The two-staged screening approach, recommended by the National Institutes of Health in the 1993 NIH Consensus Statement, has emerged as the preferred protocol and has been widely implemented with considerable success. This approach identifies the overwhelming majority of newborns with hearing loss in the most cost-effective and efficacious approach currently available. Additionally, it is only through the combined use of OAE and ABR screening that one is able to detect and identify those infants with auditory neuropathy and configured loss, ensuring a complete assessment of the auditory pathway.

In the late 1990s, the development of rapid, low-cost screening tests made it feasible to implement screening programs for all newborns for congenital hearing loss during the birth hospitalization. Two types of tests are commonly used in newborn hearing screening: otoacoustic emissions (OAEs) and auditory brainstem response (ABR). OAE testing assesses the function of outer hair-cells in the inner ear and provides frequency-specific information regarding preneural cochlear function. The ABR, in contrast, is a measure of neural synchrony of the auditory nerve through the auditory brainstem structures; the ABR as used in newborn screening does not provide frequency-specific information about the auditory system.

Figure 1. The parts of the auditory pathway assessed by OAE and ABR screening. OAE testing assesses the function of outer hair-cells in the inner ear and provides frequency-specific information regarding preneural cochlear function. The ABR, in contrast, is a measure of neural synchrony of the auditory nerve through the auditory brainstem structures.

The presence of OAEs and/or ABRs in response to low-level acoustic stimuli is consistent with normal hearing sensitivity. Because the two tests assess different aspects of auditory functioning, only through the combined use of the OAE and ABR tests does one achieve a complete screen of the auditory pathway.

The true sensitivity and specificity of newborn hearing screening is difficult to estimate from most screening programs. One large study measured the sensitivity and specificity of OAE and ABR using an independent “gold standard,” visual reinforcement audiometry, performed on infants ages 8 months to 1 year.¹ Both tests were found to be approximately equal in sensitivity; however, the two-stage protocol of OAE and ABR was more specific than testing with the ABR or OAE alone.

During the last decade an increasing number of hospitals have implemented newborn hearing screening programs. As more infants are being screened, several clinical issues have come to light. Two issues in particular—auditory neuropathy and configured hearing loss—have received widespread interest both for their far-reaching clinical implications as well as for the insight they provide on the developing auditory system.

What is auditory neuropathy?

Auditory neuropathy (also sometimes referred to as auditory dyssynchrony) is a term used to describe a range of disorders found in patients (from infants to adults) characterized by abnormal neural function at the level of the auditory nerve.^2-5 In patients with auditory neuropathy, clinical audiologic testing yields normal OAEs in the presence of absent or severely abnormal ABR. OAE testing allows the identification of individuals with normal outer hair-cell function despite showing abnormal ABRs. Patients with auditory neuropathy may possess normal hearing thresholds, and may, but do not necessarily, exhibit progressive auditory impairment.^6,7

Auditory neuropathy is not a newly discovered hearing disorder. However, the identification of the disorder has been made possible only in the last several years due to the availability and broad clinical use of OAEs. Several studies^8,9 have presented the paradox of patients with absent ABRs who were found to have auditory function. It is likely that, had OAE testing been clinically available, several of these patients would have been identified with auditory neuropathy. More recent studies^10-13 have characterized the disorder through the use of combined OAE and ABR testing.

How prevalent is auditory neuropathy?

Auditory neuropathy is quite rare. While estimates vary, most authors agree that from 2%-15% of infants with hearing loss may exhibit auditory neuropathy; that is, one can expect to identify auditory neuropathy in approximately 1-3 infants per 10,000 births.^14,15 Thus, assuming the high estimate of prevalence of 3 in 10,000, a hospital with 1,500 births per year can anticipate an auditory neuropathy case once every 2-3 years, on average. However, it should be noted that, due to the statistics related to hearing loss,¹⁶ many hospitals will go for long periods without encountering a case of auditory neuropathy. The great majority of infants with auditory neuropathy exhibit one or more high risk factors, including blood transfusion, hyperbilirubinemia, anoxia, low birth weight, NICU residence, or family history.^17-19

How is auditory neuropathy identified?

Auditory neuropathy is characterized by normal outer hair-cell function but abnormal auditory nerve function. Therefore, the appropriate auditory tests are those sensitive to cochlear and auditory nerve function—namely OAE and ABR tests. Because OAEs reflect outer hair-cell function, a child with auditory neuropathy is likely to yield good OAE responses, whereas the ABR, reflecting auditory nerve function, is likely to be absent or abnormal. Consequently, auditory neuropathies will not be detected using an OAE-only newborn hearing screening protocol but will be detected using the ABR. OAEs and the ABR, when used together, reflect functioning of the preneural as well as neural function in the auditory system and thereby form the most sensitive combination of tests currently available.⁶ Thus, the combined use of OAE and ABR screening on all high-risk infants will provide the maximum assurance that auditory neuropathy cases do not go undetected.

