Adapted from American Academy of Pediatrics, Joint Committee on Infant Hearing: Joint Committee on Infant Hearing 1994 position statement, Pediatrics 95:152, 1995.

Table 629-2 COMMON TYPES OF HEREDITARY NONSYNDROMIC SENSORINEURAL HEARING LOSS

LOCUS	GENE	AUDIO PHENOTYPE
DFN3	POU3F4	Conductive hearing loss due to stapes fixation mimicking otosclerosis; superimposed progressive SNHL
DFNA1	DIAPH1	Low-frequency loss beginning in the 1st decade and progressing to all frequencies to produce a flat audio profile with profound losses throughout the auditory range
DFNA2	KCNQ4	Symmetrical high-frequency sensorineural loss beginning in the 1st decade and progressing over all frequencies
DFNA2	GJB3	Symmetrical high-frequency sensorineural loss beginning in the 3rd decade
DFNA 6/14/38	WFS1	Early-onset low-frequency sensorinerual loss; about 75% of families dominantly segregating this audio profile carry missense mutations in the C-terminal domain of wolframin.
DFNA10	EYA4	Progressive loss beginning in the 2nd decade as a flat to gently sloping audio profile that becomes steeply sloping with age
DFNA13	COL11A2	Congenital mid-frequency sensorineural loss that shows age-related progression across the auditory range
DFNA15	POU4F3	Bilateral progressive sensorineural loss beginning in the 2nd decade
DFNA20/26	ACTG1	Bilateral progressive sensorineural loss beginning in the 2nd decade; with age, the loss increases with threshold shifts in all frequencies, although a sloping configuration is maintained in most cases
DFNB1	GJB2, GJB6	Hearing loss varies from mild to profound. The most common genotype, 35delG/35delG, is associated with severe to profound SNHL in about 90% of affected children; severe to profound deafness is observed in only 60% of children who are compound heterozygotes carrying 1 35delG allele and any other GJB2 SNHL-causing allele variant; in children carrying 2 GJB2 SNHL-causing missense mutations, severe to profound deafness is not observed.
DFNB4	SLC26A4	DFNB4 and Pendred syndrome (see Table 629-3) are allelic. DFNB4 hearing loss is associated with dilatation of the vestibular aqueduct and can be unilateral or bilateral. In the high frequencies, the loss is severe to profound; in the low frequencies, the degree of loss varies widely. Onset can be congenital (prelingual), but progressive postlingual loss also is common.
mtDNA 1555A > G	12S rRNA	Degree of hearing loss varies from mild to profound but usually is symmetrical; high frequencies are preferentially affected; precipitous loss in hearing can occur after aminoglycoside therapy.

From Smith RJH, Bale JF Jr, White KR: Sensorineural hearing loss in children, Lancet 365:879–890, 2005.

SNHL, sensorineural hearing loss.

Table 629-3 COMMON TYPES OF SYNDROMIC SENSORINEURAL HEARING LOSS

SYNDROME	GENE	PHENOTYPE
DOMINANT
Waardenberg (WS1)	PAX3	Major diagnostic criteria include dystopia canthorum, congenital hearing loss, heterochromic irises, white forelock, and an affected first-degree relative. About 60% of affected children have congenital hearing loss; in 90%, the loss is bilateral.
Waardenberg (WS2)	MITF, others	Major diagnostic criteria are as for WS1 but without dystopia canthorum. About 80% of affected children have congenital hearing loss; in 90%, the loss is bilateral.
Branchio-otorenal	EYA1	Diagnostic criteria include hearing loss (98%), preauricular pits (85%), and branchial (70%), renal (40%), and external-ear (30%) abnormalities. The hearing loss can be conductive, sensorineural, or mixed, and mild to profound in degree.
RECESSIVE
Pendred syndrome	SLC26A4	Diagnostic criteria include sensorineural hearing loss that is congenital, nonprogressive, and severe to profound in many cases, but can be late-onset and progressive; bilateral dilation of the vestibular aqueduct with or without cochlear hypoplasia; and an abnormal perchlorate discharge test or goiter.
Usher syndrome type 1 (USH1)	USH1A, MYO7A, USH1C, CDH23, USH1E, PCDH15, USH1G	Diagnostic criteria include congenital, bilateral, and profound hearing loss, vestibular areflexia, and retinitis pigmentosa (commonly not diagnosed until tunnel vision and nyctalopia become severe enough to be noticeable).
Usher syndrome type 2 (USH2)	USH2A, USH2B, USH2C, others	Diagnostic criteria include mild to severe, congenital, bilateral hearing loss and retinitis pigmentosa; hearing loss may be perceived as progressing over time because speech perception decreases as diminishing vision interferes with subconscious lip reading.
Usher syndrome type 3 (USH3)	USH3	Diagnostic criteria include postlingual, progressive sensorineural hearing loss, late-onset retinitis pigmentosa, and variable impairment of vestibular function.

