Vascularization of the spinal cord can be divided into three distinct regions (cervical, thoracic, and lumbar regions) based on blood supply and function, with all regions having approximately 62 arteries. Only a fraction of those arteries, called radiculomedullary arteries (RMAs), directly supply the spinal cord. The cervical region is most richly vascularized by these RMAs. In contrast, the thoracic and lumbar regions receive less blood via arterial branches of this type.1–4 Among these scarce thoracolumbar arteries, the artery of Adamkiewicz (AKA) serves as the main avenue of blood from the aorta to the anterior spinal cord’s lower thoracic and lumbar regions.
Because of the AKA’s significance in spinal cord vascularization, great care is taken to avoid its injury during surgical procedures and treatments. Despite the advancement of endovascular and minimally invasive techniques at the thoracolumbar region, risk of harm to the AKA is always possible. The devastating consequences of damage to this artery resulting in paraplegia and bladder dysfunction have garnered much attention.5 Efforts to reduce the risk of injury to this often-elusive artery require a comprehensive knowledge of the vascular anatomy of the thoracic and lumbar spine and the use of state-of-the-art angiographic imaging. The AKA is commonly characterized as singular in number, sourced from the left side of the aorta, and located in the lower thoracolumbar spinal region. However, exceptions to these conditions seem to be more frequent and varied than previous studies have suggested.1,5,6 We conducted a systematic surgical anatomical study of the AKA and other supporting RMAs to determine their frequency, length, diameter, laterality, spinal position, and accessory variants.
Methods
We dissected and studied 25 cadaveric specimens, which were injected with latex, fixed in formaldehyde, and preserved in alcohol. Anterior corpectomies from T4 to L5 were performed in all but one of the specimens. The dura mater was exposed and opened using microsurgical techniques; subsequently, the anterior spinal artery (ASA) and the radicular and radiculomedullary branches at their extraforaminal, intraforaminal, and subarachnoid courses were dissected. We then summarized the trajectory of the AKAs by recording the vertebral level at which they entered the spinal cord, their lateralization, their number, their length, and their diameters at different points within the course of the artery. We identified the AKAs as RMAs that join the ASA in a hairpin configuration. The RMA with the greatest diameter was considered the AKA, and any remaining RMAs of smaller caliber were also studied. Posterior bilateral laminectomies and foraminotomies from T4 to L5 were performed in only one specimen. In this sole case, the complete spinal cord, including thoracic and lumbar nerve roots, was removed as a single piece for demonstration purposes.
Results
All RMAs, including the AKA, originated either from a posterior intercostal or from a lumbar segmental artery at the foraminal level. From this point, the RMA travels up the spinal cord, where it anastomoses with the ASA in a hairpin turn. The AKA is predominantly left lateralized (81%), and most commonly enters the spinal cord in the T9–L1 region (81%). However, the origin of this artery may be found within a broader range (T8–L4), often making localization of this artery difficult.
Location of RMAs
At least one RMA was found in all the 25 anatomical specimens studied, totaling 37 RMAs among all specimens (Tables 1 and 2). The female/male sex ratio of the 25 cadaveric dissections was 3:2. Neither the laterality nor location of the RMAs was sex-specific in our study.
