Pediatric Annals

Evaluation of Hearing Loss in Infants and Young Children

John Jacobson, PhD; Claire Jacobson, AuD

Abstract

Hearing loss is an insidious handicap that may affect people of all social, ethnic, and economic classes. The limitations that hearing loss imposes are global in nature and often poorly understood. A missed diagnosis or failure to identify auditory deficits, particularly early in life, may lead to ridicule, and a hearing impaired child may be inappropriately perceived as emotionally and mentally challenged. Frustration, dejection, poor self-esteem, and social isolation are all common outcomes for these children when the detection of hearing impairment is delayed or misdiagnosed.

For infants or very young children who are unable to respond verbally, a speech detection threshold may be established. This technique is less accurate but provides a general level of awareness to speech. The examiner typically will look for behavioral changes such as head turning, cessation of activity, eye widening, and so on to establish a level of speech detection. Typically between ages 2 to 3, an SRT can be obtained in a soundfield environment. As the child reaches age 3 to 4, an SRT can be established under headphones providing ear-specific threshold levels.

The examiner will usually compare the SRT with the three-frequency average of AC thresholds at 500, 1,000, and 2,000 hertz as a measure of test reliability; the two results generally should be within 5 dB of one another. For example, an SRT of 25 dB should be associated with a three-frequency average of 20 to 30 dB. When discrepancies of greater than 5 and certainly 10 dB exist, re-instruction and repeat testing should be considered. Differences between the three-frequency average and the SRT are also a caveat for the possibility of a nonorganic (functional) hearing loss.

Speech recognition is the ability of a patient to correctly repeat a list of phonetically balanced (PB) words that represent common usage of sounds in everyday English. Special word lists have been designed for children and have been standardized for reliability. For children who speak English as a second language, standardized word lists also are available in several foreign languages. Speech recognition often is referred to in the medical community as speech discrimination, a misnomer because the patient is not asked to discriminate between words but to repeat the spoken word.

To achieve the maximum wordrecognition score (known as PB max), PB words usually are presented to children at a level 30 dB above the SRT. Another common method of presentation level is to determine the patient's most comfortable listening level, a more difficult task in the infant or young child. Word lists may be presented via AC, BC, or soundfield speakers in a quiet listening environment. Results are reported as the percentage of correct responses. Children with normal hearing and those with conductive middle ear pathology routinely score well on SR tests, usually greater than 90%, because conductive hearing loss does not distort auditory processing. Those children who present with sensory (cochlear) hearing deficits, depending on the degree of loss and the audiometrìe configuration, will tend to have more difficulty accurately responding to word lists and generally have poorer scores.

Variables associated with speech testing include test environment, live voice versus recorded word lists, accent, dialect, or other articulation patterns of the speaker, and male or female voices.

MASKING

During BC testing, the vibrator stimulates inner ear fluids in both ears, making it impossible to determine which ear is responding. In such a situation, masking is employed to isolate the ears from one another. This process entails the introduction of noise (usually white noise, narrow band noise, or speech noise) into the nontest ear during BC testing when there is a substantial difference…

Hearing loss is an insidious handicap that may affect people of all social, ethnic, and economic classes. The limitations that hearing loss imposes are global in nature and often poorly understood. A missed diagnosis or failure to identify auditory deficits, particularly early in life, may lead to ridicule, and a hearing impaired child may be inappropriately perceived as emotionally and mentally challenged. Frustration, dejection, poor self-esteem, and social isolation are all common outcomes for these children when the detection of hearing impairment is delayed or misdiagnosed.

Identification, diagnosis, and management of hearing impairment in children may be challenging but often is equally rewarding. Success is dependent on the aggressive participation of the medical community and allied health professionals to provide a comprehensive service to the families of children with hearing loss. Experience has shown that no one professional group can successfully provide comprehensive hearing services. It is only when the medical community is willing to synthesize resources that the needs of the hearing impaired infant or child can be met. Therefore, it is the collective effort of pediatricians, otolaryngologists, audiologists, speech and language pathologists, and educational specialists to share these responsibilities.

Table

TABLE 1.A Classification System for Degree of Hearing Loss

TABLE 1.

A Classification System for Degree of Hearing Loss

In general, to identify the hearing status of infants and young children, a comprehensive test battery is required that combines both behavioral and electrophysiological measures of auditory sensitivity. This article addressess the various methods of auditory evaluation from birth to, and including, the preschool-age child. Test protocols, age-appropriate testing, and recommended guidelines, sponsored by both national and professional academies, will be presented.

