Articles on Medical Diseases and Conditions

Tag: Blood Transfusions

  • Other Procedures Relevant To Transfusion Therapy

    Apheresis
    Apheresis is a technique in which blood is withdrawn from a donor, one or more substances are removed from the blood, and the blood is then returned to the donor. Most present-day apheresis equipment is based on separation of blood components by differential centrifugation. Apheresis has two major applications. One is the removal of certain blood components (e.g., platelets) to be used for transfusion into another patient. Apheresis has permitted collection of blood components in greater quantity and more frequently than ordinary phlebotomy. In addition, when all components come from the same donor, the risk of hepatitis is less than if blood from multiple donors is used. This technique is the major source of many rare blood group antibodies and certain blood components. The second major use of apheresis is the direct therapeutic removal of harmful substances or blood components from a patient. The most common application is removal of abnormal proteins from serum by plasmapheresis in patients with the hyperviscosity syndrome (associated with Waldenstцm’s macroglobulinemia or myeloma). Apheresis has also been used to remove immune complexes from patients with various disorders associated with autoimmune disease. However, use of apheresis in many of these conditions is considered experimental.

    Blood transfusion filters

    For many years it has been the custom during blood transfusion to place a filter with a 170-µ pore size between the blood donor bag and the patient. This trapped any blood clots large enough to be visible that might have formed in the donor bag. In the 1960s it was recognized that nonvisible clots or microaggregates of platelets, fibrin, degenerating leukocytes, and other debris frequently were present in stored bank blood and could produce microembolization with pulmonary and cerebral symptoms. The most severe condition was pulmonary insufficiency associated with open-heart surgery or massive blood transfusion for severe trauma. It was found that microfilters with pore sizes of 20-40 µ could trap most of these microaggregates. Some publications advocated using such a filter for every transfusion (the filter can accept 5-10 units of whole blood or packed cells before it must be replaced). Others believe that transfusions limited to one or two units do not subject the lungs to sufficient embolized material to necessitate use of a microfilter. Originally there was a flow rate problem with the 20 µ filters, but newer models have better flow characteristics. A substantial number of platelets are trapped by filter sizes less than 40 µ. However, blood more than 2 days old does not contain viable platelets.

    Blood volume measurement

    Blood volume measurement is useful in certain circumstances: (1) to differentiate anemia due to hemodilution from anemia due to RBC deficiency, (2) to differentiate polycythemia from dehydration, and (3) to quantitate blood volume for replacement or for therapeutic phlebotomy purposes. Blood volume measurement is discussed in this chapter since the most frequent indication for transfusion therapy is to replace depleted blood volume. This most commonly arises in association with surgery or from nonsurgical blood loss, acute or chronic. Immediately after an acute bleeding episode, the Hb and hematocrit values are unchanged (because whole blood has been lost), even though the total blood volume may be greatly reduced, even to the point of circulatory collapse (shock). With the passage of time, extracellular fluid begins to diffuse into the vascular system to partially restore total blood volume. Since the hematocrit is simply the percentage of RBCs compared with total blood volume (total blood volume being the RBC mass plus the plasma volume), this dilution of the blood by extracellular fluid means that the hematocrit value eventually decreases, even while total blood volume is being increased (by extracellular fluid increasing plasma volume). Hemodilution (and thus the drop in hematocrit) may be hastened if the patient has been receiving intravenous (IV) fluids. Serial hematocrit determinations (once every 2-4 hours) may thus be used as a rough indication of blood volume changes. It usually takes at least 2 hours after an acute bleeding episode for a significant drop in hematocrit value to be demonstrated. Sometimes it takes longer, even as long as 6-12 hours. The larger the extracellular blood loss, the sooner a significant hematocrit value change (> 2%) is likely to appear.

