Clinical Review

Autologous transfusion

BMJ 2002; 324 doi: (Published 30 March 2002) Cite this as: BMJ 2002;324:772
  1. Elizabeth S Vanderlinde, chief residenta,
  2. Joanna M Heal, associate clinical professor of medicinec,
  3. Neil Blumberg, director, transfusion medicine and blood bank (Neil_Blumberg{at}
  1. a Department of Pathology and Laboratory Medicine, University of Rochester Medical Center, Box 608, 601 Elmwood Avenue, Rochester, NY 14642, USA
  2. b Transfusion Medicine Unit, University of Rochester Medical Center
  3. c Department of Medicine, University of Rochester Medical Center
  1. Correspondence to: N Blumberg

    Since the AIDS epidemic of the early 1980s the interest in alternatives to allogeneic transfusion has grown, particularly for elective surgery. One alternative that currently accounts for over 5% of the blood donated in the United States and some countries in Europe is autologous transfusion, obtained primarily by preoperative donation. Although autologous transfusion is used less widely in the United Kingdom than in the United States, guidelines on its use have recently been published in the United Kingdom.1 We describe the three main types of autologous transfusion and draw attention to the advantages and disadvantages of each technique (see table A on We also review the evidence from observational and controlled trials comparing autologous with allogeneic transfusion.

    Summary points

    Autologous transfusion reduces the need for allogeneic transfusion and is most widely used in elective surgery

    Autologous transfusion is one of several techniques used to reduce the need for allogeneic transfusion

    The three main techniques are predeposit transfusion, intraoperative haemodilution, and intraoperative and postoperative salvage

    Evidence from clinical trials shows that autologous transfusion is more cost effective than allogeneic transfusion and that clinical outcomes are improved


    We searched Index Medicus for publications on autologous transfusion. Many descriptive and methodological papers have described the efficacy of autologous transfusion in reducing allogeneic transfusion. Recent books and reviews address the technical and clinical aspects of the three types of autologous transfusion in detail.24 It is accepted that these techniques reduce the use of allogeneic blood, but the quality of the evidence varies, and possible drawbacks, such as temporary anaemia, have not yet been studied thoroughly. 5 6

    Autologous transfusion driven by concerns about the safety of blood

    Transfusion is a ubiquitous and potent treatment underlying much of modern medical practice. Once an unquestioned adjunct to patient care, allogeneic transfusion is currently being re-evaluated, and alternatives to conventional practice are being considered in response to numerous concerns about the safety of the procedure (fig 1). These include decreased cell mass and occasional transient hypotension. The most recent stimulus for the use of autologous transfusion is evidence that allogeneic transfusion may lead to an increased risk of postoperative bacterial infections and multiorgan failure.7 Another potential stimulus is increased demand for blood with a declining population of qualified, willing, and healthy donors. Three main techniques for autologous transfusion are used—predeposit transfusion, acute normovolaemic haemodilution, and interoperative and postoperative blood salvage.

    Predeposit autologous transfusion

    Predeposit autologous transfusion entails repeated preoperative phlebotomy (fig 2). Blood collection begins three to five weeks before elective surgery, depending on the number of units required, usually 2-4 units (about 1-2 litres). The last donation takes place at least 48-72 hours before surgery to allow for re-equilibration of the blood volume. On each occasion, about half a litre of the patient's own blood is taken and put into sterile plastic bags. Anticoagulation is maintained with citrated glucose solution, and the blood is stored until the time of surgery.


    Predeposit autologous transfusion virtually eliminates the risks of viral transmission and immunologically mediated haemolytic, febrile, or allergic reactions. These adverse effects range in frequency from 1 in 1 000 000 (HIV) to as high as 5% (febrile reactions). In addition, it may decrease the risk of postoperative infection and recurrence of cancer because immunomodulation as a result of transfusion is avoided.24 Immunomodulation refers to decreases in cellular immune function that have been documented after allogeneic, but not autologous, transfusions.7


    Up to half of the blood that is collected may be discarded because the amount drawn off needs to exceed the median routinely needed to avoid additional allogeneic transfusions. Leftover blood can rarely be used for other patients because most autologous donors do not meet the stringent health requirements for allogeneic blood donation. This wastage of blood and the costs of administering autologous programmes result in collection costs that are higher than those for allogeneic transfusion. Volume overload, bacterial contamination, and ABO haemolytic reactions to the transfusion resulting from administrative or clerical errors are further risks.

