A, Right ear in a child with no developed semicircular canals. The arrowhead indicates the rudimentary vestibule. B, Left ear in the same child. The arrowhead indicates the rudimentary vestibule. C, Right ear in a child with hypoplastic right lateral semicircular canal with larger posterior semicircular canal. The arrowhead indicates the lateral canal, which has developed but is small. D, Left ear in the same child. The arrowhead indicates the lateral canal, which has developed but is small.
A, Axial image of the infant cochlea and intratemporal facial nerve. Note the course of the labyrinthine and tympanic segments and their relationship to the basal turn of the cochlea (arrowhead). B, Coronal image of the infant cochlea and intratemporal facial nerve. The arrowhead indicates the tympanic segment of the facial nerve.
Customize your JAMA Network experience by selecting one or more topics from the list below.
Wineland A, Menezes MD, Shimony JS, Shinawi MS, Hullar TE, Hirose K. Prevalence of Semicircular Canal Hypoplasia in Patients With CHARGE Syndrome: 3C Syndrome. JAMA Otolaryngol Head Neck Surg. 2017;143(2):168–177. doi:10.1001/jamaoto.2016.3175
How often do we observe anomalies of the semicircular canals in children with hearing loss and CHARGE syndrome?
In this study of medical record data, all children diagnosed as having CHARGE syndrome and with available temporal bone imaging demonstrated anomalies of the semicircular canals. These patients were diagnosed as having CHARGE either through genetic testing or dysmorphology using Verloes’ criteria.
Anomalous development of the semicircular canals is a major criterion for CHARGE syndrome and should be emphasized by renaming the syndrome to 3C syndrome.
CHARGE syndrome refers to a syndrome involving coloboma, heart defects, atresia choanae, retardation of growth and development, genitourinary disorders, and ear anomalies. However, Verloes revised the characteristics of CHARGE syndrome in 2005 to define this syndrome more broadly. Deficiency of the semicircular canals is now a major criterion for CHARGE syndrome.
To characterize patients with CHARGE syndrome at our center using Verloes’ criteria and to reevaluate the nomenclature for this condition.
Design, Setting, and Participants
We performed a medical chart review of patients with CHARGE syndrome and reviewed their temporal bone imaging studies at a tertiary care children’s hospital affiliated with Washington University in St Louis. Two authors independently reviewed each imaging study (A.W. and K.H.). Radiologic studies, physical findings, genetic tests, and other diagnostic tests were included. Patients with no temporal bone imaging studies were excluded.
Eighteen children were included in this study; 13 children (72%) were male, and the mean (median; range) age of patients at the time of inner ear imaging studies was 2 years (4.5 years; 8 months to 8 years). Coloboma was present in 13 patients (72%) and choanal atresia in 5 (28%); semicircular canal anomalies were present in all patients. Additionally, 13 patients (72%) were diagnosed as having hindbrain anomalies, 17 (94%) as having endocrine disorders, 17 (94%) as having mediastinal organ malformations, and all as having middle or external ear abnormalities and development delay. Cleft lip and cleft palate were found in 6 of 14 patients (43%) who did not have choanal atresia. We tested 16 patients for mutations in the CHD7 gene; 10 were positive (63%) for mutations, 4 (25%) were negative, and 2 (13%) were inconclusive.
Conclusions and Relevance
Semicircular canal anomalies were the most consistent finding in our patients with CHARGE syndrome. Given the high prevalence of semicircular canal hypoplasia and importance of imaging for diagnosing CHARGE syndrome, we propose changing the term CHARGE syndrome to 3C syndrome to emphasize the importance of the semicircular canals and to recall the 3 major criteria for diagnosis: coloboma, choanal atresia, and semicircular canal anomaly. The nomenclature would also reference the 3 semicircular canals in each ear. This new name for CHARGE syndrome would provide a mnemonic and focus the disease on the most important clinical criteria for diagnosis.
