Pediatric Annals

Medical Management of Childhood Hearing Loss

Margaret A Kenna, MD

Abstract

Hearing impairment is a problem frequently encountered in the pediatric population. The causes of hearing loss in children are varied, including etiologies that are congenital, acquired, syndromic, or nonsyndromic. Because speech and language development during the first few years of life are dependent on the child's level of hearing, early identification and management childhood hearing loss are critical.

Fortunately, advances in therapy, technology, and surgery have provided a means of addressing significant permanent hearing loss in most patients through the use of early language therapy, externally worn amplification (hearing aids, FM systems), or surgical intervention, including bone-anchored hearing aids and cochlear implants.1·2 The importance of such early intervention has been recognized by most state legislatures, which have, during the past decade, mandated newborn hearing screening programs.3

DIAGNOSIS AND MANAGEMENT OF CHILDHOOD HEARING LOSS

Identification

The diagnosis of hearing loss has come of age. With 39 states now having mandated newborn hearing screening programs, care providers have an opportunity to intervene early. Most states screen with either auditory brainstem response or otoacoustic emission testing, with confirmation by a more detailed combination of auditory brainstem response and otoacoustic emission. However, while these programs identify most losses present at birth, they may miss acquired hearing losses of delayed onset. Similarly, if a child is not screened at birth, it is possible that a congenital loss will not be appreciated until impairment of communication skills becomes apparent.49

No child is too young or too impaired to have an accurate assessment of hearing. The choice of audiometrie test is related to the age and developmental level of the child and what information is desired. Table 1 (see page 827) lists the types of auditory evaluations that can be used. A small percentage of these children develop fairly normal and usable hearing, while the majority develop more significant hearing impairments. Because hearing aids seldom help these children, those with impaired language development despite intensive language therapy are candidates for cochlear implantation.

History and Physical Exam

Once the diagnosis of hearing loss is established, a thorough history is critical to establish risk factors and potential etiologies. Patients and the families should be queried regarding suspected date of onset, hearing difficulties encountered at home or at school, school performance and language skills, any suspicion of progression, and associated symptoms of tinnitus or vertigo. Medical history should include admission of 48 hours or greater to a neonatal intensive care unit, prematurity, hyperbilirubinemia, stigmata associated with syndromes known to include hearing loss, craniofacial anomalies (including those with morphological abnormalities of the pinna and ear canal). in utero infection (eg, cytomegalovirus, herpes, toxoplasmosis, or rubella), family history of SNHL, ototoxic exposures, and head trauma.

Children with SNHL rarely have any identifiable abnormalities on otoscopy but may have many of the abnormal features of the head and neck already described. In contrast, the etiology of most cases of CHL can be identified on physical examination. In some such cases, including foreign body, cerumen impaction, or otitis media, management at the time of diagnosis can be curative. More complex diagnoses usually require surgical intervention.

Management

Patients with CHL due to cholesteatoma require surgical management to remove the cholesteatoma and any associated infection. Patients with tympanic membrane perforations often desire closure of the perforation to improve hearing, manage recurrent otorrhea and allow swimming without concerns about water exposure. Patients with isolated congenital ossicular abnormalities may consider elective surgery to improve hearing. Correction of severe congenital anomalies of the conducting system including the pinna, external auditory canal, middle ear ossicles, and tympanic membrane may be more hallenging and do not always result in significant hearing…

Hearing impairment is a problem frequently encountered in the pediatric population. The causes of hearing loss in children are varied, including etiologies that are congenital, acquired, syndromic, or nonsyndromic. Because speech and language development during the first few years of life are dependent on the child's level of hearing, early identification and management childhood hearing loss are critical.

Fortunately, advances in therapy, technology, and surgery have provided a means of addressing significant permanent hearing loss in most patients through the use of early language therapy, externally worn amplification (hearing aids, FM systems), or surgical intervention, including bone-anchored hearing aids and cochlear implants.1·2 The importance of such early intervention has been recognized by most state legislatures, which have, during the past decade, mandated newborn hearing screening programs.3

TYPES OF HEARING LX)SS

Hearing loss may be characterized as conductive, sensorineural, mixed, or central. Conductive hearing loss (CHL) results from inability of the sound signal to reach the auditory nerve efficiently. Lesions that cause CHL may involve the pinna (as in microtia), the external auditory canal, the tympanic membrane, or the middle ear space and ossicles. In children, CHL is much more common than sensorineural loss (SNHL).

The most frequent condition leading to CHL is otitis media; however, other common etiologies include otitis externa, cerumen impaction, cholesteatoma, and foreign body of the external auditory canal. Infrequent causes include otosclerosis, middle ear masses, and severe congenital or acquired abnormalities of the temporal bone. Congenital anomalies include external auditory canal atresia or stenosis with or without microtia/atresia of the pinna, abnormalities of the tympanic membrane, and ossicular malformations or fixation. Acquired abnormalities include external auditory canal stenosis (usually from otitis externa or surgery), tympanic membrane perforations (from acute otitis media, trauma, or tympanostomy tubes), and ossicular erosion due to middle ear infection or cholesteatoma.

The etiologies of SNHL can be divided into genetic, infectious, autoimmune, anatomic, traumatic, ototoxic, and idiopathic. Genetic losses are the single most common group, and in some studies represent up to 50% of identified children with SNHL. The initial SNHL loss can be congenital or acquired, including genetic losses that may manifest long after birth. Further progression of hearing loss may occur at any time and can be gradual or rapid. In cases of bilateral loss, the progression may be asymmetric, and if the child has pre-existing SNHL in one ear, the other "good" ear may become affected as well. Subgroups of children with SNHL appear particularly at risk of progression, including those with a history of congenital cytomegalovirus (CMV) infection, congenital abnormalities of the temporal bones, such as enlarged vestibular aqueduct (EVA) and Mondini deformity, or children who were on extracorporeal membrane oxygenation (ECMO). Many genetic hearing losses may progress as well.

