THE ANATOMICAL RECORD 300:300–308 (2017)
Radiographic and Magnetic Resonance Imaging Identification of Thoracolumbar Spine Variants with Implications for the Positioning of the Conus Medullaris in Rhesus Macaques MARCUS OHLSSON,1,2 JAIME H. NIETO,1 KARI L. CHRISTE,3 J. PABLO VILLABLANCA,4 AND LEIF A. HAVTON1* 1 Department of Neurology, David Geffen School of Medicine at UCLA, Los Angeles, California 2 Department of Clinical Neuroscience, Divisions of Neurosurgery and Neuroradiology, Karolinska Institute, Stockholm, Sweden 3 California National Primate Research Center, UC Davis, Davis, California 4 Department of Radiology, David Geffen School of Medicine at UCLA, Los Angeles, California
ABSTRACT The anatomy of the vertebral column in mammals may differ between species and between subjects of the same species, especially with regards to the composition of the thoracolumbar spine. We investigated, using several noninvasive imaging techniques, the thoracolumbar spine of a total of 44 adult rhesus macaques of both genders. Radiographic examination of the vertebral column showed a predominant spine phenotype with 12 ribbearing thoracic vertebrae and 7 lumbar vertebrae without ribs in 82% of subjects, whereas a subset of subjects demonstrated 13 rib-bearing thoracic vertebrae and 6 lumbar vertebrae without ribs. Computer tomography studies of the thoraco-lumbar spine in two cases with a pair of supernumerary ribs showed facet joints between the most caudal pair of ribs and the associated vertebra, supporting a thoracic phenotype. Magnetic resonance imaging (MRI) studies were used to determine the relationship between the lumbosacral spinal cord and the vertebral column. The length of the conus medullaris portion of the spinal cord was 1.5 6 0.3 vertebral units, and its rostral and caudal positions in the spinal canal were at 2.0 6 0.3 and 3.6 6 0.4 vertebral units below the thoracolumbar junction, respectively (n 5 44). The presence of a set of supernumerary ribs did not affect the length or craniocaudal position of the conus medullaris, and subjects with13 rib-bearing vertebrae may from a functional or spine surgical perspective be considered as exhibiting12 thoracic vertebrae and an L1 verteC 2016 Wiley Periodicals, Inc. bra with ribs. Anat Rec, 300:300–308, 2017. V
Key words: nonhuman primate; computed tomography; noninvasive imaging; supernumerary rib; lumbar rib
Grant sponsor: Translational Research Partnership Grant, Department of Defense Congressionally Directed Medical Research Program on Spinal Cord Injury; Grant number: SC090273; Grant Sponsor: National Institutes of Health Office of the Director; Grant number: P51 OD011107; Grant Sponsors: Dr. Miriam and Sheldon G. Adelson Medical Research Foundation, Karolinska Institutet and the Swedish Society of Medicine.
C 2016 WILEY PERIODICALS, INC. V
*Correspondence to: Leif A. Havton, Department of Neurology, David Geffen School of Medicine at UCLA, Los Angeles, California. E-mail:
[email protected] Received 31 October 2015; Accepted 23 December 2015. DOI 10.1002/ar.23495 Published online 12 October 2016 in Wiley Online Library (wileyonlinelibrary.com).
