journal of the mechanical behavior of biomedical materials 41 (2015) 108 –114
Available online at www.sciencedirect.com
www.elsevier.com/locate/jmbbm
Research Paper
Regional and depth variability of porcine meniscal mechanical properties through biaxial testing A. Kahlona, M.B. Hurtigb, K.D. Gordona,n a
School of Engineering, University of Guelph, Guelph, Ontario, Canada Clinical Studies, Ontario Veterinary School, University of Guelph, Guelph, Ontario, Canada
b
ar t ic l e in f o
abs tra ct
Article history:
The menisci in the knee joint undergo complex loading in-vivo resulting in a multi-
Received 18 June 2014
directional stress distribution. Extensive mechanical testing has been conducted to
Received in revised form
investigate the tissue properties of the knee meniscus, but the testing conditions do not
7 October 2014
replicate this complex loading regime. Biaxial testing involves loading tissue along two
Accepted 8 October 2014
different directions simultaneously, which more accurately simulates physiologic loading
Available online 19 October 2014
conditions. The purpose of this study was to report mechanical properties of meniscal
Keywords:
tissue resulting from biaxial testing, while simultaneously investigating regional variations
Meniscus
in properties. Ten left, fresh porcine joints were obtained, and the medial and lateral
Mechanical properties
menisci were harvested from each joint (twenty menisci total). Each menisci was divided
Biaxial testing
into an anterior, middle and posterior region; and three slices (femoral, deep and tibial layers) were obtained from each region. Biaxial and constrained uniaxial testing was performed on each specimen, and Young's moduli were calculated from the resulting stress strain curves. Results illustrated significant differences in regional mechanical properties, with the medial anterior (Young's modulus (E)¼11.1471.10 MPa), lateral anterior (E ¼11.5471.10 MPa) and lateral posterior (E ¼9.071.2 MPa) regions exhibiting the highest properties compared to the medial central (E¼ 5.071.22 MPa), medial posterior (E ¼4.1671.13 MPa) and lateral central (E¼ 5.671.20 MPa) regions. Differences with depth were also significant on the lateral meniscus, with the femoral (E ¼12.771.22 MPa) and tibial (E¼ 8.671.22 MPa) layers exhibiting the highest Young's moduli. This data may form the basis for future modeling of meniscal tissue, or may aid in the design of synthetic replacement alternatives. & 2014 Elsevier Ltd. All rights reserved.
n Correspondence to: School of Engineering, University of Guelph, 50 Stone Road East, Guelph Ontario, Canada, N1G 2W1. Tel.: þ1 519 824 4120x52435. E-mail address: [email protected] (K.D. Gordon).
http://dx.doi.org/10.1016/j.jmbbm.2014.10.008 1751-6161/& 2014 Elsevier Ltd. All rights reserved.
journal of the mechanical behavior of biomedical materials 41 (2015) 108 –114
1.
Introduction
The meniscus, which is composed of two C-shaped fibrocartilagenous tissues, is interposed between the articulating surfaces of the femur and tibia in the knee joint. Its functions have been described as load distribution, joint stabilization, and joint lubrication (Allen et al., 2000; Arnoczky and McDevitt, 2000; Favenesi et al., 1983; Radin et al., 1984), making the meniscus an integral component in knee biomechanics. It has been established that meniscal injuries and deterioration of this tissue can lead to osteoarthritis (Rangger et al., 1995; Ratzlaff and Liang, 2010). For this reason the standard of care in orthopedics is currently to salvage and restore damaged meniscus whenever possible (Hutchinson et al., 2014). As a result, an in-depth understanding of the indigenous structure and biomechanical functioning of the native meniscus is crucial to our progression in developing improved repair strategies and transplant alternatives. Tensile material properties of the meniscus have been studied extensively (Proctor et al., 1989; Fithian et al., 1990; Lechner et al., 2000; Goertzen et al., 1997; Skaggs et al., 1994; Tissakht and Ahmed, 1995). Although these studies vary in their methodology, including type of species, specimen dimensions and the region from which samples were extracted; one common factor is that separate uniaxial tests in the circumferential and radial directions have been performed. However, the meniscus experiences stresses simultaneously in both the circumferential and radial direction during in vivo loading-phases (Bylski-Austrow et al., 1994; Mow et al., 1992; Walker and Erkman, 1975), suggesting that a simultaneous multi-axial loading regime is more physiologically relevant. The uniaxial and biaxial tissue responses have been shown to be significantly different in other tissue types (Eilaghi et al., 2010; Gregory and Callaghan, 2011). A number of studies have looked at the biaxial tensile response of collagenous tissues (Eilaghi et al., 2010; Gregory and Callaghan, 2011; Holmes et al., 2012) but none have investigated this response in the meniscal tissue. Given the current focus on salvaging meniscal tissue and developing synthetic alternatives, a thorough investigation into the biaxial response of meniscal tissue is warranted. The distinct layers of the ultrastructure in the meniscus can be identified by the differences in the collagenous fiber alignment (Mow et al., 1992; Petersen and Tillmann, 1998). The superficial layer at the femoral and tibial surfaces consists of a smaller fibril network (bundles of approximate diameter 10 mm, fibrils of approximate diameter 35 nm) without any distinct orientation. Below this surface is the lamellar layer, identified with a larger fibril network (bundles: 20– 50 mm, fibrils: 120 nm), that has mostly randomly oriented fibers except at the periphery of the anterior and posterior segments where the fibers are radially oriented. The largest volume of the meniscus is the deep zone which has predominately circumferentially oriented fibers intermixed with radially aligned tie fibers (Mow et al., 1992; Petersen and Tillmann, 1998). Skaggs et al. (1994) found regional differences in the distribution of these radial tie fibers in medial bovine menisci with corresponding regional differences in the uniaxial tensile moduli directly correlating to the size of
109
the radial tie fiber. Tissakht and Ahmed (1995) investigated the regional and depth variations of the uniaxial tensile material properties in human meniscal tissue. With similar objectives Proctor et al. (1989) investigated bovine meniscus by examining these properties from the femoral and deep regions. The results of these previous studies suggest the mechanical response in meniscal tissue is non-homogenous. However, current finite element models (FEM) of the meniscus have modeled the meniscus as transversely isotropic without accommodating regional variation in properties (Haut Donahue et al., 2003; Bao et al., 2013; Dong et al., 2014; Mononen et al., 2013). Moreover, while little has been reported on the mechanical response of the two scaffolds currently commercially available (Spencer et al., 2012), there may be benefits to gain from synthesizing scaffolds of nonhomogenous properties. This approach to synthetic meniscal substitute design could result in better clinical outcomes following surgery. Therefore, an in-depth study of the biaxial material properties will be investigated to gain understanding of the more physiological material response of the meniscus. Furthermore, regional differences (anterior, central, and posterior) and depth variations (femoral lamellar, deep, and tibial lamellar) of porcine lateral and medial meniscus will be examined. The resulting data may be used in improve models of meniscus under load to better understand physiologic response. This may, in turn, lead to improved surgical repair method and development of synthetic alternatives and tissue engineered scaffolds.
2.
Material and methods
2.1.
Specimen preparation
Ten fresh-frozen left porcine knee joints were obtained from a local abattoir. On the day prior to dissection, the joints were thawed overnight. The lateral and medial menisci from each knee joint were carefully harvested by sharp dissection with a scalpel. Each menisci was carefully inspected for evidence of degeneration or damage from the dissection. Each menisci was then divided with a scalpel into three regions (anterior, central, and posterior) (Fig. 1). In order to maintain consistency for choosing the region, a radial center line dividing the menisci into two halves was drawn. The size of the central block was selected to be large enough to fit two 7 7 mm2 square blocks on the femoral surface with the anterior and posterior regions being adjacent to these blocks. The anterior and posterior horns were then removed. Each region was then cut into a smaller block with an approximate inner arc length of 10 mm without removing any inner or peripheral tissue. The membranes from the femoral and tibial surfaces of each block were removed. To ensure that the slices were obtained from both the femoral and tibial surfaces each block was then carefully cut circumferentially using a scalpel into three further depth regions (femoral, middle, and tibial) (Fig. 1). Finally, using a square guide, 7 7 mm2 slices from each layer of approximate thickness of 1 mm were obtained. In total, nine slices per menisci were removed. Extra care was taken to ensure that the inner and
110
journal of the mechanical behavior of biomedical materials 41 (2015) 108 –114
Fig. 1 – Each menisci was first sectioned using a scalpel into three regions (anterior, central, and posterior). Each region was then further divided into three depths (femoral, middle, and tibial), to obtain nine specimens per menisci.
the actuator along the circumferential axis while maintaining the position of the actuator along the radial axis), and lastly a constrained uniaxial test in the radial direction. All tests were performed at 1% s 1 (0.006 mm/s) for 7 s with a 7 s rest after preconditioning and after each individual uniaxial and biaxial test. This loading rate was chosen to ensure quasi-static loading conditions as outline in previous studies (Lechner et al., 2000; Tissakht and Ahmed, 1995). A final strain of 7% was chosen to ensure the linear region of the stress strain curve would be reached while incurring no damage to the tissue. Fig. 2 – A 7 7 mm2 bovine meniscus specimen mounted in the biaxial test system.
peripheral cuts of the square block were parallel to the circumferential running collagen fibers. The thickness of each slice was measured at five locations using a custom designed electrical resistant gage using a micrometer (Lee and Langdon, 1996) measuring device (resolution¼ 0.01 mm) and then averaged. The slices were kept frozen at 20710 1C until further testing. There are several studies confirming no change in mechanical properties of bone (Shaw et al., 2012) and soft tissue (Jung et al., 2011) after several (up to eight) freeze thaw cycles.
