Chapter 4: Hemostasis, Surgical Bleeding, and Transfusion

1. The life span of platelets ranges from 7 to 10 days. Drugs that interfere with platelet function include aspirin, clopidogrel, prasugrel, dipyridamole, and the glycoprotein IIb/IIIa (GP IIb/IIIa) inhibitors. Approximately 5 to 7 days should pass from the time the drug is stopped until an elective procedure is performed.

2. The acute coagulopathy of trauma results from a combination of activation of protein C and hyperfibrinolysis. It is distinct from disseminated intravascular coagulation, is present on arrival to the emergency department, and is associated with an increase in mortality.

3. Newer anticoagulants like dabigatran and rivaroxaban have no readily available method of detection of the degree of anticoagulation and may not be readily reversible.

4. Therapeutic anticoagulation preoperatively and postoperatively is becoming increasingly more common. The patient’s risk of intraoperative and postoperative bleeding should guide the need for reversal of anticoagulation therapy preoperatively and the timing of its reinstatement postoperatively.

5. Damage control resuscitation has three basic components: permissive hypotension, minimizing crystalloid-based resuscitation, and the administration of predefined blood products.

6. The need for massive transfusion should be anticipated, and guidelines should be in place to provide early and increased amounts of red blood cells, plasma, and platelets.

Hemostasis is a complex process whose function is to limit blood loss from an injured vessel. Four major physiologic events participate in the hemostatic process: vascular constriction, platelet plug formation, fibrin formation, and fibrinolysis. Although each tends to be activated in order, the four processes are interrelated so that there is a continuum and multiple reinforcements. The process is shown schematically in Fig. 4-1.

Vascular Constriction

Vascular constriction is the initial response to vessel injury. It is more pronounced in vessels with medial smooth muscles and is dependent on local contraction of smooth muscle. Vasoconstriction is subsequently linked to platelet plug formation. Thromboxane A₂ (TXA₂) is produced locally at the site if injury via the release of arachidonic acid from platelet membranes and is a potent constrictor of smooth muscle. Similarly, endothelin synthesized by injured endothelium and serotonin (5-hydroxytryptamine [5-HT]) released during platelet aggregation are potent vasoconstrictors. Lastly, bradykinin and fibrinopeptides, which are involved in the coagulation schema, are also capable of contracting vascular smooth muscle.

The extent of vasoconstriction varies with the degree of vessel injury. A small artery with a lateral incision may remain open due to physical forces, whereas a similarly sized vessel that is completely transected may contract to the extent that bleeding ceases spontaneously.

Platelet Function

Platelets are anucleate fragments of megakaryocytes. The normal circulating number of platelets ranges between 150,000 and 400,000/μL. Up to 30% of circulating platelets may be sequestered in the spleen. If not consumed in a clotting reaction, platelets are normally removed by the spleen and have an average life span of 7 to 10 days.

Platelets play an integral role in hemostasis by forming a hemostatic plug and by contributing to thrombin formation (Fig. 4-2). Platelets do not normally adhere to each other or to the vessel wall but can form a plug that aids in cessation of bleeding when vascular disruption occurs. Injury to the intimal layer in the vascular wall exposes subendothelial collagen to which platelets adhere. This process requires von Willebrand factor (vWF), a protein in the subendothelium that is lacking in patients with von Willebrand’s disease. vWF binds to glycoprotein (GP) I/IX/V on the platelet membrane. Following adhesion, platelets initiate a release reaction that recruits other platelets from the circulating blood to seal the disrupted vessel. Up to this point, this process is known as primary hemostasis. Platelet aggregation is reversible and is not associated with secretion. Additionally, heparin does not interfere with this reaction, and thus, hemostasis can occur in the heparinized patient. Adenosine diphosphate (ADP) and serotonin are the principal mediators in platelet aggregation.

Arachidonic acid released from the platelet membranes is converted by cyclooxygenase to prostaglandin G₂ (PGG₂) and then to prostaglandin H₂ (PGH₂), which, in turn, is converted to TXA₂. TXA₂ has potent vasoconstriction and platelet aggregation effects. Arachidonic acid may also be shuttled to adjacent endothelial cells and converted to prostacyclin (PGI₂), which is a vasodilator and acts to inhibit platelet aggregation. Platelet cyclooxygenase is irreversibly inhibited by aspirin and reversibly blocked by nonsteroidal anti-inflammatory agents, but is not affected by cyclooxygenase-2 (COX-2) inhibitors.

In the second wave of platelet aggregation, a release reaction occurs in which several substances including ADP, Ca²⁺, serotonin, TXA₂, and α-granule proteins are discharged. Fibrinogen is a required cofactor for this process, acting as a bridge for the GP IIb/IIIa receptor on the activated platelets. The release reaction results in compaction of the platelets into a plug, a process that is no longer reversible. Thrombospondin, another protein secreted by the α-granule, stabilizes fibrinogen binding to the activated platelet surface and strengthens the platelet-platelet interactions. Platelet factor 4 (PF4) and α-thromboglobulin are also secreted during the release reaction. PF4 is a potent heparin antagonist. The second wave of platelet aggregation is inhibited by aspirin and nonsteroidal anti-inflammatory drugs, by cyclic adenosine monophosphate (cAMP), and by nitric oxide. As a consequence of the release reaction, alterations occur in the phospholipids of the platelet membrane that allow calcium and clotting factors to bind to the platelet surface, forming enzymatically active complexes. The altered lipoprotein surface (sometimes referred to as platelet factor 3) catalyzes reactions that are involved in the conversion of prothrombin (factor II) to thrombin (factor IIa) by activated factor X (Xa) in the presence of factor V and calcium, and it is involved in the reaction by which activated factor IX (IXa), factor VIII, and calcium activate factor X. Platelets may also play a role in the initial activation of factors XI and XII.

Coagulation

Hemostasis involves a complex interplay and combination of interactions between platelets, the endothelium, and multiple circulating or membrane-bound coagulation factors. While a bit simplistic and not reflective of the depth or complexity of these interactions, the coagulation cascade has traditionally been depicted as two possible pathways converging into a single common pathway (Fig. 4-3). While this pathway reflects the basic process and sequences that lead to the formation of a clot, the numerous feedback loops, endothelial interplay, and platelet functions are not included. The intrinsic pathway begins with the activation of factor XII that subsequently activates factors XI, IX, and VIII. In this pathway, each of the primary factors is “intrinsic” to the circulating plasma, whereby no surface is required to initiate the process. In the extrinsic pathway, tissue factor (TF) is released or exposed on the surface of the endothelium, binding to circulating factor VII, facilitating its activation to VIIa. Each of these pathways continues on to a common sequence that begins with the activation of factor X to Xa (in the presence of VIIIa). Subsequently, Xa (with the help of factor Va) converts factor II (prothrombin) to thrombin and then factor I (fibrinogen) to fibrin. Clot formation occurs after fibrin monomers are cross-linked to polymers with the assistance of factor XIII.

One convenient feature of depicting the coagulation cascade with two merging arms is that commonly used laboratory tests segregate abnormalities of clotting to one of the two arms. An elevated activated partial thromboplastin time (aPTT) is associated with abnormal function of the intrinsic arm of the cascade (II, IX, X, XI, XII), while the prothrombin time (PT) is associated with the extrinsic arm (II, VII, X). Vitamin K deficiency or warfarin use affects factors II, VII, IX, and X

Expanding from the basic concept of Fig. 4-3, the primary pathway for coagulation is initiated by TF exposure following subendothelial injury. Clot propagation ensues with what is a sequence of four similar enzymatic reactions, each involving a proteolytic enzyme generating the next enzyme by cleaving its proenzyme, a phospholipid surface (e.g., platelet membrane) in the presence of ionized calcium, and a helper protein. TF binds to VIIa, and this complex catalyzes the activation of factor X to Xa. This complex is four orders of magnitude more active at converting factor X than is factor VIIa alone and also activates factor IX to IXa. Factor Xa, together with Va, calcium, and phospholipid, composes the prothrombinase complex that converts prothrombin to thrombin. The prothrombinase complex is significantly more effective at catalyzing its substrate than is factor Xa alone. Thrombin is then involved with the conversion of fibrinogen to fibrin and activation of factors V, VII, VIII, XI, and XIII.

In building on the redundancy inherent in the coagulation system, factor VIIIa combines with IXa to form the intrinsic factor complex. Factor IXa is responsible for the bulk of the conversion of factor X to Xa. This complex (VIIIa-IXa) is 50 times more effective at catalyzing factor X activation than is the extrinsic (TF-VIIa) complex and five to six orders of magnitude more effective than factor IXa alone.

Once formed, thrombin leaves the membrane surface and converts fibrinogen by two cleavage steps into fibrin and two small peptides termed fibrinopeptides A and B. Removal of fibrinopeptide A permits end-to-end polymerization of the fibrin molecules, whereas cleavage of fibrinopeptide B allows side-to-side polymerization of the fibrin clot. This latter step is facilitated by thrombin-activatable fibrinolysis inhibitor (TAFI), which acts to stabilize the resultant clot.

In seeking to balance profound bleeding with overwhelming clot burden, several related processes exist to prevent propagation of the clot beyond the site of injury.¹ First, feedback inhibition on the coagulation cascade deactivates the enzyme complexes leading to thrombin formation. Thrombomodulin (TM) presented by the endothelium serves as a “thrombin sink” by forming a complex with thrombin, rendering it no longer available to cleave fibrinogen. This then activates protein C (APC) and reduces further thrombin generation by inhibiting factors V and VIII. Second, tissue plasminogen activator (tPA) is released from the endothelium following injury, cleaving plasminogen to initiate fibrinolysis. APC then consumes plasminogen activator inhibitor-1 (PAI-1), leading to increased tPA activity and fibrinolysis. Building on the anticoagulant response to inhibit thrombin formation, tissue factor pathway inhibitor (TFPI) is released, blocking the TF-VIIa complex and reducing the production of factors Xa and IXa. Antithrombin III (AT-III) then neutralizes all of the procoagulant serine proteases and also inhibits the TF-VIIa complex. The most potent mechanism of thrombin inhibition involves the APC system. APC forms a complex with its cofactor, protein S, on a phospholipid surface. This complex then cleaves factors Va and VIIIa so they are no longer able to participate in the formation of TF-VIIa or prothrombinase complexes. This is of interest clinically in the form of a genetic mutation, called factor V Leiden. In this setting, factor V is resistant to cleavage by APC, thereby remaining active as a procoagulant. Patients with factor V Leiden are predisposed to venous thromboembolic events.

Degradation of fibrin clot is accomplished by plasmin, a serine protease derived from the proenzyme plasminogen. Plasmin formation occurs as a result of one of several plasminogen activators. tPA is made by the endothelium and other cells of the vascular wall and is the main circulating form of this family of enzymes. tPA is selective for fibrin-bound plasminogen so that endogenous fibrinolytic activity occurs predominately at the site of clot formation. The other major plasminogen activator, urokinase plasminogen activator (uPA), also produced by endothelial cells as well as by urothelium, is not selective for fibrin-bound plasminogen. Of note, the thrombin-TM complex activates TAFI, leading to a mixed effect on clot stability. In addition to inhibiting fibrinolysis directly, removal of the terminal lysine on the fibrin molecule by TAFI renders the clot more susceptible to lysis by plasmin.