LANGUAGE ACQUISITION

Hearing loss, whether congenital or acquired, represents a significant barrier to linguistic acquisition essential for subsequent psychosocial, educational, and vocational development. Yoshinaga-Itano and colleagues' demonstrated that there is a critical period for speech and language development that begins within the first 6 months of life and directly affects speech perception and cognitive skills. The presence of permanent (ie, congenital) or chronic transient (ie, middle ear effusion) hearing loss during this period has the potential to compromise speech and language acquisition. Although the severity of language sequelae secondary to otitis media remains controversial, the American Academy of Audiology cites substantial evidence that the age of onset as well as the nature, degree, and audiometric pattern of hearing loss in children with otitis media will determine their risk for developing communication and learning disorders.2

Thus, when hearing loss is present in the earliest stages of life, identification, intervention, and follow-up services, including medical or surgical treatment, and rehabilitative measures of language stimulation are necessary to minimize the detrimental affects of hearing loss. Once hearing loss is identified, national, state, and local agencies are able to provide available funding and support personnel to coordinate longitudinal efforts and services.

HEARING LOSS CLASSIFICATION

Audiological evaluation is essential to determine the degree and type of hearing loss in a patient. Hearing loss generally is characterized as conductive (pathology involving the ear canal, tympanic membrane, or middle ear including the ossicles), sensorineural (involving the cochlea or auditory nerve), or mixed. In rare cases, the loss may be central or functional (nonorganic). The protocols that are applied to diagnosis are related to the age of the patient; obviously, younger patients require more objective tests to assess hearing status appropriately.

The degree of hearing loss is described in decibels (dB), a unit of intensity or loudness (intensity's psychological equivalent). Levels of hearing handicaps are usually described as the degree of hearing loss in decibels (Table 1). If individual frequencies have been tested, the loss is usually described verbally or graphically at each frequency. Hearing handicap also may be described as a percentage of normal hearing based on formulas that involve the use of various pure tone frequency threshold measures. Although there is no uniformly accepted system, the American Medical Association, the American Academy of Otolaryngology - Head & Neck Surgery, and the American Speech and Hearing Association all have formulas for calculation of hearing handicap. In young children, 15 dB has been adopted as the "low fence" for normal hearing sensitivity.3

INCIDENCE OF HEARING LOSS

The incidence of severe-to-profound hearing loss in newborns has been estimated in the range of 1 to 6 per 1,000 live births.4,5 Comparatively, this degree of hearing loss is 20 times more prevalent than other birth defects, including phenylketonuria (PKU), sickle cell anemia, and hypothyroidism, for which screening at birth has long been routinely performed. If infants with lesser degrees of permanent sensory hearing loss were considered, these numbers would escalate dramatically.

Further, the incidence of hearing impairment is unparalleled when newborns at risk for progressive or delayedonset hearing loss are taken into account. It has been estimated that approximately 10% of all newborns are at risk for developmental disability, including hearing loss.6 Within this population, 30 to 50 out of every 1 ,000 live births are severely hearing impaired.7-8

NEWBORN HEARING SCREENING

Before the adoption of universal newborn hearing screening programs, the detection of hearing loss in the wellbaby population was typically delayed 30 to 36 months by the time appropriate testing was concluded and the diagnosis confirmed. Children who presented with milder degrees of transient, permanent, or unilateral hearing impairment typically were not identified until an initial school screening.

Acknowledging that early detection and appropriate management of hearing loss in these children is necessary for normal speech, language, and educational development, the Joint Committee on Infant Hearing (JCIH) was established to promote early detection of hearing loss in infants and children.9 The committee currently consists of representative members of the American Academy of Pediatrics (AAP), the American Academy of Otolaryngology - Head & Neck Surgery, the American Academy of Audiology, the American Speech and Hearing Association, and directors of speech and hearing programs in state health and welfare agencies. Based on decades of data in the medical literature, JCIH has identified a group of indicators that suggest a neonate may be at risk for hearing impairment. Risk indicators have evolved since the inception of JCIH and now address two groups of children: those associated with neonatal hearing impairment (Sidebar 1), and those associated with progressive or delayed-onset hearing loss (Sidebar 2, see page 814). However, it is important to note that fewer than 50% of all children with hearing loss have a known risk indicator.

For clinical health care providers, it is essential not to underestimate the role that parents and other caregivers play in the detection of hearing loss. In 1999, an AAP task force concluded that parents were as much as 12 months ahead physicians in identifying their child's hearing loss.10 Subsequently, parental concern was included as a risk indicator in the 2000 JCIH statement and in its 2003 guideline. The AAP suggests that when parents voice concerns about possible hearing loss, "the pediatrician needs to assume that such is true until the child has been evaluated objectively."" To this end, formal audiological testing can either confirm or alleviate parental concern.

