Articles on Medical Diseases and Conditions

Tag: Immunohematology

  • White Blood Cell Antigens

    The RBC ABO surface antigens are found in most tissues except the central nervous system (CNS). Some of the other RBC antigens, such as the P system, may occur in some locations outside the RBCs. White blood cells also possess a complex antigen group that is found in other tissues; more specifically, in nucleated cells. This is called the human leukocyte-A (HLA) system and is found in one site (locus) on chromosome number 6. Each locus is composed of four subloci. Each of the four subloci contains one gene. Each sublocus (gene) has multiple alleles (i.e., a pool of several genes), any one of which can be selected as the single gene for a sublocus. The four major subloci are currently designated A, B, C, and D. There is possibly a fifth sublocus, designated DR (D-related), either close to the D locus or part of it.

    HLA-A, B, and C are known as class I antigens. They have similar structure, including one polypeptide heavy chain, and can be identified using standard antiserum (antibody) methods. The class II antigen HLA-D is identified by the mixed lymphocyte culture test in which reagent lymphocytes with HLA-D antigen fail to stimulate proliferation of patient lymphocytes when patient lymphocytes havethe same HLA-D antigen but will stimulate proliferation if the patient HLA-D antigens are not compatible. HLA-DR is classified as a class II antigen with a structure that includes two polypeptide heavy chains. It includes a group of antigens found on the surface of B-lymphocytes (B antigen) and also in certain other cells such as monocytes but not in most T-lymphocytes. HLA-DR is currently tested for by antibody methods using patient lymphocytes and antibody against DR antigen (microcytotoxicity test). Two other antigen groups, MB and MT, which are closely associated with HLA-DR, have been described.

    The four subloci that form one locus are all inherited as a group (linked) in a manner analogous to the Fisher-Race theory of Rh inheritance. Again analogous to Rh, some HLA gene combinations are found more frequently than others.

    The HLA system has been closely identified with tissue transplant compatibility to such a degree that some refer to HLA as histocompatibility leukocyte-A. It has been shown that HLA antigens introduced into a recipient by skin grafting stimulate production of antibodies against the antigens that the recipient lacks, and that prior sensitization by donor leukocytes produces accelerated graft rejection. In kidney transplants from members of the same family, transplant survival was found to correlate with closeness of HLA matching between donor and recipient. On the other hand, there is evidence that HLA is not the only factor involved, since cadaver transplants frequently do not behave in the manner predicted by closeness of HLA typing using HLA-A and B antigens. There is some evidence that HLA-D, DR, and MB antigens may also be important in renal transplant compatibility.

    Platelets contain HLA antigens, and patients who receive repeated transfusions of platelets may become refractory to such transfusions due to immunization against HLA antigens. Transfusion of HLA-A and B compatible platelets improves the success rate of the platelet units. However, about one third of platelet transfusion units containing well-matched HLA-A and B platelets will not be successful once the patient is sufficiently immunized.

    HLA antigens on each chromosome are inherited as a unit in a mendelian dominant fashion. Therefore, HLA typing has proved very useful in paternity case investigations.

    Besides their association with immunologic body defenses, certain HLA antigens have been found to occur with increased frequency in various diseases. The B27 antigen is associated with so-called rheumatoid arthritis (RA) variants (Chapter 23). In ankylosing spondylitis, Reiter’s syndrome, and Yersinia enterocolitica arthritis, HLA-B27 occurs in a very high percentage of cases. The incidence of HLA-B27 in ankylosing spondylitis is 90%-95% (range, 83%-96%) in Europeans and approximately50% in African Americans. In Reiter’s syndrome the incidence is 80%-90% (range, 63%-100%) in Europeans and approximately 35% in African Americans. In juvenile rheumatoid, psoriatic, and enteropathic (ulcerative colitis and Crohn’s disease) arthritis, the incidenceof HLA-B27 depends on the presence of spondylitis or sacroiliitis. In all RA-variant patients, those with spondylitis or sacroiliitis have B27 in more than 50% of cases (some report as high as 70%-95%); without clinical disease in these locations, B27 is found in less than 25%. Increased frequency of the B27 antigen was also reported in close relatives of patients with ankylosing spondylitis.