Fibrinolysis

Fibrin clot breakdown (lysis) allows restoration of blood flow during the healing process following injury and begins at the same time clot formation is initiated. Fibrin polymers are degraded by plasmin, a serine protease derived from the proenzyme plasminogen. Plasminogen is converted to plasmin by one of several plasminogen activators, including tPA. Plasmin then degrades the fibrin mesh at various places, leading to the production of circulating fragments, termed fibrin degradation products (FDPs), cleared by other proteases or by the kidney and liver (Fig. 4-4). Fibrinolysis is directed by circulating kinases, tissue activators, and kallikrein present in vascular endothelium. tPA is synthesized by endothelial cells and released by the cells on thrombin stimulation. Bradykinin, a potent endothelial-dependent vasodilator, is cleaved from high molecular weight kininogen by kallikrein and enhances the release of tPA. Both tPA and plasminogen bind to fibrin as it forms, and this trimolecular complex cleaves fibrin very efficiently. After plasmin is generated, however, it cleaves fibrin somewhat less efficiently.

As with clot formation, fibrinolysis is also kept in check through several robust mechanisms. tPA activates plasminogen more efficiently when it is bound to fibrin, so that plasmin is formed selectively on the clot. Plasmin is inhibited by α₂-antiplasmin, a protein that is cross-linked to fibrin by factor XIII, which helps to ensure that clot lysis does not occur too quickly. Any circulating plasmin is also inhibited by α₂-antiplasmin and circulating tPA or urokinase. Clot lysis yields FDPs including E-nodules and D-dimers. These smaller fragments interfere with normal platelet aggregation, and the larger fragments may be incorporated into the clot in lieu of normal fibrin monomers. This may result in an unstable clot as seen in cases of severe coagulopathy such as hyperfibrinolysis associated with trauma-induced coagulopathy or disseminated intravascular coagulopathy. The presence of D-dimers in the circulation may serve as a marker of thrombosis or other conditions in which a significant activation of the fibrinolytic system is present. Another inhibitor of the fibrinolytic system is TAFI, which removes lysine residues from fibrin that are essential for binding plasminogen.

Coagulation Factor Deficiencies

Inherited deficiencies of all of the coagulation factors are seen. However, the three most frequent are factor VIII deficiency (hemophilia A and von Willebrand’s disease), factor IX deficiency (hemophilia B or Christmas disease), and factor XI deficiency. Hemophilia A and hemophilia B are inherited as sex-linked recessive disorders with males being affected almost exclusively. The clinical severity of hemophilia A and hemophilia B depends on the measurable level of factor VIII or factor IX in the patient’s plasma. Plasma factor levels less than 1% of normal are considered severe disease, factor levels between 1% and 5% moderately severe disease, and levels between 5% and 30% mild disease. Patients with severe hemophilia have spontaneous bleeds, frequently into joints, leading to crippling arthropathies. Intracranial bleeding, intramuscular hematomas, retroperitoneal hematomas, and gastrointestinal, genitourinary, and retropharyngeal bleeding are added clinical sequelae seen with severe disease. Patients with moderately severe hemophilia have less spontaneous bleeding but are likely to bleed severely after trauma or surgery. Mild hemophiliacs do not bleed spontaneously and have only minor bleeding after major trauma or surgery. Since platelet function is normal in hemophiliacs, patients may not bleed immediately after an injury or minor surgery as they have a normal response with platelet activation and formation of a platelet plug. At times, the diagnosis of hemophilia is not made in these patients until after their first minor procedure (e.g., tooth extraction or tonsillectomy).

Patients with hemophilia A or B are treated with factor VIII or factor IX concentrate, respectively. Recombinant factor VIII is strongly recommended for patients not treated previously and is generally recommended for patients who are both human immunodeficiency virus (HIV) and hepatitis C virus (HCV) seronegative. For factor IX replacement, the preferred products are recombinant or high-purity factor IX. In general, activity levels should be restored to 30% to 40% for mild hemorrhage, 50% for severe bleeding, and 80% to 100% for life-threatening bleeding. Up to 20% of hemophiliacs with factor VIII deficiency develop inhibitors that can neutralize FVIII. For patients with low titers, inhibitors can be overcome with higher doses of factor VIII. For patients with high titer inhibitors, alternate treatments should be used and may include porcine factor VIII, prothrombin complex concentrates, activated prothrombin complex concentrates, or recombinant factor VIIa. For patients undergoing elective surgical procedures, a multidisciplinary approach with preoperative planning and replacement is recommended.²

von Willebrand’s Disease

von Willebrand’s disease (vWD), the most common congenital bleeding disorder, is characterized by a quantitative or qualitative defect in vWF, a large glycoprotein responsible for carrying factor VIII and platelet adhesion. The latter is important for normal platelet adhesion to exposed subendothelium and for aggregation under high shear conditions. Patients with vWD have bleeding that is characteristic of platelet disorders such as easy bruising and mucosal bleeding. Menorrhagia is common in women. vWD is classified into three types. Type I is a partial quantitative deficiency, type II is a qualitative defect, and type III is total deficiency. For bleeding, type I patients usually respond well to desmopressin (DDAVP). Type II patients may respond, depending on the particular defect. Type III patients are usually unresponsive. These patients may require vWF concentrates.³

Factor XI Deficiency

Factor XI deficiency, an autosomal recessive inherited condition sometimes referred to as hemophilia C, is more prevalent in the Ashkenazi Jewish population but found in all races. Spontaneous bleeding is rare, but bleeding may occur after surgery, trauma, or invasive procedures. Treatment of patients with factor XI deficiency who present with bleeding or in whom surgery is planned and who are known to have bled previously is with fresh frozen plasma (FFP). Each milliliter of plasma contains 1 unit of factor XI activity, so the volume needed depends on the patient’s baseline level, the desired level, and the plasma volume. Antifibrinolytics may be useful in patients with menorrhagia. Factor VIIa is recommended for patients with anti-factor XI antibodies, although thrombosis has been reported.⁴ There has been renewed interest in factor XI inhibitors as antithrombotic agents, because patients with factor XI deficiency generally have only minimal bleeding risk unless a severe deficiency is present and seem to be protected from thrombosis.⁵

Deficiency of Factors II (Prothrombin), V, and X

Inherited deficiencies of factors II, V, and X are rare. These deficiencies are inherited as autosomal recessive. Significant bleeding in homozygotes with less than 1% of normal activity is encountered. Bleeding with any of these deficiencies is treated with FFP. Similar to factor XI, FFP contains one unit of activity of each per milliliter. However, factor V activity is decreased because of its inherent instability. The half-life of prothrombin (factor II) is long (approximately 72 hours), and only about 25% of a normal level is needed for hemostasis. Prothrombin complex concentrates can be used to treat deficiencies of prothrombin or factor X. Daily infusions of FFP are used to treat bleeding in factor V deficiency, with a goal of 20% to 25% activity. Factor V deficiency may be coinherited with factor VIII deficiency. Treatment of bleeding in individuals with the combined deficiency requires factor VIII concentrate and FFP. Some patients with factor V deficiency are also lacking the factor V normally present in platelets and may need platelet transfusions as well as FFP.

Factor VII Deficiency

Inherited factor VII deficiency is a rare autosomal recessive disorder. Clinical bleeding can vary widely and does not always correlate with the level of FVII coagulant activity in plasma. Bleeding is uncommon unless the level is less than 3%. The most common bleeding manifestations involve easy bruising and mucosal bleeding, particularly epistaxis or oral mucosal bleeding. Postoperative bleeding is also common, reported in 30% of surgical procedures.⁶ Treatment is with FFP or recombinant factor VIIa. The half-life of recombinant factor VIIa is only approximately 2 hours, but excellent hemostasis can be achieved with frequent infusions. The half-life of factor VII in FFP is up to 4 hours.

Factor XIII Deficiency

Congenital factor XIII (FXIII) deficiency, originally recognized by Duckert in 1960, is a rare autosomal recessive disease usually associated with a severe bleeding diathesis.⁷ The male-to-female ratio is 1:1. Although acquired FXIII deficiency has been described in association with hepatic failure, inflammatory bowel disease, and myeloid leukemia, the only significant association with bleeding in children is the inherited deficiency.⁸ Bleeding is typically delayed because clots form normally but are susceptible to fibrinolysis. Umbilical stump bleeding is characteristic, and there is a high risk of intracranial bleeding. Spontaneous abortion is usual in women with factor XIII deficiency unless they receive replacement therapy. Replacement can be accomplished with FFP, cryoprecipitate, or a factor XIII concentrate. Levels of 1% to 2% are usually adequate for hemostasis.

Platelet Functional Defects

Inherited platelet functional defects include abnormalities of platelet surface proteins, abnormalities of platelet granules, and enzyme defects. The major surface protein abnormalities are thrombasthenia and Bernard-Soulier syndrome. Thrombasthenia, or Glanzmann thrombasthenia, is a rare genetic platelet disorder, inherited in an autosomal recessive pattern, in which the platelet glycoprotein IIb/IIIa (GP IIb/IIIa) complex is either lacking or present but dysfunctional. This defect leads to faulty platelet aggregation and subsequent bleeding. The disorder was first described by Dr. Eduard Glanzmann in 1918.⁹ Bleeding in thrombasthenic patients must be treated with platelet transfusions. The Bernard-Soulier syndrome is caused by a defect in the GP Ib/IX/V receptor for vWF, which is necessary for platelet adhesion to the subendothelium. Transfusion of normal platelets is required for bleeding in these patients.

The most common intrinsic platelet defect is storage pool disease. It involves loss of dense granules (storage sites for ADP, adenosine triphosphate [ATP], Ca²⁺, and inorganic phosphate) and α-granules. Dense granule deficiency is the most prevalent of these. It may be an isolated defect or occur with partial albinism in the Hermansky-Pudlak syndrome. Bleeding is variable, depending on the severity of the granule defect. Bleeding is caused by the decreased release of ADP from these platelets. A few patients have been reported who have decreased numbers of both dense and α-granules. They have a more severe bleeding disorder. Patients with mild bleeding as a consequence of a form of storage pool disease can be treated with DDAVP. It is likely that the high levels of vWF in the plasma after DDAVP somehow compensate for the intrinsic platelet defect. With more severe bleeding, platelet transfusion is required.

Platelet Abnormalities

Acquired abnormalities of platelets are much more common than acquired defects and may be quantitative or qualitative, although some patients have both types of defects. Quantitative defects may be a result of failure of production, shortened survival, or sequestration. Failure of production is generally a result of bone marrow disorders such as leukemia, myelodysplastic syndrome, severe vitamin B₁₂ or folate deficiency, chemotherapeutic drugs, radiation, acute ethanol intoxication, or viral infection. If a quantitative abnormality exists and treatment is indicated either due to symptoms or the need for an invasive procedure, platelet transfusion is utilized. The etiologies of both qualitative and quantitative defects are reviewed in Table 4-1.

