, 2... ‘ Immflmumfi flu
V Jump»... .3, am @Wuua

c
(-

 

£83. .3...r\
.3.....3..i.{hu.... , .
£533....u... .99.? f:

. .55.. .vznlhmjsmu‘x
be. $3. ,firfirHL z
. 22....

mom

(.1 .|\ ‘ .- }\L| 5..

. In." 9.31:.3‘111 .7‘
. pl" iflBBWQ WP“
1‘1“! 3! 1....)

I .F‘ .1

own“

, Lb: -.,\..v,a.l|nhf.>u.2 .
. c .1; , ‘ .. . ‘ . . .. .éfimfi
”$234 . , ‘ . . . V E... ‘1. #311..3 i, x.

x.

 

THESVB

lllllllll‘lllllllllllllllllllIlllll‘lllllllllllllllllllll

3 1293 01772

This is to certify that the
thesis entitled

Relations of Race and Gender to
Distortion Product Otoacoustic
Emissions

presented by

Lara L. Gingerich

has been accepted towards fulfillment
of the requirements for

M.A. Audiology

degree in

 

 

    

Major professor

Date i1 NQV (fig

0-7639 MS U is an Affirmative Action/Equal Opportunity Institution

 

 

 

 

LIBRARY
Mlchlgan State
University

 

 

PLACE IN RETURN Box to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE

DATE DUE

DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

ma campus-m4

RELATIONS OF RACE AND GENDER TO DISTORTION PRODUCT
OTOACOUSTIC EMISSIONS

By

Lara L. Gingerich

A THESIS
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF ARTS
Department of Audiology and Speech Sciences

1998

ABSTRACT

RELATIONS OF RACE AND GENDER TO DISTORTION PRODUCT
OTOACOUSTIC EMISSIONS

By
LARA L. GINGERICH
There is presently considerable promise for the use of distortion product
otoacoustic emissions (DPOAE) as an audiological diagnostic tool. The pm'pose of this
study was to compare DPOAEs among racial groups and gender groups in order to
improve its clinical applicability. There was a need also to establish normative DPOAE
data using a commercially available instrument. The present investigation was designed to
answer the following questions: (1) Are there significant differences in DPOAEs recorded
across different racial groups? (2) Are there significant difl‘erences in DPOAEs recorded
between different genders? Participants included persons fiom three different racial
groups: Asian, thk and White. The gender question was answered using White men and
White women. There were no significant differences for DPOAEs between the various
racial groups. There were, however, significant differences between gender for DPOAE
amplitude. A significant difl‘erence was observed for ear canal volume between the two
genders. Thus, the gender differences in DPOAEs may be related to the larger ear canal
volumes measured in men as compared to women. In addition, an increase in DPOAEs
was observed also when SOAEs were present. These results are consistent with prior
studies in which amplitude differences were observed between genders. Thus, gender
effects may need to be taken into account when establishing normative data for clinical

audiologic purposes.

This is dedicated to my Parents who have taught me that strength, determination and
perseverance are the underlying keys to success, and to Dr. Neil West whose medical
expertise saved the life of a very sick little girl in May of 1976.

iii

ACKNOWLEDGEMENTS

I extend gratitude and appreciation to Dr. Ernest J. Moore, chairperson of my
thesis committee for his leadership and commitment to excellence that he instills in his
students.

Appreciation is also extended to the other members of my thesis committee;

To Dr. Jill Elfenbein, whose support was essential during the first brainstorming moments.
To Dr. Cynthia Fowler, who initially fueled the fire for my interest in otoacoustic
emissions. To Dr. Bharti Katbamna, whose suggestions and comments were very helpful.
To Dr. Brad Rakerd, who assisted in my receiving a grant from the Michigan State
University Hearing Research Center.

Sincere gratitude is extended to the Michigan State University Hearing Research
Center for the grant that supported this research.

My thanks is given to Mrs. Patti Hergenreder whose help was essential in
instrumentation set up and troubleshooting.

My sincere gratitude is extended to the many participants of this project. In
particular, I wish to thank Rie Oshahashi and Tomoe Kiyomiya for their assistance in
recruiting Asian women.

Sincere thanks is given to all of the subjects who participated in this project, this
would not have succeeded without them.

Gratitude is extended to Amyn Amlani for his helpful comments and suggestions.

Heartfelt appreciation is given to my Parents for lending helpful advice and to

Jason Socha whose constant support was essential to the completion of this project.

iv

TABLE OF CONTENTS

LIST OF TABLES
LIST OF FIGURES

CHAPTER I - INTRODUCTION
Otoacoustic Emission Pathway
Types of Emissions
Transient Evoked Otoacoustic Emissions (TEOAEs)
Stimulus Frequency Otoacoustic Emissions (SFOAEs)
Distortion Product Otoacoustic Emissions (DPOAEs)
Instrumentation

CHAPTER II - REVIEW OF THE LITERATURE
Normative Data
Race and Gender Effects

CHAPTER HI - METHODS AND PROCEDURES

Subjects

General Experimental Protocol
Case History
Otoscopy
Middle Ear Testing
Audiometric Testing
Real Ear Measurements
Distortion Product Otoacoustic Emission Testing Parameters
Spontaneous Otoacoustic Emission Testing Parameters

