Autoimmune Hemolytic Anemia (AIHA) is the oldest recognized autoimmune deficiency in humans (Meyer et al. 1998). It is one of many types of anemias While red blood cells (RBCs) usually circulate for 120 days before the spleen removes them from circulation, in AIHA erythrocytes are prematurely destroyed by RBC autoantibodies. When the bone marrow is unable to compensate for this hemolysis of RBCs through hemopoiesis (production of new RBCs) a person is said to have hemolytic anemia. AIHA may be acute or chronic and is sometimes fatal. Women are twice as likely to have AIHA than are men (Kennedy 2000). However, in children the disease appears more often in males and primarily affects children under 5 years of age as an acute hemolysis (Gibson 1998). While AIHA affects only .001% of the general human population (Hashimoto 1998), it is the most commonly occurring autoimmune disease in canines (Day 1999).
The mechanism for hemolysis depends upon the autoantibody idiotype. Additionally, not all autoantibodies cause hemolysis. Affinity of the autoantibody for a species specific antigen and the autoantibodies' abilities to cause hemoagglutination influence severity of the disease (Shibata et al 1991). AIHA is usually caused by IgG1, IgG3, and IgM autoantibodies. Occasionally the autoantibodies may be IgA. Hemolysis results from either activation of the classical complement pathway by IgM, IgG1, IgG3, and IgA or from phagocytosis or antibody-dependent cell cytotoxicity (ADCC) by Ig1 and Ig3 (Gibson 1998). Extravascular phagocytosis and ADCC mediated hemolysis result from recognition of the antibody by Fc receptors on macrophage, resulting in erythrophagocytosis, or on K cells in the spleen, resulting in the release of lysosomal enzymes. Studies conducted by Meyer et al suggest that IgG1 promotes erythrophagocytosis via the FcyRIII receptor and that IgG2 may also participate in erythrophagocytosis, but primarily through FcyRI (1998). In compliment mediated hemolysis, binding of the antibody initiates the compliment cascade. The compliment cascade may terminate at c3b formation, when macrophage, especially hepatic Kupffer's cells, engulf the antibody coated erythrocyte. Continuation of the complement cascade leads to formation of a membrane attack complex and intravascular hemolysis. However the results of some studies suggest that complement may not play as significant a role in hemolytic anemia as is currently thought (Meyer et al 1998).
Partial extravascular hemolysis creates sperocytes, which are spherical, rigid erythrocytes that have lost part of their cell membrane. The cells are fragile and therefore easily damaged and destroyed (Domen 1998). IgG hemolysis occurs largely in the spleen while hemolysis by IgM occurs in the liver (Merck 2000). Additionally, because splenic macrophage possess both complement and FcR receptors while hepatic macrophage display only complement, most extravascular hemolysis occurs in the spleen (Gibson 1998). Intravascular hemolysis leads to fragmentation of erythrocytes into helmet shaped schizocytes. Peripheral blood smears can be examined for sperocytes and schizocytes to provide information about the mechanism of a particular patient's AIHA (see figures 2 and 3).
Normal Red Blood Cells
Peripheral blood smear. Peripheral blood smear. Schizocytes result from fragmentation of
Note the areas of central pallor. Sperocytes formed from partial erythrocytes in intravascular hemolysis.
Note the lack of areas of central
Figure 1 Figure 2 Figure 3
Image taken from Dr Ed Uthman's Image taken from KU Pathology 851 Image taken from Dr. Ed Uthman's
web page on hemolytic anemia image web page with permission of web page on hemolytic animas with
with permission of author (Uthman 2000) author (Woodroof 2000). permission of author (Uthman 2000).
Types of AIHA
AIHA is a heterogeneous disease and includes Warm AIHA (WAIHA), Cold AIHA (CAD), Paroxysmal Cold Hemoglobinuria (PCH), and Drug Induced Hemolytic Anemias (DIHA); all of which are characterized by production of autoantibodies to RBCs . DIHA are further subdived into three classes, based upon the binding site of the autoantibody. WAIHA and CAD may be either idiopathic or may exist secondary to another autoimmune disease. Some patients have mixed AIHA, which manifests both CAD and WAIHA autoantibodies. Table 1 summarizes some of the characteristics of each AIHA discussed below.