Anatomical characterization of the AKA and additional radiculomedullary arteries in cadavers 1–12
| Specimen No. | Sex | RMA Laterality | RMA Diameter (mm) | RMA Origin | HP Turn Level | RMA Length (cm) | Diameter (mm) | |||
|---|---|---|---|---|---|---|---|---|---|---|
| RMA Foraminal Portion | Posterior Intercostal or Lumbar Artery | ASA Pre-HP | ASA Post-HP | |||||||
| 1 | F | Lt | 0.6 | T8 | T8 | 3.2 | 1.1 | 2.6 | 0.6 | 0.7 |
| Lt | 0.9 | T10 | T10 | 3.6 | 1.1 | 3.1 | 0.7 | 1 | ||
| Lt | 1 | L1 | T11–12 | 7 | 1 | 2.8 | 1 | 1.1 | ||
| 2 | F | Lt | 0.8 | T10 | T9–10 | 4 | 0.8 | 1.9 | 0.5 | 0.9 |
| 3 | M | Lt | 1 | L2 | T11 | 10.5 | 1.1 | 2.9 | 0.8 | 1 |
| 4 | F | Lt | 0.7 | T8 | T7–8 | 3.4 | 0.7 | 2.3 | 0.5 | 0.8 |
| Lt | 0.8 | T11 | T10–11 | 4.3 | 0.9 | 2.9 | 0.8 | 0.9 | ||
| 5 | M | Rt | 0.9 | T11 | T10–11 | 4.7 | 1 | 3 | 0.6 | 1.1 |
| 6 | F | Lt | 0.4 | T9 | T8–9 | 3.2 | 0.4 | 3.4 | 0.4 | 0.5 |
| Lt | 0.7 | T12 | T11–12 | 5.5 | 0.7 | 2.7 | 0.5 | 0.8 | ||
| 7 | F | Lt | 0.9 | T11 | T10–11 | 5.5 | 0.9 | 2.7 | 1 | 1.3 |
| 8 | F | Rt | 0.4 | T8 | T7–8 | 3.5 | 0.4 | 1.8 | 0.4 | 0.7 |
| Rt | 0.6 | T12 | T10–11 | 6.7 | 0.6 | 2.3 | 0.7 | 0.8 | ||
| 9 | M | Rt | 0.6 | T9 | T8–9 | 4.1 | 0.6 | 2.1 | 0.4 | 0.6 |
| Lt | 0.6 | T11 | T10–11 | 5.6 | 0.6 | 2 | 0.6 | 0.7 | ||
| 10 | F | Lt | 0.5 | T10 | T9–10 | 3 | 0.6 | 2.4 | 0.3 | 0.5 |
| 11 | F | Lt | 0.7 | T12 | T11–12 | 7.4 | 0.7 | 3.6 | 0.7 | 1 |
| 12 | M | Lt | 0.5 | T6 | T5–6 | 3.3 | 0.6 | 1.8 | 0.4 | 0.6 |
| Lt | 0.6 | T8 | T7–8 | 4 | 0.6 | 3 | 0.6 | 0.8 | ||
| 13 | F | Lt | 0.9 | T9 | T8–9 | 3.9 | 1 | 2.9 | 0.8 | 1.2 |
| 14 | M | Lt | 0.6 | T9 | T8–9 | 4.1 | 0.8 | 3.2 | 0.6 | 1 |
| 15 | M | Lt | 0.7 | T10 | T9–10 | 4.6 | 0.9 | 2.1 | 0.8 | 1 |
| 16 | M | Lt | 0.9 | T9 | T8–9 | 5.4 | 2 | 2.9 | 0.9 | 1.2 |
| 17 | M | Lt | 0.8 | T8 | T7–8 | 3.8 | 0.9 | 2.7 | 0.5 | 0.9 |
| 18 | M | Rt | 0.4 | T8 | T7–8 | 5.7 | 0.5 | 2.5 | 0.5 | 0.7 |
| 19 | M | Rt | 0.7 | T7 | T6–7 | 3.3 | 0.8 | 1.2 | 0.6 | 0.7 |
| Lt | 1 | T11 | T10–11 | 6.2 | 1.2 | 1.5 | 0.7 | 0.7 | ||
| 20 | F | Lt | 0.8 | T9 | T8–9 | 2 | 0.8 | 1.9 | 0.8 | 1.2 |
| Lt | 0.8 | T12 | T11–12 | 6 | 0.8 | 2.3 | 1.2 | 1.5 | ||
| 21 | F | Lt | 0.9 | T8 | T7–8 | 3.1 | 0.9 | 1.5 | 0.6 | 0.7 |
| Lt | 0.9 | T9 | T8–9 | 4 | 0.9 | 1.6 | 0.6 | 0.8 | ||
| Lt | 1.3 | T11 | T9–10 | 6.7 | 1.3 | 2 | 0.8 | 1.2 | ||
| 22 | F | Lt | 1.2 | T9 | T8–9 | 3.2 | 1.2 | 2.8 | 0.7 | 1.2 |
| 23 | F | Lt | 0.9 | T9 | T8–9 | 3.2 | 1.4 | 3.3 | 0.7 | 1.3 |
| 24 | F | Rt | 1.2 | T11 | T10–11 | 4.2 | 1.2 | 2.1 | 1 | 1.4 |
| 25 | F | Lt | 0.3 | T9 | T8–9 | 1.77 | 0.3 | 1.2 | 0.1 | 0.2 |
| Lt | 0.7 | L4 | T10–11 | 10.71 | 0.8 | 2.7 | 0.35 | 0.7 | ||
HP = hairpin.
Items in boldface type describe the designated AKA(s) in each specimen.
Summary of Table 1 describing 37 total RMAs in 25 cadavers
| Sex | Laterality | Mean RMA Diameter (mm) | RMA Origin | HP Turn Level | Mean RMA Length (cm) | Mean RMA Foraminal Portion (mm) | Mean Diameter (mm) | ||
|---|---|---|---|---|---|---|---|---|---|
| Posterior Intercostal or Lumbar Artery | ASA Pre-HP | ASA Post-HP | |||||||
| 15 F, 10 M | 30 lt, 7 rt | 0.76 ± 0.23 | T6, 2.7%; T7, 2.7%; T8, 18.9%; T9, 27%; T10, 10.8%; T11, 18.9%; T12, 10.8%; L1, 2.7%; L2, 2.7%; L4, 2.7% | T5–6, 2.7%; T6–7, 2.7%; T7–8, 18.9%; T8–9, 27%; T9–10, 13.5%; T10–11, 24.3%; T11–12, 10.8% | 4.71 ± 1.98 | 0.87 ± 0.32 | 2.42 ± 0.62 | 0.64 ± 0.22 | 0.90 ± 0.28 |
In our study, 30 (81%) of the 37 total RMAs arose from a left intercostal or lumbar artery, and 7 (19%) arteries arose from the right. The origins of the RMA were found from T6 to L4. Thirty-two (86%) of the RMAs traveled through an intervertebral foramen ranging from T8 to T12. The most prevalent RMA level of origin was T9 (27%), from which 10 of the 37 arteries arose.
