Acta Oto-Laryngologica
ISSN: 0001-6489 (Print) 1651-2251 (Online) Journal homepage: http://www.tandfonline.com/loi/ioto20
Audiological Findings in Acoustic Neuroma Jin Kanzaki, Kaoru Ogawa, Shigeo Ogawa, Minako Yamamoto, Shunya Ikeda & Toshiaki O-uchi To cite this article: Jin Kanzaki, Kaoru Ogawa, Shigeo Ogawa, Minako Yamamoto, Shunya Ikeda & Toshiaki O-uchi (1991) Audiological Findings in Acoustic Neuroma, Acta Oto-Laryngologica, 111:sup487, 125-132, DOI: 10.3109/00016489109130457 To link to this article: http://dx.doi.org/10.3109/00016489109130457
Published online: 08 Jul 2009.
Submit your article to this journal
Article views: 66
View related articles
Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=ioto20 Download by: [Australian National University]
Date: 26 April 2016, At: 21:52
Acta Otolaryngol (Stockh) 1991; Suppl. 487: 125-132
Audiological Findings in Acoustic Neuroma JIN KANZAKI, KAORU OGAWA, SHIGEO OGAWA, MINAKO YAMAMOTO, SHUNYA IKEDA and TOSHIAKI 0-UCHI From the Department of Otolaryngology, School of Medicine, Keio Universily, Tokyo, Japan
Downloaded by [Australian National University] at 21:52 26 April 2016
Kanzaki J, Ogawa K, Ogawa S, Yamamoto M, Ikeda S, 0-Uchi T. Audiological findings in acoustic neuroma. Acta Otolaryngol (Stockh) 1991; Suppl 487: 125- 132 Audiological examinations are vital in the diagnosis of acoustic neuroma. In interpreting their results, however, it is necessary to consider the patient’s hearing level. The most sensitive audiological examination is auditory brain stem response ( ABR) audiometry. Its most useful parameter is the IT5. A U-shaped audiometric configuration suggests AN, since it is seen in 10% of patients with small tumors. Psychological audiometric tests can be excluded from the battery of screening tests since they have low rates of positive diagnosis. The stapedius reflex (SR) test also has a low positive diagnostic rate in cases of small tumors. Even with the parameters of absence of reflex, elevated threshold, and decay combined, the overall SR test has a lower positive diagnostic rate than ABR audiometry. Nevertheless, the SR test can be employed as a screening device in cases in which the hearing level at 2 kHz and lower is 70 dB or lower, even if it is 71 dB or higher at 4 kHz and 8 kHz. At present, ABR audiometry is applicable in only about half of AN cases. Therefore, the need for early diagnosis must be further emphasized. Key words: acoustic neuroma, audiological findings.
INTRODUCTION Hearing loss is the most frequent symptom in cases of acoustic neuroma (AN). In our series, hearing loss was found in 94.6% of the patients at the initial examination. The audiological tests that are particularly important for early diagnosis of AN are pure tone audiometry, speech discrimination test, stapedius reflex (SR) test, and auditory brain stem response ( ABR) audiometry. This paper explains the battery of audiological tests conducted preoperatively in our series. MATERIAL AND METHODS Cases operated on by the authors in which AN was confirmed histopathologically were examined. The number of patients tested varies somewhat for each examination. Psychological audiometry (pure tone audiometry and speech discrimination) was carried out in 132 cases of unilateral AN. The SR threshold was measured in 64 cases (60 of unilateral AN and 4 of bilateral AN) with hearing levels (HL) of 70 dB or lower at frequencies from 0.5 kHz to 2 kHz. The Madsen ZO 72 was used in 39 cases and the Rion RS 20 In 25 cases. In the SR tests, 48 cases of unilateral sensorineural deafness were studied as controls. The SR decay test was carried out according to the method of Anderson et al. ( I ) , by applying stimulus at 10 dB above the threshold for 10 s. Measurements were made at 0.5 kHz and 1 kHz, and a reduction in amplitude of 50% or more was considered positive decay. The ABR was examined in 116 cases of unilateral AN. As controls, 120 cases of unilateral sensorineural hearing loss in which AN had been ruled out were entered into the study. Both the tumor group and the control group were divided into those with an average HL at 4 kHz and 8 kHz of 70 dB or lower (tumor group A, 60 cases, and control group, 90 cases) and those with an average HL of 71 dB or higher (tumor group B, 56 cases, and control group,
126 J. ffinzaki et al. ~~~~~~~~~~~~~~
Table I. Distribution of average 5-frequency hearing level found at initial examination Hearing level (dB)
---
10 20 30
-40 50
-60 70
Downloaded by [Australian National University] at 21:52 26 April 2016
N
80 90 100 105
Total
No.
