Relation between Spongy Bone Density in the Maxilla and Skeletal Bone Density Joe Merheb, DDS, MSc;* Andy Temmerman, DDS, MSc;* Wim Coucke, MStat, PhD;† Lars Rasmusson, DDS, PhD;‡ Alexander Kübler, DDS, PhD;§ Andreas Thor, DDS, PhD;¶ Marc Quirynen, DDS, PhD*
ABSTRACT Background and Purpose: Osteoporosis is a disease affecting more than 300 million people worldwide and is responsible for numerous medical complications. This study aimed to investigate the relation between skeletal and maxillary bone density. Materials and Methods: Seventy-three patients were recruited and divided between group A (osteoporosis), group B (healthy, control), and group C (osteopenia) on the basis of a dual-energy x-ray absorptiomery (DXA) scan. These patients also received a CT scan on which bone density measurements were performed at five sites: maxilla midline, retromolar tuberosities, incisor, premolars, and molar regions. Results: The bone density was lower in osteoporotic patients compared with the control patients. The bone mineral density (BMD) of the tuberosities showed the strongest correlations with the BMD of the hip and the spine (respectively, r = 0.50 and r = 0.61). The midline region showed moderate correlations with the hip (r = 0.47) and the spine (r = 0.46). For potential implant sites, the correlations with the BMD of the hip and spine were, however, small to insignificant. Based on measurements of bone density of the maxilla, it was possible to predict if the patient was osteoporotic or not with a sensitivity of 65% and a specificity of 83%. Conclusions: The maxillary bone density of subjects with osteoporosis is significantly lower than that of healthy patients. Moreover, there is a direct correlation between the density of the skeleton and the density of some sites of the maxilla. Using measurements of maxillary bone density in order to predict skeletal bone density might be a useful tool for the screening of osteoporosis. KEY WORDS: bone density, bone mineral density, dual-energy x-ray absorptiometry, Hounsfield units, maxilla, skeletal
INTRODUCTION
to 300 million persons worldwide are affected by osteoporosis,1 especially women.2 Its pathogenesis lies in the rupture of the balance between bone formation (osteocyte/osteoblast activity) and bone resorption (osteoclastic activity). Osteoporosis is considered one of the main causes of hip fractures in the elderly, leading to increased immobility (6 months after hip fracture, only 15% of patients can walk without assistance) and a dramatic rise in mortality rates in the year following the fracture (1 out of 5 people experiencing a hip fracture will die within a year3,4). A number of major risk factors for osteoporosis have been identified. Those include but are not restricted to sex (higher risk for females), ethnic background (higher risk for Caucasians), smoking, alcohol abuse, malnutrition (low intake of calcium or vitamin D increase the risk), and certain medications. Equally, different therapeutic alternatives have been proposed in order to, if not reverse or stop, at least slow
Osteoporosis is a systemic disease characterized by low bone mass and microarchitectural deterioration of bone tissue leading to enhanced bone fragility and a consequent increase in fracture risk.1 According to recent figures from the World Health Organization (WHO), up *Unit of Periodontology, Department of Oral Health Sciences, University of Leuven, Kapucijnenvoer, Belgium; †Scientific Institute of Public Health, Brussels, Belgium; ‡Department of Oral and Maxillofacial Surgery, Sahlgrenska Academy, University of Gothenburg, Göteborg, Sweden; §Department of Oral and Maxillofacial Plastic Surgery, University of Würzburg, Würzburg, Germany; ¶Associate Professor, Department of Oral and Maxillofacial Surgery, Institute of Surgical Sciences, Uppsala University, Uppsala, Sweden Corresponding Author: Dr. Joe Merheb, Unit of Periodontology, Department of Oral Health Sciences, University of Leuven, Belgium; e-mail: [email protected] © 2014 Wiley Periodicals, Inc. DOI 10.1111/cid.12228
1
2
Clinical Implant Dentistry and Related Research, Volume *, Number *, 2014
down bone mass loss. A great emphasis has been placed on the prevention of disease occurrence by maintaining a healthy lifestyle (regular exercise and limitation of tobacco and alcohol use) and a balanced diet ensuring a sufficient intake of calcium and other minerals. Maintaining an adequate diet during the period preceding the adult age has been deemed critical in order to reach a sufficient bone mass peak at the end of skeleton growth.5,6 Additionally, a number of medications have been introduced, among which bisphosphonates are currently the most widely used. Because of its importance and ever increasing costs to society and the social security system, osteoporosis has recently received a lot of attention, and several information and screening campaigns have been initiated. Dentists are among the few medical professionals who see their patients on a regular basis. A large benefit could be derived from involving them in a screening process. Therefore, the aim of this study was to investigate the relation between skeletal and maxillary bone density and to look into the potential to predict skeletal bone density based on density measurements from the maxilla. MATERIALS AND METHODS Patient Selection The study was designed as a multicentric trial and involved four centers: Leuven (Belgium), Gothenburg (Sweden), Uppsala (Sweden), and Würzburg (Germany). The study was reviewed and approved by the ethical committee relevant to each of the participating centers.Patients were deemed eligible for inclusion if they were female, above 60 years of age, partially or fully edentulous in the maxilla, in need of 2 to 8 implants in the maxilla, with a history of edentulism in the planned area of implantation of at least 3 months, and had never received treatment with bisphosphonates. The patients received a detailed explanation about the study, its timeline, its aims, the risks involved, the potential benefits, and the compliance expectations. Patients who met all the inclusion criteria and none of the exclusion criteria signed the informed consent form and underwent a dual-energy x-ray absorptiometry (DXA) scan using either a Hologic™ or a GE-Lunar™ DXA densitometer (Hologic; GE-Lunar) and a spiral computerized tomography (CT) scan. Calibration between the DXA sites was performed via phantom scans, and standardized bone
mineral density (sBMD) values (mg/cm2) were calculated7,8 to compensate for differences between the DXA machines used at the different sites. The DXA scans were assessed centrally at the osteoporosis unit of the University of Uppsala (Sweden). Seventy-three patients were recruited, and for all subjects two regions of interest (ROIs) were defined on the DXA scans, one for all of the hip and one for the lumbar spine (L1 to L4). T-scores for the the two ROIs were calculated based on the difference in density between the examined subject and a genderspecific reference population of white Caucasian females (NHANES III)9 for the hip ROI and on a manufacturerderived gender-specific reference population of white Caucasian females for the lumbar spine ROI. The patients were assigned to either the osteoporosis group (A; T-score 2 −2; n = 18), the control group (B; T-score 3 −1; n = 38), or the osteopenia group (C; −2 < T-score < −1; n = 16) based on the lowest T-scores for the hip or lumbar spine ROIs. (The WHO criteria for osteoporosis rely on a T-score of −2.5 as the threshold for osteoporosis.) Computerized Tomography Analysis Each patient received a CT scan. Data were stored in DICOM format. These DICOM files were loaded in planning software (SimPlant Crystal, Materialise Dental, Leuven, Belgium) and a three-dimensional model was reconstructed. On the three-dimensional reconstruction for each patient, a number of ROIs were defined, and the bone density of the spongy part of each site in Hounsfield units (HU) was recorded. The following ROIs were defined (Figure 1): 1. The maxillary midline: inferior to the anterior nasal spine, between the two central incisors (if present)
Figure 1 Visualization of the different regions of interest (ROI).
Maxillary and Skeletal Bone Density
2. The left and right tuberosities: distal to the third maxillary molar (if present) and distal and inferior to the maxillary sinus 3. The incisor region: inferior and mesial to the lateral wall of the nasal fossae 4. The premolar region: inferior to the mesial half of the maxillary sinus 5. The molar region: inferior to the distal half of the maxillary sinus Whenever possible, bilateral measurements of each ROI were performed (except for the maxillary midline region). In such cases the mean of both values was considered. However, in a number of cases, due to the presence of teeth, implants, artifacts, or other anatomical structures, only unilateral measurements were possible. The explored areas were cylindrical with a diameter of 5 mm and a length corresponding to the maximal extension of the cylinder between the coronal cortex and the cortex of the sinus or nasal floor. Statistical Analysis The difference between the osteoporosis and control groups in ROI values was assessed by a linear mixed model, and the multicenter aspect of the study was modeled by using center as a random factor. Model assumptions were tested by a normal quantile plot of the residuals and a residual dot plot. Pearson correlation coefficients and corresponding p values were calculated to assess the relations between ROI and DXA values. Stepwise model selection was applied to predict the grouping of patients into groups “A” (osteoporosis) and “Not A” (no osteoporosis) based on a generalized linear model for binomial outcomes and a logit link using ROI values as candidate predictive variables. RESULTS Comparison between the Maxillary Bone Densities of Osteoporotic and Healthy Patients The Hounsfield values of the patients in groups A and B were compared at all five reference sites. This comparison yielded a significant difference in bone density between the two groups at all sites. The significance of this difference was greater at the tuberosities and midline sites (p < .01), and in all the comparisons, the osteoporosis group recorded lower density values. The most notable difference was recorded at the tuberosity
3
sites, where the osteoporosis group averaged 13.3 1 72.4 HU in comparison to 126.9 1 102.3 HU in the control group (Figure 2). It was also noteworthy that the measurements of bone density at the tuberosity sites delivered negative HU scores in 7 out of 18 patients (38.9%) of the osteoporosis group, while only 4 out of 38 patients (10.5%) had negative HU scores in the control group. Comparison between the Maxillary and Skeletal Bone Densities The bone mineral densities (BMD) of the hip and the spine were compared with the density of the different intraoral reference sites for all three groups. Significant correlations between the BMD of the hip and those of the two nondental sites (tuberosity, midline) were found. They reached r = 0.50 (p < .01) at the tuberosity sites and r = 0.47 (p < .01) at the midline site. At the dental sites, only the BMD of the premolar region correlated slightly with the BMD of the hip (r = 0.3, p = .01) (Figure 3). A similar trend was observed when the BMD of the spine was compared with the same intraoral sites. A correlation of r = 0.61 (p < .01) was found between the BMD of the spine and that of the tuberosities. A correlation of r = 0.46 (p < .01) was found between the BMD of the spine and that of the midline. At the dental sites, the correlations between the BMD of the spine and those of the incisor (r = 0.24, p = .04) and premolar (r = 0.31, p < .01) regions reached significance (Figure 4). Prediction of Skeletal Bone Group Based on Measurements of Maxillary Bone A stepwise model selection was applied to predict the grouping of patients into groups “A” (osteoporosis) and “Not A” (groups B and C) based on a generalized linear model for binomial outcomes and a logit link. It appeared that the tuberosities and midline had the best predictive power. The following formula was applied:
Sign (3.034636 − 0.0160303 × Tuberosities − 0.005359683 × Midline) A positive value led to a prediction of group “A” and a negative value to a prediction of group “not A.”
4
Clinical Implant Dentistry and Related Research, Volume *, Number *, 2014
Tuberosity (average L + R)
Midline 800
300 600
HU
HU
200 100
400
0 –100
200
–200
A = 13.3 ± 72.4 HU
B = 126.9 ± 102.3 HU
Incisor
A = 414.0 ± 146.8 HU
Premolar
B = 552.6 ± 128.0 HU
Molar
600
600
400
400
HU
HU
400 200
HU
600
200
200 0
0
A = 260.3 ± 145.4 HU
B = 372.5 ± 178.6 HU
A = 242.2 ± 123.5 HU
B = 318.5 ± 182.9 HU
A = 139.7 ± 166.8 HU
B = 210.2 ± 167.2 HU
Figure 2 Comparison between the maxillary BMD of osteoporotic and healthy patients. *p < .05; **p < .01. A = osteoporosis group; B = control group; BMD = bone mineral density; HU = Hounsfield units; L = left; R = right.
The selected model allowed the prediction of the group to which the patient belonged with a moderate level of sensitivity (0.65%) and a high level of specificity (0.83%). DISCUSSION Several studies have investigated the relation of skeletal and jaw bone density. There was no consensus as to the existence or absence of a direct relation. In a pioneer study, Horner and colleagues10 performed DXA scans on both skeletal and jaw bone sites and found significant correlations between the densities of the femoral neck and lumbar spine and those of all the mandibular sites they tested (mandibular symphysis, mandibular body and ramus) with the strongest correlations being recorded at the mandibular body site. In a study using similar measuring devices, Drage and colleagues11 detected a significant correlation of the BMD of the ramus with that of the femur and the lumbar spine. However, contrarily to the Horner study, the BMD of the other jaw bone sites, namely the body of the mandible, the mandibular symphysis, and the anterior maxilla did not correlate well with either of the two skeletal BMD
measurements. While most studies report on the relationship of mandibular bone with skeletal bone, Lindh and colleagues12 investigated the relationship between the density of maxillary bone derived from CT scans and the skeletal bone density of the spine and hip derived from DXA scans. This investigation found inconsistent correlations between the maxillary and skeletal bone densities. The incisor sites correlated significantly with the lumbar spine sites, while most other dental sites did not correlate significantly with either the spine or hip density measurements. This discrepancy might be explained by the very low number of subjects (n = 8) enrolled in this study. When investigating the relation between jaw bone density and skeletal bone density, most studies have relied on measurements performed on the mandible, and rarely were there maxillary sites involved. According to von Wowern, this is due to the fact that the basal area of the mandible posterior to the mental foramen is the only region of the jaws that satisfies the requirements for a standard measurement site in regard to anatomical size, shape, bone structure, and function.13 The authors of the present study found, however, that the maxilla,
Maxillary and Skeletal Bone Density
B
300 200
BMD Midline
BMD Tuberosity
A
100 0 –100
5
800
600
400
200
–200 600
800
1,000
1,200
600
sBMD (hip) r = 0.5, p < 0.01 D
400 200 0 600
800
1,000
1,200
sBMD (hip) r = 0.22, p = 0.06
400 200 0 600
800
1,000
1,200
sBMD (hip) (NS)
1,200
E 600
600
BMD Molar
600
1,000
sBMD (hip) r = 0.47, p < 0.01 BMD Premolar
BMD Incisor
C
800
r = 0.30, p = 0.01
400 200 0 600
800
1,000
1,200
sBMD (hip) r = 0.20, p = 0.1
(NS)
Figure 3 Relation between the BMD of the hip and the different BMD of the maxilla. BMD = bone mineral density (g/cm ); sBMD = standardized bone mineral density (mg/cm2). 2
even if it did not offer the same possibility for standardization of measurements, still included useful measurement sites because it was more prone to osteoporosis than the mandible.14 This vulnerability to osteoporosis is probably due to the higher prevalence of trabecular structures in the maxilla as compared with the mandible.15,16 The use of Hounsfield units as an indicator of bone mineral content (BMC) has been assessed by several publications, and a positive correlation between Hounsfield values and bone mineral density has been found.17,18 The reliability of this correlation is, however, dependent on several factors, among which age plays an important role. A method to correct for age-dependent composition of trabecular bone (BMC calibration) was proposed by Nilsson and colleagues.19 While calibration of the CT equipment and sensitivity adjustment were performed daily in the context of this study, no BMC calibration or attempt to transform Hounsfield values into BMC was performed. This might be a shortcoming of the study, but one which is possibly of limited influence, given the the homogeneity of the population sample studied (exclusively women, limited age range of 60–76 years old, exclusively Caucasian, no extreme body
mass index values). Previous studies comparing jaw bone density to skeletal bone density have always performed their jaw bone density measurements on single two-dimensional slices, whereas the present study utilized three-dimensional cylinders that covered five to seven consecutive cross-sectional slices. This allows the selected ROIs to have a greater surface by the addition of the surfaces of the different areas covered by the cylinders on the consecutive slices, thereby increasing the reliability of the performed measurements. The total area exceeded the surface area of 1 cm2 defined as the threshold surface area for reliable measurements.20 Additionally, the choice of a volume ROI is justified by the great variability of bone density on a local level, even on consecutive slices as can be seen in Figure 5. In the present study, the strongest correlations between the skeletal and jaw bone were recorded at the tuberosities. A correlation of r = 0.61 with the BMD of the spine and of r = 0.50 with the BMD of the hip were found. These results support the original study hypothesis that maxillary sites are also useful for the detection of osteoporosis due to their trabecular composition. The posterior maxilla and tuberosities are the regions of the jaw with the highest proportion of trabecular bone.21,22
6
Clinical Implant Dentistry and Related Research, Volume *, Number *, 2014
B
300 200
BMD Midline
BMD Tuberosity
A
100 0 –100
800
600
400
200
–200 600
800
1,000
1,200
1,400
600
1,600
800
sBMD (spine) r = 0.61, p < 0.01 D
400 200 0 600
800 1,000 1,200 1,400 1,600
sBMD (spine) r = 0.24, p = 0.04
1,400
1,600
E 600
600
BMD Molar
600
1,200
sBMD (spine) r = 0.46, p < 0.01
BMD Premolar
BMD Incisor
C
1,000
400 200 0 600 800 1,000 1,200 1,400 1,600
sBMD (spine) r = 0.31, p < 0.01
400 200 0 600 800 1,000 1,200 1,400 1,600
sBMD (spine) r = 0.17, p = 0.16
(NS)
Figure 4 Relation between the BMD of the spine and the different BMD of the maxilla. BMD = bone mineral density (g/cm ); sBMD = standardized bone mineral density (mg/cm2). 2
Even though the calculated correlations between the tuberosity sites and the two skeletal sites are far from being linear, it is important to relativize by calculating the correlation between the two skeletal sites (hip and spine), which in this study amounted to r = 0.77. Given that even these two standard reference sites do not correlate perfectly, it can be suggested that the BMD of the tuberosities is a reliable reference that could possibly be used for the prediction of skeletal osteoporosis. The stepwise model analysis performed the extraction of the sensitivity and specificity of the selected model. While a moderate level of sensitivity and a high level of specificity were found, those results could be noticeably improved if the test could be applied to a bigger sample. The present study showed that the relation between the density of the skeleton and the density of the dental sites (incisor, premolar, molar) was weak to moderate. It also showed that the difference in spongy bone density between osteoporotic and nonosteoporotic patients was less evident at the dental sites than at the nondental sites. In the context of implant therapy, those findings seem to suggest that skeletal bone density might have only a limited influence on
implant therapy success. Those results are in line with the conclusions of previous studies on the effect of osteoporosis on implant therapy outcome. Histological observations by Shibli and colleagues23 showed no difference in bone-implant contact between the osteoporosis and control groups. While the control group showed an average of 48%, the osteoporosis group showed an average of 46%. Similarly, a large study24 on 746 women (3,224 implants) showed that patients with an established diagnosis of osteoporosis were not more likely to have dental implant failure than patients in whom the diagnosis of osteoporosis had not been established. A systematic review25 on the subject advised longer healing periods after implant therapy for patients with osteoporosis but could not find contraindications for the use of dental implants for the rehabilitation of that group of patients. This study is not the first attempt to try and use dental tools for the detection of osteoporosis. A series of articles under the label “The OSTEODENT Project” investigated the potential use of two-dimensional panoramic radiographs as a means for the detection of osteoporosis. A positive correlation was found between indices measured on panoramic radiographs (such as
Maxillary and Skeletal Bone Density
7
Figure 5 Variability of BMD on 6 consecutive cross sectionnal slices (slices 32 to 37) (0.75 mm slice thickness and 0.3 mm slice increment). BMD = bone mineral density; HU = Hounsfield units.
cortical texture or cortical width) and skeletal bone density.26–28 In the present study, similar correlations were investigated using three-dimensional imagery. Attention should be given to the choice of techniques and devices used in this study. While the DXA scan is presently recognized as the gold standard for the measurement of skeletal BMD and the detection of osteoporosis, the use of CT scanning for implant surgery purposes can be more controversial due to irradiationrelated and cost-related issues. For the authors of this study, the solution to this problem lies in the ongoing development of the technology, which will inevitably lead to a decrease of the costs and a drastic decrease in irradiation doses. The alternative lies in cone beam CT scanners, where technological advances could lead to the possibility of reliably measuring Hounsfield units or equivalent bone density units. The purpose of this study is not to propose measurements of jaw bone density as a replacement for the validated techniques and procedures currently used for the detection of osteoporosis but to offer an additional screening tool to medical professionals. Because dentists are among the medical professionals most frequently
seen by patients, involving them in the diagnostic process could help improve the screening and early detection of osteoporosis. ACKNOWLEDGMENTS Profs. Steve Boonen (deceased in May 2013, Leuven), Dan Mellström (Gothenburg), Östen Ljunggren (Uppsala), Peter Schneider (Würzburg), and their personnel at the DXA clinics are greatly acknowledged for performing the DXA examinations in this study. A special thank you is addressed to Prof. Hans Mallmin at the Osteoporosis Unit of the University of Uppsala for executing the central reading of the DXA scans. This study was supported by DENTSPLY Implants, Mölndal, Sweden (formerly Astra Tech AB). Dr. Merheb reports grants from DENTSPLY Implants during the conduct of the study. Dr. Kübler reports grants from DENTSPLY Implants during the conduct of the study and outside the submitted work. Dr. Quirynen reports grants from DENTSPLY Implants during the conduct of the study. Dr. Rasmusson reports grants from DENTSPLY Implants during the conduct of the study and outside the submitted work.
8
Clinical Implant Dentistry and Related Research, Volume *, Number *, 2014
Dr. Temmerman reports grants from DENTSPLY Implants during the conduct of the study. Dr. Thor reports grants from DENTSPLY Implants during the conduct of the study. REFERENCES 1. Kanis JA; WHO Study Group. Assessment of fracture risk and its application to screening for postmenopausal osteoporosis: synopsis of a WHO report. Osteoporos Int 1994; 4:368–381. 2. Kanis JA, Black D, Cooper C, et al. A new approach to the development of assessment guidelines for osteoporosis. Osteoporos Int 2002; 13:527–536. 3. Leibson CL, Tosteson AN, Gabriel SE, Ransom JE, Melton LJ. Mortality, disability, and nursing home use for persons with and without hip fracture: a population-based study. J Am Geriatr Soc 2002; 50:1644–1650. 4. Marottoli RA, Berkman LF, Cooney LM Jr. Decline in physical function following hip fracture. J Am Geriatr Soc 1992; 40:861–866. 5. Heaney RP, Abrams S, Dawson-Hughes B, et al. Peak bone mass. Osteoporos Int 2000; 11:985–1009. 6. Raisz LG. Pathogenesis of osteoporosis: concepts, conflicts, and prospects. J Clin Invest 2005; 115:3318–3325. 7. Hui SL, Gao S, Zhou XH, et al. Universal standardization of bone density measurements: a method with optimal properties for calibration among several instruments. J Bone Miner Res 1997; 12:1463–1470. 8. Lu Y, Fuerst T, Hui S, Genant HK. Standardization of bone mineral density at femoral neck, trochanter and Ward’s triangle. Osteoporos Int 2001; 12:438–444. 9. Looker AC, Wahner HW, Dunn WL, et al. Proximal femur bone mineral levels of US adults. Osteoporos Int 1995; 5:389–409. 10. Horner K, Devlin H, Alsop CW, Hodgkinson IM, Adams JE. Mandibular bone mineral density as a predictor of skeletal osteoporosis. Br J Radiol 1996; 69:1019–1025. 11. Drage NA, Palmer RM, Blake G, Wilson R, Crane F, Fogelman I. A comparison of bone mineral density in the spine, hip and jaws of edentulous subjects. Clin Oral Implants Res 2007; 18:496–500. 12. Lindh C, Obrant K, Petersson A. Maxillary bone mineral density and its relationship to the bone mineral density of the lumbar spine and hip. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2004; 98:102–109. 13. Von Wowern N. General and oral aspects of osteoporosis: a review. Clin Oral Investig 2001; 5:71–82. 14. Von Wowern N, Kollerup G. Symptomatic osteoporosis: a risk factor for residual ridge reduction of the jaws. J Prosthet Dent 1992; 67:656–660.
15. Klemetti E, Vainio P. Effect of maxillary edentulousness on mandibular residual ridges. Scand J Dent Res 1994; 102:309– 312. 16. Lindh C, Nilsson M, Klinge B, Petersson A. Quantitative computed tomography of trabecular bone in the mandible. Dentomaxillofac Radiol 1996; 25:146–150. 17. Lee S, Chung CK, Oh SH, Park SB. Correlation between bone mineral density measured by dual-energy x-ray absorptiometry and Hounsfield units measured by diagnostic CT in lumbar spine. J Korean Neurosurg Soc 2013; 54:384–389. 18. Schreiber JJ, Anderson PA, Rosas HG, Buchholz AL, Au AG. Hounsfield units for assessing bone mineral density and strength: a tool for osteoporosis management. J Bone Joint Surg Am 2011; 93:1057–1063. 19. Nilsson M, Johnell O, Jonsson K, Redlund-Johnell I. Quantitative computed tomography in measurement of vertebral trabecular bone mass. A modified method. Acta Radiol 1988; 29:719–725. 20. Taguchi A, Tanimoto K, Ogawa M, Sunayashiki T, Wada T. Effect of size of region of interest on precision of bone mineral measurements of the mandible by quantitative computed tomography. Dentomaxillofac Radiol 1991; 20:25–29. 21. De Oliveira RC, Leles CR, Normanha LM, Lindh C, Ribeiro-Rotta RF. Assessments of trabecular bone density at implant sites on CT images. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2008; 105:231–238. 22. Devlin H, Horner K, Ledgerton D. A comparison of maxillary and mandibular bone mineral densities. J Prosthet Dent 1998; 79:323–327. 23. Shibli JA, Aguiar KC, Melo L, et al. Histologic analysis of human peri-implant bone in type 1 osteoporosis. J Oral Implantol 2008; 34:12–16. 24. Holahan CM, Koka S, Kennel KA, et al. Effect of osteoporotic status on the survival of titanium dental implants. Int J Oral Maxillofac Implants 2008; 23:905–910. 25. Tsolaki IN, Madianos PN, Vrotsos JA. Outcomes of dental implants in osteoporotic patients. A literature review. J Prosthodont 2009; 18:309–323. 26. Devlin H, Allen P, Graham J, et al. The role of the dental surgeon in detecting osteoporosis: the OSTEODENT study. Br Dent J 2008; 204:E16, discussion 560–561. 27. Devlin H, Karayianni K, Mitsea A, et al. Diagnosing osteoporosis by using dental panoramic radiographs: the OSTEODENT project. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2007; 104:821–828. 28. Karayianni K, Horner K, Mitsea A, et al. Accuracy in osteoporosis diagnosis of a combination of mandibular cortical width measurement on dental panoramic radiographs and a clinical risk index (OSIRIS): the OSTEODENT project. Bone 2007; 40:223–229.