Unfortunately, the application high-risk criteria was insufficient to identify a significant proportion infants with hearing loss. In addition, until recently, hearing loss could not be detected accurately before age 2 or 3 due to clinical variables such as limitations in technology and the noncompliant behavior of infants and very young children. However, with rare exceptions, identification of hearing impairment can now be accomplished precisely with objective technology shortly after birth. Various techniques to identify hearing loss are currently available and advocated by such organizations as the National Institutes of Health (NIH) and the AAP.10"13

The 1993 NIH Consensus Statement on Early Identification of Hearing Impairment in Infants and Young Children and the JCIH 1994 Position Statement were the initial catalysts for implementation of newborn hearing screening within the first three months of life.12,14 Various modifications have since been adopted by national and state-affiliated organizations for the purposes of enhance existing protocol For example, AAP guidelines recommend universal detection of hearing impairment by age 3 months and initiation of intervention by age 6 months. Today, universal newborn hearing screening programs are mandated in 38 states and the District of Columbia, as well as in a variety of locations throughout Canada and Europe.

Newborn hearing screening uses objective testing of the auditory system to identify infants at risk for hearing loss. Currently, the objective measures used for newborn hearing screening are otoacoustic emissions (OAE) and the automated auditory brainstem response, which will be discussed in detail below. A newborn generally is considered to have hearing loss when a unilateral or bilateral loss of mild (30 dB hearing loss [HL]) or greater is identified within the frequency region important for speech recognition and comprehension.

For babies who pass the hearing screening, the next steps usually include documenting the results of the screening in the infant's record and in the discharge summary. In addition, the parents should be provided with the screening results and educational materials, including information on developmental milestones for communication and signs of hearing loss in young children. If a newborn's hearing screening is missed before discharge, the hospital staff should make an attempt to schedule an appointment for an outpatient screening or a reasonable effort to contact the family and schedule the hearing screening after discharge.

Babies who do not pass ("refer") following the initial hearing screening may be re- screened before discharge if feasible. This helps keep the percentage of false-positive results below the goal of approximately 4%, though ultimately this rate also depends on such factors as the screening protocol selected and the training and experience of the screeners. Re-screening also reduces the need for a follow-up outpatient screening. It has been found that most infants (80% to 90%) who fail the first screening will pass when they are re-screened. If rescreening before discharge is not possible, or passing results are not obtained, the baby will be referred for re-screening to take place after discharge.

Prompt follow-up is important for those infants who do not pass the initial hearing screening and for those infants who fail two newborn hearing screenings. Infants who have failed two hearing screenings are considered to have hearing impairment. Such children should be referred to an early intervention program for a confirmatory diagnostic audiological evaluation, which may include either a repeat screening study or more formal assessment by auditory brainstem response (ABR). Based on the results and on any concerns regarding the infant's development, the audiologist may determine that a multidisciplinary evaluation is warranted. All newborn hearing screening programs are required to submit screening data to the state department of health on a regular basis.

OBJECTIVE AUDIOMETRY

The three basic objective measures that are routinely incorporated into the assessment of infants and young children are otoacoustic emissions (OAEs), auditory evoked potentials (AEP), including ABR, and immittance audiometry (IA), also known as impedance audiometry or tympanometry. Although any of these studies yields valuable information, selecting the most appropriate combination of studies consistently results in the most reliable and valid test outcome.'5 Pediatricians should have a basic familiarity with objective audiometrie testing because most babies in the US will be undergoing these studies as a part of their evaluation in the newborn nursery.

Otoacoustic Emissions

The application of OAEs is the newest noninvasive technology in the general test battery. OAEs offer an objective clinical technique for evaluating the integrity and sensitivity of the cochlea, and they indirectly reflect the status of the middle ear. OAEs are considered a by-product of normal sensory transduction and represent frequency dispersion in response to either transient or continuously presented stimuli. These responses to stimuli are detected as measurable forms of emitted acoustic energy within the external auditory canal (EAC). Outer hair cell movement apparently is the source of this generated mechanical energy that is propagated in an outward direction via the middle ear and tympanic membrane, representing a reversal of normal acoustic transduction. OAEs are recorded with a miniature microphone seated in the EAC in response to a click or tonal stimulation.

