Key Points
- ▪
The most common causes of congenital hearing loss in infants without apparent risk factors are asymptomatic congenital human cytomegalovirus infection and recessive genetic hearing loss.
- ▪
Single-stage screening programs that use only evoked otoacoustic emissions will fail to identify children with auditory neuropathy.
- ▪
Surveillance of hearing and speech and language status is essential for all children, whether the infant passed universal newborn hearing screening or was referred for further screening.
- ▪
It is essential to recognize that not all conductive hearing losses are temporary or related to middle ear effusion. Some cases of conductive hearing loss will require early intervention.
- ▪
Follow-up from universal newborn hearing screening should involve a team of professionals with expertise in the evaluation and management of hearing loss in children.
- ▪
To optimize success, Early Hearing Detection and Intervention programs should include an infrastructure to allow for long-term surveillance of outcome measures.
- ▪
Management decisions for children with auditory neuropathy must incorporate responses to behavioral hearing tests as well as evaluations of speech and language development.
- ▪
Management decisions for children with mild or unilateral hearing loss are influenced by comorbidities or underlying pathology such as enlarged vestibular aqueduct syndrome.
Goals and Rationale for Screening
Permanent childhood hearing loss is one of the most common significant abnormalities present at birth, occurring in approximately 1 to 3 of every 1000 otherwise healthy newborns and 2 to 4 of every 100 graduates of neonatal intensive care units (NICUs). Hearing loss can be caused by many factors including both genetic and environmental etiologies. Regardless of the cause, the negative impact of undiagnosed hearing loss on language, social, emotional, cognitive, academic, and vocational development has been well established. For that reason, having a system in place that results in early identification of hearing loss when present, coupled with early and appropriate intervention, is critical to ensuring successful outcomes. When congenital hearing loss is identified early and intervention occurs by 6 months of age, infants are able to maintain language and social/emotional development that is commensurate with their physical development; this is not true if hearing loss is diagnosed later than 6 months.
National Mandate
Because of the increased availability of technology that could obtain physiologic responses to auditory stimuli from infants, in 1988 Surgeon General C. Everett Koop issued a challenge that by the year 2000, 90% of infants born with significant hearing loss would be identified by the age of 12 months. Although at the time this challenge seemed ambitious, it was incorporated into the Healthy People 2000 National Health Promotion and Disease Prevention Objectives. An additional important milestone in the effort to promote universal newborn hearing screening (UNHS) was the introduction of the Newborn and Infant Hearing Screening and Intervention Act of 1999, also referred to as the Walsh bill. Congressman James Walsh was successful in allocating appropriations for the Maternal and Child Health Bureau and the Centers for Disease Control and Prevention (CDC) into the FY2000 budget to fund states to develop and enhance their Early Hearing Detection and Intervention (EHDI) programs. As a result, 22 renewable grants of approximately $150,000 each were initially awarded to states to support their efforts. Since passage of the legislation, 53 states and territories have received federal funds to help in the development of their EHDI programs.
Before the national mandates, UNHS was initiated in the late 1980s, beginning with Rhode Island in 1989, and followed by Hawaii in 1990 and Colorado in 1993. The goal of these programs is to reliably diagnose hearing loss and intervene during infancy in order to provide children who are deaf and hard of hearing the opportunity to participate in society with their normally hearing peers. UNHS is widespread across North America, Europe, and Australia, but is still in the development phase for the majority of infants born in developing countries. In many developing countries, diagnosis is delayed until the age of 2 years or more because family suspicion of hearing loss continues to be the primary method of detection of hearing loss. The challenge is to develop screening protocols that are community based, including those that focus on educating parents about hearing and speech and language developmental milestones.
Both the 1993 National Institutes of Health Consensus Conference and the Joint Committee on Infant Hearing (JCIH) 1994 statements endorsed testing all infants before hospital discharge, with the goals of universal detection of hearing loss by 3 months of age and appropriate intervention no later than 6 months of age. Based on these recommendations, coupled with greater availability of evoked otoacoustic emission (EOAE) and automated auditory brainstem response (A-ABR) technology, the number of EHDI programs greatly expanded. A-ABR screening can be performed by trained technicians rather than audiologists, decreasing costs and helping to overcome issues of access in areas that are underserved by pediatric audiologists. The increased number of infants and young children requiring follow-up for diagnostic testing greatly increased the demand for pediatric audiology services, which are not uniformly available across the United States. By 2001, the vast majority of states had passed legislation to mandate UNHS and/or EHDI programs. The number of infants screened for hearing loss increased from 46.5% in 1999 to 97% in 2007. CDC data show that by 2009, over 97% of newborns in the United States were screened for hearing loss, and of those babies not passing the screen, 68.4% had their hearing status established (having or not having a hearing loss) before 3 months of age.
