Audiologic assessment in infants

Yvonne S. Sininger

Purpose BOA behavioral observation audiometry

VRA Visual reinforcement audiometry
The purpose of this review is to provide the reader with current OME otitis media with effusion
information regarding the standards for audiologic assessment RECD Real-ear to Coupler Difference”

of infants and very young children. The nature of the © 2003 Lippincott Williams & Wilkins
appropriate test battery and the need for adjusting test 1068-9508

procedures to meet the specific needs of infants and toddlers

are emphasized.
Recent findings Introduction
Early detection of hearing loss is the goal of newborn
The basic measures in the audiologic test battery include
hearing screening programs. Early detection must be fol-
frequency-specific threshold tests by air and bone conduction,
lowed immediately by appropriate early intervention to
predicted by electrophysiologic measures when necessary;
maximize development of auditory skills and speech and
immittance measures including tympanometry and acoustic
language [1]. The infant who does not pass screening
reflex using a high-frequency probe tone for infants under
must receive a comprehensive audiologic evaluation, suf-
4 months of age; and otoacoustic emissions. The ABR can be
ficient to characterize the degree and type of hearing loss
used with frequency-specific stimuli to predict the audiogram
in each ear. To avoid errors, the audiologic evaluation
in newborns with a great deal of accuracy. Newer techniques,
process must use a battery of tests with cross checks. The
such as Auditory Steady State Response, are promising but
test battery must incorporate objective and behavioral
need further study before they can be used reliably to predict
measures and standard audiologic tests must be tailored
hearing levels in infants. Finally, infants with hearing loss can to be developmentally appropriate.
be fit with amplification using prescriptive formulae, such as
the Desired Sensation Level, which give appropriate hearing Unlike the visual system, the peripheral auditory system
aid characteristics for infants based on their hearing and the hearing sensitivity of a newborn infant are ex-
thresholds. These fittings must be verified using objective pected to be nearly mature at birth. The human cochlea
electro-acoustic measures tailored to infants. reaches adult size and configuration by about 20 weeks
Summary gestation [2]. Cochlear tuning, as measured by distortion-
Infants failing newborn hearing screenings can be evaluated by product otoacoustic emission suppression tuning curves,
audiologists to predict all necessary audiologic data and those is adult-like for high (6000 Hz) and low (1500 Hz) re-
found to have hearing loss can be fitted with appropriate gions by term birth [3]. Allowing a few days for resolution
amplification in the newborn period. Procedures must be of amniotic fluid in the middle ear and clearing of vernix
carefully tailored to this age group. and other debris from the neonatal ear canal [4], the
otoacoustic emission (OAE), generated by active pro-
Keywords cesses in the outer hair cells of the normal newborn, can
auditory brainstem response (ABR), otoacoustic emissions be expected to be present and robust in amplitude [5,6].
(OAEs), visual reinforcement audiometry (VRA), real-ear to
coupler difference (RECD) The auditory nerve and brainstem structures continue to
develop for 18 to 24 after-birth months as reflected in a
Curr Opin Otolaryngol Head Neck Surg 11:378–382. © 2003 Lippincott Williams gradual shortening of interpeak latencies in the ABR [7].
& Wilkins.
Although the ABR can only be recorded in response to
high-level sound in neonates as young as 27 to 30 weeks
Professor, Division of Head & Neck Surgery, University of California Los Angeles, gestational age [8,9], by a few days after term birth, the
David Geffen School of Medicine, Los Angeles, California, USA ABR can be recorded with frequency-specific stimuli re-
Correspondence to Yvonne S. Sininger, PhD, 62-132 Center for the Health vealing thresholds that are within 15 dB of those pro-
Sciences, Box 951624, Los Angeles, CA 90095-1624, USA duced by young, normally hearing adult subjects [10].
E-mail: [email protected]
The ABR is a tool that has the ability to evaluate two
Current Opinion in Otolaryngology & Head and Neck Surgery 2003, distinct aspects of the auditory system. First, auditory
11:378–382
brainstem development can be studied by determining
Abbreviations patterns of peak latency and amplitude change over
AR acoustic reflex time. Normal developmental processes, such as myelina-
ART acoustic reflex threshold
ASSR auditory steady-state response
tion, or abnormal processes, for example in infants with
OAE otoacoustic emission compromised auditory system development from prena-
378
 Audiologic assessment in infants Sininger 379

