- Email: [email protected]
Transient and distortion product evoked oto-acoustic emissions in normal hearing patients with and without tinnitus
Otolaryngology–Head and Neck Surgery (2008) 138, 502-506
ORIGINAL RESEARCH— OTOLOGY AND NEUROTOLOGY
Transient and distortion product evoked oto-acoustic emissions in normal hearing patients with and without tinnitus Ronaldo C. Granjeiro, MD, MSc, Helga M. Kehrle, MD, MSc, Roberta L. Bezerra, MD, Vanessa F. Almeida, Doc Méd Sc, André L. L. Sampaio, MD, and Carlos A. Oliveira, MD, PhD, Brasília, Brasil OBJECTIVE: To test the hypothesis that tinnitus begins with outer hair cell dysfunction by recording transient (TEOAE) and distortion product evoked (DPOAE) oto-acoustic emissions in patients with normal hearing with (study group, SG) and without tinnitus (control group, CG). STUDY DESIGN: Case control study. SUBJECTS AND METHODS: SG had 32 patients with pure tone thresholds below 25 dB in the 500 to 8000 Hz interval. CG had 37 age- and gender-matched patients with similar thresholds. All patients had normal tympanograms and stapedial reflexes. TEOAE were recorded with wide band click in continuous mode at 80-dB peak SPL. DPOAE were recorded with f1/f2 ⫽ 1.22 and intensities of 65 dB (f1) and 55 dB (f2) SPL. RESULTS: DPOAE were abnormal in 68.4% of SG and in 50% of CG (P ⫽ 0.036). TEOAE were abnormal in 70.2% of SG and in 16.10% of CG (P ⫽ 0.0001). CONCLUSION: SG had significantly higher prevalence of abnormal TEOAE and DPOAE than CG. © 2008 American Academy of Otolaryngology–Head and Neck Surgery Foundation. All rights reserved.
S
ubjective tinnitus refers to sound perception that occurs in the absence of an external stimulus.1 It is a complex phenomenon, has many causes and is related to biological and psychological components.2 According to Jastreboff2 and Kaltenbach,3 this symptom is generally related to noise exposure, aging, and hearing loss, but it can occur in normal hearing people. Although tinnitus may be associated with abnormalities in any level of the auditory pathway, it very often starts in the cochlea.1 Jastreboff2 considers that tinnitus usually starts in the cochlea later and generates abnormal activity in the central pathways, which keeps the symptom alive. The central auditory pathways do not need to be structurally altered.
Simpson and Davies4 stated that 85% of patients with tinnitus also have hearing loss, with 35% of them having moderate to severe auditory impairment. However, 10% to 15% of patients with tinnitus have normal auditory thresholds in the 250 to 8000 Hz test range.5,6 In our clinical practice, we quite often see patients who complain about tinnitus and are found to have normal puretone audiometry. A recent literature search showed a paucity of publications on this subject. We hypothesized that tinnitus without hearing loss could be caused by changes in the outer hair cell (OHC) function that were not yet able to produce hearing loss. If this hypothesis is valid, transient-evoked (TEOAE) and distortion product evoked oto-acoustic emissions (DPOAE) should show the presence of OHC dysfunction in tinnitus patients with normal hearing. Consequently, the objective of the present study was to evaluate the function of the cochlear OHC by means of TEOAE and DPOAE in patients with tinnitus and normal hearing and compare this group of patients with an age- and gender-matched control group with normal hearing and no tinnitus. The results are reported in this article.
MATERIALS AND METHODS This is a case control study processed from July 2003 to July 2005 in the Hospital de Base do Distrito Federal and in the Audiology Clinic of Hospital Santa Luzia in Brasília, DF, Brazil. The ethics Committee of the Health Department of the Government of the Federal District approved the study protocol before it was started. The subjects were tested for TEOAE and DPOAE by the same investigator, who was blinded with respect to which subject had tinnitus. During the test session, subjects sat in
Received September 10, 2007; revised November 12, 2007; accepted November 14, 2007.
0194-5998/$34.00 © 2008 American Academy of Otolaryngology–Head and Neck Surgery Foundation. All rights reserved. doi:10.1016/j.otohns.2007.11.012
Granjeiro et al
Transient and distortion product evoked . . .
Table 1 Percentage of normal and altered TEOAE tests according to S/N ratio and reproducibility Test
Study group (N ⫽ 57)
Control group (N ⫽ 56)
Normal Altered
29.8% 70.2%
83.9% 16.1%
P* 0.0001
503
ered was P ⬍ 0.05. 2 test was used to compare SG and CG with respect to abnormal results according to the above criteria; Spearman rank-order correlations were used to determine the correlation between TEOAE and DPOAE results in SG at individual frequencies. Only four TEOAE and DPOAE frequencies were correlated: 1500, 2000, 3000, and 4000 Hz.
*2 test.