    An increased incidence of certain other HLA antigens has been reported in celiac disease (HLA-B8), chronic active hepatitis, and multiple sclerosis (as well as in various other diseases) but with lesser degrees of correlation than in the RA variants. The significance of this is still uncertain, and verification is needed in some instances.

  • Type and Screen

    Current recommended procedure for pretransfusion testing, as previously described, is to obtain the ABO and Rh type of the recipient RBCs and perform an antibody screen on the serum of the recipient. This has become known as type and screen. This is followed by a crossmatch on blood units actually transfused; only the immediate spin procedure is mandated by AABB rules, but testing can be more extensive. Except in certain emergencies, physicians who anticipate need for blood order an estimated amount (number of units) to be processed in case they are needed. For many years, all of these units were typed and crossmatched immediately. Some institutions now maintain a “maximum surgical blood order schedule” in which the average blood need for various surgical procedures is calculated and crossmatch is performed routinely only on these units, with others subjected only to type and screen unless actually needed. In other institutions, only type and screen is done routinely, but when the order to actually transfuse is given, a crossmatch is performed. In either case, since many blood units are ordered that are never transfused, type and screen decreases the number of crossmatches required. Some blood bankers maintain that even the immediate spin can be eliminated with acceptable safety, thereby transfusing without any crossmatch.

  • Pretransfusion Test Considerations

    A word must also be said regarding a few patients whose blood presents unexplained difficulty in crossmatching. The laboratory should be allowed to solve the problem and possibly to obtain aid from a reference laboratory. During this time 5% serum albumin or saline may temporarily assist the patient. In the absence of complete crossmatching, blood is given as a calculated risk.

    If blood is needed for emergency transfusion without crossmatch, a frequent decision is to use group O Rh-negative blood. The rationale is that no recipient (whether group O or any other ABO group) could have ABO group antibodies against group O cells, and any unexpected antibodies in donor serum against recipient RBCs would be diluted by the blood volume of the recipient. Even so, there is risk involved, since the recipient may possess antibody to some RBC non-ABO blood group antigen of the donor (e.g., anti-Rh or anti-Kell). A crossmatch would detect this. Moreover, the anti-A or anti-B antibodies in group O blood may be in high titer, and transfusion reactions may occur when this blood is used inrecipients who are group A, B, or AB. Many blood banks maintain a certain amount of low-titer O-negative blood for use in emergencies. Titers over 1:50 are considered too high for this purpose. In addition, A and B group–specific substance (Witebsky substance) may be added to the donor blood to partially neutralize anti-A and anti-B antibodies. These substances are A and B antigens manufactured from animal sources and, being foreign antigens, may sensitize the patient.

    Rather than use group O Rh-negative blood in a blind fashion for emergencies, a better method is to use blood of the same ABO and Rh type as the patient’s blood. ABO and Rh typing can be done in 5 minutes using anticoagulated specimens of the patient’s blood. This avoids interpretation problems produced by putting group O cells into a group A or B patient and subsequently attempting crossmatches for more blood.

    When repeated transfusions are needed, a new specimen should be drawn from the patient (recipient) for crossmatching purposes if blood was last given more than 48 hours earlier. Some patients demonstrate marked an amnestic responses to RBC antigens that they lack and may produce clinically significant quantities of antibody in a few hours. This antibody is not present in the original specimen from the patient.

  • Major Crossmatch

    From approximately 1960 to 1984, the purpose of the major crossmatch was to detect unexpected antibodies in the serum of the recipient, and it also acted as a check on a previous antibody screen. It also served as a check on ABO typing, since a mistake in ABO typing may result in RBCs from the donor being incompatible with naturally occurring ABO antibodies in the serum of the recipient.

    The most widely used crossmatch technique from approximately 1960 to 1984 consisted of three sequential parts. First, donor RBC and saline-diluted recipient serum were incubated at 25°C to detect bivalent antibodies. Next was incubation at 37°C with 22% albumin (or other albumin concentration), added to enhance agglutination of univalent antibodies such as the Rh group. Finally, a Coombs’ test that detected weaker univalent antibodies was performed. Many laboratories are substituting low ionic strength saline (LISS) reagent for albumin, which decreases incubation time from 30 minutes to 15 minutes. It has the same or slightly better sensitivity as the albumin reagent, with a minor disadvantage that LISS detects more antibodies that turn out to be clinicallyunimportant than does albumin.

