Search

A high incidence (39%) of positive direct antiglobulin tests (DATs) has been reported in patients taking Unasyn [ampicillin sodium plus sulbactam sodium (a beta-lactamase inhibitor)]. Three of four patients, with positive DATs, receiving Unasyn or Timentin [ticarcillin disodium plus clavulanate potassium (also a beta-lactamase inhibitor)] developed a haemolytic anaemia (HA) associated with a positive DAT, which resolved when drug therapy was stopped. The patients' sera did not react with red blood cells (RBCs) in the presence of Unasyn or Timentin, but when drug-treated RBCs were tested, patients' sera and normal sera reacted equally by indirect antiglobulin test. Following incubation in normal sera, RBCs treated with Unasyn, Timentin, Augmentin (amoxicillin + clavulanate), sulbactam and clavulanate reacted with anti-human globulin and anti-human albumin (an index of non-specific adsorption); RBCs treated with ampicillin and amoxicillin were nonreactive. The beta-lactamase inhibitors sulbactam and clavulanate seem to cause nonimmunologic adsorption of protein onto RBCs in vitro. This may explain the high incidence of positive DATs detected in patients taking Unasyn, which contains sulbactam. It was not possible to prove that there was a direct association between the nonspecific uptake of protein onto drug-treated RBCs in vitro with the positive DATs or the HA.

Summary. Kirtland et al (1980) suggested that methyldopa caused the production of red cell (RBC) autoantibodies by causing a persistent increase in lymphocyte cyclic AMP, which inhibited suppressor T cell function, leading to unregulated autoantibody production in some patients. They showed that significantly higher lymphocyte cyclic AMP concentrations were generated by lymphocytes from healthy donors after adding methyldopa, and by lymphocytes from patients who were receiving methyldopa compared to lymphocytes from healthy donors without methyldopa present. They also showed that methyldopa affected suppressor cell activity. We measured the effect of methyldopa and procainamide on suppressor cell activity, using a similar approach to Kirtland et al (1980). Suppressor cell activity was measured by measuring the amount of IgG, produced in vitro, by B cells following mitogen stimulation preceded by a 24 h incubation period. We found no significant increase in the amount of IgG generated by normal donor lymphocytes, when methyldopa or procainamide was present during the preincubation period. This is in contrast to the findings of Kirtland et al (1980). We also measured the amount of IgG generated in vitro by mitogen-stimulated lymphocytes from patients (with and without positive direct antiglobulin tests) taking methyldopa and compared this to the amount of IgG generated by lymphocytes from normal donors and patients (with and without positive direct antiglobulin tests). The results were similar for each group. This does not agree with the findings of Kirtland et al (1980) who found that lymphocytes from patients taking methyldopa produced more IgG in vitro than lymphocytes from normal donors. Our results do not support the hypothesis that methyldopa and procainamide induce autoantibodies by affecting suppressor cell function.

Cold agglutinin syndrome (CAS) is usually associated with IgM cold agglutinins with titres exceeding 1000 at 4 degrees C and a thermal amplitude of 30-32 degrees C. Occasionally patients are encountered who although having clinical and laboratory findings compatible with CAS do not have the characteristic serological findings. Thirty-two patients with a positive direct antiglobulin test due to complement sensitization were studied. Thirty-one of these patients had cold agglutinin titres greater than 64. Twenty-eight had haemolytic anaemia, including one patient with a cold agglutinin titre of only 8 against saline-suspended red cells. 53.6% of sera from patients with haemolytic anaemia reacted at 30 degrees C and 7.1% at 37 degrees C when albumin was not present, whereas in the presence of albumin all of the sera reacted at 30 degrees C and 67.9% reacted at 37 degrees C. None of the four patients without haemolytic anaemia reacted at 30 degrees C or 37 degrees C in the presence of albumin, even though one serum reacted to a titre of 1280 at 4 degrees C. Cold agglutinin titres and thermal amplitudes in the presence of bovine albumin were found to correlate better with haemolytic anaemia than reactions without albumin. If bovine albumin is utilized in compatibility testing, multiple cold autoabsorptions may be necessary before alloantibody activity at 37 degrees C can be excluded.

A 19-year-old patient with a long history of idiopathic thrombocytopenic purpura developed a potent antibody against a high-incidence antigen in the Kell blood group system. The direct antiglobulin test on his red cells was negative. His cells exhibited profound depression of Kell blood group antigens, but antigens of other blood groups were normal. Transfusion of incompatible blood was well tolerated and differential agglutination tests, using selected Rh antisera, showed in vivo survival of the transfused red cells for more than 8 weeks. However, the transfused red cells also showed acquired loss of Kell antigens. Five months after the initial findings, Kell-related antibody disappeared and Kell antigens reappeared on his red cells. The patient's serum stored from the initial investigation now reacted with his freshly collected red cells. These data suggest that an environmental agent in the patient's plasma was responsible for the temporary loss of Kell antigens from red cells in his circulation.

Summary. Current literature implies that haemolytic anaemia in infectious mononucleosis is regularly caused by the temporary production of high thermal amplitude cold agglutinins of anti-i specificity. More recently, Capra et al (1969) suggested interaction between IgG anti-i and anti-IgG antibodies as the cause of haemolytic anaemia in infectious mononucleosis. A detailed serologic evaluation of three patients during moderate to severe haemolytic anaemia in infectious mononucleosis revealed high thermal amplitude anti-i in only one. This patient's direct antiglobulin test (DAT) was negative using anti-IgG but was 3 + using anti-C3 and -C4. The serum antibody titre against cord or adult Oi cells was 512 at 4°C and 4 at 31°C. The anti-i was inhibited by mercaptoethanol, and anti-IgG was not found in the patient's serum. Patient 2 had a negative DAT and patient 3 had a 2 + DAT using anti-C3 and anti-C4. Anti-i was present only in low titre at 4°C and was unreactive at 20°C. It was inhibited by mercaptoethanol. These patients' sera contained anti-IgG antibodies. In none of our patients were warm autoantibodies detected. This data demonstrates that haemolytic anaemia in infectious mononucleosis is not necessarily associated with high thermal amplitude anti-i. Further, since the anti-i antibodies in patients 2 and 3 were of low titre, were inhibited by mercaptoethanol, and did not react at physiologic temperatures, the mechanism of haemolysis in these patients does not seem related to the interaction of anti-IgG and IgG anti-i antibodies. Further work is necessary to clarify the mechanism of haemolytic anaemia in infectious mononucleosis.