Int. J. Morphol., 21(2):123-129, 2003.

ANATOMICAL RELATIONSHIP BETWEEN VERTEBRAL ARTERIES AND
CERVICAL VERTEBRAE: A COMPUTERIZED TOMOGRAPHY STUDY

RELACIONES ANATÓMICAS ENTRE LAS ARTERIAS VERTEBRALES Y LAS VÉRTEBRAS CERVICALES: UN ESTUDIO UTILIZANDO TOMOGRAFÍA COMPUTARIZADA

^*Johnny Wesley Gonçalves Martins; ^**Riccardo Pratesi & ^*Armando José China Bezerra 

---

SUMMARY: Vertebral arteries ascend lateral to the cervical spine. Although non-contrasting tomographic studies have attempted to determine the anatomical relationship between these arteries and the spine, the presence and position of the former have yet to be clearly demonstrated. This study was thus aimed to determine the position of the vertebral arteries in relation to the midline, anterior and posterior planes of the vertebral body of each cervical vertebrae. Twenty six adult patients, without lesions that could distort normal anatomy, were employed as subjects and 52 vertebral arteries were analyzed. Two-dimensional computerized tomographies scans were conducted following the injection of an intra-venous contrast. Statistical analyses evaluated the differences between the distances of the vertebral arteries to the various cervical vertebrae. At the level of C1 and C7 the vertebral arteries were found to be farther from the midline plane (p<0.0001), compared to the other levels. The left vertebral artery was also found to be closer to the midline plane of C1 (p=0.0461) than the right artery. As these arteries ascended, greater distances from the anterior plane were verified. Finally, the vertebral arteries migrated dorsally in relation to the posterior plane, reaching significant distances at C1 (smallest distance) and C7 (greatest distance) (p<0.0001 and p<0.0001, respectivelly). The vertebral arteries, therefore, do not seem to follow a linear ascending pathway, distancing themselves from the midline plane at their upper and lower limits. Lateral, dorsal and ventral migrations were observed, in relation to the cervical vertebrae, rendering these arteries less vulnerable ventrally at C1 and at the level of the intervertebral foramina at C7.

KEY WORDS:1.Vertebral artery; 2. Sinal column; 3. Computerized tomography. 

---

INTRODUCTION

The cervical region of the spinal column is surrounded by different ligamentous, muscular, nervous and vascular structures (Heller & Pedlow, 1998 and Caovilla et al., 2000). The vertebral arteries, more specifically, constitute one of the vascular components of this region and ascend parallel to the spine (Adams & Victor, 1991; Apuzzo, 1993; Winter et al., 1995 and Osborn, 1996) through the transverse foramina of the upper six cervical vertebrae (Taveras, 1976). At the seventh vertebra, these arteries are commonly found ventrally to the transverse foramina (Goss, 1977).

Knowledge of the precise anatomical location of the vertebral arteries and the spine is essential, as various clinical conditions and surgical procedures involve these structures (Kamimura et al. and Karaikovic, 2000). Surgical approaches held laterally to the spine can lead to an iatrogenic rupture of these vessels with obvious irreversible consequences (Ebraheim et al., 1996 and Cheney & Herkowitz, 1998).

Data describing the exact position of the vertebral arteries in relation to adjacent structures are vague. Several of these studies have reported that the vertebral arteries have an almost vertical ascending pathway, lateral to the spine (Goss, 1977); ascend vertically in the neck (Apuzzo); have a long ascending pathway (Adams & Victor); or follow a relatively linear course within the transverse foramina (Taveras). However, the exact position of the vertebral arteries, relative to the spine, varies depending on the location, therefore maintaining a not entirely linear ascending pathway. This study was, therefore, aimed to describe the anatomical pathway of the vertebral arteries and their anatomical relationship with the cervical vertebrae. Computerized tomographic measurements were obtained to calculate the average distance between the vertebral arteries and the midline, anterior and posterior planes of the vertebral body of each cervical vertebra.

MATERIA AND METHOD

Subjects:Both male and female patients, 18 years old or older, were employed as subjects (Table I). Computerized tomography scans were performed as part of diagnostic tests requested by the patients' physicians to rule out cervical neoplasias. Patients with lesions or congenital abnormalities of the spinal column or neck were excluded from the study. In addition, normal X-ray radiographs of the cervical spine were performed prior to the scans to exclude possible cervical rotations or scoliosis.