RADIOGRAPHIC AND MAGNETIC RESONANCE IMAGING
In recent years, several spinal cord injury and repair models have been developed in nonhuman primates. These models represent important tools for continued progress in translational research and include injuries to the spinal cord in rhesus macaques at the cervical (Rosenzweig et al., 2010; Nout et al., 2012b), thoracic (Courtine et al., 2005) and lumbosacral levels (Ohlsson et al., 2013). Comprehensive and clinically relevant outcome measures allow for the evaluation of spontaneous plasticity of spinal cord circuitries and for preclinical testing of new treatment strategies in these nonhuman primate models (Nout et al., 2012a; Nielson et al., 2015). The development and continued refinement of translational research models for neural repair studies after a spinal cord injury requires both knowledge of the normal spine anatomy and an understanding of the human condition (Carlstedt and Havton, 2012; Dobkin and Havton, 2012). However, the surgical anatomy of the spine may differ extensively between species, and inter-species variability is especially prominent with regards to the composition of the thoracolumbar spine in primates. For instance, the human spine normally includes 12 thoracic and 5 lumbar vertebrae, whereas chimpanzees (Pan troglodytes), bonobos (Pan paniscus), and the Western gorilla (Gorilla gorilla) typically have 13 thoracic and 4 lumbar vertebrae (Williams and Russo, 2015). In contrast, old world monkeys, including macaques, have a longer back and commonly show 13 thoracic and 6 lumbar vertebrae or 12 thoracic and 7 lumbar vertebrae (Aimi, 1994; Williams and Russo, 2015). This often varied anatomy of the spine between and within primate species represents a particular challenge that needs to be taken into consideration when identifying individual lumbosacral spinal cord segments intraoperatively and developing injury models that target the conus medullaris and cauda equina portions of the spinal cord in rhesus macaques (Ohlsson et al., 2013). Although a basic understanding exists with regards to the thoracolumbar spine anatomy in several primate species, it is not well understood whether the varied length of the lower back may influence, for instance, the rostro-caudal positioning of the lumbosacral spinal cord within the spinal canal. Therefore, the overall goal for the present study was to determine the normal anatomy of the thoracolumbar spine and its relationship with the lumbosacral portion of the spinal cord in neurologically intact rhesus macaques. For this purpose, a combination of plain radiographs, computed tomography (CT), and magnetic resonance imaging (MRI) was used. An additional goal was to provide investigators with the anatomical tools to perform more precise and less invasive surgical procedures involving the lumbar spine and the lumbosacral portion of the spinal cord in rhesus macaques. The added ability to perform minimally invasive spine procedures in nonhuman primates, supported by detailed anatomical data, represents a significant refinement of existing experimental protocols and is in accordance with the guiding principles for the ethical use of animals in medical research (Russell and Birch, 1959).
MATERIALS AND METHODS All animal procedures were performed at the California National Primate Research Center at UC Davis, Davis
301
CA. The facility is accredited by the Association for Assessment and Accreditation of Laboratory Animal Care (AAALAC) International, and all subjects were maintained in accordance with the provisions provided by the Animal Welfare Act (as Amended 2007.7 USC S2131– 2159) and the Guide for Care and Use of Laboratory Animals (National Institutes of Health Publications No. 8623, revised 1985). All animal procedures were approved by the Institutional Animal Care and Use Committee at UC Davis. Noninvasive imaging data were collected from a total of 44 consecutive rhesus macaques undergoing preoperative clinical and imaging evaluation for participation in neural repair studies after lumbosacral spinal cord and nerve root injuries. The series included 33 adult female and 11 male rhesus monkeys (Macaca mulatta). The subjects were on average 9 years 4 months of age (5 years and 7 months to 15 years and 3 months) and weighed on average 9.3 kg (5.5–15.5 kg) (n 5 44). Prior to the imaging studies, all animals had undergone a comprehensive veterinary evaluation and were in good clinical health. All subjects were neurologically intact and had no prior history of any intracranial, spine, or peripheral nerve trauma or surgeries. All subjects were sedated and immobilized by an intramuscular (IM) administration of ketamine (5–30 mg/kg before undergoing spine imaging studies). For radiographic studies, including CT imaging, additional ketamine (IM) was administered as needed for sedation. For MRI studies, the subjects were initially sedated and immobilized by an IM dose of ketamine followed by intubation and administration by isoflurane for the duration of the imaging procedure. Plain X-rays of the lower thoracic, lumbar and sacral spine were taken in both antero-posterior (AP) and lateral projections (InnoVet Select, Chicago, IL). All animals were preoperatively scanned with a 1.5T MRI (Signa HDxt, GE Healthcare, Little Chalfont, UK). Two or three millimeter sagittal and 3 or 5 mm axial T1 weighted- and T2 weighted-sequences were collected, using a matrix size of 256 3 256, a field of view of 240 mm, and a 1.5 mm interslice gap, depicting the spinal cord, cauda equina and spinal roots in relation to the vertebral spinal levels. In addition, 6 animals underwent spine CT imaging (LightSpeed16 or Discovery 610, GE Healthcare) using 140 kVp with default of 50 mA and the auto mA function active, 0.625 mm slice collimation, 500 mm field of view, matrix size of 512 3 512, and bone window technique (H40s) with the Bone Plus IQ Enhance kernel for image reconstruction including 3D-reformats of the thoracolumbar transition zone. These scans were analyzed with special attention to the craniocaudal position of the conus medullaris, possible rib variations, presence of possible pairs of supernumerary ribs, and general morphology of the transitional Th12, L1 and L2 vertebrae. The tip of the conus medullaris was defined as the most caudal axial MRI section with visible medulla. The level of this section was measured as depicted in Figure 3. Briefly, the axial section was traced on the sagittal section, and the level of the axial section was counted as x/y, where x is the distance in mm from the top of the vertebra to the level of the axial section, and y is the distance in mm of the combined height of the vertebra and associated caudal intervertebral disk. The level of an
302
OHLSSON ET AL.
