Hematology, October/December 2004 Vol. 9 (5/6), pp. 433–436
Use of Intravenous Immunoglobulin and Intravenous Methylprednisolone in Hyperhaemolysis Syndrome in Sickle Cell Disease NAY WINa,*, TULLIE YEGHENb, MALCOLM NEEDSa, FREDERICK E. CHENc and IHEANYI OKPALAc a Red Cell Immunohaematology, National Blood Service—Tooting Centre, 75 Cranmer Terrace, London SW17 ORB, UK; bDepartment of Haematology, Whipps Cross Hospital, Whipps Cross Road, London E11 1NR, UK; cDepartment of Haematology, St Thomas’ Hospital, Lambeth Palace Road, Lambeth SE1 7EH, UK
(Received 4 May 2004; In final form 13 June 2004)
Hyperhaemolysis syndrome (HS), a syndrome in which there is destruction of both donor and recipient red cells after transfusion, is well recognised in patients with sickle cell disease and b-thalassaemia. It has also been reported in a patient with myelofibrosis. In acute forms of HS, evidence of red cell antibody-mediated haemolysis is lacking, and it has been proposed that the transfused and the patient’s own red blood cells were destroyed by hyperactive macrophages. Continuation of transfusion may be lethal as this can further exacerbate haemolysis. We report two cases of HS successfully treated with IVIg and IV methylprednisolone. The cessation of haemolysis following administration of IVIg and IV methylprednisolone supports the view that hyperactive macrophages contribute to the RBC destruction. IVIg and methylprednisolone appear to have a synergistic effect on suppressing the activity of macrophages. Keywords: Hyperhaemolysis syndrome; IVIg; Macrophages; IV methylprednisolone
INTRODUCTION Hyperhaemolysis syndrome (HS) in patients with sickle cell disease (SCD) is characterised by destruction of both donor and recipient erythrocytes after transfusion of crossmatch-compatible blood with significant reduction in reticulocyte count [1,2]. Awareness of this condition is important because further blood transfusion may exacerbate haemolysis, or be fatal [3]. A case of fatal HS in a non SCD patient (a patient with myelofibrosis) has been published recently [4]. Transfusion should be *Corresponding author. E-mail:
[email protected] ISSN 1024-5332 print/ISSN 1607-8454 online q 2004 Taylor & Francis Ltd DOI: 10.1080/10245330400001926
withheld and oral steroids have been used as a treatment in life-threatening haemolytic episode [1]. In life-threatening symptomatic severe haemolysis, further transfusion with intravenous immunoglobulin (IVIg) and IV steroid have been tried successfully [2,5]. We report two cases of life-threatening HS in SCD patients after the transfusion of crossmatchcompatible blood. Both patients were successfully treated with IVIg and IV methylprednisolone. Rapid rise in Hb was achieved in one case without blood transfusion. In another case further transfusion was uneventful. Pathogenesis of HS in SCD and possible mechanism of actions of IVIg, IV methylprednisolone to halt haemolysis is discussed. Case 1 The patient was a 36-year-old man with sickle cell anaemia, who was admitted with fever and right shoulder pain. He was treated with analgesics, IV fluid and antibiotics. He has also had deep vein thrombosis, and was started on anticoagulants. His blood type was group O Ro (cDe). The absolute reticulocyte count on admission was 597 £ 109 =l (normal range 50 – 100 £ 109 =l). The Hb level 2 days after admission was 5.8 g/dl. He had a history of multiple transfusions. From previous investigations, the patient was known to have anti-C, anti-E, antiFya and anti-Jkb. On this admission Anti-C, anti-E and anti-Fya only were identified in the patient’s serum. The DAT was negative. He was given four units of crossmatch-compatible blood lacking the C,
