizing dose. A ratio of 0.6 mg:100 units would result in 180 mg of protamine as the neutralizing dose. The former dose applies best when the patient receives all pump blood, unwashed, prior to protamine administration; the latter dose makes most sense for prolonged bypass during which heparin had ample opportunity for metabolism and excretion.

      2. Estimated from a heparin dose-response curve. This method depends upon construction of a heparin dose-response curve prior to or during bypass (see Fig. 19.3). The technique then estimates the blood concentration of heparin at the time of neutralization. See Fig. 19.3 legend for details. Assumptions include (i) linearity of the heparin dose–ACT response relationship; (ii) potential extrapolation beyond actual data collected; (iii) constancy of the volume of distribution of heparin and protamine. Most often, clinicians choose the arbitrary ratio of 1 mg of protamine to 100 units of heparin to calculate the protamine dose.

      3. Calculated from in vitro protamine effect, by measuring ACT both with and without protamine added to blood at the time of neutralization. The HMS system from Medtronic HemoTec automates this technique to calculate protamine dose. These curves assume a linear dose-response and extrapolate to a baseline ACT in calculating the initial protamine dose. These devices also calculate blood volume based on patient height and weight rather than on any definitive measurement, which can be a source of error. The same protamine titration curve can be performed using the TEG, but this assay has not become fully automated.

   C. Protamine administration. Always administer protamine slowly. The rate of administration is more important than the route of administration in preventing adverse hemodynamic effects (see Section III.E.1). One can either use a syringe or dilute the drug in a small volume of intravenous fluid and infuse by gravity or calibrated pump. Because the syringe technique results in multiple boluses, restricting its use to doses less than 1 mg/kg, or 20 mg in any 60-s period, appears advisable. We recommend a continuous infusion technique rather than hand-operated syringe administration, because this reduces the natural tendency to administer the protamine too quickly and it frees our hands for other important patient care activities (e.g., vasoactive drug titration, echocardiography examination) that coincide with protamine administration.


      1. The injected dose of protamine cannot neutralize heparin bound to plasma proteins or within endothelial cells. Release of heparin from these stored areas after initial protamine administration may result in reappearance of heparin anticoagulant effect (heparin rebound). Small additional doses of protamine will provide neutralization when repeat testing (e.g., ACT initially normalizes to 110 s but 30 min later is 140 s) shows a heparin effect in a bleeding patient.

      2. Protamine does not remain long in the vascular system following administration. Therefore, administration of heparinized blood, such as that remaining in the CPB machine without washing, following completion of the neutralizing protamine infusion, will likely result in renewed anticoagulation to a small extent. An additional small dose of protamine, about 1 mg/20 mL of transfused pump blood, should address this contingency.

   D. Monitoring heparin neutralization

      1. ACT. After protamine administration, the ACT test should return a value no more than 10% above the value before heparin administration. If more prolonged, residual heparin activity is likely. An ACT value that remains prolonged despite additional protamine suggests a technical error or, less commonly, some other hemostatic abnormality.

      2. Protamine titration. This test utilizes a series of tubes with increasing amounts of added protamine, beginning with none. One adds patient blood to each tube and determines which tube clots first. Because protamine has anticoagulant properties in vitro, it prolongs the coagulation time of normal blood in test tubes. Knowing which tube clots first allows identification of unneutralized heparin as well as estimation of the additional amount of protamine needed to achieve complete neutralization. This test can be performed manually or by using an automated device (Medtronic HemoTec).


   E. Adverse effects [21].

     1. Hypotension from rapid administration. Administration of a neutralizing dose of protamine (about 3 mg/kg) over 3 or fewer minutes decreases both systemic and pulmonary arterial pressures as well as venous return. This predictable response may be blunted, but not predictably avoided, by volume loading. Release of vasoactive compounds from mast cells or other sites may be responsible for this adverse response.

      2. Anaphylactoid reactions. Although protamine is a foreign protein, immune responses occur infrequently following exposure so that true allergy to protamine is uncommon. Table 19.3 lists those patients at potential risk.

Table 19.3 Patients at potential risk for true allergy to protamine

      3. Pulmonary vasoconstriction. Occasionally protamine increases pulmonary arterial pressure, resulting in right ventricular failure, decreased cardiac output, and systemic hypotension. Formation of large heparin–protamine complexes may stimulate production of thromboxane by pulmonary macrophages, causing vasoconstriction. In some animal models, the probability of experiencing this response is increased by faster rates of protamine administration.

      4. Antihemostatic effects. Protamine activates thrombin receptors on platelets, causing partial activation and subsequent impairment of platelet aggregation. Transient thrombocytopenia also occurs in the first hour after a full neutralizing dose of protamine. Inhibition of plasma coagulation can also occur.

