Abstract
Background/Aim: Atlantoaxial subluxation (AAS) is generally a congenital condition that mainly affects toy breed dogs. Previous studies in several toy breed dogs revealed that dogs with AAS had a relatively high proportion of incomplete ossification (IO) of the atlas and dens anomalies compared to dogs without AAS. These anatomical characteristics may be important in surgical decision-making. Thus, the present study evaluated morphological differences in the atlas and axis between Maltese dogs with and without AAS. Materials and Methods: The medical records of Maltese dogs with and without AAS from 2015 to 2020 were analyzed. Abnormalities of the atlas and axis were evaluated using computed tomography (CT). Results: A total of 45 dogs were reviewed. Maltese dogs with AAS revealed a higher ratio of IO of the atlas (56%) than non-affected dogs (19%). Dens anomalies were identified in 78% of the dogs with AAS and in 26% of non-affected dogs. The shape of the atlas has been identified as thin, solid compact bone in Maltese dogs. Dogs that revealed IO of the dorsal arch of the atlas showed significantly lower CT values (in Hounsfield units) than dogs without IO. The CT values of the midline of the dorsal arch were significantly lower than those of the outer surrounding region. Dens hypoplasia was defined by measuring the dens-to-axis length ratio according to a previous study. A significantly lower ratio was identified in dogs with AAS than in non-affected dogs. Conclusion: The incidence ratio of abnormalities of the atlas and axis in Maltese dogs with AAS was similar to that of previous studies. The dorsal arch of the atlas is composed of thin cortical bone with a vulnerable midline region. As dogs with AAS are more likely to be afflicted with abnormalities in the atlas and axis, considering these morphological features is important when planning the surgical stabilization of AAS.
Atlantoaxial subluxation (AAS) is an uncommon condition of the atlantoaxial joint that occurs both in humans (1) and animals (2). It is typically seen as a congenital condition in veterinary medicine (3). Although larger breeds of dogs (4, 5) and cats (6, 7) can be affected, toy breeds such as Yorkshire Terrier, Chihuahua, Toy Poodle, and Pomeranian are commonly known as AAS-predisposed breeds (3, 8, 9). Dogs with congenital AAS are likely to have other vertebral anomalies such as atlantooccipital overlap and block vertebrae (10, 11). Factors associated with AAS include the absence of ligamentous structures, incomplete ossification (IO) of the atlas, and dens anomalies (2, 12-14). IO during development may lead to midline fusion failure and structural defects or the absence of the atlas (15). These anomalies can cause weakness of the dorsal atlantoaxial ligament associated with the alignment of the atlantoaxial joint (15, 16). As another factor, the dens of the axis develops from two separate ossification centers (17). During the developmental period, the instability of the atlantoaxial joint can result from dens anomalies (aplasia, hypoplasia, or fracture) or from disorders of the transverse ligament (2). A previous study reported that 35 of 46 dogs with AAS had dens anomalies (absence or abnormal conformation) (18). Understanding the anatomical features of the atlas and axis is important for surgical decision-making (19). It may be further emphasized when planning the posterior or dorsal stabilization of AAS in human and veterinary medicine, respectively (20, 21). In addition to clinical signs in affected dogs, AAS is routinely diagnosed and evaluated by radiography and magnetic resonance imaging (MRI) (11, 14, 22). To analyze AAS more precisely, various radiographic methods have been devised (22, 23). Among them, one study analyzed anomalies of the atlas and axis in several AAS-predisposed toy breed dogs (Chihuahua, Yorkshire Terrier, Toy Poodle, and Miniature Dachshund) using computed tomography (CT). The study revealed that dogs with AAS had a high proportion of IO of the atlas (75 out of 106) and dens anomalies (70 out of 106) (24). Although Maltese is also an AAS-predisposed toy breed (25), there are few reports of AAS in this breed. In addition, studies on the anatomical evaluation of the dorsal arch of the atlas in toy breed dogs are also rare. Thus, as the shape and composition of the atlas vary between breeds (14), the purpose of this study was to identify the anatomical features of the atlas and axis between Maltese dogs with and without AAS.
