J Am Acad Audiol 11 : 162-177 (2000)

Tinnitus Retraining Therapy (TRT)

as a Method for Treatment of
Tinnitus and Hyperacusis Patients
Pawel J. Jastreboff*
Margaret M. Jastreboff*

Abstract
The aim of this paper is to provide information about the neurophysiologic model of tinnitus
and Tinnitus Retraining Therapy (TRT). With this overview of the model and therapy, profes-
sionals may discern with this basic foundation of knowledge whether they wish to pursue
learning and subsequently implement TRT in their practice . This paper provides an overview
only and is insufficient for the implementation of TRT.

Key Words : Habituation, hyperacusis, neurophysiologic model, tinnitus, Tinnitus Retraining

Therapy, treatment

Abbreviations : DPOAE = distortion product otoacoustic emission, IHC = inner hair cells,
LDL = loudness discomfort level, OHC = outer hair cells, THT = Tinnitus Habituation Therapy,
TRT = Tinnitus Retraining Therapy

innitus, commonly referred to as "ring- with hearing loss, otosclerosis, ear infections,
ing in the ears," is a common problem acoustic neuroma, Meniere's syndrome, and
T affecting many people around the world. aging. Also, more than 200 prescription and
According to studies performed in various coun- nonprescription drugs list tinnitus as a poten-
tries, tinnitus affects 10 to 20 percent of the tial side effect .
general population (McFadden, 1982 ; Coles, Much less is known about the prevalence of
1987 ; Drukier, 1989); in the United States, this increased sensitivity to sound, hyperacusis. Cur-
translates into 25 to 52 million Americans . This rently, in the literature there are limited data
affliction is even more prevalent in the elderly published by Vernon (Vernon, 1987 ; Vernon and
over the age of 65 years, with approximately 30 Press, 1998), Coles (Coles and Sood, 1988),
percent reporting tinnitus (Sataloff et al, 1987 ; Hazell and Sheldrake (Hazell and Sheldrake,
Salomon, 1989). For about 5 percent of the gen- 1992), and Jastreboff' (Jastreboff et al, 1996b,
eral population (about 13 million Americans), 1998). Data presented by Vernon are incongru-
prolonged tinnitus is moderately or significantly ent with the remaining reports, stating that
annoying, causing them to seek help (McFadden, only 0.3 percent of tinnitus patients have hyper-
1982). Consequently, this population is labeled acusis, whereas other data indicate that about
as having clinically significant tinnitus . Finally, 40 percent of tinnitus patients have some degree
1 out of 100 adults reports tinnitus as a debili- of hyperacusis . Furthermore, our data indicate
tating problem (Coles, 1996) (about 2.6 million that about 25 percent of tinnitus patients are
Americans) . Typically, tinnitus is associated bothered more by their hyperacusis than their
tinnitus, and thus require specific treatment
for hyperacusis.
We are not aware of epidemiologic data
*Tinnitus and Hyperacusis Center, Department of related to the prevalence of hyperacusis in the
Otolaryngology, Emory University School of Medicine, general population . This is an unfortunate sit-
Atlanta, Georgia uation, since these data could help in estimat-
Reprint requests : Pawel J . Jastreboff, Tinnitus and
Hyperacusis Center, Department of Otolaryngology, Emory ing the need for health services and for planning
University School of Medicine, 1365-A Clifton Road, NE, cost-effective, yet quality, services . Assuming,
Atlanta, GA 30322 however, that in our practice we are working

162
 Tinnitus Retraining Therapy/Jastreboff and Jastreboff

with patients with clinically significant tinnitus because it is dependent on the equipment used
(5% of the general population), and 25 percent and the skill of the observer, not the patho-
of those have significant hyperacusis, then about physiology of the sound.
1 .25 percent of the general population (3 .25 mil- Hyperacusis is defined as abnormally strong
lion Americans) has significant hyperacusis . reactions occurring within the auditory pathways
This is a rather conservative estimate, as there resulting from exposure to moderate sound; as
are cases of hyperacusis without tinnitus. a consequence, patients express reduced toler-
In spite of a long recorded history of tinni- ance to suprathreshold sounds . This phenome-
tus, reaching as far back as the ancient Baby- non may be, but typically is not, related to
lonian and Egyptian civilizations (Feldmann, recruitment (Moore, 1995 ; Jastreboff, 1998 ; Jas-
1988), and its high prevalence today, there is no treboff et al, 1998).
cure for tinnitus . It appears that all approaches Phonophobia is defined as abnormally strong
used in the past failed to provide systematic reactions of the autonomic and limbic systems
relief to tinnitus patients . Practically all of the (without abnormally high activation of the audi-
treatments previously used were effective only tory system by sound), resulting from enhanced
on a subpopulation of patients, had to be con- connections between the auditory and limbic
tinued through the patient's life, and were fre- systems . This can be described at the behavioral
quently accompanied by significant side effects . level as "patients being afraid of sound ."
Furthermore, the very existence of a long list of Increased sound sensitivity is abnormally
treatments that may potentially provide help, high sensitivity to a sound resulting from the
and the fact that the single most common sum effects of hyperacusis and phonophobia.
approach is telling patients "to learn to live with The seemingly simple definition of tinnitus
it," argues strongly against their effectiveness . has profound implications on proposing the
In this paper, we propose that Tinnitus Retrain- mechanisms of tinnitus and consequently on its
ing Therapy (TRT), when implemented prop- treatment. This definition stresses the involve-
erly, (1) is highly effective, (2) does not have ment of the nervous system as a key compo-
side effects, (3) needs to be implemented over a nent responsible for the emergence of tinnitus
finite amount of time, and (4) can be used on all and problems arising from its presence, thus
patients . moving its mechanisms away from the cochlea
to the central nervous system . The definition
Definitions further indicates the existence of a link between
the mechanisms of tinnitus and that of the phan-
As there are various definitions of tinnitus, tom limb and phantom pain phenomena, which
hyperacusis, and phonophobia, we are present- indeed appear to exist. Certain common aspects
ing the definitions, as proposed by us, to ensure of tinnitus and phantom pain are used for the
a clearer understanding. classification of tinnitus and hyperacusis patients
nnnitus is commonly defined as a noise in and their treatment.
the ears or head, frequently described as ring-
ing, buzzing, humming, hissing, the sound of
OUTLINE OF THE
escaping steam, etc . In 1982, the Committee on
NEUROPHYSIOLOGIC MODEL
Hearing, Bioacoustics and Biomechanics pro-
posed a definition of tinnitus as "the conscious
experience of a sound that originates in the
head of its owner" (McFadden, 1982) . The defi-
nition of tinnitus we are promoting is "the per-
P roposed in the 1980s (Jastreboff, 1990),
development of the neurophysiologic model
of tinnitus was initiated by several observa-
ception of a sound which results exclusively tions. First, epidemiologic studies revealed that
from the activity within the nervous system tinnitus induces distress in only about 25 per-
without any corresponding mechanical, vibratory cent of the tinnitus population (McFadden, 1982),
activity within the cochlea" (Jastreboff, 1995), and there is no correlation of the distress with
that is, tinnitus as an auditory phantom per- psychoacoustic characterization of tinnitus, that
ception (Jastreboff, 1990, 1995) . is, average loudness of tinnitus, its pitch, and
Somatosounds are sounds generated by maskability are similar in people who are only
structures in and adjacent to the ear, including experiencing tinnitus to those who suffer because
spontaneous otoacoustic emission . The term of it (Jastreboff, 1995). Second, the psychoa-
"objective tinnitus" has been used to describe coustic characterization of tinnitus in the patient
somatosounds . This classification is inaccurate population is not related to the severity of