WAIHA is the most commonly occurring form of AIHA. Autoantibodies produced in WAIHA are non-specific and bind to all RBCs at 37'C, with the exception of those which lack Rh antigen. The autoantibodies are of polyclonal origin and usually bind to the Rh antigen (Hashimoto 1998). Approximately 60% of WAIHA cases are idiopathic (Smith 1999). Secondary WAIHA often occurs with chronic lymphocytic leukemia and may also occur with diseases such as systematic lupus, solid tumors, myloproliferative diseases, and hepatitis A. Hemolysis is usually extravascular and occurs via partial phagocytosis or by ADCC. The autoantibodies causing hemolysis are most frequently IgG1 and IgG3. WAIHA autoantibodies are usually of polyclonal origin. However, in cases where IgM and IgA predominate, the presence of autoantibody only occasionally causes WAIHA (Gibson 1988).
In CAD, the autoantibody is usually monoclonal IgM, but occasionally IgG or IgA, with kappa light chains. The autoantibody binds best at temperatures below 4'C. However, the thermal amplitude (temperature range within which the antibody binds) varies, and higher thermal amplitudes tend to indicate a more severe form of the disease (Hashimoto 1998). A polyclonal IgM has been reported as well, but it is rarely pathological and reacts only at very low temperatures (Domen 1998). The cold agglutinin antibody is specific for either I antigen or i antigen. Idiopathic CAD usually occurs in adults, especially the elderly, as a chronic, mild anemia. Because of the antibody's temperature sensitivity, the condition worsens in winter and binds when RBCs enter peripheral circulation (Smith 1999). This form of CAD can cause both intravascular and extravascular hemolysis. Secondary CAD frequently appears in children who have recently had viral or bacterial infections such as Mycoplasma pneumoniae and infectious mononucleosis. Unlike idiopathic CAD, this condition occurs suddenly and is acute. However, the condition also tends to be transient (Hashimoto 1998). CAD also exist in a chronic form when the disease occurs secondary to B-cell lymphomas or chronic lymphocytic leukemia (Zilow et al 1994).
PCH is caused by an IgG biphasic autoantibody which binds to RBCs at temperatures below 4'C. The autoantibody is usually specific for the globoside glycosphingolipid P antigen (Rosenfield and Diamond 1981). The first two components of the complement system bind at 4'C , and the cascade is completed at 37'C. In the early 1900's PCH occurred mostly among syphilis victims as an acute disease. Most cases today occur in children after infection with measles, mumps, chickenpox or influenza (Smith 1999). This more recent condition causes severe and rapid intravascular hemolysis that may be life threatening for 10-14 days after onset (Rosenfield and Diamond 1981). However, PCH is usually a self limiting form of AIHA.
In mixed AIHA both warm agglutinate IgG and cold agglutinate IgM autoantibodies are present. The autoantibodies may or may not have specificity for I or i antigen. Both intravascular and extravascular hemolysis are observed. Approximately 50% of mixed AIHA are idiopathic, while secondary mixed AIHA commonly occurs in collagen autoimmune diseases like lupus . This form of AIHA appears as a sudden, acute disease but often becomes a chronic condition (Smith 1999).
Drug Induced AIHAs
While autoimmune hemolytic anemia is a rare disease, the incidence of DIHA is increasing significantly. Over 70 different drugs have induced either a positive Coombs' test or immune hemolysis (Wright 1999). Drugs have been observed to induce four types of autoantibody binding to erythrocytes. However, only three of these types of binding are known to cause hemolytic anemia. The characteristics of DIAHAs resemble those of WAIHA. In the hapten mechanism, the drug binds to an RBC which acts as a carrier for the drug hapten. In the immune-complex mechanism the drug first binds to the antibody and the drug-antibody complex then binds to the RBC. In the autoimmune mechanism, the autoantibody binds directly to the RBCs. (Jefferies 1994). Drugs such as cephalosporins have been shown to modify the RBC membrane. Serum proteins such as immunoglobulins and complement proteins then bind non-specifically to the RBCs. However, the weak binding of RBCs in membrane modification has not yet been demonstrated to cause hemolysis (Mueller-Eckhart and Salama 1990). Table 2 lists drugs and the specific mechanism by which they induce hemolysis. Nonetheless, not all drugs are specific to one particular mechanism. Certain drugs have been found to elicit mixed responses; for instance, administration of the drug, nonifensine, can induce both immune complex and autoantibody mechanisms of hemolytic anemia (Petz 1993).
