You are on page 1of 10


Etiologic and Audiologic Evaluations After Universal

Neonatal Hearing Screening: Analysis of 170
Referred Neonates
Frank Declau, MD, PhDa, An Boudewyns, MD, PhDa, Jenneke Van den Ende, MDb, Anouk Peeters, MDa, Paul van den Heyning, MD, PhDa

Departments of aOtorhinolaryngology, Head and Neck Surgery, and Communication Disorders and bMedical Genetics, University of Antwerp, Anwerp, Belgium

The authors have indicated they have no financial relationships relevant to this article to disclose.

What’s Known on This Subject What This Study Adds

Hearing loss is one of the most common congenital anomalies. Hearing impairment in To the best of our knowledge, this is the first report in the literature that provides data on
infants can negatively affect speech and language acquisition, academic achievement, both audiologic confirmation and etiologic assessment in a large sample of newborns
and social and emotional development. These negative effects can be diminished referred after failed screening.
through early intervention.

OBJECTIVE. The goal was to clarify the audiologic aspects and causes of congenital
hearing loss in children who failed universal neonatal hearing screening.
METHODS. A prospective analysis of 170 consecutive records of neonates referred to a peds.2007-1479
tertiary center after universal neonatal hearing screening failure, between 1998 and doi:10.1542/peds.2007-1479
2006, was performed. The data presented here represent the equivalent of ⬃87 000 Key Words
screened newborns. The screening results were validated with a clinical ear, nose, congenital sensorineural hearing loss,
and throat examination and electrophysiological testing, including diagnostic audi- cytomegalovirus, genetics, hearing
screening, newborn
tory brainstem response, automated steady state response, and/or behavioral testing.
A diagnostic evaluation protocol for identification of the cause of the hearing loss was Abbreviations
UNHS— universal neonatal hearing
also implemented, in collaboration with the departments of genetics and pediatrics. screening
ABR—auditory brainstem response
RESULTS. Permanent hearing loss was confirmed in 116 children (68.2%). Bilateral AABR—automated auditory brainstem
hearing loss was diagnosed in 68 infants (58.6%) and unilateral hearing loss in 48 response
infants (41.4%). Median thresholds for the neonates with confirmed hearing loss PCR—polymerase chain reaction
CMV— cytomegalovirus
were severe in both unilateral and bilateral cases, at 70 dB nHL and 80 dB nHL, WS—Waardenburg syndrome
respectively. In 55.8% of those cases, no risk factors for hearing loss were found. In Accepted for publication Sep 27, 2007
60.4%, the initial automated auditory brainstem response diagnosis was totally in Address correspondence to Frank Declau, MD,
agreement with the audiologic evaluation results. In 8.3% of the cases, however, a PhD, Department of Otorhinolaryngology,
unilateral refer result was finally classified as bilateral hearing loss. An etiologic factor Head and Neck Surgery, and Communication
Disorders, University Hospital Antwerp,
could be identified in 55.2% of the cases. Of the causes identified, a genetic mech- Wilrijkstraat 10, B-2650 Edegem, Belgium.
anism was present in 60.4% of the cases, peripartal problems in 20.8%, and con- E-mail:
genital cytomegalovirus infection in 18.8%. PEDIATRICS (ISSN Numbers: Print, 0031-4005;
Online, 1098-4275). Copyright © 2008 by the
CONCLUSIONS. An etiologic factor could be identified for nearly one half of the children American Academy of Pediatrics
with confirmed congenital hearing loss referred through a universal hearing screen-
ing program. Pediatrics 2008;121:1119–1126

H EARING LOSS IS one of the most common congenital anomalies, occurring in ⬃1 to 2 infants per 1000.1 The
prevalence of hearing loss has been shown to be greater than that of most other diseases and syndromes
screened for at birth (eg, phenylketonuria and sickle cell disease). The incidence of hearing loss is considerably higher
in infants in the NICU (1–2 cases per 200 infants; 1.9% bilateral and 0.6% unilateral.2 Left undetected, hearing
impairment in infants can negatively affect speech and language acquisition, academic achievement, and social and
emotional development. These negative effects can be diminished and even eliminated through early intervention at
or before 6 months of age.3 Reliable screening tests that minimize referral rates and maximize sensitivity and
specificity are available. The goal of universal neonatal hearing screening (UNHS) is to maximize linguistic and
communicative competence and literacy development for children who are hard of hearing or deaf.
In Flanders, a community-based screening program was successfully implemented by the federal health care
agency in 1998. The hearing screening program is performed with automated auditory brainstem response (AABR)
testing and aims to identify all children with a permanent unilateral or bilateral hearing impairment of ⱖ35 dB nHL.
The AABR testing is performed at the age of ⬃4 weeks. If the infant fails the initial test, then it is repeated within

