Abstract
The proximal segment of the vertebral artery most often consists of a persistent sixth cervical intersegmental artery that originates from the subclavian artery, but it may also derive from a fifth, fourth, or third cervical intersegmental artery (in decreasing order of frequency), or from a first thoracic intersegmental artery. The involvement of more cranial cervical branches is exceptional, with no known persistent first cervical intersegmental artery and possibly five cases of persistent second cervical intersegmental arteries reported so far. This report describes a patient with multiple arterial variations including right persistent second cervical intersegmental artery of common carotid origin, distal VA duplication, circumflex aortic arch, and segmental internal carotid agenesis in a context of possible PHACE syndrome.
Keywords: Vertebral artery, embryology, primitive intersegmental artery, anatomical variation
Introduction
Padget1 introduced in 1954 a modern nomenclature of the intersegmental arteries (ISAs) based on the embryological investigations of His (1882), Hochstetter (1890), and Schmeidel (1932). She emphasized the intersegmental rather than segmental nature of the dorsal aortic branches coursing along each pair of spinal nerves, confirmed the developmental association between the sixth cervical ISA and the subclavian artery, and named proatlantal intersegmental artery (ProA) the vessel associated with the first cervical nerve. The notion that the vertebral artery (VA) is a composite vessel combining originally independent segments was also established by these early contributors.2 The cranial segment of the VA derives from the ProA, the cervical segment is a series of longitudinal anastomoses between cervical ISAs, and the proximal segment is a persistent primitive ISA, which can be identified by its point of entrance into the transverse canal. The proximal segment most often consists of a persistent sixth cervical ISA and crosses the C6 transverse foramen, but it may also derive from a fifth, fourth, or third cervical ISA (in decreasing order of frequency), or from a first thoracic ISA. Persistent second cervical ISAs are exceptional, with possibly five cases reported so far.3–7 This report describes a patient with multiple arterial variations including a persistent second cervical ISA.
Case report
The patient was a 71-year-old woman with a history of left parieto-occipital stroke and arterial hypertension consulting for several episodes of vertigo and severe lightheadedness. Her neurological examination at admission was unremarkable at the exception of a left facial palsy and a right lower quadranopsia developed at the time of her prior stroke. Cerebral MRI/MRA confirmed the presence of an old left hemispheric infarct and revealed an occlusion of the distal left internal carotid artery (ICA) and a possible right anterior cerebral artery (A1 segment) aneurysm. Further investigations with CTA and DSA confirmed a 3-mm aneurysm of the posterior aspect of the right A1 segment, a 50% stenosis of the proximal left ICA, and an hypoplastic left VA ending as the left posterior-inferior cerebellar artery. In addition, CTA and DSA documented several anatomical variations, which included:
A right ascending aorta with a retroesophageal segment and a left descending thoracic aorta (right circumflex aortic arch) (Figure 1(a) and (b))
A persistent right second cervical ISA originating from the proximal portion of the right common carotid artery (CCA) and ascending along the anterolateral aspect of the cervical spine to enter the transverse canal through the right transverse foramen of the axis (C2) (Figure 1(c) to (e)). The distal VA was duplicated, with a normal V4 segment (C0–C1) and a prominent right radicular branch along the first cervical nerve (C1–C2) (Figure 1(f)).
A high right carotid bifurcation at C2 (Figure 1(e)).
A segmental agenesis of the left ICA (Figure 1(g) to (i)).
Figure 1.