RESULTS an acoustically untreated test chamber where the noise level was maintained between 20 and 40 dB SPL; this was in line with the American Speech-Language-Hearing Association standards of 50 dB SPL for background noise levels. A commercially available instrument (AuDx Plus, Biologic Systems Corp, Mundelein, IL, Brasilia/Distrito Federal) controlled by a personal microcomputer was used to test both TEOAEs and DPOAEs. The first ear to be examined was chosen randomly, but TEOAE testing always was performed first. TEOAE were performed with a wide band click in a continuous mode and with an intensity of 80 dB peak SPL. For measuring DP grams, the frequency separation of the primaries was f2/f1⫽1.22, with L1 and L2 set to 65 and 55 dB SPL, respectively. The study group (SG) had 32 patients, 14 male and 18 female, 20 to 45 years old who complained of significant subjective tinnitus. The control group (CG) had 37 patients, 11 male and 26 female, aged between 20 and 45 years, without tinnitus. Only patients who had normal audiometry (threshold below 25 dB HL at all frequencies from 250 to 8000 Hz), normal type A tympanometric curve, and present ipsi and contralateral stapedial reflexes were included. Any previous otologic disease detected in the patient’s history determined exclusion of the patient from the study. In addition, neurologic diseases, acoustic trauma, vascular diseases, metabolic problems, ototoxic drugs used in the past, middle ear disease, previous ear surgery, history of vestibular problems, and recent aspirin intake were cause for exclusion. The following parameters were considered in the TEOAE tests7,8: 1) signal-to-noise (S/N) ratio ⱖ6 dB in at least three of the four frequencies tested (1500, 2000, 3000, and 4000 Hz); 2) reproducibility of the responses of at least 70% in three of the four frequencies tested (1500, 2000, 3000, and 4000 Hz). In the DPOAE the following parameters were considered for all frequencies tested7,8: 1) S/N ratio ⱖ6 dB in all frequencies tested; 2) amplitude of the signal in the 90th percentile of the normal distribution for the frequencies tested; 3) only the frequencies from 1000 to 8000 Hz were considered. The statistical analysis was done considering the results in ears, not patients, because the results were independent between ears of the same individual. Commercial software (SPSS, v.13.0) was used. The level of significance consid-
The study group (SG) consisted of 57 ears from 32 patients. Fourteen were male (13 right and 12 left ears) and 18 were female (14 right and 18 left ears). Ages varied between 20 and 45 years (average, 36). The CG had 56 ears from 37 patients. Eleven were male (8 right and 8 left ears), 26 were female patients (19 right and 21 left ears). Age in this group varied from 20 to 41 years (average, 32). SG had 57 ears tested because 7 of 32 patients had unilateral tinnitus. CG had 56 ears tested because 18 patients had one ear excluded (according to the exclusion criteria set up). There were no significant differences between the two groups with respect to gender and age. Mean duration of tinnitus in the SG was 7.2 years (standard error of the mean ⫽ 7.6). Tinnitus was bilateral in 78.1% (25 patients); 7 (21.9%) patients exhibited unilateral tinnitus; 5 (15.6%) had the symptom in the left ear; and 2 (6.3%) in the right ear.
TEOAE Results TEOAE were abnormal in 70.2% of the patients in SG compared with 16.1% in CG. This difference was statistically significant (P ⬍ 0.05) as indicated in Table 1. The abnormal TEOAE in the SG were present for the 3000 and 4000 Hz frequencies mostly, while in the CG, the 1500 and 4000 Hz frequencies showed abnormalities more often. Table 2 shows the normal and abnormal TEOAE at each frequency tested in both the SG and CG. The difference
Table 2 Percentage of normal and altered TEOAE tests according to the frequencies evaluated in each group Study group (N ⫽ 57)
Control group (N ⫽ 56)
Frequency Normal Altered Normal Altered (Hz) % % % % 1500 2000 3000 4000 *2 test.
47.4 47.4 31.6 14.0
52.6 52.6 68.4 86.0
64.3 85.7 91.1 67.9
35.7 14.3 8.9 32.1
P* 0.070 0.001 0.001 0.001
504
Otolaryngology–Head and Neck Surgery, Vol 138, No 4, April 2008
Table 3 Percentage of normal and altered DPOAE tests according to amplitude and S/N ratio Test
Study group (N ⫽ 57)
Control group (N ⫽ 56)
P*
Normal Altered
31.60% 68.40%
50.00% 50.00%
0.036
*2 test.
between the two groups is highly significant at all frequencies except 1500 Hz.
DPOAE Results SG had 68.4% of abnormal results compared with 50.0% in CG. This difference was statistically significant (P ⫽ 0.036) as indicated in Table 3. Table 4 shows the results for DPOAE at each frequency tested for both subject groups. All frequencies had a significant difference between SG and CG (P ⬍ 0.05) and 3000, 4000, and 6000 Hz showed the greatest differences.
