Otology & Neurotology

38:118122 2016, Otology & Neurotology, Inc.

Tinnitus Suppression After Auditory Brainstem Implantation

in Patients With Neurofibromatosis Type-2
Daniel S. Roberts, ySteve Otto, zBrian Chen, yKevin A. Peng, jjMarc S. Schwartz,
yDerald E. Brackmann, and yJohn W. House
Division of Otolaryngology-Head and Neck Surgery, University of Connecticut School of Medicine, Farmington, Connecticut;
yHouse Clinic and House Ear Institute, Los Angeles, California; zDepartment of Otolaryngology-Head and Neck Surgery, William
Beaumont Army Medical Center, El Paso, Texas; House Clinic, Neurosurgery, Los Angeles, California; and jjHuntington Medical
Research Institutes, Pasadena, California

Objective: To evaluate whether an auditory brainstem
implant (ABI) can impact levels of tinnitus in neurofibroma-
tosis type-2 (NF2) patients who have undergone translabyr-
inthine craniotomy for vestibular schwannoma (VS) removal
and to evaluate the burden of tinnitus in these patients.
Study Design: A retrospective case series and patient
survey.
Setting: Tertiary neurotologic referral center.
Patients: NF2 patients who underwent translabyrinthine
removal of VS and ABI placement between 1994 and 2015.
Interventions: A survey, retrospective review and two
validated tinnitus handicap questionnaires (tinnitus handicap
inventory [THI] and tinnitus visual analogue scale [VAS])
were used to characterize the degree of tinnitus in NF2
patients and whether an ABI can alter tinnitus levels.
Main Outcome Measures(s): Survey results, THI and VAS
scores.
Results: One hundred twelve ABI users were contacted and
43 patients (38.3)% responded to our survey. Tinnitus was
reported in 83.7% of patients. The THI score for responders
was 17.8  20.5 standard deviation (SD). For survey partici-
pants, the ABI reduced tinnitus levels (mean VAS: Off 3.5;
3.5; On 1-h 2.1; p 0.048). For patients who subjectively
reported that the ABI reduced tinnitus loudness, tinnitus
levels were immediately reduced on ABI activation and after
1 hour of use (mean VAS: Off 4.8; On 2.4; On 1-h 1.8;
p < 0.01). Suppression did not continue after the device was
turned off. Audiological performance with the ABI did not
correlate with tinnitus suppression.
Conclusion: NF2 patients who have undergone removal of
VS have a significant tinnitus handicap and benefit from
tinnitus suppression through utilization of an ABI possibly
through masking or electrical stimulation of the auditory
brainstem. Key Words: Acoustic neuromaAuditory
brainstem implantationNF2TinnitusVestibular
schwannoma.
Otol Neurotol 38:118122, 2017.

Electrical stimulation of the auditory system for
tinnitus suppression is an area of active investigation
(1). Cochlear implantation (CI) reproducibly causes
tinnitus suppression in a majority of patients (2). While
subjective tinnitus is likely a multifactorial entity, it is
hypothesized that tinnitus arises from changes in neural
circuitry caused by a reduction in auditory input. Positron
emission tomography (PET) has shown altered metabolic
brain activity in the auditory cortex and recent evidence
suggests roles for the limbic, frontal, and parietal brain in
the pathophysiology of chronic tinnitus (3).
Address correspondence and reprint requests to John W. House,
M.D., 2100 West 3rd Street, Los Angeles, CA 90057;
E-mail: jhouse@houseclinic.com
Marc S. Schwartz is a consultant and receives grant support from
Cochlear Corporation for unrelated projects. All other authors disclose
no conflicts of interest.
DOI: 10.1097/MAO.0000000000001230
The development of the auditory brainstem implant
(ABI) provided a means of bypassing the cochlea and
auditory nerve to electrically stimulate the second order
neurons of the cochlear nucleus. The device was origin-
ally designed for patients with neurofibromatosis type-2
(NF2), and the vast majority of patients implanted to
date are patients with NF2 and bilateral vestibular
schwannomas (VS). Since William Hitselberger and
William House first performed the ABI in 1979, more
than 1,250 patients have been implanted worldwide with
310 implants occurring at our institution. Early results
suggested a role of tinnitus suppression after ABI implan-
tation, where tinnitus reduction occurred in six of seven
patients who reported daily use of the device; however,
long-term outcomes using validated measures of tinnitus
handicap have not been reported to date (4).
The hypothesis of this study is that ABIs reduce
tinnitus levels in NF2 patients. We investigated if ABIs
influence tinnitus levels in NF2 patients who have
received an ABI after VS resection.

