Clinical Anesthesia

Chapter 16

Hemostasis and Transfusion Medicine

John C. Drummond

Charise T. Petrovitch

Thomas A. Lane

Key Points

1. In terms of transfusion-transmitted infectious diseases, the American blood supply has never been safer than it is today.

2. Clerical and patient identification errors are the most common causes of ABO incompatibility and a large fraction of these errors typically occur in the operating theater.

3. The three leading causes of transfusion-related death in the United States are ABO incompatibility, transfusion-related acute lung injury, and sepsis caused by bacterial infections.

4. In the setting of massive transfusion, assuming maintenance of isovolemia and the absence of a consumptive coagulopathy, critical dilution of clotting factors and platelets is likely to occur after an average replacement of 140% and 230% of blood volume, respectively.

5. With the possible exception of trauma resuscitation, coagulation factor and platelet replacement should be determined by laboratory assessment and/or observation of clinical coagulopathy and not estimated blood loss-driven formulas.

6. The red blood cell (RBC) transfusion “trigger” for most patients will lie between hemoglobin values of 7 and 10 g/dL.

7. Platelet administration thresholds relevant to anesthesiologists usually will lie between 50,000 and 100,000/uL.

8. Normal coagulation can be achieved with clotting factor levels of 20 to 30% of normal. Those levels can usually be achieved by administration of 10 to 15 mL/kg of fresh-frozen plasma.

9. Recipient or donor unit identification errors will result in an acute hemolytic transfusion reaction for one of every three packed RBC (PRBC) units mistransfused.

10.     A patient who has received 10 to 12 units of group O RBCs should not be switched back to his or her own ABO group until testing has been performed to confirm that significant titers of anti-A or anti-B antibodies are not present.

11.     The classic, dual-cascade (intrinsic and extrinsic pathway) model of coagulation is an inadequate representation of coagulation, as it occurs in vivo.

12.     In vivo, coagulation is initiated principally by contact of factor VII with extravascular tissue factor leading first to platelet activation followed by the generation of large amounts of thrombin by activated clotting factors acting on the phospholipid surface provided by activated platelets.

13.     Under normal conditions, plasmin is generated only at the site of clot formation and is destroyed rapidly once released into the circulation. This localization process fails at times of accelerated fibrinolysis (disseminated intravascular coagulation, primary fibrinolysis).

14.     von Willebrand disease is the most common hereditary bleeding disorder. Some form of the disease, which may be subclinical prior to surgery, is present in approximately 1% of the population.


15.     Factors II, VII, IX, and X and proteins C and S depend on vitamin K for their synthesis. Vitamin K deficiency occurs frequently in hospitalized patients because of dietary insufficiency, gut sterilization, and malabsorption. A high index of suspicion for vitamin K deficiency should be maintained.

16.     As many as 5% of patients who receive heparin therapy for 5 days will develop heparin-induced thrombocytopenia/ thrombosis. The clinical manifestations are more often the result of thrombosis and thromboembolism than thrombocytopenia.

In the year 2004, 14.2 million units of packed red blood cells (PRBCs), 9.9 million units of platelets (84% of which were apheresis units), and 4.1 million units of fresh-frozen plasma (FFP) were administered in the United States.1 At the University of California, San Diego (UCSD) Medical Center, approximately 40% of all transfused units are administered to surgical patients by anesthesia personnel. An extrapolation of these numbers suggests that anesthesia providers may be involved in as many as 8.3 million donor unit exposures per year in the United States. Accordingly, no specialty group, save those who collect, process, and deliver blood products, has a greater incentive to have a broad grasp of the principles of transfusion medicine.

This chapter begins with a review of the risks associated with the administration of blood products, followed by a discussion of the factors that determine the necessity for the administration of the three most commonly used components, RBCs, FFP, and platelets, and then a discussion of conservation techniques for minimizing the necessity for transfusion. The remainder of the chapter presents a description of the preparation of blood products, a discussion of the physiology of hemostasis, a description of tests of the hemostatic mechanism, and finally a review of common bleeding disorders, including a discussion of the effects of pharmacologic agents on hemostasis.

The Risks of Blood Product Administration

The recognition that the human immunodeficiency virus (HIV) is transmissible by blood generated public fear of transfusion and led to dramatic changes in transfusion practices (seeChapter 13). While the transfusion-related transmission of HIV is now vanishingly rare, there remain numerous other hazards associated with blood products. The risks can be subdivided into those of infectious and noninfectious etiologies. Transfusion-transmissible infections, in particular viral infections, have generated the greatest concern and will therefore be addressed first. However, in reality, the morbidity and mortality associated with nonviral hazards are far greater concerns.

Infectious Risks Associated with Blood Product Administration

The potentially transmittable diseases/agents in blood are numerous. They include several viruses (hepatitis A, B, C, D, E), the human T-cell lymphotropic viruses (HTLV-1, HTLV-2), the human immunodeficiency viruses 1 and 2, cytomegalovirus (CMV), West Nile virus (WNV), the Epstein-Barr virus, human herpes virus-8 (the agent of Kaposi sarcoma), parvovirus B19, the GBV-C virus (also called hepatitis G), transfusion-transmitted virus, and the SEN virus. Other diseases/agents include prions (Creutzfeldt-Jakob disease [CJD] and variant Creutzfeldt-Jakob disease [vCJD]), Lyme disease, contaminating bacteria, parasites (malaria, Chagas' disease, ehrlichiosis, babesiosis), and syphilis.2,3 Several of these will not be considered further. Although GBV-C, transfusion-transmitted virus, and SEN virus are transmitted by transfusion, they do not appear to cause clinical disease; the rate of transmission of parvovirus B19 is very low and clinical disease is extremely infrequent2; there have been no reported instances of transfusion-transmitted Lyme disease and only one instance of ehrlichiosis.4

Estimates of the frequency of infectious agents in the North American blood supply are presented in Table 16-1. The rate of viral infectivity has decreased dramatically in the last 2 decades. The advent of universal (in the United States) nucleic


acid testing (NAT) for HIV and hepatitis C (HCV) has reduced the frequency of transmission of those agents to very low levels, approximately 1 in 1.7 million units transfused. Hepatitis B (HBV) remains the greatest risk, currently with about 1/269,000 donor exposures.5 All of these estimates are derived from the observed rates of seropositivity among donors and the statistical likelihood of administration of blood from donors whose infection is in the window period between contracting the virus and detectability by the available assays. The window periods from infection to detection by 16-unit minipool NAT testing for HIV and HCV are estimated to be 9 and 7.4 days, respectively.6 Minipool testing entails analysis of pooled aliquots from 6 to 24 units. It is estimated that the more expensive individual donor testing could reduce the window period for HIV and HCV to 5.6 and 4.9 days, respectively.6 For HBV, using hepatitis B surface antigen testing, the window period is 38 days.7 A NAT test for HBV is available. However, when performed on minipools rather than on individual donations, it will probably add little to the detection sensitivity achieved with the combination of hepatitis B surface antigen and the anti-hepatitis B core antigen.7,8,9

Table 16-1 Estimates of The Rate (Per Donor Exposure) of Transfusion-Transmitted Infectious Disease in North America



• Hepatitis B (HBV)


•Hepatitis C (HCV)


•Human immunodeficiency virus (HIV)


•Human T-cell lymphotrophic virus (HTLV)


•West Nile Virus (WNV)

Indeterminate/very low

• Cytomegalovirus (CMV)—Nonleukoreduced random donor


   •Leukoreduced random donor


   •CMV seronegative donor


• Epstein-Barr virus (EBV)


•Chagas'; malaria; other parasites

< 1/1,000,000

Bacterial sepsis


• Platelets (apheresis, culture tested)


•Platelets (whole-blood derived, surrogate tested)


•Platelets (untested)


Data derived from several sources (Refs. 5, 7, 14, 17, 18, 19, 52, and 276).

In terms of transfusion-transmitted infectious diseases, the North American blood supply has never been safer than it is today.

Hepatitis C

The significance of HCV is that, despite its commonly mild initial presentation, it progresses to a chronic state in 85% with significant associated morbidity and mortality of patients. Twenty percent of chronic carriers develop cirrhosis and 1 to 5% develop hepatocellular carcinoma.10,11

Hepatitis B

It is estimated that only 35% of HBV-exposed patients will develop acute disease,12 although approximately 1% will develop fulminant acute hepatitis. In approximately 85% of patients, the disease resolves spontaneously, 9% develop chronic persistent hepatitis, 3% develop chronic active hepatitis, 1% develop cirrhosis with or without chronic active hepatitis, and 1% develop hepatocellular carcinoma.

Hepatitis A

Transmission of hepatitis A virus (HAV) by transfusion has been very rare. Blood banks screen for HAV by history only and there is no carrier state for this virus. The infectious period is limited to 1 to 2 weeks. The diagnosis depends on hepatitis antibody seroconversion.

Human Immunodeficiency Virus

HIV is a retrovirus, so called because its propagation requires translation of RNA to DNA. Current screening tests are directed at both HIV-1 and HIV-2, though the latter has been an extremely infrequent cause of human disease. The incidence of transfusion-related HIV infection has decreased dramatically. As a testimonial to the effectiveness of our blood delivery system's response to the emergence of HIV, the risk of transfusion-related transmission, which was approximately 1:100 in the early 1980s and 1:400,000 in 1997,13 is currently approximately 1 per 1.7 million donor exposures.

Human T-Cell Lymphotropic Virus

HTLV-1 and HTLV-2 belong to the same retrovirus family as HIV. The incidence of clinical disease resulting from transmitted virus appears to be very low. They are associated with T-cell leukemia and lymphoma rather than the generalized immunodeficiency of the acquired immune deficiency syndrome (AIDS). In the United States, all donor units are screened for the presence of antibody to HTLV-I and HTLV-2.


Transfusion-associated CMV infections are usually benign and self-limited. However, CMV may cause serious, even fatal, infections in immunocompromised patients. Patients at risk include premature neonates, CMV-seronegative bone marrow transplant recipients, pregnant females, and those patients with severely depressed immune function. Leukoreduction and/or the use of blood from CMV-seronegative donors reduce, but do not prevent, CMV transmission.14 Restriction of immunocompromised patients to leukoreduced blood from CMV-seronegative donors is standard in many centers.

West Nile Virus

WNV is a mosquito-borne flavivirus (as is dengue fever) that became epidemic in midwestern states in 2002 and has since occurred nationwide. Although the majority of infected individuals are either asymptomatic or develop only a mild illness, encephalitis/meningitis can occur and the death rate among confirmed cases is 5 to 10%.15,16 Transmission by transfusion and organ transplantation has been confirmed. Fortunately, the window period between infection and clinical symptoms is short, approximately 3 days, and the period of infectivity also appears to be relatively brief. Universal minipool NAT testing for WNV began in 2003, with discretionary selective individual donor NAT testing in areas of high incidence.17 Transfusion transmission has subsequently been very infrequent.17

Parasitic Diseases

Transfusion-transmitted malaria is relatively common in regions where the disease is endemic but has been rare in the United States.3 Because the parasite resides within the red cell, the hazard is associated almost exclusively with RBC transfusion. Chagas' disease is caused by a protozoan (Trypanosoma cruzi) that is endemic to South and Central America (including Mexico). Significant clinical disease has been rare in North America and has occurred almost exclusively in immunocompromised transfusion recipients. Donor screening for Chagas' disease by immunoassay is now standard in all Red Cross collection centers and is increasing among other U.S. agencies, especially in the southwest and Florida.3

Bacterial Contamination of Blood Components

Bacterial contamination occurs at a much higher frequency (Table 16-1) than any of the other infections discussed in this section and is associated with substantial mortality.18,19 The incidence of sepsis is substantially greater with platelet than RBC administration because the former are stored at room temperature. The risk is less with apheresis platelets (obtained from a single donor with one venipuncture) than with whole blood-derived platelet administration, which entails pools derived from six to ten separate donor units. The source of the bacteria can be donor skin flora, donor bacteremia, or contaminants introduced during collection, processing, and storage. Numerous Gram-positive and -negative organisms can occur in platelets including Staphylococcus aureusKlebsiella pneumoniaeSerratia marcescens, and Staphylococcus epidermidis.12 Only a limited number of bacteria, including Yersinia enterocolitica and certain Serratia and Pseudomonas species can grow at RBC storage temperatures.2 Fatal sepsis is usually the result of Gram-negative organisms, and Y. enterocolitica is the most frequently implicated.

There is considerable current attention being given to the prevention of platelet-transmitted bacterial infection. Careful skin preparation is the norm and some collection centers divert


and discard the first few milliliters of the draw. In 2004, bacterial testing of all platelets became a requirement for achieving AABB (formerly, the American Association of Blood Banks) certification, and the majority of agencies now culture apheresis units. However, culture is not practical for whole blood-derived pools, and less sensitive surrogate methods (based on measurements of pH, glucose, PO2, or assay for bacterial RNA20) are employed. These measures have reduced but not eliminated transfusion of contaminated units (Table 16-1).18,19,21

The patient who receives contaminated blood transfusion will rapidly experience some combination of fever, chills, tachycardia, dyspnea, emesis, shock, and may develop disseminated intravascular coagulation (DIC) and acute renal failure. The reactions are variable in severity, and an index of suspicion should be maintained in order to distinguish these reactions from other major and minor transfusion reactions. The transfusion should be stopped immediately, blood cultures obtained, and the patient treated with broad-spectrum antibiotics. The blood bank should be notified immediately in order that it may interdict administration of other components made from the same donation and perform diagnostic testing (Gram stain and unit culture).

Prion-Related Diseases

Prions are the causative agents of CJD and vCJD. The latter is the human disease caused by the agent responsible for bovine spongiform encephalitis. All three are fatal, degenerative neurologic diseases caused by an abnormally folded variant of a protein that is constitutively present. Since the emergence of bovine spongiform encephalitis in England in 1984, approximately 200 cases of vCJD had been reported, with the large majority occurring in the United Kingdom.22 The risk of transfusion-related vCJD is undefined. CJD has never been known to have been transmitted by transfusion but there have been three reported cases of apparently transfusion-related vCJD.22 The incubation period of vCJD may be as long as 6 years. Accordingly, the true transmission rate may as yet be underrecognized. NAT testing is not feasible (prions have no nucleic acids) and there are no known antigenic or immune response markers. Therefore, it must be hoped that changes in animal husbandry practices combined with exclusion of donors who have spent time in high-risk areas will minimize whatever risk exists.

Other Infectious Risks

Many additional microbial agents can be transmitted by blood components. They include BorreliaBabesia, dengue, the viral agent of severe acute respiratory syndrome, and other herpes viruses. Transmission of these agents is apparently extremely rare. However, the recent experience with WNV reveals, once again, the potential for new agents to become a sudden threat to the blood supply and serves as a reminder of the continuing need to administer blood components only when absolutely indicated.

Noninfectious Risks Associated with Blood Product Administration

The noninfectious risks associated with blood product administration, the majority of which are immunologically mediated, and their approximate incidences are presented in Table 16-2.

Table 16-2 The Noninfectious Adverse Reactions Associated with Blood Product Administration, in the Approximate Order of Their Average Frequencies in the Published Literaturea







• Inflammatory response

(?) 100%

Increases with duration of storage

• Alloimmunization


   • RBCs



   • Plts


Reduced by leukoreduction25

•  Minor allergic reactions (urticaria, flushing)


Plts and FFP > RBCs

•  Febrile reactions


Probably reduced by leukoreduction



Most often Kell, Kidd, and Rhesus (E) antibodies



All plasma containing products;
FFP and Plts > PRBCs

•  Anaphylactic/toid reactions


Plts > PRBCs
IgA deficiency increases risk



Usually patient ID error; 2% mortality;
Plasma-incompatible Plts are a rare cause



Immunocompromised patients, especially marrow transplant recipients

TRIM, transfusion-related immunomodulation; RBCs, red blood cells; Plts, platelets; FFP, fresh-frozen plasma; DHTR, delayed hemolytic transfusion reaction; TRALI, transfusion-related acute lung injury; PRBCs, packed RBCs; IgA, immunoglobulin A; AHTR, acute hemolytic transfusion reaction; ID, identification; GVHD, graft-versus-host disease.

aThe frequencies are presented as percentages when >0.1% and otherwise as ratios.
This table draws extensively from information presented by Eder et al.26 and Klein et al.277 as well as other sources.24,278


Immunologically Mediated Transfusion Reactions

Reactions to transfused blood products can occur as a result of the presence of antibodies that are constitutive (e.g., anti-A, anti-B) or that have been formed as a result of exposure to donor RBCs, white blood cells, platelets and/or proteins, or as a consequence of the effects of transfused white cells.

Reactions to RBC Antigens

Acute Hemolytic Transfusion Reactions

The most hazardous of the immune reactions is the immediate acute hemolytic transfusion reaction (AHTR) against foreign RBCs. Hemolysis of donor RBCs can lead to acute renal failure and DIC. The mortality rate is 2%.23 There are more than 300 antigens on human red cells, but most are weak immunogens that usually do not elicit a clinically detectable antibody response. The antibodies that fix complement and commonly produce immediate intravascular hemolysis include those against A, B, Kell, Kidd, Duffy, and Ss antigens. Rh antibodies (i.e., anti-D, anti-Cc, and anti-Ee), although typically not complement binding, are also capable of causing serious acute hemolytic reactions. Transfusion of incompatible FFP to A, B, or AB patients, resulting in hemolysis of recipient red cells, has also been a rare cause of AHTRs.23

ABO incompatibility, in company with transfusion-related acute lung injury (TRALI) and bacterial contamination, is among the three leading causes of transfusion-related deaths in the United States. Clerical and patient identification errors are the most common causes of ABO incompatibility and a large fraction of these errors typically occur in the operating theater. It is an uncomfortable irony that one of principal hazards of transfusion resides not in the blood supply per se, but rather in the process whereby it is delivered to the patient.

When incompatible blood is administered, antibodies and complement in recipient plasma attack the corresponding antigens on donor RBCs. Hemolysis ensues. The hemolytic reaction will take place in the intravascular space and it may also occur extravascularly within the reticuloendothelial system (spleen, liver, bone marrow). The antigen-antibody complexes activate Hageman factor (factor XII), which in turn acts on the kinin system to produce bradykinin (see Chapter 13). The release of bradykinin increases capillary permeability and dilates arterioles, both of which contribute to hypotension. Activation of the complement system results in the release of histamine and serotonin from mast cells, resulting in bronchospasm. Thirty to 50% of patients develop DIC.

Hemolysis releases hemoglobin (Hb) into the blood. Initially it is bound to haptoglobin and albumin. When those binding sites are saturated, it circulates unbound until it is excreted by the kidneys. Renal damage occurs for several reasons. Blood flow to the kidneys is reduced in the presence of systemic hypotension and renal vasoconstriction. Free Hb in the form of acid hematin or red cell stroma may damage renal tubules. Antigen-antibody complexes may be deposited in the glomeruli. If the patient develops DIC, fibrin thrombi will also be deposited in the renal vasculature, further compromising perfusion and/or causing acute cortical necrosis, which is frequently irreversible.

The signs and symptoms of a hemolytic transfusion reaction include fever, chills, nausea and vomiting, diarrhea, and rigors. The patient is hypotensive and tachycardic (bradykinin effects) and may appear flushed and dyspneic (histamine). Chest and back pains occur and have been attributed to cytokine release. The patient is often restless, has a headache, and a sense of impending doom. Hemoglobinuria will occur if plasma Hb rises above the renal threshold (about 25 mg/dL). Diffuse bleeding occurs with the development of DIC. With renal failure, oliguria develops. During general anesthesia, many of the signs are masked. Hypotension and microvascular bleeding may be the only initial clues that a hemolytic transfusion reaction has occurred, and the diagnosis may not be suspected until hemoglobinuria is observed. A reasonable index of suspicion should be maintained during administration of RBCs to anesthetized patients in order to avoid critical delay in diagnosis.

If a reaction is suspected, the transfusion should be stopped and the identity of the patient and the labeling of the blood rechecked. Examination of the patient's plasma after brief centrifugation for the pinkish discoloration caused by free Hb is a simple, rapid screening test when a hemolytic transfusion reaction is suspected. Hemolysis can be due to other causes, but should be assumed to indicate a hemolytic transfusion reaction until proven otherwise. Management has three main objectives: maintenance of systemic blood pressure, preservation of renal function, and the prevention of DIC. Systemic blood pressure should be supported by administration of volume, pressors, and inotropes as required. Urine output should be promoted by administration of fluids and the use of diuretics, either mannitol or furosemide, or both. Sodium bicarbonate can be administered to alkalinize the urine. There is currently no specific therapy to prevent the development of DIC. However, preventing hypotension and supporting cardiac output to prevent stasis and hypoperfusion, both of which contribute to the evolution of DIC, are important.

The response should include immediate notification of the blood bank, to which the suspected unit of blood should be returned, aseptically sealed, along with a posttransfusion EDTA blood specimen. The blood bank will determine whether the unit of blood had been correctly released to the patient. Immediate tests on the posttransfusion specimen will include (1) a visual check for hemoglobinemia and (2) a direct antiglobulin (Coombs) test. The direct antiglobulin test examines recipient RBCs for the presence of surface immunoglobulins and complement. If positive, an acute hemolytic reaction may have occurred and additional testing is indicated to ascertain the cause, including repeat ABO/Rh type, antibody screen, cross-matching, and other tests as indicated. Serum haptoglobin level, plasma, and urine Hb and bilirubin assays are usually performed. However, these are evidence of hemolysis only, not specifically of an immune reaction. The unit should be cultured if bacterial sepsis, usually associated with temperature elevation, is in the differential diagnosis. Laboratory tests to establish baseline coagulation status including platelet count, prothrombin time (PT), activated partial thromboplastin time (aPTT), thrombin time (TT), fibrinogen level, and fibrin degradation products should be performed, as should baseline studies of renal function.

Delayed Hemolytic Transfusion Reactions

Numerous instances have been reported in which transfused red cells are rapidly eliminated from the circulation at a short interval (days) after an apparently “compatible” crossmatch. These delayed hemolytic transfusion reactions can be the result of a donor RBC antigen to which the recipient has been previously exposed by either transfusion or pregnancy. Over time, the recipient antibodies fall to levels too low to be detected by compatibility testing. With re-exposure, an anamnestic response results in antibody that eventually lyses the foreign RBCs. In other instances, de novo alloimmunization may be responsible. Typically, the antibody-coated RBC is sequestered extravascularly and lysis occurs in the spleen and reticuloendothelial system. Because the RBC destruction occurs extravascularly, symptoms are less severe and the reaction is unlikely to be fatal. Unlike AHTRs, which usually involve antibodies in the ABO system, delayed hemolytic transfusion reactions commonly involve antibodies against Kell, Kidd, and Rhesus antigens.24 While alloimmunization and the


appearance of new antibodies occurs with approximately 1 per 200 units transfused, clinically detectable delayed hemolytic reactions occur at a rate of only 1 per 2,000 to 2,500 transfusions24 (Table 16-2).

Evidence of hemolysis is usually detected by the first or second week following transfusion. The reaction should be suspected in the event of a low-grade fever, increased indirect bilirubin with or without mild jaundice, and/or an unexplained reduction in Hb concentration. Serum haptoglobin may also be decreased. The diagnosis is confirmed by a positive direct antiglobulin test (Coombs test) and the identification of a new antibody in the patient's plasma. The reaction is typically mild and self-limiting and the clinical manifestations resolve as the transfused cells are removed from the circulation. Supportive care includes monitoring of Hb, maintenance of hydration, and provision of compatible blood if necessary.

Reactions to Donor Proteins

Minor Allergic Reactions

Allergic reactions to proteins in donor plasma cause urticarial reactions in 0.5 to 4% of all transfusions.2,23 The reaction is most frequently associated with the transfusion of FFP or platelets. The patient may have itching, swelling, and a rash (histamine release). These mild symptoms can be treated with diphenhydramine (Chapter 12). Most mild urticarial reactions are isolated events that do not recur. Patients who experience repeated reactions or a single severe urticarial reaction may benefit from the use of saline-washed red cells. The washing of platelets is generally ineffective, and susceptible patients who require platelets or FFP can be managed by administration of antihistamine and steroids (e.g., prednisone, 1 mg/kg or equivalent) 1 hour prior to transfusion.

Anaphylactic Reactions

Infrequently, more severe, anaphylactic reactions including dyspnea, bronchospasm, angioedema, and hypotension may occur (Chapter 12). Classically, these occur when patients with hereditary immunoglobulin (Ig) A deficiency who have been sensitized by previous transfusions or pregnancy are exposed to blood with foreign IgA protein. However, other plasma protein polymorphisms (e.g., haptoglobin) may cause similar reactions. Treatment consists of discontinuation of the transfusion and administration of epinephrine and methylprednisolone. Washed red cells, frozen deglycerolized red cells, or in appropriate cases, red cells from IgA-deficient donors should subsequently be used for these patients. Platelet and FFP transfusion may be managed with pretransfusion administration of prednisone (see previous discussion), careful monitoring, and epinephrine at the bedside.

White Cell-Related Transfusion Reactions

Febrile Reactions

Patients who receive multiple transfusions of RBCs or platelets commonly develop antibodies (alloimmunization) to the human leukocyte antigens (HLAs) on the passenger leukocytes in these products. During subsequent RBC transfusions, febrile reactions may occur as a result of antibody attack on donor leukocytes. These febrile responses occur in up to 2% of platelet, FFP, and RBC transfusions (Table 16-2). Typically, the patient experiences a temperature increase of more than 1°C within 4 hours of a blood transfusion and defervesces within 48 hours. The fever is sometimes accompanied by chills, respiratory distress, anxiety, headache, myalgias, nausea, and a nonproductive cough. Febrile reactions can be treated with acetaminophen. A leukocyte-mediated febrile transfusion reaction should be distinguished from a hemolytic transfusion reaction (direct Coombs test). Leukoreduction (see later) reduces or prevents these reactions.25

Transfusion-Related Acute Lung Injury

TRALI is a noncardiogenic form of pulmonary edema occurring after blood product administration (see Chapter 12). It has been associated with all plasma-containing blood components, with platelet concentrates and FFP being implicated much more commonly than PRBCs or other products.26,27 The incidence (Table 16-2) is frequently estimated to be 1:5,000 units transfused, although it has been recently reported to be as high as 1 per 1,271 transfused units in a carefully observed, at-risk intensive care unit (ICU) patient population.28 It is likely that TRALI, which carries a mortality of at least 5%,29 has historically been both underrecognized and underreported. Awareness is increasing, and according to a report by Holness and Epstein, TRALI was responsible for 46.5% of deaths reported to the Food and Drug Administration (FDA) in the first half of 2006.30

Detailed reviews of TRALI are available.29,31 In most instances (>90%), TRALI occurs when mediators present in the plasma phase of donor blood activate leukocytes in the host. Those mediators are usually anti-HLA (Class I or Class II) or antigranulocyte antibodies in donor plasma formed as a result of previous transfusion or pregnancy. In a small percentage of instances, the inverse reaction, aggregation of donor leukocytes by recipient antibodies, may be the cause when the recipient has been alloimmunized against leukocyte antigens. In either circumstance, the activated leukocytes are sequestrated in the lung and the mediators they release cause capillary endothelial damage and increased permeability.

Because antileukocyte antibodies cannot be demonstrated in all instances of TRALI, it seems certain that other mechanisms are sometimes operative. A double insult, or “two-hit,” theory proposes that the humoral response to various physiologic stresses (e.g., trauma, surgery, sepsis, systemic inflammatory response) may first “prime” native granulocytes, causing the appearance of surface adhesion sites, which in turn results in lung sequestration. Transfusion is proposed to be the wielder of the second hit. The mediators are thought to be biologically active lipids, sometimes referred to as biological response modifiers (BRMs) that accumulate as a result of the breakdown of membranes of the cellular elements in stored blood products. It is the BRMs, notably various lysophosphatidylcholines, that activate the sequestered leukocytes. Consistent with this theory is that TRALI has been reported to be more likely to occur with longer product storage times.32 A merging of these two theories may occur, if for instance it is demonstrated that the combination of antibodies and BRMs in donor blood collaborate in some way to effect the two hits.

The clinical appearance is very similar to that of acute lung injury of other etiologies, although the mortality rate should be substantially less. Beginning within 6 hours of transfusion, and often more rapidly, the patient develops dyspnea, chills, fever, and noncardiogenic pulmonary edema. Both hypotension and hypertension may occur. Chest x-ray reveals bilateral infiltrates. Severe pulmonary insufficiency can develop. Specific diagnostic criteria for the diagnosis of TRALI have been established (Table 16-3).33

Treatment is largely supportive. The transfusion should be stopped if the reaction is recognized in time. Transfusion-associated circulatory overload (“TACO” in the vernacular of blood bankers) should be considered and ruled out. Supplemental oxygen and ventilatory support should be provided as necessary, ideally using the same low tidal volume lung protective strategies that are employed in the acute respiratory distress syndrome.34 The pulmonary edema is noncardiogenic. Accordingly, diuretics are nonwarranted. Glucocorticoids have been administered but there are no data to support the practice.

There are preventive standards at the time of this writing (April 2008). It is anticipated that universal leukoreduction


will decrease the presence of antileukocyte antibodies in both donors and recipients.35 Multiparous female donors have been identified as the most common source of the antileukocyte antibodies in TRALI fatalities.23 Beginning in 2003, the United Kingdom substantially restricted the use of plasma-containing products from this donor subgroup. A program to similarly limit the preparation of high plasma-volume components (FFP, TP, FP24, or plasma frozen within 24 hours after phlebotomy [see later discussion], apheresis platelets, whole blood) from donors known to be leukocyte-alloimmunized or at increased risk of leukocyte alloimmunization (pregnancy or prior transfusion), or to perform HLA antibody testing, is in the process of implementation in the United States.1

Table 16-3 Diagnostic Criteria for Transfusion-Related Acute Lung Injury

1.    Acute onset of hypoxemia (within 6 hours of conclusion of transfusion)

2.    Bilateral CXR infiltrates consistent with ALI

3.    Absence of evidence of left atrial hypertension

4.    Absence of other temporally related causes of ALI

CXR, chest x-ray; ALI, acute lung injury.
From Kleinman S, Caulfield T, Chan P, et al: Toward an understanding of transfusion-related acute lung injury: statement of a consensus panel. Transfusion 2004; 44: 1774 with permission.

Graft-versus-Host Disease

PRBCs and platelets both contain a significant number of viable donor lymphocytes. When transfused into immunocompromised patients, the donor lymphocytes may become engrafted, proliferate, and establish an immune response against the recipient (see Chapter 54). In essence, the engrafted lymphocytes reject the host.36

Patients at risk for graft-versus-host disease (GVHD) include organ transplant recipients, neonates who have undergone a blood-exchange transfusion, and patients immunocompromised by many other disease processes (but not AIDS; Table 16-4). GVHD typically progresses rapidly to pancytopenia. The fatality rate is very high. Transfusion-associated GVHD has also been reported in apparently immunocompetent patients when a genetic relationship exists between the donor and the recipient. In these circumstances, the recipient may share HLA antigen haplotypes with the donor lymphocytes. The patients, although immunologically competent, fail to reject the transfused cells because they do not recognize them as foreign. The transfused donor lymphocytes, however, recognize the host as foreign and a GVHD reaction takes place.