Correlation between TEOAE and DPOAE Results Spearman’s correlation coefficient was used to evaluate the degree of agreement between TEOAE and DPOAE in the frequencies tested (1500, 2000, 3000, and 4000 Hz) in SG. Significant correlation between the results of the two tests was present for both normal and abnormal cases. In the SG, the Spearman’s correlation coefficient between the two tests was 0.465 (P ⫽ 0.0001) for the normal and abnormal results. So when TEOAE results are altered there is an 82.5% (33 of 40 cases) probability for the DPOAE to be altered as well. When the DPOAE is altered, the probability for the TEOAE to be also abnormal is 84.61% (33 of 39 cases). Correlation in CG between TEOAE and DPOAE can not be tested because DPOAE are abnormal in 50% of the normal population according to the parameters for normality we used.7
One critical point in a study such as ours is to identify the criteria for normality. If we considered all tests with an S/N ratio ⱖ6 dB as normal, we certainly would not have a chance to distinguish SG from CG, because these criteria would be too lenient. In a search of the literature we found the criteria set up by Gorga et al7 to distinguish patients with normal hearing from patients with mild hearing loss with the use of the DPOAE results. We adopted this criterion: the amplitude of the signal for all frequencies tested had to be in the 90th percentile of the Gorga et al values for normality (8000 Hz: ⫺12.85 dB; 6000 Hz: ⫺6.88 dB; 4000 Hz: ⫺4.42 dB; 3000 Hz: ⫺5.55 dB; 2000 Hz: ⫺3.50 dB; 1500 Hz: 0.43 dB; 1000 Hz: 4.40 dB; 750 Hz: 2.40 dB) in order for the test to be considered normal. Of course, the S/N ratio had to be at or above 6 dB. These criteria produced a high abnormality rate in the CG, but the difference between the two groups was still statistically significant (SG, 68.40% abnormal results; CG, 50.0% abnormal results). Probst and Hauser,10 who used criteria for normality similar to ours also had a rate of abnormality similar to the ones found here. We choose TEOAE and DPOAE because they are the most commonly used OAE tests in clinical practice and because they have more standardized methodology.10 The criteria for normality were stated in the Materials and Methods section. We excluded 750 Hz from our DPOAE results because of the difficulty of the middle ear to convey these oto-acoustic emissions at low frequencies and because of the high sensitivity to external noises in this frequency.11 We have not found any previous work about TEOAE in patients with normal hearing and tinnitus. We found altered TEOAE results in 70.20% in SG and 16.1% in CG. Our normal results in the CG were different from the results of Glatke and Robinette,12 Lonsbury-Martin et al13 and Harrison and Norton,14 who found normal TEOAE in 97% to 100% of normal hearing people without tinnitus. This probably is due to our criteria of normality that includes both
Table 4 Percentage of normal and altered DPOAE tests according to amplitude and S/N ratio in each frequency tested Study group (N ⫽ 57)
DISCUSSION Our objective was to verify cochlear dysfunction in patients with normal hearing and tinnitus with the use of the TEOAE and DPOAE tests. We selected patients between 20 and 45 years because in this age group the oto-acoustic emissions are fairly stable.9 The two groups were homogeneous with respect to age and gender. A comparable number of right and left ears was also present in both groups. All subjects had normal auditory thresholds (⬍25 dB) in the frequencies from 250 to 8000 Hz; normal tympanometry results and stapedial reflexes were present.
Control group (N ⫽ 56)
Frequency Normal Altered Normal Altered (Hz) % % % % 1000 1500 2000 3000 4000 6000 8000 *2 test.
66.7 70.2 80.7 47.4 54.4 59.6 78.9
33.3 29.8 19.3 52.6 45.6 40.4 21.1
82.1 87.5 96.4 76.8 91.1 85.7 96.4
17.9 12.5 03.6 23.2 8.9 14.3 03.6
P* 0.047 0.021 0.008 0.001 0.000 0.002 0.004
Granjeiro et al
Transient and distortion product evoked . . .
S/N ratio and reproducibility instead of only the former parameter. The highest percentage of abnormality was found at 3000 and 4000 Hz for the SG. When we compared the two groups at each frequency tested (2000, 3000, 4000) there were significant differences between the two groups in all of them. These results are in agreement with the literature15-18 that reports altered TEOAE in tinnitus patients with hearing losses mainly for frequencies above 2000 Hz. The fact that normal hearing patients with tinnitus have a significantly higher percentage of abnormal TEOAE and DPOAE than normal hearing subjects without tinnitus may suggest that OHC dysfunction may be important in the generation of tinnitus. However, 29.8% of the SG had normal TEOAE results and 16.1% of the CG had abnormal results in this test. With respect to DPOAE, 31.6% of SG had normal results and 50.0% of CG had abnormal test results. It seems that OHC dysfunction detectable by TEOAE and DPOAE are neither necessary nor sufficient to cause tinnitus. Yet normal hearing people with tinnitus have significantly more abnormal results of TEOAE and DPOAE than normal hearing people without tinnitus. One way to explain this outcome is to postulate different levels of sensitivity of the higher auditory pathways to dysfunction of the OHCs. In some patients, levels of dysfunction not detectable by TEOAE, and DPOAE may be sufficient to trigger tinnitus. In others, levels of dysfunction may not be sufficient to cause the symptoms. In this way, we would agree with Jastreboff2 who states that all levels of the auditory pathways may be involved in the production of tinnitus, but the trigger is most likely at the OHC level, at least in the majority of cases. We think that patients with normal hearing, no tinnitus, and abnormal TEOAE and DPOAE results should be followed for the possibility of having tinnitus. Because we are talking about the interaction of the cochlear abnormality with the sensitivity of the central pathways, it is difficult to state that these patients are at a higher risk to develop tinnitus. A follow-up will provide the answer. This point should be studied in a prospective, longitudinal manner with additional research. We did not try to correlate intensity and annoyance of tinnitus with the results of the TEOAE and DPOAE because we would need at least 30 patients in each category: mild tinnitus (1 to 3 on the Visual Analog Scale [VAS]), intermediate intensity (4 to 7 on the VAS), and severe disabling tinnitus (8 to 10 on the VAS). Only a few of our tinnitus patients were in the mild category. The vast majority were in the intermediate one and none were in the severe disabling category. In 1995 the senior author19 (C.A.O.) published a study that included 83 temporal bones of patients with significant tinnitus during their lives in an attempt to uncover some pathological correlate(s) for the symptom. Surprisingly, 37% of the temporal bones had absolutely normal histology.