Evoked OAEs are recorded as the result of the introduction of low to moderate auditory signals generated within the EAC. Two types of evoked OAEs are most often described by the stimulus characteristics used to elicit the response. Transient evoked OAE testing (TEOAE) applies a brief transient click stimulus to elicit the hair cell response. Distortion product OAEs (DPOAEs) are recorded from the hair cells when two pure tones of different frequency are presented to the EAC simultaneously. Because TEOAEs use a click stimulus to elicit the response, they lack frequency specificity and therefore represent a broad frequency region of the cochlea, usually between 1,500 to 6,000 hertz. In contrast, DPOAEs use specific frequency stimuli to measure specific regions of the cochlea, most commonly octave band frequencies between 1,000 and 8,000 hertz (Figure 1, see page 816).

DPOAEs are always present in a nonpathological ear with normal auditory sensitivity. As a result, this test is an excellent method of estimating auditory sensitivity as a function of frequency, the objective analogy to the pure tone behavioral audiogram. The primary limiting clinical factor is that the DPOAE is usually not recorded at frequency regions that show hearing loss of greater than mild-to-moderate (greater than 30 dB HL) severity, or at frequencies where noise is a interfering factor (usually less than 1,000 hertz). In the past, the presence of ambient and physiological noise has also been a factor, but recent technology has overcome the problem.16

OAEs have been used widely and provide a quick assessment of auditory status in infants and children. The test is also helpful to the identification of infants and children with auditory neuropathy, a condition where OAEs are present but more rostral regions of the auditory pathways do not function normally. This condition frequently results in auditory processing and cognitive deficits and may be an indication for cochlear transplantation.

Auditory evoked potentials

AEPs are a group of objective electrophysiologic responses that assess auditory function and neurologic integrity. AEPs consist of a continuum of neuroelectric events that are generated along the entire length of the auditory pathway. They are described by their latency, anatomic origin, stimulusresponse relationship, or electrode placement. Most often they are clinically described by their latency from stimulus onset: early (0 to 15 milliseconds), middle (15 to 100 milliseconds), and late (100 to 500 milliseconds).

Of the early latency potentials, ABR is the most clinically useful in detection of hearing loss in newborns, infants, and other difficult-to-test children.17 ABR recordings have been observed in premature infants as early as the 26th week of gestation and may be obtained in high-risk infants shortly after birth.18 In longitudinal studies, ABR has reflected the maturation of the auditory system, with central conduction time decreasing during the first 12 to 18 months, when it reaches adult levels. Automated ABR has been applied successfully in universal newborn hearing screening programs. In addition, operating characteristics of ABR exceed any other hearing screening technique, and the procedure can be accomplished in an efficient and economic manner.

The clinical development of ABR testing has focused on two principal areas of application: the evaluation and diagnosis of the peripheral auditory system and related pathology, and the determination of the neural integrity of the acoustic nerve and brainstem pathway. ABR provides a valid estimate of auditory sensitivity based on the threshold of response and has demonstrated good correlation with behavioral measures. Because activity will compromise the recorded response, sedation typically is required to ensure a quiet and restful infant or child. Chloral hydrate is the drug of choice and allows sufficient conscious sedation to accomplish threshold measures.

Figure 1. A distortion product otoacoustic emission was obtained from the right ear of a 3-year-old child. Traces represent frequency bands from 500 to 6,000 hertz. Circles represent the distortion product, and the squares represent the noise floor. When less than 3 dB separates the noise floor and distortion product, no response is present (eg, 500 hertz).The remaining frequency recordings produced robust distortion product responses.

Figure 1. A distortion product otoacoustic emission was obtained from the right ear of a 3-year-old child. Traces represent frequency bands from 500 to 6,000 hertz. Circles represent the distortion product, and the squares represent the noise floor. When less than 3 dB separates the noise floor and distortion product, no response is present (eg, 500 hertz).The remaining frequency recordings produced robust distortion product responses.

Figure 2. Auditory brainstem response thresholds were recorded bilaterally from an 18-month-old infant using click stimuli.Traces were obtained at 70, 50, 35 and 20 dB normal hearing level. Traces at 20 dB were replicated for reliability. Typically, wave V is the benchmark used for the presence or absence of an auditory brainstem response. Reliable threshold measures can be recorded from newborns shortly after birth.

Figure 2. Auditory brainstem response thresholds were recorded bilaterally from an 18-month-old infant using click stimuli.Traces were obtained at 70, 50, 35 and 20 dB normal hearing level. Traces at 20 dB were replicated for reliability. Typically, wave V is the benchmark used for the presence or absence of an auditory brainstem response. Reliable threshold measures can be recorded from newborns shortly after birth.

