Hospital Pharmacist Vol 7 No 5 p118-123
May 2000 Special Features

Blood and blood products

Allogeneic blood transfusion - the alternatives

By N. Watson, MSc, MRPharmS and C. Taylor, MRCP, MRCPath

The second part of our special feature explores the use of products and techniques which can be used to reduce the administration of allogeneic blood. The first article in the feature is on providing a safe and cost-effective blood transfusion service

As with all other medical and surgical interventions which are of potential benefit to patients, the use of allogeneic blood products carries with it an element of risk. Allogeneic blood means blood from a donor as opposed to autologous blood transfusion where the donor and recipient are one and the same. In November, 1996, a confidential reporting line known as SHOT (serious hazards of transfusion) was introduced in recognition of the risks involved in allogeneic blood transfusion. The group behind the SHOT initiative also aims to improve understanding of the associated adverse events, and data are collated on serious adverse events following transfusion of blood or labile blood components (fresh frozen plasma, platelets, cryoprecipitate and granulocytes).
The SHOT report of 1998/99 described 252 reports for the 12-month period.¹ A summary is shown in Table 1. The full SHOT report analyses these incidents and makes recommendations designed to prevent some of them, particularly the frequently reported problem of incorrect blood component transfusion. This annual report serves to highlight the risks associated with the transfusion of blood components and strengthens the argument for the use of alternative approaches where possible.

Reducing the risks

Although this has been considered in the preceding article, the risks of blood transfusion are being considered again in the context of identifying suitable alternatives.

Donor selection The 1998/99 SHOT report shows that although the risk associated with blood transfusion is now less than it has been at any time in the history of blood transfusion, some real risks still remain, particularly of immunological complications and, to a lesser extent, transfusion transmitted infections (TTIs). A number of steps in the collection and processing of blood for transfusion are designed to reduce these risks as much as possible.
Donor selection is probably the single most important way in which the risks from blood transfusion are reduced. Volunteer blood donors are invited to fill in a questionnaire about life style and health issues and asked to exclude themselves from blood donation if they fall into various risk categories. A personal interview is carried out with all new blood donors and those who have not donated for two years in order to elucidate "at risk" behaviour and exclude these donors. It is the introduction of this form of donor selection which has caused the biggest decrease in TTIs. All blood donations are tested for infectious agents. The mandatory tests are human immunodeficiency virus (HIV) 1 and 2, hepatitis B, hepatitis C and syphilis. A proportion of donations are also tested for malaria, trypanosomiasis and cytomegalovirus (CMV).²
In spite of this testing, there is still transmission (albeit at a very low rate) of these infections because of the "window" period. This means that if a donor has recently been exposed to hepatitis B, C or HIV, they may have contracted the disease but the test used may not be able to detect antibodies. Increasingly sensitive tests shorten the window period, but there are 14 to 21 days during which the blood will test antibody negative, even in the presence of a virus. It is for this reason that donor selection procedures are so critical in reducing the risk.