    Previous dehydration, a low plasma protein level, or both will tend to delay a hematocrit drop. Besides the uncertainty introduced by time lag, other conditions may affect the hematocrit value and thus influence its interpretation as a parameter of blood volume. Anemias due to RBC hemolysis or hematopoietic factor deficiencies such as iron may decrease RBC mass without decreasing plasma or total blood volume. Similarly, plasma volume may be altered in many situations involving fluid and electrolyte imbalance without changing the RBC mass. Obviously, a need exists for accurate methods of measuring blood volume.

    Blood volume methods. The first widely used direct blood volume measurement technique was Evan’s Blue dye (T-1824). After IV injection of the dye, the amount of dilution produced by the patient’s plasma was measured, and from this the plasma volume was calculated. At present, radioisotopes are the procedure of choice. These methods also are based on the dilution principle. Chromium 51 can be used to tag RBCs; a measured amount of the tagged RBCs is then injected into the patient. After equilibration for 15 minutes, a blood sample is obtained and radioactivity is measured. Since the tagged RBCs have mixed with the patient’s RBCs, comparison of the radioactivity in the patient’s RBCs with the original isotope specimen that was injected reveals the amount that the original isotope specimen has been diluted by the patient’s RBCs; thus, the patient’s total RBC mass (RBC volume) may be calculated. If the RBC mass is known, the plasma volume and total blood volume may be derived using the hematocrit value of the patient’s blood. Another widely used method uses serum albumin labeled with radioactive iodine (RISA). This substance circulates in the plasma along with the other plasma proteins. Again, a measured amount is injected, a blood sample is withdrawn after a short period of equilibration, and the dilution of the original injected specimen is determined by counting the radioactivity of the patient’s plasma. Plasma volume is provided by RISA measurement; RBC volume must be calculated using the patient’s hematocrit value. There is no doubt that isotope techniques are much more accurate than the hematocrit value for estimating blood volume. Nevertheless, there are certain limitations to isotope techniques in general and specific limitations to both 51Cr and RISA (see box below).

    Assay problems. The main drawback of blood volume techniques is the lack of satisfactory reference values. Attempts have been made to establish reference values for males and females in terms of height, weight, surface area, or lean body mass. Unfortunately, when one tries to apply these formulas to an individual patient, there is never any guarantee that the patient fits whatever category of normal persons that the formula was calculated from. The only way to be certain is to have a blood volume measurement from a time when the patient was healthy or before the bleeding episode. Unfortunately, this information is usually not available. Another drawback is the fact that no dilution method can detect bleeding that is going on during the test itself. This is so because whole blood lost during the test contains isotope in the same proportion as the blood remaining in the vascular system (a diminished isotope dose in a

    Factors That Can Adversely Affect the Conditions Necessary for Accurate Blood Volume Results
    1. All isotope must be delivered into the fluid volume being measured:
    Extravasation outside vein (falsely increased result)

    2.Need uniform mixing of isotope throughout fluid volume being measured:

    Capillary sequestration in shock
    Slow mixing in congestive heart failure
    Focal sequestration in splenomegaly

    3. Need uniform mixing of isotope throughout fluid volume being measured:
    Male vs. female
    Obesity or malnourishment

    diminished blood volume) in contrast to the situation that would prevail if bleeding were not going on, when the entire isotope dose would remain in a diminished blood volume. Fortunately, such active bleeding would have to be severe for test results to be materially affected. Another problem is dependence on hematocrit value when results for RISA are used to calculate RBC mass or data from 51Cr are used to obtain plasma volume. It is well established that hematocrit values from different body areas or different-sized vessels can vary considerably, and disease may accentuate this variation. The venous hematocrit value may therefore not be representative of the average vascular hematocrit value.

    Single-tag versus double-tag methods. All authorities in the field agree that blood volume determination combining independent measurement of RBC mass by 51Cr and plasma volume by RISA is more accurate than the use of either isotope alone. Nevertheless, because both isotopes must be counted separately with special equipment, most laboratories use only a single isotope technique. Most authorities concede that 51Cr has a slightly better overall accuracy than RISA, although some dispute this strongly. However, most 51Cr techniques call for an extra venipuncture to obtain RBCs from the patient for tagging, plus an extra 30-minute wait while the actual tagging of the cells takes place. In addition, tagged RBCs (and their radioactivity) remain in the circulation during the life span of the cells. The main advantages of RISA are the need for one less venipuncture than 51Cr and the fact that RISA procedures can be done in less than half the time of 51Cr studies. The error with RISA blood volume has reached 300 ml in some studies, although most determinations come much closer to double-isotope results. In patients with markedly increased vascular permeability, significant quantities of RISA may be lost from blood vessels during the test, which may produce additional error. Severe edema is an example of such a situation. Nevertheless, even under adverse conditions, RISA (and 51Cr) represents a decided advance over hematocrit values for estimating blood volume.

    Central venous pressure measurement. Central venous pressure (CVP) is frequently used as an estimate of blood volume status. However, CVP is affected by cardiac function and by vascular resistance as well as by blood volume. Circumstances in which the CVP does not accurately reflect the relationship between blood quantity and vascular capacity include pulmonary hypertension (emphysema, embolization, mitral stenosis), left ventricular failure, and technical artifacts due to defects in catheter placement and maintenance.

  • Red Blood Cell Substitutes

    Attempts have been made to find an RBC substitute that will not require crossmatching, can be stored conveniently for long periods of time, can be excreted or metabolized in a reasonable period of time, is relatively nontoxic, and can provide an adequate delivery of oxygen to body tissues and return carbon dioxide to the lungs. Thus far, no perfect answer has emerged. The current two leading candidates have been hemoglobin solutions (free of RBC stroma) and synthetic substances, of which the most promising to date are fluorocarbon compounds. However, major problems still remain. Free hemoglobin can precipitate in the tubules of the kidney or alter renal function. Another difficulty involves a generalized and a coronary artery vasoconstrictor effect. Also, free Hb can interfere with some biochemical tests. Fluorocarbons usually must be oxygenated for maximum effectiveness, most commonly by having the patient breathe 100% oxygen. Elimination of fluorocarbons from the body is fairly rapid (the half-life is about 24 hours), which sometimes would necessitate continued administration. Thus far, none of these blood substitute preparations has proved entirely successful. However, several new preparations are now in clinical trials.

  • Albumin and Purified Plasma Protein Fraction

    As mentioned in the earlier discussion about plasma, 5% albumin can be used instead of plasma to restore colloid oncotic pressure, mainly in hypovolemic shock due to massive acute blood loss or extensive burns. About 40% of body albumin is intravascular, with the remainder being in extracellular fluid. In a normal-sized person, 500 ml of blood contains about 11 gm of albumin, which is about 3.5% of total body albumin and about 70% of the albumin synthesized daily by the liver. Therefore, the albumin lost in three or four units of whole blood would be replaced in about 3 days of normal production. The AABB and other investigators believe that albumin has been overused in bleeding persons. They discourage use of albumin infusions in persons with hypoalbuminemia due to chronic liver disease or albumin loss through the kidneys or gastrointestinal tract on the grounds that such therapy does not alter the underlying disease and has only a very short-term effect. They are also critical of therapeutic albumin in hypoalbuminemia due to nutritional deficiency, which should be treated with parenteral hyperalimentation or other nutritional therapy. Purified plasma protein fraction (PPPF) can be used in most cases instead of albumin but has few advantages. It is not recommended when rapid infusion or large PPPF volumes are needed since it may have a hypotensive effect under these conditions. Albumin and PPPF do not transmit viral hepatitis because they are pasteurized.

  • Fibrinogen

    Fibrinogen is a blood fraction that is essential for clotting. It is decreased in two ways, both relatively uncommon: (1) by intravascular deposition of fibrin in the form of small clots (DIC) and (2) by inactivation in the presence of primary fibrinolysin. Fibrinogen concentrates used to be prepared commercially but are no longer available due to the considerable risk of hepatitis. Cryoprecipitate contains 150 mg or more of fibrinogen per bag and is the most commonly used source of therapeutic fibrinogen.

