Auditory neuropathy: unexpectedly common in a screened newborn population

Transcription

1 DEVELOPMENTAL MEDICINE & CHILD NEUROLOGY ORIGINAL ARTICLE Auditory neuropathy: unexpectedly common in a screened newborn population ANDREW C DOWLEY FRCS ORL HNS ENT 1 WILLIAM P WHITEHOUSE FRCP FRCPCH 2 STEVE M MASON PHD 3 YVONNE COPE MSC 4 JUDITH GRANT MRCP(UK) FRCPCH 5 KEVIN P GIBBIN FRCS 6 1 South Trent Deanery, Leicestershire, UK. 2 School of Human Development, University of Nottingham, UK. 3 Queens Medical Centre, Nottingham, UK. 4 Hearing Services, Nottingham University Hospitals, UK. 5 Department of Paediatrics, Queens Medical Centre, Nottingham, UK. 6 ENT Department, Queens Medical Centre, Nottingham, UK. Correspondence to Dr William Whitehouse at Queens Medical Centre, Derby Road, Nottingham, NG7 2UH, UK. william.whitehouse@nottingham.ac.uk PUBLICATION DATA Accepted for publication 20th October Published online 24th March Auditory neuropathy, or dyssynchrony, is defined by an abnormal or absent auditory brainstem response but intact otoacoustic emissions or cochlear microphonics. It is associated with impaired hearing on behavioural pure-tone audiometry, absent acoustic reflexes, and poor speech perception, particularly in noisy environments. These results suggest a disorder of inner hair-cell and or eighth-nerve function. We describe a case-note survey of patients with and without auditory neuropathy, using data from the local newborn hearing screening programme collected prospectively from 2002 to During this period, infants were screened with otoacoustic emissions, 30 patients were diagnosed with suspected severe to profound hearing loss (16 males, 14 females), and 12 of those 30 had auditory neuropathy (six males, six females). Mean gestational age was 33 weeks 1 day in the auditory neuropathy group and 35 weeks in the non-auditory neuropathy group. The most significant risk factors for auditory neuropathy were hyperbilirubinaemia (p=0.018), sepsis (p=0.024), and gentamicin exposure (p=0.024). Children with auditory neuropathy comprise a subgroup of patients with hearing impairment involving different pathologies most commonly associated with the risk factors related to admission to neonatal intensive care units. Improvement is possible with maturity, at least in a minority. Auditory neuropathy as an entity was first suspected in the 1980s, when patients with normal pure-tone audiograms presented with difficulty in sound perception, particularly in noisy environments. With the advent of newborn hearing screening tests, auditory neuropathy has become recognized in the paediatric population. The incidence of auditory neuropathy in patients with profound hearing loss is estimated at 10%, 1,2 with a prevalence in children at risk of 0.23%. 2 The definition of auditory neuropathy implies normally functioning cochlear outer hair cells but abnormal responses in the pathway from the cochlea to the brainstem. Thus, these patients have normal cochlear microphonics and otoacoustic emissions but abnormal or absent auditory brainstem responses. The behavioural pure-tone audiogram can vary widely, from normal hearing to profound hearing loss. An inverse slope (poor thresholds in lower frequencies, better in higher) is a common finding. The acoustic reflexes (e.g. auropalpebral reflex) are absent, and speech perception is poorer than would be predicted from the behavioural pure-tone thresholds, particularly in noisy environments. The underlying pathology varies between patients, giving different clinical pictures, and may be similar to other peripheral neuropathies in some. The site of the pathology is important (receptor, synapse, or auditory nerve itself; Fig. 1). Inner hair cells are particularly sensitive to hypoxia, compared with the outer hair cells, 3 andalsotosometoxic insults such as carboplatin. 4 Synaptic disorders can cause a rate-dependent problem with saturation of the response; for example, a stimulus given 3 or 11 times a second is detected perfectly but is inadequately detected at 20 times ª The Authors. Journal compilation ª Mac Keith Press DOI: /j x

