Materials Transactions, Vol. 48, No. 3 (2007) pp. 343 to 347 Special Issue on Smart and Harmonic Biomaterials #2007 The Japan Institute of Metals EXPRESS REGULAR ARTICLE

Two-Dimensional Quantitative Analysis of Preferential Alignment of BAp c-axis for Isolated Human Trabecular Bone Using Microbeam X-ray Diffractometer with a Transmission Optical System

Sayaka Miyabe1;*1, Takayoshi Nakano1;*2, Takuya Ishimoto1;*1, Naoki Takano2, Taiji Adachi3, Hiroyoshi Iwaki4, Akio Kobayashi4, Kunio Takaoka4 and Yukichi Umakoshi1

1Divison of Materials and Manufacturing Science, Graduate School of Engineering, Osaka University, Suita 565-0871, Japan 2Department of Micro System Technology, Ritsumeikan University, Kusatsu 525-8577, Japan 3Department of Mechanical Engineering and Science, Graduate School of Engineering, Kyoto University, Kyoto 606-8501, Japan 4Department of Orthopaedic Surgery, Graduate School of Medicine, Osaka City University, Osaka 545-3851, Japan

Two-dimensional quantitative analysis of microbeam X-ray diffraction (XRD) was performed using a transmission optical system to examine biological apatite (BAp) orientation in an isolated trabecula of a human fourth lumber vertebral body. The incident X-ray beam is 20 mm in diameter, which is small enough for the isolated trabecula despite a slight beam divergence of 0.2. Integrated intensities of (002) and (310) are obtained separately by different incident angles and detector positions. Distribution of the preferential orientation of the BAp c-axis is finally calculated quantitatively as an integrated intensity ratio of (002)/(310) in a plane containing the trabecular direction. Preferential alignment of the BAp c-axis was finally determined to be perfectly parallel to the fiber direction in a rod-shaped trabecula, since accurate one-dimensional alignment is different from the alignment in the femoral cortical bone as a long bone that shows the local maximum of preferential alignment perpendicular to the longitudinal bone axis. For example, the integrated intensity ratio of (002)/(310) has a maximum value of 16 along the trabecular fiber and a minimum value of 0.09 in the perpendicular direction. Using this method, the anisotropy of BAp orientation in the trabecular bone can be quantitatively evaluated in the plane including the trabecular fiber. Thus, we successfully obtained a methodology that two-dimensionally analyzes the distribution of the BAp c-axis along all axes within a plane in a bone specimen. [doi:10.2320/matertrans.48.343]

(Received November 22, 2006; Accepted December 18, 2006; Published February 25, 2007) Keywords: trabecula, human vertebra, biological apatite crystallite, texture, bone quality, microbeam X-ray diffraction, position sensitive proportional counter (PSPC)

1. Introduction can exhibit an appropriate mechanical function by develop- ing material anisotropy on the basis of the applied stress field. Since trabecular bone shows much more metabolism In contrast, the degree of BAp orientation in pathological and activity than cortical bone, it acts as a reservoir to regulate the regenerated bones differs significantly along the representa- concentration of calcium and other mineral ions throughout tive bone axis from the original normal state.12–16) the human body.1) During modeling or remodeling, on the Vertebral bones are frequently evaluated in diagnoses of other hand, the trabecular direction is selected along the bone diseases, mainly osteoporosis, which is a severe bone stress line due to one functional adaptation of bone, as disorder in aging societies. Much attention has been paid to originally mentioned by Wolff.2) Many trabecular bone the trabecular bone in vertebrae because the trabecula is studies, which have been performed focusing on the patterns believed to sustain 30–50% of the vertebral strength.17) and distribution of trabecular fibers,3,4) have also been Moreover, the disease affects the tabecular bone more than calculated to clarify stress distribution in cancellous bone.5,6) the cortical bone.18) Many studies have focused on the At the nanoscale level, bone is a composite material relationship between trabecular architecture and mechanical mainly based on biological apatite (BAp) and collagen fibrils. properties, and actually, the notable contribution of trabec- BAp is an ionic crystal that crystallizes in an anisotropic ular architecture to the mechanical properties of cancellous hexagonal lattice in which the arrangement of ionic atoms is bone has been demonstrated.19,20) quite different along the crystallographic directions, for Previous reports have focused on BAp texture in cancel- example, the a- and c-axes in the BAp crystal.7) The c-axes of lous bone.13,21–27) Bacon et al. found preferential alignment BAp crystallites are basically distributed along the extending of BAp orientation in the portion composed of numerous collagen fibrils in bones and form texture.8,9) trabeculae in vertebral bone by using a neutron diffraction The orientation distribution of BAp in bone is crucial in technique.21,22) Both X-ray diffraction (XRD) and trans- terms of bone reinforcement.10) Nakano et al. showed that by mission electron techniques clarify the preferred orientation using the microbeam X-ray diffraction technique mature of the BAp c-axis in trabecular bone along the bone cortical bones such as rabbit ulna, rabbit skull bone, and longitudinal direction in the pig distal metaphysis of femur monkey dentulous mandible have unique preferential align- due to the preferential arrangement of the BAp c-axis along ment and texture of the BAp c-axis depending on in vivo each trabecular fiber.13) Jaschouz et al. impressively reported stress distribution.11) This indicates that normal original bone the preferential orientation of BAp crystal along the trabecula direction by recently conducting a microbeam X-ray tech- *1Graduate Student, Osaka University nique onto individual trabecula from (002) pole figure *2Corresponding author, E-mail: [email protected] analysis with a central focus on only the trabecular 344 S. Miyabe et al.

y angle, χ x 80° 100° ° ° z ω 60 120

rotation ω axis 40° 140° x-y-z table X-ray beam 20° 160° θ 2 χ ω ° ° rotation χ 0 180 axis y Collimator y-z plane Specimen z 2D PSPC trabecular x axis Specimen Fig. 1 Schematic figure of microbeam X-ray diffraction system with a transmission optical system. 2D quantitative analysis of preferential BAp orientation can be performed within y-z plane by rotation of axis. Fig. 2 Schematic drawing showing rotation of bone specimen around rotation axis to analyze distribution of BAp orientation along directions in the y-z plane. was rotated in steps of 20, and measurements were done direction.26) Scanning small-angle X-ray scattering (SAXS) nine times around from 0 to 180 . studies revealed that the particle shape of BAp in trabeculae tends to be elongated along the trabecular fiber.24–26) SAXS results only indirectly represent the BAp orientation, but their 110 mA, and the incident beam was focused onto a beam spot data imply that the orientation of BAp particles has a close of 20 mm in diameter by a mono capillary collimator with a relationship to crystallographic BAp orientation.26) Recent beam divergence of 0.2. Diffracted beams were detected reports suggest that each trabecula has preferential alignment with a two-dimensional position sensitive proportional of BAp orientation in the fiber direction, but quantitative counter (2D PSPC) (Hi-STAR, Bruker AXS) in which the analysis for the degree of BAp c-axis orientation is needed in intensity of the diffracted beam could be detected along the each trabecula between parallel and perpendicular to the fiber and 2 axes. Distances between specimen and collimator and direction. between specimen and detector were selected to be 5 mm and In this study, we performed 2D quantitative analysis of 15 cm, respectively. BAp orientation distribution in a plane with a trabecular fiber The trabecular specimen was fixed on the x-y-z tables so and then clarified the anisotropy of its distribution of BAp that the fiber direction was exactly parallel to the y-z plane. orientation parallel and perpendicular to the isolated trabec- Since the trabecular direction is always perpendicular to ular fiber. rotation axis , we can analyze the two-dimensional distribution of the BAp c-axis along all axes within the y-z 2. Materials and Methods plane including the trabecular fiber. The short axis on the cross section of the trabecula was fixed to be parallel to 2.1 Bone specimen rotation axis . An online CCD camera and a laser beam were A trabecula aligned approximately along the craniocaudal used for exact positioning of the trabecula. The center of the axis was selected from the central part of the fourth lumber trabecula’s cross section was fixed to be the center of the vertebral body of a male cadaver (66-year-old) donated at rotation of the specimen and beam path. Osaka City University Hospital. It has no evidence of We used two diffraction peaks of hexagonal-based BAp, metabolic bone disease. Analysis of this bone specimen was (002) and (310), to analyze BAp orientation. The (002) and approved by the Ethics Committee. The bone specimen was (310) diffraction peaks appear around Bragg angles of 25.9 fixed and kept in a 10% formalin neutral buffered solution to and 39.8, respectively, by Cu-K radiation. We defined the avoid infections and prevent denaturation of the organic degree of orientation of the BAp c-axis as an integrated matrix. The trabecula specimen is shown in an elliptical cross intensity ratio of the (002) diffraction to the (310) diffraction. section about 220 mm along the long axis and 160 mm along The value of the orientation degree in randomly orientated the short axis in diameter at the analysis point. apatite (calcium hydroxyapatite) powders (NIST #2910) is 0.6 in this XRD system. 2.2 Microbeam X-ray diffraction analysis To obtain diffraction information within the y-z plane 2D quantitative analysis of preferential BAp orientation including the trabecular axis, measurements for both (002) was performed by a microbeam X-ray diffractometer system and (310) were performed independently (Fig. 2). Since (D8 Discover with GADDS, Bruker AXS) with a trans- normal direction to the (002) plane corresponds to the c-axis mission optical system (Fig. 1). The x-y-z specimen table is and is parallel to (310), two diffraction peaks of (002) and mounted on a chi () axis, and the axis is further mounted (310) are suitable to determine the BAp c-axis orientation. on an omega (!) axis. The axis is always perpendicular to Incident angle (!) was selected as 13 for the (002) the ! axis and parallel to rotation axis !. Cu-K radiation diffraction and 20 for the (310) diffraction to obtain accurate was generated at a tube voltage of 45 kV and a tube current of diffraction intensity along the normal direction to the Two-Dimensional Quantitative Analysis of Preferential Alignment of BAp c-axis for Isolated Human Trabecular Bone 345

(a) (b) (211),(112), 14 14 13 (002) 13 (300),(202) (002) (211),(112), 12 12 (300),(202) 11 11 10 (310) 10 9 9 (310) 8 8 7 7 6 6 5 5 4 4 3 3 2 2 2θ 2θ χ 1 χ 1 0 0

Fig. 3 Typical two-dimensional microbeam XRD patterns for (002) in (a) and for (310) in (b). Several reflections such as (002), (310), and overlapping diffraction peaks of (211), (112), (300), and (202) appear as a function of and 2 axes. Intensity is shown in a pseudo-gray scale. detectable diffraction lattice planes in each Bragg condition in a plane with the y-z plane. In other words, the accurate 80° 100° 60° ° degree of BAp orientation for the (002) and (310) planes is (211),(112), 120 angle, χ detectable in all directions on the y-z plane including parallel (300),(202) 40° 140° and perpendicular directions along the trabecular fiber. (310) 14 In this study, the specimen was swung along the ! axis 13 (002) 12 within 5 around the accurate Bragg angle on the y-z plane 20° 160° (8 ! 18 for (002) and 15 ! 25 for (310)) to 11 10 compensate for the slight inclination of the trabecular fiber. 9 0° 180° The detector was placed at 2 ¼ 26 and 40 for (002) and 8 (310), respectively. 300µm 7 The specimen was rotated around the rotation axis to 6 5 analyze the distribution of BAp orientation along the 4 directions in the y-z plane. was rotated in steps of 20 , 3 and measurements were done nine times to cover the 2D data 2 around from 0 to 180 (Figure 2). Diffracted intensity was 1 recorded for 3600 seconds at each position to obtain 0 sufficient intensity for analysis. The profile was calculated in steps of 1, and the Fig. 4 Shape of analyzed trabecula and corresponding diffraction patterns. Clear arching of (002) reflection can be seen, and arching agrees well with integrated intensities of (002) and (310) were obtained by trabecular axis. Intensity is shown in a pseudo-gray scale. subtraction of the background intensity on the side of (002) and (310) diffraction: 24:4 2 24:9 and 26:9 2 27:4 for (002) and 37:3 2 38:3 for (310). (002) and 2 ¼ 40 for (310). Figure 4 shows the appearance of the trabecula and the 3. Results and Discussion corresponding diffraction patterns. Clear arching of (002) reflection was observed, and the range of arching corresponds Figures 3(a) and (b) show examples of two-dimensional to the trabecular axis. This suggests that the c-axis of XRD patterns for analyzing (002) and (310) diffraction, hexagonal BAp is preferentially orientated along the trabec- respectively. Several reflections such as (002), (310), and ular axis. This result is in good agreement with previous overlapping diffraction peaks of (211), (112), (300), and studies13,26,27) and strongly suggests that the trabecula is (202) can be recognized in the figures. The (002) diffraction reinforced along the trabecular direction not only by the in Fig. 3(a) and the (310) diffraction in Fig. 3(b) can be distribution of the trabecular fiber network but also by the obtained along the common axis normal to each correspond- material property based on the texture formation of BAp and ing diffraction lattice plane of the trabecula, as mentioned in the related collagen. Young’s modulus of bone tissue is well- Materials and Methods. In other words, each diffraction known to be closely related to the texture of bone tissue.28) In represents the orientation distribution along the directions contrast, the distribution of the overlapping diffraction peaks within the y-z plane containing the trabecular fiber, inde- of (211), (112), (300), and (202) along the axis seems more pendent of the rotation of the axis. The focused diffraction homogeneous than the (002) diffraction. ring appeared through the center of the 2D detector because We quantitatively evaluated the integrated peak intensities an appropriate detector position was selected as 2 ¼ 26 for of (002) and (310) as a function of from 0 to 180 in steps 346 S. Miyabe et al.

