sign in sign up
<<
Home , Avascular necrosis

Transplantation, (1998) 22, 565–569  1998 Stockton Press All rights reserved 0268–3369/98 $12.00 http://www.stockton-press.co.uk/bmt Avascular of bone following allogeneic stem cell transplantation: MR screening and therapeutic options

A Wiesmann1, P Pereira2,PBo¨hm3, C Faul1, L Kanz1 and H Einsele1

1Department of Hematology and Oncology, 2Department of Diagnostic Radiology, 3Department of Orthopedics, University of Tu¨bingen, Germany

Summary: chondral bone, whereas osteonecrosis of the metaphyseal or diaphyseal bone is commonly referred to as bone infarct.1 With improvement in long-term survival after allo- There are several causes of osteonecrosis such as trauma, geneic stem cell transplantation, late complications with benign and malignant hematologic diseases, radiation, significant morbidity are of increasing importance. We , Gaucher’s disease, collagen disease and retrospectively analysed 272 recipients of an allogeneic disease.2–9 Treatment with is stem cell transplant for the development of osteo- another risk factor in the pathogenesis of osteonecrosis as necrosis. The incidence among allograft recipients was first mentioned in 1950.10 Therefore, is 6.3% (17/272) for the whole patient population, and a well-described complication after renal and cardiac trans- 11.8% (17/144) for long-term survivors. All patients plantation, especially in patients receiving high-dose ster- were treated with high-dose prednisolone, 16 for severe oids for graft rejection.10–17 In 1987, Atkinson et al18 first acute or extensive chronic graft-versus-host disease described a 10% incidence of osteonecrosis after allogeneic (GVHD) and one patient for graft rejection. The mean bone marrow transplantation from HLA-identical sibling age at time of diagnosis was 33 years (range 16–45) and donors. Since then, few studies have addressed this late the mean time from transplant to diagnosis of complication.19–23 osteonecrosis was 13 months. Osteonecrosis was diag- We retrospectively analyzed 17 out of 272 patients after nosed by magnetic resonance (MR) imaging, which allogeneic stem cell transplantation who developed an avas- allows early detection of osteonecrosis and assessment cular necrosis of the bone diagnosed by clinical symptoms of stage. At the time of diagnosis, eight patients had and confirmed by MR imaging. stage I, three patients stage II, three patients stage III and three patients stage IV osteonecrosis according to MR criteria. All but one patient had involvement of the Patients and methods . The median total dosage of prednisolone at the time of diagnosis was 189 mg/kg (single manifes- Patients tation 150 mg/kg; multiple manifestations 313 mg/kg) Seventeen (eight female, nine male) out of 272 patients with a total range of 13–555 mg/kg. Six patients were developed avascular necrosis after allogeneic stem cell treated by conservative means, 77 patients underwent transplantation (nine patients with CML, three AML, three surgery (three core decompression, eight ALL, two SAA). Eleven patients had undergone allogeneic replacement). MR screening of patients receiving high- transplantation from a matched sibling donor and six dose steroids might help to detect osteonecrosis at an patients from a matched unrelated donor between 1987 and early stage and thus prevent progression by early inter- 1996 (see Table 1). vention, for example, by core decompression. Keywords: avascular necrosis; allogeneic stem cell trans- plantation; magnetic resonance-steroids Conditioning regimens Eight patients received TBI (12 Gy) and cyclophosphamide 60 mg/kg once daily for 2 days (total dose 120 mg/kg), with two patients receiving an additional dose of VP-16 With improvement in long-term survival after allogeneic (40 mg/kg). Five patients received busulfan 4 mg/kg p.o. in bone marrow transplantation, late complications with sig- four divided doses daily for 4 days (total dose 16 mg/kg) nificant morbidity are of increasing importance. in addition to intravenous cyclophosphamide (120 mg/kg). Osteonecrosis is a general term applied to conditions One patient was treated with cyclophosphamide resulting in necrosis of bone and marrow elements. Avascu- (120 mg/kg) and ALG (Merieux, Leimen, Germany), and lar necrosis indicates involvement of the or sub- one patient with cyclophosphamide (4 × 50 mg/kg).

Supportive care Correspondence: Dr H Einsele, Department of Hematology and Oncology, University of Tu¨bingen, Otfried Mu¨ller Str. 10, 72076 Tu¨bingen, Germany Patients were admitted to the hematologic intensive care Received 9 March 1998; accepted 20 April 1998 unit and nursed in laminar air flow units until discharge. Avascular necrosis after allogeneic stem cell transplant A Wiesmann et al 566 Table 1 Characteristics of 17 patients developing avascular necrosis of bone after allogeneic stem cell transplantation.

