DEVELOPMENTAL MEDICINE & CHILD NEUROLOGY ORIGINAL ARTICLE

A population-based study and systematic review of hearing loss in

children with cerebral palsy
SUSAN M REID 1 | MAITREYI B MODAK 2 | ROBERT G BERKOWITZ 3 | DINAH S REDDIHOUGH 2

1 Departmental Disability Research, Murdoch Childrens Research Institute and Department of Paediatrics, University of Melbourne, Melbourne, Vic., Australia. 2 Department of
Developmental Medicine, Royal Children's Hospital, Melbourne, Vic., Australia. 3 Department of Otolaryngology, Royal Children's Hospital, Melbourne, Vic., Australia.
Correspondence to Dr Sue Reid at Developmental Medicine, Royal Children's Hospital, Flemington Road, Parkville, Vic. 3052, Australia. E-mail: sue.reid@mcri.edu.au

This article is commented on by Cooley Hidecker on pages 977–978 of this issue.

PUBLICATION DATA
Accepted for publication 22nd May 2011.
Published online 6th September 2011.
ABBREVIATIONS
CHL Conductive hearing loss
SNHL Sensorineural hearing loss
VCPR Victorian Cerebral Palsy Register
AIM The aims of this study were to estimate the frequency of hearing loss in children with
cerebral palsy (CP), to examine factors associated with hearing loss, and to describe aspects of
hearing in a population sample of children with CP and hearing loss.
METHOD A systematic review of the international literature was undertaken, and data on the
frequency of hearing loss or severe hearing loss were extracted from 14 data sets based on
previously devised criteria. Six hundred and eight-five children with CP (406 males, 279 females)
born in Victoria, Australia, between 1999 and 2004 were identified from the Victorian Cerebral
Palsy Register. Children were included if they had an established post neonatal cause for their CP
before the age of 2 years. Additional information was collected on 48 children with documented
hearing loss based on a four-tone pure tone average in the better ear.
RESULTS There was considerable variation in the definitions and proportions of hearing loss
(range 4–13%) and severe hearing loss (range 2–12%) reported by CP registries in developed
countries. In Victoria, 7% of individuals with CP had bilateral hearing loss of a moderate to pro-
found degree, whereas the subgroup with a severe–profound degree of loss constituted 3% to 4%
of the CP population.
INTERPRETATION These population-based data are likely to more accurately reflect the true
frequency of defined hearing loss in children with CP than previous reviews.

The most recent consensus definition of cerebral palsy (CP)
emphasizes the importance of other conditions and impair-
ments that commonly accompany the motor disorder.1 The
American Academy of Neurology practice guidelines recom-
mend that all children with CP are assessed for intellectual,
visual, and hearing impairments, and for speech and language
disorders.2 Early assessment and identification of hearing loss
is especially important because of its potential to impact on
speech–language, cognitive, and psychosocial development.3
Sensorineural, conductive, or mixed hearing loss in children
with CP may add to the overall level of impairment. Conduc-
tive hearing loss (CHL) occurs when the conduction of sound
through the outer and middle ear is disrupted, affecting hear-
ing before the sound reaches the cochlea and the nerve recep-
tors of the inner ear. Disturbances of the conductive
mechanism in children are commonly caused by otitis media,
and most respond well to medical management.4 Sensorineu-
ral hearing loss (SNHL) results from damage to the neural
receptors of the inner ear, the nervous pathways to the brain,
or the area of the brain that receives auditory stimuli. Hearing
loss of this type can be congenital or acquired. In mixed
hearing loss there are components of both SNHL and
CHL. Among the paediatric population, there is a clinically
recognizable genetic cause in approximately 50% of cases of
SNHL.5 Conversely, although children with CP may have a
genetic cause for their hearing loss unrelated to the aetiology
of their CP, non-genetic causes are more likely because many
non-genetic risk factors are common to both conditions,
including intrauterine cytomegalovirus infection,6 severe
hypoxic–ischaemic insults,2,7 very low birthweight,2,8
extracorporeal membrane oxygenation,9 neonatal hyperbiliru-
binaemia,7,10 and neonatal meningitis.2 Permanent bilateral
hearing loss in childhood of at least moderate degree occurs in
around 0.1% of the population,11 but this increases to 1.9%
for infants surviving admission to a neonatal intensive care
unit.12
Most CP registries record hearing status, classifying hearing
into broad categories based on information gained as a result
of behavioural or physiological audiological testing. Pure tone
audiometry is the key behavioural test used to identify hearing
threshold levels in typically developing children. Test results
are plotted on audiograms, which show the difference, mea-
sured in decibels (dB), between the hearing threshold and a
reference threshold of 0dB hearing loss at each frequency. If