163
Journal of the American Academy of Audiology/Volume 11, Number 3, March 2000

tinnitus (i.e ., two people with a similar charac- autonomic function and controls emotional
terization of tinnitus often differ dramatically in expression, seizure activity, memory storage
the level of distress created by their tinnitus). and recall, and the motivational and mood states
The same observation applies to the treatment (Swanson, 1987). The limbic system plays a role
outcome, which is not correlated to the loud- in all aspects of life, which involve motivation,
ness, pitch, or maskability of tinnitus (Jastreboff mood, and emotions . Furthermore, it activates
et al, 1994). the autonomic nervous system.
The above facts argued very strongly for The autonomic nervous system, one of the
the auditory system as secondary only and other two main divisions of the nervous system, con-
systems in the brain being dominant in clinically trols the action of glands, as well as the functions
relevant tinnitus . Moreover, Heller and Bergman of the respiratory, circulatory, digestive, and
(1953) showed that the perception of tinnitus urogenital systems . The system also has some
cannot be pathologic, since essentially every- control over the production of hormones. The
one (94% of people without tinnitus experience autonomic nervous system consists of two dis-
tinnitus when isolated for several minutes in a tinct, mutually antagonistic components, the
anechoic chamber) experiences it when put in a sympathetic and parasympathetic . The sympa-
sufficiently quiet environment . thetic division stimulates the heart, dilates the
Consequently, the neurophysiologic model of bronchi, contracts the arteries, inhibits the diges-
tinnitus postulates that both abnormalities in the tive system, and prepares the organism for phys-
cochlear function and the processing of a tinni- ical action. The parasympathetic division has the
tus-related signal within the nervous system opposite effect and prepares the organism for
must be included in the analysis of tinnitus feeding, digestion, and rest (Brooks, 1987) .
phenomenon . Specifically, in the emergence of Because of the exacerbated activation of the
clinically relevant tinnitus, it is possible to dis- autonomic nervous system, many body func-
tinguish the following stages : (1) the genera- tions are affected . Typically, patients exhibit
tion of tinnitus-related neuronal activity initiated syndromes that indicate the sympathetic divi-
in the periphery of the auditory system (the sion of the autonomic nervous system as domi-
cochlea, auditory nerve), (2) the detection of this nant, which stimulates the heart, inhibits the
signal occurring in subcortical auditory centers, digestive system, and, in general, prepares the
(3) the perception and evaluation of the signal body for physical action . The overactivation of
at cortical areas (auditory and others), and the sympathetic nervous system leads to prob-
(4) the sustained activation of the limbic (emo- lems with sleep, a very common situation among
tional) and autonomic nervous systems. If neg- tinnitus patients (Coles, 1996).
ative associations are not attached to the person's Abnormally high activation of the limbic
tinnitus, then only the first three stages occur; and autonomic nervous systems results in stress,
therefore, the person only experiences tinnitus anxiety, and loss of well-being . The patients are
without being annoyed by its presence . The getting extremely annoyed by their tinnitus .
fourth stage is crucial for creating distress and, Feedback loops connecting the auditory, limbic,
consequently, clinically relevant tinnitus . Tin- and autonomic nervous systems (Fig. 1A) are get-
nitus-induced activation of the limbic and auto- ting stronger, and patients continue to get worse.
nomic nervous systems is responsible for the The feedback loops develop in the following
distress caused by tinnitus . manner. The continuous, uncontrollable presence
The involvement of these two systems was of tinnitus, or belief that something bad is going
indicated by the problems reported by tinnitus to happen, causes the tinnitus-related neuronal
patients . Patients with clinically significant tin- activity to become linked to a negative reaction
nitus exhibit a strong emotional reaction to its in the brain (i .e ., annoyance, anxiety, general
presence, a high level of anxiety, and a number stress). Due to this negative reaction, more
of psychosomatic problems (Jastreboff, 1990 ; attention is automatically directed toward tin-
Kuk et al, 1990 ; Stouffer and Tyler, 1990 ; Stouf- nitus, which, in turn, enhances the detection of
fer et al, 1991 ; Newman et al, 1995). In general, the tinnitus signal by subcortical auditory cen-
these reactions depend on the activation of the ters . This enhanced detection of the tinnitus
limbic and autonomic nervous systems . The lim- signal increases activation of the limbic and
bic system consists of an array of brain structures autonomic nervous systems, which, in turn,
including the hippocampal formation, amyg- increases the attention devoted toward tinnitus,
dala, septum, and hypothalamus . This system etc. Consequently, the sustained activation of
has a direct influence on neuroendocrine and limbic and autonomic nervous systems occurs .

164
 Tinnitus Retraining Therapy/Jastreboff and Jastreboff

to enjoy life . Frequently, in cases of severe tin-

Perception & Evaluation nitus, patients no longer enjoy activities previ-
Auditory & other Cortical Areas
T ously pleasant to them, which, in turn, may
yield depression .
From a clinical point of view, it is interest-
Detection ing to recognize that tinnitus can acquire neg-
Subcortical ative associations through (1) a prolonged,
continuous presence of a neutral stimulus (e .g.,
a neighbor's son playing the same song over
and over again); (2) fear of a new, unknown dan-
Source
Cochlea ger; and (3) "negative counseling" (e .g ., "nothing
A can be done ; you will have to learn to live with

c
it ; let's do an MRI to exclude a brain tumor") .
Perception & Evaluation Unfortunately, negative counseling is very com-
mon and triggers the development of a vicious
Auditory & other Cortical Areas

cycle, and patients devote increasingly more

time to monitoring their tinnitus and experience
l
problems with attention, work, sleep, etc.
Detection Emotional Associations
Subcortical 4;- Limbic System Once established, the reactions of the lim-
bic and autonomic nervous systems are induced,
along the principle of conditioned reflex . In
Ell the diagram presented in Figure 1, two loops
Source
Cochlea are depicted : (1) upper, cortical-verbal (Fig.
B ... . ... . . . ... . .... . .... . . ... . .... .... . . .... . . 1B), which may involve beliefs and can be
directly affected by counseling, and (2) lower,
Perception & Evaluation
Auditory & other Cortical Areas
subcortical-nonverbal (Fig . 1C), which can be
controlled only indirectly. Cognitive therapies
acting on high cortical levels affect the upper
loop . Since there is no consistently successful
outcome of cognitive therapies in a majority of
patients, this argues that the lower, subcon-
scious loop plays the dominant role .
While we postulate that the auditory system,
including the cochlea, plays a secondary rather
Cochlea
than primary role in the emergence of clinically
C . . . .... .. . . ...... . . . ... . . . ...... . . ...... . . . ..
significant tinnitus, it does, in fact, provide the
Figure 1 Block diagram outlining systems involved initial signal that starts the cascade of events
in clinically relevant tinnitus and changes occurring as resulting in the development of clinically sig-
a result of tinnitus habituation. Thickness of the arrows nificant tinnitus . The specific mechanisms trig-
indicates the significance of a given connection . A, main
gering the source of tinnitus can vary from
diagram, B, upper loop, C, lower loop .
patient to patient .
Neurophysiologic experiments provide
Certain aspects of this activation are impor- insight to potential mechanisms, yielding the
tant . First, the activation level depends on the perception of tinnitus . The emergence of tinni-
strength of the negative associations and not on tus in subjects placed in a very quiet environ-
the perceptual description of tinnitus . ment may be explained by the following
The increase of annoyance and anxiety is a observation: there is a high level of spontaneous
self-propelling process, with the maximal level activity within the auditory pathways that is ran-
depending upon the association of tinnitus with dom and is not perceived under normal condi-
something negative, as well as on the psycho- tions. This activity could be labeled as a code for
logical profile of the patient, but is not related silence and, under normal conditions, is filtered
to psychoacoustic characterization of the per- out by subcortical centers and does not result in
ceived sound of tinnitus . Second, a high level of the perception of sound. However, when the
activation of the sympathetic part of the auto- level of cochlear stimulation is decreased, it has
nomic nervous system induces the fight or flight been shown that the sensitivity of the auditory
reaction and suppresses the ability of the patient pathways increases with about 25 percent of

