Radiography is the most commonly used technique for evaluation of the cervical portion of the vertebral column in horses1 and is often performed in the diagnosis of cervical stenotic myelopathy, with or without myelography, and articular process joint arthropathy.2–4 In the diagnosis of cervical stenotic myelopathy, measurement of cervical vertebrae sagittal diameters and intra- or intervertebral ratios is useful,1 despite some limitations in repeatability and intra- and interobserver agreement.5,6 Other types of measurements, such as those of vertebral alignment, the LAPJ, and the depth of the vertebral fossa,a are also used for clinical and research purposes. However, head and neck position can influence the observed distance between thoracic spinous processes and cervical intervertebral foramina dimensions in horses.7–9
The purpose of the study reported here was to determine the influence of neck position on the most commonly performed radiographic measurements of the cervical vertebral region in horses: MSD, intra-VSD ratio, inter-VSD ratio, LAPJ ratio, VAA, and VFA. Specifically, we hypothesized that measurements of inter-VSD ratio, LAPJ ratio, and VAA would differ with neck position.
Materials and Methods
Animals
Eighteen horses (9 females and 9 geldings) with a median age of 7 years (mean, 7.1 years; range, 5 to 14 years) were included in the study. Horses were classified as Arabian (n = 7), warmblood (7), and Anglo-Arabian (4). All were regularly ridden. Each had been referred to the Veterinary Teaching Hospital of the University of Perugia for examination for lameness, and each received complete lameness and neurologic examinations.
To be included in the study, horses were required to be deemed free of cervical disease through clinical and dynamic examination (walk and trot in a straight line, walk in a figure-8 pattern, and trot in a circle on hard and soft ground) and diagnostic imaging (no clinical important findings on ultrasonography and radiography of the cervical vertebral region), and owners were required to have given consent for enrollment. The study was performed in accordance with the guidelines of the Animal Care and Use Committee of the University of Perugia.
Radiographic examination
Each horse was first sedated with detomidine hydrochlorideb (0.01 mg/kg, IV), and then radiography of the cervical vertebral region was performed with the neck in 3 positions as described elsewhere9 (Figure 1): neutral (mouth at the level of the shoulder joint; with vertebral alignment of 45°), low (mouth at the level of the carpal joint; with vertebral alignment of 0°), and high (mouth at the level of the withers [highest point of the shoulders] with vertebral alignment of 60°). For image acquisition, horses were positioned squarely on all 4 feet, and attention was paid to avoid lateroflexion or rotation of the vertebral column. In each of the 3 different positions, 4 digital laterolateral (left to right) radiographs were obtained from C1 through T1 (ie, 1 radiograph for each of C1–3, C3–5, C5–7, and C6–T1).1 Exposure values were adjusted by horse size and imaged region. The cassette holder was positioned as closely as possible to the neck to minimize the amount of distortion.
Photographs showing 3 neck positions used for acquisition of laterolateral radiographs of the cervical vertebral region in horses8: neutral (A), low (B), and high (C). The degree of vertebral alignment used for each position is indicated (red lines), along with landmarks for position of the mouth (blue lines).
Citation: American Journal of Veterinary Research 79, 10; 10.2460/ajvr.79.10.1044
Radiographic measurements
As a preliminary step, intraobserver agreement in measurements of all assessed variables was determined by use of radiographs of 6 horses with the neck in a neutral position and 3 measurements were obtained for each site and for each variable. Afterward, all radiographs of the 18 horses were evaluated and measurements recorded. Radiographs of each horse (low, neutral, and high neck positions) were separated by neck position, and a random number was assigned to each, allowing measurement without direct knowledge of neck position or horse identity. The same observer (IS; a PhD student in equine radiology) performed all measurements.
