Auditory brainstem response threshold differences in patients with vestibular schwannoma: a new diagnostic index.
Auditory brainstem response (ABR) testing is less sensitive in detecting small vestibular schwannomas than medium-size tumors. Magnetic resonance imaging (MRI) is more sensitive than ABR alone for small and large tumors, but it carries with it increased cost and issues of unavailability and patient discomfort. We conducted a prospective pilot study of 7 patients with untreated MRI-proven, unilateral vestibular schwannoma to determine if we could increase the sensitivity of ABR testing in detecting small tumors. Our method involved the use of a new ABR index that is based on threshold differences. All patients underwent pure-tone audiometry followed by a determination of behavioral threshold and neurodiagnostic threshold ABR in the normal ear, which was used as a control, and in the diseased ear. Analysis of results revealed that all 7 patients had an abnormal ABR threshold difference, and 5 patients displayed abnormal traditional ABR indices. The mean difference between the ABR and behavioral click thresholds was 41.4 dB in the diseased ears (with the ABR threshold being higher than the click threshold) and 15.8 dB in the normal ears. None of the control ears had a threshold difference >30 dB.
The use of auditory brainstem response (ABR) testing in the screening of retrocochlear pathology such as vestibular schwannomas is widespread. Since ABR testing was first described by Selters and Brackmann in 1977, (1) multiple studies (2-4) have shown that its sensitivity exceeds 90%, thus establishing it as the most sensitive audiologic test for the detection of vestibular schwannomas. However, within the past decade, magnetic resonance imaging (MRI) has emerged as the gold standard for the diagnosis and monitoring of vestibular schwannomas. MRI is capable of detecting vestibular schwannomas of any size, while ABR testing tends to be less sensitive for smaller lesions. For example, Schmidt et al reported that ABR testing was only 58% sensitive for detecting lesions 1 cm or smaller that were readily detected by MRI. (5) The usefulness of ABR testing is also limited by compromised audiologic function, which is typically seen in the setting of retrocochlear pathology. While MRI is more sensitive than ABR alone, it is not always readily available and it adds to treatment costs. (6) Moreover, patient discomfort can be an issue. Finally, MRI is contraindicated in many patients because of implants or other types of metal in the body. Computed tomography (CT) is an option, but it is less sensitive than MRI and it may also miss small lesions.
While ABR testing is a good screening tool in terms of availability, cost, and ease of administration, it would be even better if it were more sensitive in detecting smaller lesions. Detection of these lesions at a smaller size may result in earlier treatment and improved preservation of hearing and facial nerve function. (7) One advance in ABR technology, the stacked ABR, has been shown to improve sensitivity in diagnosing vestibular schwannomas, but this option is more costly and time-consuming than traditional ABR testing, (8,9)
Diagnostic indices for the detection of vestibular schwannomas have been documented by many different authors. These indices are highly dependent on the degree of the patient's hearing loss and the presence or absence of waveforms. The I-III interwave interval is a sensitive and specific marker, (10) but these waves are often absent in patients with vestibular schwannoma. The absence of ABR waveforms is a useful screening index for retrocochlear pathology. The I-V interwave interval and the absolute latency of wave V are useful indices, but they are not specific. These indices may still be of diagnostic significance when using cutoff values of [greater than or equal to] 4.4 msec and [greater than or equal to] 6.3 msec, respectively, (11)The interaural latency difference (ILD) of wave V was also described by Selters and Brackmann in 1977. (1) An ILD of [greater than or equal to] 0.3 msec has been reported to greatly assist in vestibular schwannoma detection. (11)
Marangos et al described a unique analysis of ABR thresholds in patients with cerebellopontine angle tumors. (12) They compared the pure-tone average (PTA) for frequencies between 1,000 and 6,000 Hz from the audiogram with the neurodiagnostic threshold obtained with an ABR. They found an average threshold difference of 31.2 dB in patients with cerebellopontine angle pathology and an average difference of only 3.6 dB in a control group. A threshold difference of [greater than or equal to] 30 dB was found in 40.6% of the tumor group and in 0% of the control group. Marangos et al suggested that the threshold difference represents an additional index that should trigger suspicion for retrocochlear pathology. That study provided the basis for our examination of ABR threshold differences in patients with small to medium-sized vestibular schwannomas.
Patients and methods
This prospective pilot study involved patients with untreated vestibular schwannomas. After obtaining institutional review board approval, we recruited patients who met our eligibility criteria, and we obtained informed consent. The primary inclusion criterion was the presence of an untreated unilateral vestibular schwannoma, regardless of size, that could be monitored both clinically and radiographically (MRI with gadolinium contrast) in our neurotology clinic. Patients with a PTA >60 dB in the diseased ear and those with bilateral vestibular schwannomas were excluded from the study.
During the enrollment period (~18 mo), 40 patients with unilateral vestibular schwannomas presented to our clinic. Of these, 33 were excluded on the basis of a PTA >60 dB, previous treatment, or an unwillingness to participate. The charts of the remaining 7 patients were reviewed for demographic information and for information on the site and size of the vestibular schwannoma. Recent MRIs ([less than or equal to] 1 yr old) were also reviewed.
All testing was completed in a sound-treated room or quiet listening environment to reduce ambient noise levels. All patients underwent a full audiometric evaluation using conventional clinical procedures. Thresholds for the octave frequencies from 250 to 8,000 Hz were established with a calibrated GSI 61 audiometer. Patients with air-conduction thresholds >15 dB also underwent bone-conduction testing to rule out conductive involvement. In addition, speech reception thresholds and word recognition scores were also obtained.
The ABR test was re corded with the Intelligent Biologic Master System (Intelligent Hearing Systems; Miami) using traditional montage and recording techniques. Electrodes were placed on the high forehead and at each mastoid. Impedances were maintained at <5 kOhm and balanced across the electrode array. Expanding on the previously described threshold comparison described by Marangos et al, (12) we obtained a behavioral click-stimulus threshold. The click stimulus, which involves a wide band of frequencies (2,000 to 4,000 Hz), was presented at 90 dB and decreased progressively in 10 -dB increments and increased in 5-dB steps (modified Hughson-Westlake procedure) until a threshold was established for each ear. An evoked potential threshold was then obtained and compared with the behavioral threshold, and the difference between the two was calculated. Waveform analysis was restricted to waves I, III, and V at 90 dB. The ABR indices were then recorded; they included the wave I-V interval, the ILD, and the absolute latency of wave V. Our institutional values indicating retrocochlear pathology relative to these indices are [greater than or equal to] 4.4 msec for the I-V interval, [greater than or equal to] 0.4 msec for the ILD, and [greater than or equal to] 6.2 msec for the absolute latency of V. The total amount of testing time for each patient ranged between 15 and 30 minutes.
The 7 patients in our study group were aged 49 to 70 years (mean: 59) (table). Five of the patients had at least one abnormal index. (The number of patients abnormal for each index is depicted in figure 1.) In 3 of the 7 patients (patients 4, 6, and 7), we were unable to detect a wave I, and therefore a wave I-V interval could not be calculated. This was considered an abnormal finding. All 3 of these patients had at least one other abnormal index. In 1 of those 3 patients (patient 7), we were unable to obtain a reliable waveform on the normal side, and therefore the ILD could not be calculated; this patient had a significant sensorineural hearing loss on the normal side, which may have accounted for the difficulty in obtaining normal waveforms. Failure to obtain traditional indices was considered abnormal in this study.
The behavioral threshold obtained with the ABR click was compared with the neurodiagnostic threshold (table). The mean threshold difference in the diseased ears was 41.4 dB (with the ABR threshold being higher than the click threshold), and the mean difference in the normal ears was 15.8 dB (figure 2). A comparison with the normal ear could not be performed in patient 7 because, as mentioned, a reliable waveform and threshold could not be obtained. In accordance with the criteria of Marangos et al, (12) we established a 30-dB threshold difference as indicative of retrocochlear pathology. We found that all 7 patients had an abnormal threshold difference, which meant that our test had a sensitivity of 100% in detecting retrocochlear pathology. None of the normal ears had a threshold difference of >30 dB.