    Unfortunately, the major crossmatch will not detect the most common ABO incompatibility error, which is patient identification mistakes, either crossmatch specimens obtained from the wrong patient or properly crossmatched blood transfused into the wrong patient. Also, the procedure will not detect errors in Rh typing if no Rh antibodies are present in donor or recipient blood. Since Rh antibodies do not occur naturally, either donor or recipient would have to be previously sensitized before Rh or similar antibodies would appear. Since the crossmatch is designed to detect antibodies, not antigens, the crossmatch does not demonstrate antigens and thus will not prevent immunization (sensitization) of the recipient by Rh or other non-ABO blood groups. This can be done only by proper typing of the donor and recipient cells beforehand.

    The AABB, in their 1984 Blood Bank Standards, made a major departure from the “traditional” crossmatch by relying on precrossmatch ABO and Rh typing and the antibody screen of the recipient serum plus only the “immediate spin” saline part of the major crossmatch procedure to check for technical errors in typing. The immediate spin method entails centrifugation and examination of donor blood RBC and recipient serum in saline diluent without incubation, which takes only about 5 minutes. To substitute the immediate spin crossmatch for the traditional major crossmatch it is necessary to do a pretransfusion recipient antibody screen without finding any unexpected antibodies (if the screen did detect antibodies, the traditional crossmatch would have to be done). This protocol removes considerable pressure from the blood bank since ABO typing and recipient serum antibody screen are usually performed well in advance of the actual order to transfuse. It also saves a significant amount of time and money. Studies indicate that about 0.06% (range, 0.01%-0.4%) of recipient units will contain antibody that will be missed (although the majority of these would not be likely to produce life-threatening hemolysis). Other studies found that crossmatch alone detected about 4% (range, 1%-11%) of total antibodies detected by combined antibody screen and crossmatch. There was considerable debate about how far to go in altering the traditional crossmatch procedure and whether or not to perform it in all cases. Nevertheless, the new policy has been in effect for nearly 10 years without major problems, so there is now much less controversy.

    Occasionally the laboratory is asked to do an emergency crossmatch. Those institutions using the current AABB protocol already have a single 5-minute procedure that it would not be possible to shorten. Those institutions using the traditional crossmatch should reach an agreement with the medical staff as to what would be done if an emergency crossmatch is requested. One or more steps in the crossmatch procedure can be eliminated to gain speed, but each omission produces a small but definite risk of missing an unexpected serious problem. There also should be agreement about what would be done in emergency conditions if the patient has antibodies against blood groups other than ABO and Rh or has an unidentified antibody.

    Gamma globulin concentrate for intravenous therapy (IV-GG) often contains one or more antibodies (about 50% of IV-GG lots tested in one study). In one report, about 65% of the antibodies detected were anti-D. Patients receiving IV gamma globulin therapy may receive detectable amounts of the antibody or antibodies in the gamma globulin, which might cause problems if the blood bank is not informed that IV-GG is or was being given.

  • Antibody Screen

    Even if major blood group typing has been done, transfusion reactions can occur due to unexpected antibodies in the serum of the recipient or due to incorrect RBC typing. To prevent reactions, the concept of a crossmatch evolved. The basic procedure is the “major” crossmatch, in which RBCs of the donor are placed into the serum of the recipient under standard conditions. Previously, the “minor” crossmatch (RBCs of the recipient matched with the serum of the donor) was also a standard procedure, but most blood banks no longer perform this test, and in addition, the AABB recommends that the test not be done. Even if the donor serum contains antibodies to one or more of the RBC antigens of the recipient, the relatively large blood volume of the recipient in most cases should dilute the relatively small volume of donor serum to a point at which it is harmless. The minor crossmatch has been replaced by an antibody screen on recipient serum. The antibody screen consists of the recipient serum and a commercially supplied panel of different blood group O RBCs that contain various other blood group antigens. If there is free antibody inthe patient serum (other than ABO antibody), it either will coat or will agglutinate the test RBC so that the presence of the antibody will be detected. Detection is carried out by incubation at 37°C (usually with some enhancing agent, e.g., albumin or low ionic strength medium), followed by a direct Coombs’ test. The antibody screen does not detect errors in ABO typing, since the reagent RBCs are all group O and thus nonreactive with anti-A or anti-B antibody. Once an unexpected antibody is discovered, the type of antibody has to be identified. This can usually be deduced by the pattern of reaction with the various RBCs of the test panel. In a few cases the specimen must be sent to a referencelaboratory for special tests.