  • Special Coagulation Factor Materials

    Factor VIII concentrate differs from cryoprecipitate in several ways. It is prepared from a pool of donors and is lyophilized. The two major advantages are that factor VIII activity has been assayed by the manufacturer for each concentrate bag and that treatment with solvent-detergent mixtures or adequate heat (when coupled with donor testing) can virtually eliminate infectivity by hepatitis B and C virus and HIV-I. Some other viral infections can still be transmitted. Recombinant factor VIII will eliminate infection problems when it becomes widely available. Factor VIII concentrate is not reliable for therapy of von Willebrand’s disease.

    Hemophilia A patients who develop factor VIII inhibitors and become refractory to ordinary factor VIII therapy have been treated with some success using a commercial product known as prothrombin complex concentrate. This was originally developed for use in factor IX deficiency. A newer product known as activated prothrombin complex concentrate (or antiinhibitor coagulant complex) is said to be more effective.

    Factor IX concentrate (prothrombin complex concentrate) is available for therapy of factor IX deficiency (hemophilia B; Christmas disease). Treatment of this product by solvent-detergent mixture or adequate heat reduces risk of infection by viruses surrounded by a lipid envelope such as HIV-1, the hepatitis viruses, and CMV. However, there have been cases of DIC, thrombosis and embolization, and acute myocardial infarction.

    Factor XI deficiency is usually treated with fresh frozen plasma, although some factor XI is present in all blood or blood products.

    Antithrombin III (AT-III) deficiency or protein C deficiency (Chapter 8) are usually treated with fresh frozen plasma, although AT-III activity is fairly stable in plasma that is present in any blood product. Cryoprecipitate also contains AT-III, but the amount per unit is not large. Fibronectin is a plasma glycoprotein thought to play a role in phagocytosis by acting as a nonspecific opsonin. Fibronectin is said to be fairly stable in plasma contained in any blood product, but the usually recommended therapeutic source is cryoprecipitate.

  • Cryoprecipitate

    Cryoprecipitate is prepared from fresh frozen plasma; it is the material that does not become totally liquid when fresh frozen plasma is slowly thawed and the major part has liquefied. Cryoprecipitate contains about 50% of the original factor VIII and von Willebrand factor activity, about 20%-40% of the fibrinogen and some factor XIII. The major advantage over fresh frozen plasma is reduction in the total volume of fluid that is transfused. Transfusion of one unit of cryoprecipitate carries about the same risk of hepatitis or HIV-I infection as transfusion of one unit of fresh frozen plasma or one unit of whole blood. Cryoprecipitate can also transmit infection by cytomegalovirus and Epstein-Barr virus. The amount of factor VIII activity in each unit of cryoprecipitate is reported to be highly variable (from approximately 25-150 units), although each bag is supposed to contain at least 80 units. Each unit contains about 150 mg of fibrinogen. Cryoprecipitate is useful in treating von Willebrand’s disease as well as hemophilia A. Some reports suggest that cryoprecipitate, for unknown reasons, has some corrective activity in uremic patients with a bleeding tendency. If many units of cryoprecipitate are administered, some believe that ABO-compatible units should be used, since cryoprecipitate is a plasma preparation and could contain anti-A and anti-B antibodies. These antibodies could produce a positive direct Coombs’ test result on recipient RBC, which would be confusing. However, the need for ABO compatibility has been disputed. Large amounts of cryoprecipitate might elevate plasma fibrinogen levels and through this mechanism could produce a temporary elevation of the RBC sedimentation rate.

  • Plasma

    Plasma itself may be either stored or fresh frozen. Stored plasma until the early 1970s was the treatment of choice for blood volume depletion in burns and proved very useful as an initial temporary measure in hemorrhagic shock while whole blood was being typed and crossmatched. It was also useful in some cases of shock not due to hemorrhage. Stored plasma may be either from single donors, in which case it must be crossmatched before transfusion; or more commonly, from a pool of many donors. Pooled plasma dilutes any dangerous antibodies present in any one of the component plasma units, so that pooled plasma may be given without crossmatch. For many years it was thought that viral hepatitis in plasma would be inactivated after storage for 6 months at room temperature. For this reason, pooled stored plasma was widely used. In 1968 a study reported a 10% incidence of subclinical hepatitis even after storage for prescribed time periods. The National Research Council Committee On Plasma And Plasma Substitutes then recommended that 5% albumin solution be used instead of plasma whenever possible.

    Fresh frozen plasma. Fresh frozen plasma is prepared from fresh whole blood within 6 hours after collection. Fresh frozen plasma used to be the treatment of choice for coagulation factor deficiencies such as factor VIII (hemophilia A), von Willebrand’s disease, or fibrinogen. Since large volumes were often required for hemophilia A, methods were devised to concentrate factor VIII. Concentrated factor VIII solutions and cryoprecipitate are both available commercially and have largely superseded the use of fresh frozen plasma in hemophilia A. All of these products, unfortunately, may transmit infection by viruses, including the hepatitis viruses and HIV-I. Heat treatment in conjunction with donor testing has nearly eliminated HIV-I infectivity of factor VIII concentrate and greatly reduced hepatitis B virus infections.

  • Granulocytes (Neutrophils)

    WBC transfusions are being used for treatment of infections not responding to antibiotics in patients with severe leukopenia due to acute leukemia or bone marrow depression. The AABB recommends 500 granulocytes/mm3 (or per microliter) as the cutoff point defining severe leukopenia. Clinical improvement has been reported in some of these patients but not all. Most large blood banks have the equipment to offer granulocyte transfusion as a routine procedure. The granulocytes are usually collected by apheresis methods and stored at room temperature. The AABB recommends that granulocytes be transfused within 24 hours after collection. According to the AABB, a daily dose of at least 1 x 1010 functional granulocytes appears necessary. Each granulocyte concentrate dose also contains 3 x 1011 platelets. The same recommendations regarding irradiation noted above for platelet concentrates also applies to granulocytes.

  • Platelets

    Platelets are supplied in units that are equivalent to the number of platelets in one unit of whole blood (about 5.5 x 1010). These are obtained as single platelet units from random-donor whole blood units or as multiple platelet units from a single donor by means of platelet apheresis. Platelets are stored at room temperature up to 5 (sometimes 7) days. One single-donor unit ordinarily is expected to raise the platelet count 7,000-11,000/mm3 per square meter of body surface (equivalent to 5,000-10,000/mm3 [10-25 x 109/L] in an average-size adult). It has been suggested that platelets should be transfused as soon as possible after collection to retain maximum function and that infusions should be rapid (10 ml/min). Some also suggest a microaggregate filter (usually 40-µ size) if platelet storage has been more than 24 hours from time of collection. Platelet concentrates prepared by ordinary methods usually contain some plasma (about 50 ml/platelet unit) and some WBCs. Platelets from donors who have taken aspirin within 72 hours of donation may have deficient platelet function.

    Platelet antigens. Platelets themselves contain various antigens, including ABO and HLA. Single-donor platelets can be transfused without typing (similar to random-donor platelets) or can be typed (usually ABO and HLA) for recipient compatibility before administration. Platelet ABO incompatibility usually has only a minor effect on donor platelet survival. HLA incompatibility may become a more serious problem. After repeated transfusions with random-donor nonmatched platelets about 50%-70% of patients (range, 10%-100%) become sensitized to platelet HLA antigens (or sometimes, to platelet-specific antigens), and these patients usually become refractory to additional platelet transfusions. Most can still be transfused with HLA-matched platelets. Siblings have the best chance of providing HLA-compatible blood, although nonsiblings often match adequately. Some institutions administer only HLA-matched platelets when long-term need for platelets is anticipated. However, the AABB currently recommends that patients receive non-HLA-matched platelets initially, with HLA-matched platelets reserved for those who become refractory. Some reports suggest that leukocyte depletion helps delay platelet sensitization. Some investigators perform a 1-hour and a 24-hour platelet count after transfusion. Low 1-hour recovery is said to suggest platelet antigen sensitization. When conditions that decrease platelet survival (e.g., fever, infection, or disseminated intravascular coagulation [DIC]) are present, the 1-hour count shows normal recovery but the 24-hour count is low. One report suggests that a platelet count 10 minutes after transfusion provides the same platelet values as the 1-hour count.

    Indications for platelet transfusion. Platelet transfusion can be therapeutic or prophylactic. Therapeutic transfusions are indicated when severe acute bleeding occurs and the patient is severely thrombocytopenic (<50,000 platelets/mm3 or µL). When the patient has thrombocytopenia but has very minor bleeding or is not bleeding, the question of prophylactic platelet transfusion may arise. The decision is usually based on the degree of risk and the type of disorder being treated. Until 1993, most authorities considered patients to be high-risk if their platelet counts were less than 20,000/mm3 or µL (some used a cutoff value of 10,000/mm3); moderate-risk patients (transfusion only if clinically indicated) were those with counts of 20,000-50,000mm3; and low-risk patients included those with counts over 50,000mm3. Based on more recent studies, investigators are now proposing 5,000mm3 (or µL) as the threshold “trigger” value for prophylactic platelet transfusion (rather than 20,000 or even 10,000). The bleeding time has also been used as a guide to therapy; a bleeding time value of less than twice the upper limit of the reference range would not ordinarily need platelet transfusion. In patients with conditions that require multiple platelet transfusions over relatively long time periods, an additional consideration is the probability of developing antiplatelet antibodies that would interfere with therapy if actual bleeding developed later. In idiopathic thrombocytopenic purpura, antiplatelet antibodies that destroy donor platelets are already present, so that transfusion is useless unless the patient is actively bleeding. In drug-induced thrombocytopenia, transfusion is useless if the drug is still being given; after medication has been discontinued, transfusion can be helpful since a normal platelet count will usually return in about 1 week and transfused platelets survive about 1 week.

    Platelet concentrates given to bone marrow transplant patients, severely immunodeficient or immunosuppressed patients, and blood relatives of the donor should have the platelet unit irradiated with at least 25 Gy to avoid graft-vs.-host disease.

  • Leukocyte-Poor Blood Products

    There are several indications for leukocyte removal. The most frequent reason is development of an immune reaction to antigens on “foreign” transfused leukocytes that constitutes the great majority of febrile nonhemolytic transfusion reactions and by far the most common overall transfusion reaction, especially in multitransfused patients. Second, removal of leukocytes helps prevent microaggregates (miniclots) of WBC, fibrin, platelets, and RBC debris that form in stored blood. These microaggregates have been implicated as one cause for adult respiratory distress syndrome. Another indication is to prevent immunization and subsequent reaction to class II (D-locus) HLA antigens present on lymphocytes. This may also cause nonhemolytic febrile reactions, but assumes greatest importance in patients who may need tissue transplants. Another indication is to help prevent or delay sensitization to platelets, which carry HLA class I (A, B, C loci) antigens. Leukocyte removal also helps prevent transmission of cytomegalovirus that infects lymphocytes. This is most important in neonates and in immunocompromised patients. However, the most common use of leukocyte removal is in multiply transfused patients with febrile reactions.

    Leukocyte removal is most easily accomplished in older blood units (in which some degree of spontaneous leukocyte microaggregation takes place). The original and still used method is centrifugation of the blood unit and removal of the leukocyte-rich layer (“buffy coat”). This removes about 70%-80% of the leukocytes. If the remaining blood is passed through a 20-40 micron depth-type microaggregate filter, this increases WBC removal to 90%-94%. Special granulocyte filters are now commercially available that can remove as much as 99.9% of the leucocytes; these filters can be used at the patient’s bedside. About 25% of the RBCs are lost during special leukocyte filtration. Age of the blood does not matter. When the object is to prevent immunization, there are differences in performance between commercially available filters.

