Adult total blood volume is approximately 2.5 L/m2, 75 mL/kg, or about 5 L in a 70-kg person. Whole blood transfusion would seem ideal to replace acute blood loss; however, storage of whole blood inactivates platelets and other factors. Therefore, whole blood is fractionated to its components for storage and transfusion. Packed red blood cells (PRBCs) are prepared by the centrifugation of whole blood to remove approximately 80% of the plasma; then a preservative solution is added (most commonly, citrate-phosphate-dextrose) with the additional nutrients adenosine, and mannitol (Table 238-1).
The primary reason for PRBC transfusion is to increase oxygen-carrying capacity.1,4 Emergency PRBC transfusion is usually performed for acute blood loss or, occasionally, profound anemia with impaired oxygen delivery. Transfusion thresholds assist the physician in assessing whether PRBC transfusion will benefit the patient. Current evidence is substantial that a restrictive threshold for PRBC transfusion is appropriate for most patients.5,6,7,8,9 In previously healthy adults, transfusion should be considered at hemoglobin concentrations less than 7 grams/dL (70 grams/L), and for patients with sepsis or ischemic heart or brain injury, transfusion should be considered at a hemoglobin concentration less than 8 to 9 grams/dL (80 to 90 grams/L).10,11,12,13 Transfusion threshold values for children may be higher and depend on the etiology of their anemia.14 Some patients with severe sepsis receiving early, goal-directed therapy may benefit from transfusion up to a hemoglobin of 10 grams/dL (100 grams/L) when the central venous oxygen saturation is less than 70%.15
For actively bleeding patients, transfusion is based on clinically estimated blood loss rather than hemoglobin levels, because the fall in measured hemoglobin will lag behind the clinical impact of acute blood loss. A loss of about 30% blood volume (1500 mL in an adult) generally produces symptoms and signs, but young, healthy patients can tolerate this degree of loss when treated with crystalloid. However, patients with chronic illness such as underlying anemia, cardiac diseases, or pacemakers or those on β-blockers or similar medications may not tolerate blood loss. Consider emergency PRBC transfusion for unstable trauma patients based on an inadequate response to an initial 2-L bolus of IV crystalloid or 40 mL/kg in children. The anticipated clinical course also guides the decision to transfuse the patient with acute hemorrhage; the transfusion threshold is lower if the source of bleeding cannot be controlled immediately compared to a patient whose acute hemorrhage has stopped.
Use the minimum amount of PRBCs to accomplish the desired clinical outcome.7,12 A single PRBC unit will raise the hemoglobin by 1 gram/dL (10 grams/L) and hematocrit by 3% in adults. In children, 10 to 15 mL/kg of PRBCs will raise the hematocrit by 6% to 9% and the hemoglobin level by approximately 2 to 3 grams/dL (20 to 30 grams/L).14
One unit of PRBCs, approximately 250 mL in volume, is generally transfused over 1 to 2 hours. PRBCs should be transfused more rapidly in patients with hemodynamic instability. Single-unit PRBC transfusions should not exceed 4 hours to prevent contamination. If a slow transfusion is desired (e.g., in a patient at risk for volume overload), the blood bank should be asked to split a unit so that the first half can be transfused over 4 hours while the second half waits in the blood bank refrigerator. During standard transfusions, the initial infusion rate is slower over the first 30 minutes so that if there is a transfusion reaction, the infusion may be stopped.
PRBC transfusion requires matching the recipient's and donor's red blood cells according to blood type (ABO and Rh) and screening the recipient's plasma for antibodies to the minor red blood cell antigens. Screening is done using a mixture of commercially available red blood cells that have all of the important minor antigens.12 If the screen is positive, then the recipient's plasma is cross-matched against the specific PRBC unit intended for transfusion. Blood type can be determined in approximately 15 minutes, whereas it takes about 45 to 60 minutes to perform a serologic cross-match. If an anti–red blood cell antibody is found in the recipient's plasma, cross-matching may take longer and require additional blood specimens from the patient. For most patients with no antibodies detected on the screen, serologic cross-matching can be foregone and ABO-Rh–compatible PRBC units can be released by the blood bank using a process termed electronic (or computerized) cross-match. This process has computer-based verifications to ensure the patient receives ABO-Rh–compatible blood.12 Electronic cross-matching typically takes 5 minutes or less to perform.
