Chapter 83. Blood and Blood Component Therapy

1. ABO compatibility remains the major safety consideration for blood transfusion. Clerical errors of patient identification and sample labeling remain the primary cause of mistransfusion of ABO-incompatible units.

2. Prestorage leukoreduction of cellular blood products has greatly reduced, but not eliminated, the incidence of febrile reactions, transfusion-related immunosuppression, and cytomegalovirus transmission, and has led to improved outcomes in surgical patients compared to the use of nonleukoreduced products.

3. Redirecting blood from female donors away from transfusable plasma products appears to have reduced the incidence of TRALI mediated by leucocyte reactive antibodies. However, some plasma from female donors remains in platelet and red blood cell products.

4. International travel and associated exposure to pathogens not addressed by existing screening mechanisms will increasingly limit suitable blood donors. These and other pressures on the blood supply make imperative strategies all the more important for blood conservation and promotion of bloodless surgical techniques.

Practical, safe transfusion derives from centuries of experimentation and discovery. Although the first known animal transfusion experiments took place in both England and France in the 17th century, early efforts to translate the technique to humans failed with such spectacular flair that the learned societies on both sides of the English Channel flatly banned the practice.

The first "modern" transfusion is attributed to John Syng Physick in Philadelphia in 1795. English physician James Blundell claimed success for 5 of 10 transfused patients during his professional career (1820-1840).

The 19th century saw the rapid expansion of bacteriology and a growing understanding of antisera. In 1900 Karl Landsteiner significantly advanced the cause of blood compatibility with his landmark discovery of the ABO blood groups, with O being derived from the German "ohne" or without, for which he received the 1930 Nobel Prize in Medicine and Physiology. Four decades later, in collaboration with Alex Weiner, Philip Levine, and R.E. Stetson, Landsteiner played an instrumental role in the recognition of Rh(D), a major cause of hemolytic disease of the newborn (HDN) and non–ABO-related hemolytic transfusion reactions.

Prior to effective schemes for anticoagulation, Alexis Carrel explored vascular anastomosis as a possible strategy for moving blood from donor to recipient. Although eclipsed by other approaches, this pioneering technique formed the basis of contemporary vascular anastomoses in solid-organ transplantation and won Carrel the 1912 Nobel Prize.

The use of citrate to anticoagulate stored blood by means of calcium chelation was a landmark, as heparin was unsuitable to transfuse into bleeding patients and citrate's rapid metabolic elimination in vivo proved ideal for purposes of transfusion. Its adoption paved the way for transition from direct to indirect transfusion. By 1916 Rous and Turner introduced glucose in addition to sodium citrate, extending ex vivo storage by several days.

The first blood depots were established during the First World War by the British and were stocked prior to major campaigns. The Soviets established the first permanent blood bank in a Leningrad hospital in 1932, and 5 years later Bernard Fantus established the first US blood bank at the Cook County Hospital in Chicago.

Although clinical efforts had focused largely on transfusion of whole blood, Professor Edwin Cohn at Harvard explored fractionation of plasma proteins using cold ethanol treatment and centrifugation. By 1940 these techniques yielded concentrated fractions of albumin, immunoglobulins, and fibrinogen. With the outbreak of war in Europe, the US War Department and American Red Cross launched Plasma for Britain, under the direction of Dr Charles Drew.

Despite highly successful initial resuscitation, casualties with significant blood loss treated exclusively with plasma developed "air hunger" from anemia. Improved logistics enabled a shift to the use of whole blood. With this transition, volume tolerance emerged as the next constraint to adequate replacement of oxygen-carrying capacity. The large-scale availability of red blood cell concentrates (and more general exploitation of blood component therapy) awaited development of the plastic, integrated blood-collection systems of the 1950s and 1960s.

Success was tempered by the identification of transfusion-borne infections. "Serum" hepatitis followed approximately 10% of transfusions. Worse, in some communities posttransfusion hepatitis reached rates of 30%, an observation that was explained by reliance on monetarily compensated donors. In parallel with these epidemiologic discoveries, advances in virology and radioimmunochemistry brought to market by 1969 limited supplies of tests for the detection of hepatitis B surface antigen (HBsAg, the "Australia Antigen"). With transition in the United States to an all-volunteer blood supply, rates of hepatitis B virus (HBV) transmission fell dramatically, and with full deployment of HBsAg testing in 1972, rates of transfusion-transmitted hepatitis B fell to between 0.3% and 0.9%. Although remarkably high by today's standards, this represented risk reduction by an order of magnitude.

Following the emergence of human immunodeficiency virus (HIV)/acquired immunodeficiency syndrome (AIDS), an epidemic that transformed public perceptions of blood safety occurred. Lessons learned for dealing with hepatitis led to rapid progress toward securing the safety of the blood supply. The answer to a patient's question, "Will this transfusion give me AIDS?" is, "No."

Although infectious risks of blood transfusion have diminished enormously in developed nations, high rates of endemic disease and limited testing resources in developing countries paint a variegated global picture of blood product safety. Today's tools include effective donor screening and testing. Tomorrow holds the promise of pathogen inactivation and improved controls of bedside transfusion practice as further steps on the long path toward safe transfusion.

Whole Blood

Despite some clinicians' desires for fresh whole blood, especially for use in pediatric cardiac anesthesiology, the optimal use for a limited resource demands the use of red cells for oxygen-carrying capacity, plasma for coagulation proteins, and platelets for thrombocytopenia or platelet dysfunction. The component therapy approach maximizes the number of recipients benefiting from transfusion and optimally preserves the function of essential blood elements. Colloid replacement can be provided by synthetic colloids, such as gelatins and starches, or albumin products fractionated from plasma. The time required to perform federally mandated testing makes fresh whole blood practically unavailable to most clinicians; consequently, discussion is primarily focused on blood components.

A unit of whole blood is collected in citrate-phosphate-dextrose-adenine 1 (CPDA-1) anticoagulant, giving it a shelf life of 35 days and a volume of approximately 510 mL (450 mL of blood plus 63 mL of CPDA-1). Within 24 hours of collection, the platelets are dysfunctional1 and several plasma coagulation factors are below optimal levels,2 emphasizing the importance of the component products despite the disadvantage of increased donor exposure.

Red cell concentrates, or packed red blood cells (PRBCs), are obtained from CPDA-1–anticoagulated whole blood after centrifugation of most of the plasma and platelets. At most blood centers, the red cells are then mixed with 100 mL of an additive nutrient containing an additional dextrose and adenine (NutriCell or AS-3) or dextrose, adenine, and mannitol (Adsol or AS-1) solution that extends the storage period to 42 days3 and results in flow properties similar to those of whole blood. PRBCs from CPDA-1–anticoagulated blood are stored for up to 35 days. Loss of viability during storage is caused by cellular adenosine triphosphate (ATP) depletion, which leads to loss of membrane phospholipids, increased rigidity, and reduced life span, with current standards requiring 75% of transfused cells to survive in the circulation for 24 hours.2 At 35 days, 70% of cells survive in CPDA-1 versus 88% in Adsol, with a pH of 6.7 in both solutions; intracellular ATP is 45% baseline in CPDA-1 versus 75% in Adsol, whereas 2,3-diphosphoglycerate (2,3-DPG) levels are at 10% baseline in both. In vivo, 2,3-DPG levels rapidly normalize over 24 hours and recover to more than 50% within 1 or 2 hours.4 PRBCs are the product of choice for the correction of an isolated defect in oxygen-carrying capacity, as in chronic anemia, with the advantage of lower volume load in this setting. During acute perioperative blood loss, however, attention must be paid to volume status as well as restoring oxygen-carrying capacity using crystalloid or colloid volume expanders.

Leukocyte-Reduced Red Cells

Leukocyte-reduced red cells (LRRCs) can be prepared by a variety of methods, with varying efficiency of white cell removal. The minimum standard, established by the American Association of Blood Banks (AABB), is a leukocyte number in the final component of less than 5 × 106.8 Early techniques involved centrifugation and washing with saline, repeatedly removing the buffy coat. Next the "spin-cool-filter" method was introduced, using 1-week-old red cells that were centrifuged and then cooled for 4 hours to enhance microaggregate formation, then were passed through a microaggregate filter. Currently, the most widely used method of leukoreduction is filtration, which can be performed either in the laboratory or at the bedside. The various filters on the market result in greater than 99% leukocyte reduction while depleting less than 10% of the red cells. Most recently, blood bags with in-line filters that allow prestorage leukoreduction have become available. Simple removal of the buffy coat from red cells (a routine procedure during component preparation in Europe) may provide a sufficient degree of leukoreduction.5 Because component preparation in the United States does not include buffy coat removal and because soluble white cell fragments may be capable of mediating immunosuppression, a prestorage leukoreduction method might be required to abrogate this adverse effect in transfusion recipients in the United States. Introduction of leukoreduction in the United States began around 1998. Confirmation with individual blood banks is necessary to determine their policies, as there are often regional differences.

The major indication for the use of LRRCs has been prevention of the nonhemolytic febrile transfusion reaction, which is the most common adverse effect of transfusion, particularly in multiply transfused patients or multiparous females. It is believed to be a result of preexisting antihuman leukocyte antigen (anti-HLA) antibodies. The use of LRRCs reduces the incidence of such reactions.6 Cytokines may cause these reactions, suggesting that prestorage leukodepletion may be more helpful.

A second important indication for LRRCs is the prevention of alloimmunization to HLA antigens that can adversely affect posttransfusion platelet increments.7 Platelet components also require leukodepletion. The current AABB standards mandate less 5 × 106 leukocytes for this indication, possibly with third-generation leukoreduction filters for both red cells and platelets.8 A large, multicenter study (TRAP, Trial to Reduce Alloimmunization to Platelets) confirmed that the use of leukoreduction filters significantly decreased but did not eliminate alloimmunization.9

Washed Red Cells

Red cells are washed using isotonic saline solutions with some degree of red cell loss with each wash cycle. This open system requires the resulting product to be transfused within 24 hours. The primary aim of washing is to remove plasma proteins (with some leukocytes and platelets also removed) to prevent severe allergic transfusion reactions mediated by recipient antibodies (most likely IgE) to donor plasma proteins or preformed antibodies to IgA in donor plasma, which can cause anaphylaxis in IgA-deficient patients.10 Washed cells may also be indicated in paroxysmal nocturnal hemoglobinuria.

Frozen Red Cells

Red cells can be frozen (with glycerol used as a cryoprotective agent) and stored in liquid nitrogen or mechanical freezers.11 Thawing and washing away the glycerol using a series of progressively less hypertonic saline solutions allows glycerol to diffuse gradually from the cells to prevent hemolysis before transfusion. This process removes most of the plasma, and any debris and red cells can be frozen for more than 10 years with good viability.12 As with washed cells, frozen red cells need to be transfused within 24 hours; the posttransfusion survival rate is 85% to 90%.11 Freezing, ideally using fresh blood, maintains ATP and 2,3-DPG levels, but solutions containing pyruvate, glucose, phosphate, and adenine can restore older blood.12 The major use of frozen cells is to store units of a rare phenotype. In addition, the military can transport large volumes of frozen units.

Indications for RBC Transfusion

A unit of PRBCs should increase the hemoglobin (Hgb) level by 1 g/dL or the hematocrit level by approximately 3%. However, continued blood loss causes a suboptimal response to transfusion. More chronically, hemolysis caused by either immune red cell damage or mechanical trauma shortens the survival of transfused cells, and hypersplenism can lead to sequestration and increased destruction of red cells. Also, transfusion suppresses erythropoiesis, thus sustained results of transfusion may be less than expected.

Chronic Anemia

Generally, the signs and symptoms of anemia develop at an Hgb level of less than 7 to 8 g/dL. With gradual onset, the body's compensatory mechanisms for maintaining oxygen delivery to the tissues are effective. Cardiac output and intracellular 2,3-DPG are increased, thus oxygen unloads at a lower pO2 of the tissues. When chronic anemia is a result of red cell destruction, the healthy bone marrow can respond by increasing production by up to 6-fold, provided sufficient dietary iron, folate, and vitamin B12 is ingested. The decision to transfuse depends on how a given individual manifests the signs and symptoms of anemia, determined by underlying health status, cardiorespiratory reserve, and activity level. In the perioperative setting, it is important to cross-match sufficient units to compensate for the lower initial Hgb, particularly in a multiply transfused patient with alloantibodies that may be difficult to match.