Quantitative Defects

Shortened platelet survival is seen in immune thrombocytopenia, disseminated intravascular coagulation, or disorders characterized by platelet thrombi such as thrombotic thrombocytopenic purpura and hemolytic uremic syndrome. Immune thrombocytopenia may be idiopathic or associated with other autoimmune disorders or low-grade B-cell malignancies, and it may also be secondary to viral infections (including HIV) or drugs. Secondary immune thrombocytopenia often presents with a very low platelet count, petechiae and purpura, and epistaxis. Large platelets are seen on peripheral smear. Initial treatment consists of corticosteroids, intravenous gamma globulin, or anti-D immunoglobulin in patients who are Rh positive. Both gamma globulin and anti-D immunoglobulin are rapid in onset. Platelet transfusions are not usually needed unless central nervous system bleeding or active bleeding from other sites occurs. Survival of the transfused platelets is usually short.

Primary immune thrombocytopenia is also known as idiopathic thrombocytopenic purpura (ITP). In children, it is usually acute in onset, short lived, and typically follows a viral illness. In contrast, ITP in adults is gradual in onset, chronic in nature, and has no identifiable cause. Because the circulating platelets in ITP are young and functional, bleeding is less for a given platelet count than when there is failure of platelet production. The pathophysiology of ITP is believed to involve both impaired platelet production and T cell–mediated platelet destruction.¹⁰ Management options are summarized in Table 4-2.¹¹ Treatment of drug-induced immune thrombocytopenia may simply entail withdrawal of the offending drug, but corticosteroids, gamma globulin, and anti-D immunoglobulin may hasten recovery of the count. Heparin-induced thrombocytopenia (HIT) is a form of drug-induced immune thrombocytopenia. It is an immunologic event during which antibodies against platelet factor 4 (PF4) formed during exposure to heparin affect platelet activation and endothelial function with resultant thrombocytopenia and intravascular thrombosis.¹² The platelet count typically begins to fall 5 to 7 days after heparin has been started, but if it is a re-exposure, the decrease in count may occur within 1 to 2 days. HIT should be suspected if the platelet count falls to less than 100,000 or if it drops by 50% from baseline in a patient receiving heparin. While HIT is more common with full-dose unfractionated heparin (1%–3%), it can also occur with prophylactic doses or with low molecular weight heparins. Interestingly, approximately 17% of patients receiving unfractionated heparin and 8% receiving low molecular weight heparin develop antibodies against PF4, yet a much smaller percentage develop thrombocytopenia and even fewer develop clinical HIT.¹³ In addition to the mild to moderate thrombocytopenia, this disorder is characterized by a high incidence of thrombosis that may be arterial or venous. Importantly, the absence of thrombocytopenia in these patients does not preclude the diagnosis of HIT.

The diagnosis of HIT may be made by using either a serotonin release assay (SRA) or an enzyme-linked immunosorbent assay (ELISA). The SRA is highly specific but not sensitive, so a positive test supports the diagnosis but a negative test does not exclude HIT.¹² On the other hand, the ELISA has a low specificity, so although a positive ELISA confirms the presence of anti-heparin-PF4, it does not help in the diagnosis of clinical HIT. A negative ELISA, however, essentially rules out HIT.

The initial treatment of suspected HIT is to stop heparin and begin an alternative anticoagulant. Stopping heparin without addition of another anticoagulant is not adequate to prevent thrombosis in this setting. Alternative anticoagulants are primarily thrombin inhibitors. The most recent guideline by the American College of Chest Physicians recommends lepirudin, argatroban, or danaparoid for patients with normal renal function and argatroban for patients with renal insufficiency.¹⁴ Because of warfarin’s early induction of a hypercoagulable state, warfarin should be instituted only once full anticoagulation with an alternative agent has been accomplished and the platelet count has begun to recover.

These are also disorders in which thrombocytopenia is a result of platelet activation and formation of platelet thrombi. In thrombotic thrombocytopenic purpura (TTP), large vWF molecules interact with platelets, leading to activation. These large molecules result from inhibition of a metalloproteinase enzyme, ADAMtS13, which cleaves the large vWF molecules.¹⁵ TTP is classically characterized by thrombocytopenia, microangiopathic hemolytic anemia, fever, and renal and neurologic signs or symptoms. The finding of schistocytes on a peripheral blood smear aids in the diagnosis. Plasma exchange with replacement of FFP is the treatment for acute TTP.¹⁶ Additionally, rituximab, a monoclonal antibody against the CD20 protein on B lymphocytes, has shown promise as an immunomodulatory therapy directed against patients with acquired TTP, of which the majority are autoimmune mediated.¹⁷

Hemolytic uremic syndrome (HUS) often occurs secondary to infection by Escherichia coli 0157:H7 or other Shiga toxin-producing bacteria. The metalloproteinase is normal in these cases. HUS is usually associated with some degree of renal failure, with many patients requiring renal replacement therapy. Neurologic symptoms are less frequent. A number of patients develop features of both TTP and HUS. This may occur with autoimmune diseases, especially systemic lupus erythematosus and HIV infection, or in association with certain drugs (such as ticlopidine, mitomycin C, gemcitabine) or immunosuppressive agents (such as cyclosporine and tacrolimus). Discontinuation of the involved drug is the mainstay of therapy. Plasmapheresis is frequently used, but it is not clear what etiologic factor is being removed by the pheresis.

Sequestration is another important cause of thrombocytopenia and usually involves trapping of platelets in an enlarged spleen typically related to portal hypertension, sarcoid, lymphoma, or Gaucher’s disease. The total body platelet mass is essentially normal in patients with hypersplenism, but a much larger fraction of the platelets are in the enlarged spleen. Platelet survival is mildly decreased. Bleeding is less than anticipated from the count because sequestered platelets can be mobilized to some extent and enter the circulation. Platelet transfusion does not increase the platelet count as much as it would in a normal person because the transfused platelets are similarly sequestered in the spleen. Splenectomy is not indicated to correct the thrombocytopenia of hypersplenism caused by portal hypertension.

Thrombocytopenia is the most common abnormality of hemostasis that results in bleeding in the surgical patient. The patient may have a reduced platelet count as a result of a variety of disease processes, as discussed earlier. In these circumstances, the marrow usually demonstrates a normal or increased number of megakaryocytes. By contrast, when thrombocytopenia occurs in patients with leukemia or uremia and in patients on cytotoxic therapy, there are generally a reduced number of megakaryocytes in the marrow. Thrombocytopenia also occurs in surgical patients as a result of massive blood loss with product replacement deficient in platelets. Thrombocytopenia may also be induced by heparin administration during cardiac and vascular cases, as in the case of HIT, or may be associated with thrombotic and hemorrhagic complications. When thrombocytopenia is present in a patient for whom an elective operation is being considered, management is contingent upon the extent and cause of platelet reduction. A count of greater than 50,000/μL generally requires no specific therapy.

Early platelet administration has now become part of massive transfusion protocols.^18,19 Platelets are also administered preoperatively to rapidly increase the platelet count in surgical patients with underlying thrombocytopenia. One unit of platelet concentrate contains approximately 5.5 × 10¹⁰ platelets and would be expected to increase the circulating platelet count by about 10,000/μL in the average 70-kg person. Fever, infection, hepatosplenomegaly, and the presence of antiplatelet alloantibodies decrease the effectiveness of platelet transfusions. In patients refractory to standard platelet transfusion, the use of human leukocyte antigen (HLA)-compatible platelets coupled with special processors has proved effective.

Qualitative Platelet Defects

Impaired platelet function often accompanies thrombocytopenia but may also occur in the presence of a normal platelet count. The importance of this is obvious when one considers that 80% of overall strength is related to platelet function. The life span of platelets ranges from 7 to 10 days, placing them at increased risk for impairment by medical disorders and prescription and over-the-counter medications. Impairment of ADP-stimulated aggregation occurs with massive transfusion of blood products. Uremia may be associated with increased bleeding time and impaired aggregation. Defective aggregation and platelet dysfunction are also seen in patients with thrombocythemia, polycythemia vera, and myelofibrosis.

Drugs that interfere with platelet function include aspirin, clopidogrel, prasugrel, dipyridamole, and GP IIb/IIIa inhibitors. Aspirin, clopidogrel, and prasugrel all irreversibly inhibit platelet function. Clopidogrel and prasugrel do so through selective irreversible inhibition of ADP-induced platelet aggregation.²⁰ Aspirin works through irreversible acetylation of platelet prostaglandin synthase.

There are no prospective randomized trials in general surgical patients to guide the timing of surgery in patients on aspirin, clopidogrel, or prasugrel.²¹ The general recommendation is that approximately 5 to 7 days should pass from the time the drug is stopped until an elective procedure is performed.²² Timing of urgent and emergent surgeries is even more unclear. Preoperative platelet transfusions may be beneficial, but there are no good data to guide their administration. However, new functional tests are becoming available that may better demonstrate defects in platelet function and may serve to guide the timing of operation or when platelet transfusions might be indicated.

Other disorders associated with abnormal platelet function include uremia, myeloproliferative disorders, monoclonal gammopathies, and liver disease. In the surgical patient, platelet dysfunction of uremia can often be corrected by dialysis or the administration of DDAVP. Platelet transfusion may not be helpful if the patient is uremic when the platelets are given and only serve to increase antibodies. Platelet dysfunction in myeloproliferative disorders is intrinsic to the platelets and usually improves if the platelet count can be reduced to normal with chemotherapy. If possible, surgery should be delayed until the count has been decreased. These patients are at risk for both bleeding and thrombosis. Platelet dysfunction in patients with monoclonal gammopathies is a result of interaction of the monoclonal protein with platelets. Treatment with chemotherapy or, occasionally, plasmapheresis to lower the amount of monoclonal protein improves hemostasis.

