- Email: [email protected]
A comprehensive experimental study on material properties of human brain tissue
Journal of Biomechanics 46 (2013) 2795–2801
Contents lists available at ScienceDirect
Journal of Biomechanics journal homepage: www.elsevier.com/locate/jbiomech www.JBiomech.com
A comprehensive experimental study on material properties of human brain tissue Xin Jin n, Feng Zhu, Haojie Mao, Ming Shen, King H. Yang Department of Biomedical Engineering, Wayne State University, Detroit, MI, USA
art ic l e i nf o
a b s t r a c t
Article history: Accepted 2 September 2013
A comprehensive study on the biomechanical response of human brain tissue is necessary to investigate traumatic brain injury mechanisms. Published brain material property studies have been mostly performed under a specific type of loading, which is insufficient to develop accurate brain tissue constitutive equations. In addition, inconsistent or contradictory data in the literature made it impossible for computational model developers to create a single brain material model that can fit most, if not all, experimental results. In the current study, a total of 240 brain tissue specimens were tested under tension (n ¼ 72), compression (n ¼72), and shear (n ¼ 96) loading modes at varying strain rates. Gray–white matter difference, regional difference, and directional difference within white matter were also investigated. Stress–strain relationships of human brain tissue were obtained up to 50% of engineering strain. Strain rate dependency was observed under all three loading modes. White matter was stiffer than gray matter in compression and shear. Corona radiata was found to be stiffer than cortex, thalamus, and corpus callosum in tension and compression. Directional dependency of white matter was observed under shear loading. & 2013 Elsevier Ltd. All rights reserved.
Keywords: Human brain tissue Material properties Tension Compression Shear
1. Introduction In the United States, approximately 1.7 million people sustain a traumatic brain injury (TBI) annually (CDC, 2011). TBI also contributes to one-third of all motor vehicle related deaths (CDC, 2011) and presents a major social, economic, and health problem. In order to investigate TBI mechanisms, computational modeling has been widely applied to provide insight into the local deformation field during traumatic loading. For this purpose, understanding the biomechanical response of brain tissue is necessary to provide accurate experimental data to allow for development of constitutive equations in numerical simulations. The mechanical response of brain tissue is complex. Current knowledge shows that brain tissue is anisotropic, non-linear, and viscoelastic (for example, Prange et al., 2000; Miller and Chinzei, 1997; Donnelly and Medige, 1997). Moreover, the constitutive equations derived under one loading condition do not necessarily predict the material response under another loading mode (Miller and Chinzei, 1997, 2002). During the past fifty years, numerous experimental studies have been published on brain tissue material properties. They have been mostly performed under a single mode of loading, e.g.,
n Correspondence to: Bioengineering Center, Wayne State University, 818 W. Hancock, Detroit, MI 48201, USA. Tel.: þ 313 577 8577; fax: þ313 577 8333. E-mail address: [email protected] (X. Jin).
0021-9290/$ - see front matter & 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.jbiomech.2013.09.001
compression (for example, Galford and McElhaney, 1970; Estes and McElhaney, 1970; Miller and Chinzei, 1997; Cheng and Bilston, 2007), tension (Miller and Chinzei, 2002; Velardi et al., 2006), and shear (for example, Fallenstein et al., 1969; Donnelly and Medige, 1997; Prange et al., 2000; Bilston et al., 2001; Takhounts et al., 2003; Hrapko et al., 2006). These individual studies are valuable for a better understanding of brain tissue material properties but are still insufficient to comprehensively describe the brain tissue behavior for numerical modeling purposes. Moreover, a large degree of variance observed among the published results made it impossible for computational model developers to create a single brain material model that can fit most, if not all, experimental results. There could be several different causes for this variance, such as different species, testing methods, loading conditions, temperature, post mortem time, etc. Brain tissue is heterogeneous and anisotropic due to its neuroarchitecture. Published results indicated that material properties vary between gray and white matter (e.g., Prange et al., 2000; Manduca et al., 2001). Moreover, the material properties of a given matter may also change with the different regions from where the specimens were dissected. For example, among white matters, corpus callosum is less compliant than brain stem (Arbogast and Margulies, 1997) and corona radiata (Prange et al., 2000) under large deformation. The anisotropy of brain tissue, which could be associated with the highly oriented axon fibers, was reported in some studies (Arbogast and Margulies, 1998; Prange et al., 2000). The regional and directional dependency
2796
X. Jin et al. / Journal of Biomechanics 46 (2013) 2795–2801
of brain tissue material properties during traumatic loading is important to investigate injury mechanisms at specific anatomical regions and needs to be studied carefully. In summary, the lack of experimental studies with comprehensive loading types and well-controlled testing conditions is the major limitation on the study of brain tissue biomechanical properties, especially for human brain tissue. The purpose of this study was to experimentally investigate the material properties of human brain tissue under tension, compression, and shear at varying strain rates. In addition, the gray–white matter difference, regional difference, and directional difference within white matter were also investigated. These findings will help us to better understand brain tissue biomechanics and to develop a comprehensive and accurate constitutive model of brain tissue.
2. Material and methods 2.1. Specimen preparation Brain tissue from post mortem human subjects (PMHS) was used in this study. To avoid tissue degeneration, PHMSs were kept intact at 4 1C until craniotomy was performed prior to testing, with an average post mortem time for testing of four days. Table 1 shows the information of PMHSs tested in this study. Rectangular brain specimens of 14 mm 14 mm 5 mm were cut from the cadaveric head using two similar constructs, each consisting of a pair of paralleled scalpel blades (Fig. 1). The first pair of blades (Fig. 1a) was used to dissect a thin strip of brain tissue out of the whole brain at a thickness of 5 mm and the second pair (Fig. 1b) was then used to cut the brain strip into 14 mm 14 mm square blocks. These constructs helped to obtain test samples fairly close to a rectangular block with uniform dimensions for tension and shear testing. The dissected brain samples were stored in artificial cerebral spinal fluid (aCSF) solution until testing. Brain tissues were dissected from four locations, two in the gray matter (cortex and thalamus) and the other two in the white matter (corpus callosum and corona radiata), in order to determine if there is a regional difference on mechanical properties of the brain (Fig. 2). Additionally, directional dependent properties of the white matter were investigated in two ways. Tension and compression tests were conducted along two directions and shear tests were conducted along three directions. The definition of loading with respect to the fiber orientation (direction) is illustrated in Fig. 3. During dissection, the general direction of the axon fiber was followed based on anatomical knowledge, which was lateral–medial at corpus callosum and radial at corona radiata (see Fig. 2). The directional dependency of gray matter was not investigated in this study because its neuroarchitecture did not show any clear evidence of fiber alignment and no data reported in the literature Table 1 Basic PMHS characteristics and the test matrix. Patient no.
