[Drug-induced immunocytopenia]

Drugs are responsible for the causative antibodies in 10-20% of all cases with immunocytopenias (immune hemolytic anemias, immune thrombocytopenias or immune granulocytopenias). Usually only one cell type is affected in the same patient, but rarely two cell types can be simultaneously involved. The immunocytopenia can precipitously occur within minutes to hours, or it develops gradually throughout days and weeks. In the former case the reactions are usually due to complement-activating IgG- and/or IgM-antibodies, and in the latter case to non-complement-activating IgG-antibodies without complement activation. The corresponding antigens could be complexes of the drug (metabolites) and structures of the cell membrane (drug-dependent antibodies) as well as cell antigens alone (autoantibodies). These findings indicate that the immunogenicity of haptens (metabolites and less commonly the drug itself) can result not only from their stable conjugation on macromolecules, but also from a loose interaction between the drug (particularly its metabolites) and the cell surface.

A case of immune complex hemolytic anemia, thrombocytopenia, and acute renal failure associated with doxepin use

A 46-year-old man suffering from neurotic depression complicated by alcohol and benzodiazepine dependence developed Coombs'-positive hemolytic anemia and thrombocytopenia with acute renal failure after 5 weeks of monotherapy with doxepin at a final dosage of 100 mg daily. The patient recovered completely after discontinuation of doxepin, blood exchange transfusion, and repeated hemodialyses. He had no history of hematologic abnormalities, exposure to other toxins, or glucose-6-phosphate dehydrogenase deficiency. Because other etiologic factors were ruled out, the short interval between the onset of the patient's hemolytic crisis and the administration of doxepin highly suggests that doxepin was associated with the complication. To the best of our knowledge, ours is the first report of this adverse effect following doxepin administration.

Acute immune hemolysis induced by a degradation product of amphotericin B

We report here on an eight-year-old boy who first developed acute intravascular hemolysis following therapy with amphotericin B (AmB) and subsequently a delayed hemolytic transfusion reaction due to alloantibodies. Although there is as yet no evidence for metabolism of AmB in vivo, the hemolysis appeared to be the result of sensitization against a degradation product of the drug. The patient's serum contained a hemagglutinating IgM antibody that reacted with all red blood cells (RBC) tested in the presence of plasma obtained from patients receiving AmB (ex vivo antigen), but not in the presence of their urine, AmB itself, or with AmB-pretreated RBC. These findings indicate that the antibody was directed against a degradation product of AmB, presumably a trace metabolite, that has not yet been identified.

[Immune hemolytic anemia induced by drugs]

Drugs may induce immune destruction of red blood cells through either autoantibody formation or antibody formation against the drug or its metabolites. In the former scheme the drug is believed to induce autoimmunization by causing some central dysregulation of the immune system the mechanism of which is still unclear. In the latter, data have been accumulating during the last decade pertaining to clinical manifestations, incriminated drugs, and immunohematologic diagnosis. The sharp specificity and individual variations of drug specific antibodies as well as the recognition of the receptor role of red cell blood group antigens explaining the involvement of red cells in drug-antidrug complexes are among the recent achievements in the knowledge of these anemias.

Carbimazole-induced immune haemolytic anaemia: role of drug-red blood cell complexes for immunization

We report on a patient who developed acute intravascular immune haemolysis while receiving carbimazole. Serological studies revealed a strongly (3+) and a discretely positive (1+) direct antiglobulin test due to C3d and IgG respectively, and a very weak IgG autoantibody in the eluates. Serum from the patient contained specific carbimazole-dependent red blood cell (RBC) antibodies which reacted with all normal human RBC in the presence of free carbimazole as well as with RBC coated with the drug either in vitro or in vivo, although carbimazole itself is not detectable in plasma after oral administration. The results provide direct evidence for the sequence of the drug-RBC-antibody interaction and show that the RBC (and not plasma proteins) function as 'carrier-like' macromolecules in the immune response.

