Nongenetic Hearing Loss



Nongenetic Hearing Loss


Margaret A. Kenna



INTRODUCTION

Hearing loss is the most common congenital sensory impairment. Bilateral severe to profound hearing loss is present in 1 to 2/1,000 live births, and if unilateral and mild to moderate hearing losses are included, the number rises to 4/1,000. According to a recent National Health and Nutrition Examination Survey (NHANES) study, by the age of 12 to 19 years, the rate is 19.5% for any degree of hearing loss (1). The purpose of pediatric hearing evaluation is to identify the degree and type of hearing loss and its etiology and outline a comprehensive strategy that supports language, communication, and social development. Significant strides have been made in the early identification of hearing loss, as well as in diagnostic studies and long-term management. This chapter reviews newborn hearing screening and many of the nongenetic causes of hearing loss in infants and children.


History of Newborn Hearing Screening

The evolution and institution of newborn hearing screening occurred for several reasons: (a) availability of technology that provides rapid and accurate identification of hearing loss in newborns; (b) recognition that hearing can be tested easily and accurately in newborns; (c) recognition that early intervention in an infant with hearing loss supports significantly improved language outcomes compared to those identified later; and (d) the development of technologically improved assistive listening devices, including hearing aids, FM systems, and cochlear implants, which provide markedly improved hearing and awareness of useful sound in almost any infant or child with hearing loss.

Hearing screening has been performed in schoolaged children in the public schools since 1927 and was able to identify hearing loss that could be educationally handicapping (2). Subsequently, in recognition that early identification of hearing impairment might lead to improved language, attempts were made to screen younger children aged 2 to 5 years. In 1979, Victor Goodhill, in his book Ear Diseases, Deafness and Dizziness, notes that “Parents are often incorrectly told that a child of this age cannot have accurate hearing testing” (3). However, appropriate services to support language development would not be provided until there was concrete evidence of the hearing loss.

In 1970, the Joint Committee on Infant Hearing (JCIH) recommended infant hearing screening as a research need and also recommended the widespread use of a high-risk register, which included one or more of the following: (a) history of hereditary childhood hearing impairment; (b) rubella or other nonbacterial intrauterine fetal infection (e.g., cytomegalovirus [CMV], herpes); (c) defects of the ear, nose, or throat; (d) malformed, low-set, or absent pinnae; cleft lip; or palate; (e) any residual abnormality of the otorhinolaryngologic system; (f) birth weight less than 1,500 g; and (g) bilirubin greater than 20 mg/100 mL serum. They further recommended that infants falling into any of these categories should be referred for an in-depth audiologic evaluation during the first 2 months of life, with regular follow-up evaluations even if the initial testing was normal. However, since only about 10% of newborns fall into the “high-risk” category and 50% of infants with communicatively significant hearing loss do not have any obvious high-risk features, it is apparent that the use of the high-risk register alone as a basis for hearing screening would miss up to half of the affected infants (4).

One of the significant early problems with newborn hearing screening was the lack of a reliable, rapid and easy to perform, and economical audiometric test. In 1896, Dr. Thomas Barr, in his Manual of Diseases of the Ear, noted that “we may employ the sound of a bell, a sharp whistle, or a very loud tone of voice, taking care that the
child does not see the source of the sound” (5). This essentially reflected the state of the art until the 1940s, when, in recognition of the difficulty in eliciting reliable behavioral responses in young infants, nonbehavioral audiometric techniques were developed. These included, in the 1940s and 1950s, electrodermal response audiometry, which used a conditioning paradigm that paired an auditory stimulus with a mild electric shock, with measurement of the galvanic skin response as an outcome. Additional tests based on conditionable autonomic responses included heart rate response audiometry (1950s to 1960s) and respiration audiometry (1960s to 1970s).

