Practical Transfusion Medicine 4th Ed.

7. Haemolytic transfusion reactions

Edwin J. Massey1, Robertson D. Davenport2 & Richard M. Kaufman3

1Department of Pathology, NHS Blood and Transplant, Bristol, UK

2University of Michigan Health System, Ann Arbor, Michigan, USA

3Brigham and Womens Hospital, Boston, Massachusetts, USA

Definition of a haemolytic transfusion reaction

A haemolytic transfusion reaction (HTR) is the occurrence of lysis or accelerated clearance of red cells in a recipient of a blood transfusion. With few exceptions, these reactions are caused by immunological incompatibility between the blood donor and the recipient [1].

HTRs are usually classified with respect to the time of their occurrence following the transfusion but may also be classified on the pathophysiological basis of the site of red cell destruction, intravascular or extravascular. The classification used by the Serious Hazards of Transfusion (SHOT) haemovigilance scheme in the UK is as follows [2]:

·        Acute HTRs (AHTRs) occur during or within 24 hours of the transfusion.

·        Delayed HTRs (DHTRs) occur more than 24 hours after a transfusion, typically 5–7 days later.

In general, with some exceptions, intravascular haemolysis is seen in AHTRs and extravascular haemolysis in DHTRs. During intravascular haemolysis, the destroyed red cells release free haemoglobin and other red cell contents directly into the intravascular space. These reactions are characterized by gross haemoglobinaemia and haemoglobinuria, which can potentially precipitate renal and other organ failure.

During extravascular haemolysis, red cells are removed from circulation primarily by the spleen. In these reactions the only feature may be a fall in haemoglobin (Hb), but clinically significant DHTRs can occur, which may contribute to morbidity and even mortality in patients who are otherwise compromised by single or multiple organ failure prior to the reaction.

Pathophysiology of HTRs

There are three phases involved (Figure 7.1):

·        antibody binding to red cell antigens, which may involve complement activation;

·        opsonized red cells interacting with and activating phagocytes; and

·        production of inflammatory mediators.

Fig 7.1 Pathophysiology of the haemolytic transfusion reaction (HTR). ADCC, antibody-dependent cell-mediated cytotoxicity; MAC, membrane attack complex; NK, natural killer.


Antigen–antibody interactions

Where an immunological incompatibility is responsible, the course of the reaction depends upon:

·        the class and the subclass (in the case of IgG) of the antibody;

·        the blood group specificity of the antibody;

·        the thermal range of the antibody;

·        the number, density and spatial arrangement of the red cell antigen sites;

·        the ability of the antibody to activate complement;

·        the concentration of antibody in the plasma; and

·        the amount of red cells transfused.

The characteristics of the antibody and antigen

The characteristics of the antibody (such as immunoglobulin class, specificity and thermal range) and those of the antigen sites against which antibody activity is directed (such as site density and spatial arrangement) are interrelated. Antibodies of a certain specificity, from different individuals, are often found only within a particular immunoglobulin class and have similar thermal characteristics; in addition, red cells of a certain blood group phenotype, from different individuals, tend to be relatively homogeneous regarding the attributes of the relevant antigen. It is for this reason that knowledge of the specificity of an antibody can be highly informative in predicting its clinical significance [3]. Three examples illustrate this concept.

·        Anti-A, anti-B and anti-A,B antibodies are regularly present in moderate to high titre in the plasma of group O persons. These ‘naturally occurring’ antibodies are often both IgM and IgG, having a broad thermal range up to 37°C, and are often strongly complement-binding. The A and B antigens are often present in large numbers (e.g. up to 1.2 × 106 A1 antigen sites per cell) and are strongly immunogenic (provoking an immune response in an individual lacking the antigen). If an individual who has anti-A, anti-B or anti-A,B in their plasma is transfused with donor red cells that express the cognate antigen (i.e. A and/or B), an AHTR is highly likely to occur, which may be fatal. The infusion of group O donor plasma (200–300 ml in an adult pack of platelets or plasma) can similarly cause haemolysis of the recipient's red cells if they are A, AB or B; this will be discussed later in this chapter.

·        Anti-Jka antibodies may be produced following immunization of a Jk(a–) person via pregnancy or transfusion. They are usually IgG (but may also have an IgM component), are active at 37°C and may be complement-binding. In Jk(a+b–) persons, there are about 1.4 × 104 Jka antigen sites per cell. Jka antigens are not particularly immunogenic; however, the antibody is sometimes difficult to detect in pretransfusion testing (because of the low titre of antibody); consequently, Jk(a+) blood may be inadvertently transfused to patients with pre-existing anti-JkaThese antibodies are frequently implicated in DHTRs.

