 |
|
 |
|
 |
 |
|
 |
 |
 |
 |
 |
 |
 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
PEDIATRICS Vol. 104 No. 4 October 1999, pp. 900-904
,
,,,
From the * Children's Hospital and Department of
Otolaryngology, Medical University, Lübeck, Germany; the
§ Vestische Kinderklinik Datteln, University Witten, Herdecke, Germany;
the Children's Hospital Friedrichshain, Berlin, Germany; the
¶ Children's Hospital University, Bonn, Germany; the # Children's
Hospital and the ** Department of Phoniatrics and Pedaudiology,
University Münster, Münster, Germany.
 |
ABSTRACT |
|---|
Top
Abstract
Methods
Results
Discussion
Conclusion
References
|
|---|
Objective. To determine the incidence and risk factors for hearing disorders in a selected group of neonates and the feasibility of selective hearing screening.
Settings. Multicenter prospective trial at five centers in Germany.
Methods. Enrollment criteria: in addition to previously defined risk factors by the Joint Committee on Infant Hearing (family history of hearing loss, in utero infections, craniofacial anomalies, birth weight <1500 g, critical hyperbilirubinemia, ototoxic medications, bacterial meningitis, postnatal asphyxia, mechanical ventilation >5 days, stigmata, or syndromes associated with hearing loss), the impact of maternal drug abuse, birth weight <10th percentile, persistent pulmonary hypertension, and intracranial hemorrhage more than or equal to grade III or periventricular leukomalacia on infant hearing were evaluated. The screening procedure was performed by automated auditory brainstem response (A-ABR; ALGO 1-plus; Natus Med Inc, San Carlos, CA). Statistics: univariate analyses of risk factors versus A-ABR results and a multivariate regression analysis were used; additionally, the total test time was recorded.
Results. Seven hundred seventy recordings from 777 infants enrolled consecutively constitute the basis of this analysis. Mean gestational age was 33.8 ± 4.3 weeks, birth weight 2141 ± 968 g; 431 infants being male and 339 female; 41 (5.3%) infants exhibited pathologic A-ABR results (16 bilateral and 25 unilateral). Meningitis or sepsis, craniofacial malformations, and familial hearing loss were independent significant risk factors. Median total test time was 25 minutes. Follow-up examinations in 31 infants revealed persistent hearing loss in 18 infants (13 infants sensorineural, 5 from mixed disorders), 7 requiring amplification.
Conclusion. Hearing screening in high-risk neonates revealed a total of 5% of infants with pathologic A-ABR (bilateral 2%). Significant risk factors were familial hearing loss, bacterial infections, and craniofacial abnormalities. Other perinatal complications did not significantly influence screening results indicating improved perinatal handling in a neonatal population at risk for hearing disorders. Key words: hearing screening, neonates, craniofacial malformations, sepsis, familial hearing loss, follow-up.
One to 2 of 1000 newborns suffer from congenital or
perinatally acquired hearing disorders.1-3 The prevalence of neonatal hearing disorders has been reported to be increased 10- to
50-fold in infants at risk.3-9 Moderate to severe
bilateral hearing loss (ie, >40 dB) distorts the developing child's
perception of his or her attempt at speech production. If this type of
hearing disorder remains undetected through the critical period of
language acquisition within the first year of life, a profound
impairment of receptive and expressive speech and language development
will result. This in turn will lead to a decreased acquisition of
expected milestones.10 For acquisition of hearing, a
sensitive period up to 6 months of age has been
demonstrated.11 Accordingly, introduction of hearing aids
within this period will improve subsequent hearing development among
hearing-impaired children.12 To ensure timely therapy, a
reasonable goal is to establish the diagnosis of severe neonatal
hearing impairment before the age of 6 months.