Table I. Demographic profile.

	M=male						F=female				C=caucasian				B=black				O=others
	Pacient	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15	16	17	18	19	20	21	22	23	24	25	26
	Sex	F	M	M	F	F	M	M	F	F	M	M	M	M	F	F	F	M	F	M	M	F	F	F	M	F	M
	Race	C	C	C	B	C	C	O	C	C	C	C	C	C	C	O	B	C	C	C	C	C	C	B	C	C	C
	Age	50	47	33	51	73	21	68	26	30	67	42	27	18	64	44	48	60	52	47	43	48	71	73	50	34	59

Computerized tomography scans: Data were obtained from 26 transaxial computerized tomography (CT) scans of the neck, totaling 52 vertebral arteries. CT scans were performed immediately following intra-venous administrations of an iodine contrast medium to enhance visualization of the arteries. All scans were held at the Radiology Center of the Santa Lúcia Hospital, Brasilia/Brazil, during October 1999 and December 2000, on Toshiba X vision GX CT unit.

For each cervical vertebra, data were obtained from tomographic sections which passed in a half-way distance between the upper and lower surfaces of each vertebra's body, and that were parallel to the inferior surface of the corresponding vertebra. Measurements were obtained from this location as osteophites tend to be formed above or below the vertebra and the fact that the transverse foramina are located at this plane(24). For the atlas, due to the absence of a real body, measurements were obtained from the scan sections that simultaneously included the ventral and dorsal tubercles (Goss). Data were analyzed by two independent radiologists.

Data analyses:Measurements were calculated by the following method (Fig. 1). Initially, an imaginary line was made tangential to the dorsal surface of each vertebra's body (plane X). A second line parallel to the former and tangential to the ventral plane of the vertebra's body was also determined (plane Y). Perpendicular to the preceding two, a third line was established (plane Z), passing through the midline plane of the base of the spinous process of the vertebra (Kapandji, 1990. The smallest measure between the medial wall of the vertebral artery and plane Z determined the distance between the vertebral artery and the midline plane of each vertebra (distance A). On the other hand, the depth of the vertebral artery, compared to the ventral surface of the vertebra, was determined as the smallest measure between the ventral wall of the vessel and plane Y (distance B). Finally, the smallest distance between plane X and the dorsal wall of the artery indicated the relationship between the artery and the dorsal surface of the vertebra (distance C).

Fig. 1. Axial view of a cervical vertebrae with imaginary planes (anterior Y, posterior X and midline Z).

Data were analyzed by means of an analysis of variance (ANOVA), followed by Duncan's test for pairwise comparisons to compare the distances obtained at the different levels on each side of the spinal column. ANOVAs were also employed for comparisons of both sides of the spine. Criterion for statistical significance was set at p<0.05.

Fig. 2. CT Scan with intra-venous contrast medium; see vertebral artery inside transverse foramina.

RESULTS

This study consisted mainly of a description of the anatomical structures. Similar results were observed for both male and female patients. Age of subjects varied between 18 and 80 years old, with a mean value of 49.92. Further details of the demographic profile of the subjects are described in Table I.

Distance between the vertebral arteries and the midline plane of the vertebrae. For the right side, the distance calculated at the level of C1 and C7 were found to be significantly greater than those for the remaining cervical vertebrae (p<0.0001). At these locations the mean distances were 2.37 and 1.76 cm, respectively (Table II). Similarly, the distance between the left vertebral artery at C1 and C7 was significantly greater than those observed at the other vertebrae (p<0.0001; mean C1=2.27 cm; C7=1.79 cm).

Table II : Mean distances between the right and left vertebral arteries and the midline (A), anterior (B) and posterior (C) planes of each cervical vertebra. The indicated variance of the mean represents the 95% confidence interval.