PEDIATRICS Volume 121, Number 6, June 2008 1119

Downloaded from by guest on February 8, 2017
48 hours. Children who twice the test fail or for whom TABLE 1 Diagnostic Protocols
unreliable test results are obtained are referred for re- Standard diagnostic protocol for children with congenital hearing loss
testing and diagnostic evaluation. The coverage of the General pediatric examination (dysmorphic signs)
screening with AABR testing in Flanders has been Urine sample (viral culture isolation, measurement of electrolyte levels)
⬃96% since its introduction, with a false-positive rate of Neurologic examination
only 0.06% (specificity: 99.94%).4 In many UNHS pro- Ophthalmologic examination
grams, the screening/therapy coordination is the weak- Electrocardiography
est point of the pathway, with lost to follow-up moni- High-resolution computed tomography or MRI
toring (and treatment) rates as high as 2% to 52%.5 Complementary examinations for children with congenital hearing loss
Electroretinography, slit-lamp examination
Renal ultrasonography
Heart ultrasonography, Holter monitoring
In Flanders, a community-based screening program was Radiography of thorax and spine
successfully implemented by the federal health care Karyotyping
agency (Kind & Gezin). The agency coordinates the pre- Perchlorate testing
ventive care and wellness of children in Flanders and is DNA isolation for specific syndromes (eg, pendrin, otoferlin, and connexin 30)
well organized, with 620 nurses working at 330 consul-
tation bureaus in 62 Flemish regions.6 All neonates
(NICU and non-NICU) referred after UNHS between (exon 1 and 2). If a mutation in GJB2 was found, then
March 1998 and April 2006 were included in the present polymerase chain reaction (PCR) was performed to detect
study. Referral was initiated after 2 consecutive failures the most frequent deletion in connexin 30 (GJB6-
of the AABR testing. The pattern of referral to the ter- D13S1830). Also, a toxoplasmosis, rubella, cytomegalovirus
tiary centers by the federal health care agency was based (CMV), herpes, and syphilis serologic screen was per-
on the geographical areas in which the children resided. formed for every child (toxoplasmosis, IgM and IgG; ru-
Most children came from a suburban area around Ant- bella, IgM; CMV, IgG and IgM; herpes simplex, IgM and
werp, Belgium. IgG; Treponema pallidum, rapid plasma reagin and T palli-
The standard method for validation of UNHS results is dum hemagglutination tests). These results were discussed
a combination of ear, nose, and throat and audiologic by the multidisciplinary team and, depending on the type
evaluations performed with electrophysiological testing, of hearing loss, familial history, risk factors, and associated
such as diagnostic auditory brainstem response (ABR), features, complementary examinations were performed
automated steady state response, and/or behavioral test- (Table 1).6
ing. The medical evaluation was performed by an oto- The imaging modalities used consisted of computed
rhinolaryngologist, in close collaboration with a medical tomography and MRI, and imaging was requested for all
geneticist and a pediatrician. The purpose of this evalu- children with confirmed unilateral and/or bilateral hear-
ation was not only to determine the pathogenesis of ing loss of ⱖ60 dB nHL or craniofacial malformations.
hearing loss but also to identify related medical condi- The cutoff value was set according to the report by
tions and to provide recommendations for medical treat- Bamiou et al.8 They found that profound or progressive
ment, as well as referral to other services.6 hearing loss and craniofacial abnormalities were signifi-
The evaluation included a detailed pedigree analysis cant predictors of abnormal computed tomographic find-
for congenital hearing loss, medical history, and risk ings. The radiologic imaging was usually performed at
factors, as identified by the Joint Committee on Infant ⬃6 months of age. After the initial diagnostic evaluation
Hearing 2000 position statement.7 The clinical examina- and confirmation of the hearing loss, children were
tion included tympanoscopy and an examination of the scheduled for early intervention programs (including
head and face to search for outer ear anomalies, preau- hearing aids and rehabilitation) and auditory follow-up
ricular pits or tags, and syndromic features. evaluation was performed with visual reinforcement au-
The audiologic assessment in this study was per- diometry or conditioned play audiometry according to
formed according to guidelines published by the Royal the child’s age and cooperation.
Ear Nose Throat Society.6 The examinations included All data obtained were stored in a database system
click-evoked ABR testing, measurement of auditory (Excel; Microsoft, Redmond, WA); for each child, a com-
steady state responses, and acoustic immittance audiom- plete report was sent to the federal health care agency.
etry with high-frequency tones (678 Hz) and transient Statistical analysis was performed with SPSS 13.0 (SPSS,
evoked otoacoustic emissions. This testing was per- Chicago, IL). The results were summarized as means,
formed by audiologists with technical expertise and medians, and percentages. Subgroups within the study
training for testing in infants, and results were obtained population were compared with Mann-Whitney and
for an ear-specific estimate of the type, degree, and Kruskal-Wallis tests. A P value of ⱕ.05 was accepted as
configuration of the hearing loss. indicating statistically significant results.
The infant then was examined at the department of
pediatrics and ophthalmology, where additional standard
examinations were conducted (Table 1). Mutation analysis
of connexin 26 (GJB2) was performed for every child. For Audiologic Assessment
GJB2, denaturing high performance liquid chromatogra- In Flanders, ⬃120 infants each year are referred to a
phy analysis was performed for the complete coding region tertiary center after failed newborn hearing screening

1120 DECLAU et al
Downloaded from by guest on February 8, 2017
TABLE 3 Results From Diagnostic Audiologic Workup Versus
Screening Results Obtained Through AABR Screening
AABR Finding ABR Result, n (%)

20 No Hearing Loss Unilateral Bilateral

(n ⫽ 44) (n ⫽ 44) (n ⫽ 56)
Unilateral (n ⫽ 87) 36 (41.9) 41 (47.7) 10 (11.6)
Number of cases

