| | Superiority of bilateral cochlear implantation over unilateral cochlear implantation in tone discrimination in chinese patients☆Abstract Purpose: The benefits of bilateral cochlear implants (CIs) versus unilateral CIs were evaluated by comparing the Cantonese lexical tone discrimination scores conducted in a quiet environment and against a background noise. Materials and Methods: Four postlingually deafened Cantonese-speaking adults (2 men and 2 women) with bilateral CIs were included in this study. The subjects were their own control in the monaural hearing condition. Both the Cantonese lexical tone stimuli and the speech-weighted background noise were presented at 00 azimuth and at a distance of 1 m from the subject. The speech stimuli, which were maintained at 65 dB sound pressure level, were presented in both a quiet environment and against a background noise at signal to noise ratios (SNRs) of +15, +10, +5, 0, −5, −10, and −15. Results: Against a background noise, the bilateral CIs required +5 dB SNR only to obtain significant scores in discriminating Cantonese lexical tones and to achieve discrimination scores that were comparable to the optimal discrimination scores obtained in quiet. No significant difference in the discrimination scores was observed between binaural and monaural hearing conditions when the tests were conducted in quiet. Conclusions: Our study showed that in the presence of background noise, bilateral CIs were better than unilateral CIs in discriminating Cantonese lexical tones. (Am J Otolaryngol 2003;24:19-23. Copyright 2003, Elsevier Science (USA). All rights reserved.)
Research in hearing science has indicated that hearing with both ears has the advantages of binaural summation, head shadow effect, improved sound localization, and better speech understanding in background noise.1 Most hearing-impaired subjects who wear binaural hearing aids (HAs) show the benefits of binaural hearing.2, 3 Bilateral stimulations from hearing prostheses can also avoid auditory deprivation in the nonstimulated, hearing-impaired ear.4 This is important because the adverse effect of auditory deprivation on speech can start as early as 3 to 6 months after the onset of deafness. The deprived ear, even if fitted with a hearing aid later, showed poorer speech perception performance than the ear with early stimulation.5 Taking into consideration the advantages of binaural hearing and avoidance of auditory deprivation, the binaural fitting of HAs is generally recommended for children with severe or profound hearing loss.
Reports on speech perception abilities, in the presence of background noise in subjects with unilateral cochlear implants (CIs), revealed conflicting results.6, 7, 8, 9, 10 The effects of binaural hearing on adults and children with CIs but wearing a HA in the nonimplanted ear have also been investigated.11, 12, 13, 14 The binaural advantage from this combination was minimal when noise was on the implant side while speech stimuli were on the HA side. This is because, unlike with bilateral CIs, the ear that had a hearing aid could not take advantage of the head shadow effect to enjoy improved signal-to-noise ratios (SNRs) for better speech perception.12, 15
There is growing interest in evaluating the hearing abilities of bilaterally cochlear-implanted users. The benefits of bilateral CIs are not fully known because of the small number of patients with bilateral CIs worldwide. The hearing acuity from bilateral CIs has been shown to effectively detect interaural amplitude and time difference in both binaural fusion and lateralization experiments.16, 17, 18 Patients with bilateral CIs had better speech perception scores in background noise compared with those who had unilateral CIs.15, 19 These studies were mainly from Western societies where English is the medium of speech communication.
CIs have been shown to be successful in improving auditory ability and speech perception performance in patients who speak Cantonese.20, 21, 22 Cantonese, a Chinese dialect, is widely spoken in the southern part of China and in Southeast Asia. The population using this dialect is about 100 million worldwide. Cantonese, with its 6 contrastive tones,23 has the characteristic that different tones of the same phonemic segments carry a different meaning. The voice contrastive pitch patterns produced by the vocal cord convey the lexical meanings of Cantonese tones. These tonal changes are not detectable by lipreading. The perception of these tones requires good temporal and spectral auditory abilities that theoretically can be provided by speech-coding strategies that use fast stimulation rates. Some CI systems have fast stimulation rates of more than 1,500 pulses per second per channel when used with the continuous interleaved sampling speech-coding strategies. The fast stimulation rates, which are effective in transmitting the subtle spectral and temporal information of speech, may in turn be beneficial when discriminating between various Cantonese lexical tones. The aim of this study was to compare the effectiveness between bilateral and unilateral CIs in discriminating Cantonese lexical tone in background noise and in quiet by Cantonese-speaking CI users.