Level X Dist	C1	C2	C3	C4	C5	C6	C7
Dist A Right	2.37	1.37	1.23	1.25	1.31	1.41	1.76
	(2.30-2.43)	(1.30-1.44)	(1.16-1.30)	(1.19-1.32)	(1.24-1.38)	(1.35-1.48)	(1.69-1.83)
Dist A Left	2.27	1.37	1.22	1.27	1.34	1.36	1.79
	(2.20-2.34)	(1.30-1.43)	(1.16-1.29)	(1.21-1.34)	(1.27-1.41)	(1.29-1.43)	(1.72-1.85)
Dist B Right	1.19	0.76	0.74	0.62	.061	0.67	0.43
	(1.10-1.27)	(0.67-0.85)	(0.65-0.82)	(0.54-0.71)	(0.52-0.70)	(0.59-0.76)	(0.34-0.51)
Dist B Left	1.20	0.64	0.69	0.64	0.59	0.70	0.46
	(1.10-1.30)	(0.54-0.74)	(0.60-0.79)	(0.54-0.74)	(0.49-0.68)	(0.60-0.79)	(0.36-0.56)
Dist C Right	-0.22	0.07	0.17	0.27	0.26	0.26	0.96
	(-0.29-0.14)	(-0.01-0.14)	(0.10-0.24)	(0.20-0.35)	(0.19-0.33)	(0.19-0.34)	(0.89-1.03)
Dist C Left	-0.22	0.08	0.23	0.29	0.27	0.25	0.93
	(-0,30-0,14)	(0,00-0,17)	(0,15-0,31)	(0,21-0,37)	(0,19-0,35)	(0,17-0,34)	(0,85-1,02)

Significant differences among the remaining cervical vertebrae (C2, C3, C4, C5, C6) were not verified for either the right or left side. Analysis between the two vertebral arteries indicated that only at C1 was the right artery significantly more distant from the midline plane of the vertebrae, when compared to the left artery (p=0.0461; Fig. 3)

Fig. 3. Mean distance observed for the left and the right vertebral arteries, compared to the midline plane of the respective vertebra.

Distance between the vertebral arteries and the anterior plane of the vertebrae. The mean distance between the right (1,19 cm) and the left (1,20 cm) vertebral arteries and the anterior plane of C1 were found to be significantly greater (p<0.0001) than that calculated for the other vertebrae (Table II). Conversely, the distance observed for the right artery at the level of C7 was significantly smaller (mean=0,43 cm; p<0.04444) than that observed for the remaining vertebrae. Comparisons among the right and left arteries failed to reveal any significant difference at any level (Fig. 4).

Fig. 4. Mean distance observed for the left and the right vertebral arteries, compared to the anterior plane of the respective vertebra.

Distance between the vertebral arteries and the posterior plane of the vertebral body. The distance of the right vertebral artery to the posterior plane of C7 (0,96 cm) was significantly greater than that calculated for the other levels (p<0.0001). The smallest distance (-0,22) was observed on the right side at the level of C1, differing significantly (p<0.0001) from the values of the remaining vertebrae. This negative value suggests that at C1 the right vertebral artery is located dorsally to the posterior plane of this vertebra. Similar results were observed for the left vertebral artery. When comparing the two sides, however, no significant differences were observed for the distances calculated at each vertebra (Fig. 5).

Fig. 5. Mean distance observed for the left and the right vertebral arteries, compared to the posterior plane of the respective vertebral body.

Dorsal-ventral diameter of the vertebral bodies: Analyses of the dorsal-ventral diameter of the vertebrae, including the odontoid process of the axis, revealed significantly similar values among C1, C6 and C7, as well as for C5 and C6. The dorsal-ventral diameter of C5, however, was also found to be significantly similar to that of C2, C3 and C4. Thus, the seven measurements obtained may not necessarily differ significantly between them.

Transverse diameter of the vertebrae. The transverse diameter of C4, C5, C6 and C7 were found to be significantly similar and larger than that observed for the remaining cervical vertebrae. The second largest diameter was that of C2 and C3, differing significantly from the remaining five cervical vertebrae. The smallest significant measure was shown to be at the level of C1 (p<0.0001).

