Share this article on:

Testing Otoacoustic Emissions in Children: The Known, and the Unknown

Smith, Joanna T. MS; Wolfe, Jace PhD

doi: 10.1097/

Ms. Smith is cofounder and executive director of Hearts for Hearing in Oklahoma City. Dr. Wolfe is director of audiology at Hearts for Hearing and an adjunct assistant professor at the University of Oklahoma Health Sciences Center and Salus University.

The year was 1977. Star Wars graced the silver screen. Jimmy Carter took over the Oval Office, and the Eagles ruled the Top 40 with “Hotel California.” Even more importantly, David Kemp became the first hearing scientist to measure otoacoustic emissions (OAEs) in human ears.

Dr. Kemp's discovery was met with more than a bit of skepticism, but he and others—including, but not limited to, Jay Hall, Michael Gorga, Beth Prieve, Susan Norton, Sumitrajit Dhar, Brenda Lonsbury-Martin, and Martin Robinette—continued to explore OAE through rigorous research.

We want to dedicate this column about contemporary practices for OAE use with infants and young children to the aforementioned researchers. Through their work over the past 36 years, they have unknowingly served as our mentors.

Back to Top | Article Outline


We now know that otoacoustic emissions are generated by the outer hair cells (OHCs) in the cochlea, transmitted through the middle ear, and measured in the external ear canal, but we still do not understand the exact mechanisms.

OHCs contract when they are stimulated and elongate when they are hyperpolarized. These changes in shape are largely thought to be responsible for the cochlear amplifier that gives a boost to low-level sounds. It is also possible that motility of the stereocilia is at least partially responsible for OAE generation.

Two theories have been proposed for how OAEs are propelled. First, minor irregularities may exist along the basilar membrane that “reflect” the OHC energy through the cochlear fluids (Hear Res 1988;33[1]:69-93; Hear Res 1984;13[1]:89-98

The second proposed theory involves the generation of nonlinear distortion from the interaction of the stimuli on the basilar membrane.

Back to Top | Article Outline


No real consensus exists for how pediatric clinicians should interpret OAE results from infants and children. Personally, we love Jay Hall's suggestion: OAE analysis can lead to three general findings: normal, present but abnormal, or absent. (Hall JW III, Swanepoel DW. Objective Assessment of Hearing. San Diego: Plural Publishing; 2009.)

A “normal” OAE result is constituted by two findings: The otoacoustic emission response must be at least 6 dB higher than noise at the same frequency, and the absolute amplitude must exceed a certain criterion.



An example of OAE analysis for an older child or adult is shown in the figure. The OAE responses (red circles) are much higher than the noise (green triangles). Also, the OAE response exceeds the 95th percentile for people with hearing loss, indicating that there is less than a five-percent chance that the response came from an ear with hearing loss.

An alternative criterion for absolute amplitude is the fifth percentile for people with normal hearing. If the response is below this criterion, then there is less than a five-percent chance it came from an ear with normal hearing.

A present-but-abnormal otoacoustic emission result is one that meets the 6-dB criterion but has an amplitude below what one would expect for a person with normal hearing sensitivity. This finding may be more likely to occur in cases of slight to mild sensory hearing loss or a subtle middle ear dysfunction.





In an absent OAE response, there is less than 6 dB of separation between the OAE response and the noise, which is measured at an acceptably low level. This response is likely to be obtained with patients who have a sensory hearing loss but may also occur in those with middle ear dysfunction.

Back to Top | Article Outline


Infants from one month to one year of age typically have higher OAE amplitudes than toddlers, preschool children usually possess higher-level OAEs than school-age children, and children generally have larger OAEs than adults (Hear Res 2006;212[1-2]:90-98; Brain Dev 2000;22[1]:41-46; J Acoust Soc Am 1997;102[5 pt 1]:2871-2879

When we conduct diagnostic assessments at Hearts for Hearing, our index of concern for hearing loss becomes progressively higher as the distortion product otoacoustic emission (DPOAE) amplitude of a baby approaches—or, especially, falls below—5 dB SPL, particularly for responses in the 2,000-Hz to 4,000-Hz range. Also, in a quiet infant, we don't give the all clear until we record a DPOAE with normal amplitude as well as a signal-to-noise ratio (SNR) of 10 dB.



For transient-evoked OAE (TEOAE) measurement in infants, we strive to obtain an otoacoustic emission level of at least 0 dB SPL, an SNR of at least 10 dB, and a reproducibility of at least 70 percent. In many normal hearing infants, TEOAE level far exceeds 0 dB SPL, with a robust SNR and reproducibility in the upper 90s.

If OAEs are insufficient and acoustic immittance suggests middle ear dysfunction, we attempt bone conduction auditory brainstem response (ABR), as scheduling permits, to rule out a permanent sensory hearing loss, and we refer for medical evaluation of middle ear function, after which the infant will return for repeat assessment.

If OAEs are insufficient, acoustic immittance is normal, and the ABR suggests normal hearing sensitivity, we provide the parents with a list of speech, language, and hearing developmental milestones, and we schedule the baby for return assessment at 6 months of age. At the 6-month assessment, we repeat acoustic immittance and OAE testing, and we conduct behavioral audiologic assessment.

If OAE are insufficient and the ABR indicates permanent hearing loss, we counsel the family about the results, inform them of their options, and, as appropriate, proceed with audiologic treatment and referral to a listening and spoken language specialist for evaluation.

Back to Top | Article Outline

4. SHHH!

Infants and young children tend to have higher noise levels than adults. Infants produce sucking noises when they use a pacifier or bottle, and pediatric patients are more likely to vocalize during testing and produce loud respiratory noise.

