Computed tomographic angiographic study of common carotid artery anatomic relationships in the dog

Abstract

Objective

To further understand spatial relationships of common carotid arteries to adjacent structures through evaluation of computed tomographic angiograms in dogs.

Animals

24 pet dogs.

Procedure

A database was searched for triplanar computed tomographic angiograms that included the heart base caudally and the 5th cervical vertebra cranially, without macroscopic abnormalities. Measurements included brachiocephalic trunk length, common carotid arteries’ position relative to the trachea, transverse (axial) thoracic height and width, manubrium length, and length of the 7th cervical vertebra (C7).

Results

Measurements (mean + SD) included brachiocephalic trunk length = 3.65 ± 1.34 cm (n = 24), mean thoracic inlet height = 6.51 ± 2.03 cm (n = 23), mean thoracic inlet width = 4.69 ± 1.35 cm (n = 20), mean manubrium length = 3.52 ± 1.15 cm (n = 22), and mean length of C7 = 1.93 ± 0.46 cm (n = 23).

Conclusion

Some or all measurements were feasible in all dogs. Understanding interindividual variation in spatial relationships is pertinent to differentiating normal from abnormal, for surgical planning, and possibly for elucidating the pathogenesis of certain disorders.

Clinical relevance

It is possible to obtain these measurements in dogs. This technique could be applied to subgroups of dogs (e.g., breeds) and dogs with cervical or thoracic abnormalities.

INTRODUCTION

The positions of the common carotid arteries through the neck of the dog and their relationships to adjacent structures are usually described as being uniform in the species (1). However, some descriptions exist of breed- or somatotype-associated variants. For example, the brachiocephalic trunk (BCT) of dogs is described as being “approximately 4 cm long and 8 mm in diameter” (1). The branching pattern of the vessels off the aortic arch in dogs, from right to left, is typically described as BCT [branching into the right common carotid (RCC), left common carotid (LCC), and right subclavian arteries] and left subclavian artery (1). However, additional work identified interindividual variation in the origin and branching patterns of these vessels (2–5). Common carotid arteries originate from the same point on the BCT in 68% of dogs, from different points on the BCT in 29% of dogs, and from a common, bicarotid trunk that arises from the BCT in 3% of dogs (4). In German shepherd dogs, a similar variety of morphologies exists, with some investigators dividing the bicarotid trunk group into 2 subgroups (2). A bicarotid trunk is commonly identified in dogs with a concurrent persistent right aortic arch (e.g., 17/21, 81%) but only occasionally identified in dogs without a persistent right aortic arch (e.g., 1/192, 0.5%) (5). Overall, such findings have been obtained either with macroscopic dissection of dog cadavers (1–3,6) or using computed tomographic angiography on anesthetized dogs (4,5,7,8).

Few studies of dogs have addressed spatial relationships of the common carotid arteries to adjacent thoracic and cervical structures. In 1 case study, investigators utilized computed tomographic angiography pre- and postoperatively to assess morphology of vessels where they crossed through the thoracic inlet (8). The location at which common carotid arteries are directly lateral to the trachea, associations with thoracic inlet height and width, and length of the BCT compared to skeletal structures as points of reference, have apparently not been described. Carotid arteries have been identified coursing ventrally along the malformed trachea in dogs with tracheal malformations and sometimes within the ventral tracheal groove that is characteristic of tracheal malformation. Vessel/nerve tension or thoracic inlet narrowing along soft, developing tracheal cartilage has been hypothesized as a possible cause for tracheal malformation generation; however, this is currently only a hypothesis (Chick Weisse, Schwarzman Animal Medical Center, New York, New York, USA; personal communication, 2024). Greater understanding of these spatial relationships could improve diagnosis of pathologic malformations of the vasculature, interventional treatment selection (9), and surgical planning (8) — and could offer insights into the pathogenesis of such disorders as tracheal malformation (9,10).

