Blood products that may be administered to patients in need of functional platelets include fresh whole blood (FWB), platelet-rich plasma (PRP), or platelet concentrate (PC). Platelet transfusions are indicated in the management of uncontrolled or life-threatening bleeding due to severe thrombocytopenia or thrombopathia. In clinical practice, immune-mediated thrombocytopenia (IMT) is the most common cause of severe thrombocytopenia in dogs. Blood transfusions may be required in dogs with IMT experiencing extensive mucosal surface bleeding, most commonly into the gastrointestinal tract. In such cases, packed red blood cell (PRBC) transfusions are indicated to provide additional oxygen-carrying support. Platelet transfusions are not recommended in most dogs with IMT because the platelets are destroyed within minutes to hours of transfusion. However, in dogs with IMT experiencing uncontrolled or life-threatening bleeding (e.g., into brain, myocardium, or lungs), platelet transfusions may provide short-term hemostasis despite the lack of a measurable increase in platelet count post-transfusion. In addition, bleeding associated with hereditary thrombopathias, such as δ-storage pool disease in American Cocker Spaniels and Glanzmann's thrombasthenia in Great Pyrenees, may be severe and require platelet transfusions to achieve hemostasis. Prophylactic platelet transfusions (PRP or PC) may be considered in patients with hereditary thrombopathias and a known bleeding tendency prior to surgery, with compatible PRBC available in the event of excessive bleeding.
Platelet Storage Lesion
Fresh platelet products have a short shelf-life. FWB may be stored at room temperature (22°C) for up to 8 hours, whereas PRP and PC may be stored at 22°C with continuous gentle agitation for up to 5 days, although most veterinary institutional blood banks prepare platelet products on an as needed basis, due to the limited demand and short shelf-life. The time limitation on storage of platelets at 22°C is partly related to concerns about decreased viability and function, but also to the potential for rapid proliferation of contaminating bacteria at this temperature.1 During storage, platelets undergo a variety of in vitro changes which are collectively referred to as the "platelet storage lesion". This lesion is characterized by a change in platelet shape from discoid to spherical; the generation of lactic acid from glycolysis, with an associated decreased in pH; the release of granule contents; a decrease in various assessments of in vitro platelet function; and reduction in post-transfusion platelet recovery and survival.1 There has been much interest in both human and veterinary transfusion medicine to develop alternative storage options that would increase the shelf-life and availability of platelet products in clinical practice.
Fresh Platelet Transfusions
Whole Blood Derived Platelets
Fresh whole blood is the only readily available platelet-containing product in many veterinary practices. While there are not many clinical indications for administration of FWB, it may be appropriate in the management of patients with anemia and bleeding due to thrombocytopenia or thrombopathia as described above. PRP is harvested from FWB following "soft spin" centrifugation (1000 × g for 4 minutes).2 The supernatant PRP is then expressed into a satellite bag and may be centrifuged further using a "hard spin" (2000 × g for 10 minutes) to produce PC.2 The supernatant (platelet-poor plasma) is then expressed into a second satellite bag, leaving 40-70 ml of plasma for resuspension of platelets.2 In a study evaluating preparation of canine PCs from FWB, the mean platelet yield was 8 × 1010 per PC unit, with a leukocyte content ranging from 1.0 × 108 to 2.3 × 109 WBCs and a HCT ranging from 0.1 to 26.2%, with 62% of PC units having a HCT > 1%.2 As a rough guideline, one unit of fresh whole blood or PRP/PC administered to a 30 kg dog would be expected to increase the patient's platelet count by approximately 10,000/µL. Therefore, depending on the size of the patient, large volumes of FWB (or PRP or PC) may be required to significantly increase a patient's platelet count, a practice which is costly and not feasible on a regular basis.