Joint Committee on Infant Hearing
As early as 1964, Marion Downs advocated for newborn hearing screening and demonstrated that severe to profound hearing loss could be reliably detected in neonates. Shortly thereafter, in 1969 the Joint Committee on Infant Hearing (JCIH) was established, with the goal of developing more accurate and reliable methods of identifying hearing loss in infants. Whereas the initial group included representatives from three professional organizations, there are now seven member organizations. The primary activity of the JCIH has been to publish position statements, the first of which was published in 1971, which recognized the need to detect hearing loss as soon as possible but stopped short of recommending routine screening of all newborns. The 1973 position statement identified risk indicators associated with hearing loss and recommended following children with those indicators, including a family history of childhood hearing loss, congenital perinatal infection (e.g., rubella) or other nonbacterial fetal infection (e.g., HCMV or herpes virus infection), craniofacial anomalies, low birth weight, and hyperbilirubinemia. Although the statements used the term risk factor , the term risk indicator may be more appropriate because a cause-and-effect relationship cannot always be determined. Additional risk criteria of bacterial meningitis and severe asphyxia were added in 1982.
The 1990 position statement expanded the list of risk criteria for neonates to include prolonged mechanical ventilation, exposure to ototoxic medications for more than 5 days, and stigmata associated with syndromes known to include sensorineural hearing loss (SNHL). The 1990 JCIH position statement was also the first to differentiate risk indicators for neonates from those for infants, and to indicate which indicators were associated with a probability of progressive and/or delayed-onset hearing loss. These risk criteria include parental concern, head trauma, family history of delayed-onset hearing loss, the presence of neurodegenerative disorders, meningitis, intrauterine infections or chronic lung disease, administration of potentially ototoxic doses of medications, and childhood infections known to be associated with SNHL (e.g., measles, mumps).
The statement provided recommendations for initial screening of neonates and infants, as well as for early intervention for infants with hearing loss, including referral for general medical, otologic, and audiologic services, without specifying the timing of that follow-up. The 1994 JCIH position statement also specified that a “team” of professionals be included, providing guidelines for referral for intervention. The statement emphasized the importance of developing databases and tracking systems to allow for systematic evaluation of identification techniques and assessment of outcomes.
It was not until the 2000 position statement that the JCIH formally recommended (rather than endorsed) UNHS before hospital discharge. It also promoted a system that is known as the 1, 3, 6 goals of screening using a physiologic measure before age 1 month, rescreening and identification of hearing loss by 3 months, and family-centered intervention by 6 months for all infants with confirmed permanent hearing loss. This statement outlined eight principles of effective EHDI programs and also provided guidelines for follow-up of infants who passed the initial screening but had risk factors for late-onset, progressive, or fluctuating hearing loss. In addition, the importance of guaranteeing the rights of infants and families to make informed decisions and provide consent was recognized. The statement’s guidelines for EHDI programs included benchmarks or targets by which programs could be monitored as well as quality indicators that allow comparisons with benchmarks.
The position statement was updated in 2007 and included significant changes relative to the 2000 statement. For example, the definition of hearing loss was expanded to include neural hearing loss. It also recommended a separate protocol for babies with an NICU stay over 5 days using ABR in order to detect neural hearing loss (see Auditory Neuropathy , later). In addition, it recommended rescreening both ears even when only one ear failed the initial screen, and repeat screening for readmissions within the first month of life when risk indicators for hearing loss are present. Direct referral to an audiologist is recommended for NICU infants who do not pass the A-ABR in the NICU. The position statement also recommended that diagnostic and auditory habilitation services, including the selection and fitting of amplification devices, should be done by audiologists with skill and expertise in evaluating newborn and young infants with hearing loss. Although some families may not have the goal of auditory skill and spoken language for their child, in general, infants with communicatively significant hearing loss should be fitted within 1 month of diagnosis.
Additional updates that were included in the 2007 position statement included recommendation of a genetics consultation for infants with confirmed hearing loss, evaluation by a pediatric otolaryngologist, and examination by a pediatric ophthalmologist. Further, all children with any degree of hearing loss and their families should be considered eligible for early intervention services, with clear points of entry to specialty services provided by professionals with expertise in hearing loss. The committee also recommended that both home-based and center-based intervention options be available to families.