    In summary, antibody screening is designed to demonstrate unexpected antibodies in the serum of the recipient that may destroy donor RBCs that were thought to be compatible on the basis of ABO and Rh blood group typing.

  • Other Blood Group Antigen Systems

    Besides the ABO and the Rh system there are a number of other unrelated blood group antigen systems that have some importance, either for medicolegal parenthood studies or because sensitization to these antigens causes transfusion reactions or hemolytic disease of the newborn. The most important of these systemsis Kell (K), a well-recognized blood bank problem. The Kell antibodies are similar in characteristics and behavior to the Rh system D (Rho) antibody. Fortunately, only about 10% of whites and 2% of African Americanshave the Kell antigen and are thus Kell positive, so that opportunities for sensitization of a Kell-negative person are not great. Kell antibody is univalent and, like Rh, acts best in vitro in a high-protein medium at 37°C. If reactions to Kell antibodies occur, results of the direct Coombs’ test are positive (until the affected RBCs are destroyed). A similar situation exists for the rare Duffy (Fy) and Kidd (jK) systems. There are other systems that resemble the ABO system in their antibody characteristics, and these include the MN, P, Lewis (Le), and Lutheran (Lu) systems. They are primarily bivalent antibodies and react best in vitro in a saline medium at room temperature or below. They are rare causes of transfusion reactions, and when difficulties arise, they are clinically milder than the problems associated with the univalent antibody systems.

    There are a large number of so-called minor blood group antigens. The most important is the I-i, or IH, system. This system is very weak in newborn RBCs and becomes established at approximately age 2 years. Anti-I (or anti-IH) antibodies are IgM cold agglutinins. Anti-I antibody is a frequent cause of viral or mycoplasma-associated cold agglutinins or idiopathic cold autoimmune hemolytic anemia. Some of the other systems include “public,” or “universal,” antigens, such as the Vel system. Nearly all persons have the antigen on their RBCs, so the chance is extremely small that any individual would not have the antigen and thereby would be capable of producing antibody to the antigen. Another group includes the “private,” or “low-incidence,” antigens, the most common of which is the Wright (Wra) antigen-antibody system. Very few persons have the low-incidence antibody on their RBCs, so risk of encountering the antigen is extremely small. Yet another antigen group is the “high titer–low avidity” type of antibody, which is occasionally responsible for weakly positive Coombs’ test results inrecipient or donor antibody screens. In the vast majority of cases other than high-titer anti-I antibodies, the minor group antigens or antibodies are not clinically significant. However, they are a great source of frustration to the blood bank when they occur, because any unexpected antibody must be identified and because they interfere with the crossmatch and cause donor bloods to appear incompatible with the recipient.

    In summary, RBC typing is ordinarily designed to show what ABO and Rh antigens are on the RBC and thus what blood group RBCs can be given to a recipient, either without being destroyed by antibodies the recipient is known to possess or without danger of sensitizing the recipient by introducing antigens that he or she might lack and against which he or she might produce antibodies.

  • Rh Blood Group System

    The next major blood group is the Rh system. There is considerable controversy over nomenclature of the Rh genetic apparatus between advocates of the English Fisher-Race CDE-cde nomenclature and the American Wiener’s Rh-hr labeling (Fig. 9-1)

    Comparison of the Fisher-Race and Wiener nomenclatures

    Comparison of the Fisher-Race and Wiener nomenclatures

    Fig. 9-1 Comparison of the Fisher-Race and Wiener nomenclatures. (From Hyland reference manual of immunohematology, ed 2. Los Angeles, Hyland Division of Travenol Laboratories, 1964, pp 38-39. Reproduced by permission.)

    According to Wiener, the Rh group is determined by single genes, each chromosome of a pair containing one of these genes. In most situations each gene is expected to determine one antigen, to which there may develop one specific antibody. In Wiener’s Rh system theory, each gene does, indeed, control one antigen. However, each of these antigens (agglutinogens) gives rise to several different blood factors, and it is antibodies to these factors that are the serologic components of the Rh system. There are eight such agglutinogens, each letter of which can be either big or little, on each of the two chromosomes.

