Ebook Clinical assessment of voice (2E): Part 2

15
Hearing Loss in Singers
and Other Musicians
Robert Thayer Sataloff, Joseph Sataloff, and Brian McGovern

Singers and other musicians depend on good hearing
to match pitch, monitor vocal quality, and provide
feedback and direction for adjustments during performance. The importance of good hearing among performing artists has been underappreciated. Although
well-trained musicians are usually careful to protect
their voices or hands, they may subject their ears to
unnecessary damage and thereby threaten their musical careers. The ear is a critical part of the musician’s
“instrument.” Consequently, it is important for singers to understand how the ear works, how to take care
of it, what can go wrong with it, and how to avoid
hearing loss from preventable injury.

Causes of Hearing Loss
The classification and causes of hearing loss have
been described in detail in standard textbooks of
otolaryngology and previous works by the authors,1,2
and they will be reviewed only briefly in this chapter. Hearing loss may be hereditary or nonhereditary,
and either form may be congenital (present at birth)
or acquired. There is a common misconception that
hereditary hearing loss implies the presence of the
problem at birth or during childhood. In fact, most
hereditary hearing loss occurs later in life. All otolaryngologists know families whose members begin
to lose their hearing in their third, fourth, or fifth
decade, for example. Otosclerosis, a common cause
of correctable hearing loss, often presents when people are in their 20s or 30s. Similarly, the presence of
deafness at birth does not necessarily imply hereditary or genetic factors. A child whose mother had

rubella during the first trimester of pregnancy or was

exposed to radiation early in pregnancy may be born
with a hearing loss. This is not of genetic etiology
and has no predictive value for the hearing of the
child’s siblings or future children. Hearing loss may
occur because of problems in any portion of the ear,
the nerve between the ear and the brain, or the brain.
Understanding hearing loss requires a basic knowledge of the structure of the human ear.

Anatomy and Physiology of the Ear
The ear is divided into 3 major anatomical divisions:
the outer ear, the middle ear, and the inner ear.
The outer ear has 2 parts: (1) the trumpet-shaped
apparatus on the side of the head, the auricle or pinna,
and (2) the tube leading from the auricle into the temporal bone, the external auditory canal. The opening of
the tube is called the meatus.
The middle ear is a small cavity in the temporal bone
in which 3 auditory ossicles, the malleus (hammer),
incus (anvil), and stapes (stirrup), form a bony bridge
from the external ear to the inner ear (Figure 15–1).
This bony bridge is held in place by muscles and ligaments. The tympanic membrane or eardrum stretches
across the inner end of the external ear canal, separating the outer ear from the middle ear. The middleear chamber normally is filled with air and connects
to the nasopharynx through the eustachian tube. The
eustachian tube helps to equalize pressure on both
sides of the eardrum.
The inner ear is a fluid-filled chamber divided into
2 parts: (1) the vestibular labyrinth, which functions

257

258

Clinical Assessment of Voice

Figure 15–1.  Cross section of the ear. The semicircular canals are part of the balance
system.

as part of the body’s balance mechanism, and (2) the
cochlea, which contains thousands of minute, sensory,
hairlike cells (Figure 15–2) responsible for beginning
the electrical stimulation to the brain. The organ of
Corti functions as the switchboard for the auditory
system. The eighth cranial (acoustic) nerve leads
from the inner ear to the brain, serving as the pathway for the electrical impulses that the brain will
interpret as sound.
Sound begins from a source that creates vibrations
or sound waves in the air somewhat similar to the
waves created when a stone is thrown into a pond.
The pinna collects these sound waves and funnels
them down the external ear canal to the eardrum. The
sound waves then cause the eardrum to vibrate.
These vibrations are transmitted through the middle
ear over the bony bridge or ossicular chain formed by
the malleus, incus, and stapes. The vibrations in turn
cause the membranes over the openings to the inner
ear to vibrate, causing the fluid in the inner ear to be
set in motion. The motion of the fluid in the inner ear
displaces the hair cells, which in turn excite the nerve
cells in the organ of Corti, producing electrochemical
impulses that are transmitted to the brain along the

acoustic nerve. As the impulses reach the brain, we
experience the sensation of hearing.

Establishing the Site of Damage
in the Auditory System
The cause of a hearing loss, like that of any other medical condition, is determined by obtaining a detailed
history, making a thorough physical examination,
and performing various clinical and laboratory tests.
An audiogram provides a “map” of hearing and
details the levels at which sound is detected at various frequencies. When a hearing loss is identified, an
attempt is made to localize the point along the auditory pathway where the difficulty has originated.
Every attempt to determine whether the patient’s
hearing loss is conductive, sensorineural, central,
functional, or a mixture of these is made. However,
sometimes these distinctions can be difficult to make.
In particular, it is very difficult to distinguish sensory
from neural lesions.
Details of the otologic history, physical examination,
and test protocols are detailed in many otolaryngology texts. Medical evaluation of a patient with a suspected hearing problem includes a comprehensive
history; complete physical examination of the ears,
nose, throat, head, and neck; assessment of the cranial
nerves, including testing the sensation in the external
auditory canal (Hitselberger sign); audiogram (hear-




15.  Hearing Loss in Singers and Other Musicians

Figure 15–2.  Cross-section of the organ of Corti. A. Low magnification. B. Higher magnification.

259


260

Clinical Assessment of Voice

ing test); and other tests, as indicated. Recommended
additional studies may include computed tomography (CT), magnetic resonance imaging (MRI),
dynamic imaging studies such as single-photon
emission computed tomography (SPECT), position
emission tomography (PET), specialized hearing
tests such as brainstem evoked response audiometry (ABR or BERA), electronystagmography (ENG),
computerized dynamic posturography (CDP), otoacoustic emissions, immittance measures, central
auditory processing testing, and a variety of blood
tests for the many systemic causes of hearing loss.
All patients with hearing complaints deserve a thorough examination and comprehensive evaluation to
determine the specific cause of the problem and to
rule out serious or treatable conditions that may be
responsible for the hearing impairment. Contrary to
popular misconceptions, not all cases of sensorineural hearing loss are incurable. So “nerve deafness”
should be assessed with the same systematic vigor
and enthusiasm as conductive hearing loss.

forations usually do not cause a great deal of hearing
impairment. Hearing loss from middle ear dysfunction is the most common cause of conductive hearing loss and may cause a hearing decrease of up to
60 decibels. It may occur in many ways. The middle ear may become filled with fluid because of
eustachian tube dysfunction. The fluid restricts free
movement of the tympanic membrane and ossicles,

thereby producing hearing loss. Middle-ear conductive hearing loss may also be caused by ossicular
abnormalities. These include fractures, erosion from
disease, impingement by tumors, congenital malformations, and other causes. However, otosclerosis is
among the most common. This hereditary disease
afflicts the stapes and prevents it from moving in its
normal piston-like fashion in the oval window. Hearing loss from otosclerosis can be corrected through
stapes surgery, a brief operation under local anesthesia, and it is usually possible to restore hearing.

Conductive Hearing Loss

The word sensorineural was introduced to replace the
ambiguous terms perceptive deafness and nerve deafness. The term sensory hearing loss is applied when
the damage is localized to the inner ear and auditory
nerve. The cochlea has approximately 15 000 hearing
nerve endings (hair cells). Those hair cells, and the
nerve that connects them to the brain, are susceptible
to damage from a variety of causes. Neural hearing loss
is the correct term to use when the damage is in the
auditory nerve proper, anywhere between its fibers
at the base of the hair cells and the auditory nuclei.
This range includes the bipolar ganglion of the eighth
cranial nerve. Other common names for this type of
loss are nerve deafness and retrocochlear hearing loss.
These names are useful if applied appropriately and
meaningfully, but too often they are used improperly.
Although at present it is common practice to group
together both sensory and neural components, it has
become possible through advanced diagnostic techniques to attribute a predominant part of the damage, if not all of it, to either the inner ear or the nerve.
This separation is advisable because the prognosis
and the treatment of the 2 kinds of impairment differ.

For example, in all cases of unilateral sensorineural
hearing loss, it is important to distinguish between a
sensory and neural hearing impairment, because the
neural type may be due to a tumor called an acoustic
neuroma, which could become life-threatening if left
untreated. Cases that we cannot identify as either
sensory or neural and cases in which there is damage
in both regions we classify as sensorineural.

In cases of conductive hearing loss, sound waves are
not effectively transmitted to the inner ear as a result
of some mechanical defect in the outer or middle
ear. The outer and middle ear normally enhance and
transfer sound energy to the inner ear or cochlea. In
a purely conductive hearing loss, there is no damage
to the inner ear or the neural pathway; rather, the
damage lies in the external canal or the middle ear.
Patients diagnosed as having conductive hearing
loss have a much better prognosis than those with
sensorineural loss, because modern techniques make
it possible to cure or at least improve the vast majority
of cases in which the damage occurs in the outer or
middle ear. Even if they are not improved medically
or surgically, these patients stand to benefit greatly
from a hearing aid, because what they need most is
amplification. They are not bothered by distortion
and other hearing abnormalities that may occur in
sensorineural hearing losses.
Some more common types of conductive hearing
loss may result from a complete or partial blockage of the outer ear, which will interfere with sound

transmission to the middle ear. Outer ear problems
include birth defects, total occlusion of the external
auditory canal by wax, foreign body (eg, a piece of
cotton swab or ear plug), infection, trauma, or tumor.
Large perforations in the tympanic membrane may
also cause hearing loss, especially if they surround
the malleus. However, relatively small, central per-

Sensorineural Hearing Loss




15.  Hearing Loss in Singers and Other Musicians

There are various and complex causes of sensorineural hearing loss, but certain features are characteristic and basic to all of them. Because the histories
obtained from patients are so diverse, they contribute
more insight into the etiology than into the classification of a cause. Sensorineural hearing loss often
involves not only loss of loudness but also loss of clarity. The hair cells in the inner ear are responsible for
analyzing auditory input and instantaneously coding
it. The auditory nerve is responsible for carrying this
complex coded information. Neural defects such as
acoustic neuromas (benign tumors of the auditory
nerve) are frequently accompanied by severe difficulties in discriminating sounds and words effectively, although the actual hearing threshold for
differences in sounds may not be greatly affected.
Sensory deficits in the cochlea are often associated
with distortion of sound quality, distortion of loudness (loudness recruitment), and distortion of pitch
(diplacusis). Diplacusis poses particular problems
for musicians, because it may make it difficult for
them to tell whether they are playing or singing correct pitches. This symptom is also troublesome to

conductors. Keyboard players and other musicians
whose instruments do not require critical tuning
adjustments compensate for this problem better than
singers, string players, and the like. In addition, sensorineural hearing loss may be accompanied by tinnitus (noises in the ear) and/or vertigo. However, it
is possible to have these auditory symptoms and not
demonstrate a hearing loss on a routine audiogram.
Hearing loss may be present at frequencies between
or above those usually tested and can be detected
with special audiometers that test all (or nearly all)
of the frequencies from 125 to 12 000 Hz. Special
ultrahigh-frequency audiometers are available commercially and can measure hearing thresholds up to
20 000 Hz. An evaluation of this hearing range can
show damage that could not be detected at routinely
tested frequencies.
Sensorineural hearing loss may be due to a great
number of conditions, including exposure to ototoxic
drugs (including a number of antibiotics, diuretics,
and chemotherapy agents), hereditary conditions,
systemic diseases, trauma, and noise, among other
causes. Most physicians recognize that hearing loss
may be associated with a large number of hereditary
syndromes2,3 involving the eyes, kidneys, heart, or
any other body system; but many are not aware that
hearing loss also accompanies many, very common
systemic diseases. Naturally, these occur in musicians as well as others. The presence of these systemic
illnesses should lead physicians to inquire about

261

hearing and to perform audiometry in selected cases.