Perioperative Period

Much of the dogma surrounding perioperative transfusion has few supporting data. One example is the use of 10 g/dL of Hgb as the gold standard for the red cell transfusion trigger during the perioperative period.13 A rule of thumb is that, assuming no preoperative anemia, blood losses of 5% to 10% of total blood volume require minimal replacement therapy; losses of up to 20% can be replaced exclusively with volume expansion, whereas losses of greater than 25% generally require red cell transfusion to restore oxygen-carrying capacity, along with volume expansion to restore intravascular volume and maintain perfusion. The availability of hemoglobin measurement, acid-base status, and central venous and mixed venous oximetry, combined with tools to assess intravascular volume in most hospital environments, supersedes such rules, allowing more individually tailored, closely monitored transfusion therapy designed to provide sufficient tissue oxygenation. This is still vague and subjective, therefore when to transfuse remains a clinical decision. The American Society of Anesthesiologists provides extensive guidelines, available at http://www.asahq.org. Societies, such as the Society for the Advancement of Blood Management (http://www.sabm.org), champion blood conservation techniques and appropriate safe but lower transfusion triggers aimed at minimizing allogeneic blood product usage in the perioperative period. The National Institutes of Health suggest that most surgical patients who are not actively bleeding do not need transfusion unless the Hgb level decreases to less than 7 g/dL, and such levels of Hgb may not interfere with wound healing or automatically make general anesthesia a risk.

Nonetheless, questions concerning an objective measurement to determine a safe transfusion trigger still arise. Animal models of acute normovolemic anemia can be helpful and describe the whole body oxygen extraction ratio (OER) as an indicator of when to transfuse.14 These studies make the seemingly valid assumption that the heart is the major organ at risk. With progressive hemodilution, healthy animals with normal coronary trees were able to maintain normal levels of oxygen consumption (200-250 mL/min) through a moderate increase in cardiac output, an increase in coronary blood flow, and a linear increase in the OER up to a ratio of 50%. As the hematocrit decreased to less than 10%, however, oxygen consumption began to decrease, and the animals were no longer able to increase the OER. An OER of 50% was the critical point at which the myocardium converted from aerobic to anaerobic metabolism, reflected in net lactate production. At that point, metabolic acidosis developed, resulting in hemodynamic instability. Dogs with critical coronary stenosis converted to anaerobic metabolism and went into congestive heart failure at an OER of greater than 50%, but that an OER of greater than 50% occurred at a hematocrit of 17.0% in the dogs with critical stenosis, compared with a hematocrit of 8.6% in the healthy group, is consistent with the 10% above.15,16 Dog hearts with normal coronary arteries developed subendocardial ischemia then cardiac failure at a hematocrit less than 10%; hypertrophied hearts, at higher risk of subendocardial ischemia, may fail at higher hematocrits, although this was not studied.15 OER can be calculated whenever a pulmonary artery catheter is being used, and therefore mixed venous oximetry can be performed and thus used to help assess transfusion need. An OER of 50% can be used as a red cell transfusion trigger because it appears to be a valid indicator of marginal myocardial oxygen reserve in both healthy people and individuals with coronary artery disease if it is reasonable to directly extrapolate the above animal data to humans. Given the complexity that this adds to clinical practice, most clinicians follow published guidelines based on Hgb levels.

The controversies surrounding transfusion triggers in neonates are beyond the scope of this chapter and are complicated by uncertainties concerning the diagnosis of symptomatic anemia in this group. Similarly, there are fewer data addressing transfusion in pediatric patients compared with adults. Of interest, the humoral and cellular immune systems of the neonate are immature, particularly with prematurity. This introduces a small but real risk of transfusion-induced graft versus host disease or thrombocytopenia after transfusion of leukocytes (present to some degree in PRBCs or platelet transfusions) in premature infants or in a fetus undergoing intrauterine transfusion.17 Irradiated cells are indicated in these settings. Another risk in premature infants is the development of clinical cytomegalovirus (CMV) infection in infants of CMV-seronegative mothers.18 CMV-seronegative blood is now routinely used in these neonates. The only setting in which there may be some data to support the use of CMV-seronegative units is in severe combined immune deficiency infants undergoing bone marrow transplantation. Otherwise, contemporary leukoreduced products are considered "CMV safe," equivalent to seronegative products. Leukoreduction by filtration reduces but does not eliminate the risk of these complications.19

Transfusion safety relies on the avoidance of transfusion reactions, as classified in Table 83-1, that have multiple etiologies. Immune-mediated reactions are a result of transfusion of an incompatible blood product, as described below. The single most important contribution to transfusion safety is uncompromising attention to all details of patient, sample, and blood product identification.

Red Blood Cell Compatibility

Compatibility relies on the recipient not recognizing the allogeneic blood transfusion as foreign. Sensitization to alloantigens, with formation of the alloantibodies that determine incompatibility, can occur after transfusion, transplantation, or pregnancy. The most clinically important antibodies, or isohemagglutinins, react with antigens of the ABO system; are naturally occurring, complement-fixing antibodies that require no prior sensitization; and typically result in immediate intravascular hemolysis of transfused erythrocytes and a severe hemolytic transfusion reaction (HTR). The chance of ABO incompatibility with an unmatched transfusion is 1 in 3, leading to a life-threatening HTR with approximately 10% mortality.20 Anti-A and anti-B antibodies are classically associated with severe HTRs, but reactions to anti-Rh (D, C, E, c, e, f, or ce), Kell, Duffy, or Kidd system antibodies are actually more common.21 Table 83-2 details the ABO compatibility of blood products.

Alloimmunization

The likelihood of sensitization to other antigens depends on both the variable immunogenicity of the particular red cell antigen and the immune response of the host; HTRs are 3 times more common in females.21 Such alloimmunization is best understood for the Rhesus (Rh) D antigen, studied extensively because of the important role it plays in hemolytic disease of the newborn. It is a highly immunogenic and clinically important antigen associated with severe HTRs. Up to 80% of Rh(D)-negative individuals, and up to 95% after massive transfusion, produce anti-D antibodies after exposure.22 Those who generate anti-D after an exposure are more likely to develop other red cell alloantibodies, whereas those not generating anti-D rarely develop other alloantibodies.23 Other antigen systems are less immunogenic; Kell (K) antigens stimulate anti-K in 10% of cases, followed in order of magnitude by Rh c and E, Duffy (Fya), and Kidd (Jka). Antibodies recognizing antigens from the Kell, Duffy, Kidd, Lutheran, and MNSU systems can cause HTRs, typically delayed, although in many patients they only result in reduced red cell survival.

In terms of alloantibodies detected by blood banks, the most commonly recognized antigens are in the ABO blood group system followed by anti-D, anti-K, anti-E, anti-K, anti-c, anti-Fya, anti-C, anti-Jka, anti-S, and anti-Jkb. The incidence of others is less than 1%.23 Studies of alloimmunization after transfusion have focused on chronically transfused patients with hemoglobinopathies (such as thalassemia or sickle cell disease), and an approximately 1% risk of alloimmunization has been attributed to every unit transfused.24 The incidence of red blood cell (RBC) alloimmunization is up to 50% in adults with sickle cell disease,25 increasing the likelihood of delayed HTRs, which can present as sickle crises.26

HLA Alloimmunization

Alloimmunization to antigens in the human leukocyte antigen (HLA) system occurs in multiply transfused or multiparous individuals and can be detected in 30% to 70% of patients given nonleukoreduced RBC or platelet transfusions.27 The almost ubiquitous adoption of leukoreduction techniques for RBCs and platelets during the 1990s can reduce the incidence of HLA alloimmunization, which can cause febrile nonhemolytic transfusion reactions, platelet transfusion refractoriness, and increased risk of subsequent transplant rejection.27 Alloantibodies to the human platelet antigen systems (HPA-1 to HPA-5) cause neonatal alloimmune thrombocytopenia and posttransfusion purpura; they may also lead to platelet transfusion refractoriness, but are clinically far less important than HLA alloantibodies.

Hemolytic Transfusion Reactions

Immune-mediated HTRs result from recipient alloantibodies reacting to transfused RBC surface alloantigens. HTRs can be classified as acute (within hours of transfusion), delayed (within days or weeks of transfusion), intravascular, and extravascular.

Acute HTRs

Acute HTRs are a result of existing, circulating alloantibodies to alloantigens on transfused RBCs. If the alloantibody fixes complement (such as anti-A or anti-B), acute, intravascular hemolysis occurs, which is a medical emergency. Acute HTRs have an estimated frequency of 1 per 25000 RBC units.28 The mortality of 1 in 600000 RBC units29 is primarily caused by ABO incompatibility.30

Signs and symptoms of an acute HTR can be nonspecific and confounded by the patient's clinical condition; consequently, a high index of suspicion and meticulous identification of patients and blood products is essential. Although chills; infusion site, abdominal, chest, or back pain; nausea; anxiety; and dyspnea may be reported by a conscious patient, fever, hypo- or hypertension, hemoglobinuria, anuria, shock, or coagulopathic bleeding may be the first sign of an acute HTR in anesthetized patients. The etiology of profound shock is the activation of the complement cascade with bradykinin production from factor XII activation, C3a- and C5a-induced histamine and serotonin release from mast cells leading to increased vascular permeability, bronchial and intestinal smooth muscle contraction and neutrophil degranulation, and C5a-mediated capillary vasodilatation. Numerous cytokines (tumor necrosis factor, interleukin [IL]-1, IL-6, IL-8) also play a role.31 Disseminated intravascular coagulation (DIC) results from complement-mediated activation of factor XII and extrinsic pathway activation by stromal thromboplastins from lysed erythrocytes.

Free hemoglobin does not appear to have direct toxic effects,32 and shock-induced renal failure and DIC convey most of the morbidity and mortality.33 Therefore, prompt supportive therapy with fluids, inotropes, and vasopressors as indicated is crucial, and corticosteroids may be considered in the setting of resistant vasoplegic shock. DIC can be managed with hemostatic blood products as required; the efficacy of heparin or activated protein C in this setting is not established.34

Immediate notification of the blood bank is essential, as another patient is likely to erroneously receive the unit intended for the patient experiencing the acute HTR. An anticoagulated ethylenediaminetetraacetic acid (EDTA) and a serum specimen, as well as a posttransfusion urine sample, should be sent with all suspected blood units and infusion tubing. Table 83-3 details laboratory testing for suspected HTRs.

Delayed HTRs

Delayed HTRs require a secondary or anamnestic antibody response to a transfused RBC alloantigen in a recipient previously sensitized by transfusion or pregnancy. Detectable antibody levels are not present at the time of pretransfusion serologic testing or transfusion, but appear within days of the transfusion, which is why transfused patients need to have serologic testing repeated after 4 to 5 days.35

The frequency of delayed HTRs is estimated to be 1 per 1500 units. Although fatalities have been associated with delayed HTRs, they were not believed to be causative.30 Alloantibodies are typically against antigens in the Rhesus, MNS, Kell, Kid, or Duffy systems, but rarer antigens may also be involved.

Opsonized RBCs are cleared by the reticuloendothelial system, resulting in a slower and less severe extravascular hemolysis. A history of fever, anemia, and recent blood transfusion may be the only clue to a delayed HTR, and many pass undetected.21 Rarely, a delayed HTR causes intravascular hemolysis with the clinical picture described for acute HTRs. Erythrocyte destruction is maximal 4 to 13 days posttransfusion; a positive direct antiglobulin test is found after 2 to 3 days, followed by spherocytes in peripheral blood (3-4 days) and antibodies detectable by indirect antiglobulin test, anemia, jaundice, and possible hemoglobinuria (5-7 days).

