Received 9 October 1997; accepted 16 April 1998 factors important in the etiology of the syndrome include graft-versus-host disease (GVHD), cyclosporine, ...
Bone Marrow Transplantation, (1998) 22, 777–780 1998 Stockton Press All rights reserved 0268–3369/98 $12.00 http://www.stockton-press.co.uk/bmt
Red cell fragmentation (schistocytosis) after bone marrow transplantation A Zomas1, R Saso1, R Powles1, H Mackay1, S Singhal1,2, J Treleaven1 and J Mehta2 1
Leukaemia Unit, Royal Marsden Hospital, Surrey, UK
Summary: Red cell fragmentation is often the earliest sign of thrombotic microangiopathy. Days +14, +28 and +42 blood films from 58 allograft and 32 autograft recipients were reviewed blind to determine the incidence and severity of schistocytosis (the number of fragmented red cells per 1000 red cells expressed as a percentage). Schistocytosis was graded as mild (⬍1%), moderate (1– 1.9%) or severe (⭓2%). Schistocytes were seen in 99% of day 14 films (0.1–3.0%, median 0.4%), 97% of day 28 films (0.1–3.2%, median 0.4%), and 98% of day 42 films (0.1–4.3%, median 0.5%). Nine patients (10%) had severe fragmentation and 20 patients (22%) had moderate fragmentation at some time or the other. The difference in the extent of fragmentation between the days was not significant. Allogeneic BMT was associated with more extensive fragmentation than autologous transplantation on day 28 (P = 0.008) and day 42 (P = 0.02). Age, conditioning regimen and diagnosis had no influence. None of the patients followed-up for 6 months after transplant developed full-blown thrombotic microangiopathy. The occasional patient showing mild clinical and laboratory features of hemolysis responded well to adjustment of fluid balance and cyclosporine dose where applicable. Our data indicate that mild red cell fragmentation is a common morphologic finding after transplantation with no clinical significance in the absence of other clinical and laboratory findings suggestive of thrombotic microangiopathy, although patients with moderate or severe fragmentation should be monitored closely. Keywords: bone marrow transplantation; hemolyticuremic syndrome; microangiopathic hemolytic anemia; schistocytosis; thrombotic microangiopathy; thrombotic thrombocytopenic purpura
factors important in the etiology of the syndrome include graft-versus-host disease (GVHD), cyclosporine, tacrolimus, total-body irradiation (TBI), high-dose chemotherapy and infection.1–11 Thrombotic microangiopathy is usually diagnosed on the basis of falling platelet counts and hemoglobin, Coombsnegative hemolytic anemia, schistocytosis, elevated lactate dehydrogenase (LDH), renal impairment, hypertension, and central nervous system changes in varying combinations. LDH elevation is a sensitive indicator of hemolysis, but is not specific for hemolysis secondary to microangiopathic damage.12,13 High levels of LDH are also seen with liver damage which is common in the first few weeks after BMT.14,15 Similarly, most other laboratory and clinical abnormalities seen with thrombotic microangiopathy are not specific to thrombotic microangiopathy. Schistocytosis, however, is relatively specific for microangiopathic red cell damage, and confirms microangiopathic changes when one or more of the other abnormalities are seen concomitantly. Conditioning regimens for hematopoietic stem cell transplantation cause profound alterations of normal body physiology and extensive organ impairment. These could potentially result in microangiopathic damage in a large proportion of patients. The clinical scenario of marginally elevated bilirubin in the second and third weeks after BMT,15 often in the setting of evolving acute GVHD or cyclosporine levels that are towards the higher side of the therapeutic range, is common. One of the first steps taken at this stage is to review the blood smear for schistocytes. This study was undertaken to determine the incidence of schistocytosis by blind review of the peripheral blood smear in a series of patients undergoing autologous or allogeneic BMT for hematologic malignancies. Patients and methods
Severe microangiopathic changes after blood or marrow transplantation (BMT) can result in a clinical and laboratory picture resembling thrombotic thrombocytopenic purpura (TTP) or the hemolytic-uremic syndrome (HUS).1 The