An alternative approach to preparation of a canine PC from FWB is plateletpheresis. During the plateletpheresis procedure, blood is removed from the donor, anticoagulated with citrate (typically ACD-A) in the extracorporeal circuit, and separated into components by centrifugation, allowing production of a PC, while the other blood components are returned to the donor. Plateletpheresis has become a routine procedure in human blood banks, with an estimated 75% of platelets prepared from single donors by apheresis as opposed to platelets derived from whole blood.3 The advantages of PC prepared by apheresis in comparison to standard PRP/PC prepared from a unit of FWB are greater platelet yield (3 to 4.5 × 1011 vs < 1 × 1011) and negligible RBC and WBC contamination.3
In a study at the University of Pennsylvania evaluating use of an automated blood cell separator (COBE Spectra, Gambro BCT, Lakewood, CO) for collection of platelets from 14 donor dogs weighing 18 to 27.7 kg, a high quality platelet product was safely collected from all dogs, with a mean total yield of 3.3 × 1011 platelets per PC unit. Mean donor platelet count decreased by > 50% post-apheresis but returned to baseline level by day 6. The procedure was generally well tolerated, with no evidence of hypotension. Serum citrate concentration progressively increased, causing decreases in both ionized calcium and magnesium concentrations. Despite prophylactic IV calcium supplementation, ionized calcium decreased to < 1 mmol/L (mean baseline, 1.2 mmol/L) in 10 dogs. Lip licking was noted in 3 dogs, and generalized tremors and ventricular ectopy were noted in 1 dog. The COBE Spectra may be safely utilized for plateletpheresis in dogs weighing approximately 20 kg, but due to the large amount of citrate infused to the donor, calcium supplementation with serial monitoring of ionized calcium is recommended to limit signs of hypocalcemia during the procedure.
As early as 1969, there was strong evidence that refrigerated storage of human platelets had a deleterious effect on platelet viability.4 Storage of human PRP at 4°C for as little as 18 hours with subsequent infusion into the same donor resulted in a markedly reduced survival, with a half-life of 1.3 days in comparison to 3.9 days for platelets stored at 22°C for 18 hours.4 Following this discovery, it became standard procedure for human PRP to be stored at 22°C. However, as room temperature storage is associated with a proliferation of contaminating bacteria, there has been much interest in investigating other storage options. The mechanism responsible for rapid clearance from the circulation of platelets that have been refrigerated prior to transfusion has been recently elucidated. Chilling of human and mouse platelets clusters their von Willebrand factor (vWF) receptors (GPIb), eliciting recognition of platelets by hepatic macrophage complement type 3 (CR3) receptors.5 Refrigerated and rewarmed platelets do not clear rapidly from the circulation of CR3-deficient mice.5 In addition, chilled platelets bind vWF and function normally in vitro, suggesting that the hemostatic and cold-induced clearance functions of the GPIb complex are separable.5 Subsequently, it has been demonstrated that CR3 recognizes GPIbα on cooled platelets through a lectin-mediated interaction with exposed β-N-acetylglucosamine (β-GlcNAc) residues of N-linked glycans, and the interaction between GPIbα and macrophages can be blocked by galactosylation of exposed β-GlcNAc residues.6 Galactosylation of human and murine platelets does not impair their in vitro function and restores survival of chilled platelets in the mouse.6
Further studies testing the feasibility of galactosylation of human platelets under standard blood bank conditions and assessing in vitro functions of refrigerated human platelets following long-term storage are encouraging.7 Platelets were successfully galactosylated following injection of uridine diphosphate (UDP)-galactose into PCs.7 Platelets galactosylated and refrigerated for 14 days maintained their ability to aggregate when exposed to agonists in a standard aggregometry assay, showed less pronounced changes in surface expression of GPIbα compared with room temperature-stored platelets, and were poorly phagocytized by differentiated human monocytic THP-1 cells (an in vitro model for refrigerated platelet clearance).7 Overall, these refrigerated human platelets retained in vitro function better than platelets stored at 22°C.7 However, an abstract presented at the American Society of Hematology meeting in 2006 (published manuscript not yet available) indicated that platelet survival following autologous transfusion of galactosylated human platelets stored at 4°C for 2 days was not improved in comparison to platelets stored at 4°C without galactosylation (average 2.2 days vs 2.9 days, respectively, with average survival of fresh platelets being 6.9 days).8
To date, there are no published reports on survival of chilled canine platelets. In a study of storage of canine platelet concentrates at 22°C and 4°C, it was noted that platelets stored at 22°C lost their ability to aggregate in vitro in response to ADP and collagen after 4 days, whereas platelets stored at 4°C did not lose their ability to aggregate before 8 days of storage.9 Similar loss of platelet function as assessed by the resonance thrombogram was noted over the same time frame.9 Consequently, the authors concluded that canine PC may be stored for 4 days at 22°C or 8-10 days at 4°C.9 However, in light of the effect of chilling on survival of both murine and human platelets as noted above, refrigerated storage of canine PC cannot be recommended at this time without studies documenting adequate post-transfusion survival.
