International Journal of Pediatric Otorhinolaryngology 76 (2012) 731–739

Differences between electrically evoked compound action potential (ECAP) andbehavioral measures in children with cochlear implants operated in the school agevs. operated in the first years of life

Sanja Vlahovic a,*, Branka Sindija a, Ivana Aras a, Matko Gluncic b, Robert Trotic c

a Polyclinics for Rehabilitation of Hearing and Speech SUVAG, Ljudevita Posavskog 10, HR-10000 Zagreb, Croatiab University of Zagreb, Department of Physics, Faculty of Science, Bijenicka 32, HR-10000 Zagreb, Croatiac Department of ENT and Head and Neck Surgery ‘‘Sestre milosrdnice’’ University Medical Centre, Vinogradska cesta 29, HR-10000 Zagreb, Croatia

A R T I C L E I N F O

Article history:

Received 28 November 2011

Received in revised form 9 February 2012

Accepted 11 February 2012

Available online 6 March 2012

Keywords:

Cochlear implant

ECAP

Behavioral measures

Age at operation

A B S T R A C T

Objective: The aim of this study was to identify the differences in the NRT measures, behavioral measures,

and their relationship between the group of congenitally deaf children operated in the first years of life

and the group of children operated in the school age.

Methods: The study included 40 congenitally deaf children with cochlear implants divided into two

groups. Group 1 was composed of 20 children (mean age at operation 2.3 years, range 1.4–4.6 years) and

Group 2 was composed of 20 children (mean age at operation 11.3 years, range 7.0–17.1 years). The ECAP

was recorded using the Nucleus 24 neural response telemetry (NRT) system. In each child, the responses

were evoked by the apical, middle and basal electrodes. The analyzed parameters were: the ECAP

threshold (T-NRT), N1P2 amplitude, N1 latency, slope of the amplitude growth function, response

morphology, threshold (T-) level, maximum comfort (C-) level, dynamic range (DR), T-NRT as a

percentage of the map DR, the correlation between the T-NRT and the T- and C-levels. The recordings of

parameters were performed two years after implantations.

Results: The T-NRT, DR, T-NRT as a percentage of the map DR and the correlation between T-NRT and C-

levels were significantly different between both groups of children. There were no statistically significant

differences between the groups with respect to the amplitude, latency, slope and morphology recorded

using the same electrodes. However, intragroup differences regarding NRT measures and behavioral

measures with respect to the position of stimulating electrode were more prominent in Group 2 than in

the Group 1.

Conclusions: Results of this study have also found a great variability of NRT and MAP measures within and

across patients in both groups of children, but it was still more pronounced in the group of school

children. NRT profile across electrodes follows MAP profiles better in the Group 1 then in the Group 2.

Overall findings of NRT and MAP measures are not consistent and unambiguous as we expected, but still

suggest potential differences between results in children operated in first years of life, and those operated

in school age.

� 2012 Elsevier Ireland Ltd. All rights reserved.

Contents lists available at SciVerse ScienceDirect

International Journal of Pediatric Otorhinolaryngology

jo ur n al ho m ep ag e: ww w.els evier . c om / lo cat e/ i jp o r l

1. Introduction

There is much evidence that there is a critical period for thedevelopment of hearing and spoken language [1–4]. Sensory inputduring critical periods of childhood is imperative for normaldevelopment of all sensory pathways, including hearing [5]. Theloss of hair cells in the cochlea leads to the loss of spiral ganglioncells (SGC) and to morphological and physiological changes in thecentral hearing pathways [5–8]. The pattern and degree of SGC

* Corresponding author. Tel.: +385 1 4629715; fax: +385 1 4629789.

E-mail address: svlahovic@suvag.hr (S. Vlahovic).

0165-5876/$ – see front matter � 2012 Elsevier Ireland Ltd. All rights reserved.

doi:10.1016/j.ijporl.2012.02.037

degeneration varies among pathological cochleae and depends onmultiple factors [9,10].

There is also evidence that auditory system of congenitallyhearing impaired children retains its plasticity [4,6]. Restoringauditory input early during developmental period by electricalstimulation could prevent degenerative consequences of deafnessto the extent determined by the duration of deafness [3,11,12], butthere are also other preoperative factors that influence theeffectiveness of electrical stimulation and adaptation to cochlearimplant (CI) such as etiology, extent of residual hearing,preoperative hearing stimulation, additional difficulties [3,7,9–17]. Additionally, evidence suggests that the early identification ofhearing impairment, successful use of hearing aids (HA) and

S. Vlahovic et al. / International Journal of Pediatric Otorhinolaryngology 76 (2012) 731–739732

hearing stimulation and rehabilitation may also prevent neuronalatrophy and, to a limited extent, activate the hearing pathways[3,7,11,19–21].

In a CI user each CI electrode interfaces to nearby neurons in aparticular way and the quality of the interface is also expected to be arelevant factor depending on the position and surrounding of theelectrode within the cochlea, the spread of the electrical current tothe neurons and the local neural survival pattern [22]. The extent ofcochlear pathology including peripheral dendrites and SGCincreases with duration of deafness [23]. The importance of SGCsurvival for CI outcome is significant [9,13,24–26], but hard toexamine independently of differences in the central nervous system.

Cochlear implantations are nowadays optimal treatment forprofoundly hearing impaired population, but additionally theyallow performing objective (ECAP, EABR) and behavioral (map T-and C-levels, dynamic range) measurements. The results of theseprocedures enable investigations of hearing pathways of CI users[27–31].

When the CI program began in Croatia in 1996, HealthInsurance Company did not routinely fund the program. Becausemany deaf children of various ages were waiting to undergo CIsurgery, a large humanitarian project was organized to raisemoney for the devices. As a result, among children who receivedcochlear implants within a short period of time, many underwentthe procedure in the school age. Given that congenitally deafchildren of various duration of deafness underwent implantationwithin a few months by the same surgical team, with the sametechnique, and with the same device implanted it made study ofthe influence of age at operation on behavioral and electrophysio-logical measure possible.

For the purpose of this study children were divided in twogroups; Group 1 is composed of children who received CI in thefirst years of life and the Group 2 is composed of children whoreceived CI in the school age. Behavioral (MAP levels) and neuralresponse telemetry (NRT) measures were collected, analyzed, andcompared. MAP levels reflect the anatomical and physiologicalstatus of the cochlea, but also auditory processing along the centralauditory pathways, and as a behavioral measure, these parametersare influenced by the cooperation and cognitive maturation[32,33].

NRT (i.e., recording the electrical compound action potential(ECAP) of the auditory nerve) allows the measurement of theresponse of SGC; however, a response from such cells does notdemonstrate central perception [32,33]. It is the most directmeasure of auditory nerve activity in cochlear implant users [34].Many previous studies have shown that ECAP thresholds fall insidedynamic ranges, what made that procedure useful in programmingyoung children [35,36]. According to some authors these thresh-olds usually fall within the same proportion of a subject’s dynamicrange across electrodes [33,35,37]. There are also studies thatshowed that relationship between ECAP thresholds and map levelsare not simple and uniform [38,39]. However, majority agree thatprogramming could not be based solely on ECAP thresholds, butfind NRT useful in programming especially young children [32,40–44]. Because electrically evoked potential induced by NRT provideslocation-specific information on the psychological status of theauditory nerve fibers, NRT could also be used as a diagnostic tool toevaluate status of the peripheral auditory neurons [27,28,30].