Correspondence: Dr A Zomas at his current address: St Anargyri General Cancer Hospital, Kalyftaki, Nea Kifisia, 145 64 Athens, Greece 2 Current address: University of Arkansas for Medical Sciences, Little Rock, Arkansas, USA Received 9 October 1997; accepted 16 April 1998
The transplant population at the Leukaemia Unit of the Royal Marsden Hospital includes patients undergoing allogeneic BMT for a variety of hematologic malignancies (mainly leukemia) and autologous BMT for leukemia. Ninety patients transplanted in the Leukaemia Unit between 1994 and 1995 were included in the study. Patients autografted for multiple myeloma (conditioned exclusively with 200 mg/m2 melphalan) or lymphoma (conditioned exclusively with BEAM) were not included in this study. Table 1 shows the patient characteristics. Giemsa-stained peripheral blood smears from day +14,
Schistocytosis after BMT A Zomas et al
778
Table 1
Patient characteristics
Number
90
Age
6–52 years (median 32)
Diagnosis Acute myeloid leukemia Acute lymphoblastic leukemia Acute biphenotypic/undifferentiated leukemia Chronic myeloid leukemia Otherb Type of transplant Allograft Matched sibling Mismatched family Matched unrelated Autograft
46 (51%) 19 (21%) 3 (3%) 10 (11%) 12 (13%) 58 (64%) 49 2 7 32 (36%)a
Source of cells Marrow Blood and marrow
86 (96%) 4 (4%)
Conditioning regimen Without TBI With TBI
17 (19%) 73 (81%)
Total body irradiation dose 950 cGy 1050 cGy ⬎1050 cGy
14 (19%)c 48 (66%) 11 (15%)
The combination of cyclosporine (as a 12-h intravenous infusion once a day, or in a twice-daily oral schedule) and methotrexate was used for GVHD prophylaxis in all but five of the allograft recipients. Trough cyclosporine levels (measured by high-performance liquid chromatography) were maintained between 150 and 300 ng/ml. Five allograft recipients (three from mismatched family and two from matched unrelated donors) received marrow that was depleted of T cells using the murine anti-CD52 monoclonal antibody Campath-1G (kindly provided by Drs G Hale and H Waldmann, Cambridge, UK). Acute GVHD was treated with methylprednisolone on clinical suspicion when it developed (ie even grade I), and biopsy confirmation of diagnosis was obtained only in doubtful cases. Patients received irradiated packed red cells to maintain the hemoglobin at 100 g/l and irradiated random platelets to maintain the platelet count at 20 × 109/l. The blood product transfusion policy for ABO-mismatched allogeneic transplants took into account persistent host isohemagglutinins and new isohemagglutinins being formed by the donor immune system.20 Thrombotic microangiopathy was diagnosed on the basis of falling hemoglobin and platelet count, increasing renal dysfunction, hypertension and central nervous system abnormalities such as mental status changes or seizures in the presence of schistocytosis and elevated LDH.21
a
Includes one syngeneic transplant. Includes myelodysplastic syndrome, non-Hodgkin lymphoma, multiple myeloma, and idiopathic myelofibrosis. c One patient with myelodysplasia secondary to Fanconi syndrome had ⬍900 cGy radiation. The others had 950 cGy. b
day +28 and day +42 for these 90 patients were examined blind by one physician (AZ) who was not aware of the identities of the patients – and thus of the diagnosis and therapy. The number of fragmented red cells (triangle-, helmet- or crescent-shaped) were enumerated per 1000 red cells (expressed as a percentage), and the film was examined for other morphologic features suggestive of hemolysis such as spherocytosis and polychromasia. Although liver function tests were usually performed daily, these comprised bilirubin, transaminases, and alkaline phosphatase estimation. Lactate dehydrogenase (LDH) was not routinely measured. Coagulation parameters were monitored regularly (usually once or twice a week). The various conditioning regimens used have been described before.16–18 In patients receiving total-body irradiation (TBI) as part of the conditioning regimen, single-fraction (950 or 1050 cGy) or fractionated (200 cGy twice daily for 3 days) TBI was delivered from opposed 60Co sources at a low rate (4 cGy/min). The actual regimens used were: 110 mg/m2 melphalan with 950 cGy TBI for allografting, 120 mg/kg cyclophosphamide with 1200 cGy TBI for allografting, 140 mg/m2 melphalan with 1050 cGy TBI for autografting, and 16 mg/kg busulfan with 120 mg/kg cyclophosphamide for autografting. The mismatch transplant recipients were conditioned with a complex conditioning regimen containing total lymphoid irradiation (900 cGy in five fractions), cyclophosphamide (60 mg/kg), melphalan (110 mg/m2), and TBI (1200 cGy in 6 fractions).19