One of the other important updates included in the 2007 position statement is discussion of surveillance and screening of all infants. Specifically, the primary care provider is responsible for monitoring developmental milestones including speech and language, auditory skills, middle ear status, and parental concerns on a regular basis coupled with screening using a validated tool at 9, 18, and 24 to 30 months of age if parents or health care providers have concerns. Furthermore, referral for a speech-language evaluation and audiologic assessment are recommended whenever an infant does not pass a medical home global screen or for whom there are concerns about hearing or language.
The remaining additions to the 2007 position statement are related to communication and information infrastructure. For the EHDI process to be successful, communication among the birth hospital, the parents, the medical home, the state EHDI coordinator, and the managing audiologist must be efficient. Communication with families must be culturally sensitive, and families should be made aware of all communication options and available hearing technologies in an unbiased manner. A discussion of the family’s desired outcome is an essential component to providing guidance about habilitation options. The ultimate goal is success (including communication and academic success) for all children with hearing loss.
Variations in State Requirements
Although states have taken various approaches to newborn hearing screening and early intervention for children with hearing loss, the goals of the supporters of the Walsh bill were not only to screen the hearing loss of all newborns but to also include follow-up evaluation and intervention services provided by personnel with specialized knowledge, skill, and experience needed to serve deaf and hard-of-hearing children and their families. Further, states should have complete and current listings of local, state, and national resources for providers of early intervention programs, as well as of agencies and organizations serving the deaf and hard-of-hearing communities. Most states require hearing screening for newborns, and some specify that all hospitals or birthing centers screen infants for hearing loss before discharge. For some states, the mandate is limited to facilities that are located in counties of a certain size; in Texas, for instance, the requirement for screening is limited to birthing centers in counties with a population over 50,000. Other states provide exemptions to facilities of a certain size or indicate a benchmark for the proportion of infants that must be screened. For example, Oregon requires screening within 1 month of birth only for hospitals with more than 200 live births per year, and Nevada uses a cutoff of at least 500 births annually. A number of states have established task forces or advisory committees. In Colorado, an advisory committee on newborn hearing screening was established in 1997 with a mandate that 85% of infants born in hospitals be screened using the methodologies recommended by the committee.
The EHDI legislation in some states specifies who must pay for newborn hearing screening and follow-up. Alaska requires health insurers to provide coverage for newborn and infant hearing screening within 30 days of birth, and they must also provide coverage for follow-up diagnostic hearing testing if the infant fails the initial screen. In West Virginia, the law requires health insurers to cover newborn hearing screening and indicates that the state shall pay when the child is eligible for medical assistance. On the other hand, for uninsured babies in that state, charges for testing must be paid by the hospital or other facility in which the baby was born. Many states specify to whom the results of testing will be reported (e.g., parent, primary care provider, state EHDI coordinator) and in what time frame. As of 2011, 14 states provided the option of exempting newborns from screening if parents object to the test, although in some states, the law specifies that the objection must be on religious grounds. Whereas the EHDI legislation in some states is limited to providing direction about newborn hearing screening, or hearing screening and diagnostic follow-up testing, other state laws are more comprehensive. For example, Michigan laws not only address reporting the results of infant hearing tests to the department of health, they also specify that a child’s primary health provider shall provide age-appropriate screening and that local health departments must provide preschool hearing and vision screening services.
In addition to differences in testing modalities and in requirements for which hospitals must screen and what proportion of infants must be screened, the standards for a “pass” or “refer” also vary. The goals of UNHS were to achieve a low rate of false-positives (children with normal hearing who failed the initial screen but subsequently pass) and virtually no false-negatives (children with hearing loss who pass the screen). Since initial referral rates of 2.5% to 8% were viewed as being too high, rescreening those who fail the initial screen before discharge was recommended in order to reduce the false-positive rate. Another objective was to minimize the number of children lost to follow-up, with the hope that reducing the number of babies who required follow-up would positively affect the loss to follow-up rate. However, loss to follow-up remains a significant problem, as approximately 50% of those who refer on newborn screening will not return for follow-up or receive early intervention. Many factors may contribute to the lack of follow-up, including lack of concern on behalf of the parents or primary care providers, financial concerns, and/or lack of access resulting from transportation difficulties. In some cases, what appears to be loss to follow-up is really loss to documentation of follow-up, which can occur when families move or when the birth hospital is not in the same state as the family residence.