The objective of this study was to evaluate computed tomographic angiograms (CTAs) to quantify spatial relationships between common carotid arteries and adjacent cervical and thoracic structures. We hypothesized that it is possible to i) make these measurements on the CTAs of normal dogs, and ii) index these measurements to reference points in the dog.

MATERIALS AND METHODS

Animals

A convenience sample was obtained by searching the Atlantic Veterinary College (Charlottetown, Prince Edward Island) Radiology archive for dogs that underwent simultaneous cervical and thoracic computed tomographic angiography (Aquilion 64; Canon, Otawara, Japan). The CTAs that were used had been obtained between July 2, 2020 and June 17, 2022.

Computed tomographic angiography protocol

Each dog was induced and maintained under general anesthesia while endotracheally intubated. The dog was positioned in sternal recumbency with the head facing the CT aperture. In addition, each dog had physical restraint by means of foam wedges and troughs, sandbags, and hook-and-loop straps attached to the table. Images were acquired during breath-hold sequences without contrast, and then after IV administration of iohexol (Omnipaque, 240 mg iodine/mL; GE Healthcare Canada, Mississauga, Ontario), 2.5 mL/kg, administered through a peripheral IV catheter.

Analysis of images

Exclusionary criteria were absence of images that included the heart base at the caudal extent of the CTA, absence of images that included the complete 5th cervical vertebra (C5) at the cranial extent of the CTA, and presence of cervical or intrathoracic abnormalities on CTA. Before making measurements, the investigators used CTA images from 10 dogs from a previous study (11) to establish variables to measure in the current study and to refine measurement parameters. All CTA images were assessed using open-source software (https://horosproject.org). Variables and parameters for their measurements are in Table 1. The lengths of the 7th cervical vertebra (C7) and of the manubrium were measured as reference points to calculate ratios with vascular structures; this was done to explore indexing measurements to account for differences in body size.

TABLE 1.

Variables and parameters for measuring them on computed tomographic angiograms of normal dogs.

Variable	Parameter: Definition	Parameter: Method
BCT length (Figure 1)	Distance between origin of the BCT off the aortic arch caudally and origin of the RCC, LCC, or bicarotid trunk cranially	Scroll through transverse (axial) images from caudal to cranial and identify the one where BCT is first seen as a distinct structure. Mark this origin of the BCT on the corresponding MPR-synchronized dorsal (coronal) plane image. Scroll through transverse (axial) images while remaining synchronized to the dorsal (coronal) image in (i), above, using MPR. Identify the transverse (axial) image where the RCC is first seen as a distinct structure; i.e., separate from other vascular structures. Mark this origin of the RCC on the corresponding MPR-synchronized dorsal (coronal) plane image. Measure the distance between the 2 marked points.
Thoracic inlet height (Figure 2)	Greatest distance between dorsal aspect of manubrium and corresponding ventral aspect of T1	Scroll through transverse (axial) and MPR-synchronized dorsal (coronal) images and identify the orthogonal pair of images where the right and left 1st ribs are best visualized. On the transverse image, measure the distance between the dorsal manubrium and ventral T1.
Thoracic inlet width (Figure 2)	Greatest distance between medial aspects of contralateral 1st ribs on a transverse axial image	As for thoracic inlet height, but measuring the greatest distance in a lateral plane (width) between the medial surfaces of the 2 1st ribs.
Intercarotid distance (Figure 3)	Distance between RCC and LCC at level of RCC origin	Scroll through transverse (axial) images from caudal to cranial and identify the image where the RCC and LCC are first seen as distinct structures. Measure the shortest distance between the RCC and LCC on this image. All measurements < 1 mm recorded as < 1 mm.
Carotid artery positions lateral to trachea (Figures 4, 5)	Cervical vertebra corresponding to transverse (axial) image where LCC is directly lateral to the left of the trachea and where RCC is directly lateral to the right of the trachea	Scroll through transverse (axial) images from caudal to cranial and identify 2 images: the one where the RCC is directly lateral to the right of the trachea (transverse image of trachea as clockface analogy: 9 o’clock position) and the one where the LCC is directly lateral to the left of the trachea (tracheal clockface analogy: 3 o’clock position). Using MPR, scroll through paramedian (parasagittal) images, while remaining synchronized to each of the 2 transverse (axial) images, to find the midline median (sagittal) image. Use this image to identify the vertebrae corresponding to each of the 2 transverse (axial) images described above (3 o’clock and 9 o’clock). Measure the length of each of the 2 vertebrae on the floor of the vertebral canal and identify in tenths, from cranial (zero tenths) to caudal (10 tenths), the location of each of the 2 corresponding transverse (axial) images.
Carotid artery or BCT position ventral to trachea (Figure 6)	Cervical vertebra where BCT or RCC is directly ventral to the trachea	As for carotid artery positions lateral to trachea, but using the transverse (axial) image where the BCT or RCC is ventralmost to the trachea (6 o’clock position).
Length of manubrium (Figure 7)	Length of manubrium	On midline median (sagittal) image, measure greatest length of 1st sternebra.
C7 length (Figure 7)	Length of 7th cervical vertebra	On midline median (sagittal) image, measure greatest length of body of 7th cervical vertebra (floor of vertebral canal).