2.2.
Tensile testing
Tensile testing was performed using a biaxial test system (BioTester 5000; Cell Scale, Waterloo, ON, Canada). Custom rakes with five tines were placed into the rake base and opposing rakes were positioned 6 mm apart. Specimens were carefully mounted by aligning the circumferential fiber direction with the X-direction of the actuator and then inserting rake tines on all four sides into the specimen (Fig. 2). Any abundant saline solution floating on the specimen was removed using paper towel before carefully spreading the specimen with black graphite particles using a Q-tip. The graphite particles were applied to provide contrast for the optical strain measurement calculation outlined in Section 2.3. Specimens were preconditioned at 1% s 1 (0.006 mm/s) for 7 s for ten sinusoidal cycles. This testing protocol was found to generate repeatable results in pilot testing conducted in our lab. The specimens were then stretched biaxially (equal displacement along both axes), followed by a constrained uniaxial test in the circumferential direction (displacement of
2.3.
Data analysis
Force data was obtained from the load cell output on each actuator and a digital image correlation technique was used to quantify the strains produced during testing from images of the specimen and graphite particles during testing (Veldhuis et al., 2005). Particle tracking of the graphite particles adhered to the surface of the tissue was used to then calculate the strains. Stress (σ) was calculated by taking the force data and dividing it by the initial cross-sectional area of the specimen. The average of the five thickness measurements was multiplied by the distance between the rake tines at the beginning of a test to obtain the initial crosssectional area. For each test the stress and strain values were plotted in Microsoft Excel and the linear trend-line tool was used in the linear portion of the curve to calculate Young's modulus (Fig. 3). The linear region of the stress strain curve was defined through repeated iterations of slope calculation until a maximized linear region had been defined with a Rsquared value of 0.85 or higher. In total four separate Young's modulus (E) values were calculated for the biaxial circumferential, biaxial radial, uniaxial circumferential and uniaxial radial loading. An ANOVA analysis was performed for each group of moduli to make comparisons between regions (anterior, central, and posterior), depth (femoral, middle, tibial) and side (medial, and lateral). Identified significance differences were further investigated using least square means test. A P-value of less than 0.05 was considered significant.
3.
Results
The number of specimens from each meniscus and from different locations and depths are summarized in Table 1. In some cases testing was not successfully completed due to
111
journal of the mechanical behavior of biomedical materials 41 (2015) 108 –114
human error during slicing, resulting in less than nine specimens obtained per menisci. The average thickness for all specimens was 1.0970.27 mm. For all tested specimens the average maximum difference from the average thickness was less than 2%, with the maximum difference of 5.7%. All data was normally distributed. Fig. 4 summarizes the difference between the biaxial and constrained uniaxial tests. The circumferential modulus was significantly larger (po0.0001) than the radial modulus. The modulus values for the biaxial data were also significantly larger (po0.001) than the counterpart for the constrained uniaxial test. The results of biaxial tensile testing can be found in Figs. 5 and 6. Fig. 5 summarizes the Young's modulus value for both lateral and medial specimens at three different locations (anterior, central, and posterior) and two directions (circumferential and radial) of the tissue. For the medial meniscus, in both the circumferential and radial direction, and the lateral meniscus circumferential specimens only, the anterior moduli were significantly higher (po0.0001) compared to the central and posterior locations. Specimens from the lateral anterior region tested in the radial direction were not significantly different than the lateral middle and posterior region but followed the same trend. Significant differences were also observed between the lateral central and lateral posterior region in the circumferential direction, with the posterior region displaying a higher modulus than the central (p¼ 0.04) (Fig. 5). Biaxial data for various depths (femoral, middle, and tibial) layers is summarized in Fig. 6. For the lateral menisci all three depths were significantly different (po0.004) in the
circumferential direction with the tibial layer displaying a higher modulus (p ¼0.05) compared to the middle layer in the radial direction. No significant differences were found for the medial specimens for the three depths studied.
4.
Discussion
The aim of this study was to quantify the biaxial tensile material properties of porcine meniscus with respect to regional and depth-wise variations. Young's moduli at three different anatomical locations (anterior, central and posterior), three different depths (femoral, middle, and tibial) and two sides (medial and lateral) were reported and compared. Three separate tensile tests (biaxial, constrained uniaxial in the circumferential, and constrained uniaxial in the radial direction) were performed to obtain four separate Young's modulus values (biaxial circumferential, biaxial radial, uniaxial circumferential and uniaxial radial loading). The biaxial loading regime is thought to represent increased physiological relevance as in-vivo stresses during loading are endured in multiple directions simultaneously (Bylski-Austrow et al., 1994; Mow et al., 1992; Walker and Erkman, 1975). The overall result for the biaxial and constrained-uniaxial data in the circumferential direction was significantly higher than the radial direction. This result is in accordance with the results obtained by previous studies that conducted uniaxial testing in two directions (Proctor et al., 1989; Tissakht and Ahmed, 1995) and can be explained by the predominance of circumferentially aligned collagen fibers in the meniscal
Young's Modulus (MPa)
9 8 7 6 5 4 3 2 1 0
Biaxial - X
Biaxial - Y
Uniaxial - X
Uniaxial - Y
Fig. 4 – Comparison of the biaxial and uniaxial data. For both tests the modulus values in the Y-direction (radial) were significantly lower than the result of the same test in the X-direction (circumferential). Modulus values obtained from the constrained uniaxial test were significantly lower than the biaxial test values. Values are mean7standard error.
Fig. 3 – A representative stress–strain curve illustrating the calculation of Young's modulus (E).
Table 1 – Number of specimens tested for each location (anterior, central, and posterior), each depth (femoral, middle, and tibial), and side (lateral and medial). Total number of specimens tested was 149; 71 lateral and 78 medial. Anterior
Femoral Middle Tibial
Central
Posterior
Lateral
Medial
Lateral
Medial
Lateral
Medial
8 8 8
9 8 9
8 7 8
10 8 8
8 8 8
9 8 9
112
journal of the mechanical behavior of biomedical materials 41 (2015) 108 –114
Young's Modulus (MPa)
14 12 10 8
Circumferential (X)- Modulus Radial (Y)Modulus
6 4 2 0
Anterior
Central
Posterior Anterior
Medial
Central
Posterior
Lateral
Fig. 5 – Bi-axial Young's modulus for lateral and medial porcine test samples at three different locations (anterior, central, and posterior). Each region represents the average value using data from all depths. Values are mean7the standard error of the mean. * indicates statistical significance.