505
He then thought that some change that occurred at the submicroscopic level was the primary cause of tinnitus. As 23% of the bones showed endolymphatic hydrops and 35% of the ones with normal histology also had vertigo, he imagined a continuum from normal histology to endolymphatic hydrops in patients with tinnitus and possibly with the Meniere disease syndrome. Now, dysfunction of the OHC that probably would not be detected by light microscopy is detected by TEOAE and DPOAE tests. Finally we found good correlation between the results of TEOAE and DPOAE with the Spearman’s correlation coefficient test in SG. Like Gorga et al,20 we found the best correlations at 2000 and 4000 Hz. This work has only shown a higher prevalence of OAE abnormalities in tinnitus patients with normal hearing compared with normal hearing patients without tinnitus. Additional research is needed to determine the value of routine OAE testing of normal-hearing patients with tinnitus and its potential benefit on the clinical outcome of these patients.
AUTHOR INFORMATION From the Secretaria de Saúde do Governo do Distrito Federal (Drs Granjeiro and Kehrle), Hospital Santa Luzia (Dr Almeida) and the Department of Otolaryngology–Head and Neck Surgery, Brasília University Medical School (Drs Bezerra, Sampaio, and Oliveira). Corresponding author: Ronaldo Campos Granjeiro, SQSW 305 Bloco C, Apartament 302, Brasília, DF, 70673-423, Brasil. E-mail address: [email protected]. Presented at the 140th Annual Meeting of the American Otological Society Inc.-April 27-28, 2007. San Diego, CA, U S A.
AUTHOR CONTRIBUTIONS Ronaldo C. Granjeiro, study design, data collection writer; Helga M. Kehrle, study design, data collection; Roberta L. Bezerra, Vanessa F. Almeida, and André L. L. Sampaio, study design and review; Carlos A. Oliveira, study design, review, adviser.
FINANCIAL DISCLOSURE This work was supported by Fundação de Empreendimentos Científicos e Tecnológicos-FINATEC, Brasília, DF, Brasil.
REFERENCES 1. Moller AR. Pathophysiology of tinnitus. Ann Otol Rhinol Laryngol 1984;93:39 – 44. 2. Jastreboff PJ. Review article: phantom auditory perception (tinnitus): mechanisms of generation and perception. Neurosci Res 1990;8: 221–54. 3. Kaltenbach JA. Neurophysiologic mechanisms of tinnitus. J Am Acad Audiol 2000;11:125–37. 4. Simpson JJ, Davies WE. A review of evidence in support of role for 5-HT in the perception of tinnitus. Hear Res 2000;145:1–7.
506
Otolaryngology–Head and Neck Surgery, Vol 138, No 4, April 2008
5. Axelsson A, Ringdahl A. Tinnitus: a study of its prevalence and characteristics. Br J Audiol 1989;23:53– 62. 6. Nicolas-Puel C, Faulconbridge RL, Guitton M, et al. Characteristics of tinnitus and etiology of associated hearing loss: a study of 123 patients. Inter Tin J 2002;8:37– 44. 7. Gorga MP, Neely ST, Ohirich B, et al. From laboratory to clinic: a large scale study of distortion product otoacoustic emissions in ears with normal hearing and ears with hearing loss. Ear Hear 1997;18: 440 –55. 8. Bonfils P, Avan P. Distortion product otoacoustic emissions: values for clinical use. Arch Otolaryngol Head Neck Surg 1992;118:1069 –76. 9. Probst R, Lonsbury-Martin BL, Martin GK. A review of otoacoustic emissions. J Acoust Soc Am 1991;11:236 – 43. 10. Probst R, Hauser R. Distortion product otoacoustic emissions in normal and hearing impaired ears. Am J Otolaryngol 1990;11:236 – 43. 11. Gorga MP, Neeley ST, Bergman BM, et al. Otoacoustic emission from normal hearing and hearing impaired subjects: distortion products responses. J Acoustic Soc Am 1993;93:2050 – 60. 12. Robinete MS, Glattke TJ. Transient otoacoustic emissions. In: Robinete MS and Glattke TJ, editors. Otoacoustic emission clinical applications. Stuttgart, Germany:Thieme; 1997. p. 63– 82.
13. Lonsbury-Martin BL, Whitehead ML, Martin GK. Clinical applications of otoacoustic emissions. J Speech Hear Res 1991;34:964 – 81. 14. Harrison W, Norton SJ. Characteristics of transient evoked otoacoustic emissions in normal hearing and hearing impaired children. Ear Hear 1999;20:75– 86. 15. Mirz F, Pedersen CB, Ishizu K, et al. Positron emission tomography of cortical centers of tinnitus. Hear Res 1999;134:133– 44. 16. Kaltenbach JA, Afman CE. Hyperactivity in the dorsal cochlear nucleus after intense sound exposure and its resemblance to tone-evoked activity: a physiological model for tinnitus. Hear Res 2000;140:165–72. 17. Norena A, Mickeyl M, Chéry-Croze S, et al. Psychoacoustic characterization of the tinnitus spectrum: implications of the underlying mechanisms of tinnitus. Audiol Neurootol 2002;7:358 – 69. 18. Ochi K, Ohashi T, Kenmochi M. Hearing impairment and tinnitus pitch in patients with unilateral tinnitus: comparison of sudden hearing loss and chronic tinnitus. Laryngoscope 2003;113:427–31. 19. Oliveira CA. Thinking about tinnitus. Inter Tin J 1995;1:1– 4. 20. Gorga MP, Neely ST, Bergman BM, et al. A comparison of transient evoked and distortion product otoacoustic emissions in normal hearing and hearing impaired subjects. J Acoustic Soc Am 1993;94:2639 – 48.