tal exposure to drugs or alcohol, prematurity, hyperbili- matically overestimate thresholds. If cochlear function
rubin etc., can be studied using the ABR [11,12]. Also, appears normal as indicated by present OAEs and neural
the threshold of ABR relative to the level of stimulus function appears abnormal, as with absent or abnormal
used to elicit the response, can generally be used as an ABR, neither OAE nor ABR will be useful to predict
accurate predictor of auditory sensitivity and is used to hearing thresholds [20]. For an update on AN in infants
determine the presence of hearing loss in infants who are see Sininger [21•].
too young for subjective assessment [13,14].
One reservation about the use of ASSR for measurement
Screening programs for detection of hearing loss in new- of hearing is the lack of data in infants and children
borns are now widespread. Every state in the US either [22••]. Perez-Abalo et al. [23] found that although they
has legislation mandating screening of newborns or is were able to determine hearing loss in the severe and
screening most infants in birthing hospitals by choice. profound range, in general, only fair agreement was
This has created a challenge for audiologists who must found between ASSR thresholds and hearing levels in
evaluate those infants who do not pass the newborn children with mild hearing loss or normal hearing. Fur-
screening to confirm or deny the presence of education- ther research and refinement is needed on ASSR to de-
ally significant hearing loss and, if confirmed, to deter- termine if this technique will produce accurate audio-
mine the type of loss and frequency-specific thresholds gram predictions in all infants but this is a promising
that define the degree and configuration. There is ur- technique.
gency to determine this information in each infant and to
fit appropriate amplification as soon as possible and be- Before the age of 6 months, it is possible to obtain un-
fore the infant reaches 6 months of age [15]. conditioned responses to sound such as a change in suck-
ing behavior, startle reflex, or eye widening. This is
known as behavioral observation audiometry (BOA).
Predicting the pure-tone angiogram These responses will be supra-threshold and, although
Tests of pure-tone or frequency-specific threshold in in- BOA cannot be used to rule out mild or moderate hear-
fants and children are either physiologic or behavioral (an ing loss, it is a valuable part of the test battery for infants
overt reaction is elicited from the child in response to under 6 months to substantiate overall impressions. At
sound and the action is judged by an audiologist). When this young age audiometric thresholds must be inferred
the appropriate technique is applied, physiologic tests, from physiologic tests such as ABR.
such as ABR, can predict pure-tone hearing thresholds
within a few dB in infants and young children with all Audiometric tests for children under the age of 3 years
degrees of hearing loss [14]. It is important to note that are classified in Table 1. A 6-month-old infant with
normal hearing newborns will show ABR thresholds for normal vision will naturally turn their head to find the
tone-burst stimuli, representing the full range of audio- source of an interesting sound. Visual reinforcement au-
metric frequencies, that are clinically identical to those diometry (VRA) is an operant conditioning paradigm that
obtained in young adults and clearly within the normal reinforces head turns with a pleasant visual stimulus
range of hearing by air and bone conduction [10,16]. (usually a lighted, animated toy). Tones and speech
stimuli can be used. VRA can be administered using
The auditory steady-state response (ASSR) is a new insert earphones for an ear-specific response or with a
physiologic evoked potential technique that is used to bone-conduction vibrator. If a child will not tolerate ear-
predict frequency-specific thresholds [17]. ASSR uses phones, the stimuli can be presented through a speaker
pure-tones stimuli (carriers) that are amplitude-modu- into the “sound-field” of a sound-treated chamber. Gen-
lated. For pediatric applications the modulation occurs at erally, a normal-hearing 6-month-old infant will respond
a frequency appropriate for infants and children (about to stimuli of 20 dB HL or better [24••]. Finally, depend-
80–100 Hz). Scalp-recorded activity is analyzed for evi- ing upon cooperation, a 2-year-old toddler can usually be
dence of the modulation frequency. Statistical analysis is tested using Play Audiometry, in which the child’s re-
used to determine the salience of the modulation fre- sponse to pure tones is to stack a ring on a pole or drop
quency activity to judge a response presence or absence a block into container as part of a game. Thresholds
for the carrier frequency. This technique has been obtained using Play Audiometry in a cooperative child
shown to be of value in assessing aided hearing thresh- are assumed to be adult-like in accuracy.
olds in the sound field [18]. Hearing thresholds have
been estimated within about 10 to 15 dB in adults with Immittance measures
normal hearing and hearing loss using the multi- Tympanometry and acoustic or middle-ear muscle re-
frequency ASSR [17]. It must be noted that when con- flexes are a standard, objective test of tympanic mem-
ditions of neural disease, such as auditory neuropathy brane mobility, middle-ear pressure, and brainstem au-
(AN) are found [19] neither the ABR nor the ASSR will ditory activation. These tests are a very important part of
be good indications of hearing level and they may dra- the test battery used for audiometric assessment of in-
380 Audiology

