Although hearing loss may be associated with externally visible stigmata apparent to pediatricians, more often than not it exists as an invisible disability that remains undetected for long periods before eventual diagnosis. The process of detecting hearing loss is different in newborns and infants versus school-aged children. But early detection and remediation is important in both groups to ensure normal speech and language and educational, social, psychological, and vocational development. Even unilateral hearing loss, which until recently was considered to exert little significant impact, has now been associated with poor school performance, increased likelihood of repeating a grade, and behavior problems.1 There is universal agreement that detection of hearing loss is much more important than the method used to do this: there are varying approaches that differ in extent and methodology for screening hearing.
Early diagnosis of hearing loss is necessary and should never be delayed by a "wait-andwatch" approach. If parents raise concern about a child's hearing, speech, and language development or lack of school performance due to inattention, an appropriate ear examination and audiometrie assessment should be done promptly. The first 3 years of life are the most critical for language learning and the development of normal speech. Unfortunately, our failure to respond to parental concerns and take appropriate measures to diagnose hearing loss contributes to delay. The average age when congenital or earlyonset losses are first identified is 2Vi years.2 With the widespread use of newborn hearing screening programs, this age can be reduced dramatically.3 It is now possible to measure the hearing ability of any child, regardless of age or ability to cooperate.
CLASSIFICATION OF HEARING LOSS
Hearing losses are classified as sensorineural, conductive, or mixed (a coexistence of sensorineural and conductive losses in die same ear). Sensorineural loss is due to a defect in the cochlea or the auditory pathway and may be acquired or hereditary. Acquired sensorineural hearing loss can occur at any age or may even be congenital (ie, cytomegalovirus infection in utero). Hereditary losses may be congenital or delayed in onset, becoming apparent in the first years of life or later. Although the differentiation of these various types of hearing loss is important for prognosis and for proper medical or surgical therapy, further discussion of these parameters is beyond the scope of this article and much of this is covered elsewhere in this issue. This article is about detecting all hearing losses, regardless of type.
SCREENING FOR AND DETECTION OF HEARING LOSS
Because 2 per 1,000 live births result in children with permanent hearing loss, testing of newborns by nonbehavioral techniques that are easily applicable to large numbers of newborns is desirable to begin effective habituation at the earliest possible age. Once newborns leave the nursery or the neonatal intensive care unit, parental concern, physician observation, and school hearing tests become the usual means of detecting problems.
SCREENING FOR HEARING LOSS IN NEWBORNS AND INFANTS
The National Institutes of Health issued a consensus statement in 1993 on the early detection of hearing impairment in infants and young children.2 This statement was produced following a multidisciplinary conference, and recommended universal hearing screening for all infants within the first 3 months of life. A two-stage screening process was proposed. All infants would first be screened by testing for evoked otoacoustic emissions. This uses low-level sounds generated by 98% of normal inner ears and detectable by a probe in the ear canal. Those who failed the first stage would be tested with auditory brainstem response audiometry, by which minute electrical responses to auditory stimuli generated by the auditory pathway are recorded from surface electrodes on the scalp. Those failing auditory brainstem response testing would be considered screening failures and referred for a complete diagnostic evaluation.
The Joint Committee on Infant Hearing is a multispecialty group of pediatricians, otolaryngologists, and audiologists who have endorsed the goal of universal detection of hearing loss as early as possible. The Committee has published a position statement that stresses the need to identify hearing losses that may not occur until after 3 months of age.4 The Joint Committee has developed a list of indicators of conditions possibly associated with hearing loss (Table 1). There are indicators of the need for neonatal screening (when universal screening is not available), as well as indicators for screening after the first 29 days of life and for rescreeriing or periodic monitoring of hearing tests to detect hearing loss of delayed onset. Currently, universal screening is an idea, but not a reality. Only two states (Rhode Island and Hawaii) have published the results of their universal screening programs.3,5 Therefore, in the absence of universal screening it is important for pediatricians to send neonates who have the indicators listed in the first part of Table 1 for hearing evaluation.