140 18 (a) 120 16 14 100 12

80 10 ° 60 8 54 (arbitrary unit) (arbitrary unit) 6 144° 40 4 20 2 Integrated intensity ratio of (002)/(310) Integrated intensity of the (002) diffraction 0 0 ° ° ° ° ° ° ° ° ° ° 0 20 40 60 80 100 120 140 160 180 0° 20° 40° 60° 80° 100° 120° 140° 160° 180° angle, χ 35 angle, χ (b) 30 Fig. 6 Integrated intensity ratio of (002)/(310) by using approximated intensities of (002) and (310) diffractions in Fig. 5. 25

20 perpendicular direction to the long axis of the cortical portion 15 of bovine femur in addition to strong preferential alignment (arbitrary unit) 10 along the long bone axis.9,29) In our study of human trabecular

5 fiber, not even weak alignment of BAp orientation was found along that direction other than the trabecular fiber because

Integrated intensity of the (310) diffraction 0 there is no local maximum of the intensity ratio of (002)/ 0° 20° 40° 60° 80° 100° 120° 140° 160° 180° angle, χ (310) (Fig. 6). This strongly suggests that rod-shaped trabecular fiber shows accurate one-dimensional preferential Fig. 5 Integrated peak intensity distribution of (002) in (a) and (310) in (b) alignment of the BAp c-axis along the trabecular fiber, which as a function of rotation angle after subtraction of background from integrated peak intensity. Incident angle (!) was selected to be 13 for is different from the femoral cortical bone as a long bone. (002) and 20 for (310), and the detector was placed at 2 ¼ 26 for (002) In this study, we separately measured (002) and (310) and 40 for (310). diffractions by adjusting the incident angle to the Bragg angle. This enabled us to analyze BAp orientation exactly in the directions between parallel and perpendicular to the of 1. The analysis of (002) and (310) was of course trabecular fiber. Hence, the degree of BAp orientation is performed by different diffraction patterns with different strongly reliable in the vertebral trabecula of the 66-year-old incident angles (!) to avoid different Bragg conditions. After man. In cortical bone, it is reported that the preferential subtraction of the background from the integrated diffraction degree of BAp orientation varies depending on such factors peak intensity, we obtained the actual peak intensity as age,10,30) distribution and magnitude of in vivo stress,11,31) distribution of (002) and (310) as a function of rotation degree of disease progression,14) and fracture heal- angle () (Figs. 5(a) and (b)). Intensity data were then ing.12,13,16,32) These investigations were done using a com- approximated by a polynomial fitting technique, as expressed parably larger beam, which is sufficient for the cortical bone. by a solid line in the figures. Determination coefficient of the However, since the size of the isolated trabecula is just a few fitting (R2) is 0.99 for (002) and 0.75 for (310), which means hundred mm, a small beam is needed. We used an incident a successful approximation despite scattered (310) diffraction beam whose diameter was 20 mm with a beam divergence of intensity. Since the change in the integrated intensity of (002) 0.2, which is small enough for individual trabeculae and is in direct opposition to (310) as a function of , this quantitative mapping on a trabecula within the y-z plane. measurement for both (002) and (310) diffractions is very Analysis on the limited local area for observation, several reliable. mm2, is also possible when using a synchrotron micro-focus Figure 6 shows the preferential orientation of the BAp c- X-ray beam with significant intensity, but the system is very axis, which is calculated quantitatively as an integrated ambitious and inconvenient.27,33) Therefore, the technique in intensity ratio of (002)/(310) by using approximated inten- this study conveniently allows verification of bone quality sity. A clear distribution of the preferential BAp orientation based on mechanical anisotropy to diagnose bone disease. along was obtained within the y-z plane containing parallel On the other hand, 2D quantitative analysis can be applied and perpendicular directions to the trabecular axis. Peak to cortical bone, which is needed to understand the 2D maximum appears around ¼ 54, which corresponds to anisotropy of the BAp orientation in-plane. Skull bone, for the trabecular axis, and peak minimum appears around example, has a unique two-dimensional BAp alignment on ¼ 144, which is perpendicular to the trabecular axis. The the skull surface, but preparing the bone specimen for quantitative degrees of BAp orientation as an integrated quantitative analysis is troublesome.11) This technique is intensity ratio of (002)/(310) are 16 at maximum peak and useful to analyze the distribution of the BAp c-axis two- 0.09 at minimum peak. Sasaki et al. reported that a weak dimensionally along all axes within a plane in a bone preferential orientation of BAp c-axis was present in the specimen. Two-Dimensional Quantitative Analysis of Preferential Alignment of BAp c-axis for Isolated Human Trabecular Bone 347

4. Conclusion 389–450. 3) S. Majumdar, M. Kothari, P. Augat, D. C. Newitt, T. M. Link, J. C. Lin, T. Lang, Y. Lu and H. K. Genant: Bone 22 (1998) 445–454. Quantitative analysis of preferential alignment of the BAp 4) J. H. Kinney, James S. Stflken, T. S. Smith, James T. Ryaby and c-axis for an isolated human trabecula was two-dimension- N. E. Lane: Bone 36 (2005) 193–201. ally performed using a microbeam X-ray diffractometer 5) M. Ito, K. Nakayama, A Konaka, K. Sakata, K. Ikeda and T. system with an incident beam 20 mm in diameter using a Maruyama: Bone 39 (2006) 453–459. transmission optical system. The following conclusions were 6) S. D. Rockoff, E. Sweet and J. Bleustein: Calcif. Tissue Res. 3 (1969) 163–175. reached: 7) J. C. Elliot: Structure and chemistry of the apatites and other calcium (1) An exact integrated intensity ratio of (002)/(310) for phosphates, (Elsevier, Amsterdam, 1994) pp. 1–389. bone tissue can be obtained along in all directions 8) W. J. Landis: Bone 16 (1995) 533–544. within a plane by microbeam XRD with a small 9) N. Sasaki and Y. Sudoh: Calcif. Tissue Int. 60 (1997) 361–367. incident beam 20 mm in diameter. Thus, the anisotropy 10) T. Nakano, Y. Tabata and Y. Umakoshi: Encyclopedia of Materials, Science and Technology Updates,(Texture and Bone Reinforcement, of BAp c-axis arrangement in a trabecula of human Elsevier, Oxford, 2005) MS2061 pp. 1–8. vertebra can be analyzed quantitatively in a plane 11) T. Nakano, K. Kaibara, Y. Tabata, N. Nagata, S. Enomoto, E. containing directions parallel and perpendicular to the Marukawa and Y. Umakoshi: Bone 31 (2002) 479–487. trabecular fiber. 12) T. Nakano, K. Kaibara, Y. Tabata, N. Nagata, S. Enomoto, E. (2) The rod-shaped trabecula exhibits an accurate one- Marukawa and Y. Umakoshi: Tissue engineering for therapeutic use 6, (Analysis of hydroxyapatite (HAp) texture in regenerated hard tissues dimensional alignment of BAp c-axis that is different using micro-beam X-ray diffractometer technique, Elsevier, Tokyo, from the femoral cortical bone as a long bone showing 2002) pp. 95–104. locally weak preferential alignment along a direction 13) T. Nakano, T. Ishimoto, J. W. Lee, Y. Umakoshi, M. Yamamoto, Y. perpendicular to the longitudinal bone axis. The Tabata, A. Kobayashi, H. Iwaki, K. Takaoka, M. Kawai and T. integrated intensity ratios of (002)/(310) are 16 and Yamamoto: Mater. Sci. Forum 512 (2006) 255–260. 14) J. W. Lee, T. Nakano, A. Kobayashi, K. Takaoka, Y. Tabata and 0.09 in directions parallel and perpendicular to the Y. Umakoshi: Phosph. Res. Bull. 17 (2004) 83–84. trabecular fiber, respectively. 15) J. W. Lee, T. Nakano, S. Toyosawa, N. Ijuhin, Y. Tabata, M. (3) The degree of BAp orientation as an integrated intensity Yamamoto and Y. Umakoshi: Mater. Sci. Forum 512 (2006) 265–268. ratio of (002)/(310) may be used to check bone quality 16) T. Ishimoto, T. Nakano, Y. Umakoshi, M. Yamamoto and Y. Tabata: and diagnose degree of bone disease, relating to Mater. Sci. Forum 512 (2006) 261–264. 17) R. B. Mazess: Calcif. Tissue Int. 47 (1990) 191–193. mechanical function. 18) A. Shiraishi, S. Higashi, T. Masaki, M. Saito, M. Ito, S. Ikeda and T. Nakamura: Calcif. Tissue Int. 71 (2002) 69–79. Acknowledgements 19) M. Ito, A. Nishida, A. Koga, S. Ikeda, A. Shiraishi, M. Uetani, K. Hayashi and T. Nakamura: Bone 31 (2002) 351–358. This work was supported by funds from the ‘‘Priority 20) A. M. Parfitt: Am. J. Med. 82 (1987) 68–72. 21) G. E. Bacon and R. K. Griffiths: J. Appl. Cryst. 10 (1977) 124–126. Assistance of the Formation of Worldwide Renowned 22) G. E. Bacon, P. J. Bacon and R. K. Griffiths: J. Anat. 128 (1979) 277– Centers of Research—The 21st Century COE Program 283. (Project: Center of Excellence for Advanced Structural and 23) S. Rinnerthaler, P. Roschger, H. F. Jakob, A. Nader, K. Klaushofer and Functional Materials Design)’’ and a Grant-in-Aid for P. Fratzl: Calcif. Tissue Int. 64 (1999) 422–429. Scientific Research and Development from the Ministry of 24) N. P. Camacho, S. Rinnerthaler, E. P. Paschalis, R. Mendelsohn, A. L. Boskey and P. Fratzl: Bone 25 (1999) 287–293. Education, Culture, Sports, Science and Technology of 25) P. Roschger, B. M. Grabner, S. Rinnerthaler, W. Tesch, M. Kneissel, A. Japan. The work was partly supported by Industrial Tech- Berzlanovich, K. Klaushofer and P. Fratzl: J. Struct. Biol. 136 (2001) nology Research Program (Project: 03A47002) in 2003 from 126–136. New Energy and Industrial Technology Development Or- 26) D. Jaschouz, O. Paris, P. Roschger, H. Hwang and P. Fratzl: J. Appl. ganization (NEDO) of Japan and by CREST, Japan Science Cryst. 36 (2003) 494–498. 27) E. Rokita, P. Chevallier, P. H. A. Mutsaers, Z. Tabor and A. Wrobel: and Technology Agency. The authors would like to thank Nucl. Inst. Meth. Phys. Res. B 240 (2005) 69–74. Prof. Yuji Nakajima and Prof. Hiroshi Kiyama, Osaka City 28) W. Bonfield and M. D. Grynpas: Nature 270 (1977) 453–454. University for providing the bone specimen. We wish to 29) N. Sasaki, N. Matsushima, T. Ikawa, H. Yamamura and A. Fukuda: thank the family of the donor for the generosity in the face of J. Biomech. 22 (1989) 157–164. the bereavement. 30) G. E. Bacon and R. K. Griffiths: J. Anat. 143 (1985) 97–101. 31) T. Ishimoto, T. Nakano, Y. Umakoshi, M. Yamamoto and Y. Tabata: Phosph. Res. Bull. 17 (2004) 77–82. REFERENCES 32) T. Nakano, K. Kaibara, T. Ishimoto, Y. Tabata and Y. Umakoshi: Tissue Eng. 8 (2002) 1232. 1) D. H. Copp and S. S. Shim: Oral. Surg. Oral. Med. Oral. Pathol. 16 33) H. R. Wenk and F. Heidelback: Bone 24 (1999) 361–369. (1963) 738–744. 2) J. Wolff: Virchows Arch. Pathol. Anat. Physiol. Klin. Med. 50 (1870)