Since the first dog to dog blood transfusion in 1665, transfusion medicine has evolved greatly, with many practices using volunteer donors or having their own blood bank in house (Branquinho et al, 2011). Since 2007 canine blood component products such as packed red blood cells (PBRCs), fresh frozen plasma, frozen plasma, cryo-precipitate and cryo-supernatant have been commercially available in the UK from the Pet Blood Bank UK (PBBUK) (Barnett, 2012). Historically whole blood transfusions were used in cases of hypoxic anaemia or hypovolaemic trauma (Tobin, 2011), but with the advent of component product availability, and with new evidence-based research, the applications for blood and blood products has evolved. Transfusion products are now regularly used in patients with clotting factor deficiencies, septic peritonitis, vasculitis and protein losing enteropathies and nephropathies (Boag and Walton, 2011a). Recent papers published herald the discovery of two new canine blood groups; Kai 1 and Kai 2 (Euler et al, 2016a). Current research is also ongoing for red blood cell (RBC) storage lesions and the ensuing implications to practice, xenotransfusions (Euler et al, 2016b), autologous transfusions with or without cell salvaging devices (Kellett-Gregory et al, 2013), leukoreduction (Kisielewicz and Self, 2014) and canine species-specific albumin transfusions (Craft and Powell, 2012). Goulet et al have studied the prevalence of the Dal blood group in certain canine breeds (Goulet et al, 2017) and Bebar et al (2014) have written of the identification of Von Willebrand disease (VWD) in a cat, which may well be of relevance in the future. Research in these areas closely follows that of research in human transfusion medicine and are areas which have been subject to controversy recently (Kisielewicz and Self, 2014). In human medicine advances have been made such as freeze dried plasma for use in the military field (Pusateri et al, 2016). With veterinary medicine traditionally following in the footsteps of human medicine it is reasonable to consider that similar developments in the veterinary field may be a possibility in the near future.
Plasma factor preservation
One area in transfusion medicine currently undergoing research concerns the storage times of blood, which has applications to both banking and clinical practice. Walton et al (2014) looked at whether the processing of collected blood could be delayed without detrimental effects to the end products. This would allow donor sessions to be held at further distances from the processing centre, enabling access to donors from new areas of the country, ultimately increasing stock and potentially lowering the price of blood products. They concluded that the time could lengthen from 8 to 24 hours without affecting the product. This study was based on extensive human studies by Cardigan et al (2011), but was unique in the veterinary field and was a relatively small sample size so further investigations would be recommended.
Storage lesions in RBCs
There is currently controversy over RBC storage, due to the discovery of storage lesions, which may have implications on morbidity and mortality. A study by Wang et al (2014) used blood stored for either 7 or 42 days and concluded that if the older blood was transfused for resuscitation purposes the risk of septic shock may increase. This concurred with findings by Kisielewicz and Self (2014) who reported that similar results had been found in human studies using 21 day old blood. However, they also stated that some human studies had failed to find a link between aged blood and an increase in morbidity and mortality. Obrador (2015) found that using blood older than 14 to 28 days was associated with increased morbidity and mortality, but did not specify if this was in certain situations such as resuscitation. A study using 14 day old blood by Hann et al (2014) concluded that older blood may have a negative outcome when used in patients with immune-mediated haemolytic anaemia, but not in other scenarios. A study by Soloman et al (2015) stated that there were no detrimental effects seen when using older PRBCs in certain scenarios but this study had only a small sample size of 12 dogs, whereas Wang et al (2014) had 48 dogs and Hann et al (2014) had 3095 dogs in the study. Crombie (1996) stated that there was a risk in studies with small sample sizes of clinically significant facts being missed, and that chance is more likely to have an effect, meaning that the larger studies are potentially more robust.
Currently blood in both human and veterinary medicine is stored for between 35 and 42 days depending on the preservative used (Figure 1). Hann et al (2014) highlighted the fact that most blood banks use a first in first out policy, to ensure blood does not exceed its use by date. This is a practice that may need to alter to a case specific policy depending on the outcome of further research. Holowaychuk and Musulin (2015) looked directly at how blood products were selected at two US veterinary hospitals; one selected blood on a case based decision and the other automatically selected the oldest product first. This study could not draw conclusions about the clinical effects so was of limited use; but they did find that wastage of out of date PRBCs was higher in the hospital selecting on a case basis which had cost implications for blood banking and storage. A study in Istanbul by Bala and Ozcan (2016) looked at freezing canine RBCs and found no negative effects to the cells, which may be an area for future research as the small sample size in this study may have influenced the results as facts may have been missed, or there is the possibility of chance affecting the results (Crombie, 1996).
