- Email: [email protected]
Morphometric and functional study of the canine atlantoaxial joint
Journal Pre-proof Morphometric and functional study of the canine atlantoaxial joint
Bastien Planchamp, Jasmin Bluteau, Michael H. Stoffel, Christina Precht, Fenella Schmidli, Franck Forterre PII:
S0034-5288(19)30650-2
DOI:
https://doi.org/10.1016/j.rvsc.2019.11.005
Reference:
YRVSC 3918
To appear in:
Research in Veterinary Science
Received date:
9 July 2019
Revised date:
12 October 2019
Accepted date:
12 November 2019
Please cite this article as: B. Planchamp, J. Bluteau, M.H. Stoffel, et al., Morphometric and functional study of the canine atlantoaxial joint, Research in Veterinary Science (2019), https://doi.org/10.1016/j.rvsc.2019.11.005
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
© 2019 Published by Elsevier.
Journal Pre-proof
Morphometric and functional study of the canine atlantoaxial joint Med. vet. Bastien Planchamp 1 , Dr. Jas min Bluteau 2 , Prof. Dr. med. vet. Michael H. Stoffel2 , Dr. med. vet. Christina Precht 3 , Dr. med. vet. Fenella Schmidli4 , Prof. Dr. med. vet. Franck Forterre 1 1 Div ision of Small Animal Surgery, Depart ment of Clinical Veterinary Med icine, 2 Division of Veterinary Anatomy, 3 Division of Clinical Radiology, Depart ment of Clin ical Veterinary Medicine, 4 Division of Veterinary Neuro logy, Depart ment of Clinical Veterinary Medicine, Vetsuisse Faculty, Un iversity of Berne, Switzerland
Abstract
oo f
The atlantoaxial joint is sporadically affected by instability, in most cases a congenital pathology in young small breed dogs. Causes of atlantoaxial instability (AAI) are variable but
pr
are usually attributed to a lack of ligamentous support. The purpose of the present study was
e-
to specify the role of the ligamentous structures in the stabilisation of the atlantoaxial joint
Pr
and to find possible adaptations of the ligaments’ internal structure to their specific function. Five Beagle cadavers were included in this study. Each dog was subjected to a computed
al
tomography (CT) and a magnetic resonance imaging (MRI) examination of the upper cervical
rn
region. This region was then dissected and the ligamentous structures stabilising the atlantoaxial joint were measured and removed for histological analysis. A ligament to dens
Jo u
ratio (LDR) was established in order to provide a basis for comparison with the measurements taken in other dog breeds. MRI and gross anatomical measurements were very similar, confirming the validity of the results. MRI thus seems reliable for evaluating the ligamentous structures of the canine occipitoatlantoaxial region. The movement exerting the greatest stress on the atlantoaxial ligaments and inducing the greatest distension of the alar ligaments was a head flexion combined with a rotation. A clear adaptation of the ligamentous shape and internal structure to their specific function was observed. Histologically, alar ligaments consisted of wavy collagen fibres and a high proportion of elastic fibres, providing them with a remarkable elasticity compared to the transverse ligament structure which was much more rigid.
Journal Pre-proof Keywords: Dog, atlantoaxial joint, morphometry, function, ligament, histology
1. Introduction Ensuring the transition between the cranium and the vertebral column, the first two cervical vertebrae, the atlas and the axis, possess particular anatomical properties. The first cervical vertebra supports the occiput, together building one strong unit. Both occipital condyles and
oo f
the cranial articular surfaces of the atlas fit perfectly together. The only physiological movements of the atlanto-occipital joint described in human medicine are vertical movements
pr
such as flexion and extension of the head (Bogduk and Mercer, 2000). The atlas has a unique
e-
shape and presents two expanded transverse processes that are joined by a dorsal and a ventral
Pr
arch. The transition between the atlas and the axis is provided by two almost flat articular surfaces, which are in contact with the fovea dentis, a small cartilaginous area situated on the
rn
al
floor of the ventral arch.
The corresponding articular surfaces of the axis are situated on either side of its odontoid
Jo u
process, the dens axis, which is located at the cranial extremity of the vertebral body. Continually in touch with the fovea dentis of the atlas, the dens axis provides a pivot for rotational movements of the atlas and thus of the head (Barone, 1986). According to a publication in human medicine, the principal degree of freedom of the atlantoaxial joint provides for rotational movements but to a lesser extent also allows motion in a vertical plane and lateral bending due to the unstable conformation of the cranial articular surfaces of the axis (Steinmetz et al., 2010). Therefore, the atlantoaxial joint, due to its configuration and the absence of an intervertebral disc, requires additional stabilisation by the neck musculature, the cervical fasciae and the ligamentous structures.
Journal Pre-proof The extended transverse processes of the atlas, the wings, serve as insertions for part of the musculature responsible for vertical and rotational head movements. The rectus capitis dorsalis and the obliquus capitis cranialis muscles are part of the sub-occipital muscles, and their contraction results in an extension of the head. The obliquus capitis caudalis muscle, attached laterally to the atlas wings and extending cranially to the spinous process of the axis, is one of the main muscles allowing a rotatio n of the atlantoaxial joint (König and Liebig,
oo f
2020).
However, the major contribution to atlantoaxial joint stability comes from a series of
pr
ligaments, which can be divided into two groups: the peripheral ligaments, covering the
e-
intervertebral space and separating the atlas from the axis, and the deep ligaments, located on
Pr
the floor of the vertebral canal and connecting the dens axis to the atlas and occipital bone. The peripheral ligaments include the dorsal atlantoaxial membrane, the dorsal atlantoaxial
al
ligament and the ventral atlantoaxial ligament, which link the ventral tubercle of the atlas to
rn
the ventral crest of the axis. The tectorial membrane, the apical ligament of the dens axis, the
Jo u
alar ligaments and the transverse ligament constitute the deep ligaments (Barone, 1986).
The first intervertebral joint may incidentally be affected by instability, mostly related to a congenital pathology affecting young small breed dogs (Havig et al., 2005; Reber et al., 2013; Stalin et al., 2015). The dorsal subluxation of the axis goes along with spinal cord compression, which results in neurological dysfunction. Affected dogs present variable clinical signs from cervical hyperesthesia to tetraplegia and, in severe cases, even respiratory arrest and death (Havig et al., 2005; Stalin et al., 2015). Causes of the atlantoaxial instability (AAI) are variable but may usually be attributed to a lack of ligamentous support (Parry et al., 2010; Middleton et al., 2012; Reber et al., 2013; Stalin et al., 2015).
Journal Pre-proof To our knowledge, no previous study has investigated the morphometry of the atlantoaxial joint ligaments in dogs. Consequently, the goal of the present study was to determine the exact size of the ligamentous structures stabilising the atlantoaxial joint based on the anatomical dissection of the upper cervical region in Beagle cadavers to deduce their function and to examine the histological structure of the ligaments. Our working hypothesis was that ligaments would differ in size and structure and that the ligaments’ internal structure would relate to their specific function. We expected in dogs significant structural differences
oo f
especially between the alar and transverse ligaments due to the fact that, as described in human medicine, the alar ligaments play an important role during head rotation while the
pr
transverse ligament limits the dorsal displacement of the dens axis during head flexion
2. Materials and methods
al
2.1 Samples
Pr
e-
(Dvorak and Panjabi, 1987; Steinmetz et al., 2010; Lopez et al., 2015; Dickman et al., 1991).
rn
Five Beagles, 3 males and 2 females (Beagle 2 and 3), all between 4 and 5 years old, were
Jo u
euthanised for reasons unrelated to this research project. Cadavers were stored at –25°C and then thawed at room temperature for 36 hours prior to anatomical dissection. All the dogs involved in the present study had a similar body condition (average weight 11.1 kg, range 9.7 to 12.1 kg).
2.2 Diagnostic imaging Each cadaver was subjected to a computed tomography (CT) scan of the upper cervical region to exclude any bone malformation (Philips Brilliance CT 16-slice scanner®, Philips AG Healthcare, Zürich, Switzerland). Images were evaluated in a bone window and a soft tissue window (slice thickness 0.8 mm, increment 0.4 mm). The length of the dens axis was
Journal Pre-proof determined in a 3D multiplanar reconstruction (MPR) of the axis (Fig. 1, Takahashi et al., 2017) in order to establish a ligament to dens ratio (LDR) so as to provide a reasonable basis of comparison with the measurements taken in other dog breeds. In addition, a magnetic resonance imaging (MRI) scan of the same region was obtained to identify the ligamentous structures of the atlantoaxial joint and to complete their measurements (Panorama HFO 1.0T – Diamond Select MR System®, Philips AG Healthcare, Zürich, Switzerland). The four following sequences were performed: Turbo Spin Echo T2-weighted sequence in sagittal
oo f
plane (TSE T2w sag), 3D Gradient Echo T1-weighted sequence in dorsal plane (3D GE T1w dor), Turbo Spin Echo T1-weighted sequence in transverse plane (TSE T1w tra) and 3D
pr
Water Selective sequence in transverse plane (3D WATS tra). Additional details of the MRI
Pr
with the head in slight extension.
e-
sequences are listed in Table 1. Both examinations were performed in dorsal recumbency,
al
2.3 Anatomical dissection and measurements
rn
Cadavers were placed in sternal recumbency, with the head flexed, interpo sed between the
Jo u
two front limbs to guarantee a good access to the atlantoaxial region. The dorsal approach was performed through a large skin incision from the external occipital protuberance to the spinous process of the third cervical vertebra. The neck musculature was moved aside and the muscles which insert on the spinous process of the axis were carefully severed and retracted. The dorsal atlantoaxial membrane and the dorsal atlantoaxial ligament became evident extending from the dorsal arch of the atlas to the arch of the axis and to the spinous process of the axis, respectively.
Two landmarks, the dorsal tubercle of the atlas and the most cranial point of the spinous process of the axis, allowed measurement of the length of the dorsal atlantoaxial ligament.
Journal Pre-proof The measurements were performed first with the head in a neutral position (head straight in slight flexion, L1 ) and then after a maximal head rotation to each side (L2 to the left and L3 to the right side). The maximal angle of rotation (ARMax ), as based on the rotation of the head and the atlas, but without any additional rotation of the axis and following cervical vertebrae, was also determined (Fig. 2). These measurements were repeated after sectioning of the dorsal atlantoaxial ligament.
oo f
Subsequently, the dorsal atlantoaxial membrane and the joint capsule were incised. The dorsal arch of the atlas and the cranial part of the axis spinous process were removed using an
pr
oscillating saw. The spinal cord was sectioned at the borde r of the foramen magnum and at
e-
mid-axis level and was removed. After excision of the tectorial membrane, the deep ligaments
Pr
were dissected.
al
The apical ligament of the dens axis was particularly difficult to isolate, its length could only
rn
be determined on the MR images. The length of the alar ligaments was measured precisely
Jo u
from the tip of the dens axis to the inner aspects of the occipital condyles (L4 and L5 ). Four measurements describe the transverse ligament: its central width (L6 ), its insertional widths (L7 and L8 ) and its length (L9 ) as measured between the two medial walls of the dorsal atlas arch (Fig. 3). The measurements mentioned above were taken each with the head in a neutral position, in maximal extension, in maximal flexion and when the head was rotated fully and by 45 degrees to either side.