     5. Treatment of adverse protamine reactions. Systemic hypotension within 10 min of protamine administration suggests protamine as the cause, but other causes such as hypovolemia and left ventricular dysfunction should be considered. Specific treatment depends on other hemodynamic events.

        a. Normal or low pulmonary artery pressures suggest either rapid administration or an anaphylactoid reaction. Rapid fluid administration alone often suffices to treat the former, whereas the latter cause usually requires aggressive volume resuscitation, large doses of epinephrine, and possibly other vasoactive compounds and inhaled bronchodilators. Refer to other sources for the treatment of acute anaphylaxis, including use of systemic steroids.

        b. High pulmonary artery pressures suggest a pulmonary vasoconstriction reaction. Inotropes with pulmonary dilating properties, such as isoproterenol or milrinone, will support the failing heart while facilitating movement of blood across the pulmonary circulation. Nitric oxide may also be useful. With extreme hemodynamic deterioration, reinstitution of CPB may be necessary. In this case, give a full heparin dose (at least 300 U/kg). Occasionally, heparin alone will correct the pulmonary hypertension (presumably by breaking up large heparin–protamine complexes, the putative stimulant to thromboxane production) such that reinstitution of CPB no longer becomes necessary.

     6. Prevention of adverse responses

        a. Rate of administration. Always administer fully neutralizing doses of protamine slowly (minimum duration 3 min, with a target of 10 min recommended). Rather than depend on volume loading to prevent hypotension, simply dilute the drug and give it slowly. Place the calculated dose in 50 mL or more of clear fluid and connect a small-drop (approximately 60 drops/mL) administration set to limit the infusion rate or use an infusion pump connected either to a 50 mL syringe or a 50 mL (or greater) fluid bag containing protamine. Some clinicians advocate a protamine “test dose,” such as 1 mg intravenously, prior to protamine administration. Our view is that slow initiation of a continuous protamine infusion accomplishes the same end, after which the infusion rate can be increased as tolerated down to a minimum infusion time as above.

        b. Route of administration. The preponderance of evidence suggests that peripheral vein infusion offers no benefit over central venous infusion as long as the infusion is dilute and not rapid. Injection directly into the aorta provides no reliable protection and risks introduction of embolic material, such as small bubbles, pieces of rubber stopper, or glass.

        c. High-risk subgroups. Patients without previous exposure to protamine, including those with diabetes or prior vasectomy, require no special measures before initial exposure. Even patients who have received protamine-containing insulin preparations rarely develop an adverse response. Only patients with a prior history of an adverse response to protamine warrant special treatment. See Table 19.3 for the relative risks of protamine administration to these subgroups.

        d. Prior protamine reaction. Prepare a special, dilute protamine solution of about 1 mg in 100 mL and administer over 10 min. If no adverse response occurs, administer the fully neutralizing dose as described earlier. Skin tests taken before giving protamine provide little predictive value and frequently are falsely positive. Special immunologic tests for protamine allergy, such as radioallergosorbent test (RAST) and ELISA, also demonstrate many false-positive results.

   F. Alternatives to protamine administration

      1. Allow heparin’s effect to dissipate. This approach results in continued hemorrhage with substantial transfusion requirements and bouts of hypovolemia and the potential for consumptive coagulopathy. Although this may be the only option available, ideally it should be avoided.

      2. Platelet concentrates. Platelet factor 4 (PF4) is released from activated platelets. It combines with and neutralizes heparin. However, platelet concentrates do not effectively restore coagulation following CPB. A recombinant form of PF4 failed in clinical trials as a protamine alternative.

      3. Hexadimethrine. This synthetic polycation, no longer readily available in the United States because of renal toxicity, can avoid true allergic reactions to protamine. However, like protamine, it forms complexes with heparin that can incite pulmonary vasoconstriction if administered quickly.

      4. Methylene blue. Even large doses do not effectively restore the ACT. However, inhibition of nitric oxide synthetase can incite pulmonary hypertension, making this approach potentially hazardous.

      5. Investigational substances. Heparinase I, an enzyme produced by harmless soil bacteria, failed in clinical trials as a protamine alternative. Virus-like particles, engineered from bacteriophage Qβ coat protein, have shown consistent neutralization of heparin with less variability than protamine in plasma from heparin-treated patients [22]. Cationically modified chitosan binds to heparin to form complexes similar to those formed by protamine [23].

      6. With no alternative to protamine immediately available, or even under active clinical investigation, alternatives to heparin (see Section II.E) assume greater importance in the management of patients with demonstrated severe adverse responses to protamine.

IV. Hemostatic abnormalities in the cardiac surgical patient [1,24]

   A. Management of the patient taking preoperative antithrombotic drugs. Table 19.4 lists commonly used antithrombotic drugs and their mechanisms of action.