    In the Fisher-Race system a single gene controls one single antigen, which controls one single antibody. This means that three genes would be present on each chromosome of a chromosome pair and that the three-gene group is inherited as a unit. New antibodies are assumed to be due to mutation or defective separation of the genes that make up the gene group during meiosis (“crossover” rearrangement).

    At present, most experts believe that Wiener’s theory fits the actual situation better than the Fisher-Race theory. The main drawback to Wiener’stheory is its cumbersome terminology. Actually, in the great majority of situations, the much simpler Fisher-Race terminology is adequate, because the antibodies that it names by its special letters are the same as the basic blood factors of Wiener’s system. It is only in unusual or rare situations that the Wiener system becomes indispensable. The Fisher-Race terminology has persisted because, for most practical work, one can use it while ignoring the underlying theory of gene inheritance. In the literature one often finds both the Fisher-Race and the Wiener nomenclatures, one of them being given in parentheses.

    Of the Rh antigens, D (Rho) is by far the most antigenic, and when it is present on at least one chromosome, the patient types as Rh positive. Therefore, antigen D behaves serologically like a dominant gene and persons who type as D positive can be either homozygous or heterozygous, whereas absence of D reactivity behaves serologically like a recessive gene (both chromosomes lack the D antigen). Only 20% of the population lack D (Rho) completely and are therefore considered Rh negative. Of the other antigens, c (hr’) is the next strongest, although much less important than D.

    Rh antigens lack corresponding naturally occurring antibodies in the serum. Therefore, when anti-Rh antibodies appear, they are of the immune type and are the result of exposure of an Rh-negative person to Rh antigen on RBCs of another person. This may happen from transfusion or in pregnancy. It is now well documented that RBCs from the fetus escape the placenta into the bloodstream of themother. In this way the mother can develop Rh antibodies against the Rh antigenof the fetus. One exception to this occurs when the mother’s serum contains antibodies against the ABO group of the fetus, for example, if the mother is group O and the fetus group B. In these cases the fetal RBCs are apparently destroyed in the maternal circulation before Rh sensitization can proceed to a significant extent, although this does not always happen. The syndrome of Rh-induced erythroblastosis will be discussed later. Rh incompatibility was a major cause of blood transfusion reactions, although these reactions occur much less often than ABO transfusion reactions. Rh antibody transfusion reactions may occur by transfusion of donor blood containing Rh antibodies or by previous sensitization of a recipient, who now will have the antibodies in his or her own serum. Rh sensitization after transfusion appears to have some correlation to the amount of Rh-incompatible blood received, although this varies considerably between individuals (some of whom can be sensitized from only a few milliliters of Rh-positive cells, while others have received one unit (250 ml) or even more without developing anti-D. The sensitization rate after transfusion with incompatible Rh blood varies from 8%-70% in the literature. Interestingly, infants under age 4 months usually (but not always) do not form new alloantibodies against any incompatible red cell antigens.

    Rh antigen may be typed using commercial antiserum. Preliminary screening isonly for antigen D, which establishes a person as Rh positive or negative. If aperson is Rh negative, further studies with antiserum to other components of the Rh group may be done, depending on the situation and the individual blood bank. In particular, there is a weak subgroup of D (Rho) formerly called Du (Rho variant) and now called “weak D” by the American Association of Blood Banks (AABB), which is analogous to the weak A2 subgroup of A in the ABO system. Weak D (Du) blood may fail to give a positive reaction with some commercial Rh anti-D typing serums and so may falsely type as Rh negative. Therefore, many large blood banks screen Rh-negative RBCs for weak D as well as for c (hr’) and E (rh’), the most antigenic of the minor Rh antigens. In blood banking, weak D recipients are considered Rh negative, since they may produce antibodiesto a subunit of the Rho (D) antigen of a Rh positive donor, if the weak D is the type of weak D that lacks the subunit. Weak D donors are considered Rh positive, since their cells may be destroyed by a recipient serum that contains anti-D (anti-Rho).