ABR potentials are evoked by a brief click or tone pip transmitted from an insert earphone or headphone. Click stimuli are most commonly used but have a broad frequency range (1,000 to 6,000 hertz) and provide little information about hearing in the lower frequency range. Tone pips or bursts can be used in place of clicks to obtain more frequency specificity if necessary. The elicited waveform response is measured by surface electrodes typically placed at the vertex of the scalp and ear lobes. The recorded response is both time-locked to the stimulus and averaged in order to reduce unwanted noise. The amplitude of the response is plotted against time, producing waveform peaks labeled IVII that are similar to those of an electroencephalogram (Figure 2). ABR waves I and II correspond to true action potentials, while later waves reflect postsynaptic activity in major brainstem auditory centers that concomitantly contribute to waveform peaks and troughs. Wave V usually is used to establish auditory thresholds.

In children, the electrophysiological threshold is elevated about 5 to 10 dB compared with psychophysical estimates. In addition, while ABR provides information regarding auditory function and hearing sensitivity, it is not a substitute for a formal hearing evaluation. Results should be used in conjunction with behavioral audiometry whenever possible.

Immittance Audiometry

Immittance audiometry (IA) is an objective clinical means of measuring the physical volume of the EAC and the compliance of the middle ear system. Most immittance devices also allow measurement of the acoustic reflex contraction, a bilateral phenomenon that is elicited by the introduction of a loud acoustic stimulation.

Tympanometry has been used successfully by the medical and health-professional community to establish the presence of effusion or retraction, particularly in patients whose eardrums are poorly visualized due to abnormal EACs or excessive activity. There are typically three basic patterns of measurable middle ear compliance: normal (peaked or type A), noncompliant (flat or type B), and retraction (negative pressure or type C). In the presence of middle ear pathology, tympanometry is compromised and acoustic reflexes are usually absent. When LA results in normal tympanometry without acoustic reflex measures in the presence of decreased hearing sensitivity, sensory hearing loss is a primary consideration, and other additional tests may be required to confirm the degree and type of hearing loss. An unusually large EAC volume associated with a flat tympanogram also may support the diagnosis of a perforated tympanic membrane.

Another advantage of IA is the ability to longitudinally follow the progress of effusion by repeating tympanograms on a routine basis. It is not uncommon to watch tympanometry change from totally noncompliance middle ear function to tympanic membrane retraction to normal compliance in a relatively short period of time.

In the presence of a patent tympanostomy tube, one would anticipate a flat tympanogram with a larger than normal canal volume. Furthermore, when a tube is in place and a smaller-than-expected volume is obtained, the tube may be occluded yielding a normal tympanogram; a flat tympanogram in this situation suggests recurrent middle ear disease. Tympanometry is therefore an excellent adjunct to pneumatic otoscopy and may be used to confirm pathologic findings.

TESTING BY OPERANT CONDITIONING

In infants and toddlers of sufficient developmental age, operant conditioning techniques can be used to obtain a more frequency specific audiogram than that using OAE or ABR.

Behavioral Observation Audiometry

Used in the very young infant (birth to age 6 months), behavioral observation audiometry (BOA) has been used to establish estimated levels of hearing. BOA does not require conditioning or reinforcement and is the most common behavioral technique used in infants younger than 6 months. The BOA technique involves controlled observation of an infant's overt behavioral response to sound stimulation under structured conditions. Infant responses include actions such as the auro-palpebral reflex, startle and arousal responses, and rudimentary head turning. Infant responses are initiated by the introduction of a loud sound and the examiner observes changes in the behavior of the infant. A normal developing infant age 4 to 6 months will begin to turn toward the sound source and response to a lower intensity level of stimulation. By the time an infant reaches age 12 months, minimal response for speech awareness is approximately 15 dB HL.3

Unfortunately, this procedure has several limitations, including the need for a high-intensity stimulus to evoke a response, as well as examiner bias. This results in a high number of false-positive and false-negative test results. Therefore, due to the limitations associated with BOA, this technique is best used in conjunction with objective tests that do not require cooperation or participation on the part of the infant.

Visual Reinforcement Audiometry

For infants and young children typically older than 1 2 months but younger than 3 to 4 years (depending on maturity), visual reinforcement audiometry (VRA) is most appropriate. This technique relies on the application of systematic reinforcement of behavioral responses. Simply, the acoustic stimulus serves to prompt a response from the infant that will result in some form of positive reinforcement.