·        Anti-Lua antibodies may be produced following the immunization of an Lu(a–) person, or may be ‘naturally occurring’. They are usually IgM (but often have IgA and IgG components), are only sometimes reactive at 37°C and are not usually complement-binding. The Lua antigens show variable distribution on the red cells of an individual and are poorly immunogenic. The antibody may not be detected in pretransfusion testing, because screening cells usually do not possess the Lua antigen and because antibody levels fall after immunization. Anti-Lua antibodies have not been implicated in AHTRs and have only rarely been implicated in DHTRs, which are usually mild.

Complement activation

Antibody-mediated intravascular haemolysis is caused by sequential binding of complement components (C1–C9) on the red cell membrane. IgM alloantibodies are more efficient activators of C1 than IgG antibodies, since the latter must be sufficiently close together on the red cell surface to be bridged by C1q in order to activate complement. Activation to the C5 stage leads to release of C5a into the plasma and assembly of the remaining components of the membrane attack complex (MAC) on the red cell surface, leading to lysis.

Extravascular haemolysis is caused by non-complement-binding IgG antibodies or those that bind sublytic amounts of complement. IgG subclasses differ in their ability to bind complement, with the following order of reactivity: IgG3>IgG1>IgG2>IgG4.

Activation of C3 leads to the release of C3a into the plasma and to C3b and iC3b deposition on red cells, promoting binding of the red cell to two complement receptors, CR1 (CD35) and CR3 (CD11b), which are both expressed on macrophages and monocytes. Hence, C3b and iC3b augment macrophage-mediated clearance of IgG-coated cells, and antibodies binding sublytic amounts of complement (e.g. Duffy and Kidd antibodies) often cause more rapid red cell clearance and more marked symptoms than non-complement-binding antibodies (e.g. Rh antibodies).

C3a and C5a are anaphylatoxins with potent proinflammatory effects including oxygen radical production, granule enzyme release from mast cells and granulocytes, nitric oxide production and cytokine production [4].

Fc receptor interactions

IgG alloantibodies bound to red cell antigens interact with phagocytes through Fc receptors. The affinity of Fc receptors for IgG subclasses varies, with most efficient binding to IgG1 and IgG3. After attachment to phagocytes, the red cells are either engulfed or lysed through antibody-dependent cell-mediated cytotoxicity (ADCC).


Cytokines are generated during an HTR as a consequence of both anaphylotoxin generation (C3a, C5a) and phagocyte Fc receptor interaction with red-cell-bound IgG. Some biological actions of cytokines implicated in HTRs are given in Table 7.1.

Table 7.1 Cytokines implicated in haemolytic transfusion reactions.


Biological activity

Proinflammatory cytokines





Hypotension, shock, death


Mobilization of leucocytes from marrow


Activation of T and B cells


Induction of cytokines (IL-1, IL-6, CXCL-8, TNF-α, CCL-2)


Induction of adhesion molecules




Acute-phase protein response


B-cell antibody production


T-cell activation




Chemotaxis of neutrophils


Chemotaxis of lymphocytes


Neutrophil activation


Basophil histamine release


Chemotaxis of monocytes


Induction of respiratory burst


Induction of adhesion molecules


Induction of IL-1

Anti-inflammatory cytokines



Competitive inhibition of IL-1 type I and II receptors

ABO incompatibility stimulates the release of high levels of tumour necrosis factor (TNF)-α into the plasma, within 2 hours, followed by CXCL-8 (interleukin (IL)-8) and CCL-2 (monocyte chemotactic protein (MCP)-1). In IgG-mediated haemolysis, TNF-α is produced at a lower level together with IL-1β and IL-6. CXCL-8 production follows a similar time course to that in ABO incompatibility.

IgG-mediated haemolysis, as opposed to ABO incompatibility, also results in the production of the IL-1 receptor antagonist, IL-1ra. The relative balance of IL-1 and IL-1ra may also, at least in part, account for some of the clinical differences between intravascular and extravascular haemolysis [5].

Antibody specificities associated with HTRs

These are given, together with the site of red cell destruction, in Table 7.2. A helpful review paper on the clinical significance of red cell antibodies has been written by Daniels et al. [3].

Table 7.2 Antibody specificities associated with haemolytic transfusion reactions.

Blood group







A, B, H







K, k, Kpa, Kpb, Jsa, Jsb


Jka (Jkb, Jk3)

Jka, Jkb, Jk3



Fya, Fyb, Fy3



M, S, s, U















Coa, Cob



Doa, Dob

Acute HTRs

Aetiology and incidence

These reactions arise as a result of existing antibodies, in either the recipient or donor plasma, which are directed against red cell antigens of the other party. In the developed world, transfusion reactions resulting from incompatibility are more common as a cause of morbidity and mortality than transfusion-transmitted infection. This may not be the perception of patients, the public, politicians and even clinical staff. Incompatible transfusion can occur for the following reasons:

1. Clerical error. This can occur at the point of taking and labelling the sample, laboratory compatibility testing and blood allocation, collection of the blood component from the refrigerator, freezer or agitator and bedside checking at administration.