434
N. WIN et al.
E, Fya, Jkb and K antigens. The discharge Hb was 8.1 g/dl and the serum bilirubin level was18 mmol/l (normal range 5 – 21 mmol/l). The patient was re-admitted 3 days later with a fever, cough. He also noted passing dark-coloured urine. Chest X-ray confirmed right middle lobe consolidation. He was treated with IV antibiotics. The Hb level was 5.6 g/dl. The dose of warfarin was increased because the INR was sub-therapeutic. The DAT on readmission was positive with IgG only. The eluate was non reactive, and no new red cell alloantibodies were identified in the serum. Two days after the re-admission the Hb dropped to 3.8 g/dl, and the absolute reticulocyte count dropped to 358:6 £ 109 =l: The bilirubin level had increased to 41 mmol/l, and the patient continued to pass dark-coloured urine. Abdominal ultrasound showed no evidence of internal bleeding. Repeated tests for occult blood in stool were negative. The patient received IVIg ð0:4 g=kg=day £ 5 daysÞ and IV methyl-prednisolone (0.5 g/day £ 2 days). Transfusion was commenced 4 h after infusion of the first dose of IVIg and methylprednisolone. Four units of crossmatch-compatible blood, negative for C, E, Fya, Jkb and K antigens, were transfused. Post-transfusion Hb was 7.5 g/dl. The transfusion was uneventful. A day after, the absolute reticulocyte count increased to 552:6 £ 109 =l: The Hb level remained stable, and subsequent antibody screening revealed no new antibodies at 1 and 8 weeks after the admission. Further laboratory investigations showed the presence of HLA antibodies, with the specificity of HLA-A2 and HLA-A28 in his serum. Case 2 The patient was a 36-year-old woman with sickle cell anaemia admitted to the hospital with infected chronic leg ulcer. She was started on oral antibiotics. To facilitate healing of the intractable leg ulcer, it was decided to raise the Hb level around 10 g/dl by blood transfusions. Her blood type was group O, Ro and negative for K, N, S, Fya, Fyb and Jkb antigens. The DAT was negative. Clinically insignificant antiKna antibody was identified in her serum. Anti-Kna belongs to a group of low avidity antibodies, which reacts with most of the antibody-identification cells. With an initial Hb level of 7.2 g/dl, she was given 3 units of RBCs over the next 2 days. The transfused units were crossmatch-compatible and negative for C, E and K antigens. Hb was 8.8 g/dl 3 days after the transfusion. The next day she developed fever, bone pains, haemoglobinuria and left basal pneumonia. The Hb level had dropped to 7.8 g/dl and was treated with analgesics, IV fluids, cefuroxime and flucloxacillin. There were continuing signs of haemolysis. Hb dropped to 5.5 and 4.9 g/dl in the morning and afternoon respectively the next day, and serum bilirubin rose to 125 mmol/l. The absolute
FIGURE 1 Initial blood transfusion and changes in Hb, absolute reticulocyte count after treatment with IVIg and IV methlprednisolone.
reticulocyte count was 132 £ 109/l (normal range 40 – 130 £ 10 9/l). High performance liquid chromatography (HPLC) of her blood sample showed HbS 52%, HbA 40% and HbF 8%. The rapid haemolysis and Hb level below the pre-transfusion value strongly suggested the destruction of both donor and patient’s own red cells. A working diagnosis of HS was made. Treatment began with IVIg (0.4 g/kg/day £ 5 days) and high dose methylprednisolone (500 mg IV on days 1 and 2, 1 g on day 4). The Hb level the next day was 3.5 g/dl. Serological evaluation of the post-transfusion sample revealed a strong pan-reacting antibody with underdetermined specificity. The samples were forwarded to the International Blood Group Reference Laboratory (IBGRL) for further investigation. New red cell alloantibodies Anti-S and anti-Jkb were identified in the post-transfusion sample. Anti-Kna was also confirmed. The DAT was positive for IgG only. There was insufficient sample for red cell eluate study. There was an immediate response to IVIg and methylprednisolone therapy (Fig. 1). The patient improved symptomatically, the Hb level rose gradual and therefore transfusion was avoided. The absolute reticulocyte count rose to 424 £ 109/l, Hb to 7.3 g/dl, and serum bilirubin dropped to 26 mmol/l. HLA antibodies, with no specificity, were also identified in the patient’s sample.