1038 DOI: 10.1111/j.1469-8749.2011.04069.x ª The Authors. Developmental Medicine & Child Neurology ª 2011 Mac Keith Press
possible, hearing in each ear is tested by air conduction using
earphones and by bone conduction using an oscillator in
contact with the mastoid process or forehead. Pure tone
audiograms are typically described, per ear, by the type
(SNHL, CHL, mixed), configuration (e.g. flat, sloping, rising),
and degree of loss (mild, moderate, moderately severe, severe,
profound).13 Serial audiograms are used to determine if the
hearing loss is stable, progressive, or fluctuating.
Problems may be encountered with all forms of behavioural
audiometry in children with severe CP. It may be difficult to
keep a child’s upper body steady in order to facilitate a head
turn for visual reinforcement audiometry, or for a child to use
their arms and hands for play audiometry. Many children are
not able to push a button and some do not have a yes ⁄ no
response. In many cases, severe motor impairment may be
accompanied by severe cognitive impairment, making any
form of behavioural assessment unreliable. If a child is
unsuitable for behavioural testing, or if the results are unreli-
able, further evaluation by physiological assessment of the
auditory system can be performed, for instance by auditory
brainstem response audiometry, but this may necessitate
sedation in children with CP, which may be undesirable in
children with severe motor impairment.
Estimates as high as 30 to 40% have been reported for the
prevalence of hearing loss in CP,14 but evidence from the only
available systematic review, performed on behalf of the Ameri-
can Academy of Neurology, suggested that hearing loss occurs
in approximately 12% of children with CP.2 A range of 4 to
15% was reported for the four included studies.15–18 The two
studies reporting the highest frequency of hearing loss (14%
and 15%) used clinic- or referral-based samples,15,18 whereas
the lower estimates (4% and 7%) were based on population
samples from metropolitan Atlanta and Finland respec-
tively.16,17 The Academy recommended in a practice parame-
ter that large cohorts of children with CP should be studied to
determine specific features based on clinical subtype. These
population data, the Academy claimed, would facilitate evalua-
tion of the effectiveness of early intervention strategies2 and
enable international comparisons and investigation of aetio-
logical trends.19 As most geographically based CP registries
collect data on hearing loss, we thought it likely that reports
from population registries, in addition to the two listed above,
would collectively provide a higher level of evidence for the
true prevalence of hearing loss in CP.
This study aimed to estimate the frequency of hearing loss
in CP using data from the Victorian Cerebral Palsy Register
(VCPR) in Australia, as well as from other population-based
registries from developed countries. A further objective was to
examine associations between hearing loss in CP and demo-
graphic, perinatal, and clinical variables routinely collected by
the VCPR, including motor type, topographical pattern, gross
motor function, and the presence of other associated condi-
tions and impairments. Based on previous research, we
expected that the likelihood of hearing loss would increase
with increasing extent and severity of motor impairment.20
The final aim was to describe specific features of hearing loss
in a smaller birth cohort of children with CP born in Victoria
What this paper adds
• Considerable variation exists in estimates of the frequency of hearing loss in
CP.
• Consistent definitions are needed to enable comparison of data sets.
• Unaided hearing loss >40dB in the better ear across frequencies of 500 to
4000Hz occurs in approximately 4% to 8% of individuals with CP
• Unaided hearing loss >70dB in the better ear across frequencies of 500 to
4000Hz occurs in approximately 2% to 4% of individuals with CP.
between 1999 and 2004 (Victorian cohort study), including
the type, laterality, and degree of hearing loss, investigative
findings, interventions received, and relevant causal or ante-
cedent factors.
METHOD
Setting
This epidemiological study was undertaken at the Royal Chil-
dren’s Hospital in Melbourne, Australia, and approval for the
project was granted by the Human Research Ethics Commit-
tee at that institution.
Systematic review of the literature
Searches were performed via MEDLINE for full-text
articles published in English between 1980 and 2010 con-
taining data on the frequency of hearing loss within CP
population cohorts in developed countries. The search term
used was ‘cerebral palsy ⁄ epidemiology’ as a major subject
heading. Titles and abstracts were first screened for
evidence that the data originated from a population-based
registry in a developed country, included all types of CP,
and contained information on accompanying impairments.
Subsequent screening involved perusing the body of the
manuscript for data on hearing impairment. Published reg-
istry reports and bibliographies of published papers and
review articles were searched for additional material. A flow
diagram of the process used to select studies is shown in
Figure 1.
Information was collected on each registry and each publi-
cation, including the definitions and classification systems used
for hearing loss. The most recent and largest data set was
chosen where multiple publications were found, or, where
serial reports were available, data were amalgamated to cover a
greater time period. This information and the proportion of
cases of CP with any degree of hearing loss or severe loss was
assessed and extracted independently by the first author using
a standardized form. The proportions were recalculated,
where necessary, to exclude missing data. Additional informa-
tion was sought from the Western Australian Cerebral Palsy
Register on hearing impairment in children born between
2000 and 2004 to supplement the data published in their latest
report. To minimize bias, a CP prevalence of at least 1.5 cases
per 1000 births and no more than 20% missing data were set a
priori as requirements for inclusion.
For the statistical analysis, the number of cases of CP with
either hearing loss or severe hearing loss was divided by the
total number of cases in each reported data set to determine
the proportions. The standard error and 95% confidence
intervals (CIs) around that estimate were calculated for each
Hearing Loss in CP Susan M Reid et al. 1039
 Identification
380 studies identified 6 additional studies identified
through MEDLINE search through other sources

386 studies after duplicates removed

Screening
386 studies screened 365 studies excluded:
no data on hearing

21 studies assessed 7 studies excluded:

Eligibility

for eligibility duplicate registry (6)

>20% missing data (1)

14 studies included in
qualitative synthesis
Included

14 studies included in
quantitative synthesis

Figure 1: Flow diagram of the process used to select studies.

data set. The mean proportion was then calculated over all
available data, weighted on the sample size of the study. All
statistical analyses were performed using STATA 11.0 (Stata-
Corp 2009, College Station, TX, USA).
Victorian cohort study
The VCPR was used to select 685 children with CP (406
males, 279 females) born in the Australian State of Victoria
between 1999 and 2004. Children with an established post
neonatal cause for their CP before the age of 2 years were
included in the study, whereas 24 children who had died
before the age of 5 years were excluded. From this cohort,
children with documented hearing loss at 5 years of age were
identified for more detailed study.
Data were collected from the VCPR on demographic, birth,
and clinical details, and on brain magnetic resonance imaging
(MRI) findings. The medical records of children with hearing
loss were reviewed and audiology reports were obtained from
multiple institutions to assess hearing status. Hearing assess-
ment was based on a combination of behavioural and physio-
logical methods as appropriate for the child’s developmental
age and impairments and on the expertise of the audiologist.
Behavioural assessments included behavioural observation
audiometry, visual reinforcement audiometry, visual reinforce-
ment operant conditioning audiometry, and conditioned play
audiometry. Where hearing fluctuated as a result of otitis
media, as suggested by the presence of a low-volume type B
tympanogram, data from the best hearing test were included.
The case definition for hearing loss was unaided hearing loss
of >40dB in the better ear, based on a four-tone pure tone
average covering 500, 1000, 2000, and 4000Hz air conduction
thresholds, or on non-quantitative audiological findings of
hearing loss of at least moderate degree requiring amplifi-
cation aids. Although the chosen four-tone pure tone aver-
age may not accurately reflect differing configurations of
hearing loss and frequencies higher than 4000Hz are
known to contribute to speech understanding,21 from a
practical point of view we wanted to employ a uniform
method across all children with CP that could easily be
utilized by CP registries worldwide to broadly but reliably
classify hearing status, regardless of configuration or ear-
specific differences.
For children meeting the criteria for inclusion in the Victo-
rian cohort study, the medical records and the most recent
audiology reports were used to document the type and degree
of hearing loss, intervention in the form of amplification aids
or cochlear implantation, and the presumed cause or anteced-
ent factors, if known. The type of hearing loss was determined
by the size of the difference between bone conduction and air
conduction hearing thresholds. Impedance tympanometry was
sometimes used as an additional indicator to suggest CHL
where bone conduction thresholds were unavailable. The
degree of hearing loss was categorized according to the four-
tone pure tone average in the better ear or binaurally. The
degree was described as moderate ⁄ moderately severe if the
hearing loss was 41 to 70dB, severe if the hearing loss was 71
to 90dB, and profound if the hearing loss was assessed as
>90dB based on the classification of Goodman.13
Demographic, birth, clinical, and imaging variables were
compared for children with and without hearing impairment

1040 Developmental Medicine & Child Neurology 2011, 53: 1038–1045

according to our case definition using the v2 test or the Fish-
er’s exact test if the expected frequency in any cell was <5. To
allow for multiple comparisons, a p-value of 0.01 was used to
denote statistical significance.
RESULTS
Systematic review
Reports from nine CP registries provided data on the propor-
tion of children with hearing loss, although only three of these
provided a definition based on audiometry (Table I).16,22–29
The weighted mean proportion for all cases of CP was 8%
(range 4–13%; Table I). The three reports that provided defi-
nitions used a loss of 40dB HL in the better ear as the mini-
mum level for inclusion, thereby excluding mild and unilateral
hearing loss. The mean proportion of children with hearing
loss in these three reports (5%) was lower than the mean in
reports without a minimum cut-off on audiological testing
(10%). One of these three reports also included children who
were unresponsive on audiological testing, a group that may
have included children with normal hearing but severe intel-
lectual impairment. If these children were excluded, similar
proportions (4–5%) were seen in all three reports.
Publications from 13 registries, including the VCPR,
reported the proportion of children with CP with severe hear-
ing loss.16,22–34 Five of these defined severe as hearing loss
>70dB, five as deafness but without a defined threshold on
audiometry, and one as deafness or the need for a hearing aid.
Two reports contained no specific definition of severe hearing
loss (Table I). The weighted mean percentage of all cases of
CP with a severe hearing loss was 3% (range 2–12%).
Victorian cohort study
Of the 685 children, 88 (13%) were recorded by the VCPR as
having some degree of hearing loss and were investigated fur-
ther. After reviewing the latest audiological testing results, 48
(7%) of these 685 children met our case definition for hearing
loss. The mean age of the entire cohort was 8 years and
1 month (SD 1y 7mo; range 5y 1mo to 11y) and there was no
evidence of a difference in the mean age of children with hear-
ing loss (8y 6mo [SD 1y 7m; range 5y 1mo to 10y 11mo] com-
pared with those without hearing loss (8y 1mo [SD 1y 8mo];
range 5y 1mo to 11y). The Victorian figure of 7% fell in the
middle of the range of 4 to 13% reported from the nine
reviewed CP registries (Table I). Twenty-four (3–4%) chil-
dren had severe–profound hearing loss, defined as >70dB, a
proportion that was the equal third highest of the 13 registries
contributing data to the systematic review on severe hearing
loss, where the range was 2 to 12% (Table I).
A comparison of factors relating to children with and
without hearing loss is shown in Table II. Compared with
children with CP and no hearing loss, children with hearing
loss were more likely to have a dyskinetic or hypotonic form
of CP and to be more severely affected, as evidenced by their
poorer gross motor function, more extensive motor involve-
ment, and increased likelihood of intellectual and visual
impairments.

Table I: International registry data on the frequency of hearing loss or severe hearing loss in cerebral palsy (CP), including the definitions used

Percentage
Total CP of all CP with
CP registry Years included cases hearing loss Definition

Any hearing loss

UKCP 1960–1997 6216 8 No definition
Merseyside and Cheshire, UK 1966–1989 1894 6 Bilateral loss >40dB or non-responsive
South East Thames, UK 1970–1974 322 10 No definition
Victoria, Australia 1970–2003 3003 12 No definition
Atlanta, USA 1975–1977 204 4 Bilateral pure-tone loss of >40dB unaided in better ear
at 500, 1000, and 2000Hz
Western Australia 1980–2004 1637 9 No definition
Northern Ireland 1981–1997 937 4 Loss >40dB across 500–4000Hz in better ear
West of Ireland 1990–1999 75 9 No definition
South Australia 1993–2002 262 13 No definition
Weighted mean – – 8 –
Severe hearing loss
UKCP 1960–1997 6216 2 Loss >70dB in better ear or clinical judgement
Merseyside & Cheshire, UK 1966–1989 1894 4 Bilateral loss ‡70dB or non responsive
Finland 1968–1982 112 2 No definition
South East Thames, UK 1970–1974 322 3 No definition
Victoria, Australia 1970–2003 3003 2 Bilateral deafness
Western Australia 1970–2004 1637 2 Bilateral deafness
Northern Ireland 1981–1993 937 2 Loss >70dB without aids
England 4Child 1984–2002 1292 3 Sensorineural deafness
West of Ireland 1990–1999 75 3 Loss >70dB
Iceland 1990–2003 139 2 Needing hearing aid or presence of deafness
South Australia 1993–2003 262 3 Bilateral deafness
Norway 1996–1998 280 4 Functional deafness
Quebec 1999–2002 243 12 Bilateral loss >70dB
Weighted mean – – 3 –

UKCP, United Kingdom Network of Cerebral Palsy Registers, Surveys and Databases.

Hearing Loss in CP Susan M Reid et al. 1041

 Table II: Factors associated with hearing loss in cerebral palsy (CP) for the birth years 1999 to 2004

No known hearing All known hearing Unknown hearing Total cohort

Characteristics loss (n=616) loss (n=48) status (n=21) (n=685) p value

Sex (%)
Male 60 57 43 59 0.831
Female 40 43 57 41
Gestational age (wks), %
20–27 11 19 0 11 0.384
28–31 15 10 5 14
32–36 16 15 5 15
37+ 58 56 86 59
Unknown 0 0 5 1
Birthweight (g), %
<1500 22 29 0 22 0.623
1500–2499 19 19 9 18
2500+ 56 52 86 57
Unknown 3 0 5 3
Neonatal care, %
No admission >7d 30 27 57 31 0.239
Admission >7d 47 63 10 47
Unknown 23 10 33 22
Apgar score at 5 min, %
0–6 17 10 9 16 0.228
7–10 78 88 81 79
Unknown 5 2 10 5
Neonatal seizures, %
No 55 44 52 54 0.100
Yes 22 31 10 22
Unknown 23 25 38 24
Motor type, %
Spastic 89 75 91 88 0.002a
Ataxic 5 2 0 5
Dyskinetic 4 13 0 4
Hypotonic 3 10 0 3
Unknown 0 0 9 0
Topographical pattern, %
Monoplegia ⁄ hemiplegia 36 19 43 35 <0.001
Diplegia ⁄ triplegia 33 19 38 32
Quadriplegia 31 62 9 32
Unknown 0 0 9 0
GMFCS level, %
I 37 15 33 35 <0.001
II 27 17 29 27
III 12 6 9 11
IV 13 31 5 14
V 10 31 5 12
Unknown 1 0 19 2
Intellect, %
No impairment 55 13 29 51 <0.001
Impairment 37 81 9 39
Unknown 8 6 62 10
Vision, %
No impairment 55 33 5 52 <0.001a
Strabismus only 19 10 5 18
Some impairment 19 46 5 21
Blind 5 10 5 5
Unknown 2 0 81 4
Epilepsy, %
No 72 58 57 71 0.034
Yes 27 42 10 28
Unknown 1 0 33 2
MRI finding (n=370), %
Normal 14 9 10 13 0.346a
Periventricular white matter injury 45 40 30 44
Grey matter injury 16 12 20 15
Focal vascular 10 9 0 10
Malformation 9 17 30 10
Non-specific ⁄ other 7 14 10 8
a
Fischer’s exact test. GMFCS, Gross Motor Function Classification System; MRI, magnetic resonance imaging.