2. Undetected antibody. Kidd (Jk) antibodies are a typical example of antibodies that may be missed by sensitive testing systems.

3. Intentional provision of blood components as the best available, lowest risk choice when the ‘perfect’ blood component is not available (e.g. ORhD negative cde/cde in an emergency to a patient who subsequently proves to have anti-c).

The majority of AHTRs have historically been due to the transfusion of ABO-incompatible red cells, but can also be due to the administration of donor plasma containing high titres of ABO haemolysins when platelets or less commonly fresh frozen plasma of a different ABO blood group is transfused (classically group O donor plasma into a group A recipient). ABO-incompatible red cell transfusions are the result of the ‘wrong’ blood being given to the ‘wrong’ patient because of clerical or administrative errors, occurring at any stage during the transfusion process. ABO-incompatible platelet administration is unlikely to cause a reaction and such transfusions are ‘intentional’ to utilize the short shelf-life platelet stock in an efficient manner (see below). More recently in the UK, serious errors in the transfusion process have become less frequent and morbidity and mortality following HTR is now more commonly due to antibodies other than ABO [2].

The Serious Hazards of Transfusion (SHOT) confidential reporting scheme has shown that in cases where the patient was transfused with a blood component or plasma product that did not meet the appropriate requirements or that was intended for another patient, the sites of primary error were clinical areas in 65% of cases, hospital laboratories in 34% of cases and blood establishments in 1% of cases. The reports have also highlighted that multiple errors contribute to incorrect blood component transfusion (IBCT). Examples of reported errors from several series are given in Table 7.3. Estimates of ABO-incompatible transfusions vary and may be underestimates, since some may be unrecognized or not reported, but two surveys have found a frequency of 1 in approximately 30 000 transfusions [6,7].

Table 7.3 Errors resulting in ‘wrong blood’ incidents.

Prescription, sampling and request

  Failure to identify correct recipient at sampling

  Correct patient identity at sampling but incorrectly labelled sample

  Selection of incompatible products in an emergency

Transfusion laboratory

  Took a correctly identified sample and aliquoted it into an improperly labelled test tube for testing

  Took a wrongly identified sample through testing

  Tested the correct sample but misinterpreted the results

  Tested the correct sample but recorded the results on the wrong record

  Correctly tested the sample but labelled the wrong unit of blood as compatible for the patient

  Incorrect serological reasoning, e.g. O-positive FFP to non-O-positive recipient

Collection and delivery of the wrong component to the ward

  Failure to check recipient identity with unit identity

Bedside administration error

  Recipient identity checked through case notes or prescription chart and not wristband

  Wristband absent or incorrect

Not all ABO-incompatible transfusions cause morbidity and mortality; mortality is dependent on the amount of incompatible red cells transfused and is reported to be 25% in recipients receiving 1–2 units of blood and reaches 44% with more than 2 units. However, as little as 30 mL group A cells given to a group O recipient can be fatal. Less frequently, Kell, Kidd and Duffy antibodies can be responsible and the acute reaction is due to a failure to detect, or take account of, the red cell alloantibody in either the antibody screen or cross-match.

Errors are a major cause of morbidity due to HTRs. In the UK the SHOT voluntary reporting scheme HTR has accounted for 501/8110 (6%) and IBCT 2837/8110 (35%) of errors reported. Nearly all deaths as a result of IBCT are due to ABO-incompatible transfusions and there have been 27 deaths in which IBCT was causal or contributory, between 1996 and 2010. Over the same period there have been 118 cases of major morbidity due to IBCT and 50 others attributable to acute and delayed HTR [2]. Similar findings have been noted in other countries; details of the incompatibilities resulting in deaths reported to the Food and Drug Administration between 1976 and 1985 are provided in Table 7.4 [8].

Table 7.4 Fatal acute haemolytic transfusion reactions reported to the FDA between 1976 and 1985.


Number of deaths

O recipient and A red cells


O recipient and B/AB red cells


B recipient and A/AB red cells


A recipient and B red cells


O plasma to A/AB recipient


B plasma to AB recipient


Total ABO incompatibilities












Total non-ABO incompatibilities


On a positive note, the SHOT voluntary reporting scheme has demonstrated an increase in overall reporting of errors and reactions but simultaneously the number of reported errors resulting in preventable major morbidity or mortality has fallen. HTR, in particular AHTR due to ABO incompatible transfusion, has decreased. There was also a 29% reduction of IBCT cases in 2010 [2]. This suggests that increased awareness of the hazards of transfusion has led to a lower threshold to report and an improvement in patient safety. This progress is probably due to a number of initiatives to improve hospital transfusion practice, including providing better training of the large number of staff involved at some stage of the transfusion process (see Chapter 23).

