The Prevalence of Congenital Hearing Loss in Neonates withDown Syndrome

Adrienne S. Tedeschi, MD1, Nancy J. Roizen, MD2, H. Gerry Taylor, PhD2, Gail Murray, PhD3, Christine A. Curtis, PhD4,

and Aditi Shah Parikh, MD5

Objective To determine the prevalence of hearing loss in newborns with Down syndrome.Study designWe performed a cross-sectional, retrospective chart review of all infants with Down syndrome bornat a university-affiliated hospital (n = 77) or transferred in to the associated pediatric hospital (n = 32) following birthat an outlying hospital between 1995 and 2010. We determined the rate of failure of newborn hearing screens, theproportion of infants lost to follow-up, and the rate of confirmed hearing loss, as well as the associations of risk fac-tors for hearing loss with confirmed hearing loss.Results Of the 109 patients with hearing screening data, 28 failed their newborn hearing screen. Twenty-seveninfants were referred for audiologic evaluation, and 19 completed the evaluation. Fifteen of these 19 infants(79%) had confirmed hearing loss. The prevalence of congenital hearing loss in this sample of neonates withDown syndromewas 15%. Exposure tomechanical ventilation was the sole known risk factor associated with hear-ing loss. In this study, the loss to follow-up rate for infants with positive hearing screens was 32%.Conclusion Newborns with Down syndrome have a higher prevalence of congenital hearing loss compared withthe total neonatal population (15% vs 0.25%). Continuedmonitoring of hearing is needed in children with Down syn-drome. (J Pediatr 2015;166:168-71).

Down syndrome is one of the most common chromosomal abnormalities in the US, affecting 1 in 691 births and re-sulting in approximately 6037 infants born with Down syndrome each year.1 Previous studies have reported a prev-alence of hearing loss among children and adolescents with Down syndrome ranging between 45% and 90%,2-10

compared with 2.5% in the general pediatric population.11 The variability in prevalence data is likely secondary to theuse of small convenience samples, variable hearing testing modalities, and a lack of standardized criterion defining“abnormal” hearing.

Individuals with Down syndrome are prone to otolaryngologic anomalies that complicate the diagnosis and classificationof hearing impairment. Evaluating the tympanic membrane in young children with Down syndrome is challenging second-ary to stenosis of the external auditory canal and increased rates of cerumen impaction.9 Middle ear abnormalities, includingeustachian tube dysfunction, remnant tissue in the middle ear cavity, and ossicular malformations, are frequently seen andfurther contribute to the higher rate of conductive hearing loss reported in this population.4,8,9 Inner ear hypoplasia anddysplasia of the cochlea, cochlear nerve canal, lateral semicircular canal, and cochleovestibular nerve are hypothesized toexplain the frequent and progressive nature of sensorineural hearing loss seen in individuals with Down syndrome.12 Studiesof adolescents and adults with Down syndrome have demonstrated an increasing prevalence of sensorineural hearing losswith age.3

Although numerous studies have confirmed the increased incidence of hearing loss in individuals with Down syndrome, theprevalence of congenital hearing loss had not been studied previously. Hearing loss in infants has been associated with lifelongdeficits in speech and language acquisition, poor academic performance, personal-social maladjustments, and emotional dif-ficulties.13-15 Early identification of infants with hearing loss may help mitigate the adverse consequences associated with hear-ing loss.16 The institution of universal newborn hearing screening provides a new opportunity to examine the prevalence ofcongenital vs acquired hearing loss in the population of children with Down syndrome.

From the 1Department of Neurology, Golisano Children’sHospital of Southwest Florida, Fort Myers, FL; and2Division of Developmental-Behavioral Pediatrics andPsychology, Department of Pediatrics; and Departmentsof 3Otolaryngology Head & Neck Surgery, 4Pathology,and 5Genetics and Genome Sciences, Case Western

The aim of the present study was to collect and analyze the data obtained fromuniversal newborn hearing screening and subsequent audiologic evaluations toestablish the prevalence of congenital hearing loss in neonates with Down syn-drome. We also examined the referral rate for follow-up of abnormal hearingscreening, the proportion of infants lost to follow-up, and the associations ofrisk factors for hearing loss with confirmed hearing loss.

