- Email: [email protected]
Histologic and mechanical evaluation of impacted morcellized cancellous allografts in rabbits
The Journal of Arthroplasty Vol. 15 No. 5 2000
Histologic and Mechanical Evaluation of Impacted Morcellized Cancellous Allografts in Rabbits Comparison With Hydroxyapatite Granules H. Yano, MD,* H. Ohashi, MD, PhD,* Y. Kadoya, MD, PhD,* A. Kobayashi, MD, PhD,* Y. Yamano, MD, PhD,* and Y. Tanabe, PhD†
Abstract: The bioactivity and mechanical properties of morcellized allografts and hydroxyapatite (HA) granules were evaluated in a rabbit model. Allografts were replaced by viable trabecular structures within 8 weeks. The yield strength and stiffness of allografts were within normal cancellous bone levels by 3 weeks and were maintained afterward. The amount of newly formed bone around HA granules was comparable to that around allografts. The yield strength and stiffness of HA granules were significantly higher than those of allografts at 3 and 12 weeks. Allografts offer the advantage of being replaced by host–bone without significant deterioration in mechanical properties over the course of remodeling. HA granules can also be used for a bone substitute given their bioactivity in bone conduction and superiority in mechanical properties to allografts. Key words: allograft, hydroxyapatite, impaction grafting, bone formation, mechanical properties, animal model.
In revision total joint arthroplasties, reconstruction of joints with extensive bone loss presents a major challenge. Impaction grafting using morcellized allografts was introduced for acetabular protrusion in 1984. Since then, satisfactory results have been reported with impaction allografting of morcellized allografts for acetabular [1–3] and femoral [4–7] reconstruction. Radiographically, it has been shown that impacted morcellized allografts undergo remodeling into trabecular structures with recovery of cortical bone [4,5]. Allografts have limitations for
clinical application, however, because of limited availability and disease transmission. Given the disadvantages of allograft material in clinical practice, hydroxyapatite (HA) granules are a promising material not only as a substitute for bone–graft, but also as a bioactive material for bone conduction. It has also been reported that HA granules can be used as a substitute for morcellized allografts in acetabular reconstruction [8,9]. With the increasing application of the impaction grafting technique, massive early subsidence of the stem after femoral impaction allografting has been reported [10,11]. Poor osteoconductivity of morcellized allografts has been reported in a series of animal experiments using titanium bone conduction chambers [12,13]. It is essential to evaluate the correlation of histologic findings with mechanical properties of graft materials during the course of remodeling. To our knowledge, however, there have been no in vivo studies describing the correla-
From the *Department of Orthopaedic Surgery, Osaka City University Medical School, Osaka; and †Department of Mechanical Engineering, Faculty of Engineering, Niigata University, Niigata, Japan. Submitted June 2, 1999; accepted December 7, 1999. No benefits or funds were received in support of this study. Reprint requests: H. Yano, MD, Department of Orthopaedic Surgery, Osaka City University Medical School 1-4-3 Asahimachi, Abeno-ku, Osaka 545-8585 Japan. Copyright r 2000 by Churchill Livingstonet 0883-5403/00/1505-0014$10.00/0 doi:10.1054/arth.2000.6625
635
636 The Journal of Arthroplasty Vol. 15 No. 5 August 2000 tion between histologic findings and mechanical properties of impacted cancellous allografts; consequently, the change in mechanical properties after grafting remains unknown. The aim of this study was 2-fold: first, to determine correlations between histologic findings and mechanical properties of morcellized allografts and HA granules using a cavitary bone defect model in rabbits, and second, to evaluate the usefulness of HA granules as a substitute for allograft bone chips.
Materials and Methods Preparation of Graft Material Iliac bones of 48 rabbits (donors) were harvested under sterile conditions and stored at ⫺82°C for at least 2 weeks. They were cut into small cancellous bone chips (size, 1.0–3.0 mm in diameter) just before grafting and washed with physiologic saline. HA granules (granule size, 0.9–1.5 mm in diameter; pore size, 50–300 µm; porosity, 35%–48%; sintered temperature, 1,150°C) were obtained from Sumitomo Cement Co, Ltd (Osaka, Japan). Grafting Methods Male Japanese white rabbits (body weight, approximately 3 kg) obtained from a breeder other than the donors’ were anesthetized by intramuscular injection of ketamine hydrochloride, 25 mg/kg, and xylazine, 1.5 mg/kg. Through a medial parapatellar approach, a hole (diameter, 6 mm; depth, 12 mm) was drilled in a femoral condyle (Fig. 1). The hole was rinsed with physiologic saline before implantation. The animals were divided into 3 experimental groups as follows: i) empty group—the holes were left empty; ii) allograft group—prepared cancellous allograft chips were impacted into the holes with an impactor; and iii) HA group—HA granules were impacted in the same manner. After the
aforementioned procedures, rabbits were allowed unrestricted movement in their cages. Study A: Histologic Examination Thirty-six rabbits were killed with an overdose of barbiturate at 0, 3, 8, or 12 weeks postoperatively, and the femora were resected just above the condyles. Undecalcified, methyl methacrylate–embedded specimens were prepared. Sections measuring 4 µm thick, perpendicular to the drilling axis at the midcondylar region, were stained with hematoxylin and eosin and histologically examined. Areas of grafted bone, HA granules, and newly formed bone were measured with image analysis software (NIH Image). Newly formed bone was distinguishable from the grafted bone by the presence of osteocytes in the lacunae. All measurements were confined to the central 60% of the area of the drilled hole to avoid errors in tracing the edge of the original hole. Total bone volume was defined as the sum of newly formed bone and grafted bone areas in the allograft group and as the sum of newly formed bone and HA granule areas in the HA group. The volume fractions of total bone and newly formed bone were expressed as the percentages of total bone volume and newly formed bone volume of measured hole area. Six femoral condyles in 3 untreated rabbits (male; body weight, approximately 3 kg) were examined in the same fashion and served as normal controls. Study B: Radiographs and Bone Densitometric Analysis A second group of 36 male Japanese white rabbits underwent surgery and was divided in the same fashion as in study A. The femora were resected, and soft tissue was removed. After taking radiographs, bone mineral density (BMD) in the grafted area was measured using dual-energy x-ray absorptiometry (Hologic QDR-2000, Waltham, MA). Femoral condyles were placed on a precision-milled acrylic block 4 cm in thickness. The acrylic block was used as a soft tissue substitute for accurate BMD measurement. The long axis of the hole was adjusted parallel to the x-ray beam. Ultra–highresolution scan mode (version 4.64) was used with a 13-mm collimator, 2.54-mm line spacing, and 1.27-mm resolution. Study C: Mechanical Testing
Fig. 1. Schematic anterior and medial drawings of distal femur. A hole was drilled from the medial cortex.
The specimens of study B were mounted with an epoxy-resin adhesive (SpeciFix-20, Struers A/S, Copenhagen, Denmark). Mechanical testing could not be performed on specimens in the empty group
Impacted Allografts and Hydroxyapatite Granules ●
Yano et al.
637
because the holes remained empty at all time intervals. The epoxy-resin adhesive was applied to both ends of specimens to make the surfaces parallel. Because the resin did not infiltrate through cortical bone, the strength of specimens was not affected by the mounting procedure. Rectangular test pieces, 5 ⫻ 5 ⫻ 12 mm, sandwiching the grafted area between cortical bones, were prepared (Fig. 2A). A quasistatic compressive test was performed at room temperature using an Instron-type materials testing machine (RTC-1225AS, A&D Co, Ltd, Tokyo, Japan) at a crosshead speed of 0.5 mm/min (Fig. 2B). Compressive load was applied longitudinally to the specimen under lateral confinement by a specially designed acrylic jig (Fig. 2C). All specimens were kept moist during testing. Six femoral condyles in 3 untreated rabbits (male; body weight, approximately 3 kg) were examined in the same fashion in study B & C and served as normal controls. The yield strength and stiffness of each specimen were determined from the load-displacement curves. In this study, stiffness was defined as the slope of load-displacement curves and was determined from the linear portion of the curves at the initial stage of deformation. When the load-displacement curve did not exhibit a yield point, the yield strength was defined as the turning point from linear line to nonlinear line on the load-displacement curves, according to the method proposed by Hvid and Jensen [14]. Statistical Analysis Results for each group are expressed as means ⫾ 1 standard deviation. Statistical analyses were performed with one-way analysis of variance followed by Fisher’s protected least significant difference (PLSD) posthoc test using a commercial software package (StatView 4.11, Abacus Concepts, Inc, Berkeley, CA).
Results Study A: Histologic Examination Representative low-power field photomicrographs are shown in Fig. 3. In the empty group, the holes remained vacant through 12 weeks (Fig. 3A). In the allograft group, bone formation was observed after 3 weeks, and the woven bone underwent remodeling to mature trabeculae at 8 weeks (Fig. 3B). In the HA group, bone formation was intense at 3 weeks, and fine trabeculae were observed at 8 weeks. At 12 weeks, the trabeculae were thick in the periphery of the hole, whereas most HA granules remained unreplaced (Fig. 3C). At higher mag-
Fig. 2. (A) Schematic drawing of sampling site in a condyle for the mechanical test. (B) The specimen was mounted on an Instron-type materials testing machine. (C) A quasistatic compressive test was performed. Compressive load was applied longitudinally to the specimen under lateral confinement by a specially designed acrylic jig.
638 The Journal of Arthroplasty Vol. 15 No. 5 August 2000
Fig. 3. Low-power field photomicrographs for (A) empty, (B) allograft, and (C) hydroxyapatite groups. The trabecular structure consisting of viable bone is seen after 8 weeks in the allograft group. (Hematoxylin and eosin stain.)
nification, allograft bone was surrounded by woven bone at 3 weeks (Fig. 4), and viable trabeculae had been re-established with little remnant of the allograft at 8 weeks (Fig. 5). Histomorphometric analysis revealed that the volume fraction of total bone increased at 3 weeks, then gradually decreased in 3 groups (Fig. 6). In the empty group, the volume fraction was significantly lower than normal control levels at all time intervals. In the allograft group, it increased at 3 weeks, then decreased to normal control levels at 8 weeks. In the HA group, it was significantly higher than in the other groups at all time points tested (P ⬍ .0001).
Fig. 4. High-power field photomicrograph in the allograft group at 3 weeks. Allograft chips, recognized by empty lacunae, are surrounded by woven bone. (Hematoxylin and eosin stain.)
When measurement was confined to the volume fraction of newly formed bone, the allograft and HA groups showed a similar pattern; the percentage of newly formed bone volume peaked at 3 weeks (allograft group, 14.18% ⫾ 2.72%; HA group, 18.47% ⫾ 3.52%; P ⫽ .03%), followed by a gradual decrease to approximately 10% and 14% at 12 weeks (Fig. 7). No statistically significant difference was observed between these 2 groups at 8 or 12 weeks. The volume fraction of newly formed bone
Fig. 5. Photomicrograph in the allograft group at 8 weeks. Central portion of the drilled hole is on the upper right, and peripheral portion is on the lower left. A few remnants of allografts (arrowheads) are seen only in the center. Incorporation and revascularization spread from the peripheral to the center. (Hematoxylin and eosin stain, ⫻60.)
Impacted Allografts and Hydroxyapatite Granules ●
Yano et al.
639
Study B: Radiographs and Bone Densitometric Analysis
Fig. 6. The volume fraction of total bone in the empty (䊏), allograft (䉱), and hydroxyapatite (䊉) groups (n ⫽ 6 in each). Values are mean ⫾ standard deviation. *P ⬍ .0001, **P ⬍ .01 versus normal control (夹) (n ⫽ 6).
was significantly greater in the allograft and HA groups than that in the empty group at all 3 time periods (3, 8, and 12 weeks). Table 1 shows newly formed bone volume as a percentage of total bone volume. In the allograft group, it was 60.2% ⫾ 11.6% at 3 weeks and increased significantly to 93.0% ⫾ 6.1% at 12 weeks. In the HA group, it was 32.6% ⫾ 6.1% at 3 weeks and gradually decreased to 25.7% ⫾ 3.3% at 12 weeks.
In the empty group, no significant radiographic change was observed in the drilled hole through the experimental period (Fig. 8). In the allograft group, the margin of the hole became unclear, and reorganization of the trabeculae was observed at 8 weeks. These findings correspond to the histologic observation of incorporation of the grafted bone and trabecular bone formation from the peripheral to central. At 12 weeks, the original hole was virtually indistinguishable from the surrounding intact bone. In the HA group, intense bone formation was evident within the grafted and surrounding bone areas at 3 weeks. Subsequently, the radiodensity decreased, and HA granules became partially visible after 8 weeks. In each group, the BMD closely correlated with the volume fraction of total bone in the histomorphometric analysis (Fig. 9). Study C: Mechanical Testing Representative load-displacement curves for the allograft and HA groups are shown in Fig. 10. Yield points observed commonly in normal controls were identified in all specimens except for those of the allograft group at 3 weeks. In the allograft group, the yield strength and stiffness were within normal control levels at 3 weeks and were maintained through 12 weeks (Figs. 11 and 12). The yield strength and stiffness in the HA group were significantly higher than those in the allograft group at 3 and 12 weeks but decreased to normal control levels at 8 weeks.
Discussion
Fig. 7. The volume fraction of newly formed bone in the empty (䊏), allograft (䉱), and hydroxyapatite (䊉) groups (n ⫽ 6 in each). Values are mean ⫾ standard deviation. *P ⫽ .03 versus allograft group.
The validity of the impaction allografting technique has been shown previously either histologically or mechanically. Histologically, grafted bone chips were shown to be replaced by host–bone in human retrieval studies [15–19] and animal experiments [12,13,20–23]. Mechanically, initial fixation of the femoral stem with this technique was examined using cadavers [24,25]. The mechanical properties of morcellized cancellous allografts themselves also have been reported [26–28]. Most such studies have not correlated the findings of histologic and mechanical analyses, however, and the possibility of deterioration in strength of impacted morcellized allografts in the course of incorporation has not been tested. To our knowledge, this is the first in vivo study to evaluate the relationships between histo-
640 The Journal of Arthroplasty Vol. 15 No. 5 August 2000 Table 1. Newly Formed Bone Volume as a Percentage of Total Bone Volume*
Allograft group Hydroxyapatite group
3 Weeks
8 Weeks
12 Weeks
60.2% ⫾ 11.6%† 32.6% ⫾ 6.1%
85.0% ⫾ 4.1% 28.0% ⫾ 9.7%
93.0% ⫾ 6.1% 25.7% ⫾ 3.3%
*Values are mean ⫾ standard deviation, n ⫽ 6 in each. †P ⬍ .0001 versus allograft group at 8 and 12 weeks.
logic findings and mechanical properties of impacted morcellized allografts themselves. The first point shown in this study was that morcellized cancellous allografts could be replaced by host–bone without significant deterioration in their mechanical properties during the course of remodeling. In human retrieval studies, revascularization of morcellized allografts took at least 4 months, and conversion of the graft to vital trabecular structure required about 8 months [18]. In clinical situations, a substantially longer period might be required for revascularization and remodeling. Maintenance of mechanical strength during the remodeling process is important for impacted allografts. In our study, grafted bone was extensively remodeled into reorganized viable trabecular structure after 8 weeks. The distinct yield point observed commonly in normal controls was also recognized
Fig. 8. Representative radiographs for (A) empty, (B) allograft, and (C) hydroxyapatite groups. In the allograft group, consolidation of grafted bone is recognized at 8 weeks.
after 8 weeks. These findings indicated that 8 weeks is required for allografts to re-establish normal cancellous bone structure. At 3 weeks, the mixed structure of woven bone and grafted chips did not show a yield point and was considered to be structurally immature. By 8 weeks, the yield strength and stiffness of allografts were within the range of normal cancellous bone, indicating that the mechanical properties of allografts were sustained in the course of remodeling. The second point suggested by this study is the possible use of HA granules as a substitute for morcellized allografts. The BMD and volume fraction of total bone in the HA group were significantly higher than in the other groups at all time points tested, and extensive bone apposition onto the granules indicative of the osteoconductive effect of HA granules was observed at 3 weeks. The yield strength and stiffness were significantly higher than those in the allograft group at 3 and 12 weeks but dropped to normal control levels at 8 weeks. This pattern may be attributable to the resorption of newly formed bone during remodeling secondary to the absence of the direct weight bearing in this
Fig. 9. Bone mineral density (BMD) in the empty (䊏), allograft (䉱), and hydroxyapatite (䊉) groups (n ⫽ 6 in each). Values are mean ⫾ standard deviation. *P ⬍ .0001, **P ⫽ .01 versus normal control (夹) (n ⫽ 6).
Impacted Allografts and Hydroxyapatite Granules ●
Yano et al.
641
Fig. 10. Representative loaddisplacement curves for the (A) allograft and (B) hydroxyapatite (HA) groups. Yield points (arrowheads) as observed commonly in normal controls are recognized except for 3-week specimen in the allograft group.
conditions may affect new bone formation and mechanical properties. The lack of direct loading may be 1 reason that the newly formed bone was gradually absorbed in the allograft and HA groups. The sequence of histologic events that occurred in our specimens was identical, however, to that observed in the process of incorporation of impacted morcellized allografts in loaded animal studies [20,21]. The physiologic mechanical stress may have affected the incorporation and remodeling process of allografts in our model because the yield strength and stiffness of allografts were similar to those of normal cancellous bone during the course of remodeling.
study. Subsequently, the fine trabecular structure observed at 8 weeks was remodeled further to a matured thick trabecular structure at 12 weeks. This matured thick trabecular structure was stronger than the fine trabecular structure observed at 8 weeks, as indicated by the increase of stiffness and yield strength between 8 and 12 weeks. One limitation of this study is that bone defects were created just before grafting and surrounded by cancellous bone with abundant blood supply. These conditions are not equivalent to those of bone defects encountered in revision surgery. In a study of biopsy specimens obtained at revision operations, 1 of us previously showed that new bone often forms over much of the original bone surface of the implant–bone interface [29]. The results obtained here have meaningful implications for revision surgery. Another limitation is that there was no direct loading of grafts in this study. These mechanical
We investigated the bioactivity and mechanical properties of impacted morcellized cancellous allo-
Fig. 11. Yield strength in the allograft (䉱) and hydroxyapatite (䊉) groups (n ⫽ 6 in each). Values are mean ⫾ standard deviation. *P ⬍ .0001 versus allograft group and normal control (夹) (n ⫽ 6), **P ⫽ .04 versus allograft group.
Fig. 12. Stiffness in the allograft (䉱) and hydroxyapatite (䊉) groups (n ⫽ 6 in each). Values are mean ⫾ standard deviation. *P ⬍ .0001 versus allograft group and normal control (夹) (n ⫽ 6).
Conclusion
642 The Journal of Arthroplasty Vol. 15 No. 5 August 2000 grafts and HA granules. Morcellized cancellous allografts offer the advantage of being replaced by host– bone, while maintaining yield strength and stiffness. HA granules can also be used as a bone substitute, given their bioactivity in bone conduction and superiority in mechanical properties to allografts.
Acknowledgment We thank Mr H. Maki and Mr T. Meguro, of the Department of Mechanical Engineering, Faculty of Engineering, Niigata University, for their assistance.
References 1. Slooff TJJH, Huiskes R, Horn JV, Lemmens AJ: Bone grafting in total hip replacement for acetabular protrusion. Acta Orthop Scand 55:593, 1984 2. Azuma T, Yasuda H, Okagaki K, Sakai K: Compressed allograft chips for acetabular reconstruction in revision hip arthroplasty. J Bone Joint Surg Br 76:740, 1994 3. Schreurs BW, Slooff TJJH, Buma P, et al: Acetabular reconstruction with impacted morsellised cancellous bone graft and cement: a 10- to 15-year follow-up of 60 revision arthroplasties. J Bone Joint Surg Br 80:391, 1998 4. Gie GA, Linder L, Ling RSM, et al: Impacted cancellous allografts and cement for revision total hip arthroplasty. J Bone Joint Surg Br 75:14, 1993 5. Elting JJ, Mikhail WEM, Zicat BA, et al: Preliminary report of impaction grafting for exchange femoral arthroplasty. Clin Orthop 319:159, 1995 6. Meding JB, Ritter MA, Keating EM, Faris PM: Impaction bone-grafting before insertion of a femoral stem with cement in revision total hip arthroplasty: a minimum two-year follow-up study. J Bone Joint Surg Am 79:1834, 1997 7. Duncan CP, Masterson EL, Masri BA: Impaction allografting with cement for the management of femoral bone loss. Orthop Clin North Am 29:297, 1998 8. Oonishi H, Kushitani S, Iwaki H, et al: Long term clinical results of using HAp in revision of total hip replacement involving massive bone defects in the acetabulum. Bioceramics 8:157, 1995 9. Oonishi H, Iwaki Y, Kin N, et al: Hydroxyapatite in revision of total hip replacements with massive acetabular defects: 4- to 10-year clinical results. J Bone Joint Surg Br 79:87, 1997 ¨ nnerfa¨lt R: 10. Franze´n H, Toksvig-Larsen S, Lidgren L, O Early migration of femoral components revised with impacted cancellous allografts and cement: a preliminary report of five patients. J Bone Joint Surg Br 77:862, 1995
11. Eldridge JDJ, Smith EJ, Hubble MJ, et al: Massive early subsidence following femoral impaction grafting. J Arthroplasty 12:535, 1997 12. Lamerigts N, Aspenberg P, Buma P, et al: The repeated sampling bone chamber: a new permanent titanium implant to study bone grafts in the goat. Lab Anim Sci 47:401, 1997 13. Tagil M, Aspenberg P: Impaction of cancellous bone grafts impairs osteoconduction: a titanium study in rats. Trans Orthop Res Soc 44:819, 1998 14. Hvid I, Jensen J: Cancellous bone strength at the proximal human tibia. Eng Med 13:21, 1984 15. Ling RSM, Timperley AJ, Linder L: Histology of cancellous impaction grafting in the femur: a case report. J Bone Joint Surg Br 75:693, 1993 16. Heekin RD, Engh CA, Vinh T: Morselized allograft in acetabular reconstruction: a postmortem retrieval analysis. Clin Orthop 319:184, 1995 17. Nelissen RGHH, Bauer TW, Weidenhielm LRA, et al: Revision hip arthroplasty with the use of cement and impaction grafting: histological analysis of four cases. J Bone Joint Surg Am 77:412, 1995 18. Buma P, Lamerigts N, Schreurs BW, et al: Impacted graft incorporation after cemented acetabular revision: histological evaluation in 8 patients. Acta Orthop Scand 67:536, 1996 19. Ullmark G, Linder L: Histology of the femur after cancellous impaction grafting using a Charnley prosthesis. Arch Orthop Trauma Surg 117:170, 1998 20. Schreurs BW, Buma P, Huiskes R, et al: Morsellized allografts for fixation of the hip prosthesis femoral component: a mechanical and histological study in the goat. Acta Orthop Scand 65:267, 1994 21. Schreurs BW, Huiskes R, Buma P, Slooff TJJH: Biomechanical and histological evaluation of a hydroxyapatite-coated titanium femoral stem fixed with an intramedullary morsellized bone grafting technique: an animal experiment on goats. Biomaterials 17:1177, 1996 22. Schimmel JW, Buma P, Versleyen D, et al: Acetabular reconstruction with impacted morselized cancellous allografts in cemented hip arthroplasty: a histological and biomechanical study on the goat. J Arthroplasty 13:438, 1998 23. Van der Donk S, Buma P, Schreurs BW, Aspenberg P: Tissue ingrowth in bone chambers depends on species and type of chamber: a study in rats and goats. Trans Orthop Res Soc 45:752, 1999 24. Berzins A, Sumner DR, Wasielewski RC, Galante JO: Impacted particulate allograft for femoral revision total hip arthroplasty: in vitro mechanical stability and effects of cement pressurization. J Arthroplasty 11:500, 1996 25. Malkani A, Voor MJ, Fee KA, Bates CS: Femoral component revision using impacted morsellised cancellous graft: a biomechanical study of implant stability. J Bone Joint Surg Br 78:973, 1996
Impacted Allografts and Hydroxyapatite Granules ● 26. Brodt MD, Swan CC, Brown TD: Mechanical behavior of human morselized cancellous bone in triaxial compression testing. J Orthop Res 16:43, 1998 27. Brewster NT, Gillespie WJ, Howie CR, et al: Mechanical considerations in impaction bone grafting. J Bone Joint Surg Br 81:118, 1999 28. Verdonschot N, Schreurs BW, Van Unen JMJ, et al: Cup stability after acetabulum reconstruction with
Yano et al.
643
morsellized grafts is less surgical dependent when larger grafts are used. Trans Orthop Res Soc 45:867, 1999 29. Kadoya K, Revell PA, Al-Saffer N, et al: Bone formation and bone resorption in failed total joint arthroplasties: histomorphometric analysis with histochemical and immunohistochemical technique. J Orthop Res 14:473, 1996
Histologic and Mechanical Evaluation of Impacted Morcellized Cancellous Allografts in Rabbits Comparison With Hydroxyapatite Granules H. Yano, MD,* H. Ohashi, MD, PhD,* Y. Kadoya, MD, PhD,* A. Kobayashi, MD, PhD,* Y. Yamano, MD, PhD,* and Y. Tanabe, PhD†
Abstract: The bioactivity and mechanical properties of morcellized allografts and hydroxyapatite (HA) granules were evaluated in a rabbit model. Allografts were replaced by viable trabecular structures within 8 weeks. The yield strength and stiffness of allografts were within normal cancellous bone levels by 3 weeks and were maintained afterward. The amount of newly formed bone around HA granules was comparable to that around allografts. The yield strength and stiffness of HA granules were significantly higher than those of allografts at 3 and 12 weeks. Allografts offer the advantage of being replaced by host–bone without significant deterioration in mechanical properties over the course of remodeling. HA granules can also be used for a bone substitute given their bioactivity in bone conduction and superiority in mechanical properties to allografts. Key words: allograft, hydroxyapatite, impaction grafting, bone formation, mechanical properties, animal model.
In revision total joint arthroplasties, reconstruction of joints with extensive bone loss presents a major challenge. Impaction grafting using morcellized allografts was introduced for acetabular protrusion in 1984. Since then, satisfactory results have been reported with impaction allografting of morcellized allografts for acetabular [1–3] and femoral [4–7] reconstruction. Radiographically, it has been shown that impacted morcellized allografts undergo remodeling into trabecular structures with recovery of cortical bone [4,5]. Allografts have limitations for
clinical application, however, because of limited availability and disease transmission. Given the disadvantages of allograft material in clinical practice, hydroxyapatite (HA) granules are a promising material not only as a substitute for bone–graft, but also as a bioactive material for bone conduction. It has also been reported that HA granules can be used as a substitute for morcellized allografts in acetabular reconstruction [8,9]. With the increasing application of the impaction grafting technique, massive early subsidence of the stem after femoral impaction allografting has been reported [10,11]. Poor osteoconductivity of morcellized allografts has been reported in a series of animal experiments using titanium bone conduction chambers [12,13]. It is essential to evaluate the correlation of histologic findings with mechanical properties of graft materials during the course of remodeling. To our knowledge, however, there have been no in vivo studies describing the correla-
From the *Department of Orthopaedic Surgery, Osaka City University Medical School, Osaka; and †Department of Mechanical Engineering, Faculty of Engineering, Niigata University, Niigata, Japan. Submitted June 2, 1999; accepted December 7, 1999. No benefits or funds were received in support of this study. Reprint requests: H. Yano, MD, Department of Orthopaedic Surgery, Osaka City University Medical School 1-4-3 Asahimachi, Abeno-ku, Osaka 545-8585 Japan. Copyright r 2000 by Churchill Livingstonet 0883-5403/00/1505-0014$10.00/0 doi:10.1054/arth.2000.6625
635
636 The Journal of Arthroplasty Vol. 15 No. 5 August 2000 tion between histologic findings and mechanical properties of impacted cancellous allografts; consequently, the change in mechanical properties after grafting remains unknown. The aim of this study was 2-fold: first, to determine correlations between histologic findings and mechanical properties of morcellized allografts and HA granules using a cavitary bone defect model in rabbits, and second, to evaluate the usefulness of HA granules as a substitute for allograft bone chips.
Materials and Methods Preparation of Graft Material Iliac bones of 48 rabbits (donors) were harvested under sterile conditions and stored at ⫺82°C for at least 2 weeks. They were cut into small cancellous bone chips (size, 1.0–3.0 mm in diameter) just before grafting and washed with physiologic saline. HA granules (granule size, 0.9–1.5 mm in diameter; pore size, 50–300 µm; porosity, 35%–48%; sintered temperature, 1,150°C) were obtained from Sumitomo Cement Co, Ltd (Osaka, Japan). Grafting Methods Male Japanese white rabbits (body weight, approximately 3 kg) obtained from a breeder other than the donors’ were anesthetized by intramuscular injection of ketamine hydrochloride, 25 mg/kg, and xylazine, 1.5 mg/kg. Through a medial parapatellar approach, a hole (diameter, 6 mm; depth, 12 mm) was drilled in a femoral condyle (Fig. 1). The hole was rinsed with physiologic saline before implantation. The animals were divided into 3 experimental groups as follows: i) empty group—the holes were left empty; ii) allograft group—prepared cancellous allograft chips were impacted into the holes with an impactor; and iii) HA group—HA granules were impacted in the same manner. After the
aforementioned procedures, rabbits were allowed unrestricted movement in their cages. Study A: Histologic Examination Thirty-six rabbits were killed with an overdose of barbiturate at 0, 3, 8, or 12 weeks postoperatively, and the femora were resected just above the condyles. Undecalcified, methyl methacrylate–embedded specimens were prepared. Sections measuring 4 µm thick, perpendicular to the drilling axis at the midcondylar region, were stained with hematoxylin and eosin and histologically examined. Areas of grafted bone, HA granules, and newly formed bone were measured with image analysis software (NIH Image). Newly formed bone was distinguishable from the grafted bone by the presence of osteocytes in the lacunae. All measurements were confined to the central 60% of the area of the drilled hole to avoid errors in tracing the edge of the original hole. Total bone volume was defined as the sum of newly formed bone and grafted bone areas in the allograft group and as the sum of newly formed bone and HA granule areas in the HA group. The volume fractions of total bone and newly formed bone were expressed as the percentages of total bone volume and newly formed bone volume of measured hole area. Six femoral condyles in 3 untreated rabbits (male; body weight, approximately 3 kg) were examined in the same fashion and served as normal controls. Study B: Radiographs and Bone Densitometric Analysis A second group of 36 male Japanese white rabbits underwent surgery and was divided in the same fashion as in study A. The femora were resected, and soft tissue was removed. After taking radiographs, bone mineral density (BMD) in the grafted area was measured using dual-energy x-ray absorptiometry (Hologic QDR-2000, Waltham, MA). Femoral condyles were placed on a precision-milled acrylic block 4 cm in thickness. The acrylic block was used as a soft tissue substitute for accurate BMD measurement. The long axis of the hole was adjusted parallel to the x-ray beam. Ultra–highresolution scan mode (version 4.64) was used with a 13-mm collimator, 2.54-mm line spacing, and 1.27-mm resolution. Study C: Mechanical Testing
Fig. 1. Schematic anterior and medial drawings of distal femur. A hole was drilled from the medial cortex.
The specimens of study B were mounted with an epoxy-resin adhesive (SpeciFix-20, Struers A/S, Copenhagen, Denmark). Mechanical testing could not be performed on specimens in the empty group
Impacted Allografts and Hydroxyapatite Granules ●
Yano et al.
637
because the holes remained empty at all time intervals. The epoxy-resin adhesive was applied to both ends of specimens to make the surfaces parallel. Because the resin did not infiltrate through cortical bone, the strength of specimens was not affected by the mounting procedure. Rectangular test pieces, 5 ⫻ 5 ⫻ 12 mm, sandwiching the grafted area between cortical bones, were prepared (Fig. 2A). A quasistatic compressive test was performed at room temperature using an Instron-type materials testing machine (RTC-1225AS, A&D Co, Ltd, Tokyo, Japan) at a crosshead speed of 0.5 mm/min (Fig. 2B). Compressive load was applied longitudinally to the specimen under lateral confinement by a specially designed acrylic jig (Fig. 2C). All specimens were kept moist during testing. Six femoral condyles in 3 untreated rabbits (male; body weight, approximately 3 kg) were examined in the same fashion in study B & C and served as normal controls. The yield strength and stiffness of each specimen were determined from the load-displacement curves. In this study, stiffness was defined as the slope of load-displacement curves and was determined from the linear portion of the curves at the initial stage of deformation. When the load-displacement curve did not exhibit a yield point, the yield strength was defined as the turning point from linear line to nonlinear line on the load-displacement curves, according to the method proposed by Hvid and Jensen [14]. Statistical Analysis Results for each group are expressed as means ⫾ 1 standard deviation. Statistical analyses were performed with one-way analysis of variance followed by Fisher’s protected least significant difference (PLSD) posthoc test using a commercial software package (StatView 4.11, Abacus Concepts, Inc, Berkeley, CA).
Results Study A: Histologic Examination Representative low-power field photomicrographs are shown in Fig. 3. In the empty group, the holes remained vacant through 12 weeks (Fig. 3A). In the allograft group, bone formation was observed after 3 weeks, and the woven bone underwent remodeling to mature trabeculae at 8 weeks (Fig. 3B). In the HA group, bone formation was intense at 3 weeks, and fine trabeculae were observed at 8 weeks. At 12 weeks, the trabeculae were thick in the periphery of the hole, whereas most HA granules remained unreplaced (Fig. 3C). At higher mag-
Fig. 2. (A) Schematic drawing of sampling site in a condyle for the mechanical test. (B) The specimen was mounted on an Instron-type materials testing machine. (C) A quasistatic compressive test was performed. Compressive load was applied longitudinally to the specimen under lateral confinement by a specially designed acrylic jig.
638 The Journal of Arthroplasty Vol. 15 No. 5 August 2000
Fig. 3. Low-power field photomicrographs for (A) empty, (B) allograft, and (C) hydroxyapatite groups. The trabecular structure consisting of viable bone is seen after 8 weeks in the allograft group. (Hematoxylin and eosin stain.)
nification, allograft bone was surrounded by woven bone at 3 weeks (Fig. 4), and viable trabeculae had been re-established with little remnant of the allograft at 8 weeks (Fig. 5). Histomorphometric analysis revealed that the volume fraction of total bone increased at 3 weeks, then gradually decreased in 3 groups (Fig. 6). In the empty group, the volume fraction was significantly lower than normal control levels at all time intervals. In the allograft group, it increased at 3 weeks, then decreased to normal control levels at 8 weeks. In the HA group, it was significantly higher than in the other groups at all time points tested (P ⬍ .0001).
Fig. 4. High-power field photomicrograph in the allograft group at 3 weeks. Allograft chips, recognized by empty lacunae, are surrounded by woven bone. (Hematoxylin and eosin stain.)
When measurement was confined to the volume fraction of newly formed bone, the allograft and HA groups showed a similar pattern; the percentage of newly formed bone volume peaked at 3 weeks (allograft group, 14.18% ⫾ 2.72%; HA group, 18.47% ⫾ 3.52%; P ⫽ .03%), followed by a gradual decrease to approximately 10% and 14% at 12 weeks (Fig. 7). No statistically significant difference was observed between these 2 groups at 8 or 12 weeks. The volume fraction of newly formed bone
Fig. 5. Photomicrograph in the allograft group at 8 weeks. Central portion of the drilled hole is on the upper right, and peripheral portion is on the lower left. A few remnants of allografts (arrowheads) are seen only in the center. Incorporation and revascularization spread from the peripheral to the center. (Hematoxylin and eosin stain, ⫻60.)
Impacted Allografts and Hydroxyapatite Granules ●
Yano et al.
639
Study B: Radiographs and Bone Densitometric Analysis
Fig. 6. The volume fraction of total bone in the empty (䊏), allograft (䉱), and hydroxyapatite (䊉) groups (n ⫽ 6 in each). Values are mean ⫾ standard deviation. *P ⬍ .0001, **P ⬍ .01 versus normal control (夹) (n ⫽ 6).
was significantly greater in the allograft and HA groups than that in the empty group at all 3 time periods (3, 8, and 12 weeks). Table 1 shows newly formed bone volume as a percentage of total bone volume. In the allograft group, it was 60.2% ⫾ 11.6% at 3 weeks and increased significantly to 93.0% ⫾ 6.1% at 12 weeks. In the HA group, it was 32.6% ⫾ 6.1% at 3 weeks and gradually decreased to 25.7% ⫾ 3.3% at 12 weeks.
In the empty group, no significant radiographic change was observed in the drilled hole through the experimental period (Fig. 8). In the allograft group, the margin of the hole became unclear, and reorganization of the trabeculae was observed at 8 weeks. These findings correspond to the histologic observation of incorporation of the grafted bone and trabecular bone formation from the peripheral to central. At 12 weeks, the original hole was virtually indistinguishable from the surrounding intact bone. In the HA group, intense bone formation was evident within the grafted and surrounding bone areas at 3 weeks. Subsequently, the radiodensity decreased, and HA granules became partially visible after 8 weeks. In each group, the BMD closely correlated with the volume fraction of total bone in the histomorphometric analysis (Fig. 9). Study C: Mechanical Testing Representative load-displacement curves for the allograft and HA groups are shown in Fig. 10. Yield points observed commonly in normal controls were identified in all specimens except for those of the allograft group at 3 weeks. In the allograft group, the yield strength and stiffness were within normal control levels at 3 weeks and were maintained through 12 weeks (Figs. 11 and 12). The yield strength and stiffness in the HA group were significantly higher than those in the allograft group at 3 and 12 weeks but decreased to normal control levels at 8 weeks.
Discussion
Fig. 7. The volume fraction of newly formed bone in the empty (䊏), allograft (䉱), and hydroxyapatite (䊉) groups (n ⫽ 6 in each). Values are mean ⫾ standard deviation. *P ⫽ .03 versus allograft group.
The validity of the impaction allografting technique has been shown previously either histologically or mechanically. Histologically, grafted bone chips were shown to be replaced by host–bone in human retrieval studies [15–19] and animal experiments [12,13,20–23]. Mechanically, initial fixation of the femoral stem with this technique was examined using cadavers [24,25]. The mechanical properties of morcellized cancellous allografts themselves also have been reported [26–28]. Most such studies have not correlated the findings of histologic and mechanical analyses, however, and the possibility of deterioration in strength of impacted morcellized allografts in the course of incorporation has not been tested. To our knowledge, this is the first in vivo study to evaluate the relationships between histo-
640 The Journal of Arthroplasty Vol. 15 No. 5 August 2000 Table 1. Newly Formed Bone Volume as a Percentage of Total Bone Volume*
Allograft group Hydroxyapatite group
3 Weeks
8 Weeks
12 Weeks
60.2% ⫾ 11.6%† 32.6% ⫾ 6.1%
85.0% ⫾ 4.1% 28.0% ⫾ 9.7%
93.0% ⫾ 6.1% 25.7% ⫾ 3.3%
*Values are mean ⫾ standard deviation, n ⫽ 6 in each. †P ⬍ .0001 versus allograft group at 8 and 12 weeks.
logic findings and mechanical properties of impacted morcellized allografts themselves. The first point shown in this study was that morcellized cancellous allografts could be replaced by host–bone without significant deterioration in their mechanical properties during the course of remodeling. In human retrieval studies, revascularization of morcellized allografts took at least 4 months, and conversion of the graft to vital trabecular structure required about 8 months [18]. In clinical situations, a substantially longer period might be required for revascularization and remodeling. Maintenance of mechanical strength during the remodeling process is important for impacted allografts. In our study, grafted bone was extensively remodeled into reorganized viable trabecular structure after 8 weeks. The distinct yield point observed commonly in normal controls was also recognized
Fig. 8. Representative radiographs for (A) empty, (B) allograft, and (C) hydroxyapatite groups. In the allograft group, consolidation of grafted bone is recognized at 8 weeks.
after 8 weeks. These findings indicated that 8 weeks is required for allografts to re-establish normal cancellous bone structure. At 3 weeks, the mixed structure of woven bone and grafted chips did not show a yield point and was considered to be structurally immature. By 8 weeks, the yield strength and stiffness of allografts were within the range of normal cancellous bone, indicating that the mechanical properties of allografts were sustained in the course of remodeling. The second point suggested by this study is the possible use of HA granules as a substitute for morcellized allografts. The BMD and volume fraction of total bone in the HA group were significantly higher than in the other groups at all time points tested, and extensive bone apposition onto the granules indicative of the osteoconductive effect of HA granules was observed at 3 weeks. The yield strength and stiffness were significantly higher than those in the allograft group at 3 and 12 weeks but dropped to normal control levels at 8 weeks. This pattern may be attributable to the resorption of newly formed bone during remodeling secondary to the absence of the direct weight bearing in this
Fig. 9. Bone mineral density (BMD) in the empty (䊏), allograft (䉱), and hydroxyapatite (䊉) groups (n ⫽ 6 in each). Values are mean ⫾ standard deviation. *P ⬍ .0001, **P ⫽ .01 versus normal control (夹) (n ⫽ 6).
Impacted Allografts and Hydroxyapatite Granules ●
Yano et al.
641
Fig. 10. Representative loaddisplacement curves for the (A) allograft and (B) hydroxyapatite (HA) groups. Yield points (arrowheads) as observed commonly in normal controls are recognized except for 3-week specimen in the allograft group.
conditions may affect new bone formation and mechanical properties. The lack of direct loading may be 1 reason that the newly formed bone was gradually absorbed in the allograft and HA groups. The sequence of histologic events that occurred in our specimens was identical, however, to that observed in the process of incorporation of impacted morcellized allografts in loaded animal studies [20,21]. The physiologic mechanical stress may have affected the incorporation and remodeling process of allografts in our model because the yield strength and stiffness of allografts were similar to those of normal cancellous bone during the course of remodeling.
study. Subsequently, the fine trabecular structure observed at 8 weeks was remodeled further to a matured thick trabecular structure at 12 weeks. This matured thick trabecular structure was stronger than the fine trabecular structure observed at 8 weeks, as indicated by the increase of stiffness and yield strength between 8 and 12 weeks. One limitation of this study is that bone defects were created just before grafting and surrounded by cancellous bone with abundant blood supply. These conditions are not equivalent to those of bone defects encountered in revision surgery. In a study of biopsy specimens obtained at revision operations, 1 of us previously showed that new bone often forms over much of the original bone surface of the implant–bone interface [29]. The results obtained here have meaningful implications for revision surgery. Another limitation is that there was no direct loading of grafts in this study. These mechanical
We investigated the bioactivity and mechanical properties of impacted morcellized cancellous allo-
Fig. 11. Yield strength in the allograft (䉱) and hydroxyapatite (䊉) groups (n ⫽ 6 in each). Values are mean ⫾ standard deviation. *P ⬍ .0001 versus allograft group and normal control (夹) (n ⫽ 6), **P ⫽ .04 versus allograft group.
Fig. 12. Stiffness in the allograft (䉱) and hydroxyapatite (䊉) groups (n ⫽ 6 in each). Values are mean ⫾ standard deviation. *P ⬍ .0001 versus allograft group and normal control (夹) (n ⫽ 6).
Conclusion
642 The Journal of Arthroplasty Vol. 15 No. 5 August 2000 grafts and HA granules. Morcellized cancellous allografts offer the advantage of being replaced by host– bone, while maintaining yield strength and stiffness. HA granules can also be used as a bone substitute, given their bioactivity in bone conduction and superiority in mechanical properties to allografts.
Acknowledgment We thank Mr H. Maki and Mr T. Meguro, of the Department of Mechanical Engineering, Faculty of Engineering, Niigata University, for their assistance.
References 1. Slooff TJJH, Huiskes R, Horn JV, Lemmens AJ: Bone grafting in total hip replacement for acetabular protrusion. Acta Orthop Scand 55:593, 1984 2. Azuma T, Yasuda H, Okagaki K, Sakai K: Compressed allograft chips for acetabular reconstruction in revision hip arthroplasty. J Bone Joint Surg Br 76:740, 1994 3. Schreurs BW, Slooff TJJH, Buma P, et al: Acetabular reconstruction with impacted morsellised cancellous bone graft and cement: a 10- to 15-year follow-up of 60 revision arthroplasties. J Bone Joint Surg Br 80:391, 1998 4. Gie GA, Linder L, Ling RSM, et al: Impacted cancellous allografts and cement for revision total hip arthroplasty. J Bone Joint Surg Br 75:14, 1993 5. Elting JJ, Mikhail WEM, Zicat BA, et al: Preliminary report of impaction grafting for exchange femoral arthroplasty. Clin Orthop 319:159, 1995 6. Meding JB, Ritter MA, Keating EM, Faris PM: Impaction bone-grafting before insertion of a femoral stem with cement in revision total hip arthroplasty: a minimum two-year follow-up study. J Bone Joint Surg Am 79:1834, 1997 7. Duncan CP, Masterson EL, Masri BA: Impaction allografting with cement for the management of femoral bone loss. Orthop Clin North Am 29:297, 1998 8. Oonishi H, Kushitani S, Iwaki H, et al: Long term clinical results of using HAp in revision of total hip replacement involving massive bone defects in the acetabulum. Bioceramics 8:157, 1995 9. Oonishi H, Iwaki Y, Kin N, et al: Hydroxyapatite in revision of total hip replacements with massive acetabular defects: 4- to 10-year clinical results. J Bone Joint Surg Br 79:87, 1997 ¨ nnerfa¨lt R: 10. Franze´n H, Toksvig-Larsen S, Lidgren L, O Early migration of femoral components revised with impacted cancellous allografts and cement: a preliminary report of five patients. J Bone Joint Surg Br 77:862, 1995
11. Eldridge JDJ, Smith EJ, Hubble MJ, et al: Massive early subsidence following femoral impaction grafting. J Arthroplasty 12:535, 1997 12. Lamerigts N, Aspenberg P, Buma P, et al: The repeated sampling bone chamber: a new permanent titanium implant to study bone grafts in the goat. Lab Anim Sci 47:401, 1997 13. Tagil M, Aspenberg P: Impaction of cancellous bone grafts impairs osteoconduction: a titanium study in rats. Trans Orthop Res Soc 44:819, 1998 14. Hvid I, Jensen J: Cancellous bone strength at the proximal human tibia. Eng Med 13:21, 1984 15. Ling RSM, Timperley AJ, Linder L: Histology of cancellous impaction grafting in the femur: a case report. J Bone Joint Surg Br 75:693, 1993 16. Heekin RD, Engh CA, Vinh T: Morselized allograft in acetabular reconstruction: a postmortem retrieval analysis. Clin Orthop 319:184, 1995 17. Nelissen RGHH, Bauer TW, Weidenhielm LRA, et al: Revision hip arthroplasty with the use of cement and impaction grafting: histological analysis of four cases. J Bone Joint Surg Am 77:412, 1995 18. Buma P, Lamerigts N, Schreurs BW, et al: Impacted graft incorporation after cemented acetabular revision: histological evaluation in 8 patients. Acta Orthop Scand 67:536, 1996 19. Ullmark G, Linder L: Histology of the femur after cancellous impaction grafting using a Charnley prosthesis. Arch Orthop Trauma Surg 117:170, 1998 20. Schreurs BW, Buma P, Huiskes R, et al: Morsellized allografts for fixation of the hip prosthesis femoral component: a mechanical and histological study in the goat. Acta Orthop Scand 65:267, 1994 21. Schreurs BW, Huiskes R, Buma P, Slooff TJJH: Biomechanical and histological evaluation of a hydroxyapatite-coated titanium femoral stem fixed with an intramedullary morsellized bone grafting technique: an animal experiment on goats. Biomaterials 17:1177, 1996 22. Schimmel JW, Buma P, Versleyen D, et al: Acetabular reconstruction with impacted morselized cancellous allografts in cemented hip arthroplasty: a histological and biomechanical study on the goat. J Arthroplasty 13:438, 1998 23. Van der Donk S, Buma P, Schreurs BW, Aspenberg P: Tissue ingrowth in bone chambers depends on species and type of chamber: a study in rats and goats. Trans Orthop Res Soc 45:752, 1999 24. Berzins A, Sumner DR, Wasielewski RC, Galante JO: Impacted particulate allograft for femoral revision total hip arthroplasty: in vitro mechanical stability and effects of cement pressurization. J Arthroplasty 11:500, 1996 25. Malkani A, Voor MJ, Fee KA, Bates CS: Femoral component revision using impacted morsellised cancellous graft: a biomechanical study of implant stability. J Bone Joint Surg Br 78:973, 1996
Impacted Allografts and Hydroxyapatite Granules ● 26. Brodt MD, Swan CC, Brown TD: Mechanical behavior of human morselized cancellous bone in triaxial compression testing. J Orthop Res 16:43, 1998 27. Brewster NT, Gillespie WJ, Howie CR, et al: Mechanical considerations in impaction bone grafting. J Bone Joint Surg Br 81:118, 1999 28. Verdonschot N, Schreurs BW, Van Unen JMJ, et al: Cup stability after acetabulum reconstruction with
Yano et al.
643
morsellized grafts is less surgical dependent when larger grafts are used. Trans Orthop Res Soc 45:867, 1999 29. Kadoya K, Revell PA, Al-Saffer N, et al: Bone formation and bone resorption in failed total joint arthroplasties: histomorphometric analysis with histochemical and immunohistochemical technique. J Orthop Res 14:473, 1996