Subsequently, the Crib-O-Gram was developed by Blair Simmons at Stanford University (1974). It involved a motion-sensitive transducer that was placed on or under the infant’s crib. A strip chart recorded any infant motor activity stronger than an eye blink or facial grimace before, during, and after presentation of auditory stimuli. The test was automated, noninvasive, and fairly easily performed. Using the Crib-O-Gram, the Stanford research group tested over 12,000 infants from 1974 to 1984. They found an incidence of hearing loss of 1:1,000 in the well-baby nursery and 1:52 in the neonatal intensive care unit (NICU), similar to current numbers for bilateral severe to profound hearing loss obtained using newer physiologic techniques. However, a number of variables affected the results, including a need to test when ambient noise was low and when the newborn’s behavioral state was “right.” Other issues included uncertainty about the optimal auditory stimulus, high false-positive and false-negative rates for babies with less than severe to profound losses, lack of ear-specific information, and poor test-retest reliability (6). The Auditory Response Cradle, developed in England in 1980 and similar to the Crib-O-Gram, was designed to evaluate several specific infant motor responses following programmed auditory stimuli. However, it was never widely used in the United States (7, 8).

By 1986, six states had active statewide infant hearing screening programs, and in 1990, Blake and Hall reported that although 14 states had legislative mandates for newborn hearing screening, most of the other states still had no statewide policy. In addition, many states with legislative mandates did not have full implementation due to lack of funding, diversity of geography and population, lack of consensus about the most effective screening methods, and lack of follow-up programs (9). In 1990, the United States Public Health Service established the Healthy People 2000 project (10). Recommendations included a goal to “reduce the average age at which children with significant hearing impairment are identified to no more than 12 months.” In 1993, the National Institute on Deafness and Other Communication Disorders sponsored a consensus conference on the Identification of Hearing Impairment in Infants and Young Children (11). The panel recognized that 50% of newborns with significant hearing loss did not have any high-risk factors and only about 10% of all newborns fell into any of the high-risk categories. Because use of the high-risk register alone would continue to miss many infants with congenital hearing loss, the panel recommended that universal newborn hearing screening be implemented during the first 3 months of life. A two-stage protocol of evoked otoacoustic emission (EOAE) screening followed by auditory brainstem response (ABR) screening for infants who failed the EOAE screen was recommended. In their 1994 position statement, the JCIH recommended that all newborns with hearing loss be identified by 3 months and that intervention be started by 6 months. The statement did not recommend ABR over otoacoustic emission (OAE) but recognized the advantages and limitations of each technique (12).

Three early universal hearing screening programs gave further weight to the need for universal newborn hearing screening. The first, in Florida in 1983 at Winter Park Memorial Hospital, screened 15,000 infants between 1983 and 1993 using automated ABR. Improved prenatal parent education and increased physician awareness that screening was possible were cited as two major positive outcomes. The second, in Colorado, occurred over three decades and by 1997 involved 22 hospitals. An incidence of 4:1,000 of bilateral or unilateral hearing loss was found, with no high-risk factors noted in 57% of the hearing loss group. The third study evaluated the Rhode Island Hearing Assessment Program. Based on extensive research on the detection of newborn hearing loss performed in Rhode Island in late 1980s and early 1990s, Rhode Island Public Law 23-23-13 became effective on July 1, 1993, mandating universal hearing screening of infants born in all 8 RI hospitals. Detection of transient evoked otoacoustic emissions (TEOAE) was chosen as the screening method, and all infants who failed a first TEOAE screen were rescreened using TEOAE. From January 1, 1993, to December 31, 1996, 52,659 babies were screened. Of the 10% referred for a second stage screen, 111 were found to have a permanent hearing loss, 45% NICU graduates and 55% well-baby nursery graduates. The NICU hearing loss rate was 9.75% and the well-baby nursery rate was 1.27%, for a combined rate of 2.12% (8).

In 1998, Christine Yoshinaga-Itano and her group in Colorado published the results of a study looking at language acquisition in children with hearing loss that was identified early (before 6 months) versus late (after 6 months). For all hearing impaired groups, there was significantly better language development in the early identification group (13). Finally, in 1999, the American Academy of Pediatrics published “Newborn and Infant Hearing Loss: Detection and Intervention” (14). This report recommended universal newborn hearing screening by 3 months of age with intervention by 6 months of age. The recommendations for screening were based on the fact that the five criteria for universal newborn hearing screening had been met: (a) easy to use test(s) with a high degree of sensitivity and specificity; (b) there was no other way to
detect the condition clinically (and in a timely fashion); (c) effective interventions were available; (d) early identification and intervention were associated with an improved outcome; and (e) the screening procedure was within an acceptable cost-effective range.

