Autoimmune Hemolytic Anemia, Primary Adrenal Insufficiency, and the Antiphospholipid Syndrome BRANDEN HSU, BS,
MARK M. UDDEN, MD,
ABSTRACT: Autoimmune hemolytic anemia and adrenal insufficiency are rarely associated with the anti phospholipid antibody syndrome. A 49-year-old woman with a history of deep venous thrombosis and recurrent miscarriages was found to have active autoimmune hemolytic anemia after being admitted to the hospital for cholelithiasis. The patient was treated with corticosteroids and underwent laparoscopic cholecystectomy 1 month later. Two weeks after surgery she had acute adrenal insufficiency. Activated partial thromboplastin time was prolonged, and anti phospholipid antibodies were detected in significant titer. Her illness responded well to corticosteroid therapy. Her direct Coombs' test remained positive. It appears that the antiphospholipid antibody syndrome contributed to the development of venous thrombosis, recurrent miscarriages, autoimmune hemolytic anemia, adrenal insufficiency, and indirectly, pigment stone cholelithiasis in this patient. KEY INDEXING TERMS: Autoimmune hemolytic anemia; Adrenal insufficiency; Antiphospholipid syndrome. [Am J Med Sci 1997;314(1):41-43.]
A
ntiphospholipid antibody syndrome (APS) is a
~isorder characterized by acquired venous or arterial thrombotic events, recurrent fetal loss, and specific autoantibodies to phospholipids. 1 Autoimmune hemolytic anemia (AIHA) is an uncommon occurrence in patients with APS.2 Although this relationship is not clearly defined, antiphospholipid antibodies (aPL) may play a direct role in AIHA by acting as anti-red blood cell autoantibodies. 3 Adrenal insufficiency (AI) is a rare complication of APS. From the Department of Medicine, Baylor College of Medicine, Houston, Texas. Submitted October 21, 1996; accepted in revised form February 27,1997. Correspondence: Edward C. Lynch, MD, Professor of Medicine, Baylor College of Medicine, Department of Medicine (B501), Fondren-Brown Building, 6565 Fannin, Houston, Texas 77030-2707. THE AMERICAN JOURNAL OF THE MEDICAL SCIENCES
EDWARD C. LYNCH, MD
The development of AI appears to be caused by the intrinsic hypercoagulable state in APS, leading to adrenal vein thrombosis and subsequent necrosis of the adrenal gland. 4 We report a patient with AIHA in whom primary adrenal failure developed after laparoscopic cholecystectomy, and who was subsequently found to have APS. Case Report A 49-year-old woman was admitted to Ben Taub General Hospital, May 1, 1996, for cholecystectomy. She complained of right upper quadrant pain occurring after fatty meals for the past month. She also described a 3-month history of generalized fatigue, decreased appetite, nausea, vomiting, and a 28-kg loss of weight but no fevers, chills, arthralgia, or skin rashes. The patient denied having recent infections or the use of drugs. Her medical history was remarkable for two episodes of deep venous thromboses in 1986 that was treated with warfarin and 3 miscarriages, which all occurred during the first trimester. At admission, the patient was normotensive, and serum electrolytes were within normal limits. Computerized tomography of the abdomen showed multiple gallstones and bilaterally enlarged adrenal glands with low attenuation interpreted as hyperplasia (Figure 1). The spleen was normal and no lymphadenopathy was seen. Four days after admission hematocrit value dropped from 28.7% to 20.5%. The reticulocyte percentage was 5.8. Further evaluation revealed a strongly positive direct Coombs' reaction because of warm autoantibodies. The specific Coombs' test detected immunoglobulin G (lgG) but not C3d. Antinuclear antibody titer was positive at 1:80 with a diffuse pattern, but the patient did not fulfill other criteria for the diagnosis of systemic lupus erythematosus. 