Blood group antigens and antibodies


Blood group antigens and antibodies


Blood group antigens and antibodies


Blood group antigens and antibodies

  27 Blood group antigens and antibodies

Volker Kretschmer, Tobias Legler

27.1 Blood group antigens

Blood group antigens are:

  • Genetically coded
  • Defined chemical structures of blood components
  • May elicit an immune response after their administration to an immunocompetent organism that does not possess these antigenic determinants and therefore recognizes them as foreign.

27.1.1 Antigens, epitopes

Antigens are defined on the basis of the antibodies that are directed against them. The epitope represents the smallest unit that can be recognized by an antibody. The individual blood group antigens consist of one to many epitopes. The number of antigenic determinants on a cell varies depending on the antigen, and in erythrocytes (red blood cells, RBCs) can range from 2000–4000 for antigens of the Landsteiner-Wiener (LW) system to as many as one million for antigens of the ABO system /1/. The development of the different antigens during the fetal period also varies considerably (Tab. 27-1 – Blood group antigens on the erythrocytes of newborns) and is not completed until some time during the second year of life.

Chemical structure

The chemical structure of blood group antigens is composed either of carbohydrate compounds that are based on sugars bound to proteins (glycoproteins) and/or lipids (glycolipids) or of protein molecules.


The genes are located on a single locus or on multiple closely linked loci of the corresponding chromosomes, which accounts for the fact that crossing-over events are very rare and that the mode of inheritance of these antigens is therefore very strict. The blood group antigens belong to polymorphic systems that are genetically independent from each other and that are regulated either as a primary (protein antigens) or secondary (carbohydrate antigens) product by corresponding allelic genes. For most of the genes, two alternative alleles exist (antithetically) that occur homozygously or heterozygously on both haplotypes of a chromosome pair; these alleles regulate antigen expression in a dominant/recessive or codominant manner. For the synthesis of protein antigens, only a small section of the corresponding gene is usually responsible whereas the synthesis of carbohydrate antigens in general requires the cooperation of several genes. Most of the systems are based on protein antigens. The primary structure of the majority of the most important blood group systems is almost completely known.

More than 500 erythrocyte antigens have been serologically defined /2/. The nomenclature of the International Society of Blood Transfusion Working Party lists 30 genetically independent human blood group systems that give rise to 236 different hereditary antigens /34/.


The frequency of the different blood group antigens in the population varies considerably. The distribution is characterized by geographical and ethnic differences.


Besides the blood group antigens on erythrocytes (i.e., blood groups in the narrower sense), such antigens also occur:

  • On leukocytes in the form of ABO, HLA, and granulocyte specific antigens
  • On platelets in the form of ABO, HLA, and human platelet antigens (HPA)
  • Attached to plasma factors (e.g., as the Am, Gm, and Km groups of immunoglobulins)
  • In the form of soluble blood group antigens (e.g., Lewis, Chido/Rodgers).

Furthermore, a number of blood group antigens are also detectable on other body cells and in body fluids. This chapter is restricted exclusively to the antigens found on erythrocytes (Tab. 27-2 – Biological function of blood group antigens on erythrocytes).

27.1.2 Immune reactions

Immune reactions include a variety of different reactions such as sensitization followed by the synthesis of antibodies, immunosuppression, graft rejection, and, in the presence of antibodies, humoral immune responses. As far as erythrocyte antigens are concerned, only the synthesis of erythrocyte antibodies and their detection are important for preventing the corresponding antigen-antibody reactions.

27.1.3 Clinical implications of blood groups

Biological function of blood group antigens

The biological function of the blood group antigens is to maintain cellular integrity and to assist in the process of cell maturation and cell survival. Blood group antigens are an expression of human individuality. Certain blood group antigens are located on membrane proteins that are involved in various functions such as the transport, for example, of anions (Diego), water (Colton), urea (Kidd), or ammonium (Rh proteins); cytokine receptor or adhesion molecule functions (Duffy, Lutheran, LW); enzyme functions; or the activation and regulation of complement (Chido/Rogers, Cromer, Knops).

Blood group antigens and disease

A relationship exists between certain blood groups and diseases with respect to:

  • Special susceptibility to and resistance against diseases (refer to ABO, MNSs, P, Duffy systems)
  • Alterations in blood group antigens, mainly in the form of attenuation, in hematopoietic diseases. The clinical significance of these alterations remains largely unknown.

In many tumors, a reduction in the expression of A and/or B antigens accompanied by an increase in the expression of precursor molecules is often seen in conjunction with increased malignancy /5/.

Blood transfusion related implications

The significance of blood groups as far as blood transfusions are concerned is mainly due to the antibodies that are directed against the blood group antigens. In addition, some blood groups also play a role in certain organ transplantations.

27.2 Blood group antibodies, lectins

27.2.1 Definition

Blood group antibodies define the various blood group antigens. They function as alloantibodies, autoantibodies, and heteroantibodies. Their designations (specificity) are derived from the antigens against which they are directed.


Alloantibodies are produced after a person has been exposed to foreign blood group antigens. This occurs after exposure either to erythrocytes (as of > 100 μL; 109 erythrocytes) during pregnancy or blood transfusions (immune antibodies) or to blood-group like antigens of microorganisms (ubiquitous antigens) with which humans are confronted, e.g. as a result of bacterial colonization of the intestine, food intake, infections, or vaccinations (natural antibodies). Immune antibodies generally react optimally at 37 °C (warm antibodies). The risk of developing antibodies to erythrocytes is approximately 1–5% in perioperative patients who have received transfusions of packed red blood cells /67/, 1–13% in recipients of multiple transfusions with congenital anemia (e.g. sickle cell anemia), and up to 30% in patients with myelodysplastic syndrome /8/. In transfusion recipients who are already immunized against erythrocytes, the risk of becoming immunized against additional blood group antigens is up to 20 times higher /9/. Most natural antibodies bind more readily at lower temperatures and are often released again from their antigenic binding at temperatures > 30 °C (cold antibodies).


Autoantibodies to erythrocyte antigens are not only directed against the producer’s own antigens but also against blood group antigens that occur very frequently in the population. The clinical symptoms of autoimmune hemolytic anemias depend on the thermal reaction optimum of the autoantibodies.


Heteroantibodies against erythrocyte antigens are blood group antibodies that originate from a different species. In the past, they were widely used as test sera, having been obtained by immunizing animals. Nowadays, they are mainly used as anti-human globulin reagents.


A distinction must be made between antibodies and lectins. Lectins are carbohydrate binding molecules (mostly of plant origin) that can specifically bind blood group antigens; they can therefore be used to determine various blood group antigens (Tab. 27-3 – Lectins).

27.2.2 Structure and function of antibodies

Blood group antibodies belong mainly to immunoglobulin classes G and M, and less often to class A. Except in sporadic cases of autoimmune hemolytic anemia, the blood group antibodies produced by humans are polyclonal (i.e., they are usually directed against multiple epitopes of the same antigen, monospecific) and are composed of a mixture of antibodies that belong to different immunoglobulin classes and subclasses. The same applies to test sera of human or animal (heterologous antibodies) origin.

In contrast to polyclonal antibodies, monoclonal antibodies are not only monospecific, but also directed only against individual epitopes. By selecting appropriate clones, it is possible to produce monoclonal antibodies that are far superior to polyclonal antisera as blood group reagents because of their specificity, purity, strength, and speed of reaction /10/. IgG anti-blood group antibodies

Alloantibodies produced during pregnancy and as a result of transfusions (immune antibodies) and most of the autoantibodies found in autoimmune hemolysis are IgG antibodies and frequently belong to subclass IgG1, more rarely to IgG3. These antibodies may activate complement, and after the formation of immune complexes, they show a high affinity for the corresponding receptors on macrophages. IgG2 antibodies activate significantly less complement while IgG4 antibodies do not activate complement at all and do not bind to macrophages.

A prerequisite for complement activation is that at least two IgG molecules are bound immediately adjacent to each other. Therefore, complement activation in the presence of blood group antibodies depends to a large extent on the antigen specificity. If the antigens are located too far apart from each other and are incapable of forming clusters or binding multiple antibodies, complement activation cannot take place.

The immune complexes thus produced are subsequently bound to Fc and/or C3b macrophage receptors, phagocytosed, and rapidly eliminated extravascularly (extravascular hemolysis). The activation of the complement cascade leads to intravascular cell lysis (intravascular hemolysis). IgG3 exhibit the highest degree of hemolytic activity, followed by IgG1. Only IgG antibodies, especially IgG1, pass the placental barrier.

As soon as IgG antibodies are produced, a corresponding immunological memory exists. Even when the antibodies are no longer detectable, a second exposure to the same antigen may cause a booster effect. Once IgG blood group antibodies have been detected, they must therefore be taken into consideration for the rest of the patient’s life whenever blood transfusions are necessary. IgM anti-blood group antibodies

Following primary sensitization to blood group antigens, IgM antibodies are produced initially but rarely detected. If sensitization is strong enough, IgG antibodies subsequently appear within 3 weeks to 3 months and an immunological memory develops for these antigens.

Immunization to blood group like antigens such as those found in microorganisms or food (ubiquitous antigens) is generally associated with the production of IgM antibodies only. These “natural” antibodies are usually reactive at lower temperatures. Under abnormal conditions, they may be synthesized at much higher rates and may cause, for example, cold antibody autoimmune hemolysis.

Not all IgM blood group antibodies activate complement. Their ability to activate complement also depends on other factors such as antigen specificity and thermal amplitude. IgM antibodies do not bind directly to macrophages. Therefore, only those IgM antibodies that still bind to erythrocytes at temperatures greater than 30 °C and are capable of activating complement are clinically relevant /11/. Since IgM antibodies are not subject to booster effects, they are only of interest during the period when they are detectable.

27.2.3 Serological reactivity

The various blood group antibodies (also if used as test reagents) differ in their reactivity in the basic serological test methods. Based on their serological reactivity, the following types of antibodies can be differentiated:

  • Agglutinating antibodies (agglutinins and complete antibodies)
  • Conglutinating (incomplete) antibodies
  • Blocking antibodies
  • Complement activating antibodies
  • Hemolyzing antibodies (hemolysins)
  • Enzyme reactive antibodies
  • Coombs antibodies.

Although these terms are largely historical, are not always clearly distinguishable from each other, and do not always correspond to the immunological properties of the antibodies, they have proved useful for practical purposes in clinical laboratories. Furthermore, blood group antibodies are classified according to their optimal reaction temperature, which may be 37 °C (warm antibodies) or lower (cold antibodies). Agglutinations caused by antigen-antibody complexes must be differentiated from “pseudoagglutinations.” Agglutinating (complete) antibodies

Complete antibodies, after binding to erythrocytes with the corresponding antigens, can trigger visible agglutinations of these erythrocytes (microscopically visible clumps) even in a very simple test medium (normal saline). In this milieu, erythrocytes display a negatively charged surface and, because of these charges, repel each other with a large separation distance; therefore, often only the larger IgM antibodies are capable of establishing the bridge between the erythrocytes that is necessary for agglutination.

Because blood group antibodies of the IgM class frequently show cold reactivity, agglutination tests are usually performed at room temperature.

Since the negative surface charges are mostly caused by glycoproteins and glycolipids on which the blood group antigens are located, high titers of IgG antibodies (e.g. Rh antibodies or warm antibodies) may also cause an agglutination reaction under these conditions. Because of the high antibody load, the negative surface charges are markedly diminished. This causes the distances between erythrocytes to be reduced to such a degree that even the small bivalent IgG molecules can then bridge them. The immunoglobulin class of an antibody can therefore not be deduced automatically from its detection in the agglutination test.

Agglutination tests are of immunohematological importance with regard to the determination of blood groups, the detection of abnormally increased cold agglutinins, and the investigation of irregular antibodies. It is not advisable to use these tests as part of antibody screening or cross matches (except for the detection of ABO antibodies) because the cold agglutinins that may be detected are clinically irrelevant, prompt unnecessary diagnostic investigations, and may impede the detection of clinically relevant antibodies. Conglutinating and blocking (incomplete) antibodies

Antibodies that are serologically referred to as incomplete are structurally and immunologically complete antibodies. They usually bind to erythrocytes in a saline medium but are unable to form any or any stable agglutinations due to their small size and number as well as their low binding affinity. They can sometimes block the reaction of simultaneously or subsequently added agglutinins that have the same specificity.

The detection of incomplete antibodies requires the assistance of certain media (supplements) that cause a reduction in the negative surface charge on erythrocytes (dextrans, albumin), a decrease in the water envelope or the surface tension of the erythrocytes (albumin, proteases), an increase in the binding affinity of the antibodies (low ionic strength solution, LISS), or an improvement in the steric access for the antibodies (proteases). The antibodies to be detected are usually warm IgG antibodies. Low titers of IgM antibodies or IgM antibodies with little avidity may possibly be detected as well, only by using such supplement and/or conglutination tests. Since complete antibodies can also be detected by supplement tests (often even more strongly so), positive results obtained by these tests generally do not allow conclusions to be drawn concerning the presence of IgG antibodies.

Antibodies that are only detectable in the anti-globulin test are also referred to as incomplete, but not conglutinating antibodies. Complement activating antibodies and hemolysins

Certain blood group antibodies may activate the complement cascade (Chapter 24 – The complement system):

  • Partially up to C3b (e.g., anti-K)
  • Completely as hemolysins (e.g., anti-A).

The complement activation may in general be detected in the antiglobulin test and/or, in the case of complete activation, by hemolysis tests.

Because of the different optimal temperatures for various hemolysins, the test is conducted as a:

  • Mono thermal cold hemolysin test (isohemolysins, cold autohemolysins)
  • Bithermal cold hemolysin test (Donath-Landsteiner hemolysins)
  • Warm hemolysin test (warm autohemolysins).

If hemolysis can be achieved in vitro only by using enzyme treated erythrocytes, incomplete hemolysins are considered to be present, although in general the use of enzyme treated erythrocytes also increases the sensitivity of the test.

A positive result for one of these hemolysis tests does not allow conclusions to be drawn concerning the origin of these antibodies since, for example, strong cold hemolysins may still also react (at a weaker level) in the warm hemolysin test, and vice versa.

The reaction optimum (reaction strength, titer) and other findings, such as the immunoglobulin class of the antibodies, are decisive for the diagnosis.

Complement activation and/or hemolysis is only detectable if fresh serum containing adequate complement is used or if fresh AB serum is added as a source of complement. Sometimes, complement activation is the only way to recognize a preceding immune reaction, for example, in the case of alloantibodies with low avidity such as Lewis antibodies, rare cold autoantibodies (bithermal cold hemolysins), and drug induced antibodies. These antibodies sometimes do not bind adequately or strongly enough to the erythrocytes and are therefore eluted during washing in the antiglobulin test. This is not a problem if the gel centrifugation method is used, because washing is not required. Since antibodies with lower avidity can be detected directly much more often using this method, indirect detection using complement activation has become less important. This justifies the use of citrate and EDTA as anticoagulants in diagnostic tests for blood group antibodies. Enzyme reactive antibodies

More than 70% of blood group antibodies also react in the enzyme test. Most cold antibodies react significantly more strongly (> 4 titer dilutions) in the presence of enzymes or with enzyme treated erythrocytes (Tab. 27-4 – Impact of enzymes on blood group antigens).

In 5–15%, antibodies to important (mostly Rh) and to less important blood group antigens (Lewis, P) are only detectable in the enzyme test. The clinical relevance of these “only enzyme reactive” antibodies is generally considered to be low. However, almost half of these antibodies can be identified as IgG antibodies if highly sensitive solid phase technology is used (Section – Micro column agglutination test). Because of the high frequency of positive enzyme test results due to nonspecific agglutination and the reaction of clinically irrelevant natural cold antibodies (e.g., anti-I, anti-IH, anti-H), routine performance of these tests is usually not recommended. Nevertheless, the enzyme test, conducted at 37 °C, may be useful for the early detection of immunization in certain groups of patients who are at risk of developing antibodies such as pregnant women and recipients of multiple blood transfusions /12/.

The enzyme test is also indispensable for the investigation and identification of antibodies. It is especially useful that various antigens can be destroyed by pretreating erythrocytes with certain enzymes (Tab. 27-4 – Impact of enzymes on blood group antigens), with the result that the corresponding antibodies can no longer react in the two stage enzyme test (Section – Supplement tests). In one stage enzyme tests, the rapid antibody binding may partially prevent the destruction of antigens, thus also allowing the detection of antibodies to enzyme sensitive antigens. Coombs antibodies

The term Coombs antibodies refers to incomplete antibodies (mostly IgG) that are detectable in the direct or indirect antiglobulin test (Coombs test).

The direct antiglobulin test detects immunoglobulins and complement factors on patient and control erythrocytes.

The indirect antiglobulin test allows the detection of immunoglobulins in the serum/plasma, if these immunoglobulins have bound to test erythrocytes in vitro. If the immune reaction in vitro is associated with complement activation and the antiglobulin reagent also contains antibodies to complement factors, complement factors on the erythrocytes are also detected.

Since the poly specific antiglobulin serum used for these tests usually contains not only antibodies to human IgG (possibly to IgM and/or IgA as well) but also antibodies to complement factors, especially C3d, a positive antiglobulin test by itself does not allow the conclusion that antibodies, or specifically IgG antibodies, are present. Nonetheless, greater clinical relevance is ascribed to such a finding since the antiglobulin test directly or indirectly (via their complement activation) detects clinically important immune antibodies more specifically than other hemagglutination tests (agglutination and supplement tests). This is based on the premise that clinically irrelevant complement activation due to the effect of benign cold antibodies does not occur in vitro. C1 induced (classical pathway) complement activation can be blocked by the binding of Ca2+ions (sodium citrate), or even more effectively by the simultaneous binding of Mg2+ ions (EDTA). Thus, the subsequent attachment of complement to patient erythrocytes in vitro can be ruled out by using EDTA blood or citrated blood in the direct antiglobulin test.

In the indirect antiglobulin test, the effect of benign cold antibodies should be prevented by avoiding cold incubation. Cold antibodies that are clinically relevant either agglutinate even above 30 °C or are detected indirectly via their complement binding. Irregular antibodies

Although from an immunological point of view, this term makes no sense, the classification of blood group antibodies into regular and irregular antibodies has established itself for practical purposes.

  • Regular antibodies are the regularly occurring anti-A or anti-B antibodies that correspond to the ABO blood groups
  • Irregular antibodies include all other antibodies to blood group antigens, including non corresponding ABO antibodies in the case of weak ABO phenotypes; this is regardless of which type of serological activity these antibodies display or whether they represent immune antibodies or naturally occurring antibodies. Pseudoagglutination

Pseudoagglutinations are nonspecific, non immunologically induced clumps of erythrocytes that interfere with hemagglutination tests. Microscopically, their typical appearance is referred to as rouleaux formation (like rolls of coins). However, microscopic assessment (especially by inexperienced personnel) does not allow reliable distinction between specific types of agglutination. Gel centrifugation tests are particularly susceptible to interference factors (1–2% in the indirect antiglobulin test and 2–4% in the enzyme test). Pseudoagglutinations occur on top of the gel columns or macromolecules block the capillaries of the gel columns, preventing erythrocyte sedimentation.

The causes of pseudoagglutination are manifold (Tab. 27-5 – Causes of pseudoagglutination). Pseudoagglutinations are often found in patients with large blood losses who have received large volumes of plasma expanders and who also have coagulation disorders (bleeding esophageal varices). This type of interference can be particularly troublesome in such clinical settings because urgent blood transfusions may be delayed due to diagnostic problems. Clinicians are therefore advised to collect blood samples prior to the administration of large quantities of plasma expanders if possible and/or via a separate venous access.

27.2.4 Blood-transfusion-related relevance of blood group antibodies

The different antibodies to erythrocyte antigens are of variable relevance with respect to blood transfusions, depending on the following factors:

  • Their frequency (Tab. 27-6 – Frequency of irregular erythrocyte antibodies).
  • The frequency of the antigens against which they are directed (antigen frequency)
  • Their hemolytic activity (e.g. with regard to binding to macrophages, optimal temperature, and complement activation)
  • Their concentration
  • Their ability to pass through the placenta and the expression of corresponding antigens on fetal erythrocytes
  • The detection of antibodies and the corresponding antigens under routine conditions
  • Antibody induced interference (especially by cold antibodies and autoantibodies) with immunohematological examinations. Implication of immunization for the transfusion of blood

The life span of the transfused erythrocytes that induced the primary immunization is rarely shortened since antibody production usually does not start until after the normal life cycle of the erythrocytes (three months).

However, the presence of blood group antibodies is associated with the following risks and disadvantages for future transfusions:

  • Delays in urgent transfusions due to the need to identify such antibodies
  • Inadequate availability of compatible erythrocyte components (supply problems), especially if several antibodies are present simultaneously (antibody mixtures) or if antibodies to high-frequency antigens are present
  • Risk of delayed hemolytic transfusion reactions if immunization is not detectable at the time of transfusion because of low antibody concentrations and a booster effect due to the transfusion causes these concentrations to rise /1415/. After one year, almost one-third of the irregular antibodies are no longer detectable; after five years, this figure has risen to 50% /16/.
  • The occurrence of acute hemolytic transfusion reactions if the antibodies are not taken into account (due to clerical or methodological errors) and/or cannot be taken into account (emergency situations, supply problems). Implication of immunization for pregnancy

During pregnancy, hemolytic disease of the fetus and newborn can be triggered by the transplacental transfer of an effective dose of antibodies (IgG) that are directed against fetal erythrocytes. Such immunization is most likely to occur during the last trimester, during delivery, and following invasive procedures during pregnancy. For this reason, approximately one-third of the antibodies produced during pregnancy only become detectable at the start of the last trimester /1718/. Significance of autoantibodies

In patients, abnormal autoantibodies or autoantibodies produced in abnormally high concentrations may cause hemolysis of both the autologous as well as the transfused erythrocytes. However, by no means all autoantibodies are hemolytically active. Incomplete warm autoantibodies (IgG) may very occasionally trigger fetomaternal incompatibility, but only if they have already shown hemolytic activity in the mother. Compatibility of blood components

A blood transfusion is considered compatible if it does not contain clinically significant concentrations of antigens against which the recipient has already produced antibodies or is immunized (major compatibility) or clinically relevant concentrations of antibodies against the recipient’s own antigens (minor compatibility) are not present. Donors and recipients do not need to have the same blood group type to ensure compatibility. For information about irregular erythrocyte antibodies that need to be taken into account for erythrocyte transfusions (Tab. 27-24 – Consideration of irregular erythrocyte antibodies).

27.3 Blood group systems expressing carbohydrate antigens

The ABO, H, Ii, Lewis, and P systems are systems whose antigens are formed by sugar molecules. Sugars are sequentially added to mutual or very similar precursor molecules (glycoproteins and glycolipids) by means of genetically determined transferases, thus accounting for the fact that these systems are closely related in various ways.

27.3.1 ABH(O) blood groups

Sequentially synthesized carbohydrate structures (type 2 carbohydrate chains) are produced on the erythrocyte membrane as a result of the cooperation of several genetically determined transferases. Initially, i chains (i antigen) are synthesized as precursor molecules. The i antigen is created by the action of an acetylglucosaminyltransferase that binds N-acetyl glucosamine and activates a galactosyltransferase, thus causing the carbohydrate chains to branch. Encoded by the H gene, fucose is added to the terminal galactose molecules of some of these precursor molecules by means of a fucosyltransferase.

The further synthesis of A and B depends on the presence of the precursor molecule H and again is mediated by the action of transferases (Fig. 27-1 – Biosynthesis of the blood group antigens A and B). The B gene encodes a galactosyltransferase that binds galactose to the H antigen. In blood group B, a significant number of H antigens remain unbound. The A gene encodes an N-acetylgalactosaminyltransferase, which binds N-acetylgalactosamine to the carbohydrate chain in a similar manner. Mutations in the Transferase gene produce extensive heterogeneity in the ABO system with a range of rare phenotypes /20/.

Secretors: in carriers of the Se allele of the fucosyltransferase gene FUT2 soluble ABH antigens are found in various body fluids. Homozygous carriers of the Se allele do not produce soluble ABH antigens in the body fluids and are known as non secretors. A and B phenotypes

A1 phenotypes and A2

The transferase of A2 allele carriers is 21 amino acids longer than that of A1 individuals, which has significantly reduced activity. During the synthesis of A2 less branched molecules are formed and less N-acetyl galactosamine is bound to the erythrocyte surface (approximately 290,000 per erythrocyte); hence, H is also detectable as an antigen. In the case of A1, up to one million N-acetyl galactosamine molecules are bound per erythrocyte.

Weaker A phenotypes

The situation is similar for other subtypes of blood group A. Accordingly, certain mutations of A3 and Ax also exist that lead to even less active transferases; as a result of this, even less A antigen is synthesized and correspondingly more H antigen is detectable. The serological subdivision of A variants can be difficult and is not always clear cut.

Para Bombay blood groups

The very rare para Bombay blood groups (Ah and Bh) also react like very weak A and B variants, respectively. A mutated H gene synthesizes only a very small number of H receptors; all of these receptor sites are completely occupied by the few A or B determinants and H antigen is therefore no longer detectable.

B phenotypes

These are much less common than the A phenotypes. For B3 an amino acid substitution has been described as a genetic cause, where Arg-352 has been replaced by tryptophan close to the C-terminus.

Accessory A and B antigens

B(A) or A1(B) phenomenon: since the transferases that lead to the synthesis of A1 and B molecules differ in only 4 out of 353 amino acids, errors in the synthesis may occasionally occur. Accordingly, small amounts of A specific sugars may be erroneously added to the terminus in blood group B and vice versa. Such antigens are detectable only by the use of high affinity monoclonal antibodies. The corresponding blood groups have become known as the B(A) or A1(B) phenomenon /21/. They are not relevant either for transfusions or for diagnostic aspects. The test reagents routinely used are adjusted in such a way that they do not detect this small number of antigens.

Accessory B antigens: bacterial infections (e.g., those seen in inflammatory bowel disease and colon cancer) can also occasionally produce these accessory B antigens on the surface of erythrocytes in individuals with blood group A, if bacterial enzymes (deacetylases) partially remove the acetyl residues that differentiate A from B. These acquired B like antigens are occasionally of diagnostic relevance (erroneous blood group determinations, indicators for certain diseases). Blood group 0

In blood group 0, the 0 gene usually (in approximately 96% of Caucasians with this blood group) synthesizes a transferase that remains inactive because it is a markedly shortened protein. Hence, the H characteristic remains the determinant antigen in blood group 0. In a small percentage (approximately 4%) of cases, a second form, referred to as 02, is also found. Although this second form is characterized by the presence of normal length transferase molecules, the molecules remain inactive due to another mutation. These variants need to be taken into account during blood group determination using molecular genetic methods. Further variants have been identified in the meantime. ABH(0) in newborns

During the fetal period, the described synthesis of these carbohydrate antigens occurs more slowly than the development of other blood group antigens that are, for example, components of structural proteins; the carbohydrate antigen synthesis is not completed until the end of the second year of life. Thus, newborn red blood cells (cord cells) still show significantly fewer I antigens and less I, H, and A,B antigens on their surface than adult erythrocytes (Tab. 27-7 – Density of AB0 antigens (molecules) per erythrocyte). Bombay type

In the very rare case of homozygosity for the h allele (hh), no A and/or B sugars can be bound, despite the presence not only of the A and/or B alleles but also of the transferases encoded by these alleles.

The h gene leads only to the synthesis of an inactive enzyme that is incapable of binding the necessary fucose to the I antigen. In these cases, neither H, A, nor B antigens are detectable on the erythrocytes (Bombay type, 0h).

In addition, there are several mutations of H-antigen-synthesizing fucosyltransferase that lead to inactive enzyme molecules and are also therefore associated with absent H antigen /22/. Chimerism

If donor and recipient are AB0 incompatible, two erythrocyte populations with different AB0 blood groups are seen transiently following bone marrow transplantation. This phenomenon occurs much more commonly after ABO incompatible transfusions (e.g., when red blood cells of the blood group 0 must be transfused in an emergency before the AB0 blood group has been determined).

During AB0 blood grouping, this phenomenon is recognizable by the discrepancy between the results of the antigen and the antibody determination as well as by mixed field agglutination. An overview of the most important ABH variants is presented in Tab. 27-8 – The most important ABH variants. Isoagglutinins and isohemolysins


A peculiarity of the ABH(0) system lies in the fact that, given an intact immune system, the synthesis of natural AB0 antibodies (isoagglutinins) starts during the first year of life and then continues for life.

The production of anti-A and anti-B (and anti-AB) results from a corresponding induction by bacterial antigens (e.g. E. coli) that display surface structures almost identical to the AB0 sugars. During this process, only antibodies that do not react with the individual’s own erythrocytes are produced (Tab. 27-9 – AB0 blood groups: distribution and detectability of regular and irregular antibodies).


These antibodies are cold antibodies but their thermal amplitude extends to over 37 °C. They belong mainly to the IgM class. In general, the concentration of anti-A is higher than that of anti-B. IgG antibodies are also found, especially in individuals with blood group 0 (more commonly in women than in men). Furthermore, individuals with blood group 0 produce anti-AB (i.e., antibodies that bind both to A and to B because they are evidently incapable of differentiating between galactose and N-acetylgalactosamine as the terminal sugar). The highest AB0 antibody titers are reached by the age of 5–10 years; with aging, the concentration of these antibodies declines markedly. Both IgM and IgG antibodies can activate the complement cascade fully (isohemolysins). Isoagglutinins and isohemolysins also occur as regular antibodies in the various weak A and B variants, with accessory A and B antigens, and with blood groups such as the Bombay and para-Bombay type. Irregular AB0 antibodies

Weak A antigens (more common in AB than A) are associated with increasing amounts of irregular cold antibodies of the IgM class directed against A antigenic determinants as the expression of A antigenic determinants declines (Tab. 27-9 – AB0 blood groups: distribution and detectability of regular and irregular antibodies). In general, these antibodies are clinically irrelevant.

The same applies to anti-H in blood group A1. In the Bombay blood group, however, anti-H represents a hemolytically very active antibody with a broad thermal amplitude that is as reactive as anti-A or anti-B. Other irrelevant cold antibodies that correspond with the AB0 system include anti-IH and anti-IA; they bind only in the presence of both antigens. AB0 antibodies may also rarely occur in the form of harmless cold autoantibodies.


AB0 antibodies react optimally at low temperatures in the saline test. However, they can still be detected at 37 °C. Because of the antibody specificity, regular and irregular antibodies are not detected in antibody screening but by reverse typing during the AB0 determination. Their ability to activate complement is determined by means of the isohemolysin test. IgM antibodies can be differentiated from IgG antibodies by neutralization of the IgM antibodies with soluble AB substance (AB neutralization test, refer to Section – Neutralization tests. Significance of the ABH(0) system

ABH(0) system and disease

A number of associations exist between the ABH (0) system and certain diseases; the causes for these associations have not yet been sufficiently determined and are of no practical relevance:

  • Attenuation of A, B, and H antigens( e.g., in leukemias, Hodgkin’s disease, lymphomas
  • One abnormality commonly observed in tumor cells, and which is associated with significantly increased malignancy, is reduced expression of A and/or B antigens in conjunction with increased expression of precursor H molecules /5/.
  • Individuals with blood group A have a 20% higher probability of developing gastric cancer than those with blood group 0. Blood group A is also more frequently found in patients with intestinal and salivary tumors. Various tumors have A like structures, which may explain why they are more likely to circumvent immune defense in the presence of blood group A. Soluble A antigens may also be found in higher quantities in secretors with such tumors and cause interference if unwashed patient cells are used for AB0 grouping.
  • Individuals with blood group 0, especially non secretors, have a 20% higher probability of developing gastric and duodenal ulcers and are particularly prone to gastrointestinal hemorrhage. A possible explanation for this is that less IgA is secreted in non secretors and that the concentrations of von Willebrand factor (VWF) and factor VIII:C are significantly lower in individuals with blood group O. The AB0 blood group should therefore be taken into account when selecting reference values for VWF.
  • Thromboembolic events occur more frequently in individuals with blood group A
  • Individuals with certain AB0 blood groups are more susceptible to some infections because the pathogens display similar blood group antigens and the immune response to these microorganisms is impaired. This occurs, for example, in individuals with blood group 0 who are confronted with Yersinia pestis (H like antigen), in those with blood group A who are confronted with varicella (A like antigen), and in those with blood group B who are confronted with salmonellae and shigellae (B like antigen). Blood-transfusion related implications

As far as blood transfusions are concerned, the special significance of the ABH(0) system is derived from:

  • The high immunogenecity and antigenicity of the antigenic determinants
  • The physiological occurrence of AB0 antibodies, not only in high concentrations but with a wide thermal amplitude and pronounced avidity as well as the ability to activate complement (i.e., with the ability to hemolyze erythrocytes in vitro and in vivo).

Higher antibody titers can be found in autoimmune hemolysis, alcoholic cirrhosis, chronic active hepatitis, pregnancy, and following vaccinations and bacterial infections.

Major AB0 incompatibility: acute, life threatening transfusion reactions with intravascular hemolysis may occur following the transfusion of erythrocytes with an incompatible AB0 blood group.

Minor AB0 incompatibility: in the case of passive transfer of AB0 antibodies, clinically relevant hemolysis is only seen if large volumes of plasma and/or plasma containing blood units with a high antibody titer (hemolysins, isoagglutinin titer > 100) are transfused.

Hemolytic disease of the newborn: AB0 IgG antibodies are the most common cause of hemolytic disease of the newborn. However, because of the relatively few antigenic binding sites and the adsorption of the antibodies to the AB0 antigens of the endothelial cells, severe hemolysis is a rare event. Hemolysis is more likely to occur with the blood group constellation mother 0/child A. The concentration of anti AB possibly plays a role. ABH(0) system and transplantation

The significance of the AB0 system in transplantation of organs varies. Hyperacute rejection of kidney and heart transplants is possible, depending on the antibody titers /2324/. In adults and older children at least, AB0-incompatible liver transplantation has a significantly poorer prognosis and is therefore limited to emergency situations /25/. In children (below the age of nine years), AB0 incompatibility appears to be of limited significance for heart and liver transplantation /26/. Although AB0 incompatible allogeneic bone marrow transplantation is possible, AB0 incompatibility represents an adverse risk factor /27/.

27.3.2 I system

Structure and antigens

The antigens I and i are not antithetic since i represents the biosynthetic precursor of I and both antigens generally occur simultaneously at a different level of expression. The essential difference between these antigens is the presence of branched carbohydrate chains. The i antigen still has unbranched carbohydrate chains. Whereas i dominates on the cells of fetuses and neonates, I is predominantly expressed on adult cells at individually very different levels. In adults, the i phenotype occurs at an estimated frequency of only 1 : 10,000. Antibodies


Antibodies with anti-I specificity are found in low concentrations (maximum titer at 0 °C = 64) in almost all individuals after infancy in the form of natural cold autoantibodies (IgM) with a narrow thermal optimum; they are of no clinical significance. Very commonly, the antibodies are simultaneously directed against other antigens of the carbohydrate blood groups (anti-IH, IA, IB, IP1), in which case they are best detected by erythrocytes that carry both antigens.

In cold autoimmune hemolysis (cold agglutinin syndrome), abnormally high concentrations of cold autoantibodies are detectable that show mostly anti-I specificity. In this setting, anti-I is capable of activating complement to a clinically relevant extent because of its wide thermal amplitude of up to > 30 °C (cold hemolysins).


Antibodies of this specificity are detected much more rarely and are always autoantibodies (usually cold IgM antibodies). Anti-i is found more frequently in, for example, infectious mononucleosis, alcohol induced liver cirrhosis, reticulosis, and myeloid leukemia.


The antibodies of the Ii system usually react most strongly at low temperatures; their optimal reactivity is at 0–4 °C in the saline test. The typical findings in cold autoimmune hemolysis include the detection of agglutinating antibodies at temperatures greater than 30 °C, in vitro hemolysis of erythrocytes at room temperature, and a positive direct antiglobulin test with anticomplement serum.

Clinical significance

Antigens: immature erythrocytes show more pronounced i antigenicity and less expression of I. This occurs not only physiologically in newborns but also in patients with hemolytic anemias and other hematological diseases associated with increased erythropoiesis.

Antibodies: natural cold antibodies may at times interfere significantly with the immunohematological examinations required for the preparation of blood transfusions. However, these cold autoantibodies are not clinically relevant unless they present in highly abnormal concentrations as seen in the cold agglutinin syndrome or cold autoimmune hemolysis (Section – Mixed autoimmune hemolysis).

