From the Department of Otolaryngology-Head and Neck Surgery, Division of Audiology and Electrophysiology, University of Michigan Health System, Ann Arbor, Michigan
| ABSTRACT |
|---|
|
|
|---|
Methods. Clinical data were gathered to examine 1) the variety of hearing results, 2) the average age at the time of hearing loss identification in 22 children with CHARGE using electrophysiologic and behavioral test methods, 3) the usefulness of the ABR as an early indicator of hearing sensitivity for a select group composed of children from the present study and from an earlier report from the same institution, and 4) the value of congenital facial paralysis as a predictor of sensorineural hearing loss in CHARGE children seen in the authors institution since 1983.
Results. All children had 4 or more acronymic features, including colobomatous defects or choanal atresia. Ear anomalies/hearing loss occurred at least as frequently as other primary features. A total of 81% of patients had hearing loss; in this subset, 1 child had a mild degree of loss, and the remaining children had moderate or greater losses. The average age at which ABR confirmed hearing status was 3.8 months, whereas for behavioral testing, that age was 24.7 months, a statistically significant difference. In a select group of 16 children, no statistical differences existed when comparing threshold results of early electrophysiologic testing with behavioral test findings obtained at a later date. Contingency analysis suggests that congenital facial paralysis and sensorineural hearing loss are related.
Conclusions. Hearing loss is prevalent in children with CHARGE syndrome. Within a wide range of results exists a propensity for moderate or greater hearing loss in children with sensorineural or mixed impairments. Congenital facial palsy seems to be a valid statistical predictor of sensorineural hearing loss and can be a useful device in audiologic decision making. A "window of opportunity" for audiologic intervention exists in the first few months of life. Primary care providers are encouraged to recognize the need for immediate, early audiologic referral of their patients suspected to have CHARGE. Evaluation of hearing sensitivity during infancy and, when appropriate, provision of amplification are important components in the process of auditory habilitation. These efforts are in keeping with various professional guidelines that call for early detection of hearing loss and subsequent prompt intervention to minimize effects on infant development.
Key Words: audiologic intervention auditory brainstem response behavioral hearing tests CHARGE syndrome congenital facial palsy multiple congenital anomalies
Abbreviations: CHARGE ocular coloboma heart defects atretic choanae retarded growth or development genital hypoplasia ear anomalies or hearing loss
Abbreviations: UMHS, University of Michigan Health System ABR, auditory brainstem responses
| INTRODUCTION |
|---|
|
|
|---|
The CHARGE acronym encourages medical caregivers to search for other, sometimes subtle, anomalies, particularly when a colobomatous defect or choanal atresia is identified. Important nonacronymic features have been reported3,5,21 and include orofacial clefts, tracheo-esophageal atresia, renal anomalies, facial palsy, chronic secretions,22 and vestibular anomalies.2328 Recent modifications of the 6 primary characteristics emphasize the need to identify cranial nerve deficits when choanal atresia, colobomatous defects, facial palsy, or external ear anomalies are found.29 The last feature may be characterized by a short and/or wide pinna with minimal lobule, hypoplastic helix, and triangular concha.30 The external ear is variously described as posteriorly rotated; and cupped, folded, or lopped; and square in shape.7
Determining hearing function and maximizing the communication potential of children with CHARGE syndrome is especially important because of a tendency for multisensory involvement and developmental delays. This objective can be complicated by 2 issues. First, behavioral hearing tests that incorporate visual reinforcement are considerably more difficult to conduct when the visual system is affected, a common finding in many patients with CHARGE. Reported incidences of ocular coloboma range from 86% to 92% in larger series.21,31 In isolation, small, inferior colobomas generally are not problematic.32 However, significant visual field impairments are common when the congenital defect involves chorioretinal structures or the optic nerve. Microphthalmos, cataracts, nystagmus, and bilateral congenital glaucoma also occur.32,33
Recurring otitis media with effusion in children with CHARGE has an incidence rate approaching 100% and is the second issue that affects the confirmation of peripheral hearing function. It can be a confounding variable in studies of hearing loss in children with CHARGE and is associated with reports of fluctuating hearing loss.7,23,26,34,35
No laboratory test can confirm a diagnosis of CHARGE syndrome, ie, its confirmation is formed by the clinical findings of numerous specialists. Within that process, audiologists help determine the presence, magnitude, and nature of hearing loss in children with CHARGE syndrome. They are involved in the provision and maintenance of personal and classroom amplification systems, such as FM units or soundfield devices, and also play an essential role in a core team that ensures that educational and social goals are not thwarted.30,36,37
The present report seeks to answer the following questions: Is there a propensity for degree and type of hearing loss in CHARGE syndrome? What does the literature report about hearing loss in patients with CHARGE? Do electrophysiologic techniques afford an early and accurate determination of the peripheral hearing status of infants with CHARGE? Does a predictive relationship exist between congenital facial palsy and sensorineural hearing loss in this enigmatic condition?