Acquired Hypofibrinogenemia

Disseminated Intravascular Coagulation (DIC)

DIC is an acquired syndrome characterized by systemic activation of coagulation pathways that result in excessive thrombin generation and the diffuse formation of microthrombi. This disturbance ultimately leads to consumption and depletion of platelets and coagulation factors with the resultant classic picture of diffuse bleeding. Fibrin thrombi developing in the microcirculation may cause microvascular ischemia and subsequent end-organ failure if severe. There are many different conditions that predispose a patient to DIC, and the presence of an underlying condition is required for the diagnosis. For example, injuries resulting in embolization of materials such as brain matter, bone marrow, or amniotic fluid can act as potent thromboplastins that activate the DIC cascade.²³ Additional etiologies include malignancy, organ injury (such as severe pancreatitis), liver failure, certain vascular abnormalities (such as large aneurysms), snake bites, illicit drugs, transfusion reactions, transplant rejection, and sepsis.²⁴ In fact, DIC frequently accompanies sepsis and may be associated with multiple organ failure. As of yet, scoring systems for organ failure do not routinely incorporate DIC. The important interplay between sepsis and coagulation abnormalities was demonstrated by Dhainaut et al who showed that activated protein C was effective in septic patients with DIC.²⁵ The diagnosis of DIC is made based on an inciting etiology with associated thrombocytopenia, prolongation of the prothrombin time, a low fibrinogen level, and elevated fibrin markers (FDPs, D-dimer, soluble fibrin monomers). A scoring system developed by the International Society for Thrombosis and Hemostasis has been shown to have high sensitivity and specificity for diagnosing DIC as well as a strong correlation between an increasing DIC score and mortality, especially in patients with infections.²⁶

The most important facets of treatment are relieving the patient’s causative primary medical or surgical problem and maintaining adequate perfusion. If there is active bleeding, hemostatic factors should be replaced with FFP, which is usually sufficient to correct the hypofibrinogenemia, although cryoprecipitate, fibrinogen concentrates, or platelet concentrates may also be needed. Given the formation of microthrombi in DIC, heparin therapy has also been proposed. Most studies, however, have shown that heparin is not helpful in acute forms of DIC, but may be indicated in cases where thrombosis predominates, such as arterial or venous thromboembolism and severe purpura fulminans.

Primary Fibrinolysis

An acquired hypofibrinogenic state in the surgical patient can be a result of pathologic fibrinolysis. This may occur in patients following prostate resection when urokinase is released during surgical manipulation of the prostate or in patients undergoing extracorporeal bypass. The severity of fibrinolytic bleeding is dependent on the concentration of breakdown products in the circulation. Antifibrinolytic agents, such as ε-aminocaproic acid and tranexamic acid, interfere with fibrinolysis by inhibiting plasminogen activation.

Myeloproliferative Diseases

Polycythemia, or an excess of red blood cells, places surgical patients at risk. Spontaneous thrombosis is a complication of polycythemia vera, a myeloproliferative neoplasm, and can be explained in part by increased blood viscosity, increased platelet count, and an increased tendency toward stasis. Paradoxically, a significant tendency toward spontaneous hemorrhage also is noted in these patients. Thrombocytosis can be reduced by the administration of low-dose aspirin, phlebotomy, and hydroxyurea.²⁷

Coagulopathy of Liver Disease

The liver plays a key role in hemostasis because it is responsible for the synthesis of many of the coagulation factors (Table 4-3). Patients with liver disease, therefore, have decreased production of several key non-endothelial cell-derived coagulation factors as well as natural anticoagulant proteins, causing a disturbance in the balance between procoagulant and anticoagulant pathways. This disturbance in coagulation mechanisms causes a complex paradigm of both increased bleeding risk and increased thrombotic risk. The most common coagulation abnormalities associated with liver dysfunction are thrombocytopenia and impaired humoral coagulation function manifested as prolongation of the prothrombin time and international normalized ratio (INR). The etiology of thrombocytopenia in patients with liver disease is typically related to hypersplenism, reduced production of thrombopoietin, and immune-mediated destruction of platelets. The total body platelet mass is often normal in patients with hypersplenism, but a much larger fraction of the platelets is sequestered in the enlarged spleen. Bleeding may be less than anticipated because sequestered platelets can be mobilized to some extent and enter the circulation. Thrombopoietin, the primary stimulus for thrombopoiesis, may be responsible for some cases of thrombocytopenia in cirrhotic patients, although its role is not well delineated. Finally, immune-mediated thrombocytopenia may also occur in cirrhotics, especially those with hepatitis C and primary biliary cirrhosis.²⁸ In addition to thrombocytopenia, these patients also exhibit platelet dysfunction via defective interactions between platelets and the endothelium, and possibly due to uremia and changes in endothelial function in the setting of concomitant renal insufficiency. Hypocoagulopathy is further exacerbated with low platelet counts because platelets help facilitate thrombin generation by assembling coagulation factors on their surfaces. In conditions mimicking intravascular flow, low hematocrit and low platelet counts contributed to decreased adhesion of platelets to endothelial cells, although increased vWF, a common finding in cirrhotic patients, may offset this change in patients with cirrhosis.²⁹ Hypercoagulability of liver disease has recently gained increased attention, with more evidence demonstrating the increased incidence of thromboembolism despite thrombocytopenia and a hypocoagulable state on conventional blood tests.^30,31 This is attributed to decreased production of liver-synthesized proteins C and S, antithrombin, and plasminogen levels, as well as elevated levels of endothelial-derived vWF and factor VIII, a potent driver of thrombin generation.^32,33 Given the concomitant hypo- and hypercoagulable features seen in patients with liver disease, conventional coagulation tests may be difficult to interpret, and alternative tests such as thromboelastography (TEG) may be more informative of the functional status of clot formation and stability in cirrhotic patients. Several studies imply that TEG provides a better assessment of bleeding risk than standard tests of hemostasis in patients with liver disease; however, no studies have directly tested this, and future prospective trials are needed.³⁴

Before instituting any therapy to ameliorate thrombocytopenia, the actual need for correction should be strongly considered. In general, correction based solely on a low platelet count should be discouraged. Most often, treatment should be withheld for invasive procedures and surgery. Platelet transfusions are the mainstay of therapy; however, the effect typically lasts only several hours. Risks associated with transfusions in general and the development of antiplatelet antibodies in a patient population likely to need recurrent correction should be considered. A potential alternative strategy involves administration of interleukin-11 (IL-11), a cytokine that stimulates proliferation of hematopoietic stem cells and megakaryocyte progenitors.²⁶ Most studies using IL-11 have been in cancer patients, although some evidence exists that it may be beneficial in cirrhotics as well. Significant side effects limit its usefulness.³⁵ A less well-accepted option is splenectomy or splenic embolization to reduce hypersplenism. In addition to the risks associated with these techniques, reduced splenic blood flow can reduce portal vein flow with subsequent portal vein thrombosis. Results are mixed following insertion of a transjugular intrahepatic portosystemic shunt (TIPS). Therefore, treatment of thrombocytopenia should not be the primary indication for a TIPS procedure.

Decreased production or increased destruction of coagulation factors as well as vitamin K deficiency can all contribute to a prolonged PT and INR in patients with liver disease. As liver dysfunction worsens, so does the liver’s synthetic function, which results in decreased production of coagulation factors. Additionally, laboratory abnormalities may mimic those of DIC. Elevated D-dimers have been reported to increase the risk of variceal bleeding. The absorption of vitamin K is dependent on bile production. Therefore, liver patients with impaired bile production and cholestatic disease may be at risk for vitamin K deficiency.

Similar to thrombocytopenia, correction of coagulopathy should be reserved for treatment of active bleeding and prophylaxis for invasive procedures and surgery. Treatment of coagulopathy caused by liver disease is usually done with FFP, but because the coagulopathy is usually not a result of decreased levels of factor V, complete correction is not usually possible. If the fibrinogen is less than 200 mg/dL, administration of cryoprecipitate may be helpful. Cryoprecipitate is also a source of factor VIII for the rare patient with a low factor VIII level.

Coagulopathy of Trauma

Traditional teaching regarding trauma-related coagulopathy attributed its development to acidosis, hypothermia, and dilution of coagulation factors. Recent data, however, have shown that over one third of injured patients have evidence of coagulopathy at the time of admission.³⁶ More importantly, patients arriving with coagulopathy are at a significantly higher risk of mortality, especially in the first 24 hours after injury. In light of these findings, a dramatic increase in research focused on the optimal management of the acute coagulopathy of trauma (ACoT) has been observed over the past several years. ACoT is not a simple dilutional coagulopathy but a complex problem with multiple mechanisms.³⁷ Whereas multiple contributing factors exist, the key initiators to the process of ACoT are shock and tissue injury. ACoT is a separate and distinct process from DIC, with its own specific components of hemostatic failure. Brohi et al have demonstrated that only patients in shock arrive coagulopathic and that it is the shock that induces coagulopathy through systemic activation of anticoagulant and fibrinolytic pathways.³⁸ As shown in Fig. 4-5, hypoperfusion causes activation of TM on the surface of endothelial cells. Thrombin-TM complexes induce an anticoagulant state through activation of protein C and enhancement of fibrinolysis. This same complex also limits the availability of thrombin to cleave fibrinogen to fibrin, which may explain why injured patients rarely have low levels of fibrinogen.

Acquired Coagulation Inhibitors

Among the most common acquired coagulation inhibitors is the antiphospholipid syndrome (APLS), which includes the lupus anticoagulant and anticardiolipin antibodies. These antibodies may be associated with either venous or arterial thrombosis, or both. In fact, patients presenting with recurrent thrombosis should be evaluated for APLS. Antiphospholipid antibodies are very common in patients with systemic lupus but may also be seen in association with rheumatoid arthritis and Sjögren’s syndrome. There are also individuals who will have no autoimmune disorders but develop transient antibodies in response to infections or those who develop drug-induced APLS. The hallmark of APLS is a prolonged aPTT in vitro but an increased risk of thrombosis in vivo.

Anticoagulation and Bleeding

Spontaneous bleeding can be a complication of any anticoagulant therapy whether it is heparin, low molecular weight heparins, warfarin, factor Xa inhibitors, or new direct thrombin inhibitors. The risk of spontaneous bleeding related to heparin is reduced with a continuous infusion technique. Therapeutic anticoagulation is more reliably achieved with a low molecular weight heparin. However, laboratory testing is more challenging with these medications, as they are not detected with conventional coagulation testing. However, their more reliable therapeutic levels (compared to heparin) make them an attractive option for outpatient anticoagulation and more cost-effective for the inpatient setting. If monitoring is required (e.g., in the presence of renal insufficiency or severe obesity), the drug effect should be determined with an assay for anti-Xa activity.

Warfarin is used for long-term anticoagulation in various clinical conditions including deep vein thrombosis, pulmonary embolism, valvular heart disease, atrial fibrillation, recurrent systemic emboli, recurrent myocardial infarction, prosthetic heart valves, and prosthetic implants. Due to the interaction of the P450 system, the anticoagulant effect of the warfarin is reduced (e.g., increases dose required) in patients receiving barbiturates as well as in patients with diets low in vitamin K. Increased warfarin requirements may also be needed in patients taking contraceptives or estrogen-containing compounds, corticosteroids, and adrenocorticotropic hormone (ACTH). Medications that can alter warfarin requirements are shown in Table 4-4.

Although warfarin use is often associated with a significant increase in morbidity and mortality in acutely injured and emergency surgery patients, with rapid reversal, these complications can be dramatically reduced. There are several reversal options that include vitamin K administration, plasma, cryoprecipitate, recombinant factor VIIa, and factor concentrates. Urgent reversal for life-threatening bleeding should include vitamin K and a rapid reversal agent such as plasma or prothrombin complex concentrate. In the elderly or those with intracranial hemorrhage, concentrates are preferred, whereas in situations with hypovolemia from hemorrhage, plasma should be used.