165
 Journal of the American Academy of Audiology/Volume 11, Number 3, March 2000

the neurons exhibiting abnormally low thresh- otoacoustic emission (DPOAE) suggest that, in
old to any stimulation (Boettcher and Salvi, some patients, dysfunction in the OHC system
1993 ; Salvi et al, 1996). Under a condition of may be responsible for their hyperacusis.
increased neuronal sensitivity, the auditory sys- The central type of mechanism involves
tem starts to detect the inevitable fluctuation of oversensitivity of neurons in the auditory path-
randomness of the spontaneous activity, which ways . About 25 percent of cells in the cochlear
is perceived as tinnitus . nuclei complex and the inferior colliculi exhibit
The concept of tinnitus-related neuronal abnormally high sensitivity and abnormally
activity, resulting from the compensation per- strong evoked potentials if the auditory inputs
formed by the auditory system to even mild dys- is decreased (Boettcher and Salvi, 1993 ; Gerken,
function within the cochlea or the auditory nerve, 1993). An abnormally high central gain might
has been further elaborated in the postulate of result in the detection of fluctuations in the
discordant damage/dysfunction theory (Jastre- spontaneous activity or a weak abnormal pattern
boff, 1990, 1995), which is an extension of the the- of activity that would otherwise not be detected
ory proposed by Tonndorf (Tonndorf, 1987) . as tinnitus . Thus, for some patients, tinnitus and
Typically, damage of the cochlea affects more hyperacusis can be two manifestations of the
outer (OHC) than inner (IHC) hair cells, result- same neuronal mechanism.
ing in unbalanced activity, which reaches the dor- In the case of pure hyperacusis, the abnor-
sal cochlear nucleus through Type I and Type II mal gain is constrained to the auditory system
auditory nerve fibers . If the imbalance is large and moderate sounds induce a high level of
enough, it evokes a compensation within the activity within the auditory pathways, and only
auditory pathways to such an extent that secondary in the limbic and autonomic nervous
tinnitus-related neuronal activity is generated systems, involved in preparing the subject to
as a side effect of this compensation . Potential withdraw from the unpleasant sound level. Since
mechanisms are speculative but may involve in pure hyperacusis (i .e ., without simultaneous
lateral inhibition (Liberman and Kiang, 1978 ; phonophobia), the actual amplitude of the
Liberman and Mulroy, 1982 ; Gerken, 1992, 1993) cochlear basilar membrane vibrations and the
or disinhibition (Chen and Jastreboff, 1995). processed physical intensity of the sound are the
Indeed, abnormal, bursting, epileptic-like spon- dominant factors, the physical parameters of
taneous neuronal activity has been recorded the sound are the determining factors for the
from the inferior colliculus in animals with sal- extent of discomfort. The context in which the
icylate-induced tinnitus (Chen and Jastreboff, sound occurs (e .g ., home, doctor's office, movie
1995) and recently in animals with sound over- theater) is irrelevant, and the reaction to a given
exposure (Jastreboff et al, 1999). sound is the same under all of these various
conditions . The physical parameters of the trans-
Hyperacusis and Phonophobia duction of the cochlea and the relation of dB SPL
and HL can result in a flat curve of loudness dis-
We propose that the increased sound sen- comfort levels (LDLs), with tendency of decreased
sitivity (decreased sound tolerance) consists of values for very low and very high frequencies
two components : hyperacusis and phonophobia. (Fig. 2A).
Hyperacusis may result from both peripheral and After prolonged exposure to sound perceived
central mechanisms . In the cochlea, two types by an individual as too loud, an activation of
of dysfunction in the active amplification, these systems could occur for increasingly lower
provided by the OHC, could result in hypera- sound levels, as phonophobia, the fear of sound,
cusis. Normally, OHCs amplify weak sounds may develop . Moreover, pure phonophobia (i.e .,
(<10-20 dB SPL) by 66 to 76 dB, and the ampli- without the presence of hyperacusis) can be
fication gradually decreases to 0.2 dB/dB in the found in patients believing in the harmful effects
range of 40 to 80 dB (Ruggero et al, 1997). The of sound and their attempts to overprotect their
first type of peripheral overamplification occurs ears (ear plugs, etc .) . In pure phonophobia, the
if OHCs continue to amplify louder sounds, and auditory system works normally. However, a
then IHCs become overstimulated for moderately relatively low level of the activation within the
loud sounds levels . The second type of periph- auditory system results in an overactivation of
eral overamplification occurs when OHC the limbic and autonomic nervous systems due
mechanical amplification increases to larger to the enhanced connectivity between these sys-
values than present at a normal state . Indeed, tems . The reaction to a given sound depends
measurements of the distortion product upon the context in which the sound occurs .

166
 Tinnitus Retraining Therapy/Jastreboff and Jastreboff

0
0 0 0
0 0 0 0
25 ------- - - - -Q------- 25 ------- - - - -0-------
00 0
0
0 U U U O
00 U 0
UU 0
U
U U UUUUUU U

100 I - 100
U

Fl~:: 8k
A 0 .25 1k 8k B 0 .25 1k
kHz kHz
Figure 2 Types of Loudness Discomfort Levels (LDLs)
associated with A, pure hyperacusis, B, pure phono-
0
0-~-0
phobia, C, mixture of hyperacusis and phonophobia.
O = hearing thresholds ; U= Loudness Discomfort Levels .
25 O - O -- ------

0
0
0
0
U U UU UU U
U
U
100

C 0.25 1k 8k
kHz
Patients may over-react to certain sounds, while is a need to separate this phenomenon from the
not reacting to more intense sounds . Since the temporarily enhanced phonophobia, which is
reactions depend upon the perceived loudness of treated differently.
the sound (related to dB SL) and not its physi- In summary, it is postulated that tinnitus-
cal intensity, the LDLs will keep a relatively related neuronal activity results predominantly
constant distance from the audiogram following from the compensatory action of the auditory
in consequence the shape of audiogram (Fig . pathways to a peripheral dysfunction, perhaps
2B). In patients with exclusive or strong phono- a difference in the damage of OHCs versus IHCs.
phobia, there is a tendency of extremely low In about 75 percent of tinnitus cases, this activ-
values of LDLs . Most frequently, hyperacusis and ity is contained within the auditory system and
phonophobia coexist, with typical LDLs, as is frequently blocked before it reaches the level
shown in Figure 2C . of awareness . Consequently, these people expe-
Some patients exhibit prolonged (days of rience tinnitus but do not suffer because of it.
weeks) worsening of their tinnitus and/or hyper- However, in clinically relevant cases, as a
acusis as a result of an exposure to sound . There result of initial negative associations, tinnitus-