    Rh antibodies are usually univalent and react best in vitro at 37°C in a high-protein medium. Large blood banks screen donor serum for these antibodies using a variety of techniques. When Rh antibodies attack RBCs in vivo, whether in transfusion or in hemolytic disease of the newborn, they coat the surface of the RBCs in the usual manner of univalent antibodies and then get a positive result with the direct Coombs’ test (until the affected RBCs are destroyed). This RhoGam (which is anti-Rho) is given to the mother to prevent anti-Rho(D) antibody formation in an Rh-positive mother of an Rh-negative fetus, and will prevent hemolytic disease of the newborn but occasionally may interfere with Rh typing of the newborn.

  • ABO Blood Group System

    The ABO blood group system is a classic example of agglutinogens and their corresponding isoantibodies. There are three of these antigens—A, B, and O—whose genes are placed in one locus on each of two paired chromosomes. These genes are alleles, meaning that they are interchangeable at their chromosome location. Therefore, each of the paired chromosomes carries any one of the three antigen genes. A and B are relatively strong antigens and serologically behave like dominant genes, whereas O is not detected by commercial typing sera and therefore the O antigen behaves serologically like a recessive gene. Blood group O is diagnosed by absence of reaction for either A or B antigen, so that O blood type implies O antigen on both chromosomes rather than only one. This makes four major phenotype groups possible—A, B, AB, and O—since A and B are dominant over O. Furthermore, when either A or B antigen is present on anindividual’s RBCs, the corresponding isoantibodies anti-A or anti-B will be absent from his or her serum; conversely, if an individual lacks either A or B antigen, his or her serum will contain the isoantibody to the missing isoantigen. O is so weak an antigen that for practical purposes it is considered nonantigenic. Therefore, a person who is AA or AO will have anti-B isoantibodies in his or her serum, a person who is OO will have both anti-A and anti-B isoantibodies, and so on. Why the body is stimulated to produce antibodies to the missing A or B antigens is not completely understood, but apparently antigens similar to ABO substances exist elsewhere in nature and somehow cause a natural sensitization. Anti-A and Anti-B are not detectable at birth, are weakly detectable at age 3-6 months, and gain maximum strength at age 5 years. Anti-A or anti-B incord blood or neonatal serum is usually of maternal origin.

    Anti-A and anti-B are bivalent antibodies that react in saline at room temperature. Ordinarily, little difficulty is encountered in ABO typing. However, newborn A and B antigens may be weak and may not reach full strength until age 2-4 years, presenting the potential for false negative reaction in the newborn. There is a more common potentially serious situation that arises from the fact that subgroups of agglutinogen A exist. These are called A1, A2, and A3. The most common and strongest of these is A1, which comprises about 80% of group A and AB red cells, with A2 cells comprising most of the remaining 20% of red cells. A2 is troublesome because it is sometimes so weak that some commercial anti-A serums failto detect it. This may cause A2B to be falsely typed as B or A2O to be falsely typed as O. This situation is more likely to occur with polyclonal antibody typing sera and is not frequent with present-day potent blended monoclonal antibody typing sera (2 of 7124 patients in one study). Group A subgroups weaker than A2 exist but are rare. They are easily missed, even with potent anti-A typing serums. The main importance of the A2 subgroup is that persons with A2 sometimes have antibodies to A1 (the most common subgroup of A).

    Anti-A1 is said to occur on 1%-8% of A2 persons and 22%-35% of A2B persons. These antibodies are usually not clinically important but may occasionally produce blood bank typing problems.

    Group O serum contains an antibody (anti-A1B) that reacts against group A and group B cells more strongly than separate antibodies against group Aor group B cells. Serum from group O persons with high titer of anti-A1B was used to make an antiserum that can detect weak subgroups of A or B. Blended monoclonal A-B antibody is now available that produces even stronger reactions than the naturally occurring antibody.

  • Autoantibodies

    Autoantibodies present an interesting problem, both in their clinical manifestations and in the difficulty of laboratory detection and identification. They may be either the warm or cold type and may be complete or incomplete.