In 1981, Pagon et al1 coined the acronym CHARGE for a constellation of congenital anomalies observed in newborns originally described by Hittner et al2 in 1979. The cardinal features of CHARGE syndrome were originally coloboma, heart defects, atresia choanae, retarded growth and development, genitourinary disorders, and ear abnormalities. This syndrome has been under careful reexamination since the identification of mutations in the chromodomain helicase DNA-binding protein 7 gene (CHD7; OMIM, 608892) as the underlying molecular cause of CHARGE syndrome.3 Now that genetic testing is available, the definition of CHARGE syndrome has shifted slightly, although approximately 30% of children with clinical features typical of CHARGE syndrome lack the CHD7 mutation. Verloes4 published revised criteria for CHARGE syndrome in which major and minor signs were noted to comprise 3 tiers: typical, partial, and atypical (Table 1). The major criteria include 2 anomalies associated with the original CHARGE acronym: coloboma and choanal atresia. Verloes4 appended an additional major criterion not included in the original acronym: hypoplasia of the semicircular canals. This criterion is notable in that an imaging study is required to evaluate the semicircular canals, and this finding cannot be readily determined by physical examination.
Our objective was to assess inner ear morphology in patients diagnosed as having CHARGE syndrome, either confirmed by CHD7 testing or based on Verloes’ criteria,4 and to determine the frequency of this finding in children diagnosed as having CHARGE syndrome. We reviewed the imaging studies to identify patients with semicircular canal hypoplasia and to determine if there were other associations that were linked to hypoplasia of the semicircular canals. We found that hypoplasia of the semicircular canals with a relatively normal cochlea is highly associated with the diagnosis of CHARGE syndrome, and abnormal inner ear morphology, particularly when identified early in life, may be the most notable feature that triggers further evaluation for CHARGE syndrome in children who demonstrate atypical presentations for this diagnosis.
This was a collaborative study involving input from faculty from the Departments of Radiology, Genetics, and Otolaryngology-Head and Neck Surgery from the Washington University School of Medicine. Two databases independently managed by the Division of Genetics and Genomics in the Department of Pediatrics and the Department of Otolaryngology at Washington University School of Medicine were queried. Additional patients were obtained by searching for the International Classification of Diseases, Ninth Revision diagnosis code 759.89, which categorizes “other specified congenital anomalies.” These patients were cross-referenced using other diagnosis codes that corresponded to the 3 major signs described by Verloes4: atresia choanae, coloboma, and hypoplasia of the semicircular canals. The computed tomography (CT) and magnetic resonance imaging (MRI) databases at the Mallinckrodt Institute of Radiology at the Washington University School of Medicine were queried for inner ear anomalies. Imaging studies with inner ear anomalies were reviewed, and imaging studies demonstrating characteristics consistent with CHARGE syndrome were examined and corresponding medical charts reviewed. These patients underwent further review and were enrolled in this study if they demonstrated evidence to satisfy Verloes’ criteria4 for CHARGE syndrome based on documented physical examination and radiologic studies.
A standardized data extraction form was created and used independently by 2 authors (A. W. and K. H.) to capture major and minor signs of CHARGE syndrome. The mean (median; range) age of patients at the time of inner ear imaging studies was 2 years (4.5 years; 8 months to 8 years), and 13 (72%) were male.
The Institutional Review Board at the Washington University School of Medicine approved this medical chart review. Because data were anonymous, informed consent was waived.
We identified 20 children at our institution with imaging studies of the inner ear diagnosed as having CHARGE syndrome. Their diagnoses were either supported by an identified mutation in CHD7, by traditional clinical criteria, or by meeting Verloes’ criteria4 for CHARGE syndrome based on criteria in Table 1. One child was excluded because she did not yet have imaging studies of the temporal bone. Another child is included in the discussion but likely does not have CHARGE syndrome based on an inconclusive genetic test (ie, single amino acid substitution in CHD7, not previously described) and no major criterion and 2 minor criteria. The clinical and genetic findings in our patients are shown in Table 2. Fifteen of 18 patients (83%) were diagnosed as having coloboma, 7 (39%) as having choanal atresia, and all as having semicircular canal anomalies. The anomalies observed in the semicircular canals included aplasia of the semicircular canals, where there was a small vestibule and no formation of any canals (Figure 1A and B). Other patients demonstrated hypoplasia of some of the canals, with better formation of other canals (Figure 1C and D). We did not observe any patients who fit Verloes’ criteria4 for CHARGE syndrome with unilateral semicircular canal hypoplasia or aplasia. Of the minor signs, 10 patients (56%) were diagnosed as having lower cranial nerve dysfunction, 13 (72%) as having endocrinopathies, 14 (78%) as having mediastinal organ malformations, 14 (78%) as having development delay, and all as having abnormalities in the middle or external ear. Cleft lip and cleft palate were observed in 5 patients (28%).