Pseudo HTRs

Poor transfusion practices, such as transfusion of aged, overheated, or frozen cells; osmotic hemolysis by hypotonic solutions; mechanical hemolysis from inappropriate administration equipment; hemolysis from bacterial or parasitic contamination; or coincidental presentation of congenital hemolytic conditions (eg, sickle trait or glucose-6-phosphate dehydrogenase deficiency) presenting at the time of transfusion (eg, surgery) all lead to a nonimmune-mediated or "pseudo" HTR. Nonimmune hemolysis can be induced by mishandling of blood, overheating, and mixing with nonisotonic solutions.

Plasma is separated from PRBCs through centrifugation of whole blood at the time of collection, or collected by apheresis as a single product or as a byproduct of platelet or red blood cell apheresis. Plasma can be further processed into its derivatives through the cold ethanol fractionation method of Cohn. This chapter discusses features and uses of fresh-frozen plasma (FFP), cryoprecipitate, and other products derived from pooled plasma.

Fresh-Frozen Plasma

Plasma is the fluid compartment of blood and consists of 90% water, 7% protein and colloids, and 2% to 3% nutrients, crystalloids, hormones, and vitamins. The protein fraction also contains the soluble clotting factors. Plasma frozen at −35°C (−22°F) or colder within 6 hours of donation is labeled FFP. This product may be stored up to 1 year before use, at which time it is thawed over 20 to 30 minutes. The activities of the labile coagulation factors (V and VIII) decrease after thawing but remain adequate for at least 24 hours. Plasma that is not immediately frozen as FFP becomes either "liquid plasma" (stored at 33.8°F-42.8°F [1°C-6°C]) or "source plasma" (stored at −35°C [−22°F] or colder) and is used for the preparation of plasma derivatives: albumin, clotting-factor concentrates, and immunoglobulin preparations. When FFP is thawed at 39.2°F (4°C), a precipitate separated by centrifugation forms, resulting in cryoprecipitate and cryo-poor supernatant.

Solvent detergent (SD) plasma is an alternative to FFP. SD plasma is a pooled product (2500 donor units per pool) and subjected to a solvent detergent treatment process combining the organic solvent tri(n-butyl)phosphate with a nonionic detergent, Triton X. This inactivates lipid-enveloped viruses, including HIV types 1 and 2, HBV, hepatitis C virus (HCV), human T-cell lymphotropic virus (HTLV) types 1 and 2, hepatitis G virus (HGV), vesicular stomatitis virus, and Sendai virus, while preserving the structural and functional integrity of most plasma proteins. However, it does not inactivate nonenveloped viruses, such as parvovirus B19 and hepatitis A virus (HAV), or prions, the causative agents of various rare encephalopathies including Creutzfeldt-Jakob disease. The increased prevalence of parvovirus B19 and HAV makes transmission by transfusion of pooled blood components a potential concern, although pooled SD plasma also contains significant amounts of neutralizing antibody to both parvovirus B19 and HAV. It is unclear what the real value and risk of SD plasma is compared with FFP, but, at present, the cost of SD plasma is much more than FFP.

Indications for FFP

Audits of transfusion practices have consistently demonstrated that FFP is commonly used inappropriately.36 Such inappropriate use includes reconstituting "whole blood" by coadministration with RBCs, volume expansion, and as a source of nutrients. Acceptable indications are discussed below.

Liver Disease and Transplantations

Patients with severe liver disease have low levels of the vitamin K–dependent clotting factors (II, VII, IX, and X) in addition to thrombocytopenia associated with hypersplenism and a hyperfibrinolytic state. These patients develop a prolonged prothrombin time (PT) and partial thromboplastin time (PTT). In addition, the thrombin time (TT) may be prolonged, fibrin split products may be elevated, and in later stages the fibrinogen level decreases. Hemorrhage from portosystemic varices is complicated by this multifactorial coagulopathy.

Infusion of FFP is indicated during bleeding episodes to normalize the PT or PTT. In the absence of anatomic lesions, bleeding does not usually occur until the PT is longer than 16 to 18 seconds or the PTT is longer than 55 to 60 seconds, and prophylactic FFP (eg, before a liver biopsy) is not indicated at lower values. The PT and PTT are poor predictors of surgical bleeding, and mild abnormalities in these coagulation tests may not correct despite large volumes of FFP; platelet transfusion, cryoprecipitate, antifibrinolytic drugs, and even recombinant-activated factor VII may be required. This scenario is exaggerated in the setting of orthotopic liver transplantation complicated by an anhepatic stage, massive PRBC transfusion, and DIC. Large volumes of FFP are typically used, ideally guided by clinical assessment of bleeding and coagulation test results.38

Massive Transfusion

This is the most common situation during which anesthesiologists administer combinations of blood products. FFP is often required after large quantities of PRBC (>1 blood volume in 24 hours) because of the dilutional coagulopathy arising from the PRBC and crystalloid solutions, both of which lack coagulation factors. A predetermined algorithm for FFP transfusion per units of PRBCs is often unreliable, as dilutional thrombocytopenia also complicates the coagulopathy after massive transfusion.39,40 As recommended in the transfusion guidelines published by the American Society of Anesthesiologists, FFP transfusion should ideally be guided by abnormal PT and PTT results when available as the degree of coagulopathy can be unpredictable. Current trauma practices favor a 1:1 or 2:1 RBC-to-FFP ratio. Some studies appear to show a survival advantage but the evidence is not yet conclusive.41,42

Disseminated intravascular coagulation may be secondary to sepsis, liver disease, hypotension, perioperative hypoperfusion, trauma, obstetric complications, leukemia, or underlying malignancy. Successful treatment of the underlying cause is paramount. Patients with DIC leading to coagulopathic bleeding should be given FFP to correct PT and PTT abnormalities, but as in severe liver disease, FFP alone may fail to normalize PT and PTT.43 The need for cryoprecipitate or platelet transfusion should also be evaluated.

Rapid Reversal of Vitamin K Antagonists

Warfarin (Coumadin) inhibits the hepatic synthesis of vitamin K–dependent clotting factors (II, VII, IX, and X), inducing functional deficiencies of these factors that can correct within 48 hours after drug discontinuation if diet and vitamin K absorption are normal. Vitamin K administration corrects the coagulopathy in 12 to 18 hours. In an emergency, the deficiency of clotting factors can immediately be corrected by FFP transfusion. However, a large volume of FFP may be required, and if unlikely to be tolerated, smaller volumes of FFP can be augmented with a bypassing agent such as prothrombin complex concentrates or more commonly recombinant activated factor VII.44 In the absence of bleeding, the potential danger of excessively prolonged PTs must be weighed against the risks of FFP infusion, and FFP is rarely indicated.45

Thrombotic Thrombocytopenic Purpura and Hemolytic Uremic Syndrome

In some patients with thrombotic thrombocytopenic purpura (TTP); hemolytic uremic syndrome (HUS); or the syndrome of hemolysis, elevated liver enzymes, and low platelets associated with preeclampsia (HELLP), FFP exchange transfusion has been clinically therapeutic; plasmapheresis is preferred if there is concern for fluid overload.46,47 The mechanism of the therapeutic effect of FFP is likely to involve transfused anticoagulant proteins such as protein C and antithrombin III, which may restore endothelial prostaglandin I2 (prostacyclin) production and inhibit the agglutinating activity of the abnormal plasma.48 Alternatively, it may inhibit the release of large-molecular-weight von Willebrand factor (vWF) multimers or restore their normal processing in the circulation by deficient ADAMTS13 enzyme.49 The use of cryo-poor supernatant for refractory TTP has been suggested, but the advantage is unestablished.50

Dosage

One unit of FFP derived from a unit of whole blood contains 200 to 280 mL; apheresis may contain as much as 800 mL. On average, there are 0.7 to 1 unit/mL of activity of each coagulation factor per milliliter of FFP and 1 to 2 mg/mL fibrinogen. For perioperative use, the FFP dose may be estimated at 8 to 10 mL/kg and should be ordered as the number of milliliters to be infused for pediatric patients or units for adults. Laboratory measurements and clinical response determine the need for additional doses.

Compatibility and Side Effects

Fresh-frozen plasma is screened for unexpected RBC antibodies and should be ABO-type compatible as it contains anti-A (groups O and B) and anti-B (groups O and A). Cross-matching is not performed. The Rh(D) type is not always matched because immunization to Rh(D) antigen has rarely been reported as a result of transfusion of Rh(D)-positive plasma to Rh(D)-negative individuals. Hemolytic reactions as a consequence of infusion of an undetected antibody directed toward recipient RBC antigens are rarely seen, as is alloimmunization to RBC antigens.51

Fever, chills, and allergic reactions may occur and are treated symptomatically. Typical urticarial reactions are believed to result from recipient antibodies interacting with donor plasma protein antigens. Rarely, severe allergic reactions occur. Some are believed to be caused by donor antibodies reacting with recipient leukocytes or protein antigens. Anaphylactic reactions may occur after infusion of plasma (containing IgA) into patients with IgA deficiency and antibodies to IgA.52 IgA-deficient plasma from donors in the national registry can be obtained for these occasional patients. Transmission of infectious disease by FFP has been significantly reduced but not eliminated. Cell-associated CMV is not transmitted by FFP.53

Intravenous immunoglobulin (IVIg) is prepared by fractionation of large pools of human plasma. Indications for IVIg use include primary congenital immunodeficiency syndromes, children with HIV infection,54 chronic lymphocytic leukemia complicated by hypogammaglobulinemia, CMV pneumonia complicating bone marrow transplantation (in combination with ganciclovir),55 a possible role in preventing severe graft versus host disease after bone marrow transplantation, acute and chronic idiopathic thrombocytopenic purpura, and mucocutaneous lymph node syndrome (Kawasaki disease).56

Hyperimmune immunoglobulin (HIg) is prepared from large pools of plasma known to contain elevated antibody titers against specific infectious agents. Specifically, CMV HIg is indicated in transplantation recipients seronegative for CMV who receive a seropositive donor organ or in bone marrow transplantation patients with CMV interstitial pneumonia.55 Hepatitis B HIg is used to provide passive immunity to hepatitis B virus associated with inoculation by or a liver transplantation in HBsAg-positive individuals.

Rhesus HIg is used when fetal Rh(D)-positive RBCs may have entered the maternal circulation of an Rh(D)-negative mother. Rhesus HIg is given to Rh(D)-negative mothers after abortion or amniocentesis, as well as before delivery and again postpartum if the child proves to be Rh(D) positive.57,58 The therapeutic effect is believed to be caused by antibody feedback with T-cell suppression of the B-cell clone responsible for the formation of anti-Rh antibody. Its routine use in obstetrical practice has virtually eliminated alloimmunization of Rh-negative mothers.

Antithymocyte globulin (ATG) is purified from the hyperimmune serum of horses immunized with human T lymphocytes. ATG is used in transplant patients as an adjunct therapy in the treatment of graft rejection.

Cryoprecipitate is prepared from 1 unit of FFP thawed at 39.2°F (4°C). The precipitate is then refrozen in 10 to 15 mL of plasma and stored at −35°C (−22°F) or colder for up to 1 year. Cryoprecipitate contains 80 to 100 units of factor VIII, 100 to 250 mg of fibrinogen, 50 to 60 mg of fibronectin, 40% to 70% of normal vWF levels, and anti-A and anti-B antibodies.

Cryoprecipitate is effectively the blood component containing the highest concentration of factor VIII, vWF, factor XIII, fibrinogen, and fibronectin. Cryoprecipitate may be used to treat bleeding associated with hypo- or dysfibrinogenemia or factor XIII deficiency states, although an equivalent amount of fibrinogen to the usual adult dose of 100 mL is also found in 4 units of FFP. Its use in vWF and factor VIII deficiency has been superseded by individual factor replacement therapy.