Patient Sex Age Diagnosis Conditioning Donor GVHD GVHD Max. pred. Total pred. No. regimen prophylaxis stage (mg/kg/day) (mg/kg)

1 F 20 ALL TBI/VP-16/Cy related CsA II 4.4 58 2 M 16 SAA Cy/ALG related CsA (rejection) 6 222 3 M 19 ALL TBI/Cy related CsA II 10 328 4 M 33 CML TBI/Cy unrelated CsA/MTX III 23 281 5 F 19 SAA Cy unrelated CsA IV 27 408 6 M 41 CML Bu/Cy related CsA/MTX II 5 146 7 F 45 CML Bu/Cy unrelated CsA/MTX II 20 181 8 F 45 AML TBI/Cy related CsA/MTX II 2 197 9 F 33 CML Bu/Cy related CsA II 2 13 10 M 35 CML TBI/Cy unrelated CsA/MTX II 15 555 11 F 32 AML TBI/VP-16/Cy related CsA II 10 404 12 M 40 CML TBI/Cy unrelated CsA/MTX II 7.5 518 13 F 27 CML TBI/Cy related CsA II 5.0 285 14 M 21 ALL TBI/Cy related CsA II 2.5 120 15 M 22 AML Bu/Cy related CsA II 2 45 16 F 26 CML TBI/Cy unrelated CsA/MTX I 5 155 17 M 26 CML Bu/Cy related CsA/MTX II 2 44

Max. pred. = maximal prednisolone dose.

Table 2 Characteristics of avascular bone necrosis after allogeneic stem cell transplantation

Patient Day of Stage of AVN Follow-up Localisation No. of Therapy Follow-up No. diagnosis (progression) time of involved (months after of AVN AVN diagnosis) (months)

1 +135 I(III) 36 1 conservative alive 2 +284 I(IV) 24 hip 2 decompression (9) alive 3 +304 III/IV 84 hip 1 (36) alive 4 +340 III/IV 48 hip 1 arthroplasty (36) died, 4.5 years after Tx (pneumonia) 5 +436 III 84 hip, humerus 3 conservative alive 6 +402 II 12 hip 1 conservative alive 7 +95 II 15 hip 2 conservative alive 8 +345 III 12 hip 1 arthroplasty (5) alive 9 +289 III 9 hip 1 conservative alive 10 +865 I(II) 18 hip 2 decompression (3) alive 11 +419 I(III) 24 hip, humerus 3 decompression (6) alive 12 +813 II 36 hip 1 arthroplasty (2) died 5 years after Tx (sepsis) 13 +372 I(IV) 106 hip 1 arthroplasty (42) alive 14 +309 I(III) 66 hip, knee 3 conservative alive 15 +457 IV 12 hip 1 arthroplasty (12) alive 16 +783 I 10 hip 1 conservative died 3 years after Tx (sepsis) 17 +542 I(III) 24 hip 2 conservative alive

AVN = avascular necrosis; Tx = transplantation.

Supportive infusion therapy (parenteral feeding, vitamins) performed to 10 and finally 20 or more mg/kg/day. Taper- was administered through a tunneled Hickman catheter. ing of the steroid dosage was initiated when clinical signs Patients received prophylactic antibiotics, antifungal and and symptoms of acute GVHD improved. antiviral drugs as well as passive immunoglobulin prophy- laxis (0.25 mg/kg every 3 weeks).

Diagnosis and staging of osteonecrosis GVHD prophylaxis and therapy All patients received cyclosporin from day −1 until at least Symptomatic patients (pain, stiffness etc) were diagnosed day +100 or longer, eight patients additionally received a by conventional and MR imaging. Osteo- short course of methotrexate (15 mg/m2 day 1, 10 mg/m2 necrosis was classified according to MR criteria: stage I: days 3 and 6). Acute graft-versus-host disease grade II or non-homogenous loss in signal intensity in T1; II: wedge- more was treated with 2 mg/kg/day prednisolone. In non- shaped ; III: crescent sign sequestra/cortical responding patients, the dose of prednisolone was increased collapse; IV: degenerative changes with narrowed joint to 5 mg/kg/day. If necessary, a further dose escalation was space.24–27 Avascular necrosis after allogeneic stem cell transplant A Wiesmann et al 567 Results