Recombinant alfa-2b-interferon therapy in untreated, stages A and B chronic lymphocytic leukemia. A preliminary report

Ten patients with B-chronic lymphocytic leukemia (B-CLL) (Six Stage A and four Stage B), who had not received therapy previously, were treated with recombinant alfa-2b-interferon (Schering Corporation, Kenilworth, NJ). The low dose of 1.5 MU was administered by intramuscular (IM) injection three times a week for the first week. The dose was increased to 3.0 MU thereafter until 3 months of therapy were completed. In the responding patients, treatment was continued in the same dose and schedule for 3 additional months. Interferon was tolerated without major toxicity by most patients. Objective tumor response (one complete response and four partial hematologic responses [PHR]) was observed in five of ten patients (50%). Severe autoimmune hemolytic anemia developed in one of the nonresponders at 8 weeks. Therefore, treatment had to be discontinued. Our study demonstrated single alfa-2b-interferon antitumor activity in untreated B-CLL patients with stable disease, and indicated that further trials of alfa-2b-interferon, possibly combined with chemotherapy, may be justified.

[Autoimmune hemolytic anemia during treatment with cyclosporin after liver transplantation]

Hemolytic anemia was observed in a 36 year-old liver transplant patient. The immunosuppressive regimen included cyclosporine A and prednisolone. Hemolysis appeared a few days after amoxicillin treatment. The presence of anti-D allo-antibodies, auto-antibodies active against erythrocytes, and an immuno-allergic phenomenon against erythrocytic membrane coated by cyclosporine was demonstrated. Increase of daily dose of prednisolone without modification of cyclosporine doses was followed by the disappearance of allo- and auto-sensitization. The role of amoxicillin in promoting these hemolytic phenomenons may be suggested.

Fenoprofen-induced immune hemolysis. Difficulties in diagnosis and complications in compatibility testing

A 70-year-old man developed severe immune intravascular hemolysis and renal failure following ingestion of fenoprofen, a nonsteroidal, anti-inflammatory drug. The patient's red blood cells were sensitized with both IgG and C3d. The serum reacted with normal red blood cells in the presence and absence of the drug. Addition of albumin to the serum inhibited the reactivity with both neat and drug-treated serum. These atypical serologic findings for drug-related immune hemolytic anemia were explained by (1) the measurement of fenoprofen by high performance liquid chromatography (HPLC) in the neat serum; and (2) solid-phase adsorption studies showing that albumin can bind drug, leading to the disappearance of agglutination when albumin is added. This case demonstrates the utility of drug levels and adsorption techniques to confirm the diagnosis of drug-induced immune hemolytic anemia despite the anomalous serologic results which obscured the diagnosis and management of the patient.

Methyldopa-induced autoantibodies against red blood cells

Methyldopa therapy results in the formation of red cell autoantibodies in 10-20% of patients taking the drug for longer than 4 months. These red cell antibodies are true autoantibodies, that is they are directed against an autoantigen on the red blood cell membrane and not against the drug or against a drug-altered antigen. The target membrane antigen is usually within the Rhesus system, although often the antibody specificity cannot be defined. Red cell antibody is usually detectable in the patient's sera as well as on the red cells. The autoantibody is usually a warm reacting IgG antibody. Most patients who develop these autoantibodies do not go on to develop hemolytic anemia in spite of high titres of antibodies on their red cells. In addition, these patients do not tend to develop hemolysis if methyldopa therapy is continued. Rarely patients develop hemolytic anemia which can be severe. Differences in antibody characteristics, including subclass restriction, complement-binding ability, or titre do not explain why some patients with autoantibody hemolyze while most do not. One group of investigators found that hemolyzing patients had IgM on their red cells while those who did not had IgG only. But while this observation could explain why some patients (IgM-sensitized red cells) hemolyze, it does not explain why most patients with IgG-sensitized red cells do not hemolyze. Why the autoantibody forms is not known but some investigators have proposed that the drug may directly affect B or T cells with resulting impairment of immune tolerance.(ABSTRACT TRUNCATED AT 250 WORDS)

Toxic effects of drugs on erythrocytes

The erythrocyte abnormality most often associated with the toxic effects of commonly used drugs is premature destruction. The mechanisms of erythrocyte destruction include: denaturation of unstable hemoglobins, oxidation of sulfhydril groups in hemoglobin and the erythrocyte membrane in the presence of glucose-6-phosphate dehydrogenase deficiency, direct effects on enzymes, cholesterol or phospholipids of the erythrocyte membrane, and various autoimmune reactions. Therapy includes stopping the drug and transfusions when anemia is severe. Splenectomy and steroids are rarely needed. A careful medical history and use of drugs only for good indications may avoid many of these reactions.