|AIHA||% of Cases||Pathogenesis||Predominating Blood Group||Antibody Type||DAT Results||Antibody in Eluate||Treatment Options|
|WAIHA||80%||AuAb bind RBC at 37 'C||Rh||IgG||IgG, IgG+C, C(rare)||IgG||variable: cortecosteroids. immunosuppression, danazol, IV gamma globulin|
|CAD||20-25%||AuAb bind to RBC at 4' C||I, i||IgM, IgG||C3d||IgM||frequently not needed|
|AIHA Mixed||7-8%||broad amplitude of reactivity to 37 'C||Possibly I, i||IgM,IgG||IgG, C3d||IgG||cortecosteroids|
|PCH||1%||AuAb binds RBC at 4 'C; fixes complement;complement cascade completed at 37 'C||P||IgG||C||nonreactive||self-limiting,; possibly transfusion|
|DIHA||12-18%||AuAb binds drug, or binds drug then RBC, or binds drug-rbc||---||---||---||---||discontinue drug, occasionally transfusion|
|Hapten Mechanism||Penicillin, Cephalothin, Ampicillin, Carbenicillin, Methicillin, Cephaloridine|
|Immune Complex Mechanism||Quinine, Quinidine, Rifampin, Antihistamines, Sulfonamides, Tetracyclin, Insulin, Streptomycin, Acetaminophen, Cephaosporin, Dipyrone, Isoniazid, Tolmetin|
|Autoantibody Mechanism||a-Methyldopa, L-Dopa, Ibuprofen, Procainamide, Thioridazine|
Many symptoms of AIHA resemble those of other anemias and include nosebleeds, bleeding gums, chills, fatigue, paleness, shortness of breath, and jaundice (Kennedy 2000). Symptoms of AIHA may also include an enlarged spleen, due to excessive RBC destruction, and dark urine, due to an excess of unprocessed catabolites resulting from RBC hemolysis. Patients with CAD may experience numbness and pain in cooler temperatures as a result of cyanosis (Domen 1998). Because the bone marrow attempts to compensate for the loss of RBCs through elevated hemapoiesis laboratory test results such as high reticulocyte (developing RBCs) counts are suggestive of a hemolytic anemia. Click here to view a map of diagnostic procedures to identify various types of anemias. After hemolytic anemia has been diagnosed clinical history, Coombs' test (direct antiglobulin test or DAT), and blood smear morphology aid in determination of its origin (Uthman 2000). DAT is the most important assay for distinguishing AIHA from other types of hemolytic anemias (Jefferies 1994). A DAT to determine the presence of either c3 or IgG bound to erythrocytes is performed and a positive test results in erythrocyte agglutination (Figure 4). After the initial positive DAT, additional DATs are conducted to determine whether c3, IgG or both proteins are binding to the RBC. In CAD, only c3 binding will usually be detected when the test is conducted at temperatures around 37'C, but in WAIHA the DAT may be positive for IgG alone or both IgG and c3. If IgG is detected, the autoantibody may be eluted and tested for antigen specificity, especially when cross-matching for a transfusion.
RBC agglutination caused by cold agglutination autoantibody.
Image taken from KU Pathology 851 image web page with permission of author (Woodroof 2000).
AIHA is a drug induced condition and tests against drug-treated RBC can confirm the mechanism of the drug induced reaction (Wright and Smith 1999). The Donath-Landsteiner test is preformed to detect PCH. In this test, the IgG autoantibody is incubated with normal RBC and serum at 4'C and then warmed to 37'C to cause hemolysis (Jefferies 1994).
New Zealand Black mice (NZB) provide the current animal model to study AIHA, while methyl-dopa drug induced AIHA has provided researchers with a human model of both AIHA as well as autoimmune diseases as a whole. Nevertheless, the etiology of AIHA is still not understood.