From Smith RJH, Bale JF Jr, White KR: Sensorineural hearing loss in children, Lancet 365:879–890, 2005.

Sudden SNHL in a previously healthy child is uncommon but may be due to otitis media or other middle ear pathologies. Usually these causes are obvious from the history and physical examination. Sudden loss of hearing in the absence of obvious causes often is the result of a vascular event affecting the cochlear apparatus or nerve, such as embolism or thrombosis (secondary to prothrombotic conditions). Additional causes include perilymph fistula, drugs, trauma, and the first episode of Ménière syndrome. In adults, sudden SNHL is often idiopathic and unilateral; it may be associated with tinnitus and vertigo. Identifiable causes of sudden SNHL include infections (Epstein-Barr virus, varicella zoster virus, herpes simplex virus), vascular injury to the cochlea, endolymphatic hydrops, and inflammatory diseases.

Infectious Causes

The most common infectious cause of congenital SNHL is cytomegalovirus (CMV), which infects 1/100 newborns in the USA (Chapters 247 and 630). Of these, 6,000-8,000 infants each year have clinical manifestations, including approximately 75% with SNHL. Congenital CMV warrants special attention because it is associated with hearing loss in its symptomatic and asymptomatic forms, and the hearing loss may be progressive. Some children with congenital CMV have suddenly lost residual hearing at 4-5 yr of age. Much less common congenital infectious causes of SNHL include toxoplasmosis and syphilis. Congenital CMV, toxoplasmosis, and syphilis also can manifest with delayed onset of SNHL months to years after birth. Rubella, once the most common viral cause of congenital SNHL, is very uncommon because of effective vaccination programs. In utero infection with herpes simplex virus is rare, and hearing loss is not an isolated manifestation.

Other postnatal infectious causes of SNHL include neonatal group B streptococcal sepsis and bacterial meningitis at any age. Streptococcus pneumoniae is the most common cause of bacterial meningitis that results in SNHL after the neonatal period and has become less common with the routine administration of pneumococcal conjugate vaccine. Haemophilus influenzae type b, once the most common cause of meningitis resulting in SNHL, is rare owing to the Hib conjugate vaccine. Uncommon infectious causes of SNHL include Lyme disease, parvovirus B19, and varicella. Mumps, rubella, and rubeola, all once common causes of SNHL in children, are rare owing to vaccination programs.

Genetic Causes

Genetic causes of SNHL probably are responsible for as many as 50% of SNHL cases (see Tables 629-2 and 629-3). These disorders may be associated with other abnormalities, may be part of a named syndrome, or can exist in isolation. SNHL often occurs with abnormalities of the ear and eye and with disorders of the metabolic, musculoskeletal, integumentary, renal, and nervous systems.

Autosomal dominant hearing losses account for about 10% of all cases of childhood SNHL. Waardenburg (types I and II) and branchio-otorenal syndromes represent 2 of the most common autosomal dominant syndromic types of SNHL. Types of SNHL are coded with a 4-letter code and a number, as follows: DFN = deafness, A = dominant, B = recessive, and number = order of discovery, for example DFNA 13. Autosomal dominant conditions in addition to those just discussed include DFNA 1-11, 13, 15, 17, 20, 22, 28, 36, 48 and mutations in the crystallin gene (CRYM).

Autosomal recessive genetic SNHL, both syndromic and nonsyndromic, accounts for about 80% of all childhood cases of SNHL. Usher syndrome (types 1, 2, and 3), Pendred syndrome, and the Jervell and Lange-Nielsen syndrome (one form of the long Q-T syndrome) are 3 of the most common syndromic recessive types of SNHL. Other autosomal recessive conditions include Alström syndrome, type 4 Bartter syndrome, biotinidase deficiency and DFNB1-4, 6-9, 12, 16, 18, 21-23, 28-31, 36, 37, 67.