Bilateral (n ⫽ 57) 8 (14.0) 3 (5.3) 46 (80.7)


screening population were found to have normal hear-

10 ing. The male/female ratio remained almost constant at
Bilateral hearing loss was diagnosed in 68 infants
(58.6%) and unilateral hearing loss in 48 infants
(41.4%). Bilateral involvement was more frequently
seen in boys. The bilateral/unilateral ratio was 1.54 for
boys and 1.29 for girls. Considering only the infants (n ⫽
0 116) with confirmed hearing loss (⬎35 dB nHL), the
30 40 45 50 55 60 65 70 75 80 90 100110 120 median thresholds were 70 dB nHL in unilateral cases
ABR thresholds and 80 dB nHL in bilateral cases. The distribution of
dB nHL hearing loss in the 184 ears (116 infants) involved is
FIGURE 1 presented in Fig 1. There was no statistical difference
Hearing loss distribution. between the hearing loss levels in the unilateral and
bilateral hearing impairment populations (Mann-Whit-
ney test, P ⫽ .098). However, patients with bilateral
(mean birth rate: 61 673 births per year during the years hearing loss were more apt to have profound hearing
1998 –20059). The data presented here represent the loss than were those with unilateral hearing loss (41.0%
equivalent of ⬃87 000 screened newborns. vs 20.8%) (Table 2).
The referral population after failed screening (n ⫽ The relationship between the initial AABR screening
170) consisted of 91 boys and 79 girls (gender ratio: results from the federal health care agency (n ⫽ 144)
1.15), with a median age at first presentation in the and the final audiologic diagnoses is shown in Table 3. In
referral center of 50 days (range: 36-86 days). A total of 60.4%, the initial AABR screening diagnosis was totally
157 infants (92.4%) were referred as non–NICU infants; in agreement with the audiologic evaluation. However,
148 infants were referred directly from the UNHS pro- 11.6% of neonates with a unilateral refer result revealed
gram, and 9 were sent by referral hospitals for a second a permanent bilateral hearing loss. Children with a uni-
opinion. In addition, 13 infants (7.6%) came from a lateral refer result were more likely to have normal
NICU. Indications for further audiologic evaluation after hearing than were those with a bilateral refer result
AABR screening (n ⫽ 148) were unilateral refer results (41.9% vs 14.0%). For children with bilateral refer re-
(n ⫽ 87), bilateral refer results (n ⫽ 57), and failed tests sults at screening, this diagnosis was confirmed in 80.7%
(n ⫽ 4). The median period between referral by the of cases.
federal health care agency and the first visit at the au- At first presentation at the referral center, risk factors
diologic center was 8 days. A total of 96.5% (n ⫽ 164) of for hearing loss were identified by using the guidelines of
the referral population was of white origin and 3.5% the Joint Committee on Infant Hearing 2000 position
(n ⫽ 6) black. statement.7 Ninety-six children (57.8%) had no risk fac-
Audiologic evaluation after screening confirmed the tors, there were 4 children for whom risk factors were
presence of hearing loss in 121 children. Five children unknown because of unavailable questionnaires, and 70
(2.9%) initially had secretory otitis media, which re- children (42.2%) had 1 (n ⫽ 58) or more (n ⫽ 12) risk
solved during the testing period. Consequently, 116 factors. The relationship between the presence or ab-
(68.2%) of the referred children had permanent hearing sence of risk factors and the hearing status after audio-
loss of ⱖ35 dB nHL; 31.8% of the children from the metric evaluation is presented in Table 4. A total of
65.6% of children without risk factors were found to

TABLE 2 Patients Categorized According to Maximal Hearing Loss

Hearing n TABLE 4 Relationship Between Risk Factors and Hearing Status
Loss Risk Factor ABR, n
40–70 dB 71–95 dB ⬎95 dB
nHL nHL nHL Hearing Loss Normal Hearing
Bilateral 25 15 28 (n ⫽ 113) (n ⫽ 53)
Unilateral 27 11 10 Present (n ⫽ 70) 50 20
Total 52 26 38 Absent (n ⫽ 96) 63 33

PEDIATRICS Volume 121, Number 6, June 2008 1121

Downloaded from by guest on February 8, 2017
Familial deafness
In utero infections
Craniofacial anomalies
No risk factor
Distribution of risk factors according the guidelines of the Joint Weight < 1500 g
Committee on Infant Hearing 2000 position statement. 6.1%

