Bone turnover is a continuous process, whereby existing bone tissue is renewed by groups of bone multicellular units (Klein-Nulend et al.,2005). The rate of bone turnover influences the degree of mineralization of bone (DMB) (Roschger et al.,2008). In case of high-turnover the time for deposition of mineral is on average reduced, resulting in a lower DMB (Grynpas,1993; Boivin and Meunier,2003). Therefore, the amount of DMB can be considered a valid indicator of the rate of bone turnover in a certain location (Meunier and Boivin,1997; Renders et al.,2006). Bone turnover has been shown to be strongly dependent on mechanical stresses and strains. Indeed, changes in local strain frequencies, magnitudes, and types have been related to regional differences in rates of bone turnover (O'Connor et al.,1982; Turner,1998; Robling et al.,2002; Skedros et al.,1994).
The DMB is influenced by strains related to muscular function (Berg et al.,2007). Biological markers for bone formation and bone resorption react strongly on new exercise programs (Fujimura et al.,1997). As a matter of fact, intraindividual differences in bone mineral content have been related to different muscular loading conditions (Colletti et al.,1989; Bass et al.,2005; Thomopulos et al.,2007), whereas asymmetric muscular activities have been related to intraindividual variations in bone mineralization (Calbet et al.,1998).
The effect of loading and strain frequencies on the muscle-bone relationship has been discussed previously (Frost,2004; Fricke and Schoenau,2007). Nevertheless, it is not clearly known to which extent muscle forces might influence the local DMB, that is, at their bony attachment sites. Recent reports have shown that in the cortical bone of the human mandibular condyle, the DMB is heterogeneous, and its variations were suggested to be related to differences in bone turnover (Renders et al.,2006; Cioffi et al,2007). At the anterior surface of the human mandibular condyle, the lateral pterygoid muscle inserts into the pterygoid fovea (El Haddioui et al.,2005). In craniofacial orthopaedics, the loading of the lateral pterygoid has been object of interest because of its possible involvement in mandibular growth and development (Whetten and Johnston,1985; Yamin-Locouture et al.,1997; Hiyama et al.,2000; Voudouris et al.,2003).
To assess whether the lateral pterygoid muscle might have an influence on the mineralization degree in the mandibular condyle, the DMB at the attachment site of the lateral pterygoid muscles was measured by means of a micro CT scanner, and compared with a control region where no muscular attachment was present. It was hypothesized that DMB at the attachment sites of lateral pterygoid muscles was lower than at the control regions, because of the larger number of loadings and subsequently higher remodeling rates. Furthermore, as the human lateral pterygoid muscle is heterogeneous in its internal architecture (Hems and Tillmann,2000), variations in DMB within the attachment sites were expected.
Materials and Methods
Ten left mandibular condyles were obtained from embalmed human male cadavers (mean age ± SD: 69.8 ± 14.4 years, range, 43–92 years). The subjects had 10 ± 4 teeth in the upper jaw and 11 ± 3 in the lower jaw. The specimens were used for a previous study (Cioffi et al.,2007).
The condyles were separated from the mandible at the transition from the neck to the ramus (Fig. 1a) with a hand saw; condylar bone marrow was left in situ. The specimens were fixated in 60% formalin.
The bony attachments of the lateral pterygoid muscle were identified. The bone was preserved and the muscle tissue of the upper and lower heads and the temporomandibular joint (TMJ) capsule was left in situ. Using cyanoacrylate (Histoacryl blue, Braun Melsungen AG Melsungen, Germany), slices of radiopaque markers (hand rolled guttapercha points for dental use, size #30, Demedis, Dusseldorf, Germany) were glued to the external surface of the muscle close to the bone surface at the medial, lateral, and inferior boundaries of the muscle attachment region.
The use of the condyles conforms to a written protocol that was reviewed and approved by the Department of Anatomy and Embryology of the Academic Medical Center of the University of Amsterdam.
