Epidemiology, aetiology and diagnosis of congenital hearing loss via hearing screening of 153 913 newborns

Abstract

Background

Congenital hearing loss (HL), one of the most common paediatric chronic conditions, significantly affects speech and language development. Its early diagnosis and medical intervention can be achieved via newborn hearing screening. However, data on the prevalence and aetiology of congenital HL in infants who fail newborn hearing screening are limited.

Methods

The sample population included 153 913 infants who underwent newborn hearing screening, and the prevalence of congenital HL, defined as moderate to profound bilateral HL (BHL) or unilateral HL (UHL) (≥40 dB HL), in one prefecture of Japan was measured to minimize the loss-to-follow-up rate, a common factor affecting the screening procedure. Comprehensive aetiological investigation, including physiology, imaging, genetic tests, and congenital cytomegalovirus screening, was performed on children diagnosed with congenital HL.

Results

The calculated prevalence of congenital HL was 1.62 per 1000 newborns (bilateral, 0.84; unilateral, 0.77). More than half of the cases with congenital bilateral or severe to profound UHL showed genetic aetiology or cochlear nerve deficiency (CND), respectively. Approximately 4% and 6% of the cases of congenital BHL and UHL were associated with congenital cytomegalovirus infection and auditory neuropathy spectrum disorder, respectively.

Conclusions

This is an epidemiological and comprehensive aetiological study of congenital HL, as determined via newborn hearing screening according to its severity and laterality, in a large-scale general population of a developed country. Our findings can serve as a reference for optimizing care and intervention options for children with HL and their families.

Key Messages.

• We evaluated the prevalence and aetiology of congenital hearing loss determined after newborn hearing screening failure according to its severity and laterality in a general population of a developed country.

• The calculated prevalence of congenital hearing loss was 1.62 per 1000 newborns (bilateral, 0.84; unilateral, 0.77).

• More than half of the cases with severe to profound congenital bilateral or unilateral hearing loss showed genetic aetiology or cochlear nerve deficiency, respectively.

• We highlighted the feasibility and benefits of comprehensive aetiological study, including physiologic, imaging, genetic tests, and congenital cytomegalovirus screening.

• Our findings can serve as a basis for providing care and intervention options for children with hearing loss and their families.

Introduction

Congenital hearing loss (HL) is one of the most common paediatric chronic conditions.¹ According to Morton and Nance,² the prevalence of congenital HL, defined as moderate to profound bilateral HL (BHL) or unilateral HL (UHL), is 1.86 per 1000 newborns. For children with congenital HL, early diagnosis, intervention, and treatment help improve developmental outcomes in later childhood.³ Additionally, children whose HL is identified by 6 months of age have substantially more optimized receptive and expressive skills than children whose HL is identified later.³ Therefore, in 1993, the National Institutes of Health recommended newborn hearing screening (NHS) within the first 3 months of life.⁴ Since then, NHS programs have spread throughout most developed countries; ¹ therefore, congenital HL can be identified in the early stages, leading to considerable improvement in the management of care for infants with HL.⁴

Although the early diagnosis of congenital HL following NHS is possible, medical and supportive treatments for congenital HL depend on the aetiology and HL type (sensorineural, conductive, or mixed).¹ However, only a few comprehensive aetiological studies have been performed on children with NHS-detected congenital HL. The most common aetiologies of deafness are genetic in nature, and approximately two-thirds of congenital/early onset sensorineural HL cases in developed countries are likely caused by genetic aetiology.² However, these aetiological data were merged with the results from multiple reports among different populations without comprehensive genetic testing; therefore, the data should be evaluated in one large-scale population through a comprehensive aetiological study. Genetic testing, specifically targeted genome resequencing using massively parallel sequencing (MPS), and retrospective testing for congenital cytomegalovirus (cCMV) infection have advanced remarkably; thus, cases of genetic and cCMV-associated deafness can now be properly diagnosed, thereby improving the diagnostic rate.^5–7 Given the aforementioned background and technological advances, epidemiological and comprehensive aetiological studies of congenital HL should be revisited. To the best of our knowledge, no study has yet described the aetiology of HL with respect to its severity and laterality. Herein, we investigated 153 913 newborns in Nagano prefecture from 2009 to 2019 in terms of: (1) referral rate and positive predictive value of NHS; (2) prevalence of congenital BHL and UHL; and (3) aetiology of congenital BHL and UHL according to the degree of HL. Our findings could be useful to paediatricians, otologists, and audiologists conducting follow-ups for children with HL and provide their family members with evidence for the prevalence and aetiology of newborns with congenital HL.

Methods

Patients

Between 2009 and 2019, 156 038 babies were born in Nagano Prefecture in central Japan (Figure 1A). Among them, 153 913 infants who underwent NHS at 41 hospitals were enrolled. The aforementioned data were provided by the local government. Most of the babies tested were Japanese. The overall coverage for NHS was 98.6%. NHS was performed using an automated auditory brainstem response (AABR) screening device [Natus algo^® 3i, ATOM MEDICAL, Tokyo, Japan] at 38 birth hospitals, or an automated otoacoustic emissions (AOAE) screening device [OAE screener ER-60, RION, Tokyo, Japan] at 3 hospitals a couple of days after birth. In both protocols, a pass or refer was evaluated automatically, but testing was performed two or more times. When babies failed the AOAE test, AABR was subsequently performed to confirm the AOAE results. Babies cared for in the neonatal intensive care unit for a prolonged period received auditory brainstem response (ABR) rather than NHS procedures in Nagano prefecture, suggesting that only healthy babies were enrolled in this study. As a neonatal intensive care unit stay of >5 days is one of the risk factors for congenital HL, we were unable to include such cases, which was one of the limitations of this study. After NHS, 661 bilateral and unilateral cases were referred. These babies underwent ABR screening using the Neuropack [Nihon Koden Co., Tokyo, Japan] at approximately 1 month of age at 16 ear, nose and throat hospitals in Nagano prefecture. When the ABR threshold, measured for clicks, was below 35 dB nHL, the hearing level was considered to be normal. Children found to have normal hearing were excluded, and 403 children who failed both NHS and 1-month ABR screening visited Shinshu University Hospital. Final diagnoses were made using the results of auditory steady-state response evaluations (MASTER, Natus, CA, USA) (Figure 1B). Until 2019, Shinshu University Hospital was the only institution in Nagano Prefecture at which these diagnoses were made. All subjects with HL underwent audiometric and diagnostic testing before the age of 6 months.

Figure 1.

Location of Nagano Prefecture in Japan (A), hearing screening protocol (B), and the number of tested newborns at each screening step (C). ABR, auditory brainstem response; ENT, ear, nose, and throat; HL, hearing loss; ASSR, auditory steady-state response; NHS, newborn hearing screening; OME, otitis media with effusion

Audiometric assessment

Audiologic assessments, including conditioned orientation response audiometry, distortion product otoacoustic emission testing, and auditory steady-state response under sedation, were performed on infants who did not pass the NHS or ABR screening within 3–4 months, to identify cases of moderate to profound congenital UHL or BHL. For auditory steady-state response, the mean hearing ability of both ears was defined as the average air-conduction thresholds at 0.5, 1, 2, and 4 kHz, and HL was characterized as moderate or severe. Permanent HL and potentially curable HL, such as congenital aural atresia (CAA), congenital aural stenosis (CAS), and middle ear malformations, were included. The extent of hearing impairment was set as follows: moderate, 41–70 dB HL; severe, 71–90 dB HL; and profound, ≥90 dB HL. Patients with mild HL (≤40 dB HL) were not included, because: (1) distinguishing mild HL from normal hearing was clinically complicated, and (2) the goal of this study was to determine the frequency of patients with HL requiring medical intervention; i.e., moderate to profound HL either bilaterally or unilaterally. If the initial auditory test revealed that both ears had HL, the patient was diagnosed with BHL. If one ear had normal hearing ability but the other ear demonstrated HL, the patient was diagnosed with UHL.

Diagnostic testing

Clinical history was taken, and a physical examination was performed. Subsequently, all babies diagnosed with HL either bilaterally or unilaterally underwent computed tomography (CT) imaging. Except for the children for whom the aetiology was clearly identified using the imaging test, all of the children with bilateral sensorineural or mixed HL underwent genetic testing as well as cCMV testing. CT imaging was conducted to study the causes of HL and diagnose cochlear nerve deficiency (CND), CAA, CAS, or inner or middle ear malformations. CND was diagnosed on the basis of bony cochlear nerve canal measurement (1.4 mm or less) and/or internal auditory canal width (3 mm or less).⁸ Notably, the cases with HL due to otitis media with effusion were excluded because otitis media with effusion was considered to be easily curable, unlike other outer and/or middle ear abnormalities.