The intersection of the RMA hairpin turn with the ASA in the spinal cord marks the termination of the RMA. The location of these RMA hairpins ranged from T7 to T12, with 22 (59%) located from the T8 to T11 vertebral levels. The most prevalent hairpin turn level was T8–9 (27%).
Location of the AKA
Twenty-seven AKAs, which we defined as the largest RMA in each specimen, were found among our 25 specimens (Tables 1 and 2). Similarly, we found the majority of AKAs to be left lateralized. Twenty-two (81%) arteries arose from a left intercostal or lumbar artery (Figs. 1–3). The AKA was found to originate from an intervertebral foramen within the T8–L4 range (Figs. 1–4). The majority (81%) of AKAs originated from the T9–L1 range. The most prevalent levels of origin were T9 (7 of 27 AKAs) and T11 (7 of 27 AKAs).
Specimen 19. Cadaveric dissection. The anterior and middle spinal columns have been removed. Anterior view of the ASA supplied by a left AKA, which originated from the left T11 intercostal artery. Dent. l. = dentate ligament; Dorsal b. = dorsal branch; Filum T. a. = artery of the filum terminale; Interc. n. = intercostal nerve; Interc. v. = intercostal vein; Musc. b. = muscular branch; Post. Segm. a. = posterior segmental artery; Rad. Med. b. = radiculomedullary branch. © Jorge Alvernia, published with permission.
Specimen 19. A: Cadaveric dissection. The anterior and middle spinal columns have been removed. Anterior view of the ASA supplied by a left T11 AKA and from a contralateral right T7 RMA. B: Schematic drawing of cadaveric dissection in panel A. Filum T. a. = artery of the filum terminale; Interc. a. = intercostal artery; Interc. n. = intercostal nerve; Interc. v. = intercostal vein; Musc. b. = muscular branch; Post. b. = posterior branch; Post. Rad. a. = posterior radicular artery; Post. Segm. a. = posterior segmental artery; Rad. med. b. = radiculomedullary branch. © Jorge Alvernia, published with permission.
Specimen 21. Cadaveric dissection. The anterior and middle spinal columns have been removed. The ASA is supplied by a left AKA originating from a T11 posterior segmentary artery (AKA) and from two ipsilateral RMAs, one originating from a T8 intercostal artery and the other from a T9 intercostal artery. © Jorge Alvernia, published with permission.
Specimen 25. A: Cadaveric dissection. The anterior, middle, and posterior spinal columns have been removed, and the spinal cord and cauda equina have been removed for better visualization. The ASA is supplied by an AKA originating from a radicular artery accompanying the L4 nerve root (L4 n.r.). B: Cadaveric dissection of the same specimen as in panel A. The anterior, middle, and posterior spinal columns have been removed. The ASA is supplied by an additional RMA originating from a T9 intercostal artery. C: Schematic drawing depicting the ASA supplied by both the AKA originating from an L4 radicular artery and an RMA originating from a left T9 intercostal artery. C.I.a. = common iliac artery; Interc. n. = intercostal nerve; n.r. = nerve root. © Jorge Alvernia, published with permission.
The termination of the AKA at its hairpin loop ranged from T7 to T12, with 20 (74%) hairpin turns located from T8 to T11. The most common hairpin turn level was T10–11, at which 8 of 27 AKAs terminated.
AKA Segmental Measurements
From its entry point to the spinal cord at the intervertebral foramen, the AKA was found to travel the distance of 1 to 7 vertebrae up the spinal cord, averaging 36.1 ± 17.7 mm (± SD). The length of the artery itself from the intervertebral foramen to its conjunction with the ASA averaged 52.6 ± 20.3 mm (Table 1). The diameter of this artery decreased as it branched off from the intercostal or lumbar artery, which had a diameter of 2.56 ± 0.50 mm. At its origin in the intervertebral foramen, the AKA itself was 0.92 ± 0.32 mm in diameter, and at its ascending segment in the spinal cord it was 0.81 ± 0.22 mm in diameter.
Effects of the AKA on the ASA
As previously mentioned, the AKA is the first rich source of blood to the ASA after the sparsely vascularized thoracic region of the spinal cord. The impact of the AKA on thoracolumbar spinal cord vascularization was demonstrated by the substantial increase in the ASA diameter after meeting the AKA (mean 0.99 mm ± 0.26 mm) compared with the prejunction diameter (0.70 mm ± 0.21 mm) (Tables 1 and 2).
