- Email: [email protected]
Quantitative evaluation of maxillary alveolar cortical bone thickness and density using computed tomography imaging
ORIGINAL ARTICLE
Quantitative evaluation of maxillary alveolar cortical bone thickness and density using computed tomography imaging Henry Ohiomoba,a Andrew Sonis,b Alfa Yansane,c and Bernard Friedlandd Boston, Mass
Introduction: Primary stability is essential to the success of orthodontic mini-implants (OMIs) and heavily depends on the mechanical retention between OMIs and their supporting bone. Alveolar cortical bone commonly serves as the supporting bone for OMIs during treatment. The purposes of this study were to characterize alveolar cortical bone thickness and density in the maxilla and to explore patient factors that may significantly affect these bone properties. Methods: Sixty medical computed tomography scans of the maxilla were analyzed from a selected sample of patients seen at the Radiology Department of Boston Children's Hospital. Interradicular alveolar bone thickness and density were measured at 2, 4, 6, and 8 mm from the buccal and palatal alveolar bone crests using the Synapse 3D software (version 4.1; FUJIFILM Medical Systems USA, Stamford, Conn). Analyses were conducted with STATA /1C (version 12.0 for Windows; StataCorp, College Station, Tex) using multivariate mixed-effects regression models and paired t tests. Results: Mean age and body mass index of the study sample were 17.88 years and 22.94 kg/m2, respectively. Cortical bone density and thickness significantly increased from the coronal (2 mm) to the apical (8 mm) regions of the alveolar bone (P \0.05). At 8 mm from the alveolar crest, interradicular buccal cortical bone was thickest (1 mm) and densest (1395 Hounsfield units) between the first and second molars. On the palatal side, the thickest bone (1.15 mm) was found between the canine and first premolar; it was similarly densest (1406 Hounsfield units) between the first premolar and canine, and between the first premolar and second premolar interradicular bones. On average, palatal cortical bone was thicker and denser compared with buccal; this difference was statistically significant (P \0.01) in the anterior and middle maxilla, with the anterior maxillary region showing the greatest difference. Female subjects have significantly denser bone compared with male subjects; however, sex is not significantly associated with bone thickness. Body mass index and age are positively associated with bone thickness and density. Radiologic absence of bone was more commonly seen in the anterior maxilla. Conclusions: Alveolar bone properties vary in the maxilla in patterns that could guide clinicians in selecting sites best suited for placement of OMIs. (Am J Orthod Dentofacial Orthop 2017;151:82-91)
A
bsolute anchorage is often a desired goal during orthodontic treatment; this can be difficult to achieve, since most anchorage biomechanics are tooth-supported, leading to unwanted side effects From the Harvard School of Dental Medicine, Boston, Mass. a Orthodontic resident. b Pediatric Dentistry and Orthodontics, and clinical professor; Department of Dentistry, Boston Children's Hospital, Boston, Mass. c Oral Health Policy and Epidemiology. d Department of Oral Medicine Infection and Immunity. All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest, and none were reported. Address correspondence to: Henry Ohiomoba, Holyoke Health Center - 230 Maple St, Holyoke, MA 01040; e-mail, [email protected]. Submitted, July 2015; revised and accepted, May 2016. 0889-5406/$36.00 Ó 2017 by the American Association of Orthodontists. All rights reserved. http://dx.doi.org/10.1016/j.ajodo.2016.05.015
82
such as translation or tipping of anchoring teeth. In recent years, orthodontic mini-implants (OMIs) have provided a solution to the problem of unpredictable anchorage because they are bone-supported. A 2008 survey reported that approximately 80% of orthodontists use OMIs, and 78% of them believe that OMIs improve treatment outcomes.1 Common sites for placement of OMIs are the maxillary and mandibular buccal alveolar bones, infrazygomatic crest, palatal alveolar bone, and paramedian palate.2 A major limitation of OMIs is loosening (mobility) and subsequent failure, which could lead to premature removal and longer treatment times. Failure rates of OMIs under orthodontic loading have been reported to range from 11% to 30%,3 whereas success rates of 76% to 87% have also
Ohiomoba et al
been reported.4 Possible risk factors that affect the stability of OMIs are supporting bone quality and quantity, screw design, force loads, and safe placement techniques.5 Some controversy exists over the significance of some of these risk factors with regard to what is optimal; however, it is generally accepted that increased thickness of cortical bone is favorable for OMI stability. Iijima et al6 in 2012 showed that cortical bone thickness, total bone mineral density, cortical bone mineral density, and bone hardness were significantly related to the mean failure force of OMIs. Studies that look to characterize cortical bone thickness and density in the alveolar bone have the limitation of small sample sizes, in-vitro studies on skulls, or failing to explore patient factors such as age and sex on cortical bone properties.7 In addition, cone-beam computerized tomography (CBCT) has several limitations compared with standard medical computerized tomography (MCT) for bone evaluation; Hounsfield units that measure density are not accurate, and increased scatter radiation and artifacts caused by beam hardening are all associated with CBCT.8-11 In this study, we used MCT to quantitatively characterize cortical bone thickness and density in the maxillary alveolar bone and to investigate the influence of patient factors to determine optimal locations for placement of OMIs during orthodontic treatment. MATERIALS AND METHODS
The study sample consisted of 60 high-resolution MCT scans of the maxillofacial region taken from a pool of patients who had been seen in the Radiology Department of Boston Children's Hospital for reasons unrelated to this study such as chronic sinusitis and presence of a foreign body in the sinuses (30 female, 30 male). Sample inclusion criteria included patients aged 12 years or older with permanent dentition and a diagnosis of “normal” or “unremarkable” in the radiologist's report of the MCT scan. Exclusion criteria included significant radiographic signs of periodontal disease, moderate to severe overlapping of crowns or roots of adjacent teeth, trauma to the maxillofacial region, multiple missing teeth, and a positive history of systemic conditions, pathologies, or therapy that may significantly affect cortical bone properties. Scans were captured using a GE LightSpeed Pro 32 Slice CT unit (GE Healthcare, Little Chalfont, United Kingdom) at 120 KVp, 50 mA, 0.6-mm slice thickness, and mean spatial resolution of 0.63 mm. The images were analyzed using the dental multiplanar reconstruction interphase in the Synapse 3D software (version 4.1; FUJIFILM Medical Systems USA, Stamford, Conn). All reconstructions were
83
Fig 1. CT cross-sectional view perpendicular to the axial plane showing thickness (mm) and average density measurements.
done at 0.6 mm with a high spatial frequency bone algorithm. A reference plane parallel to the maxillary occlusal plane was established to generate a 2-dimensional panoramic image of the maxilla and to aid in generating 2dimensional cross-sectional images. Two-dimensional cross-sectional images of the maxilla were reconstructed at 1-mm intervals (150 cross-sections per scan) and viewed under settings appropriate for bone details (window width, 2000; window level, 400) and sharpened to best delineate cortical boundaries.12 Best cross-sectional slices through each contact area that clearly delineated interradicular bone without root anatomy were selected for measurements, beginning distally to the right second molar and ending distally to the left second molar. Thickness measurements were made perpendicular to the bone surface using the function “ruler,” and average density measurements were made using the corresponding oval density tool, both in the Synapse 3D software. The demarcation between cancellous and cortical bone was drawn manually by visual gray-white discrimination; gray was perceived as cancellous bone, and white was perceived as cortical bone. Palatal and buccal alveolar cortical bone thickness and density were documented in millimeters and Hounsfield units, respectively. These were measured at 2, 4, 6, and 8 mm from the alveolar bone crest along the surface of the cortical bone (Fig 1). Thus, a maximum of 120 alveolar thickness and 120 mean alveolar density measurements were documented per MCT scan. For additional analysis, measurements were further grouped into anterior maxilla (interradicular bone from the mesial aspect of the canine to the mesial aspect of the contralateral canine), middle maxilla (interradicular bone between the distal aspect of the canines to the mesial aspect of the first molars), and posterior maxilla (interradicular bone between the distal aspect of the first molars to the mesial aspect of the third molars). Training support for image analysis was provided
American Journal of Orthodontics and Dentofacial Orthopedics
January 2017 Vol 151 Issue 1
Ohiomoba et al
84
by the Radiology Advanced Image Analysis Laboratory at Boston Children's Hospital. All measurements were performed by a primary rater (H.O.) and entered into an Excel 2010 spreadsheet version 14.0; (Microsoft, Redmond, Wash). Two hundred randomly selected measurements were repeated, 1 month later, by both the primary rater and a second calibrated rater (A.S.) to assess for intrarater and interrater reliabilities. Both raters were required to recreate new 2-dimensional crosssectional images using the technique described earlier during reliability measurements. Protocol approval for the study was received from the Committee on Clinical Investigation at Boston Children's Hospital, and the Committee on Human Studies at Harvard Medical School and Harvard School of Dental Medicine. Statistical analysis
Descriptive analyses of patient demographics were summarized using means and standard deviations for continuous variables and frequencies with corresponding proportions for the categorical variables. The intraclass correlation coefficient was used to determine both interrater and intrarater reliabilities. Bland-Altman plots were also used to assess agreement between the raters’ measurements. The 2-sample paired t test was used to determine differences in thickness and density between palatal vs buccal cortical bone; because of the expected correlation of intrapatient measurements, the patient was used as the unit of analysis by averaging the total buccal and palatal measurement values for each patient. Multivariate mixedeffects regression models (patient identity was included as a random effect to account for within-patient correlation) were used to explore the effects of alveolar bone height and multiple patient factors such as sex, age, body mass index (BMI), and ethnicity on alveolar bone thickness and density. To test for multiple comparisons on the effect of alveolar bone height on both thickness and density, we used the Bonferroni method for multiple comparisons to adjust for type 1 error inflation. For the regression models, the BMI and age variables were divided into 3 ordinal groups, less than 21 Kg/m2, 21 to 25 Kg/m2, and greater than 25 Kg/m2; and 12 to 16 years, more than 16 but less than 20 years, and 20 years or more, respectively. All tests were conducted at the significance level of a 5 0.05 as is standard, and all statistical analysis were performed using STATA /1C software (version 12.0 for Windows; StataCorp, College Station, Tex). RESULTS
The mean age of our study sample was 17.88 years (SD, 4.71 years), and the mean BMI was 22.94 kg/m2 (SD, 4.13 kg/m2). Table I shows the frequency and
January 2017 Vol 151 Issue 1
Table I. Demographic distribution Age group (y) 12-16 \16-\20 $20 BMI group \21 21-25 .25 NA Ethnicity White Black Hispanic Asian American Indian NA
Male (%)
Female (%)
Total (%)
8 (13.3) 17 (28.3) 5 (8.3)
12 (20) 13 (21.7) 5 (8.3)
20 (33.3) 30 (50) 10 (16.7)
14 (23.3) 8 (13.3) 8 (13.3) -
5 (8.3) 13 (21.67) 8 (13.3) 4 (6.7)
19 (31.7) 21 (35) 16 (26.7) 4 (6.7)
11 (18.3) 9 (15) 3 (5) 3 (5) 4 (6.7)
18 (30) 10 (16.7) 1 (1.7) 1 (1.7)
29 (48.3) 19 (31.7) 3 (5) 3 (5) 1 (1.7) 5 (8.3)
NA, Not answered.
