University of Pretoria etd - De Koker, E (2004)
To evaluate the usefulness of the electrophysiological tests available,
specifically auditory evoked potentials, in the audiological evaluation of
pseudohypacusic patients.
The main question addressed is:
contribution electrophysiological tests can make to the detection of
pseudohypacusis and the determination of thresholds in the difficult-to-test
population of mine workers.
“All diagnostic procedures are designed to identify the presence of the
disorder as early as possible. That is, the procedure must accurately identify
those patients with the disorder while clearing those patients without the
disorder” (Roeser et al., 2000b:12). This requirement for audiological test
procedures is met by the tests described in Chapter 2, in that the conventional
tests can identify pseudohypacusis.
The audiologist’s responsibility goes further: it is not only to identify the
presence of a disorder, but to quantify it, thus to determine frequency-specific
hearing thresholds for all patients assessed, in order to provide guidance for
the rehabilitation process, as well as to facilitate recommendations and
decisions regarding patient referrals (Stach, 1998; Roeser et al., 2000b).
Schmulian (2002) supports this position, commenting that poor and inaccurate
diagnostic procedures result in sub-standard recommendations regarding the
rehabilitation of the disorder.
In the field of medico-legal investigations, there is a further reason for not only
identifying but also quantifying the hearing loss, namely financial loss. Coles
University of Pretoria etd - De Koker, E (2004)
and Mason (1984:71) clarify the importance of true hearing threshold
estimation as follows: “In medico-legal investigations of all kinds, precautions
have to be taken against falsification of disability by the patient since there is a
clear motivation for him to exaggerate and thereby obtain greater financial
advantage. This is particularly necessary where the disability claimed can
only be fully characterized by including subjective aspects, as in the case of
hearing loss.”
This is certainly of particular relevance for mine workers with noise-induced
hearing loss who present with pseudohypacusis.
A pseudohypacusic
worker’s lack of co-operation confounds efforts to obtain accurate frequencyspecific information, and often leads to large numbers of pending cases.
These workers have to be retested, which increases the cost of audiological
and other specialist assessments. Retesting workers also leads to additional
expenditure (further financial implications), since these workers miss shifts
and the mining company thus loses production. Additional transport costs may
also be involved if workers are referred for further assessments.
The lack of the availability of accurate hearing thresholds results in situations
where compensation is not paid to deserving cases and in overcompensation
inconsistencies are not detected. The frustration of audiologists, occupational
health centre staff and the mining industry in general is understandable.
Abramovich (1990), Martin (1994) and Schmulian (2002) state that a lack of
patient co-operation, irrespective of the cause or motivation, necessitates the
use of additional, more objective (and sometimes more costly) procedures,
and that other responses apart from behavioural responses to acoustic
signals should be explored for the estimation of hearing thresholds.
In the assessment of hearing, audiologists have always used a test battery
approach (Hall & Mueller, 1997) to ensure acceptable service delivery to
clients. The various tests available to audiologists are used in conjunction
University of Pretoria etd - De Koker, E (2004)
with each other and allow for cross checks to confirm results.
With a
pseudohypacusic patient, such efforts generally confirm the presence of
pseudohypacusis without quantifying its extent, in other words, tests fail to
provide frequency-specific hearing thresholds.
The most reliable means of cross checking is provided by test procedures that
require no voluntary response from the patient (Schmulian, 2002). Gorga
(1999) indicates that assessments of pseudohypacusic patients require the
use of test procedures that do not rely on voluntary behavioural responses.
The quest for measures not requiring a behavioural response has led to the
development of electrophysiological tests, which provide an objective
assessment of auditory sensitivity (Hall, 1992).
Rintelmann et al., (1991);
Stach, (1998) and Roeser et al. (2000b) also promote the use of physiological
tests for difficult-to-test populations.
Today, audiologists have a wide range of electrophysiological assessment
tools to select from (Roeser et al., 2000b).
These are examined and
evaluated in this chapter. Particular attention is focused on auditory evoked
potential (AEP) methods, which have been shown to provide estimates of
hearing thresholds.
The objective is to identify and evaluate audiological
solutions and test procedures for the population of mine workers, in which
noise-induced hearing loss is frequent and pseudohypacusis is rife.
Discoveries in the field of audiology (and other related fields, including
neurology and electronics) have recently led to rapid advances in the
development of electrophysiological tests (Ferraro & Durrant, 1994; De Waal,
2000; Roeser et al., 2000b). Audiological assessment techniques no longer
need to be limited to traditional behaviour-based psycho-acoustic tests, now
University of Pretoria etd - De Koker, E (2004)
that EP tests can help satisfy the need to assess auditory sensitivity at
specific frequencies objectively (Schmulian, 2002).