Multiple AKAs
Most of our results showed a single AKA supplying the ASA. However, 2 (8%) of our specimens had two RMAs of the same caliber at different vertebral levels (Tables 1 and 2). In each case, we assumed that both arteries were equally weighted in their contribution to spinal cord blood supply because of their same caliber, so both branches were considered AKAs. The two AKAs in one of the two specimens were ipsilateral to each other and left lateralized. The two AKAs in the other specimen were contralateral with a right-lateralized superior branch and a left-lateralized inferior branch.
Other RMAs
Although most thoracolumbar vascularization is usually attributed to the AKA, our results demonstrate that a significant percentage of individuals have additional RMAs of a lower caliber, which certainly play an important role. We found at least one of these additional RMAs in 8 (32%) of the 25 cadaveric specimens (Tables 1 and 2). The average length of these additional RMAs from the intervertebral foramen to the ASA was substantially shorter (32 ± 5.8 mm) than that of the AKA, and their average diameter was smaller (0.63 ± 0.23 mm) than that of the AKA. The additional RMA(s) predominantly originated from a left-sided segmental artery within the T7–10 range. However, they were found as high as T6 and as low as T10. Additional RMAs were often present when their respective AKAs originated from the T11 level or lower. Notably, all additional RMAs originated superiorly to their respective AKAs (Figs. 1–4).
Additionally, all but one (Fig. 2) of the additional RMAs were ipsilateral to their AKA (Figs. 3 and 4). In each of two specimens, two additional RMAs were found, and both were left lateralized and ipsilateral to their AKAs (Fig. 4).
Discussion
The AKA, the dominant RMA of the thoracolumbar region of the spine, is responsible for the majority of the blood flow to the caudal portion of the ASA and is also known as the artery of lumbar enlargement, great anterior RMA, and arteria radicularis magna.2,7,8 The discovery of the AKA is predominantly attributed to Albert Wojciech Adamkiewicz (1850–1921), who noted the uneven arterial supply to the spinal cord by spinal arteries of varying caliber and was the first to coin the name "arteria radicularis magna" for the largest of these arteries. Nevertheless, there remains a lack of recent anatomical studies concerning the AKA and its possible variants.
Anatomy
The blood supply of the AKA can be sourced back to the descending thoracolumbar aorta from which a segmental (intercostal or lumbar) artery branches off and divides into posterior and anterior branches. The posterior segmental artery splits into a muscular branch, a dorsal branch, and a radiculomedullary branch. The RMA divides into the posterior branch and the main anterior branch. From its origin, the AKA travels through the intervertebral foramen and up the spinal cord, where it eventually takes a sharp, caudal turn and joins the ASA.5,7,9,10 The portion of the ASA supplied by the AKA descends until it anastomoses with the ends of the two posterior spinal arteries (PSAs) on the underside of the conus medullaris, forming the arterial basket, from which the merged PSAs and ASA continue down the cauda equina as one artery.1,11
The difficulty in identifying the AKA lies in the high variability in its lateralization and spinal level from person to person. Previous studies have shown that the AKA commonly arises from a left-sided intercostal or lumbar artery between T9 and T12. However, a substantial percentage of AKAs arise from the right side and can be located within an even broader range of vertebral levels.3,5 The AKA is almost always located within the top half of the intervertebral foramen through which it travels.5,10 The diameter of this artery varies throughout its course, ranging from 0.5 to 1.3 mm, and the distinct hairpin turn formed where the AKA meets the ASA is its main identifying characteristic (Fig. 1).2,6,12–15
Clinical Importance: From the Belief of a Single AKA to Evidence of a Collateral Network
Several studies have attempted to find the maximum number of thoracic and/or lumbar segmental artery levels that can be sacrificed without a postoperative neurological deficit. Animal studies have approached this subject extensively.16–18 Kato et al., in a study of 25 dogs, showed that, although spinal cord blow decreases as the number of bilateral segmental arteries are interrupted, including those that supply the AKA, it is necessary to interrupt four or more consecutive bilateral segmental arteries to produce ischemic spinal cord dysfunction.18
Tan et al.19 recently reviewed the spine and vascular and surgery literature on this topic using the PRISMA guidelines. They found that when 1 to 6 pairs of spinal segmental arteries (without knowledge of the AKA location) were ligated, the postoperative neurological deficit was just 0.6%, compared with 5.4% when more than 6 bilateral pairs of spinal segmental arteries were ligated. On the other hand, the postoperative risk of neurological deficit did not exceed 1.3% when 2 to 9 unilateral spinal segmental arteries were interrupted.
There is only one study from the spine literature so far addressing the question of the results of ligating the AKA and the presence of postoperative neurological deficits. In that study, Murakami et al.20 reported on 15 patients who underwent total en bloc spondylectomies for spine tumors, including ligature of the AKA at the nerve root level, without changes in the patients’ neurological status. They concluded that sacrificing three pairs of segmental arteries, even including the AKA, can be done safely.