Symptoms and signs

These may become apparent within receiving as little as 20 mL of ABO-incompatible red cells. Initial clinical presentations include the following:

·        fever, chills or both;

·        pain at the infusion site or localized to the lower back/flanks, abdomen, chest or head;

·        hypotension, tachycardia or both;

·        agitation, distress and confusion, particularly in the elderly;

·        nausea or vomiting;

·        dyspnoea;

·        flushing; and

·        haemoglobinuria.

In anaesthetized patients, the only signs may be uncontrollable hypotension or excessive bleeding from the operative site, as a result of disseminated intravascular coagulation (DIC).

Some of these symptoms and signs can also be features of other transfusion reactions including bacterial contamination of the unit, allergic reactions, transfusion-related acute lung injury and febrile nonhemolytic reactions (see Chapters 8, 9 and 14).


Acute kidney injury develops in up to 36% of patients with AHTR as a result of acute tubular necrosis induced by both hypotension and DIC. Thrombus formation in renal arterioles may also cause cortical infarcts.

DIC develops in up to 10% of patients. TNF-α can induce tissue factor expression by endothelial cells and together with IL-1 can reduce the endothelial expression of thrombomodulin. Thromboplastic material is also liberated from leucocytes during the course of complement activation [5].

Immediate management of suspected AHTR (see Chapter 6)

Actions for nursing staff

In the presence of a fever of more than 1°C above the patient's pretransfusion temperature and/or any symptoms or signs mentioned above, the nursing staff should:

·        stop the transfusion, leaving the infusion line (‘giving set’) attached to the blood pack;

·        use a new giving set and keep an intravenous infusion running with normal saline;

·        call a member of the medical staff;

·        check that the patient identity as provided on the wristband corresponds with that given on the label on the blood pack and on the compatibility form;

·        save any urine the patient passes for later examination for haemoglobinuria; and

·        monitor the patient's pulse (P), blood pressure (BP) and temperature (T) at 15-minute intervals.

Actions for medical staff

The immediate actions depend upon the presenting symptoms and signs, and are summarized in Table 7.5.

Table 7.5 Immediate medical management of an acute transfusion reaction.


Likely diagnosis


Isolated fever or fever and shivering, stable observations, correct unit given

Febrile nonhaemolytic transfusion reaction (FNHTR)

Paracetamol 1 g orally/per os (PO) (in the US acetaminophen 625 mg), continue transfusion slowly observations of P, BP and T every 15 min for 1 h, then hourly. If no improvement then call haematology medical staff

Pruritus and/or urticaria

Allergic transfusion reaction

Chlorpheniramine 10 mg IV (US: diphenhydramine 25–50 mg PO or IV) and other actions as for suspected FNHTR

Any other symptoms/signs, hypotension or incorrect unit

Assume to be an acute haemolytic transfusion reaction in first instance

Discontinue transfusion, normal/saline to maintain urine output >1 mL/kg/h. Full and continuous monitoring of vital signs. Call haematology medical and transfusion laboratory staff immediately for further advice/action. Send discontinued unit of blood with attached giving set and other empty packs, after clamping securely, to the transfusion laboratory

Investigation of suspected AHTR

Blood samples should be taken from a site other than the infusion site for the investigations listed in Table 7.6.

Table 7.6 Laboratory investigation of suspected acute haemolytic transfusion reaction.

Blood test


Full blood count

Baseline parameters, red cell agglutinates on film

Plasma/urinary haemoglobin

Evidence of intravascular haemolysis

Haptoglobin, bilirubin, LDH

Evidence of intravascular or extravascular haemolysis

Blood group

Comparison of posttransfusion and retested pretransfusion samples, to detect ABO error not apparent at bedside. Unexpected ABO antibodies posttransfusion may result from transfused incompatible plasma. The donor ABO group should be confirmed

Direct antiglobin test (DAT)

Positive in majority, pretransfusion sample should be tested for comparison. May be negative if all incompatible cells destroyed

Compatibility testing

An IAT antibody screen and IAT cross-match using the pre- and posttransfusion sample provide evidence for the presence of alloantibody. Elution of antibody from posttransfusion red cells may aid identification of antibody or confirm specificities identified in serum in cases of non-ABO incompatibility. Red cell phenotype should also be performed on recipient pretransfusion sample and unit in cases of non-ABO incompatibility, to confirm absence in patient and presence in unit of corresponding antigen

Urea/creatinine and electrolytes

Baseline renal function

Coagulation screen

Detection of incipient DIC

Blood cultures from the patient and implicated pack(s)

In event of septic reaction caused by bacterial contamination of unit, which may be suspected from inspection of pack for lysis, altered colour or clots

Other reactions characterized by haemolysis

In patients with autoimmune haemolytic anaemia, transfusion may exacerbate the haemolysis and be associated with haemoglobinuria.