ORIGINAL RESEARCH— OTOLOGY AND NEUROTOLOGY
Transient and distortion product evoked oto-acoustic emissions in normal hearing patients with and without tinnitus Ronaldo C. Granjeiro, MD, MSc, Helga M. Kehrle, MD, MSc, Roberta L. Bezerra, MD, Vanessa F. Almeida, Doc Méd Sc, André L. L. Sampaio, MD, and Carlos A. Oliveira, MD, PhD, Brasília, Brasil OBJECTIVE: To test the hypothesis that tinnitus begins with outer hair cell dysfunction by recording transient (TEOAE) and distortion product evoked (DPOAE) oto-acoustic emissions in patients with normal hearing with (study group, SG) and without tinnitus (control group, CG). STUDY DESIGN: Case control study. SUBJECTS AND METHODS: SG had 32 patients with pure tone thresholds below 25 dB in the 500 to 8000 Hz interval. CG had 37 age- and gender-matched patients with similar thresholds. All patients had normal tympanograms and stapedial reflexes. TEOAE were recorded with wide band click in continuous mode at 80-dB peak SPL. DPOAE were recorded with f1/f2 ⫽ 1.22 and intensities of 65 dB (f1) and 55 dB (f2) SPL. RESULTS: DPOAE were abnormal in 68.4% of SG and in 50% of CG (P ⫽ 0.036). TEOAE were abnormal in 70.2% of SG and in 16.10% of CG (P ⫽ 0.0001). CONCLUSION: SG had significantly higher prevalence of abnormal TEOAE and DPOAE than CG. © 2008 American Academy of Otolaryngology–Head and Neck Surgery Foundation. All rights reserved.
S
ubjective tinnitus refers to sound perception that occurs in the absence of an external stimulus.1 It is a complex phenomenon, has many causes and is related to biological and psychological components.2 According to Jastreboff2 and Kaltenbach,3 this symptom is generally related to noise exposure, aging, and hearing loss, but it can occur in normal hearing people. Although tinnitus may be associated with abnormalities in any level of the auditory pathway, it very often starts in the cochlea.1 Jastreboff2 considers that tinnitus usually starts in the cochlea later and generates abnormal activity in the central pathways, which keeps the symptom alive. The central auditory pathways do not need to be structurally altered.
Simpson and Davies4 stated that 85% of patients with tinnitus also have hearing loss, with 35% of them having moderate to severe auditory impairment. However, 10% to 15% of patients with tinnitus have normal auditory thresholds in the 250 to 8000 Hz test range.5,6 In our clinical practice, we quite often see patients who complain about tinnitus and are found to have normal puretone audiometry. A recent literature search showed a paucity of publications on this subject. We hypothesized that tinnitus without hearing loss could be caused by changes in the outer hair cell (OHC) function that were not yet able to produce hearing loss. If this hypothesis is valid, transient-evoked (TEOAE) and distortion product evoked oto-acoustic emissions (DPOAE) should show the presence of OHC dysfunction in tinnitus patients with normal hearing. Consequently, the objective of the present study was to evaluate the function of the cochlear OHC by means of TEOAE and DPOAE in patients with tinnitus and normal hearing and compare this group of patients with an age- and gender-matched control group with normal hearing and no tinnitus. The results are reported in this article.
MATERIALS AND METHODS This is a case control study processed from July 2003 to July 2005 in the Hospital de Base do Distrito Federal and in the Audiology Clinic of Hospital Santa Luzia in Brasília, DF, Brazil. The ethics Committee of the Health Department of the Government of the Federal District approved the study protocol before it was started. The subjects were tested for TEOAE and DPOAE by the same investigator, who was blinded with respect to which subject had tinnitus. During the test session, subjects sat in
Received September 10, 2007; revised November 12, 2007; accepted November 14, 2007.
0194-5998/$34.00 © 2008 American Academy of Otolaryngology–Head and Neck Surgery Foundation. All rights reserved. doi:10.1016/j.otohns.2007.11.012
Granjeiro et al
Transient and distortion product evoked . . .
Table 1 Percentage of normal and altered TEOAE tests according to S/N ratio and reproducibility Test
Study group (N ⫽ 57)
Control group (N ⫽ 56)
Normal Altered
29.8% 70.2%
83.9% 16.1%
P* 0.0001
503
ered was P ⬍ 0.05. 2 test was used to compare SG and CG with respect to abnormal results according to the above criteria; Spearman rank-order correlations were used to determine the correlation between TEOAE and DPOAE results in SG at individual frequencies. Only four TEOAE and DPOAE frequencies were correlated: 1500, 2000, 3000, and 4000 Hz.
*2 test.