Much research supports an antigen induction model of AIHA. In mice, band 3, an erythrocyte anion exchange protein, appears to be the predominate antigen for RBC autoantibodies. However, not all autoantibodies binding band 3 produce pathological effects. The protein appears to serve a natural role in the elimination of aged RBCs; in aged RBCs, band 3 aggregates and antibody binds at higher density to facilitate clearance of the cells (Diilulio et al 1997). However, NZB mice with band 3 reactive CD4 T cells do produce pathogenic autoantibodies (Perry et al 1996). In a study by Barker et al on humans, many AIHA patients were also found to express T helper cells which bind to the Rh antigen on human RBCs. B cells, however, were found not to react with same epitopes on the Rh antigen which is recognized by the T helper cells (1997). As a result of these studies, researchers have proposed that induced changes in MHCII autoantigen processing results in the presentation of previously cryptic epitopes to which naive Rh reactive T cells respond (Barker et al 1997, Shen et al 1996). This theory is supported by a study by Diiulio et al on NZB mice which found yet another autoantibody for murine RBCs which binds to a partially masked epitope when the RBCs are treated with protease to enhance expression of the epitope (1997). The TH-1 predominated response to band 3 elicits IFN-Y production, and this cytokine may promote presentation of the cryptic epitopes (Shen et al 1996). However, how a self-reactive T cells might escape clonal deletion to respond to the self-antigen is not yet understood.
Evidence from other studies supports a polyclonal activation model instead of an antigen-induced model of AIHA. For instance, Hernandez et al found RBC autoantibodies in both healthy and AIHA individuals but that autoantibody levels were much higher in AIHA individuals. Higher levels of RBC autoantibodies could be induced by polyclonal activation (1990). Yet other research suggests B-1 involvement in AIHA. Unlike conventional self-reactive B cells in the periphery and the bone marrow, self-reactive B-1 cells in the peritoneal cavity are separated from RBCs and may therefore escape clonal deletion. Oral administration of lipopolysaccharides (LPS) to HL mice with H and L chains derived from NZB mice induces peritoneal B-1 cell secretion of RBC autoantibodies in the gut lumen and results in AIHA (Nisitani et al 1997). In contrast to other studies, this study also found that TH2 cells could cause AIHA through Il-5 and Il-10 induction of autoantibody secreting B-1 cells. Additionally, elimination of B-1 cells reduces not only the amount of IgM autoantibody but also the amount of IgG autoantibody, demonstrating B-1 cell involvement in IgG production as well as IgM production (Murakami et al 1995). As a result of the evidence for both polyclonal and specific antigen-induced responses, a unifying model whereby antigen-induced specific responses are preceded by polyclonal activation has been proposed for all autoimmune diseases (Dziarski 1988).
Mueller and Eckhart have proposed yet another mechanism to explain DIHAs. They suggest that, rather than through inhibition of T cell suppresser function or a failure of immune tolerance, all forms of drug induced antibodies are caused by the formation of a composite antigenic structure upon binding of the drug or drug metabolites to a site on the RBC. Antibodies elicited by an altered membrane structure can react with the drug, the drug-RBC complex, and/or the RBC alone (1990).
|Potential Defects in Self Tolerance
|Development of Autoimmunity
-B cell hyper-reactivity
Whether or not transfusions should be used to treat AIHA is still controversial. Serological evaluations routinely done before blood transfusions are complicated in AIHA, especially in the WAIHA form of the disease because special procedures most be preformed to separate autoantibodies from alloantibodies before an indirect agglutination test is performed (Jefferies 1994). Additionally WAIHA antibodies may destroy transfused cells as rapidly as they destroy self-RBCs, unless the transfused blood is Rh- (Smith 1999). CAD and PCH present less risk but transfused cells may also be incompatible in these forms of AIHA as well (Domen 1998). Additionally repeated transfusions may increase the risk of alloimmune response. Thus some researchers argue that the temporary benefits of transfusion are not warranted (Smith 1999, Gibson 1998). Other researchers argue that the transfusions do not result in intensified hemolysis nor alloimmunuzation (Salama and Berghofer 1992). However, when patients experience acute AIHA and are at high risk for central nervous system or cardiac failure, transfusion is warranted, even where blood has not been thoroughly cross-matched (Hashimoto 1998).