Transfusion need As a result of these residual risks of transfusion, it is very important that blood components are used appropriately, and alternatives must always be considered. By employing a haematologist with an interest in transfusion in every hospital, as well as a nurse practitioner or haemovigilance officer, and the active involvement of the hospital pharmacy, much can be done to limit the unnecessary use of blood components.³ Doctors and nurses can be educated in the appropriate use of blood products, and local guidelines can be written, based on the British Committee for Standards in Haematology (BCSH) guidelines, incorporating agreed trigger levels of haemoglobin or platelets, which should be attained before transfusion is implemented.
A consensus was reached recently on platelet transfusion⁴ which resulted in most centres reducing the threshold platelet count for prophylactic platelet transfusion from 20 x 10⁹ per litre to 10 x 10⁹ per litre. This has resulted in a great saving in the number of platelets transfused, which translates not just into cost saving but also reduced donor exposure which lowers the risk from blood transfusion. It is harder to come to an agreement on the trigger level for haemoglobin. However, recent data suggest that a haemoglobin level of 8g per decilitre is not unreasonable in most individuals, including the critically ill, and it is the exceptional patient in whom a higher haemoglobin level needs to be maintained.⁵ Accepting a lower post-operative haemoglobin level in surgical patients achieves a significant reduction in blood transfused and reduced donor exposure. This can be coupled with the proper use of haematinics (any therapeutic agent giving rise to an increase in haemoglobin content in the blood), colloids and oxygen therapy post-operatively.
Fresh frozen plasma (FFP) is the blood component that is probably abused the most in the hospital setting. It should only be used to treat coagulopathies on the basis of clotting screen results and not as a volume expander or formula replacement. BCSH guidelines should be followed in each institution and, when FFP is used, it should be given in adequate doses to correct coagulopathy.⁶
One other way of reducing donor exposure is to use products from one donor for the same recipient. This can be done in neonatal transfusions where a single donor unit is divided into eight paediatric packs which can then be given in series to a baby on the special care baby unit.⁷ A similar approach may be possible with platelet transfusion, especially now that many platelet donors are donating triple doses on an apheresis machine. The efficacy and feasibility of this approach is currently being investigated.

Viral and bacterial inactivation Processes such as solvent and detergent treatment, methylene blue addition and psoralen treatment can be implemented, either to treat pooled products, such as FFP, or individual components.⁸
Solvent and detergent treated products are widely available, including FFP, anti-D, and intravenous immunoglobulin. This treatment is effective in inactivating enveloped viruses, although it does not inactivate hepatitis A, parvovirus or bacteria. Methylene blue (with white light) treatment is in use for FFP, although it has some drawbacks, such as dyeing the patient blue. New methylene blue compounds are in development for treatment of red cells. Psoralen and ultra violet A treatment is still in the development phase but has the advantage that it is applicable to FFP, platelets and red cells, It also inactivates all viruses and bacteria.⁹

Alternatives to RBCs

It is clear that avoiding blood loss in the first instance is an effective method of limiting the use of allogeneic blood components such as red blood cells (RBCs). But how can this be achieved? There are a number of techniques that can be applied to prevent blood loss, ranging from pharmacological intervention to altered surgical techniques.

Pharmacological intervention Tranexamic acid, a lysine analogue which inhibits plasminogen activation and fibrinolysis, can be used to prevent excessive blood loss in a number of clinical settings. These include prophylaxis and treatment of bleeding associated with prostatectomy and menorrhagia, and for the management of dental extraction in haemophiliacs or Jehovah's Witnesses.
Aprotinin, another antifibrinolytic agent, inhibits proteolytic enzymes and thus reduces blood loss and transfusion requirements. It has a wide range of indications that are characterised by the need to reduce blood loss during surgery, the most notable being cardiac surgery with cardiopulmonary bypass.
A recently published meta-analysis¹⁰ examined the pharmacological strategies used to decrease excessive blood loss during cardiac surgery. The results demonstrated an almost two-fold decrease in mortality when using aprotinin, and a significant decrease in the proportion of patients receiving allogeneic blood transfusions when treated with either aprotinin or tranexamic acid.
Fibrin sealant or glue has been studied in a wide variety of surgical operations. It is prepared typically from a mixture of fibrinogen, factor XIII, thrombin, aprotinin and calcium chloride. The topical application of fibrin glue has been effective in the reduction of intra-operative bleeding in a number of settings, including managing haemorrhage after dental extraction in patients on anticoagulant therapy or with haemorrhagic disorders.
The use of haemostatic agents can be useful in preventing excessive blood loss, and are therefore a useful means of reducing transfusion requirements. The added cost of using such agents may well be offset by the savings in transfusion costs. In addition, the risks associated with blood transfusion are avoided.