Unlike children with an easily identified syndrome or with anomalies of the outer ear, who may be identified as being at risk for hearing loss and consequently monitored adequately, children with nonsyndromic hearing loss present greater diagnostic difficulty. Mutations of the connexin-26 and -30 genes have been identified in autosomal recessive (DNFB 1) and autosomal dominant (DNFA 3) SNHL and in sporadic patients with nonsyndromic SNHL; up to 50% of nonsyndromic SNHL may be related to a mutation of connexin-26. Mutations of the GJB2 gene co-localize with DFNA 3 and DFNB 1 loci on chromosome 13, are associated with autosomal nonsyndromic susceptibility to deafness, and are associated with as many as 30% of cases of sporadic severe to profound congenital deafness and 50% of cases of autosomal recessive nonsyndromic deafness. Sex-linked disorders associated with SNHL, thought to account for 1-2% of SNHL, include Norrie disease, the otopalatal digital syndrome, Nance deafness, and Alport syndrome. Chromosomal abnormalities such as trisomy 13-15, trisomy 18, and trisomy 21 also can be accompanied by hearing impairment. Patients with Turner syndrome have monosomy for all or part of 1 X chromosome and can have CHL, SNHL, or mixed hearing loss. The hearing loss may be progressive. Mitochondrial genetic abnormalities also can result in SNHL (see Table 629-2).

Many genetically determined causes of hearing impairment, both syndromic and nonsyndromic, do not express themselves until some time after birth. Alport, Alström, and Down syndromes, von Recklinghausen disease, and Hunter-Hurler syndrome are genetic diseases that can have SNHL as a late manifestation.

Physical Causes

Agenesis or malformation of cochlear structures including the Scheibe, Mondini (Fig. 629-1), Alexander, and Michel anomalies; enlarged vestibular aqueducts (which may be associated with Pendred Syndrome); and semicircular canal anomalies may be genetic. These anomalies probably occur before the 8th wk of gestation and result from arrest in normal development, aberrant development, or both. Many of these anomalies also have been described in association with other congenital conditions such as intrauterine CMV and rubella infections. These abnormalities are quite common; in as many as 20% of children with SNHL, obvious or subtle temporal bone abnormalities are seen on high-resolution CT scanning or MRI.

Figure 629-1 Mondini’s dysplasia shown by CT of the temporal bone in a child with Pendred syndrome. Both dilatation of the vestibular aqueduct and cochlear dysplasia are present in this section. In the larger of the two inset images of a normal temporal bone, the vestibular aqueduct is visible but much smaller (arrow). The cochlea appears normal, and in the smaller inset image of a more inferior axial section, the expected number of cochlear turns can be clearly counted. *Internal auditory canal.

(From Smith RJH, Bale JF Jr, White KR: Sensorineural hearing loss in children, Lancet 365:879–890, 2005.)

Conditions, diseases, or syndromes that include craniofacial abnormalities may be associated with conductive hearing loss and possibly with SNHL. Pierre Robin, Treacher Collins, Klippel-Feil, Crouzon, and branchio-otorenal syndromes and osteogenesis imperfecta often are associated with hearing loss. Congenital anomalies causing CHL include malformations of the ossicles and middle-ear structures and atresia of the external auditory canal.

SNHL also can occur secondary to exposure to toxins, chemicals, and antimicrobials. Early in pregnancy, the embryo is particularly vulnerable to the effects of toxic substances. Ototoxic drugs, including aminoglycosides, loop diuretics, and chemotherapeutic agents (cisplatin) also can cause SNHL. Congenital SNHL can occur secondary to exposure to these drugs as well as to thalidomide and retinoids. Certain chemicals, such as quinine, lead, and arsenic, can cause hearing loss both pre- and postnatally.

Trauma, including temporal bone fractures, inner ear concussion, head trauma, iatrogenic trauma (e.g., surgery, extracorporeal membrane oxygenation [ECMO]), radiation exposure, and noise, also can cause SNHL. Other uncommon causes of SNHL in children include immune disease (systemic or limited to the inner ear), metabolic abnormalities, and neoplasms of the temporal bone.

Effects of Hearing Impairment

The effects of hearing impairment depend on the nature and degree of the hearing loss and on the individual characteristics of the child. Hearing loss may be unilateral or bilateral, conductive, sensorineural, or mixed; mild, moderate, severe, or profound; of sudden or gradual onset; stable, progressive, or fluctuating; and affecting a part or all of the audible spectrum. Other factors, such as intelligence, medical or physical condition (including accompanying syndromes), family support, age at onset, age at time of identification, and promptness of intervention, also affect the impact of hearing loss on a child.