Genetic testing was performed in 16 of 19 patients (84%). Two children (CCC07 and CCC16) were tested for CHD7 mutation by fluorescent in situ hybridization only. Their test results were normal; however, they both met criteria for typical CHARGE syndrome (ie, met 2 of 3 major criteria and 2 or more of 5 minor criteria). Sequencing of CHD7 was not completed in these 2 patients. Both children (CCC01 and CCC02) with all 3 major criteria—coloboma, choanal atresia, and semicircular canal anomalies—were found to have pathogenic mutations in CHD7. Six of 8 children (CCC04, CCC05, CCC07, CCC08, CCC09, and CCC14) with typical CHARGE syndrome (ie, met 2 of 3 major criteria and 2 minor criteria) who underwent molecular testing for CHD7 were found to have pathogenic mutations. Of the 2 patients (CCC14 and CCC19) that did not demonstrate pathogenic mutations, 1 child (CCC19) had a paternally inherited CHD7 mutation, p.Met2527Leu, which is classified as a benign polymorphism. In another child (CCC11), whole exome sequencing failed to detect a diagnostic mutation. Both of these 2 children who failed to show CHD7 mutations were diagnosed as having CHARGE syndrome based on 2 of 3 major criteria and 2 or more of the 5 minor criteria. Of the 3 children who were diagnosed as having partial CHARGE syndrome (ie, met 1 of 3 major criteria and 3 of 5 minor criteria), 2 (CCC16 and CCC18) were positive for CHD7 mutations and 1 was found to be normal by sequencing. Overall, of 14 children who met Verloes’ criteria4 for CHARGE syndrome who were tested for CHD7 mutations, 10 (71%) were positive (Table 2).
Imaging studies demonstrated 2 cardinal findings that together are highly suggestive of CHARGE syndrome (Table 3). Rudimentary semicircular canals or an absence of the canals altogether was seen in all patients diagnosed as having CHARGE syndrome. While the inner ear criterion was added to the CHARGE diagnosis only in 2005, this finding appears to be a robust and important criterion for diagnostic purposes. Semicircular canal hypoplasia/aplasia was more common in our population than coloboma or choanal atresia.
In addition, a specific radiologic anomaly of the intratemporal facial nerve was found to be virtually pathognomonic for patients with CHARGE syndrome (Figure 2). In these patients, the facial nerve exits the labyrinthine segment more inferiorly than normal and emerges into the tympanic segment directly lateral to the cochlear promontory, such that the descending segment of the nerve is in the middle ear and not in the mastoid bone and can obscure the view of the round window. This observation was noted in 7 of 18 patients (39%) and is not as common as hypoplasia of the semicircular canals but is highly suggestive of CHARGE syndrome. Anomalies of the facial nerve, including abnormalities in the internal auditory canal, were noted in 12 of 18 children (67%) who underwent temporal bone imaging. Congenital facial palsy was observed in 6 of 18 patients (33%) in our cohort. Three of 7 children (43%) with an anomalous intratemporal course of the facial nerve demonstrated perfectly normal facial nerve function without evidence of facial weakness in any of the divisions.
The diagnosis of CHARGE syndrome is currently based on clinical findings, but genetic testing now plays a major role in augmenting our diagnostic accuracy. The initial presentation of CHARGE syndrome is highly variable, leading at some times to diagnostic uncertainty. It is in these patients that genetic testing can be exceptionally helpful. In addition, the importance of semicircular canal morphology in making the diagnosis of CHARGE syndrome renders temporal bone imaging an important part of the assessment of children with possible CHARGE syndrome. A number of the anomalies that traditionally signaled the diagnosis of CHARGE syndrome may be mild or absent. If a child has all 3 major criteria for CHARGE syndrome (ie, coloboma, choanal atresia, and semicircular canal hypoplasia), then they require no further abnormalities to arrive at the diagnosis of CHARGE syndrome (ie, growth retardation, genitourinary anomalies, cardiac anomalies, or ear/pinna abnormalities). This shifting of criteria has resulted in various diagnostic schemas describing “major” and “minor” criteria. Numerous researchers have provided their own interpretation on which criteria are major or minor.1,4,5 Other researchers have noted anomalous semicircular canals as an overlooked diagnostic clue in evaluating patients with CHARGE syndrome.4,6
In 1987, Guyot et al7 described semicircular canal agenesis and Mondini malformation in a patient with CHARGE syndrome, but at that time, they did not suggest that such inner ear morphology be included in the definition of CHARGE syndrome. In 1993, Morgan et al8 stated that the combination of a hypoplastic incus and abnormal semicircular canals were pathognomonic of CHARGE syndrome. In 1998, Lemmerling et al9 stated that the “absence of the semicircular canals is the most specific change in patients with the CHARGE association.” Since then, several others have reported similar inner ear anomalies in patients with CHARGE syndrome.6,8,10-15 Vesseur et al16 recently reported findings from a cohort of 42 patients with CHARGE syndrome with temporal bone imaging. They reported a 77% rate of canal aplasia and a 20% rate of semicircular canal hypoplasia.16 These numbers are similar to what we observed, with most children having no canal formation and a small subset having partial development of some of the canals.