Low Apgar score


Ototoxic medication
Artificial breathing > 5
Deafness syndromes days
3.9% 4.5%

have a hearing loss, whereas 28.6% of the children with cases. Dropout of 29 (25.0%) of 116 children was ob-
risk factors had normal hearing. The distribution of the served, in which case the complete diagnostic protocol
different risk factors is presented in Fig 2. could not be conducted. A specific etiologic diagnosis
Familial history of deafness was the most frequently was established for 48 (55.2%) of the 87 children for
encountered risk factor for congenital hearing loss in our whom a permanent hearing loss was confirmed. For the
population. Risk factors were not statistically different remaining 39 children (44.8%), the cause underlying
between the normal hearing and hearing loss groups the congenital hearing loss remained unknown. The
(Kruskal-Wallis test, P ⫽ .44). distribution of the etiologic diagnoses is presented in Fig
The screening population of NICU infants demon- 3. Of the causes identified, a genetic cause was present in
strated a permanent hearing loss in 61.5% of cases. The 60.4% of cases, peripartal problems in 20.8%, and con-
male/female ratio was 3.0. The median hearing loss of genital CMV infection in 18.8%. With respect to the
the hearing-impaired children was 60 dB nHL. The genetic causes of hearing loss in our population, muta-
most-prevalent risk factors were mechanical ventilation, tions in the GJB2 gene accounted for 37.9% of the cases,
low birth weight, and hyperbilirubinemia. chromosomal aberrations were found in 13.8% (n ⫽ 4),
specific syndromic deafness for which the underlying
Etiologic Evaluation molecular basis was known was proven in 6.9% (n ⫽ 2),
Additional diagnostic testing for the permanently hear- and a yet unknown genetic factor was involved in
ing-impaired infants (n ⫽ 116) could be performed in 87 41.4% (n ⫽ 12). The last category was based on pedigree

Asphyxia Cerebral bleeding

Bacterial meningitis 3.4% 3.4%
ABO incompatibility

Syndromic deafness
FIGURE 3 No diagnosis 10.3%
Etiologic diagnoses for children with congenital permanent 44.8%
hearing loss.
CX 26

Fetal Alcohol syndrome


Familial hearing loss 4.6%

1122 DECLAU et al
Downloaded from by guest on February 8, 2017
TABLE 5 Genotype-Phenotype Correlation in Children With GJB2 ABO incompatibility (n ⫽ 2), bacterial meningitis (n ⫽
Mutations 1), and fetal alcohol syndrome (n ⫽ 1). The other labo-
ratory tests demonstrated various findings in a number
Patient GJB2 Mutation Hearing Threshold, dB nHL
of cases, without any contribution to the final etiologic
Right Ear Left Ear diagnosis. Ophthalmologic examination revealed 5 in-
1 35delG homozygote ⬎100 ⬎100 fants with eye abnormalities, including astigmatism (n ⫽
2 35delG homozygote ⬎100 ⬎100 2), eye ptosis, pseudofakia, and strabismus. In the
3 35delG/1VS1 ⫹ 1G⬎A 70 90 present population, ophthalmologic evaluations were of
4 35delG/1VS1 ⫹ 1G⬎A 50 65 only limited importance for the diagnosis of deafness.
5 35delG homozygote ⬎100 ⬎100
Radiologic imaging was performed for 37 patients
6 35delG homozygote ⬎100 20
7 35delG homozygote 60 80
(38.5%) exhibiting sensorineural hearing loss of ⱖ60 dB
8 35delG homozygote ⬎100 90 nHL (n ⫽ 96). Of those patients, 11 (29.7%) showed
9 71G⬎A homozygote ⬎120 ⬎120 positive radiologic findings. Computed tomography had
10 C101T⬎C ⫹ C71G⬎A 50 70 a slightly higher rate of abnormality detection (n ⫽ 33;
11 35delG ⫹ 35dupG 120 120 27%) than did MRI (n ⫽ 23; 21%).
In aural atresia, the classical malformations of the
external and middle ear were found. Minor inner ear
malformations at the semicircular canals were described
characteristics demonstrating a mendelian type of inher-
in 3 cases, but no Mondini-like defects were observed.
