Computed Tomographic Evaluation of Adjacent Segment Motion after Ex Vivo Fusion of Equine Third and Fourth Cervical Vertebrae

Abstract Objective Surgical fusion of vertebral segments is a treatment option for horses with cervical stenotic myelopathy or cervical fracture. Degenerative disease affecting adjacent vertebral segments is a reported complication following surgical vertebral fusion in other species, termed adjacent segment disease. The aim of this study was to evaluate the impact of cervical vertebral fusion on the biomechanics of adjacent vertebral segments in the horse. Study Design Neck specimens of 12 horses were assessed using computed tomographic imaging. Range of motion (ROM) was determined by measuring the maximum sagittal flexion, extension and lateral bending between C2 and C5. C3/4 was subsequently fused using a standard locking compression plate and locking head screws and computed tomographic scans and ROM measurements were repeated. Results Prior to intervertebral fusion, a significant increase in ROM along the vertebral segments from cranial to caudal was observed. Range of motion measurements of C3/4 decreased significantly after fusion (p = 0.01). Range of motion of the adjacent segments (C2/3 and C4/5) did not change significantly after fusion. Conclusion Fusion of one cervical intervertebral joint did not affect the ROM of the adjacent vertebral segments. Further research investigating the implications of vertebral fusion on the intervertebral pressure in the equine patient is indicated.


Introduction
The cervical spine is currently receiving increased attention as a source of dysfunction in the equine patient. 1,2 The range of motion of the seven cervical vertebrae (C 1-7 ) has major implications for equine locomotor function as the neck acts as a lever and aids the horses balance. 3,4 The three main types of movement occurring in the motion segments of the equine spine include flexion and extension, lateral bending and axial rotation. Additionally, vertical and transversal translation as well as longitudinal compression play a minor role in the equine cervical spine range of motion. 5 A spinal motion segment is a functional unit, composed of two adjacent articulating vertebrae and the connecting tissues binding them. 6 The range of motion of the spinal motion segments is defined as the difference between full extension and flexion as well as full right and left lateral bending, respectively. 7 The first two cervical joints have a unique morphology, utilizing specific types of movement. Compared with the other cervical motion segments, the atlantooccipital joint has the maximum dorsoventral flexion-extension range of motion whereas maximum rotational movement occurs between C 1 and C 2 . 3 The caudal cervical vertebrae (C 3 -C 7 ) are more homogenous in their morphology. 8 A kinematic study conducted on cadaveric specimens identified an increase in range of motion from the cranial to the caudal motion segments (C 2/3 -C 6/7 ) for dorsoventral flexion-extension as well as lateral bending. 3 However, in an in vivo study more range of motion was observed in the cranial cervical spine (C 1/2 -C 4/5 ), while C 5/6 showed the smallest range of motion in both directions. 9 Severe pathological conditions including traumatic vertebral fractures or luxations as well as developmental cervical vertebral malformations are commonly treated with surgical ventral cervical vertebral fusion in the horse. [10][11][12] The surgical techniques described include the use of locking compression plates, 13 stainless steel Cloward Bagby baskets, 14 titaniumbased kerf cut cylinders 11,12 or a polyaxial pedicle screw and rod construct. 1 Locking compression plate fusion had similar mechanical properties to the use of a kerf cut cylinder in a biomechanical evaluation in cadaveric specimens which may justify its use in clinical cases. 14 The fusion of a motion segment in the human cervical spine can result in a condition termed adjacent segment disease. Adjacent segment disease is characterized by pathological processes including joint and intervertebral disc degeneration in the motion segments adjacent to the fused motion segment. The condition is thought to be caused by an increase in range of motion and higher loads acting upon adjacent motion segments after intervertebral fusion. 15 A higher incidence of adjacent segment disease has been observed when plates were used instead of kerf cut cylinder for intervertebral fusion in humans. 15 The compensatory increase in motion of C 4/5 was also observed in a canine cadaveric study after fusion of C 5/6 . 16 As cervical fusion is performed in horses with various pathological conditions, it is plausible that adjacent segment disease is of concern in horses as well. However, to the authors' knowledge there is no study investigating the range of motion of adjacent motion segments after cervical vertebral fusion in the equine species.
The aim of this study was to evaluate the momentary state of the biomechanical impact of cervical vertebral fusion on adjacent motion segments in the horse. The authors hypothesized that the range of motion of the C 2/3 and C 4/5 segments will increase when force is applied in flexion and extension, as well as in lateral bending after cervical fusion of C 3/4 using locking compression plate fixation.