A digital image analysis programc was used to evaluate the radiographs and obtain the measurements and ratios for each vertebra from C2 through T1 as described elsewhere1,a (Figure 2). The MSD was measured across the narrowest part of the vertebral foramen in a dorsoventral direction. The intra-VSD ratio was calculated by dividing the MSD by the dorsoventral height of the cranial aspect of the vertebral body of the same vertebra. The inter-VSD ratio was calculated by dividing the distance between the cranioproximal aspect of the vertebral body and the ventrocaudal aspect of the lamina of the immediate cranial vertebra (minimum intervertebral distance) by the dorsoventral height of the cranial aspect of the vertebral body. The LAPJ ratio was calculated by dividing the length of the superimposed left and right synovial joints, drawn by following the longest axis in a cranioventral to caudodorsal direction, by the MSD. The VAA was measured as the angle between 2 adjacent vertebrae by drawing 2 lines following the line of the floor of the vertebral canal of each vertebra, with the angle under the intersection of the 2 lines taken as the VAA (Figure 3). The VFA was measured as the angle between the line representing the height of the vertebral fossa (from the dorsal to the ventral angle) and the floor of the vertebral canal.
Laterolateral (left to right) radiographic views of C4–5 in a horse free of cervical disease illustrating how various measurements were performed. A—The intra-VSD ratio was obtained by dividing the MSD of the vertebra of interest (a) by the dorsoventral height of the cranial aspect of the vertebral body of the same vertebra (b). B—The inter-VSD ratio was obtained by dividing the distance between the cranioproximal aspect of the vertebral body of the vertebra of interest and the ventrocaudal aspect of the lamina of the immediately cranial vertebra (c) by the dorsoventral height of the cranial aspect of the vertebral body (b) of the vertebra of interest. C—The LAPJ ratio was obtained by dividing the length of the superimposed left and right articular processes (d) of the vertebra of interest by the respective MSD (a). The VFA (e) is shown, representing the angle between the height of the vertebral fossa and the floor of the vertebral canal. D—The VAA was measured as the angle between 2 adjacent vertebrae (f).
Citation: American Journal of Veterinary Research 79, 10; 10.2460/ajvr.79.10.1044
Laterolateral radiographic views of C3–5 in the same horse, with its neck in a neutral (A) and high (B) position. Notice the greater VAA in the high versus neutral position. See Figure 1 for definitions of neck positions.
Citation: American Journal of Veterinary Research 79, 10; 10.2460/ajvr.79.10.1044
Statistical analysis
A worksheet applicationd and statistical softwaree were used for statistical analyses. Intraobserver agreement for each assessed variable was evaluated by means of repeated-measures ANOVA and calculation of intraclass correlation coefficients. Measurements for the rest of the study are reported as mean ± SD. Data were considered as continuously dependent, with horse as the experimental unit of interest. The Shapiro-Wilk and Levene tests were used to verify homoscedasticity of the data. The influence of neck position on the performed measurements was assessed with the paired t test (2 tailed); differences from the neutral position for the high and low positions were assessed through pairwise comparison. Values of P < 0.05 were considered significant.
Results
Results of preliminary measurements indicated high intraobserver agreement, with intraclass correlations ranging from 0.74 to 0.99 and no significant differences identified among measurements (Supplementary Table S1, available at http://avmajournals.avma.org/doi/suppl/10.2460/ajvr.79.10.1044). Data for measurements of C2 through C7 derived from laterolateral radiographs of the 18 horses obtained with their necks in 3 positions were summarized after confirming that the data were normally distributed with equal variances (Tables 1 and 2). Compared with the neutral neck position, the high position resulted in a significant (P = 0.004) increase in VAA (mean increase of 2°) at C3–4, and the low position resulted in a significant (P = 0.03) decrease in VAA (mean decrease of 2°) at C4–5. Also compared with the neutral neck position, the high position resulted in a significant (P = 0.04) increase in the inter-VSD ratio at C4–5 (mean increase of 4%) and a decrease in the LAPJ ratio at C2–3 (P = 0.01), C3–4 (P < 0.001), and C4–5 (P < 0.001; mean decreases of 12%, 27%, and 27%, respectively). No significant differences were identified between the high or low neck position and the neutral position regarding MSD, intra-VSD ratio, and VFA.