The EP methods used in the past thirty years have included immittance
testing, acoustic reflex (AR) measurements, oto-acoustic emission (OAE)
tests and auditory evoked potential (AEP) tests.
Of the many electrical
responses to auditory stimuli, most originate in the central nervous system.
Some are generated in the cochlea, while others are reflexive muscular
responses (Glasscock, Jackson & Josey, 1987).
Immittance and OAE measurements are not measures of hearing per se, but
are means of evaluating the status of the auditory system at specific
peripheral levels, although never as an entire system (De Waal, 2000). These
measures do not provide frequency-specific thresholds, but merely confirm
the suspicion of pseudohypacusis, thus serving as a means of cross checking.
Acoustic immittance measures (tympanometry, static compliance and acoustic
reflex) have been well established as a routine part of audiological evaluation
(Rintelmann et al., 1991). The primary application of acoustic immittance is
the evaluation of organic hearing disorders. It can also be useful in the
detection of pseudohypacusis.
Martin (1994) claims that the sophistication of automated middle ear tests may
discourage pseudohypacusis, and is therefore very valuable in the detection
or prevention of pseudohypacusis.
Clinicians should thus remember to
suggest to the patient that the test is fully automatic and that no response is
needed, thereby removing the temptation to feign a hearing loss.
It is
therefore generally good practice to perform an immittance test first if this test
can be used to avoid pseudohypacusis. This is a valid approach, but goes
against the recommendation of Rintelmann et al. (1991) that supra-threshold
tests should be performed after air- and bone-conduction tests. Experience
University of Pretoria etd - De Koker, E (2004)
has shown that completing threshold testing before embarking on suprathreshold tests does save time and prevents unco-operative patients from
finding a supra-threshold reference level (Dobie, 2001; De Koker, 2003).
The AR threshold is the most useful measurement in the detection of
pseudohypacusis. In persons who have normal hearing, an acoustic reflex is
usually elicited by means of contralateral stimulation at sensation levels that
range from 70 to 95 dB.
For persons with cochlear lesions, as in mine
workers exposed to noise, the reflex may be obtained between 15 and 60 dB
(Rintelmann et al., 1991). When the difference between the AR threshold and
the voluntary threshold is extremely low (5 dB or less), the pure-tone threshold
must be questioned on the basis of organic pathology (Martin, 1994; 2000).
Claims of a profound unilateral or bilateral hearing loss can be refuted if the
AR is present at normal stimulus levels, but the phenomenon of recruitment
may limit the usefulness of AR measurements in estimating hearing
thresholds, especially in cases of noise-induced hearing loss.
Tympanometry provides an immediate evaluation of the middle ear status.
Present ARs and normal middle ear function are not compatible with
conductive hearing loss (Qiu et al., 1998). If conductive hearing loss is present
with normal middle ear function pseudohypacusis can be expected.
reason being mine workers’ unfamiliarity with bone conduction testing.
AR measurements may be useful in estimating actual hearing thresholds by
performing the sensitivity prediction with the acoustic reflex test (SPAR).
Middle ear reflex thresholds for pure tones are compared with those for wideband noise, as well as for filtered low- and high-frequency wide-band noise
(Martin, 1994; 2000). In the researcher’s experience, the high incidence of
absent ARs in this population makes the use of the SPAR test impossible.
Dobie (2001) also claims that the SPAR test has no clinical utility in
pseudohypacusic populations. Some reasons for this, although Dobie (2001)
does not mention them, could be elevated and absent ARs.
University of Pretoria etd - De Koker, E (2004)
In conclusion as was the case with the behavioural tests described in
Chapter 2 immittance testing predicts and detects pseudohypacusis but is not
quantitative in nature.
Oto-acoustic emissions (OAE)
Small changes in the biomechanical function of the cochlea can be monitored
by measuring OAEs, which are generated within the cochlea by active nonlinear processes involving the outer hair cells (Kvaerner et al., 1996).
It is impossible for a patient with compensable hearing loss to have normal
OAEs, and OAE testing is thus advocated as a quick and objective means of
confirming hearing status in suspected cases of pseudohypacusis (Qiu et al.,
1998). A patient with normal OAEs should have normal hearing thresholds.
Unfortunately, the usefulness of OAE testing is limited in cases of noiseexposed patients, as such individuals often exhibit abnormal or absent OAEs
with normal hearing as a result of pre-symptomatic cochlear damage (Hall,
2000; De Koker et al., 2003). So far, it has also been difficult to correlate
OAEs and behavioural thresholds (Hall, 2000). OAEs are another qualitative
assessment tool which is useful in the detection of pseudohypacusis.
Despite the considerable interest that has been generated by all the
conventional tests described in Chapter 2 and the electrophysiological tests of
immittance and oto-acoustic emissions, as the foregoing discussion has
indicated, none have provided accurate hearing thresholds in the case of
pseudohypacusic mine workers. The problem faced when compensation is
involved is that the audiologist must obtain ten hearing thresholds (South
African legislation) that are accurate enough to be duplicated in a second test.