Young's Modulus (MPa)
16 14 12 10
Circumferential (X)-Modulus
8
Radial (Y)Modulus
6 4 2 0 Femoral
Middle
Medial
Tibial
Femoral
Middle
Tibial
Lateral
Fig. 6 – Bi-axial Young's modulus for lateral and medial porcine test samples at three different depths (femoral, middle, and tibial). Each depth represents the average value using data from all regions. Values are mean7the standard error of the mean. * indicates statistical significance. tissue (Mow et al., 1992; Petersen and Tillmann, 1998). The circumferential modulus values were also dependent on the anatomical location within the menisci with the anterior specimens having significantly higher values than the central and posterior regions. Similar to the results for the circumferential modulus, the radial modulus for the anterior region was significantly higher than the radial central and posterior regions for the medial menisci. Although not significant, the lateral anterior region had a higher modulus value (4.6471.22 MPa) than both the lateral central (2.2571.24 MPa) and lateral posterior (2.8671.22 MPa) regions in the radial direction. Farinaccio (1989) also reported the circumferential anterior region to be stiffer than the central and posterior in human menisci. Lechner et al. (2000) found higher modulus values in the anterior region of human menisci, but the result was not significant. Tissakht and Ahmed (1995) tested human meniscal samples and did not see any significant effect for region, and in contrast with the current study the posterior regions had slightly higher modulus values than the anterior
and central regions. Proctor et al. (1989) reported that the posterior segments of the deep regions were significantly stiffer than the anterior region in bovine meniscus, however an overall comparison between the posterior and anterior region was not provided. Finally, Fithian et al. (1990) reported that the posterior two-thirds of the medial meniscus had lower modulus values compared to the anterior and the lateral human meniscus. There is certainly variability in the results of these studies, but overall there is some consensus that the anterior and posterior regions of both the lateral and medial menisci have higher mechanical properties compared to the middle region. This is intuitive since there is likely more load applied in the anterior posterior direction during flexion and extension of the knee. This also supports the idea of regionally varying mechanical properties in a model of the meniscus. A comparison to past studies that have investigated the mechanical properties of the meniscal tissue using different experimental procedures is also enlightening. Baro et al. (2012) conducted indentation tests on medial bovine meniscus to study the variation of contact modulus with anatomical location (anterior, central, and posterior). Baro et al. (2012) found that the contact modulus was significantly higher in the anterior region and lowest at the posterior region, supporting the result of the current study with highest properties in the anterior region. Shear modulus has also been reported in bovine meniscus at three anatomical locations (Abraham et al., 2011). Although not significantly different, the anterior region had a higher shear modulus than the central region. The current study found that modulus also varied with depth for the lateral menisci. All three depths (tibial, middle, and femoral) were significantly different from each other with the femoral layer having the highest modulus and the middle layer the lowest modulus values. This may be explained due to higher amount of glycosaminoglycan (GAG), the proteoglycan responsible for water retention, in the deep zone of the meniscus (Moyer et al., 2013) which may be responsible for a lower modulus value (Sanchez-Adams et al., 2011). This may also be explained by the variation in collagen fiber orientation. The randomly oriented collagen fibers near the femoral and tibial surfaces result in stiffer mechanical properties while the predominately unidirectional fibers in the middle or deep layer result in lower mechanical properties. Tissakht and Ahmed (1995) obtained similar results for their radial specimens only, and saw similar trends for the circumferential specimens. However, their study included the approximately 100 mm thick meniscal membrane which has shown to be significantly stiffer in isolation than specimens from the surface layers with the membrane (Whipple et al., 1985). Interestingly, Baro et al. (2012) also observed an effect from the membrane when calculating the effective contact modulus. The mean value for the femoral layer was significantly higher than the deep and tibial layers and the mean tibial layer value was slightly higher than the deep layer (Baro et al., 2012). The average biaxial Young's modulus values obtained in this study (7.73 MPa circumferential and 2.80 MPa for radial specimens) are much lower than the values reported by previous relevant studies who conducted uniaxial testing
journal of the mechanical behavior of biomedical materials 41 (2015) 108 –114
only (Proctor et al., 1989; Tissakht and Ahmed, 1995; Lechner et al., 2000; Fithian et al., 1990). Tissakht and Ahmed (1995) obtained Young's modulus in the range of 90.22–102.12 MPa for circumferential and 7.82–13.04 MPa for radial human meniscal specimens. Proctor et al. (1989) reported values within a much wider range of 48.4–139 MPa for circumferential and (2.8–71.4 MPa) for radial specimens. Lechner et al (2000) reported values between 43.4 and 141.2 MPa, varying with anatomical region and sample size. We hypothesize that these variable results are arising from the different methods of sample preparation and testing. A literature review on biaxial testing of various collagenous tissues reveals results that are more harmonized with the current study. A study of the tensile material properties of the annulus fibrosis reported an average modulus value of 6.95 MPa in the direction parallel to collagen fiber alignment (Gregory and Callaghan, 2011). A comparison between the uniaxial, unconfined results (Wollensak and Spoerl, 2004) and biaxial testing results (Eilaghi et al., 2010) of the human sclera showed that the unconfined, uniaxial results were approximately 2 times higher than the biaxial data with the average biaxial modulus reported at 2.9 MPa. Other collagenous tissue studies have reported Young's modulus values in the range of 0.13– 2.76 MPa for the transverse carpal ligament under biaxial testing (Holmes et al., 2012) and between 38 MPa and 53 MPa for unconfined uniaxial testing of the entire ligament (Brett et al., 2014). These biaxial results are on the same order of magnitude as the results presented in the current study. We hypothesize that there are likely several reasons for the discrepancy between biaxial and unconfined uniaxial testing. The most obvious explanation would be the disruption of the network of collagen fibers when the specimens are prepared. The fibers that are not in-plane with the specimen dimensions are severed, and can therefore no longer offer the same resistance to applied load, lowering mechanical properties. When comparing unconfined to confined uniaxial testing, we would hypothesize that the unconfined testing state would allow the transverse and out-of-plane support fibers to strain in the direction of applied load, potentially increasing the overall resistance to load, increasing modulus. The length of the specimen and length of the collagen fibrils also likely plays a role, in that the individual collagen fibers may or may not run from grip to grip depending on the size of the specimen, also contributing to differing results. Further work is needed to explore these questions related to the preparation of samples and testing modalities of soft tissues. Currently, permanent and biodegradable synthetic meniscal scaffolds with isotropic mechanical properties are the focus of extensive research. These implants rely on the infiltration of cells to develop the anisotropic properties seen in indigenous meniscal tissues. Early clinical and experimental outcomes from these procedures are still not optimal (Sacks, 1999). This study provides evidence that varying the material properties from the anterior to the posterior regions as well as from the femoral to the tibial surface in synthetic manufactured scaffolds may lead to improved patient outcomes following surgery. The limitations of this study include the necessity to slice the specimens manually, which led to some variability in thickness and may have some impact on subsequent stress
113
calculations. Preliminary work using a vibratome also did not result in consistent slice thickness across a 7 7 mm2 slice, in fact careful dissection by hand generated more consistent results. Another limitation of this study was that quantification of collagen fiber orientation was not performed, and instead previous literature was used to determine the predominant collagen fiber direction. During mounting of the specimens for testing on the biaxial equipment, if the axis of collagen fiber orientation is not aligned with the actuator axis, shearing load can be introduced in the specimen (Sacks, 1999). It is possible that collagen fiber orientation may not have been perfectly aligned in all cases, and subsequently would impact mechanical properties calculations. The order of testing (biaxial tests conducted prior to uniaxial tests) was not randomized, and small changes in material properties may have occurred during the three minute duration of the testing due to dehydration. Finally the samples tested in this study were porcine, and thus may demonstrate different mechanical properties compared to human owing to the loading history on the meniscus tissue. The most notable difference between human and porcine anatomy is the posterior meniscal ligament, which attaches to the femur in the pig and to the tibia in a human. This variation in anatomy may result in altered loading of the meniscus, and subsequently varying material properties.
5.
Conclusions
In conclusion, this study is the first to perform biaxial testing on meniscal samples obtained from porcine knee joints. Biaxial testing provides an improved loading regime over previously conducted uniaxial testing in that more accurately represents loading seen in-vivo. The results of this study illustrate significant differences in regional mechanical properties, with the medial anterior, lateral anterior and lateral posterior regions exhibiting the highest properties. Differences with depth were also significant on the lateral meniscus, with the femoral and tibial layers exhibiting the strongest mechanical properties. This data may form the basis for future modeling of meniscal tissue, or design of synthetic replacement alternatives.
r e f e r e nc e s
Abraham, A.C., Edwards, C.R., Odegard, G.M., Donahue, T.L.J., 2011. Regional and fiber orientation dependent shear properties and anisotropy of bovine meniscus. Mech. Behav. Biomed. Mater. 4 (8), 2024–2030, http://dx.doi.org/10.1016/j. jmbbm.2011.06.022. Allen, C.R., Wong, E.K., Livesay, G.A., 2000. Importance of the medial meniscus in the anterior cruciate ligament-deficient knee. J. Orthop. Res. 18, 109–115. Arnoczky, S.P., McDevitt, C.A., 2000. The meniscus: structure, function, repair, and replacement. In: Buckwalter, J.A., Einhorn, T.A., Simon, S.R. (Eds.), Orthopaedic Basic Science: Biology and Biomechanics of the Musculoskeletal System. Rosemont, Illinois, pp. 531–545. Bao, H.R.C., Zhu, D., Gong, H., Gu, G.S., 2013. The effect of complete radial lateral meniscus posterior root tear on the knee contact
114
journal of the mechanical behavior of biomedical materials 41 (2015) 108 –114
mechanics: a finite element analysis. J. Orthop. Sci. 18 (2), 256–263, http://dx.doi.org/10.1007/s00776-012-0334-5. Baro, V.J., Bonnevie, E.D., Lai, X., Price, C., Burris, D.L., Wang, L., 2012. Functional characterization of normal and degraded bovine meniscus: rate-dependent indentation and friction studies. Bone 51 (2), 232–240, http://dx.doi.org/10.1016/j. bone.2012.03.009. Bylski-Austrow, D.I., Ciarelli, M.J., Kayner, D.C., Matthews, L.S., Goldstein, S.A.,, 1994. Displacements of the menisci under joint load: an in vitro study in human knees. J. Biomech. 27 (4), 421–431. Brett, A.W., Oliver, M.L., Agur, A.M., Edwards, A.M., Gordon, K.D., 2014. Quantification of the transverse carpal ligament elastic properties by sex and region. Clin. Biomech. 29 (6), 601–606. Dong, Y., Hu, G., Dong, Y., Hu, Y., Xu, Q., 2014. The effect of meniscal tears and resultant partial meniscectomies on the knee contact stresses: a finite element analysis. Comput. Methods Biomech. Biomed. Eng. 17 (13), 1452–1463, http://dx. doi.org/10.1080/10255842.2012.753063. Eilaghi, A., Flanagan, J.G., Tertinegg, I., Simmons, C.A., Brodland, W.G., Ethier, R.C., 2010. Biaxial mechanical testing of human sclera. J. Biomech. 43 (9), 1696–1701, http://dx.doi.org/10.1016/ j.jbiomech.2010.02.031. Farinaccio, R., 1989. An Experimental and Numerical Study of the Mechanical Response of the Knee Menisci (M. Eng. thesis). McGill University, Montreal, Quebec, Canada. Favenesi, J.A., Shaffer, J.C., Mow, V.C., 1983. Biphasic mechanical properties of the knee meniscus. Trans. Orthop. Res. Soc. 8, 57. Fithian, D.C., Kelly, M.A., Mow, V.C., 1990. Material properties and structure-function relationships in the menisci. Clin. Orthop. Relat. Res. 252, 19–31. Gregory, D.E., Callaghan, J.P., 2011. A comparison of uniaxial and biaxial mechanical properties of the annulus fibrosus: a porcine model. J. Biomech. Eng. 133 (2) (doi: 0.1115/1.4003327). Goertzen, D.J., Budney, D.R., Cinats, J.G., 1997. Methodology and apparatus to determine material properties of the knee joint meniscus. Med. Eng. Phys. 19 (5), 412–419. Haut Donahue, T.L., Hull, M.L., Rashid, M.M., Jacobs, C.R., 2003. How the stiffness of meniscal attachments and meniscal material properties affect tibio-femoral contact pressure computed using a validated finite element model of the human knee joint. J. Biomech 36 (1), 19–34. Holmes, M.W.R., Howarth, S.J., Callaghan, J.P., Keir, P.J., 2012. Biomechanical properties of the transverse carpal ligament under biaxial strain. J. Orthop. Res. 30 (5), 757–763, http://dx. doi.org/10.1002/jor.21583. Hutchinson, I.D., Moran, C.J., Potter, H.G., Warren, R.F., Rodeo, S. A., 2014. Restoration of the meniscus form and function. Am. J. Sports Med. 42 (4), 987–998, http://dx.doi.org/10.1177/ 0363546513498503. Jung, H., Vangipuram, G., Fisher, M.B., Yang, G., Hsu, S., Bianchi, J., Ronholdt, C., Woo, S.L.Y., 2011. The effects of multiple freeze– thaw cycles on the biomechanical properties of the human bone-patellar tendon-bone allograft. J. Orthop. Res. 29 (8), 1193–1198. Lechner, K., Hull, M.L., Howell, S.M., 2000. Is the circumferential tensile modulus within a human medial meniscus affected by the test sample location and cross-sectional area?. J. Orthop. Res. 18 (6), 945–951. Lee, M.J., Langdon, S.E., 1996. Thickness measurement of soft tissue biomaterials: a comparison of five methods. J. Biomech. 29 (6), 829–832.
Mononen, M.E., Jurvelin, J.S., Korhonen, R.K., 2013. Effects of radial tears and partial meniscectomy of lateral meniscus on the knee joint mechanics during the stance phase of the gait cycle—a 3D finite element study. J. Orthop. Res. 31 (8), 1208–1217, http://dx.doi.org/10.1002/jor.22358. Mow, V.C., Arnoczky, S., Jackson, D., 1992. Structure and function relationships of the meniscus in the knee, Knee Meniscus: Basic and Clinical Foundations. Raven Press, New York37–58. Moyer, J.T., Priest, R., Bouman, T., Abraham, A.C., Donahue, T.L., 2013. Indentation properties and glycosaminoglycan content of human menisci in the deep zone. Acta Biomater. 9 (5), 6624–6629, http://dx.doi.org/10.1016/j.actbio.2012.12.033. Petersen, W., Tillmann, B., 1998. Collagenous fibril texture of the human knee joint menisci. Anat. Embryol. 197 (4), 317–324. Proctor, C.S., Schmidt, M.B., Whipple, R.R., Kelly, M.A., Mow, V.C., 1989. Material properties of the normal medial bovine meniscus. J. Orthop. Res. 7, 771–782. Radin, E.L., de Lamotte, F., Maquet, P., 1984. Role of the meniscus in the distribution of stress in the knee. Clin. Orthop. Relat. Res. 185, 290–294. Rangger, C., Klestil, T., Gloetzer, W., Kemmler, G., Benedetto, L.P., 1995. Osteoarthritis after arthroscopic partial meniscectomy. Am. J. Sports Med. 23, 240–244. Ratzlaff, C.R., Liang, M.H., 2010. New developments in osteoarthritis. Prevention of injury-related knee osteoarthritis: opportunities for the primary and secondary prevention of knee osteoarthritis. Arthritis Res. Ther. 12 (4), 215–220. Sacks, M., 1999. A method for planar biaxial testing that includes in-plane shear. J. Biomech. Eng. 121 (5), 551–555. Sanchez-Adams, J., Willard, V.P., Athanasiou, K.A., 2011. Regional variation in the mechanical role of the knee meniscus glycosaminoglycans, J. App. Phys. 111 (6), 1590–1596, http://dx. doi.org/10.1152/japplphysiol.00848.2011. Skaggs, D.L., Warden, W.H., Mow, V.C., 1994. Radial tie fibers influence the tensile properties of the bovine medial meniscus. J. Orthop. Res. 12, 176–185. Shaw, J.M., Hunter, S.A., Gayton, J.C., Boivin, G.P., Prayson, M.J., 2012. Repeated freeze-thaw cycles do not alter biomechanics properties of fibular allograft bone. Clin. Orthop. Relat. Res. 470 (3), 937–943. Spencer, S.J., Saithna, A., Carmont, M.R., Dhillon, M.S., Thompson, P., Spalding, T., 2012. Meniscal scaffolds: early experience and review of the literature. Knee 19, 760–765, http://dx.doi.org/10.1016/j.knee.2012.01.006. Tissakht, M., Ahmed, A.M., 1995. Tensile stress–strain characteristics of the human meniscal material. J. Biomech. 28 (4), 411–422. Veldhuis, J.H., Brodland, G.W., Wiebe, C.J., Bootsma, G.J., 2005. Multiview robotic microscope reveals the in-plane kinematics of amphibian neurulation. Ann. Biomed. Eng. 33 (6), 821–828, http: //dx.doi.org/10.1007/s10439-005-3309-2. Walker, P.S., Erkman, M.J., 1975. The role of the menisci in force transmission across the knee. Clin. Orthop. Relat. Res. 109, 184–192. Whipple, R.R., Wirth, C.R., Mow, V.C., 1985. Anisotropic and zonal variations in the tensile properties of the meniscus. Trans. Orthop. Res. Soc. 10, 367. Wollensak, G., Spoerl, E., 2004. Collagen crosslinking of human and porcine sclera. J. Cataract Refract. Surg. 30 (3), 689–695, http://dx.doi.org/10.1016/j.jcrs.2003.11.032.