percentage distributions of the study population by age, sex, BMI, and ethnicity. Approximately half (48%) of the study sample self-identified as white, 32% selfidentified as black, and the rest identified as Hispanic (5%), Asian (5%), or American Indian (2%), or chose not to self-identify (8%). Intrarater reliabilities (absolute agreement) within the primary rater were r 5 0.98 for density and r 5 0.90 for thickness; this is near perfect agreement. Interrater reliability (absolute agreement) comparing the primary rater with the secondary rater was substantial for density (r 5 0.78) and moderate for thickness (r 5 0.60) measurements.13 BlandAltman plots (Figs 2 and 3) exploring the agreement between the primary rater vs the secondary rater for the thickness and density measurements showed no evidence of a difference in variability between the raters (P value for Pitman's test of difference in variability in both plots were .0.05); the limits of agreement were 0.512 to 0.599 (mean difference, 0.043) for thickness measurements and 299.250 to 310.336 (mean difference, 5.543) for density measurements. The results of the paired t test showed that, on average, palatal bone was thicker and denser than buccal bone; this difference was statistically significant (\0.01) at the anterior and middle regions of the maxilla but not at the posterior maxilla (Tables II and III). Multivariate mixed-effects models controlling for patient identity, alveolar height, sex, ethnicity, BMI group, and age group were used to explore the effect of multiple patient factors on alveolar bone properties simultaneously (Table IV). On average, alveolar bone thickness and density increased significantly the farther we measured from the alveolar crest, with regions 8 mm
American Journal of Orthodontics and Dentofacial Orthopedics
Ohiomoba et al
85
Fig 2. Bland-Altman plot of differences in alveolar cortical bone thickness (mm) measurements between raters 1 and 2. Limits of agreement (reference range for difference), 0.512 to 0.599; mean difference, 0.043 (95% CI, 0.016, 0.102); Pitman's test of difference in variance, r 5 0.119, P 5 0.268.
Fig 3. Bland-Altman plot of differences in alveolar cortical bone density (Hounsfield units) measurements between raters 1 and 2. Limits of agreement (reference range for difference), 299.250 to 310.336; mean difference, 5.543 (95% CI, 26.195, 37.281); Pitman's test of difference in variance, r 5 0.035, P 5 0.741.
from the alveolar crest having the greatest thickness and density. At 8 mm from the alveolar crest, buccal interradicular bone is thickest and densest between the second molar and first molar, followed by between the first molar and second premolar; it is least thick between the lateral incisor and central incisor, and least dense
between the central incisors (Figs 4 and 5). With regard to the palatal aspect at 8 mm, the bone is thickest between the first premolar and canine, followed by between the premolars; it is similarly densest between the first premolar and canine, and between the premolars. It is least thick between the
American Journal of Orthodontics and Dentofacial Orthopedics
January 2017 Vol 151 Issue 1
Ohiomoba et al
86
Table II. Palatal vs buccal thickness in millimeters
Anterior maxilla Middle maxilla Posterior maxilla
P Mean (SD) 0.90 (0.18) 0.96 (0.18) 0.78 (0.14)
B Mean (SD) 0.71 (0.11) 0.78 (0.16) 0.76 (0.14)
P B Diff (SD) 0.19 (0.14) 0.18 (0.16) 0.02 (0.13)
Degrees of freedom 56 59 59
t test value 10.50 8.50 1.52
P value \0.001 \0.001 0.134
Degrees of freedom 56 59 59
t test value 3.88 3.43 1.69
P value \0.001 0.001 0.096
Paired t test. P, Palatal; B, buccal; Diff, average difference.
Table III. Palatal vs buccal density in Hounsfield units
Anterior maxilla Middle maxilla Posterior maxilla
P Mean (SD) 1288 (117) 1340 (104) 1252 (145)
B Mean (SD) 1227 (121) 1291 (138) 1224 (174)
P B Diff (SD) 61 (118) 49 (110) 28 (130)
Paired t test. P, Palatal; B, buccal; Diff, average difference.
second molar and third molar, and least dense between the central incisors (Figs 4 and 5). Cortical bone could not be delineated at all the 8-mm measurement points. On the buccal side, the interradicular regions with the highest incidence of this occurrence compared with other buccal sites were located between the central incisors (56%), between the first and second premolars (12%), and between the lateral and central incisors (10%). A similar pattern was seen on the palatal side, with the highest incidences compared with other palatal sites seen between the central incisors (89%) and between the lateral and central incisors (6%). The results of the multivariate models (Table IV) show that sex had no significant effect on thickness (P 5 0.77), but female subjects had significantly denser bone compared with male subjects (P \0.0001). With regard to ethnicity, most groups had 5 or fewer persons; in comparing the 2 largest groups—white (n 5 29) vs black (n 5 19) patients—there appeared to be no significance difference in alveolar bone thickness (P 5 0.28), but black subjects may have less dense alveolar bone (P 5\0.0001) compared with white subjects. Increases in BMI were significantly associated with increases in bone thickness and density, with patients in the highest BMI group (.25 kg/m2) having the highest density and thickness. Increase in age was significantly associated with increased alveolar bone density, with the age group greater than 20 years having the highest density compared with the 2 younger groups. However, the effect of age on thickness appeared to plateau by age 16, with subjects 16 years or older having similar cortical bone thickness, a value that was significantly greater (P \0.0001) compared with the youngest age group between 12 and 16 years.
January 2017 Vol 151 Issue 1
DISCUSSION
The reported absolute numbers for alveolar cortical bone thickness differ in the literature. This could be due to the underreporting and nonstandardization of visualization parameters such as window width and window length by authors. For our study, we used a window width of 2000 and a window length of 400; this is comparable with the parameters used at the Boston Children's Hospital Department of Radiology for diagnosing bone pathology. Arctander et al12 also used similar window width (2100) and window length (500) for appropriately viewing bone details. Knowledge of the pattern of variation in alveolar bone properties in the maxilla may, however, prove to be more clinically useful than the absolute values for these properties. Thus, the purpose of this study was to highlight predictable patterns of alveolar bone properties in the maxilla and the influence of patient factors in the context of bone thickness and density. Although a CT scan of every patient before insertion of OMIs may prove beneficial, the downside of this protocol (eg, radiation exposure, expense, and accessibility to a CT scanner) makes this approach impractical. Results from our study provide a guideline and may largely negate the need for CT scans before OMI insertion. Also, ours is one of the few studies to characterize alveolar bone density in the maxilla, information that is not easily obtained from CBCT imaging.14-16 The reliability results showed excellent intrarater agreements but moderate to substantial interrater agreements; this could be expected because of the subjectivity of the manual visual gray-white
American Journal of Orthodontics and Dentofacial Orthopedics
Ohiomoba et al
87
Table IV. Multivariate mixed-effects regression anal-
ysis showing influence of patient factors on alveolar bone properties (thickness and density) Coef Thickness Alveolar height: 2 mm 4 mm 0.11 6 mm 0.21 8 mm 0.26 Sex: male Female 0.00 Ethnicity: white Asian 0.09 American Indian 0.12 Black 0.01 Hispanic 0.05 NA 0.10 BMI group: \21 21-25 0.04 .25 0.09 Age group: 12-16 y .16-\20 y 0.09 .20 y 0.08 Density Alveolar height: 2 mm 4 mm 109.18 6 mm 179.45 8 mm 208.37 Sex: male Female 42.22 Ethnicity: white Asian 94.93 American Indian 32.16 Black 22.38 Hispanic 24.64 NA 34.80 BMI group: \21 21-25 39.78 .25 87.88 Age group: 12-16 y \16-\20 y 124.11 #20 y 174.23
SE
Sig
0.01 \0.01 0.01 \0.01 0.01 \0.01 0.01
0.84
95% CI
0.09, 0.12 0.19, 0.22 0.24, 0.28 0.02, 0.02
0.02 \0.01 0.03 \0.01 0.01 0.38 0.02 0.02 0.01 \0.01
0.12, 0.05 0.07, 0.18 0.03, 0.01 0.01, 0.08 0.13, 0.07
0.01 \0.01 0.01 \0.01
0.02, 0.05 0.07, 0.11
0.01 \0.01 0.01 \0.01
0.07, 0.11 0.06, 0.11
8.55 \0.01 8.52 \0.01 8.56 \0.01
92.42, 125.94 162.74, 196.15 191.58, 225.16
7.23 \0.01
28.04, 56.40
14.25 \0.01 24.11 0.18 7.36 \0.01 15.19 0.11 11.54 \0.01
122.87, 66.98 15.09, 79.41 38.81, 7.95 54.41, 5.14 12.19, 57.41
7.60 \0.01 8.60 \0.01
24.88, 54.68 71.00, 104.75
7.13 \0.01 10.05 \0.01
110.14, 138.09 154.53, 193.93
Patient identity included as a random effect. Coef, Coefficient; Sig, significance; NA, not answered.