The focus is thus on quantitative data.
University of Pretoria etd - De Koker, E (2004)
Accordingly, attention needs to be paid to auditory evoked potential methods
as the most useful and effective electrophysiological measure of auditory
system function (Rance et al., 1998) with due consideration to both the
peripheral and central auditory systems. Hood (1998) emphasises that EP
tests are not tests of hearing, but tests of synchronous neural function and the
ability of the central nervous system (CNS) to respond to external stimuli in a
synchronous manner.
Nevertheless, numerous authors have shown how
closely thresholds from AEP testing correspond with behavioural thresholds
(Reneau & Hnatiow, 1975; Rance et al., 1998; Barrs et al., 1994). This fact,
combined with the statement of Abramovich (1990) that the verification of
hearing loss and the validation of the pure-tone audiogram is important in
dealing with compensation claims, supports the necessity of evaluating AEP
tests within the framework of this study.
Hyde et al. (1986) argue even more strongly that, if AEPs are accepted as the
ultimate arbiter in medico-legal evaluations, the rationale for interposing
confirmatory tests (detection) between a suspicion of and the quantification of
pseudohypacusis is suspect.
Background: the development of the use of AEPs
AEP procedures are not really a “new” technique. Glasscock et al. (1987)
trace the origins of auditory brainstem response (ABR) testing to animal
experiments in the nineteenth century, citing Caton, who reported electrical
activity in the brain of a rabbit in 1875.
They also mentioned other
researchers who investigated electrical activity in the brains of other animals
between 1883 and 1891.
Not only the technique but also the apparatus used to evoke and record the
electrical responses has developed over the years.
photographed an apparatus to record animal electro encephalograms (EEGs)
using a string galvanometer (Glasscock et al., 1987).
During the 1930s,
oscilloscope images were bright enough and electrical amplifiers stable
University of Pretoria etd - De Koker, E (2004)
enough to allow neurophysiologists to concentrate on experimental work
rather than on equipment problems (Abramovich, 1990).
Berger first observed spontaneous electrical activity of the type now routinely
recorded during EEGs in 1929 (Abramovich, 1990; Ferraro & Durrant, 1994).
In searching for electrical activity in the inner ear, Wever and Bray (1930)
recorded potentials in response to auditory stimuli from the round window of a
cat. These potentials have since been termed cochlear microphonic or CM
(Abramovich, 1990).
The main problem facing early researchers was the difficulty of measuring
very small potentials in isolation from other electrical activity within the brain.
Particular difficulty was experienced when the stimuli were of low intensity, as
EEG voltage was much greater than the voltage of the evoked potential
(Reneau & Hnatiow, 1975). The development of averaging computers has
facilitated more accurate analysis of small bio-electrical signals (Abramovich,
1990). Glasscock et al. (1987) note that Davis acquired a digital computer in
1961, after which he began using it in his electrophysiological experiments.
The ABR, currently the most popular AEP used in clinical contexts, was first
described and defined by Jewett and Williston in the early 1970s (Glasscock
et al., 1987).
In 1963, the New York Academy of Arts and Sciences sponsored a
symposium of investigators of averaged potentials (including visual,
somatosensory, auditory, myogenic and neurogenic), followed by the founding
of the International Electrical Response Audiometry study group in 1968
(Abramovich, 1990).
Much of the research in the field of AEPs tries to correlate the electrical
responses with auditory behavioural thresholds. Reneau and Hnatiow (1975)
cite difficulties in relating physiological thresholds (such as evoked responses)
to behavioural response thresholds.
It was believed that behavioural
responses are binary measures in which a subject decides between “yes” or
University of Pretoria etd - De Koker, E (2004)
“no”, while physiological thresholds are graded, or quantitative, and that
graded measures are mathematically different from binary ones.
It was
concluded that these two types of tests can be expected to yield different
results. Nevertheless, as a result of subsequent advances in electronics, and
a far greater understanding of brain function, there has been a move in the
field of AEPs, supported in this study, to relate behavioural and physiological
The enthusiasm for auditory evoked potentials in the 1970s resulted in this
type of testing, being incorporated in test batteries for unco-operative patients
such as small children (Martin, 1994). It is thus logical that the use of this
quantitative procedure was also extended to cases of pseudohypacusis
(Roeser et al., 2000b).
As early as 1990, the use of auditory evoked potentials was recommended in
the assessment of pseudohypacusic patients by Abramovich (1990), who also
cites the use of slow vertical responses, auditory brain stem response, middle
latency responses and transtympanic electrocochleograms as possible
auditory evoked potentials to be used with pseudohypacusic patients. Today,
a mere decade, later is it predicted that in future, AEPs will become even
more prominent in the field of Audiology (Roeser, Buckley & Sichney, 2000).