BCT — Brachiocephalic trunk; LCC — Left common carotid artery; MPR — Multiplanar reformation; RCC — Right common carotid artery; T1 — 1st thoracic vertebral body.

Measurements were obtained using 2D multiplanar reformation (MPR) and the software’s length measurement function. The bone setting was used for measuring bony structures and the abdomen setting for measuring contrast-filled vascular structures. Additional features of the software, including 3D image rendering and 3D MPR, were used as needed to increase the precision and accuracy of measurements through confirmation of the points of origin of vessels, enhancement of visualization of the course of vessels, and confirmation of the reference vertebra as C7.

Statistical methods

Data were analyzed using commercial software (GraphPad Prism 10.2.2; GraphPad Software, La Jolla, California, USA). Analysis for normality was done using the Shapiro-Wilk test. Normally distributed data are reported as mean ± SD. Non-normally distributed data are reported as median (range). Investigation of correlations between variables was established a priori and was done using Pearson correlation for normally distributed data or Spearman correlation for non-normally distributed data (both 2-tailed). For vertebral lengths, results are shown to tenths of a vertebra, for precision. For localization using vertebral lengths as reference points, cervical vertebral lengths were treated numerically without the designation of “cervical” (e.g., C5.3 as 5.3 during statistical analysis). Similarly, thoracic vertebrae were treated numerically without the designation of “thoracic,” and as extensions of cervical vertebrae by extrapolation (e.g., T1.6 as 8.6 during statistical analysis). P < 0.05 was considered statistically significant, except for correlation analyses to which a Bonferroni correction was applied.

RESULTS

Of the 35 CTAs retrieved, 11 were excluded due to space-occupying masses (4), megaesophagus (1), esophageal perforation/tear (1), lack of contrast (2), inability to visualize the BCT (2), and a technical issue (missing images) (1). In the 24 dogs studied, some individual measurements could not be made for the following reasons: absence of parallel right and left 1st ribs in a transverse (axial) view (thoracic width, n = 4), oblique image of manubrium (length of manubrium, n = 2), image series did not reach sufficiently cranially (carotid artery position lateral to trachea (3 o’clock, n = 2), carotid artery position lateral to trachea (9 o’clock, n = 2), ill-defined margins of C7 (length of C7, n = 1), and oblique image of thoracic inlet (thoracic height, n = 1).

There were 13 males (10 castrated) and 11 females (9 spayed). On transverse (axial) images, the vessel directly ventral to the trachea was the RCC in 17 dogs (71%) and the BCT in 7 dogs (29%) (Tables 1, 2). No dog had a bicarotid trunk. Measurements and descriptive statistics are presented in Table 2 and correlation analyses in Table 3. After Bonferroni correction for 18 analyses, only the association between BCT length and weight remained significant.