Sex
Age
IIAM 0293 WSU 362 WSU 364 WSU 501 WSU 526 UM 33,769 WSU 590 WSU 603 UM 33,881
M F M M F F F M M
66 67 45 85 80 60 94 64 67
Tension
thus far indicates any directional dependency for the gray matter. A total of 240 brain tissue specimens were tested under tension (n ¼72), compression (n¼ 72), and shear (n¼ 96) loading modes under varying strain rates. The test matrices are shown in Tables 2 and 3. 2.2. Mechanical testing 2.2.1. Tension and compression Tension and compression tests were performed on the brain tissue samples using a test apparatus as shown in Fig. 4. Testing was performed within an environmental chamber installed on the Instron testing table. Steam generated from an electric kettle was directed into the chamber through a pipe to humidify and heat the air inside. All tests were conducted at 37 1C monitored using a thermometer positioned next to the brain tissue samples. No pre-conditioning was performed and only one loading cycle was executed on each sample. Prior to tension testing, the bottom surface of the brain tissue block was glued (Crazy glue, Elmer's Products, Inc., Columbus, OH) onto a polyethylene plate connected to a 22.24 N capacity load cell (MDB-5, Transducer Technology Inc, Temecula, CA). The same glue was applied on the top surface of the brain tissue sample. The top polyethylene plate, which was connected to the actuator of an Instron Machine (Model 1321 frame with a Model 8500 controller, Canton, MA), was then lowered to adhere the top plate to the brain tissue sample. The position of Instron actuator was recorded at the moment the plate surfaces contacted brain tissue specimen. A pre-compression of 1 mm was applied and maintained for one minute to allow glue set. Prior to applying tension loading, the pre-compression was released by lifting up the actuator to the pre-recorded position. The compression test was conducted using the same setup as the tension test but glue was not applied to allow free expansion of specimens. 2.2.2. Shear test The testing fixture for the shear test is shown in Fig. 5. The brain tissue specimen was glued on the fixed block. The micrometer head mounted on the moving block served to horizontally move the moving block to apply precompression to allow glue to set. Similar to the tension and compression test, the reading of the micrometer head was recorded at the initial contact between brain specimen and moving block. After the glue was set, the moving block was reversed back to release the pre-compression. The Instron actuator then lifted up to apply shear loading on the specimen. The shear test was also conducted at 37 1C and only one loading cycle was executed on each sample without pre-conditioning. 2.3. Data acquisition and analysis Force–displacement histories were recorded from tension, compression, and shear testing at a sampling rate of 10 kHz. Engineering stress was calculated as the ratio of force and nominal area of specimen and engineering strain was calculated as the ratio of actuator displacement and nominal initial thickness of specimen.
Fiber directions Compression
Shear
Fig. 2. Locations of the brain tissue samples tested: gray matter (cortex and thalamus) and white matter (corpus callosum and corona radiata).
Fig. 1. Uniform dimension cutting constructs.
X. Jin et al. / Journal of Biomechanics 46 (2013) 2795–2801
2797
Fig. 3. (a) Loading direction defined for tension and compression tests. D1: fibers are perpendicular to loading surface; D2: fibers are parallel to loading surface. (b) Loading direction defined for shear test. D1: fibers are perpendicular to loading surface; D2: fibers are parallel to loading surface and parallel to loading direction; D3: fibers are parallel to loading surface and perpendicular to loading direction.
Table 2 Test matrix for tension and compression tests. Strain rate
Gray Matter
White Matter
Cortex Thalamus Corpus callosum
Low (0.5/s) 4 Medium (5/s) 4 High (30/s) 4 Subtotal 12
4 4 4 12
Subtotal Corona radiata
D1
D2
D1
D2
4 4 4 12
4 4 4 12
4 4 4 12
4 4 4 12
Actuator
24 24 24 72
Brain tissue
Glue
Fig. 4. Tension test configuration.
Table 3 Test matrix for shear test. Strain Rate
Gray matter
White matter
Cortex Thalamus Corpus callosum
Low (0.5/s) 4 Medium (5/s) 4 High (30/s) 4 Subtotal 12
4 4 4 12
Total Corona radiata
D1
D2
D3
D1
D2
D3
4 4 4 12
4 4 4 12
4 4 4 12
4 4 4 12
4 4 4 12
4 4 4 12
32 32 32 96
Three velocities were chosen to conduct testing at 2.5, 25, and 150 mm/s, which correspond to the strain rates of 0.5 (low), 5 (medium), and 30 (high) s 1, respectively. A maximum strain level of 50% was performed during the experiments. Stress at the maximum strain was taken and ANOVA was applied to study the significance of all the factors, including the strain rate, fiber direction, region, and material (white or gray matter). Since the factor of fiber direction is fully crossing, a separate test within white matter was carried out to study direction effects (Test 1). Then, regional effects were studied (Test 2) by ignoring directional effects within white matter. Tukey's post hoc test was conducted to do the pairwise comparisons if significance was observed in ANOVA. Finally, to illustrate the difference between the white and gray matter, all test data for white and gray matter were combined and compared (Test 3). Linear regression was conducted to investigate the age effect on brain material properties. Statistical analysis was performed using SPSS (version 12.0, SPSS Inc. Chicago, IL). A probability value of less than 0.05 (po0.05) was considered significant in all analysis.
3. Results In all tests, strain rate dependency of brain tissue is clear. The maximum stress of three loading rates is significantly different
Fig. 5. Fixture of shear testing.
from each other (p o0.001). Fig. 6 summarizes the average stress of gray and white matters at 50% of strain for the tension, compression, and shear tests. Stress–strain relationships are shown in Fig. 7 (see the detailed data sheet in APPENDIX, and the full data set is available as online supplemental material) for tension, compression, and shear at varying strain rates. The presence of regional effects was found in tension and compression tests. To be specific, in compression tests the corona radiata was found significantly stiffer than the cortex, thalamus, and corpus callosum (po0.003). In tension tests, the corona radiata was significantly stiffer than cortex (p¼0.013), and marginally stiffer than corpus callosum (p¼0.069). No significant difference was found among the cortex, thalamus, and corpus callosum. Fig. 8 shows the peak stress for all four regions at the three strain rates.
X. Jin et al. / Journal of Biomechanics 46 (2013) 2795–2801
Gray-white Matter and Strain Rate Effect
Gray-white Matter and Strain Rate Effect Grey 16
**
White
12
8
** **
4
0 low
medium
Gray-white Matter and Strain Rate Effect 2.5
40
Compressive Stress at 50% Strain (kPa)
Tensile Stress at 50% Strain (kPa)
20
Gray
Gray **
White **
30
20
Shear Stress at 50% Strain (kPa)
2798
**
10
1.5
** **
1
0.5
0
0
high
**
White 2
low
medium
Strain Rate
high
low
medium
Strain Rate
high
Strain Rate
12
8
4
0
Gray Low White Low Gray Med White Med Gray Hige White High
30
20
Engineering Stress (kPa)
Gray Low White Low Gray Med White Med Gray Hige White High
16
Engineering Stress (kPa)
Engineering Stress (kPa)
Fig. 6. Summary of average stress at 50% strain of gray and white matter at varying strain rates. **p o 0.001. (a) Tension, (b) Compression, and (c) Shear.
