Acta Otorrinolaringológica Gallega

Artículo Original

Hearing loss: are children with
risk indicators actually at risk?

Surdez em idade pediátrica: os indicadores
de risco conferem um risco acrescido?

Joana Silva, José Peneda, Joana Ferreira,

Sandra Alves, Fernanda Castro,

Aida Sousa, Joana Dias, Artur Condé.

Department of Otolaryngology,

Centro Hospitalar Vila Nova de Gaia/Espinho,

Porto, Portugal

Corresponding Author:
Joana Silva.

Department of Otolaryngology,

Centro Hospitalar Vila Nova de Gaia/Espinho,

Porto, Portugal

E-mail: joanavilela.silva@gmail.com

Fecha de envío: 7/8/2020

Fecha de aceptación: 29/1/2020

ISSN:
2340-3438

Edita:
Sociedad Gallega de Otorrinolaringología

Periodicidad:
continuada.

Web:
www.sgorl.org/ACTA

Correo electrónico:
actaorlgallega@gmail.com

Abstract

Introduction: Despite the broad use of risk indicators in hearing loss management, their association with sensorineural hearing (SNHL) loss remains debated.

The aim of this study was to assess the role of risk indicators among infants referred for targeted surveillance program and estimate their association with SNHL.

Material and Methods: We conducted a retrospective study. Infants referred for Otolaryngology assessment between January 2014 and December 2017 were included, using a consecutive sampling procedure.

We reviewed all clinical records and assessed the risk of hearing loss.

Statistical analysis was conducted on IBM SPSS version 24. Statistical significance was considered for p values below 0.05.

Results: The sample included 326 infants, 176 males. Median age of the first evaluation was 2 months. Targeted surveillance infants included 38 cases with a REFER result in TOAEs and 288 infants with risk indicators. Hearing loss was present in 21 infants. There were no differences between the hearing loss group and the normal hearing group regarding the most common risk indicators: ototoxic exposure (38% vs 51%), prematurity (28.6% vs 27.5%), and low birth weight (28.6% vs 22%). Whereas almost half of the patients in the hearing loss group had no identified risk factors, multiple factors were observed in 29%.

Conclusion: Our study emphasizes the importance of universal hearing screening as we observed a weak association between isolated risk indicators and hearing loss. The presence of multiple risk factors seems to be more important. Further studies would be necessary to assess its clinical value.

Keywords

hearing loss, risk indicators, hearing screening

Resumen

Introdução: Apesar da utilização generalizada dos indicadores de risco na avaliação auditiva, a contribuição individual destes fatores não é consensual.

O objetivo deste trabalho foi avaliar a contribuição dos indicadores de risco na referenciação à consulta de Otorrinolaringologia, e estimar a sua associação com surdez neurossensorial.

Material e Métodos: Realizámos um estudo retrospetivo. Foram incluídos, através de uma amostragem consecutiva, recém-nascidos referenciados à consulta de Otorrinolaringologia para rastreio auditivo, entre Janeiro de 2014 e Dezembro de 2017. Procedeu-se a uma revisão dos processos clínicos e foi estimado o risco de surdez neurossensorial. Os dados foram analisados com recurso ao IBM SPSS versão 24, com significância estatística para valores de p inferiores a 0.05.

Resultados: Incluímos 326 participantes, 176 do sexo masculino. A idade mediana da primeira avaliação foi de 2 meses. 38 casos foram referenciados devido a REFER no rastreio neonatal e 288 apresentavam indicadores de risco. 21 casos foram diagnosticados com surdez neurossensorial. Não se verificaram diferenças entre o grupo de crianças com surdez e o grupo de crianças normouvintes relativamente aos indicadores de risco mais frequentes: exposição ototóxica (38% vs 51%), prematuridade (28.6% vs 27.5%), e baixo peso ao nascimento (28.6% vs 22%). Aproximadamente 50% das crianças com surdez não apresentava indicadores de risco, enquanto que 29% tinham vários indicadores.

Conclusão: Este estudo enfatiza a importância do rastreio universal neonatal, uma vez que metade das crianças com surdez não apresentava indicadores de risco. A presença de múltiplos indicadores de risco parece ter uma maior contribuição nos casos de surdez. São necessários mais estudos para avaliar a significância clínica dos indicadores de risco.