As additional parameters characterising the transverse ligament, the covering rate of the transverse ligament over the dens axis and the contact area between the transverse ligament and the dens axis were calculated. The covering rate denotes the central width of the transverse ligament (L6 ) divided by the length of the dens axis. The contact area corresponds
Journal Pre-proof to the product of the central width of the transverse ligament (L6 ) by the width of the dens axis (as measured in the middle of the dens in the dorsal plane of the MPR of CT images).
2.4 Histological analysis Once the measurements were completed, the alar and transverse ligaments were sectioned at their insertions and placed individually in test tubes filled with 4% buffered formaldehyde solution. Following the fixation, ligaments were dehydrated and embedded in paraffin using
oo f
an automated system (Shandon Histokinette Citadel 2000®, Histocom AG, Zug, Switzerland)
pr
according to standard protocols. Paraffin sections (3 µm thickness, one longitudinal and one transverse section) of the alar and transverse ligaments were cut using a Reichert-Jung 2030
e-
BioCut® manual rotary microtome (Leica, Biosystems, Muttenz, Switzerland) and processed
Pr
for Hematoxylin and Eosin (HE), Picrosirius and Verhoeff-Van Gieson (VVG) stainings.
al
The first series of sections was stained using a standard HE staining protocol (IHCWorld
rn
Protocol Database) and analysed under bright field illumination using a Zeiss AxioImager
Jo u
Z1® (Zeiss, Feldbach, Switzerland, 10x and 40x magnification) and a digital microscope VHX-5000® (Keyence, Urdorf, Switzerland).
The second series of sections was stained with Picrosirius red stain (IHCWorld Protocol Database; Junqueira et al., 1978; Dapson et al., 2011; Lattouf et al., 2014) and examined using polarised light microscopy (Linear Polarising Microscope BX61 ®, Olympus, Volketswil, Switzerland, 10x magnification) in order to reveal possible differences in fibre composition between the alar and transverse ligaments.
Journal Pre-proof Picrosirius red stain is highly specific for collagen and highlights its birefringence (refractive properties). Under polarised light, collagen bundles appear green, red or yellow allowing a quantitative analysis. The percentage of each birefringence colo ur obtained under polarised light microscopy was assessed semiquantitatively with the colour segmentation plug- in (Sage, D.) running in the Fiji ImageJ analysis programme (Rasband, W.S., 1997-2018). Five colours (black, green, yellow, orange and red) were chosen and the same RGB values per colo ur (combination of red, green and blue colours) were used to analyse the percentage of this
oo f
specific colour in each analysed image.
pr
To highlight the elastic components, the third series of sections was stained with VVG stain
Pr
AxioImager Z1®, 40x magnification).
e-
(IHCWorld Protocol Database) and analysed under a bright field microscope (Zeiss
al
Finally, two additional sections of the transverse ligament were stained with azan and alcian
rn
blue (IHCWorld Protocol Database) and analysed under bright field illumination using a Zeiss
Jo u
AxioImager Z1 ® (Zeiss, Feldbach, Switzerland, 10x and 40x magnification) in order to reveal the presence of fibrocartilage.
3. Results
3.1 Diagnostic imaging Diagnostic imaging (CT and MRI) did not reveal any abnormalities in the occipitoatlantoaxial region of the five Beagle cadavers. MRI could visualise all the ligamentous structures of the atlantoaxial joint in each dog examined. Except for the dorsal atlantoaxial ligament, the measurements of the ligamentous structures performed on MRI images were very congruent with the gross anatomical measurements. The TSE T2w sag sequence provided the best
Journal Pre-proof representation of the apical ligament of the dens axis (Fig. 4a). The 3D GE T1w dor sequence revealed the apical ligament of the dens axis, the alar ligaments and the transverse ligament (Fig. 4b). However, the transverse ligament was more distinct in 3D WATS tra (Fig. 4c) and TSE T1w tra sequences.
The average length of the dens axis was 10.6 mm. The ligament to dens ratio (LDR) of each
pr
3.2 Anatomical and histological analysis
oo f
ligament is presented in Table 2.
e-
3.2.1 Dorsal atlantoaxial ligament
Pr
The length L1 is the most representative relative measurement of its real size. In anatomical specimens, the length of L1 was 12.8 mm. The TSE T2w sag MRI sequence allowed to
al
estimate its average length at 4.0 mm (range 3.6 to 4.3 mm). Further anatomical
Jo u
rn
measurements are presented in Table 3.
The mean distance between the dorsal tubercle of the atlas and the spinous process of the axis (L2-3 ) as well as the maximal rotation angle (ARMax ) increase after transection of the dorsal atlantoaxial ligament.
3.2.2 Apical ligament of the dens axis The insertion of the apical ligament of the dens axis on the basilar part of the occipital bone could not be determined precisely, thus preventing its length to be measured during anatomical dissection. However, the apical ligament can easily be identified on the TSE T2w
Journal Pre-proof sag MRI sequence (Middleton et al., 2012). Its average size, established on this basis, is 10.4 mm (range 8.8 to 12.3 mm).
3.2.3 Alar ligaments
oo f
Measurements of the alar ligaments are presented in Table 4 and Graph 1.
Histologically, the structure of the alar ligaments is reminiscent of dense regular connective
pr
tissue. On the longitudinal sections stained with HE, the collagen fibres appear for the most
e-
part as wavy bundles. Their thickness is variable and their orientation is often parallel,
Pr
sometimes crisscrossed, but if so, usually at acute angles (Fig. 5a and b). The alar ligaments are particularly well vascularised. Many blood vessels run between the collagen fibres both
al
within the ligament itself as well as in the adjacent loose connective tissue. In longitudinal
rn
sections of the alar ligaments, a substantial amount of straight, evenly distributed, non-wavy
Jo u
elastic fibres was observed running alongside the collagen fibres (Fig. 5c).
Collagen composition of left and right alar ligaments was almost identical in terms of percentage of birefringence colours as revealed by linearised polarised light microscopy (Graph 2). The histological sections of the two alar ligaments contained on average 52.6% of collagen fibres and 47.4% of amorphous ground substance.
3.2.4 Transverse ligament The transverse ligament bridges the dens axis and secures it in contact with the fovea dentis of the atlas. Its length variations depend only on the movements of the dens axis ( Table 5, Graph 3).
Journal Pre-proof
As additional parameters characterising the transverse ligament, the covering rate and the contact area are presented in Table 6.
Histologically, the structure of the transverse ligaments is also evocative of regular dense connective tissue, the transverse ligament being particularly homogeneous. Ordered next to each other and pointing in the same direction, the fibres run perfectly in parallel as seen in the
oo f
longitudinal sections stained with HE. The thickness of the fibres is relatively constant, and their contours are straight (Fig. 6a and b). Sections of the transverse ligament rarely show the
pr
presence of blood vessels. The transverse ligaments appear to be poorly vascularised, and the
Pr
e-
few blood vessels present mainly run in the adjacent connective tissue.
The transverse ligaments contained many elastic fibres, though fewer than the alar ligaments.
al
Compared to the alar ligaments, the distribution of the elastic fibres was not as uniform, and
Jo u
part of the ligament.
rn
their density was considerably higher at the insertions and in the periphery than in the central
Collagen fibres do not necessarily show a uniform birefringence colo ur throughout their length. A combination of red and green is commonly present in any region. However, the fibres located at the ligament’s periphery in the transition zone to the adjacent connective tissue rather displayed an orange-red colouration (Fig. 6c). Distribution of the birefringence colours in the transverse ligament is different compared to the alar ligaments (Graph 2).
The azan and alcian blue staining methods did not reveal fibrocartilage in the central part of the transverse ligament.
Journal Pre-proof According to these results, the main functions of the ligamentous structures stabilising the atlantoaxial joint are illustrated in Fig. 7.
4. Discussion To our knowledge, no previous study has investigated the morphometry of the atlantoaxial joint ligaments in correlation with MRI findings in dogs. As the gross anatomical and MRI
oo f
measurements were alike, magnetic resonance imaging seems reliable for evaluating the ligamentous structures of the canine occipitoatlantoaxial region. Elongated and thicke ned
pr
apical, alar or transverse ligaments, atlantoaxial subluxation, hypo- or aplastic dens as well as focal spinal cord signal changes are pathological findings related to AAI in dogs (Middleton
e-
et al., 2012). However, many dogs suffer from AAI due to incomplete ossification of the atlas,
Pr
associated with a secondary deficiency of ligamentous structures. In those cases, the appropriate imaging technique seems to be CT. The transverse ligament is well visible on a
al
CT scan (soft tissue window), while the apical and alar ligaments are difficult to ide ntify.
rn
According to the histological analysis, the alar ligaments are particularly well vascularised.
Jo u
The administration of contrast medium could, therefore, possibly favour their visualisation. The CT facilitates the preoperative surgical planning (Vizcaíno Revés et al., 2013) but cannot illustrate spinal cord changes.
According to the literature, the main function of the atlantoaxial joint is to allow head rotation (Barone, 1986; Bogduk and Mercer, 2000; Lopez et al., 2015). The dorsal atlantoaxial ligament and the alar ligaments directly restrict this movement in order to avoid overrotation. The increase of L2-3 and of the maximal rotation angle (ARMax ) after the transection of the dorsal atlantoaxial ligament attest this statement.
Journal Pre-proof The dorsal atlantoaxial ligament is stretched during the flexion of the head and therefore limits its degree of maximal flexion. The discrepancy between the gross anatomical measurement and the MRI measurement of L1 can be explained by the alternative positioning required for these two approaches, the landmarks having remained identical for the gross anatomical and MRI measurements. As opposed to the slight extension during the MRI, the head was positioned in a neutral position during the clinical measurement of L1 . By analogy, the ventral atlantoaxial ligament probably limits the head’s rotation too, as well as its
oo f
extension.
pr
Extending from the occipital condyles to the dens axis, the alar ligaments are subject to
e-
considerable dynamic stress and their lengths vary substantially according to the position of
Pr
the head. A recent study has shown that alar ligaments play a major role in the stabilisation of the atlantoaxial joint under shear load (Reber et al., 2013). The significant stretching of the
al
alar ligaments during the flexion and rotation of the head indicates an obvious limitation of
rn
these movements. It is important to distinguish between two types of movements, i.e. purely
Jo u
vertical movements and rotational movements, because the lengths of the alar ligaments adjust differently. During vertical movements, the length of the alar ligaments changes in a bilaterally symmetrical way, increasing during the flexion of the head and decreasing during its extension. During rotational movements, however, adaptations in the length of the alar ligaments are inversely related. The length of the ligament located to the side of the rotation decreases while the length of the contralateral ligament increases. Briefly, both alar ligaments are stretched simultaneously during the flexion of the head, whereas the rotation is limited only by the contralateral alar ligament. Caution is clearly advisable when drawing interspecies conclusions because of the upright posture in humans and because of the snout’s lever action in dogs. Notwithstanding, publications from human medicine also report a limitation of the
Journal Pre-proof head’s rotation angle by the alar ligaments (Dvorak and Panjabi, 1987; Steinmetz et al., 2010; Lopez et al., 2015).