A 71-year-old patient with multiple arterial variations. (a) DSA, left anterior oblique view. A right-sided ascending thoracic aorta (ASC) passes behind the esophagus (re) and continues as a left-sided descending thoracic aorta (DESC). Supra-aortic branches originate in the following sequence: left common carotid artery (LCCA), right common carotid artery (RCCA), right subclavian artery (RSCA), and left subclavian artery (LSCA). (b) CTA, sagittal reconstruction, showing the retroesophageal segment of the circumflex aorta (re) passing behind the esophagus (E) and the trachea (TR). (c) DSA, right CCA injection, posteroanterior projection, documenting the proximal origin of the right vertebral artery (VA) (gray arrow). The RCCA (large black arrow) and VA (white arrow) have parallel ascending courses lateral to the cervical spine. The duplication of the distal VA is appreciated (black and white arrowheads). (d) CTA, volume-rendered reconstruction, right lateral view. The course of the right VA (white arrow) is anterior and lateral to the cervical spine until it reaches the C2 level, at which point the vessel passes through the right C2 transverse foramen (i.e. persistent second cervical ISA). The distal VA is duplicated, with a dominant limb following a normal course at the C0–C1 level (i.e. spinal branch of the ProA), and a lesser limb at the C1–C2 level (i.e. spinal branch of the first cervical ISA). (e) CTA, volume-rendered reconstruction, left anterior oblique view. The parallel courses of the right CCA, right VA, and right internal jugular vein (IJV) are documented. This projection also documents the high position of the carotid bifurcation (asterisk) at C2. (f), CTA, volume-rendered reconstruction, posterior view, detailing the morphology of the distal right VA duplication, with a “normal” limb at C0–C1 and a smaller additional limb at C1–C2. (g) DSA, left CCA injection, lateral projection, showing the left common (white arrow) and internal (black arrow) carotid arteries. The left internal carotid artery (ICA) ends as the left posterior communicating artery. (h) DSA, right VA injection, posteroanterior projection, showing the right VA (white arrow) and its duplicated segment (black and white arrowheads). The basilar artery (BA) contributes to the supply of both cerebral hemispheres, including the left and right middle cerebral arteries (LMCA, RMCA) and the left and right anterior cerebral arteries (LACA, RACA). (i) CTA, volume-rendered reconstruction, superior-posterior view, detailing the morphology of the intracranial circulation. The BA is connected to the agenetic LICA via the left posterior communicating artery (LPCOM). The BA is also connected to the distal right ICA (RICA) via a prominent right posterior communicating artery (RPCOM), through which it participates to the supply of both hemispheres. The white asterisk indicates an enlarged right A1 segment, with a small proximal aneurysm protruding from its posterior wall, and the black asterisk identifies the anterior communicating artery. There is no connection between the LICA and the left hemispheric circulation (LMCA and LACA).
Further cardiovascular investigations showed no additional heart or large vessels anomalies. The right A1 segment aneurysm was managed conservatively, while the vertigo and lightheadedness subsided after readjustment of the patient’s blood pressure medications.
Discussion
Persistent second cervical ISA
The development of a normal VA originating from the subclavian artery via a persistent sixth cervical ISA is illustrated using the still-unsurpassed scheme proposed by Kemmetmüller (Figure 2(a)).8 This figure also shows an example of aberrant VA origin from the aortic arch via a persistent third cervical ISA.
Figure 2.
Developmental anatomy of vertebral arteries (VA) of normal and aberrant origins (aortic arch scheme based on Kemmetmüller8). (a) Normal developmental pattern, in which the right VA (1) is formed by a succession of anastomoses between the ProA and the first six cervical ISAs. The left side illustrates an aberrant origin of the VA (2) from the aortic arch via a persistent third cervical ISA. (b) The developmental mechanism suspected in the cases published by Parkinson, Chan, Littooy et al.3,4,6 is illustrated on the right side: a normal VA (3) takes off from the right subclavian artery and supplies an isolated right ICA via a persistent second cervical ISA. Note that the variant implicates a partially persistent DC. The left side provides an example of persistent ProA (4). (c) Hypothetical developmental pattern leading to the variant reported by Fazan et al.,5 which included an aberrant origin of the right subclavian artery and a common carotid origin of the right vertebral artery (VA) (5). The latter is not uncommon in association with aberrant subclavian arteries, but the VA then usually derives from a more caudal cervical ISA (fourth or fifth). In this case, the VA is a composite vessel made of the right fourth aortic arch (IV), a persistent portion of the ductus caroticus (DC), and a persistent right second cervical ISA. A persistence of a distal (dotted line) rather than proximal portion of the DC could lead to a persistent second cervical ISA of internal carotid origin. The left VA (6) has a normal origin. (d) Hypothetical developmental pattern leading to the variant described in this report, i.e. a circumflex right arch with separate origin of the left and right CCAs. The proximal segment of the right CCA is made of a persistent right DC, from which arises the right VA (7). The right VA is a persistent second cervical ISA passing through the transverse foramen of C2 (tf-C2). The next branch is the right SCA, followed by the left SCA. The left SCA remains attached to the proximal left CCA by an atretic but not interrupted left arch (dotted line), which completes the vascular ring and keeps the retroesophageal segment in a high position. The left VA has a normal origin (8). Other developmental patterns are conceivable, for example, a regular origin of the right CCA with persistence of a cranial segment of the DC (dotted line in panel c). However, a CCA origin via a completely persistent DC accounts, in addition, for the distance separating the origin of the left and right CCAs (see Figure 1(a)) and for the high carotid bifurcation. A high carotid bifurcation may also result from a partially persistent second aortic arch11 (not shown).