Palabras clave

surdez, indicadores de risco, rastreio auditivo neonatal

Introduction

Sensorineural hearing loss (SNHL) is among the most common disorders at birth affecting 1-3 in every 1000 children. 1

Regardless of the incomparable importance afforded by universal screening methods, several conditions have been associated with an increased risk of SNHL and received special concern in daily practice. These factors are highlighted in the latest statement of the American Academy of Pediatrics Joint Committee on Infant Hearing (JCIH) which further emphasizes the need to monitor at-risk children. 1

However, despite the broad use of these recommendations in clinical practice, concerns regarding the true role of these factors were raised by different authors. 2,3

While previous studies identified conditions associated with SNHL, including severe birth asphyxia, neurological disorders, syndromes associated with hearing loss, TORCH infections and family history of deafness, the incidence of deaf or hard of hearing children was low. 4

One study that assessed neurodevelopment outcomes at 18 months among extremely low birth weight (LBW) and very LBW infants, found no evidence of significant hearing loss. 5 A large study conducted by Beswick also found no evidence to support monitoring low birth weight (LBW) infants.6

In another study that included hard of hearing children, no association was found between expected risk factors and progressive HL, including neonatal intensive care unit admission, family history, hearing loss syndromes and postnatal infections.7

Le Roux et al conducted a study that included 264 children diagnosed with profound hearing loss elected to cochlear implant amplification. Hearing loss associated risk factors were absent in almost half of the subjects, although 28% of these children had a history of NICU admittance. 8

While the association between these factors and sensorineural hearing loss remains debated, the incidence of hearing loss among these infants is highly variable, with some studies reporting high rates of hearing loss 9 and others reporting a low yield 10,11.

The aim of this study is to assess the prevalence of risk indicators among infants referred for targeted surveillance and estimate the association between these factors and SNHL among these children perceived to be at-risk.

Material and Methods

We conducted a retrospective study. Newborns were screened for SNHL with transient evoked otoacoustic emissions (TOAE) before hospital discharge. For the presence of TOAE the response was PASS, while REFER for absent. For those infants who presented a refer result in one or both ears, a second test was performed and both ears were retested. Patients who presented a refer result in either ear or those with PASS result in both ears with risk indicators for HL were referred to target surveillance program. This included automated auditory brainstem responses (A-ABR) and serial clinical assessment with periodic diagnostic ABR (D-ABR) as relevant.

We collected data obtained through clinical records. Information gathered included physical examination, otoscopy, tympanometry, A-ABR and D-ABR as relevant. While infants with normal A-ABR and no HL suspected were scheduled to be reassessed with D-ABR between 24 and 30 months, infants at risk for delayed HL were reassessed with D-ABR every six months. Infants with altered tests from middle ear causes were reassessed a few weeks later.

We included infants referred for A-ABR and Otolaryngology assessment between January 2014 and December 2017 using a consecutive sampling procedure.

Exclusion criteria included older children with language development delay referred for hearing assessment, infants who missed all audiometric test scheduled and infants referred for other motives who had their hearing tested.

We consulted individual charts to evaluate general information (gender, birth date), relevant family history, in utero exposures (gestational age, medication exposure, infection, trauma or hospital admission duration), anthropometric parameters (birth weight), APGAR score and reanimation interventions, admission care (well-infant nursery, neonate care intensive unit (NICU)), admission time before discharge, relevant therapeutic interventions (ototoxic medication, mechanical ventilatory support) and overall health status including congenital anomalies detected on physical examination.

We obtained a series of audiometric evaluations performed on each infant and a normal versus an abnormal result was recorded.

Statistical analysis was conducted on IBM SPSS version 24. We analyzed categorical variables using chi-square test and continuous variables with T-test.

Statistical significance was considered for p values below 0.05.

Results

A total of 375 infants were enrolled in this study. Twenty-five (6.7%) infants were excluded due to lack of data and 24 (6.4%) because they did not meet inclusion criteria. The final sample included 326 infants, 176 (54%) males and 150 (46%) females. A resume of the sample’s characteristics is depicted in table 1.

Table 1

Characteristics of the study sample (n = 326).

Age of target program inclusion (months, min-max)	2 (1-10)
Follow-up (median (min-max))	18 (1-48)
Gender (boys/girls, %)	176/150 (54/46)
Birth admission (WIN/NICU, %)	156/170 (47.9/52.1)
Gestational age (weeks, median (min-max))	37 (24-41)
Birth Weigh (grams, median (min-max))	2890 (520-4630)

Abbreviations: WIN= well infant nursery; NICU= neonate care intensive unit

The median age of first Otolaryngology assessment was 2 months (1-10) and 123 infants remained in follow up surveillance for 18 months (10-27).

Overall, 156 (47.9%) infants were admitted to a well-infant nursery after birth, while 170 (52.1%) were admitted to a NICU. Neonate care admission varied between 3 and 166 days (median 18).

