Guidelines for red blood cell and plasma transfusion for adults and children Plasma transfusion Plasma for transfusion is prepared by centrifuging anticoagulated whole blood from a single donor followed by storage at or below -18°C. Larger volumes of plasma may be collected using automated apheresis units and similarly stored. A typical unit of plasma has a volume of 200­250 mL if obtained from a whole blood donation or 400­600 mL when obtained by plasmapheresis. Plasma frozen within 8 hours of donation contains at least 0.70 U/mL of factor VIII:C and is referred to as fresh-frozen plasma (FFP). In plasma frozen 8 to 72 hours after collection the concentration of coagulation factors V and VIII:C may be reduced as much as 15%; it is usually referred to as frozen plasma (FP). In Canada, there is some regional variation as to which plasma products (FP and FFP) are available. However, in most therapeutic applications, there is little reason to choose one over the other, and the generic term plasma is used in this document. Cryosupernatant or cryo-poor plasma is the supernatant fluid obtained after preparation of a factor-VIII:C cryoprecipitate from plasma. It contains greatly reduced concentrations of factors V and VIII:C and other large-molecular-weight proteins such as fibrinogen, fibronectin and von Willebrand factor. Excluding viruses such as human cytomegalovirus (CMV) and human T-cell lymphotropic virus type II (HTLV-II), which are transmitted only by infected leukocytes in cellular blood products, plasma carries the same risk of viral transmission per donor exposure as that of red blood cells.71 Virus-inactivated purified or recombinant concentrates of many physiologically important plasma proteins are now commercially available. These products have been shown to be safer and more effective than plasma in correcting specific protein or coagulation factor deficiencies. Hence, the consensus in published guidelines2,5,72­75 is that plasma transfusion should be avoided when a safer and more effective product can be used to achieve the same therapeutic goal (Table 3). This applies specifically to intravascular volume expansion or repletion where crystalloids, synthetic colloids or purified human albumin solutions are preferred (including therapeutic plasmapheresis where plasma should not be used as routine replacement fluid) the correction of hypoalbuminemia or protein malnutrition, where purified human albumin or synthetic amino acid solutions are preferred the correction of hypogammaglobulinemia, where purified human immunoglobulin concentrates are preferred the treatment of hemophilia and von Willebrand's disease where desmopressin (DDAVP) or existing virus-free factor concentrates are preferred the treatment of any other isolated congenital procoagulant or anticoagulant factor deficiency, where virus-inactivated or recombinant factor concentrates are preferred if they exist. In the past, plasma has been used for life-threatening complications of hereditary angioneurotic edema due to deficiency of C1 esterase inhibitor. A concentrate in which viruses have been inactivated now also exists for the treatment of this disorder.76 Plasma transfusion is often used to prevent or stop bleeding in patients with acquired multiple coagulation factor deficiencies, which may occur in the conditions described below. Vitamin K deficiency and warfarin effect When severe bleeding occurs in patients with laboratory evidence of a deficiency of vitamin K-dependent factor (usually a prolonged prothrombin time [PT] or increased international normalized ratio [INR]) or is expected to occur from an emergency surgery or invasive procedure, correction of the hemostatic defect may be obtained by administration of plasma. There is evidence that prothrombin complex concentrates can achieve a more rapid or complete reversal of oral anticoagulation than plasma alone.77 Accordingly, some physicians advocate such concentrates as primary therapy for patients with life or limb-threatening hemorrhagic complications associated with warfarin therapy. In addition, parenteral administration of vitamin K is advised in these circumstances. In nonurgent situations, parenteral administration of synthetic vitamin K alone is usually recommended but may require up to 6 hours to bring about significant correction of the hemostatic defect.78 Severe liver disease Elevated PT, INR or partial thromboplastin time (aPTT) in patients with liver disease is a cause for concern when surgery or percutaneous liver biopsy must be undertaken. Studies79,80 attempting to demonstrate the efficacy of plasma have shown that its ability to correct abnormal coagulation in these patients is often poor or quite variable. Although it is generally agreed that people who are not bleeding or do not face an invasive procedure should not receive plasma merely to correct abnormal coagulation tests, guidelines have recommended the use of plasma in bleeding patients with liver disease.2,73 