ABSTRACT Background and Purpose: Osteoporosis is a disease affecting more than 300 million people worldwide and is responsible for numerous medical complications. This study aimed to investigate the relation between skeletal and maxillary bone density. Materials and Methods: Seventy-three patients were recruited and divided between group A (osteoporosis), group B (healthy, control), and group C (osteopenia) on the basis of a dual-energy x-ray absorptiomery (DXA) scan. These patients also received a CT scan on which bone density measurements were performed at five sites: maxilla midline, retromolar tuberosities, incisor, premolars, and molar regions. Results: The bone density was lower in osteoporotic patients compared with the control patients. The bone mineral density (BMD) of the tuberosities showed the strongest correlations with the BMD of the hip and the spine (respectively, r = 0.50 and r = 0.61). The midline region showed moderate correlations with the hip (r = 0.47) and the spine (r = 0.46). For potential implant sites, the correlations with the BMD of the hip and spine were, however, small to insignificant. Based on measurements of bone density of the maxilla, it was possible to predict if the patient was osteoporotic or not with a sensitivity of 65% and a specificity of 83%. Conclusions: The maxillary bone density of subjects with osteoporosis is significantly lower than that of healthy patients. Moreover, there is a direct correlation between the density of the skeleton and the density of some sites of the maxilla. Using measurements of maxillary bone density in order to predict skeletal bone density might be a useful tool for the screening of osteoporosis. KEY WORDS: bone density, bone mineral density, dual-energy x-ray absorptiometry, Hounsfield units, maxilla, skeletal
INTRODUCTION
to 300 million persons worldwide are affected by osteoporosis,1 especially women.2 Its pathogenesis lies in the rupture of the balance between bone formation (osteocyte/osteoblast activity) and bone resorption (osteoclastic activity). Osteoporosis is considered one of the main causes of hip fractures in the elderly, leading to increased immobility (6 months after hip fracture, only 15% of patients can walk without assistance) and a dramatic rise in mortality rates in the year following the fracture (1 out of 5 people experiencing a hip fracture will die within a year3,4). A number of major risk factors for osteoporosis have been identified. Those include but are not restricted to sex (higher risk for females), ethnic background (higher risk for Caucasians), smoking, alcohol abuse, malnutrition (low intake of calcium or vitamin D increase the risk), and certain medications. Equally, different therapeutic alternatives have been proposed in order to, if not reverse or stop, at least slow
Osteoporosis is a systemic disease characterized by low bone mass and microarchitectural deterioration of bone tissue leading to enhanced bone fragility and a consequent increase in fracture risk.1 According to recent figures from the World Health Organization (WHO), up *Unit of Periodontology, Department of Oral Health Sciences, University of Leuven, Kapucijnenvoer, Belgium; †Scientific Institute of Public Health, Brussels, Belgium; ‡Department of Oral and Maxillofacial Surgery, Sahlgrenska Academy, University of Gothenburg, Göteborg, Sweden; §Department of Oral and Maxillofacial Plastic Surgery, University of Würzburg, Würzburg, Germany; ¶Associate Professor, Department of Oral and Maxillofacial Surgery, Institute of Surgical Sciences, Uppsala University, Uppsala, Sweden Corresponding Author: Dr. Joe Merheb, Unit of Periodontology, Department of Oral Health Sciences, University of Leuven, Belgium; e-mail: [email protected] © 2014 Wiley Periodicals, Inc. DOI 10.1111/cid.12228
1
2
Clinical Implant Dentistry and Related Research, Volume *, Number *, 2014
down bone mass loss. A great emphasis has been placed on the prevention of disease occurrence by maintaining a healthy lifestyle (regular exercise and limitation of tobacco and alcohol use) and a balanced diet ensuring a sufficient intake of calcium and other minerals. Maintaining an adequate diet during the period preceding the adult age has been deemed critical in order to reach a sufficient bone mass peak at the end of skeleton growth.5,6 Additionally, a number of medications have been introduced, among which bisphosphonates are currently the most widely used. Because of its importance and ever increasing costs to society and the social security system, osteoporosis has recently received a lot of attention, and several information and screening campaigns have been initiated. Dentists are among the few medical professionals who see their patients on a regular basis. A large benefit could be derived from involving them in a screening process. Therefore, the aim of this study was to investigate the relation between skeletal and maxillary bone density and to look into the potential to predict skeletal bone density based on density measurements from the maxilla. MATERIALS AND METHODS Patient Selection The study was designed as a multicentric trial and involved four centers: Leuven (Belgium), Gothenburg (Sweden), Uppsala (Sweden), and Würzburg (Germany). The study was reviewed and approved by the ethical committee relevant to each of the participating centers.Patients were deemed eligible for inclusion if they were female, above 60 years of age, partially or fully edentulous in the maxilla, in need of 2 to 8 implants in the maxilla, with a history of edentulism in the planned area of implantation of at least 3 months, and had never received treatment with bisphosphonates. The patients received a detailed explanation about the study, its timeline, its aims, the risks involved, the potential benefits, and the compliance expectations. Patients who met all the inclusion criteria and none of the exclusion criteria signed the informed consent form and underwent a dual-energy x-ray absorptiometry (DXA) scan using either a Hologic™ or a GE-Lunar™ DXA densitometer (Hologic; GE-Lunar) and a spiral computerized tomography (CT) scan. Calibration between the DXA sites was performed via phantom scans, and standardized bone
mineral density (sBMD) values (mg/cm2) were calculated7,8 to compensate for differences between the DXA machines used at the different sites. The DXA scans were assessed centrally at the osteoporosis unit of the University of Uppsala (Sweden). Seventy-three patients were recruited, and for all subjects two regions of interest (ROIs) were defined on the DXA scans, one for all of the hip and one for the lumbar spine (L1 to L4). T-scores for the the two ROIs were calculated based on the difference in density between the examined subject and a genderspecific reference population of white Caucasian females (NHANES III)9 for the hip ROI and on a manufacturerderived gender-specific reference population of white Caucasian females for the lumbar spine ROI. The patients were assigned to either the osteoporosis group (A; T-score 2 −2; n = 18), the control group (B; T-score 3 −1; n = 38), or the osteopenia group (C; −2 < T-score < −1; n = 16) based on the lowest T-scores for the hip or lumbar spine ROIs. (The WHO criteria for osteoporosis rely on a T-score of −2.5 as the threshold for osteoporosis.) Computerized Tomography Analysis Each patient received a CT scan. Data were stored in DICOM format. These DICOM files were loaded in planning software (SimPlant Crystal, Materialise Dental, Leuven, Belgium) and a three-dimensional model was reconstructed. On the three-dimensional reconstruction for each patient, a number of ROIs were defined, and the bone density of the spongy part of each site in Hounsfield units (HU) was recorded. The following ROIs were defined (Figure 1): 1. The maxillary midline: inferior to the anterior nasal spine, between the two central incisors (if present)
Figure 1 Visualization of the different regions of interest (ROI).
Maxillary and Skeletal Bone Density
2. The left and right tuberosities: distal to the third maxillary molar (if present) and distal and inferior to the maxillary sinus 3. The incisor region: inferior and mesial to the lateral wall of the nasal fossae 4. The premolar region: inferior to the mesial half of the maxillary sinus 5. The molar region: inferior to the distal half of the maxillary sinus Whenever possible, bilateral measurements of each ROI were performed (except for the maxillary midline region). In such cases the mean of both values was considered. However, in a number of cases, due to the presence of teeth, implants, artifacts, or other anatomical structures, only unilateral measurements were possible. The explored areas were cylindrical with a diameter of 5 mm and a length corresponding to the maximal extension of the cylinder between the coronal cortex and the cortex of the sinus or nasal floor. Statistical Analysis The difference between the osteoporosis and control groups in ROI values was assessed by a linear mixed model, and the multicenter aspect of the study was modeled by using center as a random factor. Model assumptions were tested by a normal quantile plot of the residuals and a residual dot plot. Pearson correlation coefficients and corresponding p values were calculated to assess the relations between ROI and DXA values. Stepwise model selection was applied to predict the grouping of patients into groups “A” (osteoporosis) and “Not A” (no osteoporosis) based on a generalized linear model for binomial outcomes and a logit link using ROI values as candidate predictive variables. RESULTS Comparison between the Maxillary Bone Densities of Osteoporotic and Healthy Patients The Hounsfield values of the patients in groups A and B were compared at all five reference sites. This comparison yielded a significant difference in bone density between the two groups at all sites. The significance of this difference was greater at the tuberosities and midline sites (p < .01), and in all the comparisons, the osteoporosis group recorded lower density values. The most notable difference was recorded at the tuberosity
3
sites, where the osteoporosis group averaged 13.3 1 72.4 HU in comparison to 126.9 1 102.3 HU in the control group (Figure 2). It was also noteworthy that the measurements of bone density at the tuberosity sites delivered negative HU scores in 7 out of 18 patients (38.9%) of the osteoporosis group, while only 4 out of 38 patients (10.5%) had negative HU scores in the control group. Comparison between the Maxillary and Skeletal Bone Densities The bone mineral densities (BMD) of the hip and the spine were compared with the density of the different intraoral reference sites for all three groups. Significant correlations between the BMD of the hip and those of the two nondental sites (tuberosity, midline) were found. They reached r = 0.50 (p < .01) at the tuberosity sites and r = 0.47 (p < .01) at the midline site. At the dental sites, only the BMD of the premolar region correlated slightly with the BMD of the hip (r = 0.3, p = .01) (Figure 3). A similar trend was observed when the BMD of the spine was compared with the same intraoral sites. A correlation of r = 0.61 (p < .01) was found between the BMD of the spine and that of the tuberosities. A correlation of r = 0.46 (p < .01) was found between the BMD of the spine and that of the midline. At the dental sites, the correlations between the BMD of the spine and those of the incisor (r = 0.24, p = .04) and premolar (r = 0.31, p < .01) regions reached significance (Figure 4). Prediction of Skeletal Bone Group Based on Measurements of Maxillary Bone A stepwise model selection was applied to predict the grouping of patients into groups “A” (osteoporosis) and “Not A” (groups B and C) based on a generalized linear model for binomial outcomes and a logit link. It appeared that the tuberosities and midline had the best predictive power. The following formula was applied:
Sign (3.034636 − 0.0160303 × Tuberosities − 0.005359683 × Midline) A positive value led to a prediction of group “A” and a negative value to a prediction of group “not A.”
4
Clinical Implant Dentistry and Related Research, Volume *, Number *, 2014
Tuberosity (average L + R)
Midline 800
300 600
HU
HU
200 100
400
0 –100
200
–200
A = 13.3 ± 72.4 HU
B = 126.9 ± 102.3 HU
Incisor
A = 414.0 ± 146.8 HU
Premolar
B = 552.6 ± 128.0 HU
Molar
600
600
400
400
HU
HU
400 200
HU
600
200
200 0
0
A = 260.3 ± 145.4 HU
B = 372.5 ± 178.6 HU
A = 242.2 ± 123.5 HU
B = 318.5 ± 182.9 HU
A = 139.7 ± 166.8 HU
B = 210.2 ± 167.2 HU
Figure 2 Comparison between the maxillary BMD of osteoporotic and healthy patients. *p < .05; **p < .01. A = osteoporosis group; B = control group; BMD = bone mineral density; HU = Hounsfield units; L = left; R = right.
The selected model allowed the prediction of the group to which the patient belonged with a moderate level of sensitivity (0.65%) and a high level of specificity (0.83%). DISCUSSION Several studies have investigated the relation of skeletal and jaw bone density. There was no consensus as to the existence or absence of a direct relation. In a pioneer study, Horner and colleagues10 performed DXA scans on both skeletal and jaw bone sites and found significant correlations between the densities of the femoral neck and lumbar spine and those of all the mandibular sites they tested (mandibular symphysis, mandibular body and ramus) with the strongest correlations being recorded at the mandibular body site. In a study using similar measuring devices, Drage and colleagues11 detected a significant correlation of the BMD of the ramus with that of the femur and the lumbar spine. However, contrarily to the Horner study, the BMD of the other jaw bone sites, namely the body of the mandible, the mandibular symphysis, and the anterior maxilla did not correlate well with either of the two skeletal BMD
measurements. While most studies report on the relationship of mandibular bone with skeletal bone, Lindh and colleagues12 investigated the relationship between the density of maxillary bone derived from CT scans and the skeletal bone density of the spine and hip derived from DXA scans. This investigation found inconsistent correlations between the maxillary and skeletal bone densities. The incisor sites correlated significantly with the lumbar spine sites, while most other dental sites did not correlate significantly with either the spine or hip density measurements. This discrepancy might be explained by the very low number of subjects (n = 8) enrolled in this study. When investigating the relation between jaw bone density and skeletal bone density, most studies have relied on measurements performed on the mandible, and rarely were there maxillary sites involved. According to von Wowern, this is due to the fact that the basal area of the mandible posterior to the mental foramen is the only region of the jaws that satisfies the requirements for a standard measurement site in regard to anatomical size, shape, bone structure, and function.13 The authors of the present study found, however, that the maxilla,
Maxillary and Skeletal Bone Density
B
300 200
BMD Midline
BMD Tuberosity
A
100 0 –100
5
800
600
400
200
–200 600
800
1,000
1,200
600
sBMD (hip) r = 0.5, p < 0.01 D
400 200 0 600
800
1,000
1,200
sBMD (hip) r = 0.22, p = 0.06
400 200 0 600
800
1,000
1,200
sBMD (hip) (NS)
1,200
E 600
600
BMD Molar
600
1,000
sBMD (hip) r = 0.47, p < 0.01 BMD Premolar
BMD Incisor
C
800
r = 0.30, p = 0.01
400 200 0 600
800
1,000
1,200
sBMD (hip) r = 0.20, p = 0.1
(NS)
Figure 3 Relation between the BMD of the hip and the different BMD of the maxilla. BMD = bone mineral density (g/cm ); sBMD = standardized bone mineral density (mg/cm2). 2
even if it did not offer the same possibility for standardization of measurements, still included useful measurement sites because it was more prone to osteoporosis than the mandible.14 This vulnerability to osteoporosis is probably due to the higher prevalence of trabecular structures in the maxilla as compared with the mandible.15,16 The use of Hounsfield units as an indicator of bone mineral content (BMC) has been assessed by several publications, and a positive correlation between Hounsfield values and bone mineral density has been found.17,18 The reliability of this correlation is, however, dependent on several factors, among which age plays an important role. A method to correct for age-dependent composition of trabecular bone (BMC calibration) was proposed by Nilsson and colleagues.19 While calibration of the CT equipment and sensitivity adjustment were performed daily in the context of this study, no BMC calibration or attempt to transform Hounsfield values into BMC was performed. This might be a shortcoming of the study, but one which is possibly of limited influence, given the the homogeneity of the population sample studied (exclusively women, limited age range of 60–76 years old, exclusively Caucasian, no extreme body
mass index values). Previous studies comparing jaw bone density to skeletal bone density have always performed their jaw bone density measurements on single two-dimensional slices, whereas the present study utilized three-dimensional cylinders that covered five to seven consecutive cross-sectional slices. This allows the selected ROIs to have a greater surface by the addition of the surfaces of the different areas covered by the cylinders on the consecutive slices, thereby increasing the reliability of the performed measurements. The total area exceeded the surface area of 1 cm2 defined as the threshold surface area for reliable measurements.20 Additionally, the choice of a volume ROI is justified by the great variability of bone density on a local level, even on consecutive slices as can be seen in Figure 5. In the present study, the strongest correlations between the skeletal and jaw bone were recorded at the tuberosities. A correlation of r = 0.61 with the BMD of the spine and of r = 0.50 with the BMD of the hip were found. These results support the original study hypothesis that maxillary sites are also useful for the detection of osteoporosis due to their trabecular composition. The posterior maxilla and tuberosities are the regions of the jaw with the highest proportion of trabecular bone.21,22
6
Clinical Implant Dentistry and Related Research, Volume *, Number *, 2014
B
300 200
BMD Midline
BMD Tuberosity
A
100 0 –100
800
600
400
200
–200 600
800
1,000
1,200
1,400
600
1,600
800
sBMD (spine) r = 0.61, p < 0.01 D
400 200 0 600
800 1,000 1,200 1,400 1,600
sBMD (spine) r = 0.24, p = 0.04
1,400
1,600
E 600
600
BMD Molar
600
1,200
sBMD (spine) r = 0.46, p < 0.01
BMD Premolar
BMD Incisor
C
1,000
400 200 0 600 800 1,000 1,200 1,400 1,600
sBMD (spine) r = 0.31, p < 0.01
400 200 0 600 800 1,000 1,200 1,400 1,600
sBMD (spine) r = 0.17, p = 0.16
(NS)
Figure 4 Relation between the BMD of the spine and the different BMD of the maxilla. BMD = bone mineral density (g/cm ); sBMD = standardized bone mineral density (mg/cm2). 2
Even though the calculated correlations between the tuberosity sites and the two skeletal sites are far from being linear, it is important to relativize by calculating the correlation between the two skeletal sites (hip and spine), which in this study amounted to r = 0.77. Given that even these two standard reference sites do not correlate perfectly, it can be suggested that the BMD of the tuberosities is a reliable reference that could possibly be used for the prediction of skeletal osteoporosis. The stepwise model analysis performed the extraction of the sensitivity and specificity of the selected model. While a moderate level of sensitivity and a high level of specificity were found, those results could be noticeably improved if the test could be applied to a bigger sample. The present study showed that the relation between the density of the skeleton and the density of the dental sites (incisor, premolar, molar) was weak to moderate. It also showed that the difference in spongy bone density between osteoporotic and nonosteoporotic patients was less evident at the dental sites than at the nondental sites. In the context of implant therapy, those findings seem to suggest that skeletal bone density might have only a limited influence on
implant therapy success. Those results are in line with the conclusions of previous studies on the effect of osteoporosis on implant therapy outcome. Histological observations by Shibli and colleagues23 showed no difference in bone-implant contact between the osteoporosis and control groups. While the control group showed an average of 48%, the osteoporosis group showed an average of 46%. Similarly, a large study24 on 746 women (3,224 implants) showed that patients with an established diagnosis of osteoporosis were not more likely to have dental implant failure than patients in whom the diagnosis of osteoporosis had not been established. A systematic review25 on the subject advised longer healing periods after implant therapy for patients with osteoporosis but could not find contraindications for the use of dental implants for the rehabilitation of that group of patients. This study is not the first attempt to try and use dental tools for the detection of osteoporosis. A series of articles under the label “The OSTEODENT Project” investigated the potential use of two-dimensional panoramic radiographs as a means for the detection of osteoporosis. A positive correlation was found between indices measured on panoramic radiographs (such as
Maxillary and Skeletal Bone Density
7
Figure 5 Variability of BMD on 6 consecutive cross sectionnal slices (slices 32 to 37) (0.75 mm slice thickness and 0.3 mm slice increment). BMD = bone mineral density; HU = Hounsfield units.
cortical texture or cortical width) and skeletal bone density.26–28 In the present study, similar correlations were investigated using three-dimensional imagery. Attention should be given to the choice of techniques and devices used in this study. While the DXA scan is presently recognized as the gold standard for the measurement of skeletal BMD and the detection of osteoporosis, the use of CT scanning for implant surgery purposes can be more controversial due to irradiationrelated and cost-related issues. For the authors of this study, the solution to this problem lies in the ongoing development of the technology, which will inevitably lead to a decrease of the costs and a drastic decrease in irradiation doses. The alternative lies in cone beam CT scanners, where technological advances could lead to the possibility of reliably measuring Hounsfield units or equivalent bone density units. The purpose of this study is not to propose measurements of jaw bone density as a replacement for the validated techniques and procedures currently used for the detection of osteoporosis but to offer an additional screening tool to medical professionals. Because dentists are among the medical professionals most frequently
seen by patients, involving them in the diagnostic process could help improve the screening and early detection of osteoporosis. ACKNOWLEDGMENTS Profs. Steve Boonen (deceased in May 2013, Leuven), Dan Mellström (Gothenburg), Östen Ljunggren (Uppsala), Peter Schneider (Würzburg), and their personnel at the DXA clinics are greatly acknowledged for performing the DXA examinations in this study. A special thank you is addressed to Prof. Hans Mallmin at the Osteoporosis Unit of the University of Uppsala for executing the central reading of the DXA scans. This study was supported by DENTSPLY Implants, Mölndal, Sweden (formerly Astra Tech AB). Dr. Merheb reports grants from DENTSPLY Implants during the conduct of the study. Dr. Kübler reports grants from DENTSPLY Implants during the conduct of the study and outside the submitted work. Dr. Quirynen reports grants from DENTSPLY Implants during the conduct of the study. Dr. Rasmusson reports grants from DENTSPLY Implants during the conduct of the study and outside the submitted work.
8
Clinical Implant Dentistry and Related Research, Volume *, Number *, 2014
Dr. Temmerman reports grants from DENTSPLY Implants during the conduct of the study. Dr. Thor reports grants from DENTSPLY Implants during the conduct of the study. REFERENCES 1. Kanis JA; WHO Study Group. Assessment of fracture risk and its application to screening for postmenopausal osteoporosis: synopsis of a WHO report. Osteoporos Int 1994; 4:368–381. 2. Kanis JA, Black D, Cooper C, et al. A new approach to the development of assessment guidelines for osteoporosis. Osteoporos Int 2002; 13:527–536. 3. Leibson CL, Tosteson AN, Gabriel SE, Ransom JE, Melton LJ. Mortality, disability, and nursing home use for persons with and without hip fracture: a population-based study. J Am Geriatr Soc 2002; 50:1644–1650. 4. Marottoli RA, Berkman LF, Cooney LM Jr. Decline in physical function following hip fracture. J Am Geriatr Soc 1992; 40:861–866. 5. Heaney RP, Abrams S, Dawson-Hughes B, et al. Peak bone mass. Osteoporos Int 2000; 11:985–1009. 6. Raisz LG. Pathogenesis of osteoporosis: concepts, conflicts, and prospects. J Clin Invest 2005; 115:3318–3325. 7. Hui SL, Gao S, Zhou XH, et al. Universal standardization of bone density measurements: a method with optimal properties for calibration among several instruments. J Bone Miner Res 1997; 12:1463–1470. 8. Lu Y, Fuerst T, Hui S, Genant HK. Standardization of bone mineral density at femoral neck, trochanter and Ward’s triangle. Osteoporos Int 2001; 12:438–444. 9. Looker AC, Wahner HW, Dunn WL, et al. Proximal femur bone mineral levels of US adults. Osteoporos Int 1995; 5:389–409. 10. Horner K, Devlin H, Alsop CW, Hodgkinson IM, Adams JE. Mandibular bone mineral density as a predictor of skeletal osteoporosis. Br J Radiol 1996; 69:1019–1025. 11. Drage NA, Palmer RM, Blake G, Wilson R, Crane F, Fogelman I. A comparison of bone mineral density in the spine, hip and jaws of edentulous subjects. Clin Oral Implants Res 2007; 18:496–500. 12. Lindh C, Obrant K, Petersson A. Maxillary bone mineral density and its relationship to the bone mineral density of the lumbar spine and hip. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2004; 98:102–109. 13. Von Wowern N. General and oral aspects of osteoporosis: a review. Clin Oral Investig 2001; 5:71–82. 14. Von Wowern N, Kollerup G. Symptomatic osteoporosis: a risk factor for residual ridge reduction of the jaws. J Prosthet Dent 1992; 67:656–660.
15. Klemetti E, Vainio P. Effect of maxillary edentulousness on mandibular residual ridges. Scand J Dent Res 1994; 102:309– 312. 16. Lindh C, Nilsson M, Klinge B, Petersson A. Quantitative computed tomography of trabecular bone in the mandible. Dentomaxillofac Radiol 1996; 25:146–150. 17. Lee S, Chung CK, Oh SH, Park SB. Correlation between bone mineral density measured by dual-energy x-ray absorptiometry and Hounsfield units measured by diagnostic CT in lumbar spine. J Korean Neurosurg Soc 2013; 54:384–389. 18. Schreiber JJ, Anderson PA, Rosas HG, Buchholz AL, Au AG. Hounsfield units for assessing bone mineral density and strength: a tool for osteoporosis management. J Bone Joint Surg Am 2011; 93:1057–1063. 19. Nilsson M, Johnell O, Jonsson K, Redlund-Johnell I. Quantitative computed tomography in measurement of vertebral trabecular bone mass. A modified method. Acta Radiol 1988; 29:719–725. 20. Taguchi A, Tanimoto K, Ogawa M, Sunayashiki T, Wada T. Effect of size of region of interest on precision of bone mineral measurements of the mandible by quantitative computed tomography. Dentomaxillofac Radiol 1991; 20:25–29. 21. De Oliveira RC, Leles CR, Normanha LM, Lindh C, Ribeiro-Rotta RF. Assessments of trabecular bone density at implant sites on CT images. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2008; 105:231–238. 22. Devlin H, Horner K, Ledgerton D. A comparison of maxillary and mandibular bone mineral densities. J Prosthet Dent 1998; 79:323–327. 23. Shibli JA, Aguiar KC, Melo L, et al. Histologic analysis of human peri-implant bone in type 1 osteoporosis. J Oral Implantol 2008; 34:12–16. 24. Holahan CM, Koka S, Kennel KA, et al. Effect of osteoporotic status on the survival of titanium dental implants. Int J Oral Maxillofac Implants 2008; 23:905–910. 25. Tsolaki IN, Madianos PN, Vrotsos JA. Outcomes of dental implants in osteoporotic patients. A literature review. J Prosthodont 2009; 18:309–323. 26. Devlin H, Allen P, Graham J, et al. The role of the dental surgeon in detecting osteoporosis: the OSTEODENT study. Br Dent J 2008; 204:E16, discussion 560–561. 27. Devlin H, Karayianni K, Mitsea A, et al. Diagnosing osteoporosis by using dental panoramic radiographs: the OSTEODENT project. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2007; 104:821–828. 28. Karayianni K, Horner K, Mitsea A, et al. Accuracy in osteoporosis diagnosis of a combination of mandibular cortical width measurement on dental panoramic radiographs and a clinical risk index (OSIRIS): the OSTEODENT project. Bone 2007; 40:223–229.