%
8 8 15
0.8 6.2 10.9 12.4 14.0 10.9 13.2 7.8 6.2 6.2 11.6
129
100.0
1
8 14 16 18 14 17
LO
30 cases). ABR recordings were made from the vertex and mastoid using click stimuli at 70 dBnHL. The average of 2 048 recordings made over a range of 100 Hz- 1 000 Hz with an evoked potential analyser (MEB 5100, Nihon Koden) was calculated. RESULTS 1. Pure tone audiometric findings
The average HL at 5 frequencies determined by pure tone audiometry at the initial examination ranged from 6 dB (near-normal) to 103 dB (totally deaf). The distribution pattern showed two peaks, one within the range of moderate hearing loss at 31-70 dB and one within the range of severe hearing loss at 101 dB or higher. The average HL overall was 58.4 dB (Table I). Next, the cases were divided into three periods according to the date of the patient’s first visit: Period I (1980 and earlier), Period I1 (1981-1985), and Period I11 (1986- 1989). There were no significant differences in the distribution patterns of the average HL among the three periods. The relationship between the time elapsed from the onset of symptoms to the first visit and the average HL was investigated. While the findings varied widely, there was a tendency for the HL to be higher among those cases with longer time elapsed from onset to the first visit. The average audiogram of the 132 cases showed a high-frequency gradual-loss type of configuration. The highest proportion of cases, i.e., 25.8%, showed flat audiometric configurations, while 19.7% showed high-frequency sudden-loss configurations, 12.1% showed high-frequency gradual-loss configurations, 1 1.4% were deaf, and 10.6% showed irregular configurations. The relationship of the audiometric configuration, time elapsed from onset to first visit, and tumor size was investigated. The time elapsed from onset to first visit tended to be longer in cases with high-frequency sudden-loss audiometric configurations, high-frequency gradualloss configurations, and deafness, while tumor size tended to be larger in those with total deafness and near-total deafness. In addition, in the few cases with low-frequency hearing loss, tumor size tended to be large even in cases with little time elapsed from onset to first visit. A U-shaped audiometric configuration was frequently seen in cases of small tumors. In a large proportion of these cases, the HL was most elevated at 2 kHz. While the average audiogram of these U-shaped configurations showed a high-frequency gradual-loss type, the hearing level at 2 kHz tended to be slightly higher than that at 4 kHz.
Audiological findings in acoustic neuroma
Table 11. Distribution of maximum speech discrimination Max. Speech Discrim. Score (YO)
---
0 10 20 * 30 40
--
50
* 60
70
Downloaded by [Australian National University] at 21:52 26 April 2016
80 * 90
-
100
Total
No.