    Suitability of patients

    Predeposit autologous donation is practical only for elective surgery. Patients must be willing and able to travel to a donation centre before their operation, which can be inconvenient and stressful and may decrease their productivity at work. Because preoperative donation results in perioperative anaemia (which may not be completely resolved before surgery) blood volume, venous access, packed cell volume, and haemodynamic stability are important determinants of who is an appropriate candidate for the procedure. Children who weigh less than 30-40 kg are usually not suitable, but adult patients are deferred from donation only if they have severe haemodynamic problems, active systemic infections, or a history of serious reactions to donation (such as seizure). Patients with diarrhoeal illnesses in the days or weeks before donation should not donate as they may be at increased risk of bacterial contamination of their donated blood. Although autologous donors have a higher incidence of reactions such as fainting or dizziness than voluntary donors (presumably because they are inexperienced donors and not as young and fit), their reactions are seldom severe.

    Fig 1
    Fig 1

    The degree of anxiety and concern about the safety of blood transfusion over the centuries

    Intraoperative acute normovolaemic haemodilution

    Acute normovolaemic haemodilution (“haemodilution”) is a type of autologous donation that is performed preoperatively in the operating theatre or anaesthetic area. It is usually restricted to patients in whom substantial blood loss is predicted (>1 litre or 20% of blood volume). Whole blood (1.0-1.5 litres) is removed, and simultaneously intravascular volume is replaced with crystalloid or colloid, or both, to maintain blood volume. The anticoagulated blood is then reinfused in the operating theatre during or shortly after surgical blood loss has stopped. The blood sparing benefit of haemodilution is the result of the reduced red cell mass lost during surgical bleeding.


    Haemodilution provides the advantages of predeposit autologous donation and some additional benefits. It may be used before any type of surgical procedure, and systemic infection does not preclude its use. The patient is under anaesthesia during the procedure, which reduces stress, and the anaesthetist can ensure expert monitoring of blood circulation. Blood is stored at room temperature for a short time, so deterioration of clotting factors and cells is minimal. Additional advantages include a lower cost than for predeposit transfusion (because testing and cross matching are not usually required) and minimal wastage, as most or all blood is reinfused. Blood is maintained at the point of care, incurring little or no administrative expense, and the risk of ABO incompatibility because of administrative or clerical error is further minimised.


    The circulating red cell mass is lowered appreciably and acutely. If colloid is used for volume replacement the risk of allergic reactions or haemostatic abnormalities increases. Other disadvantages are the additional expense of, and inconvenience to, the anaesthetist who performs the procedure. The procedure may require additional training and experience on the anaesthetist's part. No large studies have investigated morbidity or mortality that may occur with acute anaemia, so the general belief that haemodilution is safe is largely anecdotal at this time.

    Suitability of patients

    Elective operations with typical blood losses of 1-2 litres are particularly suitable for haemodilution (for example, replacement of cardiac valves, revision of hip arthroplasty, or spinal reconstruction). The major limiting factor in choosing candidates for haemodilution is the patient's ability to tolerate a low volume of red blood cells. Patients with severe anaemia are usually poor candidates.

    Salvage autologous transfusion

    Intraoperative red blood cell salvage entails the collection and reinfusion of blood lost during surgery. Shed blood is aspirated from the operative field into a specially designed centrifuge. Citrate or heparin anticoagulant is added, and the contents are filtered to remove clots and debris. Centrifuging concentrates the salvaged red cells, and saline washing may be used. This concentrate is then reinfused. Devices used can vary from simple, inexpensive, sterile bottles filled with anticoagulant to expensive, sophisticated, high speed cell washing devices. Postoperative salvage refers to the process of recovering blood from wound drains and reinfusing the collected fluid with or without washing.


    Salvage is considered a safe and efficacious alternative to allogeneic red cell transfusion, but fewer data are available about clinical outcomes than for predeposit autologous donation or haemodilution.1 These techniques offer advantages similar to those of haemodilution but do not require infusions of crystalloid or colloid to preserve blood volume. Many litres of blood can be salvaged intraoperatively during extensive bleeding, far more than with other autologous techniques.

    Fig 2
    Fig 2

    Blood collection before preoperative autologous transfusion begins several weeks before surgery, and phlebotomy may be carried out several times



    Although the oxygen transport properties and survival of red cells are similar to that of allogeneic blood, salvaged blood is not haemostatically intact compared with blood derived by haemodilution. Coagulation in the wound leads to consumption of coagulation factors and platelets. Salvaged blood that is not washed contains raised concentrations of various tissue materials. Uncommon complications of extensive intraoperative salvage include disturbances to pH and electrolytes, systemic dissemination of non-sterile material, infectious agents or malignant cells, air or fluid embolism, and dilutional coagulopathy. A “salvaged blood syndrome” has been described, which entails multiorgan failure and consumption coagulopathy.8

    Clinical outcomes of randomised trials of autologous versus allogeneic blood transfusions

    View this table:

    Suitability of patients

    Intraoperative salvage is used extensively in cardiac surgery, trauma surgery, and liver transplantation. Contraindications to its use are bacterial infection or malignant cells in the operative field, and use of microfibrillar collagen or other foreign material at the operative site. Salvage can be one of the most expensive autologous techniques because costly capital equipment and disposables are used, and it is usually restricted to procedures resulting in substantial blood loss (>1-2 litres).

