key: cord-026021-b8vtmr9h authors: Hohenhaus, Ann E. title: Blood Transfusion and Blood Substitutes date: 2011-06-22 journal: Fluid, Electrolyte, and Acid-Base Disorders in Small Animal Practice DOI: 10.1016/b978-1-4377-0654-3.00031-7 sha: doc_id: 26021 cord_uid: b8vtmr9h nan Ann E. Hohenhaus Blood transfusions have many things in common with fluid therapy. Like crystalloid and colloid solutions, blood products are not used to treat disease; they are supportive therapies given to correct deficiencies in the patient until the underlying disease process can be treated. For example, a red blood cell transfusion is given to replace red blood cells lost as a result of a traumatic laceration. The transfusion of red blood cells increases the oxygen-carrying capacity of the blood, allowing for surgical repair of the laceration; it is not the primary treatment for hemorrhage. Likewise, sodium chloride is used to replace sodium, chloride, and water in a dehydrated patient with hypoadrenocorticism until adrenal hormones can be replaced. The use of both blood transfusions and fluid therapy must be carefully assessed before inclusion in a patient's treatment plan, and the veterinarian should evaluate the risk/benefit ratio for each patient. Volume overload, electrolyte disturbances, and transmission of infection can occur from administration of pathogen-contaminated blood products or fluids. 31, 66, 130 Despite the potential negative effects of transfusion, most veterinarians view it as lifesaving therapy, allowing the transfusion recipient to receive other necessary treatments such as surgery, chemotherapy, or medical care. 56 Three major differences exist between the more commonly used fluids and blood products. The differences between crystalloid or colloid solutions and blood products are their immunogenicity, availability, and cost. The immunogenicity of blood products stems from the proteins and cellular material in the blood. Because crystalloid solutions lack proteins and cellular material, they are not considered immunogenic; however, certain colloid solutions such as hydroxyethyl starch have been reported to cause acute anaphylaxis in rare instances in humans. 96 The mechanism of this reaction is unknown. Limited availability differentiates blood products from crystalloid and colloid solutions. Crystalloid and colloid solutions are readily available because they can be manufactured according to market demand. Only a living animal can produce blood, and the donor's physiologic capability limits production. The small number of commercial canine and feline blood banks providing a convenient source of blood for the veterinary practitioner further limits availability of blood for transfusion (Box 24-1). Furthermore, blood products require a more regulated storage environment and have a significantly shorter shelf life than crystalloid or colloid solutions, making blood a less convenient product to store and use in a veterinary hospital. Nearly 20 years ago, veterinarians estimated costs associated with transfusions, but an exact analysis of cost is lacking. In 1992, the estimated cost of a 500-mL whole blood transfusion ranged from $25 to more than $300. 56 The cost of 500 mL of lactated Ringer's solution is about $1. Despite the fact that the first documented transfusion was given to a dog in 1665 by Richard Lower at Oxford University, veterinary transfusion medicine scientifically and technologically lags behind its counterpart in human medicine. 76 Information in this chapter is based on veterinary studies whenever possible. When none is available, currently accepted guidelines from human medicine will be applied to the veterinary patient. The purpose of this chapter is to provide the reader with the following: 1. A basic understanding of the theory of blood component therapy 2. Information on the technical aspects of obtaining blood for transfusion 3. Suggestions for the administration and monitoring of transfusions 4. A description of the clinical applications of a veterinary blood substitute Blood is the body's largest connective tissue. When collected from the donor, it contains all the elements of blood: red blood cells, white blood cells, platelets, coagulation factors, immunoglobulins, and albumin. Whole blood can be transfused into the recipient as it is collected from the donor, but it is neither a specific therapy nor an economical use of blood. The optimal method of preservation of blood for transfusion is to separate whole blood into its component parts. Appropriate use of blood components not only conserves the products but also allows the most specific and safe product to be administered for each patient. When blood components are used instead of whole blood for transfusion, two dogs can benefit from 1 unit of whole blood. A plasma transfusion counteracts the anticoagulant effects of rodenticide intoxication in one dog, and red blood cells from the same donor provides enhanced oxygen-carrying capacity in a second, anemic dog. Component transfusions also have been used in cats, but preparation of components is more difficult because of the small volume of blood collected from donor cats. 23, 53, 68, 99 Production of components is not feasible for most veterinary practices. Most will purchase their blood inventory because they lack the time and equipment to recruit donors, and collect and process whole blood into components. Blood components predominate in the inventory of commercial blood banks, requiring veterinarians to become familiar with their usage. The technical aspects of component production are not included in this chapter but can be found elsewhere. 85, 99 A brief summary follows. Preparation of blood components from whole blood requires that the blood from the donor be collected into the anticoagulant-containing bag of a multibag plastic blood collection system. The whole blood then is separated into packed red blood cells (PRBCs) and plasma by differential centrifugation in a refrigerated blood bank centrifuge, and the plasma is transferred into one or more of the attached satellite bags via the sterile tubing linking the bags. The bags are separated, and PRBCs are stored in a refrigerator and plasma is stored in a freezer. Blood collected into glass bottles is not amenable to centrifugation and cannot be processed into components. Additionally, storage of canine blood in a glass bottle results in lower levels of 2,3-diphosphoglycerate and adenosine triphosphate (ATP) than blood stored in plastic bags; consequently, plastic bags are the preferred storage container for blood. 28 The most commonly used blood products, their indications, and suggested dosages are described later. The dosage of a blood product depends on the physical state of the patient and the response of the patient to the treatment: in essence, the treatment is "to effect." Whole blood is the blood collected from the donor plus the anticoagulant. In veterinary medicine, no standards have been established for the volume of blood that constitutes 1 unit. When a human blood collection system is used for dogs, 450 AE 45 mL of blood is collected and combined with 63 mL of anticoagulant, and often is designated as 1 unit. Whole blood contains red blood cells, clotting factors, proteins, and platelets and is the product most commonly transfused into dogs and cats. 56 Once whole blood is refrigerated, the white blood cells and platelets become nonfunctional. As a starting point, the dosage for whole blood is 10 to 22 mL/kg. PRBCs are the cells and the small amount of plasma and anticoagulant that remains after the plasma is removed from 1 unit of whole blood. If 450 mL of blood are collected, the volume of PRBCs obtained is approximately 200 mL. Because the plasma has been removed, the total volume transfused is less than 1 unit of whole blood but contains the same oxygen-carrying capacity as 450 mL of whole blood. In cats, the increase in packed cell volume (PCV) after transfusion of 1 unit of PRBCs has been shown to be equivalent to the increase after transfusion of 1 unit of whole blood. 68 PRBCs are used only to treat clinically symptomatic anemia because they do not contain platelets or clotting factors. Red blood cell transfusions are administered to cats for a variety of reasons. Data on 126 cats administered whole blood or PRBCs indicated 52% were transfused for blood loss anemia, 38% for erythropoietic failure, and 10% for hemolytic anemia. 68 Similar reasons for transfusion of cats have been reported elsewhere. 22, 126 Dogs more commonly are transfused for blood loss anemia (70%) with 14% to 22% being transfused for hemolytic anemia and 8% to 14% for erythropoietic failure. 17, 64 The initial dosage of PRBCs is 6 to 10 mL/kg, and transfusion is continued until the clinical signs of anemia resolve. Fresh frozen plasma is the plasma obtained from whole blood plus the anticoagulant solution, which is frozen within 8 hours of collection. When whole blood is centrifuged to produce plasma and PRBCs, the anticoagulant segregates with the plasma fraction. Fresh frozen plasma contains all clotting factors, which, if frozen at À30 C in a blood bank freezer, maintains activity for 12 months. 121 Fresh frozen plasma maintained in an upright freezer at À20 C maintains clotting factor activity for 6 months. When frozen, the plastic storage bag becomes brittle and if not carefully handled can crack, rendering the plasma unusable. For this reason, plasma is stored in special boxes to protect the plastic bag and must be handled carefully before transfusion. If plasma is thawed, and not transfused, it can be refrozen within 1 hour of thawing without loss of coagulation factor activity. 131 Fresh frozen plasma has been used to treat a wide variety of clinical patients. A retrospective analysis of fresh frozen plasma usage in dogs identified multiple indications for administration of fresh frozen plasma, including replacement of coagulation factors, albumin, a 2 -macroglobulin, and immunoglobulin despite the recommendation that fresh frozen plasma should not be used as a source of albumin, for volume expansion, or nutritional support. 