118

Copyright 2016 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited.
 TINNITUS SUPPRESSION AFTER AUDITORY BRAINSTEM IMPLANTATION 119
TABLE 1. Patient characteristics METHODS
(n 43)
A retrospective chart review and patient survey were
Age (yrs, mean [SD]) 46.7 (15.6) performed of NF2 patients undergoing translabyrinthine
Age at ABI placement (yrs, mean [SD]) 36.1 (14.7) removal of VS and placement of an ABI with the goal of
Follow-up to ABI implantation (mo, mean [SD]) 131.2 (64.4) hearing restoration between 1994 and 2015. Institutional review
Follow-up to audiological testing (mo, mean [SD]) 43.3 (39.2) board approval was obtained (St. Vincent Medical Center,
Sex (% male) 37.2% Los Angeles, CA, 15-012 J. House).
NF2 100% The multichannel ABI in use at our center is the 21-electrode
Bilateral ABIs (%) 5 (11.6) system (Cochlear Corp, Englewood, CO) that has been
Bilateral acoustic neuroma surgery (%) 86.8%a approved by the US Food and Drug Administration. Trans-
ABI usage (hours per day)a labyrinthine resection of VS and placement of ABI are
0 hours 5 (13.1) described elsewhere (5).
02 hours 0 (0) Questionnaires were mailed to patients to assess their levels
24 hours 1 (2.6) of tinnitus. The questionnaire contained a set of questions
48 hours 1 (2.6)
regarding patient demographics, tinnitus symptomatology,
and ABI utilization (see results Table 1). A tinnitus handicap
812 hours 8 (21.1)
inventory (THI) was also included to quantify the overall levels
>12 hours 23 (60.5)
of tinnitus (6,7). The THI is a well-validated and widely used
a
Thirty eight patients reporting. self-report measure with 25 questions. Each answer is given a
ABI indicates auditory brainstem implant; SD, standard deviation. score of 0 (never), 2 (sometimes), or 4 (always), for a total
score of 100. THI scores may be graded as slight (016), mild

Turn on ABI. Wait for 1 hour aer Turn ABI O. Wait for
Awake from sleep in your ABI is on. 1 hour.
morning, rank severity
  
Immediately rank
of nnitus. severity of nnitus. Rank the severity of Rank the severity of
your nnitus. your nnitus.
ABI OFF ABI ON ABI ON ABI OFF

10

9 p=0.048

8
p<0.01 p<0.01
7
p<0.01 p<0.01
6

V 5
A
S All paents
4
Tinnitus
3 present,
perceived
2 benet from
ABI
1

0
C AM (Device O) Device On (Immediate) Device On (1 hr) Device O (1 hr)

FIG. 1. Modulation of tinnitus levels with an ABI. A, Patients were asked to rank tinnitus levels in four conditions with the ABI on or off
according to the diagram (B), visual analogue scale (VAS) used to score tinnitus levels (C), quantification of tinnitus levels. Data are mean
VAS scores  SEM. Dark bars are all patients responding (n 33). Light bars are for patients who thought the ABI reduced tinnitus loudness
subjectively (n 16). ABI indicates auditory brainstem implant; SEM, standard error of mean.

Otology & Neurotology, Vol. 38, No. 1, 2017

Copyright 2016 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited.
120 D. S. ROBERTS ET AL.