Characterization of antibody and selection of alternative drug therapy in hydrochlorothiazide-induced immune hemolytic anemia

The authors report the clinical and laboratory findings of a patient who had severe immune hemolytic anemia due to hydrochlorothiazide (HCTZ). In this case, the HCTZ antibody reacted not only with other thiazide and thiazide-like drugs, but also with a chemically unrelated diuretic, ethacrynic acid. These results indicate that HCTZ antibody activity is not restricted solely to the thiazides and imply that therapy with any of the reactive drugs would be contraindicated for this patient. The serologic screening for drug reactivity may be useful for selecting alternative therapy for patients with drug-induced immune hemolytic anemia.

[Autoimmune hemolysis after Catergen ([+]-cyanidanol-3)]

Clinical and serological data in 5 cases of autoimmune hemolysis following therapy with Catergen are reported and compared to the data in similar literature reports. The main argument in favour of Catergen as causative agent in our 5 cases was rapid remission of hemolysis within 2 1/2 to 10 weeks of withdrawing Catergen treatment. Besides causing hemolysis mediated by drug dependent antibodies, long term treatment with Catergen may induce formation of IgG autoantibodies against red blood cells with or without overt hemolysis.

Cianidanol and its metabolites bind tightly to red cells and are responsible for the production of auto- and/or drug-dependent antibodies against these cells

Cianidanol ((+)-2-(3,4-dihydroxyphenyl)-3, 5, 7-chromantriol) is a flavonoid which has been associated with severe immune haemolysis by as yet unclear mechanisms. We report six patients who developed haemolysis while receiving the drug. The disorder was episodic in all patients and resolved after discontinuing the drug. The causative antibodies could be demonstrated in all six cases, even when the haemolytic episode was more than 1 year prior to this study. One patient had developed drug-independent IgG autoantibodies, another simultaneously developed autoantibodies and drug-dependent antibodies (ddab) of the IgG class, while the remaining four patients had only ddab of the IgM and/or the IgG classes. All ddab were reactive with red blood cells (RBC) in the presence of the drug and/or its metabolites (ex vivo antigens), and, quite unexpectedly, also with RBC coated with the drug (metabolites) in vitro or in vivo. This reactivity did not change either by preincubating the antibodies with the drug or by adding large amounts of the drug to the mixture of drug-coated cells plus antibody. It seems that the stable association of cianidanol with RBC generates antigenic sites against which a heterogeneous immune response is elicited giving rise to long-lasting drug-dependent antibodies as well as autoantibodies.

On the mechanisms of sensitization and attachment of antibodies to RBC in drug-induced immune hemolytic anemia

The mechanisms of sensitization and attachment of drug-dependent antibodies to RBC in drug-induced immune hemolytic anemias are largely speculative. Nomifensine has been incriminated in causing immune hemolysis in a large number of patients. The hemolysis was usually of the so-called immune complex type, less commonly of the autoimmune type, and more surprisingly, few patients had developed both types of hemolysis. To determine whether nomifensine (metabolite)-dependent antibodies (ndab) exhibit specificity for antigenic structures of RBC membranes, 30 ndab were tested against large panels of RBC with common and rare antigens. We found that only 14 out of 30 ndab were invariably reactive with all cells tested. Nine antibodies were, similar to the majority of idiopathic or drug-induced autoantibodies, not or only weakly reactive with Rhnull RBC. Three antibodies did not react with cord RBC and could be inhibited by soluble I antigen. The remaining four antibodies gave inhomogeneous reaction patterns or were even negative with selected RBC; their specificity could not be identified. On a Scatchard plot analysis of one ndab, a maximum of 173,000 drug-dependent antibodies of the IgG class can specifically bind per RBC in the presence of the drug. Although nomifensine and its metabolites do not attach tightly onto RBC, our results clearly indicate that RBC do not act as "innocent bystanders," but rather serve as a surface for a loose attachment of drugs that possibly cause a subtle structural change in the cell antigens and, by this means, allow in vivo sensitization; and a specific binding of the resultant antibodies. This concept would explain why these antibodies can be directed against drug-cell complexes, against cell antigens alone (autoantibodies), or against both in the same patient.

