- Email: [email protected]
Sudden hearing loss as a presenting symptom of acoustic neuroma
Sudden Hearing Loss as a Presenting Symptom of Acoustic Neuroma Moshe Chaimoff, MD, Ben I. Nageris, MD, Jaqueline Sulkes, PhD, Thomas Spitzer, MD, and Marian Kalmanowitz, MD Purpose: Retrocochlear pathological findings may be suggested by findings on the brainstem response and acoustic sensorineural reflex (AR) tests. We describe the incidence of acoustic neuroma presenting as sudden hearing loss (SHL) and the effectiveness of the discrimination (DISC) test, the brainstem-evoked response, and AR test in predicting acoustic neuroma in patients with SHL. Methods: We retrospectively reviewed the charts of all adult patients who presented to our center with sensorineural SHL between 1989 and 1995. Two groups were defined: those with cerebellopontine angle (CPA) tumor and those with negative imaging findings. Results of the brainstem-evoked response, AR, and DISC tests were compared. Results: Forty patients were admitted with sensorineural SHL, of whom 19 (47.5%) had a CPA tumor. The latter group showed a significantly lower mean age and better results for the low frequencies on pure tone audiometry, as well as better brainstem-evoked response test results than the patients with negative imaging findings. There was also a significant difference between the groups for both the affected and unaffected ears on the discrimination test. Conclusion: Acoustic tumors may be a more common cause of sudden sensorineural hearing loss than previously suspected. The DISC test is a useful screening tool for acoustic tumor, whereas the brainstem-evoked response test shows poorer results in affected patients with sensorineural hearing loss than in other subgroups with different signs of acoustic neuroma. We recommend that young patients presenting with mild SHL who have normal results on the AR and brainstem-evoked response tests undergo magnetic resonance imaging to rule out CPA tumor. (Am J Otolaryngoll999;20:157-160. Copyright 0 1999 by W.B. Saunders Company)
Gradually progressive hearing loss is an accepted sign of acoustic neuromas, whereas sudden sensorineural hearing loss (SHL) is usually attributable to a viral or vascular cause, perilymphatic fistula, acute labyrinthitis, or trauma. However, since the early reports of Edwards and Paterson1 and Hallberg, the increasingly widespread use of computerized tomography and magnetic resonance imaging (MRI) has indicated the rate of SHL in acoustic neuromas may be as high as 10% to 19%.3-5 Retrocochlear pathological findings can be suggested by findings on auditory brainstemFrom the Department of Otolaryngology, Head and Neck Surgery, and Epidemiology Unit, Rabin Medical Center, Petah Tiqva, Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel. Address reprint requests to Moshe Chaimoff, MD, Department of Otolaryngology, Head & Neck Surgery, Rabin Medical Center, Beilinson Campus, Petah Tiqva 49100, Israel. Copyright 0 1999 by W.B. Saunders Company 0196-0709/99/2003-001 0$10.00/O American
Journal
of Otolaryngology,
Vol20,
evoked response (ABR) and acoustic reflex (AR) tests6 The aim of the present study was to describe the incidence of acoustic neuroma presenting as SHL and to examine the effectiveness of the discrimination (DISC) test, AR test, and ABR in predicting acoustic neuroma in patients with SHL.
MATERIALS
AND METHODS
We retrospectively reviewed the charts of all adult patients who presented to our center with SHL between 1989 and 1995. Criteria for SHL were isolated unilateral hearing impairment occurring suddenly or within a few days, an interaural difference of at least 10 dB in at least two frequencies on pure tone audiometry, and absence of external or middle ear pathology, Meniere’s disease, perilymphatic fistula, or systemic disease. In all cases, pure tone audiometry, DISC test, AR test, and ABR were performed on admission. Hearing evaluations consisted of pure tone air and bone conduction, speech reception thresholds, and word No 3 (May-June),
1999:
pp 157-l
60
157
158
CHAIMOFF
recognition testing. Audiometric testing was conducted in double-walled soundproof rooms. The AR test used a maximum stimulus level of 120 decibel hearing level (dBHL); any reflex level that was higher was considered absent. ABR was defined as abnormal when the interaural difference for the wave I to III or III to V interpeak latencies was 0.2 milliseconds or greater. All patients underwent computed tomography or MRI, and findings on the hearing tests were correlated with the size of the acoustic neuroma, if present. Two groups of SHL patients were defined: (1) those diagnosed as having cerebellopontine angle (CPA) tumor (study group); and (2) those with negative imaging findings (control group).
Statistical
Analysis
Pearson’s correlation coefficients (r) and their significance (P) were calculated between parameters. Significance of the differences in continuous parameters (age, audiometry, ABR) between the two groups of patients was determined with Student’s t-test and, in categorical parameters (affected side, sex distribution, AR), with chi-squared test or Fisher’s exact test, as appropriate. P of .05 or less were considered statistically significant.
RESULTS Forty patients admitted with SHL were identified; 22 women and 18 men aged 23 to 79 years (mean, 51.5 years). Sixteen patients (40%) had SHL in the right ear and 24 patients in the left ear. Nineteen patients (47.5%) were diagnosed as having a CPA tumor and 21 were
ET AL
lesion free. Comparison of these groups yielded a significantly lower mean age in the tumor group (47 + 13.4 ~55.6 ? 13.3 years; P = .047),
but no differences in sex distribution or side affected. The tumor group also showed significantly better results for the low frequencies (0.25 and 0.5 kHZ) on pure tone test (33.9 and 42.1 dBHL, respectively v 60 and 64 dBHL, respectively; P = .0003; Fig 1) and a significantly prolonged interaural difference on ABR (0.257 + 0.17 ~0.134 2 0.13; P = .265). DISC scores showed a significant difference between the groups for both the affected and unaffected ears (Table 1). The correlations of tumor size with ABR scores (r = .75; P = .14) and of the AR test for the affected and nonaffected ear with tumor size (r = .44; P = .27; and r = -.22; P = .60, respectively) were not statistically significant. DISCUSSION Four theories of the cause of SHL in acoustic neuroma patients have been suggested: (1) direct pressure on the cochlear nerve, (2) pressure on the vessels supplying the inner ear, (3) biochemical changes in inner ear fluid, and (4) conduction block of the remaining cochlear nerve fibers.7,8 The present study indicates that SHL may be the presenting symptom of acoustic neuroma in a higher number of patients than currently accepted when more careful atten-
db 0 10 20 30 40 50 60 70 80 su 100 110 W
Fig 1. Audiometric test: frequency versus pure tone Intensity. (X), Tumor group; (O), control group.
SUDDEN
HEARING
TABLE 1.
LOSS
Discrimination
AND
Score
ACOUSTIC
in Tumor
159
NEUROMA
and Control
Groups
Affected ear Nonaffected ear NOTE.
Values
Tumor Group (n = 19)
Control Group (n = 21)
P
60.13 98.31
k 27.42 ? 2.42
78.22 95.57
.03 .02
as mean
2 standard
expressed
? 18.1 2 4.96 deviation.
tion is directed to the presence of a small tumor. Our 47.5% rate (19 of 40 patients) is very high compared with the 13.9% reported by Pensak et al3 and the 19% by others.4-5 However, it may be partially attributable to the pecularities of the Israeli population, which consists of a large number of immigrants from developing countries and partially to our very high index of suspicion for acoustic neuroma in every SHL patient, even those with minor impairment of 10 dB. It should be emphasized that normally we would define SHL as a greater than lo-dB change in three frequencies. Our definition, mentioned in the Material and Methods section, regards SHL as a change of greater than 10 dB in two frequencies to better delineate the tumor group. This is supported by the audiometry test: as shown in Fig 1, the presence of a statistically significant difference between the tumor and control groups at 250 HZ and 500 HZ. At 1 kHZ, there is a tendency toward a difference between the groups that did not reach statistical significance. We will continue to check the incidence of acoustic neuroma to confirm these findings. Although the age of presentation for the entire SHL cohort did not differ from that for acoustic tumor in the literature (30 to 60 years), the age of the CPA group, when analyzed separately (47 2 13.44 years), was significantly lower than that of the controls (55.6 + 13.28 years). This may indicate that the younger the patient with SHL, the more likely the cause is acoustic tumor. Neither the side affected nor sex was a predisposing factor for acoustic neuroma. Acoustic neuroma may present with any audiometric pattern, even normal hearing. Based on our audiometry findings, however, we were able to delineate a pure tone threshold pattern in SHL patients that may be predictive of acoustic neuroma (Fig 1). The lower level of hearing loss among the tumor group may be caused by well-circumscribed pres-
sure damage at the small site of a growing tumor, as opposed to the more general and severe damage caused by viral infection, vascular accident, perilymphatic fistula, acute labyrinthitis, or skull base trauma. It is well established that DISC findings of asymmetric SHL or speech impairment that are disproportionate to the pure tone loss requires a specific investigation for a retrocochlear lesion. This was also true in our study. However, as opposed to reports in the literature,5 we could not distinguish the acoustic tumor from the nontumor group based on AR testing. This may suggest that the eighth nerve is affected by the same mechanism in all SHL patients regardless of the cause. Conversely, the tumors in our group may have been diagnosed at an early stage, when the AR is still normal.g The ABR is considered the most reliable and sensitive electrophysiological examination for acoustic neuroma, with an associated falsenegative rate of 5% or less.lOJ1 We obtained abnormal ABR results for all tumor patients except three, for a false-negative rate of 15%. This emphasizes the importance of imaging in all suspected cases. The lack of correlation between tumor size and ABR findings may have been caused by the small number of patients for whom information on tumor size was available. It should be emphasized that the smaller the tumor at diagnosis, the better the surgical results with regard to preservation of hearing and facial nerve function.
CONCLUSION The increasing use of state-of-the-art technologies has indicated that acoustic tumors may be a more common cause of SHL than previously suspected. The presence of a specific audiometric pure tone threshold at the low frequencies (0.25 and 0.5 kHZ) may be predictive of tumor in some SHL patients. DISC is also useful as a screening tool for acoustic tumor. ABR shows poorer results in SHL patients than in other subgroups with different signs of acoustic neuroma, emphasizing the importance of imaging in these specific populations. We recommend that patients presenting with mild SHL should undergo MRI to
CHAIMOFF
160
rule out CPA tumor, regardless of AR and ABR test results. REFERENCES 1. Edwards CH, Paterson JH: A review of the symptoms and signs of acoustic neurofibromata. Brain 74:144-190, 1951 2. Hallberg OE: Sudden deafness of obscure origin. Laryngoscope 66:1237-1267,1956 3. Pensak ML, Glasscock ME III, Josey AF, et al: Sudden hearing loss and cerebellopontine angle tumors. Laryngoscope 95:1188-1193,1985 4. Berg HM, Cohen NL, Hammerschlag PE, et al: Acoustic neuroma presenting as sudden hearing loss with recovery. Otolaryngol Head Neck Surg 94:15-22,1986 5. Yanagihara N, Asai M: Sudden hearing loss induced by acoustic neuroma: Significance of small tumors. Laryngoscope 103:308-311,1993
ET AL
6. Burky JM, Rizer FM, Lippy WH, et al: Acoustic reflexes, auditory brainstem response, and MRl in the evaluation of acoustic neuromas. Laryngoscope 106:839-
841,1996 7. DeMoura
LFP, Hayden RD, Conner GH: Further observation on acoustic neuroma. Trans Am Acad Ophthalmol Otolaryngol 73:60-70,1969 8. Neely JG: Gross and microscopic anatomy of the eighth cranial nerve in relationship to the solitary schwannoma. Laryngoscope 91:1512-1531,198l 9. Sakakibara A, Aoyagi A, Koike Y: Acoustic neuroma presented as repeated hearing loss. Acta Otolaryngol
511:77-80,1994 10. Josey AF, Glasscock ME, Musiek ABR and medical imaging in patients tine angle tumors. Am J Otol9:6-12,1988 11. Telian SA, Kleny PR, Niparko auditory response in patients with Laryngoscope 99:4-lo,1989
FE: Correlation with cerebelloponJK, et al: Normal acoustic neuroma.
of
Gradually progressive hearing loss is an accepted sign of acoustic neuromas, whereas sudden sensorineural hearing loss (SHL) is usually attributable to a viral or vascular cause, perilymphatic fistula, acute labyrinthitis, or trauma. However, since the early reports of Edwards and Paterson1 and Hallberg, the increasingly widespread use of computerized tomography and magnetic resonance imaging (MRI) has indicated the rate of SHL in acoustic neuromas may be as high as 10% to 19%.3-5 Retrocochlear pathological findings can be suggested by findings on auditory brainstemFrom the Department of Otolaryngology, Head and Neck Surgery, and Epidemiology Unit, Rabin Medical Center, Petah Tiqva, Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel. Address reprint requests to Moshe Chaimoff, MD, Department of Otolaryngology, Head & Neck Surgery, Rabin Medical Center, Beilinson Campus, Petah Tiqva 49100, Israel. Copyright 0 1999 by W.B. Saunders Company 0196-0709/99/2003-001 0$10.00/O American
Journal
of Otolaryngology,
Vol20,
evoked response (ABR) and acoustic reflex (AR) tests6 The aim of the present study was to describe the incidence of acoustic neuroma presenting as SHL and to examine the effectiveness of the discrimination (DISC) test, AR test, and ABR in predicting acoustic neuroma in patients with SHL.
MATERIALS
AND METHODS
We retrospectively reviewed the charts of all adult patients who presented to our center with SHL between 1989 and 1995. Criteria for SHL were isolated unilateral hearing impairment occurring suddenly or within a few days, an interaural difference of at least 10 dB in at least two frequencies on pure tone audiometry, and absence of external or middle ear pathology, Meniere’s disease, perilymphatic fistula, or systemic disease. In all cases, pure tone audiometry, DISC test, AR test, and ABR were performed on admission. Hearing evaluations consisted of pure tone air and bone conduction, speech reception thresholds, and word No 3 (May-June),
1999:
pp 157-l
60
157
158
CHAIMOFF
recognition testing. Audiometric testing was conducted in double-walled soundproof rooms. The AR test used a maximum stimulus level of 120 decibel hearing level (dBHL); any reflex level that was higher was considered absent. ABR was defined as abnormal when the interaural difference for the wave I to III or III to V interpeak latencies was 0.2 milliseconds or greater. All patients underwent computed tomography or MRI, and findings on the hearing tests were correlated with the size of the acoustic neuroma, if present. Two groups of SHL patients were defined: (1) those diagnosed as having cerebellopontine angle (CPA) tumor (study group); and (2) those with negative imaging findings (control group).
Statistical
Analysis
Pearson’s correlation coefficients (r) and their significance (P) were calculated between parameters. Significance of the differences in continuous parameters (age, audiometry, ABR) between the two groups of patients was determined with Student’s t-test and, in categorical parameters (affected side, sex distribution, AR), with chi-squared test or Fisher’s exact test, as appropriate. P of .05 or less were considered statistically significant.
RESULTS Forty patients admitted with SHL were identified; 22 women and 18 men aged 23 to 79 years (mean, 51.5 years). Sixteen patients (40%) had SHL in the right ear and 24 patients in the left ear. Nineteen patients (47.5%) were diagnosed as having a CPA tumor and 21 were
ET AL
lesion free. Comparison of these groups yielded a significantly lower mean age in the tumor group (47 + 13.4 ~55.6 ? 13.3 years; P = .047),
but no differences in sex distribution or side affected. The tumor group also showed significantly better results for the low frequencies (0.25 and 0.5 kHZ) on pure tone test (33.9 and 42.1 dBHL, respectively v 60 and 64 dBHL, respectively; P = .0003; Fig 1) and a significantly prolonged interaural difference on ABR (0.257 + 0.17 ~0.134 2 0.13; P = .265). DISC scores showed a significant difference between the groups for both the affected and unaffected ears (Table 1). The correlations of tumor size with ABR scores (r = .75; P = .14) and of the AR test for the affected and nonaffected ear with tumor size (r = .44; P = .27; and r = -.22; P = .60, respectively) were not statistically significant. DISCUSSION Four theories of the cause of SHL in acoustic neuroma patients have been suggested: (1) direct pressure on the cochlear nerve, (2) pressure on the vessels supplying the inner ear, (3) biochemical changes in inner ear fluid, and (4) conduction block of the remaining cochlear nerve fibers.7,8 The present study indicates that SHL may be the presenting symptom of acoustic neuroma in a higher number of patients than currently accepted when more careful atten-
db 0 10 20 30 40 50 60 70 80 su 100 110 W
Fig 1. Audiometric test: frequency versus pure tone Intensity. (X), Tumor group; (O), control group.
SUDDEN
HEARING
TABLE 1.
LOSS
Discrimination
AND
Score
ACOUSTIC
in Tumor
159
NEUROMA
and Control
Groups
Affected ear Nonaffected ear NOTE.
Values
Tumor Group (n = 19)
Control Group (n = 21)
P
60.13 98.31
k 27.42 ? 2.42
78.22 95.57
.03 .02
as mean
2 standard
expressed
? 18.1 2 4.96 deviation.
tion is directed to the presence of a small tumor. Our 47.5% rate (19 of 40 patients) is very high compared with the 13.9% reported by Pensak et al3 and the 19% by others.4-5 However, it may be partially attributable to the pecularities of the Israeli population, which consists of a large number of immigrants from developing countries and partially to our very high index of suspicion for acoustic neuroma in every SHL patient, even those with minor impairment of 10 dB. It should be emphasized that normally we would define SHL as a greater than lo-dB change in three frequencies. Our definition, mentioned in the Material and Methods section, regards SHL as a change of greater than 10 dB in two frequencies to better delineate the tumor group. This is supported by the audiometry test: as shown in Fig 1, the presence of a statistically significant difference between the tumor and control groups at 250 HZ and 500 HZ. At 1 kHZ, there is a tendency toward a difference between the groups that did not reach statistical significance. We will continue to check the incidence of acoustic neuroma to confirm these findings. Although the age of presentation for the entire SHL cohort did not differ from that for acoustic tumor in the literature (30 to 60 years), the age of the CPA group, when analyzed separately (47 2 13.44 years), was significantly lower than that of the controls (55.6 + 13.28 years). This may indicate that the younger the patient with SHL, the more likely the cause is acoustic tumor. Neither the side affected nor sex was a predisposing factor for acoustic neuroma. Acoustic neuroma may present with any audiometric pattern, even normal hearing. Based on our audiometry findings, however, we were able to delineate a pure tone threshold pattern in SHL patients that may be predictive of acoustic neuroma (Fig 1). The lower level of hearing loss among the tumor group may be caused by well-circumscribed pres-
sure damage at the small site of a growing tumor, as opposed to the more general and severe damage caused by viral infection, vascular accident, perilymphatic fistula, acute labyrinthitis, or skull base trauma. It is well established that DISC findings of asymmetric SHL or speech impairment that are disproportionate to the pure tone loss requires a specific investigation for a retrocochlear lesion. This was also true in our study. However, as opposed to reports in the literature,5 we could not distinguish the acoustic tumor from the nontumor group based on AR testing. This may suggest that the eighth nerve is affected by the same mechanism in all SHL patients regardless of the cause. Conversely, the tumors in our group may have been diagnosed at an early stage, when the AR is still normal.g The ABR is considered the most reliable and sensitive electrophysiological examination for acoustic neuroma, with an associated falsenegative rate of 5% or less.lOJ1 We obtained abnormal ABR results for all tumor patients except three, for a false-negative rate of 15%. This emphasizes the importance of imaging in all suspected cases. The lack of correlation between tumor size and ABR findings may have been caused by the small number of patients for whom information on tumor size was available. It should be emphasized that the smaller the tumor at diagnosis, the better the surgical results with regard to preservation of hearing and facial nerve function.
CONCLUSION The increasing use of state-of-the-art technologies has indicated that acoustic tumors may be a more common cause of SHL than previously suspected. The presence of a specific audiometric pure tone threshold at the low frequencies (0.25 and 0.5 kHZ) may be predictive of tumor in some SHL patients. DISC is also useful as a screening tool for acoustic tumor. ABR shows poorer results in SHL patients than in other subgroups with different signs of acoustic neuroma, emphasizing the importance of imaging in these specific populations. We recommend that patients presenting with mild SHL should undergo MRI to
CHAIMOFF
160
rule out CPA tumor, regardless of AR and ABR test results. REFERENCES 1. Edwards CH, Paterson JH: A review of the symptoms and signs of acoustic neurofibromata. Brain 74:144-190, 1951 2. Hallberg OE: Sudden deafness of obscure origin. Laryngoscope 66:1237-1267,1956 3. Pensak ML, Glasscock ME III, Josey AF, et al: Sudden hearing loss and cerebellopontine angle tumors. Laryngoscope 95:1188-1193,1985 4. Berg HM, Cohen NL, Hammerschlag PE, et al: Acoustic neuroma presenting as sudden hearing loss with recovery. Otolaryngol Head Neck Surg 94:15-22,1986 5. Yanagihara N, Asai M: Sudden hearing loss induced by acoustic neuroma: Significance of small tumors. Laryngoscope 103:308-311,1993
ET AL
6. Burky JM, Rizer FM, Lippy WH, et al: Acoustic reflexes, auditory brainstem response, and MRl in the evaluation of acoustic neuromas. Laryngoscope 106:839-
841,1996 7. DeMoura
LFP, Hayden RD, Conner GH: Further observation on acoustic neuroma. Trans Am Acad Ophthalmol Otolaryngol 73:60-70,1969 8. Neely JG: Gross and microscopic anatomy of the eighth cranial nerve in relationship to the solitary schwannoma. Laryngoscope 91:1512-1531,198l 9. Sakakibara A, Aoyagi A, Koike Y: Acoustic neuroma presented as repeated hearing loss. Acta Otolaryngol
511:77-80,1994 10. Josey AF, Glasscock ME, Musiek ABR and medical imaging in patients tine angle tumors. Am J Otol9:6-12,1988 11. Telian SA, Kleny PR, Niparko auditory response in patients with Laryngoscope 99:4-lo,1989
FE: Correlation with cerebelloponJK, et al: Normal acoustic neuroma.
of