Most hearing-impaired children have some usable hearing. Only 6% of those in the hearing-impaired population have bilateral profound hearing loss. Hearing loss very early in life can affect the development of speech and language, social and emotional development, behavior, attention, and academic achievement. Some cases of hearing impairment are misdiagnosed because affected children have sufficient hearing to respond to environmental sounds and can learn some speech and language but when challenged in the classroom cannot perform to full potential.

Even mild or unilateral hearing loss can have a detrimental effect on the development of a young child and on school performance. Children with such hearing impairments have greater difficulty when listening conditions are unfavorable (e.g., background noise and poor acoustics), as can occur in a classroom. The fact that schools are auditory-verbal environments is unappreciated by those who minimize the impact of hearing impairment on learning. Hearing loss should be considered in any child with speech and language difficulties or below-par performance, poor behavior, or inattention in school (Table 629-5).

Table 629-5 HEARING HANDICAP AS A FUNCTION OF AVERAGE HEARING THRESHOLD LEVEL OF THE BETTER EAR

Children with moderate, severe, or profound hearing impairment and those with other handicapping conditions often are educated in classes or schools for children with special needs. The auditory management and choices regarding modes of communication and education for children with hearing handicaps must be individualized, because these children are not a homogeneous group. A team approach to individual case management is essential, because each child and family unit has unique needs and abilities.

Hearing Screening

Hearing impairment can have a major impact on a child’s development, and because early identification improves prognosis, screening programs have been widely and strongly advocated. The National Center for Hearing Assessment and Management estimates that the detection and treatment at birth of hearing loss saves $400,000 per child in special education costs; screening costs approximately $8-50/child. Data from the Colorado newborn screening program suggest that if hearing-impaired infants are identified and treated by age 6 mo, these children (with the exception of those with bilateral profound impairment) should develop the same level of language as their age-matched peers who are not hearing impaired. This is compelling support for the establishment of mandated newborn hearing screening programs for all children. The American Academy of Pediatrics endorses the goal of universal detection of hearing loss in infants before 3 mo of age, with appropriate intervention no later than 6 mo of age. Currently, hearing screening has been mandated in 39 states in the USA, plus the District of Columbia and Puerto Rico.

Until mandated screening programs are established universally, many hospitals will continue to use other criteria to screen for hearing loss. Some use the high-risk criteria (see Table 629-1) to decide which infants to screen; some screen all infants who require intensive care; and some do both. The problem with using high-risk criteria to screen is that 50% of cases of hearing impairment will be missed, either because the infants are hearing impaired but do not meet any of the high-risk criteria or because they develop hearing loss after the neonatal period.

Many children become hearing impaired after the neonatal period and therefore are not identified by newborn-screening programs. Often it is not until children are in preschool or kindergarten that further hearing screening takes place. Primary care physicians and pediatricians should be alert to the signs and symptoms of childhood hearing impairment, so that children with hearing impairment who have not been screened formally can be identified as early as possible. Recommendations for postneonatal screening are noted in Figure 629-2.

Figure 629-2 Hearing-assessment algorithm within an office visit. CMV, cytomegalovirus; ENT, ear, nose, and throat.

Modified from Harlor AD Jr, Bower C: Clinical report—hearing assessment in infants and children: recommendations beyond neonatal screening, Pediatrics 124:1252–1263, 2009.)

Identification of Hearing Impairment

The impact of hearing impairment is greatest on an infant who has yet to develop language; therefore, identification, diagnosis, description, and treatment should begin as soon as possible. In general, infants with a prenatal or perinatal history that puts them at risk (see Table 629-2) or those who have failed a formal hearing screening should be monitored closely by an experienced clinical audiologist until a reliable assessment of auditory function has been obtained. Pediatricians should encourage families to cooperate with the follow-up plan. Infants who are born at risk but who were not screened as neonates (often because of transfer from one hospital to another) should have a hearing screening by age 3 mo.

Hearing-impaired infants, who are born at risk or are screened for hearing loss in a neonatal hearing screening program, account for only a portion of hearing-impaired children. Children who are congenitally deaf because of autosomal recessive inheritance or subclinical congenital infection often are not identified until 1-3 yr of age. Usually, those with more severe hearing loss are identified at an earlier age, but identification often occurs later than the age at which intervention can provide an optimal outcome. Children who hear normally develop an extensive language by 3-4 yr of age (Table 629-6) and exhibit behavior reflecting normal auditory function (Table 629-7). Failure to fulfill these criteria should be the reason for an audiologic evaluation. Parents’ concern about hearing and any delayed development of speech and language should alert the pediatrician, because parents’ concern usually precedes formal identification and diagnosis of hearing impairment by 6 mo to 1 yr of age.

