Veterinary clinicians play a key role in providing safe and effective transfusion therapy. Blood typing is clinically important to ensure blood compatibility and, therefore, is recommended for any dog in need of a transfusion or considered to become a blood donor. Moreover, previously transfused dogs also should be crossmatched. Unless blood typing is performed regularly in practice, blood may be sent to a clinical pathology laboratory for typing. Different viewpoints exist regarding the extent and methods used for compatibility testing.

Canine Blood Types

Blood types are genetic markers on erythrocyte surfaces that are antigenic and species specific. A set of blood types of two or more alleles makes up a blood group system. Dogs have likely more than a dozen blood group systems mostly known as dog erythrocyte antigens (DEA). However, there is no DEA 2 blood group and some may be rather labeled high frequency or common red blood cell (RBC) antigens (e.g., DEA 4) and some have not yet received a DEA designation (e.g., Dal). Canine erythrocytes are either positive or negative for a blood type (e.g., DEA 4+ or DEA 4-), and these blood types are likely codominantly inherited. The DEA 1 system was thought to be an exception with DEA 1.1 (A1), DEA 1.2 (A2) and potentially DEA 1.3 (A3) being allelic. Thus, a dog could apparently be DEA 1.1+ or DEA 1.1- and DEA 1.1- dogs can be DEA 1.2+ or DEA 1.2-. However, these studies were based upon weak polyclonal antibodies (DEA 1.1 and 1.X) requiring Coombs’ reagents. Recent studies with a monoclonal antibody showed that the DEA 1 blood group is a continuum from DEA 1- to weakly to strongly DEA 1+; hence DEA 1.2 typing is no longer offered. The degree of DEA 1 expression is constant and DEA 1+ appears to be dominantly inherited. A recent survey in North America indicates that most dogs are either DEA 1- or strongly DEA 1+ with fewer dogs being weakly to moderately DEA 1+. The biochemical structure of the DEA 1 remains still unknown, but a genome wide association study has identified a likely single locus.

Recent surveys revealed that the Dal- type is not restricted to Dalmatians but is also seen in Doberman pinschers, Lhasa apsos, and Shih Tzus and thus typing for this blood type is becoming more important, particularly for those requiring multiple transfusions. In a related study dogs from North America were screened for two new blood types, preliminarily called Kai 1 and Kai 2. Most dogs were Kai 1+ and only few dogs were Kai 2+ or Kai 1-/Kai 2-. The clinical importance is yet to be determined, albeit anecdotally dogs can develop anti­-Kai 1 alloantibodies. The PennGen Laboratory currently offers Dal and Kai 1 and Kai 2 typing.

The clinically most important canine blood type is DEA 1, which elicits a strong alloantibody response after sensitization of a DEA 1- dog by a transfusion and thus can be responsible for a transfusion reaction in a DEA 1- dog previously transfused with DEA 1+ blood. It is currently unknown if DEA 1- dogs are equally sensitized by weakly to strongly DEA 1+ blood, or if weakly DEA 1+ dogs are sensitized by strongly DEA 1+ blood. Furthermore, transfusion reactions against other blood types or common antigens have rarely been observed and reported. They include reactions against the DEA 4, Dal, Kai 1, and other common RBC antigens; other clinically important blood types may be found in the future. No reagents currently are available against several antigens or are only available on a limited basis, and additional blood types continue to be recognized. Only limited surveys on the frequency of these blood types have been reported, which suggest possible geographic and breed-associated differences.

Strongly antigenic blood types are of great clinical importance because they can elicit a potent alloantibody response. These alloantibodies may be of the immunoglobulin G (IgG) or IgM class and may be hemagglutinins or hemolysins. Based upon experimental and clinical data, dogs can become sensitized after receiving a mismatched transfusion (i.e., a blood unit positive for one or more blood types not found on the recipient’s RBCs). There are no clinically important, naturally occurring alloantibodies (also known as isoantibodies) present before sensitization of a dog with a transfusion. Sensitizing dogs in experimental studies in the 1950s led to the documentation of some transfusion reactions caused by blood group incompatibilities and to the characterization of new blood types.

Clinically, the most antigenic blood type in dogs is the DEA 1. Transfusion of DEA 1+ RBCs to a DEA 1- dog invariably elicits a strong alloantibody response. Following a first transfusion, anti-DEA 1 antibodies develop after more than 4 days and may cause a delayed transfusion reaction (rarely clinically documented). However, a previously sensitized DEA 1- dog can develop an acute hemolytic reaction after a second transfusion of DEA 1+ blood. Transfusion reactions also may occur after a sensitized dog receives blood that is mismatched for a RBC antigen other than DEA 1 (e.g., DEA 4 and Dal). However, in most cases the incompatible blood type has not been determined. Because administration of a small (<1 ml) amount of incompatible blood can result in life-threatening reactions, the practice of giving small “test volumes” of donor blood to assess blood-type compatibilities is unacceptable. In contrast, pregnancy does not cause sensitization in dogs, because of a complete placenta, and does not induce alloantibody production; thus, dogs with prior pregnancies can be used safely as blood donors.