Table 1. Audiologic tests appropriate for infants and toddlers by age

Age in months 0 2 4 6 8 10 12 18 24 30 36

Threshold ABR [------------------------------------------------------------------------------------------------------→

Air or
Bone VRA [----------------------------------------------------------------]

Play [--------------------→

Middle-ear Tymp 600– [------------------------------------------------------------------------------------------------------→

function 1000 Hz

Tymp [---------------------------------------------------------------------------→
226 Hz

Cochlear OAEs [------------------------------------------------------------------------------------------------------→

VRA, Visual reinforcement audiometry; Tymp, Tympanogram.

fants and toddlers. This is especially important when one physiologic noise and thus, absent OAEs in low- fre-
considers the high incidence of otitis media with effusion quency regions should not be given great weight in
(OME) in children in this age group. In a recent study of interpretation.
over 3000 infants aged 8 to 12 months, receiving follow-
up audiologic examinations, 30% had OME at the time Absence of OAEs can be due to a variety of causes, from
of testing [25]. middle-ear dysfunction to sensorineural disorders pro-
ducing hearing loss of any degree. The absence of an
Tympanometry for young infants, when performed with OAE should not be interpreted as indication of signifi-
a standard probe tone frequency of 220 or 226 Hz, can be cant hearing loss. In contrast, OAE presence indicates
invalid. Infants with OME can reveal a normal-appearing good hair cell function and generally indicates that the
tympanogram when tested using this probe frequency hearing thresholds should be better than 30 to 40 dB.
[26]. This finding may be due to extensibility of the skin However, OAEs cannot be used to determine exact hear-
of the ear canal in these infants. Valid tympanograms can ing thresholds.
be obtained by raising the probe frequency to 600 to
1000 Hz for infants aged 4 months or less [27,28]. The The presence of an OAE alone cannot insure that hear-
acoustic reflex (AR) can be a very useful part of the ing sensitivity is normal. Disease that spares the cochlea
audiologic evaluation in infants. A present reflex is and impairs function in the auditory nerve or low brain-
added support for normal middle-ear function. Because stem (eg, acoustic neuroma or AN) can also cause signifi-
the reflex arc involves the seventh and eighth nerve and cant hearing loss. The OAE must be included in a bat-
the low brainstem, a normal or present reflex can be tery of tests for accurate interpretation.
useful in ruling out abnormalities such as AN [20]. In
Fitting of amplification in infants
addition, although there is no direct relation between The goal of early identification and characterization of
acoustic reflex threshold (ART) and hearing threshold, hearing loss in infants is the appropriate fitting of ampli-
the ART does set an upper bound for hearing level (ie, fication as early as possible. If an infant is identified
the ART will never be elicited at levels below the true before 6 months of age, the fitting must proceed based
auditory threshold). It is particularly important to use a on hearing thresholds obtained by ABR. The infant will
high-frequency probe (600–1400 Hz) to measure the AR be able to do very little to help in the initial fitting of the
in infants under 6 months of age [29]. amplification. For that reason, we rely on objective
physical measurements along with behavioral observa-
Otoacoustic emissions
tions and parent report in this age group.
Otoacoustic emissions are a noninvasive, objective mea-
sure of cochlear functioning. OAEs are generated exclu- Formulas such as the Desired Sensation Level (DSL)
sively by outer hair cells [30]. Outer hair cells are gen- [31] prescribe the output characteristics needed in a
erally more vulnerable to disease and damage than inner hearing aid for children based on the child’s hearing
hair cells. Therefore when OAEs are normal, it is rea- thresholds. Infant ear canals are considerably smaller
sonable to assume that the inner hair cells are function- than those of adults and it is critical to take the acoustic
ing as well. When OAEs are present for a range of fre- characteristics of the individual infant’s ear into consid-
quencies, hearing thresholds are generally 30 to 40 dB eration when the amplification levels from the hearing
HL or better for those frequency regions. Caution should aids are set. Consequently, real-ear measures are an in-
be used in over-interpretation of very narrow, low- dispensable part of the fitting process in infants.
amplitude regions of OAE, which can be spurious noise.
Also, OAEs are not strong in low-frequency regions Hearing aid characteristics (such as gain and maximum
(1000 Hz and below) in infants and toddlers due to output) can be measured in one of two ways. Tradition-
 Audiologic assessment in infants Sininger 381

