Case Report

Top Case Report Discussion Author & Article Info References

A 43-year-old man was well until 1984, when he had two episodes of lower-extremity deep venous thrombosis and pulmonary embolism; he was noted to have thrombocytopenia (platelet count, 9 x 10⁹/L). Treatment with warfarin was initiated and was continued for 1 year. Prednisone (40 mg/d) increased the platelet count, but thrombocytopenia recurred when attempts were made to taper the dose, so a splenectomy was done. Despite a transient increase in the platelet count, corticosteroids were again required to maintain the platelet count in the normal range. During the next 6 years, repeated attempts to taper prednisone to a dose less than 30 mg/d were unsuccessful. On initial evaluation at our center in 1991, the patient had several adverse sequelae related to long-term corticosteroid use, including obesity (body mass index, 58 kg/m²), hypertension, glucose intolerance, and posterior subcapsular cataracts. No clinical characteristics suggested systemic lupus erythematosus. Laboratory evaluation showed negative serologic test results for the human immunodeficiency virus (HIV) and an absence of antinuclear antibodies. In addition, tests for antibodies to native DNA and a panel of extractable nuclear antigens (Sm, nRNP, SS-A, and SS-B) were also negative. The erythrocyte sedimentation rate, hemoglobin concentration, leukocyte count and differential, partial thromboplastin time, prothrombin time, and serum complement proteins C3 and C4 were normal. Enzyme-linked immunoassay for IgG anticardiolipin antibodies was strongly positive (74.2 Harris units; normal < 12 units). No other tests for lupus anticoagulant activity were done. On the basis of history and laboratory test results as noted, a diagnosis of thrombocytopenia associated with the antiphospholipid antibody syndrome was made.

Reduction of the corticosteroid dose led to a marked decrease in the platelet count (31 x 10⁹/L; (Figure 1). Thereafter, the dose of prednisone was temporarily increased, and treatment with danazol was initiated at a dose of 400 mg/d. As the prednisone was tapered, the platelet count remained within the normal range with danazol alone. After several months, the dose of danazol was decreased to 200 mg/d and the platelet count subsequently fell. On commencement of a higher dose of danazol [800 mg/d], the platelet count increased and has remained normal at a lower dose (600 mg/d).

View larger version (49K): [in this window] [in a new window]   Figure 1. Response of the platelet count to corticosteroids and danazol over time.

Discussion

Top Case Report Discussion Author & Article Info References

Our patient presented with recurrent venous thrombosis and thrombocytopenia and was subsequently noted to have high-titer serum IgG anticardiolipin antibodies. Although further thrombotic episodes did not occur, thrombocytopenia remained a problem. Splenectomy was unsuccessful, and long-term use of corticosteroids had caused substantial morbidity. The efficacy of danazol in treating this patient is underscored by the steroid-sparing effect noted when the drug was first instituted and by the response to an increased dose of danazol after tapering of the danazol had caused a decrease in the platelet count. Danazol has been used effectively in other conditions associated with immune-mediated thrombocytopenia, including idiopathic thrombocytopenic purpura and systemic lupus erythematosus [3-5]. In many cases, long-term therapy with danazol has been possible, and adverse effects related to its use have often been dose dependent and reversible [3, 6, 7].

The mechanism underlying the association between anticardiolipin antibodies and thrombocytopenia has not been defined. Most patients with idiopathic thrombocytopenic purpura have antibodies to the platelet surface glycoproteins IIb-IIIa or Ib-IX [8]. Removal of antibody-coated platelets by the reticuloendothelial system is thought to be a relevant means of platelet elimination in that disease. Unfortunately, accurate testing to determine the presence of antibodies to specific platelet surface antigens is not universally available. Studies defining the prevalence of such antibodies in patients with thrombocytopenia associated with the antiphospholipid antibody syndrome remain to be done. Testing for antibodies to specific platelet surface antigens was not done in this case. An alternative mechanism might be that the thrombocytopenia associated with the antiphospholipid antibody syndrome results from the binding of platelet antigens other than glycoproteins IIb-IIIa or Ib-IX. Although resting platelets do not express negatively charged phospholipids on their surfaces, activation or injury of platelets can cause the expression of phosphatidylserine residues on their membranes [9]. These phospholipid residues may in turn be recognized and bound by anticardiolipin antibodies, resulting in thrombocytopenia. Such a mechanism might also explain an apparent paradox of the antiphospholipid syndrome, that is, a thrombotic diathesis coexisting with thrombocytopenia, without evidence for massive intravascular platelet consumption. Thus, perturbation of the endothelium associated with the antiphospholipid antibody syndrome may result in platelet activation. This might lead to thromboses but for some patients might also result in the removal of platelets by the reticuloendothelial system.

The mechanism of action of danazol in immune thrombocytopenias has also not been completely defined. It has been shown that danazol therapy may modulate the expression of Fc receptors on mononuclear phagocytes, thereby impeding the opsonization and destruction of platelets [4]. Danazol may also have an immunomodulatory function. Patients treated with danazol have had alterations in circulating T-cell subsets, which could affect antibody production in immune thrombocytopenias [10]. However, successful therapy with danazol has been reported not to alter concentrations of platelet-bound IgG antibodies [5]. The effect of danazol on antiphospholipid antibodies remains to be defined. Interestingly, therapy with danazol has been shown to modify the membranes of erythrocytes, rendering them less susceptible to osmotic lysis [11]. Because danazol is similar in structure to cholesterol, it might be incorporated into and affect the functions of cellular membranes. It may be hypothesized that by such a mechanism, danazol could modify the interaction of anticardiolipin antibodies with their antigens in platelet membranes.

Danazol is an effective and well-tolerated therapy in the immune thrombocytopenia of idiopathic thrombocytopenic purpura and systemic lupus erythematosus. Our case shows that it may also be a useful therapy in thrombocytopenia associated with the antiphospholipid antibody syndrome.