For infants with BHL, simultaneous genetic testing, which is covered by social health insurance in Japan, was performed.⁹ Blood samples were collected until the age of 6 months, but all genetic testing was reanalysed in 2021–2022. As the methods for genetic testing have advanced, 63 genes reported to cause non-syndromic HL (Hereditary Hearing Loss; http://hereditaryhearingloss.org/) (see Supplementary Table S1, available as Supplementary data at IJE online) were analysed using the same analysis platform, filtering method, and pathogenicity assessment described in our earlier study.⁹ All genetic results were confirmed via segregation analysis. DNA was collected not only from the patients but also from their parents, enabling us to perform segregation analysis. The pathogenicity of the identified variants was evaluated in accordance with the American College of Medical Genetics standards and guidelines¹⁰ and the ClinGen Hearing Loss Clinical Domain Working Group Expert specifications.¹¹ All patients with genetic deafness were also enrolled in earlier studies including the associated review article.⁹

For children with both BHL and UHL, cCMV DNA testing was performed in accordance with our previously described method.^5,6 Preserved dried umbilical cord samples were collected from patients with HL. In Japan, all obstetric hospitals customarily provide the dried umbilical cord to every parent as a symbol of the bond between mother and child, thus facilitating the retrospective diagnosis of cCMV infection.

Results

Referral rate and positive predictive value of NHS

Of the 153 913 infants who underwent NHS, 661 failed bilaterally or unilaterally, indicating a referral rate of 0.43%. Of the 403 newborns who underwent audiometric assessment, 249 were diagnosed with HL, excluding the cases with mild HL or otitis media with effusion (Figure 1C). These results indicated that the positive predictive value of NHS was 37.7%.

Prevalence of congenital BHL and UHL

Of the 249 newborns diagnosed with HL, 130 (74 males and 56 females) and 119 (59 males and 60 females) had BHL and UHL, respectively (Figure 1C). Given that 153 913 newborns underwent NHS, congenital BHL and UHL had prevalences of 0.84 and 0.77 per 1000 newborns, respectively. Overall, the prevalence of congenital HL was 1.62 per 1000 newborns. Of the 130 BHL cases, 66 had moderate and 64 had severe to profound HL (Figure 2). Of the 119 UHL cases, 49 had moderate and 70 had severe to profound HL (Figure 4).

Figure 2.

Aetiology of children with congenital bilateral hearing loss (HL). (A) Overall, (B) moderate HL cases, and (C) severe to profound HL cases. CND, cochlear nerve deficiency; CAA, congenital aural atresia; CAS, congenital aural stenosis; cCMV, congenital cytomegalovirus; OME, otitis media with effusion

Figure 4.

Aetiology of children with congenital unilateral hearing loss (HL). (A) Overall, (B) moderate HL cases, and (C) severe to profound HL cases. cCMV, congenital cytomegalovirus; CAA, congenital aural atresia; CAS, congenital aural stenosis; CND, cochlear nerve deficiency

Aetiology of congenital BHL

In our cohort of congenital BHL, the most prevalent cause was genetic (56.2%, including 42.3% nonsyndromic, 6.9% syndromic, and 6.9% chromosomal abnormality), followed by cCMV and CAA/CAS (Figure 2). For newborns with moderate BHL, the most prevalent causes were also genetic (33.3%). For newborns with severe to profound BHL, the most prevalent cause was genetic (65.6%), followed by cCMV and inner ear malformation. Among the newborns with congenital BHL, 58 patients were diagnosed with genetic HL; for nonsyndromic cases, 11 genes related to BHL were identified. The most frequently implicated gene was GJB2, followed by SLC26A4, STRC, and OTOF (Figure 3). Most of the cases of genetic aetiology with chromosomal abnormality had 21 trisomy, which can be identified using ultrasound and subsequently diagnosed by chorionic villus sampling or amniocentesis prenatally.

Figure 3.

Frequency of causative genes in children with congenital bilateral nonsyndromic hearing loss

Aetiology of congenital UHL

In our cohort of patients with congenital UHL, the most prevalent cause was CND (40.3%), followed by CAA/CAS and cCMV (Figure 4). For newborns with moderate UHL, the most prevalent cause was CAA/CAS (20.4%), followed by CND. For newborns with severe to profound UHL, the most prevalent cause was CND (55.7%), followed by CAA/CAS and cCMV.

