Predictors of language and auditory skills in Egyptian children with a cochlear implant

Aim The aim of this study was to investigate the prognostic factors contributing toward the comprehension and production of Arabic language as well as auditory abilities after cochlear implantation of Egyptian children. Participants and methods A total of 150 children with severe to profound sensory neural congenital hearing loss were implanted with multichannel cochlear implants. Children had received their implants 3 months to 2 years before the onset of the study. They had received their implants between the age range of 3 and 6 years. Tests of receptive, expressive language, and sets of auditory abilities were administered. Characteristics of the child and the family (age before implant and duration of implant use, preimplant use of hearing aids and language therapy, parents′ involvement in therapy, mode of communication at home, type of the implant, and geographic distribution) were considered the predictors. Predictors of total language and auditory abilities were determined using statistical analysis by univariate and multivariate analyses. Results Some significant predictors of language and auditory skills in this study included duration of implant use, use of auditory mode of communication and parent interaction. There is a positive correlation between language age and both individual and total auditory abilities scores. Conclusion Prognostic factors for language and auditory abilities are useful in a rehabilitation program after cochlear implantation.


Introduction
Cochlear implant systems comprise an externally worn microphone and a microprocessor programmed to extract intensity, frequency, and timing cues from acoustic signals. Th e system transforms these acoustic cues into an electrical code. Internally, a surgically placed receiver relays the transmitted code to an implanted array of contacts in the cochlea to stimulate surviving auditory neuron [1]. With experience, children understand speech, environmental sounds, and music with varying degrees of success [2][3][4][5].
Young children who experience severe to profound sensorineural hearing loss ( SNHL) face challenges in developing spoken language because of an inability to detect acoustic-phonetic cues that are essential for speech recognition, even when fi tted with traditional amplifi cation devices (hearing aids). More than half of such children are treated with cochlear implantation [6].
Cochlear implantation is associated with signifi cant improvements in comprehension and expression of spoken language over the fi rst 3 years of implant use. Infants and toddlers who receive implants show rapid improvements in auditory skills during the fi rst year of device use irrespective of age at implantation, although younger children achieve higher scores. Children who undergo implantation at a younger age acquire auditory skills closer to those of their peers with normal hearing (at a younger age) [7].
Much of the recent research in pediatric cochlear implantation is focused not only on documenting language achievements in this population but also determining what factors might infl uence the outcome. Documented reasons for poor performance include late age of implantation, poor nerve survival, inadequate fi tting, insuffi cient cognitive skills, educational

Postoperative assessment of history
Th e history was assessed with a particular focus on the following variables: (1) Age before implantation.
(3) Preimplant use of hearing aids and language therapy [type and duration (0 = no hearing aid use or language therapy; 1 = use hearing aid for 6 months or less; 2 = use hearing aid more than 6 months)]. (4) Postoperative means of communication at home (auditory, gestural, or both). (5) Maternal involvement in therapy, the duration of time that the mother spends talking to the child (0 = mother does not talk with the child, 1 = mother talks throughout daily activities, 2 = mother talks throughout daily activities in addition to structured therapy with her child). General examination, vocal tract examination, ear and nose examination, and neurological examination were performed. and social environments emphasizing manual communication, and limited parental support [8].
Th is study was carried out to gain a better understanding of the prognostic factors that promote spoken language development and auditory abilities after cochlear implantation in Egyptian children; these predictors help in the implementation of a rehabilitation program for better language and learning outcome after cochlear implantation, and help these children develop language normally and to be enrolled in mainstream education.

Aim of the study Aim
Th e aim of this study was to investigate prognostic factors contributing toward the comprehension and production of Arabic language as well as auditory abilities after cochlear implantation in Egyptian children.

