AudiologyOnline Phone: 800-753-2160

Evaluating Hearing in Infants and Small Children

Evaluating Hearing in Infants and Small Children
Michelle R. Petrak, PhD, CCC-A
November 3, 2000
This article is sponsored by ICS.

In the past several years, considerable data has been obtained supporting the need for newborn and infant hearing screening while emphasizing the benefits of early intervention for very young children with hearing impairments (1,2. Yoshinaga-Itano, 1995, Yoshinaga-Itano et al 1998). According to the National Institutes of Health (NIH), otoacoustic emissions (OAE) and auditory brainstem response (ABR) testing are recommended for infant hearing screenings (3. NIH, 1993).

The primary objective of infant screening is to obtain reliable, ear-specific and frequency-specific information on auditory function, as quickly as possible (4. Bachmann, 1998). Once hearing loss is confirmed, the audiologist must assess the type and degree of hearing loss, the needs of the child and the family and the appropriate rehabilitation instruments and strategies so intervention can begin as quickly as possible.

ABR, when applied to newborns and very young children, is a reliable tool for approximating hearing thresholds and audiometric configurations. (5. Oates et al, 1998). The use of ototoxic drugs, oxygenation and ventilation procedures, as well as the incidence of bilirubin problems, and other abnormalities in newborns has helped force the rebirth of the ABR. ABR is the 'gold standard' in the demonstration of hearing abnormalities in infants.

It is well accepted that the most reliable measurement of hearing in children, is conventional audiometry. Techniques such as visual reinforcement audiometry (VRA) are available and are very successful assessment tools for children 6 months of age and older. However, in the first few months of life, behavioral audiometry tests are unreliable.
This is why audiologists rely more on physiological tests to determine hearing sensitivity in newborns and very young children.

For many audiologists, ABR test protocols for the adult population are well known and accepted. However, when addressing infants, protocols and practices need to be revised. Eventually, standardized infant protocols will be developed and will become widely accepted. This paper reviews some basic issues and methods related to recording auditory responses using ABR and other auditory evoked potentials in newborns and infants.

Auditory Brainstem Response Testing

ABR is not a hearing test. ABR is a measure of synchronized nerve firing along the auditory brainstem pathways in response to auditory stimuli. ABR is essentially unaffected by the state of consciousness and responses can be recorded at or near auditory threshold. ABR can identify and quantify hearing loss in infants and other patients unable or unwilling to participate in behavioral testing. Importantly, information obtained from ABR can be used to select hearing aids and to preliminarily set their electroacoustic parameters.

I. Electrodes and Electrode Application

It is important to select electrodes designed specifically for infants. There are a variety of smaller electrodes designed for use with infants. Sterile electrodes are necessary.

Types of electrodes:

a. Disposable electrodes and leads. These provide the safest connection. They are used once and then the entire unit is disposed of.

b. Disposable electrodes with reusable leads. The electrodes are individually packaged to provide sterility. After use, the electrode is disposed of and the lead wire must be re-sterilized.

c. Reusable electrodes and leads. These are completely reusable; there is nothing to dispose of. Both the electrode and the lead are re-sterilized after each use.

The infant's skin should be carefully and gently abraded before the electrodes are applied. The impedances should be less than 5000 ohms. A single channel recording is acceptable. Contralateral recordings in infants can be considerably different than ipsilateral recordings and hence comparisons are not always useful. (Edwards, 1985)

a. The negative electrode is connected to the nape of the neck or ipsilateral earlobe or mastoid.

b. The positive electrode is connected to the high forehead as near to Cz as possible.

c. The common electrode is connected to the forehead (vertical montage). Some research findings suggest the nape of the neck provides a larger wave V response than does the ear/mastoid location (King and Sininger, 1992). If wave I is required for neurodiagnostic procedures, a horizontal montage is recommended.

II. Transducers:

a. Insert Earphones (a.k.a. 'inserts')

Inserts have several advantages including disposable neonatal eartips, reduced stimulus artifact, decreased background noise, less chance of crossover, higher interaural attenuation, decreased likelihood of collapsed canals, increased comfort, and smaller tubing with a better chance of fitting an infants ear. A correction factor is needed to account for the length of the tubing and the associated acoustic delay. The correction factor is usually 0.9 ms and most manufacturers of ABR equipment automatically account for this difference.

b. Head Phones

Typical TDH-49 headphones are still available with most ABR systems. Headphones are calibrated separately from the insert phones. When testing infants and small children, the size of the headphones can be a barrier. It is often necessary to remove the head phone cups from the headband and hold them (the cup portion) to the baby's ear. It is also important to insure that the receiver is lined up with the ear canal. Newborn ears are also prone to collapse from pressure applied from the earphone.

c. Bone conduction

A conductive hearing loss can be effectively ruled out by performing ABR using a standard B-70 bone conductor (BC). Importantly, the B-70 oscillator used for standard BC testing must be recalibrated for ABR use. Clinically, the two cannot be interchanged. The accuracy of bone conducted responses is dependent on proper pressure applied to the bone conductor, proper calibration of the bone conduction oscillator for ABR purposes, and optimal anatomic placement. It is good practice to apply pressure to the bone conductor using a single finger, when testing an infant (Bachman and Hall, 1998). The best placement appears to be the mastoid (Yang et al, 1987). Bone conduction ABR's can be obtained with either click or tonal stimuli.