Material and methods  Patients From April 2000 to February 2001, at the University of Hong Kong Medical Centre, Queen Mary Hospital, Hong Kong, 4 postlingually deafened adults (2 men and 2 women) received bilateral CIs. The age of the patients ranged from 23 years to 38 years (mean = 33 years). All patients were implanted with the MED-EL Tempo (MED-EL Medical Electronics, Innsbruck, Austria) + CI devices and programmed with the continuous interleaved sampling speech-coding strategies. Three patients who had been fitted with a unilateral CI requested a second CI to benefit from binaural hearing and optimal aural rehabilitation. The patients underwent staged surgery with the second implant performed at about 4 to 7 months after their first implantation. The remaining patient, at the recommendation of the other patients, received bilateral CIs at a single procedure. The demographic data of the patients with bilateral CIs are shown in Table 1.
| | |  | Subjects | Sex | Age at 1st implant (yr) | Interval for 2nd implant (mo) | Length of Deafness (yr) | Cause of Deafness | Device | |
|---|
| 1 | M | 23 | 6 | 12 | Unknown | MED-EL | |
| 2 | M | 42 | 7 | 14 | Otitis media | MED-EL | |
| 3 | F | 28 | 4 | 16 | Unknown | MED-EL | |
| 4 | F | 38 | 0 | 20 | Heredity | MED-EL | |
| | | | | |
The average preoperative unaided pure-tone thresholds (averaged at .5K Hz, 1K Hz, and 2K Hz of the left and right ear combined) was 103.33 dB hearing level. The average unilateral CI aided pure-tone thresholds in free field was 32.50 dB sound pressure level (SPL). The average bilateral CI aided pure-tone threshold in free field was 27.92 dB SPL. In this study, the evaluations were performed with the bilaterally cochlear-implanted subjects acting as their own controls in the monaural hearing condition. There were 4 bilateral and 8 unilateral CIs for comparisons in the binaural and monaural hearing conditions. Speech material The speech material used in this study was 4 lists of the Tone Discrimination Test, which form part of the Hong Kong Speech Perception Test Manual.24 The manual, used for assessing the suitability of adult candidates for cochlear implantation, was developed through the collaborative effort of a working group recruited from the 3 CI centers in Hong Kong. Each list has 30 test items, and each item has a pair of contrastive tones. The implanted subjects had to determine by verbal response whether the 2 presenting tones were the same or different. The speech material, recorded on a compact disc, was delivered through an audiometer (Madsen, Model OB822, Minnetonka, MN) to a loudspeaker. The intensity level of the speech material was maintained at 65 dB SPL, which corresponded to the loudness level in daily life conversation. The implanted subjects sat at a distance of 1 m from and at 00 azimuth toward the loudspeaker. The background noise used in this study was speech-weighted noise, which was filtered to mimic the long-term spectrum of speech. The background noise was generated from the same audiometer and played through the same loudspeaker. There were 8 conditions of measurements (7 conditions of SNRs and 1 condition of quiet). The speech-weighted noise was delivered at SNRs of +15, +10, +5, 0, −5, −10, and −15. A SNR of −10 meant that the speech stimuli were maintained at 65 dB SPL during the evaluation, whereas the noise level was delivered at 75 dB SPL. A SNR of 0 meant that both the speech stimulus and noise were delivered at 65 dB SPL. The speech lists were assigned randomly to each of the SNR conditions, and the order of the testing conditions among the subjects was randomized. No feedback of the correct answers was given in all conditions of testing. The mean percentage of the correct Cantonese lexical tones discrimination scores obtained at each SNR and in quiet were subjected to binomial analysis with correction for the continuity of discrete variables. A score of 66.77% or above was regarded as being truly correct (binomial test, Z = 1.643, P < .05). Nonparametric tests were used to compare the mean scores. Results The mean percentage of the correct Cantonese lexical tone discrimination scores of the subjects in the binaural and monaural hearing conditions obtained at the various SNRs and in quiet are shown in Figure 1.