Problems implicated in hearing impairment include
Rh incompatibility, hypoxia, jaundice, rubella, mumps,
rubeola, fungal infections, meningitis, tuberculosis,
sarcoidosis, Wegener granulomatosis, vasculitis, histiocytosis X, allergy, hyperlipoproteinemia, syphilis,
hypothyroidism, hypoadrenalism, hypopituitarism,
renal failure, autoimmune disease, coagulopathies,
aneurysms, vascular disease, multiple sclerosis, infestations, diabetes, hypoglycemia, cleft palate, and others.2
Prolonged exposure to very loud noise is a common cause of hearing loss in our society. Noiseinduced hearing loss is seen most frequently in heavy
industry. However, occupational hearing loss caused
by musical instruments is a special problem, as discussed below.
Mixed Hearing Loss
For practical purposes, a mixed hearing loss should be
understood to mean a conductive hearing loss accompanied by a sensory, neural (or a sensorineural) loss
in the same ear. However, the clinical emphasis is
on the conductive hearing loss, because available
therapy is so much more effective for this disorder.
Consequently, the otologic surgeon has a special
interest in cases of mixed hearing loss in which there
is primarily a conductive loss complicated by some
sensorineural damage. In a musician, curing the
correctable component may be sufficient to convert
hearing from unserviceable to satisfactory for performance purposes.
Functional Hearing Loss
Functional hearing loss occurs as a condition in which
the patient does not seem to hear or to respond, yet
the handicap is not caused by any organic pathology in the peripheral or central auditory pathways.
The hearing difficulty may have an entirely psychological etiology, or it may be superimposed on some
mild organic hearing loss, in which case it is called a
functional or a psychogenic overlay. Often, the patient
has normal hearing, but the secondary gain from a

hearing loss, even if it is not organic, motivates the
patient to behave as though he or she has a legitimate
hearing loss. In some cases, the patient may not even
realize that the loss is nonorganic. A careful history
usually will reveal some hearing impairment in the
patient’s family or some personally meaningful reference to deafness that generated the patient’s psychogenic hearing loss. The important challenge for
the clinician in such a case is to classify the condition


262

Clinical Assessment of Voice

properly, so that effective treatment can be initiated.
Functional hearing loss occurs not only in adults, but
also in children. This diagnosis should be considered
whenever hearing problems arise in musicians under
great pressure regardless of age, including young
prodigies.
Central Hearing Loss (Central Dysacusis)
In central hearing loss, the damage is situated in the
central nervous system at some point in the brain
between the auditory nuclei (in the medulla oblongata) and the cortex. Formerly, central hearing loss
was described as a type of “perceptive deafness,” a
term now obsolete.
Although information and research about central
hearing loss has developed, it remains complex and
unclear. Physicians know that some patients cannot
interpret or understand what is being said and that
the cause of the difficulty is not in the peripheral

mechanism but somewhere in the central nervous
system. In central hearing loss, the problem is not a
lowered pure-tone threshold but the patient’s ability to interpret what he or she hears. Obviously, it
is a more complex task to interpret speech than to
respond to a pure-tone threshold; consequently, the
tests necessary to diagnose central hearing impairment must be designed to assess a patient’s ability to
handle complex information.

Psychological Consequences
of Hearing Loss
Performing artists are frequently sensitive, somewhat “high-strung” people who depend on physical perfection in order to practice their crafts and
earn their livelihoods. Any physical impairment that
threatens their ability to continue as musicians may
be greeted with dread, denial, panic, depression, or
similar responses, which may be perceived as exaggerated, especially by physicians who do not specialize in caring for performers. In the case of hearing
loss, such reactions are common even in the general
public. Consequently, it is not surprising that psychological concomitants of hearing loss in musicians are
seen in nearly all cases.
Many successful performers are communicative
and gregarious and anything that impairs their ability to interact in their usual manner can be problematic. Their vocational hearing demands are much
greater than those required in most professions, and
often musicians’ normal reactions to hearing loss
are amplified by legitimate fears about interruption

of their artistic and professional futures. The problems involved in accurately assessing the disability
associated with such impairments are addressed
below in the discussion of occupational hearing loss
in musicians.

Occupational Hearing Loss

Performing artists are required to accurately match
frequencies over a broad range, including frequencies above those required for speech comprehension.
Even mild pitch distortion (diplacusis) may make it
difficult or impossible for musicians to play or sing in
tune. Elevated high-frequency thresholds may lead
to excessively loud playing at higher pitches and to
an artistically unacceptable performance, which may
end the career of a violinist or conductor, for example.
It is extremely important for singers and other musicians to be protected from hearing loss. However, the
musical performance environment poses not only
critical hearing demands, but also noise hazards.
Review of the literature reveals convincing evidence
that music-induced hearing loss occurs, but there is
a clear need for ongoing research to clarify incidence,
predisposing factors, and methods of prevention. It
is interesting to note that, in direct contrast to many
other publications, Johnson and Sherman evaluated
60 orchestra members and 30 nonmusicians from 250
to 20 000 Hz and found no substantive differences.4
This study suggested that there is no additional risk
to hearing as a result of exposure to orchestral music.
Similarly, Schmidt and colleagues showed that the
students of Rotterdam Conservatory did not show
any decreased hearing loss when compared to a
group of medical students of the same age, despite
the music students’ exposure to music.5
As mentioned previously, noise exposure can
cause both temporary and/or permanent hearing
loss. In a study to evaluate temporary threshold shift
in performers and listeners, Axelsson and Lindgren

determined that the performers showed less of a
shift than the audience did.6 It was surmised by the
authors that this may be explained by pre-exposure
hearing levels. The performers had poorer hearing
levels than listeners did before being exposed to the
study noise. Another interesting finding was that the
male listeners showed more of a temporary shift than
the females. The authors suggested that exposure to
live pop music should be limited to 100 dBA or less
for no more than 2 hours. When symphony orchestra musicians from the Royal Danish Theater were
studied, Ostri et al found that 58% of the 95 subjects
demonstrated a hearing loss when using 20 dB HL




15.  Hearing Loss in Singers and Other Musicians

as the “normal” cutoff value. The male subjects were
more affected by noise exposure than the female subjects.7 The authors concluded that symphonic music
does indeed cause hearing loss. In 2014, Schmidt
et al8 studied the hearing levels of 182 professional
symphony orchestra musicians with varying degrees
of exposure time and intensity. For most of the musicians tested, the level of hearing loss was less than
expected based on the 1999 International Organization for Standardization’s measure for predicting
permanent threshold shifts based on duration and
intensity of noise exposure.9 Although the level of
hearing loss was generally less than predicted, they
found that the ears with the highest exposure (above
90.4 dBA and a mean exposure time of 41.7 years)

had an additional threshold shift of 6.3 dB compared
to musicians with the lowest exposure. In 1992,
McBride et al set out to determine whether noise
exposure affected the classical musician.10 Contrary
to other studies, audiograms showed no significant
differences between participants of the same sex and
age. They did prove that the musicians were exposed
to high doses of noise, which do pose an occupational
hazard. Drake-Lee studied 4 heavy-metal musicians
before and after performance.11 It was determined
that exposure does cause a temporary threshold shift
with a maximum effect at lower frequencies. An article by Bu in 1992, examined hearing loss in Chinese
opera orchestra musicians.12 Bu discovered that the
incidence of hearing loss in this group was exceedingly high and apparently associated with the types
of instruments used.12 Bu has suggested measures
to combat the noise exposure other than the use of
ear protectors, such as percussion musicians seated 1
meter lower than the other musicians in order to better preserve hearing of the instrumentalists around
them.12 However, such suggestions have a variety of
drawbacks and practical limitations.
Occupational hearing loss is usually bilateral,
fairly symmetrical, sensorineural hearing impairment
caused by exposure to high-intensity workplace noise
or music. This subject has been discussed in this text,
and specifically with regard to musicians in a previous review, in general in detail elsewhere.13 Music is
one of the professions that can produce a somewhat
asymmetrical hearing loss in selected cases.
It has been well established that selected symphony orchestra instruments, popular orchestras,
rock bands, and personal stereo headphones produce sound pressure levels (SPL) intense enough
to cause both temporary and permanent hearing

loss. Such hearing loss may also be accompanied
by tinnitus and may be severe enough to interfere
with performance, especially in violinists. The vio-

263

lin is the highest-pitched string instrument in routine use. The amount of hearing loss is related to the
intensity of the noise, duration and intermittency
of exposure, total exposure time over months and
years, and other factors. Rosanowski and Eysholdt
published a case study on a violinist with bilateral
tinnitus.14 They recorded peak sound pressure levels
of over 90 dB. The violinist showed a 20-dB drop in
hearing between 2 and 8 kHz on the side with which
the violin rests (left). This phenomenon (asymmetry)
is produced by the head shadow, the same mechanism that causes asymmetrical hearing loss in rifle
shooters. The authors point out the potential hazard
of other auditory symptoms (ie, tinnitus) as a result
of noise exposure. In their 1999 study of hearing, tinnitus, and exposure to amplified recreational noise,
Metternich and Brusis found a very high risk of tinnitus even when subjects were exposed to short durations of amplified music.15 This risk appears to be
greater than the risk of permanent hearing loss from
the same exposure.
Various methods have been devised to help protect
the hearing of performers. For example, many singers and other musicians (especially in rock bands)
wear ear protectors. They may not feel comfortable
wearing ear protection during a performance but
may take precautionary measures during practice.
Ear protectors have changed tremendously over the
years, and there are more sophisticated and suitable
models available now that cater to the musician. The

importance of using new, more appropriate ear protectors for professional musicians should be stressed,
especially because the previously unappreciated
relationship between orchestral music exposure
and noise-induced hearing loss has become clear. In
their 1983 publication, “The Hearing of Symphony
Orchestra Musicians,” Karlsson et al determined that
the criteria used to evaluate noise exposure in industry must be different from the criteria used to assess
acoustic instrument levels in symphonic music.16
This complex issue is discussed later in this chapter.
Singers need to be made more aware of the hazards
of noise exposure and find ways to avoid or reduce
its effects whenever possible. They should also be
careful to avoid exposure to potentially damaging
avocational noise such as loud music through headphones, chainsaws, snowmobiles, gunfire, motorcycles, and power tools. Hoppmann has reviewed the
hazards of being an instrumental musician, including
hearing loss, and he emphasizes the need for a team
approach to comprehensive arts-medicine diagnosis
and care.17 Noise exposure has a cumulative effect,
and exposure to these other types of noise just compounds the damage to a performer. In his article


