Hearing Impairment Workup

Depending on the patient's history and physical findings, biochemical evidence or genetic testing may help to determine the etiology of deafness if a genetic syndrome is suspected.

Some have recommended that any child with a diagnosis of SNHL should have blood studies to search for evidence of thyroid and renal disease. Such an evaluation involves testing thyroid function, measuring BUN and creatinine levels, and urinalysis.

ECG may be useful in diagnosing a prolonged QT interval, leading to a diagnosis of Jervell Lange-Nielsen syndrome. A prolonged QT may lead to cardiac arrhythmias, resulting in syncopal episodes or sudden death.

Connexin-26 is a marker for genetic deafness (DFNB1); therefore, a test for connexin-26 might be helpful.^{[8, 24, 25]} This gene codes for GJB2 (a gap junction protein regulating the potassium concentration in the endolymph, which is important for signal transduction). There are more than 150 distinct mutations of the gene. It has been said to account for 8-60% of nonsyndromic deafness in various populations.^[26] In a European population, it accounted for 34-50% of autosomal recessive deafness and 10-37% of sporadic deafness (probably new mutations).^[27] Its carrier frequency there was 2% of the general population. A similar prevalence was found in the Midwest United States, with the gene accounting for 42% of nonsyndromic deafness and a carrier frequency of 3%.^[28]

For patients with acquired bilateral hearing loss, markers of general inflammatory disease (eg, erythrocyte sedimentation rate, rheumatoid factor) or specific markers for autoimmune inner ear disease (eg, 68-kd protein) may be evaluated.

The yield of positive findings is low; however, laboratory studies are safe and inexpensive. Positive findings raise important considerations in the management of hearing loss.

In the past, the benefit of imaging studies has been questioned. Although a positive finding on MRI or CT scanning may occasionally help to explain the defect, it does not lead to treatment options. However, some abnormalities uncovered during imaging (eg, enlarged vestibula aqueduct) may indicate a child with a sensitive ear in whom minor head trauma could worsen his or her hearing. However, MRI may help in identifying a malformation of the cochlea or the cochlear nerve (see the image below).

Such information may be critical when cochlear implants are being placed in profoundly deaf individuals. Recent work suggests the superiority of MRI in preoperative planning for candidates for cochlear implants.^[29]

At this time, accurate evaluation of hearing for children of all ages is possible. Therefore, if any adult involved in the care of a child suspects the possibility of hearing loss, an immediate referral should be made for appropriate diagnostic evaluation. Universal newborn screening does not rule out the possibility of a newly acquired hearing loss or a progressive loss that had been previously undiagnosed. The development of symptoms or physical examination findings consistent with a syndrome should be used to direct further diagnostic testing.[#screeningininfants]

Prior to reviewing specific test results for hearing loss, examining the Joint Committee on Infant Hearing (JCIH) updated position statement is crucial.^[30] Specifically, the JCIH recommends hearing screening in all infants by age 1 month; those who fail the initial test should have a thorough audiologic evaluation by age 3 months, with appropriate intervention by age 6 months. In this update, the JCIH also included auditory neuropathy and dyssynchrony in the category of neural hearing loss.^[31] Recommendations were also made regarding babies who remain in the NICU for longer than 5 days. These patients should undergo audio-evoked brainstem response testing rather than the otoacoustic emissions testing that frequently is used for newborn screening.

Specific tests for hearing loss include ABR (formally called the brainstem audio-evoked response [BAER] or automated ABR), otoacoustic emissions (OAEs), and audiometry.

The BAER is occasionally referred to as an ABR when it means audio-evoked brainstem response; in this case, the ABR is then called AABR for automated audio-evoked brainstem response. (BAER and ABR, the most common and least confusing abbreviations, are used in this article.)

ABR and BAER testing

ABR testing is based the same principle as electroencephalography (EEG). When a hearing ear is given a stimulus, the resulting electrographic activity can be followed from the ear to central areas of the brain. In the formal testing procedure for BAER, clicks or specific frequencies at different volumes can be the stimuli. CHL cannot be distinguished from SNHL with the screening test, but formal BAER testing can be performed using bone conduction testing. The sensitivity and specificity of this testing are near 100%. BAER tests frequently require sedation, and they take time and are expensive. Abnormal brain-wave activity (eg, seizure activity, significant prematurity, movement artifact) can render the results uninterpretable.