During the past several decades, platelet cryopreservation has been extensively investigated as a means to provide long-term storage and immediate availability of platelet products for transfusion. The American Association of Blood Banks (AABB) has approved cryopreservation of human platelets in 6% DMSO as an acceptable storage procedure, although platelet cryopreservation is largely considered a research technique, and human platelet products for transfusion are still stored at room temperature.1 Based on the numerous studies evaluating cryopreservation of human platelets, it is clear that regardless of the methodology, cryopreserved platelets inevitably demonstrate impaired function and shortened post-transfusion survival in comparison to fresh platelets. Recent studies with human platelets have shown that cryopreservation with 2% DMSO plus Thrombosol (a mixture of amiloride, adenosine, and sodium nitroprusside, second-messenger effectors that inhibit platelet activation) resulted in a significant improvement in platelet recovery and function in comparison to 6% DMSO alone.10,11
The few published reports on canine platelet cryopreservation offer somewhat differing conclusions. In studies of autologous platelet transfusions, canine platelets cryopreserved in 6% DMSO and stored at -80°C for up to one year demonstrated an in vitro recovery of 70% and an in vivo survival 1 to 2 hour post-transfusion 40% that of fresh platelets.12 The half-life of the cryopreserved platelets was 2 days, in comparison to 3.5 days for fresh platelets.12 In addition, autologous cryopreserved platelets administered to lethally irradiated thrombocytopenic dogs were reported to be hemostatically effective, resulting in a reduction in clinical bleeding (GI bleeding, ecchymoses, oozing from venipuncture sites).12 The authors concluded that canine platelets could be "satisfactorily preserved" in 6% DMSO in plasma with storage at -80°C for 1 year.12 A subsequent study evaluating in vitro function of canine platelets cryopreserved in 6% DMSO for 6 months demonstrated marked impairment in platelet aggregation and response to hypotonic shock, leading the authors to question how well these platelets would have performed in vivo.13 However, it has been documented that human platelets with poor aggregation responses in vitro may still have good function and viability in vivo.14
In a study at the University of Pennsylvania, we compared two methods of canine platelet cryopreservation, the standard rapid freeze (-80°C) in 6% DMSO (DMSO) vs. 2% DMSO plus Thrombosol (Thrombosol), to determine if either cryopreservation method produces a platelet product with acceptable function and survival as compared to fresh platelets. Platelet concentrates collected via apheresis from 10 healthy adult mixed breed dogs were each split into 3 units: fresh and cryopreserved in DMSO and Thrombosol. Cryopreserved units were evaluated 1-10 weeks post-freezing. In vitro platelet evaluation included optical aggregometry, baseline and post-thrombin stimulated P-selectin expression, and platelet morphology via phase microscopy. In vivo platelet survival was determined by administration of biotinylated platelets to 30 healthy research dogs (not previously transfused). Both γ-thrombin- and convulxin-induced platelet aggregation were markedly diminished (< 15% increase in light transmittance) in DMSO and Thrombosol cryopreserved platelets in comparison to fresh platelets (> 60% increase in light transmittance). Baseline P-selectin expression was < 1% for fresh and frozen platelets. To determine if platelets could be activated, P-selectin expression was evaluated post-thrombin stimulation. There was no difference in thrombin-induced P-selectin expression between the cryopreserved platelets (~20%), but