Ovid: Denissen: J Periodontal Res, Volume 31(4).May 1996.265-270

© 1996 Munksgaard International Publishers Ltd.

Page 1 of 11

Volume 31(4), May 1996, pp 265-270

Dual X-ray Absorptiometry for Alveolar Bone: Precision of Peri-implant Mineral Measurements ex vivo [Original Article] Denissen, Harry1; Verhey, Hans2; de Blieck, Jolanda1; Corten, Frans3; Klein, Christel1; van Lingen, Arthur4 1 Department of Implantology, Academic Center for Dentistry, Amsterdam, The Netherlands, 2Department of Methodology and Statistics, Academic Center for Dentistry, Amsterdam, The Netherlands, 3Department of Oral Function and Prosthetic Dentistry, University of Nijmegen, Nijmegen, The Netherlands, 4Department of Nuclear Medicine, Free University Hospital, Amsterdam, The Netherlands Harry Denissen, Department of Implantology, Academic Center for Dentistry Amsterdam, Louwesweg 1, 1066 EA Amsterdam, The Netherlands Accepted for publication July 21, 1995

Abstract The precision of measurements of minor mineral changes in alveolar bone mineral content (ABMC) and alveolar bone mineral density (ABMD) on implant surfaces was determined in small regions (0.03±0.005 cm2) using dual X-ray absorptiometry (DXA). Dog hemimandibles with alveolar processes containing 17 implants were studied ex vivo. The precision was expressed as the coefficient of variation in percent (c.v.%). The ultra-high resolution protocol was applied to the mesial, distal and apical subregions of each implant. The line spacing was 0.0254 cm and the point resolution was 0.0127 cm. The mean c.v. (%)±s.d. for the ABMC in the mesial, distal and apical regions were 0.42±0.17, 0.47±0.21 and 0.48±0.18, respectively. For the ABMD these values were 0.42±0.16, 0.47±0.19 and 0.48±0.16. For each region approximately 68% of the 17 c.v. values were distributed within 1 s.d. from the mean c.v. These results indicate that measurements are highly reproducible (better than 0.48%) and that there are no differences in precision between several peri-implant regions. Changes as small as 0.85% in ABMC and ABMD in small areas adjacent to implant surfaces are measurable with a confidence level of 95%. Therefore the DXA technique will be expedient for our research evaluating the efficacy of the ceramic hydroxyapatite implant releasing agents affecting or inducing alveolar bone- and root cementum-like materials on its surface.

It has become clear that it is desirable that alveolar bone implants may require biologically active additives which may promote, maintain or induce periodontal-like materials such as alveolar bone and root cementum (1-5). Porous ceramic hydroxyapatite (CHA) as a bulk ceramic or as a ceramic coating applied on a metal substrate has been developed and is suitable for the slow release of bioactive agents (6-8). The rationale was that alveolar bone- and root cementum-like material could theoretically be effected or induced on implant surfaces by releasing resorptioninhibiting factors, growth hormones, cytokines or enzymes (9-11). The development of CHA implants as release systems and the availability of agents influencing or creating mineralized peri-implant materials have necessitated a method to determine the efficacy of these implants. A precise, non-destructive and convenient digital imaging bone

http://gateway.ut.ovid.com/gw2/ovidweb.cgi

5/26/2006

Ovid: Denissen: J Periodontal Res, Volume 31(4).May 1996.265-270

Page 2 of 11

densitometer technique seems to be the method of choice to compare CHA implant surfaces with and without release of a particular agent. A suitable method to determine bone mineral content (BMC) and bone mineral density (BMD) for the post-cranial skeleton is by dual X-ray absorptiometry (DXA) (12-16). The DXA unit uses an X-ray tube with two different energy levels. The two energy levels enable correction for the additional absorption by soft tissue. The system has an internal calibration wheel composed of materials equivalent to bone and soft tissue as well as an empty air segment (Fig. 1). These materials rotate through the X-ray beam between the tube and the patient, thus providing continuous calibration on a pixel (step)-by pixel basis. Measurements in each pixel are made for both energies with all three calibration materials (bone, soft tissue and air). The attenuation of photons is converted to BMC in grams (g) or, when corrected for the projected bone area, to BMD data, in g/cm2. The values obtained by DXA are not volumetric, as the measurements are not made per unit volume but rather per cross-section of bone.

Fig. 1. Schematic representation of mineral measurement on alveolar bone implant surfaces using dual X-ray absorptiometry. The right side shows the patient bench with, in the centre the tiny hemimandible of a dog (indicated by*). The left side shows an enlarged cross-sectional view of the hemimandible with 4 implants in the edentulous third and fourth premolar region. The X-ray source (X) is rigidly coupled by a “C-arm” to the X-ray detector (D). Individual scans are made across the implant regions from the lower border of the hemimandible to the crest of the alveolar process and vice versa. The system moves from the left (first molar) to the right (second premolar). The source collimator (SC) is mounted below the patient bench.

The feasibility of measuring minor alveolar BMC (ABMC) and alveolar BMD (ABMD) changes on implant surfaces will depend on the precision of the employed method. The precision of a measurement, which is expressed as the coefficient of variation in percent (c.v.%), is the smallness of random variation that occurs when measurements are repeated in the same specimen. This study concerns the precision of ABMC and ABMD measurements in small peri-implant regions in hemimandibles of dogs by DXA ex vivo.

http://gateway.ut.ovid.com/gw2/ovidweb.cgi

5/26/2006

Ovid: Denissen: J Periodontal Res, Volume 31(4).May 1996.265-270

Page 3 of 11

Material and methods Implants, retrieved mandibles and processing CHA implants were prepared from commercially available HA powder (Merck, Darmstadt, Germany) (17, 18). The implants were cylindrical, 3-5 mm in length and 3 mm in diameter. The implants were placed in the third and fourth premolar regions of mandibles of Beagle dogs immediately after extraction of teeth and followed-up with conventional intra-oral radiographs (Fig. 2).

The mandibles were retrieved 6 months after implantation and cut in two in the midline. For

this study 3 hemimandibles with 4 implants each and 1 hemimandible with 5 implants were measured directly after retrieval. The hemimandibles were from 4 different animals and all soft tissues were left in situ.

Fig. 2. Conventional intraoral radiograph of the premolar region of the mandible of a dog with 4 ceramic hydroxyapatite implants 6 months after implantation. Minor mineral changes on the

http://gateway.ut.ovid.com/gw2/ovidweb.cgi

5/26/2006

Ovid: Denissen: J Periodontal Res, Volume 31(4).May 1996.265-270

Page 4 of 11

implant surfaces with release of agents are potentially useful to study the efficacy of chemicals and biologicals affecting or inducing alveolar bone- and cementum-like materials.

Bone densitometry Scanning The ABMC and the ABMD measurements in the hemimandibles were made with a Hologic QDR-1000 densitometer (Hologic, Inc., Waltham, MA, USA). The densitometer uses an X-ray tube with 75 and 150 kV pulses. For this study the small animal option was applied with a regional ultra-high resolution scanning protocol and a single X-ray beam. A small X-ray source collimator was used with a diameter of 0.06 cm. The length and the width of the scan area were 4±1 cm. The line spacing was 0.254 mm and the point resolution 0.127 mm. The hemimandibles were placed on the patient bench in open air (Fig. 1). The laser dot indicating the starting point of the scan was positioned on the mesial side of the first molar. After the scan was started, an image of the hemimandible with the alveolar process and the implants appeared on the monitor screen one line at a time, from the bottom up. Once the desired anatomy of the hemimandible section with the implants had appeared in full on the screen the scan was stopped.

Analysis After adjusting the brightness and contrast of the image on the monitor the global region of interest was selected. The global region was a square area which outlined the hemimandible section with the implants (Fig. 3). The global region must include approximately 1 cm of air above the soft tissue crest of the alveolar ridge as well as 1 cm of air below the soft tissue layer of the lower border of the hemimandible to establish a soft tissue base line needed for the system to outline the bone edges correctly. After the global region of interest was established the following step was to select subregions surrounding the implant. For this study subregions were selected on the mesial, distal and apical surfaces of each implant. The length of the peri-implant region was 0.3±0.05 cm, dependent on the length of the projection of the implant image on the screen, but the width was always 0.1 cm. Once all the peri-implant region were demarcated, the image in the regions was calculated with respect to area, ABMC and ABMD. After editing the image, the system produced a report ready for printing.

Fig. 3. Print of DXA measurements with images of 4 CHA implants and resulting calculations. The

http://gateway.ut.ovid.com/gw2/ovidweb.cgi

5/26/2006

Ovid: Denissen: J Periodontal Res, Volume 31(4).May 1996.265-270

Page 5 of 11

global region of interest (large square box) as well as the mesial, distal and apical subregions on the CHA implant surfaces (small rectangular boxes with numbers) have been created. The subregions at the cervical surfaces of the implants (R6 and R7) were not used in this study.

To determine the precision of the measurements 4 more consecutive scans were done while the position of the hemimandible remained unchanged on the patient bench. The system repeated the whole scanning procedure of the first scan. The values found for the ABMC and ABMD of the second, third, fourth and fifth scans were compared with the first scan. The analysis template which had been created for the first scan was therefore employed via the “compare” feature of the system to standardize the measurement regions for the consecutive scans. The size, shape and the location of the subregions were thus duplicated by the system from the analysis made of the first scan.

Statistics For each implant and for each subregion the mean AMBC and the mean ABMD and the corresponding s.d. were calculated using the five measurements of the subregions. Thereafter the c.v. (%) for both the AMBC and ABMD were calculated in the usual way for each implant and each subregion. Finally the mean c.v. (%) and the corresponding s.d. were calculated for both the ABMC and ABMD. These statistics were calculated for each of the 3 peri-implant subregions. The probability distributions of the c.v. values were tested for normality by counting the measurements which were within the mean c.v. (%)±s.d. Differences between the c.v. (%) of the 3 peri-implant subregions were tested for statistical significance using analysis of variance for repeated measurements.

Results For the ABMC the range of the measurements was 0.0180-0.0429 g, 0.0179-0.0444 g and 0.0148-0.0456 g for the mesial, distal and apical subregions, respectively. For the c.v. (%) the range was 0.16-0.77, 0.19-0.87 and 0.00-0.72, respectively. The corresponding box plots are presented in Fig. 4. For the ABMD the range of the measurements was 0.5316-0.9833 g/cm2, 0.5264-1.0186 g/cm2 and 0.0148-0.0456 g/cm2. For the c.v. (%) the range was 0.12-0.69, 0.210.90 and 0.16-0.67. The corresponding box plots are shown in Fig. 5.

http://gateway.ut.ovid.com/gw2/ovidweb.cgi

5/26/2006

Ovid: Denissen: J Periodontal Res, Volume 31(4).May 1996.265-270

Page 6 of 11

Fig. 4. Box plots showing the precision (c.v.%) of the measurements of the mineral content in the mesial, distal and apical regions on the alveolar bone implants surfaces. The median c.v. (%) value is denoted as* (n=17).

http://gateway.ut.ovid.com/gw2/ovidweb.cgi

5/26/2006

Ovid: Denissen: J Periodontal Res, Volume 31(4).May 1996.265-270

Page 7 of 11

Fig. 5. Box plots showing the precision (c.v.%) of the measurements of the mineral density in the regions. The median c.v. (%) value is denoted as* (n=17).

Of the ABMC and ABMD values, 68% were within 1 s.d. of the mean c.v. (%) value. The means of the c.v. (%) values for the mesial, distal and apical regional measurements were for the ABMC 0.42±0.17, 0.47±0.21 and 0.48±0.18, respectively, and for the ABMD 0.42±0.16, 0.47±0.19 and 0.48±0.16. Analysis of variance did not show statistical significant differences between the 3 peri-implant subregions for both the ABMC (F=0.5299, d.f.=2.15, p=0.599) and the ABMD (F=0.5618, d.f.=2.15, p=0.582).

