Speech perception and spoken language of children with auditory neuropathy spectrum disorder: a systematic literature review

ABSTRACT Purpose: to analyze the test results of speech perception and spoken language in children with hearing loss and auditory neuropathy spectrum disorder, users of hearing aids or cochlear implants. Methods: a systematic review of the literature based on the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. Consultation was performed in databases, considering studies from 1996 to 2021, selecting the studies that presented the results of speech perception or spoken language in children with bilateral auditory neuropathy spectrum disorder, with no structural alterations of the ear and/or vestibulocochlear nerve, or other associated impairments. Descriptive analysis was performed. Literature Review: among 1,422 studies found, 15 were included. Variability in the sample size, types of studies, evaluation procedures and methodological questions were observed. The cochlear implants and hearing aids contributed to the development of speech perception and spoken language in children with auditory neuropathy spectrum disorder, since skills such as auditory comprehension and speech intelligibility, were achieved. Conclusion: 15 studies suggest that cochlear implants and hearing aids may be effective for speech perception and spoken language development in children with auditory neuropathy spectrum disorder, with no other associated impairments. The need for further research with a high methodological rigor is highlighted.


INTRODUCTION
The term "Auditory Neuropathy" was first published by Starr et al. (1996)  1 to describe a possible impairment of the inner hair cells, the auditory nerve and/or the failure between the synapses of the inner hair cells and the auditory nerve 1,2 .Over the years, studies in the area have used different nomenclatures, such as "Auditory Neuropathy", "Auditory Dyssynchrony" and "Auditory Neuropathy Spectrum Disorder" -ANSD 1,3,4 .
The diagnosis of auditory neuropathy is verified by the presence of Evoked Otoacoustic Emission, register of abnormal or absent Brainstem Auditory Evoked Potentials (BAEPs) in the neural response and/or the presence of Cochlear Microphonism, observed in the BAEPs or in the Electrocochleography 5,6 .In addition, when there is a hearing loss, the degree can vary from mild to profound and the difficulty in understanding speech in silence and, especially, in noise, is disproportionate to the hearing loss presented 4,7 .
The variability found in the Auditory Neuropathy Spectrum (ANS) diagnostic tests demonstrates the heterogeneity of its pathophysiology 8,9 .Individuals with ANS may present alterations in the perception of sound stimuli and in the ability of temporal auditory processing, due to the failure in neural synchrony 7,10 .
The use of electronic devices to access speech sounds, such as the Hearing Aid (HA) and the Cochlear Implant (CI), brings the possibility of minimizing the impact of auditory sensory deprivation and enabling the development of spoken language 11,12 .
When the results provided by the HA regarding auditory speech perception are limited, the CI may be indicated, since this device partially replaces the function of the cochlea hair cells, compensating for the alteration of neural synchrony 13,14 .However, the literature reinforces the importance of considering the individuality of each case, and caution is necessary in relation to the conduct of indicating the cochlear implant 6,9,13 .
A previous study systematically reviewed the existing evidence regarding the results of the auditory abilities of children using CI with auditory neuropathy spectrum disorder, verifying advances in the detection of speech, in the discrimination and recognition of words and sentences after surgery, and no differences were observed in these abilities when compared to children with sensorineural hearing loss who used cochlear implants 15 .spoken language, with no structural alterations of the ear or vestibulocochlear nerve (CN VIII) or other associated impairments, such as syndromes.
• Intervention: use of the bilateral hearing aid or unilateral or bilateral cochlear implant or use of the unilateral cochlear implant with contralateral (bimodal) hearing aid.
• Clinical outcomes: results expressed as a percentage of correct answers or classifications of scales that demonstrate the performance in the perception of speech and/or spoken language.

Study Selection Criteria
For the selection of primary studies, the following criteria were considered: • Types of studies: those in Portuguese, English or Spanish, published in the period from 1996 to 2021.The delimitation of studies starting from 1996 is justified by the fact that it was in this year the term "auditory neuropathy" and its description was first published.When publication was available, dissertations and theses were replaced by published work.To determine the level of evidence, the classification adapted by Cox (2004) was used 18 .
To assess the risk of bias, the critical appraisal tools appropriate for this type of study from the Joanna Briggs Institute were used 19 .The risk of bias was characterized according to the frequency of "yes" answers, which meant adequacy of the study.The risk of "high" bias corresponded to the frequency of "yes" responses less than or equal to 49%, "moderate" between 50% and 69% and "low" greater than 70%.These tools were applied by the same reviewers who evaluated the articles in the previous stages, through consensus and, in case of disagreements, a discussion was held to reach consensus.
It is worth noting that the literature review has been updated in order to contemplate all possible studies that were published by the end of 2021.

LITERATURE REVIEW
Considering the importance of monitoring the results of the intervention, both with the hearing aid and with the CI, this study aimed to analyze the results of tests of speech perception and spoken language of children with auditory neuropathy spectrum disorder.This objective aims to answer the guiding question of the study, which analyzes the effectiveness of the devices in this population.
Figure 1 shows the diagram of the results of this search.

Data Analysis
The studies were extracted from the databases and organized using the EndNote software (online version), and repeated work was manually excluded.
The evaluation took place in two stages, by the Rayyan selection platform.
In the first stage, a blind evaluation by two reviewers selected the studies with potential relationship with the theme addressed by reading the titles and abstracts.In the second stage, two reviewers through blind evaluation analyzed the full text of the studies in which inclusion or exclusion was not possible only by reading the title and abstract, so as to define which studies fulfilled the selection criteria.
When there were disagreements among the judges regarding the inclusion or exclusion of the studies in the previous stages, another judge was called as a tie-breaking criterion.The third judge was able to analyze the title, abstract and full text (when it was not possible to include or exclude the study only by reading the title and abstract).
A protocol was used to analyze the research resulting from the previous stages, containing the following information: authors, year of publication, objectives, population, procedures for assessing speech perception and/or language, main results, level of evidence of the studies and risk of bias.One author organized these data, which were analyzed by another author and, if there were disagreements, a discussion was held to reach consensus.60 studies were excluded for different reasons, such as language, population and clinical outcomes, and 15 were included in this study.Taking into account the period elapsed between the first description of auditory neuropathy and the current advances in audiological diagnosis and intervention, it can be inferred that the number of studies found is relatively low.
Chart 2 shows the synthesis of the characteristics of the included studies.
The research identified 1,422 studies in the databases used.After the duplicates were removed, 845 papers were screened.Of these, four dissertations/ theses had already been published, being disregarded from the analysis and 765 studies did not contemplate the selection criteria.
After reading the title and abstract, 76 studies showed a relationship with the proposed theme and in only one the full text was not recovered, totaling 75 studies included for the second stage.After full analysis,    13 Case report

Chart 2. Summary of included studies
To present the potential benefits of CI in a case of a child with ANSD.

Not provided
The child showed significant improvement in speech perception after implantation.
When compared with 10 children who had cochlear hearing loss and used CI, their performance in the skills evaluated was similar.se report To report a case of a child with ANSD and CI.