discrimination technique of demarcating between cortical and cancellous bone on CT scans. However, a further analysis of the interrater agreement using Bland-Altman plots showed no significant biases between the raters toward underestimating or overestimating thickness or density values, with differences randomly scattered above and below the zero line on the y-axis, and most values within the limits of agreement. On average, we found that palatal alveolar bone was thicker and denser compared with buccal bone; this difference in properties was most significant in the anterior region but not statistically significant in the posterior
region. Edentulous maxillary sites also appeared to follow the same pattern, with the cortical bone usually thicker at the lingual aspect than at the facial aspect, as shown by Flanagan17 in 2008. Deguchi et al18 in 2006 reported significantly thicker cortical bone on the lingual side in comparison with the buccal side at the distal aspect of the second molar region (P \0.01) but no significant difference at the mesial counterpart. In both the palatal and buccal regions, alveolar bone density and thickness progressively increased the farther away from the alveolar crest we measured, with the highest values seen at a distance of 8 mm; this is consistent with the expectation that cortical bone should get thicker and denser as we move from the alveolar bone crest toward the basal bone. Measurement was stopped at 8 mm because of the clinical limitation on the buccal aspect of inserting OMIs in the attached gingiva for improved success.19-21 In their study on skulls, Baumgaertel and Hans22 and Kim et al23 found similar results, with the greatest thickness on the buccal side at the maximum measured distance from the alveolar crest (or cementoenamel junction). However, they both showed that a thinning of bone thickness could be seen at 4 mm in the buccal segments, a finding not seen in our study. Kim et al23 also examined palatal cortical bone and found a general thickening toward the basal bone. Thus, to ensure maximum primary stability, inserting OMIs as high as possible in the attached gingiva would ensure the greatest mechanical retention in alveolar cortical bone. After we established that the thickest and densest bone was usually 8 mm apical to the alveolar crest, our secondary analysis of this region showed the following results. On average, the interradicular regions with the greatest buccal thickness and density were seen in the posterior maxilla, with the region between the first and second molars having the greatest value. An exception to this rule was found between the second and third molars. Other authors have reported similar findings, with thin buccal cortical bone commonly found distally to the second molar.18,22 On the palatal side, maximum thickness and density were found more anteriorly in 2 sites: between the premolars and between the first premolar and the canine. Table V gives recommendations for placement of OMIs. Sites are arbitrarily divided into high, moderate, and low prognoses for OMI success based on the average thickness and density of alveolar cortical bone at 8 mm from the alveolar crest. Table V can be used chair side by clinicians as an easy-to-use guide when making decisions during OMI insertion. It is unclear at this point which property (thickness or density) is more relevant to OMI survivability, and this conflict is highlighted in Table V in regions where average
American Journal of Orthodontics and Dentofacial Orthopedics
January 2017 Vol 151 Issue 1
Ohiomoba et al
88
Fig 4. Average thickness (mm) of alveolar cortical bone at 8-mm alveolar bone height (61 SD). I, Incisor; L, lateral incisor; C, canine; PM, premolar; M, molar; 1, first; 2, second; 3, third.
Fig 5. Average density (Hounsfield units) of alveolar cortical bone at 8-mm alveolar bone height (61 SD). I, Incisor; L, lateral incisor; C, canine; PM, premolar; M, molar; 1, first; 2, second; 3, third.
density and thickness values are not directly correlated; it is left to the clinician's discretion to reconcile this difference when making treatment decisions. As more clinical studies that report OMI success rates at various regions in the maxilla become available, our understanding of this issue should improve. Bone density exhibits an opposite pattern in edentulous areas compared with dentate ones. Misch24 reported that basal bone is denser in the edentulous anterior maxilla compared with the edentulous posterior maxilla; this contrasts to the pattern seen in alveolar bone as demonstrated in our results. The borders of the palatal and buccal cortical bone could not be adequately delineated in some regions, and thickness values from these areas were excluded from the final analysis. These regions
January 2017 Vol 151 Issue 1
of possible dehiscence or fenestration could be false negatives, with thin bone present when viewed clinically, a phenomenon that could be described as radiologic absence of bone.25 The highest incidence of this was seen between the incisors compared with other regions; the cortical bone between the premolars on the buccal side also showed a high incidence. Patient factors played a significant role in bone density and thickness values; in our study, we explored the effect of the following patient variables: sex, ethnicity, BMI, and age. Overall, female subjects had significantly denser bones compared with male subjects. This finding is not surprising, since most of our study subjects were in their teenage years (mean age, 17.88 years). In their 2009 study on the differences in peak bone density
American Journal of Orthodontics and Dentofacial Orthopedics
Ohiomoba et al
89
Table V. Recommendations for selection of OMI sites
based on the thickness and density of the bone Buccal Probability of success High Moderate
Low
Thickness 1M-2M 2PM-1M C-1PM 1PM-2PM L-C 2M-3M I-I I-L
Density
1PM-2PM L-C C-1PM I-L 2M-3M I-I
Palatal Thickness C-1PM 1PM-2PM L-C I-L 2PM-1M 1M-2M I-I 2M-3M
Density 1PM-2PM C-1PM 2PM-1M L-C 1M-2M I-L 2M-3M I-I
Example: C-1PM means between the canine and the first premolar. I, Central incisor; L, lateral incisor; C, canine; PM, premolar; M, molar; 1, first; 2, second; 3, third.
between male and female students aged 19 to 25 years, Avdagic et al26 suggested that boys achieve peak bone density later than girls. Other studies have also reported that female patients tend to have denser palatal cortical bone compared with male patients.27,28 The opposite relationship may be seen in an older nongrowing population of patients because of the expectation that men will eventually develop a higher peak bone density26 and also experience a slower rate of bone loss over time compared with women.29 It was difficult to come to any definite conclusions regarding the influence of ethnicity on bone properties because the sample sizes were too small: ie, 5 or fewer subjects in some groups; thus a study with larger samples is needed to validate our findings. In comparing the 2 largest groups—white vs black subjects—in our study, ethnicity was not a significant factor for bone thickness; however, black patients may have less dense alveolar bone compared with white patients in this age group. Our findings are, however, pivotal, since a 2012 systemic review by AlSamak et al30 on this subject reports no comparative assessment on the effect of race in any study. BMI is positively associated with bone density and thickness in the maxilla, and this relationship is statistically significant. This is consistent with reports in the medical literature, with a positive correlation between BMI and bone mineral density.31-33 Salamat et al34 in 2013 also concluded that both BMI and weight are associated with bone mineral density of the hip and vertebrae. Equally, Zlataric et al35 reported that heavier people have higher bone mineral density than lighter persons do. Skeletal mass is mostly completed by age 20 and continues to grow slowly until around the third decade.26 Late adolescence has also been proposed as the time period during which most of the bone mass at multiple skeletal locations will be accumulated.36 Theintz
et al37 reported that the bone mass accumulation rates declined by 16 years of age in girls and several years later in boys at the lumbar spine and femoral neck. In our study, increase in age was positively associated with cortical bone thickness and density, but this relationship plateaued by the age of 16 with regard to thickness. Similar results were reported by Fayed et al7; subjects aged 19 to 27 years were reported to have significantly greater buccal and palatal cortical thicknesses at specific sites compared with a younger group of patients aged 13 to 18 years. A limitation of this study was the small sample size in some ethnic groups during analysis; this may significantly affect the validity of the results associated with this variable. In addition, the slice thickness (0.6 mm) of the images used in this study compared with the smaller slice thickness commonly used in CBCT images could negatively affect measurement accuracy. However, spatial resolution—more relevant for the visualization of thin bone—is better with MCT compared with CBCT. Ballrick et al8 showed that the spatial resolution of common voxel sizes used for orthodontic scans (0.3 and 0.4 mm) averaged 0.7 mm. Finally, other patient variables such as skeletal pattern, biting strength, and diet (soft processed foods vs raw foods), which were not controlled for in this study, may be associated with alveolar bone properties. Further studies that continue to explore the influence of patient factors on bone properties and ultimately on OMI survivability are necessary. Further research is also needed to elucidate whether bone density or thickness is of greater importance in predicting the success of OMIs or whether there is clinical significance to statistical differences in bone properties. CONCLUSIONS
Alveolar bone properties vary in the maxilla in patterns that could guide clinicians in selecting sites best suited for OMI placement. A summary of these patterns is itemized below. 1.
2.
3.
4.
American Journal of Orthodontics and Dentofacial Orthopedics
Cortical bone density and thickness significantly increase from the coronal (2 mm) to the apical (8 mm) regions of the alveolar bone. On average, palatal cortical bone is thicker and denser compared with buccal; this difference is greatest in the anterior part of the maxilla. Sex is significantly associated with alveolar bone density (female . male) but not thickness in the age range in our study (12-41 years). Ethnicity may play a significant role with regard to alveolar bone properties; however, future studies with larger samples per ethnic group are needed to fully determine this relationship.
January 2017 Vol 151 Issue 1
Ohiomoba et al
90
5. 6. 7.
8.
9.
10.
Increased BMI is significantly associated with increased bone thickness and density. In our study population, patient age was directly associated with alveolar bone density. Subjects in the youngest age group (12-16 years) had significantly thinner alveolar cortical bone compared with older age groups. Alveolar buccal cortical bone is thickest and densest between the first molar and second molar interradicular bone, followed by between the second premolar and first molar interradicular bone. Alveolar palatal cortical bone is thickest between the canine and first premolar interradicular bone, followed by between the first premolar and second premolar interradicular bone. It is similarly densest between the canine and first premolar interradicular bone, and between the first premolar and second premolar interradicular bone. Radiologic absence of bone is more commonly seen in the anterior maxilla.
9.
10.
11.
12.
13. 14.
15.
16.
17.
ACKNOWLEDGMENT
We thank Sanjay Prabhu, director of the Advanced Image Analysis Laboratory at Boston Children's Hospital Department of Radiology, for his expert technical assistance.
18.
19. REFERENCES 1. Buschang PH, Carrillo R, Ozenbaugh B, Rossouw PE. 2008 survey of AAO members on miniscrew usage. J Clin Orthod 2008;42: 513-8. 2. Farnsworth D, Rossouw PE, Ceen RF, Buschang PH. Cortical bone thickness at common miniscrew implant placement sites. Am J Orthod Dentofacial Orthop 2011;139:495-503. 3. Motoyoshi M, Uemura M, Ono A, Okazaki K, Shigeeda T, Shimizu N. Factors affecting the long-term stability of orthodontic mini-implants. Am J Orthod Dentofacial Orthop 2010;137: 588.e1-5:discussion, 588-9. 4. Crismani AG, Bertl MH, Celar AG, Bantleon HP, Burstone CJ. Miniscrews in orthodontic treatment: review and analysis of published clinical trials. Am J Orthod Dentofacial Orthop 2010;137:108-13. 5. Kravitz ND, Kusnoto B. Risks and complications of orthodontic miniscrews. Am J Orthod Dentofacial Orthop 2007;131(Suppl 4): S43-51. 6. Iijima M, Takano M, Yasuda Y, Mugurama T, Nakagaki S, Sakakura Y, et al. Effect of the quantity and quality of cortical bone on the failure force of a miniscrew implant. Eur J Orthod 2013;35:583-9. 7. Fayed MM, Pazera P, Katsaros C. Optimal sites for orthodontic mini-implant placement assessed by cone beam computed tomography. Angle Orthod 2010;80:939-51. 8. Ballrick JW, Palomo JM, Ruch E, Amberman BD, Hans MG. Image distortion and spatial resolution of a commercially available cone-
January 2017 Vol 151 Issue 1
20.
21.
22.
23.
24. 25.
26.
beam computed tomography machine. Am J Orthod Dentofacial Orthop 2008;134:573-82. Loubele M, Guerrero ME, Jacobs R, Suetens P, van Steenberghe D. A comparison of jaw dimensional and quality assessments of bone characteristics with cone-beam CT, spiral tomography, and multislice spiral CT. Int J Oral Maxillofac Implants 2007;22:446-54. Li J, Yao W, Xiao Y, Yu Y. Feasibility of improving cone-beam CT number consistency using a scatter correction algorithm. J Appl Clin Med Phys 2013;14:4346. Molteni R. Prospects and challenges of rendering tissue density in Hounsfield units for cone beam computed tomography. Oral Surg Oral Med Oral Pathol Oral Radiol 2013;116:105-19. Arctander K, Kolbenstvedt A, Aalokken TM, Abyholm F, Froslie KF. Computed tomography of alveolar bone grafts 20 years after repair of unilateral cleft lip and palate. Scand J Plast Reconstr Surg Hand Surg 2005;39:11-4. Viera AJ, Garrett JM. Understanding interobserver agreement: the kappa statistic. Fam Med 2005;37:360-3. Cassetta M, Stefanelli LV, Pacifici A, Pacifici L, Barbato E. How accurate is CBCT in measuring bone density? A comparative CBCTCT in vitro study. Clin Implant Dent Relat Res 2014;16:471-8. Mah P, Reeves TE, McDavid WD. Deriving Hounsfield units using grey levels in cone beam computed tomography. Dentomaxillofac Radiol 2010;39:323-35. Silva IM, Freitas DQ, Ambrosano GM, Boscolo FN, Almeida SM. Bone density: comparative evaluation of Hounsfield units in multislice and cone-beam computed tomography. Braz Oral Res 2012;26:550-6. Flanagan D. A comparison of facial and lingual cortical thicknesses in edentulous maxillary and mandibular sites measured on computerized tomograms. J Oral Implantol 2008;34:256-8. Deguchi T, Nasu M, Murakami K, Yabuuchi T, Kamioka H, TakanoYamamoto T. Quantitative evaluation of cortical bone thickness with computed tomographic scanning for orthodontic implants. Am J Orthod Dentofacial Orthop 2006;129:721.e7-12. Palumbo P. The Anatomy and physiology of the healthy periodontium. In Panagakos FS, editor. Gingival Diseases: Their Aetiology, Prevention and Treatment. Rijeka, Croatia: InTech; 2011, p. 1-22. Available at: http://www.intechopen.com/books/gingival-disea ses-their-aetiology-prevention-and-treatment/the-anatomy-and -physiology-of-the-healthy-periodontium. Accessed November 14, 2016. Kuroda S, Sugawara Y, Deguchi T, Kyung HM, TakanoYamamoto T. Clinical use of miniscrew implants as orthodontic anchorage: success rates and postoperative discomfort. Am J Orthod Dentofacial Orthop 2007;131:9-15. Tenenbaum H, Tenenbaum M. A clinical study of the width of the attached gingiva in the deciduous, transitional and permanent dentitions. J Clin Periodontol 1986;13:270-5. Baumgaertel S, Hans MG. Buccal cortical bone thickness for miniimplant placement. Am J Orthod Dentofacial Orthop 2009;136: 230-5. Kim HJ, Yun HS, Park HD, Kim DH, Park YC. Soft-tissue and cortical-bone thickness at orthodontic implant sites. Am J Orthod Dentofacial Orthop 2006;130:177-82. Misch CE. Contemporary Implant Dentistry. 3rd ed. St. Louis, MO: Mosby Inc; 2008. Mondina M, Bonnard D, Barreau X, Darrouzet V, Franco-Vidal V. Anatomo-radiological study of the superior semicircular canal dehiscence of 37 cadaver temporal bones. Surg Radiol Anat 2013;35:55-9. Avdagic SC, Baric IC, Keser I, Cecic I, Satalic Z, Babic J, et al. Differences in peak bone density between male and female students. Arh Hig Rada Toksikol 2009;60:79-86.
American Journal of Orthodontics and Dentofacial Orthopedics
Ohiomoba et al
27. Moon SH, Park SH, Lim WH, Chun YS. Palatal bone density in adult subjects: implications for mini-implant placement. Angle Orthod 2010;80:137-44. 28. Ozdemir F, Tozlu M, Germec Cakan D. Quantitative evaluation of alveolar cortical bone density in adults with different vertical facial types using cone-beam computed tomography. Korean J Orthod 2014;44:36-43. 29. Warming L, Hassager C, Christiansen C. Changes in bone mineral density with age in men and women: a longitudinal study. Osteoporos Int 2002;13:105-12. 30. AlSamak S, Gkantidis N, Bitsanis E, Christou P. Assessment of potential orthodontic mini-implant insertion sites based on anatomical hard tissue parameters: a systematic review. Int J Oral Maxillofac Implants 2012;27:875-87. 31. Felson DT, Zhang Y, Hannan MT, Anderson JJ. Effects of weight and body mass index on bone mineral density in men and women: the Framingham study. J Bone Miner Res 1993;8:567-73. 32. Nguyen TV, Center JR, Eisman JA. Osteoporosis in elderly men and women: effects of dietary calcium, physical activity, and body mass index. J Bone Miner Res 2000;15:322-31.
91
33. Ravn P, Cizza G, Bjarnason NH, Thompson D, Daley M, Wasnich RD, et al. Low body mass index is an important risk factor for low bone mass and increased bone loss in early postmenopausal women. Early Postmenopausal Intervention Cohort (EPIC) study group. J Bone Miner Res 1999;14:1622-7. 34. Salamat MR, Salamat AH, Abedi I, Janghorbani M. Relationship between weight, body mass index, and bone mineral density in men referred for dual-energy x-ray absorptiometry scan in Isfahan. Iran. J Osteoporos 2013;2013:205963. 35. Zlataric DK, Celebic A, Kobler P. Relationship between body mass index and local quality of mandibular bone structure in elderly individuals. J Gerontol A Biol Sci Med Sci 2002;57:M588-93. 36. Matkovic V, Jelic T, Wardlaw GM, Ilich JZ, Goel PK, Wright JK, et al. Timing of peak bone mass in Caucasian females and its implication for the prevention of osteoporosis. Inference from a crosssectional model. J Clin Invest 1994;93:799-808. 37. Theintz G, Buchs B, Rizzoli R, Slosman D, Clavien H, Sizenenko PC, et al. Longitudinal monitoring of bone mass accumulation in healthy adolescents: evidence for a marked reduction after 16 years of age at the levels of lumbar spine and femoral neck in female subjects. J Clin Endocrinol Metab 1992;75:1060-5.
American Journal of Orthodontics and Dentofacial Orthopedics
January 2017 Vol 151 Issue 1
Quantitative evaluation of maxillary alveolar cortical bone thickness and density using computed tomography imaging Henry Ohiomoba,a Andrew Sonis,b Alfa Yansane,c and Bernard Friedlandd Boston, Mass
Introduction: Primary stability is essential to the success of orthodontic mini-implants (OMIs) and heavily depends on the mechanical retention between OMIs and their supporting bone. Alveolar cortical bone commonly serves as the supporting bone for OMIs during treatment. The purposes of this study were to characterize alveolar cortical bone thickness and density in the maxilla and to explore patient factors that may significantly affect these bone properties. Methods: Sixty medical computed tomography scans of the maxilla were analyzed from a selected sample of patients seen at the Radiology Department of Boston Children's Hospital. Interradicular alveolar bone thickness and density were measured at 2, 4, 6, and 8 mm from the buccal and palatal alveolar bone crests using the Synapse 3D software (version 4.1; FUJIFILM Medical Systems USA, Stamford, Conn). Analyses were conducted with STATA /1C (version 12.0 for Windows; StataCorp, College Station, Tex) using multivariate mixed-effects regression models and paired t tests. Results: Mean age and body mass index of the study sample were 17.88 years and 22.94 kg/m2, respectively. Cortical bone density and thickness significantly increased from the coronal (2 mm) to the apical (8 mm) regions of the alveolar bone (P \0.05). At 8 mm from the alveolar crest, interradicular buccal cortical bone was thickest (1 mm) and densest (1395 Hounsfield units) between the first and second molars. On the palatal side, the thickest bone (1.15 mm) was found between the canine and first premolar; it was similarly densest (1406 Hounsfield units) between the first premolar and canine, and between the first premolar and second premolar interradicular bones. On average, palatal cortical bone was thicker and denser compared with buccal; this difference was statistically significant (P \0.01) in the anterior and middle maxilla, with the anterior maxillary region showing the greatest difference. Female subjects have significantly denser bone compared with male subjects; however, sex is not significantly associated with bone thickness. Body mass index and age are positively associated with bone thickness and density. Radiologic absence of bone was more commonly seen in the anterior maxilla. Conclusions: Alveolar bone properties vary in the maxilla in patterns that could guide clinicians in selecting sites best suited for placement of OMIs. (Am J Orthod Dentofacial Orthop 2017;151:82-91)
A
bsolute anchorage is often a desired goal during orthodontic treatment; this can be difficult to achieve, since most anchorage biomechanics are tooth-supported, leading to unwanted side effects From the Harvard School of Dental Medicine, Boston, Mass. a Orthodontic resident. b Pediatric Dentistry and Orthodontics, and clinical professor; Department of Dentistry, Boston Children's Hospital, Boston, Mass. c Oral Health Policy and Epidemiology. d Department of Oral Medicine Infection and Immunity. All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest, and none were reported. Address correspondence to: Henry Ohiomoba, Holyoke Health Center - 230 Maple St, Holyoke, MA 01040; e-mail, [email protected]. Submitted, July 2015; revised and accepted, May 2016. 0889-5406/$36.00 Ó 2017 by the American Association of Orthodontists. All rights reserved. http://dx.doi.org/10.1016/j.ajodo.2016.05.015
82
such as translation or tipping of anchoring teeth. In recent years, orthodontic mini-implants (OMIs) have provided a solution to the problem of unpredictable anchorage because they are bone-supported. A 2008 survey reported that approximately 80% of orthodontists use OMIs, and 78% of them believe that OMIs improve treatment outcomes.1 Common sites for placement of OMIs are the maxillary and mandibular buccal alveolar bones, infrazygomatic crest, palatal alveolar bone, and paramedian palate.2 A major limitation of OMIs is loosening (mobility) and subsequent failure, which could lead to premature removal and longer treatment times. Failure rates of OMIs under orthodontic loading have been reported to range from 11% to 30%,3 whereas success rates of 76% to 87% have also
Ohiomoba et al
been reported.4 Possible risk factors that affect the stability of OMIs are supporting bone quality and quantity, screw design, force loads, and safe placement techniques.5 Some controversy exists over the significance of some of these risk factors with regard to what is optimal; however, it is generally accepted that increased thickness of cortical bone is favorable for OMI stability. Iijima et al6 in 2012 showed that cortical bone thickness, total bone mineral density, cortical bone mineral density, and bone hardness were significantly related to the mean failure force of OMIs. Studies that look to characterize cortical bone thickness and density in the alveolar bone have the limitation of small sample sizes, in-vitro studies on skulls, or failing to explore patient factors such as age and sex on cortical bone properties.7 In addition, cone-beam computerized tomography (CBCT) has several limitations compared with standard medical computerized tomography (MCT) for bone evaluation; Hounsfield units that measure density are not accurate, and increased scatter radiation and artifacts caused by beam hardening are all associated with CBCT.8-11 In this study, we used MCT to quantitatively characterize cortical bone thickness and density in the maxillary alveolar bone and to investigate the influence of patient factors to determine optimal locations for placement of OMIs during orthodontic treatment. MATERIALS AND METHODS
The study sample consisted of 60 high-resolution MCT scans of the maxillofacial region taken from a pool of patients who had been seen in the Radiology Department of Boston Children's Hospital for reasons unrelated to this study such as chronic sinusitis and presence of a foreign body in the sinuses (30 female, 30 male). Sample inclusion criteria included patients aged 12 years or older with permanent dentition and a diagnosis of “normal” or “unremarkable” in the radiologist's report of the MCT scan. Exclusion criteria included significant radiographic signs of periodontal disease, moderate to severe overlapping of crowns or roots of adjacent teeth, trauma to the maxillofacial region, multiple missing teeth, and a positive history of systemic conditions, pathologies, or therapy that may significantly affect cortical bone properties. Scans were captured using a GE LightSpeed Pro 32 Slice CT unit (GE Healthcare, Little Chalfont, United Kingdom) at 120 KVp, 50 mA, 0.6-mm slice thickness, and mean spatial resolution of 0.63 mm. The images were analyzed using the dental multiplanar reconstruction interphase in the Synapse 3D software (version 4.1; FUJIFILM Medical Systems USA, Stamford, Conn). All reconstructions were
83
Fig 1. CT cross-sectional view perpendicular to the axial plane showing thickness (mm) and average density measurements.
done at 0.6 mm with a high spatial frequency bone algorithm. A reference plane parallel to the maxillary occlusal plane was established to generate a 2-dimensional panoramic image of the maxilla and to aid in generating 2dimensional cross-sectional images. Two-dimensional cross-sectional images of the maxilla were reconstructed at 1-mm intervals (150 cross-sections per scan) and viewed under settings appropriate for bone details (window width, 2000; window level, 400) and sharpened to best delineate cortical boundaries.12 Best cross-sectional slices through each contact area that clearly delineated interradicular bone without root anatomy were selected for measurements, beginning distally to the right second molar and ending distally to the left second molar. Thickness measurements were made perpendicular to the bone surface using the function “ruler,” and average density measurements were made using the corresponding oval density tool, both in the Synapse 3D software. The demarcation between cancellous and cortical bone was drawn manually by visual gray-white discrimination; gray was perceived as cancellous bone, and white was perceived as cortical bone. Palatal and buccal alveolar cortical bone thickness and density were documented in millimeters and Hounsfield units, respectively. These were measured at 2, 4, 6, and 8 mm from the alveolar bone crest along the surface of the cortical bone (Fig 1). Thus, a maximum of 120 alveolar thickness and 120 mean alveolar density measurements were documented per MCT scan. For additional analysis, measurements were further grouped into anterior maxilla (interradicular bone from the mesial aspect of the canine to the mesial aspect of the contralateral canine), middle maxilla (interradicular bone between the distal aspect of the canines to the mesial aspect of the first molars), and posterior maxilla (interradicular bone between the distal aspect of the first molars to the mesial aspect of the third molars). Training support for image analysis was provided
American Journal of Orthodontics and Dentofacial Orthopedics
January 2017 Vol 151 Issue 1
Ohiomoba et al
84
by the Radiology Advanced Image Analysis Laboratory at Boston Children's Hospital. All measurements were performed by a primary rater (H.O.) and entered into an Excel 2010 spreadsheet version 14.0; (Microsoft, Redmond, Wash). Two hundred randomly selected measurements were repeated, 1 month later, by both the primary rater and a second calibrated rater (A.S.) to assess for intrarater and interrater reliabilities. Both raters were required to recreate new 2-dimensional crosssectional images using the technique described earlier during reliability measurements. Protocol approval for the study was received from the Committee on Clinical Investigation at Boston Children's Hospital, and the Committee on Human Studies at Harvard Medical School and Harvard School of Dental Medicine. Statistical analysis
Descriptive analyses of patient demographics were summarized using means and standard deviations for continuous variables and frequencies with corresponding proportions for the categorical variables. The intraclass correlation coefficient was used to determine both interrater and intrarater reliabilities. Bland-Altman plots were also used to assess agreement between the raters’ measurements. The 2-sample paired t test was used to determine differences in thickness and density between palatal vs buccal cortical bone; because of the expected correlation of intrapatient measurements, the patient was used as the unit of analysis by averaging the total buccal and palatal measurement values for each patient. Multivariate mixedeffects regression models (patient identity was included as a random effect to account for within-patient correlation) were used to explore the effects of alveolar bone height and multiple patient factors such as sex, age, body mass index (BMI), and ethnicity on alveolar bone thickness and density. To test for multiple comparisons on the effect of alveolar bone height on both thickness and density, we used the Bonferroni method for multiple comparisons to adjust for type 1 error inflation. For the regression models, the BMI and age variables were divided into 3 ordinal groups, less than 21 Kg/m2, 21 to 25 Kg/m2, and greater than 25 Kg/m2; and 12 to 16 years, more than 16 but less than 20 years, and 20 years or more, respectively. All tests were conducted at the significance level of a 5 0.05 as is standard, and all statistical analysis were performed using STATA /1C software (version 12.0 for Windows; StataCorp, College Station, Tex). RESULTS
The mean age of our study sample was 17.88 years (SD, 4.71 years), and the mean BMI was 22.94 kg/m2 (SD, 4.13 kg/m2). Table I shows the frequency and
January 2017 Vol 151 Issue 1
Table I. Demographic distribution Age group (y) 12-16 \16-\20 $20 BMI group \21 21-25 .25 NA Ethnicity White Black Hispanic Asian American Indian NA
Male (%)
Female (%)
Total (%)
8 (13.3) 17 (28.3) 5 (8.3)
12 (20) 13 (21.7) 5 (8.3)
20 (33.3) 30 (50) 10 (16.7)
14 (23.3) 8 (13.3) 8 (13.3) -
5 (8.3) 13 (21.67) 8 (13.3) 4 (6.7)
19 (31.7) 21 (35) 16 (26.7) 4 (6.7)
11 (18.3) 9 (15) 3 (5) 3 (5) 4 (6.7)
18 (30) 10 (16.7) 1 (1.7) 1 (1.7)
29 (48.3) 19 (31.7) 3 (5) 3 (5) 1 (1.7) 5 (8.3)
NA, Not answered.
percentage distributions of the study population by age, sex, BMI, and ethnicity. Approximately half (48%) of the study sample self-identified as white, 32% selfidentified as black, and the rest identified as Hispanic (5%), Asian (5%), or American Indian (2%), or chose not to self-identify (8%). Intrarater reliabilities (absolute agreement) within the primary rater were r 5 0.98 for density and r 5 0.90 for thickness; this is near perfect agreement. Interrater reliability (absolute agreement) comparing the primary rater with the secondary rater was substantial for density (r 5 0.78) and moderate for thickness (r 5 0.60) measurements.13 BlandAltman plots (Figs 2 and 3) exploring the agreement between the primary rater vs the secondary rater for the thickness and density measurements showed no evidence of a difference in variability between the raters (P value for Pitman's test of difference in variability in both plots were .0.05); the limits of agreement were 0.512 to 0.599 (mean difference, 0.043) for thickness measurements and 299.250 to 310.336 (mean difference, 5.543) for density measurements. The results of the paired t test showed that, on average, palatal bone was thicker and denser than buccal bone; this difference was statistically significant (\0.01) at the anterior and middle regions of the maxilla but not at the posterior maxilla (Tables II and III). Multivariate mixed-effects models controlling for patient identity, alveolar height, sex, ethnicity, BMI group, and age group were used to explore the effect of multiple patient factors on alveolar bone properties simultaneously (Table IV). On average, alveolar bone thickness and density increased significantly the farther we measured from the alveolar crest, with regions 8 mm
American Journal of Orthodontics and Dentofacial Orthopedics
Ohiomoba et al
85
Fig 2. Bland-Altman plot of differences in alveolar cortical bone thickness (mm) measurements between raters 1 and 2. Limits of agreement (reference range for difference), 0.512 to 0.599; mean difference, 0.043 (95% CI, 0.016, 0.102); Pitman's test of difference in variance, r 5 0.119, P 5 0.268.
Fig 3. Bland-Altman plot of differences in alveolar cortical bone density (Hounsfield units) measurements between raters 1 and 2. Limits of agreement (reference range for difference), 299.250 to 310.336; mean difference, 5.543 (95% CI, 26.195, 37.281); Pitman's test of difference in variance, r 5 0.035, P 5 0.741.
from the alveolar crest having the greatest thickness and density. At 8 mm from the alveolar crest, buccal interradicular bone is thickest and densest between the second molar and first molar, followed by between the first molar and second premolar; it is least thick between the lateral incisor and central incisor, and least dense
between the central incisors (Figs 4 and 5). With regard to the palatal aspect at 8 mm, the bone is thickest between the first premolar and canine, followed by between the premolars; it is similarly densest between the first premolar and canine, and between the premolars. It is least thick between the
American Journal of Orthodontics and Dentofacial Orthopedics
January 2017 Vol 151 Issue 1
Ohiomoba et al
86
Table II. Palatal vs buccal thickness in millimeters
Anterior maxilla Middle maxilla Posterior maxilla
P Mean (SD) 0.90 (0.18) 0.96 (0.18) 0.78 (0.14)
B Mean (SD) 0.71 (0.11) 0.78 (0.16) 0.76 (0.14)
P B Diff (SD) 0.19 (0.14) 0.18 (0.16) 0.02 (0.13)
Degrees of freedom 56 59 59
t test value 10.50 8.50 1.52
P value \0.001 \0.001 0.134
Degrees of freedom 56 59 59
t test value 3.88 3.43 1.69
P value \0.001 0.001 0.096
Paired t test. P, Palatal; B, buccal; Diff, average difference.
Table III. Palatal vs buccal density in Hounsfield units
Anterior maxilla Middle maxilla Posterior maxilla
P Mean (SD) 1288 (117) 1340 (104) 1252 (145)
B Mean (SD) 1227 (121) 1291 (138) 1224 (174)
P B Diff (SD) 61 (118) 49 (110) 28 (130)
Paired t test. P, Palatal; B, buccal; Diff, average difference.
second molar and third molar, and least dense between the central incisors (Figs 4 and 5). Cortical bone could not be delineated at all the 8-mm measurement points. On the buccal side, the interradicular regions with the highest incidence of this occurrence compared with other buccal sites were located between the central incisors (56%), between the first and second premolars (12%), and between the lateral and central incisors (10%). A similar pattern was seen on the palatal side, with the highest incidences compared with other palatal sites seen between the central incisors (89%) and between the lateral and central incisors (6%). The results of the multivariate models (Table IV) show that sex had no significant effect on thickness (P 5 0.77), but female subjects had significantly denser bone compared with male subjects (P \0.0001). With regard to ethnicity, most groups had 5 or fewer persons; in comparing the 2 largest groups—white (n 5 29) vs black (n 5 19) patients—there appeared to be no significance difference in alveolar bone thickness (P 5 0.28), but black subjects may have less dense alveolar bone (P 5\0.0001) compared with white subjects. Increases in BMI were significantly associated with increases in bone thickness and density, with patients in the highest BMI group (.25 kg/m2) having the highest density and thickness. Increase in age was significantly associated with increased alveolar bone density, with the age group greater than 20 years having the highest density compared with the 2 younger groups. However, the effect of age on thickness appeared to plateau by age 16, with subjects 16 years or older having similar cortical bone thickness, a value that was significantly greater (P \0.0001) compared with the youngest age group between 12 and 16 years.
January 2017 Vol 151 Issue 1
DISCUSSION
The reported absolute numbers for alveolar cortical bone thickness differ in the literature. This could be due to the underreporting and nonstandardization of visualization parameters such as window width and window length by authors. For our study, we used a window width of 2000 and a window length of 400; this is comparable with the parameters used at the Boston Children's Hospital Department of Radiology for diagnosing bone pathology. Arctander et al12 also used similar window width (2100) and window length (500) for appropriately viewing bone details. Knowledge of the pattern of variation in alveolar bone properties in the maxilla may, however, prove to be more clinically useful than the absolute values for these properties. Thus, the purpose of this study was to highlight predictable patterns of alveolar bone properties in the maxilla and the influence of patient factors in the context of bone thickness and density. Although a CT scan of every patient before insertion of OMIs may prove beneficial, the downside of this protocol (eg, radiation exposure, expense, and accessibility to a CT scanner) makes this approach impractical. Results from our study provide a guideline and may largely negate the need for CT scans before OMI insertion. Also, ours is one of the few studies to characterize alveolar bone density in the maxilla, information that is not easily obtained from CBCT imaging.14-16 The reliability results showed excellent intrarater agreements but moderate to substantial interrater agreements; this could be expected because of the subjectivity of the manual visual gray-white
American Journal of Orthodontics and Dentofacial Orthopedics
Ohiomoba et al
87
Table IV. Multivariate mixed-effects regression anal-
ysis showing influence of patient factors on alveolar bone properties (thickness and density) Coef Thickness Alveolar height: 2 mm 4 mm 0.11 6 mm 0.21 8 mm 0.26 Sex: male Female 0.00 Ethnicity: white Asian 0.09 American Indian 0.12 Black 0.01 Hispanic 0.05 NA 0.10 BMI group: \21 21-25 0.04 .25 0.09 Age group: 12-16 y .16-\20 y 0.09 .20 y 0.08 Density Alveolar height: 2 mm 4 mm 109.18 6 mm 179.45 8 mm 208.37 Sex: male Female 42.22 Ethnicity: white Asian 94.93 American Indian 32.16 Black 22.38 Hispanic 24.64 NA 34.80 BMI group: \21 21-25 39.78 .25 87.88 Age group: 12-16 y \16-\20 y 124.11 #20 y 174.23
SE
Sig
0.01 \0.01 0.01 \0.01 0.01 \0.01 0.01
0.84
95% CI
0.09, 0.12 0.19, 0.22 0.24, 0.28 0.02, 0.02
0.02 \0.01 0.03 \0.01 0.01 0.38 0.02 0.02 0.01 \0.01
0.12, 0.05 0.07, 0.18 0.03, 0.01 0.01, 0.08 0.13, 0.07
0.01 \0.01 0.01 \0.01
0.02, 0.05 0.07, 0.11
0.01 \0.01 0.01 \0.01
0.07, 0.11 0.06, 0.11
8.55 \0.01 8.52 \0.01 8.56 \0.01
92.42, 125.94 162.74, 196.15 191.58, 225.16
7.23 \0.01
28.04, 56.40
14.25 \0.01 24.11 0.18 7.36 \0.01 15.19 0.11 11.54 \0.01
122.87, 66.98 15.09, 79.41 38.81, 7.95 54.41, 5.14 12.19, 57.41
7.60 \0.01 8.60 \0.01
24.88, 54.68 71.00, 104.75
7.13 \0.01 10.05 \0.01
110.14, 138.09 154.53, 193.93
Patient identity included as a random effect. Coef, Coefficient; Sig, significance; NA, not answered.
discrimination technique of demarcating between cortical and cancellous bone on CT scans. However, a further analysis of the interrater agreement using Bland-Altman plots showed no significant biases between the raters toward underestimating or overestimating thickness or density values, with differences randomly scattered above and below the zero line on the y-axis, and most values within the limits of agreement. On average, we found that palatal alveolar bone was thicker and denser compared with buccal bone; this difference in properties was most significant in the anterior region but not statistically significant in the posterior
region. Edentulous maxillary sites also appeared to follow the same pattern, with the cortical bone usually thicker at the lingual aspect than at the facial aspect, as shown by Flanagan17 in 2008. Deguchi et al18 in 2006 reported significantly thicker cortical bone on the lingual side in comparison with the buccal side at the distal aspect of the second molar region (P \0.01) but no significant difference at the mesial counterpart. In both the palatal and buccal regions, alveolar bone density and thickness progressively increased the farther away from the alveolar crest we measured, with the highest values seen at a distance of 8 mm; this is consistent with the expectation that cortical bone should get thicker and denser as we move from the alveolar bone crest toward the basal bone. Measurement was stopped at 8 mm because of the clinical limitation on the buccal aspect of inserting OMIs in the attached gingiva for improved success.19-21 In their study on skulls, Baumgaertel and Hans22 and Kim et al23 found similar results, with the greatest thickness on the buccal side at the maximum measured distance from the alveolar crest (or cementoenamel junction). However, they both showed that a thinning of bone thickness could be seen at 4 mm in the buccal segments, a finding not seen in our study. Kim et al23 also examined palatal cortical bone and found a general thickening toward the basal bone. Thus, to ensure maximum primary stability, inserting OMIs as high as possible in the attached gingiva would ensure the greatest mechanical retention in alveolar cortical bone. After we established that the thickest and densest bone was usually 8 mm apical to the alveolar crest, our secondary analysis of this region showed the following results. On average, the interradicular regions with the greatest buccal thickness and density were seen in the posterior maxilla, with the region between the first and second molars having the greatest value. An exception to this rule was found between the second and third molars. Other authors have reported similar findings, with thin buccal cortical bone commonly found distally to the second molar.18,22 On the palatal side, maximum thickness and density were found more anteriorly in 2 sites: between the premolars and between the first premolar and the canine. Table V gives recommendations for placement of OMIs. Sites are arbitrarily divided into high, moderate, and low prognoses for OMI success based on the average thickness and density of alveolar cortical bone at 8 mm from the alveolar crest. Table V can be used chair side by clinicians as an easy-to-use guide when making decisions during OMI insertion. It is unclear at this point which property (thickness or density) is more relevant to OMI survivability, and this conflict is highlighted in Table V in regions where average
American Journal of Orthodontics and Dentofacial Orthopedics
January 2017 Vol 151 Issue 1
Ohiomoba et al
88
Fig 4. Average thickness (mm) of alveolar cortical bone at 8-mm alveolar bone height (61 SD). I, Incisor; L, lateral incisor; C, canine; PM, premolar; M, molar; 1, first; 2, second; 3, third.
Fig 5. Average density (Hounsfield units) of alveolar cortical bone at 8-mm alveolar bone height (61 SD). I, Incisor; L, lateral incisor; C, canine; PM, premolar; M, molar; 1, first; 2, second; 3, third.
density and thickness values are not directly correlated; it is left to the clinician's discretion to reconcile this difference when making treatment decisions. As more clinical studies that report OMI success rates at various regions in the maxilla become available, our understanding of this issue should improve. Bone density exhibits an opposite pattern in edentulous areas compared with dentate ones. Misch24 reported that basal bone is denser in the edentulous anterior maxilla compared with the edentulous posterior maxilla; this contrasts to the pattern seen in alveolar bone as demonstrated in our results. The borders of the palatal and buccal cortical bone could not be adequately delineated in some regions, and thickness values from these areas were excluded from the final analysis. These regions
January 2017 Vol 151 Issue 1
of possible dehiscence or fenestration could be false negatives, with thin bone present when viewed clinically, a phenomenon that could be described as radiologic absence of bone.25 The highest incidence of this was seen between the incisors compared with other regions; the cortical bone between the premolars on the buccal side also showed a high incidence. Patient factors played a significant role in bone density and thickness values; in our study, we explored the effect of the following patient variables: sex, ethnicity, BMI, and age. Overall, female subjects had significantly denser bones compared with male subjects. This finding is not surprising, since most of our study subjects were in their teenage years (mean age, 17.88 years). In their 2009 study on the differences in peak bone density
American Journal of Orthodontics and Dentofacial Orthopedics
Ohiomoba et al
89
Table V. Recommendations for selection of OMI sites
based on the thickness and density of the bone Buccal Probability of success High Moderate
Low
Thickness 1M-2M 2PM-1M C-1PM 1PM-2PM L-C 2M-3M I-I I-L
Density
1PM-2PM L-C C-1PM I-L 2M-3M I-I
Palatal Thickness C-1PM 1PM-2PM L-C I-L 2PM-1M 1M-2M I-I 2M-3M
Density 1PM-2PM C-1PM 2PM-1M L-C 1M-2M I-L 2M-3M I-I
Example: C-1PM means between the canine and the first premolar. I, Central incisor; L, lateral incisor; C, canine; PM, premolar; M, molar; 1, first; 2, second; 3, third.
between male and female students aged 19 to 25 years, Avdagic et al26 suggested that boys achieve peak bone density later than girls. Other studies have also reported that female patients tend to have denser palatal cortical bone compared with male patients.27,28 The opposite relationship may be seen in an older nongrowing population of patients because of the expectation that men will eventually develop a higher peak bone density26 and also experience a slower rate of bone loss over time compared with women.29 It was difficult to come to any definite conclusions regarding the influence of ethnicity on bone properties because the sample sizes were too small: ie, 5 or fewer subjects in some groups; thus a study with larger samples is needed to validate our findings. In comparing the 2 largest groups—white vs black subjects—in our study, ethnicity was not a significant factor for bone thickness; however, black patients may have less dense alveolar bone compared with white patients in this age group. Our findings are, however, pivotal, since a 2012 systemic review by AlSamak et al30 on this subject reports no comparative assessment on the effect of race in any study. BMI is positively associated with bone density and thickness in the maxilla, and this relationship is statistically significant. This is consistent with reports in the medical literature, with a positive correlation between BMI and bone mineral density.31-33 Salamat et al34 in 2013 also concluded that both BMI and weight are associated with bone mineral density of the hip and vertebrae. Equally, Zlataric et al35 reported that heavier people have higher bone mineral density than lighter persons do. Skeletal mass is mostly completed by age 20 and continues to grow slowly until around the third decade.26 Late adolescence has also been proposed as the time period during which most of the bone mass at multiple skeletal locations will be accumulated.36 Theintz
et al37 reported that the bone mass accumulation rates declined by 16 years of age in girls and several years later in boys at the lumbar spine and femoral neck. In our study, increase in age was positively associated with cortical bone thickness and density, but this relationship plateaued by the age of 16 with regard to thickness. Similar results were reported by Fayed et al7; subjects aged 19 to 27 years were reported to have significantly greater buccal and palatal cortical thicknesses at specific sites compared with a younger group of patients aged 13 to 18 years. A limitation of this study was the small sample size in some ethnic groups during analysis; this may significantly affect the validity of the results associated with this variable. In addition, the slice thickness (0.6 mm) of the images used in this study compared with the smaller slice thickness commonly used in CBCT images could negatively affect measurement accuracy. However, spatial resolution—more relevant for the visualization of thin bone—is better with MCT compared with CBCT. Ballrick et al8 showed that the spatial resolution of common voxel sizes used for orthodontic scans (0.3 and 0.4 mm) averaged 0.7 mm. Finally, other patient variables such as skeletal pattern, biting strength, and diet (soft processed foods vs raw foods), which were not controlled for in this study, may be associated with alveolar bone properties. Further studies that continue to explore the influence of patient factors on bone properties and ultimately on OMI survivability are necessary. Further research is also needed to elucidate whether bone density or thickness is of greater importance in predicting the success of OMIs or whether there is clinical significance to statistical differences in bone properties. CONCLUSIONS
Alveolar bone properties vary in the maxilla in patterns that could guide clinicians in selecting sites best suited for OMI placement. A summary of these patterns is itemized below. 1.
2.
3.
4.
American Journal of Orthodontics and Dentofacial Orthopedics
Cortical bone density and thickness significantly increase from the coronal (2 mm) to the apical (8 mm) regions of the alveolar bone. On average, palatal cortical bone is thicker and denser compared with buccal; this difference is greatest in the anterior part of the maxilla. Sex is significantly associated with alveolar bone density (female . male) but not thickness in the age range in our study (12-41 years). Ethnicity may play a significant role with regard to alveolar bone properties; however, future studies with larger samples per ethnic group are needed to fully determine this relationship.
January 2017 Vol 151 Issue 1
Ohiomoba et al
90
5. 6. 7.
8.
9.
10.
Increased BMI is significantly associated with increased bone thickness and density. In our study population, patient age was directly associated with alveolar bone density. Subjects in the youngest age group (12-16 years) had significantly thinner alveolar cortical bone compared with older age groups. Alveolar buccal cortical bone is thickest and densest between the first molar and second molar interradicular bone, followed by between the second premolar and first molar interradicular bone. Alveolar palatal cortical bone is thickest between the canine and first premolar interradicular bone, followed by between the first premolar and second premolar interradicular bone. It is similarly densest between the canine and first premolar interradicular bone, and between the first premolar and second premolar interradicular bone. Radiologic absence of bone is more commonly seen in the anterior maxilla.
9.
10.
11.
12.
13. 14.
15.
16.
17.
ACKNOWLEDGMENT
We thank Sanjay Prabhu, director of the Advanced Image Analysis Laboratory at Boston Children's Hospital Department of Radiology, for his expert technical assistance.
18.
19. REFERENCES 1. Buschang PH, Carrillo R, Ozenbaugh B, Rossouw PE. 2008 survey of AAO members on miniscrew usage. J Clin Orthod 2008;42: 513-8. 2. Farnsworth D, Rossouw PE, Ceen RF, Buschang PH. Cortical bone thickness at common miniscrew implant placement sites. Am J Orthod Dentofacial Orthop 2011;139:495-503. 3. Motoyoshi M, Uemura M, Ono A, Okazaki K, Shigeeda T, Shimizu N. Factors affecting the long-term stability of orthodontic mini-implants. Am J Orthod Dentofacial Orthop 2010;137: 588.e1-5:discussion, 588-9. 4. Crismani AG, Bertl MH, Celar AG, Bantleon HP, Burstone CJ. Miniscrews in orthodontic treatment: review and analysis of published clinical trials. Am J Orthod Dentofacial Orthop 2010;137:108-13. 5. Kravitz ND, Kusnoto B. Risks and complications of orthodontic miniscrews. Am J Orthod Dentofacial Orthop 2007;131(Suppl 4): S43-51. 6. Iijima M, Takano M, Yasuda Y, Mugurama T, Nakagaki S, Sakakura Y, et al. Effect of the quantity and quality of cortical bone on the failure force of a miniscrew implant. Eur J Orthod 2013;35:583-9. 7. Fayed MM, Pazera P, Katsaros C. Optimal sites for orthodontic mini-implant placement assessed by cone beam computed tomography. Angle Orthod 2010;80:939-51. 8. Ballrick JW, Palomo JM, Ruch E, Amberman BD, Hans MG. Image distortion and spatial resolution of a commercially available cone-
January 2017 Vol 151 Issue 1
20.
21.
22.
23.
24. 25.
26.
beam computed tomography machine. Am J Orthod Dentofacial Orthop 2008;134:573-82. Loubele M, Guerrero ME, Jacobs R, Suetens P, van Steenberghe D. A comparison of jaw dimensional and quality assessments of bone characteristics with cone-beam CT, spiral tomography, and multislice spiral CT. Int J Oral Maxillofac Implants 2007;22:446-54. Li J, Yao W, Xiao Y, Yu Y. Feasibility of improving cone-beam CT number consistency using a scatter correction algorithm. J Appl Clin Med Phys 2013;14:4346. Molteni R. Prospects and challenges of rendering tissue density in Hounsfield units for cone beam computed tomography. Oral Surg Oral Med Oral Pathol Oral Radiol 2013;116:105-19. Arctander K, Kolbenstvedt A, Aalokken TM, Abyholm F, Froslie KF. Computed tomography of alveolar bone grafts 20 years after repair of unilateral cleft lip and palate. Scand J Plast Reconstr Surg Hand Surg 2005;39:11-4. Viera AJ, Garrett JM. Understanding interobserver agreement: the kappa statistic. Fam Med 2005;37:360-3. Cassetta M, Stefanelli LV, Pacifici A, Pacifici L, Barbato E. How accurate is CBCT in measuring bone density? A comparative CBCTCT in vitro study. Clin Implant Dent Relat Res 2014;16:471-8. Mah P, Reeves TE, McDavid WD. Deriving Hounsfield units using grey levels in cone beam computed tomography. Dentomaxillofac Radiol 2010;39:323-35. Silva IM, Freitas DQ, Ambrosano GM, Boscolo FN, Almeida SM. Bone density: comparative evaluation of Hounsfield units in multislice and cone-beam computed tomography. Braz Oral Res 2012;26:550-6. Flanagan D. A comparison of facial and lingual cortical thicknesses in edentulous maxillary and mandibular sites measured on computerized tomograms. J Oral Implantol 2008;34:256-8. Deguchi T, Nasu M, Murakami K, Yabuuchi T, Kamioka H, TakanoYamamoto T. Quantitative evaluation of cortical bone thickness with computed tomographic scanning for orthodontic implants. Am J Orthod Dentofacial Orthop 2006;129:721.e7-12. Palumbo P. The Anatomy and physiology of the healthy periodontium. In Panagakos FS, editor. Gingival Diseases: Their Aetiology, Prevention and Treatment. Rijeka, Croatia: InTech; 2011, p. 1-22. Available at: http://www.intechopen.com/books/gingival-disea ses-their-aetiology-prevention-and-treatment/the-anatomy-and -physiology-of-the-healthy-periodontium. Accessed November 14, 2016. Kuroda S, Sugawara Y, Deguchi T, Kyung HM, TakanoYamamoto T. Clinical use of miniscrew implants as orthodontic anchorage: success rates and postoperative discomfort. Am J Orthod Dentofacial Orthop 2007;131:9-15. Tenenbaum H, Tenenbaum M. A clinical study of the width of the attached gingiva in the deciduous, transitional and permanent dentitions. J Clin Periodontol 1986;13:270-5. Baumgaertel S, Hans MG. Buccal cortical bone thickness for miniimplant placement. Am J Orthod Dentofacial Orthop 2009;136: 230-5. Kim HJ, Yun HS, Park HD, Kim DH, Park YC. Soft-tissue and cortical-bone thickness at orthodontic implant sites. Am J Orthod Dentofacial Orthop 2006;130:177-82. Misch CE. Contemporary Implant Dentistry. 3rd ed. St. Louis, MO: Mosby Inc; 2008. Mondina M, Bonnard D, Barreau X, Darrouzet V, Franco-Vidal V. Anatomo-radiological study of the superior semicircular canal dehiscence of 37 cadaver temporal bones. Surg Radiol Anat 2013;35:55-9. Avdagic SC, Baric IC, Keser I, Cecic I, Satalic Z, Babic J, et al. Differences in peak bone density between male and female students. Arh Hig Rada Toksikol 2009;60:79-86.
American Journal of Orthodontics and Dentofacial Orthopedics
Ohiomoba et al
27. Moon SH, Park SH, Lim WH, Chun YS. Palatal bone density in adult subjects: implications for mini-implant placement. Angle Orthod 2010;80:137-44. 28. Ozdemir F, Tozlu M, Germec Cakan D. Quantitative evaluation of alveolar cortical bone density in adults with different vertical facial types using cone-beam computed tomography. Korean J Orthod 2014;44:36-43. 29. Warming L, Hassager C, Christiansen C. Changes in bone mineral density with age in men and women: a longitudinal study. Osteoporos Int 2002;13:105-12. 30. AlSamak S, Gkantidis N, Bitsanis E, Christou P. Assessment of potential orthodontic mini-implant insertion sites based on anatomical hard tissue parameters: a systematic review. Int J Oral Maxillofac Implants 2012;27:875-87. 31. Felson DT, Zhang Y, Hannan MT, Anderson JJ. Effects of weight and body mass index on bone mineral density in men and women: the Framingham study. J Bone Miner Res 1993;8:567-73. 32. Nguyen TV, Center JR, Eisman JA. Osteoporosis in elderly men and women: effects of dietary calcium, physical activity, and body mass index. J Bone Miner Res 2000;15:322-31.
91
33. Ravn P, Cizza G, Bjarnason NH, Thompson D, Daley M, Wasnich RD, et al. Low body mass index is an important risk factor for low bone mass and increased bone loss in early postmenopausal women. Early Postmenopausal Intervention Cohort (EPIC) study group. J Bone Miner Res 1999;14:1622-7. 34. Salamat MR, Salamat AH, Abedi I, Janghorbani M. Relationship between weight, body mass index, and bone mineral density in men referred for dual-energy x-ray absorptiometry scan in Isfahan. Iran. J Osteoporos 2013;2013:205963. 35. Zlataric DK, Celebic A, Kobler P. Relationship between body mass index and local quality of mandibular bone structure in elderly individuals. J Gerontol A Biol Sci Med Sci 2002;57:M588-93. 36. Matkovic V, Jelic T, Wardlaw GM, Ilich JZ, Goel PK, Wright JK, et al. Timing of peak bone mass in Caucasian females and its implication for the prevention of osteoporosis. Inference from a crosssectional model. J Clin Invest 1994;93:799-808. 37. Theintz G, Buchs B, Rizzoli R, Slosman D, Clavien H, Sizenenko PC, et al. Longitudinal monitoring of bone mass accumulation in healthy adolescents: evidence for a marked reduction after 16 years of age at the levels of lumbar spine and femoral neck in female subjects. J Clin Endocrinol Metab 1992;75:1060-5.
American Journal of Orthodontics and Dentofacial Orthopedics
January 2017 Vol 151 Issue 1