Mean ± SD values of MSD and intra-VSD ratio for C2 through C7 derived from laterolateral radiographs of 18 horses with their necks in 3 positions.
| Measurement | Neck position | C2 | C3 | C4 | C5 | C6 | C7 |
|---|---|---|---|---|---|---|---|
| MSD (mm) | Low | 27.9 ± 2.9 | 24.7 ± 2.7 | 23.4 ± 1.8 | 24.2 ± 2.4 | 29.1 ± 3.7 | 30.9 ± 2.6 |
| Neutral | 28.1 ± 2.6 | 24.9 ± 2.7 | 23.6 ± 2.1 | 25.3 ± 2.0 | 28.7 ± 3.5 | 32.1 ± 4.0 | |
| High | 27.9 ± 2.4 | 25.0 ± 2.4 | 23.5 ± 1.8 | 25.6 ± 2.7 | 28.5 ± 2.2 | 31.9 ± 2.7 | |
| Intra-VSD ratio | Low | 0.78 ± 0.07 | 0.70 ± 0.01 | 0.61 ± 0.04 | 0.58 ± 0.08 | 0.61 ± 0.07 | 0.66 ± 0.07 |
| Neutral | 0.78 ± 0.08 | 0.69 ± 0.04 | 0.64 ± 0.05 | 0.60 ± 0.06 | 0.61 ± 0.05 | 0.70 ± 0.09 | |
| High | 0.76 ± 0.05 | 0.70 ± 0.03 | 0.64 ± 0.03 | 0.61 ± 0.07 | 0.62 ± 0.06 | 0.69 ± 0.05 |
See Figure 1 for full descriptions of the 3 neck positions.
Mean ± SD values of inter-VSD, LAPJ ratio, VAA, and VFA for C2 through C7 derived from the laterolateral radiographs in Table 1.
| Measurement | Neck position | C2–3 | C3–4 | C4–5 | C5–6 | C6–7 |
|---|---|---|---|---|---|---|
| Inter-VSD ratio | Low | 1.00 ± 0.12 | 0.84 ± 0.08 | 0.82 ± 0.06 | 0.81 ± 0.09 | 0.83 ± 0.08 |
| Neutral | 1.00 ± 0.09 | 0.82 ± 0.07 | 0.84 ± 0.07 | 0.82 ± 0.11 | 0.83 ± 0.1 | |
| High | 1.00 ± 0.11 | 0.84 ± 0.06 | 0.88 ± 0.07* | 0.88 ± 0.08 | 0.84 ± 0.01 | |
| LAPJ ratio | Low | 1.47 ± 0.23 | 1.71 ± 0.17 | 1.84 ± 0.22 | 1.92 ± 0.26 | 1.93 ± 0.19 |
| Neutral | 1.48 ± 0.18 | 1.65 ± 0.18 | 1.84 ± 0.22 | 1.90 ± 0.26 | 1.94 ± 0.21 | |
| High | 1.31 ± 0.14* | 1.20 ± 0.20* | 1.35 ± 0.11* | 1.80 ± 0.25 | 1.86 ± 0.31 | |
| VAA (°) | Low | 168 ± 5 | 173 ± 4 | 174 ± 3* | 175 ± 3 | 174 ± 3 |
| Neutral | 169 ± 4 | 172 ± 4 | 176 ± 2 | 176 ± 3 | 175 ± 3 | |
| High | 170 ± 4 | 174 ± 4* | 177 ± 3 | 177 ± 2 | 175 ± 3 | |
| VFA (°) | Low | 114 ± 2 | 123 ± 4 | 124 ± 4 | 124 ± 3 | 124 ± 5 |
| Neutral | 115 ± 2 | 122 ± 4 | 124 ± 4 | 125 ± 4 | 125 ± 3 | |
| High | 116 ± 4 | 124 ± 3 | 124 ± 3 | 125 ± 3 | 125 ± 3 |
Value differs significantly (P < 0.05) from the respective value for the neutral neck position.
See Figure 1 for full descriptions of the 3 neck positions.