ally, the hearing aid is connected to a hard-walled cou- Conclusion

pler with dimensions designed to mimic the size and Audiologists have the technology necessary to perform
acoustic characteristics of the human external ear and all the basic audiologic assessments on newborn infants.
canal. A microphone is connected to the coupler to mea- It is crucial that time not be lost in identifying hearing
sure the hearing aid output in approximately the position loss and fitting appropriate amplification. Extra care
that the tympanic membrane would occupy. The hearing must be taken to avoid mistakes. For that reason test
aid is turned on and placed in a test chamber where batteries with attention to crosschecking results must be
calibrated sounds are delivered to the hearing aid micro- used. (For an excellent review see Gravel [35••]). When-
phone. A comparison of the input sound to the output in ever possible objective measures are used in conjunction
the coupler describes the performance characteristics of with behavioral measures and test batteries are adjusted
the hearing aid. as appropriate for the age of the child as they grow. Once
the audiogram is accurately predicted, electro-acoustic
real ear measures and prescriptive formulae can be ap-
Another way of measuring hearing aid acoustic perfor-
plied to accurately fit and monitor amplification. No in-
mance is termed a “real-ear” measurement. The mea-
fant with hearing loss should be misdiagnosed or improp-
suring microphone is connected to a tiny probe tube that
erly fit with amplification, if proper protocols are
is placed in the patient’s ear canal terminating near the
understood and followed.
tympanic membrane. The hearing aid is then placed in
the ear with the probe tube undisturbed. The measure-
References and recommended reading
ment device again presents calibrated sounds directed
Papers of particular interest, published within the annual period of review,
toward the subject wearing the hearing aid. By compar- have been highlighted as:
ing the sound levels at the entrance to the ear and those • Of special interest
amplified by the aid near the tympanic membrane, the •• Of outstanding interest
amplification characteristics of that aid on that patient 1 Sininger YS, Doyle KJ, Moore JK: The case for early identification of hearing
can be determined. When real-ear measures are used, loss in children. Pediatr Clin North Am 1999, 46:1–14.