Newer anticoagulants like dabigatran and rivaroxaban have no readily available method of detection of the degree of anticoagulation. More concerning is the absence of any available reversal agent. Unlike warfarin, the nonreversible coagulopathy associated with dabigatran and rivaroxaban is of great concern to those providing emergent care to these patients.³⁹

The only possible strategy to reverse the coagulopathy associated with dabigatran may be emergent dialysis. Unfortunately, the ability to rapidly dialyze the hemodynamically unstable bleeding patient or rapidly dialyze the anticoagulated patient with an intracranial bleed is challenging even at large medical centers. Recent data suggest that rivaroxaban, however, may be reversed with the use of prothrombin complex concentrates (four-factor concentrates only: II, VII, IX, and X).⁴⁰ In less urgent states, these drugs can be held for 36 to 48 hours prior to surgery without increased risk of bleeding in those with normal renal function. Alternatively, activated clotting time (stand alone or with rapid TEG) or ecarin clotting time can be obtained in those on dabigatran, and anti-factor Xa assays can be obtained in those taking rivaroxaban.

Bleeding complications in patients on anticoagulants include hematuria, soft tissue bleeding, intracerebral bleeding, skin necrosis, and abdominal bleeding. Bleeding secondary to anticoagulation therapy is also not an uncommon cause of a rectus sheath hematomas. In most of these cases, reversal of anticoagulation is the only treatment that is necessary. Lastly, it is important to remember that symptoms of an underlying tumor may first present with bleeding while on anticoagulation.

Surgical intervention may prove necessary in patients receiving anticoagulation therapy. Increasing experience suggests that surgical treatment can be undertaken without full reversal of the anticoagulant, depending on the procedure being performed.⁴¹ When the aPTT is less than 1.3 times control in a heparinized patient or when the INR is less than 1.5 in a patient on warfarin, reversal of anticoagulation therapy may not be necessary. However, meticulous surgical technique is mandatory, and the patient must be observed closely throughout the postoperative period.

Certain surgical procedures should not be performed in concert with anticoagulation. In particular, cases where even minor bleeding can cause great morbidity, such as the central nervous system and the eye, surgery should be avoided. Emergency operations are occasionally necessary in patients who have been heparinized. The first step in these patients is to discontinue heparin. For more rapid reversal, protamine sulfate is effective. However, significant adverse reactions, especially in patients with severe fish allergies, may be encountered when administering protamine.⁴² Symptoms include hypotension, flushing, bradycardia, nausea, and vomiting. Prolongation of the aPTT after heparin neutralization with protamine may also be a result of the anticoagulant effect of protamine. In the elective surgical patient who is receiving coumarin-derivative therapy sufficient to effect anticoagulation, the drug can be discontinued several days before operation and the prothrombin concentration then checked (a level >50% is considered safe).⁴³ Rapid reversal of anticoagulation can be accomplished with plasma or prothrombin complex concentrates in the emergent situation. Parenteral administration of vitamin K also is indicated in elective surgical treatment of patients with biliary obstruction or malabsorption who may be vitamin K deficient. However, if low levels of factors II, VII, IX, and X (vitamin K–dependent factors) exist as a result of hepatocellular dysfunction, vitamin K administration is ineffective.

The perioperative management of patients receiving long-term oral anticoagulation therapy is an increasingly common problem. Definitive evidence-based guidelines regarding which patients require perioperative “bridging” anticoagulation and the most effective way to bridge are lacking. However, the American College of Chest Physicians Evidence-Based Clinical Practice Guidelines do serve as best practice for these situations.⁴⁴ A few clinical scenarios exist where the patient should be transitioned to intravenous heparin from oral anticoagulants. A heparin infusion should be held for 4 to 6 hours before the procedure and restarted within 12 to 24 hours of the end of its completion. The primary indication for this level of aggressiveness is patients with mechanical heart valves. Other indications include a recent (within 30 days) myocardial infarction, stroke, or pulmonary embolism. Situations such as thromboembolic events greater than 30 days prior, hypercoagulable history, and atrial fibrillation do not require such stringent restarting strategies.

Cardiopulmonary Bypass

Under normal conditions, homeostasis of the coagulation system is maintained by complex interactions between the endothelium, platelets, and coagulation factors. In patients undergoing cardiopulmonary bypass (CPB), contact with circuit tubing and membranes results in abnormal platelet and clotting factor activation, as well as activation of inflammatory cascades, that ultimately result in excessive fibrinolysis and a combination of both quantitative and qualitative platelet defects. Platelets undergo reversible alterations in morphology and their ability to aggregate, which causes sequestration in the filter, partially degranulated platelets, and platelet fragments. This multifactorial coagulopathy is compounded by the effects of shear stress in the system, induced hypothermia, hemodilution, and anticoagulation.⁴⁵

While on pump, activated clotting time measurements are obtained along with blood gas measurements; however, conventional coagulation assays and platelet counts are not normally performed until rewarming and after a standard dose of protamine has been given. TEG may give a better estimate of the extent of coagulopathy and may also be used to anticipate transfusion requirements if bleeding is present.⁴⁵

Empiric treatment with FFP and cryoprecipitate is often used for bleeding patients; however, there are no universally accepted transfusion thresholds. Platelet concentrates are given for bleeding patients in the immediate postoperative period; however, studies have shown that indiscriminate platelet therapy conferred no therapeutic advantage.⁴⁶ It is in these patients where rapid coagulation testing is required to assist with directed transfusion therapy.⁴⁷ Many institutions now use antifibrinolytics, such as ε-aminocaproic acid and tranexamic acid, at the time of anesthesia induction after several studies have shown that such treatment reduced postoperative bleeding and reoperation. Aprotinin, a protease inhibitor that acts as an antifibrinolytic agent, has been shown to reduce transfusion requirements associated with cardiac surgery.⁴⁸ Desmopressin acetate stimulates release of factor VIII from endothelial cells and may also be effective in reducing blood loss during cardiac surgery. The use of recombinant factor VIIa has also been studied but with conflicting results between improved hemostasis and thrombotic events and mortality, and thus its use is often employed only as a measure of last resort.^45,49

Local Hemostasis

Significant surgical bleeding is usually caused by ineffective local hemostasis. The goal is therefore to prevent further blood loss from a disrupted vessel that has been incised or transected. Hemostasis may be accomplished by interrupting the flow of blood to the involved area or by direct closure of the blood vessel wall defect.

Mechanical Procedures

The oldest mechanical method of bleeding cessation is application of direct digital pressure, either at the site of bleeding or proximally to permit more definitive action. An extremity tourniquet that occludes a major vessel proximal to the bleeding site or the Pringle maneuver for liver bleeding are good examples. Direct digital pressure is very effective and has the advantage of being less traumatic than hemostatic or even “atraumatic” clamps.

When a small vessel is transected, a simple ligature is usually sufficient. However, for larger pulsating arteries, a transfixion suture to prevent slipping is indicated. All sutures represent foreign material, and selection should be based on their intrinsic characteristics and the state of the wound. Direct pressure applied by “packing” a wound with gauze or laparotomy pads affords the best method of controlling diffuse bleeding from large areas, such as in the trauma situation. Packing bone wax on the raw surface to effect pressure can control bleeding from cut bone.

Thermal Agents

Heat achieves hemostasis by denaturation of protein that results in coagulation of large areas of tissue. Electrocautery generates heat by induction from an alternating current source, which is then transmitted via conduction from the instrument directly to the tissue. The amplitude setting should be high enough to produce prompt coagulation, but not so high as to set up an arc between the tissue and the cautery tip. This avoids thermal injury outside of the operative field and also prevents exit of current through electrocardiographic leads, other monitoring devices, or permanent pacemakers or defibrillators. A negative grounding plate should be placed beneath the patient to avoid severe skin burns, and caution should be used with certain anesthetic agents (diethyl ether, divinyl ether, ethyl chloride, ethylene, and cyclopropane) because of the hazard of explosion.

A direct current also can result in hemostasis. Because the protein moieties and cellular elements of blood have a negative surface charge, they are attracted to a positive pole where a thrombus is formed. Direct currents in the 20- to 100-mA range have successfully controlled diffuse bleeding from raw surfaces, as has argon gas.

Topical Hemostatic Agents

Topical hemostatic agents can play an important role in helping to facilitate surgical hemostasis. These agents are classified based on their mechanism of action, and many act at specific stages in the coagulation cascade and take advantage of natural physiologic responses to bleeding.⁵⁰ The ideal topical hemostatic agent has significant hemostatic action, minimal tissue reactivity, nonantigenicity, in vivo biodegradability, ease of sterilization, low cost, and can be tailored to specific needs.⁵¹

In 2010, Achneck et al published a comprehensive overview of absorbable, biologic, and synthetic agents.⁵² Absorbable agents include gelatin foams (Gelfoam), oxidized cellulose (Surgicel), and microfibrillar collagens (Avitene). Both gelatin foam and oxidized cellulose provide a physical matrix for clotting initiation, while microfibrillar collagens facilitate platelet adherence and activation. Biologic agents include topical thrombin, fibrin sealants (FloSeal), and platelet sealants (Vitagel). Human or recombinant thrombin derivatives, which facilitate the formation of fibrin clots and subsequent activation of several clotting factors, take advantage of natural physiologic processes, thereby avoiding foreign body or inflammatory reactions.⁵¹ Caution must be taken in judging vessel caliber in the wound because thrombin entry into larger caliber vessels can result in systemic exposure to thrombin with a risk of disseminated intravascular clotting or death. They are particularly effective in controlling capillary bed bleeding when pressure or ligation is insufficient; however, the bovine derivatives should be used with caution due to the potential immunologic response and worsened coagulopathy. Fibrin sealants are prepared from cryoprecipitate (homologous or synthetic) and have the advantage of not promoting inflammation or tissue necrosis.⁵³ Platelet sealants are a mixture of collagen and thrombin combined with plasma-derived fibrinogen and platelets from the patient, which requires the additional need for centrifugation and processing.

Topical agents are not a substitute for meticulous surgical technique and only function as adjuncts to help facilitate surgical hemostasis. The advantages and disadvantages of each agent must be considered, and use should be limited to the minimum amount necessary to minimize toxicity, adverse reactions, interference with wound healing, and procedural costs.

Background

Human blood replacement therapy was accepted in the late nineteenth century. This was followed by the introduction of blood grouping by Landsteiner who identified the major A, B, and O groups in 1900, resulting in widespread use of blood products in World War I. Levine and Stetson in 1939 followed with the concept of Rh grouping. These breakthroughs established the foundation from which the field of transfusion medicine has grown. Whole blood was considered the standard in transfusion until the late 1970s when component therapy began to take prominence. This change in practice was made possible by the development of improved collection strategies, infectious disease testing, and advances in preservative solutions and storage.