In addition to conventional treatment, more recent therapies have been explored. Danazol, a modified androgen which reduces both the amount of c3 bound to RBCs and the number of Fc receptors on macrophage, may alleviate WAIHA (Eckman 1998). Cyclosporin A has also recently been used to successfully treat WAIHA through inhibition of T cell activation and proliferation (Smith 1999). Trials with intravenous IgG immunoglobulins (IV-IgG) have shown variable success (Domen 1998). IV-IgG anti-idiotypic immunoglobulins appear to neutralize autoantibodies by forming idiotype-anti-idiotype complexes to prevent coating of the RBC, bind B cell receptors to decrease autoantibody production, and regulate T cell function (Choudry, Mahapatra, and Kashyap 1998). Another experimental treatment has effectively reduced hemolysis through administration of monoclonal antibodies for the IgG Fc receptor (Gibson 1998).
CAD does not respond well to many of the conventional treatments; however,
the disease can often be treated through supportive methods alone such
as by keeping the patient warm and by drinking lots of fluids. If
CAD or PCH is severe, transfusions or cytotoxic agents may be administered
(Gibson 1998). Plasmaphoresis to remove autoantibody is sometimes
effective in temporary treatment of CAD, but usually not WAIHA, because
at 37'C IgM is no longer bound to RBCs and is intravascularly distributed
(Gibson 1998). Because hemolysis occurs at decreased temperatures,
cardiac patients should be tested for CAD prior to surgery, as cold heart
surgery could lead to severe hemolysis (Hashimoto 1998). Conventional treatment
is usually not used in DIHA either; discontinuation of the reactive drug
usually resolves RBC hemolysis.
Barker, R., Hall, A., Standen, G., Jones, J., Elson, C. 1997. Identification of T-Cell Epitopes on the Rhesus Polypeptides in Autoimmune Hemolytic Anemia. 90 (7) : 2701-2715.
Choudhry, V., Mahaptra, M., Kashyap, R. 1998. Immunoglobulin Therapy in Immunohematological Disorders. Indian Journal of Pediatrics. 65 (5) : 681-690.
Cornell University Medical School. 1996 Oct 3. Cornell Pathology
Image Collection-Classification of Anemias.
2000 April 19.
Day, M. 1999. Antigen Specificity in Canine Autoimmune Haemolytic Anaemia. Veterinary Immunology and Immunopathology. 69 (2-4) : 215-224.
Diiulio, N., Fairchild, R., Caulfield, M. 1997. The Anti-Erythrocyte Autoimmune Response of NZB Mice. Identification of Two Distinct Autoantibodies. Immunology. 91 (2) : 246-251.
Domen, R. 1998. An Overview of Immune Hemolytic Anemias. Cleveland Journal of Medicine. 65 (2) : 89-99.
Dziarski, R. 1988. Autoimmunity : Polyclonal Activation or Antigen Induction? Immunology Today. 9 (11) : 340-342.
Eckman, J. 1998 April 14. Disorders of Red Cells. <http://www.emory.edu/INT_MED_REV/Atlanta/paper/paper.htm> Accessed 2000 April 15.
Gibson, J. 1998. Autoimmune Hemolytic Anemias : Current Concepts. Australian and New Zealand Journal of Medicine. 18 (4) : 625-637.
Hashimoto, C. 1998. Autoimmune Hemolytic Anemia. Clinical Reviews in Allergy and Immunology. 16 (3) : 285-295.
Hernandez-Jodra, M., Hudnall, S., Petz., L. 1990. Studies of In Vitro Red Cell Autoantibody Production in Normal Donors and in Patients with Autoimmune Hemolytic Anemia. Transfusion. 30 (5) : 411-416.
Jefferies, L. 1994. Transfusion Therapy in Autoimmune Hemolytic Anemia. Transfusion Medicine. 8 (6) : 1087-1104.
Kennedy, R. 2000. The Doctors' Medical Library-Hemolytic
> Accessed 2000 April 7.
Merck. 2000 March 19. The Merck Manual of Diagnosis and Therapy- Anemias Caused by Excessive Hemolysis. < http:/www.merck.com/pubs/mmanual/section11/chapter127/127d.htm > Accessed 2000 April 10.
Meyer, D., Schiller, C., Westermann, J., Izui, S., Hazenbos, W., Verbeek, J., Schmidt, R., Gessner, J. 1998. FcYRIII (CD16)-Deficient Mice Show IgG Isotype-Dependent Protection to Experimental Autoimmune Hemolytic Anemia. Blood. 92 (11) : 3997-4002.