| METHODS |
|---|
|
|
|---|
Evaluation of hearing sensitivity was conducted in the manner most appropriate for a patients age and developmental level. Behavioral testing procedures were typically used with older children, whereas threshold auditory brainstem responses (ABR) were chosen to evaluate infants and children who were otherwise unable to participate. For newborns and other infants who were younger than approximately 6 months, ABR were obtained as patients slept unsedated in the audiology clinic. Inpatients were usually sedated orally with chloral hydrate or with injectable midazolam. Otherwise, children were taken to the operating room, where ABRs were obtained under general anesthesia, often in conjunction with otologic or plastic surgery procedures. Electrophysiologic methods used to predict peripheral hearing outcomes were described previously.38,39 Briefly, recording electrodes were positioned high on the forehead and the mastoids or earlobes. Air conduction thresholds were obtained by presenting click and 500 Hz or 1000 Hz tonal stimuli monaurally through insert earphones.
For identifying threshold ABR, waveforms were inspected for reductions in amplitude and prolongations of absolute latency as stimulus intensity decreased. Depending on the auditory stimulus, a threshold latency range of 9 to 12 milliseconds was typical (Fig 1).
|
Information about cochlear function was often needed, particularly when a conductive component of hearing loss was encountered. Stimuli were delivered via a bone-conduction transducer positioned on the mastoid after repositioning of the reference electrode away from the mastoid (often on the posterior earlobe). The nontest ear was masked using white noise. Bone-conduction threshold responses were obtained as described previously after consideration of impedance differences and possible alterations in spectral composition of the stimuli presented through the bone transducer.
ABR threshold testing was conducted in various settings throughout the institution, often not in a sound-treated and electrically shielded audiologic suite. Considerable efforts were made to minimize ambient noise in the vicinity. In settings such as intensive care units, where stray electromagnetic signals were encountered, 60-Hz notch filters and extended periods of averaging were used. ABR measures were repeated; post hoc statistical analyses made of these trials verified their replicability. Audiometric equipment used during this study was calibrated quarterly and met current American National Standards Institute standards.40
When a childs age and developmental level permitted, behavioral testing was conducted in the soundfield of a sound-treated booth. For older and developmentally advanced children, stimuli were presented through insert earphones during play audiometry to obtain tonal and speech thresholds for each ear. Bone-conduction testing was included, with contralateral masking used as necessary. Often, 2 or more appointments were needed to obtain satisfactory results.
Determination of the type of hearing loss was based on air and bone-conduction results, the presence or absence of air-bone gaps, and/or abnormal immittance findings in the better-hearing ear. Tympanometry was performed using higher-frequency probe tones for infants. Otoscopy and computed tomography results were also used to assist in this effort. For example, when anomalous ossicles or a hypoplastic cochlea was identified on temporal bone computed tomography, conductive or sensory hearing losses, respectively, were presumed to exist. Recurring ear disease confounded attempts to identify the type and degree of hearing loss, and so the category that best characterized a patients hearing was selected. For instance, if a patient with a normal ABR in the operating room later had recurring otitis media that clearly responded to medical treatment, then his results were placed in the normal hearing category. A different patient who had moderate sensorineural loss and later developed refractory middle ear effusions would be placed in a severe mixed category if audiologic results later demonstrated an additional conductive component.
| RESULTS |
|---|
|
|
|---|
|
Figure 2 illustrates that retarded growth/development (100%), ear anomalies/hearing loss (100%), and heart defects (95%) occur most frequently. Colobomatous defects (73%), choanal atresia/stenosis (67%), and genital hypoplasia (59%) occur somewhat less often. Congenital facial palsy is present in 48% of the 21 children who could be analyzed for this feature.
|
|
Patients were on average 3.8 months old (range: 1 week14 months) when ABR results determined peripheral hearing sensitivity. In contrast, when hearing loss was first confirmed behaviorally, patients were 24.7 months old (range: 5 months5 years). An unpaired Students t test suggested that these differences were statistically significant (t = 3.12, degrees of freedom = 18, P = .006).