Neck Specimens
Cervical specimens of 12 Warmblood horses humanely destroyed for reasons unrelated to this study and with no clinical history or findings indicating cervical pathology were included in the study. Samples were obtained from six geldings and six mares, aged 3 to 31 years (mean 12 years). Informed owner consent for tissue retention was obtained and approval for the study was given by the local Committee on Research Ethics.
Specimens were excised at the atlantooccipital joint and at the cervicothoracic junction. The musculature was removed to the level of the intertransversarii cervices and longus colli muscles, all joint capsules were left intact. The specimens were frozen at -20°C immediately after dissection and subsequently thawed over 24 hours at 10°C prior to biomechanical testing. 14

Testing Device
A custom-made motorized testing device was used to position and manipulate the neck specimens to perform dynamic computed tomographic (CT) scanning (►Fig. 1). The device held the specimen in neutral position and additionally enabled consistent, controlled movement of the motion segments in flexion, extension and lateral bending. The caudal part of the cervical spine was fixed in the device and C 1 was attached to a rope which led through a pulley and was attached to a rope winch. Additionally, a wooden frame was used to support the weight of the specimen during motion. The two motion segments caudal to the motion segment that was evaluated were not attached to the frame to allow full range of motion of these adjacent segments. To standardize testing conditions, the applied force was limited to 50 Nm and the speed to 0.05 m/s. 17 Fig. 1 Equine neck specimen (C 1 -C 7 ) positioned in custom-made motorized biomechanical testing device. The specimen is placed in neutral position to perform computed tomography (CT) scanning (320-detector-row CT scanner, Aquilion One, Canon Medical Systems) of the motion segment C 2/3 (field of view indicated in red). Note the rope that is attached to C1 and leads through a pulley (white arrow) to a rope winch (asterix), facilitating controlled movement of the specimen. Additionally, the caudal part of the cervical spine is fixed in the device (blue lines). The two motion segments caudal to the motion segment that is evaluated are not attached to the frame to allow full range of motion of these adjacent segments.

Computed Tomographic Imaging
The CT scans were performed using an intermittent sequential mode with a 320-detector-row CT scanner (Aquilion One; Canon Medical Systems, Zoetermeer, the Netherlands). A field-of-view of 512 Â 512 pixels, 0.5 mm slice thickness and a tube rotation time of 0.35 seconds at 100 kVp and 280 mAs (100 effective mAs) were used. A total of five CT scans including neutral, maximal flexion, maximal extension, maximal left and right lateral bending were performed of each neck specimen prior to surgical vertebral fusion. All five scan sequences were repeated after fusion.

Intervertebral Fusion
Following the initial CT scans, surgical fusion of the third and fourth cervical vertebrae was performed within 48 hours of thawing based on a previously described protocol. 14 In brief, the longus colli muscles were separated in midline (10-15 cm incision) at the level of C 3/4 . The muscle bellies were retracted abaxially to allow access to the ventral aspect of the vertebrae. The surface of the ventral spinous process of the body of C 3/4 was flattened using a curved osteotome and disc material was removed from the intercentral articulation between C 3/4 with a high-speed drill. A standard 4.5-mm broad seven-hole locking compression plate (DePuy Synthes; Umkirch, Germany) plate was contoured and six 5.0 mm self-tapping locking head screws were subsequently inserted under radiographic guidance. The screws were applied in the following order: 3, 5, 1, 7, 2, 6 from cranial to caudal (►Fig. 2). Accurate screw placement was achieved by ensuring that the third screw consistently engaged the most caudal aspect of the body of C 3 without interfering with the intercentral articulation. The central screw hole of the plate, 4 was positioned over the intercentral articulation; no screw was inserted here. Following C 3/4 fusion, the five CTscans of the motion segments C 2 to C 5 were repeated as described earlier.

Range of Motion Measurements
Based on the acquired CTscans, three-dimensional reconstructions of the vertebra and their articulations were created using a commercially available software (OsiriX, Bernex, Switzerland). Measurements were performed on multiplanar reconstructed images in neutral position and maximum flexion, extension, left and right lateral bending for the motion segments C 2/3 , C 3/4 and C 4/5 (►Figs. 3 and 4). The range of motion of each vertebral motion unit was defined as the difference between the neutral position and the maximum angles of flexion-extension and right/left lateral bending. 4 The total dorsoventral range of motion and the total lateral range of motion of each motion segment unit was calculated by adding

Data Analysis
Data were recorded in Excel (Microsoft Inc., Redmond, Washington, United States) and analysed in SPSS (IBM SPSS Statistics Version 24; Armonk, New York, United States).
Descriptive analysis was performed to compare the range of motion of the cervical vertebral motion segments (C 2-5 ) before and after vertebral fusion. Data distribution was assessed and found not to be normally distributed. Data were analysed using the Wilcoxon signed-rank test for pairwise comparison of nonparametric data. The intra-observer agreement was assessed by taking all measurements twice within 1 month (first author) and the inter-observer agreement was based on a comparison of the assessment of the first and the last author. For the intra-and inter-observer agreement, Cohen's Kappa as measure of agreement was calculated. For all tests, a p-value of < 0.05 was considered significant.