III, IV, VI: 3rd, 4th, and 6th aortic arches; DC: ductus caroticus; ECA: external carotid artery; ICA: internal carotid artery; OCC: occipital bone; PT: pulmonary trunk; SCA: subclavian artery; tf: transverse foramen.
There are possibly five previously documented cases of persistent second cervical ISAs3–7 (Tables 1 and 2). A recent case of “second persistent cervical ISA” is excluded from this review, as the provided images show both VAs passing through C3 transverse foramina (bilateral persistent third ISAs).9 The first observation, published by Tönnis and Schiefer in 1959, was simply labeled as a VA “originating from the common carotid artery”7; this case has since been considered as an example of persistent first cervical ISA, but we believe this interpretation to be erroneous based on the image Tönnis and Schiefer provided. In fact, there is at this time—to our knowledge—no case of persistent first cervical ISA documented.
Table 1.
Published cases of VA originating from the CCA and passing through the C2 transverse foramen (i.e. true persistent cervical ISA with anterograde flow).
| Year | Authors | Authors’ description |
|---|---|---|
| 1959 | Tönnis and Schiefer7 | “VA of CCA origin”. This case is often erroneously quoted as a persistent first cervical ISA. |
| 2004 | Fazan et al.5 | “Unusual origin and course of the right VA” |
| 2020 | Current case | Persistent second cervical ISA |
CCA: common carotid artery; ICA: internal carotid artery; VA: vertebral artery.
Table 2.
Published cases of anastomosis between VA and cervical ICA (i.e. reverse flow in a context of ICA occlusion/agenesis.).
| Year | Authors | Authors’ description |
|---|---|---|
| 1979 | Parkinson et al.4 | “Anastomosis vessels extending from the right VA to the right ICA” |
| 1988 | Littooy et al.6 | “Muscular branch of the cervical VA collateralizing with a posterior branch of the ICA” |
| 2003 | Chan et al.3 | “Anastomotic channel […] located between the cervical portion of the ICA and the cervical portion of the VA” |
CCA: common carotid artery; ICA: internal carotid artery; VA: vertebral artery.
In the cases published by Parkinson et al. in 19794 and by Chan and coauthors in 2003,3 an anastomotic vessel identified as a persistent second cervical ISA took off from a normal right VA to supply an isolated right ICA (Figure 2(b)). The case labeled as “an anomalous branch of the cervical internal carotid artery” by Littooy and colleagues in 19886 was identical to Parkinson’s observation. The images provided in Parkinson’s and Littooy’s cases make it difficult to confirm that the vessel connecting the VA to the ICA was a persistent cervical ISA. Littooy et al. described the anastomotic channel in their patient as a “muscular collateral vessel”.6 Muscular connections involved, for example, in so-called persistent ProA type II or second cervical ISA originating from the external carotid artery are likely not—in our opinion—persistent primitive ISAs, which are never seen to originate from an external carotid component in fetuses, but secondary collateral pathways, possibly established in utero, similar to anastomotic circles seen in adults (e.g., the Bosniak node). This review therefore only considers as true persistent primitive vessels those joining the internal or common carotid arteries to the VA. Whether the VA-ICA connections seen in Parkinson’s and Littooy’s observations were persistent second cervical ISAs or secondary muscular anastomoses remains an open question.
VAs of common carotid origin represent a small fraction of VAs with an aberrant origin: a recent meta-analysis found 41 cases out of 1231 aberrant VAs (3.3%), including a single left-sided occurrence.10 The case described by Fazan et al. in 2004 featured, like Tönnis and Schiefer’s case and the current observation, a right VA derived from a second cervical ISA taking off from the ipsilateral CCA5 (Figure 2(c)). Fazan et al.’s observation was associated, like most VAs of common carotid origin, with an aberrant right subclavian artery (SCA); rare exceptions include patients with a right aortic arch, with double outlet right ventricle, or with no observed or specified coexistent anomaly.