Discussion
Several previous in vivo and ex vivo studies7–9 have shown that, in horses, the position of the head and neck during acquisition of radiographs can influence the distance between the spinous processes of the thoracic vertebrae and the intervertebral foramina dimensions of the cervical vertebral region. In the present study, we investigated the influence of various neck positions on some measurements performed on laterolateral radiographs of this region that were used to identify compression of the cervical segment of the spinal cord by vertebral malformation and malarticulation or cervical osteoarthropathy.1,2,5,10,a Minimum sagittal diameter, intra-VSD ratio, and inter-VSD ratio were chosen because these are commonly used in the diagnosis of cervical vertebral stenotic myelopathy. The VAA and VFA were also chosen because they can be used to assess vertebral malalignment and vertebral fossa dysplasia, respectively.a
The LAPJ was measured and a ratio calculated to avoid confounding effects of different horse sizes and radiographic magnification. The size of the cervical articular processes joint is commonly evaluated subjectively or on a semiquantitative scale; however, the articular processes of the caudal cervical vertebrae, particularly at the C5–6 joint, can enlarge with age, and therefore, enlargement is not always associated with clinical signs.4 Previous studies10,11 in which objective measurements of the cervical articular processes were obtained on lateral and oblique radiographic views showed that the cranial articular process of C6 (C5–6 joint) and C7 (C6–7 joint) are significantly larger in horses with cervical arthropathy than in horses without cervical arthropathy.
The operator who performed the radiographic measurements in the present study was unaware of the identities of the horses and neck positions being evaluated; however, complete blinding with respect to neck position was impossible because the neck angle changes between the different positions. To minimize the possible bias, radiographs of each horse were separated by neck position and numbered in random order.
Supporting our hypotheses, some significant differences in values were identified among the neck positions of the horses in the present study, affecting more commonly the C3-C5 area and specifically the VAA, inter-VSD ratio, and LAPJ ratio. No changes were identified regarding the MSD, intra-VSD ratio, and VFA. The MSD and intra-VSD ratio are among the most common measurements performed on lateral radiographs of the cervical vertebral region in horses with spinal ataxia for the diagnosis of vertebral stenosis and malalignment, respectively.1,2,5 The finding that these values did not change significantly with neck position would be useful information in clinical practice given that the use of α2-adrenoceptor agonists is recommended for sedation of horses for radiographic evaluation of the vertebral column,12 but these drugs can induce a certain grade of ataxia and thus worsen an existing ataxia. In the authors' experience, clinicians may reduce the dosage for horses with signs of ataxia, resulting in suboptimal positioning during radiography. However, according to the results of the present study for horses free of cervical disease, the MSD and intra-VSD ratio would not be affected, although the inter-VSD ratio might be.
Although the difference between the high and neutral neck positions in the inter-VSD ratio at C4–5 was significant in the present study, the mean difference was small (4%). This amount of variation could be unlikely to result in a false-negative diagnosis, but additional research would be necessary to verify whether the various neck positions have a different influence on horses with clinical signs of ataxia (eg, those with cervical spondylomyelopathy). In fact, radiographs of the cervical vertebrae are often obtained during hyperflexion and extension for ataxic horses to identify small pathological changes.
Compared with values in the neutral neck position, an increase in the VAA at C3–4 was identified with the high position and a decrease in VAA at C4–5 with the low position for horses of the present study. As observed for the inter-VSD ratio at C4–5, these mean differences were low (2°). Such differences may be explained by the fact that with the horse's neck in the high position, extension of the cervical area occurs, consisting of sliding of the cranial aspect of each vertebral body ventrally against the vertebral fossa of the preceding vertebra. On the other hand, with the neck in the low position, the opposite condition occurs, consisting of mild flexion with sliding of the cranial aspect of each vertebral body dorsally against the vertebral fossa.13 Extension of the cervical area results in an increased distance between the cranioproximal aspect of the vertebral body and the ventrocaudal aspect of the lamina of the immediate cranial vertebra and in a greater inter-VSD ratio as the cranioproximal aspect of the vertebral body moves ventrally.13,14 Because of these movements, the angle between the floor of 2 adjacent vertebrae increases during extension and decreases during flexion.
Changes in vertebral alignment with different neck positions were identified in horses in a previous study,7 in which angles increased between cervical vertebrae during extension of the neck and decreased during flexion, similar to findings in the present study. However, in that ex vivo study,7 most of the changes in vertebral alignment were in the caudal cervical area. It is possible that the lack of proximal forelimb and cervical musculatures (except for the multifidus and longies colli muscles) in the previous study7 was responsible for the differences in some findings between that study and the present study. Indeed, considering the observed mean values in the present study without regard to whether they differed significantly from those with the neutral neck position, the VAA increased in the high position and decreased in the low position at every considered segment. Comparison with kinematic studies15,16 involving evaluation of different angulations of the cervical vertebrae in cervical positions commonly used in equestrian sporting events is not easy, given that the neck position in kinematics studies differs from those used in the present study for radiographic purposes.
Other significant differences were identified in the present study that affected the LAPJ ratio such that it was significantly lower than in the neutral position at C2–3, C3–4, and C4–C5 when the head was in the high position. Again, this result was likely the consequence of the biomechanical movements of the vertebrae during flexion and extension. In fact, the amount of overlap between caudal and cranial articular processes of 2 adjacent vertebrae increases during neck extension,13,14 resulting in a lower ratio given that the MSD is not influenced by neck position. If objective measurements of the articular processes were used to assist the diagnosis of cervical arthropathy in horses, an incorrect neck position could lead to false-negative results. However, in the clinical context, it is important to remember that a single measurement should not be used alone to define the presence of cervical arthropathy and that radiographic findings such as bone remodelling, new bone proliferation, asymmetry of the articular processes, or bone fragmentation, together with clinical signs, provide stronger evidence to confirm or exclude cervical arthropathy.1 Nevertheless, the LAPJ ratio for the C5–6 and C6–7 articular processes joints, which are common sites for arthrotic changes of the cervical vertebral region,1,3,4,11 did not differ significantly in the present study, even when the ratio decreased from the low or neutral neck position to the high position.
The lack of an influence of neck position on measurements obtained for single or multiple parts of the same vertebra in horses was demonstrated in the present study, as indicated by values for MSD, intra-VSD ratio, and VFA. These findings were logical because although movement can change the anatomic disposition and relationship between vertebrae, it cannot influence the anatomic features of a single vertebra.
The size, shape, and visibility of the intervertebral foramina were not assessed in the study reported here because radiographic abnormalities in this area are commonly responsible for forelimb lameness as a consequence of nerve root impingement secondary to cervical articular processes arthropathy, and the included horses had been evaluated for lameness. However, a previous study9 showed how the distance between the caudal epiphysis and the cranial articular process of the subsequent vertebra (intervertebral foramen height) from C4 through C7 as measured on radiographs is significantly greater in horses when the neck is in flexion than when it is in a resting position or in extension. In agreement, an ex vivo study7 showed that extension of the cervical vertebral region causes a decrease in intervertebral foramina dimensions at C4-T1 as determined via CT.
Overall, the present study showed that commonly used radiographic measurements of the cervical vertebral region in horses free of cervical disease can be affected by neck position during imaging, but the changes induced by various neck positions were unlikely to have great clinical relevance. Although some commonly performed cervical radiographic measurements have poor reliability and intra- and interobserver agreement,6,17,18 these measurements continue to be used in equine clinical practice and research.5,a For this reason, clinicians need to be aware of another potential source of bias during radiograph acquisition (neck position) and should standardize the neck position of a horse to ensure high-quality imaging. However, given that some of these measurements were unaffected or only minimally affected by neck position in our study, measurements obtained for horses in which neck position cannot be standardized during radiography (eg, horses that cannot be sedated) may still be reasonably informative provided that the image quality is optimal.
ABBREVIATIONS
| LAPJ | Length of the articular processes joint |
| MSD | Minimum sagittal diameter |
| VAA | Vertebral alignment angle |
| VFA | Vertebral fossa angle |
| VSD | Vertebral sagittal diameter |
Footnotes
Guedj J. Anomalies cervicales identifiees radiographiquement chez des chevaux de CSO cliniquement sains. PhD thesis, École Nationale Veterinarie D'Alfort, France, 2009.
Domosedan, Pfizer Italia srl, Latina, Italy.
Osirix, Pixmeo SARL, Bernex, Switzerland.