2 Inner hair cell Tectorial membrane Spiral ganglion Cochlear nerve Reissner s membrane Outer hair cells Basilar membrane Scala vestibuli Scala tympani Scala media Figure 1: Section through the cochlea, expanded to show the organ of Corti. a second. 5 Disorders of the auditory nerve can arise from either demyelination, which causes slowing of the compound action potential, current blockage, and ephaptic transmission (current leakage to adjacent neurons), or primary axonal disease with loss of fibres and small compound action potentials. These two disorders would most likely affect the action potentials from the longest fibres (as is the case in peripheral neuropathies) because of degeneration along the length and, as these fibres supply the apex of the cochlea, would be expected to cause most impairment at lower frequencies. This is consistent with the observations described above. In the UK, otoacoustic emission testing is the only test routinely performed on the neonatal population, unless there are significant risk factors for profound hearing loss, such as a family history or admission to neonatal intensive care, when auditory brainstem response is also performed. This is the only group in the UK population undergoing both otoacoustic emission and auditory brainstem response testing, so it is the predominant group in which auditory neuropathy is currently being diagnosed. We describe a case-note survey of patients with auditory neuropathy, two of whom have been described in a previous report, 6 and discuss the risk factors involved, compared with those for patients diagnosed with severe to profound hearing loss without auditory neuropathy. METHOD Audiological data from the local newborn hearing screening programme from its initiation in 2002 to 2007 was collected prospectively, as part of a service evaluation. The medical notes of children identified with severe to profound hearing impairment (thresholds 60dB) were then reviewed. This non-intervention evaluation was registered as a clinical audit with the Nottingham University Hospitals NHS Trust and as such did not require ethical approval. Audiological information collected included age at diagnosis (defined as the date that the parents or guardians were informed of the diagnosis), age at which hearing aids were provided, and age at cochlear implantation, if suitable. The latest behavioural audiogram (visual reinforcement audiogram, either sound field or ear specific) was also reviewed if the child was developmentally able to undertake the assessment. The information collected from medical physics included the results of otoacoustic emissions and auditory brainstem responses. We were particularly interested in whether there had been any sign of maturation or evidence of a shift in thresholds with time. This is revealed by an improvement in the auditory brainstem response compared with previous tests, as these children are retested at 3, 9, and 12 months of age. 7 The medical notes review included detailed antenatal, birth, and neonatal histories, including any information on potential risk factors such as gestation, birthweight, neonatal intensive care unit admission, hypoxia, jaundice, or sepsis. For simplicity, all ages are actual rather than corrected for gestation. Patients with auditory neuropathy were compared with those with hearing impairment without auditory neuropathy, and possible causative factors were identified. Simple descriptive statistics are given, using Fisher s exact test for nominal values and paired Student s t-test for interval data. RESULTS During the study, infants were screened with otoacoustic emissions, 30 patients were diagnosed with suspected severe to profound hearing loss, and 12 of these 30 had auditory neuropathy. This gives an annual incidence of for cases of severe to profound hearing loss and for auditory neuropathy. The mean followup was 2 years 10 months (range 10mo 5y 1mo). Table I shows comparison of the two groups. The two groups were closely comparable in terms of gestation, numbers reaching term, and birthweight. Both groups included some very unwell infants, but this was particularly evident in the auditory neuropathy group. As shown in Table I, all of the measures of severity of illness were significantly worse for the auditory neuropathy group than for the non-auditory neuropathy group. This would be expected as the non-auditory neuropathy group included those with non-syndromic hearing loss who were often well, and, before the neonatal screening programme, would not have been diagnosed as having hearing problems until later in life. All of the infants with auditory neuropathy were admitted to neonatal intensive care units; this suggests that neonatal illness, especially associated with preterm delivery (e.g. respiratory failure requiring mechanical ventilation, Auditory Neuropathy in a Screened Newborn Population Andrew C Dowley et al. 643