(1836), moderate (3856), severe (5876), and catastrophic TABLE 2. ABI utilization and characteristics of tinnitus
(78100) (6,7). Thus, the higher the THI score, the greater the modulation
handicap. Finally, levels of tinnitus, while using (and not using)
Yes No
the ABI were then evaluated using the tinnitus visual analogue
scale (VAS) at four sequential determinations. The VAS is a Does ABI provide hearing benefit? 35 (81.4) 8 (18.6)
10 cm visual scale (010) where 0 describes no tinnitus, Usable hearing without ABI (%)? 5 (11.9) 37 (88.1)
10 describes the worst possible tinnitus (Fig. 1A) that is easily Does ABI change tinnitus loudness? 20 (55.6) 16 (44.4)
applied to different experimental conditions and correlated with Does ABI reduce tinnitus loudness? 20 (54.1) 17 (45.9)
THI (8). The four sequential scenarios are described (Fig. 1B): Does ABI increase tinnitus loudness? 2 (5.4) 35 (94.6)
Does ABI change quality of tinnitus? 18 (47.4) 20 (52.6)
(1) After you awake from sleep in the morning, immedi- (Example changes the frequency)
ately rank your tinnitus on the following scale with your If your tinnitus is reduced by your ABI, 14 (60.9) 9 (39.1)
ABI off (0 describes no tinnitus, 10 describes the worst does the device need to be on to have
possible tinnitus). a reduction?
(2) Turn on your ABI. Then, immediately rank the severity How long is your tinnitus reduced after you turn your ABI off?
of your tinnitus. Circle a number. (0 describes no NA 16 (66.7)
tinnitus, 10 describes the worst possible tinnitus). 0 hour 0 (0)
(3) Wait for 1 hour after your ABI is on. Then, rank the <1 hour 2 (8.3)
severity of your tinnitus. Circle a number. (0 describes 12 hours 3 (0.13)
no tinnitus, 10 describes the worst possible tinnitus). 24 hours 1 (4.2)
(4) Turn off your ABI. Wait for 1 hour after your ABI is off. 412 hours 2 (8.3)
Then, rank the severity of your tinnitus. Circle a >12 hours 0 (0)
number. (0 describes no tinnitus, 10 describes the worst
possible tinnitus). (%) indicates percent; ABI, auditory brainstem implant; SD,
standard deviation.
Patients were asked to gauge the level of background noise
during these tasks. The listening environment (background
noise) was designated as low, moderate, or high. ear in 10.2% of cases. THI score for all participants was
17.8  20.5 SD. Mean VAS score with the device off was
Audiometric Analysis 3.7  2.5 SD.
Speech perception performance was determined for each A series of survey questions were used to investigate
patient using the Northwestern University-Childrens Percep- how the ABI is used among responders and if tinnitus
tion of Speech (NU-CHIPS) word test, a 4-choice monosyllabic levels were modulated by the device (Table 2). 81.4% of
word test given in the sound-only mode (9). The best post- patients described hearing benefits from the ABI and
implantation performance was recorded and compared with 88.1% had no useable hearing without the device. Tin-
tinnitus levels using these above described measures. nitus loudness was modulated in 55.6% and a reduction
of tinnitus loudness was reported in 54.1%. Notably,
Data Analysis tinnitus loudness was subjectively increased in 5.4%
Data were extracted directly from patient surveys and by the ABI. For patients who had a reduction in tinnitus
medical records. Survey data were reported for all patients
levels, the majority (60.9%) thought that the device
despite incomplete survey responses for some patients.
Microsoft Excel (Redmond, WA), and SPSS (Armonk, NY) needed to be active for tinnitus suppression. Eight of
were used for analysis. Both parametric and nonparametric forty-three (18.6%) patients describe some subjective
statistics were used, including correlation and analysis of persistent degree of tinnitus suppression after the device
variance for continuous/interval level data and Games-Howell was turned off with suppression lasting less than l to
post hoc test. Criterion for statistical significance was set 12 hours.
at p < 0.05. To experimentally investigate levels of tinnitus while
using and not using the ABI, the tinnitus VAS was used in
RESULTS four conditions described in Fig. 1A. For all patients
completing this component of our survey (n 32), tin-
From 1994 to 2015, 279 patients underwent ABI nitus was reduced when the device was turned on for
placement at our institution. All patients implanted with 1 hour (mean VAS: Off 3.5; On 1-h 2.1; p 0.048)
the multichannel 21-electrode system (Cochlear Corp, (Fig. 1C). For patients who thought the ABI reduced
Englewood, CO) with known contact information who tinnitus loudness subjectively, the ABI reduced tinnitus
resided in the United States and who were adults were scores immediately on activation and after 1 hour of
contacted (n 133). One hundred-twelve patients use (mean VAS: Off 4.8; On 2.4; On 1-h 1.8;
received our patient survey with 43 patients (38.3%) Off 1-h 4.8 p < 0.01) (Fig. 1C). Suppression did not
responding. Patient characteristics are shown in Table 1. persist after the device was off for 1 hour. During these
Tinnitus was present in 83.7% of patients and the tasks, the listening environment (background noise) was
duration of tinnitus was substantial (18.8  11.7 SD self-reported in 20 of 43 patients (46.5%) and designated
years). More commonly, tinnitus was perceived in both as low, moderate, or high. Eight of 20 patients described
ears (74.4%). Patients were unable to localize tinnitus in a low level of background noise while 12 of 20 reported
15.4% of cases. Tinnitus was isolated to the right or left a moderate level.