Surgical and anaesthetic techniques Surgical and anaesthetic techniques are frequently employed to reduce blood loss. Diathermy is commonplace in theatres to keep the surgical field clear and to reduce blood loss.
During surgical procedures, it is now common practice to use heated water blankets to maintain body temperature. The maintenance of normothermia, both intraoperatively and postoperatively, contributes to improved haemostasis. The use of insulated caps on patients' heads postoperatively also helps to maintain body temperature. In cardiac surgery, where hypothermia is used while the patient is on bypass, there is no risk of bleeding since circulation is in a closed circuit. However, when the patient comes off bypass, normothermia is achieved rapidly in order to reduce the need for blood transfusion.

Haematinics and erythropoietin The use of haematinics and erythropoietin can reduce the need for blood transfusions.
A frequent finding in audits of red blood cell usage in hospitals is that patients are transfused because of a low haemoglobin resulting from haematinic deficiency.¹¹ This includes pre-operative surgical patients and medical patients, particularly the elderly. There may be constraints about dates of surgery and the haemoglobin having to be corrected within a short space of time, but frequently there is adequate opportunity for the investigation and correction of anaemia.
Oral iron therapy and vitamin C will raise the haemoglobin in an iron-deficient patient by 2g per decilitre in three weeks and greatly reduces the need for transfusion. A pre-clerking surgical clinic a month before elective surgery allows this to take place. Less frequently, patients who are folate and vitamin B12 deficient with macrocytic anaemias are transfused. This is rarely, if ever, necessary and can cause electrolyte imbalance and cardiac problems. Even haemoglobin levels below 5g per decilitre due to haematinic deficiency can be safely treated by replacement therapy with iron, folic acid or vitamin B12, rather than by transfusion. Parenteral iron therapy may be given if gastric toxicity or malabsorption are a problem.¹² Post-operatively, a patient who has bled down to a haemoglobin level of 8g per decilitre may be safely managed by treatment with iron and folic acid rather than by exposure to the risks of transfusion.
Interest in the use of erythropoietin in non-uraemic patients has been stimulated recently by the increased cost of RBCs, due to universal leucodepletion being instituted in the UK, and public perception of the high risk of blood components. Erythropoietin has been shown in numerous studies to be effective in increasing the pre-operative haematocrit, thereby reducing the need for allogeneic transfusion, particularly in orthopaedic surgery. It has also been compared and found to be at least as good as (or possibly better than) pre-operative autologous donation. It also has a well-established safety profile.¹³
A major predictor of transfusion risk is pre-operative haematocrit. Patients with a haemoglobin level above 14g per decilitre have a low risk of transfusion (less than 10 per cent) in hip and knee surgery, while those with a haemoglobin level below 10g per decilitre have a higher risk (greater than 60 per cent) and pre-donation is impossible. Patients with haemoglobin levels between 10 to 13g per decilitre may benefit the most from peri-operative erythropoietin, in that allogeneic transfusion may be avoided all together.
Pre-donation may be impossible in patients who are unable to participate due to anaemia, co-morbidity or logistical problems. Erythropoietin may be very effective in these patients. It has also found use in the management of Jehovah's Witnesses, many of whom will accept recombinant human erythropoietin as an alternative to blood transfusion. In paediatric patients, for instance those requiring spinal surgery, erythropoietin may be more acceptable than blood transfusion.
A decision to start erythropoietin therapy may be made at the pre-admission surgical clinic four weeks prior to the elective surgical procedure. A dose of 600 IU per kg may be given three times weekly with supplementary iron and folate, prior to the procedure. There is no indication that there is any increase in thromboembolic problems when erythropoietin is used in this way. Newer erythropoietin products are being developed which will require less frequent administration.
There is increasing interest in using recombinant human erythropoietin in patients having surgery for cancers. The immunomodulatory effect of blood transfusion has been demonstrated to cause a slight but significant increase in infectious complications and relapse rates of solid tumours. Hence, attempts are being made to reduce the need for allogeneic blood in these patients. A strong rationale for treatment with erythropoietin in anaemic cancer patients is the blunted endogenous erythropoietin production. In both orthopaedic and cancer surgery, the use of erythropoietin may be combined with autologous blood donation pre-operatively.
In the majority of studies, erythropoietin has still proved to be a more expensive option than allogeneic transfusion if the straightforward costs of erythropoietin and units of RBCs are compared. In health economic terms, however, the costs of managing the complications of transfusion, the additional drugs and therapies that may be necessary, long-term follow-up and the potential for litigation, may make transfusion more costly than the safer erythropoietin therapy. However, detailed health economic evaluations of the real costs of transfusion are not readily available.