Type O Rh-negative (universal donor) blood may be used in critical circumstances because these transfused red cells do not contain major blood group antigens (A or B). Type O Rh-positive blood may be used if type O Rh-negative is not available, but should be avoided in girls and women of childbearing potential. Approximately 20% of Rh-negative patients transfused with 1 unit of Rh-positive PRBCs will develop anti-Rh(D) antibodies, creating the risk for hemolytic disease of the newborn with subsequent pregnancies. This is usually clinically inconsequential for men or postmenopausal women.
PRBCs may be further treated for specific clinical applications: leukocyte-reduced PRBCs, irradiated PRBCs, washed PRBCs, and frozen PRBCs.16 Leukocyte-reduced PRBCs have 70% to 85% of the white cells removed. Leukocyte-reduced PRBCs are used (1) to decrease the occurrence of nonhemolytic febrile reactions due to cytokines from transfused white cells, (2) to prevent sensitization to human leukocyte antigen antibodies found on white cells in patients who may be eligible for bone marrow transplantation, and (3) to minimize the risk of intracellular virus transmission, such as cytomegalovirus. Leukocytes can be reduced by filtration or other methods before storage of the PRBCs or during transfusion. Irradiation of PRBCs eliminates the capacity of T lymphocytes to proliferate, thereby preventing the donor's T lymphocytes from reacting to the recipient's cells and causing graft-versus-host disease. Irradiated cells are used in transplant patients, neonates, and immunocompromised patients, and with directed donations from relatives of the patient. Washed PRBCs are indicated in patients who have a hypersensitivity to plasma, such as immunoglobulin A deficiency or persistent febrile reactions. For rare blood types, red cells may be frozen and saved for up to 10 years for later use. Freezing red blood cells is more expensive than normal storage, and once thawed, the blood must be washed and transfused within 24 hours.
Massive transfusion is the replacement of one blood volume or approximately 10 units of PRBCs in an adult within a 24-hour period. If only PRBCs are used, platelets and coagulation factors lost or consumed will not be replaced, potentially producing increased bleeding. The military medical experience during the past decade finding good results with using fresh whole blood transfusion for trauma patients17 has led to the concept of incorporating the platelets and plasma in addition to PRBCs to closer mimic whole blood during a massive transfusion.18,19 Studies routinely using platelets and fresh frozen plasma (FFP) with PRBCs during massive transfusion have yielded mixed results on reducing mortality.20,21
Institution-specific massive transfusion protocol is recommended to guide the clinician in correct ordering of the individual products and facilitate release from the blood bank.22 The best ratio of PRBCs to platelets to FFP during a massive transfusion is controversial.23 Some experts advocate a 1:1:1 ratio, although lower ratios of platelets and FFP have been used without clear evidence of inferiority.24 Including cryoprecipitate,20 fibrinogen concentrate,24 or coagulation factor VIIa (recombinant)20,24 during a massive transfusion has produced mixed results, depending on the outcome measured.
If fixed ratios of platelets or FFP are not used, suggested indications for their administration during massive transfusion include the following: (1) when the platelet count is <50,000/mm3(<50 × 109/L), a platelet transfusion is warranted; (2) if the INR is >1.5, FFP may be given; and (3) if the fibrinogen level is <100 milligrams/dL (<1 g/L), it may be replaced with cryoprecipitate or fibrinogen concentrate.
Draw sufficient specimens early in the course from massive transfusion patients because once the patient has received close to one blood volume of transfused products, new blood specimens will contain so much donor blood that it will confuse further cross-matching of subsequent units. Hypothermia is a risk during massive transfusion, so blood and crystalloid should be warmed, in addition to instituting warming measures for the patient. Hypocalcemia from the preservative citrate chelating calcium may occur with a massive transfusion.22
Platelet transfusions are used either prophylactically to prevent bleeding in thrombocytopenia or therapeutically when patients with thrombocytopenia are actively bleeding.25,26 One apheresis-collected, single-donor platelet concentrate is the standard product in developed countries (Table 238-1). Platelets collected from six different donors (a "six pack") can be combined for transfusion but are not recommended because this increases the risk of disease transmission and transfusion reaction.