The importance of accessory RMAs cannot be overemphasized. As our results show, one or two additional RMAs from T4 to L5 were found in one-third of the series, including two AKAs in 8% of the cases. Furthermore, other posterior RMAs were not part of our analysis, but their contribution to the collateral circulation of the spinal cord should also be taken into account.
The old concept that the AKA is the primary determinant factor for thoracic spinal cord supply may need to be reevaluated. Although most articles provide level IV evidence, enough animal research has demonstrated the same results as available clinical series. While a randomized controlled trial or even a prospective comparative study on this topic will be challenging, building evidence shows that collateral circulation is generally available in most cases when several segmentary arteries are ligated, even the ones supplying the AKA.
Some questions without clear answers remain. The proportion of patients with vascular pathology who developed neurological deficits is higher than those with spinal pathology (4.0% vs 0.5%) when spinal segmental arteries are interrupted with and without occlusion of the AKA.19 Different physiopathology and/or different surgical techniques employed may explain this difference. Additionally, it is crucial to mention that the point of vascular occlusion between the spine and vascular procedures is not the same. While in spine surgery the interruption of the segmentary arteries is generally 1–2 cm from the intervertebral foramen, in vascular surgery (endovascular or open) it is proximal at the aortic origin. The role of additional distal segmental branches at the level of the foramen as a source of collateral circulation and vascular network, such as muscular or dorsal branches at the foramen level, remains to be elucidated. Moreover, the physiopathology of ischemia by ligation differs from that of a thromboembolic event. The recruitment and collateral circulation are likely in the first while absent in the second.
Imaging Techniques
Because of the variable location of the AKA, preoperative identification of this artery in interventional procedures regarding the thoracic and lumbar spinal cord is paramount. High-quality imaging techniques are necessary to prevent misidentification of the AKA with similar blood vessels, such as the great anterior radiculomedullary vein, which is usually found at a lower level than the AKA and possesses a characteristic hairpin turn very similar to that of the AKA.6,7 The proper use of imaging tools in combination with a detailed knowledge of the anatomical course of the AKA may be helpful in planning applicable endovascular spinal procedures and treatments.
Digital Subtraction Angiography
DSA has been the gold standard for preoperative arterial identification because of its consistently superior image quality. It often allows for the visualization of smaller arterial feeder vessels that may be missed using other methods. In a study comparing DSA with contrast-enhanced MRA, DSA was found to be superior in terms of vessel conspicuity, sharpness, continuity, and background homogeneity.21 However, a significant risk posed by DSA to the thoracolumbar spine lies in the possible occlusion of arteries or branches through which the catheter may travel. This risk is counterintuitive to the protection of the AKA by preprocedural localization. Additionally, the intercostal arteries of the thoracolumbar spine have an increased risk of atherosclerosis, especially in the elderly, which makes identifying this artery much more difficult.6
CTA and MRA
The methods for preprocedural localization of the AKA have evolved over the past century. While DSA is currently the gold standard for preoperative arterial identification, current developments in MRA and CTA show that these noninvasive techniques may eventually provide a promising alternative.7,12 Because these techniques do not require catheterization, the risk of arterial injury is nearly eliminated. Aside from a significantly decreased neurological risk factor, the use of MRA or CTA over catheter angiography is less costly and generally exposes patients to lesser amounts of harmful radiation compared with DSA.5 The quality of MRA and CTA imaging can be enhanced by employing specific techniques, including the contrast-enhanced 3D MRA approach. Table 3 summarizes findings on the various locations of the AKA from different studies using multiple imaging techniques.5,7,8,10,12,14,15,21–24 The studies using MRA and CTA produced a mean recognition rate of 82.3%. While this recognition rate is still not high enough to warrant using these noninvasive techniques over DSA, these results look promising in the evolution of spinal angiography toward more patient-friendly approaches.