Replacement Therapy

Typing and Cross-Matching

Serologic compatibility for A, B, O, and Rh groups is established routinely. Cross-matching between the donors’ red blood cells and the recipients’ sera (the major cross-match) is performed. Rh-negative recipients should be transfused only with Rh-negative blood. However, this group represents only 15% of the population. Therefore, the administration of Rh-positive blood is acceptable if Rh-negative blood is not available. However, Rh-positive blood should not be transfused to Rh-negative females who are of child-bearing age.

In emergency situations, type O-negative blood may be transfused to all recipients. O-negative and type-specific red blood cells are equally safe for emergency transfusion. Problems are associated with the administration of four or more units of O-negative blood because there is a significant increase in the risk of hemolysis. In patients with clinically significant cold agglutinins, blood should be administered through a blood warmer. If these antibodies are present in high titer, hypothermia is contraindicated.

In patients who have been multiply transfused and who have developed alloantibodies or who have autoimmune hemolytic anemia with pan-red blood cell antibodies, typing and cross-matching is often difficult, and sufficient time should be allotted preoperatively to accumulate blood that might be required during the operation. Cross-matching should always be performed before the administration of dextran because it interferes with the typing procedure.

The use of autologous transfusion is growing. Up to 5 units can be collected for subsequent use during elective procedures. Patients can donate and store their own blood if their hemoglobin concentration exceeds 11 g/dL or if the hematocrit is greater than 34%. The first procurement is performed 40 days before the planned operation, and the last one is performed 3 days before the operation. Donations can be scheduled at intervals of 3 to 4 days. Recombinant human erythropoietin (rHuEPO) accelerates generation of red blood cells and allows for more frequent harvesting of blood.

Banked Whole Blood

Once the gold standard, whole blood is rarely available in Western countries. With sequential changes in storage solutions, the shelf life of red blood cells is now 42 days. Recent evidence has demonstrated that the age of red cells may play a significant role in the inflammatory response and incidence of multiple organ failure.⁵⁴ The changes in the red blood cells that occur during storage include reduction of intracellular ADP and 2,3-diphosphoglycerate (2,3-DPG), which alters the oxygen dissociation curve of hemoglobin, resulting in a decrease in oxygen transport. Stored RBCs progressively becomes acidodic with elevated levels of lactate, potassium, and ammonia.

Red Blood Cells and Frozen Red Blood Cells

Red blood cells are the product of choice for most clinical situations requiring resuscitation. Concentrated suspensions of red blood cells can be prepared by removing most of the supernatant plasma after centrifugation. The preparation reduces but does not eliminate reactions caused by plasma components. Frozen red blood cells are not currently available for use in emergencies, as the thawing and preparation time is measured in hours. They are used for patients who are known to have been previously sensitized. The red blood cell viability is improved, and the ATP and 2,3-DPG concentrations are maintained.

Leukocyte-Reduced and Leukocyte-Reduced/Washed Red Blood Cells

These products are prepared by filtration that removes about 99.9% of the white blood cells and most of the platelets (leukocyte-reduced red blood cells) and, if necessary, by additional saline washing (leukocyte-reduced/washed red blood cells). Leukocyte reduction prevents almost all febrile, nonhemolytic transfusion reactions (fever and/or rigors), alloimmunization to HLA class I antigens, and platelet transfusion refractoriness and cytomegalovirus transmission. In most Western nations, it is the standard red blood cell transfusion product. Supporters of universal leukocyte reduction argue that allogenic transfusion of white cells predisposes to postoperative bacterial infection and multiorgan failure. Reviews of randomized trials and meta-analyses have not provided convincing evidence either way,^55,56 although a large Canadian retrospective study suggests a decrease in mortality and infections.⁵⁷

Platelet Concentrates

The indications for platelet transfusion include thrombocytopenia caused by massive blood loss and replacement with platelet-poor products, thrombocytopenia caused by inadequate production, and qualitative platelet disorders. The shelf life of platelets is 120 hours from time of donation. One unit of platelet concentrate has a volume of approximately 50 mL. Platelet preparations are capable of transmitting infectious diseases and can account for allergic reactions similar to those caused by red blood cell transfusion. A therapeutic level of platelets is in the range of 50,000 to 100,000/μL but is very dependent on the clinical situation. Recent evidence suggests that earlier use of platelets may improve outcomes in bleeding patients.⁵⁸

In rare cases, in patients who become alloimmunized through previous transfusion or patients who are refractory from sensitization through prior pregnancies, HLA-matched platelets can be used.

Fresh Frozen Plasma

Fresh frozen plasma (FFP) prepared from freshly donated blood is the usual source of the vitamin K-dependent factors and is the only source of factor V. FFP carries similar infectious risks as other component therapies. Use of plasma as a primary resuscitation modality in patients who are rapidly bleeding has received attention over the last few years, and ongoing studies are under way to evaluate this concept. FFP can be thawed and stored for up to 5 days, greatly increasing its immediate availability. In an effort to increase the shelf life and avoid the need for refrigeration, lyophilized plasma is being tested. Preliminary animal studies suggest that it preserves the beneficial effects of FFP.⁵⁹

Concentrates and Recombinant DNA Technology

Technologic advancements have made the majority of clotting factors and albumin readily available as concentrates. These products are readily available and carry none of the inherent infectious risks as other component therapies.

Tranexamic Acid

Tranexamic acid (TXA; trade name: Cyklokapron), an antifibrinolytic agent, has been used to decrease bleeding and the need for blood transfusions in coronary artery bypass grafting (CABG), orthotopic liver transplantation, hip and knee arthroplasty, and other surgical settings. The safety and efficacy of using TXA to treat trauma patients was recently evaluated in a large randomized, placebo-controlled clinical trial.⁶⁰ In this trial, 20,211 adult trauma patients in 274 hospitals in 40 countries with significant hemorrhage (heart rate >110 beats per minute and systolic blood pressure <90 mmHg or both) or judged to be at risk for significant hemorrhage were randomized to either TXA or placebo administered as a loading dose of 1 g over 10 minutes followed by an infusion of 1 g over 8 hours. It is important to understand that the responsible physician did not randomize patients with either a clear indication or a clear contraindication to TXA. The overall mortality rate in the cohort studied was 15.3%, of whom 35.3% died on the day of randomization. A total of 1063 patients died due to hemorrhage, and the majority died on the day of randomization. The authors reported that TXA use resulted in a statistically significant reduction in the relative risk (RR) of all-cause mortality of 9% (14.5 vs. 16.0%, RR 0.91, confidence interval [CI] 0.85–0.97; P = .0035). A recent post hoc analysis of the CRASH-2 data showed that the greatest benefit of TXA administration occurred when patients received the medication soon after injury.⁶¹ In this analysis, TXA given between 1 and 3 hours after trauma reduced the risk of death due to bleeding by 21% (147/3037 [4.8%] vs. 184/2996 [6.1%], RR 0.79, CI 0.64–0.97; P = .03). Treatment given after 3 hours increased the risk of death due to bleeding (144/3272 [4.4%] vs. 103/3362 [3.1%], RR 1.44, CI 1.12–1.84; P = .004). Finally, a recent meta-analysis reported that TXA is effective for preventing blood loss in surgery and reducing transfusion and was not associated with increased vascular occlusive events.⁶²

Adverse events associated with TXA use have been reported. These include acute gastrointestinal disturbances (nausea, vomiting, and diarrhea, generally dose-related), visual disturbances (blurry vision and changes in color perception, especially with prolonged use), and occasional thromboembolic events (e.g., deep venous thrombosis and pulmonary embolism, generally observed in the setting of active intravascular clotting). Its use is thus contraindicated in the settings of acquired defective color vision and active intravascular clotting. TXA should be used with caution in the setting of urinary tract bleeding because ureteral obstruction due to clotting has been reported. TXA is contraindicated in patients with aneurysmal subarachnoid hemorrhage; however, there have been no reported complications associated with intra- or extracranial hemorrhage associated with trauma. TXA should not be given with activated prothrombin complex concentrate or factor IX complex concentrates because these may increase the risk of thrombosis.

TXA is an antifibrinolytic that inhibits both plasminogen activation and plasmin activity, thus preventing clot breakdown rather than promoting new clot formation. TXA is an inhibitor of plasminogen activation and an inhibitor of plasmin activity. It occupies the lysine binding sites on plasminogen, thus preventing its binding to lysine residues on fibrin. This reduces plasminogen activation to plasmin. Similarly, blockade of lysine-binding sites on circulating plasmin prevents binding to fibrin and thus prevents clot breakdown. TXA is 10 times more potent in vitro than aminocaproic acid. At therapeutically relevant concentrations, TXA does not affect platelet count or aggregation or coagulation parameters. It is excreted largely unchanged in urine and has a half-life of about 2 hours in circulation. While prolonged use requires that dosing be adjusted for renal impairment, use in the acute trauma situation does not appear to require adjustment. No adjustment is needed for hepatic impairment. Based on the CRASH-2 trial, TXA is becoming more widely used in the United States for patients with ongoing bleeding, especially those with documented evidence of fibrinolysis. Careful analysis of recently ongoing trials will further elucidate the safety profile of this powerful drug.⁶³

Indications for Replacement of Blood and Its Elements

Improvement in Oxygen-Carrying Capacity

Oxygen-carrying capacity is primarily a function of the red blood cells. Thus, transfusion of red blood cells should augment oxygen-carrying capacity. Additionally, hemoglobin is fundamental to arterial oxygen content and thus oxygen delivery. Despite this obvious association, there is little evidence that actually supports the premise that transfusion of red blood cells equates with enhanced cellular delivery and utilization. The reasons for this apparent discrepancy are related to changes that occur with storage of blood. The decrease in 2,3-DPG and P50 impair oxygen offloading, and deformation of the red cells impairs microcirculatory perfusion.⁶⁴

Treatment of Anemia: Transfusion Triggers

A 1988 National Institutes of Health Consensus Report challenged the dictum that a hemoglobin value of less than 10 g/dL or a hematocrit level less than 30% indicates a need for preoperative red blood cell transfusion. This was verified in a prospective randomized controlled trial in critically ill patients that compared a restrictive transfusion threshold to a more liberal strategy and demonstrated that maintaining hemoglobin levels between 7 and 9 g/dL had no adverse effect on mortality. In fact, patients with APACHE II scores of ≤20 or patients age <55 years actually had a lower mortality.⁶⁵

Despite these results, change in daily clinical practice has been slow. Critically ill patients still frequently receive transfusions, with the pretransfusion hemoglobin approaching 9 g/dL in a recent large observational study.⁶⁶ This outdated approach unnecessarily exposes patients to increased risk and little benefit.

One unresolved issue related to transfusion triggers is the safety of maintaining a hemoglobin of 7 g/dL in a patient with ischemic heart disease. Data on this subject are mixed, and many studies have significant design flaws, including their retrospective nature. However, the majority of the published data favors a restrictive transfusion trigger for patients with non-ST elevation acute coronary syndrome, with many reporting worse outcomes in those patients receiving transfusions.^67,68

Volume Replacement

The most common indication for blood transfusion in surgical patients is the replenishment of the blood volume; however, a deficit is difficult to evaluate. Measurements of hemoglobin or hematocrit levels are frequently used to assess blood loss. These measurements can be occasionally misleading in the face of acute loss. Both the amount and the rate of bleeding are factors in the development of signs and symptoms of blood loss.

Loss of blood in the operating room can be roughly evaluated by estimating the amount of blood in the wound and on the drapes, weighing the sponges, and quantifying blood suctioned from the operative field. In patients with normal preoperative values, blood loss up to 20% of total blood volume can be replaced with crystalloid or colloid solutions. Blood loss above this value may require the addition of a balanced resuscitation including red blood cells, FFP, and platelets (detailed later in this chapter) (Table 4-5).

New Concepts in Resuscitation

Traditional resuscitation algorithms are sequentially based on crystalloid followed by red blood cells and then plasma and platelet transfusions and have been in widespread use since the 1970s. No quality clinical data supported this concept. Recently the damage control resuscitation (DCR) strategy, aimed at halting and/or preventing rather than treating the lethal triad of coagulopathy, acidosis, and hypothermia, has challenged traditional thinking on early resuscitation strategies.⁶⁹

Rationale

In civilian trauma systems, nearly half of all deaths happen before a patient reaches the hospital, and many are nonpreventable.⁷⁰ Patients who survive to an emergency center have a high incidence of truncal hemorrhage, and deaths in this group of patients may be potentially preventable. Truncal hemorrhage patients in shock often present with the early coagulopathy of trauma in the emergency department and are at significant risk of dying.^71,72,73

Many of these patients have suffered substantial bleeding and may receive a significant transfusion, generally defined as the administration of ≥4 to 6 units of red blood cells within 4 to 6 hours of admission. This definition is admittedly arbitrary. Although 25% of all trauma admissions receive a unit of blood early after admission, only a small percentage of patients receive a massive transfusion. In the military setting, the percentage of massive transfusion patients almost doubles.⁷⁴

Damage Control Resuscitation

Standard advanced trauma life support guidelines start resuscitation with crystalloid, followed by packed red blood cells.⁷⁵ Only after several liters of crystalloid have been transfused does transfusion of units of plasma or platelets begin. This conventional massive transfusion practice was based on a several small uncontrolled retrospective studies that used blood products containing increased amounts of plasma, which are no longer available.⁷⁶ Because of the known early coagulopathy of trauma, the current approach to managing the exsanguinating patient involves early implementation of damage control resuscitation (DCR). Although most of the attention to hemorrhagic shock resuscitation has centered on higher ratios of plasma and platelets, DCR is actually composed of three basic components: permissive hypotension, minimizing crystalloid-based resuscitation, and the immediate release and administration of predefined blood products (red blood cells, plasma, and platelets) in ratios similar to those of whole blood.

In Iraq and Afghanistan, DCR practices are demonstrating unprecedented success with improved overall survival.⁷⁷ Civilian data also suggest that a balanced resuscitation approach yields improved outcome in severely injured and bleeding trauma patients.⁶⁹ To verify military and single-institution civilian data on DCR, a multicenter retrospective study of modern transfusion practice at 17 leading civilian trauma centers was performed.⁷⁸ It was found that plasma:platelet:red blood cell ratios varied from 1:1:1 to 0.3:0.1:1, with corresponding survival rates ranging from 71% to 41%. A significant center effect was seen, documenting wide variation in both transfusion practice and outcomes between Level 1 trauma centers. This variation correlated with blood product ratios. Increased plasma- and platelet-to-RBC ratios significantly decreased truncal hemorrhagic death and 30-day mortality without a concomitant increase in multiple organ failure as a cause of death. A prospective observational study evaluating current transfusion practice at 10 Level 1 centers was recently published, again documenting the wide variability in practice and improved outcomes with earlier use of increased ratios of plasma and platelets.⁷⁹ Patients receiving ratios less than 1:2 were four times more likely to die than patients with ratios of 1:1 or higher.

Regardless of the optimal ratio, it is essential that the trauma center has an established mechanism to deliver these products quickly and in the correct amounts to these critically injured patients. In fact, several authors have shown that a well-developed massive transfusion protocol is associated with improved outcomes independent of the ratios chosen.⁸⁰ This aggressive delivery of predefined blood products should begin prior to any laboratory-defined anemia or coagulopathy.

An example of an adult massive transfusion clinical guideline specifying the early use of component therapy is shown in Table 4-6. Specific recommendations for the administration of component therapy during a massive transfusion are shown in Table 4-7. Because only a small percentage of trauma patients require a massive transfusion and because blood products in general are in short supply, the need for early prediction models has been studied and a comparison of results from both civilian and military studies is shown in Table 4-8.^{81,82,83,84,85} While compelling, none of these algorithms have been prospectively validated.

Complications of Transfusion

(Table 4-9)

Transfusion-related complications are primarily related to blood-induced proinflammatory responses. Transfusion-related events are estimated to occur in approximately 10% of all transfusions, but less than 0.5% are serious in nature. Transfusion-related deaths, although rare, do occur and are related primarily to transfusion-related acute lung injury (TRALI) (16%–22%), ABO hemolytic transfusion reactions (12%–15%), and bacterial contamination of platelets (11%–18%).⁸⁶

Nonhemolytic Reactions

Febrile, nonhemolytic reactions are defined as an increase in temperature (>1°C) associated with a transfusion and are fairly common (approximately 1% of all transfusions). Preformed cytokines in donated blood and recipient antibodies reacting with donated antibodies are postulated etiologies. The incidence of febrile reactions can be greatly reduced by the use of leukocyte-reduced blood products. Pretreatment with acetaminophen reduces the severity of the reaction.

Bacterial contamination of infused blood is rare. Gram-negative organisms, which are capable of growth at 4°C, are the most common cause. Most cases, however, are associated with the administration of platelets that are stored at 20°C or, even more commonly, with apheresis platelets stored at room temperature. Cases from FFP thawed in contaminated water baths have also been reported.⁸⁷ Bacterial contamination can result in sepsis and death in up 25% of patients.⁸⁸ Clinical manifestations includes systemic signs such as fever and chills, tachycardia and hypotension, and gastrointestinal symptoms (abdominal cramps, vomiting, and diarrhea). If the diagnosis is suspected, the transfusion should be discontinued and the blood cultured. Emergency treatment includes oxygen, adrenergic blocking agents, and antibiotics.

Allergic Reactions

Allergic reactions are relatively frequent, occurring in about 1% of all transfusions. Reactions are usually mild and consist of rash, urticaria, and flushing. In rare instances, anaphylactic shock develops. Allergic reactions are caused by the transfusion of antibodies from hypersensitive donors or the transfusion of antigens to which the recipient is hypersensitive. Allergic reactions can occur after the administration of any blood product but are commonly associated with FFP and platelets. Treatment and prophylaxis consist of the administration of antihistamines. In more serious cases, epinephrine or steroids may be indicated.

Respiratory Complications

Respiratory compromise may be associated with transfusion-associated circulatory overload (TACO), which is an avoidable complication. It can occur with rapid infusion of blood, plasma expanders, and crystalloids, particularly in older patients with underlying heart disease. Central venous pressure monitoring should be considered whenever large amounts of fluid are administered. Overload is manifest by a rise in venous pressure, dyspnea, and cough. Rales can generally be heard at the lung bases. Treatment consists of diuresis, slowing the rate of blood administration, and minimizing fluids while blood products are being transfused.

The syndrome of TRALI is defined as noncardiogenic pulmonary edema related to transfusion.⁸⁹ It can occur with the administration of any plasma-containing blood product. Symptoms are similar to circulatory overload with dyspnea and associated hypoxemia. However, TRALI is characterized as noncardiogenic and is often accompanied by fever, rigors, and bilateral pulmonary infiltrates on chest x-ray. It most commonly occurs within 1 to 2 hours after the onset of transfusion but virtually always before 6 hours. Toy et al recently reported a decrease in the incidence of TRALI with the reduction transfusion of plasma from female donors, due to a combination of reduced transfusion of strong cognate HLA class II antibodies and HNA antibodies in patients with risk factors for acute lung injury.⁹⁰ Treatment of TRALI entails discontinuation of any transfusion, notification of the transfusion service, and pulmonary support, which may vary from supplemental oxygen to mechanical ventilation.

Hemolytic Reactions

Hemolytic reactions can be classified as either acute of delayed. Acute hemolytic reactions occur with the administration of ABO-incompatible blood and can be fatal in up to 6% of cases. Contributing factors include errors in the laboratory of a technical or clerical nature or the administration of the wrong blood type. Immediate hemolytic reactions are characterized by intravascular destruction of red blood cells and consequent hemoglobinemia and hemoglobinuria. DIC can be initiated by antibody-antigen complexes activating factor XII and complement, leading to activation of the coagulation cascade. Finally, acute renal insufficiency results from the toxicity associated with free hemoglobin in the plasma, resulting in tubular necrosis and precipitation of hemoglobin within the tubules.

Delayed hemolytic transfusion reactions occur 2 to 10 days after transfusion and are characterized by extravascular hemolysis, mild anemia, and indirect (unconjugated) hyperbilirubinemia. They occur when an individual has a low antibody titer at the time of transfusion, but the titer increases after transfusion as a result of an anamnestic response. Reactions to non-ABO antigens involve immunoglobulin G-mediated clearance by the reticuloendothelial system.

If the patient is awake, the most common symptoms of acute transfusion reactions are pain at the site of transfusion, facial flushing, and back and chest pain. Associated symptoms include fever, respiratory distress, hypotension, and tachycardia. In anesthetized patients, diffuse bleeding and hypotension are the hallmarks. A high index of suspicion is needed to make the diagnosis. The laboratory criteria for a transfusion reaction are hemoglobinuria and serologic criteria that show incompatibility of the donor and recipient blood. A positive Coombs’ test indicates transfused cells coated with patient antibody and is diagnostic. Delayed hemolytic transfusions may also be manifest by fever and recurrent anemia. Jaundice and decreased haptoglobin usually occur, and low-grade hemoglobinemia and hemoglobinuria may be seen. The Coombs’ test is usually positive, and the blood bank must identify the antigen to prevent subsequent reactions.

If an immediate hemolytic transfusion reaction is suspected, the transfusion should be stopped immediately, and a sample of the recipient’s blood drawn and sent along with the suspected unit to the blood bank for comparison with the pretransfusion samples. Urine output should be monitored and adequate hydration maintained to prevent precipitation of hemoglobin within the tubules. Delayed hemolytic transfusion reactions do not usually require specific intervention.

Transmission of Disease

Malaria, Chagas’ disease, brucellosis, and, very rarely, syphilis are among the diseases that have been transmitted by transfusion. Malaria can be transmitted by all blood components. The species most commonly implicated is Plasmodium malariae. The incubation period ranges from 8 to 100 days; the initial manifestations are shaking chills and spiking fever. Cytomegalovirus (CMV) infection resembling infectious mononucleosis also has occurred.