A family history of hereditary sensorineural hearing loss is an important indicator for neonatal hearing screening. However, only approximately 50% of childhood hearing losses are of genetic origin. The remainder are due to acquired or developmental factors, or are of unknown etiology (some of which may be genetic). Only approximately 20% of genetic hearing loss is inherited as an autosomal dominant trait. Recessive inheritance is responsible for approximately 80% of cases and sex-linked inheritance for only 1% to 2%. Because most genetic hearing loss is of a recessive type, there is a high likelihood of children with normal hearing from two deaf parents. This is because it is highly unlikely that two deaf parents would be carriers of the same recessive gene. Genetic hearing loss should be suspected if a neonate fails screening tests for hearing and (1) has a family history of hearing loss, (2) is the product of a consanguineous marriage, or (3) if the mother has a history of multiple stillbirths or miscarriages (indicative of a possible genetic disorder).
Approximately half of recessive hearing losses are associated with recognizable syndromes, of which there are hundreds. The physical findings associated with hearing loss are listed in Table 2, as modified from Brookhouser.6 These signs should be searched for in children who fail screening tests because they may define the cause of the hearing loss. Their presence should also alert the pediatrician to the possibility of hearing loss and the need for a hearing evaluation, especially for those cases with findings associated with craniofacial syndromes.
Indicators for Hearing Screening or Rescreening
Neonatal screening tests miss those who will have hearing losses after they leave hospital. The more common conditions that into this category are hereditary hearing loss delayed onset and sequelae of membrane oxygenation, meningitis, drugs, and cytomegalovirus infection. Also, tory neuropathy, an uncommon condition with a normal peripheral hearing mechanism but a sumed defect in the auditory pathway, may be detected by neonatal screening programs use only otoacoustic emissions. Continuing surveillance is required in the first years of life to diagnose hearing loss in all of these conditions
Soon after the National Institutes of issued its consensus statement in 1993 recommending universal screening of newborns for hearing loss, there was some controversy questioning the feasibility, risks, and benefits of such a program and challenging its justification.7'8 The publication of the Rhode Island experience with universal screening demonstrated feasibility of universal screening. Of 99.7% of newborns screened by technicians, nurses' aides, or nurses, 101 cases of confirmed permanent hearing loss were found in 53,121 infants tested during 4 years.3 The mean age of hearing loss confirmation declined from 8.7 to 3.5 months, and the mean age at which hearing aids were applied declined from 13.3 to 5.7 months as experience grew dur-
Findings Associated With Hearing Loss Syndromes
ing the 4-year initial period. This study has left little doubt concerning the effectiveness and cost-benefit ratio of universal screening, at least for the United States. Until screening of all newborns for hearing loss becomes a reality, pediatricians must be alert to those newborns who should be screened and must remain vigilant in observing those infants who pass screening but are at risk for later hearing loss.
DETECTION OF HEARING LOSS IN YOUNG CHILDREN
Between the newborn period and the routine audiometrie testing done in preschool, kindergarten, and above, hearing loss is usually found because of parental concern or because the child's primary care physician had doubts about the child's speech milestones. Such doubts are raised if there is no startling or awakening to loud sounds in the first 3 months of life. By 6 to 9 months of age, parents should expect baby sounds having some inflection and first words (typically "mama"). By the first year, babies respond to their names and follow simple instructions. They mimic sounds and have a vocabulary of several words. After VA to 2 years, there is a cascade of speech development with a vocabulary of 25 to 50 words. By 3 years there should be concern if a child is not speaking in brief intelligible sentences and using a vocabulary of several hundred words. Although there is a wide range of normal in attainment of these milestones, failure to demonstrate them should arouse suspicion of hearing loss and induce consideration for a hearing evaluation.
Other conditions that should alert physicians to the possibility of hearing loss are all of the risk factors shown in Table 1, declining school performance, and behavioral changes (especially the withdrawal from interaction with playmates). A tendency to play sound systems or television sets at high volumes has alerted many parents to hearing loss in their children. Although these indicators do not translate universally into the presence of hearing loss, they do raise the suspicion and this suspicion should not be ignored. Parental concern alone is probably of greater value in predicting hearing loss than are the simple behavioral tests of hearing that can be done by a primary care physician during physical examination.