Our observation was associated with a right circumflex aortic arch, i.e. an arch variant made of a right ascending aorta and a left descending aorta connected by a retroesophageal segment (Figure 2(d)). Right circumflex aortas correspond to double aortic arches with left arch atresia; they may be difficult to differentiate radiologically from other arch anomalies, for example, right arches with mirror branching or with an aberrant left subclavian artery.12 The difference lies in the partial or complete regression of a portion of the embryonic left arch: a right circumflex arch forms when the regression is partial (atresia) rather than complete (interruption), leaving a fibrous band that leads to the formation of a vascular ring. In our case, the atretic segment extended between the left common carotid and left subclavian arteries (i.e. a double aortic arch Type II–Subtype 3 of Shuford and colleagues).12
Distal VA duplication
Distal VA duplications are found at the craniocervical junction.13 One limb of the duplication follows a normal course along the first cervical nerve to pass between C0 and C1 (i.e. spinal branch of the ProA), while the other follows the second cervical nerve to enter the intradural space one level lower (i.e. spinal branch of the first cervical ISA). The two limbs join to participate in the formation of the vertebrobasilar junction but can vary in importance; when the normal component (C0-C1) is diminutive or absent, the variant appears as an aberrant course of the VA rather than a duplication (“intradural course of the vertebral system from C2”13). In our case, the “normal” limb remained dominant (Figure 1(f)).
Segmental ICA agenesis
A segmental ICA agenesis is the congenital absence of one or more ICA embryonic segments adequately compensated by collateral pathways,14 which often involve the persistence of embryonic channels. An agenesis (or aplasia) of the cervical ICA, which was present in one prior observation of persistent second cervical ISA3 and suspected in another two,8,10 is consistent with an anomaly of the third aortic arch (segment 1).14 In these three cases, the persistent second cervical ISA played the role of collateral pathway compensating for the ICA agenesis.
The missing segment in our observation was distal rather than proximal, between the take-off of the posterior communicating artery, i.e. the most cranial carotid-basilar anastomosis, and the ICA termination (segment 8).15 It was compensated by collateral pathways including the left and right A1 segments and the anterior communicating artery, and by an enlarged right posterior communicating artery (Figure 1(h) and (i)). The increased risk of intracranial aneurysm development in patients with a segmental ICA agenesis has been well documented. It is likely related to increased hemodynamic stress through the collateral channels.
Possible PHACE syndrome
Several anatomical variants observed in our patient represent major criteria for the diagnosis of PHACE syndrome,16 including the absence, hypoplasia, or aberrant origin or course of a large cerebral artery, a cerebral saccular aneurysm, and an aortic arch anomaly with an aberrant origin of the subclavian artery. In addition, the persistence of an extracranial embryonic artery is considered as a minor criterion for that syndrome. There was, on the other hand, no facial, cervical, or thoracic hemangioma, no structural brain anomalies, and no ocular or midline defects (sternal or pituitary). Our patient thus satisfied the conditions for the diagnosis of possible PHACE syndrome with at least two major criteria in the absence of hemangioma.16,17
Clinical implications
It is unclear whether our patient’s unusual right VA anatomy played a role in her current presentation (i.e. vertigo and lightheadedness) or her prior left hemispheric stroke. The combination of distal left ICA occlusion, proximal left ICA stenosis, and hypoplastic left VA certainly highlights the importance of the blood supplied through her right VA. Shimizu et al. recently reported a case of transient ischemic attack in a case of PHACE syndrome with multiple cerebral and cervical arterial anomalies, and emphasized the risk of ischemic events in these patients, notably in situations potentially leading to hypoperfusion (e.g. excessive blood pressure medication in our case).18 VAs derived from persistent cervical ISAs present an increased risk of arterial dissection during certain neck motion, in particular when they are stretched over the anterior aspect of the transverse process caudal to their level of passage into the transverse canal. Transient flow impairment during neck rotation could also have triggered episodes of vertigo and lightheadedness in our patient. Unfortunately, no suggestive history was recorded during her visit.
From a surgical standpoint, it is important to remember that, while exceptional, variants such as the one reported here can appear as unexpected but critical vessels in the surgical field. A very proximal VA origin from the CCA can also be missed during selective angiography if the catheter tip is placed too far past the VA ostium and, for example, delay a time-sensitive intervention.
Conclusion
Six observations of persistent second cervical ISAs have been published. In three cases (including the present report), an aberrant VA of common carotid origin was seen entering the transverse canal at C2 (“true persistent cervical ISA”). The other three instances consisted in anastomotic connections established between a VA and a proximally occluded ICA. The exact nature of the connection in those cases—i.e. persistent cervical ISA segment versus secondary muscular anastomoses—is difficult to confirm.
Footnotes
Declaration of conflicting interests: The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: P. Gailloud is a consultant for Cerenovus and received research grant money from Siemens Medical.
Funding: The author(s) received no financial support for the research, authorship, and/or publication of this article.