Literature review of AKA recognition and characterization using various imaging methods
| Authors & Year | Imaging Method | No. of Subjects | % AKA Recognition | % Single AKA | % Multiple AKA | Lateralization of AKA | Vertebral Level of AKA Origin | ||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| % Lt | % Rt | % T5–7 | % T8–10 | % T11–12 | % L1–5 | ||||||
| Bley et al., 20107 | MRA | 68 | 88 | 100 | 0 | 65 | 35 | 1.6 | 68.9 | 21.3 | 8.2 |
| Guziński et al., 20175 | MD CTA | 200 | 21.5 | 97.7 | 2.3 | 83.7 | 14 | 2.3 | 29.5 | 59.1 | 9.1 |
| Kawaharada et al., 20048 | MRA | 120 | 82.5 | 88.9 | 11.1 | 95.5 | 4.5 | 0.9 | 54.6 | 43.6 | 0.9 |
| Kudo et al., 200312 | MD CTA | 19 | 68.4 | 100 | 0 | 69.2 | 30.8 | 0 | 30.8 | 46.1 | 23.1 |
| Melissano et al., 200914 | CTA & OsiriX | 67 | 67.1 | 86.3 | 13.7 | 66.1 | 33.9 | NA | NA | NA | NA |
| Murthy et al., 201010 | DSA or MSA | 248 | 46.4 | 96.5 | 3.5 | 83.3 | 16.7 | NA | NA | NA | NA |
| Nijenhuis et al., 200621 | CE-MRA & DSA | 15 | 100 | 100 | 0 | 67 | 33 | 0 | 26.7 | 40 | 33.3 |
| Ou et al., 200715 | CTA | 40 | 95 | 100 | 0 | 71.1 | 28.9 | 2.6 | 89.5 | 7.9 | 0 |
| Uotani et al., 200822 | IA CTA/IV CTA | 32 | 78.1 | 100 | 0 | 72 | 28 | 4 | 72 | 16 | 8 |
| Williams et al., 199123 | DSA | 47 | 55.3 | 100 | 0 | 84.6 | 15.4 | 3.9 | 61.5 | 19.2 | 15.4 |
| Yoshioka et al., 200624 | MRA/CTA | 30 | 96.7 | 100 | 0 | 72.4 | 27.6 | 0 | 55.2 | 31 | 13.8 |
CE = contrast-enhanced; IA = intra-arterial; IV = intravenous; MD CTA = multidetector CTA; MSA = manual subtraction angiography; NA = not applicable.
Clinical Applications
Preprocedural localization of the AKA may prove important in the maintenance of proper thoracolumbar spinal cord function in various treatments and procedures ranging from major spinal surgeries and spinal vascular malformation treatments to common spinal surgery pain injections.
Pain Therapy
Spinal injections in the thoracolumbar region, such as transforaminal epidural injections used to treat radicular pain, may pose risks to the safety of the AKA. Transforaminal epidural injections may injure the portion of the AKA that runs through the intervertebral foramen. This injury may cause thrombosis or emboli and create a significant risk of spinal cord ischemia and significant neurological deficit.25 The use of nonsoluble, or particulate, steroids in these epidural injections may result in an arterial embolism and consequential infarction of the nervous tissue fed by the AKA or other RMAs.4 As shown in this study, the AKA may originate from as low as L4, a common location for this kind of injection. Therefore, using nonparticulate steroids and ensuring the absence of arterial uptake during contrast injection are mandatory when administering transforaminal epidural injections.
Metastatic Tumor Surgery
The use of en bloc vertebrectomy to treat primary or thoracolumbar metastatic tumors such as renal cell carcinoma often involves removing the whole or most of the vertebra. Thus, pedicle osteotomies are often an essential part of the procedure, so special care to avoid damage to major vessels such as the AKA, whose course runs close to the pedicle, must be taken.26,27 In addition, preoperative surgical preparation for open surgery of a spinal tumor may involve embolization of its feeders, such as the AKA or other radiculomedullary branch, to prevent significant bleeding.25 Therefore, the location of the AKA and other RMAs traveling anteriorly or posteriorly into the vertebral foramen near the tumor site must also be visualized in the preprocedural imaging of the spine to avoid any iatrogenic injury and neurological deficit.
Spinal Malformation Treatment
Spinal dural arteriovenous fistulas (SDAVFs) result from abnormal vascular shunting between one radicular artery or arteries to the radiculomedullary veins. They are typically found in the thoracolumbar region of the spine and can cause progressive paraplegia.28 The standard treatment for SDAVFs is occlusion of the shunting zone of the abnormal artery by endovascular treatment or open surgery, or a combination of both. In most cases, this treatment relies on cutting off the abnormal shunting provided by one or more radicular arteries or RMAs.29 The possible involvement of the AKA and additional anterior or posterior RMAs, including the PSAs, makes preoperative knowledge of the patient’s thoracolumbar vasculature imperative in SDAVF treatment.
Variants of the AKA
Most previous studies have defined the AKA as a singular artery responsible for the vascularization of the thoracolumbar spinal cord. In contrast, some studies have recognized rare occurrences of two or three AKAs or a single supplementary artery to the AKA if the AKA originates from a higher spinal level.1,2,8,10 One previous study even recognized vessels of a smaller caliber that travel through the intervertebral foramen and are thought to contribute to spinal cord vascularization cooperatively with the AKA.5
One study by Gailloud focused on a prominent and consistent RMA at the T5 level and attributed the discovery of this artery to Albrecht von Haller (1708–1777).30 The so-called artery of von Haller may play an essential role in the vascularization of the upper to middle thoracic region, previously recognized as a mere watershed area with no significant direct supply to the ASA. However, Gailloud suggested that the small caliber of this artery is proportional to the vascular needs of this spinal region and should still be considered as an essential spinal contributor. Additionally, he concluded that the presence of the artery of von Haller is independent of the presence of the AKA and that this smaller caliber contributor is frequently overlooked in angiographic imaging.
