We read with interest Gailloud et al’s “Segmental Agenesis of the Internal Carotid Artery Distal to the Posterior Communicating Artery Leading to the Definition of a New Embryologic Segment.”1 In the article, the authors describe a case of segmental agenesis of the internal carotid artery (ICA) distal to the origin of the posterior communicating artery (PcomA) that was well documented by angiography and surgical inspection secondary to an associated anterior cerebral artery aneurysm.
The ICA is constituted by a number of successive embryologically distinct segments, each of them located between embryonic arteries or their remnants. Each of these segments can be absent, representing as a focal agenesis. In the ICA developmental anatomy proposed by Lasjaunias and Santoyo-Vazquez,2 the ICA ends with the bifurcation into a rostral branch (from which in adult life the ICA distal to the PcomA, the anterior choroidal [AchoA], the anterior cerebral [ACA], and the middle cerebral artery [MCA] are derived) and a caudal branch from which the PcomA, parts of the basilar artery (BA), and the posterior cerebral artery (PCA) are derived in later life. This concept is challenged by the authors taking their case of a segmental agenesis of the ICA distal to the PComA into account. They regard the caudal division of the ICA no longer as a terminal branch of the ICA but instead argue that the PcomA is simply another embryonic vessel bridging the anterior and posterior circulation, being the most cranial of the carotid-basilar anastomoses. From this perspective, a new segment (the eighth segment) of the ICA distal to the PComA has to be defined that ends with the bifurcation into MCA and ACA.
We, on the other hand, argue that the ICA terminates with the bifurcation into a caudal and rostral branch and that, therefore, no eighth segment of the ICA can be present as such. What, on first sight seems to be just a problem of nomenclature (ie, why can’t we simply call the “rostral branch of the ICA” the “eighth segment of the ICA”) is, when phylogeny and embryology are taken into account, a misnomer that creates a misunderstanding.
The MCA is a recent phylogenetic acquisition and must be considered as a collateral branch of the ACA.3 The ACA and AchoA are phylogenetically old vessels, forerunners of which are present in fish. The MCA appears as late as in reptiles, though not as a single trunk but instead as a series of small anastomosed vessels arising from the olfactory artery as the forerunner of the ACA. It continues to evolve in mammals and primates to finally become the single stem that we know. In most species, the caudal branch supplies the posterior fossa.
When considering the embryology of the human ICA development, further contradictions against the hypothesis that the ICA ends with the bifurcation into MCA and ACA can be found. The embryonic period is characterized by the shaping of the rostral extremity of the neural tube and, with it, the simultaneous shaping of the arterial tree. At the end of the fifth week (the prechoroidal stage), the ICA ends with a rostral and a caudal division. The caudal branch reaches the cephalic end of the ipsilateral ventral neural artery to constitute the PcomA, the so-called P1 segment and the upper half of the BA. This leads to a regression of pre-existing transient carotid-basilar anastomoses—ie, the trigeminal and hypoglossal arteries.4 The caudal division, therefore, supplies the diencephalon, the mes-, and metencephalon. The rostral division, on the other hand, may be called the telencephalic branch. It subdivides further into the ACA and AchoA that both encircle the neck of the telencephalic vessel and anastomose with each other to form a ring. Lateral branches of the AchoA will later become the telencephalic portion of the PCA (ie, the so-called P2, P3, and P4 segments). Similarly, lateral branches of the pericerebral network of the hemispheres supplied by the ACA will become, at the end of the choroidal stage (ie, during the seventh and eighth week) the future MCA. Because the MCA, therefore, has to be regarded from a morphogenetic point of view as a branch of the ACA, there cannot be a bifurcation of the ICA into MCA and ACA, because, if it was a bifurcation, both vessels should appear at the same time in phylogeny and embryology.