In addition to hereditary hearing loss, a number of in utero and
neonatal complications (eg, infections, immaturity, asphyxia, ototoxic
medications, and hyperbilirubinemia) have been described as risk
factors of neonatal hearing disorders.11 This resulted in
the recommendations of the Joint Committee on Infant Hearing to screen
all infants with risk factors within the neonatal period. This
selective screening was considered a first step toward the introduction
of universal hearing screening.13 Two methods for neonatal
hearing screening have been developed during the past 2 decades: the
registration of a modified auditory brainstem response (ABR) first
described by Davis14 in 1976 and the registration of
otoacoustic emissions (OAE) introduced by Kemp15 in 1978. In contrast to ABR, the use of OAE does not identify postcochlear
hearing disorders. Both methods further were adapted for routine
screening by an automated auditory brainstem response (A-ABR) or
corresponding OAE techniques.
The objective of the present study was to obtain actual data on the
incidence and risk factors of neonatal hearing disorders in light of
improved outcome of high-risk neonates.16,17 We thus
performed a hearing screening in all infants with risk factors for
hearing disorders,6-818-20 in a prospective multicenter
clinical trial, to identify the incidence of moderate to severe
neonatal hearing disorders. We added further risk factors to those
proposed by the Joint Committee,13 which were demonstrated
specifically to affect the auditory system: intracranial hemorrhage
more than or equal to grade III or periventricular leukomalacia,6,21 maternal drug abuse,22 intrauterine growth restriction leading to small-for-gestational age
infants,23 and persistent pulmonary hypertension of the
newborn (PPHN).24 In addition, the feasibility of a
neonatal screening for hearing disorders and the time necessary for
screening procedures were investigated.
The study was performed from October 1995 through November 1997 at 5 pediatric hospitals in the Federal Republic of Germany (University
Children's Hospitals Bonn, Münster, and Lübeck; Vestische
Kinderklinik Datteln, University Witten-Herdecke; and Pediatric
Hospital Berlin-Friedrichshain).
Inclusion Criteria
All infants were enrolled in the trial, if informed parental
consent was obtained and at least one of the risk factors according to
the Joint Committee on Infant Hearing was diagnosed:
METHODS
Top
Abstract
Methods
Results
Discussion
Conclusion
References
5 at 1 minute or 6 at 5 minutes)
Definition of additional variables: intracranial hemorrhage grade III to IV was defined according to Papile et al26 and periventricular leukomalacia according to Hill et al27; small-for-gestational age infants were classified according to the percentiles worked out recently by Voigt et al28 for German newborns. PPHN was diagnosed according to Fox and Duara.29 Maternal drug abuse was assumed based on prenatal history and confirmed by analyses of urine and/or meconium postnatally including alcohol, cocaine, opiates, and derivatives. Infants who required neonatal intensive care were tested before discharge from hospital, with all other newborns tested between days 2 to 7 after birth.
This procedure resulted in a final sample size of 777 infants.
Materials and Methods
An A-ABR, the ALGO-1-Plus (Natus Med Inc, San Carlos, CA) with a click stimulus of 100-ms duration, an intensity of 35 dB nHL, and an alternating polarity with an acoustic frequency spectrum of 700 to 5000 Hz (±5 dB), was used. A built-in artifact rejection for either myogenic, electric, or environmental noise interference ensured that data collection was halted given unfavorable testing conditions. The automated screener provided a pass-fail report.30 All infants with initially abnormal A-ABR screening results were reexamined in the hospitals with A-ABR, ie, a two-stage screening was scheduled in the study protocol.
In 3 out of 5 hospitals additional investigations using OAE (ILO 88, Hatfield, UK31) were performed after having obtained A-ABR. The transient click-evoked OAE are produced by an active process related to outer hear-cell activity in the cochlea. These OAE were rated by an experienced audiologist in consideration of the stability of the stimulus and its reproducibility. When the whole response reproducibility was >60%, a pass was obtained, and testing was halted. All infants with pathologic A-ABR on two occasions further were scheduled for a detailed pediatric audiologic work-up and therapy if indicated.