The addition of partly purified prothrombin complex concentrates has been shown to be more effective than plasma alone,80 but the use of these products as the first-line therapy is usually not recommended because of their associated risk of causing thrombosis and disseminated intravascular coagulation (DIC), particularly in liver disease. Although a threshold PT, INR or aPTT value of 1.5 times normal has been suggested in other guidelines to indicate a need for plasma administration before hemostatic challenges, a clear correlation has not been demonstrated between the occurrence of bleeding and the actual results of coagulation tests obtained prior to surgery or liver biopsy.81,82 Three retrospective studies81,83,84 found that patients with liver disease and mild coagulopathy did not have excess bleeding with invasive procedures (liver biopsy, paracentesis, thoracentesis). These studies reported PT, rather than INR, values. The relation between PT and INR is INR = (PTpatient/PTmean)ISI where PTpatient is the patient's PT, PTmean is the mean of the normal range for PT (measured in a laboratory) and ISI (international sensitivity index) is the measure of the responsiveness of the thromboplastin used to measure the PT to a reduction in vitamin-K dependent coagulation factors. The higher the ISI, the less responsive the thromboplastin and the smaller the increase in PT when testing plasma from patients receiving oral anticoagulants. Using relatively insensitive rabbit thromboplastins (ISI values about 2.081,83 and 2.584), these studies reported no increased bleeding in patients whose PTs were as high as 1.5 times the mean of the normal range, corresponding to INR values of approximately 2.2 and 2.7 respectively (i.e., 1.52.0 and 1.52.5). In formulating our recommendation, we chose the lower value (2.2) as a preliminary basis. However, it should be noted that although INR values are comparable among different thromboplastins when used to test patients receiving oral anticoagulants, they are not comparable in patients with liver disease.85 Moreover, in Canada, most laboratories use relatively sensitive (low ISI) thromboplastins. To determine an appropriate INR recommendation for patients with liver disease, we took into account a Canadian study 85 that compared INR values using different thromboplastins for patients with liver disease. It suggested that for patients with liver disease, an INR of 2.2 derived using a rabbit thromboplastin corresponded to an INR of 2.0 derived using a human placental thomboplastin (Thromborel S [Behring Diagnostics, Kanata, Ont.; ISI about 1.09], widely used in Canada) and to an INR of 1.8 obtained using a recombinant human thromboplastin (Innovin [Dade Diagnostics, Mississauga, Ont.; ISI about 0.86]). Therefore, we adopted a value of 2.0 as the threshold INR in the final formulation of our recommendation for patients with liver disease who will undergo certain invasive procedures (see recommendation 13b). Current evidence neither proves nor disproves the efficacy of plasma transfusion in preventing bleeding during surgery or liver biopsy in patients with liver disease. Finally, it should be recognized that the coagulopathy of liver disease is often complex and may include specific defects that are not reflected by the INR value alone (i.e., coexisting thrombocytopenia or dysfibrinogenemia). Disseminated intravascular coagulation Acute DIC is characterized by the abnormal consumption of coagulation factors and platelets and may lead to thrombocytopenia, hypofibrinogenemia and increased PT, INR or aPTT with uncontrollable bleeding from wound and puncture sites. Retrospective and uncontrolled evidence suggests that the transfusion of plasma, along with other blood components, may be useful in limiting hemorrhage, provided aggressive measures are simultaneously undertaken to overcome the triggering disease.85,86 Plasma transfusion is generally not recommended in the absence of bleeding or in chronic DIC, where it is believed to be ineffective.5 Massive transfusion Massive blood transfusion is defined as the transfusion of more than 10 units of packed red blood cells in adults or the replacement of more than 1 blood volume in 24 hours. The degree and rapidity of blood loss necessitating massive transfusion can be markedly variable; the requirement for blood components and complications of massive transfusion are more likely with greater and more rapid blood loss. Massive blood transfusion may be associated with a number of complications that are usually related to the underlying condition as well as to the large volumes of crystalloid and blood components administered. Some of the adverse effects of massive transfusion may be ameliorated by using blood warming devices and filters and by closely monitoring the patient using clinical, laboratory and hemodynamic measurements.43,44,46,87 Abnormal hemostasis can occur in association with massive transfusion.88­90 A reduction in platelet count after