%
30 13 19 7 4 11 13 4 13 9 7
24.8
121
100.0
10.7
8.3 5.8 3.3 9.1 10.7
3.3 10.7
7.4 5.8
2. Speech discrimination The speech discrimination score in 132 cases of AN ranged from 0% to loo%, and averaged 36.7% (Table 11). The relationship between the average 5-frequency HL and speech discrimination was investigated. Speech discrimination tended to be poorer in cases with higher HL, but speech discrimination was worse than the pure-tone HL in many cases. The relationship between the time elapsed from onset of symptoms to the first visit and speech discrimination was also investigated. As with pure tone audiometry, speech discrimination tended to be poorer in cases with longer time elapsed from onset to first visit. However, the discrepancy among cases was large, so no apparent characteristic was discerned. There was also no apparent relationship between tumor size and speech discrimination. 3. Bikisy audiometry The findings of Bekesy audiometry were generally classified according to the Jerger type. BCkesy audiometry was performed in 81 of the 132 cases. Of these, 26 (32.1%) were classified as type I, 28 (34.6%) as type 11, 13 ( 16.0%) as type 111, and 14 (17.3%) as type IV. Only 33.3% of the cases were types I11 and IV, although these types are considered characteristic of AN. 4. ABLB and SISI Tests of supra-threshold have conventionally been carried out as part of the test battery for AN in order to determine whether the patient exhibits recruitment. Since the interpretation of the results is indefinite and the rate of obtaining a positive diagnosis is not very high, however, it has become less valuable in the diagnosis of AN. For these reasons, these examinations were carried out only in 45 (34.1%) of the cases in our series. Negative recruitment was found either by the ABLB test or the SISI test in 26 cases (57.8%). Positive recruitment was found by either or both of the tests in 15 cases (33.3%). Evaluation was impossible in the other 4 cases (8.9%). 5 . Stapedius refex test
Because none of the cases in the control group with pure tone HL of 70dB or lower had absence of the SR at a maximum sound pressure of 120 dB, the absence of the SR in cases with HL of 70 dB or lower was considered pathological.
127
Downloaded by [Australian National University] at 21:52 26 April 2016
128 J. Kanzaki et al. The SR reflex was absent in 18 of 64 cases (28.1%) at 0.5 kHz, 20 of 64 cases (31.3%) at 1.O kHz, and 30 of 63 cases (47.6%) at 2.0 kHz, indicating that the rate of absence tended to increase at higher frequencies. The difference in the SR threshold between the affected side and healthy side was 15 dB or more in 10 of 40 cases (25.0%) at 0.5 kHz, 11 of 40 cases (27.5%) at 1.0 kHz, and 12 of 26 cases (46.2%) at 2.0 kHz. Positive SR decay was found in 8 of 26 cases (30.8%) at 0.5 kHz and 9 of 24 cases (37.5%)at 1.0 kHz. Out of the total of 26 cases in which this test was carried out, 13 (50.00/0) had positive decay at one or more frequencies. The relationship between SR threshold elevation and SR decay was investigated. There was only positive SR decay and no threshold elevation in 3 of 21 cases (14.3%) at 0.5 kHz and 5 of 20 cases (25.0%) at 1.0 kHz; these rates were both higher than those of cases with only threshold elevation and no SR decay. When the absence of SR and threshold elevation (a difference of 15 dB or more in the SR threshold between the affected ear and healthy ear) were used as indices for the diagnosis of AN, the positive diagnostic rate (sensitivity) in the present series was 48.3%(28 of 58 cases) at 0.5 kHz, 51.7% (31 of 60) cases) at 1.0 kHz, and 75.00/0(42 fo 56 cases) at 2.0 kHz. There was no difference in the SR findings between cases with tumors of 20 mm or smaller and those with tumors of 21 mm or larger. However, the rate of absence of the SR was higher in cases of cystic tumors than in those of solid tumors. 6. ABR Wave I was present in only 43% of the cases in the B group but in approximately 70-85% of the cases in the other groups. Wave 111 was present in an extremely low proportion of cases (30% or less) in both the A and B tumor groups. Wave V was present in 60% of the cases in the A tumor group but in only 20% of the cases in the B tumor group. The relation between tumor size and the parameters of latency was investigated in the A tumor group. As a result of examination of the correlation coefficients of tumor sue and latency parameters, as well as the Spearman order correlation coefficient, it was found that no correlation existed between tumor size and the following 4 latency parameters: Wave I, Wave 111, IT3, and IPL 1-111. The correlation coefficients for the 4 latency parameters of Wave V, IT5, IPL I-V, and IPD I-V were 0.555, 0.572, 0.564, and 0.573, respectively, showing a linear correspondence ( p < 0.01). A similar correlation between these parameters and tumor size were seen by the Spearman order correlation coefficients as well ( p
ISSN: 0001-6489 (Print) 1651-2251 (Online) Journal homepage: http://www.tandfonline.com/loi/ioto20
Audiological Findings in Acoustic Neuroma Jin Kanzaki, Kaoru Ogawa, Shigeo Ogawa, Minako Yamamoto, Shunya Ikeda & Toshiaki O-uchi To cite this article: Jin Kanzaki, Kaoru Ogawa, Shigeo Ogawa, Minako Yamamoto, Shunya Ikeda & Toshiaki O-uchi (1991) Audiological Findings in Acoustic Neuroma, Acta Oto-Laryngologica, 111:sup487, 125-132, DOI: 10.3109/00016489109130457 To link to this article: http://dx.doi.org/10.3109/00016489109130457
Published online: 08 Jul 2009.
