Persistent Trigeminal Artery: A Novel Imaging Finding in CHARGE Syndrome

CHARGE Syndrome

CHARGE syndrome (or Hall-Hittner syndrome) was first described as an association of congenital anomalies in 1979.^1,2 The acronym CHARGE is based on the cardinal features identified when the syndrome was delineated: Coloboma, Heart malformation, choanal Atresia, Retardation of growth and/or development, Genital and/or urinary anomalies, and Ear anomalies.³ Diagnosis is based on clinical criteria, which have been updated across time, with emphasis on the 3C triad (Coloboma, Choanal atresia, and abnormal semicircular Canals).^4,5 It was subsequently labeled as the CHARGE syndrome because the association of multiple anomalies was pathogenetically related following the identification of the CHD7 gene.⁷ Since that time, several new clinical findings have been added to the clinical spectrum of CHARGE syndrome, such as dysmorphic features, rhombencephalic dysfunction, cranial nerve dysfunction, hypoplasia of the semicircular canals, olfactory hypoplasia and/or arhinencephaly, feeding difficulties, esophageal anomalies, facial clefts, and hypothalamic-hypophyseal dysfunction.^{8⇓⇓⇓⇓-13} The clinical spectrum of CHD7 mutations and CHARGE continues to expand, and recent literature has focused on the presence of venous and skull base anomalies.^14,15

Genetically, CHARGE syndrome is a monogenic autosomal dominant disorder with an incidence of approximately 1:8500 to 1:12,500 in North America.^5,12,16 It results from a dysblastogenetic and dysneurulative process and could be related by a common pathogenetic mechanism resulting in disturbed neural crest development.¹² Mutations in the gene CHD7 (in 8q12) were identified as causative for CHARGE syndrome in most (approximately two-thirds) patients with a clinical diagnosis of CHARGE syndrome.⁷ CHD7 belongs to a large family of evolutionarily conserved proteins thought to play a role in chromatin organization and is a regulatory element that potentially affects a large number of developmental pathways, explaining the pleiotropic nature of its phenotypic spectrum.¹²

While congenital cardiovascular anomalies are commonly seen in CHARGE syndrome, anomalies of the head and neck vessels have been rarely described, with occasional case reports of isolation of the carotid artery, isolation of the left subclavian artery from the pulmonary artery, and complex cervical arteriovenous fistulas.^17⇓-19

PTA

The PTA is one of the primitive carotid-vertebrobasilar anastomoses; it is the largest of these anastomotic vessels seen in fetal life and is seen in approximately 0.1%–1% of angiographic studies,^20,21 with a recent meta-analysis demonstrating a pooled prevalence of 0.4%.⁶ In utero, the trigeminal artery supplies the basilar artery before development of the PcomA and vertebral arteries, with subsequent involution. The PTA arises from the junction between the petrous and cavernous ICA and runs posteriorly along the trigeminal nerve or crosses over or through the dorsum sellae. Vertebral, posterior communicating, and caudal basilar arteries are often hypoplastic. In 1959, Saltzman²² reported 8 cases and proposed an angiographic classification for the PTA into 2 main types. In Saltzman type 1, or the fetal variant of the PTA, the basilar artery proximal to the insertion of the PTA may be hypoplastic and the PcomA may be absent, making the PTA a very important vessel supplying the entire basilar artery system distal to the anastomosis and providing flow to a large part of the posterior circulation, whereas in Saltzman type 2, or the adult variant, the PTA is a relatively less important vessel, with the posterior cerebral arteries receiving their blood supply predominantly through patent PcomAs and the basilar artery filling by the vertebral arteries.^21,23 While we were not able to classify all PTAs according to the Saltzman system, in our series, the basilar artery was hypoplastic below the PTA in 36% (5/14) of patients and the PcomAs were hypoplastic in 6/6 patients who had undergone an MRA, suggesting that these were probably type 1, thereby making the PTA a potentially significant source of supply to the posterior circulation.

The exact etiology of the persistence of this vessel into postnatal life is unclear. The PTA is the most prominent of the carotid-basilar anastomoses, existing in early fetal life (4- to 5-mm embryonic stage) and providing an important source of flow to the developing rhombencephalon, arising from the developing ICA to connect with the paired longitudinal neural arteries, which would eventually form the basilar artery. Growth of the posterior fossa structures, interposition of the developing basisphenoid cartilages, and evolution of the carotid and vertebrobasilar systems result in obliteration of the PTA at around the 7- to 14-mm embryonic stage, with the PcomA superseding the PTA from above and the vertebral arteries developing from below.²¹ The PTA has been reported in association with vascular malformations, aneurysms, trigeminal neuralgia/compression, and PHACE syndrome, with additional implications in surgical and endovascular treatment-planning, making it a relevant finding to note, with significance potentially emerging later in life.^20,21,24 One of our patients had a mild ICA hypoplasia contralateral to the PTA, but we did not find any vascular malformations or aneurysms in our study.