10
0 0.0
0.1
0.2
0.3
0.4
0.5
Gray Low White Low Gray High White High Gray Med White Med
2.0
1.5
1.0
0.5
0.0 0.0
0.1
Engineering Strain
0.2
0.3
0.4
0.5
0.0
0.1
Engineering Strain
0.2
0.3
0.4
0.5
Engineering Strain
Fig. 7. Stress–strain relationships of gray and white matter at varying strain rates. (a) Tension, (b) Compression, and (c) Shear.
Regional and Strain Rate Effect
16
12
8
4
0 Cortex (ct)
Thalamus Corpus Corona (th) Callosum (cc) Radiata(cr)
* 30
20
10
medium
2
1.5
1
0.5
0
0 Cortex(ct)
Thalamus (th) Corpus Corona Callosum(cc) Radiata(cr)
Region low
2.5
40 Shear Stress at 50% Strain (kPa)
*
Regional and Strain Rate Effect
Regional and Strain Rate Effect Compressive Stress at 50% Strain (kPa)
Tensile Stress at 50% Strain (kPa)
20
Cortex(ct)
Thalamus (th) Corpus Corona Callosum(cc) Radiata(cr)
Region high
Low
Medium
Region High
low
medium
high
Fig. 8. Summary of regional dependency. *p o0.05. (a) Tension, (b) Compression, and (c) Shear.
In tension and compression tests, no significance was found for the directional factor. However, in shear tests, shear stress along direction 2 was found to be significantly higher than the other two directions (p ¼0.022). To study the age effect of brain tissue properties, data points of corona radiata were excluded from compression and tension because this region was found significantly stiffer than other regions under these two loading types. For each loading type and strain rate, the average stress at 50% strain was used to conduct linear regression with respect to the age. Regression results were shown in Fig. 9.
No significance was found for the linear regression between stress and age.
4. Discussion Estes and McElhaney (1970) conducted uniaxial, constant rate compression tests on human brain tissue. Specimens dissected from corona radiata were loaded up to 65% of engineering strain at strain rates of 0.08, 0.8, 8, and 40 s 1. General agreement
X. Jin et al. / Journal of Biomechanics 46 (2013) 2795–2801
Stress vs. Strain in Compression Test
Compression
Engineering Stress (kPa)
Stress at the maximum strain(kPa)
25 y = 0.1284x + 11.851 R² = 0.4834, p=0.305
20
y = 0.2523x + 0.5659 R² = 0.8413, p=0.083
15
10
High Medium
y = 0.3372x -8.4389 R² = 0.4492, p=0.33
5
2799
50
Current Study
8 s-1
30 30 s-1
0.8 s-1
20
5 s-1 0.5 s-1
10 0 0
Low
40 s-1
Estes and McElhaney, 1970
40
0.1
0.2
0.3
0.4
0.5
0.6
0.7
Engineering Strain Fig. 10. Comparison of brain tissue compression properties.
0 40
45
50
55
60
65
70
Age (year) Tension
12
Stress at the maximum strain (kPa)
High Medium
10
Low
y = -0.0874x + 16.075 R² = 0.6766, p=0.177
8
6 y = -0.0306x + 6.3062 R² = 0.1188, p=0.655 4
2 y = -0.0061x + 2.3135 R² = 0.0211, p=0.855
Fig. 11. Comparison of brain tissue shear properties.
0 55
60
65
70
75
80
85
90
Age (year)
Shear
Stress at the maximum strain (kPa)
2.5
High Medium Low
2
y = -0.0018x + 1.9484 R² = 0.0171, p=0.869
1.5
y = 0.0056x + 0.3844 R² = 0.1661, p=0.592
1
0.5
y = 0.0011x + 0.4997 R² = 0.0389, p=0.803
0 65
70
75
80
85
90
95
100
Age (year) Fig. 9. Linear regression between stress at the maximum strain and age for (a) compression, (b) tension and (c) shear.
was observed from comparison of the compression properties of white matter in the current study with Estes' data (Fig. 10). Donnelly and Medige (1997) studied material properties of human brain white matter. Constant velocity, parallel plate shear tests were performed at strain rates of 0, 30, 60, 90, 120, and 180 s 1. Engineering strain level tested in this study was up to 100%. Shear test results of the white matter from the current study correlate well with Donnelly's data (Fig. 11). Corona radiata was found to be stiffer than the other three regions under tension and compression. In the shear test, this
difference was not significant. One possible reason could be the relatively small sample size adopted in the current study. Because the shear property is the weakest compared to the other two loading directions, regional effects are even more difficult to detect with the current sample size. In addition, the regional dependency found in the current study is partially supported by the study of Prange et al. (2000), who reported corona radiata is stiffer than corpus callosum in shear. Directional dependency was also observed in the shear test. The stress level along direction 2 was found to be significantly higher than the other two directions. This finding is in accord with the study of Arbogast et al. (1995). Another white matter directional dependency study from Prange et al. (2000) showed contrary results. They found that along the fiber direction, brain tissue is stiffer (D2 4D1) in corpus callosum but more compliant in corona radiata (D2 oD1). The inconsistent results could be a result of the curve fitting procedure during data analysis. In their study, the shear relaxation test was conducted and material parameters of the Ogden model were obtained and compared to study the directional dependency. Recent studies have shown that the Ogden model parameters are sensitive to loading modes and cannot accurately describe the brain tissue behavior based on a single loading mode (Miller and Chinzei, 1997, 2002). Since the stress measurement was not reported in Prange's study, no further comparison can be performed with the current results. As concluded in most published studies, strain rate dependency of the brain tissue is prominent in all three loading modes (p o0.001). White matter (corpus callosum and corona radiata) was found to be stiffer than gray matter (cortex and thalamus) under compression (p ¼0.024) and shear (p ¼0.022) loading. In the tension test, the average maximum stress of white matter was 20% higher than that of gray matter at the highest strain rate (Fig. 6)
2800
X. Jin et al. / Journal of Biomechanics 46 (2013) 2795–2801
but, overall, this difference was not significant (p ¼0.102) with the current sample size. As a result, a study with a larger sample size is needed in the future. Due to the difficulty of specimen procurement and experimental setup, material properties of human brain tissue in tension have never been conducted previously. Data from the current study provides us with useful information for better understanding of brain tissue biomechanics. Age-dependency of brain biomechanics has been investigated since 1990s and great discrepancy was observed among the published literatures. Thibault and Margulies (1998) reported adult pig brain is stiffer than the infant pig brain. Prange and Margulies (2002) reported that infant pig brain is about twice as stiff as adult pig brain. For studies using rat brain, Gefen et al. (2003) found immature brain is stiffer while Finan et al. (2012) observed brain stiffens with age. In the current study, human brain tissue specimens were tested using PMHS with ages ranging from 45 to 94. Linear regression between the stress and age did not show any significance (p 40.05 for all loading types and strain rate groups) with the current sample size (N ¼ 4 for each loading type and strain rate). More studies are needed in the future to clarify this issue. Post-mortem time could potentially affect the brain tissue material properties but this effect has not been fully understood. Garo et al. (2007) have reported that the shear modulus of pig brain tissue began to increase 6 h post-mortem. Nicolle et al. (2004) also observed 6% increase of shear modulus between samples measured at 24 and 48 hours post-mortem. However, Darvish and Crandall (2001) did not find correlation between storage time (3–16 days) and the variation in mechanical properties of bovine brain tissue. In the current study, constrained by the specimen procurement protocol (such as blood test, transportation, etc.), brain tissue testing was conducted 3–5 days postmortem. The post-mortem effect on brain tissue material properties was not investigated, which is the major limitations of the current study. Future work is needed to clarify the post-mortem effect on human brain tissue biomechanics to better interpret the data in the current study. In the current study, brain tissue specimens were stored in aCSF, the most physiological-like environment currently available until testing. During the mechanical testing, wet air (100% humidity) was applied to prevent dehydration of the specimen. This is commonly adopted in brain material property testing (e.g., Thibault and Margulies, 1998; Nicolle et al., 2004; Hrapko et al., 2006). As a result, even though the testing was not conducted in a liquid-surrounded environment, brain material properties obtained should largely represent in-situ conditions under the current experimental setup. In summary, published material property studies have been mostly conducted on animals and experimental data for human brain tissue is very limited. Although the biological basis and neuroarchitecture for animal brain and human brain are close, there is no commonly accepted scaling law available that allows us to apply animal results to human. Experimental results from the current study will greatly improve our understanding of human brain tissue biomechanics. Finally, experimental results obtained from different loading modes, large deformation level, and large range of strain rate provided useful data to develop accurate constitutive equations of human brain tissue for TBI mechanism study.