While 38 infants were referred due to refer TOAEs, 288 (88.34%) had risk indicators for further assessment. The median number of risk indicators present was 1 (1-3), although 75 (24.8%) infants had three or more.

The prevalence of SNHL in this sample was 6.4% (n=21). Of these, 10 (47.6%) had no risk indicators for HL and 6 (28.57%) had three or more indicators. Isolated risk factors were present in only four (19%) patients and included a family history of HL, craniofacial anomalies and congenital CMV infection. Hearing loss among NICU admitted infants was 2.4% and all of these (n= 4) were admitted for more than 50 days.

When we compared the hearing loss group with the normal hearing group, we found no difference between both regarding the most common risk indicators for referral (table 2).

Table 2.

Comparison between normal hearing infants and hearing loss cases regarding different risk indicators

RI	NH (n=305)	HL (n=11)	OR	CI 95%	p value
Ototoxicity	156 (51%)	8 (38%)	0.59	0.24-1.46	0.25
Prematurity	84 (27.5%)	6 (28.6%)	1.05	0.4-2.8	0.92
Family History	76 (24.9%)	5 (9.5%)	0.32	0.07-1.4	0.11
LBW	67 (22%)	6 (28.6%)	1.42	0.53-3.8	0.59
Low AGAR	>60 (19.7%)	4 (19%)	0.96	0.31-2.96	1
Septicemia	60 (19.7%)	3 (14.3%)	0.68	0.19-2.39	0.78
CMV	10 (3.3%)	1 (4.8%)	1.48	0.18-12.1	0.53
Craniofacial anomalies	9 (3%)	3 (14.3%)	5.48	1.37-22	0.04*
BW <1000 g	13 (4.3%)	4 (19%)	5.29	1.56-17.95	0.018*

Abbreviations: RI= risk indicator; NH= normal hearing; HL= hearing loss; BW= birth weight

Craniofacial abnormalities was the only indicator associated with an increased risk of SNHL (OR= 5.48 1.37-22, p=0.04) though the number of affected participants was low. LBW was only associated with an increased risk when birth weight below 1000g was considered (OR=5.29, 1.56-17.95, p=0.018). These infants had prolonged NICU admissions and multiple risk indicators. Among participants referred for presenting risk indicators, having three or more indicators combined increased the risk of hearing loss (2.4% vs 1.7%, p= 0.021).

Discussion

While NICU admittance is associated with an increased SNHL risk among neonates, several factors are pointed as possible contributors, including central nervous system and circulatory system conditions, TORCH infections, prolonged artificial ventilation, noise exposure, and ototoxic medication. 1,12,13

In our sample, most infants were included in the hearing surveillance program because they presented at least one risk indicator for hearing loss. A minority represented healthy neonates with a refer result in TOAE. The prevalence of HL among NICU neonates was low and similar to that of previous studies 13 which reported a range between 0.7-3.7% 14

While some studies found an association between craniofacial anomalies and conductive hearing loss only, other authors found it to be associated with different subtypes of SNHL 4,15 We found this risk indicator more prevalent among hearing loss cases, however, the number of participants was low.

LBW was a frequent cause of referral, however, due to its inconsistent association with SNHL and the low yield of targeted surveillance, some authors are now recommending against routine monitoring 6,16. Other studies report a consistent increase in SNHL risk with decreased gestational age and birth weight, however, a causal relation could not be stablished. 14,17 In our study, we only found an increased risk of HL among premature weighing less than 1000g. This is not unexpected and was previously reported on an univariate analysis performed by van Dommelen et al.13 These premature had prolonged admissions and presented combined risk indicators. We could not perform multivariate analysis in this study because of the low number of HL cases.

Infants referred because of ototoxic medication exposure were most frequently treated with gentamycin and other aminoglycosides. While these agents are more commonly vestibulotoxic, a high-frequency hearing loss is a possible consequence.18,19 Variable factors are commonly associated with different hearing outcomes and include peak level, duration of treatment course and cumulative exposures possibly accounting for a variable effect.13,19-20 Also, point mutations in mitochondrial DNA are associated with an increased susceptibility to aminoglycosides and deafness-associated gene mutations can be more prevalent among NICU admitted infants. 21-22

In our sample, SNHL following aminoglycoside treatment was very low and associated with other risk indicators in all cases. However, due to cochlear tonotopic selectivity, ototoxic aggressors may first cause high-frequency SNHL.23 As the clicks of the D-ABR screening technique have an acoustic spectrum ranging from 2000-3000Hz, a high frequency hearing loss would be missed by this method.24-25

Other reported risk indicators were not evaluated because of the low prevalence in this sample.