5 Treatment was started with intravenous corticosteroids, resulting in marked improvement symptomatically. At discharge she was advised to take 60 mg prednisone daily and gradually reduce dose. Surgery was postponed for 4 weeks. Follow-up hematology clinic visits disclosed increasing hematocrit values as the dose of prednisone was reduced from 60 mg to 5 mg during a period of 6 weeks. The patient underwent elective laparoscopic cholecystectomy, June 17, 1996, which revealed multiple small pigmented gallstones. She was discharged 2 days later in good condition and took her last dose of prednisone that day. Symptoms of fatigue, decreased appetite, nausea, and vomiting appeared during the next several days. She was readmitted, June 30, 1996, with severe hypotension. Blood pressure was 75/65 mm Hg. Physical examination was remarkable for mildly icteric sclerae. No skin rash, lymphadenopathy, or hepatosplenomegaly were present. Serum sodium was 125 mmollL, potassium 5.9 mmollL, bicarbonate 24 mmollL, chloride 93 mmollL, BUN 29 mg/dL, creatinine 1.2 mg/dL, glucose 93 mg/dL, total bilirubin 2.0 mg/dL, direct bilirubin 0.3 mg/dL, and serum LDH 212 UIL. Hemoglobin and hematocrit values were 10.6 g/dL and 28.9%, respectively; reticulocyte percentage was 0.6%. White blood cell count yielded 6,100
41
The Antiphospholipid Syndrome
Figure 1. Initial abdominal computerized tomography scan showing bilaterally enlarged adrenal glands (black arrows) with low attenuation, suggestive of hyperplasia and pigmented gallstones. (open arrow).
leukocytes with 1% eosinophils. Platelet count was 180,000//-LL. Peripheral blood smear showed no schistocytes. A clinical diagnosis of acute adrenal insufficiency was made and resuscitation with intravenous normal saline and hydrocortisone was begun. A low-dose (1 /-Lg) cosyntropin stimulation test showed no response; baseline cortisol level was 3.0 /-Lg/dL, I-hour cortisol was 3.0 /-Lg/dL (normal > 20 /-Lg/mL), and 2-hour cortisol was 2.5 /-Lg/dL. Results of urinalysis and thyroid function tests were normal. Activated partial thromboplastin time was elevated at 50.6 seconds (normal, 20 to 35 seconds). The prolonged value of the partial thromboplastin time did not change with a 1:1 mixture with normal plasma. The IgG anticardiolipin antibody (ACA) level was elevated moderately at 30 U (n = < 13 U ACA IgG), but ACA immunoglobulin M was normal at 5 U (n = < 7 U ACA IgM). Dilute Russell vipor venom time was prolonged at 34 seconds (normal 0 to 27 seconds). The addition of platelet extracts to a platelet neutralization procedure corrected the dilute Russell vipor venom time. Direct Coombs' test remained strongly positive. Antinuclear antibody titer was 1:160 with a diffuse pattern. Results of native DNA, ribonucleoprotein, Smith, and thyroid microsomal antibody assays were negative. Rheumatoid factor, follicle stimulating hormone, growth hormone, and somatomedin C levels were all within normal limits. The patient improved rapidly with treatment with corticosteroids. She was discharged on day 5 with instructions to take prednisone and fiudrocorti.sone. Long-term anticoagulation therapy with warfarin was recommended but the patient declined. Treatment was initiated with daily low-dose aspirin. At 1 month follow-up the patient tested negative for antiadrenal antibodies. Repeat direct Coombs' test remained positive, although her anemia had resolved. Computerized tomography of the adrenal glands 3 months after the initial scan showed a marked decrease in the size of both adrenal glands; low attenuation continued, but no evidence of old hemorrhage was seen (Figure 2). Immunoglobulin G ACA levels drawn at a 3 month follow-up visit remained elevated at 24 U.