ORCID iD: Philippe Gailloud https://orcid.org/0000-0003-0768-3273
References
- 1.Padget DH. Designation of the embryonic intersegmental arteries in reference to the vertebral artery and subclavian stem. Anat Rec 1954; 119: 349–356. [DOI] [PubMed] [Google Scholar]
- 2.Hochstetter F. Uber die entwicklung der A. Vertebralis beim kaninchen, nebst bemerkungen über die entstehung des ansa vieussenii. Morphol Jahrb 1890; 16: 572–586. [Google Scholar]
- 3.Chan PN, Yu SC, Boet R, et al. A rare congenital anastomosis between the vertebral artery and internal carotid artery. AJNR Am J Neuroradiol 2003; 24: 1885–1886. [PMC free article] [PubMed] [Google Scholar]
- 4.Parkinson D, Reddy V, Ross RT. Congenital anastomosis between the vertebral artery and internal carotid artery in the neck. Case report. J Neurosurg 1979; 51: 697–699. [DOI] [PubMed] [Google Scholar]
- 5.Fazan VP, Caetano AG, Filho OA. Anomalous origin and cervical course of the vertebral artery in the presence of a retroesophageal right subclavian artery. Clin Anat 2004; 17: 354–357. 2004/04/27. [DOI] [PubMed] [Google Scholar]
- 6.Littooy FN, Baker WH, Field TC, et al. Anomalous branches of the cervical internal carotid artery: two cases of clinical importance. J Vasc Surg 1988; 8: 634–637. [DOI] [PubMed] [Google Scholar]
- 7.Tönnis W, Schiefer W. Anatomie der hirngefäße In: Tönnis W, Schiefer W. (eds) Zirkulationsstörungen des gehirns im serienangiogramm. Berlin, Heidelberg: Springer, 1959, pp.3–49. [Google Scholar]
- 8.Kemmetmüller H. Über eine seltene varietät der art. Vertebralis. Anatomische Hefte 1911; 44: 305–360. [Google Scholar]
- 9.Sturda C, Steyn C, Olivi A, et al. Extraforaminal vertebral artery until C2 transverse foramen in down syndrome patient affected by atlantoaxial subluxation: first observation and review of literature. World Neurosurg 2019; 131: 230–233. [DOI] [PubMed] [Google Scholar]
- 10.Yuan SM. Aberrant origin of vertebral artery and its clinical implications. Braz J Cardiovasc Surg 2016; 31: 52–59. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Suzuki T, Moriyama T, Moriwaki H, et al. Anomalous artery directly connecting the external and internal carotid arteries. Annals Anatomy 2000; 182: 59–63. [DOI] [PubMed] [Google Scholar]
- 12.Shuford WH, Sybers RG, Weens HS. The angiographic features of double aortic arch. Am J Roentgenol Radium Ther Nucl Med 1972; 116: 125–140. [DOI] [PubMed] [Google Scholar]
- 13.Lasjaunias P, Braun JP, Hasso AN, et al. True and false fenestration of the vertebral artery. J Neuroradiol 1980; 7: 157–166. [PubMed] [Google Scholar]
- 14.Lasjaunias P, Santoyo-Vazquez A. Segmental agenesis of the internal carotid artery: angiographic aspects with embryological discussion. Anat Clin 1984; 6: 133–141. [DOI] [PubMed] [Google Scholar]
- 15.Gailloud P, Clatterbuck RE, Fasel JH, et al. Segmental agenesis of the internal carotid artery distal to the posterior communicating artery leading to the definition of a new embryologic segment. AJNR Am J Neuroradiol 2004; 25: 1189–1193. [PMC free article] [PubMed] [Google Scholar]
- 16.Metry D, Heyer G, Hess C, et al. ; PHACE Syndrome Research Conference, Consensus statement on diagnostic criteria for PHACE syndrome. Pediatrics 2009; 124: 1447–1456. [DOI] [PubMed] [Google Scholar]
- 17.Nabatian AS, Milgraum SS, Hess CP, et al. PHACE without face? Infantile hemangiomas of the upper body region with minimal or absent facial hemangiomas and associated structural malformations. Pediatr Dermatol 2011; 28: 235–241. [DOI] [PubMed] [Google Scholar]
- 18.Shimizu Y, Tsuchiya K, Fujisawa H. Transient ischemic attack in elderly patient with PHACE syndrome. Surg Neurol Int 2019; 10: 188–123. [DOI] [PMC free article] [PubMed] [Google Scholar]