Nomenclature and definitions
Picton and Scherg (1990) argue that one of the most important clinical
applications of AEPs is their use in objectively evaluating the hearing of
patients who are unable to respond during conventional testing. However, in
order to evaluate this application, it is important first to define auditory evoked
potentials and to highlight controversial issues.
Stach (1998:293) describes the measurement of AEPs as follows:
University of Pretoria etd - De Koker, E (2004)
The brain processes information by sending small electrical impulses
from one nerve to another. This electrical activity can be recorded by
placing sensing electrodes on the scalp and measuring the ongoing
changes in electrical potentials throughout the brain. This technique is
called electroencephalography, or EEG, and is the basis for recording
evoked potentials. The passive monitoring of EEG activity reveals the
brain in a constant state of activity; electrical potentials of various
frequencies and amplitudes are measured continually.
If a sound is
introduced to the ear, the brain’s response to that sound is just another of
a vast number of electrical potentials that occur at that instant of time.
Evoked potential measurement techniques are designed to extract these
tiny signals from the ongoing electrical activity.
This described electrical activity can be spontaneous or event-related (Picton,
2001). Responses that are time-linked to some event or stimulus are called
event-related potentials (ERPs), and can be responses to a sensory stimulus
(such as a visible flash or a sound), a mental event, or the interruption, delay
or omission of a stimulus (Picton, 2001).
Auditory evoked potentials (AEP) are a type of ERP in which the stimulus is a
sound, and the response takes the form of very small electrical potentials
originating in the nervous system and recorded at the scalp (Picton, 2001).
AEPs originate from structures such as the auditory cortex, the auditory
brainstem and the auditory cranial nerve (VIII or 8th).
These electrical
potentials are very small: 2 to 10 micro volts for cortical AEPs, and less than
one microvolt for deeper brainstem structures (Picton, 2001).
The measurement of these potentials in response to auditory stimuli has
become a valuable diagnostic tool (Stach, 1998) but is still an evolving field in
which there are problematic issues that need to be resolved.
University of Pretoria etd - De Koker, E (2004)
Problematic issues in the field of AEP
It should be noted that the terms “evoked potential” and “evoked response”
are used interchangeably in the literature (Hood, 1998). The term “response”
is derived from the procedure of pure-tone audiometry in which a stimulus is
presented and a response (motor action) is subsequently recorded. In AEP
testing, a response is not recorded, but a potential is measured. Furthermore,
electrical activity is elicited by a signal, and not a stimulus (Goldstein &
Aldrich, 1999).
The term “stimulus” implies perception, but in tests of auditory brain stem
response and auditory steady state response, electrical activity is measured
sub cortically and only up to brainstem level.
It should therefore be
remembered that the terms “stimulus” and “signal” are interchangeable, as are
“potential” and “response” (Schmulian, 2002).
The field of evoked potentials has been burdened with inconsistencies in
terminology and definitions and its classification system has lacked uniformity
and clarity (Ferraro & Durrant, 1994; Schmulian, 2002). Schmulian (2002)
attributes this lack of clarity to the presence of specialists from the different
fields of audiology, neurology and otolaryngology who all work in the field of
evoked potentials. Classifications of AEPs in the literature can be divided into
those based on anatomical generators, on the type of potential, on the types
of stimuli used, on the location of recording electrodes and on latency
characteristics (the time between stimulus onset and response) (Schmulian,
The most common classification of AEPs is based on their time domain
(Goldstein & Aldrich, 1999), in which the time between the stimulus and the
response is termed the “latency epoch”. Ferraro and Durrant (1994) mention
that, although this classification system is the easiest to apply, the
classification of latency is not standardised. A familiar method is to classify
response latency as short, middle or late latency responses, depending on the
University of Pretoria etd - De Koker, E (2004)
time between the presentation of the stimulus and the responses’ becoming
evident as an AEP. Short latency is referred to as “fast” by Glasscock et al.
(1987), and as “early” by Abramovich (1990), while “late” latency responses
are also referred to as “slow”. These types of inconsistency create confusion.
characteristics and pathology, it has been found that authors attribute different
latency epochs to different AEPs.
So, for example, according to Picton
(2001), the ABR is seen 1.5 to 15 milliseconds (ms) after the stimulus, which
contradicts Abramovich (1990), who states that an auditory brain stem
response (ABR) is seen within the first 10 ms after the stimulus. Different
nomenclatures are also used to identify major peaks for AEPs, for example
Roman and Arabic numerals are used for ABR waves, and “No” or “SN10” are
used to identify the slow negative wave appearing in the ABR after 10 ms.
The use of different potentials in pseudohypacusis
The use of auditory evoked potentials in the estimation of hearing in patients
that cannot or will not co-operate during behavioural tests has been
advocated by numerous authors (Abramovich, 1990; Mc Pherson & Starr,
1993; Stach, 1998). Schmulian (2002) expresses a stronger opinion, saying
that AEP testing is the only procedure in the audiologists’ test battery that can
quantify the hearing sensitivity of unco-operative patients.
If an audiologist has to rely on a single test in a battery (due to an uncooperative patient), AEP testing needs to meet the following requirements
(Roeser et al., 2000b):
The test should diagnose the nature of the hearing loss (conductive or
sensory neural).
The degree of hearing loss (from normal hearing to profound hearing
loss) has to be established.
University of Pretoria etd - De Koker, E (2004)
The configuration of the hearing loss (across a range from 250 to
8 000 Hz) is important clinical information and must be determined.
Frequency-specific hearing thresholds need to be estimated for both
The above requirements are used in the discussion below to evaluate the use
of different auditory evoked potentials in pseudohypacusic patients.
Early potentials
The first three AEPs identified (cochlear microphonic (CM); action potential
(AP) and summating potential (SP) are very early-stage potentials seen during
the first 5 ms after stimulation with a sound (Stach, 1998). Responses to
sound originate in the cochlea and the distal portion of the auditory nerve.
They are also grouped together in clinical use as the electrocochleogram
(EcochG). Tone burst and click stimuli are used to elicit responses, and two
different electrode placements for near-field measurements are possible,
transtympanic placement, where an electrode is invasively placed
through the tympanic membrane onto the promontory of the temporal
bone; and
the external auditory meatus (EAM) near the tympanic membrane
(Abramovich, 1990).
The value of the EcochG lies in its usefulness for assessing the hearing of
young children who are difficult to control in clinical situations, and in the fact
that these potentials are not altered by anaesthesia. The EcochG provides
information on inner ear function (Abramovich, 1990) in conditions such as
tinnitus, Meniere’s disease and sudden hearing loss (Halliday, 1993).
disadvantages are that low frequency function is almost impossible to assess,
and the surgical procedures required for transtympanic placement make the
EcochG invasive (Abramovich, 1990).
University of Pretoria etd - De Koker, E (2004)
The use of electrocochleography in pseudohypacusic populations (Qiu et al.,
1998; Rintelmann et al., 1991; Abramovich, 1990) has been reported.
Rintelmann and his co-authors opine that EcochG does not measure the
ability to hear. The invasive nature of the surgical procedures for the EcochG
and the resultant need for an Ear-, Nose- and Throat (ENT) specialist
(Schmulian, 2002), together with the ability of the test to measure only the
most peripheral functions of the auditory system limit its clinical use to a small
number of highly specialised diagnostic centres (Abramovich, 1990; Stach,
It can be concluded that pseudohypacusic patients are not adequately
evaluated by early potential testing, as it fails to include all of the frequencies
required for compensation assessments, and the invasiveness of the
procedure is unacceptable for Occupational Health applications.
ABR is a big misnomer in the field of AEPs (Schmulian, 2002). Since the ABR
is the most widely used AEP (Hood, 1995), all AEPs have come to being
perceived as ABRs, irrespective of the latency epoch and the equipment used
(Goldstein & Aldrich, 1999).
Ferraro and Durrant (1994) have found ten
different names for ABRs in a literature review, including “brainstem auditory
evoked potential”, “brainstem auditory evoked response”, and “auditory
brainstem evoked response”, to list but a few.
In ABR testing, electrical potentials are generated by the VIII (8th) cranial
nerve and neural centres within the brainstem (Stach, 1998). The ABR uses
far-field potentials recorded at the scalp (vertex), and comprises five or more
waves generated in the auditory pathway up to the level of the inferior
The procedure is firmly established in clinical practice for
estimating audiometric thresholds and for neurological/neuro-otological
diagnosis (Abramovich, 1990). In South Africa, ABR has for many years been
the test of choice among the available AEP procedures, particularly for
University of Pretoria etd - De Koker, E (2004)
difficult-to-test patients for whom the configuration and severity of hearing loss
have to be determined. The waves are robust and easily recorded, and are
unaffected by the patient’s state of consciousness (the patient can even be
asleep or sedated).
ABR potentials are minute, rarely reaching amplitudes greater than 1 micro
volt (µV), and thus it requires a great deal of averaging to distinguish
potentials from background noise and other artefacts (Arnold, 2000).
Furthermore, ABR tests rely on transient responses elicited by brief acoustic
stimuli (Arnold, 2000), as the more abrupt the stimulus, the more clearly
defined the ABR. The most widely used stimulus is a broadband click,
because of its rapid onset (100 µsec) and broad frequency content, which
stimulates a large portion of the basilar membrane to give a reasonable
indication of hearing thresholds between 2 000 and 4 000 Hz. However, due
to the structural and mechanical properties of the cochlea, ABR can only
predict auditory sensitivity in the upper part of this range to within 5 to 20 dB
of behavioural thresholds (Rance et al., 1998). This limitation has led to the
development of other stimuli, including tone bursts, filtered clicks and various
masking techniques to provide more precise information on narrower
frequency ranges (Hood, 1998).
According to Swanepoel (2001), tone bursts are the stimulus of choice where
low frequency threshold information is required.
Tone bursts are more
frequency-specific than clicks, and their gradual stimulus onset ensures good
frequency specificity (Weber, 1994). Unfortunately, the resulting stimulus does
not elicit a clear ABR and, therefore, an abrupt stimulus onset is necessary to
improve the quality of the response. However, this introduces high-frequency
energy into the test stimulus, necessitating the use of masking techniques to
eliminate the effects of unwanted high frequency energy.
Stapells et al.
(1990) have obtained good agreement between ABR and behavioural
thresholds by using tone burst stimuli embedded in notched noise.
University of Pretoria etd - De Koker, E (2004)
Unfortunately, the time needed to obtain a single ABR threshold for each ear
exceeds 30 minutes, making a full audiogram impractical (Weber, 1994). With
children the test often lasts for as long as the child sleeps and, even with
adults, a long test is tiring and undesirable (Swanepoel, 2001).
At the
moment, the best method for determining hearing loss configuration is to
present first a low-frequency tone burst and then a click ABR. This procedure
is an attempt to shorten the procedure, but should still allow the audiologist to
get an idea of the configuration of the hearing loss.
An advantage of ABR is that the latencies of the various waves are quite
stable within and among different patients (Abramovich, 1990). In addition,
time intervals between peaks are prolonged by auditory disorders central to
the cochlea, making ABR useful in differentiating cochlear and retrocochlear
pathology (Weber, 1994).
A disadvantage is that the interpretation of wave forms is subjective (Martin,
1994), and the interpretation of tone bursts requires considerable expertise
and experience (Abramovich, 1990; Swanepoel, 2001). The ABR is also timeconsuming, and the maximum stimulus level for clicks and tone bursts is
restricted, resulting in a risk that the audiologist may fail to identify residual
hearing at high loudness levels. Furthermore, the high cost of instrumentation
and software are added negative considerations (Schmulian, 2002). Qiu et al.
(1998) point out the further disadvantage that involuntary responses are
generated only by sub-cortical structures and, hence, can never provide a
measure of true hearing thresholds. These authors also criticise the great
technical demands with regard to stimulus, filter settings, recording methods
and response interpretation with bone-conduction ABRs.
This limits the
clinical application of the technique.
In a study by Barrs et al. (1994), it was found that an ABR was a useful
procedure in the threshold confirmation needed in cases of noise-induced
hearing loss, but that middle latency responses were more useful than the
ABR because of the ABRs tendency to overestimate hearing loss in down-
University of Pretoria etd - De Koker, E (2004)
sloping audiograms.
Middle latency responses were also more frequency
specific which is important in the case of noise-induced hearing loss
From the preceding discussion, it is apparent that the ABR has up to now
been the most widely used electrophysiological procedure, and is the only
electrophysiological procedure prescribed in South Africa for the formal
assessment of pseudohypacusic patients (RMA guidelines, 2003). Despite
the limitations discussed above, frequency-specific threshold determinations
are possible, but only through a long and expensive process requiring a great
deal of skill and experience on the part of the audiologist. These are two
important limitations that hinder the consistent use of ABRs in hearing
assessment in the mining industry.
Middle latency responses
It is generally accepted that there are two main reasons for the use of auditory
electrophysiological tests, namely the need to make inferences regarding
hearing thresholds and the need to obtain information regarding the functional
and structural integrity of the auditory pathway’s neural components (Kraus,
Kileny & McGee, 1994). The purpose of this section is to provide a basis for
understanding the principles and applications of middle latency response
(MLR) testing and to evaluate the contribution of MLRs in meeting the above
two aims.
An MLR is a series of waveforms occurring 10 to 80 ms after the onset of an
auditory stimulus (Kraus et al., 1994).
Here, again, contradictory
classifications abound in the literature. Abramovich (1990) classifies MLR as
having a latency of 8 to 50 ms, while Picton (2001) and Glasscock et al.
(1987) set latency at 25 to 50 and 12 to 50 ms respectively.
Within the
continuum of components comprising scalp-recorded AEPs, MLRs follow
ABRs and precede late latency responses (LLRs), while evoked potentials No,
Po, Na, Pa, Nb and Pb are classified as MLRs ( Kraus et al., 1994).
University of Pretoria etd - De Koker, E (2004)
According to Kraus et al. (1994), Geisler and his co-workers were the first
investigators to describe MLRs (in 1958). MLRs are measured at the scalp,
using an electrode montage identical to that used for ABR recordings, and
MLR generators include many brain structures central to the midbrain, as well
as structures outside the primary auditory pathway, such as the auditory
thalamocortical pathway, the reticular formation and the multi-sensory
divisions of the thalamus (Kraus et al., 1994).
MLR is used clinically for electrophysiological determination of hearing
thresholds at lower frequencies, for the assessment of cochlear implants and
auditory pathway function, and for the localisation of auditory pathway lesions.
They are also used intra-operatively (McPherson & Ballachanda, 2000). It is
thus clear that MLR has many applications in audiology, but unfortunately, the
disadvantages of MLRs overshadow the advantages.
The most important limitations include:
the inconsistency of responses as specifically observed in the
paediatric population (Kraus et al., 1994);
including EEG for example;
the need for the patient to be awake, co-operative and alert (Hood,
1995). Ferraro and Durant (1994) state that sensitivity to the patient’s
state of consciousness limits the acceptance of MLR techniques;
Thorton et al. (1984) show that MLRs are distorted and delayed during
sedation, and those potentials are poorly detected in stage IV sleep;
electrophysiological test (Mc Pherson & Ballachanda, 2000), and are
regarded as difficult to record (Abramovich, 1990), causing a lack of
facilities where these procedures could be tested; and
University of Pretoria etd - De Koker, E (2004)
reports that MLR potentials can be contaminated by muscle potentials
from the neck or peri-auricular region (McPherson & Ballachanda,
The question that needs to be answered is whether MLRs can be used as a
technique to identify pseudohypacusic mine workers and quantify their
hearing loss.
Abramovich, (1990) advocates the use of MLRs in
pseudohypacusic patients. He is of the opinion that a stimulation rate of 40
per second instead of the usual 10 per second can cause a superimposition of
the peaks of MLRs and an augmentation of the response. He specifies that
MLRs are to be used in this population when slow vertical response (SVR)
measurement conditions are poor. Barrs et al. (1994) mention the possibility
of using MLRs to detect work–related noise-induced hearing loss, stating that
MLRs are more effective in threshold estimation than ABRs, as a result of the
steepness of the audiometric curve in noise-induced hearing loss. Barrs et al.
(1994) also advocate the use of MLRs to verify low frequency thresholds.
McPherson and Ballachanda (2000) argue that the biggest problem in testing
and verifying these MLRs is the fact that these evoked potentials are not
considered to be mainstream electrophysiological tests in audiology practice.
Hence, equipment and facilities are not readily available.
Late latency responses (LLR)
As indicated previously, confusing nomenclature also exists regarding the
potentials evoked at later latencies. These potentials are described as “slow”
(Halliday, 1993), while Stach (1998) favours the term “late latency response”
“Slow vertical response” (Abranovich, 1990) and “cortically evoked
responses” (Rickards et al., 1996) are other nomenclature in the existing
literature. The confusing nomenclature stated above is further compounded
by a lack in uniformity in the latency epochs of LLRs. Ferraro and Durrant
(1994) define LLRs as potentials manifesting 50 to 800 ms after the stimulus,
while Glasscock et al. (1987) and Picton (1991) relate latencies in this subclass to 250 to 600 and 50 to 200 ms, respectively.
University of Pretoria etd - De Koker, E (2004)
These potentials have been found to be greatly affected by subject state
(Abramovich, 1990; Stach, 1998), and the potentials are best recorded when
the patient is awake and attending carefully to the sounds presented. It is
thus understandable that these methods are only used in adult, difficult-to-test
populations. Stach (1998) mentions that LLRs are robust and easily recorded
in adults and that the response can estimate auditory sensitivity independently
of behavioural response.
As is the case with other potentials, the late latency response generators are
still unknown. Halliday (1993) attributes the P3 or P300 AEP to widespread
activity of the frontal cortex involving the parietal association areas.
An important disadvantage of LLRs is the fact that the procedure is timeconsuming.
Abramovich (1990) estimates the time requirement for four
thresholds in two ears at 60 minutes.
pseudohypacusic population, it is worth noting that several authors have
promoted LLRs as a medico-legal test (Halliday, 1993; Rickards et al., 1996;
Rickards & De Vidi, 1995; Abramovich, 1990; Dobie, 2001; McCandless &
Lentz, 1968; Hyde et al., 1986; Coles & Mason, 1984).
As early as 1968 McCandless and Lentz tested LLRs on adults with
pseudohypacusis using pure-tone stimuli with a 700 msec duration. They
found a very good correlation between the electrophysiological and
behavioural thresholds (5dB).