In 2004, Vissers et al3 reported a 2.3 megabase de novo microdeletion on chromosome 8q12 identified by comparative genomic hybridization in 2 individuals with CHARGE syndrome. Sequencing of candidate genes in this region detected specific mutations in CHD7 in 10 of 17 patients (59%) with CHARGE syndrome. This finding initiated a number of subsequent studies that confirmed the role of CHD7 in both identifying children with CHARGE syndrome and developing models to improve our understanding of the pathogenesis of CHARGE syndrome. Zentner et al17 reviewed a series of 379 patients with CHARGE syndrome who underwent CHD7 testing; 67% tested positive for a CHD7 mutation. Compared with CHD7 mutation–negative individuals, those who tested positive demonstrated a higher rate of coloboma, facial nerve dysfunction, and semicircular canal hypoplasia (98% vs 75%) and a lower rate of delayed growth and development.17 Basic research has demonstrated that CHD7 functions as a transcription regulator by binding preferentially to methylated histones in enhancer regions and near transcription start sites.18 CHD7 binds to enhancer elements in the presence of protein cofactors, and its chromatin and gene regulation effects in modulating the transcriptional output of a target gene depend on the composition of large protein complexes with which it binds.17,18
With respect to the otolaryngologic aspects of CHARGE syndrome, abnormalities in the course of the facial nerve are not frequently observed in children with hearing loss, and for those patients in this study who had only an MRI scan and not a temporal bone CT scan, the precise course of the facial nerve through the petrous bone was difficult to discern. In a number of the children with normal cochleas and internal auditory canals on MRI and normal facial nerve function by physical examination, CT scan was not typically performed. However, if we were to have performed CT and MRI systematically in children with CHARGE syndrome, the prevalence of facial nerve anomalies may have been different. With high-resolution temporal bone CT scans, we can detect the anomalous course of the facial nerve readily; the path of the facial nerve in the population with CHARGE syndrome can be characterized by its entering the middle ear over the face of the cochlea and descending sharply inferiorly over the promontory, then passing lateral to the round window niche and traveling anteriorly and laterally along the cochlear promontory instead of posterior to the cochlea in the facial recess. There were also instances of bifid facial nerves where the imaging studies appeared to show 2 facial nerves within the tympanic segment. This anomalous path of the facial nerve increases the risk of facial nerve injury during otologic surgery, particularly during cochlear implantation. Because of these anomalies, it is good practice to obtain both MRI and CT scan studies prior to cochlear implantation of the patients with CHARGE syndrome.
The overwhelming majority of children with CHARGE syndrome are diagnosed as having hearing loss on their newborn hearing screen. Because these studies are performed at birth, hearing loss is sometimes the first anomaly noted in children with syndromic forms of hearing loss. Thus, it is often hearing loss that drives the decision to obtain neuroimaging in these patients. Incidental findings on brain MRI are often found while querying the inner ear anatomy associated with hearing loss. In our cohort, all children with CHARGE syndrome and available audiometric studies demonstrated some degree of hearing loss. In fact, all children in our cohort with CHARGE syndrome had at least 1 ear with profound hearing loss, and many of them had bilateral hearing loss requiring intervention. One child had a normal hearing ear on 1 side and a profound hearing loss on the other. Thirteen of 17 children (76%) had bilateral severe to profound sensorineural hearing loss. Three of 17 children (18%) with hearing loss demonstrated deficient cochlear nerves on MRI scan. These children with thin or undetectable cochlear nerves on imaging studies generally are profoundly deaf at birth and perform poorly, even with cochlear implantation, because of insufficient innervation of the cochlea for electrical stimulation. Seven of 17 children (41%) with hearing loss underwent cochlear implantation. Outcomes in these children were mixed, but generally, acquiring spoken language was not the goal for cochlear implantation in this population. The parents of these children are counseled conservatively, in part because of the anatomic variations that are prevalent in these children but particularly because of the cognitive delays that are often observed in children with CHARGE syndrome, making speech language prognosis difficult to discern.