Fluorouracil (5-FU) induced acute immune haemolytic anaemia

A female patient on Fluorouracil (5-FU) therapy for rectal carcinoma developed acute intravascular haemolysis with her fifteenth drug injection; a similar episode occurred later after a controlled challenge with the drug. A 5-FU-dependent complement-activating IgM antibody, which reacted with RBC in an indirect antiglobulin test, was detected in her serum. The antibody did not react with cord RBC, but a blood group specificity could not be determined. An excess of 5-FU in the indirect antiglobulin tests resulted in inhibition of these reactions.

Rh blood group-specific antibodies in immune hemolytic anemia induced by nomifensine

Nomifensine (Merital, Alival; Hoechst, Frankfurt, FRG), an antidepressant drug, may cause immune hemolytic anemia (IHA) of the so-called immune complex type that is believed to occur by means of an innocent-bystander mechanism. In this report we describe findings that are not consistent with this mechanism in a patient with nomifensine-induced intravascular IHA associated with renal failure. In vitro studies showed a transitory positive direct antiglobulin test (DAT) due to IgG, IgM, and C3 fixation. The causative antibodies were found to be a drug-independent IgM antibody in the serum and eluate that reacted only with E-positive RBC, although the patient's RBC were E-negative; an IgG antibody in the serum and initial eluates that showed a stronger reaction with e-positive than with e-negative or Rhnull RBC, but only in the presence of ex vivo antigen (ie, urine containing the drug and all its metabolites); and an IgM antibody in the serum and initially also on the patient's RBC that, in the presence of ex vivo antigen as well as in the presence of known metabolites of the drug, agglutinated all RBC equally strongly, but was hemolytically more active against E-positive than E-negative cells. Within a few days of stopping the drug the hemolysis rapidly resolved without administration of prednisone, the DAT became negative with anti-IgG and anti-IgM, and the drug-independent anti-E disappeared, but both metabolite-dependent antibodies remained detectable in the patient's serum. We conclude that the production and specificity of the causative antibodies in this case were controlled by a larger antigenic site, presumably consisting of the drug and/or its metabolites plus RBC antigens, rather than by epitopes of the drug or metabolites alone.

Induction of hemolytic anemia by nonsteroidal antiinflammatory drugs

The incidence of immune hemolytic anemia (IHA) is increasing. The proliferation of pharmaceuticals is a contributing factor to this increase. IHA is an uncommon, though significant, adverse effect of a wide variety of drugs. Several recent case reports have implicated the nonsteroidal antiinflammatory drugs (NSAIDs). Because of the extensive use of this class of drugs, a review of case reports, clinical studies, and in vitro research was conducted on NSAID-induced IHA. Mefenamic acid, ibuprofen, sulindac, naproxen, tolmetin, feprazone, and aspirin are reported to cause IHA, with mefenamic acid most frequently implicated. Mefenamic acid appears to cause hemolytic anemia by an autoimmune mechanism similar to methyldopa and aspirin by an immune complex mechanism. However, there is insufficient information concerning ibuprofen, sulindac, naproxen, tolmetin, and feprazone to assign specific mechanisms of immune hemolysis. In individuals with glucose-6-phosphate dehydrogenase (G-6-PD) deficiency, aspirin at usual therapeutic doses is not a predisposing factor to hemolysis unless other risk factors are present. Although individuals with G-6-PD deficiency are at increased risk of developing hemolytic anemia when exposed to oxidizing stresses, the use of NSAIDs does not appear to increase this risk significantly. Because NSAID-induced IHA occurs infrequently and the sensitivity of currently used tests to detect drug-dependent antibodies is limited, routine serologic testing in patients receiving NSAIDs is not justified. If hemolytic anemia occurs in a NSAID-treated patient and the history is consistent with a drug-induced etiology, the NSAID should be discontinued. With discontinuation of the offending agent, the prognosis is good. There is a rapid hematologic recovery, with a slow resolution of abnormal serologic findings.

The mechanisms of sulfonylurea-induced immune hemolysis: case report and review of the literature

Chlorpropamide, a sulfonylurea antidiabetic drug, was found to be the etiologic agent in a patient with immune hemolytic anemia. Hemolysis was severe, and ceased promptly when the drug was discontinued. The direct antiglobulin test was positive for complement, and the indirect antiglobulin test was also positive when the drug, patient serum, and fresh serum complement were all present. In this patient, and in four other patients described in the literature, hemolysis was mediated by the immune complex mechanism. In reports of two other patients taking a sulfonylurea drug, a milder form of hemolysis was mediated by the hapten mechanism. Sulfonylurea drugs are widely used. Since immune complex-mediated hemolysis in these patients is dramatic, it is not likely to be missed. However, the slower, milder hemolysis mediated by the hapten mechanism may be more common than is realized. Such patients may have only mild anemia and a near-normal reticulocyte count, making the diagnosis difficult. It is thus important to consider drug-immune hemolysis in patients who become anemic while taking sulfonylurea drugs.

Two types of nomifensine-induced immune haemolytic anaemias: drug-dependent sensitization and/or autoimmunization

Thirty-one patients who developed immune haemolytic anaemia while receiving nomifensine were studied. We provide evidence that nomifensine can cause two forms of immune haemolytic anaemia: one that is associated with an abrupt haemolytic episode due to drug-dependent antibodies, and a less acute form associated with IgG autoantibodies. The majority of patients' serum samples (23 cases) contained IgG and/or IgM antibodies reacting to a highly variable extent with red blood cells (RBC) only in the presence of the drug and/or its metabolites. Sera of six patients contained IgG autoantibodies which reacted, like those in warm autoimmune haemolytic anaemia, with RBC in the absence of drugs. Three patients had developed both types of antibodies. From a diagnostic viewpoint, nine of the drug-dependent antibodies could not be identified by using the drug itself, but by its known (three cases) or unknown urine-born metabolites (ex vivo antigens). We conclude that nomifensine can induce in vivo the production of RBC drug- and/or metabolite-dependent antibodies, autoantibodies, or both in the same patient.

Intravascular haemolysis and acute renal failure induced by nomifensine

Nomifensine is a widely used antidepressant agent with few side-effects but haemolytic anaemia with acute renal failure has been reported with both intermittent therapeutic dosage and self-inflicted overdose. We report here a case of acquired immune haemolytic anaemia with acute renal failure induced by second exposure to a single capsule of nomifensine.

Immune haemolytic anaemia and nomifensine treatment in north west England 1984-85: report of six cases

Nomifensine, an antidepressant used for eight years in Britain, was particularly popular in the north west of England. Haemolytic anaemia was recognised as a rare side effect, but in 1984 a cluster of six cases was referred to the North West Regional Transfusion Centre, reflecting an incidence of 0.006%. These were collected within 18 months and showed variable serological features, indicating that antibodies associated with nomifensine treatment are neither rare nor of one particular type. The accumulation of this and similar data contributed to its withdrawal from the British market in January, 1986.

Probenecid induced immune hemolytic anemia

We report a patient who, while receiving probenecid and colchicine for acute gouty arthritis, developed severe hemolytic anemia in association with a generalized rash. The hemolysis was immune mediated as shown by a positive direct Coombs' test. In vitro hematologic studies showed that a probenecid dependent antibody was present.

Severe reversible autoimmune haemolytic anaemia and thrombocytopenia associated with diclofenac therapy

Severe immune haemolytic anaemia and thrombocytopenia developed in a 71-year-old female within 10 d of starting diclofenac (Voltarol) therapy. These complications resolved within 3 weeks of discontinuation of the drug and corticosteroid therapy. A warm autoantibody of the IgG type together with C3 was found in the direct antiglobulin test of the patient's RBC. The patient's serum and RBC eluate contained a warm autoantibody which reacted with all commercial panel cells without the addition of diclofenac, and gave a negative reaction with Rh null and -D- RBC. This pattern of interactions is similar to haemolysis associated with alpha-methyldopa, indicating the presence of autoantibodies directed against structural components common to all Rh antigens. The coexistence of immune thrombocytopenia and immune haemolytic anaemia is suggestive of an autoimmune disease caused by modified T-cell regulation. Although immune haemolytic anaemia is a rare complication of diclofenac therapy, our observations illustrate the severity of haemolytic anaemia in the occasional patient and stress the need for increased awareness of such a development.