In the binaural hearing condition, the Cantonese lexical tones discrimination scores were correct at SNRs of +15, +10, and +5 and in quiet. In the monaural hearing condition, the Cantonese lexical tones discrimination scores were correct at SNRs of +15 and +10 and in quiet. Comparison between the binaural and monaural hearing conditions of the Cantonese lexical tone discrimination scores conducted in quiet yielded no difference (Wilcoxon matched pair test, Z = 000, P = 1.000). The lexical tone discrimination scores achieved at different levels of background noise were also compared with the optimal scores obtained in quiet. Only those SNRs for which the discrimination scores were correct with the binomial analysis were included for comparison. In the bilateral CI mode of audition, at SNRs of +15, +10, and +5, subjects could achieve discrimination scores comparable to those obtained in quiet. In the unilateral CI mode of audition, however, only at SNRs of +15 and +10 could subjects attain similar effects as that obtained in quiet (Table 2).
| | | | | | SNR | |
|---|
| Audition Modes | Quiet | +15 | +10 | +5 | |
|---|
| Bilateral CIs | 86.67 | 88.34 | 89.17 | 74.17 | |
| Unilateral CI | 83.33 | 77.50 | 72.50 | | |
| | | | | |
Discussions Our study showed that subjects with bilateral CIs performed better than those with unilateral CI in discriminating Cantonese lexical tones from background noise. The subjects in the bilateral CI mode of audition required a minimum of +5 dB SNR only (ie, the speech stimuli were 5 dB louder than the background noise) to achieve significant discrimination scores. Subjects in the unilateral CI mode of audition required a SNR of at least +10 dB to achieve a similar effect. The reason for the better discrimination scores in the binaural over the monaural hearing condition in the presence of background noise may be explained by the “cocktail party” effect.1, 25, 26 This refers to the ability, when listening with both ears, to perceptually segregate sound sources, thus allowing the listener to selectively focus on a topic of interest or conversation in a noisy environment. Not all the mechanisms of this process are completely understood. It is assumed that binaural hearing has the ability for spatial localization of sounds to reduce the intrusion of noise, thus making detection of signals possible against a noisy background. When an individual with normal hearing is listening monaurally with 1 ear plugged, the masking effect is at its maximum in the unplugged ear. As a result, speech or conversation is masked and cannot be fully understood. This explains why the monaural hearing from a unilateral CI does not have the advantage of the “cocktail party” effect found in binaural hearing. Thus, patients with a unilateral CI have a poorer Cantonese lexical tones discrimination performance compared with those with bilateral CIs. The nonsignificant discrimination scores of the Cantonese lexical tones between the binaural and monaural hearing conditions in quiet showed the importance of binaural hearing when there was background noise. In quiet, despite the fact that binaural hearing has the advantage of binaural summation effect1 in improving the SNR by about 3 dB, this study showed that bilateral CIs were not significantly better than unilateral CI in the discriminating between Cantonese lexical tones. However, speech perception in quiet is not a situation that is encountered often in everyday life. In environments such as classrooms, restaurants, and busy highways, high levels of background noise are regularly present, and this may render speech understanding difficulty. Background noise can reduce the intelligibility and loudness of a speech signal even though its intensity is unchanged. The fact that the bilaterally implanted subjects required only a small difference of 5 dB SNR to achieve significant discrimination scores in background noise showed the importance and superiority of binaural hearing. The discrimination scores obtained in a quiet environment can serve as a benchmark for comparing with the discrimination scores obtained against a background noise. We also observe that the Cantonese lexical tones discrimination scores obtained from the bilateral CI mode of audition required a SNR of +5 dB to achieve discrimination scores comparable to the optimal discrimination scores attained in quiet. In the unilateral CI mode of audition, a SNR of +10 dB is necessary to achieve similar results. Noise did affect the Cantonese lexical tone discrimination scores in both the binaural and monaural hearing conditions. These deficits could be compensated by increasing the SNRs, that is, to make the speech stimuli increasingly louder than the background noise. The superiority of bilateral CIs over the unilateral CI was that the speech stimuli had to be 5 dB louder than the background noise to make the discrimination scores comparable to the optimal scores obtained in quiet. Bilateral CIs are expensive. In some countries, cochlear implantation is a luxury even for those hearing-impaired patients who cannot be helped by hearing aids. In Hong Kong, 1 CI is provided free of charge to each suitable candidate. The fitting of bilateral CIs is a major financial undertaking for most patients. Studies on bilateral CIs are still at the early stage. Our study has highlighted some benefits of binaural hearing from bilateral CIs over monaural hearing from unilateral CIs in discriminating Cantonese lexical tones from a background noise. However, advantages gained by fitting binaural CIs must be balanced against the major costs incurred. Conclusions Binaural hearing from bilateral CIs gave a better performance in discriminating Cantonese lexical tones from background noise compared with monaural hearing from unilateral CIs. In a noisy environment, bilateral CIs required the speech stimuli to be only 5 dB louder than the background noise to achieve significant discrimination scores and to obtain discrimination scores that were comparable to those attained in quiet. The unilateral CI required higher SNRs than the bilateral CIs to achieve similar effects. References 1.
1
In: ed 3.
Durrant JD, Lovrinic JH editor.
Bases of Hearing Science. Baltimore, MD: Williams & Wilkins; 1995;. 2.
2
Mueller HG, Hawkins DB.
Three important considerations in hearing aid selection.
In:
Sandlin RE editors.
Handbook of Hearing Aid Amplification. vol 2:Boston, MA: College-Hill Press; 1990;p. 31–60. 3.
3
Byrne D.
Binaural hearing aid fitting: Research findings and clinical application.
In:
Libby R editors.
Binaural Hearing and Amplification. Chicago, IL: Zenetron Inc; 1980;p. 23–73. 4.
4
Hurley RM.
Onset of auditory deprivation.
J Am Acad Audiol. 1999;10:529–534. MEDLINE 5.
5
Silverman CA, Emmer MB.
Auditory deprivation and recovery in adults with asymmetric sensori-neural hearing impairment.
J Am Acad Audiol. 1993;4:338–346. MEDLINE 6.
6
Hochberg I, Boothroyd A, Weiss M.
Effects of noise and noise suppression speech perception by cochlear implant users.
Ear Hear. 1992;13:262–270. 7.
7
Dorman MF, Loizou PC, Fitzke J.
Identification of speech in noise by cochlear implant patients and normal-hearing listeners using 6-channel digital processors.
Ear Hear. 1998;19:481–484. MEDLINE |
CrossRef
8.
8
Fetterman BL, Domico EH.
Speech recognition in background noise of cochlear implant patients.
Otolaryngol Head Neck Surg. 2002;126:257–263. Abstract | Full Text |
Full-Text PDF (94 KB)
|
CrossRef
9.
9
Kiefer J, Muller J, Pfennigdorff T, et al.
Speech understanding in quiet and in noise with the CIS speech coding strategy (MED-EL Combi-40) compared to the Multipeak and Spectral Peak strategies (Nucleus).
ORL J Otorhinolaryngol Relat Spec. 1996;58:127–135. MEDLINE |
CrossRef
10.
10
Hamzavi J, Franz P, Baumgartner WD, et al.
Hearing performance in noise of cochlear implant patients versus severely-profound hearing-impaired patients with hearing aids.
Audiology. 2001;40:26–31. MEDLINE 11.
11
Armstrong M, Pegg P, James C, et al.
Speech perception in noise with implant and hearing aid.
Am J Otolaryngol. 1997;18:S140–S141. 12.
12
Tyler RS, Parkinson AJ, Wilson BS, et al.