Typically, VRA uses an animated lighted toy as a reinforcement tool once a correct behavioral response is observed. Any number of stimuli (speech, tones, or noise) can be presented, typically through sound field speakers. Animated toys are placed so that when a child appropriately responds to sound by turning in the direction of the signal, the animated toy is lit to reinforce the behavioral response. The infant or child usually is seated in the lap of the parent or caregiver facing at 45 degrees azimuth equidistant and approximately 3 feet from each speaker. Stimulation of a third centering toy is intended to turn the head back to a neutral position between the two speakers following a positive response. Infants and young children tend to naturally seek the sound source by localizing to the stimuli. When immediate visual reinforcement is applied, there is a relatively short learning curve and accurate frequency-specific thresholds can be gained from infants younger than 1.

Figure 3.This basic audiogram model uses two graphs, one for each ear. Frequency is on the vertical axis, and intensity is on the horizontal axis. Data summary includes the three-frequency average, speech reception thresholds, and speech recognition scores (PBM). Threshold measures are marked by a standardized symbol system.

Figure 3.This basic audiogram model uses two graphs, one for each ear. Frequency is on the vertical axis, and intensity is on the horizontal axis. Data summary includes the three-frequency average, speech reception thresholds, and speech recognition scores (PBM). Threshold measures are marked by a standardized symbol system.

Play Audiometry

As young children, usually age 2 and older, become more attentive and develop longer attention spans, additional operant conditioning techniques become feasible. The most commonly used measure is play audiometry, which requires the child to respond to sound by performing a task such as dropping blocks, placing a ring on a peg, or stacking blocks. The learning curve is brief, and children enjoy the task sufficiently to establish auditory thresholds. The technique may be performed with either soundfield speakers or headphones.

Pure Tone and Speech Audiometry

Older children and adolescents usually are tested using a combination of pure tone and speech audiometry. In pure tone assessment, the patient is asked to respond to signals generated by a calibrated audiometer. There are two types of pure tone assessment, air conduction (AC) and bone conduction (BC) measures. Hearing thresholds are determined at various frequencies and are reported in decibels hearing level (dB HL). In AC testing, the tone is delivered through either headphones or insert transducers, while a bone vibrator, usually placed over the mastoid process, is used for BC testing.

Because AC testing is generated through a transducer placed either in or around the EAC, AC is a measure of the entire auditory pathway. Thus, any auditory pathology will be reflected in the established thresholds. When testing infants or young children who will not allow the use of headphones, AC testing is conducted via two calibrated speakers and is termed soundfield testing. The primary limitation associated with soundfield testing is that test results reflect only the better-hearing ear; it is virtually impossible to differentiate hearing levels between ears. With a relatively cooperative child, both pure tone AC and speech testing can be conducted using a variety of operant conditioning techniques.

The use of a BC transducer stimulates fluids within the inner ear by vibrating the skull - in theory, bypassing any contribution of mechanical transduction by the middle ear. By doing so, BC testing helps to differentiate conductive hearing loss from sensory hearing loss. BC testing is considered an estimation of cochlear function reserve only. BC testing has several limitations, including audiometer output, potential vibrotactile responses from the patient at high intensities, especially when employing low frequency stimulation (principally at 250 and 500 hertz), and interaural attenuation (the crossing of sound from one ear to the other; see also the section on masking later in this article). The recent introduction of insert transducers has increased the levels at which interaural attenuation occurs, but tactile stimulation remains problematic. Regardless of whether AC or BC is used, several testing variables must be considered. These include ambient room noise levels, technical skill of the examiner, cooperation of the patient, and realistic goals within specified time limitations.

THEAUDIOGRAM

The results of a hearing test are most easily described graphically by means of an audiogram (Figure 3. The audiogram displays auditory threshold in decibels as a function of frequency. Frequency is measured in hertz and is most often associated with pitch. Typically, thresholds are measured at octave band frequencies of 250, 500, 1,000, 2,000, 4,000 and 8,000 hertz. When necessary, the examiner may also test at halfoctaves (1,500, 3,000 and 6,000 hertz). In evaluating young children, octave frequencies usually are adequate to estimate auditory sensitivity.

Figure 4. Audiogram results from a 4-year-old child with unilateral effusion.The right ear illustrates a rising audiometrie configuration consistent with conductive pathology. Tympanometry produced a noncompliant middle ear function in the right ear and normal compliance in the normal left ear. Crossed and uncrossed acoustic reflexes are absent, with the exception of an uncrossed reflex in the normal left ear.