Further research is required before protocol changes are recommended as currently there is no agreed cut off point for the age of the blood used for transfusion, and there are large variations between the scenarios in which blood is used. Many of the previous studies raised ethical and welfare issues, as they involved dogs tested under laboratory settings with subsequent euthanasia post study, for example the study by Wang et al (2014). In contrast the study by Hann et al (2014) had the largest sample size and involved dogs treated in a clinical setting at a hospital, and the study by Walton et al (2014) was approved by the ethics committee for their research model, and used units already collected for processing and made available for use.
Leukoreduction
Another interesting area of development is leukoreduction, which could potentially be of significance to both canine blood banking, and transfusion procedures. Leukoreduction involves removing platelets and leukocytes from donated whole blood prior to storage or transfusion (Kisielewicz and Self, 2014). It is common practice in human transfusion medicine, but still a relatively new subject in veterinary medicine (Kisielewicz and Self, 2014). It is carried out because leukocytes release cytokines in storage which contribute to the development of storage lesions, adversely affecting storage times for PRBCs (Kisielewicz and Self, 2014). Studies in human medicine have linked the use of leukoreduced cells to a decreased risk of post-operative infections, acute respiratory distress syndrome and a decrease in morbidity (Pertinhez et al, 2016). Several countries, including the UK, use leukoreduced blood for all human transfusion cases. There is evidence that leukoreduction could reduce acute transfusion reactions, this would support the case for leukoreduction in veterinary medicine, especially as a study by Bruce et al (2015), showed that 22% of dogs receiving a transfusion of PRBCs experience a transfusion reaction. With nearly a quarter of transfused dogs being affected by transfusion reactions, it may be worth investigating if using leukoreduction as part of canine blood banking protocol would be worth the additional costs. Jagodich et al (2016) felt there was sufficient evidence to support leukoreduction, in order to reduce the incidence of inflammatory transfusion reactions, and found that only 4% of respondents to a US internet survey currently used leukoreduced components. It is understood that PBBUK is constantly looking into any developments that may improve either the blood products or the services available (personal communication); however the current reported transfusion reaction rate across all the products is only 0.2%, and the impact of leukoreduction is currently unknown, therefore the PBBUK at this time has made an active decision not to use leukoreduced blood, based on the enhanced costs of processing a unit, the increase in processing time, potential loss of units during the leukoreduction process and the reduction in available PRBCs for use post leukoreduction. The Royal Veterinary College (RVC) however, carries out leukoreduction on all products in their in-house blood bank, which is likely to reduce febrile reactions in transfusions (Jagodich et al, 2016), and has stated that although not usual practice in this country, they consider it best practice (RVCS, 2016).
Leukoreduced cells have been reported to be much more stable in storage (Pertinhez et al, 2016), making leukoreduction especially beneficial in blood banking. Ergul Ekiz et al (2012) looked specifically at oxygen transportation post transfusion and concluded that leukoreduced stored blood may actually improve oxygen delivery in comparison to non-leukoreduced stored blood. This is of particular significance in patients with acute anaemia or undergoing general anaesthesia, and adds weight to the evidence in favour of leukoreduction. A study at Istanbul University declared that leukoreduced blood products had a longer shelf life (Bala et al, 2016), which is in accordance with Pertinez et al (2016). They concluded that leukoreduced PRBCs could potentially be frozen to increase storage times therefore potentially negating the additional costs of leukoreduction in stored blood by decreasing the wastage.
The inflammation seen in transfusion reactions was often contributed to age-related conditions of stored blood and it has been suggested that leukoreduction would have no effect on this (Kisielewicz and Self, 2014). Kisielewicz and Self (2014) questioned the cost to benefit ratio of leukoreduction and concurred with Holowaychuk and Musulin (2015) that the age of the transfused PRBCs was the risk factor regardless of whether leukoreduction had been carried out or not. All the studies agreed that further investigation is needed in this area to establish whether the increased costs of leukoreduction would be feasible in UK canine blood banks. According to some the cost increases are small and leukoreduction would improve the safety of canine transfusion medicine (McMichael, 2010), and certainly the RVC concur with this as they already incorporate leukoreduction into their protocol.
