Localized Marked Elongation of the Distal Internal Carotid Artery with or without PHACE Syndrome: Segmental Dolichoectasia of the Distal Internal Carotid Artery

Embryologic Pathogenesis of ICDE of the Distal ICA

During the embryologic period of cerebral artery formation, angiogenesis is mainly driven by hypoxia and related growth factors in the target tissue.^11,12 We hypothesized that exposure of a specific segment of the intracranial artery to a vasculogenetic trigger resulted in arterial lengthening. According to the ICA developmental anatomy proposed by Lasjaunias and Santoyo-Vazquez,¹³ the ICA branches into a cranial and caudal rami at stages I and II (3.5–4 weeks); the former gives rise to the anterior cerebral artery and AchoA, whereas the latter gives rise to the PcomA, P1 segment, and upper BA. In this study, dolichoectasia at the anterior (A1 segment, AchoA, and CS segment) and posterior (PcomA, P1 segment, and upper BA) divisions of the ICA suggested that these regions were triggered during embryonic development. We observed no involvement of the superior cerebellar artery in our cases (data not shown); because the superior cerebellar artery existed before ICA branching, the vasculogenetic event may have occurred after the appearance of superior cerebellar artery.¹³ Although the remaining PCA (segments P2–P4) comprises the posterior choroidal branch of the caudal ramus at stage V, the involvement of the P2 segment in several cases suggests that the PCA precursor had also been affected.¹³

PHACE Syndrome

Cerebral arterial anomalies are observed in 91% of patients with PHACE syndrome,¹⁴ and previous studies of PHACE syndrome have reported a presentation of dolichoectasia of the internal carotid arteries similar to that observed in our cases.^2,4,15,16 Therefore, we searched for common features between our cases and PHACE syndrome cases.

First, we observed a strong female predominance in our patient group in agreement with previous studies of PHACE syndrome (up to 8:1).^15,17,18 Second, the timing of PHACE syndrome was consistent with our speculated time course. Several studies of PHACE syndrome have reported that the teratogenic influence might occur from gestational weeks 3 to 5.5,^8,14 concurrent with the regression of the embryonic capillary bed and active stemming of the craniocervical vasculature. Therefore, any influences on these 2 processes may result in a cutaneous hemangioma and trigeminal artery persistence.¹⁹

Third, several vascular anomalies have been reported in both patients with PHACE syndrome and in our patient group. In a previous study, A1 hypoplasia was reported as an intracranial anomaly affecting 8 of 12 patients with PHACE syndrome; this is similar to the findings of our study (15/20).²⁰ ICDE of the ICA was accompanied by dolichoectasia of the posterior circulation in 6 of 7 patients with PHACE syndrome in a previous study.²⁰ An aberrant origin or course of the principal cerebral arteries, a major or minor PHACE syndrome criterion, was observed in 9 of 20 patients (45%; 5 ectopic ophthalmic arteries and 4 other arteries).

Fourth, 4 of the 20 patients in this study met the diagnostic criteria of PHACE syndrome, and an additional 7 patients were classified as possible cases of PHACE syndrome. In addition, some patients exhibited ipsilateral midbrain hypoplasia. Because a lack of direct contact with the adjacent vessel does not support compression-induced midbrain deformation (Fig 2E), we suspected that dolichoectasia of the arteries feeding the midbrain (BA, P1, or P2) mildly altered the blood supply and caused further hypoplasia. In 1 patient with PHACE (patient 10), hypoplasia of the ipsilateral cerebellum and the ipsilateral midbrain was found to coexist, suggesting that these 2 structural anomalies shared a common origin. Therefore, if a smaller midbrain was defined as a posterior fossa anomaly associated with PHACE syndrome, an additional 4 patients in our study would meet the criteria for possible PHACE syndrome. Such lesions might broaden the PHACE syndrome phenotype. According to a study by Heyer et al,²⁰ moderate effacement of the right pons (Fig 2C in the article by Heyer et al) and cerebral peduncle along with corresponding vascular anomalies was observed via MR imaging in a patient with PHACE syndrome. However, the author did not propose this finding as an anomaly.²⁰

In our study, we found no significant differences in arterial component dysplasia and brain structures between patients positive and negative for PHACE (Table 2), suggesting that the 2 groups of patients share the same features and pathogenesis. Furthermore, although we did not detect an obvious cutaneous hemangioma in many of our patients, a previous study found that very small cutaneous hemangiomas might be absent or regress spontaneously without prior recognition or reporting.⁴ The above-mentioned aspects raise the intriguing possibility that a marked ICDE of the ICA might indicate an otherwise-unrecognized partial phenotypic expression of PHACE syndrome. However, the spontaneous regression of cutaneous hemangiomas at an early age may cause the underestimation of the incidence of PHACE syndrome in this group of patients, which further induces underestimation of the relationship between the ICDE and PHACE syndrome.

Acquired Changes in the Involved Arterial Wall

Normally, an abnormal mural angiogenesis likely causes an increase in the luminal caliber because the correct remodeling signals induce apoptosis of the unnecessary vessel wall components. A lack of remodeling causes centripetal and longitudinal proliferation and luminal reduction; however, this might also cause ectasias, elongated arteries, and aneurysms.¹³ Although many cases involving ICDE of the ICA, with or without PHACE syndrome, have been reported, little is known about the evolution of vessel wall lesions and relevant complications. Results from a follow-up study of vessel lesions may provide valuable prognostic information.