Range of Motion after Intervertebral Fusion
The total range of motion for the C 3/4 motion segments did decrease significantly after fusion in both flexion and extension (C 3/4 before fusion vs. C 3/4 after fusion p ¼ 0.01) and in left and right lateral bending (C 3/4 before fusion vs. C 3/4 after fusion p ¼ 0.01) (►Table 1). The total range of motion of the adjacent cranial motion segments (C 2/3 ) did not change significantly after surgical fusion of C 3/4 in flexion-extension (C 2/3 vs. C 3/4 p ¼ 0.31) or in left/right lateral bending (C 2/3 vs. C 3/4 p ¼ 0.18). There was also no significant difference in the total range of motion of the adjacent caudal motion segment (C 4/5 ) after fusion of C 3/4 in flexion-extension (C 3/4 vs. C 4/5 p ¼ 0.64) or lateral bending (C 3/4 vs. C 4/5 p ¼ 0.07) when compared with the baseline values before surgical fusion (►Figs. 5 and 6).

Intra-and Inter-Observer Agreement
The assessment of the intra-and inter-observer agreement identified kappa values of 0.91 and 0.97 for flexion-extension and 0.87 and 0.97 for left and right lateral bending.

Discussion
The range of motion of equine cervical spine specimens was assessed before and after fusion of the motion segments C 3/4 using CT imaging. No significant increase in the range of motion was identified in the cervical motion units adjacent to C 3/4 after fusion. The observations of this study differ from the findings described in studies assessing the biomechanical properties of the canine and human cervical spine range of motion after vertebral fusion and contradict the authors' hypothesis. 16,18   The range of motions determined before intervertebral fusion in the current study were comparable to those previously reported. 3 The range of motion in the caudal cervical spinal motion segments was significantly larger than in the cranial cervical motion segments in flexion-extension and lateral bending. One possible explanation for the larger range of motion in the caudal cervical spine is thought to be the more medial orientation of the articular process joints in this region, facilitating an increased range of motion. 3,9,19 The total range of motion of the C 2/3 segment in flexion-extension was approximately 10°less when compared with the previous report. 3 This finding might be explained by the different preparation of the specimens. In contrast to the previous study, the intertransversarii cervices and longus colli muscles were left intact in the current study. 3 Additionally, it should be considered that the muscle tone in the living animal might further increase the stiffness of the vertebral segments.
The pathogenesis of adjacent segment disease is widely unknown and the investigation of the condition is mainly based on in vitro studies. 20 Depending on the assumed postoperative motion behaviour of the patient, three in vitro protocols are available for biomechanical testing of the spine in man. 20 The spinal range of motion is assessed based on a flexibility-, stiffness-or hybrid protocol and load can be applied using eccentric forces or a pure moment. The force applied with a pure moment is currently defined as the 'gold standard' for in vitro biomechanical testing of the spine. 20 The purpose of pure moment loading is to supply the same magnitude of force at each cross-section throughout the whole length of the construct. 21 The flexibility protocol used in the current study is similar to the previously performed canine study in which a predefined load was applied to the free end of the spinal specimens resulting in a pure bending moment. 16 In contrast to the current study where the intrinsic musculature remained on the vertebrae, all soft tissue structures were removed in the canine model and only the joint capsules and major ligaments were left intact. 16 The lack of soft tissue attachments may explain the increased range of motion in the adjacent motion segments after vertebral fusion in the canine specimens.
Instead of pure moments, eccentric compressive forces with cyclic loading were applied to the cervical vertebrae in the main human study describing adjacent segment disease. 18 The compressive flexion of the specimen that changes the lever arm at each motion segment complicated this testing approach. 22 Cyclic loading is generally used to apply load on a structural component like an implant, to assess the structure's point of fatigue. 23 Cyclic loading does not reflect the physiological movement of the intact motion segments of the cervical spine in horses. The flexibility protocol with the use of pure moments for load application was also utilized in several biomechanical in vitro studies for the assessment of the human spine; however, an increase in range of motion of adjacent segments after vertebral fusion was only observed in one out of eight of those studies. 20 Additionally, anatomical differences between species might explain the varying results of the described protocols. 24 An increased range of motion of adjacent motion segments has been identified in canine specimens after vertebral fusion was performed using a titanium plate system, or metal implants and polymethylmethacrylate. 16 Similar findings were observed after vertebral fusion in the area of the human cervical spine using a custom designed external fixation device. 