Indications

Fibrinogen Deficiency

Fibrinogen deficiency may be a result of rare congenital afibrinogenemias or dysfibrinogenemias, or more commonly as a result of severe liver disease, DIC, or massive transfusion.45 These patients often have multifactorial coagulopathies and may require the coadministration of FFP, platelets, and antifibrinolytic agents. Fibrinogen measurements are an important component of a coagulation screen because levels less than 100 mg/dL can cause prolongation of both the PT and PTT despite adequate levels of other clotting factors. Low levels of fibrinogen occur during liver transplantation, necessitating cryoprecipitate transfusion.38

von Willebrand Disease

Initial platelet adhesion to disrupted endothelium is mediated by von Willebrand factor (vWF) interacting with the platelet surface GP1b complex; vWF is low or dysfunctional in the various types of von Willebrand disease. DDAVP (desmopressin acetate; 1-deamino-8-D-arginine vasopressin), which releases stored vWF from platelet granules and endothelial cells, is the initial treatment, but severe bleeding requires cryoprecipitate or vWF concentrate.59

Fibrin Glue

Fibrin glue results from the mixture of a fibrinogen source (FFP; platelet-rich plasma; heterologous, or more recently autologous, cryoprecipitate) with bovine thrombin, which leads to fibrin formation and enhanced local hemostasis. It has been used for diffuse local bleeding during cardiac surgery60 and to reduce blood loss and the factor concentrate requirement in surgical patients with von Willebrand disease.61 Other uses include meniscal tear repair and sealing postoperative neonatal chylothorax.62 Autologous cryoprecipitate avoids any infectious risks; bovine thrombin can cause anaphylaxis, formation of antibodies to factor V, or activation of coagulation.63

Uremic Bleeding

Uremia causes coagulopathy, and historically, cryoprecipitate or DDAVP has been used to treat bleeding associated with this.64 There are no data to support this, although cryoprecipitate could be considered if the bleeding is unresponsive to other therapies.65

Dosage

The dosage of cryoprecipitate is calculated on the basis of the amount of fibrinogen present in 1 unit of cryoprecipitate, the plasma volume, and the desired increment. The fibrinogen content of cryoprecipitate is variable, but even if relatively low, sufficient response is seen with transfusion of 2 to 4 U/10 kg. In practice, most adults are given a pooled 10-unit or 100-mL dose that is repeated if the desired response is not seen.

Compatibilty and Side Effects

Cryoprecipitate can contain anti-A or anti-B antibodies; therefore, infused units should be ABO plasma compatible. The risks of fevers, chills, allergic reactions, and infectious disease transmission are similar to those of FFP.

In the 1940s Dr Edwin Cohn developed an ethanol fractionation procedure (the Cohn fractionation procedure) that allowed the fractionation of human plasma into various components, greatly expanding this limited resource by deriving more products from each unit of donated plasma. Sufficient amounts of factor concentrates, however, require pooling of plasma. Present pathogen inactivation technologies have virtually eliminated infectious risks associated with plasma derivatives.

The development of recombinant products has been fueled by concerns of infectious disease risks, but recombinant expression processes are complex and expensive. Mammalian cell culture is most commonly used to optimize posttranslational modifications required for biologic activity, with transgenic recombinant technology being developed to decrease or eliminate reliance on donated human plasma resources or mammalian cell-culture–based recombinant production methods. Recombinant transgenic human antithrombin expressed in goat milk and human 1-antitrypsin produced in sheep's milk are currently clinically available.

The most common congenital deficiencies of coagulation proteins are hemophilia A (classic hemophilia or factor VIII deficiency) or hemophilia B (Christmas disease or factor IX deficiency). The genes for these coagulation factors are located in close proximity on the long arm of the X chromosome. Limited to males, hemophilia A affects 1 in 10000 males, whereas hemophilia B affects 1 in 30000. Familial cases predominate, but more than 30% of cases arise from spontaneous mutations.

Arthropathy is the major morbidity of the hemophiliac secondary to recurrent spontaneous joint bleeding, and such patients are frequently encountered for orthopedic surgery. The major cause of mortality (other than transfusion-related infection) is bleeding, with central nervous system (CNS) hemorrhage occurring in 3% to 14% of patients, conferring a mortality of 20% to 50%. CNS hemorrhage occurs predominantly in patients with severe disease (<1% factor level).

Citrated plasma was first used in 1923 for the treatment of hemophilia by Feissly in Paris, and the development of modern blood banking in the 1930s and expansion of transfusion during and after World War II allowed for more widespread use of whole blood and, subsequently, frozen plasma in the treatment of these fatal hemophilias. The advent of cryoprecipitate revolutionized hemophilia therapy. Cryoprecipitate derived from a single whole-blood collection contains approximately 125 Units of factor VIII, and it quickly replaced frozen plasma to treat hemophilia.

The fractionation of plasma using ethanol, glycine, polyethylene glycol, or a combination of glycine and polyethylene glycol, and calcium or barium to precipitate plasma proteins led to the production of factor VIII and factor IX concentrates for clinical use.66 This enabled high concentration, low volume, intensive infusion therapy for serious and life-threatening bleeding complications, such as intracranial, retroperitoneal, and retropharyngeal hemorrhages, and major surgery. From an infectious standpoint, these large pools of more than 1000 donations were universally contaminated with viral pathogens, such as hepatitis B or C, and between 1978 and 1985 most hemophiliacs were tragically infected with HIV from concentrates. Pasteurization and dry heat inactivate HIV and limit hepatitis B transmission; other strategies to limit infection include screening of potential donors for risk factors, surveillance of the blood supply for new pathogens, screening for markers of infectious agents, and other purification steps, including physical and chemical viral inactivation methods. After cloning of the genes factor VIII and factor IX, recombinant factor VIII and factor IX products were made available that are effective and free from pathogens.67

Treatment Regimens

The use of prophylactic regimens to maintain trough factor levels at more than 1% of normal reduces the incidence of arthropathy and CNS hemorrhage but requires dosing every 2 to 3 days with attendant problems and complications associated with long-term venous access.68

For the treatment of bleeding, the goal of therapy is to achieve a plasma factor VIII or IX level of between 30% and 50% for non–life-threatening bleeding episodes and a level of 100% for life-threatening bleeds or prophylaxis for surgical procedures using repeated boluses or continuous infusion for a duration of 10 to 14 days or longer, depending on the severity of the bleed or surgical intervention.

Adjunctive therapy includes DDAVP, a vasopressin analogue used for the treatment of patients with mild or moderate hemophilia A. This synthetic octapeptide causes a release of factor VIII (and von Willebrand factor) from endothelial cells, raising plasma factor VIII by approximately 3-fold (range, 2- to 12-fold) in patients with hemophilia in whom the disease is caused by decreased production or secretion of a functional protein or a protein that has decreased activity. Patients with severe hemophilia do not benefit from its use, as the mutation results in no synthesis or secretion, or a protein with no activity. The DDAVP response is unpredictable but reproducible in a given individual and should be documented by measuring factor VIII levels. The recommended intravenous dose is 0.3 g/kg infused slowly, as vasodilatation may occur. There have been reports of myocardial infarction and cerebral thrombosis with its use in patients at risk of arterial thrombosis; therefore, it should only be used in the setting of a documented coagulopathy.69

Dosing of factor VIII concentrates assumes an increase in plasma factor VIII activity of 2% for every 1 IU/kg infused. Cryoprecipitate replacement assumes approximately 80 to 150 IU per bag of factor VIII per bag of cryoprecipitate (derived from a 450-mL single whole-blood donor collection unit). Thus a typical 1750-IU dose (50% correction for a 70-kg patient) requires between 10 and 21 units.

Highly purified plasma-derived factor VIII concentrates or recombinant factor VIII concentrates have been available in North America, Europe, and Japan since 1992. These are either full-length or B-domain deleted molecules (the B domain is not required for activity in coagulation) that are expressed in mammalian cell culture (Chinese hamster ovary or baby hamster kidney cell lines) and are purified using immunoaffinity techniques. These products are formulated with added albumin as a stabilizer, but some have been developed that stabilize the factor VIII molecule with nonprotein molecules; despite no reliance on donated blood, the supply of these products is not abundant because of the complex manufacturing process.

Factor IX Concentrates

Intermediate- and high-purity plasma products and a recombinant factor IX product are available for the treatment of hemophilia B, all of which undergo at least 1 viral inactivation or exclusion step in manufacture. Recovery of factor IX after infusion is therefore approximately 1%/IU/kg. The recovery observed with recombinant factor IX is approximately 20% lower than that observed with plasma-derived factor IX, likely because of differences in posttranslational modifications between the recombinant and plasma-derived factor proteins.

Intermediate-purity factor IX products use anion exchange chromatography-selecting proteins containing highly negatively charged epitopes such as the vitamin K–dependent coagulation factors with γ-carboxyglutamic acid–rich domains. Therefore, the other vitamin K–dependent coagulation factors, factor VII, factor X, and prothrombin, are copurified with factor IX and referred to as "prothrombin complex concentrates" (PCCs). PCCs are contaminated with trace amounts of activated forms of these factors, which is the likely explanation for thromboses complicating their use.70 To minimize the risk of thrombosis with the use of PCCs, it is advisable to achieve a peak factor IX level no higher than 50%; some manufacturers have formulated PCCs with heparin or antithrombin III.70

High-purity factor IX products are produced from plasma using techniques of ligand affinity or immunoaffinity chromatography, and typically have specific activities greater than 150 IU/mg. A single high-purity recombinant factor IX product has been marketed. Thrombotic complications have not been reported with these high-purity products.

Hemophilia with Circulating Inhibitors

Replacement therapy may be complicated by the development of inhibitory antibodies against the infused factors that interfere with factor activity. The incidence of factor VIII inhibitors is approximately 30%, although titers and inhibitory activity vary.72

Treatment of acute bleeding episodes can be accomplished by 2 methods. For patients with low titer inhibitors, hemostasis can be accomplished with large doses of factor VIII or factor IX (eg, as high as 200 IU/kg given at frequent intervals) in an effort to neutralize, or "override," the circulating inhibitor. Porcine factor VIII can be used to treat an actively bleeding patient with inhibitors to human replacement factors, but problems include inhibitors cross-reacting with porcine factor VIII, allergic reactions, anaphylaxis, and thrombocytopenia caused by binding of porcine von Willebrand factor present in the concentrate to human platelets.

Alternatively, a bypassing agent is required. These bypass agents include PCCs in which contaminant-activated forms of these proteins, factors VIIa, Xa, IXa, and IIa (thrombin), likely enhance the production of a fibrin clot by activating the coagulation pathway later than the level of action of factors VIII and IX. Activated PCCs (aPCCs) have been treated to increase the levels of these activated factors and may be more effective in the treatment of bleeding episodes in patients with inhibitors. They were used as first-line therapy for patients with inhibitors, but because of the risk of thrombotic complications with aPCCs,73 recombinant activated factor VII (rFVIIa) has been licensed as an alternative bypass agent in the treatment of hemophiliacs with inhibitors.71 The dosage of rFVIIa is 90 g/kg every 2 hours as necessary, compared with 50 to 75 IU/kg every 8 to 12 hours for PCCs or aPCCs. Thrombosis has also been reported with the use of rFVIIa.77 Coadministration of antifibrinolytic agents, such as ε-aminocaproic acid, is often used as an adjunct to replacement or bypass therapy.

von Willebrand Disease

von Willebrand disease (vWD) is an autosomal dominant mucosal, platelet-type bleeding disorder first described by Erik von Willebrand in 1926, which has a prevalence of up to 2% of the general population. The penetrance and severity of the disease vary depending on the type of vWD (type 1, 2, or 3), the specific mutation, the number of affected genes, the patient's blood type, and numerous drug, hormonal, and other parameters. The type of vWD and the patient's response to DDAVP influence the recommended treatment for acute bleeding episodes or for prophylaxis against bleeding.66 DDAVP releases stored vWF as well as factor VIII from platelets and endothelium, and is often effective for type 1 disease and may be useful in certain patients with type 2 disease. It is not appropriate for use in patients with type 3 disease in whom no stores of vWF exist.

As described for hemophilia A, antifibrinolytic agents are useful adjuncts for replacement therapy for vWD such as in the setting of dental surgery to inhibit fibrinolysis. Estrogens upregulate vWF synthesis and may be useful especially in women, and may ameliorate menorrhagia.