The 3D distribution of the DMB in the condyles was measured with a high-resolution micro CT system (μCT 40, Scanco Medical AG, Brüttisellen, Switzerland). Each condyle was positioned to be scanned in frontal slices. The scan resolution was 30 μm and the beam energy 55 kV. To minimize noise, an integration time of 1,250 ms was used. To minimize beam-hardening artefacts, the system was equipped with a specially developed algorithm (Scanco Medical AG, Brüttisellen, Switzerland) and an aluminum filter (0.5 mm) was used to remove the low-energy photons. Noise level and the effect of beam hardening were maximally 6% and 7%, respectively (Mulder et al.,2004).
The DMB of each volume element (voxel) was computed from the attenuation coefficient using a linear relation, which was calibrated with a phantom containing hydroxyapatite (HA) densities of 0, 50, 200, 800, and 1200 mg/cm3. Bone tissue was distinguished from bone marrow by applying a visually determined threshold of 500 mg HA/cm3, which was calculated by averaging the thresholds determined in several slices of four condyles. From the bone tissue the outer two surface layers were excluded from all DMB calculations to exclude any partial volume effect.
Volumes of Interest
In each condyle, two regions were defined, namely the attachment site of the lateral pterygoid muscle, located at the anterior surface, and a region at the posterior surface without muscle attachments, serving as control.
In the anterior surface of the condyle, the muscle attachment was delimited inferiorly, medially, and laterally by the radiopaque markers. The upper limit of the muscle attachment to the bone was considered as the boundary to the articular surface. To analyze subregional differences, the attachment area was further divided into eight subregions as shown in Figure 1b. In each subregion a volume of interest (VOI) containing only cortical bone was selected. For the definition of the control region, each condyle was divided into four equal supero-inferior regions. In the third one from the top, the control region was selected in the posterior surface (region P, Fig. 1c), containing only cortical bone. None of the specimens showed a muscular or capsular attachment in this area. Means and standard deviations of DMB were calculated for each VOI. The same parameters were also determined for the anterior region by averaging the subregional results. Interindividual means and standard deviations were then calculated over all 10 condyles.
One-tailed paired Student's T-test was used to compare the DMB of the attachment sites with the control regions. A general linear model (repeated measures) was used to identify medio-lateral and supero-inferior differences in DMB within the attachment site of the muscle. Pearson coefficient was calculated to assess correlations between DMBs at different regions. Statistical analysis was performed using SPSS Software (version 12.0.1 Inc., Chicago, IL).
The 3D distributions of the DMB at the attachment sites of the lateral pterygoid muscles are depicted in Figure 2. Considering that the DMB ranges, from 600 mg HA/cm3 (red) to 1300 mg HA/cm3 (purple), large differences can be observed between individuals and subregions.
The mean DMBs at the anterior region and the posterior control region are shown in Figure 3a. At the attachment site, the DMB was significantly lower than at the posterior control region (respectively, 1036.5 ± 70.3 and 1079.3 ± 62.3 mg HA/cm3, p = 0.003). Within the attachment site, the DMB increased in medio-lateral and in supero-inferior directions. The mineralization in the lateral subregions C–F (1052.2 ± 74.0 mg HA/cm3) was significantly higher (p = 0.016) than in the medial subregions A–D (1004.0 ± 66.8 mg HA/cm3). The subregions G–H showed a higher mineralization (1062.0 ± 87.6 mg HA/cm3; p = 0.049) as compared with the subregions D–E–F (1027.5 ± 89.6 mg HA/cm3). No significant differences were found between the subregions A–B–C (1028.6 ± 58.4 mg HA/cm3) and the lower subregions. Interregional differences explained 7.4% of the total variation, when calculated as the variance. Interindividual differences explained 63.3% of the total variation.