167
Journal of the American Academy of Audiology/Volume 11, Number 3, March 2000

related neuronal activity inappropriately acti- (2) automize as many tasks as possible into the
vates the limbic and autonomic nervous sys- subconscious, nonverbal reflexes (driving a car,
tems, resulting in the development of annoyance, eye movements when reading, walking, run-
anxiety, sleep disturbances, and a number of ning, etc.) ; (3) prioritize all remaining tasks ;
somatic problems . Connections between the and (4) perform one task at a time, starting
auditory and other systems are based upon the from the most important.
principle of conditioned reflexes and as such Note that the selection and blockage of
cannot be easily and directly changed. Fur- unimportant signals have to occur at a subcon-
thermore, the limbic and autonomic nervous scious level on the basis of past experiences .
systems may be completely normal, and the The selection process cannot be done on a con-
problem arises from their activation by inap- scious level, as it would consume our attentional
propriate stimulus (i .e ., tinnitus-related neu- abilities and nullify the purpose of selection.
ronal activity). On the basis of the neurophysiologic model,
Hyperacusis and phonophobia activate the it has been proposed to treat tinnitus by induc-
limbic and autonomic nervous systems as well, ing its habituation . To achieve habituation, a spe-
but with different mechanisms than tinnitus . cific approach based on the general principles of
Once established, reactions of these systems the brain function (e .g ., the physiologic mecha-
are controlled by the conditioned reflex princi- nisms of perception, the role of subcortical audi-
ple but triggered by external sounds rather than tory pathways, the functional properties of the
tinnitus-related neuronal activity. limbic and autonomic nervous systems and their
interaction with the auditory system, mecha-
How to Treat Tinnitus and Hyperacusis nisms of conditioned reflexes, and the physiologic
mechanisms of the brain plasticity) was sug-
The neurophysiologic model offers an gested (Jastreboff, 1990, 1995 ; Jastreboff et al,
approach to treat both tinnitus and hyperacu- 1996a) . Use of habituation to help tinnitus
sis. For tinnitus, the approach is based on obser- patients was first proposed by Hallam et al
vations that, although there is no reliable method (1984), but their model was psychological rather
for attenuating the tinnitus source (cure), the than physiologic . Consequently, their approach
brain exhibits a high level of plasticity. It is pos- to induce and sustain habituation was different,
sible to habituate to any sensory signal, as long focused on reassurance, relaxation, and atten-
as the signal is not associated with any negative tion distraction (Jakes et al, 1986), and did not
implications . Thus, the conclusion is to induce appear to produce significant, sustained improve-
habituation of tinnitus (by interfering with ment in tinnitus patients .
tinnitus-related neuronal activity above the tin- Habituation of tinnitus has two main goals.
nitus source). The primary clinical goal is to habituate reac-
Habituation is a normal, common, and nec- tions of the limbic and autonomic nervous sys-
essary function of the brain, as pointed out by tems (Fig . 3A). Tinnitus-related neuronal activity
Konorski in 1967 (Konorski, 1967) following the is blocked from reaching both systems. Conse-
original postulate from Pavlov in 1928 . Its neces- quently, patients who achieve full habituation
sity results from the fact that although the brain of their reactions do not experience annoyance,
can detect very weak sound patterns if signifi- anxiety, or any other negative reactions of their
cant (our name called out in a noisy room, cry brain and the body. Note that the patients still
of our baby, our language), it cannot handle perceive their tinnitus and, in case of exclusive
more than one conscious task at any given time habituation of reactions, the proportion of the
(inability to read a book and write a letter ; to time when they are aware of tinnitus is still
understand someone talking while reading; to the same . Nevertheless, since these patients no
listen to two people talking at the same time). longer are bothered by their tinnitus, even when
The question is how do we manage the huge perceived, tinnitus ceases to be an issue.
amount of sensory stimulations that bombard us The secondary goal is to achieve a habitu-
all of the time? For example, how are we able to ation of the perception (Fig . 3B). In this case,
drive a car? tinnitus-related neuronal activity is blocked
The solution to this problem performed by before it reaches the level of awareness, and
the brain is to (1) select and block all unimpor- patients are unaware of the presence of tinnitus .
tant stimuli from reaching our awareness at Note that even when a very high level of habit-
the subconscious level and block reactions that uation of the reaction and perception is achieved,
these stimuli would otherwise evoke (habituate) ; it is still not a cure for tinnitus, as patients can

168
 Tinnitus Retraining Therapy/Jastreboff and Jastreboff

TINNITUS RETRAINING THERAPY

Perception & Evaluation
Auditory 8 other Cortical Areas
Outline of the Treatment
Habituation can be achieved or facilitated by
a number of approaches, including counseling,
combined, for example, with medications, biofeed-
back, hypnosis, etc. However, the easiest imple-
mentation of THT involves a parallel use of
HR retraining counseling with sound therapy. This
Annoyance
therapy has become popular under the name of
Autonomic Nervous System Tinnitus Retraining Therapy (TRT). These two
components perform different functions and both
B have to be used for a method to be called TRT.
Perception & Evaluation
Auditory & other Cortical Areas
Retraining counseling acts to decrease the
level of stimulation from the cortical areas of
Hp the brain to the limbic and autonomic nervous
systems and to decrease the general level of
Emotional Associations activity within these two systems. During coun-
Detection
Subcortical H E Limbic System seling, a number of points are presented to the
II

HR patient : (1) the perception of tinnitus results

I from a compensation occurring within the audi-
Source Annoyance tory system, (2) tinnitus is a problem because
Cochlea Autonomic Nervous System of the activation of emotional (limbic) and auto-
nomic nervous systems, and (3) by using the
Figure 3 Classes of habituation. A, habituation of plasticity of the nervous system, it is possible
reactions, B, habituation of perception and reactions . to retrain the brain to achieve habituation of
Abbreviations: HR = habituation of reactions of the auto-
tinnitus-induced reactions and tinnitus per-
nomic nervous system, HE = habituation of emotional reac-
tions involving the limbic system, Hp = habituation of ception. Once the patient accepts these ideas
perception. as realistic and "making sense" to them, this
puts tinnitus into the category of neutral stim-
uli, to which they may gradually habituate .
hear their tinnitus anytime when their attention The crucial point to recognize, for both the
is focused on it ; also, the tinnitus pitch and therapist and the patient, is that tinnitus-
loudness are the same as at the beginning of the induced reactions are governed by the condi-
treatment. As predicted by the model, tinnitus tioned reflex principle . Consequently, the
maskability may decrease a few decibels (Jas- conscious realization of the benign nature of tin-
treboff et al, 1994), but the change is too small nitus is not sufficient to remove these reac-
to have a practical significance for patients . tions, and a significant amount of time is needed
Although some patients achieve a high level for their gradual extinction .
of tinnitus habituation, to be of clinical signifi- Sound therapy provides significant help in
cance, habituation does not have to be complete, the process of habituation by decreasing the
and patients can experience significant improve- strength of tinnitus-related neuronal activity
ment even with partial habituation of reactions within the auditory system and from the audi-
and perception . The final goal of the treatment tory system to the limbic and autonomic nervous
is that tinnitus ceases to have an impact on the systems . Therefore, the strength of tinnitus-
patient's life . related activation within all loops presented in
In conclusion, Tinnitus Habituation Ther- Figure 1 decreases, making habituation of tin-
apy (THT), aimed at inducing tinnitus habitu- nitus easier.
ation, has been proposed as a treatment for Specifically, sound therapy acts by provid-
tinnitus on the basis of the neurophysiologic ing the auditory system with constant, low-
model of tinnitus (Jastreboff, 1990). Since the level, neutral auditory signals to (1) decrease the
habituation-related modifications of the ner- contrast between tinnitus-related neuronal activ-
vous system occur above the source of tinnitus, ity and background neuronal activity, (2) inter-
the etiology of tinnitus is irrelevant, and any type fere with the detection of the tinnitus signal, and
of tinnitus, as well as somatosounds, can be (3) decrease enhanced gain within the auditory
treated by inducing habituation . pathways .