75, 87 Calculations suggest 45 mL/kg of plasma would need to be given to increase albumin serum concentration by 1 g/dL. 120 In cases of coagulation factor deficiencies, plasma should be given to effect (i.e., until active bleeding ceases). 70 For the treatment of coagulation disorders, 6 to 10 mL/kg is the recommended starting dosage. Multiple doses may be required to control bleeding because of the short half-life of clotting factors, especially in patients with disseminated intravascular coagulation. Normalization of previously abnormal coagulation tests can be used as a guide for discontinuation of plasma therapy. Cryoprecipitate is prepared by thawing fresh frozen plasma at 0 C to 6 C. A white precipitate forms, the liquid plasma is removed after centrifugation, and both aliquots are refrozen. The cryoprecipitate is a concentrated source of von Willebrand's factor, fibrinogen, and factors XIII and VIII (antihemophilia factor). It is useful in the treatment of deficiencies of these clotting factors and is handled in the same manner as fresh frozen plasma. Two studies have shown cryoprecipitate to be the blood product of choice for the treatment of von Willebrand's disease because it concentrates the larger, more hemostatically active von Willebrand's multimers into a smaller volume than fresh frozen plasma. 23,108 A preliminary study suggests cryoprecipitate corrects the hypocoagulable state associated with disseminated intravascular coagulation when administered at a dose of 5 to 7 ml/kg. 117 Cryoprecipitate is equivalent to fresh frozen plasma for the treatment of hemophilia A. The dose is 1 unit per 10 kg body weight. 84 Cryo-poor plasma is the supernatant plasma removed from the cryoprecipitate. Cryo-poor plasma contains factors II, VII, IX, and X, which make it useful for the treatment of rodenticide intoxication. Storage and handling of cryo-poor plasma is similar to fresh frozen plasma. The initial dose is 1 unit per 10 kg of body weight. Platelet-rich plasma is prepared from fresh whole blood by centrifugation at a slower rate than is used for production of PRBCs and plasma. 85 The platelets are suspended in a small amount of plasma to facilitate transfusion. Storage of fresh platelets is impractical outside a blood bank because of their requirement for storage at 20 C to 24 C in special plastic bags and continuous agitation. 2 Transfused platelets are rapidly destroyed in human patients with immune-mediated thrombocytopenia, and because immune-mediated thrombocytopenia is a common cause of profound thrombocytopenia in dogs, most cases of thrombocytopenia-mediated hemorrhage may not be amenable to successful platelet transfusion. If a platelet transfusion is given, the dose is the platelets collected from 1 unit of whole blood per 10 kg of body weight. Cryopreserved canine platelets are collected from a single donor via plateletpheresis, and the manufacturer reports one bag contains 1 Â 10 11 platelets preserved in dimethyl sulfoxide (DMSO). 57 The bag also contains a small amount of fresh frozen plasma. Efficacy data on this product have not been published, but the manufacturer recommends this product be used for the treatment of immune-mediated thrombocytopenia. The dose is 1 unit of frozen platelets per 10 kg of body weight. According to the manufacturer, anticipated increase in platelet count is 20,000/ mL 1 to 2 hours posttransfusion. Because the product contains DMSO, it must be infused slowly to prevent bradycardia. Cryopreserved canine platelet concentrate was compared with fresh platelet rich plasma in the laboratory. 48 This study identified decreases in platelet number and function as a result of the freeze-and-thaw process. Platelet number decreased 59% compared with the manufacturer's reported platelet count and platelets demonstrated multiple features of activation. The impact of cryopreservation on platelet function and number in vitro has not been studied. The use of serum has been recommended for the treatment of kittens and puppies with failure of passive transfer. Kittens treated with 5 mL subcutaneously or intraperitoneally three times in 24 hours achieved immunoglobulin G (IgG) concentrations comparable to kittens receiving colostrum. 72 Treatment of puppies with 22 mL/kg of serum given orally or subcutaneously at birth did not result in equivalent IgG and IgA concentrations when serum-treated puppies were compared with nursing puppies. 90 IgM was higher in the puppies treated with subcutaneously administered serum. Human albumin is a concentrate of albumin derived from pooled human plasma. Homology between canine and human albumin is approximately 79%, and human albumin is antigenic in dogs. 30, 80 Previous human albumin transfusion does not appear to be required for production of antibodies in dogs. 80 Hypoalbuminemia predicts a negative outcome in several canine diseases; consequently, the ability to correct hypoalbuminemia by albumin transfusion would be a medically desirable intervention. 1, 20 Because canine albumin was not previously available, human albumin has been used in dogs. Two retrospective studies have evaluated transfusion of human albumin to critically ill dogs. 81, 115 One associated improved albumin levels and blood pressure with human albumin administration and did not report serious adverse events. 81 The second concluded the serious nature of the diseases treated with precluded recognition of complications of the transfusion. 115 A recent study performed in normal dogs has identified serious adverse events suggestive of anaphylactic and fatal type III hypersensitivity reactions. 25 Transfusion of dogs with human albumin should be undertaken with great caution especially because lyophilized canine albumin is available. 58 Human intravenous immunoglobulin (hIVIG) is a highly purified preparation of immunoglobulin G, obtained from large pools of donated human plasma. The manufacturer provides the product as a lyophilized powder, which is reconstituted before transfusion. Sporadic availability of hIVIG limits its use, as does its high cost. Estimates indicate the cost of the drugs alone may be as high as $3000 to treat a 20 kg dog. 129 Reconstituted hIVIG is infused over 6 hours. Most report a single administration of the drug at a dosage of 0.5 to 1.0 g/kg, but in some cases the dose is administered three times on 3 consecutive days. 8, 9, 113, 129 Because of its immunomodulatory properties, transfusion of hIVIG has become more common in veterinary patients. 94 The two major diseases treated with hIVIG have been immune-mediated hemolytic anemia (IMHA) and immune-mediated thrombocytopenia (ITP), but hIVIG has also been used to treat some immune-mediated dermatologic disorders as well. 8, 9, 113, 129 Randomized, controlled prospective studies of glucocorticoids with and without hIVIG for the treatment of IMHA and ITP have been published. 8, 129 The ITP study demonstrated reduction in platelet recovery time without a concurrent increase in associated charges in the group randomized to receive glucocorticoids and hIVIG. The IMHA study did not show an advantage to treatment with hIVIG and glucocorticoids compared with glucocorticoids alone, but the study was underpowered to distinguish a difference between the two treatment groups. Administration of hIVIG to normal dogs promoted a hypercoagulable state, but in clinically ill dogs causality of thromboembolism is difficult to determine given the complexity of diseases undergoing hIVIG transfusion. 100, 116 The most convenient source of blood for a veterinary clinic is a commercial blood bank. Currently, there are only a limited number of commercial veterinary blood banks in the United States, and they cannot adequately supply all the small animal practices in the country with blood (see Box 24-1). Veterinary school blood donor programs may serve as an additional source of blood for the practitioner. 56 Because of the limited supply of blood from commercial animal blood banks, small animal practitioners typically borrow a donor from an employee or maintain a blood donor on the premises. 56 Borrowing a donor from either an employee or a client is a frequently used, if less convenient, option and is less expensive than maintaining an in-hospital donor. Maintaining a donor on the premises is advantageous because they are readily available for donation and their health status and disease exposure can be controlled, but the expense associated with feeding, housing, and caring for a blood donor is significant. 54 Volunteer blood donor programs have replaced many on-site blood donors. 14, 59 Collecting blood from stray animals is unsafe because infectious disease exposure and health status are unknown. Identification of donor dogs and cats before blood is needed is essential to allow blood type to be determined and the health status of the donor to be assessed before blood collection, thus ensuring the safety of the blood being transfused. The American College of Veterinary Internal Medicine has published recommendations on infectious disease screening for canine and feline blood donors as a consensus statement. 124 The recommendations have been incorporated into the following sections. For nearly 60 years, the best blood donor was believed to be a large, quiet dog not requiring anesthesia during blood collection. 78 The current recommendation is unchanged. A canine blood donor weighing more than 27 kg can safely donate 450 mL of blood in one donation, allowing collection of blood into commercially manufactured blood collection bags designed to facilitate sterile processing of components. Dogs weighing 27 kg or more have been shown to consistently donate 1 unit of blood for 2 years at 3-week intervals. 92 Dogs selected as donors also should have an easily accessible jugular vein to facilitate venipuncture. Prior pregnancy does not exclude female dogs from donation because pregnancy does not induce alloantibodies. 12 Greyhounds have been promoted as ideal blood donors because of their gentle disposition, high hematocrits, and lean body type, which simplifies blood collection. 45 Many greyhounds are euthanized because of poor racing performance, and these dogs are available from racetracks, breeders, and rescue organizations. 