    Warm autoantibodies react at body temperature and are most often of the incomplete type. They comprise about 70% of autoantibodies. They may be idiopathic or secondary to certain diseases. The main disease categories responsible are leukemias and lymphomas (particularly chronic lymphocytic leukemia and Hodgkin’s disease); collagen diseases (especially disseminated lupus); and, uncommonly, a variety of other diseases, including cirrhosis and extensive carcinoma. Results of the direct Coombs’ test are usually but notalways positive, both in the “idiopathic acquired” and the secondary autoimmune hemolytic anemias. In one report, 2%-4% of patients with autoimmune hemolytic anemia had a negative direct Coombs’ test result. If the Coombs’ test result is negative, demonstration of warm autoantibodies is very difficult, often impossible.

    Cold autoantibodies react at 4°C-20°C and are found so frequently in normal persons that titers up to 1:32 are considered normal. Theyare hemagglutinating and are believed to be due to infection by organisms having antigenic groups similar to some of those on the RBCs. These antibodies behave mostly as bivalent types and require complement for reaction. In normally lowtiter they need refrigerator temperatures to attack RBCs. In response to a considerable number of diseases, these cold agglutinins are found in high titer, sometimes very high, and may then attack RBCs at temperatures approaching body levels, causing hemolytic anemia. High-titer cold agglutinins may befound in nonbacterial infections, especially mycoplasma pneumonia (primary atypical pneumonia), influenza, and infectious mononucleosis; in collagen diseases, including rheumatoid arthritis; in malignant lymphomas; and occasionally in cirrhosis. Fortunately, even when cold autoantibodies are present in high titer there usually is no trouble, and generally only very high titers are associated with in vivo erythrocyte agglutination or hemolytic anemia. This is not always true, however. The direct Coombs’ test result is usually negative. When cold agglutinin studies are ordered, an indirect Coombs’ test is generally done, with the first stage being incubation of RBCs and the patient’s serum at 4°C-10°C.

  • Antibody Detection Methods

    There are two methods of detecting and characterizing antibodies: (1) the direct Coombs’ test and (2) a group of procedures that try to determine if an antibody is present, and if present, attempt to identify the antibody by showing what the antibody will do in various controlled conditions.

    Direct Coombs’ test

    To prepare reagents for the Coombs’ test, human globulin, either gamma (IgG), nongamma (IgM), or a mixture of the two, is injected into rabbits. The rabbit produces antibodies against the injected human globulin. Rabbit serum containing these antihuman globulin antibodies is known as Coombs’ serum. Since human antibodies are globulin, usually gamma globulin, addition of Coombs’ serum (rabbit antibody against human gamma globulin) to anything containinghuman antibodies will result in the combination of the Coombs’ rabbit antibody with human antibody. There also has to be some indicator system that reveals that the reaction of the two antibodies has taken place. This can be seen visually if the Coombs’ rabbit antibody has been tagged with a fluorescent dye; or if the reaction takes place on the surface of RBCs, lysis or agglutination of the RBC can be produced.

    The direct Coombs’ test demonstrates that in vivo coating of RBCs by antibody has occurred. It does not identify the antibody responsible. It is a one-stage procedure. The Coombs’ serum reagent is simply added to a preparation of RBCs after the RBCs are washed to remove nonspecific serum proteins. If the RBCs are coated with antibody, the Coombs’ reagent will attack this antibody on the RBC and will cause the RBCs to agglutinate to one another, forming clumps. The antibody on the RBC is most often univalent but sometimes is polyvalent. Although antibodies on RBCs that are detected by the direct Coombs’ test are most often antibodies to RBC blood group antigens, certain medications (e.g., methyldopa and levodopa) in some patients may cause autoantibodies to beproduced against certain RBC antigens. Also, in some cases antibodies not directed against RBC antigens can attach to RBCs, such as antibodies developed in some patients against certain medications such as penicillin or autoantibodies formed in the rheumatoid-collagen diseases or in some patients with extensive cancer. In addition, some reports indicate an increased incidence of apparently nonspecific positive direct Coombs’ reactions in patients with elevated serum gamma globulin levels.

    The reagent for the direct Coombs’ test can be either polyspecific or monospecific. The polyspecific type detects not only gamma globulin but also the C3d subgroup of complement. Complement may be adsorbed onto RBCs in association with immune complexes generated in some patients with certain conditions, such as the rheumatoid-collagen diseases and certain medications, such as quinidine and phenacetin. Monospecific Coombs’ reagents are specific either for IgG immunoglobulin (and therefore, for antibody) or for complement C3d. If the polyspecific reagent produces a positive result, use of the monospecific reagents (plus elution techniques discussed later) can narrow down the possible etiologies.