Aetiology of auditory neuropathy spectrum disorder

In terms of auditory neuropathy spectrum disorder (ANSD), its overall frequency of occurrence was 5.6%. Bilateral ANSD was found in 5 cases (80%), among which OTOF gene-related deafness was the most frequently occurring type (4 cases). ANSD occurred unilaterally in 9 cases (77.8%), among which CND was the most frequently implicated cause (7 cases).

Discussion

The findings of this study showed that the calculated prevalence of congenital HL in a developed country was 1.62 per 1000 newborns. This was consistent with other reports (1.0–6.4 per 1000 newborns).^12,13 In Nagano prefecture, almost all infants suspected of having HL based on NHS and ABR visited Shinshu University Hospital to receive an auditory steady-state response, which minimized the loss-to-follow-up rate, a common factor affecting screening procedures. Therefore, our study offers a more accurate measure of the prevalence of congenital HL. Unlike other studies, this study included temporary HL as well as permanent HL; except for otitis media with effusion, some cases of conductive HL resulting from CAA, CAS, or middle ear malformation are not curable. As such conductive HL and permanent sensorineural HL have a similar impact on the families of children with HL, any aetiology of HL was included in this study.

Traditionally, normal hearing on only one side is considered necessary for normal development of speech and language; consequently, UHL remains underexamined.¹⁴ Lieu et al.¹⁵ concluded that speech and language delays may occur in some children with UHL, indicating that analysing the prevalence of UHL is also important. Our findings reveal that the prevalence of congenital UHL is comparable to that of congenital BHL. The findings of Chu et al.¹⁶ support this result; however, Nance et al.¹⁷ reported, nine years earlier, that the frequency of congenital UHL is 30–40% that of congenital BHL. Interestingly, these values differ, likely because of the worldwide spread of NHS, allowing for the early and accurate diagnosis of UHL, which is often missed without NHS.¹⁴

We found that the referral rate from NHS was 0.43%. Some studies have reported referral rates between 0.7% and 3.4%, but other studies with smaller populations have documented higher rates.^18,19 Our data are consistent with those of studies with relatively large populations. The positive predictive value of NHS was 37.7%, which is much higher than the rates reported in previous studies (5.3%–27.8%).^12,20 It is possible that the referral rate from NHS and the positive predictive value of NHS were relatively high because of differences in the definitions of HL, including conductive HL.

In previous studies describing the aetiology of congenital HL, each frequency of HL aetiology (e.g., genetic cause, cCMV, and CND) was reported separately.²¹ However, in this study, a comprehensive aetiological study was initially conducted via imaging tests, genetic test, and cCMV screening, which improved the diagnostic rate.

Based on the genetic screening results of 63 deafness genes, we found that more than half of the cases of BHL had a genetic aetiology. Additionally, at least 11 genes were involved in congenital BHL, which suggested that comprehensive genetic testing is required for aetiological investigation. In this study, we chose the 63 frequent genes which were responsible for genetic deafness. However, as over 100 non-syndromic deafness-causing genes have been reported to date, the diagnosis could have been more accurate by including more genes. As we reported previously, ⁹ genetic testing contributes not only to making accurate diagnoses but also to administering optimal medical intervention (i.e., hearing aids or cochlear implants). Therefore, genetic analysis is essential for comprehensive aetiological studies of congenital BHL. The diagnostic rate via genetic testing differed among severity groups, indicating that the severity of HL affects the diagnostic rate, which is consistent with previous reports.^9,22 We determined that the overall diagnostic rate was 56.2%, which is comparable to that reported in our earlier nationwide study in Japan (48.6%)⁹ and in a previous worldwide study (48%).²² The frequency of causative genes found by the present study was similar to that described in our earlier nationwide studies.^7,9 While further investigation is warranted for children with unknown causes of HL reported in this study, it is also necessary to enhance the analysis by including children with BHL as well as those with UHL, as UHL may have a genetic aetiology in some patients.²²

cCMV represents an important cause of congenital HL that can be present at birth or develop over time (delayed onset). We previously demonstrated that approximately 10% of cases of deafness in children with congenital and delayed-onset HL are associated with cCMV infection.^5,6 In the present study, approximately 4% of cases of congenital BHL and UHL were associated with cCMV infection, and this result is consistent with the findings of Beswick et al.²³ (3.64%). Although cCMV testing using urine is a more accurate method to identify cCMV-related HL,²¹ most of the patients enrolled in this study visited the hospital after the neonatal period, meaning that it was very difficult to make a diagnosis through standard urine screening. The best available strategy was to diagnose cCMV infection using dried umbilical cords as described by Koyano et al.²⁴ Hearing ability-related outcomes can reportedly be improved by the administration of antiviral drugs within the first month of life for symptomatic cCMV infection,⁶ meaning that the early diagnosis of cCMV would facilitate appropriate and effective interventions.