Autologous transfusion This refers to the collection and re-transfusion into a patient of their own blood. This is done in three ways.¹⁴
The first way is by pre-operative autologous deposit (PAD). During the four weeks preceding planned elective surgery, blood can be venesected from patients who have normal haemoglobin levels. Blood has a shelf life of 35 days and a maximum of four units can usually be taken at weekly intervals prior to surgery. This blood can then be used instead of allogeneic blood during the procedure. PAD should not be undertaken unless there is a definite surgical date, which is unlikely to be cancelled, and a definite need for transfused blood. A pre-operative surgical clerking clinic is a great help when planning PAD. Supplementary iron and folate can be given during the procedure. Although other risks are reduced, the likelihood of bacterial contamination and of receiving the wrong blood is the same as for allogeneic blood.
Erythropoietin has been used as an adjunct to PAD and may increase the amount of blood that can safely be collected. However, PAD is not a cost-effective method of avoiding the use of allogeneic blood, as a large amount of time is involved in counselling the patient, ensuring they are fit enough and in collecting the blood. The blood is tested for infectious agents in the same way as allogeneic blood. The wastage rate is very high, with up to 55 per cent of autologous units collected in the USA being wasted.
A second method of autologous infusion is acute normovolaemic haemodilution (ANH) which is carried out in the immediate pre-operative period. Blood is withdrawn from the patient during the induction of anaesthesia and concurrently replaced with colloid or crystalloid, resulting in red blood cell mass dilution. This results in a circulating blood of low haematocrit and therefore the number of red blood cells lost due to surgical bleeding is less. The blood with a high haematocrit is replaced at the end of the procedure, when surgical haemostasis has been achieved. Generally, this is a safe and cost-effective alternative to allogeneic transfusion, although there is still some debate as to its efficacy and benefit.
The advantages of this method of autologous transfusion are:

• It can be performed by most anaesthetic specialists
• The risk of blood-borne disease is eliminated
• There is a full complement of 2,3 diphosphoglycerate (DPG) in the cells as there is no storage period

However, there are practical limitations, such as the patient's medical condition and co-existing disease (pulmonary, cardiac and renal), the amount of blood withdrawn, the limits of tissue oxygenation, effects on the coagulation system and the length of time the ANH blood can be stored.
Clinical studies, for instance in urological surgery, have demonstrated that PAD and ANH have similar efficacy and outcomes. There are early data available to show that the efficacy of ANH may be improved by the administration of pre-operative erythropoietin. However, it must be remembered that neither of these procedures allows avoidance of allogeneic blood transfusion if more than three or four units of blood are likely to be needed.
The third method of autologous transfusion is intra-operative cell salvage where shed blood from the operative field is collected and re-transfused. A number of different devices are available, all of which pass the collected blood through a filter. Some of them then re-transfuse the blood without further processing, while others wash the blood to remove haemolysed cells, free haemoglobin and other impurities.
Numerous randomised and non-randomised studies have reported a clinically important decrease in the frequency of allogeneic transfusion. This system is most commonly used in areas where there is a higher requirement for allogeneic transfusion, such as open heart surgery, abdominal aortic aneurysm repair and liver transplantation. The main side effects include air embolism, coagulation abnormalities and disseminated intravascular coagulation.
Coagulopathies, with disseminated intravascular coagulation or "salvaged blood syndrome", occur after transfusion of salvaged blood and may result in large intra-operative blood losses and increased post-operative bleeding, which in turn may be counteracted by the large amount of autologous blood that can be salvaged and transfused. Some studies have reported an increase in post-operative bleeding, particularly in those receiving unwashed blood. Salvage should not be used during surgery for infected areas such as abscesses or bowel perforations.
There is also concern about its use in cancer surgery. However, there is no doubt that this is a major method of reducing the quantity of allogeneic blood to which patients having major surgical procedures are exposed. In addition, continuous circuit devices are acceptable for use in many Jehovah's Witnesses' patients. Blood collected from thoracic and abdominal drains post-operatively may also be recycled using this system.