Plasma Protein—Based Therapy: Factor VIII Concentrates

Some intermediate-purity factor VIII concentrates are manufactured from plasma using methods that copurify significant amounts of vWF, but no product has the pattern of multimers that is present in normal plasma.

von Willebrand Factor Concentrates (Plasma-Derived and Recombinant)

Several chromatography-purified plasma-derived concentrates enriched in vWF have been studied and may be effective and amenable to viral inactivation steps. Recombinant vWF has been successfully isolated and partially characterized; it is being developed clinically.

Cryoprecipitate was the mainstay of plasma-based therapy for bleeding in patients with vWD until the availability of virally inactivated intermediate-purity factor VIII concentrates that retained functional vWF protein. The factor VIII concentrates are the products of choice for the treatment of bleeding in patients with types 1 and 2 vWD that are unresponsive to DDAVP, and for bleeding in patients with type 3 vWD, as they have a lower risk of transfusion-associated infection than cryoprecipitate.

Other Factor Deficiencies

Approximately 15% of inherited bleeding disorders are caused by deficiencies of fibrinogen and factors II, V, VII, X, XI, and XIII.66 As the genes for these factors are not located on the X chromosome, homozygous defects are typically required for symptomatic disease. Hereditary deficiencies of factor XII, prekallikrein, and high-molecular-weight kininogen do not result in bleeding diatheses. The mainstays of therapy for these disorders are cryoprecipitate (for fibrinogen and factor XIII deficiencies) and plasma. Bypass therapy with PCCs, aPCCs, or rFVIIa can be considered for refractory bleeding, and adjunctive therapy with antifibrinolytics may be useful.

Acquired Disease

Antibodies that inhibit the activity of factor VIII can develop in previously normal patients. Most of these patients have no underlying disease; however, factor VIII antibodies can arise in the setting of autoimmune disorders or in the postpartum period. One-third of these inhibitors resolve spontaneously, but supportive therapy is required while they persist.

Acquired vWD is a rare autoimmune disorder that can occur in association with other autoimmune disorders or lymphoproliferative disease. Antibodies may interact with functional epitopes, and the immune complexes increase the clearance of vWF. Plasmapheresis or immunoadsorption can reduce the circulating levels of inhibitors. Immunosuppressive agents and intravenous immunoglobulin (IgG) are useful adjuncts to abrogate production of the autoantibody. Otherwise, the treatment of these conditions is as for congenital disease with associated circulating inhibitors.

Dilutional coagulopathy develops following resuscitation during trauma and major surgery or following hemodilution and cell salvage during cardiac surgery. It is a multifactorial deficiency and treatment should be guided by laboratory tests as recommended in the American Society of Anesthesiologists (ASA) guidelines for blood product usage.74 In Europe there is increasing interest in using factor concentrates in these settings to instantly boost certain factor levels,75,76 although this is preliminary and thrombotic complications may be similar to those plaguing the use of rFVIIa.77

Anticoagulant Proteins

Plasma-derived concentrates enriched in protein C and antithrombin III are available for the treatment of congenital deficiencies that result in thrombosis, and FFP contains normal levels of these circulating anticoagulants. Recombinant-activated protein C (Xigris, Eli Lilly and Company, Indianapolis, IN) is now licensed for the treatment of severe sepsis, and recombinant, transgenic antithrombin III is also available for the treatment of venoocclusive disease in patients who are undergoing bone marrow transplantation, heparin resistance in patients who are undergoing cardiopulmonary bypass, and sepsis.78

Preparations

In 1910 W.W. Duke described the Duke bleeding time, the importance of platelets in hemorrhagic disease, and the value of platelet transfusion. However, it was not until the 1970s before certain specialized centers could reliably offer platelet transfusions because of the preparation and storage problems resulting in loss of platelet viability. Platelets are prepared by centrifuging whole blood then separating the buffy coat (in Europe) or platelet-rich plasma (United States) for a second centrifugation to produce platelet concentrates (∼50 mL/U). A typical adult dose pools units from 6 donors. Apheresis platelets are collected from single donors using various pheresis devices and reduce donor exposure and thus infection risk. Leukocyte counts are less in apheresis or buffy coat–derived units than in US concentrates (106 rather than 108), but the introduction of leukoreduction techniques makes this distinction obsolete. Similarly, the risk of alloimmunization to HLA antigens is equal for concentrates or apheresis platelets undergoing leukoreduction.9 Platelets are stored at room temperature in plastic containers porous to the atmosphere to avoid anaerobic metabolism and acidosis, and are licensed for only 5 days of storage.79 Beyond this time, a storage lesion develops that limits platelet viability.

Platelet preparations contain appreciable volumes of plasma. If inventory constraints or HLA compatibility considerations lead to transfusion of plasma-incompatible products, mild recipient hemolysis may result. This is rarely clinically important unless large volumes are transfused in small patients. Crystalloid-based preservation solutions may allow the removal of plasma; such processes could also facilitate viral inactivation by photosensitive dyes. Cryopreserved and lyophilized platelets are areas of future development, with lyophilized platelets being close to clinical trials.80

Transfusion Triggers

Anesthesiologists encounter the need for platelet transfusions in intensive care and operative settings. Most (86%) platelets are given to patients with hematologic malignancies; 68% are given prophylactically, and 32% to treat bleeding episodes.81 Other indications include dilutional thrombocytopenia after massive transfusion-, autoimmune-, or drug-induced thrombocytopenia. Use may be relatively contraindicated in microangiopathic platelet destruction because of TTP, HUS, or DIC. Certain qualitative platelet disorders, either rare congenital conditions (Glanzmann thrombasthenia, Bernard-Soulier syndrome) or drug-induced dysfunction (aspirin, clopidogrel, Persantine, abciximab), respond to platelet transfusion. In contrast, high circulating plasma levels of eptifibatide or clopidogrel also render transfused platelets dysfunctional.

The current transfusion trigger for the most common indication of prophylaxis in oncology patients is 10000/mm3;82 in patients who are bleeding or undergoing invasive procedures the trigger is typically higher.83 A platelet count less than 50000/mm3 is quoted from the American Society of Anesthesiologists' guidelines. The use of recombinant activated factor VII has been used to augment or replace platelet transfusion, especially in patients refractory to transfusion.84,85

Refractoriness to Platelet Transfusion

Refractoriness to platelet transfusion can be caused by immune or nonimmune mechanisms.86 In immune mechanisms, alloantibodies, autoantibodies, drug-associated antibodies, or immune complexes can be responsible. Platelet membrane–associated anti-class I HLA-A and -B antibodies are responsible for most platelet refractoriness, and matching of so-called "public" HLA antigens, such as Bw4 or Bw6, forms the mainstay of compatibility testing. Importantly, ABO incompatibility can reduce platelet survival by up to 20%. Other rarer antibodies are HPA or human platelet antigens (polymorphic platelet surface glycoproteins) or platelet surface glycoproteins absent in the recipient (such as GP1b in Bernard-Soulier syndrome or GPIIb/IIIa in Glanzmann thrombasthenia). An antibody does not guarantee refractoriness, and underlying disease processes or immunosuppressive therapy may modulate the clinical picture.9 Nonimmune mechanisms include microangiopathic destruction (HUS or TTP), sepsis, or fever, especially in the presence of DIC, splenic sequestration, hepatic dysfunction, and certain drugs (such as vancomycin and amphotericin).

Posttransfusion purpura is a rare complication occurring 7 to 10 days after an immunogenic blood transfusion. It most often affects previously nontransfused, multiparous women. As with neonatal alloimmune thrombocytopenia, the risk for developing posttransfusion purpura is increased among HLA-DR3–positive individuals, and HPA-1a is the antigen most often implicated (in Western populations). At this time, there is no clear understanding of the exact pathology of posttransfusion purpura.

Human error is the root cause of almost all fatal acute HTRs, with transfusion of a unit into the wrong patient occurring in 50% of cases. Mislabeling of patient specimens sent to the blood bank and clerical errors in the blood bank occur more often than serologic testing errors, thus attention to design and implementation of safe transfusion practice is of paramount importance, more so than technological advances in serotyping. The aim of compatibility testing is to avoid immune-mediated HTRs resulting from recipient alloantibodies reacting to transfused RBC surface alloantigens. The direct (Fig. 83-1) and indirect (Fig. 83-2) agglutination tests are the mainstay of laboratory compatibility testing.

Pretransfusion testing assures ABO and Rh(D) compatibility, and screens recipient plasma for antibodies to less common but clinically important antigens. Fig. 83-3 describes the laboratory testing involved in grouping, typing, and cross-matching blood prior to transfusion. When performed in the setting of appropriate controls to prevent clerical error, compatibility determination effectively eliminates risk of acute intravascular hemolytic transfusion reactions. However, it does not eliminate the risk of alloimmunization to non-ABO/Rh antigens, nor does it afford complete protection against delayed hemolytic reactions.

In 2009 the AABB's Transfusion-Transmitted Diseases Committee made up of volunteer members with expertise in infectious disease, including representation from the Food and Drug Administration, Centers for Disease Control and Prevention, the US Department of Defense, the American Society of Hematology, and the Association of Public Health Laboratories, convened to publish a supplement to the journal Transfusion on the threat of emerging infectious diseases (EIDs).87 They prioritized specific agents into categories: red agents had the highest priority, followed by orange and yellow. White agents have not yet been given a priority status. The prioritization was based on several features of the organism including scientific/epidemiological evidence regarding blood safety, public perception and/or regulatory concern regarding blood safety, and public concern regarding the disease agent. Curious readers are directed to the reference for further understanding of the classification system.87

Red agents include human variant Creutzfeldt-Jacob disease, dengue viruses, and Babesia species. Orange agents include Chikungunya virus, St Louis encephalitis virus, Leishmania species, Plasmonium species, and Trypanosoma cruzi. Yellow agents include chronic wasting disease prions, human herpes virus 8, HIV variants, human parvovirus B19, influenza A virus subtype H5N1, simian foamy virus, Borrelia burgdorferi, and hepatitis A virus. White agents include hepatitis E and Anaplasma phagocytophilum.87

Bacterial Contamination

Bacterial contamination of blood components can be asymptomatic or induce sepsis with a high mortality. It occurs in random donor-pooled platelets (5-30 in 10000 units) and apheresis platelets (0.5-23 in 10000 units) stored at room temperature, PRBCs (0.25 in 10000 units) stored at 39.2°F (4°C), and, rarely, in FFP or cryoprecipitate contaminated during thawing in water baths.88,89 Bacterial contamination of platelet products is acknowledged as the most frequent infectious risk from transfusion, occurring in approximately 1 of 2000 to 3000 platelet units, and is considered one of the most common causes of death from transfusion (along with TRALI and clerical errors resulting in ABO mismatch) with mortality rates ranging from 1:20000 to 1:85000 donor exposures.88

Although bacterial contamination rate estimates vary, they are generally approximately 0.2% to 0.3%.89,90 Clinically recognized septic reactions have been reported at a rate of 1:2500 to 1:11400 for whole-blood–derived platelet concentrate pools and 1:15400 for apheresis platelets. Symptoms occurred after 17% to 42% of contaminated platelet transfusions, with a 17% mortality rate.90 The incidence of severe septic episodes has not been clearly established but is probably approximately 200 per million platelet units transfused (50% sensitivity).28,91 Given the 5-day storage life and the persistent risk of platelet shortage, in September 2005 the Food and Drug Administration (FDA) approved the use of 7-day apheresis platelets under the surveillance program PASSPORT to determine the safety of extending the storage life. The study was discontinued early after 2 true positive cultures were detected in 2571 day-8 platelets (778/million).92 However, after discontinuation, a group was convened to conduct a risk assessment. Based on their modeling, overall recipient risk may not have been improved by withdrawal of 7-day apheresis platelets, given the platelet pools that would have replaced them (surrogate-tested whole-blood platelets vs culture-tested apheresis or whole-blood platelets) and risk of delaying a TRALI risk reduction strategy, which could not be implemented without the additional platelet stores provided by the use of 7-day apheresis platelets.93

Bacteria can enter the blood bag during venipuncture as a result of inadequate skin preparation; during component preparation, transient bacteremia concurrent with the blood donation, or a break in technique during pooling or sealing; or through disruption of container integrity. Bacterial proliferation occurs more rapidly in platelet concentrates stored for up to 5 days at room temperature than in red cells refrigerated for up to 42 days.