The current national goals for hearing screening were developed by the federal Early Hearing Detection and Intervention program. The first three goals, often called the 1-3-6 plan, are (a) all newborns will be screened for hearing loss by the age of 1 month, preferably prior to discharge from the hospital; (b) all newborns who screen positive will have a diagnostic audiologic evaluation before the age of 3 months; and (c) all infants with an identified hearing loss will receive appropriate early intervention services by the age of 6 months (15). The 2007 JCIH guidelines have been adopted by their member organizations, including the Alexander Graham Bell Association for the Deaf and Hard of Hearing, the American Academy of Audiology, the American Academy of Otolaryngology-Head and Neck Surgery, the American Academy of Pediatrics, the American Speech-Language-Hearing Association, the Council on Education of the Deaf, and the Directors of Speech and Hearing Programs in State Health and Welfare Agencies (16). In addition, hearing screening by age 1 month has been incorporated as one of the Healthy People 2020 Objectives (17, 18). Currently all states and the District of Columbia screen over 95% of their newborns (19). However, neither the newborn hearing screening laws nor the implementation are uniform. Payment for hearing aids and early intervention services vary greatly, and follow-up continues to be a significant challenge in most states. Some of the reasons for loss to follow-up include underfunded newborn screening programs, challenging geography (rural areas with difficult access to follow-up facilities), cultural diversity, home births, between hospital transfers, and out of state births.


What Are the Downsides of Newborn Hearing Screening?

Newborn hearing screening is not without areas of concern. The physiologic tests used, ABR and OAE, are not 100% sensitive and specific; both false positives and false negatives do occur. The tests may need to be repeated several times, first in a screening mode and then in a diagnostic mode to give a definitive diagnosis. An infant may pass a screen only to be found after discharge to have a hearing loss, this may represent a false-negative test, the infant may have hearing loss that is too mild or in a frequency range that it is not detected by the screen, the hearing loss develops after discharge, or the screen used OAEs and the infant has auditory dyssynchrony. Errors in pass/fail parameters (i.e., how the machine is set) may “pass” patients when they should actually “fail” or “refer.” It is also statistically possible that an infant can “pass” if the screen is repeated often enough, as most screens are designed with a balance in their sensitivity and specificity; for this reason, most newborn screening programs only allow one repeat screen, and if the infant still fails, then a diagnostic auditory evaluation is recommended. A “pass” may actually delay a hearing loss diagnosis. If the baby “passed” in error, or if a hearing loss occurred after birth but the baby passed the initial screen, the parents (and the primary care physician) will have a false sense of security, even if they feel the baby does not hear normally. A third possibility, recognized since screening programs became widespread, is that in protocols that use OAEs to screen rather than ABR, patients with auditory dyssynchrony spectrum disorder will often be missed.


Tests Used for Newborn Hearing Screening

Since auditory dyssynchrony spectrum disorders comprise up to 10% of hearing loss in infants, the use of both ABR and OAE as part of a hearing screening program in many ways would be ideal. In many cases, ABR or OAEs alone are able to detect even mild conductive or sensorineural hearing impairments. However, the use of both modalities is needed to make a definitive diagnosis of auditory dyssynchrony, especially in a neonate. In an infant with auditory dyssynchrony, robust OAEs may be present, yet on ABR tracings only a cochlear microphonic, followed by indistinct or absent ABR waveforms, will be seen. In screening programs using only OAEs, the diagnosis of auditory dyssynchrony may therefore be missed. It is important, therefore, that no matter which test is used for screening, follow-up diagnostic testing should utilize both modalities (20) (see Chapter xxx for further discussion of ABR and OAE testing).


