Basilar artery origin of an orbital artery – A rare variant and review of the embryology of the orbital arterial supply

Abstract

We present a patient with separation of the arterial supply to the globe and the extra-ocular muscles. The ophthalmic artery originates from the typical adult location and supplies only the globe. Arising from the basilar artery was a branch that supplies the extra-ocular muscles. There was no apparent connection between these vessels around the optic nerve and no evidence of supply from the external carotid artery. We discuss the embryology of the ophthalmic artery from the point of view of Padget and Lasjaunias and offer our opinion on the on-going controversy. We believe this is the first case to highlight the trigeminal-primitive maxillary-stapedial anastamotic pathway.

Introduction

The adult ophthalmic artery represents the end result of a complex series of anastomoses between a variety of vessels during embryogenesis that begins at the 4-8mm stage of the embryo. The original description of the embryonic steps was provided by Dorcas Padget in her seminal work based on anatomical studies.¹ An alternative theory was put forward more recently by Lasjaunias² based mainly on angiographic findings. A wide variety of anatomical variants to the ophthalmic artery have been described with one of the rarest types being the basilar artery origin.^3–5

Here we describe a rare variant of an orbital artery derived from the basilar artery in the typical location of a persistent trigeminal artery with the ocular artery supply from the ophthalmic artery that originates from the normal location. We discuss the embryology of the ophthalmic artery and discuss the differing views of Padget and Lasjaunias and how this case may help to resolve the on-going controversy.^6,7

Case report

A 49-year-old male patient was referred to our neurovascular unit after magnetic resonance imaging (MRI) of the brain for headaches revealed an incidental aneurysm arising from the ophthalmic segment of the left internal carotid artery (ICA). His past medical history, aside from a previous diagnosis of Lyme’s disease, was unremarkable. His neurological examination was normal.

After discussion in the multi-disciplinary team (MDT) meeting digital subtraction angiography (DSA) was performed and this revealed a para-ophthalmic aneurysm that measured approximately 5.2 mm in maximum dome width. Angiography of the contralateral ICA revealed a further mirror aneurysm on the right ICA. Both of these aneurysms were subsequently treated with p64 flow diverters (phenox, Bochum, Germany). It was noted at the time of the angiography that, a hypoplastic ophthalmic artery could be seen arising from the base of the aneurysm. This artery could be seen to split into two divisions with the choroidal blush arising from this vessel and the divisions likely to represent the medial and lateral posterior ciliary arteries (Figure 1(a) and (b)). There was no evidence that the larger division of the hypoplastic ophthalmic artery supplied the extra-ocular structures. Left vertebral angiography demonstrated an artery originating from the basilar artery between the anterior inferior cerebellar artery (AICA) and superior cerebellar artery (SCA) that coursed towards the orbit on the right (Figure 1(c) and (d)). This artery travelled initially laterally before sharply turning anteriorly and running anterolateral to the ICA in the inferior aspect of the cavernous sinus before entering the orbit via the superior orbital fissure (SOF). The intra-orbital portion of the vessel could be seen resting upon the superior aspect of the optic nerve before dividing into superior and inferior divisions, medial to the optic nerve, both of which appeared to supply the extra-orbital structures (Figures 2 to 4). It is important to note that a lacrimal blush could not be seen on the angiography however, the juxtaposition of the branches of this artery lead us to believe it did supply the extra-ocular muscles and remaining structures. Angiography of the external carotid artery (Figure 5) did not show any evidence of supplying the orbital structures on the right (Figure 5(b)) or the left (Figure 5(c)).

Figure 1.

Angiography of the internal carotid artery (a and b) in early and late arterial phase showing a hypoplastic ophthalmic artery that originates from the ‘typical’ anatomical location. The artery can be seen to divide and this was thought to represent the division into the medial and lateral posterior ciliary arteries. A prominent choroidal blush can be seen. Angiography of the basilar artery demonstrated an anteriorly directed artery that could be seen to travel towards the orbit (c and d) that supplied the extraocular muscles. There was no evidence of a choroidal blush from this vessel suggesting it did not provide arterial supply to the globe itself.