1042 Developmental Medicine & Child Neurology 2011, 53: 1038–1045

 Of the 48 children in the 6-year Victorian cohort study with two for whom MRIs were available, one had changes within
well-defined hearing loss, 29 (60%) were known to have sen- the globus pallidus and the other was reported as normal. In
sorineural hearing loss, 14 (29%) had mixed sensorineural and the two children with perinatal vascular insults, imaging
conductive loss, and one (2%) had conductive loss based on findings were consistent with major vascular territory infarcts,
audiology after the age of 4 years. The type of hearing loss and two of the three children with perinatal hypoxic–ischae-
was unknown for four children. Details of the hearing loss for mic injury had grey matter lesions typically associated with
the 48 children with hearing loss are shown in Table III. In term insults; one had no imaging. Based on positive brain
half the children, the presumed cause for the hearing loss was MRI findings, the timing of acquisition of the brain abnormal-
unknown. One child had Waardenburg syndrome, a genetic ity was considered to be antenatal in 13 of 34 children, perina-
syndrome known to be associated with hearing loss. Intrauter- tal in 18, and post neonatal in three. Of the 24 children in
ine cytomegalovirus infection was the most common anteced- whom the aetiology of hearing loss was unknown, no imaging
ent factor, while kernicterus, intrapartum asphyxia, and post findings were available in four. Eleven children had changes in
neonatal infection were antecedents in three cases each. the periventricular white matter suggestive of preterm injury,
Recurrent otitis media was recognized as the main cause in the one had a malformation, one had primarily grey matter
two cases of CHL, but was also a present in the 14 cases of changes, six had other types of non-specific abnormalities, and
mixed loss and also in some children with SNHL. Computed in one imaging findings were normal.
tomography and ⁄ or MRI of the ear were known to have been
performed in 18 children, but only a small number of anoma- DISCUSSION
lies were found (Table III). This study explored the frequency and nature of hearing loss
Brain MRIs were available for 37 children. Of eight children in children with CP. Thirteen per cent of children born in
with intrauterine infection, brain imaging demonstrated the Victoria between 1999 and 2004 had documented hearing loss
presence of malformations consistent with early infection in of any degree up to the age of 5 years, but only 7% had bilat-
five, one had changes consistent with periventricular white eral hearing loss of >40dB on their latest testing, the cut-off
matter injury, one was normal, and the other had no MRI used to denote a moderate degree of hearing loss in the Victo-
available. Three children had hyperbilirubinaemia and, of the rian cohort study and by the three reviewed CP registries
whose publications included a definition for hearing impair-
Table III: Details of 48 children with hearing loss born in Victoria 1999– ment. Although we used 40dB hearing loss as our cut-off,
2004 lesser degrees of hearing loss can still have an impact on
communication. In fact, the World Health Organization
Characteristics Hearing loss (n=48) defines disabling hearing impairment in children as hearing
loss >30dB in the better ear, in acknowledgement of the belief
Age at latest audiology (y), %
0–2 44
that children acquiring language need greater hearing sensitiv-
3–5 15 ity than adults.35 We also determined that 3 to 4% of children
6+ 31 born in Victoria with CP were categorized as having severe
No report available 10
Type of hearing loss, %
hearing loss, based on bilateral loss of at least 70dB. These
Conductive 2 percentages were within the range reported by other CP regis-
Sensorineural 60 tries, although valid comparisons were limited by the lack of a
Mixed 29
Unknown type 8
consensus definition of hearing loss.
Degree of hearing loss, % We observed that the most recent international data series,
Moderate ⁄ moderately severe 50 including this one, showed the highest proportions of
Severe 4
Profound 46
severe ⁄ profound hearing loss, and this led us to speculate
Management of hearing loss, % whether there may have been a rise in the rates of bilateral
Hearing aids 69 severe ⁄ profound hearing loss over time. It is possible that the
Cochlear implant 38
Presumed cause of hearing loss, %
increase in the numbers of survivors of extremely preterm
Genetic syndrome 2 birth may have contributed to the overall higher rates of hear-
Antenatal vascular injury 2 ing loss, since the period between 20 and 33 weeks’ gestation
Intrauterine infection 17
Intraventricular haemorrhage 2
is one of rapid fetal audiological development, and the
Hyperbilirubinaemia 6 frequency of SNHL in this group is many times the frequency
Intrapartum asphyxia 6 in unselected populations.36,37 The exact pathophysiological
Perinatal vascular injury 4
Post neonatal infection 6
mechanisms for hearing loss remain unclear, but it is likely
Otitis media 4 that hyperbilirubinaemia, hypoxia, infection, and use of
Unknown 50 ototoxic drugs may all have a part to play, either singly or in
Investigative findings, %
Normal 27
combination.37 Although numbers were very small, there was
Large vestibular aqueduct 6 an increase in deafness in extremely preterm infants born in
Large internal auditory meatus 2 Victoria from 0.9% in 1991–1992 to 2.5% in 2005.38
Cochlear calcification 2
No investigations available 63
Another explanation for a possible rise in the frequency of
hearing loss in children with CP may be that changes in the

Hearing Loss in CP Susan M Reid et al. 1043

availability and methods of testing for hearing loss have
resulted in an increase in the number of infants diagnosed,
particularly in those in whom other impairments may have
made testing difficult in the past. In Victoria, the Victorian
Infant Hearing Screening Program has been operating since
1992. However, it is only since 2003 that all infants admitted
to one of the four Victorian neonatal intensive care units have
been automatically screened, in addition to those with other
identified risk factors. Another relevant consideration is that
advances in audiological testing equipment and techniques
have allowed more accurate hearing screening of the newborn,
using either otoacoustic emission screeners or automated audi-
tory brainstem-evoked response audiometry.
In half of the 48 children meeting our definition for hear-
ing loss, we were unable to determine a probable cause,
although 18 of the 24 children in whom no cause could be
identified had a perinatal risk factor such as preterm birth or
neonatal unit admission. In a retrospective study of severe or
profound hearing loss >70dB in children born in three French
administrative departments between 1976 and 1985, the aeti-
ology was also undetermined in more than half of cases. The
aetiologies in the French study included 12% perinatal risk
factors and 21% hereditary causes.39 Our data support the
notion that clinically recognized genetic causes of SNHL are
less common in CP than in the wider population, with only
one child having an identified genetic cause. Factors previ-
ously reported as being associated with SNHL in CP were
also identified in this study. These included intrauterine or
post neonatal infection, severe hypoxic–ischaemic insults,
extreme preterm birth, very low birthweight, neonatal
kernicterus, and neonatal care longer than 7 days. In general,
the MRI data findings were consistent with the presumed
aetiology.
A limitation of the Victorian cohort study was the amount
of missing data on hearing, including the type of loss. Some
testing was done at very young developmental ages when
responses were sometimes unreliable. The frequency of CHL
found in this study does not represent the true incidence of
CHL because CHL tends to be age related. As we based our
classification on most recent or best audiology, many children
had recovered. Others may have recovered, but the lack of
recent audiology made it difficult to confirm this.
CONCLUSION
These population-based data are likely to more accurately
reflect the true incidence of hearing loss in children with CP
compared with previous reviews.
ACKNOWLEDGEMENTS
The authors are grateful to Anna Lanigan and Christine Poulis, who
assisted with data collection for this study, and the Victorian Perinatal
Data Collection Unit, which provided birth data. We also
acknowledge the support of the Victorian Medical Insurance Agency
Ltd, the name behind Professional Services Australia, and the Victo-
rian Department of Health who provided funding for the Victorian
Cerebral Palsy Register. The first author’s doctoral research is sup-
ported by a scholarship (2009–2011) from the Australian National
Health and Medical Research Council.

REFERENCES
1. Rosenbaum P, Paneth N, Leviton A, Goldstein M, Bax M. A
report: the definition and classification of cerebral palsy April
2006. Dev Med Child Neurol 2007; 49: 8–14.
2. Ashwal SM, Russman BSM, Blasco PAM, et al. Practice
parameter: diagnostic assessment of the child with cerebral
palsy: report of the Quality Standards Subcommittee of the
American Academy of Neurology and the Practice Commit-
tee of the Child Neurology Society. Neurology 2004; 62:
851–63.
3. Yoshinaga-Itano C, Sedey AL, Coulter DK, Mehl AL. Lan-
guage of early- and later-identified children with hearing
loss. Pediatrics 1998; 102: 1161–71.
4. Stach BA, Ramachandran VS. Hearing disorders in children.
In: Madell JR, Flexer CA, editors. Paediatric Audiology:
Diagnosis, Technology, and Management. New York: Thi-
eme, 2008: 3–12.
5. Lowe LH, Vezina LG. Sensorineural hearing loss in chil-
dren. Radiographics 1997; 17: 1079–93.
6. Williamson WD, Demmler G, Percy A. Progressive hearing
loss in infants with asymptomatic congenital cytomegalovirus
infection. Pediatrics 1992; 90: 862–6.
7. Sano M, Kaga K, Kitazumi E, Kodama K. Sensorineural
hearing loss in patients with cerebral palsy after asphyxia and
hyperbilirubinemia. Int J Pediatr Otorhinolaryngol 2005; 69:
1211–7.
8. Robertson CMT, Watt M-J, Dinu IA. Outcomes for the
extremely premature infant: what is new and where are we
going? Pediatr Neurol 2009; 40: 189–96.
9. Fligor B, Neault M, Mullen C. Factors associated with senso-
rineural hearing loss among survivors of extracorporeal mem-
brane oxygenation therapy. Pediatrics 2005; 115: 1519–28.
10. Kaga K, Kitazumi E, Kodama K. Auditory brainstem
responses of kernicterus infants. Int J Pediatr Otorhinolaryn-
gol 1979; 1: 255–64.
11. Davis A, Wood S. The epidemiology of childhood hearing
impairment: factors relevant to planning of services. Br J Au-
diol 1992; 26: 77–90.
12. van Straaten HLM, Hille ETM, Kok JH, Verkerk PH;
Dutch NICU Neonatal Hearing Screening Working Group.
Implementation of a nation-wide automated auditory brain-
stem response hearing screening programme in neonatal
intensive care units. Acta Paediatr 2003; 92: 332–8.
13. Goodman A. Reference zero levels for pure-tone audiome-
ters. ASHA 1965; 7: 262–3.
14. Pellegrino L. Cerebral palsy. In: Batshaw ML, Pellegrino L,
Roizen NJ, editors. Children with Disabilities. Baltimore:
Paul H. Brookes Publishing Co. Inc, 2007: 443–66.
15. Kolawole T, Patel P, Mahdi A. Computed tomographic
(CT) scans in cerebral palsy. Pediatr Radiol 1989; 20: 23–7.
16. Murphy C, Yeargin-Allsopp M, Decoufle P. Prevalence of
cerebral palsy among ten-year-old children in metropolitan
Atlanta, 1985 through 1987. J Pediatr 1993; 123: S13–20.
17. von Wendt L, Rantakallio P, Saukkonen A. Cerebral palsy
and additional handicaps in a 1-year birth cohort from north-
ern Finland-a prospective follow-up study to the age of 14
years. Ann Clin Res 1985; 17: 156–61.
18. Zafeiriou D, Kontopoulos E, Tsikoulas I. Characteristics and
prognosis of epilepsy in children with cerebral palsy. J Child
Neurol 1999; 14: 289–94.
19. Bax M. Diagnostic assessment of children with cerebral palsy.
Lancet Neurol 2004; 3: 395.
20. Stanley F, Blair E, Alberman E. Cerebral Palsies: Epidemiol-
ogy and Causal Pathways. Clinics in Developmental Medi-
cine No. 151. London: Mac Keith Press, 2000.
21. Fletcher H. A method for calculating hearing loss for speech
from an audiogram. J Acoust Soc Am 1950; 22: 1–5.
22. Evans P, Elliott M, Alberman E, Evans S. Prevalence and dis-
abilities in 4 to 8 year olds with cerebral palsy. Arch Dis Child
1985; 60: 940–5.
23. Hutton JL, Pharoah PO. Effects of cognitive, motor, and
sensory disabilities on survival in cerebral palsy. Arch Dis
Child 2002; 86: 84–9.
24. Mongan D, Dunne K, O’Nuallain S, Gaffney G. Prevalence
of cerebral palsy in the West of Ireland 1990–1999. Dev Med
Child Neurol 2006; 48: 892–5.
25. Parkes J, Dolk H, Hill N. Children and Young People
with Cerebral Palsy in Northern Ireland – Birth Years
1977–1997. A Comprehensive Report from the Northern
Ireland Cerebral Palsy Register. Belfast: Queen’s University,
2005.
26. Peek A, van Essen P, Gibson C, Scott H, Chan A, Haan E.
2007 Annual Report of The South Australian Cerebral Palsy
Register. Adelaide: Children, Youth and Women’s Health
Services, 2008.