What are “configured” hearing losses?

Hearing loss is generally not uniform across frequencies; instead, hearing losses have a shape or “configuration.” Losses can be “flat” (approximately the same level of loss at all frequencies), “sloping”, in which hearing is normal at low frequencies with the level of loss increasing above a certain frequency (ie, a high frequency loss), “rising” in which a loss is exhibited at low frequencies but hearing is normal at high frequencies (ie, a reverse-slope or low frequency loss), and “notching” in which hearing is normal at high and low frequencies but there is a loss in the mid-frequencies (ie, cookie-bite loss).²⁰ Hearing loss of any configuration in infants is highly likely to progress with age.

How prevalent is configured hearing loss?

Hearing loss in the general newborn population appears to have a prevalence of approximately 3-5 per 1,000 births²¹; the probability of hearing loss is significantly higher in the high-risk populations. Configured hearing loss appears to be much more common in young children than adults. While data are currently limited, a recent study²² indicates that only approximately 18% of children with hearing loss have a “flat” audiogram, while a “sloping” audiogram is the most common configuration, found in 33% of children with hearing loss. Overall, approximately 82% of children with hearing loss exhibit a non-flat audiogram (ie, a configured loss). Thus, assuming a prevalence of hearing loss in the newborn population of 3 per 1,000, a hospital with 1,500 births per year can anticipate encountering 2-3 infants with configured loss per year, on average.

Figure 2. Two-stage screening protocol endorsed by NIH. All infants are initially screened using the OAE test. Those that refer with a possible hearing loss are then retested with the ABR. If the infant refers on the ABR, then they are referred for a full diagnostic evaluation.

How is a configured loss identified?

In order to detect configured losses, an audiometric test that is sensitive to frequency-specific hearing loss is required; in other words, the test must be able to assess auditory function on a frequency-by-frequency basis (ie, analagous to an audiogram). The OAE test meets this criterion,²³ hence configured losses can be identified using an OAE-based screening protocol. The ABR— which tests the synchronized discharge of auditory neurons in response to a brief, wide-spectrum signal—lacks frequency specificity and is therefore likely to miss, sloping, rising, and notched losses,²⁴ resulting in an elevation of false negative tests. Given that approximately 82% of children with hearing loss exhibit a non-flat audiogram, there is a significant risk of configured losses being missed if an ABR-only screening protocol is employed.

How can a screening program using two-stage testing minimize missing infants with hearing loss?

In order to maximize the probability of detecting children with hearing loss— both the more usual cases of configured loss and the rarer cases of auditory neuropathy—the ideal approach using currently available technology would be to test every baby using both the OAE and ABR screen. While in some programs this combined protocol has been implemented to great success, in many instances this is impractical for the general well-baby population due to cost or time considerations.

The most cost-effective and efficacious approach is, therefore, the two-stage protocol endorsed by many leading clinicians and medical organizations, including the National Institutes of Health.^25-27 A recent clinical study by Gorga et al.²⁷ published in JAAA, following recommendations made in the 1993 NIH Consensus Statement, showed that a two-stage newborn screening program using OAE testing, followed by an ABR test of infants who do not pass, resulted in the lowest costs-per-baby tested. The NIH model presents a screening protocol that allows for the effective identification of a broad class of hearing impairments and minimizes false negatives, while at the same time prevents excessive false positive tests from burdening the system for follow-up diagnostic evaluation.

The preferred Two-Stage Protocol for the well-baby population is shown in Figure 2 (page 32). Following this protocol, all infants not exhibiting high-risk factors are initially screened using the OAE test. Those infants passing the initial test are discharged; those not passing the initial OAE are screened using an automated ABR. Infants passing the second-stage ABR are discharged with recommended follow-up and monitoring; those referring receive a full diagnostic evaluation on an out-patient basis.

This Two-Stage Protocol identifies the majority of infants with hearing loss, including those infants with configured hearing loss, as well as those rare cases with auditory neuropathy. For the high-risk population with elevated risks of both configured loss, as well as auditory neuropathy, the current consensus is that all infants should be screened using both OAE and ABR tests. Because the high-risk population represents a small percentage of all births, such a combined protocol places little added burden (in time or dollars) on the hospital.

William F. Dolphin, PhD, is director of Sonamed Corp, Waltham, Mass, and a professor in the Department of Biomedical Engineering at Boston University.