Results None of the 90 patients in this series, who were followedup for at least 6 months, developed the full-fledged syndrome of thrombotic microangiopathy (thrombocytopenia, anemia, renal dysfunction, hypertension, etc). However, as Table 2 shows, schistocytosis was seen to a variable extent in the majority of blood films examined, and there was no significant difference in either incidence (number of patients showing schistocytosis) or severity (amount of schistocytosis) between the 3 days examined. Most patients had mild schistocytosis, and although the proportion of patients with severe schistocytosis increased from days 14 to 28 to 42 (Table 3), the difference was not statistically significant. No abnormalities suggestive of disseminated intravascular coagulation were seen in any patient. Schistocytosis was more marked in patients who had undergone allogeneic transplantation than autograft recipients 28 and 42 days post-transplant, although there was no difference on day 14 (Table 4). All seven patients with severe schistocytosis on day 42 were allograft recipients (P = 0.025), whereas only two of five patients with severe schistocytosis on day 28 were allograft recipients (P = NS). The conditioning regimen, age, diagnosis, or development of GVHD (70% of the allograft recipients clinically; grades III–IV in 25%) did not affect the severity of schistocytosis. The frequency of schistocytosis was not affected by any of these factors since most patients had a variable degree of schistocytosis (data not shown). There was no consistent relationship between the cyclosporine levels and schistocytosis. A number of patients with severe schistocytosis had trough cyclosporine levels which
Schistocytosis after BMT A Zomas et al
Table 2
779
Incidence and extent of schistocytosis (comparison: t-test)
Day after transplant
No. of patients with schistocytosis
14 28 42
Extent of schistocytosis (%)
89 of 90 (99%) 85 of 88 (97%) 78 of 80 (98%)
Range
Mean (s.e.)
Median
0.1–3.0 0.1–3.2 0.1–4.3
0.5 (0.05) 0.6 (0.07) 0.7 (0.09)
0.4 0.4 0.5
Day 14 vs Day 28: P = 0.39. Day 28 vs Day 42: P = 0.41. Day 14 vs Day 42: P = 0.11.
Table 3
Extent of schistocytosis (P = 0.6, 2 test)
Day after transplant
14 28 42
No. of patients (%) None
Mild
Moderate
Severe
1 (1%) 3 (3%) 2 (3%)
77 (86%) 71 (81%) 62 (78%)
10 (11%) 9 (10%) 9 (11%)
2 (2%) 5 (6%) 7 (9%)
Table 4 Mean (s.e.) per cent schistocytes in relation to patient characteristics and day after transplant (t-test) %
Day 14
Day 28
Day 42
Autografts Allografts P
0.4 (0.06) 0.5 (0.06) 0.22
0.4 (0.08) 0.7 (0.1) 0.008
0.5 (0.08) 0.8 (0.1) 0.02
TBI No TBI P
0.5 (0.05) 0.6 (0.09) 0.30
0.6 (0.08) 0.9 (0.2) 0.10
0.7 (0.1) 0.9 (0.2) 0.27
Acute leukemia Other P
0.5 (0.05) 0.6 (0.09) 0.81
0.6 (0.07) 0.7 (0.2) 0.39
0.7 (0.09) 0.8 (0.2) 0.68
⬍32 years ⬎32 years P
0.5 (0.07) 0.5 (0.05) 0.96
0.6 (0.09) 0.6 (0.1) 0.77
0.8 (0.1) 0.7 (0.1) 0.35
were well within the therapeutic range, and a number of patients with high levels had minimal schistocytosis. Severe schistocytosis seen in allograft recipients on cyclosporine resolved with adjustment in the cyclosporine dose and hydration. Discussion Thrombotic microangiopathy is a well-recognized disorder which occurs in up to 15% of patients following bone marrow transplantation.4,7 This study shows that morphologic detection of schistocytosis is almost universal in the first 6 weeks after bone marrow transplantation. However, the presence of schistocytosis is not necessarily indicative of clinically significant thrombotic microangiopathy. Although the etiology of this morphologic alteration is
likely to be multifactorial, the most likely contributory cause is probably the cytoreductive preparative regimen. More severe schistocytosis after allogeneic transplants than autologous in the fourth and sixth weeks suggests that clinical or subclinical GVHD and cyclosporine therapy probably contribute to this. It has been proposed that post-transplant thrombotic microangiopathy be divided into four subgroups: multifactorial fulminant thrombotic microangiopathy, HUS associated with the conditioning regimen, cyclosporine-induced nephrotoxicity with microangiopathic hemolytic anemia, and cyclosporine-associated neurotoxicity with microangiopathic hemolytic anemia. Although none of our patients