1044 Developmental Medicine & Child Neurology 2011, 53: 1038–1045

27. Reid SM, Lanigan A, Walstab JE, Reddihough DS. Third
Report of the Victorian Cerebral Palsy Register. Melbourne:
Murdoch Childrens Research Institute, 2005.
28. Surman G, Bonellie S, Chalmers J, et al. UKCP: a collabora-
tive network of cerebral palsy registers in the United King-
dom. J Public Health (Oxf) 2006; 28: 148–56.
29. Watson L, Blair E, Stanley F. Report of the Western Austra-
lian Cerebral Palsy Register. West Perth: Telethon Institute
for Child Health Research, 2006.
30. Andersen GL, Irgens LM, Haagaas I, Skranes JS, Meberg
AE, Vik T. Cerebral palsy in Norway: prevalence, subtypes
and severity. Eur J Paediatr Neurol 2008; 12: 4–13.
31. Riikonen R, Raumavirta S, Sinivuori E, Seppala T. Changing
pattern of cerebral palsy in the southwest region of Finland.
Acta Paediatr Scand 1989; 78: 581–7.
32. Shevell MI, Dagenais L, Hall N. Comorbidities in cerebral
palsy and their relationship to neurological subtype and
GMFCS level. Neurology 2009; 72: 290–6.
33. Sigurdardottir S, Thorkelsson T, Halldorsdottir M, Thora-
rensen O, Vik T. Trends in prevalence and characteristics of
cerebral palsy among Icelandic children born 1990 to 2003.
Dev Med Child Neurol 2009; 51: 356–63.
34. Surman G, Newdick H, King A, Gallagher M, Kurinczuk JJ.
4Child: Four Counties Database of Cerebral Palsy, Vision
Loss and Hearing Loss in Children. Annual Report 2008,
including Data for Births 1984 to 2002. Oxford: National
Perinatal Epidemiology Unit, 2008.
35. World Health Organization. Table of grades of hearing
impairment. Available from http://apps.who.int/pbd/pdh/
Docs/GRADESTable-DEFs.pdf (accessed 8 April 2011).
36. Doyle LW, Anderson PJ. Improved neurosensory outcome
at 8 years of age of extremely low birthweight children born
in Victoria over three distinct eras. Arch Dis Child Fetal Neo-
natal Ed 2005; 90: 484–8.
37. Marlow ES, Hunt LP, Marlow N. Sensorineural loss and
prematurity. Arch Dis Child Fetal Neonatal Ed 2000; 82:
141–4.
38. Doyle LW, Roberts G, Anderson PJ; Victorian Infant Col-
laborative Study Group. Outcomes at age 2 years of infants
<28 weeks’ gestational age born in Victoria in 2005. J Pediatr
2010; 156: 49–53.
39. Baille MF, Arnaud C, Cans C, Grandjean H, du Mazaubrun
C, Rumeau-Rouquette C. Prevalence, aetiology, and care of
severe and profound hearing loss. Arch Dis Child 1996; 75:
129–32.
Hearing Loss in CP Susan M Reid et al. 1045