itance or specific phenotypes for which the gene or the
Electrocardiographic findings were abnormal in only 1
sequence is not yet known.
patient, with Jervell and Lange-Nielsen syndrome.
For 11 children, hearing loss was attributed to muta-
tions in the GJB2 gene. A detailed description of the
different GBJ2 mutations and the resulting hearing loss is
Hearing loss is one of the most common congenital
shown in Table 5. All of these children were either
anomalies, occurring in ⬃1 to 2 infants per 1000. Left
homozygous or compound heterozygous. Homozygotic
undetected, hearing impairment in infants can nega-
carriage of the 35delG mutation was the most common
tively affect speech and language acquisition, academic
pattern but demonstrated variable clinical expression.
achievement, and social and emotional development.
Apart from GBJ2 mutations, the following specific
These negative effects can be diminished and even elim-
genetic phenotypes were found: trisomy 21 (n ⫽ 2),
inated through early intervention at or before 6 months
partial trisomy 19p (n ⫽ 1), trisomy 18 (n ⫽ 1), Jervell
of age.3 To the best of our knowledge, this is the first
and Lange-Nielsen syndrome (n ⫽ 1), Waardenburg
report in the literature that provides data on audiologic
syndrome (WS) (n ⫽ 1), and craniofacial malformations
confirmation and etiologic assessment in a large sample
(n ⫽ 6). The last category included 4 cases of aural
of newborns referred after failed UNHS.
Many UNHS programs are experiencing difficulty get-
For 9 children, a diagnosis of congenital CMV infec-
ting all or most of the infants referred from the screening
tion was made. CMV was detected through PCR assays
program to complete the diagnostic process and to be
with the blood spots on Guthrie cards, viral cultures, or
enrolled in intervention programs. Attrition rates as high
blood serologic tests. A detailed description of these chil-
as 60% between the initial referral and diagnostic con-
dren and their hearing thresholds at birth and at fol-
firmation are not unusual.10 Moreover, an important
low-up evaluations is presented in Table 6.
deficiency of virtually all UNHS programs is that they
Perinatal causes of hearing loss were cerebral bleeding
lack information on diagnostic evaluations.6 The current
(n ⫽ 3), asphyxia (n ⫽ 3), kernicterus attributable to
tracking system organized by the federal health care
agency seems to be very effective, because virtually all
patients sent by the organization were actually seen for
TABLE 6 Hearing Thresholds at Time of Diagnosis and Evolution in follow-up audiologic confirmation. In addition, organiz-
Children With Hearing Loss Attributable to Congenital ing the etiologic evaluation was quite efficient, with a
CMV Infection dropout rate of only 25.0%. The search for an underly-
Patient Hearing Threshold, Age at Last Follow-up Evolution ing cause must be undertaken through a multidisci-
dB nHL Evaluation, y plinary approach involving an ear, nose, and throat sur-
Right Ear Left Ear geon, a pediatrician, a geneticist, an ophthalmologist, a
radiologist, and other specialists when indicated by the
1 40 80 3.0 Stable
2 ⬎100 40 2.5 Stable
clinical findings. An efficiently managed diagnostic pro-
3 55 45 Lost to follow-up ? tocol decreases the burden for the parents and child and
monitoring limits the risk of noncompliance.
4 ⬎100 ⬎100 1.2 Stable The presence of hearing loss (either unilateral or bi-
5 70 60 2.4 Progressive lateral) was confirmed in 68.2% of the referrals. In
6 120 120 1.8 Stable 31.8%, hearing was found to be normal after a thorough
7 50 70 0.9 Stable audiometric evaluation. In 60.4% of the cases, exactly
8 40 120 1.1 Progressive analogous results were found in screening and diagnos-
9 60 100 0.8 Stable tic testing. Our results suggest that the accuracy of new-