Available online at www.sciencedirect.com
www.elsevier.com/locate/jmbbm
Research Paper
Regional and depth variability of porcine meniscal mechanical properties through biaxial testing A. Kahlona, M.B. Hurtigb, K.D. Gordona,n a
School of Engineering, University of Guelph, Guelph, Ontario, Canada Clinical Studies, Ontario Veterinary School, University of Guelph, Guelph, Ontario, Canada
b
ar t ic l e in f o
abs tra ct
Article history:
The menisci in the knee joint undergo complex loading in-vivo resulting in a multi-
Received 18 June 2014
directional stress distribution. Extensive mechanical testing has been conducted to
Received in revised form
investigate the tissue properties of the knee meniscus, but the testing conditions do not
7 October 2014
replicate this complex loading regime. Biaxial testing involves loading tissue along two
Accepted 8 October 2014
different directions simultaneously, which more accurately simulates physiologic loading
Available online 19 October 2014
conditions. The purpose of this study was to report mechanical properties of meniscal
Keywords:
tissue resulting from biaxial testing, while simultaneously investigating regional variations
Meniscus
in properties. Ten left, fresh porcine joints were obtained, and the medial and lateral
Mechanical properties
menisci were harvested from each joint (twenty menisci total). Each menisci was divided
Biaxial testing
into an anterior, middle and posterior region; and three slices (femoral, deep and tibial layers) were obtained from each region. Biaxial and constrained uniaxial testing was performed on each specimen, and Young's moduli were calculated from the resulting stress strain curves. Results illustrated significant differences in regional mechanical properties, with the medial anterior (Young's modulus (E)¼11.1471.10 MPa), lateral anterior (E ¼11.5471.10 MPa) and lateral posterior (E ¼9.071.2 MPa) regions exhibiting the highest properties compared to the medial central (E¼ 5.071.22 MPa), medial posterior (E ¼4.1671.13 MPa) and lateral central (E¼ 5.671.20 MPa) regions. Differences with depth were also significant on the lateral meniscus, with the femoral (E ¼12.771.22 MPa) and tibial (E¼ 8.671.22 MPa) layers exhibiting the highest Young's moduli. This data may form the basis for future modeling of meniscal tissue, or may aid in the design of synthetic replacement alternatives. & 2014 Elsevier Ltd. All rights reserved.
n Correspondence to: School of Engineering, University of Guelph, 50 Stone Road East, Guelph Ontario, Canada, N1G 2W1. Tel.: þ1 519 824 4120x52435. E-mail address: [email protected] (K.D. Gordon).
http://dx.doi.org/10.1016/j.jmbbm.2014.10.008 1751-6161/& 2014 Elsevier Ltd. All rights reserved.
journal of the mechanical behavior of biomedical materials 41 (2015) 108 –114
1.
Introduction
The meniscus, which is composed of two C-shaped fibrocartilagenous tissues, is interposed between the articulating surfaces of the femur and tibia in the knee joint. Its functions have been described as load distribution, joint stabilization, and joint lubrication (Allen et al., 2000; Arnoczky and McDevitt, 2000; Favenesi et al., 1983; Radin et al., 1984), making the meniscus an integral component in knee biomechanics. It has been established that meniscal injuries and deterioration of this tissue can lead to osteoarthritis (Rangger et al., 1995; Ratzlaff and Liang, 2010). For this reason the standard of care in orthopedics is currently to salvage and restore damaged meniscus whenever possible (Hutchinson et al., 2014). As a result, an in-depth understanding of the indigenous structure and biomechanical functioning of the native meniscus is crucial to our progression in developing improved repair strategies and transplant alternatives. Tensile material properties of the meniscus have been studied extensively (Proctor et al., 1989; Fithian et al., 1990; Lechner et al., 2000; Goertzen et al., 1997; Skaggs et al., 1994; Tissakht and Ahmed, 1995). Although these studies vary in their methodology, including type of species, specimen dimensions and the region from which samples were extracted; one common factor is that separate uniaxial tests in the circumferential and radial directions have been performed. However, the meniscus experiences stresses simultaneously in both the circumferential and radial direction during in vivo loading-phases (Bylski-Austrow et al., 1994; Mow et al., 1992; Walker and Erkman, 1975), suggesting that a simultaneous multi-axial loading regime is more physiologically relevant. The uniaxial and biaxial tissue responses have been shown to be significantly different in other tissue types (Eilaghi et al., 2010; Gregory and Callaghan, 2011). A number of studies have looked at the biaxial tensile response of collagenous tissues (Eilaghi et al., 2010; Gregory and Callaghan, 2011; Holmes et al., 2012) but none have investigated this response in the meniscal tissue. Given the current focus on salvaging meniscal tissue and developing synthetic alternatives, a thorough investigation into the biaxial response of meniscal tissue is warranted. The distinct layers of the ultrastructure in the meniscus can be identified by the differences in the collagenous fiber alignment (Mow et al., 1992; Petersen and Tillmann, 1998). The superficial layer at the femoral and tibial surfaces consists of a smaller fibril network (bundles of approximate diameter 10 mm, fibrils of approximate diameter 35 nm) without any distinct orientation. Below this surface is the lamellar layer, identified with a larger fibril network (bundles: 20– 50 mm, fibrils: 120 nm), that has mostly randomly oriented fibers except at the periphery of the anterior and posterior segments where the fibers are radially oriented. The largest volume of the meniscus is the deep zone which has predominately circumferentially oriented fibers intermixed with radially aligned tie fibers (Mow et al., 1992; Petersen and Tillmann, 1998). Skaggs et al. (1994) found regional differences in the distribution of these radial tie fibers in medial bovine menisci with corresponding regional differences in the uniaxial tensile moduli directly correlating to the size of
109
the radial tie fiber. Tissakht and Ahmed (1995) investigated the regional and depth variations of the uniaxial tensile material properties in human meniscal tissue. With similar objectives Proctor et al. (1989) investigated bovine meniscus by examining these properties from the femoral and deep regions. The results of these previous studies suggest the mechanical response in meniscal tissue is non-homogenous. However, current finite element models (FEM) of the meniscus have modeled the meniscus as transversely isotropic without accommodating regional variation in properties (Haut Donahue et al., 2003; Bao et al., 2013; Dong et al., 2014; Mononen et al., 2013). Moreover, while little has been reported on the mechanical response of the two scaffolds currently commercially available (Spencer et al., 2012), there may be benefits to gain from synthesizing scaffolds of nonhomogenous properties. This approach to synthetic meniscal substitute design could result in better clinical outcomes following surgery. Therefore, an in-depth study of the biaxial material properties will be investigated to gain understanding of the more physiological material response of the meniscus. Furthermore, regional differences (anterior, central, and posterior) and depth variations (femoral lamellar, deep, and tibial lamellar) of porcine lateral and medial meniscus will be examined. The resulting data may be used in improve models of meniscus under load to better understand physiologic response. This may, in turn, lead to improved surgical repair method and development of synthetic alternatives and tissue engineered scaffolds.
2.
Material and methods
2.1.
Specimen preparation
Ten fresh-frozen left porcine knee joints were obtained from a local abattoir. On the day prior to dissection, the joints were thawed overnight. The lateral and medial menisci from each knee joint were carefully harvested by sharp dissection with a scalpel. Each menisci was carefully inspected for evidence of degeneration or damage from the dissection. Each menisci was then divided with a scalpel into three regions (anterior, central, and posterior) (Fig. 1). In order to maintain consistency for choosing the region, a radial center line dividing the menisci into two halves was drawn. The size of the central block was selected to be large enough to fit two 7 7 mm2 square blocks on the femoral surface with the anterior and posterior regions being adjacent to these blocks. The anterior and posterior horns were then removed. Each region was then cut into a smaller block with an approximate inner arc length of 10 mm without removing any inner or peripheral tissue. The membranes from the femoral and tibial surfaces of each block were removed. To ensure that the slices were obtained from both the femoral and tibial surfaces each block was then carefully cut circumferentially using a scalpel into three further depth regions (femoral, middle, and tibial) (Fig. 1). Finally, using a square guide, 7 7 mm2 slices from each layer of approximate thickness of 1 mm were obtained. In total, nine slices per menisci were removed. Extra care was taken to ensure that the inner and
110
journal of the mechanical behavior of biomedical materials 41 (2015) 108 –114
Fig. 1 – Each menisci was first sectioned using a scalpel into three regions (anterior, central, and posterior). Each region was then further divided into three depths (femoral, middle, and tibial), to obtain nine specimens per menisci.
the actuator along the circumferential axis while maintaining the position of the actuator along the radial axis), and lastly a constrained uniaxial test in the radial direction. All tests were performed at 1% s 1 (0.006 mm/s) for 7 s with a 7 s rest after preconditioning and after each individual uniaxial and biaxial test. This loading rate was chosen to ensure quasi-static loading conditions as outline in previous studies (Lechner et al., 2000; Tissakht and Ahmed, 1995). A final strain of 7% was chosen to ensure the linear region of the stress strain curve would be reached while incurring no damage to the tissue. Fig. 2 – A 7 7 mm2 bovine meniscus specimen mounted in the biaxial test system.
peripheral cuts of the square block were parallel to the circumferential running collagen fibers. The thickness of each slice was measured at five locations using a custom designed electrical resistant gage using a micrometer (Lee and Langdon, 1996) measuring device (resolution¼ 0.01 mm) and then averaged. The slices were kept frozen at 20710 1C until further testing. There are several studies confirming no change in mechanical properties of bone (Shaw et al., 2012) and soft tissue (Jung et al., 2011) after several (up to eight) freeze thaw cycles.