Reserve University School of Medicine, Cleveland, OH

Supported by Health Resources and Services Adminis-tration (T77MC00004B0). The authors declare no conflictof interest.

0022-3476/$ - see front matter. Copyright ª 2015 Elsevier Inc.

All rights reserved.

http://dx.doi.org/10.1016/j.jpeds.2014.09.005

AABR Automated auditory brainstem response

NICU Neonatal intensive care unit

168

Vol. 166, No. 1 � January 2015

Methods

Following Institutional Review Board approval, a retrospec-tive chart review was performed of all infants with Down syn-drome born at University Hospitals in Cleveland ortransferred in following birth at an outlying hospital betweenMarch 8, 1995, and December 17, 2010. Deidentified data,including newborn hearing screening results, risk factorsassociated with hearing loss, referral to audiology, and resultsof subsequent audiologic evaluations, were collected.

The hearing loss risk factors assessed included neonatalintensive care unit (NICU) admission for $5 days, maternalprenatal illness (ie, documented infections associated withhearing loss to include cytomegalovirus, herpes, syphilis,rubella, and toxoplasmosis), family history of hereditarychildhood hearing loss, low Apgar scores (0-4 at 1 minuteor 0-6 at 5 minutes), birth weight <1500 g, presence of a syn-drome associated with hearing impairment, indirect bilirubin>20 mg/dL, exchange transfusion, meningitis, craniofacialanomalies, exposure to ototoxic drugs, and mechanical venti-lation. These risk factors were initially selected from the 1994Position Statement of the Joint Committee on Infant Hearingand modified after subsequent revisions.

The screening protocol and tools did not change duringthe 15-year study period. All infants born at or admitted toUniversity Hospitals in the neonatal period underwent uni-versal newborn hearing screening using automated auditorybrainstem response (AABR) technology. If an infant receiveda nonpass result (referred) during the initial screen, then asecond AABR was attempted before discharge. Infants whodid not pass the hearing screening after 2 AABR screenswere referred to audiology for further evaluation, whichincluded auditory brainstem response testing, distortionproduct otoacoustic emissions testing, tympanography, andbone and air conduction studies. Infants identified as atincreased risk for hearing loss (based on the risk factors listedabove) underwent an audiologic evaluation before discharge.Lower-risk infants were scheduled for outpatient audiologicevaluation on discharge to home.

Association of known risk factors for hearing loss with thepresence of confirmed hearing loss were analyzed using theFisher exact test. P# .05 was considered to indicate statisticalsignificance.

Table. Associations between risk factors for hearing lossand presence of hearing loss

Risk factorsFisher exact

testInfants with riskfactor, n (%)

NICU stay >5 days 0.159 77 (70.6)Maternal illness 1.000 2 (1.8)Family history of hearing loss 1.000 1 (0.9)Low Apgar score 0.279 17 (15.6)Low birth weight 0.097 8 (7.3)Bilirubin >20 mg/dL 1.000 1 (0.9)Meningitis 0.150 1 (0.9)Defects of face/head 1.000 5 (4.6)Ototoxic drugs (prenatal and postnatal) 0.101 53 (48)Mechanical ventilation 0.008 19 (17.4)

Results

Of the 131 infants who met the inclusion criteria, 126 hadcharts available for review. The remaining 5 infants’ chartscould not be located by the medical records department.Six infants died before screening, and 11 infants did nothave a documented hearing screen, resulting in a study popu-lation of 109 infants (57 males). Of the 109 infants included inthis study, 77 were born at University Hospitals and 32 weretransferred in from outlying community hospitals after birth.