METHODS Automated reticulocyte counting (Flow Cytometer System/ADVIA) was used in Case 1: absolute reticulocyte count reference range (50 – 100 £ 109/l). Manual reticulocyte counting was used in Case 2. By use of a supravital staining technique (new methylene blue stain) a stained blood film was examined under microscopy and the reticulocyte percentage was calculated. With the reticulocyte
HYPERHAEMOLYSIS IN SCD
percentage and total number of RBCs, the absolute reticulocyte counts were calculated; (reference range 40 –139 £ 109/l). HPLC of blood sample in Case 2 was undertaken using an HPLC analyser and reagents (Variant, Bio-Rad Laboratories UK, Hemel Hempstead, UK) as described previously [6].
DISCUSSION HS is an atypical form of delayed haemolytic transfusion reaction (DHTR) and is well described in SCD patients. The term “syndrome” was introduced by Petz et al. [1] The manifestations include: post-transfusion Hb lower than the pretransfusion value, a fall in absolute reticulocyte count with recovery manifested by reticulocytosis and symptoms suggestive of painful sickle crisis. Hyperbilirubinemia, raised LDH, and haemoglobinuria are common findings. HS can be manifested as either acute or delayed forms. The acute form occurs 48– 72 h after transfusion. Usually, the DAT is negative and no red cell alloantibodies identified in both pre-transfusion and post-transfusion samples. In cases with preformed alloantibodies, transfusion of antigen-negative crossmatch-compatible blood does not prevent HS. Serological investigation of post-transfusion and follow-up blood samples may not reveal formation of new red cell alloantibodies [2,5,7]. Delayed HS occurs 7 –10 days after transfusion. New alloantibodies are identified in post-transfusion samples, and the DAT is commonly positive [1,8]. Case one represented an acute form, as additional alloantibodies were not identified. Case two represented a delayed form, as new alloantibodies, anti-S and anti-Jkb, were formed. The exact pathogenesis of HS is not well understood. Petz et al. [1], who documented reticulocytopenia, have proposed the erythropoiesis suppression theory, and recommended to withhold transfusion in non life-threatening haemolysis and to prescribe corticosteroids. Steroid therapy seems beneficial in HS, but usually gave a delayed response. Response to steroids suggest another element may be involved in the pathogenesis of HS, rather than the marrow suppression itself. Awareness of HS is important as fatal cases of HS have been reported in both SCD and non SCD patients [3,4]. In life-threatening symptomatic severe haemolysis, further transfusion with IVIg and IV steroid have been tried successfully [2,5]. There were only three cases reported in the literature [2,5], of use of IVIg/steroids in HS in SCD patients. The mechanism of HS is poorly understood. In the acute form of HS, evidence of red cell antibodymediated haemolysis is lacking, therefore Win et al.[2] have proposed that the patient’s own RBCs and transfused cells are destroyed by hyperactive
435
macrophages. Evidence of destruction of patient’s own cells and transfused cells were documented by sequential examination of RBC Hb electrophoresis and HPLC analysis of the urine during haemolysis. Reticulocytopenia (a significant decrease from the patient’s usual absolute reticulocyte level) is one of the presenting features of HS. Win et al. have suggested that the reticulocytopenia is not due to erythropoiesis suppression, but rather, the result of peripheral destruction by macrophages. That hypothesis was supported by the bone marrow finding of erythroid hyperplasia and the reticulocyte response to IVIg/steroids [2]. The drop in reticulocyte count during haemolysis and the rise in reticulocyte count after IVIg and methylprednisolone was also documented in our two cases (Fig. 1). Some of the red cells express HLA antigen and HLA antibody can cause acute and DHTR [9,10]. HLA antibody formation is common among SCD patients [11], and Win et al. [2] proposed that transfused red cells expressing HLA antigens are destroyed via HLA mediated immune mechanisms by hyperactive macrophages and those transfused cells not expressing HLA antigens are destroyed