Blood must not only be maintained as a fluid in normal circulation, but also be capable of forming a solid clot to staunch leaks in the vascular wall—and then dismantling the clot when the need has passed. This delicate equilibrium between anticoagulation and coagulation is maintained by a complex system of counterbalanced blood proteins and cells (platelets). Many congenital and acquired disorders can push the system toward either bleeding or thrombosis. The patient care team has a number of tests to evaluate the system and many therapeutic modalities to correct these imbalances.

Platelets adhere to sites of endothelial disruption, undergo activation to recruit more platelets and amplify the platelet response, and then cross-link with fibrin, the end product of the plasma clotting factor cascade, to form a platelet plug. Primary hemostasis (Fig. 16-1) describes the initiation of the platelet plug and clotting mechanism.

Clotting factors in the plasma are activated at sites of endothelial injury and assemble in enzymatic complexes to activate thrombin. This initiates secondary hemostasis. Thrombin then amplifies production of itself by activating other more efficient enzymes which propagate a thrombin burst. Thrombin also converts fibrinogen to fibrin, which cross-links with activated platelets to form the platelet plug. Each of the three enzymatic complexes in the clotting process consists of four parts: An enzyme in the serine protease family, a cofactor, a plasma membrane phospholipid surface such as the platelet, and calcium ion (Ca²⁺). The proteases convert other clotting factors from their inactive circulating configuration to an active form (termed [factor number]a).⁵

The extrinsic pathway (Fig. 16-2A–E). The process begins when endothelial disruption exposes tissue factor (TF) on underlying cell membranes (A), extrinsic to the circulation—hence
the term “extrinsic pathway.” TF binds both VII and VIIa, which circulates at low levels, and is a cofactor for the activation of factor VII (B). VIIa enzyme, TF cofactor, cell membrane phospholipid, and Ca²⁺ form the first complex, a low-efficiency extrinsic-pathway “tenase” which activates factor X and factor IX (C). Then Xa enzyme, its cofactor Va (derived in large part from factor V released from activated platelet α-granules), phospholipid, and Ca²⁺ assemble to form the second complex, a “prothrombinase,” which converts prothrombin (II) to thrombin (IIa) (E).⁵

The intrinsic pathway (Fig. 16-2F–J). Thrombin has several central functions. It activates platelets via surface receptors PAR-1 and PAR-4 (see Primary Hemostasis), cleaves more V to Va (F), and initiates the “intrinsic” (intravascular) coagulation pathway by cleaving factor XI to XIa (G). XIa cleaves more IX to IXa (H). Thrombin also activates VIII to VIIIa (G). (VIII is carried and stabilized in the plasma by vWF until needed, and vWF deficiency results in low plasma VIII levels also.) The third complex is then formed: IXa enzyme, VIIIa cofactor, phospholipid, and Ca²⁺. This is a high-efficiency intrinsic-pathway “tenase” (I) which provides many times more Xa for more prothrombinase complex (J). Ultimately, thrombin cleaves fibrinogen to fibrin monomers, which then polymerize extensively. Fibrin polymers are cross-linked by factor XIIIa (also activated by thrombin) to form the stable fibrin clot (K). Fibrin also cross-links activated platelets by their GP IIb/IIIa receptors to enmesh platelets and fibrin in the platelet plug (see Primary Hemostasis).⁵

All of these clotting factors are primarily produced in the liver, except for VIII, which is also released by endothelial cells and is well maintained in liver disease. The plasma half-life of most clotting factors is around 1.5 to 3 days, but those of the initiating factor VII (6 hours) and the cofactors V and VIII (8 to 12 hours) are much shorter. Four critical enzyme factors—VII in the extrinsic tenase, IX in the intrinsic tenase, X in the prothrombinase, and prothrombin (II)—must be carboxylated at multiple glutamic acid residues after translation, in order to interact with phospholipid and Ca²⁺. Vitamin K in its reduced form is the cofactor for the glutamyl-carboxylase enzyme, and thus these factors are vitamin K–dependent.⁵

Fibrin clots must be broken down after their job is done, and fibrinolysis is also a complex process with checks and balances. In the end, plasminogen is activated to plasmin, which breaks down fibrin polymers (Fig. 16-4). Plasminogen activation has a minor and a major pathway. In the minor pathway, IXa, XIIa, and kallikrein can each activate plasminogen. The latter two factors are part of the “contact factors” which initiate in vitro clotting in the activated partial thromboplastin (activated partial thromboplastin time [aPTT]) test, but are not thought to take part in clotting in vivo. The minor pathway is thought to account for a small fraction of fibrinolysis.

The major activator of plasminogen in the blood is tissue plasminogen activator (tPA), which is secreted from endothelial cells. When associated with cross-linked fibrin, tPA becomes much more efficient. Once some plasmin is formed, it cleaves tPA to a more active form. tPA also directly cleaves fibrin polymers. In tissues, urokinase plasminogen activator is important in fibrinolysis. Urokinase is secreted from the endothelial cells and the kidney. Plasmin also activates urokinase to a more active form.