both exhibited significantly less activation than fresh platelets (44%). The percentage of discoid and spherical platelets was not different between the 3 platelet groups. Fresh biotinylated platelets survived 7-9 days in all 10 recipients, with a mean half-life of 3.8 days. In the DMSO and Thrombosol groups, 4 of 10 dogs in each group had > 1% circulating biotinylated platelets on day 7, but the mean half-life was 1.7 days and 2.5 days, respectively, both significantly less than fresh platelets but not different from each other. We conclude that Thrombosol did not provide any appreciable benefit for maintaining in vitro function or prolonging in vivo survival of cryopreserved canine platelets in comparison to DMSO. Cryopreserved platelets can be activated, as demonstrated by thrombin-induced P-selectin expression, and survive in the circulation long enough to potentially be of benefit in the management of life-threatening hemorrhage in severely thrombocytopenic or thrombopathic patients. However, further studies are needed to assess in vivo function of cryopreserved platelets.
Platelet Transfusion Refractoriness
Dogs have served as a platelet transfusion model in the evaluation of methods of preventing alloimmunization in human patients requiring long-term platelet therapy because of disease-induced bone marrow failure or chemo/radiation therapy.15,16 Weekly transfusion of PCs from single unrelated canine donors resulted in development of alloimmune platelet refractoriness in 95% (20 of 21) of recipients after an average of 3 transfusions.15 When DLA non-identical littermates were used as platelet donors, the percentage of recipient dogs developing platelet refractoriness was reduced to 31% (4 of 13), and the average number of transfusions required for platelet alloimmunization increased to 7.3 transfusions.15 Use of DLA identical littermates as platelet donors resulted in a similar rate of platelet refractoriness (31%, or 4 of 12 dogs) as with DLA non-identical littermates, but the average number of transfusions administered prior to development of platelet refractoriness increased to 14 transfusions.15 In clinical practice, most dogs in need of platelet transfusions will receive platelets from random unrelated donors, so it can be expected that recipients will develop platelet refractoriness after just a few transfusions. Fortunately, most canine patients do not require repeated platelet transfusions, but in dogs with hereditary thrombopathias in particular, repeated platelet transfusions may be necessary to control bleeding tendencies throughout the lives of these patients, and platelet refractoriness may be a significant problem. Leukocyte reduction of platelet products by centrifugation and filtration has reduced platelet alloimmunization in canine studies16 and may be considered for dogs expected to require multiple platelet transfusions. As indicated above, apheresis platelets are leukoreduced and do not require filtration.
1. Brecher ME, ed. AABB Technical Manual, 15th ed, 2005.
2. Abrams-Ogg ACG, et al. Am J Vet Res 1993;54(4):635.
3. Vassallo RR, et al. Transfusion Med Rev 2004;18(4):257.
4. Murphy S, et al. N Engl J Med 1969;280(20):1094.
5. Hoffmeister KM, et al. Cell 2003;112:87.
6. Hoffmeister KM, et al. Science 2003;310:1531.
7. Babic AM, et al. Transfusion 2007;47:442.
8. Slichter SJ, et al. Blood 2006;108:175a.
9. Klein A, et al. Beil Munch Tierarztl Wschr 1999;112:243.
10. Currie LM, et al. Transfusion 1998;38:160.
11. Currie LM, et al. Br J Haematol 1999;105:826.
12. Valeri CR, et al. Cryobiology 1986;23:387.
13. Allyson K, et al. Am J Vet Res 1997;58(11):1338.
14. Murphy S, et al. Transfusion Med Rev 1994;8(1):29.
15. Slichter SJ, et al. Br J Haematol 1986;63:713.
16. Slichter SJ, et al. Blood 2005;105:847.