Donor units of red cells may also be haemolysed as a result of:

·        bacterial contamination;

·        excessive warming;

·        erroneous freezing;

·        addition of drugs or intravenous fluids;

·        trauma from extracorporeal devices; or

·        red cell enzyme deficiency.

Management of a confirmed AHTR

The management of haemolytic transfusion reactions should be determined by the severity of the clinical manifestations.

·        Maintain adequate renal perfusion while avoiding volume overload by:

a. maintenance of circulating volume with crystalloid and/or colloid infusions and,

b. if necessary, inotropic support.

·        Transfer to a high dependency area where continuous monitoring can take place.

·        Repeat coagulation and biochemistry screens 2- to 4-hourly.

·        If urinary output cannot be maintained at 1 mL/kg/h, seek expert renal advice.

·        Haemofiltration or dialysis may be required for the acute tubular necrosis.

·        In the event of the development of DIC, blood component therapy may be required.

·        Having ascertained the nature of the incompatibility causing the AHTR, transfusion of compatible blood may be required for life-threatening anaemia.

Prevention of AHTRs

Prevention of ‘wrong blood’ incidents

·        Prevention of the multiplicity of errors that can contribute to the transfusion of ABO-incompatible red cells must depend upon the creation of an effective quality system for the entire process, which will involve:

a. adherence to national guidelines and standards;

b. local procedures that are agreed, documented and validated;

c. training and retraining of key staff;

d. regular error analysis and review;

e. reporting to local Risk Management/Assurance Committee; and

f. reporting to regulatory bodies such as the Medicines and Healthcare Products Regulatory Agency (MHRA) in the UK, the Food and Drug Administration in the USA and to national haemovigilance schemes to contribute to the understanding of the extent and underlying causes.

These aspects are specifically covered in Chapter 23.

·        Since the majority of errors leading to an ABO-incompatible transfusion are due to misidentification of the patient or patient's sample, due attention must be paid to the comprehensive use of unique patient identifiers throughout the hospital and automation within the laboratory [2, 9, 10].

·        Access to previous transfusion records containing historical ABO groups should be available at all times.

·        It is desirable that computerized systems are used to verify at the bedside the matches between the patient and the sample taken for compatibility testing, and at the time of transfusion between the patient and the unit of blood.

Prevention of non-ABO AHTRs

·        In the case of recurrently transfused patients, due attention should be paid to the interval between sampling and transfusion, to optimize the detection of newly developing antibodies. In the UK for patients transfused within the previous 14 days, the pretransfusion sample should not be taken more than 3 days before the next transfusion [11,12]. In the USA the pretransfusion sample must be obtained within 3 days (72 hours) if the patient has been transfused or pregnant within the past three months. Similar requirements exist in other countries.

·        In the presence of multiple red cell alloantibodies, when it is not feasible to obtain compatible red cells in an emergency, intravenous immunoglobulin (1 g/kg/day for 3 days) and/or steroids (hydrocortisone 100 mg 6-hourly or methylprednisolone 1 g daily for 3 days) have been used with anecdotal reports of ameliorating a potential haemolytic or ‘hyperhaemolytic’ episode (see below).

Delayed HTRs

Aetiology and incidence

With few exceptions, DHTRs are due to secondary immune responses following re-exposure to a given red cell antigen. The recipient has been primarily sensitized to the antigen in pregnancy or as a result of a previous blood transfusion and a few days after a subsequent transfusion there is a rapid increase in the antibody concentration, resulting in the destruction of red cells.

·        The antibodies most commonly implicated and reported to SHOT between 1996 and 2006 were those from the Kidd blood group system followed by those from the Rh, Duffy and Kell systems. One analysis showed that in approximately 10% of reported cases, more than one alloantibody was found in the serum [13].

·        Frequently, there are no clinical signs of red cell destruction, but subsequent patient investigations reveal a positive direct antiglobulin test (DAT) and the emergence of a red cell antibody. This situation has been termed a delayed serological transfusion reaction (DSTR) [14].

·        Kidd and Duffy antibodies are more likely to cause symptoms and be associated with a DHTR rather than a DSTR.

·        Estimates of the frequency of DHTR and DSTR vary, but in a series reported from the Mayo Clinic, the frequency of DHTR was 1 in 5405 units and of DSTR was 1 in 2990 units, giving a combined frequency of 1 in 1899 units transfused [15].

·        DHTRs are in themselves rarely fatal, although in association with the underlying disease can lead to mortality.

·        Of transfusion fatalities reported to the United States Food and Drug Administration (FDA) between 1976 and 1985 10% were due to DHTR; in 75% of cases, more than one alloantibody was present in the serum and the same proportion involved non-Rh antibodies.

