The SOMs are three groups of short muscles that cover all surfaces of the atlas and axis. In dogs, the SOMs consists of rectus capitis dorsalis, oblique capitis caudalis, and oblique capitis cranialis [15]. Contraction of the rectus capitis dorsalis and the obliquus capitis cranialis muscles results in an extension of the head. The obliquus capitis caudalis muscle, which attaches laterally to the atlas wings and extends cranially to the spinous process of the axis, is one of the main muscles allowing rotation of the AAJ [16]. As they contribute to the movement of the AAJ, they could affect the stability in this region, and be affected by diseases affecting it.
In this study, the SOMs of AAI dogs appeared smaller and showed increased fat infiltration as compared to those of normal dogs, similar to the results from human studies on the relationship between pain and muscles [10]. This is also consistent with the findings on the epaxial muscles of the TL region in dogs (13). The smaller SOMs in AAI dogs could be a sign of denervation atrophy; the painful and compressive characteristics and increased fat infiltration into the muscles is likely due to disuse and denervation [13, 17]. However, there is a disagreement between human studies on the influence of pain on muscles; for example, Yuan et al. (2017) reported a relationship between hypertrophy of this muscle and chronic headaches, while Hvedstrup et al. (2020) reported no association between pain and muscle properties [12, 18]. Skeletal muscles and bone share common embryological origins from mesodermal cell populations [19], and it has been reported that congenital abnormalities of bones in this region, such as incomplete ossification of the atlas, are associated with AAI [20]; therefore, it is possible that AAI dogs have congenital abnormalities of the SOMs as well as the bones.
Previous studies in dogs have reported that the severity of clinical symptoms had a significant effect on fat infiltration [13]. However, in this study, among AAI dogs, fat infiltration was more severe in group A dogs with mild symptoms than in group B dogs with severe symptoms. Studies on humans have reported that fat infiltration into skeletal muscles increases with age [19, 21]. In this study, group A was significantly older (7.58 years ± 3.08) than group B (2.56 years ± 1.49, P = 0.001). Therefore, it is possible that the fat infiltration identified in group A is age related, rather than associated with AAI. Although this study only evaluated a small number of dogs, there was no correlation between the fat ratio of muscles and age. In the long term, ongoing effects of pain and inflammatory mechanisms exert additional effects on muscle structure, such as atrophy, fatty infiltration, fibrosis, and function [22]. It is possible that these chronically altered SOMs reduce movement in AAJ, so that group A dogs showed mild symptoms for a long time, despite the AAI. To confirm this possibility, future studies are required to compare the histology of SOMs between normal and AAI dogs. However, intramuscular fat is reliably quantified using the signal intensity obtained by MRI [13, 14, 23]. The T1 relaxation time for fat tissue is short, and the signal intensity is high compared with skeletal muscle on the T1-weighted images; therefore, fat appears hyperintense relative to muscle, which is hypointense [23]. T1-weighted images have been used successfully to evaluate fat infiltration into various muscles in humans and dogs [13, 24–26]. Moreover, it has reported that increased fat infiltration and decreased muscle volume on T1-weighted images correlated well with subsequent histological analysis in isolated skeletal muscles of dogs [27]. Therefore, although histological confirmation was not possible, the method of quantifying fat infiltration used in this study may be useful as an index to non-invasively evaluate changes of the SOMs in AAI dogs.
There was no difference in the CSA of the muscle between the AAI groups in this study. It has been reported that nerve lesions affecting the multifidus CSA atrophy progress within three days [28]; most AAI dogs in this study showed neurological symptoms for more than three days; therefore, the possibility that there is no significant difference in the CSA of muscles in the AAI groups was considered.
This study identified a difference in the ratio of the spinal cord between normal and AAI dogs. The spinal cord compression in AAI dogs may have affected atrophy and fat infiltration in the SOMs, similar to the results on the epaxial muscles in dogs [13]. However, the correlations between spinal cord compression and muscle atrophy, and between spinal cord compression and fat infiltration in AAI dogs were not statistically significant. A study on humans suggested that the degree of instability in AAI can be underestimated by MRI when compared with radiology [29]. All MRI in this study were taken in a neutral, not a flexed position; therefore, it is highly likely that spinal cord compression due to the degree of instability was not reflected in the measurement.
The main limitation of this study was that the number of dogs in each group was small, and therefore, we used a non-parametric test. Future studies are required with larger groups to investigate the SOMs in dogs. Due to ethical reasons, we were unable to conduct a subsequent histological verification for the muscle changes identified by MRI. In this study, intra-class reproducibility was verified; however, there was no inter-class reproducibility verification because an assessor requires training on MRI images and SOM anatomy for a significant time period. In this regard, further research related to inter-class verification according to the training level is needed for measurement of the muscles. Since the increase in the CSA of the SOMs according to body weight was identified even within toy-breed dogs with ≤ 5 kg, to compensate for possible discrepancies in body weight, the size of the SOMs was evaluated using the CSA ratio of muscles, which was not correlated with body weight because the disc measurement method used in other studies cannot be used in this area [13]. Despite the above limitations, this study provides fundamental quantitative information about MRI of the SOMs in normal and AAI dogs.