36 Blood banks choosing greyhounds as blood donors should be aware of certain breed idiosyncrasies that will impact on the management of a greyhound donor. The greyhound idiosyncrasy most important in transfusion medicine is the high red blood cell count, PCV and hemoglobin concentration, and low white blood cell counts and platelet count compared with mixed breed dogs. 91, 109 Greyhounds in Florida have a 46% seroprevalence of babesiosis. 111 Because the geographic origin of greyhounds serving as blood donors cannot always be determined, all greyhounds being screened as donors should have serologic testing for Babesia canis performed, and dogs with positive titers should be excluded as donors. Greyhounds with negative titers against B. canis should have B. canis polymerase chain reaction (PCR) performed, and if the test is positive, the dog should be excluded as a donor. In addition to the tendency of greyhounds to be asymptomatic carriers of B. canis, some other breeds of dogs should be used cautiously as blood donors because they are known to be asymptomatic carriers of infectious organisms transmitted by transfusion. American pit bull terriers and Staffordshire bull terriers recently have been recognized as carriers of Babesia gibsoni. 10, 77 Use of these dogs as blood donors should be restricted to those dogs that are seronegative and PCR-negative for B. gibsoni. Leishmaniasis has been identified in American foxhounds and transfusion of Leishmania infantum-infected blood from American foxhounds resulted in clinical leishmaniasis in transfusion recipients. 47, 89 All potential foxhound donors should be screened for Leishmania sp. Determination of blood type is critical to the selection of a blood donor dog. Although seven canine blood groups or blood type systems have received international standardization, typing sera are available for only five types (Box 24-2). Red blood cells can be negative or positive for a given blood type, except for the dog erythrocyte antigen (DEA) 1 system, which has three subtypes: DEA 1.1, 1.2, and 1.3. Canine red blood cells can be negative for all three subtypes (a DEA 1-negative blood type) or positive for any one of the three subtypes. Naturally occurring alloantibodies occur infrequently and without previous sensitization from transfusion do not appear to cause transfusion incompatibility in the dog 51 (see . A new canine red blood cell antigen, Dal, has recently been described. 11 This antigen appears to be common in the general canine population and lacking in Dalmatians. Transfusion with Dal-positive blood, induced an anti-Dal antibody producing multiple incompatible crossmatch tests. Dogs producing anti-Dal antibodies are at risk for hemolytic transfusion reactions. The blood type of the ideal canine blood donor is not uniformly agreed on among transfusion experts. Of the five blood groups for which typing sera are available, a transfusion reaction has been attributed to antibodies against DEA 1.1 induced by a DEA 1.1-positive transfusion in a DEA 1.1-negative recipient and to an antibody induced by a DEA 4-positive transfusion in a DEA 4-negative dog. 40, 83 In theory, red blood cells expressing DEA 1.2 can sensitize a DEA 1.2-negative transfusion recipient, resulting in an acute hemolytic transfusion reaction if a second transfusion of DEA 1.2-positive blood is given. In a laboratory setting, antibodies against DEA 1.2 have been reported to cause transfusion reactions, but clinical reports of hemolytic transfusion reactions mediated by anti-DEA 1.2 antibodies are lacking. DEA 7 is believed to be structurally related to an antigen found in common bacteria. A naturally occurring antibody against DEA 7 has previously been described in 20% to 50% of DEA 7-negative dogs, but recently revised down to 9.8% of dogs. 51 This antibody may result in accelerated removal of DEA 7-positive cells from a DEA-negative donor with anti-DEA 7 antibodies. 51, 102 Based on this information, the recommendation has been made to select donors that are negative for DEA 1.1, 1.2, and 7. Others suggest the donor dog should also have red blood cells positive for DEA 4 to be designated as a universal donor. 50 The recent description of a transfusion reaction attributed to antibodies against DEA 4 in a dog with DEA 4-negative red blood cells calls into question this recommendation. 119 Ninety-eight percent of dogs are DEA 4-positive, making it easy to find donors of this blood type. The importance of DEA 3 and 5 and Dal in blood donor selection remains to be determined. One other feature that should be considered before selection as a blood donor is the dog's plasma von Willebrand factor concentration. Von Willebrand's disease is the most common inherited coagulopathy in dogs and has been reported in many breeds of dogs and in dogs of mixed breeding as well. Because of the high frequency of this disease in the canine population, plasma from a canine blood donor will likely be used to transfuse a dog with von Willebrand's disease-induced hemorrhage, and a donor with a normal concentration of von Willebrand's factor is essential to replace the deficient coagulation factor. The physical requirements for a feline blood donor are similar to those for a canine donor. The ideal feline donor is a large cat, more than 5 kg of body weight, with a pleasant disposition. Easily accessible jugular veins facilitate collection of blood, and choosing a shorthair cat decreases the clipping required before phlebotomy. It is essential to determine the blood type of potential donors. One feline blood group system has been identified with three blood types: A, B, and AB (see Box 24-2) and recently a new common red blood cell antigen, Mik has been identified. 6, 128 Unlike dogs, cats have naturally occurring alloantibodies. Type A cats have naturally occurring alloantibodies against type B cells and type B cats against type A cells. 38 Cats of blood type B have strong hemagglutinating antibodies of the IgM type against type A cells, and cats of blood type A have weak hemolysin and hemagglutinating antibodies of the IgM and IgG type against type B cells. The clinical significance of these alloantibodies is threefold in transfusion medicine. First and most importantly, a cat may have a transfusion reaction without sensitization from a previous transfusion; second, type A kittens born to a type B queen are at risk for neonatal isoerythrolysis 21 ; and third, the antibodies are useful in determining the blood type of a cat. Mik appears to be a common red blood cell antigen. Only a few cats lacking Mik have been identified and they all produce anti-Mik alloantibodies. Donors of both type A and type B blood must be available because there is no universal donor in cats. Incompatible transfusions result in shortened red blood cell survival and potentially death in the transfusion recipient; therefore the serologic compatibility between recipient and donor must be determined before every transfusion in cats. 38 Donors of type A blood are easy to find because more than 99% of the domestic cats in the United States are type A. 42 The prevalence of domestic cats with type B blood varies geographically. In the United States, the western states have the highest percentage of type B cats, 4% to 6%. 42 Australia has the highest reported percentage of type B cats in their domestic cat population, 73%. 6 In Europe, the frequency of blood type B in domestic cats varies from 0% in Finland, 14.9% in France, and 24.6% in Turkey. 3, 41 Some purebred cats have a higher frequency of type B in their population. 39 The British shorthair, Devon rex and Turkish van have been reported to have the highest proportion of type B individuals, approximately 50% to 60%. 4 The Siamese, Oriental shorthair, Burmese, Tonkinese, American shorthair, and Norwegian forest cat breeds have not been reported to have any members with type B blood. Blood type AB is extremely rare, occurring in 0.14% of cats in the United States and Canada. 46 Fortunately, a type AB donor is not required to successfully transfuse a type AB cat. Blood from a type A cat is adequate. Screening blood donors for infectious diseases transmitted by blood transfusion is an integral step in maintaining a safe blood supply. Infectious disease screening of canine and feline blood donors varies within the different geographic regions of the United States and with the breed of the blood donor. An American College of Veterinary Internal Medicine consensus statement, developed by a committee consisting of members of the Infectious Disease Study Group and the Association of Veterinary Hematology and Transfusion Medicine should serve as the guideline for donor screening. 124 Organisms infectious to dogs and known to be transmitted by blood transfusion include B. canis, B. gibsoni, Haemobartonella canis, and Leishmania sp. 31 124 Dogs should not donate if they are ill or have fever, vomiting, or diarrhea; using donors with these clinical signs has resulted in Yersinia enterocolitica contamination of human units of blood. 32 Organisms infectious to cats and known to be transmitted by blood transfusion include: feline leukemia virus (FeLV), feline immunodeficiency virus (FIV) Bartonella henselae, Anaplasma phagocytophilum, Ehrlichia spp. and Neorickettsia spp., and the organisms formerly classified as Haemobartonella sp. (Mycoplasma haemofelis and Candidatus Mycoplasma haemominutum). 33, 49 Potential donor cats should be screened for FeLV and FIV. Because the prepatent period for FeLV infection can be 3 months, cats being considered as donors should be screened monthly for FeLV for 3 consecutive months. Testing for FIV antibodies can be performed simultaneously. Bartonella henselae is an emerging feline infectious disease and has been transmitted to cats by infected blood. 