boy who received unilateral CI.
There is no reported degree of hearing loss.Captions: CI: cochlear implant; ANSD: auditory neuropathy spectrum disorder; HA: hearing aid studied children with ANSD and moderate hearing loss 26,27 .This result emphasizes the importance of neonatal hearing screening programs, which involve both evoked otoacoustic emissions and brainstem auditory evoked potentials, because the combination of both tests will avoid errors, allowing to verify babies with possible findings compatible with auditory neuropathy spectrum disorder 4,28,29 .
Most of the children evaluated had severe or profound hearing loss and used cochlear implants 13,22- 26,30-36 .Studies show that the cochlear implant can bring greater neural synchrony 13,24,31 , becoming an alternative treatment when the hearing aid is not effective for the development of auditory skills and language 23 .However, the CI indication process should be thorough, considering that a common clinical conduct is not possible for this group of children, due to the fact that the ANSD presents different degrees of impairment along the auditory pathway, with fluctuation in auditory performance, making the decision-making process about the best rehabilitation option for this population challenging 26,31,33,34 .
Two studies showed that children who use hearing aids may present benefits in speech perception skills, with regard to auditory recognition and comprehension, when evaluating children with degrees of hearing loss that ranged from mild to severe 12,26 .However, another study, which evaluated the language development of children using hearing aids with moderate to severe hearing loss, found superior performance for participants with lower degree of hearing loss, stating that amplification can help in hearing, but is not enough to modify the neural coding pattern that is altered in the different degrees of hearing loss, especially in the severe degree 27 .
To this extent, it is emphasized that a careful preoperative evaluation is necessary, which takes into account different aspects, such as the individual benefits of hearing aids, the parents' perception of the child's development, the parents' expectations regarding the results of the CI and the active participation of the family in the rehabilitation process 13,22,32 .
It is worth mentioning that, after the conduct is established, the success of the devices (HA and CI) for the development of auditory and language skills involves factors such as the motivation to use the device, family support, speech therapy, among others 31,37 .
Regarding the clinical outcomes of speech perception, it was possible to observe that results were obtained through procedures that involved inventories Regarding the population of the included studies, the chronological age ranged from two years and two months to 12 years and two months, with a sample size between one and 136 participants.Regarding the degree of hearing loss, two studies included participants with mild hearing loss, two studies addressed the moderate degree, seven studies evaluated children with severe hearing loss, 12 encompassed the profound degree, one article mentioned the nomenclature "degree of severe hearing loss" and one study did not mention the degree of hearing loss.
When analyzing the criteria related to the population included in this study, we selected the studies with children with no structural alterations of the ear or vestibulocochlear nerve (cranial nerve VIII) and with no other associated impairments, such as syndromes.

It is known that the diagnosis of the Auditory
Neuropathy Spectrum Disorder may be accompanied by other comorbidities, due to risk factors, genetic factors and peri-or postnatal complications, which makes the population extremely heterogeneous 8,9,20 .
In addition, it is worth noting that the term "spectrum" brought an expansion to the concept of the lesion site, which contemplates other sites of alteration, in addition to the auditory nerve 21 .Although the pathophysiological mechanism of auditory neuropathy is not yet fully understood, the literature reports that the lack of synchronization of neural discharges along the auditory pathway causes changes in the accuracy of the coding or transmission of auditory information, which may explain this condition 7,9,10 .Thus, in view of the different manifestations that individuals with ANSD may present, the referred criteria for the population were established aiming at greater homogeneity of the participants of the included studies, which may be related to the number of studies found.
Among the included studies, the number of the sample varied widely, observing case report studies 13,[22][23][24] that evaluated the speech perception and language of children who received unilateral cochlear implant and a multicenter study 25 , which proposed to evaluate the auditory performance and speech production of 136 children who received the cochlear implant between 2003 and 2015, performing a follow-up in the period of two years post-implantation.Due to the longitudinal character, the authors obtained the complete evaluations of 79 participants.
Regarding the degree of hearing loss, which ranged from mild to profound, only two studies involved participants with ANSD and mild hearing loss 12,26 and two These different findings in the literature show that knowing the site of the alteration along the auditory pathway is fundamental in assisting in the choice of the technological resource for auditory rehabilitation and in understanding the prognosis of the individual with ANSD.When there is a sensory impairment, a favorable prognosis is expected with the use of hearing aids, but if there is a dysfunction that also involves the neural component, the auditory responses may be unpredictable, regardless of the device used (HA or CI) and, specifically with the CI, the results may depend on the ability of the auditory nervous system to deal with electrical stimulation 12 .
It is worth noting that among the tests applied with the child, only two studies evaluated the perception of speech in the presence of competitive noise.In both studies, children with CI and ANSD were compared with children with CI and sensorineural hearing loss, in relation to the recognition of sentences in noise 33,36 and spondaic words 33 , in which the performance of these groups was similar.
Considering that monitoring of the development of the auditory skills of children who use devices is an important factor to verify the effectiveness of the intervention with the HA and the CI and that this population can reach the plateau in the tests applied in silence, the evaluation with competitive noise becomes relevant, because it inserts a listening situation closer to everyday life 12,25,38 .
Regarding the clinical outcomes of spoken language, the receptive and expressive aspects were evaluated, using different procedures, as shown in Chart 2.
In relation to the receptive aspect, only one study investigated this ability, through the Language Development Assessment (ADL) procedure.Three children using bilateral hearing aids, with moderate to severe hearing loss, who underwent speech therapy with emphasis on hearing and spoken language, were evaluated, verifying differences in receptive age between the pre-and post-therapy moments, which ranged from six months to three years 27 .Regarding the expressive aspect, the same study obtained results by ADL before and after the intervention, with differences in expressive age ranging from one year to one year and seven months 27 .
Although the number of participants was low, this study demonstrates the importance of speech therapy, with emphasis on the development of hearing and spoken language, as children with hearing loss, using answered by the guardians, scales scored by the evaluator, tests with the child him/herself or the procedures agreed upon.
Among the inventories answered by the parents/ guardians, the Meaningful Auditory Integration Scale (MAIS) was applied in two studies 22,32 and the Infant-Toddler Meaningful Auditory Integration Scale (IT-MAIS) in three studies 12,32,33 .The use of these procedures is relevant, considering that the report of the parents/guardians is an important tool to evaluate the effectiveness of the devices in significant contexts of daily life, especially for young children 22 .
The scores on these scales ranged from zero to 44% before CI surgery 22,33 , while after surgery scores of up to 100% were achieved 12,22,32,33 .In one study 22 , the results were obtained after three and seven months of use of the implant, achieving development of hearing skills, with a maximum score of 81% at the end of this period.Another study found that children implanted before 24 months of life had higher scores for the MAIS and IT-MAIS scales than children implanted after this period 32 .
The early age for cochlear implant surgery is reported in the literature as a factor capable of bringing benefits to the speech perception of children with ANSD 25,32,34 .Other authors, who evaluated the auditory skills of this group, through the Categories of Auditory Performance (CAP) scale, found superior results for participants implanted early 25,34 .
Regarding the tests applied with the child him/ herself, the results showed that different skills were evaluated through a variety of evaluation procedures, as shown in Chart 2.
One of the main aspects considered was to understand whether there were differences in performance in speech perception between children with ANSD who used hearing aids or CI and children with sensorineural hearing loss 13,26,33,34,36 .It was noted that four studies found similar results between the groups with regard to the abilities of detection of speech sounds 13 , attention to environmental and speech sounds of everyday life 33,34 , auditory discrimination 13,33,34 , recognition in closed and open set 13,33,34,36 , sequential memory 13 , background figure 13 and auditory comprehension 13,34 .