DISCUSSION

Maintaining the vertebral arteries intact constitutes a frequent concern during cervical procedures (Riew, 1988 and Sonntag & Dickman, 1996) given that minor lesions may result in serious hemorrhages or even death. Anatomical studies have, in fact, been undertaken in an attempt to minimize intra-surgery lesions of these arteries (An et al., 1991 and Cooper et al., 1988). Data from the present CT scans indicated that at the level of C1 and C7 the vertebral arteries were located at a significant greater distance from the midline plane (Fig. 3), compared to that of the other cervical vertebrae. These results suggest that coronally the vertebral arteries do not follow a truly linear pathway. Thus, it would seem safer to perform lateral resections at the upper and/or lower limits of the cervical spinal column. The midline plane was defined in this case as the imaginary line passing through the midline of the base of the spinous process of the vertebra in order to separately measure the distance between this plane and the right/left vertebral artery. This parameter has been previously calculated by Vaccaro et al. (1994) as half the distance between the vertebral arteries, implying that this measure is symmetrical. However, this assumption does not seem to be the case, given that significant differences were presently observed between the right and the left vertebral artery at C1. The left vertebral artery was significantly closer to the midline plane of C1 than that observed for the right artery. To secure C1-C2, when placing lateral mass screws, lesion thus seem more probable on the left side. Accordingly, distinct placement angles should be used when performing such procedures.

Nonetheless, the depth at which the vertebral arteries lay is also an important factor during surgery. Similar to previous reports (Vaccaro et al.), the vertebral arteries become increasingly more distant from the anterior plane as they ascend within the cervical region. At the level of C7, the right vertebral artery was located near the anterior plane, migrating dorsally while ascending through the next upper five cervical vertebrae, compared to the values found at C7. The distance calculated for these five vertebrae, however, did not vary significantly, indicating that this migration seems to occurs abruptly between C7 and C6 (Fig. 4). On the left side, the increased distance between the artery and the anterior plane of the vertebrae was less pronounced, such that significant differences were not observed between the lower six cervical vertebrae. At thelevel of C1, on the other hand, both vertebral arteries were found to be at a greater distance from the anterior plane. At this location the vertebral arteries were, therefore, more distance from the midline plane, as well as located more deeply in relation to the anterior plane, which should facilitate surgical procedures.

The posterior plane of the vertebrae corresponds, in this study, to the ventral side of vertebral canal where the transverse foramina are located. Resections of the foramina, through radiculopaties, are common due to hernias or osteophytes. Therefore, the exact location of the vertebral arteries in relation to this plane is essential. At the level of C7, the vertebral arteries were found to be significantly more distant from posterior plane of this vertebra, compared to other vertebrae, and as a result less susceptible to lesions at this point. The distance between the vertebral arteries and the posterior plane of the upper six vertebrae decreased as the former ascend through the neck, although not significantly.

In general, the distance between the vertebral arteries has been previously calculated to be 3 cm(Gokaslan & Cooper, 1996), although the level from which this measure was obtained was not reported. Results from the present investigation suggest that the distance should not be standardized to this level, as values as small as 2.45 cm at the level of C3 and as large as 4.64 cm at C1 were verified. Surgical interventions of cervical hernias, for instance, are reported to cause a 2.90 cm transverse decompression (Riew), inducing lesions to the vertebral arteries at C3.

In previous investigations CT scans of the cervical spinal column have been performed without the use of a contrast, simply assuming that the vertebral arteries lay within the transverse foramina (Vaccaro et al.). Nevertheless, this assumption is not always valid, inasmuch as the anatomical location of the vertebral arteries has been found to vary, as previously reported for C7 (Goss). In fact, these arteries were located externally to the transverse foramina in 100% and 3,84% of the CT scans at C7 and C6, respectively. Therefore, data obtained without proper verification of the location of the vertebral arteries may be imprecise. This does not seem to be the case in the present investigation as measures were determined through the use of an intra-venous contrast medium, which confirmed the location of the vertebral arteries (Fig 2).

Furthermore, CT scans have been reported to be a more accurate method to analyze bone tissues(Esses, 1995 and Gokaslan & Cooper). In contrast, surgery of the neck is known to influence the result for the manipulations involved in this procedure may alter the original distances between structures. This is particularly valid when the vertebral arteries lay externally to the transverse foramina of the vertebrae. In addition, through surgical procedures, data related to C7 could only be obtained via removal of structures. CT scans thus provide a more precise in vivo assessment of the location of the vertebral arteries.