69 The use of cats with positive serology or cultures for B. henselae as blood donors is controversial and eliminating these cats from the donor pool is the safest approach. Testing for hemoplasma should include both light microscopy and PCR, and infected cats should be eliminated from the donor pool. 33 Screening of donor cats for feline infectious peritonitis (FIP) is problematic because there is no reliable test to identify the FIP-causing coronavirus. Feline blood donors should be screened for infection with Cytauxzoon felis and the agents causing feline ehrlichiosis if they reside in or are known to have traveled to endemic locations. A safe blood supply begins with healthy blood donors. All blood donors should undergo a complete physical examination each time they donate blood. Complete and differential blood counts, biochemical profile, and fecal examination should be performed annually. Donor cats and dogs with exposure to the outdoors or to ectoparasites should be routinely screened for infectious disease. Blood donors should be tested for heartworms, treated for ectoparasites, and vaccinated for the diseases on the schedule recommended for pets residing in the geographic region of the blood bank. Because the ideal feline blood donor lives in an indoor environment and is not exposed to other cats, the author believes vaccinations against FeLV, FIV, and FIP are unnecessary in donor cats. Exposure to the outdoors or to fleas approximately doubles the prevalence of hemoplasma infections in donor cats and restricting access to the outdoors, fleas, and other cats can prevent most infectious diseases in feline blood donors. 49 Strict aseptic technique must be used during the blood collection process to prevent contamination of blood with microorganisms. Whenever possible, solutions and equipment used for the collection process should be single-use products to prevent inadvertent contamination of blood. 55 After clipping the hair over the venipuncture site, the skin is surgically scrubbed. The ideal skin preparation regimen is yet to be determined in animals; however, in human blood donors, a 30-second, 70% isopropyl scrub followed by a 2% iodine tincture resulted in better skin surface disinfection than alcohol followed by chlorhexidine or green soap. 44 The phlebotomist wears sterile surgical gloves and performs venipuncture without touching the scrubbed area. Several different solutions are available to anticoagulate and preserve blood for transfusion and species-specific storage times are listed (Table 24-1) . Anticoagulants provide no nutrients to preserve red cell metabolism during storage. Blood collected in anticoagulants should be transfused immediately. Anticoagulant-preservative solutions have been designed to provide nutrients to maintain red blood cell function during storage. One common anticoagulant solution for preservation of canine red blood cells, citrate phosphate dextrose adenine (CPDA-1), is found in commercially prepared, multiple-bag systems. Maximal storage time for feline blood in CPDA-1 has yet to be determined but may be as long as 35 days. 15 Acid citrate dextrose or anticoagulant citrate dextrose (ACD) formula B can be used to store either canine or feline blood. 29, 79 It can be purchased in 500-mL bags and placed in syringes for collection of blood. Cat and dog red blood cells maintain adequate viability following storage in ACD for 30 and 21 days, respectively. Additive solutions are contained in a multibag system containing citrate phosphate dextrose (CPD) or citrate phosphate dextrose 2 (CPD-2) as the anticoagulant. The additive solution is contained in a bag separate from the main bag and is added to PRBCs after the plasma is removed. Additive solutions that have been evaluated in dogs are Adsol (Fenwal Laboratories, Baxter Health Care Corporation, Deerfield, Ill.) and Nutricel (Miles Pharmaceutical Division, West Haven, Conn.). 123, 125 Additive solutions have not been evaluated for storage of feline blood, but are sometimes used as storage media for feline red blood cells. White blood cells are responsible for some adverse effects of transfusion and donot contribute to transfusion efficacy(see "Adverse Effects of Transfusion"). An integral filter to remove white blood cells from whole blood is incorporated into some blood bag systems. One system has been evaluated using canine blood and effectively removed white blood cells without affecting red blood cell viability. 13 98 The choice is strictly a matter of personal preference and skill. Acepromazine is not recommended because it causes hypotension and platelet dysfunction. The flow of blood into the bag can occur by gravity or suction. Blood collected by suction does not have a greater rate of hemolysis than that collected by gravity flow, and it can be collected more rapidly. 27 Suction collection of blood is facilitated using a vacuum chamber manufactured by the Animal Blood Resources International (Stockbridge, Mich.). This device requires an external vacuum source during collection of blood. It is unusual to find a feline blood donor that does not require sedation during blood donation. The author prefers a combination of ketamine (10 mg) and diazepam (0.5 mg) intravenously for cats, whereas others recommend using midazolam, and isoflurane or sevoflurane. 98, 114 If the sedative agent is to be given intravenously, a peripheral vein (cephalic or medial saphenous) should be used to preserve the jugular veins for blood collection. No commercially available system is manufactured for the collection of blood from cats because of the small volume of blood that can safely be withdrawn from a cat. Typically, anticoagulant can be withdrawn from a blood bag port using a syringe. It is placed in one or two large syringes (25 to 60 mL) depending on the volume of blood to be collected (see Table 24 -1). A large (19gauge) butterfly needle is used for jugular venipuncture so that if a second syringe of blood is to be collected, the full syringe can be removed and the second syringe connected without a second venipuncture. By the definition of the American Association of Blood Banks, this is an "open" system, and blood collected in this manner should not be transfused more than 4 hours after collection. 118 Alternatively, a standard blood collection bag containing CPDA-1 can be used. All CPDA-1 is expelled from the bag except for the amount remaining in the tubing. Feline blood is collected directly into the bag. 93 A commercially available vacuum system can be used for collecting blood from cats, but some authors find this system less satisfactory than the syringe method. 63, 98 Selection and transfusion of compatible blood is one component of the process to provide a safe and efficacious red cell transfusion. With the identification of a new red blood cell antigen in both dogs and cats, recommendations for appropriate compatibility testing before the first transfusion are a currently being revaluated. Because each unit of red blood cells is antigenically distinct, the recipient may form antibodies after transfusion of any unit of blood. The immune system will take a minimum of 5 days to make antibodies against transfused red blood cells; therefore, a crossmatch should be performed if more than 4 days elapse between transfusions. Performing a crossmatch will not prevent an immune reaction to subsequent transfusions; it can only identify those units of blood with potential to cause acute hemolytic transfusion reactions. Because of the lack of clinically significant preformed alloantibodies in the dog, blood typing and crossmatching are not routinely performed before the first transfusion. When DEA 1.1-positive blood is transfused, ideally it would be given to a DEA 1.1-positive recipient to prevent sensitization of a DEA 1.1-negative dog. DEA 1.1 status can be determined by using the typing systems described below. Crossmatching should be performed before any subsequent transfusion to identify a compatible unit of red blood cells. Blood typing or crossmatching is not required before transfusion of canine plasma. Previously, blood typing was considered adequate pretransfusion testing before administration of red blood cells or plasma to cats. Blood typing prevented administration of type B blood to a type A cat and vice versa. However, transfusion of A-B mismatched blood results in decreased red blood cell survival or a potentially fatal, acute hemolytic transfusion reaction. 38 Blood typing will not identify the Mik antigen and its naturally occurring alloantibody; however crossmatching will detect the anti-Mik antibody and crossmatching may become the preferred compatibility test for all feline transfusions to prevent incompatible transfusions due to anti-A, anti-B, or anti-Mik alloantibodies. Determining A-B blood type in the cat has been simplified by the availability of in-clinic typing systems (Figure 24-1) . A special situation with regard to blood typing and crossmatching exists in cats. When blood typing is unavailable, crossmatching will administration of an incompatible transfusion due to A, B, or Mik alloantibodies. When crossmatching is performed with a known type A donor, an incompatible major crossmatch strongly suggests the potential recipient is a type B or a Mik negative cat because of the naturally occurring alloantibodies in these cats. If cat plasma is administered, it should be the same blood type as the recipient because plasma will contain anti-A, anti-B, and anti-Mik antibodies. Crossmatching the donor to the recipient cat will prevent a reaction because of A, B, or Mik alloantibodies. The person administering the blood should pay careful attention to the blood bag label before transfusion. The most common reason for an acute hemolytic transfusion reaction in human patients is clerical error-the wrong unit of blood is released from the blood bank or a unit of blood is given to a patient who was not intended to receive a transfusion. 110 In veterinary medicine, it is crucial to confirm that the blood comes from the correct species of blood donor in addition to being typed and matched to the patient requiring a transfusion. The contents of the bag also should be examined for normal color and consistency. Bacterially contaminated blood often appears brown or purple because of deoxygenation, hemolysis, and formation of methemoglobin. 