The Role of the Otolaryngologist

The otolaryngologist is often one of the first physicians that the family and infant or child will meet in the course of the diagnosis of hearing loss. They may meet them at the time of their initial diagnostic ABR, for medical clearance for a hearing aid, or for further investigation into the etiology of the hearing loss. To be effective, the otolaryngologist must have some basic understanding of the support services needed and how to access them, and they should also be able to provide the family with some basic information about the possible causes of hearing loss. The otolaryngologist should be prepared to have at least a working knowledge of OAEs and ABR. and behavioral audiometry, as well as interventions including hearing aids, FM systems, and cochlear implants.


CAUSES OF HEARING LOSS IN INFANTS AND CHILDREN

Defining the cause of the hearing loss can provide prognostic as well as educational information to the family. Careful evaluation can now pinpoint a definite or probable cause of the hearing loss 50% to 60% of the time. The etiologies
of hearing loss have often been divided into congenital and acquired. However, many of the causes that are congenital were “acquired” in utero and may only present at a later time (e.g., delayed onset of sensorineural hearing loss [SNHL] from CMV); this is in contrast to those truly acquired after birth (e.g., secondary to extracorporeal membrane oxygenation [ECMO], meningitis, or trauma). Although it is important to know if the hearing loss is pre- or postlingual in onset for management purposes, a more useful way to look at hearing loss etiology is to divide the etiologies into infectious, anatomic, genetic, traumatic, ototoxic (and other issues related to NICU stays), and other. The degree and laterality of the hearing loss in many of the causes of hearing loss are similar, with potential overlap between some of these causes, making the identification of a definite etiology more challenging. Some examples of this include children with enlarged vestibular aqueducts (EVAs) and mutations in the SLC26A4 gene (a genetic cause with an anatomic presentation); NICU stays involving possible combinations of ototoxicity, ECMO, prolonged ventilation, and hyperbilirubinemia; and more than one statistically common cause presenting in the same child, for example, congenital CMV and mutations in the GJB2 (Connexin 26) gene. Similarly, “genetic” may be a more accurate way to describe “hereditary” hearing loss, as many families think hearing loss can be genetic only if it presents in a dominant pattern, while most genetic congenital and early onset hearing loss is recessive (so the parents do not usually manifest the hearing loss). Other common inheritance patterns that may result in hearing loss include X-linked recessive (the mothers/sisters who are carriers do not usually have the hearing loss), X-linked dominant (mother would be symptomatic), mitochondrial (which is highly variable so the mother may be asymptomatic), or multifactorial (gene and environment interaction). Similarly, some genetic causes of hearing loss may be due to denovo mutations (arising in germ cells or in the embryo itself) so that neither parent would be expected to have that same mutation. See Chapter xxxx for an in-depth discussion of genetic causes of hearing loss.


Infectious Causes

Infectious causes of hearing loss can occur both before and after birth. Historically, the TORCHES (Toxoplasmosis, Other, Rubella, Cytomegalovirus, Herpes virus, Syphilis) organisms are described as common causes of congenital hearing loss due to prenatal exposure. However, as the epidemiology of these organisms has changed, only one, CMV, is currently a substantial cause of congenital hearing loss in many countries.