264

Clinical Assessment of Voice

entitled, “Binaural hearing in music performance,”
Donald Woolford evaluated the effects of hearing
impairment on performance and found no direct correlation between degree of hearing impairment and
level of performance.18
Clinical observations in the authors’ practice suggest that the rock performance environment may be

another source of asymmetrical noise-induced hearing loss, a relatively unusual situation because most
occupational hearing loss is symmetrical. Rock singers and instrumentalists tend to have slightly greater
hearing loss in the ear adjacent to the drum and cymbal, or the side immediately next to a speaker, if it is
placed slightly behind the musician. Various methods have been devised to help protect the hearing
of rock musicians. For example, most of them stand
beside or behind their speakers, rather than in front
of them. In this way, they are not subjected to peak
intensities, as are the patrons in the first rows.
The problem of occupational hearing loss among
classical singers and other musicians is less obvious but equally important. In fact, in the United
States, it has become a matter of great concern and
negotiation among unions and management. Various reports have found an increased incidence of
high-frequency sensorineural hearing loss among
professional orchestra musicians as compared to the
general public, and sound levels within orchestras
have been measured between 83 and 112 dBA, as
discussed below. The size of the orchestra and the
rehearsal hall are important factors, as is the position of the individual instrumentalist within the
orchestra. Players seated immediately in front of the
brass section appear to have particular problems,
for example. Individual classical instruments may
produce more noise exposure for their players than
assumed. In their study entitled “Hearing assessment of orchestral musicians,” Kahari et al reported
that male musicians have a more pronounced highfrequency hearing loss than females exposed to the
same musical noise.19 They also noted that percussion and woodwind players demonstrated a slightly
more pronounced hearing loss than musicians of
large string instruments.
Because many singers and instrumentalists practice or perform 4 to 8 hours a day (sometimes more),
such exposure levels may be significant. An interesting review of the literature may be found in the report
of a clinical research project on hearing in classical

musicians by Axelsson and Lindgren.20 They also
found asymmetrical hearing loss in classical musicians, greater in the left ear. This is a common finding, especially among violinists. A brief summary of

most of the published works on hearing loss in musicians is presented below.
In the United States, various attempts have been
made to solve some of the problems of the orchestra musician, including placement of Plexiglas barriers in front of some of the louder brass instruments;
alteration in the orchestra formation, such as elevation of sections or rotational seating; changes in
spacing and height between players; use of specialized musicians, ear protectors; and other measures.
These solutions have not been proven effective, and
some of them appear impractical, or damaging to
the performance. The effects of the acoustic environment (concert hall, auditorium, outdoor stage, etc)
on the ability of music to damage hearing have not
been studied systematically. Recently, popular musicians have begun to recognize the importance of this
problem and to protect themselves and educate their
fans. Some performers are wearing ear protectors
regularly in rehearsal and even during performance,
as noted in the press in 1989.21 In a 5-year study of
the health of 377 professional orchestra musicians,
Ackermann et al reported that 64% of the musicians
who responded to the survey used earplugs at least
intermittently. They also highlighted the need for
hearing protection by reporting that “For average
reported practice durations (2.1 hour per day, 5 days
a week), 53% would exceed accepted permissible
daily noise exposure in solitary practice, in addition
to sound exposure during orchestral rehearsals and
performance.”22(p8) Considerable additional study is
needed to provide proper answers and clinical guidance for this very important occupational problem.
In fact, a review of the literature on occupational
hearing loss reveals that surprisingly little information is available on the entire subject. Moreover, all

of it is concerned with instrumentalists; few similar
studies in singers were found. In 2008, Hamdan et
al recorded the transient-evoked otoacoustic emissions (TEOAEs) of 23 normal hearing singers and
found them to be less robust than those of the control
group. These results suggest subtle cochlear dysfunction possibly resulting from increased noise exposure
during practice and performance. The authors propose using TEOAE measurement as a tool to identify
ears as “at risk for music-induced hearing loss.”23
Study of the existing reports reveals a variety of
approaches. Unfortunately, neither the results nor
the quality of the studies is consistent. Nevertheless,
familiarity with the research already performed provides useful insights into the problem. In 1960, Arnold
and Miskolczy-Fodor studied the hearing of 30 pianists. SPL measurements showed that average levels




15.  Hearing Loss in Singers and Other Musicians

were approximately 85 dB SPL, although periods of
92 to 96 dB SPL were recorded.24 The A-weighting
network was not used for sound level measurements
in this study. No noise-induced hearing loss was identified. The pianists in this study were 60 to 80 years of
age; and, in fact, their hearing was better than normal
for their age. Flach and Aschoff,25 and later Flach,26
found sensorineural hearing loss in 16% of 506 music
students and professional musicians, a higher percentage than could be accounted for by age alone,
although none of the cases of hearing loss occurred
in students. Hearing loss was most common in musicians playing string instruments. Flach and Aschoff
also noticed asymmetrical sensorineural hearing loss
worse on the left in 10 of 11 cases of bilateral sensorineural hearing loss in instrumentalists.25 In one case

(a flautist), the hearing was worse on the right. In
4% of the professional musicians tested, hearing loss
was felt to be causally related to musical noise exposure. Histories and physical examinations were performed on the musicians, and tests were performed
in a controlled environment. This study also included
interesting measurements of sound levels in a professional orchestra. Unfortunately, they are reported in
DIN-PHONS, rather than dBA.
In 1968, Berghoff 27 reported on the hearing of 35
big band musicians and 30 broadcasting (studio)
musicians. Most had performed for 15 to 25 years,
although the string players were older as a group and
had performed for as many as 35 years. In general,
they played approximately 5 hours per day. Hearing loss was found in 40- to 60-year-old musicians
at 8000 and 10 000 Hz. Eight musicians had substantial hearing loss, especially at 4000 Hz. Five out
of 64 (8%) cases were felt to be causally related to
noise exposure. No difference was found between
left and right ears, but hearing loss was most common in musicians who were sitting immediately
beside drums, trumpets, or bassoons. Sound level
measurements for wind instruments revealed that
intensities were greater 1 meter away from the instrument than they were at the ear canal. Unfortunately,
sound levels were measured in PHONS. Lebo and
Oliphant studied the sound levels of a symphony
orchestra and two rock-and-roll orchestras.28 They
reported that sound energy for symphony orchestras
is fairly evenly distributed from 500 to 4000 Hz, but
most of the energy in rock-and-roll music was found
between 250 and 500 Hz. The SPL for the symphony
orchestra during loud passages was approximately
90 dBA. For rock-and-roll bands, it reached levels
in excess of 110 dBA. Most of the time, during rock
music performance, sound energy was louder than

265

95 dBA in the lower frequencies; symphony orchestras rarely achieved such levels. However, Lebo and
Oliphant made their measurements from the auditorium, rather than in immediate proximity to the
performers.28 Consequently, their measurements are
more indicative of distant audience noise exposure
than that of the musicians or audience members in
the first row. In 2008, O’Brien, Wilson, and Bradley29
studied orchestral SPLs. They found the musicians
at the greatest risk of sustained noise exposure to
be the principal trumpeter, first and third hornists,
and principal trombonist. They also noted the highest peak SPLs in the percussion and timpani sections. Rintelmann and Borus studied noise-induced
hearing loss in rock-and-roll musicians, measuring
SPL at various distances from 5 to 60 ft from center
stage.30 They studied 6 different rock-and-roll groups
in 4 locations and measured a mean SPL of 105 dB.
Their analysis revealed that the acoustic spectrum
was fairly flat in the low- and mid-frequency region
and showed gradual reduction above 2000 Hz. They
also detected hearing loss in only 5% of the 42 high
school and college student rock-and-roll musicians
they studied. The authors estimated that their experimental group had been exposed to approximately
105 dB (SPL) for an average of 11.4 hours a week for
2.9 years.
In 1970, Jerger and Jerger studied temporary
threshold shifts (TTSs) in rock-and-roll musicians.31
They identified TTSs greater than 15 dB in at least one
frequency between 2000 and 8000 Hz in 8 of 9 musicians studied prior to performance and within 1 hour
after the performance. Speaks and coworkers32 examined 25 rock musicians for threshold shifts, obtaining

measures between 20 and 40 minutes following performance. In this study, shifts of only 7 to 8 dB at 4000
and 6000 Hz were identified. TTSs occurred in about
half of the musicians studied. Six of the 25 musicians
had permanent threshold shifts. Noise measurements
were also made in 10 rock bands. Speaks et al found
noise levels from 90 to 110 dBA. Most sessions were
less than 4 hours, and actual music time was generally 120 to 150 minutes. The investigators recognized
the hazard to hearing posed by this noise exposure.
In 1972, Rintelmann, Lindberg, and Smitley studied
the effects of rock-and-roll music on humans under
laboratory conditions.33 They exposed normal hearing females to rock-and-roll music at 110 dB SPL in a
sound field. They also compared subjects exposed to
music played continuously for 60 minutes with others
in which the same music was interrupted by 1 minute of ambient noise between each 3-minute musical
selection. At 4000 Hz, they detected mean TTSs of


266

Clinical Assessment of Voice

26 dB in the subjects exposed to continuous noise, and
22.5 dB in those exposed intermittently. Both groups
required approximately the same amount of time for
recovery. TTSs sufficient to be considered potentially
hazardous for hearing occurred in slightly over 50%
of the subjects exposed to intermittent noise and in
80% of subjects subjected to continuous noise.
A study by Samelli et al34 in 2012, compared different areas of the auditory pathway of professional
pop/rock musicians with that of nonmusicians. The

participants included 16 young male pop/rock musicians who had been performing for at least 5 years,
and a group of age-matched peers. Although the
researchers found damage to the peripheral auditory
system evidenced by higher pure-tone thresholds and
smaller TEOAE amplitudes, they found no damage to
the central auditory nervous system. The researchers
assessed the central auditory system using auditory
brainstem response (ABR) testing. Both groups were
within the normal range. However, the group of pop/
rock musicians had earlier neural responses to the
acoustic stimuli, suggesting better or faster processing of acoustic information. Samelli et al attribute this
to musical training providing improved processing
of acoustic information and improved spontaneous
attention to sound. While their speculation might be
correct, the findings also could be explained by mild
hyperacusis associated with noise-induced sensorineural impairment. Alternatively, it might be that
their superior hearing performance was present from
birth and predisposed them to choose careers as musicians. The findings also could be irrelevant clinically.
In 1972, Jahto and Hellmann35 studied 63 orchestra musicians playing in contemporary dance bands.
Approximately one-third of their subjects had measurable hearing loss, and 13% had bilateral highfrequency loss suggestive of noise-induced hearing
damage. They also measured peak SPL of 110 dB
(the A scale was not used). They detected potentially
damaging levels produced by trumpets, bassoons,
saxophones, and percussion instruments. In contrast,
in 1974, Buhlert and Kuhl36 found no noise-induced
hearing loss among 17 performers in a radio broadcasting orchestra. The musicians had played for an
average of 20 years and were an average of 30 years
of age. In a later study, Kuhl37 studied members of
a radio broadcasting dance orchestra over a period
of 12 days. The average noise exposure was 82 dBA.