    Data on clinical outcomes

    Observational studies

    Many studies have examined whether patients who donate and receive autologous blood fare better clinically than those who receive allogeneic blood only (see table B on Of 16 studies, 10 found statistically significant reductions in unfavourable postoperative outcomes (primarily infections) in patients receiving autologous blood. Five found trends to improved outcomes that did not reach significance; one study found significantly better outcomes in patients receiving allogeneic transfusions.

    Randomised trials

    The number of randomised studies is small and the quality of the reporting variable. In four of the five studies to date, patients randomised to receive autologous rather than allogeneic transfusions had better clinical outcomes (table).914 Improved outcomes included a reduction in postoperative infections.1013 One study found a trend to reductions in recurrence of colorectal cancer with autologous transfusions.14 One third of the patients randomised to receive autologous blood also received allogeneic transfusions because their blood loss was too high to be treated with the autologous blood alone. The data from randomised trials thus confirm the results of observational studies: postoperative complications of surgery may be reduced by using autologous transfusions. These results currently provide one of the strongest arguments for the use of autologous transfusions.

    Additional educational resources

    National Audit Office, National Blood Service, Department of Health ( This website contains the proceedings of the conference on better blood transfusion, hosted by the United Kingdom's chief medical officers in October 2001, and links to many related sites.

    Scottish Intercollegiate Guidelines Network ( The introductory section of the recently prepared Scottish Intercollegiate Guidelines Network on perioperative blood transfusion for elective surgery.

    University of Pisa, Bloodless Medicine Research ( This website contains current research on alternatives to allogeneic transfusion and links to other academic and clinical centres specialising in bloodless medicine and surgery.

    Network for the Advancement of Transfusion Alternatives ( Nataonline is the home page of the network for the advancement of transfusion alternatives (NATA), a recently formed international academic and clinical society.

    NoBlood. NoBlood is of particular interest to patients, organised by proponents of bloodless medicine and surgery, especially relevant to addressing the needs of Jehovah's Witnesses. Links to hospitals with programmes, primarily in the United States.

    New Jersey Institute for the Advancement of Bloodless Medicine and Surgery ( This is the home page of Englewood Hospital and Medical Center of New Jersey, which has a longstanding commitment to alternatives to allogeneic transfusion,.

    Johns Hopkins University ( This is the home page of the recently established Eugene and Mary B Meyer Center for Advanced Transfusion Practices and Blood Research in Baltimore, Maryland.

    Cost effectiveness of autologous transfusion

    Some studies take into account increases in the risks of postoperative infection mediated by immunomodulation with allogeneic but not autologous transfusions. These studies have found autologous transfusion to be cost effective and perhaps even cost saving. 15 16 A study that did not address the possible immunomodulatory effects of transfusion found that autologous transfusion is not cost effective.17

    In the United States issues of cost effectiveness were secondary to the desire of patients to minimise risks associated with transfusion through autologous donation during the early years of the AIDS epidemic. The demand for autologous transfusion has decreased as patients have become less concerned over the safety of transfusion, primarily because of improved testing for viral agents such as HIV and hepatitis.

    Other concepts that spare blood transfusion

    Erythropoietin, the red cell production hormone, can reduce the need for transfusion in stable medical patients with cancer and premature newborn infants. It can also be used to reduce the need for allogeneic transfusion in surgical patients, with or without concomitant autologous collection. 18 19 Perioperative anaemia and blood loss can also be dealt with by reducing the amount of blood lost at surgery through improving mechanical haemostasis, using antifibrinolytics such as aprotinin, limiting phlebotomy to essential diagnostic tests, and using microsample laboratory techniques.20 Autologous transfusions also form part of a new concept of blood management called bloodless medicine and surgery.21 This includes the use of erythropoietin, surgical techniques that minimise blood loss, and drugs that inhibit fibrinolysis; greater degrees of anaemia are tolerated, and phlebotomy undertaken for diagnostic testing is minimal.20 Some of these methods are neither technically demanding nor expensive and may be adaptable to medical practice in less developed settings. The excellent results obtained in patients who are Jehovah's Witnesses, who refuse allogeneic transfusions, and the potential advantage of using fewer transfusions in patients in critical care support the promise of this concept for allogeneic transfusion.22


    • Competing interests NB has received lecture honorariums and consulting fees from Ortho Biotech.

    • Embedded ImageExtra tables appear on


    1. 1.
    2. 2.
    3. 3.
    4. 4.
    5. 5.
    6. 6.
    7. 7.
    8. 8.
    9. 9.
    10. 10.
    11. 11.
    12. 12.
    13. 13.
    14. 14.
    15. 15.
    16. 16.
    17. 17.
    18. 18.
    19. 19.
    20. 20.
    21. 21.
    22. 22.
    View Abstract

    Sign in

    Log in through your institution