Table 16-4 Irradiation of Cellular Blood Products for Patients at Risk of Graft-Versus-Host Disease

Widely recommended:
   Bone marrow transplant recipients
   Transfusion from consanguineous donor
   Hodgkin's disease
   Intrauterine transfusions
   HLA-matched platelet or granulocytes transfusion
   Ongoing treatment with purine analog antimetabolites
   Severe combined immunodeficiency syndrome
Lesser risk/practices vary with intensity of immunosuppression:
   Acute leukemias
   Non-Hodgkin's lymphoma
   Solid tumors
   Solid organ transplant recipients

Modified from Schroeder ML: Transfusion-associated graft-versus-host disease. Br J Haematol 2002; 117: 275.

GVHD has been reported only after the transfusion of cellular blood components. It has not occurred following transfusion of FFP or cryoprecipitate. The AABB recommends that HLA-matched platelets and directed donations from first-degree relatives be irradiated to inactivate donor lymphocytes.a Leukoreduction may reduce the incidence of GVHD, but it does not prevent it2 or reduce the mortality if it occurs. Irradiation remains the only effective means for preventing GVHD.37 Anesthesiologists will encounter patients in operating rooms and ICUs who are at risk for GVHD; they should be prepared to ask, “Should the blood we administer to this patient be gamma irradiated?”

The three leading causes of transfusion-related death in the United States are TRALI, ABO incompatibility, and sepsis caused by bacterial contamination.

Transfusion-Related Immunomodulation (TRIM)

Allogeneic transfusion has long been known to cause alteration of immune responsiveness. The initial observations were of decreased rates of transplant rejection38 and decreased rates of spontaneous abortion among patients who had received allogeneic transfusions. That some modification of immune surveillance occurs seems inescapable, and the occurrence of numerous transfusion-associated changes in immune-related processes, including T-lymphocyte helper/suppressor ratio, the function of killer T cells, lymphocyte responsiveness, and delayed hypersensitivity has been demonstrated.39 While transfused mononuclear white cells are thought to be principally responsible, other mechanisms may be involved.39 Numerous adverse effects, presumed to reflect this attenuation of immunocompetence, have been reported, including increased mortality, accelerated recurrence of malignancy, increased rates of infection, and more rapid progression of HIV/AIDS. Although many of the typical observational studies have left it less than absolutely clear whether transfusion was the cause of the adverse outcome or merely a reflection of the concomitant processes that necessitated blood product administration, the weight of the accumulated investigations, including some that have controlled carefully for confounding variables,40,41,42,43 argue that the adverse effect of transfusion on infection rates and mortality is a real one in at least some contexts.44 One investigation is particularly revealing. Hebert et al.45 prospectively compared transfusion strategies based on a liberal (10 g/dL) versus a restrictive (7 g/dL) transfusion threshold in an ICU population in whom the potential confounders were balanced at the time of patient enrollment. They observed lesser severity of multiple-organ dysfunction, reduced length of ICU and hospital stay, and reduced mortality at all follow-up intervals in the restrictive group. Although this investigation is strongly suggestive of an adverse effect of allogeneic blood and further supports the importance of avoiding unnecessary transfusion, it should be acknowledged that there is no certainty that the adverse effect was entirely a function of immune suppression (see next section).


Transfusion-Induced Inflammatory Response

It seems probable that, in addition to any TRIM effect, transfusion induces an inflammatory response in the recipient. (Note that “TRIM” is now sometimes used to encompass boththe immune suppressant and proinflammatory effects of transfusion.)

Numerous bioactive substances, including cytokines, membrane lipid breakdown products, and complement, accumulate during blood product storage and are suspected of contributing to an inflammatory response in the recipient and to the progression of multiorgan dysfunction.39,46,47 It is possible that some of the adverse effects of transfusion on mortality are a function of a proinflammatory rather than an immune suppressant effect. Because the concentrations of these mediators increase during storage, several investigations have sought to determine whether the duration of storage has any relation to outcome. A correlation between the age of transfused PRBCs and the severity of multiorgan failure in trauma patients,48 life-threatening outcomes and mortality in an ICU population,49 and mortality, renal dysfunction, and length of stay in cardiac surgical patients50,51 have been reported. If these results are borne out by larger prospective trials, it will bring pressure on our blood delivery system to achieve shorter “shelf times” (which currently average about 20 days52), at least for patients in the more critical circumstances.


The suspicion that transfused leukocytes are the mediators of the immunity-attenuating effects of transfusion mentioned previously led to the development and progressive application of techniques for leukocyte depletion of donor blood products. If leukocytes are responsible for TRIM, leukoreduction should attenuate the adverse effects. However, meta-analysis of studies comparing white blood cell (WBC)-reduced and non–WBC-reduced blood has revealed a reduction of mortality that is evident only in the context of cardiac surgery.53 A recent retrospective comparison of patients who received allogeneic blood after the onset of acute lung injury subsequently reported increased mortality among patients who received nonleukoreduced blood.43 That study was limited by some noncurrency of the leukoreduced and nonleukoreduced groups. Further meta-analysis, limited to the most carefully performed investigations, also revealed an effect on the rate of postoperative infections.53 However, when the impact of leukoreduction on the progression of HIV/AIDS was studied in a blinded prospective manner, no effect was identified.54

Albeit that the extent to which white cells are responsible for the adverse immunologic effects of transfusion is not absolutely clear, for that and other reasons, many countries—including Canada, France, Portugal, and the United Kingdom—and certain states and regions within the United States have already adopted the practice of leukoreduction of 100% of their blood supplies. The entire United States is moving towards that objective. At present (2008), about 70% of platelets and 40 to 50% of FFP and PRBCs are leukoreduced. There are several other well-confirmed benefits of leukoreduction25 including reduction in the development of alloimmunization and platelet refractoriness, reduction in the incidence of febrile nonhemolytic transfusion reactions, and reduction in (but not prevention of14) the transmission of CMV. However, it has been argued that (less expensive) selective leukoreduction could readily be applied for the patients to whom these benefits are relevant. The advocacy of universal leukoreduction is based on the premise that it might serve to accomplish the various unconfirmed benefits listed in Table 16-5, and/or that selective leukocyte reduction might result in many patients receiving nonleukoreduced blood before the indication for leukocyte reduction became apparent e.g., a severely anemic, undiagnosed, acute leukemic patient. In individual institutions with dual inventories, it will be the responsibility of the clinician at the bedside to request leukoreduced blood when he or she perceives it to be in the patient's best interest.

Table 16-5 The Benefits of Leukoreduction

·         Confirmed benefits

o    Decreased alloimmunization/platelet refractoriness in multiply transfused leukemics

o    Prevention of febrile reactions to RBC transfusions

o    Reduction of CMV transmission

o    Reduced inflammatory mediator accumulation during storage

·         Reported but unconfirmed benefits:

o    Shortened hospitalization

o    Decreased postoperative mortality in cardiac surgery

o    Decreased postoperative infections

o    Prevention of transfusion-related increase in tumor recurrence

·         Suggested but unconfirmed:

o    Reduced incidence of GVHD

RBC, red blood cell; CMV, cytomegalovirus; GVHD, graft-versus-host disease.

While many of the putative benefits are unconfirmed, additional reports of benefits attributed to leukoreduction are being added to the literature.55,56 Although skepticism persists, the common view is that, in spite of the associated costs, because the hazards of leukoreduction are minimal, the possible benefits justify proceeding with universal leukoreduction. Universal leukoreduction, when fully implemented, will employ prestorage depletion rather than bedside leukoreduction filters at the time of administration. Prestorage depletion avoids the accumulation of cytokines released by WBCs during storage. Clinicians should also be attentive to the possibility of severe, apparently bradykinin-mediated hypotension in patients who receive bedside-filtered blood. The reaction appears to occur more frequently, although not exclusively, in patients receiving angiotensin-converting enzyme inhibitors (which reduce breakdown of bradykinin).57

Other Noninfectious Risks Associated with Transfusion

Massive Transfusion

The rapid transfusion of large volumes of stored blood can have several consequences (Table 16-6). Some of these are functions of properties of the blood itself, some of the agents used to preserve and anticoagulate it, and some of the biochemical reactions that occur during storage (see Chapter 36). Other complications are not unique to blood transfusions, but may occur with the rapid transfusion of large volumes of any fluid.


Hypothermia slows hemostasis (as it does all enzymatically mediated processes) and causes sequestration of platelets. The administration of one unit of PRBCs at 4°C will reduce the core temperature of a 70-kg patient approximately 0.25°C. At 29°C (the temperature at which the risk of cardiac dysrhythmias is critical), PT and aPTT will increase approximately 50% over normothermic values, and platelet count will decrease by approximately 40%.58 Dysrhythmias may be seen at higher core temperatures if unwarmed blood is administered


rapidly; in particular, through central catheters. With decreasing body temperature, cardiac output declines, tissue perfusion is impaired (as a consequence of both vasoconstriction and a left shift of the oxygen-hemoglobin [O2-Hb] dissociation curve), and metabolic acidosis may develop. Shivering on emergence can increase oxygen consumption by 400%.

Table 16-6 Hazards Associated with Massive Transfusion

·         Hypothermia

·         Volume overload

·         Dilutional coagulopathy

·         Reduced oxygen-carrying capacity (decreased 2,3-DPG)

·         Metabolic acidosis

·         Hyperkalemia

·         Citrate intoxication

2,3-DPG, 2,3-diphosphoglycerate.

A meta-analysis concluded that even mild hypothermia increases blood loss.59 Hypothermia, after attempting to correct for covariates, is an independent predictor of mortality in trauma patients.60,61 Hypothermia has been associated with increased postoperative morbidity and mortality including increased rates of postoperative infection.62 However, in studies of this nature, it is difficult to separate the effects of the common clinical concomitants of hypothermia (e.g., acidosis, shock, massive transfusion, massive tissue injury) from those of hypothermia per se. Furthermore, the significance of hypothermia may lie in the interaction with other variables as suggested by the observation that temperatures of 33°C have been used extensively in elective neurosurgery without clinically apparent coagulopathy. Nonetheless, hypothermia should be carefully avoided and aggressively corrected in the patient receiving massive transfusion. Accordingly, transfusions administered rapidly or in substantial volume should be warmed.

Volume Overload

Circulatory volume overload occurs when blood or fluid is transfused too rapidly for compensatory fluid redistribution to take place.

In the setting of massive transfusion, assuming maintenance of isovolemia and the absence of a consumptive coagulopathy, critical dilution of clotting factors and platelets is likely to occur after an average replacement of 140% and 230% of blood volume, respectively.

Dilutional Coagulopathy

Administration of large volumes of fluid deficient in platelets and clotting factors will result in a coagulopathy as a consequence of dilution. In contemporary practice, in which patients receive principally PRBCs with only very limited amounts of residual plasma, factor deficiencies develop before thrombocytopenia. With large-volume isovolemic dilution, clinically significant dilution of fibrinogen, factors II, V, and VIII, and platelets would be expected to occur after volume exchanges of approximately 140%, 200 to 230% and 230% (i.e., 1.4, 2, and 2.3 blood volumes), respectively.63 However, fibrinogen is an acute phase reactant, and levels will often be greater than would be predicted by dilution calculations. Resuscitation from hypovolemia will result in reaching these thresholds at smaller percentage volume exchanges. Note that calculations of this nature should not be used as a guide to blood product administration but merely as a means of anticipating clinically relevant occurrences. The decision to administer FFP or platelets will depend on clinical and laboratory evidence of coagulopathy, or frequently on the uncertainties associated with rapid and ongoing blood loss. In the setting of trauma, PT/international normalized ratio (INR) has been shown to reveal factor deficiencies with greater sensitivity than aPTT.64 This is probably because the aPTT assay is very FVIII-sensitive, and FVIII, like fibrinogen, is an acute phase reactant that is often increased in the setting of trauma. The most common initial factor deficiencies in the setting of trauma are FV and FX.64

With the possible exception of trauma resuscitation, coagulation factor and platelet replacement should be determined by laboratory assessment and/or observation of clinical coagulopathy and not estimated blood loss-driven formulas.

Decreases in 2,3-Diphosphoglycerate

Storage of RBCs is associated with a progressive decrease in intracellular ATP and 2,3-diphosphoglycerate (2,3-DPG) with a resultant left shift of the O2-Hb dissociation curve. Accordingly, transfusion of the 2,3-DPG–depleted blood, while increasing the patient's Hb value, will result in less efficient oxygen delivery than would occur with native Hb at the same hematocrit. After transfusion, 2,3-DPG levels return toward normal over 12 to 24 hours.65

Acid-Base Changes

When citrate-phosphate-dextrose (CPD) solution is added to a unit of freshly drawn blood, pH decreases to approximately 7.0 to 7.1 (see Chapter 14). Further reduction of pH will occur during storage as a consequence of ongoing metabolism of glucose to lactate. At the end of 21 days, the pH may be as low as 6.9, but much of this is the result of the production of CO2 that is rapidly eliminated following the transfusion. Whether rapid infusion of this acidic bank blood leads to metabolic acidosis is debated. When the liver is adequately perfused, citrate from the CPD solution is metabolized to bicarbonate and any acid-base disturbance should therefore be self-correcting. Clinically, in the injured patient who is hypotensive, poorly perfused, and has inadequate tissue oxygenation, it will be difficult to differentiate what portion of the metabolic acidosis is due to rapid transfusion, and what portion is due to the production of lactic acid. The appropriate course is to base bicarbonate therapy on blood gas analysis.


During storage, potassium moves out of the RBCs, in part to maintain electrochemical neutrality as hydrogen ions generated during storage redistribute. The potassium concentration in plasma may reach levels variously reported to be between 19 and 35 mEq/L in blood stored for 21 days. Hazard exists if large volumes of stored blood are administered rapidly. While there are only 20 to 60 mL of plasma in a unit of PRBCs, contemporary infusion devices allow blood to be transfused at rates of 500 to 1,000 mL/min. At these infusion rates, critical hyperkalemia can occur and intraoperative arrests have been documented.66 Premature neonates are especially susceptible to hyperkalemia, and typically therefore receive either fresh (<8 day old), plasma-reduced, or washed PRBCs if rapid transfusion (>10 to 15 mL/kg per 2 hours) is required.

Citrate Intoxication

When large volumes of stored blood (more than one blood volume) are administered rapidly, the citrate can cause a temporary reduction in ionized calcium levels. Citrate is normally metabolized efficiently by the liver and decreased ionized calcium levels should not occur unless the rate of transfusion exceeds 1 mL/ kg per minute or about 1 unit of blood per 5 minutes in an average-sized adult. The now-common additive solution blood preservatives have a much smaller citrate content than citrate-phosphate-dextrose-adenine (CPDA). This should further reduce the hazard of


citrate intoxication with PRBC administration. However, most of the citrate administered during massive transfusion is in the FFP rather than the PRBCs. Impaired liver function or perfusion will lower the rate threshold for developing citrate intoxication. Note also that critical cardiac consequences that occur before hypocalcemia have significant implications for coagulation. Signs of citrate intoxication (hypocalcemia) include hypotension, narrow pulse pressure, and elevated intraventricular end-diastolic pressure and central venous pressure, prolonged QT interval, widened QRS complexes, and flattened T waves.

Microaggregate Delivery

Stored blood contains microaggregates. Platelet aggregates form during the second to fifth day of storage and after approximately 10 days, larger aggregates composed of fibrin, degenerated white cells, and platelets appear. Macroaggregates of RBCs also develop. Standard fluid administration sets contain 170 micron filters, which will remove these larger aggregates, and are appropriate for RBCs, FFP, cryoprecipitate, and platelet administration. Microaggregates have been suspected in the pathogenesis of pulmonary insufficiency after large-volume transfusion. However, concomitant physiologic disturbances (hypotension, sepsis, tissue injury) may be the actual causes, and some of what has been attributed to microaggregates may in fact be TRALI. Micro-pore filters, typically with a 40-micron pore size, were once advocated for RBC administration but were of no demonstrated benefit (with the exception of the arterial cannula of the cardiopulmonary bypass [CPB] circuit).

RBCs are frequently diluted with crystalloid solutions to increase the rate at which the blood can be transfused. In the ideal situation, normal saline, Normosol, or other diluents accepted by the FDA and AABB should be used in preference to lactated Ringer solution (LR). In fact, the amount of citrate present in stored blood is more than sufficient to bind the small amounts of calcium in the 100 to 300 mL of LR typically used for dilution.67 There is no evidence that any clinically significant sequelae have resulted from the use of LR as an RBC diluent.68

Blood Products and Transfusion Thresholds

Red Blood Cells

The question of what hemoglobin/hematocrit (Hb/Hct) level justifies the risks associated with the administration of blood has been widely discussed. The once all but inviolable “10–30” rule has been abandoned. Experience with several patient subpopulations (renal failure, military casualties, Jehovah's Witnesses) and systematic study has revealed that considerable greater degrees of anemia can be well tolerated and that, in many situations, morbidity and mortality rates did not increase until Hb levels fell below 7 g/dL.45,69 As significant as the identification of a 7-g/dL threshold for increased morbidity was the observation that stable general medical-surgical managed to a target Hb of 10 g/dL fared less well than a parallel group managed with a transfusion trigger of 7g/dL.45 That observation implies an adverse effect of transfusion (see “Transfusion-Related Immunomodulation”). Accordingly, the contemporary transfusion trigger for stable general medical-surgical patients is 21%/7.0 g/dL (Hb/Hct). However, there is evidence that the threshold for patients with cardiac disease should be higher.45 That evidence includes an investigation supporting a threshold of 30%/10 g/dL (Hb/Hct) in patients who have suffered a recent acute myocardial infarction (MI),70 and an observational study suggests better outcomes in patients with several cardiac diagnoses (cardiac and vascular surgery, ischemic heart disease, dysrhythmias) above a threshold of 9.5 g/dL71 (see Chapter 42). The Practice Guidelines for Blood Component Therapy developed by the American Society of Anesthesiologists (ASA) state that “red blood cell transfusion is rarely indicated when the hemoglobin concentration is greater than 10 g/dL and is almost always indicated when it is less than 6 g/dL. The indications for autologous transfusion may be more liberal than for allogeneic (homologous) transfusion.”72

The clinician's responsibility is to anticipate, on a patient-by-patient basis, the minimum Hb level (probably in the range of 7 to 10 g/dL) that will avoid organ damage due to oxygen deprivation. Determining this individual “transfusion trigger” requires reference to the many elements of patient condition that determine demand for the delivery of oxygen and the physiologic reserve (Table 16-7),73 including ongoing blood loss and the potential for sudden blood loss. Ultimately, the decision to transfuse RBCs should be made on the basis of the clinical judgment that the oxygen-carrying capacity of the blood must be increased to prevent oxygen consumption from outstripping oxygen delivery. That judgment is based on an understanding of the physiologic mechanisms that compensate for anemia and the limits of those mechanisms.

The RBC transfusion “trigger” for most patients will lie between hemoglobin values of 7 and 10 g/dL.

Table 16-7 Conditions that May Decrease Tolerance for Anemia and Influence the Red Blood Cell Transfusion Threshold

·         Increased oxygen demand

o    Hyperthermia

o    Hyperthyroidism

o    Sepsis

o    Pregnancy

·         Limited ability to increase cardiac output

o    Coronary artery disease

o    Myocardial dysfunction (infarction, cardiomyopathy)

o    β-Adrenergic blockade

o    Inability to redistribute cardiac output

o    Low SVR states

§  Sepsis

§  Post–cardiopulmonary bypass

·         Occlusive vascular disease (cerebral, coronary)

·         Left shift of the O2-Hb curve

o    Alkalosis

o    Hypothermia

·         Abnormal hemoglobins

o    Presence of recently transfused Hb (decreased 2,3-DPG)

o    Hb S (sickle cell disease)a

o    Acute anemia (limited 2,3-DPG compensation)

·         Impaired oxygenation

o    Pulmonary disease

o    High altitude

·         Ongoing or imminent blood loss

o    Traumatic/surgical bleeding

o    Placenta previa or accreta, abruption, uterine atony

o    Clinical coagulopathy

SVR, systemic vascular resistance; O2-Hb, oxygen-hemoglobin; 2,3-DPG, 2,3-diphosphoglycerate.
aTotal Hb should not be increased to >10 g/dL unless Hb S <30%.73


Compensatory Mechanisms During Anemia

When anemia develops, but blood volume is maintained (iso-volemic hemodilution), four compensatory mechanisms serve to maintain oxygen delivery: (1) an increase in cardiac output, (2) a redistribution of blood flow to organs with greater oxygen requirements, (3) increases in the extraction ratios of some vascular beds, and (4) alteration of oxygen-Hb binding to allow the Hb to deliver oxygen at lower oxygen tensions.

1.   Increased cardiac output.

With isovolemic hemodilution, cardiac output increases primarily because of an increase in stroke volume brought about by reductions in systemic vascular resistance (SVR). The two principal determinants of SVR are vascular tone and blood viscosity.74 As Hct decreases, reduction of blood viscosity decreases SVR. This decrease in SVR increases stroke volume and consequently cardiac output and blood flow to the tissues. Over a wide range of Hcts, isovolemic hemodilution is self-correcting. Linear decreases in the oxygen-carrying capacity of the blood are matched by improvements in oxygen transport. Because oxygen transport is optimal at Hcts of 30%, oxygen delivery may remain constant between the Hcts of 45 and 30%.74 Further reductions in Hct are accompanied by increases in cardiac output, which reach 180% of control as the Hct approaches 20%. The exact Hb value at which cardiac output rises varies among individuals and is influenced by age and whether the anemia is acute or develops slowly.

2.   Redistribution of cardiac output.

With isovolemic hemodilution, blood flow to the tissues increases, but this increased flow is not distributed equally to all tissue beds. Organs with higher extraction ratios (brain and heart) receive disproportionately more of the increase in blood flow than organs with low extraction ratios (muscle, skin, viscera). Because basal extraction ratio (ER) is already high in the coronary circulation (see next paragraph), increased flow must be the principal means by which the healthy heart compensates for anemia.74 Coronary blood flow can increase by as much as 500%.75 When the heart can achieve no further increase in cardiac output and coronary blood flow, the limits of isovolemic hemodilution have been reached. Thereafter, further decreases in oxygen delivery will result in myocardial injury. Acute isovolemic Hb reductions to 5 g/dL can occur without impairment of total-body oxygen delivery in healthy, otherwise unstressed adults.76 However, in the same experimental paradigm (acute isovolemic reduction), reversible impairment of cognitive function occurred when Hb concentrations fell below 7.0 g/dL.77 The latter observation serves as a reminder that measures of global oxygen delivery may conceal critical occurrences in individual circulatory beds.

3.   Increased oxygen extraction.

Increasing oxygen ER is a critical compensatory mechanism when Hct drops below 25%. As isovolemic Hct decreases to 15%, the whole-body oxygen ER increases from 38 to 60%, and the mixed venous oxygen saturation decreases from 70 to 50% or less.68 Some organs (brain and heart) already have high ERs under basal conditions, and have a limited capacity to further increase oxygen delivery by this mechanism. The heart, under basal conditions, extracts between 55 and 70% of the oxygen delivered.78,79 The brain's ER is 30 to 35%. This contrasts with ERs of 7 to 30%, in most other tissues. In clinical practice, the measurement of the ERs of individual organs is usually not feasible. Because the heart has the highest ER, it is commonly said to be the organ at greatest risk under conditions of isovolemic anemia (although the work of Weiskopf et al.76,77 cited in the previous paragraph argues that it may in fact be the brain).

4.   Changes in oxygen-hemoglobin affinity.

The sigmoid O2-Hb dissociation curve describes the relationship between the partial pressure of oxygen (PO2) in the blood and the percentage saturation of the Hb molecule (seeChapter 11). The P50, the PO2 at which the Hb molecule is 50% saturated with oxygen at 37°C and a pH of 7.4, is 27 mm Hg. When the curve is shifted to the left (hypothermia, alkalosis), the P50 is reduced. The Hb molecule is more “stingy” and requires lower PO2 to release oxygen to the tissues; that is, the Hb molecule does not release 50% of its oxygen until an ambient PO2 <27 mm Hg is reached. This may impair tissue oxygenation. Right-shifting of the curve (increased temperature, acidosis) results in an increase of P50, decreased Hb affinity for the oxygen molecules and release of oxygen to tissues at higher partial pressures of oxygen.

When anemia develops slowly, the affinity of Hb for oxygen may be decreased, that is, the curve is right-shifted as a result of the accumulation of 2,3-DPG in RBCs. Synthesis of supranormal levels of 2,3-DPG begins at a Hb of 9 g/dL. Stored RBCs become depleted of 2,3-DPG. Temperature reduction and storage-related pH decreases also reduce the P50 of stored blood. These changes, however, are reversed in vivo, but the resynthesis of 2,3-DPG by RBCs will require from 12 to 24 hours.

Isovolemic Anemia Versus Acute Blood Loss

Although the same compensatory mechanisms are operative in acute and chronic anemias, they have different degrees of importance and occur at different Hb concentrations. In chronically anemic patients, the accumulation of 2,3-DPG in the RBCs, thereby increasing the P50 of Hb, is the important first mechanism for compensation. Cardiac output increases as Hb decreases to approximately 7 to 8 g/dL. With acute blood loss, vasoconstriction occurs and cardiac output does not increase. Redistribution and increased extraction are the compensatory mechanisms.


While published guidelines for platelet administration are available, there is once again a substantial requirement for clinician judgment. The indications for platelet administration presented in Table 16-8 are an amalgam of recommendations presented by the ASA in 1996 and 2006, the British Committee for Standards in Haematology in 2003 and 2006, and the French Safety Agency for Health Products in 2003.27,72,80,81,82

Table 16-8 makes it apparent that the platelet administration thresholds that will most often be relevant to anesthesiologists will lie between 50,000 and 100,000/uL.83,84 The threshold within that range at which platelets are administered should be based on the likelihood of the intended procedure to cause bleeding, the hazard of bleeding should it occur (e.g., intracranial neurosurgery > peripheral orthopaedics), and the presence or possibility of additional causes of coagulation disturbance (e.g., recent administration of antiplatelet agents, CPB, DIC, dilution due to large-volume administration). Bleeding manifestations can vary substantially from patient to patient in the face of similar platelet counts. This occurs because some platelets are more effective than others. When thrombocytopenia results from peripheral destruction of platelets, the bone marrow continues to produce normal, young, large platelets that are hemostatically very effective. A patient with these platelets may have more effective primary hemostasis than a patient with the same platelet count but whose platelets were produced by a less active, less healthy bone marrow.


Table 16-8 Indications, Expressed as Platelet Count Thresholds or Target Levels, Commonly Warranting the Administration of Platelets

•Nonbleeding patients without other abnormalities of hemostasis25


•Lumbar puncture, epidural anesthesia,a central line placement, endoscopy with biopsy, liver biopsy or laparotomy in patients without other abnormalities of hemostasis, vaginal delivery


•To maintain platelet count during ongoing bleeding and transfusion not less than


•To maintain platelet count during DIC with ongoing bleeding not less than


•To maintain platelets during management of massive blood loss


•Intended procedures in which closed cavity bleeding might be especially hazardous (e.g., neurosurgery)


•Microvascular bleeding attributed to platelet dysfunction such as uremia,b post–cardiopulmonary bypass, or in association with massive transfusion.


DIC, disseminated intravascular coagulation.
aThe French Safety Agency for Health Products recommends 50,000/µL for spinal and 80,000/µL for epidural anesthesia.82
bAfter a trial of DDAVP, if permitted by the clinical situation.83,84

Platelet administration thresholds relevant to anesthesiologists will lie usually between 50,000 and 100,000/µL.

A platelet concentrate derived from a single unit of donor blood will increase the platelet count of a 70-kg recipient by 5,000 to 10,000/µL. However, the majority of platelets (>70%) are now obtained by apheresis (see “Collection and Preparation of Blood Products”). One apheresis unit will increase platelet count by 30,000 to 60,000/µL. A common practice is to administer either one unit of apheresis platelets to an adult or one unit of platelet concentrate/10 kg of body weight. The increase in platelet count must be verified by platelet count, especially in patients who may have been alloimmunized by frequent platelet administration.

Fresh-Frozen Plasma

In spite of the fact that over 2,000,000 units of FFP are administered annually in the United States, there is remarkably little systematically-derived evidence of efficacy.85Nonetheless, the use of FFP to restore coagulation factor levels is inevitably valid in many clinical circumstances. The indications for FFP administration presented in Table 16-9 are an amalgam of recommendations presented by the ASA in 1996 and 2006, and the British Committee for Standards in Haematology in 2004.72,80,86 Effective coagulation can usually occur with clotting factor levels of 20 to 30% of normal. Levels that are 30% of normal can usually be achieved by administration of 10 to 15 mL/kg of FFP.80

Fresh-Frozen Plasma/Thawed Plasma in Trauma Resuscitation

It is the traditional dogma that administration of blood products, in particular FFP and platelets, should not be formula-driven, but should occur in response to a clinical coagulopathy, ideally with laboratory demonstration of abnormality. However, there is an evolving sentiment in the area of trauma resuscitation that this approach results in “falling behind” in the struggle against the tightening spiral of bleeding, hypotension, stasis, acidosis, hypothermia, and DIC87,88,89 (see Chapter 36). Because of the very high incidence of coagulopathy in multiple trauma victims,87,88,90 empiric approaches that involve the once-taboo formulas (e.g., two units of FFP or


thawed plasma [TP; see later for description of TP] with every five units of PRBC) when massive transfusion is anticipated or ongoing are already in routine use.87,88,91 An RBC-to-plasma ratio of 1:1 has been advocated and reported effective in military trauma.89 Formula-driven administration of platelets is also occurring. However, platelet administration may be less urgent because, as Ho and colleagues88 have observed, thrombocytopenia is not a central element of the insidious spiral just mentioned, and it may be easier to “catch up” from the consequences of a low platelet count than from coagulopathy-driven bleeding and the associated hypovolemia.

Table 16-9 Indications for the Administration of Fresh-Frozen Plasma

·         Correction of multiple coagulation factor deficiencies (e.g., DIC) with evidence of microvascular bleeding and PT and/or aPTT >1.5 times normal

·         Correction of microvascular bleeding during massive transfusion (more than one blood volume) when PT/aPTT cannot be obtained in a timely manner

·         Urgent reversal of warfarin therapya

·         Heparin resistance (antithrombin III [AT] deficiency) in a patient requiring heparin when AT concentrate is not available

·         Thrombotic thrombocytopenic purpura or hemolytic uremic syndrome

·         Correction of single coagulation factor deficiencies for which specific concentrates are not available (principally factor V)

·         ? Formula management of trauma/massive blood loss (see text)

DIC, disseminated intravascular coagulation; PT, prothrombin time; aPTT, activated partial thromboplastin time.
aProthrombin complex concentrate (II, VII, IX, X) is an alternative that has been reported to be more effective than fresh-frozen plasma.