Several studies have reported a correlation between thepsychophysical and electrophysiological measures and the neuro-nal density in the vicinity of a stimulating electrode [18,24,26].According to some other studies, the correlation between ECAPthresholds and T- and C-levels is a potential indicator of successfulstimulation i.e., stimulation that results in perception [32].Furthermore, previous studies have shown that there are uniquecharacteristics of the NRT response, such as shorter latencies,

bigger amplitudes, steeper ECAP growth functions, lower thresh-olds and type 2 morphology (double-positive peak responses), thatcan predict better neural responsiveness that could mean betterneural survival or greater synchrony within the population ofneurons [7,27,29–31,38,43,45–47].

Moreover, lower MAP levels, wider dynamic ranges and lowerT-NRT levels within a dynamic range were also shown to beindicators of good neural responsiveness [13,29,32,33,44,45,48–51].

Our hypothesis was that negative effects of auditory depriva-tion on hearing pathways, including peripheral neural survival inall cochlear segments, are more prominent in children with longerduration of deafness, and that they caused differences in the MAPand NRT parameters between two groups of children operated indifferent age. Thus, the aim of this study was to identify thedifferences in the NRT measures and behavioral measures betweenthe group of congenitally deaf children operated in the first years oflife and the group of children operated in the school age. Intergroupand intragroup (basal vs. middle vs. apical electrodes) differencesin NRT and behavioral measures and their relationship wereinvestigated.

2. Methods

2.1. Subjects

Forty prelingually deaf children who received the Nucleus 24 CIsystem and used the implant for a minimum of 24 months wereenrolled in the study. They were included in the study only if bothNRT and behavioral measurements were available. Taking intoaccount findings about sensitive periods children were divided intotwo groups according to the age at which they received the implant.According to Sharma et al. a sensitive period of 3.5 y is the period ofmaximal plasticity. In some, but not all children, plasticity remainsuntil the age of 7 [11], therefore Group 1 was composed of 20children (mean age 2.3 y; range 1.4–4.6 y) and Group 2 wascomposed of 20 children (mean age 11.3 y; range 7.0–17.1 y).

All of the children were congenitally deaf. Preoperative hearingstatus of children from both groups was very similar; bilateral,symmetric, profound hearing loss with residual hearing in lowfrequencies (Table 1).

In some of the children, we confirmed a genetic etiology (6children with confirmed genetic etiology in Group 1, 7 children inGroup 2); however, in the majority of them, the etiology wasunknown. Children with postmeningitic deafness or deafness ofother known causes, or risk factors, except genetic, were notincluded in the study. Most of the children received the Nucleus24 R (CS) CI, and some children in both groups received the Nucleus24 M or Nucleus 24 R (ST) CI. The selection of the device wasrandom, depending only on the availability of devices at the time ofsurgery. All of the surgeries were performed at the Department ofENT and Head and Neck Surgery ‘‘Sestre milosrdnice’’ UniversityMedical Centre in Zagreb, Croatia, by the same surgical team andusing the same techniques within 6 months period. In addition, adistribution of contour and straight electrodes, a factor that wasrelevant because the type of electrode array placed within the scalatympani may influence the results of this study [52] was equal inboth groups. Only children with normal preoperative imaging andfull insertions and activation of the electrode array were included.Although all aforementioned cannot guarantee identical positionsof electrode arrays, it increases the likelihood that arrays areconsistently placed within cochlea. Speech processors wereprogrammed at the Cochlear Implant Center of the SUVAGPolyclinic by the same experienced audiologist. All of the childrenwho participated in the study underwent preoperative Verbotonalrehabilitation [53–59].

Table 1Hearing thresholds (according to AEP) on frequency specific stimuli of 500 Hz, 1 kHz and click stimulus for each ear, for both groups (mean, minimum and maximum values,

standard deviations and standard errors).

Minimum Maximum Mean Std. error Std. deviation

Group 1 0.5 kHz 80 120 102.00 2.471 11.050

Right 1 kHz 90 120 118.00 1.556 6.959

click 120 120 120.00 0.000 0.000

0.5 kHz 90 120 102.50 2.161 9.665

Left 1 kHz 105 120 118.75 0.879 3.932

click 100 120 119.00 1.000 4.472

Group 2 0.5 kHz 90 120 105.00 2.351 10.513

Right 1 kHz 95 120 117.25 1.601 7.159

click 100 120 117.00 1.638 7.327

0.5 kHz 90 120 105.75 2.273 10.166

Left 1 kHz 95 120 118.00 1.423 6.366

click 100 120 118.00 1.376 6.156

S. Vlahovic et al. / International Journal of Pediatric Otorhinolaryngology 76 (2012) 731–739 733

Tables 2 and 3 show the characteristics of the subjects withrespect to gender, age at implantation, ear implanted and type ofimplant. All are programmed with fast stimulation rate ACE 900 or1200 except one child with 720 and one child with SPEAK strategyin the Group 2.

2.2. Evaluation procedures

The ECAP of the auditory nerve was recorded using NRTsoftware versions 3.0 and 3.1. We used the NRT technique with theforward masking paradigm, which has been extensively describedby multiple studies [29,32,35,43]. The responses to stimulationwere evoked through the apical (19), middle (11) and basal (5)electrodes in each child. The active recording electrodes wereseparated from the active stimulating electrodes by 2 positions inthe apical direction, and the MP1 electrode was used as thereference. We used a default stimulation rate of 80 Hz with anamplifier gain of 60 dB and 100 average sweeps and in some cases astimulation rate of 30 Hz or a gain of 40 dB was used. When reliablerecordings could not be obtained using these electrodes, adjacentelectrodes were used. The masker level was set at 10 current unitsabove the probe level, and the masker probe interval was set to500 ms. The default delay setting of 50 ms was used.

NRT recordings were started at a low level and when typicalwaveform was observed stimulus level was increased until loudestacceptable presentation level (LAPL) was achieved. Waves werevisually assessed and marked. Amplitudes and latencies were

Table 2Demographic and cochlear implant characteristics of the 20 study participants from

Group 1.

Subject Age at implantation Gender Ear Type of implant

1.1 2y 3mo m R CI 24 M

1.2 1y 9mo m R CI 24 M

1.3 2y 6mo m R CI 24 R(ST)

1.4 1y 6mo m R CI 24 R(CS)

1.5 2y 8mo m R CI 24 R(CS)

1.6 4y 7mo m R CI 24 M

1.7 1y 8mo f R CI 24 R(CS)

1.8 1y 11mo f L CI 24 R(CS)

1.9 1y 7mo m R CI 24 R(CS)

1.10 2y 6mo f L CI 24 R(CS)

1.11 2y 7mo m R CI 24 R(CS)

1.12 2y 4mo m L CI 24R(CS)

1.13 2y 1mo f R CI 24R(CS)

1.14 1y 5mo m R CI 24R(CS)

1.15 2y 2mo m R CI 24R(CS)

1.16 1y 5mo m R CI 24R(CS)

1.17 2y 8mo m R CI 24M

1.18 1y 5mo m R CI 24R(CS)

1.19 4y 3mo f R CI 24R(CS)

1.20 3y 3mo f R CI 24R(CS)

measured on LAPL. Stimulus level was then decreased until theresponse could no longer be visualized. Once all of the recordings wereobtained the amplitude growth function feature of the NRT softwarewas utilized to measure NRT slope. The responses to stimulation werecategorized into two main types depending on number of positivepeakes following the negative, Type 1 with single-positive peakresponse and Type 2 with double-positive peak response.

