- Email: [email protected]
Histological and biomechanical studies of two bone colonizable cements in rabbits
Bone Vol. 25, No. 2, Supplement August 1999:41S– 45S
Histological and Biomechanical Studies of Two Bone Colonizable Cements in Rabbits J. X. LU,1 I. ABOUT,1 G. STEPHAN,1 P. VAN LANDUYT,2 J. DEJOU,1 M. FIOCCHI,3 J. LEMAˆITRE,2 and J-P. PROUST1 1
Laboratoire Interface Matrice Extracellulaire Biomate´riaux, Faculte´ d’Odontologie, Marseille, France Laboratoire de Technologie des Poudres-Ecole Federale Polytechnique de Lausanne, Lausanne, Switzerland 3 De´partement Me´canique et Ge´nie Civil, Ecole Sure´rieure d’Inge´nieurs de Marseille, Marseille, France 2
tissue at the cement-bone interface,4,7 and adverse biological responses.1,3 In addition, it is neither bioresorbable nor colonizable by bone tissue. Moreover, micromovements at the cementbone interface13 and insufficient fatigue strength of PMMA bone cement itself were reported.5,18 Because of these inconveniencies, some other calcium phosphate cements were abundantly studied over the past 20 years. These studies demonstrated an excellent biocompatibility and biodegradation as well as a slight exothermic reaction. However, the mechanical properties of these calcium phosphate cements are weaker than those of PMMA cement.8,15 The objective of this work was to test new bone cements designed to replace the PMMA cement that are biocompatible, are colonizable, and present good biomechanical properties and a slight exothermic reaction in order to be used either in orthopedics or odontology. Two bone-colonizable cements were studied histologically and biomechanically: a partially resorbable bisphenol-␣-glucidyl methacrylate (Bis-GMA)-based cement (PRC) and a calcium phosphate cement (CPC).
We have developed two colonizable bone cements: the first is a partially resorbable bisphenol-␣-glycidyl methacrylate (Bis-GMA)-based cement (PRC) and the second is a calcium phosphate cement (CPC). PRC is composed of aluminous silanized ceramic and particles of a bioresorbable polymer embedded in a matrix of Bis-GMA. CPC consisted of tricalcium phosphate, monocalcium phosphate monohydrate, dicalcium phosphate dihydrate, and xanthane. Both cements were implanted into cavities drilled in rabbit femoral and tibial condyles. After 2, 4, 12, and 24 weeks of implantation, histological observations and biomechanical tests were performed. With CPC, a progressive osteointegration with a concomitant biodegradation in the presence of macrophages were observed. The mechanical study revealed a decrease of the compressive strength until the 4th week, followed by a slight increase. There was a general decrease in the elastic modulus with time. Moreover, by week 4, the histological study showed that the new bone was in direct contact with CPC margins. No inflammation was observed during the observation period. With PRC, the osteointegration as well as the biodegradation were slight, but its compressive strength was higher than that of cancellous bone and CPC (p < 0.05) at all observation periods. Its elastic modulus was greater than that of cancellous bone and CPC until the 4th week, then fell under the values of the cancellous bone. (Bone 25: 41S– 45S; 1999) © 1999 by Elsevier Science Inc. All rights reserved.
Materials and Methods Bone Cements PRC is composed of mineral particles (0.1–5 m) of aluminous silanized ceramic (40.60% wt) and beads (100 –300 m) of a bioresobable polymer (23.67% wt) embedded in an organic matrix (35.73% wt) of Bis-GMA, (Degussa, Frankfurt, Germany). CPC consists of -tricalcium phosphate [54.70% wt, -TCP, Ca3(PO4)2] and monocalcium phosphate monohydrate [13.3% wt, MCPM, Ca(H2PO4)2 䡠 H2O] that transforms into brushite (CaHPO4 䡠 H2O) in aqueous solutions. This cement contains 12% wt of granules (about 300 m) of -TCP and a weak quantity of xanthan gum to improve the injectability.
Key Words: Bone cement; Bisphenol-␣-glycidyl methacrylate; Calcium phosphate; Rabbit; Osteointegration; Biodegradation; Biomechanics. Introduction Polymethylmethacrylate (PMMA) bone cement has been used in surgery as an artificial bone-filling cement for the fixation of prostheses for about 30 years. It presents good mechanical properties, but its polymerization is exothermic (66 – 83°C)6 and the monomers can provoke bone necrosis,9 formation of fibrous
Animal Experiments Twenty-four adult New Zealand white rabbits weighting from 3.0 to 4.5 kg were used. They were kept in wire cages under standardized conditions and received tap water ad libitum. The rabbits were assigned to one of four groups: the 2-, 4-, 12-, and 24-week groups. Each group consisted of six rabbits, and each rabbit was used for the two cements (PRC and CPC). Animals were operated on under general anesthesia performed with intravenous injection of 1.25 mg/kg of ketamine
Address for correspondence and reprints: J. X. Lu, Ph.D., M.D., Laboratoire Interface Matrice Extracellulaire Biomate´riaux (IMEB), Faculte´ d’Odontologie, 27 Boulevard Jean Moulin, 13385 Marseille Cedex 05, France. E-mail: [email protected] © 1999 by Elsevier Science Inc. All rights reserved.
41S
8756-3282/99/$20.00 PII S8756-3282(99)00131-3
42S
J. X. Lu et al. Bone-colonizable cements in rabbits
Bone Vol. 25, No. 2, Supplement August 1999:41S– 45S
(Ketalar, Parke-Davis, UK) and 6.25 mg/kg of chlorpromazine (Largactyl, Spe´cia Rhoˆne Poulence, Paris, France) under aseptic conditions. The rabbits were operated on through the external lateral knee. A cavity of 6 mm in diameter and 12 mm in depth was drilled manually in the femoral and tibial condyles. The cavities were drilled perpendicular to the longitudinal and sagital axes of the femur or the tibia. After carefully washing with a physiological saline solution, the cavities were filled with PRC on one side and with CPC on the other side. The cavities were randomly filled with one of the cements. Surgeries on the bilateral legs were performed at the same time. Fourteen days before death, a first label was performed by an intramuscular injection of 25 mg/kg of oxytetracycline (Vibraveineuse, Pfizer, Orsay, France). Two days before death, a second label was performed by an intravenous injection of 30 mg/kg of alizarin complexon (Sigma Chemical Co., St. Louis, MO). All animals were killed by an overdose of thiopental sodium (Nesdonal, Vetoquinol, Magny-Vernios-Lure, France), then the femoral and tibial condyles were removed. Histological Techniques Soft tissues were removed and the tibial condyles fixed in 4% neutral buffered formaldehyde (pH 7.2) for 2 weeks, then rinsed for 12 h under tap water. They were dehydrated in successive alcohol concentrations (80% to absolute) and cleared with toluene before being embedded in polymethylmethacrylate. After hardening, the blocks were debited to orient the section perpendicular to the cavity, then 200 –300-m-thick sections were cut under cooling water with a sawing microtome (Isomet 2000, Buehler, Lake Bluff, IL). The sections glued onto a plastic support were polished manually (Struers, Rodovre, Denmark) to a thickness of 50 ⫾ 5 m under cooling water. Finally, they were stained with Van Gieson’s Picro-Fuchsine. Biomechanical Techniques Soft tissues of the femoral condyles were removed. They were then frozen at ⫺20°C. The cylinders composed of the cement containing newly formed bone were obtained using a small dental hand piece and a saw blade disk of 22 mm in diameter. The staining with Alizarin complex pointed out the border of the cement and the original bone. They were shaped into cylinders 6 mm in diameter and 12 mm in height and used for the mechanical tests. Their diameters and lengths were precisely measured with a gauge and recorded. For the control group, cancellous bone of the femoral condyle was shaped into a cylinder of the same size. A compression test was performed for all the six specimens of each group. Compression force was applied along their axes using an Instron-type testing machine (Adamel Lhomargy-MTS
Figure 1. After 2 weeks, CPC cylinder started to be resorbed and replaced by woven bone trabeculae (NB) on its peripheral portion. The newly formed bone did not show any direct contact (asterisk) with the cement (original magnification ⫻20).
Systemes, Ivry-Sur-Seine, France) at a cross-head speed of 5 mm/min. The values of the following two parameters were calculated from the load-deformation curve of the test: compressive strength (MPa) ⫽ failure load/compression area; elastic modulus (GPa) ⫽ stress/deformation rate. Results Histological Results (Table 1) After 2 weeks, the CPC cylinder started to be resorbed, and replaced by woven bone trabeculae on its peripheral portion. Numerous macrophages were observed. The newly formed bone did not show any direct contact with the cement (Figure 1). From 4 to 24 weeks, the newly formed bone became lamellar, and a direct contact between it and the cement increased with time, associating osteoid tissue and numerous osteoblasts (Figure 2). The cement degradation continued and paralleled the new bone formation in the majority of the cement cylinders until 24 weeks, when 60% of CPC cylinders were resorbed. Particles of the residual cement were phagocytosed by macrophages and remained inside the bone marrow tissue (Figure 3). PRC did not show any degradation of the cement or any bone
Table 1. Bone colonization and cement biodegradation after implantation
Bone colonization
2 4 12 24
weeks weeks weeks weeks
Cement biodegradation
CPC
PRC
CPC
PRC
⫺ ⫺ ⫹ ⫹
⫹ ⫹⫹ ⫹⫹⫹ ⫹⫹⫹⫹
10% 15% 30% 60%
weak weak weak weak
PRC, partially resorbable Bis-GMA-based cement; CPC, calcium phosphate cement.
Figure 2. After 24 weeks with CPC, the newly formed bone became lamellar (NB), and a direct contact (open arrow) between it and the cement increased with time, associating osteoid tissue and numerous osteoblasts (filled arrow) (original magnification ⫻200).
Bone Vol. 25, No. 2, Supplement August 1999:41S– 45S
Figure 3. After 2 weeks with CPC, particles of the residual cement were phagocytosed by macrophages (asterisk) (original magnification ⫻200).
formation from the 2nd to the 4th week. In the remaining bone fragments resulting from the cavity preparation, the osteocytes were absent or less numerous than in normal bone tissue (Figure 4). From weeks 12 to 24, a new bone tissue was formed around the PRC and increased with time, but the bone was not in direct contact with the cement and did not colonized PRC cylinders (Figure 5). Numerous macrophages were found in PRC margins (Figure 6). In CPC samples, the quantity of labeled bone increased with time from 2 to 4 weeks. The bone was more labeled by the second marker than by the first. In PRC samples, a decrease of the first labeling and an increase of the second labeling were found with time. The labeling of CPC was higher than that of PRC (Figure 7).
J. X. Lu et al. Bone-colonizable cements in rabbits
43S
Figure 5. After 24 weeks with PRC, in the remaining bone fragments resulting from the cavity preparation, the osteocytes were absent or less numerous than in normal bone tissue. The bone was not in direct contact (asterisk) with the cement and did not colonized PRC cylinders (original magnification ⫻200).
the values of the cancellous bone. The elastic modulus of CPC was equivalent to that of the cancellous bone before implantation, but became weaker after its implantation (p ⬍ 0.05) as compared with PRC and cancellous bone (Figure 9). Discussion
The compressive strength of PRC was higher than that of CPC and normal cancellous bone whatever the interval after implantation (p ⬍ 0.05). However, the strength of CPC before implantation was higher than that of the cancellous bone (p ⬍ 0.05). Compressive strength of CPC decreased until the 4th week, then increased from the 12th week, but always remained under the values of the cancellous bone (Figure 8). The elastic modulus of PRC was greater than that of the cancellous bone until the 4th week (p ⬍ 0.05), then it fell under
In this study, two bone cements were histologically and biomechanically investigated after implantation in rabbit femoral and tibial condyles. CPC already had good osteointegration at 4 weeks. More than 60% of the CPC has been resorbed after 24 weeks, mainly by macrophages. Its biomechanical properties (compressive strength and elastic modulus) were weak. In spite of an important decrease of the elastic modulus after 12 weeks, PRC had good biomechanical properties. A slight new bone formation and a weak cement biodegradation were observed. Several models are used for the evaluation of biomaterials, especially in rabbit and sheep, in cortical or cancellous bone sites of the femur or tibia. The tissue reactions to calcium phosphate ceramics are significantly different with implantation in cortical, cancellous, or medullar bone sites. The bone regeneration is observed best in the cortical bone, and it is better in cancellous bone than in the bone medullar tissue. Material biodegradation is
Figure 4. After 2 weeks, PRC did not show any degradation of the cement or any bone formation. Remaining bone fragments resulting from the cavity preparation (asterisk) (original magnification ⫻20).
Figure 6. After 12 weeks, numerous macrophages and a foreign body giant cell (asterisk) were found in PRC margins (original magnification ⫻200).
Biomechanical Results
44S
J. X. Lu et al. Bone-colonizable cements in rabbits
Bone Vol. 25, No. 2, Supplement August 1999:41S– 45S
Figure 7. Quantity of labeled bone. Figure 9. Elastic modulus.
highest in the bone medullar tissue and greater in the cancellous bone than in the cortical bone. A cancellous bone site is a suitable site for the evaluation of the biomaterials’ biofunctionality because of the equilibrium between osteogenesis and biodegradation activities.11 Therefore, the cancellous bone site (femoral and tibial condyles) was chosen to implant the bone cements for this study. Moreover, one of the cements has been implanted in femoral and tibial condyles of one side of the rabbit and the other cement in the other side. Femoral samples were used for biomechanical tests and the tibial samples for the histological investigation. This method allows obtaining of the biomechanical and histological results in the same animal and same leg to avoid the influence of these factors. Though the tibial condyle is smaller
Figure 8. Compressive strength.
than the femoral condyle, there is no difficulty performing the cavity in the tibial condyle. Since Bowen2 reported that Bis-GMA composite was an appropriate restorative material, it was largely used in dentistry. Although the strength and solubility of this material are superior to any other type of methacrylate cement, it is not frequently used by orthopedic surgeons, because of the film viscosity, short working time, difficulty in the removal of excess material, and irritation.4 From a biological point of view, it is not resorbable and not bone colonizable. We have developed a cement that has the potential to be partially resorbed, made of mineral particles of aluminous silanized ceramic and granules of 100 –300 m of a bioresobable polymer embedded in an organic matrix of BisGMA. The idea is that these polymer granules could create a porosity inside the cement to allow bone colonization. However, our results did not show tissue colonization and only a new bone formation around the PRC. Its elasticity fell after the 12th week. This decrease may be due to the degradation of the bioresorbable polymer granules inside the PRC. Why was this degradation not followed by tissue colonization? We suppose there are too few pores and/or too small-sized connections between pores. The size of interconnection must be greater than 10 m for osteoblast penetration, greater than 20 m for osteoid and chondroid tissue penetrations, and greater than 50 m for bone colonization.12 Other Bis-GMA-based cements associating the matrix with hydroxyapatite,16 or CaO-SiO2-P2O5-CaF2 glass powder, or glassceramic powder,17 present neither bone colonization nor biodegradation of the material. A cross-head speed of 0.1–10 mm/min was used for mechanical tests of bone cements,8,10,17 and 20 –25.4 mm/min was recommended to measure compressive strength, and 5 ⫾ 1 mm/min for bending modulus.14 In this study, the cross-head speed of 5 mm/min was employed to measure and calculate compressive strength and elastic modulus. The results of compressive strength were 27 MPa for PRC and 11 MPa for CPC before implantation. Another study carried out in our laboratory with a cross-head speed of 20 mm/min in the same cements showed values above that with 5 mm/min.
Bone Vol. 25, No. 2, Supplement August 1999:41S– 45S
Ikenaga et al.,8 using a cross-head speed of 1 mm/min for the mechanical measurement, showed that an increase in the compressive strength and fracture toughness of calcium phosphate cement was due to massive new bone formation until the 12th week. Then, a reduction between 12 and 16 weeks was the consequence of gradual maturation of bone in accordance with mechanical needs. However, its elastic modulus always increased with time. This study shows that the compressive strength of the CPC decreased until the 4th week, then it increased with time. This increase corresponded to a bone colonization inside cement. Therefore, its elastic modulus decreased with time. According to us, the cross-head speed (1 mm/min instead of 5 mm/min) may explain, at least in part, the discordance between the results obtained by Ikenaga et al. and our results.
Acknowledgments: This work was partially supported by the Swiss Priority Program on Materials Research (PPM 4.2.D), Dr. h. c. Robert Mathys Foundation and Stratec Medical.
References 1. Amstutz, H. C., Campbell, P., Kossovsky, N., and Clarke, I. C. Mechanism and clinical significance of wear debris-induced osteolysis. Clin Orthop 276:7–18; 1992. 2. Bowen, R. L. Properties of a silica-reinforced polymer for dental restorations. J Am Dent Assoc 66:57– 64; 1963. 3. Coe, M. R., Fechner, R. E., Jeffrey, J. J., Balian, G., and Whitehill, R. Characterization of tissue from the bone- Polymethylmethacrylate interface in a rat experimental model. J Bone Joint Surg 71-A:863– 874; 1989. 4. Council on Dental Materials and Devices. Polymers used in dentistry: part II. Resins containing a Bis-GMA: coating and cementing uses. J Am Dent Assoc 90:841– 843; 1975. 5. Davies, J. P., O’Connor, D. O., Burke, D. W., Jasty, M., and Harris, W. H. The effect of centrifugation on the fatigue life of bone cement in the presence of surface irregularities. Clin Orthop 229:156 –161; 1988.
J. X. Lu et al. Bone-colonizable cements in rabbits
45S
6. Harving, S., Søballe, K., and Bu¨nger, C. A method for bone cement interface thermometry. Acta Orthop Scand 62:546 –548; 1991. 7. Herman, J. H., Sowder, W. G., Anderson, D., Appel, A. M., and Hopson, C. N. Polymethylmethacrylate-induced release of bone-resorbing factors. J Bone Joint Surg 71:1530 –1541; 1989. 8. Ikenaga, M., Hardouin, P., Lemaıˆtre, J., Andrianjatovo, H., and Flautre, B. Biomechanical characterization of a biodegradable calcium phosphate hydraulic cement: a comparison with porous biphasic calcium phosphate ceramics. J Biomed Mater Res 40:139 –144; 1998. 9. Linder, L. Tissue reaction to methyl-methacrylate monomer. Acta Orthop Scand 47:3–10; 1976. 10. Liu, C. S., and Shen, W. Effect of crystal seeding on the hydration of calcium phosphate cement. J Mater Sci: Mater Med (special) 8:803– 807; 1997. 11. Lu, J. X., Gallur, A., Flautre, B., Anselme, K., Descamps, M., Thierry, B., and Hardouin, P. Comparative study of tissue reactions to calcium phosphate ceramics among cancellous, cortical, and medullar bone sites in rabbits. J Biomed Mater Res 42:357–367; 1998. 12. Lu, J. X., Flautre, B., Anselme, K., Gallur, A., Descamps, M., Thierry, B., and Hardouin, P. Role of interconnections in porous bioceramics on bone recolonization in vitro and in vivo. J Mater Sci: Mater Med 10:111–120; 1999. 13. Noble, P. C., and Swarts, E. Penetration of acrylic bone cements into cancellous bone. Acta Orthop Scand 54:566 –573; 1983. 14. Standard ISO 5833. Implants for Surgery: Acrylic Resin Cements. Geneva, Switzerland: International Standards Organization; 1992; 13–18. 15. Ohura, K., Bohner, M., Hardouin, P., Lemaıˆtre, J., Pasquier, G., and Flautre, B. Resorption of, and bone formation from, new -tricalcium phosphate-monocalcium phosphate cements: an in vivo study. J Biomed Mater Res 30:193–200; 1996. 16. Saito, M., Maruoka, A., Mori, T., Sugano, N., and Hino, K. Experimental studies on a new bioactive bone cement: Hydroxyapatite composite resin. Biomaterials 15:156 –160; 1994. 17. Tamura, J., Kawanabe, K., Kobayashi, M., et al. Mechanical and biological properties of two types of bioactive bone cements containing MgO-CaO-SiO2P2O5-CaF2 glass and glass-ceramic powder. J Biomed Mater Res 30:85–94; 1996. 18. Wixson, R. L., Lautenschlager, E. P., and Novak, M. A. Vacuum mixing of acrylic bone cement. J Arthroplasty 2:141–149; 1987.
Histological and Biomechanical Studies of Two Bone Colonizable Cements in Rabbits J. X. LU,1 I. ABOUT,1 G. STEPHAN,1 P. VAN LANDUYT,2 J. DEJOU,1 M. FIOCCHI,3 J. LEMAˆITRE,2 and J-P. PROUST1 1
Laboratoire Interface Matrice Extracellulaire Biomate´riaux, Faculte´ d’Odontologie, Marseille, France Laboratoire de Technologie des Poudres-Ecole Federale Polytechnique de Lausanne, Lausanne, Switzerland 3 De´partement Me´canique et Ge´nie Civil, Ecole Sure´rieure d’Inge´nieurs de Marseille, Marseille, France 2
tissue at the cement-bone interface,4,7 and adverse biological responses.1,3 In addition, it is neither bioresorbable nor colonizable by bone tissue. Moreover, micromovements at the cementbone interface13 and insufficient fatigue strength of PMMA bone cement itself were reported.5,18 Because of these inconveniencies, some other calcium phosphate cements were abundantly studied over the past 20 years. These studies demonstrated an excellent biocompatibility and biodegradation as well as a slight exothermic reaction. However, the mechanical properties of these calcium phosphate cements are weaker than those of PMMA cement.8,15 The objective of this work was to test new bone cements designed to replace the PMMA cement that are biocompatible, are colonizable, and present good biomechanical properties and a slight exothermic reaction in order to be used either in orthopedics or odontology. Two bone-colonizable cements were studied histologically and biomechanically: a partially resorbable bisphenol-␣-glucidyl methacrylate (Bis-GMA)-based cement (PRC) and a calcium phosphate cement (CPC).
We have developed two colonizable bone cements: the first is a partially resorbable bisphenol-␣-glycidyl methacrylate (Bis-GMA)-based cement (PRC) and the second is a calcium phosphate cement (CPC). PRC is composed of aluminous silanized ceramic and particles of a bioresorbable polymer embedded in a matrix of Bis-GMA. CPC consisted of tricalcium phosphate, monocalcium phosphate monohydrate, dicalcium phosphate dihydrate, and xanthane. Both cements were implanted into cavities drilled in rabbit femoral and tibial condyles. After 2, 4, 12, and 24 weeks of implantation, histological observations and biomechanical tests were performed. With CPC, a progressive osteointegration with a concomitant biodegradation in the presence of macrophages were observed. The mechanical study revealed a decrease of the compressive strength until the 4th week, followed by a slight increase. There was a general decrease in the elastic modulus with time. Moreover, by week 4, the histological study showed that the new bone was in direct contact with CPC margins. No inflammation was observed during the observation period. With PRC, the osteointegration as well as the biodegradation were slight, but its compressive strength was higher than that of cancellous bone and CPC (p < 0.05) at all observation periods. Its elastic modulus was greater than that of cancellous bone and CPC until the 4th week, then fell under the values of the cancellous bone. (Bone 25: 41S– 45S; 1999) © 1999 by Elsevier Science Inc. All rights reserved.
Materials and Methods Bone Cements PRC is composed of mineral particles (0.1–5 m) of aluminous silanized ceramic (40.60% wt) and beads (100 –300 m) of a bioresobable polymer (23.67% wt) embedded in an organic matrix (35.73% wt) of Bis-GMA, (Degussa, Frankfurt, Germany). CPC consists of -tricalcium phosphate [54.70% wt, -TCP, Ca3(PO4)2] and monocalcium phosphate monohydrate [13.3% wt, MCPM, Ca(H2PO4)2 䡠 H2O] that transforms into brushite (CaHPO4 䡠 H2O) in aqueous solutions. This cement contains 12% wt of granules (about 300 m) of -TCP and a weak quantity of xanthan gum to improve the injectability.
Key Words: Bone cement; Bisphenol-␣-glycidyl methacrylate; Calcium phosphate; Rabbit; Osteointegration; Biodegradation; Biomechanics. Introduction Polymethylmethacrylate (PMMA) bone cement has been used in surgery as an artificial bone-filling cement for the fixation of prostheses for about 30 years. It presents good mechanical properties, but its polymerization is exothermic (66 – 83°C)6 and the monomers can provoke bone necrosis,9 formation of fibrous
Animal Experiments Twenty-four adult New Zealand white rabbits weighting from 3.0 to 4.5 kg were used. They were kept in wire cages under standardized conditions and received tap water ad libitum. The rabbits were assigned to one of four groups: the 2-, 4-, 12-, and 24-week groups. Each group consisted of six rabbits, and each rabbit was used for the two cements (PRC and CPC). Animals were operated on under general anesthesia performed with intravenous injection of 1.25 mg/kg of ketamine
Address for correspondence and reprints: J. X. Lu, Ph.D., M.D., Laboratoire Interface Matrice Extracellulaire Biomate´riaux (IMEB), Faculte´ d’Odontologie, 27 Boulevard Jean Moulin, 13385 Marseille Cedex 05, France. E-mail: [email protected] © 1999 by Elsevier Science Inc. All rights reserved.
41S
8756-3282/99/$20.00 PII S8756-3282(99)00131-3
42S
J. X. Lu et al. Bone-colonizable cements in rabbits
Bone Vol. 25, No. 2, Supplement August 1999:41S– 45S
(Ketalar, Parke-Davis, UK) and 6.25 mg/kg of chlorpromazine (Largactyl, Spe´cia Rhoˆne Poulence, Paris, France) under aseptic conditions. The rabbits were operated on through the external lateral knee. A cavity of 6 mm in diameter and 12 mm in depth was drilled manually in the femoral and tibial condyles. The cavities were drilled perpendicular to the longitudinal and sagital axes of the femur or the tibia. After carefully washing with a physiological saline solution, the cavities were filled with PRC on one side and with CPC on the other side. The cavities were randomly filled with one of the cements. Surgeries on the bilateral legs were performed at the same time. Fourteen days before death, a first label was performed by an intramuscular injection of 25 mg/kg of oxytetracycline (Vibraveineuse, Pfizer, Orsay, France). Two days before death, a second label was performed by an intravenous injection of 30 mg/kg of alizarin complexon (Sigma Chemical Co., St. Louis, MO). All animals were killed by an overdose of thiopental sodium (Nesdonal, Vetoquinol, Magny-Vernios-Lure, France), then the femoral and tibial condyles were removed. Histological Techniques Soft tissues were removed and the tibial condyles fixed in 4% neutral buffered formaldehyde (pH 7.2) for 2 weeks, then rinsed for 12 h under tap water. They were dehydrated in successive alcohol concentrations (80% to absolute) and cleared with toluene before being embedded in polymethylmethacrylate. After hardening, the blocks were debited to orient the section perpendicular to the cavity, then 200 –300-m-thick sections were cut under cooling water with a sawing microtome (Isomet 2000, Buehler, Lake Bluff, IL). The sections glued onto a plastic support were polished manually (Struers, Rodovre, Denmark) to a thickness of 50 ⫾ 5 m under cooling water. Finally, they were stained with Van Gieson’s Picro-Fuchsine. Biomechanical Techniques Soft tissues of the femoral condyles were removed. They were then frozen at ⫺20°C. The cylinders composed of the cement containing newly formed bone were obtained using a small dental hand piece and a saw blade disk of 22 mm in diameter. The staining with Alizarin complex pointed out the border of the cement and the original bone. They were shaped into cylinders 6 mm in diameter and 12 mm in height and used for the mechanical tests. Their diameters and lengths were precisely measured with a gauge and recorded. For the control group, cancellous bone of the femoral condyle was shaped into a cylinder of the same size. A compression test was performed for all the six specimens of each group. Compression force was applied along their axes using an Instron-type testing machine (Adamel Lhomargy-MTS
Figure 1. After 2 weeks, CPC cylinder started to be resorbed and replaced by woven bone trabeculae (NB) on its peripheral portion. The newly formed bone did not show any direct contact (asterisk) with the cement (original magnification ⫻20).
Systemes, Ivry-Sur-Seine, France) at a cross-head speed of 5 mm/min. The values of the following two parameters were calculated from the load-deformation curve of the test: compressive strength (MPa) ⫽ failure load/compression area; elastic modulus (GPa) ⫽ stress/deformation rate. Results Histological Results (Table 1) After 2 weeks, the CPC cylinder started to be resorbed, and replaced by woven bone trabeculae on its peripheral portion. Numerous macrophages were observed. The newly formed bone did not show any direct contact with the cement (Figure 1). From 4 to 24 weeks, the newly formed bone became lamellar, and a direct contact between it and the cement increased with time, associating osteoid tissue and numerous osteoblasts (Figure 2). The cement degradation continued and paralleled the new bone formation in the majority of the cement cylinders until 24 weeks, when 60% of CPC cylinders were resorbed. Particles of the residual cement were phagocytosed by macrophages and remained inside the bone marrow tissue (Figure 3). PRC did not show any degradation of the cement or any bone
Table 1. Bone colonization and cement biodegradation after implantation
Bone colonization
2 4 12 24
weeks weeks weeks weeks
Cement biodegradation
CPC
PRC
CPC
PRC
⫺ ⫺ ⫹ ⫹
⫹ ⫹⫹ ⫹⫹⫹ ⫹⫹⫹⫹
10% 15% 30% 60%
weak weak weak weak
PRC, partially resorbable Bis-GMA-based cement; CPC, calcium phosphate cement.
Figure 2. After 24 weeks with CPC, the newly formed bone became lamellar (NB), and a direct contact (open arrow) between it and the cement increased with time, associating osteoid tissue and numerous osteoblasts (filled arrow) (original magnification ⫻200).
Bone Vol. 25, No. 2, Supplement August 1999:41S– 45S
Figure 3. After 2 weeks with CPC, particles of the residual cement were phagocytosed by macrophages (asterisk) (original magnification ⫻200).
formation from the 2nd to the 4th week. In the remaining bone fragments resulting from the cavity preparation, the osteocytes were absent or less numerous than in normal bone tissue (Figure 4). From weeks 12 to 24, a new bone tissue was formed around the PRC and increased with time, but the bone was not in direct contact with the cement and did not colonized PRC cylinders (Figure 5). Numerous macrophages were found in PRC margins (Figure 6). In CPC samples, the quantity of labeled bone increased with time from 2 to 4 weeks. The bone was more labeled by the second marker than by the first. In PRC samples, a decrease of the first labeling and an increase of the second labeling were found with time. The labeling of CPC was higher than that of PRC (Figure 7).
J. X. Lu et al. Bone-colonizable cements in rabbits
43S
Figure 5. After 24 weeks with PRC, in the remaining bone fragments resulting from the cavity preparation, the osteocytes were absent or less numerous than in normal bone tissue. The bone was not in direct contact (asterisk) with the cement and did not colonized PRC cylinders (original magnification ⫻200).
the values of the cancellous bone. The elastic modulus of CPC was equivalent to that of the cancellous bone before implantation, but became weaker after its implantation (p ⬍ 0.05) as compared with PRC and cancellous bone (Figure 9). Discussion
The compressive strength of PRC was higher than that of CPC and normal cancellous bone whatever the interval after implantation (p ⬍ 0.05). However, the strength of CPC before implantation was higher than that of the cancellous bone (p ⬍ 0.05). Compressive strength of CPC decreased until the 4th week, then increased from the 12th week, but always remained under the values of the cancellous bone (Figure 8). The elastic modulus of PRC was greater than that of the cancellous bone until the 4th week (p ⬍ 0.05), then it fell under
In this study, two bone cements were histologically and biomechanically investigated after implantation in rabbit femoral and tibial condyles. CPC already had good osteointegration at 4 weeks. More than 60% of the CPC has been resorbed after 24 weeks, mainly by macrophages. Its biomechanical properties (compressive strength and elastic modulus) were weak. In spite of an important decrease of the elastic modulus after 12 weeks, PRC had good biomechanical properties. A slight new bone formation and a weak cement biodegradation were observed. Several models are used for the evaluation of biomaterials, especially in rabbit and sheep, in cortical or cancellous bone sites of the femur or tibia. The tissue reactions to calcium phosphate ceramics are significantly different with implantation in cortical, cancellous, or medullar bone sites. The bone regeneration is observed best in the cortical bone, and it is better in cancellous bone than in the bone medullar tissue. Material biodegradation is
Figure 4. After 2 weeks, PRC did not show any degradation of the cement or any bone formation. Remaining bone fragments resulting from the cavity preparation (asterisk) (original magnification ⫻20).
Figure 6. After 12 weeks, numerous macrophages and a foreign body giant cell (asterisk) were found in PRC margins (original magnification ⫻200).
Biomechanical Results
44S
J. X. Lu et al. Bone-colonizable cements in rabbits
Bone Vol. 25, No. 2, Supplement August 1999:41S– 45S
Figure 7. Quantity of labeled bone. Figure 9. Elastic modulus.
highest in the bone medullar tissue and greater in the cancellous bone than in the cortical bone. A cancellous bone site is a suitable site for the evaluation of the biomaterials’ biofunctionality because of the equilibrium between osteogenesis and biodegradation activities.11 Therefore, the cancellous bone site (femoral and tibial condyles) was chosen to implant the bone cements for this study. Moreover, one of the cements has been implanted in femoral and tibial condyles of one side of the rabbit and the other cement in the other side. Femoral samples were used for biomechanical tests and the tibial samples for the histological investigation. This method allows obtaining of the biomechanical and histological results in the same animal and same leg to avoid the influence of these factors. Though the tibial condyle is smaller
Figure 8. Compressive strength.
than the femoral condyle, there is no difficulty performing the cavity in the tibial condyle. Since Bowen2 reported that Bis-GMA composite was an appropriate restorative material, it was largely used in dentistry. Although the strength and solubility of this material are superior to any other type of methacrylate cement, it is not frequently used by orthopedic surgeons, because of the film viscosity, short working time, difficulty in the removal of excess material, and irritation.4 From a biological point of view, it is not resorbable and not bone colonizable. We have developed a cement that has the potential to be partially resorbed, made of mineral particles of aluminous silanized ceramic and granules of 100 –300 m of a bioresobable polymer embedded in an organic matrix of BisGMA. The idea is that these polymer granules could create a porosity inside the cement to allow bone colonization. However, our results did not show tissue colonization and only a new bone formation around the PRC. Its elasticity fell after the 12th week. This decrease may be due to the degradation of the bioresorbable polymer granules inside the PRC. Why was this degradation not followed by tissue colonization? We suppose there are too few pores and/or too small-sized connections between pores. The size of interconnection must be greater than 10 m for osteoblast penetration, greater than 20 m for osteoid and chondroid tissue penetrations, and greater than 50 m for bone colonization.12 Other Bis-GMA-based cements associating the matrix with hydroxyapatite,16 or CaO-SiO2-P2O5-CaF2 glass powder, or glassceramic powder,17 present neither bone colonization nor biodegradation of the material. A cross-head speed of 0.1–10 mm/min was used for mechanical tests of bone cements,8,10,17 and 20 –25.4 mm/min was recommended to measure compressive strength, and 5 ⫾ 1 mm/min for bending modulus.14 In this study, the cross-head speed of 5 mm/min was employed to measure and calculate compressive strength and elastic modulus. The results of compressive strength were 27 MPa for PRC and 11 MPa for CPC before implantation. Another study carried out in our laboratory with a cross-head speed of 20 mm/min in the same cements showed values above that with 5 mm/min.
Bone Vol. 25, No. 2, Supplement August 1999:41S– 45S
Ikenaga et al.,8 using a cross-head speed of 1 mm/min for the mechanical measurement, showed that an increase in the compressive strength and fracture toughness of calcium phosphate cement was due to massive new bone formation until the 12th week. Then, a reduction between 12 and 16 weeks was the consequence of gradual maturation of bone in accordance with mechanical needs. However, its elastic modulus always increased with time. This study shows that the compressive strength of the CPC decreased until the 4th week, then it increased with time. This increase corresponded to a bone colonization inside cement. Therefore, its elastic modulus decreased with time. According to us, the cross-head speed (1 mm/min instead of 5 mm/min) may explain, at least in part, the discordance between the results obtained by Ikenaga et al. and our results.
Acknowledgments: This work was partially supported by the Swiss Priority Program on Materials Research (PPM 4.2.D), Dr. h. c. Robert Mathys Foundation and Stratec Medical.
References 1. Amstutz, H. C., Campbell, P., Kossovsky, N., and Clarke, I. C. Mechanism and clinical significance of wear debris-induced osteolysis. Clin Orthop 276:7–18; 1992. 2. Bowen, R. L. Properties of a silica-reinforced polymer for dental restorations. J Am Dent Assoc 66:57– 64; 1963. 3. Coe, M. R., Fechner, R. E., Jeffrey, J. J., Balian, G., and Whitehill, R. Characterization of tissue from the bone- Polymethylmethacrylate interface in a rat experimental model. J Bone Joint Surg 71-A:863– 874; 1989. 4. Council on Dental Materials and Devices. Polymers used in dentistry: part II. Resins containing a Bis-GMA: coating and cementing uses. J Am Dent Assoc 90:841– 843; 1975. 5. Davies, J. P., O’Connor, D. O., Burke, D. W., Jasty, M., and Harris, W. H. The effect of centrifugation on the fatigue life of bone cement in the presence of surface irregularities. Clin Orthop 229:156 –161; 1988.
J. X. Lu et al. Bone-colonizable cements in rabbits
45S
6. Harving, S., Søballe, K., and Bu¨nger, C. A method for bone cement interface thermometry. Acta Orthop Scand 62:546 –548; 1991. 7. Herman, J. H., Sowder, W. G., Anderson, D., Appel, A. M., and Hopson, C. N. Polymethylmethacrylate-induced release of bone-resorbing factors. J Bone Joint Surg 71:1530 –1541; 1989. 8. Ikenaga, M., Hardouin, P., Lemaıˆtre, J., Andrianjatovo, H., and Flautre, B. Biomechanical characterization of a biodegradable calcium phosphate hydraulic cement: a comparison with porous biphasic calcium phosphate ceramics. J Biomed Mater Res 40:139 –144; 1998. 9. Linder, L. Tissue reaction to methyl-methacrylate monomer. Acta Orthop Scand 47:3–10; 1976. 10. Liu, C. S., and Shen, W. Effect of crystal seeding on the hydration of calcium phosphate cement. J Mater Sci: Mater Med (special) 8:803– 807; 1997. 11. Lu, J. X., Gallur, A., Flautre, B., Anselme, K., Descamps, M., Thierry, B., and Hardouin, P. Comparative study of tissue reactions to calcium phosphate ceramics among cancellous, cortical, and medullar bone sites in rabbits. J Biomed Mater Res 42:357–367; 1998. 12. Lu, J. X., Flautre, B., Anselme, K., Gallur, A., Descamps, M., Thierry, B., and Hardouin, P. Role of interconnections in porous bioceramics on bone recolonization in vitro and in vivo. J Mater Sci: Mater Med 10:111–120; 1999. 13. Noble, P. C., and Swarts, E. Penetration of acrylic bone cements into cancellous bone. Acta Orthop Scand 54:566 –573; 1983. 14. Standard ISO 5833. Implants for Surgery: Acrylic Resin Cements. Geneva, Switzerland: International Standards Organization; 1992; 13–18. 15. Ohura, K., Bohner, M., Hardouin, P., Lemaıˆtre, J., Pasquier, G., and Flautre, B. Resorption of, and bone formation from, new -tricalcium phosphate-monocalcium phosphate cements: an in vivo study. J Biomed Mater Res 30:193–200; 1996. 16. Saito, M., Maruoka, A., Mori, T., Sugano, N., and Hino, K. Experimental studies on a new bioactive bone cement: Hydroxyapatite composite resin. Biomaterials 15:156 –160; 1994. 17. Tamura, J., Kawanabe, K., Kobayashi, M., et al. Mechanical and biological properties of two types of bioactive bone cements containing MgO-CaO-SiO2P2O5-CaF2 glass and glass-ceramic powder. J Biomed Mater Res 30:85–94; 1996. 18. Wixson, R. L., Lautenschlager, E. P., and Novak, M. A. Vacuum mixing of acrylic bone cement. J Arthroplasty 2:141–149; 1987.