The accuracy with which ABR predicted peripheral hearing sensitivity (as established later behaviorally) was then evaluated in a select group of 16 children who had both ABR and behavioral hearing testing in the authors institution in the past 18 years. This select group was composed of 9 children from a previous report7 and 7 children from the present article. Results were classified normal or abnormal as described in the Methods section. Table 2 shows that ABR and behavioral test results were in agreement in 13 of the 16 (81%) children. Of the 3 cases that were incorrectly predicted, ABR results were inaccurate by 1 category level for each child. That is, hearing thresholds for 2 patients (A7 and A19) were overestimated and for the remaining patient (B5) underestimated by
30 dB. In the first 2 cases, more accurate predictions of hearing sensitivity were precluded by middle ear effusion present at the time of the outpatient ABR. For patient B5, no effusion was present at the time of early ABR testing; later, the child had chronic middle ear effusions with associated changes in hearing thresholds. In this select group, no child with severe hearing loss was misidentified using ABR.
|
Mean scores (2.8 and 2.7) and 95% confidence intervals (2.03.5 and 2.03.4) were similar for each group, whereas the maximum difference between the cumulative distributions ("D"), 0.06, with a corresponding P of 1.0 suggests that no significant difference exists between the control group and treatment group scores. The data were noted to be abnormally distributed, ie, median scores of 3 and 4 reflect the prevalence of moderate or greater hearing loss in CHARGE syndrome.
| DISCUSSION |
|---|
|
|
|---|
Earlier studies found that children with risk factors for hearing loss were identified by approximately 16 to 22 months of age.4648 An inverse relationship between age of identification and degree of hearing loss was reported, ie, children with less severe hearing loss took longer to identify, and there was a larger interval between age at identification and hearing aid fitting. Recently, in a large, multicenter study in New York state, Dalzell et al49 reported a significant decrease in the average age at the time of hearing loss identification. On the basis of their data, 52 neonatal intensive care unit graduates with any degree of hearing loss in either ear were detected by 5.9 months of age; amplification was fit to 36 infants with bilateral hearing loss at approximately 10 months of age, resulting in a 3-month interval between identification and hearing aid fitting. These data may reflect a reduction in the average age of a child at the time of hearing loss identification and at the moment of hearing aid fitting and approximate the guidelines from the Joint Committee on Infant Hearing.45
Early audiologic intervention is particularly important for children with CHARGE because vision and hearing impairments coexist.29,32,50 In the present article, 81% of children have hearing loss that ranges from mild to severe or greater. However, with the entire spectrum of hearing function represented in this and other large reports,7,34 no presumptions about a young CHARGE patients peripheral hearing abilities should be made, making it important to quantify a childs hearing in a timely manner to facilitate the process of hearing habilitation. The incidence rate and hearing levels from the present report compare favorably with data in a different study from this same institution.7 In that article, three fourths of patients had hearing loss, and all but 2 of the children with hearing impairment and CHARGE had moderate or greater losses in at least the better hearing ear.
ABR measures and behavioral hearing tests provide information about a childs hearing, either indirectly or directly. However, because the former method requires only a calm patient and a functional afferent auditory system, one can obtain information about peripheral hearing sensitivity much earlier than if one waits many months for adequate development of these multiply involved children. The average age of children when ABR results were obtained was approximately 4 months, and when using behavioral hearing tests, the average age at the time of initial hearing loss identification was approximately 25 months. Significant interventions should begin as soon as a hearing impairment is identified so that the impact of deprived sensory inputs on development is minimized. These include otologic care, the fitting of amplification, referral to local school districts for creation of individualized education programs, and the initiation of support that ultimately assists the child in meeting his or her maximum performance levels. The ABR offers the best opportunity for early identification of hearing loss in these children and should be requested early in life.
The data reflecting differences in the age of hearing loss identification by form of hearing test are surely influenced by several covarying factors, including an infants availability for testing before hospital discharge, recognition of the need for referral to audiology by the primary care physician, and ability of the care providers to understand the need for early confirmation of hearing status. Despite these issues, the age at identification of hearing loss using ABR in UMHS patients with CHARGE syndrome approximates the guidelines of the Joint Committee on Infant Hearing Screening45 that call for identification by 3 months of age, followed by provision of amplification several months later. It is important that the "window of opportunity" for audiologic intervention that exists in the first several months of life be recognized by primary care providers as an opportune time in which to request that hearing function be evaluated with threshold ABR measures. This is essential to order before the childs hospital discharge if multiple system involvement is suspected. When early identification of hearing loss in deaf and hard-of-hearing children is realized, delays in language acquisition and development can be minimized effectively.51
A question posed in an earlier article considered the relationship between congenital facial palsy and sensorineural hearing loss in young patients with CHARGE.7 In that report, facial paralysis was found to predict reliably sensorineural hearing loss (P < .025) in 20 children with CHARGE. Sensorineural hearing loss in the setting of congenital facial palsy occurred 3 times more frequently than did sensorineural hearing loss in the absence of facial palsy. Establishing this link was helpful both in the evaluation of hearing function and in planning for amplification and educational support for children with CHARGE syndrome.