Mixed hearing loss refers to a combination of CHL and SNHL and means that a problem occurs in both the external or middle and the inner ear. A central hearing loss results from damage or impairment to the nerves or nuclei of the central nervous system, either in the pathways to the brain or in the brain itself.

INCIDENCE

The incidence of childhood CHL is difficult to determine due to its varying presentations. Infants not infrequently fail a newborn hearing screen due to CHL caused by otitis media or vernix/cerumen impactions; these losses often have resolved by the time of follow-up testing. Many infants and toddlers also experience intermittent mild CHL from these etiologies or from persistent tympanic membrane perforation, and in most cases the hearing loss resolves with appropriate debridement, medical therapy, tympanostomy tubes, or surgical repair of the perforation. CHL due to ossicular abnormalities, microtia/atresia, or other congenital anatomic abnormalities usually is more severe and does not resolve without surgical correction. However, the incidence of such cases is not well documented.

The incidence of congenital bilateral severe-to-profound SNHL (hearing loss of greater than 70 dB) is 1 to 2 in 1,000. If cases of bilateral mild to moderately severe loss also are considered (mild is 20 to 40 dB, moderate 41 to 59 dB, and moderately severe 60 to 70 dB), the incidence increases to 2 to 4 in 1 ,000.4 The inclusion of patients with significant unilateral hearing loss further increases this number to nearly 10 in 1,000. The incidence of fluctuation or progression among children with SNHL is unknown, although most reports have placed it at 2% to 6%. In a recent study, threshold variation, defined as fluctuation, progression, or both, was found in 21% of children with SNHL (229 of 1,1 08).5 Meticulous diagnostic audiometrie testing and physical exam of the ears is necessary to confirm the diagnosis and to distinguish CHL from SNHL.

Childhood SNHL is due to genetic causes in 30% to 50% of cases, which include both syndromic and nonsyndromic genetic deafness. Prenatal insults such as congenital infections (eg, CMV, herpes, rubella, syphilis, toxoplasmosis, varicella), as well as fetal exposure to teratogens such as alcohol, account for an additional 5% to 10% of cases. Perinatal causes, identified in 5% to 15% of patients with SNHL, include prematurity or low birth weight, hypoxia, and hyperbilirubinemia. Childhood infections such as meningitis or mumps, exposure to ototoxic medications such as aminoglycoside antibiotics, furosemide, or chemotherapy agents, and major head trauma account for 10% to 20% of cases of SNHL. The remaining 20% to 30% of hearing impairment cases in children have no identifiable etiology.

ETIOLOGY OF HEARING LOSS

Hearing Loss from Genetic Causes

Approximately 80% of congenital SNHL is recessive.6 Hearing loss genes generally are referred to as DFNA (dominant), DFNB (recessive), DFN (Xlinked), and mitochondrial. However, recent identification of many of the important genes for hearing loss has resulted in alterations in the classification of deafness. For example, syndromic and nonsyndromic forms of deafness sometimes can be manifestations of mutations in the same gene (ie, Pendred syndrome and DFNB4). Also, dominant and recessive mutations may occur in the same gene (ie, DFNA3 and DFNBl), and apparently acquired causes actually may represent a later onset of a congenital genetic abnormality, (ie, mitochondrial mutations).

For nonsyndromic hearing loss, 39 genetic loci have been identified for recessive hearing loss, 41 for dominant, 5 for X-linked, and 6 for mitochondrial.7 At these loci, 42 genes have been cloned for nonsyndromic deafness (21 recessive, 20 dominant, and 1 X-linked). Additionally, more than 500 syndromic forms of deafness have been described.8 Of these, there are several for which hearing loss is common, and many of the genes are known. These include Waardenburg, Usher, Alport, Jervell and Länge-Nielsen, Norrie, Branchio-oto-renal, Stickler, Pendred, and Treacher-Collins. Most have substantial genetic heterogeneity, often with more than one gene, and frequently multiple mutations within that gene, identified at the loci involved in each particular syndrome.7 Additionally, many mitochondrial genes for both syndromic and nonsyndromic hearing impairment have been identified.

In 1997, the gap junction beta-2 (GJB2) gene encoding the Connexin 26 (Cx26) protein was the first nuclear gene to be implicated in nonsyndromic recessive SNHL.9 Recessive Cx26 mutations account for up to 50% of all genetic hearing loss in some populations, and more than 90 mutations have been described.10 The 35delG mutation is the most common, particularly in Caucasian populations, while a second mutation, 167delT, has a high frequency in the Ashkenazi Jewish population.11·12 Although mutations in Cx26 mainly cause recessive nonsyndromic SNHL, at least five dominant mutations are associated with nonsyndromic SNHL and four with syndromic SNHL.10 The hearing loss associated with Cx26 mutations usually is moderate to severe and bilateral, but can be mild, asymmetric, and progressive.13

Pendred syndrome previously was defined as the association of recessive SNHL with goiter involving abnormal organification of iodine. In 1998, Cremers and colleagues14 reported progressive SNHL, EVA, and positive testing for pendrin (PDS) mutations despite normal thyroid function tests and the lack of physical findings on thyroid examination. The presence of EVA now is recognized as a nearly invariable part of the syndrome, although not all patients with EVA have Pendred syndrome. DNA tests for PDS mutations can be diagnostic for Pendred syndrome, and sites for this testing are increasingly available.15"17

Waardenburg syndrome (WS), the most common dominant genetic hearing loss syndrome, is characterized by dystopia canthorum (lateral displacement of the inner canthus of each eye), pigmentary abnormalities of the hair, iris, and skin (often white forelock and heterochromia iridis), and SNHL. WS I, the most common type of Waardenburg, is characterized by the presence of dystopia canthorum, although only about 20% of WS I patients have SNHL. WS ? differs from WS I in that there is no dystopia canthorum and hearing loss is observed approximately 50% of the time. There are two other types of Waardenburg syndrome; patients with WS DJ have features of WS I plus abnormalities of the upper limbs, while those with WS IV have features of WS ? plus Hirschsprung's disease. Six loci and at least seven genes have been found to be associated with WS.