Submit your article to this journal
Article views: 66
View related articles
Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=ioto20 Download by: [Australian National University]
Date: 26 April 2016, At: 21:52
Acta Otolaryngol (Stockh) 1991; Suppl. 487: 125-132
Audiological Findings in Acoustic Neuroma JIN KANZAKI, KAORU OGAWA, SHIGEO OGAWA, MINAKO YAMAMOTO, SHUNYA IKEDA and TOSHIAKI 0-UCHI From the Department of Otolaryngology, School of Medicine, Keio Universily, Tokyo, Japan
Downloaded by [Australian National University] at 21:52 26 April 2016
Kanzaki J, Ogawa K, Ogawa S, Yamamoto M, Ikeda S, 0-Uchi T. Audiological findings in acoustic neuroma. Acta Otolaryngol (Stockh) 1991; Suppl 487: 125- 132 Audiological examinations are vital in the diagnosis of acoustic neuroma. In interpreting their results, however, it is necessary to consider the patient’s hearing level. The most sensitive audiological examination is auditory brain stem response ( ABR) audiometry. Its most useful parameter is the IT5. A U-shaped audiometric configuration suggests AN, since it is seen in 10% of patients with small tumors. Psychological audiometric tests can be excluded from the battery of screening tests since they have low rates of positive diagnosis. The stapedius reflex (SR) test also has a low positive diagnostic rate in cases of small tumors. Even with the parameters of absence of reflex, elevated threshold, and decay combined, the overall SR test has a lower positive diagnostic rate than ABR audiometry. Nevertheless, the SR test can be employed as a screening device in cases in which the hearing level at 2 kHz and lower is 70 dB or lower, even if it is 71 dB or higher at 4 kHz and 8 kHz. At present, ABR audiometry is applicable in only about half of AN cases. Therefore, the need for early diagnosis must be further emphasized. Key words: acoustic neuroma, audiological findings.
INTRODUCTION Hearing loss is the most frequent symptom in cases of acoustic neuroma (AN). In our series, hearing loss was found in 94.6% of the patients at the initial examination. The audiological tests that are particularly important for early diagnosis of AN are pure tone audiometry, speech discrimination test, stapedius reflex (SR) test, and auditory brain stem response ( ABR) audiometry. This paper explains the battery of audiological tests conducted preoperatively in our series. MATERIAL AND METHODS Cases operated on by the authors in which AN was confirmed histopathologically were examined. The number of patients tested varies somewhat for each examination. Psychological audiometry (pure tone audiometry and speech discrimination) was carried out in 132 cases of unilateral AN. The SR threshold was measured in 64 cases (60 of unilateral AN and 4 of bilateral AN) with hearing levels (HL) of 70 dB or lower at frequencies from 0.5 kHz to 2 kHz. The Madsen ZO 72 was used in 39 cases and the Rion RS 20 In 25 cases. In the SR tests, 48 cases of unilateral sensorineural deafness were studied as controls. The SR decay test was carried out according to the method of Anderson et al. ( I ) , by applying stimulus at 10 dB above the threshold for 10 s. Measurements were made at 0.5 kHz and 1 kHz, and a reduction in amplitude of 50% or more was considered positive decay. The ABR was examined in 116 cases of unilateral AN. As controls, 120 cases of unilateral sensorineural hearing loss in which AN had been ruled out were entered into the study. Both the tumor group and the control group were divided into those with an average HL at 4 kHz and 8 kHz of 70 dB or lower (tumor group A, 60 cases, and control group, 90 cases) and those with an average HL of 71 dB or higher (tumor group B, 56 cases, and control group,
126 J. ffinzaki et al. ~~~~~~~~~~~~~~
Table I. Distribution of average 5-frequency hearing level found at initial examination Hearing level (dB)
---
10 20 30
-40 50
-60 70
Downloaded by [Australian National University] at 21:52 26 April 2016
N
80 90 100 105
Total
No.