transfusion may occur with the decrease proportional to the number of units of blood transfused. Thrombocytopenia is an important factor linked with microvascular bleeding in this setting. In addition, coagulation factors in the recipient may be both consumed and diluted as a result of the underlying process and the resuscitation, resulting in coagulopathy.91 Abnormal tests of coagulation are common in this setting, but unless they are more than 1.5 times the normal, they are not likely to be associated with microvascular bleeding.92 However, most authors have noted that, as coagulation measures become more abnormal, the occurrence of microvascular bleeding is more likely.92,93 Past recommendations advocated routine transfusion of plasma (i.e., administer 2 units of plasma for every 5 units of red blood cells transfused) to reduce the risk of abnormal bleeding due to factor depletion during massive transfusion.94 However, there is no evidence to support the routine or prophylactic administration of plasma in this scenario.88,94 Several groups90,94 found that prophylactic administration of plasma did not reduce the incidence of hemostatic disorders following massive transfusion. Cardiopulmonary bypass Open heart surgery using cardiopulmonary bypass (CPB) is frequently complicated by postoperative bleeding.95,96 Approximately 3­5% of patients undergo a second operation for excessive bleeding that, at least half the time, is caused by inadequate surgical hemostasis; the remainder are presumed to be the result of a poorly defined defect in hemostasis.95 Acquired platelet dysfunction has historically been considered the major hemostatic defect following CPB.96 It is now also recognized that despite the use of high doses of heparin, there is progressive activation of the coagulation pathway during CPB, as shown by markers of thrombin generation, and concurrent activation of the fibrinolytic system.97­99 Several randomized, double-blind clinical trials have shown that prophylactic use of antifibrinolytic agents, such as aprotinin,100­102 episilon-aminocaproic acid103 or tranexamic acid,104 significantly reduces bleeding in adults undergoing heart surgery. There is some evidence that these agents can also be effective for patients who bleed during the postoperative period.105,106 Despite adequate protamine reversal of heparin after surgery involving CPB, routine coagulation tests (INR, aPTT, thrombin clotting times) usually show above-normal values, primarily because of hemodilution. These abnormal values are commonly used to justify giving plasma to patients who are bleeding following cardiac surgery; unfortunately, there is no correlation between the abnormal values and clinical bleeding.106 Moreover, the concentration of coagulation factors is generally thought to be adequate for hemostasis,95,106,107 and are not raised significantly by plasma transfusion.106 (These routine coagulation tests do not reflect activation of the fibrinolytic system, which occurs to a variable extent among post-CPB patients.108) No studies indicate that prophylactic or therapeutic plasma administration improves hemostasis after CPB. Martinowitz and colleagues109 randomly selected 40 post-CPB patients to receive either the plasma or cell fraction of whole blood; the cell fraction led to higher platelet counts and a decrease in bleeding time, whereas the plasma fraction did not cause any improvement. Evidence110 suggests, but has not been established in comparative clinical trials that antifibrinolytic agents such as episilon-aminocaproic acid, rather than plasma, are reasonable first-line therapy for nonsurgical bleeding after CPB. Thrombotic thrombocytopenic purpura The fatal course of thrombotic thrombocytopenic purpura (TTP) has been dramatically altered since the first report of the empirical use of plasma in a small series of patients in 1977.111 Other uncontrolled studies have confirmed the benefit of plasma therapy in TTP and the closely related adult hemolytic­uremic syndrome. Two well-designed prospective, randomized studies112,113 have shown that plasma exchange is better than plasma infusion in the treatment of TTP. Why plasma works in TTP is not known; however, large multimers of von Willebrand factor are known to contribute to the pathogenesis of TTP.114,115 Unlike plasma, cryosupernatant is devoid of any von Willebrand factor multimers and is effective treatment in patients unresponsive to standard plasma therapy.116,117 Issues specific to plasma transfusion in children Although very few studies address the use of plasma in childhood, experts generally recommend applying the same principles as for adults to guide transfusion decisions for children. This approach seems reasonable for children 6 or more months old because, by 6 months, the concentration of coagulation factors and natural inhibitors of coagulation generally approach adult levels.118 Infants under 6 months of age have relatively lower levels of the vitamin K-dependent