The autosomal-dominant branchiooto-renal syndrome (BOR) is characterized by branchial arch anomalies, renal abnormalities, and mixed hearing loss. The syndrome is the manifestation of mutations found in the human homologue of the eyes-absent gene in fruit flies (EYAI) and maps to chromosome 8ql3.3; a second BOR gene recently has been identified on lq31. Many patients with BOR also have abnormalities of the temporal bones, including Mondini deformities, EVA, hypoplasia of the cochlea, bulbous internal auditory canals, deep posterior fossa, and acutely angled basal cochlear turns.18'20

Usher syndrome is the most common genetic form of deaf-blindness. Usher syndrome Type 1 (USHl), seen most frequently, involves progressive blindness due to retinitis pigmentosa (RP), congenital bilateral profound SNHL, and absent vestibular function. USH2 patients have moderate SNHL, progressive visual loss due to RP, and normal vestibular function. USH3, which is much less common, features progressive hearing loss, variable onset of RP, and variable vestibular function. Eleven different genetic loci have been found to be associated with the various forms of Usher syndrome; in USHl alone, seven different genetic variations have been noted.21

Mitochondrial mutations, which follow a maternal inheritance pattern, are associated with both syndromic and nonsyndromic hearing loss. The A1555G mutation is a classic example of the interaction between genetics and the environment.22·23 People with this mutation are more susceptible to aminoglycoside-induced ototoxicity at low drug levels and first-time administration, and hearing loss even without exposure to aminoglycosides has been reported. Many other mitochondrial mutations causing syndromic, nonsyndromic and ototoxicity-related hearing loss also have been described.24

Hearing Loss from Infectious Causes

There are many infectious causes of SNHL, including bacteria, viruses, and other pathogens, such as spirochetes. The incidence of many of the viral etiologies has decreased significantly due to successful vaccination programs, including those for measles, mumps, and rubella. In the United States, CMV is now the most commonly identified virus associated with SNHL at birth; in up to half of these children, the hearing loss will progress. Approximately 40,000 newborns, or 1% of the newborn population, shed CMV in their urine and saliva at the time of birth, making it the most common congenital infection in the US. Of these 40,000 neonates, approximately 6,000 to 8,000 develop clinical CMV disease sometime during infancy, and about three-quarters develop SNHL. Additionally, up to 10% 15% of infants with initially asymptomatic congenital CMV develop SNHL, mental retardation, microcephaly, or some combination of the three.

Children with congenital CMV may pass a newborn hearing screen only to manifest as later onset and progressive SNHL, especially during the first few years of life, although progression may occur over many years. The hearing losses are usually bilateral, although often asymmetric; progression can occur unevenly between the two ears, and progression to profound hearing loss is not uncommon. Children with congenital CMV often have other clinical manifestations, including prematurity, microcephaly, developmental delay, and visual impairment, making the hearing loss more difficult to manage.25

Other viruses that may cause congenital SNHL include herpes simplex, mumps, measles, and rubella. Historically, bilateral SNHL was the most common long-term clinical manifestation of congenital rubella and was present in up to two-thirds of the cases, while mumps was a common cause of unilateral profound SNHL. Fortunately, vaccination programs in many countries have reduced the incidence of these congenital infections by more than 95 %.26 However, unvaccinated children or children who have a poor response to vaccination still may develop SNHL as a result of these infections. The hearing loss associated with congenital rubella may be progressive up to 25% of the time.27

Fetal herpes simplex virus (HSV) infection is uncommon, with many pediatric cases arising in the perinatal period as a result of maternal genital infection at the time of birth. Neonatal HSV infection rates vary from 1 in 2,500 to 1 in 10,000 live births, with only 5% representing intrauterine exposure. Most neonatal HSV infections are due to HSV ?. Hearing loss occurs as a result of meningoencephalitis and is most likely due to central nervous system involvement.

Another virus that has been identified as a cause for acquired progressive and fluctuating SNHL is parvovirus B- 19, the virus that causes erythema infectiosum (Fifth disease).28 The hearing loss that occurs may be autoimmune in nature and is very uncommon compared with the incidence of Fifth disease. HIV infection also can cause acquired SNHL that is progressive.29

Bacterial meningitis continues to be a major cause of acquired SNHL in children, although effective vaccination programs are changing this. With the neareradication of Haemophilus influenzae type B meningitis in the US because to vaccination, pneumococcal meningitis has become the leading cause of meningitis and subsequent SNHL. A I heptavalent conjugated polysaccharide vaccine against Streptococcus pneumoniae was introduced in 2000, and the invasive 5. pneumoniae rate is dropping; other conjugate 5. pneumoniae vaccines are being tested as well.30

Hearing loss due to bacterial meningitis nearly always appears early in the disease course and may remain stable or progress.31 If the loss is mild or moderate, there may be some improvement with treatment of the meningitis, but profound losses seldom improve. Neonatal sepsis, often caused by group B Streptococcus or gram-negative enteric pathogens, also may result in meningitis with subsequent SNHL.27,32