55, 65 Blood and plasma can be administered using several routes. Most commonly, blood is given intravenously. The diameter of the catheter used for transfusion is important in determining the rate of blood flow because blood flows more slowly through a small catheter; however, small diameter catheters have not been associated with increased risk of hemolysis during transfusion. 118 The intraosseous route can be used successfully for administration of blood and plasma. 88 In normal dogs, 93% to 98% of red blood cells administered through an intraosseous catheter are found in the peripheral circulation within 5 minutes. 24 This rapid and simple method is especially useful in animals with vascular collapse and in extremely young puppies and kittens. Special intraosseous catheters are available, but a spinal needle, bone marrow aspiration needle, over-the-needle catheter, or even an ordinary hypodermic needle can be used. Sites for the placement of the intraosseous catheter include the trochanteric fossa of the femur, the medial tibia, and the iliac crest. Blood flows very rapidly through an intraosseous catheter, and rate of administration should be monitored closely. Plasma can be administered intraperitoneally in emergency situations, but red blood cells are slowly and poorly absorbed when administered by this route, and it is not recommended for red blood cell transfusions. A blood transfusion administration set is required for administration of red blood cells or plasma to remove blood clots and debris, which form during storage and which could cause embolism. The filter typically used in veterinary medicine is 170 mm in size. For smallvolume transfusions, an 18-mm filter attached to intravenous tubing is useful. An 18-mm filter does not work well for large-volume transfusions because it rapidly becomes obstructed with debris, and transfusion rate slows. The risk of an air embolism is increased when blood is collected into glass bottles. A blood administration set does not remove air from stored blood; accordingly, glass bottles are not recommended for collection and storage of blood. The American Association of Blood Banks explicitly states that medications should not be added to blood or components. 118 In addition, no fluid should be added to blood excep. 0.9% sodium chloride when it is necessary to decrease the viscosity of PRBCs. Fluids containing calcium such as lactated Ringer's solution may overcome the anticoagulant properties of citrate, resulting in coagulation of the blood. Solutions such as 5% dextrose in water are hypotonic and may induce hemolysis. The recommended rate of transfusion of red blood cells depends on the status of the recipient. In massive hemorrhage, the transfusion should be given as rapidly as possible. In a normovolemic, stable transfusion recipient, some clinicians recommend a rate of 0.25 mL/ kg for the first 30 minutes, after which the rate is increased if no reaction is seen. 112 In patients with heart disease, a rate of 4 mL/kg/hr should not be exceeded. 45 Transfusion rates of 10 mL/kg/hr, 4 mL/kg/hr, and up to 60 mL/kg/hr were used to transfuse red blood cells to cats with normovolemia, cardiovascular dysfunction. and hypovolemic shock, respectively. 126 Plasma can be given more rapidly (4 to 6 mL/min). 67 Whatever the rate chosen, it should be rapid enough to complete the transfusion within 4 hours of initiation because of the risk of bacterial growth in blood maintained at room temperature for a prolonged period. Control of blood product delivery rate can be accomplished by use of infusion pumps to deliver a preset volume over a specific period of time. The use of infusion pumps must be limited to devices approved for use with blood because some infusion pumps can result in hemolysis of red blood cells as a result of excessive pressure. 107 Because blood does not contain any antibacterial agents, it must be refrigerated until used to retard bacterial growth and maintain red blood cell viability. If the clinical status of the animal requires that the transfusion be given over a period greater than 4 hours, the blood can be split into smaller units with a transfer bag. One portion of the blood is transfused while the other is returned to the refrigerator until the first half of the transfusion is completed. In patients with cardiac disease at risk for volume overload, the risk can be further minimized by use of PRBCs, which require infusion of a lower volume than whole blood. Warming of blood before transfusion has been recommended to prevent hypothermia in the transfusion recipient. Warming of blood probably is only necessary if a large volume of blood is to be given or if the recipient is a neonate. For adult animals receiving a single unit of blood, the blood can be administered directly from the refrigerator. Warming blood has the potential for excessive heating, causing red blood cell membrane damage and hemolysis or promoting bacterial growth if contamination is present. Blood warming devices that use dry heat, radio waves, microwaves, or electromagnetic energy are available, but cost often is prohibitive. Refrigerated human blood can be warmed quickly by admixing it with warm (45 C to 60 C) 0.9% saline in a ratio of 1:1 without damage to red blood cells. 61 This method has not been tested for dogs or cats. Once blood is warmed to 37 C, it deteriorates rapidly and, if not used, should be discarded. Fresh frozen plasma must be thawed before transfusion. A method for thawing canine fresh frozen plasma in a microwave oven has been described, but the author has found this unsatisfactory because of uneven heating by household microwave ovens. 60 Plasma can be thawed at room temperature, and if the thawing time needs to be shortened, the plasma can be placed into a plastic bag and thawed in a 37 C water bath. The plastic bag is necessary to prevent contamination of the infusion ports in the water bath. If thawed and not used within 1 hour, it maybe refrozen with out loss of anticoagulant activity. 131 Plasma should be used within 4 hours of thawing or discarded. Transfusion recipients should be monitored during transfusion to allow early detection of a transfusion reaction. Rectal temperature, heart rate, and respiratory rate should be recorded every 10 minutes during the first 30 minutes and then every 30 minutes thereafter. The patient should be monitored for vomiting, diarrhea, urticaria, and hemoglobinuria or hemoglobinemia. Changes in vital signs or clinical status may indicate a transfusion reaction. Patients developing volume overload will become tachypneic or dyspneic, and tachycardic. Massive transfusions (1 blood volume in 24 hours) have been reported in both dogs and cats. 16, 62, 97 Patients receiving massive transfusions of stored blood may develop specific abnormalities. Consequently, patients receiving massive transfusions should be monitored for changes in serum potassium, ionized calcium, and ionized magnesium concentrations, as well as hypothermia and coagulation abnormalities. 16, 62, 97 An adverse effect of transfusion or transfusion reaction consists of the range of immunologic and metabolic changes that occur during or after administration of a blood product. Four classes of adverse effects of transfusion have been described (Box 24-3). Acute transfusion reactions occur during or within a few hours after a transfusion, and delayed transfusion reactions occur after the completion of the transfusion. The delay may be hours to years. Reports describing adverse effects of transfusion in dogs and cats are limited to case reports and retrospective series.* Acute immunologic transfusion reactions occur because antibodies that elicit an immune response are present in the plasma of either the donor or recipient. The sequelae of an acute immunologic transfusion reaction are rapid, often irreversible, and sometimes fatal. Current theories on the pathogenesis of acute hemolytic transfusion reaction in humans propose that hemolysis induces the release of cytokines, such as tumor necrosis factor, interleukin 1 (IL-1), IL-6, and IL-8, complement, endotheliumderived relaxing factor (nitric oxide), and endothelin, resulting in the clinical syndrome of disseminated intravascular coagulation, shock, and acute renal failure. 9 The pathophysiology of acute hemolytic transfusion reaction in dogs and cats must differ in some manner from that described in humans because acute renal failure is not reported to be a feature in dogs and cats. 5, 37, 40, 132 The best example of an acute hemolytic transfusion reaction in veterinary medicine is the administration of type A red blood cells to a type B cat. In the recipient cat, naturally occurring alloantibodies and complement bind to the transfused red blood cells and cause hemolysis. Clinical signs described in cats having an acute hemolytic transfusion reaction include fever, vomiting, lethargy, icterus, and death. 5 Results of laboratory testing often show a positive Coombs test, rapidly declining PCV, and increasing serum bilirubin concentration. Dogs experiencing an acute hemolytic transfusion reaction show clinical signs similar but not identical to those observed in cats. Most affected dogs exhibit fever, restlessness, salivation, incontinence, and vomiting. Some dogs develop shock, and an occasional dog experiences acute death. Plasma and urine hemoglobin concentrations increase within minutes of transfusion. Incompatible cells are cleared from circulation in less than 2 hours. Dogs whose red blood cells lack the DEA 1.1 antigen that have previously been sensitized by transfusion of DEA 1.1-positive cells are at the greatest risk for an acute hemolytic transfusion reaction. 40 Other acute immunologic transfusion reactions reported in dogs and cats include nonhemolytic fever and urticaria. 