He concluded that such symphony orchestras were
exposed to safe noise levels, in disagreement with
Jahto and Hellmann.35 Zeleny et al studied members
of a large string orchestra with intensities reaching
104 to 112 dB SPL.38 Hearing loss greater than 20 dB
in at least one frequency occurred in 85 of 118 subjects

(72%), usually in the higher frequencies. Speech frequencies were affected in 6 people (5%). Conversely,
in 2007, Reuter and Hameroshoi39 found no evidence
of TTSs or changes in otoacoustic emissions for 12
normal hearing symphony orchestra musicians, both
before and after rehearsals.
In 1976, Siroky et al reported noise levels within
a symphony orchestra ranging between 87 and 98
dBA, with a mean value of 92 dBA.40 Audiometric
evaluation of 76 members of the orchestra revealed
16 musicians with hearing loss, 13 of them sensorineural. Hearing loss was found in 7.3% of string
players, 20% of wind players, and 28% of brass players. All percussionists had some degree of hearing
loss. Hearing loss was not found in players who had
performed for fewer than 10 years but was present
in 42% of players who had performed for more than
20 years. This study needs to be reevaluated in consideration of age-matched controls. At least some of
the individuals reported have hearing loss not causally related to noise (eg, those with hearing levels of
100 dB HL in the higher frequencies). In a companion report, Folprechtova and Miksovska also found
mean sound levels of 92 dBA in a symphony orchestra with a range of 87 to 98 dBA.41 They reported that
most of the musicians performed between 4 and 8
hours daily. They reported the sound levels of various instruments as seen in Table 15–1.
A study by Balazs and Gotze, also in 1976, agreed
that classical musicians are exposed to potentially
damaging noise levels.42 The findings of Gryczynska

and Czyzewski supported the concerns raised by
other authors.43 In 1977, they found bilateral normal

Table 15–1.  Sound Levels of Various Instruments
Instrument

Sound Level (dBA)

Violin

84–103

Cello

84–92

Bass

75–83

Piccolo

95–112

Flute

85–111

Clarinet

92–103

French horn

90–106

Oboe

80–94

Trombone

85–114

Xylophone

90–92

Source:  Data from Folprechtova and Miksovska.41




15.  Hearing Loss in Singers and Other Musicians

hearing in only 16 of 51 symphony orchestra musicians who worked daily at sound levels between 85
and 108 dBA. Five of the musicians had unilateral
normal hearing; the rest had bilateral hearing loss.
In 2009, a team of researchers from the University
of Sydney began a large-scale comprehensive health

study of 377 professional orchestra musicians. The
Sound Practice Project evaluated the health of musicians from 8 professional orchestras over a 5-year
period. Although the focus was on physical and psychological impacts of the profession, they found that
the majority of subjects were exposed to noise that
exceeded daily acceptable levels.22
In 1977, Axelsson and Lindgren studied factors
increasing the risk for hearing loss in pop musicians.44 They reported that aging, length of exposure
per musical session, long exposure time in years,
military service, and listening to pop music with
headphones all had a statistically significant influence on hearing. They noted that the risk and severity
of hearing loss increase with increasing duration of
noise exposure and increasing sound levels. In pop
music, the exposure to high sound levels was felt to
be limited in time, and less damaging low frequencies predominated.
Also in 1977, Axelsson and Lindgren published an
interesting study of 83 pop musicians and noted a
significant incidence of hearing loss.45 They reanalyzed previous reports investigating a total of 160
pop musicians, which identified an incidence of only
5% hearing loss. In their 1978 study, Axelsson and
Lindgren tested 69 musicians, 4 disk jockeys, 4 managers, and 6 sound engineers.46 To have hearing loss,
a subject had to have at least 1 pure-tone threshold
exceeding 20 dB HL at any frequency between 3000
and 8000 Hz. Thirty-eight musicians were found to
have sensorineural hearing loss. In 11, only the right
ear was affected; in 5, only the left ear was affected.
Thirteen cases were excluded because their hearing
loss could be explained by causes other than noise.
Thus, 25% of the pop musicians had sensorineural hearing loss probably attributable to noise. The
most commonly impaired frequency was 6000 Hz,
and very few ears showed hearing levels worse than

35 dB HL. After correction for age and other factors,
25 (30%) had hearing loss as defined above. Eleven
(13%) had hearing loss defined as a pure-tone audiometric average greater than 20 dB HL at 3000, 4000,
6000, and 8000 kHz in at least one ear. Of these 11,
7 (8%) had unilateral hearing loss. The authors concluded that it seemed unlikely that sensorineural
hearing loss would result from popular music presented at 95 dBA with interruptions and with relatively short exposure durations and low-frequency

267

emphasis. Axelsson and Lindgren published further articles on the same study.47,48 They also noted
that TTS measurements in pop music environments
showed less shift in musicians than in the audience.
They also found that female listeners were more
resistant to TTS than males. In a follow-up study
to their 1975 work, published in 1977,44,45 Axelsson
and Eliasson re-evaluated 53 out of the original 83
pop/rock musicians they had studied.49 Their findings indicate a rather slow progression of hearing
loss. The authors were surprised to find that hearing remained so stable and fell within 20 dB of original thresholds. They surmised that this presentation
may have something to do with the stapedius reflex,
but this theory has yet to be proven. In our opinion,
this finding is a manifestation of the asymptotic pattern seen in noise-induced hearing loss (NIHL) from
other sources and is not surprising. In a 2001 followup study to a 1979 assessment done by Axelsson and
Lindgren (published in 1981),20 Kahari et al re-evaluated 56 of the original 139 orchestral musicians in
Sweden.50 Interestingly, 16 years later, those studies
showed no significant decreases in pure-tone threshold. The male participants continued to demonstrate
a greater hearing loss than the females in the highfrequency range.
In 1981, Westmore and Eversden studied a symphony orchestra and 34 of its musicians.51 They
recorded SPL for 14.4 hours. Sound levels exceeded
90 dBA for 3.51 hours and equaled or exceeded
110 dBA for 0.02 hours. In addition, there were brief

peaks exceeding 120 dBA. They interpreted their
audiometric testing as showing noise-induced hearing loss in 23 of 68 ears. Only 4 of the 23 ears had
a hearing loss greater than 20 dB HL at 4000 Hz.
There was a “clear indication” that orchestral musicians may be exposed to damaging noise. However,
because of the relatively mild severity, they speculated that, “it is unlikely that any musician is going
to be prevented from continuing his artistic career.”
In Axelsson and Lindgren’s 1981 study, sound level
measurements were performed in 2 theaters, and 139
musicians underwent hearing tests.20 Sound levels for
performances ranged from 83 to 92 dBA. Sound levels
were slightly higher in an orchestra pit, although this
is contrary to the findings of Westmore.51 Fifty-nine
musicians (43%) had pure-tone thresholds worse than
expected for their ages. French hornists, trumpeters,
trombonists, and bassoonists were found to be at
increased risk for sensorineural hearing loss. Asymmetric pure-tone thresholds were common in musicians with hearing loss and in those still classified
as having normal hearing. The left ear demonstrated
greater hearing loss than the right, especially among


268

Clinical Assessment of Voice

violinists. Axelsson and Lindgren also found that the
loudness comfort level was unusually high among
musicians. Acoustic reflexes also were elicited at comparatively high levels, being pathologically increased
in approximately 30%. TTSs were also identified,
supporting the assertion of noise-related etiology.
Also, in 1983, Lindgren and Axelsson attempted

to determine whether individual differences of TTS
existed after repeated controlled exposure to noninformative noise and to music having equal frequency,
time, and sound level characteristics.52 They studied
10 subjects who were voluntarily exposed to 10 minutes of recorded pop music on 5 occasions. On 5 other
occasions they were exposed to equivalent noise.
Four subjects showed almost equal sensitivity in
measurements of TTS, and 6 subjects showed marked
differences, specifically, greater TTS after exposure to
the nonmusic stimulus. This research suggests that
factors other than the physical characteristics of the
fatiguing sound contributed to the degree of TTS.
The authors hypothesized that these factors might
include the degree of physical fitness, stress, and
emotional attitudes toward the sounds perceived.
The authors concluded that high sound levels perceived as noxious cause greater TTS than high sound
levels that the listener perceived as enjoyable.
In 1983, Karlsson and coworkers published a
report with findings and conclusions substantially
different from those of Axelsson and others.53 Karlsson et al investigated 417 musicians, 123 of whom
were investigated twice at an interval of 6 years.
After excluding 26 musicians who had hearing loss
for reasons other than noise they based their conclusions on the remaining 392 cases. Karlsson et al
concluded that there was no statistical difference
between the hearing of symphony orchestra musicians and those of a normal population of similar
age and sex.53 Their data revealed a symmetric dip
of 20 dB HL at 6000 Hz in flautists and a 30 dB HL left
high-frequency sloping hearing loss in bass players.
Overall, a 5-dB HL difference between ears was also
found at 6000 and 8000 Hz, with the left side being
worse. Although Karlsson and coworkers concluded

that performing in a symphonic orchestra does not
involve an increased risk of hearing damage, and that
standard criteria for industrial noise exposure are not
applicable to symphonic music, their data are similar
to previous studies. Only their interpretation varies
substantially.
In 1984, Woolford studied SPLs in symphony
orchestras and hearing.54 Woolford studied 38 Australian orchestral musicians and measured SPLs
using appropriate equipment and technique. He

found potentially damaging sound levels, consistent
with previous studies. Eighteen of the 38 musicians
had hearing losses. Fourteen of those had threshold
shifts in the area of 4000 Hz, and 4 had slight losses
at low frequencies only.
Johnson et al studied the effects of instrument type
and orchestral position on the hearing of orchestra
musicians.55 They studied 60 orchestra musicians
from 24 to 64 years in age, none of whom had symptomatic hearing problems. The musicians underwent
otologic histories and examinations and pure-tone
audiometry from 250 through 20 000 Hz. Unfortunately, this study used previous data from other
authors as control data. In addition to the inherent
weakness in this design, the comparison data did not
include thresholds at 6000 Hz. There appeared to be
a 6000-Hz dip in the population studied by Johnson
et al, but no definitive statement could be made. The
authors concluded that the type of instrument played
and the position on the orchestra stage had no significant correlation with hearing loss, disagreeing with
findings of other investigators. In another paper produced from the same study,56 Johnson et al reported
no difference in the high-frequency thresholds (9000–

20 000 Hz) between musicians and nonmusicians.
Again, because he examined 60 instrumentalists, but
used previously published reports for comparison,
this study is marred. This shortcoming in experimental design is particularly important in high-frequency
testing during which calibration is particularly difficult and establishment of norms on each individual
piece of equipment is advisable.
In their 1986 article entitled, “The level of the musical loud sound and noise induced hearing impairment,” Ono et al determined the importance of
measuring the cumulative effects of noise over long
periods of time on a single individual. They developed a compact noise dosimeter designed to evaluate long-term and varied exposure called “Noise
Badge.”57
In 1987, Swanson et al studied the influence of subjective factors on TTS after exposure to music and
noise of equal energy,58 attempting to replicate Lindgren and Axelsson’s 1983 study. Swanson’s study
used 2 groups of subjects, 10 who disliked pop music,
and 10 who liked pop music. Each subject was tested
twice at 48-hour intervals. One session involved
exposure to music for 10 minutes. The other session
involved exposure to equivalent noise for 10 minutes.
Their results showed that individuals who liked pop
music experienced less TTS after music than after
noise. Those who disliked the music showed greater
TTS in music than in noise. Moreover, the group that