Toxoplasmosis, caused by the intracellular protozoan Toxoplasma gondii, is another pathogen that is associated with congenital SNHL. The incidence of congenital toxoplasmosis is about 2 to 6 per 1 ,000 live births; this compares with the 1 % of births estimated to have congenital CMV infection. Congenital toxoplasmosis resembles congenital CMV in many of its presenting clinical features, including SNHL. As in CMV, hearing loss may progress and may stabilize with treatment. Approximately 15% to 25% of infants who are not appropriately diagnosed and managed develop SNHL. Medical treatment of the pregnant mother with toxoplasmosis prevents approximately 50% of transmissions, and timely medical treatment of infected infants can reduce chorioretinitis and SNHL substantially.33

Syphilis, caused by the spirochete Treponema pallidum, historically has been associated with SNHL in both its congenital and acquired forms. In 2000-2002, the overall incidence of congenital syphilis in the US was 1 1.2 per 100,000 live births, the lowest rate since 1991. Because hearing loss in some newborns with congenital syphilis is not present at birth, longterm follow-up is mandatory.34,35

Many other infectious causes less frequently result in SNHL. Borrelia bergdoferi, the causative agent of Lyme disease, may cause SNHL and facial paralysis among its neurologic manifestations. Other bacterial causes of meningitis besides S. pneumoniae occasionally may be associated with SNHL, including Listeria monocytogenes and other types of Haemophilus. Even varicella may rarely cause SNHL, and many other viruses, usually unidentifiable at the time of illness, are suspected to cause SNHL, especially in adults.36,37

Before the antibiotic era, otitis media was a leading cause of SNHL. Fortunately, SNHL due to acute otitis media is very rare now in countries where it is routinely treated with antibiotics. However, chronic otitis, especially chronic suppurative otitis media with or without cholesteatoma, is still not uncommonly associated with SNHL, with progression of hearing loss over time, often years. These losses usually are seen first in the higher frequencies and can be handicapping. More recently, there has been some concern that even otitis media with effusion may be associated with SNHL in the very high frequencies (above 8,000 hertz); these losses, usually undetected during routine audiometrie evaluations, can cause significant hearing difficulty in the presence of background noise and can be very frustrating.38,39

Table

TABLE 1.Methods of Hearing Assessment

TABLE 1.

Methods of Hearing Assessment

Hearing Loss from Anatomic Abnormalities

Anatomic abnormalities resulting in hearing loss can be divided into congenital and acquired. Many obvious and subtle anatomic abnormalities of the bony and membranous labyrinth are now well seen with high-resolution computed tomography (HRCT) and MRI imaging of the temporal bones.40,41 In some clinical studies, up to 20% of children with SNHL have abnormal HRCT. The most commonly identified abnormality is EVA, with abnormalites of the cochlea and semicircular canals also identified frequently.

Hearing loss that accompanies congenital abnormalities of the temporal bone ranges from mild to profound, is often progressive or fluctuating, and is frequently asymmetric. The hearing losses may be purely SNHL or mixed (sensory and conductive). These children also may have intermittent vertigo or be described as "off-balance." A precipitous drop or fluctuation in hearing can occur after seemingly minor head trauma, barotrauma (eg, airplane flights, diving, playing a wind instrument), sneezing, weight lifting, or even exposure to sudden extremely loud noises. Vertigo or imbalance also may become much more apparent during these episodes as well.

Because children with congenital temporal bone abnormalities frequently have substantial usable hearing at birth, they often acquire significant spoken language before their hearing loss worsens. This makes them very good cochlear implant candidates if the hearing loss becomes severe to profound and bilateral.

Some acquired abnormalities of the temporal bone are associated with hearing loss that is usually conductive. These include otosclerosis, osteogenesis imperfecta, fibrous dysplasia, and osteopetrosis. Benign neoplasms involving the temporal bone in children are very uncommon but include Langerhan's cell histiocytosis, Wegener's granulomatosis, acoustic schwannomas (either isolated or usually associated with neurofibromatosis-2), and vascular anomalies. Rarely, malignant neoplasms such as rhabdomyosarcoma, lymphoma, squamous cell carcinoma, mucoepidermoid carcinoma, and adenoid cystic carcinoma occur in the temporal bone in children. Metastatic malignant disease, such as leukemia, occasionally may present in this area in children. Unfortunately, radiation therapy for head and neck neoplasms may involve the temporal bone and result in SNHL.42

Hearing Loss from Autoimmune and Metabolic Causes

The incidence of autoimmune SNHL in children is unknown. SNHL may uncommonly occur as part of an identifiable systemic disease, such as juvenile rheumatoid arthritis or systemic lupus erythematosis, or as an apparently isolated finding. One autoimmune syndrome that has SNHL as one of its major presenting symptoms is Cogan's syndrome; other findings include vertigo, tinnitus, aural fullness, interstitial keratitis, and systemic symptoms including aortitis. Autoimmune SNHL loss often is asymmetric and may progress or fluctuate asymmetrically. Metabolic causes of SNHL may include hypothyroidism and type I diabetes (thought to be autoimmune in nature).4344

Hearing Loss from Trauma

Trauma resulting in SNHL in children may include temporal bone fractures, labyrinthine concussion secondary to head trauma, noise, foreign bodies in the ear canal that penetrate into the inner ear through the oval or round windows, and iatrogenic trauma, including radiation therapy, surgery, ototoxic medications, and ECMO.45 Longitudinal temporal bone fractures account for approximately 80% of all temporal bone fractures and are uncommonly associated with SNHL. Transverse fractures, however, usually result in SNHL, which over time may be stable or may progress.46

Noise is a very common cause of SNHL, even in children. The pattern of damage depends on the sound source, its frequency, intensity, and duration, and whether it is continuous or intermittent. The US Department of Labor sets the boundary between acceptable and damaging noise in the workplace at 85 dB for continuous noise. Exposure to intense (140 dB) noise of short duration may result in permanent hearing loss due to acoustic trauma, as the intensity of such sound can damage virtually any structure in the ear, including the organ of Corti and the tympanic membrane.

Among 114 children ages 1 to 19 with noise-induced SNHL, the loss was unilateral in 37% and bilateral in 63%; 90% were boys.47 Of the 47 children followed with serial audiograms, 17% had progression of SNHL. Definitive causes of SNHL included firearms (hunting and rifle range), while fireworks, loud music, recreational vehicles (motorcycle, snowmobile), and power tools or equipment were identified as other common noise sources.

Iatrogenic trauma, including radiation therapy, ECMO, and surgery, may lead to SNHL. Surgical trauma generally is secondary to drilling on the basal turn of the cochlea or on the ossicles, or due to transmitted impulses from the drill. Occasionally, SNHL results from subluxation or dislocation of the stapes. Hearing loss may occur in up to 40% of patients after ECMO, with progressive mid- and high-frequency SNHL. With both ECMO and radiation therapy, the SNHL may not be apparent intially, resulting in extensive evaluations that may not have considered these therapies as a possible cause of the SNHL.

Hearing Loss from Ototoxic Exposures

Ototoxic drugs and chemicals are an important cause of SNHL. Antibiotics, especially the aminoglycosides, are well known for their vestibulotoxicity and ototoxicity. Among these, neomycin is the most ototoxic; streptomycin, gentamicin, and tobramycin are more vestibulotoxic than ototoxic, while amikacin and kanamycin are primarily cochleotoxic.

Factors that may predispose to aminoglycoside ototoxicity include the presence of mitochondrial mutations that increase susceptibility, pre-existing SNHL, temporally related noise exposure, duration of therapy, degree of illness, prior aminoglycoside exposure, and concomitant use of other potentially ototoxic drugs, including loop diuretics and other aminoglycosides. The resultant hearing loss generally is symmetric, initially affecting the high frequencies. Although the loss may stabilize, or even improve occasionally, after the drug is stopped, progressive loss also may occur.48

Other potentially ototoxic drugs that may cause SNHL include other antibiotics (erythromycin, ampicillin, chloramphenicol, vancomycin, Vibramycin, and minocycline); antimalarial agents, especially quinine; anti-inflammatory agents, including sodium salicylate and naproxen; and antineoplastic agents, including cisplatin. Among agents that may cause congenital SNHL if used by the mother during pregnancy are alcohol (resulting in fetal alcohol syndrome), isotretinoin, thalidomide, cisplatin. and some aminoglycosides.

Hyperbilirubinemia in the newborn period also is associated with SNHL and may result in an unusual type of SNHL called auditory dyssynchrony (auditory neuropathy). The exact level of bilirubin that can result in SNHL or other manifestations of kernicterus is unclear, but the effects occur more frequently in the presence of prematurity, sepsis, and liver dysfunction.

Table

TABLE 2.Tests to Consider in the Evaluation of SNHL in Children

TABLE 2.

Tests to Consider in the Evaluation of SNHL in Children

DIAGNOSIS AND MANAGEMENT OF CHILDHOOD HEARING LOSS

Identification

The diagnosis of hearing loss has come of age. With 39 states now having mandated newborn hearing screening programs, care providers have an opportunity to intervene early. Most states screen with either auditory brainstem response or otoacoustic emission testing, with confirmation by a more detailed combination of auditory brainstem response and otoacoustic emission. However, while these programs identify most losses present at birth, they may miss acquired hearing losses of delayed onset. Similarly, if a child is not screened at birth, it is possible that a congenital loss will not be appreciated until impairment of communication skills becomes apparent.49

No child is too young or too impaired to have an accurate assessment of hearing. The choice of audiometrie test is related to the age and developmental level of the child and what information is desired. Table 1 (see page 827) lists the types of auditory evaluations that can be used. A small percentage of these children develop fairly normal and usable hearing, while the majority develop more significant hearing impairments. Because hearing aids seldom help these children, those with impaired language development despite intensive language therapy are candidates for cochlear implantation.

History and Physical Exam

Once the diagnosis of hearing loss is established, a thorough history is critical to establish risk factors and potential etiologies. Patients and the families should be queried regarding suspected date of onset, hearing difficulties encountered at home or at school, school performance and language skills, any suspicion of progression, and associated symptoms of tinnitus or vertigo. Medical history should include admission of 48 hours or greater to a neonatal intensive care unit, prematurity, hyperbilirubinemia, stigmata associated with syndromes known to include hearing loss, craniofacial anomalies (including those with morphological abnormalities of the pinna and ear canal). in utero infection (eg, cytomegalovirus, herpes, toxoplasmosis, or rubella), family history of SNHL, ototoxic exposures, and head trauma.

Children with SNHL rarely have any identifiable abnormalities on otoscopy but may have many of the abnormal features of the head and neck already described. In contrast, the etiology of most cases of CHL can be identified on physical examination. In some such cases, including foreign body, cerumen impaction, or otitis media, management at the time of diagnosis can be curative. More complex diagnoses usually require surgical intervention.

Management

Patients with CHL due to cholesteatoma require surgical management to remove the cholesteatoma and any associated infection. Patients with tympanic membrane perforations often desire closure of the perforation to improve hearing, manage recurrent otorrhea and allow swimming without concerns about water exposure. Patients with isolated congenital ossicular abnormalities may consider elective surgery to improve hearing. Correction of severe congenital anomalies of the conducting system including the pinna, external auditory canal, middle ear ossicles, and tympanic membrane may be more hallenging and do not always result in significant hearing improvement. Many patients find equivalent hearing results and less risk with bone-conduction or bone-anchored hearing aids that do not require the presence of an ear canal.

If the reason for hearing loss is not immediately apparent, a search for an etiology should be attempted. Reasons include the potential for treatment (ie, congenital syphilis), establishment of prognosis for purposes of rehabilitation, identification of associated medical issues that may be syndromally linked, and prognosis for hearing loss in siblings or progeny. Additionally, as diagnostic imaging, genetic testing, and identification of infectious organisms improve, the child who initially does not have an identifiable etiology for a permanent hearing loss may have a definite cause confirmed on follow-up a few years later.