Table 16-10 Indications for the Administration of Cryoprecipitate

·         Microvascular bleeding when there is a disproportionate decrease in fibrinogen, such as DIC and very massive transfusion,a with fibrinogen <80–100 mg/dL (or assay result not available)

·         Fibrin sealant (if virus-inactivated concentrate is unavailable)

·         Bleeding due to uremia that is unresponsive to DDAVP

·         Prophylaxis before surgery or treatment of bleeding in hemophilia A and vWD (if virus-inactivated concentrates are unavailable or ineffective)

·         Prophylaxis before surgery or treatment of bleeding in patients with congenital dysfibrinogenemias

·         FXIII deficiency

DIC, disseminated intravascular coagulation; vWD, von Willebrand disease.
aFresh-frozen plasma is the first-line component for the factor depletion associated with massive transfusion.

Normal coagulation can be achieved with clotting factor levels of 20 to 30% of normal. Those levels can usually be achieved by administration of 10 to 15 mL/kg of FFP.


Cryoprecipitate contains factor VIII, the von Willebrand factor (vWF), fibrinogen, fibronectin, and factor XIII. Virally inactivated factor VIII concentrates, some of which contain clinically effective concentrations of vWF (e.g., Haemate P, Alphanate) are now available. As a result, hemophilia A and von Willebrand disease (vWD) are usually treated (in consultation with a hematologist) with those concentrates92 rather than cryoprecipitate, which is now generally used for fibrinogen-deficient states. The remaining indications for cryoprecipitate are presented in Table 16-10.

Blood Conservation Strategies

Because of the many hazards of blood product administration, numerous techniques and alternatives have been explored (Table 16-11).

Autologous Donation

Preoperative donation and perioperative salvage of autologous blood have been used extensively as part of programs to reduce allogeneic blood administration. Autologous blood may be collected days to weeks prior to surgery (predonation); it may be collected immediately prior to surgery (isovolemic hemodilution); or it may be salvaged from the surgical field or wound drains and reinfused (blood salvage). In spite of the demonstration of modest efficacy,93 enthusiasm for many of these approaches has declined pari passu with the progressive reduction of the risk of transfusion-transmitted infections.

Table 16-11 Blood Conservation Techniques

·         Preoperative autologous donation

·         Acute normovolemic hemodilution

·         Intraoperative blood salvage

·         Postoperative blood salvage

·         Pharmacologic agents

o    Erythropoietin

o    Blood substitutes (hemoglobin- and nonhemoglobin-based)

o    DDAVP

o    Anti-fibrinolytics

Preoperative Autologous Donation

Preoperative donation of autologous blood (PAD) has been applied principally in patients undergoing major orthopaedic procedures (total hip and knee replacement, scoliosis procedures) and prostatic and cardiac surgery. However, with limited exception,94 the systematic experience has generally failed to demonstrate a reduction in allogeneic blood exposure.95 Effectiveness has probably been limited because the patients' erythropoietic response is often not vigorous, in which case the process may simply result in an anemia at the time of surgery. Furthermore, the PAD procedure is more expensive than the collection of allogeneic blood, and if autologous blood is not transfused it is usually discarded. The wastage rate was 59% in 2004.1 Note also that the transfusion of autologous blood does not eliminate the chance of human error during blood collection, processing, and reinfusion or the risk of bacterial contamination or the adverse effects of the storage lesion (bioactive lipids, cytokines). The use of PAD has decreased substantially since the initial enthusiasm for the technique.95,96,97 PAD nonetheless may be a useful alternative in alloimmunized patients for whom compatible allogeneic blood is difficult to obtain.

The medical condition of the patient must be considered prior to recommending PAD. Severe aortic stenosis, significant coronary disease or myocardial dysfunction, low initial Hct and blood volume (body weight <50 kg) are relative contraindications to PAD. If the patient's Hb level, cardiac status, and general condition permit, blood can be donated at weekly intervals prior to surgery. Four units is typically the maximum donation because of the shelf life of the first unit collected. Patients making PAD should receive supplemental iron (e.g., 2 mg/kg/day for 3 weeks). In addition, PAD can be supplemented with administration of recombinant erythropoietin (Epo).


The effectiveness of Epo in hastening recovery of Hct in conjunction with PAD and in improving Hct in patients not submitted to PAD has been demonstrated.98,99,100,101,102,103,104However, the practice has not become widespread in part because of the expense of the agent and in part because of the necessity for frequent (e.g., weekly injections for 3 weeks and two additional injections in the final week) parenteral (subcutaneous or intravenous) administration. Administration of Epo to


presurgical patients has resulted in reduction in allogeneic blood administration,105 and selective administration to anemic patients has been more obviously effective106,107 than has administration to “all comers.”108,109 Epo, a recombinant product, is often accepted by Jehovah's Witnesses, and its efficacy in that population has been demonstrated.110,111 The demonstration of the reduction by Epo of transfusion requirements in critically ill patients112 may increase awareness and encourage its systematic use in anemic elective surgical patients. Erythropoietic agents with longer half-lives (e.g., darbepoetin alpha) are under development and may serve to overcome one of the logistic limitations (frequent parenteral administration) to the preoperative use of Epo.113

Acute Normovolemic Hemodilution

Acute normovolemic hemodilution (ANH) entails withdrawal of the patient's blood early in the intraoperative period with simultaneous administration of crystalloids or colloids to maintain normovolemia. The rationale is that during the ensuing surgery, the patient will lose blood of low Hct, and the withdrawn blood will be available for reinfusion at the end of the operation. The end point for the initial withdrawal is a Hct of 27 to 33%, depending on the patient's cardiovascular and respiratory reserve. Selection for this technique should rely on careful evaluation of the patient for coronary or cerebral vascular disease. ANH evolved in the anticipation that it would reduce total red cell loss and allogeneic blood administration. However, both mathematical modeling and empiric experience have revealed only a modest benefit.114 By way of example, Goodnough115 calculated that, in a 100-kg patient from whom three units of blood is withdrawn and replaced by asanguinous fluid, if the subsequent blood loss is 2,800 mL, 215 mL of RBCs (about one unit) will be saved. For patients of limited body size, low starting Hct, or blood loss <70% of one blood volume,116 avoidance of allogeneic blood might be difficult to achieve. A recent meta-analysis reported that ANH does not achieve complete avoidance of allogeneic blood, but that when transfusion is necessary the amount transfused is reduced by one to two units per patient. The authors concluded that “widespread adoption of ANH cannot be encouraged.”114 Nonetheless, there are reports of favorable experiences in liver resection, prostatectomy, total hip arthroplasty, and abdominal aortic surgery.101,117,118,119,120 It is possible that in the future the efficacy of ANH will be enhanced by administration of preoperative erythropoietics and/or by the use of either Hb-based oxygen-carrying compounds or perfluorocarbon emulsions to permit withdrawal of larger volumes of blood.

ANH has also been employed for the purpose of making fresh autologous blood available at the end of procedures in which either a dilutional or CPB-related coagulopathy may occur. The efficacy in this context has not been confirmed by systematic study. Blood collected and reinfused for this purpose should not be passed through a 40-micron filter in order to avoid platelet elimination.

Perioperative Blood Salvage

Perioperative blood salvage refers to the recovery of shed blood from the surgical field or wound drains and readministration to the patient. In most instances, the process involves “washing” of the salvaged material with return of only the RBC component of blood. In some instances, usually those involving wound drainage, blood is returned filtered but otherwise unprocessed.

Intraoperative Blood Salvage

Intraoperative blood salvage (IBS) is employed with many surgical procedures that have the potential to require allogeneic transfusion. Contemporary cell-salvage devices anticoagulate the salvaged blood as it leaves the surgical field, separate the RBCs from other liquid and cellular elements by centrifugation, and then wash the salvaged RBCs extensively with saline. The RBCs are typically returned to the patient suspended in saline in aliquots of 125 or 225 mL with a Hct of 45 to 65%.121 Higher Hcts can be achieved at the expense of the additional time required for slower filling of the centrifuge chamber.

IBS has been used commonly during cardiovascular surgical procedures, aortic reconstruction, spinal instrumentation, joint arthroplasty, liver transplantation, resection of arteriovenous malformations,122 and occasionally in the management of trauma patients.123 There have been numerous demonstrations that IBS can reduce total blood loss and/or the use of allogeneic RBCs.124,125,126 The presence of infection, malignant cells, urine, bowel contents, and amniotic fluid in the operative field have been viewed as contraindications. However, although malignant cells are known to be retained with RBCs after the washing process, IBS has been applied in the management of hepatic and urologic malignancies without evidence of metastasis.127,128 At least one IBS washing device has also been shown to remove the critical procoagulant factors present in amniotic fluid129 and IBS has been employed successfully in cesarean section.130 However, the safety of IBS use in that context is unconfirmed and should not be routine.131

The potential complications of IBS are largely a function of the reinfusion of materials that might remain after the washing process. These include fat, microaggregates such as platelets and leukocytes, air, red cell stroma, free Hb, heparin, bacteria, and debris from the surgical field. Most of these are removed quite efficiently by contemporary cell-salvage equipment. Bacteria are the exception, and contamination of cell salvage return with skin organisms is relatively common.124 Leukocyte-reduction filters have been shown to remove most bacteria132 and may be relevant to the use of IBS in trauma and cesarean section. Massive air embolism has occurred as a result of user error. Direct return from the cell-salvage apparatus has now been largely abandoned in favor of return via an intermediary bag under the control of the anesthesiologist. Care should still be taken in the event that pressure infuser devices are applied to these bags.

A dilutional coagulopathy in association with large-volume IBS is to be expected because essentially all clotting factors and most platelets are removed by the washing process. A DIC-like coagulopathy was once associated with IBS; however, it seems likely that this syndrome was the result of inadequate preparation of blood by older cell-salvage devices. Unwashed, salvaged blood has been shown to contain numerous constituents that influence the coagulation process: thromboplastic material, interleukins, complement, fibrin-degradation products, and factors released from activated leukocytes and platelets122 and to activate the coagulation process in recipients.133 The majority of these are quite efficiently removed by contemporary IBS processing devices. However, their presence is used as an argument against the return of unprocessed blood from wound drains (see “Postoperative Blood Salvage”).134 Similar elements, including lipids and cytokines, in blood shed into the mediastinum during CPB are suspected of contributing to postprocedure morbidity, including cognitive dysfunction.135 Accordingly, it is an increasingly common practice to return mediastinal blood to the patient via IBS devices rather than the CPB reservoir.

An additional coagulopathy risk arises with the use of thrombin and microfibrillar collagen or cellulose products in


the surgical field.136 These agents are not reliably removed by the washing process, and suctioning of blood into the IBS device should be discontinued during the use of these agents and resumed after the field has been irrigated.

The clinician should appreciate that the efficiency of the recovery of shed RBCs by the IBS process is on the order of 50%. Allogeneic blood will therefore frequently be necessary in spite of the IBS, and blood and fluid replacement calculations should take this into account.137,138 The efficiency of RBC recovery is improved by prompt recovery of blood from the surgical field (i.e., before clotting occurs) by limiting the negative pressure used, and by minimizing the mechanical air-blood interface during suctioning.139

Postoperative Blood Salvage

Postoperative recovery of blood from mediastinal chest tubes and wound drains after hip and knee replacement with immediate reinfusion of the “unwashed” blood has been employed quite commonly. The many substances present in the unprocessed blood (see previous section) suggest that coagulation dysfunction might result, and many are skeptical regarding the wisdom of this practice121,134 (with one editorialist going so far as to characterize the technique as “repugnant”140). However, there have been several reports of efficacy in reducing allogeneic blood exposure without apparent adverse effects,141,142,143 and only occasional reports of apparent adverse consequences.144 This may reflect the fact that the reinfused volumes are usually small.

Hemoglobin-Based Oxygen-Carrying Solutions

Hemoglobin-based oxygen-carrying solutions would offer numerous advantages: long shelf live, minimal infection hazard, absence of alloimmunization, immediate availability (no typing), little likelihood of TRIM, and no storage lesion/ inflammatory response. However, while numerous polymerized Hb products have been studied, only one, Hemopure (Biopure Inc Evanston IL), is approved for human use (in South Africa) and only one, PolyHeme (Northfield Laboratories Cambridge, MA) is currently in phase III trial in the United States.145 One additional product is at the phase II trial stage (Hemospan, Sangart, San Diego, CA). The many products studied have used bovine, outdated human, or recombinant Hb that has been entirely separated from red cell membranes (stroma) and polymerized to increase half-life. The initial difficulties with renal failure caused by residual stroma and excessive free Hb have been overcome. However, there are several remaining difficulties with which clinicians will probably have to contend including methemoglobinemia, interference with some calorimetrically based laboratory assays (including creatinine, total bilirubin, and lactate dehydrogenase), some degree of vasoconstriction caused by nitric oxide binding by free Hb, and a relatively short half-life. The nitric oxide effect is probably the basis for the failure of the several products that have been withdrawn from clinical trials. Polymerization increases half-life to 18 to 36 hours but that period is sufficiently short such that oxygen-carrying capacity will usually become inadequate before native reticulocytosis can compensate.146 Perfluorocarbon emulsions146 appear to be further from potential clinical application than hemoglobin-based oxygen-carrying solutions and will not be discussed here.

Jehovah's Witnesses

In general, on the basis of New Testament admonitions (Acts 15: 20, 29), Jehovah's Witnesses will accept neither administration of most allogeneic blood products nor the readministration of autologous products that have left the circulation. According to a recent Jehovah's Witness publication, “they reject all transfusions involving whole blood or the four primary components—red cells, plasma, white cells, and platelets. As for the various fractions, derived from those components—and products that contain such fractions—the Bible does not comment on these. Therefore each Witness makes his own personal decision on such matters.”147 Accordingly, the wishes of each patient must be clarified carefully. Few will permit the administration of PRBCs, FFP, platelets, or granulocytes, but other components and fractions may be acceptable. The majority will decline PAD. However, many will accept procedures that maintain extracorporeal blood in continuity with the circulation. The acceptability of CPB, acute normo-volemic hemodilution, and perioperative cell salvage must be clarified with each patient individually. Most will permit administration of Epo.

Collection and Preparation of Blood Products for Transfusion

Red Blood Cells

Whole blood is first collected in bags containing CPDA or CPD solution. The citrate chelates the calcium present in blood and prevents coagulation. Sequential centrifugation at various spin speeds and durations is used to separate whole blood into components including PRBCs, platelet concentrates, cryoprecipitate, and cell-free plasma. The two common PRBC preparations ultimately delivered to the clinician have either CPDA or so-called additive solution as the preservative. CPDA blood has an Hct of about 70 to 75%, contains 50 to 70 mL of residual plasma in a total volume of 250 to 275 mL, and has a shelf life of 35 days. With the additive solution preparation, the original preservative and most of the plasma (10 to 15 mL remains) is removed and replaced with 100 mL of additive solution. This results in a lower Hct (60%) in a total volume of 250 to 350 mL, less citrate per unit, 75 to 80% fewer microaggregates, and a longer shelf life (42 days). Additive solution RBCs are thought to regenerate 2,3-DPG more rapidly. The pH and K+ content of the two preparations are similar. The smaller plasma volume in additive solution blood results in smaller amounts of coagulation factors in PRBCs but also a potentially lesser risk of minor allergic reactions and TRALI (Table 16-2).

There are alternative RBC preparations that eliminate the various “passengers.” Saline-washed RBCs may be used for patients who experience reactions to foreign proteins. RBCs can be frozen and stored indefinitely. Preservatives to prevent freeze-thaw–associated damage must be added and subsequently removed before administration, which must occur within 24 hours of thawing. The process is expensive and therefore not widely used. Lymphocytes can be rendered incapable of division (and therefore unable to induce GVHD) by irradiation.

The administration of one unit of PRBCs will increase the Hb and Hct of a 70-kg adult by approximately 1 g/dL and 3%, respectively. However, both the freeze-thaw process and washing to reduce allergic reactions result in an RBC wastage of at least 20%.

Compatibility Testing

Compatibility testing involves three separate procedures: ABO Rhesus blood type identification, antibody screening of donor and recipient plasma, and the donor/recipient crossmatch.


ABO, Rhesus Typing

The first step is to determine the ABO blood group type and the Rh status of both donor and recipient blood. This is a critical step because most of the fatal hemolytic transfusion reactions result from the transfusion of ABO-incompatible blood. Blood types are defined according to the antigens present on the surface of the RBCs. Patients with type A blood have type A antigens on the surface of their red cells. Type B blood has B antigens. When both antigens are present the patient is said to have type AB blood, and when both are lacking the patient is has type O blood. By 6 to 12 months of age, the serum constitutively contains antibodies to the A and B antigens that are lacking on the RBC. Patients with type A blood have antibodies against the B antigen and vice versa. Patients with no antigens on their cells, type O blood, will have both anti-A and anti-B antibodies in the plasma.

Patients with the D antigen of the Rhesus group of antigens are said to be Rh-positive. Approximately 85% of the population is Rh-positive. In contrast to the A and B blood groups, anti-D antibodies are not constitutively present in the serum of an Rh-negative patient. However, 60 to 70% of Rh-negative patients exposed to donor Rh-positive RBCs will develop anti-D antibodies. There is a latency before these antibodies are synthesized. As a consequence, the reaction between the Rh-positive donor cells and the anti-D evolves slowly and may not be clinically apparent on first exposure. This process whereby a foreign antigen stimulates the synthesis of the corresponding antibody is termed alloimmunization. Subsequent exposure of these Rh-negative individuals to Rh-positive cells may result in an AHTR.

AHTRs are most often caused by antibodies in recipient plasma directed against A, B, or D antigens on donor RBCs. The antibody-antigen interaction activates complement and leads to intravascular hemolysis. “O-positive” recipients (type O, Rh[D]-positive) will have both anti-A and anti-B antibodies, but not the anti-D antibody in their plasma. These patients must not receive type A, type B, or type AB blood. They must receive type O blood, but it may be Rh-positive or Rh-negative. In contrast, patients with blood type AB-negative (type AB, Rh-negative) will lack both the A and B antibodies in their plasma and may or may not have the anti-D antibody in their plasma. They can receive A-, B-, AB-, or O- blood. Individuals with the greatest number of antigens on their RBCs (i.e., type AB-positive) have the fewest constitutive antibodies in their plasma and can receive all blood types (types A+, A-, B+, B-, AB+, AB-, O+, and O-) and are referred to as universal recipients. Individuals with the fewest antigens on their cells (type O) have the greatest number of antibodies in their plasma. Type O-negative RBCs can be administered to all ABO, Rh types and these individuals are referred to as universal donors. The distribution of A, B, O, and D phenotypes in the U.S. population is presented in Table 16-12. A derivative of those distributions is that, assuming the same representation among donors and recipients (and the absence of superimposed alloimmunization that would compound the risk), random administration of blood (or recipient identification errors) will result in an AHTR with one of every three PRBC units administered.

Table 16-12 Major Red Blood Cell Surface Antigen Incidence (%) in the U.S. Population
















Rh (D)



The Antibody Screen

The antibody screen, an indirect Coombs test, is performed to identify recipient antibodies against RBC antigens. Commercially supplied RBCs, selected for strong expression of 25 to 30 potentially hemolytic antigens, are mixed with recipient serum. Only about 4 in 1,000 potential recipients demonstrate unexpected antibodies. The likelihood that the antibody screen will miss a potentially dangerous antibody has been estimated to be much less than 1 in 10,000. If the recipient plasma screen is positive, the antibody must be identified and appropriate antigen-negative donor units selected. The antibody screening of recipient plasma should be repeated at 3-day intervals if the patient is receiving ongoing transfusion.

The Crossmatch

The predictive power of a negative antibody screen is such that most hospitals perform no further crossmatch procedures in patients who have no history of antibody formation. In institutions with validated blood bank computer systems, eligible patients may receive blood solely on the basis of an “electronic [computer database] crossmatch” of the recipient and the available units. Alternatively, some institutions perform an “immediate spin” (30 seconds at room temperature) crossmatch of recipient plasma and donor RBCs and examine for gross agglutination, which is predictive of ABO incompatibility. The requisite time for ABO/Rh typing and antibody screen, from sample arrival in the blood bank to blood release, is 30 to 45 minutes when the antibody screen is negative.

A formal crossmatch is performed if an antibody is identified, if the patient has a history of antibody formation, or if the patient is deemed to be at high risk for alloimmunization. Current procedures, which use a variety of enhancement techniques (e.g., low ionic strength solutions, polyethylene glycol, gels, or solid phase technology) allow antibody screens and/or crossmatches to be accomplished in approximately 20 minutes. The various incubation phases that were once used, necessitating 2 hours for a complete crossmatch, are no longer performed in the majority of institutions. Crossmatch procedures vary, but at a minimum will entail incubation of recipient plasma with donor RBCs at 37°C for 10 to 15 minutes followed by an indirect antiglobulin test and examination for agglutination.

In patients who have been transfused previously or who may have been exposed to foreign RBC antigens during pregnancy, the rate of development of an anti-RBC antibody to other than the A, B antigens is about 1 per 200 exposures (and is cumulative with multiple exposures).24 Determining the ABO and Rh status alone is sufficient to assure that the transfusion will be compatible in 99.8% of patients who have not previously been transfused or pregnant (i.e., the likelihood of incompatible transfusion is about 1 in 1,000). That latter rate will rise in proportion to the number of prior donor exposures or pregnancies. The addition of the 30-to 45-minute antibody screen further increases the likelihood of a compatible transfusion (to >99.9% in UCSD Medical Center's experience with >50,000 transfused units). These data reveal that the administration, in emergency situations, of type-specific, uncrossmatched blood to patients with no history of pregnancy or transfusion should entail relatively little risk.

Type and Screen Orders

When blood is ordered preoperatively for surgical cases in which it is unlikely that the blood will actually be transfused, the orders should be for “type and screen” only. The ABO,


Rh status of the patient is determined and the antibody screen (see previous discussion) is performed to determine the presence of antibodies other than ABO in the potential recipient's plasma. If the antibody screen is negative, type-specific otherwise uncrossmatched blood will result in a hemolytic reaction in <1/50,000 units. If the screen is positive, the blood bank will proceed to identify a pool of potentially compatible units.

Emergency Transfusions

The exsanguinating patient may require RBCs before complete compatibility testing can be performed. If testing is to be abbreviated, there is a preferred order for selecting partially tested blood (see Chapter 36). The first choice is to transfuse type-specific partially crossmatched blood or type-specific uncrossmatched blood (although verification of blood type by analysis of two separately drawn specimens must be performed before releasing any uncrossmatched blood). In urgent situations in which the patient's ABO and Rh type is unknown, group O RBCs should be administered until there is time to complete ABO and Rh testing. Rh-negative blood is preferable, particularly if the patient is a woman of child-bearing age. If Rh-negative blood is not available for a critically ill, bleeding Rh-negative patient, Rh-positive blood is frequently used. If a non–group O patient receives a large volume of group O red cells, the combined amount of anti-A and/or anti-B present in small amounts in the residual plasma of each PRBC unit may react with the patient's own A, B, or AB red cells and cause some hemolysis. For this reason, a non–group O patient who has received group O red cells approximating one patient blood volume (10 to 12 units) during the period of acute blood loss should not be switched back to his or her own ABO group until testing has been performed to confirm that significant titres of anti-A or anti-B antibodies are not present. That testing is typically performed automatically by contemporary blood banks. When FFP or TP (see later discussion) are administered prior to ABO typing, type AB plasma is preferable,90 although sometimes not feasible because of limited supply (Table 16-12).


Platelets are separated from plasma by centrifugation.25 More than 70% of the platelets used in the United States are now derived by apheresis. A single apheresis unit (referred to asapheresis platelets), which is obtained from a single donor at a single session, supplies 3 × 1011 platelets in a volume of 200 to 400 mL. An apheresis unit supplies the equivalent of the platelets derived by concentrating platelets from six to eight individual donor units of whole blood. The latter, when combined, are referred to as platelet packs or concentrates. The use of apheresis platelets substantially reduces donor exposures with the attendant risks of alloimmunization and infection (viral and bacterial). Platelet viability is optimal at 22°C. Although platelets are potentially viable for as long as 10 days (the normal in vivo lifespan), by FDA mandate, storage is limited to 5 days because of the time-related risk of bacterial growth.25 Platelets should be delivered through the standard 170-micron blood set filter. A micropore filter should not be used.

Platelets bear ABO, HLA, and other platelet-specific antigens. ABO compatibility is ideal, although not absolutely required. ABO incompatibility reduces platelet survival. In addition, it appears to increase immune responsiveness to HLA and other platelet surface antigens, thereby increasing the incidence of alloimmunization.25 ABO/HLA-matched platelets, crossmatched platelets, and HLA antigen-negative platelets can be used for patients who become refractory to random donor platelets. Platelets do not carry the Rh antigen. However, administration of platelets from an Rh-positive donor to an Rh-negative female of child-bearing age should be avoided if delay does not impose hazard in order to prevent sensitization as a result of passenger RBCs in the platelet preparation. The sensitization risk is small because of the very limited number of RBCs in contemporary platelet preparations and is effectively prevented by Rh immune globulin, which should be administered. ABO compatibility of platelets is also desirable because the antibodies present in the plasma phase can cause hemolysis of recipient RBCs.148 Hemolytic events have invariably involved administration of O type platelets to a non-O recipient, and blood bank procedures typically avoid large-volume administration of those pairings.148

Fresh-Frozen Plasma/Thawed Plasma

Plasma is separated from the RBC component of whole blood by centrifugation. One unit has a volume of 200 to 250 mL. It will contain the preservative added at the time of collection, usually CPDA. To preserve the two labile clotting factors (V and VIII), FFP is frozen promptly and thawed only immediately prior to administration. FFP must be ABO compatible. Avoiding Rh-positive plasma in Rh-negative patients seems unnecessary because there has been no reported instance of alloimmunization in over 40 years.

TP is being ever more widely used in lieu of FFP, especially in the management of trauma. (Many of the readers of this chapter may discover that they have already been administering TP.) It is obtained from thawed FFP that is maintained at 6°C for a maximum of 5 days. Its advantage is immediate availability (and reduction of wastage of thawed FFP not administered within 24 hours). Levels of FV and FVIII decline during storage. However, it is believed that there is sufficient residual FV even after 5 days to achieve FV levels of 25 to 30% readily. Factor VIII is an acute phase reactant that is usually present in sufficient amounts in trauma victims. TP can be used interchangeably with FFP in most situations, with the exception of patients with specific deficiencies of FV or FVIII, in DIC, and in neonates. Some clinicians will encounter FP24 (plasma frozen within 24 hours after phlebotomy), which is typically used interchangeably with FFP.

Solvent Detergent Plasma

One of the principal hazards of FFP administration has been virus transmission. Three procedures—pasteurization, photochemical treatment, and solvent detergent (SD) treatment—have been used to inactivate viruses. The SD technique is highly effective in inactivating all of the lipid encapsulated viruses (i.e., HIV, HCV, HBV, and HTLV). The disadvantage of the SD technique is that the process involves pooling of large numbers of single FFP units (>1,000) and is not effective against non–lipid-enveloped viruses (HAV, parvovirus) or the agent of CJD. The concern with SD plasma is that the pooling process might result in wide dissemination of an infectious agent. The incidence of parvovirus viremia among donors is estimated to be nearly 1.0%.149 Parvovirus B19 infection has been reported as a consequence of transfusion. While the disease is usually self-limited, significant morbidity, such as red cell aplasia and/or meningitis, especially in immunocompromised patents can occur.150 SD plasma is now tested for B19 and HAV and is widely used in Europe but is no longer available in the United States.



Cryoprecipitate is the precipitate that remains when FFP is thawed slowly at 4°C. It is a concentrated source of FVIII, FXIII, vWF, and fibrinogen. One unit of cryoprecipitate (the yield from one unit of FFP) contains sufficient fibrinogen to increase fibrinogen levels by 5 to 7 mg/dL.151 Accordingly, it is usually provided in bags that contain 10 or 20 units. ABO compatibility is not essential because of the limited antibody content of the associated plasma vehicle (10 to 20 mL). Viruses can be transmitted with cryoprecipitate. It is stored at -20°C and thawed immediately prior to use.

Factor VIII and IX

Recombinant and virally inactivated plasma-derived FVIII and FIX concentrates are available.152

Antithrombin III

Virus-treated antithrombin III (AT) concentrates are available. They can be used in the treatment of congenital and acquired AT deficiencies, including heparin resistance, DIC, and fulminant hepatic failure.153,154,155

The Hemostatic Mechanism

Normal hemostasis involves a series of physiologic checks and balances that assure that blood remains invariably in a liquid state as it circulates throughout the body but, once the vascular network is violated, transforms rapidly to a solid state. That transformation to a solid state (i.e., coagulation) must inevitably be complemented by processes for eliminating clot that is no longer needed for hemostasis. The latter is accomplished by fibrinolysis.

The Nomenclature of Coagulation

The nomenclature of coagulation is unfortunately complex. The first 4 of the 12 originally identified factors are usually referred to by their common names—fibrinogen, prothrombin, tissue factor (TF), and calcium—and not by their Roman numerals. FVI no longer exists; it proved to be activated FV. The more recently discovered clotting factors (e.g., prekallikrein and high-molecular-weight kininogen) have not been assigned Roman numerals. Some factors have more than one name (Table 16-13).

The Coagulation Mechanism

The classic dual cascade (intrinsic and extrinsic pathway) model of coagulation (Fig. 16-1) is now recognized to be an inadequate representation of in vivo coagulation. It fails to explain several clinical phenomena. First, persons lacking FXII, prekallikrein, or high-molecular-weight kininogen do not bleed abnormally, suggesting that contact activation is not critical for in vivo hemostasis. Second, patients with only trace quantities of FXI withstand major trauma without unusual bleeding, and those completely lacking factor XI (hemophilia C) have only a mild hemorrhagic disorder. FXI therefore appears to have a more minor role in coagulation than ascribed to it by classic theory. Next, deficiencies of FVIII and FIX (both intrinsic pathway factors) lead to hemophilia A and B, respectively. However, the classic description of two pathways of coagulation leaves it unclear clear why either type of hemophiliac could not simply clot via the unaffected pathway. Most importantly, it is now appreciated that while the classic theories may provide a reasonable model of in vitro coagulation tests (i.e., the aPTT and PT), they fail to incorporate the central role of cell-based surfaces in the in vivo coagulation process. The three stages of that process that have been thoroughly defined and described by Hoffman and


Monroe156 are summarized in the following sections and in Figure 16-2.

Table 16-13 Factor Nomenclature and Half-Lives











Tissue factor, thromboplastin



Calcium ion



Proaccelerin, labile factor



Serum prothrombin conversion accelerator, stable factor



Antihemophilic factor (AHF), AHF-A, factor VIII:C



von Willebrand factor



Christmas factor, AHF-B



Stuart-Prower factor, Stuart factor, Autoprothrombin



Plasma thromboplastin antecedent, AHF-C



Hageman factor, contact factor



Fibrin stabilizing factor



Fletcher factor


HMW kininogen

Fitzgerald, Flaujeac, or Williams factor; contact activation cofactor


HMW, high molecular weight.

Figure 16-1. The classic intrinsic and extrinsic pathways of coagulation. Intrinsic pathway (left). A cascade initiated by contact with a foreign surface (contact activation) leads to the formation of fibrin (Ia). Extrinsic pathway (right). This pathway, also leading to fibrin formation, is depicted as it was originally thought to occur, that is, largely extravascularly and independent of the classic intrinsic pathway (cf: Fig. 16-2). The dotted arrows indicate the occurrence of an enzymatically mediated conversion of an inactive factor to its active form. The shaded spheroids represent the procoagulant surfaces provided in the extrinsic pathway both in vivo and in vitro by tissue factor (TF) and, in the intrinsic pathway, by phospholipids in vitro and platelets in vivo.