Otology & Neurotology, Vol. 38, No. 1, 2017

Copyright 2016 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited.
 TINNITUS SUPPRESSION AFTER AUDITORY BRAINSTEM IMPLANTATION
Lastly, we investigated whether audiological per-
formance with the ABI and device utilization related
to levels of tinnitus suppression. Best audiological
performance using the NU-CHIPS test had no corre-
lation to THI scores (R2 0.006). When only patients
who used the ABI for some period of the day were
considered, the length of time since implantation had
no influence on THI scores (R2 0.03). THI scores
were then stratified by daily ABI usage. Higher daily
ABI usage was not associated with differences in THI
scores.
DISCUSSION
Our data represent a single institution series character-
izing levels of tinnitus in NF2 patients with ABIs.
Importantly, we illustrate that tinnitus levels may be
suppressed by utilization of an ABI in NF2 patients
who have undergone translabyrinthine craniotomy for
VS removal.
Our data are consistent with two other studies that have
noted some degree of tinnitus suppression among NF2
ABI recipients. In a study of 10 NF2 patients who
received an ABI at the time of translabyrinthine surgery
for removal of an acoustic neuroma from our institution,
six of seven who reported daily use of their ABIs noted
some suppression of their tinnitus (4). In another study of
20 patients with NF2 receiving the ABI through a retro-
sigmoid approach, the majority achieved some tinnitus
suppression (10). However, meaningful conclusions
could not be deduced from these studies because vali-
dated measures of tinnitus handicap where not used in the
assessment of outcomes.
Findings in the current study suggest tinnitus was
reduced when the device was turned on for 1 hour.
For patients who thought the ABI reduced tinnitus
loudness subjectively, the ABI reduced tinnitus VAS
scores immediately on activation and after 1 hour of use.
Suppression did not persist after the device was off for
1 hour. Together, these data illustrate that suppression
occurs rapidly after the device is activated and does not
persist for greater than 1 hour after the device is
turned off.
A logical question is whether these changes represent a
meaningful and clinically significant level of tinnitus
suppression. Amoodi et al. (11) showed that tinnitus
suppression mediated by CI leads to a reduction of
THI by 16 points in one series (preoperative THI
36.18; 12 months after CI, 20.18). Tinnitus reduction
did not correlate with audiological performance, how-
ever, tinnitus reduction did correlate with improved
quality-of-life (QOL) domains. These data suggest that
a 16-point change in THI is clinically significant and
may impact QOL. We report reduced VAS scores by
1.4 (3.52.1) for all patients completing the survey. For
patients with tinnitus and a subjective improvement when
using the ABI, VAS scores were reduced three points
(4.81.8). A three-point change in VAS is estimated to
correlate with approximately an 18-point change in THI
Copyright 2016 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited.
121
(8), suggesting clinically significant findings in some
patients relating to improved QOL through tinnitus
suppression. However, QOL was not directly tested in
our study and further studies are needed to make
these conclusions.
The mechanism by which the ABI may modulate
tinnitus can only be speculated. A fundamental question
is whether tinnitus suppression occurs primarily due to
masking, or that tinnitus suppression occurs due to
electrical stimulation of the auditory brainstem which
may alter neurological activity associated with peripheral
deafferentation. Evidence supporting a masking mech-
anism includes early experience with CI and tinnitus
suppression through transcutaneous electrical stimu-
lation. Early CI recipients who responded to CI for
tinnitus suppression also benefited from a hearing aid
for tinnitus masking (12). Shulman noted that the
majority of patients who achieved tinnitus suppression