3 Table I: Clinical features of infants with and without auditory neuropathy Clinical feature Auditory neuropathy Non-auditory neuropathy Significance Number of patients Males females Severe hearing impairment 2 4 Profound hearing impairment Admission to neonatal intensive care unit p=0.001 Gestation Mean (SD) 33wks 1d (32d) 35wks (43d) a, p=0.748 Range 25wks 2d 40wks 23wks 0d 40wks Born at term p=0.722 Birthweight, mean (SD) g 1863 (990) 1976 (1060) a, p=0.774 Mechanical ventilation Patients a, p=0.397 Duration, mean (SD) d 9.2 (9.1) 34.3 (65.8) Hypoxia Hyperbilirubinaemia p=0.018 Intracranial bleeding p=0.128 Sepsis p=0.024 Gentamicin exposure p=0.024 Cerebral palsy p=0.548 a Student s t test. hypoxia, sepsis, hyperbilirubinaemia, and intracranial haemorrhage) plays a causative role in auditory neuropathy, even though the sample was biased towards those with neonatal risk factors. This is also shown by the fact that the most significant risk factors for auditory neuropathy in this group were hyperbilirubinaemia (p=0.018), sepsis (p=0.024), and gentamicin exposure (p=0.024). Of the infants who were ventilated, those with non-auditory neuropathy were ventilated on average for longer; however, this group included one outlier with 6 months total ventilation. If this infant is discounted, then the mean period of ventilation for the non-auditory neuropathy group was 10.3 days, which was not significantly different from the auditory neuropathy group (Table I). Other previously reported risk factors such as hypoxia and cerebral palsy (CP) were not more frequent in the auditory neuropathy group than in the non-auditory neuropathy group in our series of at-risk neonates. The mean time to diagnosis was 3 months 25 days for the auditory neuropathy group (range 5d 9mo 26d) and 3 months 15 days for the non-auditory neuropathy group (range 28d 8mo 25d). All patients received hearing aids, and the mean time to first use of a hearing aid was 4 months 27 days for the auditory neuropathy group (range 1mo 28d 8mo 7d) and 6 months 1 day for the non-auditory neuropathy group (range 1mo 9d 1y 11mo 12d). Three patients showed signs of maturation, with improvement in the auditory brainstem response thresholds, from either no response at all or response to only high volumes, to a response detectable at lower volumes. One child in the non-auditory neuropathy group, born at 29 weeks gestation, who had chronic lung disease, sepsis, and gentamicin exposure, had evidence of hearing maturation at 7 months of age. Two children with auditory neuropathy had evidence of hearing maturation: one at 10 months, who had been born at 37 weeks gestation and developed kernicterus and CP; the other at 7 months, who had been born at 29 weeks gestation and developed hypoxia, intraventricular haemorrhage, sepsis, and chronic lung disease. Currently seven of the children with non-auditory neuropathy and just one of the auditory neuropathy group have undergone cochlear implantation, at a mean age of 1 year 7 months (range 10mo 2y 8mo). These figures may change as more of the younger children complete their assessments. DISCUSSION Risk factors for hearing impairment have been well documented. Vohr et al. 8 prospectively reviewed 4478 high-risk infants on neonatal intensive care units and 2701 on wellinfant units (2348 infants with no risk factors and 353 with risk factors). In the neonatal intensive care unit population, the four most common risk factors were ototoxic medications (44.4%), very low birthweight (17.8%), assisted ventilation for longer than 5 days (16.4%), and low Apgar 644 Developmental Medicine & Child Neurology 2009, 51;

4 scores at 1 or 5 minutes (13.9%). The six most common risk factors in the well-infant nurseries were family history (6.6%), craniofacial abnormalities (3.4%), low Apgar scores (2.8%), syndromes or stigmata associated with hearing loss (0.5%), ototoxic medications (0.2%), and congenital infection (0.1%). Patients with auditory neuropathy comprise a subgroup of patients with hearing impairment, involving different pathologies most commonly associated with the risk factors related to admission to neonatal intensive care units. Although patients may present with none of the known risk factors, in our series of auditory neuropathy patients, the most significant risk factors appear to be hyperbilirubinaemia, intracranial haemorrhage, sepsis, and gentamicin treatment. These have been identified in other clinical series of auditory neuropathy with rates of hyperbilirubinaemia of 50 to 73%, preterm birth of 30 to 46%, ototoxic drug exposure of 41 to 80%, family history of hearing loss of 36 to 38%, mechanical ventilation of 36%, and CP of 9 to 15%. 