17, 53, 64, 120 In humans, nonhemolytic fever is a result of antibodies against donor white blood cells, and urticaria occurs as a result of antibodies-against donor plasma proteins. Nonhemolytic febrile transfusion reactions do not require treatment, but antipyretics may be used if the patient is uncomfortable (Table 24 -2). Urticaria is the most common reaction to plasma transfusion in dogs. 120 If urticaria caused by plasma administration is diagnosed, it should be treated with short-acting corticosteroids and antihistamines. The plasma transfusion then may be restarted at a slower rate and the recipient carefully observed. Delayed immunologic transfusion reactions are classified as delayed hemolytic, transfusion-induced immunosuppression, posttransfusion purpura, and graft-versus-host disease. These reactions are not preventable by crossmatching or blood typing. Delayed hemolytic transfusion reactions invariably occur in persons who have been previously sensitized to allogenic red blood cell antigens by transfusion or pregnancy. Even though compatible blood is given to a patient, the recipient may develop antibodies against any one of the hundreds of red blood cell antigens present on the transfused cells. An anamnestic response to the antigens on the transfused red blood cells results in a delayed hemolytic transfusion reaction that occurs 7 to 10 days after a transfusion and is a well-described complication of red cell transfusion in humans. It has not been reported in dogs, but there is no reason it could not occur. Fever is the most common sign of a delayed hemolytic transfusion reaction in humans. Icterus also may be noticed 4 to 7 days after a transfusion. The only delayed immunologic transfusion reaction that has been reported in veterinary medicine is posttransfusion purpura occurring in a previously transfused dog with hemophilia A. 122 Five to eight days after subsequent transfusion, thrombocytopenia and petechiation were evident. Blood collected during a thrombocytopenic episode was positive for plateletbound IgG, indicating an immune mechanism for platelet destruction. Acute Immunologic Acute nonimmunologic transfusion reactions are caused by physical changes in the red blood cells during collection, storage, or administration. Improper collection of blood can result in an adverse reaction to transfusion. Collection of blood from an inadequately screened donor can result in transmission of bacteria, spirochetes, or protozoa and eventually clinical signs of the associated disease in the recipient. Transfusion of blood contaminated by bacteria can cause shock, which is managed with volume expansion and pressor agents, as well as empirical antibiotic administration based on results of a Gram stain. Endotoxic shock results from transfusion of blood heavily contaminated with endotoxin-producing bacteria. Clinical signs in cats transfused with blood contaminated by bacteria include collapse, vomiting, diarrhea, and acute death, but most cats did not exhibit clinical signs after receiving bacterially contaminated blood. 55 Hypotensive shock developed in a dog that received a B. canis-infected transfusion. 18 During storage, the ATP content of red blood cells decreases, and some cells undergo hemolysis resulting in leakage of potassium out of the cells into the storage medium. The increase in potassium in the storage medium is a contributing factor in the development of hyperkalemia in patients receiving large volume transfusions of stored blood. A large-volume transfusion of stored blood can cause hyperkalemia, but this is rare unless the patient has renal failure or preexisting hyperkalemia. 62 Hyperkalemia in a transfusion recipient is as it would be in any patient with hyperkalemia. The transfusion should be discontinued and 0.9% NaCl administered because 0.9% NaCl does not contain added potassium and will facilitate renal excretion of potassium. Intravenous administration of insulin, followed by administration of 50% dextrose and frequent monitoring of blood glucose and potassium concentrations until serum potassium concentration normalizes, is all that is necessary. Physical damage (such as freezing or overheating) to red blood cells during storage causes hemolysis. While being transfused, the patient exhibits hemoglobinuria and hemoglobinemia without evidence of other signs of an acute hemolytic transfusion reaction, such as fever, vomiting, or collapse. During storage of blood, formation of clots or introduction of air into the bag may occur, resulting in embolism during transfusion. A rare adverse event associated with transfusion is an embolism. Venous air embolism causes sudden onset pulmonary vascular obstruction, a precordial murmur, hypotension, and death as a result of respiratory failure. Administration of large-volume transfusions can result in multiple adverse events. Ionized hypocalcemia or ionized hypomagnesemia can result from the citrate used as an anticoagulant complexing with calcium or magnesium, and lead to myocardial dysfunction and potential cardiac arrest and tetany. 66 Routine empirical administration of calcium to transfusion recipients cannot be recommended because of the risk of hypercalcemia and increased myocardial irritability, but animals with ionized hypocalcemia resulting from large transfusion should be treated with calcium gluconate or calcium chloride to effect. 26 Hypothermia is common after large-volume transfusion in veterinary patients, and use of warming blankets should be instituted whenever possible. Dilution of coagulation factors by large-volume transfusion of coagulation factor-depleted stored blood results in prolongation of coagulation times. In dogs receiving large-volume transfusions, prolongation of coagulation times is associated with a poor prognosis. 62 Administration of fresh frozen plasma is indicated to correct the coagulation abnormalities. Any transfusion can cause circulatory overload. Dogs and cats with chronic severe anemia or compromised cardiac and pulmonary systems are at greater risk for circulatory overload and pulmonary edema than are those without cardiopulmonary disease. Dogs and cats developing volume overload from transfusion are treated with oxygen supplementation, diuretics, and vasodilators. Improvement should be seen within 1 to 2 hours. In humans, human immunodeficiency virus, hepatitis virus, and cytomegalovirus infections are documented as late effects of transfusion. One late complication of transfusion described in veterinary medicine is hemochromatosis. 104 A Schnauzer received blood transfusions every 6 to 8 weeks for 3 years to treat chronic anemia. Hemochromatosis was confirmed by necropsy when the dog was euthanized because of progressive liver disease. When an acute transfusion reaction is suspected, immediate intervention is critical because of the life-threatening nature of acute transfusion reactions. In all animals suspected of having some form of acute transfusion reaction, the transfusion should be stopped and samples of patient blood and urine obtained for baseline evaluation of biochemical, hematologic, and coagulation values. The unit of blood should be inspected to ensure it is from the appropriate species and is the intended unit based on the crossmatch or blood type. A Gram stain and bacterial culture of the blood remaining in the blood bag should be submitted to the laboratory. Urine can be visually inspected to determine the presence or absence of hemoglobin. Rectal temperature of the recipient should be compared with the pretransfusion value. A transfusionassociated fever is defined as an increase in 1 F over the pretransfusion temperature. 118 The cardiovascular system should be monitored by electrocardiogram and blood pressure measurement. Immediate evaluation of serum ionized calcium and potassium concentrations are critical, but certain electrocardiographic changes serve as surrogate markers of hypocalcemia (long QT-interval with a normal heart rate) or hyperkalemia (decreased height of P waves, loss of P waves, or widening of the QRS complex with large T waves) if rapid measurement of serum electrolyte concentrations cannot be obtained. Venous access and blood pressure should be maintained by an infusion of a crystalloid solution such as lactated Ringer's solution or 0.9% NaCl. Intravenous administration of short-acting glucocorticoids may suppress some of the mediators of acute hemolytic transfusion reactions and lessen the clinical progression, but their efficacy in transfusion reactions has not been evaluated in veterinary patients. When the evaluation of a patient with a suspected transfusion reaction suggests that an acute hemolytic transfusion reaction is occurring, the blood typing and crossmatching must be repeated to determine whether a laboratory error is responsible for the reaction. In patients with fever, without evidence of hemolysis, the transfusion may be restarted if the Gram stain is negative for bacterial contamination. It is important to recognize the late effects of transfusion and not mistake them for another disease process. Delayed transfusion reactions usually are managed with supportive care. The only specific treatment for a delayed transfusion reaction consists of treating a transfusionacquired infection appropriately. A special effort is not necessary to prevent transfusion reactions. By simply following the transfusion guidelines discussed here with reference to donor selection, blood typing, blood storage, and administration, most transfusion reactions can be prevented. Crossmatching should be included in the guidelines for providing a safe blood transfusion. Major and minor crossmatches detect antibodies in the plasma of the donor or recipient capable of causing an acute hemolytic transfusion reaction; however, a transfusion reaction may still occur despite a compatible crossmatch. Crossmatching does not prevent sensitization to red blood cell antigens, which may result in a hemolytic reaction during future transfusions because it detects only antibodies that are currently present in the donor or recipient. It should be performed routinely in veterinary clinics either by a commercially available gel tube method (DMS Laboratories, Inc., Flemington, N.J.) or by the tube method. A tube crossmatch is described below. Performing a crossmatch is an intimidating but simple procedure once all the equipment is assembled (Box 24-4) . Several descriptions of the procedure have been published, all of which describe the same basic procedure with minor variations. 