The diagnostic tests ordered in the search for a cause should be based on the type, duration, and progression of the hearing loss (Table 2, see page 829). HRCT is very useful in identifying structural abnormalities of the temporal bones in children of any age. In the newborn with SNHL, testing for CMV, toxoplasmosis, and thyroid function should be considered. Most states mandate neonatal testing for thyroid function, and some require testing for toxoplasmosis, so these often have been performed by the time infant sees an otolaryngologist or the results of confirmatory hearing testing are known. However, if the infant's birth history is unclear, a common situation related to adoption or immigration, testing for s treatable infectious disease such as syphilis and toxoplasmosis should be strongly considered. If there is any indication that the mother may have had rubella, this diagnosis should be considered as well. Herpes is the fifth letter in 'TORCHS*' (toxoplasmosis, other, rubella. CMV herpes. syphilis), but hearing loss as its presenting symptom would be distinctly rare.

Routine testing for CMV at birth is not usually legally mandated. The definitive test for CMV is culturing the virus from the urine, although nasopharyngeal cultures and IgG and IgM titers for CMV may be highly suggestive. If CMV, toxoplasmosis, or syphilis is diagnosed, systemic therapy should be urgently considered. When ganciclovir is given early in life to newborns with congenital CMV, hearing loss may stabilize, and the overall neurologic picture may improve. This effect may or may not persist, however, when ganciclovir is stopped, and extended courses of ganciclovir have been given in some cases. In all cases of suspected congenital or acquired infection input from pediatric infectious disease specialists should be obtained, as both the diagnostic tests (which may be difficult to interpret) and the treatments for these infections are in evolution.

Although hearing loss at birth is a distinctly unusual presentation of thyroid dysfunction, thyroid function studies should be obtained if the baby has not been tested at birth or has developmental delays (babies born at home, outside the country, or in certain cultural settings may not have had the usual state -mandated studies). If the patient has a goiter or Pendred syndrome is suspected, routine thyroid studies may be normal, and more specific thyroid tests should be considered. The Perchlorate washout test was previously suggested if Pendred syndrome was suspected. However, because EVA are nearly always present in Pendred, DNA testing for PDS gene mutations is now available and can confirm the diagnosis.

After the newborn period, congenital infection is harder to document. Because congenital syphilis and toxoplasmosis may not result in hearing loss until later in childhood, they should always be considered, as they are treatable. Antibody titers for rubella, CMV, and other congenital viral infections may be positive later in childhood, but the time of infection or immunization may be impossible to pinpoint. Other infectious causes for SNHL are uncommon, but include Lyme disease and Parvovirus B 19, and EpsteinBarr virus.

Other studies that should be considered at any age include electrocardiograms and ophthalmology consultation. Long Q-T syndrome has varying clinical manifestations and, although the associated hearing loss usually is severe or profound, this may not always be the case. If there is any question about the electrocardiogram findings, a pediatric cardiologist should be consulted. Early ophthalmologic assessment may be useful to look for signs of congenital CMV or toxoplasmosis, or early vision impairment. However, although retinitis pigmentosa is one of the diagnostic features of Usher syndrome, it usually will not be evident on routine ophthalmologic examination in young children. If retinitis pigmentosa is suspected, an eiectroretinogram is needed. Early ophthalmologic assessment may be useful to identify retinitis pigmentosa but is more important to be certain that there is no impairment of a second sensory system.

Although no direct treatment for genetic SNHL is available, the hearing loss can be managed. DNA testing for many of the genes, including Cx26, Cx30, the many of the mitochondrial mutations including Al 555G, and PDS (DFNB4), is more widely available. Genetic counseling should be offered before any genetic testing, with follow-up counseling after results are known. If a specific genetic test is positive, the family may be interested in testing of other family members or prenatal genetic testing for any future children.

Ototoxic causes of SNHL may be stabilized with a change in therapy. For example, if a child receiving cancer chemotherapy develops a high-frequency SNHL, it may be possible to change the chemotherapeutic agent. Children with aminoglycoside or other ototoxicity who need further treatment may be able to be treated with something that is less ototoxic. When a child is identified with a mitochondrial gene conferring susceptibility to aminoglycoside-related hearing loss, all aminoglycosides should be avoided.

The hearing of children with noiseinduced SNHL can be stabilized by eliminating identifiable significant noise exposures. Ear protection for young hunters and target shooters is advised if the patient will not completely avoid these activities. Exposure to loud music or any anticipated work-related exposure also should be modified by avoidance or use of noise protection.

Most patients with hearing loss benefit from amplification. In a patient with a CHL, consideration usually is given initially to correction of the loss, medically or surgically. If the patient is an unsuitable surgical candidate or has significant residual loss after medical or surgical therapy, a hearing aid should be strongly considered. Most patients with predominantly conductive hearing loss derive significant benefit from appropriately fitted hearing aids. Children with SNHL also usually derive benefit from a hearing aid, although the fitting of an appropriate instrument may be more difficult, depending on the magnitude of loss and the degree of associated distortion. Patients with bilateral severe to profound SNHL who do not benefit significantly from conventional hearing aids often are candidates for cochlear implantation.

Children with congenital anatomic anomalies such as external auditory canal stenosis/atresia and ossicular and middle ear anomalies do well with bone-conduction hearing aids. This technology converts sound to movement of inner ear fluids via a bone oscillator held tightly against the mastoid process by a headband. The device must be tight to be effective through the skin and may therefore be uncomfortable. An implantable version of a bone-conduction hearing aid, the bone-anchored hearing aid, often is preferable for slightly older children with permanent conductive hearing losses who are not good candidates for or do not gain benefit from conventional hearing aids. All children being considered for amplification and other assistive devices should be evaluated by an otolaryngologist and audiologist familiar with pediatric hearing problems.