Participants and methods
In all, 150 cochlear implant (CI) children with preoperative bilateral congenital severe-to-profound hearing loss were included in this study. Th e children were implanted with multichannel CI at Wadi El Neel Hospital as a part of CROP (Cochlear Implant Rehabilitation Outreach Program) and were seen during their routine postimplant follow-up.
Full insertion of the electrode array was performed in all the children using three diff erent implant systems: Medel Sonata (Austria) with Opus II processor, Cochlear Freedom (Australian) CI system, and Advanced Bionics (American) HiRes 90k with a Harmony processor. Th e children had received their implants for periods of 3 months to 2 years before the onset of the study. Th ey received their implants in age range between 2 and 6 years. Th e children were from diff erent Egyptian governorates. Th ey were classifi ed according to the geographic distribution of their area of residence into four classes. Th e study was carried out from the period from October 2013 to February 2014.
All the children were subjected to the protocol of cochlear implant assessment applied in Kasr Al Aini Hospital in addition to some auditory tasks from the EARS Assessment Program [9].

Preoperative selection
Patients were selected according to the following criteria.
(4) Familiar sentence repetition: Ten familiar words were spoken to the child and he/she was asked to repeat it correctly: (with a total score of 10). (5) Language testing was performed using the Arabic language test [12]: Each child was assigned receptive, expressive, and total language ages at the time of examination (postoperatively).
Data were collected, tabulated, and analyzed statistically.

Statistical analysis
Precoded data were statistically analyzed using the Statistical Package of Social Science Software program ( SPSS), version 21. Data were summarized using frequency and percentage for qualitative variables or mean and SD for quantitative variables. Comparison between groups was performed usin g the 2 -test or Fisher's exact test for qualitative variables, or independent-sample t-test or one-way analysis of variance (with post-hoc Tukey's test) for quantitative variables. Pearson or Spearman correlation coeffi cients were calculated to signify the association between quantitative and ordinal variables, respectively. Linear regression models were constructed to obtain the signifi cant predictors of auditory and language scores. P values less than 0.05 were considered statistically signifi cant and those less than 0.01 were considered highly signifi cant.

Results
Th is study was carried out on 150 Egyptian children with cochlear implants; language and auditory assessments were performed to determine the predictors for better outcome using univariate and multivariate analyses.

Univariate analysis of predictors showed the following:
(1) Non signifi cant predictors (P value <0.05) for language and auditory skills included: sex, age, type of implant, and duration of preoperative therapy (Table 1).
(2) High signifi cant predictors (P > 0.01) for language and auditory abilities: duration of implant use and high participation of mothers in therapy (always talks to her child).
(3) For residence, the best predictor was residence in Cairo, followed by residence in the Delta and residence in upper Egypt. (4) For mode of communication at home, the best predictor was the auditory mode, followed by mixed auditory and gestural modes, and gestural mode of communication.
Local examination of the scar of the implant was performed and the device was checked.
An adaptation of auditory skills checklist [11], which is an assessment tool that uses a combination of a structured parent interview and clinician observation to obtain information about functional auditory skills development, was used; this checklist includes the following: (1) Auditory detection and localization.
All abilities were assigned a score of 0 = does not have the skill, 1 = emerging skill development, and 2 = consistently shows the skill (the total score is 22).

Auditory abilities were assessed from Arabic modi cations of tests by the EARS (Evaluation of Auditory Responses to Speech) test [9]
(1) Monosyllabic word discrimination/identifi cation: Five monosyllabic words pictures are introduced to the child; he/she is asked to repeat and/or point to the correct picture ( 1 point for repeating and 1 point for pointing). Th e child was assigned a score out of 10.
(2) Polysyllabic word discrimination/identifi cation: Five polysyllabic words pictures are introduced to the child; he/she is asked to repeat and/or point to the correct picture (1 point for repeating and 1 point for pointing). Th e child was assigned a score out of 10.
(3) Open set word discrimination/identifi cation: Five monosyllabic words were spoken to child without any cues and the child was asked to guess the word spoken by the examiner once; he/she was assigned a score (out of 10) according to the number of correct sounds and words.
(2) Female sex and duration of preoperative language therapy (>6 months) were highly signifi cant in terms of language output (P > 0.01).
Correlations between total language age with both total and individual auditory skills were highly signifi cant (P > 0.01) (Table 3).