14, 35, 103 Not all protocols recommend the use of phosphate-buffered saline; others have an additional step at the end using species-specific Coombs reagent to increase test sensitivity, and some recommend that tubes be incubated at 4 C, 37 C, and 42 C. The following is the protocol the author uses: 1. Obtain EDTA-anticoagulated blood from the recipient and the potential donor or the tube segments of blood from the units being considered for transfusion. 2. Centrifuge both donor and recipient blood for 5 minutes at 1000 g. 3. Using pipettes, remove the plasma, and save in separate labeled tubes. 4. Wash the red blood cells by adding phosphate-buffered saline to the red cells to fill the tube. Resuspend the red cells in the saline by tapping the bottom of the tube with a finger. 5. Centrifuge the red cells and saline for 5 minutes at 1000 g. Pipette off saline, and discard. 6. Repeat step. 4 and 5 twice. 7. After the third washing of the red cells in saline, resuspend the red cells to a 3% to 5% solution. It will appear bright cherry red. 8. For each potential donor, mix two drops of recipient plasma and one drop of donor red cell suspension for the major crossmatch. Mix gently. 9. For each potential donor, mix two drops of donor plasma and one drop of recipient red cell suspension for the minor crossmatch. Mix gently. 10. For the recipient control, mix two drops of recipient plasma and one drop of recipient red cell suspension. Mix gently. 11. Incubate the tubes at room temperature for 15 minutes. 12. Centrifuge the tubes for 15 seconds at 1000 g. 13. Observe the plasma for hemolysis. 14. Resuspend the centrifuged cells by shaking gently. 15. Observe the red blood cells for agglutination. Interpretation. Hemolysis or agglutination in a crossmatch indicates transfusion incompatibility. The degree of agglutination is graded 0 to 4þ (Box 24-5 and Figure 24 -2). Units of blood that are incompatible should not be used. If all available units are incompatible, the least reactive unit should be chosen. When the recipient control shows hemolysis or agglutination, the crossmatch cannot be interpreted. This is common in patients with hemolytic anemia. Multiple methods of blood typing dogs and cats have been described, including tube tests, typing cards, slide tests, immunochromatography, and gel tubes. 43, 106 A reference laboratory can perform blood typing for DEA 3, 4, 5, and 7 and the recently described Dal and Mik of dogs and cats, respectively. In clinical situations in the United States, the commercially available blood typing cards for feline types A, B, and AB are commonly used (DMS Laboratories, Inc., Flemington, N.J.). However, when results indicate type AB, the results should be interpreted with caution as the card typing method commonly gives false positive results as type AB when the cat is actually type A. 7 Any cat typed as AB should be confirmed by a second typing method. Gel tube typing tests, available in Europe, but not currently available in the United States, give accurate results when used in cats of type A, B, and AB. A simple immunochromatography method of blood typing has recently become available in the United States. 57 A paper strip impregnated with anti-A and anti-B monoclonal antibodies is placed in a red blood cell solution, allowing the cells to migrate up the strip and bind to the antibodies. Results are rapidly available and easily interpreted. When using any blood Equipment for Performing Crossmatch ) is an ultrapurified, polymerized hemoglobin of bovine origin (13 g/dL) in a modified Ringer's lactate solution with a physiologic pH (7.8). The hemoglobin polymers range in molecular mass from 65 to 500 kDa, with an average of 200 kDa. The viscosity is low compared with blood (1.3 and 3.5 centipoise, respectively), and the solution is isosmotic (300 mOsm/kg) with blood. The concentration of methemoglobin, the inactive form of hemoglobin, is 10%. Oxyglobin can be stored at room temperature or refrigerated (2 C to 30 C) for up to 3 years. Its intravascular half-life is dose dependent (18 to 43 hours, at a dosage of 10 to 30 mL/kg), as measured in healthy dogs. It is expected that more than 90% of the administered dose will be eliminated from the body in 5 to 7 days after infusion. The oxygen half-saturation pressure (P-50) of Oxyglobin is greater than that of canine blood (38 vs. 30 mm Hg, respectively). This increase in P-50 facilitates off-loading of oxygen to the tissues. The hemoglobin is packaged in the deoxygenated state in an overwrap that is impermeable to oxygen. Complications of severe anemia result from poor oxygenation of tissues. Restoration of adequate tissue oxygenation typically is achieved by administering a blood transfusion. Improvement in the clinical signs of anemia results from a corresponding increase in hemoglobin concentration, which in turn increases the arterial oxygen content of the blood. The increased oxygen content of the blood supplied by Oxyglobin also relieves the clinical signs of anemia. Two prospective randomized trials have evaluated Oxyglobin for the treatment of anemia. 95, 133 One was a multicenter clinical trial for dogs with moderate to severe anemia (PCV, 6% to 23%). 95 Sixty-four dogs in need of blood transfusion were studied, including those with anemia caused by blood loss (n ¼ 25), hemolysis (n ¼ 30), or ineffective erythropoiesis (n ¼ 9). Dogs in both groups were monitored for a decrease in hemoglobin concentration or deterioration in physical condition at which time they received additional oxygen-carrying support. If additional oxygen-carrying support was needed, Oxyglobin-treated dogs received PRBCs (n ¼ 1), and untreated control dogs received Oxyglobin (n ¼ 19). Treatment success was defined as the lack of need for additional oxygen-carrying support for 24 hours. The success rate in the 30 treated dogs (95%) was significantly greater than the success rate in the 34 control dogs (32%). This difference between treated and control dogs was significant, regardless of the cause of anemia. The other trial randomized 12 dogs with severe anemia (PCV ¼ 10% to 20%) secondary to babesiosis to receive either 20 mL/kg of Oxyglobin, or packed red blood cells. 133 Blood gas, acid-base, and blood pressure were objective measures of response to treatment. Similar overall improvements were seen in both the Oxyglobin and PRBC transfusion groups. Although Oxyglobin is approved only for use in dogs, other species have been infused with the solution. Oxyglobin administration to cats has been retrospectively evaluated. 34, 127 The median dosage was approximately 10 to 11 mL/kg/24 hours. Oxyglobin has also been administered to other species to increase oxygen carrying capacity: mallard duck, miniature horse, and serval cat. 73, 82, 101 One published dosage for birds is 5 mL/kg IV or interosseously for the treatment of shock and for the treatment of shock in small mammals, 2 mL/kg as a 10 to 15 minute intravenous bolus followed by a continuous rate infusion at 0.2 to 0.4 mL/kg/hr. 74 Because it lacks the antigenic red blood cell membrane, Oxyglobin is not only useful in multiple species, but it eliminates some of the pretransfusion testing required with red blood cell transfusions. Blood typing and crossmatching are not necessary because the red blood cell membrane, which is the major cause of transfusion incompatibility, has been removed during the manufacturing process. Repeated dosing of Oxyglobin was reported in both feline retrospective studies. 34, 127 No allergic reactions were reported. A laboratory study of repeated dosing in dogs showed antibodies to Oxyglobin did form, but those antibodies did not decrease binding of oxygen to Oxyglobin and did not result in systemic allergic reactions. 52 Adverse effects of treatment with Oxyglobin are similar in dogs and cats. After treatment, a transient discoloration (yellow, brown, or red) of the mucous membranes, sclera, urine, and sometimes skin occurs. Overexpansion of the vascular volume may occur, especially in normovolemic animals. Rates of administration greater than 10 mL/kg/hr in anemic, clinically ill dogs sometimes resulted in increased central venous pressure, with or without pulmonary edema or other respiratory signs of circulatory overload. Pleural effusion and pulmonary edema were found commonly in cats given Oxyglobin, but evidence was insufficient to directly link either to the administration of Oxyglobin. 34, 127 Cats recently administered blood transfusions or having underlying cardiac disease appear to be more likely to develop pleural effusion and pulmonary edema. Low infusion rates are recommended in these cats (<5 mL/kg/hr). In the canine clinical trial, vomiting occurred in 35% of the treated dogs. Diarrhea, fever, and death also were seen in approximately 15% of Oxyglobin-treated dogs; however, an association with Oxyglobin or the underlying disease could not be determined. These findings were most common in dogs with immune-mediated hemolytic anemia that received Oxyglobin. The presence of Oxyglobin in serum may cause artifactual changes in the results of serum chemistry tests. Interference by Oxyglobin depends on the type of analyzers and reagents used but is not typical of hemolysis. 