27.3.3 Lewis system

Structure and antigens

The Lewis (Lea, Leb) system is characterized by the following peculiarities:

  • It includes soluble, non antithetical antigens on glycosphingolipids that are secondarily adsorbed to erythrocytes from the plasma. The secretion of Lea substance does not commence until two weeks to six months after birth; in the case of Leb it starts even later. In general, therefore, fetal erythrocytes do not yet possess these Lewis antigens. During the first year of life, up to 80% of children become Lea positive.
  • The genetically determined phenotype is not detectable until the second year of life. During erythrocyte storage, Lewis antigens also tend to dissolve again; this explains why they are poorly detectable on erythrocytes after prolonged storage.
  • Three independent genes determine the phenotype as an indirect gene product (Le, Se, H). Le encodes the fucosyltransferases that bind the fucose molecules to the type 1 carbohydrate chains. In non secretors (sese), only one fucose molecule is bound to each carbohydrate chain, thus yielding Lea (Le(a+b–)). Only if the H gene is present, a second fucose molecule is bound in secretors (Se), thus leading to the formation of Leb (phenotype Le(a–b+)). The absence of the Le gene (lele) always results in the phenotype Le(a–b–).
  • The ABO phenotype has an impact on the expression of the Lewis antigens since both systems arise from mutual precursor molecules. Therefore, in blood group A1 the expression of Lea and Leb may be attenuated.

The different genes and the resultant phenotypes of the Lewis antigens are listed in Tab. 27-10 – Lewis-System: genes and phenotypes. Lea and Leb are the important antigens while Lec and Led are only of theoretical interest. Within the entire Lewis system, up to seven different fucosyltransferases may be involved that are regulated by different gene loci. For practical reasons, Tab. 27-10 – Lewis-System: genes and phenotypes therefore shows only a simplified version of this complex system. Antibodies

Lewis antibodies are usually naturally occurring cold antibodies that usually belong to the IgM class (often with an IgG portion); they are capable of activating complement. They are only of clinical relevance (hemolytic transfusion reactions) if they still react at temperatures above 30 °C, if they are detectable in the indirect antiglobulin test, and/or if they are capable of in vitro hemolysis as well.

Because the antigens are not expressed on fetal erythrocytes, even IgG Lewis antibodies are not capable of causing hemolytic disease of the newborn.

Anti-Lea: these are the most important and most common Lewis antibodies. Antibody carriers possess the Le(a–b–) phenotype.

Anti-Leb: these are rarely of clinical relevance. Antibody carriers are mainly Le(a–b–) and rarely Le(a+b–). Therefore, anti-Leb occasionally occur simultaneously with anti-Lea. The reactivity of anti-Leb is also determined by the AB0 blood group. According to their optimum reactivity, anti-LebH and anti-LebL specificities are distinguishable:

  • Anti-LebH are found almost exclusively in individuals with blood group A1 or A1B; these antibodies react most with Leb cells of blood group 0. They can already be neutralized by H substance [saliva of Le(a–b–), ABH secretors].
  • Anti-LebL occur much more rarely and do not show a preference for a certain AB0 blood group. They react with all AB0 blood groups to the same extent.

Anti-Lec, anti-Led: these antibodies are not clinically relevant because they are so rare and because they are cold antibodies. They react with erythrocytes of Le(a–b–) phenotype.


Lewis antibodies react preferentially in the enzyme test at 4 °C and in the gel centrifugation test at room temperature. They can be neutralized by LeaLeb substance. Since they activate complement, they can be detected more easily using fresh serum samples. Due to the storage instability of the antigens, the erythrocytes should not be too old and dissolved antigens should be removed prior to the test by washing in saline.

Clinical significance

Men with the Le(a–b–) phenotype are thought to have a higher risk of coronary heart disease. Sjögren’s syndrome also occurs more frequently in individuals with this phenotype.

The relevance of the Lewis system in transplant recipients is controversial. Like HLA antibodies, cytotoxic Lewis antibodies are thought to significantly shorten graft survival (kidney) /28/.

In general, the blood transfusion related implications of the Lewis system tend to be overestimated.

27.3.4 P system

Structure and antigens

The antigens of the P system are based on carbohydrate chains with galactose and N-acetyl galactosamine terminally bound to lactosylceramide.

A distinction is made between the antigens P, Pk, P1, and p as well as between the phenotypes P1, P2, P1k, P2k, and p; at least two different genes encode each of these antigens and phenotypes (Tab. 27-11 – P system).

P1 is the most common phenotype. On fetal erythrocytes, the development of antigen P1 exists but is still very weak. In adults, the genetically determined expression of this antigen varies (P1s = P1 strong, P1, P1w = P1 weak); it is one of the most labile antigens in terms of storage. Antibodies


Using very sensitive techniques, naturally occurring cold antibodies of the anti-P1 specificity are detectable in almost all persons with the P2 phenotype (almost exclusively IgM without complement binding). Very rarely, these antibodies still react at 37 °C (also IgM with complement binding) and can cause hemolytic transfusion reactions. As far as blood transfusions are concerned, anti-P1 antibodies need only be taken into consideration if they are detectable in the indirect antiglobulin test. Anti-P1 antibodies do not cause hemolytic disease of the newborn because these antibodies are not capable of placenta transfer.

Anti-P and anti-Tja (anti-P, anti-P1, anti-Pk)

These antibodies are extremely rare since very few individuals do not have these antigens themselves (Tab. 27-11 – P system). In these rare cases, allo sensitizations do, however, occur during early childhood because these antigens are ubiquitous. Although the antibodies are cold, naturally occurring allo antibodies (IgM), they display pronounced hemolytic activity because of their wide thermal amplitude and their ability to activate complement. Furthermore, as IgG antibodies, they cause hemolytic disease of the newborn as well as abortions. Anti-P antibodies also occur as bithermal autohemolysins in paroxysmal cold hemoglobinuria (refer to Section – Postnatal diagnostic tests).


Anti-P1: these antibodies react preferentially in the enzyme test at 4–22 °C and can be neutralized by P1 substance. Their detection in the indirect antiglobulin test is of some clinical relevance. However, their detectability depends heavily on antigen expression and the quality of the test cells. Neutralization with P1 substance is advisable both for confirming the presence of the antibodies and for eliminating their interfering effect during the detection and/or exclusion of other more relevant antibodies.

Anti-P, anti-Tja: these antibodies react most strongly in the enzyme test and are characterized by a wide thermal amplitude; they produce positive results in the indirect antiglobulin test with antiglobulin sera containing complement. Sometimes, they can also be detected using the hemolysis test, especially when enzyme-treated erythrocytes are used (Section – Bithermal cold hemolysins of the Donath-Landsteiner type regarding bithermal cold autohemolysins).

Clinical significance

P-like structures are located on certain cancer cells and on various microorganisms (e.g., E. coli). Furthermore, E. coli and its toxins are more easily bound by P1 and P2 antigens, thus explaining why individuals with P1 are more commonly affected by urinary tract infections. In addition, the P antigen has been described as a receptor for Parvovirus B19.

The blood transfusion related implications are relatively minor. The common antibodies to P1 are in most cases clinically irrelevant; on the other hand, because of the relatively high antigen frequency and reactivity of P1, they do interfere with the AB0 determination (reverse typing) and possibly with the antibody screening and cross match. In contrast to this, hemolytically active antibodies (anti-P, anti-Tja) are extremely rare, either as allo antibodies or as autoantibodies.

27.4 Blood group systems with protein antigens

Most of the blood group systems on erythrocytes are based on protein polymorphism and these proteins are components of structural proteins. In contrast to the carbohydrate blood group systems, they are primary gene products.

27.4.1 Rh system

Structure and antigens

The most important protein based blood group system is the Rhesus (Rh) system. Two genes are distinguishable at the RH locus (chromosome 1); they regulate the expression of the five most important antigens D, C, c, E, and e /4/:

  • The RHD gene, which encodes the RhD protein
  • The RHCE gene, which is responsible for the synthesis of the RhCE protein.
  • The C, c, E, and e antigens are located on the RhCE protein.

Rh molecules are transmembrane proteins composed of 416 amino acids that are situated in a meander conformation within the membrane, which they cross 12 times (Fig. 27-2 – Rh transmembrane protein and assignment to the exons of the RH genes).

  • The RhD protein differs from the RhCE protein by approximately 31–35 amino acid substitutions. With a few exceptions, RhD negative Europeans lack the RHD gene and therefore also lack the RhD protein (Rh(D) factor).
  • A difference in four amino acids is responsible for the C specific and c specific proteins. However, the C/c specificity seems to be determined by only one amino acid (position 103: serine for C, proline for c).
  • The E/e polymorphism can be traced back to one amino acid substitution (position 226: proline for E, alanine for e).

The formation of epitopes in the Rh system depends on the conformation and steric composition of the individual protein loops that are expressed on the erythrocyte membrane. For instance, the cysteine residue at position 285 plays an important role in the binding of antibodies /29/.

Besides these five major antigens, more than 40 other antigens exist that are determined by position effects of the RH genes (cis and trans effects) as well as by suppressor genes, further Rh sub loci (satellite antigens), and the effect of gene complexes (combined antigens).

Three nomenclatures are available for these antigens; each nomenclature takes different aspects into consideration and offers certain advantages, depending on its field of application (Tab. 27-12 – Antigens of the Rh system). The Fisher-Race nomenclature has proven its worth for routine laboratory purposes in Germany.

The presence of the five most important Rh blood group antigens C, c, D, E, and e is generally summarized as the Rh genotype and documented by the serologically determined phenotype. Without family testing, the actual haplotypes (genotypes) can only be presumed at a certain level of probability (Tab. 27-13 – Frequency of Rh phenotypes and genotypes). Rh factor D, weak D, partial D, and DEL


After A and B, the D antigen displays the highest immunogenicity (around 20 times higher than the other major Rh blood group antigens) because the antigen expression D-negative (d) does not imply that the antigen is present in another, polymorphic form. It is completely absent. Following the transfusion of D-positive packed red blood cells, around 50% of D-negative recipients develop antibodies with anti-D specificity. Due to the presence of erythrocytes, immunization to D is possible even following the transfusion of platelet concentrates, but not of frozen plasma.

For the individual major Rh antigens C, c, E, and e alone, the immunization rate is < 1%. In this regard, no differences are noted between Rh(D)-positive and Rh(D)-negative recipients.

Antigen dose

Ten thousand to more than 30,000 D antigens are located on one red blood cell. The number of antigens, and hence the immunogenicity and antigenicity, depend on the gene dose (homozygous or heterozygous) and the Rh phenotype (Rh genotype). However, differences in antigen doses cannot be determined using routine serological methods since the normal deviation range overlaps too much in homozygous and heterozygous individuals. Thus, conclusions cannot be drawn regarding the genotype.


The complete D antigen is composed of at least nine epitopes (Tab. 27-14 – D epitopes on D variant cells). Using monoclonal antibodies, up to 30 subepitopes can be defined. Although the reactivity of monoclonal anti-D reagents with particular epitopes is important in terms of their characterization and licensing, the classification of epitopes does not have a role in routine laboratory investigation. Weak D

In less than 1% of cases, significantly fewer D antigens are expressed on the cells (D variants). Many of these weakly reacting D variants were formerly referred to as Du and, according to the degree of attenuation, were arbitrarily subdivided into low and high grade Du. In most cases, a positional effect of C on the gene locus of the other haplotype (trans effect) is responsible as the genotype in CDe/Cde for the reduced expression of D antigens. If suitable monoclonal anti-D reagents are used, most of the antigenic determinants that were formerly determined as Du should no longer be distinguished from normal D. The term Du is of historical significance only.

Nowadays, the most clinically important classification of D variants is into weak D and partial D.

With few exceptions, individuals with weak D express all D epitopes. However, binding of monoclonal antibodies may be very weak and differences in the coding sequence between normal RHD and weak D exist /4/. To date, alloimmunization to erythrocytes that express normal D has not been observed in individuals with the most common weak D types 1, 2, and 3. Therefore, it is unnecessary to refer to individuals with weak D as Rh-negative and transfuse them with Rh(D)-negative blood.

With respect to the other rare weak D types, however, women in prenatal care should be considered as an exception. Women with weak D type are at risk of developing anti-D antibodies against Rh(D)-positive fetuses. Rh immune globulin (RhIG) prophylaxis is therefore recommended for these women in the same way as for RH(D)-negative women.

Weak D type 2 is associated with a very low number of expressed antigens (approximately 500 per erythrocyte). It is relatively common, accounting for 18% of all individuals with weak D. Because weak D type 2 can also be associated with the production of anti-D in Rh(D)-negative recipients, anti-D reagents (for donor investigations in particular) should be able to detect this.

If anti-D reagents are used that simultaneously contain IgG and IgM antibodies, blocking antibodies may mimic an attenuation of D.

To avoid confusion with the specific antigen referred to as Dw, Dw is no longer used to refer to weak D. Partial D

In contrast to weak D, the less common partial D phenotypes (also mosaic D) are characterized by the absence of individual or multiple epitopes of the D antigen (Tab. 27-14 – D epitopes on D variant cells). The number of expressed antigens may be reduced or increased.

By using monoclonal antisera of the IgG class, six different D categories (II–VII) can be distinguished according to the number of antigens and epitopes on the erythrocytes. In addition to the D categories, rarer partial D variants also exist.

Of these, D category VI is the most common (0.02–0.03%) and most clinically relevant. Since D VI has the lowest number of D epitopes, individuals with this D variant are more likely to become immunized to D (or more precisely, to the missing epitopes of D) than individuals with other D variants; at the same time, they are significantly less likely than D-negative individuals to become immunized. The synthesized allo antibodies are not distinguishable from anti-D, apart from the fact that they do not react with the erythrocytes of the immunized individual. Partial D variants can always be presumed if D-positive individuals have allo antibodies to D. Recipients with DVI are classified as Rh(D) negative and blood donors with this variant are classified as Rh(D) positive, although the immunogenicity of weak D variants is low in comparison to D. DEL

The designation Rh(D) negative is based on the reaction of erythrocytes with anti-D in various agglutination methods including the antiglobulin test. However, these methods may not detect D variants that express very little D antigen; these variants can only be detected using anti-D adsorption/elution tests. These phenotypes, which are relatively common among Asians, are referred to collectively as DEL variants /30/. Although the immunogenicity of this very weak D variant is low, isolated cases of immunization have been described /3132/. The DEL variant is very rarely observed in Europe and donors with this variant are considered as Rh(D) negative. However, this is a controversial point of view. Many blood donation services therefore perform molecular genetic examinations on Rh(D)-negative donors and consider DEL packed red cells to be Rh(D) positive. Other Rh antigens

In the same way as described for D, quantitative and qualitative deviations are also found with regard to other major antigens of the Rh system; these deviations are caused by positional effects or corresponding gene mutations (Cw, Cu, Eu, Ew etc.). With the exception of Cw, these variants are of no relevance as far as blood transfusions are concerned.

Very limited importance is ascribed to the compound antigens cE, Ce, ce, and CE that are encoded by the corresponding gene complexes of a haplotype. The corresponding antibodies react only with cells that express both antigens as a complex. In contrast, antibodies to the G antigen react not only with cells that express C and D simultaneously but also with cells that express C or D in isolation. Rh “deletion” phenotypes

Very rarely, various deletion phenotypes occur as a result of silent, functionally inactive, or minimally active genes (-D-, •D• , ---/---). The deletion type ---/--- (Rhnull) can also be caused by a suppressor or regulator gene that is independent of the Rh locus. Individuals with Rhnull have hemolytic anemia with morphologically atypical erythrocyte forms (stomacytosis).

In Rhnull, the antigen LW (Landsteiner-Wiener) is missing. This antigen is encoded independently from the Rh system but it requires Rh antigens for its expression on erythrocytes /34/. LW antigens are more abundant on D-positive cells than on D-negative cells. They are also expressed more strongly on the erythrocytes of neonates.

27.4.2 Rh antibodies

Rh antibodies are mostly warm immune antibodies of the IgG type, mainly IgG1 and IgG3. They generally do not activate the complement system, probably because the antigens are located too far apart and only one IgG molecule is bound per antigen. Differences in the hemolytic activity of the antibodies cannot be detected by simple serological methods. Therefore, in the case of erythrocyte transfusions, Rh antibodies are generally taken into account by using antigen negative cells. The rare natural Rh antibodies are mainly IgG but most show cold reactivity; they are clinically irrelevant. Anti-D

Anti-D remains the most common and most important Rh antibody. Immunization can be traced back to pregnancies prior to the era of Rh immune globulin prophylaxis, to failure or omission of Rh immune globulin prophylaxis, and to the transfusion of Rh(D)-positive RBCs. Anti-D is very occasionally found in combination with D variants and shows the same hemolytic activity as in Rh-negative individuals. In almost one third of cases, anti-D occurs in combination with anti-C; more rarely (approximately 2%), it occurs with anti-E or with both. In addition, autoantibodies sometimes display anti-D specificity. Anti-E

Anti-E is the second most common Rh antibody after anti-D. Anti-E reactivity is almost always characterized by a marked dosage effect. If the phenotype CCDee is present, latent immunization to c is possible; this constellation (CCDee + anti-E) should prompt transfusions of Rh-matched blood components. Anti-E is found more commonly than other Rh antibodies without a preceding transfusion or pregnancy. Anti-c, anti-c anti-E, anti-cE, anti-ce

From a clinical point of view, anti-c is the second most important Rh antibody after anti-D. Anti-c is relatively often responsible for delayed hemolytic transfusion reactions. After anti-D and anti-A, anti-c is the third most frequent cause of hemolytic disease of the newborn.

Following immunization, the concentration of anti-c usually declines again rapidly. Furthermore, the frequently observed dosage effect interferes with the detectability of the antibody. Finally, if anti-c is present in patients who require transfusion, supply problems develop very quickly because of the relatively high antigen frequency (80%).

If the individual carrying the antibody has the phenotype CCDee, anti-c is relatively often accompanied by anti-E. In addition, these mixed antibodies sometimes mask anti-cE or anti-ce that are directed against the compound antigens cE and/or ce. If this Rh phenotype is present, especially in girls or women of childbearing age as well as in patients with chronic intermittent transfusion requirements, RBC units with an identical or “selected”* Rh genotype should therefore be used. If anti-c has already been detected, this approach is strongly recommended.

* Note: “Selected” Rh genotype means that the transfusion does not contain any Rh blood group antigens that the recipient does not already have. An Rh-genotype-compatible blood transfusion can contain any Rh antigens, provided Rh antibodies are not present. Anti-C, anti-Cw

Anti-C seldom occurs in isolation. As far as immunization is concerned, it does not matter whether the immunizing erythrocytes are simultaneously D positive or D negative. In most cases, sera containing anti-C also contain anti-Cw.

If the individual carrying the antibody has the Rh blood group antigens ccDEE, it is prudent to assume that immunization to e has also occurred simultaneously and to use blood units from a donor with an identical Rh genotype. The antibody is often not detectable but a delayed transfusion reaction may develop following a booster effect. On the other hand, anti-C is detected mostly in combination with anti-D and anti-G.

Isolated anti-Cw is produced by individuals who are Cw negative or who themselves possess CC, Cc, or cc. An isolated anti-Cw is also possible without preceding transfusions or pregnancies. Anti-e, anti-C anti-e, anti-Ce

As an allo antibody, anti-e is thankfully very rare. On the other hand, due to the high antigen frequency (98%) and low immunogenicity, this antibody is associated with significant supply problems. If the phenotype ccDEE is present, anti-C, anti-ce, and anti-Ce should be taken into consideration as concomitant antibodies and blood units from a donor with an identical Rh genotype should be used. In spite of the detectability of the e antigen, its polymorphism means that anti-e can occasionally be found as an allo antibody.

Anti-e is more commonly found as a warm autoantibody. However, it rarely displays a clear specificity. In most cases, the warm autoantibodies react only weakly with E cells and in particular with EE cells. Furthermore, they can be adsorbed by EE cells. Anti-G

As a concomitant antibody, anti-G is often masked by the presence of mixed antibodies to C and D. In contrast to the antibodies directed against compound antigens, anti-G reacts not only with cells that express C and D (as regulated by different haplotypes) such as cDE/Cde, but also with those that carry only C or D on their surface (e.g., Ccdee and ccDee).

Anti-G differs from mixed antibodies in the fact that it can be adsorbed equally well to cells that express only one of the two antigens. This explains why in the presence of anti-G, a mother may seem to have anti-C and anti-D although the father of the child and the child itself do not possess C. If anti-G is detected without anti-D during prenatal care, Rh immune globulin prophylaxis is recommended /4/. Anti-LW (Landsteiner-Wiener)

Anti-LW occurs very rarely as an allo antibody. More commonly, it is a benign autoantibody that not infrequently precedes and/or occurs in parallel to an immunization to D. Anti-LW is not an Rh antibody, although it acts like one. Anti-LW does not react with Rhnull cells but shows the strongest reaction with Rh(D)-positive cells. Weak anti-LW therefore acts like anti-D while strong anti-LW antibodies react like autoantibodies with a preference for D-positive cells. However, anti-LW can be completely adsorbed by Rh(D)-negative cells. Determination of Rh antibodies

Rh antibodies react like incomplete antibodies. They bind particularly easily in the presence of low ion concentrations (low ionic strength solution, LISS) or if enzymes (e.g. papain, bromelin, or ficin) or enzyme-treated test erythrocytes are used. Therefore, enzyme tests and the indirect antiglobulin test (LISS technique) are the most sensitive methods for detecting these antibodies.

If conventional tube methods are used 10–15% of the Rh antibodies are detected only in the enzyme test. In 80% of cases, these antibodies are anti-E. Overall, because of their limited clinical relevance, it is debatable whether it is necessary to detect such antibodies (that react only after enzyme treatment of erythrocytes) by including enzyme in the antibody screening test. Approximately half of these antibodies that react only with enzymes are identifiable as IgG antibodies by more sensitive methods such as gel centrifugation and solid phase technology; thus, if such methods are used, the performance of additional enzyme tests is not warranted.

If, however, conventional manual tests are performed, the inclusion of enzyme tests at 37 °C may be useful for antibody screening in at least certain groups of patients (e.g., pregnant women, recipients of multiple blood transfusions, transfusion reactions) to identify primary immunization as early as possible. Clinical significance of Rh antibodies

The blood transfusion related implications of the Rh system are essentially determined by the immune antibodies and their hemolytic activity in acute and delayed hemolytic transfusion reactions as well as in hemolytic disease of the newborn.

27.4.3 Kell system

Structure and antigens

The structure of the Kell system is determined by two gene loci on different chromosomes. A smaller glycoprotein (37 kDa) is determined by a gene (Xk) located on the X chromosome; this glycoprotein basically represents the primary substance and is referred to as Kx /35/.

If Kx is not expressed (McLeod phenotype, extremely rare), the production of Kell antigens is very weak. Their expression is regulated by a specific Kell gene on chromosome 7 (19 exons). These antigens and the high-frequency para-Kell antigens K12, 13, 14, 18, 19, and 22 are located on a glycoprotein composed of 732 amino acids (93 kDa). Twenty of these amino acids are found in the transmembrane region while 46 are found intracellularly and the majority are found on the outer surface. Twenty-four antigens of mostly high and low frequency are regulated by this gene, of which 4–5 are considered to be antithetically related (Tab. 27-15 – The most important antigens of the Kell system).

The two most important antigens K and k differ by the substitution of one amino acid (i.e., Thr (k) for Met (K) at position 193).

Depending on homozygosity or heterozygosity, 3,000–6,000 antigenic binding sites are found per erythrocyte for each specificity. If the specific Kell gene (K0 phenotype, extremely rare) is absent, no Kell antigens are expressed although more of the primary substance Kx is detectable. Since the antigenic structure is essentially determined by the steric composition of the epitopes (with disulfide bonds playing a significant role), it is possible to destroy Kell antigens by using dithiothreitol, and thus to artificially produce K0 cells for experimental purposes. However, this also results in the destruction of other high-frequency antigens, especially Yta and LW.

As in the Rh system, the Kell system also displays positional effects (cis effects), between Kpa and k and/or Jsa (attenuation of k and/or Jsb).

The Kell antigen (K, KEL1) is highly immunogenic. Up to 10% of K-negative individuals synthesize anti-K following the transfusion of K-positive blood. In Germany, it is recommended that transfusions with K-positive blood be avoided if at all possible in K-negative girls and women of childbearing age. Kell antibodies

The antibodies of the Kell system are mainly immune antibodies, almost exclusively belong to class IgG1, and may frequently activate complement (but only up to C3b). Thus, in most cases, they are hemolytically active and may cause hemolytic transfusion reactions as well as hemolytic disease of the newborn. Anti-K

Anti-K is the most common irregular erythrocyte antibody outside the Rh system (Tab. 27-6 – Frequency of irregular erythrocyte antibodies). Diagnostic problems are rare. Because of the low antigen frequency, supply problems do not occur. Anti-K is only rarely the cause of hemolytic disease of the newborn, which often runs a severe course and is characterized by suppression of erythropoiesis /4/. On rare occasions, anti-K is temporarily detectable as a natural cold antibody (IgM). Infections with different microorganisms such as E. coli, M. tuberculosis, Streptococcus sp. or Campylobacter sp. are thought to be responsible for immunization.

Anti-K occasionally occurs as an autoantibody as well. Misinterpretations are possible if the expression of the antigen is markedly reduced at the same time. Anti-k, anti-Kpb, anti-Jsb

These antibodies are very rare since very few individuals are subject to immunization due to the antigen distribution in the population. The presence of these antibodies creates major blood transfusion related problems. First, they are difficult to identify, in particular because they react with nearly all test cells due to the high frequency of their corresponding antigens and can mask other antibodies. Second, the high antigen frequency also results in supply problems.

As part of diagnostic evaluation, antigen testing of the patient’s erythrocytes is the test most likely to be useful, provided corresponding test sera are available. Because the antibodies may in principle be hemolytically active, they need to be taken into consideration for blood transfusions.

The tremendous supply problems in these cases can be successfully managed only by cooperation at national and international level including the use of cryo-preserved packed RBCs from special blood banks, the use of blood units donated by family members, and the use of autologous hemotherapy. Anti-Kpa, anti-Jsa

In terms of characteristics and pathogenicity, these antibodies possess the same general properties as the other antibodies in the Kell system. Because of the low antigen frequency in Caucasians, the antibodies are extremely rare and are not associated with major blood-transfusion-related problems. Anti-KEL 11, 14, 18, 19, 22

These antibodies are also extremely rare due to the corresponding antigen frequency of > 99.9%. The same information applies as for anti-k. However, these antibodies do not always display hemolytic activity.


Antibodies of the Kell system usually do not display dosage effects. If tube tests are used, they are best detected in the indirect antiglobulin test using the albumin technique. LISS techniques sometimes detect anti-K poorly; however, this is no longer a problem since the introduction of gel centrifugation tests and solid phase technology. In the enzyme test, antibodies of the Kell system usually react only if they are natural antibodies (IgM).

Clinical significance

Erythrocytes of the McLeod phenotype have a shortened life span and present as acanthocytes. In addition, this phenotype commonly occurs in combination with chronic granulomatous diseases because the responsible genes are located close to one another on the X chromosome.

27.4.4 Duffy system

Structure and antigens

Two codominant alleles on chromosome 1 determine the two most important antigens Fya and Fyb (Tab. 27-16 – The most important Duffy antigens in Caucasians). Structurally, these antithetically related antigens are transmembrane glycoproteins composed of 338 amino acids; Fya and Fyb differ by only one amino acid substitution at position 42 (i.e., glycine versus asparagine). Approximately 17,000 Fya and Fyb antigens are found per erythrocyte. The immunogenicity of Fya is significantly higher than that of Fyb and is similar to that of the Rh antigens c and E (approximately 1% in the case of transfusions). Both antigens are very sensitive to proteolytic enzymes (with the exception of trypsin) and are therefore labile during storage, especially at a low pH.

A third allele, Fy, is rarely present in Caucasians but occurs in 68% of Black individuals; it encodes the phenotype Fy(a–b–).

A fourth allele (Fyx) regulates a very weak Fyb; for all practical purposes, the latter antigen is of no significance because it is only detectable by antibodies at very high titers or by molecular genetic examination methods. Antibodies

The antibodies of the Duffy system are mostly immune antibodies of the IgG class (IgG1) and can activate complement in about half of cases. Various Duffy antibodies with different specificities may cause severe acute and delayed hemolytic transfusion reactions; very rarely, they are also the cause of hemolytic disease of the newborn. Anti-Fya

This is the most common antibody in the Duffy system. It occurs mainly in conjunction with the phenotype Fy(a–b+) and rarely with Fy(a–b–). Occasionally, anti-Fya is found as a natural antibody (IgM, partial dosage effects). Anti-Fyb

This occurs much more rarely than anti-Fya and is often associated with other antibodies. Individuals carrying this antibody have the Fy(a+b–) phenotype. Anti-Fy3, anti-Fy4, anti-Fy5

These antibodies are of no clinical significance because of their rarity. They are directed against enzyme-resistant antigens. Anti-Fy3 and anti-Fy5 reacts equally well with all Fya and Fyb-positive erythrocytes and are produced by individuals who are themselves Fy(a–b–). Anti-Fy3 react with cord blood different in whites and blacks, and unlike anti-Fy5, does not display attenuation with respect to Rhnull cells. Anti-Fy4 reacts with Fy(a–b–) and most Fy(a+b–) as well as Fy(a–b+).


The best way to detect these antibodies is by using the direct antiglobulin test (albumin and LISS techniques), which is hardly associated with any dosage effects. The enzyme test, especially if performed as a two stage test (enzyme treated erythrocytes), is not suitable for detecting anti-Fya and Fyb directly, but it may be useful for differentiating mixtures of antibodies. Most Duffy antibodies are relatively difficult to elute and/or poorly detectable in the eluate.

Clinical significance

The Duffy antigens Fya and Fyb represent binding sites for Plasmodium knowlesi and Plasmodium vivax, thus explaining why the absence of these antigenic determinants [Fy(a–b–)] is associated with resistance to these pathogens.

27.4.5 Kidd system

Structure and antigens

Two codominant alleles (Jka, Jkb) and a silent gene (Jk) on chromosome 18 determine three antigens (Jka, Jkb, Jk3) and four phenotypes (Tab. 27-17 – Kidd antigens). The phenotype Jk(a–b–) is extremely rare. The Kidd polymorphism is associated with an Asp280Asn amino acid polymorphism in the Kidd glycoprotein.

The HUT11 gene encodes the urea transporter of erythrocytes, a polypeptide chain composed of 389 amino acids with ten transmembrane domains. The HUT11 gene product, also known as the HUT11 protein, is identical to the Kidd glycoprotein /36/. The number of antigens is reported to be 14,000–18,000 per erythrocyte. The immunogenicity is lower than that of Fya. Jk-like structures also occur on microorganisms such as P. mirabilis and S. faecium. Antibodies

Anti-Jka in combination with phenotype Jk(a–b+) occurs significantly more frequently than anti-Jkb in the presence of Jk(a+b–). Kidd antibodies are mostly immune antibodies (IgG3) that are capable of activating complement. They can hemolyze enzyme treated erythrocytes in vitro. Sometimes, a marked discrepancy exists between the detectability of these antibodies in vitro and the severity of the hemolytic reaction in vivo. Detection of the antibodies is difficult because they:

  • Display pronounced dosage effects
  • Often occur in combination with other antibodies (antibody mixtures)
  • Often are, due to their low concentration, only detectable indirectly via the binding of complement to the erythrocyte surface or using solid phase technology
  • Usually are detectable for only a few days to weeks after their occurrence /16/.

They are the cause of about half of the cases of delayed hemolytic reactions. Occasionally, they cause hemolytic disease of the newborn. Anti-Jka in particular may also occur as an autoantibody with different degrees of hemolytic activity or as a drug-induced antibody (e.g., due to α-methyldopa or chlorpropamide). Natural Kidd antibodies (IgM) are also found rarely.


Antibodies are determined using the indirect antiglobulin test (LISS, enzyme, gel centrifugation, and solid phase technology). Solid phase technology is the most sensitive detection method. It is used mainly to evaluate transfusion reactions and in cases where the direct antiglobulin test is positive following transfusion. Because of the pronounced dosage effect and instability of the antigens during storage, the use of homozygous, relatively fresh erythrocytes is recommended.

Clinical significance

The significance of the Kidd system is based on the antibodies. The Kidd antigens are located on the same protein that is responsible for the urea transport functions of erythrocytes.

27.4.6 MNS system

Primary structure

The major polymorphism of this system is located on two transmembrane glycoproteins that are rich in sialic acid (glycophorins A [GPA] and B [GPB]) and form part of the erythrocyte membrane. The number of glycoprotein molecules per cell varies: for GPA (131 amino acids, of which only 23 are within the membrane), it is approximately 1,000,000 and for GPB (72 amino acids), it is approximately 250,000.

MN antigens

The two codominant antigens M and N are located on GPA and differ by only two amino acid substitutions at position 1 (serine leucine) and position 5 (glycine glutamic acid). In addition, neuraminic acid molecules are bound at positions 2, 3, and 4; these are essential components of the epitopes of many anti-M and anti-N antibodies as well as being the reason for the sensitivity of the antigens to sialidases.

GPB displays the same sequence for the first five terminal amino acids as GPA in the expression of N. This antigen on GPB is therefore referred to as ’N.’ Thus, anti-N can also react with homozygous M cells. However, since significantly fewer GPB molecules are present on the cells, the reaction is much weaker and can be prevented by adjusting the test reagents and/or the test procedure.

The absence of GPA is associated with the very rare phenotype En(a–) M–N–’N’+.

Alternatively, En(a–) cells may display N–’N’– in combination with a weak M+ antigenic determinant, as a result of simultaneous changes in GPB. Because of this, in the presence of En(a–), different antibodies with a high antigen frequency can be found (e.g., anti-Ena and anti-Wrb).

Ss antigens

The two codominant antigens S and s are determined by the alleles of the GYPB gene, which is located close to the MN gene GYPA. These antigens are located on GPB and differ by only one amino acid at position 29 (methionine threonine). S is often sensitive to enzymes while s is not.

U antigen

U is encoded as another antigen by the same gene locus as Ss. It is located on GPB and is a prerequisite for the synthesis of S and/or s. The antigen determinant is absent only in Black individuals (0.2%); in this case, the phenotype is S–s–.


The MNS system includes numerous, mainly rare variants whose antigens are also mainly located on GPB. Due to their lack of clinical relevance, no further details are mentioned here.


The most important antigens and phenotypes of the MNS system are listed in Tab. 27-18 – Antigens and phenotypes in the MNS system.


The Gerbich system, with its high and low-frequency antigens, is located on the glycophorins GPC and GPD; it is also regulated by another gene locus. Antibodies

The various antibodies of the MNS system show relatively few similarities in their serological reactivity except for their marked dosage effects and their general inability to activate complement.

In ascending order, the probability of these antibodies being clinically relevant, warm immune antibodies (IgG) is as follows: anti-N, anti-M, anti-S, anti-s. The probability of these antibodies being cold, naturally occurring IgM antibodies that cause neither hemolytic transfusion reactions nor hemolytic disease of the newborn increases in the reverse order (i.e., most frequently anti-N).

The very rare antibodies to the high incidence antigens Ena, Wrb, and U are mainly immune antibodies that may cause hemolytic transfusion reactions and hemolytic disease of the newborn. In addition, they also occur as hemolytically active autoantibodies. Anti-M, anti-N

Anti-M is the most commonly found antibody of the MNS system. It occurs mainly as a natural cold antibody; in children, it occurs more commonly in conjunction with bacterial infections. 50–80% of the sera contain IgG antibodies in addition to IgM antibodies. Anti-M is a rare cause of hemolytic transfusion reactions or hemolytic disease of the newborn.

Occasionally, anti-M can occur as an alloantibody in antibody carrying individuals despite the detectability of M. In such cases, the antibody is directed against an epitope of the M antigen (MA) that is absent in the antibody-carrying individual. Autoantibodies of this specificity also occur rarely. Anti-M is clinically relevant only if it is currently detectable and if it shows reactivity above 30 °C and/or positivity in the indirect antiglobulin test.

Anti-N occurs much more rarely and almost always displays cold reactivity (IgM). Hemolytically active anti-N is found almost exclusively in conjunction with the phenotype S–, s–. Like anti-M, it can rarely occur as an alloantibody (anti-NA) in N-positive individuals. Anti-S, anti-s, anti-U

The much less frequent anti-S, but also anti-s and anti-U, are mostly immune antibodies (IgG) that have significantly greater clinical relevance. Anti-Gerbich

Anti-Ge are mainly IgG antibodies and may cause hemolytic transfusion reactions but not hemolytic disease of the newborn. Sporadically, anti-Ge also occur as warm autoantibodies.


Anti-M and anti-N are best detected in the indirect antiglobulin test using the gel centrifugation method. As cold antibodies, they react mostly at room temperature. They also display a dosage effect. In the enzyme test, especially the two stage tests, the antibodies do not react at all or react only weakly. The enzyme test can therefore be useful for differentiating antibodies in an antibody mixture.

Anti-S, anti-s, and anti-U are best detected in the indirect antiglobulin test. Since the antigens are not destroyed by enzymes, the enzyme test may also be useful for differentiation in certain cases. Their reactivity, even as IgG antibodies, may be enhanced by incubation at lower temperatures, which can be diagnostically useful. On the other hand, antibody screening and cross matching should only detect clinically relevant antibodies that still display warm reactivity and/or Coombs antibodies.