Table 19.4 Common antithrombotic drugs

      1. Anticoagulant therapy. Patients receiving warfarin anticoagulant medications should be advised to discontinue the medication 3 to 5 days before the anticipated cardiac surgery. Generally an INR value less than 2 reflects an acceptable recovery of vitamin K-dependent clotting factors. In fact, some residual inhibition of the extrinsic coagulation pathway advantageously accentuates anticoagulation for CPB. If anticoagulation is so vitally important that it must be maintained until the time of surgery, an intravenous infusion of heparin may be started preoperatively. Heparin may be discontinued a few hours before surgery or continued into the operative period.

        a. In urgent or semi-urgent surgery, the effects of warfarin may need rapid reversal which can be accomplished by giving FFP until INR correction occurs [25].

        b. In clinical studies, prothrombin complex concentrate was found more effective with quicker time to INR correction and no volume overload observed compared to FFP [26].

      2. Antiplatelet therapy

        a. Aspirin. Many studies support the use of aspirin in the prevention of thrombosis in coronary and cerebral vascular disease. The patients taking aspirin therapy who are undergoing cardiac surgery have a propensity for increased bleeding postoperatively; however, the benefits of aspirin therapy, weighed against a potential for bleeding, often lead to preoperative continuation of aspirin therapy. Most patients do not bleed excessively with this approach. An increase in bleeding, if it exists, is not necessarily accompanied by an increase in transfusion requirements due to blood conservation strategies in use.

        b. Glycoprotein IIb/IIIa (GPIIb/IIIa) inhibitors. The GPIIb/IIIa receptor is the platelet fibrinogen receptor, which causes fibrinogen bridging of adjacent platelets and subsequent platelet aggregation. GPIIb/IIIa inhibitors inhibit platelet aggregation and have been increasingly used during interventional cardiology procedures. Their beneficial effects include reductions in mortality and cardiac events after angioplasty and stent procedures. However, there is strong potential for hemorrhagic complications if these patients present for emergent cardiac surgery. Currently, the three intravenous GPIIb/IIIa inhibitors in clinical use are abciximab, tirofiban, and eptifibatide. Abciximab is a monoclonal antibody to the GPIIb/IIIa receptor that inhibits fibrinogen binding and covalently alters the GPIIb/IIIa receptor. Tirofiban and eptifibatide are smaller competitive receptor blockers whose effects are reversible after discontinuation of therapy. Their short duration of action may mitigate some perioperative bleeding complications [26] and may actually provide some platelet protection during CPB.

        c. ADP receptor inhibitors. The thienopyridine derivatives ticlopidine, clopidogrel, and prasugrel noncompetitively antagonize at a platelet ADP receptor known as the P2Y12 receptor. Blockade of this receptor by one of these agents elevates cyclic adenosine monophosphate levels to induce profound and rapid platelet disaggregation. Clopidogrel use in conjunction with percutaneous coronary intervention or in acute coronary syndromes reduces the occurrence of adverse ischemic events [27]. Antiplatelet activity is permanent for the life span of the platelet because the P2Y12 receptor is permanently altered. Clopidogrel is a pro-drug that is metabolized by cytochrome P450 (2C19 and 3A4) to its active metabolite. The combination of clopidogrel and aspirin is synergistic. Meta-analyses of comparative trials demonstrate that clopidogrel pre-treatment is associated with more bleeding than that observed in non-exposed patients [28].

   B. Abnormalities acquired during cardiac surgery

      1. Endothelial dysfunction. Contact of blood with extracorporeal surfaces initiates a “total body inflammatory response” characterized by activation of coagulation, fibrinolysis, and inflammation. This leads to an abnormal cellular–endothelial interaction.

      2. Persistent heparin effect. This is uncommon because most clinicians fully neutralize the administered heparin, although heparin rebound (resumption of heparin effect after complete neutralization) is relatively common within the first 2 hrs following CPB, and usually responds well to small (e.g., 25 mg) incremental doses of protamine.

      3. Platelet abnormalities (Table 19.5)

Table 19.5 Causes of platelet dysfunction in cardiac surgery

        a. Thrombocytopenia occurs frequently after CPB due to dilution of the blood volume with extracorporeal circuit volume and to platelet consumption or sequestration. This thrombocytopenia can be severe (<50,000/μL) but, in the absence of other hemostasis abnormalities, often does not lead to excessive bleeding. With modern techniques, platelet counts after CPB most often exceed 100,000/μL.

        b. Platelet dysfunction. The most prevalent yet elusive cause of hemostatic abnormalities after CPB is platelet dysfunction. Platelets are rendered inactive by contact activation from the extracorporeal surfaces, hypothermia, receptor downregulation, and by heparin and protamine. Heparin activates platelets to render them less functional after CPB, and protamine also depresses platelet function. The use of antithrombotic drugs preoperatively leads to an even greater degree of platelet dysfunction after CPB.