Histologically, the alar ligaments are mainly composed of wavy unorganised collagen fibres. This conformation and the high proportion of elastic fibres seem to provide them with appropriate elasticity, thus allowing them to adapt ideally to the numerous head movements generating forces in many different directions. The movement requiring the greatest
oo f
distension of the alar ligaments is a flexion combined with a rotation of the head. Serious trauma in this position may lead to ligament rupture and to AAI. In human medicine, this type
pr
of traumatic movement often occurs in road accidents, in which the patient suffers an uppe r
e-
cervical spine injury, more commonly known as whiplash (Dvorak and Panjabi, 1987;
Pr
Saldinger et al., 1990; Johansson, 2006). Coupled with high velocity, ball games may constitute an additional risk factor for young small breed dogs that are already predisposed to
rn
al
AAI.
Jo u
During the anatomical dissection, the insertion of the apical ligament of the dens axis on the occipital bone could not be determined precisely. Measured only on static MRI-sequences, its exact length could be established, but no information on the dynamic variations of its length could be collected. However, the evolution of its size could be observed via measurements of the other ligaments. The apical ligament has nearly the same length as the alar ligament (LDR 0.99 and 1.1, respectively). Its width, however, is clearly smaller, approximately half of the width of the alar ligament. During the extension of the head, the apical ligament seems to be the only deep ligamentous structure still taut. In human medicine, the apical ligament is considered as a vestigial structure that offers no significant added stability to the craniocervical junction (Tubbs et al., 2000). In the same study, about 20% of the examined human cadavers were devoid of an apical ligament. In dogs, the apical ligament seems to have
Journal Pre-proof kept a certain function in the stabilisation of the dens axis during the head extension, even if its constitution appears weaker than that of the alar ligaments.
In human medicine, the transverse ligament is considered the most important stabilising ligamentous structure of the entire atlantoaxial complex (Dickman et al., 1991). Similarly, even though the alar ligaments seem to be loaded with the greatest force by carrying the head (Reber et al., 2013), the contribution of the transverse ligament to the stability of the canine
oo f
atlantoaxial joint is essential. During head flexion, the dens axis tilts dorsally due to the slight ventral displacement of the atlas. Although attenuated by the alar ligaments’ function, the
pr
force applied by the dens axis on the transverse ligament intensifies, exerting an increased
e-
pressure on its central part and generating an enlargement of its central width as well as a
Pr
decrease in its insertional widths. The contact area and the covering rate increase, relieving the pressure on the transverse ligament and thus minimising the risk of rupture. By extension
al
of the head, the opposite happens. However, the length of the transverse ligament does not
rn
vary and only depends on the width of the atlas vertebral foramen. Neither the gender of the
Jo u
dog nor the weight of the specimen within the same breed seems to affect this parameter.
Histologically, the transverse ligament shows a much denser structure than the one of the alar ligaments. The highly organised collagen fibres run between the two bone attachment points on the atlas, suggesting stretch along this interconnecting line. The transverse ligament also contains many elastic fibres, their distribution, however, is not uniform and their density is considerably higher at the insertions than in the central part of the ligament. This statement implies that areas of increased elastic fibres near the attachments likely allow for some movement of the dens axis in relation to the atlas without compromising the strength of the portion directly covering the dens.
Journal Pre-proof Publications from human medicine (Milz et al., 2001) and veterinary medicine (Kupczynska et al., 2012) report the presence of fibrocartilage at attachment points of the occipitoatlantoaxial ligaments as well as in the central part of the transverse ligament. Fibrocartilage reinforce connective tissue where it is exposed to both stretch and compression, a twofold stress which also applies to the transverse ligament. The central part of the transverse ligament is exposed to intensive stretching and compression forces. The compression force, perpendicular to the ligament, results from the dorsal tilt of the dens axis,
oo f
which occurs during head flexion. However, the two additional staining methods used (azan and alcian blue) did not reveal any sizeable fibrocartilage in the central area of the transverse
e-
pr
ligament.
Pr
Although there is still some debate between authors whether colour coding reflects different collagen types (yellow-orange for type I, green for type III) or whether polarised colours only
al
reflect fibre thickness and packing, the literature supports the co ntention that Picrosirius
rn
specifically stains collagen and reveals its molecular order, organisation and the heterogeneity
Jo u
of collagen fibre orientation. Reports describe the fact that under circularised polarised light, thinner and more disorganised collagen fibres appear greenish while thicker and more organised fibres appear yellow to red (Junqueira et al., 1978; Lattouf et al., 2014). However, with Picrosirius red staining the types of collagen cannot be defined since the absorbed amount of polarised light depends on the orientation of the collagen bundles (Lattouf et al., 2014). The transverse ligament shows about 10% more yellow-orange collagen fibres, i.e. thicker and more regularly ordered fibres than the alar ligaments. With a small amount of amorphous ground substance in the range of 10% of the histological section, the transverse ligament is also more homogeneous than the alar ligaments.
Journal Pre-proof Compared to the alar ligaments, the transverse ligament is longer (LDR 1.1 and 1.6, respectively) and wider. However, this comparison is not pertinent, since the mechanism of dynamic compensation of the alar ligaments is very different from the one of the transverse ligament. The alar ligaments have a better elasticity and limit the flexion and the rotation of the head when their maximal stretch capacity is reached. By analogy, the dorsal atlantoaxial ligament and the apical ligament of the dens axis can be assigned to the same category. As a common denominator, these ligaments are thin, short (LDR smaller or close to 1), elastic and
oo f
designed for a restrictive function in order to limit a particular movement. In contrast, the transverse ligament is long (LDR 1.6), wide, and its insertions are fixed in space within a
pr
single bone. Thus, its denser structure is in accordance with its function, which consists in
e-
fastening the dens axis to the fovea dentis of the atlas in order to prevent a dorsal subluxation
Pr
of the second cervical vertebra.
al
Consequently, there seems to be a clear adaptation of the ligamentous shape and internal
rn
structure related to their specific function. Furthermore, several ligaments are responsible for
Jo u
restricting the same movement together. A major trauma might therefore not lead to injury or isolated rupture of one single ligament but rather to a lesion of the entire group and a possible destabilisation of the atlantoaxial joint.
Finally, the Beagle is not commonly affected by the atlantoaxial instability. This fact constitutes the main limitation of the present study. Because they are widely recognised as a reference breed in scientific studies, the Beagles were used as a homogenous standard not being affected by atlantoaxial instability..
The ligament to dens ratio (LDR) was established in order to provide a basis for comparison with the measurements taken in other dog breeds. However, the sizes of the ligaments and
Journal Pre-proof bones may certainly differ between individuals; therefore it remains to be established whether the LDR can be applied to all canine breeds. Future studies will be needed to corroborate the validity of this parameter. Nevertheless, the general considerations generated by the results, such as the ligamentous functions, may be expected to hold true in small breed dogs as well.
5. Conflict of interest
oo f
The authors declare no conflict of interest related to this study.
pr
6. References
Barone, R., 1986. Anatomie comparée des mammifères domestiques: Tome 1 : Ostéologie,
e-
Tome 2 : Arthrologie et myologie. Vigot Freres, Paris.
Pr
Bogduk, N., Mercer, S., 2000. Biomechanics of the cervical spine. I: Normal kinematics. Clinical biomechanics (Bristol, Avon) 15 (9), 633–648.
al
Dapson, R.W., Fagan, C., Kiernan, J.A., Wickersham, T.W., 2011. Certification procedures
rn
for sirius red F3B (CI 35780, Direct red 80). Biotechnic & histoche mistry : official
Jo u
publication of the Biological Stain Commission 86 (3), 133–139. Dickman, C.A., Mamourian, A., Sonntag, V.K., Drayer, B.P., 1991. Magnetic resonance imaging of the transverse atlantal ligament for the evaluation of atlantoaxial instabilit y. Journal of neurosurgery 75 (2), 221–227. Dvorak, J., Panjabi, M.M., 1987. Functional anatomy of the alar ligaments. Spine 12 (2), 183– 189. Havig, M.E., Cornell, K.K., Hawthorne, J.C., McDonnell, J.J., Selcer, B.A., 2005. Evaluation of nonsurgical treatment of atlantoaxial subluxation in dogs: 19 cases (1992-2001). Journal of the American Veterinary Medical Association 227 (2), 257–262.
Journal Pre-proof IHCWorld Protocol Database: Histology, Immunohistochemistry, Molecular and Cell Biology. General histology protocols and special stain protocols. https://www.ihcworld.com/protocol_database.htm. Accessed 10 October 2019. Johansson, B.H., 2006. Whiplash injuries can be visible by functional magnetic resonance imaging. Pain research & management 11 (3), 197–199. Junqueira, L.C., Cossermelli, W., Brentani, R., 1978. Differential staining of collagens type I, II and III by Sirius Red and polarization microscopy. Archivum histologicum Japonicum
oo f
= Nihon soshikigaku kiroku 41 (3), 267–274.
König, H.E., Liebich, H.-G. (Eds.), 2020. Veterinary Anatomy of Domestic Animals :
pr
Textbook and Color Atlas, 7th ed. Thieme, Stuttgart, 864 pp.
e-
Kupczynska, M., Wieladek, A., Janczyk, P., 2012. Craniocervical junction in dogs revisited--
Pr
new ligaments and confirmed presence of enthesis fibrocartilage. Research in veterinary science 92 (3), 356–361.
al
Lattouf, R., Younes, R., Lutomski, D., Naaman, N., Godeau, G., Senni, K., Changotade, S.,
rn
2014. Picrosirius red staining: a useful tool to appraise collagen networks in normal and
Jo u
pathological tissues. The journal of histochemistry and cytochemistry : official journal of the Histochemistry Society 62 (10), 751–758. Lopez, A.J., Scheer, J.K., Leibl, K.E., Smith, Z.A., Dlouhy, B.J., Dahdaleh, N.S., 2015. Anatomy and biomechanics of the craniovertebral junction. Neurosurgical focus 38 (4), E2. Middleton, G., Hillmann, D.J., Trichel, J., Bragulla, H.H., Gaschen, L., 2012. Magnetic resonance imaging of the ligamentous structures of the occipitoatlantoaxial region in the dog. Veterinary radiology & ultrasound : the official journal of the American College of Veterinary Radiology and the International Veterinary Radiology Association 53 (5), 545– 551.
Journal Pre-proof Milz, S., Schlüter, T., Putz, R., Moriggl, B., Ralphs, J.R., Benjamin, M., 2001. Fibrocartilage in the transverse ligament of the human atlas. Spine 26 (16), 1765–1771. Parry, A.T., Upjohn, M.M., Schlegl, K., Kneissl, S., Lamb, C.R., 2010. Computed tomography variations in morphology of the canine atlas in dogs with and without atlantoaxial subluxation. Veterinary radiology & ultrasound : the official journal of the American College of Veterinary Radiology and the International Veterinary Radiology Association 51 (6), 596–600.
oo f
Rasband, W.S., 1997-2018. ImageJ. U. S. National Institutes of Health. https://imagej.nih.gov/ij/.
pr
Reber, K., Bürki, A., Vizcaino Reves, N., Stoffel, M., Gendron, K., Ferguson, S.J., Forterre,
e-
F., 2013. Biomechanical evaluation of the stabilizing function of the atlantoaxial
Pr
ligaments under shear loading: a canine cadaveric study. Veterinary surgery : VS 42 (8), 918–923.
al
Sage, D. Color Segmentation : ImageJ plugin to cluster color pixel driven by the user input.
rn
bigwww.epfl.ch/sage/soft/colorsegmentation.
Jo u
Saldinger, P., Dvorak, J., Rahn, B.A., Perren, S.M., 1990. Histology of the alar and transverse ligaments. Spine 15 (4), 257–261. Stalin, C., Gutierrez-Quintana, R., Faller, K., Guevar, J., Yeamans, C., Penderis, J., 2015. A review of canine atlantoaxial joint subluxation. Veterinary and comparative orthopaedics and traumatology : V.C.O.T 28 (1), 1–8. Steinmetz, M.P., Mroz, T.E., Benzel, E.C., 2010. Craniovertebral junction: biomechanical considerations. Neurosurgery 66 (3 Suppl), 7–12. Takahashi, F., Hakozaki, T., Kanno, N., Harada, Y., Yamaguchi, S., Hara, Y., 2017. Evaluation of the dens-to-axis length ratio and dens angle in toy-breed dogs with and without atlantoaxial instability and in healthy Beagles. American journal of veterinary research 78 (12), 1400–1405.
Journal Pre-proof Tubbs, R.S., Grabb, P., Spooner, A., Wilson, W., Oakes, W.J., 2000. The apical ligament: anatomy and functional significance. Journal of neurosurgery 92 (2 Suppl), 197–200. Vizcaíno Revés, N., Stahl, C., Stoffel, M., Bali, M., Forterre, F., 2013. CT scan based determination of optimal bone corridor for atlantoaxial ventral screw fixation in miniature breed dogs. Veterinary surgery : VS 42 (7), 819–824.
Figure 1:
oo f
7. Figures Legends Median image of the axis reconstructed from CT images by 3D MPR. To
pr
determine the length of the dens axis (orange line), a line (continuous line) was first drawn passing through the tip of the dens and the dorsocaudal aspect of the body of the axis. Another
e-
line (dashed line) was then drawn perpendicular to the first, passing through the base of the
Pr
ventral aspect of the dens. The length of the dens was defined as the distance from the tip of
al
the dens to the ventral base of the dens (Takahashi et al., 2017).
rn
Figure 2: 3D rendering of CT images and outlining of the peripheral ligaments between the
Jo u
atlas (C1) and the axis (C2) with the corresponding anatomical preparations (lateral and dorsal view). Dorsal atlantoaxial ligament (1), dorsal atlantoaxial membrane (2), measurements (L1 to L3 ), maximal angle of rotation (ARMax ).
Figure 3: 3D rendering of CT images and outlining of the deep ligaments between the atlas (C1) and the axis (C2) with the corresponding anatomical preparation. Apical ligament of the dens axis (1), alar ligaments (2), transverse ligament (3), dens axis (Δ), measurements (L4 to L9 ). The black dashed arrows indicate the internal surface of the occipital condyles.
Journal Pre-proof Figure 4: MR images of TSE T2w sag sequence (a), 3D GE T1w dor sequence (b) and 3D WATS tra (c) of the upper cervical region, atlas (C1), axis (C2), dens axis (Δ), dorsal atlantoaxial ligament (1), apical ligament of the dens axis (2), alar ligaments (3), transverse ligament (4).
Figure 5: Longitudinal histological sections of the alar ligaments, HE stain (a. and b.), VVG stain (c.), longitudinal section with parallel fibre orientation, magnification 10x (b.) and
oo f
magnification 40x (c.); black dashed arrows indicate elastic fibres.
pr
Figure 6: Longitudinal histological sections of the transverse ligament, HE stain (a. and b.),
e-
Picrosirius red stain (c.), longitudinal section of one-half transverse ligament (a.) and
Pr
longitudinal sections, magnification 10x (b. and c.).
al
Figure 7: Representation of the dorsal atlantoaxial ligament (a) and the deep ligaments (b) on
rn
a 3D rendering of CT images of the upper cervical vertebrae showing their main functions in
Jo u
the stabilisation of the atlantoaxial joint. Blue: limitation of flexion
Green: limitation of extension
Red: limitation of rotation to the left Yellow: limitation of rotation to the right
Journal Pre-proof MRI Sequences Repetition time [ms] Echo time [ms] Slice thickness [mm] Gap [mm] Flip angle [°]
TSE T2w sag 1500 90 2.0 0.5 –
3D GE T1w dor 25 6.89 0.8 0.8 –
TSE T1w tra 533 7.51 3.0 0.5 –
3D WATS tra 22 11 25 – 0.75
Table 1: MRI parameters for Turbo spin echo T2-weighted (TSE T2w), 3D Gradient echo T1weighted (3D GE T1w), Turbo spin echo T1-weighted (TSE T1w) and 3D Water Selective (WATS) sequences. Repetition time and echo time are shown in ms, slice thickness and gap in mm and flip angle in degrees.
e-
Pr Beagle 2 15.2 15.6 – – 52 49 57 59
al
rn
L2 | L3 (ligament intact) L2 | L3 (ligament cut) Angle (ligament intact) Angle (ligament cut)
1 14 – 65 70
pr
a. 0.39 b. 0.29 c. 1.66
Table 2: Ligament to dens ratio (LDR)
Beagle 14.5 – 63 73
oo f
Ligament to dens ratio (LDR) 0.37 0.99 1.1
Dorsal atlantoaxial ligament (L1 ) Apical ligament of the dens axis Alar ligaments (L4-5 ) Transverse ligament : a. Central width (L6 ) b. Insertional widths (L7-8 ) c. Length (L9 )
Beagle 3 14.0 13.8 15.9 15.3 60 57 65 63
Beagle 4 14.3 13.8 17.2 16.1 71 67 76 74
Beagle 5 13.8 15.5 16.4 18.2 70 66 74 71
Jo u
Table 3: Measurements of the dorsal atlantoaxial ligament Lengths L2 (left column) and L3 (right column) in mm as well as maximal rotation angles in degrees (the left column refers to the rotation of the head to the left side and the right column to the right side) before and after transecting the dorsal atlantoaxial ligament. Delayed introduction of the second parameter explains the lack of data for Beagle 1 and 2.
Journal Pre-proof Beagle 1
Beagle 2
Beagle 3
Beagle 4
Beagle 5
Maximal rotation left
9.9
12.7
10.8
13.3
9.9
13.6
10.4
12.8
10.2
13.1
Rotation 45° left
10.2
12.2
11.1
12.8
10.4
12.2
11.0
11.9
11.1
12.6
Physiological position
11.6
11.4
12.6
12.3
11.5
11.6
11.4
11.5
11.5
11.7
Rotation 45° right
12.0
10.6
13.2
10.7
12.0
10.5
12.2
10.8
11.6
10.8
Maximal rotation right
13.1
8.9
13.8
10.0
13.5
10.1
13.1
10.1
12.7
10.4
Maximal extension
9.8
10.2
11.2
11.6
10.2
9.9
10.7
10.9
10.8
10.9
Maximal flexion
12.5
12.8
13.3
13.1
12.5
12.8
12.6
12.8
12.1
12.4
Beagle 2 4.4 3.9 3.0 3.2 3.1 3.1 16.6
4.6 3.3 3.2
3.9 3.4 3.3
Beagle 3 4.5 4.1 2.9 3.1 2.9 3.0 18.4
pr
Beagle 1 4.7 4.6 3.1 3.2 3.1 3.2 19.8
3.9 3.1 3.0
Beagle 4 4.7 4.3 2.7 2.9 2.5 2.8 16.2
4.2 3.1 3.0
Beagle 5 4.3 3.8 3.0 3.2 2.9 3.1 16.5
e-
Central width (L6 ) Insertional width left (L7 ) Insertional width right (L8 ) Length (L9 )
oo f
Table 4: Measurements of the alar ligaments The first column indicates the position of the head. For each dog, the left column refers to the length of the left alar ligament (L4 ) and the right column to the length of the right alar ligament (L5 )
Beagle 2 10.7 5.8 36.3 22.6
rn
Beagle 1 10.4 5.9 43.9 26.9
Jo u
Length [mm] Width [mm] Covering rate [%] Contact area [mm2 ]
al
Pr
Table 5: Measurements of the transverse ligament The three squares of a column show the size of the transverse ligament in mm in the following head positions: maximal flexion, physiological position and maximal extension – respectively from left to right.
Beagle 3 10.8 4.9 38.0 20.1
Beagle 4 10.5 5.5 41.1 23.7
Beagle 5 10.4 5.3 36.6 20.2
Average 10.6 5.5 39.2 22.7
Table 6: Measurements of the dens axis The length and the width of the dens axis are shown in mm, the covering rate in % and the contact area in mm 2 .
3.6 3.3 3.2
Journal Pre-proof
Graph 1: Mean lengths of the alar ligaments The horizontal axis legends indicates the head position during each measurement of the alar ligaments. Grey: Left alar ligament (L4 ) Blue: Right alar ligament (L5 )
pr
oo f
Graph 2: Graphical representation of the average distribution of the collagen fibres’ birefringence colours in histological sections of the alar and transverse ligaments taking into account (a.) and disregarding (b.) the amorphous ground substance. The columns’ colours correspond to the real birefringence colour observed on histological sections under polarized light. The black columns represent the amorphous ground substance. The values are shown in percent.
Jo u
rn
al
Pr
e-
Graph 3: Mean lengths of the transverse ligament The horizontal axis indicates the four measurements describing the transverse ligament. Representing the length (L9 ), the last columns only refer to the y-scale on the right hand side. The three columns in each group refer to the measurements as a function of the head’s position, whereby the colours represent the following positions: Grey: maximal flexion Light blue: neutral position Dark blue: maximal extension
Journal Pre-proof Highlights MRI reveals all the ligamentous structures of the atlantoaxial joint.
Head flexion with rotation results in maximum distension of the alar ligaments.
Histologically, the transverse ligament’s structure was particularly homogeneous.
Alar and transverse ligaments differ in amount and distribution of elastic fibres.
The ligaments’ internal structure is closely related to their specific function.