However, one study found different results, in which children with ANSD who used CI or HA had lower performance in the ability to recognize phonemes, when compared to children with sensorineural hearing loss 26 .differences in performance between the groups 30,34 .These findings demonstrate that the access to speech sounds provided by the adequate adaptation of the device can bring benefits to the spoken language of children with hearing loss.
It was also pointed out in the studies that children with higher scores for speech intelligibility had shorter sensory deprivation time and had undergone CI surgery at an early age 25,32,34 .In addition, there was an evolution in the performance of speech intelligibility, as a function of the duration of the post-adaptation follow-up period with the CI 25 .These results show the importance of early intervention and systematic monitoring of the development of this group.
Regarding the level of evidence, eight studies were classified in "level 4", which consists of descriptive studies, five case-control studies 26,30,33,34,36 , two cohort studies 25,32 and one cross-sectional study 31 , while seven studies were classified in "level 5", which represents the case studies 12,13,[22][23][24]27,35 . It ispossible to observe that the 15 included studies are between levels 4 and 5 of the classification proposed by Cox (2004)  18 , which consist of lower levels related to the quality of the study.
Chart 3 shows the risk of bias of the studies, analyzed by the critical appraisal tools of the Joanna Briggs Institute 19 .devices, need an intervention that helps in the use of hearing information obtained by hearing aids or CI for the establishment of effective oral communication 24,27 .
It is worth mentioning that the main expressive aspect evaluated in the studies was speech intelligibility, in which four studies 22,25,32,34 used the Speech Intelligibility Rating (SIR) scale.One study 35 applied a speech intelligibility test based on percentage and one study 30 used the ABFW phonology test.
The literature reports that intelligibility is an aspect that can bring representative results about the child's competence for spoken language, quickly, reflecting their performance in a more natural communication situation 39,40 .
All studies that evaluated speech intelligibility were performed with implanted children and verified that the intervention with CI led to an increase in the scores obtained with the scales used 22,25,30,32,34,35 .Considering that the development of spoken language in children who use hearing aids may also be influenced by the quality of the acoustic signal provided by the device, impacting on the expressive aspect, the evaluation of speech intelligibility in this group is also relevant 27 .
Two studies compared speech intelligibility between children with ANSD who used CI and implanted children with sensorineural hearing loss, finding no      The risk of bias in three studies was classified as moderate, with blanks in the description of the patients' history 24 , pairings of cases and controls 34 , description of the method used to evaluate clinical outcomes 24 , description of intervention procedures 24 and lack of identification of confounding factors, as well as strategies for resolving these factors 30,34 .
There were also 12 studies identified with low risk of bias, although five studies presented methodological aspects that were not contemplated, such as: whether there was the inclusion of all participants of interest from the place where the research was developed 12,35 , clear description of the demographic characteristics of the participants 27 , results of speech and/or language perception before the intervention with the devices 25,32 , clearly reported clinical outcomes 35 , description of the reasons for the loss of participants throughout the period in which the research was developed, as well as strategies for adjusting the analysis of the outcomes, when patients did not participate in the entire study period 25,32 .
Although variability was observed in relation to sample size, types of studies, evaluation procedures used and methodological aspects, the results of this literature review demonstrated that CI and HA can be effective for the development of speech perception and spoken language in children with auditory neuropathy spectrum disorder, once complex levels of these skills were achieved, such as listening comprehension and speech intelligibility.Due to the heterogeneity of the studies, it was not possible to perform a meta-analysis.
Considering the specificities of ANSD, the continuity of research that seeks to evaluate the speech perception and spoken language of this population, with standardized methods and longitudinal character, may add information about the effectiveness of hearing aids and CI and contribute to the intervention in this pathology.

CONCLUSION
Considering the heterogeneity of the auditory neuropathy spectrum, over 25 years of research, 15 studies were verified suggesting that CI and HA may be effective for the development of speech perception and spoken language of children with auditory neuropathy spectrum disorder with no other associated impairments.The need for studies with high methodological rigor that demonstrate the effectiveness of the intervention with the devices (HA and/or CI) in the Source: Page et al. (2020)17

Figure 1 .
Figure 1.Diagram of systematic review results

Chart 3 .
Risk of bias of included studies

36
Were the criteria for inclusion in the sample clearly defined?Y 2. Were the study subjects and the setting described in detail?Y 3. Was the exposure measured in a valid and reliable way?Y 4. Were objective, standard criteria used for measurement of the condition?Y 5. Were confounding factors identified?NA 6. Were strategies to deal with confounding factors stated?NA 7. Were the outcomes measured in a valid and reliable way?