Fig. 1. Spinal radiographs demonstrating dominant and variant spine anatomy in rhesus macaques. A subject with dominant spine anatomy shows 12 thoracic vertebrae with ribs and 7 lumbar vertebrae without ribs in anteroposterior (AP) and lateral views (A, C). Note that subject with a variant spine anatomy shows a pair of supernumerary ribs extending from the L1 vertebra in AP and lateral views (B,D).
individual axial section was expressed as the distance from the transition between the T12 and L1 vertebra. Radiological images were analyzed in RadiAnt DICOM viewer (64-bit) (Medixant, Poznan, Poland) and/or Efilm Lite (Essential Enterprise Solutions, Tarpon Springs, FL), depending on system compatibility and further processed for publishing in Photoshop CCV (Adobe Systems Inc., San Jose, CA). Graphs as well as numerical and statistical R Software analyses were created using Microsoft ExcelV (Microsoft, Redmond, WA). Quantitative data were expressed as the mean 6 standard deviation. R
RESULTS A total of 44 adult rhesus macaques were examined by spine radiography and magnetic resonance imaging (MRI) techniques. All subjects were neurologically intact and included 33 female and 11 male subjects.
Radiographic and CT Imaging of the Thoracolumbar Spine Radiographic studies of the thoracolumbar spine in the AP and lateral views allowed for the identification of individual vertebrae and ribs in all subject. For the majority of subjects, there were 12 rib-bearing vertebrae forming the thoracic portion of the spine and 7 lumbar vertebrae without ribs (n 5 36) (Fig. 1A,C). In the remaining subjects (n 5 8), a variant spine anatomy was present with a total of 13 rib-bearing vertebrae and 6
lumbar vertebrae without ribs (Fig. 1B,D). The variant spine anatomy was detected in 3 male and 5 female subjects. CT studies of the entire spine, including the thoracolumbar transition and upper sacrum, were performed in 6 select subjects. A total of 4 of the subjects showed 12 rib-bearing vertebrae and 2 subjects exhibited a variant spine anatomy with 13 rib-bearing vertebrae (Fig. 2). For all of the 4 subjects with 12 rib-bearing thoracic vertebrae, the CT imaging studies confirmed the absence of ribs associated with the L1 vertebra, which also did not show any facet joint, bony processes or specializations that may accommodate a ribs head (Fig. 2C). In contrast, the 2 subjects with an atypical thoracolumbar spine presentation showed bilateral bony projections with shallow fossae along the L1 posterolateral vertebral body margins to accommodate the supernumerary ribs (Fig. 2D).
Magnetic Resonance Imaging of the Thoracolumbar Spine and Lumbosacral Spinal Cord MRI studies of the thoracic and lumbar spine were performed to characterize size differences between individual vertebrae of the lower thoracic and upper lumbar spine as well as determine the position of the conus medullaris within the spinal canal and its anatomical relationship to the lumbar spine. MR imaging of the thoracolumbar spine allows for the identification of
RADIOGRAPHIC AND MAGNETIC RESONANCE IMAGING
Fig. 2.
303
304
OHLSSON ET AL.