The clinical response may include fever, rigors, skin flushing, abdominal cramps, myalgia, DIC, renal failure, cardiovascular collapse, and cardiac arrest; reactions may be immediate or delayed by several hours. This may be clinically indistinguishable from a febrile nonhemolytic transfusion reaction (FNHTR). If this picture presents, blood samples must be sent to rule out incompatible transfusion, and simultaneously the blood unit must be sent for culture. Patient evaluation includes blood culture; treatment includes stopping the transfusion, supportive measures for developing shock, and broad-spectrum antibiotics.

Gram-negative bacteria, including? Pseudomonas, Yersinia, Enterobacter, and Flavobacterium, are organisms commonly associated with a contaminated unit of refrigerated blood. In contrast, platelet concentrates contaminated with Bacillus, Escherichia coli, Klebsiella, Staphylococcus aureus, Staphylococcus epidermidis, Serratia marcescens, and Streptococcus account for 85% of fatal reactions.90,94 Bacterial contamination may be masked clinically by concomitant antibiotic administration.

Syphilis

Screening for syphilis was mandated in 1958, but the last reported case of transfusion-related disease was in 1966. Any risk of transmission is reduced by sensitivity to most penicillin or cephalosporin antibiotics. Treponema pallidum is not currently a threat to the US blood supply.

Viral Contamination

The risk for infection by HIV and hepatitis C viruses is extremely rare in the United States (<1 case per 2 million units transfused) after implementation of nucleic acid testing (NAT) in 1999.95,96 Contrast this blood safety record to the risk of morbidity or mortality in people with indications for transfusion for which blood products are withheld, and HIV or HCV should not affect transfusion decisions. However, a rare but real risk exists, so inappropriate transfusion should be avoided.

Other Viral Pathogens

Hepatitis A

The HAV, a nonenveloped RNA virus in the Picornaviridae family, is transmitted predominantly through the fecal–oral route. Because acute HAV infection is generally symptomatic, infected prospective donors are typically eliminated before donation.

Hepatitis B

The HBV is a DNA virus in the Hepadnaviridae family. The infective virion is known as the Dane particle and has surface and core components: HBsAg and hepatitis B core antigen (HBcAg). Among blood donors, the prevalence of new HBV infections declined from 1.97 per 100000 person years to 1.27 between 1998 and 2001. Most current HBV transfusion transmission cases correspond to blood donations by asymptomatic donors during acute infections before HBsAg appearance and therefore detection (a "window period" of 37-87 days). With current screening, the risk of HBV transmission per unit is approximately 1:205000 units, with likely reduction in risk with the introduction of NAT testing.95,96

Hepatitis C

The HCV is an RNA virus in the Flaviviridae family that has 6 genotypes. In the United States, genotypes 1, 2, and 3 cause 75%, 10%, and 10% of infections, respectively, and have similar replication and transmission rates and natural history. The risk of HCV transmission by transfusion declined after the introduction of serologic testing for HCV antibody in May 1990, and NAT testing for the HCV viral genome in 1999 reduced the test-negative window period to 40 days. Combined NAT and serologic testing reduced the risk from an estimated 1:276000 units to 1:1 935000 units.

Hepatitis D

Originally called the delta agent, hepatitis D virus is a defective RNA-containing passenger virus that requires HBV to act as a "helper" for assembly of envelope proteins. Screening for hepatitis B therefore helps prevent transfusion-associated hepatitis D.

Hepatitis E

The hepatitis E virus (HEV) is an RNA calicivirus associated with fecally contaminated water supplies, which usually causes a self-limited illness; it is not a known transfusion-related pathogen.

Human Immunodeficiency Virus-1

The Centers for Disease Control and Prevention (CDC) report 9352 AIDS cases in the United States linked to transfusion through December 31, 2001, including 41 adults or adolescents and 2 children who received blood from HIV-seronegative donors. In addition, during the 1980s, 4799 hemophiliacs and other patients with coagulation deficiencies acquired AIDS as a result of therapy with plasma derivatives.

Among 37 million donations screened for HIV-1 RNA by NAT between 1999 and 2002, only 12 were NAT positive and antibody negative.96 The risk with plasma derivatives, such as coagulation factor concentrates and albumin and immunoglobulin preparations, can be expected to be even lower because of additional viral inactivation processes, including heat (pasteurization), physicochemical processes (SD treatment), and nanofiltration, that have been implemented since the mid-1980s; since then, no new HIV infections have been attributed to manufactured blood products.

Human Immunodeficiency Virus-2

Human immunodeficiency virus-2, a retrovirus linked more closely to the simian immunodeficiency virus than to HIV-l, was recognized initially in West Africa. In 1998, the CDC reported 79 cases of HIV-2 infection in the United States, mostly in natives of West African countries. Transmissibility of HIV-2 appears to be lower, the course of infection milder, and the interval between infection and AIDS longer than that associated with HIV-1, presenting a lower risk for HIV-2 transmission by transfusion.

Human T-Cell Lymphotropic Virus-I and -II

HTLV-I and -II are closely related retroviruses in the Oncovirinae group. In contrast with HIV, HTLV is rarely present in cell-free plasma and shows little active replication in infected humans. HTLV is found around the globe, with endemic foci in southern Japan, the Caribbean, South America, and the Middle East. A sensitive HTLV-I/II combination assay was introduced in 1998, and the threat of acquiring HTLV infection from screened blood is currently minimal.

Human Herpesvirus

Human herpesviruses (HHVs) are enveloped, structurally complex, double-stranded DNA viruses that cause common infectious diseases, usually associated with lifelong carrier states and the possibility of recurrent reactivation infections. However, the common alpha herpesviruses, herpes simplex and varicella zoster, are not linked to transfusion-transmitted infections.

Cytomegalovirus Infection

CMV, a beta herpesvirus (HHV-5), can infect a wide range of cell types, primarily leukocytes; cell-depleted blood components (plasma, cryoprecipitate) do not transmit CMV. In immunocompetent people, this community-acquired infection is either asymptomatic or associated with a mild, self-limited infectious mononucleosis-like syndrome. However, latent virus persists permanently intracellularly, allowing lifelong potential for reactivation of infections or viral transmission by transfusion of cellular blood products or transplanted donor organs.

In immunosuppressed patients, CMV infection leads to morbid and occasionally lethal pneumonitis, hepatitis, gastroenteritis, retinitis, and other inflammatory conditions requiring treatment with ganciclovir and other antiviral agents. Primary infection in pregnancy may be associated with severe fetal malformations and congenital infections complicated by jaundice, hepatosplenomegaly, microencephaly, and thrombocytopenia with significant mortality. Screening for CMV status is essential for neonates, bone marrow transplant recipients, and transplant recipients. A CMV-seronegative transplant recipient is not denied an organ from a seropositive donor but is treated prophylactically and tested for infection. If leukoreduced blood components are not available, transfusion from CMV seropositive donors is avoided to reduce viral load.

At least 50% of blood donors have antibodies against CMV, thereby indicating previous CMV infection. CMV infection develops in 2% of CMV-seronegative transplant recipients given CMV-seronegative blood, compared with 30% to 60% historically seen after infusion of CMV-unscreened blood components.66 Additional safety is conferred by prestorage leukocyte reduction and apheresis platelet collection.

Epstein-Barr Virus

Epstein-Barr virus (EBV), a gamma herpesvirus (HHV-4), causes infectious mononucleosis and is closely associated with Burkitt lymphoma, nasopharyngeal carcinoma, and posttransplantation lymphoproliferative disease. More than 90% of the adult population has evidence of previous exposure, and virus-specific cytotoxic T lymphocytes lyse EBV-infected B lymphocytes, so transfusion-transmitted EBV infection is rare in both immunocompetent and immunosuppressed recipients.

Parvovirus

Patients with sickle cell anemia, thalassemia, and other conditions associated with shortened red blood cell survival are at risk for developing acute aplastic or hypoplastic anemia after infection. Others at risk for aplasia after parvovirus infection include immunodeficient patients, HIV-infected patients, solid-organ transplant recipients, and children with malignancies. Persistent parvovirus infection may cause severe chronic anemia in immunocompromised patients. Acute parvovirus infection during pregnancy may result in fetal loss, neurologic abnormalities, and congenital infection. Red blood cell aplasia and chronic anemia caused by parvovirus infection often respond to infusion with immunoglobulin preparations.

Parvovirus can be expected to be present in 1:20000 to 1:50000 blood donors or a higher incidence during epidemic periods. Reactive antibodies may be present in the donor or the recipient, and only 3 cases of transfusion-related transmission have been reported. Pooled plasma products theoretically pose a greater threat, and factor VIII concentrates demonstrated a 40% risk of parvovirus infection.97 The pasteurization process inactivates virus in albumin-based products.

West Nile Virus

West Nile virus (WNV) is a member of the Flaviviridae family, the genus Flavivirus, and the Japanese encephalitis virus serocomplex that includes Japanese encephalitis virus and St Louis encephalitis virus. Viruses in this complex are arthropod-borne viruses, or arboviruses (ie, transmitted by mosquitoes and other arthropod vectors), with the potential to cause meningoencephalitis. WNV was first isolated in 1937 in the West Nile district of northern Uganda and derives its name from that region. The natural life cycle of the virus includes female mosquitoes as vectors, with birds serving as the primary vertebrate hosts. Humans are incidental hosts, with transmission occurring through bites of infected mosquitoes. Peak transmission occurs in the late summer and early fall.

Since 1999 there have been sporadic outbreaks in the United States, with more than 4000 cases—277 attributed to a 2002 epidemic. In 2002, 23 cases of transfusion (any blood component) or transplant-related transmission were reported; 12 patients developed meningoencephalitis and more severe outcomes were seen in immunocompromised patients.98

Currently, donors are asked about fevers or exposure to endemic areas, and recommendations exclude donors with suspected or confirmed diagnosis of WNV within 120 days of donation. Since June 2003 NAT was instituted on pooled donor samples, and 800 of 6 million US blood donations were WNV NAT positive. Despite this, 6 cases of transfusion-transmitted WNV infection occurred in susceptible individuals.99 The risk of WNV infection depends on the immune status of the recipient, the presence of donor antibodies, and the rate of endemic disease prevalence.

Dengue Virus

Dengue virus was prioritized in the red category based on scientific/epidemiological evidence regarding blood safety, specifically related to the asymptomatic viremia that poses a significant transfusion risk.87 It is a mosquito-borne disease. Asia, Oceania, Africa, Australia, and the Americas contain at-risk populations. Transmission through blood transfusion has been documented. The clinical disease most commonly is limited to dengue fever with fever and rash, conjunctivitis, myalgia, nausea/vomiting, and prostration. Other more severe forms including dengue hemorrhagic fever and dengue shock syndrome are more rare and occur primarily in children. The mortality in this group is high, 10% to 20%. There is currently no FDA-licensed screening test.87

Lyme Disease and Other Tick-Borne Illnesses

Borrelia burgdorferi, the agent responsible for Lyme disease, is transmitted to humans by deer tick bites. Spirochetemia probably occurs after infection and may be present in asymptomatic individuals, but again, only rarely documented transfusion-related transmissions have been reported for this or other tick-borne pathogens: Babesia species (babesiosis) and Rickettsia rickettsii (Rocky Mountain spotted fever).

Malaria

Malaria in the United States is typically limited to travelers, military personnel, and immigrants from endemic countries. Prevention of transfusion-transmitted malaria relies on deferral of blood donors immigrating or returning from malaria-endemic regions; there are no available screening tests. Approximately 3 transfusion-associated malaria cases occur per year in the United States, with a reported incidence of 0 to 0.2 cases per million units.