coagulation factors (FII, FVII, FIX, FX), the 4 contact factors and the vitamin K-dependent inhibitors of coagulation.118 (PT and aPTT values are correspondingly greater.118) Thus, these factors likely are more rapidly depleted in situations such as acute hemorrhage or DIC, and it may be reasonable to administer plasma to infants less than 6 months old relatively sooner than for older children and adults. Where feasible, transfusion decisions should be guided both by the clinical situation and appropriate laboratory monitoring, although the results of coagulation tests may be more difficult to interpret in young infants. Plasma has long been used to treat congenital deficiencies of hemostatic or anticoagulant proteins; however, more appropriate alternatives now exist for most disorders and, as new treatments are rapidly becoming available, recommendations for treatment are changing. Physicians with special expertise in pediatric hemostasis or thrombosis should supervise the care of children with these disorders. The available alternatives to plasma include: specific factor concentrates, recombinant or derived from human plasma factor concentrates, containing multiple coagulation factors (e.g., prothrombin complex concentrates for the treatment of FII or FX deficiencies) other medical treatments such as synthetic desmopressin (DDAVP) for the most common form of von Willebrand disease (Table 3). Treatment with plasma may occasionally be necessary for deficiencies for which an alternative does not exist. Plasma has also been used to treat severe thrombotic complications in patients with congenital deficiencies of the anticoagulant proteins antithrombin III and protein C, but has now been replaced by diverse protein concentrates. Observational reports and 2 randomized trials119,120 have addressed the role of plasma in treatment of hemolytic­uremic syndrome (HUS) in childhood. Experts have reached consensus that plasma is not indicated for classic childhood HUS, i.e., the syndrome characterized by microangiopathic hemolytic anemia, thrombocytopenia and acute renal failure following diarrhea associated with enterohemorragic E. coli infection. HUS and TTP may be indistinguishable pathologically, and the clinical manifestations of HUS occasionally approach those of TTP. In the absence of definitive studies, and in light of the adult TTP studies,112,113 plasma exchange seems a reasonable consideration in treating children with unusually complicated HUS, particularly those with neurologic complications. Recommendations regarding plasma transfusion Plasma transfusion should be considered for patients with acquired multiple coagulation-factor deficiencies under the following circumstances. Plasma is recommended when serious bleeding has occurred or when preparing for an emergency surgical or invasive procedures in patients with vitamin K deficiency or on warfarin therapy with significantly increased PT, INR or aPTT.      Level of evidence: III Plasma is recommended when there is actual bleeding in patients with liver disease and increased PT, INR or aPTT. Plasma may be administered to prepare for surgery or liver biopsy when the results of PT, INR, aPTT or other appropriate coagulation assay are deemed sufficiently abnormal. Prophylactic plasma transfusion is not indicated for certain invasive procedures (e.g., percutaneous liver biopsy, paracentesis, thoracentesis) in patients with liver disease if their INR is 2.0 or less.      Level of evidence: II Plasma is recommended in patients with acute disseminated intravascular coagulation with active bleeding associated with increased PT, INR or aPTT, provided that the triggering condition can also be treated effectively.      Level of evidence: II Plasma should be administered in the context of massive transfusion (more than 1 blood volume) if there is microvascular bleeding associated with a significantly increased PT, INR or aPTT. If PT, INR or aPTT cannot be measured quickly, plasma may be transfused in an attempt to stop diffuse nonsurgical bleeding.      Level of evidence: II Plasma should be used in the treatment of TTP or adult HUS, followed as soon as possible by daily plasmapheresis with either cryosupernatant or plasma as replacement fluid. Plasma transfusion or exchange is not recommended in the classic form of pediatric HUS.      Level of evidence: I Plasma should be used in patients with acquired deficiencies of a single coagulation factor only when DDAVP or appropriate factor concentrates are ineffective or unavailable. Plasma should be used in these patients only when bleeding has occurred or is reasonably expected to occur from surgery or other invasive procedures. Plasma may be used depending on the specific factor involved.      Level of evidence: III Next: The risks of blood transfusion Previous:   Autologous blood transfusion [Table of Contents] | CMAJ June 1, 1997 (vol 156, no 11) / JAMC le 1er juin 1997 (vol 156, no 11) | CPG Infobase / Infobanque des GPC