    The direct Coombs’ test may be done by either a test tube or a slide method. The direct Coombs’ test must be done on clotted blood and the indirect Coombs’ test on serum, since laboratory anticoagulants may interfere. A false positive direct Coombs’ test result may be given by increased peripheral blood reticulocytes using the test tube method, although the slide technique will remain negative. Therefore, one should know which method the laboratoryuses for the direct Coombs’ test.

    In summary, positive direct Coombs’ test results can be due to blood group incompatibility, may be drug induced, may be seen after cardiac valve operations, and may appear in rheumatoid-collagen diseases, malignancy, idiopathic autoimmune hemolytic anemia, and other conditions. The overall incidence of a positive direct Coombs’ test result in hospitalized patients is reported to be about 7%-8% (range, 1%-15%).

    The main indications for the direct Coombs’ test include the following (most are discussed later in detail):

    1. The diagnosis of hemolytic disease of the newborn.
    2. The diagnosis of hemolytic anemia in adults. These diseases include manyof the acquired autoimmune hemolytic anemias of both idiopathic and secondary varieties. Results of the direct Coombs’ test at normal temperatures are usually negative with cold agglutinins.
    3. Investigation of hemolytic transfusion reactions.

    In these clinical situations the indirect Coombs’ test should not be done if the direct test result is negative, since one is interested only in those antibodies that are coating the RBCs (and thus precipitating clinical disease).

    Antibody detection and identification

    Indirect Coombs’ test. The indirect Coombs’ test is a two-stage procedure. The first stage takes place in vitro and may be done in either of two ways:

    1. RBCs of known antigenic makeup are exposed to serum containing unknown antibodies. If the antibody combines with the RBCs, as detected by the second stage, this proves that circulating antibody to one or more antigens on the RBC is present. Since the RBC antigens are known, this may help to identify that antibody more specifically.
    2. Serum containing known specific antibody is exposed to RBCs of unknown antigenic makeup. If the antibody combines with the RBCs, as detected by the second stage, this identifies the antigen on the RBCs.

    The second stage consists in adding Coombs’ serum to the RBCs after the RBCs have been washed to remove nonspecific unattached antibody or proteins. Ifspecific antibody has coated the RBCs, Coombs’ serum will attack this antibody and cause the cells to agglutinate. The second stage is thus essentially adirect Coombs’ test done on the products of the first stage.

    Therefore, the indirect Coombs’ test can be used either to detect free antibody in a patient’s serum or to identify certain RBC antigens, depending on how the test is done.

    The major indications for the indirect Coombs’ test are the following:

    1. Detection of certain weak antigens in RBCs, such as Du or certain RBC antigens whose antibodies are of the incomplete type, such as Duffy or Kidd (see antibody screen).
    2. Detection of incomplete antibodies in serum, either for pretransfusion screening or for purposes of titration.
    3. Demonstration of cold agglutinin autoantibodies.

    The indirect Coombs’ test is almost never needed routinely. In most situations, such as cold agglutinins or antibody identification, simply ordering atest for these substances will automatically cause an indirect Coombs’ test to be done. The indirect Coombs’ test should be thought of as a laboratory technique rather than as an actual laboratory test.

    False positives and false negatives may occur with either the direct or indirect Coombs’ technique due to mixup of patient specimens, clerical error when recording results, technical error (too much or not enough RBC washing; also failure to add reagents or adding the wrong reagent), contamination by 5% or 10% glucose in water (but not glucose in saline) from intravenous tubing, and, rarely, use of faulty commercial Coombs’ reagent.

    Antibody elution. When a direct Coombs’ test yields positive results, especially when thecause is thought to be a blood group–specific antibody, it is desirable to attempt elution (removal or detachment) of the antibody from the RBC to determine the antigen against which it is reacting. This is usually done by changing the physical conditions surrounding the antibody to neutralize the attachment forces. The most common current methods are heat, freeze-thaw, and chemical. Once the antibody is isolated from the RBCs, it can be tested with a panel of RBCs containing known antigens to establish its identity.