individual vertebrae and inter-vertebral discs as well as the spinal cord and nerve roots in different anatomical planes (Fig. 3). First, vertebral height progression at the thoracolumbar spine transition was examined in all subjects. For this purpose, the height ratio was calculated for three successive vertebral pairs at the transition between the thoracic and lumbar spine (Fig. 4). The height ratios for the T11/T12, T12/L1, and L1/L2 vertebral pairs were 0.90 6 0.06, 0.85 6 0.07, and 0.92 6 0.04, respectively (n 5 44). Next, the vertebral height ratio was calculated for subjects with 12 and 13 rib-bearing vertebrae as separate subgroups. For this purpose and to facilitate comparisons between groups, rhesus macaques with a set of supernumerary ribs were conidered as having 12 thoracic vertebrae and an L1 vertebra with ribs. The height ratios for the T11/T12, T12/L1, and L1/L2 vertebral pairs were 0.89 6 0.05, 0.86 6 0.07 and 0.92 6 0.04, respectively, for the anatomically dominant group (n 5 36), and the same ratios were 0.90 6 0.07, 0.82 6 0.05 and 0.92 6 0.07, respectively, for the anatomical variant sub-group (n 5 8). No statistical difference in vertebral height progression was detected for any of the three vertebral pairs between the sub-groups. MRI studies were also performed to determine the caudal extent of the spinal cord and position of the conus medullaris in the spinal canal (Fig. 5). The rostral part of the conus medullaris was identified as the beginning of the caudal tapering of the lumbosacral spinal cord. The terminal end of the conus medullaris was determined as the caudal tapering apex of the spinal cord and the formation of the filum terminalis. The thoracolumbar junction between the T12 and L1 vertebral units was used as a reference point for describing the caudal projection of the conus medullaris in all subjects. The beginning or the top part of the conus medullaris was at the level of 2.0 6 0.3 vertebral units caudal to the thoracolumbar junction when all subjects were considered (n 5 44). Separate analyses were also performed for the subgroup of subjects demonstrating the typical anatomy of 12 rib-bearing vertebrae (n 5 36) and for the subgroup of subjects with a pair of supernumerary ribs (n 5 8). The top of the conus medullaris was at the level of 2.0 6 0.3 vertebral units caudal to the thoracolumbar junction for the dominant subgroup (n 5 36) and at the level of 2.3 6 0.3 vertebral units caudal to the thoracolumbar junction for the subgroup exhibiting a pair of supernumerary ribs (n 5 8). There was no statistical difference between the subgroups for the location of the top part of the conus medullaris. The tip of the conus medullaris was at the level of 3.6 6 0.4 vertebral units caudal to the thoracolumbar junction when all subjects were included (n 5 44). For the dominant subgroup and for the variant subgroup with a pair of supernumerary ribs, the conus medullaris showed a caudal termination at 3.5 6 0.4 (n 5 36) and 3.9 6 0.3 (n 5 8) vertebral units caudal to the
Fig. 2. Three-dimensional CT-images of dominant (A, C, and E) and variant (B, D, and F) spine anatomy in adult female rhesus macaques. Subjects with both dominant and variant spine anatomy show 7 cervical, 12 thoracic vertebrae, and 7 lumbar vertebrae. Note the presence of a pair of supernumerary ribs attached to the L1 vertebra in the
thoracolumbar junction, respectively. There was no significant difference for the position of the tip of the conus medullaris in the spinal canal between the subgroups. Based on the location of the top and inferior positions for the conus medullaris, the rostro-caudal length of the conus medullaris was calculated for each subject. The length of the conus medullaris was 1.5 6 0.3 vertebral units when all subjects were considered (n 5 44). For the dominant subgroup with 12 rib-bearing vertebrae and the variant subgroup with a pair of supernumerary ribs, the length of the conus medullaris was 1.5 6 0.3 vertebral units (n 5 36) and 1.5 6 0.2 vertebral units (n 5 8), respectively. There was no significant difference in conus medullaris length between the groups.
DISCUSSION This study evaluated the anatomy of the thoracolumbar spine and its relationship with the lumbosacral spinal cord in adult rhesus macaques. The vast majority of subjects demonstrated 12 thoracic vertebrae with ribs and 7 lumbar vertebrae without ribs. A subset of animals showed 13 rib-bearing vertebrae and 6 lumbar vertebrae without ribs. In the latter and smaller subset of subjects, the lowest rib-bearing vertebra was considered an L1 vertebra with ribs. Next, combined radiography and MRI demonstrated extensive individual variability between subjects with regards to the caudal extent of the spinal cord and the position of the conus medullaris. For all subjects combined and for each of the two subgroups analyzed separately, the average position for the upper part of the conus medullaris was at the level of the L2 vertebra, and the caudal end of the spinal cord with the location of the tip of the conus medullaris was at the level of the mid-portion of the L4 vertebral unit.
Radiographic Aspects of the Thoracolumbar Spine The mammalian spine may demonstrate extensive inter-species variability in the number and types of vertebrae with most of the variability in vertebral numbers affecting the most caudal portions of the spine. Almost all mammals have 7 cervical vertebras, with the exception of sloths, armadillos and anteaters that may show between 5-9 vertebrae in the cervical spine (Oostraet al., 2005). In humans and nonhuman primates, including both great apes and old world monkeys, it is very rare to encounter a subject that does not have 7 vertebrae in the cervical spine (Williams and Russo, 2015). In contrast, the number of vertebrae that contribute to the thoracolumbar spine and sacrum may show extensive variability between different species of nonhuman primates, and a shorter back with a reduced number of lumbar vertebrae is generally associated with an improved adaptation for upright posture and locomotion (Schultz, 1961; Whitcome et al., 2007; Williams et al., 2013). Humans also demonstrate variability in the thoracolumbar spine with atypical number of
subject with a variant spine anatomy (B,D,F). Detailed view of the thoracolumbar transition area in the anterior oblique right (AOR) view shows dominant spine anatomy with absence of lumbar ribs and variant spine anatomy with the presence of a pair of supernumerary ribs associated with the L1 vertebra.
RADIOGRAPHIC AND MAGNETIC RESONANCE IMAGING
Fig. 3.
305
306
OHLSSON ET AL.
rib-bearing vertebrae in about 11% of subjects and 2-8% of humans demonstrating a reduced number of 11 ribbearing vertebrae (Spencer et al., 2014). The present findings of a predominant phenotype of 12 thoracic and 7 lumbar vertebrae in normal rhesus macaques is in good agreement with prior studies on old world monkeys, including macaques, as is our demonstration of a subset of subjects with 13 rib-bearing vertebrae and 6 lumbar vertebrae without ribs (Aimi et al., 1994; Russo et al., 2010; Williams and Russo, 2015). We did not encounter any subjects with a reduced number of rib-bearing vertebrae in our sample of rhesus macaques.
Nomenclature for Thoracolumbar Spine Variants The classification of individual vertebrae for the thoracolumbar spine in primates has varied in prior studies. A traditional approach requires the presence of a pair of ribs for a vertebra to be classified as thoracic (Schultz and Straus, 1945; Williams, 2012). A variant to the original costal definition requires also the presence of a costal facet in order for a vertebra to be identified as thoracic, and it therefore classifies vertebrae without costal facets but with free and movable ribs as lumbar (Bornstein and Peterson, 1966; Haeusler et al., 2002). Another definition is based on a functional consideration and takes into account the detailed anatomy of the synovial joints present in the spine, the zygapophyses, which may allow or restrict intervertebral movements along the spine (Pal and Routal, 1999; Russo, 2010). According to the zygapophyseal definition, the diaphragmatic vertebra represents the ultimate thoracic vertebra (Williams, 2012).In the present study, we used an alternative approach, and all subjects were considered as possessing 12 thoracic and 7 lumbar vertebrae. For the subset of animals with 13 rib-bearing vertebrae, the most caudal rib-bearing vertebra was identified as an L1 vertebra with ribs. Subsequent MRI studies on the thoracolumbar vertebral height progression and the relationship between the lumbar spine and the lumbosacral spinal cord validated this approach as a useful and reasonable nomenclature to use during the planning of surgical procedures targeting the conus medullaris and cauda equina.
medullaris to the lumbar spine was developed using each single lumbar vertebra and its associated intervertebral disc, as a measurement unit. This approach provided a practical strategy and eliminated the influence of normal size differences between vertebrae and the progressively increasing height of the vertebral bodies at the thoracolumbar region, as was also demonstrated in both subgroups of subjects in the present study. The latter aspect was particularly useful, as both male and female subjects were included in the studies, and differences in overall vertebral column height and in areas of vertebral weight bearing surfaces have been reported in nonhuman primates of different genders (Galloway et al., 1996). In rhesus macaques, the top of the conus was located at the level of the L2 vertebral unit, and the tip of the conus was located at the level of the mid-portion of the L4 vertebral unit. This location appears to be markedly more caudal than the corresponding location for the caudal end of the spinal cord in humans. In comprehensive MRI studies of the lumbar spine in adult humans, the normal location of the conus medullaris may range from the T11-T12 to L2-L3 vertebrae with an average location of the conus medullaris at the L1 vertebrae (Wilson and Prince, 1989; Morimoto et al., 2013). In rhesus macaques, a similar extensive range for the normal rostrocaudal positioning of the conus medullaris in the spinal canal was detected.
Surgical Considerations
Our MRI studies of the lumbar spine in rhesus macaques identified the lumbosacral spinal cord, including the tip of the conus, and determined the relationship between the conus medullaris and the vertebral column. In order to perform quantitative comparisons between subjects, a system to relate the tip of the conus
Palpation of bony landmark is a common technique in clinical practice to identify a variety of anatomical structures, including specific vertebral levels. However, this approach has limitations, also for experienced clinicians. Relying on bony landmarks only to determine spinal levels in the lower back resulted in the identification of correct lumbar levels in 74% of attempts, and in the presence of an atypical spine anatomy, this success rate was reduced to 55% (Snider et al., 2011). In rhesus macaques, there is also a relatively common occurrence of normal anatomical variants of the spine, and this also limits the utility surface anatomy reliance for diagnostic or surgical procedures involving the thoracolumbar spine. In addition, there is extensive variability for the positioning of the caudal end of the spinal cord in the within the spinal canal. Based on the present findings, a two part approach is suggested to determine the anatomical features of the lumbar spine, lumbosacral spinal cord, and the cauda equina in macaques prior to performing surgical procedures targeting these structures. First, initial radiography studies in the anteroposterior and lateral views of the entire thoraco-lumbar spine and the upper portion of the sacrum will determine the number and location of thoracic and lumbar
Fig. 3. Magnetic resonance imaging (MRI) of the lower thoracic and lumbar spine in an adult female rhesus macaque. Sagittal T2 views of the thoracolumbar spine shows relationship between the vertebral column and the lumbosacral spinal cord. The conus medullaris is visible as the caudal tapering of the spinal cord. The rostral part of the conus medullaris is identified as the beginning of the caudal tapering of the lumbosacral spinal cord (A). The caudal end of the conus medullaris is identified as the end of the tapered portion of the spinal cord (B). The
filum terminalis is readily identified below the tip of the conus medullaris (C). To determine the position of the conus medullaris in relation to the spinal column, a vertebral unit is considered as a single vertebra and its corresponding caudal disk, indicated as y (D). The position of the axial section in A-C is determined by tracing the axial section in the sagittal plane, and the position is calculated as x/y, with x representing the distance from the cranial surface of the spinal vertebra to the level of the axial section of interest, and y being the full height of the vertebral unit.
Location of the Conus Medullaris in the Spinal Canal
RADIOGRAPHIC AND MAGNETIC RESONANCE IMAGING
307
will allow for pre-surgical determination of vertebral levels as well as planning of the location of the skin incision and laminectomy to visualize individual spinal cord segments and lumbosacral nerve root levels as they emerge from the surface of the spinal cord (Ohlsson et al., 2013).
LITERATURE CITED
Fig. 4. Ratios of height difference between adjacent thoracolumbar vertebral units based on MRI measurements. No significant difference was seen in height ratios for the T11/T12, T12/L1, or L1/L2 in between the dominant and variant groups of animals.
Fig. 5. Position of the conus medullaris within the spinal canal in relation to the vertebral column in rhesus macaques. The Y-axis indicates the level of the cranial vertebral end plate for the lumbar vertebras. The average cranial and caudal levels of the conus medullaris are indicated. There was no significant difference between the dominant and variant groups of subjects.