Table 629-6 CRITERIA FOR REFERRAL FOR AUDIOLOGIC ASSESSMENT

AGE (mo)	REFERRAL GUIDELINES FOR CHILDREN WITH “SPEECH” DELAY
12	No differentiated babbling or vocal imitation
18	No use of single words
24	Single-word vocabulary of ≤10 words
30	<100 words; no evidence of 2-word combinations; unintelligible
36	<200 words; no use of telegraphic sentences; clarity <50%
48	<600 words; no use of simple sentences clarity ≤80%

From Matkin ND: Early recognition and referral of hearing-impaired children, Pediatr Rev 6:151–156, 1984. Reproduced by permission of Pediatrics.

Table 629-7 GUIDELINES FOR REFERRAL OF CHILDREN WITH SUSPECTED HEARING LOSS

AGE (mo)	NORMAL DEVELOPMENT
0-4	Should startle to loud sounds, quiet to mother’s voice, momentarily cease activity when sound is presented at a conversational level
5-6	Should correctly localize to sound presented in a horizontal plane, begin to imitate sounds in own speech repertoire or at least reciprocally vocalize with an adult
7-12	Should correctly localize to sound presented in any plane Should respond to name, even when spoken quietly
13-15	Should point toward an unexpected sound or to familiar objects or persons when asked
16-18	Should follow simple directions without gestural or other visual cues; can be trained to reach toward an interesting toy at midline when a sound is presented
19-24	Should point to body parts when asked; by 21-24 mo, can be trained to perform play audiometry

From Matkin ND: Early recognition and referral of hearing-impaired children, Pediatr Rev 6:151–156, 1984. Reproduced by permission of Pediatrics.

Clinical Audiologic Evaluation

Even the youngest infants can be evaluated for auditory function. When hearing impairment is suspected in a young child, reliable and valid estimates of auditory function can be obtained. Successful treatment strategies for hearing-impaired children rely on prompt identification and ongoing assessment to define the dimensions of auditory function. Cooperation among the pediatrician and specialists in areas such as audiology, speech and language pathology, education, and child development is necessary to optimize auditory-verbal development. Therapy for hearing-impaired children includes considering and often fitting an amplification device, using an FM system in the classroom, monitoring hearing and auditory skills, counseling parents and families, advising teachers, and dealing with public agencies.

Audiometry

The technique of the audiologic evaluation varies as a function of the age or developmental level of the child, the reason for the evaluation, and the child’s otologic condition or history. An audiogram provides the fundamental description of hearing sensitivity (Fig. 629-3). Hearing thresholds are assessed as a function of frequency using pure tones (sine waves) at octave intervals from 250-8,000 Hz. Earphones typically are used when age-appropriate, and hearing is assessed independently for each ear. Air-conducted signals are presented through earphones (or loudspeakers) and are used to provide information about the sensitivity of the auditory system. These same test sounds can be delivered to the ear through an oscillator that is placed on the head, usually on the mastoid. Such signals are considered bone-conducted because the bones of the skull transmit vibrations as sound energy directly to the inner ear, essentially bypassing the outer and middle ears. In a normal ear, and also in children with SNHL, the air- and bone-conduction thresholds are the same. In those with CHL, the air- and bone-conduction thresholds differ. This is called the air-bone gap, which indicates the amount of hearing loss attributable to dysfunction in the outer and/or middle ear. With mixed hearing loss, both the bone- and air-conduction thresholds are abnormal, and there is an air-bone gap.

Figure 629-3 Audiogram showing bilateral conductive hearing loss.

Speech-Recognition Threshold

Another measure useful for describing auditory function is the speech-recognition threshold (SRT), which is the lowest intensity level at which a score of approximately 50% correct is obtained on a task of recognizing spondee words. Spondee words are 2-syllable words or phrases that have equal stress on each syllable, such as baseball, hotdog, and pancake. Listeners must be familiar with all the words for a valid test result to be obtained. The SRT should correspond to the average of pure-tone thresholds at 500, 1,000, and 2,000 Hz, the pure-tone average (PTA). The SRT is relevant as an indicator of a child’s potential for development and use of speech and language; it also serves as a check of the validity of a test because children with nonorganic hearing loss (malingerers) might show a discrepancy between the PTA and SRT.

The basic battery of hearing tests concludes with an assessment of a child’s ability to understand monosyllabic words when presented at a comfortable listening level. Performance on such word intelligibility tests assists in the differential diagnosis of hearing impairment and provides a measure of how well a child performs when speech is presented at loudness levels similar to those encountered in the environment.

Play Audiometry

Hearing testing is age-dependent. For children at or above the developmental level of a 5-6 yr old, conventional test methods can be used. For children 30 mo to 5 yr of age, play audiometry can be used. Responses in play audiometry usually are conditioned motor activities associated with a game, such as dropping blocks in a bucket, placing rings on a peg, or completing a puzzle. The technique can be used to obtain a reliable audiogram for a preschool child. For those who will not or cannot repeat words clearly for the SRT and word intelligibility tasks, pictures can be used with a pointing response.

Visual Reinforcement Audiometry

For children between the ages of about 6 mo and 30 mo, visual reinforcement audiometry (VRA) commonly is used. In this technique, the child is observed for a head-turning response on activation of an animated (mechanical) toy reinforcer. If infants are properly conditioned, by giving sounds associated with the visual toy cue, VRA can provide reliable estimates of hearing sensitivity for tones and speech sounds. In most applications of VRA, sounds are presented by loudspeakers in a sound field, so no ear-specific information is obtained. Assessment of an infant often is designed to rule out hearing loss that would affect the development of speech and language. Normal sound-field response levels of infants indicate sufficient hearing for this purpose despite the possibility of different hearing levels in the 2 ears.

Behavioral Observation Audiometry

Used as a screening device for infants <5 mo of age, behavioral observation audiometry (BOA) is limited to unconditioned, reflexive responses to complex (not frequency-specific) test sounds such as noise, speech, or music presented using calibrated signals from a loudspeaker or uncalibrated noisemakers. Response levels can vary widely within and among infants and usually do not provide a reliable estimate of sensitivity.

Assessment of a child with suspected hearing loss is not complete until pure-tone hearing thresholds and SRTs (a reliable audiogram) have been obtained in each ear. BOA and VRA in sound-field testing give estimates of hearing responsivity in the better-hearing ear.

Acoustic Immittance Testing

Acoustic immittance testing is a standard part of the clinical audiologic test battery and includes tympanometry. It is a useful objective assessment technique that provides information about the status of the middle ear. Tympanometry can be performed in a physician’s office and is helpful in the diagnosis and management of OM with effusion, a common cause of mild to moderate hearing loss in young children.

Tympanometry

Tympanometry provides a graph of the middle ear’s ability to transmit sound energy (admittance, or compliance) or impede sound energy (impedance) as a function of air pressure in the external ear canal. Because most immittance test instruments measure acoustic admittance, the term admittance is used here. The principles apply to whatever units of measurement are used.

A probe is inserted into the entrance of the external ear canal so that an airtight seal is obtained. The probe varies air pressure, presents a tone, and measures sound pressure level in the ear canal through the probe assembly. The sound pressure measured in the ear canal relative to the known intensity of the probe signal is used to estimate the acoustic admittance of the ear canal and middle-ear system. Admittance can be expressed in a unit called a millimho (mmho) or as a volume of air (mL) with equivalent acoustic admittance. The test is performed so that an estimate can be made of the volume of air enclosed between the probe tip and TM. The acoustic admittance of this volume of air is deducted from the overall admittance measure to obtain a measure of the admittance of the middle-ear system alone. Estimating ear canal volume also has a diagnostic benefit, because an abnormally large value is consistent with the presence of an opening in the TM (perforation or tube).

Once the admittance of the air mass in the external auditory canal has been eliminated, it is assumed that the remaining admittance measure accurately reflects the admittance of the entire middle-ear system. Its value is controlled largely by the dynamics of the TM. Abnormalities of the TM can dictate the shape of tympanograms, thus obscuring abnormalities medial to the TM. In addition, the frequency of the probe tone, the speed and direction of the air pressure change, and the air pressure at which the tympanogram is initiated can all influence the outcome.

When air pressure in the ear canal is equal to that in the middle ear, the middle-ear system is functioning optimally. Therefore, the ear canal pressure at which there is the greatest flow of energy (admittance) should be a reasonable estimate of the air pressure in the middle-ear space. This pressure is determined by finding the maximum or peak admittance on the tympanogram and obtaining its value on the x-axis. The value on the y-axis at the tympanogram peak is an estimate of peak admittance based on admittance tympanometry (Table 629-8). This peak measure sometimes is referred to as static acoustic admittance, even though it is estimated from a dynamic measure (Fig. 632-4A).

Table 629-8 NORMS FOR PEAK (STATIC) ADMITTANCE USING A 226-HZ PROBE TONE FOR CHILDREN AND ADULTS

Tympanometry in Otitis Media with Effusion

Children who have OM with effusion often have reduced peak admittance or high negative tympanometric peak pressures (see Fig. 632-4C). However, in the diagnosis of effusion, the tympanometric measure with the greatest sensitivity and specificity is the shape of the tympanogram rather than its peak pressure or admittance. This shape sometimes is referred to as the tympanometric gradient or width; it measures the degree of roundness or peakedness of the tympanogram. The more rounded the peak (or, in an absent peak, a flat tympanogram), the higher is the probability that an effusion is present (see Fig. 632-4B). It is important to know which instrument is used, because some compute gradient automatically but others do not.

Acoustic Reflex Test

The acoustic reflex test (ART) also is part of the immittance test battery. With a properly functioning middle-ear system, admittance at the TM changes on activation of the stapedius and tensor tympani muscles. In healthy ears, the stapedial reflex occurs after exposure to loud sounds. Admittance instruments are designed to present reflex activating signals (pure tones of various frequencies or noise), either to the same or the contralateral ear, while monitoring admittance. Very small admittance changes that are time-locked to presentations of the signal are considered to be a result of middle-ear muscle reflexes. Admittance changes may be absent when the hearing loss is sufficient to prevent the signal from reaching the loudness level necessary to elicit the reflex or when a middle-ear condition affects the ear’s ability to monitor a small admittance change. Reflexes usually are absent in patients with CHL due to the presence of an abnormal transfer system; thus, the ART is useful in the differential diagnosis of hearing impairment. The ART also is used in the assessment of SNHL and the integrity of the neurologic components of the reflex arc, including cranial nerves VII and VIII.

Auditory Brainstem Response

The ABR test is used to screen newborn hearing, confirm hearing loss in young children, obtain ear-specific information in young children, and test children who cannot, for whatever reason, cooperate with behavioral test methods. It also is important in the diagnosis of auditory dysfunction and of disorders of the auditory nervous system. The ABR test is a far-field recording of minute electrical discharges from numerous neurons. The stimulus, therefore, must be able to cause simultaneous discharge of the large numbers of neurons involved. Stimuli with very rapid onset, such as clicks or tone bursts, must be used. Unfortunately, the rapid onset required to create a measurable ABR also causes energy to be spread in the frequency domain, reducing the frequency-specificity of the response.

The ABR result is not affected by sedation or general anesthesia. Infants and children from about 4 mo to 4 yr of age routinely are sedated to minimize electrical interference caused by muscle activity during testing. The ABR also can be performed in the operating room when a child is anesthetized for another procedure. Children <4 mo of age might sleep for a long enough period of time after feeding to allow an ABR to be done.

The ABR is recorded as 5-7 waves. Waves I, III, and V can be obtained consistently in all age groups; waves II and IV appear less consistently. The latency of each wave (time of occurrence of the wave peak after stimulus onset) increases, and the amplitude decreases with reductions in stimulus intensity or loudness; latency also decreases with increasing age, with the earliest waves reaching mature latency values earlier in life than the later waves.

The ABR test has two major uses in a pediatric setting. As an audiometric test, it provides information on the ability of the peripheral auditory system to transmit information to the auditory nerve and beyond. It also is used in the differential diagnosis or monitoring of central nervous system pathology. For audiometry, the goal is to find the minimum stimulus intensity that yields an observable ABR. Plotting latency versus intensity for various waves also aids in the differential diagnosis of hearing impairment. A major advantage of auditory assessment using the ABR test is that ear-specific threshold estimates can be obtained on infants or patients who are difficult to test. ABR thresholds using click stimuli correlate best with behavioral hearing thresholds in the higher frequencies (1,000-4,000 Hz); responsivity in the low frequencies requires different stimuli (tone bursts or filtered clicks) or the use of masking, neither of which isolates the low-frequency region of the cochlea in all cases, and this can affect interpretation.