169
Journal of the American Academy of Audiooogy/Volume 11, Number 3, March 2000

All of our senses are acting upon the prin- "Mixing point " -Tinnitus changing
ciple of gradient or the difference between a sig-
nal and its background . The absolute physical
intensity of stimulus is secondary. The same
sound appears louder and evokes more detect-
able change in the neuronal activity when there
are no other competing sounds. When the same
sound is presented with some additional audi-
Tinnitus
tory background, it will appear to be softer, and suppressed
the neuronal activity evoked by it will become "masking"
more difficult to detect . Presently, we cannot
directly decrease the tinnitus-related neuronal
activity. However, by enhancing background
neuronal activity, through exposing patients to
Sound intensity
low-level sounds, the relative strength of the
tinnitus signal decreases, thereby making habit- Figure 4 Theoretical dependence of the effectiveness
uation easier. of habituation on the intensity of the sound used for the
sound therapy.
Enhancement of the background sound can
be provided by enrichment of environmental
sounds, which can be further amplified by hear-
ing aids for patients with a significant hearing nificant information, essential for a specific diag-
loss or by the use of special sound generators . nosis to be gathered and for the discernment of
Notably, the sound is of importance and not any an individualized treatment for the patient. The
particular means or device providing it . initial contact with a patient is made through a
From the perspective of achieving habitua- questionnaire sent in the mail to those individ-
tion, the masking of tinnitus is counterproduc- uals who expressed an interest to be treated in
tive, since it prevents detection oftinnitus (signal the center. This questionnaire is further
to be habituated), thus preventing, by definition, expanded by an interview, performed before an
habituation . Even partial masking is not rec- audiologic or medical evaluation, which is guided
ommended, as it will change the tinnitus signal, by the Tinnitus/Hyperacusis History form. The
and habituation would occur to this modified sig- main goals of the interview are to (1) identify
nal . Once the external sound is removed and the complaints and any resulting problems, (2) deter-
tinnitus signal is restored to its initial charac- mine the impact of tinnitus on the patient's life,
teristics, it will not be effectively habituated. The (3) assess emotional status and the degree of dis-
relationship of the effectiveness of habituation tress, and (4) evaluate the influence of sound
from the sound intensity is presented in Figure 4 . exposure on the problems .

Stages of the Protocol Audiologic Evaluation

The audiologic evaluation is helpful in

The treatment consists of an initial appoint-
assessing and separating issues of hearing, tin-
ment and a number of follow-up visits . The ini-
nitus, hyperacusis, and phonophobia in each
tial visit, which is described in some detail below,
case . The results are used as the basis for indi-
consists of (1) initial contact with the patient,
vidualized counseling, choosing appropriate
(2) audiologic evaluation, (3) medical evalua-
treatment, and as a reference for evaluation of
tion, (4) diagnosis with decision regarding the
the treatment outcome during follow-up visits .
treatment category, (5) retraining counseling,
After otoscopic examination of the ear canal
and (6) fitting/counseling. Although the initial ap-
and tympanic membrane, a series of tests are
pointment is of great importance, the follow-up
performed. Routine audiologic testing, consist-
visits, even if not very frequent, are crucial in
ing of an evaluation of pure-tone thresholds (up
order to achieve a high level of TRT effectiveness .
to 12 kHz) and word recognition scores, pro-
vides an assessment of a patient's hearing and
Initial Contact with the Patient a basis for subsequent tinnitus measurements .
Specific tinnitus/hyperacusis measurements
The first contact with a patient establishes consist of the evaluation of pitch matching (for
the basis for future interactions and allows sig- the most troublesome tinnitus), loudness match-

170
 Tinnitus Retraining Therapy/Jastreboff and Jastreboff

ing, the minimal masking level, and LDLs . Pitch During the medical evaluation, the main
and loudness matching provide information use- information regarding tinnitus and hyperacusis
ful for counseling but not for diagnosis . The cru- is rechecked, followed by a detailed otolaryngo-
cial measurement is that of LDLs, evaluated logic and general medical evaluation .
using pure tones up to 12 kHz, as well as the fre- Somatosounds might also be detected during
quency that corresponds to the tinnitus pitch. this evaluation . However, since TRT is effective
Measurements are performed twice, and the in their treatment, typically, somatosounds are
second set is recorded . not treated medically or surgically.
DPOAEs allow assessment of the function A frequently asked question from profes-
and integrity of the OHC. This information is sionals who work in an audiologic clinic without
predominantly used for counseling patients, but physicians on site is whether they can treat tin-
it can also be useful in the characterization of nitus patients . The answer is yes, but they
OHC-related hyperacusis (Jastreboff et al, 1998). should require a statement from an otolaryn-
Acoustic immittance, which provides some gologist that the patient's tinnitus can be treated
insight into the integrity of the ascending path- without the risk of removing tinnitus as a symp-
ways of the auditory nerve, is not routinely tom of another medically treatable disease .
tested . This test is not performed if probe tone
levels or acoustic reflex levels exceed LDLs Patient Categories
(probe tone signal of 226 Hz is approximately
60 dB HL) . The patients are placed into one of five gen-
Other tests, such as auditory brainstem eral categories (Table 1) (Jastreboff, 1998).
response or electronystagmography, are per- Although all patients receive counseling and
formed only if there is an indication of medical sound therapy, including the advice to "avoid
problems extending beyond tinnitus. Both tests silence," there are substantial differences for
are interesting for research purposes, but at both components in each category.
this stage they do not provide any clinically use- Categories of patients and their treatments
ful information for patient treatment. are based on four factors . The first is the extent
In summary, a basic audiogram with LDLs of impact tinnitus or hyperacusis has on the
is the crucial test for diagnosis and assessment patient's life and the duration of tinnitus . This
of the treatment outcome . The remaining mea- reflects the strength of the connection formed
surements are useful for individualized coun- between the auditory system and the limbic and
seling and population studies . autonomic nervous systems . The second is the
patient's subjective perception of hearing loss,
Medical Evaluation of with stress placed on the subjectivity of this
TinnituslHyperacusis Patients perception . The third is the presence or absence
of hyperacusis ; threshold of significant hypera-
The medical evaluation of patients with cusis is defined as average LDLs below 100 dB
tinnitus/hyperacusis is directed at identifying HL . It is necessary to assess the relative con-
medical conditions that may cause, contribute tribution of hyperacusis and phonophobia since
to, or have an impact on the treatment of tin- LDLs reflect the sum of both phenomena. The
nitus . The main goal is to exclude any known fourth characteristic is a prolonged worsening
medical condition that has tinnitus as one of its of hyperacusis and/or tinnitus following exposure
symptoms . Typical examples would include an to moderate or loud sounds . This effect is of par-
acoustic neuroma, Meniere's disease, or Lyme ticular significance, as it is a characteristic fea-
disease . In fact, if such a condition is diagnosed, ture of patients with hyperacusis difficult to
the treatment is focused on alleviating this (pri- treat, including hyperacusis resulting from Lyme
mary) cause of tinnitus . There is a danger of disease . Forty-eight percent of patients with
delaying appropriate treatment of the medical Lyme disease have hyperacusis, which exhibits
problem if TRT is successfully used to treat tin- prolonged worsening of the symptoms as a result
nitus before proper medical diagnosis . More- of exposure to moderate or even very low sound
over, during counseling, the perception of tinnitus levels (Fallon et al, 1992) . Some patients with-
is presented to patients as benign, resulting out Lyme disease may also exhibit a similar
from a compensatory mechanism within the effect .
auditory system. Therefore, it is necessary to be Although a number of patients report wors-
sure that this statement is true before present- ening of their tinnitus or hyperacusis as a result
ing it to patients . of exposure to sound, in most cases, this wors-

171
Journal of the American Academy of Audiology/Volume 11, Number 3, March 2000

Table 1 Categories of Tinnitus and Hyperacusis Patients

Prolonged Subjective Impact

Sound-Induced Hearing on
Category Hyperacusis Exacerbation Loss Life Treatment

0 Low Counseling only

1 High Sound generators set at mixing point
2 Present High Hearing aid with stress on enrichment of the
auditory background
3 Present Not relevant High Sound generators set above threshold of
hearing
4 Present Present Not relevant High Sound generators set at the threshold ; very
slow increase of sound level
Hyperacusis : significant sensitivity to environmental sounds typically associated with LDLs below 100 dB HL ; prolonged
sound-induced exacerbation of tinnitus/hyperacusis when the effects persist to the following day; subjective hearing loss : perceived
subjectively by a patient as having a significant impact on patient's life ; impact on life : the extent of impact of tinnitus and/or hyperacusis
on patient's life ; common treatment for each category involves counseling and the use of enriched auditory background .