Ideally, OAE testing should be timed for when the child is sleep deprived and ready for a nap. If the child will not sleep, it is usually possible to get him or her to cooperate for OAE testing by showing a children's video on a monitor, tablet, or smartphone; blowing bubbles; or holding a puppet, stuffed animal, mirror, or small illuminated toy in front of the child.

Most children with normal hearing have robust responses, so it is possible to record normal OAE in cases where a child sucks on a pacifier or bottle.

Many instruments allow for the plotting of the 95th percentile of noise level obtained from normative data in a large study. When an instrument does not, the clinician may observe a general rule of thumb. Since physiologic noise is mainly comprised of low-frequency energy, noise levels should roughly fall at or below -5 dB SPL from 1,000 Hz and below, and at or below -10 dB SPL at 2,000 Hz and lower.

OAE analysis should occasionally be considered inconclusive when noise levels exceed acceptable values. In the case of robust OAEs, the response may well exceed the noise and be considered normal.

Back to Top | Article Outline


In order to generate a robust OAE response, it is critical that the clinician ensure the right test environment.

The level of the stimulus should be acceptable throughout the duration of the test. For DPOAE testing, it is ideal for the f1 and f2 primary tones to be presented at 65 and 55 dB SPL, respectively, at all evaluated frequencies.

The primary tone levels should resemble flat lines throughout the test, as shown at the top of the figure.

For TEOAE testing, the displayed stimulus spectrum should be flat throughout the frequency range of interest. Also, the display reflecting temporal aspects of the stimulus should indicate little to no stimulus ringing.

We have found that a snug probe fit is more likely with foam ear tips. In the case of infants, we often trim off much of the foam from the tip to allow placement into the small ear canal.

Back to Top | Article Outline


In rare cases, it is possible to obtain abnormal TEOAE in the presence of normal DPOAE, or absent TEOAE in the case of present but reduced DPOAE (Prieve BA; personal communication; March 1, 2013).

These children should be followed closely until the audiologist is certain that any potential auditory dysfunction will not result in a delay in speech and language development.

DPOAE is better suited than TEOAE for assessing auditory function at 6,000 Hz and above, and can even be measured well beyond 8,000 Hz using special instrumentation. In particular, the clinician should consider use of DPOAE when monitoring for the presence of high-frequency hearing loss, such as in children receiving ototoxic medications.

TEOAE may be a better choice for the evaluation of OAE in the 750- to 1,000-Hz range.

Back to Top | Article Outline


To overcome the impedance mismatch for sound transmission from air in the outside world to the cochlear fluid of the inner ear, the middle ear boosts the signal from the outside going inward.

Unfortunately, this design results in significant attenuation as the OAE response attempts to travel from the cochlea back to the ear canal—as much as 30 dB, according to some estimates (J Acoust Soc Am 2003;113[5]:2773-2789

Considering this, we should not be surprised that the presence of any type of middle ear dysfunction typically means we're unable to record OAE responses. Usually, such responses are absent or significantly reduced in amplitude when a child has otitis media (Ear Hear 2000;21[3]:212-217; Am J Otol 1993;14[1]:34-40;Int J Audiol 2003;42[3]:117-131

Since infants have such large OAEs, it is more probable to record otoacoustic emissions in infants who have middle ear effusion than in older children or adults. In fact, these infants will occasionally show flat tympanograms with present OAEs.

Back to Top | Article Outline


OAE measurements are most powerful when interpreted in conjunction with other audiologic measures. James Jerger and colleagues taught us that the acoustic reflex is our most sensitive measure of middle ear dysfunction (Arch Otolaryngol 1974;99[3]:165-171

If acoustic reflexes are present at normal limits, but OAEs are absent, then the clinician should be concerned about a possible mild to moderate sensory hearing loss.

If acoustic reflexes are present but elevated, and OAEs are absent, then the clinician should be suspicious of a moderately severe to severe sensory hearing loss.

If tympanometry is normal, but acoustic reflexes and OAE are both absent, then we should be on the lookout for a severe to profound sensorineural hearing loss.

Finally, if OAEs are present and robust but acoustic reflexes are absent or abnormal, then we conduct additional evaluations to assess for retrocochlear disorder.

Back to Top | Article Outline


Just like Bill Murray lived the same day over and over again in Groundhog Day, OAE results, like any physiologic response in the auditory system, should be similar from one measure to the next.

Previous research has suggested that OAE amplitude should fall within +/- 2 dB in successive measures. (Dhar S, Hall JW III. Otoacoustic Emissions: Principles, Procedures, and Protocols. San Diego: Plural Publishing; 2011.)

Jay Hall is fond of saying that “if the test does not repeat, then it is not complete.” We totally agree. Repeatable OAEs that are present but abnormal enhance concern for middle ear dysfunction or a mild hearing loss.

Back to Top | Article Outline

10. OAEconomics

There are three CPT codes for OAE measurements in children:

92558: This code describes OAE screening assessment when automated analysis is conducted.

92587: This code, which is supported by a written report from the responsible clinician, describes DPOAE diagnostic assessment conducted at three to six frequencies per ear, or TEOAE with analysis by an audiologist or physician. Additional reimbursement may be pursued by billing the -22 modifier if the clinician conducts both distortion product and transient-evoked otoacoustic emissions.

92588: This code describes a comprehensive DPOAE diagnostic assessment conducted at a minimum of 12 frequencies per ear with analysis by an audiologist or physician.

May the force be with you as you use OAE to provide the best possible care for the children you serve.

© 2013 by Lippincott Williams & Wilkins, Inc.