PEDIATRICS Volume 121, Number 6, June 2008 1123

Downloaded from by guest on February 8, 2017
born hearing screening remains an issue, even if the northern European ancestry and has a carrier frequency
positive predictive value for the UNHS program orga- of ⬃2.5% in the general population. It was also the most
nized by the federal health care agency in Belgium has common allele variant of GJB2 in our population. As
been reported to be as high as 32%.6 illustrated in Table 5, there was a variable genotype-
Children with unilateral refer results were more likely phenotype correlation even among children homozy-
to have normal hearing, but 11.6% of those children gous for 35delG, which is known to be associated with
were diagnosed as having bilateral hearing loss. These profound hearing loss. Our data are in accordance with
findings underscore the need for a comprehensive au- those from a large multicenter study reported by Sno-
diometric evaluation in both unilateral and bilateral re- eckx et al.16 Those authors found that the degree of
ferral cases. Limiting the audiometric protocol to bilat- hearing loss associated with biallellic truncating muta-
eral referral cases cannot be justified, because unilateral tions (such as 35delG) was more severe than that asso-
referral cases with bilateral hearing loss (and the need ciated with biallellic nontruncating mutations. A diag-
for rehabilitation) would go undetected. nosis of GJB2 mutations has important implications for
The same holds true for limiting screening to new- parental counseling. Hearing loss attributable to GJB2
borns with risk factors for congenital hearing loss. A total mutations is usually nonprogressive, and prenatal diag-
of 55.8% of children with hearing loss had no risk fac- nosis is possible in the case of future children.14–16
tors at all. The presence of ⬎1 risk factor does not WS is the most common type of autosomal dominant
necessarily imply a cumulative risk for hearing loss, be- syndromic hearing loss, in association with minor de-
cause 4 children with multiple risk factors had normal fects in structures arising from the neural crest and pig-
hearing. mentation anomalies.17 It consists of variable degrees of
In some European countries, such as the United King- sensorineural hearing loss and pigmentary abnormalities
dom, an etiologic investigation is usually conducted only of the skin, hair (white forelock), and eyes (heterochro-
for children with bilateral hearing loss of ⬎35 dB nHL mia iridis). Four types are recognized (ie, WS I, WS II,
identified through the National Newborn Hearing WS III, and WS IV), on the basis of the presence of other
Screening Program.11 Interestingly, in both the GJB2 and abnormalities. WS I and WS II share many features but
congenital CMV groups, some children had only 1 af- have an important phenotypic difference; WS I is char-
fected ear. If an etiologic diagnosis had been sought only acterized by the presence of dystopia canthorum,
for children with bilateral hearing loss, then those chil- whereas WS II is characterized by its absence. In WS III,
dren would have been missed. upper-limb abnormalities are present. In WS IV, Hirsch-
According to the literature, prelingual hearing loss is sprung disease is present. Mutations in PAX3 cause WS I
attributed to genetic causes in 50% of cases and to and WS III. Mutations in MITF cause some cases of WS
environmental causes in the other 50%.12 In the present II. Mutations in EDNRB, EDN3, and SOX10 cause WS IV.
study, an etiologic diagnosis could be established in Jervell and Lange-Nielsen syndrome is an autosomal
55.2% of the cases. Of the causes identified, a genetic recessive form of profound bilateral congenital deafness
cause was present in 60.4% of the cases and environ- associated with prolongation of the QT interval, as de-
mental problems in 39.6%. tected through electrocardiography (the abnormal QTc is
Among the genetic causes, ⬃30% were reported pre- ⬎440 milliseconds). It is caused specifically by mutation
viously to be syndromic and 70% nonsyndromic.12 of the KCNE1 and KCNQ1 genes. Among untreated indi-
Within the prelingual, nonsyndromic, hearing loss viduals with Jervell and Lange-Nielsen syndrome,
group, inheritance is reported to be 75% to 80% auto- ⬃50% die by the age of 15 years, as a result of ventric-
somal recessive, 20% to 25% autosomal dominant, and ular arrhythmias.18
15 to 1.5% X-linked. Although nonsyndromic sensori- Fetal alcohol spectrum disorder describes a spectrum
neural hearing loss is very heterogeneous, mutations in of permanent and often devastating birth defect syn-
the GJB2 gene account for nearly 50% of cases of con- dromes caused by maternal consumption of alcohol dur-
genital, autosomal recessive, nonsyndromic, sensorineu- ing pregnancy. The main effect of fetal alcohol exposure
ral hearing loss in the white population.13 is brain damage. Church and Abel19 observed that fetal
Looking specifically at the genetic causes of hearing alcohol spectrum disorder is associated with 4 kinds of
loss in our population (n ⫽ 29), syndromic sensorineural hearing disorders, that is, (1) developmentally delayed
hearing loss accounted for 31.0% and nonsyndromic speech and language development, (2) sensorineural
phenotypes for 69.0%. Mutations in the GJB2 gene ac- hearing loss, (3) intermittent conductive hearing loss
counted for 38% of the genetic cases, specific syndromic attributable to recurrent serous otitis media, and (4)
deafness for 7%, and chromosomal aberrations for 14%; central hearing loss. As is the case with other syndromes
in 41% of cases, a mendelian phenotype without known associated with craniofacial anomalies and hearing im-
gene and/or sequence was involved. This finding is in pairments, speech and language pathologic conditions
agreement with the finding that GJB2 mutations account also are common in patients with fetal alcohol spectrum
for 30% to 40% of genetic causes of prelingual hearing disorder. In our population, 1 child demonstrated clini-
loss throughout the world. cal signs of fetal alcohol spectrum disorder.
Hearing impairment in subjects in the biallellic non- Craniofacial malformations were observed in 12.5%
truncating GJB2 mutations ranges from mild to profound of cases (n ⫽ 6). In 4 cases, aural atresia was determined.
and is usually nonprogressive.14,15 The c35delG (35delG According to Das,20 chromosomal aberrations may
mutation) variant is the most common in children of account for 3.8% of congenital hearing loss. In our pop-