Mueller-Eckhart, C., Salama, A. 1990. Drug-Induced Immune Cytopenias : A Unifying Pathogenetic Concept with Special Emphasis on the Rule of Drug Metabolites. Transfusion Medicine Reviews. 4 (1) : 69-77.
Murakami, M., Honjo, T. 1995. B-1 Cells and Autoimmunity. Annals of the New York Academy of Sciences : Vol 764. Eds. Boland, B, Cullinan, J., Kimball, C. New York : New York Academy of Sciences. 402-409.
Nisitani, S., Murakami, M., Honjo, T. 1997. Anti-Red Blood Cell Immunoglobulin transgenic Mice. An Experimental Model of Autoimmune Hemolytic Anemia. Annals of the New York Academy of Sciences : Vol 815. Eds. Boland, B, Cullinan, J., Kimball, C. New York : New York Academy of Sciences. 246-252.
Perry, F., Barker, R., Mazza, G., Day, M., Wells, A., Shen, C., Schofield, A., Elson, C. 1996. Autoreactive T Cell Specificity in Autoimmune Hemolytic Anemia of the NZB Mouse. European Journal of Immunology. 26 (1) : 136-141.
Petz, L. 1993. Drug-Induced Autoimmune Hemolytic Anemia. Transfusion Medicine Reviews. 7 (4) : 242-254.
Rosenfield, R., Diamond, S. 1981. Diagnosis and Treatment of the Immune Hemolytic Anemias. Haematologia. 14 (3) : 247-256.
Salama, A., Berghofer, H. 1992. Red Blood Cell Transfusion in Warm-Type Autoimmune Haemolytic Anaemia. Lancet. 340 (8834-8835) : 1515-1526.
Shen, C., Mazza, G., Perry, F., Beech, J., Thompson, S., Corato, A., Newton, S., Barker, R., Elson, C. 1996. T-Helper 1 Dominated Responses to Erythrocyte Band 3 in NZB Mice. Immunology. 89 (2) : 195-199.
Shibata, T., Berney, T., Reininger, L., Chicheportiche, Y., Ozaki, S., Shirai, T., Izui, S. 1991. Monoclonal Anti-Erythrocyte Autoantibodies Derived from NZB Mice Cause Autoimmune Hemolytic Anemia by Two Distinct Pathogenic Mechanisms. International Immunology. 2 (12) : 1133-1142.
Smith, L. 1999. Autoimmune Hemolytic Anemia : Characterization and Classification. Clinical Laboratory Science. 12 (2) : 110-114.
Sullivan, J. 2000 Jan 29 Cells Alive- The Gallery. < http://www.cellsalive.com/ > Accessed 2000 April 14.
Territo, M. 1997 Dec 8. Pathophysiology of Disease- Hematopathology-Anemia. <http://www.pathnet.medsch.ucla.edu/med-edu/ppd/book/image30.gif> Accessed 2000 April 14.
Uthman, E. 2000 March 18. Blood Cells and the CBC.
< http://www.neosoft.com/~uthman/blood_cells.html> Accessed 2000 April 10.
Uthman, E. 2000 March 18. Hemolytic Anemias.
< http://www.neosoft.com/~uthman/hemolytic_anemia/hemolytic_anemia.html > Accessed 2000 April 10.
Woodroof, J. 2000 Jan 10. Pathology 851 Blood and Lymphoid Tissues I Supplemental Image Database. < http:/www.kumc.edu/instruction/medicine/pathology/ed/ch_20a/ch20a_nf.html> Accessed 2000 April 10.
Wright, M., Smith, L. 1999. Laboratory Investigation of Autoimmune Hemolytic Anemias. Clinical Laboratory Science. 12 (2) : 119-122.
Wright, M. 1999. Drug-Induced Hemolytic Anemias- Increasing Complications to Therapeutic Interventions. Clinical Laboratory Science. 12 (2) : 115-118.
Zilow, G., Kirschfink, M., Roelcke, D. 1994. Red Cell Destruction
in Cold Agglutinin Disease. Infusionstherapie und Transfusionmedizin.
21 (6) : 410-415.
Davidson College Immunology Home Page.
Send comments, questions, and suggestions to: email@example.com