Toxoplasmosis

Toxoplasmosis is caused by the protozoan parasite Toxoplasma gondii and, if acquired prenatally, can cause SNHL. Toxoplasmosis is acquired by exposure to undercooked meat, cat feces, soil, or water containing the parasite. The overall prevalence and incidence of congenital toxoplasmosis varies by region and country. In much of the United States, toxoplasmosis is not considered a reportable disease, making it difficult to get accurate incidence figures. Reporting is seriously hampered by the fact that primary infection in most adults (and therefore pregnant mothers) is asymptomatic. In 2000, the Centers for Disease Control and Prevention (CDC) reported that there were an estimated 400 to 4,000 cases of congenital toxoplasmosis in the United States, based on the seroconversion rate of pregnant mothers (21). Massachusetts began screening newborns for toxoplasmosis in 1986, and in 1994 a study from the New England Regional Newborn Screening Program found the incidence of toxoplasmosis in newborns to be 1/10,000 (22). From 1986 to 1998, a total of 99 cases was detected through this program in Massachusetts, but at least 6 cases eventually diagnosed cases were missed by the screening. The incidence of congenital toxoplasmosis varies outside the United States, estimated at 6 per 1,000 births in France, 2 per 1,000 births in Poland, 7 to 10 per 1,000 births in Colombia, and 3 per 1,000 births in Slovenia (23).The NHANES from 1999 to 2000 reported that of the 2,221 women aged 18 to 49 years tested for IgG antibodies to Toxoplasma 14.9% were positive, meaning that the remainder of the women were susceptible to primary infection. Mothers who become primarily infected with T. gondii have approximately a 30% to 50% chance of transmitting the infection to their fetus. Although the most severe disease is experienced by fetuses becoming infected during the first trimester, transmission of infection, which is transplacental, is much more common later in pregnancy, especially during the third trimester, or during delivery. Eighty-five percent of infants with congenital toxoplasmosis are asymptomatic at birth, with those infected later in pregnancy most likely to be asymptomatic. Although vertical transmission of infection is most likely to occur with primary maternal infection, immunocompromised women with chronic infection may also pass on the disease. If new infection is detected in the mother early in pregnancy, treatment with oral spiramycin, which does not cross the placenta, can prevent transmission to the fetus. If infection occurs later in pregnancy, and/or if the fetus is also infected, treatment with pyrimethamine, sulfadiazine, and folinic acid (pyrimethamine is a folic acid antagonist) may prevent transmission to the infant by treating the mother and may also treat the fetus (24).

The classic symptoms of congenital toxoplasmosis are chorioretinitis, hydrocephalus, and intracranial calcifications. Other symptoms, which can also be seen in other congenital infections including CMV, herpes, and rubella, may include hepatosplenomegaly, jaundice, anemia, microcephaly, and lymphadenopathy. Although a recent systematic review by Brown et al. (25) did not find any cases of hearing loss in infants with congenital toxoplasmosis, others have reported an incidence from 0% to 20%. Earlier studies with a higher incidence of SNHL
often included untreated infants, while most infants in recent studies had been treated, suggesting that adequate treatment decreases the occurrence of SNHL. Treatment of congenitally infected infants with pyrimethamine and sulfadiazine is generally effective, although infants who present without neurologic disease have somewhat better outcomes than those who present with neurologic impairment. In addition, babies asymptomatic at birth may develop signs later in life, including chorioretinitis (25).


Rubella

Rubella is caused by a togavirus of the genus Rubivirus. Until the introduction of the rubella vaccine, rubella was the most common viral cause of congenital SNHL. The rate of congenital rubella syndrome (CRS) for infants born to women infected during their first 11 weeks of pregnancy is 90%; if infected during the first 20 weeks of pregnancy, it is 20%. The most common features of CRS are cataracts, heart defects, and hearing impairment. During the 1962 to 1965 global rubella pandemic, there were an estimated 12.5 million rubella cases in the United States, and 20,000 infants developed CRS. Following the licensure of live attenuated rubella vaccines in the United States in 1969, the number of reported cases of CRS declined 99%, from 77 cases in 1970 to one imported case in 2004. Although rubella is no longer endemic in the United States, it continues to be endemic in many other countries, where vaccination rates may also not be uniformly high in postpubertal females. Therefore, babies with congenital hearing loss born outside the United States in areas where rubella remains endemic, or born to mothers who may not have been vaccinated, may be more likely to have CRS as a cause of their hearing loss. According to the World Health Organization, the number of countries that have incorporated rubella vaccine into their immunization programs increased from 83 countries (13% of the birth cohort) in 1996 to 127 countries in 2008. In addition, the development of CRS is a theoretical risk in infants of mothers vaccinated during pregnancy, although this has not been reported in the literature. Any infant with CRS can shed the virus for up to 1 year after birth, so that all unvaccinated female caretakers should be aware that they may be at increased risk to become infected (26, 27) (http://www.who.int/immunization/newsroom/Global_Immunization_Data.pdf).