·        Six deaths reported to SHOT between 1996 and 2006 have been due to DHTRs [13]. Tragically, in some instances there were delays in diagnosis, investigation and provision of compatible units, which led to marked anaemia and contributed to mortality.

Signs and symptoms

These usually appear within 5–10 days following the transfusion, but intervals as short as 24 hours and as late as 41 days have been recorded. The exact onset may be difficult to define since haemolysis can be initially insidious and may only be appreciated from results of posttransfusion samples. The commonest features are:

·        fever;

·        fall in haemoglobin concentration; and

·        jaundice and hyperbilirubinemia.

Hypotension and kidney injury are uncommon (6% of cases). In the postoperative period in particular, the diagnosis may be overlooked and the symptoms and signs incorrectly attributed to continuing haemorrhage or sepsis. In the setting of sickle cell disease, DHTR can be particularly severe with destruction of autologous red cells (hyperhaemolysis) (see below).


The majority of DHTRs require no treatment because red cell destruction occurs gradually as antibody synthesis increases. Haemolysis may, however, contribute to the development of life-threatening anaemia, particularly in patients with ongoing bleeding, and urgent investigations are required to ensure the timely provision of antigen-negative units.

Expert medical advice may be required for treatment of the hypotension and renal failure. When accompanied by circulatory instability and renal insufficiency, a red cell exchange transfusion with antigen-negative units can curtail the haemolytic process. Future transfusions of red cells should also be negative for the antigen in question.

Investigation of suspected DHTR (see Chapter 6)

·        The peripheral blood film is likely to show spherocytosis.

·        Other evidence of haemolysis – namely hyperbilirubinaemia, elevated lactate dehydrogenase (LDH), reduced serum haptoglobin, haemoglobinaemia, haemoglobinuria and haemosiderinuria – is useful to confirm the nature of the reaction and to monitor progress.

·        The DAT usually becomes positive within a few days of the transfusion until the incompatible cells have been eliminated.

·        Further serological testing on pre- and posttransfusion samples should be undertaken in accordance with the schedule provided for AHTR.

·        The antibody may not be initially apparent in the posttransfusion serum but can be eluted from the red cells. If the red cell eluate is inconclusive, then a repeat sample should be taken after 7–10 days, to allow for an increase in antibody titre. However, additional, more sensitive techniques may have to be employed to detect the antibody and it is advisable to seek the help of a reference laboratory.

·        Since a significant proportion of cases have more than one alloantibody in the serum, it is important that the panels used for antibody identification have sufficient cells of appropriate phenotypes to exclude additional specificities.


Access to previous transfusion records may disclose the presence of antibodies undetectable at the time of cross-matching, and all patients should be questioned regarding previous transfusions and pregnancies. Patients found to have developed a clinically significant red cell alloantibody should be provided with an antibody card. When the care of patients requiring transfusion support is shared between hospitals, there must be adequate communication between laboratories and clinical teams.

Laboratories should ensure that their antibody screen is effective in detecting weak red cell alloantibodies and that screening cells are taken from homozygotes where the corresponding antibodies show a dosage effect (i.e. they are less easy to detect when red cells with heterozygous expression of the relevant antigen are used rather than cells with homozygous expression). Pretransfusion testing is covered in detail in Chapter 23.

Haemolysis resulting from haemopoietic stem cell transplantation (see Chapter 41)

Major ABO incompatible transplants

Infusion of bone marrow during major ABO-incompatible transplants can result in an AHTR (the recipient has antibodies against the donor's red cells, e.g. group A donor, group O recipient). The risk is dependent on the antibody titre of the recipient and the volume of red cells in the marrow harvest. Peripheral blood stem cell products rarely have enough red cells to result in clinical AHTR even if there is ABO incompatibility.

Minor ABO incompatible transplants

Most patients transplanted with minor ABO-incompatible marrow (the donor has antibodies against the recipient's red cells, e.g. O donor, A recipient) develop a positive DAT, but only 10–15% of patients develop clinically significant haemolysis. Haemolysis in minor ABO incompatibility is short-lived and exchange transfusion is rarely required. Red cells and plasma-containing components (platelets, FFP and cryoprecipitate) should be compatible with both recipient and donor.

It has been suggested that the use of peripheral blood stem cells may increase the risk of significant haemolysis since the number of lymphocytes infused with the graft is increased, and three deaths due to an AHTR were reported between 1997 and 1999 in minor ABO-incompatible transplants. Several cases due to anti-D have been described, and antibody production has persisted for up to 1 year [16].