Babesiosis

Babesia was placed in the red category based on scientific/epidemiologic evidence regarding blood safety.87 Babesiosis is a malaria-like zoonosis in which humans are infected incidentally, usually through the bite of an infected tick. Blood transfusion has been documented. Babesia infections are usually asymptomatic or cause mild flu-like symptoms but can be fatal in the immunocompromised. Endemic throughout the United States, more than 40 cases of cellular transfusion-transmitted babesiosis have occurred since 1980. A history of babesiosis precludes blood donation, but no screening test is available.

Toxoplasmosis

Toxoplasmosis is caused by the intracellular protozoal parasite Toxoplasma gondii, whose usual host is the domestic cat. Transmission is via cats or from undercooked pork, goat, lamb, beef, or wild game. There is evidence of infection in 20% to 25% of the US population, but transfusion-associated disease has only been described in immunocompromised patients.

Chagas Disease

Chagas was assigned category orange with primary concern with public perception and/or regulatory concern regarding blood safety.87

The flagellate protozoan parasite Trypanosoma cruzi causes Chagas disease. The disorder is widespread in South America, but T. cruzi is also found in North and Central America. Infection is not endemic in the United States, but it is considered emergent in the United States and Canada based on increased immigration. Routes of infection include exposure to feces from infected vectors (largely insects), blood transfusion and organ transplantation, and congenital transmission and breast-feeding. Severe complications include myocarditis, meningoencephalitis, cardiomyopathy, megacolon, or achalasia. Transfusion-associated cases of Chagas disease are more common in immunocompromised patients. In North America, testing is available but is not required by the FDA, and even testing of donors with risk factors or those from endemic areas has not been implemented.87

Human Variant Creutzfeldt-Jakob Disease

Human variant Creutzfeldt-Jakob disease (vCJD) is a prion disease given red priority based on public perception and/or regulatory concern regarding blood safety.87 Exposure is through consumption of bovine spongiform encephalopathy–infected neural beef products. Current to October 2008, 206 cases and 203 deaths had occurred in the United Kingdom or secondary to UK beef products. Blood transmissibility of the disease has been documented and a potentially long incubation period increases the theoretical possibility of transfusion-related transmission. The likelihood of clinical disease after transmission is hypothesized to be high, with mortality estimated to be 100% for those who are symptomatic. There are no screening tests available and there is no treatment. Screening questions that eliminate US donors with UK travel history, leukoreduction, and filtration methods may reduce prion content in US blood stores.87

New molecular assays, such as NAT, can enhance safety but at a high cost compared with the serologic screening that was effective for virtually negating the HBV, HIV, and HCV risk of transfusion. The advent of newer infectious agents, such as vCJD, WNV, severe acute respiratory syndrome (SARS), and potentially Avian influenza A (H5N1) virus, may lead to the deferral of increasing numbers of potential blood donors, with implications for blood supply.

The discussion of infectious agents is mostly based on North American data, but additional challenges exist for ensuring a safe blood supply in developing countries. With global travel and immigration, proactive and collaborative surveillance on an international level is essential to protect any country's blood supply.100

An FNHTR occurs when temperature increases above 1.8°F (1°C) during or after transfusion when no other cause can be found. These reactions are caused either by cytotoxic or agglutinating antibodies in the patient's plasma reacting against antigens present on transfused donor lymphocytes, granulocytes, or platelets, or by donor plasma containing antibodies against the recipient's cellular antigens. These antigens are typically, but not always, HLA antigens.101 The febrile reaction results from antibody–leukocyte or antibody–platelet interactions leading to the release of inflammatory cytokines and pyrogens.

These cytokines are released from recipient leukocytes or donor leukocytes in response to antigen–antibody interaction, but could also be in the supernatant of the transfused unit, derived from the donor leukocytes during storage before transfusion.102 Such generation of cytokines, mostly from mononuclear leukocytes, occurs during storage and is proportional to the leukocyte count of the unit and the duration of storage.103 Febrile reactions to platelet transfusions are more likely caused by proinflammatory cytokines in the supernatant than by cellular elements.104 Use of prestorage leukocyte reduction lowers, but not eliminates, the incidence of febrile reactions, as leukoreduction filters do not remove cytokines.105

The frequency of febrile reactions is approximately 0.5% per unit transfused, most commonly in recipients exposed to multiple white cell or platelet antigens such as oncology patients or multiparous women.28 Previous alloimmunization leads to anti-HLA, granulocyte, or platelet-specific antibodies that react with white cells or platelets on subsequent exposure.

Clinically, an FNHTR is characterized by fever and chills occurring shortly after the transfusion begins. FNHTRs are usually self-limited, with fever persisting for no more than 8 to 10 hours. Elderly patients with a tenuous cardiovascular status and critically ill patients may develop respiratory complications, hypotension, or shock. With any transfusion reaction, the transfusion must be stopped, antipyretics and supportive measures instituted as necessary, and laboratory evaluation initiated.

A febrile reaction may be the first sign of an acute hemolytic reaction or infusion of a unit of red cells or platelets contaminated with bacteria. Repeat cross-matching is performed to confirm patient–donor ABO compatibility, examining the results of pre- and posttransfusion direct antiglobulin tests (DATs), evaluating the serum for hemolysis, and confirming the accuracy of paperwork. The posttransfusion DAT should yield negative findings, as an FNHTR does not involve red cell alloantibodies. Sending blood cultures from the patient and the blood product, and administration of broad-spectrum antibiotics can be considered depending on the severity of the reaction.

The diagnosis of FNHTR is one of exclusion after ruling out a hemolytic reaction, a septic transfusion, or other miscellaneous causes of fever. Routine premedication is not indicated, but antipyretics and corticosteroids can be reserved for patients with a prior history of FNHTR. The use of leukocyte-depleted blood components is not universal, but as discussed earlier, this practice reduces the incidence of FNHTRs and should be mandated for patients with a previous history of FNHTRs.106

Allergic transfusion reactions are most commonly caused by the infusion of plasma proteins. Manifestations include skin erythema with associated mild urticaria and pruritus, a confluent rash that is intensely pruritic, extensive urticaria, severe vasomotor instability, bronchospasm, and anaphylaxis. The severity of these reactions may not be dose related, so a patient developing hives or a mild allergic reaction during a blood transfusion may not necessarily progress to severe anaphylaxis by completing the transfusion.

Causative antibodies have not been conclusively demonstrated, but milder allergic reactions are usually IgG or IgE mediated; anaphylactic reactions are most often IgG mediated. Allergic transfusion reactions are quite common, occurring in approximately 1% of all transfusions and typically leading to pruritus followed by hives. Treatment involves holding the transfusion, administering antihistamines, and assessing the wisdom of continuing the transfusion based on improvement of symptoms and no progression to more serious symptoms such as fever, chills, bronchospasm, dyspnea, or hemodynamic instability.

Such mild urticarial reactions rarely progress to serious reactions and often do not recur with subsequent transfusions. The pathogenesis is believed to be recipient-antibody directed against donor plasma proteins, although the etiologic antibody is rarely detected, making the diagnosis of an allergic transfusion reaction a diagnosis of exclusion. Leukocyte depletion filters do not remove plasma proteins, so washed red cells are required to prevent reactions.107 However, washed cells are reserved for patients with severe or recurrent reactions.

Anaphylactic Reactions

Other anaphylactic reactions can occur when plasma containing IgA is transfused to patients with IgG anti-IgA antibodies or after transfusion of red cells or, more commonly, platelets administered through certain bedside leukoreduction filters.108 The latter scenario is believed to be a consequence of contact activation by the negatively charged surface of the filter converting prekallikrein to kallikrein, in turn converting high-molecular-weight kininogen to bradykinin. Bradykinin causes pain, cutaneous flushing, and hypotension without fever or chills, and its effects are exaggerated in patients taking angiotensin-converting enzyme (ACE) inhibitors, as ACE is identical to kininase II, which is responsible for degrading bradykinin. A common theme throughout this chapter is that prestorage leukoreduction avoids this problem.

Microaggregate Debris and Adult Respiratory Distress Syndrome

Microaggregate debris consists of dead platelets, granulocytes, and fibrin strands that form in blood during storage. Theoretically, this debris (20-120 micron) passes through standard 170-m pore filters, infusing these microaggregates into the pulmonary vasculature, obstructing pulmonary capillaries, and causing pulmonary failure. Microaggregate filters have been suggested to prevent adult respiratory distress syndrome (ARDS) after massive transfusion, but they may reduce transfusion flow rates and do not appear to alter the onset of ARDS.109 Hypotension and sepsis are postulated to trigger ARDS rather than microaggregate debris, and improvements in the general care of the critically ill treating shock and sepsis have likely decreased the incidence of postperfusion respiratory distress syndrome more than the use of microaggregate filters.

Transfusion-Associated Circulatory Overload (TACO)

It is obvious to any practicing anesthesiologist that rapidly transfusing a patient who is euvolemic and not actively bleeding is unlikely to be of benefit and may cause harm. Infants, patients with cardiovascular disease, and the elderly are especially at risk. Importantly, transfusion-induced hypervolemia may be initially difficult to distinguish from hemolytic transfusion reaction, a febrile nonhemolytic transfusion reaction, transfusion-related acute lung injury (TRALI), or allergic reactions. Hypervolemia results in headache, dyspnea, tachycardia, tachypnea, and congestive heart failure. The absence of hemoglobinuria and hemoglobinemia or a positive posttransfusion DAT distinguishes hypervolemia from immune hemolysis, and the absence of fever, chills, or urticaria from febrile or allergic reactions.

Treatment involves stopping transfusion until the diagnosis is determined, administering diuretics and supportive therapy as indicated, and resuming the transfusion at a slower rate while monitoring for recurrence of symptoms and signs of hypervolemia.

Transfusion-Related Acute Lung Injury

As TRALI recognition has grown there has been an international effort to formalize a definition, identify current etiologic theories, and provide insight into methods of prevention. In 2004 the Canadian Blood Services convened a consensus conference on TRALI, and in 2005 published their proceedings where they reviewed TRALI and proposed a definition.110 Additionally, in 2005 the US National Heart, Lung, and Blood Institute offered insights from their work group.111 It is clinically indistinguishable from acute lung injury (ALI)/ARDS and hence the definition includes this diagnosis,112 noncardiogenic pulmonary edema and hypoxemia, but with the addition of a temporal association with transfusion.110,111,113,114 The symptoms of TRALI include fever, chills, cyanosis, dyspnea, tachypnea, and hypotension or hypertension, and copious amounts of pulmonary edema fluid.110 If a patient shows signs of pulmonary insufficiency, as with all other reactions, transfusion should be stopped, diagnosis sought, and supportive measures (including mechanical ventilation) initiated as indicated.

The incidence of TRALI in the United States is 1 in 1300 to 1 in 5000, and although mortality estimates range widely between 5% and 25%, it is the most common fatal transfusion-related reaction.110,113,115 All blood products have been associated with TRALI; however, plasma-containing products are associated with a higher incidence.110,116

The pathophysiology is likely multifactorial. Two well-described mechanisms involve the transfusion of antileukocyte antibodies and/or the transfusion of older stored cellular products. Anti-HLA I, anti-HLA II, or antigranulocyte (antileukocyte) antibodies recognize recipient leukocytes, leading to complement activation and leukocyte aggregation.113,117 Activated leukocytes express adhesive molecules on their surfaces, permitting attachment to pulmonary endothelial cells and entry into the interstitial space by diapedesis, with degranulation leading to pulmonary edema as a result of microvascular leukostasis/occlusion and capillary leakage.118 These alloantibodies result from pregnancy, transfusion, and transplantation. Their prevalence has been reported to be less than 2% among women with no pregnancies, and nontransfused and transfused men, but 24.4% in previously pregnant women.119 Although many studies are able to find a clear antibody–antigen link, this theory is troubled by the lack of consistent measureable specificity of the donor antibody–recipient antigen interaction. It appears that this antibody–antigen interaction is neither necessary nor reliably sufficient to cause TRALI; it has been reported to occur without a clear antigen–antibody interaction, and in cases where there is an antibody and a cognate antigen, TRALI is not assured. Nonetheless, female donors have been implicated in many TRALI cases,116 and avoidance of transfusion of female plasma may decrease the incidence of TRALI, leading to the diversion of female plasma donation in Europe and the United States.120,121 This does not have an impact on all female blood donations, however, as female donors only increased the risk of TRALI in plasma-rich products, not RBCs.122

A complementary theory, and one that may explain antibody-negative TRALI, is the idea of a 2-event model proposed by Silliman et al.115,123,124 The first event is a systemic illness or the condition of the patient responsible for the priming of neutrophils and the activation of the pulmonary endothelium with neutrophil sequestration. The second event is the transfusion of biologic response modifiers (BRMs), which include lipids, and/or soluble CD40 ligand (a platelet-derived proinflammatory mediator),115,123–125 both of which accumulate during the storage of blood. The BRMs activate the sequestered neutrophils and lead to the damage of pulmonary endothelial cells, the accumulation of lung water, and the clinical syndrome of ALI, temporally associated with the transfusion of blood product.110,115,116,124–129 Therefore, based on this mechanism, avoidance of female donation is unlikely to eliminate TRALI.