Cytomegalovirus

Prenatal exposure to CMV, a beta-herpes virus, is the most common congenital viral infection and currently the most common viral cause of congenital SNHL. A 2008 study in Dallas, Texas, by Stehel et al. documented a 6% incidence of congenital CMV in babies with confirmed hearing loss. The prevalence of congenital CMV infection is approximately 0.4% to 2.3% of all newborns. If the mother has a primary CMV infection during pregnancy, there is a 40% chance that the infant will become infected, and if she has a reactivation infection, there is about a 2% chance. The chance of transmission of primary infection from mother to child is 25% in the first trimester, 50% in the second, and 75% in the third. However, infants exposed to the virus during the first trimester are the most likely to develop serious neurologic sequelae. Although 90% of infants with congenital CMV are asymptomatic at birth, 10% to 15% will develop hearing loss, while 65% to 70% of infants with symptomatic CMV infection will develop hearing loss. In the study by Stehel et al. (28), 75% of the infants with congenital CMV were only diagnosed because they had hearing loss. However, since CMV hearing loss may be present at birth or it may not develop until later in infancy/childhood, the diagnosis can easily be missed (29). Infants with symptomatic CMV may present at birth with microcephaly, intrauterine growth restriction, developmental delay, hepatosplenomegaly, chorioretinitis, jaundice, petechiae, thrombocytopenia, hyperbilirubinemia, anemia, and hearing loss. The diagnosis of congenital CMV is made in the first few weeks of life by shell vial culture using saliva or urine. Recently, investigators have evaluated CMV polymerase chain reaction (PCR) using dried blood spots, liquid and dried saliva, and dried umbilical cord (30). Although PCR from any of these sources has not been found to be as sensitive as shell vial culture in the newborn period, positive PCR using dried blood spots can confirm a diagnosis of congenital CMV outside the newborn period when shell vial or CMV IgG cannot differentiate between congenital and postnatally acquired infection (31). In addition, PCR technology may be a basis for large-scale screening. Currently, treatment for the hearing loss involves at least 6 weeks of ganciclovir, or, more recently, valganciclovir, a pro drug of ganciclovir (32, 33). The optimal duration of therapy remains unclear, with some evidence supporting a longer course of therapy (up to 6 to 12 months). Although several studies support improved hearing with either of these drug regimens, the numbers of treated patients remain small, the follow-up is variable, and progression of the hearing loss may occur after the drugs are stopped. Side effects of ganciclovir and valganciclovir include bone marrow suppression (neutropenia, thrombocytopenia, anemia) and kidney and liver toxicity. Teratogenesis, mutagenesis, inhibition of spermatogenesis, and impaired fertility have been reported, primarily in animal studies, as a result of exposure to these agents. In addition, treatment of older children with progressive hearing loss due to congenital CMV has not been studied.

Once the baby is out of the newborn period, making a definite diagnosis of congenital CMV is difficult. It can be suspected, however, based on findings of developmental delay, microcephaly, and hearing loss. Visual impairment may be present in 10% to 20% of symptomatic babies and includes strabismus, optic atrophy, macular scarring, pigmentary retinitis, and visual loss. Boppana et al. (34) has reported late onset or reactivation chorioretinitis in 7/31 (23%) of children after 1 to 10 years of age. Imaging findings can help narrow the diagnosis in older children, or
confirm the diagnosis in newborns. Intracranial calcifications, migrational abnormalities (e.g., polymicrogyria), cerebral and cerebellar volume loss, ventriculomegaly, and white matter disease are commonly seen on a combination of ultrasound, magnetic resonance imaging (MRI), and computed tomography (CT). Characteristic imaging findings in association with neurodevelopmental delay strongly suggests a diagnosis of congenital CMV.

The prevention of congenital CMV is a major public health issue. CMV infection in healthy adults and children is common, often silent, and chronic, with viral shedding in urine and saliva continuing for months or years. This makes avoidance of infected individuals by pregnant women, although desirable, almost impossible. The timing of pregnancy related to when the mother developed a primary CMV infection has been found to affect whether subsequent infants develop CMV. Fowler et al. (35) found that for mothers who were initially CMV nonimmune or immune, a shorter interval between the birth at which the mother’s serologic status was established and the subsequent birth was associated with an increased risk of congenital CMV. It was also clear from this study that maternal CMV infection that occurs before getting pregnant still results in a substantial risk of congenital CMV.