the true characteristics of the aid in the subject’s ear can 2 Moore KL, Persaud TVN: The Developing Human, edn 5. Philadelphia:
WB Saunders;1993.
be determined and the aid can be manipulated to obtain
3 Abdala C: A developmental study of distortion product otoacoustic emission
the desired levels according to the prescriptive formula (2f1-f2) suppression in humans. Hear Res 1998, 121:125–138.
being used. 4 Doyle KJ, Burggraaff B, Fujikawa S, et al.: Neonatal hearing screening with
otoscopy, auditory brain stem response, and otoacoustic emissions. Oto-
laryngol Head Neck Surg 1997, 116:597–603.
Because it can be difficult to get cooperation from infants 5 Sininger YS: Clinical Applications of Otoacoustic Emissions. Advances in
for the real-ear measures, a technique known as the Otolaryngology—Head and Neck Surgery 1993, 7:247–269.
“Real-ear to Coupler Difference” (RECD) is recom- 6 Norton SJ, Widen JE: Evoked otoacoustic emissions in normal-hearing infants
and children: Emerging data and issues. Ear Hear 1990, 11:121.
mended for small children [32]. The probe tube is fit into
the infant’s ear followed by an inserted earphone. The 7 Salamy A, McKean CM, Buda FB: Maturational changes in auditory transmis-
sion as reflected in human brainstem potentials. Brain Res 1975, 96:361–
earphone presents a calibrated sound and a quick acous- 366.
tic measurement is recorded from the probe microphone. 8 Starr A, Amlie RN, Martin WH, et al.: Development of auditory function in
The time needed for infant cooperation is less than a newborn infants revealed by auditory brainstem potentials. Pediatrics 1977,
60:831–839.
minute. Next, the earphone is used to present the same
9 Galambos R, Hecox KE: Clinical applications of the auditory brainstem re-
sound via the standard coupler and the sound level is sponse. Otolaryngol Clin North Am 1978, 11:709–722.
recorded. By comparing the levels recorded in the cou- 10 Sininger YS, Abdala C, Cone-Wesson B: Auditory threshold sensitivity of the
pler to those recorded in the infant’s ear, the infant’s human neonate as measured by the auditory brainstem response. Hear Res
1997, 104:27–38.
RECD can be determined. Generally, the level recorded
is higher in the infant’s ear and the degree of difference 11 Shih L, Cone-Wesson B, Reddix B: Effects of maternal cocaine abuse on the
neonatal auditory system. Int J Pediatr Otorhinolaryngol 1988, 15:245–251.
is greatest for the high frequencies. Now, hearing aids
12 Lee JA, Schoener EP, Nielsen DW, et al.: Alcohol and the auditory brain-stem
can be adjusted and tested for that infant using the cou- response, brain temperature, and blood alcohol curves: explanation of a para-
pler (rather than the squirming baby) and RECD values dox. Electroencephalogr Clin Neurophysiol 1990, 77:362–375.
can be applied to correct coupler measures indicating 13 Sininger YS, Abdala C: Hearing threshold as measured by auditory brain
stem response in human neonates. Ear Hear 1996, 17:395–401.
how the aid will perform in the infant’s ear. Whether or
14 Stapells DR, Gravel JS, Martin BA: Thresholds for auditory brainstem re-
not the infant will cooperate for standard real-ear mea- sponses to tones in notched noise from infants and young children with nor-
sures, a physical measurement is used to determine mal hearing or sensorineural hearing loss. Ear Hear 1995, 16:361–371.
that the appropriate hearing aid fitting has been achieved. 15 Joint Committee on Infant Hearing: Year 2000 position statement. Audiology
This allows audiology to move ahead with hearing aid Today 2000 (Special Issue 2000):3–23.

fittings long before the infant can offer a behavioral re- 16 Cone-Wesson B, Ramirez GM: Hearing sensitivity in newborns estimated
from ABRs to bone-conducted sounds. J Am Acad Audiol 1997, 8:299–307.
sponse when using the aids. For excellent review of hear-
17 Dimitrijevic A, John MS, Van Roon P, et al.: Estimating the audiogram using
ing aid fittings and verification for infants see Beauchaine multiple auditory steady-state responses. J Am Acad Audiol 2002, 13:205–
•
[33••] and Scollie and Seewald [34••]. 224.
382 Audiology