All data records were reviewed by the study coordinator (L.G.). In case of unclear or incomplete reports, they were sent back to the respective hospital for correction or completion.
Statistics
We analyzed the variables, assumed to be risk factors for
neonatal hearing loss, according to the criteria of the American Academy of Pediatrics13 and additionally the above
outlined variables: intracranial hemorrhage grade III to IV and/or
periventricular leukomalacia, birth weight <10th percentile, PPHN, and
maternal drug abuse. Neonatal characteristics given in Table
1 were not included in the analysis. In a
first step univariate analyses with appropriate methods, the
Mann-Whitney U test and 
2 test,
SPSS-program32 were performed to identify significant risk
factors. The validity of single risk factors was further assessed by
multivariate logistic regression analysis, with inclusion of all
variables with a significance level of P .20 in the
univariate analyses.
|
| |
RESULTS |
|---|
|
Top
Abstract
Methods
Results
Discussion
Conclusion
References
|
|---|
During the study period, a total of 777 consecutive newborns were
enrolled with 770 valid records available. Three hundred and
fifty-eight (42%) had a gestational age <32 completed weeks, 241 (32.1%) had a birth weight <1500 g, and 74 (9.6%) had a birth weight
1000 g. Neonatal characteristics of study infants are given in Table
1. The median number of risk factors per infant was 1 (range, 1-8).
Twenty-three infants had 
5 risk factors with mechanical ventilation
5 days (n = 23), birth weight <1500 g
(n = 22), ototoxic medications (n = 22), and intracranial hemorrhage more than or equal to grade III or
periventricular leukomalacia (n = 12) being most
prevalent. Table 2 gives an overview on
prenatal and neonatal risk factors in study infants. The first A-ABR
testing gave pathologic results in 62 infants, which could be confirmed
at rescreening in 41 infants (5.3%). Sixteen of these infants (2%)
had bilateral pathologic A-ABR, whereas unilateral pathologic A-ABR was
observed in 25 infants (3.3%).
|
On univariate analysis sepsis and/or meningitis, stigmata and/or syndromal disorders, and craniofacial malformations were identified as significant risk factors of A-ABR testing (Table 3). Cleft palate was registered in 6 infants as the most common disorder in the latter group. Meningitis or sepsis, observed in 7 infants with pathologic A-ABR, was caused by five Gram-positive and two Gram-negative germs. Four out of 5 infants with stigmata or syndromes were diagnosed as Down-Syndrome and 1 with fragile X-syndrome by karyotyping. A weight at birth of <1500 g was no specific independent risk factor for pathologic A-ABR in the present study population. Breaking down birth weight categories for infants with 1000 to 1500 g and those with <1000 g also exhibited no association with pathologic A-ABR testing in both subgroups. Chronic lung disease, defined as oxygen dependency (fraction of inspired oxygen >0.21 on day 28), was not found to be associated with pathologic A-ABR testing. Mean (±SD) serum bilirubin concentrations in infants with critical hyperbilirubinemia were 14.6 (±1.7) mg/dL in infants <1500 g birth weight, 16.0 (±2.7) mg/dL in infants of 1500 to 2500 g birth weight, and 23.7 (±1.9) mg/dL in infants >2500 g birth weight.
|
The result of the multivariate regression analysis is given in Table 4.
|
In addition to A-ABR testing, OAE were measured in 464 infants (only possible in three hospitals). Pathologic OAE results were observed in 137 infants (29.5%; unilateral in 61 and bilateral in 76 infants). Comparison of OAE and A-ABR testing is given in Table 5. Assuming A-ABR to be the gold standard, OAE predicted pathologic A-ABR with a sensitivity of 71% and a specificity of 73%.