Audiologic evaluations can be performed on children of any age, with some limitations. From birth onward, the auditory brainstem response test is feasible because it does not depend on patient cooperation. This test is an electrophysiologic measure of auditory nerve and brainstem activity. Because it does not measure cortical activity, it is not truly a test of "hearing." From approximately 9 months to 2% half years of age, visual reinforcement audiometry is used to determine a child's behavioral responses to speech and sounds of specific frequency and amplitude delivered by loudspeakers. The child's behavioral conditioning is reinforced by lighting up toys in response to correct turning toward the loudspeaker delivering a stimulus. Between 2?i and 4 years, play audiometry can usually be performed. The sound stimuli are delivered through earphones so ear-specific information can be obtained. This is not possible when loudspeakers are used for younger children. In play audiometry, a child is conditioned to move a toy (such as a peg or a block) in response to sounds. After the age of 4 years, conventional audiometry can be done, depending on the maturity and the cooperation of the child. At all ages, auditory brainstem response and evoked otoacoustic emission testing can be done when patient cooperation is not possible, and they provide a good estimate of hearing ability.
Audiologic testing (and retesting) is needed to characterize the type of hearing loss (sensorineural, conductive, or mixed), its severity, and the audiometrie pattern of frequency involvement, symmetry, and stability or progression over time. Tympanometry may be needed to define the contributon of middle ear disease or conductive loss to hearing impairment. Habilitation or rehabilitation is dependent on the early detection and accurate characterization of a hearing loss. In infancy, the early application of a hearing aid is almost always the initial step to prevent loss of critical developmental time. In older children, other alternatives are added, such as myringotomy and tubes, tympanoplasty surgery, and cochlear implants. These are selected based on reliable audiometrie assessments. Older children may need to be placed in special educational programs for the hard of hearing or may need preferential seating or FM amplification systems in their classrooms to hear teachers and fellow students.
Enormous strides have been made in the second half of the 20th century, and we are now able to detect hearing loss in newborns and children of all ages. Contemporary techniques of audiology and radiologic imaging allow accurate characterization and etiology of most childhood hearing loss. However, the cause of some impairments remains elusive. The great progress in understanding genetics has clarified our concepts of inherited hearing loss and will continue to do so at an accelerating rate in the near future. The etiology of hearing loss in childhood is constantly changing as some infectious causes are reduced by vaccines (eg, rubella, measles, mumps, and Haemophilus influenzae) and new technologies increase other problems (eg, survival of low-birth-weight infants, or ototoxic drugs). Aggressive attitudes toward detecting hearing loss in neonates and in young children are justified by feasible screening programs, dependable hearing tests available for all ages, and effective therapeutic modalities.
1. Brookhouser PE, Worthington DW, Kelly WJ. Unilateral hearing loss in children. Laryngoscope. 1990;101:1264-1272.
2. National Institutes of Health. Early identification of hearing impairment in infants and young children: NIH consensus statement. 1993;11:1-24.
3. Vohr BR, Carty LM, Moore PE, Letourneau K. The Rhode Island Hearing Assessment Program: experience with statewide hearing screening (1993-1996). / Pediatr. 1998;133:353-357.
4. Joint Committee on Infant Hearing. 1994 position statement. ASHA. 1994;36:38-41.
5. Johnson JL, Kuntz NL, Sia CC, White KR, Johnson RL. Newborn hearing screening in Hawaii. Hatoaii Med J. 1997;56:352-355.
6. Brookhouser PE. Sensorineural hearing loss in children. Pediatr Clin North Am. 1996;43:1195-1216.
7. Bess FH, Paradise JL. Universal screening for infant hearing impairment: not simple, not risk-free, not necessarily beneficial and not presently justified. Pediatrics. 1994;93:330-334.
8. White KR, Maxon AB. Universal screening for infant hearing impairment: simple, beneficial, and presently justified, Int J Pediatr Otorhinolaiyngol. 1995;32:201-211.
Indicators for Hearing Screening or Rescreening
Findings Associated With Hearing Loss Syndromes