1124 DECLAU et al
Downloaded from by guest on February 8, 2017
ulation, trisomies 18, 21, and 19p accounted for 8.3% dyssynchrony.31 In this particular case, the disorder is
(n ⫽ 4) of the identified causes of prelingual hearing loss. attributed to functional alterations in the inner hair cells.
In the literature, the most important environmental In 1 child, auditory dyssynchrony was caused by hy-
factors responsible for prelingual hearing loss are con- perbilirubinemia attributable to Rh factor incompatibil-
genital infections (mainly rubella in nonvaccinated areas ity. The child was treated with exchange transfusion
and CMV), ototoxicity, prematurity, and asphyxiation.12 during the newborn period, and later she received a
In our study group, 39.6% of the cases of permanent cochlear implant. For the other child, no risk factors
hearing loss were attributable to environmental causes. were present and mutations in the OTOF gene were
Congenital CMV infection was recognized as the most excluded.
frequent cause of acquired hearing loss in neonates. In It is obvious that identification of the cause of the
our population, congenital CMV infection was present hearing loss provides new information relevant to hear-
in 18.8% of cases. A diagnosis of congenital CMV infec- ing loss management, coexisting medical problems, and
tion was based on PCR assays with dried blood spots on the prognosis for the child and family. In addition, these
Guthrie cards, viral cultures from urine or saliva sam- investigations clarify the epidemiological features of con-
ples, and the presence of specific IgM antibodies. Con- genital deafness, which may facilitate the planning of
genital CMV infection is a challenging diagnosis.21 Once effective hearing loss prevention and surveillance pro-
children are older than 2 to 3 weeks, the diagnosis grams.
becomes sometimes difficult, because viral excretion in
the urine might be intermittent and a postnatal infection CONCLUSIONS
cannot be ruled out. In more-recent years, implementa- A diagnosis of congenital hearing loss (either unilateral
tion of PCR analysis of viral DNA in dried blood spots has or bilateral) was made for 68.2% of neonates referred for
significantly improved the diagnostic power for congen- audiometric evaluation after failed UNHS. For 55.2% of
ital CMV infection.22,23 An early diagnosis of congenital those children, an etiologic diagnosis could be made after
CMV has important implications from a therapeutic a comprehensive etiologic evaluation. This evaluation
point of view. Clinical studies with treatment protocols allows for appropriate, individually tailored rehabilita-
based on ganciclovir are underway and promising, al- tion, audiometric follow-up monitoring, and parental
though not applied in the present study.24,25 Hearing loss counseling. Our diagnostic abilities are likely to improve
attributable to congenital CMV may be progressive in in the near future, with advances in molecular genetics
⬎50% of cases,26 and regular audiometric follow-up and the rapidly expanding knowledge concerning genes
monitoring is required until the age of 6 years. Both involved in nonsyndromic sensorineural hearing loss.
aspects underscore the need for correct early diagnosis This would significantly improve the strength of UNHS
and appropriate counseling of the parents. programs and allow for the rapid identification of pre-
According to Fortnum and Davis,27 bacterial menin- lingual hearing loss.
gitis accounts for 6% of sensorineural hearing loss in
children. In a study by Koomen et al, 75% of patients ACKNOWLEDGMENT
were ⬍2 years of age.28 Bacterial meningitis was a cause We thank the audiologists in our department for their
of acquired neonatal hearing loss in 1 child (2.1%) in outstanding professional help and diligence in screening
our population. newborns.
A diagnosis of auditory dyssynchrony was estab-
lished for 2 children (4.2%). This disorder is charac- REFERENCES
terized by a dysfunction in neural/brainstem transmis- 1. Parving A, Hauch AM, Christensen B. Hearing loss in children:
sion of auditory stimuli in the presence of normal epidemiology, age at identification and causes through 30 years
outer hair cell function. This condition translates into [in Danish]. Ugeskr Laeger. 2003;165(6):574 –579
absent or abnormal ABR responses and normal oto- 2. van Straaten HL, Tibosch CH, Dorrepaal C, Dekker FW, Kok
acoustic emissions. The clinical significance is that this JH. Efficacy of automated auditory brainstem response hearing
may result in impaired development of normal audi- screening in very preterm newborns. J Pediatr. 2001;138(5):
674 – 678
tory behavior or oral language, often leading to co-
3. Yoshinaga-Itano C, Coulter D, Thomson V. Developmental
chlear implantation. According to Foerst et al,29 a outcomes of children with hearing loss born in Colorado hos-
prevalence of 0.94% was found within the group at pitals with and without universal newborn hearing screening
risk for hearing loss, compared with 8.44% among programs. Semin Neonatol. 2001;6(6):521–529
profoundly hearing-impaired children. Protocols that 4. Van Kerschaver E, Stappaerts L. ALGO Gehoorsscreening: Rapport
use otoacoustic emission screening should refer in- van de Werkjaren 2001 & 2002 [ALGO Hearing Screening: Report of
fants with hyperbilirubinemia for ABR testing, be- the Work Years 2001 and 2002; in Dutch]. San Carlos, CA: ALGO
cause there is mounting evidence in the literature Natus Medical Inc; 2003
5. Thompson DC, McPhillips H, Davis RL, Lieu TL, Homer CJ,
showing linkage between auditory dyssynchrony and
Helfand M. Universal newborn hearing screening: summary of
hyperbilirubinemia.30 In the present study, every child evidence. JAMA. 2001;286(16):2000 –2010
who needed at least phototherapy was considered at 6. Declau F, Doyen A, Robillard T, Janssens de Varebeke S. Uni-
risk for hearing loss. versal newborn hearing screening. B-ENT. 2005;1(suppl 1):
Mutations in the gene encoding otoferlin (OTOF), 16 –23
known as DFNB9, also may cause a pattern of auditory 7. Joint Committee on Infant Hearing, American Academy of