18, 86 Blood samples for analysis should be collected before infusion. A list of valid chemistry tests by analyzer is included in the product labeling. Results of any clinical chemistry test performed on serum containing Oxyglobin should be interpreted with consideration of the validity of the test. In general, all tests using colorimetric techniques are invalid, but other methodologies also show some interference. No interference is seen with hematologic or coagulation parameters except when optical methods are used for measuring prothrombin time and activated partial thromboplastin time. Dipstick measurements (pH, glucose, ketones, protein) of urine are inaccurate when gross discoloration of the urine is present. The urine sediment is not affected. Chronic enteropathies in dogs: evaluation of risk factors for negative outcome Room temperature storage and cryopreservation of canine platelet concentrates Blood type A and B frequencies in Turkish van and Angora cats in Turkey Frequencies of blood type A, B and AB in non-pedigree domestic cats in Turkey Blood transfusion reactions in the cat The AB blood group system in cats Erythrocytic pyruvate kinase deficiency and AB blood types in Australian Abyssinian and Somali cats A prospective, randomized, double-blinded, placebo-controlled study of human intravenous immunoglobulin for the acute management of presumptive primary immune-mediated thrombocytopenia in dogs Treatment of severe immune-mediated thrombocytopenia with human IV immunoglobulin in 5 dogs Serosurvey of anti-Babesia antibodies in stray dogs and American pit bull terriers and American Staffordshire terriers from North Carolina Canine Dal blood type: a red cell antigen lacking in some Dalmatians Lack of evidence of pregnancy-induced alloantibodies in dogs Use of a prestorage leukoreduction filter effectively removes leukocytes from canine whole blood while preserving red blood cell viability Problems in veterinary medicine. Philadelphia: JB Lippincott Storage of feline and canine whole blood in CPDA-1 and determination of the posttransfusion viability Massive transfusion and surgical management of iatrogenic aortic laceration associated with cystocentesis in a dog Canine red blood cell transfusion practice In vitro effects of a novel hemoglobin-based oxygen carrier on the routine chemistry, therapeutic drug, coagulation, hematology, and blood bank assays Acute hemolytic transfusion reaction, a paradigm of the systemic inflammatory response: new insights into pathophysiology and treatment Prognostic factors for mortality and thromboembolism in canine immune-mediated hemolytic anemia: a retrospective study of 72 dogs Transfer of colostral antibodies from queens to their kittens Clinical use of blood products in cats: a retrospective study Effect of cryoprecipitate and plasma on plasma von Willebrand factor multimeters and bleeding time in Doberman pinschers with type-I von Willebrand's disease The absorption of red blood cells after parenteral injection at various sites Response of healthy dogs to infusions of human serum albumin Calcium chloride versus calcium gluconate: comparison of ionization and cardiovascular effects in children and dogs Post transfusion viability of stored canine red blood cells after vacuum facilitated collection Evaluation of preservatives and containers for storage of canine blood Use of biochemical measures to estimate viability of red blood cells in canine blood stored in acid citrate dextrose solution with and without added ascorbic acid Adverse reactions suggestive of type III hypersensitivity in six healthy dogs given human albumin Hypotensive shock syndrome associated with acute Babesia canis infection in a dog Transfusion acquired Yersinia enterocolitica Survival of Mycoplasma haemofelis and "Candidatus Mycoplasma haemominutum" in blood of cats used for transfusions Use of a hemoglobin-based oxygen-carrying solution in cats: 72 cases Problems in veterinary medicine. Philadelphia: JB Lippincott Where to get blood donors? Acute hemolytic transfusion reaction in an Abyssinian cat with blood type B Transfusion of type-A and type-B blood to cats Frequency and inheritance of A and B blood types in feline breeds of the United States An acute hemolytic transfusion reaction caused by dog erythrocyte antigen 1.1 incompatibility in a previously sensitized dog Frequencies of feline A and B blood types in Europe Geographical variation of the feline blood type frequencies in the United States Comparison of various canine blood-typing methods Evaluation of donor skin disinfection methods Blood transfusion therapy: an updated overview Blood type AB in the feline AB blood group system Seroprevalence of antibodies against Leishmania spp among dogs in the United States Assessment of a dimethyl sulfoxide-stabilized frozen canine platelet concentrate Anaplasma phagocytophilum and species of Bartonella, Neorickettsia and Ehrlichia in cats used as blood donors in the United States Canine blood groups and their importance in veterinary transfusion medicine Incidence of canine serum antibody to known dog erythrocyte antigens in potential donor population Absence of immunopathology associated with repeated IV administration of bovine Hb-based oxygen carrier in dogs Feline blood component therapy: retrospective study of 246 transfusions Problems in veterinary medicine. Philadelphia: JB Lippincott; 1992 Serratia marcescens contamination of feline whole blood in a hospital blood bank Transfusion practices and costs in dogs Evaluation of microwave-thawed canine plasma for transfusion Blood warming: current applications and techniques Massive transfusion in dogs Problems in veterinary medicine. Philadelphia: JB Lippincott Packed red blood cell transfusions in dogs: 131 cases (1989) Pseudomonas fluorescens contamination of a feline packed red blood cell unit and studies of canine units Response of canine plasma-ionized calcium and magnesium to the rapid infusion of acid-citrate-dextrose (ACD) solution Use of blood and blood components for feline and canine patients Red blood cell transfusions in cats: 126 cases (1999) in press Relapsing bacteremia after blood transmission of Bartonella henselae to cats General principles of small animal blood component administration Haemobartonella canis infection following splenectomy and transfusion Use of adult cat serum to correct failure of passive transfer in kittens Comparison of fluid types for resuscitation after acute blood loss in mallard ducks (Anas platyrhynchos) Emergency care and managing toxicoses in the exotic animal Clinical indications for use of fresh frozen plasma in dogs: 74 dogs A treatise on the heart on the movement and colour of the blood and on the passage of the chyle into the blood Babesia gibsoni infection among dogs in the southeastern United States The blood and plasma bank Posttransfusion viability of feline erythrocytes stored in acid citrate dextrose solution Serum antibodies against human albumin in critically ill and healthy dogs The use of 25% human serum albumin: outcome and efficacy in raising serum albumin and systemic blood pressure in critically ill dogs and cats Use of a bovine hemoglobin preparation in the treatment of cyclic ovarian hemorrhage in a miniature horse A hemolytic transfusion reaction due to DEA 4 alloantibodies in a dog Canine von Willebrand's disease; pathobiology, diagnosis and shortterm treatment Problems in veterinary medicine. Philadelphia: JB Lippincott Effect of Hemopure W on the performance of Ektachem and Hitachi clinical analyzers National Institutes of Health Consensus Conference Fresh-frozen plasma: indications and risks Intraosseous infusion of fluids and therapeutics Transmission of visceral leishmaniasis through blood transfusions from infected English foxhounds to anemic dogs Use of adult dog serum as a substitute for colostrums in the neonatal dog Hematologic values in mongrel and greyhound dogs being screened for research use Effects of collection interval, body weight, and season on the hemograms of canine blood donors A method for collecting and storing feline whole blood Effect of human intravenous immunoglobulin on canine monocytes and lymphocytes A clinical trial of a hemoglobin based oxygen carrier (HBOC) fluid in the treatment of anemia in dogs Incidence and severity of anaphylactoid reactions to colloid volume substitutes ): indications, complications and outcomes Blood components: collection, processing and storage Principles of blood collection and processing Intravenous administration of human immune globulin in dogs with immune-mediated hemolytic anemia Successful treatment of a southern Pacific rattlesnake (Crotalus viridis helleri) bite in a caracal (Caracal caracal) Transfusion medicine: the challenge of practical use Hemochromatosis secondary to repeated blood transfusions in a dog Transfusion-associated Babesia gibsoni infection in a dog Comparison of various blood-typing methods for the feline AB blood group system Hemolysis of canine fresh and stored blood associated with peristaltic pump infusion Efficacy of fresh frozen plasma and cryoprecipitate in dogs with von Willebrand's disease or hemophilia A Platelet concentration and hemoglobin function in greyhounds Reports of 355 transfusion associated deaths: 1976-1985 Seroprevalence of babesiosis in greyhounds in Florida Blood transfusion in dogs and cats. Part II. Administration, adverse effects and component therapy Treatment of severe adverse cutaneous drug reaction with human intravenous immunoglobulin in two dogs Comparing chemical restraint and anesthetic protocols used for blood donation in cats: one teaching hospital's experience Evaluation of use of human albumin in critically ill dogs: 73 cases Prothrombotic and inflammatory effects of intravenous administration of human immunoglobulin G in dogs Hemostatic effects of cryoprecipitate in dogs with disseminated intravascular coagulation Technical manual. 11th ed. Bethesda, Md: American Association of Blood Banks New red blood cell antigens in dogs and cats -a welcome discovery Canine plasma therapy Stability of hemostatic proteins in canine fresh frozen plasma units Posttransfusion purpura in a dog with hemophilia A Evaluation of an additive solution for preservation of canine red blood cells Consensus statement on canine and feline blood donor screening for infectious disease Evaluation of canine red blood cells stored in a saline, adenine and glucose solution for 35 days Whole blood transfusions in 91 cats: a clinical evaluation Clinical use of a haemoglobin-based oxygen carrying solution (Oxyglobin) in 48 cats A newly recognized blood group in domestic shorthair cats: The Mik red cell antigen Use of human immunoglobulin in addition to glucocorticoids for the initial treatment of dogs with immune-mediated hemolytic anemia Severe cardiomegaly secondary to anemia in a kitten Comparative stability of canine hemostatic factors in freeze-thaw-cycled fresh frozen plasma Hemolytic reactions produced in dogs by transfusion of incompatible dog blood and plasma A prospective, randomized comparison of Oxyglobin (HB-200) and parked red blood cells transfusion for canine babesiosis