While our findings show a significant number of additional radiculomedullary contributors in the middle to lower thoracic region of the spinal cord, a prominent artery at the T5 level was not consistently found among our specimens. In our study, all additional radiculomedullary branches were found lower than T5 and as low as T10. Furthermore, our results suggest that other radiculomedullary contributors exist only in the presence of an established AKA of a larger caliber.
Our results yielded a small but significant percentage of specimens with two RMAs of the same caliber. We considered both to be AKAs due to their equal caliber and contribution to the ASA blood supply. In previous imaging studies, multiple AKAs were reported in approximately 11%–14% of patients.8,14 The growing recognition of the presence of multiple AKAs and additional RMAs in anatomical studies of the thoracolumbar spine calls for the standardization of terminology used to label additional RMAs and for new studies centered on the localization of the AKA and its variants whose fundamental role in the vascularization of the thoracolumbar spinal cord affects the maintenance of the patient’s neurological function.
Limitations of Our Study
Even though our study was targeted to evaluate the RMAs supplying the thoracic and lumbar spinal cord, we did not study the cervicothoracic junction, including the first three thoracic segments. Hence, additional arteries such as the artery of Lazorthes and its variations that supply the cervical enlargement were not analyzed. Additionally, posterior RMAs supplying the PSAs were not studied, so their role and frequency remains to be elucidated.
Conclusions
A thorough understanding of the vascular anatomy of thoracolumbar spinal cord is essential to avoid neurological complications that might arise in various spinal procedures and treatments. The segmental RMAs supplying the thoracic and lumbar spinal cord can be unilateral, bilateral, or multiple. Multiple AKAs or additional RMAs are common and should be considered when dealing with the spinal cord at the thoracolumbar level.
Our study adds to the growing evidence that the importance given to the AKA, as the main and only source of vascular supply to the thoracic and lumbar spinal cord, is overrated and somehow outdated. A collateral network concept has emerged as a better way to thoroughly understand the vascular architecture of the spinal cord.
Acknowledgments
We thank Professor Marc Sindou for his support and continuous guidance in perfectioning surgical microanatomy techniques, which helped to enhance the results of our anatomical study.
Disclosures
The authors report no conflict of interest concerning the materials or methods used in this study or the findings specified in this paper.
Author Contributions
Conception and design: Alvernia, Simon. Acquisition of data: Alvernia, Simon, Khandelwal. Analysis and interpretation of data: Alvernia, Khandelwal, Ramos, Perkins, Kim, Mertens. Drafting the article: Alvernia, Ramos. Critically revising the article: all authors. Reviewed submitted version of manuscript: all authors. Approved the final version of the manuscript on behalf of all authors: Alvernia. Statistical analysis: Khandelwal, Ramos. Administrative/technical/material support: Simon, Perkins, Mertens, Messina, Luzardo, Diaz. Study supervision: Alvernia, Perkins, Kim, Luzardo.
Supplemental Information
Previous Presentations
Portions of this paper were presented at the 2019 CNS Annual Meeting and won the Best Pain Basic Science Poster Award Presentation, October 21, 2019, San Francisco, California.
References
- 2↑
Hoehmann C, Hitscherich K, Cuoco J. The artery of Adamkiewicz: vascular anatomy, clinical significance and surgical considerations. Int J Cardiovasc Res. 2016;5(6).
- 3↑
Lazorthes G, Gouaze A, Zadeh JO, Santini JJ, Lazorthes Y, Burdin P. Arterial vascularization of the spinal cord. Recent studies of the anastomotic substitution pathways. J Neurosurg. 1971;35(3):253–262.
- 4↑
Soubeyrand M, Court C, Fadel E, et al. Preoperative imaging study of the spinal cord vascularization: interest and limits in spine resection for primary tumors. Eur J Radiol. 2011;77(1):26–33.
- 5↑
Guziński M, Bryl M, Ziemińska K, Wolny K, Sąsiadek M, Garcarek JS. Detection of the Adamkiewicz artery in computed tomography of the thorax and abdomen. Adv Clin Exp Med. 2017;26(1):31–37.
- 6↑
Backes WH, Nijenhuis RJ. Advances in spinal cord MR angiography. AJNR Am J Neuroradiol. 2008;29(4):619–631.
- 7↑
Bley TA, Duffek CC, François CJ, et al. Presurgical localization of the artery of Adamkiewicz with time-resolved 3.0-T MR angiography. Radiology. 2010;255(3):873–881.
- 8↑
Kawaharada N, Morishita K, Hyodoh H, et al. Magnetic resonance angiographic localization of the artery of Adamkiewicz for spinal cord blood supply. Ann Thorac Surg. 2004;78(3):846–852.
- 10↑
Murthy NS, Maus TP, Behrns CL. Intraforaminal location of the great anterior radiculomedullary artery (artery of Adamkiewicz): a retrospective review. Pain Med. 2010;11(12):1756–1764.