PEDIATRICS Volume 121, Number 6, June 2008 1125

Downloaded from by guest on February 8, 2017
Audiology, American Academy of Pediatrics, American speech, language, and vestibular disorders. Obstet Gynecol Clin
Speech-Language-Hearing Association, Directors of Speech North Am. 1998;25(1):85–97
and Hearing Programs in State Health and Welfare Agencies. 20. Das VK. Aetiology of bilateral sensori-neural deafness in chil-
Year 2000 position statement: principles and guidelines for dren. Scand Audiol Suppl. 1988;30:43–52
early hearing detection and intervention programs. Pediatrics. 21. Liberek A, Rytlewska M, Szlagatys-Sidorkiewicz A, et al. Cy-
2000;106(4):798 – 817 tomegalovirus disease in neonates and infants: clinical presen-
8. Bamiou DE, Phelps P, Sirimanna T. Temporal bone computed tation, diagnostic and therapeutic problems, own experience.
tomography findings in bilateral sensorineural hearing loss. Med Sci Monit. 2002;8(12):CR815–CR820
Arch Dis Child. 2000;82(3):257–260 22. Barbi M, Binda S, Primache V, Luraschi C, Corbetta C. Diag-
9. General Executive Board on Statistics and Economic Information. nosis of congenital cytomegalovirus infection by detection of
Bevolkingsstatistieken [Population Statistics]. Brussels, Belgium: FOD viral DNA in dried blood spots. Clin Diagn Virol. 1996;6(1):
Economie, KMO, Middenstand en Energie; 2006 27–32
10. Deben K, Janssens de Varebeke S, Cox T, Van de Heyning P. 23. Barbi M, Binda S, Caroppo S, Primache V. Neonatal screening
Epidemiology of hearing impairment at three Flemish insti- for congenital cytomegalovirus infection and hearing loss.
tutes for deaf and speech defective children. Int J Pediatr Oto- J Clin Virol. 2006;35(2):206 –209
rhinolaryngol. 2003;67(9):969 –975 24. Smets K, De Coen K, Dhooge I, et al. Selecting neonates with
congenital cytomegalovirus infection for ganciclovir therapy.
11. Bamford J, Uus K, Davis A. Screening for hearing loss in
Eur J Pediatr. 2006;65(12):885– 890
childhood: issues, evidence and current approaches in the UK.
25. Kimberlin DW, Lin C-Y, Sanchez PJ, et al. Effect of ganciclovir
J Med Screen. 2005;12(3):119 –124
therapy on hearing in symptomatic congenital cytomegalovi-
12. Smith RJ, Van Camp G. Non-syndromic hearing impairment:
rus disease involving the central nervous system: a random-
gene linkage and cloning. Int J Pediatr Otorhinolaryngol. 1999;
ized, controlled trial. J Pediatr. 2003;143(1):16 –25
49(suppl 1):S159 –S163
26. Fowler KB, Boppana SB. Congenital cytomegalovirus (CMV)
13. Smith RJH, Bale JF, White KR. Sensorineural hearing loss in
infection and hearing deficit. J Clin Virol. 2006;35(2):226 –231
children. Lancet. 2005;365(9462):879 – 890
27. Fortnum H, Davis A. Hearing impairment in children after
14. Denoyelle F, Marlin S, Weil D, et al. Clinical features of the bacterial meningitis: incidence and resource implications. Br J
prevalent form of childhood deafness, DFNB1, due to a con- Audiol. 1993;27(1):43–52
nexin-26 gene defect: implications for genetic counselling. Lan- 28. Koomen I, Grobbee DE, Roord JJ, et al. Hearing loss at school
cet. 1999;353(9161):1298 –1303 age in survivors of bacterial meningitis: assessment, incidence,
15. Murgia A, Orzan E, Polli R, et al. Cx26 deafness: mutation and prediction. Pediatrics. 2003;112(5):1049 –1053
analysis and clinical variability. J Med Genet. 1999;36(11): 29. Foerst A, Beutner D, Lang-Roth R, et al. Prevalence of auditory
829 – 832 neuropathy/dyssynchrony in a population of children with
16. Snoeckx RL, Huygen PLM, Feldmann D, et al. GJB2 mutations profound hearing loss. Int J Pediatr Otorhinolaryngol. 2006;
and degree of hearing loss: a multicenter study. Am J Hum 70(8):1415–1422
Genet. 2005;77(6):945–957 30. Berg AL, Spitzer JB, Towers HM, et al. Newborn hearing
17. Newton VE. Clinical features of the Waardenburg syndromes. screening in the NICU: profile of failed auditory brainstem
Adv Otorhinolaryngol. 2002;61:201–208 response/passed otoacoustic emission. Pediatrics. 2005;116(4):
18. Tyson J, Tranebjaerg L, McEntagart M, et al. Mutational spec- 933–938
trum in the cardioauditory syndrome of Jervell and Lange- 31. Rodriguez-Ballesteros M, del Castillo FJ, Martin Y, et al. Au-
Nielsen. Hum Genet. 2000;107(5):499 –503 ditory neuropathy in patients carrying mutations in the oto-
19. Church MW, Abel EL. Fetal alcohol syndrome: hearing, ferlin gene (OTOF). Hum Mutat. 2006;22(6):451– 456

1126 DECLAU et al
Downloaded from by guest on February 8, 2017
Etiologic and Audiologic Evaluations After Universal Neonatal Hearing Screening:
Analysis of 170 Referred Neonates
Frank Declau, An Boudewyns, Jenneke Van den Ende, Anouk Peeters and Paul van den
Pediatrics 2008;121;1119
DOI: 10.1542/peds.2007-1479
Updated Information & including high resolution figures, can be found at:
Services /content/121/6/1119.full.html

References This article cites 27 articles, 6 of which can be accessed free at:
Citations This article has been cited by 12 HighWire-hosted articles:
Subspecialty Collections This article, along with others on similar topics, appears in the
following collection(s):
Ear, Nose & Throat Disorders
Fetus/Newborn Infant
Permissions & Licensing Information about reproducing this article in parts (figures, tables)
or in its entirety can be found online at:
Reprints Information about ordering reprints can be found online:

PEDIATRICS is the official journal of the American Academy of Pediatrics. A monthly publication, it
has been published continuously since 1948. PEDIATRICS is owned, published, and trademarked by
the American Academy of Pediatrics, 141 Northwest Point Boulevard, Elk Grove Village, Illinois,
60007. Copyright © 2008 by the American Academy of Pediatrics. All rights reserved. Print ISSN:
0031-4005. Online ISSN: 1098-4275.

Downloaded from by guest on February 8, 2017

Etiologic and Audiologic Evaluations After Universal Neonatal Hearing Screening:
Analysis of 170 Referred Neonates
Frank Declau, An Boudewyns, Jenneke Van den Ende, Anouk Peeters and Paul van den
Pediatrics 2008;121;1119
DOI: 10.1542/peds.2007-1479

The online version of this article, along with updated information and services, is located
on the World Wide Web at:

PEDIATRICS is the official journal of the American Academy of Pediatrics. A monthly

publication, it has been published continuously since 1948. PEDIATRICS is owned, published,
and trademarked by the American Academy of Pediatrics, 141 Northwest Point Boulevard, Elk
Grove Village, Illinois, 60007. Copyright © 2008 by the American Academy of Pediatrics. All
rights reserved. Print ISSN: 0031-4005. Online ISSN: 1098-4275.

Downloaded from by guest on February 8, 2017