This relationship was re-examined in an expanded group of 40 children with CHARGE seen in the authors institution in the past 18 years. This larger set includes 20 children from the earlier study and 20 children from the present report (Table 3). As seen, facial palsy and sensorineural hearing loss coexist in 16 of 19 patients. Without congenital facial palsy, sensorineural hearing loss is present in 8 of 21 patients. In this larger group, congenital facial paralysis accurately predicts sensorineural hearing loss (P = .003), suggesting that congenital facial paralysis and sensorineural hearing loss are related.
|
Although sparse, recent reports strengthen the notion that electrophysiologic measures have value in the early identification of hearing levels in children with CHARGE. Brown and Israel52 described fraternal twins who had CHARGE and were evaluated using click-evoked ABR at 2 weeks of age. Subsequent behavioral testing confirmed the initial findings that suggested moderate to severe hearing loss for which amplification was provided shortly thereafter. Toriello53 reported on 3 unrelated children with CHARGE. Patients 2 and 3 from that report had initial testing conducted at 1 to 2 months of age using the ABR. Results for their second patient suggested a moderate to severe hearing loss, and behavioral results approximately 2 years later confirmed the electrophysiologic results. Follow-up results for the third patient were not reported.
A review of other clinical investigations of CHARGE that include hearing results suggests that a minority of patients have normal or near-normal hearing.23,26,34,35 Results from these studies and those of the present investigation are shown in Table 4. Results are arranged to reflect the categories of hearing loss used in the present study. By reviewing these data, one can discern a trend toward greater degrees of hearing loss across studies. The cumulative data suggest that 80% of patients in the 5 studies had peripheral hearing thresholds greater than an average of approximately 40 to 45 dB in the best-hearing ear. The inclusion of children with mild hearing loss increases the rate of hearing loss in CHARGE to 87%. Given the rate of hearing loss in this population, referral for early ABR testing, preferably before hospital discharge, is reasonable.
|
| CONCLUSION |
|---|
|
|
|---|
Chronic illness, numerous surgeries, and many medical appointments keep the families of these children very busy. Furthermore, multisensory involvement and subsequent developmental delays hinder behavioral confirmation of hearing loss that, in the present report, was found to occur at an average age of 24.7 months. In contrast, threshold ABR results were available by 3.8 months of age. To avoid a delay in hearing loss identification of approximately 21 months, a "window of opportunity" for audiologic intervention that may often coincide with early hospitalization should be used to arrange for audiologic evaluation of the newborn with CHARGE features.
When these data were combined with those of a previous report, early indications of auditory sensitivity using threshold ABR measures were found to predict accurately behavioral results obtained many months later, validating ABR as an effective tool for early assessment of hearing function in children with multisystem involvement. Finally, statistical analyses confirmed that congenital facial palsy has validity as a predictor of sensorineural hearing loss; the rate of sensorineural hearing loss in children with facial palsy was twice that found in the group without facial involvement. The combination of this predictor tool and the frequency of moderate or greater degrees of hearing loss may be useful when planning for the provision of various systems of amplification.
| ACKNOWLEDGMENTS |
|---|
| FOOTNOTES |
|---|
Reprint requests to (B.M.E.) University of Michigan Health System, Department of Otolaryngology-Head and Neck Surgery, Division of Audiology and Electrophysiology, 1500 E Medical Center Dr, TC 1904, Ann Arbor, MI 48109. E-mail: bedwards{at}umich.edu
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
E. A. Bosman, A. C. Penn, J. C. Ambrose, R. Kettleborough, D. L. Stemple, and K. P. Steel Multiple mutations in mouse Chd7 provide models for CHARGE syndrome Hum. Mol. Genet., November 15, 2005; 14(22): 3463 - 3476. [Abstract] [Full Text] [PDF] |
||||
![]() |
Radiology Quiz Case 1: Diagnosis Arch Otolaryngol Head Neck Surg, October 1, 2005; 131(10): 925 - 926. [Full Text] [PDF] |
||||
![]() |
D. S. Theodoropoulos, G. A. Theodoropoulos, B. M. Edwards, P. R. Kileny, and L. A. Van Riper Immune Deficiency and Hearing Loss in CHARGE Association Pediatrics, March 1, 2003; 111(3): 711 - 712. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||