CND is an increasingly recognized cause of congenital UHL.²⁵ Consistent with previous results,²⁶ we found that CND was the most common cause within our cohort, accounting for more than half of the cases of severe to profound congenital UHL. HL due to CND is often severe to profound; however, various audiometric patterns have been observed.⁸ Imaging modalities, such as CT and magnetic resonance imaging, are essential. Cases with internal auditory canal and bony cochlear nerve canal stenosis that can be observed by CT are highly likely to have CND. However, children with congenital HL for whom the internal auditory canal and bony cochlear nerve canal are normal should undergo cochlear nerve assessment by magnetic resonance imaging as well, considering the potential for CND.⁸ This study included only the CT results; therefore, it is possible that some CND cases were not detected. This is one of the limitations of this study. Additionally, as CT basically exposes them to ionizing radiation, magnetic resonance imaging may be a safer choice in infants. When unilateral CND is diagnosed, it can be explained to their parents that normal hearing in the opposite ear will not be affected. In addition, hearing aids or cochlear implants may not dramatically improve hearing in patients with CND.

ANSD is a specific form of HL with an abnormal ABR and the presence of otoacoustic emission.²⁷ ANSD can significantly impede language development, making early identification critical.²⁸ Thus, late diagnosis is a disadvantage for such ANSD patients.²⁹OTOF-related ANSD is reportedly the most prevalent form of ANSD. Our findings suggest that 80% of bilateral ANSD cases were caused by pathogenic variants in the OTOF gene, consistent with previous reports (23–90.9%).²⁹ In our earlier study, otoacoustic emission responses were observed in approximately 10% of the cases with CND.⁸ Accordingly, CND can be observed in 60–100% of cases with UHL diagnosed with ANSD.^25,30 Similarly, we found that 77.8% of unilateral ANSD cases were caused by CND, suggesting that cases clinically diagnosed with unilateral ANSD should be assessed using imaging-based tests, considering the possibility of CND. Although environmental factors such as neonatal intensive care unit stay, prematurity, perinatal anoxia, severe hyperbilirubinemia, and low birth weight can also cause ANSD,²⁸ the result in this study was likely underestimated.

Conclusion

This study evaluates the prevalence and aetiology of congenital HL determined after NHS failure according to its severity and laterality in the general population of a developed country and presents the overall picture of congenital HL. Additionally, it highlights the feasibility and benefits of comprehensive aetiological studies, including physiology, imaging, genetic tests, and cCMV screening for newborns, which result in appropriate clinical decision-making, treatment approaches, and early intervention strategies, as well as counselling for families with children with HL.

Ethics approval

This study was approved by the ethics committee of the Shinshu University School of Medicine (Approval numbers: 4313).

Data availability

The data underlying this article cannot be shared publicly due to privacy concerns restricting availability of register data for research.

Supplementary data

Supplementary data are available at IJE online.

Author contributions

H.Y., S.N. and S.U. conceived the study and its design and had full access to the data. H.Y., T.O. and JS contributed to data analyses. Y.T. and S.N. were responsible for the results of newborn hearing screenings and genetic testing, respectively. H.Y., S.N. and S.U. wrote the first draft of the current manuscript, and all co-authors provided feedback and approved the final version for submission. The corresponding author attests that all listed authors meet authorship criteria and that no others meeting the criteria have been omitted.

Funding

This research was funded by a Health and Labor Sciences Research Grant for Research on Rare and Intractable Diseases and Comprehensive Research on Disability Health and Welfare from the Ministry of Health, Labor and Welfare of Japan (S.U. 20FC1048); the Aiba Works Research Scholarship for Young Physicians (H.Y.); and the Shinshu Public Utility Foundation for the Promotion of Medical Sciences (H.Y.).

Conflict of interest

None declared.