The ABR test does not assess “hearing.” It reflects auditory neuronal electric responses that can be correlated to behavioral hearing thresholds, but a normal ABR result only suggests that the auditory system, up to the level of the midbrain, is responsive to the stimulus used. Conversely, a failure to elicit an ABR indicates an impairment of the system’s synchronous response but does not necessarily mean that there is no “hearing.” The behavioral response to sound sometimes is normal when no ABR can be elicited, such as in neurologic demyelinating disease. The ABR test may be used to infer whether and at what level of the auditory system impairment exists.

Hearing losses that are sudden, progressive, or unilateral are indications for ABR testing. Although it is believed that the different waves of the ABR reflect activity in increasingly rostral levels of the auditory system, the neural generators of the response have not been precisely determined. Each ABR wave beyond the earliest waves probably is the result of neural firing at many levels of the system, and each level of the system probably contributes to several ABR waves. High-intensity click stimuli are used for the neurologic application. The morphology of the response and wave and interwave latencies are examined in respect to age-appropriate forms. Delayed or missing waves in the ABR result often have diagnostic significance.

The ABR and other electrical responses are extremely complex and difficult to interpret. A number of factors, including instrumentation design and settings, environment, degree and configuration of hearing loss, and patients’ characteristics, can influence the quality of the recording. Therefore, testing and interpretation of electrophysiologic activity as it possibly relates to hearing should be carried out by trained audiologists to avoid the risk that unreliable or erroneous conclusions will affect a patient’s care.

Otoacoustic Emissions

During normal hearing, OAEs originate from the hair cells in the cochlea and are detected by sensitive amplifying processes. They travel from the cochlea through the middle ear to the external auditory canal, where they can be detected using miniature microphones. Transient evoked OAEs (TEOAEs) may be used to check the integrity of the cochlea. In the neonatal period, detection of OAEs can be accomplished during natural sleep, and TEOAEs can be used as screening tests in infants and children for hearing at the 30 dB level of hearing loss. They are less time-consuming and elaborate than ABR and are more sensitive than behavioral tests in young children. TEOAEs are reduced or absent owing to various dysfunctions in the middle and inner ears. They are absent in patients with >30 dB of hearing loss and are not used to determine hearing threshold; rather, they provide a screen for whether hearing is present at >30-40 dB. Diseases such as OM or congenitally abnormal middle ear structures reduce the transfer of TEOAEs and may incorrectly indicate a cochlear hearing disorder. If a hearing loss is suspected based on the absence of OAE, the ears should be examined for evidence of pathology, and then ABR testing should be used for confirmation and identification of the type, degree, and laterality of hearing loss.

Acoustic Reflectometry

In acoustic reflectometry, a hand-held instrument is placed next to the opening of a child’s ear canal and 80-dB sound is delivered that varies in frequency from 2,000-4,500 Hz in a 100-msec period. The instrument measures the total level of reflected and transmitted sound. Some physicians have found this device useful to help gauge the presence or absence of middle-ear fluid, and a commercial version is marketed to parents as a way to monitor ear fluid. The instrument does not provide any information about hearing; if the presence of chronic fluid is suggested, audiometric evaluation should be obtained.

Treatment

With the use of universal hearing screening in the majority of states within the USA, the early diagnosis and treatment of children with hearing loss is common. Testing for hearing loss is possible even in very young children, and it should be done if parents suspect a problem. Any child with a known risk factor for hearing loss should be evaluated in the 1st 6 mo of life.

Once a hearing loss is identified, a full developmental and speech and language evaluation is needed. Counseling and involvement of parents are required in all stages of the evaluation and treatment or rehabilitation. A conductive hearing loss often can be corrected through treatment of a middle-ear effusion (i.e., ear tube placement) or surgical correction of the abnormal sound-conducting mechanism. Children with SNHL should be evaluated for possible hearing aid use by a pediatric audiologist. Hearing aids may be fitted for children as young as 2 mo of age. Compelling evidence from the hearing screening program in Colorado shows that identification and amplification before age 6 mo makes a very significant difference in the speech and language abilities of affected children, compared with cases identified and amplified after the age of 6 mo. In these children, repeat audiologic testing is needed to reliably identify the degree of hearing loss and to fine tune the use of hearing aids.

Infants and young children with profound congenital or prelingual onset of deafness have benefited from multichannel cochlear implants (see Fig. 629-4). These implants bypass injury to the organ of Corti and provide neural stimulation by way of an external microphone and a signal processor that digitizes auditory stimuli into digital radiofrequency impulses. Cochlear implantation before age 2 yr (and even 1 yr) improves hearing and speech, enabling more than 90% of children to be in mainstream education. Most develop age-appropriate auditory perception and oral language skills.