15.  Hearing Loss in Singers and Other Musicians

liked pop music exhibited less TTS than the group
that disliked the music. These findings support the
notion that sounds perceived as offensive produce

greater TTS than sounds perceived as enjoyable.
A particularly interesting review of hearing impairment among orchestra musicians was published by
Woolford et al in 1988.59 Although this report pre­
sents only preliminary data, the authors have put
forward a penetrating review of the problem and
interesting proposals regarding solutions, including
an international comparative study. They concluded
that among classical musicians the presence of hearing loss from various etiologies including noise has
been established, that some noise-induced hearing
impairments in musicians are permanent (although
usually slight), and that efforts to reduce the intensity
of noise exposure can be successful.
In addition to concern about hearing loss among
performers, in recent years there has been growing
concern about noise-induced hearing loss in audiences. Those at risk include not only people at rock
concerts, but also people who enjoy music through
stereo systems, especially modern personal headphones. Concern about hearing loss from this source
in high school students has appeared to the lay press
and elsewhere.60,61 Because young music lovers are
potentially performers, in addition to other reasons,
this hazard should be taken seriously and investigated further.
In 1990, West and Evans studied 60 people aged
15 to 23 years at the University of Keele, looking for
hearing loss caused by listening to amplified music.62
They found widening of auditory bandwidths to be
a sensitive, early indicator of noise-induced hearing
loss that was detectable before threshold shift at 4000
or 6000 Hz occurred. They advocated the use of frequency resolution testing and high-resolution Békésy
audiometry for early detection of hearing impairment. West and Evans found that subjects extensively
exposed to loud music were significantly less able to

differentiate between a tone and its close neighbors.
Reduced pitch discrimination was particularly common in subjects who had experienced TTS or tinnitus
following exposure to amplified music.
In 1991, van Hees published an extensive thesis
on noise-induced hearing impairment in orchestral
musicians.63 He agreed that noise levels were potentially damaging in classical and wind orchestras.
Unlike other researchers, van Hees found it more
useful to classify the instruments by orchestral zone,
rather than by instrument or instrument group.
However, he found a much greater incidence of hearing loss among both symphony and wind orchestra

269

musicians than was reported in previous literature.
In contrast to previous investigators, he also did not
find evidence of asymmetric hearing loss in violinists
and cello players.
In 1991, the musicians of the Chicago Symphony
Orchestra were evaluated by Royster, Royster, and
Killion. Subjects were given individual dosimeters,
and 68 measurements were made during various
rehearsals and performances. More than half of the
musicians had audiograms consistent with NIHL and
showed a high prevalence of “notch” patterns.64 The
authors measured Leq values and hearing threshold levels on 32 musicians and found that pure-tone
thresholds between 3 and 6 kHz correlated to the
measured equivalent sound level (Leq), which was
determined to range from 79 to 99 dBA SPL.63
In their 1998 evaluation of choir singers and hearing loss, Steurer and colleagues made some unusual
discoveries that warrant additional research. The

authors discovered that low-frequency hearing, in
particular 250 Hz to 1 kHz, was most affected in
these subjects.65 They were not able to explain the
demonstrated hearing losses below 100 Hz but have
speculated that there may be increased endolymphatic pressure when singing that could account for
the loss in this range.
Reviewing these somewhat confusing and contradictory studies reveals that a great deal of important work remains to be done to establish the risk
of hearing loss among various types of musicians,
the level and pattern of hearing loss that may be sustained, practical methods of preventing hearing loss,
and advisable programs for monitoring and early
diagnosis. However, a few preliminary conclusions
can be drawn. First, the preponderance of evidence
indicates that noise-induced hearing loss occurs in
both pop and classical musicians and is causally
related to exposure to loud music. Second, in most
instances, especially among classical musicians, the
hearing loss is not severe enough to interfere with
speech perception. Third, the effects of mild highfrequency hearing loss on musical performance have
not been established. Fourth, it should be possible to
devise methods to conserve hearing in performing
artists without interfering with their performance.
In 1991, Chasin and Chong reported on an ear protection program for musicians.66 They provide an
interesting discussion of the use of ear protectors in
musicians, although several aspects of their paper
are open to challenge. In particular, their assertion
that some vocalists (particularly sopranos) have selfinduced hearing loss caused by singing has not been
substantiated.


270

Clinical Assessment of Voice

Legal Aspects of Hearing Loss
in Singers and Musicians
The problem of hearing loss in musicians raises
numerous legal issues, especially the implications of
occupational hearing loss, and hearing has become
an issue in some orchestra contracts. Traditionally,
workers’ compensation legislation has been based on
the theory that workers should be compensated when
a work-related injury impairs their ability to earn a
living. Ordinarily, occupational hearing loss does not
impair earning power (except possibly in the case of
musicians and a few others). Consequently, current
occupational hearing loss legislation broke new legal
ground by providing compensation for interference
with quality of life — that is, loss of living power.
Therefore, all current standards for defining and
compensating occupational hearing loss are based on
the communication needs of the average speaker, and
losses are usually compensated in accordance with
the recommendations of the American Academy of
Otolaryngology.13 Because music-induced hearing
loss appears to rarely affect the speech frequencies,
it is not compensable under most laws. However,
although a hearing loss at 3000, 4000, or 6000 Hz with
preservation of lower frequencies may not pose a
problem for a boilermaker, it may be a serious problem for a violinist. Under certain circumstances, such
a hearing loss may even be disabling. Because professional instrumentalists require considerably greater

hearing acuity throughout a larger frequency range,
we must investigate whether the kinds of hearing
loss caused by music are severe enough to impair
performance. If so, new criteria must be established
for compensation for disabling hearing impairment
in musicians, in keeping with the original intent of
the workers’ compensation law.
There may also be unresolved legal issues regarding hearing loss not caused by noise in professional
musicians. Like people with other disabilities, numerous federal laws protect the rights of those with
hearing impairment. In the unhappy situation in
which an orchestra must release a hearing-impaired
violinist who can no longer play in tune, for example,
legal challenges may arise. In such instances, and in
many other circumstances, an objective assessment
process is in the best interest of performers and management. Objective measures of performance are
already being used in selected areas for singers, and
they have proven beneficial in helping the performer
assess dispassionately certain aspects of performance
quality and skill development. Such technologic
advances will probably be used more frequently

in the future to supplement traditional subjective
assessment of performing artists for musical, scientific, and legal reasons.

Ear Protectors for the Musician
Current ear protectors offer a much more suitable
solution to noise protection than their predecessors.
There are several models available that cater to musicians and their specific requirements. The design
strategy has improved to allow more accurate music
and speech perception at lower intensity levels. Various models provide differing attenuation levels ranging from 9 to 25 decibels. These protectors are custom

fitted to the individual and thus provide better performance than preformed in-the-ear hearing protectors. They can be purchased through an audiologist
or hearing aid dispenser.

Noise Exposure and the Audience
The focus of this chapter thus far has been on the
musician and noise exposure. A few more recent
studies have looked at the effects of noise on an audience. In particular, Gunderson, Moline, and Catalano, in a 1997 publication, evaluated the effect of
noise exposure on employees of urban music clubs.
Average sound levels ranged from 94.9 to 106.7 dBA.
Only 16% of the employees used ear protection.67 The
authors recommended the development of a hearing conservation program for this often overlooked
population. Similarly, in 1998, Sataloff, Hickey, and
Robb evaluated noise levels at an outdoor rock concert. Results showed audience exposure to levels of
119 dBA. These studies indicate that future efforts
need to focus on hearing conservation not only for
the performer, but for audiences as well.68

Treatment of Occupational Hearing
Loss in Singers and Other Musicians
For a complete discussion of the treatment of hearing loss, the reader is referred to other sources1 and
to standard otolaryngology texts. Most cases of sensorineural hearing loss produced by aging, hereditary factors, and noise cannot be cured. When they
involve the speech frequencies, modern, properly
adjusted hearing aids are usually extremely helpful.
However, these devices are rarely satisfactory for
musicians during performance. More often, appro-




15.  Hearing Loss in Singers and Other Musicians

priate counseling is sufficient. The musician should
be provided with a copy of his or her audiogram and
an explanation of its correspondence with the piano
keyboard. Unless a hearing loss becomes severe,
this information usually permits musicians to make
appropriate adjustments. For example, a conductor
with an unknown high-frequency hearing loss will
call for violins and triangles to be excessively loud. If
he or she knows the pattern of hearing loss, this error
may be reduced. Musicians with or without hearing
loss should routinely be cautioned against avocational loud noise exposure without ear protection
(hunting, power tools, motorcycles, etc) and ototoxic
drugs. In addition, they should be educated about
the importance of immediate evaluation if a sudden
hearing change occurs. When diplacusis (pitch distortion) is present, compensation is especially difficult,
especially for singers and string players. Auditory
retraining may be helpful in some cases. Hopefully,
electronic devices will be available in the future to
help this problem, as well.

Conclusion
Good hearing is of great importance to musicians, but
the effects on performance of mild high-frequency
hearing loss remain uncertain. It is most important
to be alert for hearing loss from all causes in performers, to recognize it early, and to treat it or prevent its
progression whenever possible. Musical instruments
and performance environments are capable of producing damaging noise. Strenuous efforts must be
made to define the risks and nature of music-induced
hearing loss in musicians, to establish damage-risk

criteria, and to implement practical means of noise
reduction and hearing conservation.
Singers depend on their hearing almost as much
as they do on their voices. It is important not to take
such valuable and delicate structures as our ears for
granted. Like the voice, the ear must be understood
and protected if a singer is to enjoy a long, happy, and
successful career.

References
1. Sataloff J, Sataloff RT. Hearing Loss. 3rd ed. New York,
NY: Marcel Dekker; 1993.
2. Sataloff RT, Sataloff J. Occupational Hearing Loss. 2nd ed.
New York, NY: Marcel Dekker; 1993.
3. Konigsmark BW, Gorlin RJ. Genetic and Metabolic Deafness. Philadelphia, PA: WB Saunders; 1976.

271

4.Johnson DW, Sherman RE, Aldridge J, Lorraine A.
Extended high frequency hearing sensitivity: a normative threshold study in musicians. Ann Otol Rhinol Laryngol. 1986;95(2, pt 1):196–202.
5. Schmidt JM, Verschuure J, Brocaar MP. Hearing loss
in students at a conservatory. Audiology. 1994;33(4):​
185–194.
6. Axelsson A, Lindgren F. Temporary threshold shift after
exposure to pop music. Scand Audiol. 1978;7(3):127–135.
7. Ostri B, Eller N, Dahlin E, Skylv G. Hearing impairment in orchestral musicians. Scand Audiol. 1989;18(4):​
243–249.
8. Schmidt, JH, Pedersen, ER, Paarup, HM, et al. Hearing loss in relation to sound exposure of professional
symphony orchestra musicians. Ear Hear. 2014;35(4):​
448–460.

9. International Organization for Standardization. 1999.
ISO/TC 43, Acoustics.
10. McBride D, Gill F, Proops D, et al. Noise and the classical musician. BMJ. 1992;305(6868):1561–1563.
11. Drake-Lee AB. Beyond music: auditory temporary
threshold shift in rock musicians after a heavy metal
concert. J R Soc Med. 1992;85(10):617–619.
12. Bu G. [Prevention of noise induced hearing loss in
musicians of Chinese opera.] Zhonghua Er Bi Yan Hou
Ke Za Zhi. 1992;27(2):73–75, 124.
13. Cocchiarella L, Andersson GBJ. Guides to the Evaluation of Permanent Impairment. 5th ed. Chicago, IL: AMA
Press; 2001:245–275.
14. Rosanowski F, Eysholdt U. [In situ sound pressure
measurement in a professional violinist with bilateral
tinnitus. Laryngorhinootologie. 1996;75(9):514–516.
15. Metternich FU, Brusis T. [Acute hearing loss and tinnitus caused by amplified recreational music. Laryngorhinootologie. 1999;78(11):614–619.
16. Karlsson K, Lundquist PG, Olaussen T. The hearing
of symphony orchestra musicians. Scand Audiol. 1983;​
12(4):257–264.
17.Hoppmann RA. Instrumental musicians’ hazards.
Occup Med. 2001;6(4):619–631.
18. Woolford D. Binaural hearing in music performance.
Acoustics Australia.1998;26(2):68.
19. Kahari KR, Axelsson A, Hellstrom PA, Zachau G. Hearing assessment of classical orchestral musicians. Scand
Audiol. 2001;30(1):13–23.
20. Axelsson A, Lindgren F. Hearing in classical musicians.
Acta Otolaryngol Suppl. 1981;(377):3–74.
21. Time. September 29, 1989:78.
22.Ackermann BJ, Kenny DT, O’Brien I, Driscoll TR.
Sound Practice — improving occupational health and
safety for professional orchestral musicians in Australia. Front Psychol. 2014;5:973.