While the absence of semicircular canals is highly prevalent in this population, very few children receive attention for balance disturbance. Balance problems are reported in as many as 53% of adolescents and adults with CHARGE syndrome.19,20 Reports of sensory deficits in these patients include visual and hearing impairments, but balance disturbance is also cited as a shortcoming in the quality of life for these individuals. Those children who are also affected by coloboma have poor vision and therefore are not as well equipped to compensate for vestibular weakness using visual cues. We do not routinely obtain vestibular functional testing results in these children with semicircular canal deficits, but if we were to obtain physiologic testing, it would be expected that they would have bilateral severe vestibular hypofunction or even vestibular areflexia. These children should be observed to provide support for vestibular dysfunction and to offer therapies or transportation options for children whose activities are limited by severe balance disturbance.
The formal classification of CHARGE syndrome has undergone several changes since the original descriptions by Hittner et al2 and Hall21 of this constellation of birth defects. Two years later, Pagon et al1 coined the term CHARGE to describe the syndrome. However, the recent update of CHARGE criteria by Verloes4 in 2005 with the creation of a tiered stratification for diagnosing CHARGE into typical, partial, and atypical and the division of criteria into major and minor created a profound change in our practice. With the addition of genetic testing (ie, sequencing of CHD7), we are now finding children who meet the criteria for CHARGE syndrome who would not have been given this diagnosis in the past based on dysmorphology and other tests, such as renal or cardiac ultrasonography. Importantly, the role of skull base imaging is now critical in diagnosing CHARGE syndrome, given the fact that inner ear abnormalities are a major criterion and the prevalence of semicircular canal anomalies in children with CHARGE syndrome is high.
Semicircular canal anomalies were the most consistent finding in patients diagnosed as having CHARGE syndrome in this cohort. Given the high prevalence of semicircular canal hypoplasia and importance of imaging for diagnosing CHARGE syndrome, we propose changing the term CHARGE syndrome to 3C syndrome to emphasize the importance of the semicircular canals and to recall the 3 major criteria for diagnosis, ie, coloboma, choanal atresia, and semicircular canal anomaly. The nomenclature would also reference the 3 semicircular canals in each ear. This new nomenclature for CHARGE syndrome would provide a mnemonic and focus the disease on the most important clinical criteria for this diagnosis.
The finding of semicircular canal hypoplasia on an imaging study should immediately raise the possibility of CHARGE syndrome, particularly in children with hearing loss and other syndromic features. An otolaryngologist may be the first to recognize this feature in individuals with atypical or partial CHARGE syndrome and can be instrumental in guiding the child to special therapies and further evaluation beyond the hearing evaluation for these children. In addition, the diagnosis of CHARGE syndrome imparts a prognosis for hearing rehabilitation that is qualitatively different than hearing loss in nonsyndromic children. Thus, counseling regarding therapeutic options must take into account comorbidities associated with CHARGE syndrome, both related to inner ear anomalies and cognitive issues, such that expectations and interventions can be tailored to the individual needs of the child and the family.
Corresponding Author: Keiko Hirose, MD, Campus Box 8115, 660 S Euclid Ave, Washington University School of Medicine, St Louis, MO 63110 (firstname.lastname@example.org).
Accepted for Publication: August 19, 2016.
Published Online: November 10, 2016. doi:10.1001/jamaoto.2016.3175
Author Contributions: Dr Hirose had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Concept and design: Wineland, Hirose.
Acquisition, analysis, or interpretation of data: Wineland, Menezes, Shimony, Shinawi, Hirose.
Drafting of the manuscript: Wineland, Hirose.
Critical revision of the manuscript for important intellectual content: Menezes, Shimony, Shinawi, Hirose.
Statistical analysis: Wineland.
Administrative, technical, or material support: Wineland, Menezes, Shimony, Hirose.
Study supervision: Hirose.
Conflict of Interest Disclosures: All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Dr Hullar is on the medical advisory board for Advanced Bionics and Med El. No other disclosures were reported.
Previous Presentation: An early version of this study was presented at the American Society of Pediatric Otolaryngology; April 24, 2015; Boston, Massachusetts.
Create a personal account or sign in to: