Assessing Auditory Nerve Condition by Tone Decay in Deaf Subjects with a Cochlear Implant

International Journal of Audiology

ISSN: 1499-2027 (Print) 1708-8186 (Online) Journal homepage: https://www.tandfonline.com/loi/iija20

Assessing auditory nerve condition by tone decay in deaf subjects with a cochlear implant

Jan-Willem A. Wasmann, Ruben H. M. van Eijl, Huib Versnel & Gijsbert A. van Zanten

To cite this article: Jan-Willem A. Wasmann, Ruben H. M. van Eijl, Huib Versnel & Gijsbert A. van Zanten (2018) Assessing auditory nerve condition by tone decay in deaf subjects with a cochlear implant, International Journal of Audiology, 57:11, 864-871, DOI: 10.1080/14992027.2018.1498598 To link to this article: https://doi.org/10.1080/14992027.2018.1498598

Published online: 27 Sep 2018.

Submit your article to this journal

Article views: 182

View Crossmark data

Full Terms & Conditions of access and use can be found at https://www.tandfonline.com/action/journalInformation?journalCode=iija20 INTERNATIONAL JOURNAL OF AUDIOLOGY 2018, VOL. 57, NO. 11, 864–871 https://doi.org/10.1080/14992027.2018.1498598

ORIGINAL ARTICLE Assessing auditory nerve condition by tone decay in deaf subjects with a cochlear implant

Jan-Willem A. Wasmanna,b, Ruben H. M. van Eijla,c, Huib Versnela,c and Gijsbert A. van Zantena,c aDepartment of Otorhinolaryngology and Head & Neck Surgery, University Medical Center Utrecht, Utrecht University, Utrecht, The Netherlands; bDepartment of Otorhinolaryngology, Radboud University Medical Center, Nijmegen, The Netherlands; cBrain Center Rudolf Magnus, University Medical Center Utrecht, Utrecht University, Utrecht, The Netherlands

ABSTRACT ARTICLE HISTORY The condition of the auditory nerve is a factor determining hearing performance of cochlear implant (CI) Received 8 June 2016 recipients. Abnormal loudness adaptation is associated with poor auditory nerve survival. We examined Revised 26 June 2018 which stimulus conditions are suitable for tone decay measurements to differentiate between CI recipi- Accepted 2 July 2018 ents with respect to their speech perception. Tone decay was defined here as occurring when the per- KEYWORDS cept disappears before the stimulus stops. We measured the duration of the percept of a 60-s pulse train. Cochlear implant; Current levels ranged from below threshold up to maximum acceptable loudness, pulse rates from 250 behavioural measures; to 5000 pulses/s, and duty cycles (percentages of time the burst of pulses is on) from 10% to 100%. Ten speech perception; adult CI recipients were included: seven with good and three with poor speech perception. Largest differ- psychoacoustics/hear- ences among the subjects were found at 5000 pulses/s and 100% duty cycle. The well performing sub- ing science jects had a continuous percept of the 60-s stimulus within 3 dB above threshold. Two poorly performing subjects showed abnormal loudness adaptation, that is, no continuous percept even at levels greater than 6 dB above threshold. We conclude that abnormal loudness adaptation can be detected via an electric tone decay test using a high pulse rate and 100% duty cycle.

Introduction stimulation compared to normal (Shepherd and Javel 1997; Sly et al. 2007). We assume that this leads to abnormal loudness Profound sensorineural hearing loss or deafness is usually caused adaptation, which is also referred to as marked or abnormal tone by loss of sensory hair cells in the cochlea. In many cases, it is decay (Huss and Moore 2003). treated by cochlear implantation. A cochlear implant (CI) In the early years of cochlear implantation, an electrical ana- bypasses the hair cells and stimulates the auditory nerve via elec- logue of the tone decay test was used for CI recipients by trical currents. This provides auditory perception, and in the Brimacombe and Eisenberg (1984). Three out of seventeen sub- most successful cases leads to almost normal speech perception jects exhibited abnormal loudness adaptation even at maximum in low-noise acoustic conditions (Wilson and Dorman 2008; acceptable loudness. Those three subjects became deaf at an ear- Blamey et al. 2013). However, not all recipients experience such lier age, had more years of profound hearing loss and shorter a successful outcome; the variability in speech perception out- experience with cochlear implant and hearing aid use than the come among CI recipients is large (Blamey et al. 2013). Auditory other subjects (Brimacombe and Eisenberg 1984). Sennaroglu nerve degeneration, occurring as a result of severe hair cell loss et al. (2001) investigated the effect of electric pulse rate on loud- (Spoendlin 1975; Versnel et al. 2007), is one of many variables ness adaptation in seven CI recipients. They expected stronger identified to affect speech perception in CI recipients (Cosetti adaptation with a higher rate, but they did not find an effect. and Waltzman 2012; Seyyedi et al, 2014). The extent of nerve Current subjective clinical tests for CI recipients, such as degeneration varies widely among patients (Fayad and Linthicum measurements of threshold and maximum acceptable loudness 2006; Seyyedi et al. 2013). (MAL) or speech reception tests, do not provide a measure of Loudness adaptation reflects a reduction in loudness during the condition of the auditory nerve. Before the era of auditory continued exposure to a stationary sound with constant level. brainstem response (ABR) and MRI for assessment of the audi- For people with normal hearing, acoustic loudness adaptation tory nerve, the tone decay test was one of the tests in the coch- occurs at levels close to the threshold (within 5 dB) and is more lear-retrocochlear differentiation test battery (Clemis and Gee, prominent at frequencies above 1000 Hz (Bray et al. 1973; 1979). We hypothesise that a tone decay test provides such a Hellman et al. 1997). Tone decay is a specific type of loudness measure in CI recipients and that a recipient’s speech perception adaptation in which the loudness is reduced to no-percept, that performance is inversely related to the degree of tone decay. is, “silence” (Owens 1964). A degenerated auditory nerve may This is based on the assumption that tone decay reflects auditory cause a more rapid decrease of neural response to steady nerve degeneration and that tone decay increases with extent of

CONTACT Jan-Willem A. Wasmann [email protected] Department of Otorhinolaryngology, Radboud University Medical Center, Philips van Leydenlaan 15, P.O. Box 9101, 6500 HB, Nijmegen, The Netherlands ß 2018 British Society of Audiology, International Society of Audiology, and Nordic Audiological Society. Published by Informa UK Limited, trading as Taylor & Francis Group. This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives License (http://creativecommons.org/licenses/by-nc-nd/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited, and is not altered, transformed, or built upon in any way. INTERNATIONAL JOURNAL OF AUDIOLOGY 865 nerve degeneration. Therefore, we conducted an exploratory Helsinki (version 2013, Fortaleza) and the Medical Research method-finding study to find parameters for a test that might Involving Human Subjects Act (WMO). The study was approved link loudness adaptation measures with speech perception by the Medical Research Ethics Committee of the UMC Utrecht performance. (protocol number 13-648/D) and informed consent was obtained from all subjects. Methods Equipment Subjects Tests were performed using a stand-alone personal computer Ten profoundly deaf adults with a multichannel CI participated (PC) with a built-in data acquisition card (National Instruments in the study. Subjects were recruited from the outpatient clinic NI PCI-6533), which was connected to a Research Interface Box of the UMC Utrecht. All subjects used a SONATAti100 or 2 (RIB2) that in turn was connected to the implant by means of PULSARci100 device manufactured by MED-EL. Each subject an electrical coil with the common radio frequency link. The had at least 6 months of experience with the CI. Six subjects had RIB2 developed by the Leopold-Franzens-University of suffered progressive hearing loss, one with onset of hearing loss Innsbruck ensured the electrical isolation of computer and sub- in childhood. Two subjects had experienced sudden deafness, ject and enabled control of the implant by the computer one in childhood. Two subjects were congenitally deaf. Table 1 (Bahmer et al. 2010). The experiments were written in MATLAB summarises subject characteristics. (version 7.12.0.635; Mathworks, Natick, MA, USA), using the The MAL level, which is the highest current level on a specific electrode for a certain pulse rate that is still comfortably Psychophysics toolbox (version 3.0.11; Brainard 1997) and the loud, for each electrode was taken from the patients’ clinical RIB2.dll (version 1.12). records. As a measure of speech perception performance, consonant-vowel-consonant (CVC) phoneme recognition scores Stimuli were collected in quiet. Seven subjects, C1–C6 and C13, were considered good performers since their CVC scores were above All stimuli consisted of (a series of concatenated) 700-ms long 70%, and three subjects with scores below 50% were considered basic time frames. A basic time frame started with a number (N) poor performers (see Table 1). of repeated pulses at amplitude (A), that is, a pulse train, and The experiment consisted of two test sessions in which elec- the rest of the time frame contained no stimulation (Figure 1). tric detection thresholds and tone decay were measured as The percentage of the time the burst is on is referred to as the reported by the subject, with at least one week in between ses- duty cycle. The duty cycle was varied from 10% to 100%. With sions. Testing was in accordance with the Declaration of duty cycles less than 100% the effect of a silent gap on tone