Patients utilizing a hearing aid and a cochlear implant: Speech perception and localization.
Ear Hear. 2002;23:98–105. MEDLINE |
CrossRef
13.
13
Chmiel RA, Clark J, Jerger J, et al.
Speech perception and production in children wearing a cochlear implant in one ear and a hearing aid n the opposite ear.
Ann Otol Rhinol Laryngol. 1995;108(suppl 166):314–316. MEDLINE 14.
14
Ching TYC, Psarros C, Hill M, et al.
Should children who use cochlear implants wear hearing aids in the opposite ear?.
Ear Hear. 2001;22:365–380. MEDLINE |
CrossRef
15.
15
van Hoesel R, Ramsden R, O'Driscoll M.
Sound-direction identification, interaural time delay discrimination, and speech intelligibility advantages in noise for a bilateral cochlear implant user.
Ear Hear. 2002;23:137–149. MEDLINE |
CrossRef
16.
16
van Hoesel RJM, Tong YC, Hollow RD, et al.
Psychophysical and speech perception studies: A case report on a binaural cochlear implant subject.
J Acoust Soc Am. 1993;94:3178–3189. MEDLINE |
CrossRef
17.
17
van Hoesel RJM, Clark GM.
Fusion and lateralization study with two binaural cochlear implant patients.
Ann Otol Rhinol Laryngol. 1995;104(suppl 166):233–235. 18.
18
van Hoesel RJM, Clark GM.
Psychophysical studies with two binaural cochlear implant subjects.
J Acoust Soc Am. 1997;102:495–507. MEDLINE |
CrossRef
19.
19
Muller J, Schon F, Helms J.
Speech understanding in quiet and noise in bilateral users of the MED-EL Combi 40/40+ cochlear implant system.
Ear Hear. 2002;23:198–206. MEDLINE |
CrossRef
20.
20
Wei WI, Wong R, Hui Y, et al.
Chinese tonal language rehabilitation following cochlear implantation in children.
Acta Otolaryngol. 2000;120:218–221. MEDLINE |
CrossRef
21.
21
Tang SO, Luk WS, Lau CC, et al.
Cochlear implant in Hong Kong Cantonese.
Am J Otolaryngol. 1990;11:421–426. 22.
22
Lee KY, van Hasselt CA, Chiu SN, et al.
Cantonese tone perception ability of cochlear implant children in comparison with normal-hearing children.
Int J Pediatr Otorhinolaryngol. 2002;63:137–147. Abstract | Full Text |
Full-Text PDF (161 KB)
|
CrossRef
23.
23
Gandour J.
Perceptual dimensions of tones: Evidence from Cantonese.
J Chin Linguist. 1981;9:20–36. 24.
24
Hong Kong Hospital Authority: Hong Kong Speech Perception Test Manual.
Hong Kong, Hong Kong Hospital Authority; 2000;. 25.
25
In:
Gulick WL, Gescheider GA, Frisina RD editor.
Hearing: Physiological, Acoustical, Neural Coding, and Psychoacoustics. New York, NY: Oxford; 1989;. 26.
26
Mitchell OM, Ross CA, Yates GH.
Signal processing for a cocktail party effect.
J Acoust Soc Am. 1971;50:656–660.
CrossRef
Department of Surgery, Division of Otorhinolaryngology, University of Hong Kong Medical Centre, Queen Mary Hospital, Hong Kong ☆ Address correspondence to: William Ignace Wei, FRCS, FRCSE, FACS, FHKAM, Department of Surgery, Division of Otorhinolaryngology, University of Hong Kong Medical Centre, Queen Mary Hospital, 102 Pokfulam Road, Hong Kong. E-mail: hrmswwi@hkucc.hku.hk. PII: S0196-0709(02)32410-4 doi:10.1053/ajot.2003.8 © 2003 Published by Elsevier Inc. | |
|
FULL TEXT32410-4/journalimage?img=PIIS0196070902324104.gr1.lrg.gif&fig=&kwhquery=&issn=0196-0709&ishighres=false&allhighres=false&free=yes','journalimage');)