One apheresis single-donor platelet unit will increase the platelet count by up to 50,000/mm3 (50 × 109/L), an amount sufficient to stop most spontaneous and minor traumatic bleeding. Check platelet levels at 1 and 24 hours after transfusion completion because the response is variable. Failure of platelets to rise appropriately may be due to increased consumption of platelets from an underlying process, active thrombosis due to ongoing hemorrhage, destruction due to platelet antibodies, or sequestration due to hypersplenism. Transfused platelets should survive 3 to 5 days unless there is a platelet-consumptive process.
The decision to transfuse platelets depends on the severity of thrombocytopenia and clinical circumstances (Table 238-3).13,25,26,27 Patients with comorbid conditions, such as infection, fever, medications, and CNS involvement, may be more likely to bleed or be at higher risk if they bleed; therefore, the threshold for platelet transfusion is more liberal.13,28
TABLE 238-3General Indications for Platelet Transfusion ||Download (.pdf) TABLE 238-3 General Indications for Platelet Transfusion
Platelet count <5000/mm3 (<5 × 109/L)
Platelet count <20,000/mm3 (<20 × 109/L) with a coagulation disorder, low-risk procedure, or during outpatient treatment
Platelet count <50,000/mm3 (<50 × 109/L) with active bleeding or invasive procedure within 4 h
Platelet count <100,000/mm3 (<100 × 109/L) with neurologic or cardiac surgery
As part of a massive transfusion protocol
There are no clear recommendations concerning platelet transfusions in patients with nonfunctioning platelets (antiplatelet medications, uremia, von Willebrand's disease, or hyperglobulinemia) and active bleeding. In von Willebrand's disease, normal platelets may help deliver von Willebrand factor to the bleeding site. Conversely, in uremic patients, the transfused platelets may not function any better than native platelets. In these complex cases, consult with a hematologist or transfusion medicine specialist for recommendations.
Relative contraindications to the transfusion of platelets are disorders associated with platelet activation, such as thrombotic thrombocytopenic purpura or heparin-induced thrombocytopenia, in which transfusion may worsen thrombosis. In these conditions, ongoing bleeding or the need to perform procedures may necessitate platelet transfusion in consultation with the appropriate specialist.
Platelet transfusions are usually ABO-type specific because the platelets are bathed in plasma, although a serologic cross-match is usually not done. As a result, patients receiving platelets are subject to many of the same complications described for plasma transfusion. Depending on availability, non–type-specific platelets may sometimes be transfused. This practice is usually avoided in children or patients receiving multiple transfusions because they are at higher risk for complications. Transfusing non–type-specific platelets may also shorten the half-life of the transfused platelets.
As with PRBCs, platelets can be leukocyte reduced or washed. Patients who have had repeated transfusions may become alloimmunized and refractory to platelet transfusion, noted by the lack of expected rise in platelet count after transfusion. Such patients need human leukocyte antigen–matched or cross-matched platelets. Other factors may affect the efficacy of platelet transfusion, including bacterial sepsis in the recipient, antibiotics forming an antigen complex epitope with the platelet, disseminated intravascular coagulation, and splenomegaly.
FRESH FROZEN PLASMA TRANSFUSION
FFP is plasma obtained after the separation of whole blood from erythrocytes and platelets and then frozen within 8 hours of collection.29 FFP takes approximately 20 to 40 minutes to thaw, and this process cannot be sped up through artificial heating. Once thawed, FFP can be transfused up to 5 days later. Trauma centers and other specialty hospitals may keep prethawed units of FFP available.
Transfused FFP should be ABO-type compatible, and Rh compatibility is unnecessary. A common misconception is that type O plasma is the universal FFP donor, as it is for PRBCs. This is not the case, because type O plasma contains antibodies to A and B blood group antigens. Type AB is the universal donor for FFP, and in emergencies, universal donor FFP can be given minutes after thawing. Each unit of FFP has a volume of 200 to 250 mL and contains approximately 1 unit of each coagulation factor and 2 milligrams of fibrinogen per milliliter (Table 238-1). FFP is used for replacement of multiple coagulation deficiencies in cases such as liver failure, warfarin-induced overanticoagulation, disseminated intravascular coagulation, and massive transfusion in bleeding patients, although evidence of benefit is weak (Table 238-4).30 FFP is also used with bleeding due to an individual coagulation factor deficiency when a specific replacement factor is not available. FFP is unlikely to reverse anticoagulation from oral anticoagulants such as dabigatran and rivaroxaban, which are specific inhibitors of thrombin and factor Xa, respectively (see chapter 239, "Thrombotics and Antithrombotics" and the "Prothrombin Complex Concentrate" section, below).31
TABLE 238-4General Indications for Fresh Frozen Plasma Transfusion ||Download (.pdf) TABLE 238-4 General Indications for Fresh Frozen Plasma Transfusion
Reversal of warfarin overanticoagulation (not the primary agent)
Bleeding with multiple coagulation defects
Correction of coagulation defects for which no specific factor is available
As a component of a massive transfusion protocol
As part of plasma exchange when treating thrombotic microangiopathies or neurologic disorders
Response to FFP treatment is monitored by tests of the coagulation system: the prothrombin time, INR, and activated partial thromboplastin time. Using fresh frozen plasma to achieve complete normalization of coagulation studies is neither necessary nor realistic in most circumstances. Clinically adequate hemostasis is generally present with functional coagulation factor levels of 30% to 40% of normal, which corresponds to an INR of about 1.7. Although it is common to administer FFP before a procedure when the INR exceeds 1.5,13,27 there is little evidence to support this practice,32 and it will likely have little effect on patients with an INR less than 1.85.33 If rapid reversal of a vitamin K antagonist coagulopathy is needed, prothrombin complex concentrate or coagulation factor VIIa (recombinant) is faster and more reliable.34
Administering FFP prophylactically to nonbleeding patients is not indicated, and prophylaxis is not needed before procedures in patients with a coagulopathy.32 Procedures such as abdominal paracentesis35 or endoscopy with variceal banding36 can be performed safely with a coagulopathy. If correction is desired, the effect of FFP is transient, dose dependent, and may subject the patient to volume overload.
Other possible indications for FFP include hereditary angioedema if C1 esterase inhibitor is not available (see chapter 14, "Anaphylaxis, Allergies, and Angioedema").37,38 FFP is used during plasma exchange for treatment of diseases such as thrombotic thrombocytopenic purpura and Guillain-Barré syndrome.39,40
For isolated factor deficiencies, specific factor replacement is preferred over FFP for major bleeding, with fresh frozen plasma sometimes used for minor bleeding episodes (Table 238-5).
TABLE 238-5Replacement Therapy for Congenital Factor Deficiencies ||Download (.pdf) TABLE 238-5 Replacement Therapy for Congenital Factor Deficiencies
|Coagulation Factor ||Approximate Incidence* ||Replacement Therapy |
|Factor I (fibrinogen) ||1 per million || |
|Factor II (prothrombin) ||1 per 2 million ||3- or 4-factor PCC for major bleeding |
|Factor V ||1 per million ||FFP |
|Factor VII ||1 per 500,000 || |
4-factor PCC for major bleeding
Coagulation factor VIIa (recombinant)
|Factor VIII† ||1 per 5000–10,000 males || |
Recombinant factor VIII
Desmopressin for mild hemophilia
|Von Willebrand's disease‡ ||Up to 1 per 100 persons ||Desmopressin or factor VIII concentrates (or cryoprecipitate if either unavailable) |
|Factor IX† ||1 per 30,000 males || |
Recombinant factor IX
3- or 4-factor PCC
|Factor X ||1 per million || |
FFP for minor bleeding episodes
3- or 4-factor PCC for major bleeding
|Factor XI‡ || |
3 per 10,000 Ashkenazi Jews
1 per million in general population
|Factor XII ||25 per 1000 ||No bleeding manifestations, replacement not required |
|Factor XIII ||1 per million ||FFP or cryoprecipitate |
The increase in individual coagulation factors seen after FFP infusion varies depending on the specific factor. In general, 1 unit of FFP will increase most coagulation factors by 3% to 5% in a 70-kg adult. Administering 2 units of FFP to an adult (approximately 7 to 8 mL/kg) will increase coagulation factors up to 10%, a clinically inconsequential benefit in most circumstances. For clinically relevant correction of coagulation factor deficiencies, a dose of 15 mL/kg (or 4 units in a 70-kg adult) is often required (Table 238-1). After transfusion, coagulation studies should be repeated and further FFP transfusion guided by the results.