18 All of the different devices used for intervertebral fusion, including the locking compression plate construct used in the current study, resulted in a significant decrease in the range of motion of the fused motion segment. However, in contrast to the canine and human studies, the range of motion of the adjacent segments did not change significantly after fusion in the equine cervical spine specimens. In human medicine the debate is ongoing as to whether adjacent segment disease develops due to the increased biomechanical load on the intervertebral discs adjacent to the cervical fusion sites or is caused by pre-existing degenerative disc disease, characterized by disc fragmentation or protrusion. 15 Current literature suggests a multivariate aetiology of adjacent segment disease rather than the condition being a secondary compensation of the lost motion in a fused segment. 15 This assumption is mainly based on the fact that spinal fusion is generally performed in spines with pre-existing disease and that the degenerative changes observed in the adjacent segments after fusion may be part of a pre-existing and ongoing process of degenerative disc disease. [25][26][27] Disc degeneration is the most commonly observed pathological feature of adjacent segment disease in humans. 28 As shown in human and canine biomechanical in vitro studies, the development of disc degeneration is generally attributed to an increased range of motion and pressure acting on adjacent motion segments after spinal fusion. 16,29,30 Different grades for intervertebral disc degeneration in horses have recently been evaluated and reported. 2 The equine intervertebral disc including the nucleus pulposus is generally described to be of fibrous or fibrocartilaginous consistency. 31,32 However, a recent anatomical study revealed that the equine annulus fibrosus and nucleus pulposus have a higher similarity (grossly and histologically) to those intervertebral structures in humans and dogs than previously reported. 2 A small number of case reports suggest that neurological deficits might be related to lesions of the intervertebral discs in horses. [33][34][35] To the authors' knowledge scientific research investigating the significance of intervertebral disc degeneration in horses is lacking. However, slightly different anatomical conditions may be a possible explanation for a lower or even a lack of susceptibility to adjacent segment disease in the equine patient. The intervertebral pressure in the area of the intervertebral discs was not assessed in the current study. Further investigation of the clinical significance of intervertebral disc degeneration as well as intervertebral pressure measurements of adjacent segments before and after intervertebral fusion is required to further investigate the implications of adjacent segment disease in the horse.
The current study is limited by the ex vivo protocol used for the assessment of the cervical spine range of motion and the observation that the data were not normally distributed. The forces acting upon the equine cervical spine during physiological movement are described to lie at approximately 40 to 80 Joules. 17 When a force of 50 Nm was applied in the current study this equated 25 to 50 times the force applied in the canine study and no further movement could be elicited by application of a force greater than 50 Nm (1 Joules ¼ 1 Nm). 16 For the purposes of this study, the spine was excised at the level of the atlantooccipital and the cervicothoracic junction and it is conceivable that this might influence the range of motion of the cervical motion segments when compared with conditions found in a complete spine. Additionally, the fixation of the cervical spine specimen could have impaired on the motion segments range of motion. However, the two motion segments cranial and caudal to the fused segment were not attached to the testing device during CT evaluation. The post mortem evaluation of the specimens does not take the entire function of the ligaments and musculature into account and only gives a momentary state rather than an assessment over time. While osteoarthritis and secondary changes can strongly influence the kinetics of the cervical spine, no pre-existing pathological conditions were noted in the specimens used for the study. Axial rotation as one of the three main movements of the equine intervertebral joints was not considered in the current study since previous studies demonstrated that the axial rotation in the motion segments of interest (C 2-5 ) is minimal. 3 In conclusion, the results of this study did not show any significant association between fusion of a cervical motion segment and change in range of motion of the adjacent vertebral segments. However, apart from adjacent segment disease due to an increase in range of motion of adjacent vertebral segments, adjacent degenerative disc disease is commonly diagnosed in other species after vertebral fusion. Increased pressure acting on the intervertebral joint after adjacent vertebral fusion may also affect the intervertebral disc in the equine patient. Further research investigating the implications of vertebral fusion on the intervertebral pressure in the equine cervical spine is indicated.