The behavioral measures of threshold and comfortably loudstimulation levels that we analyzed were based on those usedroutinely in MAPs, when EAP recordings were made. For childrenin Group 1 the T- and C-levels were obtained using playaudiometry in an ascending–descending fashion and a picturedloudness scale. In the youngest children, behavioral observationswere performed by an experienced audiologist. T-levels were beingset at the level that evoked a change in behaviour. C-levels werebeing set at the highest level that did not cause any negativebehavioral response. For children in Group 2 the T- and C-levelswere determined using an ascending–descending technique in 1-level increments and a loudness scale. The C-levels weresubsequently balanced across the electrodes.

All data for the NRT and the behavioral T- and C-levels werecollected from the same electrodes. The parameters that werecompared included the NRT threshold, N1P2 amplitude, slope of theamplitude growth function, N1 latency, response morphology, T- andC-levels, dynamic range, T-NRT as a percentage of the map dynamicrange (in which T-level represents 0% of range and C-levels represents100%), and correlation between T-NRT and T- and C-levels. All of the

Table 3Demographic and cochlear implant characteristics of the 20 study participants from

Group 2.

Subject Age at implantation Gender Ear Type of implant

2.1 10y 8mo f R CI 24R(CS)

2.2 9y 9mo m R CI 24R(ST)

2.3 8y 1mo m R CI 24R(CS)

2.4 11y 11mo f R CI 24R(CS)

2.5 12y m L CI 24R(CS)

2.6 16y 4mo m R CI 24R(CS)

2.7 12y 6mo m L CI 24R(CS)

2.8 10y 9mo f R CI 24R(CS)

2.9 13y 6mo m R CI 24R(CS)

2.10 7y f R CI 24M

2.11 15y 9mo f R CI 24R(CS)

2.12 11y 11mo m R CI 24R(CS)

2.13 17y 1mo m R CI 24R(CS)

2.14 12y 4mo f R CI 24R(CS)

2.15 11y 6mo m R CI 24R(CS)

2.16 10y 5mo f L CI 24R(CS)

2.17 12y f R CI 24R(CS)

2.18 8y 9mo f R CI 24M

2.19 7y 4mo m R CI 24M

2.20 7y 6mo m R CI 24R(CS)

S. Vlahovic et al. / International Journal of Pediatric Otorhinolaryngology 76 (2012) 731–739734

recordings were performed 2 years after surgery under the sameconditions. Intergroup and intragroup differences based on thespecific electrodes were also compared.

2.3. Data analysis

When comparing intergroup differences, a parametric twosample t-test for the comparison of two independent samples wasused if the data fit into a normal distribution. When the data were notnormally distributed, we performed a non-parametric Mann–Whitney test. To compare intragroup differences, a one way analysisof variance (ANOVA) was used when the data were normallydistributed, and a Kruskal–Wallis analysis of variance by ranks wasused when the data were not normally distributed. Correction formultiple testing was done by post hoc MANOVA, Tukey and Scheffetests. The strength of the straight line association between variableswas assessed using the Pearson’s correlation coefficient. P-values of�0.05 were considered statistically significant.

3. Results

3.1. Intragroup differences with respect to NRT parameters and

behavioral measures between electrodes

Table 4 shows the mean and median values for the T- and C-levels, T-NRT, ECAP latency, amplitude, slope, morphology,

Table 4Mean and median values for the T- and C-levels, Map dynamic range, T-NRT, T-NRT as a

significance level.

Group 1

Mean Median Min Max SD

T-level (CU)

Apical 141.8 144.5 122 157 10.1

Middle 141.5 142.5 122 157 9.8

Basal 145.0 146.5 126 158 9.0

C-level (CU)

Apical 186.1 185.5 170 202 8.3

Middle 187.0 190.0 166 203 9.8

Basal 189.3 189.0 175 204 7.5

Map dynamic range (CU)

Apical 44.3 44.5 26 57 8.2

Middle 45.5 45.5 32 55 7.2

Basal 44.4 44.5 25 59 8.02

T-NRT (CU)

Apical 180.5 177.0 166 206 11.0

Middle 183.4 185.5 135 211 16.2

Basal 183.7 182.0 166 206 9.5

T-NRT (% of dynamic range)

Apical 87.7 89.0 47 126 25.01

Middle 83.4 91.0 59.1 124 31.9

Basal 84.2 89.0 30 116 19.5

ECAP latency (ms)

Apical 0.3 0.31 0.22 0.42 0.05

Middle 0.31 0.32 0.2 0.38 0.04

Basal 0.33 0.32 0.27 0.37 0.03

ECAP amplitude (mV)

Apical 136.7 105.0 58 315 72.3

Middle 117.2 104.5 42 261 53.9

Basal 149.0 131.0 49 346 92

ECAP slope (mV/CU)

Apical 10.1 8.5 2.0 24 6.32

Middle 8.9 8.5 2.0 18 3.4

Basal 8.5 7.0 5 16 3.4

Group 1

Type 1 Type 2

ECAP morphology

Apical 47% 53%

Middle 72% 28%

Basal 82% 18%

dynamic ranges, T-NRT as a percentage of dynamic range in bothgroups of children, and significance level for intergroup compar-isons, while Table 5 shows significance level for intragroupcomparisons. A comparison of the electrodes in the children withinGroup 1 revealed that there were no statistically significantdifferences along the array with respect to T-NRT, amplitude,latency, slope, T- and C-levels, dynamic range, T-NRT as apercentage of the dynamic range, and correlation between theT-NRT and T- and C-levels on the apical and basal electrodes.However, we did observe differences with respect to theproportion of Type 2 morphologies, which was greater on theapical electrode than on the middle or basal electrodes, but theywere not statistically significant. Pearson’s correlation coefficientwas higher between the T-NRT and T-levels and for the T-NRT andC-levels on the middle electrode than on the basal and apicalelectrodes, and that was statistically significant.

A comparison of the electrodes in the children within Group 2revealed more variability along the array. The largest differenceswere noticed between the apical and basal part of the cochlea. T-NRT was lower on the apical than on the middle (p = 0.045) and onthe basal electrode (p = 0.007) (Fig. 1). Furthermore, the T-NRTvalues were lower on the dynamic range on the apical than on themiddle and basal electrode, but difference was statisticallysignificant between the apical and basal electrodes (p = 0.029)(Fig. 1). We also found that there were more Type 2 morphologieson the apical electrode than on the middle and basal electrodes, but

percentage of dynamic range, ECAP latency, amplitude, slope and morphology, and

Group 2 p

Mean Median Min Max SD

134.6 135.0 104 164 14.6 0.2

136.6 137.5 111 174 15.05 0.4

139.3 139.0 119 182 16.2 0.4

174.9 175.0 141 199 13.6 0.03

177.6 178.5 158 208 13.3 0.05

177.2 176.5 148 220 16.0 0.04

40.4 40.5 18 65 10.6 0.27

41.0 39.5 19 64 10.3 0.18

37.9 38.0 18 58 10.02 0.09

163.7 167.0 130 182 15.6 0.001

172.2 171.0 146 192 11.7 0.016

178.6 179.0 163 193 9.1 0.06

67.3 70.0 11 118 30.5 0.015

85.5 91.0 30 120 26.5 0.38

98.3 103.0 16 161 35.1 0.32

0.3 0.31 0.26 0.36 0.03 0.5

0.32 0.32 0.17 0.37 0.05 0.69

0.32 0.33 0.13 0.37 0.05 0.64

134.0 105.0 48 451 110 0.88

121.5 83.0 40 377 97.6 0.69

123.0 77.0 47 545 124.6 0.64

7.9 7.0 1.0 25.0 6 0.22

9.5 8.5 2.0 44.0 9.6 0.16

9.5 8.0 3.0 23 5.8 0.53

Group 2 p

Type 1 Type 2

72% 32% 0.35

80% 22% 0.99

79% 21% 1.00

120

130

140

150

160

170

180

190

200

Sti

mu

lati

on

Un

its

Electrode

apical middle basal

C – optimal age

TNRT – optimal age

T – optimal age

87%84%

Fig. 3. Mean T and C-levels, T-NRT values and T-NRT as a percentage of dynamic

range in the group of children who underwent surgery in the first years of life

(Group 1).

Table 5Significance level for intragroup comparisons.

p

Apical Middle 0.88

Group 1 Apical Basal 0.85

Middle Basal 0.99

T-NRT

Apical Middle 0.05

Group 2 Apical Basal 0.007

Middle Basal 0.08

Apical Middle 0.99

Group 1 Apical Basal 0.84

Middle Basal 0.76

T-NRT %

Apical Middle 0.24

Group 2 Apical Basal 0.03

Middle Basal 0.6

120

130

140

150

160

170

180

190

200

Sti

mu

lati

on

Un

its

Electrode

apical middle basal

C – school ageTNRT – school age

T – school age

67%

86%

103%

Fig. 1. Mean T and C-levels, T-NRT values and T-NRT as a percentage of dynamic

range in the group of children who underwent surgery at school age (Group 2).

S. Vlahovic et al. / International Journal of Pediatric Otorhinolaryngology 76 (2012) 731–739 735

this difference was not statistically significant. The correlationbetween the T-NRT and C-level was better on the middle and apicalelectrodes than on the basal electrode, but we only observed astatistically significant difference between the apical and basalelectrodes (Fig. 2).

3.2. Intergroup differences with respect to NRT parameters and

behavioral measures collected from the same electrodes

Figs. 1 and 3 show the mean T-and C-levels and the T-NRTvalues (ECAP threshold) obtained from each electrode in both

GROUP 2

apical

r = 0.74

TN

RT

C

140 150 160 170 200

150

160

170

180

190

130

140

180 190

Fig. 2. Correlation between T-NRT and C-levels on apica

groups. A comparison of the mean T-NRT values between groupsindicated that the T-NRT was lower in the apical (p = 0.001) andmiddle (p = 0.016) electrodes in Group 2 compared to Group 1,whereas there were no statistically significant differences betweenthe groups in the basal electrode. There were no statisticallysignificant differences between the groups with respect to theECAP latency recorded on the same electrodes between groups.There were also no statistically significant differences with respectto amplitudes, although the biggest difference was between basalelectrodes (mean amplitude in the Group 2 showed a tendency tobe smaller than in the Group 1). The largest slope was recorded onthe apical electrode in the children from Group 1; however, thisdifference was not statistically significant. In addition, the slopesrecorded on the middle and basal electrodes were also notstatistically significant between the groups.

Most of the Type 2 (double peak) morphologies were recordedon the apical electrodes in both groups, but the larger number ofthis type of morphology was obtained for children in Group 1(Table 4). However, the differences between the groups were notstatistically significant. The proportion of these two morphologytypes recorded on the middle and basal electrodes was similar inboth groups.

Fig. 4 shows the C-level values for both groups on electrodesthat represent the basal, middle and apical cochlear segments. TheC-levels were lower on all of the electrodes in children from Group2 when compared to children from Group 1, and these differenceswere statistically significant.

GROUP 2

basal

r = 0.33

TN

RT

C

150 160 190 210 220

170

180

190

160

170 180 200

l and basal electrodes for the children in Group 2.

Table 6Correlation (Pearson’s correlation coefficient) between T-NRT and T and C levels on

the apical, middle and basal electrodes for children in Groups 1 and 2.

r (Group 1) r (Group 2)

Apical T-NRT, T 0.097 0.4

T-NRT, C 0.375 0.743

Middle T-NRT, T 0.571 0.391

T-NRT, C 0.741 0.679

Basal T-NRT, T 0.193 0.151

T-NRT, C 0.465 0.33

120

130

140

150

160

170

180

190

200

Sti

mu

lati

on

Un

its

Electrode

apical middle basal

C – optimal age

C – school age

TNRT – optimal ageTNRT – school age

T – optimal age

T – school age

Fig. 4. C-levels in both groups of children.

S. Vlahovic et al. / International Journal of Pediatric Otorhinolaryngology 76 (2012) 731–739736

The MAP dynamic range values showed a tendency to be wideron all electrodes especially on basal (p = 0.09) in children fromGroup 1 than in those from Group 2, but these differences were notstatistically significant. The relationship between T-NRT and T- andC-levels in both groups of children are shown in Figs. 1 and 3. Themean T-NRT values expressed as a percentage of the dynamicrange were not significantly different for the middle and basalelectrode, but they were significantly different for the apicalelectrode (p = 0.015). It is important to note that, for the children inGroup 1, the value for T-NRT as a percentage of dynamic range wason the apical electrode higher than for the children in Group 2, buton the basal electrode it was lower than for the children in Group 2.

The correlation between the T-NRT and C-levels was better thanthe correlation between the T-NRT and T- levels in both groups(Table 6). However, the correlation between the T-NRT and T-levels was only significant on the middle electrode in both groupsof children, with Group 1 having a higher r-value (Pearson’scorrelation coefficient). According to Pearson’s correlation coeffi-cient, the correlation between the T-NRT and C-levels on the apicalelectrode was significantly higher in the children from Group 2(r = 0.74) than in those from Group 1 (r = 0.38) (Fig. 5). In addition,the correlation between the T-NRT and C-levels on the middleelectrode was good in both groups, with Group 1 having asignificantly better correlation.

4. Discussion

There are numerous studies of NRT. Many of them wereperformed with adult subjects i.e., in groups of postlingually deaf

GROUP 1

r = 0.38

TN

RT

C

170 180 190 200 210

170

180

190

200

210

160

Fig. 5. Correlation between T-NRT and C-levels on the apical e

adults with various, mostly acquired, deafness etiologies[34,35,38,44,46,52,60]. Some other studies, in accordance withthe design of the studies, included subjects of wide range of ages –children and adults together [29,31,61]. Some studies includedprelingually deaf children of various ages [32,33,36,39] and someincluded only young children [37,43,47,62]. There are also studiesthat compared results of children vs. adults. Brown et al. found outsteeper ECAP growth function for children than for adults [31] andHughes et al. found higher C-levels in children than in adultsprobably as a consequence of greater distance between electrodeand the activated nerve fiber then in adults [45]. The possibleexplanation for that finding is that children may have fewerdendritic processes than postlingually deaf adults and/or developsmore fibrous tissue within the cochlea after operation than adults.There are also data that excessive scarring after surgery is morecommon in children and adolescents than in adults [45].

Some of the findings of the studies performed with adults areconsistent with a pathogenesis that is specific for a particularetiology. For example, high T-NRT in patients with otosclerosis andpostmeningitic deafness is attributed to pathological bone growthand, in the case of meningitis, to a lower number of vital, functionalneurons [29,52].

Several studies have reported higher thresholds and smallerECAP amplitudes in the basal electrodes than in the apicalelectrodes [29,36,38,43,60]. The dominant paradigm to explainthese findings has been that there is a smaller number of preservedvital neurons in the basal part of the cochlea, which is consistentwith postmortem pathohistological studies of the temporal bones[9,17,29,33,38,43,46]. An alternative explanation is that theelectrode array in the basal part of the cochlea is more distantfrom SGC or that there is a higher risk for surgical trauma in thatpart of the cochlea [29,60]. Furthermore, several studies havereported variability in the ECAP thresholds, amplitudes, morphol-ogies, and MAP levels along the array, indicating that the spatialdistribution of neuronal populations stimulated by a particularelectrode is heterogeneous [24,38]. The impact of a particularapical–basal profile and the survival of peripheral nerves on the

GROUP 2

r = 0.74

TN

RT

C

140 150 170 180 200

150

160

170

180

190

130

140

160 190

lectrode for the children in Group 1 and those in Group 2.

S. Vlahovic et al. / International Journal of Pediatric Otorhinolaryngology 76 (2012) 731–739 737

effectiveness of cochlear implants is still unclear [18], but seems tobe important for providing optimal stimulation parameters alongthe array, and is expected to affect listening development aftercochlear implantation [17].

In both groups of children that we evaluated there were also agreat variability of results between subjects and along the array,but in the Group 1 comparison of average NRT and MAP valuesbetween electrodes, even apical and basal, did not revealsignificant differences. Multiple studies have underscored thatchildren implanted during the first years of life have a betterchance of developing hearing and speech than those implanted inolder age [12,63–66]. Because of less differences regarding NRTmeasures and behavioral measures with respect to the position ofstimulating electrode observed in young children maybe it can beassumed that such finding also reflects better conditions fordevelopment of hearing and speech with CI. In the group of school-age children, we observed more variability along the array, whichincluded differences with respect to T-NRT and the relationshipbetween the T-NRT and map levels. Some studies performed inadults indicated that the pronounced variations in behavioralthresholds among stimulation sites, which were attributed tovariability in the neuropathology along the length of the cochlea,are an unfavorable factor for speech recognition [51,67,68].

Due to longer deprivation of auditory stimulation, we expectedto find more indicators of poor neuronal survival in all cochlearsegments in the Group 2. According to previous studies thoseindicators would be: higher T-NRT values, lower ECAP amplitudes,longer ECAP latencies, decreased B morphologies, narrower DRvalues, higher MAP levels, lack of correlation between T-NRT andT- and C-levels [7,13,27,29,32,33,38,43,45,51].

In this study latencies were similar in both groups of children inall parts of the cochlea which was not unexpected becauseprevious studies have shown that latencies are the most stablevalues, likely due to direct electrical stimulation [29,47].

There were also no significant intergroup differences betweenECAP amplitudes. ECAP amplitudes were equal along the array forchildren in Group 1 while for children in Group 2 they were higheston the apical electrode although not statistically significant, but wealso noticed extremely large dispersion of results in that group ofchildren that complicates the interpretation. According to previousstudies, amplitudes reflect the sum of the activities of neurons (i.e.,the number of neurons that react to a stimulus), which in turnindicate better preservation of neuronal function [29,30,46,47].Indication for better preservation of neuronal function is also lowT-NRT values [7,24,29,46]. In our study T-NRT was significantlylower in the apical and middle electrodes in Group 2 compared tochildren from Group 1. The lowest T-NRT value was observed onthe apical electrode for children in Group 2. Therefore, these resultsindicate that the neurons with the best function in the childrenfrom Group 2 are located in the apical area of the cochlea and thattheir function is comparable, or might even be better then in theGroup 1. There is no simple and univocal explanation for theseresults. But if we take in consideration some facts; residual hearing(in the low frequency field) in children from Group 2 was intenselystimulated for years with hearing aids and electroacoustic filtereddevices, second; among deaf school age children only those whodeveloped a certain level of speech discrimination and under-standing through such stimulation were considered as candidatesfor CI surgery, and third; auditory stimulation is a trophic factor forneurons along the hearing pathways [19], therefore we mayconsider whether chronic stimulation of low-frequency cochlearregion during the preoperative period could have contributed toimproved neuronal survival and/or better neuronal synchrony ofthe peripheral neurons in apical part of cochlea? Or, in other words,did preoperative rehabilitation, activation and development ofcentral auditory pathways contributed to functional maintenance

of spiral ganglion cells in apical part of cochlea? Nowdays, theimportance of the stimulation of low frequencies has beendemonstrated by the success of hybrid stimulation and thesignificance of preserving residual hearing in the apex usingatraumatic surgery [18].

In both groups we observed more often Type 2 morphologiescompared to previously published studies; however, most of thesestudies were performed on adult CI users [27,38]. Highestincidence of Type 2 morphology was observed on the apicalelectrodes in both groups, but still higher in the Group 1, althoughnot statistically significant (Group 1 53%; Group 2 32%). Severalexplanations have been proposed to account for Type 2 waveformmorphologies [27]. If we assume that degenerative effects due toauditory deprivation did not develop to a greater extent because ofbetter residual hearing and better auditory stimulation in that partof cochlea and additionally, in the case of children from Group 1,due to the short duration of their deafness, such a result wouldsupport the explanation that double peaks are the result ofimproved preservation of neurons i.e., preserved dendrites [27,38].This would be consistent with the expectation that dendrites arebetter preserved in children with short-lasting deafness, especiallyin the apical part of the cochlea, where better residual hearing islocated. Hence, Type 2 morphology could be an indicator of goodperipheral status at the time of surgery.

Behavioral measures as well as analysis of their relationship toT-NRT are influenced by a child’s age, cooperation, cognitive status,previous hearing experience, and the duration of time after thesurgery. Because some changes occur during the first months aftersurgery [31–33,45,47], or according to some authors [33,62] evenfirst years after surgery, this study was performed two years afterthe children underwent implantation, when their maps and ECAP-swere stable and the results of the behavioral measures were quitereliable even in the group of young children.

