- Email: [email protected]
Quantitative morphometric evaluation of critical size experimental bone defects by microcomputed tomography
British Journal of Oral and Maxillofacial Surgery 45 (2007) 203–207
Quantitative morphometric evaluation of critical size experimental bone defects by microcomputed tomography d, ¨ Candan Efeoglu a,∗ , Sheila E. Fisher b , Selda Ert¨urk c , Fikri Oztop Sevtap G¨unbay c , Aylin Sipahi c a
Department of Oral and Maxillofacial Surgery, Level 6 Worsley Building, Leeds Dental Institute, Clarendon Way, Leeds, West Yorkshire LS2 9LU, UK b Department of Maxillofacial Surgery, Faculty of Medicine and Health, University of Leeds, UK c Department of Oral Surgery, School of Dentistry, Ege University, Izmir, Turkey d Department of Pathology, School of Medicine, Ege University, Izmir, Turkey Accepted 27 May 2006 Available online 18 July 2006
Abstract Our aim was to show that microcomputed tomography is a useful tool for acquiring high-resolution three-dimensional tomographic images to assess bone healing, the interface with materials, and the biocompatibility of bone substitutes. Acquired images can be used for non-invasive quantitative morphometric analysis of regenerating bone, leaving the option for conventional histology to be an adjunct used at defined intervals. The temporal characterisation of the mineralisation of bone potentially has a critical role in the understanding of the dynamics of mineralisation of healing bone. This has applications both for degradable and bioactive materials and for pharmaceutical products that act on bone. Formal validation of this promising new technique will be a critical part of continuing studies. © 2006 The British Association of Oral and Maxillofacial Surgeons. Published by Elsevier Ltd. All rights reserved. Keywords: Microcomputed tomography; Histomorphometrics; Critical size bone defect; Quantitative morphometrics
Introduction Orofacial bone defects cause considerable morbidity and distress to patients. Reconstruction can be difficult or occasionally impossible using current surgical techniques. Numerous materials are under development to aid fixation and subsequent bone remodelling, but each has associated problems and disadvantages. Ethical concerns and the high costs involved in animal studies make it essential for the investigators to reduce the number of animals used in research, reduce costs, refine ∗
Corresponding author. Fax: +44 113 3436165. E-mail address: [email protected] (C. Efeoglu).
experimental settings for optimum results, and look for other methods of testing that can be substituted for in vivo testing on animals. Reduction, refinement, and replacement summarise the current approach to research on animals. Conventional histomorphometric evaluations, radiography, and microradiography have established places in the evaluation of substitutes for bone; however, they fall short of what we require. Microcomputed tomography (CT) is a non-invasive method that enables quantitative morphometric evaluation of harvested tissue in three dimensions. Variables calculated by CT have been devised that allow the interpretation of results, but they rely on characteristics of mature rather than healing bone.1–5
0266-4356/$ – see front matter © 2006 The British Association of Oral and Maxillofacial Surgeons. Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.bjoms.2006.05.007
204
C. Efeoglu et al. / British Journal of Oral and Maxillofacial Surgery 45 (2007) 203–207
Data can be saved, which gives the opportunity to develop comparative indices that can be applied to the future assessment of materials for bony reconstruction.
Materials and methods Animals Samples were obtained during a study of bony regeneration in a series of 33 New Zealand white rabbits. Full thickness critical-size calvarial defects 15 mm in diameter were prepared and implanted with a mixture of platelet rich plasma6 and -tricalcium phosphate (Vitoss® Scaffold, Orthovita, USA) under general anaesthesia, after approval by the local ethics committee on animal studies.7–12 Microcomputed tomographic analysis Animals were killed at 4, 12, and 24 weeks postoperatively. Four animals were excluded from the study because of postoperative complications including systemic upset, poor oral intake, and signs of distress, which had previously been chosen as termination points. Calvarial bone samples from 29 animals were excised, fixed with 10% formalin, and later scanned and evaluated using a CT 80 (Scanco Medical, Bassersdorf Switzerland) and the software provided by the manufacturer (Scanco Medical, Software revision 3.1). All the samples were positioned and scanned in a standard manner using an airtight cylindrical sample holder filled with formaldehyde to preserve them for the duration of the measurement. When the scan was finished, images with slices of 25 m thick and a resolution of 2048 × 2048 pixels had been obtained. On the data set, a cylindrical region of interest that corresponded to the complete thickness of the calvarial bone sample was chosen. It had a diameter of 14.8 mm to ensure that it remained within the borders of the surgical defect. The region was first positioned in the slices that best showed the borders of the surgical defect, and then it was extended to all slices of the data set. Grey values of the images are shown on a scale from −1000 to +1000 in the workstation of the CT scanner. The grey value of an image is proportional to the linear attenuation coefficient of the tissue. We used the “adaptive thresholding” procedure from the software provided and calculated
optimum threshold values (grey values) of 296 and 116 for rabbit calvarial bone samples and -tricalcium phosphate, respectively. This allowed segmentation, whereby mineralised bone was differentiated from non-mineralised tissue and -tricalcium phosphate. Quantitative morphometric analysis of the tissue inside the region was carried out on voxels that corresponded to bone (values between 296 and 1000), which excluded the voxels that corresponded to -tricalcium phosphate and non-mineralised tissues. Three-dimensional images comprising cubic voxels were obtained, and the following morphometric variables were evaluated1–4,13–16 : volume of bone in the region of interest; surface area of bone in the region of interest; connectivity density (that quantifies the connectivity of histological architecture of bone). It is calculated by subtracting “1” from the number of trabecules in mm3 . In addition, we propose a new measurement to permit assessment of the healing bone at different times. This allows characterisation of bone mineralisation in relation to changes in bone volume. This was assessed by dividing the grey values of bone into quartiles of increasing mineralisation, and then the distribution of bone volume of each animal in these quartiles was expressed as a percentage. We used the statistical package for the social sciences (SPSS 11.0 for Windows, SPSS, Chicago, IL) for statistical calculations. We applied Kolmogorov–Smirnov and Shapiro–Wilk tests of normality and then one-way analysis of variance and post hoc tests where appropriate.
Results Animals were killed at the 4th, 12th, and 24th weeks postoperatively (Table 1). There were no differences between the samples at the respective periods (figures not shown). Representative three-dimensional CT images from an unoperated sample and operated samples in each of the three groups that correspond to the surgical defect are shown in Fig. 1. All our morphometric variables were normally distributed. The median values for bone volume, bone surface, and connectivity density are shown in Fig. 2A–C, respectively. There were no differences between the groups. Table 1 shows bone mineralisation in relation to the changes in bone volume.
Table 1 Distribution of bone volume in quartiles of mineralisation Group
No
First quartile
Second quartile
Third quartile
Fourth quartile
Killed at: 4th week 12th week 24th week
10 10 9
55 (2) 54 (3) 51 (4)
26 (1) 27 (1) 27 (1)
12 (1) 12 (1) 13 (2)
8 (1) 7 (2) 9 (2)
The mean bone volume of each group is expressed as mean (S.D.)% for quartiles of increasing mineralisation.
C. Efeoglu et al. / British Journal of Oral and Maxillofacial Surgery 45 (2007) 203–207
205
Fig. 1. Representative microcomputed tomograms that correspond to the surgical defect: (A), an unoperated sample; (B), group killed 4 weeks postoperatively; (C), group killed 12 weeks postoperatively; (D), group killed 24 weeks postoperatively.
Discussion We know of few similar studies that have used CT to assess regeneration in experimental bone defects. Cacciafesta et al.,13 Lutolf et al.,15 and Verna et al.16 have reported on the critical size calvarial defect in rats, whereas Jones et al.17 used an inhouse CT to assess bone healing in pigs’ orbital walls. The thickness of slices in these images was 15, 18, 15, and 9.4 m, respectively. Lu and Rabie18 investigated the microarchitecture of healing in rabbit mandibular defects and used a slice thickness of 25 m, as we did. The -tricalcium phosphate had a grey value of 116, which approximated to the grey value of rabbit calvarium (296). For an unbiased quantitative morphometric analysis of the CT images the grey values that corresponded to -tricalcium phosphate were excluded, which left the grey values that corresponded only to bone. Jones et al.17 used a similar phase separation technique that was based on intensity histograms to differentiate their polymer scaffold from mineralised bone. Three-dimensional CT images showed that none of the defects healed completely, which means that the combination of platelet-rich plasma and -tricalcium phosphate does not
induce sufficient osteoinduction for complete closure within 6 months. Variables such as surface area and volume of bone, and connectivity density, which apply to woven bone, were calculated for quantitative morphometric analysis. An understanding of the difference on imaging between healing woven bone and trabecular bone is fundamental to the assessment of new materials and pharmaceutical products that act on bone. The most important new finding in this study is the development of a new measurement that defines the dynamics of mineralisation of healing bone. Its key components are that it attempts to assess the quantity of mineralisation against time: it is a dynamic measure, which can be compared across materials and pharmaceutically active products, and which acts on bone. Hilderbrand et al.1 have previously used volume and surface area of bone to study the effects of osteoporosis on bone samples from cadavers, and Kurth and M¨uller2 used the same measures to quantify the destructive effects of an experimental bone tumour. Connectivity density was used to quantify resorption of bone experimentally by Odgaard Agundersen.4 Another variable that defines “the amount of mineralisation” was used by Kurth and M¨uller2 to
206
C. Efeoglu et al. / British Journal of Oral and Maxillofacial Surgery 45 (2007) 203–207
Fig. 2. Diagram of the histomorphometric variables: bone volume, surface, and connectivity density: (A), scattergram showing the distribution of bone volume (the horizontal line indicates the median); (B), scattergram showing the distribution of surface area of bone (the horizontal line indicates the median); (C), scattergram showing the distribution of connectivity density (the horizontal line indicates the median).
quantify the resorption of bone from an experimental tumour. Cacciafesta et al.13 and Verna et al.16 used volume and “the amount of mineralisation in terms of grey values” to assess regeneration in critical size defects in rats. Lutolf et al.15 used connectivity density as well as volume for the same purpose. Jones et al.17 used CT data for qualitative assessment of bony ingrowth and finite element modelling of the regenerated bone. The volume of bone in a healing defect is likely to increase with time, but we found no significant difference between the groups killed at 4 and 12 weeks. Those killed at 24 weeks had a higher bone volume than the other two groups. This may be because modelling of the immature bone that had developed during the first 4 weeks continued until the end of 12 weeks, with no increase in bone volume. We acknowledge that the sample size was small, and that to achieve significance a larger sample would have been required. This study was conducted using an in vitro CT scanner. The more recently developed in vivo CT scanners permit serial imaging during healing, which increases the reliability of preclinical studies and reduces the number of animals required in the assessment of biocompatibility. Non-invasive methods for both assessment and comparison are an important part of the regulatory process and allow clinicians to make informed choices in clinical practice. In future, it is possible that CT technology will be used in patients. The effect of new pharmacological agents for the control of bone disease or to improve healing can be explored
systematically, permitting both serial and comparative studies. We hope that this technique will be suitable for adoption as a regulatory standard for quality control in the testing of bone-related pharmaceuticals, materials, and tissue engineering, thereby reducing the use of animals in research without compromising standards. Acknowledgements This study was funded by Ege University Research Fund project number 2000/DIS/014. The authors would like to thank to Dr. Richard Hall (School of Mechanical Engineering, University of Leeds, UK) for granting the use of the CT scanner; Dr. Andres Laib (Scanco Medical, Swtzerland); Dr. Ruth Wilcox, and Dr. Ruth Goodridge (School of Mechanical Engineering, University of Leeds, UK); the technical staff of Centre for Experimental Surgery and Central Research Laboratory (School of Medicine, Ege University, Turkey) for their expert technical assistance.
References 1. Hilderbrand T, Laib A, M¨uller R, Dequeker J, R¨uegsegger P. Direct three-dimensional morphometric analysis of human cancellous bone: microstructural data from spine, femur, iliac crest, and calceneus. J Bone Miner Res 1999;14:1167–75.
C. Efeoglu et al. / British Journal of Oral and Maxillofacial Surgery 45 (2007) 203–207 2. Kurth AA, M¨uller R. The effect of an osteolithic tumor on the threedimensional trabecular bone morphology in an animal model. Skeletal Radiol 2001;30:94–8. 3. Muller R, Van Campenhout H, Van Damme B, et al. Morphometric analysis of human bone biopsies: a quantitative structural comparison of histological sections and micro-computed tomography. Bone 1998;23:59–66. 4. Odgaard Agundersen HJ. Quantification of connectivity in cancellous bone, with special emphasis on 3-D reconstructions. Bone 1993;14:173–82. 5. R¨uegsegger P, Koller B, Muller R. A microtomographic system for the non-destructive evaluation of bone architecture. Calcif Tiss Int 1996;58:24–9. 6. Efeoglu C, Delen Akc¸ay Y, Ert¨vrk S. A modified method for preparing platelet rich plasma: an experimental study. J Oral Maxillofac Surg 2004;62:1403–7. 7. Beck L, Deguzman L, Lee W. Rapid publication. TGF-beta 1 induces bone closure of skull defects. J Bone Miner Res 1991;6:1257–65. 8. Frame JW. A convenient animal model for testing bone substitute materials. J Oral Surg 1980;38:176–80. 9. Kim ES, Park EJ, Choung PH. Platelet concentration and its effect on bone formation in calvarial defects: an experimental study in rabbits. J Prosthet Dent 2001;86:428–33. 10. Schmitz JP, Hollinger JO. The critical size defect as an experimental model for craniomandibulofacial nonunions. Clin Orthop Relat Res 1986;205:299–308.
207
11. Vikjaer D, Blom S, Hjorting-Hansen E, Pinholt EM. Effect of plateletderived growth factor-bb on bone formation in calvarial defects: an experimental study in rabbits. Eur J Oral Sci 1997;105:59–66. 12. Zellin G, Beck S, Hardwick R, Linde A. Opposite effects of recombinant human transforming growth factor beta1 on bone regeneration in vivo: effects of exclusion of periosteal cells by microporous membrane. Bone 1998;22:613–20. 13. Cacciafesta V, Dalstra M, Bosch C, Melsen B, Andreassen TT. Growth hormone treatment promotes guided bone regeneration in rat calvarial defects. Eur J Orthod 2001;23:733–40. 14. Gowen M, Lazner F, Dodds R, et al. Cathepsin K knockout mice develop osteoporosis due to a deficit in matrix degradation but not demineralization. J Bone Miner Res 1999;14:1654–63. 15. Lutolf MP, Weber FE, Schmoekel HG, et al. Repair of bone defects using synthetic mimetics of collageneous extracellular matrices. Nat Biotechnol 2003;21:513–8. Epub 2003 April 21. doi:10.1038/nbt818. 16. Verna C, Dalstra M, Wikesj¨o UM, Trombelli L, Bosch Carles. Healing patterns in calvarial bone defects following guided bone regeneration in rats. A micro-CT scan analysis. J Clin Periodontol 2002;29:865– 70. 17. Jones AC, Sakellariou A, Limaye A, et al. Investigation of microstructural features in regenerating bone using microcomputed tomography. J Mater Sci Mater Med 2004;15:529–32. 18. Lu M, Rabie ABM. Microarchitecture of rabbit mandibular defects grafted with intramembraneous or endochondral bone shown by microcomputed tomography. Br J Oral Maxillofac Surg 2003;41:385–91.
Quantitative morphometric evaluation of critical size experimental bone defects by microcomputed tomography d, ¨ Candan Efeoglu a,∗ , Sheila E. Fisher b , Selda Ert¨urk c , Fikri Oztop Sevtap G¨unbay c , Aylin Sipahi c a
Department of Oral and Maxillofacial Surgery, Level 6 Worsley Building, Leeds Dental Institute, Clarendon Way, Leeds, West Yorkshire LS2 9LU, UK b Department of Maxillofacial Surgery, Faculty of Medicine and Health, University of Leeds, UK c Department of Oral Surgery, School of Dentistry, Ege University, Izmir, Turkey d Department of Pathology, School of Medicine, Ege University, Izmir, Turkey Accepted 27 May 2006 Available online 18 July 2006
Abstract Our aim was to show that microcomputed tomography is a useful tool for acquiring high-resolution three-dimensional tomographic images to assess bone healing, the interface with materials, and the biocompatibility of bone substitutes. Acquired images can be used for non-invasive quantitative morphometric analysis of regenerating bone, leaving the option for conventional histology to be an adjunct used at defined intervals. The temporal characterisation of the mineralisation of bone potentially has a critical role in the understanding of the dynamics of mineralisation of healing bone. This has applications both for degradable and bioactive materials and for pharmaceutical products that act on bone. Formal validation of this promising new technique will be a critical part of continuing studies. © 2006 The British Association of Oral and Maxillofacial Surgeons. Published by Elsevier Ltd. All rights reserved. Keywords: Microcomputed tomography; Histomorphometrics; Critical size bone defect; Quantitative morphometrics
Introduction Orofacial bone defects cause considerable morbidity and distress to patients. Reconstruction can be difficult or occasionally impossible using current surgical techniques. Numerous materials are under development to aid fixation and subsequent bone remodelling, but each has associated problems and disadvantages. Ethical concerns and the high costs involved in animal studies make it essential for the investigators to reduce the number of animals used in research, reduce costs, refine ∗
Corresponding author. Fax: +44 113 3436165. E-mail address: [email protected] (C. Efeoglu).
experimental settings for optimum results, and look for other methods of testing that can be substituted for in vivo testing on animals. Reduction, refinement, and replacement summarise the current approach to research on animals. Conventional histomorphometric evaluations, radiography, and microradiography have established places in the evaluation of substitutes for bone; however, they fall short of what we require. Microcomputed tomography (CT) is a non-invasive method that enables quantitative morphometric evaluation of harvested tissue in three dimensions. Variables calculated by CT have been devised that allow the interpretation of results, but they rely on characteristics of mature rather than healing bone.1–5
0266-4356/$ – see front matter © 2006 The British Association of Oral and Maxillofacial Surgeons. Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.bjoms.2006.05.007
204
C. Efeoglu et al. / British Journal of Oral and Maxillofacial Surgery 45 (2007) 203–207
Data can be saved, which gives the opportunity to develop comparative indices that can be applied to the future assessment of materials for bony reconstruction.
Materials and methods Animals Samples were obtained during a study of bony regeneration in a series of 33 New Zealand white rabbits. Full thickness critical-size calvarial defects 15 mm in diameter were prepared and implanted with a mixture of platelet rich plasma6 and -tricalcium phosphate (Vitoss® Scaffold, Orthovita, USA) under general anaesthesia, after approval by the local ethics committee on animal studies.7–12 Microcomputed tomographic analysis Animals were killed at 4, 12, and 24 weeks postoperatively. Four animals were excluded from the study because of postoperative complications including systemic upset, poor oral intake, and signs of distress, which had previously been chosen as termination points. Calvarial bone samples from 29 animals were excised, fixed with 10% formalin, and later scanned and evaluated using a CT 80 (Scanco Medical, Bassersdorf Switzerland) and the software provided by the manufacturer (Scanco Medical, Software revision 3.1). All the samples were positioned and scanned in a standard manner using an airtight cylindrical sample holder filled with formaldehyde to preserve them for the duration of the measurement. When the scan was finished, images with slices of 25 m thick and a resolution of 2048 × 2048 pixels had been obtained. On the data set, a cylindrical region of interest that corresponded to the complete thickness of the calvarial bone sample was chosen. It had a diameter of 14.8 mm to ensure that it remained within the borders of the surgical defect. The region was first positioned in the slices that best showed the borders of the surgical defect, and then it was extended to all slices of the data set. Grey values of the images are shown on a scale from −1000 to +1000 in the workstation of the CT scanner. The grey value of an image is proportional to the linear attenuation coefficient of the tissue. We used the “adaptive thresholding” procedure from the software provided and calculated
optimum threshold values (grey values) of 296 and 116 for rabbit calvarial bone samples and -tricalcium phosphate, respectively. This allowed segmentation, whereby mineralised bone was differentiated from non-mineralised tissue and -tricalcium phosphate. Quantitative morphometric analysis of the tissue inside the region was carried out on voxels that corresponded to bone (values between 296 and 1000), which excluded the voxels that corresponded to -tricalcium phosphate and non-mineralised tissues. Three-dimensional images comprising cubic voxels were obtained, and the following morphometric variables were evaluated1–4,13–16 : volume of bone in the region of interest; surface area of bone in the region of interest; connectivity density (that quantifies the connectivity of histological architecture of bone). It is calculated by subtracting “1” from the number of trabecules in mm3 . In addition, we propose a new measurement to permit assessment of the healing bone at different times. This allows characterisation of bone mineralisation in relation to changes in bone volume. This was assessed by dividing the grey values of bone into quartiles of increasing mineralisation, and then the distribution of bone volume of each animal in these quartiles was expressed as a percentage. We used the statistical package for the social sciences (SPSS 11.0 for Windows, SPSS, Chicago, IL) for statistical calculations. We applied Kolmogorov–Smirnov and Shapiro–Wilk tests of normality and then one-way analysis of variance and post hoc tests where appropriate.
Results Animals were killed at the 4th, 12th, and 24th weeks postoperatively (Table 1). There were no differences between the samples at the respective periods (figures not shown). Representative three-dimensional CT images from an unoperated sample and operated samples in each of the three groups that correspond to the surgical defect are shown in Fig. 1. All our morphometric variables were normally distributed. The median values for bone volume, bone surface, and connectivity density are shown in Fig. 2A–C, respectively. There were no differences between the groups. Table 1 shows bone mineralisation in relation to the changes in bone volume.
Table 1 Distribution of bone volume in quartiles of mineralisation Group
No
First quartile
Second quartile
Third quartile
Fourth quartile
Killed at: 4th week 12th week 24th week
10 10 9
55 (2) 54 (3) 51 (4)
26 (1) 27 (1) 27 (1)
12 (1) 12 (1) 13 (2)
8 (1) 7 (2) 9 (2)
The mean bone volume of each group is expressed as mean (S.D.)% for quartiles of increasing mineralisation.
C. Efeoglu et al. / British Journal of Oral and Maxillofacial Surgery 45 (2007) 203–207
205
Fig. 1. Representative microcomputed tomograms that correspond to the surgical defect: (A), an unoperated sample; (B), group killed 4 weeks postoperatively; (C), group killed 12 weeks postoperatively; (D), group killed 24 weeks postoperatively.
Discussion We know of few similar studies that have used CT to assess regeneration in experimental bone defects. Cacciafesta et al.,13 Lutolf et al.,15 and Verna et al.16 have reported on the critical size calvarial defect in rats, whereas Jones et al.17 used an inhouse CT to assess bone healing in pigs’ orbital walls. The thickness of slices in these images was 15, 18, 15, and 9.4 m, respectively. Lu and Rabie18 investigated the microarchitecture of healing in rabbit mandibular defects and used a slice thickness of 25 m, as we did. The -tricalcium phosphate had a grey value of 116, which approximated to the grey value of rabbit calvarium (296). For an unbiased quantitative morphometric analysis of the CT images the grey values that corresponded to -tricalcium phosphate were excluded, which left the grey values that corresponded only to bone. Jones et al.17 used a similar phase separation technique that was based on intensity histograms to differentiate their polymer scaffold from mineralised bone. Three-dimensional CT images showed that none of the defects healed completely, which means that the combination of platelet-rich plasma and -tricalcium phosphate does not
induce sufficient osteoinduction for complete closure within 6 months. Variables such as surface area and volume of bone, and connectivity density, which apply to woven bone, were calculated for quantitative morphometric analysis. An understanding of the difference on imaging between healing woven bone and trabecular bone is fundamental to the assessment of new materials and pharmaceutical products that act on bone. The most important new finding in this study is the development of a new measurement that defines the dynamics of mineralisation of healing bone. Its key components are that it attempts to assess the quantity of mineralisation against time: it is a dynamic measure, which can be compared across materials and pharmaceutically active products, and which acts on bone. Hilderbrand et al.1 have previously used volume and surface area of bone to study the effects of osteoporosis on bone samples from cadavers, and Kurth and M¨uller2 used the same measures to quantify the destructive effects of an experimental bone tumour. Connectivity density was used to quantify resorption of bone experimentally by Odgaard Agundersen.4 Another variable that defines “the amount of mineralisation” was used by Kurth and M¨uller2 to
206
C. Efeoglu et al. / British Journal of Oral and Maxillofacial Surgery 45 (2007) 203–207
Fig. 2. Diagram of the histomorphometric variables: bone volume, surface, and connectivity density: (A), scattergram showing the distribution of bone volume (the horizontal line indicates the median); (B), scattergram showing the distribution of surface area of bone (the horizontal line indicates the median); (C), scattergram showing the distribution of connectivity density (the horizontal line indicates the median).
quantify the resorption of bone from an experimental tumour. Cacciafesta et al.13 and Verna et al.16 used volume and “the amount of mineralisation in terms of grey values” to assess regeneration in critical size defects in rats. Lutolf et al.15 used connectivity density as well as volume for the same purpose. Jones et al.17 used CT data for qualitative assessment of bony ingrowth and finite element modelling of the regenerated bone. The volume of bone in a healing defect is likely to increase with time, but we found no significant difference between the groups killed at 4 and 12 weeks. Those killed at 24 weeks had a higher bone volume than the other two groups. This may be because modelling of the immature bone that had developed during the first 4 weeks continued until the end of 12 weeks, with no increase in bone volume. We acknowledge that the sample size was small, and that to achieve significance a larger sample would have been required. This study was conducted using an in vitro CT scanner. The more recently developed in vivo CT scanners permit serial imaging during healing, which increases the reliability of preclinical studies and reduces the number of animals required in the assessment of biocompatibility. Non-invasive methods for both assessment and comparison are an important part of the regulatory process and allow clinicians to make informed choices in clinical practice. In future, it is possible that CT technology will be used in patients. The effect of new pharmacological agents for the control of bone disease or to improve healing can be explored
systematically, permitting both serial and comparative studies. We hope that this technique will be suitable for adoption as a regulatory standard for quality control in the testing of bone-related pharmaceuticals, materials, and tissue engineering, thereby reducing the use of animals in research without compromising standards. Acknowledgements This study was funded by Ege University Research Fund project number 2000/DIS/014. The authors would like to thank to Dr. Richard Hall (School of Mechanical Engineering, University of Leeds, UK) for granting the use of the CT scanner; Dr. Andres Laib (Scanco Medical, Swtzerland); Dr. Ruth Wilcox, and Dr. Ruth Goodridge (School of Mechanical Engineering, University of Leeds, UK); the technical staff of Centre for Experimental Surgery and Central Research Laboratory (School of Medicine, Ege University, Turkey) for their expert technical assistance.
References 1. Hilderbrand T, Laib A, M¨uller R, Dequeker J, R¨uegsegger P. Direct three-dimensional morphometric analysis of human cancellous bone: microstructural data from spine, femur, iliac crest, and calceneus. J Bone Miner Res 1999;14:1167–75.
C. Efeoglu et al. / British Journal of Oral and Maxillofacial Surgery 45 (2007) 203–207 2. Kurth AA, M¨uller R. The effect of an osteolithic tumor on the threedimensional trabecular bone morphology in an animal model. Skeletal Radiol 2001;30:94–8. 3. Muller R, Van Campenhout H, Van Damme B, et al. Morphometric analysis of human bone biopsies: a quantitative structural comparison of histological sections and micro-computed tomography. Bone 1998;23:59–66. 4. Odgaard Agundersen HJ. Quantification of connectivity in cancellous bone, with special emphasis on 3-D reconstructions. Bone 1993;14:173–82. 5. R¨uegsegger P, Koller B, Muller R. A microtomographic system for the non-destructive evaluation of bone architecture. Calcif Tiss Int 1996;58:24–9. 6. Efeoglu C, Delen Akc¸ay Y, Ert¨vrk S. A modified method for preparing platelet rich plasma: an experimental study. J Oral Maxillofac Surg 2004;62:1403–7. 7. Beck L, Deguzman L, Lee W. Rapid publication. TGF-beta 1 induces bone closure of skull defects. J Bone Miner Res 1991;6:1257–65. 8. Frame JW. A convenient animal model for testing bone substitute materials. J Oral Surg 1980;38:176–80. 9. Kim ES, Park EJ, Choung PH. Platelet concentration and its effect on bone formation in calvarial defects: an experimental study in rabbits. J Prosthet Dent 2001;86:428–33. 10. Schmitz JP, Hollinger JO. The critical size defect as an experimental model for craniomandibulofacial nonunions. Clin Orthop Relat Res 1986;205:299–308.
207
11. Vikjaer D, Blom S, Hjorting-Hansen E, Pinholt EM. Effect of plateletderived growth factor-bb on bone formation in calvarial defects: an experimental study in rabbits. Eur J Oral Sci 1997;105:59–66. 12. Zellin G, Beck S, Hardwick R, Linde A. Opposite effects of recombinant human transforming growth factor beta1 on bone regeneration in vivo: effects of exclusion of periosteal cells by microporous membrane. Bone 1998;22:613–20. 13. Cacciafesta V, Dalstra M, Bosch C, Melsen B, Andreassen TT. Growth hormone treatment promotes guided bone regeneration in rat calvarial defects. Eur J Orthod 2001;23:733–40. 14. Gowen M, Lazner F, Dodds R, et al. Cathepsin K knockout mice develop osteoporosis due to a deficit in matrix degradation but not demineralization. J Bone Miner Res 1999;14:1654–63. 15. Lutolf MP, Weber FE, Schmoekel HG, et al. Repair of bone defects using synthetic mimetics of collageneous extracellular matrices. Nat Biotechnol 2003;21:513–8. Epub 2003 April 21. doi:10.1038/nbt818. 16. Verna C, Dalstra M, Wikesj¨o UM, Trombelli L, Bosch Carles. Healing patterns in calvarial bone defects following guided bone regeneration in rats. A micro-CT scan analysis. J Clin Periodontol 2002;29:865– 70. 17. Jones AC, Sakellariou A, Limaye A, et al. Investigation of microstructural features in regenerating bone using microcomputed tomography. J Mater Sci Mater Med 2004;15:529–32. 18. Lu M, Rabie ABM. Microarchitecture of rabbit mandibular defects grafted with intramembraneous or endochondral bone shown by microcomputed tomography. Br J Oral Maxillofac Surg 2003;41:385–91.