Cryoprecipitate is the cold-insoluble protein fraction of fresh frozen plasma.34 With the development of recombinant factor VIII products for use in hemophilia, the current role for cryoprecipitate is as replacement of fibrinogen.41 Cryoprecipitate may be used in bleeding patients with fibrinogen levels <100 milligrams/dL (< 1 g/L) due to severe liver disease, uremia, disseminated intravascular coagulation, and dilutional coagulopathy, although there is controversy over dosing and efficacy (Table 238-6).42,43 Five units of cryoprecipitate are typically pooled for use, with adults receiving one to three pooled infusions. Cryoprecipitate may also be included in some massive transfusion protocols.20,24
TABLE 238-6General Indications for Cryoprecipitate Transfusion ||Download (.pdf) TABLE 238-6 General Indications for Cryoprecipitate Transfusion
Bleeding with a fibrinogen level of <100 milligrams/dL (< 1 g/L)
Bleeding in some subtypes of von Willebrand's disease that are unresponsive to desmopressin, and factor VIII concentrates are unavailable
Fibrinogen concentrate is derived from pooled human plasma and used to treat bleeding episodes in patients with congenital fibrinogen deficiency.44 Fibrinogen has been investigated for benefit in other hemorrhagic conditions with an observed ability to reduce bleeding and transfusion requirements, but without a measurable effect on mortality.45 Four products are commercially available. The advantages over cryoprecipitate are minimal risk of disease transmission due to viral inactivation, accurate dosing because each vial is assayed for fibrinogen content, a lower volume for infusion, no need for thawing, no requirement of ABO testing and compatibility, and a rapid reconstitution for infusion.
Fibrinogen is dosed according the patient's baseline fibrinogen level, the target level (in most circumstances >150 milligrams/dL),46 volume of distribution, and body weight. If the baseline fibrinogen level is unknown, the initial dose is 70 milligrams/kg. The most common adverse reactions include allergic reactions, fever, chills, nausea, and vomiting.
PROTHROMBIN COMPLEX CONCENTRATE
Prothrombin complex concentrates are blood-derived concentrations of three or four vitamin K–dependent clotting factors: prothrombin and factors VII, IX, and X.47 Some prothrombin complex concentrate formulations may also contain the anticoagulant proteins C, S, and antithrombin, as well as heparin. Three-factor prothrombin complex concentrate is approved for treatment of hemophilia B (factor IX deficiency) (Table 238-2). Four-factor prothrombin complex concentrate is approved for urgent reversal of overanticoagulation from vitamin K antagonists (such as warfarin),48,49 resulting in a more reliable reduction in the elevated INR than three-factor prothrombin complex concentrate.50,51 The four-factor prothrombin complex concentrate dose in this circumstance is administered using factor IX units and adjusted according to the pretreatment INR value. Off-label use of three- or four-factor prothrombin complex concentrate includes bleeding in patients with congenital factor II, IX, or X deficiency.
Prothrombin complex concentrate does not require thawing, does not necessitate ABO-compatibility testing, and does not carry the risk of volume overload, all of which can hinder fresh frozen plasma use. Because prothrombin complex concentrate's effects are transient, vitamin K should usually be co-administered for sustained warfarin reversal. Prothrombin complex concentrate is part of some protocols for reversal of rivaroxaban and dabigatran in the setting of life-threatening bleeding, despite limited evidence for their effectiveness.31,52,53 Thrombosis is the major complication of prothrombin complex concentrate, observed in about 5% of treated patients, although this incidence is not much higher than in similar patients treated with fresh frozen plasma.
COAGULATION FACTOR VIIa (RECOMBINANT)
Coagulation factor VIIa (recombinant) is primarily used for treatment of hemophilia A and B in patients who have developed inhibitor antibodies to factors VIII or IX, respectively. Other uses for this agent have been investigated, such as coagulation support in liver failure, multisystem trauma, intracranial hemorrhage, and postpartum bleeding, but evidence for overall safety and efficacy in these expanded indications is lacking.54,55,56,57,58 The major drawbacks to this product are risk of thrombosis (up to 4% in patients with acquired hemophilia) and the high cost.