Delayed haemolysis following organ transplantation (passenger lymphocyte syndrome)

Donor-derived B lymphocytes within the transplanted organ may mount an anamnestic response against the recipient's red cell antigens. Donor-derived antibodies are usually directed against antigens within the ABO and Rh systems. If ABO mismatched organs are transplanted, the frequency of occurrence of donor-derived antibodies and haemolysis increases with the lymphoid content of the graft, from kidney to liver to heart–lung transplants. The figures for haemolysis are 9%, 29% and 70%, respectively. The frequency of haemolysis increases with an O donor and an A recipient. Pretransplant isohaemaglutinin titres do not appear to predict the incidence or severity of haemolysis. The ABO antibodies, which appear 7–10 days after transplant, last for approximately one month. Haemolysis is usually mild, although several cases of renal failure and one death have been reported. It can be prevented by switching to group O cells, either at the end of surgery or postoperatively if the DAT becomes positive.

Rh antibodies have been described following kidney, liver and heart–lung transplants. They can cause haemolysis for up to 6 months, which can be sufficiently severe to merit therapy.

Haemolysis occurs 7–10 days after transplantation, with an unpredictable and abrupt onset [17].

Hyperhaemolytic transfusion reactions and haemolytic transfusion reactions in sickle cell disease

The frequency of alloimmunization in sickle cell anaemia is dependent upon the nature and success of the extended red cell antigen-matching policy employed. Approximately, 40% of patients who are alloimmunized have experienced or will experience a DHTR.

Although DHTRs are characteristically mild in other groups of recipients, they can be responsible for major morbidity in sickle cell disease. The term ‘sickle cell haemolytic transfusion reaction (SCHTR) syndrome’ has been suggested to capture some of the distinctive features that can be seen to accompany a reaction. A similar syndrome has been described in other transfusion-dependent patients, so the term ‘hyperhaemolytic transfusion reaction (HHTR)’ may be more appropriate. These features are as follows:

·        symptoms suggestive of a sickle cell pain crisis that develop or are intensified during the HTR;

·        marked reticulocytopenia (relative to pretransfusion levels);

·        development of a more severe anaemia after transfusion than was present before. This may be due to the suppression of erythropoiesis as a result of the transfusion but hyperhaemolysis of autologous red cells (bystander immune haemolysis) has also been suggested. There have been reports that bone marrow aspirates performed on patients suffering from this complication have shown evidence of active erythropoiesis during the reticulocytopenic phase and haemophagocytosis. This has led to the suggestion that erythroid precursors and reticulocytes are removed by adhesion to monocytes via other mechanisms, in addition to IgG and Fc receptors, such as the integrins α4β1 and VCAM-1;

·        subsequent transfusions may further exacerbate the anaemia and it may become fatal; and

·        patients often have multiple red blood cell alloantibodies and may also have autoantibodies, which make it difficult or impossible to find compatible units of red blood cells.

However, in other patients the DAT may be negative, no alloantibodies are identified and serological studies may not provide an explanation for the HTR; even red cells that are phenotypically matched with multiple patient antigens may be haemolysed.

Management involves withholding further transfusion and treating with corticosteroids (hydrocortisone 100 mg 6-hourly or methylprednisolone 1 g daily for 3 days), while IVIG (1 g/kg/day) may have been beneficial in some cases [18].

It is recommended that patients with sickle cell disease are phenotyped prior to transfusion and that blood is matched for Rhc, C, D, e, E and K (see Chapter 28).

Acute haemolysis from ABO-incompatible platelet transfusions

Rarely, the passive transfusion of anti-A or anti-B present in a platelet pack will cause haemolysis in the recipient. This is most commonly seen in type A recipients of type O platelets. Clinically significant reactions are rare: passive anti-A/B becomes diluted in the recipient's plasma and it will also bind to A or B antigen, both soluble in the recipient's plasma and on endothelial cells. The typical anti-A/B titre in a platelet donor is on the order of 1:128, but occasionally donors will have very high titres exceeding 10 000. Severe and even fatal AHTRs have been reported, particularly in cases where a large amount of incompatible ABO antibody is transfused into a recipient with a small plasma volume (e.g. pediatric patient). In the UK, all platelet units are required to be screened for anti-A/B using a cut-off titre of 1:100. Packs from donors who have titres below this level are marked ‘HT-negative’ or high titre negative. Approximately 10% of platelet units will have titres above 1:100 and will not be marked ‘HT-negative’; these are restricted to ABO-identical recipients. In the USA, no preventative strategy is currently mandated and local practices vary. AABB accredited transfusion services are simply required to have a policy concerning the transfusion of products having significant amounts of incompatible ABO antibodies.

Key points

1. HTRs are the second commonest cause in the UK and the USA of immediate morbidity and mortality following a transfusion (the most common cause is transfusion-related acute lung injury, TRALI).

2. The clinical presentations are diverse and they can be unrecognized or misdiagnosed.

3. Most fatal AHTRs have historically been due to the transfusion of ABO-incompatible red cells but there is evidence that increased transfusion safety awareness has reduced the frequency of ABO incompatible red cell transfusion. Other causes of AHTR are overtaking ABO in countries where haemovigilance schemes have been successful.