by “bystander mechanism”. HLA antibodies were identified in both case 1 and case 2. Belcher et al. [12] reported that monocytes are activated in SCD patients. Fever is a common presenting feature in HS, possibly signifying infection, which may activate macrophages and may trigger hyperhaemolysis. The exact pathogenesis of HS is poorly understood and complex as it involves destruction of transfused and autologous cells. A fatal case of acute severe haemolysis after transfusion has been reported in a patient with myelofibrosis. There was no evidence of red cell antibody-mediated haemolysis in that case as no free red cell antibodies were identified in both pre- and post-transfusion serum samples. The patient was receiving splenic irradiation and was suggested that the red cells were destroyed by the malfunctioning spleen [4]. This case report, further supports the proposed theory of haemolysis caused by hyperactive macrophages in SCD [2]. Recently, Jasinski et al. [13] have demonstrated a novel mechanism of complement-independent clearance of RBCs, deficient in glycosyl phosphatidylinositol (GPI)-linked proteins, in a mouse model Paroxysmal Nocturnal Haemoglobinuria. They have shown that the clearance of RBCs deficient in GPIlinked proteins is mediated by macrophages. Liposome encapsulated clodronate which depletes macrophages, effectively normalised the half-life of these RBCs. This study highlighted the mechanism of non red-cell antibody-mediated haemolysis by macrophages. We report two cases of life-threatening HS, both were successfully treated with IVIg and IV methylprednisolone. Further transfusion was only given in
436
N. WIN et al.
case one and was uneventful. Rapid rise in Hb without blood transfusion was achieved in the latter case (Fig. 1). There are several possible mechanisms of the action of IVIg and steroids on macrophages. IVIg contains a broad range of antiviral/antibacterial antibody specificity, which may neutralise infections, and may thus reduce the activation of macrophages [14]. IVIg might suppress hyperactive macrophages through immunomodulatory mechanism [15]. It may block the adhesion of sickle cells and reticulocytes to macrophages. In vitro studies in sickle cell mice demonstrates that IVIg prevents venular vasoocclusion by inhibiting leucocyte adhesion and the interactions between sickle erythrocytes and adherent leucocytes [16]. Steroids suppress macrophage activity [17]. The cessation of hemolysis following administration of IVIg and IV methylprednisolone in our two cases further supports the view that hyperactive macrophages contribute to red cell destruction in HS. Case two responded to IVIg and steroids therapy and further transfusion support was avoid in that case. Although the exact mechanism is not known, IVIg and methylprednisolone appear to have a synergistic effect in suppressing the activity of macrophages. Infusion of IVIg has been associated with renal toxicity [18,19] thrombosis [20] and estimated 0.6% risk of stroke [21]. Caution is necessary as patient’s with SCD may be predispose to such adverse events. Therefore, use of IVIg should be selective and reserved for those cases with acute severe life-threatening haemolysis.
References [1] Petz, L.D., Calhoun, L., Shulman, I.A., Johnson, C. and Herron, R.M. (1997) “The sickle cell haemolytic transfusion reaction syndrome”, Transfusion 37, 382 –392. [2] Win, N., Doughty, H., Telfer, P., Wild, B.J. and Pearson, T.C. (2001) “Hyperhemolytic transfusion reaction”, Transfusion 41, 323–328. [3] Friedman, D.F., Kim, H.C. and Manno, C.S. (1993) “Hyperhemolysis associated with red cell transfusion in sickle cell disease (abstract)”, Transfusion 33(Suppl.), 14S. [4] Treleaven, J.G. and Win, N. (2004) “Hyperhaemolysis syndrome in a patient with myelofibrosis”, Hematology 9, 147 –149. [5] Cullis, J.O., Dudley, J.M., Kaye, T. and Win, N. (1995) “Post-transfusion hyperhaemolysis in a patient with sickle cell disease: use of steroids and intravenous immunoglobulin to prevent further red cell destruction”, Vox Sanguinis 4, 355–357. [6] Wild, B.J. and Stephens, A.D. (1997) “The use of automated HPLC to detect and quantiate haemoglobins”, Clinical and Laboratory 19, 171– 176.
[7] King, K.E., Shirey, R.S., Lankiewicz, M.W., et al. (1997) “Delayed haemolytic transfusion reactions in sickle cell disease: simultaneous destruction of recipients’ red cells”, Transfusion 37, 376–381. [8] Aygun, B., Padmanabhan, S., Paley, C. and Chandrasekaran, V. (2002) “Clinical significance of RBC alloantibodies and autoantibodies in sickle cell patients who received transfusion”, Transfusion 42, 37–43. [9] Benson, K., Agosti, S.J., Latoni-Benedetti, G.E. and Leparc, G.F. (2003) “Acute and delayed hemolytic transfusion reactions secondary to HLA alloimmunization”, Transfusion 43, 753 –757. [10] Panzer, S., Matyr, W.R., Graninger, W., Puchler, K., Hocker, P. and Lechner, K. (1987) “Haemolytic transfusion reactions due to HLA antibodies. A prospective study combining red cell serology with investigations of Chromium-51-labelled red cell kinetics”, The Lancet 28, 474–478. [11] Friedman, D.F., Lukas, M.B., Jawad, A., Larson, P.J., OheneFrempong, K. and Manno, C.S. (1996) “Alloimmunization to platelets in heavily transfused patients with sickle cell disease”, Blood 88, 3216– 3222. [12] Belcher, J.D., Marker, P.H., Weber, J.P., Hebbel, R.P. and Vercellotti, G.M. (2000) “Activated monocytes in sickle cell disease potential role in the activation of vascular endothelium and vaso-occlusion”, Blood 96, 2451–2459. [13] Jasinski, M., Pantazopoulos, P., Rother, R.P., van Rooijen, N., Song, W.-C., Molina, H. and Bessler, M. (2004) “A novel mechanism of complement-independent clearance of red cells deficient in glycosyl phosphatidylinositol-linked proteins”, Blood 103(7), 2827–2834. [14] Krause, I., Wu, R., Sherer, Y., Patanik, M., Peter, J.B. and Shoenfeld, Y. (2002) “In vitro antiviral and antibacterial activity of commercial intravenous immunoglobulin preparations—a potential role for adjuvant intravenous immunoglobulin therapy in infectious disease”, Transfusion Medicine 12, 133 –139. [15] Rhoades, C.J., Williams, M.A., Kelsey, S.M. and Newland, A.C. (2000) “Monocyte-macrophage system as targets for immunomodulation by intravenous immunoglobulin”, Blood Reviews 14, 14 –30. [16] Turhan, A., Jenab, P., Bruhns, P., Ravetch, J.V., Coller, B.S. and Frenette, P.S. (2004) “Intravenous immune globulin prevents venula vaso-occlusion in sickle cell mice by inhibiting leukocyte adhesion and the interactions between sickle erythrocytes and adherent leukocytes”, Blood 15(103), 2397– 2399. [17] Packer, J.T., Greendyke, R.M. and Swisher, S.N. (1960) “The inhibition of erythrophagocytosis in vitro by corticosteroids”, Transfusion Association of American Physicians 73, 93–102. [18] Cantu, T.G., Hoehn-Saric, E.W., Burgess, K.M., Racusen, L. and Scheel, P.J. (1995) “Acute renal failure associated with immunoglobulin therapy”, American Journal of Kidney Diseases 25, 228 –234. [19] Gupta, N., Ahmed, I., Nissel-Horowitz, S., Patel, D. and Mehrotra, B. (2001) “Intravenous gammglobulin-associated acute renal failure”, American Journal of Hematology 66, 151 –152. [20] Go, R.S. and Call, T.G. (2000) “Deep venous thrombosis of the arm after intravenous immunoglobulin infusion: case report and literature review of intravenous immunoglobulinrelated thrombotic complications”, Mayo Clinic Proceedings 75, 83–85. [21] Caress, J.B., Cartwright, M.S., Donofrio, P.D. and Peacock, J.E., Jr. (2003) “The clinical features of 16 cases of stroke associated with administration of IVIg”, Neurology 60, 1822–1824.