Normative data collection is highly recommended to assure that bone conducted ABR's have a well established and constant relationship to traditional, air conduction derived audiometric thresholds.


a. Clicks

A click stimulus is an electrical impulse, typically 100 microseconds in duration. The click is a broadband signal containing a wide range of frequencies in its spectral presentation. Importantly, the click demonstrates the highest energy in the frequency range from 2 kHz - 4KHz. An ABR generated with a click can be adequate for screening hearing, but it cannot provide frequency specific information across the entire speech region, which is necessary for proper hearing aid fittings.

b. Tone burst

Gating an ongoing electrical signal through an electronic switch or some other type of modulation technique generates a tone burst. The stimulus rise time reacts with the frequency in a complicated way (see ramp and plateau). Responses to tone bursts provide relatively accurate estimates of auditory sensitivity and can accurately predict the audiogram. Ideally, a tone burst concentrates energy at a single pure-tone frequency. This causes activation of the basilar membrane of the cochlea that is sensitive to that specific frequency. A pitfall of tone bursts is that a stimulus with a very brief onset may produce spectral splatter in and around unwanted frequencies. Several types of masking and envelope patterns are used to reduce spectral splattering caused by tone bursts and hence concentrate on specific tonal regions of the cochlea. A 500 Hz tone burst produces a response with a much different morphology compared to a click response. The toneburst produces a broad wave V without readily defined peaks.

III Stimulus Parameters

a. Rate

Repetition rate is a significant parameter variable. Site of lesion studies are usually obtained with slow repetition rates of 20 pulses per second, or less. Slower repetition rates tend to preserve waveform morphology. As rate is increased, waveform morphology becomes poorer. It is also important to assure that an adequate interstimulus interval is chosen. Even though slower rates provide better morphology, faster repetition rates can be used to expedite threshold testing. Infants can easily be screened using rates up to 39.1 clicks/sec. Nonetheless, it is often useful to reduce the presentation rate when approaching threshold. Odd number stimulus rates are advisable to reduce interference with 60 Hz electrical noise.

b. Ramp, plateau and duration

With tone bursts, frequency and duration can be controlled separately. Higher frequencies can rise from zero to maximum amplitude in a very brief time period because the duration of their cycles is shorter. Lower frequencies require longer rise times. A very brief rise time causes stimulus energy to spread to frequencies on both sides of the intended signal. Since the rise time, however specified, interacts with the stimulus frequency and spectral dispersion, some uncertainty about point of stimulation in the cochlea accompanies the use of tone bursts. A Blackman ramp helps maximize synchrony and minimize spectral splatter. This type of window works best without a plateau, such as a ramp 2-0-2 setting, which has a rise and fall time of 2 ms each without a plateau present.

c. Frequency

Most ABR systems on the market today can cover a wide range of frequencies. The most common frequency range tested is 500 Hz-4KHz. This information is needed to determine intervention strategies.

d. Polarity

Polarity refers to the phase of the signal. Rarefaction is produced by initially pulling the earphone diaphragm away from the tympanic membrane and condensation is produced by initially pushing the diaphragm toward the tympanic membrane. Alternating polarity switches back and forth between condensation and rarefaction. Alternating polarity is often used to reduce noise and eliminate the cochlear microphonic. The initial phase of the stimulus and the intensity of the stimulus interact in a complex manner to produce changes in waveform latency. Currently, there are no industry standards for these parameters. Hence, knowledge of how stimulus and recording parameters impact the waveform is necessary for interpreting test results and for efficiently maximizing test parameters and clinical efficacy.

Additionally, it is important to understand stimulus and recording parameters when comparing data from one clinic to another with respect to normative data and unusual test results. One useful 'rule of thumb' regarding polarity is to make sure the polarity you are testing with is the same as was used to collect the normative data.

IV. Acquisition parameters

(a) Gain

Gain of 100,000 is typical and is sufficient to record ABR's from babies.