Progressive cerebral arterial stenosis and arterial occlusion and a Moyamoya-like vasculopathy leading to stroke have been described in infants with PHACE syndrome.^2,20 This progressive cerebral vasculopathy corresponds with the proliferative phase of hemangioma growth, and as a result, the average age of experiencing stroke among patients with PHACE syndrome is 8.8 months.²¹ However, no ischemic stroke events were reported in our present study, even among patients with a Moyamoya-like vasculopathy (patients 4, 8, 12). Notably, we observed vessel wall enhancement more frequently among young patients than older patients, indicating a regression in inflammation with aging. Accordingly, we speculate that most arterial stenoses and occlusions formed within a short time during the prenatal or infant stage. Patients with mild lesions might pass through that period asymptomatically, and adult PHACE syndrome diagnoses may be incidental.^1,22 Due to the long-standing nature of the lesion, there is a good chance of good collateral circulation formation secondary to arterial stenoses, which may present as nonsymptomatic steno-occlusive disease.

We further propose that ICDE of the distal ICA, with or without PHACE syndrome, might stabilize after a period of rapid progression. Bracken et al¹⁵ followed up several cases of PHACE in neurodevelopmentally healthy patients for 1–12 years. McLaughlin et al⁵ reported a 24-year-old female patient with a pure arterial malformation involving the distal ICA, PcomA, and PCA that was found on a CT scan obtained to determine the cause of a headache. When this patient was later followed up at 54 years of age, the abnormal vessels had not changed with time on MR images, and no symptoms relevant to the abnormal vessels were reported during the 31-year interval.⁵ Similarly, our review of imaging data collected during a long follow-up demonstrated a slow evolution of vessel wall enhancement (patient 1) and slow progression of both calcification and stenosis (patient 9). The vessel wall enhancement in patient 1 could be explained by the immature nature of the affected vessel wall, which may increase the permeability of the endothelium, with contrast leakage from the lumen into the arterial wall, and may be simultaneously associated with an atherosclerotic-like process in the dysplastic segments.

Treatment and Follow-Up

Many cases of ICDE with tortuous ICAs were identified incidentally, without relevant symptoms^5,7,15; in these cases, the lesions appeared stable on follow-up images and the patients did not receive medical treatment. However, several reports and our observations suggested the need for regular imaging follow-up as well as medication in some cases. However, no specific treatment exists for dolichoectasia, and the surgical and medical therapies used to treat this condition have not been systematically evaluated. Although anticoagulation and antiplatelet therapies might help in preventing an ischemic episode, some studies have indicated that aspirin and warfarin or both do not effectively reduce the stroke recurrence rates in patients with dolichoectasia and might increase the risk of hemorrhage in this population.²³ However, we note that these previous data were all with respect to the BAs.

Treatment for PHACE syndrome should address the aforementioned symptoms.² Corticosteroids and interferon have been previously used to treat hemangiomas associated with PHACE syndrome; however, their efficacy in the treatment of acute-phase vessel wall inflammation remains unknown.¹⁶ Occasionally, a pial synangiosis procedure has been suggested for severe stenosis or occlusion of the distal ICA.²⁰

We believe that attention should be paid to several cases in this study. One patient (patient 1) exhibited simultaneous vessel wall enhancement and calcification at 7 years of age, leading to our hypothesis that the affected vessel wall was prone to atherosclerosis formation and secondary calcification. Questions also remain regarding the use of antiatherosclerosis therapies in young patients. Another patient (patient 11, Fig 5) exhibited vessel wall enhancement in an aneurysm and its parent artery, which may be a risk factor for aneurysm rupture; accordingly, a pre-emptive aneurysm embolization was performed.²⁴ In another patient (patient 16), asymptomatic ICDE of the right ICA and hypoplasia of the right A1 segment were detected at 54 years of age, and a blood flow–related aneurysm of the left anterior communicating artery was observed at 69 years of age. This patient was later treated with coiling embolization. Therefore, we suggested a follow-up comprising regular angiography studies (CTA or MRA) to demonstrate overall luminal changes and, if possible, vessel wall imaging to detect inflammation in the lesion.

Limitations

This study had several limitations. First, we found it difficult to objectively define “segmental dolichoectasia.” To overcome this problem, we included only cases with noncontroversial elongation and unusual tortuosity relative to other segments or the contralateral ICA. Accordingly, we might have skipped many mild elongation cases and underestimated the number of relevant cases. Second, the definition of segmental arterial tortuosity is rather subjective. Lasjaunias et al⁹ defined the ICA segments according to embryogenic evolution. In this study, we considered ICDE of the distal ICA to be a congenital disease that may occur segmentally. Although we used this system to describe the observed lesions, we were unable to conclude that the lesions could be attributed to a similar congenital origin. Third, 2 patients in our study had bilateral ICA involvement, which has also been reported in patients with PHACE syndrome. However, a satisfactory interpretation of the bilateral pathogenesis could not be attained.²⁰