ening lasts only a few minutes or hours. How- nificant subjective hearing loss . For these
ever, in some cases, it can last for days or weeks. patients to achieve improvement in both tinni-
If the patient experiences worsening of their tus and hearing, we recommend hearing aids. We
tinnitus and/or hyperacusis the morning after instruct the patient to wear them all of the time
the sound exposure, then the patient is classi- while enriching their sound environment. It is
fied as having a prolonged impact to noise expo- stressed to the patient that sound is important
sure . The resulting categories from this for the treatment and not the hearing aids . The
classification are presented in Table 1 . main purpose of the hearing aids is to amplify
Category 0 consists of patients who do not sound, whereas providing better communica-
have hyperacusis or any significant hearing tion is secondary.
loss, and whose tinnitus has little impact on Category 3 consists of patients with signif-
their life. For these patients, the directive coun- icant hyperacusis, which is not enhanced, for a
seling session, including the advice to avoid prolonged period of time, as a result of sound
silence and to enrich their sound environment, exposure . Tinnitus may or may not be present.
is usually sufficient and there is no need for Sound generators are necessary to help desen-
any instrumentation . Patients with a recent sitize the auditory system and consequently to
onset of tinnitus, not exceeding more than 2 decrease/remove hyperacusis. The desensitiza-
months, and who have not received any "nega- tion protocol begins with the sound level set
tive counseling," leaving them with many con- close to, but clearly above, the threshold of hear-
cerns and a little hope for the potential ing. This level is increased during the treat-
improvement (Jastreboff and Hazell, 1993), ment to the level appropriate to their tinnitus
belong to this category as well . (if present). These patients tend to recover faster
Category 1 consists of patients who have than patients with tinnitus only.
significant tinnitus but no hyperacusis and no Category 4 consists of patients who have
subjective hearing loss . For these patients, the tinnitus and/or hyperacusis and exhibit pro-
most effective approach is the use of sound gen- longed worsening of their symptoms as a result
erators set at the level close to the "mixing"/ of sound exposure . This is the most difficult cat-
"blending" point. This is the sound level corre- egory of patients to treat, and the success rate
sponding to the beginning of partial suppression is lower than in the other four categories . In this
("partial masking"). Patients describe it as when case, we set the level of sound generators at the
the external sound and tinnitus can be heard sep- threshold of hearing. In cases where there is gen-
arately but start to interfere or intertwine with eral hypersensitivity of perception of any type,
each other. Most tinnitus patients belong to this not just sound, the patients are advised to wear
category and exhibit a high level of success in the devices for a week without turning them
controlling their tinnitus . on . This is done in order to desensitize the
Category 2 consists of patients with the patient's perception of the touch to devices in
characteristics of Category 1 but additional sig- their ears . As the treatment progresses, the

172
 Tinnitus Retraining Therapy/Jastreboff and Jastreboff

sound level is increased very slowly. These prevention of the attenuation of low-frequency
patients need continuous monitoring and typi- environmental sound . In-the-canal hearing aids
cally exhibit profound phonophobia . are not recommended except in cases with a
significant low-frequency hearing loss, since
Specific Issues of Treatment of they will attenuate low-frequency environmen-
Hyperacusis and Phonophobia tal sound, which, in turn, typically results in the
increase of tinnitus .
In most cases, hyperacusis can be treated In cases of unilateral deafness and tinnitus,
directly by a process of gradual desensitization CROS, BICROS, or transcranial stimulation,
of the auditory system . If hyperacusis is present combined with training to improve space local-
(e .g ., Categories 3 and 4), then it must be treated ization of the sound, is recommended . The goal
first, before the tinnitus . After the patient shows is to reactivate parts of the auditory pathways
improvement in his/her hyperacusis, the tinni-
that received decreased input as a result of deaf-
tus is addressed more directly. Frequently, how- ness . This approach is based on recent devel-
ever, as the patient gets the hyperacusis under opments in neuroscience .
control, the tinnitus becomes less of an issue . For It is common knowledge that the nervous
the hyperacusis patient, it is even more impor- system exhibits an enormous amount of plas-
tant than for patients with tinnitus only to have ticity and that information from various sensory
an enriched sound environment in addition to systems is integrated into a coherent entity. The
the use of instruments . It is also important to visual and vestibular systems are classic exam-
discontinue the overuse of ear protection as it ples of such a collaboration. It also has been
causes an increase in the sensitivity of the audi- recognized that, in the absence of sensory input,
tory system due to decreased auditory input . phantom perception occurs (phantom limb, phan-
Typically, patients combine and confuse tom pain, tinnitus), with accompanied reorga-
hyperacusis and phonophobia . LDLs, although
nization of receptive fields . Afew years ago, a new
essential, are not sufficient for the diagnosis of dramatic development was reported for con-
hyperacusis ; therefore, a detailed interview is trolling phantom pain and phantom limb by
necessary. Once hyperacusis is at least partially using multisensory interaction (Ramachandran
under control, patients are treated for their and Rogers-Ramachandran, 1996). Phantom
phonophobia by training involving engagement pain and phantom limb frequently cannot be
in activities that they enjoy and have sound as controlled by any pharmacologic or surgical
an inevitable component . approach . However, by introducing visual input,
Instruments it turned out to be possible to control phantom
pain in patients with one of their hands ampu-
In theory, all patients who do not have hyper- tated. These patients were instructed to put the
acusis can be treated without instrumentation . healthy hand into the box with a glass top and
In practice, however, it is advisable to use some the mirror inside, so they saw only the healthy
form of instrumentation for most patients (except hand and its mirror reflection, which mimicked
Category O) due to the following reasons . First, the missing hand, and to move the hand . After
about 40 percent of patients have hyperacusis several sessions, the phantom pain, which could
(Jastreboff et al, 1996b) . For these patients, not be controlled by other means, disappeared.
there is a need for a well-controlled, stable sound Presumed mechanisms of action involved reor-
source, such as sound generators . Second, the ganization of receptive fields of somatosensory
increased ease of implementing sound therapy representation of the hands by visual input and
by patients who use sound generators results in partially restoring the balance disturbed by the
better compliance with the protocol . Third, for lack of sensory input from the missing hand .
patients with significant subjective hearing loss Recent data with fMRI strongly supported this
(Category 2), the use of hearing aids provides an postulate (Borsook et al, 1998).
additional benefit in improved hearing . A wide This information had direct effect on the
variety of hearing aids may be used . The gen- treatment of tinnitus in patients with profound
eral principle is the use of a high-quality, pro- unilateral hearing loss or unilateral deafness .
grammable hearing aid, fit with an open mold, Although the high level of plasticity of the ner-
to ensure significant improvement of hearing vous system was recognized long ago, the extent
under various environmental conditions, pro- of plasticity and reorganization of receptive fields
tection against overstimulation (by use of higher within the auditory system was not sufficiently
than typically selected compression ratio), and appreciated. Recent data changed this situation