However, because preconception maternal CMV immunity may confer a significant degree of protection against vertical transmission of the virus compared to primary infection during pregnancy, vaccination of young women before they get pregnant is an obvious route to protection (35). Although the development of an effective CMV vaccine has been difficult, Schleiss (36) has recently reported interim results of a phase II trial of a deoxyribonucleic acid (DNA) vaccine against CMV in a guinea pig model. Further population-based studies of transmission rates, severity of disease related to when the mother and fetus become infected, and vaccine development will be key to decreasing the overall prevalence and transmission of CMV in the population.


Herpes Simplex

Herpes simplex virus types 1 and 2 (HSV-1 and HSV-2) are double-stranded DNA viruses that commonly infect humans (37). Neonatal HSV infections occur in approximately 1/3,200 births in the United States, with the majority being HSV-2 (37). Kimberlin (38) reported that 85% of transmission is peripartum, 10% is postnatal, and 5% is in utero, with primary maternal infection resulting in a higher transmission rate than reactivation disease. Approximately 50% of neonatal HSV infections have central nervous system (CNS) involvement. The incidence of hearing loss in children with neonatal HSV is unclear. A recent systematic review of the incidence of SNHL in neonates exposed to HSV was published in 2008 (39). They found that SNHL occurs rarely in infants with exposure to HSV and that there was no evidence of delayed onset of SNHL in infants with asymptomatic perinatal HSV infection. The few papers that reported audiometric data found hearing loss in association with intrauterine HSV-2 infection 0% to 33% of the time, with all infants also having severe CNS disease. However, all studies were small and audiometric techniques varied. The standard treatment for symptomatic perinatal HSV infection is acyclovir, but the effect of treatment on hearing loss is unclear. Based on the limited information available, Westerberg et al. (39) did not find that routine serologic screening for HSV infection in otherwise asymptomatic neonates with SNHL was justified.


Syphilis

Transplacental mother-to-child transmission of syphilis has been recognized since the 15th century. Congenital syphilis (CS) and syphilis acquired after birth are caused by Treponema pallidum, a gram-negative spirochete bacterium. Worldwide, 2 million mothers test positive for syphilis during pregnancy, comprising 1.5% of all pregnancies, and can lead to stillbirths and neonatal death. In the United States, however, CS is uncommon, with the result that hearing loss due to CS may be missed. A systematic review of CS cases and articles contained in multiple databases, including Medline (1950 to March 2008), EMBASE (1980 to March 2008), CINAHL (1982 to March 2008), BIOSIS Previews (1969 to March 2008), and Cochrane databases, was published in 2009 (40). These authors found no reported cases of SNHL in infants with CS born to mothers with syphilis acquired during pregnancy in any of the articles they reviewed. Between 2003 and 2005, the number of CS cases reported annually in the United States declined from 10.6 cases per 100,000 live births in 2003 to 8.2 in 2005, but then increased back to 10.1 per 100,000 live births in 2008. This increase in CS followed a 38% increase in the primary and secondary syphilis rate among US females from 2004 to 2007. In 2008, infants born to black mothers accounted for 50% of CS cases, infants of Hispanic mothers for 31% of cases, and infants of white, Asian/Pacific Islander, and American Indian/Alaskan Native mothers for 15%, 2%, and 1%, respectively. Despite this recent increase, however, CS remains uncommon in the United States, and therefore CS and any associated hearing loss may not be as readily recognized as in countries where it is more common. Hutchinson described the classic triad of late CS as interstitial keratitis, notched incisors, and SNHL (41). In 1970, Fiumara and Lessell (42) added to this triad Clutton joint (symmetrical hydrarthrosis especially of the knees) and mulberry molars (malformed first molar resembling a mulberry). Although any infant who presents with nonimmune hydrops, osteochondritis, periosteitis, hepatosplenomegaly, jaundice, rhinitis, mucocutaneous skin rash, and/or pseudoparalysis of an extremity should be suspected of having CS, there are no studies looking at the array of symptoms in infants who are actually serologically positive, raising the question whether hearing loss could be the only symptom of CS. Prenatal diagnosis involves testing pregnant mothers, with nearly complete elimination
of any serious complications in the baby if the mother is treated before 24 weeks gestation. For babies with CS identified after birth, appropriate treatment with penicillin (dose and duration dependent on formulation) usually results in a cure. If there is uncertainty about whether the mother has been adequately treated, evaluation of both the mother and infant should be undertaken shortly after birth. Syphilis acquired after birth may present as progressive and/or sudden hearing loss or vertigo. Therefore, the diagnosis of acquired syphilis should be considered in any sexually active patient who presents with recent onset hearing loss and/or vertigo, as these symptoms often improve with penicillin and steroids.