4. The transfusion of ABO-incompatible red cells is the result of an error occurring at any stage in the transfusion process. Patient identification errors are the most frequent culprit.

5. Devising and successfully implementing measures to overcome these preventable and fatal errors is a challenge but should be a priority for those involved in hospital transfusion.


1. Beauregard P & Blajchman MA. Haemolytic and pseudo-haemolytic transfusion reactions: an overview of the haemolytic transfusion reactions and the clinical conditions that mimic them. Transfus Med Rev 1994; 8: 184–199.

2. Serious Hazards of Transfusion. Annual Report 2010. Manchester: SHOT Office. ISBN 978-0-9558648-3-4. Available at:

3. Daniels G, Poole J, de Silva M, Callaghan T, MacLennan S & Smith N. The clinical significance of blood group antibodies. Transfusion Medicine 2002; 12(5): 287–295.

4. Davenport RD. Hemolytic transfusion reactions. In: MA Popovsky (ed.), Transfusion Reactions, 2nd edn. Bethesda, MD: AABB Press; 2001, pp. 2–36.

5. Davenport RD. Pathophysiology of hemolytic transfusion reactions. Seminars in Hematology 2005 July; 42(3): 165–168.

6. Linden JV, Wagner K, Voytovich AE & Sheehan J. Transfusion errors in New York State: an analysis of 10 years experience. Transfusion 2000; 40(10): 1207–1213.

7. Stainsby D. ABO incompatible transfusions – experience from the UK Serious Hazards of Transfusion (SHOT) scheme. Transfusion Clinique et Biologique 2005; 12(5): 385–388.

8. Sazama K. Reports of 355 transfusion-associated deaths: 1976 through 1985. Transfusion 1990; 30: 583–590.

9. National Patient Safety Agency. Right Patient, Right Blood. Safer Practice Notice No. 14. London: NPSA; 2006. Available at:

10. National Patient Safety Agency. Standardising Patient Wristbands Improves Patient Safety. Safer Practice Notice No. 24. London: NPSA; 2007. Available at:

11. British Committee for Standards in Haematology. Guidelines for pre-transfusion compatibility procedures in blood transfusion laboratories. Available at (acc-essed November 15, 2012).

12. Milkins CE, Berryman J, Cantwell C, Elliott C & Rowley M. Timing of sample collection in relation to previous transfusion: a proposal for changing the recommendations. British Blood Transfusion Society Annual Conference prize winning poster 2010. (p. 27).

13. Stainsby D, Jones H, Asher D, Atterbury C, Bonicelli A, Brant L, Chapman C, Davidson K, Gerrard R, Gray A, Knowles S, Love E, Milkins C, McClelland B, Norfolk D, Soldan K, Taylor C, Revill J, Williamson L & Cohen H. Serious hazards of transfusion: a decade of haemovigilance in the UK. Transfus Med Reviews 2006; 20(4): 273–282.

14. Vamvakas EC, Pineda AA, Reisner R, Santrach PJ & Moore SB. The differentiation of delayed haemolytic and delayed serologic transfusion reactions: incidence and predictors of haemolysis. Transfusion 1995; 35: 26–32.

15. Pineda AA, Vamvakas EC, Gorden LD, Winters JL & Moore SB. Trends in the incidence of delayed hemolytic and delayed serologic transfusion reactions. Transfusion 1999; 39(10): 1097–1103.

16. Daniel-Johnson J & Schwartz J. How do I approach ABO-incompatible hematopoietic progenitor cell transplantation? Transfusion 2011; 51(6): 1143–1149.

17. Petz LD. Immune hemolysis associated with transplantation. Semin Hematol 2005; 42: 145–155.

18. Win N, Sinha S, Lee E & Mills W. Treatment with intravenous immunoglobulin and steroids may correct severe anemia in hyperhemolytic transfusion reactions: case report and literature review. Transfus Med Rev 2010; 24 (1): 64–67.

Further reading

Daniels G. Human Blood Groups, 2nd edn. Oxford: Blackwell Science; 2002.

Food and Drug Administration. Fatalities Reported to FDA Following Blood Collection and Transfusion: Annual Summary for Fiscal Year 2010. Available at: ReportaProblem/TransfusionDonationFatalities/ucm254802.htm.

Klein H & Anstee D. Haemolytic transfusion reactions. In: HG Klein & D Anstee (eds), Mollison's Blood Transfusion in Clinical Medicine. Oxford: Backwell Science; 2006.

Popovsky MA. Transfusion Reactions, 3rd edn. Bethesda, MD: AABB Press; 2007.

Serious Hazards of Transfusion (SHOT) resources and reports. Available at:

Win N. The clinical significance of blood group alloantibodies and the supply of blood for transfusion; 2007. Available at: