Materials Science and Engineering C 39 (2014) 100–104
Contents lists available at ScienceDirect
Materials Science and Engineering C journal homepage: www.elsevier.com/locate/msec
Strain rate effects on the mechanical properties and fracture mode of skeletal muscle Michael Shapiro a, Nick Tovar a, Daniel Yoo a, Micheal Sobieraj b, Nikhil Gupta c, Ryan C. Branski d, Paulo G. Coelho a,⁎ a
Biomaterials and Biomimetics, New York University College of Dentistry, United States Orthopedic Surgery, Hospital for Joint Diseases, United States Mechanical and Aerospace Engineering, NYU-Poly, United States d Dept of Otolaryngology, New York University School of Medicine, United States b c
a r t i c l e
i n f o
Article history: Received 27 October 2013 Accepted 17 February 2014 Available online 22 February 2014 Keywords: Muscle Biomechanical Testing Fracture Dog
a b s t r a c t The present study aimed to characterize the mechanical response of beagle sartorius muscle fibers under strain rates that increase logarithmically (0.1 mm/min, 1 mm/min and 10 mm/min), and provide an analysis of the fracture patterns of these tissues via scanning electron microscopy (SEM). Muscle tissue from dogs' sartorius was excised and test specimens were sectioned with a lancet into sections with nominal length, width, and thickness of 7, 2.5 and 0.6 mm, respectively. Trimming of the tissue was done so that the loading would be parallel to the direction of the muscle fiber. Samples were immediately tested following excision and failures were observed under the SEM. No statistically significant difference was observed in strength between the 0.1 mm/min (2.560 ± 0.37 MPa) and the 1 mm/min (2.702 ± 0.55 MPa) groups. However, the 10 mm/min group (1.545 ± 0.50 MPa) had a statistically significant lower strength than both the 1 mm/min group and the 0.1 mm/min group with p b 0.01 in both cases. At the 0.1 mm/min rate the primary fracture mechanism was that of a shear mode failure of the endomysium with a significant relative motion between fibers. At 1 mm/min this continues to be the predominant failure mode. At the 10 mm/min strain rate there is a significant change in the fracture pattern relative to other strain rates, where little to no evidence of endomysial shear failure nor of significant motion between fibers was detected. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Soft tissue injuries of the musculoskeletal system can be challenging to treat. A thorough understanding of the underlying mechanical behavior of these tissues is necessary to understand the pathology and for adequate treatment and surgical intervention if necessary. Much work has been done investigating the mechanical properties and the repair of ligaments (i.e. ACL/PCL) and of tendons (rotator cuff). Additionally there can be a significant injury to the muscle in isolation and further characterization of muscle biomechanical behavior and fracture mode is warranted. Muscle tears and partial muscle tears can occur as a consequence of overexertion of the muscle [1,2], steroid use [3], and trauma [4]. These injuries are often the result of eccentric loading and can take weeks to months to heal with loss of productivity in the short term and possible long term decrease of function [5]. Understanding of the mechanical behavior of this tissue can help us to better formulate preventative ⁎ Corresponding author at: Dept. of Biomaterials and Biomimetics, 345 E 24th st 804s, New York, NY 10010, United States. Tel.: +1 212 998 9214. E-mail address: [email protected] (P.G. Coelho).
http://dx.doi.org/10.1016/j.msec.2014.02.032 0928-4931/© 2014 Elsevier B.V. All rights reserved.
measures and to potentially develop interventional modalities in high demand individuals such as professional athletes and military personnel, where prolonged removal from activity may have significant professional consequences. An understanding of the dynamic mechanical properties of biological muscle tissue is needed to determine the response of the human body to traumatic loading. Skeletal muscle tissue is a complex fiber-oriented structure composed of approximately 80% water, 3% fat and 10% collagenous tissues [6]. It can, therefore, be expected to display similar mechanical properties resulting from interactions between these constituents as shown for soft tissues with similar structures, such as ligaments and tendons [7,8]. An understanding of the mechanism of muscle failure can help in understanding the damage mechanisms, designing physical therapy routines, surgical repair techniques, and providing key data for informed design rationale in developing medical devices or protective systems [9–11]. However, this failure mechanism is not well understood for skeletal muscle tissues, especially when tensile loading is considered. Palevski et al. [12] have noted the sparsity on the viscoelastic properties of skeletal muscle. The published data consists of a few compression and tensile tests, but they lack micrographic explanations as to the mechanism of muscle failure. Bosboom et al. [13]
M. Shapiro et al. / Materials Science and Engineering C 39 (2014) 100–104
conducted compression studies on the tibialis anterior muscles of rats and developed a method to determine the passive transverse properties of skeletal muscle. Best et al. [14] performed live tensile tests on the tibialis anterior and extensor digitorum longus muscles of rabbits and established that the quasi-linear model is the most reliable way to characterize the structural responses of muscle. Zheng et al. [15] conducted ultrasound indentation experiments on human upper and lower limb soft tissues at differing indentation rates (0.75 mm/s to 7.5 mm/s) and found that the extracted young's moduli of limb soft tissues under similar conditions were repeatable. Van Loocke et al. [16] investigated the elastic properties of passive skeletal muscle tissue by means of quasi-static uniaxial unconfined compression tests performed on fresh porcine muscle tissue in vitro. While the results indicated that muscle elastic behavior is nonlinear and transversely isotropic, the study did not detail the fracture pattern and fracture mechanism of the muscle fibers. There is little published literature that provides micrographic analysis of the fracture properties of muscle tissue. Carroll et al. [17] investigated the micrographic effects of tension on bovine semitendinosus muscle. They reported that stress applied parallel to the fiber axis resulted in the initial rupture of the muscle fiber-endomysium sheath, while perpendicular stress caused the initial rupture at the endomysium– perimysium junction without disturbing the muscle fibers. However, that study did not consider the effects of various strain rates on tension, nor did it consider the modes of muscle fracture, both of which are detailed in the present study. The present study aimed to characterize the mechanical response of beagle sartorius muscle fibers under strain rates that increase logarithmically (0.1 mm/min, 1 mm/min and 10 mm/min), and provide an analysis of the fracture patterns of these tissues via scanning electron microscopy (SEM).
101
multiple two-tailed heteroscedastic t-tests with statistical significance set to α = 0.05. The specimens were then examined via SEM for the characterization of failure mode.
2.3. Freeze fracture Untested samples were freeze-fractured to compare the effects of tensile forces and failure modes due to absolute abrupt fracture on muscle tissue. Untested muscle samples that had previously been placed in a liquid ethanol environment were placed in a liquid nitrogen bath for 75 s, and then fractured at the edges using two pairs of tweezers. The frozen segments were returned to liquid ethanol to thaw before drying at room temperature.
2.4. Characterization of failure mode A Hitachi S-3500 N variable pressure SEM (Hitachi Ltd, Tokyo, Japan) was used to analyze structural damage caused by the tensile forces. Samples were sputter coated with gold for 60 s prior to SEM examination.
2. Materials and methods 2.1. Specimen preparation and test materials All of the experiments were conducted within the biological ethics protocol of NYU and NYU-Poly. Five beagles were euthanized for orthopedic bone related research under the bioethics approval using an overdose of anesthesia; muscle tissue from the sartorius was excised and stored in a 70% ethanol solution at 5–7 °C. The test specimens were sectioned with a lancet into sections with nominal length, width, and thickness of 7, 2.5 and 0.6 mm, respectively. Trimming of the tissue was done so that the loading would be parallel to the direction of the muscle fiber. The initial cross-sectional area was determined by averaging three measurements carried out in correspondence with the central region of the specimen [18]. Samples were immediately tested following excision and then stored in 70% ethanol prior to scanning electron microscopy (SEM). 2.2. Microtensile experimental setup The machine employed for testing was an Instron 5566 universal test system (Instron, Norwood, MA, USA) equipped with an Intsron 100 N load cell. The specimen ends were bonded to the plates of the machine with cyanoacrylate glue (Scotch Super Glue AD117, 3 M Co., St. Paul, MN, USA). To preserve integrity of the muscle, the muscle strips were moistened with distilled water prior to bonding and throughout the entire test procedure. Tests were conducted at room temperature (25 °C). Twenty specimens were tested at three different strain rates each, for a total of 60 specimens. Three logarithmically increasing loading rates of 0.1, 1.0, and 10 mm/min were selected for testing. Load–displacement data was obtained using Bluehill® 2.0 software (Instron, Norwood, MA, USA). Tests were conducted until the specimen completely failed. Improperly fractured specimens, such as those fractured at the grips, were excluded from the study. Statistical analysis was performed by
Fig. 1. Representative engineering stress–strain curves of beagle sartorius muscle tested at different strain rates. (a) Initial region of the curves to show the near linear region upon immediate loading, yield, ultimate stress and the initiation of gross failure of the specimen; (b) representative curves in their entirety through complete gross failure.
102
M. Shapiro et al. / Materials Science and Engineering C 39 (2014) 100–104
Table 1 Strength values of beagle muscles at different strain rates (mean ± SD). The 10 mm/min group had a lower strength than both the 1 mm/min group and the 0.1 mm/min group (p b 0.01 in both cases). There was no statistically significant difference between the 1 mm/min and the 0.1 mm/min groups (the gray shading indicates that there was no difference detected in the groups).
The stress–strain behavior of the muscle was not unexpected. At all three strain rates the specimens showed an initial region that was linear
in nature, followed by evidence of yielding (deviation from linearity), reaching an eventual maximum load followed by slow complete failure as a greater percentage of muscle fibers failed to complete fracture of the total specimen (Fig. 1). There was no statistically significant difference in strength between the 0.1 mm/min (2.560 ± 0.37 MPa) and the 1 mm/min (2.702 ± 0.55 MPa) groups (Table 1). However, the 10 mm/min group (1.545 ± 0.50 MPa) had a statistically significant lower strength than both the 1 mm/min group and the 0.1 mm/min group with p b 0.01 in both cases. SEM fractographs were compared to images of intact beagle sartorius in order to elucidate the fracture micromechanisms at the different strain rates (Fig. 2). At the 0.1 mm/min rate the primary fracture mechanism was that of a shear mode failure of the endomysium with a significant relative motion between fibers (Fig. 3). At 1 mm/min this continues to be the predominant failure mode (Fig. 4). At the 10 mm/min strain rate there is a significant change in the fracture pattern (Fig. 5). There is little to no evidence of endomysial shear failure nor of a significant motion between fibers. The overwhelming majority of the fracture appears to be one of fast fracture across the individual fibers. The freeze fracture specimens (Fig. 6), also exhibit a similar pattern, with no evidence of shearing or motion between fibers.
Fig. 2. SEM Micrographs of untested beagle sartorius muscle: (a) 100× magnification, depicting fasciculi (black arrows); (b) 500× magnification. Individual muscle fibers are indicated (white arrows), and the endomysium (gray arrows) is indicated as well. Characteristic banding of fibers is evident even at this low magnification (black arrow).
Fig. 3. SEM fractograph of beagle sartorius muscle specimen tested at 0.1 mm/min: (a) fracture zone is indicated with white arrows: failure mode at this strain rate is mainly by shearing of endomysium (gray arrow); (b) higher magnification of the same specimen depicting shearing (white arrow) and cross-slipping (gray arrow) of endomysium.
Strain rate (mm/min)
Strength (MPa)
10
1.545 ± 0.50
1
2.702 ± 0.55
0.1
2.560 ± 0.37
3. Results
M. Shapiro et al. / Materials Science and Engineering C 39 (2014) 100–104
Fig. 4. SEM fractograph of beagle sartorius tested at 1 mm/min: (a) overview of fracture zone of this specimen; and (b) higher magnification of same specimen with endomysial motion relative to fibrils (gray arrow) and shearing (white arrow) highlighted.
103
4. Discussion
Fig. 5. SEM fractograph of beagle sartorius tested at 10 mm/min: (a) failure mode of this strain rate is mostly via abrupt fracture of muscle fibers with little shearing (gray arrow) and their motion in relation to the endomysium (white arrow); (b) higher magnification showing that shearing is absent with fracture of individual muscle fibers highlighted (white arrows); and (c) higher magnification with enhanced view of individual muscle fibers highlighted (white arrows).
The failure features for the skeletal muscle specimens tested at logarithmically increasing strain rates showed failure mode trends as strain rate was increased. SEM analysis indicated that tensile failure mode evolves as the strain rate is increased. At the two lower strain rates failure is dominated by endomysium shearing. This is expected since the endomysium has been shown to be the essential series-elastic element in muscle fiber [19] and, therefore, contributes greatly to the mechanical properties of skeletal muscle [20,21]. At the higher strain rate the mode switches to one of bulk failure of the fibers, with little evidence of shear or motion between fibers. Several reports have shown the endomysium as a random network of collagen fibrils [21–23], and Purslow [24] has shown that the arrangement of collagen fibrils in the endomysium affects the tensile stiffness parallel to the axis of muscle fibers. The endomysial network is composed of wavy collagen fibrils arranged in a near-random planar array [24], which enables the endomysial sheets of relatively inextensible collagen fibers [25] to adapt to the length changes in skeletal muscle by aligning the endomysial fibers along the edges of the muscle [20,26]. Thus, the regular wave pattern of the endomysium is well-suited for transmitting forces as muscle fibers contract and expand, while simultaneously avoiding severe strain in any of its regions [27]. As the muscle
tissue is elongated the randomly oriented fibers of the endomysium become aligned in the direction of the applied stretch [28]. As tension continued, the fibers of the endomysium become fully aligned (linear region). Increased stress beyond this point (plastic deformation) will lead to rupture [17], with rupture planes usually occurring along the path of the weaker structures [29]. At slow strain rates (0.1 mm/min), the muscles endured a longer duration of tensile force, resulting in total shear failure of the endomysium. Such slow tensile strain caused the muscle fibers to stretch out of the endomysium and with a significant motion relative to one another prior to fracture. This observation corroborates Trotter's assertion that tension is transmitted between adjacent myofibers through the endomysium [30]. These types of failure likely result in damage accumulation in the myofibrils and muscular units. At a strain rate of 1 mm/min, the muscle specimens demonstrated subtle differences relative to samples tested at 0.1 mm/min, demonstrating less shearing. Nonetheless, the fracture analysis did suggest that the primary mode of failure is tearing of the fasciculi as they move relative to one another with a significant endomysial shearing present. In addition, the fracture pattern suggests that less time was allowed for microstructural rearrangement in response to the applied load. Specimens tested at rates
104
M. Shapiro et al. / Materials Science and Engineering C 39 (2014) 100–104
data suggests that muscle fibrils had reached substantial endomysial alignment prior to the failure point at the two lower strain rates; the lower strength value noted at 10 mm/min suggests that a lower degree of structural alignment was achieved prior to the onset of fracture. Lastly, it should be noted that the structural alignment arguments made here are based load–displacement curves and on post-mechanical testing fracture analysis. Microscopic evaluation during mechanical testing or microscopic analysis of specimens tested only to distinct regions of the stress–strain curves (not necessarily to complete failure) would further elucidate if the current explanation for the rate dependent fracture behavior is correct. This type of analysis will be conducted in future work. Conflict of interest statement The authors declare that they have no conflict of interest.
Fig. 6. SEM fractograph of beagle sartorius freeze fractured in liquid Nitrogen: no relative motion between endomysium and fibrils is visible nor is shearing.
of 10 mm/min resembled the untested freeze fractured sample, which implies a strong rate dependence of failure mode in the beagle sartorius muscle. At a higher strain rate less time is allowed for hierarchical anatomic structure rearrangement prior to failure. Therefore, a minimal amount of shearing of the endomysium was visible in the micrographs, while the primary mode of failure in these specimens was an abrupt rupture of the fasciculi and muscle fibers. Relative to the two other rates, 10 mm/min fracture demonstrated lack of shearing of the endomysium and the presence of multiple abrupt ruptures of the muscle fibers. The muscle displays a “stair-step” fracture appearance, indicating that the muscle rupture can be found through and around the fibers [29]. These results seem to confirm that the endomysium may act as a mechanical coupling between skeletal muscle fibers [31], and that a necessary amount of time is required for the endomysial fibers to properly align in order to achieve maximal tension. Overall, the fracture behavior as a function of strain rate presented a behavior consistent with time–temperature equivalence in polymeric materials [32]. According to the mechanical testing results within the limited amount of strain rates applied, the “optimal” strength of the muscle tissue was visible at 1 mm/min, and this is likely the result of the time allowed for alignment of the muscle fibers prior to fracture. The specimen tested at 10 mm/min presented with lower strength values, suggesting that tensile forces at this speed do not allow for the optimal alignment of the muscle fibers prior to fracture. The stress–strain curves show regions of linear behavior followed by yield and ultimately sequential failure of the muscle. The differences in failure features explained for the various strain rates displayed what we speculate to be the aligning of endomysium and fasciculi, and optimal alignment with subsequent failure is indicated by the peak load. This
References [1] G.M. Kerkhoffs, N. van Es, T. Wieldraaijer, I.N. Sierevelt, J. Ekstrand, C.N. van Dijk, Knee Surg. Sports Traumatol. Arthrosc. 21 (2) (2013) 500–509, http://dx.doi.org/ 10.1007/s00167-012-2055-x [Epub 2012 May 24]. [2] U.M. Kujala, S. Orava, M. Jarvinen, Sports Med. 23 (1997) 397–404. [3] W.J. Rijnberg, B. van Linge, Arch. Orthop. Trauma Surg. 112 (1993) 104–105. [4] T. Marom, D. Itskoviz, S. Kutikov, J. Naftal, I. OStfeld, J. R. Army Med. Corps 155 (2009) 24–26. [5] M. MIller, S. Thomspon, J. Hart, Review of Orthopedics, Sixth edition Saunders, Philadelphia, 2012. [6] M. Van Loocke, C.G. Lyons, C.K. Simms, J. Biomech. 41 (2008) 1555–1566. [7] S.D. Abramowitch, S.L. Woo, T.D. Clineff, R.E. Debski, Ann. Biomed. Eng. 32 (2004) 329–335. [8] J.A. Weiss, J.C. Gardiner, C. Bonifasi-Lista, J. Biomech. 35 (2002) 943–950. [9] D.S. Cronin, Proceedings of the Eudem2-Scot Conference, 2003. [10] Y.C. Deng, W. Kong, H. Ho, SAE Technical Paper, 01, 1999, pp. 165–172. [11] N. Gupta, K. Cho, Adv. Mater. Process. 168 (2010) 32–33. [12] A. Palevski, I. Glaich, S. Portnoy, E. Linder-Ganz, A. Gefen, J. Biomech. Eng. 128 (2006) 782–787. [13] E.M. Bosboom, M.K. Hesselink, C.W. Oomens, C.V. Bouten, M.R. Drost, F.P. Baaijens, J. Biomech. 34 (2001) 1365–1368. [14] T.M. Best, J. McElhaney, W.E. Garrett Jr., B.S. Myers, J. Biomech. 27 (1994) 413–419. [15] Y. Zheng, A.F. Mak, B. Lue, J. Rehabil. Res. Dev. 36 (1999) 71–85. [16] M. Van Loocke, C.G. Lyons, C.K. Simms, J. Biomech. 39 (2006) 2999–3009. [17] R.J. Carroll, F.P. Rorer, S.B. Jones, J.R. Cavanaugh, J. Food Sci. 43 (1978) 1181–1187. [18] F. Velardi, F. Fraternali, M. Angelillo, Biomech. Model. Mechanobiol. 5 (2006) 53–61. [19] E. Otten, Exerc. Sport Sci. Rev. 16 (1988) 89–137. [20] R.W.D. Rowe, J. Food Technol. 9 (1974) 501–508. [21] J.A. Trotter, P.P. Purslow, J. Morphol. 212 (1992) 109–122. [22] T.K. Borg, J.B. Caulfield, Tissue Cell 12 (1980) 197–207. [23] T. Nishimura, A. Hattori, K. Takahashi, Acta Anat. 151 (1994) 250–257. [24] P.P. Purslow, J.A. Trotter, J. Muscle Res. Cell Motil. 15 (1994) 299–308. [25] A.J. Bailey, J. Sci. Food Agric. 23 (1972) 995. [26] A. Viidik, Int. Rev. Connect. Tissue Res. 6 (1973) 127–215. [27] T. Nishimura, K. Ojima, A. Liu, A. Hattori, K. Takahashi, Tissue Cell 28 (1996) 527–536. [28] D.W. Stanley, L.M. McKnight, W.G.S. Hines, W.R. Usborne, J.M. Deman, J. Texture Stud. 3 (1972) 51–68. [29] S.B. Jones, R.J. Carroll, J.R. Cavanaugh, J. Food Sci. 42 (1977) 125–131. [30] J.A. Trotter, J. Morphol. 206 (1990) 351–361. [31] S.F. Street, J. Cell. Physiol. 114 (1983) 346–364. [32] B. Erman, J. Mark, Structures and Properties of Rubberlike Networks, Oxford University Press, 1997.
Contents lists available at ScienceDirect
Materials Science and Engineering C journal homepage: www.elsevier.com/locate/msec
Strain rate effects on the mechanical properties and fracture mode of skeletal muscle Michael Shapiro a, Nick Tovar a, Daniel Yoo a, Micheal Sobieraj b, Nikhil Gupta c, Ryan C. Branski d, Paulo G. Coelho a,⁎ a
Biomaterials and Biomimetics, New York University College of Dentistry, United States Orthopedic Surgery, Hospital for Joint Diseases, United States Mechanical and Aerospace Engineering, NYU-Poly, United States d Dept of Otolaryngology, New York University School of Medicine, United States b c
a r t i c l e
i n f o
Article history: Received 27 October 2013 Accepted 17 February 2014 Available online 22 February 2014 Keywords: Muscle Biomechanical Testing Fracture Dog
a b s t r a c t The present study aimed to characterize the mechanical response of beagle sartorius muscle fibers under strain rates that increase logarithmically (0.1 mm/min, 1 mm/min and 10 mm/min), and provide an analysis of the fracture patterns of these tissues via scanning electron microscopy (SEM). Muscle tissue from dogs' sartorius was excised and test specimens were sectioned with a lancet into sections with nominal length, width, and thickness of 7, 2.5 and 0.6 mm, respectively. Trimming of the tissue was done so that the loading would be parallel to the direction of the muscle fiber. Samples were immediately tested following excision and failures were observed under the SEM. No statistically significant difference was observed in strength between the 0.1 mm/min (2.560 ± 0.37 MPa) and the 1 mm/min (2.702 ± 0.55 MPa) groups. However, the 10 mm/min group (1.545 ± 0.50 MPa) had a statistically significant lower strength than both the 1 mm/min group and the 0.1 mm/min group with p b 0.01 in both cases. At the 0.1 mm/min rate the primary fracture mechanism was that of a shear mode failure of the endomysium with a significant relative motion between fibers. At 1 mm/min this continues to be the predominant failure mode. At the 10 mm/min strain rate there is a significant change in the fracture pattern relative to other strain rates, where little to no evidence of endomysial shear failure nor of significant motion between fibers was detected. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Soft tissue injuries of the musculoskeletal system can be challenging to treat. A thorough understanding of the underlying mechanical behavior of these tissues is necessary to understand the pathology and for adequate treatment and surgical intervention if necessary. Much work has been done investigating the mechanical properties and the repair of ligaments (i.e. ACL/PCL) and of tendons (rotator cuff). Additionally there can be a significant injury to the muscle in isolation and further characterization of muscle biomechanical behavior and fracture mode is warranted. Muscle tears and partial muscle tears can occur as a consequence of overexertion of the muscle [1,2], steroid use [3], and trauma [4]. These injuries are often the result of eccentric loading and can take weeks to months to heal with loss of productivity in the short term and possible long term decrease of function [5]. Understanding of the mechanical behavior of this tissue can help us to better formulate preventative ⁎ Corresponding author at: Dept. of Biomaterials and Biomimetics, 345 E 24th st 804s, New York, NY 10010, United States. Tel.: +1 212 998 9214. E-mail address: [email protected] (P.G. Coelho).
http://dx.doi.org/10.1016/j.msec.2014.02.032 0928-4931/© 2014 Elsevier B.V. All rights reserved.
measures and to potentially develop interventional modalities in high demand individuals such as professional athletes and military personnel, where prolonged removal from activity may have significant professional consequences. An understanding of the dynamic mechanical properties of biological muscle tissue is needed to determine the response of the human body to traumatic loading. Skeletal muscle tissue is a complex fiber-oriented structure composed of approximately 80% water, 3% fat and 10% collagenous tissues [6]. It can, therefore, be expected to display similar mechanical properties resulting from interactions between these constituents as shown for soft tissues with similar structures, such as ligaments and tendons [7,8]. An understanding of the mechanism of muscle failure can help in understanding the damage mechanisms, designing physical therapy routines, surgical repair techniques, and providing key data for informed design rationale in developing medical devices or protective systems [9–11]. However, this failure mechanism is not well understood for skeletal muscle tissues, especially when tensile loading is considered. Palevski et al. [12] have noted the sparsity on the viscoelastic properties of skeletal muscle. The published data consists of a few compression and tensile tests, but they lack micrographic explanations as to the mechanism of muscle failure. Bosboom et al. [13]
M. Shapiro et al. / Materials Science and Engineering C 39 (2014) 100–104
conducted compression studies on the tibialis anterior muscles of rats and developed a method to determine the passive transverse properties of skeletal muscle. Best et al. [14] performed live tensile tests on the tibialis anterior and extensor digitorum longus muscles of rabbits and established that the quasi-linear model is the most reliable way to characterize the structural responses of muscle. Zheng et al. [15] conducted ultrasound indentation experiments on human upper and lower limb soft tissues at differing indentation rates (0.75 mm/s to 7.5 mm/s) and found that the extracted young's moduli of limb soft tissues under similar conditions were repeatable. Van Loocke et al. [16] investigated the elastic properties of passive skeletal muscle tissue by means of quasi-static uniaxial unconfined compression tests performed on fresh porcine muscle tissue in vitro. While the results indicated that muscle elastic behavior is nonlinear and transversely isotropic, the study did not detail the fracture pattern and fracture mechanism of the muscle fibers. There is little published literature that provides micrographic analysis of the fracture properties of muscle tissue. Carroll et al. [17] investigated the micrographic effects of tension on bovine semitendinosus muscle. They reported that stress applied parallel to the fiber axis resulted in the initial rupture of the muscle fiber-endomysium sheath, while perpendicular stress caused the initial rupture at the endomysium– perimysium junction without disturbing the muscle fibers. However, that study did not consider the effects of various strain rates on tension, nor did it consider the modes of muscle fracture, both of which are detailed in the present study. The present study aimed to characterize the mechanical response of beagle sartorius muscle fibers under strain rates that increase logarithmically (0.1 mm/min, 1 mm/min and 10 mm/min), and provide an analysis of the fracture patterns of these tissues via scanning electron microscopy (SEM).
101
multiple two-tailed heteroscedastic t-tests with statistical significance set to α = 0.05. The specimens were then examined via SEM for the characterization of failure mode.
2.3. Freeze fracture Untested samples were freeze-fractured to compare the effects of tensile forces and failure modes due to absolute abrupt fracture on muscle tissue. Untested muscle samples that had previously been placed in a liquid ethanol environment were placed in a liquid nitrogen bath for 75 s, and then fractured at the edges using two pairs of tweezers. The frozen segments were returned to liquid ethanol to thaw before drying at room temperature.
2.4. Characterization of failure mode A Hitachi S-3500 N variable pressure SEM (Hitachi Ltd, Tokyo, Japan) was used to analyze structural damage caused by the tensile forces. Samples were sputter coated with gold for 60 s prior to SEM examination.
2. Materials and methods 2.1. Specimen preparation and test materials All of the experiments were conducted within the biological ethics protocol of NYU and NYU-Poly. Five beagles were euthanized for orthopedic bone related research under the bioethics approval using an overdose of anesthesia; muscle tissue from the sartorius was excised and stored in a 70% ethanol solution at 5–7 °C. The test specimens were sectioned with a lancet into sections with nominal length, width, and thickness of 7, 2.5 and 0.6 mm, respectively. Trimming of the tissue was done so that the loading would be parallel to the direction of the muscle fiber. The initial cross-sectional area was determined by averaging three measurements carried out in correspondence with the central region of the specimen [18]. Samples were immediately tested following excision and then stored in 70% ethanol prior to scanning electron microscopy (SEM). 2.2. Microtensile experimental setup The machine employed for testing was an Instron 5566 universal test system (Instron, Norwood, MA, USA) equipped with an Intsron 100 N load cell. The specimen ends were bonded to the plates of the machine with cyanoacrylate glue (Scotch Super Glue AD117, 3 M Co., St. Paul, MN, USA). To preserve integrity of the muscle, the muscle strips were moistened with distilled water prior to bonding and throughout the entire test procedure. Tests were conducted at room temperature (25 °C). Twenty specimens were tested at three different strain rates each, for a total of 60 specimens. Three logarithmically increasing loading rates of 0.1, 1.0, and 10 mm/min were selected for testing. Load–displacement data was obtained using Bluehill® 2.0 software (Instron, Norwood, MA, USA). Tests were conducted until the specimen completely failed. Improperly fractured specimens, such as those fractured at the grips, were excluded from the study. Statistical analysis was performed by
Fig. 1. Representative engineering stress–strain curves of beagle sartorius muscle tested at different strain rates. (a) Initial region of the curves to show the near linear region upon immediate loading, yield, ultimate stress and the initiation of gross failure of the specimen; (b) representative curves in their entirety through complete gross failure.
102
M. Shapiro et al. / Materials Science and Engineering C 39 (2014) 100–104
Table 1 Strength values of beagle muscles at different strain rates (mean ± SD). The 10 mm/min group had a lower strength than both the 1 mm/min group and the 0.1 mm/min group (p b 0.01 in both cases). There was no statistically significant difference between the 1 mm/min and the 0.1 mm/min groups (the gray shading indicates that there was no difference detected in the groups).
The stress–strain behavior of the muscle was not unexpected. At all three strain rates the specimens showed an initial region that was linear
in nature, followed by evidence of yielding (deviation from linearity), reaching an eventual maximum load followed by slow complete failure as a greater percentage of muscle fibers failed to complete fracture of the total specimen (Fig. 1). There was no statistically significant difference in strength between the 0.1 mm/min (2.560 ± 0.37 MPa) and the 1 mm/min (2.702 ± 0.55 MPa) groups (Table 1). However, the 10 mm/min group (1.545 ± 0.50 MPa) had a statistically significant lower strength than both the 1 mm/min group and the 0.1 mm/min group with p b 0.01 in both cases. SEM fractographs were compared to images of intact beagle sartorius in order to elucidate the fracture micromechanisms at the different strain rates (Fig. 2). At the 0.1 mm/min rate the primary fracture mechanism was that of a shear mode failure of the endomysium with a significant relative motion between fibers (Fig. 3). At 1 mm/min this continues to be the predominant failure mode (Fig. 4). At the 10 mm/min strain rate there is a significant change in the fracture pattern (Fig. 5). There is little to no evidence of endomysial shear failure nor of a significant motion between fibers. The overwhelming majority of the fracture appears to be one of fast fracture across the individual fibers. The freeze fracture specimens (Fig. 6), also exhibit a similar pattern, with no evidence of shearing or motion between fibers.
Fig. 2. SEM Micrographs of untested beagle sartorius muscle: (a) 100× magnification, depicting fasciculi (black arrows); (b) 500× magnification. Individual muscle fibers are indicated (white arrows), and the endomysium (gray arrows) is indicated as well. Characteristic banding of fibers is evident even at this low magnification (black arrow).
Fig. 3. SEM fractograph of beagle sartorius muscle specimen tested at 0.1 mm/min: (a) fracture zone is indicated with white arrows: failure mode at this strain rate is mainly by shearing of endomysium (gray arrow); (b) higher magnification of the same specimen depicting shearing (white arrow) and cross-slipping (gray arrow) of endomysium.
Strain rate (mm/min)
Strength (MPa)
10
1.545 ± 0.50
1
2.702 ± 0.55
0.1
2.560 ± 0.37
3. Results
M. Shapiro et al. / Materials Science and Engineering C 39 (2014) 100–104
Fig. 4. SEM fractograph of beagle sartorius tested at 1 mm/min: (a) overview of fracture zone of this specimen; and (b) higher magnification of same specimen with endomysial motion relative to fibrils (gray arrow) and shearing (white arrow) highlighted.
103
4. Discussion
Fig. 5. SEM fractograph of beagle sartorius tested at 10 mm/min: (a) failure mode of this strain rate is mostly via abrupt fracture of muscle fibers with little shearing (gray arrow) and their motion in relation to the endomysium (white arrow); (b) higher magnification showing that shearing is absent with fracture of individual muscle fibers highlighted (white arrows); and (c) higher magnification with enhanced view of individual muscle fibers highlighted (white arrows).
The failure features for the skeletal muscle specimens tested at logarithmically increasing strain rates showed failure mode trends as strain rate was increased. SEM analysis indicated that tensile failure mode evolves as the strain rate is increased. At the two lower strain rates failure is dominated by endomysium shearing. This is expected since the endomysium has been shown to be the essential series-elastic element in muscle fiber [19] and, therefore, contributes greatly to the mechanical properties of skeletal muscle [20,21]. At the higher strain rate the mode switches to one of bulk failure of the fibers, with little evidence of shear or motion between fibers. Several reports have shown the endomysium as a random network of collagen fibrils [21–23], and Purslow [24] has shown that the arrangement of collagen fibrils in the endomysium affects the tensile stiffness parallel to the axis of muscle fibers. The endomysial network is composed of wavy collagen fibrils arranged in a near-random planar array [24], which enables the endomysial sheets of relatively inextensible collagen fibers [25] to adapt to the length changes in skeletal muscle by aligning the endomysial fibers along the edges of the muscle [20,26]. Thus, the regular wave pattern of the endomysium is well-suited for transmitting forces as muscle fibers contract and expand, while simultaneously avoiding severe strain in any of its regions [27]. As the muscle
tissue is elongated the randomly oriented fibers of the endomysium become aligned in the direction of the applied stretch [28]. As tension continued, the fibers of the endomysium become fully aligned (linear region). Increased stress beyond this point (plastic deformation) will lead to rupture [17], with rupture planes usually occurring along the path of the weaker structures [29]. At slow strain rates (0.1 mm/min), the muscles endured a longer duration of tensile force, resulting in total shear failure of the endomysium. Such slow tensile strain caused the muscle fibers to stretch out of the endomysium and with a significant motion relative to one another prior to fracture. This observation corroborates Trotter's assertion that tension is transmitted between adjacent myofibers through the endomysium [30]. These types of failure likely result in damage accumulation in the myofibrils and muscular units. At a strain rate of 1 mm/min, the muscle specimens demonstrated subtle differences relative to samples tested at 0.1 mm/min, demonstrating less shearing. Nonetheless, the fracture analysis did suggest that the primary mode of failure is tearing of the fasciculi as they move relative to one another with a significant endomysial shearing present. In addition, the fracture pattern suggests that less time was allowed for microstructural rearrangement in response to the applied load. Specimens tested at rates
104
M. Shapiro et al. / Materials Science and Engineering C 39 (2014) 100–104
data suggests that muscle fibrils had reached substantial endomysial alignment prior to the failure point at the two lower strain rates; the lower strength value noted at 10 mm/min suggests that a lower degree of structural alignment was achieved prior to the onset of fracture. Lastly, it should be noted that the structural alignment arguments made here are based load–displacement curves and on post-mechanical testing fracture analysis. Microscopic evaluation during mechanical testing or microscopic analysis of specimens tested only to distinct regions of the stress–strain curves (not necessarily to complete failure) would further elucidate if the current explanation for the rate dependent fracture behavior is correct. This type of analysis will be conducted in future work. Conflict of interest statement The authors declare that they have no conflict of interest.