- 11↑
Martirosyan NL, Kalani MYS, Lemole GM Jr, Spetzler RF, Preul MC, Theodore N. Microsurgical anatomy of the arterial basket of the conus medullaris. J Neurosurg Spine. 2015;22(6):672–676.
- 12↑
Kudo K, Terae S, Asano T, et al. Anterior spinal artery and artery of Adamkiewicz detected by using multi-detector row CT. AJNR Am J Neuroradiol. 2003;24(1):13–17.
- 13
Martirosyan NL, Feuerstein JS, Theodore N, Cavalcanti DD, Spetzler RF, Preul MC. Blood supply and vascular reactivity of the spinal cord under normal and pathological conditions. J Neurosurg Spine. 2011;15(3):238–251.
- 14↑
Melissano G, Bertoglio L, Civelli V, et al. Demonstration of the Adamkiewicz artery by multidetector computed tomography angiography analysed with the open-source software OsiriX. Eur J Vasc Endovasc Surg. 2009;37(4):395–400.
- 15↑
Ou P, Schmit P, Layouss W, Sidi D, Bonnet D, Brunelle F. CT angiography of the artery of Adamkiewicz with 64-section technology: first experience in children. AJNR Am J Neuroradiol. 2007;28(2):216–219.
- 16↑
Ueda Y, Kawahara N, Tomita K, Kobayashi T, Murakami H, Nambu K. Influence on spinal cord blood flow and function by interruption of bilateral segmental arteries at up to three levels: experimental study in dogs. Spine (Phila Pa 1976). 2005;30(20):2239–2243.
- 17
Fujimaki Y, Kawahara N, Tomita K, Murakami H, Ueda Y. How many ligations of bilateral segmental arteries cause ischemic spinal cord dysfunction? An experimental study using a dog model. Spine (Phila Pa 1976). 2006;31(21):E781–E789.
- 18↑
Kato S, Kawahara N, Tomita K, Murakami H, Demura S, Fujimaki Y. Effects on spinal cord blood flow and neurologic function secondary to interruption of bilateral segmental arteries which supply the artery of Adamkiewicz: an experimental study using a dog model. Spine (Phila Pa 1976). 2008;33(14):1533–1541.
- 19↑
Tan T, Rutges J, Marion T, Fisher C, Tee J. The safety profile of intentional or iatrogenic sacrifice of the artery of Adamkiewciz and its vicinity’s spinal segmental arteries: a systematic review. Global Spine J. 2020;10(4):464–475.
- 20↑
Murakami H, Kawahara N, Tomita K, Demura S, Kato S, Yoshioka K. Does interruption of the artery of Adamkiewicz during total en bloc spondylectomy affect neurologic function? Spine (Phila Pa 1976). 2010;35(22):E1187–E1192.
- 21↑
Nijenhuis RJ, Mull M, Wilmink JT, Thron AK, Backes WH. MR angiography of the great anterior radiculomedullary artery (Adamkiewicz artery) validated by digital subtraction angiography. AJNR Am J Neuroradiol. 2006;27(7):1565–1572.
- 22↑
Uotani K, Yamada N, Kono AK, et al. Preoperative visualization of the artery of Adamkiewicz by intra-arterial CT angiography. AJNR Am J Neuroradiol. 2008;29(2):314–318.
- 23↑
Williams GM, Perler BA, Burdick JF, et al. Angiographic localization of spinal cord blood supply and its relationship to postoperative paraplegia. J Vasc Surg. 1991;13(1):23–35.
- 24↑
Yoshioka K, Niinuma H, Ehara S, Nakajima T, Nakamura M, Kawazoe K. MR angiography and CT angiography of the artery of Adamkiewicz: state of the art. Radiographics. 2006;26(suppl 1):S63–S73.
- 25↑
Arici T, Kurçaloğlu M, Eyıgor C, Uyar M. Transforaminal epidural steroid injection and infraneural approach. Agri. 2019;31(2):104–106.
- 26↑
Bohinski RJ, Rhines LD. Principles and techniques of en bloc vertebrectomy for bone tumors of the thoracolumbar spine: an overview. Neurosurg Focus. 2003;15(5):E7.
- 27↑
Langdon J, Way A, Heaton S, Bernard J, Molloy S. The management of spinal metastases from renal cell carcinoma. Ann R Coll Surg Engl. 2009;91(8):649–652.
- 28↑
Miyasaka K, Asano T, Ushikoshi S, Hida K, Koyanagi I. Vascular anatomy of the spinal cord and classification of spinal arteriovenous malformations. Interv Neuroradiol. 2000;6(suppl 1):195–198.
- 29↑
Krings T, Geibprasert S. Spinal dural arteriovenous fistulas. AJNR Am J Neuroradiol. 2009;30(4):639–648.
- 30↑
Gailloud P. The artery of von Haller: a constant anterior radiculomedullary artery at the upper thoracic level. Neurosurgery. 2013;73(6):1034–1043.