A significant correlation was found between the DMB of the anterior regions and the DMB of the posterior control regions (r = 0.83, p = 0.003)
To our knowledge, this is the first study that aimed to measure the DMB at the attachment site of the lateral pterygoid muscle and to compare it with a location where no muscle is attached. In this study, condyles belonging to the same individuals used for two preceding reports (Renders et al.,2006; Cioffi et al.,2007) were examined. In the first study in which the mineralization of the cortical bone was measured in the anterior and in the posterior surfaces of the mandibular condyle, a lower DMB in the anterior than in the posterior cortex was found, suggesting a higher turnover rate in the anterior than in the posterior surface. In the second, the relationship of strain distribution with the variation in DMB in the cortex of the condyles by finite element models was examined. In this study, the region of interest in the anterior surface was selected below the pterygoid fovea, and the authors found no statistically significant difference in DMB between anterior and posterior cortices. Since the condyles used for both the studies originated from the same individuals, hormonal or metabolic factors cannot be responsible for this inconsistency. In this study the DMB measured at the attachment site of the lateral pterygoid muscle was found to be lower than in the control posterior region were no muscle is attached. The results of this report along with the interpretation of the outcomes of the studies cited above, suggests that the lateral pterygoid muscle might affect the DMB at this site. In the study by Renders et al, the DMBs in both the anterior and posterior regions were slightly lower as compared with the current study. This might be in part related to the more cranial selection of their anterior and posterior regions of interest, which might have been more affected by TMJ stresses. This, in turn, might have resulted in a lower mean DMB.
A possible explanation for the lower DMB found at the attachment site is given in a previous study (Skedros et al.,2001). The authors reported a difference in bone mineralization between cortices subjected to tension and cortices subjected to compression, the mineral density being higher in the latter. The lower DMB, we found at the attachment site, might be related to the tension exerted by the muscle on the bone surface. Also, the exclusion of the two outer layers from the selection of VOIs reduces the chance of including low mineralized superficial tissues of the muscle enthesis considerably (Hems and Tillmann,2000).
Although statistically significant, the difference found in DMB between the two regions was very low. The presence of fibrocartilage, which has higher Young's moduli than cortical bone, within the attachment site might have acted as stress absorber (Evans et al.,1990) eventually reducing the amount of strains and the rate of bone turnover at the bony attachment sites. It might be questioned to which extent the lower mineralization found at the pterygoid fovea can be related to the presence of the lateral pterygoid muscle, or to other local or metabolic conditions. In this study, the interindividual differences explain much more variation in mean DMB than interregional differences. This suggests that differences in metabolic factors between individuals might have had a larger influence on bone turnover than locally occurring strains. Also, the large correlation found in DMB between the two regions suggests that the turnover rate is rather determined by metabolic factors. Nevertheless, it is also possible that the DMBs between the two regions are largely correlated because of a high-correlation between the anterior and posterior strains.
The attachment site of the lateral pterygoid muscle was found to be heterogeneous in its DMB. This resembles the heterogeneous histological structure of the enthesis of the muscle (Hems and Tillmann,2000). In particular, the DMB was found to increase in medio-lateral direction. A possible explanation for medio-lateral differences at the attachment site of the lateral pterygoid could be related to the tendinous insertion at the most lateral side of the attachment area, we found during dissections. Indeed, the higher forces exerted by the tendon at this site might benefit from the support by tissue with higher mineral density (Evans et al.,1991). Also, it could be supposed that the functional heterogeneity of the lateral pterygoid muscle might partially account for the heterogeneity in the DMB of the pterygoid fovea (Phanachet et al.,2002,2003). However, further investigations are needed to address this point.
In conclusion, the results of this study reveal that, the DMB at the attachment site of the lateral pterygoid muscle is lower than at a posterior region, where no muscle is attached. This could suggest that the loadings generated by the lateral pterygoid muscle might intensify bone turnover at this site and could be the cause for the mineral heterogeneity of the pterygoid fovea.
This research was institutionally supported by the Inter-University Research School of Dentistry, through the Academic Center for Dentistry Amsterdam and by the University of Naples “Federico II” through the Dottorato di Ricerca in Scienze Odontostomatologiche. The authors are grateful to Jan Harm Koolstra, Lars Mulder, and Roberto Martina for their research support. The authors thank Peter Brugman for the technical assistance.