Twenty-eight infants (25.7%) failed the newborn hearingscreen (Figure; available at www.jpeds.com). Twenty-seven

infants were referred for follow-up audiologic evaluation,and 19 infants completed the evaluation. Fifteen of the19 infants with audiologic data (78.9%) had confirmedcongenital hearing loss, including 4 with unilateral hearingloss and 11 with bilateral hearing loss. The screening processonly determined a pass/fail result, and the diagnosticevaluations did not consistently distinguish betweenconductive and sensorineural hearing loss.In this population of newborns with Down syndrome, the

prevalence of congenital hearing loss was determined to be15%, based on 15 infants with confirmed hearing loss in apopulation of 100 (after having subtracted the 9 infants lostto follow-up). Given the 32% loss to follow-up rate andour concern that excluding these infants could have underes-timated or overestimated the true prevalence of hearing loss,we calculated an estimated prevalence. Based on the high de-gree of agreement between failed AABR screening andconfirmed hearing loss demonstrated on audiologic testing(78.9%), we estimated that an additional 7 infants withhearing loss would be identified in the lost to follow-upgroup (0.79 � 9 infants lost to follow-up). We determinedan estimated prevalence of hearing loss in newborns withDown syndrome of 20% after adding the estimated 7 addi-tional cases to the known 15 cases, resulting in 22 total casesof hearing loss in a population of 109 infants. If all of thechildren lost to follow-up had normal hearing, then theprevalence of congenital hearing loss would be 13.8%; ifall had hearing loss, then the prevalence of hearing losswould be 22%.Associations between known risk factors for hearing loss

and presence of hearing loss were examined using the Fisherexact test, owing to the low frequency of some risk factors inthe study population (Table). There were only 2 infants withdocumented maternal illness during pregnancy and 1 infanteach with family history of hearing loss, bilirubin level>20 mg/dL, and meningitis. No infants received exchangetransfusions for an elevated bilirubin level, and thus we didnot include this risk factor in our calculations.We compared the presence of risk factors for hearing loss

in infants with confirmed hearing loss and infants who passednewborn hearing screening or had normal audiologic evalu-ations. The sole risk factor identified as positively correlatedwith the presence of hearing loss was mechanical ventilation

169

THE JOURNAL OF PEDIATRICS � www.jpeds.com Vol. 166, No. 1

(P= .008). The low incidence of some hearing loss risk factorsadversely affected our analysis of hearing loss risk factors.Our inclusion criteria identified 2 populations of infantswith Down syndrome: infants born at our institution and in-fants transferred in from outlying community hospitals. Weincluded both hospitals with the goal of increasing the sizeand power of our study population.

Each sample was examined independently to assess for dif-ferences in the occurrence of hearing loss risk factors and therates of confirmed hearing loss. There was no statistical dif-ference in the prevalence of hearing loss between transferredinfants (12.5%) and inborn infants (14%). Both were high-risk groups, because all children with Down syndrome whoare transferred in are routinely admitted to the NICU. Theobstetric service is a level 3 high-risk service, and approxi-mately 35% of neonates go to the NICU for care. This studydid not include an evaluation of which group might be athigher risk.

There were differences in the rates of hearing loss riskfactors between the 2 populations for both NICU admission(96.9% of infants transferred in vs 71.4% of inborn infants;P = .003) and exposure to ototoxic medications (81.2%transferred in vs 41.2% inborn; P = .001). This is a reflectionof the direct transfer to the NICU of those infants born atoutlying hospitals. A larger proportion of these infantsreceived 48-hour aminoglycoside therapy as part of arule-out sepsis protocol. No statistical differences were iden-tified in the incidence of other hearing loss risk factors be-tween the 2 groups, including mechanical ventilation, lowbirth weight, hyperbilirubinemia, meningitis, and maternalinfections.