%
8 8 15
0.8 6.2 10.9 12.4 14.0 10.9 13.2 7.8 6.2 6.2 11.6
129
100.0
1
8 14 16 18 14 17
LO
30 cases). ABR recordings were made from the vertex and mastoid using click stimuli at 70 dBnHL. The average of 2 048 recordings made over a range of 100 Hz- 1 000 Hz with an evoked potential analyser (MEB 5100, Nihon Koden) was calculated. RESULTS 1. Pure tone audiometric findings
The average HL at 5 frequencies determined by pure tone audiometry at the initial examination ranged from 6 dB (near-normal) to 103 dB (totally deaf). The distribution pattern showed two peaks, one within the range of moderate hearing loss at 31-70 dB and one within the range of severe hearing loss at 101 dB or higher. The average HL overall was 58.4 dB (Table I). Next, the cases were divided into three periods according to the date of the patient’s first visit: Period I (1980 and earlier), Period I1 (1981-1985), and Period I11 (1986- 1989). There were no significant differences in the distribution patterns of the average HL among the three periods. The relationship between the time elapsed from the onset of symptoms to the first visit and the average HL was investigated. While the findings varied widely, there was a tendency for the HL to be higher among those cases with longer time elapsed from onset to the first visit. The average audiogram of the 132 cases showed a high-frequency gradual-loss type of configuration. The highest proportion of cases, i.e., 25.8%, showed flat audiometric configurations, while 19.7% showed high-frequency sudden-loss configurations, 12.1% showed high-frequency gradual-loss configurations, 1 1.4% were deaf, and 10.6% showed irregular configurations. The relationship of the audiometric configuration, time elapsed from onset to first visit, and tumor size was investigated. The time elapsed from onset to first visit tended to be longer in cases with high-frequency sudden-loss audiometric configurations, high-frequency gradualloss configurations, and deafness, while tumor size tended to be larger in those with total deafness and near-total deafness. In addition, in the few cases with low-frequency hearing loss, tumor size tended to be large even in cases with little time elapsed from onset to first visit. A U-shaped audiometric configuration was frequently seen in cases of small tumors. In a large proportion of these cases, the HL was most elevated at 2 kHz. While the average audiogram of these U-shaped configurations showed a high-frequency gradual-loss type, the hearing level at 2 kHz tended to be slightly higher than that at 4 kHz.
Audiological findings in acoustic neuroma
Table 11. Distribution of maximum speech discrimination Max. Speech Discrim. Score (YO)
---
0 10 20 * 30 40
--
50
* 60
70
Downloaded by [Australian National University] at 21:52 26 April 2016
80 * 90
-
100
Total
No.