TABLE 2.

Measurements of variables on computed tomographic angiograms of normal dogs.

Variable	Mean ± SD or median (range)	Number of dogs
Body weight (kg)	21.5 ± 12.3	24
Length of BCT (cm)	3.65 ± 1.34	24
BCT length/C7	1.92 ± 0.35	23
BCT length/manubrium length	1.06 ± 0.26	22
Thoracic inlet height (cm)	6.51 ± 2.03	23
Thoracic inlet width (cm)	4.69 ± 1.35	20
Thoracic inlet height:width	1.38 ± 0.21	20
Intercarotid distance (mm)	0.57 (< 1 to 8.1)	24
LCC position lateral to trachea (3 o’clock) (vertebra)	C5.5 ± 1	22
RCC position lateral to trachea (9 o’clock) (vertebra)	C5.8 ± 0.78	22
Artery position (BCT or RCC) ventral to trachea (6 o’clock) (vertebra)	T1.1 ± 0.75	24
Length of C7 (cm)	1.93 ± 0.46	23
Length of manubrium (cm)	3.52 ± 1.15	22

BCT — Brachiocephalic trunk; C7 — 7th cervical vertebra; LCC — Left common carotid artery; RCC — Right common carotid artery.

Results reported as mean ± SD if normally distributed and median (range) if non-normally distributed.

TABLE 3.

Correlation analyses between various variables for computed tomographic angiograms of normal dogs.

IV	DV	r	95% CI	P
BCT length	Weight	0.82	0.62 to 0.92	< 0.0001
BCT length/C7 length	Weight	0.37	−0.05 to 0.68	0.08
BCT length/manubrium length	Weight	0.15	−0.29 to 0.54	0.5
LCC @ 3 o’clock	BCT length	−0.16	−0.54 to 0.28	0.48
RCC or BCT @ 6 o’clock	BCT length	0.16	−0.26 to 0.53	0.46
RCC @ 9 o’clock	BCT length	−0.2	−0.57 to 0.25	0.38
LCC @ 3 o’clock	Weight	−0.25	−0.61 to 0.19	0.26
RCC or BCT @ 6 o’clock	Weight	0.007	−0.4 to 0.41	0.97
RCC @ 9 o’clock	Weight	−0.14	−0.53 to 0.3	0.54
Thoracic inlet height:width	Weight	0.02	−0.42 to 0.46	0.93
Thoracic inlet height:width	BCT length	0.23	−0.23 to 0.61	0.33
Intercarotid distance	Weight	0.13	−0.3 to 0.51	0.56
Intercarotid distance	BCT length	0.34	−0.09 to 0.66	0.11
Intercarotid distance	BCT length/C7 length	0.42	−0.005 to 0.71	0.047
Intercarotid distance	BCT length/manubrium length	0.52	0.11 to 0.78	0.013
Intercarotid distance	LCC @ 3 o’clock	0.1	−0.35 to 0.51	0.67
Intercarotid distance	RCC or BCT @ 6 o’clock	0.04	−0.38 to 0.45	0.86
Intercarotid distance	RCC @ 9 o’clock	−0.12	−0.52 to 0.33	0.61

BCT — Brachiocephalic trunk; CI — Confidence interval; C7 — 7th cervical vertebra; DV — Dependent variable; IV — Independent variable; LCC — Left common carotid artery; RCC — Right common carotid artery; 3 o’clock — LCC position directly left-lateral to trachea on transverse (axial) image; 6 o’clock — RCC or BCT position directly ventral to trachea on transverse (axial) image; 9 o’clock — RCC position directly right-lateral to trachea on transverse (axial) image.

Boldface font indicates statistical significance.