|
Follow-up in 31 infants with pathologic A-ABR revealed persisting severe hearing loss in 18 infants (13 infants suffered from sensorineural, 5 from mixed hearing disorders). Underlying disorders included craniofacial anomalies (n = 7), familial hearing loss (n = 3), sepsis and/or meningitis (n = 3), very low weight at birth and mechanical ventilation >5 days (n = 2), congenital rubella (n = 1), postnatal asphyxia (n = 1), and Down Syndrome (n = 1). Seven infants required early amplification at 3 to 6 months of age. A transient hearing loss was diagnosed in 10 infants, as follow-up findings were normal. In 3 infants with moderate conductional hearing disorders, no amplification was necessary. Two infants died after discharge from the hospital, 8 infants were lost to follow-up because reexaminations were refused by the parents.
The median time for the A-ABR screening was 25 minutes including 15 minutes for preparation and 10 minutes for registration compared with a total of 8 minutes for measuring OAE.
| |
DISCUSSION |
|---|
|
Top
Abstract
Methods
Results
Discussion
Conclusion
References
|
|---|
Criteria for establishing screening programs for newborns have been published in the past.33 At present, screening in Europe and North America include at least blood tests for congenital hypothyroidism, phenylketonuria, and galactosemia during the first postnatal week (incidence in Europe and in the United States: congenital hypothyroidism ~30 per 100 000; phenylketonuria ~10 per 100 000, galactosemia ~2 per 100 000 live births).34,35 Currently considerations are ongoing to include other tests into routine screening procedures. Some individual states have already added tests for maple syrup disease, homocystinuria, biotinidase deficiency, adrenal hyperplasia, cystic fibrosis, tyrosinemia, hemoglobinopathies, and toxoplasmosis.34 For example, maple syrup disease currently is diagnosed in 1 per 250 000 to 400 000 newborns.35 The prevalence of neonatal hearing disorders is in the range of 1 to 2 per 1000 newborns. As it recently has been demonstrated, language abilities of early-identified children to be superior to late-identified children with identical treatment,36 it seems reasonable to also include hearing screening into routine programs. Thus, screening in a population at risk as performed in the present study can only be regarded to be the first step toward a universal screening.
Hearing disorders in the range of 5 to 60 per 1000 were observed in at-risk study populations enrolling between 117 and 1401 infants.2-9 Most of the studies used inclusion criteria similar to the guidelines of the Joint Committee on Infant Hearing. Our observed incidence of persisting hearing disorder was 18 out of 770 infants (2.3%) and therefore is consistent with the concept of a 10-fold increased risk for neonatal hearing disorders in high-risk groups.
Familial hearing loss, ie, hereditary factors, sepsis and/or meningitis, and craniofacial malformations were identified to be independent significant risk factors for pathologic A-ABR results in the present study population. Prematurity <32 weeks and weight at birth <1500 g did not significantly increase the risk for neonatal hearing disorders. This is somewhat contradictory to the findings of previous studies.6-818-22,37,38 However, this finding may be explained by improved conditions of perinatal care and overall reduced incidence of complications of prematurity.16 Although being major predictors for impaired neurodevelopmental outcome, cerebral complications of prematurity (eg, intraventricular hemorrhage or periventricular leukomalacia) also were not significantly associated with pathologic A-ABR testing. Ototoxic medications and hyperbilirubinemia at concentrations observed in our study infants have been demonstrated to be risk factors for hearing loss by several authors in the past.618-20,37 However, we were unable to confirm those findings. Strict monitoring of aminoglycoside serum concentrations and dose adjustment may have contributed to the avoidance of toxic side effects in study infants. Furthermore comorbidity of infants with hyperbilirubinemia may have influenced the results of the aforementioned studies. In addition, PPHN was not associated with an increased risk for neonatal hearing disorders in the present trial. This finding might be explained by the fact, that hyperventilation was performed less aggressively; ie, targeted hypocarbia was less pronounced during the study period compared with guidelines used in the past.29,39 Our data are further supported by two recently published trials indicating that PPHN and its treatment are not risk factors for hearing disorders.40,41
The high number of infants with either familial hearing disorders, craniofacial malformations, and syndromal disorders, who per se did not require intensive care after birth emphasizes the need, to screen all newborns, not just those treated in neonatal intensive care units. Recently published controlled multicenter trials on universal newborn hearing screening do support this view.42,43
Limitations of the conclusion drawn from the multivariate regression model potentially could result from the preselection of risk factors being the basis of enrollment criteria,13 and their uneven distribution within our study population. The addition of 4 further potential risk factors did not contribute significant associations.