Seventeen out of 272 recipients of an allogeneic marrow transplant developed avascular necrosis as a long-term complication. The incidence was 6.3% among all patients and 11.8% (17/144) for long-term survivors. All 17 patients received prednisolone treatment post-transplant; 16 patients for treatment of acute severe GVHD (one patient grade I, 13 patients grade II, one patient grade III and one patient grade IV) and one patient for graft rejection. The average maximum prednisolone dose was 13 mg/kg/day with a range of 2–27 mg/kg/day. At the time of diagnosis of osteonecrosis the average age was 33 years (range 16–45 years) and the mean time from transplant to manifestation of osteonecrosis was 13 months (Table 2). On the appear- ance of clinical symptoms such as pain and stiffness, only 8/17 patients were diagnosed as having stage I osteonecrosis according to MR criteria (Figure 1), all other patients had already progressed to more advanced stages: three patients to stage II, three to stage III and three to stage III/IV. All but one patient developed osteonecrosis of the femoral head, 10 patients had unilateral and four patients bilateral manifestations. Three patients showed involvement of three joints, two with bilateral disease of the femoral head and one of the humerus. A further patient had involvement of both knees and one humerus. The median total dose of prednisolone at the time of diagnosis was 189 mg/kg (single manifestation 150 mg/kg, multiple manifestations 313 mg/kg) with a total range of 13– 555 mg/kg. All patients were initially treated conservatively with reduced -bearing and analgesia. Core decompression was performed in three patients (stage I) and six patients required . All but one (patient No. 16) patient diagnosed in stage I and treated conservatively pro- gressed to stages III and IV after a median follow-up of 2 years (Figure 2). Two patients who had core decompression performed in stage I progressed to stages III and IV, whereas one patient undergoing core decompression remained in stage I/II during an 18 months follow-up period.

Discussion

Avascular necrosis of the bone has been recognised as a complication of systemic therapy since 1950.10 In 1987, Atkinson et al18 reported osteonecrosis for the first time following allogeneic bone marrow transplan- tation. Although few reports have addressed this compli- cation of allogeneic stem cell transplantation,19–23 improve- ment in the detection of this late complication with its attendant morbidity is of increasing importance. Figure 1 (a) Coronal T2 weighted image with fat saturation demonstrat- Early diagnosis of avascular necrosis appears essential ing typical bone marrow edema of the right femoral head (arrow) sugges- tive of epiphyseal necrosis in an early stage (stage I). (b) Plain radiograph to prevent progression. MR imaging allows the detection of the right hip at the same time as magnetic resonance imaging (a) show- of osteonecrosis while conventional radiographs still show ing normal appearance of bone structures without changes of bone no signs of osteonecrosis (Figure 1). Therefore, MR imag- . ing is clearly the technique of choice for early diagnosis and follow-up evaluations.24,28–31 Breitenseher et al32 showed that MR imaging achieved the best agreement with the definite diagnosis in distinguishing early avascular Avascular necrosis after allogeneic stem cell transplant A Wiesmann et al 568 In our study, all patients had received prednisolone as treatment of GVHD or graft rejection. The mean maximal dose of prednisolone was 3 mg/kg/day (range 2–27). The median total dose of prednisolone was 189 mg/kg with a range of 13–555 mg/kg. Socie et al22 reported a range of the cumulative steroid dose of 60–840 mg/kg in their patients with osteonecrosis, but others observed osteo- necrosis in a patient who had received a total dose of only 14 mg/kg.18 The mean age at diagnosis was 33 years (range 16–44) with only two patients younger than 20 years. These data are comparable to other series which demonstrated increas- ing age as an independent risk factor.19,22 We observed an incidence of osteonecrosis after allogeneic stem cell trans- plantation of 11.8% among long-term survivors, compara- ble to previous reports.18,19,22 No male predominance was observed in our study and only one report shows male gen- der as an independent risk factor.22 Even if 10 out of 17 patients (59%) showed multifocal manifestations it is noteworthy that the femoral head was involved in all but one of our patients. In contrast to Socie et al,22 there was a significant difference in median total dosage of prednisolone at the time of diagnosis between single and multiple manifestations (single manifestation, median prednisolone dose 150 mg/kg, mean prednisolone dose 132 mg/kg; multiple manifestation, median predniso- lone dose 313 mg/kg, mean prednisolone dose 333 mg/kg). However, there was no significant correlation between total dose and stage of osteonecrosis at the time of diagnosis (mean total prednisolone dose stage I and II, 240 mg/kg; mean total prednisolone dose stage III and IV, 212 mg/kg). Enright et al19 reported similar results. The mean time from transplant to diagnosis was 13 months (range day +95 to day +865), similar to all other studies.18–20,22 Thus, osteonecrosis seems to occur earlier after bone marrow transplantation than in patients treated with steroids for systemic erythematosus (3.5–5.5 years)2 or after cardiac transplantation (2.5 years).14–17 All our patients were initially treated conservatively. At follow-up, three patients underwent core decompression and eight required joint replacement. Unfortunately, all patients diagnosed at an early stage and treated conserva- tively experienced progression of osteonecrosis after a median time of 2 years. Two further patients undergoing core decompression at a more advanced stage showed pro- Figure 2 (a) Coronal T1 weighted image from both showing an gression soon afterwards. Only one patient presented with early stage necrosis (stage I) with hypointense lesions in epiphyseal and stable disease (stage I/II) continuing 18 months after core metaphyseal regions on both sides. (b) Coronal T1 weighted image from the same patient (9 weeks following (a)) with a bilateral wedge-shaped decompression. Controversial results are reported following crescent sign (arrows) demonstrating progression of the infarction on both core decompression, even in the early stages of avascular sides (stage II). necrosis, but many surgeons feel the technique may be effective if the diagnosis is established early.24,33,34 There are reports of complete resolution or stable disease in 40% necrosis from bone marrow edema compared to radiogra- of patients when treated in stages I and II, but also of pro- phy, CT and scintigraphy. Glickstein et al33 reported a gression of avascular necrosis in 60% of patients treated specificity of 98% and a sensitivity of 97% in differen- with core decompression.25,34,35 tiating avascular necrosis from normal hips. In contrast to The role of intertrochanteric for the treatment Atkinson et al,18 we were able to diagnose osteonecrosis of avascular necrosis stages II or III of the femoral head in eight patients (44%) at stage I with MR criteria while in patients after allogeneic stem cell transplantation is not radiographs remained normal. Sixty-five percent of our yet known. In patients with stage III disease with total hip patients were diagnosed prior to developing advanced dis- involvement or stage IV, total hip arthroplasty is the pro- ease (III–IV) compared to 50% reported by Enright et al.19 cedure of choice. Avascular necrosis after allogeneic stem cell transplant A Wiesmann et al 569 Further studies will hopefully show whether diagnosis 18 Atkinson K, Cohen M, Biggs J. Avascular necrosis of the fem- of avascular necrosis by MR imaging and subsequent core oral head secondary to corticosteroid therapy for graft-versus- decompression at an early stage is able to prevent disease host disease after marrow transplantation: effective therapy progression. with hip arthroplasty. Bone Marrow Transplant 1987; 2: 421–426. 19 Enright H, Haake R, Weisdorf D. Avascular necrosis of bone: a common serious complication of allogeneic bone marrow References transplantation. Am J Med 1990; 89: 733–738. 20 Mascarin M, Giavitto M, Zanazzo GA et al. Avascular necrosis of bone in children undergoing allogeneic bone mar- 1 Munk PL, Helms CA, Holt RG. Immature bone infarcts: find- row transplantation. Cancer 1991; 68: 655–659. ings of plain radiographs and MR scans. Am J Routgenol 21 Bernaudin F, Souillet G, Vannier JP et al. Bone marrow trans- 1989; 152: 547–549. plantation (BMT) in 14 children with severe 2 Zizic TM, Marcoux C, Hungerford DS et al. Corticosteroid (SCD): the French experience. Bone Marrow Transplant therapy associated with ischemic necrosis of bone in systemic 1993; 12 (Suppl. 1): 118–121. lupus erythematosus. Am J Med 1985; 79: 596–604. 22 Socie G, Selimi F, Sedel L et al. Avascular necrosis of bone 3 McCallum RI, Walder DN. Bone lesions in compressed air after allogeneic bone marrow transplantation: clinical findings, workers. J Bone Joint Surg 1966; 48: 207–235. incidence and risk factors. Br J Haematol 1994; 86: 624–628. 4 Hiller E, Wolf-Hornung B. Avascular necrosis of the hip fol- 23 Balduzzi A, Gooley T, Anasetti C et al. Unrelated donor mar- lowing combination chemotherapy of acute lymphoblastic leu- row transplantation in children. Blood 1995; 86: 3247–3256. kemia. Blut 1986; 53: 246–247. 24 Berquist TH, Ehman RL, Richardson ML. Magnetic Reson- 5 Prindell G, Weigel W, Jentsch E et al. Aseptic osteonecrosis ance of the Musculoskeletal System. Raven Press: New York, in children treated for acute lymphoblastic leukemia and aplas- 1990, p 176. tic anemia. Eur J Pediatr 1982; 139: 48–51. 25 Camp JF, Colwell CW. Core decompression of the femoral 6 Benz G, Brandeis E, Willich E. Radiological aspects of leuke- head for osteonecrosis. J Bone Joint Surg 1986; 68A: 1313– mia in childhood. An analysis of 89 children. Pediatr Radiol 1319. 1976; 4: 201–213. 26 Ficat RF. Treatment of avascular necrosis of the femoral head. 7 Rossleigh MA, Smith J, Strauss DJ, Engel IA. Osteonecrosis Hip 1983; 2: 279–295. in patients with malignant lymphoma: a review of 31 cases. 27 Totty WG, Murphy WA, Ganz WI et al. Magnetic resonance Cancer 1986; 58: 1112–1116. imaging of the normal and ischemic femoral head. Am J Rout- 8 Smith RE, Chelmowski MK, Anderson T. Osteonecrosis in genol 1984; 143: 1273–1280. patients with immune thrombocytopenia purpura. Am J Hema- 28 Bieber E, Hungerford D, Leunox DW. Factors in the diagnosis tol 1987; 26: 97–99. of avascular necrosis of the femoral head. Adv Orthop Surg 9 Zizic TM, Hungerford DS, Stevens MB. Ischemic bone 1985; 147: 221–226. necrosis in systemic lupus erythematosus. Medicine 1980; 59: 29 Ehman RL, Bequist TH, McLeod RA. MR imaging of the 134–142. musculoskeletal system: a 5-year appraisal. Radiology 1988; 10 Boland EW, Headley NE. Management of 166: 313–320. with smaller (maintenance) doses of acetate. JAMA 30 Mitchell DG, Roa VM, Dalinka MK et al. Femoral head avas- 1950; 144: 365–372. cular necrosis: correlation of MR imaging, radiographic stag- 11 Ibels LS, Allen CA, William EH, Weil R. Aseptic necrosis of ing, radionuclide imaging, and clinical findings. Radiology bone following renal transplantation: experience in 194 trans- 1987; 162: 709–715. plant recipients and review of literature. Medicine 1978; 57: 31 Mitchell MD, Kunkel HL, Steinberg ME et al. Avascular 25–45. necrosis of the hip: comparison of MR, CT, and scintigraphy. 12 Elmstedt E. Incidence of skeletal complications in renal graft Am J Radiol 1986; 147: 67–71. recipients. Acta Orthop Scand 1982; 53: 853–856. 32 Breitenseher MJ, Hoffmann S, Rand T et al. Differentiation 13 Harrington KD, Murray WR, Kountz SI, Belzer FO. Avascu- of bone marrow edema from early vascular necrosis and other lar necrosis of bone after renal transplantation. J Bone Joint hip pathologies: assessment of radiography, CT, MR imaging, Surg (Am) 1971; 53: 203–215. and scintigraphy. Radiology 1996; 201 (Suppl.): 539 14 Feletti C, Di Felice A, Scolari MP, Bonomini V. Glucocorti- (Abstr. 922). coids and avascular bone necrosis in renal transplantation. Adv 33 Glickstein MF, Burk DL Jr, Schiebler ML et al. Avascular Exp Med Biol 1984; 171: 361–368. necrosis versus other diseases of the hip: sensitivity of MR 15 Patton PR, Pfaff WW. Aseptic bone necrosis after renal trans- imaging. Radiology 1988; 169: 213–215. plantation. Surgery 1988; 103: 63–68. 34 Hopson CN, Siverhus SW. Ischemic necrosis of the femoral 16 Elmstedt E. Avascular bone necrosis in the renal transplant head: treatment by core decompression. J Bone Joint Surg patient: a discriminant analysis of 144 patients. Clin Orthop 1988; 70: 1048–1051. 1981; 158: 149–157. 35 Iida S, Harada Y, Ikenoue S et al. High correlation between 17 Isono SS, Woolson ST, Schurman DJ. Total joint arthroplasty bone marrow edema and collapse of the femoral head: pro- for steroid-induced osteonecrosis in cardiac transplant spective study of steroid-induced osteonecrosis in high risk patients. Clin Orthop 1987; 217: 201–208. patients. Radiology 1996; 201 (Suppl.): 199 (Abstr. 392).