    Other methods of leukocyte removal are the washing of red cells or as a side effect of preparing frozen RBCs. These are discussed separately since these methods accomplish other purposes besides only leukocyte removal.

    Washed red blood cells. Washed RBCs are packed cells that have received several washes with saline, followed by centrifugation. This removes more than 90% of the WBCs with most of the platelets and plasma proteins. About 10%-20% of the RBCs are lost. Indications for washed cells are relatively few. Cell washing removes donor antibodies and is useful in IgA immune reactions. Washed RBCs are traditionally used in patients with paroxysmal nocturnal hemoglobinuria. Washed cell methods remove 70%-93% of total WBCs and are reported to cause fewer leukoagglutinin reactions than ordinary centrifuged leukocyte-poor RBCs. However, current leukocyte filters remove more leukocytes than cell washing. The saline wash process reduces but does not completely eliminate the risk of viral hepatitis.

    Washed RBCs must be discarded if not transfused within 24 hours after preparation (washing). No more units should be ordered than definitely will be used.

    Frozen red blood cells. Fresh citrate-anticoagulated RBCs may have their storage life greatly prolonged by freezing. Glycerol is added to packed RBCs to protect them during freezing; this substance prevents intracellular fluid from becoming ice. The blood is slowly frozen and is maintained at below-zero temperatures until ready for use. Thereafter it must be thawed, after which the glycerol is removed (to avoid osmotic hemolysis), and the cells are suspended in saline. This technique will maintain packed RBCs for up to 5 years. Advantages to the blood bank include the ability to maintain a much larger inventory of blood types without fear of outdating and better control over temporary fluctuations in donor supply or recipient demand. It also permits stockpiling of rare RBC antigen types. Advantages to the patient include approximately 95% elimination of leukocytes, platelets, and plasma proteins, thus removing sources of immunization and febrile reactions; removal of most potassium, ammonium, and citrate, three substances that might be undesirable in large quantities; and considerable reduction of risk for viral hepatitis. Unfortunately, risk of viral hepatitis or HIV-I (HTLV-III) infection is not completely eliminated. Transfused frozen RBCs contain less plasma than washed RBCs.

    Disadvantages include considerably greater cost; significant time lost in thawing and preparing the RBCs (1 hour or more); equipment limitation on the number of units that can be prepared simultaneously, which might cause difficulty in emergencies; the fact that once thawed, cells must be used within 24 hours (by current AABB rules; some data suggest this period could be extended to 7 days using a special plastic bag closed system to remove the glycerine); and the presence of variable amounts of free Hb, which might be of sufficient quantity to be troublesome if tests are needed for possible hemolytic transfusion reaction.

    Frozen RBCs are becoming more widely available as larger blood banks acquire the necessary equipment. In addition to use in persons with rare blood types, current practice favors their use in some circumstances with greater risk of certain types of immunohematologic sensitization. These include circumstances in which reactions to plasma proteins may occur (e.g., IgA-deficient persons or persons already sensitized to IgA or other serum proteins). Some advocate frozen RBCs as therapy for paroxysmal nocturnal hemoglobinuria. The new leukocyte filters are more effective in removing WBCs than are frozen RBCs. RBCs containing sickle Hb are difficult to deglycerolize and require special techniques.

    Irradiated red blood cells. Donor lymphocytes can induce the graft-vs.-host reaction (Chapter 11) in recipients who are severely immunocompromised or are blood relatives of the donor. Gamma radiation (at doses of 25 Gy, equivalent to 2500 rads) affects lymphocytes but not RBCs, granulocytes, or platelets. Such radiation substantially reduces the risk of graft-vs.-host reaction. Radiation of this magnitude will not inactivate viral hepatitis or HIV-I. If blood product irradiation is necessary, all blood products that contain any lymphocytes must be irradiated; these include whole blood, the various RBC preparations (including frozen RBCs), platelet concentrates, and granulocyte concentrates.