developed the full-blown syndrome of thrombotic microangiopathy, some patients could have belonged to the second or third categories. The absence of severe thrombotic microangiopathy in our group of patients is somewhat surprising, but may be partially attributable to the fact that particular attention was paid to cyclosporine levels and hydration whenever schistocytes were detected. We certainly have seen severe thrombotic microangiopathy before and after the study period (unpublished observations). Thrombocytopenia is common in the first several weeks after transplantation, and its etiology is multifactorial. Falling platelet counts in some of the patients may have been due to ‘subclinical’ thrombotic microangiopathy – but the absence of regular LDH measurements does not permit any conclusions to be drawn. Holler et al4 found morphological and biochemical changes indicative of generalized endothelial damage in 49 of 66 allograft recipients on cyclosporine, but not in 11 patients treated with methotrexate for GVHD prophylaxis. Severe microangiopathy was observed in 10 patients, and was fatal in seven. There was an association between microangiopathy and severe acute GVHD and the use of cyclosporine for GVHD prophylaxis. Zeigler et al11 devised a system of grading thrombotic microangiopathy in transplant recipients based upon LDH and the amount of schistocytosis. Our patients cannot be categorized by this classification because LDH levels were not routinely performed, and the primary aim of our study was to determine the incidence of schistocytosis rather than investigate thrombotic microangiopathy. However, based upon the percentage of fragmented red cells, the most severe grade of thrombotic microangiopathy in our patients would have been grade 2 (and the majority would have been grade 0 or 1).
Schistocytosis after BMT A Zomas et al
780
Since this index11 was based mainly upon patients with known thrombotic microangiopathy, it needs to be evaluated prospectively in a series of consecutive patients. It would have been useful to evaluate blood smears from normal controls (perhaps the marrow donors) concurrently in a blinded fashion. Unfortunately, this was not done. However, we do not believe this to be a major limitation because, while minor schistocytosis in an otherwise normal person may not even be noticed, smears in transplanted patients are frequently examined specifically for schistocytosis. Therefore, irrespective of what may be seen in normal controls, our findings retain their practical significance. The frequency of schistocytosis after transplantation means that this is not a good indicator of clinically significant endothelial damage either. The diagnosis of thrombotic microangiopathy after transplantation should therefore be based upon a combination of features including schistocytosis, elevated LDH, thrombocytopenia, anemia, renal impairment and hypertension in an appropriate clinical setting. We conclude that mild red cell fragmentation is a common morphologic finding after transplantation. Its presence is not clinically significant in the absence of other clinical and laboratory findings suggestive of thrombotic microangiopathy. However, patients with moderate or severe fragmentation should be monitored closely for the development of thrombotic microangiopathy. References 1 Pettitt AR, Clark RE. Thrombotic microangiopathy following bone marrow transplantation. Bone Marrow Transplant 1994; 14: 495–504. 2 Marshall RJ, Sweny P. Haemolytic-uraemic syndrome in recipients of bone marrow transplants not treated with cyclosporin A. Histopathol 1986; 10: 953–962. 3 Chappell ME, Keeling DM, Prentice HG, Sweny P. Haemolytic uraemic syndrome after bone marrow transplantation: an adverse effect of total body irradiation? Bone Marrow Transplant 1988; 3: 339–347. 4 Holler E, Kolb HJ, Hiller E et al. Microangiopathy in patients on cyclosporine prophylaxis who developed acute graftversus-host disease after HLA-identical bone marrow transplantation. Blood 1989; 73: 2018–2024. 5 Cohen H, Bull HA, Seddon A et al. Vascular endothelial cell function and ultrastructure in thrombotic microangiopathy following allogeneic bone marrow transplantation. Eur J Haematol 1989; 43: 207–214. 6 Loomis LJ, Aronson AJ, Rudinsky R, Spargo BH. Hemolytic uremic syndrome following bone marrow transplantation: a case report and review of the literature. Am J Kidney Dis 1989; 14: 324–328.