Impaired reticuloendothelial function in patients treated with methyldopa

In about 20 per cent of patients taking the antihypertensive agent methyldopa, IgG autoantibodies form against red cells, but most such patients do not have hemolysis. The reason for this is uncertain; it does not appear to be explained by known characteristics of the autoantibody. Since antibody-dependent reticuloendothelial function is an important determinant of cell clearance, we measured reticuloendothelial function in nine patients taking methyldopa. We did this by measuring the rate of clearance of radiolabeled autologous red cells sensitized with anti-D alloantibody. Five patients had a positive direct antiglobulin test, and four did not. Only one patient had laboratory evidence of hemolysis. The patients without hemolysis had significantly impaired reticuloendothelial clearance. In contrast, the patient with hemolysis did not have impaired reticuloendothelial function. This study suggests that in patients taking methyldopa who have a positive direct antiglobulin test the absence of hemolysis may be caused by an impairment in reticuloendothelial function. That the drug itself may be responsible for the impairment is suggested by the occurrence of abnormal reticuloendothelial function in patients taking the drug who have a negative direct antiglobulin test.

Tolmetin-induced hemolysis

A case of drug-associated immune hemolysis in a patient taking tolmetin for arthritic pain is described. Serologic tests showed that in the absence of tolmetin, the patient had a negative antibody screening test but a strongly positive direct antiglobulin test. An eluate prepared from the patient's red cells caused agglutination of all cells tested. However, addition of tolmetin revealed a high-titered tolmetin-dependent antibody in the patient's serum; the addition of tolmetin did not affect the results obtained with the eluate. within 3 months after the patient discontinued tolmetin, his hematocrit had increased to 45 percent, and his jaundice and bilirubinuria had disappeared. These results are similar to those described for zomepirac, another of the group of nonsteroidal anti-inflammatory medications.

Immune hemolytic anemia associated with probenecid

Upon hospital admission a patient was found to have severe anemia and a strongly positive direct antiglobulin test (DAT). The patient was taking probenecid periodically for gout. An antibody was detected in the patient's serum that only reacted with red blood cells (RBCs) when probenecid was added. Eluates from the patient's RBCs, with and without the presence of drug, were nonreactive. Upon the discontinuation of probenecid, the patient's hemoglobin level improved steadily. We believe this to be the first reported case of immune hemolytic anemia associated with probenecid.

Drug induced red blood cell autoantibodies co-developed with drug specific antibodies causing haemolytic anaemias

Four individuals with anti-glafenine, anti-latamoxef and anti-teniposide antibodies were found to have an associated red blood cell autoantibody. The two components could be separated by selective absorption and showed distinct time course patterns. In three patients a well-defined blood group antigen was recognized as the receptor for both auto- and drug specific antibodies. Similarities between this type of immune response to drugs and the well-known hapten and carrier specificities developed in animals immunized by hapten-carrier conjugates are discussed.

Immune hemolytic anemia associated with sulindac

A number of adverse reactions have been associated with sulindac, a nonsteroidal anti-inflammatory drug, but acute hemolytic anemia associated with sulindac has not yet been reported, to our knowledge. We encountered two cases of severe immune hemolytic anemia, one fatal, which appear to have been related to sulindac. In vitro studies provided evidence that antibodies to sulindac and its metabolites were present in the serum of both patients.

Immune hemolytic anemia following administration of antithymocyte globulin

Antithymocyte globulin is commonly used as a therapy for rejection of transplanted organs. Its use can be associated with many side effects. We report a significant hemolytic anemia following therapy with horse-derived antithymocyte globulin. Several commercial Coombs' sera failed to yield a positive Coombs' test with the patient's erythrocytes; however, we were able to detect horse immunoglobulin on the patient's cells as well as on control red cells incubated with this particular preparation of antithymocyte globulin.