Autologous transfusion
There are very few studies in veterinary medicine looking at autologous transfusions, although they are common place in human transfusion medicine. An autologous transfusion is where blood is collected from, and then used for transfusion back into the same patient (Hofbauer et al, 2016). In veterinary medicine autologous transfusion can involve the use of a cell salvaging device to wash the collected cells in saline to remove any additives, such as anticoagulants, prior to the cells being transfused back into the patient (Kellett-Gregory et al, 2013).
Another method of autologous transfusion is more basic, requiring less equipment or specialised training, and involves pooled blood being collected using a needle and syringe, and then being transferred into collection bags before transfusion back into the same patient (Higgs et al, 2015). There is also a UK developed cell salvaging device called a Hemosepvet (Advancis Veterinary) which has become recently available.
Autologous transfusion is controversial; in human literature studies have stated evidence in favour of certain techniques, but in the veterinary field there is very little research or evidence looking specifically at canine and feline autologous blood transfusion techniques. This may explain why it appears to be a fairly infrequent transfusion method at this time. Hirst and Adamantos (2012) concluded that transfusion of cells salvaged intra-operatively from the abdominal cavity by suction and processed through a cell salvaging device was successful in situations of large volume haemorrhage, however there were only three dogs in the study all of which had also had homologous transfusions before or after the autologous blood. Therefore, the study concluded that they were unable to comment on possible benefits of these transfusions other than the decreased reliance on banked blood.
Kellett-Gregory et al (2013) discussed the use of a cell salvaging device to wash red blood cells in saline and highlighted the advantages of autologous transfusions over homologous transfusions, these transfusions having less risk of an inflammatory transfusion reaction caused by storage lesions in stored blood products; Kellett-Gregory et al (2013) also stated that cell salvaging devices and leukoreduction resolved two of the historical controversies around autotransfusion: that of bacterial contamination and neoplastic cell spread (Kellett-Gregory et al, 2013). These authors have also highlighted the lack of studies into whether cell salvaging devices would remove endoparastic larvae, which in light of the increasing cases in the UK of vector transmitted blood-borne diseases such as babesiosis (Helm, 2013) and angiostrongylosis (Eastwood, 2013), is likely to be of clinical significance in the future. The use of cell salvaging devices does incur additional costs and staff training, and would need to be shown to be cost effective compared with homologous banked products in order for autologous transfusions to be considered for regular use in veterinary practice. In one study the costs in the human field of using a cell salvaging device was shown to be equal to the cost of two cross-matching procedures (Hirst and Adamantos, 2012), and it has been calculated that the cost of cell salvaging in a patient was less than the cost of two banked units of PRBCs (Kellett and Gregory et al, 2013).
Risks associated with cell salvaging devices include the possibility of disseminated intravascular coagulation, acute respiratory distress syndrome, renal failure and death, which has given rise to the term salvage syndrome. However, Kellett-Gregory et al (2013) contributed this syndrome to problems with using the equipment and inadequate staff training. It has also been recognised that risks were associated with technique and equipment malfunction or failure (Hirst and Adamantos, 2012). An Austrian study highlighted the fact that changes occurred to canine blood cells following processing in a cell salvaging unit, and recommended further studies into this new area of concern over autologous transfusions (Hofbauer et al, 2016).
An alternative to using a cell salvaging device was investigated in a retrospective study (Higgs et al, 2015); in this study the blood for autotransfusion was collected by direct aspiration from either an abdominal or a thoracic cavity in a variety of different clinical scenarios over a 5-year period. The authors concluded that autologous blood collection and transfusion without a cell salvaging device or a leukoreduction filter was suitable for emergency situations where there is a large volume of blood loss. The study did however have limitations as it was a retrospective study, and 68% of the dogs involved also received homologous banked blood products in addition to the autologous transfusion, making it difficult to attribute any clinical effects to the autologous blood alone.
Some human studies suggested that cell salvaged autologous transfusions were potentially advantageous over homologous transfusions due to reduced risks of acute transfusion reactions or transfusion of mismatched blood (Hirst and Adamantos, 2012; Higgs et al, 2015). One of the greatest risks in human blood transfusions is the administration of wrongly matched or incorrectly selected blood products, but to the author's knowledge there is no evidence of the risk of this occurrence in veterinary practice, possibly due to the smaller number of transfusions performed concurrently, and the likelihood of fewer staff members being involved in the procedure.
The veterinary studies available at this time are unable to conclusively determine any benefits of autologous transfusions, as almost all of the patients involved in the studies concurrently received stored blood products. Therefore, the main conclusion gained from these studies was that the use of autologous blood decreased the amount of banked blood products required, thereby reducing reliance on blood bank resources at a time when demand is ever increasing.
Xenotransfusion
Xenotransfusion is the transfusion of blood or blood component products from one species to another. In human medicine there are both historical and current studies published involving blood transfusions using an animal donor and a human recipient, the conclusions from these studies have been varied with some success and some fatal outcomes (Weingram, 2014). Recent studies in veterinary xenotransfusion medicine have looked at the viability of transfusing canine whole blood or blood components into feline recipients. Cats have always presented a greater challenge in transfusion medicine than dogs (Figure 2) both on the donor and the recipient side, and due to manufacturing problems there was a loss from the market of products such as Oxyglobin, and stabilisation of anaemic cats has become more challenging. Transfusions in cats may cause a fatal reaction due to naturally occurring antibodies present from birth, meaning there is no safe option for transfusion without prior blood typing and cross matching both the donor and the recipient (Branquinho et al, 2012). Feline blood donors should be health screened for occult disease and viruses, and unlike dogs usually require sedation as a means of chemical restraint for the donation procedure. The sedation along with the risk of removing a significant volume of circulating blood from a relatively small animal, brings increased risks and ethical considerations in comparison with canine donations (Harvey, 2011). Feline blood banking therefore also has ethical and welfare implications to consider.
In the US banked feline blood is widely available, but in the UK although there are small feline blood donor programmes within some referral institutes, and some UK practices have their own in-house donor schemes which they use, banked feline blood is not as readily available. PBBUK has an active programme looking specifically into the banking of feline blood, but currently does not store and supply feline blood or blood products (personal communication). The complications and associated cost implications of sourcing, screening, typing and collecting a donation mean that feline blood transfusions is beyond the scope of some practices or owners, especially in emergency situations where whole blood or blood components are needed for immediate use (Weingram, 2014).
Xenotransfusion of canine blood to feline patients has been evaluated in situations where feline blood was not an option (Weingram, 2014; Klainbart, 2015). Weingram (2014) highlighted the benefits of xenotransfusion from canine to feline as being more easily accessible and negating the need for AB blood typing in cats. Weingram (2014) used 10 ml/kg of canine PRBCs in a 4-week-old kitten with parasite induced anaemia, no blood typing or cross matching was carried out. They concluded that multiple xenotransfusions would most likely be fatal but a one off, un-typed, un-cross matched, canine to feline PRBC transfusion was successful in this case as a lifesaving procedure. Weingram (2014) further theorised that there may be benefits of xenotransfusion in cases of feline immune-mediated haemolytic anaemia, as the cat's anti-RBC antibodies would not attack the foreign erythrocytes in the same way and therefore haemolysis would not occur.
A review of nine cases of canine to feline xenotransfusion concurred with the opinion of Weingram (2014) that xenotransfusion was a legitimate consideration in emergency situations where compatible feline blood was not available or there was not time, funding or facilities for blood typing and cross matching prior to transfusion (Klainbart et al, 2015). However, in contrast to these studies, current research carried out in the US reported that cats had naturally occurring allanto-antibodies against canine erythrocytes (Euler et al, 2016b). The presence of these antibodies accounted for the well documented short lifespan of 4 days for canine erythrocytes transfused into feline patients, causing the intravascular haemolysis and the acute haemolytic transfusion reactions observed (Euler et al, 2016b). The study concluded that xenotransfusion was not a suitable solution due to incompatibility issues and recommended the use of Oxyglobin in emergency situations where typed and cross matched blood was unavailable as it was the haemolysis of the canine RBCs causing the harmful reaction.
In the recent xenotransfusion studies the sample sizes have been small. Clearly, further research is required into this potentially promising area of transfusion medicine, especially in relation to alleviating the current difficulties encountered in practice with sourcing and supplying suitably compatible banked products for use in cats in emergency situations.