Multivariate analysis of predictors showed the following
(1) High signifi cant predictors for language and auditory abilities were duration of CI use, optimal level of care by mothers, and auditory mode of communication at home (P > 0.01) ( Table 2).  of 5 years was not a signifi cant predictor of language performance as children between 2 and 3 years of age did not show a signifi cant advantage over those who had received implantation between the ages of 4 and 5 years.
As regards gender diff erence in language outcome, it was found that girls have better language outcome than boys, and this is evidenced that girls show verbal advantage over boys in both hearing [17] and hearingimpaired populations [18].
Preoperative fi tting of hearing aids and regular language therapy over months were found to be signifi cant predictors for auditory abilities; duration of language therapy of more than 6 months is a signifi cant predictor for language outcome (Table 2). Th ese fi ndings are in agreement with those of a study that reported that preimplantation performance across candidates provided an estimate of a language-learning trajectory for children with severe to profound SNHL and without cochlear implants [19].
Auditory mode of communication with the child was also found to be the best one, followed by mixed auditory and gestural communication than just the gestural mode of communication ( Table 2). Use of a visual (i.e. sign) language system did not provide the linguistic advantage that had been anticipated. Children educated without the use of signs showed a signifi cant advantage in their language as a study showed that children in oral programs (including both auditory-verbal and traditional oral approaches) performed signifi cantly better on a battery of speech perception tests than did those children using total communication [20]. Geers and Moog [21] found that those in oral programs showed better speech and auditory skills than those in TC settings. However, Miyamoto et al. [22] described the speech perception results of 19 prelinguistically deafened children and found that communication mode did not play a signifi cant role in speech-perception abilities.
Maternal involvement in therapy was found to be a signifi cant predictor (Table 2) as language exposure and caregivers' mentoring provides the context for language learning. Spencer [23] examined diff erent behavioral indicators of parental involvement related to their children's education and development before and after cochlear implantation. Findings indicated an association between high levels of parental involvement, for example, learning sign language, advocating for their child's needs, devoting time and eff ort to take their child to the cochlear implantation clinic for follow-up, and monitoring children's language achievement. DesJardin et al. [24] found that mothers' higher sense

Discussion
Th e children included in this study represented a broad spectrum of characteristics. Children of both sexes had been using their implants for periods of 3 months to 2 years. Th ey had received their implants at age ranges between 3 and 6 years. Th eir preoperative language and auditory levels diff ered according to whether they had received early hearing aids and preoperative therapy; mode of communication at home diff ered according to the care and awareness of parents. Patients from diff erent Egyptian governorates were represented; all these variables were represented in the study as predictors that helped determine whether these are valuable in the prognosis of language and auditory abilities in CI children or not and to which extension as regard each variable alone (univariate analysis) or with regards of other variables (multivariate analysis).
Duration of CI use was found to be signifi cant in both univariate (Table 1) and multivariate analyses (Table 2) for total language and auditory skills. Th is is in agreement with a previous study that indicated the importance of duration of use of cochlear implants when evaluating benefi ts of the device [13]. Magmata et al. [14] reported that length of implant use accounted for the greatest variance in performance among a group of 61 children with implants.
However, age of implantation was found to be a signifi cant predictor in prognosis in other studies [7]. Early implantation may take advantage of neuronal fl exibility inherent in critical periods of auditory-based learning [15], but in this study, it was found to be a nonsignifi cant predictor (Table 1). A study by Geers et al. [16] reached the conclusion that children who had received implantation early (before the age of 5 years) achieved language skills close to their normally hearing peers after 4-7 years of use of this implant and also found that age of implantation in the group of children who had received implantation before the age Correlations between total language age with individual auditory skills were highly signi cant for all skills; The correlation between total language age with the total auditory score was signi cant (r = 0.824 and P < 0.001).
of involvement was associated with mothers' enhanced language facilitation strategies and their children's improvement in language abilities.
Maternal engagement in early communication refl ected in higher scores of parent-child interactions was associated with increased development in spoken language skills. Language comprehension and expression are infl uenced by parent-child interactions in bidirectional spoken communication [25,26].

Conclusion
Signifi cant predictors of language and auditory abilities including duration of implant use, auditory mode of communication, parents' involvement in therapy, and optimal level of parental communication with the child showed a signifi cant advantage in prognosis. Th ere is a positive correlation between language age and both individual and total auditory abilities' scores. Th ese predictors should be considered during rehabilitation to achieve better language and learning outcomes.