Discussion
Antiphospholipid antibody syndrome gained widespread medical attention after it was identified in a
42
subclass of patients with systemic lupus erythematosus who suffered frequent thrombotic events. Further studies revealed a heterogenous class of aPL in the sera of systemic lupus erythematosus and nonsystemic lupus erythematosus patients, consisting of either lupus anticoagulant or anticardiolipin antibody.6 Antiphospholipid antibody syndrome has been associated with numerous hematologic disorders, including venous or arterial thrombosis and thrombocytopenia, although AIHA and neutropenia have been described less frequently.l Autoimmune hemolytic anemia is characterized by the destruction of red blood cells by the production of antierythrocyte autoantibodies. A European study reported a 4.3% incidence of a positive direct Coombs' test in patients with primary APS.2 The association of AIHA with aPL has been characterized in detai1. 7 - lo Interestingly, aPL appear to be involved intimately in the pathogenesis of AIHA by acting as anti-red blood cell autoantibodies. Cabral et al 7 demonstrated the specific binding ofIgM anticardiolipin antibodies to erythrocytes in a patient with primary APS. Similarly, Del Papa et alB showed that IgG anticardiolipin antibodies, eluted from erythrocytes of a patient with AIHA, displayed significant cardiolipin binding activity. More recently, Guzman et al 9 reported that a significant number of patients with AIHA have aPL. These findings contribute to the hypothesis that aPL may playa direct role in the hemolytic process. In this patient, symptoms of cholelithiasis were the initial manifestation of her illness. The gallstones likely were the result of persistent AIHA, as supported by the finding of pigmented bilirubin stones. In vitro studies demon-
Figure 2. Follow-up abdominal computerized tomography scan 3 months later, showing a markedly decreased size of both adrenal glands (black arrows) compared with the initial scan with continued low attenuation, suggestive of old adrenal infarction. July 1997 Volume 314 Number 1
Hsu, Udden, and Lynch
strated IgG coating but no evidence of C3 coating on the patient's red blood cells, consistent with the earlier findings of Del Papa8 and arguing against an immune complex-mediated process. Primary adrenal insufficiency is a rare complication of APS.4 Characterized by a marked predilection to intravascular thrombosis, antiphospholipid antibody syndrome appears to he associated with an increased risk of adrenal infar«.tion. The anatomic structure of the single adrenal vein with poor blood flow may predispose a patient to thrombosis, leading to adrenal infarction and adrenal failure. This association has been described in numerous cases and is supported by histologic findings. 4 Another presumed mechanism by which AI occurs in these patients is the development of bilateral adrenal hemorrhage. In these cases acute hemorrhage usually occurred after the patient had either undergone surgery or had received anticoagulant therapy for recurrent thromboembolism. 4 ,1l,12 This was not the scenario in this patient. At initial admission, computerized tomography demonstrated large, swollen adrenal glands bilaterally, but there was no evidence of hemorrhage. However, the patient did not exhibit clinical signs or symptoms suggestive of AI. Only 2 months later, after the postsurgical stress and the withdrawal of corticosteroids for AIHA, did the patient abruptly develop overt manifestations of adrenal failure. The significance of the adrenal gland enlargement seen on a computerized tomogrl;lphy scan 2 months before the acute illness is unclear, although it is possible that the patient had a chronic autoimmune process affecting her adrenal glands. Antiadrenal antibodies were not found in this patient; however, only two thirds of patients with autoimmune AI have these antibodies. 