| Sign In to gain access to subscriptions and/or personal tools. |
Regional Dynamic Tensile Properties of the TMJ DiscDepartment of Chemical and Petroleum Engineering, University of Kansas, 1530 W. 15th St., Room 4132, Lawrence, KS 66045-7609, USA Correspondence: * corresponding author, detamore{at}ku.edu
Although the TMJ disc has been well-characterized under tension and compression, dynamic viscoelastic regional and directional variations have heretofore not been investigated. We hypothesized that the intermediate zone under mediolateral tension would exhibit lower dynamic moduli compared with the other regions of the disc under either mediolateral or anteroposterior tension. Specimens were prepared from porcine discs (3 regions/direction), and dynamic tensile sweeps were performed at 1% strain over a frequency range of 0.1 to 100 rad/sec. Generally, the intermediate zone possessed the lowest storage and loss moduli, and the highest loss tangent. This study further accentuates the known distinct character of the intermediate zone by showing for the first time that these differences also extend to dynamic behavior, perhaps implicating the TMJ disc as a structure primarily exposed to predominantly anteroposterior tension via anterior and posterior attachments, with a need for great distension mediolaterally across the intermediate zone.
Key Words: temporomandibular joint disc • dynamic • viscoelasticity • biomechanics
The temporomandibular joint (TMJ) disc experiences static loading during clenching, grinding, and bruxism, and dynamic loading during talking and chewing (Kuboki et al., 1997; Beatty et al., 2001; Tanaka and van Eijden, 2003). Biomechanical characterization of the TMJ disc is essential to the development of a tissue-engineered replacement disc, and is necessary for finite element modeling. Finite element models of the disc characterize compressive, tensile, and shear forces during mandibular motion (Beek et al., 2000; Koolstra, 2003; Donzelli et al., 2004; Tanaka et al., 2004a), and may be able to address questions of pressing clinical significance in the future (Detamore et al., 2007). As a viscoelastic material, the disc exhibits a time-dependent stress-strain response. The TMJ disc exhibits distinct regional variations, leading to commonly defined regions (Rees, 1954). The disc is heterogeneous and anisotropic, with the greatest structural and functional regional distinction being between (1) the intermediate zone and (2) the anterior and posterior bands (Detamore et al., 2005). Variations of compressive properties and tensile properties do not correspond to each other, due to both the differing structural factors contributing to compressive and tensile integrity and to the different axes for tensile (anteroposterior or mediolateral) and compressive (supero-inferior) testing. Single-deformation (as opposed to dynamic) tensile testing of the disc has elucidated the dramatic differences in tensile behavior among regions in the mediolateral and anteroposterior directions, which can be attributed to heterogeneous and anisotropic orientation of collagen fibers (Teng and Xu, 1991; Detamore and Athanasiou, 2003; Scapino et al., 2006). Dynamic loading has been investigated with regard to shear, compressive, and tensile stress (Beek et al., 2001; Tanaka et al., 2002, 2003a, Tanaka et al., b,d, 2004b; Beatty et al., 2003; Koolstra et al., 2007). The current study focused specifically on the dynamic tensile properties of the TMJ disc under dynamic tension. Previous dynamic tensile studies of the TMJ disc have examined the effects of age-related changes (Tanaka et al., 2002), proteoglycan content (Tanaka et al., 2003a), and impulsive compression prior to tension (Tanaka et al., 2003c), all in the anteroposterior direction. The purpose of this study was to analyze the dynamic uni-axial tensile properties of the porcine TMJ disc, examining for the first time its regional and directional differences in dynamic behavior. Pigs were selected based on the similarity of their TMJ to the human TMJ (Herring, 2003) and recommended use for TMJ tissue biomechanics studies (Detamore et al., 2007). Based on previous single-deformation studies that compared the intermediate zone with other regions in tension (Teng et al., 1991; Beatty et al., 2001; Detamore and Athanasiou, 2003;), we hypothesized that the intermediate zone would have lower dynamic moduli in comparison with the other regions of the disc.
Specimen Preparation TMJ discs were dissected from pig heads (Yorkshire cross, female, weighting 70–85 kg each) obtained at a local slaughterhouse. Discs were harvested within 24 hrs of death, wrapped in gauze, soaked in 0.01 M phosphate-buffered saline (PBS – 0.138 M sodium chloride, 0.0027 M potassium chloride), and stored at –20°C. Left and right discs from 6 pigs were obtained (n = 6 for each of the 6 regions). From each left disc, 3 specimens were made in the mediolateral direction (Fig. 1  ). These specimens comprised the anterior and posterior bands, and the intermediate zone. From each right disc, 3 specimens (medial, lateral, and central regions) were made in the anteroposterior direction (Fig. 1).
