Otoacoustic Emissions

Otoacoustic emissions are faint sounds produced by the outer hair cells of the cochlea, either spontaneously or in response to a sound. The test is quick, objective, and needs no response from the patient.

Outer hair cells amplify sound inside the cochlea, and the amplified energy travels back through the middle ear to the ear canal, where a sensitive microphone can record it. A normal emission means that the middle ear and the outer hair cells of the cochlea work.

A small probe with a loudspeaker and a microphone is placed in the ear canal. Two types of emission are recorded. Transient evoked emissions (TEOAE) are produced in response to a brief click, are recorded over about 20 milliseconds, and cover about 500 to 4000 Hz.

Distortion product emissions (DPOAE) are produced in response to two pure tones of different frequencies. The response arises at a frequency determined by the two tones, and can be tested at specific frequencies from about 1000 to 8000 Hz, giving a frequency-specific result.

Results are shown as a pass (emission present, with enough signal over the background noise) or refer, or as a plot of the amplitude by frequency (DP-gram).

The test takes 1 to 5 minutes for each ear, and is done in a quiet room, with a sleeping or quiet baby, or with an adult sitting still.

Otoacoustic emissions are used for newborn hearing screening throughout the world, since permanent hearing loss affects about 1 to 3 in 1000 babies. Emissions are present in nearly all ears with normal hearing, and are absent when hearing loss is over about 30 to 40 dB HL, with a sensitivity of about 90 to 95% for permanent hearing loss.

Indications.

• Newborn and infant hearing screening, in the first days or weeks of life.
• Follow-up screening of children at risk of hearing loss (family history, premature birth, intensive care stay, meningitis, congenital infection such as cytomegalovirus, craniofacial anomalies, and exposure to ototoxic drugs).
• Children and adults who cannot do behavioural hearing tests (young children, learning disability, non-cooperative patients).
• Detection of auditory neuropathy spectrum disorder, where emissions are present but the auditory brainstem response is abnormal or absent.
• Monitoring of hearing in patients on ototoxic drugs (aminoglycosides, cisplatin, loop diuretics), including high-frequency DPOAE, which change before pure tone hearing.
• Monitoring of noise exposure effects in workers and musicians.
• Distinguishing cochlear from retrocochlear loss, and suspected non-organic hearing loss (emissions present with reported severe loss).
• Part of the test battery in tinnitus, sudden hearing loss, and in research.
• Not a test of hearing in the full sense: it does not measure the hearing threshold, or the function of the nerve and brain.

Contraindications and precautions.

• No contraindications.
• Wax, vernix (in newborns in the first hours after birth), debris, or fluid in the ear canal, and middle ear fluid or infection, cause a refer result without hearing loss, so ears are examined and the test is repeated.
• The probe is placed carefully in infants, with disposable tips, and the test is not done in painful or infected ears, and discharging ears.
• Noise in the room, and a restless, crying, feeding or sucking baby, increase background noise and cause a false refer result.
• Recent loud noise exposure reduces emissions temporarily.
• Patients with very small or collapsed canals may need a different tip.
• Equipment is calibrated, and probes are checked daily, since blocked probes give false refer results.

Preparation and technique.

Otoscopy is done, and any visible wax and debris are removed gently. In a baby, the test is done when the child is asleep or quiet, usually after the first feed, within a quiet room, in the first 1 to 2 days of life (and after 24 hours, since vernix and fluid clear). The probe tip is placed in the ear canal with a good seal, and the stimulus level and noise level are checked on the screen.

The device presents clicks (TEOAE) or tone pairs (DPOAE), and records and averages the responses. A pass requires a reproducible response and a signal-to-noise ratio of about 6 dB or more at several frequencies (for example, 3 of 4 or 5 frequencies for DPOAE, and reproducibility of over 70% for TEOAE), depending on the device settings.

If the result is refer, the test is repeated, and the ear is rechecked.

Normal appearances.

A normal ear shows a clear emission, with a good signal-to-noise ratio (over 6 dB) and reproducibility, at the frequencies tested. DPOAE amplitude is typically 0 to 20 dB SPL, and TEOAE responses are at levels of about 10 to 20 dB SPL in newborns, and lower in adults. Emissions decrease with age, and are higher in newborns and young children, and in women. The ears are similar.

Interpretation.

• Pass (emission present): cochlear outer hair cell function and middle ear function are normal, and hearing loss greater than about 30 to 40 dB is unlikely. It does not exclude mild loss, a neural problem (auditory neuropathy), a central problem, or later onset hearing loss.
• Refer (absent or low emission): a failure to find an emission shows a possible hearing loss of more than about 30 to 40 dB, a middle ear problem (fluid, negative pressure, perforation), or debris and noise. It is not a diagnosis of hearing loss, and needs follow-up.
• Emissions present with abnormal or absent auditory brainstem response: auditory neuropathy spectrum disorder, which needs specialist assessment.
• Emissions absent with a normal tympanogram: cochlear hearing loss, with outer hair cell damage.
• Emissions absent in one ear with an abnormal tympanogram: middle ear disease. They are rechecked after the middle ear clears.
• Reduced high-frequency DPOAE in patients on ototoxic drugs: early outer hair cell damage, before change in the audiogram.
• Present emissions with a reported severe hearing loss: suggests non-organic hearing loss, or neural loss.
• A refer result at newborn screening is followed by a repeat test, and then a diagnostic test with auditory brainstem response.

Critical values and findings needing immediate action.

• A baby who does not pass screening in either ear: referral to audiology for diagnostic testing without delay. The aim is to screen by 1 month, diagnose by 3 months, and start intervention by 6 months of age.
• A baby with risk factors (neonatal intensive care, meningitis, congenital cytomegalovirus, craniofacial anomalies) and a refer result: urgent diagnostic assessment.
• Falling emissions in a patient on cisplatin or aminoglycosides: urgent review with the treating team, to consider changes to treatment.
• Absent emissions with sudden hearing loss: emergency ear, nose and throat assessment.
• Present emissions with absent auditory brainstem response in a child: urgent referral for neuropathy assessment.

Limitations and artefacts.

• Middle ear problems (fluid, negative pressure), wax, vernix, and debris give false refer results. The false refer rate in the first 24 hours of life is higher (up to 5 to 10%) than later, which is why testing is often repeated.
• Noise in the environment, infant movement, breathing, sucking, and crying, and electrical noise from equipment lower the signal-to-noise ratio and reduce reliability.
• Emissions give no information on the type or severity of hearing loss, nor on thresholds.
• A present emission does not exclude auditory neuropathy, retrocochlear disease, central hearing disorders, mild loss, and late-onset or progressive loss.
• Emissions are very small and vary with probe fit, so the probe position is checked and repeated measurements are done.
• Normal values vary with age, device, and stimulus levels.
• Spontaneous emissions are present in some normal ears and are not used for diagnosis.
• Screening in a well baby with a pass does not rule out later hearing loss, and babies with risk factors need further checks.
• False reassurance is a danger, and parents should still be advised to seek assessment for concerns about hearing or speech development.

Special groups.

• Newborns: the main application. Babies from the neonatal intensive care unit have higher rates of hearing loss and neural loss, and need auditory brainstem response in addition to emissions.
• Infants with risk factors: follow-up at 6 to 12 months and in the preschool period, even after a pass.
• Children with developmental delay and learning difficulty: emissions help in those who cannot do behavioural tests.
• Patients on ototoxic treatment: baseline and serial DPOAE, with audiometry.
• Noise-exposed workers: early detection of damage in the outer hair cells.
• Patients with suspected non-organic hearing loss: objective evidence that the cochlea is working.
• Regions: newborn hearing screening is limited by availability of equipment and trained staff, and hearing loss is often found late, after speech delay. Portable emission devices can be used by trained nurses and community health workers.
• Causes of childhood hearing loss in many regions include infections (meningitis, measles, mumps, rubella, cytomegalovirus), ototoxic drugs, neonatal jaundice, and birth complications. Many are preventable by immunisation, safe drug use, and good perinatal care.

Follow-up tests.

• Repeat otoacoustic emissions after a short interval, and tympanometry.
• Auditory brainstem response for threshold estimation in infants and for neural function.
• Behavioural audiometry (visual reinforcement or play audiometry) as the child grows.
• Pure tone audiometry in older children and adults.
• Genetic testing (GJB2, and hearing loss gene panels), cytomegalovirus testing within the first 3 weeks of life, and metabolic and imaging studies for the cause of loss.
• Imaging (MRI, CT) for suspected malformations of the inner ear and for cochlear implant planning.
• Referral for hearing aid fitting, early intervention, and cochlear implant assessment.