Clinical significance

Glycophorins A and B are crucial for the structure of cells and represent binding sites for the invasion of individual strains of Plasmodium falciparum ; for this reason, individuals who have phenotypes with abnormal GPA or GPB are resistant to these microorganisms to a greater or lesser degree.

27.4.7 Lutheran system

Structure and antigens

The Lutheran gene is located on chromosome 19. The structure of the antigens resembles that of the immunoglobulins. The Lutheran system is therefore a member of the immunoglobulin super family. The system consists of 18 antigens (Lu1–9, Lu11–14, and Lu16–20). The most important antigens are Lua (Lu1) and Lub (Lu2) (Tab. 27-19 – The most important Lutheran antigens), which are encoded by two codominant alleles. Lutheran antigens are expressed more weakly on fetal cells.

The Lutheran (Lu1–3) and para-Lutheran antigens (Lu4–20) are located on two glycoproteins with a molecular mass of 78 kDa and 85 kDa respectively. The number of molecules per cell ranges from 1,500 to 4,000.

The extremely rare Lu(a–b–) (Lunull) genotype is defined by a second gene that is responsible for suppressing not only the antigens in the Lutheran system but also the expression of antigens in other systems e.g., P1. Antibodies

Lutheran antibodies are rarely detected. Anti-Lua

The main reason that anti-Lua is detected so rarely is that the corresponding antigens on the erythrocytes are rarely offered as part of antibody screening and cross matching. Anti-Lua occurs mostly as a natural cold antibody and is of minor clinical relevance. The transfusion of red blood cell units that react negatively in the cross match is considered to be safe as far as hemolytic transfusion reactions are concerned (Tab. 27-24 – Consideration of irregular erythrocyte antibodies). Anti-Lub

This is rare and tends to be IgG; it is capable of activating complement in certain circumstances but rarely causes hemolytic transfusion reactions. Because of the high antigen frequency, anti-Lub is associated with significant problems of blood supply. In urgent clinical situations, the indications for transfusing “incompatible” erythrocytes should not be too restrictive.


Anti-Lua is detectable mainly by agglutination tests (including the single stage enzyme test), while anti-Lub is detected mainly by using the indirect antiglobulin test. Since Lub-negative test cells are usually unavailable, the use of cord erythrocytes is recommended instead (Tab. 27-1 – Blood group antigens on the erythrocytes of newborns).

Clinical significance

Lutheran glycoproteins belong to the immunoglobulin super family (IgSF), a group of immunoglobulin-like adhesion molecules with receptor functions that are thought to be important in signal transduction /36/. The presence of antibodies causes diagnostic and supply problems (anti-Lub).

27.4.8 Xg


Xga is a protease-sensitive, weakly immunogenic antigen (sialoglycoprotein) that is encoded on the X chromosome and for which no antithetical antigen has been found. 67% of all men and 89% of all women possess this antigen. Its expression on neonatal erythrocytes is significantly weaker. Antibodies

Antibodies of the anti-Xga specificity are very rare. They are usually immune antibodies (IgG) with complement-binding capacity; they are detectable in the indirect antiglobulin test using poly specific antiglobulin sera. In most cases, they are presumed to be clinically irrelevant; nevertheless, autoantibodies of this specificity are known to have caused severe autoimmune hemolysis.

27.4.9 Diego/Wright


The antigens Dia (< 0.01%) and Dib (> 99.9%) as well as Wra (0.08%) and Wrb (> 99.9%) are each encoded by two codominant alleles on chromosome 17; both antigens are located on the band 3 protein. The expression of Wrb seems to also require glycophorin A, thus explaining the relationship to the MNS system and the Ena antigen.

Since Europeans almost all exclusively possess Dib, the Diego antigens are for all practical purposes irrelevant in Europe. The antigens are located on proteins to which a function in the cellular transport of chloride and bicarbonate has been attributed. Antibodies

Diego antibodies are extremely rare. They are detected using the indirect antiglobulin test. Only in the case of anti-Dia hemolytic activity has been observed.

The Wright antibodies are much more important. Antibodies of the anti-Wra specificity occur in 1–3% of patients’ sera since they are mostly the product of immunization, not to erythrocytes but to “natural” antigens. Because of the low antigen frequency of Wra, they are often not detected.

Severe hemolytic transfusion reactions are sporadically described in conjunction with these naturally occurring antibodies also. They may occasionally cause hemolytic disease of the newborn as well. If the cause of hemolytic disease of the newborn cannot be determined and/or the causative antibodies are not detectable, anti-Wra should also be considered and paternal erythrocytes should be included in the investigations.

Sera with warm autoantibodies often also contain antibodies of the specificity anti-Wra or anti-Wrb. In general, antibodies of this specificity seem to be more frequent if the elimination of autologous erythrocytes is increased in the presence of a very active immune system.

In individuals with the rare phenotype En(a–), allo antibodies of the anti-Wrb specificity normally occur.

The Wright antibodies are almost always detectable without difficulty by the indirect antiglobulin test.

Clinical significance

The Wrb antigen is also a receptor for Plasmodium falciparum because the invasion of the pathogen can be inhibited by anti-Wrb.

27.4.10 Cartwright


The antithetic, codominantly inherited antigens Yta (99.7%) and Ytb (8.1%) are located on acetylcholinesterase and are encoded by the ACHE gene. Anti-Yta

This antibody occurs relatively frequently and displays warm reactivity. In most cases, it is an IgG (often IgG4) and is detectable in the indirect antiglobulin test, possibly enhanced by proteases. The antibodies are only of partial clinical relevance. Because of supply problems in conjunction with blood transfusions, it is advisable to determine the erythrocyte life span. Although anti-Yta is subject to transplacental transfer, hemolytic disease of the newborn does not occur because of the low expression of Yta on fetal cells (Tab. 27-1 – Blood group antigens on the erythrocytes of newborns). Because of the lack of antigen-negative test cells, it is recommended to include cord blood erythrocytes in the diagnostic evaluation of the antibodies. Anti-Ytb

Ytb is a weak immunogen that is normally expressed on fetal cells. Anti-Ytb is therefore rare and usually of no clinical significance. It is detected using the direct antiglobulin test.

The investigation of Cartwright antibodies, in particular anti-Ytb, is complicated by the fact that they often occur together with other antibodies.

27.4.11 Dombrock


The antigens of this system are located on the glycosylphosphatidylinositol (GPI) molecule, which anchors a group of proteins in the lipid bilayer of the erythrocytes. In paroxysmal nocturnal hemoglobinuria, the synthesis of this molecule is impaired. The most important antigens of this system are the weakly immunogenic, codominantly inherited, antithetic antigenic determinants Doa (66.7%) and Dob (82.8%). Antibodies

The corresponding antibodies rarely occur; they are most likely to occur in conjunction with other antibodies that are directed against erythrocytes. They are immune antibodies (IgG) without the ability to bind complement and are detectable by means of the indirect antiglobulin test (detectability is enhanced if enzyme-treated erythrocytes are used). The antibodies may cause hemolytic transfusion reactions. No cases of hemolytic disease of the newborn have been described to date, although the antigens are fully developed on neonatal erythrocytes.

27.4.12 Colton


Colton antigens are located on the aquaporin molecule, which occurs in large numbers (120,000–160,000 per cell) on erythrocytes and plays a role in the cellular transport of water. The codominantly (chromosome 7) inherited antigens Coa (CO1 99.8%) and Cob (CO2 8.5%) may also be absent in rare cases (Co(a–b–) or CO3 < 0.01%). Antibodies

Antibodies of the anti-Cob specificity are relatively common, but are seldom detected by antibody screening because of the low antigen frequency. As immune antibodies (IgG, sporadically with complement-binding capacity), they are readily detectable by the indirect antiglobulin test (detectability is enhanced if enzyme-treated erythrocytes are used). As far as blood transfusions are concerned, these antibodies are of no significance. However, they must in principle be considered as clinically relevant because delayed transfusion reactions and/or a shortened erythrocyte life span have been described in conjunction with anti-Cob.

27.4.13 Vel


Vel is a high-incidence (> 99.9%), relatively strongly immunogenic antigen with pronounced individual variation that is expressed minimally on neonatal erythrocytes. Antibodies

Although antibodies of anti-Vel specificity (mainly IgM with strong complement activation) are rare, they are associated with significant transfusion-related problems. Because of their hemolytic activity, these antibodies should generally be taken into account; however, this results in significant supply problems. It can be useful to determine the life span of erythrocytes that are positive in the cross match because not all antibodies lead to accelerated erythrocyte elimination.

Anti-Vel is detectable in the enzyme test and the indirect antiglobulin test, if poly specific antihuman globulin sera (complement binding) are used.

In general, the presence of anti-Vel should also be considered if serum reacts reproducibly with almost all test cells (with varying strength) but hardly reacts with cord erythrocytes. If an individual who carries the antibody does not possess the Vel antigen, this virtually proves the presence of the antibody. Anti-Vel is difficult to adsorb and to elute.

27.4.14 Bg


On young erythrocytes in particular, the expression of various HLA antigens shows marked individual variation; on erythrocytes, these antigens are referred to as Bg antigens. Bga correlates with HLA-B7, Bgb with HLA-B17, and Bgc with HLA-A28 (cross reactivity with HLA-A2). By using highly sensitive test methods such as gel centrifugation, even more HLA antigens can be detected on erythrocytes. The frequency with which these antigens are detected on erythrocytes is far lower than for the corresponding HLA antigens. HLA antigens can be eluted using chloroquine. Antibodies

Only high titers of HLA antibodies are serologically detectable as Bg antibodies. To date, only a few cases of hemolytic disease of the newborn and rare hemolytic transfusion reactions have been attributed to these antibodies. Usually, they appear as unidentifiable antibodies to low-incidence antigens in the indirect antiglobulin test. The adsorption of serum samples using pooled platelets and subsequent testing of the adsorbate in comparison to serum diluted equally with normal saline allows inferences to be made at least with regard to HLA antibodies. The specificity can be determined by performing the lymphotoxicity test.

27.4.15 Cryptantigens


Various enzymes can remove surface structures of erythrocytes to reveal underlying structures that are normally not accessible (cryptantigens). The most important cryptantigen is the T antigen. Because cryptantigen structures are ubiquitous in nature, humans are confronted with them in many ways and in turn produce clinically irrelevant natural cold antibodies with anti-T specificity. This is why almost all adult sera agglutinate altered erythrocytes that expose cryptantigens. This type of agglutination reaction is therefore called poly agglutination.


If cryptantigens are exposed in vivo under abnormal conditions (e.g., by bacterial enzymes during bacterial infections) antibodies already present may contribute to accelerated elimination of erythrocytes.

Furthermore, test sera may be contaminated with anti-T, thus leading to erroneous determinations of antigens if T antigens are exposed on erythrocytes in vivo or in vitro (e.g., in blood samples contaminated by bacteria, which most commonly occurs when material is mailed to the laboratory.

Very rarely, cryptantigens may also occur as the result of impaired biosynthesis of the covering surface structures (Tn, HEMPAS, NOR). HEMPAS (hereditary erythroblastic multinuclearity with a positive acidified serum test) is a congenital condition that is associated with dyserythropoietic anemia.

Cryptantigens can be detected and distinguished by means of agglutination tests including the use of commercial lectin kits (Tab. 27-3 – Lectins).

27.5 Immune hemolysis

Definition and pathogenic mechanism

Immune hemolysis refers to the immunologically induced destruction and/or accelerated elimination of erythrocytes. The cause is usually an antigen-antibody reaction with or without activation of the complement system.

Three different pathogenic mechanisms can be distinguished:

  • Antigen-antibody activation of the classic complement pathway (rarely the alternative pathway), the reaction leads to intravascular hemolysis. In most cases, IgM antibodies are responsible.
  • Erythrocytes to which IgG antibodies are attached bind to Fc receptors on macrophages (mostly in the spleen) and are subsequently phagocytosed and subjected to intracellular sequestration (extravascular hemolysis).
  • As part of the antigen-antibody reaction, partial activation of the complement cascade up to C3 occurs on the erythrocyte surface. Erythrocytes to which C3b is attached bind to C3b receptors on macrophages (especially on Kupffer cells in the liver) and are subsequently phagocytosed and sequestered intracellularly. However, this mechanism of extravascular hemolysis is not very effective, especially since the binding to C3b receptors dissociates again in the face of competitive inactivation of C3b to C3dg. As a result, more than 50% of the erythrocytes are able to return to the circulation and display an almost normal life span. On the other hand, the elimination of erythrocytes is markedly accelerated if they bind simultaneously to Fc receptors on macrophages.

Intravascular hemolysis usually runs a more rapid and clinically more severe course. Free Hb occurs in the serum and urine; in extravascular hemolysis, only the most severe courses are characterized by such findings.

Clinically overt hemolytic anemia occurs if a stimulation of erythropoiesis is unable to compensate for the hemolytic process. Often, however, the immunohematological findings of immune hemolysis occur without any clinical signs and symptoms of hemolysis and/or anemia (serological immune hemolysis). In such cases, the antibodies are not pathogenic. This may change over the course of time, however, so appropriate follow-up examinations are required at longer time intervals.

Depending on the pathogenesis and reactivity of the antibodies, various types of immune hemolysis can be distinguished.

27.5.1 Alloimmune hemolysis

Alloimmune hemolysis is caused by allo antibodies and leads to accelerated elimination of erythrocytes against which the antibodies are directed.

Clinically, alloimmune hemolysis presents in the form of hemolytic transfusion reaction or hemolytic disease of the fetus and newborn.

Hemolytic transfusion reaction

Acute or delayed hemolytic transfusion reaction (major reaction) may be triggered in patients who receive erythrocyte containing blood components with blood group antigens to which they have been sensitized.

Acute hemolytic transfusion reaction is also seen, albeit more rarely and with a much milder clinical course, following the transfusion of plasma containing blood components, due to the transfer of antibodies directed against the blood group antigens of the recipient (minor reaction). The rate of hemolytic transfusion reactions in relation to units of blood transfused is 0.02–0.2 ‰ /37/. Acute hemolytic transfusion reaction

In acute hemolytic transfusion reaction, hemolytically active allo antibodies (rarely, autoantibodies) are already present at the time of the transfusion at a sufficiently high concentration. These antibodies are almost always detectable by standard methods such as blood group determination, antibody screening, and cross matching, when performed correctly. Acute hemolytic transfusion reaction is therefore due in the majority of cases to logistical (organizational) mistakes in the preparation of the transfusion and much more rarely to errors in the performance of immunohematological tests. Because of the high hemolytic activity of the antibodies, AB0 mismatches during transfusion are especially dangerous. They account for the major portion of the acute hemolytic transfusion reactions that have a lethal outcome. Such occurrences are also referred to as hemolytic transfusion incidents. Delayed hemolytic transfusion reaction

If a recipient is exposed by transfusion to blood group antigens to which he or she is already immunized, greatly increased synthesis of hemolytically active allo antibodies ensues within 3–14 days (booster effect).

Delayed hemolytic transfusion reaction is caused especially by Rh antibodies (mainly anti-c) and antibodies of the Kidd system (mostly anti-Jka). Such reactions are unavoidable, but rarely fatal, complications of blood transfusion. The rate of these complications can be reduced by meticulously documenting all IgG antibodies detected in an individual (emergency and maternity records) and by taking these into account for all transfusions, even if they are no longer detectable. Women in particular are at an increased risk since they may be immunized as a result of pregnancies.

According to older statistical reports, about two thirds of all hemolytic transfusion reactions are characterized by a delayed course. However, they occur much more rarely nowadays since the introduction of more sensitive antibody detection methods such as gel centrifugation test and solid phase technology.

Two to four times more common is the occurrence of the typical immunohematological findings associated with a delayed hemolytic transfusion reaction in the days to weeks following transfusion, in the absence of clinical symptoms or hemolysis /1415/. These delayed serological transfusion reactions are more likely to be identified if more sensitive antibody detection methods are used.

Diagnostic tests

For the diagnostic investigation of hemolytic transfusion reactions, refer to Section 27.6.5 – Antibody screening.

27.5.2 Hemolytic disease of the newborn (HDN)

Pathogenic mechanism

HDN (also known as HDFN, hemolytic disease of the fetus and newborn) is caused by maternal, hemolytically active allo antibodies (rarely also autoantibodies) that are directed against blood group antigens of the fetus/newborn and that can cross the placenta during pregnancy. These antibodies are usually IgG1 antibodies (rarely IgG3) that pass through the placenta relatively early during pregnancy and subsequently accumulate in the fetal blood. The fetal/neonatal erythrocytes to which the IgG antibodies are bound are eliminated at an accelerated rate by the reticuloendothelial.

Clinical symptoms

The full blown clinical picture consists of anemia with reticulocytosis and erythroblastosis, jaundice, and hydrops fetalis. Jaundice does not develop fully until after birth, when maternal elimination of bilirubin is no longer occurs and the liver of the neonate is not yet capable (due to immaturity) of conjugating bilirubin with glucuronic acid in order to excrete it via the biliary tract. With rising levels of unconjugated bilirubin, the risk of cerebral damage due to kernicterus increases because unconjugated bilirubin, due to its fat solubility, is capable of crossing the blood-CSF barrier. Hydrops is a condition characterized by generalized water retention due to hypoxia and hypoalbuminemia that can lead acutely to death as a result of heart failure. Antepartum blood group diagnostic tests

Antepartum examinations include:

  • Blood grouping and antibody screening in all pregnant women during the first trimester. Under these circumstances, it is always advisable to use very sensitive techniques such as enzyme tests in order to detect immunization as early as possible.
  • Repeat antibody screening during the 24th to 27th gestational week
  • In the event of a positive antibody screening test, investigation (specificity, Ig class, titration) of the antibodies that may cause hemolytic disease of the newborn. Regular monitoring (antibody differentiation, titration) while simultaneously running the first or preceding sample as a titer control; every four weeks up to the 30th gestational week and every two weeks thereafter (or if there is a rise in titer).
  • If antibodies are detectable, determination of the paternal blood group antigens. If anti-D is detected and the father is Rh(D) positive, additional molecular genetic examination of the RH(D) factor is recommended, since this is the only way to determine the RHD zygosity (hemizygous or homozygous) /3839/.

Comments: the determination and quantification of IgG subclasses by gel centrifugation or flow cytometry provides preliminary information regarding the hemolytic activity of irregular antibodies. Bioassays such as the examination of the antibody dependent cell mediated cytotoxicity (ADCC) of mononuclear cells, on the other hand, have not established themselves despite good correlation with in vivo hemolytic activity, because of standardization problems and extensive methodological demands /40/. Presence of irregular erythrocyte antibodies in pregnancy

If antepartum examinations indicate the presence of irregular erythrocyte antibodies that may cause hemolytic disease of the newborn (HDN), a thorough prenatal diagnostic evaluation is required. During invasive procedures, precautions must be taken to avoid damaging blood vessels and introducing fetal erythrocytes into the maternal circulation, where they might result in immunization and/or a booster effect.

Noninvasive examinations are therefore performed initially:

  • Fetal anemia can be diagnosed by measuring the velocity of blood flow in the fetal middle cerebral artery using Doppler ultrasonography /41/.
  • If the father is heterozygous for a Rhesus blood group antigen against which maternal antibodies are directed, the fetal blood group can be determined from maternal blood /4243/. Cell free DNA is isolated from plasma and molecular biological methods are used to detect allele specific nucleic acid polymorphisms.
  • Fetal blood group antigens from other blood group systems can also be determined from maternal blood in specialized laboratories by isolating fetal DNA from amniotic fluid
  • Examination of cordocentesis blood (e.g., cell content and blood group antigens) is possible from the 18th week of gestation. This invasive procedure should be considered if (Doppler) ultrasonography has indicated possible fetal anemia and an intrauterine transfusion is planned. Postpartum diagnostic tests

In newborns with unexplained anemia as well as following protracted delivery or manual removal of the placenta, it is important to determine that fetal-maternal macro transfusion has not occurred, in which case the customary standard dose of anti-D would not be sufficient for Rh immune globulin (RhIg) prophylaxis. The rare (< 0.3%) situation of fetal maternal macro transfusion occurs when more than 30 mL of fetal blood enters the maternal circulation (i.e., > 5 ‰ of fetal erythrocytes are detectable in the maternal blood sample).

Examinations to check for the presence of fetal-maternal macro transfusion include (refer to Section 38.7 – Fetomaternal hemorrhage):

  • Kleihauer-Betke test: in contrast to HbF, HbA can be removed from alcohol treated adult erythrocytes at a pH of 3.3. Based on their HbF content, fetal cells that have been introduced into the maternal circulation are microscopically distinguishable from adult cells (ghosts) after an eosin stain has been performed and can be determined semi quantitatively.
  • Quantification of D-positive fetal erythrocytes in maternal blood by flow cytometry. Anti-D (IgG) attaches to D-positive fetal erythrocytes, which are labeled using fluorescent secondary antibodies (e.g., murine antihuman IgG).

On the other hand, universal postpartum monitoring of Rh immune globulin prophylaxis is no longer indicated, in particular since the introduction of antepartum anti-D prophylaxis, since residual anti-D is often still detectable using sensitive detection methods. Postnatal diagnostic tests

The following examinations are carried out:

  • In children of Rh(D)-negative mothers: determination of the ABO blood group and Rh factor as well as the direct antiglobulin test
  • In children of mothers with blood group O: blood group determination in the child, including the direct antiglobulin test
  • In children of mothers who did not undergo antibody screening during pregnancy: direct antiglobulin test
  • In children of mothers with irregular antibodies to erythrocytes that may cause hemolytic disease of the newborn: direct antiglobulin test, including titer, corresponding antigens on neonatal erythrocytes, antibody screening, and antibody differentiation (and titration, where appropriate)
  • In suspected hemolytic disease of the newborn: direct antiglobulin test, complete blood count including a differential blood count and reticulocyte count as well as a bilirubin determination. The bilirubin levels are considered in relation to the maturity and age of the neonate, thus allowing the optimal timing for the initiation of phototherapy and/or exchange transfusion to be determined.
  • In newborns with anemia, hemolysis, or jaundice of undetermined cause: direct antiglobulin test, antibody screening with maternal and neonatal serum, exclusion of AB0 incompatibility (see below)
  • In newborns with a positive direct antiglobulin test: all examinations that are required to detect and/or identify the antibodies.

Examinations for identifying antibodies:

  • Antibody screening with serum from the neonate and/or mother
  • Inclusion of paternal erythrocytes if antibody screening is negative
  • Differentiation and possibly quantification of the antibodies if antibody screening is positive
  • Antibody screening of the eluate of neonatal cells and differentiation of the antibodies that can be eluted, with the possible inclusion of paternal erythrocytes
  • Exclusion of AB0 incompatibility.

Comments: because of antepartum Rh immunoglobulin prophylaxis and the use of sensitive techniques such as gel centrifugation, the direct antiglobulin test is relatively often positive in Rh(D)-positive newborns of Rh-negative mothers although this finding is of no clinical relevance. In these cases, no further diagnostic investigations are necessary unless special or unusual clinical findings are present.

If the differential diagnostic question arises of whether a woman has anti-D because of RhIG prophylaxis or because of active immunization (e.g. due to the failure of RhIG prophylaxis), monitoring of antibody titers provides valuable information. RhIG prophylaxis practically never leads to titers > 8 if the titration is based on the indirect antiglobulin test using the tube method (> 32 for the gel centrifugation test). However, the administered antibodies may be detectable for up to six months. AB0 incompatibility


Hemolytic disease of the newborn occurs most commonly (approximately 1 in 100 births) in conjunction with AB0 incompatibility. However, the disease is usually characterized by a mild course with prolonged jaundice lasting for several weeks. Severe forms of the disease requiring an exchange transfusion are rare (0.02–0.03% of all births) /44/. This is because the number of AB0 antigens on the fetal/neonatal erythrocytes is relatively low and these show little branching; furthermore, the antibodies bind preferentially to endothelial or soluble AB0 antigens. Therefore, AB0 incompatibility usually occurs only in neonates born at term or post term. The corresponding IgG antibodies are not produced as a result of pregnancy or transfusion, thus accounting for the fact that AB0 incompatibility may occur during the first pregnancy. AB0 incompatibility is most commonly associated with the pattern mother 0, child A (less commonly, B). From a differential diagnostic point of view, fetal-maternal transfusion of a significant magnitude should be considered if pronounced anemia is present in the newborn in conjunction with increased erythropoiesis or hydrops fetalis.

Diagnostic tests

Antepartum and prenatal diagnostic investigations for hemolytic disease of the newborn are not required unless the history reveals any unusual findings. The diagnosis depends on the clinical picture in combination with the corresponding immunohematological findings. These findings include:

  • The detection of spherocytes
  • The detection of the corresponding blood group constellation between mother and child, possibly also the father (A1, A1B, or B)
  • A positive direct antiglobulin test of the IgG type (gel centrifugation)
  • A high anti-A (IgG) titer (> 32 in the AB neutralization test)
  • Detection of the causative AB0 antibodies, possibly also of anti-AB, using Rh(D)-negative test cells of the corresponding AB0 blood group (indirect antiglobulin test), both in the serum and the eluate (acid eluate).

The immunohematological findings correlate only minimally with the clinical picture. Since the introduction of sensitive techniques such as gel centrifugation or solid phase technology, the immunohematological picture of AB0 incompatibility is a common laboratory finding, even if there is no clinical evidence for the presence of hemolytic disease of the newborn. Clinically relevant AB0 incompatibility, on the other hand, can be reliably ruled out if the relevant antibodies are not detectable in the serum and eluate. The detection of hemolysins is of no significance.

A sizable fetal-maternal transfusion can be assumed if mixed field agglutinations are detectable during determination of the maternal blood group.


If a transfusion is required, group 0 packed RBCs with the Rh factor of the child are used. If anti-D is detectable in the maternal serum (even if this is due to RhIG prophylaxis), Rh(D)-negative packed RBCs are used. Maternal serum/plasma can be used for the cross match (up to four weeks post partum). Rh incompatibility


Hemolytic disease of the newborn (HDN) due to anti-D has become a rare event thanks to the use of antepartum and postpartum Rh immune globulin prophylaxis (< 0.11%). Nonetheless, severe HDN necessitating treatment is still more frequently caused by anti-D than by any other incompatibility. Immunization usually results from preceding pregnancies during which RhIG prophylaxis was ineffective or was not administered at all (migrants in particular). Thus, Rh incompatibility usually does not affect the first born child. HDN may also be caused by other Rh antibodies, especially by anti-c and more rarely by anti-E and anti-C. In these cases, however, HDN tends to be mild. Diagnostic tests

Necessary examinations include:

  • The determination of the paternal Rh genotype in the case of women with Rh antibodies. This allows a prediction to be made with a certain degree of probability about whether the child will inherit the corresponding antigens.
  • If the father is positive for anti-D and Rh(D), the blood group of the fetus is determined from the serum of the mother using molecular genetic methods. Some authors even advocate genotyping for all Rh-negative mothers.
  • Molecular genetic examination is also advisable if other antibodies that can cause HDN are present and the father is heterozygous for the corresponding blood group antigen
  • The concentration of Rh antibodies must be monitored regularly. However, the concentration correlates to a limited degree only with the hemolytic activity (subclasses). However, antibody titers of less than 16 in the conventional tube test (or less than 64 in the gel centrifugation test) are usually associated with mild hemolysis that generally does not necessitate premature delivery.

Because titration results are subject to methodological variation, mainly due to differences in the age and antigenicity of test cells or antibody carry over effects during the transfer of serum, the following approach is recommended:

  • Parallel titration with earlier serum samples (stored at < –25 °C) from the same patient
  • Parallel titration of diluted anti-D (produced from anti-D for RhIG prophylaxis) with a defined antibody concentration, in order to calculate the anti-D concentration in the patient serum sample. Anti-D concentrations of > 10 μg/mL are considered to be critical. This corresponds to a titer of ≥ 32 in the indirect antiglobulin test using the tube method.
  • If possible, use of the same methodology with test cells of comparable antigenicity (cDE/cDE if possible). With regard to the comparability and the clinical assessment of the findings, the controls should, if possible, always be conducted in the same laboratory using generally accepted standard methods that are not too sensitive (e.g., the indirect antiglobulin test in a test tube using the albumin technique). In the case of high titers (> 1,000), it is advisable to change the disposable tips of the pipettes at each titer dilution in order to avoid carry over effects. If the sensitive gel centrifugation method is used, the results, which are usually about two titer dilutions higher, must be evaluated accordingly.
  • Titer increases of ≥ two dilutions and/or an increase in the added agglutination score of serial titration of ≥ 10 are considered to be significant.

Comments: if neonatal erythrocytes are coated with anti-D (IgG), no further anti-D may be bound during the Rh determination because all binding sites are already occupied. In this case, the determination may erroneously give Rh negative or weak D as the result.


Maternal serum/plasma rather than neonatal serum should be used for the cross match. The blood for transfusion is then selected based on the respective AB0 constellation between mother and child. Cross match testing should not be abandoned generally. Cross matching does not need to be repeated for up to four weeks if blood from the same donor is transfused in small portions (baby units).

Incompatibilities outside the AB0 and Rh systems

Rarely, hemolytic disease of the newborn (HDN) can also be caused by antibodies outside the AB0 and Rh systems. Sporadic cases with severe clinical courses are also seen. In such cases, the mothers have usually been immunized as the result of previous transfusion. Anti-K antibodies play a particularly significant role in these cases, followed by anti-Jka, anti-Fya, anti-S, and, very rarely, anti-k, anti-s, and anti-Fyb. In principle, however, almost all IgG antibodies to fetal blood group antigens are capable of inducing HDN.

In particular, the possibility of antibodies to rare, so called private antigens should be considered if hemolysis of undetermined etiology is present in the newborn and the direct antiglobulin test is positive in the face of negative antibody screening. In such cases, paternal erythrocytes must be included in the detection of antibodies. AB0 antibodies in the serum of the mother and/or child may need to be neutralized prior to the test by means of AB substance or adsorbed by corresponding test cells. Alternatively, the eluate of neonatal erythrocytes can also be used.

27.5.3 Autoimmune hemolysis

Definition and classification

Autoimmune hemolysis is defined as hemolysis that is mainly caused by autoantibodies with or without activation of complement. Three different types of autoimmune hemolysis are distinguished, based on the in vivo behavior patterns of the autoantibodies pathophysiologically responsible for the condition; these behavior patterns are also documented in vitro and are characterized mainly by differences in temperature reactivity. Accordingly, autoimmune hemolysis may be caused by:

  • Warm autoantibodies
  • Mono thermal cold autoantibodies
  • Bithermal cold hemolysins.

If autoimmune hemolysis occurs in association with another disease, it is referred to as symptomatic or secondary autoimmune hemolysis.

If no primary disease is found despite an extensive diagnostic evaluation, the autoimmune hemolysis is said to be idiopathic.

If all symptoms, including the immunohematological changes, disappear within a period of six months, this type of autoimmune hemolysis is considered to be acute reversible; otherwise, it is classified as chronic irreversible.

If pronounced symptoms of hemolysis and anemia are present, the disease is referred to as autoimmune hemolytic anemia. Warm autoantibodies

Incomplete warm autoantibodies (mostly of the IgG class), with and without partial complement activating capacity, cause accelerated extravascular hemolysis via the pathogenic mechanisms described in Section 27.4 – Immune hemolysis.


Approximately 70–80% of all cases of autoimmune hemolysis are caused by warm autoantibodies. More than half of them are found to be secondary, if the appropriate immunohematological investigations are performed when hemolysis occurs in diseases where secondary autoimmune hemolysis is likely. The acute reversible type of autoimmune hemolysis is found in patients with infections such as CMV, EBV, and Yersinia, whereas the chronic irreversible type is primarily seen in patients with hemoblastoses, especially CLL and Hodgkin’s disease, as well as in autoimmune diseases (e.g., disseminated lupus erythematosus). Immunohematological diagnostic tests do not allow differentiation between the various clinical types.

Clinical symptoms

The clinical presentation varies from milder forms of the disease with anemia and discrete signs of hemolysis, to pronounced hemolytic jaundice with compensated anemia, and finally to severe decompensated hemolytic anemia with cardiogenic shock. Splenomegaly and/or hepatomegaly are found, depending on the duration of the disease and the site of erythrocyte breakdown.

Sometimes, patients present primarily with complications of autoimmune hemolysis such as gallstones and thrombosis or with side effects such as persistent weakness or upper abdominal discomfort. Autoimmune hemolysis is often an incidental diagnosis during the investigation of secondary findings such as an elevated erythrocyte sedimentation rate.

Evans syndrome with simultaneous autoimmune thrombocytopenia represents a prognostically unfavorable special type of autoimmune hemolysis. Immunohematological findings

The direct anti-globulin test using the tube method is considered to be the gold standard for the detection of warm autoimmune hemolytic anemia /45/. It has the highest specificity (i.e., the highest correlation between a positive test result and clinically overt autoimmune hemolytic anemia). However, only 98–99% of all autoimmune hemolysis cases yield a positive result in this test /1146/. The direct antiglobulin test (tube method) may be negative in Coombs negative autoimmune hemolytic anemia if erythrocytes are coated with too few IgG autoantibodies or lgG autoantibodies with low avidity (easily removed), or with IgA and/or IgM autoantibodies. If the direct antiglobulin test is conducted using more sensitive gel centrifugation tests and solid phase technology, Coombs negative autoimmune hemolysis is relatively rare. Therefore, in cases of hemolytic anemia of undetermined cause and/or suspected autoimmune hemolytic anemia, the direct anti-globulin test should always be performed using a number of different methods and with various anti-human globulin reagents with different specificities (anti-IgG, anti-IgA, anti-IgM, anti-C3d) /45/.

In more than 75% of cases, the erythrocytes are shown to be coated with IgG (in approximately one third, only IgG is found), while binding of C3d is found to be present in the same percentage of cases (rarely without simultaneously detectable IgG). In more than 90% of cases, the antibodies involved are IgG1, whereas in less than 5% they belong to IgG3, almost always in association with clinically overt hemolysis.

By using antibody elution, it is often possible to detect IgG antibodies in cases that show complement binding alone. Complement activation is more likely to occur in the presence of autoantibodies that have more complex specificity or that are not directed against the Rh system. It also tends to be more common in mixed types of warm and cold autoimmune hemolysis. In some cases associated with the detection of complement alone, incomplete warm hemolysins are found that are mainly IgM.

Sporadically, the autoantibodies belong to the IgA class and do not show any complement binding. Together with IgG and/or complement, IgA antibodies are found in up to 20% while IgM antibodies are present in approximately 8% /46/.

The type and extent of complement binding do not allow conclusions to be drawn concerning the severity of hemolysis. However, cases with simultaneous binding of both IgG and complement are more frequently associated with marked hemolysis, whereas those cases with complement binding alone are characterized by milder forms of hemolysis. The immunohematological pattern tends to change over the course of time. In individual cases, a correlation can be seen between the clinical course of the disease and changes in the type and extent of red cell coating.

In the indirect antiglobulin test using the tube technique, only about one third of patients do not display excessive free autoantibodies in the serum. The sensitivity of the enzyme test for the detection of autoantibodies is approximately 60%. The detection of Coombs antibodies in the serum correlates more frequently with overt hemolysis and is therefore more relevant clinically. This statement, however, no longer applies to the same extent when more sensitive test methods such as the gel centrifugation test or solid phase technology are used. Particularly when solid phase technology is used, autoantibodies are very often detected that are of no clinical relevance.

Specificity of the autoantibodies

In about 75% of cases, warm autoantibodies do not display a particular specificity. Therefore, they react not only with the patient erythrocytes but also with all normal donor erythrocytes (anti-nl = anti-normal). Specificities are detectable at least in part by performing examinations with very rare test cells (e.g., Rhnull, D, En(a–), and Wr(b–) cells). Under such conditions, it has been demonstrated that in half of cases, the antibodies are directed against the Rh complex (no reaction with D and/or Rhnull cells) while in one third, they are directed against the high incidence antigens Ena and/or Wrb. This type of differentiation, however, is of no practical relevance.

More important is the fact that in about 30% of cases, warm autoantibodies also apparently display special Rh specificity (anti-e in particular). In general, however, the antibodies are not mixed antibodies with anti-nl and (e.g., anti-e specificity); instead, this finding is based only on an increased reactivity of the nonspecific Rh autoantibodies with e cells. The apparent anti-e can also be adsorbed by EE cells. Often, the specificities of the serum antibodies and the eluted antibodies do not even correspond with each other.

Specific Rh autoantibodies are also rarely found (< 5%). In some cases, they cannot be distinguished from allo antibodies without adsorption tests if the patient does not possess the corresponding antigens or if these antigens are suppressed as part of the disease.

Warm autoantibodies may also show a host of other specificities (e.g., against Ge, Jka, K1, K4, K5, K13, Xga, LW, and U).