SUMMARY

Hearing loss in children is common. Advances in the identification of infectious diseases at birth or in utero, genetic testing, and diagnostic imaging now permit many infants and children to be identified and treated sooner. Treatment and rehabilitation should be instituted early so that the effects of hearing loss on communication are minimized and the child's social and academic skills maximized. CHL can usually be managed medically or surgically with subsequent return to normal or near normal hearing. Children with SNHL, and their future hearing will benefit from new antiviral agents, less ototoxic antibiotics, more focused chemotherapy, and possibly genetic therapy. Digital and programmable hearing aids, more accessible FM systems, cochlear implants, and bone-anchored hearing aids provide significant rehabilitation potential for children with even very significant hearing losses. Early identification and prevention, however, remain the best strategies to combat hearing loss in children.

REFERENCES

1. Yoshinaga-Itano C. Early intervention after universal neonatal hearing screening: impact on outcomes. Ment Retard Dev Disabit Res Rev. 2003;9(4):252-266.

2. Wake M, Hughes EK, Poulakis Z, Collins C, Rickards FW. Outcomes of children with mildprofound congenita] hearing loss at 7 to 8 years: a population study. Ear Hear. 2004:25(1): 1-8.

3. National Center for Hearing Assessment and Management Utah State University. Available at: http.7/www.infanthearing.org/legislative. Accessed October 28, 2004.

4. Nozza RD. The assessment of hearing and middle-ear function in children. In: Bluestone CD, Stool SE, Alper CM, et al., eds. Pediatric Otolaryngology. 4th ed. Philadelphia, PA: WB Saunders; 2003:165-206.

5. Brookhouser PE, Worthington DW, Kelly WJ. Fluctuating and/or progressive sensorineural hearing loss in children. Laryngoscope. 1994; 104(8 Pt l):958-964.

6. Nance WE. The genetics of deafness. Ment Retard Dev Disabil Res Rev. 2003;9(2): 1 09- 19.

7. Hereditary Hearing Loss Homepage. Available at: http://www.uia.ac.be/dnalab/hhh. Accessed October 27, 2004.

8. Gorlin RJ, Tonello HV, Cohen MM, eds. Hereditary Hearing Loss and Its Syndromes. New York, NY: Oxford University Press; 1995.

9. Kelsell DP, Dunlop J, Stevens HP, et al. Connexin 26 mutations in hereditary non-syndromic SNHL deafness. Nature. 1997; 387(6628):80-83.

10. Ballana E. Ventayol M, Rabionet R, Gasparini P, Estivili X. The connexin-deafness homepage. Available at: http://www.crg.es/deafness. Accessed October 27, 2004.

11. Morel! RJ, Kim HJ. Hood LJ, et al. Mutations in the connexin 26 gene (GJB2) among Ashkenazi Jews with nonsyndromic recessive deafness. Af Engl J Med. 1 998;339(2 1 ): 1 500- 1 505.

12. Denoyelle F, Weil D, Maw MA, et al. Prelingual deafness: high prevalance of a 30delG mutation in the connexin-26 gene. Hum MoI Genet. 1997;6(12):2173-2177.

13. Kenna MA, Wu BL, Cotanche DA, Korf BR, Rehm HL. Connexin 26 studies in patients with sensorineural hearing loss. Arch Otolaryngol Head Neck Surg. 2001 ; 127(9): 1037-42.

14. Cremers CW, Admiraal RJ, Huygen PL, et al. Progressive hearing loss, hypoplasia of the cochlea and widened vestibular aqueducts are very common features in Pendred's syndrome. Int J Pediatr Otorhinolaryngol. 1998; 45(2): 113-123.

15. GeneTests [database online]. Available at: http://www.genetests.org. Accessed October 27, 2004.

16. Usami S, Abe S, Weston MD, et al. Non-syndromic hearing loss associated with enlarged vestibular aqueduct is caused by PDS mutations. Hum Genet. 1999: 104(2): 188- 192.

17. Everett LA. Glaser BA, Beck JC. et al. Pendred syndrome is caused by mutations in a putative sulphate transponer gene (PDS). Nat Genet. 1997:17(4):41 1-422.

18. Chang EH. Menezes M. Meyer NC, et al. Branchio-oto-renal syndrome: the mutation spectrum in EYAI and its phenotypic consequences. Hum Mutat. 20O4:23(6):582-589.

19. Kemperman MH. Koch SM. Joosten FB, et al. Inner ear anomalies are frequent but nonobligatory features of the branchio-oto-renal syndrome. Arch Otolaryngol Head Neck Surg. 2002;128(9): 1033-1038

20. Abdelhak S, Kalatzis V, Heilig R, et al. A human homologue of the drosophila eyes absent gene underlies branchio-oto-renal (BOR) syndrome and identifies a novel gene family. Nat Genet. 1997; 15(2): 157-164.

21. Ahmed ZM, Riazuddin S, Riazuddin S, Wilcox ER. The molecular genetics of Usher syndrome. Clin Genet. 2003;63(6):43 1-444.

22. Prezant TR, Agapian JV, BohJman MC, et al. Mitochondrial ribosomal RNA mutation associated with bom antibiotic-induced and non-syndromic deafness. Nat Genet. 1993;4(3):289-94.

23. Fischel-Ghodsian N. Mitochondrial diseases. N Engl J Med. 2003 ;349( 1 3): 1 293- 1 294.

24. Guan MX. Molecular pathogenetic mechanism of maternally inherited deafness. Ann N Y Acad Sci. 2004 Apr;101 1 :259-271.