Figure 2.

Contrast enhanced dyna CT with arterial injection of the basilar artery with axial and coronal images through the orbit (a and b) and sphenoid sinuses (c and d). The basilar orbital artery can be seen to enter the orbit through the superior orbital fissure and then course over the optic nerve medially (a and b). The origin of the vessel can be seen from the basilar artery and it can be seen to travel laterally before turning sharply anteriorly (large white arrow) before travelling inferolateral to the ICA (clearly seen because of the implanted p64 flow diverter, white star). (SOF – superior orbital foramen, SS – sphenoid sinus, BA – basilar artery).

Figure 3.

Contrast enhanced dyna CT with arterial injection of the basilar artery and coronal sections through the orbit. The artery (white arrow) can be seen superior to the optic nerve before travelling medially and then dividing to supply the extra-ocular muscles. (so = superior oblique muscle, io = inferior oblique muscle, lr = lateral rectus muscle, mr = medial rectus muscle, sr = superior rectus muscle, ir = inferior rectus muscle, lps = levator palpebrae superior muscle, lg = lacrimal gland).

Figure 4.

Maximum intensity projection (MIP) with sagittal reconstruction from the dyna CT with contrast injection into the vertebrobasilar system. The artery (white arrows) can be clearly seen to pass through the superior aspect of the superior orbital fissure (black arrows) and resting atop the optic nerve as was seen on the axial imaging. The small white arrows point to the posterior markers of the p64 flow diverter in the ICA.

Figure 5.

Lateral projection of right ECA injection showing zygomatic branch of superficial temporal artery (STA) proceeding to the orbit (a, black arrows). The middle meningeal artery can also be clearly delineated (a, white arrows). An anterior-posterior projection after selective right-sided STA injection demonstrates the prominent zygomatic branch supplying the upper eyelid (b, black arrows) however, a lacrimal blush could not be seen. An anterior-posterior projection after injection of the left ECA did not reveal opacification of the orbital structures (c).

Discussion

When considering the anatomy and embryology of the ophthalmic artery it is important to first consider the seminal work of Padget. According to her work several arteries are believed to contribute to the classical anatomical disposition namely the primitive maxillary artery (PMA), the primitive ventral ophthalmic artery (PVOA), the primitive dorsal ophthalmic artery (PDOA), the stapedial artery and possibly the primitive olfactory artery (POlfA). In the 4–5 mm (crown-rump length) embryo, the blood supply is highly plexiform in nature and the PMA and POlfA encircle the optic vesicle. The lateral branch of the PMA supplies the vesicle whereas its medial branch supplies the Rathke’s pouch region. This vessel starts to regress in the 7 12–mm embryos. The PMA arises from the ICA caudal to the origin of the OA. In the adult the remnant of the lateral branch of the PMA may persist and become the anastomosis between the deep recurrent OA and the inferolateral trunk. The POlfA, a branch of the cranial division of the ICA, gives rise to the cortical branches of the future anterior cerebral artery (ACA) and transiently provides a number of small vessels to the network of arteries at the base of the optic vesicle. In the original drawings of Padget a small ‘budding’ PDOA can be seen. At the 5–6 mm stage the PDOA can be seen more clearly arising from the junction of the cranial and caudal divisions of the primitive ICA opposite to where the future posterior communicating artery will arise. The PVOA emerges very high opposite the origin of the anterior choroidal artery. It first appears in 9 mm embryos and supplies the common nasal ciliary artery (future medial ciliary artery) to the adult OA whereas the PDOA branches to form a common temporal ciliary artery (future lateral ciliary artery) and the hyaloid artery (future central retinal artery). According to Padget the site of origin of the OA as is seen in adults results from a caudal migration of the PDOA along the ICA, which Padget suggests is due to a process of anastomotic progression, and this process occurs during the 16-18mm stage. In the 20 mm embryo the stapedial artery contributes to the orbital blood supply via its supra-orbital branch that enters the orbit via the superior orbital fissure (SOF) so that the optic nerve is surrounded by a ring that is formed from the three different supplies – the stapedial artery, the PDOA and PVOA. The stapedial artery arises as a branch of the hyoid artery which itself develops at the 9 mm stage the remnant of which remains as the caroticotympanic trunk. The stapedial artery grows to supply the non-neural structures of the developing orbit and face and it divides into a superior and inferior branch. The superior branch, termed the supra-orbital artery, supplies dura and the non-neural orbital structures with the artery accompanying the ophthalmic division of the trigeminal nerve through the SOF.