Our study population was primarily screened using A-ABR because this
method had been shown to have a sensitivity of 98% and a specificity
of 96%.44-47 Although not the primary goal of the
present study, our data indicate a superior sensitivity and specificity
of the A-ABR screening in comparison to OAE. The high number of
false-positive results47 and the problem of missing single
infants with severe hearing loss48 further add some
concern about OAE testing in at-risk groups. Psychologic problems
resulting from the high false-positive rate of screening using OAE have
further been discussed.49 At present, OAE measurements
should probably not be used in hearing screening for high-risk
neonates.
| |
CONCLUSION |
|---|
|
Top
Abstract
Methods
Results
Discussion
Conclusion
References
|
|---|
In summary, our data indicate a change in risk factors for neonatal hearing disorders. Craniofacial malformations, familial hearing disorders, and neonatal bacterial infections were significant factors associated with pathologic A-ABR testing. In contrast, very low birth weight and complications of prematurity were not independent risk factors for pathologic screening results in our study population. This change in the risk profile for neonatal hearing impairment in a high-risk population is speculated to be related to changes in perinatal and neonatal care.
| |
ACKNOWLEDGMENTS |
|---|
This study was supported by the "Stiftung für das behinderte Kind," Frankfurt am Main, Germany.
We thank I. Dickau for careful preparation of the manuscript, B. Roenspiess for skilled technical assistance, and H.-J. Friedrich, PhD, for statistical advice.
| |
FOOTNOTES |
|---|
Received for publication Sep 14, 1998; accepted Apr 8, 1999.
Address correspondence to Ludwig Gortner, MD,
Justus-Liebig-University, Pediatric Hospital, Feulgenstra
e 12, D-35392, Giessen, Germany.
| |
ABBREVIATIONS |
|---|
ABR, auditory brainstem response; OAE, otoacoustic emissions; A-ABR, automated auditory brainstem response; PPHN, persistent pulmonary hypertension of the newborn.
| |
REFERENCES |
|---|
|
Top
Abstract
Methods
Results
Discussion
Conclusion
References
|
|---|
past, present and future.
Eur J Pediatr
1994;
153:S2-S8 [Medline]This article has been cited by other articles:
 |
|
 |
|
  D Ari-Even Roth, M Hildesheimer, A Maayan-Metzger, C Muchnik, A Hamburger, R Mazkeret, and J Kuint Low prevalence of hearing impairment among very low birthweight infants as detected by universal neonatal hearing screening Arch. Dis. Child. Fetal Neonatal Ed., July 1, 2006; 91(4): F257 - F262. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Keren, M. Helfand, C. Homer, H. McPhillips, and T. A. Lieu Projected Cost-Effectiveness of Statewide Universal Newborn Hearing Screening Pediatrics, November 1, 2002; 110(5): 855 - 864. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Nafstad, S. O. Samuelsen, L. M. Irgens, and T. Bjerkedal Birth Weight and Hearing Impairment in Norwegians Born From 1967 to 1993 Pediatrics, September 1, 2002; 110(3): e30 - 30. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
||
|
||
Home | My Pediatrics | Journal Information | Current Issue | Past Issues | Subscriptions & Services | Contact Us Click here for faster international access © 1999 American Academy of Pediatrics. All rights reserved. |
||
|
||