Figure 4. Audiogram results from a 4-year-old child with unilateral effusion.The right ear illustrates a rising audiometrie configuration consistent with conductive pathology. Tympanometry produced a noncompliant middle ear function in the right ear and normal compliance in the normal left ear. Crossed and uncrossed acoustic reflexes are absent, with the exception of an uncrossed reflex in the normal left ear.

Graphically, pure tone audiometry results are represented by a symbol system. The American Speech and Hearing Association has recommended a system that is universally recognized.19 Other symbol systems have been adopted by many clinics; one such system, depicting unilateral middle ear effusion, is illustrated in Figure 4.20 Two graphs are created representing the right and left ears. Air conduction thresholds are represented by circles; when masking is employed, the circles are filled. Bone conduction thresholds are represented by triangles and also filled for masking. For infants and young children who will not allow the use of insert transducer or headphones, soundfield thresholds are recorded on a single graph using the symbol "S" to denote the threshold at each frequency.

SPEECH AUDIOMETRY

In addition to pure tone testing, speech reception threshold (SRT) measures and speech recognition (SR) scores are a routine component of a basic auditory test battery. Speech thresholds are defined as the lowest (softest) level in dB at which a patient can repeat approximately 50 percent of a series of spondaic words. Spondees are two-syllable words with equal stress on both syllables (eg, baseball, doorbell, playground) that are used to represent a normal sampling of English speech sounds. When young children will not respond verbally on the test, pointing to objects, pictures, and body parts can be introduced as a substitute measure. Gaining the confidence of the child is critically important, and the examiner must also keep in mind the variables inherent in testing infants and young children, including short attention spans and limited vocabularies.

Table

TABLE 2.Hearing Assessment: Test Protocol by Age

TABLE 2.

Hearing Assessment: Test Protocol by Age

For infants or very young children who are unable to respond verbally, a speech detection threshold may be established. This technique is less accurate but provides a general level of awareness to speech. The examiner typically will look for behavioral changes such as head turning, cessation of activity, eye widening, and so on to establish a level of speech detection. Typically between ages 2 to 3, an SRT can be obtained in a soundfield environment. As the child reaches age 3 to 4, an SRT can be established under headphones providing ear-specific threshold levels.

The examiner will usually compare the SRT with the three-frequency average of AC thresholds at 500, 1,000, and 2,000 hertz as a measure of test reliability; the two results generally should be within 5 dB of one another. For example, an SRT of 25 dB should be associated with a three-frequency average of 20 to 30 dB. When discrepancies of greater than 5 and certainly 10 dB exist, re-instruction and repeat testing should be considered. Differences between the three-frequency average and the SRT are also a caveat for the possibility of a nonorganic (functional) hearing loss.

Speech recognition is the ability of a patient to correctly repeat a list of phonetically balanced (PB) words that represent common usage of sounds in everyday English. Special word lists have been designed for children and have been standardized for reliability. For children who speak English as a second language, standardized word lists also are available in several foreign languages. Speech recognition often is referred to in the medical community as speech discrimination, a misnomer because the patient is not asked to discriminate between words but to repeat the spoken word.

To achieve the maximum wordrecognition score (known as PB max), PB words usually are presented to children at a level 30 dB above the SRT. Another common method of presentation level is to determine the patient's most comfortable listening level, a more difficult task in the infant or young child. Word lists may be presented via AC, BC, or soundfield speakers in a quiet listening environment. Results are reported as the percentage of correct responses. Children with normal hearing and those with conductive middle ear pathology routinely score well on SR tests, usually greater than 90%, because conductive hearing loss does not distort auditory processing. Those children who present with sensory (cochlear) hearing deficits, depending on the degree of loss and the audiometrìe configuration, will tend to have more difficulty accurately responding to word lists and generally have poorer scores.

Variables associated with speech testing include test environment, live voice versus recorded word lists, accent, dialect, or other articulation patterns of the speaker, and male or female voices.

MASKING

During BC testing, the vibrator stimulates inner ear fluids in both ears, making it impossible to determine which ear is responding. In such a situation, masking is employed to isolate the ears from one another. This process entails the introduction of noise (usually white noise, narrow band noise, or speech noise) into the nontest ear during BC testing when there is a substantial difference in auditory thresholds of between ears during AC testing. With the introduction of insert earphones, dependence on masking is now limited to those cases in which the attenuation between ears exceeds 70 dB. In infants and young children, masking may not always be possible, and when not applied, the BC responses should be considered a unilateral response from the better inner ear regardless of where the bone vibrator is placed on the skull.

Two variables associated with masking are the "occlusion effect," an increase in perceived loudness due to a reduction in ambient noise reaching the eardrum, and the "masking dilemma," a condition that exists when there is a moderate or greater bilateral hearing loss. Audiologists are trained in the use of masking paradigms during testing and when its use is deemed necessary.