Other Infectious Diseases Associated with Hearing Loss


HIV

Audiologic and vestibular symptoms can occur in children infected with human immunodeficiency virus type 1. These include conductive hearing loss (often due to otitis media), SNHL, tinnitus, vertigo, and ataxia. Hearing loss may occur due to direct infection of the inner ear with human immunodeficiency virus (HIV), or secondary to opportunistic infections (e.g., CMV, syphilis, tuberculosis, cryptococcal meningitis) or ototoxic drug therapy. In a 2008 study, 23 Mexican children with HIV on highly active antiretroviral therapy aged 5 months to 16 years underwent pure tone audiometry, speech discrimination testing, ABR testing, electronystagmography, and rotary chair testing. Most children had acquired their HIV perinatally. Although not every child was able to be tested using all techniques, six children had conductive hearing loss (related to a history of otitis media), one had SNHL, one had a mixed hearing loss, others had abnormalities on ABR reflecting dysfunction at varying levels, and several had vestibular hyporeflexia. Although not statistically significant, there was a suggestion that lower CD4 counts, longer time with HIV, and higher viral load were associated with an increased prevalence of audiologic or vestibular symptoms (43).


Measles (Rubeola)

Measles is a paramyxovirus spread via the respiratory route and, in the prevaccine era, was an identified cause of acquired hearing loss, although the incidence was not well defined. Although measles was reported to have been eradicated in the United States in 2000, hundreds of cases have been reported since then, occurring mainly in children born outside the United States, children too young to vaccinate, and in those children whose parents have intentionally not vaccinated them. Live attenuated measles virus is included in the standard measles vaccine. A paper using information from the Vaccine Adverse Event Reporting System in the United States from 1990 to 2003 estimated the reporting rate of SNHL possibly related to mumps/measles vaccine to be one case in 6 to 8 million doses (44). Measles virus has also been possibly implicated in the development of otosclerosis, although whether this is an association or causation remains unclear (45, 46).


Lyme Disease

Lyme disease is caused by the bacterium Borrelia burgdorferi and is transmitted to humans by the bite of an infected black-legged tick. The incidence is highest in the United States in east coast states and Wisconsin and Minnesota but has been reported in all 50 states. Although people of any age and both genders are susceptible to tick bites, Lyme disease is most common among boys aged 5 to 19 and persons aged 30 or older. The CDC reported an 8% incidence of facial palsy among 119,965 patients with Lyme disease between 1992 and 2004. Hearing loss in association with Lyme disease, however, is infrequently reported (47). A 2010 retrospective review by Wilson et al. of patients 18 years of age and older presenting with asymmetric SNHL of uncertain etiology showed that of the 88 who were tested for Lyme disease, 3 were positive. Although the endemic nature of Lyme disease in many parts of the country can make confirming a relationship between Lyme disease and hearing loss challenging, Wilson et al. (48) felt that of the many tests ordered in this study, Lyme titers and tests for syphilis were potentially the most useful and cost-effective.

May 24, 2016 | Posted by in OTOLARYNGOLOGY | Comments Off on Nongenetic Hearing Loss

Full access? Get Clinical Tree

Get Clinical Tree app for offline access