Jo u
rn
al
Pr
e-
pr
oo f
Figure 1
Figure 2
Figure 3
Figure 4
Figure 5
Figure 6
Figure 7
Bastien Planchamp, Jasmin Bluteau, Michael H. Stoffel, Christina Precht, Fenella Schmidli, Franck Forterre PII:
S0034-5288(19)30650-2
DOI:
https://doi.org/10.1016/j.rvsc.2019.11.005
Reference:
YRVSC 3918
To appear in:
Research in Veterinary Science
Received date:
9 July 2019
Revised date:
12 October 2019
Accepted date:
12 November 2019
Please cite this article as: B. Planchamp, J. Bluteau, M.H. Stoffel, et al., Morphometric and functional study of the canine atlantoaxial joint, Research in Veterinary Science (2019), https://doi.org/10.1016/j.rvsc.2019.11.005
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
© 2019 Published by Elsevier.
Journal Pre-proof
Morphometric and functional study of the canine atlantoaxial joint Med. vet. Bastien Planchamp 1 , Dr. Jas min Bluteau 2 , Prof. Dr. med. vet. Michael H. Stoffel2 , Dr. med. vet. Christina Precht 3 , Dr. med. vet. Fenella Schmidli4 , Prof. Dr. med. vet. Franck Forterre 1 1 Div ision of Small Animal Surgery, Depart ment of Clinical Veterinary Med icine, 2 Division of Veterinary Anatomy, 3 Division of Clinical Radiology, Depart ment of Clin ical Veterinary Medicine, 4 Division of Veterinary Neuro logy, Depart ment of Clinical Veterinary Medicine, Vetsuisse Faculty, Un iversity of Berne, Switzerland
Abstract
oo f
The atlantoaxial joint is sporadically affected by instability, in most cases a congenital pathology in young small breed dogs. Causes of atlantoaxial instability (AAI) are variable but
pr
are usually attributed to a lack of ligamentous support. The purpose of the present study was
e-
to specify the role of the ligamentous structures in the stabilisation of the atlantoaxial joint
Pr
and to find possible adaptations of the ligaments’ internal structure to their specific function. Five Beagle cadavers were included in this study. Each dog was subjected to a computed
al
tomography (CT) and a magnetic resonance imaging (MRI) examination of the upper cervical
rn
region. This region was then dissected and the ligamentous structures stabilising the atlantoaxial joint were measured and removed for histological analysis. A ligament to dens
Jo u
ratio (LDR) was established in order to provide a basis for comparison with the measurements taken in other dog breeds. MRI and gross anatomical measurements were very similar, confirming the validity of the results. MRI thus seems reliable for evaluating the ligamentous structures of the canine occipitoatlantoaxial region. The movement exerting the greatest stress on the atlantoaxial ligaments and inducing the greatest distension of the alar ligaments was a head flexion combined with a rotation. A clear adaptation of the ligamentous shape and internal structure to their specific function was observed. Histologically, alar ligaments consisted of wavy collagen fibres and a high proportion of elastic fibres, providing them with a remarkable elasticity compared to the transverse ligament structure which was much more rigid.
Journal Pre-proof Keywords: Dog, atlantoaxial joint, morphometry, function, ligament, histology
1. Introduction Ensuring the transition between the cranium and the vertebral column, the first two cervical vertebrae, the atlas and the axis, possess particular anatomical properties. The first cervical vertebra supports the occiput, together building one strong unit. Both occipital condyles and
oo f
the cranial articular surfaces of the atlas fit perfectly together. The only physiological movements of the atlanto-occipital joint described in human medicine are vertical movements
pr
such as flexion and extension of the head (Bogduk and Mercer, 2000). The atlas has a unique
e-
shape and presents two expanded transverse processes that are joined by a dorsal and a ventral
Pr
arch. The transition between the atlas and the axis is provided by two almost flat articular surfaces, which are in contact with the fovea dentis, a small cartilaginous area situated on the
rn
al
floor of the ventral arch.
The corresponding articular surfaces of the axis are situated on either side of its odontoid
Jo u
process, the dens axis, which is located at the cranial extremity of the vertebral body. Continually in touch with the fovea dentis of the atlas, the dens axis provides a pivot for rotational movements of the atlas and thus of the head (Barone, 1986). According to a publication in human medicine, the principal degree of freedom of the atlantoaxial joint provides for rotational movements but to a lesser extent also allows motion in a vertical plane and lateral bending due to the unstable conformation of the cranial articular surfaces of the axis (Steinmetz et al., 2010). Therefore, the atlantoaxial joint, due to its configuration and the absence of an intervertebral disc, requires additional stabilisation by the neck musculature, the cervical fasciae and the ligamentous structures.
Journal Pre-proof The extended transverse processes of the atlas, the wings, serve as insertions for part of the musculature responsible for vertical and rotational head movements. The rectus capitis dorsalis and the obliquus capitis cranialis muscles are part of the sub-occipital muscles, and their contraction results in an extension of the head. The obliquus capitis caudalis muscle, attached laterally to the atlas wings and extending cranially to the spinous process of the axis, is one of the main muscles allowing a rotatio n of the atlantoaxial joint (König and Liebig,
oo f
2020).
However, the major contribution to atlantoaxial joint stability comes from a series of
pr
ligaments, which can be divided into two groups: the peripheral ligaments, covering the
e-
intervertebral space and separating the atlas from the axis, and the deep ligaments, located on
Pr
the floor of the vertebral canal and connecting the dens axis to the atlas and occipital bone. The peripheral ligaments include the dorsal atlantoaxial membrane, the dorsal atlantoaxial
al
ligament and the ventral atlantoaxial ligament, which link the ventral tubercle of the atlas to
rn
the ventral crest of the axis. The tectorial membrane, the apical ligament of the dens axis, the
Jo u
alar ligaments and the transverse ligament constitute the deep ligaments (Barone, 1986).
The first intervertebral joint may incidentally be affected by instability, mostly related to a congenital pathology affecting young small breed dogs (Havig et al., 2005; Reber et al., 2013; Stalin et al., 2015). The dorsal subluxation of the axis goes along with spinal cord compression, which results in neurological dysfunction. Affected dogs present variable clinical signs from cervical hyperesthesia to tetraplegia and, in severe cases, even respiratory arrest and death (Havig et al., 2005; Stalin et al., 2015). Causes of the atlantoaxial instability (AAI) are variable but may usually be attributed to a lack of ligamentous support (Parry et al., 2010; Middleton et al., 2012; Reber et al., 2013; Stalin et al., 2015).
Journal Pre-proof To our knowledge, no previous study has investigated the morphometry of the atlantoaxial joint ligaments in dogs. Consequently, the goal of the present study was to determine the exact size of the ligamentous structures stabilising the atlantoaxial joint based on the anatomical dissection of the upper cervical region in Beagle cadavers to deduce their function and to examine the histological structure of the ligaments. Our working hypothesis was that ligaments would differ in size and structure and that the ligaments’ internal structure would relate to their specific function. We expected in dogs significant structural differences
oo f
especially between the alar and transverse ligaments due to the fact that, as described in human medicine, the alar ligaments play an important role during head rotation while the
pr
transverse ligament limits the dorsal displacement of the dens axis during head flexion
2. Materials and methods
al
2.1 Samples
Pr
e-
(Dvorak and Panjabi, 1987; Steinmetz et al., 2010; Lopez et al., 2015; Dickman et al., 1991).
rn
Five Beagles, 3 males and 2 females (Beagle 2 and 3), all between 4 and 5 years old, were
Jo u
euthanised for reasons unrelated to this research project. Cadavers were stored at –25°C and then thawed at room temperature for 36 hours prior to anatomical dissection. All the dogs involved in the present study had a similar body condition (average weight 11.1 kg, range 9.7 to 12.1 kg).
2.2 Diagnostic imaging Each cadaver was subjected to a computed tomography (CT) scan of the upper cervical region to exclude any bone malformation (Philips Brilliance CT 16-slice scanner®, Philips AG Healthcare, Zürich, Switzerland). Images were evaluated in a bone window and a soft tissue window (slice thickness 0.8 mm, increment 0.4 mm). The length of the dens axis was
Journal Pre-proof determined in a 3D multiplanar reconstruction (MPR) of the axis (Fig. 1, Takahashi et al., 2017) in order to establish a ligament to dens ratio (LDR) so as to provide a reasonable basis of comparison with the measurements taken in other dog breeds. In addition, a magnetic resonance imaging (MRI) scan of the same region was obtained to identify the ligamentous structures of the atlantoaxial joint and to complete their measurements (Panorama HFO 1.0T – Diamond Select MR System®, Philips AG Healthcare, Zürich, Switzerland). The four following sequences were performed: Turbo Spin Echo T2-weighted sequence in sagittal
oo f
plane (TSE T2w sag), 3D Gradient Echo T1-weighted sequence in dorsal plane (3D GE T1w dor), Turbo Spin Echo T1-weighted sequence in transverse plane (TSE T1w tra) and 3D
pr
Water Selective sequence in transverse plane (3D WATS tra). Additional details of the MRI
Pr
with the head in slight extension.
e-
sequences are listed in Table 1. Both examinations were performed in dorsal recumbency,
al
2.3 Anatomical dissection and measurements
rn
Cadavers were placed in sternal recumbency, with the head flexed, interpo sed between the
Jo u
two front limbs to guarantee a good access to the atlantoaxial region. The dorsal approach was performed through a large skin incision from the external occipital protuberance to the spinous process of the third cervical vertebra. The neck musculature was moved aside and the muscles which insert on the spinous process of the axis were carefully severed and retracted. The dorsal atlantoaxial membrane and the dorsal atlantoaxial ligament became evident extending from the dorsal arch of the atlas to the arch of the axis and to the spinous process of the axis, respectively.
Two landmarks, the dorsal tubercle of the atlas and the most cranial point of the spinous process of the axis, allowed measurement of the length of the dorsal atlantoaxial ligament.
Journal Pre-proof The measurements were performed first with the head in a neutral position (head straight in slight flexion, L1 ) and then after a maximal head rotation to each side (L2 to the left and L3 to the right side). The maximal angle of rotation (ARMax ), as based on the rotation of the head and the atlas, but without any additional rotation of the axis and following cervical vertebrae, was also determined (Fig. 2). These measurements were repeated after sectioning of the dorsal atlantoaxial ligament.
oo f
Subsequently, the dorsal atlantoaxial membrane and the joint capsule were incised. The dorsal arch of the atlas and the cranial part of the axis spinous process were removed using an
pr
oscillating saw. The spinal cord was sectioned at the borde r of the foramen magnum and at
e-
mid-axis level and was removed. After excision of the tectorial membrane, the deep ligaments
Pr
were dissected.
al
The apical ligament of the dens axis was particularly difficult to isolate, its length could only
rn
be determined on the MR images. The length of the alar ligaments was measured precisely
Jo u
from the tip of the dens axis to the inner aspects of the occipital condyles (L4 and L5 ). Four measurements describe the transverse ligament: its central width (L6 ), its insertional widths (L7 and L8 ) and its length (L9 ) as measured between the two medial walls of the dorsal atlas arch (Fig. 3). The measurements mentioned above were taken each with the head in a neutral position, in maximal extension, in maximal flexion and when the head was rotated fully and by 45 degrees to either side.