In this study, we evaluated patients with untreated, MRI-proven unilateral vestibular schwannomas with the use of a traditional ABR test and a determination of threshold differences for the purpose of increasing ABR sensitivity. Our goal was not to attempt to replace MRI with ABR. Rather, we advocate the development of a cost-effective yet accurate algorithm for the diagnostic evaluation of patients with asymmetrical auditory symptoms. The application of the threshold difference is completely dependent on the examiner's ability to determine a reliable neurodiagnostic threshold; this is not possible in all patients, but it does hold promise for the development of new indices in vestibular schwannoma detection. The results of ABR testing in patients with significantly diminished auditory function are typically unreliable, and these patients should be evaluated radiographically if asymmetrical symptoms exist. As reflected in our exclusion criteria, we attempted to examine patients with PTAs <60 dB in an attempt to gain more reliable ABR results.
There is much debate in the ABR literature regarding the appropriate index parameters used to identify small vestibular schwannomas. Zappia et al used an ILD of 0.2 msec and reported a sensitivity of 89% in diagnosing tumors < 1 cm. (13) However, Schmidt et al (5) emphasized the importance of using the Brackmann correction factor (1) to account for cochlear loss. We use a parameter of [greater than or equal to] 0.4 msec for ILD, realizing that more lenient parameters may increase our ABR sensitivity but may also increase the number of false positives.
With a sample size of 7 patients, we intended only to document a trend toward abnormal threshold differences in acoustic vestibular schwannoma patients. Testing a larger number of acoustic vestibular schwannoma patients might help validate this index. Our main focus was to examine an additional index of ABR analysis by comparing the thresholds in vestibular schwannoma patients. We found that all 7 patients had an abnormal threshold difference based on the criteria developed by Marangos et al, (12) and 5 of the 7 had an abnormal traditional ABR index. In fact, our data demonstrated 100% sensitivity for the threshold difference.
[FIGURE 1 OMITTED]
[FIGURE 2 OMITTED]
Marangos et al (12) discussed the neural dyssynchrony that occurs in vestibular schwannomas. This phenomenon may account for the disparity between the PTA and the neurodiagnostic threshold, as well as the disparity that is seen between the PTA and word discrimination scores in acoustic vestibular schwannoma patients. While Marangos et al (12) used a PTA of 1,000 to 6,000 Hz, we decided to use the click stimulus (2,000 to 4,000 Hz) to develop our baseline threshold. We feel that since this click presents multiple frequencies simultaneously, it provides a more accurate threshold and it provides internal consistency within the test since the neurodiagnostic threshold is obtained with the same stimulus. We found a mean difference of 41.4 dB between the ABR and the behavioral thresholds in the diseased ear and a mean difference of 15.8 dB in the normal ear. A threshold difference >30 dB may represent an additional index to indicate suspicion of retrocochlear pathology. These audiologic tests are easily performed and can be used in patients who have slightly asymmetrical audiologic or subjective symptoms.
In conclusion, this study focused on developing a more sensitive vestibular schwannoma screening measure using the ABR test that requires only a minimal amount of additional time. We found that ABR threshold differences can augment the sensitivity of the ABR test, especially in patients with small vestibular schwannomas. This study provides insight into improving the value of the ABR test as a screening measure without adding significant time and cost. Further validation of this new index may assist in the subsequent diagnosis and early management of vestibular schwannomas.
(1.) Selters WA, Brackmann DE. Acoustic tumor detection with brain stem electric response audiometry. Arch Otolaryngol 1977; 103(4):181-7.
(2.) Josey AF, Glasscock ME III, Musiek FE. Correlation of ABR and medical imaging in patients with cerebellopontine angle tumors. Am J Otol 1988;9(Suppl):12-16.
(3.) Selesnick SH, Jackler RK. Atypical hearing loss in acoustic neuroma patients. Laryngoscope 1993;103(4 Pt 1):437-41.
(4.) Dornhoffer JL, Helms J, Hoehmann DH. Presentation and diagnosis of small acoustic tumors. Otolaryngol Head Neck Surg 1994; 111 (3 Pt 1):232-5.
(5.) Schmidt RI, Sataloff RT, Newman J, et al. The sensitivity of auditory brainstem response testing for the diagnosis of acoustic neuromas. Arch Otolaryngol Head Neck Surg 2001;127(1):19-22.
(6.) Rupa V, Job A, George M, Rajshekhar V. Cost-effective initial screening for vestibular schwannoma: Auditory brainstem response or magnetic resonance imaging? Otolaryngol Head Neck Surg 2003; 128 (6):823-8.
(7.) Wiegand DA, Ojemann RG, Fickel V. Surgical treatment of acoustic neuroma (vestibular schwannoma) in the United States: Report from the Acoustic Neuroma Registry. Laryngoscope 1996;106(1 Pt 1):58-66.
(8.) Don M, Kwong B, Tanaka C, et al. The stacked ABR: A sensitive and specific screening tool for detecting small acoustic tumors. Audiol Neurootol 2005;10(5):274-90.
(9.) Don M, Masuda A, Nelson R, Brackmann D. Successful detection of small acoustic tumors using the stacked derived-band auditory brain stem response amplitude. Am J Otol 1997;18(5):608-21; discussion 682-5.
(10.) Musiek F, Bornstein SP, Hall JW, Schwaber M. Auditory brainstem response: Neurodiagnostic and intraoperative applications. In: Katz J, ed. Handbook of Clinical Audiology. 4th ed. Baltimore: Williams & Wilkins; 1994:351-74.
(11.) Musiek FE, Shinn JB, Jirsa RE. The auditory brainstem response in auditory nerve and brainstem dysfunction. In: Burkard RF, Eggermont II, Don M, eds. Auditory Evoked Potentials. Philadelphia: Lippincott Williams & Wilkins; 2007:291-312.
(12.) Marangos N, Schipper J, Richter B. Objective auditory brainstem response threshold deficits in patients with cerebellopontine angle tumors [in German]. HNO 1999;47(9):804-8.
(13.) Zappia II, O'Connor CA, Wiet RJ, Dinces EA. Rethinking the use of auditory brainstem response in acoustic neuroma screening. Laryngoscope 1997;107(10):1388-92.
Matthew L. Bush, MD; Raleigh O. Jones, MD; Jennifer B. Shinn, PhD
From the Division of Otolaryngology-Head and Neck Surgery, Department of Surgery, University of Kentucky College of Medicine, Lexington.
Corresponding author: Matthew L. Bush, MD, University of Kentucky College of Medicine, Otolaryngology-HNS, 800 Rose St., Suite C-236, Lexington, KY 40536-0293. Phone: (859) 257-5097; fax: (859) 257-5096; e-mail: firstname.lastname@example.org
Table. Summary of patient demographic information and ABR results * I-V interval ILD Tumor Tumor (msec) (msec) Pt. Age side diameter ([dagger]) ([dagger]) 1 52 Left 3.0 mm 4.2 0.0 2 49 Left 4.0 mm 4.6 0.2 3 60 Left 1.1 cm 5.0 0.8 4 70 Left 1.6 cm CND ([section]) 0.4 5 70 Right 5.0 mm 4.3 0.0 6 58 Left 1.2 mm CND 0.5 7 54 Right 5.0 mm CND CND Absolute latency of V Behavioral ABR Threshold (msec) threshold threshold difference Pt. ([dagger]) (dB) (dB) (dB) 1 6.0 10 (10) 60 (20) 50 ([double dagger]) (10) 2 6.0 10 (25) 50 (30) 40 (5) 3 6.7 15 (5) 50 (20) 35 (15) 4 6.1 35 (15) 90 (40) 55 (25) 5 6.1 25 (25) 60 (50) 35 (25) 6 6.6 15 (15) 50 (30) 35 (15) 7 6.6 45 (35) 85 (CND) 40 (CND) * Figures in parentheses represent findings in the normal ear. ([dagger]) Our institutional values indicating retrocochlear pathology relative to these indices are [greater than or equal to] 4.4 cosec for the I-V interval, [greater than or equal to] 0.4 cosec for the interaural latency difference (ILD), and [greater than or equal to] 6.2 msec for the absolute latency of V. ([double dagger]) Figures in boldface indicate abnormal values. ([section]) CND = could not determine.