13 If the patient had AI during the first hospital stay, the disease was treated properly with corticosteroids, albeit for a different medical diagnosis, hemolytic anemia. Repeat computerized tomography showed atrophy of both adrenal glands without any swelling or hemorrhage, suggesting that the mechanism of adrenal failure was caused by thrombosis with consequent adrenal infarction. Clinically, acute AI is difficult to recognize, often because these patients are ill with other diseases. However, early recognition is crucial because the mortality rate remains high. Until recently, the management of patients with APS outside the setting of pregnancy remained controversial because ofthe lack of controlled treatment trials. 14 Khamashta et al,15 in a 1995 large retrospective study, compared antithrombotic treatments in-
THE AMERICAN JOURNAL OF THE MEDICAL SCIENCES
volving high-intensity warfarin (INR > 3), low-intensity warfarin (INR < 3), low-dose aspirin, and no therapy for patients with a history of thrombosis. The authors concluded that lifelong high-intensity warfarin offered the best protection for patients with APS in preventing recurrent thrombosis. No guidelines exist in the treatment of patients who have suffered adrenal infarction because of APS. In summary, this patient exhibited the tetrad of venous thrombosis, recurrent miscarriages, Coombs' positive AIHA, and AI as part of a unique clinical manifestation of APS. References 1. Lockshin MD. Antiphospholipid antibody syndrome. Rheum Dis Clin North Ani. 1994;20(1):45-59. 2. Vianna JL, Khamashta MA, Ordi-Ros J, Font J, Cervera R, Lopez-Soto A, et aI. Comparison of the primary and secondary antiphospholipid syndrome: A European multicenter study of 114 patients. Ani J Merl.. 1994;96:3-9. 3. Alarcon-Segovia D, Cabral AR. Antiphospholipid antibodies. Where do they come from? Where do they go? J Rheumatol. 1994;21(6):982-9. 4. Amason JA, Graziano FM. Adrenal insufficiency in the antiphospholipid antibody syndrome. Semin Arthritis Rheum. 1995;25(2):109-16. 5. Tan EM, Cohen AS, Fries JF, Masi AT, McShan DJ, Rothfield NF, et aI. The 1982 revised criteria for classification of systemic lupus erythematosus. Arthritis Rheum. 1982;25:1271-7. 6. Devine DV, Brigden ML. The Anti-phospholipid syndrome: When does the presence of anti phospholipid antibodies require therapy? Postgrad Med. 1996;99(6):105-25. 7. Cabral AR, Cabiedes J, Alarcon-Segovia D. Hemolytic anaemia related to an IgM autoantibody to phosphatidylcholine that binds in vitro to stored and to bromelain-treated human erythrocytes. J Autoimmun. 1990;3(6):773-87. 8. Del Papa N, Meroni PL, Barcellini W, Borghi MO, Fain C, Khamashta M, et aI. Antiphospholipid antibodies crossreacting with erythrocyte membranes. A case report. Clin Exp Rheumatol. 1992; 10:395-9. 9. Guzman J, Cabral AR, Cabiedes J, Pita-Ramirez L, Alarcon-Segovia D. Antiphospholipid antibodies in patients with idiopathic autoimmune haemolytic anemia. Autoimmunity. 1994; 18:51-6. 10. Sthoeger Z, Sthoeger D, Green L, Geitner D. The role of anticardiolipin autoantibodies in the pathogenesis of autoimmune hemolytic anemia in systemic lupus erythematosus. J Rheumatol. 1993;20(12):2058-61. 11. Papadopoulos KI, Jonsson A, Berntorp E, Tornquist C, Hulthen UL. Primary antiphospholipid syndrome associated with postoperative primary adrenal failure. J Intern Med.1995;238:175-8. 12. WaIz B, Kong HHP, Silver R. Adrenal failure and the primary antiphospholipid syndrome. J Rheumatol. 1990; 17(6): 836-7. 13. Muir A, Schatz DA, Maclaren, NK. Autoimmune Addison's disease. Springer Semin Immunopathol. 1993; 14:275-84. 14. Lockshin MD. Answers to the the anti phospholipid syndrome (editorial)? N Engl J Med. 1995;332(15):1025-7. 15. Khamashta MA, Cuadrado MJ, Mujic F, Taub NA, Hunt BJ, Hughes GRV. The management of thrombosis in the antiphospholipid syndrome. N Engl J Med. 1995;332(15): 993-7.
43