173
 Journal of the American Academy of Audiology/Volume 11, Number 3, March 2000

dramatically, with results showing reorganiza- actively promoted, habituation acts to decrease
tion of the tonotopic cortical maps due to the the strength of those connections. The outcome
presence of tinnitus (Muhlnickel et al, 1998). depends upon the relative strength of the
Based on the results with phantom pain, the patients' annoyance and habituation. Conse-
ideas was to use the combined actions of the quently, the continuous push toward habituation
auditory and visual systems by fitting patients is needed to counteract continuous negative
with CROS, BICROS, or transcranial systems. reinforcement and enhancement of reflex loops
This provided them with auditory information connecting the tinnitus signal with reactions of
from a whole auditory space, which, in combi- the limbic and autonomic nervous systems pro-
nation with the information from the visual sys- vided by tinnitus-evoked annoyance. Therefore,
tem, would restore spatial localization of the the counseling component of the follow-up vis-
sound and modify receptive fields in the auditory its is essential in order to provide a patient with
pathways . The clinical results confirmed that continuous promotion and facilitation of habit-
these patients had partially restored their space uation . The sound therapy alone without direc-
localization of auditory stimuli (tested with closed tive counseling is not sufficient. However, once
eyes). Furthermore, as hoped, this method also weakening of the insidious connections reaches
was helpful for their tinnitus . A systematic study a sufficient level, further habituation occurs
on a large number of cases is needed, but results spontaneously.
obtained so far are very encouraging.
There are a variety of sound generators . Results
The ideal sound generator should (1) minimally
interfere with the perception of external sound, The process of habituation is slow and grad-
(2) allow for a smooth increase in the sound ual, with fluctuations of tinnitus severity, includ-
level from the threshold of hearing, (3) provide ing transient perception of tinnitus worsening
a reasonably wide frequency range of generated observed in some patients about 3 to 4 weeks
sound, (4) provide a stable amplitude and spec- after the initiation of the treatment. This para-
trum of sound, and (5) be cosmetically pleasing. doxical phenomenon is actually a positive sign,
Although none of the 18 models of sound gen- as it probably indicates the beginning of habit-
erators evaluated by us were ideal, a number of uation . Patients experience temporary relief
them may be used for treatment. At the moment, due to partial habituation, but when tinnitus
we recommend the three most interesting returns to the previous state it appears to be
devices : (1) Silent Star (Viennatone/ReSound, worse, due to the contrast with the previous
USA), (2) Tranquil (General Hearing Instru- period of habituation. Initial improvement is
ments, class D model with enhanced high fre- typically seen within the first few months, fol-
quencies), and (3) the sound generator from lowed by further gradual improvement. Clear
Audifon (Spain). results can be seen in about a year, but to pre-
Two sound generators are always recom- vent a relapse it is recommended that treat-
mended, including cases of unilateral tinnitus . ment last at least 18 months . Some patients
This is to ensure symmetric stimulation of the follow the protocol even longer and experience
auditory pathways and avoidance of abnormal further improvement.
modification of connections and receptive fields Due to the lack of an objective method for
within the auditory system, thereby pushing measuring tinnitus, selecting the proper crite-
perception of tinnitus to one side . ria for assessing the effectiveness of the treat-
ment outcome is of fundamental importance .
Follow-up Visits Multidimensional evaluation is necessary with
an emphasis on changes and increased involve-
To achieve the habituation of tinnitus, TRT ment in life activities that were previously pre-
is aimed at the reversal or retraining of the vented or interfered with by tinnitus or
feedback loops formed between the auditory, hyperacusis (e .g., the patient has resumed square
limbic, and autonomic nervous systems. Modi- dancing) . Comparison of the initial question-
fications of these loops occur in a dynamic bal- naire with the follow-up questionnaire allows
ance scenario . Under a dynamic balance observation of changes that may have occurred
condition, the annoyance that the person expe- in life activities .
riences acts (continuously) to increase the Moreover, patients are asked to assess their
strength of connections between the auditory sys- tinnitus awareness, annoyance, and effect on
tem and other systems in the brain . When life on a scale of 0 to 10 before, during, and after

174
 Tinnitus Retraining Therapy/Jastreboff and Jastreboff

treatment is performed . To classify a patient as nitus and hyperacusis, can be used to treat all
showing "significant improvement," the follow- types of patients, does not require frequent vis-
ing criteria are used : (1) at least one activity its, and does not interfere with hearing, and
previously prevented/interfered with is no longer there are no negative side effects . The protocol
affected or all activities show improvement ; requires limited time for treatment, and many
(2) tinnitus awareness is decreased by at least 20 successful patients who completed the treat-
percent, the impact of tinnitus on life is decreased ment and stopped using the devices have no
by at least 20 percent, and tinnitus annoyance need to use them several years later; tinnitus is
is decreased by at least 20 percent ; (3) evalua- not a problem in their life . Moreover, approxi-
tion was performed after at least 6 months of mately 20 percent of these patients achieved a
treatment and is repeated at least once, with the block of the perception of their tinnitus to the
last assessment performed not later than 3 years extent that they could not hear tinnitus even
after initiation of the treatment ; and (4) an when focusing their attention on it (Sheldrake
improvement in more than one category. If there et al, 1996) .
is improvement in only one category, then the The main negative aspect of TRT is that
patient is classified as showing no improvement . the protocol has to focus on the individual needs
Note that the results described below are not and profile of a patient, consequently requiring
an outcome of the planned studies designed to significant time involvement of the personnel
validate TRT protocol but rather a brief sum- providing the treatment, who have to be specif-
mary of the retrospective analysis of the treat- ically trained. Moreover, the development of
ment outcome of a nonpreselected sample of the specific plastic changes within the nervous sys-
University of Maryland Tinnitus and Hypera- tem (leading to habituation of tinnitus) requires
cusis Center patients . The progress of the treat- about 18 to 24 months, as estimated from our ret-
ment was monitored by the use of questionnaires rospective observation of the patients .
during initial and follow-up visits or telephone Although we are not claiming that TRT is
interviews . Of 263 patients with whom we were the only method to treat tinnitus patients, nor
able to keep contact, 90 .1 percent received instru- that it is finalized, we believe, that when imple-
ments (82 .5% sound generators and 7 .6% hear- mented properly, TRT is effective in helping tin-
ing aids) and stated that they were following nitus and hyperacusis patients . Since it is based
TRT. Of these patients, 9 .9 percent received one upon a scientific model, it can be further tested
session of counseling, including information and refined. It is not a cure, since it does not
about sound therapy and no instruments . These remove tinnitus, but by inducing habituation of
patients typically did not follow TRT. tinnitus-induced reactions and tinnitus per-
Combining results obtained from all 263 ception, it allows patients to achieve control of
patients, including those who decided not to fol- their tinnitus, live a normal life, and participate
low TRT, revealed that 75 percent of patients in everyday activities . TRT does not promise a
reached the level of significant improvement as 100 percent success rate, but we feel it is justi-
defined above. The results are even more opti- fied to promote it as an effective treatment for
mistic (above 80%) for patients using noise gen- tinnitus and hyperacusis patients . Neverthe-
erators or hearing aids as a part of TRT. On less, this should not stop efforts to search for
average, the indices for awareness, annoyance, better methods, particularly research on mech-
and life quality decreased to about half of their anism-based tinnitus alleviation .
pretreatment values . Additionally, the analysis
of the relationship of the treatment outcome
REFERENCES
with the presence of hyperacusis performed on
163 cases revealed that patients with hypera-
cusis (combined Categories 3 and 4) showed a Boettcher FA, Salvi RJ . (1993) . Functional changes in
the ventral cochlear nucleus following acute acoustic
higher rate of improvement than patients with overstimulation . JAcoust Soc Am 94 :2123-2134 .
tinnitus only (Categories 1 and 2) .
Borsook D, Becerra L, Fisherman S, Edwards A, Jennings
SL, Stojanovic M, Papinicolas L, Ramachandran VS,
CONCLUSIONS Gonzales RG, Breiter H. (1998) . Acute plasticity in the
human somatosensory cortex following amputation .
NeuroReport 9:1013-1017 .

A s with any method, TRT has both positive

and negative aspects. On the positive side,
TRT appears to be highly effective for both tin-
Brooks CM . (1987) . Autonomic nervous system, nature
and functional role . In : Adelman G, ed . Encyclopedia of
Neuroscience . Boston : Birkhauser, 96-98.