RESULTS an acoustically untreated test chamber where the noise level was maintained between 20 and 40 dB SPL; this was in line with the American Speech-Language-Hearing Association standards of 50 dB SPL for background noise levels. A commercially available instrument (AuDx Plus, Biologic Systems Corp, Mundelein, IL, Brasilia/Distrito Federal) controlled by a personal microcomputer was used to test both TEOAEs and DPOAEs. The first ear to be examined was chosen randomly, but TEOAE testing always was performed first. TEOAE were performed with a wide band click in a continuous mode and with an intensity of 80 dB peak SPL. For measuring DP grams, the frequency separation of the primaries was f2/f1⫽1.22, with L1 and L2 set to 65 and 55 dB SPL, respectively. The study group (SG) had 32 patients, 14 male and 18 female, 20 to 45 years old who complained of significant subjective tinnitus. The control group (CG) had 37 patients, 11 male and 26 female, aged between 20 and 45 years, without tinnitus. Only patients who had normal audiometry (threshold below 25 dB HL at all frequencies from 250 to 8000 Hz), normal type A tympanometric curve, and present ipsi and contralateral stapedial reflexes were included. Any previous otologic disease detected in the patient’s history determined exclusion of the patient from the study. In addition, neurologic diseases, acoustic trauma, vascular diseases, metabolic problems, ototoxic drugs used in the past, middle ear disease, previous ear surgery, history of vestibular problems, and recent aspirin intake were cause for exclusion. The following parameters were considered in the TEOAE tests7,8: 1) signal-to-noise (S/N) ratio ⱖ6 dB in at least three of the four frequencies tested (1500, 2000, 3000, and 4000 Hz); 2) reproducibility of the responses of at least 70% in three of the four frequencies tested (1500, 2000, 3000, and 4000 Hz). In the DPOAE the following parameters were considered for all frequencies tested7,8: 1) S/N ratio ⱖ6 dB in all frequencies tested; 2) amplitude of the signal in the 90th percentile of the normal distribution for the frequencies tested; 3) only the frequencies from 1000 to 8000 Hz were considered. The statistical analysis was done considering the results in ears, not patients, because the results were independent between ears of the same individual. Commercial software (SPSS, v.13.0) was used. The level of significance consid-
The study group (SG) consisted of 57 ears from 32 patients. Fourteen were male (13 right and 12 left ears) and 18 were female (14 right and 18 left ears). Ages varied between 20 and 45 years (average, 36). The CG had 56 ears from 37 patients. Eleven were male (8 right and 8 left ears), 26 were female patients (19 right and 21 left ears). Age in this group varied from 20 to 41 years (average, 32). SG had 57 ears tested because 7 of 32 patients had unilateral tinnitus. CG had 56 ears tested because 18 patients had one ear excluded (according to the exclusion criteria set up). There were no significant differences between the two groups with respect to gender and age. Mean duration of tinnitus in the SG was 7.2 years (standard error of the mean ⫽ 7.6). Tinnitus was bilateral in 78.1% (25 patients); 7 (21.9%) patients exhibited unilateral tinnitus; 5 (15.6%) had the symptom in the left ear; and 2 (6.3%) in the right ear.
TEOAE Results TEOAE were abnormal in 70.2% of the patients in SG compared with 16.1% in CG. This difference was statistically significant (P ⬍ 0.05) as indicated in Table 1. The abnormal TEOAE in the SG were present for the 3000 and 4000 Hz frequencies mostly, while in the CG, the 1500 and 4000 Hz frequencies showed abnormalities more often. Table 2 shows the normal and abnormal TEOAE at each frequency tested in both the SG and CG. The difference
Table 2 Percentage of normal and altered TEOAE tests according to the frequencies evaluated in each group Study group (N ⫽ 57)
Control group (N ⫽ 56)
Frequency Normal Altered Normal Altered (Hz) % % % % 1500 2000 3000 4000 *2 test.
47.4 47.4 31.6 14.0
52.6 52.6 68.4 86.0
64.3 85.7 91.1 67.9
35.7 14.3 8.9 32.1
P* 0.070 0.001 0.001 0.001
504
Otolaryngology–Head and Neck Surgery, Vol 138, No 4, April 2008
Table 3 Percentage of normal and altered DPOAE tests according to amplitude and S/N ratio Test
Study group (N ⫽ 57)
Control group (N ⫽ 56)
P*
Normal Altered
31.60% 68.40%
50.00% 50.00%
0.036
*2 test.
between the two groups is highly significant at all frequencies except 1500 Hz.
DPOAE Results SG had 68.4% of abnormal results compared with 50.0% in CG. This difference was statistically significant (P ⫽ 0.036) as indicated in Table 3. Table 4 shows the results for DPOAE at each frequency tested for both subject groups. All frequencies had a significant difference between SG and CG (P ⬍ 0.05) and 3000, 4000, and 6000 Hz showed the greatest differences.