(b) Filters

Filter settings primarily affect the auditory response with respect to amplitude. Most audiologists agree there is little significant ABR information obtained with the high frequency (low-pass) filter set above 3000 Hz. There is much less agreement on low frequency (high-pass) settings, since there can be a significant increases in noise as this setting is lowered. There are two types of filters. (a) High frequency filters allow low frequency sounds to pass through and cut off high frequency sounds. When testing infants, these filters are often set at 1500 Hz. (b) Low frequency filters allow high frequency sounds to pass through and cuts-off lower frequencies. These filters are often set at 30 Hz. A notch filter is used only when artifact is uncontrollable by other means. A notch filter provides some protection against electrical interferences, such as the hum that can be heard from fluorescent lighting.

(b) Analysis Time (Epoch) and Sweeps

The analysis time or epoch is the period of time, after the stimulus is presented, in which data are collected and appear in the analysis window. This window needs to be long enough to display the entire response. Babies have longer latencies for wave V than adults. Hence, the window should be at least 20-22 ms. When selecting an epoch it important to allow enough time for possible interactions in analysis time and stimulus rate. The interstimulus interval is the time period between two successive stimuli. The interstimulus interval decreases as the rates of stimulation increases and vice vera.

Each time a stimulus is presented and data is recorded, one sweep occurs. In ABR and in most auditory evoked potentials, it is necessary to average many sweeps. Generically, this is referred to as signal averaging. The component of the sweep, which contains the 'signal', is essentially held constant and is therefore amplified across many repetitions, whereas the more random physiologic background 'noise' is averaged out to essentially zero. Some authors have described this as 'allowing the ABR to emerge from the noisy background'. It is common to collect 1500-2000 sweeps. More sweeps are used in noisier backgrounds. It is permissible to stop data collection after 1000 sweeps, if you have well defined waveforms and peaks.


When testing with ABR equipment, the signal presented is much shorter in duration than pure tone signals used in conventional audiometry. This means that an 80 dB HL 1 kHz tone burst is not perceived with equal loudness compared to a 1 kHz puretone. There are no international standards for ABR.

It is the responsibility of either the manufacturer or the clinician performing the test to determine the normative values for clicks and tones. This is easily accomplished by performing listening tests on normal hearing listeners. To calibrate a toneburst, the point at which the listener can just barely detect the toneburst is designated '0 dB nHL'. This allows the audiologist to compare the tone burst data collected with audiometric estimates of hearing.

Many manufacturers provide default dB nHL values. However, you must bear in mind that those values are only valid if used under similar recording parameters. If your test protocols differ from those that were used by the manufacturer, then you must provide corrections to the dB nHL default values. dB nHL values change with parameter settings such as rate and ramp/plateau. For example, the faster the click rate the easier it is to perceive a stimulus, and hence the lower the intensity needed to achieve threshold.


Suggested test parameters for Auditory Brainstem Response when testing Newborns and Infants:


1. Yoshinaga-Itano, C. (1995). Efficacy of early identification and intervention. Seminars in Hearing, 16; 115-120.

2. Yoshinaga_Itano, C., Sedey, A., Coulter, D.K., and Mehl, A.L. (1998). Language of early and later identified children with hearing loss. Pediatrics, 102, 1161-1171.

3. National Institute of Health (1993). NIH panel recommends hearing screening gor al newborns. NIH Office of Medical Applications of Research.

4. Bachmann, K.R., and Hall, J.W. (1998) Pediatric auditory brainstem assessment: The crosscheck principle twenty years later. Seminars in Hearing, 19:1

5. Oates, Peggy. and Stapells. D., (1998) Auditory brainstem response estimates of the pure-tone audiogram: current status. Seminars in Hearing, 9:1, 61-85.

6. Edwards, C.G., Durieux-Smith, A., and Picton, T.W., (1985). Neonatal Auditory Brain Stem Responses from ipsilateral and contralateral recording montages. Ear and Hearing, 6:175.

7. Yang, E.Rupert, A., and Moushegian, G. (1987). A developmental study of bone conduction auditory. Ear and Hearing. 8:244-251.

8. Bachman, K.R., and Hall, J.W. (1998). Pediatric auditory brainstem response assessment: the crosscheck principle twenty years later. Seminars in Hearing. 10:1,41-60.

9. King, A.J., and Sininger, Y.S. (1992) Electrode configuration for auditory brainstem response audiometry. American Journal of Audiology. 19:40.

Manufacturers of electrodes:

1. ICS Medical Corp. Schaumburg, IL 1-800-289-2150
2. The Electrode Store, Enumclaw, WA 1-800-537-1093
3. Neuromed, Sterling, VA

For more information on ICS Medical click here.

Click here to visit the ICS Medical website.

20Q with Gus Mueller | Hearing Loss & Dementia - Highlights from Key Research | Author: Nicholas Reed, Aud |

Michelle R. Petrak, PhD, CCC-A