SUMMARY

The identification, assessment, and management of hearing impairment in the pediatric population can be a challenging endeavor. Nevertheless, newer technology, improved techniques, and the cooperative efforts of various professional organizations and their constituencies have made significant strides toward achieving this goal. As more precise objective technologies are introduced, there will be a tendency to rely more heavily on their application. Both IA and OAEs have already made significant impact in pediatric practices because of their ease and simplicity. Within a short period of time, trained technical staff can become proficient in their usage and test interpretation. Their application in conjunction with basic audiometry can provide a global picture of auditory status (Table 2). However, it is critical to recall that the basic building block of auditory assessment is the audiogram, a true measure of behavioral threshold sensitivity. Therefore, when test results suggest hearing impairment, appropriate audiological referral to the will ensure continuity of services and provision of rehabilitative measures. It is equally essential for primary care providers to understand the therapeutic needs of their patients and to manage and coordinate the medical aspects of the infant or child when hearing loss is suspected.

REFERENCES

1. Yoshinaga-Itano C, Sedey AL, Coulter DK, Mehl AL. Language of early-and later-identified children with hearing loss. Pediatrics. 1998; 102(5): 1 161-1 171.

2. American Academy of Audiology. Audiologic guidelines for die diagnosis and treatment of otitis media in children. Available at: http:// www.audiology.org/professional/positions/ otitis.php. Accessed November 17, 2004.

3. Northern JL, Down MP. Hearing in Children. 4th ed. Baltimore, MD: Williams & Wilkins; 1991.

4. Bachmann KR, Arvedson JC. Early identification and intervention for children who are hearing impaired. Pediatr Rev. 1998; 19(5): 155-165.

5. Northern JL, Hayes D. Universal screening for infant hearing impairment: necessary, beneficial and justifiable. Bull Am Acad Audiol. 1994:6:10-13.

6. Mahoney T. High risk hearing screening of large general newborn populations. Semin Hear. 1984;5:25-37.

7. Galambos R, Hicks G, Wilson M. Hearing loss in graduates of a tertiary intensive care nursery. Ear Hear. 1982;3(2):87-90.

8. Galambos R, Hick GE, Wilson MJ. The auditory brainstem response reliably predicts hearing loss in graduates of tertiary intensive care nursery. Ear Hear. 1984;5(4):254-260.

9. Joint Committee on Infant Hearing position statement 1982. Ear Hear. 1983;4(l):3-4.

10. Erenberg A, Lemons J, Sia C, Trunkel D, Ziring P. Newborn and infant hearing loss: detection and intervention.American Academy of Pediatrics. Task Force on Newborn and Infant Hearing, 1998-1999. Pediatrics. 1999;103(2):527-530.

1 1 . Cunningham M. Cox EO; Committee on Practice and Ambulatory Medicine and the Section on Otolaryngology and Bronchoesophagology. Hearing assessment in infants and children: recommendations beyond neonatal screening. Pediatrics. 2003;! 11(2): 436-440.

12. Early Identification of Hearing Impairment in infants and Young Children. NIH Consens Statement Online. 1993;11:1-24.

13. Early identification of hearing impairment in infants and young children. NlH Consens Statement. 1993; 11(1): 1-24.

14. Joint Committee on Infant Hearing. 1994 position statement. Audiol Today. 1994;6(6):6-9.

15. Jerger J. Hayes D. The cross-check principle in pediatric audiometry. Arch Otolaryngol. J976;102(10):6I4-620.

16. Jacobson JT, Jacobson CA. The effects of noise in transient EOAE newborn hearing screening. Int J Pediatr Otorhinolaryngol. l994;29(3):235-248.

17. Hecox G, Galambos R. Brain stem auditory evoked responses in human infants and adults. Arch Otolaryngol. 1974:99(1 ):30-33.

1 8. Starr A, Amlie RN, Martin WH, Sanders S. Development of auditory function in newborn infants revealed by auditory brainstem potentials. Pediatrics. 1977;60(6):83 1-839.

19. Guidelines for audiometrie symbols. ASHA. 1988;30(12):39-42.

20. Jerger J. A proposed audiometrie symbol system for scholarly publications. Arch Otolaryngol. 1976;102(l):33-36.

TABLE 1.

A Classification System for Degree of Hearing Loss

TABLE 2.

Hearing Assessment: Test Protocol by Age

10.3928/0090-4481-20041201-07

Sign up to receive

Journal E-contents