%
30 13 19 7 4 11 13 4 13 9 7
24.8
121
100.0
10.7
8.3 5.8 3.3 9.1 10.7
3.3 10.7
7.4 5.8
2. Speech discrimination The speech discrimination score in 132 cases of AN ranged from 0% to loo%, and averaged 36.7% (Table 11). The relationship between the average 5-frequency HL and speech discrimination was investigated. Speech discrimination tended to be poorer in cases with higher HL, but speech discrimination was worse than the pure-tone HL in many cases. The relationship between the time elapsed from onset of symptoms to the first visit and speech discrimination was also investigated. As with pure tone audiometry, speech discrimination tended to be poorer in cases with longer time elapsed from onset to first visit. However, the discrepancy among cases was large, so no apparent characteristic was discerned. There was also no apparent relationship between tumor size and speech discrimination. 3. Bikisy audiometry The findings of Bekesy audiometry were generally classified according to the Jerger type. BCkesy audiometry was performed in 81 of the 132 cases. Of these, 26 (32.1%) were classified as type I, 28 (34.6%) as type 11, 13 ( 16.0%) as type 111, and 14 (17.3%) as type IV. Only 33.3% of the cases were types I11 and IV, although these types are considered characteristic of AN. 4. ABLB and SISI Tests of supra-threshold have conventionally been carried out as part of the test battery for AN in order to determine whether the patient exhibits recruitment. Since the interpretation of the results is indefinite and the rate of obtaining a positive diagnosis is not very high, however, it has become less valuable in the diagnosis of AN. For these reasons, these examinations were carried out only in 45 (34.1%) of the cases in our series. Negative recruitment was found either by the ABLB test or the SISI test in 26 cases (57.8%). Positive recruitment was found by either or both of the tests in 15 cases (33.3%). Evaluation was impossible in the other 4 cases (8.9%). 5 . Stapedius refex test
Because none of the cases in the control group with pure tone HL of 70dB or lower had absence of the SR at a maximum sound pressure of 120 dB, the absence of the SR in cases with HL of 70 dB or lower was considered pathological.
127
Downloaded by [Australian National University] at 21:52 26 April 2016
128 J. Kanzaki et al. The SR reflex was absent in 18 of 64 cases (28.1%) at 0.5 kHz, 20 of 64 cases (31.3%) at 1.O kHz, and 30 of 63 cases (47.6%) at 2.0 kHz, indicating that the rate of absence tended to increase at higher frequencies. The difference in the SR threshold between the affected side and healthy side was 15 dB or more in 10 of 40 cases (25.0%) at 0.5 kHz, 11 of 40 cases (27.5%) at 1.0 kHz, and 12 of 26 cases (46.2%) at 2.0 kHz. Positive SR decay was found in 8 of 26 cases (30.8%) at 0.5 kHz and 9 of 24 cases (37.5%)at 1.0 kHz. Out of the total of 26 cases in which this test was carried out, 13 (50.00/0) had positive decay at one or more frequencies. The relationship between SR threshold elevation and SR decay was investigated. There was only positive SR decay and no threshold elevation in 3 of 21 cases (14.3%) at 0.5 kHz and 5 of 20 cases (25.0%) at 1.0 kHz; these rates were both higher than those of cases with only threshold elevation and no SR decay. When the absence of SR and threshold elevation (a difference of 15 dB or more in the SR threshold between the affected ear and healthy ear) were used as indices for the diagnosis of AN, the positive diagnostic rate (sensitivity) in the present series was 48.3%(28 of 58 cases) at 0.5 kHz, 51.7% (31 of 60) cases) at 1.0 kHz, and 75.00/0(42 fo 56 cases) at 2.0 kHz. There was no difference in the SR findings between cases with tumors of 20 mm or smaller and those with tumors of 21 mm or larger. However, the rate of absence of the SR was higher in cases of cystic tumors than in those of solid tumors. 6. ABR Wave I was present in only 43% of the cases in the B group but in approximately 70-85% of the cases in the other groups. Wave 111 was present in an extremely low proportion of cases (30% or less) in both the A and B tumor groups. Wave V was present in 60% of the cases in the A tumor group but in only 20% of the cases in the B tumor group. The relation between tumor size and the parameters of latency was investigated in the A tumor group. As a result of examination of the correlation coefficients of tumor sue and latency parameters, as well as the Spearman order correlation coefficient, it was found that no correlation existed between tumor size and the following 4 latency parameters: Wave I, Wave 111, IT3, and IPL 1-111. The correlation coefficients for the 4 latency parameters of Wave V, IT5, IPL I-V, and IPD I-V were 0.555, 0.572, 0.564, and 0.573, respectively, showing a linear correspondence ( p < 0.01). A similar correlation between these parameters and tumor size were seen by the Spearman order correlation coefficients as well ( p