Table 1. Subject demographics. duration of Experience Congenital/ Subject nr. M/F Age (yr) deafness (yr) with CI (yr) Aetiology progressive Mean CVC (%) C1 F 59 28 0.7 Genetic Progressive 87 C2 M 60 25 4 Otosclerosis Progressive 92 C3 M 72 6 5 Meniere’s Disease Progressive 87 C4 M 83 21 1 Genetic Progressive 85 C5 M 55 26 1 Genetic Progressive 90 C6 M 72 3 8 Infection Sudden 88 C13 M 34 22 7 Genetic Progressive 88 C7 M 62 45 5 Meningitis Sudden 43 C8 F 51 45 5 Rubella Congenital <20 C9 F 50 46 4 Rubella Congenital 38 CVC: Consonant-Vowel-Consonant recognition score.

Figure 1. Examples of two 60-s pulse train stimuli with duty cycles of 100% and 80%. 866 J.-W. A. WASMANN ET AL.

Table 2. Stimulus characteristics. Nr. of Pulse Inter-pulse Train Duty Stimulus label pulses rate (pps) interval (ms) length (ms) cycle (%) PR250 100% 175 250 4 700 100 PR250 80% 140 250 4 560 80 PR250 20% 35 250 4 140 20 PR1000 100% 700 1000 1 700 100 PR1000 80% 560 1000 1 560 80 PR1000 20% 140 1000 1 140 20 PR1000 10% 70 1000 1 70 10 PR5000 100% 3500 5000 0.2 700 100 PR5000 80% 2800 5000 0.2 560 80 PR5000 20% 700 5000 0.2 140 20 PR5000 10% 350 5000 0.2 70 10

decay can be explored. We expect less tone decay with longer the mean of the last six ascending and descending turning silent gaps, because the nerve has a longer time to recover from points. A similar procedure was followed for determining the adaptation before the next pulse train (Chimento and Schreiner MAL, except that subjects had to press the button when 1990; Shannon 1990; Killian et al. 1994). In addition, the effect the sound was loud but comfortable and release the button when of number of pulses can be disentangled from that of pulse rate. the sound was medium loud. A total of three ascending and Stimuli used for threshold and MAL estimation were concaten- two descending turning points were obtained. The MAL was ated basic time frames, in which the next basic time frame had estimated as the highest turning point. an increased or decreased amplitude. Stimuli used for tone-decay measurements consisted of basic time frames with the same Tone decay test amplitude, which were repeated for 60 s. All pulse trains consisted of charge-balanced biphasic current A stimulus was presented at a constant current level for 60 s. m pulses, with a fixed pulse duration of 25 s/phase and an inter- The subjects were instructed to press the button when the stimu- m phase gap of 2.1 s. The pulse rate was 250, 1000 or 5000 lus was perceived and to release the button when perception pulses/s (pps). The amplitude could vary between stimulus pre- stopped. They were instructed that the stimulus duration varied sentations, with a step size between 1 and 60 cu, and could range between 1-s and 2 min, and that they had to track how long they from zero to MAL. Table 2 lists the applied stimuli. The strength perceived the stimulus. If the stimulus was not perceived at all, of the pulses was defined in current units (cu), a custom unit or if tone decay occurred, the current level was increased. A l used by MED-EL, which approximately equals 1 A. Afterwards recovery period of 5 s was used between stimulus presentations. m the detection thresholds were converted to dB relative to 100 A Every tone decay pattern was characterised by two values, the ¼ m (20 log10(I/Iref) with I being the current and Iref 100 A). The detection threshold of the stimulus and the minimum stimula- supra-threshold stimulation levels were expressed in dB sensation tion level for a sustained percept of the stimulus, from now on ’ level (dB SL), which is the level in decibels above the subject s called “continuous percept level”. Percept duration was defined detection threshold. as the period of time from stimulus onset until cessation of perception. The detection threshold was approximated by the high- Procedure est stimulation level at which there was no percept, that is, the just inaudible stimulus. The continuous percept level was the Each subject started with practice trials to familiarise the subject lowest stimulus level with a percept duration of 60 s. with the procedure. As soon as the investigator (JW) concluded The test was stopped as soon as the detection threshold and that the subject understood the procedure, the actual test was continuous percept level were determined or if the stimulation level started. Whenever the investigator observed that a subject lost reached the MAL, which subjects could also indicate during this concentration, the test was aborted and restarted after a short test.Despitethesimpletestparadigm,thetonedecaytestwasdiffi- break. The measurements were performed if possible on three cult for subjects. It required a long concentration span, and it was electrodes: 3, 6 and 9. This range of electrodes was chosen to hard to decide whether the stimulus was still heard, which was cover a reasonable range of electrode positions (apical, middle even more difficult close to hearing threshold. Therefore, if a con- and basal). Not all stimulus conditions were measured for all tinuous percept level was measured, a short “check” stimulus in a subjects due to lack of time, lack of concentration by the subject range from 5 to 10 s duration was presented after a break of 5 s at or inactive electrodes. thesameleveltoverifythatthesubjectreallyrespondedtothat Prior to the tone decay measurements, detection thresholds stimulus level. The check stimulus was not applied if the maximum for all stimulus conditions and MAL for stimulus conditions test level for that run was reached or if a preset number of presen- with a higher pulse rate than used in a subject’s default processor tations (5–8) was reached to prevent excessive experiment time. settings were estimated using a Bekesy tracking method (Jerger False positive responses were recorded for a particular stimulus if a 1960). To determine threshold, subjects were instructed to press subject responded longer than 70 s or did not respond consistently the button when the stimulus was perceived and to release the to the associated check stimulus, but otherwise false positive button when not. If the button was pressed by the subject at any responses could not be recognised with any degree of certainty. time during a basic time frame, the current level was decreased After the measurements, all the tone decay data were visually in the next basic time frame. Otherwise the current level was inspected to assess measurement validity. The measurement increased. A total of eight ascending and eight descending turn- results were categorised as follows. Category 1 included both ing points were obtained. The threshold was estimated by taking threshold and continuous percept level. Category 2 included a INTERNATIONAL JOURNAL OF AUDIOLOGY 867