5. Conclusions In this study, uniaxial tension, compression, and shear tests were performed on human brain tissue at varying strain rates. Stress– strain relationships were obtained up to 50% of engineering strain under all loading modes. Strain rate dependency was observed under all three loading modes. White matter was stiffer than gray matter in
compression and shear. Among the four tested regions, the corona radiata was found to be stiffer than the cortex, thalamus, and corpus callosum in tension and compression. Directional dependency was also observed in white matter under shear loading. Experimental results from the current study provide useful information to better understand brain tissue biomechanics and will eventually help to develop accurate constitutive equations for brain tissue.
Conflict of interest statement Authors declare no conflict of interest regarding the submitted manuscript.
Acknowledgments This study was supported by the Global Human Body Model Consortium. The authors would like to express their appreciation to Abrar Wazir for his assistance with grammar correcting on the manuscript.
Appendix A. Supplementary material Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.jbiomech.2013.09.001. References Arbogast, K.B., Meaney, D.F., and Thibault, L.E., 1995. biomechanical characterization of the constitutive relationship for the brainstem. In: Proceedings of the 39th Stapp Car Crash Conference. Arbogast, K.B., and Margulies, S.S., 1997. Regional differences in mechanical properties of the porcine central nervous system. In: Proceedings of the 41st Stapp Car Crash Conference, pp. 293–300. Arbogast, K.B., Margulies, S.S., 1998. Material characterization of the brainstem from oscillatory shear tests. Journal of Biomechanics 31, 801–807. Bilston, L.E., Liu, Z., Phan-Thien, N., 2001. Large strain behavior of brain tissue in shear: some experimental data and differential constitutive model. Biorheology 34, 377–385. CDC, Traumatic Brain Injury. Updated October 6 2011. Available from: 〈http:/www. cdc.gov/TraumaticBrainInjury/index.html〉 (accessed 12.06.13). Cheng, S., Bilston, L.E., 2007. Unconfined compression of white matter. Journal of Biomechanics. Donnelly, B.R., Medige, J., 1997. Shear properties of human brain tissue. ASME Journal of Biomechanical Engineering. Fallenstein, G.T., Hulce, V.D., Melvin, J.W., 1969. Dynamic mechanical properties of human brain tissue. Journal of Biomechanics 2, 217–226. Finan, J.D., Elkin, B.S., Pearson, E.M., Kalbian, I.L., Morrison III, B., 2012. Viscoelastic properties of the rat brain in the sagittal plane: effects of anatomical structure and age. Annals of Biomedical Engineering 40 (1), 70–78. Estes, M.S., McElhaney, J.H., 1970 .Response of Brain Tissue of Compressive Loading. American Society of Mechanical Engineers. Paper no. 70-BHF-13. Galford, J.E., McElhaney, J.H., 1970. A viscoelastic study of scalp, brain, and dura. Journal of Biomechanics. Garo, A., Hrapko, M., van Dommelen, J.A.W., Peters, G.W.M., 2007. Towards a reliable characterization of the mechanical behaviour of brain tissue: the effects of post-mortem time and sample preparation. Biorheology. Gefen, A., Gefen, N., Zhu, Q., Raghupathi, R., Margulies, S.S., 2003. Age-dependent changes in material properties of the brain and braincase of the rat. Journal of Neurotrauma 20, 1163–1177. Hrapko, M., van Dommelen, J.A.W., Peters, G.W.M., Wismans, J.S.H.M., 2006. The mechanical behavior of brain tissue: large strain response and constitutive modelling. Biorheology 43, 623–636. Manduca, A., Oliphant, T.E., Dresner, M.A., Mahowald, J.L., Kruse, S.A., Amromin, E., Felmlee, J.P., Greenleaf, J.F., Ehman, R.L., 2001. Magnetic resonance elastography: non-invasive mapping of tissue elasticity. Medical Image Analysis 5 (4), 237–254. Miller, K., Chinzei, K., 1997. Constitutive modelling of brain tissue: experiment and theory. Journal of Biomechanics 35, 483–490. Miller, K., Chinzei, K., 2002. Mechanical properties of brain tissue in tension. Journal of Biomechanics 35, 483–490. Nicolle, S., Lounis, M., Willinger, R., 2004. Shear properties of brain tissue over a frequency range relevant for automotive impact situations: new experimental results. Stapp Car Crash Journal 48, 239–258.
X. Jin et al. / Journal of Biomechanics 46 (2013) 2795–2801
Prange, M.T., Meaney, D.F., Margulies, S.S. 2000. Defining Brain Mechanical Properties: Effects of Region, Direction, and Species. In: Proceedings of the 44th Stapp Car Crash Conference. Prange, M.T., Margulies, S.S., 2002. Regional, directional, and age- dependent properties of the brain undergoing large deformation. Journal of Biomechanical Engineering 124, 244–252. Takhounts, E.G., Crandall, J.R., Darvish, K.K., 2003. On the importance of nonlinearity of brain tissue under large deformations. Stapp Car Crash Journal.
2801
Thibault, K.L., Margulies, S.S., 1998. Age-dependent material properties of the porcine cerebrum: effect on pediatric inertial head injury criteria. Journal of Biomechanics 31 (12), 1119–1126. Velardi, F., Fraternali, F., Angelillo, M., 2006. Anisotropic constitutive equations and experimental tensile behavior of brain tissue. Biomechanics and Modeling in Mechanobiology 5, 53–61.
Contents lists available at ScienceDirect
Journal of Biomechanics journal homepage: www.elsevier.com/locate/jbiomech www.JBiomech.com
A comprehensive experimental study on material properties of human brain tissue Xin Jin n, Feng Zhu, Haojie Mao, Ming Shen, King H. Yang Department of Biomedical Engineering, Wayne State University, Detroit, MI, USA
art ic l e i nf o
a b s t r a c t
Article history: Accepted 2 September 2013
A comprehensive study on the biomechanical response of human brain tissue is necessary to investigate traumatic brain injury mechanisms. Published brain material property studies have been mostly performed under a specific type of loading, which is insufficient to develop accurate brain tissue constitutive equations. In addition, inconsistent or contradictory data in the literature made it impossible for computational model developers to create a single brain material model that can fit most, if not all, experimental results. In the current study, a total of 240 brain tissue specimens were tested under tension (n ¼ 72), compression (n ¼72), and shear (n ¼ 96) loading modes at varying strain rates. Gray–white matter difference, regional difference, and directional difference within white matter were also investigated. Stress–strain relationships of human brain tissue were obtained up to 50% of engineering strain. Strain rate dependency was observed under all three loading modes. White matter was stiffer than gray matter in compression and shear. Corona radiata was found to be stiffer than cortex, thalamus, and corpus callosum in tension and compression. Directional dependency of white matter was observed under shear loading. & 2013 Elsevier Ltd. All rights reserved.
Keywords: Human brain tissue Material properties Tension Compression Shear
1. Introduction In the United States, approximately 1.7 million people sustain a traumatic brain injury (TBI) annually (CDC, 2011). TBI also contributes to one-third of all motor vehicle related deaths (CDC, 2011) and presents a major social, economic, and health problem. In order to investigate TBI mechanisms, computational modeling has been widely applied to provide insight into the local deformation field during traumatic loading. For this purpose, understanding the biomechanical response of brain tissue is necessary to provide accurate experimental data to allow for development of constitutive equations in numerical simulations. The mechanical response of brain tissue is complex. Current knowledge shows that brain tissue is anisotropic, non-linear, and viscoelastic (for example, Prange et al., 2000; Miller and Chinzei, 1997; Donnelly and Medige, 1997). Moreover, the constitutive equations derived under one loading condition do not necessarily predict the material response under another loading mode (Miller and Chinzei, 1997, 2002). During the past fifty years, numerous experimental studies have been published on brain tissue material properties. They have been mostly performed under a single mode of loading, e.g.,
n Correspondence to: Bioengineering Center, Wayne State University, 818 W. Hancock, Detroit, MI 48201, USA. Tel.: þ 313 577 8577; fax: þ313 577 8333. E-mail address: [email protected] (X. Jin).
0021-9290/$ - see front matter & 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.jbiomech.2013.09.001
compression (for example, Galford and McElhaney, 1970; Estes and McElhaney, 1970; Miller and Chinzei, 1997; Cheng and Bilston, 2007), tension (Miller and Chinzei, 2002; Velardi et al., 2006), and shear (for example, Fallenstein et al., 1969; Donnelly and Medige, 1997; Prange et al., 2000; Bilston et al., 2001; Takhounts et al., 2003; Hrapko et al., 2006). These individual studies are valuable for a better understanding of brain tissue material properties but are still insufficient to comprehensively describe the brain tissue behavior for numerical modeling purposes. Moreover, a large degree of variance observed among the published results made it impossible for computational model developers to create a single brain material model that can fit most, if not all, experimental results. There could be several different causes for this variance, such as different species, testing methods, loading conditions, temperature, post mortem time, etc. Brain tissue is heterogeneous and anisotropic due to its neuroarchitecture. Published results indicated that material properties vary between gray and white matter (e.g., Prange et al., 2000; Manduca et al., 2001). Moreover, the material properties of a given matter may also change with the different regions from where the specimens were dissected. For example, among white matters, corpus callosum is less compliant than brain stem (Arbogast and Margulies, 1997) and corona radiata (Prange et al., 2000) under large deformation. The anisotropy of brain tissue, which could be associated with the highly oriented axon fibers, was reported in some studies (Arbogast and Margulies, 1998; Prange et al., 2000). The regional and directional dependency
2796
X. Jin et al. / Journal of Biomechanics 46 (2013) 2795–2801
of brain tissue material properties during traumatic loading is important to investigate injury mechanisms at specific anatomical regions and needs to be studied carefully. In summary, the lack of experimental studies with comprehensive loading types and well-controlled testing conditions is the major limitation on the study of brain tissue biomechanical properties, especially for human brain tissue. The purpose of this study was to experimentally investigate the material properties of human brain tissue under tension, compression, and shear at varying strain rates. In addition, the gray–white matter difference, regional difference, and directional difference within white matter were also investigated. These findings will help us to better understand brain tissue biomechanics and to develop a comprehensive and accurate constitutive model of brain tissue.
2. Material and methods 2.1. Specimen preparation Brain tissue from post mortem human subjects (PMHS) was used in this study. To avoid tissue degeneration, PHMSs were kept intact at 4 1C until craniotomy was performed prior to testing, with an average post mortem time for testing of four days. Table 1 shows the information of PMHSs tested in this study. Rectangular brain specimens of 14 mm 14 mm 5 mm were cut from the cadaveric head using two similar constructs, each consisting of a pair of paralleled scalpel blades (Fig. 1). The first pair of blades (Fig. 1a) was used to dissect a thin strip of brain tissue out of the whole brain at a thickness of 5 mm and the second pair (Fig. 1b) was then used to cut the brain strip into 14 mm 14 mm square blocks. These constructs helped to obtain test samples fairly close to a rectangular block with uniform dimensions for tension and shear testing. The dissected brain samples were stored in artificial cerebral spinal fluid (aCSF) solution until testing. Brain tissues were dissected from four locations, two in the gray matter (cortex and thalamus) and the other two in the white matter (corpus callosum and corona radiata), in order to determine if there is a regional difference on mechanical properties of the brain (Fig. 2). Additionally, directional dependent properties of the white matter were investigated in two ways. Tension and compression tests were conducted along two directions and shear tests were conducted along three directions. The definition of loading with respect to the fiber orientation (direction) is illustrated in Fig. 3. During dissection, the general direction of the axon fiber was followed based on anatomical knowledge, which was lateral–medial at corpus callosum and radial at corona radiata (see Fig. 2). The directional dependency of gray matter was not investigated in this study because its neuroarchitecture did not show any clear evidence of fiber alignment and no data reported in the literature Table 1 Basic PMHS characteristics and the test matrix. Patient no.
Sex
Age
IIAM 0293 WSU 362 WSU 364 WSU 501 WSU 526 UM 33,769 WSU 590 WSU 603 UM 33,881
M F M M F F F M M
66 67 45 85 80 60 94 64 67
Tension
thus far indicates any directional dependency for the gray matter. A total of 240 brain tissue specimens were tested under tension (n ¼72), compression (n¼ 72), and shear (n¼ 96) loading modes under varying strain rates. The test matrices are shown in Tables 2 and 3. 2.2. Mechanical testing 2.2.1. Tension and compression Tension and compression tests were performed on the brain tissue samples using a test apparatus as shown in Fig. 4. Testing was performed within an environmental chamber installed on the Instron testing table. Steam generated from an electric kettle was directed into the chamber through a pipe to humidify and heat the air inside. All tests were conducted at 37 1C monitored using a thermometer positioned next to the brain tissue samples. No pre-conditioning was performed and only one loading cycle was executed on each sample. Prior to tension testing, the bottom surface of the brain tissue block was glued (Crazy glue, Elmer's Products, Inc., Columbus, OH) onto a polyethylene plate connected to a 22.24 N capacity load cell (MDB-5, Transducer Technology Inc, Temecula, CA). The same glue was applied on the top surface of the brain tissue sample. The top polyethylene plate, which was connected to the actuator of an Instron Machine (Model 1321 frame with a Model 8500 controller, Canton, MA), was then lowered to adhere the top plate to the brain tissue sample. The position of Instron actuator was recorded at the moment the plate surfaces contacted brain tissue specimen. A pre-compression of 1 mm was applied and maintained for one minute to allow glue set. Prior to applying tension loading, the pre-compression was released by lifting up the actuator to the pre-recorded position. The compression test was conducted using the same setup as the tension test but glue was not applied to allow free expansion of specimens. 2.2.2. Shear test The testing fixture for the shear test is shown in Fig. 5. The brain tissue specimen was glued on the fixed block. The micrometer head mounted on the moving block served to horizontally move the moving block to apply precompression to allow glue to set. Similar to the tension and compression test, the reading of the micrometer head was recorded at the initial contact between brain specimen and moving block. After the glue was set, the moving block was reversed back to release the pre-compression. The Instron actuator then lifted up to apply shear loading on the specimen. The shear test was also conducted at 37 1C and only one loading cycle was executed on each sample without pre-conditioning. 2.3. Data acquisition and analysis Force–displacement histories were recorded from tension, compression, and shear testing at a sampling rate of 10 kHz. Engineering stress was calculated as the ratio of force and nominal area of specimen and engineering strain was calculated as the ratio of actuator displacement and nominal initial thickness of specimen.
Fiber directions Compression
Shear
Fig. 2. Locations of the brain tissue samples tested: gray matter (cortex and thalamus) and white matter (corpus callosum and corona radiata).
Fig. 1. Uniform dimension cutting constructs.
X. Jin et al. / Journal of Biomechanics 46 (2013) 2795–2801
2797
Fig. 3. (a) Loading direction defined for tension and compression tests. D1: fibers are perpendicular to loading surface; D2: fibers are parallel to loading surface. (b) Loading direction defined for shear test. D1: fibers are perpendicular to loading surface; D2: fibers are parallel to loading surface and parallel to loading direction; D3: fibers are parallel to loading surface and perpendicular to loading direction.
Table 2 Test matrix for tension and compression tests. Strain rate
Gray Matter
White Matter
Cortex Thalamus Corpus callosum
Low (0.5/s) 4 Medium (5/s) 4 High (30/s) 4 Subtotal 12
4 4 4 12
Subtotal Corona radiata
D1
D2
D1
D2
4 4 4 12
4 4 4 12
4 4 4 12
4 4 4 12
Actuator
24 24 24 72
Brain tissue
Glue
Fig. 4. Tension test configuration.
Table 3 Test matrix for shear test. Strain Rate
Gray matter
White matter
Cortex Thalamus Corpus callosum
Low (0.5/s) 4 Medium (5/s) 4 High (30/s) 4 Subtotal 12
4 4 4 12
Total Corona radiata
D1
D2
D3
D1
D2
D3
4 4 4 12
4 4 4 12
4 4 4 12
4 4 4 12
4 4 4 12
4 4 4 12
32 32 32 96
Three velocities were chosen to conduct testing at 2.5, 25, and 150 mm/s, which correspond to the strain rates of 0.5 (low), 5 (medium), and 30 (high) s 1, respectively. A maximum strain level of 50% was performed during the experiments. Stress at the maximum strain was taken and ANOVA was applied to study the significance of all the factors, including the strain rate, fiber direction, region, and material (white or gray matter). Since the factor of fiber direction is fully crossing, a separate test within white matter was carried out to study direction effects (Test 1). Then, regional effects were studied (Test 2) by ignoring directional effects within white matter. Tukey's post hoc test was conducted to do the pairwise comparisons if significance was observed in ANOVA. Finally, to illustrate the difference between the white and gray matter, all test data for white and gray matter were combined and compared (Test 3). Linear regression was conducted to investigate the age effect on brain material properties. Statistical analysis was performed using SPSS (version 12.0, SPSS Inc. Chicago, IL). A probability value of less than 0.05 (po0.05) was considered significant in all analysis.
3. Results In all tests, strain rate dependency of brain tissue is clear. The maximum stress of three loading rates is significantly different
Fig. 5. Fixture of shear testing.
from each other (p o0.001). Fig. 6 summarizes the average stress of gray and white matters at 50% of strain for the tension, compression, and shear tests. Stress–strain relationships are shown in Fig. 7 (see the detailed data sheet in APPENDIX, and the full data set is available as online supplemental material) for tension, compression, and shear at varying strain rates. The presence of regional effects was found in tension and compression tests. To be specific, in compression tests the corona radiata was found significantly stiffer than the cortex, thalamus, and corpus callosum (po0.003). In tension tests, the corona radiata was significantly stiffer than cortex (p¼0.013), and marginally stiffer than corpus callosum (p¼0.069). No significant difference was found among the cortex, thalamus, and corpus callosum. Fig. 8 shows the peak stress for all four regions at the three strain rates.
X. Jin et al. / Journal of Biomechanics 46 (2013) 2795–2801
Gray-white Matter and Strain Rate Effect
Gray-white Matter and Strain Rate Effect Grey 16
**
White
12
8
** **
4
0 low
medium
Gray-white Matter and Strain Rate Effect 2.5
40
Compressive Stress at 50% Strain (kPa)
Tensile Stress at 50% Strain (kPa)
20
Gray
Gray **
White **
30
20
Shear Stress at 50% Strain (kPa)
2798
**
10
1.5
** **
1
0.5
0
0
high
**
White 2
low
medium
Strain Rate
high
low
medium
Strain Rate
high
Strain Rate
12
8
4
0
Gray Low White Low Gray Med White Med Gray Hige White High
30
20
Engineering Stress (kPa)
Gray Low White Low Gray Med White Med Gray Hige White High
16
Engineering Stress (kPa)
Engineering Stress (kPa)
Fig. 6. Summary of average stress at 50% strain of gray and white matter at varying strain rates. **p o 0.001. (a) Tension, (b) Compression, and (c) Shear.
10
0 0.0
0.1
0.2
0.3
0.4
0.5
Gray Low White Low Gray High White High Gray Med White Med
2.0
1.5
1.0
0.5
0.0 0.0
0.1
Engineering Strain
0.2
0.3
0.4
0.5
0.0
0.1
Engineering Strain
0.2
0.3
0.4
0.5
Engineering Strain
Fig. 7. Stress–strain relationships of gray and white matter at varying strain rates. (a) Tension, (b) Compression, and (c) Shear.
Regional and Strain Rate Effect
16
12
8
4
0 Cortex (ct)
Thalamus Corpus Corona (th) Callosum (cc) Radiata(cr)
* 30
20
10
medium
2
1.5
1
0.5
0
0 Cortex(ct)
Thalamus (th) Corpus Corona Callosum(cc) Radiata(cr)
Region low
2.5
40 Shear Stress at 50% Strain (kPa)
*
Regional and Strain Rate Effect
Regional and Strain Rate Effect Compressive Stress at 50% Strain (kPa)
Tensile Stress at 50% Strain (kPa)
20
Cortex(ct)
Thalamus (th) Corpus Corona Callosum(cc) Radiata(cr)
Region high
Low
Medium
Region High
low
medium
high
Fig. 8. Summary of regional dependency. *p o0.05. (a) Tension, (b) Compression, and (c) Shear.
In tension and compression tests, no significance was found for the directional factor. However, in shear tests, shear stress along direction 2 was found to be significantly higher than the other two directions (p ¼0.022). To study the age effect of brain tissue properties, data points of corona radiata were excluded from compression and tension because this region was found significantly stiffer than other regions under these two loading types. For each loading type and strain rate, the average stress at 50% strain was used to conduct linear regression with respect to the age. Regression results were shown in Fig. 9.
No significance was found for the linear regression between stress and age.
4. Discussion Estes and McElhaney (1970) conducted uniaxial, constant rate compression tests on human brain tissue. Specimens dissected from corona radiata were loaded up to 65% of engineering strain at strain rates of 0.08, 0.8, 8, and 40 s 1. General agreement
X. Jin et al. / Journal of Biomechanics 46 (2013) 2795–2801
Stress vs. Strain in Compression Test
Compression
Engineering Stress (kPa)
Stress at the maximum strain(kPa)
25 y = 0.1284x + 11.851 R² = 0.4834, p=0.305
20
y = 0.2523x + 0.5659 R² = 0.8413, p=0.083
15
10
High Medium
y = 0.3372x -8.4389 R² = 0.4492, p=0.33
5
2799
50
Current Study
8 s-1
30 30 s-1
0.8 s-1
20
5 s-1 0.5 s-1
10 0 0
Low
40 s-1
Estes and McElhaney, 1970
40
0.1
0.2
0.3
0.4
0.5
0.6
0.7
Engineering Strain Fig. 10. Comparison of brain tissue compression properties.
0 40
45
50
55
60
65
70
Age (year) Tension
12
Stress at the maximum strain (kPa)
High Medium
10
Low
y = -0.0874x + 16.075 R² = 0.6766, p=0.177
8
6 y = -0.0306x + 6.3062 R² = 0.1188, p=0.655 4
2 y = -0.0061x + 2.3135 R² = 0.0211, p=0.855
Fig. 11. Comparison of brain tissue shear properties.
0 55
60
65
70
75
80
85
90
Age (year)
Shear
Stress at the maximum strain (kPa)
2.5
High Medium Low
2
y = -0.0018x + 1.9484 R² = 0.0171, p=0.869
1.5
y = 0.0056x + 0.3844 R² = 0.1661, p=0.592
1
0.5
y = 0.0011x + 0.4997 R² = 0.0389, p=0.803
0 65
70
75
80
85
90
95
100
Age (year) Fig. 9. Linear regression between stress at the maximum strain and age for (a) compression, (b) tension and (c) shear.
was observed from comparison of the compression properties of white matter in the current study with Estes' data (Fig. 10). Donnelly and Medige (1997) studied material properties of human brain white matter. Constant velocity, parallel plate shear tests were performed at strain rates of 0, 30, 60, 90, 120, and 180 s 1. Engineering strain level tested in this study was up to 100%. Shear test results of the white matter from the current study correlate well with Donnelly's data (Fig. 11). Corona radiata was found to be stiffer than the other three regions under tension and compression. In the shear test, this
difference was not significant. One possible reason could be the relatively small sample size adopted in the current study. Because the shear property is the weakest compared to the other two loading directions, regional effects are even more difficult to detect with the current sample size. In addition, the regional dependency found in the current study is partially supported by the study of Prange et al. (2000), who reported corona radiata is stiffer than corpus callosum in shear. Directional dependency was also observed in the shear test. The stress level along direction 2 was found to be significantly higher than the other two directions. This finding is in accord with the study of Arbogast et al. (1995). Another white matter directional dependency study from Prange et al. (2000) showed contrary results. They found that along the fiber direction, brain tissue is stiffer (D2 4D1) in corpus callosum but more compliant in corona radiata (D2 oD1). The inconsistent results could be a result of the curve fitting procedure during data analysis. In their study, the shear relaxation test was conducted and material parameters of the Ogden model were obtained and compared to study the directional dependency. Recent studies have shown that the Ogden model parameters are sensitive to loading modes and cannot accurately describe the brain tissue behavior based on a single loading mode (Miller and Chinzei, 1997, 2002). Since the stress measurement was not reported in Prange's study, no further comparison can be performed with the current results. As concluded in most published studies, strain rate dependency of the brain tissue is prominent in all three loading modes (p o0.001). White matter (corpus callosum and corona radiata) was found to be stiffer than gray matter (cortex and thalamus) under compression (p ¼0.024) and shear (p ¼0.022) loading. In the tension test, the average maximum stress of white matter was 20% higher than that of gray matter at the highest strain rate (Fig. 6)
2800
X. Jin et al. / Journal of Biomechanics 46 (2013) 2795–2801
but, overall, this difference was not significant (p ¼0.102) with the current sample size. As a result, a study with a larger sample size is needed in the future. Due to the difficulty of specimen procurement and experimental setup, material properties of human brain tissue in tension have never been conducted previously. Data from the current study provides us with useful information for better understanding of brain tissue biomechanics. Age-dependency of brain biomechanics has been investigated since 1990s and great discrepancy was observed among the published literatures. Thibault and Margulies (1998) reported adult pig brain is stiffer than the infant pig brain. Prange and Margulies (2002) reported that infant pig brain is about twice as stiff as adult pig brain. For studies using rat brain, Gefen et al. (2003) found immature brain is stiffer while Finan et al. (2012) observed brain stiffens with age. In the current study, human brain tissue specimens were tested using PMHS with ages ranging from 45 to 94. Linear regression between the stress and age did not show any significance (p 40.05 for all loading types and strain rate groups) with the current sample size (N ¼ 4 for each loading type and strain rate). More studies are needed in the future to clarify this issue. Post-mortem time could potentially affect the brain tissue material properties but this effect has not been fully understood. Garo et al. (2007) have reported that the shear modulus of pig brain tissue began to increase 6 h post-mortem. Nicolle et al. (2004) also observed 6% increase of shear modulus between samples measured at 24 and 48 hours post-mortem. However, Darvish and Crandall (2001) did not find correlation between storage time (3–16 days) and the variation in mechanical properties of bovine brain tissue. In the current study, constrained by the specimen procurement protocol (such as blood test, transportation, etc.), brain tissue testing was conducted 3–5 days postmortem. The post-mortem effect on brain tissue material properties was not investigated, which is the major limitations of the current study. Future work is needed to clarify the post-mortem effect on human brain tissue biomechanics to better interpret the data in the current study. In the current study, brain tissue specimens were stored in aCSF, the most physiological-like environment currently available until testing. During the mechanical testing, wet air (100% humidity) was applied to prevent dehydration of the specimen. This is commonly adopted in brain material property testing (e.g., Thibault and Margulies, 1998; Nicolle et al., 2004; Hrapko et al., 2006). As a result, even though the testing was not conducted in a liquid-surrounded environment, brain material properties obtained should largely represent in-situ conditions under the current experimental setup. In summary, published material property studies have been mostly conducted on animals and experimental data for human brain tissue is very limited. Although the biological basis and neuroarchitecture for animal brain and human brain are close, there is no commonly accepted scaling law available that allows us to apply animal results to human. Experimental results from the current study will greatly improve our understanding of human brain tissue biomechanics. Finally, experimental results obtained from different loading modes, large deformation level, and large range of strain rate provided useful data to develop accurate constitutive equations of human brain tissue for TBI mechanism study.
5. Conclusions In this study, uniaxial tension, compression, and shear tests were performed on human brain tissue at varying strain rates. Stress– strain relationships were obtained up to 50% of engineering strain under all loading modes. Strain rate dependency was observed under all three loading modes. White matter was stiffer than gray matter in
compression and shear. Among the four tested regions, the corona radiata was found to be stiffer than the cortex, thalamus, and corpus callosum in tension and compression. Directional dependency was also observed in white matter under shear loading. Experimental results from the current study provide useful information to better understand brain tissue biomechanics and will eventually help to develop accurate constitutive equations for brain tissue.
Conflict of interest statement Authors declare no conflict of interest regarding the submitted manuscript.
Acknowledgments This study was supported by the Global Human Body Model Consortium. The authors would like to express their appreciation to Abrar Wazir for his assistance with grammar correcting on the manuscript.
Appendix A. Supplementary material Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.jbiomech.2013.09.001. References Arbogast, K.B., Meaney, D.F., and Thibault, L.E., 1995. biomechanical characterization of the constitutive relationship for the brainstem. In: Proceedings of the 39th Stapp Car Crash Conference. Arbogast, K.B., and Margulies, S.S., 1997. Regional differences in mechanical properties of the porcine central nervous system. In: Proceedings of the 41st Stapp Car Crash Conference, pp. 293–300. Arbogast, K.B., Margulies, S.S., 1998. Material characterization of the brainstem from oscillatory shear tests. Journal of Biomechanics 31, 801–807. Bilston, L.E., Liu, Z., Phan-Thien, N., 2001. Large strain behavior of brain tissue in shear: some experimental data and differential constitutive model. Biorheology 34, 377–385. CDC, Traumatic Brain Injury. Updated October 6 2011. Available from: 〈http:/www. cdc.gov/TraumaticBrainInjury/index.html〉 (accessed 12.06.13). Cheng, S., Bilston, L.E., 2007. Unconfined compression of white matter. Journal of Biomechanics. Donnelly, B.R., Medige, J., 1997. Shear properties of human brain tissue. ASME Journal of Biomechanical Engineering. Fallenstein, G.T., Hulce, V.D., Melvin, J.W., 1969. Dynamic mechanical properties of human brain tissue. Journal of Biomechanics 2, 217–226. Finan, J.D., Elkin, B.S., Pearson, E.M., Kalbian, I.L., Morrison III, B., 2012. Viscoelastic properties of the rat brain in the sagittal plane: effects of anatomical structure and age. Annals of Biomedical Engineering 40 (1), 70–78. Estes, M.S., McElhaney, J.H., 1970 .Response of Brain Tissue of Compressive Loading. American Society of Mechanical Engineers. Paper no. 70-BHF-13. Galford, J.E., McElhaney, J.H., 1970. A viscoelastic study of scalp, brain, and dura. Journal of Biomechanics. Garo, A., Hrapko, M., van Dommelen, J.A.W., Peters, G.W.M., 2007. Towards a reliable characterization of the mechanical behaviour of brain tissue: the effects of post-mortem time and sample preparation. Biorheology. Gefen, A., Gefen, N., Zhu, Q., Raghupathi, R., Margulies, S.S., 2003. Age-dependent changes in material properties of the brain and braincase of the rat. Journal of Neurotrauma 20, 1163–1177. Hrapko, M., van Dommelen, J.A.W., Peters, G.W.M., Wismans, J.S.H.M., 2006. The mechanical behavior of brain tissue: large strain response and constitutive modelling. Biorheology 43, 623–636. Manduca, A., Oliphant, T.E., Dresner, M.A., Mahowald, J.L., Kruse, S.A., Amromin, E., Felmlee, J.P., Greenleaf, J.F., Ehman, R.L., 2001. Magnetic resonance elastography: non-invasive mapping of tissue elasticity. Medical Image Analysis 5 (4), 237–254. Miller, K., Chinzei, K., 1997. Constitutive modelling of brain tissue: experiment and theory. Journal of Biomechanics 35, 483–490. Miller, K., Chinzei, K., 2002. Mechanical properties of brain tissue in tension. Journal of Biomechanics 35, 483–490. Nicolle, S., Lounis, M., Willinger, R., 2004. Shear properties of brain tissue over a frequency range relevant for automotive impact situations: new experimental results. Stapp Car Crash Journal 48, 239–258.
X. Jin et al. / Journal of Biomechanics 46 (2013) 2795–2801
Prange, M.T., Meaney, D.F., Margulies, S.S. 2000. Defining Brain Mechanical Properties: Effects of Region, Direction, and Species. In: Proceedings of the 44th Stapp Car Crash Conference. Prange, M.T., Margulies, S.S., 2002. Regional, directional, and age- dependent properties of the brain undergoing large deformation. Journal of Biomechanical Engineering 124, 244–252. Takhounts, E.G., Crandall, J.R., Darvish, K.K., 2003. On the importance of nonlinearity of brain tissue under large deformations. Stapp Car Crash Journal.
2801
Thibault, K.L., Margulies, S.S., 1998. Age-dependent material properties of the porcine cerebrum: effect on pediatric inertial head injury criteria. Journal of Biomechanics 31 (12), 1119–1126. Velardi, F., Fraternali, F., Angelillo, M., 2006. Anisotropic constitutive equations and experimental tensile behavior of brain tissue. Biomechanics and Modeling in Mechanobiology 5, 53–61.