Correlation between TEOAE and DPOAE Results Spearman’s correlation coefficient was used to evaluate the degree of agreement between TEOAE and DPOAE in the frequencies tested (1500, 2000, 3000, and 4000 Hz) in SG. Significant correlation between the results of the two tests was present for both normal and abnormal cases. In the SG, the Spearman’s correlation coefficient between the two tests was 0.465 (P ⫽ 0.0001) for the normal and abnormal results. So when TEOAE results are altered there is an 82.5% (33 of 40 cases) probability for the DPOAE to be altered as well. When the DPOAE is altered, the probability for the TEOAE to be also abnormal is 84.61% (33 of 39 cases). Correlation in CG between TEOAE and DPOAE can not be tested because DPOAE are abnormal in 50% of the normal population according to the parameters for normality we used.7
One critical point in a study such as ours is to identify the criteria for normality. If we considered all tests with an S/N ratio ⱖ6 dB as normal, we certainly would not have a chance to distinguish SG from CG, because these criteria would be too lenient. In a search of the literature we found the criteria set up by Gorga et al7 to distinguish patients with normal hearing from patients with mild hearing loss with the use of the DPOAE results. We adopted this criterion: the amplitude of the signal for all frequencies tested had to be in the 90th percentile of the Gorga et al values for normality (8000 Hz: ⫺12.85 dB; 6000 Hz: ⫺6.88 dB; 4000 Hz: ⫺4.42 dB; 3000 Hz: ⫺5.55 dB; 2000 Hz: ⫺3.50 dB; 1500 Hz: 0.43 dB; 1000 Hz: 4.40 dB; 750 Hz: 2.40 dB) in order for the test to be considered normal. Of course, the S/N ratio had to be at or above 6 dB. These criteria produced a high abnormality rate in the CG, but the difference between the two groups was still statistically significant (SG, 68.40% abnormal results; CG, 50.0% abnormal results). Probst and Hauser,10 who used criteria for normality similar to ours also had a rate of abnormality similar to the ones found here. We choose TEOAE and DPOAE because they are the most commonly used OAE tests in clinical practice and because they have more standardized methodology.10 The criteria for normality were stated in the Materials and Methods section. We excluded 750 Hz from our DPOAE results because of the difficulty of the middle ear to convey these oto-acoustic emissions at low frequencies and because of the high sensitivity to external noises in this frequency.11 We have not found any previous work about TEOAE in patients with normal hearing and tinnitus. We found altered TEOAE results in 70.20% in SG and 16.1% in CG. Our normal results in the CG were different from the results of Glatke and Robinette,12 Lonsbury-Martin et al13 and Harrison and Norton,14 who found normal TEOAE in 97% to 100% of normal hearing people without tinnitus. This probably is due to our criteria of normality that includes both
Table 4 Percentage of normal and altered DPOAE tests according to amplitude and S/N ratio in each frequency tested Study group (N ⫽ 57)
DISCUSSION Our objective was to verify cochlear dysfunction in patients with normal hearing and tinnitus with the use of the TEOAE and DPOAE tests. We selected patients between 20 and 45 years because in this age group the oto-acoustic emissions are fairly stable.9 The two groups were homogeneous with respect to age and gender. A comparable number of right and left ears was also present in both groups. All subjects had normal auditory thresholds (⬍25 dB) in the frequencies from 250 to 8000 Hz; normal tympanometry results and stapedial reflexes were present.
Control group (N ⫽ 56)
Frequency Normal Altered Normal Altered (Hz) % % % % 1000 1500 2000 3000 4000 6000 8000 *2 test.
66.7 70.2 80.7 47.4 54.4 59.6 78.9
33.3 29.8 19.3 52.6 45.6 40.4 21.1
82.1 87.5 96.4 76.8 91.1 85.7 96.4
17.9 12.5 03.6 23.2 8.9 14.3 03.6
P* 0.047 0.021 0.008 0.001 0.000 0.002 0.004
Granjeiro et al
Transient and distortion product evoked . . .
S/N ratio and reproducibility instead of only the former parameter. The highest percentage of abnormality was found at 3000 and 4000 Hz for the SG. When we compared the two groups at each frequency tested (2000, 3000, 4000) there were significant differences between the two groups in all of them. These results are in agreement with the literature15-18 that reports altered TEOAE in tinnitus patients with hearing losses mainly for frequencies above 2000 Hz. The fact that normal hearing patients with tinnitus have a significantly higher percentage of abnormal TEOAE and DPOAE than normal hearing subjects without tinnitus may suggest that OHC dysfunction may be important in the generation of tinnitus. However, 29.8% of the SG had normal TEOAE results and 16.1% of the CG had abnormal results in this test. With respect to DPOAE, 31.6% of SG had normal results and 50.0% of CG had abnormal test results. It seems that OHC dysfunction detectable by TEOAE and DPOAE are neither necessary nor sufficient to cause tinnitus. Yet normal hearing people with tinnitus have significantly more abnormal results of TEOAE and DPOAE than normal hearing people without tinnitus. One way to explain this outcome is to postulate different levels of sensitivity of the higher auditory pathways to dysfunction of the OHCs. In some patients, levels of dysfunction not detectable by TEOAE, and DPOAE may be sufficient to trigger tinnitus. In others, levels of dysfunction may not be sufficient to cause the symptoms. In this way, we would agree with Jastreboff2 who states that all levels of the auditory pathways may be involved in the production of tinnitus, but the trigger is most likely at the OHC level, at least in the majority of cases. We think that patients with normal hearing, no tinnitus, and abnormal TEOAE and DPOAE results should be followed for the possibility of having tinnitus. Because we are talking about the interaction of the cochlear abnormality with the sensitivity of the central pathways, it is difficult to state that these patients are at a higher risk to develop tinnitus. A follow-up will provide the answer. This point should be studied in a prospective, longitudinal manner with additional research. We did not try to correlate intensity and annoyance of tinnitus with the results of the TEOAE and DPOAE because we would need at least 30 patients in each category: mild tinnitus (1 to 3 on the Visual Analog Scale [VAS]), intermediate intensity (4 to 7 on the VAS), and severe disabling tinnitus (8 to 10 on the VAS). Only a few of our tinnitus patients were in the mild category. The vast majority were in the intermediate one and none were in the severe disabling category. In 1995 the senior author19 (C.A.O.) published a study that included 83 temporal bones of patients with significant tinnitus during their lives in an attempt to uncover some pathological correlate(s) for the symptom. Surprisingly, 37% of the temporal bones had absolutely normal histology.