References
1. Norton SJ, Gorga MP, Widen JE, et al. Identification of neonatal hearing impairment: evaluation of transient evoked otoacoustic emission, distortion product otoacoustic emission, and auditory brain stem response test performance. Ear Hear. 2000;21:508-28.
2. Starr A, Picton TW, Siniger Y, Hood LJ, Berlin CI. Auditory neuropathy. Brain. 1996;119: 741-753.
3. Harrison R. Models of auditory neuropathy based on inner hair cell damage. In: Sininger Y, Starr A, eds. Auditory Neuropathy: A New Perspective on Hearing Disorders. San Diego: Singular Publishing Group Inc;2000: 51-66.
4. Berlin C, Hood L, Rose K. On renaming auditory neuropathy as auditory dys-synchrony. Audiol Today. 2001;13: 15-17.
5. Starr A, Picton T, Kim R. Pathophysiology of auditory neuropathy. In: Sininger Y, Starr A, eds. Auditory Neuropathy: A New Perspective on Hearing Disorders. San Diego: Singular Publishing Group Inc;2001:67-82.
6. Hood L. Auditory neuropathy: What is it and what can we do about it? Hear Jour. 1998;5(8).
7. Berlin C, Hood L, Rose K. On renaming auditory neuropathy as auditory dys-synchrony. Audiology Today. 2001; 13:15-17.
8. Worthington DW, Peters JF. Quantifiable hearing and no ABR: Paradox or error? Ear Hear. 1980;1:281-285.
9. Kraus N, Àzdamar à, Stein L, Reed N. Absent auditory brainstem response: Peripheral hearing loss or brain stem dysfunction? Laryngoscope. 1984;94:400-406.
10. Starr A, McPherson D, Patterson J, Don M, Luxford W, Shannon R, Sininger Y, Tonokawa L, Waring M. Absence of both auditory evoked potentials and auditory percepts depending on timing cues. Brain. 1991;114:1157-1180.
11. Gravel JS, Stapells DR. Behavioral, electrophysiologic and otoacoustic measures from a child with auditory processing dysfunction: Case report. J Am Acad Audiol. 1993;4:412-419.
12. Gorga MP, Stelmachowicz PG, Barlow SM, Brookhouser PE. Case of recurrent, reversible, sudden sensorineural hearing loss in a child. J Am Acad Audiol. 1995;6:163-172.
13. Sininger YS, Hood LJ, Starr A, Berlin CI, Picton TW. Hearing loss due to auditory neuropathy. Audiol Today. 1995;7:10-13.
14. Rance G, Beer DE, Cone-Wesson B, Shepard R, Dowell RC, King AM, Rickards FW, GM Clark. Clinical findings for a group of infants and young children with auditory neuropathy. Ear Hear. 1999;20:238-252.
15. Sininger Y. Identification of auditory neuropathy in infants and children. Sem Hear. 2002;23(3):193-200.
16. Burrows DL, Owen W. Estimating positive test results in UNHS programs. Hearing Review. 2000;2: 8-12.
17. Stein LK, Tremblay K, Pasternak J, Banerjee S, Lindemann K. 1996. Auditory brainstem neuropathy and elevated bilirubin levels. Sem Hear. 17,197-213.
18. Deltenre P, Mansbach AL, Bozet C, Clercx A, Hecox KE. Auditory neuropathy: A report on three cases with early onsets and major neonatal illnesses. Electroencephalography Clin Neurophysiol. 1997;104:17-22.
19. Berlin CI, Bordelon J, St John P, Wilensky D, Hurley A, Kluka E, Hood LJ. Reversing click polarity may uncover auditory neuropathy in infants. Ear Hear. 1998;19:37-47.
20. Hall, JW, and HG Mueller. 1997. Audiologists’ Desk Reference. Vol. 1. Singular Publishing Group, San Diego. 904pp.
21. American Academy of Pediatrics. Policy Statement. Newborn and Infant Hearing Loss: Detection and Intervention (RE9846). Pediatrics. 1999;103(2):527-530
22. Pittman A, Stelmachowicz P. Hearing loss in children and adults: Audiometric configuration, asymmetry, and progression. Ear Hear. 2003; 24:198-205.
23. Hall JW. Handbook of Otoacoustic Emissions. San Diego: Singular Publishing; 2000:635.
24. Joint Committee on Infant Hearing. Position Statement. Pediatrics. 2000;106;798-817.
25. Kezirian EJ, White KR, Yueh B, Sullivan SD. Cost and cost-effectiveness of universal screening for hearing loss in newborns. Otolaryngol Head Neck Surg. 2001;124(4):359-67.
26. National Institutes on Health. NIH Consensus Statement: Early Identification of Hearing impairment in Infants and Young Children. Washington, DC: NIH, 1993:11, 1-24.
27. Gorga MP, Preissler K, Simmons J, Walker L, Hoover B. Some issues relevant to establishing a universal newborn hearing screening program. J Am Acad Audiol. 2001;12(2):101-12.

Correspondence can be addressed to HR or William Dolphin, PhD, Sonamed Corp, 1250 Main St, Waltham, MA 02451; email: wdolphin@SonaMed.com.