Given this variability, it is remarkable that virtually all states have established EHDI and UNHS programs in the years since 1993–1994. Unfortunately, most states are still unable to provide documentation of the outcomes resulting from their UNHS and EHDI programs. In addition, it is not clear what the optimal balance is between a low false-positive rate and the risk of failing to identify children with mild degrees of hearing loss or hearing loss involving only one or two frequencies. Screening at one level using a stimulus covering a broad frequency range is insensitive to detecting certain hearing loss configurations. For example, mutations in the WFS1 gene account for about 80% of hereditary low-frequency SNHL (affecting ≤1000 Hz), but this configuration is not typically detected by UNHS. It is hoped that the evolution of these programs over the coming years will include not only diagnosis and intervention of hearing loss in infants and young children but also tracking systems that enable surveillance of outcomes.
Risk Indicators for Delayed-Onset Hearing Loss and Follow-Up Requirements
In the early years of newborn hearing screening, providers’ focus was solely on those babies with risk indicators associated with hearing loss, and the initial list did not distinguish whether the risk was for hearing loss at birth or later in life. Even the best high-risk screening programs faced challenges in identifying all children with risk indicators for hearing loss. One large series found that the best follow-up was for children with ototoxic drug exposure, and the lowest rate of diagnostic testing occurred in infants with subtle craniofacial dysmorphisms such as ear pits or tags and infants with a family history of hearing loss. Physical examination of craniofacial dysmorphisms in newborns may require the expertise of a pediatric geneticist, and knowledge of the family history is also difficult to elicit. One of the obvious limitations of a screening program that focuses only on infants with risk factors is that hearing loss also occurs in infants without risk factors. The most common causes of congenital hearing loss in infants without apparent risk factors are asymptomatic congenital HCMV infection and recessive genetic hearing loss.
Implementation of UNHS has allowed early identification of congenital hearing loss in all infants whether or not they have risk indicators for hearing loss. However, despite an initial perception that UNHS obviated the need to identify or track risk indicators, it soon became apparent that a significant number of infants passed the newborn screen but later on developed communicatively significant hearing loss. Increased vigilance is required both to identify when risk indicators are present for late-onset and progressive hearing loss and to ensure that there is appropriate follow-up at regular intervals, especially for infants passing the newborn screen. In contrast to earlier recommendations that all infants with risk indicators be reevaluated at 6-month intervals for the first 3 years of life, current guidelines recommend that the timing and number of repeat hearing assessments be personalized based on the relative likelihood of a subsequent delayed-onset hearing loss. Specifically, the 2007 statement indicated that all infants who pass the newborn hearing screen and have a risk factor should have at least one follow-up diagnostic hearing test by age 24 to 30 months. In addition, earlier and more frequent testing may be indicated for children with risk indicators that are of greater concern for delayed-onset hearing loss, such as caregiver concerns regarding hearing, speech, language, or developmental delay; family history of permanent childhood hearing loss; NICU stay of greater than 5 days or for those receiving extracorporeal membrane oxygenation regardless of length of stay; congenital HCMV infection; syndromes associated with progressive hearing loss; head trauma resulting in skull fracture; neurodegenerative disorders; postnatal infections associated with SNHL including meningitis; and children receiving chemotherapy.
The rationale for this change in approach to follow-up from newborn hearing screening on a more individualized basis includes the potential burden that is placed on audiologists and on families for frequent hearing testing for the high number of infants who are cared for in NICUs in the United States, coupled with the variable likelihood of progressive hearing loss and potential need for intervention across risk indicators. For example, the follow-up and potential need for intervention of a child with mild unilateral SNHL may vary depending on comorbidities and other potential risk factors for speech and language delay. Rather than assuming that all risk indicators carry equal weight and that following a particular schedule for all children is optimal, it is essential to use a team approach to surveillance that includes the primary care provider. In addition, because not all hearing loss in children is present at birth or is associated with a risk indicator, it cannot be assumed that passing a newborn hearing screen completely rules out the possibility of hearing loss. As a result, a more inclusive strategy of monitoring all children within the medical home is recommended.
Diagnostic Methods
It is important to distinguish hearing screening from diagnostic hearing testing. The screening test is the initial test used in a larger population, usually at less cost, which identifies the subset of patients who require more comprehensive or diagnostic testing. Although hearing involves a behavioral response to sound, identification of hearing loss in newborns is not possible using behavioral techniques. For that reason, objective physiologic methods are employed for hearing screening, including EOAE and A-ABR. Screening units that are available for both technologies use algorithms to determine whether a given response falls within defined acceptable parameters and is considered to be a pass. When the response falls outside the permissible response parameters, it is labeled “refer,” which indicates the need for additional testing. The terms “fail” and “failure” are avoided as these have potential negative connotations for parents. For a UNHS program to be successful, the methods employed must be easy to use, reliable, accurate, and relatively low cost. Although both EOAE and A-ABR meet these criteria, there are strengths and limitations of each.
