Calcium hydroxyapatite ceramic implants in bone tumour surgery

Transcription

1 Calcium hydroxyapatite ceramic implants in bone tumour surgery A LONG-TERM FOLLOW-UP STUDY A. Matsumine, A. Myoui, K. Kusuzaki, N. Araki, M. Seto, H. Yoshikawa, A. Uchida From Mie University School of Medicine and Osaka University Medical School, Japan We reviewed the results of 51 patients with benign bone tumours treated by curettage and implantation of calcium hydroxyapatite ceramic (CHA). The mean follow-up was 11.4 years (10 to 15.5). Post-operative fractures occurred in two patients and three had local recurrences; three had slightly limited movement of the adjacent joint and one had mild osteoarthritis. There were no allergic or neoplastic complications. In all cases, radiographs showed that the CHA was well incorporated into the host bone. Statistical analysis showed that absorption of the implanted CHA was greater in males (odds ratio, 6.2; 95% CI, 1.6 to 23.7) and younger patients (odds ratio, 0.6 for increase in age of 10 years; 95% CI, 0.91 to 0.99). However, the implanted CHA was not completely absorbed in any patient. We conclude that CHA is a useful and safe bone substitute for the treatment of benign bone tumours. A. Matsumine, MD, Assistant K. Kusuzaki, MD, Associate M. Seto, MD, Assistant A. Uchida, MD, Department of Orthopaedic Surgery, Mie University School of Medicine, Edobashi, Tsu-city, Mie , Japan. A. Myoui, MD, Assistant H. Yoshikawa, MD, Department of Orthopaedic Surgery, Osaka University Medical School, 2-2 Yamadaoka, Suita, Osaka , Japan. N. Araki, MD, Chief of the Department Department of Orthopaedic Surgery, Osaka Medical Center for Cancer and Cardiovascular Diseases, Nakamichi, Higashinariku, Osaka , Japan. Correspondence should be sent to Dr A. Matsumine British Editorial Society of Bone and Joint Surgery doi: / x.86b $2.00 J Bone Joint Surg [Br] 2004;86-B: Received 5 February 2003; Accepted after revision 20 January 2004 Large bone defects remaining after resection and curettage of benign bone tumours should be filled with a substitute to restore mechanical strength as soon as possible. Autologous bone grafts have the advantage of low risk of disease transmission and good osteoinduction. However, it may be difficult to obtain the quantity of bone graft required, particularly when the tumour is large or the patient is a child. The choices available include corticocancellous autograft, cortical or cancellous allograft, composite graft, ceramics, polymethylmethacrylate bone cement, bone marrow and demineralised bone matrix. The ideal bone substitute must be easily fabricated and preserved, biocompatible and biodegradable. In 1990, our short-term follow-up study of hydroxyapatite ceramic (CHA) for bone tumour surgery was the first to show that this material is biocompatible with bone and bone marrow. 1 Others have also reported short-term results 2-7 but there have been no reports of long-term clinical and radiographic studies of CHA in bone tumour surgery. Here, we present a series of 51 patients with benign bone tumours followed for a minimum of ten years after treatment by resection and curettage followed by implantation of CHA. Patients and Methods We retrospectively reviewed the results of 51 patients with benign bone tumours sequentially treated by curettage followed by implantation of CHA between April 1986 and January 1991 (Table I). The series comprised 32 males and 19 females, with a mean age of 24.6 years (7 to 59). The histological diagnoses were as follows: simple bone cysts, 13 patients; giant cell tumour of bone, 11; enchondroma, nine; fibrous dysplasia, eight; nonossifying fibroma, three; osteofibrous dysplasia, three; aneurysmal bone cyst, three; chondroblastoma, one. The locations were as follows: tibia, 16 patients; femur, 14; humerus, nine; phalanx, five; calcaneum, two; pelvis, two; scapula, one; fibula, one; talus, one. The mean follow-up was 11.4 years (10 to 15.5). Calcium hydroxyapatite ceramic implants. We used CHA in the form of granules (granule size 300 to 3000 µm) or porous blocks resembling pumice (porosity 35% to 48%; pore diameter 50 to 300 µm) (BONECERAM; Central Research Institute of Sumitomo Cement, Tokyo, Japan). The hydroxyapatite was sintered at 1150 C; the bending strength of each block was between 130 and 200 kgf/cm 2. The spatial dimensions of the blocks varied according to the shape and volume of the cavities. Operative technique. The tumour was exposed through an appropriately sized bone window. Careful and thorough curettage was performed using a curette and an airtome until cancellous bone was visible. Blocks and granules of gas-sterilized CHA were pressed in to VOL. 86-B, No. 5, JULY

2 720 A. MATSUMINE, A. MYOUI, K. KUSUZAKI, N. ARAKI, M. SETO, H. YOSHIKAWA, A. UCHIDA Table I. ceramic Details of 51 patients with benign bone tumours treated by resection and curettage followed by implantation of calcium hydroxyapatite Number Age (yrs) Gender Bone* Site in bone* Diagnosis Symptom at first visit Volume of the cavity (ml) Combination with autologous bone graft Complication Duration of followup (yrs) Radiographic staging at final follow-up 1 32 F Femur d-me GCT Knee pain 36 Yes No F Tibia p-me GCT Knee pain 72 Yes No A 3 17 M Humerus p-m SBC Shoulder pain 26 No No A 4 7 M Humerus D SBC No symptom 57 No Recurrence M Calcaneus SBC Heel pain 4 No No B 6 12 M Tibia D FD Leg pain 23 No No B 7 11 M Femur d-m EC Knee pain 12 Yes No A 8 37 M Femur p-m FD Thigh pain 24 No No A 9 53 M Humerus p-m SBC No symptom 48 No No M Femur p-md FD Thigh pain 48 Yes Fracture B F Scapula GCT Shoulder pain 6 No No M Iliac bone SBC No symptom 16 No No B M Femur D EC No symptom 48 No No M Femur p-m GCT Hip pain 72 Yes No A M Tibia p-me NOF No symptom 112 No No M Tibia p-m GCT Knee pain 80 Yes No B M Femur p-me GCT Knee pain 130 Yes No M Humerus p-m ABC No symptom 36 No Limited ROM F Femur p-m FD Thigh pain 8 No No 11 3B F Tibia D FD Leg pain 32 No No 11 3A F Tibia D FD Leg pain 11 No No F Humerus d-md FD Fracture 24 No Limited ROM A F Humerus p-md SBC Shoulder pain 21 No Recurrence A M Femur p-m SBC Hip pain 18 No No B M Femur p-m SBC Hip pain 20 No No B M Humerus D SBC Brachialgia 11 No No B M Tibia D OFD No symptom 19 Yes Fracture B M Femur p-m SBC Hip pain 6 No No B 29 5 M Femur p-m SBC Thigh pain 2 No No 11 3B 30 8 M Tibia D OFD Leg pain 12 Yes No B F p-phalanx p-m EC Finger pain 0.5 No No M Tibia p-me GCT Knee pain 80 Yes OA B F Tibia D OFD Leg pain 40 No No M Femur d-me GCT Knee pain 110 Yes No B M Tibia p-m GCT Knee pain 30 Yes No A M Tibia d-m GCT Ankle pain 3 No No A F Pubis SBC Pubic pain 10 No No F Tibia p-md FD No symptom 63 No No F p-phalanx p-md EC Swelling 1.5 No No A M Metacarpal D EC Swelling 0.8 No No B M Humerus p-m ABC Shoulder pain 60 No No B F p-phalanx D EC Swelling 0.5 No No F Tibia p-me GCT Knee pain 120 Yes Recurrence B F Tibia D EC Leg pain 12 Yes No F Calcaneus CB Heel pain 0.5 No No M Tibia D NOF No symptom 5 No No M Talus ABC Ankle pain 4 No No A F m-phalanx p-m EC Swelling 0.6 No Limited ROM M Femur d-m NOF No symptom 48 No No A F Humerus p-md EC Brachialgia 29 No No M Fibula D SBC Leg pain 6 No No B * p-; proximal; m, middle; d-, distal E, epiphysis; M, metaphysis; D, diaphysis GCT, giant cell tumour; SBC, simple bone cyst; FD, fibrous dysplasia; OFD, osteofibrous dysplasia; EC, enchondroma; NOF, non-ossifying fibroma; ABC, aneurysmal bone cyst; CB, chondroblastoma fill the bone defect. No other adjuvant treatment such as phenol or cryotherapy was used. The mean volume of the bone cavity was 32.5 ml (0.5 to 130). In 14 patients, CHA was used in combination with autologous bone graft. In six patients, internal fixation using plates and screws was used in addition to curettage, if the patient s bone was deemed to be too fragile. In 19 patients, the growth plate was still open at the time of the procedure. Assessment and statistical analysis. Assessment was based on clinical examination and radiography at the final followup. All assessments were performed by the musculoskeletal tumour service teams of Osaka University Medical School THE JOURNAL OF BONE AND JOINT SURGERY

3 CALCIUM HYDROXYAPATITE CERAMIC IMPLANTS IN BONE TUMOUR SURGERY 721 Stage 1 Clear margin Stage 2 Hazy margin Stage 3 Absorbed < 50% of the margin > 50% of the margin A B Fig. 1 Diagram showing the three stage radiographic classification. Stage 1, clear margin; stage 2, hazy margin; stage 3, absorbed margin. Stage 3 was subdivided into stage 3A, absorption less than 50% of the margin of CHA and stage 3B, absorption more than 50% of the margin of CHA. Table II. Relationship between radiographic staging at final follow-up and various factors Number of Radiographic staging at final follow-up Characteristics patients 2 3A 3B p value Number of patients Gender F M Age (yrs) < Location in body Inferior limb Non-inferior limb Type of bone Long tubular bone Non-long tubular bone Site in the long bone Diaphysis only Including metaphysis or epiphysis Volume of the cavity <20 (ml) (ml) Combination with autologous bone graft Yes No Duration of the follow-up <11 (yrs) (yrs) and Mie University School of Medicine, under the direct supervision of the senior author (AU). Radiographic findings were classified into three stages: stage 1, clear margin; stage 2, hazy margin; stage 3, absorbed margin. Stage 3 was subdivided further into stage 3A, absorption <50% of the margin of CHA and stage 3B, absorption >50% of the margin of CHA (Fig. 1). Post-operative CT or MRI was performed to detect local recurrence. In order to study factors affecting the absorption of the CHA, statistical analysis was performed using StatView statistical software (version 5.0; SAS Institute Inc. Cary, North Carolina). Univariate analysis was performed using VOL. 86-B, No. 5, JULY 2004

4 722 A. MATSUMINE, A. MYOUI, K. KUSUZAKI, N. ARAKI, M. SETO, H. YOSHIKAWA, A. UCHIDA Fig. 2a Fig. 2b Fig. 2c Radiographs of a 28-year-old man with an aneurysmal bone cyst of the proximal humerus (case 41), showing a) an anteroposterior view of the right shoulder with extensive osteolysis, b) clear margin of the CHA one month after operation, and c) absorbed margin of CHA more than 50% (stage 3B) 15.5 years after operation with increased density of the CHA implant and fusion of granules to each other. the Mann-Whitney U test for nonparametric data. When one-factor analysis of variance indicated a significant difference, multiple logistic regression analysis was used to estimate the odds ratio (OR) of absorption of the CHA implant; less than p < 0.05 was considered significant. Results Clinical findings. There were no adverse local effects such as excessive post-operative drainage or erythema, during the early post-operative period. No patient developed postoperative infection; no toxicity was detected by routine serological examinations; no allergic or neoplastic disease was observed during the follow-up period. Functionally three patients, one with an aneurysmal bone cyst of the proximal humerus, one with fibrous dysplasia of the diametaphyseal humerus, and one with enchondroma of the middle phalanx had a limited range of movement in joints adjacent to the lesions. One patient, with a giant cell tumour beneath the subchondral bone plate of the proximal tibia, had some knee pain with mild radiographic osteoarthritic changes eight years after surgery, but he suffered no limitation of daily activities. There were three local recurrences of tumour; two were from the group of 13 patients with a simple bone cyst, and one from the group of 11 patients with a giant cell tumour. The two patients with recurrence of a bone cyst did not require further surgery because there was no expansive growth of the lesions. The patient with recurrent giant cell tumour required repeat curettage and CHA implantation two years after the initial surgery. Two patients, one with fibrous dysplasia of the femur, and one with osteofibrous dysplasia of the tibia, suffered a post-operative fracture within two months of the initial procedure. Both these fractures united without deformity after immobilisation in a cast. Radiographic findings. All patients had good filling of their bone defects by CHA. Radiographs at the final follow-up were staged as follows: stage 2, in 19 patients; stage 3A, in 13; stage 3B, in 19 (Table II). No clear margins (stage 1) were detected at the final follow-up. Initial radiolucent lines and halos around CHA implants faded with time, and eventually disappeared. Radiographic density at the implant sites increased with time and some CHA blocks and granules fused to each other (Fig. 2). There was no displacement of an implant from its intra-osseous site. In skeletally immature patients, implanted CHA granules scattered within the bone and were eventually absorbed (Fig. 3). Progressive remodelling of any bony deformity resulting from the tumour occurred over time (Fig. 4). Longitudinal bone growth was not disturbed, even when CHA was implanted close to the growth plate. At the final follow-up, radiographs showed that the implanted CHA was well incorporated into the surrounding host bone in every patient but in no patient was it seen to be completely absorbed. Table II shows the radiographic staging at the final follow-up in the two groups. Univariate analysis demon- THE JOURNAL OF BONE AND JOINT SURGERY

5 CALCIUM HYDROXYAPATITE CERAMIC IMPLANTS IN BONE TUMOUR SURGERY 723 Radiographs of a nine-year-old boy with a simple bone cyst of the fibula (case 51), showing a) lytic change of the diaphysis, b) clear margin of the CHA one month after operation, and c) absorbed margin of CHA more than 50% (stage 3B) 10.4 years after operation. Fig. 3a Fig. 3b Fig. 3c Fig. 4a Fig. 4b Fig. 4c Anteroposterior radiographs of a 10-year-old girl with enchondroma of proximal phalanx (case 39), showing a) expansive growth of the tumour with thin cortex, b) clear margin of the implanted CHA one month after operation, and c) absorbed margin of the CHA less than 50% (stage 3A) 11.6 years after operation with remodelling of the deformity of the metaphysis. VOL. 86-B, No. 5, JULY 2004

6 724 A. MATSUMINE, A. MYOUI, K. KUSUZAKI, N. ARAKI, M. SETO, H. YOSHIKAWA, A. UCHIDA Table III. Multivariate analysis of factors predictive of radiographic absorption of the implanted CHA Variable Odds ratio 95% CI p value Gender Female 1 Male to Age (increasing by 10-year increments) to Table IV. Reports of results of CHA implantation following curettage of bone tumour Study Type of CHA Number of patients Uchida et al 1 CHA Inoue et al 2 CHA Yamaguchi et al 3 CHA 7 86 Yamamoto et al 4 CHA Joosten et al 5 HA cement 8 12 Irwin et al 6 Coralline HA Present study CHA Duration of follow-up (mths) strated that the odds rate of absorption of the implanted CHA was significantly higher for male patients (p < 0.01) and in patients less than 30 years of age (p < 0.05). Our data do not indicate a significant association between the odds rate of absorption of the implanted CHA and location in the body (lower limb vs upper; p = 0.09), type of the bone (long tubular bone vs short; p = 0.24), site in the bone (diaphysis only vs diaphysis with metaphysis or epiphysis; p = 0.72), volume of the cavity (p = 0.52), combination with autologous bone graft (p = 0.15) or duration of the followup period (p = 0.33). Multivariate analysis showed that absorption of the implanted CHA was significantly greater in males (OR, 6.2; 95% CI, 1.6 to 23.7) and younger patients (OR, 0.6 for an increase in age of 10 years; 95% CI, 0.91 to 0.99) (Table III). Discussion Calcium hydroxyapatite ceramic has been used for repair of bony defects resulting from severely comminuted fractures, 8 in spinal surgery 9 and arthroplasty. 10,11 The orthopaedic oncologist is often faced with large bony defects during surgical procedures to remove bone tumours. When the length of the cortical defect exceeds 75% of the cortical diameter, the open-section effect results, and this can greatly reduce the load-carrying capacity of the cortical bone, particularly under torsional loading. 12 Immediately after surgery, there is no difference in mechanical strength between bone filled with bone substitute and bone left empty. However, because bone substitute can act as a scaffold for new bone formation, bone filled with substitute can gradually become stronger against mechanical stress than a bone defect which is left empty. Although autologous bone graft has long been considered the ideal bone substitute, its disadvantages include limited availability, tumour implantation into the donor site, and the risk of pain, infection and nerve damage 13 at the donor site. Porous and granular CHA implanted into bony defects can provide a suitable framework for osteogenesis, and compares well with allografts and xenografts. 1,4 Shortterm follow-up studies have demonstrated the benefits of implanted CHA both biologically and mechanically 1-7 (Table IV). We noticed a slightly limited range of movement in joints adjacent to the lesions in three patients with an aneurysmal bone cyst of the proximal humerus, fibrous dysplasia of the distal metadiaphyseal humerus and an endochondroma of the middle phalanx, respectively. However, this stiffness was not due to hydroxyapatiteinduced fibrosis or arthritis, but rather to soft-tissue adhesion. There was no displacement of hydroxyapatite granules into the soft tissues or the joint space. One other patient, with a giant cell tumour beneath the subchondral bone plate of the proximal tibia, had slight knee pain with mild radiographic degenerative changes eight years after surgery. Yamamoto et al 4 did not detect degenerative changes in any joint in 75 patients who received CHA implants after resection of benign bone tumours, but the mean follow-up was short (41.3 months). Itokazu et al 8 successfully treated tibial plateau fractures in 17 patients using CHA as a bone substitute, without progression of osteoarthritis. In two cases with lesions in the lower limb (fibrous dysplasia of the femur, and osteofibrous dysplasia of the tibia), post-operative fractures occurred within two months of CHA implantation. In the early post-operative period, the mechanical strength of CHA-implanted bone tends to be overestimated because of its dense appearance on radiographs. Hamson et al 14 reported that, within six months of implantation, there was no significant difference in mechanical strength between CHA and autologous cortical bone. Vuola et al 15 showed that, at 12 weeks, the compressive strength of CHA was greater than that of cancellous bone. For patients with bone tumours in the lower limbs in which it occupies more than half of the medullary cavity, as shown by CT scan, we recommend that full weight-bearing is delayed for at least three months. The fractures in the present series united after immobilisation, with retention of the CHA implant and without deformity. In our study, although thorough curettage of the tumour was performed, there were three local recurrences; two of these patients had simple bone cysts (rate of recurrence 15%) but they did not require re-operation, because the lesion did not expand. Curettage with autogenous bone graft is the most common treatment for simple bone cysts, despite reported local recurrence rates of 30% to 50%. 16 The other patient with recurrence had a giant cell tumour (1 of 11; rate of recurrence 9%). There have been reports of recurrence rates of between 12% and 25% for giant cell tumour The small number of patients and the noncom- THE JOURNAL OF BONE AND JOINT SURGERY

7 CALCIUM HYDROXYAPATITE CERAMIC IMPLANTS IN BONE TUMOUR SURGERY 725 parative design of our study are not suitable for statistical comparison of local recurrence rates. In patients implanted with CHA, there was no disease transmission, no potential increase of infection, no toxic effects, and no allergic or neoplastic induction. In our study, radiographs show: 1) new bone formation within the CHA; 2) hazy or partially absorbed margin of the CHA; 3) more significant absorption of the CHA in skeletally immature patients and in male patients; 4) normal fracture healing and 5) corrective remodelling of the affected bone. Because no quantitative measurements are available for evaluation of bone ingrowth into porous material, it is not possible accurately to compare ingrowth rates of different bone substitutes. However, we have previously reported bone ingrowth into the pore structures of implanted CHA at one year after implantation. 1 Such human histological findings are consistent with the results of animal experiments, 20,21 although bone ingrowth into pore structures took place more slowly in humans. Bioabsorption of CHA in vivo over time is not completely understood, due to the lack of long-term follow-up studies (Table IV). Radiographically, in no patient was the CHA seen to disappear completely, but it was partially absorbed at final follow-up. Yamaguchi et al 3 described the degradation-resistant character of synthetic hydroxyapatite blocks used to fill bony defects, followed for 78 to 109 months. By contrast, Goto et al 22 recently reported that synthetic hydroxyapatite blocks implanted into an iliac bone defect harvested for grafting, were completely absorbed and replaced by newly formed bone, 79 months after implantation. A recent epidemiological study has shown that peak bone mass occurs between the ages of 21 and 30 years, and that bone density correlates with youth and male gender. 23 Static histomorphometric analysis using human iliac crest bone biopsies, has demonstrated that there is a linear decline in trabecular bone volume with age. 24 Therefore absorption of implanted CHA should correlate with normal bone turnover, although biodegradability of CHA appears to be influenced by other factors including pore diameter, porosity, calcined temperature and the shape of the implant. 22 In our study, CHA showed several favourable features including lack of adverse effects, ease of manufacture, ease of preservation and good adjustment of shape. Our results indicate that CHA is a useful and safe bone substitute in the treatment of benign bone tumours. No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. References 1. Uchida A, Araki N, Shinto Y, et al. The use of calcium hydroxyapatite ceramic in bone tumour surgery. J Bone Joint Surg [Br] 1990;72-B: Inoue O, Ibaraki K, Shimabukuro H, Shingaki Y. Packing with high-porosity hydroxyapatite cubes alone for the treatment of simple bone cyst. Clin Orthop 1993; 293: Yamaguchi K, Hirano T, Yoshida G, Iwasaki K. Degradation-resistant character of synthetic hydroxyapatite blocks filled in bone defects. Biomaterials 1995;16: Yamamoto T, Onga T, Mariu T, Mizuno K. Use of hydroxyapatite to fill cavities after excision of benign bone tumours: clinical results. J Bone Joint Surg [Br] 2000;82-B: Joosten U, Joist A, Frebel T, Walter M, Langer M. The use of an in situ curing hydroxyapatite cement as an alternative to bone graft following removal of enchondroma of the hand. J Hand Surg [Br] 2000;25: Irwin RB, Bernhard M, Biddinger A. Coralline hydroxyapatite as bone substitute in orthopedic oncology. Am J Orthop 2001;30: Heise U, Osborn JF, Duwe F. Hydroxyapatite ceramic as a bone substitute. Int Orthop 1990;14: Itokazu M, Matsunaga T, Ishii M, Kusakabe H, Wyni Y. Use of arthroscopy and interporous hydroxyapatite as a bone graft substitute in tibial plateau fractures. Arch Orthop Trauma Surg 1996;115: Cook SD, Dalton JE, Tan EH, et al. In vivo evaluation of anterior cervical fusions with hydroxyapatite graft material. Spine 1994;19: Hernandez Cortes P, Najera Sagastume OO, Mesa Ramos F, Pajares Lopez M. Hydroxyapatite-coated stems with metaphyseal and diaphyseal press-fit: eleven-year follow-up result. Acta Orthop Belg 2002;68: Epinette JA, Manley MT, D Antonio JA, Edidin AA, Capello WN. A 10-year minimum follow-up of hydroxyapatite-coated threaded cups: clinical, radiographic and survivorship analyses with comparison to the literature. J Arthroplasty 2003;18: Frankel VH, Burstein AH. Load capacity of tubular bone. In: Kenedi, RM, ed. Biomechanics and related bio-engineering topics. Oxford, etc: Pergamon Press, 1965: Younger EM, Chapman MW. Morbidity at bone graft donor site. J Orthop Trauma 1986;3: Hamson KR, Toth JM, Stiehl JB, Lynch KL. Preliminary experience with a novel model assessing in vivo mechanical strength of bone grafts and substitute materials. Calcif Tissue Int 1995;57: Vuola J, Taurio R, Goransson H, Asko-Seljavaara S. Compressive strength of calcium carbonate and hydroxyapatite implants after bone-marrow-induced osteogenesis. Biomaterials 1998;19: Neer CS, Francis KC, Johnston AD, Kiernan HA Jr. Current concepts on the treatment of solitary unicameral bone cyst. Clin Orthop 1973;97: Trieb K, Bitzan P, Lang S, Dominkus M, Kotz R. Recurrence of curetted and bonegrafted giant-cell tumours with and without adjuvant phenol therapy. Eur J Surg Oncol 2001;27: Blackley HR, Wunder JS, Davis AM, et al. Treatment of giant-cell tumors of long bones with curettage and bone-grafting. J Bone Joint Surg [Am] 1999;81-A: Turcotte RE, Wunder JS, Isler MH, et al. Giant cell tumor of long bone: a Canadian Sarcoma Group study. Clin Orthop 2002;397: Uchida A, Nade SM, McCartney ER, Ching W. The use of ceramics for bone replacement: a comparative study of three different porous ceramics. J Bone Joint Surg [Br] 1984;66-B: Uchida A, Nade S, McCartney E, Ching W. Bone ingrowth into three different porous ceramics implanted into the tibia of rats and rabbits. J Orthop Res 1985;3: Goto T, Kojima T, Lijima T, et al. Resorption of synthetic porous hydroxyapatite and replacement by newly formed bone. J Orthop Sci 2001;6: Lin JD, Chen JF, Chang HY, Ho C. Evaluation of bone mineral density by quantitative ultrasound of bone in 16,862 subjects during routine health examination. Br J Radiol 2001;74: Thomsen JS, Ebbesen EN, Mosekilde L. Relationships between static histomorphometry and bone strength measurements in human iliac crest bone biopsies. Bone 1998;22: VOL. 86-B, No. 5, JULY 2004