DISCUSSION

This study presented a novel method for assessing spatial relationships of the BCT and common carotid arteries not using cadavers. Measurement of all variables was feasible in most dogs. Inability to measure certain variables in some dogs, such as LCC position lateral to trachea (9 o’clock) or thoracic inlet width, were related to technical issues such as cranial extent of acquisition of images or patient positioning, respectively. Such issues can be addressed proactively at the time computed tomographic angiography is done if the present variables are intended to be acquired and measured, or through 3D reconstruction of CTAs.

Although the BCT is broadly described as being ~4 cm long in dogs (1), there was a strong correlation between BCT length and body weight. Therefore, a single value for BCT length is not expected to be accurate for dogs of all body weights.

Quantitative information on the origin and course of the common carotid arteries was obtained from CTAs of dogs using the methods of this study. The RCC and LCC origins on the BCT were separated by a distance that varied widely among individuals and was not correlated to body weight. When examining the course of the common carotid arteries from caudal to cranial, both the RCC and LCC first were directly lateral to the trachea at approximately the level of C5 in most dogs. The artery (BCT, RCC, or LCC) that first appeared directly ventral to the trachea when proceeding from caudal to cranial differed among dogs. Indexing BCT length to length of the manubrium or of C7 reduced the influence of body weight on BCT length. Such indexing could be evaluated further as part of developing reference intervals for BCT length that apply to dogs of various body weights. The RCC origin on the BCT was noted to occur at the level of the 1st rib (1), and similarly, the present study identified the level of the 1st thoracic vertebra as the mean position of the origin of the RCC. Together, these findings increased the feasibility of quantitatively assessing BCT and common carotid artery structure and course in CTAs of healthy dogs. The feasibility of this approach means that formal validation can be undertaken.

Some limitations could be attributed to the retrospective nature of this study. Many CTAs could not be included because they imaged the neck or the thorax, but not both. This selection bias reduced the number of cases simply because protocols were applied for computed tomographic angiography of the anatomic region of interest.

Positioning during imaging differed among dogs. In 4 dogs, this meant that thoracic inlet width could not be measured, and thoracic inlet height:width could not be calculated, because the right and left 1st ribs could not be visualized on the same transverse (axial) images. Three-dimensional reconstruction could be applied to circumvent this, and similar limitations, in the future.

The mean body weight of dogs in this study was higher than the typical body weight of dogs with tracheal malformation or tracheal collapse. This difference would limit the extent to which the present results can be applied to dogs with such tracheal diseases.

In conclusion, it was feasible to obtain measurements involving the common carotid arteries and BCT of dogs on CTAs. These findings could have clinical implications for surgical planning and for understanding embryogenesis of disorders of the vasculature in the necks of dogs.

FIGURE 1.

Brachiocephalic trunk length in a dog.

FIGURE 2.

Thoracic inlet height (A) and width (B) in a dog.

FIGURE 3.

Intercarotid distance in a dog.

FIGURE 4.

Transverse (left image), dorsal (center image), and median (right image) plane views demonstrating the left common carotid artery at the left lateralmost (3 o’clock) position relative to the trachea (short arrow) in the dog. Using multiplanar reformation, the transverse image is identified as corresponding to vertebral level C5.5 (long arrow).

FIGURE 5.

Transverse (left image), dorsal (center image), and median (right image) plane views demonstrating the right common carotid artery at the right lateralmost (9 o’clock) position relative to the trachea (short arrow) in the dog. Using multiplanar reformation, the transverse image is identified as corresponding to vertebral level C6.1 (long arrow).

FIGURE 6.

Transverse (left image), dorsal (center image), and median (right image) plane views demonstrating the brachiocephalic trunk as the first major artery to be apparent directly ventral to the trachea (6 o’clock position) when scrolling from the heart base cranially (short arrow) in the dog. Using multiplanar reformation, the transverse image is identified as corresponding to vertebral level C7.9 (long arrow).

FIGURE 7.

Measured lengths of the 7th cervical vertebra (A) and manubrium (B) in the dog.

Funding Statement

Generous funding was provided by the Ptarmigan Foundation and by the Atlantic Veterinary College — University of Prince Edward Island Student Research and Leadership Program.