Discussion As far as we know no research reports have been published on the repeatability of DXA measurements for alveolar bone density. The one big draw-back to using DXA for measuring ABMC and ABMD is that with the currently available commercial instruments the X-ray beam hits first one half of the mandible or maxilla and then the other (19). Therefore the limitation of DXA is that the mandible or maxilla must be cut in two ex vivo. The hemimandible or hemimaxilla can then be placed on the patient bench of the DXA scanner and mineral measurements can be obtained of the alveolar process without superimposition of other bony structures.

http://gateway.ut.ovid.com/gw2/ovidweb.cgi

5/26/2006

Ovid: Denissen: J Periodontal Res, Volume 31(4).May 1996.265-270

Page 8 of 11

Our findings compare well with those of DXA measurements for the post-cranial skeleton. The CV (%) values for measurements by DXA reported in the literature are 0.5-3 (13-14). Markel et al. using the proximal femur of dogs, demonstrated that the presence of a titanium prosthesis

(20),

did not significantly affect the precision of measurements of the BMC and BMD in the surrounding bone, being 2.7 and 3% with and without the prosthesis, respectively. Other techniques for measuring BMC and BMD presently available are conventional X-ray densitometry, computer-aided densitometric image analysis of X-rays (CADIA), quantitative computed tomography (QCT), and dual photon absorptiometry (DPA). Conventional bite-wing Xrays are able to detect in vitro changes of 5% or greater in ABMC (21). By using CADIA in vitro under optimal conditions with a simple aluminium phantom c.v. (%) values of 1.014-1.639 have been reported (22). To attain a high level of precision for QCT careful and constant attention is required (23). However, precision of measurements was found to be 1-3% (24). Another aspect of this sensitive technique is the high radiation dosage and the cost of the tool which is therefore not readily available. The long-term in vivo precision that can be achieved with DPA is 2-4% (25). To assess the potential of DXA we did not have to apply special scanning conditions such as using water immersion at different depths or using a special positioning device as are used in other ex vivo experimental studies (15). The ex vivo results in hemimandibles of dogs indicate that conventional ultra-high resolution DXA measurements are applicable to alveolar bone sections and that satisfactory precision is obtained. The distribution of the c.v. values appeared to be normal because approximately 68% of the 17 values of the mesial, distal and apical subregions were within 1 s.d. from the mean c.v. The results of our ex vivo study indicate that DXA measurements are highly reproducible (better than 0.48%) and that there are no differences in precision between several peri-implant regions. The question remains whether a difference in ABMC and ABMD of 0.48% reflects a real mineral change of 0.48%. The statistical nature of the precision error has been discussed by several authors (26-28). For the mesial region the 2-sided 95% confidence interval runs from 0.08% to 0.66% which means that the real ABMC may be as low as 0.08% and as high as 0.66%. For the other 5 measures these intervals can also be computed. The lowest limit is 0.05% and the highest upper limit is 0.85%, which means that the real mineral change can be as low as 0.05% and the worst precision is 0.85%. In dental research we often have more than one measurement in the same individual. This is also the case in this study where there were 4 measurements for 3 dogs and 5 measurements for 1 dog (for each region). Therefore, one may question the assumption of statistical independence which is necessary for the analysis of variance and the computation of the confidence intervals. As can be seen from Figs 4 and 5 the median c.v. for all 3 regions were almost equal. Furthermore, the lowest c.v. (%) observed is 0.00 and the highest is 0.90. These observations are in line with the results of the statistical methods. Thus, it appears that our results are valid although some measurements are from the same dog. A single-tailed confidence limit is adequate because we want to have reasonable confidence that the precision error is less than some preassigned value (16, 27).

For example, we want to have a 95% confidence level (single tailed) that the precision

error is less than 1%. As can be concluded from our results, this was the case for all 6 measurements. Therefore, changes larger than 1% in ABMC and ABMD in small peri-implant regions (0.03±0.005 cm2) are most likely not due to lack of measurement precision but reflect real mineral gain or loss. Our results indicate a higher precision because the worst possible

http://gateway.ut.ovid.com/gw2/ovidweb.cgi

5/26/2006

Ovid: Denissen: J Periodontal Res, Volume 31(4).May 1996.265-270

Page 9 of 11

precision is 0.85%. Because of this (high) precision the DXA technique will be expedient for our intended research evaluating the efficacy of the CHA implant as release system for bisphosphonate, growth hormone, morphogenetic protein or alkaline phosphatase affecting or inducing mineralized periodontal-like materials on its porous surface.

Acknowledgements We gratefully acknowledge Joke Denissen-Gruter for the graphic work.

References 1. Denissen HW, Kalk W, Veldhuis AAH, Van den Hooff A. An eleven year study of hydroxyapatite implants. J Prosthet Dent 1989; 61: 706-711. Bibliographic Links Library Holdings [Context Link] 2. Groeneveld MC, Everts V, Beertsen W. A quantitative enzyme histochemical analysis of the distribution of alkaline phosphatase activity in the periodontal ligament of the rat incisor. J Dent Res 1993; 72: 13441350. Bibliographic Links Library Holdings [Context Link] 3. Lynch SE, Buser D, Hernandez RA, Weber HP, Stich H, Fox CH, Williams RC. Effects of the plateletderived growth factor/insulin-like growth factor-1 combination on bone regeneration around titanium dental implants. J Periodont 1991; 62: 710-716. [Context Link] 4. Wang X, Jin Y, Liu B, Zhou S, Yang L, Xi Y, White FH. Tissue reactions to titanium implants containing bovine bone morphogenetic protein: a scanning electron microscopic investigation. Int J Oral Maxillofac Surg 1994; 23: 115-119. Bibliographic Links Library Holdings [Context Link] 5. Beertsen W, Van den Bos T, Niehof J. Mineralization of dentinal collagen sheets complexed with alkaline phosphatase and integration with newly formed bone following subperiosteal implantation over osseus defects in rat calvaria. Bone Mineral 1993; 20: 41-55. Bibliographic Links Library Holdings [Context Link] 6. Denissen H, Van Beek E, Löwik C, Papapoulos S, Van den Hooff A. Ceramic hydroxyapatite implants for the release of bisphosphonate. Bone Mineral 1994; 25: 123-134. Bibliographic Links Library Holdings [Context Link]

7. Denissen HW, Kalk W, De Nieuport HM, Maltha JC, Van den Hooff A. Mandibular bone response to plasma-sprayed coatings of hydroxyapatite. Int J Prosthodont 1990; 3: 53-58. Bibliographic Links Library Holdings [Context Link]

8. Denissen HW, Kalk W, De Nieuport HM, Mangano C, Maltha JC. Preparation-induced stability of bioactive coatings. Int J Prosthodont 1991; 4: 432-439. Bibliographic Links Library Holdings [Context Link] 9. Denissen H, Kalk W, Van Beek E, Löwik C, Papapoulos S, van den Hooff A. Composites of hydroxyapatite and bisphosphonate: properties and alveolar bone response. J Mater Sci Mater Med 1995; 6: 35-40. Library Holdings [Context Link]

10. Clayden AM, Young WG, Zhang CZ, Harbrow D, Romaniuk K, Waters MJ. Ultrastructure of cementogenesis as effected by growth hormone in the molar periodontium of the hypophysectomized rat. J Periodont Res 1994; 29: 266-275. Bibliographic Links Library Holdings [Context Link] 11. Beertsen W, van den Bos T. Alkaline phosphatase induces the deposition of calcified layers in relation to dentin: an in vitro study to mimic the formation of afibrillara cellular cementum. J Dent Res 1991; 70: 176-181. Bibliographic Links Library Holdings [Context Link] 12. Mazess R, Chesnut III CH, McClung M, Genant H. Enhanced precision with dual-energy X-ray absorptiometry. Calcif Tissue Int 1992; 51: 14-17. Bibliographic Links Library Holdings [Context Link] 13. Lang PL, Steiger P, Faulkner K, Gluer C, Genant HK. Current techniques and recent developments in quantitative bone densitometry. Radiol Clin North Am 1991; 29: 49-76. [Context Link]

http://gateway.ut.ovid.com/gw2/ovidweb.cgi

5/26/2006

Ovid: Denissen: J Periodontal Res, Volume 31(4).May 1996.265-270

Page 10 of 11

14. Johnson J, Dawson-Hughes B. Precision and stability of dual energy X-ray absorptiometry measurements. Calcif Tissue Int 1991; 49: 174-178. Bibliographic Links Library Holdings [Context Link] 15. Kaymakci B, Wark JD. Precise accurate mineral measurements of excised sheep bones using X-ray densitometry. Bone Mineral 1994; 25: 231-246. Bibliographic Links Library Holdings [Context Link] 16. Wahner HW, Steiger P, von Stetten E. Instrument and measurement techniques. In: Wahner HW, Fogelman I. The evaluation of osteoporosis: dual energy X-ray absorptiometry in clinical practice. United Kingdom: Martin Dunitz Ltd, 1994; 14-34. [Context Link] 17. Denissen HW, De Groot K. Immediate dental root implants from synthetic dense calcium hydroxyapatite. J Prosthet Dent 1979; 42: 551-556. [Context Link] 18. Denissen HW, Van Dijk HJA, Gehring AP, De Groot K. Preparation of densely sintered calcium hydroxylapatite. In: Proceedings of 57th General Meeting of International Association of Dental Research, New Orleans: 1978; 613. [Context Link] 19. Corten FGA, Van't Hof MA, Buÿs WCAM, Hoppenbrouwers P, Kalk W, Corstens FHM. Measurement of mandibular bone density ex vivo and in vivo by dual-energy X-ray absorptiometry. Archs Oral Biol 1993; 38: 215-219. [Context Link] 20. Markel MD, Gottsauner-Wolf F, Wahner HW. Dual energy X-ray absorptiometry of implanted femora after cemented total hip arthoplasty in a canine model. In: Wahner HW, Fogelman I. The evaluation of osteoporosis: dual energy X-ray absorptiometry in clinical practice. United Kingdom: Martin Dunitz Ltd, 1994; 276-280. [Context Link] 21. Hildebolt CF, Rupich RC, Vannier MW, et al. Inter-relationships between bone mineral content measures. Dual energy radiography (DER) and bitewing radiography (BWX). J Clin Periodontal 1993; 20: 739-745. [Context Link] 22. Zubery Y, Dove SB, Ebersole J. An in vitro study of the characteristics of a computer-aided radiographic evaluation (CARE) system for longitudinal assessment of density changes. J Periodont Res 1993; 28: 233-240. Bibliographic Links Library Holdings [Context Link] 23. Cann CE, Genant HK. Precise measurement of vertebral mineral content using computed tomography. J Comput Assist Tomogr 1990; 4: 493-498. [Context Link] 24. Steiger P, Block JE, Steiger S. Spinal bone mineral density by quantitative computed tomography: effect of region of interest, vertebral level, and technique. Radiology 1990; 175: 537-543. Bibliographic Links

Library Holdings [Context Link]

25. Nilas L, Hassager C, Christiansen C. Long-term precision of dual photon absorptiometry in the lumbar spine in clinical settings. Bone Mineral 1988; 3: 305-315. Bibliographic Links Library Holdings [Context Link]

26. Cummings SR, Black D. Should perimenopausal women be screened for osteoporosis? Ann Int Med 1986; 104: 817-823. Bibliographic Links Library Holdings [Context Link] 27. Genant HK, Block JE, Steiger P. Appropriate use of bone densitometry. Radiology 1989; 170: 817-822. Bibliographic Links

Library Holdings [Context Link]

28. Goodwin P. Methodologies for the measurement of bone density and their precision and accuracy. Semin Nucl Med 1987; 17: 293-304. Bibliographic Links Library Holdings [Context Link]

Key words: alveolar bone mineral; implants; X-ray absorptiometry; repeatability

Accession Number: 00005201-199605000-00006

http://gateway.ut.ovid.com/gw2/ovidweb.cgi

5/26/2006

Ovid: Denissen: J Periodontal Res, Volume 31(4).May 1996.265-270

Page 11 of 11

Copyright (c) 2000-2006 Ovid Technologies, Inc. Version: rel10.2.2, SourceID 1.11354.1.251

http://gateway.ut.ovid.com/gw2/ovidweb.cgi

5/26/2006