Children and adolescent patients were excluded from the study. Accurate measures, reflecting the general population, can not be obtained from these subjects since the spine is not yet fully developed. In addition, the observed predominance of caucasian subjects indicates the type of patient seeking treatment at the Radiology Center of the Santa Lúcia Hospital. Finally, as the nature of this study was mainly descriptive, verifying the distribution pattern within the general population, a control group was not deemed necessary (Pereira, 1995).

CONCLUSION

The vertebral arteries do not seem to follow a linear pathway as they ascending lateral to the cervical spinal column. At their upper and lower limits (C1 and C7), these arteries distance themselves from the respective vertebra's midline plane. In addition, when comparing the right and the left side, the distances calculated were found to be

asymmetrical (significantly at C1). Dorsal and ventral migrations of the vertebral arteries were also demonstrated, compared to the anterior and posterior planes of the cervical vertebrae. Taken together, these results suggest that the vertebral arteries seem to be less vulnerable ventrally at C1 and at the level of the intervertebral foramina at C7. 

---

Las arterias vertebrales ascienden lateralmente a la columna vertebral. Estudios para definir las relaciones entre estos vasos y la columna han sido realizados por medio de exámenes tomográficos no contrastados, no habiendo mostrado directamente la presencia ni la posición de las arterias vertebrales. El objetivo de este estudio es mostrar la localización de las arterias vertebrales en relación a los planos medio, anterior y posterior de los cuerpos vertebrales cervicales. Se trata de un estudio anatómico descriptivo realizado con tomografías computarizadas bidimensionales, después de la inyección de contraste venoso.

Se consideraron para este estudio, pacientes adultos, de ambos sexos de quienes fueron evaluadas 52 arterias vertebrales, sin lesiones estructurales que pudieran distorsionar la anatomía normal. Al final, fue efectuada una prueba estadística para evaluar diferencias en las distancias entre las arterias vertebrales y los distintos cuerpos vertebrales cervicales. Fue encontrado que en C1 y C7, las arterias vertebrales están más alejadas del plano mediano que en las demás regiones (p<0,0001). Al nivel de C1, la arteria vertebral izquierda se encuentra más cercana que la arteria vertebral derecha, con respecto al plano mediano (p=0,0461). A medida que ascienden en el cuello, las arterias vertebrales se distancian del plano anterior. Con relación al plano posterior, las arterias vertebrales realizan una migración posterior que muestra diferencia estadística en C7 y C1 (respectivamente p<0,0001 y p<0,0001).

A medida que ascienden en el cuello las arterias vertebrales, no tienen trayecto rectilíneo. En los extremos superior e inferior, éstas se apartan del plano mediano. Existen migraciones laterales, posteriores y anteriores con respecto a los planos de los cuerpos vertebrales. Estos desplazamientos dejan las arterias vertebrales menos vulnerables en los abordajes anteriores en C1 y en las descompresiones de forámenes intervertebrales a nivel de C7.

PALABRAS CLAVE: 1. Arteria vertebral; 2. Columna vertebral; 3. Tomografía computarizada. 

---

REFERENCES

Adams, R. D. & Victor, M. Principles of Neurology. 5 ed. Editora McGraw Hill, 1993. Chapter 34, pp 669-748.

An, H. S.; Gordin, R. & Renner K. Anatomic considerations for plate-screw fixation pf the cervical spine. Spine,16 (suppl 10):548-51, 1991.

Apuzzo. M. L. J. Brain Surgery-Complications avoidance and Management. Mayer P. L. & Kier E. L. The ontogenetic and phylogenetic basis of cerebrovascular anomalies and variants. Churchill Livingstone Inc, 1993. V.1. Cap 24, pp 691-792.

Caovilla, H. H.; Ganança, M. M.; Munhoz M. S. L.; Silva M. L. G. & Ganança F. F. Síndrome cervical. Silva M. L. G.; Munhoz M. S. L.; Ganança M. M. & Caovilla H. H. Quadros Clínicos Otoneurológicos Mais Comuns. Atheneu, São Paulo, 2000. V. 3. Cap 11, pp 95-100.

Cheney, R. A . & Herkowitz H. N. Anterior Cervical Corpectomy and Fusion for Radiculophaty. Clark C. R. The Cervical Spine-The Cervical Spine Research Society. 3 ed. Lippincott-Raven, Philadelphia, 1998. Chapter 59, pp 805-37.