Recommendation for further studies
In some areas of transfusion medicine large retrospective studies had been undertaken (Hann et al, 2014), but in other areas such as xenotransfusion very small sample sizes were used, in some studies only single cases (Weingram, 2014). This can often lead to clinical data being missed (Crombie, 1996); studies using larger sample sizes would be advised. Retrospective studies over 5 years such as that by Higgs et al (2015) which looked into autologous transfusions stated limitations due to the retrospective nature of the study and the fact that there had been a lack of data collected at the time of the study; another retrospective study by Loyd et al (2016) on xenotransfusion also had limitations, as medications administered during the trial varied among the subjects, and different sources for products were used. With the drawbacks of retrospective studies being highlighted by researchers, recommendations for further study would be for prospective multicentre cohort trials with standardised protocols to improve the quality of the results. Literature on effects of aged PRBCs were again mainly retrospective studies, in which many of the patients had received multiple transfusion products sometimes from multiple sources. Future prospective studies that included only patients that had received a single transfusion would improve these trials.
Recommendations for practice
Good quality in-house training, and standardised protocols for transfusion practices within the work place are essential for ensuring gold standard care to patients. RVNs must be aware of the correct handling and storage of blood and its component products, as well as the risks involved to both donor and recipients with collection and administration of blood. Continuously improving knowledge and tasks such as performing more in-house blood typing and cross matching would help to support the blood banks and reduce wastage of products. Increased use of specific condition component products would also help reduce the demand on banked blood. Following current evidence-based recommendations for all aspects of transfusion medicine is essential, and reviewing blood product selection protocols in light of recent evidence regarding storage lesions would be advisable. Choosing to request or source leukoreduced PRBCs and performing autologous transfusions are areas where discussions and training may need to be implemented as deemed necessary.
For now, it should be recognised that xenotransfusions are high risk and should only ever be considered when there are no other available options and only with informed consent. Within the practice where this author is employed, component therapy is regularly carried out and typing and cross matching in-house has recently been made part of the transfusion protocol. An autologous transfusion has been performed in one patient with a traumatic abdominal bleed, with a successful outcome. Blood products are currently used on an oldest first policy although this is potentially undergoing review in the near future. There are currently studies ongoing in both human and veterinary transfusion medicine, and with human transfusion medicine rapidly advancing, the potential possibilities in veterinary transfusion medicine are huge (Figure 3).
Conclusion
Transfusion medicine is a dynamic area of veterinary medicine and is rapidly evolving. Currently the UK differs from the US in various transfusion protocols, especially in that the UK has a low reliance on banked feline blood compared with the US. However, the ethical and welfare implications surrounding feline blood collection may mean that this is not an area that is likely to change in the UK. Despite this feline blood transfusion medicine has great scope for further discoveries, and in comparison to human and canine transfusion medicine it is a relatively new area.
New evidence has been revealed attempting to optimise collection and storage protocols, potentially advancing blood banking collection capabilities (Walton et al, 2014). The identification of storage lesions and their implications, has highlighted issues with using aged PRBCs in cases of sepsis, but in other situations a storage lesion associated link with morbidity and mortality is yet to be confirmed, in either human or animal transfusion medicine. This means that evidence-based decisions cannot currently be made regarding length of storage in veterinary transfusion medicine. Reducing the length of time blood products are stored, along with changes to the protocol with which blood is selected for use, is a current topic of research both within the human and veterinary world. Controversy surrounds the issue, with some veterinary hospitals in America already choosing to alter the selection of product based on the individual case (Holowaychuk and Masulin, 2015).
Xenotransfusions are still drawing opposing conclusions, although this may well become an area of interest in the future; the current evidence does not appear to support xenotransfusions as there are stated risks to life associated with this practice, and only a few very isolated cases of successful xenotransfusions have been reported.
Leukoreduction is already being carried out in some hospitals and current evidence seemed in favour of the practice, although some authors have questioned the cost to benefit ratio (Pertinhez et al, 2016; McMichael et al, 2010), so further studies looking into this aspect have been recommended.
Current indications for autologous transfusions are positive for use alongside homologous transfusions in practice, although they seem to be performed very infrequently. However, autologous transfusion is likely to be a viable option, particularly when cell salvaging devices are used.