2.2.
Tensile testing
Tensile testing was performed using a biaxial test system (BioTester 5000; Cell Scale, Waterloo, ON, Canada). Custom rakes with five tines were placed into the rake base and opposing rakes were positioned 6 mm apart. Specimens were carefully mounted by aligning the circumferential fiber direction with the X-direction of the actuator and then inserting rake tines on all four sides into the specimen (Fig. 2). Any abundant saline solution floating on the specimen was removed using paper towel before carefully spreading the specimen with black graphite particles using a Q-tip. The graphite particles were applied to provide contrast for the optical strain measurement calculation outlined in Section 2.3. Specimens were preconditioned at 1% s 1 (0.006 mm/s) for 7 s for ten sinusoidal cycles. This testing protocol was found to generate repeatable results in pilot testing conducted in our lab. The specimens were then stretched biaxially (equal displacement along both axes), followed by a constrained uniaxial test in the circumferential direction (displacement of
2.3.
Data analysis
Force data was obtained from the load cell output on each actuator and a digital image correlation technique was used to quantify the strains produced during testing from images of the specimen and graphite particles during testing (Veldhuis et al., 2005). Particle tracking of the graphite particles adhered to the surface of the tissue was used to then calculate the strains. Stress (σ) was calculated by taking the force data and dividing it by the initial cross-sectional area of the specimen. The average of the five thickness measurements was multiplied by the distance between the rake tines at the beginning of a test to obtain the initial crosssectional area. For each test the stress and strain values were plotted in Microsoft Excel and the linear trend-line tool was used in the linear portion of the curve to calculate Young's modulus (Fig. 3). The linear region of the stress strain curve was defined through repeated iterations of slope calculation until a maximized linear region had been defined with a Rsquared value of 0.85 or higher. In total four separate Young's modulus (E) values were calculated for the biaxial circumferential, biaxial radial, uniaxial circumferential and uniaxial radial loading. An ANOVA analysis was performed for each group of moduli to make comparisons between regions (anterior, central, and posterior), depth (femoral, middle, tibial) and side (medial, and lateral). Identified significance differences were further investigated using least square means test. A P-value of less than 0.05 was considered significant.
3.
Results
The number of specimens from each meniscus and from different locations and depths are summarized in Table 1. In some cases testing was not successfully completed due to
111
journal of the mechanical behavior of biomedical materials 41 (2015) 108 –114
human error during slicing, resulting in less than nine specimens obtained per menisci. The average thickness for all specimens was 1.0970.27 mm. For all tested specimens the average maximum difference from the average thickness was less than 2%, with the maximum difference of 5.7%. All data was normally distributed. Fig. 4 summarizes the difference between the biaxial and constrained uniaxial tests. The circumferential modulus was significantly larger (po0.0001) than the radial modulus. The modulus values for the biaxial data were also significantly larger (po0.001) than the counterpart for the constrained uniaxial test. The results of biaxial tensile testing can be found in Figs. 5 and 6. Fig. 5 summarizes the Young's modulus value for both lateral and medial specimens at three different locations (anterior, central, and posterior) and two directions (circumferential and radial) of the tissue. For the medial meniscus, in both the circumferential and radial direction, and the lateral meniscus circumferential specimens only, the anterior moduli were significantly higher (po0.0001) compared to the central and posterior locations. Specimens from the lateral anterior region tested in the radial direction were not significantly different than the lateral middle and posterior region but followed the same trend. Significant differences were also observed between the lateral central and lateral posterior region in the circumferential direction, with the posterior region displaying a higher modulus than the central (p¼ 0.04) (Fig. 5). Biaxial data for various depths (femoral, middle, and tibial) layers is summarized in Fig. 6. For the lateral menisci all three depths were significantly different (po0.004) in the
circumferential direction with the tibial layer displaying a higher modulus (p ¼0.05) compared to the middle layer in the radial direction. No significant differences were found for the medial specimens for the three depths studied.
4.
Discussion
The aim of this study was to quantify the biaxial tensile material properties of porcine meniscus with respect to regional and depth-wise variations. Young's moduli at three different anatomical locations (anterior, central and posterior), three different depths (femoral, middle, and tibial) and two sides (medial and lateral) were reported and compared. Three separate tensile tests (biaxial, constrained uniaxial in the circumferential, and constrained uniaxial in the radial direction) were performed to obtain four separate Young's modulus values (biaxial circumferential, biaxial radial, uniaxial circumferential and uniaxial radial loading). The biaxial loading regime is thought to represent increased physiological relevance as in-vivo stresses during loading are endured in multiple directions simultaneously (Bylski-Austrow et al., 1994; Mow et al., 1992; Walker and Erkman, 1975). The overall result for the biaxial and constrained-uniaxial data in the circumferential direction was significantly higher than the radial direction. This result is in accordance with the results obtained by previous studies that conducted uniaxial testing in two directions (Proctor et al., 1989; Tissakht and Ahmed, 1995) and can be explained by the predominance of circumferentially aligned collagen fibers in the meniscal
Young's Modulus (MPa)
9 8 7 6 5 4 3 2 1 0
Biaxial - X
Biaxial - Y
Uniaxial - X
Uniaxial - Y
Fig. 4 – Comparison of the biaxial and uniaxial data. For both tests the modulus values in the Y-direction (radial) were significantly lower than the result of the same test in the X-direction (circumferential). Modulus values obtained from the constrained uniaxial test were significantly lower than the biaxial test values. Values are mean7standard error.
Fig. 3 – A representative stress–strain curve illustrating the calculation of Young's modulus (E).
Table 1 – Number of specimens tested for each location (anterior, central, and posterior), each depth (femoral, middle, and tibial), and side (lateral and medial). Total number of specimens tested was 149; 71 lateral and 78 medial. Anterior
Femoral Middle Tibial
Central
Posterior
Lateral
Medial
Lateral
Medial
Lateral
Medial
8 8 8
9 8 9
8 7 8
10 8 8
8 8 8
9 8 9
112
journal of the mechanical behavior of biomedical materials 41 (2015) 108 –114
Young's Modulus (MPa)
14 12 10 8
Circumferential (X)- Modulus Radial (Y)Modulus
6 4 2 0
Anterior
Central
Posterior Anterior
Medial
Central
Posterior
Lateral
Fig. 5 – Bi-axial Young's modulus for lateral and medial porcine test samples at three different locations (anterior, central, and posterior). Each region represents the average value using data from all depths. Values are mean7the standard error of the mean. * indicates statistical significance.
Young's Modulus (MPa)
16 14 12 10
Circumferential (X)-Modulus
8
Radial (Y)Modulus
6 4 2 0 Femoral
Middle
Medial
Tibial
Femoral
Middle
Tibial
Lateral
Fig. 6 – Bi-axial Young's modulus for lateral and medial porcine test samples at three different depths (femoral, middle, and tibial). Each depth represents the average value using data from all regions. Values are mean7the standard error of the mean. * indicates statistical significance. tissue (Mow et al., 1992; Petersen and Tillmann, 1998). The circumferential modulus values were also dependent on the anatomical location within the menisci with the anterior specimens having significantly higher values than the central and posterior regions. Similar to the results for the circumferential modulus, the radial modulus for the anterior region was significantly higher than the radial central and posterior regions for the medial menisci. Although not significant, the lateral anterior region had a higher modulus value (4.6471.22 MPa) than both the lateral central (2.2571.24 MPa) and lateral posterior (2.8671.22 MPa) regions in the radial direction. Farinaccio (1989) also reported the circumferential anterior region to be stiffer than the central and posterior in human menisci. Lechner et al. (2000) found higher modulus values in the anterior region of human menisci, but the result was not significant. Tissakht and Ahmed (1995) tested human meniscal samples and did not see any significant effect for region, and in contrast with the current study the posterior regions had slightly higher modulus values than the anterior
and central regions. Proctor et al. (1989) reported that the posterior segments of the deep regions were significantly stiffer than the anterior region in bovine meniscus, however an overall comparison between the posterior and anterior region was not provided. Finally, Fithian et al. (1990) reported that the posterior two-thirds of the medial meniscus had lower modulus values compared to the anterior and the lateral human meniscus. There is certainly variability in the results of these studies, but overall there is some consensus that the anterior and posterior regions of both the lateral and medial menisci have higher mechanical properties compared to the middle region. This is intuitive since there is likely more load applied in the anterior posterior direction during flexion and extension of the knee. This also supports the idea of regionally varying mechanical properties in a model of the meniscus. A comparison to past studies that have investigated the mechanical properties of the meniscal tissue using different experimental procedures is also enlightening. Baro et al. (2012) conducted indentation tests on medial bovine meniscus to study the variation of contact modulus with anatomical location (anterior, central, and posterior). Baro et al. (2012) found that the contact modulus was significantly higher in the anterior region and lowest at the posterior region, supporting the result of the current study with highest properties in the anterior region. Shear modulus has also been reported in bovine meniscus at three anatomical locations (Abraham et al., 2011). Although not significantly different, the anterior region had a higher shear modulus than the central region. The current study found that modulus also varied with depth for the lateral menisci. All three depths (tibial, middle, and femoral) were significantly different from each other with the femoral layer having the highest modulus and the middle layer the lowest modulus values. This may be explained due to higher amount of glycosaminoglycan (GAG), the proteoglycan responsible for water retention, in the deep zone of the meniscus (Moyer et al., 2013) which may be responsible for a lower modulus value (Sanchez-Adams et al., 2011). This may also be explained by the variation in collagen fiber orientation. The randomly oriented collagen fibers near the femoral and tibial surfaces result in stiffer mechanical properties while the predominately unidirectional fibers in the middle or deep layer result in lower mechanical properties. Tissakht and Ahmed (1995) obtained similar results for their radial specimens only, and saw similar trends for the circumferential specimens. However, their study included the approximately 100 mm thick meniscal membrane which has shown to be significantly stiffer in isolation than specimens from the surface layers with the membrane (Whipple et al., 1985). Interestingly, Baro et al. (2012) also observed an effect from the membrane when calculating the effective contact modulus. The mean value for the femoral layer was significantly higher than the deep and tibial layers and the mean tibial layer value was slightly higher than the deep layer (Baro et al., 2012). The average biaxial Young's modulus values obtained in this study (7.73 MPa circumferential and 2.80 MPa for radial specimens) are much lower than the values reported by previous relevant studies who conducted uniaxial testing
journal of the mechanical behavior of biomedical materials 41 (2015) 108 –114
only (Proctor et al., 1989; Tissakht and Ahmed, 1995; Lechner et al., 2000; Fithian et al., 1990). Tissakht and Ahmed (1995) obtained Young's modulus in the range of 90.22–102.12 MPa for circumferential and 7.82–13.04 MPa for radial human meniscal specimens. Proctor et al. (1989) reported values within a much wider range of 48.4–139 MPa for circumferential and (2.8–71.4 MPa) for radial specimens. Lechner et al (2000) reported values between 43.4 and 141.2 MPa, varying with anatomical region and sample size. We hypothesize that these variable results are arising from the different methods of sample preparation and testing. A literature review on biaxial testing of various collagenous tissues reveals results that are more harmonized with the current study. A study of the tensile material properties of the annulus fibrosis reported an average modulus value of 6.95 MPa in the direction parallel to collagen fiber alignment (Gregory and Callaghan, 2011). A comparison between the uniaxial, unconfined results (Wollensak and Spoerl, 2004) and biaxial testing results (Eilaghi et al., 2010) of the human sclera showed that the unconfined, uniaxial results were approximately 2 times higher than the biaxial data with the average biaxial modulus reported at 2.9 MPa. Other collagenous tissue studies have reported Young's modulus values in the range of 0.13– 2.76 MPa for the transverse carpal ligament under biaxial testing (Holmes et al., 2012) and between 38 MPa and 53 MPa for unconfined uniaxial testing of the entire ligament (Brett et al., 2014). These biaxial results are on the same order of magnitude as the results presented in the current study. We hypothesize that there are likely several reasons for the discrepancy between biaxial and unconfined uniaxial testing. The most obvious explanation would be the disruption of the network of collagen fibers when the specimens are prepared. The fibers that are not in-plane with the specimen dimensions are severed, and can therefore no longer offer the same resistance to applied load, lowering mechanical properties. When comparing unconfined to confined uniaxial testing, we would hypothesize that the unconfined testing state would allow the transverse and out-of-plane support fibers to strain in the direction of applied load, potentially increasing the overall resistance to load, increasing modulus. The length of the specimen and length of the collagen fibrils also likely plays a role, in that the individual collagen fibers may or may not run from grip to grip depending on the size of the specimen, also contributing to differing results. Further work is needed to explore these questions related to the preparation of samples and testing modalities of soft tissues. Currently, permanent and biodegradable synthetic meniscal scaffolds with isotropic mechanical properties are the focus of extensive research. These implants rely on the infiltration of cells to develop the anisotropic properties seen in indigenous meniscal tissues. Early clinical and experimental outcomes from these procedures are still not optimal (Sacks, 1999). This study provides evidence that varying the material properties from the anterior to the posterior regions as well as from the femoral to the tibial surface in synthetic manufactured scaffolds may lead to improved patient outcomes following surgery. The limitations of this study include the necessity to slice the specimens manually, which led to some variability in thickness and may have some impact on subsequent stress
113
calculations. Preliminary work using a vibratome also did not result in consistent slice thickness across a 7 7 mm2 slice, in fact careful dissection by hand generated more consistent results. Another limitation of this study was that quantification of collagen fiber orientation was not performed, and instead previous literature was used to determine the predominant collagen fiber direction. During mounting of the specimens for testing on the biaxial equipment, if the axis of collagen fiber orientation is not aligned with the actuator axis, shearing load can be introduced in the specimen (Sacks, 1999). It is possible that collagen fiber orientation may not have been perfectly aligned in all cases, and subsequently would impact mechanical properties calculations. The order of testing (biaxial tests conducted prior to uniaxial tests) was not randomized, and small changes in material properties may have occurred during the three minute duration of the testing due to dehydration. Finally the samples tested in this study were porcine, and thus may demonstrate different mechanical properties compared to human owing to the loading history on the meniscus tissue. The most notable difference between human and porcine anatomy is the posterior meniscal ligament, which attaches to the femur in the pig and to the tibia in a human. This variation in anatomy may result in altered loading of the meniscus, and subsequently varying material properties.
5.
Conclusions
In conclusion, this study is the first to perform biaxial testing on meniscal samples obtained from porcine knee joints. Biaxial testing provides an improved loading regime over previously conducted uniaxial testing in that more accurately represents loading seen in-vivo. The results of this study illustrate significant differences in regional mechanical properties, with the medial anterior, lateral anterior and lateral posterior regions exhibiting the highest properties. Differences with depth were also significant on the lateral meniscus, with the femoral and tibial layers exhibiting the strongest mechanical properties. This data may form the basis for future modeling of meniscal tissue, or design of synthetic replacement alternatives.