This paper describes the use of simultaneous multiple test tones (4 per ear) for 26 Paradise J, Smith C, Bluestone C: Tympanometric detection of middle ear
rapid measurement of ASSR in both ears at once. The technique is able to predict effusion in infants and young children. Pediatrics 1976, 58:198–210.
hearing levels in adults with normal hearing and hearing loss within 15 dB.
27 Marchant CD, McMillan PM, Shurin PA, et al.: Objective diagnosis of otitis
18 Picton TW, Durieux-Smith A, Champagne SC, et al.: Objective evaluation of media in early infancy by tympanometry and ipsilateral acoustic reflex thresh-
aided thresholds using auditory steady-state responses. J Am Acad Audiol olds. J Pediatr 1986, 109:590–595.
1998, 9:315–331. 28 McKinley AM, Grose JH, Roush J: Mulit-frequency tympanometry and evoked
19 Starr A, Picton TW, Sininger Y, et al.: Auditory neuropathy. Brain 1996, otoacoustic emissions in neonates during the first 24 hours of life. J Am Acad
119:741–753. Audiol 1997, 8:218–223.
29 Weatherby M, Bennett M: The neonatal acoustic reflex. Scand Audiol 1980,
20 Sininger YS, Oba S: Patients With Auditory Neuropathy: Who Are They and
9:103–110.
What Can They Hear. In: Sininger YS, Starr A, editors. Auditory Neuropathy:
A New Perspective on Hearing Disorders. Albany, New York: Thompson 30 Schrott A, Puel J-L, Rebillard G: Cochlear origin of 2ƒ1-ƒ2 distortion products
Learning; 2001:15–35. assessed by using 2 types of mutant mice. Hear Res 1991, 52:245–254.

21 Sininger YS: Auditory neuropathy in infants and children: implications for early 31 Seewald RC, Moodie KS, Sinclair ST, et al.: Predictive validity of a procedure
• hearing detection and intervention programs. Audiology Today 2002 (Special for pediatric hearing instrument fitting. Amer J Audiol 1999, 8:143–152.
Issue):16–21. 32 Moodie KS, Seewald RC, Sinclair ST: Procedure for predicting real-ear hear-
This article summarizes data from a variety of studies on the incidence and risk ing aid performance in young children. Amer J Audiol 1994:23–31.
factors involved in infants with auditory neuropathy. Implications for test batteries
appropriate for newborn screening are discussed. 33 Beauchaine KA: An amplification protocol for infants. In A Sound Foundation
•• Through Early Amplification 2001; Proceeding of the Second International
22 Stapells DR: The tone-evoked ABR: Why it’s the measure of choice for young Conference. Edited by Seewald RC, Gravel JS. Phonak AG;2002:105–12.
•• infants. The Hearing Journal 2002, 55:14–18. This paper is found in the proceedings of an excellent meeting on amplification for
This paper is part of an entire issue devoted to the same topic as this paper—the children sponsored by Phonak. The text gives a step-by-step procedure for transi-
follow-up evaluations for infants who are referred from newborn hearing screening. tioning from hearing thresholds to hearing aid fitting for infants.
This manuscript discusses tricks on getting a rapid and reliable threshold measure
using ABR and also discusses the prematurity of ASSR for this application. 34 Scollie SD, Seewald RC: Electroacoustic verification measures with modern
•• hearing instrument technology. In A Sound Foundation Through Early Ampli-
23 Perez-Abalo MC, Savio G, Torres A, et al.: Steady state responses to multiple fication 2001; Proceeding of the Second International Conference.. Edited
amplitude-modulated tones: an optimized method to test frequency-specific by Seewald RC, Gravel JS. Phonak AG;2002:121–127.
thresholds in hearing-impaired children and normal-hearing subjects. Ear This paper is also from the pediatric amplification proceedings. This paper dis-
Hear 2001, 22:200–211. cusses the theoretical and practical aspects of electro-acoustic verification of hear-
ing aid fittings on infants and young children.
24 Widen JE, O’Grady G: Using visual reinforcement audiometry in the assess-
ment of hearing in infants. The Hearing Journal 2002, 55:28–36. 35 Gravel JS: Potential pitfalls in the audiological assessment of infants and
This is another paper in the issue on follow-up testing that provides audiologists •• young children. In A Sound Foundation Through Early Amplification 2001;
with excellent practical tips on how and when to use VRA for the assessment of Proceeding of the Second International Conference. Edited by Seewald RC,
behavioral thresholds. Gravel JS. Phonak AG;2002:85–101.
The third recommended paper from the Phonak conference, which provides excel-
25 Widen JE, Folsom RC, Cone-Wesson B, et al.: Identification of neonatal hear- lent case studies demonstrating the importance of the use of age-appropriate mea-
ing impairment: Hearing status at 8 to 12 months corrected age using a visual sures and of the test battery crosscheck principle. Many of the points stressed in
reinforcement audiometry protocol. Ear Hear 2000, 21:471–487. this review are emphasized by these case studies.