Evoked Otoacoustic Emission
EOAEs are sounds that are produced by the inner ear in response to an auditory signal and can be recorded using a microphone that is placed in the ear canal. Although they are measured in the ear canal, the sounds occur as a by-product of a mechanism known as the “cochlear amplifier,” which is generated by active motility of the outer hair cells and contributes to hearing sensitivity and discrimination. Because EOAEs are transmitted backward from the cochlea through the middle ear, the measurement of EOAE responses requires normal outer hair cell function as well as normal transmission of the response through the middle ear. Consequently, the presence of ear canal obstruction or middle ear pathology may interfere with obtaining EOAE responses even when cochlear function is normal.
Two EOAE methods are available for use in newborn hearing screening. Transient EOAEs are generated in response to a very brief broadband stimulus known as a click presented at a relatively high stimulus level (80 to 86 dB peak equivalent sound pressure level SPL), providing information over a wide frequency range. Robust responses occur only when hearing is 20 dB hearing level or better ; thus the presence of responses rules out most hearing losses that would require amplification. Distortion product otoacoustic emissions (DPOAE) occur in response to the simultaneous presentation of two pure tones that result in a nonlinear intermodulation between the stimulus tones within the cochlea, which generates several new acoustic components that travel to the ear canal. The lower frequency tone is typically presented at a higher intensity than the higher frequency tone, and the frequencies of the tones are established to optimize the generation of distortion products (e.g., f2/f1 ~ 1.2). Clinical DPOAE analysis typically tracks only one of the distortion products, and the stimuli are varied across the frequency range in order to generate an estimate of hearing.
Auditory Brainstem Response
The auditory brainstem response (ABR) is a series of scalp recorded electrical potentials that are generated in the auditory nerve and brainstem in response to sound. The response occurs within the first 20 msec following the onset of a short-duration stimulus, which may be a broadband click or a tone burst. Generation of the response requires transmission of the signal through the peripheral auditory system as well as synchronous activity in the auditory nervous system. Wave I of the ABR is the cochlear microphonic potential. Although a normal auditory periphery is required to generate a normal ABR response, an abnormal response may not be a result of peripheral hearing loss. Specifically, lack of neural synchrony in the auditory nervous system may result in an absent or abnormal ABR response.
For the purposes of screening, most programs are likely to use A-ABR systems. Electrodes are coupled to the baby’s head in order to record the electrical response. A click stimulus is presented at one intensity level (typically 30 to 40 dB normal hearing level), and the equipment algorithms compare the response from the baby with the formulas in the system to determine whether the response is pass or refer. Ideal systems also save the recorded response so that an audiologist can determine whether the algorithm made the correct determination.
Diagnostic Auditory Brainstem Response
Although some institutions may use diagnostic ABR test methods for their newborn hearing screening, doing so is not recommended because it is less efficient than using automated screeners. Diagnostic testing is performed by an audiologist, whereas automated screening may be performed by a technician. Diagnostic ABR testing is typically used to more thoroughly assess the auditory responses of infants who are referred during their newborn screening. Diagnostic ABR equipment allows the audiologist to control the stimulus and recording parameters in order to obtain threshold or near-threshold responses not only to broadband click stimuli but also to stimuli with greater frequency specificity such as tone bursts or pips. The characteristics of the tonal stimuli are defined in such a way that the need for an abrupt stimulus onset in order to generate neural synchrony is balanced against the desire to have frequency specificity. Because the initial degree and configuration of an infant’s hearing loss are typically estimated on the basis of diagnostic physiologic methods including the ABR, it is important to use stimuli that are closely correlated with those that are used in behavioral testing that is performed at a later age. For most infants diagnosed with significant hearing loss, the initial hearing aid fitting is based on the ABR threshold responses to air- and bone-conducted stimuli for each ear.
Auditory Neuropathy
Auditory neuropathy (AN) was first described in 1996 and is characterized by intact outer hair cell function, as evidenced by normal EOAEs and/or cochlear microphonic responses, coupled with absent or severely abnormal ABR responses ( Fig. 12-1 ). AN is more commonly identified in the NICU population than in well babies, but acquired causes of AN such as anoxia, prematurity, and hyperbilirubinemia are more prevalent in this population as well. It is estimated that 10% of infants with absent ABRs will in fact have measurable otoacoustic emissions.