      4. Coagulopathy. Hemodilution and consumption of coagulation factors by microvascular coagulation combine to cause the deficiencies of coagulation seen after CPB. Despite the use of large doses of heparin, contact activation causes microvascular activation of factor XII and initiates the intrinsic pathway of coagulation.

      5. Fibrinolysis. Fibrinolysis can be primary or secondary during CPB (Fig. 19.4). Primary fibrinolysis occurs from release of endothelial plasminogen activators. Secondary fibrinolysis describes activation of plasmin as a result of a feedback response to fibrin formation. Circulating plasmin degradation products adversely affect platelet function.

Figure 19.4 Schematic diagram of the fibrinolytic system displaying endogenous and exogenous activators and inhibitors of fibrinolysis. The antihemostatic actions of plasmin and fibrin(ogen) split products (FSPs) are illustrated. aPC, activated protein C; EACA,-aminocaproic acid; TA, tranexamic acid; tPA, tissue plasminogen activator; UK, urokinase.

      6. Pharmacology. As noted earlier, heparin and protamine impair platelet function. Other drugs commonly used during CPB (milrinone, nitroglycerin, nitroprusside) adversely affect platelet function in vitro, but in vivo their effects appear to be clinically undetectable.

      7. Hypothermia. Hypothermia impairs the enzymatic cascades of the coagulation pathway. Platelets are activated during mild hypothermia and are depressed during moderate-to-severe hypothermia.

   C. Pharmacologic and protective prophylaxis

      1. Platelet protection

        a. Antifibrinolytic agents. See IV.C.2.

        b. Coated surfaces. Heparin-bonded circuits attenuate the inflammatory response to CPB and may confer some platelet protective properties.

        c. Antiplatelet agents that are active during CPB may confer some platelet protection so long as they are short-acting. Patients who have emergency surgery after having been exposed to tirofiban or eptifibatide do not experience excessive postoperative bleeding and receive equivalent or reduced transfusion volumes.

      2. Antifibrinolytic agents [29]

        a. Synthetic antifibrinolytic agents: ε-aminocaproic acid (EACA) and tranexamic acid (TA). EACA and TA act as lysine analogs that bind to the lysine-binding sites of plasmin and plasminogen (Fig. 19.4). TA is a more potent analog of EACA that has a higher affinity for plasminogen than does EACA. Fibrin degradation products inhibit platelet function. Thus, plasmin inhibition may protect platelets. The benefits of EACA and TA have been demonstrated in multiple meta-analyses to reduce bleeding in cardiac surgery, when these agents are used prophylactically (i.e., initiated before CPB and maintained throughout CPB) rather than as rescue agents [30]. Dosing: EACA 100 to 150 mg/kg bolus, 10 to 15 mg/kg/hr, or 4 to 10 g bolus, 1 g/hr. Reports suggest constant plasma activity may be best achieved with smaller initial boluses (approximately 50 mg/kg) followed by higher maintenance doses (20 to 25 mg/kg/hr). Dosing range: Low dose: TA 10 to 20 mg/kg bolus, 1 to 2 mg/kg/hr; Moderate dose: 30 to 50 mg/kg bolus, 15 to 30 mg/kg/hr; or High dose: 5-g bolus, repeat bolus to total 15 g. The latter dosing scheme probably is much higher than necessary and there is concern that high-dose TA may be associated with central nervous system adverse events.

        b. Aprotinin, a high-molecular-weight proteinase inhibitor of bovine origin, inhibits plasmin, kallikrein, and other serine proteases. Aprotinin decreases activation of the hematologic system during CPB and subsequent fibrinolysis, resulting in a 30% to 40% decrease in chest tube drainage. It is the only agent found to be successful in reducing the rate of reoperation for bleeding in meta-analyses. However, an increased rate of mortality and renal morbidity has been reported with aprotinin that caused the drug to be removed from commercial use until the data were reevaluated [3133]. Commercial availability of aprotinin going forward remains uncertain, and this may vary from country to country. Newer protease inhibitors are currently under investigation for use in cardiac surgery.

        c. Investigational substances: Carbon monoxide releasing molecule-2 (CORM-2) significantly improved velocity of clot formation and clot strength in plasma in patients both before and following CPB. Pending further trials, CORM-2 may be of use in improving coagulation and decreasing fibrinolysis in patients with persistent bleeding after CPB [34].

      3. Accurate heparin and protamine dosing. Attempts to individualize heparin and protamine doses in order to minimize bleeding rely on patient-specific doses of each drug based on patient sensitivity. A number of different heparin and protamine management strategies have been reported to result in reduced perioperative bleeding.

        a. Higher heparin concentrations during CPB have been associated with increased mediastinal tube bleeding postoperatively. Higher doses might predispose to greater bleeding as a result of heparin rebound or platelet dysfunction. This leads to the practice of giving just enough heparin to maintain a “threshold” minimum acceptable ACT.

        b. Some investigators postulate that higher heparin levels allow for reduced activation of the coagulation cascade and may blunt the consumptive coagulopathy that occurs with microvascular coagulation. This leads to the practice of maintaining heparin at a specific concentration in the blood, which leads to large doses of heparin and higher ACTs. Heparin management strategies are still highly variable and institution-specific.

        c. Lower protamine doses have been used successfully to neutralize heparin after CPB and have been associated with reduced bleeding and transfusion requirements. This relationship between higher heparin and lower protamine doses has been suggested to result in less postoperative bleeding.

      4. Inhibition of inflammation

        a. Coated surfaces. Heparin-bonded CPB circuits make the extracorporeal circuit more biocompatible and thus effectively attenuate the inflammatory response to CPB. Despite this benefit, use of these circuits has not uniformly reduced morbidity. Use of a reduced heparin dose in conjunction with heparin-bonded circuits has shown reductions in postoperative chest tube drainage and transfusion requirements, but reduced heparin dosing in this scenario is not yet fully endorsed as the overall effect on safety for CPB is unclear.

        b. Steroids. Methylprednisolone 500 to 1,000 mg has proved helpful in some studies but has not been universally adopted due to risks of hyperglycemia and immune suppression.

        c. Aprotinin acts by kallikrein inhibition to reduce the inflammatory response, but this can only be achieved in the high-dose range (e.g., full Hammersmith protocol). As noted above, clinical availability of aprotinin in the future is uncertain (see Section C.2.b).

        d. Modified ultrafiltration has a beneficial effect of reducing postoperative morbidity and improving organ function in pediatric patients.

        e. Complement inhibitors are mostly experimental agents. They act to prevent activation of complement by kallikrein inhibition or direct complement antagonism. Reduction of the inflammatory response theoretically would reduce morbidity. Clinical trials are underway to evaluate different complement inhibitors but no one agent has been statistically proven to reduce adverse events.

V. Management of postbypass bleeding [1]

   A. Evaluation of hemostasis

      1. Achieve surgical hemostasis

      2. Confirm adequate heparin neutralization

        a. Tests of heparin neutralization: heparinase ACT, protamine titration test, heparinase thromboelastography (TEG). Note: The standard ACT is not a specific test for complete heparin neutralization, that is, it is possible to have ACT return to normal while some residual heparin effect remains. If the ACT has returned to baseline and residual heparin is either suspected or identified using a protamine titration technique, treat this by titrating in additional protamine. Most often protamine 25 to 50 mg completes the neutralization.

      3. Point-of-care testing to diagnose and treat bleeding. Point-of-care tests should be used appropriately in order to accurately and rapidly pinpoint the cause of a hemostasis defect. Etiologies of post-CPB bleeding should be prioritized by the clinician and should be tested in logical order. These tests should include heparin neutralization (see V.A.2.a), platelet function, platelet number, coagulation, and fibrinolysis (Table 19.6).

Table 19.6 Point-of-care tests of platelet function

        a. Tests of platelet function: TEG maximal amplitude, Sonoclot, Platelet Works, Platelet Function Analyzer-100, VerifyNow, and whole blood aggregometry to name a few. Treatment of platelet dysfunction may include administration of desmopressin acetate (DDAVP) 0.3 mg/kg slowly (over approximately 15 to 20 min, as vasodilation may occur). If platelet number is reduced or platelet dysfunction persists after DDAVP, a transfusion of platelet concentrates is recommended.

        b. Coagulation tests: PT, aPTT, thrombin time, ACT, TEG-R, or TEG-K value. Treatment of this abnormality includes transfusion of FFP, or occasionally administration of recombinant factor VIIa (rVIIa) when the coagulopathy is unresponsive to FFP. Administration of rVIIa will also treat CPB or drug-induced platelet dysfunction by creating a thrombin burst; however, its safety has not been fully established.

        c. Tests of fibrinolysis: Euglobulin lysis time, TEG lysis index, fibrin degradation products, D-dimers. Treatment of primary fibrinolysis includes administration of an antifibrinolytic agent. If an antifibrinolytic agent has already been administered and discontinued, this same agent should be re-started. Starting a different class of antifibrinolytic drug in the same patient is not recommended. Secondary fibrinolysis should be treated by replacement of consumed coagulation factors (FFP or cryoprecipitate).

        d. Note: Treatment of any abnormal laboratory value should not take place unless warranted by the clinical situation, that is, observation of clinical coagulopathy. Treat the patient, not the number!

   B. Treatment of postbypass hemostatic disorders

Follow the nine steps below in order. They reflect the causes of postoperative bleeding in decreasing frequency of occurrence. Remember to maintain intravascular volume to avoid generating or exacerbating a consumptive coagulopathy. These steps are designed to address the situation where urgency dictates that one must treat a presumed cause before obtaining laboratory results. Do not embrace any one presumed cause; rather, constantly re-evaluate and question your assumptions. Instead of bedside empiricism, however, we encourage the development and use of transfusion algorithms using point-of-care testing (see V.C).


      1. Rule out a surgical cause. Copious chest tube drainage usually results from a vessel in need of suture. A generalized ooze suggests a non-surgical cause. Keep the blood pressure in the low normal range while the surgeons effect repair, and optimally for sometime thereafter to maximize the potential for clot formation.

      2. Maintain normothermia. In the effort to restore intravascular volume rapidly, clinicians must infuse refrigerated blood products only with adequate warming, lest they cause or accentuate hypothermia, thereby decreasing platelet function and enzymatic activity of clotting proteins.

      3. Determine the cause. While surgeons are checking anastomoses, assure complete heparin neutralization with an ACT and aPTT, the latter showing prolongation at smaller blood concentrations of heparin. Consider measuring fibrinogen concentration, D-dimer, and thrombin time, the last of which is prolonged only by residual heparin, inadequate amount or functionality of fibrinogen, and fibrin degradation products.

      4. Give more protamine if the ACT exceeds its baseline (pre-heparin) value by >10 s (or the aPTT is more than 1.3 times its control value). A dose of 25 to 50 mg usually suffices. Do so upon learning the ACT results and while awaiting other laboratory results.

      5. In the absence of hypovolemia, consider application of 5 cm of positive end-expiratory pressure (PEEP) to help tamponade open blood vessels in the chest. This maneuver may be most effective after the sternum is closed because of limitations in the ability of PEEP to effectively increase mediastinal end-expiratory pressure while the sternum remains open.

      6. Platelet transfusion and/or DDAVP. If testing uncovers platelet dysfunction, or if it is highly suspected, give 1 unit of platelet concentrates per 10 kg body weight for an estimated effective platelet count below 100,000/μL. While awaiting platelet concentrates from the blood bank, consider administration of DDAVP 0.3 ug/kg especially if there is laboratory evidence of platelet dysfunction (e.g., decreased maximum amplitude [MA] on TEG). This may resolve the coagulopathy and avoid the need for platelet concentrates.

      7. FFP. Give 15 mL/kg for a PT in excess of 1.5 times control, or an INR in excess of 2.0.

      8. Give antifibrinolytic medication. Although these agents work best when administered prophylactically before and during CPB, about half of their benefit occurs if given (or continued) in the post-CPB period. An increased D-dimer value, or teardrop-shaped TEG tracing, suggests active fibrinolysis that would warrant antifibrinolytic agent administration or higher doses if antifibrinolytic agents are already being administered.

      9. Give rVIIa or cryoprecipitate. rVIIa has been associated with “miracle cures” of post-CPB coagulopathy, but its use has been limited to rescue situations and very few prospective studies about its use have been published. In cardiac surgery, reductions in bleeding have clearly been documented; however, a dose of 80 ug/kg has been associated with some hypercoagulable complications [35]. The rVIIa probably makes the most sense when given as a secondary intervention when two “rounds” of FFP 10 to 15 mL/kg and platelet concentrates 1 unit/10 kg have not resolved the coagulopathy. Overwhelming coagulopathy may at times call for earlier use of this potentially lifesaving agent, for which an initial dose of 30 to 40 ug/kg is recommended. Cryoprecipitate 1 unit/4 kg body weight (generally 15 to 20 units in adults) will correct fibrinogen deficiency (<100 mg/dL). Its use is best reserved for situations where hypofibrinogenemia has been documented.

   C. Transfusion medicine and the use of algorithms. Allogeneic transfusions after CPB are common because of the wide range of hemostatic insults incurred. The lack of adequate testing of hemostasis and the subjectivity of a diagnosis of microvascular bleeding lead to indiscriminate transfusion practices.

      1. Transfusion of red blood cells, platelets, and plasma is fraught with adverse effects, not least of which are infectious disease transmission, acute lung injury, and immunomodulation. The rational use of transfusion algorithms should create an approach to transfusion medicine that is stepwise, logical, and based on the hemostatic defects that are most common and easily treated. This usually starts with a specific test of heparin neutralization. After residual heparin is ruled out, other coagulation tests are measured.

     2. One of the most critical tests that should be measured “early” in a transfusion algorithm is an accurate point-of-care test of platelet function. This will minimize the occurrence of indiscriminate transfusion practices because subjective assessment of microvascular bleeding often leads to empiric transfusion practices. Rapid and accurate diagnosis of a hemostasis abnormality after CPB is critical. TEG predicts abnormal bleeding after CPB and has been used successfully in a number of algorithms to reduce the incidence of transfusion [36,37]. Other point-of-care monitors (e.g., PT, aPTT, and platelet count) used in rational algorithms have proved effective in reducing transfusions in routine CPB patients and in patients who have been exposed to antithrombotic agents preoperatively [38].

      3. Aside from confirming neutralization of heparin, the routine use of coagulation tests after CPB has not proven beneficial in the absence of a clinical coagulopathy.


 1. Horrow JC, Mueksch JN. Coagulation testing. In: Gravlee GP, Davis RF, Stammers AH, Ungerleider RM, eds. Cardiopulmonary Bypass: Principles and Practice. 3rd ed. Philadelphia, PA: Lippincott Williams & Wilkins; 2008:459–471.

 2. Shore-Lesserson L, Gravlee GP. Anticoagulation for cardiopulmonary bypass. In: Gravlee GP, Davis RF, Stammers AH, Ungerleider RM, eds. Cardiopulmonary Bypass: Principles and Practice. 3rd ed. Philadelphia, PA: Lippincott Williams & Wilkins; 2008:472–501.

 3. Merton RE, Curtis AD, Thomas DP. A comparison of heparin potency estimates obtained by activated partial thromboplastin time and British pharmacopoeial assays. Thromb Haemost. 1985;53:116–117.

 4. Gravlee GP, Angert KC, Tucker WY, et al. Early anticoagulation peak and rapid distribution after intravenous heparin. Anesthesiology. 1988;68:126–129.

 5. Heres EK, Speight K, Benckart D, et al. The clinical onset of heparin is rapid. Anesth Analg. 2001;92:1391–1395.

 6. Olsson P, Lagergren H, Ek S. The elimination from plasma of intravenous heparin. An experimental study on dogs and humans. Acta Med Scand. 1963;173:619–630.

 7. Bull BS, Korpman RA, Huse WM, et al. Heparin therapy during extracorporeal circulation. I. Problems inherent in existing heparin protocols. J Thorac Cardiovasc Surg. 1975;69:674–684.

 8. Cohen EJ, Camerlengo LJ, Dearing JP. Activated clotting times and cardiopulmonary bypass I: The effect of hemodilution and hypothermia upon activated clotting time. J Extracorp Technol.1980;12:139–141.

 9. Gravlee GP, Case LD, Angert KC, et al. Variability of the activated coagulation time. Anesth Analg. 1988;67:469–472.

10. Despotis GJ, Summerfield AL, Joist JH. Comparison of activated coagulation time and whole blood heparin measurements with laboratory plasma anti-Xa heparin concentration in patients having cardiac operations. J Thorac Cardiovasc Surg. 1994;108:1076–1082.

11. Koster A, Despotis G, Gruendel M, et al. The plasma supplemented modified activated clotting time for monitoring of heparinization during cardiopulmonary bypass: A pilot investigation. Anesth Analg. 2002;95:26–30.

12. Chavez JJ, Foley DE, Snider CC, et al. A novel thrombelastograph tissue factor/kaolin assay of activated clotting times for monitoring heparin anticoagulation during cardiopulmonary bypass. Anesth Analg. 2004;99:1290–1294.

13. Hellstern P, Bach J, Simon M, et al. Heparin monitoring during cardiopulmonary bypass surgery using the one-step point-of-care whole blood anti-factor-Xa clotting assay heptest-POC-Hi. J Extra Corpor Technol.2007;39:81–86.

14. Ganter MT, Monn A, Tavakoli R, et al. Kaolin-based activated coagulation time measured by sonoclot in patients undergoing cardiopulmonary bypass. J Cardiothorac Vasc Anesth. 2007;21:524–528.

15. Bennett-Guerrero E, Slaughter TF, White WD, et al. Preoperative anti-PF4/heparin antibody level predicts adverse outcome after cardiac surgery. J Thorac Cardiovasc Surg. 2005;130:1567–1572.

16. Warkentin TE, Kelton JG. Temporal aspects of heparin-induced thrombocytopenia. N Engl J Med. 2001;344:1286–1292.

17. Koster A, Meyer O, Hausmann H, et al. In vitro cross-reactivity of danaparoid sodium in patients with heparin-induced thrombocytopenia type II undergoing cardiovascular surgery. J Clin Anesth.2000;12:324–327.

18. Greinacher A, Volpel H, Janssens U, et al. Recombinant hirudin (lepirudin) provides safe and effective anticoagulation in patients with heparin-induced thrombocytopenia. A prospective study. Circulation. 1999;99:73–80.