The C-levels in Group 1 were higher than those in Group 2 on allelectrodes. If it is assumed, as is commonly accepted [44,67] thatlarger nerve survival requires less current to reach C-level, thensuch finding would lead to the conclusion that there is a betterneural survival in children from Group 2. Another, moreacceptable explanation is that younger children require higherstimulus intensities for speech understanding [45], while childrenoperated in school age have a reduced ability to tolerate loudsounds due to a long period of auditory deprivation. Pfingst and Xuhave found that subjects with poor speech recognition tended tohave low mean C-levels and small mean DR. They assumed that apossible mechanism underlying such finding is saturation ofneurons at high stimulus levels in those subjects [51]. Previousstudies found out that C-levels are higher in congenitally deafchildren with CI than in postlingually impaired adult [45].According to our study there are differences in this parameteramong congenitally deaf children with CI depending on age atsurgery.

Additionally, wider DR and lower T-NRT as a percentage of DRindicate a better ability for discrimination and a higher tolerance ofloudness [29,32,51], and according to some studies, they are alsoindicators of better neuronal survival [29,51]. The majority ofprofoundly hearing impaired patients before implantation man-aged to hear only low frequencies. As a consequence, severalmonths after surgery the high pitched sound perception is stilllimited [32]. In the Group 2 MAP DR was narrower on allelectrodes, especially on basal. Although these differences did notmeet the desired level of significance (p = 0.05) it is reasonable topoint out the difference on basal electrode (p = 0.09) whichindicate that two years after surgery the dynamic range of thechildren in Group 2 still have a tendency to be narrower on thebasal electrode. Moreover, in the Group 1 mean T-NRT fell withinthe same proportion of DR across electrodes (it falls approximately

S. Vlahovic et al. / International Journal of Pediatric Otorhinolaryngology 76 (2012) 731–739738

around 85% of the DR), i.e. the MAP profiles parallel the NRT profile.Some authors found that NRT thresholds were typically obtained at91% of the DR across electrodes for adults, and at 53% of the DRacross electrodes for children [35,36]. However in the Group 2 thisparameter was significantly different with respect to electrode (itfalls on the 67% on apical, 86% on middle and on the 98% on basalelectrode).

In any case, it appears that even two years after performedimplantation, in children operated in school age tolerance toloudness was reduced, especially for the high frequency sounds.This finding should be taken into consideration when program-ming children who received CI at an older age especially becausethe use of inappropriate levels of electrical stimulation can resultin further loss of peripheral neurons [31]. In addition, this impliesthat even prior to school age deaf children suffer from irreversiblechanges that affect their ability to broaden dynamic range duringpostoperative period and their ability to perceive high-frequenciesdue to the lack of residual hearing in the high frequency region, aswell as the inability of hearing aids and electroacoustic devices tostimulate this part of the auditory system.

Previous studies reported that the good correlation between T-NRT and T- and C-levels indicates that stimulation of the nervefibers in the area examined activates the central auditorypathways i.e. hearing really occurs and that this transmissionimproves by time [32]. Thus, obtaining a good correlationbetween the T-NRT and C-levels along the array could also be agood prognostic factor for listening development with cochlearimplants. The best correlation were reported between the T-NRTand T- and C-levels in the apical part of the cochlea and with thetime improved correlation in T-NRT and T-and C-levels on thebasal and middle electrodes and the authors attributed suchimprovement to changes in the perception of high-frequencytones after surgery [32]. In our study in both groups of children weobserved a better correlation between the T-NRT and C-levelsthan between the T-NRT and T-levels. In addition, the weakestcorrelations in both groups of children were found in the basalparts of the cochlea; the best correlations were observed in theapical parts of the cochlea in the school-age children and in themiddle part of the cochlea in the younger group of children. Asalready mentioned, the high correlation indicates a goodtransmission between the periphery and cortex. In addition, thishigh correlation on the middle electrode also indicates thatoptimal hearing field of children who undergo surgery early in lifeis in the speech frequency range similar to what is observed innormal hearing children and that this occurs early duringpostoperative rehabilitation. In contrast, in the children fromGroup 2 who were preoperatively rehabilitated only throughattainable low-frequency hearing area and whose optimal hearingfield was widened towards middle frequencies over time, the bestcorrelation was observed on the apical electrode even two yearsafter cochlear implantation. That finding indicates that low-frequencies are still significantly involved in hearing in school-agechildren, whereas for younger children those frequencies are notof such significance.

However, after operation with better peripheral input thesechildren can progress better and faster, albeit with limitations,while their hearing pathways influenced over long preoperativetime by low-frequencies seem to still play an important role intheir hearing.

Overall differences in NRT and behavioral measures that wereidentified between two groups of children are not consistent andunambiguous as we expected, but still points to possibledistinctions in perceptual abilities which may be relevant forprogramming CI and users ability to benefit from electricalstimulation, and could be associated with age at surgery andpreoperative rehabilitation.

5. Conclusions

1. Results of this study are in agreement with previous studies thathave found a great variability of NRT and behavioral measureswithin and across patients, but still it was more pronounced inthe group of school children.

2. NRT profile across electrodes follows MAP profiles better in thegroup of children operated in the first years of life than in thegroup of school children.

3. Overall findings of NRT and MAP measures, as well as theirrelationship, are not consistent and unambiguous as weexpected, but still suggest potential differences between resultsin children operated in the first years of life and those operatedin the school age.

4. There is a need for more studies performed on congenitally deafchildren that would be divided into groups according to age atcochlear implantation and whose results would then becorrelated with progress in hearing development.

Acknowledgments

The authors would like to thank two anonymous reviewers fortheir very helpful and constructive comments which significantlyimproved the first version of the manuscript. This study wassupported by the Polyclinic for Rehabilitation of Hearing andSpeech SUVAG and Ministry of Science, Education and Sports of theRepublic of Croatia.

References

[1] J.J. Eggermont, G.R. Bock (Eds.),Critical periods in auditory development ActaOtolaryngol. Suppl. (Stockh.) 429 (1986) 1–64.

[2] P.W. Jusczyk, Some critical developments in acquiring native language sound organi-zation during the first year, Ann. Otol. Rhinol. Laryngol. Suppl. 111 (2002) 11–15.

[3] D.M. Hammes, M.A. Novak, L.A. Rotz, M. Willis, D.M. Edmondson, J.F. Thomas,Early identification and cochlear implantation: critical factors for spoken lan-guage development, Ann. Otol. Rhinol. Laryngol. Suppl. 111 (2002) 74–78.

[4] M.G. Gilley, A. Sharma, M.F. Dorman, Cortical reorganization in children withcochlear implants, Brain Res. 1239 (2008) 56–65.

[5] N.A. Hardie, The consequences of deafness and chronic intracochlear electricalstimulation on the central auditory pathways, Clin. Exp. Pharmacol. Physiol. 25(1998) 303–309.

[6] C.W. Ponton, M. Don, J.J. Eggermont, M.D. Waring, B. Kwong, A. Masuda, Auditorysystem plasticity in children after long periods of complete deafness, Neuroreport8 (1996) 61–65.

[7] A. Coco, S.B. Epp, J.B. Fallon, J. Xu, R.E. Millard, R.K. Shepherd, Does cochlearimplantation and electrical stimulation affect residual hair cells and spiral gan-glion neurons? Hear. Res. 225 (1–2) (2007) 60–70.

[8] A. Sharma, A.A. Nash, M. Dorman, Cortical development, plasticity and re-organi-zation in children with cochlear implants, J. Commun. Dis. 42 (2009) 272–279.