with transcutaneous electrical stimulation have pre-
viously benefited from tinnitus masking (13). A masking
mechanism may account for our findings because the
onset of tinnitus suppression occurred immediately with
device activation and suppression did not persist after the
device is inactivated for 1 hour.
A second possibility is that ABIs exert a direct inhibi-
tory effect in the auditory system at the level of the brain
or brainstem, possibly at the cochlear nucleus, auditory
cortex, or inferior colliculus. A plausible electrophysio-
logical mechanism suggests that chronic electrical stimu-
lation of the deafferented auditory nerve can reverse the
downregulation of glycinergic receptor (GlyRalpha1)
and glutamic acid decarboxylase (GAD) 67 expression
in the central nucleus of inferior colliculus (CIC) (14).
These changes or other molecular changes associated
with neural plasticity after electrical stimulation could
mediate tinnitus suppression. A second report suggests
that subjective tinnitus may be controlled by intraco-
chlear electrical stimulation, generated by a CI in a
participant with single-sided deafness through stimu-
lation that is independent of environmental sounds
(15). Our survey data show that 39.1% of patients report
tinnitus reduction that continues when the ABI is off
suggesting residual inhibition of tinnitus. A plausible
mechanism includes an alteration of neural plasticity
that persists even when the device is off. However,
our experimental data suggest that tinnitus suppression
requires the device to be on. Further efforts will focus on
the question of whether tinnitus suppression after ABI
occurs through masking or by altering neural plasticity.
Our data report the presence of tinnitus in 83.7% of
patients, a prevalence that is consistent with other data in
NF2 patients after translabyrinthine removal (16). To our
knowledge, our data expand upon our understanding of
the severity of tinnitus in this population by establishing
levels of tinnitus handicap through the THI and VAS
scores. The reported level of THI 17.8  20.5 SD in our
NF2 patients is similar to THI 1-year after resection of a
unilateral sporadic VS (13.5  19.5 SD) (17). Tinnitus
levels are lower when compared with CI patients before
Otology & Neurotology, Vol. 38, No. 1, 2017
122 D. S. ROBERTS ET AL.
and after implantation in one series (preoperative THI
36.18; 12 months after CI 20.18) (11). Overall, the
established THI for our patients is categorized as mild
tinnitus (18). Mild tinnitus is typically masked with
environmental sounds and may interfere with sleep but
typically is not intrusive to daily activities (18). However,
in patients with NF2 with no usable hearing, who
accounted for 88.1% of our patients, conventional mask-
ing is not an option illustrating the importance of ABIs
for tinnitus reduction for some NF2 patients.
As with all retrospective and survey studies, there are
several caveats with data interpretation. There is the risk
of bias from this type of evaluation because we are
depending on voluntary survey responses that may under
or over-report the incidence and degree of tinnitus
severity. Our survey response rate was 38% and partial
responses to our survey were present in some cases
suggesting possible reporting bias. Additionally, our
study design asked patients to rank tinnitus levels while
the ABI was on and off, however, we were not able to
control for the levels of background noise during these
tasks. The degree of tinnitus suppression may indeed vary
depending on the level of background noise. 88.1% of our
cohort has no useable hearing without the ABI device
with variable ABI performance. It may not be possible
for our patients to gauge background noise during these
tasks. A prospective study at our institution is underway
to further characterize our patient population controlling
for these variables.
CONCLUSIONS
NF2 patients who have undergone removal of VS have
a notable tinnitus handicap. This study shows that tinnitus