23. Hamdan A, Abouchacra K, Zeki A, Zatoun G. Transient-evoked otoacoustic emissions in a group of professional singers who have normal pure-tone hearing
thresholds. Ear and Hearing. 2008;29(3):360–377.
24. Arnold GE, Miskolczy-Fodor F. Pure-tone thresholds of
professional pianists. Arch Otolaryngol. 1960;71:938–947.


272

Clinical Assessment of Voice

25. Flach M, Aschoff E. [On the question of occupationally
related hearing difficulties in musicians.] Z Laryngol
Rhinol Otol. 1966;45:595–605.
26. Flach M. [Hearing musicians from the otologic viewpoint.] Monatsschr Ohrenheilkd Laryngorhingol. 1972;9:​
424–432.
27. Berghoff F. Höbrleistung und berufsbedingte Hörsch­
ädigung des Orchestermusikers mit einem Beitrag zur
Pathophysiologie des Larmtraumatischen Hörscha­
dens. 1968, Dissertation. Cited in Axelsson A, Lindgren F. Hearing in classical musicians. Acta Otolaryngol
Suppl. 1981;(377):3–74.
28. Lebo CP, Oliphant KP. Music as a source of acoustic
trauma. Laryngoscope. 1968;72(2):1211–1218.
29. O’Brien I, Wilson W, Bradley A. Nature of orchestral
noise. J Acoustic Soc Am. 2008;124(2):926–939.
30. Rintelmann WF, Borus JF. Noise-induced hearing loss
and rock and roll music. Arch Otolaryngol. 1968;88:​
377–385.
31. Jerger J, Jerger S. Temporary threshold shift in rockand-roll musicians. J Speech Hear Res. 1970;13:221–224.
32. Speaks C, Nelson D, Ward WD. Hearing loss in rockand-roll musicians. J Occup Med. 1970;12:216–219.
33. Rintelmann WF, Lindberg RF, Smitley EK. Temporary

threshold shift and recovery patterns from two types
of rock and roll music presentations. J Acoust Soc Am.
1972;​51:1249–1255.
34. Samelli A, Matas C, Carvallo R, et al. Audiological and
electrophysiological assessment of professional pop/
rock musicians. Noise Health. 2012;14(56):6–12.
35. Jahto K, Hellmann H. [The problem of acoustic trauma
in orchestra musicians.]. HNO. 1972;20(1):21–29.
36. Buhlert P, Kuhl W. Höruntersuchungen im freien Schallfeld zum Altershörverlust. Acustica. 1974;31:168–177.
37. Kuhl W. Keine Gehörschädigung durch Tanzmusik,
simfonische Musik und Maschinengeräusche beim
Rundfunk. Kampf dem Larm. 1976;23(4):105–107.
38. Zeleny M, Navratilova Z, Kamycek Z, et al. [Relation of
hearing disorders to the acoustic composition of working environment of musicians in a wind orchestra.]
Cesk Otolaryngol. 1975;24(5):295–299.
39. Reuter K, Hammershoi D. Distortion product optoacoustic emission of symphony orchestra musicians
before and after rehearsal. J Acoustic Soc Am. 2007;121:​
327–336.
40. Siroky J, Sevcikova L, Folprechtova A, et al. Audiological examination of musicians of a symphonic orchestra
in relation to acoustic conditions. [Czech]. Cesk Otolaryngol. 1976;25(5):288–294.
41. Folprechtova A, Miksovska O. [The acoustic conditions
in a symphony orchestra.] Pracov Lek. 1978;28:1–2.
42. Balazs B, Gotze A. [Comparative examinations between
the hearing of musicians playing on traditional instruments and on those with electrical amplifications.]
[Czech]. Ful-orr-gegegyogyaszat. 1976;22:116–118.
43. Gryczynska D, Czyzewski I. [Damaging effect of music
on the hearing organ in musicians.] Otolaryngol Pol.
1977;31(5):527–532.

44. Axelsson A, Lindgren F. Factors increasing the risk for

hearing loss in “pop” musicians. Scand Audiol. 1977;6:​
127–131.
45. Axelsson A, Lindgren F. Does pop music cause hearing
damage? Audiology. 1977;16:432–437.
46. Axelsson A, Lindgren F. Hearing in pop musicians.
Acta Otolaryngol. 1978;85:225–231.
47.Axelsson A, Lindgren F. Horseln hos popmusiker.
Lakartidningen. 1978;75(13):1286–1288.
48. Axelsson A, Lindgren F. Pop music and hearing. Ear
Hear. 1981;2(2):64–69.
49. Axelsson A, Eliasson A, Israelsson B. Hearing in pop/
rock musicians: a follow-up study. Ear Hear. 1995;16(3):​
245–253.
50. Kahari KR, Axelsson A, Hellstrom PA, Zachau G. Hearing development in classical orchestral musicians: a
follow-up study. Scand Audiol. 2001;30(3):141–149.
51. Westmore GA, Eversden ID. Noise-induced hearing
loss and orchestral musicians. Arch Otolaryngol. 1981;​
107(12):761–764.
52. Lindgren F, Axelsson A. Temporary threshold shift
after exposure to noise and music of equal energy. Ear
Hear. 1983;4(4):197–201.
53. Karlsson K, Lundquist PG, Olaussen T. The hearing of
symphony orchestra musicians. Scand Audiol. 1983;​12:​
257–264.
54.Woolford DH. Sound pressure levels in symphony
orchestras and hearing. Preprint 2104 (B-1), Australian
Regional Convention of the Audio Engineering Society,
September 25–27, 1984; Melbourne, Australia.
55.Johnson DW, Sherman RE, Aldridge J, Lorraine A.
Effects of instrument type and orchestral position on

hearing sensitivity for 0.25 to 20 kHZ in the orchestral
musician. Scand Audiol. 1985;14:215–221.
56. Johnson DW, Sherman RE, Aldridge J, et al. Extended
high frequency hearing sensitivity. A normative threshold study in musicians. Ann Otol Rhinol Laryngol. 1986;​
95:196–202.
57. Ono H, Deguchi T, Ino T, et al. [The level of the musical
loud sound and noise induced hearing impairment.] J
UOEH. 1986;8(suppl):151–161.
58. Swanson SJ, Dengerink HA, Kondrick P, Miller CL. The
influence of subjective factors on temporary threshold shifts after exposure to music and noise of equal
energy. Ear Hear. 1987;8(5):288–291.
59.Woolford DH, Carterette EC, Morgan DE. Hearing
impairment among orchestral musicians. Music Percept. 1988;5(3):261–284.
60. Gallagher G. Hot music, high noise, and hurt ears. Hear
J. 1989;42(3):7–11.
61. Lewis DA. A hearing conservation program for highschool level students. Hear J. 1989;42(3):19–24.
62. West PD, Evans EF. Early detection of hearing damage
in young listeners resulting from exposure to amplified
music. Br J Audiol. 1990;24:89–103.
63.van Hees OS. Noise Induced Hearing Impairment in
Orchestral Musicians. Amsterdam, Holland: University
of Amsterdam Press; 1991.




15.  Hearing Loss in Singers and Other Musicians

64. Royster JD, Royster LH, Killion MC. Sound exposures
and hearing thresholds of symphony orchestra musicians. J Acoust Soc Am. 1991;89(6):2793–2803.

65. Steurer M, Simak S, Denk DM, Kautzky M. Does choir
singing cause noise-induced hearing loss? Audiology.
1998;37(1):38–51.
66. Chasin M, Chong J. An in situ ear protection program
for musicians. Hearing Instrument. 1991;42(12):26–28.

273

67. Gunderson E, Moline J, Catalano P. Risks of developing noise-induced hearing loss in employees of urban
music clubs. Am J Ind Med. 1997;31(1):75–79.
68. Sataloff RT, Hickey K, Robb J. Rock concert audience
noise exposure: a preliminary study. J Occupat Hear
Loss. 1998;1(2):97–99.



16
Endocrine Function
Timothy D. Anderson, Dawn D. Anderson, and
Robert Thayer Sataloff

Endocrine problems are worthy of special attention.
The human voice is extremely sensitive to endocrinologic changes, with many endocrine changes occurring in the voice throughout the normal human life
cycle, in addition to the potential for derangements
of these complex systems at any time. The laryngologist caring for voice patients must be familiar
with the broad range of normal changes, as well
as the potential for hormonal imbalances, that may
affect the voice in order to recognize them promptly
and generate appropriate treatment and referrals.
Although accurate diagnoses of problems involving the endocrine system often can be made by the

otolaryngologist through a comprehensive history,
physical examination, and appropriate laboratory
investigations, the value of a good endocrinologist
interested in consulting in the care of professional
voice users cannot be overestimated. As in other
areas of professional voice care, recognizing abnormalities and prescribing therapy can be challenging.
Otolaryngologists are encouraged strongly to enlist
the services of an endocrinologist who is interested
in arts-medicine and to assist in his or her education
in the special problems of professional voice users.
Unfortunately, the more we learn about endocrine
disorders, the more complex these systems appear,
and hormonal effects on the voice are poorly studied
and understood.

Sex Hormones
Voice changes associated with sex hormones are
encountered commonly in clinical practice and have
been investigated more than have the voice effects of

most other hormonal changes. The status of the voice
as a secondary sexual characteristic is well established, and through the work of doctors Jean and
Beatrice Abitbol and others, the status of the larynx
as a sex-hormone responsive organ has been emphasized.1–4 In an elegant demonstration, the Abitbols
showed that superficial laryngeal and vaginal smears
both exhibited significant changes throughout a
woman’s normal menstrual cycle. Surprisingly, the
smears from cervix and larynx were indistinguishable at each phase of the cycle. This work has been
supported by localization of estrogen, progesterone,
and androgen receptors in both the mucosa and

deeper tissues of the larynx,4–11 although other studies have failed to find these receptors and propose
other mechanisms by which the larynx is affected by
hormonal changes, such as differential expression of
various growth factors.12–14
The voice changes in response to changing sex hormones throughout life. The initial and most dramatic
changes occur in both males and females at puberty.
In females, cyclic voice and laryngeal changes occur
with each menstrual cycle, and a second permanent
change occurs at menopause. Males undergo a more
dramatic initial pubertal voice change, but then have
relatively stable circulating levels of sex hormones
across their life span and undergo fewer subsequent
voice changes.
Males
Puberty is the process of sexual development that
lasts between 2 and 5 years and normally begins at
age 12 to 17. It is still not known what triggers the
onset of puberty. In males, it seems as if the pituitary
275


276

Clinical Assessment of Voice

gland becomes less sensitive to the suppressive effects
of testosterone on the release of gonadotrophicreleasing hormone. This causes escalating levels of
circulating testosterone that reach adult levels by the
end of puberty.15 These high testosterone levels are
maintained until senescence, when they gradually

drop. The initial physical sign of puberty is enlargement of the testes. This is followed by rapid growth,
increases in muscle mass especially in the chest and
shoulder girdle, and development of male secondary
sex characteristics. The development of hair in the
axilla, face, and groin is initiated by adrenal androgens and then facilitated by the presence of high levels of testosterone and dihydrotestosterone (DHT).
Although DHT was thought to be the primary active
form of testosterone, it is now clear that both testosterone and DHT interact in a complex manner, often
binding to the same receptor complex but mediating
different cellular effects through poorly understood
mechanisms.16,17 Disorders of normal development
occur with any disturbances in the production,
receptors or effects of testosterone, DHT, or adrenal androgens. Although no pathologic process has
been ascribed to high levels of natural testosterone,
exogenous administration of pharmacologic doses
of sex steroids has significant effects on both males
and females. Anabolic steroids can cause testicular
atrophy, acne, voice changes, hirsutism, psychologic
changes, and severe heart problems. Because these
drugs do have the potential to increase lean body
mass and athletic performance in both men18–20 and
women,21 they are widely available for illicit use by
both professional and recreational atheletes.22 Use
of exogenous androgens has been banned in every
major competitive sport and should be strongly discouraged, especially in professional voice users. In
both males and females, these drugs may cause permanent lowering and coarsening of the voice.
Puberty’s effect on the male larynx is to increase
the size and mass of the intrinsic cartilages of the larynx, with formation of a prominent Adam’s apple.
The muscles and ligaments of the larynx also become
bulkier and change shape contributing to the drop in
fundamental frequency of the voice.23 These changes

occur at variable rates and require constant retuning of the delicate voice production mechanisms
until pubertal development has ceased. While this
retraining occurs, voice breaks and uncontrolled
pitch changes can cause considerable embarrassment
to the affected adolescent male.
The most obvious derangement of the normal
pubertal voice changes is exemplified by castrati.
When castrato singers were in vogue, castration of
males at about age 7 or 8 resulted in failure of puber-

tal laryngeal development due to lack of the normal
amounts of testosterone. The lack of circulating testosterone led to delayed closure of bony growth centers and tall stature with large lung capacities. The
combination of immature larynges and exceptional
power produced voices in the alto or soprano range
that boasted a unique quality of sound that cannot be
replicated in any other way.24 The use of castrati in
religious music continued from the 16th century until
1903, when it was officially banned by Pope Pius X.25
The presence of high levels of androgens has been
hypothesized to be causally related to the increased
incidence of coronary heart disease and atherosclerosis in men as compared to women. This suggests
that there may have been at least some advantage to
being a castrato. However, in an interesting study,
Nieschlag and coworkers compared the life span
of 50 famous castrati born between 1581 and 1858
with 50 intact, equally famous male singers born
during the same period and found no trend toward
increased longevity in castrati.26 This may point to
the cardioprotective effects of estrogen instead of
adverse cardiovascular effects of androgens.

Failure of male voice change at puberty is uncommon. Medical causes of hormonal deficiencies include
Kallmann syndrome, cryptorchidism, delayed sexual
development, Klinefelter syndrome, and Fröhlich
syndrome.27 In these cases, the persistently highpitched voice may be the complaint that brings the
patient to medical attention. In some ways, these disorders can mimic castration. For example, patients
with Klinefelter syndrome become tall at puberty and
have long legs and gynecomastia. They have small
testes that do not produce sperm. In patients in whom
the levels of circulating testosterone are very low,
pubic hair is absent, and the voice remains soprano.
Voice changes may be less dramatic in patients with
Klinefelter syndrome in whom more testosterone is
produced. Patients with low circulating androgen
levels can achieve a normal male fundamental frequency with the administration of exogenous testosterone.28 Professional singers with hypoandrogenism
should be warned that the posttherapy singing voice
may be substantially altered, and may no longer be
perceived as a “professional-level” voice.29 It should
be recognized that medical or surgical castration is
still occasionally required therapeutically in adult
males, particularly in the treatment of prostate or testicular cancer. In such cases, the mild increase in fundamental frequency of the speaking voice that occurs
physiologically with aging may be accelerated, but
survival considerations must take precedence over
vocal concerns. Rarely, the larynx may fail to respond
to hormonal changes at the time of puberty, and an




16.  Endocrine Function

infantile larynx and soprano voice may be seen in a
male with normal secondary sex characteristics and
testosterone levels. It is hypothesized that there is an
androgen receptor abnormality confined to the larynx in these cases, and exogenous androgen administration has not resulted in further voice change in
these patients. After puberty, sex-hormone-related
problems are encountered uncommonly in men.
Although there appears to be a correlation between
sex hormone levels and depths of male voices (higher
testosterone and lower estradiol levels in basses than
in tenors),30,31 the most important hormonal considerations in males occur during puberty.23 The most
common hormonal problems encountered in postpubertal males are probably those related to ingestion of
anabolic steroids, as discussed earlier. Use of testosterone in middle-aged and elderly men has become
much more widespread since the development of
topically administered testosterone. The diagnosis
and treatment of “low-T” is currently under scrutiny,
and the US Food and Drug Administration (FDA) has
recently pointed out that slow decline in levels of
testosterone in older men are normal, and has urged
that health-care providers only use testosterone in
those patients with recognized disorders of the testicles or pituitary due to the risk of adverse effects,
especially heart attack and stroke.32 Interestingly, the
voice change in elderly men seems to be related more
strongly to estrogen than testosterone levels, so testosterone supplementation of male singers may be
ineffective in avoiding senescent voice changes.33
Females
The female professional voice user must weather
multiple changes in the milieu of her sex hormones
during her life. At puberty, estrogen and progesterone levels rise, secondary sex characteristics develop,
and the menstrual cycle is established. Throughout
a woman’s reproductive life, the female voice professional has cyclic changes in the relative serum

concentration of estrogen and progesterone that
may cause important laryngeal changes. The rapid
physical and physiologic changes that occur during pregnancy also have the potential to affect the
professional voice user. Voice changes of pregnancy
may be similar to those encountered premenstrually
or mimic those produced by exogenous androgens.
They are occasionally perceived as desirable by the
patient. In some cases, alterations produced by pregnancy are permanent.32,33 After the climacteric, serum
levels of estrogen and progesterone fall while testosterone levels remain relatively stable, once again
affecting the larynx.

277

Androgenic medications should be avoided in
female singers, if there are any reasonable therapeutic alternatives. Clinically, these drugs are now used
most commonly to treat endometriosis or (illicitly) to
enhance athletic performance.21,34 Exogenous androgens cause unsteadiness of the female voice, rapid
changes of timbre, and lowering of fundamental
voice frequency.34–40 These changes often are irreversible, and can occur within weeks of initiation of
androgens. It should be stressed that these changes
can occur even with “bio-identical” hormones, which
many patients regard as more natural and therefore
more safe. It also appears that the timing of androgen
exposure is important, with early childhood androgen exposure from adrenocortical tumors causing
only rare vocal virilization.41 In the past, voices with
androgenic damage have been considered “ruined.”
In our experience, voices are altered permanently, but
not necessarily ruined. Through a slow, meticulous
retraining process, it has been possible in some cases
to return singers with androgenic voice changes to

a professional singing career. However, their “new
voice” has fewer high notes and fuller low notes than
before androgen exposure.
Puberty
The trigger causing the onset of puberty in females
is not known. Gradual increases in estrogen and progesterone are seen with eventual establishment of a
normal menstrual cycle with cyclic hormonal changes
that can affect the voice.42,43 Secondary sexual maturation is dependent on several factors besides normal
hormonal changes. Normal menstrual cycles require
adequate energy stores and caloric intake.44 This is
particularly important in dancers, gymnasts, and
other athletes who maintain a very low body weight
and body fat and may experience delayed menarche
or secondary amenorrhea. Secondary amenorrhea
also has been observed in normal weight athletes
with inadequate caloric intakes. In some women,
once training is reduced, normal cycling can resume,
although sequelae of delayed menarche or secondary amenorrhea can be observed throughout the
woman’s life. Because maximal bone mass is accumulated by age 25 to 30, osteoporosis can be particularly severe if sexual maturation is delayed.
Because initial pituitary hormone secretion occurs
at night, adequate sleep also is essential in adolescents to initiate puberty. Once sexual maturity is
attained, gonadotropin secretion is independent of
sleep. Finally, optic exposure to sunlight is essential
for normal timing of sexual development. Blind adolescents have delayed menarche. Decreased exposure


278

Clinical Assessment of Voice

to sunlight is thought to be capable of delaying development in humans, although experimental evidence
is lacking. In females, although estrogen and progesterone are the primary sex hormones, adrenal and
ovarian androgens and other pituitary hormones also
play an essential role (Table 16–1). The female voice
undergoes changes during puberty with a decrease
in fundamental frequency of about one-third of an
octave.45 Changes in the resonance properties of the
upper airways contribute to the change from a distinctively childlike voice to an adult female voice.
Because the pubertal changes in the vocal instrument
are not as large or rapid as those in the male, obvious
voice breaks and register changes are unusual. Nonetheless, maintenance of a professional singing voice
during these changes can be difficult and requires
constant adjustments in technique and expectations
to avoid the development of maladaptive behaviors.
Disorders of puberty are rare, but they occur more
commonly in females than males. Voice complaints
are not a common feature of these disorders in
females with the exception of processes with a relative excess of androgens. Abnormalities of pubertal
development can be divided into delayed maturation
and precocious puberty. As the initial sign of puberty
should be formation of breast buds, delayed sexual
maturation is identified by failure of breast budding
by age 13 or sexual hair growth that precedes breast
budding by more than 6 months. The differential
diagnosis for delayed puberty includes premature
ovarian failure (as seen in Turner syndrome), inadequate gonadotropin-releasing hormones (GnRH)
secretion from the hypothalamus (as in Kallmann
syndrome), inadequate gonadotropin (luteinizing
hormone [LH] and/or follicle-stimulating hormone
[FSH]) secretion, and inadequate body fat (anorexia

nervosa or excessive exercise). Theoretically, heavy
marijuana use could delay puberty as marijuana
blocks the release of GnRH from the hypothalamus,

preventing normal gonadotropin secretion,44 but no
human cases have been reported.
Precocious puberty is defined as the onset of breast
budding before age 8. There are many potential
causes of precocious puberty. Physicians caring for
the voice are most likely to see precocious puberty
as a result of androgen excess, as these conditions
carry the risk of irreversible deepening and coarsening of the voice at an early age. One representative disorder is congenital adrenal hyperplasia as a
result of an incomplete 21-hydroxylase deficiency.
21-Hydroxylase is an enzyme that converts progesterone to desoxycorticosterone during the synthesis
of cortisol in the adrenal gland. Minor deficiencies in
this enzyme lead to accumulation of adrenal androgens, which produce premature adrenarche. In addition to the premature development of pubic and
axillary hair, prolonged exposure to adrenal androgens can cause virilization of the voice. With early
diagnosis and continuous adequate treatment, vocal
virilization can be avoided.46,47
Females can suffer significant vocal problems
from ingesting anabolic steroids. Marked virilization
of the voice is common and is usually irreversible.
Physicians must be familiar with these side effects,
because virilizing agents are not only used by women
bodybuilders,34 but are also prescribed for postmenopausal sexual dysfunction and other problems. Such
treatments may substantially masculinize and “age”
a voice and may end a vocal career. Patients with
virilization have higher baseline scores on the voice
handicap index, and complain because they are frequently mistaken for a man on the telephone.47 Risks
must be weighed carefully against benefits before

such medication is prescribed, and the voice carefully monitored in voice professionals during therapy. Female-to-male transsexuals are a special case
where virilizing effects of androgens on the voice are
desirable. Cases of female-to male-transsexual pro-

Table 16–1.  Sequence of Female Puberty24
Sign
Breast budding
Adult breast development
Sexual hair growth (initiation)
Adult sexual hair pattern
Growth spurt
Menarche

Age (years)

Hormone

10–11

Estradiol

12.5–15

Progesterone

10.5–11.5

Androgens

13.5–16

Androgens

11–12
11.5–13

Growth hormone
Estradiol




16.  Endocrine Function

fessional voice users continuing to successfully work
have been reported.48
Menstrual Cycle
Probably the most common hormonal voice complaints in female voice professionals are related to
the normal hormonal fluxes encountered during the
menstrual cycle. Although voice changes associated
with the normal menstrual cycle may be difficult to
quantify,49–52 there is no question that they occur.
Most of the ill effects are seen in the immediate premenstrual period and are known as laryngopathia premenstrualis. This condition is common and is caused
by physiologic, anatomic, and psychologic alterations secondary to endocrine changes. The subjective symptoms are characterized by decreased vocal
efficiency, loss of the highest notes in the voice, vocal
fatigue, slight hoarseness, and some muffling of the

279

voice; it is often more apparent to the singer than to

the listener. Submucosal hemorrhages in the larynx
are more common in the premenstrual period.52,53
Singers used to be excused from singing in European
opera houses during premenstrual and early menstrual days (called grace days). This practice is not
followed in the United States and is no longer practiced widely in Europe. Voice dysfunction similar to
laryngopathia premenstrualis is also relatively common at the time of ovulation.54
Familiarity with the normal ovarian cycle is helpful
in understanding these problems (Figure 16–1). The
cycle begins with the menstrual period and ends just
prior to the next menses. The first portion of the cycle
is known as the follicular phase. It is characterized
by gradually increasing levels of estrogen and low
levels of progesterone. The follicular phase normally
occupies the first 14 days of the cycle, but is variable.
Ovulation begins the luteal phase, which continues

Figure 16–1.  Normal ovarian cycle. FSH = follicle-stimulating hormone; LH = luteinizing hormone.


280

Clinical Assessment of Voice

for 14 days. Progesterone levels increase during the
first half of luteal phase. Estrogen decreases but rises
again slightly premenstrually.
The cause of menstrual cyclic dysphonia remains
incompletely understood. The combined activity of
estrogen/progesterone in the premenstrual period
causes venodilatation by relaxing smooth muscles,

thereby increasing blood volume. These changes
result in engorgement of vocal fold blood vessels and
vocal fold edema. In addition, polysaccharides break
down into smaller molecules in the vocal folds and
bind water, increasing fluid accumulation. Vasodilatation also causes changes in nasal patency and selfperception (audition). In addition, the premenstrual
hormonal environment decreases gastric motility,
exacerbating laryngopharyngeal reflux. Abdominal
bloating and cramps impair effective support. In
addition, Abitbol et al1 showed a strong correlation
between premenstrual dysphonia and luteal insufficiency. The incidence of premenstrual hoarseness is
unknown, but anecdotally it appears to be significant
in about one-third of women. Although most authors
and clinicians have been more impressed with premenstrual voice changes than with those occurring
during midcycle, at least one author has suggested
that voice changes occurring at the time of ovulation may actually be more prominent.2 In a survey
of female singers regarding premenstrual symptoms,
the most frequent general symptom was abdominal
bloating, and the most frequent voice symptom was
difficulty singing high notes.55 Other commonly
reported vocal symptoms were changes in voice quality and impairment in flexibility. A study of vocally
untrained young women revealed no spectrographic
changes in the voice through the menstrual cycle,56
although other studies have reported cyclic hoarseness or even periodic aphonia during the premenstrual period of nonsingers.57–59 Speech dysfluencies
have also been reported to occur more frequently in
the premenstrual period.60
Although ovulation inhibitors have been shown to
mitigate some of these symptoms,51 in rare patients,
older birth control pills containing more androgenic
progesterones were reported to alter voice range
and character deleteriously after only a few months

of therapy.61–64 Current formulations of oral contraceptives use much lower hormone doses and voice
changes have not been reported.65 Indeed, many
women find that the oral contraceptives decrease
vocal fluctuations and stabilize the voice.66 When oral
contraceptives are used, the voice should be monitored closely. Androgenic progesterone-containing
oral contraceptives are still available in some countries and may cause permanent masculinization of

the voice. Oral contraceptives marketed in the United
States generally do not contain these progesterones.
Under crucial performance circumstances, oral contraceptives can be used to alter the time of menstruation. While common in athletes, the effects of
this manipulation on the voice are unknown. When
cyclical voice changes are incapacitating, endocrinologic or gynecologic assessment and appropriate hormonal therapy certainly should be considered.
The hormonal cycle is also associated with menstrual cramping, or dysmenorrhea. Cramps occur in
approximately one-half to three-quarters of menstruating women, and about 10% are disabled for 1 to 3
days monthly.64,67 Muscle cramping associated with
menstruation causes pain and compromises abdominal contraction. This undermines support and makes
singing or projected speech (acting and public speaking) difficult. Dysmenorrhea is also associated with
diarrhea and low back pain, which further impair
support. It also may be accompanied by fatigue,
headache, dizziness, emesis, nausea, all of which
are distracting and potentially impair technique and
performance. Menstrual cramps seem to be due to
uterine contraction mediated by local prostaglandin
release. Hence, prostaglandin inhibitors such as ibuprofen are prescribed commonly for dysmenorrhea.
The combination of such drugs with capillary fragility and other hormonally mediated vascular changes
puts a professional voice user at higher risk of vocal
hemorrhage.53 Medications that impair coagulation
should therefore be avoided. The development of the
selective cyclooxygenase-2 (COX-2) inhibitors offers
an alternative to nonsteroidal anti-inflammatory

drugs (NSAIDs). The selective COX-2 inhibitors are
marketed as not causing anticoagulation or gastric
complications. Dysmenorrhea pain relief has been
documented with celecoxib,68 offering a potentially
safer alternative for professional voice users, although
these medications are not widely used by gynecologists for this purpose.
The hormonal cycle also is associated in some
women with premenstrual syndrome (PMS). This
syndrome may include emotional lability, depression, anxiety, irritability, decreased concentration
ability, abdominal bloating, edema, nausea, diarrhea, palpitations, water and salt retention, and other
changes that may affect voice performance adversely.
Insomnia occurs commonly, as well, and subsequent
sleep deprivation may further impair voice function.
In controlled, double-blind studies, fluoxetine (Prozac, Lilly) has been shown to reduce the psychiatric
symptoms associated with severe premenstrual dysphoric disorder, especially when taken only during
the premenstrual period.69,70 Interestingly, further




16.  Endocrine Function

studies have shown improvements in physical symptoms including bloating, breast tenderness, and headache with intermittent dosing of fluoxetine.71 The
manufacturer is now marketing fluoxetine under the
name Serafem with specific labeling for premenstrual
use. Although no information has been published on
the effects of fluoxetine on premenstrual dysphonia,
the reduction in physical symptoms may be helpful
in facilitating voice performance. More study on the
use of this medication is required before its use can be

recommended strongly for voice professionals. Oral
contraceptives are also frequently helpful in mitigating premenstrual symptoms and are widely used for
this purpose.
One common disorder of the menstrual cycle is
polycystic ovary syndrome (PCOS). Anovulation
and physiologic androgen excess can lead to marked
virilization. Although hirsutism, oily skin, and acne
are most common, androgen-mediated vocal changes
can also occur. Treatment of PCOS is evolving constantly, but oral contraceptives, spironolactone, and
metformin are often able to control some of the
symptoms.72 In obese patients with PCOS (80% of
patients), weight loss is one of the most effective
treatments.
Endometriosis commonly presents as severe
dysmenorrhea. Ectopic areas of endometrial tissue
respond to the hormones regulating the normal menstrual cycle and undergo a normal cycle of proliferation and shedding (menstruation) on the surface of
pelvic or abdominal organs. The pathophysiologic
mechanism resulting in endometriosis is unknown.
Control of endometriosis symptoms can sometimes
be obtained with use of oral contraceptive agents or
other medications. Hormonal intrauterine devices
(IUDs) are highly effective in controlling endometriosis symptoms. Danazol (Danocrine, Samofi) is no
longer used widely due to significant side effects and
high cost. Danazol induces a high-androgen, lowestrogen hormonal environment that can coarsen and
lower the voice.73 Leuprolide acetate, while effective,
causes a hypoestrogenism similar to menopause that
also may be harmful to voice strength and quality.
Because leuprolide acetate causes medical menopause it should not be used for more than 6 months
even in patients without professional voice needs,
because it causes significant osteoporosis as well as

other symptoms of menopause. The estrogen suppression caused by leuprolide acetate is sufficiently
great that it should rarely be used in voice professionals. Laparoscopy with ablation of the areas of
endometriosis may provide months of symptom
control but does not commonly result in a permanent cure. In patients in whom all visible disease can

281

be treated, the 5-year recurrence rate is 20%. If oral
contraceptives and hormonal IUD are ineffective,
we frequently recommend that voice professionals
undergo early laparoscopy for definitive diagnosis
and surgical control in order to avoid the long-term
use of potential voice-altering medications.

Sexual Activity
Many myths and unproven admonitions are promulgated by performers regarding the effects of sexual
activity on performance. Many professional performers believe that coitus prior to a performance is detrimental. Similar prohibitions on sexual activity before
athletic competition are widespread. A few studies in
the sports medicine literature have been performed
to determine whether sexual activity can affect athletic performance. Sztajzel et al found a very slight
difference in heart rate recovery following exercise
in a group of male patients 2 hours after sexual activity.74 This difference disappeared in a second test
performed 10 hours after sexual activity.75 An earlier
study in male athletes failed to show a difference in
performance 12 hours after sexual activity.76 Unfortunately, there are no more recent studies about the
effects of sexual activity on athletic performance, and
no studies at all regarding the effects of sexual activity
on the voice. There are several limitations in applying
the limited information about athletic performance
to voice professionals. All of the studies above were

performed using male subjects. There is no published study detailing the effects of sexual activity
on the athletic or vocal performance of females. It is
not known whether studies on athletic performance
are applicable to vocal performance. More research
is required to answer the questions about the vocal
effects of sexual activity.
Multiple methods of contraception are available
and effective. In the female voice professional, barrier
methods or nonhormonal IUD (Paragard, Duramed)
are probably the safest choice as no hormonally active
medications are taken. Patients taking oral contraceptives must be monitored carefully for voice changes,
and androgenic compounds should obviously be
avoided. Medroxyprogesterone acetate (Depo-Provera, Pharmacia, and Upjohn) is a long-acting progestational agent that is administered by injection
every 3 months. Medroxyprogesterone acetate causes
adverse changes in the lipoprotein profile, probably
accelerates osteoporosis, and has been associated
with depressive episodes. Depo-Provera induces a
hypoestrogenic state that may cause voice changes
similar to those encountered during menopause.