Use of the automated testing procedure for ABR has been recommended for universal newborn hearing screening. Sound clicks are presented to each ear, and 2 electrodes placed on the scalp record brain-wave activity. An internal template of what the waveforms should resemble is used to determine if the baby passes the test (waveforms match) or not (waveforms do not match).

This is like the automatic readings that appear on many ECGs; ECGs have multiple templates stored that indicate a close match to normal sinus rhythm, right bundle-branch block, anterior infarct, or other conditions. Just like the ECG, muscle movement produces changes in the waveform that the ECG reads as artifact, but the ABR reads as "fail." Additionally, like the ECG, staff with relatively little training can perform the ABR test quickly and inexpensively. The test itself has a sensitivity and specificity of about 100% and 96%, respectively.

Because ABR reflects only nerve impulses that reach the brain, it cannot be used to distinguish CHL from SNHL. In neonatal screening, the false-positive rate can be as high as 10-15% because amniotic fluid and cellular debris are retained in the neonate's ear canal. Repeat testing is often completed before discharge from the nursery, but it is optimally performed after the fluid clears (up to 1 wk). However, if the baby passes the repeat test, then it is very likely that the baby does not have a significant hearing loss.

As with the BAER, prematurity or seizure disorders may cause failing results on ABR testing because the abnormal brain-wave activity does not match the machine's internal template for passing results. In this case, formal BAER testing may be necessary because the important waves might be distinguishable from the background abnormalities when read by a trained professional. Use of the OAE is a reasonable alternative because does not depend on brain waves (except for neonatal ICU babies).

OAE testing

The concept of OAE is that certain sounds generated by the inner ear can be recorded. These sounds are present in ears that can hear and likely reflect the presence and function of structures responsible for hearing. The sounds may be spontaneous or evoked. How they are produced and why they are not produced in people with SNHL is unclear, but they are well correlated with hearing loss. Also used for newborn screening, OAE tests can be performed quickly and inexpensively by personnel with relatively little training. An earphone is placed over the ear of a resting neonate, and the machine generates a sound and then records the evoked response. The sensitivity and specificity reported with evoked OAE are 100% and 82%, respectively.

By definition, OAE cannot be used to diagnose retrocochlear deafness nor can it be used to distinguish CHL from SNHL. OAE had slightly elevated false-positive rates in most studies of neonatal hearing screening, probably because a sound must pass both in and out of the obstructed canal to be recorded. (With ABR, the sound only needs to enter in through the ear canal). OAE also seems to have a high failure rate when it is used in the neonatal ICU. For healthy term babies, the “2-step protocol” recommends that babies who fail an initial OAE may have follow-up testing with either another OAE or ABR testing.

Audiometry

Routine audiometry is performed by placing headsets over the ears of children whose developmental age is at least 4-5 years and who can be instructed to raise the corresponding hand when a sound is heard. Pure tone sounds can be presented so that specific volumes at specific frequencies can be documented. CHL and SNHL can be differentiated, and speech recognition can also be tested. It is the criterion standard for patients with normal mental status and the ability to cooperate with the testing procedure. The only limits to the sensitivity and specificity of the test are the patient's ability to understand the instructions and his or her willingness to cooperate.

Pure-tone audiometry can be performed as a quick and easy screening test. It has proven to be an effective tool in schools. The disadvantages of pure-tone audiometry are that formal evaluation takes time and considerable equipment and that it can be fully performed only in older, cooperative patients.

Behavioral (visual reinforcement) and conditioned play audiometry can be completed in children as young as 6 months. Children can be conditioned to look at a puppet or a light show when a pure sound stimulus is presented or their name is called from one side of the room. If the evaluators are reliable, they can judge whether the child is cooperative and responding to cues other than the sound stimulus. In general, this test is fairly successful for identifying hearing loss in children. Disadvantages are that it requires considerable time and equipment, it cannot be used to distinguish CHL from SNHL, and it succeeds only if the child is cooperating.

Tests to avoid

Assessing responses to clapping, rattling keys, and snapping are poor tests of hearing. A child may respond to the visual or tactile stimulation rather than to the noise (eg, detect the slight breeze from a clap, accidental touching of the face with keys or fingers, or see the hands moving). Furthermore, the noise created is frequently more than 50 dB and, therefore, not useful in detecting mild and moderate losses.

A few companies market a small wand that produces white noise or a click at fixed or variable volumes. These wands have limited utility in rapid screening done in the office. However, if the time is taken to use them properly, they may provide some useful information.