What is called the ICA bifurcation in the manuscript is, therefore, the termination of the rostral division of the ICA, whereas the termination of the ICA is in fact located at the level of the PcomA regardless of its size. No additional segment of the ICA distal to this point, named as such, can be envisaged.
Reply:
We are grateful to Krings and Lasjaunias for their comment. It gives us an opportunity to re-emphasize a point central to our recent publication describing a case of segmental agenesis of a portion of the internal carotid artery (ICA) located distal to the origin of the posterior communicating artery (PcomA).1 In our report, the absence of the involved segment was unequivocally documented by angiography and direct surgical observation.
It is often assumed that the embryonic ICA originally bifurcates into a rostral branch and a caudal branch. Following this view, principally based on the outstanding work of Padget,2 the rostral branch is the precursor of the anterior cerebral artery (ACA) and its secondary branches, including the middle cerebral artery (MCA) and the anterior choroidal artery (AchoA), while the caudal branch corresponds to the posterior cerebral artery (PCA), including a proximal portion that later becomes known as the PcomA. In their theory on the segmental development of the ICA, Lasjaunias and Santoyo-Vazquez3 use this assumption to set the distal limit of the embryonic ICA segments at the PcomA, the adult equivalent of the embryonic caudal division. According to this perspective, as they state in their letter, no segment can be defined distal to the PcomA, because it would then belong to the rostral branch of the fetal ICA, not to the ICA per se.
More recent investigations, however, offer a convincing alternate developmental scenario for the distal ICA. Van Overbeeke et al4 looked at the relative role of the PcomA and the proximal PCA (P1 segment) in the formation of the circle of Willis in human fetuses and infants from 12 to 60 weeks of age. These authors defined 3 patterns of blood supply to the PCA territory: (1) an adult configuration, in which the P1 segment is the dominant source of blood supply, (2) a transitional configuration, in which the P1 segment and the PcomA are equivalent in size, and (3), a fetal or embryonic configuration, in which the PcomA is the dominant source of blood. Note that the name given to the 3 configurations is derived from the developmental pattern expected from Padget’s conception of the distal ICA development—ie, a progressive transfer of the blood supply for the PCA territory from the ICA (fetal or embryonic configuration) to the basilar artery (BA; adult configuration), with an intermediate state of equivalent contribution (transitional configuration). The actual observations made by Van Overbeeke et al showed, however, a different pattern of development, with a largely dominant transitional configuration during the early fetal stages, which then progressively regresses in favor of either the adult or fetal/embryonic configurations. These findings strongly suggest that, in fact, the so-called caudal branch of the fetal ICA behaves as a carotid-basilar anatomosis, not as a dominant branch later annexed by the posterior circulation. In a superb publication dealing with the development of the PCA in the rat, Moffat5 had already shown that the future territory of the PCA was, in the embryo, vascularized by the AchoA and not by the PcomA as initially believed, and later transferred to the cephalic end of the ipsilateral longitudinal neural artery (the future BA). This finding again suggests that the PcomA is, indeed, but the most cranial of the carotid-basilar anastomoses. If the concept of the ICA bifurcation into cranial and caudal branches is rejected, the assumption that the ICA terminates at the PcomA level becomes arbitrary, as does the rebuttal of a new embryonic segment based solely on that assumption.
We agree with Krings and Lasjaunias that the new segment documented in our publication is not located between the PcomA and the fetal ICA termination. Most likely, the distal end of this new segment is the AchoA, which is, as mentioned, a phylogenetically older vessel that plays a prominent role in the early development of the PCA. Unfortunately, the AchoA was visible neither angiographically nor during surgical exploration in our patient.
In summary, we have documented a segmental agenesis involving a previously unrecognized segment located distal to the origin of the PcomA.1 This anatomic and angiographic fact does not fit in Lasjaunias and Santoyo-Vazquez theory of segmental development of the ICA as presently stated, but we think that it can be used, when combined with modern insight of the distal ICA development, to complement this theory.
- Copyright © American Society of Neuroradiology