2,9,10 Not all studies have elicited these factors. Berg et al 11 studied patients admitted to a paediatric intensive care unit, looking at sex, gestational age, birthweight, respiratory distress, hyperbilirubinaemia, use of potentially ototoxic medications, multiple births, seizures, and family history. A comparison of children with and without auditory neuropathy did not show a significant difference in exposure or presence of risk factors, apart from hyperbilirubinaemia, vancomycin, and furosemide. Although neonates with auditory neuropathy have been found to have significantly higher levels of neuron-specific enolase (an indicator of neuronal damage) than those without auditory neuropathy, no correlation has been found between neuron-specific enolase levels and serum bilirubin levels. 12 On histopathological studies, loss of inner hair cells have consistently been found with hyperbilirubinaemia 9 and ototoxic medication. 13,14 Hyperbilirubinaemia has been found to cause severe degeneration of spiral ganglion neurons and a paucity of myelinated axons in a jaundiced Gunn rat model compared with a non-jaundiced model. 15 In many cases, the times to diagnosis and first use of a hearing aid are lengthy because some of the children were too unwell or immature to be tested or have hearing aids fitted. Because of uncertainties surrounding the prognosis of auditory neuropathy, the main issues in the management of children with this diagnosis are the provision of information and support for the family, repeated audiological assessment (both objective [electrophysiological] and behavioural), assessment of the child s communication, and decisions about provision of amplification (in terms of a conventional hearing aid or cochlear implant), and other medical management. These requirements are complex and cannot be provided by one or two individuals. Therefore a multidisciplinary team of professionals including medical physicists, teachers of the deaf, audiologists, audiovestibular physicians, speech and language therapists, community developmental paediatricians, paediatric neurologists, otolaryngologists, and the family themselves should undertake the diagnosis and treatment of these children. Many children with auditory neuropathy are able to make good use of their hearing and benefit from hearingaid fitting, if there is sufficient behavioural hearing loss to warrant provision of amplification. 2 However, in some cases, although there may be an improvement in detection and awareness of sound and speech, the long-term benefit of hearing aids in terms of understanding speech may be very poor and not comparable to the benefit achieved in a child with a non-auditory neuropathy sensorineural hearing loss of a similar degree. 2,16 After the diagnosis of auditory neuropathy it is usual to undertake a trial with hearing aids where there are reliable and raised behavioural hearing thresholds. As the electrophysiological auditory brainstem response thresholds may differ significantly from the behavioural hearing thresholds, the hearing aid prescription should be based on behavioural audiological assessment results. 2 In cases where the behavioural assessment is difficult or inconclusive, subjective observations by the family and professionals of the child s response to use of a hearing aid are used. Surprisingly, the literature has shown positive outcomes in terms of detection and communication for some children with a diagnosis of auditory neuropathy who receive a cochlear implant. 16,18,19 This implies functional integrity of the eighth nerve proximal (brain-wards) to the implant in these cases. However, the results are variable. 20,21 Concerns about implantation include the potential of maturation with improvement and the pathology being more proximal, therefore rendering stimulation via the implant useless. It is, therefore, usual to offer a trial with conventional hearing-aid amplification before consideration of cochlear implantation. The mode of communication that the child and family use should be an informed choice by the family themselves. However, visual inputs such as total communication, signed or cued speech are effective options for children with a diagnosis of auditory neuropathy. 22 CONCLUSION Our understanding of auditory neuropathy is growing as we start to recognize the risk factors and pathology Auditory Neuropathy in a Screened Newborn Population Andrew C Dowley et al. 645