Activation of the coagulation process begins when a breach in the vascular endothelium exposes blood to TF. TF is a membrane-bound protein, with adjacent membrane phospholipids, that is constitutively expressed in extravascular tissue, principally on fibroblasts (Fig. 16-2A). TF also appears on the surface of vascular endothelium and circulating monocytes in response to mechanical injury or inflammation.157 TF activates FVII (Fig. 16-2B) to yield a complex of TF and activated FVII (FVIIa) on the phospholipid surface. The TF-VIIa in turn activates FX, yielding a complex of TF-VIIa-Xa (Fig. 16-2C). The FXa, still on the phospholipid surface, then binds with FVa to form the “prothrombinase complex.” The FVa that participates in this reaction is liberated from the alpha granules of platelets that were activated at the site of injury as a result of binding to subendothelial vWF (Figs. 16-2Eand 16-3). The prothrombinase complex catalyzes the conversion of prothrombin (FII) to thrombin (FIIa; Fig. 16-2E). However, generation of IIa by this pathway is limited by tissue factor pathway inhibitor (TFPI). TFPI, a protein that is constitutively present in endothelium and platelets,158 binds to and inhibits the Xa component of the TF-VIIa-Xa complex and, once bound, inhibits adjacent TF-VIIa complexes from further activation of FX158 (Fig. 16-2D). As a consequence, only very limited amounts of thrombin can be generated by this mechanism (which explains why hemophiliacs bleed in spite of an intact intrinsic pathway). But this initial formation of small amounts of thrombin is sufficient to advance the coagulation process to the more efficient “amplification” phase that follows.


While it was the surface provided by membrane-bound TF and adjacent phospholipid that initiated the coagulation process, it is now the phospholipid surface provided by platelets that serves to perpetuate it. The breach in the vascular tree that


began the activation process also exposed platelets to collagen to which they become bound via vWF and the GPIb receptor on the platelet surface (Fig. 16-4). That binding results in platelet surface changes, most notably the appearance of the GPIIb/IIIa receptor, and in the release of the contents of alpha and dense platelet granules (Fig. 16-3).159 The latter contain numerous substances that contribute to additional platelet activation (adenosine diphosphate [ADP], Ca++, serotonin), platelet aggregation (vWF, fibronectin, thrombospondin, fibrinogen), clot formation and stabilization (calcium, fibrinogen, factors V, XI and XIII, plasminogen activator inhibitor [PAI-1]), and to adhesion and activation of additional platelets. The thrombin just generated by the TF-bound prothrombinase complex supports the amplification of the coagulation process in four ways. First, thrombin, a serine protease, further activates the adjacent platelets (Fig. 16-2F) via protease-activated surface receptors.160 Thrombin's second effect is to promote the activation FV in plasma to FVa (Fig. 16-2F). Third, thrombin releases circulating FVIII from its carrier molecule (vWF) and activates it (Fig. 16-2G). Fourth, thrombin activates FXI. FXIa in turn activates FIX (Fig. 16-2H). Note that some FIXa was also generated by the TF-VIIa during the activation phase (Fig. 16-2C). This may explain why FXI deficiency results in such a minor coagulation disturbance. The net result of this amplification stage is the availability of additional activated platelets and activated Factors V, VIII, and IX.

Figure 16-2. The coagulation mechanism. See text for details. TF, membrane-bound tissue factor on a extravascular cell surface; TFPI, tissue factor pathway inhibitor; vWF-VIII:C, circulating factor VIII bound to its carrier protein, the von Willebrand factor.

Figure 16-3. Platelet release reaction. Platelets undergo a release reaction in response to adherence to the subendothelium or to physiologic agonists including epinephrine, adenosine diphosphate (ADP), and thrombin. The numerous substances released from the alpha and dense granules of platelets contribute to additional platelet activation (ADP, Ca++, serotonin), platelet aggregation (von Willebrand factor [vWF], fibronectin, thrombospondin, fibrinogen), and clot formation (calcium, fibrinogen, factors V, XI and XIII, plasminogen activator inhibitor [PAI-1]). ATP, adenosine triphosphate; TFPI, tissue factor pathway inhibitor; GP, glycoprotein; EC, endothelial cell.

Figure 16-4. Platelet adhesion and aggregation. When the endothelium is denuded, platelets adhere to the collagen in the subendothelium via their glycoprotein glycoprotein (GP) 1b receptors and von Willebrand factor, present in both plasma and the subendothelial matrix. Platelets aggregate to one another by cross-linking via fibrinogen (or von Willebrand factor, not shown) between GPIlb/IIIa receptors expressed on the platelet surface during the process of platelet activation.


The platelet then provides the phospholipid surface on which two coagulation factor complexes form and act to produce the explosive generation of thrombin. First, FVIIIa and FIXa form the “tenase complex,” which activates FX (Fig. 16-2H). The resultant FXa forms additional prothrombinase complex (Xa-Va), and large amounts of thrombin are elaborated (Fig. 16-2J). (For mnemonic purposes it is “eight-nine and nickel-dime” that together are responsible for the thrombin burst.) Thrombin (FIIa) catalyzes the formation of fibrin from fibrinogen, and fibrin acts to crosslink the platelets, largely via the IIb/IIIa receptors (Fig. 16-4), to reinforce the friable platelet plug. Thrombin also activates FXIII (Figs. 16-2K and16-5) and thrombin-activatable fibrinolysis inhibitor (TAFI; Fig. 16-5). Fibrin


monomers initially aggregate relatively loosely to form clot composed of fibrin S (soluble), which is held together only by hydrogen bonds. FXIII (fibrin-stabilizing factor) mediates the formation of covalent peptide bonds between the fibrin monomers. FXIII may be an underappreciated cause of clinical coagulation disturbance.161 TAFI functions to prevent lysis of the newly formed clot (Fig. 16-5). In the presence of subnormal amounts of thrombin, although fibrin clot can form, it may not achieve normal strength and stability162 and may not be protected by adequate concentrations of TAFI.163

Figure 16-5. The formation and lysis of fibrin. Fibrin is formed from fibrinogen by the action of thrombin (FIIa). Thrombin also converts factor XIII (FXIII) to activated factor XIII (FXIIIa), which in turn stabilizes the evolving fibrin clot by cross-linkage. Circulating plasminogen binds to fibrin and is converted to plasmin by tissue plasminogen activator (tPA) released from normal endothelium in areas remote from sites of vascular injury. Plasmin digests fibrin to its various degradation products (FDPs). The action of tPA can be inhibited by plasminogen activator inhibitor (PAI-1) released by endothelium and platelets. The action of plasmin is also inhibited by thrombin-activated fibrinolysis inhibitor (TAFI).

In vivo, coagulation is initiated principally by contact of factor VII with extravascular TF leading first to the generation of small amounts of thrombin. Thereafter, activated clotting factors, acting intravascularly on the phospholipid surface provided by activated platelets, lead to the generation of large amounts of thrombin.

Additional Principles of Coagulation

A few additional facts will aid in achieving a broader understanding of coagulation.

1.   Most clotting factors circulate in an inactive proenzyme, or zymogen, form. During the process of coagulation, a portion of the molecule is cleaved off, resulting in active enzymes (designated by the addition of a lower case “a” after the Roman numeral, e.g., Xa), most of which are serine proteases.

2.   Most clotting factors are synthesized by the liver. The probable exception is factor VIII, which probably also has some extrahepatic synthesis.

3.   Factor VIII is actually a large, two-molecule complex (vWF and coagulant factor VIII). Factor VIII circulates as a very large complex of two distinct protein components. The high-molecular-weight portion (VIIIR:Ag) encompasses both the FVIII antigen and vWF. The vWF portion serves as a carrier protein for the second and smaller component of this macromolecular complex, VIIIC, which contains the factor VIII coagulant activity. The vWF has a second function. During the process of primary hemostasis, when the endothelial lining has been denuded, vWF in the subendothelial matrix mediates adhesion of platelets to collagen. Absence of the smaller portion of the factor VIII complex (VIII:C), results in hemophilia A. vWF deficiency causes two hemostatic abnormalities: (1) a defect in primary hemostasis because of a failure of platelet adhesion to the sites of vascular injury, and (2) the clinical equivalent of hemophilia A because of deficiency of circulating factor VIII:C. Restoration of vWF levels restores normal hemostasis. Synthesis of the vWF occurs in endothelial cells and megakaryocytes. The site of synthesis of the coagulant portion of factor VIII is unknown but may be located in the hepatic sinusoidal endothelial cells.

4.   Four clotting factors are vitamin K-dependent. Factors II, VII, IX, and X require vitamin K for completion of their synthesis in the liver. Each undergoes a final enzymatic addition of a carboxyl group that requires the presence of vitamin K. The carboxyl group enables these factors to bind (using calcium as a cofactor) to phospholipid surfaces. Without vitamin K, factors II, VII, IX, and X are produced in normal amounts but are nonfunctional.

The anticoagulant action of vitamin K antagonists is the result of their ability to inhibit this final carboxylation step. The warfarin-like drugs compete with vitamin K for binding sites on the hepatocyte. With sufficient warfarin administration, vitamin K is displaced and the vitamin K-dependent factors are not carboxylated. Of the four vitamin K-dependent factors, factor VII has the shortest half-life. It is the first clotting factor to disappear from the circulation when a patient is given warfarin or begins to develop vitamin K deficiency.

5.   Factors V and VIII have short storage half-lives. Factors V and VIII are also referred to as the labile factors because their coagulant activity is not durable in stored blood. While PRBCs contain some residual plasma with clotting factors, massive transfusion with stored blood will nonetheless lead to a dilutional coagulopathy because of diminished activity of factors V and VIII.


Fibrinolysis serves to dissolve or remodel fibrin clots and thereby “recanalize” vessels that have been occluded by thrombosis.

The Formation of Plasmin

Plasminogen is the inactive form of the fibrinolytic enzyme plasmin. Conversion of plasminogen to plasmin is accomplished principally by tissue plasminogen activator (tPA; Fig. 16-5). Plasmin is rapidly degraded by circulating antiplasmins and therefore cannot circulate freely. Plasminogen, however, can circulate. It binds to fibrin on contact and is incorporated in the evolving fibrin clot where it is converted to plasmin by tPA. While bound plasmin is protected from attack by circulating antiplasmins, any plasmin that is released from the clot is immediately neutralized by circulating α2-antiplasmin. Thus, like the coagulation cascade, the fibrinolytic system relies on surface-mediated reactions that limit both plasmin formation and fibrinolysis to the site of vascular injury.

Plasminogen Activation

The principal activator of plasmin is tPA. tPA is synthesized by vascular endothelial cells. In the event of clot formation (which requires the presence of thrombin), thrombin forms a complex with thrombomodulin (present on the vascular endothelial surface) that activates protein C. Activated protein C (APC) stimulates the release of tPA. tPA is also released from the endothelium in response to venous occlusion, physical activity, stress, or vasoactive drugs (such as epinephrine, vasopressin, and DDAVP).164 tPA binds to the adjacent fibrin and converts plasminogen to plasmin (Fig. 16-5). This mechanism serves to localize fibrinolysis to the site of vascular injury, thereby preventing vascular injury at a single location from initiating widespread fibrinolysis. As a further “check” on the fibrinolytic process, the vascular endothelium and platelets also synthesize an inhibitor of tPA, PAI-1, which reduces the amount of plasmin formed and serves to slow the fibrinolytic process (Fig. 16-5). Some patients with thrombotic disorders have been found to have increased levels of this inhibitor.164 A similar inhibitor is found in placental tissue, and it may be that the progressive hypercoagulable state associated with pregnancy is related to increased levels of this tPA inhibitor.164

There are other plasminogen activators. Urokinase is present in prostatic tissue and urine but not in circulating blood. Physiologic activators of the fibrinolytic system include vigorous exercise, anoxia, and stress. Exogenous plasminogen activators include streptokinase, urokinase, and recombinant tPA. These fibrinolytic agents all differ with respect to their action, clot specificity, systemic fibrinolytic effect, antigenic effect, and efficacy. Proteins derived from streptococci and staphylococci have also been found to be activators of the fibrinolytic system. The therapeutic fibrinolytic agents, streptokinase and urokinase, differ from tPA in that they will activate circulating plasminogen. These lead to more widespread fibrinolysis. Fibrinolytic therapy has been used in the treatment of unstable


angina, acute thrombotic stroke, acute peripheral arterial occlusions, deep vein thrombosis, pulmonary embolism (PE), and occluded indwelling catheters and arteriovenous shunts.

Plasmin Inactivation/Inhibition

Under normal circumstances, free plasmin is rapidly inactivated by antiplasmins. In the event of deficiency of α2-antiplasmin or when antiplasmin capacity is exceeded in primary fibrinolysis or DIC, plasmin circulates. Circulating plasmin will contribute to the bleeding diathesis because plasmin, in addition to degrading fibrin, is a serine protease that can also degrade other coagulation process components including fibrinogen, FV, FVIII, FXIII, vWF, and the GPIb platelet receptor.164

Fibrin Degradation Products

The structure of the products of fibrin breakdown, called fibrin degradation products (FDPs) or fibrin split products (FSPs), varies according to whether plasmin cleaves fibrinogen, fibrin that is cross-linked, or fibrin that is not cross-linked. FDPs are removed from the blood by the liver, kidney, and reticuloendothelial system. If they are produced at a rate that exceeds their normal clearance, they will accumulate. In high concentrations, FDPs impair platelet function, inhibit thrombin, and prevent the cross-linking of fibrin strands. The defective polymerization of the fibrin monomers results in a clot that is more readily degraded by plasmin.164

Under normal conditions, plasmin is generated only at the site of clot formation and is destroyed rapidly once released into the circulation. This localization process fails at times of accelerated fibrinolysis (DIC, primary fibrinolysis).

Control of Coagulation—The Checks and Balances

Coagulation must be precisely regulated to prevent rampant, uncontrolled clotting, such as that which occurs with DIC. Several mechanisms regulate and control coagulation.

Endothelial Inhibition

The first line of defense is the vascular endothelium. The intact endothelium has antithrombotic properties that serve to limit both platelet aggregation and coagulation and to induce fibrinolysis should a clot begin to form on normal endothelium. These properties are summarized in Table 16-14.

Table 16-14 Endothelial Control of Platelet Aggregation, Coagulation, and Fibrinolysis

·         Endothelial control of platelet aggregation

o    Synthesis of prostacyclin

o    Synthesis of ADPases and nitric oxide

·         Endothelial inhibition of coagulation

o    Synthesis of thrombomodulin

o    Synthesis of heparan sulfate

·         Endothelial control of fibrinolysis

o    Synthesis of tPA

ADPase, adenosine diphosphatase; tPA, tissue plasminogen activator.

1.   The thromboxane-prostacyclin balance. Primary hemostasis is, in part, controlled by the balance between the effects of two prostaglandins, thromboxane A2 (TxA2) and prostacyclin. TxA2 is synthesized at the site of vascular damage by activated platelets. TxA2 has two hemostatic effects: (1) it is a potent vasoconstrictor that limits flow to the site of injury, and (2) it stimulates additional ADP release from platelets, thereby recruiting additional platelets. Remote from the site of vascular damage, normal endothelial cells synthesize prostacyclin (Fig. 16-6). Prostacyclin has actions opposite those of TxA2. Prostacyclin inhibits platelet activation, secretion, and aggregation and is a potent vasodilator and thereby serves to prevent platelet aggregation and clot formation on the endothelial surface beyond the site of injury.

2.   Nitric oxide and adenosine diphosphatase (ADPase). The effects of prostacyclin are potentiated by nitric oxide, which is constitutively synthesized by normal endothelium and which also has vasodilatory and platelet antiaggregant effects (Fig. 16-6). As an additional means of preventing clot formation on the surface of normal endothelium, ADPases are expressed on the outer membrane of endothelial cells and serve to degrade “surplus” ADP that might otherwise initiate platelet aggregation on normal surfaces.

3.   Heparan sulfate. One of the constituents of the mucopolysaccharide glycocalyx that covers normal endothelium is a naturally occurring heparinlike substance, heparan sulfate (Fig. 16-6). Like heparin, heparan has the ability to accelerate the binding of AT to thrombin and the other activated clotting factors of the classic intrinsic pathway. This heparan sulfate is well positioned because it is at this blood-endothelial interface that activated factors of the coagulation cascade are being generated.

Figure 16-6. Five antithrombotic mechanisms. Five mechanisms that serve to prevent unrestrained coagulation are depicted. 1. Tissue factor pathway inhibitor (TFPI) inhibits the initial activation of factor X by the extrinsic pathway. 2. A complex of thrombomodulin (TM) and thrombin (IIa) activates protein C, which, with protein S (Prot S) as a cofactor, inhibits activated factors V and VIII. 3. Intact vascular endothelium releases several substances that have a platelet-inhibiting or clot-lysing effect, including nitric oxide (eNO), prostacyclin (PgI2), adenosine diphosphatase (ADPase), and tissue plasminogen activator (tPA). 4. In addition to TM, other coagulation-inhibiting substances including heparan sulphate and dermatan sulphate (latter not shown) are present in the intact glycocalyx. 5. Antithrombin III binds, and thereby inhibits, several activated clotting factors (XIIa, XIa, IXa, Xa, and IIa).


4.   Thrombin, thrombomodulin, and proteins C and S. Thrombin, in a negative feedback process, can decrease its own synthesis by inhibition of factors V and VIII. That inhibition is accomplished via protein C. Protein C circulates in plasma as an inactive precursor. Thrombomodulin is a glycoprotein located on the vascular endothelial cell surface (Fig. 16-6). The binding of thrombin to thrombomodulin alters the thrombin molecule such that it can no longer directly activate clotting factors V and VIII or catalyze the conversion of fibrinogen to fibrin. In addition, the thrombin-thrombomodulin complex rapidly converts protein C to activated protein C (APC). APC, with protein S as a cofactor, cleaves and inactivates factors Va and VIIIa (Fig. 16-6). Like protein C, protein S is vitamin K-dependent. Where the endothelium is intact, the thrombomodulin-thrombin-protein C interaction will inhibit coagulation and maintain the “nonthrombogenic” property of the endothelial lining. Where the endothelium has been stripped away or damaged, this anticoagulant mechanism will be absent and clotting can continue unopposed.

5.   Endothelial synthesis of tPA. Endothelial synthesis of tPA is one of several mechanisms by which the normal endothe-lial surface is maintained in a nonthrombogenic state (Fig. 16-6). Should clot begin to form on the normal endothelial surface, the associated thrombin induces the release of tPA, which, in the absence of other promoters of coagulation, leads rapidly to dissolution of the incipient clot.

Other Modulators of Coagulation

Several additional factors serve to limit and localize clot formation. First, the clotting factors themselves circulate in an inactive form. Once activated at an injury site, normal blood flow dilutes their concentration and clears them away from sites of injury, limiting clot formation. Activated clotting factors are preferentially removed from the circulation by the liver and the reticuloendothelial system. Finally, most of the interactions of the coagulation pathway require the presence of a phospholipid surface, which localizes clot formation to those surfaces (TF, activated platelets). Several specific coagulation-inhibiting systems are operative. Five of them are depicted in Figure 16-6. TFPI and AT are described later. The others have been previously described in “Endothelial Inhibition.”

1.   Tissue Factor Pathway Inhibitor (TFPI). Superficially, the description of the cell-based coagulation mechanism still leaves in place one of the inadequacies of the classic cascade theories of coagulation, that is, if activated factor X, and subsequently thrombin, can be formed via the direct action of the VIIa/TF complex, why is it that hemophiliacs bleed? Why do they appear to be dependent on factors VIII and IX to produce activated factor X? The answer lies in a feedback inhibitor of the extrinsic pathway known as TFPI (Figs. 16-2and 16-6). TFPI, the precursor molecule of activated TFPI (TFPIa), is constitutively present on the endothelial surface and bound to circulating lipoproteins.158 It is activated by contact with the Xa-VIIa-TF complex, that is, it is not activated until coagulation has been initiated. It inactivates factor Xa and causes internalization of membrane bound VIIa/TF complexes.158 In the presence of TFPI, extensive activation of factor X appears to require the reaction sequences of the classic intrinsic pathway. The TF pathway can initiate the first flurry of thrombin generation—enough to activate platelets and stimulate cofactors V and VIII. Thereafter, continued thrombin production appears to require the action of factors VIIIa and IXa.165

2.   Antithrombin (AT). AT is a circulating serine protease inhibitor that binds to thrombin and thereby inactivates it. AT can bind and inactivate each of the activated clotting factors of the classic “intrinsic” coagulation cascade—factors XIIa, XIa, IXa, and Xa (see Fig. 16-6). The AT molecule has two critical binding sites, one of which reacts with thrombin and the other activated clotting factors and a second to which heparin can bind (see Chapter 41). In the absence of heparin, AT has a relatively low affinity for thrombin. Heparin binding to AT increases the efficiency of binding of AT to thrombin and the other factors dramatically. Congenital AT deficiency (levels 40 to 50% of normal) can lead to a prothrombotic diathesis. Acquired AT deficiency can occur with liver disease, prolonged heparin administration, nephrotic syndrome, DIC, sepsis, pre-eclampsia, fatty liver of pregnancy, oral contraceptive use, and during CPB.166,167 AT concentrates have been used in AT deficiency states, including heparin resistance.167,168,169

3.   Thrombin Activatable Fibrinolysis Inhibitor (TAFI). An additional feedback mechanism to prevent excessive fibrinolysis and premature clot breakdown exists in the form of TAFI (Fig. 16-5). TAFI is activated by low concentrations of thrombin when thrombomodulin is present or directly by greater concentrations. TAFI's role in abnormalities of hemostasis is not well-defined.163

The Complexities of the Hemostatic Mechanism

Many mechanisms interact to maintain the liquid state of the blood under normal circumstances and to transform blood into a solid clot when injury occurs. These mechanisms include numerous feedback processes. The complexity is revealed by the existence of “double agents,” which act at some times as procoagulants and at other times as anticoagulants. Chief among them is thrombin. Thrombin is primarily a procoagulant. It promotes primary hemostasis by activating platelets, and promotes coagulation by direct activation of factors V, VIII, and XIII. Thrombin, in the final step of the coagulation cascade, cleaves fibrinogen to fibrin. However, it also has anticoagulant effects. It inhibits coagulation through its interaction with thrombomodulin and protein C. APC stimulates the release of tPA from endothelial cells, and by this mechanism thrombin has a fibrinolytic effect while simultaneously activating the fibrinolysis inhibitor TAFI. Accordingly, thrombin through its effects at many stages of the feedback-controlled hemostasis process, functions as platelet proaggregant, a procoagulant, an anticoagulant, a profibrinolytic, and an anti-fibrinolytic.

The Hemostatic Mechanism: Summary

Under normal circumstances, the hemostatic mechanism is quiescent with many of the potential participants circulating in an inactive form. Only when the endothelial lining is breached is the hemostatic mechanism set in motion. With collagen and TF exposed, the intertwined processes of platelet-mediated primary hemostasis and factor-mediated coagulation begin. Vascular injury is sealed rapidly by a platelet mass into which are incorporated fibrinogen, thrombin, plasminogen, and tPA. The completion of the coagulation process converts fibrinogen into fibrin and the platelet plug is transformed into a fibrin clot. Simultaneously, several properties of adjacent intact endothelium (elaboration of ADPases, prostacyclin, thrombomodulin, heparans, and tPA) serve to prevent extension of the clot beyond the site of injury. Within the clot, plasmin, generated by the action of tPA on the trapped plasminogen, begins the process of fibrinolysis. Over time, the entire fibrin clot dissolves, new endothelial cells line the vessel, and flow is restored.


Laboratory Evaluation of the Hemostatic Mechanism

Laboratory Evaluation of Primary Hemostasis

Platelet Count

A platelet count should be the first test ordered in the evaluation of primary hemostasis. The platelet count is quick, accurate, and reproducible. However, it reveals only platelet numbers and gives no information regarding their function. Normal platelet counts range between 150,000 and 440,000/mm3. Counts below 150,000/mm3 are defined asthrombocytopenia. Spontaneous bleeding is unlikely in patients with platelet counts >10,000 to 20,000/mm3. With counts from 40,000 to 70,000/ mm3, bleeding induced by surgery, may be severe. A detailed review of the many methods for testing platelet function is available.170 Only the more widely used methods are mentioned here.

Bleeding Time

The Ivy bleeding time (BT) is the most widely accepted clinical test of platelet function. A blood pressure cuff is placed around the upper arm and inflated to 40 mm Hg. A cut is made on the volar surface of the forearm and the wound blotted at 30-second intervals until bleeding stops. The Simplate Bleeding Time (Organon Telenika Corp., Durham, NC) device, which uses a spring-loaded lancet, standardizes the size and depth of the cut. The normal range is 2 to 9 minutes. Variations in venous pressure, blotting technique, and patient cooperation result in a lack of precision and reproducibility that make this test somewhat less reliable than other coagulation tests. The BT is purported to evaluate the time necessary for a platelet plug to form following vascular injury. This requires a normal number of circulating platelets, platelets with normal function (which can adhere and aggregate), and an appropriate platelet interaction with the blood vessel wall. A prolongation of the BT may be because of (1) thrombocytopenia, (2) platelet dysfunction (adhesion, aggregation), and (3) vascular abnormalities such as scurvy or the Ehlers-Danlos syndrome. BTs are prolonged in patients with many conditions that cause platelet dysfunction (e.g., use of aspirin, uremia). However, prolonged BTs have been observed with numerous disorders that are not associated with platelet dysfunction, such as vitamin K deficiency of the newborn, amyloidosis, congenital heart disease, the presence of factor VIII inhibitors, or anemia.171 Whether or not the BT test represents a specific measure of in vivo platelet function is much debated. The test is unpleasant for the patient and leaves a small scar. In spite of the correlation of BT with conditions known to influence platelet function, and in spite of BT quite reliably becoming progressively prolonged as platelet count falls below 80,000/µL, there are no convincing data to confirm that BT is a reliable predictor of the bleeding that will occur in association with surgical procedures.

Platelet Aggregometry

Platelet aggregometry quantifies platelet aggregation either spectrophotometrically or by impedance changes in response to stimulation with ADP, epinephrine, collagen, arachidonic acid, or ristocetin. The tests are sufficiently well standardized to allow distinctions among normal function, drug-related impairment of function, and intrinsic platelet defects. However, the tests are time-consuming and require absolutely fresh blood, and are therefore not widely used in acute patient management.170

The Platelet Function Analyser

The PFA-100 (Dade-Behring, Marburg, Germany) is a point of care, flow cytometry device. The test is based on the time to occlusion as the specimen passes through a small aperture impregnated with platelet activators (e.g., collagen, ADP). In one investigation, it proved less sensitive to the effect of aspirin than aggregometry.172 The PFA-100 has also been reported to be very insensitive to the platelet-inhibiting effect of clopidogrel.172,173 Its predictive value has not been well confirmed, and a report by the Platelet Physiology Subcommittee of the Scientific and Standardization Committee of the International Society on Thrombosis and Hemostasis offered the opinion that, “Although the PFA-100 closure time is abnormal in some forms of platelet disorders, the test does not have sufficient sensitivity or specificity to be used as a screening tool for platelet disorders.”174

Clot Retraction

Clot retraction is another function of platelets that can be assessed grossly and by thromboelastography. When maintained at 37°C, a clot should begin to retract within 2 to 4 hours. This test is difficult to quantify and only qualitative results (retraction vs. no retraction) are usually reported.

Laboratory Evaluation of Coagulation

When blood is placed in a glass test tube, clot formation occurs in response to contact with the foreign surface. No exogenous reagents are required because all of the factors necessary for contact initiated coagulation are “intrinsic” to blood. The time to formation of a clot via this pathway can be prolonged by deficiencies of any factors in the classic intrinsic pathway. However, the observation that, even in hemophiliacs, the addition of thromboplastin (now more commonly called TF) to the test tube could shorten the time to clot formation suggested the presence of an alternative pathway of fibrin formation. That pathway required the addition of something “extrinsic to blood” and did not require the presence of factors VIII or IX. In 1936, when Quick introduced the prothrombin time (PT) to clinical medicine, sufficient “thromboplastin” was used to yield a clot formation time of approximately 12 seconds. Under these circumstances, even patients lacking factors VIII or IX showed normal clot formation times.175 However, when “dilute” (partial) thromboplastin, which lacked the TF-equivalent activity necessary to activate FVII, was used in lieu of the “12-second reagent,” hemophiliacs showed much longer clotting times than did healthy controls. The two different pathways could be tested individually. With “complete thromboplastin,” coagulation proceeds via reactions that are independent of factors VIIIa and IXa. With “partial thromboplastin,” coagulation must proceed via a sequence of reactions that requires factors VIII and IX. For both tests, calcium is added because of the chelating agent in the blood specimen container. The time to fibrin strand formation is then measured.

Prothrombin Time

The PT measures the time to fibrin strand formation via a short sequence of reactions involving only TF, factors VII, X, V, II (prothrombin) and I (fibrinogen), that is, the classic extrinsic coagulation pathway (Fig. 16-1). The normal PT is 10 to 12 seconds and will be prolonged by deficiencies, abnormalities, or inhibitors of factors VII, X, V, II, or I. The PT has limitations. First, it is not very sensitive to deficiencies of any of these factors. The coagulant activity of these factors must drop to 30% of normal before the PT is prolonged. The PT is most


sensitive to a decrease in FVII and least sensitive to changes in prothrombin (FII). When prothrombin levels are only 10% of normal, the increase in the PT may be only 2 seconds. PT will not be prolonged until the fibrinogen level is below 100 mg/dL. If the aPTT (see later discussion) is normal, then a prolonged PT is most likely to represent a deficiency or abnormality of factor VII. Because FVII has the shortest half-life among the clotting factors synthesized in the liver, it is the factor that first becomes deficient with liver disease, vitamin K deficiency, or warfarin therapy. Prolongation of the PT may also be due to deficiencies of multiple factors. However, when multiple factor deficiencies occur, the aPTT is usually also prolonged.

International Normalized Ratio

The variation in thromboplastin reagents used resulted in wide variation in normal values and made comparison of PT results between laboratories difficult. The INR was introduced to circumvent this difficulty.176 Each thromboplastin is compared with an internationally accepted standard thromboplastin and assigned an International Sensitivity Index. PT test times obtained with individual reagents can thereby be normalized and reported as an INR.177

Activated Partial Thromboplastin Time

The aPTT assesses the function of the classic intrinsic and final common pathways (Fig. 16-1). Patient blood is combined with three reagents. In addition to calcium, there is a contact activator (e.g., diatomaceous earth, kaolin, celite, and ellagic acid) on the basis of which the test is called an activated PTT; and a partial thromboplastin (often a phospholipid extracted from rabbit brain or human placenta), which substitutes for the phospholipid surface provided by platelets in vivo. The aPTT will reveal deficiencies, abnormalities, or inhibitors of one or more coagulation factors: high-molecular-weight kininogen (HMWK), prekallikrein, XII, XI, IX, VIII, X, V, II, and I. Surface activation in the laboratory parallels the (clinically relatively unimportant) contact activation phase involving factors XII and XI, prekallikrein, and HMWK that initiates the intrinsic pathway in vivo. Normal aPTT values are between 25 and 35 seconds. The aPTT is prolonged when there is a deficiency, abnormality, or inhibitor of factors XII, XI, IX, VIII, X, V, II, and I (i.e., all factors except VII and XIII). The aPTT is most sensitive to deficiencies of factors VIII and IX, but, as is the case with the PT, levels of these factors must be reduced to approximately 30% of normal values, before the test is prolonged. The assay is also very sensitive to inhibition of thrombin (e.g., by unfractionated heparin and direct thrombin inhibitors). Heparin initially prolongs the aPTT, but with high levels will also prolong PT. As with the PT, the level of fibrinogen must be reduced to 100 mg/dL before the aPTT is prolonged. FXII deficiency, which is a relatively common cause of aPTT prolongation, does not cause a clinical coagulopathy. FXIII deficiency, which is associated with a significant bleeding diathesis, does not alter aPTT (or any other common coagulation test). aPTT results (like those of the PT) vary from laboratory to laboratory because of nonstandardization of the phospholipids and activators.

Activated Clotting Time

The activated clotting time (ACT) is similar to the aPTT in that it depends on factors that are all “intrinsic” to blood (the classic intrinsic pathway of coagulation; see Chapter 41). Fresh whole blood is added to a test tube that contains a particulate surface activator of factors XII and XI. The time to clot formation is measured. Neither partial thromboplastin nor phospholipid substitute is added. Coagulation therefore depends on adequate amounts of platelet phospholipid being present in the blood sample. The automated ACT is widely used to monitor heparin therapy in the operating room. Normal values are in the range of 90 to 120 seconds. The ACT is less sensitive than the aPTT to factor deficiencies in the classic intrinsic coagulation pathway.

Thrombin Time

TT, also called thrombin clotting time, is a measure of the ability of thrombin to convert fibrinogen to fibrin. This test, which is performed by adding exogenous thrombin to citrated plasma, bypasses all the preceding reactions. TT may be prolonged by conditions that affect either the substrate, fibrinogen, or the action of the enzyme, thrombin. TT is prolonged when there is an inadequate amount of fibrinogen (<100 mg/dL) or fibrinogen is abnormal (dysfibrinogenemia), as in advanced liver disease. Thrombin's enzymatic function can be inhibited by heparin (complexed to antithrombin III), direct thrombin inhibitors (see later discussion), FDPs (see previous discussion), or by inhibitors that may occur in patients with plasma cell myeloma and other immunoproliferative conditions.178 The normal TT is 10 to 15 seconds.

Reptilase Time

When TT is prolonged, the reptilase time can be used to differentiate between the effects of heparin and FDPs. Reptilase, which is derived from snake venom, converts fibrinogen to fibrin. The action of reptilase is unaffected by heparin but is inhibited by FDPs. A prolonged TT and a normal reptilase time suggest the presence of heparin. Prolongation of both TT and reptilase time will occur in the presence of FDPs, or when fibrinogen level is low. The normal reptilase time is 14 to 21 seconds.

Ecarin Clotting Time

Direct thrombin inhibitors (DTIs) such as hirudin, lepirudin, argatroban, and bivalirudin are frequently used in patients with heparin-induced thrombocytopenia/thrombosis (HIT/T). At low DTI concentrations, TT, aPTT, and ACT provide reasonable correlations with DTI concentration, and on the limited occasions when monitoring is deemed necessary (most often patients in renal failure), the aPTT is commonly used. But with the levels required for CPB, the correlation becomes poor and the risk of overdose with these agents, for which there are no antagonists, becomes significant. The ecarin clotting time provides a better correlation and can be used for monitoring in that context.179 The test employs the venom of the saw-scaled (also known as sawtooth) viper (Echis carinatus). A metalloprotease in the venom converts normal prothrombin to a form (meizothrombin) that is still capable of converting fibrinogen to fibrin but that is inhibited by DTIs in a reliably dose-dependent manner.180 A thromboelastographic method in which ecarin is used to initiate coagulation has also been reported to provide a much better correlation with bivalirudin levels than the ACT.181

Anti-Xa Activity Assay

The anti-Xa activity assay is used to monitor the effects of low-molecular-weight heparins, indirect Xa inhibitors and occasionally unfractionated heparin. Patient plasma is mixed with a reagent containing a known amount of Xa and excess antithrombin. A chromogenic substrate of Xa is added, and a color change reaction occurs in proportion to the Xa not bound by anti-Xa activity in the patient's serum.

Fibrinogen Level

Normal fibrinogen values are between 160 and 350 mg/dL. Below 100 mg/dL, fibrinogen may be inadequate. Fibrinogen is rapidly depleted during DIC. A marked increase in fibrinogen


may occur in response to stress, including surgery and trauma. Levels in excess of 700 mg/dL may occur. Because of this increase, in spite of rapid fibrinogen consumption during a hypercoagulable state such as DIC, the fibrinogen level may still appear to be “normal.”

Evaluation of Fibrinolysis-Fibrin Degradation Products and D-Dimer

The FDP test identifies the breakdown products of fibrin (cross-linked or uncross-linked) and fibrinogen. The D-dimer assay is specific for breakdown products of cross-linked fibrin. FDPs will be increased in any state of accelerated fibrinolysis, including advanced liver disease, fibrinolysis associated with CPB, exogenous thrombolytics (e.g., streptokinase), and DIC. D-dimer is specific to conditions in which extensive lysis of the cross-linked fibrin of a mature thrombus is occurring, as occurs in DIC, but also with deep vein thrombosis (DVT) and PE.

The Thromboelastogram

Thromboelastography provides a measure of the mechanical properties of evolving clot as a function of time. A principal advantage is that the processes it measures require the integrated action of all the elements of the hemostatic process: platelet aggregation, coagulation, and fibrinolysis. The thromboelastogram is obtained by placing a specimen of blood in a rotating cuvette containing a contact activator and calcium. (Heparinase can also be added to eliminate heparin effect.) A “piston” is lowered into the cuvette. As clot formation begins, the piston rotates as a function of the adherence of the evolving fibrin clot to the piston. The rotation of the piston results in a to-and-fro excursion of a stylus, the amplitude of which is proportional to the speed of piston rotation.

Figure 16-7 depicts a normal thromboelastogram. Several parameters are derived from the thromboelastogram. The most commonly used ones and their interpretation are as follows.182 R, the reaction time, is the interval until initial clot formation. It requires thrombin formation, and prolongation is usually indicative of an intrinsic pathway factor deficiency. K, the clot formation time, is the interval required after R for the thromboelastogram to achieve a width of 20 mm. Prolongation occurs with deficiencies of thrombin formation or generation of fibrin from fibrinogen. The alpha angle, like K, is a measure of the speed of clot formation. A decrease of the alpha angle has similar significance to a prolongation of K. MA, the maximum amplitude, is a measure of the strength of the fully formed clot. It reflects primarily platelet number and function, although it also requires proper fibrin formation to achieve normal values. MA typically occurs between 30 and 60 minutes. The (MA + x)/ MA, is the ratio of the amplitude at a specific time interval (x) after MA divided by MA, is used as a measure of the rate of fibrinolysis. The (MA + 60)/MA ratio has been used most widely.183 A ratio of <0.85 is evidence of abnormal fibrinolysis.184 In clinical practice, particularly in liver transplantation, a nonquantitative appreciation of the typical teardrop shape (Fig. 16-8) is used more often to support a diagnosis of increased fibrinolysis than are specific numerical values. F, the interval from MA to return to a zero amplitude, is a measure of the rate of fibrinolysis. F is sufficiently long in normal subjects so that the test is usually terminated before this time elapses.

Figure 16-7. The normal thromboelastogram and the variables commonly derived from it. See text for details. (From Kang Y, Lewis JH, Navalgund A, et al: Epsilon-aminocaproic acid for treatment of fibrinolysis during liver transplantation. Anesthesiology 1987; 66: 766, with permission.)

Figure 16-8. Thromboelastogram patterns seen in normal subjects and in subjects with four abnormalities of hemostasis. (From Kang Y: Monitoring and treatment of coagulation, Hepatic Transplantation: Anesthetic and Periperative Management. Edited by Winter K, Kang Y. New York, Praeger, 1986, pp 151, with permission.)

The thromboelastogram has been employed in cardiac surgery, major trauma, and hepatic transplantation. It is in the latter that it is used most frequently. Commonly, in that context, an increased R prompts the administration of FFP, a decreased MA leads to platelet administration, and the teardrop configuration of fibrinolysis leads to the administration of antifibrinolytics. The use of the thromboelastogram to guide transfusion in liver transplantation has been shown to decrease the amounts of RBCs and FFP administered.185

Interpretation of Tests of the Hemostatic Mechanism

An effective approach to the interpretation of coagulation tests is to appreciate in advance the constellation of test results (the coagulation “profile”) that is likely to occur with each of the common bleeding disorders (Table 16-15). The most commonly ordered coagulation tests are the platelet count, PT, aPTT, and occasionally BT. When a greater disruption of the hemostatic mechanism is suspected, further tests including fibrinogen, TT, and assays for FDPs and D-dimer may be ordered. Note that some significant clinical bleeding diatheses, including deficiencies of FXIII and α2-antiplasmin and mild degrees of vWD, will not be revealed by routine coagulation testing.

Because the coagulation defects that appear most often are revealed by abnormal values of PT and/or aPTT, Figure 16-9 provides an algorithm for the evaluation of those abnormalities.

Common Coagulation Profiles

1. Platelet count decreased (normal aPTT and PT). Differential diagnosis: decreased platelet production (see later discussion), excess consumption, platelet destruction, or sequestration in the spleen (see bleeding disorders, thrombocytopenia).

Figure 16-9. An approach to the evaluation of prolonged prothrombin time (PT) and/or activated partial thromboplastin time (aPTT). TT, thrombin time; Fbg, fibrinogen; DD, D-dimers; APA, antiphospholipid antibody (e.g., lupus anticoagulant, anticardiolipin, and anti-B2-GPI antibodies); DIC, disseminated intravascular coagulation. (Modified from Bombeli T, Spahn DR: Updates in perioperative coagulation: physiology and management of thromboembolism and haemorrhage. Br J Anaesth 2004; 93: 275, with permission.)



2. Prolonged BT (normal platelet count, aPTT, PT). Differential diagnosis: antiplatelet drug ingestion (e.g., nonsteroidal anti-inflammatory drugs, acetylsalicylic acid, clopidogrel), uremia, vWD (although factor VIII:C levels may be decreased with vWD [type 1], only 25 to 30% of VIII:C coagulant activity is necessary to produce a normal aPTT).

3. Prolonged aPTT (normal platelet count and PT). Differential diagnosis: heparin, the lupus anticoagulant or other antiphospholipid antibodies such as anticardiolipin and anti-B2-GPI antibodies,186 deficiency of FXII, HMWK, or prekal-likrein, hemophilia A or B, vWD, acquired factor inhibitors, and poor collection technique.

Disorders that produce this combination affect factors of the intrinsic pathway (prekallikrein, HMWK, factors XII, XI, IX, and VIII) and/or the common pathway (X, V, II, and I). With heparin therapy, initially only the aPTT is prolonged. At higher doses both the aPTT and PT are prolonged. Note that some common causes of a prolonged aPTT are not associated with a bleeding diathesis. The aPTT prolongation caused by the lupus “anticoagulant” and other antiphospholipid antibodies is the result of the binding of the phospholipid used to support coagulation in vitro. These patients actually have a prothrombotic tendency. Deficiencies of FXII, HMWK, or prekallikrein, in particular FXII, are also common causes of aPTT prolongation. They are not usually associated with a significant clinical hemostatic defect. Collection technique can prolong the aPTT either by heparin contamination or because factors V and VIII, the labile factors, may be consumed if the blood becomes partially clotted prior to delivery to the laboratory. The aPTT is very sensitive to factor VIII deficiency. When the aPTT is prolonged in isolation, is it less likely to be due to a bleeding disorder that involves multiple factor deficiencies (such as liver disease, vitamin K deficiency, the administration of warfarin, or the coagulopathy associated with massive transfusion or DIC). Heparin therapy or congenital disorders of hemostasis are more probable.

4. Prolonged PT (normal platelet count and aPTT). Differential diagnosis: vitamin K deficiency, warfarin administration, early liver dysfunction, FVII deficiency, and acquired coagulation factor inhibitors.

Because factor VII has the shortest half-life among the vitamin K-dependent factors, depletion of the vitamin K-dependent factors will first prolong the PT and only later the aPTT. Similarly, the development of liver disease will lead to deficiencies of factor VII first and initial prolongation of only the PT. With further deterioration of liver function, both the PT and the aPTT will be prolonged. Advanced liver disease can also lead to thrombocytopenia and platelet dysfunction.84 Acquired coagulation factor inhibitors are rare but can occur in patients with lymphoma or collagen vascular disease.

5. Prolonged PT and aPTT (normal platelet count). Differential diagnosis: vitamin K deficiency, warfarin, and heparin.

6. Prolonged PT, aPTT, and TT (normal platelet count). Differential diagnosis: heparin, DTIs, FDPs, hypofibrinogenemia, and dysfibrinogenemia.

Although advanced liver disease can also produce multiple factor deficiencies and this pattern, the platelet count is usually decreased. FDPs will also be elevated (see later discussion).

Simultaneous prolongation of the TT makes the diagnosis of simple vitamin K deficiency or warfarin therapy unlikely. TT is sensitive to minute levels of heparin. Addition of protamine or a reptilase time will identify heparin. FDPs may be elevated with fibrinolytic therapy, DIC, or liver disease. DIC and liver disease usually result in thrombocytopenia as well. A normal platelet count makes heparin or extensive fibrinolysis more likely.

7. Prolonged PT, aPTT, TT, decreased platelet count. Differential diagnosis: DIC, dilution by massive transfusion, liver disease, and heparin therapy.

Table 16-15 Interpretation of Coagulation Tests










N or ↓






↓ Production sequestration
↑ Consumption
Immune destruction

Radiation, chemotherapy
Extensive tissue damage







Platelet dysfunction

Drugs: ASA, NSAIDs, Clopidogrel, IIb/IIIa inhibitors; uremia; mild vWD






Severe vWF deficiency








Factor deficiency
Factor inhibition
Antiphospholipid antibody

Hemophilia A or B
Low-dose heparin, LMWHa
Poor collection technique
Lupus anticoagulant







Factor VII deficiency

Early liver disease
Early vitamin K deficiency
Early Coumadin therapy





Multiple factor deficiencies

Late vitamin K deficiency
Late Coumadin therapy
Heparin therapyb


Dilution of factors and platelets

Massive transfusion

Hypercoagulable state ± ↓ production of factors

Advanced liver disease

↑, increased; ↓, decreased; N, normal; ASA, aspirin; aPTT, “activated” partial thromboplastin time; PT, prothrombin time; TT, thrombin time; FDPs, fibrin degradation products; HIT, heparin-induced thrombocytopenia; vWF, von Willebrand factors; vWD, von Willebrand' s disease; LMWH, low– molecular-weight heparin; NSAIDs, nonsteroidal anti-inflammatory drugs; DIC, disseminated intravascular coagulation; H.I.T., heparin induced thrombocytopenia.
aaPTT prolongation is more likely to occur with LMWHs with lower Xa/IIa effect ratios, for example, tinzaparin, than with greater ratios, for example, enoxaparin.
bBleeding time may also be prolonged in association with a marked aPTT increase.
cDIC may be distinguished by the presence of D-dimers.

FDPs and D-dimer are elevated in DIC and allow differentiation from dilutional effects and excess heparin. Heparin causes thrombocytopenia only when prolonged exposure results in HIT/T. FDPs, but not D-dimer, are elevated in severe liver disease.

The interpretation of coagulation tests may be made more difficult by the fact that patients who develop a bleeding diathesis in the perioperative period may have more than one bleeding disorder (e.g., DIC and coagulopathy related to massive transfusion) and may also have a surgical cause for bleeding.

Disorders of Hemostasis: Diagnosis and Treatment

The hemostatic mechanism involves an intricate balance that serves to limit blood loss in the event of vascular injury while maintaining the liquid character of blood at other times. Under normal circumstances, an equilibrium between clotting and bleeding is maintained with the help of multiple activators, inhibitors, cofactors, and feedback loops, both positive and negative. Under pathologic circumstances, that equilibrium may be lost, leading to either hemorrhagic or thrombotic complications. Accordingly, disorders of hemostasis can be broadly classified into those that lead to abnormal bleeding and those that lead to abnormal clotting. The disorders may be further categorized according to whether they involve platelets, clotting factors, and/or the presence or absence of inhibitors (such as FDPs). Finally, disorders may be hereditary or acquired. Treatment may require administration of hemostatic blood products (platelets and/or clotting factors) or pharmacologic agents. The latter may be chosen for effects on platelets (desmopressin, antiplatelet drugs), on clotting factors (vitamin K, warfarin, heparin), or on naturally occurring inhibitors (antifibrinolytic agents, protamine, fibrinolytics).

The preoperative history is invaluable. Abnormalities of primary hemostasis, usually caused by reduced platelet number or function, will be revealed by evidence of “superficial” (skin and mucosal) bleeding including easy bruising, petechiae, prolonged bleeding from minor skin lacerations, recurrent epistaxis, and menorrhagia (see Chapter 23). Coagulation abnormalities are associated with “deep” bleeding events including hemarthroses or hematomas after blunt trauma.

Hereditary Disorders of Hemostasis

Inherited Platelet Disorders

The Bernard-Soulier syndrome involves various abnormalities of the GPIb receptor and therefore results in deficiencies of platelet adhesion (Fig. 16-3). In Glanzmann's thrombasthenia, abnormalities of the GPIIb-IIIa receptor complex result in defective platelet aggregation (Fig. 16-4). Both are extremely rare. There are other even less common abnormalities affecting virtually every phase of platelet function including synthesis of TXA2, synthesis and release of the contents of alpha and dense granules (Fig. 16-3), and receptor (ADP, TXA2) morphology and function.187

von Willebrand Disease

vWD is the most common hereditary bleeding disorder. Some form of the disease is present in approximately 1% of the general population, although it is overtly symptomatic in only about 10% of those afflicted.188 vWD is the result of the synthesis of an abnormal vWF or normal vWF in reduced amount. The vWF is a protein synthesized by endothelial cells,


megakaryocytes, and platelets. It is important for both primary hemostasis, that is, the binding of platelets to sites of vascular injury, and for coagulation, the latter through its role as a carrier protein/stabilizer for FVIII. vWF has several distinct binding domains responsible for its several hemostatic functions. Those domains include sites that are specific for collagen (for adherence to the subendothelium), for the platelet GPIb receptor (for platelet adhesion to collagen), for the platelet GP IIb/IIIa receptor (for platelet aggregation), and for factor VIII:C (for vWF's carrier protein function). There are at least 50 genetic variations of vWD, which accounts for its phenotypic heterogeneity. There are three principal subtypes. Type 1, which comprises 70 to 80% of vWD, is a quantitative defect. vWF is present but is secreted in reduced amount. Patients with type 1 vWD present with a pattern of bleeding that is characteristic of abnormalities of primary hemostasis. Type 2 vWD, which comprises 20 to 30% of patients with vWD, includes a host of qualitative defects of vWF. Some mutations affect the platelet interactions of vWF and others the factor VIII interaction. Type 2 is subdivided into four subtypes. Type 2B is characterized by a variant of the vWF that causes abnormal aggregation of platelets and thrombocytopenia. The abnormal vWF has a high affinity for the platelet GPIb receptor. The bleeding diathesis is probably the result of formation and clearance of vWF-platelet complexes and the resultant thrombocytopenia. In the 2N (Normandy) subtype, the vWF has a markedly reduced affinity for factor VIII. These patients demonstrate normal platelet function, but bleed because of decreased factor VIII coagulant activity. These patients are readily misdiagnosed as having mild hemophilia A. Type 3 vWD, which is very rare, entails a complete absence of vWF, resulting in a severe abnormality of both primary hemostasis and coagulation.

The Role of vWF in Hemostasis

vWF is essential for platelet plug formation. It mediates platelet adhesion to the subendothelial surface of blood vessels. After binding to the subendothelium, vWF undergoes a conformational change that only then allows platelets to adhere via their glycoprotein GPIb receptors. The antibiotic, ristocetin, can induce the platelet GPIb-vWF interaction and, accordingly, is the basis for one laboratory test of platelet function. vWF also participates in platelet to platelet aggregation. Platelet aggregation occurs by binding of vWF molecules to the GPIIb/IIIa receptors on the surface of several platelets. The vWF also acts as a carrier protein for the coagulant activity of factor VIII, referred to as VIII:C, with which it circulates in a complexed form that prolongs the circulation time of VIII:C.

Diagnosis and Treatment of vWD

History will commonly reveal abnormal bleeding from mucosal surfaces. Sixty percent of the patients will report epistaxis, 50 will report menorrhagia, and 35 will acknowledge gingival bleeding, easy bruising, and hematomas.189 vWD should be considered in patients who give a history of unexplained postoperative bleeding, particularly following tonsillectomy or dental extraction. Although vWD is a hereditary disease, a clear family history is not always evident because disease severity varies substantially.

Specialized laboratory tests, ideally directed by a hematologist, may be required to confirm the diagnosis and type of vWD. One or more vWF markers, including vWF factor antigen (vWF:Ag), vWF ristocetin cofactor activity (vWF:RCo), and/or vWF collagen binding activity (vWF:CB) will be diminished or absent. Because vWD is a carrier protein/stabilizer of FVIII, FVIII half-life is diminished, and FVIIII levels are characteristically also decreased. What is important for the anesthesiologist to appreciate is that the results of the most commonly ordered coagulation tests, the platelet count, the aPTT, and the PT, may be normal in the patient with vWD. Although the half-life of VIII:C is diminished in vWD, there is usually sufficient VIII:C to yield a normal aPTT in basal conditions.

The two established treatments for vWD are DDAVP (1-deamino-8-D-arginine vasopressin) and factor concentrates.190,191 DDAVP, which promotes release of vWF, is effective first-line therapy for the large majority (approximately 80%) of patients with vWD, including those with type 1 and type 2A disease.192 However, the recognition of subtype 2B (see previous discussion) is important because DDAVP will cause thrombocytopenia in these patients.193 DDAVP, given intravenously in a dose of 0.3 µg/kg, increases factor VIII:C and vWF two to fivefold in most patients. Its effect is maximal after 30 minutes, and elevated levels persist for 6 to 8 hours189,192 (see “Pharmacologic Therapy: Desmopressin”). For the 20% of patients who do not respond adequately to DDAVP, virally inactivated factor concentrates (e.g., Haemate-P) will be appropriate. Their efficacy is well confirmed.194,195Antifibrinolytic agents, ε-aminocaproic acid (EACA) and tranexamic acid (TXA), are sometimes used in combination with DDAVP to manage these patients during the perioperative period.193 These drugs may be given intravenously or orally. They have also been administered topically, as mouthwashes, in patients with vWD undergoing dental extractions. Oral contraceptives (estrogens) have been used to treat patients with vWD who have menorrhagia, or who are undergoing elective surgery.193 The mechanism of action of the estrogens is not well understood, although a connection with vWF synthesis is suspected. Antiplatelet drugs should be avoided in patients with vWD.

The Hemophilias

Hemophilia A results from mutations that lead to either deficient or functionally defective factor VIII:C. Hemophilia B (Christmas disease) and hemophilia C are caused by deficiency or abnormality of factors IX and XI, respectively.196 The relative frequencies of the three hemophilias are factor VIII:C, 85%; factor IX, 14%; and factor XI, 1%. Rare inherited deficiencies of factors II, VII, V, and X also occur.196 Both hemophilia A and B are sex-linked recessive disorders, which therefore occur almost exclusively in males. Hemophilia C is an autosomal recessive disorder that occurs almost exclusively in Ashkenazi Jews.196 About 50% of operations in hemophiliacs are orthopaedic procedures required for treatment of the arthritic consequences of hemarthroses.

Hemophilia A

Factor VIII:C circulates bound to and protected by vWF. In hemophilia A, patients have normal levels of vWF but have reduced or defective factor VIII:C. Hemophilia A occurs in approximately 1 in 10,000 males. Hemophiliacs experience deep tissue bleeding, hemarthroses, and hematuria most commonly. Patients with mild disease have factor levels of 5 to 30% of normal and usually bleed abnormally only following trauma. Patients with moderate disease have factor levels of 1 to 5% and occasionally bleed spontaneously. The great majority of hemophiliacs have the severe form of the disease. Factor VIII:C levels are <1% of normal and they frequently experience spontaneous bleeding episodes. The severity of clinical symptoms usually correlates with the level of clotting factor activity. Like the patient with vWD, hemophiliacs should avoid aspirin and other platelet-inhibiting agents.

Diagnosis and Treatment

Patients with hemophilia A will commonly report a history that reveals the X-linked recessive pattern of disease inheritance. Diagnosis is made on the basis of a prolonged aPTT and specific factor assays demonstrating a deficiency of factor VIII coagulant activity with normal levels of vWF, factor IX, and factor XI. PT and BT will be


normal. Hemophilia A is treated with plasma-derived, virally attenuated concentrates or with recombinant factor VIII.196

In the event of an episode of spontaneous bleeding (most often a hemarthrosis), a procoagulant level of 25% is a common target. For elective surgical procedures, the level of factor VIII:C activity is usually raised to 50 to 100% of normal by administration of virally inactivated factor concentrate. Many hemophiliacs develop inhibitors to factor VIII:C, which increases the amount of concentrate that will be required. Recombinant activated FVIIa (see later discussion) may be necessary for the patient with inhibitors.

DDAVP will also increase plasma factor VIII:C and vWF concentrations and is often effective in mild hemophilia A. The effect may be partly the result of “protection” of available FVIII by increased concentrations of the carrier molecule, vWF. However, DDAVP is also thought to cause the release of factor VIII:C from liver endothelial cells.197 There is a large variation in patient response to DDAVP, and it is most effective in patients with factor VIII:C levels >5%.83,192,198 It is given intravenously in a dose of 0.3 µg/kg in 50 mL of saline over 15 to 30 minutes. It causes a prompt increase in factor VIII:C. However, tachyphylaxis does develop, which limits its usefulness. The antifibrinolytics EACA and TXA have been used to treat hemophiliac patients prior to dental procedures. The agents are contraindicated in bleeding episodes involving joints or the urinary tract because the clots that do form may not be lysed for a long period of time.

Hemophilia B

Factor IX deficiency is also an X-linked recessive disorder, occurring in approximately 1/25,000 males.196 It produces a bleeding diathesis that is clinically indistinguishable from hemophilia A. Typically, minor hemorrhage is managed by achieving FIX levels of 20 to 30% of normal. Levels of 50 to 100% are sought for more severe hemorrhage and in anticipation of surgery. Recombinant and virally attenuated FIX factor concentrates are available and are the preferred treatment.

Protein C and Protein S Deficiency

Hereditary deficiencies of protein C and protein S are associated with thromboembolic events originating on the venous side of the circulation (e.g., DVT, PE, and paradoxical embolization causing stroke). The complete absence of protein C is associated with death in infancy. Patients who experience thromboembolic events and have decreased levels of protein C or protein S should remain on anticoagulant therapy indefinitely.

Acquired Disorders of Hemostasis

For mnemonic purposes, it is helpful to classify bleeding disorders according to which of the three hemostatic processes is involved: primary hemostasis (platelet disorders), coagulation (clotting factor disorders), fibrinolysis (production of inhibitors such as FDPs), or some combination of the three. Similarly, it is useful to use the results of coagulation tests to determine whether the clinical problem involves primary hemostasis (e.g., decreased platelet count, increased BT), coagulation (e.g., prolonged PT and aPTT, decreased factor levels), fibrinolysis (increased FDPs, increased D-dimer), or some combination of the three. Ultimately, therapeutic decisions (e.g., administration of platelets, FFP, or an antifibrinolytic agent) will similarly be oriented to treatment of one or more of these processes.

Acquired Disorders of Platelets

The clinical conditions that cause an isolated disorder of primary hemostasis typically involve abnormalities of either platelet number or function.


Platelets are derived from megakaryocytes in the bone marrow in response to thrombopoietin, which is synthesized by the liver. The causes of thrombocytopenia may be categorized as (1) inadequate production by the bone marrow, (2) increased peripheral consumption or destruction (non–immune-mediated), (3) increased peripheral destruction (immune-mediated), (4) dilution of circulating platelets, and (5) sequestration.

1.   Bone marrow production of platelets can be impaired in many ways. Physical and chemical agents (radiation and chemotherapy), various drugs (thiazide diuretics, sulfonamides, diphenylhydantoin, alcohol), infectious agents (hepatitis B, TB, overwhelming sepsis), and chronic disease states (uremia, liver disease) can all cause bone marrow suppression. Infiltration of the bone marrow by cancer cells or replacement by fibrosis will also result in inadequate platelet production.

2.   Accelerated nonimmunologically mediated consumption can occur in many conditions that cause extensive activation of coagulation with or without the occurrence of DIC. After extensive tissue damage (e.g., burns, crush injuries), which denude vascular endothelium, the normal process of hemostasis activates platelets and leads to their consumption and to thrombocytopenia. In a similar fashion, the interaction of platelets with nonendothelialized structures such as large vascular grafts can also lead to a transient thrombocytopenia. Platelets are consumed in patients with an extensive vasculitis such as occurs with toxemia of pregnancy. The many conditions that cause DIC (see later discussion) will also cause platelets to be consumed or destroyed more rapidly than they can be produced.

3.   Immunologically mediated consumption can be caused by various drugs (heparin, quinidine, cephalosporins, vancomycin) and autoimmune disorders (systemic lupus erythematosus, rheumatoid arthritis, thrombotic thrombocytopenic purpura). Alloimmunization resulting from previous transfusions or pregnancy can cause refractoriness to platelet transfusions.

4.   Dilution of platelets will occur in the context of massive transfusion (see later discussion and “Massive Transfusion”).

5.   Under normal conditions, approximately one third of platelets are sequestered in the spleen. When the spleen enlarges, an increasing number of platelets are sequestered and thrombocytopenia may result. This may occur with the splenomegaly associated with myelodysplastic syndromes and cirrhosis of the liver, although in the latter condition, decreased production also contributes to thrombocytopenia.

Disorders of Platelet Function


Platelet dysfunction is common in uremia. Thorough reviews are available.84,199 The accumulation of several metabolites is thought to interfere with vWF formation and release and to cause abnormal function of the GPIIb-IIIa receptor. Synthesis of prostacyclin and nitric oxide synthesis, both of which have platelet inhibitory effects, is increased in uremia. Dialysis frequently improves the hemostatic defect. There are several other potential treatment modalites.84,200 Cryoprecipitate (a source of vWF) was once used for uremic bleeding but has now been supplanted by DDAVP, which induces immediate release of vWF from endothelial cells and rapidly improves platelet adhesiveness. Severe anemia, per se, contributes to bleeding because in the lower viscosity state, platelets have a reduced tendency to travel in the periphery of the blood column, along the endothelial surface. Improvement of the hemostatic defect associated with uremia has been observed with administration of erythropoietin (probably by correction of anemia201) and conjugated estrogens84 (perhaps by reduction of nitric oxide formation).


When life-threatening bleeding occurs in the uremic patient, platelet concentrates should be administered.

Antiplatelet Agents

Numerous medications are administered expressly for the purpose of platelet inhibition to reduce the risk of MI, stroke, and other thromboembolic complications. They induce platelet dysfunction by several mechanisms, which include inhibition of cyclo-oxygenase (Cox), inhibition of phosphodiesterase, ADP receptor antagonism, and blockade of the GP IIb/IIIa receptor.

Cyclo-oxygenase Inhibitors

Aspirin is the prototype. Aspirin produces irreversible inhibition of platelet Cox, which prevents synthesis of TxA2, a potent platelet proaggregant and vasoconstrictor. In moderate doses, there is selective sparing of the synthesis of prostacyclin (antiaggregant, vasodilator), which results in “tilting” the balance substantially in favor of platelet inhibition. The platelet-inhibiting effectiveness of aspirin varies substantially. Increased rates of new platelet synthesis, simultaneous administration of other drugs that temporarily bind and thereby protect Cox-1 (e.g., ibuprofen), and polymorphisms of the Cox-1 enzyme may be responsible.202

Indomethacin, phenylbutazone, and all the nonsteroidal anti-inflammatory agents (e.g., Naprosyn, ibuprofen) also inhibit Cox. However, unlike aspirin, their inhibition is promptly reversible with clearance of the drug. The more recent Cox-2 inhibitors selectively inhibit Cox-2, the isoform responsible for generating the mediators of pain and inflammation, while sparing Cox-1, the inhibition of which causes both gastric damage and decreased renal blood flow and inhibition of platelet TxA2. Accordingly, platelet function should not be impaired. However, it has become apparent that Cox-2 inhibitors reduce prostacyclin generation by vascular endothelial cells and may thereby tilt the natural balance toward platelet aggregation. That procoagulant effect is not uniform among Cox-2 inhibitors. Celecoxib simultaneously decreases endothelial expression of TF and may thereby produce a compensatory “counter-tilt.”203 An increased rate of myocardial ischemic events resulted in the withdrawal of some Cox-2 inhibitors from the market in 2004.

Phosphodiesterase Inhibitors

Cyclic adenosine monophosphate is an inhibitor of platelet aggregation, and levels are increased by inhibition of phosphodiesterase. Dipyridamole, which is used for stroke prophylaxis (usually in combination with aspirin), and cilostazol appear to act primarily by this mechanism. Caffeine, aminophylline, and theophylline will also similarly produce mild, reversible platelet inhibition.

ADP Receptor Antagonists

Activation of the platelet ADP receptor leads to surface expression of the IIb/IIIa receptor. Clopidogrel, which is administered for prevention of stent occlusion as well as stroke and MI prophylaxis, blocks the ADP receptor in a noncompetitive and irreversible manner.

Glycoprotein IIb/IIIa Receptor Antagonists

The GPIIb/IIIa platelet surface receptor, by which fibrinogen cross-links platelets, is the final common pathway for platelet aggregation. The IIb/IIIa antagonists have been used principally for the management of acute coronary syndromes. They include abciximab (ReoPro), a monoclonal antibody, tirofiban (Aggrastat), and eptifibatide (Integrilin). These agents all require intravenous administration. Their effect is reversible. The half-lives are approximately 2.5 hours for tirofiban and eptifibatide (both increased with renal dysfunction) and 12 hours for abciximab.204 However, abciximab has a relatively high affinity for the IIb/IIIa receptor, and platelet dysfunction lasts longer (approximately 48 hours) than implied by half-life. All of these agents have also been associated with thrombocytopenia, the incidence of which has been greater for abciximab (2.5%) than tirofiban and eptifibatide (0.5%).205 The thrombocytopenia caused by abciximab can be either delayed (antibody mediated) or immediate.174 Note that these agents cause prolongation of the ACT.204

Herbal Medications and Vitamins

Several herbal medications may cause inhibition of platelet function84 (see Chapter 22). Among the more common agents identified by the ASA Practice Advisory are feverfew, flaxseed oil, garlic, ginger, gingko biloba, grape seed extract, and saw palmetto.80 Because the actual risks are not well defined, they should be discontinued before surgery, and in particular, before cardiac, neurologic, and cosmetic surgical procedures. Vitamin E and ginseng are also platelet/coagulation inhibitors and should similarly be discontinued.206,207

Other Conditions

Myeloproliferative and myelodysplastic syndromes can produce intrinsic defects in platelets. In these disorders, the platelets may be abnormal in both morphology and function. Platelet dysfunction occurs in conjunction with conditions that also cause other hemostatic abnormalities (liver disease, fibrinolytic states including DIC, storage defects), which are discussed in the following section.

Acquired Disorders of Clotting Factors (Including Anticoagulant Therapy)

Vitamin K Deficiency

Hepatic synthesis of clotting factors II, VII, IX, and X as well as protein C and protein S requires the presence of vitamin K. Vitamin K is necessary for the enzymatic carboxylation of these factors. The carboxyl group enables binding to phospholipid surfaces during the coagulation process. With vitamin K deficiency, these factors are depleted in an order determined by their half-lives. Factor VII has the shortest half-life and is the first to be depleted, followed by FIX, FX, and finally FII (prothrombin). Vitamin K deficiency occurs frequently in hospitalized patients because of dietary insufficiency, gut sterilization, and malabsorption. A high index of suspicion should be maintained.

Vitamin K occurs naturally in two forms.206 Vitamin K1 (phylloquinone) is found in leafy green vegetables. The greatest concentrations occur in brussels sprouts. Vitamin K2(menaquinone) is synthesized by the normal intestinal flora. It is uncommon for patients to develop vitamin K deficiency solely because of dietary deficiency, but it may occur in patients who are receiving parenteral nutrition without vitamin K supplementation, and who are being treated concurrently with broad-spectrum antibiotics that destroy the gut flora. Because the body has no appreciable stores of vitamin K, deficiencies can develop in as little as 7 days. Newborns, who have a sterile gut at birth, have been noted to develop vitamin K deficiency. Vitamin K is fat-soluble and therefore requires bile salts for absorption from the jejunum. Biliary obstruction, malabsorption syndromes, gastrointestinal obstruction, or rapid gastrointestinal transit can result in vitamin K deficiency because of inadequate absorption.

Diagnosis and Treatment of Vitamin K Deficiency

Vitamin K deficiency will cause prolongation of the PT. PT is an FVII-sensitive assay and with vitamin K deficiency, FVII is the first factor to be depleted. With more prolonged deficiency, aPTT (a very FIX-sensitive assay) will also increase. Platelet count will be normal. Vitamin K may be administered orally, intramuscularly, or intravenously. Urgent treatment of vitamin K deficiency is best accomplished by the intramuscular or intravenous administration of vitamin K (Aquamephyton),


usually in doses of 1 to 5 mg. Vitamin K should be administered slowly to avoid the occurrence of hypotension. Improvement of the coagulation disturbance will begin to be apparent in 6 to 8 hours.

Warfarin Therapy

Warfarin is administered for the prevention of DVT and PE and to patients with atrial fibrillation, some prosthetic heart valves, and ventricular mural thrombi in the setting of acute MI. Patients with protein S or protein C deficiency may also be treated with long-term anticoagulation with warfarin. Warfarin produces its anticoagulant effect by competition with vitamin K for the carboxylation binding sites and leads to the depletion of factors II, VII, IX, X, protein C, and protein S. As with vitamin K deficiency (previous paragraph), FVII is the first factor to be depleted and initially only the PT will be prolonged. With higher doses, FIX levels will decrease and the aPTT will increase. Warfarin therapy is adjusted according to the INR (see “Tests of the Hemostatic Mechanism”). The primary untoward effect of warfarin therapy is bleeding. Rapid reversal (12 to 24 hours) of warfarin effect208 can be accomplished by intravenous administration of vitamin K. Doses of 5 to 10 mg intravenously are recommended for urgent situations.209 Smaller doses, 0.5 to 3 mg, and the oral route should be used in less urgent situations when the objective is to reduce rather than normalize INR. INR should be rechecked at 6-hour intervals. Vitamin K administration may have to be repeated at 12-hour intervals. In situations of greater urgency, FFP, TP (which is immediately available in facilities that provide it), or prothrombin complex concentrate (PCC) will all provide the relevant factors and can be employed. In patients who might not tolerate the requisite volume of FFP or TP (>15 mL/kg), PCC, which contains FII, FVII, FIX, and FX, is an alternative. PCC dosing recommendations vary. However, 15 IU/kg when INR is <5 and 30 IU/kg for INR >5 appear reasonable.210 Recombinant FVIIa (rFVIIa; see later discussion) has also been used to achieve rapid normalization of INR.211 Note, however, that the action of rFVIIa requires the participation of FX and FII (prothrombin), both of which are depleted at greater degrees of warfarin effect (as witnessed by aPTT prolongation). In this circumstance, rFVIIa may not provide effective reversal of anticoagulation. PCC contains FII, VII, FIX, and FX and is more likely to be effective.212 Thrombotic events have occurred with the administration of both PCC and rFVIIa.212,213 If FFP, TP, PCC, or rFVIIa are administered for rapid reversal and sustained reversal is desired, vitamin K should be administered simultaneously209 because of the short half-life of FVII (6 hours for native FVII, 2 hours for rFVIIa).

Heparin Therapy

Unfractionated heparin (UFH) is used widely for anticoagulation in vascular surgery and in procedures requiring CPB. It inhibits coagulation principally through its interaction with AT (see Chapter 41). UFH binds to AT, and in so doing causes a conformational change that greatly increases AT's inhibitory activity. In spite of its name, anti-“thrombin,” AT also inhibits several activated factors including, in addition to IIa (thrombin), Xa, IXa, XIa, and XIIa (Fig. 16-6). It is most active against thrombin and Xa. UFH also increases the activity of a second native antithrombin, heparin cofactor II. Heparin cofactor II inhibits thrombin and not the other activated factors. Its contribution to the clinical effects of UFH is not clear. Resistance to UFH can occur in patients who are deficient in AT on either a hereditary or an acquired basis. The latter may occur in patients on sustained UFH therapy, in the presence of depletion by a consumptive coagulopathy or during CPB. UFH responsiveness can be restored by administration of AT concentrates168,169 or FFP.

Low-Molecular-Weight Heparin

Low-molecular-weight fractions of heparin (LMWH) have been employed principally for DVT prophylaxis and treatment, and are supplanting subcutaneous UFH and warfarin for these indications.214 There are several available agents including certoparin, dalteparin, enoxaparin, reviparin, and tinzaparin. These agents do not appear to differ in their efficacy,215 and enoxaparin is used most widely in the United States. LMWHs, which also act via AT, have greater activity against FXa than thrombin (FIIa). However, the ratio of that activity varies among the agents (e.g., enoxaparin, 3.8:1; tinzaparin, 1.9:1).216 Accordingly, the effect of these agents on standard coagulation tests will vary (minimal for enoxaparin217) as will the effect of protamine neutralization, which is very incomplete for enoxaparin. Monitoring is usually not required or performed. If it is deemed necessary (e.g., renal failure, extreme obesity), the anti-Xa activity level (see previous discussion) is the appropriate test. The LMWHs cause less platelet inhibition and are associated with a lesser incidence of HIT/T than UFH.218 While twice-daily dosing with enoxaparin has been common in North America, once-daily regimens are usually sufficient. Because of the relatively long half-life of enoxaparin, twice-daily dosing poses a problem with respect to removal of epidural catheters because there is no anticoagulant nadir (see Chapter 53.)

Heparin Induced Thrombocytopenia/Thrombosis (HIT/T)

One to five percent of patients who receive UFH therapy for 5 days will develop thrombocytopenia as a result of antibodies (usually IgG) directed against platelet factor 4 (PF4)-heparin complexes on the platelet surface.219,220 Onset requires several days in the heparin-naive patient but can occur much more quickly (10 to 12 hours) in those who have been exposed within the preceding 100 days. Occurrence appears to be dose-related and is more common with bovine than porcine heparin. HIT/T is relatively uncommon with LMWH and requires longer periods of exposure.218,221 When LMWH does induce antibodies, they are more commonly of the IgM or IgA type and often do not cause thrombocytopenia. However, patients who have developed IgG antibodies and HIT/T in response to UFH will frequently develop HIT/T on exposure to LMWH.218 Although HIT/T is most often identified because of thrombocytopenia, not all patients become markedly thrombocytopenic. Thrombotic and thromboembolic events including DVT, PE, limb or acral ischemia, MI, or stroke frequently reveal the occurrence of HIT/T. Diagnosis is complicated by the fact that not all patients who develop antiplatelet antibodies have clinical HIT/T. A hematologist should be consulted.

Treatment entails withdrawal of heparin and administration of a nonheparin anticoagulant. The DTIs (lepirudin, argatroban, and bivalirudin) and the indirect Xa inhibitor danaparoid are approved for this use in various countries (although danaparoid is not available in the United States). Several other anticoagulants are under development, including orally administered DTIs and direct inhibitors of FXa and FXIa.222 A LMWH is not appropriate. Fondaparinux (an indirect Xa inhibitor; see later discussion) has a negligible (perhaps zero) incidence of antiheparin/PF4 cross-reactivity but is not formally approved. Warfarin is contraindicated because the combination of protein C and S inhibition by warfarin in the face of ongoing platelet clumping may aggravate thrombosis. Platelets similarly should not be administered unless thrombocytopenia is extreme.

Cardiac Surgery and HIT/T

Several alternatives have been employed for the patient with HIT/T who requires CPB180 (see Chapter 41).


The most common, when antiheparin/PF4 antibodies are still present, is the use of nonheparin anticoagulants, usually a DTI (see later discussion). An alternative is to provide profound inhibition of platelet activation with either iloprost (synthetic prostacyclin) or a IIb/IIIa inhibitor (tirofiban at UCSD) during CPB and proceed with UFH administration, protamine reversal, and a nonheparin anticoagulant in the postoperative period.223 When antiheparin/PF4 antibodies have decreased to undetectable levels in a patient with a history of HIT/T, heparin may be employed during CPB, although it must be rigidly avoided during the remainder of the hospitalization. Antibody generation requires 5 days by which time heparin will be absent.219

As many as 5% of patients who receive UFH therapy for 5 days will develop heparin-induced thrombocytopenia/ thrombosis. The clinical manifestations are more often the result of thrombosis and thromboembolism than thrombocytopenia.

Heparin in Cardiopulmonary Bypass

A comprehensive discussion of this topic is beyond the scope of this chapter. Extensive reviews are available224 (see Chapter 41). In brief, the common practice is to maintain ACT >480 to 500 seconds for the duration of bypass. There is substantial variation in the UFH-ACT dose-response relationship, probably because of variability in UFH binding to many native surfaces including platelets, WBCs, endothelium, and plasma proteins including the vWF and AT.224 There is hazard, in terms of activation of both platelets and coagulation, in allowing ACT to be on the “low side.” Platelet and coagulation activation can be demonstrated at ACT levels of 400 seconds.225 Evidence of activation is less apparent when longer ACTs are maintained.226 Protamine is administered for reversal of UFH effect. Many clinicians employ a “milliliter for milliliter” technique. However, a more careful titration of protamine against ACT is ideal to avoid excessive administration of protamine, which has inherent anticoagulant effects including platelet inhibition, stimulation of tPA release from endothelium, and inhibition of fibrinogen cleavage by thrombin.227

Direct Thrombin Inhibitors

DTIs produce their anticoagulant effect by directly binding to thrombin.180 Hirudin is a naturally occurring compound; lepirudin is its recombinant equivalent. Argatroban and bivalirudin are synthetic. By contrast with UFH, LMWH, and fondaparinux, all of which act via AT to inhibit only unbound thrombin, DTIs inhibit both unbound and fibrin-bound thrombin. DTIs therefore inhibit all of thrombin's numerous effects on hemostasis (Figs. 16-2 and 16-5). Clot-bound thrombin can continue to promote coagulation by activation of platelets, by activation of FV, FVIII, FXI, and FXIII, and by conversion of fibrinogen to fibrin.228 There is no antidote to the anticoagulant effect of DTIs. Termination of the effect of hirudin and lepirudin (half-life, 80 minutes) depends on renal elimination. Argatroban (half-life, 40 to 50 minutes) is metabolized by the liver. Bivalirudin (half-life, 25 minutes) is largely cleared by proteolysis by plasma proteases with some contribution by renal clearance.

DTIs do not bind to PF4 and are widely used to anticoagulate patients with HIT/T, in particular those who require CPB. For the latter, bivalirudin is the most widely used agent because of its relative independence of hepatic and renal clearance and a relatively short half-life. Monitoring of anticoagulation is problematic. As noted in the section “Laboratory Evaluation of Coagulation,” with the greater degrees of anticoagulation required for CPB, the correlation between ACT and DTI serum level is poor, and unnecessary overdose can occur easily. The ecarin clotting time (see previous discussion) is the preferred test. However, it is not widely available. Fixed dosage regimens have therefore been employed (e.g., loading dose, 1.0 to 1.5 mg/kg; infusion, 2.5 mg/kg/hr).229,230 ACT should be >400. Note that the relatively short half-life and enzymatic degradation mean that blood that is static (CPB or cell salvage reservoirs) may clot. The technique has to be adjusted accordingly. In the patient with renal failure or in urgent situations, elimination can be accomplished by dialysis or hemofiltration.231

Ximelagatran is a direct thrombin inhibitor. It has been withdrawn from the market on the basis of hepatotoxicity.

Indirect Inhibitors of Xa

Fondaparinux and idraparinux are synthetic agents that act via AT to produce a highly specific inhibition of FXa.232 Fondaparinux is an increasingly popular alternative for DVT prophylaxis, in part because of its very predictable uptake (after once daily subcutaneous administration) and kinetics that make monitoring and dosage adjustment unnecessary. Fondaparinux can form a complex with PF4. However, heparin/PF4 antibodies do not react with that complex in a manner that produces platelet activation and HIT/T.233Nonetheless, fondaparinux is not yet approved for use in HIT/T. These agents have long half-lives (fondaparinux, 17 hours; idraparinux, 80 hours), and there is no antidote. Excretion is via the kidneys. With therapeutic doses PT, aPTT, and ACT remain within the normal range.

Danaparoid, which is not available in the United States, is a mixture of three glycosaminoglycans (heparan sulphate, dermatan sulfate, and chondroitin sulfate) derived from porcine intestine. Anti-Xa to anti-IIa activity occurs in a ratio of 22:1. Half-life is 25 hours. Elimination is renal. PT and aPTT are unaffected by therapeutic doses. Cross-reactivity with heparin/PF4 antibodies occurs infrequently, and danaparoid is approved for use in HIT/T.180

Acquired Combined Disorders of Platelets and Clotting Factors with Increased Fibrinolysis

Liver Disease

Chronic liver disease is associated with abnormalities of all three phases of hemostasis: primary hemostasis, coagulation, and fibrinolysis. Table 16-16 provides an overview of these abnormalities.

Table 16-16 The Etiology of Hemostatic Abnormalities in Liver Disease

·         Thrombocytopenia

o    Decreased production

o    Hypersplenism

o    Increased consumption (DIC)

·         Impaired platelet function

o    Decreased FDP clearance

·         Decreased factor synthesis

o    Decreased hepatocyte function

o    Vitamin K deficiency (diet, malabsorption)

·         Increased factor consumption

o    Decreased clearance of activated factors

o    Decreased synthesis of inhibitors (protein C, protein S)

·         Increased fibrinolysis

o    Decreased synthesis of α2-antiplasmin

o    Decreased clearance of tPA

o    Decreased synthesis of PAI-1

DIC, disseminated intravascular coagulation; FDP, fibrin degradation product; tPA, tissue plasminogen activator; PAI-1, plasminogen activator inhibitor-1.


Impaired Primary Hemostasis

Impaired primary hemostasis occurs as a result of both thrombocytopenia and impaired platelet function. The former is largely the result of decreased production, which in turn is probably the result of decreased thrombopoietin secretion by the liver. Hypersplenism may also contribute, but its role has been overemphasized. Platelet dysfunction can occur when liver disease is sufficiently advanced that clearance of FDPs is impaired, or when DIC complicates the coagulation disturbance. The FDPs coat the surface of platelets and impair aggregation.84 Ethanol can also directly contribute to platelet dysfunction by inhibition of the synthesis of ADP, ATP, and TxA2.84 Accordingly, when faced with a patient with liver disease who is bleeding, a normal platelet count cannot be assurance of intact primary hemostasis. DDAVP may be helpful, but transfusion of platelet concentrates may be necessary.

Disturbances of Coagulation

With liver disease, factor production decreases and consumption increases. The liver synthesizes all of the clotting factors (with the probable exception of factor VIII). As with vitamin K deficiency, hepatic disease first leads to a deficiency of factor VII as it has the shortest half-life. Thereafter, deficiencies will develop in factors IX, X, and II. Dietary deficiency of vitamin K, as may occur in alcoholics, combined with diminished secretion of bile salts leading to malabsorption, will exaggerate these deficiencies. If impaired coagulation is the result of vitamin K deficiency and not hepatic damage, then parenteral vitamin K may be helpful in restoring factor levels of II, VII, IX, and X. Further deterioration of hepatic function will affect the remaining factors, I, V, XI, XII, and XIII.

Impaired liver function can also cause a thrombotic tendency, which leads to increased consumption of clotting factors. This occurs for two reasons. First, synthesis of the natural anticoagulants, AT, protein C, and protein S, may be diminished, thereby altering the balance of pro- and anticoagulant forces. Second, clearance of activated clotting factors from the circulation may be impaired, thereby allowing persistent activation of the coagulation cascade.

Increased Fibrinolysis

Increased fibrinolysis occurs as a result of decreased clearance of tPA from the circulation by the impaired liver and decreased hepatic synthesis of α2-antiplasmin. Production of the natural inhibitor of the plasmin system, PAI-1, is also diminished.234 The combination of accelerated coagulation and increased fibrinolysis in patients with advanced liver disease can lead to a persistent, low-grade DIC. The release into the circulation of the breakdown products of necrotic hepatocytes may contribute to the development of DIC.235

Diagnosis and Treatment of Coagulation Abnormalities Associated with Liver Disease

The initial laboratory evaluation should include platelet count, PT, aPTT, fibrinogen level, and D-dimer. In the event of thrombocytopenia and clinical bleeding or pending surgery, platelet transfusion may be appropriate. If the PT is prolonged (>1.5 times control), vitamin K should be administered speculatively. In the absence of a response to vitamin K (which requires a minimum of 8 hours), factor deficiencies should be treated with FFP, with attention to the possibility of volume overload. Cryoprecipitate is appropriate in the event of hypofibrinogenemia (fibrinogen <100 to 125 g/dL). While antifibrinolytics have been applied in the context of liver transplantation, they should not otherwise be used for bleeding associated with liver disease because of the catastrophic consequences of administering these agents in the face of an unrecognized DIC. However, making the diagnosis of DIC (see later discussion) is often difficult because the laboratory tests used to identify DIC are already abnormal in patients with liver dysfunction. Thrombocytopenia, prolonged PT and aPTT, decreased fibrinogen level, and circulating FDPs will commonly occur in the absence of DIC. Elevated D-dimer is somewhat more specific for the occurrence of DIC.


Detailed reviews of DIC are available.236,237,238 DIC is characterized by excessive deposition of fibrin throughout the vascular tree, with simultaneous depression of the normal coagulation inhibitory mechanisms and impaired fibrin degradation (see Chapter 56). It is triggered by the appearance of procoagulant material (TF or equivalent) in the circulation in amounts sufficient to overwhelm the mechanisms that normally restrain and localize clot formation. That appearance may be the result of either extensive endothelial injury, which exposes TF, or the release of TF into the circulation as occurs with amniotic fluid embolus, extensive soft-tissue damage, severe head injury, or any cause of a systemic inflammatory response. The native pathways that inhibit coagulation are either inhibited or overwhelmed: AT levels are depleted by excess thrombin formation, as reflected by elevated levels of thrombin-AT complexes; thrombomodulin expression in vascular endothelium is reduced in response to inflammation thereby reducing protein C formation; and the capacity of TFPI to restrain the TF-driven extrinsic pathway may be exceeded because of excessive TF.238 The accelerated process of clot formation causes both tissue ischemia and, ultimately, critical depletion of platelets and factors. Simultaneously, the fibrinolytic system is activated, and plasmin is generated to lyse the extensive fibrin clots. FDPs appear in the circulation. FDPs stimulate release of PAI-1 from the endothelium, and thrombolysis becomes impaired. The FDPs also inhibit platelet aggregation and prevent the normal cross-linking of fibrin monomers. Depleted of platelets and clotting factors and inhibited by FDPs, the coagulation system fails and the patient bleeds. Simultaneously, the microvascular occlusion by fibrin causes both cutaneous (purpura fulminans) and deep tissue ischemia, with the latter contributing to multiorgan failure.

Table 16-17 lists the numerous clinical conditions that have been associated with DIC. It reveals that several clinical entities that are encountered frequently in anesthetic and critical


care practice are associated with the development of DIC. Sepsis is the most common. Endotoxins or lipopolysaccharide breakdown products from Gram-negative and positive bacteria, respectively, incite an inflammatory response that includes the generation of cytokines (tumor necrosis factor-α, various interleukins). These cytokines in turn stimulate the release or expression of TF by endothelial cells and monocytes, and the DIC sequence is initiated.

Table 16-17 Clinical Conditions Associated with Disseminated Intravascular Coagulopathy

·         Sepsis (Gram-positive or -negative)

·         Viremias

·         Obstetric conditions

o    Amniotic fluid embolus

o    Fetal death in utero

o    Abruptio placentae

o    Pre-eclampsia

·         Extensive tissue damage

o    Burns

o    Trauma

·         Liver failure

·         Extensive cerebral injury

o    Head injury

o    Stroke

·         Extensive vascular endothelial damage

o    Vasculitis

o    Pre-eclampsia

·         Hemolytic transfusion reactions

·         Metastatic malignancies

·         Leukemia

·         Snake venoms

Several obstetric conditions can cause DIC. Amniotic fluid embolism, placental abruption, and fetal death in utero result in the direct release of TF-equivalent material into the circulation. Pre-eclampsia is characterized by a systemic vasculitis. The associated endothelial damage causes an initially low-grade DIC that accelerates as vasculitis-related damage leads to release of TF from ischemic tissues, in particular, placenta.

Large burns, extensive traumatic soft-tissue injuries, severe brain injury, and hemolytic transfusion reactions can also liberate TF-equivalent material into the circulation and incite DIC. Certain malignancies, most notably promyelocytic leukemia and adenocarcinomas, are associated with DIC. However, with malignancy-associated DIC, thrombotic manifestations are more likely to appear first, whereas with the others mentioned here, the hemorrhagic diathesis is often the first clinical manifestation.

A few general conditions such as acidosis, shock, and hypoxia are associated with DIC. Shock promotes coagulation because one of the control mechanisms (rapid blood flow) is compromised. Clearance of activated clotting factors is reduced when blood flow is decreased. Acidosis and hypoxia may contribute to both tissue and endothelial damage.

The clinical manifestations of DIC are a consequence of both thrombosis and bleeding. Bleeding is a more common clinical presentation in patients with acute, fulminant DIC. Petechiae, ecchymoses, epistaxis, gingival/mucosal bleeding, hematuria, and bleeding from wounds and puncture sites may be evident. With the chronic forms of DIC, thrombotic manifestations are more likely. Organs with the greatest blood flow (e.g., kidney and brain) typically sustain the greatest damage. Pulmonary function may deteriorate as a consequence of microthrombus accumulation.

Diagnosis of DIC

There is no absolutely consistent constellation of laboratory findings among routine tests. Increased PT, aPTT, thrombocytopenia, decreased fibrinogen level, and the presence of FDPs and D-dimer may all be noted. The peripheral smear may reveal schistocytes (fragmented RBCs reflecting the microangiopathy that occurs as a consequence of widespread fibrin deposition). Thrombocytopenia (<100,000 /µL) is not always evident early in the process, but true DIC without sequential reduction in platelet count is very unlikely. PT and aPTT may remain normal in spite of decreasing factor levels because of the presence of high levels of activated factors including thrombin and Xa. Fibrinogen levels may not be decreased, that is, <100 mg/dL, initially. Fibrinogen is an “acute phase reactant” that increases in response to stress, and the early consumption of fibrinogen may simply reduce its levels to “normal”. FDPs are a sensitive measure of fibrinolytic activity although they are not specific for DIC. D-dimer (a breakdown product of the cross-linked fibrin in a mature clot) is somewhat more specific for DIC, but not entirely so, and should be measured when that diagnosis is suspected. The 3-P (plasma-protamine-paracoagulation) test is a relatively specific, although not very sensitive, assay sometimes used to confirm a diagnosis of DIC. It tests for the presence of soluble complexes composed of fibrin monomers (generated by excess thrombin) and FDPs. The addition of protamine desolublizes these complexes resulting in a precipitate.

Various other laboratory assays have been employed to support a diagnosis of DIC236 but should probably not be considered part of the anesthesiologist's routine. They include levels of prothrombin fragments F1 + F2 (a marker of prothrombin conversion to thrombin—increased), thrombin-AT complexes (increased), AT (decreased), α2-antiplasmin (decreased by binding to excess plasmin), protein C (decreased), plasminogen (decreased), and factor VIII (decreased in DIC but normal with hepatic failure without DIC).

Treatment of DIC

Treatment should focus on management of the underlying condition. Septicemia will require antibiotic therapy. The obstetric conditions are frequently self-limited, although evacuation of the uterus or hysterectomy may be warranted. Hypovolemia, acidosis, and hypoxemia should be corrected to prevent their contribution to the DIC process. When bleeding is or may become life-threatening, the consumptive coagulopathy must be treated. Platelets will be required for thrombocytopenia (e.g., <50,000/mm3). FFP will replace the clotting factor deficiencies. Fibrinogen level should be raised to >100 mg/dL. When hypofibrinogenemia is severe (<50 mg/dL), cryoprecipitate may be required. Six units of cryoprecipitate will increase fibrinogen level by approximately 50 mg/dL in a 70-kg patient.239

Heparin has been advocated. However, the contemporary practice is to restrict its use to only those situations where thrombosis is clinically problematic, principally DIC associated with malignancies. There is no proven benefit in situations in which bleeding is the predominant manifestation. Administration of antifibrinolytics in the face of widespread thrombosis is potentially disastrous, and they should not be used. AT concentrates have been administered. The hope is that its administration will serve to slow the runaway coagulation process. However, a beneficial effect on outcome from DIC has not been confirmed (see data review by Levi237), and its use should be viewed as experimental. An insufficiency in the protein C endogenous coagulation inhibition system is thought to contribute to the prothrombotic state in DIC (see previous discussion). APC has been shown to decrease mortality and organ failure in patients with severe sepsis, and that improvement is also evident among patients with sepsis with overt DIC.240 Its use should be considered in any sustained episode of DIC.240

Cardiopulmonary Bypass and Coagulation

Limited mention of this topic has been made in the earlier sections “Heparin in Cardiopulmonary Bypass, Cardiac Surgery, and HIT/T” and “Direct Thrombin Inhibitors.” The management of anticoagulation and post-CPB bleeding is addressed in detail in Chapter 41.

Pharmacologic Therapy

Recombinant Factor VIIa

Recombinant FVIIa (rFVIIa) (NovoSeven) was developed for the treatment of patients with hemophilia A or B and inhibitors to exogenous FVIII or FIX preparations. The only current “on-label” indications for rFVIIa in the United States are those two conditions plus congenital FVII deficiency. Glanzmann's thrombasthenia is an approved indication in some other countries. However, rFVIIa has become a hemostatic agent of last (and sometimes earlier) resort in many clinical situations. Its use has been reported in trauma, hepatic failure, gastrointestinal bleeding, obstetric hemorrhage, acute intracerebral hemorrhage, and in cardiac, prostatic, hepatic, spinal, neurologic, and hepatic transplantation surgery. It has been used to reverse the anticoagulant effect of warfarin,


LMWHs, and selective Xa inhibitors. It has been administered to patients with vWD, FXI deficiency, thrombocytopenia, and with both congenital (Bernard-Soulier syndrome, Glanzmann's thrombasthenia) and acquired (uremia, aspirin, ADP and IIb/IIIa antagonists) platelet abnormalities.211,241,242 However, most of these uses are supported by only anecdotal reports, among which there may be significant publication bias; that is, apparent success is reported more often than obvious failure. Of the off-label applications, only the use in prostate surgery, trauma, cardiac surgery (a very small series), and intracerebral hemorrhage are supported by randomized, blinded prospective trials.243,244,245,246

The mechanism of action is more than an augmentation of the native functions of FVII. Were that the case, rFVIIa would not be effective in hemophilia (Fig. 16-2). It seems probable that rFVIIa directly activates FX on platelet surfaces and thereby effects, without the participation of factors VIII, IX, and XI, the generation of the large amounts of thrombin necessary to produce a firm fibrin clot.247 While the preferred ligand of FVIIa is TF, it also undergoes low-affinity binding to activated platelets. The serum concentrations achieved by typical rFVIIa dosing are several hundred times those that occur physiologically and are probably sufficient to activate FX on the platelet surface.

In a survey of experience with trauma patients, it was reported that acidosis (pH <7.20) appeared to decrease the efficacy of rFVIIa but that moderate hypothermia did not.248 Note that while several reports speak to the efficacy of rFVIIa in reversing the effects of warfarin, LMWH, and fondaparinux, production of the fibrin burst in response to rFVIIa requires the availability of some FX. Warfarin, LMWH, and fondaparinux, all either inhibit the synthesis or the activity of FX. It seems reasonable to expect that in severe overdoses, rFVIIa may not be effective.

The appropriate dosing of this expensive agent (approximately US $1 per microgram at UCSD) is not well defined. The dose used most often in hemophilia has been 90 µg/kg and that dose has widely, and arbitrarily, we think, been adopted in other clinical situations. Doses as low as 20 µg/kg have been effective in some reports including the prostate surgery investigation just mentioned243 and the reversal of warfarin effect.249 In the prospective trauma investigation by Boffard et al.,244 the initial dose was 200 µg/kg, although others have reported apparent efficacy in trauma patients with doses of 75 µg/kg.250 It seems reasonable that the appropriate dosage may vary with the clinician's perception of the severity of the physiologic disturbance and the urgency of the situation. A one unit per hour gastrointestinal bleed and an exsanguinating trauma patient with escalating acidosis and hypothermia may warrant different doses. The current, somewhat arbitrary, algorithm in place at UCSD provides for the administration of 60 µg/kg for profuse bleeding that is unresponsive to conventional therapy. That dose is rounded to the nearest 1,200 µg in recognition that the agent is supplied in vials of 1.2 mg. A U.S. consensus panel recommended 20 to 40 µg/kg for “non-emergent reversal” and 41 to 90 µg/kg “for all other scenarios.”251 The half-life is approximately 2 hours and repeat dosing at that interval may be required.

Because rVIIa is an active procoagulant only when it is in contact with TF or activated platelets, thrombosis in locations remote from sites of vessel disruption has been infrequent. However, thrombotic complications, some fatal, have been reported,213 and the use of rFVIIa should be undertaken with an awareness of that hazard. When the exigencies of the clinical situation permit, modest initial doses of 20 to 40 µg/kg with supplementary doses at 15-minute intervals as warranted by clinical response seem prudent. rFVIIa should probably be viewed as relatively contraindicated in clinical states in which TF may be widely exposed or circulating freely, that is, in most of the conditions associated with DIC.

The effectiveness of available laboratory tests in monitoring the clinical effect of rFVIIa is uncertain.241 It has not been confirmed that the effect of high serum concentrations of rFVIIa on in vitro tests (PT, PTT) will reflect effects on coagulation in vivo. Furthermore, the use of “normal values” as a comparator for post-rFVIIa aPTT values may have little meaning if, in vitro, rFVIIa directly activates FX on the phospholipid reagent, and thereby bypasses contact activation and all the earlier steps of the intrinsic pathway (Fig. 16-1).


Desmopressin, 1-deamino-8-D-arginine vasopressin (DDAVP), is a synthetic analogue of the natural hormone vasopressin.192,197 The actions of vasopressin are mediated by two general classes of receptors: V1, which mediate smooth muscle contraction in the peripheral vasculature, and V2, which regulate water reabsorption in the collecting ducts of the nephron. DDAVP is active only at V2 receptors. Accordingly, it is a potent antidiuretic with no vasoconstrictor effect. DDAVP was used primarily for clinical conditions such as diabetes insipidus until its hemostatic effects were recognized. DDAVP causes rapid release of vWF and tPA from vascular endothelium via stimulation of endothelial V2 receptors. DDAVP also causes increases in serum levels of FVIII, perhaps by release from hepatic sinusoidal endothelial cells,197 and increased expression of platelet surface GP1b receptors.252 In vWD, the DDAVP-induced increases in FVIII level are mediated in part by increased serum life of FVIII because of the availability of its protective carrier protein vWF. In mild hemophilia A, DDAVP can increase the circulating factor VIII:C concentration two- to sixfold. DDAVP increases platelet adhesiveness and shortens the BT.


DDAVP is effective treatment for type I and some type II variants of vWD and for mild hemophilia A (see “von Willebrand Disease” and “Hemophilia A”). DDAVP has been shown to reduce BT in several conditions associated with platelet dysfunction, including uremia (see “Uremia”) and advanced liver disease.84 DDAVP also decreases the prolonged BTs caused by many drugs including aspirin, nonsteroidal anti-inflammatory drugs, dextran, ticlopidine, and heparin. It is effective for some congenital platelet abnormalities, including the Bernard-Soulier syndrome (but not Glanzmann's thrombasthenia).192

Because platelet dysfunction and thrombocytopenia are common in cardiac surgery, studies of prophylactic administration of DDAVP have been performed. Those that have revealed decreased blood loss or blood product administration have involved principally patients who were predisposed to blood loss (e.g., those having repeat procedures253,254,255,256) and patients receiving aspirin.257 It has not proven effective at reducing blood loss in unselected surgical populations.258


DDAVP, when given for its procoagulant effect, is usually administered intravenously in a dose of 0.3 µg/kg. (Note that this dose is not appropriate for the management of acute central diabetes insipidus, for which the total initial intravenous dose should be 0.2 to 0.4 µg.) Administration over 30 minutes is recommended because DDAVP induces endothelial release of nitric oxide, and mild degrees of hypotension may occur. Peak levels of factor VIII:C and vWF are achieved within 30 to 60 minutes, and the effect lasts for several hours. DDAVP administration may be repeated after 8 to 12 hours. When used in cardiac surgery, the drug should be administered after termination of CPB. Water balance should be monitored. However, while congestive cardiac failure and


hyponatremia and seizures in children have been reported, clinically significant water retention is relatively uncommon.


Antifibrinolytic agents have been used frequently in situations in which exaggerated fibrinolysis is suspected of contributing to intraoperative bleeding. The situations in which favorable effects on blood loss and replacement have been reported include CPB procedures, hepatic transplantation, scoliosis surgery, total joint replacement, and prostate surgery259,260,261,262 (see Chapter 41). The use of antifibrinolytic mouthwashes in the context of dental procedures in patients with hemophilia has been mentioned elsewhere in this chapter. Three antifibrinolytics have been widely employed: the lysine analogues, EACA, and TXA, and the serine protease inhibitor aprotinin (AP). AP was withdrawn from the market in November 2007 because of reports of renal dysfunction and increased mortality after CPB.105,263,264 Some discussion of AP has been included in case the withdrawal is only temporary, as anticipated by the manufacturer.

ε-Aminocaproic Acid and Tranexamic Acid

EACA and TXA bind to produce a structural change in both plasminogen and plasmin. That structural change prevents the conversion of plasminogen to plasmin and also prevents plasmin from degrading fibrinogen and fibrin. The dual action of these agents results in two effects on the hemostatic mechanism. First, decreased conversion of plasminogen to plasmin results in reduced fibrinolysis. The second effect, the inactivation of plasmin, decreases the formation of degradation products of fibrinogen and fibrin. FDPs have anticoagulant effects, including the inhibition of platelet aggregation and the inhibition of the cross-linking of fibrin strands, which are thereby avoided. Their effectiveness in reducing blood loss in the wide variety of surgical situations previously mentioned is well confirmed.261


AP produces its antifibrinolytic effect by a different mechanism. It is an inhibitor of numerous serine protease enzymes including plasmin and kallikrein. The latter participates in the process of contact activation of factor XII. As a consequence of its inhibition of plasmin, AP, like EACA and TXA, prevents degradation of fibrinogen and fibrin. As is the case with EACA and TXA, the reduction in FDPs should improve both platelet and coagulation function. However, AP is believed to have additional beneficial effects on the inflammatory response to CPB in general, and on platelets in particular.265,266 The mechanism of these effects is not known with certainty. However, thrombin is a serine protease that can activate platelets via a “protease-activated receptor” on the platelet surface.267 Better preservation of the GP1b receptor (which is necessary for initial platelet adhesion to vascular defects) has been reported during CPB in patients who received AP.268 AP also appears to reduce neutrophil activation and transmigration across capillary endothelium, perhaps via an effect on an endothelial protease-activated receptor, and may therefore also blunt the neutrophil-mediated component of the response to endothelial injury.266

Use of Antifibrinolytics in Cardiac Surgery

Meta-analyses of the many studies performed in the context of CPB confirm that blood loss and the administration of allogeneic blood are diminished by the use of all three agents.259,261,269,270 Concern has been expressed that antifibrinolysis might lead to an increased rate of graft occlusion, MI, and renal failure. While meta-analysis had not borne out any of those concerns,261,269,270 as previously noted, increased renal dysfunction and mortality have recently been attributed to AP.263,264,271 The mechanism of the renal dysfunction remains a matter of speculation, and concern about selection bias in those investigations (i.e., “sicker” patients were more likely to receive AP) has been expressed.266 As a result, at least three other retrospective reviews of existing databases, with attempts to control for covariates, have been performed. Two did not provide support for an adverse effect. Furnary et al.272 reported no effect of AP on renal function, and Dietrich et al.273 reported the absence of any dose-related effect of AP on renal function. However, Schneeweiss et al.274 compared 33,517 patients who received AP with 44,682 who received EACA, and reported an adverse effect of AP on both mortality and renal function. AP is not currently available.

There is not a clear consensus as to which of the remaining two agents is most appropriate in the context of CPB. Meta-analysis has revealed both to be effective.269 However, there is more evidence in support of TXA, and at least one study reported greater reduction of blood loss with TXA than EACA.275

The patterns of use of antifibrinolytic agents in cardiac surgery vary substantially among institutions. Few appear to use these agents for all CPB procedures. Most reserve them for situations more likely to be associated with post-CPB bleeding (e.g., repeat and circulatory arrest procedures). Still others appear to reserve antifibrinolytics for refractory bleeding post-CPB. The latter seems less logical because much of the activation of the hemostatic mechanism occurs during CPB.

Use of Antifibrinolytics in Liver Transplantation

Accelerated fibrinolysis occurs commonly during hepatic transplantation. This is probably, in part, the consequence of decreased clearance of activated clotting factors by the diseased liver. More importantly, hepatic clearance ceases entirely during the anhepatic phase. In addition, with reperfusion of the donor liver, there is a release of tPA into the systemic circulation. All three agents have all been used, and meta-analysis confirms reduced blood loss with AP and TXA, with too little information to draw conclusions about EACA.262 Some advocate prophylactic administration to all patients, while others administer these agents only in response to the demonstration, typically by thromboelastography, of hyperfibrinolysis.

Use of Antifibrinolytics in Orthopaedic and Other Surgery

There have been numerous investigations of the effect of TXA and EACA on blood loss and transfusion requirement in scoliosis and joint replacement surgery. A meta-analysis confirmed the efficacy of TXA, but not EACA, in those circumstances.260


The approach to the bleeding patient requires a knowledge of the basic hemostatic mechanism and of common bleeding disorders, an ability to interpret coagulation tests, and an appreciation of the risks inherent to blood component therapy. The hemostatic balance is delicate and complex, and it is the responsibility of the anesthesiologist to anticipate, prevent, and treat disturbances of that balance. Preoperative evaluation must identify those patients whose inherited or acquired medical conditions or whose current medications may influence these processes. With respect to medications, there are a rapidly increasing number of agents that are administered specifically for the purpose of altering the hemostatic balance, such as clopidogrel, tPA, and LMWH. As the patient proceeds through the perioperative period, the anesthesiologist must determine whether bleeding is surgical in nature or is the result of a pre-existing or evolving hemostatic defect that will require the transfusion of hemostatic blood components—platelets, FFP, or cryoprecipitate—or the administration of pharmacologic agents.



The authors are grateful to Dzung Le, MD, Professor of Pathology, UCSD School of Medicine, for time spent in discussion of coagulation mechanisms and testing.


1. Whitaker BI, Sullivan M: The 2005 nationwide blood collection and util-ization survey report. US Dept of Health and Human Services.

2. Kleinman S, Chan P, Robillard P: Risks associated with transfusion of cellular blood components in Canada. Transfus Med Rev 2003; 17: 120.

3. Dodd RY: Current risk for transfusion transmitted infections. Curr Opin Hematol 2007; 14: 671.

4. Cable RG, Leiby DA: Risk and prevention of transfusion-transmitted babesiosis and other tick-borne diseases. Curr Opin Hematol 2003; 10: 405.

5. Busch MR: Evolving approaches to estimate risks of transfusion-transmitted viral infections: incidence-window period model after ten years. Dev Biol (Basel) 2007; 127: 87.

6. Busch MP, Glynn SA, Stramer SL, et al: A new strategy for estimating risks of transfusion-transmitted viral infections based on rates of detection of recently infected donors. Transfusion 2005; 45: 254.

7. Kleinman SH, Busch MP: Assessing the impact of HBV NAT on window period reduction and residual risk. J Clin Virol 2006; 36(Suppl 1): S23

8. Kuhns MC, Busch MP: New strategies for blood donor screening for hepatitis B virus: nucleic acid testing versus immunoassay methods. Mol Diagn Ther 2006; 10: 77

9. Satake M, Taira R, Yugi H, et al: Infectivity of blood components with low hepatitis B virus DNA levels identified in a lookback program. Transfusion 2007; 47: 1197

10. Conry-Cantilena C, VanRaden M, Gibble J, et al: Routes of infection, viremia, and liver disease in blood donors found to have hepatitis C virus infection. N Engl J Med 1996; 334: 1691

11. Tong MJ, el-Farra NS, Reikes AR, et al: Clinical outcomes after transfusion-associated hepatitis C. N Engl J Med 1995; 332: 1463

12. Goodnough LT, Brecher ME, Kanter MH, et al: Transfusion medicine. First of two parts—blood transfusion. N Engl J Med 1999; 340: 438

13. Shander A: Emerging risks and outcomes of blood transfusion in surgery. Semin Hematol 2004; 41: 117

14. Nichols WG, Price TH, Gooley T, et al: Transfusion-transmitted cytomegalovirus infection after receipt of leukoreduced blood products. Blood 2003; 101: 4195

15. Stephenson J: Investigation probes risk of contracting West Nile virus via blood transfusions. JAMA 2002; 288: 1573

16. Alter HJ: Emerging, re-emerging and submerging infectious threats to the blood supply. Vox Sang 2004; 87(Suppl 2): 56

17. Montgomery SP, Brown JA, Kuehnert M, et al: Transfusion-associated transmission of West Nile virus, United States 2003 through 2005. Transfusion 2006; 46: 2038

18. Brecher ME, Hay SN: Bacterial contamination of blood components. Clin Microbiol Rev 2005; 18: 195

19. Eder AF, Kennedy JM, Dy BA, et al: Bacterial screening of apheresis platelets and the residual risk of septic transfusion reactions: the American Red Cross experience (2004–2006). Transfusion 2007; 47: 1134

20. Stroncek DF, Rebulla P: Platelet transfusions. Lancet 2007; 370: 427

21. Stramer SL: Current risks of transfusion-transmitted agents: a review. Arch Pathol Lab Med 2007; 131: 702

22. Seitz R, von Auer F, Blumel J, et al: Impact of vCJD on blood supply. Biologicals 2007; 35: 79

23. Eder AF, Chambers LA: Noninfectious complications of blood transfusion. Arch Pathol Lab Med 2007; 131: 708

24. Heddle NM, Soutar RL, O'Hoski PL, et al: A prospective study to determine the frequency and clinical significance of alloimmunization post-transfusion. Br J Haematol 1995; 91: 1000

25. Slichter SJ: Platelet transfusion therapy. Hematol Oncol Clin North Am 2007; 21: 697

26. Eder AF, Herron R, Strupp A, et al: Transfusion-related acute lung injury surveillance (2003–2005) and the potential impact of the selective use of plasma from male donors in the American Red Cross. Transfusion 2007; 47: 599

27. Stainsby D, MacLennan S, Thomas D, et al: Guidelines on the management of massive blood loss. Br J Haematol 2006; 135: 634

28. Kleinman S: A perspective on transfusion-related acute lung injury two years after the Canadian Consensus Conference. Transfusion 2006; 46: 1465

29. Silliman CC, McLaughlin NJ: Transfusion-related acute lung injury. Blood Rev 2006; 20: 139

30. Wendel S, Biagini S, Trigo F, et al: Measures to prevent TRALI. Vox Sang 2007; 92: 258

31. Sheppard CA, Logdberg LE, Zimring JC, et al: Transfusion-related acute lung injury. Hematol Oncol Clin North Am 2007; 21: 163

32. Silliman CC, Boshkov LK, Mehdizadehkashi Z, et al: Transfusion-related acute lung injury: epidemiology and a prospective analysis of etiologic factors. Blood 2003; 101: 454

33. Kleinman S, Caulfield T, Chan P, et al: Toward an understanding of transfusion-related acute lung injury: statement of a consensus panel. Transfusion 2004; 44: 1774

34. Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome. The Acute Respiratory Distress Syndrome Network. N Engl J Med 2000; 342: 1301

35. Insunza A, Romon I, Gonzalez-Ponte ML, et al: Implementation of a strategy to prevent TRALI in regional blood centre. Transfusion Medicine 2004; 157

36. Schroeder ML: Transfusion-associated graft-versus-host disease. Br J Haematol 2002; 117: 275

37. Lane TA: Leukocyte depletion of cellular blood components. Curr Opin Hematol 1994; 1: 443

38. Opelz G, Sengar DP, Mickey MR, et al: Effect of blood transfusions on subsequent kidney transplants. Transplantation Proceedings 1973; 5: 253

39. Vamvakas EC, Blajchman MA: Transfusion-related immunomodulation (TRIM): an update. Blood Rev 2007; 21: 327

40. Taylor RW, O'Brien J, Trottier SJ, et al: Red blood cell transfusions and nosocomial infections in critically ill patients. Crit Care Med 2006; 34: 2302

41. Malone DL, Dunne J, Tracy JK, et al: Blood transfusion, independent of shock severity, is associated with worse outcome in trauma. J Trauma 2003; 54: 898

42. Murphy GJ, Reeves BC, Rogers CA, et al: Increased mortality, postoperative morbidity, and cost after red blood cell transfusion in patients having cardiac surgery. Circulation 2007; 116: 2544

43. Netzer G, Shah CV, Iwashyna TJ, et al: Association of RBC transfusion with mortality in patients with acute lung injury. Chest 2007; 132: 1116

44. Vamvakas EC: Why have meta-analyses of randomized controlled trials of the association between non-white-blood-cell-reduced allogeneic blood transfusion and postoperative infection produced discordant results? Vox Sang 2007; 93: 196

45. Hebert PC, Wells G, Blajchman MA, et al: A multicenter, randomized, controlled clinical trial of transfusion requirements in critical care. Transfusion Requirements in Critical Care Investigators, Canadian Critical Care Trials Group. N Engl J Med 1999; 340: 409

46. Silliman CC, Moore EE, Johnson JL, et al: Transfusion of the injured patient: proceed with caution. Shock 2004; 21: 291

47. Hebert PC, Tinmouth A, Corwin HL: Controversies in RBC transfusion in the critically ill. Chest 2007; 131: 1583

48. Zallen G, Offner PJ, Moore EE, et al: Age of transfused blood is an independent risk factor for postinjury multiple organ failure. Am J Surg 1999; 178: 570

49. Hebert PC, Chin-Yee I, Fergusson D, et al: A pilot trial evaluating the clinical effects of prolonged storage of red cells. Anesth Analg 2005; 100: 1433

50. Basran S, Frumento RJ, Cohen A, et al: The association between duration of storage of transfused red blood cells and morbidity and mortality after reoperative cardiac surgery. Anesth Analg 2006; 103: 15

51. Koch CG, Li L, Sessler DI, et al: Duration of red-cell storage and complications after cardiac surgery. N Engl J Med 2008; 358: 1229

52. Spiess BD: Risks of transfusion: outcome focus. Transfusion 2004; 44: 4S

53. Vamvakas EC: White-blood-cell-containing allogeneic blood transfusion and postoperative infection or mortality: an updated meta-analysis. Vox Sang 2007; 92: 224

54. Collier AC, Kalish LA, Busch MP, et al: Leukocyte-reduced red blood cell transfusions in patients with anemia and human immunodeficiency virus infection: the Viral Activation Transfusion Study: a randomized controlled trial. JAMA 2001; 285: 1592

55. Hebert PC, Fergusson D, Blajchman MA, et al: Clinical outcomes following institution of the Canadian universal leukoreduction program for red blood cell transfusions. JAMA 2003; 289: 1941

56. Fergusson D, Hebert PC, Lee SK, et al: Clinical outcomes following institution of universal leukoreduction of blood transfusions for premature infants. JAMA 2003; 289: 1950

57. Nightingale S: Hypotension and bedside leukocyte reduction filters. JAMA 1999; 281: 1978

58. McLoughlin TM, Greilich PE: Preexisting hemostatic defects and bleeding disorders, Blood: Hemostasis, Transfusion, and Alternatives in the Perioperative Period. Edited by Lake CL MR. New York, Raven Press, 1995, pp 25.

59. Rajagopalan S, Mascha E, Na J, et al: The effects of mild perioperative hypothermia on blood loss and transfusion requirement. Anesthesiology 2008; 108: 71

60. Ferrara A, MacArthur JD, Wright HK, et al: Hypothermia and acidosis worsen coagulopathy in the patient requiring massive transfusion. Am J Surg 1990; 160: 515

61. Wang HE, Callaway CW, Peitzman AB, et al: Admission hypothermia and outcome after major trauma. Crit Care Med 2005; 33: 1296

62. Sessler DI: Mild perioperative hypothermia. N Engl J Med 1997; 336: 1730

63. Hiippala ST, Myllyla GJ, Vahtera EM: Hemostatic factors and replacement of major blood loss with plasma-poor red cell concentrates. Anesth Analg 1995; 81: 360


64. Yuan S, Ferrell C, Chandler WL: Comparing the prothrombin time INR versus the APTT to evaluate the coagulopathy of acute trauma. Thromb Res 2007; 120: 29

65. AuBuchon JP: Minimizing donor exposure in hemotherapy. Arch Pathol Lab Med 1994; 118: 380

66. Jameson LC, Popic PM, Harms BA: Hyperkalemic death during use of a high-capacity fluid warmer for massive transfusion. Anesthesiology 1990; 73: 1050

67. Rock G, Tittley P, Fuller V: Effect of citrate anticoagulants on factor VIII levels in plasma. Transfusion 1988; 28: 248

68. Crosby ET: Perioperative haemotherapy: I. Indications for blood component transfusion. Can J Anaesth 1992; 39: 695

69. Hebert PC, McDonald BJ, Tinmouth A: Overview of transfusion practices in perioperative and critical care. Vox Sang 2004; 87(Suppl 2): 209

70. Wu WC, Rathore SS, Wang Y, et al: Blood transfusion in elderly patients with acute myocardial infarction. N Engl J Med 2001; 345: 1230

71. Hebert PC, Wells G, Tweeddale M, et al: Does transfusion practice affect mortality in critically ill patients? Transfusion Requirements in Critical Care (TRICC) Investigators and the Canadian Critical Care Trials Group. Am J Respir Crit Care Med 1997; 155: 1618

72. Practice Guidelines for blood component therapy: A report by the American Society of Anesthesiologists Task Force on Blood Component Therapy. Anesthesiology 1996; 84: 732

73. Swerdlow PS: Red cell exchange in sickle cell disease. Hematology Am Soc Hematol Educ Program 2006; 48

74. Robertie PG, Gravlee GP: Safe limits of isovolemic hemodilution and recommendations for erythrocyte transfusion. Int Anesthesiol Clin 1990; 28: 197

75. Roth DM, Maruoka Y, Rogers J, et al: Development of coronary collateral circulation in left circumflex Ameroid-occluded swine myocardium. Am J Physiol 1987; 253: H1279

76. Weiskopf RB, Viele MK, Feiner J, et al: Human cardiovascular and metabolic response to acute, severe isovolemic anemia. JAMA 1998; 279: 217

77. Weiskopf RB, Kramer JH, Viele M, et al: Acute severe isovolemic anemia impairs cognitive function and memory in humans. Anesthesiology 2000; 92: 1646

78. Fluit CR, Kunst VA, Drenthe-Schonk AM: Incidence of red cell antibodies after multiple blood transfusion. Transfusion 1990; 30: 532

79. Tuman K: Tissue Oxygen Delivery- the Physiology of Anemia. Anesthesiol Clin North Am 1990; 8: 451

80. Practice guidelines for perioperative blood transfusion and adjuvant therapies: an updated report by the American Society of Anesthesiologists Task Force on Perioperative Blood Transfusion and Adjuvant Therapies. Anesthesiology 2006; 105: 198

81. Guidelines for the use of platelet transfusions. Br J Haematol 2003; 122: 10

82. Samama CM, Djoudi R, Lecompte T, et al: Perioperative platelet transfusion: recommendations of the Agence Francaise de Securite Sanitaire des Produits de Sante (AFSSaPS) 2003. Can J Anaesth 2005; 52: 30

83. Levi MM, Vink R, de Jonge E: Management of bleeding disorders by prohemostatic therapy. Int J Hematol 2002; 76(Suppl 2): 139

84. Shen YM, Frenkel EP: Acquired platelet dysfunction. Hematol Oncol Clin North Am 2007; 21: 647

85. Stanworth SJ, Brunskill SJ, Hyde CJ, et al: Is fresh frozen plasma clinically effective? A systematic review of randomized controlled trials. Br J Haematol 2004; 126: 139

86. O'Shaughnessy DF, Atterbury C, Bolton Maggs P, et al: Guidelines for the use of fresh-frozen plasma, cryoprecipitate and cryosupernatant. Br J Haematol 2004; 126: 11

87. Johansson PI, Hansen MB, Sorensen H: Transfusion practice in massively bleeding patients: Time for a change? Vox Sang 2005; 89: 92

88. Ho AM, Larmakar MK, Dion PW: Are we giving enough coagulation factors during major trauma resuscitation? Am J Surg 2005; 190: 479

89. Borgman MA, Spinella PC, Perkins JG, et al: The ratio of blood products transfused affects mortality in patients receiving massive transfusions at a combat support hospital. J Trauma 2007; 63: 805

90. Hess JR, Holcomb JB, Hoyt DB: Damage control resuscitation: the need for specific blood products to treat the coagulopathy of trauma. Transfusion 2006; 46: 685

91. Forestner JE: Massive transfusion protocol for trauma. American Society of Anesthesiologists Newsletter 2005; 69: 7

92. Lethagen S, Kyrle PA, Castaman G, et al: von Willebrand factor/factor VIII concentrate (Haemate P) dosing based on pharmacokinetics: a prospective multicenter trial in elective surgery. J Thromb Haemost 2007; 5: 1420

93. Carless P, Moxey A, O'Connell D, et al: Autologous transfusion techniques: a systematic review of their efficacy. Transfus Med 2004; 14: 123

94. Bern MM, Bierbaum BE, Katz JN, et al: Autologous blood donation and subsequent blood use in patients undergoing total knee arthroplasty. Transfus Med 2006; 16: 313

95. Boulton FE, James V: Guidelines for policies on alternatives to allogeneic blood transfusion. 1. Predeposit autologous blood donation and transfusion. Transfus Med 2007; 17: 354

96. Rock G, Berger R, Bormanis J, et al: A review of nearly two decades in an autologous blood programme: The rise and fall of activity. Transfus Med 2006; 16: 307

97. Goodnough LT, Shander A: Blood management. Arch Pathol Lab Med 2007; 131: 695

98. Goodnough LT: The use of erythropoietin in the enhancement of autologous transfusion therapy. Curr Opin Hematol 1995; 2: 214

99. Laupacis A, Fergusson D: Erythropoietin to minimize perioperative blood transfusion: a systematic review of randomized trials. The International Study of Peri-operative Transfusion (ISPOT) Investigators. Transfus Med 1998; 8: 309

100. Milbrink J, Birgegard G, Danersund A, et al: Preoperative autologous donation of 6 units of blood during rh-EPO treatment. Can J Anaesth 1997; 44: 1315

101. Monk TG, Goodnough LT, Brecher ME, et al: A prospective randomized comparison of three blood conservation strategies for radical prostatectomy. Anesthesiology 1999; 91: 24

102. Monk TG: Preoperative recombinant human erythropoietin in anemic surgical patients. Crit Care 2004; 8(Suppl 2): S45

103. Rosencher N, Poisson D, Albi A, et al: Two injections of erythropoietin correct moderate anemia in most patients awaiting orthopedic surgery. Can J Anaesth 2005; 52: 160

104. MacLaren R, Sullivan PW: Cost-effectiveness of recombinant human erythropoietin for reducing red blood cells transfusions in critically ill patients. Value Health 2005; 8: 105

105. Karkouti K, McCluskey SA, Evans L, et al: Erythropoietin is an effective clinical modality for reducing RBC transfusion in joint surgery. Can J Anaesth 2005; 52: 362

106. Pierson JL, Hannon TJ, Earles DR: A blood-conservation algorithm to reduce blood transfusions after total hip and knee arthroplasty. J Bone Joint Surg Am 2004; 86-A: 1512

107. Couvret C, Laffon M, Baud A, et al: A restrictive use of both autologous donation and recombinant human erythropoietin is an efficient policy for primary total hip or knee arthroplasty. Anesth Analg 2004; 99: 262

108. Marchetti M, Barosi G: Cost-effectiveness of epoetin and autologous blood donationin reducing allogeneic blood transfusions incoronary artery bypass graft surgery. Transfusion 2000; 40: 673

109. Avall A, Hyllner M, Bengtson JP, et al: Recombinant human erythropoietin in preoperative autologous blood donation did not influence the haemoglobin recovery after surgery. Acta Anaesthesiol Scand 2003; 47: 687

110. Rosengart TK, Helm RE, Klemperer J, et al: Combined aprotinin and erythropoietin use for blood conservation: results with Jehovah's Witnesses. Ann Thorac Surg 1994; 58: 1397

111. Price S, Pepper JR, Jaggar SI: Recombinant human erythropoietin use in a critically ill Jehovah's witness after cardiac surgery. Anesth Analg 2005; 101: 325

112. Corwin HL, Gettinger A, Pearl RG, et al: Efficacy of recombinant human erythropoietin in critically ill patients: a randomized controlled trial. JAMA 2002; 288: 2827

113. Egrie JC, Dwyer E, Browne JK, et al: Darbepoetin alfa has a longer circulating half-life and greater in vivo potency than recombinant human erythropoietin. Exp Hematol 2003; 31: 290

114. Segal JB, Blasco-Colmenares E, Norris EJ, et al: Preoperative acute normovolemic hemodilution: a meta-analysis. Transfusion 2004; 44: 632

115. Goodnough LT: Acute normovolemic hemodilution. Vox Sang 2002; 83(Suppl 1): 211

116. Weiskopf RB: Efficacy of acute normovolemic hemodilution assessed as a function of fraction of blood volume lost. Anesthesiology 2001; 94: 439

117. Matot I, Scheinin O, Jurim O, et al: Effectiveness of acute normovolemic hemodilution to minimize allogeneic blood transfusion in major liver resections. Anesthesiology 2002; 97: 794

118. Goodnough LT, Despotis GJ, Merkel K, et al: A randomized trial comparing acute normovolemic hemodilution and preoperative autologous blood donation in total hip arthroplasty. Transfusion 2000; 40: 1054

119. Wolowczyk L, Lewis DR, Nevin M, et al: The effect of acute normovolaemic haemodilution on blood transfusion requirements in abdominal aortic aneurysm repair. Eur J Vasc Endovasc Surg 2001; 22: 361

120. Bennett J, Haynes S, Torella F, et al: Acute normovolemic hemodilution in moderate blood loss surgery: a randomized controlled trial. Transfusion 2006; 46: 1097

121. Williamson KR, Taswell HF: Intraoperative blood salvage: A review. Transfusion 1991; 31: 662

122. Ereth MH, Oliver WC, Jr., Santrach PJ: Perioperative interventions to decrease transfusion of allogeneic blood products. Mayo Clin Proc 1994; 69: 575

123. Hughes LG, Thomas DW, Wareham K, et al: Intra-operative blood salvage in abdominal trauma: a review of 5 years' experience. Anaesthesia 2001; 56: 217

124. Desmond MJ, Thomas MJ, Gillon J, et al: Consensus conference on autologous transfusion. Perioperative red cell salvage. Transfusion 1996; 36: 644

125. Huet C, Salmi LR, Fergusson D, et al: A meta-analysis of the effectiveness of cell salvage to minimize perioperative allogeneic blood transfusion in cardiac and orthopedic surgery. International Study of Perioperative Transfusion (ISPOT) Investigators. Anesth Analg 1999; 89: 861

126. Hashimoto T, Kokudo N, Orii R, et al: Intraoperative blood salvage during liver resection: a randomized controlled trial. Ann Surg 2007; 245: 686

127. Fujimoto J, Okamoto E, Yamanaka N, et al: Efficacy of autotransfusion in hepatectomy for hepatocellular carcinoma. Arch Surg 1993; 128: 1065

128. Hart OJ, 3rd, Klimberg IW, Wajsman Z, et al: Intraoperative autotransfusion in radical cystectomy for carcinoma of the bladder. Surg Gynecol Obstet 1989; 168: 302


129. Bernstein HH, Rosenblatt MA, Gettes M, et al: The ability of the Haemonetics 4 Cell Saver System to remove tissue factor from blood contaminated with amniotic fluid. Anesth Analg 1997; 85: 831

130. Potter PS, Waters JH, Burger GA, et al: Application of cell-salvage during cesarean section. Anesthesiology 1999; 90: 619

131. Weiskopf RB: Erythrocyte salvage during cesarean section. Anesthesiology 2000; 92: 1519

132. Waters JH, Tuohy MJ, Hobson DF, et al: Bacterial reduction by cell salvage washing and leukocyte depletion filtration. Anesthesiology 2003; 99: 652

133. Biagini D, Filippucci E, Agnelli G, et al: Activation of blood coagulation in patients undergoing postoperative blood salvage and re-infusion of unwashed whole blood after total knee arthroplasty. Thromb Res 2004; 113: 211

134. Tawes RL, Jr., Sydorak GR, DuVall TB: Postoperative salvage: a technological advance in the ‘washed’ versus ‘unwashed’ blood controversy. Semin Vasc Surg 1994; 7: 98

135. Djaiani G, Fedorko L, Borger MA, et al: Continuous-flow cell saver reduces cognitive decline in elderly patients after coronary bypass surgery. Circulation 2007; 116: 1888

136. McKie JS, Herzenberg JE: Coagulopathy complicating intraoperative blood salvage in a patient who had idiopathic scoliosis. A case report. J Bone Joint Surg Am 1997; 79: 1391

137. Waters JH, Lee JS, Karafa MT: A mathematical model of cell salvage efficiency. Anesth Analg 2002; 95: 1312

138. Drummond JC, Petrovitch CT: Intraoperative blood salvage: Fluid replacement calculations. Anesth Analg 2005; 100(3): 645

139. Waters JH, Williams B, Yazer MH, et al: Modification of suction-induced hemolysis during cell salvage. Anesth Analg 2007; 104: 684

140. Waters JH, Dyga RM: Postoperative blood salvage: outside the controlled world of the blood bank. Transfusion 2007; 47: 362

141. Strumper D, Weber EW, Gielen-Wijffels S, et al: Clinical efficacy of postoperative autologous transfusion of filtered shed blood in hip and knee arthroplasty. Transfusion 2004; 44: 1567

142. Munoz M, Cobos A, Campos A, et al: Impact of postoperative shed blood transfusion, with or without leucocyte reduction, on acute-phase response to surgery for total knee replacement. Acta Anaesthesiol Scand 2005; 49: 1182

143. Moonen AF, Knoors NT, van Os JJ, et al: Retransfusion of filtered shed blood in primary total hip and knee arthroplasty: a prospective randomized clinical trial. Transfusion 2007; 47: 379

144. Griffith LD, Billman GF, Daily PO, et al: Apparent coagulopathy caused by infusion of shed mediastinal blood and its prevention by washing of the infusate. Ann Thorac Surg 1989; 47: 400

145. Winslow RM: Current status of oxygen carriers (‘blood substitutes’): 2006. Vox Sang 2006; 91: 102

146. Spahn DR, Kocian R: The place of artificial oxygen carriers in reducing allogeneic blood transfusions and augmenting tissue oxygenation. Can J Anaesth 2003; 50: S41

147. The real value of blood (accessible at Awake! August: 2006

148. Fung MK, Downes KA, Shulman IA: Transfusion of platelets containing ABO-incompatible plasma: a survey of 3156 North American laboratories. Arch Pathol Lab Med 2007; 131: 909

149. Kleinman SH, Glynn SA, Lee TH, et al: Prevalence and quantitation of parvovirus B19 DNA levels in blood donors with a sensitive polymerase chain reaction screening assay. Transfusion 2007; 47: 1756

150. Parsyan A, Candotti D: Human erythrovirus B19 and blood transfusion - an update. Transfus Med 2007; 17: 263

151. Reiner A: Massive Transfusion, Perioperative Transfusion Medicine. Edited by Spiess B, Counts R, Gould S. Philadelphia: Williams & Wilkins, 1998, pp 351.

152. Key NS, Negrier C: Coagulation factor concentrates: past, present, and future. Lancet 2007; 370: 439

153. Vinazzer H: Clinical use of antithrombin III concentrations. Vox Sang 1997; 53: 193

154. Avidan MS, Levy JH, Scholz J, et al: A phase III, double-blind, placebo-controlled, multicenter study on the efficacy of recombinant human antithrombin in heparin-resistant patients scheduled to undergo cardiac surgery necessitating cardiopulmonary bypass. Anesthesiology 2005; 102: 276

155. Wiedermann CJ, Kaneider NC: A systematic review of antithrombin concentrate use in patients with disseminated intravascular coagulation of severe sepsis. Blood Coagul Fibrinolysis 2006; 17: 521

156. Hoffman M, Monroe DM: Coagulation 2006: a modern view of hemostasis. Hematol Oncol Clin North Am 2007; 21: 1

157. Tilley R, Mackman N: Tissue factor in hemostasis and thrombosis. Semin Thromb Hemost 2006; 32: 5

158. Monroe D, MKey NS: The tissue factor-factor VIIa complex: procoagulant activity, regulation, and multitasking. J Thromb Haemost 2007; 5: 1097

159. Zarbock A, Polanowska-Grabowska R, KLey K: Platelet-neutrophil-interactions: linking hemostasis and inflammation. Blood Rev 2007; 21: 99

160. Ofosu FA: Protease activated receptors 1 and 4 govern the responses of human platelets to thrombin. Transfus Apher Sci 2003; 28: 265

161. Wettstein P, Haeberli A, Stutz M, et al: Decreased factor XIII availability for thrombin and early loss of clot firmness in patients with unexplained intraoperative bleeding. Anesth Analg 2004; 99: 1564

162. Wolberg AS: Thrombin generation and fibrin clot structure. Blood Rev 2007; 21: 131

163. Mosnier LO, Bouma BN: Regulation of fibrinolysis by thrombin activatable fibrinolysis inhibitor, an unstable carboxypeptidase B that unites the pathways of coagulation and fibrinolysis. Arterioscler Thromb Vasc Biol 2006; 26: 2445

164. Nilsson IM: Coagulation and fibrinolysis. Scand J Gastroenterol Suppl 1987; 137: 11

165. Hoffman M: A cell-based model of coagulation and the role of factor VIIa. Blood Rev 2003; 17(Suppl 1): S1

166. Buller HR, ten Cate JW: Acquired antithrombin III deficiency: laboratory diagnosis, incidence, clinical implications, and treatment with antithrombin III concentrate. Am J Med 1989; 87: 44S

167. Sniecinski RM, Chen EP, Levy JH, et al: Coagulopathy after cardiopulmonary bypass in Jehovah's Witness patients: management of two cases using fractionated components and factor VIIa. Anesth Analg 2007; 104: 763

168. Bucur SZ, Levy JH, Despotis GJ, et al: Uses of antithrombin III concentrate in congenital and acquired deficiency states. Transfusion 1998; 38: 481

169. Lemmer JH, Jr., Despotis GJ: Antithrombin III concentrate to treat heparin resistance in patients undergoing cardiac surgery. J Thorac Cardiovasc Surg 2002; 123: 213

170. Cardigan R, Turner C, Harrison P: Current methods of assessing platelet function: relevance to transfusion medicine. Vox Sang 2005; 88: 153

171. Rodgers RP, Levin J: A critical reappraisal of the bleeding time. Semin Thromb Hemost 1990; 16: 1

172. Agarwal S, Coakley M, Reddy K, et al: Quantifying the effect of antiplatelet therapy: a comparison of the platelet function analyzer (PFA-100) and modified thromboelastography (mTEG) with light transmission platelet aggregometry. Anesthesiology 2006; 105: 676

173. Dyszkiewicz-Korpanty A, Olteanu H, Frenkel EP, et al: Clopidogrel anti-platelet effect: an evaluation by optical aggregometry, impedance aggregometry, and the platelet function analyzer (PFA-100). Platelets 2007; 18: 491

174. Hayward CP: Further complexities in diagnosing acquired thrombocytopenia: unexpected parallels between antibody-mediated delayed thrombocytopenia with abciximab and heparin induced thrombocytopenia. Thromb Haemost 2004; 92: 674

175. Nemerson Y: The tissue factor pathway of blood coagulation. Semin Hematol 1992; 29: 170

176. Poller L: The British system for anticoagulant control. Thromb Diath Haemorrh 1975; 33: 157

177. Henriksen R: Instrumentation and quality control of hemostasis, Clinical Hematology. Edited by Lotspiech-Steininger C S-ME, Koepke J. Philadelphia, J.B. Lippincott, 1992, pp 695

178. Triplett DA: Overview of hemostasis, Hemostatic Disorders and the Blood Bank. Edited by Menitove JE ML. Arlington, Va, American Association of Blood Banks, 1984, pp 1

179. Casserly IP, Kereiakes DJ, Gray WA, et al: Point-of-care ecarin clotting time versus activated clotting time in correlation with bivalirudin concentration. Thromb Res 2004; 113: 115

180. Murphy GS, Marymont JH: Alternative anticoagulation management strategies for the patient with heparin-induced thrombocytopenia undergoing cardiac surgery. J Cardiothorac Vasc Anesth 2007; 21: 113

181. Carroll RC, Chavez JJ, Simmons JW, et al: Measurement of patients' bivalirudin plasma levels by a thrombelastograph ecarin clotting time assay: a comparison to a standard activated clotting time. Anesth Analg 2006; 102: 1316

182. Traverso CI, Caprini JA, Arcelus JI: The normal thromboelastogram and its interpretation. Semin Thromb Hemost 1995; 21(Suppl 4): 7

183. Tuman KJ, Spiess BD, McCarthy RJ, et al: Effects of progressive blood loss on coagulation as measured by thrombelastography. Anesth Analg 1987; 66: 856

184. Kang W: Blood coagulation during liver, kidney, and pancreas transplantation, Blood: Hemostasis, Transfusion, and Alternatives in the Perioperative Period. Edited by Lake CL MR. New York, Raven Press, 1995, pp 529

185. Zuckerman L, Cohen E, Vagher JP, et al: Comparison of thrombelastography with common coagulation tests. Thromb Haemost 1981; 46: 752

186. Levine JS, Branch DWRauch J: The antiphospholipid syndrome. N Engl J Med 2002; 346: 752

187. Neunert CE, Journeycake JM: Congenital platelet disorders. Hematol Oncol Clin North Am 2007; 21: 663

188. Ewenstein BM: Von Willebrand's disease. Annu Rev Med 1997; 48: 525

189. Vischer UM, de Moerloose P: von Willebrand factor: from cell biology to the clinical management of von Willebrand's disease. Crit Rev Oncol Hematol 1999; 30: 93

190. Federici AB, Mazurier C, Berntorp E, et al: Biologic response to desmopressin in patients with severe type 1 and type 2 von Willebrand disease: results of a multicenter European study. Blood 2004; 103: 2032

191. Rodeghiero F, Castaman G: Treatment of von Willebrand disease. Semin Hematol 2005; 42: 29

192. Franchini M: The use of desmopressin as a hemostatic agent: a concise review. Am J Hematol 2007; 82: 731

193. Mannucci PM: Treatment of von Willebrand's Disease. N Engl J Med 2004; 351: 683

194. Michiels JJ, Berneman ZN, van der Planken M, et al: Bleeding prophylaxis for major surgery in patients with type 2 von Willebrand disease with an intermediate purity factor VIII-von Willebrand factor concentrate (Haemate-P). Blood Coagul Fibrinolysis 2004; 15: 323


195. Franchini M, Gandini G, Veneri D, et al: Safety and efficacy of subcutaneous bolus injection of deferoxamine in adult patients with iron overload: an update. Blood 2004; 103: 747

196. Lee JW: von Willebrand disease, hemophilia A and B, and other factor deficiencies. Int Anesthesiol Clin 2004; 42: 59

197. Kaufmann JE, Vischer UM: Cellular mechanisms of the hemostatic effects of desmopressin (DDAVP). J Thromb Haemost 2003; 1: 682

198. Warrier AI, Lusher JM: DDAVP: a useful alternative to blood components in moderate hemophilia A and von Willebrand disease. J Pediatr 1983; 102: 228

199. Sohal AS, Gangji AS, Crowther MA, et al: Uremic bleeding: pathophysiology and clinical risk factors. Thromb Res 2006; 118: 417

200. Gangji AS, Sohal AS, Treleaven D, et al: Bleeding in patients with renal insufficiency: a practical guide to clinical management. Thromb Res 2006; 118: 423

201. Moia M, Mannucci PM, Vizzotto L, et al: Improvement in the haemostatic defect of uraemia after treatment with recombinant human erythropoietin. Lancet 1987; 2: 1227

202. Tran HA, Anand SS, Hankey GJ, et al: Aspirin resistance. Thromb Res 2007; 120: 337

203. Steffel J, Luscher TF, Ruschitzka F, et al: Cyclooxygenase-2 inhibition and coagulation. J Cardiovasc Pharmacol 2006; 47(Suppl 1): S15

204. Kam PC, Egan MK: Platelet glycoprotein IIb/IIIa antagonists: pharmacology and clinical developments. Anesthesiology 2002; 96: 1237

205. Merlini PA, Rossi M, Menozzi A, et al: Thrombocytopenia caused by abciximab or tirofiban and its association with clinical outcome in patients undergoing coronary stenting. Circulation 2004; 109: 2203

206. Couris RR: Vitamins and minerals that affect hemostasis and antithrombotic therapies. Thromb Res 2005; 117: 25

207. Szuwart T, Brzoska T, Luger TA, et al: Vitamin E reduces platelet adhesion to human endothelial cells in vitro. Am J Hematol 2000; 65: 1

208. Makris M, Watson HG: The management of coumarin-induced over-anticoagulation Annotation. Br J Haematol 2001; 114: 271

209. Baglin TP, Keeling DM, Watson HG: Guidelines on oral anticoagulation (warfarin): third edition—2005 update. Br J Haematol 2006; 132: 277

210. Makris M: Optimisation of the prothrombin complex concentrate dose for warfarin reversal. Thromb Res 2005; 115: 451

211. Ghorashian S, Hunt BJ: “Off-license” use of recombinant activated factor VII. Blood Rev 2004; 18: 245

212. Dickneite G: Prothrombin complex concentrate versus recombinant factor VIIa for reversal of coumarin anticoagulation. Thromb Res 2007; 119: 643

213. O'Connell KA, Wood JJ, Wise RP, et al: Thromboembolic adverse events after use of recombinant human coagulation factor VIIa. JAMA 2006; 295: 293

214. Geerts WH, Pineo GF, Heit JA, et al: Prevention of venous thromboembolism: the Seventh ACCP Conference on Antithrombotic and Thrombolytic Therapy. Chest 2004; 126: 338S

215. Bombeli T, Spahn DR: Updates in perioperative coagulation: physiology and management of thromboembolism and haemorrhage. Br J Anaesth 2004; 93: 275

216. Groce JB 3rd: Treatment of deep vein thrombosis using low-molecular-weight heparins. Am J Manag Care 2001; 7: S510

217. Boneu B, de Moerloose P: How and when to monitor a patient treated with low molecular weight heparin. Semin Thromb Hemost 2001; 27: 519

218. Walenga JM, Jeske WP, Prechel MM, et al: Decreased prevalence of heparin-induced thrombocytopenia with low-molecular-weight heparin and related drugs. Semin Thromb Hemost 2004; 30(Suppl 1): 69

219. Warkentin TE: Heparin-induced thrombocytopenia. Hematol Oncol Clin North Am 2007; 21: 589

220. Levy JH, Tanaka KA, Hursting MJ: Reducing thrombotic complications in the perioperative setting: an update on heparin-induced thrombocytopenia. Anesth Analg 2007; 105: 570

221. Gruel Y, Pouplard C, Nguyen P, et al: Biological and clinical features of low-molecular-weight heparin-induced thrombocytopenia. Br J Haematol 2003; 121: 786

222. Hirsh J, O'Donnell M, Eikelboom JW: Beyond unfractionated heparin and warfarin: current and future advances. Circulation 2007; 116: 552

223. Koster A, Kukucka M, Bach F, et al: Anticoagulation during cardiopulmonary bypass in patients with heparin-induced thrombocytopenia type II and renal impairment using heparin and the platelet glycoprotein IIb-IIIa antagonist tirofiban. Anesthesiology 2001; 94: 245

224. Despotis GJ, Joist JH: Anticoagulation and anticoagulation reversal with cardiac surgery involving cardiopulmonary bypass: an update. J Cardiothorac Vasc Anesth 1999; 13: 18

225. Paparella D, Galeone A, Venneri MT, et al: Activation of the coagulation system during coronary artery bypass grafting: comparison between on-pump and off-pump techniques. J Thorac Cardiovasc Surg 2006; 131: 290

226. Paparella D, Al Radi OO, Meng QH, et al: The effects of high-dose heparin on inflammatory and coagulation parameters following cardiopulmonary bypass. Blood Coagul Fibrinolysis 2005; 16: 323

227. Body SC, Morse DS: Coagulation, transfusion and cardiac surgery, Perioperative Transfusion Medicine. Edited by Spiess BD, Counts RB, Gould SA. Baltimore, Williams & Wilkins, 1998, pp 419

228. White CM: Thrombin-directed inhibitors: pharmacology and clinical use. Am Heart J 2005; 149: S54

229. Wasowicz M, Vegas A, Borger MA, et al: Bivalirudin anticoagulation for cardiopulmonary bypass in a patient with heparin-induced thrombocytopenia. Can J Anaesth 2005; 52: 1093

230. Dyke CM, Smedira NG, Koster A, et al: A comparison of bivalirudin to heparin with protamine reversal in patients undergoing cardiac surgery with cardiopulmonary bypass: the EVOLUTION-ON study. J Thorac Cardiovasc Surg 2006; 131: 533

231. Koster A, Chew D, Grundel M, et al: An assessment of different filter systems for extracorporeal elimination of bivalirudin: an in vitro study. Anesth Analg 2003; 96: 1316

232. Bramlage P, Pittrow D, Kirch W: Current concepts for the prevention of venous thromboembolism. Eur J Clin Invest 2005; 35(Suppl 1): 4

233. Warkentin TE, Cook RJ, Marder VJ, et al: Anti-platelet factor 4/heparin antibodies in orthopedic surgery patients receiving antithrombotic prophylaxis with fondaparinux or enoxaparin. Blood 2005; 106: 3791

234. Kahl B, Schwartz B, Mosher D: Profound imbalance of pro-fibrinolytic and anti-fibrinolytic factors (tissue plasminogen activator and plasminogen activator inhibitor type 1) and severe bleeding diathesis in a patient with cirrhosis: correction by liver transplantation. Blood Coagulation and Fibrinolysis 2003; 14: 741

235. Staudinger T, Locker GJ, Frass M: Management of acquired coagulation disorders in emergency and intensive-care medicine. Semin Thromb Hemost 1996; 22: 93

236. Bick RL: Disseminated intravascular coagulation current concepts of etiology, pathophysiology, diagnosis, and treatment. Hematol Oncol Clin North Am 2003; 17: 149

237. Levi M: Current understanding of disseminated intravascular coagulation. Br J Haematol 2004; 124: 567

238. Zeerleder S, Hack CE, Wuillemin WA: Disseminated intravascular coagulation in sepsis. Chest 2005; 128: 2864

239. Carey MJ, Rodgers GM: Disseminated intravascular coagulation: clinical and laboratory aspects. Am J Hematol 1998; 59: 65

240. Dempfle CE: Coagulopathy of sepsis. Thromb Haemost 2004; 91: 213

241. Roberts HR, Monroe DM, White GC: The use of recombinant factor VIIa in the treatment of bleeding disorders. Blood 2004; 104: 3858

242. Welsby IJ, Monroe DM, Lawson JH, et al: Recombinant activated factor VII and the anaesthetist. Anaesthesia 2005; 60: 1203

243. Friederich PW, Henny CP, Messelink EJ, et al: Effect of recombinant activated factor VII on perioperative blood loss in patients undergoing retropubic prostatectomy: a double-blind placebo-controlled randomised trial. Lancet 2003; 361: 201

244. Boffard KD, Riou B, Warren B, et al: Recombinant factor VIIa as adjunctive therapy for bleeding control in severely injured trauma patients: two parallel randomized, placebo-controlled, double-blind clinical trials. J Trauma 2005; 59: 8

245. Diprose P, Herbertson MJ, O'Shaughnessy D, et al: Activated recombinant factor VII after cardiopulmonary bypass reduces allogeneic transfusion in complex non-coronary cardiac surgery: randomized double-blind placebo-controlled pilot study. Br J Anaesth 2005; 95: 596

246. Mayer SA, Brun NC, Begtrup K, et al: Recombinant activated factor VII for acute intracerebral hemorrhage. N Engl J Med 2005; 352: 777

247. Hedner U: Mechanism of action of factor VIIa in the treatment of coagulopathies. Semin Thromb Hemost 2006; 32(Suppl 1): 77

248. Martinowitz U, Michaelson M: Guidelines for the use of recombinant activated factor VII (rFVIIa) in uncontrolled bleeding: a report by the Israeli Multidisciplinary rFVIIa Task Force. J Thromb Haemost 2005; 3: 640

249. Firozvi K, Deveras RA, Kessler CM: Reversal of low-molecular-weight heparin-induced bleeding in patients with pre-existing hypercoagulable states with human recombinant activated factor VII concentrate. Am J Hematol 2006; 81: 582

250. Khan AZ, Parry JM, Crowley WF, et al: Recombinant factor VIIa for the treatment of severe postoperative and traumatic hemorrhage. Am J Surg 2005; 189: 331

251. Shander A, Goodnough LT, Ratko T, et al: Consensus Recommendations for the Off-Label Use of Recombinant Human Factor VIIa (NovoSeven®) Therapy. P&T Journal 2005; 30: 644

252. Gordz S, Mrowietz C, Pindur G, et al: Effect of desmopressin (DDAVP) on platelet membrane glycoprotein expression in patients with von Willebrand's disease. Clin Hemorheol Microcirc 2005; 32: 83

253. Cattaneo M, Harris AS, Stromberg U, et al: The effect of desmopressin on reducing blood loss in cardiac surgery—a meta-analysis of double-blind, placebo-controlled trials. Thromb Haemost 1995; 74: 1064

254. Salzman EW, Weinstein MJ, Weintraub RM, et al: Treatment with desmopressin acetate to reduce blood loss after cardiac surgery. A double-blind randomized trial. N Engl J Med 1986; 314: 1402

255. Czer LS, Bateman TM, Gray RJ, et al: Treatment of severe platelet dysfunction and hemorrhage after cardiopulmonary bypass: reduction in blood product usage with desmopressin. J Am Coll Cardiol 1987; 9: 1139

256. Mongan PD, Hosking MP: The role of desmopressin acetate in patients undergoing coronary artery bypass surgery. A controlled clinical trial with thromboelastographic risk stratification. Anesthesiology 1992; 77: 38

257. Gratz I, Koehler J, Olsen D, et al: The effect of desmopressin acetate on postoperative hemorrhage in patients receiving aspirin therapy before coronary artery bypass operations. J Thorac Cardiovasc Surg 1992; 104: 1417


258. Carless PA, Henry DA, Moxey AJ, et al: Desmopressin for minimising perioperative allogeneic blood transfusion. Cochrane Database Syst Rev 2004; CD001884

259. Diprose P, Herbertson MJ, O'Shaughnessy D, et al: Reducing allogeneic transfusion in cardiac surgery: a randomized double-blind placebo-controlled trial of antifibrinolytic therapies used in addition to intra-operative cell salvage. Br J Anaesth 2005; 94: 271

260. Zufferey P, Merquiol F, Laporte S, et al: Do antifibrinolytics reduce allogeneic blood transfusion in orthopedic surgery? Anesthesiology 2006; 105: 1034

261. Henry DA, Carless PA, Moxey AJ, et al: Anti-fibrinolytic use for minimising perioperative allogeneic blood transfusion. Cochrane Database Syst Rev 2007; CD001886

262. Molenaar IQ, Warnaar N, Groen H, et al: Efficacy and safety of antifibrinolytic drugs in liver transplantation: a systematic review and meta-analysis. Am J Transplant 2007; 7: 185

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

264. Mangano DT, Miao Y, Vuylsteke A, et al: Mortality associated with aprotinin during 5 years following coronary artery bypass graft surgery. JAMA 2007; 297: 471

265. Cirino G, Napoli C, Bucci M, et al: Inflammation-coagulation network: are serine protease receptors the knot? Trends Pharmacol Sci 2000; 21: 170

266. McEvoy MD, Reeves ST, Reves JG, et al: Aprotinin in cardiac surgery: a review of conventional and novel mechanisms of action. Anesth Analg 2007; 105: 949

267. Day JR, Taylor KM, Lidington EA, et al: Aprotinin inhibits proinflammatory activation of endothelial cells by thrombin through the protease-activated receptor 1. J Thorac Cardiovasc Surg 2006; 131: 21

268. van Oeveren W, Harder MP, Roozendaal KJ, et al: Aprotinin protects platelets against the initial effect of cardiopulmonary bypass. J Thorac Cardiovasc Surg 1990; 99: 788

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

270. Sedrakyan A, Treasure T, Elefteriades JA: Effect of aprotinin on clinical outcomes in coronary artery bypass graft surgery: a systematic review and meta-analysis of randomized clinical trials. J Thorac Cardiovasc Surg 2004; 128: 442

271. Karkouti K, Beattie WS, Dattilo KM, et al: A propensity score case-control comparison of aprotinin and tranexamic acid in high-transfusion-risk cardiac surgery. Transfusion 2006; 46: 327

272. Furnary AP, Wu Y, Hiratzka LF, et al: Aprotinin does not increase the risk of renal failure in cardiac surgery patients. Circulation 2007; 116: I127

273. Dietrich W, Busley R, Boulesteix AL: Effects of aprotinin dosage on renal function: an analysis of 8,548 cardiac surgical patients treated with different dosages of aprotinin. Anesthesiology 2008; 108: 189

274. Schneeweiss S, Seeger JD, Landon J, et al: Aprotinin during coronary-artery bypass grafting and risk of death. N Engl J Med 2008; 358: 771

275. Casati V, Guzzon D, Oppizzi M, et al: Hemostatic effects of aprotinin, tranexamic acid and epsilon-aminocaproic acid in primary cardiac surgery. Ann Thorac Surg 1999; 68: 2252

276. Bowden RA, Slichter SJ, Sayers MH, et al: Use of leukocyte-depleted platelets and cytomegalovirus-seronegative red blood cells for prevention of primary cytomegalovirus infection after marrow transplant. Blood 1991; 78: 246

277. Klein HG, Spahn DR, Carson JL: Red blood cell transfusion in clinical practice. Lancet 2007; 370: 415

278. Walker RH: Special report: transfusion risks. Am J Clin Pathol 1987; 88: 374

279. Kang Y, Lewis JH, Navalgund A, et al: Epsilon-aminocaproic acid for treatment of fibrinolysis during liver transplantation. Anesthesiology 1987; 66: 766

280. Kang Y: Monitoring and treatment of coagulation, Hepatic Transplantation: Anesthetic and Periperative Management. Edited by Winter K, Kang Y. New York, Praeger, 1986, pp 151

Editors: Barash, Paul G.; Cullen, Bruce F.; Stoelting, Robert K.; Cahalan, Michael K.; Stock, M. Christine