As additional parameters characterising the transverse ligament, the covering rate of the transverse ligament over the dens axis and the contact area between the transverse ligament and the dens axis were calculated. The covering rate denotes the central width of the transverse ligament (L6 ) divided by the length of the dens axis. The contact area corresponds
Journal Pre-proof to the product of the central width of the transverse ligament (L6 ) by the width of the dens axis (as measured in the middle of the dens in the dorsal plane of the MPR of CT images).
2.4 Histological analysis Once the measurements were completed, the alar and transverse ligaments were sectioned at their insertions and placed individually in test tubes filled with 4% buffered formaldehyde solution. Following the fixation, ligaments were dehydrated and embedded in paraffin using
oo f
an automated system (Shandon Histokinette Citadel 2000®, Histocom AG, Zug, Switzerland)
pr
according to standard protocols. Paraffin sections (3 µm thickness, one longitudinal and one transverse section) of the alar and transverse ligaments were cut using a Reichert-Jung 2030
e-
BioCut® manual rotary microtome (Leica, Biosystems, Muttenz, Switzerland) and processed
Pr
for Hematoxylin and Eosin (HE), Picrosirius and Verhoeff-Van Gieson (VVG) stainings.
al
The first series of sections was stained using a standard HE staining protocol (IHCWorld
rn
Protocol Database) and analysed under bright field illumination using a Zeiss AxioImager
Jo u
Z1® (Zeiss, Feldbach, Switzerland, 10x and 40x magnification) and a digital microscope VHX-5000® (Keyence, Urdorf, Switzerland).
The second series of sections was stained with Picrosirius red stain (IHCWorld Protocol Database; Junqueira et al., 1978; Dapson et al., 2011; Lattouf et al., 2014) and examined using polarised light microscopy (Linear Polarising Microscope BX61 ®, Olympus, Volketswil, Switzerland, 10x magnification) in order to reveal possible differences in fibre composition between the alar and transverse ligaments.
Journal Pre-proof Picrosirius red stain is highly specific for collagen and highlights its birefringence (refractive properties). Under polarised light, collagen bundles appear green, red or yellow allowing a quantitative analysis. The percentage of each birefringence colo ur obtained under polarised light microscopy was assessed semiquantitatively with the colour segmentation plug- in (Sage, D.) running in the Fiji ImageJ analysis programme (Rasband, W.S., 1997-2018). Five colours (black, green, yellow, orange and red) were chosen and the same RGB values per colo ur (combination of red, green and blue colours) were used to analyse the percentage of this
oo f
specific colour in each analysed image.
pr
To highlight the elastic components, the third series of sections was stained with VVG stain
Pr
AxioImager Z1®, 40x magnification).
e-
(IHCWorld Protocol Database) and analysed under a bright field microscope (Zeiss
al
Finally, two additional sections of the transverse ligament were stained with azan and alcian
rn
blue (IHCWorld Protocol Database) and analysed under bright field illumination using a Zeiss
Jo u
AxioImager Z1 ® (Zeiss, Feldbach, Switzerland, 10x and 40x magnification) in order to reveal the presence of fibrocartilage.
3. Results
3.1 Diagnostic imaging Diagnostic imaging (CT and MRI) did not reveal any abnormalities in the occipitoatlantoaxial region of the five Beagle cadavers. MRI could visualise all the ligamentous structures of the atlantoaxial joint in each dog examined. Except for the dorsal atlantoaxial ligament, the measurements of the ligamentous structures performed on MRI images were very congruent with the gross anatomical measurements. The TSE T2w sag sequence provided the best
Journal Pre-proof representation of the apical ligament of the dens axis (Fig. 4a). The 3D GE T1w dor sequence revealed the apical ligament of the dens axis, the alar ligaments and the transverse ligament (Fig. 4b). However, the transverse ligament was more distinct in 3D WATS tra (Fig. 4c) and TSE T1w tra sequences.
The average length of the dens axis was 10.6 mm. The ligament to dens ratio (LDR) of each
pr
3.2 Anatomical and histological analysis
oo f
ligament is presented in Table 2.
e-
3.2.1 Dorsal atlantoaxial ligament
Pr
The length L1 is the most representative relative measurement of its real size. In anatomical specimens, the length of L1 was 12.8 mm. The TSE T2w sag MRI sequence allowed to
al
estimate its average length at 4.0 mm (range 3.6 to 4.3 mm). Further anatomical
Jo u
rn
measurements are presented in Table 3.
The mean distance between the dorsal tubercle of the atlas and the spinous process of the axis (L2-3 ) as well as the maximal rotation angle (ARMax ) increase after transection of the dorsal atlantoaxial ligament.
3.2.2 Apical ligament of the dens axis The insertion of the apical ligament of the dens axis on the basilar part of the occipital bone could not be determined precisely, thus preventing its length to be measured during anatomical dissection. However, the apical ligament can easily be identified on the TSE T2w
Journal Pre-proof sag MRI sequence (Middleton et al., 2012). Its average size, established on this basis, is 10.4 mm (range 8.8 to 12.3 mm).
3.2.3 Alar ligaments
oo f
Measurements of the alar ligaments are presented in Table 4 and Graph 1.
Histologically, the structure of the alar ligaments is reminiscent of dense regular connective
pr
tissue. On the longitudinal sections stained with HE, the collagen fibres appear for the most
e-
part as wavy bundles. Their thickness is variable and their orientation is often parallel,
Pr
sometimes crisscrossed, but if so, usually at acute angles (Fig. 5a and b). The alar ligaments are particularly well vascularised. Many blood vessels run between the collagen fibres both
al
within the ligament itself as well as in the adjacent loose connective tissue. In longitudinal
rn
sections of the alar ligaments, a substantial amount of straight, evenly distributed, non-wavy
Jo u
elastic fibres was observed running alongside the collagen fibres (Fig. 5c).
Collagen composition of left and right alar ligaments was almost identical in terms of percentage of birefringence colours as revealed by linearised polarised light microscopy (Graph 2). The histological sections of the two alar ligaments contained on average 52.6% of collagen fibres and 47.4% of amorphous ground substance.
3.2.4 Transverse ligament The transverse ligament bridges the dens axis and secures it in contact with the fovea dentis of the atlas. Its length variations depend only on the movements of the dens axis ( Table 5, Graph 3).
Journal Pre-proof
As additional parameters characterising the transverse ligament, the covering rate and the contact area are presented in Table 6.
Histologically, the structure of the transverse ligaments is also evocative of regular dense connective tissue, the transverse ligament being particularly homogeneous. Ordered next to each other and pointing in the same direction, the fibres run perfectly in parallel as seen in the
oo f
longitudinal sections stained with HE. The thickness of the fibres is relatively constant, and their contours are straight (Fig. 6a and b). Sections of the transverse ligament rarely show the
pr
presence of blood vessels. The transverse ligaments appear to be poorly vascularised, and the
Pr
e-
few blood vessels present mainly run in the adjacent connective tissue.
The transverse ligaments contained many elastic fibres, though fewer than the alar ligaments.
al
Compared to the alar ligaments, the distribution of the elastic fibres was not as uniform, and
Jo u
part of the ligament.
rn
their density was considerably higher at the insertions and in the periphery than in the central
Collagen fibres do not necessarily show a uniform birefringence colo ur throughout their length. A combination of red and green is commonly present in any region. However, the fibres located at the ligament’s periphery in the transition zone to the adjacent connective tissue rather displayed an orange-red colouration (Fig. 6c). Distribution of the birefringence colours in the transverse ligament is different compared to the alar ligaments (Graph 2).
The azan and alcian blue staining methods did not reveal fibrocartilage in the central part of the transverse ligament.
Journal Pre-proof According to these results, the main functions of the ligamentous structures stabilising the atlantoaxial joint are illustrated in Fig. 7.
4. Discussion To our knowledge, no previous study has investigated the morphometry of the atlantoaxial joint ligaments in correlation with MRI findings in dogs. As the gross anatomical and MRI
oo f
measurements were alike, magnetic resonance imaging seems reliable for evaluating the ligamentous structures of the canine occipitoatlantoaxial region. Elongated and thicke ned
pr
apical, alar or transverse ligaments, atlantoaxial subluxation, hypo- or aplastic dens as well as focal spinal cord signal changes are pathological findings related to AAI in dogs (Middleton
e-
et al., 2012). However, many dogs suffer from AAI due to incomplete ossification of the atlas,
Pr
associated with a secondary deficiency of ligamentous structures. In those cases, the appropriate imaging technique seems to be CT. The transverse ligament is well visible on a
al
CT scan (soft tissue window), while the apical and alar ligaments are difficult to ide ntify.
rn
According to the histological analysis, the alar ligaments are particularly well vascularised.
Jo u
The administration of contrast medium could, therefore, possibly favour their visualisation. The CT facilitates the preoperative surgical planning (Vizcaíno Revés et al., 2013) but cannot illustrate spinal cord changes.
According to the literature, the main function of the atlantoaxial joint is to allow head rotation (Barone, 1986; Bogduk and Mercer, 2000; Lopez et al., 2015). The dorsal atlantoaxial ligament and the alar ligaments directly restrict this movement in order to avoid overrotation. The increase of L2-3 and of the maximal rotation angle (ARMax ) after the transection of the dorsal atlantoaxial ligament attest this statement.
Journal Pre-proof The dorsal atlantoaxial ligament is stretched during the flexion of the head and therefore limits its degree of maximal flexion. The discrepancy between the gross anatomical measurement and the MRI measurement of L1 can be explained by the alternative positioning required for these two approaches, the landmarks having remained identical for the gross anatomical and MRI measurements. As opposed to the slight extension during the MRI, the head was positioned in a neutral position during the clinical measurement of L1 . By analogy, the ventral atlantoaxial ligament probably limits the head’s rotation too, as well as its
oo f
extension.
pr
Extending from the occipital condyles to the dens axis, the alar ligaments are subject to
e-
considerable dynamic stress and their lengths vary substantially according to the position of
Pr
the head. A recent study has shown that alar ligaments play a major role in the stabilisation of the atlantoaxial joint under shear load (Reber et al., 2013). The significant stretching of the
al
alar ligaments during the flexion and rotation of the head indicates an obvious limitation of
rn
these movements. It is important to distinguish between two types of movements, i.e. purely
Jo u
vertical movements and rotational movements, because the lengths of the alar ligaments adjust differently. During vertical movements, the length of the alar ligaments changes in a bilaterally symmetrical way, increasing during the flexion of the head and decreasing during its extension. During rotational movements, however, adaptations in the length of the alar ligaments are inversely related. The length of the ligament located to the side of the rotation decreases while the length of the contralateral ligament increases. Briefly, both alar ligaments are stretched simultaneously during the flexion of the head, whereas the rotation is limited only by the contralateral alar ligament. Caution is clearly advisable when drawing interspecies conclusions because of the upright posture in humans and because of the snout’s lever action in dogs. Notwithstanding, publications from human medicine also report a limitation of the
Journal Pre-proof head’s rotation angle by the alar ligaments (Dvorak and Panjabi, 1987; Steinmetz et al., 2010; Lopez et al., 2015).