5 involved, more patients are diagnosed, and treatment options are developed. Although it is an uncommon problem, it may be underdiagnosed because of lack of familiarity on the part of the professionals and the way that we currently screen newborn infants, although universal screening with auditory brainstem response is not a viable option. In this population, auditory neuropathy is much more common than hitherto expected, and the infants generally have had more severe neonatal illness than those with nonauditory neuropathy hearing impairments. However, there is a real chance of improvement over time as the child matures, at least in some cases, and further systematic follow-up is needed to quantify this. REFERENCES 1. Kraus N, Ozdamar O, Stein L, Reed N. Absent auditory brain stem response: peripheral hearing loss or brain stem dysfunction? Laryngoscope 1984; 94: Rance G, Beer DE, Cone-Wesson B, et al. Clinical findings for a group of infants and young children with auditory neuropathy. Ear Hear 1999; 20: Shirane M, Harrison RV. The effects of hypoxia on sensory cells of the cochlea in chinchilla. Scanning Microsc 1987; 1: Takeno S, Harrison RV, Mount RJ, Wake M, Harada Y. Induction of selective inner hair cell damage by carboplatin. Scanning Microsc 1994; 8: Starr A, Picton TW, Kim R. Pathophysiology of auditory neuropathy. In: Sininger Y, Starr A, editors. Auditory neuropathy: a new perspective on hearing disorders. Canada: Singular, Thomson Learning Inc, 2001: Dunkley C, Farnsworth A, Mason S, Dodd M, Gibbin K. Screening and follow up assessment in three cases of auditory neuropathy. Arch Dis Child 2003; 88: Wood S. NHS newborn hearing screening programme: auditory neuropathy auditory dys-synchrony timeline of investigation and management. nhs.uk/getdata.php?id=147 (accessed 11 November 2008). 8. Vohr BR, Widen JE, Cone-Wesson B, et al. Identification of neonatal hearing impairment: characteristics of infants in the neonatal intensive care unit and wellbaby nursery. Ear Hear 2000; 21: Amatuzzi MG, Northrop C, Liberman MC, et al. Selective inner hair cell loss in premature infants and cochlea pathological patterns from neonatal intensive care unit autopsies. Arch Otolaryngol Head Neck Surg 2001; 127: Madden C, Rutter M, Hilbert L, Greinwald JH Jr, Choo DI. Clinical and audiological features in auditory neuropathy. Arch Otolaryngol Head Neck Surg 2002; 128: Berg AL, Spitzer JB, Towers HM, Bartosiewicz C, Diamond BE. Newborn hearing screening in the NICU: profile of failed auditory brainstem response passed otoacoustic emission. Pediatrics 2005; 116: Akman I, Ozek E, Kulekci S, Türkdogan D, Cebeci D, Akdas F. Auditory neuropathy in hyperbilirubinemia: is there a correlation between serum bilirubin, neuronspecific enolase levels and auditory neuropathy? In J Audiol 2004; 43: Harrison RV. An animal model of auditory neuropathy. Ear Hear 1998; 19: Salvi RJ, Wang J, Ding D, Stecker N, Arnold S. Auditory deprivation of the central auditory system resulting from selective inner hair cell loss: animal model of auditory neuropathy. Scand Audiol Suppl 1999; 51: Shaia WT, Shapiro WM, Spencer RF. The jaundiced Gunn rat model of auditory neuropathy dyssynchrony. Laryngoscope 2005; 115: Rance G, Cone-Wesson B, Wunderlich J, Dowell R. Speech perception and cortical event related potentials in children with auditory neuropathy. Ear Hear 2002; 23: Shallop JK, Peterson A, Facer GW, Fabry LB, Driscoll CLW. Cochlear implants in five cases of auditory neuropathy: prospective findings and progress. Laryngoscope 2001; 111: Berlin CI, Li L, Hood LJ, Morlet T, Rose K, Brashears S. Auditory neuropathy dys-synchrony: after the diagnosis, then what? Semin Hear 2002; 23: Trautwein PG, Sininger YS, Nelson R. Cochlear implantation of auditory neuropathy. J Am Acad Audiol 2000; 11: Miyamoto RT, Kirk KI, Renshaw J, Hussain D. Cochlear implantation in auditory neuropathy. Laryngoscope 1999; 109: Buss E, Labadie RF, Brown CJ, Gross AJ, Grose JH, Pillsbury HC. Outcome of cochlear implantation in pediatric auditory neuropathy. Otol Neurotol 2002; 23: Sutton G, Gravel J, Hood L, et al. Assessment and management of auditory neuropathy auditory dys-synchrony. A recommended protocol. Version 1.1 May (accessed 11 November 2008). 646 Developmental Medicine & Child Neurology 2009, 51;