19. Koster A, Hansen R, Grauhan O, et al. Hirudin monitoring using the TAS ecarin clotting time in patients with heparin-induced thrombocytopenia type II. J Cardiothorac Vasc Anesth. 2000;14:249–252.

20. Metz S, Horrow JC. Pharmacologic manipulation of coagulation: protamine and other heparin antagonists. In: Lake CL, Moore RA, eds. Blood: Hemostasis, Transfusion, and Alternatives in the Perioperative Period. New York, NY: Raven Press; 1995:119–130.

21. Horrow JC. Protamine: A review of its toxicity. Anesth Analg. 1985;64:348–361.

22. Gale AJ, Elias DJ, Averell PM, et al. Engineered virus-like nanoparticles reverse heparin anticoagulation more consistently than protamine in plasma from heparin-treated patients. Thromb Res.2011;128(suppl 4):e9–e13. Epub 2011 Apr 14.

23. Kamiński K, Szczubiałka K, Zazakowny K, et al. Chitosan derivatives as novel potential heparin reversal agents. J Med Chem. 2010;53:4141–4147.

24. Spiess BD, Horrow JC, Kaplan JA. Transfusion medicine and coagulation disorders. In: Kaplan JA, Reich DL, Savino JS, eds. Kaplan’s Cardiac Anesthesia: The Echo Era. 6th ed. Philadelphia, PA: Elsevier Saunders; 2011:949–991.

25. Demeyere R, Gillardin S, Arnout J, et al. Comparison of fresh frozen plasma and prothrombin complex concentrate for the reversal of oral anticoagulants in patients undergoing cardiopulmonary bypass surgery: a randomized study. Vox Sang. 2010;99:251–260.

26. Brown DL, Fann CS, Chang CJ. Meta-analysis of effectiveness and safety of abciximab versus eptifibatide or tirofiban in percutaneous coronary intervention. Am J Cardiol. 2001;87:537–541.

27. Bertrand ME, Rupprecht HJ, Urban P, et al. Double-blind study of the safety of clopidogrel with and without a loading dose in combination with aspirin compared with ticlopidine in combination with aspirin after coronary stenting: the Clopidogrel Aspirin Stent International Cooperative Study (CLASSICS). Circulation. 2000;102:624–629.

28. Purkayastha S, Athanasiou T, Malinovski V, et al. Does clopidogrel affect outcome after coronary artery bypass grafting? A meta-analysis. Heart. 2006;92:531–532.

29. Levi M, Cromheecke ME, de Jonge E, et al. Pharmacologic strategies to decrease excessive blood loss in cardiac surgery: A meta-analysis of clinically relevant endpoints. Lancet. 1999;354:1940–1947.

30. Brown JR, Birkmeyer NJO, O’Connor JT. Meta-analysis comparing the effectiveness and adverse outcomes of antifibrinolytic agents in cardiac surgery. Circulation. 2007;115:2801–2813.

31. Mangano DT, Tudor IC, Dietzel C. The risk associated with aprotinin in cardiac surgery. N Engl J Med. 2006;354:353–365.

32. Fergusson DA, Hebert PC, Mazur CD, et al. A comparison of aprotinin and lysine analogues in high-risk cardiac surgery. N Engl J Med. 2008;358:2319–2331.

33. Henry D, Carless P, Fergusson D, et al. The safety of aprotinin and lysine-derived antifibrinolytic drugs in cardiac surgery: A meta-analysis. CMAJ. 2009;180(suppl 2):183–193.

34. Malayaman SN, Entwistle JW 3rd, Boateng P, et al. Carbon monoxide releasing molecule-2 improves coagulation in patient plasma in vitro following cardiopulmonary bypass. Blood Coagul Fibrinolysis. 2011;22:362–368.

35. Gill R, Herbertson M, Vuylsteke A, et al. Placebo-controlled trial in the setting of bleeding after cardiac surgery. Safety and efficacy of recombinant activated factor VII: a randomized placebo-controlled trial in the setting of bleeding after cardiac surgery. Circulation. 2009;120:21–27.

36. Shore-Lesserson L, Manspeizer HE, DePerio M, et al. Thromboelastography-guided transfusion algorithm reduces transfusions in complex cardiac surgery. Anesth Analg. 1999;88:312–319.

37. Nuttall GA, Oliver WC, Santrach PJ, et al. Efficacy of a simple intraoperative transfusion algorithm for nonerythrocyte component utilization after cardiopulmonary bypass. Anesthesiology.2001;94:773–781.

38. Chen L, Bracey AW, Radovancevic R, et al. Clopidogrel and bleeding in patients undergoing elective coronary artery bypass grafting. J Thorac Cardiovasc Surg. 2004;128:425–431.