Materials and Methods
Inclusion criteria and data collection. The medical records of Maltese dogs with and without AAS in the Veterinary Teaching Hospital from 2015 to 2020 were reviewed. AAS was diagnosed based on CT and MRI. The medical records included age, sex, and body weight. CT images of each dog were obtained using a four-row multi-detector CT scanner (HiSpeed QX/i; GE Medical Systems, Milwaukee, WI, USA) and analyzed using a digital image measurement program (eFilm Workstation 4.1; Merge Healthcare, Hartland, WI, USA).
Incomplete ossification of the atlas. CT values in Hounsfield units (HU) were measured on the midline region of the dorsal arch of the atlas in transverse images. Three different parts of the dorsal arch of the atlas (cranial, central, and caudal) were measured and evaluated (Figure 1A). According to a previous study (24), if one or more of the three parts of the dorsal arch of the atlas had a CT value below 600 HU, it was considered IO. In addition, considering the shape of the dorsal arch of the atlas, CT values measured 1, 2, and 4 mm laterally from the midline region of the cranial part of the atlas were also evaluated (Figure 1B).
To identify incomplete ossification of the dorsal arch of the atlas, computed tomography (CT) values (in Hounsfield units) were measured in the cranial, central, and caudal parts of the atlas (red dots). Measurements were performed on the midline of the dorsal arch of the atlas using transverse CT images of the corresponding positions (A). To evaluate bone density according to the shape of the dorsal arch, CT values on the midline of the dorsal arch (red dashed line) and CT values measured 1, 2, and 4 mm laterally from the midline (red dots) were compared (B). Dens hypoplasia was defined according to the dens-to-axis length ratio. The length of the dens was obtained by measuring the length between the ventral base and the tip of the dens (red double-arrow). To define the length of the axis, a line that passed the tip of the dens and the dorsocaudal extremity of the axis was made (white dashed line). Another line was made by drawing a line perpendicular to the previous line and passing it through the ventral base of the dens (red dashed line). The length of the axis was defined as the length between the intersection of the two dashed lines and the dorsocaudal extremity of the axis (white double-arrow) (C).
Dens anomalies. Dens anomalies included dens aplasia, hypoplasia, and fracture. Hypoplasia of the dens was decided by the dens-to-axis length ratio (DALR) according to a previous study (23). In the present study, dogs with a DALR below the lower 95% confidence interval (CI) of normal Maltese dogs (without AAS and IO) were considered to have dens hypoplasia (Figure 1C).
Data analysis. Data between two groups or more than two groups were analyzed by the Mann-Whitney U-test or the Kruskal-Wallis test with Dunn’s multiple comparisons, respectively. Analyses were performed using GraphPad Prism version 6.01 for Windows and a p-value of <0.05 was considered significant. All data are expressed as the mean±SD.
Results
Demographics. A total of 45 Maltese dogs (9 dogs with AAS and 36 dogs without AAS) were reviewed in the present study. IO of the atlas was observed in 56% (5 out of 9) of the dogs with AAS compared to 19% (7 out of 36) of the dogs without AAS. No significant differences between dogs with IO (n=12) and dogs without IO (n=33) were identified in age (mean; 61 and 90 months, respectively) or body weight (mean; 2.8 and 3.3 kg, respectively) (Table I). Dens anomalies were revealed in 78% (7 out of 9) of the dogs with AAS. Six dogs showed dens hypoplasia and a dens fracture was identified in one dog. Dens hypoplasia was identified in 26% (9 out of 34) of the dogs without AAS.
Demographics of dogs with and without incomplete ossification (IO).
Comparison of CT values of the atlas. Compared to non-affected dogs (Figure 2A), the dogs with IO showed a relatively radiolucent dorsal arch of the atlas (Figure 2B). The mean CT values obtained from all three regions (cranial, central, and caudal parts of the dorsal arch of atlas) were significantly lower in dogs with IO compared to non-affected dogs (Table II). According to the shape of the cranial part of the dorsal arch, the mean CT values increased laterally. The CT values measured 4 mm laterally from the midline were significantly higher than the CT values measured from the midline and 1 mm laterally from the midline (Table III).
Transverse computed tomography (CT) images of a non-affected dog (59-month-old intact male weighing 1.9 kg) with a CT value of 1,003 Hounsfield units (HU) measured from its dorsal arch (arrow) (A) and a dog (167-month-old intact female weighing 3.6 kg) showing incomplete ossification of the dorsal arch (arrow) with a CT value of 464 HU (B).
Comparison of the degree of ossification of the atlas between incomplete ossification (IO) and non-IO groups using computed tomography values (in Hounsfield units).
Computed tomography values (in Hounsfield units) according to the distance from the midline of the cranial dorsal arch of the atlas.
Dens hypoplasia. The lower 95%CI (<0.34) of Maltese dogs was obtained from normal Maltese dogs (n=27) in the present study. The mean DALR of dogs with AAS (mean±SD, 0.31±0.06) was significantly less than that of dogs without AAS (mean±SD, 0.36±0.05). When the dogs with AAS were excluded, no significant difference in DALRs was observed between dogs with IO (mean±SD, 0.37±0.04) and non-affected dogs (mean±SD, 0.36±0.05).
Discussion
Among the various toy breed dogs prone to developing congenital AAS, previous studies reported that Chihuahuas, Yorkshire Terriers, and Toy Poodles accounted for a relatively high proportion (14, 25). In addition to the toy breeds mentioned above, as another predisposed toy breed to develop congenial AAS (25), the present study evaluated the atlas and axis of Maltese dogs with and without AAS. As factors that contribute to AAS development, IO of the atlas and dens anomalies were previously reviewed (14, 23). IO of the atlas may be highly associated with AAS (14, 16). In the present study, the IO of the atlas was also prevalent in Maltese dogs with AAS compared to dogs without AAS. IO occurred more often in the cranial (10 out of 12) and central (9 out of 12) parts than in the caudal (5 out of 12) part of the dorsal arch of the atlas. The CT values from all three regions of the dogs with IO in this study were significantly lower than those of dogs without IO, consistent with previous literature (24). When considering Mitch’s classification of bone density, the dorsal arch of the dogs with IO was classified as D3 (composed of thin, porous cortical bone), whereas the dorsal arch of dogs without IO was classified as D2 (composed of dense, porous cortical bone) (26). As a limitation in measuring the CT values of each part, it was hard to set a uniform region of interest (ROI) as Maltese dogs have a thin dorsal arch. Therefore, the maximum ROI value rather than the average was used to avoid interference from surrounding tissues. No significant difference between the two methods was identified (data not shown). A previous study described that an osseous defect mainly affected the dorsal arch on the midline (14). A thin dorsal arch composed of solid compact bone was identified in the Maltese dogs in the present study. The dorsal arch also had significantly lower bone density on the midline compared to adjacent areas, implying that it was the weakest area of the dorsal arch. Taken together, the dorsal arch of the atlas is more vulnerable to fracture (especially on the midline of the dorsal arch) in Maltese dogs with IO compared to non-affected individuals. When evaluating dens hypoplasia, a previous study compared the DALR in four AAS-predisposed breeds and reported that the breeds with AAS showed significantly lower DALRs than the same breeds without AAS. The study also suggested a reference range of dens hypoplasia as a DALR below the lower 95%CI of the DALR in normal dogs (23). As no reference range for Maltese dogs was suggested, the lower 95%CI of the DALR in normal Maltese dogs was calculated in the present study. The reference range for dens hypoplasia obtained from normal Maltese dogs (<0.34) was comparable to other AAS-predisposed breeds (Chihuahua, <0.36; Yorkshire Terrier, <0.34) previously reported (24). In conclusion, similar results were obtained in the present study as in other toy breed dogs (14, 24). Maltese dogs showed anatomically thin dorsal arches without trabecular bone and had the weakest midline region. AAS-affected Maltese dogs are more prone to other atlas and axis anomalies. Dogs with IO were shown to have a significantly lower bone density of the dorsal arch than non-affected dogs. Like other toy breeds in previous studies, AAS-affected Maltese dogs revealed relatively high rates of dens anomalies with significantly lower DALRs than Maltese dogs without AAS. Taken together, understanding these characteristics of cervical vertebrae and co-existing anomalies may be critical when planning the dorsal stabilization of AAS in toy breed dogs.
Footnotes
Authors’ Contributions
DK, DC and GK designed the study. DK analyzed the data. The manuscript was written by DK and GK. All Authors critically revised the manuscript and approved the final version.
Conflicts of Interest
The Authors declare no conflicts of interest in relation to this study.
- Received August 21, 2022.
- Revision received September 15, 2022.
- Accepted September 16, 2022.
- Copyright © 2022, International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved
This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY-NC-ND) 4.0 international license (https://creativecommons.org/licenses/by-nc-nd/4.0).