Especially if sensitive test methods are used, erythrocyte bound warm autoantibodies and/or IgG molecules are relatively often found, not only in various diseases, but also in healthy blood donors, without any signs of hemolysis (serological immune hemolysis; for differential diagnosis, refer to Section 27.5.2 – Hemolytic disease of the newborn. Therefore, the diagnosis of clinically relevant autoimmune hemolysis is generally not based on immunohematological findings alone. Nonetheless, most cases of autoimmune hemolysis are detected not because the clinician has specifically requested the appropriate tests, but because of abnormal results in the immunohematological tests used for the preparation of a blood transfusion.

Abnormal immunohematological results due to warm autoantibodies:

  • The AB0 determination rarely gives abnormal results. If many antibodies are attached to the patient’s erythrocytes and unwashed patient erythrocytes are used (serum as supplement), tests using AB0 reagents may yield false positive results. However, this problem can be recognized easily based on the reaction pattern of the reverse typing and the positive auto control or positive antiglobulin test.
  • The Rh determination yields a false positive result especially if a supplement test is used; however, this is recognizable from the Rh control (auto control) or positive antiglobulin test. The same applies to any other blood group antigen determination, especially if it employs the indirect antiglobulin test. If these problems occur when determining the AB0/Rh antigens, the test can first be undertaken to exclude the influence of cold agglutinins.
  • The antibody screening becomes positive not only in the auto control of the indirect antiglobulin test but also in the test cells, if free serum autoantibodies are present. The same applies to cross match testing.

The problems described concerning the determination of blood group antigens can be at least partially eliminated by performing the tests in a saline medium using monoclonal test reagents and by using anti-IgG instead of poly specific antiglobulin serum in cases where patient erythrocytes are coated exclusively with complement. Intact antigens can be exposed by eluting the antibodies; however, this method is only partially successful and requires a lot of time. Its performance is limited to specialized laboratories. Refer to Section – Elution). Diagnostic tests

If autoimmune hemolysis is suspected and to clarify the aforementioned findings, the following approach is recommended:

  • Antibody screening with patient serum in the enzyme and indirect antiglobulin test, including auto controls
  • Direct antiglobulin test using at least two poly specific antiglobulin reagents that are directed against IgG and C3d (gel centrifugation and/or solid phase technology) and, if indicated and if this test is negative, also including mono specific anti-IgA and anti-IgM.

If the direct antiglobulin test using the sensitive gel centrifugation test and solid phase technology is positive and there are no clinical indications for performing the tube test (hemolysis and/or anemia of undetermined cause), the following approach is recommended:

  • Cold agglutinin test in saline medium with adult and neonatal erythrocytes to enable differentiation from cold agglutinin autoimmune hemolysis and to detect mixed autoimmune hemolysis
  • Antibody screening (gel centrifugation test or solid phase technology) with an acid eluate of patient erythrocytes
  • Possibly hemolysis tests with enzyme treated test erythrocytes at 37 °C, 22 °C, and under bithermal conditions.

If there is evidence of the presence of warm autoantibodies, the following examinations should be added:

  • Identification of serum antibodies and exclusion of allo antibodies; possible titration with one or more test cells or differentiation using diluted serum to identify mixed antibodies and for monitoring
  • Possibly differential adsorption to test cells with different antigen patterns and elution of the serum antibodies in order to differentiate allo antibodies and autoantibodies; if the patient’s erythrocytes are coated with complement alone and/or only weakly with IgG, auto adsorbates are preferable. If the patient’s erythrocytes are heavily coated with immunoglobulin, auto adsorption can only take place after the autoantibodies have been eluted.
  • If cold agglutinins are simultaneously present, investigation of these as described below
  • Identification of the globulins detected on the erythrocytes by mono specific antiglobulin reagents in the direct antiglobulin test (anti-IgG, anti-IgA, anti-IgM, anti-C3d); if the patient’s erythrocytes are coated with complement alone, controls with EDTA blood or citrated blood instead of clotted blood and possibly titration using the tube test for monitoring purposes; a marked prozone phenomenon suggests the presence of benign warm autoantibodies
  • Identification of the different types of eluted autoantibodies, thus allowing better differentiation from allo antibodies; possibly titration
  • Warm hemolysis test, at least if patient’s erythrocytes are coated with complement alone; possibly titration.

Determination of the antibody IgG subclasses and possibly quantification of the globulins coating the erythrocytes (e.g. using the gel centrifugation test or flow cytometry) is desirable. This allows a better assessment of the clinical relevance of the immunohematological findings, especially during monitoring /47/.


If excess autoantibodies are detectable in the indirect antiglobulin test (including when less sensitive, but therefore more clinically relevant, test methods such as the albumin tube test are used), transfusions must be restricted to acute, life-threatening indications since a compatible transfusion is generally not feasible in the presence of autoantibodies. In addition, the differentiation between autoantibodies and allo antibodies is difficult under routine conditions and every transfusion is associated to a small extent with the risk of activating autoimmune hemolysis. Allo antibodies must be taken into consideration. This also applies, where possible, to specific autoantibodies directed against low frequency antigens such as anti-E and anti-K. It does not make sense to consider common autoantibodies with apparent Rh specificity (e.g. anti-e). Even in the absence of excess serum autoantibodies, the indication for transfusion should be restricted due to the risk of autoimmune activation.

Note: when sensitive test methods (in particular solid phase technology) are used, clinically irrelevant free warm autoantibodies are detected relatively commonly in patients who do not have hemolytic anemia. In such cases, cross match testing should be carried out using diluted patient serum/plasma (1 volume of serum/plasma + 1–2 volumes of saline) or the tube method in the indirect antiglobulin test. Despite the lower sensitivity of these methods, a negative cross match result means that relevant acute reactions are unlikely. Cold agglutinins and mono thermal cold hemolysins


Cold agglutinins belong to the IgM class and are produced in low concentrations by almost all humans during the first years of life. However, they are only of clinical relevance if they are capable of agglutinating erythrocytes at a temperature > 30 °C and hence, under normal conditions, of activating complement and of hemolyzing erythrocytes (cold hemolysins) to a significant extent.

Under abnormal conditions, these cold agglutinins are present at higher concentrations and/or have a wider thermal amplitude. In the case of cold exposure, they then agglutinate the erythrocytes of patients. This causes the blood flow to slow down in the small vessels of the parts of the body exposed to cold (acra), which leads to the development of cyanosis (acrocyanosis) due to deoxygenation. Simultaneously, the complete complement cascade is activated on the surface of the erythrocytes, which is associated with acute intravascular hemolysis.

If erythrocytes coated with C3b are not phagocytosed in the liver prior to the inactivation of the complement factors (C3dg), they recirculate and subsequently display an almost normal life span. Furthermore, the bound inactive complement factors inhibit further activation of complement, thereby protecting the erythrocytes from a repeat attack by cold antibodies.


Approximately 10–20% of cases of autoimmune hemolysis are part of the cold agglutination syndrome, of which about half are idiopathic (cold agglutinin disease). Symptomatic forms of cold agglutinin syndrome with an acute reversible course are found in infections (e.g., caused by Mycoplasma and the Epstein-Barr virus, whereas those with a chronic irreversible course are seen in conjunction with lymphoproliferative diseases (e.g., malignant lymphoma and Waldenstroem’s macroglobulinemia) and various malignant tumors such as gastric cancer.

Clinical symptoms

Usually, only mild anemia is present but with relatively pronounced signs of hemolysis. Depending on exposure to cold, the symptoms show seasonal variation. Patients usually consult a physician because of circulatory problems (acrocyanosis) or post exposure hemoglobinuria. Another diagnosis that frequently prompts clinical admission is tumor anemia and a high erythrocyte sedimentation rate.

Immunohematological findings

The immunohematological picture is characterized by:

  • The abnormally high cold agglutinin titer at 0 °C (> 64)
  • Complement coating of the patient’s erythrocytes (positive direct antiglobulin test with polyspecific antiglobulin reagent and anti-C3d)
  • The detectability of cold hemolysins, at least if enzyme treated erythrocytes are used. Only rarely are borderline cold agglutinin titers found at 0 °C, although distinct agglutinations are still seen at temperatures ≥ 30 °C, especially in the albumin test. The specificity of the cold agglutinins is almost always anti-I, rarely anti-i (often in infectious mononucleosis), and very rarely anti-Pr or anti-Gd. Determination of the specificity is of little clinical relevance. Anti-Pr antibodies of the IgA class do not cause hemolysis. Laboratory findings in the presence of cold autoantibodies

Increased cold agglutinins can interfere with clinical laboratory tests to a greater extent than warm autoantibodies, in particular with immunohematological examinations:

  • Blood and bone marrow smears are difficult to prepare unless the specimens are warmed
  • Citrated blood and EDTA blood appear to be clotted (agglutinated)
  • Blood counts obtained by hematology analyzers show abnormal results
  • The erythrocyte sedimentation rate shows varying acceleration, depending on the ambient temperature and the time interval until the blood is collected into the sedimentation pipette
  • During AB0 and Rh determination, all tests (including the auto control/Rh control) can yield more or less positive results, if the tests are not strictly performed at 37 °C; in the presence of very high antibody titers, the patient erythrocytes, patient serum, and all the reagents as well as the materials and test equipment must be preheated to 37 °C.
  • In antibody screening and cross matching, all tests (including auto controls) may yield positive results despite strict warm incubation.

In order to detect allo antibodies, the effect of the cold autoantibodies needs to be eliminated. This can be done using the following measures:

  • Usually, preheating all reaction components and performing the test strictly at 37 °C, using appropriately preheated saline as the wash solution, is sufficient.
  • In addition, it may be necessary to use anti-IgG instead of poly specific antiglobulin reagent in the indirect antiglobulin test in order to eliminate the interference caused by complement binding.
  • In some cases, it may be necessary to adsorb the cold antibodies from the serum using patient cells.
  • The destruction of the antibodies by dithiothreitol or 6-mercaptoethanol is limited to specialized laboratories.
  • Antigen determinations in the indirect antiglobulin test yield false positive results if poly specific Coombs sera are used. This can be avoided by using anti-IgG as the antiglobulin serum.

Note: Especially if the gel centrifugation method is used, false positive results are to be expected for antigen testing if the patient’s erythrocytes auto agglutinate prior to centrifugation. Diagnostic tests

For the diagnosis of cold agglutinin autoimmune hemolysis, it is recommended that blood samples are kept warm and separated at 37 °C (clotted blood or EDTA blood). However, blood samples can also be collected and transported under routine conditions. In this case, they have to be reheated to 37 °C, followed by sedimentation, using a water bath. The serum and/or plasma should be separated without further cooling. The following tests are recommended:

  • Direct antiglobulin test with at least two poly specific antiglobulin reagents using warm washed patient erythrocytes from EDTA blood; if positive, follow up examination with mono specific antiglobulin reagents, (e.g., anti-IgG and anti-C3d) and possibly titration
  • Cold agglutinin test; if positive, cold agglutinin titer and determination of the thermal amplitude (serum or EDTA plasma)
  • If the cold agglutinin titer is elevated, mono thermal cold hemolysis tests with untreated and enzyme treated erythrocytes; if positive, possibly titration (serum)
  • If the direct antiglobulin test is positive but the cold agglutinin titer is borderline, possibly determination of the thermal amplitude of the cold agglutinins
  • Determination of antibody specificity, which can be carried out according to Tab. 27-20 – Determination of the specificity of cold autoantibodies, based on examples /49/.

Note: caution should be used when interpreting the immunohematological findings. In warm autoimmune hemolysis, elevated levels of cold agglutinins are also found relatively often. A corresponding note should be entered in the transfusion record sheet.


The immunohematological tests necessary for the release of blood units that are compatible with regard to AB0 and any allo antibodies that may be present are described above. Usually, there are no restrictions on blood transfusion because of the presence of cold autoantibodies, if the blood units are preheated prior to transfusion (37 °C). Mixed autoimmune hemolysis

In approximately 8% of autoimmune hemolysis cases, agglutinating cold autoantibodies (IgM, especially anti-I) with a slightly elevated cold agglutinin titer at 0 °C but a wide thermal amplitude and the capacity for complement activation are detectable in addition to the usual warm autoantibodies of the IgG class. IgG and complement are bound to the erythrocytes of patients. Pronounced anemia is often present but responds well to the standard therapy for warm antibody autoimmune hemolysis. Bithermal cold hemolysins of the Donath-Landsteiner type

After cold exposure, intravascular hemolysis is triggered by incomplete autoantibodies that bind reversibly at a low concentration to erythrocytes in the cold and strongly activate complement up to C9 during rewarming.


Autoimmune hemolytic anemia of the Donath-Landsteiner type is rare (< 2% of autoimmune hemolysis cases). It occurs almost exclusively in the form of an acute reversible disorder in young children during acute viral infections of the upper respiratory tract. Autoimmune hemolysis in childhood is thought to be relatively frequently caused by bithermal cold hemolysins /50/.

Difficulties are encountered in the differential diagnosis of a positive hemolysis test (refer to Section – Hemolysis methods). In the past, bithermal cold hemolysins were frequently observed in a chronic form during the late stages of syphilis. Rarely, they also occur as idiopathic chronic antibodies in elderly individuals.

Clinical symptoms

In most cases, autoimmune hemolysis of the Donath-Landsteiner type presents clinically as an acute, severe hemolytic anemia with chills, fever, nausea, abdominal discomfort, and hemoglobinuria.

Immunohematological findings

Often no autoantibodies, only complement factors, are detectable on the erythrocytes of patients. Therefore, autoantibodies cannot usually be eluted from the patient’s erythrocytes using the usual methods. The antibodies are detected by means of the hemolysis test using biphasic incubation; this is why they are referred to as bithermal cold hemolysins. Sometimes, they also react as incomplete antibodies in the enzyme test, and, more rarely, also in the indirect antiglobulin test. They usually display anti-P specificity (anti-P1+Pk+p or -Tja). Bithermal cold hemolysins of anti-I (-HI), anti-i, and anti-Pr specificity have also occasionally been described. Diagnostic tests

Samples (clotted blood and EDTA blood) should be kept warm and separated at 37 °C in order to avoid the auto adsorption of autoantibodies. A notable finding is the spontaneous occurrence of hemolysis in the blood samples. The interference with immunohematological investigations such as blood group determination, antibody screening, and cross matches is usually less than that seen in warm antibody autoimmune hemolysis. However, the test samples may be washed out as a result of unnoticed hemolysis during automatic washing (e.g., in the indirect antiglobulin test using the tube method).

If autoimmune hemolysis is suspected on clinical grounds (hemolytic uremic syndrome, hemoglobinuria), the following diagnostic approach is recommended:

  • Direct anti-globulin test using the tube method, after cold incubation (30 min. at 4 °C) with the patient’s plasma (EDTA plasma), followed by washing with cold saline (4 °C) or, preferably, using the gel centrifugation test after cold incubation with the patient’s plasma. If anti-IgG is used as the antiglobulin serum, the IgG autoantibodies are more likely to be detected directly. The in vivo binding of complement to the erythrocytes of patients is detected using anti-C3d. The test is only valuable, however, if no interfering cold agglutinins are present. In this case, an attempt can be made to detect the IgG antibodies at a higher incubation temperature. However, the chance of detecting IgG, in addition to complement, on the patient’s erythrocytes decreases at higher temperatures.
  • In the case of a positive direct antiglobulin test with only complement on the patient’s erythrocytes or if autoimmune hemolysis of the Donath-Landsteiner type is suspected clinically (even in the case of a negative direct antiglobulin test), the biphasic cold hemolysis test with untreated and enzyme treated erythrocytes should always be performed.


Even if the antibody specificities are not taken into consideration and antigen positive erythrocytes are used, problems do not generally arise provided transfused units are always preheated. Sialidase induced hemolysis

During bacterial infections (e.g. with Streptococcus pneumoniae or E. coli) the action of bacterial sialidases (neuraminidases) or proteases may result in the exposure of crypt antigens to which humans develop natural antibodies during early childhood. The accelerated elimination of erythrocytes that is sometimes observed in such situations is, however, more likely to be due to enzymatic damage to erythrocytes than to antigen-antibody reactions.

When monoclonal test reagents free of anti T are used, this problem is found less frequently. It should, however, be taken into consideration if the patient erythrocytes agglutinate in a control test containing AB serum or if only complement is attached to them.

The corresponding erythrocyte abnormality can be detected easily using lectins. If anemia of unclear etiology develops during the course of bacterial infections, especially in children, this type of hemolysis should always be considered.

27.5.4 Drug induced immune hemolysis

The frequency of immune hemolysis induced by therapeutic drugs is estimated to be one case per million of the population or approximately 12% of cases of immune hemolysis (not including alloimmune hemolysis) /465152/. The number of drugs responsible has increased in the past 20 years to more than 125 /51/.


Although the underlying pathogenic mechanisms have not been fully clarified, at least three can be distinguished /4651/:

  • Autoimmune mechanism: certain drugs (e.g. α-methyldopa, fludarabine) induce the synthesis of incomplete warm autoantibodies that are not distinguishable from the autoantibodies present in warm antibody autoimmune hemolysis
  • Adsorption mechanism (hapten mechanism): certain drugs, especially penicillins and cephalosporins, bind covalently as haptens to the erythrocyte membrane. The antibodies directed against these drugs or their metabolites bind to those erythrocytes coated with the adsorbed drug.
  • Immune complex (neoantigen) mechanism: for the majority of drugs that cause immune hemolysis, this is presumed to be the underlying mechanism. According to this hypothesis, the drugs and their metabolites (sulfonamides and nonsteroidal anti-inflammatory drugs) bind non covalently to erythrocyte proteins to produce immunogenic structures known as neoantigens that induce the production of antibodies. The drugs are bound loosely and can be removed from the erythrocytes by washing. The antibodies are bound to the erythrocyte membrane mainly via their Fab domain and in part have defined blood group specificities. The immune complexes thus formed often activate complement.

In addition, a number of drugs such as cisplatin and β-lactamase inhibitors may cause a positive direct antiglobulin test due to non immunological adsorption of immunoglobulins as a result of alterations to the erythrocyte membrane. In most cases, this is clinically irrelevant. However, sporadic cases in which hemolytic anemia was triggered have also been observed /51/.

Because a number of drugs can simultaneously induce the production of antibodies that exert their effects via different mechanisms, models also exist that combine all three pathogenic mechanisms /51/. Tab. 27-21 – Therapeutic drugs that can induce immune hemolysis contains a selection of some of the drugs involved that are still used in Germany today. Drug induced hemolysis due to autoimmune mechanisms


Depending on the dose, up to 15% of patients taking α-methyldopa produce erythrocyte autoantibodies. Immune hemolysis, however, develops in only 0.5–1% of patients. For a time, the majority of drug induced immune hemolysis was due to α-methyldopa. Because of the decreased use of this drug as a hypertensive agent, this form of drug induced immune hemolysis is now rare. Nowadays, this type of immune hemolysis is mainly due to the cytostatic drug fludarabine. After dis continuation of the drug, clinical symptoms and antibodies generally disappear within 1–2 months.

Clinical symptoms

Clinical symptoms do not differ from the symptoms of warm antibody autoimmune hemolysis. However, the course tends to be rather mild.

Immunohematological findings

The antibodies are incomplete warm autoantibodies (IgG) that are almost exclusively directed against Rh structures and show hardly any complement activating capacity.

IgG antibodies, and rarely also complement, are attached to the patient’s erythrocytes. The antibodies can be eluted from the patient’s erythrocytes and are detectable in the eluate without the presence of the drug. Diagnostic tests

Refer to Section – Immunohematological findings about diagnosing warm antibody autoimmune hemolysis. In general, if warm autoantibodies of the IgG type are detected, the possibility of a drug induced immune hemolysis must be considered and confirmed or excluded based on the medication history.


The same rules apply as for warm antibody autoimmune hemolysis. Drug induced immune hemolysis due to adsorption mechanisms


Cephalosporins and derivatives of penicillin may bind to proteins on erythrocytes; this adsorption is completely nonspecific and without any clinical relevance. Approximately 3% of patients who are treated with high doses of i.v. cephalosporins or penicillins do, however, synthesize specific antibodies to the erythrocyte bound drugs. Most of these antibodies belong to the IgM class and have no hemolytic activity. In a small number of drugs (e.g., cefotetan, ceftriaxone, and piperacillin in particular), especially at high doses, IgG antibodies occur in conjunction with extravascular immune hemolysis. These antibodies and penicillin allergy are not related to each other because penicillin allergy is caused by IgE antibodies.

Clinical symptoms

Usually mild extravascular hemolysis that is reversible within hours of discontinuing the drug. Immunohematological findings

Incomplete antibodies are detectable only in the presence of the offending drugs or their metabolites (after their intake or their in vitro addition). They usually do not activate complement. Therefore, only IgG is usually attached to the erythrocytes of patients. The antibodies can be detected in the eluate only after these drugs have been added or have previously been attached to the test erythrocytes. Diagnostic tests

The following workup is recommended:

  • Antibody screening in the indirect antiglobulin test and/or enzyme test (gel centrifugation test) using the patient’s serum after pretreating the test erythrocytes with the drug. The auto control is the test most likely to be positive.
  • Drugs that are already available in dissolved form (drops, ampoules) should be used where possible and adjusted to match the expected plasma concentrations during high dose therapy. For adsorption to the erythrocytes, a veronal buffer solution with a pH of 9.6 (with 40 mg of the drug/mL) is recommended. For the adsorption of cefalotin, acidic phosphate buffer solutions (30 mg/mL, pH 6.0) should be used in order to avoid nonspecific globulin binding.
  • Direct anti-globulin test with at least two polyspecific antiglobulin reagents and, if positive, with mono specific sera in order to detect the attachment of IgG to the patient’s erythrocytes.
  • Acid eluate of the patient’s erythrocytes and testing in the antibody screen using erythrocytes that have been pretreated with the drug (as described above).


The transfusion of blood components is not a problem as long as any allo antibodies can be distinguished from the drug-induced antibodies and are taken into consideration. Drug induced hemolysis due to immune complex mechanisms


The number of therapeutic drugs that may cause immune hemolysis by this method is large. However, the reported number of cases that have been attributed to various drugs is low. Tab. 27-21 – Therapeutic drugs that can induce immune hemolysis therefore lists only the most important drugs. Since the detection of antibodies is rarely successful, despite clear clinical indications of their existence, the frequency of this type of immune hemolysis is probably higher than reported by immunohematologists.

Clinical symptoms

Drug induced hemolysis due to an immune complex mechanism often presents as an acute and life threatening intravascular hemolysis; acute renal failure occurs in up to 50% of cases. It is typical of this type of immune hemolysis that the offending drugs have already been administered to the patient before and/or that there has been a drug free interval prior to renewed intake. The severity of hemolysis is largely independent of the dose. Clinical symptoms subside rapidly once the drug has been discontinued. Immunohematological findings

The strong hemolytic activity is due to the fact that the attachment of the immune complexes to the erythrocytes is loose; thus, even a small number of immune complexes are capable of activating complement on many erythrocyte surfaces. Therefore, immunoglobulins (IgG and/or IgM) are only rarely detectable on erythrocytes by the direct antiglobulin test or by elution. Usually, only complement binding is found on the erythrocytes. In fulminant hemolysis, the direct antiglobulin test may even yield a negative result because of the rapid elimination of the erythrocytes involved. Antibodies are detectable only in the presence of the drug and/or its metabolites. Diagnostic tests

The diagnostic spectrum includes the examinations described in Section – Cold agglutinins and monothermal cold hemolysins. However, for the examination of serum/plasm and eluates, the offending drug and/or its metabolites must be added to the test samples at different concentrations. If test methods are used that involve washing the samples (indirect antiglobulin test using the tube method or solid phase technology), the drug must also be added to the wash solution. The drugs are prepared as described (refer to Section – Drug-induced hemolysins). Diagnostic examinations for this type of immune hemolysis should be supplemented by hemolysis tests with the addition of the drug and/or its metabolites

The inclusion of the metabolites in the various test methods may be decisive for detection of the antibodies. If it is medically safe and justifiable, a relative of the patient should therefore be asked to take one dose of the drug; this is followed by collection of blood samples at 1 hour and 6 hours post ingestion (no sooner than 1 hour), in order to obtain sera containing the metabolites. A urine collection obtained 6–12 hours after ingestion can also be used in the same way, if the urine is made isotonic by the addition of hypertonic saline solution.


The suspicion of drug induced immune hemolysis must be raised on clinical grounds. Detailed information concerning the dose and the duration of drug administration as well as the occurrence of the first signs of hemolysis are essential for the immunohematological diagnosis. This allows the offending drugs to be narrowed down and thus prompts focused testing.

In the laboratory, the possibility of drug induced antibodies should be considered if:

  • The immunohematological picture is consistent with autoimmune hemolysis
  • A positive direct antiglobulin test is found in isolation
  • Antibody screening initially yields positive results in almost all tests including the auto control and subsequently (after the drug has been discontinued) only remains positive in the auto control.

The clinician should not stop diagnostic investigations if hemolysis stops after dis continuation of the putative offending drugs (trial withdrawal), because often multiple drugs with different constituents may have been ingested and the causative substance remains unclear. The advantage of immunohematological investigation is that, given a successful diagnostic evaluation with a positive result, the responsible substance and thus a large number of other intolerable drugs can be identified. Targeted diagnostic investigations for drug induced immune hemolysis are only possible in specialized laboratories. Nevertheless, causative antibodies are often not detected, which is why a thorough therapeutic drug history is the most important diagnostic tool.


Refer to Section – Cold agglutinins and monothermal cold hemolysins

27.6 Immunohematological methods

The tests presented in this section are basic immunohematological methods that are performed manually in most cases.

27.6.1 Tube and slide tests

Most of the tube and slide tests, routinely used methods are also derived from these basic procedures. Refer to Section 27.6.2 – Test methods suitable for automation. Agglutination tests


Antibodies cause visible agglutination of erythrocytes in a normal saline solution. Depending on whether antigens or antibodies are used as the known test parameter, the test serves either to detect agglutinating (complete) antibodies (mostly cold antibodies) or antigens.

Test antibodies are used to detect corresponding antigens on erythrocytes whereas test erythrocytes with known antigen patterns are used to detect antibodies in serum or plasma.


Serum or plasma and/or antibody-containing test reagents; test erythrocytes or the patient’s own erythrocytes (at a concentration of 2–10%, depending on the serum volume) in normal saline; single specimen or multiple specimen slides, glass tubes, or micro titer plates; possibly a water bath, warming block or incubator; possibly a centrifuge, illuminated box.

Test protocol

One to three drops (one drop = approximately 50 μL) of the test reagent or serum are mixed with one drop of the test erythrocytes or the patient’s erythrocytes, followed by incubation for 5–120 min. within a temperature range that corresponds to the thermal optimum of the antibodies; in tube tests and micro titer plates, if necessary, low speed centrifugation (1,000 rpm = 120 × g) for a period of 15–30 sec. before macroscopic reading of agglutination over an illuminated box. During this procedure, it is important to dissolve the loose agglutinations that result from centrifugation by gently tilting the tube, without shaking up the specific agglutinations; these need to be read close to the light source.

Microscopic reading should only be performed by experienced examiners under special circumstances (e.g. mixed field agglutination) because of the risk of possible misinterpretation.


Positive factors: optimal temperature range, high antigen density per cell, low cell concentration, high serum volume and/or high antibody content (2–3 drops serum for the detection of antibodies), optimal pH range (mainly between 6.5 and 7.0), thorough mixing of the reagents.

Negative factors: dissolved antigens, blocking antibodies, old test cells, excessively short or long incubation period (optimum usually between 30 and 60 min.).

Antigen testing: the recommendations of the reagent manufacturer should be followed. In general, the best results are obtained by using washed erythrocytes in as fresh a state as possible from citrated blood or EDTA blood in a dilute suspension (concentration 2–5%). Erythrocyte suspensions and test antibody reagents are usually employed in a ratio of 1 : 1 and a shorter incubation time (5–15 min.) is chosen.

Antibody detection: serum and test erythrocyte suspensions are often used in a ratio of 2 : 1 (antibody excess) and the incubation time is longer (15–120 min.) to increase the sensitivity.


Antigen testing: determination of the blood group antigens of the ABO, Rh, and MN systems as well as other blood group antigens (e.g., Lea, Leb, P1, or crypt antigens) with agglutinating sera, lectins, and/or monoclonal reagents. The ever increasing spectrum of monoclonal test antibodies means that more and more antigens can be determined quickly and easily using the agglutination test (e.g., Jka, Jkb, and K).

Controls: in order to rule out pseudoagglutination, poly agglutination, and autoagglutination, auto controls of the erythrocytes under examination against the antibody reagent medium or AB serum always need to be run using the same method. In addition, with the exception of ABO and Rh determination, the test antibodies need to be checked regularly using antigen positive (heterozygous where possible) and antigen negative test cells (positive and negative controls).

Antibody detection: reverse typing of ABO grouping, ABO cross matching, detection of cold agglutinins (cold agglutinin test and cold agglutinin titer), determination of the thermal amplitude of cold agglutinins. Lateral flow technology

Lateral flow technology is a special procedure for blood group determination. In this slide method, monoclonal antibodies directed against different blood group antigens are applied along a strip. Erythrocytes added to the strip (in a 10% suspension) bind within a few minutes as a result of specific antigen-antibody reactions. Unbound erythrocytes are rinsed off using a few drops of diluent. After five minutes at the latest, positive tests due to erythrocyte binding are clearly identifiable as red bands. Because of the specificity of binding, erroneous determinations due to nonspecific interference factors and autoantibodies are unlikely. Therefore, isoagglutinin detection (reverse typing) is no longer necessary when this method is used for AB0 determination. This test is therefore particularly suitable for determining AB0 and Rh blood groups in emergency situations. Supplement tests


Antibodies cause visible erythrocyte agglutination in the presence of auxiliary (supplementary) media. Depending on whether antigens (test erythrocytes) or antibodies (test sera) are used as the known test parameter, the test serves either to detect incomplete antibodies (mostly warm antibodies) or antigens. The detection limit of these test methods for agglutinating antibodies is often higher than that of the simple agglutination test.


Bovine albumin (20–30% or polymerized), modified low ionic strength solution (LISS), enzymes (e.g., bromelin, papain, sialidase), polybrene, colloids (refer to Section – Agglutination tests).

Test protocol

Supplement tests are usually conducted as single stage tests (with the exception of the papain and sialidase tests). First, erythrocytes and serum/plasma are mixed as for the agglutination test (refer to Section Supplementary media (1–4 drops) are then added to the test before incubation.

Albumin test: two drops of albumin are added to the agglutination test, followed by incubation at 37 °C for 30 min. For the remaining steps of the test protocol, refer to Section

LISS test: depending on the ionic strength of the solution, 2–4 drops of LISS are added to the agglutination test, followed by incubation for 5–30 min. For the remaining steps of the test protocol, refer to Section

Single-stage enzyme test: one to two drops of an enzyme (often bromelin) are added to the agglutination test, followed by incubation at 37 °C for 10–30 min., depending on the activity of the enzyme used (usually, 15 min.). The enzyme activity can vary in different batches. For the remaining steps of the test protocol, refer to Section Incubation at lower temperatures is only suitable for special indications since commonly occurring cold antibodies display very wide thermal amplitudes in the enzyme test.

Two-stage enzyme test: under routine conditions, this test is only recommended if commercially produced enzyme-treated test erythrocytes are used. The enzyme treatment of test cells for agglutination tests (supplement tests) depends on their proposed use and requires extensive monitoring. More gentle treatment is required for the destruction of protease-labile antigens, e.g. Fya/Fyb, that are used to differentiate antibody mixtures than for the exposure of T antigen (control cells for examinations using lectins). In the former case, antigen destruction must be verified using a corresponding test serum, e.g. anti-Duffy, and poly agglutination as a result of T antigen exposure must be excluded using lectins or AB serum (contains anti-T). Because the enzyme activity varies among different commercial products, a general recommendation cannot be made regarding the enzyme treatment of erythrocytes. Gentle pretreatment with commercially available enzymes can be attempted by mixing two volumes of dry cell sediment with one volume of enzyme, warming the mixture in a water bath at 37 °C for 1 min., and immediately washing the erythrocytes three times with cold saline. The success of enzyme treatment must be assessed using the checks described above before the cells are used in the test.

Like the agglutination test, the two-stage enzyme test is performed at 37 °C (refer to Section – Agglutination tests).

Polybrene test: due to its complexity, the polybrene test is no longer used, especially since its analytical sensitivity is inferior to that of modern methods such as gel centrifugation.


Interference factors: refer to Section Specific negative factors include: nonadherence to exact volume ratios, direct addition of the medium to the erythrocytes, significant difference between the temperature of reagents and the incubation temperature for the test immediately before the reagents are added to the mixture (especially cold reagents, e.g. those taken directly from the refrigerator).


Antigen testing: Since the introduction of monoclonal test reagents, albumin tests are no longer needed for antigen testing, e.g. for the Rh genotype.

Antibody detection: The following methods are used:

  • Albumin test: This is employed as a supplementary test for the detection and determination of the thermal amplitude of cold antibodies. Otherwise, albumin and LISS tests are only used in combination with the indirect antiglobulin test.
  • Single and two-stage enzyme tests: At 37 °C, their sensitivity for detecting Rh antibodies is especially high; these tests are recommended as an addition to normal antibody screening. Because of the relatively high frequency of nonspecific reactions, enzyme tests must not be used uncritically. Appropriate controls must always be run simultaneously.

Controls: In general, positive and negative controls as well as auto-controls must be run simultaneously using the same method. This applies to:

  • Antigen testing: The controls consist of, if possible, heterozygous antigen-positive (positive control) and homozygous antigen-negative (negative control) test cells mixed with the test antibody as well as the patient’s erythrocytes mixed with the test antibody medium or AB serum (auto-control).
  • Antibody detection procedures: In all test methods used for the detection of antibodies (including cross match), diluted test antibodies are checked against antigen-positive and antigen-negative test cells. The auto-control is performed with the patient’s serum and erythrocytes using the same test procedure. Direct antiglobulin test


Antihuman globulin (antiglobulin serum) agglutinates erythrocytes that are coated with the corresponding globulins. Depending on the specificity of the antiglobulin serum (poly specific or mono specific), the bound globulins can be specified further e.g., IgG, IgA, IgM, C3d, C4d. By using EDTA blood or citrated blood, the in vitro activation of complement e.g., by cold autoantibodies, is inhibited. Thorough washing of the erythrocytes being examined removes soluble and non specifically attached globulins, which allows the detection of immunoglobulins and complement factors that are specifically bound to erythrocytes in vivo.


Erythrocytes obtained from clotted blood, or preferably from EDTA blood or citrated blood, are washed four to six times using a normal saline solution or a phosphate buffer, followed by the preparation of a 2–3% saline suspension, at least two poly specific antiglobulin reagents (containing at least anti-IgG and anti-C3d), mono specific antiglobulin reagents (at least anti-IgG and anti-C3d), test tubes or slides, centrifuge, illuminated box.

Test protocol

Screening test: the direct antiglobulin test is performed using at least two undiluted poly specific antiglobulin reagents (1–2 drops of antiglobulin reagent are mixed with 1 drop of erythrocyte suspension), followed by incubation at room temperature for 15 min. and centrifugation at 120 × g for 15–30 sec.; the result is read over the light source (refer to Section – Agglutination tests). The analytical sensitivity of the test can be improved by reading the result again after 1 h at room temperature, after repeated centrifugation.

Differentiation: a positive direct antiglobulin test can be differentiated further using mono specific antiglobulin reagents.

Controls: in general, positive and negative controls must be run simultaneously. The positive control uses antibody coated test cells (e.g., Coombs control cells) mixed with poly specific antiglobulin reagents and/or anti-IgG; the negative control uses untreated test cells.


Methodological aspects: prolonged incubation may significantly enhance the detection of complement on the erythrocytes. A single washing with normal saline (including centrifugation at 1,000 × g for two minutes) after the second reading may enhance the reaction and prevent the occurrence of prozone phenomena. The use of a phosphate buffered wash solution (pH 7.0) with EDTA and bovine albumin as additives prevents the elution of specific but loosely bound antibodies as well as the nonspecific activation of complement. The tube test is clearly superior to the slide test because it includes centrifugation.

Inadequate cell washing and a high immunoglobulin concentration in the patient serum have an unfavorable impact on the test.


The direct antiglobulin test is diagnostically useful in the following clinical settings:

  • Investigation of transfusion reactions
  • Autoimmune hemolysis
  • Hemolytic disease of the newborn
  • Drug induced immune hemolysis
  • Positive auto controls as part of blood grouping, antibody screening, and cross matches.

Differential diagnosis of a positive direct antiglobulin test

In the case of a positive direct antiglobulin test, the following possibilities need to be considered:

  • Antepartum Rh immune globulin prophylaxis in the mother if a newborn is Rh positive
  • Benign warm autoantibodies without hemolytic activity
  • Modification of the erythrocyte membrane (e.g., drug-induced; cephalosporins), with subsequent nonspecific globulin adsorption
  • Nonspecific immunoglobulin adsorption in conjunction with hypergammaglobulinemia, paraproteinemia
  • In vitro complement activation due to benign cold autoantibodies (e.g., in clotted blood after cooling)
  • Poly agglutination if crypt antigens are exposed and antiglobulin reagents are contaminated with anti-T. Indirect antiglobulin test


The indirect antiglobulin test serves to detect serum antibodies that specifically bind to erythrocytes in vitro. In general, the poly specific antiglobulin reagents used detect the in vitro bound antibodies (IgG, IgM) not only directly, but also via their complement activation (C3d). By using reagents with defined antibody specificity (e.g., test sera) the indirect antiglobulin test allows specific and sensitive antigen testing.


Poly specific antiglobulin reagent (anti-IgG, anti-C3c, anti-C3d; Coombs serum), wash centrifuge, wash solution (normal saline solution, phosphate buffered saline solution), Coombs control cells, further materials, depending on the selected procedure. Refer to:

Test protocol

The indirect antiglobulin test is performed as a tube test following agglutination and supplement tests (usually albumin test or LISS test at 37 °C) (Fig. 27-3 – Performance of the indirect antiglobulin test for the detection of serum antibodies (tube test)). Interim reading of the tests (and centrifugation) before the indirect antiglobulin test are usually unnecessary. Following incubation, the test samples are washed thoroughly with wash solution. The supernatant is then removed by aspiration or decantation in order to provide a dry button of cells. Then 1–2 drops of antiglobulin serum are added, followed by low speed centrifugation (15–30 sec. at 120 × g). The results are read carefully over the light source. Thorough washing means that 3 mL of wash solution is used for each of at least three cycles with complete erythrocyte resuspension after each wash cycle and that the wash solution is removed completely after each cycle. If the indirect antiglobulin test is negative, a Coombs control is subsequently performed with antibody coated test cells. If strongly coated erythrocytes are added (one drop), the test must turn positive without centrifugation within 5 min. If weakly coated erythrocytes are used, a soft spin (15–30 sec. at 120 × g) is necessary.


Controls must be run simultaneously for each batch in the tube test as described in Section – Agglutination tests. Because of the sensitivity of the washing process, controls are very important.


For interference factors, refer to Section – Agglutination tests. IgM and weakly binding, complement-activating antibodies are detectable in the indirect antiglobulin test only via the detection of complement. Complement activation only occurs in vitro if serum (clotted blood) rather than plasma is used.


Antigen testing: most erythrocyte antigens (e.g., Fya, Fyb, K, k, S, s, Jka, and Jkb) as well as rare antigens such as Lua, Wra, and Kpa and high frequency antigens such as Coa, Jsb, and Kpb, and weak antigen variants (weak D, DVI in particular) can be determined reliably by means of the indirect globulin test using the saline technique and tube method. Nowadays, however, this method is increasingly being replaced by gel centrifugation tests and solid-phase assays (refer to Section – Micro column agglutination test). If an initial direct antiglobulin test and/or auto control in these more sensitive test methods is positive, it is sometimes possible to perform an antigen determination subsequently using the less sensitive tube method. The result is significant only if the simultaneously run auto control is negative.

Antibody detection: the albumin technique (with 30 min. incubation at 37 °C) is the most sensitive of the test tube methods. It can reliably detect clinically relevant irregular antibodies that can cause acute hemolytic transfusion reactions both in the cross match and (if the corresponding test cell antigen pattern is used) in antibody screening. In addition, interference due to nonspecific agglutinations (pseudoagglutinations) and clinically irrelevant autoantibodies is rare.

The test tube method with LISS detects Rh antibodies more sensitively, but occasionally fails to detect anti-K. It is suitable for cross matching in emergency situations because of the rapid antibody binding and short incubation time (5–10 min.).

Tube tests have largely been replaced by significantly more sensitive methods for detecting antibodies, such as gel centrifugation tests and solid phase technology (refer to Section 27.6.2 – Test methods suitable for automation). However, these newer methods are far more susceptible to interference from nonspecific factors, cold antibodies, and clinically irrelevant warm autoantibodies. The indirect antiglobulin test using the albumin technique continues to be the decisive reference method for evaluating the clinical relevance of the antibodies detected by these methods. Furthermore, it should continue to be used as the standard test for antibody titration. Refer to Section – Supplement tests.

Methods that employ enzyme treated erythrocytes are generally more sensitive but also significantly more susceptible to interference factors. Furthermore, they fail to detect antibodies that are directed against enzyme sensitive antigens (Tab. 27-4 – Impact of enzymes on blood group antigens). Therefore, test methods that involve enzyme treated erythrocytes or the addition of enzymes can only be used as supplementary procedures /53/. Hemolysis methods


In the presence of complement, various antibodies hemolyze untreated and/or enzyme treated erythrocytes that carry the corresponding blood group antigens. The optimal reactivity of the different antibodies varies with regard to pH and temperature.

A number of allo antibodies are capable of causing in vitro hemolysis (ABO antibodies, some Lewis and Kidd antibodies). Since no clear relationship has been found to exist between hemolytic activity in vitro and in vivo (except in the case of ABO antibodies), hemolysis tests are not diagnostically useful in the case of allo antibodies. On the contrary, the occurrence of hemolysis in blood grouping, antibody screening, and cross match can lead to misinterpretation due to the erroneous reading of positive tests as negative.

Therefore, the test conditions for these procedures should be chosen in such a way that antigen-antibody reactions are detectable by agglutination.

The isoagglutinin concentration instead of the isohemolysin titer should be used to confirm that blood units are free of hemolysins. The variability of the hemolysis test, especially due to differences in serum complement activity, means that reliable results cannot be guaranteed. On the other hand, the isohemolysin titer correlates closely with the isoagglutinin titer (refer to Section – A and B phenotypes).


Patient serum; 0.1 mol/L HCl; 0.1 mol/L NaOH; fresh or freshly frozen AB serum; bromelin; fresh O Rh-negative and P1-positive test erythrocytes; possibly fresh A test cells; auto hemolysin containing control serum (possibly isohemolysin-containing 0 serum); normal saline or phosphate buffer solution; if indicated, therapeutic drugs and veronal buffer (refer to Section – Drug-induced hemolysins); pH meter.


The hemolytic complement activity of the patient serum is usually unknown. In the presence of immune hemolysis, it may be markedly reduced. Therefore, the addition of fresh AB serum is generally recommended as a source of complement.

Since the auto hemolysins associated with the different types of immune hemolysis do not react specifically, specification can sometimes only be accomplished by comparative examination of the different methods.

A conclusion regarding the type of hemolysis can be drawn based on the test method with the most strongly positive result. Accordingly, the different tests for detecting auto hemolysins should always be run in parallel using the same test material, including the test cells.

Since bithermal cold hemolysins react most strongly with P1s cells, the test cells used should in general have this specificity.

Under certain circumstances, the differentiation between, for example, bithermal and mono thermal cold hemolysins can only be achieved by determining their immunoglobulin class, by pretreating the serum with dithiothreitol or adsorbing the antibodies by means of affinity chromatography.

It is only necessary to run positive controls in one hemolysis test method. Warm hemolysins

Method of determination

Preparation: the test erythrocytes are divided into two portions after thorough washing (e.g., 500 μL of sediment is washed three times using 10 mL of normal saline for each wash cycle). The sediment of one portion is mixed with an identical volume of bromelin, followed by incubation at 37 °C for 5 min. The erythrocytes are then washed three times using cold (4 °C) normal saline. Treated and untreated erythrocytes are suspended at a ratio of 1 : 1 in normal saline or phosphate buffer solution (50% erythrocyte suspension) before they are added to the test.  The suspension can only be used on the day of its production and must not show spontaneous hemolysis. In the event of spontaneous hemolysis, the treatment time or the bromelin concentration needs to be reduced.

The patient’s serum is mixed with AB serum in a ratio of 1 : 1. By using 0.1 mol/L HCl (and 0.1 mol/L NaOH), the pH of the serum mixture is carefully adjusted to 6.5 (the acidic or alkaline solution is allowed to run down the wall of the glass tube).

Test method: 100 μL of serum (1 : 1 mixture of AB serum and the patient’s serum, with a pH of 6.5) is added to each of two Coombs tubes. 20 μL of enzyme treated and/or untreated erythrocyte suspension is added to each tube. This is followed by incubation at 40 °C in a water bath for 60 min. The test tubes need to be thoroughly agitated every 15 min. and at the end of the incubation period. Next, the tubes are centrifuged for 2 min. at 1,000 × g. In front of a white sheet of paper, the supernatant (without prior agitation) is assessed for the presence of hemolysis; in each case, this is compared to the corresponding negative control.

Controls: for the negative controls, two serum mixtures (each 1 : 1 AB serum and patient serum, with a pH of 6.5) are inactivated at 56 °C for 15 min. prior to the addition of enzyme treated or untreated erythrocytes. Mixtures with auto hemolysin containing control serum and enzyme treated erythrocytes should be run simultaneously as positive controls.

If no auto hemolysin containing control serum is available, controls should be run using isohemolysin-containing 0 serum with untreated test cells of blood group A; this allows at least the complement activity of the AB serum to be checked. Cold hemolysins

Test preparation and test mixtures are as described in Section The tubes are then placed in a water bath at 37 °C for 20 sec., during which time they are agitated. This is followed by incubation at room temperature (22 °C) for 2 h. The positive controls can be limited to tests with auto hemolysin containing serum and untreated erythrocytes. For further methodological details, refer to Section – Warm hemolysins. Bithermal cold hemolysins


Preparation: as described in Section, but the pH of the serum needs to be set to 7.4.

Test method: as described in Section, except for the following: the test tubes are initially incubated in iced water for 30 min. and then quickly transferred to a water bath at 40 °C, where they remain for another 60 min. after thorough agitation. Throughout the incubation periods, the test tubes need to be agitated at regular intervals (refer to Section – Warm hemolysins). Drug-induced hemolysins

Preparation: intravenously and intramuscularly administered drugs are added to the test mixtures in an undiluted as well as a diluted form i.e., at ratios of 1 : 5 and 1 : 10 in a phosphate buffer solution. Orally administered drugs are used in the form of drops, if available. Tablets must be crushed in a mortar. Drops and pulverized tablets are dissolved in phosphate buffer solution, using heating if necessary. If the drug is not soluble, veronal buffer (pH 7.5) is used. The final drug concentration should correspond to the in vivo concentration. For many drugs, a concentration of approximately 1 mg/mL is adequate. Further dilution in phosphate buffer solution may be necessary in order to prevent hemolysis due to a direct, toxic effect exerted by the drug (negative controls). Undissolved drug residues are removed by filtration.

Enzyme-treated and untreated erythrocytes are incubated with the appropriate drug solutions at 37 °C for 30–60 min., during which time they are agitated at regular intervals. They are then washed three times with phosphate buffer solution at a pH of 7.4 and added to the test as a 50% suspension.

The pH of the AB serum and the patient’s serum is adjusted to 7.0–7.4.

Test method: Hemolysis tests are performed as described for warm hemolysins (Section, with the addition of 20 μL of solution containing the drug or the use of drug coated erythrocytes. If these test mixtures yield a negative result even though there is a strong clinical suspicion of drug induced immune hemolytic anemia, 50 μL of serum or isotonic urine samples obtained from healthy volunteers after drug exposure may be used instead of the drug solution (refer to Section – Mixed autoimmune hemolysis).

Controls: in addition to the controls listed in Section – Warm hemolysins, test mixtures must be run simultaneously as controls in which the drugs are tested with AB serum only instead of the patient’s serum in order to detect a direct toxic effect of the drugs on the erythrocytes. If these negative controls yield a positive result, the drug concentration needs to be reduced. Titration of hemolysins

Titration of hemolysins may be necessary to exclude prozone phenomena, which can occur as a result of anti- complement activity in serum (e.g., as observed after freezing). Furthermore, comparative titration between different methods makes it easier to determine the reaction optimum of hemolysins, thus contributing to their specification.

100 μL of fresh AB serum (pH adjusted according to the method used) is added to each test tube. A geometric series of dilutions is produced by adding 100 μL of patient serum (pH adjusted) to the first tube; after thorough mixing, 100 μL of serum is removed again and added to the second tube, and so on. The remaining methodological details correspond to the specifications provided for each method in Sections to

27.6.2 Test methods suitable for automation

It has been possible since the 1960s to automate blood grouping and antibody screening by means of continuous flow procedures using LISS and polybrene and/or enzymes combined with polymers /5455/. These procedures were not suitable for hospital laboratories and detected numerous unidentifiable and clinically irrelevant antibodies.

Newer procedures using cards or micro titer plates are more reliable, require fewer follow up examinations, and detect clinically relevant antibodies with a higher detection limit than the classic manual methods. These tests can also still be performed manually (refer to Section 27.6.6 – Automation of blood group determination and antibody screening). Micro column agglutination test


Plastic cards, each composed of 6 or 8 micro tubes, are used. Each micro tube contains an upper reaction chamber into which sample material and reagents are added and in which any required incubation takes place. Antibody binding, agglutination, and supplement reactions take place here during incubation. The reaction chambers taper to form capillaries filled with sepharose gel. Gel cards can contain neutral gel (neutral cards) or gel that has been impregnated with specific test antibodies (e.g. against blood group antigens or anti-human globulins). During subsequent standardized centrifugation in special centrifuges (specific time and g force), agglutinated erythrocytes are retained in the gel sooner or later (gel filtration), depending on the degree of agglutination, whereas unagglutinated erythrocytes pass through the gel and accumulate at the bottom of the capillaries.

Neutral gel cards are used for antigen testing. In this process, test antibodies and erythrocytes to be examined are added to the reaction chambers, followed by incubation and centrifugation.

If the gel contains test antibodies( e.g., anti-A, anti-B, anti-Rh antibodies, or anti-K) antigen-positive erythrocytes are agglutinated during centrifugation and retained in the gel (Fig. 27-4 – Principle of the gel centrifugation test).

If the gel contains antiglobulin serum (Coombs or AHG cards), erythrocytes to which incomplete antibodies and/or complement are attached are agglutinated during centrifugation and retained in the gel. Unbound antibodies and other plasma components are not relevant here because their penetration into the gel at the selected centrifugation speed is too slow. Therefore, there is no need for the washing that is otherwise necessary in the indirect antiglobulin test. However, hemoglobin molecules, fibrin clots, and large lipid particles may block the gel capillaries if they penetrate the gel during the incubation phase. Erythrocytes may be retained in the gel because of this, which can result in a false positive or unclear test result.

Strong agglutinations (including nonspecific reactions) that occur in the reaction chamber are identifiable by a red line on the gel. Such reactions that are triggered by test antibodies or antiglobulin serum in the gel lead to the formation of a red line below the level of the gel. As the strength of the reaction decreases, the smaller agglutinates penetrate more deeply into the gel and distribute themselves throughout the gel column. Very weak agglutination reactions are only recognizable by indistinct margins and/or tails of erythrocyte sediment. The agglutination reactions remain stable for at least 12–16 h so they do not need to be read immediately.

By now, many manufacturers have produced a range of cards for the gel centrifugation test with different shapes, test antibody sequences, and gel matrices (Sephadex, polyacrylamide, or glass pellets), which in turn affect the density and transparency of the gel. As a result, different cell densities and dilution media need to be used and the representation of the reaction differs in terms of the size and location of agglutinations. Furthermore, the antiglobulin reagents used by different manufacturers differ (particularly with respect to the anti-C3d content) so that cards vary, for example, in their ability to detect complement activating antibodies (especially clinically irrelevant cold antibodies). A high anti-C3d content is not necessarily an advantage for the detection of allo antibodies (indirect antiglobulin test), particularly if plasma is used.


Because of the wide variety of commercially available test cards, it is only possible to make some general comments. The manufacturer’s instructions should be followed. Materials are:

  • Patient serum, patient erythrocytes and test erythrocytes in LISS, saline, or bromelin suspension (0.6–5%, depending on the manufacturer)
  • Cards with test antibodies, neutral cards, cards with poly specific and mono specific antiglobulin reagents, cards with anti-IgG, ready to use diluent solutions (LISS, bromelin)
  • Adjustable volume pipettes; incubators; special centrifuges for standardized centrifugation of the cards.


Antigen determination: the simplest approach uses cards in which test antibodies have already been added to the separation medium.

Antigen determination can also be performed using separate test antibodies in neutral cards or Coombs cards, by adding the test antibodies to the cells in the reaction chamber. This approach is particularly likely to be necessary in situations where ready to use cards with antibodies against the antigens to be investigated are not available but the higher detection limit of the gel centrifugation test is required (e.g., in the case of rare antigens or partial antigens).

Appropriate preliminary investigations and test modifications may be necessary in order to exclude the presence of prozone phenomena and other contaminating antibodies /56/.

Antibody detection

Neutral cards are used for the detection of AB0 antibodies as part of reverse typing; if enzymes are added or enzyme treated erythrocytes are used, these cards are also useful for detecting and identifying enzyme reactive antibodies.

For the detection and identification of antibodies in the indirect antiglobulin test, cards containing poly specific antiglobulin serum are usually employed.

For the detection and differentiation of transfusion-relevant allo antibodies from complement activating cold autoantibodies with a wide thermal amplitude, cards are available that contain only antihuman IgG. It is essential, however, that the preincubation of reagents and incubation of test mixtures take place at exactly 37 °C in order to prevent agglutination of the erythrocytes by the cold antibodies prior to centrifugation.

A more detailed differentiation of the bound human globulins as detected by the direct antiglobulin test is possible by using test cards with various mono specific Coombs sera (antihuman IgG, IgA, IgM, C3c, and C3d).

Test protocol

Antigen testing

Three different methodological approaches are recommended:

  • The simplest method uses cards that already contain the relevant monoclonal test antibodies. Erythrocytes suspended in LISS are pipetted into the reaction chambers. The cards are then centrifuged in a special centrifuge without further incubation. The results are read over a light source by visually inspecting the front and back of the cards. This method is available for the determination of ABO, Rh, and MN antigens.
  • Particularly if human (polyclonal) antibodies are used, the erythrocytes must be pretreated with a protease. Unwashed erythrocyte sediment is added to bromelin solution, followed by incubation at room temperature for 10 min. Of this solution, 10 μL is pipetted into the reaction chambers of the test cards that already contain the corresponding test antibodies. Otherwise, the method is the same as that described above. This method can be used to determine the antigens of the AB0, Rh, Kell, Lutheran, Lewis, and Kidd systems.
  • If separate test antibodies are to be used, neutral gel cards can be used for complete antibodies (IgM) and Coombs test cards can be used for incomplete antibodies (IgG). Erythrocytes suspended in LISS (e.g., 50 μL) and the test antibody (e.g. 25 μL) are pipetted into the reaction chambers of the cards. The cards are then incubated for up to 15 min. at the optimal temperature for the test antibody and than centrifuged. Finally, the results are read. Directly agglutinating test antibodies for tests using neutral cards may require the addition of auxiliary media/supplements to produce sufficiently robust agglutinations.

Enzyme test

Test erythrocytes, the serum/plasma to be examined, and bromelin solution are pipetted one after the other into the reaction chambers of neutral gel cards, followed by incubation for 15 min. at 37 °C and centrifugation. The test should not be performed at room temperature as part of antibody screening. Incubation at room temperature is only useful for improved differentiation and/or identification of cold antibodies within the scope of targeted tests (e.g., Lewis antibodies).

The enzyme test is even more sensitive if papain treated test erythrocytes are used. Under these circumstances, the addition of bromelin is no longer necessary. The reagents and instructions for treating test erythrocytes with papain can be obtained from the manufacturer. Enzyme treated fresh cells can be stored for four weeks in a special preservative solution.

Indirect antiglobulin test

Coombs test cards are used. The test is prepared in the same way as the enzyme test, but without the addition of bromelin. The detection limit of the test can be increased by doubling the serum/plasma volume and/or extending the incubation period. If the test is part of cross matching, the donor erythrocytes must be suspended in the card manufacturer’s diluent. Washing the donor erythrocytes once with normal saline can reduce the occurrence of non specific test results and improve the detection limit of the test. The “Coombs control” that is required for the indirect antiglobulin test using the tube method in the case of a negative result is not necessary because, unlike the tube test, the gel method does not produce false negative results due to inadequate washing or an excess of immunoglobulins.

Direct antiglobulin test

50 μL of a suitable erythrocyte suspension is pipetted into the reaction chambers of the cards containing poly specific and mono specific antiglobulin sera; this is followed by immediate centrifugation. As far as the detection of complement binding is concerned, a more sensitive alternative is to use cards with neutral gel and to incubate the erythrocytes with different poly specific and mono specific antiglobulin reagents (50 μL each) prior to centrifugation at room temperature for at least 15 min.

Controls: in the antibody screen gel centrifugation test, it is advisable to simultaneously run an auto control for each centrifuge run. However, if it makes more sense to perform antibody screening with three test populations, the auto control usually has to be omitted due to a lack of space. The auto control is not necessary if the direct antiglobulin test is performed as part of blood group determination. Furthermore, the auto control can be run later if antibody screening is positive.

Because the interference prone washing procedure is omitted, once daily positive and negative controls are sufficient to check the detection limit and test specificity.


Methodologically critical aspects include the transport and storage of the cards, the temperature of the sera, reagents, and incubation, the sequence and placement of reagents during pipetting, and the speed and duration of centrifugation.

The unused gel cards must show a thin layer of fluid on top of the gel; otherwise, a high frequency of nonspecific reactions is likely /13/.

Erythrocytes must first be pipetted into the reaction chamber in a way that prevents the premature diffusion of serum into the gel. Otherwise, the same technical aspects apply as described in Section and Section

Precise adherence to the centrifugation specifications affects the detection limit and specificity. Inadequate centrifugation causes an increase in false positive results, whereas excessive centrifugation has a negative impact on the detection limit. If the card is not maintained strictly in a horizontal plane during centrifugation, the erythrocyte sediment shows indistinct lateral margins and/or a lateral erythrocyte tail. Therefore, the film that seals the gel column must be removed completely or, in the case of unused columns, trimmed accordingly.

In general, unwashed erythrocytes can be used in all tests. In the case of nonspecific or weak reactions, washing the erythrocytes may increase the analytical specificity and sensitivity. It is therefore recommended to wash donor erythrocytes before repeating doubtful positive cross matches. The use of suspensions with a higher concentration of erythrocytes (especially in tests that use glass pellets) also causes an increase in the rate of nonspecific positive results.

The gel centrifugation test is characterized by a significantly higher detection limit than tube tests /13575859/. It allows faster and more objective reading as well as a second reading up to a day later. It also requires less reagent and sample material and results in less waste. The test procedure is easier and can be automated. During the determination of blood group antigens, blood mixing and mixed field agglutination are easily detectable.

The test should be read from both sides.

The disadvantages of the gel centrifugation test depend primarily on the test conditions selected. The high detection limit, also as far as the detection of clinically irrelevant, interfering cold antibodies is concerned, can be circumvented by avoiding incubation temperatures below 37 °C and by preheating blood samples and reagents to at least room temperature prior to testing /13/. Recommendations to perform antibody screening at room temperature or under bithermal conditions as part of blood group determination /60/ do not make sense because, on the one hand, this results in less sensitive detection of clinically relevant allo antibodies while, on the other hand, clinically irrelevant cold antibodies often prompt unnecessary and time consuming follow up examinations.

Because of its high detection limit, the gel centrifugation test consistently detects antibodies without any recognizable specificity or antibodies that, despite having defined specificity, are of questionable clinical relevance. If there is any doubt about the clinical relevance of such antibodies, decisions concerning urgent blood transfusions should be based on the results of the indirect antiglobulin test using the albumin technique (tube test). Based on the many years of experience with tube tests, it is unlikely that antibodies that are not detected by the indirect antiglobulin test using the albumin technique will cause severe acute transfusion reactions. Analysis of the IgG subclasses can contribute to improved assessment of the clinical significance of such antibodies.

When the gel centrifugation method is used, the auto control that is run simultaneously as part of antibody screening often yields positive results. In such cases, the direct antiglobulin test usually indicates the binding of IgG to the patient’s erythrocytes. These findings are of clinical significance only if the patient displays signs of hemolysis and/or anemia of undetermined etiology or if it is suspected on the basis of other findings that erythrocyte allo antibodies or autoantibodies and/or drug-induced antibodies may be present. Without further clinical evidence of immune hemolysis and/or anemia of undetermined etiology, further immunohematological investigations are usually not indicated if the reactions are weak (< 2+) or the direct antiglobulin test using the tube method yields negative results.

Antibody screening should be performed using test erythrocytes from three donors. This makes it easier to detect antibodies that display dosage effects and antibodies directed against rare antigens because test cells that are homozygous for the corresponding antigens and/or that express low incidence antigens such as Cw and Luacan be provided.

Due to the high detection limit of the gel centrifugation test, antigen determinations performed using this method often yield meaningless false positive results if IgG binding is present (positive direct antiglobulin test) or if there is a history of previous transfusions (mixed field agglutination). This difficulty can often be circumvented by using less sensitive tube tests.

Test cells are generally obtained from the manufacturer of the test cards. If test cells from another manufacturer or from blood products are to be used, they must first be suspended in the suspension medium (diluent) provided by the card manufacturer. Micro titer plate tests

Micro titer plates can be used for simple agglutination tests (refer to Section – Agglutination tests) or in solid phase immunoassays.

Simple agglutination tests using micro titer plates are performed as part of blood group determination (antigens, isoagglutinins) and to detect agglutinating antibodies (reverse typing) (Fig. 27-5 – Determination of ABO blood group antigens and ABO antibodies on uncoated micro titer plates).

In solid-phase immunoassays, micro titer plates display a particular capacity for binding cells and antibodies, which can become firmly attached and coat the wells of the plates. Three fundamentally different approaches are possible:

  • Methods that use indicator cells
  • Methods that use protein-A-coated plates
  • Erythro-magnetic technology.

By using monoclonal test antibodies, it is possible to determine, for example, AB0 and Rh antigens under the same test conditions on the same plate. These tests offer the advantage of being amenable to automation, which is useful for large numbers of samples. Solid-phase immunoassay Tests that use indicator cells

The test is used in three ways:

  • The wells of the plates are already coated with antibodies (anti-A, anti-B, anti-D, etc.) for determining blood group antigens and thus bind patient erythrocytes that express the corresponding antigens. Unbound cells are removed by washing.
  • The wells of the plates are pre coated with erythrocyte membranes for antibody screening, antibody differentiation, reverse typing , and thus bind antibodies in the patient serum/plasma that are directed against the corresponding antigens. Unbound globulins are removed by washing).
  • The tester needs to coat the wells of the plates with a single layer of erythrocytes (cross match and antigen determinations with soluble test antibodies). Otherwise, the procedure is as described under point above.

Refer to:

In all three test applications, complexes consisting of erythrocytes or erythrocyte membranes coated with antibodies bind strongly to the bottom of the wells. The indicator cells coated with antihuman IgG that are added at the end attach to the bound IgG antibodies. Following centrifugation, a positive test is indicated by an even distribution of erythrocytes on the bottom of the wells, whereas in negative tests, the indicator erythrocytes accumulate as a small button in the middle of the well floor.


Serum or plasma of the patient; micro titer plates to be coated with test cells or donor cells for cross matching and antigen determinations with soluble test antibodies, test antibodies, test cells, donor cells; micro titer plates coated with dried test erythrocyte membranes for antibody screening and possible detection of isoagglutinins; micro titer plates coated with test antibodies for antigen determination, LISS, indicator erythrocytes carrying antihuman IgG, incubator, centrifuge for micro titer plates, washing device, illuminated box for reading results.


In the case of uncoated plates, 50 μL of erythrocytes in 1–3% saline suspension are first placed into each of the wells. By centrifuging the plates (5 min. at 190 × g), a coating of erythrocytes is produced. In all cases, two drops (approx. 100 μL) of LISS and one drop (approx. 50 μL) of the patient’s serum are pipetted into each test mixture. After incubation for at least 15 min. at 37 °C, the supernatants are decanted or removed by aspiration and the plates are washed six times with normal saline. Then, 50 μL of indicator cells are added to each test mixture, immediately followed by centrifugation of the plates for 3 min. at 370 × g. The results are read over an illuminated box.


For antibody screening and identification, assays using plates that are pre coated with erythrocytes offer a clear advantage. The production of erythrocyte coated plates is cumbersome and the test results are more difficult to assess reliably because of the hemoglobin content of the erythrocytes.

Tests using plates already coated with erythrocyte membranes are significantly more sensitive at detecting IgG antibodies than tube tests (by 3–5 titer dilutions) and even more sensitive than gel centrifugation in the direct antiglobulin test (by approx. 1 titer dilution) /61/. Such tests are possibly even more sensitive than other solid phase immunoassays /62/.

In contrast to other solid phase immunoassays and tube tests, assays using plates that are pre coated with erythrocyte membranes do not detect IgA or IgM antibodies or complement activation. This represents a problem for cross matching with regard to ABO antibodies. Additional tests are therefore required to ensure AB0 compatibility (refer to Section 27.6.7 – Cross match). Due to the fact that these solid phase assays sometimes fail to detect antibodies that are of definite clinical relevance (e.g. several complement-activating anti-Fya antibodies of the IgM class) they cannot be recommended as the sole test for conducting immunohematological investigations in transfusion recipients /61/. On the other hand, this method is significantly better than any other at detecting anti-Jk.

Clinically irrelevant allo antibodies and autoantibodies that are not identifiable under routine conditions but that react with all test cells are detected even more frequently by solid phase immunoassays than by the gel centrifugation test (refer also to Section – Micro column agglutination test).

With regard to the immunohematological investigations for transfusion recipients, another disadvantage of solid phase immunoassays is that the auto controls do not yield directly comparable results and involve a greater amount of effort. They can therefore be omitted.

Nevertheless, solid phase immunoassays are widely employed in routine investigations. Automated procedures in particular significantly reduce the effort involved and yield reliable results when performed correctly and in a controlled manner. Tests that use micro titer plates coated with protein A

The plates are coated with protein A, which selectively (but non specifically) binds IgG . Test erythrocytes carrying irregular antibodies (IgG, IgA, IgM) and C3d bind antihuman globulin serum (of class IgG) that couples the erythrocytes non specifically to protein A via IgG.

Refer to Fig. 27-8 – Determination of irregular erythrocyte antibodies using protein A-coated test plates


Patient serum and erythrocytes, test plates coated with protein A, papain treated or untreated test or donor erythrocytes in Alsever’s solution, LISS, antihuman globulin serum (modified poly specific antiglobulin with monoclonal anti-C3d and anti-C3b) or modified IgG alone, phosphate buffer, agitator, incubator, washing device, centrifuge for micro titer plates, illuminated box.


Initially, 50 μL of serum/plasma and 50 μL of test erythrocytes or donor erythrocytes as well as the patient’s erythrocytes (auto control) as a 1% LISS suspension are pipetted into the wells; this is followed by mixing and incubation for 20 min. at 37 °C. Alternatively, papain treated cells in Alsever’s solution can be used. Next, the plates are centrifuged (3 min. at 1500 × g) and immediately washed five times with phosphate buffer. 100 μL of antihuman globulin serum is then added and the erythrocytes are completely resuspended by using the agitator (2–3 min. at high speed). Finally, the plates are centrifuged for 3 min. at 35–40 μg. Antiglobulin coated erythrocytes bind evenly to the protein A on the plate while uncoated erythrocytes accumulate on the floor of the wells. The results are read over an illuminated box. Positive reactions are characterized by an even distribution of cells whereas negative results are indicated by a small button of erythrocytes (Fig. 27-8 – Determination of irregular erythrocyte antibodies using protein A-coated test plates).


The solid phase immunoassay is a highly sensitive method for detecting and identifying antibodies to erythrocytes. Depending on the antihuman globulin serum used, it detects IgG, IgA, and IgM antibodies as well as C3d or antibodies from the different immunoglobulin classes only.

The use of papain treated cells increases the detection limit to Rh antibodies. However, the number of non specific results also increases.

If papain treated cells are used, this test can only be used in combination with a second test system because, otherwise, antibodies to enzyme sensitive antigens are not detectable. Refer to Tab. 27-4 – Impact of enzymes on blood group antigens.

If the procedure is automated, the determination of blood group antigens is performed using simple agglutination and supplement tests. Plates already coated with dried test antibodies or uncoated plates into which all reagents must be pipetted are used for this. Erythro-magnetic technology

Erythro-magnetic technology involves the use of magnetism to attract erythrocytes to the bottom of the plate, thereby eliminating the need for centrifugation. The erythrocytes are pre magnetized by adding iron chloride solution to the erythrocyte suspension. A magnetic force is applied that draws magnetized patient erythrocytes that have been incubated with dried test antibodies (antigen determination) or magnetized test erythrocytes that have been incubated with patient serum/plasma (reverse typing) to the base of the plate. Following a final agitation, agglutinated erythrocytes form a central button whereas non agglutinated erythrocytes are distributed evenly across the floor of the well. Refer to Fig. 27-5 – Determination of AB0 blood group antigens and antibodies.

Micro titer plates that have been coated with antihuman IgG and pre magnetized test erythrocytes (antibody screening) or donor/patient erythrocytes to which iron chloride solution has been added (cross match, testing for rare antigens) are used for the detection of irregular antibodies (antibody screening, cross match) and the determination of rare antigens. A high-density solution is first added to the wells, followed by a diluent (screen solution), patient serum/plasma, and test erythrocytes. The high-density solution prevents a reaction between antihuman IgG and IgG antibodies in the patient sample. Following incubation, a magnetic force is used to draw the erythrocytes to the base of the plate. A positive result is indicated by an even distribution of antibody-coated erythrocytes on the bottom of the wells, whereas, in negative tests, non agglutinated erythrocytes accumulate as a button on the well floor. The results are represented in Fig. 27-8 – Determination of irregular erythrocyte antibodies using protein A-coated test plates.

Like other micro titer plate tests, assays that use erythro-magnetic technology can also be automated. Another advantage of this method is that it eliminates the need for centrifugation. It is therefore less dependent on technique. Initial studies of this method show promising results /636465/.

27.6.3 Blood grouping and antibody screening

Blood grouping usually includes the determination of ABO antigens and the Rh factor D. It is generally supplemented by antibody screening for irregular antibodies to erythrocytes.

Prior to any invasive or surgical procedure potentially associated with bleeding complications that may require transfusion, a blood group and a current antibody screening result from the laboratory in charge must be available. In addition, blood grouping and antibody screening are part of the antepartum and postnatal immunohematological monitoring of mother and child. Refer to Section – Postpartum diagnostic tests).

The ABO blood group, Rh genotype, and Kell antigen must be determined in all blood donors and indicated on the labels of whole blood and red cell products. Furthermore, regular antibody screening must be performed in blood donors to ensure that blood components containing plasma are free of transfusion-relevant irregular antibodies to erythrocytes. ABO determination

Test spectrum

A complete AB0 grouping includes the determination of the AB0 antigens using the test antibodies anti-A and anti-B in separate test mixtures and detection of the corresponding ABO antibodies using test cells from the blood groups A1, A2, B, and 0 from a blood sample. Since the introduction of monoclonal test antibodies, it is no longer necessary to test with anti-AB /37/. Anti-AB sera were used in the past to detect weak AB0 variants, which are even more sensitively detected using monoclonal anti-A and anti-B.

Test mixtures containing A2 test cells are therefore unnecessary when monoclonal test antibodies are used for AB0 grouping.


Agglutination tests, including manual tube or slide tests, gel centrifugation tests, micro titer plate tests, or solid phase immunoassays at room temperature.


A definite AB0 blood group can only be identified if obvious agglutination with test antibodies and test cells is present, no extra positives are found, and the AB0 antigens determined correspond to the isoagglutinins. Otherwise, further investigations are required for clarification. The results of the controls (auto controls, Rh controls) and the antibody screen should also be taken into account.

The following points must also be considered:

  • Incorrect blood group determinations usually result from mixing up blood samples and/or patients or from clerical errors. Technical errors and misinterpretation of results are the exception. A number of irregular findings can make determination of the AB0 antigens more difficult (Tab. 27-22 – Problems associated with blood group determination).
  • Strict attention must be paid to verifying the identity of the individuals and samples being examined and results should be confirmed where possible by a second, independently collected blood sample (confirmatory test). The results of manual blood group determinations must be read by two individuals (principle of dual control). All procedural steps that might lead to misidentification of samples and/or patients, incorrect allocation or transfer of results, or misinterpretation of results must be checked.
  • Many problems indicated in Tab. 27-22 by an asterisk (*) can be largely avoided by using monoclonal test antibodies.
  • The neutralization of test antibodies due to a high concentration of soluble antigens occurs only if unwashed patient erythrocytes are used in the tube or slide test.

Chimerism and previous AB0 incompatible transfusions (two cell populations) are easily recognizable microscopically as mixed field agglutination (particularly if the gel centrifugation method is used) due to the simultaneous presence of agglutinated and non agglutinated cells.

The presence of Bombay and para-Bombay blood groups can be established relatively rapidly by subsequent testing with anti-H (negativity).

If isoagglutinins are not detectable, tube tests with a longer incubation period and final centrifugation as well as gel centrifugation tests (due to their higher analytical sensitivity) may be of value. The AB0 blood group can also be verified using a second antigen test (anti-A, anti-B) instead of reverse typing. In this case, the auto controls must always be negative, to prevent the erroneous determination of blood group AB as a result of autoagglutination (pseudoagglutinations or autoantibodies).

Cold antibodies can be eliminated as interference factors in most cases by pre incubating the blood samples and reagents and by performing the test at 37 °C (for further options, refer to Section – Cold agglutinins and mono thermal cold hemolysins.

Pseudoagglutinations occur less frequently when monoclonal test reagents are used since they require fewer colloidal additives. Otherwise, washing the cells is useful in most cases. The problem is often no longer present if a new blood sample (preferably citrated blood or EDTA blood) is examined because, in the meantime, the interference factors have been eliminated or are no longer in use (e.g., certain i.v. solutions or drugs). Rh determination

Test spectrum

The only Rh blood group antigen that needs to be included in routine blood grouping is Rh factor D [Rh(D)]. There is no general indication for determining the Rh blood group antigens C or E (using anti-CDE) or the Rh genotype unless the Rh factor is detectable in the agglutination test. The Rh genotype should only need to be determined in blood donors, girls and women of childbearing age, chronically transfused patients, and transfusion recipients with irregular antibodies to erythrocytes.

The labels on packed red cells must indicate the Rh genotype so that compatible blood products can be provided quickly to recipients with Rh antibodies.

According to current German guidelines, the different Rh blood group antigens must each be examined with two different test antibodies and auto controls must be included in the tests /37/. Monoclonal test antibodies from different cell clones must be used to examine Rh(D) /10/. Given the high quality of today’s monoclonal Rh reagents, repeat determination of the CcEe antigens in patients would appear to be generally unnecessary. Repeat determination is only justified if corresponding antibodies are present or need to be considered, or in the case of blood donors.


If monoclonal antibodies or agglutinating polyclonal test sera are used, simple agglutination tests employing tube, slide, or micro titer plate methods can be performed with a brief incubation at room temperature. Refer to Section – Agglutination tests.

For the agglutination test, IgM anti-D test reagents that do not detect the DVI category should be used /10/. If incomplete polyclonal test sera are used, supplement tests with incubation for 15–30 min. at 37 °C need to be performed. The manufacturer’s instructions must always be followed. Meanwhile, gel centrifugation tests have proven their effectiveness for determining the partial D antigen type. Either ready to use cards containing antibodies or separate test antibodies can be used.

Controls: in auto controls (Rh control medium), the erythrocytes to be examined must be tested against the medium of the test antibodies under identical test conditions.


In addition to the problems that apply to both AB0 determination (refer to Section – AB0 determination and Rh determination(refer to Section – Rh determination), various weak D variants can also cause diagnostic difficulty and transfusion related problems. These antigens, previously known as Du antigens, can nowadays be determined more sensitively, more specifically, and with a higher degree of differentiation using monoclonal test antibodies /10/.

Under routine conditions, however, the only thing that matters in the context of recipient investigations is to define weak D forms as Rh(D) positive if possible and DVI as Rh(D) negative so that transfusion patients can be managed appropriately. For this reason, monoclonal test antibodies that do not recognize the category DVI should be used. In this way, it is consciously accepted that DVI as well as very weak forms of weak D and D negative are not differentiated under routine conditions, since to do so would require more elaborate examinations.

Examinations in blood donors should include additional tests such as the indirect antiglobulin test using polyclonal anti-D or an anti-D that also recognizes DVI (e.g. anti-Dblend), so that very weak D and DVI are also detected. However, it is not necessary to differentiate between these by using specific subsets of monoclonal test antibodies.

The measures suggested in Section – AB0 determination to solve various problems can also be applied to Rh determination.


The examination spectrum and the interpretation of results in patients differ from those in blood donors. For practical reasons, and due to the fact that the risk of maternal immunization to very weak neonatal weak D or DVI erythrocytes, newborns do not need to be examined in the same way as donors.

Interpretation of the results of Rh(D) determination in patients and in newborns if two monoclonal anti-D reagents that do not recognize DVI are used:

1. Rh(D) positive, if the test reagents yield clearly positive results (≥ 2+) with both anti-D test antibodies but the auto control is negative

2. Rh(D) negative, if the test reagents yield clearly negative results with both anti-D test antibodies regardless of the result of the auto control

3. Further testing, e.g. the indirect antiglobulin test with incomplete, polyclonal anti-D or anti-Dblend, including an auto control as well as positive and negative controls, is recommended if both anti-D test reagents yield very weakly positive (< 2+) or discrepant results in the face of a negative auto control, with a view to recognizing nonspecific positivities and autoantibodies as the cause.

The interpretation is:

  • Rh(D) negative if the indirect anti-globulin test is negative, provided the positive and negative controls yield the respectively correct results. This pattern of findings can be explained by nonspecific reactions in the agglutination test. It can also be found in the presence of RH27, which has a frequency of < 1 : 60,000 /66/ and can only be detected by IgM anti-D. Recipients who already possess this Rh blood group antigen do not become immunized to D by receiving Rh(D)-positive blood.
  • Rh(weak D) positive if the indirect antiglobulin test yields a positive result in the face of a negative auto control as well as correct results in the positive and negative controls. Since a direct agglutination test with monoclonal antibodies that do not detect DVI was already at least weakly positive, the result cannot be caused by DVI.

4. As described under point 3, further tests are required if both the test reagents with anti-D and the auto control are positive. The interpretation is:

  • Rh(D) positive if the indirect globulin test with anti-D is positive but the simultaneously run auto control in the indirect antiglobulin test is negative. The result for the auto control in the previous direct test was evidently nonspecific.
  • Rh(D) not determinable if the test reagent with anti-D and the auto control are also positive in the indirect antiglobulin test. This pattern of findings is observed if autoantibodies are present. Patients with these findings should receive Rh-negative blood products. However, specific results may be obtained by using antihuman IgG as the antiglobulin serum (in the case of complement-activating autoantibodies. Refer to Section – Cold agglutinins and mono thermic hemolysins. For further possibilities to determine Rh factors in autoimmune hemolytic disease (refer to Section 27.5.3. – Autoimmune hemolysis).

In principal, anti-Dblend may also be used in addition to monoclonal anti-D. Anti-Dblend is a mixture of (monoclonal) IgM anti-D and IgG anti-D. The IgM anti-D reacts with D-positive cells in the direct agglutination test but does not detect DVI. In the subsequent indirect antiglobulin test, the IgG anti-D also detects DVI. If, however, unwashed cells are used for the agglutination test and thus a protein medium like the one in a supplement test is present, DVI may also be detected by the direct test in the first stage (agglutination test), leading to an erroneous interpretation.

Polyclonal, complete anti-D can also be used as a second reagent parallel to a monoclonal anti-D that does not detect DVI; however, this makes it even more difficult to interpret the results.

Any results that are not unequivocally positive or negative should be investigated further in the indirect antiglobulin test using anti-Dblend or polyclonal antiglobulin serum.

If an incomplete, polyclonal anti-D is used in parallel with a complete anti-D that does not detect DVI, the results are interpreted as follows:

1. If a discrepant or only weakly positive result is obtained in the direct test with one or both of the anti-D reagents in the face of appropriate results in the controls:

  • Rh(D) negative if the indirect antiglobulin test is negative, irrespective of which of the two anti-D reagents was positive in the direct test
  • Rh(weak D) positive in the event of a positive indirect antiglobulin test and appropriate results in the controls, if the direct test with monoclonal anti-D was already positive (since the presence of DVI should be ruled out under such circumstances)
  • Rh(D) negative as a recipient and Rh(D) positive as a donor in the case of a positive indirect antiglobulin test, if the test was negative using complete or monoclonal anti-D and positive using polyclonal antiserum, since the latter could be due to the presence of DVI
  • Rh(D) negative as a recipient and Rh(D) positive as a donor in the case of a positive indirect antiglobulin test, if the test was negative using complete, monoclonal anti-D since this could be due to the presence of DVI.

2. Positive result in the auto controls: the interpretation does not differ from that of tests using two monoclonal anti-D reagents (see above).

By using two monoclonal test reagents, fewer Rh-negative results are determined, thus preventing the unnecessary transfusion of Rh-negative blood products in many patients.

Rh(D) determination in blood donors

Even if the tests yield negative results with both anti-D reagents, additional tests must generally be carried out (e.g. the indirect antiglobulin test using IgG anti-D that also detects DVI or a suitable RHD PCR procedure to exclude the presence of a weak D variant). If the indirect antiglobulin test is positive in the face of a negative auto control, blood units from this donor need to be labeled as Rh positive. In the emergency health card or donor card, it must be noted that this individual is Rh negative as a recipient and Rh positive as a donor. Confirmatory test

The ABD test with anti-A, anti-B, and anti-D is suitable for confirming blood groups using a second or further blood sample and for verifying the identity (e.g. of blood units). It can be performed using the least labor-intensive method for the clinical laboratory in question (e.g. plate test). In the case of blood samples from patients, an auto control (e.g., with Rh control medium) must also be performed. Blood groups indicated on emergency health cards should be confirmed by performing a confirmatory test on a second blood sample where possible. Determination in emergency situations

AB0 and Rh(D) grouping must be complete even in emergency situations and must include reverse typing, unless a valid blood group result is already on record at the laboratory responsible. Until a full blood group determination has been completed, universally compatible blood products must be transfused /67/. Alternatively, lateral flow technology can be used for emergency blood grouping (refer to Section – Lateral flow technology). This method is reliable and significantly faster than other methods. Determination in newborns

Conclusive results cannot be obtained from blood group examinations in newborns and infants (refer to Section – ABH(0) system). Accordingly, blood group results in this group must be indicated as preliminary. Nonetheless, the determination of blood group antigens is relatively reliable if monoclonal test antibodies are used, especially in combination with sensitive test methods such as gel centrifugation.

If blood products that are not universally compatible are transfused, the results should be verified by performing tests twice with reagents from different manufacturers. Bedside testing

The bedside test, also known as the AB0 identity test, is merely used to confirm a patient’s AB0 blood group that has already been determined in the laboratory. Its purpose is to discover potential mistakes in the identification of patients and/or blood units immediately prior to transfusion, thus helping to prevent the dreaded complications of AB0-incompatible transfusions. This test is not intended to replace blood group determination in the laboratory since it is less sensitive and more susceptible to interference. It must always be used directly at the patient’s bedside.

Although the additional determination of Rh(D) is not mandatory, the current availability of rapidly and strongly reactive monoclonal test reagents means that it should be considered for transfusions in girls and women of childbearing age at least. Bedside testing of blood units is only required in the case of autologous transfusions; however, it can help inexperienced staff with evaluating patient blood results in other types of transfusion also.

27.6.4 Other blood group antigens

According to German guidelines, if corresponding test antibodies are available, all antigens to be investigated should be determined using two different reagents and tests should include auto-controls, positive controls (antigen-positive test cells, heterozygous if possible), and negative controls (antigen-negative test cells) /37/. The test methods used depend on the specifications of the reagent manufacturers. More reliable results can be obtained with fresh, washed erythrocytes from citrated blood or EDTA blood. The assessment of the results should take previous blood transfusions into account.

The test antibodies currently on the market (monoclonal antibodies in particular) can determine the Rh genotype as well as, for example, the K, k, Fya, and Fyb antigens, with such a high degree of reliability that repeat determinations in recipients are no longer necessary.

27.6.5 Antibody screening

Antibody screening is used to detect clinically relevant and potentially relevant allo antibodies and autoantibodies against erythrocytes outside the AB0 system with a high detection limit. It is not normally available for detecting antibodies to low incidence antigens such as Cob, Wra, Lua, Kpa, Bga, and “private” antigens with cells from two or three test blood donors. Therefore, the antibody screening test cannot replace the cross match. Test conditions (including incubation temperature) should be selected accordingly to ensure that irrelevant cold antibodies are not detected by antibody screening.

Test spectrum

Antibody screening always includes the indirect antiglobulin test or a test based on this principle. Additional tests such as the enzyme test can also be useful (refer to Section 27.2.3 – Serologic reactivity). Using two or three test cell suspensions/populations, the following antigens, at least, should be present on the test cells: C, Cw, c, D, E, e, K, k, Fya, Fyb, Jka, Jkb, Lea, Leb, M, N, S, s, and P1 /37/. Other antigens such as f, Xga, Kpb, Jsb, Lub, H, and I are also offered because of their high frequency.

The antigen distribution on the different test cells should take into account that Rh, Kidd, MNSs, and rarely also Fy antibodies show dosage effects; if possible, these antigens should therefore be present in their homozygous form. To ensure this, test cells from three blood donors are required.

If the indirect antiglobulin test using the tube method is negative, a Coombs control with antibody coated erythrocytes must follow.

Auto controls can be omitted if the direct antiglobulin test is run simultaneously as part of blood grouping or auto controls are run simultaneously as part of cross matching. Otherwise, an auto control must be run later if antibody screening is positive (at the latest, during antibody differentiation).

Because of the sensitivity of the washing process in the tube test, controls should be run with every test series to check detection limit and specificity if this method is used. In automated test methods or those that do not involve washing, it is only necessary to run positive and negative controls once per shift or even once per working day.

Test frequency

For transfusions of RBC units that are cross matched with current blood samples, the most recent antibody screen in the recipient should not be older than 2–4 weeks. The occurrence of new, irregular antibodies due to primary immunization after transfusion does not need to be considered after shorter intervals. In contrast, antibodies that are subject to a booster effect due to secondary immunization have concentrations that allow their reliable detection even in potentially less sensitive cross matches.

As a precautionary measure, however, current German guidelines recommend repeating antibody screening with a new blood sample after three days if the patient has been transfused with RBCs or been pregnant in the previous three months /37/. On the other hand, if cross matches are normally omitted (type and screen), antibody screening must always be updated after three days before repeated transfusions.

In Rh-negative women, antibody screening as part of antepartum monitoring should be repeated between the 24th and 27th gestational weeks (refer to Section 27.5.2 – Hemolytic disease of the newborn).

In blood donors, this test should be repeated at least every two years as well as after pregnancies and transfusions.


When using the tube method and for subsequent investigation of transfusion reactions, the use of clotted blood (serum) is preferred since loosely bound antibodies are removed by washing and it may only be possible to detect them based on complement activation (C3d binding). If highly sensitive methods such as the gel centrifugation test or solid phase technology are used, the required degree of analytical sensitivity can also be achieved using citrated blood and EDTA blood (for exceptions, refer to Section 27.6.2 – Test methods suitable for automation.

The basic method consists of the indirect antiglobulin test (albumin or LISS technique) with incubation at 37 °C as a tube test, as a gel centrifugation test, or as a solid phase immunoassay. The single stage enzyme test at 37 °C as a tube test, gel centrifugation test, or solid phase immunoassay can be used as a supplementary method. Although the two stage enzyme test is significantly more sensitive, it frequently yields nonspecific positive results.

Refer to

Weakly reacting test antibodies and test cells that are heterozygous for the corresponding antigens are used as positive controls. If manufactured reagents are not used, the test antibodies should be diluted in AB serum to a titer of 4–8. A test reagent that does not contain antibodies is used as a negative control.


Negative antibody screening

Antibody screening is negative if, given the appropriate results in the positive and negative controls, all tests remain negative and the Coombs controls (in the tube test only) yield unequivocally positive results. A negative antibody screen rules out the most important irregular antibodies to erythrocytes (with the exception of AB0 antibodies). Antibodies to rare antigens (antigen frequency less than 9%) are not usually detected by antibody screening since these antigens, with the exception of Cw, are not usually offered on the test cells provided.

Negative antibody screening, negative Coombs controls

If the Coombs control remains negative in the tube test, this indicates that the indirect anti-globulin test was not performed correctly.

  • Inadequate automated washing accounts for most cases (e.g., omission of a wash cycle, insufficient wash solution, incomplete decantation, or inadequate resuspension of erythrocytes)
  • In the case of full automation, the addition of antiglobulin serum is sometimes omitted. In the case of an automated washer, every step must be checked and manual washing may possibly be required. In the case of insufficiently coagulated blood samples, the formation of clots in the test may interfere with the washing of cells.
  • Antibody screening must be repeated with completely coagulated blood (if necessary, blood is mixed with one drop of thrombin per mL, followed by incubation at 37 °C for 5 min. before centrifugation and serum separation) or with citrated blood or EDTA blood.

Negative antibody screening, negative result for positive control

This result raises the suspicion that a methodological error occurred during antibody screening or that the test is not sensitive enough (e.g., due to poor quality of the test cells). Antibody screening must be repeated using freshly prepared controls and, possibly, fresh reagents.

Positive antibody screening

This result always requires further investigation; if the result is clinically relevant, the treating clinician and the patient must be informed in writing (an entry regarding the antibody must be made in the blood group records or maternity records). This information must include the specificity and clinical significance of the antibody; if antibodies are detected during the course of a pregnancy, the antibody titer must be specified as well.

The clarification of a positive antibody screening test must be completed well in advance of any invasive diagnostic procedures or surgery that may lead to hemorrhagic complications requiring transfusion. Interpretation of findings:

  • Positive enzyme test with positive auto control in enzyme test: in most cases, nonspecific reactions or irrelevant benign cold antibodies are responsible. Further examinations are not necessary.
  • Positive enzyme test with one or more test cell suspensions and positive auto controls in the enzyme test and the indirect antiglobulin test: this result raises the suspicion of autoantibodies (cold autoantibodies) or, following a preceding transfusion, also of allo antibodies. The following additional examinations are required: antibody differentiation in the enzyme test, cold agglutinin test, direct antiglobulin test.
  • Positive indirect antiglobulin test with one or more test cell suspensions and positive auto controls in the enzyme test and the indirect antiglobulin test: this result raises the suspicion of autoantibodies (warm autoantibodies) or, following a preceding transfusion, also of allo antibodies.

The following additional examinations are required:

  • Antibody differentiation in the test methods that yielded a positive result, direct antiglobulin test, possibly cold agglutinin test, elution of antibodies that are attached to the patient’s erythrocytes.
  • Positive indirect antiglobulin test and/or enzyme test with one or more test cell suspensions and negative auto controls: this result raises the suspicion of allo antibodies. The following additional examination is required: antibody differentiation in the test method that yielded positive results.
  • Isolated positive auto control in the enzyme test: insignificant
  • Isolated positive auto control in the indirect antiglobulin test: this result raises the suspicion of immune hemolysis (autoantibodies, drug-induced antibodies, allo antibodies after transfusion). If the gel centrifugation method is used, a result with a reaction strength of up to 2+ is a relatively frequent, irrelevant finding (refer to Section – Micro column agglutination test).

27.6.6 Automation of blood group determination and antibody screening

Great progress has been made in the automation of blood grouping, antibody screening, and cross matching. This progress is mainly due to improvements in automatic pipetting equipment and reading instruments, the development of computer hardware and software for analyzing and documenting results, and improvements in individual test procedures. An important precondition is the use of anticoagulated blood (citrated blood or EDTA blood). Due to the high detection limit of the procedures, plasma can be used instead of serum for antibody detection. Refer to Section 27.2.3 – Serologic reactivity: complement activating antibodies.

Automation offers significant improvements in terms of safety. The use of barcode readers enables samples and reagents to be identified more reliably and reduces the risk of mixing up test reagents. Clerical errors in particular are eliminated. Electronic access to previous findings and the ability to compare findings and interpret findings automatically prevent and compare findings as well as automatic interpretation of findings reduce the risk of aberrant results (e.g. due to confusion of samples or erroneous readings). The AB0 compatibility of blood units can be verified automatically (electronic cross match). The use of automated test procedures also ensures that tests are performed in accordance with a defined standard profile (e.g. volumes, pipetting sequence, incubation temperature, incubation time) with minimum variation. Any significant deviations are detected by pipette checks, volume monitoring, and clot detection. Furthermore, automation contributes significantly to optimal quality management (monitoring of materials, complete documentation of batches and process steps). In addition, test results can be protected against methodological problems by incorporating the results of quality checks into the interpretation of findings by means of corresponding algorithms.

Automated reading of test reactions makes it possible to distinguish between reliably assessed and doubtful findings. Reliably assessed findings should not require visual inspection. Questionable findings must be subjected to visual inspection and assessment and the examination may need to be repeated either automatically or manually or supplemented by additional examinations. Findings reports can be generated either by the automated test equipment or by the laboratory IT system using algorithms, by comparing the results of different tests (including controls) internally. Requirement profiles are sent to automated equipment and findings are submitted online via bidirectional interfaces to the laboratory information system (LIS), donor information system (DIS), and hospital information system (HIS). In this way, clerical errors relating to patient data and blood group findings are avoided and blood products can be labeled reliably. As soon as the electronic transmission of findings released by automated equipment to the LIS, DIS, and HIS has been proven to be reliable, general visual comparison between immunohematological tests and the data stored in these systems is no longer necessary.

The accuracy of readings has been significantly improved by the use of high resolution charge-coupled device (CCD) cameras. Nevertheless, the reliability of the assessment of test results still varies between test procedures: in micro titer plate tests, for example, visual inspection should not generally be omitted. It is much more difficult to differentiate between positive and negative agglutination test results and to recognize duplicate populations in micro titer plate tests than in gel centrifugation tests, particularly since minor procedural variations during centrifugation or agitation of the plates can significantly affect the result. Problems of detection limit are particularly common in the determination of particular blood groups (e.g., K and e). If test reactions have to be read visually and/or automatically generated findings have to be corrected relatively often or determined by the user, a second reading (as in the case of manual determinations) would also appear to be necessary. In this check, the findings sent to the LIS must be compared directly with the plates. Password protection and various password levels are used to ensure that only correspondingly qualified staff can make corrections.

Gel centrifugation tests can be used to distinguish reliably between unequivocally positive test reactions, unequivocally negative test reactions, and questionable positive test reactions, especially if the cards are read from both sides. Therefore, only questionable results from the system need to be assessed visually. However, the percentage of questionable findings is still quite high (up to 10%). Second readings can be limited to checking these findings against the results transferred to the LIS. If visual assessment of results is incorporated as described above, fewer than 1% of blood group determinations performed on blood samples from blood donors using automated equipment require a follow-up examination. If patient blood samples are used, this figure is significantly higher due to patient-related problems and problems during blood collection. It is important that the blood samples are adequately anticoagulated, which can sometimes be a problem in the case of patient samples.

Due to the instability of reagents, in particular test cells, integrated cold storage is desirable. Uncooled test cells may need to be replaced every day, since certain antigens such as Jk, Fy, MNSs, Le, and P1 are highly unstable when stored and can partially dissolve. In this case, the quality of test cells and the specificity and sensitivity of the test procedure must be checked during each shift using positive and negative controls. However, test antibodies to antigens that are unstable in storage should be used. In addition, the cell suspensions should be mixed thoroughly prior to pipetting. Leftover reagents from automated equipment can be used in manual procedures, e.g. emergency determinations, on the following day. In this case, the cells must be washed again first using normal saline solution.

An appropriate security system is required in hospital laboratories in case of failure of automated equipment (Tab. 27-23 – Safety measures when using automated blood group analyzers). This must include appropriate back up procedures. If manual methods need to be used in the event of equipment failure, the same methods as those used by the automated equipment should be used.

While automated test procedures can certainly also be used to reduce the number of staff required or to enable the use of less qualified staff, specialist staff must be available as required to troubleshoot problems and perform minor, but relatively common, repairs such as exchanging needles and syringes and replacing tubing or seals. Automation can also contribute to cost reduction in immunohematology by reducing the amount of reagents and consumables (tubes, pipettes, gloves, paper) required and enabling leftover reagents to be reused.

The routine blood grouping of blood donors has been successfully automated for many years, since the samples can be serially processed without any urgency. Similarly, it makes sense to use automated equipment in central laboratories if this is economically justified by the number of samples to be examined. Nowadays, the use of automated equipment is worthwhile in many hospital laboratories as well. However, the workflow should be organized in such a way that as many samples as possible can be processed serially. Problems still arise in connection with urgent samples and different test profiles. Particularly urgent examinations (emergencies) must either be performed manually or using automated equipment reserved specifically for that purpose.

27.6.7 Cross match

Cross matching is used to verify the compatibility of erythrocytes prior to their transfusion (serological compatibility testing). It detects clinically relevant blood group incompatibility between donors and recipients. Unlike antibody screening, it also detects major ABO incompatibility and antibodies to rare (private) antigens.

Because of the low proportion of plasma in packed red cells and the minimal effect of irregular erythrocyte antibodies in donor blood, cross matching is limited to checking compatibility between the recipient’s serum and donor erythrocytes (major cross match).

Cross matches must show minimal susceptibility to errors and interferences while being simple and fast to perform. Therefore the test spectrum and complexity should be appropriately limited in comparison to antibody screening. Both this fact and the fact that the quality of donor cells used for cross match is inferior to that of test erythrocytes used for antibody screening automatically reduces the detection limit of cross matching as compared to antibody screening. Therefore, the cross match usually cannot replace antibody screening but can merely supplement it.

Test spectrum

Cross matching is performed as a major cross match with the indirect antiglobulin test including an auto control. As is the case for antibody screening (refer to Section 27.6.5 – Antibody screening), positive and negative controls must be run simultaneously using the same test method. Depending on the method used, these controls must be run for each test series (tube tests), once per shift (automated procedures), or once daily (manual gel centrifugation test).

Enzyme tests or other sensitive tests that are susceptible to interference are not recommended since urgent transfusions may be delayed by mostly nonspecific findings.

If the indirect antiglobulin test using the tube method is negative, a Coombs control must follow.

Minor AB0 incompatibility cannot be determined in the cross match. Certain solid-phase assays (refer to Section – Solid-phase immunoassay) detect antibodies of immunoglobulin class G only, so even major AB0 incompatibility often cannot be recognized. This can also occasionally occur in gel centrifugation tests as a result of the short incubation time, the use of plasma, and a lack of anti-C3d in the antiglobulin reagent. Therefore, it is advisable to include an AB0 (with anti-A and anti-B) or ABD confirmatory test using the patient’s blood as part of the cross match. Alternatively, an additional cross match can be performed using the agglutination test on a micro titer plate. Finally, an electronic cross match can also be used.

Additional testing of blood units in addition to the testing performed by the manufacturer is only required if a transfusion has to be administered despite a positive cross match (e.g., in the case of autoantibodies).

Blood samples from patients

The following blood samples are required:

  • For tube tests: clotted blood (serum), no older than three days (complement activity). In newborns and in emergency transfusions, citrated blood or EDTA blood can also be used. In emergency blood transfusions, citrated blood or EDTA blood may even be preferable because of the problems frequently caused by insufficient coagulation of native blood samples.
  • For gel centrifugation tests: clotted blood, citrated blood, or EDTA blood
  • For micro titer plate tests: citrated blood, EDTA blood.

Donor erythrocytes

Donor erythrocytes must be added to the solution specified for the respective test method. The detection limit and specificity of the method is improved by washing the cells with normal saline prior to the test. Ideally, the cells to be used should be taken from a freshly sealed tube segment of the RBC unit. Repeated or prolonged use of these cell suspensions is no longer recommended due to the instability of certain blood group antigens and the risk of bacterial contamination. Pilot tubes are associated with an increased risk of identification errors; also, the RBC quality (antigenicity) is poorer if the pilot tubes have been stored for a prolonged period of time and have been used repeatedly. Pilot tubes are filled under non sterile conditions. They contain a stabilizer that guarantees less storage stability than, for example, the additive solution in the units of packed RBCs. Furthermore, repeated cross matching implies repeated interruption of cooling and the repeated use of non sterile pipettes, both of which have a negative impact on the quality of the RBCs in the pilot tubes.

However, pilot tubes offer advantages in automated cross matches. In addition, packed RBC units stored in external blood banks can be cross matched in the laboratory if patient blood samples and pilot tubes are stocked in the laboratory. However, careful attention must be paid to the correct assignment of pilot tubes (preferably using barcodes) and blood units. Furthermore, interruptions to pilot tube cooling must be minimized. Pilot tubes can be used for up to five weeks if CPD-A is used as a stabilizer.

The blood from tube segments can be used for 8 to 14 days if it is transferred to special preservative or stabilizing solutions (e.g., Alsever’s solution) and the tubes are labeled accordingly.


The usual procedure for performing a cross match is the gel centrifugation test (15 min. incubation), in which the erythrocytes must always be suspended in the solution recommended by the manufacturer. Because of the susceptibility of this test to harmless cold agglutinins, it must be ensured that cell suspensions, recipient plasma, and reagents are at room temperature or above.

The indirect anti-globulin test using the albumin technique (30 min. incubation, tube test) is still considered an acceptable and sufficiently sensitive procedure. Because of its low susceptibility to nonspecific factors and irrelevant autoantibodies, it can be used in cases where more sensitive methods are largely positive but corresponding allo antibodies cannot be specified.


The result of the cross match must be documented on a form attached to each blood unit that includes all relevant personal data pertaining to the patient, the blood unit number, and the signature of the person who conducted the cross match. If possible, this form should remain attached to each RBC unit until the end of the transfusion.

Negative cross match

A negative result in a correctly performed cross match implies that the results of the controls are as expected (refer to Section 27.6.5 – Antibody screening).

A negative result also occurs in the case of minor AB0 incompatibility. If minor AB0 incompatible blood products are transfused intentionally (e.g., if packed RBCs of blood group A are used for a recipient of blood group AB) this should be stated on the attached form. By doing so, it is clear that the blood units have not been administered as the result of an error in the blood bank or laboratory.

Negative cross match in the presence of a currently negative antibody screen

If the current antibody screen is negative and identification errors have been ruled out, immune mediated acute hemolytic transfusion reactions are unlikely.

However, delayed hemolytic transfusion reactions due to unrecognized prior immunization cannot be ruled out (refer to Section – Acute hemolytic transfusion reaction). If transfusion-relevant irregular erythrocyte antibodies have ever been detected at any point in time, these must be taken into consideration for the rest of the patient’s life by selecting antigen-negative packed RBCs. Antigen testing must be performed using defined test reagents.

Previously detected antibodies to the Le, MN, P1, H, and I antigens (naturally occurring antibodies to ubiquitous antigens) that are not currently detectable no longer need to be taken into account. Even if these antibodies are currently detectable, their presence must not be allowed to delay life saving transfusions in emergency situations and should therefore be ignored.

Refer to:

Negative cross match in the absence of a current antibody screen

If a current antibody screen is unavailable, underlying incompatibility may go unnoticed due to the lower detection limit of the cross match.

Negative cross match in the presence of a positive antibody screen

Underlying incompatibility may not be detected due to the lower detection limit of the cross match (e.g., because of possible dosage effects). The selection of blood units by cross matching with the antibody containing serum of the patient is therefore not sufficiently reliable. The antibodies should be specified if possible and antigen tested blood units should be checked for compatibility. Blood units with a negative cross match result can be released with reservations, but only in an emergency situation. The treating physician needs to be informed and a note must also be made on the attached form stating that the compatibility of the unit is uncertain due to an unidentified antibody and that it is only to be used for transfusion if there are life saving indications.

Discrepant results for Coombs controls and positive/negative controls

Refer to antibody screening (Section 27.6.5).

Positive cross match

If cross matches yield positive results, all blood units provided for the patient in question need to be put on hold, except for emergency situations. Even the blood units with an apparently negative cross match may be incompatible if the negative result was due to a dosage effect and/or storage related poor RBC quality. Furthermore, the risk of mixing up blood units within the cross match series for a given patient is high. In general, the compatible blood units must therefore be identified unequivocally (antibody identification, antigen testing). If the antibodies detected by highly sensitive methods such as the gel centrifugation test cannot be identified, there are few concerns in emergency situations about transfusing RBCs that are cross match negative in the indirect antiglobulin test using the albumin technique.

Positive cross match in the presence of a negative antibody screen

Possible reasons for this constellation of findings include:

  • Major AB0 incompatibility
  • The use of different samples for the antibody screen and the cross match (mixing up samples, drug related effects)
  • The use of different methods in the antibody screen and cross match (further investigation using the positive test method)
  • Irregular anti-A1; verification of the A phenotypes of donor erythrocytes, reverse typing with recipient serum using the relevant test method
  • Autoagglutination of donor erythrocytes; auto control with donor erythrocytes and AB serum using the relevant test method
  • Antibodies to rare erythrocyte antigens.

Positive auto control

If only the auto control is positive while all other tests (apart from the positive control) are negative, an urgent transfusion can be administered without reservations if the patient has not been transfused within the last four weeks. The reasons listed in Section 27.5.4 – Drug induced immune hemolysis also need to be investigated.

Transfusion in the presence of a positive cross match

In exceptional cases where there are life saving indications for transfusion, even cross match positive blood may need to be administered if antibody investigation is not possible within the time available and/or compatible blood units are not available. After all available alternatives have been exhausted, cross match incompatible transfusion is justified in order to prevent injury caused by inadequate oxygen carriage. Even clinically relevant antibodies do not always cause serious hemolytic transfusion reactions. However:

  • AB0 incompatibility must be ruled out
  • The transfusion must be conducted with all necessary precautions
  • A specialist in transfusion medicine should be consulted if possible.


According to current German guidelines, the results of cross matches are valid for a maximum of three days (blood collection date plus three days) in patients currently undergoing transfusion or those who have been transfused in the last three months, due to the risk of booster effects /37/. If these patients are to receive further transfusions, blood units that have already been reserved must be cross matched again after this time with a new blood sample. If a transfusion or pregnancy in the last three months can be ruled out, the results of cross matches may be valid for up to seven days. Rapid and emergency cross match

Rapid cross matching should be capable of detecting major AB0 incompatibility as well as high concentrations of clinically relevant irregular antibodies to erythrocytes. It is indicated in emergency situations if antibody screening has to be postponed due to a lack of time. The indirect antiglobulin test using the gel centrifugation method (with 5–10 min. incubation at 37 °C) including auto control meets the relevant requirements. If the antibody screen is negative, the cross match does not need to be repeated using a standard method. The antibody screen, however, should be performed using the first blood sample. Electronic cross match

Electronic cross matching refers to the automatic comparison of the recipient’s blood group (AB0 and Rh factor) with the blood groups of the cross matched RBC units. To do this, the blood groups of the patient and the blood units must be stored correctly in an accessible form in the IT system of the automated blood grouping machine or laboratory. Ideally, any clinically relevant antibodies that have been detected in the patient at any point in time should also be included in the comparison. Electronic cross matching eliminates the need for AB0 confirmatory tests as part of the cross match, at least if the ABO confirmatory test has already been carried out on the blood sample (e.g., when the patient is admitted). It is particularly useful for uncross matched transfusions in emergency situations. Minor cross match

The minor cross match is conducted in order to investigate transfusion reactions that are associated with plasma containing blood units. It is performed as an indirect antiglobulin test (gel centrifugation test) using donor plasma and recipient erythrocytes. Omission of cross match

If antibody screening is carried out with three test cell populations using gel centrifugation or solid phase technology and AB0 compatibility is ensured by an electronic cross match, simple agglutination test, or regular monitoring of the AB0 blood groups of recipient and donor blood, the risk of hemolytic transfusion reactions is extremely low /68/, which means that blood can usually be transfused safely without performing a full cross match using the indirect antiglobulin test. This has been common practice in a number of countries (e.g., United States, Great Britain, the Netherlands) for many years /69/. In Germany, however, a full cross match is mandatory /37/. Cross matching can be omitted without hesitation in the case of urgent transfusions only and subject to the conditions specified.

27.6.8 Antibody identification

If an irregular erythrocyte antibody is suspected on the basis of a positive antibody screen, positive cross match, extra positivities in reverse typing, or in the case of an unexplained hemolytic transfusion reaction, additional investigations are required in order to detect and identify the antibody in question.

The test methods used are the same as those employed for antibody screening. However, it may be necessary to expand the test spectrum, optimize the test procedure, or (e.g., in the case of suspected cold antibodies) to adjust the incubation temperature.

It may sometimes be necessary to prove that a positive test reaction is due to an antibody and a specific immune reaction. Reproducibility of findings and detectability of the Ig class are essential aspects of this process. In general, the possibility of antibody mixtures must always be taken into account by examining a larger number of test cell populations or by attempting to separate and identify the various antibodies in question by adsorption and, possibly, elution.

The following approach has proved to be useful for the identification of antibodies:

  • Reproduction of the results under optimized test conditions (e.g., washed test cells, completely coagulated blood samples: serum); preheating of the samples, reagents, and materials; prolonged incubation
  • Identification using one or multiple test cell panels (each consisting of 8–15 cell populations) under optimized test conditions using the test methods that yielded positive results. Identification should not be limited to the most strongly reacting test method. Antibody mixtures and antibodies with dosage effects are easier to identify if test methods in which the antibody reaction was weaker are also included. The number of panel cells depends on the number of positively reacting test cells and on additional antigens hidden behind positively reacting cells, if these antigens are not presented adequately (possibly homozygous) on the nonreactive cells.
  • If there is interference by cold antibodies, the samples, reagents, and materials (including the wash solution) are preheated to 37 °C and the tests are performed strictly at 37 °C.
  • If cold antibodies with a wider thermal optimum are suspected, warm antibodies can be excluded by using anti-IgG as antiglobulin serum or by using test procedures that are less influenced by complement activation or IgM antibodies (e.g., solid phase technology)
  • Use of complement inactivated sera
  • Destruction of IgM antibodies by dithiothreitol or 6-mercaptoethanol
  • Expansion of the range of methods used (e.g., gel centrifugation, solid phase technology)
  • If the undetermined reactions potentially involve specific cold antibodies, the tests are performed at room temperature in order to positively identify the cold antibodies in question. As far as transfusions are concerned, however, it is more important to rule out warm antibodies.
  • If specific antibodies are suspected, optimal test conditions for these antibodies should be selected. For example, serum acidification to a pH of 6.5 for anti-M, room temperature for anti-S or anti-s (even in the indirect antiglobulin test), enzyme tests for Rh antibodies, and solid phase technology for Kidd antibodies.
  • Antigen testing in the antibody carrier. Usually, the affected individual does not actually possess the antigen against which the antibodies are produced. Exceptions to this rule are found in the case of anti-M, anti-N, anti-e, rarely anti-D, and following transfusions. If the antibodies involved are directed against high frequency antigens such as anti-k or anti-P1, the absence of these antigens has a high predictive value, and vice versa. This also applies to anti-Lea, anti-Leb, and anti-Lea, anti-Leb, whereby Le(a–b–) is almost always found (refer to Section 27.3.3 – Lewis system).
  • Adsorption of cold antibodies such as anti-I, anti-IH, and anti-P with formalin-treated rabbit erythrocytes
  • In the presence of cold autoantibodies at high concentrations or with a wide thermal optimum, cold auto adsorption is used to identify warm antibodies if the aforementioned methods are not sufficient.
  • Adsorption and elution of IgG antibodies using affinity chromatography, in order to eliminate high-titer cold autoantibodies
  • Neutralization of IgM antibodies by soluble antigens (anti-A and anti-B by AB substance, anti-P1 by P substance, anti-Lea, anti-Leb by Le(ab) substance)
  • Proteolytic destruction of enzyme sensitive antigens (Tab. 27-4 – Impact of enzymes on blood group antigens).
  • Inclusion of A1, A2, and O cord blood cells. Cells of blood group A1 do not react with anti-H and anti-IH, while A2 cells react significantly more weakly than 0 cells with these antibodies. Cord blood cells react significantly less strongly with anti-H, anti-IH, anti-I, and other high-incidence antibodies (Tab. 27-1 – Blood group antigens on the erythrocytes of newborns).
  • Antibody cross match against a number of donor cell suspensions in parallel with test antibodies of the presumed specificity (e.g., antigens that the antibody carrier does not possess)
  • Differentiation using diluted sera in order to detect any antibody mixtures and dosage effects
  • Differential adsorption and differential elution in the case of antibodies with high antigen frequency (broad reaction) in order to identify antibody mixtures and autoantibodies. If no information about the antibody specificity is available (e.g., from antigen testing), the use of cells with the following antigen pattern is recommended: CCDee/ccDEE/ccddee (K neg), homozygous for the antigens Fya/Fyb/Jka/Jkb.
  • In the presence of warm autoantibodies, auto adsorption can be performed easily if there is little immunoglobulin attachment to patient erythrocytes (e.g., if only the complement antiglobulin test is positive). If there is significant immunoglobulin binding, the autoantibodies have to be carefully eluted from the patient’s erythrocytes before adsorption.
  • Determination of subclasses (IgG1–4) by means of gel centrifugation or flow cytometry in order to estimate the clinical relevance of allo antibodies and autoantibodies.

27.6.9 Antibody titration

Antibody titrations are of limited value only in terms of providing a reliable and clinically useful measure of the antibody concentration. In geometric dilution, a titer increase of one dilution corresponds to a doubling of the antibody concentration. On the other hand, because of methodological variations due to changes in the antigenicity of the test cells and the relatively low accuracy of manual pipetting, a change of at least two titer dilutions is required for differences in titers to be considered as significant (even those that are determined under standardized conditions in the same laboratory). This corresponds to a total difference in score of 10 between two serial titrations when the reaction strengths of the individual titer dilutions are added up (based on an assessment of the reaction strength using a score ranging from 0–4). Because of methodological differences, the determination of antibody titers in different laboratories at best allows conclusions to be drawn from trends if the methods used are specified.

Therefore (at least for monitoring during pregnancy), standard methods should be used, parallel titration of earlier samples from the patient should be performed, and controls with a defined antibody concentration should be run simultaneously as standard. This allows direct comparisons between different laboratories.

The titer corresponds to the last positive dilution. According to international practice, the titer is expressed as the reciprocal value of this dilution (e.g., a dilution of 1 : 32 = a titer of 32).


Checks during pregnancy for estimating the risk of hemolytic disease of the newborn. A clear prediction of the activity and effects of the antibodies cannot be made based on absolute titers alone; other factors are also involved. However, serial antibody titers allow conclusions to be drawn in relation to the further diagnostic and therapeutic approach.

Cold antibody autoimmune hemolysis is associated with the detection of abnormal cold agglutinin titers (> 64) in addition to the detection of complement attachment to patient erythrocytes.

Weak blood group antigens (e.g., Ax, DEL) can be detected by adsorption and subsequent titration of corresponding antibodies. If the erythrocytes have the corresponding antigen, the antibody titer in the adsorbate is significantly lower after adsorption with antigen-positive cells than with antigen-negative cells.

Serial autoantibody titrations are suitable to a limited extent for assessing the clinical course. The titration of auto hemolysins is of special diagnostic value (Section – Hemolysis methods). Cold agglutinin test and cold agglutinin titer

Since cold agglutinins are not associated with the prozone phenomenon, determination of the cold agglutinin titer is only useful if at least one test in the cold agglutinin screening tests is very strongly positive (one agglutination).


Patient serum, adult test erythrocytes of blood group 0 or the AB0 blood group of the patient, cord erythrocytes as a 5% suspension in normal saline. Refer to Section – Cold agglutinins and mono thermic cold hemolysins.

Cold agglutinin test

Two tests as tube tests with two drops (100 μL) of the patient’s serum and one drop (50 μL) of erythrocyte suspension (one test with adult erythrocytes and one test with cord erythrocytes). Incubation takes place for 2–4 h in the refrigerator at 4 °C. The results are read immediately over an illuminated box without prior agitation or heating. An elevated cold agglutinin titer is only likely if very strong agglutination (one agglutination) is found in one or more of the tests.

If specific cold antibodies such as anti-H, anti-IH, anti-P1, are suspected based on AB0 reverse typing, antibody screening, and/or antibody identification tests, the corresponding antigen-negative and antigen-positive erythrocytes should be used as the adult test cells. If irregular warm antibodies are present simultaneously, the test cells should not have the corresponding antigens, so that their influence on the cold agglutinin test can be ruled out.

In order to determine the thermal amplitudes of cold agglutinins, the cold agglutinin test is run in parallel at different incubation temperatures (4 °C, 20 °C, 30 °C, and 37 °C) in different water baths.

Cold agglutinin titer

The same tests are used as in the cold agglutinin screening test. Initially, 100 μL of the patient’s serum is diluted geometrically in 100 μL of normal saline solution (two series). The tip of the pipette must be changed after each titer dilution. If the titer cannot be crudely predicted on the basis of previous findings, dilutions of up to a titer of 1 : 1,000 are initially prepared, leaving the pipette tip in the last tube, thus allowing further titration in the case of higher titers. Adult erythrocytes and cord erythrocytes in normal saline suspension are subsequently pipetted into each dilution series (50 μL in each tube). The tubes are incubated for 2–6 h in ice water (melting ice) placed into a refrigerator at 4–8 °C.

The results are read as described in the previous section. Because the agglutinations start to dissolve as the temperature increases, the results should be read starting from the end (highest dilution). The titer corresponds to the last dilution that yields a clearly positive result. The cold agglutinin titer refers to the titer of the test using adult cells or the highest titer obtained. A cold agglutinin titer of greater than 64 is considered to be elevated. If the cold agglutinins possess anti-I specificity, the titer with adult erythrocytes is at least two dilutions higher than the one with cord erythrocytes (refer also to Tab. 27-20 – Determination of the specificity of cold autoantibodies, based on examples). Allo antibodies

Titration of allo antibodies is usually only indicated as part of monitoring during pregnancy. The anti-D titer in the indirect antiglobulin test (tube test with albumin technique) is the only one for which clinically relevant cut-off values have been identified. We therefore recommend using this method for titration.


Current and possibly previous serum samples from the patient; fresh, antigen-positive test cells (preferably homozygous) as a 5% suspension in normal saline, e.g. for the titration of anti-D test erythrocytes of blood group 0 ccDDEE (R2R2). As the standard, control serum, such as anti-D for Rh immune globulin prophylaxis, diluted in phosphate buffer (approx. 1 : 10, target 50 IU/mL), corresponding to a titer of approximately 1,000 in the indirect antiglobulin test using the albumin technique, is stored frozen in 200 μL aliquots. Reagents for the indirect antiglobulin test.

Test protocol and interpretation

Geometric dilution of patient serum in normal saline. Parallel serial titrations using patient serum from previous examinations as well as control serum.

After test erythrocytes and albumin have been added, incubation at 37 °C for 30–45 minutes. The remaining steps in the procedure and the reading of results are performed as described in Section – Indirect antiglobulin test. For titration using the gel centrifugation test, the patient’s serum must first be diluted in a tube.

The remaining steps of the procedure are performed as described in Section – Micro column agglutination test. For most antibodies (Rh antibodies) the titers are 2–4 levels higher than in the tube test using the albumin method.

The target titer of the control serum is determined by repeat titrations any time after the standard is produced. The target titer plus or minus one dilution must always be reached. Otherwise, all titration series have to be repeated.

Compared to the previous sample, a rise of two titer dilutions (or a score of 10 when the agglutination strengths of the individual titer dilutions are added) is considered to be significant. In the case of anti-D, the antibody concentration can be calculated as follows:

x (IU/ml) = Titer of the patient’s serum × IU of the control serum Titer of the control serum Hämolysine titer

Refer to Section – Hemolysis methods. Titration in the direct antiglobulin test

The degree of immunoglobulin and/or complement binding to erythrocytes can be measured by the direct antiglobulin test using geometric dilution of the antiglobulin reagents in normal saline.

In individual cases, these antiglobulin titers may correlate with the clinical course. In general, however, they do not allow conclusions to be drawn regarding the severity of hemolysis. Prozone phenomena suggest the presence of benign warm autoantibodies.

27.6.10 Special immunohematological methods Neutralization tests


By means of soluble blood group substances (AB, P, Sda, and the Lewis blood group substance), IgM antibodies of the corresponding specificity can be precipitated. This results in the neutralization of these antibodies. By this method, antibodies with a different specificity or IgG antibodies with the same specificity (e.g., anti-A in the AB neutralization test) can be detected more easily. In addition, the specificity of a suspected antibody can be confirmed if it can be neutralized by the corresponding blood group substance. This may be particularly useful in the case of Lewis and P1 antibodies, which may only react with individual antigen-positive test cells due to the storage duration (and subsequent quality) of the test cells or a low antibody concentration.

AB neutralization test and titer

One volume of AB substance is added to one volume of the patient’s serum (e.g. 50 μL), followed by incubation for 60 min. at 2–8 °C. This mixture is then used for cross matches or titrations (in normal saline) using the indirect antiglobulin test. If subsequent examinations are performed using the gel centrifugation test, the sera should be centrifuged at high speed after neutralization, (e.g., in a hematocrit centrifuge for 3 min. at 16,000 × g) and the supernatant used for the subsequent tests.


The cross match between maternal serum and paternal erythrocytes is performed in order to detect incompatibility between mother and child caused by antibodies to rare paternal antigens. Test erythrocytes with the paternal AB0 group are run simultaneously as a control; these erythrocytes must react negatively after adequate neutralization of the AB0 antibodies.

If AB0 incompatibility is suspected between mother and child, the titer is determined using maternal serum in the indirect antiglobulin test. A titer of < 32 in the AB neutralization test generally rules out AB0 incompatibility. On the other hand, a higher titer is not necessarily associated with AB0 incompatibility; nevertheless, the probability increases significantly with titers > 1,000.

Neutralization with P or Lewis blood group substances

As with the AB neutralization test, the corresponding blood group substance is added to the serum of the patient, followed by incubation. An identical test using normal saline instead of the blood group substance is run simultaneously so that the result of the neutralization test can be interpreted independently of the dilution effect of the reagent. The volumes selected depend on the subsequent examinations (antibody screen, antibody identification, cross match). If the subsequent examinations are performed using the gel centrifugation test, the immune complexes that form should be removed prior to the examination.


Confirmation of rudimentarily reacting P1 and Lewis antibodies. These antibodies can be neutralized whereas the dilution control remains positive.

Neutralization of interfering P1 and Lewis antibodies in sera with antibody mixtures in order to identify the more relevant immune antibodies. Adsorption


Erythrocytes are capable of binding antibodies directed against their antigens, thus removing these antibodies from the serum. The efficacy of adsorption correlated with antigen excess and optimal antibody binding. Accordingly, different types of antibodies are adsorbed by erythrocytes with the corresponding antigens under particular conditions (e.g., cold antibodies at low temperatures, warm antibodies at 37 °C, enzyme-reactive antibodies in the presence of enzymes).

IgG antibodies can also be separated by immunoadsorption under routine conditions, thus allowing the differentiation of clinically relevant IgG antibodies after elution from the columns without interference from cold antibodies.


Indications for removing antibodies from the serum:

  • To determine their specificity
  • To separate antibody mixtures
  • To remove interfering antibodies (cold antibodies, cold and warm autoantibodies), thus facilitating the detection of more relevant antibodies in the adsorbate.
  • To differentiate between antibody induced and nonspecific reactions, since the latter usually cannot be removed by adsorption
  • To detect weak or blocked antigens
  • To separate adsorbed and subsequently eluted antibodies.

Adsorption of allo antibodies


Antibody containing patient serum; thoroughly washed test erythrocyte sediment (if possible, from citrated blood or EDTA blood e.g., 1 mL of sediment washed three times with 10 mL of normal saline); LISS or bromelin solution. The test cells should have the antigens to which the interfering or already known antibodies bind but not the antigens to which those antibodies bind that are to be demonstrated and identified in the adsorbate. Sera with hemolytic activity in vitro have to be inactivated at 56 °C for 15 min. prior to adsorption.

Test protocol

One volume of serum and two volumes of erythrocyte sediment are thoroughly mixed and incubated (with thorough mixing every 15 minutes). A half volume to two volumes of bromelin solution or LISS may be added. The incubation temperature depends on the thermal optimum for each type of antibody that is adsorbed. Cold antibodies are incubated for 1 h at 4–20 °C, warm antibodies for 30 to 60 min. at 37 °C, undetermined antibodies initially for 30–60 min. at 37 °C and then for 1 h at ≤ 20 °C. This is followed by centrifugation for 5 min. at 2500 rpm (1500 × g). The adsorbate (supernatant) is collected.

Adsorption control: the adsorbate is checked in the test procedure in which the antibody was reactive using fresh test erythrocytes as used for the adsorption. If this control yields a negative result, this indicates that adsorption was successful. If this control is positive, adsorption needs to be repeated using the adsorbate and fresh cells.

A series of consecutive adsorptions may be necessary. In certain circumstances, adsorption is only successful if enzyme treated erythrocytes are used (enzyme reactive antibodies) or if bromelin or LISS are added. However, the addition of these supplements must be taken into account in subsequent tests, in which excessively high concentrations of bromelin or LISS can cause nonspecific reactions.

Adsorption of cold antibodies

Cold antibodies such as anti-I, anti-IH, anti-H, and anti-P can be adsorbed using formalin treated rabbit erythrocytes.

Adsorption of cold autoantibodies


Patient serum; patient erythrocytes (sediment), thoroughly washed at 37 °C.

Test protocol

The same test mixtures are used as for allo antibodies. In the case of antibodies (especially of high titer) against protease resistant antigens, enzyme treated erythrocytes may be used. Incubation for one hour at 4 °C, centrifugation in a cold centrifuge (4 °C) at 2500 rpm (1500 × g). The rest of the procedure is the same as for allo antibodies.

Adsorption of warm autoantibodies

Auto adsorption is relatively ineffective if the patient’s erythrocytes are coated with IgG. In such cases, the autoantibodies must first be gently eluted, thus avoiding hemolysis (refer to Section – Elution).


Patient serum, thoroughly washed patient erythrocytes (sediment) after gentle elution.

Test protocol

The test protocol is the same as that described for warm allo antibodies. However, the control tests with the patient’s cells often do not yield a completely negative result since a small amount of IgG may be left on the cells, even after elution.

Adsorption of antibodies to plasma proteins

Some HTLA antibodies (HTLA, high titer, low avidity) that react with almost all test erythrocytes are antibodies to plasma proteins (e.g. complement factors, anti-Chido, and anti-Rogers). These antibodies can be bound by incubating the sera with fresh, complement containing AB serum (pooled plasma from at least four different individuals). Patient serum and AB serum are mixed in a ratio of 1 : 1 and incubated for one hour at room temperature. Even if the sera then show a negative reaction, this could also be a result of dilution. Therefore, the adsorption result can be considered positive only if the simultaneously run control using normal saline (instead of AB serum) remains positive. Elution


Antibodies may be eluted by neutralizing the binding forces, changing or destroying the antigen structure, or altering their tertiary structure. Within certain limits, antibodies eluted in this manner are detectable in the same way as serum antibodies.


Numerous elution methods are available, the descriptions of which are beyond the scope of this chapter. For details of these methods, readers are referred to the special immunohematological literature /1153/.

On account of their different indications, some of these methods are presented here.

Heat elution

As a result of the effect of heat, the binding steady state that follows the law of mass action is shifted toward dissociation. Depending on the optimal binding temperature of each of the corresponding antibodies, these antibodies can be eluted at 37–45 °C (cold antibodies) or at 50–60 °C (warm antibodies). Changes in the structure of antigens and antibodies as a result of the selected temperature are minimal.

Hapten elution

By adding excessive amounts of water soluble antigens (drugs, AB substance), the corresponding cell bound antibodies may be displaced from their binding sites. These procedures are rarely used nowadays.

Freeze/thaw elution

Cell membranes are destroyed by freezing and subsequent thawing. The antibodies dissociate due to changes in the antigens and ion concentration. The eluates are hemolytic.

Ultrasonic elution

Antibody dissociation occurs due to shear forces and thermal energy from high-frequency ultrasound.

Acid elution

Acidification of the medium induces changes in the charges (cationic state) and tertiary structure of proteins, thus loosening antibody binding. However, acids also destroy cell membranes if the pH is sufficiently low and the reaction time is sufficiently long. The degree of hemolytic activity of the eluates depends on the degree of acidity.

Elution using organic solvents

Alcohol, ether, xylene, and chloroform reduce the surface tension of the solution, thereby leading to the dissociation of antibodies. At higher concentrations, they destroy the antigen structure and cause largely reversible changes in the tertiary structure of antibodies. These methods are particularly effective for eluting high concentrations of antibodies. However, they are too aggressive for certain antibodies (e.g., Duffy and Lewis antibodies) which are not always detectable in the eluate. The eluates are usually hemolytic.

Chloroquine method

Chloroquine diphosphate neutralizes oppositely charged groups on amino acids, thus causing dissociation of antigen-antibody binding. During this process, antigens are not (or only minimally) denatured. Complement components are not removed. IgG antibodies eluted by chloroquine are not suitable for subsequent antibody identification procedures because the chloroquine remains bound to the eluted antibodies and cannot be neutralized. Further, the treatment of erythrocytes with chloroquine can selectively destroy HLA antigens that are expressed to varying degrees on (mainly young) erythrocytes /53/.


Under routine conditions, even specialized laboratories require at most three different elution methods. First, a routine method should be established that is capable of eluting the most important antibodies as effectively as possible; it also needs to correspond with the subsequently performed detection method. Significantly hemolytic eluates are not suitable for direct examination using the gel centrifugation test. In these cases, the eluates must be incubated in a tube with the test erythrocytes, which must then be washed once. The erythrocytes suspended in the diluent can then be added to the gel centrifugation test as for the direct antiglobulin test. Hemolytic eluates can also be examined using tube tests or solid phase technology.

A second method should be available that allows a particularly gentle elution of antibodies that does not change their structure and that preserves their complement binding ability (e.g. heat elution). If, however, this second method does not leave the cell membrane largely intact or does not elute antibodies satisfactorily, a third method may be necessary. This method should allow the determination of erythrocyte antigens and the absorption of warm autoantibodies (e.g. the chloroquine method or modified acid elution).

General aspects

  • The erythrocytes must be washed thoroughly prior to elution and transferred to a new tube in order to prevent serum antibodies being carried over
  • The test conditions during adsorption, but at the latest during the subsequent washing of the erythrocytes, must ensure complete resuspension during washing; otherwise, serum antibodies may be carried over
  • Erythrocytes must be washed with appropriately cold solutions if cold antibodies are to be eluted subsequently
  • After the container has been opened, organic solvents should not be used for more than four weeks or after the residual volume is down to a quarter because the pH decreases during storage, which may result in nonspecific reactions in the eluates
  • Eluates change rapidly following their production and should therefore be examined as soon as possible. Precipitation of proteins can result in pseudoagglutination. Precipitates can be removed by high speed centrifugation prior to examination.
  • During prolonged storage, eluates remain stable only if they are frozen in a protein solution, e.g. by diluting the eluates with serum or by adding bovine albumin (final concentration 6%). Elution methods are very sensitive to methodological variations, which is why the methodological recommendations should be strictly observed.

Heat elution

Materials: thoroughly washed erythrocyte sediment (e.g. 1 mL of erythrocytes washed 5–6 times with > 10 mL of normal saline), antibody free AB serum or 6% bovine albumin solution, shaking water bath, pre warmed centrifuge cups, possibly filled with warm water.

Test protocol: erythrocyte sediment and AB serum or bovine albumin solution are incubated in equal aliquots in a warm shaking water bath. For the elution of warm antibodies, the incubation time is 5 min. at 56 °C, whereas for cold antibodies, it ranges from 30–60 min. at 37 °C (agitation is required every 15 min.).

A final thorough agitation without cooling is followed by centrifugation (e.g. for 3 min. at 1,000 × g) using appropriately warmed centrifuge cups. The supernatant (eluate) is collected.

Comments: this procedure is particularly well suited to the elution of AB0 antibodies and IgM antibodies in general.

Ether elution

Materials: thoroughly washed erythrocyte sediment, diethyl ether, water bath, water jet pump, possibly a vacuum pump.

Test protocol: erythrocyte sediment is mixed with an equal volume of normal saline and two volumes of ether in a glass tube that is tightly closed with a rubber or cork stopper, followed by vigorous agitation for one minute. The stopper is removed carefully (positive pressure). The tube is then centrifuged at high sped (for 10 min. at 2000 × g). The upper two layers (ether, cell debris) of the three layers thus formed are suctioned off using a water jet pump while the lowest, hemoglobin containing layer (eluate) is transferred to a new tube. Ether residues must then be allowed to evaporate (15 min. at 37 °C). After repeated centrifugation at high speed, precipitated proteins and cell residues are removed. Experience has shown that the ether eluate displays fewer nonspecific agglutinations if it is kept in a refrigerator for 6–12 h prior to centrifugation. In order to provide ether eluates more quickly, ether can also be extracted with a vacuum pump without incubation at 37 °C prior to the final 10-minute centrifugation /70/.

Comments: ether elution is a particularly effective method that allows elution of most warm allo antibodies and warm autoantibodies in high concentrations. Only antibodies such as anti-S, anti-s, and sometimes anti-Fya, anti-Fyb as well as certain cold antibodies (e.g., Lewis antibodies) are less effectively eluted and/or are not detectable in the subsequent test methods.

Acid elution

The combination of acid elution with subsequent testing using the gel centrifugation test is highly suitable for detecting erythrocyte bound allo antibodies and autoantibodies. This method can be used, for example, to detect anti-A on a child’s erythrocytes even in very mild cases of 0/A incompatibility.

This chapter describes a simple method that yields very good results and can also used in a modified version to obtain largely intact erythrocytes.

Materials: erythrocyte sediment, commercial elution test kit consisting of concentrated wash solution, elution (acid) solution, and buffer solution.

Eluate production (follow any instructions provided by the test kit manufacturer): the erythrocyte sediment is thoroughly washed once with normal saline and then four times with a special wash solution. Twenty drops (approximately 1 mL) of the dry erythrocyte sediment are transferred to another tube and 20 drops of elution solution are added; the tube is sealed, carefully mixed four times, and, after 60 sec., centrifuged at high speed (for 5 min. at 1500 × g). Without delay, the supernatant is collected and neutralized by approximately 20 drops of buffer solution (while being shaken) until the color of the solution changes to blue.

Because of the potential for protein precipitation, the eluate should be centrifuged again at high speed before testing.

Erythrocyte collection: if erythrocytes need to remain as intact as possible following the elution of antibodies, the same procedure is used, but with less elution solution (12 drops of elution solution per 20 drops of erythrocyte sediment). After the supernatant is removed, the erythrocytes are immediately washed thoroughly with wash solution until the supernatant remains colorless.

Chloroquine elution

This method is the most gentle elution method for erythrocytes.

Materials: thoroughly washed erythrocyte sediment, 20% chloroquine diphosphate.

Test protocol: one volume of erythrocyte sediment (e.g. 0.2 mL) is mixed with four volumes of chloroquine solution and incubated for 30 min. at room temperature (thorough mixing is required every 15 min.); finally, the mixture is washed four times with normal saline.


If the erythrocytes are to be used for antigen testing, 20 μL of erythrocytes are removed from the sediment after incubation and washed four times with normal saline. Residual immunoglobulin binding is then checked using poly specific antihuman globulin. Antigen testing is only reliable if the antiglobulin test is negative. If the antiglobulin test yields a positive result, the incubation of the erythrocytes with chloroquine has to be extended up to a maximum of 120 min.

Indication for elution methods

  • In the case of a positive direct antiglobulin test: differentiation between allo antibodies and autoantibodies as well as investigation of antibody specificity (Section – Direct anti-globulin test)
  • In the case of a positive antiglobulin test after previous erythrocyte transfusion: detection of allo antibodies present at low concentrations and/or of delayed hemolytic or serological transfusion reactions (Section – Acute hemolytic transfusion reaction)
  • In the case of complement binding only in the direct antiglobulin test: detection of masked antibodies.
  • In the case of Coombs negative immune hemolysis: detection of autoantibodies present at low concentrations
  • Investigation of hemolytic transfusion reactions.
  • In the case of difficulties in determining antigens from erythrocytes coated with antibodies (generally autoantibodies): improved antigen determination after gentle elution
  • Adsorption of autoantibodies in order to detect and/or to rule out allo antibodies in the adsorbate
  • Identification of adsorbed and subsequently eluted allo antibodies, especially as part of the investigation of antibody mixtures
  • Detection of weak antigens after the adsorption of specific test antibodies. Detection of crypt antigens and poly agglutination

Materials: 5% erythrocyte suspension in normal saline, commercial lectin test kit.

Test protocol: two drops of each of the different lectins are each mixed with one drop of erythrocyte suspension in a tube. This is followed by incubation at room temperature for 5 min. and centrifugation for 20–30 sec. at 200 × g. The results are read over an illuminated box. With regard to the interpretation of results, refer to Tab. 27-3 – Lectins.

27.6.11 Molecular genetic blood group determination

Molecular biological methods can be used to determine the genes for different blood group antigens. Blood group phenotypes can be predicted on the basis of single nucleotide polymorphisms. Molecular genetic investigations are based on large scale genotype-phenotype association studies and selected samples with rare phenotypes.

Molecular genetic methods are used for blood group determination in the following situations /471/:

  • Analysis of fetal blood group antigens in the case of maternal immunization and impending hemolytic disease of the newborn. Where possible, the examination is performed using fetal DNA in the maternal blood /7172/. In a pregnant Rh(D)-negative woman who is undergoing invasive procedures for other indications, the risk of hemolytic disease of the newborn due to anti-D can be predicted by analyzing samples obtained from chorionic biopsy or amniocentesis.
  • Blood group determination in previously transfused patients with unknown blood group antigens (e.g. when the transfusion history is incomplete) /73/
  • Determination of blood group antigens in autoimmune hemolysis and allo immune hemolysis, because in immune hemolytic anemia, it is often difficult to determine the antigen pattern of patients reliably because of autoagglutination, positive antiglobulin tests, and possibly ongoing transfusions
  • Investigation of blood group antigens in the case of discrepant or weak serological reactions /74/. Genotyping for RHD can be used to detect DEL, D chimerism, and very weak D variants in blood donors. In pregnant women with D variants, genotyping for weak D provides information about the risk of immunization and therefore about whether Rh immune globulin prophylaxis is indicated. Patients with serological type Fy(a–b–) can be investigated for the presence of the FY*B-33 or FY*X allele.
  • Determination of blood group antigens for which no antisera (or only very expensive antisera) exist
  • Determination of RHD zygosity in the partners of women with anti-D allo immunization and who are pregnant or wish to become pregnant
  • Characterization of test cells (e.g., for antibody identification and differentiation) to check the results of serological tests
  • Determination of rare antigens or antigen combinations in blood donors in cases where molecular genetic investigations are less expensive than serological methods
  • Paternity determinations and other forensic determinations.

27.7 Preparation of blood transfusion

The preparation of blood transfusion must fulfill the following objectives:

  • Ensure that erythrocyte transfusions are AB0 compatible (major compatibility)
  • Avoid unnecessary immunization, especially to Rh(D)
  • Detect clinically relevant and potentially clinically relevant erythrocyte antibodies with a sufficient degree of sensitivity
  • Reliably determine antigens on the erythrocytes of patients and blood donors to which corresponding antibodies are present.

In general terms, a blood group laboratory must enable a compatible transfusion of blood components to be provided within a reasonable period of time (Fig. 27-9 – Blood supply for elective surgery with risk of bleeding complications/75/. For more about the compatibility of blood components, refer to Section 27.2.4 – Blood-transfusion-related relevance of blood group antibodies.

27.7.1 Routine cases

As a general rule, a valid blood group and a current antibody screening result from the laboratory in charge must be available prior to any invasive or surgical procedure potentially associated with bleeding complications that may require transfusion. The examination should be initiated on the day of admission if possible.

If the antibody screen is negative, a sufficient number of compatible units of packed RBCs (in Germany, cross matched) should be reserved to meet the anticipated requirements for elective surgery and/or planned transfusions.

The cross match should be performed with a second blood sample on the day before surgery or transfusion if possible. An AB0 check or ABD confirmation test should be performed for each new blood sample in order to verify the identity.

27.7.2 Positive antibody screening test

A positive antibody screen must always be investigated further. Elective procedures associated with a risk of bleeding complications that may require transfusion must be postponed until the investigation has been completed.

Irregular antibodies

If clinically relevant irregular antibodies are present, the RBC units must be cross matched and the number of units provided needs to cover not only the regular anticipated requirements but also possible bleeding complications. This is done by providing appropriately antigen tested blood units (that are not cross matched) that are available either through the hospital’s own blood bank or regional transfusion services (refer to Tab. 27-24 – Consideration of irregular erythrocyte antibodies).

Undetermined irregular antibodies, incompatible transfusions

First of all, it must be established whether the antibodies are cold or warm antibodies. Generally, the following applies:

  • At most, cold antibodies are taken into account for planned transfusions or elective surgery
  • If the specificity of warm antibodies cannot be determined, this represents a particularly strong indication for autologous hemotherapy procedures
  • Transfusions of allogeneic RBC units should be restricted to life saving indications
  • In the case of undetermined allo antibodies, an attempt must be made to find compatible RBC units by cross matching with the patient’s serum. The reference test in this case is the indirect antiglobulin test using the albumin technique (tube test at 37 °C). If this is negative, there is usually no acute risk associated with the transfusion.
  • If only autoantibodies are present, searching for compatible units by cross matching is pointless
  • If blood components have to be transfused without prior investigation of the antibodies or as cross match incompatible products, the transfusion is stated to be of uncertain compatibility (this must be indicated on the accompanying form). The patient is managed in accordance with the guidelines for incompatible transfusions.

27.7.3 Urgent transfusions

The preparation of urgent and emergency transfusions depends on the time available and existing immunohematological findings.

Bleeding complications during elective procedures

In these cases, the results of blood grouping and the antibody screen are available.

  • If the antibody screen is negative, additional unmatched RBC units (or units that have been tested using a rapid cross match only) can be administered without hesitation, depending on the urgency of the bleeding complications. This also applies to the antigen tested blood units that should be held in reserve if antibodies are present.
  • Cold antibodies and/or antibodies that are rarely hemolytically active, such as anti-P1, anti-Lea, anti-Leb, and anti-M should be ignored in emergencies
  • If compatible blood units are not available in sufficient quantities and/or cannot be procured in the time available, the same recommendations apply as for cases in which undetermined irregular antibodies are present.

Emergency hospital admissions

In patients who are admitted to hospital as an emergency, the provision of blood supplies needs to be ensured in accordance with Tab. 27-25 – Timeline for emergency and massive transfusions in emergency admissions. If potential antibodies are found in the cross match and/or antibody screen, the same recommendations apply as for cases in which undetermined irregular antibodies are present.

27.8 Investigation of transfusion reactions

Neither the type nor the severity of transfusion reactions allows conclusions to be drawn regarding the cause. Therefore, the investigation should proceed according to a defined protocol (Tab. 27-26 – Investigation of transfusion reactions). Hemolysis, in particular intravascular hemolysis, must be ruled out in the first instance. Evidence of errors in the identification of patients or samples or the use of incompatible blood units must be sought, as well as erythrocyte incompatibilities that were not detected prior to transfusion. Even in emergency blood transfusions, intravascular hemolysis and AB0 mismatches should be ruled out prior to further transfusions. The possibility of bacterial contamination must also be considered. If the cause of the transfusion reaction remains unclear, further investigation is required. Screening for leukocyte antibodies, especially HLA antibodies, makes sense only if there are > 108 leukocytes in the transfused RBC units or if platelets have been transfused. This type of transfusion reaction has become rare since the introduction of universal leukocyte depletion for cellular blood components. In addition, HLA and granulocyte-specific antibodies should be sought in cases of suspected transfusion associated acute lung injury (TRALI syndrome), especially following the transfusion of plasma containing blood products from female donors. These investigations need to be performed in specialized laboratories.

In the immunohematological follow up examination, the test spectrum should be expanded and methodologically optimized in comparison to the pre transfusion diagnostic tests. It is imperative that sensitive methods such as gel centrifugation or solid phase technology are used for the investigation of transfusion reactions. The methodology can be optimized by, for example, using serum instead of plasma, washing donor erythrocytes thoroughly, prolonging the incubation time, and performing interim readings after incubation. The examinations must be conducted using both pre transfusion and post transfusion blood samples from the patient. On the one hand, the pre transfusion sample could belong to another patient (misidentification) while, on the other hand, antibodies that were overlooked prior to the transfusion may no longer be detectable because they are bound to the transfused erythrocytes.

If biochemical hemolysis parameters suggest the presence of hemolysis that cannot be explained by immunohematological investigations or by the hemolysis of transfused blood components, further tests (e.g., antibody identification or antibody elution, are required). Furthermore, these patients should be followed up after one week, by which time the levels of any antibodies present should have increased as a result of booster effects.

In the case of reactions to plasma containing blood units, minor incompatibilities should also be ruled out.

Because of secondary bacterial contamination of blood units, cultures of blood units are only useful if the culture bottles are inoculated under sterile conditions immediately after the transfusion has been discontinued. The culture should indicate the predominant growth of a single species of bacteria within a matter of hours and the same bacteria should also be cultivated from the patient’s post transfusion blood sample.

27.9 Quality assurance

Every laboratory that performs serological blood group examinations must implement quality assurance measures regularly, take part in internal and internal quality controls, and document the results accordingly /37/.


The reasons for conducting internal and external quality assurance in immunohematology are as follows:

  • Results cannot be validated clinically
  • Immunohematological methods are not quantitative
  • Biological reagents with a high degree of variability and storage instability are used
  • Reagents are subject to only limited controls by regulatory authorities
  • Reagents and methods need to be compatible
  • Numerous sources of potential methodological errors and patient-specific interference factors exist.


The objectives of quality assurance in immunohematology are:

  • Confirmation that the methods used are sufficiently sensitive and specific by means of external quality control (e.g., inter laboratory surveys)
  • Regular internal review of the quality of the reagents used
  • Regular internal inspection of the equipment used
  • Ongoing monitoring of the accuracy of test performance, test sensitivity, and test specificity by internal quality controls (in-process controls)
  • Identification and differentiation of patient-specific interference factors by internal quality control.

27.9.1 Internal quality control

Internal quality control includes in-process controls, reagent controls, and equipment controls.

In-process controls

The in-process controls (test controls) performed for the individual test methods are described in Section 27.6 – Immunohematological methods. They include positive and negative controls as well as auto-controls. In tube tests for the detection of antibodies, these controls must always be run (even in emergency examinations). In gel centrifugation tests and automated procedures in general, only auto-controls must always be run. For gel centrifugation tests, once-daily positive and negative controls are adequate.

Reagent controls

All patient samples, test cards, test cells, and other reagents must be inspected visually immediately prior to use in order to check the identity and ensure that no abnormalities (hemolysis, turbidity, precipitation) are present:

  • Test cards and reagents for detecting antibodies (antibody screening, cross match) must be checked once during each working day using positive and negative controls in the methods used. If automated equipment is used, the corresponding quality controls for the test cells must be performed at least once per shift.
  • A batch control must be carried out each time a new batch is used
  • The pH of unbuffered, opened wash solutions must be measured daily (target value ≥ 6.5).

All other reagents used on a daily basis should be examined under identical conditions (quality control location) at least once a week according to a standard protocol, if possible, in the test methods used. The results of these examinations should be compared with a target value protocol. If abnormalities are found, the test must be repeated. If the abnormal result is confirmed, the test is repeated again using a new sample from the same batch.

Depending on whether the quality of an individual sample or of the entire batch is compromised, the reagents involved must be discarded. Reagents that are used only sporadically (e.g., test antibodies for determining rare antigens) are checked as part of the tests in which they are used (by means of positive and negative controls).

Controls of the reactivity of reagents

The detection limit and specificity are checked as part of the weekly reagent controls. Due to the storage instability of various blood group antigens, it is recommended when examining test cells to use test antibodies directed against such antigens.

Equipment controls

The required equipment controls in immunohematology do not differ from the other equipment specific controls customarily performed in laboratories; some controls are also performed according to manufacturer specifications. Therefore, no further details are provided here.

27.9.2 External quality control

External quality control is primarily based on regular participation (at least twice, but preferably four times, each year) in inter laboratory surveys. As part of inter laboratory surveys, all parameters and target values that are investigated in a particular laboratory should always be checked.

In immunohematological laboratories, this applies in particular to AB0 and Rh blood groups, including phenotypes and weak variants; antibody screening, identification, and quantification; and the characterization of autoantibodies. If different methods are used for antibody screening and cross matches, these must be tested independently of each other.

Analyses for inter laboratory surveys must be performed with coded material under routine conditions.

Furthermore, any problematic immunohematological cases that cannot be clarified should be sent to reference laboratories for further investigation.

This chapter contains sections from earlier editions of this text book that were co-authored by Dr. H. H. Sonneborn, whose involvement as a co-author ceased following his subsequent retirement. We would like to express our sincere thanks to Dr. Sonneborn for allowing us to use the texts, tables, and figures in question.

We would also like to thank Ms. Karin Schmidt, former head TA at the Institute for Transfusion Medicine and Hemostaseology, University Hospital Marburg and Dr. med. Monika Weippert-Kretschmer for reviewing the manuscript and providing valuable advice regarding methodological details. We are also grateful to Dr. med. Adam for her support in producing the figures used in the chapter.


1. Mallinson G, Martin PG, Anstee DJ, Tanner MJA, Merry AH, Tills D, Sonneborn HH. Identification and partial characterisation of the human erythrocyte membrane component(s) which express the antigens of the LW blood group system. Biochem J 1986; 234: 649–52.

2. Tippett P. What is important on the red blood cell surface? Immunol Invest 1995; 24: 173–86.

3. Cartron JP, Colin Y. Structural and functional diversity of blood group antigens. Transfus Clin Biol 2001; 8: 163–99.

4. Flegel WA, Wagner FF. Blutgruppen: Alloantigene auf Erythrozyten. In: Kiefel V (ed). Transfusionsmedizin und Immunhämatologie, 4. Auflage. Heidelberg: Springer, 2010: 133–75.

5. Hakomori S: Antigen structure and genetic basis of histo-blood groups A, B and O: their changes associated with human cancer. Biochemica et Biophysica Acta 1999; 1473: 247–66.

6. Schonewille H; Brand A. Alloimmunization to red blood cell antigens after universal leucodepletion. A regional multicentre retrospective study. Brit J Haematol 2005; 129 (1): 151–6.

7. Schonewille H; Haak HL; van Zijl AM. Alloimmunization after blood transfusion in patients with hematologic and oncologic diseases. Transfusion 1999; 39 (7): 763–71.

8. Fluit CR, Kunst VA; Drenthe-Schonk AM. Incidence of red cell antibodies after multiple blood transfusion. Transfusion 1990; 30: 532–35.

9. Schonewille H; van de Watering LM; Brand A. Additional red blood cell alloantibodies after blood transfusions in a nonhematologic alloimmunized patient cohort: is it time to take precautionary measures? Transfusion 2006; 46 (4); 630–5.

10. Sonneborn HH, Dahr W, Helmbold W, Kasulke D. Monoklonale Antikörper in der Blutgruppenserologie. Infusionsther Transfusionsmed 1993; 20: 328–43.

11. Mollison PL, Engelfriet CP, Contreras M. Blood transfusion in clinical medicine. London: Blackwell Scientific, 10th edition 1997.

12. Redman M, Regan F, Contreras M: A prospective study of the incidence of red cell allo-immunisation following transfusion. Vox Sang 1996; 71: 216–20.

13. Kretschmer V, Heuckeroth A, Schulzki T, Dietrich G. Superiority of gel centrifugation in antibody screening and identification. Infusionsther 1992; 19: 226–30.

14. Ness PM, Shirey RS, Thomen SK, Buck SA: The differentiation of delayed serological and delayed hemolytic transfusion reactions: incidence, long-term serologic findings, and clinical significance. Transfusion 1990; 30: 688–93.

15. Pineda AA, Vamvakas EC, Gordon LD, Winters JL, Moore SB. Trends in the incidence of delayed haemolytic and delayed serologic transfusion reactions. Transfusion 1999; 39: 1097–1103.

16. Schonewille H; Haak HL; van Zijl AM. RBC antibody persistence. Transfusion 2009; 40 (9): 1127–31.

17. Koelewijn JM; Vrijkotte TG; van der Schoot CE; Bonsel GJ; de Haas M. Effect of screening for red cell antibodies, other than anti-D, to detect hemolytic disease of the fetus and newborn: a population study in the Netherlands. Transfusion 2008; 48 (5): 941–52.

18. Dajak S; Stefanovic V; Capkun V. Severe hemolytic disease of fetus and newborn caused by red blood cell antibodies undetected at first-trimester screening. Transfusion 2011; 51:1380–8.

19. Working party of the British Committee for Standards in Haematology Blood Transfusion Task Force. Guidelines for compatibility procedures in blood transfusion laboratories. Transfusion Medicine 2004; 14: 50–73.

20. Yamamoto F: Molecular genetics of ABO. Vox Sang 2000; 78 (Suppl 2): 91–103.

21. Voak D, Sonneborn H, Yates A. The A1(B) phenomenon: a monoclonal anti-B (BS85) demonstrates low levels of B determinants on A1 red cells. Transf Med 1992; 2 T: 119–27.

22. Fernandez-Mateos P, Cailleau A, Henry S, Costache M, Elmgre A, Svensson L, Larson G, Samuelsson BE, Oriol R, Mollicone R: Point mutations and deletion responsible for the Bombay H null and the Reunion H weak blood groups. Vox Sang 1998; 75: 37–46.

23. Schousboe K; Titlestad K; Baudier F; Hansen LU; Bistrup C. ABO-incompatible kidney transplantation. Danish Medical Bulletin 2010; 57 (10): A4197.

24. Tobian AA; Shirey RS; Montgomery RA; Cai W; Haas M; Ness PM; King KE. ABO antibody titer and risk of antibody-mediated rejection in ABO-incompatible renal transplantation. American Journal of Transplantation 2010; 10 (5): 1247–53.

25. Stewart ZA; Locke JE; Montgomery RA; Singer AL; Cameron AM; Segev DL. ABO-incompatible deceased donor liver transplantation in the United States: a national registry analysis. Liver Transplantation 2009; 15 (8): 883–93.

26. Dipchand AI; Pollock BarZiv SM; Manlhiot C; West LJ; Van der Vliet M; McCrindle BW. Equivalent outcomes for pediatric heart transplantation recipients: ABO-blood group incompatible versus ABO-compatible. American Journal of Transplantation 2010; 10 (2): 389–97.

27. Benjamin RJ, McGurk S, Ralston MS, Churchill WH, Antin JH: ABO incompatibility as an adverse risk factor for survival after allogeneic bone marrow transplantation. Transfusion 1999; 39: 179–87.

28. Boratynska M; Banasik M; Halon A; Patrzalek D; Klinger M. Blood group Lewis alloantibodies cause antibody-mediated rejection in renal transplant recipients. Transplantation Proceedings 2007; 39 (9): 2711–4.

29. Schmitz G, Sonneborn HH, Ernst M, Blanchard D, Gielen W, Dahr, W. The effect of cysteine modification and proteinases on the major antigens (D, C, c, E and e) of the Rh blood group system. Vox Sang 1996; 70: 34–9.

30. Körmöczi GF, Gassner C, Shao CP, Uchikawa M, Legler TJ. A comprehensive analysis of DEL types: partial DEL individuals are prone to anti-D alloimmunization. Transfusion. 2005; 45: 1561–7.

31. Wagner T, Körmöczi GF, Buchta C, Vadon M, Lanzer G, Mayr WR, Legler TJ. Anti-D immunization by DEL red blood cells. Transfusion. 2005; 45: 520–6.

32. Yasuda H, Ohto H, Sakuma S, Ishikawa Y. Secondary anti-D immunization by Del red blood cells. Transfusion 2005; 45: 1581–4.

33. Lomas C, Tippett P, Mannessier L. Abolition of the DVc subcategory. Transfusion 1993; 27: 535.

34. Bailly P, Hermand P, Callebaut I, Sonneborn HH, Khamlichi S, Morou JP, Cartron JP. The LW blood group glycoprotein is homologous to intercellular adhesion molecules. Proc Natl Acad Sci 1994; 91: 5306–10.

35. Marsh WL, Redman CM. The Kell blood group system: a review. Transfusion 1990; 30: 158–67.

36. Daniels G. Human Blood Groups. Second Edition, Blackwell Science Ltd. 2002

37. Richtlinien der Bundesärztekammer zur Gewinnung von Blut und Blutbestandteilen und zur Anwendung von Blutprodukten (Hämotherapie). Bundesanzeiger 2010; 62 (101).

38. Wagner FF, Flegel WA. RHD gene deletion occurred in the Rhesus box. Blood 2000; 95: 3662–8.

39. Perco P, Shao CP, Mayr WR, Panzer S, Legler TJ. Testing for the D zygosity with three different methods revealed altered Rhesus boxes and a new weak D type. Transfusion 2003; 43: 335–9.

40. Arndt PA; Garratty G. A retrospective analysis of the value of monocyte monolayer assay results for predicting the clinical significance of blood group alloantibodies. Transfusion 2004; 44 (9): 1273–81.

41. Mari G for the Collaborative Group for Doppler Assessment of the Blood Velocity in Anemic Fetuses: Noninvasive diagnosis by doppler ultrasonography of fetal anemia due to maternal red cell alloimmunization. N Engl J Med 2000; 342: 9–14.

42. Müller SP, Bartels I, Stein W, Emons G, Gutensohn K, Köhler M, Legler TJ. The determination of the fetal RhD status from maternal plasma for decision making on Rh prophylaxis is feasible. Transfusion 2008; 48: 2292–301.

43. Gutensohn K, Müller SP, Thomann K, Stein W, Suren A, Körtge-Jung S, Schlüter G, Legler TJ. Diagnostic accuracy of noninvasive polymerase chain reaction testing for the determination of fetal rhesus C, c and E status in early pregnancy. BJOG 2010; 117: 722–9.

44. Kroll H, Maier RF. Perinatale und pädiatrische Transfusionsmedizin. In: Kiefel V (ed). Transfusionsmedizin und Immunhämatologie, 4. Auflage. Heidelberg: Springer, 2010: 420–48.

45. Barcellini W; Revelli N; Imperiali FG; Villa MA; Manera MC; Paccapelo C; Zaninoni A; Zanella A. Comparison of traditional methods and mitogen-stimulated direct antiglobulin test for detection of anti-red blood cell autoimmunity. International Journal of Haematology 2010; 91 (5): 762–9.

46. Petz LD, Garratty G. Immune hemolytic anemias. New York: Churchill Livingstone, 2004.

47. Lynen R, Krone O, Legler TJ, Köhler M, Mayr WR. A newly developed gel centrifugation test for quantification of RBC-bound IgG antibodies and their subclasses IgG1 and IgG3: comparison with flow cytometry. Transfusion 2002; 42: 612–8.

48. Chaudhary R; Das SS; Gupta R; Khetan D. Application of flow cytometry in detection of red-cell-bound IgG in Coombs-negative AIHA. Hematology (Amsterdam, Netherlands) 2006; 11 (4): 295–300.

49. Roelcke D. Cold agglutination. Transfusion Medicine Reviews 1989; 3: 140–66.

50. Gottsche B, Salama A, Mueller-Eckhardt C. Donath-Landsteiner autoimmune haemolytic anaemia in children. Vox Sang 1990; 58: 281–6.

51. Garratty G, Arndt PA. An update on drug-induced immune hemolytic anemia. Immunohematology 2007; 23 (3): 105–119.

52. Johnson ST; Fueger JT; Gottschall JL. One center’s experience: the serology and drugs associated with drug-induced immune hemolytic anemia – a new paradigm. Transfusion 2007; 47 (4): 697–702.

53. Technical Manual of the American Association of Blood Banks, 17th edition, Bethesda, 2010.

54. Marsh WL, Nichols ME, Margaret E, Jenkins WJ. Automated detection of blood group antibodies. J Med Lab Technol 1968; 25: 275.

55. Lalezari P. A new method for detection of red cell antibodies. Transfusion 1968; 8: 372.

56. Rogge Th, Fabricius HA: Reaktionsverhalten von Antiseren der Spezifitäten Anti-S, Anti-s, Anti-Fya, anti-Fyb, Anti-Jka und Anti-Jkb bei Verwendung des Gel-Zentrifugationstests. J Lab Med 1998; 22: 647 (B-6).

57. Eichler H, Klinge U, Kretschmer V. Gel test and column agglutination technology: comparative study of two red cell antibody screening and identification systems. Lab Med 1995; 19: 22–6.

58. Lapierre Y, Rigal D, Adam J, Josef D, Greber S, Drot C. The gel test: a new way to detect red cell antigen-antibody reactions. Transfusion 1990; 30: 109–13.

59. Hitzler W. Vergleichsstudie zur Antikörperidentifizierung im Geltest (ID-Microtyping-System), Festphasen-Antihumanglobulintest (Solidscreen, Capture R, Ready ID) und Röhrchentest. Ärztl Lab 1991; 37: 137–46.

60. Rünz T, Wisser, H. Bitherme Inkubation – eine Optimierung des Antikörperscreenings mit Hilfe des Geltestes. Klin Lab 1995; 3: 107–11.

61. Zeiler T, Thiele S, Kretschmer V. Festphasentechnik versus Gelzentrifugation zum Nachweis erythrozytärer Antikörper – eine prospektive Studie zum Vergleich zweier Antikörpersuchtests. In: Sibrowski W, Stangel W, Blauhut B (eds). Transfusionsmedizin, Beiträge Infusionsther Transfusionsmed, Vol 27. Basel: Karger, 1996: 17–21.

62. Schrem A, Flegel WA. Comparison of solid-phase antibody screening tests with pooled red cells in blood donors. Vox Sang 1996; 71: 37–42.

63. Bouix O., Ferrera V., Delamaire M., Redersdorff J.C., Roubinet F.: Erythrocyte-magnetized technology:  an original and innovative method for blood group serology. Transfusion 2008; 48 (9); 1878-85.

64. Schoenfeld H., Bulling K, von Heymann C. Neuner B., Kalus U., Kiesewetter H., Pruss A. Evaluation of immunohematologic routine methods using the new erythrocyte-magnetized technology on the QWALYS 2 system. Transfusion 2008; 48: 1878–85.

65. Schoenfeld H; Pretzel KJ; von Heymann C; Neuner B; Kalus U; Kiesewetter H; Pruss A. Validation of a hospital-laboratory workstation for immunohematologic methods. Transfusion; 2009; 50 (1); 26–31.

66. Flegel WA, Wagner FF. The frequency of RHD protein variants in Caucasians. Transfus Clin Biol 1996; 3: 10s (Abstract P1A4).

67. Kretschmer V, Weippert-Kretschmer M: Notfall- und Massivtransfusion. In: Kiefel V (ed). Transfusionsmedizin und Immunhämatologie, 4. Auflage. Heidelberg: Springer, 2010; 405–18.

68. Lange J, Selleng K, Heddle NM, Traore A, Greinacher A. Coombs’ crossmatch after negative antibody screening – a retrospective observational study comparing the tube test and microcolumn technology. Vox Sang 2009; 98 (3): 269–75.

69. Schonewille H; van Zijl AM; Wijermans PW. The importance of antibodies against low-incidence RBC antigens in complete and abbreviated cross-matching. Transfusion 2003; 43 (7): 939–44.

70. Beck KH, Eichler H, Kretschmer V. Vakuum-Äther-Elution. Eine Modifikation der Ätherelutionsmethode nach Rubin. In: Sibrowski W, Stangel W, Wegener S (eds). Transfusionsmedizin, Beiträge Infusionsther Transfusionsmed. Basel: Karger, 1994: 167–8.

71. Legler TJ, Lynen R, Maas JH, Pindur G, Kulenkampff D, Suren A, Osmers R, Köhler M. Prediction of fetal Rh D and Rh CcEe phenotype from maternal plasma with real-time polymerase chain reaction. Transfus Apher Sci. 2002; 27: 217–23.

72. Bettelheim D, Panzer S, Reesink HW, Csapo B, Pessoa C, Guerra F, Wendel S, Calda P, Sprogøe U, Dziegiel M, Aitokallio-Tallberg A, Koskinen S, Kuosmanen M, Legler TJ, Stein W, Villa S, Villa MA, Trespidi L, Acaia B, Vandenbussche FP, Brand A, de Haas M, Kanhai HH, Gounder D, Flanagan P, Donegan R, Parry E, Sefonte C, Skulstad SM, Hervig T, Flesland Ø, Zupańska B, Uhrynowska M, Lapaire O, Zhong XY, Holzgreve W. Monitoring and treatment of anti-D in pregnancy. Vox Sang. 2010; 99: 177–92.

73. Legler TJ, Eber SW, Lakomek M, Lynen R, Maas JH, Pekrun A, Repas-Humpe M, Schröter W, Köhler M. Application of RHD and RHCE genotyping for correct blood group determination in chronically transfused patients. Transfusion. 1999; 39: 852–5.

74. Legler TJ, Maas JH, Köhler M, Wagner T, Daniels GL, Perco P, Panzer S. RHD sequencing: a new tool for decision making on transfusion therapy and provision of Rh prophylaxis. Transfus Med 2001; 11: 383–8.

75. Kretschmer v, Gombotz H, Rump G. Transfusionsmedizin – Klinische Hämotherapie. Thieme Verlag, Stuttgart, New York, 2008.

Table 27-1 Blood group antigens on the erythrocytes of newborns



A, B, H, I

Lea, Leb

Lua, Lub




M, N, S, s


Table 27-2 Biological function of blood group antigens on erythrocytes, modified from Ref. /4/


Most important antigens
and frequency (%)

Number of possible blood groups*

Function of the antigen
carrying molecules


A (42), A1 (27), B (11), H (67)

> 20*



I (> 99.9), i (> 99.9)

>  2*


i (> 99.9)


Lea (22), Leb (72)


Binding of H. pylori


P1 (80), P (> 99.9), p (< 0.1), Pk (0.1)


Binding of E. coli, Parvovirus B19


D (85), C (70), E (30), c (80), e (98)

> 50*

ammonium transport


K (9), k (99.8), Kpa (2), Kpb (> 99.9), Jsa (< 1), Jsb (> 99.9)

> 20*



Fya (67), Fyb (80.5), Fy3 (> 99.9), Fy4 (< 0.1)


Binding of IL-8,
malaria receptors (Fya/Fyb)


Jka (76.9), Jkb (72.5), Jk3 (> 99.9)


Urea transport


M (78), N (72), S (55), s (89), U (> 99.9), Ena (99.9)


malaria receptors


Lua (7.7), Lub (92.3)


Receptor (adhesion molecules)


Xga* (men 67, women 89)




Dia (< 0.01), Dib (> 99.9), Wra (0.08), Wrb (> 99.9)


Structure (band 3),
chloride and bicarbonate transport


Yta (99.7), Ytb (8.1)




Doa (66.7), Dob (82.8)


ADP ribosyltransferase


Coa (99.8), Cob (8.5)


H2O transport


Vel (> 9.9)




Bga (2), Bgb (2), Bgc (1)

> 3*


* Including variants

Table 27-3 Lectins


Derived from

React with


Dolichos biflorus

A1 (Tn, Cad)


Molucella laevis

A and N


Vicia graminea

N and T*


Arachis hypogaea



Salvia sclarea



Leonurus cardiaca

Cad and Sd


Lotus tetragonolobus

A2, A2B, A3 etc., O, B


Ulex europaeus

A2, A2B, A3 etc., O, B


Helix pomatia


* Refer to Section 27.4.15 (Cryptantigens)

Table 27-4 Impact of enzymes on blood group antigens



Fya, Fyb; K, k; Xga

Rh (especially E); Jka, Jkb; Kna, Kpb, LKE, Lua, Lea, P1, Coa, Cob, LW


Pr, Ina, Inb

I, i; Gd; T, Tn, Tk; EL, Ena

Table 27-5 Causes of pseudoagglutination

Plasma expanders (dextran, HES*)

Coagulation disorders, incomplete coagulation of the blood sample

Hyperfibrinogenemia, fibrin degradation products, fibrinolytic agents

Dysproteinemia, monoclonal Ig, cryoglobulinemia

Increase in acute-phase proteins α1-glycoprotein, α2-macroglobulin, α1-antitrypsin, ceruloplasmin, and IgM

Drugs (e.g., high dose heparin, protamine sulfate)

Increase in free hemoglobin, cholesterol, bilirubin

Rheumatoid factors

Leukocytosis, leukocyte enzymes

Wharton’s jelly (hyaluronidase)

* HES, hydroxyethyl starch

Table 27-6 Frequency of irregular erythrocyte antibodies /13/*



Individual specificity/number



D 11; E 7; C 1; Cw 1; C,D 5; c,E 2; c,Cw 1; D,E 1; C,D,E 3;



K 4; K, Jka 2



Jka 2; Jkb 1, Jka,K 2



Fya 2, Fya,C,D,E 1



S 2



Lea 19, Leb 3, Leab 1



P1 12



Warm AAB 5

* Investigation of 3000 patients using the gel centrifugation test, enzyme test, and indirect antiglobulin test

** One serum also contained anti-Fya

*** AAB = Autoantibodies

Table 27-7 Density of ABO antigens (molecules) per erythrocyte



Number of antigens


























O cord




Table 27-8 The most important ABH variants


Erythrocyte reaction with



Phenotype frequency
in Caucasians**












A1 34%, A1B 3%








A2 10%, A2B 1%








1–2 of 200 A


± MF

± MF





2 of 150,000 A


± MF

± MF



1 of 6000 A


± MF

± MF





4% of A in South Africa








1 of 40,000–80,000 A

Am (Ay)

– (±)





1 of 150,000 A







Very rare














3 of 350,000 B


























Very rare




Very rare






46 of 500,000 Chinese

* When anti-B is adsorbed to rabbit erythrocytes, ** Europeans, unless otherwise specified; MF, mixed field agglutination

Table 27-9 ABO blood groups: distribution and detectability of regular and irregular antibodies










































* German population

Table 27-10 Lewis system: genes and phenotypes



in the saliva






LeLe/ Lele

HH/ Hh




Leb (Lea)*

LeLe/ Lele

HH/ Hh






HH/ Hh






HH/ Hh




Lele/ Lele




Very rare






Very rare






Very rare





Very rare

* ABH substance according to the ABO blood group

Table 27-11 P system



Reaction with anti-


















Rarely HTR



Tja = P,P1,Pk















* Extremely rare; HTR, hemolytic transfusion reaction; HDN, hemolytic disease of the newborn; PCH, paroxysmal cold hemoglobinuria

Table 27-12 Antigens of the Rh system*





RH 1




RH 2




RH 3




RH 4




RH 5




RH 6

ce(f) f



RH 7




RH 8




RH 9



< 1

RH 10

C(ces) ces

rhv hrv

< 1

RH 11


rhw2 rh2w

< 1

RH 12




RH 17



> 99

RH 18



> 99

RH 19




RH 20

VS(es) es,


< 1

RH 21




RH 22



< 1

RH 23



< 1

RH 26




RH 27




RH 28



< 1

RH 29

(total Rh’)


> 99

RH 30



< 1

RH 31




RH 32



< 1

RH 33



< 1

RH 34



> 99

RH 35



< 1

RH 36



< 1

RH 37



< 1

RH 39



> 99

RH 40



< 1

RH 41




RH 42

Ces Cces


< 1

RH 43



< 1

RH 44



> 99.9

RH 45



< 1

RH 46



> 99.9

RH 47



> 99.9

RH 48



< 1

RH 49




RH 50




RH 51




* Nomenclature of Rosenfield, Fisher and Race, and Wiener. European population.

Table 27-13 Frequency of Rh phenotypes and genotypes





Fisher, Race









































































* Nomenclature of Rosenfield, Fisher and Race, and Wiener.

Table 27-14 D epitopes on D variant cells /33/





















































































Table 27-15 The most important antigens of the Kell system





Kell (K)


K pos.




k pos.




Kpa pos.




Kpb pos.

> 99.9




Kpc pos.

< 0.1



Jsa pos.

< 1.0



Jsb pos.

> 99.9

* Numerical nomenclature of the ISBT /3/, ** according to the guidelines of the German Medical Association /37/

Table 27-16 The most important Duffy antigens in Caucasians















* Identical to the phenotype

Table 27-17 Kidd antigens















* Identical to the phenotype

Table 27-18 Antigens and phenotypes in the MNS system

































































* Identical to the phenotype, ** rarely in Black individuals

Table 27-19 The most important Lutheran antigens















* Identical to the phenotype

Table 27-20 Determination of the specificity of cold autoantibodies, based on examples /49/*


Adult erythrocytes

Cord blood
















< 8

< 8





< 8


* Titers indicated, ** Reaction of enzyme treated cord blood erythrocytes is analogous to that of adult erythrocytes

Table 27-21 Therapeutic drugs that can induce immune hemolysis*



4-aminosalicylic acid














  • 1st generation


  • 2nd generation


  • 3rd generation








































































* Modified from Ref. /51/

Abbreviations: AI, autoimmune; DA, drug adsorption; IC, immune complex; NPA, nonspecific protein adsorption

Table 27-22 Problems associated with blood group determination*

Weak antigens

  • Neonates and premature infants*
  • Hemoblastoses*
  • Weak ABO phenotypes, weak D, partial D*
  • High serum concentration of soluble A antigens associated with malignancy

Accessory antigens*


Following bone marrow transplantation

Bombay and para Bombay blood groups

Prior transfusion of blood with different blood group

Polyagglutination* (refer to Section 27.3.4 – P system)

Absent or decreased isoagglutinins

  • Neonates and premature infants
  • Congenital/acquired antibody deficiency syndrome
  • Advanced age
  • Major blood loss

Irregular ABO antibodies (weak phenotypes)

Irregular alloantibodies or autoantibodies (especially cold antibodies)

Pseudoagglutination (refer to Section 27.2.3 – Serological reactivity)

* Less common if monoclonal test antibodies are used.

Table 27-23 Safety measures when using automated blood group analyzers

  • Error protocols
  • Detailed error journals with instructions for error resolution
  • Printout of interim reports and findings, automatic data storage
  • Memory system to enable analysis to continue following an interruption
  • Remote maintenance via secure internet connection (hotline)
  • Backup devices
  • Option to resume analysis manually in the event of an interruption
  • Use of the same procedures as those used for manual processing
  • Specially trained team supervisors (troubleshooting, minor repairs)
  • Company service on site within 12–24 h

Table 27-24 Consideration of irregular erythrocyte antibodies /19/





Anti-A1, anti-HI
with A1 and A1B

crossmatch at 37 °C


-C, -c, -E, -e

Antigen negative*


Negative crossmatch


Anti-K, -k

Antigen negative*


Negative crossmatch


Anti-Jka, -Jkb

Antigen negative*


active at 37 °C,
anti-S, -s, -U

Antigen negative*

inactive at 37 °C,

Negative crossmatch
at 37 °C


Anti-Fya, -Fyb

Antigen negative*



Negative crossmatch
at 37 °C


Anti-Lea, -Leb

Negative crossmatch
at 37 °C



Negative crossmatch
at 37 °C



Negative crossmatch

* Antigen negative and negative crossmatch

Table 27-25 Timeline for emergency and massive transfusions in emergency admissions

Immediate transfusion required

  • Blood sample for complete blood count, coagulation tests, electrolytes, creatinine, AB0 and Rh determination, antibody screen, crossmatch
  • 4–6 units of unmatched group 0 packed RBCs (Rh negative if possible), FFP of blood group AB
  • AB0 bedside test with the patient’s blood

Transfusion required within approximately 10 min

  • Regular AB0/Rh determination
  • 4–6 units of unmatched AB0/Rh identical or ABO/Rh compatible packed RBCs
  • AB0 identical FFP
  • AB0 bedside test with the patient’s blood

Transfusion required within approximately 30 min

  • ≤ 10 units of AB0/Rh identical or AB0/Rh compatible packed RBCs, crossmatched using a rapid or emergency crossmatch: major crossmatch in the indirect antiglobulin test (gel centrifugation test) including autocontrols; depending on urgency, incubation for 5–10 min at 37 °C
  • AB0-identical FFP
  • AB0 bedside test with the patient’s blood

Transfusion required within approximately 45 min

  • ≤ 10 units of AB0/Rh identical or AB0/Rh compatible packed RBCs, crossmatched using a standard crossmatch (gel centrifugation test)
  • ABO-identical FFP
  • ABO bedside test with the patient’s blood

Transfusion not required in the next hour

  • Standard antibody screen in the recipient, using the serum/plasma of the initial blood sample if possible
  • Further procedure as described for transfusion in 45 min

FFP, fresh frozen plasma

Table 27-26 Investigation of transfusion reactions

1. Secure the blood units involved, pilot tubes, and pre transfusion patient blood samples.

2. Perform bacteriological cultures (aerobic and anaerobic) of the patient’s blood and the blood units.

3. Obtain post transfusion samples of the patient’s blood for determination of hemolysis parameters (EDTA blood, citrated blood, and clotted blood) and immunohematological investigations (clotted blood and EDTA blood or citrated blood).

4. Assess the supernatant after centrifugation of the blood samples (free Hb) in order to detect hemolysis in the blood units and in post transfusion samples of the patient’s blood (EDTA blood or citrated blood)**.

5. Recheck the AB0 and Rh blood groups of the blood units (or of the pilot tube, if the blood units are no longer available) and of the patient (pre transfusion and post transfusion blood sample)**; units of plasma or platelets can be investigated by determining the isoagglutinins.

6. Repeat cross matching of the pre transfusion and post transfusion patient blood samples and, if possible, RBCs from the blood units (tube segment) using enhanced and optimized methods*; in the case of plasma containing units, these tests can also be performed as minor cross matches.

7. Perform a direct anti-globulin test on the post transfusion blood sample (citrated blood or EDTA blood) using a highly sensitive method (gel centrifugation) and multiple polyspecific Coombs sera*.

8. Repeat antibody screening on pre transfusion and post transfusion patient blood samples (clotted blood) using enhanced and optimized methods*, and, if necessary, also on donor plasma; if hemolysis is present, repeat after one week.

9. If irregular antibodies are present, determine the corresponding antigens in the pre transfusion patient blood sample and the blood units.

10. Perform further tests in order to detect antibodies to RBCs if signs of hemolysis are present or if one of the tests listed in points 1 to 8 yields an abnormal result.

11. Screen for HLA antibodies if the examinations listed in points 2 to 8 are negative, possibly also in the donor plasma.

12. Screen for HLA specific and granulocyte specific antibodies if the TRALI syndrome is suspected.

13. Determine IgA if the examinations listed in points 2 to 8 and point 11 are negative.

* In general prior to further blood transfusions, ** Also in emergencies

Figure 27-1 Biosynthesis of the blood group antigens A and B.

Gluc Gal Fuc Gal Gal Gluc Fuc Gal Gal Gluc Fuc AcNH R (R residue) AcNH B AcNH A R UDP UDP-GalNac A gene α-3-N-acetyl-galactosaminyl-transferase UDP UDP-Gal B gene α-3-D-galactosyl-transferase H (0) R AcNH

Figure 27-2 Rh transmembrane protein and assignment to the exons of the RH genes.

Csy ( C) Trp ( c) Ile ( C) Leu ( c) Ser ( C) Pro ( c) Pro ( C) Ala ( c) Ser ( C) Asn ( c) Rh-trans- membrane protein Outside Inside Er ythrocyte- membrane COOH 417 amino acid residues NH 2 1 112 162 33 53 117 284 103 152 226 98 60 68 12 16 81 132 139 186 207 212 259 273 285 369 358 291 384 389 309 332 313 408 49 = cysteine residues = differences in amino acid sequence between RhD and RhCE proteins. * very similar and also consisting of 10 exons, is the RHCE gene. The numbers mean amino acid positions in the protein. RHD gene* exons 1 2 3 4 5 6 7 8 9 10

Figure 27-3 Performance of the indirect anti-globulin test for the detection of serum antibodies (tube test)

1 drop test-erythro-cytes 2 dropspatientserum Mixing 2 dropssupplement Mixing,incubate 3 x washing/decantation 2 dropsCoombsserum/mixing Centrifu-gation andreading + + + + + +

Figure 27-4 Principle of the gel centrifugation test. The micro column on the left contains antibodies bound to gel globules and the erythrocytes to be tested prior to centrifugation. The center micro column shows a positive reaction following centrifugation, while the micro column on the right shows a negative reaction.

Antibodies Erythrocytes ++++ Gel spheres

Figure 27-5 Determination of AB0 blood group antigens (left) and AB0 antibodies (right) on uncoated microtitre plates. The blood of the tested individual is B.

Antisera Negative Positive Incubation and centrifugation Serum + A1-+ Proband serum + Erythrocytes+ Anti-A + Anti-B B- O-Ery Negative Negative Positive

Figure 27-6 Determination of AB0 blood group antigens (left) and AB0 antibodies (right) on antibody coated or antigen-coated microtitre plates (solid phase technology). The blood of the tested individual is B.

Serum Negative Negative Positive Positive Anti-A Anti-B A1-Ag + Proband erythrocytes + Proband serum + Fermented test cells A B B-Ag Plate coated with: Incubationwashing Centrifugation

Figure 27-7 Determination of irregular erythrocyte antibodies using microtitre plates coated with erythrocytes/erythrocyte membranes (solid phase technology).

Ag coated plates + Serum/incubation + Anti-human lgGloadedindicator cells Serum Ab with serum Wash Positive negative aftercentrifugation + Anti-human lgGloadedindicator cells

Figure 27-8 Determination of irregular erythrocyte antibodies using protein A-coated test plates.

Serum Protein Acoatedplates Test erythrocytes +serumincubation Mixed Positive Negative Positive Negative aftercentrifugation Wash + + + Ab Coombs serum

Figure 27-9 Blood supply for elective surgery with risk of bleeding complications.

Negative PreparationAB0/Rh + Antibody screening test Conspicuous Always clarify a. Cold antibody (Anti-x) b. Transfusion related Ab (Anti-x) Provide normal requirement, AB0/Rh equal, -compatible,antigen x-negative, crossbred Provide normal requirement,AB0/Rh equal, crossbred Ensure reserve requirements,AB0/Rh compatible uncrossed Ensure reserve requirement, AB0/Rh-compatible antigen x-negative, uncrossed Blood management practices, autologous procedural hemotherapy Blood management practices, autologous procedural hemotherapy On complications depending on urgency uncrossed/abbreviated crossed transfusions + + + + a. b.
Goto top <