The normal automated platelet count in adults is approximately 150,000 to 400,000/μL. The peripheral blood smear should be examined in specimens with abnormal platelet counts or size. Microscopic review may reveal clotted specimens, artifactual platelet clumping in vitro, or abnormal platelet morphology. Large platelets are seen in some congenital disorders. One of the first platelet function tests (PFTs) was the template bleeding time, in which a standardized small cut is made on the subject’s forearm and the bleeding duration timed. However, this test is invasive, labor-intensive, impractical to repeat frequently, poorly reproducible, and only modestly predictive for bleeding problems.

In vitro PFTs use various platelet agonists to activate and aggregate the patient’s platelets.⁷ For example, the PFA-100® device (Siemens, Munich, Germany) simulates capillary blood flow through a chamber after activation by collagen and either epinephrine or ADP. Prolonged “closure time” with collagen/epinephrine but not collagen/ADP suggests aspirin or other antiplatelet medications, whereas when both pairs are abnormal, other congenital or acquired platelet dysfunctions may be present. This type of testing is sometimes used as a screen in patients with a history suggesting platelet problems or von Willebrand disease (vWD). However, the sensitivity and specificity are low. False negatives are common, and abnormal results can also be caused by thrombocytopenia, uremia, or anemia. Several other devices test for specific antiplatelet medication effects from aspirin or P2Y12 inhibitors.⁷

Platelet aggregation is the most detailed overall PFT. Platelets are tested with multiple isolated agonists to assess their patterns of physical aggregation and, in turn, the platelets’ own agonist release. Some uncommon congenital disorders lack responses to specific agonists in a characteristic fashion. More detailed testing may be needed for a specific diagnosis, such as electron microscopy for granule defects, flow cytometry for surface receptors and granule markers, or genetic testing.⁸

vWD is a factor deficiency which imparts clinical features of platelet dysfunction, due to the central role of vWF in cross-linking activated platelets to form the platelet plug.⁹ Up to 1% of all patients have vWD, with a wide range of severity due to either quantitative or functional defects of vWF. Diagnostic testing is integral to deciphering the specific defect and type of vWD to ensure the appropriate treatment. Since vWF is the carrier for factor VIII (FVIII) in plasma, vWF protein levels usually correlate with FVIII levels. Initial testing for vWD should include the vWF antigen level, vWF activity level, and FVIII activity level for comparison with vWF. Blood group O persons have shorter plasma half-life and lower normal levels of vWF, so ABO blood typing may be needed to interpret a borderline vWF level. Type 1 vWD is a quantitative deficiency, with decreased antigen and activity. Type 2 vWD may have normal antigen levels, but decreased activity from a defective protein. Within type 2, there are several subtypes with different molecular defects, and specialized identification is needed in order to determine the best therapy. Type 3 vWD is a rare, very severe autosomal recessive deficiency.⁹ The clinical features and management of vWD are discussed later in this chapter.

A general oversight of plasma clotting factor activity is obtained by the prothrombin time (PT) for the intrinsic (tissue) pathway and the aPTT for the extrinsic (contact) pathway (Fig. 16-2), with both tests completed through the common pathway.⁶ These clotting tests are performed in blood specimens collected in a chelator (3.2% citrate) which binds Ca²⁺ to prevent clotting in the tube. The in vitro clotting test is activated by TF in the PT or negatively charged surfaces in the aPTT, using phospholipid as a platform (substituting for platelets). Ca²⁺ is then added to overcome the specimen chelation, and the time is measured
until complete fibrin clotting is observed. Representative normal ranges are around 12 to 15 seconds for the PT and 25 to 35 seconds for the aPTT, but are defined by each laboratory using its own equipment, reagents, and normal specimens. Testing is routinely performed at 37°C, but hypothermia in the patient impairs the enzymatic reactions of clot formation.

Clotting physiology is more complicated than the traditional diagrams of separate cascade pathways for these two tests. We have seen that thrombin from the intrinsic pathway activates the extrinsic pathway. In vitro, the aPTT clotting test is activated by synthetic contact materials which initiate via factor XII, so deficiencies of XII and other related contact factors cause a prolonged aPTT. However, deficiencies of these contact factors do not cause bleeding and may in fact be associated with impaired fibrinolysis and thrombosis. Fibrinogen activity is also a critical parameter. Most assays measure the functional conversion of fibrinogen to fibrin, although the fibrinogen protein level can also be measured for comparison to assess fibrinogen dysfunction. Normal fibrinogen levels are around 150 to 400 mg/dL.

Whole-blood clotting and fibrinolysis can be assessed by viscoelastic testing in thromboelastography (TEG®, Haemoscope Corp., Niles, IL, USA) or rotation thromboelastometry (ROTEM®, Pentapharm GmBH, Munich, Germany). These tests measure the rate, strength, and lysis, if any, of clot formation. Numerous parameters can be measured with these tests; accordingly the TEG–ROTEM working group attempted to standardize the parameters obtained from both testing modalities in order to make them more clinically relevant. There are minor differences in the mechanisms for TEG versus ROTEM; however, both involve the use of whole blood in a heated cup with the addition of a sensor pin. The cup or the pin oscillates while the blood clots. The increasing resistance to oscillation is transmitted through the sensor pin and a graphic depiction of clot formation is displayed in the thromboelastogram.¹² The patterns obtained can implicate defects in factor levels, platelet function, fibrinogen concentration, and the presence of abnormal fibrinolysis, the latter which is difficult to measure rapidly otherwise. Testing can be performed in the presence of inhibitors of heparin or fibrinolysis to help judge whether these drugs would be effective. This test format has also been adapted to assess antiplatelet therapy in patients with ventricular assist devices.¹³ Thromboelastography is helpful in determining the appropriate therapy, including platelets, plasma, fibrinogen replacement, or antifibrinolytics, as complex bleeding syndromes such as massive hemorrhage with consumptive or dilutional coagulopathy progress.

The risk for DVT, PE, VTE, and other thromboses is increased by intercurrent factors such as physical inactivity or immobilization, malignancy, oral contraceptives, estrogen therapy, and pregnancy. However, in addition to or especially in the absence of such factors, laboratory testing often identifies an underlying congenital
or acquired predisposing abnormality which tips the hemostatic systems toward thrombosis. Discovery of one or more risk factors may influence the course of therapy and suggest benefit from family studies.

Most anticoagulant therapies need ongoing or selective testing for assessment of therapeutic effect. Appropriate monitoring ensures that these agents are maintained within the therapeutic range; otherwise patients are at risk of thromboembolism and bleeding complications both of which can have devastating consequences.

The production of blood components is highly controlled by regulations and accreditation requirements in the interests of donor and recipient safety and therapeutic efficacy. Blood donors are carefully screened and tested, and blood products are made in specialized laboratories and other facilities. Regional blood centers collect and provide most blood components for transfusion, although some hospitals collect blood or platelets to augment their supply. Virtually all blood components come from volunteer unpaid donors. Pharmaceutical companies process plasma into various derivatives or synthesize some desired proteins for infusion.

Blood donors undergo a confidential interview to screen for medical problems in their donation safety and for risks of disease transmission to their recipients.²¹ They are questioned about risk factors, exposure, or signs of human immunodeficiency virus (HIV), hepatitis, and other infections. There are geographically based deferral criteria for tropical exposure to malaria and (in American donors) European exposure to variant Creutzfeldt–Jakob disease (vCJD). The donor’s pulse, blood pressure, and hemoglobin/hematocrit level (US minimum 12.5 g/dL or 38%) are checked. Phlebotomy is performed with validated antiseptic measures to reduce the risk of bacterial contamination in the blood bags. In the United States, donors are deferred for 8 weeks after a whole-blood donation to avoid iron deficiency.

Table 16-1 shows the contents and storage parameters for blood components. In whole-blood donations, 450 to 500 mL of blood is collected into citrate anticoagulant and then separated by centrifugation into RBCs, platelets, and/or plasma. The RBC units usually have most plasma removed and replaced with preservative. In the United States, the plasma must be frozen within 6 hours of collection to be labeled fresh frozen plasma (FFP). A large proportion of plasma is now made as plasma frozen within 24 hours, with minimal effect on clotting factor content compared to FFP. Cryoprecipitate is made from barely thawed FFP, which has a precipitate enriched in fibrinogen; the precipitate is isolated by centrifugation and refrozen.²² Five bags of “cryo” comprise a typical adult dose. Whole-blood–derived platelets (sometimes called “random-donor platelets”) are derived from platelet-rich plasma in the United States and from the buffy-coat centrifugation layer in other countries.²³ Four to six units are pooled to yield one adult-sized dose of platelets. Traditionally, pooling was done at the hospital just before transfusion, but blood centers can now provide pre-pooled platelets to hospitals.

Blood components are also collected by apheresis, in which the donor’s blood is processed by ex vivo centrifugation, the
desired component(s) are siphoned off into citrate anticoagulant, and the rest returned to the donor. Most platelets are produced by apheresis (sometimes called single-donor platelets). Plasma and RBCs can also be collected by apheresis, and if the donor’s blood volume and cell counts permit, two doses of the desired component can be obtained in one collection session.

All donors are checked against files of deferred donors, and all donations are tested for blood-borne infectious agents. Tests for HIV, hepatitis B virus (HBV), and hepatitis C virus (HCV) are universally required.²⁴ The addition of sensitive nucleic acid testing (NAT; e.g., polymerase chain reaction) to routine serology shortens the window period in a recently infected donor down to 7 to 10 days for HIV and HCV and 1 month for HBV. The US FDA also requires testing for syphilis, human T-cell lymphotropic virus (HTLV), and West Nile virus (WNV). Also available for donors at risk is a test for Trypanosoma cruzi (Chagas disease). Cytomegalovirus (CMV) is mainly in WBCs; selected patients at risk for complications of infection from cellular components can receive either CMV-negative units or leukoreduced units.

At some centers, directed donations can be made by family or friends for a specific patient. These donations augment the overall supply of blood and donors, and could reduce the number of donor exposures by using the same donor(s) more than once. However, the infectious risk from these units is not considered any less than from community volunteer donors. Cellular components from blood relatives must be irradiated to prevent the risk of transfusion graft-versus-host disease (GVHD) from closely matched donor lymphocytes which are not rejected by the recipient.

Autologous donations by the patient can be made in advance of scheduled surgery, with physician approval. The usual minimum hemoglobin or hematocrit is 11 g/dL or 33%, respectively, which is lower than the minimum allowed for regular donors. Patients should be on erythropoietin or iron supplementation to support RBC replenishment. In order to achieve a net gain in RBC mass, donations should be scheduled with some lead time within the 6-week storage period to permit erythropoiesis before surgery. Erythropoietin assists with multiple donations. Autologous units are only used for the patient and cannot be given to anyone else, since they often have a lower RBC content than normal requirements, and surgical patients frequently have conditions disqualifying them for regular donation.

Leukoreduction (LR) to remove WBCs from RBCs and platelets reduces the risk of human leukocyte antigen (HLA) alloimmunization, febrile nonhemolytic transfusion reactions (FNHTRs), and CMV transmission in patients who require these precautions. LR is usually performed by filtration, although some apheresis collections have sufficiently precise cell separation to minimize WBC content. The WBC content is reduced from around 10⁹ WBCs to ≤1 × 10⁶ to 5 × 10⁶ WBCs/unit.

In addition to the routine indications above, LR has been studied for the prevention of transfusion-related acute lung injury (TRALI) and so-called transfusion-related immunomodulation (TRIM), that is, possible immunosuppressive and pro-inflammatory effects mediated by donor WBCs. “Universal” LR of all cellular components is done in many countries, for this and other reasons. However, randomized controlled trials have been mostly inconclusive for benefit. Meta-analysis of available data has highlighted cardiac surgery as a setting in which there may be postoperative survival benefit associated with leukoreduced blood components, based on a small number of studies to date.²⁵

Washing cellular components with saline is mostly done to remove plasma in patients with allergic transfusion reactions such as those who are IgA deficient. It does not affect the antigens on the cells and does not remove enough WBCs to prevent GVHD or HLA alloimmunization.

Irradiation of cellular components is performed to prevent transfusion GVHD from directed-donor units from blood relatives and in highly immunosuppressed patients at risk for this complication: Leukemia, lymphoma, hematopoietic stem cell transplants, congenital cellular immunodeficiencies.²⁶ The units are exposed to gamma irradiation (2,500 cGy) to damage donor WBC DNA and prevent a cellular immune proliferative response to the recipient’s tissues. Irradiation is usually performed in cesium-137 blood irradiators. The blood units do not come in contact with the radioisotope and are not radioactive, but laboratory personnel must comply with radiation safety regulations. Blood units can also be irradiated by cobalt-60 beams in radiation therapy facilities, but in the United States, the FDA regulates this function closely. In recent years, X-ray generators which do not contain radioisotopes have been approved for blood components.

Platelets are stored at room temperature to preserve clotting function, but this increases the risk of bacterial growth in contaminated units, compared to other blood components. Accordingly, bacterial detection has become routine in many countries. Platelet testing must be done after a short period of storage in order to detect growing bacteria. Blood centers can take culture samples from plateletpheresis units and pre-pooled whole-blood–derived platelets before product release. However, culturing is not
feasible from the small volumes of individual whole-blood–derived platelets when the transfusion service is pooling the units just before use. In this situation, bacterial antigen testing is available to transfusion services and is required by accrediting agencies in the United States.

RBC preservation solutions use CDPA—Citrate for anticoagulation, Phosphate as a buffer, Adenine, and 1 to 2 g of dextrose (D-glucose)—to maintain adenosine triphosphate (ATP) levels and RBC membrane integrity. However, despite preservatives, several metabolic changes occur during RBC storage. 2,3-Diphosphoglycerate (2,3-DPG) is depleted in the first 2 weeks, thus shifting the oxygen dissociation curve to the left and making hemoglobin more avid for O₂, although this reverses after transfusion. Furthermore, by the end of the 42-day shelf life of additive-solution RBCs, the pH is 6.5, the plasma K⁺ is 50 mmol/L from RBC leakage and hemolysis, and 15% to 20% of the RBCs are nonviable.²⁷ RBC storage lesions could be involved in observations linking older units to adverse outcomes such as short-term mortality and multiple organ failure.²⁸ Randomized trials are underway to address this important question.

Plasma derivatives are proteins processed from plasma for therapeutic infusions. They include albumin, immunoglobulins (IGs), clotting factors, and other proteins. Derivatives are purified from plasma using physicochemical fractionation methods developed by Edwin J. Cohn in the 1930s. Donors for plasma derivatives are screened and tested just as blood component donors, and the separation techniques provide a degree of purification from microbial pathogens. However, the large numbers of donor units which must be pooled together for production of plasma derivatives introduce the possibility of contamination of an entire lot by a small number of infected donors. Such was the case in the early years of HIV when factor concentrates infected a large proportion of hemophiliacs before the risk was known. Because intact cells are not required, many derivatives can undergo purification and pathogen inactivation methods which are impractical for blood components.

Albumin is produced in large quantities for intravascular volume support and is pasteurized at 60°C for sterility. IGs are given for immune support or for immunomodulation to suppress native antibody production. IGs can also withstand robust pathogen inactivation processes, and in some cases are nanofiltered to remove viruses. To be given intravenously, IGs need extra processing to avoid reaction-provoking protein aggregates. Hyperimmune globulins are fractionated from the plasma of donors with high levels of antibody to specific antigens of interest, such as viruses (HBV, CMV, varicella zoster) or the Rh blood group D antigen (RhIG to prevent anti-D formation in RhD-negative women).

Factor concentrates for patients with congenital deficiencies are made with special techniques to ultrapurify factors VIII, IX, XIII, vWF, fibrinogen, and in some countries, factor XI—while also applying pathogen inactivation methods and viral filtration to remove microbes. However, plasma-free recombinant factors VIII and IX are also available to further allay concern about disease transmission, and recombinant activated factor VII (rFVIIa) is approved for rare VII-deficient patients. Some hemophiliacs develop inhibitory antibodies to the FVIII or factor IX they are missing. These inhibitor patients often need products which “bypass” their missing clotting step. The therapeutic potential of prothrombin complex concentrates (PCCs) containing multiple factors, or FVIIa to bypass secondary hemostasis and generate a thrombin burst is discussed later in this chapter. On the antithrombotic side of hemostasis, antithrombin concentrate is available.

Other plasma proteins which are purified for selected deficiencies are complement C1 esterase inhibitor (for hereditary angioedema) and α₁-antitrypsin.

Techniques have been developed to kill microbial pathogens in blood components. This approach has the major advantage of protection against unrecognized infectious agents. None of these processes have been approved in the United States to date. The methods vary by blood component. In Europe and other countries, solvent-detergent and methylene blue treatments have been approved for plasma, and psoralen and riboflavin treatments for plasma and platelets. Cell-free plasma, either in pools or as individual units, can be treated with membrane-disrupting solvent-detergent treatments for enveloped viruses. Non-enveloped viruses, most prominently hepatitis A and parvovirus, are less susceptible, but NAT can be added to detect these agents.

More robust methods involve agents which damage nucleic acids. When methylene blue, psoralen, or riboflavin is added to the blood bag, they bind to nucleic acids. Then photoactivation is performed with specific wavelengths of light, cross-linking DNA and RNA to prevent microbial function. Not yet approved but under investigation is ultraviolet light treatment of platelets. RBC units are less amenable to photoactivation through the hemoglobin, although riboflavin is being explored. Certain alkylating agents have shown promising microbicidal activity, but modify the RBC membrane enough that subjects have formed antibodies against the treated RBCs.

The adverse effects that limit these techniques include the potential for reduced platelet count increments, alterations in plasma clotting factor levels, and potential toxicity from some of the added agents if they are not sufficiently removed after treatment.

Many years of effort have gone into the search for an oxygen-carrying substitute for RBCs. Oxygen-avid perfluorocarbon chemicals underwent international trials, but 100% O₂ administration was needed and clinical trials were unsuccessful (Castro). Several hemoglobin solutions have been made from pools of human or bovine hemoglobin, or from recombinant hemoglobin, all of which were chemically modified to facilitate extracellular O₂ offloading.²⁹ However, the potentially toxic effects of free hemoglobin, including intravascular binding of the vasodilator nitrous oxide (NO), are not sufficiently understood for regulatory approval. US clinical trials were unsuccessful in demonstrating clinical advantage in pre-hospital trauma management. None of these agents appear to hold much promise for clinical application in the near future. A more recent avenue of exploration is the possibility of “growing” RBCs in cell culture systems. They would be biocompatible and low in pathogen risk, and their RBC antigens could be engineered to some extent to maximize compatibility. The biology of cultivating mature normal RBCs in large scale is under investigation.

Platelets are so complex that it would be difficult to replace their functions fully. However, some early work has emerged on preserved platelets (e.g., freeze-dried or fixed) and on biocompatible platforms bearing hemostatic proteins, such as fibrinogen-coated
albumin beads, which could supply some degree of platelet-like clotting function. Culturing platelets in vitro for therapeutic uses is also being pursued.³⁰

Routine RBC compatibility testing includes ABO and RhD typing, an antibody screen for IgG non-ABO RBC antibodies, and an RBC cross-match. RBCs must be ABO-compatible to avoid intravascular hemolysis, and RhD-negative patients should receive D-negative RBCs to avoid anti-D alloimmunization. Rh, Kell, Kidd, Duffy, and some other non-ABO antibodies can also hemolyze transfused RBCs; 1% of all patients and 5% to 20% of heavily transfused patients have such antibodies. If no antibodies are present, the cross-match can be electronic; that is, RBC units are selected by the laboratory computer to be ABO- and RhD-compatible. But if hemolytic antibodies are detected or are on record, RBC units negative for the incompatible antigen(s) must be found, and a serologic cross-match of patient plasma versus donor RBCs is performed to confirm compatibility. Most hospitals have blood order schedules for their most common surgical procedures, which set forth a recommended number of RBC units to cover 80% to 90% of patients undergoing each procedure.

RBC compatibility testing takes 45 to 60 minutes, and much longer if antibodies are found. Hence, testing in advance of scheduled surgery is desirable. In emergencies, uncross-matched group O RBCs can be given, albeit with the risk of non-ABO antibody incompatibility. Group AB is the universal plasma which avoids transfusing anti-A or -B versus the patient’s RBCs.

Over the past decade, transfusion practices for medical and surgical patients shifted from a liberal strategy to more restrictive management with lower thresholds and careful consideration of the balance between transfusion risks and the physiologic consequences of anemia. The American Society for Anesthesiologists (ASA) Practice Guidelines for Blood Component Therapy 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”.³¹ This leaves little guidance for the majority of anemic patients whose hemoglobin lies between 7 and 10 g/dL. In 1999, the Transfusion Requirements in Critical Care (TRICC) Investigators, Hebert et al., set a new precedent for the tolerance of hemoglobin levels as low as 7 g/dL without increasing morbidity or mortality.³² The literature clearly supports a lower transfusion trigger for young healthy patients as evidence by trauma studies and case discussions of Jehovah’s Witnesses.³³ However, the question of when to transfuse those at risk for tissue hypoxia such as patients with cardiovascular, cerebrovascular, or renal insufficiency remains controversial.

The original TRICC trial included critically ill but euvolemic patients in a large multicenter randomized control trial. The investigators compared the survival of patients transfused to hemoglobin levels greater than 10 g/dL in the liberal group with patients under a restrictive strategy who were not transfused until their hemoglobin fell to less than 7 g/dL. Overall survival for more than 800 patients did not differ between the groups.³² Subsequent subgroup analysis of the TRICC trial and several meta-analysis and systematic reviews over the past 10 years not only confirmed these results, but established a possible morbidity and mortality risk associated with liberal transfusion management for a variety of different patient populations including those less than 55 years old, trauma patients, and those with stable cardiovascular disease.³³,³⁴,³⁵,³⁶

Traditionally, patients with cardiovascular disease and anemia were considered to be at significant risk of tissue ischemia and therefore thought to benefit from higher hemoglobin goals in the perioperative and critical care setting. These recommendations were based on two studies which showed an association between anemia in patients with cardiovascular disease and mortality that improved with transfusion.³⁷,³⁸ Several retrospective studies and systematic reviews have since contradicted these findings by documenting the safety of hematocrits less than 24% during cardiac surgery with cardiopulmonary bypass.³⁴,³⁹ The Transfusion Requirements After Cardiac Surgery (TRACS) trial randomized postcardiac surgery patients to restrictive (hematocrit >24%) versus liberal (hematocrit >30%) transfusion strategies and found no difference in 30-day mortality or severe morbidity. Furthermore, they found that transfusion was an independent risk factor for morbidity and mortality.³⁴,⁴⁰ The most recent update to the Society of Thoracic Surgeons and the Society of Cardiovascular Anesthesiologists Blood Conservation Clinical Practice Guidelines recognizes that all cardiac surgery patients are at risk for tissue hypoxia and transfusion. These guidelines report that tissue oxygenation does not improve with transfusion for hemoglobin levels greater than 10 g/dL; and they state that “Transfusion is reasonable in most postoperative patients whose hemoglobin is less than 7 g/dL.” For patients with hemoglobin levels between 7 and 10 g/dL the two societies recommend transfusion in patients with “critical noncardiac end-organ ischemia,” active blood loss, or clinical indication of tissue hypoxia (low mixed venous oxygen saturation or electrocardiographic or echocardiographic evidence of myocardial ischemia).⁴¹

Patients with cerebrovascular disease or acute neurologic illness such as ischemic stroke, subarachnoid hemorrhage, and traumatic brain injury are at significant risk for secondary injury from tissue hypoxia. There are several studies and reviews devoted to the question of optimal brain tissue oxygen delivery during acute neurologic illness. Acute and chronic anemia initiates physiologic compensatory mechanisms discussed below as well as neuroprotective strategies for tolerating decreases in cerebral oxygen delivery; however, the hemoglobin level at which anemia induces detrimental patient outcomes remains unclear.⁴² Evidence indicates that a hemoglobin level less than 9 g/dL is independently predictive of poor outcome especially in patients with cerebrovascular injury.⁴³,⁴⁴ RBC transfusion perioperatively or during critical illness is also independently associated with morbidity and mortality.⁴³,⁴⁵,⁴⁶ This leaves the clinician with little guidance to decide when and who should be transfused. Clearly transfused RBCs do not function as well as endogenous erythrocytes and it remains unclear if transfused blood improves overall outcomes.

Numerous recommendations provide guidance for the transfusion management of thrombocytopenia and acquired or inherited platelet disorders.³¹,⁵²,⁵³,⁵⁵,⁵⁶,⁵⁷,⁵⁸ The indications for platelet transfusion depend on both quantitative and qualitative measures of platelet activity as well as the clinical setting. Table 16-4 outlines typical platelet thresholds for transfusion based on the clinical situation and patient history. For stable patients with severe thrombocytopenia, transfusion can be held until counts fall below 10,000 to 20,000/μL as long as they have no signs or symptoms of bleeding.⁵³,⁵⁵,⁵⁶,⁵⁸ Prophylactic platelet transfusion is necessary for patient with severe thrombocytopenia <50,000/μL about to undergo major surgery or invasive procedures such as lumbar puncture, liver biopsy, neuraxial anesthesia, central venous catheterization, or endoscopy with biopsy.⁵⁶,⁵⁷ In preparation for surgery on the eye or the central nervous system, the platelet count should be raised to >100,000/μL.⁵⁶,⁵⁷

Transfusion is not necessary for platelet counts >100,000/μL in clinically stable patients without suspicion of platelet dysfunction. Most patients with counts <50,000/μL and clinical bleeding require therapeutic transfusion.³¹,⁵⁵,⁵⁶ There are several relative indications for transfusion of platelet counts between 50,000 and 100,000/μL. In particular, patients having surgery on vital organs such as the eye or central nervous system, or patients with multiple traumatic injuries, benefit from higher transfusion goals.⁵³,⁵⁵,⁵⁶ Furthermore, patients undergoing massive transfusion or hemorrhage with estimated blood loss of more than two blood volumes and ongoing bleeding should have a transfusion threshold of at least 75,000/μL to ensure the level does not fall below 50,000/μL.⁵³,⁵⁴

Platelet transfusion should not be guided by platelet counts alone, but also by the clinical suspicion of platelet dysfunction which can be inherited or acquired. Qualitative dysfunction is often associated with systemic diseases such as uremia, liver failure, and disseminated intravascular coagulopathy. It also occurs after cardiopulmonary bypass, extracorporeal circulation such as dialysis or plasmaphoresis, and as a result of medication side effects⁵⁹ (Table 16-5). Regardless of the platelet count, if bleeding is out of proportion to the level of thrombocytopenia, qualitative deficiency should be suspected and treated.⁵³,⁵⁶

The average dose of platelets is one concentrate from apheresis donation or pools of five to eight concentrates from whole-blood or buffy-coat collections. These “units” generally contain about 3 × 10¹¹ to 4 × 10¹¹ platelets each.⁵³,⁵⁵,⁵⁶ Once platelets are given, a posttransfusion response should be followed to guide further therapy and to rule out ongoing consumptive pathophysiology. Refractoriness at 20 to 24 hours is generally associated with older platelets or increased consumption secondary to fever, infection, bleeding, or medications.⁵⁶ Adjunctive therapy for platelet dysfunction can be established with antifibrinolytics, DDAVP (1-deamino-8-D-arginine vasopressin, desmopressin), PCCs, or rFVIIa as discussed later in this chapter.⁵⁵,⁵⁸

FFP is a blood component that contains all the factors involved in hemostasis. (As used here, “FFP” also includes plasma frozen within 24 hours.) The preparation and storage method discussed previously in this chapter involves either separation from whole
blood or apheresis; both maintain normal levels of stable factors and at least 70% normal levels of labile factors such as FVIII and FV. Prior to use, each unit must be thawed at 30 to 37°C, but can be stored at 4 ± 2°C for up to 24 hours as thawed FFP or 5 days as thawed plasma.⁵³,⁵⁶ FFP must be ABO-compatible with the recipient to avoid transfusion of donor anti-A and anti-B antibodies that may lead to hemolysis.⁶⁰

FFP is indicated for the treatment of coagulopathy secondary to congenital or acquired factor deficiencies. The specific indications for FFP outlined in Table 16-6 are based on a compilation of guidelines from the ASA, the British Committee for Standards in Haematology, and the Italian Society of Transfusion Medicine and Immunohaematology (SIMTI).³¹,⁵³,⁵⁶,⁶⁰,⁶¹ The initial therapeutic dose of FFP averages 10 to 15 mL/kg in an attempt to obtain at least 30% factor activity. Repeat dosing should follow the results of serial diagnostic coagulation tests such as the PT and aPTT.³¹,⁵³,⁵⁶,⁶⁰,⁶¹ Although the guidelines for FFP continue to recommend prophylactic transfusions in patients at risk for bleeding, there is no evidence to support the efficacy of FFP reducing red cell transfusion, morbidity, or mortality, especially when given to patients with mild derangements in PT or aPTT and no clinical signs of bleeding.⁶¹,⁶² The transfusion of plasma carries risks of several transfusion reactions, most notably TRALI, allergic reactions, and cardiopulmonary overload. The burden of adverse reactions in conjunction with limited prophylactic benefits and high frequency of inappropriate use makes FFP arguably the riskiest blood component transfused.⁶⁰,⁶¹,⁶³,⁶⁴

Over the past decade there has been growing debate over formula-driven resuscitation protocols for trauma patients with major bleeding. Traditionally trauma patients were resuscitated first with fluid and red cell units only to receive FFP based on the results of coagulation tests.⁵⁴ This method perpetuates dilutional coagulopathy resulting in prolonged microvascular bleeding.⁶⁵ More recent studies show improved outcomes with higher ratios of red cell units to FFP (more than 3:2).⁶⁵,⁶⁶ However, these findings are mostly retrospective and likely confounded by injury severity and the potential “survivor bias” that occurs when patients die of massive hemorrhage before FFP can be thawed and transfused.⁶⁶,⁶⁷ The optimal management strategy for immediate resuscitation of trauma patients and those with acute massive hemorrhage is still under investigation. However, it is clear that coagulopathy driven by ongoing bleeding requires rapid response, and the best treatment strategies involve active management protocols to maintain intravascular volume, avoid platelet dysfunction, and facilitate hemoglobin and factor replacement.

Cryoprecipitate is created by a controlled thaw of frozen plasma which allows for precipitation of large molecules. It is then centrifuged with the final product containing fibrinogen, fibronectin, vWF, FVIII, and factor XIII.⁶⁸ One unit of plasma typically generates 10 to 20 mL of cryoprecipitate. These small concentrates are combined for a single adult dose of five bags and frozen at −20°C.⁶⁸,⁶⁹ Current use of cryoprecipitate is limited to a few countries (United States and Canada) and generally only indicated for hypofibrinogenemia (Table 16-7). Compared to FFP, cryoprecipitate contains more fibrinogen per volume (15 vs. 2.5 g/L in FFP)⁶⁰ and therefore functions as a low-volume alternative for fibrinogen replacement especially during acquired hypofibrinogenemia states such as disseminated intravascular coagulopathy and massive hemorrhage. It carries similar transfusion risks as FFP and consequently has been replaced in several countries including Europe with fibrinogen concentrates, currently unavailable in the United States.⁶⁹

Fibrinogen is primarily responsible for clot stabilization, but it also assists with platelet activation and aggregation and, at adequate concentrations, may compensate for low thrombin states.⁶⁹ Traditionally, the threshold for fibrinogen replacement has been levels less than 80 to 100 mg/dL.³¹,⁶⁰,⁶⁸ However, a few studies of acquired hypofibrinogenemia with massive bleeding document the need for higher levels (150 to 200 mg/dL) to optimize clot stabilization.⁷⁰,⁷¹ A single dose of five pooled bags typically raises the fibrinogen concentration by 50 mg/dL.⁶⁸

Since the introduction of NAT for the major transfusion-transmittable viral infections (HCV in 2000, HIV in 2003, and HBV in 2006) the residual risk of infection from blood product transfusion has decreased substantially.⁷² Prior to the institution of NAT, the blood supply was simply tested for the presence of viral antibodies leaving a long window period when the blood is infectious, but there was not sufficient time for the donor to mount a large enough antibody response to be detected by conventional testing. NAT increased the sensitivity of donor screening by testing for the presence of specific viral DNA. This significantly shortens the window period between when a donor gets infected and when the viral load is detectable. The residual risk of transfusion-transmitted infections depends mostly on the relative length of this window period determined by the reproductive rate of each virus and the prevalence of the disease. There is an additional risk of false negatives or mistaken release of quarantined blood products; however, these events account for less than 0.5% of the residual risk of transfusion-transmitted viral infection.⁷³

The true risk of transfusion-transmittable infections is difficult if not impossible to accurately quantify given the variability of donor response to inoculation and the immune state of the recipient. However, the residual risk can be mathematically modeled from the prevalence of infection in donors and the known window period of each viral illness. The estimated residual risk of major viral infection and the viral-specific window periods are shown in Table 16-8 based on the reported incidence of infection in donors for the American Red Cross Blood Services.⁷³ These residual risks are likely overestimates since not all transfusions of known infected blood products result in recipient infection.⁷⁴,⁷⁵ The mathematical modeling of residual risk assumes 100% infectivity with even one infectious unit of viral particles per blood bag. It also presumes that the collection and storage process is harmless to viral reproduction. Furthermore, viral infectivity will vary depending on the acute phase of infection when an immunocompetent recipient could resist transmission, or the chronic phase of infection when antibodies are present and reduce infectivity.⁷⁵