Overall, our study shows a low diagnostic yield for targeted risk indicator surveillance. As risk indicators represent a major referral cause, a better comprehension of the factors that increase the risk of HL is necessary, in order to promote more effective case selection and avoid overcoming the health system capacity.

However, despite the fact that most infants with risk indicators present normal hearing thresholds, our knowledge of the hearing apparatus continues to increase. Recent studies on animal models have shown that the synapses between inner hair cells and low spontaneous rate auditory nerve fibers, are more susceptible to damage following noise exposure than other cochlear elements. These fibers do not participate in liminal responses but are important to suprathreshold sound detection, and are associated with hearing skills in complex auditory environments, like understanding speech in background noise. 26-27 Evidence of cochlear synaptopathy in humans is emerging in the literature, as evidenced by transitory and permanent hearing threshold shift following noise exposure. Also, reduction in wave I amplitude be an additional indicator of cochlear nerve fibers loss.28 However, a definitive diagnostic test is still under study.29 Accordingly, while most neonates with risk indicators do not present increased hearing threshold levels, a hidden hearing loss may not be evident by current screening tests.

Some limitations of this study include its unicentric nature and the small number of hearing loss cases, which precluded multivariate analysis and may miss important associations. Reporting biases and coding differences are also limitations inherent to its retrospective nature.

Conclusion

Hearing loss risk indicators are an important cause of Otolaryngology referral, however, we observed a weak association between isolated risk indicators and hearing loss. Our study emphasizes the importance of universal hearing screening. Further prospective and larger studies would be necessary to assess the clinical value of different risk indicators.

Conflict of interest

None declared.

Data Confidentiality

This study is in accordance with the ethics standards of the institute where it was carried out.

Compliance with ethical standards

The authors state that this study was performed in accordance with the ethical standards of the Declaration of Helsinki. This study was approved by Vila Nova de Gaia/Espinho Hospital Centre Ethical Committee.

References

1. Year 2007 position statement: Principles and guidelines for early hearing detection and intervention programs. Pediatrics. 2007 Oct;120(4):898-921.

2. Karaca CT, Oysu C, Toros SZ, Naiboglu B, Verim A. Is hearing loss in infants associated with risk factors? Evaluation of the frequency of risk factors. Clin Exp Otorhinolaryngol. 2014;7(4):260-263.

3. Vasconcellos AP, Colello S, Kyle ME, Shin JJ. Societal-level Risk Factors Associated with Pediatric Hearing Loss: A Systematic Review. Otolaryngol Head Neck Surg. 2014;151(1):29-41.

4. Ohl C, Dornier L, Czajka C, Chobaut JC, Tavernier L. Newborn hearing screening on infants at risk. International journal of pediatric otorhinolaryngology. 2009;73(12):1691-1695.

5. Zhang Q, Wu YQ, Zhuang Y, Cao J, Gao XR. [Neurodevelopmental outcomes of extremely low birth weight and very low birth weight infants and related influencing factors]. Zhongguo Dang Dai Er Ke Za Zhi. 2016;18(8):683-687.

6. Beswick R, Driscoll C Fau - Kei J, Kei J Fau - Khan A, Khan A Fau - Glennon S, Glennon S. Which risk factors predict postnatal hearing loss in children? J Am Acad Audiol. 2013 Mar;24(3):205-13. doi: 10.3766/jaaa.24.3.6.

7. Barreira-Nielsen C, Fitzpatrick E Fau - Hashem S, Hashem S Fau - Whittingham J, Whittingham J Fau - Barrowman N, Barrowman N Fau - Aglipay M, Aglipay M. Progressive Hearing Loss in Early Childhood. Ear Hear. 2016 Sep-Oct;37(5):e311-21. doi: 10.1097/AUD.0000000000000325.

8. le Roux T, Swanepoel de W, Louw A, Vinck B, Tshifularo M. Profound childhood hearing loss in a South Africa cohort: risk profile, diagnosis and age of intervention. Int J Pediatr Otorhinolaryngol. 2015;79(1):8-14.

9. Martines F, Martines E Fau - Mucia M, Mucia M Fau - Sciacca V, Sciacca V Fau - Salvago P, Salvago P. Prelingual sensorineural hearing loss and infants at risk: Western Sicily report. Int J Pediatr Otorhinolaryngol. 2013 Apr;77(4):513-8. doi: 10.1016/j.ijporl.2012.12.023. 

10. Fowler KB, Dahle Aj Fau, Boppana SB, Pass RF. Newborn hearing screening: will children with hearing loss caused by congenital cytomegalovirus infection be missed?  Pediatr. 1999 Jul;135(1):60-4.