[9] J.B. Nadol Jr., Patterns of neural degeneration in the human cochlea and auditorynerve: implications for cochlear implantation, Otolaryngol. Head Neck Surg. 117(1997) 220–228.

[10] N.A. Hardie, R.K. Shepherd, Sensorineural hearing loss during development:morphological and physiological response of the cochlea and auditory brainstem,Hear. Res. 128 (1–2) (1999) 147–165.

[11] A. Sharma, M. Dorman, A. Spahr, N.W. Todd, Early cochlear implantation inchildren allows normal development of central auditory pathways, Ann. Otol.Rhinol. Laryngol. Suppl. 111 (2002) 38–41.

[12] K. Iler Kirk, R.T. Miyamoto, C.L. Lento, E. Ying, T. O’Neill, B. Fears, Effects of age atimplantation in young children, Ann. Otol. Rhinol. Laryngol. Suppl. 111 (2002) 69–73.

[13] A. Incesulu, J.B. Nadol Jr., Correlation of acoustic threshold measures and spiralganglion cell survival in severe to profound sensorineural hearing loss: implica-tions for cochlear implantation, Ann. Otol. Rhinol. Laryngol. 107 (1998) 906–911.

[14] R.K. Shepherd, N.A. Hardie, Deafness-induced changes in the auditory pathway:implications for cochlear implants, Audiol. Neurootol. 6 (6) (2001) 305–318.

[15] S. Vlahovic, B. Sindija, The influence of potentially limiting factors on paediatricoutcomes following cochlear implantation, Int. J. Pediatr. Otorhinolaryngol. 68(2004) 1167–1174.

[16] B. Sindija, S. Vlahovic, B. Kekic, Rezultati rane ugradnje umjetne puznice, Paediatr.Croat. Suppl. 49 (2) (2005) 65–68.

[17] K.A. Gordon, B.C. Papsin, R.V. Harrison, Auditory brainstem activity and develop-ment evoked by apical versus basal cochlear implant electrode stimulation inchildren, Clin. Neurophysiol. 118 (2007) 1671–1684.

[18] P.A. Leake, O. Stakhovsaya, G.T. Hradek, A.M. Hetherington, Factors influencingneurotrophic effects of electrical stimulation in the deafened developing auditorysystem, Hear. Res. 242 (1–2) (2008) 86–99.

S. Vlahovic et al. / International Journal of Pediatric Otorhinolaryngology 76 (2012) 731–739 739

[19] N.A. Hardie, L.M. Martsi-McClintock Aitkin, R.K. Sheperd, Neonatal sensorineuralhearing loss affects synaptic density in the auditory midbrain, Neuroreport 9(1998) 2019–2022.

[20] C. Yoshinaga-Itano, M.L. Apuzzo, Identification of hearing loss after age 18 monthsis not early enough, Am. Ann. Deaf 143 (5) (1998) 380–387.

[21] C. Yoshinaga-Itano, M.L. Apuzzo, The development of deaf and hard of hearingchildren identified early through the high risk registry, Am. Ann. Deaf 143 (5)(1998) 416–424.

[22] J.A. Bierrer, K.F. Faulkner, Identifying cochlear implant channels with poorelectrode-neuron interface: partial tripolar, single-channel threshodls and psy-chophysical tuning curves, Ear Hear. 31 (2) (2010) 247–258.

[23] R.K. Shepherd, S. Hatsushika, G.M. Clark, Electrical stimulation of the auditorynerve: the effect of electrode placement on neural excitation, Hear. Res. 66 (1993)108–120.

[24] A.M. Khan, D.M. Whiten, J.B. Nadol, D.K. Eddington, Histopathology of humancochlear implants: correlation of psychophysical and anatomical measures, Hear.Res. 205 (2005) 83–93.

[25] J.Y. Nadol, B.J. Shiao, D.R. Burgess, D.R. Ketten, D.K. Eddington, B.J. Gantz, et al.,Histopathology of cochlear implants in humans, Ann. Otol. Rhinol. Laryngol. 110(9) (2001) 883–891.

[26] A. Kawano, H.L. Seldon, G.M. Clark, R.T. Ramsden, C.H. Raine, Intracochlear factorscontributing to psychophysical percepts following cochlear implantation, ActaOtolaryngol. 118 (3) (1998) 313–326.

[27] W.K. Lai, N. Dillier, A simple two-component model of the electrically evokedcompound action potential in the human cochlea, Audiol. Neurootol. 5 (2000)333–345.

[28] F.G. Zeng, G. Grant, J. Niparko, Speech dynamic range and its effect on cochlearimplant performance, J. Acoust. Soc. Am. 111 (2002) 377–386.

[29] M.C. Guedes, R.V. Brito Neto, M.V. Gomez, S.B. Sant’Anna, C.G. Peralta, A.M.Castilho, Neural response telemetry measures in patients implanted with Nucleus24, Rev. Bras. Otorrinolaringol. 71 (5) (2005) 660–667.

[30] C.A. Miller, C.J. Brown, P.J. Abbas, S.L. Chi, The clinical application of potentialsevoked from the peripheral auditory system, Hear. Res. 242 (2008) 184–197.

[31] C.J. Brown, P.J. Abbas, C.P. Etler, S. O’Brient, J.J. Oleson, Effects of long term use of acochlear implant on the electrically evoked compound action potential, J. Am.Acad. Audiol. 21 (1) (2010) 5–15.

[32] H. Thai-Van, J.M. Chanal, C. Coudert, E. Veuillet, E. Truy, L. Collet, Relationshipbetween NRT measurements and behavioral levels in children with the Nucleus24 cochlear implant may change over time: preliminary report, Int. J. Pediatr.Otorhinolaryngol. 58 (2001) 153–162.

[33] Y. Henkin, R. Kaplan-Neeman, C. Muchik, J. Kronenberg, M. Hildesheimer, Changesover time in electrical stimulation levels and electrode impedances values inchildren using the Nucleus 24M cochlear implant, Int. J. Pediatr. Otorhinolaryngol.67 (8) (2003) 873–880.

[34] P.J. Abbas, C.J. Brown, J.K. Shallop, J.B. Firszt, M.L. Hughed, S.H. Hong, et al.,Summary of results using the nucleus CI24 M implant to record the electricallyevoked compound action potential, Ear Hear. 20 (1) (1999) 45–59.

[35] C.J. Brown, M.L. Hughes, B. Luk, P.J. Abbas, A. Wolaver, J. Gervais, The relationshipbetween EAP and EABR thresholds and levels used to program the nucleus 24speech processor: data from adults, Ear Hear. 21 (April (2)) (2000) 151–163.

[36] M.L. Hughes, C.J. Brown, P.J. Abbas, A.A. Wolaver, J.P. Gervais, Comparison of EAPthresholds with MAP levels in the Nucleus 24 cochlear implant: data fromchildren, Ear Hear. 21 (2) (2000) 164–174.

[37] H.A. Muhaimeed, F.A. Anazy, O. Hamed, E. Shubair, Correlation between NRTmeasurement level and behavioral levels in pediatrics cochlear implant patients,Int. J. Pediatr. Otorhinolaryngol. 74 (4) (2010) 356–360.