Canine Blood-Typing Procedures

Because of the strong antigenicity of DEA 1, typing of donors for DEA 1 is recommended. Whenever possible, the recipient also should be typed to allow the use of DEA 1+ blood for DEA 1+ recipients. Canine blood typing tests are generally based on serologic identification by agglutination reactions but chromatographic strip methods are also offered. Originally serum from sensitized dogs has been used for typing, but such polyvalent alloantibodies vary from batch to batch, may require Coombs’ reagent to enhance agglutination, and may not be always available and are, therefore, not optimal. Two monoclonal antibodies against DEA 1 have been developed. The gel column technology, widely used in human blood banking, was found to be an excellent standardized laboratory method (DiaMed), but is unfortunately no longer commercially available. A blood typing card has been available with modifications since the mid-1990s as a simple in-practice kit to classify dogs as DEA 1- or DEA 1+ (degree of reaction can vary). a standardized simple immunochromatographic technique became available in the mid-2000s from Alvedia. Another cartridge with a similar strip technique was introduced by DMS/AgroLabo, but has not been evaluated. Moreover, a third cartridge method in which blood flows through the cartridge is also available (DMS/Abaxis) but seems to produce inconsistent results.

Polyclonal reagents against other DEA types are currently only available on a limited basis for DEA 3, 4, and 7 from Animal Blood Resource International (prior Michigan State University and Midwest Blood Services). And only limited anti-Dal reagents from sensitized dogs are currently available in a couple of laboratories like Montreal University and PennGen, monoclonal anti-Kai 1 and anti-Kai 2 alloantibodies have been developed in South Korea. DEA 1 typed and matched patients in need of a transfusion may be typed for DEA 4, Dal, and Kai 1/2, which may then permit the localization of a type-matched donor dog.

Caution should be exercised whenever the patient’s blood is autoagglutinating or has a low hematocrit (<10%). If autoagglutination is not too severe, it does not appear to affect the Alvedia strip technique because only free RBCs are moving up the strip. Clinicians and technicians should check for autoagglutination of blood with buffer/sailne on a slide or the card. Autoagglutinating blood may be first washed three times with ample physiological saline to overcome the apparent autoagglutination similar to what is done for the Coombs’ and crossmatch testing. However, if autoagglutination after three washes persists at more than 1+, it is considered to reflect true autoagglutination, which may preclude typing (as well as Coombs’ testing and crossmatching), because it always looks like DEA 1+ blood. In such circumstances, DEA 1- blood should be used, until the patient does not agglutinate anymore and can be retyped. DEA 1+ blood from severely anemic animals may not agglutinate when exposed to the anti­DEA 1 or other reagents because of a prozone effect. In these cases, some of the patient’s plasma may be discarded before applying a drop of blood onto the card. Finally, recently transfused dogs may display a mixed field reaction, with only the transfused or recipient cells agglutinating if they were DEA 1 mismatched.

Blood Crossmatching Test

Whereas blood typing tests reveal the blood group antigens on the red blood cell surface, blood crossmatching tests assess the serologic compatibility or incompatibility between donor and recipient. Thus, the crossmatch test checks for the presence or absence of naturally occurring and induced alloantibodies in serum (or plasma) without determining the blood type and thus does not replace blood typing. These antibodies may be hemagglutinins and/or hemolysins and can be directed against known blood groups or other RBC surface antigens. Many laboratories commonly use a standardized tube crossmatching procedure, but the interpretation of the agglutination reaction is highly variable. The crossmatching test requires some technical expertise, may be accomplished through a veterinary laboratory along with blood typing, and is done with washed EDTA­anticoagulated blood from recipient and potential donor(s.) The DiaMed gel column technique and more recently the in-clinic DMS gel tube assay have been evaluated and were found to be simple, sensitive, and standardized methods to crossmatch dogs and cats. In addition, Alvedia introduced a simple strip crossmatch test with a Coombs’ phase.

The major crossmatch tests search for alloantibodies in the recipient’s plasma against donor cells, whereas the minor crossmatch test looks for alloantibodies in the donor’s plasma against the recipient’s RBCs. Generally, tube segments from collection bags are used for this purpose in dogs. The presence of autoagglutination or severe hemolysis may preclude the crossmatch testing. A major crossmatch incompatibility is of greatest importance, because it predicts that the transfused donor cells will be attacked by the patient’s plasma, thereby causing a potentially life-threatening acute hemolytic transfusion reaction. Because fatal reactions may occur with less than 1 ml of incompatible blood, compatibility testing by administering a small amount of blood is not appropriate; this has been shown in experimental studies to potentially result in fatal reactions. A minor crossmatch incompatibility should not occur in dogs if canine donors have not been transfused previously and is of lesser concern because donor’s plasma volume is small, particularly with packed red cell products, and is diluted markedly in the patient. Do not use previously used dogs as donors.