Figure 2. (A–F) Selection of response patterns observed in tone decay testing for three good (A–C) and three poor performers (D–F). The filled circles mark the percept duration for single recordings. Detection thresholds are marked with T, continuous percept levels with CP, and the dashed line indicates the time limit for the continuous percept. threshold and at least 4 results above threshold, but a continuous acquired on different electrodes could be pooled. If multiple percept was not observed up to MAL (i.e. abnormal). Category 3 measurements were made, the mean value is presented. included a continuous percept level but did not include a threshold. Instead, the detection threshold was estimated if a response Detection thresholds less than 10 s was included in the pattern. Category 3 occurred when the first stimulus was perceived continuously, and the fol- The detection thresholds for the three pulse rates at a duty cycle lowing current levels were insufficiently decreased. A result was of 100%, and for four duty cycles at a pulse rate of 5000 pps are discarded if a lapse in concentration had occurred, if percept plotted in Figure 3. A clear drop in detection threshold with duration was greater than 70 s, or if it did not fit in a category. increasing pulse rate was seen for all subjects (as expected from Statistical analyses were conducted using SPSS version 23 the literature; Shannon 1985; Kreft et al. 2004; Zhou et al. 2012). (IBM, Armonk, NY, USA). Mean differences within subjects The 5000 pps stimulus gave significantly lower thresholds than were calculated with dependent two-tailed t-tests or Wilcoxon the 1000 pps stimulus (by 6 dB, t-test p < 0.001), and the 1000 signed rank tests. Comparisons between the good and poor per- – pps stimulus gave lower detection thresholds than the 250 pps formance groups were made with Mann Whitney U tests. Since ¼ we could not obtain measurements for all stimulus conditions stimulus (by 2 dB, t-test p 0.010). A linear fit of detection for each patient, we had missing data. Stimulus conditions with thresholds at 100% duty cycle versus pulse rate in octaves, 2 ¼ too few measurements (PR250 80%, PR1000 80%, PR1000 20% showed a decrease of 2 dB per octave (R 0.923). and PR1000 10%) were not included in the statistical analysis. For the 5000 pps stimulus, the duty cycle had a significant but Measurements for different electrodes were pooled. small effect on detection threshold (Figure 3). The 100% duty cycle resulted in significantly higher thresholds (1-2 dB) than for the 80% or 20% duty cycles (t-test, p ¼ 0.044 and p ¼ 0.041 Results respectively). Duty cycles of 80% versus 20% (t-test, p ¼ 0.553), and 80% versus 10% (t-test, p ¼ 0.423), did not significantly differ. A total of 147 (131 þ 5 þ 11, category 1, 2 and 3, respectively) In addition, comparison of two stimulus conditions with the same complete tone decay measurements were performed. number of pulses (PR1000 100% vs. PR5000 20%) showed a lower Additionally, 28 detection thresholds were included from incom- detection threshold for the higher pulse rate (t-test, p ¼ 0.002). plete tone decay measurements. Figure 2 shows representative There were no significant differences in detection thresholds response patterns. A commonly seen pattern was non-monotonic between good and poor performers for PR250 100% (B, D, E, F). Two subjects showed tone decay even at MAL (E, – ¼ – F). All tone decay patterns showed an overall increase in percept (Mann Whitney U, p 0.383), PR1000 100% (Mann Whitney U, ¼ – ¼ duration with increasing current level. p 0.250), and PR5000 100% (Mann Whitney U, p 0.714). In a within-subject pairwise comparison of electrodes 3, 6 and 9, little effect of stimulation electrode was seen on threshold or Continuous percept continuous percept levels (<1 dB). In the subsequent analysis, it was assumed that the effect of electrode position is negligible The continuous percept levels are displayed in Figure 4 for the compared to the effect of pulse rate, and that therefore data samesetofstimulusconditionsasshowninFigure 3.Mostwell 868 J.-W. A. WASMANN ET AL.

Figure 3. Detection thresholds for several stimuli. Data for the poor performers are identified. Points in a stimulus group are separated horizontally to avoid superimposition.

Figure 4. Continuous percept levels. Data for the poor performers are identified. The symbol indicates that the MAL was reached before a continuous percept was measured. Points in a stimulus group are separated horizontally to avoid superimposition. performing subjects (5 out of 7) had a continuous percept of the stimuli. As she stated, the perception of stimuli with a 100% 60-s stimulus at or below 3dB SL, regardless of stimulus condition. duty cycle was lost within the stimulus duration, but continuous Two of the well performing subjects showed continuous percepts perception occurred for lower duty cycles. between 3 and 6 dB SL in some (but not all) of the stimulus conditions. No significant effect of pulse rate (t-test, p 0.263) or duty cycle (t-test, p 0.131) on continuous percept level was found. Reproducibility The continuous percept level for PR5000 100% was greater than 4 dB SL for subjects C8 and C9, who were poor performers, At least one stimulus condition was measured twice in the first while the mean across good performers was 1.6 dB SL ±1.1 dB. and second sessions for each subject. Repetitions were also made However, the difference between the poor and good performers within a session. The repeated measurements do not reflect a was not significant (Mann–Whitney U, p ¼ 0.072), because the fully representative random sample due to incomplete measure- continuous percept level of poor performer C7 (3 dB SL for ments. The Bland Altman analysis provides a measure of repro- PR5000 100%) was similar to those for the good performers ducibility (Bland and Altman 1986). Here it was used to evaluate (Figure 4). The displayed values for subjects C8 and C9 under- the agreement among two measurements for the same stimulus estimate the true values because the MAL was reached for at condition (Figure 5). For the detection threshold, the difference least one or more measurements. Subject C9 confirmed that the was less than 0.5 dB within and between sessions, and 95% of sound percept disappeared after 20–40 s, even for initially loud the differences were smaller than 3 dB. For the continuous INTERNATIONAL JOURNAL OF AUDIOLOGY 869

Figure 5. Bland–Altman test–retest plot within a session and between visits for threshold measurements and tone decay measurements for various conditions. A. Threshold within session. B. Tone decay within session. C. Threshold between sessions. D. Tone decay between sessions. percept, the difference was less than 1 dB within and between seen in Figure 4, a 140 ms silent gap (i.e. 80% duty cycle) or sessions, and 95% of the differences were smaller than 3 dB. more, was enough to prevent loudness adaptation for all subjects, and led to slightly lower thresholds than for a 100% duty cycle. Discussion Continuous percept This pilot study shows that a wide variation in loudness adaptation can be measured among CI users, which may reflect different states In agreement with the findings of Brimacombe and Eisenberg of auditory nerve survival. The widest variation was observed for (1984), the percept duration increased with increasing stimula- continuous stimuli with a pulse rate of 5000 pps, where good tion level. All well performing subjects in this study had a con- speech perception performers showed continuous percepts a few tinuous percept for levels below 4 dB SL, independent of pulse dB above threshold and where two out of three poor performers rate or duty cycle. When comparing our results with those of experienced tone decay at high levels, at or close to MAL, which other studies, differences in devices and reported units have to was well above the range of the good performers. be taken into account. Sennaroglu et al. (2001) set the limit of abnormal loudness adaptation to 30 Nucleus current levels (Cochlear Corporation) above threshold for the Nucleus CI 24 M Detection thresholds devices, which is equivalent to 5.3 dB SL in this study. The results for good performers are thus in the range that Studies that investigated detection threshold as a function of Sennaroglu et al. (2001) reported as normal. We suggest that a pulse rate have shown that electrical thresholds decrease by reasonable criterion for abnormal loudness adaptation is the – 1 4 dB per doubling of pulse rate (Shannon 1985; Kreft et al. mean value of the continuous percept level of the good perform- 2004; Zhou et al. 2012). The decrease in detection threshold with ers plus two times the standard deviation. For the stimulus con- increasing pulse rate by 2 dB per octave, as found in this study, dition PR5000 100%, this criterion was 4 dB SL. is within this range. The condition of the auditory nerve varies among CI patients, Another effect to take into account is forward masking, which and C7 may have a nerve that is comparable to the auditory occurs when two sounds are presented within 200 ms (Zwislocki nerves of CI patients with good speech perception. The poor et al. 1959). Temporal rules for forward masking for CI recipi- perception of C7 may be caused by meningitis (Blamey et al. ents and normal listeners appear to be similar (Shannon 1990). 2013). There are many more factors that can affect outcome If forward masking is representative of temporal recovery from (Blamey et al. 2013). Hence sensitivity for detecting poor per- loudness adaptation, then a silent gap of the order of 200 ms formers with the tone decay test may be low (here, 2 out of 3), would provide enough recovery time to the auditory system to but the specificity (here, 7 out of 7) is expected to be high since prevent abnormal loudness adaptation. In this study, as can be the condition of the auditory nerve is vital for CI outcome. 870 J.-W. A. WASMANN ET AL.

Two poor performers stand out compared to the other sub- duty cycle is recommended. However, taking into account the jects. Both C8 and C9 were congenitally deaf due to maternal small group of poor performers evaluated in this study, and the rubella. They were the only subjects without tinnitus complaints spread in outcomes, no clear conclusions can be drawn regarding before implantation. (We do not know if their tinnitus was still the relation between speech perception and tone decay outcome. absent at the time of the tone decay measurements.) A recent A greater number of poorly performing subjects should be tested multicentre retrospective study on postlingually deaf adults has to determine the sensitivity and specificity of the electric tone shown that a duration of deafness of more than 40 years has a decay test. We hypothesise that the sensitivity will be low, but negative impact on speech perception (Blamey et al. 2013). the specificity will be high. Aetiology is also a factor that affects speech perception. In the future, auditory performance and nerve degeneration However, maternal rubella as a factor was not separ- might be studied with a combination of imaging and electro- ately reported. physiological techniques (Ramekers et al. 2015; Vos et al. 2015), The tone decay test suffers from some drawbacks. Tinnitus electric tone decay and/or post mortem cochlear histopathology perception at the time of measurement could make it harder to (Seyyedi et al. 2014). discern the electric stimuli in the threshold range, which is the region of interest in this study. The perception of tinnitus could even increase during a test session due to fatigue and periods of Disclosure statement silence (Tucker et al. 2005). The subjects did not receive any input between stimulus presentations, because the processor unit No potential conflict of interest was reported by the authors. was not worn during the test sessions. Another drawback for the tone decay method is that the criterion of threshold is set by the Funding subject and not controlled by the investigator. This introduces an uncertainty when comparing among subjects, because differ- The study was funded by the University Medical Center Utrecht. ences could stem from different subjective criteria about hearing or not hearing the weak stimulus. The subjects found it hard to determine if they still heard the stimulus. Often, if they released References the button, and stimulation was stopped, they noticed a change “ “ ” Bahmer, A., O. Peter, and U. Baumann. 2010. Recording and Analysis of to real silence. In those cases, subjects were encouraged to con- Electrically Evoked Compound Action Potentials (ECAPs) with MED-EL sider a very weak sound still as a sound. Therefore, the oscilla- Cochlear Implants and Different Artifact Reduction Strategies in Matlab.” tory pattern observed in Figure 2 could be a result of subjects Journal of Neuroscience Methods 191 (1): 66–74. doi:10.1016/ changing their criterion for “silence” between stimulus presenta- j.jneumeth.2010.06.008. tions. We conclude that the investigator should be aware of con- Blamey, P., F. Artieres, D. Baskent, F. Bergeron, A. Beynon, E. Burke, N. Dillier, et al. 2013. “Factors Affecting Auditory Performance of centration lapses, tinnitus, subjective decision criteria about Postlinguistically Deaf Adults Using Cochlear Implants: An Update with hearing or not hearing the weak stimulus, and false positive 2251 Patients.” Audiology and Neurotology 18 (1): 36–47. doi:10.1159/ responses; the investigator should instruct and coach subjects 000343189 accordingly. Bland, J. M., and D. Altman. 1986. “Statistical Methods for Assessing ” Possibly, a shortened version of the test procedure applied in Agreement between Two Methods of Clinical Measurement. Lancet 327 (8476): 307–310. doi:10.1016/S0140-6736(86)90837-8 this pilot study could be used in an early stage of the CI fitting Brainard, D. H. 1997. “The Psychophysics Toolbox.” Spatial Vision 10 (4): process to distinguish suspected poor performers who might 433–436. doi:10.1163/156856897X00357 need an adjusted fitting and evaluation process. Abnormal tone Bray, D. A., D. D. Dirks, and D. E. Morgan. 1973. “Perstimulatory Loudness decay disappeared when a lower duty cycle was introduced Adaptation.” The Journal of the Acoustical Society of America – (Figure 4), which is interesting for CI coding strategies. Silent 53 (6):1544 8. doi:10.1121/1.1913499 Brimacombe, J. A., and L. S. Eisenberg. 1984. “Tone Decay in Subjects with gaps (like we used with duty cycles of 80% or lower) could delib- the Single-Channel Cochlear Implant.” International Journal of Audiology erately be introduced in daily stimulation. Alternatively, the aver- 23: 321–332. doi:10.3109/00206098409072843 age load on the nerve can be reduced by reducing the pulse Chimento, T. C., and C. E. Schreiner. 1990. “Time Course of Adaptation and repetition rate. The first method gives the nerve non-activation Recovery from Adaptation in the Cat Auditory-Nerve Neurophonic.” The – periods for recovery, the second gives more time to recover Journal of the Acoustical Society of America 88 (2): 857 864. doi:10.1121/ 1.399735 between successive pulses. Likely candidates for such stimulus Clemis, J. D., and T. M. Gee. 1979. “Brain Stem Electric Response Audiometry in paradigm would be patients with a suspected poor nerve condi- the Differential Diagnosis of Acoustic Tumors.” Laryngoscope 89 (1): 31–42. tion, that is, patients with an infection or acoustic neuroma as doi:10.1288/00005537-197901000-00004 cause of deafness, long duration of deafness or prelingually deaf Cosetti, M. K., and S. B. Waltzman. 2012. “Outcomes in Cochlear adults. Figures 3 and 4 show that it is likely that a procedure Implantation: Variables Affecting Performance in Adults and Children.” – using a 5000 pps 100% duty cycle stimulus at either 4 dB SL or Otolaryngologic Clinics of North America 45 (1): 155 171. doi:10.1016/ j.otc.2011.08.023 at MAL (whichever is lowest) might identify future poor per- Fayad, J. N., and F. H. Linthicum. Jr. 2006. “Multichannel Cochlear Implants: formers. Subsequently, in case of an abnormally high continuous relation of Histopathology to Performance.” Laryngoscope 116 (8): percept level, a stimulus with a duty cycle below 100% could be 1310–1320. doi:10.1097/01.mlg.0000227176.09500.28 used as a control to confirm that the subject understands the Hellman, R. P., A. Miskiewicz, and B. Scharf. 1997. “Loudness Adaptation ” procedure, since then normal loudness adaptation is expected. and Excitation Patterns: Effects of Frequency and Level. The Journal of the Acoustical Society of America 101 (4): 2176–2185. doi:10.1121/1.418202 Huss, M., and B. C. J. Moore. 2003. “Tone Decay for Hearing-Impaired Listeners with and without Dead Regions in the Cochlea.” The Journal of Conclusions the Acoustical Society of America 114 (6 Pt 1): 3283–3294. doi:10.1121/ 1.1624063 For prelingually deaf adults or for CI recipients with poor speech Jerger, J. 1960. “Bekesy Audiometry in Analysis of Auditory Disorders.” perception, the electrical tone decay test might give insights into Journal of Speech Language and Hearing Research 3 (3): 275–287. the outcome with CI. A stimulus with high pulse rate and 100% doi:10.1044/jshr.0303.275 INTERNATIONAL JOURNAL OF AUDIOLOGY 871

Killian, M. J., S. F. Klis, and G. F. Smoorenburg. 1994. “Adaptation in the Shannon, R. V. 1990. “Forward Masking in Patients with Cochlear Implants.” Compound Action Potential Response of the guinea Pig VIIIth Nerve to The Journal of the Acoustical Society of America 88 (2): 741–744. Electric Stimulation.” Hearing Research 81 (1-2): 66–82. doi:10.1016/0378- doi:10.1121/1.399777 5955(94)90154-6 Shepherd, R. K., and E. Javel. 1997. “Electrical Stimulation of the Auditory Kreft, H. A., G. S. Donaldson, and D. A. Nelson. 2004. “Effects of Pulse Rate Nerve. I. Correlation of Physiological Responses with Cochlear on Threshold and Dynamic Range in Clarion Cochlear-Implant Users Status.” Hearing Research 108 (1–2): 112–144. doi:10.1016/S0378- (L).” The Journal of the Acoustical Society of America 115 (5 Pt 1): 5955(97)00046-4 1885–1888. doi:10.1121/1.1701895 Spoendlin, H. 1975. “Retrograde Degeneration of the Cochlear Nerve.” Acta “ ” Owens, E. 1964. Tone Decay in VIIIth Nerve and Cochlear Lesions. The Oto-Laryngologica 79 (3-4): 266–275. doi:10.3109/00016487509124683 – Journal of Speech and Hearing Disorders 29: 14 22. doi:10.1044/jshd.2901.14 Tucker, D. A., S. L. Phillips, R. A. Ruth, W. A. Clayton, E. Royster, and Ramekers, D., H. Versnel, S. B. Strahl, S. F. L. Klis, and W. Grolman. 2015. A. D. Todd. 2005. “The Effect of Silence on Tinnitus Perception.” “Recovery Characteristics of the Electrically Stimulated Auditory Nerve in ” Otolaryngology-Head and Neck Surgery: Official Journal of American Deafened guinea Pigs: Relation to Neuronal Status. Hearing Research 321: Academy of Otolaryngology-Head and Neck Surgery 132 (1): 20–24. 12–24. doi:10.1016/j.heares.2015.01.001 “ doi:10.1016/j.otohns.2005.08.016 Sennaroglu, G., L. Sennaroglu, E. E. Yucel, and E. Belgin. 2001. Cochlear Versnel, H., M. J. H. Agterberg, J. C. M. J. de Groot, G. F. Smoorenburg, Implants: Is There Any Relationship between Stimulation Rates and and S. F. L. Klis. 2007. “Time Course of Cochlear Electrophysiology and Adaptation?.” Otology & Neurotology 22 (5): 634–636. doi:10.1097/ Morphology after Combined Administration of Kanamycin and 00129492-200109000-00013 Furosemide.” Hearing Research 231 (1–2): 1–12. doi:10.1016/ Seyyedi, M., D. K. Eddington, and J. B. Nadol. Jr. 2013. “Effect of Monopolar and Bipolar Electric Stimulation on Survival and Size of Human Spiral j.heares.2007.03.003 ” Vos, S. B., W. Haakma, H. Versnel, M. Froeling, L. Speleman, P. Dik, M. A. Ganglion Cells as Studied by Postmortem Histopathology. Hearing “ Research 302: 9–16. doi:10.1016/j.heares.2013.04.007 Viergever, A. Leemans, and W. Grolman. 2015. Diffusion Tensor Seyyedi, M., L. M. Viana, and J. B. Nadol. Jr. 2014. “Within-Subject Imaging of the Auditory Nerve in Patients with Long-Term Single-Sided ” – Comparison of Word Recognition and Spiral Ganglion Cell Count in Deafness. Hearing Research 323: 1 8. doi:10.1016/j.heares.2015.01.010 “ Bilateral Cochlear Implant Recipients.” Otology & Neurotology 35: 1–1450. Wilson, B. S., and M. F. Dorman. 2008. Cochlear Implants: A Remarkable ” – doi:10.1097/MAO.0000000000000443 past and a Brilliant Future. Hearing Research 242 (1-2): 3 21. Sly, D. J., L. F. Heffer, M. W. White, R. K. Shepherd, M. G. J. Birch, R. L. doi:10.1016/j.heares.2008.06.005 Minter, N. E. Nelson, A. K. Wise, and S. J. O’Leary. 2007. “Deafness Zhou, N., L. Xu, and B. E. Pfingst. 2012. “Characteristics of Detection Alters Auditory Nerve Fibre Responses to Cochlear Implant Stimulation.” Thresholds and Maximum Comfortable Loudness Levels as a Function of European Journal of Neuroscience 26 (2): 510–522. doi:10.1111/j.1460- Pulse Rate in Human Cochlear Implant Users.” Hearing Research 284 (1- 9568.2007.05678.x 2): 25–32. doi:10.1016/j.heares.2011.12.008 Shannon, R. V. 1985. “Threshold and Loudness Functions for Pulsatile Zwislocki, J., E. Pirodda, and H. Rubin. 1959. “On Some Poststimulatory Stimulation of Cochlear Implants.” Hearing Research 18 (2): 135–143. Effects at the Threshold of Audibility.” Journal of the Acoustical Society of doi:10.1016/0378-5955(85)90005-X America 31 (1): 9–14. doi:10.1121/1.1907619