Fig. 6. SEM fractograph of beagle sartorius freeze fractured in liquid Nitrogen: no relative motion between endomysium and fibrils is visible nor is shearing.
of 10 mm/min resembled the untested freeze fractured sample, which implies a strong rate dependence of failure mode in the beagle sartorius muscle. At a higher strain rate less time is allowed for hierarchical anatomic structure rearrangement prior to failure. Therefore, a minimal amount of shearing of the endomysium was visible in the micrographs, while the primary mode of failure in these specimens was an abrupt rupture of the fasciculi and muscle fibers. Relative to the two other rates, 10 mm/min fracture demonstrated lack of shearing of the endomysium and the presence of multiple abrupt ruptures of the muscle fibers. The muscle displays a “stair-step” fracture appearance, indicating that the muscle rupture can be found through and around the fibers [29]. These results seem to confirm that the endomysium may act as a mechanical coupling between skeletal muscle fibers [31], and that a necessary amount of time is required for the endomysial fibers to properly align in order to achieve maximal tension. Overall, the fracture behavior as a function of strain rate presented a behavior consistent with time–temperature equivalence in polymeric materials [32]. According to the mechanical testing results within the limited amount of strain rates applied, the “optimal” strength of the muscle tissue was visible at 1 mm/min, and this is likely the result of the time allowed for alignment of the muscle fibers prior to fracture. The specimen tested at 10 mm/min presented with lower strength values, suggesting that tensile forces at this speed do not allow for the optimal alignment of the muscle fibers prior to fracture. The stress–strain curves show regions of linear behavior followed by yield and ultimately sequential failure of the muscle. The differences in failure features explained for the various strain rates displayed what we speculate to be the aligning of endomysium and fasciculi, and optimal alignment with subsequent failure is indicated by the peak load. This
References [1] G.M. Kerkhoffs, N. van Es, T. Wieldraaijer, I.N. Sierevelt, J. Ekstrand, C.N. van Dijk, Knee Surg. Sports Traumatol. Arthrosc. 21 (2) (2013) 500–509, http://dx.doi.org/ 10.1007/s00167-012-2055-x [Epub 2012 May 24]. [2] U.M. Kujala, S. Orava, M. Jarvinen, Sports Med. 23 (1997) 397–404. [3] W.J. Rijnberg, B. van Linge, Arch. Orthop. Trauma Surg. 112 (1993) 104–105. [4] T. Marom, D. Itskoviz, S. Kutikov, J. Naftal, I. OStfeld, J. R. Army Med. Corps 155 (2009) 24–26. [5] M. MIller, S. Thomspon, J. Hart, Review of Orthopedics, Sixth edition Saunders, Philadelphia, 2012. [6] M. Van Loocke, C.G. Lyons, C.K. Simms, J. Biomech. 41 (2008) 1555–1566. [7] S.D. Abramowitch, S.L. Woo, T.D. Clineff, R.E. Debski, Ann. Biomed. Eng. 32 (2004) 329–335. [8] J.A. Weiss, J.C. Gardiner, C. Bonifasi-Lista, J. Biomech. 35 (2002) 943–950. [9] D.S. Cronin, Proceedings of the Eudem2-Scot Conference, 2003. [10] Y.C. Deng, W. Kong, H. Ho, SAE Technical Paper, 01, 1999, pp. 165–172. [11] N. Gupta, K. Cho, Adv. Mater. Process. 168 (2010) 32–33. [12] A. Palevski, I. Glaich, S. Portnoy, E. Linder-Ganz, A. Gefen, J. Biomech. Eng. 128 (2006) 782–787. [13] E.M. Bosboom, M.K. Hesselink, C.W. Oomens, C.V. Bouten, M.R. Drost, F.P. Baaijens, J. Biomech. 34 (2001) 1365–1368. [14] T.M. Best, J. McElhaney, W.E. Garrett Jr., B.S. Myers, J. Biomech. 27 (1994) 413–419. [15] Y. Zheng, A.F. Mak, B. Lue, J. Rehabil. Res. Dev. 36 (1999) 71–85. [16] M. Van Loocke, C.G. Lyons, C.K. Simms, J. Biomech. 39 (2006) 2999–3009. [17] R.J. Carroll, F.P. Rorer, S.B. Jones, J.R. Cavanaugh, J. Food Sci. 43 (1978) 1181–1187. [18] F. Velardi, F. Fraternali, M. Angelillo, Biomech. Model. Mechanobiol. 5 (2006) 53–61. [19] E. Otten, Exerc. Sport Sci. Rev. 16 (1988) 89–137. [20] R.W.D. Rowe, J. Food Technol. 9 (1974) 501–508. [21] J.A. Trotter, P.P. Purslow, J. Morphol. 212 (1992) 109–122. [22] T.K. Borg, J.B. Caulfield, Tissue Cell 12 (1980) 197–207. [23] T. Nishimura, A. Hattori, K. Takahashi, Acta Anat. 151 (1994) 250–257. [24] P.P. Purslow, J.A. Trotter, J. Muscle Res. Cell Motil. 15 (1994) 299–308. [25] A.J. Bailey, J. Sci. Food Agric. 23 (1972) 995. [26] A. Viidik, Int. Rev. Connect. Tissue Res. 6 (1973) 127–215. [27] T. Nishimura, K. Ojima, A. Liu, A. Hattori, K. Takahashi, Tissue Cell 28 (1996) 527–536. [28] D.W. Stanley, L.M. McKnight, W.G.S. Hines, W.R. Usborne, J.M. Deman, J. Texture Stud. 3 (1972) 51–68. [29] S.B. Jones, R.J. Carroll, J.R. Cavanaugh, J. Food Sci. 42 (1977) 125–131. [30] J.A. Trotter, J. Morphol. 206 (1990) 351–361. [31] S.F. Street, J. Cell. Physiol. 114 (1983) 346–364. [32] B. Erman, J. Mark, Structures and Properties of Rubberlike Networks, Oxford University Press, 1997.