r e f e r e nc e s
Abraham, A.C., Edwards, C.R., Odegard, G.M., Donahue, T.L.J., 2011. Regional and fiber orientation dependent shear properties and anisotropy of bovine meniscus. Mech. Behav. Biomed. Mater. 4 (8), 2024–2030, http://dx.doi.org/10.1016/j. jmbbm.2011.06.022. Allen, C.R., Wong, E.K., Livesay, G.A., 2000. Importance of the medial meniscus in the anterior cruciate ligament-deficient knee. J. Orthop. Res. 18, 109–115. Arnoczky, S.P., McDevitt, C.A., 2000. The meniscus: structure, function, repair, and replacement. In: Buckwalter, J.A., Einhorn, T.A., Simon, S.R. (Eds.), Orthopaedic Basic Science: Biology and Biomechanics of the Musculoskeletal System. Rosemont, Illinois, pp. 531–545. Bao, H.R.C., Zhu, D., Gong, H., Gu, G.S., 2013. The effect of complete radial lateral meniscus posterior root tear on the knee contact
114
journal of the mechanical behavior of biomedical materials 41 (2015) 108 –114
mechanics: a finite element analysis. J. Orthop. Sci. 18 (2), 256–263, http://dx.doi.org/10.1007/s00776-012-0334-5. Baro, V.J., Bonnevie, E.D., Lai, X., Price, C., Burris, D.L., Wang, L., 2012. Functional characterization of normal and degraded bovine meniscus: rate-dependent indentation and friction studies. Bone 51 (2), 232–240, http://dx.doi.org/10.1016/j. bone.2012.03.009. Bylski-Austrow, D.I., Ciarelli, M.J., Kayner, D.C., Matthews, L.S., Goldstein, S.A.,, 1994. Displacements of the menisci under joint load: an in vitro study in human knees. J. Biomech. 27 (4), 421–431. Brett, A.W., Oliver, M.L., Agur, A.M., Edwards, A.M., Gordon, K.D., 2014. Quantification of the transverse carpal ligament elastic properties by sex and region. Clin. Biomech. 29 (6), 601–606. Dong, Y., Hu, G., Dong, Y., Hu, Y., Xu, Q., 2014. The effect of meniscal tears and resultant partial meniscectomies on the knee contact stresses: a finite element analysis. Comput. Methods Biomech. Biomed. Eng. 17 (13), 1452–1463, http://dx. doi.org/10.1080/10255842.2012.753063. Eilaghi, A., Flanagan, J.G., Tertinegg, I., Simmons, C.A., Brodland, W.G., Ethier, R.C., 2010. Biaxial mechanical testing of human sclera. J. Biomech. 43 (9), 1696–1701, http://dx.doi.org/10.1016/ j.jbiomech.2010.02.031. Farinaccio, R., 1989. An Experimental and Numerical Study of the Mechanical Response of the Knee Menisci (M. Eng. thesis). McGill University, Montreal, Quebec, Canada. Favenesi, J.A., Shaffer, J.C., Mow, V.C., 1983. Biphasic mechanical properties of the knee meniscus. Trans. Orthop. Res. Soc. 8, 57. Fithian, D.C., Kelly, M.A., Mow, V.C., 1990. Material properties and structure-function relationships in the menisci. Clin. Orthop. Relat. Res. 252, 19–31. Gregory, D.E., Callaghan, J.P., 2011. A comparison of uniaxial and biaxial mechanical properties of the annulus fibrosus: a porcine model. J. Biomech. Eng. 133 (2) (doi: 0.1115/1.4003327). Goertzen, D.J., Budney, D.R., Cinats, J.G., 1997. Methodology and apparatus to determine material properties of the knee joint meniscus. Med. Eng. Phys. 19 (5), 412–419. Haut Donahue, T.L., Hull, M.L., Rashid, M.M., Jacobs, C.R., 2003. How the stiffness of meniscal attachments and meniscal material properties affect tibio-femoral contact pressure computed using a validated finite element model of the human knee joint. J. Biomech 36 (1), 19–34. Holmes, M.W.R., Howarth, S.J., Callaghan, J.P., Keir, P.J., 2012. Biomechanical properties of the transverse carpal ligament under biaxial strain. J. Orthop. Res. 30 (5), 757–763, http://dx. doi.org/10.1002/jor.21583. Hutchinson, I.D., Moran, C.J., Potter, H.G., Warren, R.F., Rodeo, S. A., 2014. Restoration of the meniscus form and function. Am. J. Sports Med. 42 (4), 987–998, http://dx.doi.org/10.1177/ 0363546513498503. Jung, H., Vangipuram, G., Fisher, M.B., Yang, G., Hsu, S., Bianchi, J., Ronholdt, C., Woo, S.L.Y., 2011. The effects of multiple freeze– thaw cycles on the biomechanical properties of the human bone-patellar tendon-bone allograft. J. Orthop. Res. 29 (8), 1193–1198. Lechner, K., Hull, M.L., Howell, S.M., 2000. Is the circumferential tensile modulus within a human medial meniscus affected by the test sample location and cross-sectional area?. J. Orthop. Res. 18 (6), 945–951. Lee, M.J., Langdon, S.E., 1996. Thickness measurement of soft tissue biomaterials: a comparison of five methods. J. Biomech. 29 (6), 829–832.
Mononen, M.E., Jurvelin, J.S., Korhonen, R.K., 2013. Effects of radial tears and partial meniscectomy of lateral meniscus on the knee joint mechanics during the stance phase of the gait cycle—a 3D finite element study. J. Orthop. Res. 31 (8), 1208–1217, http://dx.doi.org/10.1002/jor.22358. Mow, V.C., Arnoczky, S., Jackson, D., 1992. Structure and function relationships of the meniscus in the knee, Knee Meniscus: Basic and Clinical Foundations. Raven Press, New York37–58. Moyer, J.T., Priest, R., Bouman, T., Abraham, A.C., Donahue, T.L., 2013. Indentation properties and glycosaminoglycan content of human menisci in the deep zone. Acta Biomater. 9 (5), 6624–6629, http://dx.doi.org/10.1016/j.actbio.2012.12.033. Petersen, W., Tillmann, B., 1998. Collagenous fibril texture of the human knee joint menisci. Anat. Embryol. 197 (4), 317–324. Proctor, C.S., Schmidt, M.B., Whipple, R.R., Kelly, M.A., Mow, V.C., 1989. Material properties of the normal medial bovine meniscus. J. Orthop. Res. 7, 771–782. Radin, E.L., de Lamotte, F., Maquet, P., 1984. Role of the meniscus in the distribution of stress in the knee. Clin. Orthop. Relat. Res. 185, 290–294. Rangger, C., Klestil, T., Gloetzer, W., Kemmler, G., Benedetto, L.P., 1995. Osteoarthritis after arthroscopic partial meniscectomy. Am. J. Sports Med. 23, 240–244. Ratzlaff, C.R., Liang, M.H., 2010. New developments in osteoarthritis. Prevention of injury-related knee osteoarthritis: opportunities for the primary and secondary prevention of knee osteoarthritis. Arthritis Res. Ther. 12 (4), 215–220. Sacks, M., 1999. A method for planar biaxial testing that includes in-plane shear. J. Biomech. Eng. 121 (5), 551–555. Sanchez-Adams, J., Willard, V.P., Athanasiou, K.A., 2011. Regional variation in the mechanical role of the knee meniscus glycosaminoglycans, J. App. Phys. 111 (6), 1590–1596, http://dx. doi.org/10.1152/japplphysiol.00848.2011. Skaggs, D.L., Warden, W.H., Mow, V.C., 1994. Radial tie fibers influence the tensile properties of the bovine medial meniscus. J. Orthop. Res. 12, 176–185. Shaw, J.M., Hunter, S.A., Gayton, J.C., Boivin, G.P., Prayson, M.J., 2012. Repeated freeze-thaw cycles do not alter biomechanics properties of fibular allograft bone. Clin. Orthop. Relat. Res. 470 (3), 937–943. Spencer, S.J., Saithna, A., Carmont, M.R., Dhillon, M.S., Thompson, P., Spalding, T., 2012. Meniscal scaffolds: early experience and review of the literature. Knee 19, 760–765, http://dx.doi.org/10.1016/j.knee.2012.01.006. Tissakht, M., Ahmed, A.M., 1995. Tensile stress–strain characteristics of the human meniscal material. J. Biomech. 28 (4), 411–422. Veldhuis, J.H., Brodland, G.W., Wiebe, C.J., Bootsma, G.J., 2005. Multiview robotic microscope reveals the in-plane kinematics of amphibian neurulation. Ann. Biomed. Eng. 33 (6), 821–828, http: //dx.doi.org/10.1007/s10439-005-3309-2. Walker, P.S., Erkman, M.J., 1975. The role of the menisci in force transmission across the knee. Clin. Orthop. Relat. Res. 109, 184–192. Whipple, R.R., Wirth, C.R., Mow, V.C., 1985. Anisotropic and zonal variations in the tensile properties of the meniscus. Trans. Orthop. Res. Soc. 10, 367. Wollensak, G., Spoerl, E., 2004. Collagen crosslinking of human and porcine sclera. J. Cataract Refract. Surg. 30 (3), 689–695, http://dx.doi.org/10.1016/j.jcrs.2003.11.032.