25. Kimberlin DW, Lin CY, Sanchez PJ, et al.; National Institute of Allergy and Infectious Diseases Collaborative Antiviral Study Group. Effect of ganciclovir therapy on hearing in symptomatic congenital cytomegalovirus disease involving the central nervous system: a randomized, controlled trial. J Pediatr. 2003:143(1): 16-25.

26. Morzaria S, Westerberg Bd, Kozak FK. Systematic review of the etiology of bilateral sensorineural hearing loss in children. Int J Pediatr Otorhinolaryngol. 2004;68(9):1 193-1 198.

27. Dyer JJ, Strasnick B, Jacobsen JT. Teratogenic hearing loss: a clinical perspective. Am J Otol. 1998;19(5):671-678.

28. Cotter CS, Singleton GT, Coman LC. Immune-mediated inner ear disease and parvovirus B-I9. Laryngoscope. I994;104(l): 1235-1239.

29. Roland JT Jr, Alexiades G, Jackman AH, Hulman D, Shapiro W Cochlear implantation in human immunodeficiency virus-infected patients. Otol Neurotol. 2003;24(6):892-895.

30. Poehling KA, Lafleur BJ, Szilagyi PG, et al. Population-based impact of pneumococcal conjugate vaccine in young children. Pediatrics. 2004;114(3):755-761.

31. Brookhouser PE. Auslander MC, Meskan ME. The pattern and stability of postmeningitis hearing loss in children. Laryngoscope. 1988:98(9):940-948.

32. Litwin CM, Hill HR. Serologic and DNAbased testing for congenital and perinatal infections. Pediatr Infect Dis J. 1997; I6(12):l 166-1 175.

33. Breugelmans M, Naessens A, Foulon W. Prevention of toxoplasmosis during pregnancy - an epidemiologic survey over 22 consecutive years. J Perinat Med. 2004:32(3):21 1-214.

34. Congenital Syphilis - United States, 2002. MMWR Morb Mortal WkIy Rep. 2004: 53(31 ):716-719.

35. Peeling RW, Ye H. Diagnostic tools for preventing and managing maternal and congenital syphilis: an overview. Bull World Health Organ. 2004:82(6):439-46.

36. Chan YC. Wilder-Smith A. Ong BK, Kumarasinghe G, Wilder-Smith E. Adult community acquired bacterial meningitis in a Singaporean teaching hospital. A seven-year overview (1993-2000). Singapore Med J. 2002; 43(12):632-6.

37. Lorenzi MC, Binar RS, Pedalini ME, et al. Sudden deafness and Lyme disease. Laryngoscope. 2003; 1 13(2):312-315.

38. de Beer BA, Schilder AG, Ingels K, Snik AF, Zielhuis GA, Graamans K. Hearing loss in young adults who had ventilation tube insertion in childhood. Ann Otol Rhinol Laryngol. 2004;113(6):438-444.

39. Bluestone CD, Klein JO. Intratemporal complications and sequelae of otitis media. In: Bluestone CD, Stool SE, Alper CM, et al., eds. Pediatric Otolaryngology. 4th ed. Philadelphia, PA: WB. Saunders; 2003:687-763.

40. Purcell DD, Fischbein N, Lalwani AK. Identification of previously "undetectable" abnormalities of the bony labyrintii with computed tomography measurement. Laryngoscope. 2003;) 13(1 1):1908-1911.

41 . Hamsberger HTR, Dahlen RT, Shelton C, Gray SD, Parkin JL. Advanced techniques in magnetic resonance imaging in the evaluation of the large endolymphatic duct and sac syndrome. Laryngoscope. 1995; 105(1 0)1 037- 1042.

42. Van de Graaf, Cass SP. Tumors of the ear and temporal bone. In: Bluestone CD, Stool SE, Alper CM, et al., eds. Pediatric Otolaryngology. 4m ed. Philadelphia, PA: WB Saunders; 2003:849-860.

43. Grasland A, Pouchot J, Hachulla E, et al.; Study Group for Cogan's Syndrome. Typical and atypical Cogan's syndrome: 32 cases and review of the literature. Rheumatology (Oxford). 2004;43(8):1007-1015.

44. Papadimitraki ED, Kyrmizakis DE, Kritikos I, Boumpas DT. Ear-nose-diroat manifestations of autoimmune rheumatic diseases. Clin Exp Rheumatol. 2004;22(4):485-94.

45. Rasheed A, Tindall S, Cueny DL, Klein MD, Delaney-Black V. Neurodevelopmental outcome after congenital diaphragmatic hernia: Extracorporeal membrane oxygenation before and after surgery. J Pediatr Surg. 2001;36(4):539-544.

46. Pansier SC, Fayad JN, McGuiit WF. Injuries of me ear and temporal bone. In: Bluestone CD, Stool SE, Alper CM, et al., eds. Pediatric Otolaryngology. 4fh ed. Philadelphia, PA: WB Saunders: 2003:829-848.

47. Brookhouser PE Worthington DW, Kelly WJ. Noise-induced hearing loss in children. Laryngoscope. 1992:I02(6):645-655.

48. de Hoog M, van Zanten BA, Hop WC. Overbosch E, Weisglas-Kuperus N. van den Anker JN. Newborn hearing screening: tobramycin and vancomycin are not risk factors for hearing loss. J Pediatr. 2003; 142( 1 ):41^t6.

49. Gnindfast KM, Siparsky NF. Hearing loss. In: Bluestone CD. Stool SE, Alper CM, et al., eds. Pediatric Otolaryngology. 4tfi ed. Philadelphia. PA: WB Saunders: 2003:306-358.

TABLE 1.

Methods of Hearing Assessment

TABLE 2.

Tests to Consider in the Evaluation of SNHL in Children

10.3928/0090-4481-20041201-08

Sign up to receive

Journal E-contents