TMJ discs were thawed and then cut into 500-µm-thick sections by means of a cryotome (Microm HM550, Richard-Allan Scientific, Kalamazoo, MI, USA). To obtain uniform sections from a consistent location, we marked the superior surface of each dissected disc with India ink. The tissue was frozen such that the superior surface was flat, and then it was placed on a stage and covered in embedding medium. We made a series of 10-µm sections from the superior surface of the specimen until the ink was removed, and then made a 500-µm cut to obtain the disc section. We removed the surface to test the internal architecture immediately below the superior surface, which has a documented collagen fiber architecture (Detamore and Athanasiou, 2003; Scapino et al., 2006) crucial to tensile behavior. It was anticipated that removal of the surface zone may result in indeterminate changes in storage and loss moduli. From the 500-µm section, the disc was divided into its 3 respective regions. Rectangular specimens were made from each region by means of parallel razor blades, uniformly spaced at 2 mm. These specimens were wrapped in gauze, soaked in PBS, then frozen at –20°C until tested (Allen and Athanasiou, 2005).
Dynamic Tensile Tests
Waterproof sandpaper was glued to the grip surfaces by cyanoacrylate adhesive to prevent slippage, and caused no damage to the specimen (Detamore and Athanasiou, 2003). TMJ disc specimens were fixed in the tensile testing grips, and tests were performed at room temperature in a bath of PBS (Detamore and Athanasiou, 2003; Park and Ateshian, 2006) following a 1-g tare load. During the dynamic tensile tests, a strain of 1% was applied to the specimen over a frequency range of 0.1 to 100 rad/sec (preliminary tests with a strain amplitude of 0.1% gave comparable results; absence of pre-strain beyond the tare load provides the lower bound on E' and E'', as indicated by work now under way to measure E' and E'' as a function of pre-strain). To rule out slippage and specimen damage, we conducted several preliminary studies, including confirming independence of the direction of the frequency sweep (high to low vs. low to high). Values for the storage modulus (E'), loss modulus (E''), and loss tangent (tan
Statistical Analysis
In general, storage and loss moduli increased as frequency increased from 0.1 to 10 rad/sec, and began to decrease from 10 to 100 rad/sec (Fig. 2 ). Generally, viscous damping, as measured by the loss tangent, was relatively higher at lower frequencies, and decreased and leveled off by 10 rad/sec (Fig. 3). For a frequency of 1 rad/sec, storage moduli ranged from 0.52 MPa in the intermediate zone to 2.9 MPa in the medial region, and loss moduli ranged from 0.075 MPa in the intermediate zone to 0.30 MPa in the medial region (Fig. 4). Correspondingly, the loss tangents at a frequency of 1 rad/ sec ranged from 0.10 to 0.15 (Fig. 4). At frequencies of 1 rad/ sec and 10 rad/sec, the medial region, followed by the anterior region, exhibited the highest mean values of storage and loss moduli. In contrast, the intermediate zone, followed by the posterior band, exhibited the lowest mean values of storage and loss moduli. At 1 rad/sec, the storage modulus was significantly greater in the medial region than in the intermediate zone (p < 0.005) and posterior band (p < 0.05), and the central region (p = 0.05) and anterior band (p < 0.05) also had larger storage moduli than the intermediate zone.
The intermediate zone, followed by the posterior zone, exhibited the greatest loss tangents at frequencies of 1 rad/sec and 10 rad/sec, whereas the medial region was found to have the smallest loss tangent (Fig. 4  ). At 1 rad/sec, the loss tangent was significantly higher in the intermediate zone and posterior band relative to the anterior band (p < 0.01), medial region (p < 0.005), and lateral region (p < 0.05). In pooling the regions in each of the two directions, we found that the respective values in the mediolateral and anteroposterior directions at 1 rad/sec were 1.3 ± 1.5 and 2.3 ± 1.3 for the storage modulus (p < 0.05), 0.14 ± 0.13 and 0.25 ± 0.13 for the loss modulus (p < 0.05), and 0.14 ± 0.03 and 0.12 ± 0.02 for the loss tangent (p < 0.05).
In comparison with previous TMJ disc dynamic tensile studies (Tanaka et al., 2002, 2003a,Tanaka et al., c), the removal of the surface in the current study should be noted, and only the anteroposterior central region data can be directly compared. Overall, findings were in close agreement. Loss moduli in this region generally ranged from 0.1 MPa to 0.33 MPa in the current study, comparable with previously reported loss moduli of about 0.07 to 0.25 MPa. Loss tangents ranged from 0.12 to 0.17, compared with a range of 0.1 to 0.25 found previously, indicating a well-cross-linked material with a relatively small amount of viscous damping. Storage moduli in the current study generally ranged from 0.8 to 2.3 MPa, compared with 0.4-2 MPa previously reported. The current study and the studies by Tanaka et al. were all performed in the non-linear low-strain regions of the stress-strain curves (the "toe region"), which extends to ~ 6% tensile strain for the TMJ disc (Detamore and Athanasiou, 2003); increasing pre-strains within this region is hypothesized to increase storage and loss moduli (as seen in follow-up studies currently under way), reaching constant values in the linear region of the stress-strain curve. A decline in the storage modulus was typically observed above ~ 70 rad/sec. The exact frequency where the downturn begins depends on experimental parameters such as the length of the sample between the grips. This is consistent with resonance effects that arise in dynamic mechanical testing (Ferry, 1980; Menard, 1999). Such effects have been observed at similar frequencies in other soft tissues, such as muscle (Wakeling and Nigg, 2001), tendon (Wang et al., 2007), and pericardium (Mavrilas et al., 2005), as well as in hyaluronic acid and collagen gels (Klemuk and Titze, 2004). Data above the downturn frequency were discarded in these cases, since lower-frequency data were not affected by this phenomenon (Menard, 1999; Klemuk and Titze, 2004; Mavrilas et al., 2005). Since the conclusions of this work rest on interpretation of the data below 70 rad/sec, further analysis of this phenomenon is separate from the goals of this report, though its quantitative analysis is under investigation. With regard to the resonance effect, the variation in heterogeneity between the anteroposterior and mediolateral directions may have played a role in determining the degree of storage moduli decrease between these directions. Based on regional analysis of dynamic tensile moduli and the loss tangent, the intermediate zone was found to have statistically significant differences in properties in comparison with most of the other regions. In general, the intermediate zone was found to have the lowest storage modulus, lowest loss modulus, and highest loss tangent. This indicates that the disc in this region is softer and less cross-linked than elsewhere. Previous studies examining regional variations under static tension have demonstrated that the intermediate zone has a lower modulus than all regions in the mediolateral direction and anteroposterior directions (Teng et al., 1991; Beatty et al., 2001; Detamore and Athanasiou, 2003). The results of the current study underscore this relationship between the intermediate zone and all other regions. Indeed, lower dynamic moduli in the intermediate zone correspond nicely with lower static moduli found in previous studies. Collagen fibers, which provide the tensile integrity of the disc, run anteroposteriorly in the intermediate zone (Scapino et al., 2006). Thus the fibers provide less resistance in this region than in the anterior and posterior bands of the disc under mediolateral tension, and in all regions under anteroposterior tension, in which fibers are oriented, to a large extent, parallel to the direction of tension. The results indicated that the intermediate zone also had a greater viscous character than the other regions of the disc. It should be noted that the samples representing the posterior band in this study were taken from a position slightly anterior to the location used in a related previous study (Detamore and Athanasiou, 2003), which likely resulted in the relatively similar behavior observed between the posterior band and intermediate zone in the current study, as opposed to the striking contrast between the two regions in previous reports (Teng et al., 1991; Detamore and Athanasiou, 2003). Future studies may investigate numerous regions in the mediolateral direction for better characterization of the transition from the behavior of the posterior band to that of the intermediate zone. To the best of our knowledge, this is the first study to present a multi-regional, multi-directional characterization of the dynamic tensile properties of the TMJ disc. These results help to paint a bigger picture of TMJ disc biomechanics by further elucidating the distinct character of the intermediate zone relative to all other regions of the disc. Under tension, the intermediate zone under mediolateral tension is known to be less stiff, less strong, less tough, and more ductile than the other regions (Teng et al., 1991; Beatty et al., 2001; Detamore and Athanasiou, 2003). The current study has shown not only that the lower stiffness shown in previous studies is also observed in the dynamic moduli, but also that the intermediate zone generally appears to be more viscous. Therefore, although the intermediate zone is more prone to fracture and deformation, it is much more yielding, with a much higher fracture strain and viscous character. These biomechanical property observations collectively stand as a testament to the complex biomechanical environment of the TMJ, perhaps implicating the TMJ disc as a structure primarily exposed to predominantly anteroposterior tension via its anterior and posterior attachments, with a need for great distension mediolaterally across the intermediate zone.
This research was supported by the University of Kansas General Research Fund, and partial support of NSF IOS-0726412 is also acknowledged. We thank the reviewers for helpful comments for improving the manuscript. Received for publication August 16, 2007. Revision received July 23, 2008. Accepted for publication August 1, 2008.
Journal of Dental Research, Vol. 87, No. 11,
1053-1057 (2008)
|
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
|
 |
|
Sign In to gain access to subscriptions and/or personal tools.
TOP
ABSTRACT
INTRODUCTION
) were obtained over this frequency range. 