Discussion

This study establishes a prevalence of congenital hearingloss of 15% in this group of high risk newborns withDown syndrome. This is considered a conservative esti-mate, because universal newborn hearing screening at ourinstitution via AABR is designed to detect hearing loss ata threshold of 35 dB, which may miss infants with mildand/or progressive hearing impairment. Although there isa rich body of evidence demonstrating that the majorityof individuals with Down syndrome will develop hearingloss within their lifetime, to our knowledge no other studieshave investigated and established the rate of congenitalhearing loss at birth.

Historically, studies of hearing loss in this population havebeen based on small convenience samples that included chil-dren and often adults, used different hearing testing modal-ities, and reported rates of hearing loss calculated usingvariable criterion. In addition, there are many otolaryngo-logic comorbidities that make diagnosing and managingchronic middle ear problems challenging. Also confoundingthe picture are the accelerated effects of aging on hearing. In-dividuals with Down syndrome have been shown to haveearly-onset presbycusis8 and an increasing prevalence ofsensorineural hearing loss with age.3

170

With the advent of universal newborn hearing screening,further understanding of the frequency of hearing loss inthe population with Down syndrome is possible. Furtherprospective studies are needed to confirm the prevalenceof congenital hearing loss in newborns with Down syn-drome and to follow the course of their hearing loss,speech development, and social development in conjunc-tion with the appropriate management of their underlyinghearing deficits.The strengths of this study include a large cross-sectional

population of infants with Down syndrome that receivedstandardized hearing screening and audiologic evaluationthroughout a 15-year period at a single institution. Limita-tions of the study are its retrospective design, lack of powerfor the evaluation of hearing loss risk factors, and a loss tofollow-up rate of 32%. Although our rate of patients lostto follow-up is elevated, it is lower than the national lossto follow-up rate of 43.7%-64.0% reported annually bythe Centers for Disease Control and Prevention from 1999to 2007.17

Shulman et al18 surveyed universal newborn hearingscreening and intervention programs nationwide and re-ported that only 52% of infants who needed a diagnosticevaluation after failed newborn hearing screening receivedone by age 3 months. Lack of provider knowledge, data sys-tems inaccessible to providers, and a “wait and see” attitudeby pediatricians were identified as barriers to the follow-up offailed screenings.According to the Healthcare Supervision for Children with

Down Syndrome Guidelines19 and the Joint Committee onInfant Hearing,20 follow-up audiologic assessments shouldbe completed by age 3 months, followed by appropriateintervention and monitoring if a hearing loss is identified.If normal hearing is established, then the guidelines recom-mend an audiogram and tympanometry every 6 monthsbetween age 1 year and 5 years and then annually. Theserecommendations for continued monitoring despite docu-mented normal hearing stem from studies identifyingincreased frequency of fluctuating and progressive hearingloss, with 45%-90% of children and adolescents with Downsyndrome having hearing loss in cross-sectional studies.In conclusion, the present study establishes the prevalence

of congenital hearing loss in this group of newborns withDown syndrome and identifies areas of further prospectivestudy. Normal hearing plays an important and critical rolein the development of communication skills, as well as inbehavioral, social, and cognitive development. This studyserves as a reminder that all infants with Down syndrome,even those who pass the newborn hearing screen, requireobjective newborn hearing screening and continued audio-logic follow-up. The primary care physician has the impor-tant and critical role of ensuring prompt, timely, andaccurate initial and ongoing evaluation of hearing in thisat-risk population. In addition, the elevated loss to follow-up rate in our at-risk population compared with the highernational loss to follow-up rate among all newborns under-scores the importance of the medical home and the need

Tedeschi et al

January 2015 ORIGINAL ARTICLES

for proactive follow-up of all newborns with failed hearingscreenings. n

Submitted for publication May 2, 2014; last revision received Aug 5, 2014;

accepted Sep 4, 2014.

References

1. Presson AP, Partyka G, Jensen KM, Devine OJ, Rasmussen SA,

McCage LL, et al. Current estimate of Down syndrome population prev-

alence in the United States. J Pediatr 2013;163:1163-8.

