| Sign In to gain access to subscriptions and/or personal tools. |
Dynamic Compressive Properties of the Mandibular Condylar Cartilage
Department of Orthodontics and Craniofacial Developmental Biology, Hiroshima University Graduate School of Biomedical Sciences, 1-2-3 Kasumi, Minami-ku, Hiroshima 734-8553, Japan; and Correspondence: * corresponding author, etanaka{at}hiroshima-u.ac.jp
The mandibular condylar cartilage plays an important role as a stress absorber during function. However, relatively little information is available on its dynamic properties under compression. We hypothesized that these properties are region-specific and depend on loading frequency. To characterize the viscoelastic properties of the condylar cartilage, we performed dynamic indentation tests over a wide range of loading frequencies. Ten porcine mandibular condyles were used; the articular surface was divided into 4 regions, anteromedial, anterolateral, posteromedial, and posterolateral. The dynamic complex, storage, and loss moduli increased with frequency, and these values were the highest in the anteromedial region. Loss tangent decreased with frequency from 0.68 to 0.17, but a regional difference was not found. The present results suggest that the dynamic compressive modulus is region-specific and is dependent on the loading frequency, which might have important implications for the transmission of load in the temporomandibular joint.
Key Words: temporomandibular joint • condylar cartilage • dynamic property
Functionally, articular cartilage is a relatively compliant material that covers the much stiffer bone ends of a joint with a relatively thin layer. Without this intervening layer, joint loads would have to be transferred through point contacts where excessive stresses would be developed, resulting in bone destruction. Like the hyaline cartilage found in most joints, the cartilage of the mandibular condyle is composed of cells and extracellular matrix. Collagens and proteoglycans are essential components of the extracellular matrix (Mao et al., 1998; Sharawy et al., 2003). Of the extracellular matrix, collagens are considered to provide mainly tensile strength to the cartilage, whereas resistance to compressive forces is due to the presence of proteoglycans (Stegenga et al., 1991; Kuboki et al., 1997; Mao et al., 1998). When cartilage is loaded by compression, the small permeability of the collagen network impedes the interstitial fluid to flow through the collagen network (Mow et al., 1993). Therefore, the load acting on the cartilage is initially carried by pressurization of the incompressible fluid, without much deformation of the collagen network (Soltz and Ateshian, 1998). These features contribute to the viscoelastic properties of cartilage. Proteoglycans have a hydrophilic character and are associated with the production of hydrostatic pressure in the interstitial fluid. The resistance to compression is mainly dependent on the density of proteoglycans, especially of the large chondroitin sulfated proteoglycans (Tanaka et al., 2003). Since the distribution of the proteoglycans is different in various regions of the cartilage, regional differences in its compressive stiffness can be expected (Del Santo et al., 2000; Shibata et al., 2001; Hu et al., 2001). To measure the resistance to compression, indentation is typically applied to a cartilage sample up to the yield point. This kind of characterization of material properties is limited to static loading conditions that are in contrast to the dynamic nature of biological stresses to which skeletal tissues are subjected (Mow et al., 1989; Athanasiou et al., 1995; Hu et al., 2001; Tanaka and van Eijden, 2003). The mechanical properties depend on the rate of applied strain. In general, stiffness of the articular cartilage increases with the loading rate (Silyn-Roberts and Broom, 1990; Oloyede et al., 1992). Thus far, however, relatively little information is available on the dynamic properties of the condylar cartilage under compression (Hu et al., 2001). To characterize these properties, we investigated them in the mandibular condylar cartilage from pigs over a wide range of loading frequencies.
Description of the Sample Ten mandibular condyles from 10 pigs (ages 6–9 mos; gender not specified) were obtained at a slaughterhouse (Japan Agriculture, Hiroshima, Japan). The protocol of the experiment was approved by the Animal Care and Use Committee at Hiroshima University. The condyles were carefully dissected soon after the animals’ death. The articular disc was maintained as a cover of the articular cartilage surface until the indentation test was conducted. Immediately post mortem, the condyle-disc complexes were placed in 0.1 M phosphate buffer (pH 7.3) at 4°C; indentation tests were conducted within 4 hrs after resection.
The articular surface of the condyle was divided into 4 quadrants: anteromedial, anterolateral, posteromedial, and posterolateral (Fig. 1A
Testing Apparatus and Conditions The dynamic viscoelastic properties of the specimens were determined by means of an automatic dynamic viscoelastometer (Rheovibron DDV-25FP, Orientec Corp., Tokyo, Japan) (Fig. 1B  ). This device consists of a measurement operation block, a main unit, a power unit, a data-processing device, and testing jigs. The measurement operation block consists of a magnetic exciter, a load control motor, a displacement-detecting sensor and amplifier, a load-detecting load cell, and 2 metal plates for indentation. The bony surface of the specimen was dried and rigidly glued onto a metal plate with a fast-drying cyanoacrylate. The other metal plate was equipped with a cylindrical indenter with a diameter of 1.2 mm (Fig. 1B). The articular cartilage was compressed by cyclic displacements of the top of this indenter. During these procedures, each sample was constantly irrigated with phosphate-buffered saline (pH 7.3) at 37°C, which was monitored by means of a thermocouple.
During the test, a dynamic compression was applied to the specimen by a sinusoidal strain of
Dynamic Viscoelastic Parameters
For calculating the dynamic complex modulus |E*|, we assumed that |E*|
where v is the Poisson’s ratio, S the contact stiffness = The moduli were determined after a steady-state response of more than 20 cycles. After a recovery time of 5 min, a second series of indentation tests was conducted at the same location and in reverse order. The results of the second series did not differ significantly from those of the first series, which implied that the testing order and the degree of recovery had no effect on the stress-strain relationship.
In each region, the mean and standard deviation of |E*|, E', E'', and tan
In all 4 regions, the values of |E*| and E' were proportional with the frequency (Fig. 2  ). |E*| and E' increased faster with the frequency in the specimens from the anteromedial and anterolateral regions than in the other 2 regions. The values of E'' showed only a small change with frequency. The loss modulus E'' increased between 0.01 and 0.1 Hz, but it exhibited almost no changes from 0.1 to 10 Hz. In the anteromedial region, the largest values for E'' were found. The ANOVA revealed a significant effect (p < 0.001) of the strain frequency on the values of |E*| and E' for all 4 regions.
The loss tangent decreased considerably with the frequency between 0.01 and 0.1 Hz. This decrease diminished with frequencies beyond 0.1 Hz (Fig. 3 ). Throughout the entire range of frequencies tested, the specimens from the anteromedial region had a slightly larger loss tangent than those from the remaining 3 regions. The tan decreased from 0.68 to 0.17.
In the anteromedial region, the |E*| and E' moduli were 1.40 ± 0.39 MPa and 1.36 ± 0.38 MPa, respectively, at a frequency of 1.0 Hz (Fig. 4 ). These values were highest among the 4 regions. The |E*| and E' moduli in the anteromedial and anterolateral regions were significantly (P < 0.05) larger than those in the posteromedial and posterolateral regions. The highest E'' (0.34 ± 0.07 MPa) was also found in the anteromedial region, and it was significantly (P < 0.01) larger than those in the remaining 3 regions. The loss tangent ranged from 0.21 to 0.27 at 1.0 Hz; the differences among the 4 regions were not significant.
Dynamic indentation tests on the mandibular condylar cartilage have been carried out at a frequency of 14 Hz (Hu et al., 2001). The viscoelastic properties, however, depend on the frequency. Therefore, for precise characterization of the biomechanical behavior of the condylar cartilage during function, its mechanical properties have to be examined over a wide range of frequencies. To our knowledge, this is the first study in which the dynamic compressive properties of the condylar cartilage have been examined for such a wide frequency range. With respect to joint morphology, the porcine TMJ has a shape more or less similar to that in humans (Bermejo et al., 1993). Functionally, the porcine TMJ shows both translational and rotational movements during mastication, like that of humans (Langenbach and van Eijden, 2001; Sun et al., 2002), although the chewing frequency in the pig is typically 2–3 Hz (Druzinsky, 1993), which is larger than that in the human (approx. 1 Hz) (Gallo et al., 2000). The pig can perhaps be assumed to be the best animal model for the TMJ (Sun et al., 2002; Herring, 2003). In this study, we investigated the influence of frequency on the dynamic properties by examining a range of frequencies (from 0.01 to 10 Hz). This range can be considered sufficient to cover habitual loading in both pigs and humans (Druzinsky, 1993; Langenbach and van Eijden, 2001). In our study, the dynamic complex and storage moduli increased as the frequency increased, regardless of the region. This implies that the stiffness of the condylar cartilage enlarges with frequency. The loss modulus E'' increased but slightly from 0.01 to 0.1 Hz, and was almost constant thereafter. E'' describes the viscous behavior of the cartilage and is proportional to the average dissipation of energy through heat by deformation (Tanaka and van Eijden, 2003). Two mechanisms are responsible for the viscoelastic behavior of cartilage (Stolz et al., 2004): (1) a flow-independent mechanism, intermolecular friction, exhibited in all polymeric materials; and 2) a flow-dependent mechanism that is present if loading conditions allow water to move through the structure. Our findings indicate that fluid flow within and out of the cartilage is slower during cyclic compression. Below a certain frequency (approx. 0.1 Hz), the fluid flow may match up to the applied frequency, resulting in a flow-dependent mechanism. Meanwhile, at higher frequencies (more than 0.1 Hz), the proteoglycans occupying the interfibrillar spaces interfere with smooth fluid flow, which leads to strain energy dissipation, resulting in a higher stiffness. This is probably due to a flow-independent mechanism. Within the frequency range examined in the present study, values of 1.0 Hz reflect chewing conditions (Druzinsky, 1993; Gallo et al., 2000). These values are considered to be important in assessment of the clinical significance of the results. Therefore, in this study, the regional difference of the dynamic viscoelasticity was evaluated by use of the results at 1.0 Hz, although chewing frequency in pigs is typically 2–3 Hz. A similar significant difference can be expected at 2–3 Hz. The sinusoidal strain used in this study was 1%, because the tissue could maintain contact with the plate during the compression/tension cycles. The anterior regions exhibited significantly higher values than the posteromedial and posterolateral regions in complex modulus |E*| and storage modulus E'. Furthermore, the loss modulus E'' showed significant regional differences, i.e., the anteromedial region had a significantly higher value of E'' than did the other regions. Therefore, it can be concluded from the present study that the viscoelasticity of the condylar cartilage is a region-specific behavior. Hu et al.(2001), using fresh fibrocartilage samples of rabbit condyles, conducted nanoindentation with atomic force microscopy, and showed that regional differences of its ultrastructure and viscoelastic properties appear to be correlated. They also revealed that the gradient distribution of Young’s moduli from the highest anteromedial region (mean, 2.34 MPa) to the lowest posterolateral region (mean, 0.95 MPa) further differentiated between the stress-bearing capacities among different regions. Since they conducted nanoindentation tests at 14 Hz, the values of elastic moduli were greater than those in our study at 1 Hz. However, the regional difference, that the anteromedial region exhibited the highest moduli in the 4 regions, is consistent with our result. It is known that highly stressed regions of articular cartilage are stiffer in compression than regions that experience less-compressive stresses (Ahmed and Burke, 1983). These regional differences are statistically significant, indicating that different regions may have been constructed to withstand a gradient propagation of shear stress. Furthermore, according to Hu et al.(2001), the porcine condylar surface shows a decreasing gradient of surface roughness from the anteromedial to the posterolateral region, indicating a co-regulation of the ultrastructural properties of fibrocartilage with its viscoelastic properties. The regional difference of the surface roughness might result in regional differences of compressive moduli and subsequent stresses. However, histochemical studies have not detected obvious regional differences of proteoglycans, such as aggrecan, in the condylar cartilage, although versican, which is the large interstitial chondroitin sulfate/keratan sulfate-proteoglycan, was predominantly localized in the anterior area of the condylar cartilage (Shibata et al., 2001). These GAGs are considered to be a determinant for the capacity for resistance to compression. Consequently, the regional difference in the viscoelastic properties in the condylar cartilage, as observed in the present study, may reflect the distribution of GAG associated with compressive resistance of the condylar cartilage.
We thank Japan Agriculture Hiroshima (Hiroshima) for provision of porcine TMJs. This research was supported by a grant (No. 16592045) for Science Research from the Ministry of Education, Science and Culture, Japan. Received for publication July 12, 2005. Revision received January 19, 2006. Accepted for publication February 16, 2006.
Journal of Dental Research, Vol. 85, No. 6,
571-575 (2006) This article has been cited by other articles:
|
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
|
 |
|
Sign In to gain access to subscriptions and/or personal tools.
TOP
ABSTRACT
INTRODUCTION
, angular velocity) (B). (A) The articular fibrocartilage was divided into 4 regions; anteromedial, anterolateral, posteromedial, and posterolateral. For each region, the center of the crosspoint, as illustrated in the anteromedial region, was selected for indentation. (B) The sinusoidal strain was produced by a tension control motor in the driver, and the stress 
and strain 
were measured by means of load and displacement detectors and transmitted to a data processor. In a viscoelastic material, the phase difference between stress and strain is somewhere between (0 < 
/2), and the complex modulus E* is resolved into two components: the storage modulus E' and the loss modulus E'', shown in vector format. The tangent of the phase angle (
E, where the indentation modulus, E is determined by 
–1, and tan