Abramovich (1990) claims that SVR testing is the test of choice for assessing
non-organic hearing loss. He argues that SVRs most closely approximates
the results of conventional frequency-specific audiometry (within 10 dB), and
that SVR is insensitive to neurological dysfunction. Pseudohypacusic patients
are instructed to pay attention, and those who deliberately exaggerate their
level of attention due to anxiety can be clearly identified. The stimulus used is
a 100 ms tone burst with rise and fall times of 10 ms.
University of Pretoria etd - De Koker, E (2004)
Coles and Mason (1984) used a 50 to 300 ms latency epoch and have proven
that these latency responses have by far the greatest value for verifying puretone thresholds in adult patients, in comparison to brainstem and cochlear
potentials. The tonal signals that these authors used had a duration of 200
ms and a rise and fall time of 10 ms. A specific advantage of LLRs mentioned
by these researchers is the frequency specificity at low frequencies where
non-organic overlay is maximal.
They also argue for the use of LLRs in
medico-legal investigations because of the non-invasive nature of the
procedure and because the procedure tests up to a much higher dB level
than, for instance, the ABR.
Hence, there is a less likelihood of a non-
peripheral disorder causing a discrepancy between the AEP and the
behavioural threshold.
Hyde et al. (1986) have expressed the opinion that the verification of puretone audiometry is a long-standing problem in Departments of Veterans‘
administration, compensation assessment for noise-induced hearing loss and
medico-legal evaluation. These authors have found a correlation between the
slow vertical response and behavioural thresholds of within 10 dB. The stimuli
used are tone bursts with 10 ms rise and fall times, and a 40 ms plateau.
Despite a very good threshold estimation ability, and although by 1986 the
procedure had been used in the Mount Sinai hospital (Toronto), for a decade,
the authors emphasise the following disadvantages of using SVRs:
testers in a clinical setting need to be experienced and carefully trained
audiologists whose performance is monitored ( it is clear that the skill
requirement is very high);
the test procedure is demanding and the skill requirements are often
testers need to have an adequate caseload to maintain their skill;
all clinical interpretation is subjective and on-line;
slow vertex response audiometry is problematic in 5 per cent of cases
due to high levels of rhythmic activity:
University of Pretoria etd - De Koker, E (2004)
the time exceeds 1.5 hours in 95 per cent of cases (Hyde et al., 1986);
from the above it is clear there is still a limited acceptance of the
technique even in North America (Hyde et al., 1986; Dobie, 2001).
Picton (2001) indicates that the British Columbia Workers Compensation
Board has used LLRs, and Rickards et al. (1996) state that cortical evoked
response audiometry (CERA) has been used to assess noise-induced hearing
loss in the Australian state of Victoria for the past 15 years, with 18 per cent of
all noise-induced hearing loss cases referred for CERA.
This seems to
indicate some positive experience with AEP procedures.
CERA thresholds have been found to be within 10 dB of pure-tone thresholds,
but, again, the procedure has failed to gain wide acceptance. Rickards et al.
(1996) indicate that reliance on subjective interpretations of tracings, and the
high levels of skill and training required of testers have hampered acceptance
of CERA as a routine test for pseudohypacusis.
As far as can be determined, late latency responses have not so far been
used in South Africa for the assessment of noise-induced hearing loss or the
evaluation of compensation claims. Although it is clear that, as in any clinical
population, no single AEP method is always the best (Hyde et al., 1986), the
main reason for searching for a better method is a lack of objectivity in
deciding whether the evoked potential is present.
This chapter has discussed electrophysiological tests, particularly AEPs, as a
pseudohypacusic mine workers. Nomenclature, selected definitions and the
historic evolution of AEPs have also been discussed, and the value of various
AEP methods for estimating hearing thresholds have been examined.
University of Pretoria etd - De Koker, E (2004)
A summary of the disadvantages of currently used AEPs based on the above
discussion, is set out in Table 3.1 below.
Although LLRs have been used internationally in medico-legal evaluations, an
even better solution is still sought for. A recent development in the field of
AEPs is auditory steady state responses (ASSRs), which is discussed
comprehensively in the next chapter.
Lins et al. (1995) have found that
results obtained from ASSR testing can be presented as an audiogram,
thereby providing information about the extent, nature and configuration of
hearing loss. Most importantly, it has been reported that ASSR provides true
objectivity, as thresholds are not determined subjectively, through a clinician’s
interpretation of wave forms, as is the case with ABR and LLRs, but are rather
calculated objectively by a computer (ERA systems Pty. Ltd, 2000).
The latter crucial benefit motivated this researcher to investigate this type of
AEP as a possible means for testing pseudohypacusic patients, particularly
those with noise-induced hearing loss in the South African mining industry.
skilled tester
not frequency
sub cortical
influenced by
age sensitive
Type of AEP
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF