Skip to main content
NCBI home page
Journal of Anatomy logoLink to Journal of Anatomy
. 2015 Jul 21;227(4):506–513. doi: 10.1111/joa.12354

MRI-determined lumbar muscle morphometry in man and sheep: potential biomechanical implications for ovine model to human spine translation

Stephanie Valentin 1,✉, Theresia F Licka 1,2, James Elliott 3,4
PMCID: PMC4580108  EMSID: EMS74072  PMID: 26200090

Abstract

The sheep is a commonly used animal model for human lumbar spine surgery, but only in vitro investigations comparing the human and ovine spine exist. Spinal musculature has previously not been compared between man and sheep. This additional knowledge could further indicate to what extent these species are biomechanically similar. Therefore, the purpose of the study was to investigate spinal muscle morphometric properties using magnetic resonance imaging (MRI) in different age groups of healthy human participants and sheep in vivo. Healthy human participants (n = 24) and sheep (n = 17) of different age groups underwent T1-weighted MRI of the lumbar spine. Regions of interest of the muscles erector spinae (ES), multifidus (M) and psoas (PS) were identified. The ratio of flexor to extensor volume, ratio of M to ES volume, and muscle fat relative to an area of intermuscular fat were calculated. Sheep M to ES ratio was significantly smaller than in the human participants (sheep 0.16 ± 0.02; human 0.37 ± 0.05; P < 0.001), although flexor to extensor ratio was not significantly different between species (human 0.39 ± 0.08; sheep 0.43 ± 0.05; P = 0.06). Age did not influence any muscle ratio outcome. Sheep had significantly greater extensor muscle fat compared with the human participants (M left human 40.64%, sheep 53.81%; M right human 39.17%, sheep 51.33%; ES left human 40.86%, sheep 51.29%; ES right human 35.93%, sheep 44.38%; all median values; all P < 0.001), although PS did not show any significant between-species differences (PS left human 36.89%, sheep 33.67%; PS right human 32.78%, sheep 30.09%; P < 0.05). The apparent differences in the size and shape of sheep and human lumbar spine muscles may indicate dissimilar biomechanical and functional demands, which is an important consideration when translating to human surgical models.

Keywords: lumbar spine, magnetic resonance imaging, muscle, muscle fatty infiltrate, ovine model

Introduction

Sheep are commonly used as an animal model for the human spine due to their apparent anatomical similarities (Wilke et al. 1997; Hasler et al. 2010; Sheng et al. 2010). Animal model investigations are generally limited to in vitro testing of functional spinal units (FSUs) with the majority of musculature removed (Veres et al. 2010; Leckie et al. 2012). However, a single FSU displays different biomechanical properties in comparison to a series of multiple FSUs (Dickey & Kerr, 2003), and the removal of muscle tissue considerably reduces spinal stiffness (Valentin et al. 2012). In addition, active neuromuscular control of the spine in vivo is crucial for spinal functionality (Reeves et al. 2007). Therefore, single or a limited number of FSUs in vitro that are devoid of muscle tissue are likely to display a considerably different biomechanical behaviour from the spine in vivo. Muscle parameters such as size, composition and activity can indicate the force-producing ability and functionality of a muscle that directly influences the stability of the skeletal system. If such parameters are dissimilar between man and sheep, caution may be warranted as outcomes from animal model studies (i.e. investigations into the biomechanical effects of spinal implants) might not accurately reflect the expected outcomes in a clinical population if based solely on comparative in vitro studies.

Axial spinal loading in quadrupeds is thought to be similar to that of bipeds despite the differences in spinal orientation (Smit, 2002). However, one might assume that the quadrupedal spine requires different stabilisation strategies to resist the extension moments exerted by gravity and the abdomen when compared with man. This may be achieved by specific muscle activation strategies, differences in passive spinal stiffness or differences in the force-generating capacity of trunk muscles. The volume of a muscle is related to its force-production ability, and this has been described for the trunk muscles in healthy human participants (Reid & Costigan, 1987) and in patients with low back pain (Lee et al. 2012). As a result, a comparison of the muscle volume ratios between bipeds and quadrupeds could provide an insight in the stabilisation strategies of the spine. To the authors’ knowledge, the relationship of trunk flexor to extensor muscle volume in sheep or other quadrupeds has not been quantified and compared with bipeds. Such information interpreted alongside other inter-species differences such as thoracic spinous processes length (Wilke et al. 1997), which influences extensor muscle moment arms, could provide a more detailed insight into the likely forces acting on the quadrupedal and bipedal spine in vivo.

In man, the lumbar paraspinal muscles multifidus (M) and erector spinae (ES) are frequently investigated for the purpose of understanding motor control of the spine, with a particular reference to patients with low back pain (Hodges et al. 2003; Kavcic et al. 2004). Composition of the M and ES muscles has been described from histological data (Mazis et al. 2009; Rossi et al. 2010). Although histological analysis is very useful to quantify muscle fibre typing and fat content (Boyd-Clark et al. 2001; Rossi et al. 2010), it is not easily implemented, or even available, in a clinical environment. Rather, imaging techniques provide a non-invasive method to obtain data on muscle morphometry. Magnetic resonance imaging (MRI) is considered the gold standard for the investigation of muscle morphometric parameters, such as cross-sectional area (CSA) and muscle fatty infiltrate (MFI; Akagi et al. 2010; Smith et al. 2014). These parameters are of research interest in patients with low back pain, as altered CSA and MFI have been explored with varying results (Mengiardi et al. 2006; Yanik et al. 2012; Fortin et al. 2014).

While fat content in sheep is analysed in the long spinal muscles for the purpose of meat production (McPhee et al. 2008), these data are obtained post-slaughter and are thus aquired using very different evaluation methods than the non-invasive imaging methods used in man. Only one study investigated ovine lumbar M atrophy and MFI using MRI, where the effects of different surgical approaches on M muscle damage were compared (Liu et al. 2010). In that study, no other muscles were investigated and the data were not compared with man. Nisolle et al. (2014) have provided a detailed description of ex vivo ovine lumbar spine anatomy using 1.5 Tesla MRI and compared these data with existing human literature. However, little is known about spinal muscle CSA and MFI in the ovine model in vivo and its potential relationship to human data. Comparing in vivo spinal muscle morphometry between ovine and man using MRI could provide an insight into the similarities or differences in spinal biomechanics between man and sheep in vivo. This knowledge could lead to a better understanding of whether the ovine spine is or should be considered a suitable model for human spinal research (Alini et al. 2008).

Furthermore, the spines of young, healthy sheep are typically utilised in animal model studies. This would, in the authors’ opinion, make it difficult to generalise the findings to older human patients with varying types of spinal disorders that require surgical correction (e.g. stenosis or radiculopathy). Therefore, the purpose of this study was to investigate lumbar spine muscle morphometry using MRI in different age groups of healthy human participants and sheep.

Materials and methods

Study population

Twenty-four healthy male and female human participants were recruited from a University population. Inclusion criteria were age of 18–25 years (young group) or 45–60 years (mature group), and a body mass index of 25 kg m−2 or less. Participants were excluded if they had current or a history of low back pain in the last 12 months, previous spinal surgery or fracture, neurological or orthopaedic disease, open abdominal surgery, or if they were determined by institutional standards to be unsuitable to undergo MRI (i.e. due to non-compatible magnetic resonance implants). All participants received a participant information sheet and gave written informed consent. Ethical approval for the human aspect of the study was granted by the Medical University of Vienna Ethics Committee (1609/2012).

For the ovine aspect of the study, 20 male and female Austrian Mountain sheep were allocated to either an immature group (6–9 months), young group (1–3 years) or a mature group (6–9 years). Due to the difficulties in obtaining enough sheep that could be allocated to one of two distinct age groups as was the case in the human participants, three age groups in the sheep were formed in an attempt to still allow potential age-effects to be explored. All sheep were assessed by an experienced orthopaedic veterinarian, and were deemed free from spinal or neurological disorders. The female sheep underwent an ultrasound investigation to exclude pregnancy. Ethical approval was granted from the Austrian Federal Ministry of Science and Research (13/10/97/2011).

Data collection

Human participants underwent MRI at a dedicated human imaging facility. Axial T1-weighted MRI of the lumbar spine were obtained (1.5T Philips Achieva, Best, The Netherlands) using a body coil and a coil integrated into the patient table. The participants were positioned in supine. The following parameters were used: slice thickness 10 mm; slice gap 1 mm; repetition time (TR) 9.3 ms; echo time (TE) 4.6 ms; flip angle 15 °; field of view (FOV) 350 mm; rectangular FOVy 350 × 273 mm; voxel size 1 × 1 × 10 mm. An anterior–posterior phase direction was used.

Sheep were anaesthetised for the imaging procedure of computed tomography (CT) followed by MRI. CT (Siemens Somatom Emotion 16, Erlangen, Germany) was performed prior to MRI as per University departmental protocol, to ensure they did not have any metal in their bodies (i.e. ingested foreign objects) that could move when introduced to a magnetic field. Both CT and MRI were carried out at a veterinary university. Premedication for anaesthesia consisted of intravenous butorphanol (0.1 mg kg−1) and xylazine (0.15 mg kg−1). After sedation was achieved, general anaesthesia was induced with ketamine (5 mg mL−1) and maintained with inhalation of vaporised sevoflurane. Sheep were placed in dorsal recumbency for CT and MRI data collection. At the end of the imaging procedure, sheep were placed in lateral recumbency until they showed active breathing movements, at which time they were extubated and the orogastric tube was removed. Following extubation, sheep were placed in sternal recumbency in a stable and were continuously observed until they were stable at walk.

For sheep MRI data collection, lumbar spine T1-weighted MRI were collected using a 1.5-Telsa MRI (Siemens Magnetom Esprit, Erlangen, Germany) with a body coil and a Spine Matrix Coil integrated into the patient table. The following parameters were used: slice thickness 4 mm; slice gap 2 mm; flip angle180°, TR 448ms, TE 11ms, FOV = 450mm, rectangular FOV × 218 × 450mm, voxel size 1.8 × 2.1 × 4mm. It was not always possible to obtain images of the entire lumbar spine in sheep in one package due to increased specific absorption rates, therefore axial images of the lumbar spine were obtained in up to three packages. An anterior–posterior phase direction was used. Where this was not possible due to breathing artefacts, a left to right phase direction was used.

Data analysis

analyzedirect software (Mayo Clinic, Rochester, MN, USA; Version 11.0) was used for data analysis. Axial slices from the caudal aspect of the last lumbar vertebral body (L5 in humans and L7 or L6 in sheep) to the cranial aspect of the first lumbar vertebral body were included. The left and right ES in human participants and longissimus dorsi in sheep (for the purpose of this study also referred to as ES), M and psoas (PS) were manually traced and the region of interest quantified (Fig.1). The ratio of flexor to extensor muscle volume (FE ratio) was calculated by dividing the combined left and right PS volume by the combined left and right ES and M volume. The ratio of M to ES volume (MES ratio) was calculated by dividing the combined left and right M volume by the combined left and right ES volume.

Figure 1.

Figure 1

Open in a new tab

Regions of interest of the muscles psoas (PS; green), erector spinae (ES; yellow) and multifidus (M; blue) in a human (top) and sheep (bottom) axial image at the level of L4 in both species.

For MFI, mean pixel intensity (MPI) from the left and right sides of each muscle across all included slices was reported as a percentage of MPI relative to an area of intermuscular fat from the left or right side of the body, taken from an axial slice from the last lumbar vertebrae between the PS and quadratus lumborum (Fig.2).

Figure 2.

Figure 2

Open in a new tab

Regions of interest of intermuscular fat from the left (blue) and right (red) sides of the last lumbar vertebrae in a human (top) and sheep (bottom) axial image.

All data analysis was conducted twice by the same assessor to obtain a measure of intra-rater reliability.

Statistical analysis

Intra-rater reliability of the human data in this study was determined using an average measures interclass correlation (ICC3,1) with absolute agreement. The ICC3,1 is used when all data are rated by the same assessors who are assumed to be the entire population of assessors (Elliott et al. 2013). The same assessor also performed this for the sheep data.

Distribution of the data was analysed before a 2 × 3 independent factorial anova was performed with species and age group as independent variables. Where significant main effects for species, age group, and interactions of species and age group were found, estimated marginal means using a Bonferroni post hoc test, chosen to keep the family-wise alpha at 0.05, were reported. Where homogeneity of variance had been violated, the data were transformed. If following transformation heterogeneity of variance remained, species were directly compared ignoring age group using a Mann–Whitney U-test. For all statistical analysis, spss (IBM; version 19) was used.

Results

All 24 human participants and 17 of the 20 sheep completed the study. Three sheep were excluded due to pregnancy, resulting in six sheep in the immature group, six in the young group and five in the mature group. No peri- or post-anaesthesia complications occurred in any sheep except one, which required a tracheotomy after biting off parts of the intubation tube on sudden awakening after anaesthesia. Treatment consisted of daily wound cleaning, antibiotics and anti-inflammatories (carprofen 1.5 mg kg−1 per day for 2 days and cefquinome 1 mg kg−1 per day for 2 days). Further recovery of this sheep was uneventful. For right-sided MFI, data from three further sheep were excluded (middle group n = 2, mature group n = 1), as they did not have sufficient right-sided intermuscular fat to determine MFI.

Intra-rater reliability data are presented in Tables1 and 2. Tables3 and 4 show the mean and standard error of the first and second analysis of muscle volume and MFI for combined age groups of humans and sheep.

Table 1.

ICC3,1 and CI for PS, ES and M muscle volume of combined age groups in human participants and sheep

Human Sheep
ICC CI ICC CI
Lower Upper Lower Upper
PS
 Left 0.993 0.984 0.997 0.997 0.991 0.999
 Right 0.996 0.991 0.998 0.997 0.989 0.999
ES
 Left 0.991 0.977 0.996 0.997 0.991 0.999
 Right 0.991 0.980 0.996 0.998 0.991 0.999
M
 Left 0.882 0.232 0.965 0.942 0.840 0.979
 Right 0.866 0.467 0.953 0.939 0.832 0.978
Open in a new tab

CI, confidence interval; ES, erector spinae; ICC, intraclass correlation coefficient; M, multifidus; PS, psoas.

Table 2.

ICC3,1 and CI for PS, ES and M MFI of combined age groups in human participants and sheep

Human Sheep
ICC CI ICC CI
Lower Upper Lower Upper
PS
 Left 0.944 0.870 0.976 0.996 0.988 0.999
 Right 0.948 0.871 0.978 0.998 0.992 0.999
ES
 Left 0.927 0.830 0.968 0.999 0.997 1.000
 Right 0.952 0.787 0.984 0.998 0.993 0.999
M
 Left 0.918 0.813 0.964 0.995 0.984 0.999
 Right 0.962 0.912 0.983 0.998 0.994 0.999
Open in a new tab

CI, confidence interval; ES, erector spinae; ICC, intraclass correlation coefficient; M, multifidus; PS, psoas.

Table 3.

Mean (standard error) of analysis 1 and 2 for PS, ES and M muscle volume (in mm3 × 103) of combined age groups in human participants and sheep

Human Sheep
Analysis 1 (mm3 × 103) Analysis 2 (mm3 × 103) Analysis 1 (mm3 × 103) Analysis 2 (mm3 × 103)
PS
 Left 155.70 (11.967) 154.79 (11.23) 136.34 (10.58) 138.52 (10.67)
 Right 149.64 (10.57) 150.14 (10.71) 131.01 (11.08) 133.59 (11.08)
ES
 Left 290.82 (17.09) 297.00 (18.64) 274.66 (24.78) 269.45 (24.78)
 Right 282.17 (16.14) 283.39 (17.79) 268.60 (22.66) 263.44 (23.03)
M
 Left 104.31 (5.61) 91.94 (4.13) 41.06 (2.87) 41.90 (3.35)
 Right 106.24 (5.24) 95.80 (3.97) 44.34 (2.97) 43.90 (3.19)
Open in a new tab

ES, erector spinae; M, multifidus; PS, psoas.

Table 4.

Mean (standard error) of analysis 1 and 2 for PS, ES and M MFI (reported as a percentage) of combined age groups in human participants and sheep

Human Sheep
Analysis 1 (%) Analysis 2 (%) Analysis 1 (%) Analysis 2 (%)
PS
 Left 40.44 (0.78) 40.44 (0.73) 20.13 (1.18) 20.23 (1.18)
 Right 32.39 (0.50) 31.93 (0.54) 17.80 (0.84) 17.94 (0.80)
ES
 Left 45.08 (0.70) 45.23 (0.66) 30.08 (1.61) 30.05 (1.66)
 Right 36.25 (0.60) 35.46 (0.60) 26.40 (1.22) 26.63 (1.26)
M
 Left 45.97 (0.79) 46.43 (0.74) 33.94 (1.57) 34.36 (1.55)
 Right 38.60 (0.84) 38.42 (0.74) 32.25 (1.69) 32.47 (1.63)
Open in a new tab

ES, erector spinae; M, multifidus; PS, psoas.

The MES ratio had a significant main effect for species (P < 0.001), with estimated marginal means of human 0.37 ± 0.05 and sheep 0.16 ± 0.02 (P < 0.001). The FE ratio did not have a significant main effect for species (human 0.39 ± 0.08, sheep 0.43 ± 0.05, P = 0.06). There were no significant main effects for age or interaction for either ratio.

Table 5.

Mean (standard deviations) of muscle ratios (FE; MES) and MFI of PS, ES and M for humans and sheep in their respective age categories

Human Sheep
Young Mature Immature Young Mature
Ratios
 FE 0.42 (0.07) 0.36 (0.09) 0.38 (0.04) 0.45 (0.07) 0.46 (0.03)
 MES 0.39 (0.04) 0.36 (0.05) 0.17 (0.01) 0.16 (0.04) 0.15 (0.02)
MFI (%)
 PS left 37.65 (4.11) 36.73 (4.56) 36.75 (4.66) 32.45 (3.83) 37.72 (7.16)
 PS right 33.20 (2.47) 32.42 (2.13) 31.87 (4.02) 26.12 (5.94) 36.44 (6.77)
 M left 42.46 (4.47) 42.11 (4.74) 52.01 (4.81) 54.90 (5.10) 76.90 (17.49)
 M right 38.28 (2.79) 39.75 (3.20) 48.61 (5.44) 55.68 (17.63) 73.34 (9.23)
 ES left 42.56 (5.30) 40.49 (3.97) 50.15 (7.20) 49.67 (4.96) 60.17 (7.40)
 ES right 35.97 (2.48) 37.40 (2.01) 43.41 (5.51) 42.53 (8.92) 54.64 (7.19)
Open in a new tab

ES, erector spinae; FE, ratio of flexor to extensor muscle volume; M, multifidus; MES, ratio of multifidus to erector spinae volume; MFI, muscle fatty infiltrate; PS, psoas.

Heterogeneity of variance occurred in four out of six muscles for MFI, therefore between-species MFI comparisons were made without age sub-groups. Sheep had significantly greater MFI than humans for M and ES (all P < 0.001), whereas PS was not significantly different (P < 0.05; Tables6, Fig.3).

Table 6.

Median values of MFI of human participants and sheep (in percentage) for the muscles PS, M and ES

Muscle Human MFI (%) Sheep MFI (%)
PS left 36.89 33.67
PS right 32.78 30.09
M left 40.64a 53.81a
M right 39.17b 51.33b
ES left 40.86c 51.29c
ES right 35.93d 44.38d
Open in a new tab

Pairs of superscripts indicate significant differences between the species (all P < 0.001).

Figure 3.

Figure 3

Open in a new tab

Median muscle fatty infiltrate (MFI) as a percentage in humans (blue) and sheep (red) for the muscles psoas (PS), erector spinae (ES) and multifidus (M). The double asterisks indicate significant differences between species for that particular muscle and side (all P < 0.001).

Discussion

This study uniquely compared differences in lumbar spine muscle volume and MFI between man and sheep. A significantly smaller M relative to ES was found in sheep in comparison to the human participants. The physiological CSA of a muscle and its force-producing ability are intrinsically linked (Asaka et al. 2010; Lee et al. 2012). In the human spine, M is important for FSU stability (Kavcic et al. 2004). Although the relevance of M in sheep spine stability has not been investigated previously, a significant increase in thoracolumbar M thickness in healthy horses following a 3-month exercise protocol composed of dynamic head–neck and trunk mobilisation and stabilisation exercises has been observed (Stubbs et al. 2011). This suggests the possibility of a similar stabilising function of M in quadrupeds. Therefore, a relative reduction in M in sheep compared with man might indicate that sheep do not require the same degree of force production for segmental stability despite the ability of this muscle to perform a similar function, as stiffness provided by the skeletal system itself (or other muscles not investigated in this study) may already be sufficient. Furthermore, fibre typing of the extensor muscles should also be considered, as this will influence muscle function. In sheep, longissmus dorsi post-slaughter in cross-breed lambs has been reported to contain 11.7% type I fibres (Carpenter et al. 1996), although fibre composition of M specifically has not been reported in sheep. In cadaveric human deep and superficial M and ES, the percentage of type I muscle fibres was reported to range from 57% to 63% (Sirca & Kostevc, 1985). Although caution should be warranted when data from very young sheep are compared with those of mature human adults, in addition to the differences in extensor muscle sampling locations, the smaller proportion of slow oxidative fibres of the spinal extensors in sheep and the relative smaller M volume suggest a lesser need of M to provide spinal stability. Alternatively, it may be that sheep require a greater volume of ES relative to M in comparison with humans, as ES is important in controlling the lateral bend of the spine during quadrupedal locomotion (Licka et al. 2009). Regardless, the difference in ES to M volume is indicative of different biomechanical and functional needs, which is relevant for ovine model to human spine translation.

Human participants in this study had significantly less MFI in all dorsal muscles compared with sheep, although ventral muscle MFI was similar. An increase in MFI can lead to a reduction in muscle functionality (Ali et al. 2011), and greater MFI content in lumbar muscles has been identified in patients with low back pain (Mengiardi et al. 2006). Furthermore, increased trunk muscle fat has been shown to reduce functional capacity in older adults (Hicks et al. 2005). As the human participants and sheep in this study were considered free from spinal pain, it can be assumed that the greater amount of MFI in sheep is inherent to this species and not a pathological process. Longissimus dorsi fat content is important for meat palatability (Pannier et al. 2014), therefore greater MFI in dorsal musculature in sheep may be due to breeding selection for production. Disparities in MFI between healthy humans and sheep found in this study are thus relevant for translational research, as it may further indicate different muscle recruitment needs between species, making it difficult to generalise.

It should be noted that MFI outcomes were higher for both species compared with those published in other studies, which is likely to be due to methodological differences. Mengiardi et al. (2006) reported MFI in healthy human participants of 14.5% for M and 23% for ES measured at L4–5 obtained using MR spectroscopy, whereas Pezolato et al. (2012) found MFI of 1.2–11.5% in M in L1–S1 segments using T2-weighted MRI. To the authors’ knowledge, detailed fat quantification in sheep spinal muscles has not been investigated using similar measurement tools as used in man. Even though one study reported MFI in sheep using MRI and histology (Liu et al. 2010), only one control sheep was included and a categorical approach used to quantify M MFI. Therefore, comparisons of sheep muscle MFI data obtained in the present study cannot be made to other published data. However, fat content in longissimus dorsi was reported to range from 3.8 to 4.8% in Merino cross-breed sheep for meat production using near-infra-red spectroscopy (McPhee et al. 2008; Pannier et al. 2014). Validation studies are recommended where sheep MFI obtained from similar MRI sequences (Reeder et al. 2012) are compared with those obtained from histology (Gaeta et al. 2011). This has recently been conducted in porcine and rabbit models, where spinal muscle fat obtained from a two-point Dixon at two different field strengths was comparable to biopsy (Smith et al. 2014).

Similar FE ratios between species that were not influenced by age were observed in the present study. Previous studies have shown that sheep PS, similar to the spinal extensors, contains a smaller proportion of type I muscle fibres compared with those muscles in humans. Specifically, PS samples obtained from young human cadavers contained 40.09–40.21% type I fibres (Arbanas et al. 2009), whereas sheep PS obtained post-slaughter contained 27.1–30.8% type I fibres (Suzuki & Tamate, 1988). As a result, it is anticipated that the smaller proportion of type I fibres in both the flexors and extensors in sheep identified in the previous studies would not considerably influence the FE ratio findings of the present study. Therefore, this would suggest that sheep do not require greater PS volume relative to the dorsal musculature to counteract the effects of gravity. This might be due to the PS muscle having several other primary functions, such as creating hip flexion and hip moment in addition to stabilising the lumbar spine and pelvis complex in both humans (Yoshio et al. 2002) and in a quadrupedal model (Schilling et al. 2005). Therefore, it may be that other muscles in the sheep such as the abdominals play a greater role in providing support against extension moments. Even though abdominal muscle volume could be quantified in the human participants, this was not possible in the sheep due to the large rumen in the field of view. Another explanation for the similarity in FE ratios despite differences in spinal orientation may be that in vivo sheep thoracolumbar spines are inherently stiffer than the human spine (unpublished communication). This requires further investigation.

There are acknowledged limitations in this study. Initially, it was planned to have two age groups (young and mature) for both species, although obtaining healthy sheep with specific ages was difficult therefore three age groups were created. Furthermore, the average life expectancy of a sheep has not been reported and it will also be skewed by the majority of animals being bred for meat production with a resultant short life span, therefore selecting age groups that are representatively similar in humans and sheep poses a challenge. As a factorial anova could not be performed on the MFI data, species were directly compared without considering age. The effect of age could have had an influence on the MFI outcomes and this should be explored in future studies. Different MRI scanners were used for the human participants and sheep due to the logistics of scanning an animal in a human facility and vice versa. However, it has been shown that muscle data obtained from different species, different scanners and different field strengths provide comparable findings (Smith et al. 2014). Regarding the imaging protocols, differences between the human and sheep MRI should be acknowledged. In all human participants and 13 sheep, scans were obtained in an anterior to posterior phase direction, although in four sheep a left to right phase direction was used to minimise movement artefacts. This may have created phase-dependent signal intensity changes. Further, there were differences in slice thickness and voxel size between the human participants and sheep. While this effect is thought to be minimal, the data may be prone to differences in spatial resolution and signal to noise ratio, and may contain partial volumes, thus some caution may be warranted in interpretation of the data. Increasing scanning averages would have led to higher quality images, which is particularly important for diagnostic imaging; however, this would have required longer scanning times that may have been problematic for the anaesthetised sheep for the duration of the imaging protocol. Furthermore, costs were prohibitive. Also, there were apparent differences in signal intensity between the dorsal subcutaneous and intermuscular fat in sheep, whereas in the human participants, fat signal intensity appeared more consistent throughout the images. This may have had an effect on sheep MFI outcomes and further studies should identify which fat source is optimal for T1-weighed MFI calculation and these data compared against histology. Alternatively, different sequences could be obtained such as the Dixon method (Dixon, 1984) to identify the most reliable method for muscle fat/water quantification within different species. Lastly, the possible effect of positioning on the human participants and sheep for MRI should be considered. Stemper et al. (2010) demonstrated in their MRI study on cervical spine muscle volume that only the sternocleidomastoid muscle CSA was influenced by scanning posture of the participants (supine vs. upright). Even though both the human participants and the sheep were positioned supine/dorsal, this position may have influenced muscle CSA. However, as muscle volume values were obtained across the entire lumbar spine, it is likely that any changes in CSA from individual axial images caused by being positioned supine will have balanced out across the imaging data set.

In conclusion, the results indicate that human and sheep spinal muscles differ in morphology, which may have biomechanical implications. This should be considered when in vitro ovine model data are interpreted for translation to the human spine.

Acknowledgments

This study was made possible with financial support from the Austrian Science Fund (FWF, grant number P24020). The authors would like to thank Professor Sibylle Kneissl and Sabine Dengg at the University of Veterinary Medicine, Vienna, Austria, and Dr Friedrich Vorbeck and staff at the Diagnosezentrum Donaustadt, Vienna, Austria, for their assistance in MRI data collection.

Conflicts of interest

The authors have no conflicts of interest to declare.

References

  1. Akagi R, Takaj Y, Kato E, et al. Development of an equation to predict muscle volume of elbow flexors for men and women with a wide age range. Eur J Appl Physiol. 2010;108:689–694. doi: 10.1007/s00421-009-1269-1. [DOI] [PubMed] [Google Scholar]
  2. Ali I, Ulbricht C, McGregor AH. Degeneration of the extensor muscle group in a surgical low back and leg pain population. J Back Musculoskelet Rehabil. 2011;24:23–30. doi: 10.3233/BMR-2011-0271. [DOI] [PubMed] [Google Scholar]
  3. Alini M, Eisenstein SM, Ito K, et al. Are animal models useful for studying human disc disorders/degeneration? Eur Spine J. 2008;17:2–19. doi: 10.1007/s00586-007-0414-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Arbanas J, Klasan GS, Nikolic M, et al. Fibre type composition of the human psoas major muscle with regard to the level of its origin. J Anat. 2009;215:636–641. doi: 10.1111/j.1469-7580.2009.01155.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Asaka M, Usui C, Ohta M, et al. Elderly oarsmen have larger trunk and thigh muscles and greater strength than age-matched untrained men. Eur J Appl Physiol. 2010;108:1239–1245. doi: 10.1007/s00421-009-1337-6. [DOI] [PubMed] [Google Scholar]
  6. Boyd-Clark LC, Briggs CA, Galea MP. Comparative histochemical composition of muscle fibres in a pre- and a postvertebral muscle of the cervical spine. J Anat. 2001;199:709–716. doi: 10.1046/j.1469-7580.2001.19960709.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Carpenter CE, Rice OD, Cockett NE, et al. Histology and composition of muscles from normal and callipyge lambs. J Anim Sci. 1996;74:388–393. doi: 10.2527/1996.742388x. [DOI] [PubMed] [Google Scholar]
  8. Dickey JP, Kerr DJ. Effect of specimen length: are the mechanics of individual motion segments comparable in functional spinal units and multisegment specimens? Med Eng Phys. 2003;25:221–227. doi: 10.1016/s1350-4533(02)00152-2. [DOI] [PubMed] [Google Scholar]
  9. Dixon W. Simple proton spectroscopic imaging. Radiology. 1984;153:189–194. doi: 10.1148/radiology.153.1.6089263. [DOI] [PubMed] [Google Scholar]
  10. Elliott JM, Walton DM, Rademaker A, et al. Quantification of cervical spine muscle fat: a comparison between T1-weighted and multi-echo gradient echo imaging using a variable projection algorithm (VARPRO) BMC Med Imaging. 2013;13:30. doi: 10.1186/1471-2342-13-30. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Fortin M, Videman T, Gibbons LE, et al. Paraspinal muscle morphology and composition: a 15-yr longitudinal magnetic resonance imaging study. Med Sci Sports Exerc. 2014;46:893–901. doi: 10.1249/MSS.0000000000000179. [DOI] [PubMed] [Google Scholar]
  12. Gaeta M, Scribano E, Mileto A, et al. Muscle fat fraction in neuromuscular disorders: dual-echo dual-flip-angle spoiled gradient-recalled MR imaging technique for quantification – a feasibility study. Radiology. 2011;259:487–494. doi: 10.1148/radiol.10101108. [DOI] [PubMed] [Google Scholar]
  13. Hasler C, Sprecher CM, Milz S. Comparison of the immature sheep spine and the growing human spine. Spondylitic database for growth modulating research. Spine. 2010;35:E1262–E1272. doi: 10.1097/BRS.0b013e3181e24196. [DOI] [PubMed] [Google Scholar]
  14. Hicks GE, Simonsick EM, Harris TB, et al. Trunk muscle composition as a predictor of reduced functional capacity in the health, aging and body composition study: the moderating role of back pain. J Gerontol A Biol Sci Med Sci. 2005;60A:1420–1424. doi: 10.1093/gerona/60.11.1420. [DOI] [PubMed] [Google Scholar]
  15. Hodges P, Moseley GL, Gabrielsson A, et al. Experimental muscle pain changes feedforward postural responses of the trunk muscles. Exp Brain Res. 2003;151:262–271. doi: 10.1007/s00221-003-1457-x. [DOI] [PubMed] [Google Scholar]
  16. Kavcic N, Grenier S, McGill SM. Determining the stabilizing role of individual torso muscles during rehabilitation exercises. Spine. 2004;29:1254–1265. doi: 10.1097/00007632-200406010-00016. [DOI] [PubMed] [Google Scholar]
  17. Leckie AE, Akens MK, Woodhouse KA, et al. Evaluation of thiol-modified hyaluronan and elastin-like polypeptide composite augmentation in early-stage disc degeneration: comparing 2 minimally invasive techniques. Spine. 2012;37:1296–1303. doi: 10.1097/BRS.0b013e318266ecea. [DOI] [PubMed] [Google Scholar]
  18. Lee HJ, Lim WH, Park JW, et al. The relationship between cross sectional area and strength of back muscles in patients with chronic low back pain. Ann Rehabil Med. 2012;36:173–181. doi: 10.5535/arm.2012.36.2.173. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Licka T, Frey A, Peham C. Electromyographic activity of the longissimus dorsi muscles in horses when walking on a treadmill. Vet J. 2009;180:71–76. doi: 10.1016/j.tvjl.2007.11.001. [DOI] [PubMed] [Google Scholar]
  20. Liu X, Wang Y, Wu X, et al. Impact of surgical approaches on the lumbar multifidus muscle: an experimental study using sheep as models. J Neurosurg Spine. 2010;12:570–576. doi: 10.3171/2009.11.SPINE09174. [DOI] [PubMed] [Google Scholar]
  21. Mazis N, Papachristou DJ, Zouboulis P, et al. The effect of different physical activity levels on muscle fiber size and type distribution of lumbar multifidus. A biopsy study on low back pain patient groups and healthy control subjects. Eur J Phys Rehabil Med. 2009;45:459–467. [PubMed] [Google Scholar]
  22. McPhee MJ, Hopkins DL, Pethick DW. Intramuscular fat levels in sheep muscle during growth. Aust J Exp Agr. 2008;48:904–909. [Google Scholar]
  23. Mengiardi B, Schmid MR, Boos N. Fat content of lumbar paraspinal muscles in patients with chronic low back pain and in asymptomatic volunteers: quantification with MR spectroscopy. Radiology. 2006;240:786–792. doi: 10.1148/radiol.2403050820. [DOI] [PubMed] [Google Scholar]
  24. Nisolle JF, Wang XQ, Squélart M, et al. Magnetic resonance imaging (MRI) anatomy of the ovine lumbar spine. Anat Histol Embryol. 2014;43:203–209. doi: 10.1111/ahe.12063. [DOI] [PubMed] [Google Scholar]
  25. Pannier L, Pethick DW, Geesink GH. Intramuscular fat in the longissimus muscle is reduced in lambs from sires selected for leanness. Meat Sci. 2014;96:1068–1075. doi: 10.1016/j.meatsci.2013.06.014. [DOI] [PubMed] [Google Scholar]
  26. Pezolato A, de Vasconcelos EE, Defino HLAD, et al. Fat infiltration in the lumbar multifidus and erector spinae muscles in subjects with sway-back posture. Eur Spine J. 2012;21:2158–2164. doi: 10.1007/s00586-012-2286-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Reeder SB, Hu HH, Sirlin CB. Proton density fat-fraction: a standardized mr-based biomarker of tissue fat concentration. J Magn Reson Imaging. 2012;36:1011–1014. doi: 10.1002/jmri.23741. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Reeves NP, Narendra KS, Cholewicki J. Spine stability: the six blind men and the elephant. Clin Biomech. 2007;22:266–274. doi: 10.1016/j.clinbiomech.2006.11.011. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Reid JG, Costigan PA. Trunk muscle balance and muscular force. Spine. 1987;12:783–786. doi: 10.1097/00007632-198710000-00013. [DOI] [PubMed] [Google Scholar]
  30. Rossi A, Zoico E, Goodpaster BH, et al. Quantification of intermuscular adipose tissue in the erector spinae muscle by MRI: agreement with histological evaluation. Obesity. 2010;18:2379–2384. doi: 10.1038/oby.2010.48. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Schilling N, Arnold D, Wagner H, et al. Evolutionary aspects and muscular properties of the trunk – implications for human low back pain. Pathophysiology. 2005;12:233–242. doi: 10.1016/j.pathophys.2005.09.005. [DOI] [PubMed] [Google Scholar]
  32. Sheng SR, Wang X-Y, Xu H-Z, et al. Anatomy of large animal spines and its comparison to the human spine: a systematic review. Eur Spine J. 2010;19:46–56. doi: 10.1007/s00586-009-1192-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Sirca A, Kostevc V. The fibre composition of thoracic and lumbar paravertebral muscles in man. J Anat. 1985;141:131–137. [PMC free article] [PubMed] [Google Scholar]
  34. Smit TH. The use of a quadruped as an in vivo model for the study of the spine – biomechanical considerations. Eur Spine J. 2002;11:137–144. doi: 10.1007/s005860100346. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Smith AV, Parrish TB, Abbott R, et al. Muscle-fat magnetic resonance imaging: 1.5 Tesla and 3.0 Tesla versus histology. Muscle Nerve. 2014;50:170–176. doi: 10.1002/mus.24255. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Stemper BD, Baisden JL, Yoganandan N, et al. Determination of normative neck muscle morphometry using upright MRI with comparison to supine data. Aviat Space Environ Med. 2010;81:878–882. doi: 10.3357/asem.2758.2010. [DOI] [PubMed] [Google Scholar]
  37. Stubbs NC, Kaiser LJ, Hauptman J, et al. Dynamic mobilisation exercises increase cross sectional area of musculus multifidus. Equine Vet J. 2011;43:522–529. doi: 10.1111/j.2042-3306.2010.00322.x. [DOI] [PubMed] [Google Scholar]
  38. Suzuki A, Tamate H. Distribution of myofiber types in the hip and thigh musculature of sheep. Anat Rec. 1988;221:494–502. doi: 10.1002/ar.1092210106. [DOI] [PubMed] [Google Scholar]
  39. Valentin S, Groesel M, Licka T. The presence of long spinal muscles increases stiffness and hysteresis of the caprine spine in-vitro. J Biomech. 2012;45:2506–2512. doi: 10.1016/j.jbiomech.2012.07.022. [DOI] [PubMed] [Google Scholar]
  40. Veres SP, Robertson PA, Broom ND. The influence of torsion on disc herniation when combined with flexion. Eur Spine J. 2010;19:1468–1478. doi: 10.1007/s00586-010-1383-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Wilke HJ, Kettler A, Wenger KH, et al. Anatomy of the sheep spine and its comparison to the human spine. Anat Rec. 1997;247:542–555. doi: 10.1002/(SICI)1097-0185(199704)247:4<542::AID-AR13>3.0.CO;2-P. [DOI] [PubMed] [Google Scholar]
  42. Yanik B, Keyik B, Conkbayir I. Fatty degeneration of multifidus muscle in patients with chronic low back pain and in asymptomatic volunteers: quantification with chemical shift magnetic resonance imaging. Skeletal Radiol. 2012;42:771–778. doi: 10.1007/s00256-012-1545-8. [DOI] [PubMed] [Google Scholar]
  43. Yoshio M, Murakami G, Sato T, et al. The function of the psoas major muscle: passive kinetics and morphological studies using donated cadavers. J Orthop Sci. 2002;7:199–207. doi: 10.1007/s007760200034. [DOI] [PubMed] [Google Scholar]

Articles from Journal of Anatomy are provided here courtesy of Anatomical Society of Great Britain and Ireland

RESOURCES