11. Barreira-Nielsen C, Fitzpatrick E, Hashem S, Whittingham J, Barrowman N, Aglipay M. Progressive Hearing Loss in Early Childhood. Ear and hearing. 2016;37(5):e311-321.

12. Fuchs A, Zimmermann L, Bickle Graz M, et al. Gentamicin Exposure and Sensorineural Hearing Loss in Preterm Infants. PLoS One. 2016 Jul 8;11(7):e0158806. doi: 10.1371/journal.pone.0158806.

13. van Dommelen P, Mohangoo Ad Fau, Verkerk PH, et al. Risk indicators for hearing loss in infants treated in different neonatal intensive care units. Acta Paediatr. 2010 Mar;99(3):344-9. doi: 10.1111/j.1651-2227.2009.

14. van Dommelen P, Verkerk PH, van Straaten HL. Hearing loss by week of gestation and birth weight in very preterm neonates. The Journal of pediatrics. 2015;166(4):840-843.e841.

15. Coenraad S, Goedegebure A, van Goudoever JB, Hoeve LJ. Risk factors for sensorineural hearing loss in NICU infants compared to normal hearing NICU controls. Int J Pediatr Otorhinolaryngol. 2010 Sep;74(9):999-1002. doi: 10.1016/j.ijporl.2010.05.024.

16. Beswick R, Driscoll C, Kei J, Glennon S. Targeted surveillance for postnatal hearing loss: a program evaluation. Int J Pediatr Otorhinolaryngol. 2012 Jul;76(7):1046-56. doi:

17. Wroblewska-Seniuk K, Greczka G, Dabrowski P, Szyfter-Harris J, Mazela J. Hearing impairment in premature newborns-Analysis based on the national hearing screening database in Poland. PloS one. 2017;12(9):e0184359.

18. Lanvers-Kaminsky C, Ciarimboli G. Pharmacogenetics of drug-induced ototoxicity caused by aminoglycosides and cisplatin.

19. Kent A, Turner MA, Sharland M, Heath PT. Aminoglycoside toxicity in neonates: something to worry about? Expert review of anti-infective therapy. 2014;12(3):319-331.

20. Farzal Z, Kou YF, St John R, Shah GB, Mitchell RB. The role of routine hearing screening in children with cystic fibrosis on aminoglycosides: A systematic review. Laryngoscope. 2016 Jan;126(1):228-35. doi: 10.1002/lary.25409.

21. Soini Hk, Karjalainen MK, Hinttala R, Rautio A, Hallman M, Uusimaa J. Mitochondrial hearing loss mutations among Finnish preterm and term-born infants. Audiol Res. 2017 Nov 3;7(2):189. doi: 10.4081/audiores.2017.

22. Yang SM, Liu Y, Liu C, et al. Hearing-loss-associated gene detection in neonatal intensive care unit. The journal of maternal-fetal & neonatal medicine : the official journal of the European Association of Perinatal Medicine, the Federation of Asia and Oceania Perinatal Societies, the International Society of Perinatal Obstet. 2018;31(3):284-288.

23. Fettiplace R, Nam JH. Tonotopy in calcium homeostasis and vulnerability of cochlear hair cells. Hearing research. 2018.

24. Baldwin M, Watkin P. Predicting the degree of hearing loss using click auditory brainstem response in babies referred from newborn hearing screening. Ear and hearing. 2013;34(3):361-369.

25. Lu TM, Wu FW, Chang H, Lin HC. Using click-evoked auditory brainstem response thresholds in infants to estimate the corresponding pure-tone audiometry thresholds in children referred from UNHS. International journal of pediatric otorhinolaryngology. 2017;95:57-62.

26. Jensen JB, Lysaght AC, Liberman MC, Qvortrup K, Stankovic KM. Immediate and delayed cochlear neuropathy after noise exposure in pubescent mice. PLoS One. 2015 May 8;10(5):e0125160. doi: 10.1371/journal.pone.0125160.

27. Pienkowski M. On the Etiology of Listening Difficulties in Noise Despite Clinically Normal Audiograms. Ear Hear. 2017 Mar/Apr;38(2):135-148. 

28. Shi L, Chang Y, Li X, Aiken S, Liu LA-OX, Wang JA-O. Cochlear Synaptopathy and Noise-Induced Hidden Hearing Loss. Neural Plast. 2016;2016:6143164. Epub 2016 Sep 21.

29. Plack CJ, Leger A, Prendergast G, Kluk K, Guest H, Munro KJ. Toward a Diagnostic Test for Hidden Hearing Loss. Trends Hear. 2016 Sep 7;20. pii: 2331216516657466. doi: 10.1177/2331216516657466.