175
 Journal of the American Academy of Audiology/Volume 11, Number 3, March 2000

Chen GD, JastreboffPJ . (1995) . Salicylate-induced abnor- Jastreboff PJ, Gray WC, Mattox DE . (1998) . Tinnitus
mal activity in the inferior colliculus of rats . Hear Res and hyperacusis . In : Cummings CW Fredrickson JM,
82 :158-178 . Harker LA, Krause CJ, Richardson MA, Schuller DE,
eds. Otolaryngology Head & Neck Surgery. St . Louis:
Coles RRA. (1987) . Epidemiology of tinnitus. In : Hazell Mosby, 3198-3222.
JWP, ed. Tnnitus. Edinburgh: Churchill Livingstone,
46-70. Jastreboff PJ, Hazell JWP (1993) . Aneurophysiological
approach to tinnitus : clinical implications . Br J Audiol
Coles RRA. (1996). Epidemiology, aetiology and classifi- 27 :1-11.
cation . In: Vernon JA, Reich G, eds. Proceedings of the
Fifth International Tinnitus Seminar, 1995 . Portland, Jastreboff PJ, Hazell JWP, Graham RL . (1994) .
OR : American Tinnitus Association, 25-30. Neurophysiological model of tinnitus : dependence of the
minimal masking level on treatment outcome . Hear Res
Coles RRA, Sood SK. (1988) . Hyperacusis and phono- 80 :216-232 .
phobia in tinnitus patients . Br JAudiol 22 :228 .
JastreboffPJ, Jastreboff MM, Kwon O, Shi J, Hu S. (1999).
Drukier GS. (1989) . The prevalence and characteristics An animal model of noise induced tinnitus . In : Hazell
of tinnitus with profound sensori-neural hearing impair- JWP, ed . Proceedings of the Sixth International Tinnitus
ment. Am Ann Deaf 134 :260-264 . Seminar, Cambridge, UK. London : 198-202.

Fallon BA, Nields JA, Burrascano JJ, Liegner K, DelBene Jastreboff PJ, Jastreboff MM, Sheldrake JB . (1996b).
D, Liebowitz MR . (1992) . The neuropsychiatric mani- Utilization of Loudness Discomfort Levels in the treat-
festation of Lyme borreliosis. Psychiatric Q 63:95-117. ment of hyperacusis, tinnitus, and hearing loss . Assoc
Res Otolaryngol 19 .
Feldmann H. (1988) . Pathophysiology of tinnitus . In :
Kitahara M, ed . Tinnitus : Pathophysiology and Konorski J. Integrative Activity of the Brain. Chicago:
Management . Tokyo : Igaku-Shion, 7-35 . University of Chicago Press, 1967 .

Gerken GM . (1992). Central auditory temporal process-

Kuk FK, Tyler RS, Russell D, Jordan H . (1990) . The psy-
ing: alterations produced by factors involving the cochlea.
chometric properties of a tinnitus handicap questionnaire .
In : Dancer A, Henderson D, Salvi R, Hamernik R, eds.
Ear Hear 11 :434-445 .
Effect of Noise on the Auditory System . Philadelphia :
Mosby, 146-155.
Liberman MC, Kiang NYS . (1978) . Acoustic trauma in
cats . Acta Otolaryngol Suppl 358 :1-63 .
Gerken GM. (1993) . Alteration of central auditory pro-
cessing of brief stimuli: a review and a neural model. J
Acoust Soc Am 93 :2038-2049 . Liberman MC, Mulroy MJ . (1982) . Acute and chronic
effects of acoustic trauma : cochlear pathology and audi-
tory nerve pathophysiology. In : Hamernik RP, Henderson
Hallam RS, Rachman S, Hinchcliffe R . (1984) . D, Salvi R, eds. New Perspectives on Noise-Induced
Psychological aspects of tinnitus . In : Rachman S, ed .
Hearing Loss. New York : Raven Press, 105-136.
Contribution to Medical Psychology . Vol . 3 . Oxford :
Pergamon Press, 31-34 .
McFadden D. (1982) . Tinnitus : Facts, Theories, and
Treatments. Washington, DC : National Academy Press.
Hazell JWP, Sheldrake JB . (1992) . Hyperacusis and tin-
nitus. In : Aran J, Dauman R, eds. Tinnitus 91 . Proceedings
of the IV International Tinnitus Seminar. Amsterdam: Moore BCJ. (1995) . An Introduction to the Psychology of
Kugler, 245-248. Hearing. San Diego: Academic Press, 1-350.

Heller MF, Bergman M. (1953) . Tinnitus in normally Muhlnickel W, Elbert T, Taub E, Flor H. (1998). Reorgani-
hearing persons . Ann Otol 62 :73-93 . zation of auditory cortex in tinnitus . Proc Natl Acad Sci
USA 95 :10340-10343 .
Jakes SC, Hallam RS, Rachman S, Hinchcliffe R. (1986) .
The effects of reassurance, relaxation training and dis- Newman CW Wharton JA, Jacobson GP. (1995) . Retest
traction on chronic tinnitus sufferers. Behav Res Ther stability ofthe tinnitus handicap questionnaire . Ann Otol
24:497-507 . Rhinol Laryngol 104:718-723 .

JastreboffPJ . (1990) . Phantom auditory perception (tin- Ramachandran VS, Rogers-Ramachandran D. (1996).
nitus) : mechanisms of generation and perception . Neurosci Synaesthesia in phantom limbs induced with mirrors.
Res 8:221-254. Proc R Soc Lond B Biol Sci 263:377-386 .

JastreboffPJ . (1995). Tinnitus as a phantom perception : Ruggero MA, Rich NC, Recio A, Narayan SS, Robles L.
theories and clinical implications . In : Vernon J, Moller (1997) . Basilar-membrane responses to tones at the base
AR, eds. Mechanisms of 7lnnitus. Boston : Allyn & Bacon, of the chinchilla cochlea . JAcoust Soc Am 101 :2151-2163.
73-94.
Salomon G. (1989) . Hearing problems and the elderly.
JastreboffPJ . (1998) . Tinnitus ; the method of. In: Gates Dan Med Bull 33(Suppl 3) :1-22 .
GA, ed . Current Therapy in Otolaryngology Head and
Neck Surgery. St . Louis: Mosby, 90-95. Salvi RJ, Wang J, Powers N. (1996). Rapid functional
reorganization in the inferior colliculus and cochlear
JastreboffPJ, Gray WC, Gold SL. (1996a). Neurophysio- nucleus after acute cochlear damage . In : Salvi RJ,
logical approach to tinnitus patients . Am J Otol Henderson D, Fiorino F, Colletti V, eds. Auditory System
17 :236-240 . Plasticity and Regeneration. New York : Thieme, 275-296.

176
 Tinnitus Retraining Therapy/Jastreboff and Jastreboff

Sataloff J, Sataloff RT, Lueneburg W (1987) . Tinnitus Swanson LW (1987) . Limbic system . In : Adelman G, ed .
and vertigo in healthy senior citizens without a history Encyclopedia of Neuroscience . Boston : Birkhauser,
of noise exposure . Am J Otol 8 :87-89 . 589-591.

Sheldrake JB, Jastreboff PJ, Hazell JWP. (1996) . Tonndorf J. (1987) . The analogy between tinnitus and
Perspectives for total elimination of tinnitus perception . pain: a suggestion for a physiological basis of chronic tin-
In : Vernon JA, Reich G, eds. Proceedings of the Fifth nitus. Hear Res 28 :271-275 .
International tinnitus Seminar. Portland, OR : American
Tinnitus Association, 531-536.
Vernon JA. (1987) . Pathophysiology of tinnitus: a special
Stouffer JL, Tyler RS . (1990) . Characterization of tinni- case-hyperacusis and a proposed treatment. Am J Otol
tus by tinnitus patients . J Speech Hear Disord 8:201-202 .
55 :439-453 .

Stouffer JL, Tyler RS, Kileny PR, Dalzell LE . (1991) . Vernon J, Press L. (1998) . Treatment for hyperacusis. In :
Tinnitus as a function of duration and etiology : coun- Vernon JA, ed . tinnitus D-eatment and Relief. Boston :
selling implications . Am J Otol 12 :188-194 . Allyn and Bacon, 223-227.