2. Kwong KL, Wong V. Neurodevelopmental profile of Down syndrome in

Chinese children. J Paediatr Child Health 1996;32:153-7.

3. McPherson B, Lai SP, Leung KK, Ng IH. Hearing loss in Chinese school

children with Down syndrome. Int J Pediatr Otorhinolaryngol 2007;71:

1905-15.

4. SchwartzDM, Schwartz RH.Acoustic impedance and otoscopic findings in

young children with Down syndrome. Arch Otolaryngol 1978;104:652-6.

5. Downs MP. The hearing of Down’s individuals. Semin Speech Lang

Hear 1980;1:25-38.

6. Roizen NJ, Wolters C, Nicol T, Blondis TA. Hearing loss in children with

Down syndrome. J Pediatr 1993;123:S9-12.

7. Dahle AJ, McCollister FP. Hearing and otologic disorders in children

with Down syndrome. Am J Ment Defic 1986;90:636-42.

8. Driscoll C, Kie J, Bates D. Tympanometry and TEOAE testing of children

with Down syndrome in special schools. Aust NZ J Audiol 2003;25:85-93.

9. Balkany TJ. Otologic aspects of Down’s syndrome. Semin Speech Lang

Hear 1980;1:39-48.

10. Balkany TJ, Mischke RE, Downs MP, Jafek BW. Ossicular abnormalities

in Down syndrome. Otolaryngol Head Neck Surg 1979;88:372-84.

The Prevalence of Congenital Hearing Loss in Neonates with Dow

11. Hans PS, England R, Prowse S, Young E, Sheehan PZ. UK and Ireland

experience of cochlear implants in children with Down syndrome. Int

J Pediatr Otorhinolaryngol 2010;74:260-4.

12. Blaser S, Propst EJ, Martin D, FeigenbaumA, James AL, Shannon P, et al.

Inner ear dysplasia is common in children with Down syndrome. Laryn-

goscope 2006;116:2113-9.

13. CunninghamM, Cox EO, Committee on Practice and AmbulatoryMed-

icine and Section on Otolaryngology and Bronchoesophagology. Hear-

ing assessment in infants and children: recommendations beyond

neonatal screening. Pediatrics 2003;111:436-40.

14. Stoel-Gammon C. Phonological development in Down syndrome.

MRDD Res Rev 1997;3:300-6.

15. Yoshinaga-Itano C, Sedey AL, Coulter DK, Mehl AL. Language of early

and later-identified children with hearing loss. Pediatrics 1998;102:

1161-71.

16. Korver AM, Konings S, Dekker FW, Beers M, Wever CC, Frijnstt J, et al.

Newborn hearing screening vs later hearing screening and develop-

mental outcomes in childhood with permanent childhood hearing

impairment. JAMA 2010;304:1701-8.

17. Centers for Disease Control and Prevention. Identifying infants with

hearing loss—United States, 1997-2007. MMWR Morb Mortal Wkly

Rep 2010;59:220-3.

18. Shulman S, Besculides M, Saltzman A, Ireys H, White KR, Forsman I.

Evaluation of the Universal Newborn Hearing Screening and Interven-

tion Program. Pediatrics 2010;126(Suppl 1):S19-27.

19. Bull MJ, Committee on Genetics. Health supervision for children with

Down syndrome. Pediatrics 2011;128:393-406.

20. American Academy of Pediatrics, Joint Committee on Infant Hearing.

Year 2007 position statement: principles and guidelines for early

hearing detection and intervention programs. Pediatrics 2007;120:

898-921.

n Syndrome 171

Figure. Failed hearing screening, lost to follow-up, andfollow-up audiologic evaluation results.

THE JOURNAL OF PEDIATRICS � www.jpeds.com Vol. 166, No. 1

171.e1 Tedeschi et al