The initial blood crossmatch between two dogs that have never before received a transfusion should be compatible, because dogs do not have naturally occurring alloantibodies. Therefore, a crossmatch may be omitted before the first transfusion in clinical practice for dogs. Because the crossmatch does not determine the blood type of the patient and donor, a compatible crossmatch does not prevent sensitization of the patient against donor cells within 1 to 2 weeks. Thus, previously transfused dogs should always be crossmatched, even when receiving again blood from the same donor. The time span between the initial transfusion and incompatibility reactions may be as short as 4 days and the induced alloantibody can last for many months to years (i.e., years after the last transfusion alloantibodies may be present). Again, a blood donor never should have received a blood transfusion to avoid sensitization. The practice of transfusing patients with the least compatible unit does not have any scientific basis. Nevertheless, some minor agglutination results in crossmatching a patient may be unrelated to alloantibodies and unspecific (e.g., patient’s RBC damage by uremia and other illnesses, donor cells after extended storage of unit in the refrigerator). Of course, any patient with true/persistent autoagglutination may not be matched to any donor.

Although transfusion of blood and its components is usually a safe and temporarily effective form of therapy, there is always a risk for potential hazards. Adverse reactions usually occur during or shortly after the transfusion and can be due to any component of whole blood. Most transfusion reactions can be avoided by carefully selecting only healthy donors; using appropriate collection, storage, and administration techniques; performing blood typing and crossmatching; and administering only the needed blood components.

Transfusion Reactions

While transfusion of blood and its components is usually a safe and temporarily effective form of therapy, there is always a risk for potential hazards. Adverse reactions usually occur during or shortly after the transfusion and can be due to any component of whole blood. Most transfusion reactions can be avoided by carefully selecting only healthy donors, using appropriate collection, storage, and administration techniques, performing blood typing and crossmatching, and administering only needed blood components. The most common clinical sign of transfusion reaction is fever, followed by vomiting, and hemolysis. Hemolytic transfusion reactions can be fatal and are, therefore, most important, while fever and vomiting are usually self-limiting. Adverse effects of transfusions can be divided into non-immunologic (pyrogen-mediated fever, transmission of infectious agents, vomiting, mechanical hemolysis, congestive heart failure, hypothermia, citrate toxicity, pulmonary complications) and immunologic reactions (acute and delayed hemolytic transfusion reactions, urticaria to anaphylaxis, acute respiratory distress, graft versus host disease). Note that some clinical signs may be caused by both mechanisms. Despite the variety of blood types and the limited degree of compatibility testing in clinical practice, transfusion reactions are rarely reported.

Blood Donors and Sources

Many larger veterinary hospitals have permanent canine and/or feline blood donors to cover their transfusion requirements or in case fresh whole blood or platelet-rich plasma (concentrate) is needed. Several larger voluntary blood donor programs have emerged with client or staff owned dogs. More than a dozen commercial canine blood banks have been established in the United States and deliver overnight blood products. Autologous (self) transfusion refers to the donation of blood by a patient four weeks to a few days prior to surgery when major surgical blood loss is anticipated. Blood can also be collected immediately prior to surgery. The patient will be hemodiluted with crystalloid and colloid solution and receives the blood when excessive bleeding occurs or after surgery. Autotransfusion is another autologous transfusion technique in which freshly shed blood salvaged intraoperatively or following trauma can be reinfused after careful filtering.

Blood donors should be young adult, lean, and good-tempered animals, and weigh at least 23 kg for dogs (to donate 450 ml); have no history of prior transfusion; have been regularly vaccinated and are healthy as determined by history, physical examination, and laboratory tests (complete blood cell count, chemistry screen, and fecal parasite examination every 6–12 months) as well as free of infectious diseases (testing depends on geographic area but may include regular microfilaria, Brucella, hemomycoplasma, Babesia, Ehrlichia, Anaplasma, Borrelia, Leishmania spp. testing in dogs. Donors should receive a well-balanced, high performance diet, and may be supplemented twice weekly with ferrous sulfate (Feosal, 10 mg/kg), if bled frequently. Packed cell volume (PCV) or hemoglobin (Hb) should be >40% and >13 g/dl in canine donors.