Histologically, the alar ligaments are mainly composed of wavy unorganised collagen fibres. This conformation and the high proportion of elastic fibres seem to provide them with appropriate elasticity, thus allowing them to adapt ideally to the numerous head movements generating forces in many different directions. The movement requiring the greatest
oo f
distension of the alar ligaments is a flexion combined with a rotation of the head. Serious trauma in this position may lead to ligament rupture and to AAI. In human medicine, this type
pr
of traumatic movement often occurs in road accidents, in which the patient suffers an uppe r
e-
cervical spine injury, more commonly known as whiplash (Dvorak and Panjabi, 1987;
Pr
Saldinger et al., 1990; Johansson, 2006). Coupled with high velocity, ball games may constitute an additional risk factor for young small breed dogs that are already predisposed to
rn
al
AAI.
Jo u
During the anatomical dissection, the insertion of the apical ligament of the dens axis on the occipital bone could not be determined precisely. Measured only on static MRI-sequences, its exact length could be established, but no information on the dynamic variations of its length could be collected. However, the evolution of its size could be observed via measurements of the other ligaments. The apical ligament has nearly the same length as the alar ligament (LDR 0.99 and 1.1, respectively). Its width, however, is clearly smaller, approximately half of the width of the alar ligament. During the extension of the head, the apical ligament seems to be the only deep ligamentous structure still taut. In human medicine, the apical ligament is considered as a vestigial structure that offers no significant added stability to the craniocervical junction (Tubbs et al., 2000). In the same study, about 20% of the examined human cadavers were devoid of an apical ligament. In dogs, the apical ligament seems to have
Journal Pre-proof kept a certain function in the stabilisation of the dens axis during the head extension, even if its constitution appears weaker than that of the alar ligaments.
In human medicine, the transverse ligament is considered the most important stabilising ligamentous structure of the entire atlantoaxial complex (Dickman et al., 1991). Similarly, even though the alar ligaments seem to be loaded with the greatest force by carrying the head (Reber et al., 2013), the contribution of the transverse ligament to the stability of the canine
oo f
atlantoaxial joint is essential. During head flexion, the dens axis tilts dorsally due to the slight ventral displacement of the atlas. Although attenuated by the alar ligaments’ function, the
pr
force applied by the dens axis on the transverse ligament intensifies, exerting an increased
e-
pressure on its central part and generating an enlargement of its central width as well as a
Pr
decrease in its insertional widths. The contact area and the covering rate increase, relieving the pressure on the transverse ligament and thus minimising the risk of rupture. By extension
al
of the head, the opposite happens. However, the length of the transverse ligament does not
rn
vary and only depends on the width of the atlas vertebral foramen. Neither the gender of the
Jo u
dog nor the weight of the specimen within the same breed seems to affect this parameter.
Histologically, the transverse ligament shows a much denser structure than the one of the alar ligaments. The highly organised collagen fibres run between the two bone attachment points on the atlas, suggesting stretch along this interconnecting line. The transverse ligament also contains many elastic fibres, their distribution, however, is not uniform and their density is considerably higher at the insertions than in the central part of the ligament. This statement implies that areas of increased elastic fibres near the attachments likely allow for some movement of the dens axis in relation to the atlas without compromising the strength of the portion directly covering the dens.
Journal Pre-proof Publications from human medicine (Milz et al., 2001) and veterinary medicine (Kupczynska et al., 2012) report the presence of fibrocartilage at attachment points of the occipitoatlantoaxial ligaments as well as in the central part of the transverse ligament. Fibrocartilage reinforce connective tissue where it is exposed to both stretch and compression, a twofold stress which also applies to the transverse ligament. The central part of the transverse ligament is exposed to intensive stretching and compression forces. The compression force, perpendicular to the ligament, results from the dorsal tilt of the dens axis,
oo f
which occurs during head flexion. However, the two additional staining methods used (azan and alcian blue) did not reveal any sizeable fibrocartilage in the central area of the transverse
e-
pr
ligament.
Pr
Although there is still some debate between authors whether colour coding reflects different collagen types (yellow-orange for type I, green for type III) or whether polarised colours only
al
reflect fibre thickness and packing, the literature supports the co ntention that Picrosirius
rn
specifically stains collagen and reveals its molecular order, organisation and the heterogeneity
Jo u
of collagen fibre orientation. Reports describe the fact that under circularised polarised light, thinner and more disorganised collagen fibres appear greenish while thicker and more organised fibres appear yellow to red (Junqueira et al., 1978; Lattouf et al., 2014). However, with Picrosirius red staining the types of collagen cannot be defined since the absorbed amount of polarised light depends on the orientation of the collagen bundles (Lattouf et al., 2014). The transverse ligament shows about 10% more yellow-orange collagen fibres, i.e. thicker and more regularly ordered fibres than the alar ligaments. With a small amount of amorphous ground substance in the range of 10% of the histological section, the transverse ligament is also more homogeneous than the alar ligaments.
Journal Pre-proof Compared to the alar ligaments, the transverse ligament is longer (LDR 1.1 and 1.6, respectively) and wider. However, this comparison is not pertinent, since the mechanism of dynamic compensation of the alar ligaments is very different from the one of the transverse ligament. The alar ligaments have a better elasticity and limit the flexion and the rotation of the head when their maximal stretch capacity is reached. By analogy, the dorsal atlantoaxial ligament and the apical ligament of the dens axis can be assigned to the same category. As a common denominator, these ligaments are thin, short (LDR smaller or close to 1), elastic and
oo f
designed for a restrictive function in order to limit a particular movement. In contrast, the transverse ligament is long (LDR 1.6), wide, and its insertions are fixed in space within a
pr
single bone. Thus, its denser structure is in accordance with its function, which consists in
e-
fastening the dens axis to the fovea dentis of the atlas in order to prevent a dorsal subluxation
Pr
of the second cervical vertebra.
al
Consequently, there seems to be a clear adaptation of the ligamentous shape and internal
rn
structure related to their specific function. Furthermore, several ligaments are responsible for
Jo u
restricting the same movement together. A major trauma might therefore not lead to injury or isolated rupture of one single ligament but rather to a lesion of the entire group and a possible destabilisation of the atlantoaxial joint.
Finally, the Beagle is not commonly affected by the atlantoaxial instability. This fact constitutes the main limitation of the present study. Because they are widely recognised as a reference breed in scientific studies, the Beagles were used as a homogenous standard not being affected by atlantoaxial instability..
The ligament to dens ratio (LDR) was established in order to provide a basis for comparison with the measurements taken in other dog breeds. However, the sizes of the ligaments and
Journal Pre-proof bones may certainly differ between individuals; therefore it remains to be established whether the LDR can be applied to all canine breeds. Future studies will be needed to corroborate the validity of this parameter. Nevertheless, the general considerations generated by the results, such as the ligamentous functions, may be expected to hold true in small breed dogs as well.
5. Conflict of interest
oo f
The authors declare no conflict of interest related to this study.
pr
6. References
Barone, R., 1986. Anatomie comparée des mammifères domestiques: Tome 1 : Ostéologie,
e-
Tome 2 : Arthrologie et myologie. Vigot Freres, Paris.
Pr
Bogduk, N., Mercer, S., 2000. Biomechanics of the cervical spine. I: Normal kinematics. Clinical biomechanics (Bristol, Avon) 15 (9), 633–648.
al
Dapson, R.W., Fagan, C., Kiernan, J.A., Wickersham, T.W., 2011. Certification procedures
rn
for sirius red F3B (CI 35780, Direct red 80). Biotechnic & histoche mistry : official
Jo u
publication of the Biological Stain Commission 86 (3), 133–139. Dickman, C.A., Mamourian, A., Sonntag, V.K., Drayer, B.P., 1991. Magnetic resonance imaging of the transverse atlantal ligament for the evaluation of atlantoaxial instabilit y. Journal of neurosurgery 75 (2), 221–227. Dvorak, J., Panjabi, M.M., 1987. Functional anatomy of the alar ligaments. Spine 12 (2), 183– 189. Havig, M.E., Cornell, K.K., Hawthorne, J.C., McDonnell, J.J., Selcer, B.A., 2005. Evaluation of nonsurgical treatment of atlantoaxial subluxation in dogs: 19 cases (1992-2001). Journal of the American Veterinary Medical Association 227 (2), 257–262.
Journal Pre-proof IHCWorld Protocol Database: Histology, Immunohistochemistry, Molecular and Cell Biology. General histology protocols and special stain protocols. https://www.ihcworld.com/protocol_database.htm. Accessed 10 October 2019. Johansson, B.H., 2006. Whiplash injuries can be visible by functional magnetic resonance imaging. Pain research & management 11 (3), 197–199. Junqueira, L.C., Cossermelli, W., Brentani, R., 1978. Differential staining of collagens type I, II and III by Sirius Red and polarization microscopy. Archivum histologicum Japonicum
oo f
= Nihon soshikigaku kiroku 41 (3), 267–274.
König, H.E., Liebich, H.-G. (Eds.), 2020. Veterinary Anatomy of Domestic Animals :
pr
Textbook and Color Atlas, 7th ed. Thieme, Stuttgart, 864 pp.
e-
Kupczynska, M., Wieladek, A., Janczyk, P., 2012. Craniocervical junction in dogs revisited--
Pr
new ligaments and confirmed presence of enthesis fibrocartilage. Research in veterinary science 92 (3), 356–361.
al
Lattouf, R., Younes, R., Lutomski, D., Naaman, N., Godeau, G., Senni, K., Changotade, S.,
rn
2014. Picrosirius red staining: a useful tool to appraise collagen networks in normal and
Jo u
pathological tissues. The journal of histochemistry and cytochemistry : official journal of the Histochemistry Society 62 (10), 751–758. Lopez, A.J., Scheer, J.K., Leibl, K.E., Smith, Z.A., Dlouhy, B.J., Dahdaleh, N.S., 2015. Anatomy and biomechanics of the craniovertebral junction. Neurosurgical focus 38 (4), E2. Middleton, G., Hillmann, D.J., Trichel, J., Bragulla, H.H., Gaschen, L., 2012. Magnetic resonance imaging of the ligamentous structures of the occipitoatlantoaxial region in the dog. Veterinary radiology & ultrasound : the official journal of the American College of Veterinary Radiology and the International Veterinary Radiology Association 53 (5), 545– 551.