Microsoft Excel, version 14.7.7, Microsoft Corp, Redmond, Wash.
R, version 3.3.3, R Foundation for Statistical Computing, Vienna, Austria.
References
1. Butler J, Colles C, Dyson S, et al. The spine. In: Butler J, Colles C, Dyson S, et al, eds. Clinical radiology of the horse. 4th ed. Oxford, England: Wiley-Blackwell, 2017;531–568.
2. Johnson AL, Reed S. Cervical vertebral stenotic myelopathy. In: Furr M, Reed S, eds. Equine neurology. 2nd ed. Oxford, England: Wiley-Blackwell, 2015:349–367.
3. Hepburn R. Cervical articular process disease, fractures, and other axial skeletal disorders. In: Furr M, Reed S, eds. Equine neurology. 2nd ed. Oxford, England: Wiley-Blackwell, 2015:386–400.
4. Down SS, Henson F. A radiographic retrospective study of the caudal cervical articular process joints in the horse. Equine Vet J 2009;41:518–524.
5. Hahn CN, Handel I, Green SI, et al. Assessment of the utility of using intra- and intervertebral minimum sagittal diameter ratios in the diagnosis of cervical vertebral malformation in horses. Vet Radiol Ultrasound 2008;49:1–6.
6. Scrivani PV, Levine JM, Holmes NL, et al. Observer agreement study of cervical-vertebral ratios in horses. Equine Vet J 2011;43:399–403.
7. Sleutjens J, Voorhout G, van der Kolk JH, et al. The effect of ex vivo flexion and extension on intervertebral foramina dimensions in the equine cervical spine. Equine Vet J Suppl 2010;42(suppl 38):425–430.
8. Berner D, Winter K, Brehm W, et al. Influence of head and neck position on radiographic measurement of intervertebral distances between thoracic dorsal processes in clinically sound horses. Equine Vet J Suppl 2012;43:21–26.
9. Berner D, Brehn W, Gerlach K. The influence of head and neck position on radiographic examination of the intervertebral foramina of the neck of horses at latero-lateral projection. Pferdeheilkunde 2012;28:39–45.
10. Lautenschläger I, Baumann I, Schulze M, et al. Die röntgenologische Darstellung der caudalen Facettengelenke der Halswirbelsäule beim Warmblutferd. Pferdeheilkunde 2009;25:544–553.
11. Hett AR, Busato A, Ueltschi G. Radiologische Messungen an der arthrotisch veränderten Halswirbelsäule des Pferdes – eine retrospektive, statistische Studie. Pferdeheilkunde 2006;22:241–249.
12. Weaver M, Barakzai S. Radiation safety and patient preparation. In: Weaver M, Barakzai S, eds. Handbook of equine radiography. London: Saunders Elsevier, 2010;9–12.
13. Denoix JM, Pailloux JP. Basic anatomy and physiology. In: Denoix JM, Pailloux JP, eds, Physical therapy and massage for the horse, 2nd ed, London: Manson Publishing, 2001;19–57.
14. Barone R. Articolazioni della colonna vertebrale. In: Barone R, ed, Anatomia comparata dei mammiferi domestici. Vol 2°-parte I Artrologia. 3rd ed. Bologna, Italy: Edagricole, 2004;53–86.
15. Elgersma AE, Wijnberg ID, Sleutjens J, et al. A pilot study on objective quantification and anatomical modelling of in vivo head and neck positions commonly applied in training and competition of sport horses. Equine Vet J Suppl 2010;38:436–443.
16. Clayton HM, Kaise LJ, Lavagnino M, et al. Dynamic mobilizations in cervical flexion: effects on intervertebral angulations. Equine Vet J Suppl 2010;38:688–694.
17. Lischer CJ, Withers JM, Parkin T. Accuracy of radiographic measurements of the cervical articular process joints of the horse. Pferdeheilkunde 2010;26:553–558.
18. Hugles KJ, Laidlaw EH, Reed SM, et al. Repeatability and intra- and inter-observer agreement of cervical vertebral sagittal diameter ratios in horses with neurological disease. J Vet Intern Med 2014;28:1860–1870.