suppression occurs in NF2 patients implanted with an
ABI. In subgroup analyses of patients who reported
tinnitus suppression, tinnitus levels were immediately
reduced on ABI activation and suppression persisted
during use. Suppression did not continue after the device
was turned off for 1 hour. Audiological performance,
ABI usage, and length of time since implantation did
not correlate with tinnitus suppression. These findings
show that the auditory brainstem is a target for tinnitus
suppression for NF2 patients and suggest that the coch-
lear nucleus may be a putative target for tinnitus sup-
pression in patients with idiopathic tinnitus.
Acknowledgments: The authors are grateful for our patients
who enthusiastically participated in this study.
Otology & Neurotology, Vol. 38, No. 1, 2017
Copyright 2016 Otology & Neurotology, Inc. Unauthorized reproduction of this article is prohibited.
REFERENCES
1. Arts RA, George EL, Stokroos RJ, Vermeire K. Review: coch-
lear implants as a treatment of tinnitus in single-sided
deafness. Curr Opin Otolaryngol Head Neck Surg 2012;20:
398403.
2. Roberts DS, House JW. Tinnitus suppression and cochlear implan-
tation.Tinnitus: Diagnosis and Treatment, 2nd ed. New York:
Shulman; 2016.
3. Schecklmann M, Landgrebe M, Poeppl TB, et al. Neural correlates
of tinnitus duration and distress: a positron emission tomography
study. Hum Brain Map 2013;34:23340.
4. Soussi T, Otto SR. Effects of electrical brainstem stimulation on
tinnitus. Acta Otolaryngol 1994;114:13540.
5. Schwartz MS, Otto SR, Shannon RV, Hitselberger WE, Brackmann
DE. Auditory brainstem implants. Neurotherapeutics 2008;5:
12836.
6. Newman CW, Jacobson GP, Spitzer JB. Development of the
tinnitus handicap inventory. Arch Otolaryngol Head Neck Surg
1996;122:1438.
7. Newman CW, Sandridge SA, Jacobson GP. Psychometric adequacy
of the tinnitus handicap inventory (THI) for evaluating treatment
outcome. J Am Acad Audiol 1998;9:15360.
8. Figueiredo RR, Azevedo AA, Oliveira Pde M. Correlation
analysis of the visual-analogue scale and the tinnitus handicap
inventory in tinnitus patients. Braz J Otorhinolaryngol 2009;75:
7679.
9. Elliot LL, Katz DR. The Northwestern University Childrens Per-
ception of Speech Test: NU-CHIPS. St Louis: Auditec; 1980.
10. Behr R, Muller J, Shehata-Dieler W, et al. The high rate CIS
auditory brainstem implant for restoration of hearing in NF-2
patients. Skull Base 2007;17:91107.
11. Amoodi HA, Mick PT, Shipp DB, et al. The effects of unilateral
cochlear implantation on the tinnitus handicap inventory and the
influence on quality of life. Laryngoscope 2011;121:153640.
12. Berliner KI, Cunningham JK. Tinnitus suppression in cochlear
implantation. In: Hazell J, editor. Tinnitus. London: Churchill
Livingstone; 1987. pp. 11830.
13. Shulman A. External electrical stimulation in tinnitus control. Am J
Otol 1985;6:1105.
14. Argence M, Vassias I, Kerhuel L, Vidal PP, de Waele C. Stimu-
lation by cochlear implant in unilaterally deaf rats reverses the
decrease of inhibitory transmission in the inferior colliculus. Eur J
Neurosci 2008;28:1589602.
15. Zeng FG, Tang Q, Dimitrijevic A, Starr A, Larky J, Blevins NH.
Tinnitus suppression by low-rate electric stimulation and its electro-
physiological mechanisms. Hear Res 2011;277:616.
16. Moffat DA, Lloyd SK, Macfarlane R, et al. Outcome of trans-
labyrinthine surgery for vestibular schwannoma in neurofibroma-
tosis type 2. Br J Neurosurg 2013;27:44653.
17. Baguley DM, Humphriss RL, Axon PR, Moffat DA. Change in
tinnitus handicap after translabyrinthine vestibular schwannoma
excision. Otol Neurotol 2005;26:10613.
18. McCombe A, Baguley D, Coles R, McKenna L, McKinney C,
Windle-Taylor P, British Association of Otolaryngologists, Head
and Neck Surgeons. Guidelines for the grading of tinnitus severity:
the results of a working group commissioned by the British Associ-
ation of Otolaryngologists, Head and Neck Surgeons. 1999. Clin
Otolaryngol Allied Sci 2001;26:38893.