Transmission of hepatitis C and HIV-1 has been dramatically minimized by the introduction of better antibody and nucleic acid screening for these pathogens. The residual risk among allogeneic donations is now estimated to be less than 1 per 1,000,000 donations. The residual risk of hepatitis B is approximately 1 per 300,000 donations.⁹¹ Hepatitis A is very rarely transmitted because there is no asymptomatic carrier state. Improved donor selection and testing are responsible for the decreased rates of transmission. Recent concerns about the rare transmission of these and other pathogens, such as West Nile virus, are being addressed by current trials of “pathogen inactivation systems” that reduce infectious levels of all viruses and bacteria known to be transmittable by transfusion. Prion disorders (e.g., Creutzfeldt-Jakob disease) also are transmissible by transfusion, but there is currently no information on inactivation of prions in blood products for transfusion.

The initial approach to assessing hemostatic function is a careful review of the patient’s clinical history (including previous abnormal bleeding or bruising), drug use, and basic laboratory testing. Common screening laboratory testing includes platelet count, PT or INR, and aPTT. Platelet dysfunction can occur at either extreme of platelet count. The normal platelet count ranges from 150,000 to 400,000/μL. Whereas a platelet count greater than 1,000,000/μL may be associated with bleeding or thrombotic complications, increased bleeding complications may be observed with major surgical procedures when the platelets are below 50,000/μL and with minor surgical procedures when counts are below 30,000/μL, and spontaneous hemorrhage can occur when the counts fall below 20,000/μL. Despite a lack of evidence supporting their use, platelet transfusions are still recommended in ophthalmologic and neurosurgical procedures when the platelet count is less than 100,000/μL.

The PT and aPTT are variations of plasma recalcification times initiated by the addition of a thromboplastic agent. The PT reagent contains thromboplastin and calcium that, when added to plasma, leads to the formation of a fibrin clot. The PT test measures the function of factors I, II, V, VII, and X. Factor VII is part of the extrinsic pathway, and the remaining factors are part of the common pathway. Factor VII has the shortest half-life of the coagulation factors, and its synthesis is vitamin K dependent. The PT test is best suited to detect abnormal coagulation caused by vitamin K deficiencies and warfarin therapy.

Due to variations in thromboplastin activity, it can be difficult to accurately assess the degree of anticoagulation on the basis of PT alone. To account for these variations, the INR is now the method of choice for reporting PT values. The International Sensitivity Index (ISI) is unique to each batch of thromboplastin and is furnished by the manufacturer to the hematology laboratory. Human brain thromboplastin has an ISI of 1, and the optimal reagent has an ISI between 1.3 and 1.5.

The INR is a calculated number derived from the following equation:

INR = (measured PT/normal PT)^ISI

The aPTT reagent contains a phospholipid substitute, activator, and calcium, which in the presence of plasma leads to fibrin clot formation. The aPTT measures function of factors I, II, and V of the common pathway and factors VIII, IX, X, and XII of the intrinsic pathway. Heparin therapy is often monitored by following aPTT values with a therapeutic target range of 1.5 to 2.5 times the control value (approximately 50 to 80 seconds). Low molecular weight heparins are selective Xa inhibitors that may mildly elevate the aPTT, but therapeutic monitoring is not routinely recommended.

The bleeding time is used to evaluate platelet and vascular dysfunction, although not as frequently as in the past. Several standard methods have been described; however, the Ivy bleeding time is most commonly used. It is conducted by placing a sphygmomanometer on the upper arm and inflating it to 40 mmHg, and then a 5-mm stab incision is made on the flexor surface of the forearm. The time is measured to cessation of bleeding, and the upper limit or normal bleeding time with the Ivy test is 7 minutes. A template aids in administering a uniform test and adds to the reproducibility of the results. An abnormal bleeding time suggests platelet dysfunction (intrinsic or drug-induced), vWD, or certain vascular defects. Many laboratories are replacing the template bleeding time with an in vitro test in which blood is sucked through a capillary and the platelets adhere to the walls of the capillary and aggregate. The closure time in this system appears to be more reproducible than the bleeding time and also correlates with bleeding in vWD, primary platelet function disorders, and patients who are taking aspirin.

Additional medications may significantly impair hemostatic function, such as antiplatelet agents (clopidogrel and GP IIb/IIIa inhibitors), anticoagulant agents (hirudin, chondroitin sulfate, dermatan sulfate), and thrombolytic agents (streptokinase, tPA). If abnormalities in any of the coagulation studies cannot be explained by known medications, congenital abnormalities of coagulation or comorbid disease should be considered.

Unfortunately, while these conventional tests (PT, aPTT) capture the classic intrinsic and extrinsic coagulation cascade, they do not reflect the complexity of in vivo coagulation.⁹² Although they are useful to follow warfarin and heparin therapies, they poorly reflect the status of actively bleeding patients. This is not surprising given that these tests use only plasma and not whole blood to provide their assessment of the patient’s clotting status. To better assess the complex and rapidly changing interactions of an actively bleeding patient, many centers have moved to whole blood-viscoelastic testing such as TEG or rotational thromboelastometry (ROTEM). In addition, some centers have demonstrated that the graphical display options allow for more rapid return of results and that these tests are actually less expensive than standard coagulation panels.

TEG was originally described by Hartert in 1948.⁹³ Continuous improvements in this technique have made this test a valuable tool for the medical personnel interested in coagulation. The TEG monitors hemostasis as a dynamic process rather than revealing information of isolated conventional coagulation screens.⁹⁴ The TEG measures the viscoelastic properties of blood as it is induced to clot under a low-shear environment (resembling sluggish venous flow). The patterns of change in shear-elasticity enable the determination of the kinetics of clot formation and growth as well as the strength and stability of the formed clot. The strength and stability provide information about the ability of the clot to perform the work of hemostasis, while the kinetics determines the adequacy of quantitative factors available for clot formation. A sample of celite-activated whole blood is placed into a prewarmed cuvette, and the clotting process is activated with kaolin with standard TEG and kaolin plus tissue factor with rapid TEG. A suspended piston is then lowered into the cuvette that moves in rotation of a 4.5-degree arc backward and forward. The normal clot goes through acceleration and strengthening phase. The fiber strands that interact with activated platelets attach to the surface of the cuvette and the suspended piston. The clot forming in the cuvette transmits its movement onto the suspended piston. A “weak” clot stretches and therefore delays the arc movement of the piston, which is graphically expressed as a narrow TEG. A strong clot, in contrast, will move the piston simultaneously and proportionally to the cuvette’s movements, creating a thick TEG. The strength of a clot is graphically represented over time as a characteristic cigar-shape figure (Fig. 4-6).

Several parameters are generated from the TEG tracing. The r-value (reaction time) represents the time between the start of the assay and initial clot formation. This reflects clotting factor activity and initial fibrin formation and is increased with factor deficiency or severe hemodilution. The k-time (clot kinetics) is the time needed to reach specified clot strength and represents the interactions of clotting factors and platelets. As such, the k-time is prolonged with hypofibrinogenemia and significant factor deficiency. Prolonged r-value and k-time are commonly addressed with plasma transfusions. The alpha or angle (∝) is the slope of the tracing and reflects clot acceleration. The angle reflects the interactions of clotting factors and platelets. The slope is decreased with hypofibrinogenemia and platelet dysfunction. Decreased angles are treated with cryoprecipitate transfusion or fibrinogen administration. The maximal amplitude (mA) is the greatest height of the tracing and represents clot strength. Its height is reduced with dysfunction or deficiencies in platelets or fibrinogen. Decreased mA is addressed with platelet transfusion and, in cases where the angle is also decreased, with cryoprecipitate (or fibrinogen) as well. The G-value is a parametric measure derived from the mA value and reflects overall clot strength or firmness. An increased G-value is associated with hypercoagulability, whereas a decrease is seen with hypocoagulable states. Finally, the LY30 is the amount of lysis occurring in the clot, and the value is the percentage of amplitude reduction at 30 minutes after mA is achieved. The LY30 represents clot stability and when increased fibrinolysis is present.

TEG is the only test measuring all dynamic steps of clot formation until eventual clot lysis or retraction. TEG has also been shown to identify on admission those patients likely to develop thromboembolic complications after injury and postoperatively.^95,96,97

Recent trauma data have shown TEG to be useful in predicting early transfusion of red blood cells, plasma, platelets, and cryoprecipitate.⁹⁸ TEG can also predict the need for life-saving interventions shortly after arrival and to predict 24-hour and 30-day mortality.⁹⁹ Lastly, TEG can be useful to guide administration of TXA to injured patients with hyperfibrinolysis.¹⁰⁰ Our center now uses TEG rather than PT and a PTT to evaluate injured patients in the emergency room.¹⁰¹

Excessive bleeding during or after a surgical procedure may be the result of ineffective hemostasis, blood transfusion, undetected hemostatic defect, consumptive coagulopathy, and/or fibrinolysis. Excessive bleeding from the operative field unassociated with bleeding from other sites usually suggests inadequate mechanical hemostasis.

Massive blood transfusion is a well-known cause of thrombocytopenia. Bleeding following massive transfusion can occur due to hypothermia, dilutional coagulopathy, platelet dysfunction, fibrinolysis, or hypofibrinogenemia. Another cause of hemostatic failure related to the administration of blood is a hemolytic transfusion reaction. The first sign of a transfusion reaction may be diffuse bleeding. The pathogenesis of this bleeding is thought to be related to the release of ADP from hemolyzed red blood cells, resulting in diffuse platelet aggregation, after which the platelet clumps are removed out of the circulation.

Transfusion purpura occurs when the donor platelets are of the uncommon PlA¹ group. This is an uncommon cause of thrombocytopenia and associated bleeding after transfusion. The platelets sensitize the recipient, who makes antibody to the foreign platelet antigen. The foreign platelet antigen does not completely disappear from the recipient circulation but attaches to the recipient’s own platelets. The antibody then destroys the recipient’s own platelets. The resultant thrombocytopenia and bleeding may continue for several weeks. This uncommon cause of thrombocytopenia should be considered if bleeding follows transfusion by 5 or 6 days. Platelet transfusions are of little help in the management of this syndrome because the new donor platelets usually are subject to the binding of antigen and damage from the antibody. Corticosteroids may be of some help in reducing the bleeding tendency. Posttransfusion purpura is self-limited, and the passage of several weeks inevitably leads to subsidence of the problem.

DIC is characterized by systemic activation of the coagulation system, which results in the deposition of fibrin clots and microvascular ischemia and may contribute to the development of multiorgan failure. Consumption and subsequent exhaustion of coagulation proteins and platelets due to the ongoing activation of the coagulation system may induce severe bleeding complications.

Lastly, severe hemorrhagic disorders due to thrombocytopenia have occurred as a result of gram-negative sepsis. Defibrination and hemostatic failure also may occur with meningococcemia, Clostridium perfringens sepsis, and staphylococcal sepsis. Hemolysis appears to be one mechanism in sepsis leading to defibrination.