Plasma expanders

Haemodynamic equilibrium and tissue oxygenation is dependent on the maintenance of blood volume. Nevertheless, blood volume does not need to be maintained acutely with allogeneic blood.
There are two forms of plasma expanders: crystalloid and colloid. Crystalloids include basic intravenous infusions such as normal saline, 5 per cent glucose solution and various combinations thereof. Colloids include albumin, dextrans, starches and gelatins.
There is still some controversy surrounding the choice of plasma expanders, and complex decisions have to be made about the use of albumin and the relative merits of gelatins, dextrans and starches. However, it appears that crystalloids should remain as first choice intravenous solutions for hypovolaemia, with colloids reserved for patients who cannot tolerate the larger volumes required during intravascular resuscitation.¹⁵

Other oxygen carriers

Unfortunately, there is as yet no oxygen-carrying solution for use in patients who require a blood transfusion in order to maintain oxygenation. However, if it is accepted that the use of allogeneic blood always carries a risk, the need for such a product becomes obvious. Four technologies are being developed to tackle the problem.

Modified red blood cell antigen preparations Given the issue of cross-matching red blood cells using the ABO system, it would be useful to be able to administer a universal red blood cell product to all patients. Two methods to deal with red blood cell antigens are in development: enzymatic removal of antigens¹⁶ and antigenic modulation using polyethylene glycol.¹⁷

Cell-free haemoglobin preparations Cell-free haemoglobin has been studied for a considerable period of time and three key problems have been identified in its use. First, free haemoglobin, with or without cellular debris (stroma) is nephrotoxic. Secondly, free haemoglobin also has a greater binding affinity for oxygen, which means that oxygen is not released into tissues as readily as it is from haemoglobin inside erythrocytes. As a direct result of these difficulties, polymerised agents have been developed to overcome the renal toxicity and oxygen affinity. Lastly, there is the issue of the source of haemoglobin. Many of the preparations under investigation utilise human or bovine haemoglobin, which creates concerns over viral transmission and immunological incompatibility. The development of recombinant DNA-produced haemoglobin is a major advance in this respect.¹⁸
Despite the development of these preparations, phase III studies are still going on in the USA, and we are some way from seeing the first of these licensed in the UK.

Liposome-enclosed haemoglobin Liposome-encapsulated haemoglobin strives to overcome some of the problems with cell-free haemoglobin preparations. Pre-clinical studies are continuing with this novel technology.¹⁹

Perfluorocarbon emulsions PFC emulsions offer a completely different approach to the problem of oxygen carriers. PFCs are highly soluble in non-polar gases such as the respiratory gases. When formed into emulsions (in their pure form they are oils), they may be administered to patients. Following initial difficulties with first generation PFCs, such as liver toxicity and limited oxygen-carrying capacity, newer, second-generation agents are well tolerated and have greater oxygen-carrying capacity.¹⁸ Interestingly, the potential use for PFCs extends beyond being an alternative to allogeneic blood. PFCs may be able to support oxygenation to areas of poor blood flow, such as solid tumours or even ischaemic cardiac tissue. PFCs also carry nitrogen, and their ability to help with air embolism may prove beneficial.
These differing forms of oxygen carriers, each with its own potential advantages and disadvantages, hold hope for future alternatives to allogeneic blood. At present, PFC emulsions appear to be the most promising.