Therefore, according to this theory the POlfA can account for the rare variants seen to arise from the ACA in the adult, the lateral branch of the primitive maxillary artery accounts for the cavernous origin of the OA and the persistent PVOA and PDOA account for duplicated arteries.⁸

According to Lasjaunias et al, whose description is based on angiography, there are two ophthalmic arteries – the ventral and dorsal ophthalmic arteries (VOA and DOA). The VOA is the true ocular artery and originates from the anterior cerebral artery passing through the optic canal. The DOA originates from the ICA at the carotid siphon and passes through the superior orbital fissure (SOF). At some point two anastomoses are formed, one near the optic nerve between the VOA and the DOA and another near the intradural optic canal between the VOA and the ICA. The proximal part of the VOA and the DOA regress with the resultant adult configuration. According to this theory the remnant of the DOA is the inferolateral trunk (ILT). The supraorbital artery, a branch of the stapedial artery, passes through the SOF and contributes supply to the orbit e.g. the ethmoido-nasal and lacrimal arteries.

Given these two differing opinions we must first try to elucidate which we believe to be true prior to an exploration on the anatomy and embryology of our case. Although it has been suggested that the ILT and its deep recurrent ophthalmic artery branch are remnants of the PDOA this is likely to be an incorrect observation. In Padgett’s work there is no evidence of an ophthalmic artery arising from the cavernous portion of the ICA. The ILT is developmentally linked to the primitive maxillary artery. The arteria anastomotica in dogs connects the cavernous ICA to the base of the external ophthalmic artery, which is a branch of the orbital artery in canines and this has been used to support the concept of a cavernous ophthalmic in humans.⁹ However, Netsky and de la Torre¹⁰ have shown that the this artery is related to the primitive maxillary artery (MA) in agreement with Padget who located the trunk of the primitive MA at the expected location of the ILT, just lateral to the inferior hypohyseal artery (the adult derivative of the medial branch of the primitive MA). In Padget’s original work the division of the primitive maxillary artery into branches supplying the developing optic vesicle and Rathke’s pouch are clearly seen in Figures 3(a) and (b). Similarly, she went on to state ‘Also seen in some embryos is the small primitive maxillary artery, which is now directed more mesially than ventrally toward the dwindling craniopharyngeal canal along the caudal border of the hypophysis; this becomes a component of the inferior hypophyseal artery’. The lateral branch of the primitive MA developing into the deep recurrent ophthalmic artery was supported by the findings of De La Torre and Netsky. In their anatomical study of 6 human foetuses aged between 5-9months they described the inferior hypophyseal artery that was directed medially toward the posterior part of the sella turcica and that anastomosed with its contralateral counter part. At the approximately the same level they saw a large lateral branch of the cavernous ICA supplying the adjacent dura mater, orbital structures and semilunar ganglion and that anastomosed with the middle and lesser (accessory) meningeal systems. The ophthalmic artery arising from the cavernous ICA and entering into the orbit via the medial aspect of the superior orbital fissure represents a dominant anastomosis between the ILT and the ophthalmic artery via the deep recurrent ophthalmic artery, therefore corresponding to the persistence of the lateral branch of the primitive MA.

Upon careful review of the seminal work by Daniel et al.¹¹ it can be seen that the ILT and MHT are homologues of the carotid rete that is present in other mammals and is particularly well developed in artiodactyls such as sheep, goats, and oxen as well as pigs. Studying the supply to the carotid rete across the various different species it is apparent that the anterior branch of the ILT a homologue of the arteria anastomotica, which has branches that pass through the superior orbital fissure and the foramen rotundum whereas the ramus anastomotica gives branches that pass through the foramen ovale and foramen spinosum. Therefore, we believe that the ILT represents a fusion of these two vessels. Similarly the MHT can be thought of as a homologue of the posterior aspect of the carotid rete that is best seen in the pig and largely supplied by the ascending pharyngeal artery. Given that these vessels may represent the homologues of the carotid rete and the distribution of these vessels are the homologues of the differential supplies e.g. arteria anastomotica, ramus anastomotica, ascending pharyngeal etc. it is possible that the MHT and ILT actually share a common origin and certainly a single trunk has been described and is actually not uncommon.¹² The typical MHT may represent the more medial and posterior aspect and the ILT the more anterior and lateral aspect of a common originator vessel that is likely to be the primitive maxillary artery. The separation of the vessels over time may occur due to a differential in the growth of the brain and face in a similar way to the anterior movement of the OA as described by Padget. The anatomical pathway to explain the findings in our case is simply then the well established anastomosis between a persistent trigeminal to the ILT resulting in a trigeminal-primitive maxillary-stapedial variant in our case. The persistent trigeminal remnant as seen in our case represents the lateral type (Saltzman type 2).^13,14 To our knowledge the basilar origin of an ophthalmic artery has been reported infrequently in the past^3–5 with the first case related to an orbital AVM. Two theories have been proposed to explain this variant anatomy. As explained by Sade et al. this configuration could arise from an anastomosis between the vertebrobasilar system and the stapedial artery however, such a connection is not known to occur in humans. Lasjaunias proposed that this configuration could result from a persistent connection between primitive trigeminal artery and the PDOA (Lasjaunias theory)/primitive maxillary artery (Padget theory). Our case represents an anastomosis between the primitive trigeminal, ILT (remnant of the primitive maxillary) and the supra-orbital branch of stapedial artery which has not been reported previously.

In addition our case lacks the arterial ring that normally surrounds the optic nerve and consists of supply from the primitive ventral and dorsal ophthalmic arteries as well as the stapedial artery.¹⁵ In our case the normal anatomical position of the true ocular artery is seen and demonstrated by the choroidal blush whereas the supply to the orbital structures is provided by a separate vessel that enters the orbit via the superior orbital fissure and does not anastomose with the ophthalmic artery as would typically be seen. This confirms that this vessel represents a remnant of the stapedial system. It is unclear why this may be the case and why two abnormalities have occurred and certainly it may be the case that they are causally linked. As only an ocular supply is seen arising from the ophthalmic artery and with the clear division of this artery into the medial and lateral posterior ciliary arteries, with the central retinal artery not visualised but also likely arising from this vessel, this vessel can be said to be the remnant of the process described by Padget. The complete disconnection between orbital and ocular supply in our case and the entry of the basilar orbital supply through the superior orbital fissure helps to confirm this although our case does not enable the theory proposed by Lasjaunias to be completely disqualified.

Conclusion

We report an extremely rare anatomical variant of a basilar artery supplying the orbital structures but with the true ocular supply originating from the ophthalmic artery that is located in its normal position.

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: PB, MA are consultants for phenox. HH is co-founder of phenox. The other authors report no conflicts of interest.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Pervinder Bhogal https://orcid.org/0000-0002-5514-5237

Stefan Schob https://orcid.org/0000-0003-2846-5443

Muhammad AlMatter https://orcid.org/0000-0002-6405-656X

Hans Henkes https://orcid.org/0000-0002-6534-036X