vertebrae, including any variants with regards to the presence of ribs. Second, a lumbar MRI study in longitudinal and transverse planes will determine the caudal extent of the conus medullaris. Combined, these findings
Aimi M. 1994. Numerical variation of vertebrae in Japanese macaques, Macaca fuscata. Anthropol Sci 102:1–10. Bornstein PE, Peterson RR. 1966. Numerical variation of the presacral vertebral column in three population groups in North America. Am J Phys Anthropol 25:139–146. Carlstedt T, Havton L. 2012. The longitudinal spinal cord injury: Lessons from intraspinal plexus, cauda equina and medullary conus lesions. Handbook Clin Neurol 109:337–354. Courtine G, Roy RR, Raven J, Hodgson J, McKay H, Yang H, Zhong H, Tuszynski MH, Edgerton VR. 2005. Performance of locomotion and foot grasping following a unilateral thoracic corticospinal tract lesion in monkeys (Macaca mulatta). Brain 128: 2338–2358. Dobkin BH, Havton LA.2012. Paraplegia and Spinal Cord Syndromes. In: Daroff RB, Fenichel GM, Jankovic J, Mazziotta JC, editors. Bradley’s Neurology in Clinical Practice. 6th ed. Elsevier: Elsevier. p 286–292. Galloway A, Morbeck M, Zihlman A. 1996. Sex differences in the vertebral column of gombe chimpanzees. Primates 37:443–455. Haeusler M, Martelli SA, Boeni T. 2002. Vertebrae numbers of the early hominid lumbar spine. J Human Evol 43:621–643. Morimoto T, Sonohata M, Kitajima M, Konishi H, Otani K, Kikuchi S, Mawatari M. 2013. The termination level of the conus medullaris and lumbosacral transitional vertebrae. J Orthop Sci 18: 878–884. Nielson JL, Haefeli J, Salegio EA, Liu AW, Guandique CF, St€ uck ED, Hawbecker S, Moseanko R, Strand SC, Zdunowski S, etal. 2015. Leveraging biomedical informatics for assessing plasticity and repair in primate spinal cord injury. Brain Res 1619:124–138. Nout YS, Ferguson AR, Strand SC, Moseanko R, Hawbecker S, Zdunowski S, Nielson JL, Roy RR, Zhong H, Rosenzweig ES, et al. 2012. Methods for functional assessment after C7 spinal cord hemisection in the rhesus monkey. Neurorehabil Neural Repair 26:556–569. Nout YS, Rosenzweig ES, Brock JH, Strand SC, Moseanko R, Hawbecker S, Zdunowski S, Nielson JL, Roy RR, Courtine G, et al. 2012. Animal models of neurologic disorders: A nonhuman primate model of spinal cord injury. Neurotherapeutics 9:380– 392. Ohlsson M, Nieto JH, Christe KL, Havton LA. 2013. Long-term effects of a lumbosacral ventral root avulsion injury on axotomized motor neurons and avulsed ventral roots in a non-human primate model of cauda equina injury. Neuroscience 250:129–139. Oostra RJ, Hennekam RC, de Rooij L, Moorman AFM. 2005. Malformations in the axial skeleton in Museum Vrolik I: homeotic transformations and numerical anomalies. Am J Med Genet 134A:268–281. Pal GP, Routal RV. 1999. Mechanism of change in the orientation of the articular process of the zygapophyseal joint at the thoracolumbar junction. J Anat 195:199–209. Rosenzweig ES, Courtine G, Jindrich DL, Brock JH, Ferguson AR, Strand SC, Nout YS, Roy RR, Miller DM, Beattie MS, et al. 2010. Extensive spontaneous plasticity of corticospinal projections after primate spinal cord injury. Nat Neurosci 13:1505–1510. Russell WMS, Burch, RL. 1959. The principle of humane experimental technique. London: Methuen and Co Ltd. Russo GA. 2010. Prezygoapophyseal articular facet shape in the catharrine thoracolumbar vertebral column. Am J Phys Anthropol 142:600–612. Schultz AH. 1961. Vertebral column and thorax. Primatologia 4: 1–66.
308
OHLSSON ET AL.
Schultz AH, Straus WL. 1945. The numbers of vertebrae in primates. Proc Am Philos Soc 89:601–626. Snider KT, Snider EJ, Degenhardt BF, Johnson JC, Kribs JW. 2011. Palpatory accuracy of lumbar spinous processes using multiple bony landmarks. J Manipulative Physiol Ther 34:306–313. Spencer HT, Gold ME, Hresko MT. 2014. Abnormal rib count in scoliosis surgery: Impact on the reporting of spinal fusion levels. J Child Orthop 8:497–503. Whitcome KK, Shapiro LJ, Lieberman DE. 2007. Fetal load and the evolution of lumbar lordosis in bipedal hominins. Nature 450: 1075–1078.
Williams SA. 2012. Variation in anthropoid vertebral formulae: Implications for homology and homoplasy in hominoid evolution. J Exp Zool B Mol Dev Evol 318:134–147. Williams SA, Ostrofsky KR, Frater N, Churchill SE, Schmid P, Berger LR. 2013. The vertebral column of Australopithecus sediba. Science 340:1232996. Williams SA, Russo GA. 2015. Evolution of the hominoid vertebral column: The long and the short of it. Evol Anthropol 24:15–32. Wilson DA, Prince JR. 1989. MR imaging determination of the location of the normal conus medullaris throughout childhood. 10: 259–262.