Blood Collection and Component Preparation

Canine donors are generally not sedated. Blood is collected aseptically by gravity or blood bank vacuum pump from the jugular vein over 5- to 10-minute period. Plastic bags containing citrate-phosphate-dextrose­adenine (CPD-A1) with or without satellite bags for blood component separation are optimal. These commercial blood bags represent a closed collection system in which the blood does not come into contact with the environment at any time during collection or separation into blood components, thus minimizing the risk of bacterial contamination and allowing storage of the blood products. The maximal blood volume to be donated is 20 ml blood/kg or one regular blood bag unit of 450±45 ml per ≥25 kg dog.

Blood components are prepared from a single donation of blood by centrifugation generally within 8 hours from collection; thereby, fresh whole blood can be separated into packed red cells, platelet-rich plasma or concentrate, fresh frozen plasma, and cryoprecipitate and cryo-poor plasma. Fluctuations in storage temperature significantly alter the length of storage; thus, temperature needs to be monitored and the refrigerator/freezer are not too frequently opened. Partially used or opened blood bags should be used within 24 hours because of the risk of contamination.

Administration of Blood Products

For routine transfusion in the treatment of anemia, it is not necessary to warm blood after removal from the refrigerator. A temperature-controlled waterbath (37°C) is ideal to warm frozen blood products. A warm water bowl in which the water is periodically changed may be used to warm blood products. Care should be taken to maintain absolute sterility and to not overheat the blood products.

Blood bags are connected to blood infusion sets that have an in-line microfilter. A long (85 cm) blood infusion set with a dripping chamber and a short infusion set for small dogs to connect with syringes are available. Use a latex-free infusion sets for platelet administration to avoid aggregation. Microfilter with 170 µm pores are commonly used to remove clots and larger red cell and platelet aggregates. Finer filters with 40 µm pores will remove most platelets and microaggregates and clog after 100 ml. Leukocyte reduction filters (expensive) may be used to decrease febrile adverse reactions to WBC components prior to storage.

Blood components are best administered intravenously with an indwelling catheter (16–22-gauge depending on size of animal). An intramedullary (or intraosseous) infusion at the trochanteric fossa (or other sites) may be used when no venous access can be obtained while the intraperitoneal administration is not recommended. Avoid concurrent administration of drugs or fluids other than physiologic saline through the same catheter in order to prevent lysis of erythrocytes and blood coagulation.

Rate of transfusion depends on the hydration status, degree of anemia, and general health condition of an animal. Initial rate is slow, starting with 1–3 ml over the first 5 minutes to observe for any transfusion reactions, even with blood typed and/or crossmatched transfusions. In animals with cardiac failure, do not exceed 4 ml/kg/h. Transfusion of a single bag should be completed within 4 hours to prevent functional loss or bacterial growth. Volume of blood component to be administered depends on the type of deficiency and size of the animal. In anemia: volume (ml) of whole blood = 2 x PCV rise desired (%) x body weight (kg) or in other words, administration of 2 ml whole blood/kg body weight raises the PCV by 1%. If packed red cells are used without prior resuspension in a red cell preservative, closer to half the volume is administered, since packed red cells have a PCV of 70–80%. In the absence of bleeding and hemolysis, at least 80% of transfused erythrocytes survive 24 hours (required blood bank standard) and transfused erythrocytes may be thereafter expected to have a normal life-span (∼110 days in dogs). Response to transfusion is carefully monitored by obtaining PCV/TP readings prior to, immediately, 2, 4, 6, and 24 hours post-transfusion, and observing the clinical parameters of a patient.

In thrombocytopenia or thrombopathia, platelet transfusions are only used with life-threatening bleeding. One unit of platelet concentrate, platelet rich plasma, or fresh whole blood will increase the platelet count by 10,000/µL in a recipient weighing 30 kg. Platelet counts are monitored prior, 1 hour and 24 hours after the platelet transfusion.

In coagulopathies and von Willebrand’s disease, fresh frozen plasma at 6–10 ml/kg is an initial dose to stop bleeding or avoid excessive bleeding during surgery. In some cases, larger volumes may be needed to control bleeding. Depending on the coagulopathy, repeated administration of FFP may be required. Because of the short half-life of factor VII and VIII and von Willebrand factor, deficient animals need to be treated twice to four times daily. Other coagulopathies may be treated daily. Cryoprecipitate at a dose of 1 Cryoprecipitate unit/10 kg or 2–4 ml/kg body weight twice daily is ideal to treat hemophilia A and von Willebrand’s disease. Plasma support should be provided for an additional 1–3 days after the bleeding has been controlled to allow for healing and prevent rebleeding.

Supported in part by a grant from the NIH (OD 010939). The author’s laboratory PennGen is offering quantitative DEA 1, Dal, and Kai typing. Alvedia and DMS Laboratories kindly provided reagents and kits for the author's studies.