7 Rabinowe SN, Soiffer RJ, Tarbell NJ et al. Hemolytic-uremic syndrome following bone marrow transplantation in adults for hematologic malignancies. Blood 1991; 77: 1837–1844. 8 Juckett M, Perry EH, Daniels BS, Weisdorf DJ. Hemolytic uremic syndrome following bone marrow transplantation. Bone Marrow Transplant 1991; 7: 405–409. 9 Carlson K, Smedmyr B, Hagberg H et al. Haemolytic uraemic syndrome and renal dysfunction following BEAC (BCNU, etoposide, ara-C, cyclophosphamide) ⫾ TBI and autologous BMT for malignant lymphomas. Bone Marrow Transplant 1993; 11: 205–208. 10 Oursler DP, Holley KE, Wagoner RD. Hemolytic uremic syndrome after bone marrow transplantation without total body irradiation. Am J Nephrol 1993; 13: 167–170. 11 Zeigler ZR, Shadduck RK, Nemunaitis J et al. Bone marrow transplant-associated thrombotic microangiopathy: a case series. Bone Marrow Transplant 1995; 15: 247–253. 12 Gomez S, Nagler A, Naparstek E, Slavin S. Transient elevation of serum lactic dehydrogenase following autologous bone marrow transplantation. Bone Marrow Transplant 1991; 7: 487–488. 13 Sierra J, Conde E, Iriondo A et al. Frozen vs. nonfrozen bone marrow for autologous transplantation in lymphomas: a report from the Spanish GEL/TAMO Cooperative Group. Ann Hematol 1993; 67: 111–114. 14 McDonald GB, Shulman HM, Sullivan KM, Spencer GD. Intestinal and hepatic complications of human bone marrow transplantation. Gastroenterol 1986; 90: 460–477, 770–784. 15 Mehta J, Powles R, Horton C et al. The relationship between donor–recipient blood group incompatibility and serum bilirubin after allogeneic bone marrow transplantation from HLAidentical siblings. Bone Marrow Transplant 1995; 15: 853– 858. 16 Mehta J, Powles RL, Mitchell P et al. Graft failure after bone marrow transplantation from unrelated donors using busulphan and cyclophosphamide for conditioning. Bone Marrow Transplant 1994; 13: 583–587. 17 Powles R, Mehta J, Singhal S et al. Autologous bone marrow or peripheral blood stem cell transplantation followed by maintenance chemotherapy for adult acute lymphoblastic leukemia in first remission: 50 cases from a single center. Bone Marrow Transplant 1995; 16: 241–247. 18 Mehta J, Powles R, Singhal S et al. Autologous bone marrow transplantation for acute myeloid leukemia in first remission: identification of modifiable prognostic factors. Bone Marrow Transplant 1995; 16: 499–506. 19 Mehta J, Powles R, Horton C et al. Bone marrow transplantation for primary refractory acute leukaemia. Bone Marrow Transplant 1994; 14: 415–418. 20 Mehta J, Powles R, Singhal S et al. Transfusion requirements after bone marrow transplantation from HLA-identical siblings: effects of donor–recipient ABO incompatibility. Bone Marrow Transplant 1996; 18: 151–156. 21 Hamblin M, Powles R, Treleaven J et al. Defibrotide for refractory thrombotic thrombocytopenic purpura (TTP) after bone marrow transplantation. Blood 1996; 88 (Suppl. 1): 60b.