Cooper, P. R.; Cohen, A.; Rosiello, A. et al. Posterior stabilization of cervical spine fractures and subluxations using plates and screws. Neurosurgery, 23:300-6, 1988.

Ebraheim, N.; Lu, J. & Brown J. Vulnerability of vertebral artery in anterolateral decompression for cervical spondylosis. Clin. Orthop., 322:146-51, 1996.

Esses, S. I. Degenerative disease of the spine. Esses S. I. Textbook of Spinal Disorders. J. B. Lippincott Company, Philadelphia, 1995. Chapter 8, pp 173-84.

Gokaslan, Z. L. & Cooper, P. R. Treatment of disc and ligamentous diseases of the cervical spine by anterior approach. Youmans J. R. Neurological Surgery. 4 ed, Saunders, 1996. V.3. Chapter 96, pp 2253-61.

Goss, C. M. Gray Anatomia. 29 ed. Guanabara Koogan, Rio de Janeiro, 1977, pp 80-237 and 462-551.

Heller, J. G. & Pedlow, F. X. Jr. Anatomy of the cervical spine. Clark C. R. The Cervical Spine-The Cervical Spine Research Society. 3 ed, Lippincott-Raven, Philadelphia, 1998. Chapter 1, pp 3-36.

Kamimura, M.; Ebara, S.; Itoh, H.; Tateiwa, Y.; Kinoshita, T. & Takaoka K: Cervical pedicle screw insertion: Assessment of safety and accuracy with computer-assisted image guidance. J. Spinal Disorders, 13(3):218-24, 2000.

Kapandji, I. A. Fisiologia Articular. Manole, São Paulo, 1990. V.3., Cap 1, pp 7-51.

Karaikovic, E. E.; Kunakornsawat, S.; Daubs, M. D.; Madsen, R. W. & Gaines, R W Jr. Surgical anatomy of the cervical pedicles: landmarks for posterior cervical pedicle entrance localization. J. Spinal Disorders, 13(1):63-72, 2000.

Osborn, A. N. Neurorradiologia Diagnostica. Davis W. L. & Jacobs J. Vasculatura Cerebral: anatomia normal y patologica. Editora Mosby, 1996. Cap 6, pp 117-53.

Pereira, M. G. Epidemiologia Teórica e Prática. Guanabara, 1995. Cap 12, pp 269-88.

Riew, K. D. Microscope-assisted anteriorcervical decompression and plating techniques for multilevel cervical spondylosis. Operative Techniques in Orthopaedics, 8,(1):22-33, 1988.

Sonntag, V. K. H. & Dickman, C. A. Posterior occipital C1-C2 instrumentation. Menezes A. H. & Sonntag V. K. H. Principles of Spinal Surgery. McGraw-Hill, New York, 1996. V.2., Chapter 68, pp 1067-79.

Taveras, J. M. Diagnostic Neuroradiology. 2 ed. Willians & Wilkins, Baltimore, 1976. Vol II, pp 543-986.

Vaccaro, R.; Ring D. & Scuderi, G. Vertebral artery location in relation to the vertebral body as determined by two-dimensional computed tomography evaluation. Spine, 19:2637-41, 1994.

Winter, R. B.; Lonstein, J. W.; Denis, F. & Smith M. D. Atlas of Spine Surgery.W. B. Saunders, 1995. Chapter IV, pp 47-71 and Chapter V, pp 73-103.

Correspondence to:
Prof. Dr. Johnny Wesley Gonçalves Martins
QRSW 8, Bloco B1, Apt. 101, Setor Sudoeste,
CEP 70675-821
Brasília/DF
BRAZIL
Telefone: 55 (61) 344-7979
Fax: 55 (61) 245-7621
Email: colunavertebral@zipmail.com.br

Received : 30-12-2002
Accepted : 06-02-2003 

---

^* Neuranatomy Department, Catholic University of Brasília, Brasília/DF, Brazil.
^** Neuropediatrics Department, University of Brasília, Brasília/DF, Brazil.
^*** Neuranatomy Department, Catholic University of Brasília, Brasília/DF, Brazil.