Association of the PTA and CHARGE Syndrome

To the best of our knowledge, this is the first report describing a compelling association of the PTA with CHARGE syndrome. This finding was readily identified on imaging studies and particularly when thin-section T2WI was available. While it is important for the neuroradiologists to recognize this association, it may also have an additional supplementary value when other criteria are lacking, particularly in cases of atypical or partial/incomplete CHARGE syndrome. The presence of the PTA on neuroimaging studies, particularly in the context of a child with complex cardiovascular anomalies or rhombencephalic dysfunction, should prompt a review for other hitherto unsuspected imaging features of CHARGE syndrome.

The prevalence of PTA of 56% (14/25) in cases of CHARGE syndrome is particularly high, given the relatively low prevalence of the PTA (0.4%).⁶ Similarly, we have not found any reference to the PTA or intracranial arterial anomalies in the context of CHARGE syndrome, including large reviews on this subject.^25,26 This could potentially be due to the lack of thin-section imaging of the posterior fossa or lack of MR angiographic imaging. The PTA was clearly identified on the thin-section T2WI performed for assessment of the inner ears or when thin-section T2WI of the head was available, but it could also be demonstrated in 57% (8/14) on the standard T2WI of the head (Fig 2). There was no particular correlation with sex. We also found that a substantial proportion of patients with CHARGE syndrome, who did not have one of the major or minor diagnostic criteria, had the presence of the PTA, as shown in the Table, in more than half of the cases. This is a strong argument for including the PTA as an additional diagnostic criterion of CHARGE syndrome, to further support the other diagnostic criteria, increase diagnostic confidence, and direct appropriate investigations. Indeed, 2 patients who did not have a typical CHARGE diagnosis (but were genetically confirmed) had a PTA.

CHD7 is expressed ubiquitously in fetal and adult tissues, including eye, olfactory epithelium, inner ear, and the vascular system.⁷ The in situ hybridization analysis of the CHD7 gene during early human development has shown a good correlation between the specific CHD7 expression pattern and the developmental anomalies observed in CHARGE syndrome.¹² Therefore, it is plausible that mutations in this gene in some could result in vascular maldevelopment and persistence of embryonic vessels. Embryologically, the PTA is a significant source of supply to the developing hindbrain in very early fetal life at around 4–6 weeks but has a very limited time span of approximately a week or so, with normal regression of the PTA paralleling the simultaneous development of the PcomA and vertebral arteries, which then take over the function of providing flow to the posterior circulation and the rapidly developing hindbrain.^21,27,28 It is also well-established that hindbrain dysfunction is a characteristic feature of CHARGE syndrome; therefore, the finding of parallel anomalous vascular development is not surprising. Embryologically, CHARGE is thought to be a complex neurocristopathy, related to abnormalities of neural crest cells and abnormal differentiation of the cephalic mesoderm and ectoderm occurring between 3 and 9 weeks of gestation.⁵ This timing also corresponds to and overlaps the time of development and regression of the PTA, thereby potentially explaining its persistence in CHARGE.

PTA versus Other Criteria in CHARGE

Compared with the other accepted diagnostic criteria, the PTA fares well, and the prevalence of PTA in our series (56%, 14/25) is higher than some major and minor criteria (Table). It is well-documented that semicircular canal aplasia is seen in almost all cases of CHARGE syndrome, as was seen in 100% of our cases. However, regarding other major criteria, ocular colobomas were seen in only 64% (16/25), and choanal atresia, in 40% (10/25) of our cohort. Previous studies have indicated the prevalence of ocular colobomas to be 75%–90% and choanal atresia to be 35%–65%.¹² The prevalence of minor criteria in our study was variable, and the PTA also compared favorably. This finding is a strong argument in support of including the PTA as an additional diagnostic criterion. It is also topographically distinct from the other established clinical criteria and adds the missing arterial angle to the pre-existing broad clinicoradiologic phenotype of CHARGE syndrome. In addition to our findings of the high prevalence of the PTA, recent studies have also demonstrated a high prevalence of clival pathology in CHARGE syndrome, with characteristic clival clefts found in nearly 90% of cases, and cerebellar malformations/heterotopia, in >70%, further arguing for expansion of the existing clinical criteria.^15,29,30

Limitations

There are limitations to a retrospective study. In particular, there were variable MR imaging sequences and technical parameters, and MRA was available in only 6/25 patients. However, most of our patients (20/25) had undergone thin-section T2WI, primarily for assessment of the vestibulocochlear nerves and inner ears, and the PTA was well-demonstrated on this sequence. In 3 patients without MRA or thin-section T2WI, the axial head T2WI did not show an obvious PTA, and we acknowledge the possibility that a small PTA may have been present. When one compares the utility of the PTA with other diagnostic criteria, one must consider their prevalence in other syndromic and nonsyndromic populations. We did not look at the prevalence of the PTA in other syndromes or similar conditions; therefore, we are unable to assess the specificity of this finding and how frequently it is present in other related syndromes, though we did not find any reference to the PTA in chromosomal syndromes overlapping with CHARGE syndrome such as 22q11.2 deletion or cat eye syndromes. We also acknowledge that the available clinical data may have been limited in some of our patients. Additional prospective multi-institutional studies would be helpful in further validation of this interesting finding and bringing into focus the association of arterial anomalies with CHARGE syndrome, which may have significance in later life.