[38] D. Cafarelli Dees, N. Dillier, W.K. Lai, E. von Wallenberg, B. van Dijk, F. Akdas,Normative findings of electrically evoked compound action potential measure-ments using the neural response telemetry of the Nucleus CI 24M cochlearimplant system, Audiol. Neurootol. 10 (2005) 105–116.

[39] B.A. Holstad, V.G. Sonneveldt, B.T. Fears, L.S. Davidson, R.J. Aaron, M. Richter, et al.,Relation of electrically evoked compound action potential thresholds to behav-ioral T- and C-levels in children with cochlear implants, Ear Hear. 30 (1) (2009)115–127.

[40] J.K. Shallop, G.W. Facer, A. Peterson, Neural response telemetry with the nucleusCI24M Cochlear Implant, The Laryngoscope 109 (1999) 1755–1759.

[41] K.H. Franck, S.J. Norton, Estimation of psychophysical levels using the electricallyevoked compound action potential measured with the neural response telemetrycapabilities of Cochlear Corporation’s CI24M device, Ear Hear. 22 (4) (2001) 289–299.

[42] G.F. Smoorenburg, C. Willeboer, J.E. van Dijk, Speech perception in Nucleus CI24Mcochlear implant users with processor settings based on electrically evokedcompound action potential thresholds, Audiol. Neurootol. 7 (2002) 335–347.

[43] K.A. Gordon, K.A. Ebinger, J.E. Gilden, W.H. Shapiro, Neural response telemetryin 12- to 24-month-old children, Ann. Otol. Rhinol. Laryngol. Suppl. 111 (2002)42–48.

[44] J.E. King, M. Polak, A.V. Hodges, S. Payne, F.F. Telischi, Use of neural responsetelemetry measures to objectively set the comfort levels in the Nucleus 24Cochlear Implant, J. Am. Acad. Audiol. 17 (2006) 413–431.

[45] M.L. Hughes, K.R. Van der Werff, C.J. Brown, P.J. Abbas, D.M. Kelsay, H.F. Teagle,et al., A longitudinal study of electrode impedance, the electrically evokedcompound action potential, and behavioral measures in nucleus 24 cochlearimplant users, Ear Hear. 22 (6) (2001) 471–486.

[46] M. Polak, A.V. Hodges, J.E. King, T.J. Balkany, Further prospective findings withcompound action potentials from Nucleus 24 cochlear implants, Hear. Res. 188(1–2) (2004) 104–116.

[47] L.F. Tanamati, M.C. Bevilacqua, O.A. Costa, Longitudinal study of the ecap mea-sured in children with cochlear implants, Braz. J. Otorhinolaryngol. 75 (1) (2009)90–96.

[48] B.E. Pfingst, D. Sutton, J.M. Miller, B.A. Bohne, Relation of psychophysical data tohistopathology in monkeys with cochlear implant, Acta Otolaryngol. 92 (1–6)(1981) 1–13.

[49] A.L. Miller, Effects of chronic stimulation on auditory nerve survival in ototoxi-cally deafened animals, Hear. Res. 151 (2001) 1–14.

[50] B.E. Pfingst, L. Xu, Across-site variation in detection thresholds and maximumcomfortable loudness levels for cochlear implants, JARO 5 (2004) 11–24.

[51] B.E. Pfingst, L. Xu, Psychophysical metrics and speech recognition in cochlearimplant users, Audiol. Neurootol. 10 (2005) 331–341.

[52] L.G. Potts, M.W. Skinner, B.D. Gotter, Relation between neural response telemetrythresholds T- and C-levels, and loudness judgements in 12 adult nucleus 24cochlear implant recipients, Ear Hear. 28 (4) (2007) 495–511.

[53] P. Guberina, Verbotonal method and its application to the rehabilitation of thedeaf, in: Report of the Proceedings of the International Congress on Education ofthe Deaf and of the Forty-first Meeting of the Convention of American Instructorsof the Deaf, US Government Printing Office, Washington, (1964), pp. 279–293.

[54] A. Persillon, N. Badez, M. Dorey, G. Diebolt, Resume of rehabilitation with theSUVAG apparatus and by the verbotonal method of Guberina after 3(1/2) years ofuse at the Centre de la rue Pierre-Corneille, in Lyons, Rev. Laryngol. Otol. Rhinol.(Bord.) 86 (11) (1965) 1090–1107.

[55] C.W. Asp, P. Guberina, Verbo-tonal method for rehabilitating people with com-munication problems, in: The Verbo-Tonal Method for Rehabilitating People withCommunication Problems, World Rehabilitation Fund, International Exchange ofInformation in Rehabilitation, New York, 1981.

[56] C.W. Asp, The verbotonal method for management of young, hearing-impairedchildren, Ear Hear. 6 (1) (1985) 39–42.

[57] V. Mildner, B. Sindija, K.V. Zrinski, Speech perception of children with cochlearimplants and children with traditional hearing aids, Clin. Linguist. Phon. 20 (2–3)(2006) 219–229.

[58] C. Asp, Verbotonal Speech Treatment, Plural Publishing, San Diego, Oxford, 2005.[59] C.W. Asp, K.J. Koike, M. Kline, Breaking News:, Verbotonal rehabilitation: are we

doing enough? Hear. J. 65 (1) (2012) 28–34.[60] K. Pedley, C. Psarros, K. Gardner-Berry, A. Parker, S.C. Purdy, P. Dawson, et al.,

Evaluation of NRT and behavioral measures for mapping elderly cochlear implantusers, Int. J. Audiol. 46 (2007) 254–262.

[61] M.L. Hughes, J. Stille, Psychophysical and physiological measures of electrical-field interaction in cochlear implants, J. Acoust. Soc. Am. 125 (1) (2009) 247–260.

[62] R. Mittal, S.S. Panwar, Correlation between intra-operative high rate neuralresponse telemetry measurements and behaviourally obtained threshold andcomfort levels in patients using Nucleus 24 cochlear implants, Cochlear ImplantsInt. 10 (2) (2009) 103–111.

[63] T. Morita, Y. Naito, T. Nakamura, S. Yamaguchi, J. Tsuji, J. Ito, Chronologicalchanges of stimulation levels in prelingually deafened children with cochlearimplant, Acta Otolaryngol. Suppl. 551 (2004) 60–64.

[64] T. Lenarz, Cochlear implants: selection criteria and shifting borders, Acta Otorhi-nolaryngol. (Belg.) 52 (1998) 183–199.

[65] T.P. Nikolopoulos, G.M. O’Donoghue, S. Archbold, Age at implantation: its impor-tance in pediatric cochlear implantation, Laryngoscope 109 (1999) 595–599.

[66] M.J. Osberger, S. Zimmerman-Philips, D. Burton Koch, Cochlear implant candidacyand performance trends in children, Ann. Otol. Rhinol. Laryngol. Suppl. 111 (2002)62–65.

[67] M.F. Dorman, A. Sharma, P. Gilley, K. Martin, P. Roland, Central auditory develop-ment: Evidence from CAEP measurements in children fit with cochlear implants, J.Commun. Dis. 40 (2007) 284–294.

[68] B.E. Pfingst, L. Xu, C.S. Thompson, Across-site threshold variation in cochlearimplants: relation to speech recognition, Audiol. Neurootol. 9 (2004) 341–352.