In addition to multiparous females, exclusion policies include the deferral of donors with known antileukocyte antibodies or those who have been linked to 1 or more TRALI events. Future product management policies may include washing of cellular components to eliminate plasma or using fresher blood products to decrease BRMs; current policies do not address the controversy surrounding the storage age of RBCs.130 More research is needed to clarify the benefits of each of these actions weighed against their cost to the blood inventory by reducing the donor pool.

Hypothermia

Hypothermia can occur with rapid infusion of large quantities of cold blood. Rapid infusion of cold blood (∼5 min/U) may lower the temperature of the sinoatrial node to below 86°F (30°C), leading to serious arrhythmias. Numerous in-line blood-warming devices are available for clinically indicated rapid transfusion, as in major trauma, and may reduce the incidence of hypothermia-induced coagulopathy and platelet dysfunction that occurs with temperatures below 95°F (35°C). Only approved blood warmers should be used for this purpose. Such devices monitor and regulate temperature to below 107.6°F (42°C) to avoid causing hemolysis. The use of hot water or microwave devices is unacceptable, as hot spots develop that can cause hemolysis.

Electrolyte Toxicity

Citrate is used for anticoagulation during storage, acting by chelating calcium and interfering with the coagulation cascade. Rapid transfusion of citrated blood thus can be associated with a decrease in ionized calcium levels. Typically, citrate is rapidly metabolized in the liver to bicarbonate, but mild to severe citrate toxicity can be seen after infusion of large volumes of citrated products or in the setting of transient or established liver dysfunction.

The effects of hypocalcemia range from mild circumoral paresthesia to frank tetany. If prolonged Q-T intervals or signs of tetany are seen, calcium can be administered. In practice, the routine availability of arterial blood gas analysis with ionized calcium measurement allows the close monitoring and guided treatment of developing hypocalcemia. Calcium should never be added to a unit of blood, as it recalcifies the unit and leads to clot formation; similarly, the use of calcium-containing solutions, such as Ringer lactate, should not be coadministered with blood products.

Hypomagnesemia, presumably caused by chelation of magnesium by citrate, has also been reported, but its clinical relevance has not been determined. In practice, administration of magnesium can accompany calcium administration in response to a large citrate load.

Hyperkalemia caused by infusion of stored blood is rare, and hypokalemia may be more common. With storage, leakage of potassium from red cells to the extracellular fluid occurs. However, after infusion into a recipient, the red cells reverse the biochemical storage lesion, and intracellular potassium levels are restored. In addition, citrate is metabolized to bicarbonate, causing alkalosis contributing to hypokalemia. In massive transfusion, this may not uncommonly result in the need for administration of potassium. Extracellular potassium increases at the rate of approximately 1 mEq/d during the first few weeks of storage. However, hyperkalemia is still a concern for neonates or patients with renal failure, and washed or fresher units can be requested. Potassium leakage from red cells is increased after exposure to 25-Gy irradiation to prevent posttransfusion graft versus host disease; the shelf life of irradiated units is a maximum of 28 days.

Plasticizers

Plasticizers are chemicals used to make the rigid polyvinyl chloride plastic used in blood bags more malleable; traditionally, diethylhexyl phthalate (DEHP) is used. After concerns about carcinogenic properties and the possible production of peroxisomes by the metabolite monoethylhexyl phthalate, DEHP is being replaced with a citrate-based plasticizer.131

Graft versus Host Disease

Graft versus host disease (GVHD) occurs when immunologically competent lymphocytes are introduced into an immunoincompetent host who cannot destroy the donor lymphocytes. The immunocompetent donor lymphocytes engraft, recognize the host as foreign, and then attack host tissues. GVHD occurs after allogeneic bone marrow transplantation and less often after transfusion of nonirradiated cellular blood components, especially when the blood donor and recipient share some HLA antigens. There is an increased danger from posttransfusion GVHD in part because of the frequent failure of physicians to recognize and treat the reaction promptly. Another major factor, however, is the propensity of the donor's lymphocytes to attack and produce recipient bone marrow aplasia in contrast to after bone marrow transplantation GVHD, in which the bone marrow is of donor origin, and bone marrow aplasia does not occur.

Posttransfusion GVHD is fatal in more than 90% of cases, primarily because of aplasia of the recipient's bone marrow. Reports have shown that haploidentical-directed donor units of blood may produce fatal posttransfusion GVHD even in immunocompetent recipients. The use of irradiated blood (>2500 cGy) is now recommended in clinical situations in which transfusion poses a risk of GVHD such as when patients receive blood transfusions from their relatives or HLA-matched donors. Leukocyte reduction is not adequate prophylaxis against GVHD, as the exact number of leukocytes needed to produce the disease is unknown.132

Hemosiderosis

A unit of red cells contains approximately 250 mg of iron, and 1 g of iron is roughly the amount stored in the bone marrow; men and nonmenstruating women lose only approximately 1 mg of iron per day. Consequently, repeated transfusion in individuals with an extravascular hemolytic anemia, such as thalassemia or sickle cell anemia, in which iron is recycled rather than excreted, can result in the accumulation of excessive tissue stores of iron. Chronically, iron stored in parenchymal cells results in cell death and eventual organ failure. Iron chelation therapy, such as deferoxamine, is the mainstay of therapy to avoid predictable iron overload and maintain patients with chronic hemolytic anemias in negative iron balance.

Air Emboli

Air is currently expelled from plastic blood bags but still may be pumped into patients by pressurized transfusion devices, especially after apheresis and intraoperative salvage. All such devices currently manufactured, however, contain air in-line sensors. Regardless, clinicians must remain alert to the potential risk of air embolization at all times while the patient is being treated. Patients suffering large air emboli (boluses of 3-8 mL/kg) can experience acute cyanosis, pain, cough, arrhythmia, acute right ventricular outflow obstruction, cardiogenic shock, and circulatory arrest. Infusions should be stopped and checked for air, and the patient should be placed head-down on the left side. This usually displaces the air bubble from the pulmonary valve and may be amenable to aspiration via a central venous catheter.

Transfusion-Related Immunomodulation

Deleterious effects of transfusion can be caused by a reaction to or incompatibility with a component of the transfused blood product or by the subtler transfusion-related immunomodulation (TRIM). TRIM has been linked to 2 different forms of immunosuppression, one believed to predispose patients to infectious complications and the other pertaining to tolerance of allogeneic transplants. TRIM has a number of postulated immunosuppressive mechanisms related to leukocytes and cytokines present in blood components limiting the potential effect to transfused red cell concentrates and platelets. Leukocytes and leukocyte-derived bioactive substances, such as histamine, myeloperoxidase (MPO), eosinophil cationic protein (ECP), eosinophil protein X, plasminogen activator inhibitor 1 (PAI-1), and IL-6 are known to contaminate PRBCs133 and platelet units,134 accumulating in a storage time–dependent manner. Suppression of lymphocyte proliferation has been induced by these mediators, providing a mechanism for TRIM. Length of storage of transfused allogeneic PRBCs, but not plasma, is associated with pneumonia after coronary artery bypass graft surgery.135

In addition, phosphatidylserine is expressed on the surface of apoptotic leukocytes in PRBCs and platelets and erythrocytes, mediating phagocytosis of apoptotic cells. This induces transforming growth factor (TGF-α) secretion, resulting in an anti-inflammatory effect and suppression of proinflammatory mediators136; enhanced production of the anti-inflammatory cytokines IL-4, IL-10, and transforming growth factor is observed after blood transfusion. Anti-inflammatory cytokine release and decreased cellular immune function occur after allogeneic, but not autologous, transfusions.

Prestorage leukofiltration appears to reduce storage time–dependent immunosuppression after blood transfusion and the incidence of febrile transfusion reactions.137 Indeed, leukoreducing PRBCs decrease postoperative mortality after cardiac surgery and the incidence of multiorgan failure after major surgery.138 However, leukoreduction is not a panacea, as leukocytes are not eliminated. Leukoreduced PRBCs can still cause hypotensive transfusion reactions and HLA alloimmunization, and express surface phospholipids. It may be that a greater volume of leukoreduced products is required to produce the deleterious effects of transfusion compared with nonleukoreduced products.

In summary, the safety of the US blood supply is exemplary, with the chance of a fatal reaction to transfusion in the realm of 1 in every 100000 patients transfused. Such risk is acceptable should there be definite benefit, which emphasizes the responsibility placed on the physician to use appropriate triggers for blood component therapy with physiologic evidence of a deficiency in oxygen-carrying capacity for red cell transfusion and an accurately defined coagulopathy for hemostatic blood product transfusion. Second only to hematology nurses, anesthesiologists transfuse more blood products than any physician group, and it is incumbent on them to lead blood conservation strategies, including research into safe and effective blood substitutes, and reinforce surgical philosophies that avoid unnecessary blood loss at all costs. Such approaches provide the best possible care for patients and prepare anesthesiologists well for practice in areas with less safe blood supplies or during unforeseen safety concerns with the US blood supply.

In 2008 the United States launched the Hemovigilance Module of the US Biovigilance initiative from the National Healthcare Safety Network (NHSN), jointly developed by the CDC and AABB, charged with data collection and analysis of transfusion practices and outcomes. In the coming years, this centralized database designed to report and monitor recipient adverse reactions and quality control incidents related to blood transfusion will undoubtedly provide important insight and direction in transfusion medicine.

• Leukocyte reduction and ultraviolet B irradiation of platelets to prevent alloimmunization and refractoriness to platelet transfusions. The Trial to Reduce Alloimmunization to Platelets Study Group. N Engl J Med. 1997;337:1861-1869.
• Hoeltge GA, Domen RE, Rybicki LA, Schaffer PA. Multiple red cell transfusions and alloimmunization. Experience with 6996 antibodies detected in a total of 159,262 patients from 1985 to 1993. Arch Pathol Lab Med. 1995;119:42-45.
• American Society of Anesthesiologists Practice Guidelines for Perioperative Blood Transfusion and Adjuvant Therapies. http://www.asahq.org/publicationsAndServices/practiceparam.htm#blood (accessed).
• Stramer SL, Hollinger FB, Katz LM, et al. Emerging infectious disease agents and their potential threat to transfusion safety. Transfusion. 2009;49(Suppl 2):1S-29S.
• Kleinman S, Chan P, Robillard P. Risks associated with transfusion of cellular blood components in Canada. Transfus Med Rev. 2003;17:120-162.
• Allain JP, Bianco C, Blajchman MA, et al. Protecting the blood supply from emerging pathogens: the role of pathogen inactivation. Transfus Med Rev. 2005;19:110-126.
• Silliman CC, Kelher M. The role of endothelial activation in the pathogenesis of transfusion-related acute lung injury. Transfusion. 2005;45:109S-116S.
• Vamvakas EC. Possible mechanisms of allogeneic blood transfusion-associated postoperative infection. Transfus Med Rev. 2002;16:144-160.
• 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: 1941-1949.