505
He then thought that some change that occurred at the submicroscopic level was the primary cause of tinnitus. As 23% of the bones showed endolymphatic hydrops and 35% of the ones with normal histology also had vertigo, he imagined a continuum from normal histology to endolymphatic hydrops in patients with tinnitus and possibly with the Meniere disease syndrome. Now, dysfunction of the OHC that probably would not be detected by light microscopy is detected by TEOAE and DPOAE tests. Finally we found good correlation between the results of TEOAE and DPOAE with the Spearman’s correlation coefficient test in SG. Like Gorga et al,20 we found the best correlations at 2000 and 4000 Hz. This work has only shown a higher prevalence of OAE abnormalities in tinnitus patients with normal hearing compared with normal hearing patients without tinnitus. Additional research is needed to determine the value of routine OAE testing of normal-hearing patients with tinnitus and its potential benefit on the clinical outcome of these patients.
AUTHOR INFORMATION From the Secretaria de Saúde do Governo do Distrito Federal (Drs Granjeiro and Kehrle), Hospital Santa Luzia (Dr Almeida) and the Department of Otolaryngology–Head and Neck Surgery, Brasília University Medical School (Drs Bezerra, Sampaio, and Oliveira). Corresponding author: Ronaldo Campos Granjeiro, SQSW 305 Bloco C, Apartament 302, Brasília, DF, 70673-423, Brasil. E-mail address: [email protected]. Presented at the 140th Annual Meeting of the American Otological Society Inc.-April 27-28, 2007. San Diego, CA, U S A.
AUTHOR CONTRIBUTIONS Ronaldo C. Granjeiro, study design, data collection writer; Helga M. Kehrle, study design, data collection; Roberta L. Bezerra, Vanessa F. Almeida, and André L. L. Sampaio, study design and review; Carlos A. Oliveira, study design, review, adviser.
FINANCIAL DISCLOSURE This work was supported by Fundação de Empreendimentos Científicos e Tecnológicos-FINATEC, Brasília, DF, Brasil.
REFERENCES 1. Moller AR. Pathophysiology of tinnitus. Ann Otol Rhinol Laryngol 1984;93:39 – 44. 2. Jastreboff PJ. Review article: phantom auditory perception (tinnitus): mechanisms of generation and perception. Neurosci Res 1990;8: 221–54. 3. Kaltenbach JA. Neurophysiologic mechanisms of tinnitus. J Am Acad Audiol 2000;11:125–37. 4. Simpson JJ, Davies WE. A review of evidence in support of role for 5-HT in the perception of tinnitus. Hear Res 2000;145:1–7.
506
Otolaryngology–Head and Neck Surgery, Vol 138, No 4, April 2008
5. Axelsson A, Ringdahl A. Tinnitus: a study of its prevalence and characteristics. Br J Audiol 1989;23:53– 62. 6. Nicolas-Puel C, Faulconbridge RL, Guitton M, et al. Characteristics of tinnitus and etiology of associated hearing loss: a study of 123 patients. Inter Tin J 2002;8:37– 44. 7. Gorga MP, Neely ST, Ohirich B, et al. From laboratory to clinic: a large scale study of distortion product otoacoustic emissions in ears with normal hearing and ears with hearing loss. Ear Hear 1997;18: 440 –55. 8. Bonfils P, Avan P. Distortion product otoacoustic emissions: values for clinical use. Arch Otolaryngol Head Neck Surg 1992;118:1069 –76. 9. Probst R, Lonsbury-Martin BL, Martin GK. A review of otoacoustic emissions. J Acoust Soc Am 1991;11:236 – 43. 10. Probst R, Hauser R. Distortion product otoacoustic emissions in normal and hearing impaired ears. Am J Otolaryngol 1990;11:236 – 43. 11. Gorga MP, Neeley ST, Bergman BM, et al. Otoacoustic emission from normal hearing and hearing impaired subjects: distortion products responses. J Acoustic Soc Am 1993;93:2050 – 60. 12. Robinete MS, Glattke TJ. Transient otoacoustic emissions. In: Robinete MS and Glattke TJ, editors. Otoacoustic emission clinical applications. Stuttgart, Germany:Thieme; 1997. p. 63– 82.
13. Lonsbury-Martin BL, Whitehead ML, Martin GK. Clinical applications of otoacoustic emissions. J Speech Hear Res 1991;34:964 – 81. 14. Harrison W, Norton SJ. Characteristics of transient evoked otoacoustic emissions in normal hearing and hearing impaired children. Ear Hear 1999;20:75– 86. 15. Mirz F, Pedersen CB, Ishizu K, et al. Positron emission tomography of cortical centers of tinnitus. Hear Res 1999;134:133– 44. 16. Kaltenbach JA, Afman CE. Hyperactivity in the dorsal cochlear nucleus after intense sound exposure and its resemblance to tone-evoked activity: a physiological model for tinnitus. Hear Res 2000;140:165–72. 17. Norena A, Mickeyl M, Chéry-Croze S, et al. Psychoacoustic characterization of the tinnitus spectrum: implications of the underlying mechanisms of tinnitus. Audiol Neurootol 2002;7:358 – 69. 18. Ochi K, Ohashi T, Kenmochi M. Hearing impairment and tinnitus pitch in patients with unilateral tinnitus: comparison of sudden hearing loss and chronic tinnitus. Laryngoscope 2003;113:427–31. 19. Oliveira CA. Thinking about tinnitus. Inter Tin J 1995;1:1– 4. 20. Gorga MP, Neely ST, Bergman BM, et al. A comparison of transient evoked and distortion product otoacoustic emissions in normal hearing and hearing impaired subjects. J Acoustic Soc Am 1993;94:2639 – 48.