Alternatives to WBCs

Granulocyte transfusion has been used to treat neutropenic patients with severe life-threatening sepsis. Several prospective randomised controlled clinical trials have shown an increase in survival of patients with Gram-negative septicaemia and a neutropenia of less than 0.1 x 10⁹ per litre treated with granulocyte infusion.²⁰
The use of therapeutic granulocyte transfusions has diminished over the past decade, despite reports of its efficacy. This can be explained by the availability of more effective antimicrobial drugs to prevent and treat infections and by the development of recombinant haemopoietic cytokines. Peripheral blood progenitor cell transfusions also hasten stem cell engraftment and shorten the duration of severe neutropenia following myeloablation.
Granulocyte transfusions may yet have a role, if a higher dose of white blood cells (WBCs) is given. Data from the USA suggest that from 5 to 7 x 10¹⁰ WBCs are needed per adult dose and this number can only be collected from donors pre-treated with granulocyte-colony stimulating factor (G-CSF), who then undergo leucophoresis. The UK practice of treatment with buffy coats (the retrieved white cell layer after centrifugation) from donated units of whole blood results in a daily dose of at most 1 x 10¹⁰ WBCs. Such doses have rarely been associated with improved survival in randomised studies. Volunteer donors in the UK cannot be treated with G-CSF for leucocyte collection purposes and therefore reliance is placed on the alternatives to leucocyte transfusion.
Colony stimulating factors, which include G-CSF and GM-CSF (granulocyte macrophage-colony stimulating factor) have found their place in therapy, primarily in supporting oncology and haematological oncology treatments by stimulating the production of neutrophils, rather than directly as alternatives to blood transfusion.

Alternatives to platelets

The platelet is the blood component that most commonly contains bacterial contamination and it can result in life-threatening infection or death from endotoxic shock syndrome. The majority of platelet concentrates are used as prophylaxis in patients receiving chemotherapy and haemopoietic stem cell transplantation and the risk from transfusion of platelets has been reduced by lowering the platelet trigger level to 10 x 10⁹ per litre. Methods of viral and bacterial inactivation of platelet concentrates have been discussed above. Psoralen plus ultra violet (UV) light treatment is the most promising as it inactivates both bacteria and viruses. This is currently in phase III trials and may be licensed in the next two years.⁹
It was hoped that thrombopoietin might be used for the treatment of thrombocytopenia, in the same way that erythropoietin is offering a new therapy for chronic anaemia. Studies in normal donors given multiple doses of the factor known as pegylated recombinant human megakaryocyte growth and development factor (PEG-rHuMGDF) demonstrated development of antibodies to this protein which resulted in thrombocytopenia and bleeding, and this has curtailed all work on this project. Patient-based studies with the full-length protein, thrombopoietin, are continuing, but these are unlikely to be extended to donors and will have no impact on platelet supply.²¹ Interleukin-11 is also under assessment for its potential effect in increasing platelet counts.²²
A number of platelet substitutes are in development. Fixed, freeze-dried platelets have been developed and shown to shorten the bleeding time in thrombocytopenic rabbits. These products require harvesting from volunteer donors and provision of large quantities would therefore be limited. Infusible platelet membrane (IPM) products have been explored, which can be made from out-dated platelets and this has also shortened bleeding time in the rabbit model. Some synthetic and semi-synthetic platelet substitutes have been developed based on fibrinogen-coated albumin microspheres or red blood cells coated with fibrinogen. Human phase I trials are just beginning in these products.²³

Mr Watson is pharmacy manager at Harefield hospital, Royal Brompton and Harefield NHS trust, London, and Dr Taylor is a consultant in haematology and transfusion medicine at the Royal Free hospital NHS trust, London, and the National Blood Service

References