Journal Pre-proof Milz, S., Schlüter, T., Putz, R., Moriggl, B., Ralphs, J.R., Benjamin, M., 2001. Fibrocartilage in the transverse ligament of the human atlas. Spine 26 (16), 1765–1771. Parry, A.T., Upjohn, M.M., Schlegl, K., Kneissl, S., Lamb, C.R., 2010. Computed tomography variations in morphology of the canine atlas in dogs with and without atlantoaxial subluxation. Veterinary radiology & ultrasound : the official journal of the American College of Veterinary Radiology and the International Veterinary Radiology Association 51 (6), 596–600.
oo f
Rasband, W.S., 1997-2018. ImageJ. U. S. National Institutes of Health. https://imagej.nih.gov/ij/.
pr
Reber, K., Bürki, A., Vizcaino Reves, N., Stoffel, M., Gendron, K., Ferguson, S.J., Forterre,
e-
F., 2013. Biomechanical evaluation of the stabilizing function of the atlantoaxial
Pr
ligaments under shear loading: a canine cadaveric study. Veterinary surgery : VS 42 (8), 918–923.
al
Sage, D. Color Segmentation : ImageJ plugin to cluster color pixel driven by the user input.
rn
bigwww.epfl.ch/sage/soft/colorsegmentation.
Jo u
Saldinger, P., Dvorak, J., Rahn, B.A., Perren, S.M., 1990. Histology of the alar and transverse ligaments. Spine 15 (4), 257–261. Stalin, C., Gutierrez-Quintana, R., Faller, K., Guevar, J., Yeamans, C., Penderis, J., 2015. A review of canine atlantoaxial joint subluxation. Veterinary and comparative orthopaedics and traumatology : V.C.O.T 28 (1), 1–8. Steinmetz, M.P., Mroz, T.E., Benzel, E.C., 2010. Craniovertebral junction: biomechanical considerations. Neurosurgery 66 (3 Suppl), 7–12. Takahashi, F., Hakozaki, T., Kanno, N., Harada, Y., Yamaguchi, S., Hara, Y., 2017. Evaluation of the dens-to-axis length ratio and dens angle in toy-breed dogs with and without atlantoaxial instability and in healthy Beagles. American journal of veterinary research 78 (12), 1400–1405.
Journal Pre-proof Tubbs, R.S., Grabb, P., Spooner, A., Wilson, W., Oakes, W.J., 2000. The apical ligament: anatomy and functional significance. Journal of neurosurgery 92 (2 Suppl), 197–200. Vizcaíno Revés, N., Stahl, C., Stoffel, M., Bali, M., Forterre, F., 2013. CT scan based determination of optimal bone corridor for atlantoaxial ventral screw fixation in miniature breed dogs. Veterinary surgery : VS 42 (7), 819–824.
Figure 1:
oo f
7. Figures Legends Median image of the axis reconstructed from CT images by 3D MPR. To
pr
determine the length of the dens axis (orange line), a line (continuous line) was first drawn passing through the tip of the dens and the dorsocaudal aspect of the body of the axis. Another
e-
line (dashed line) was then drawn perpendicular to the first, passing through the base of the
Pr
ventral aspect of the dens. The length of the dens was defined as the distance from the tip of
al
the dens to the ventral base of the dens (Takahashi et al., 2017).
rn
Figure 2: 3D rendering of CT images and outlining of the peripheral ligaments between the
Jo u
atlas (C1) and the axis (C2) with the corresponding anatomical preparations (lateral and dorsal view). Dorsal atlantoaxial ligament (1), dorsal atlantoaxial membrane (2), measurements (L1 to L3 ), maximal angle of rotation (ARMax ).
Figure 3: 3D rendering of CT images and outlining of the deep ligaments between the atlas (C1) and the axis (C2) with the corresponding anatomical preparation. Apical ligament of the dens axis (1), alar ligaments (2), transverse ligament (3), dens axis (Δ), measurements (L4 to L9 ). The black dashed arrows indicate the internal surface of the occipital condyles.
Journal Pre-proof Figure 4: MR images of TSE T2w sag sequence (a), 3D GE T1w dor sequence (b) and 3D WATS tra (c) of the upper cervical region, atlas (C1), axis (C2), dens axis (Δ), dorsal atlantoaxial ligament (1), apical ligament of the dens axis (2), alar ligaments (3), transverse ligament (4).
Figure 5: Longitudinal histological sections of the alar ligaments, HE stain (a. and b.), VVG stain (c.), longitudinal section with parallel fibre orientation, magnification 10x (b.) and
oo f
magnification 40x (c.); black dashed arrows indicate elastic fibres.
pr
Figure 6: Longitudinal histological sections of the transverse ligament, HE stain (a. and b.),
e-
Picrosirius red stain (c.), longitudinal section of one-half transverse ligament (a.) and
Pr
longitudinal sections, magnification 10x (b. and c.).
al
Figure 7: Representation of the dorsal atlantoaxial ligament (a) and the deep ligaments (b) on
rn
a 3D rendering of CT images of the upper cervical vertebrae showing their main functions in
Jo u
the stabilisation of the atlantoaxial joint. Blue: limitation of flexion
Green: limitation of extension
Red: limitation of rotation to the left Yellow: limitation of rotation to the right
Journal Pre-proof MRI Sequences Repetition time [ms] Echo time [ms] Slice thickness [mm] Gap [mm] Flip angle [°]
TSE T2w sag 1500 90 2.0 0.5 –
3D GE T1w dor 25 6.89 0.8 0.8 –
TSE T1w tra 533 7.51 3.0 0.5 –
3D WATS tra 22 11 25 – 0.75
Table 1: MRI parameters for Turbo spin echo T2-weighted (TSE T2w), 3D Gradient echo T1weighted (3D GE T1w), Turbo spin echo T1-weighted (TSE T1w) and 3D Water Selective (WATS) sequences. Repetition time and echo time are shown in ms, slice thickness and gap in mm and flip angle in degrees.
e-
Pr Beagle 2 15.2 15.6 – – 52 49 57 59
al
rn
L2 | L3 (ligament intact) L2 | L3 (ligament cut) Angle (ligament intact) Angle (ligament cut)
1 14 – 65 70
pr
a. 0.39 b. 0.29 c. 1.66
Table 2: Ligament to dens ratio (LDR)
Beagle 14.5 – 63 73
oo f
Ligament to dens ratio (LDR) 0.37 0.99 1.1
Dorsal atlantoaxial ligament (L1 ) Apical ligament of the dens axis Alar ligaments (L4-5 ) Transverse ligament : a. Central width (L6 ) b. Insertional widths (L7-8 ) c. Length (L9 )
Beagle 3 14.0 13.8 15.9 15.3 60 57 65 63
Beagle 4 14.3 13.8 17.2 16.1 71 67 76 74
Beagle 5 13.8 15.5 16.4 18.2 70 66 74 71
Jo u
Table 3: Measurements of the dorsal atlantoaxial ligament Lengths L2 (left column) and L3 (right column) in mm as well as maximal rotation angles in degrees (the left column refers to the rotation of the head to the left side and the right column to the right side) before and after transecting the dorsal atlantoaxial ligament. Delayed introduction of the second parameter explains the lack of data for Beagle 1 and 2.
Journal Pre-proof Beagle 1
Beagle 2
Beagle 3
Beagle 4
Beagle 5
Maximal rotation left
9.9
12.7
10.8
13.3
9.9
13.6
10.4
12.8
10.2
13.1
Rotation 45° left
10.2
12.2
11.1
12.8
10.4
12.2
11.0
11.9
11.1
12.6
Physiological position
11.6
11.4
12.6
12.3
11.5
11.6
11.4
11.5
11.5
11.7
Rotation 45° right
12.0
10.6
13.2
10.7
12.0
10.5
12.2
10.8
11.6
10.8
Maximal rotation right
13.1
8.9
13.8
10.0
13.5
10.1
13.1
10.1
12.7
10.4
Maximal extension
9.8
10.2
11.2
11.6
10.2
9.9
10.7
10.9
10.8
10.9
Maximal flexion
12.5
12.8
13.3
13.1
12.5
12.8
12.6
12.8
12.1
12.4
Beagle 2 4.4 3.9 3.0 3.2 3.1 3.1 16.6
4.6 3.3 3.2
3.9 3.4 3.3
Beagle 3 4.5 4.1 2.9 3.1 2.9 3.0 18.4
pr
Beagle 1 4.7 4.6 3.1 3.2 3.1 3.2 19.8
3.9 3.1 3.0
Beagle 4 4.7 4.3 2.7 2.9 2.5 2.8 16.2
4.2 3.1 3.0
Beagle 5 4.3 3.8 3.0 3.2 2.9 3.1 16.5
e-
Central width (L6 ) Insertional width left (L7 ) Insertional width right (L8 ) Length (L9 )
oo f
Table 4: Measurements of the alar ligaments The first column indicates the position of the head. For each dog, the left column refers to the length of the left alar ligament (L4 ) and the right column to the length of the right alar ligament (L5 )
Beagle 2 10.7 5.8 36.3 22.6
rn
Beagle 1 10.4 5.9 43.9 26.9
Jo u
Length [mm] Width [mm] Covering rate [%] Contact area [mm2 ]
al
Pr
Table 5: Measurements of the transverse ligament The three squares of a column show the size of the transverse ligament in mm in the following head positions: maximal flexion, physiological position and maximal extension – respectively from left to right.
Beagle 3 10.8 4.9 38.0 20.1
Beagle 4 10.5 5.5 41.1 23.7
Beagle 5 10.4 5.3 36.6 20.2
Average 10.6 5.5 39.2 22.7
Table 6: Measurements of the dens axis The length and the width of the dens axis are shown in mm, the covering rate in % and the contact area in mm 2 .
3.6 3.3 3.2
Journal Pre-proof
Graph 1: Mean lengths of the alar ligaments The horizontal axis legends indicates the head position during each measurement of the alar ligaments. Grey: Left alar ligament (L4 ) Blue: Right alar ligament (L5 )
pr
oo f
Graph 2: Graphical representation of the average distribution of the collagen fibres’ birefringence colours in histological sections of the alar and transverse ligaments taking into account (a.) and disregarding (b.) the amorphous ground substance. The columns’ colours correspond to the real birefringence colour observed on histological sections under polarized light. The black columns represent the amorphous ground substance. The values are shown in percent.
Jo u
rn
al
Pr
e-
Graph 3: Mean lengths of the transverse ligament The horizontal axis indicates the four measurements describing the transverse ligament. Representing the length (L9 ), the last columns only refer to the y-scale on the right hand side. The three columns in each group refer to the measurements as a function of the head’s position, whereby the colours represent the following positions: Grey: maximal flexion Light blue: neutral position Dark blue: maximal extension
Journal Pre-proof Highlights MRI reveals all the ligamentous structures of the atlantoaxial joint.
Head flexion with rotation results in maximum distension of the alar ligaments.
Histologically, the transverse ligament’s structure was particularly homogeneous.
Alar and transverse ligaments differ in amount and distribution of elastic fibres.
The ligaments’ internal structure is closely related to their specific function.
Jo u
rn
al
Pr
e-
pr
oo f
Figure 1
Figure 2
Figure 3
Figure 4
Figure 5
Figure 6
Figure 7