Data Analysis

CHAPTER IV - RESULTS
Pure Tone Hearing Thresholds
Equivalent Ear Canal Volume
Height
Distortion Product Otoacoustic Emissions
Real Ear Unoccluded Responses
Spontaneous Otoacoustic Emissions
Normative Data

CHAPTER V - DISCUSSION AND SUMMARY
Sunmary of the Study
Discussion of Relevant Findings
Conclusions

Implications for Further Research

@dhAwN—‘I— é:

OO\I\)

l4
14
14
14
15
15
15
16
16
17
18

20
20
23
23
23
29
29
39

41
41
41
47
48

LIST OF TABLES

Table 1 - One-Way ANOVA of Hearing Thresholds (dB HL): All Groups

Table 2 - One-Way ANOVA of Ear Canal Volume (cm3): All Groups

Table 3 - One-Way ANOVA of Height (in): All Groups

Table 4 - General Linear Model method of DPOAE Amplitude (dB SPL): All
Groups

Table 5 - One-Way AN OVA of Real Ear Unoccluded Response: By Frequency

Table 6 - One-Way ANOVA of number of SOAEs

Table 7 - General Linear Model of DPOAEs with noise floor at 2 SD

Table 8 - General Linear Model of DPOAEs with and without SOAEs

50
51
52

53
54
55
56
57

LIST OF FIGURES

Figure 1 - Mean Pure Tone Hearing Thresholds Among Racial Groups of Women 21

Figure 2 - Mean Pure Tone Hearing Thresholds between Men and Women 22
Figure 3 - Equivalent Ear Canal Volume for All Groups 24
Figure 4 - Mean Height for All Groups 25
Figure 5 - Mean DPOAE Amplitudes of All Groups 26
Figure 6 - Mean DPOAE Amplitudes among Racial Groups of Women 27
Figure 7 - Mean DPOAE Amplitudes between Men and Women 28
Figure 8 - Mean DPOAE Amplitudes for All Four Groups Taking Into Account
the Noise Floor 31
Figure 9 - Mean Real Ear Unoccluded Response Among All Four Groups 32
Figure 10 - Mean Real Ear Unoccluded Response Among All Racial Groups of
Women 33
Figure 11 - REUR Response of All Four Groups at 2000 Hz 34
Figure 12 - REUR Response of All Four Groups at 8000 Hz 35
Figure 13 - SOAB Amplitude Distribution across Frequency among All Groups 36
Figure 14 - Number of SOAEs Occurring among All Groups 37
Figure 15 - Comparison of DPOAE Amplitudes between Persons Who Demonstrate
SOAEs and Those Who Do Not 38

Figure 16 - Normative DPOAE Amplitudes across Frequency for Women with One
Standard Deviation Indicated 4O

vii

CHAPTER I

INTRODUCTION

Otoacoustic emissions (OAEs) are acoustical responses that originate from a
normal functioning organ of Corti. They can be detected in the external auditory meatus
when the middle ear system is normal (Kemp, 1996). Brownell (1990) defined OAEs as
a normal byproduct of the outer hair cells of the cochlea that are generated in response to
an acoustic stimulus, or in the case of spontaneous OAEs (SOAEs), occur without
acoustic stimulation. OAEs were first discovered by Kemp in 1978, although Gold had
published a theoretical treatise on this topic in 1948. Since their discovery, OAEs have
elicited the curiosity of scientific researchers, as well as clinicians, of the possibility that
they may be used for research and clinical purposes. At the present time, there is
considerable focus on the clinical utility of OAEs in audiologic practice.

Otoacoustic Emission Pathway

OAEs originate in the cochlea when energy is produced in response to acoustical-
mechanical stimulation, or occur inherently in the normal functioning cochlea (e. g.,
SOAEs). The energy travels fiom the hair cells of the cochlea to the oval window, across
the ossicular chain, thus vibrating the tympanic membrane and causing an acoustical
wave to radiate out into the external auditory canal (Lonsbury-Martin and Martin, 1990).
Kemp (1996) stated that OAEs are not electrical in nature, but are vibratory, and are
created by the motion of the eardrum driven by the cochlea through the middle ear chain.
Miniature microphones placed into the external auditory canal are used to detect OAEs.

The recording microphone converts the vibratory energy into an electrical impulse and is

processed via computer analysis (Kemp, 1996).

Given that OAEs are not very high in amplitude, Kemp (1996) stated that OAEs
would perhaps go undetected if the ear canal was not acoustically sealed with a probe.
That is, closing ofl‘ the ear canal is essential for a more efficient system for compressing
and rarefying the trapped air that would otherwise escape from the ear canal.
Measurement of OAEs depend also on the condition of the middle ear cavity. The
middle car must function normally in order to accurately transmit these low energy
signals to the outer ear canal. If middle ear pathologies exist, OAEs are most likely to be
absent (Glattke and Kujawa, 1991). Measurements of otoacoustic emissions are non-
invasive and do not require active patient participation. They do, however, require that
the person being tested sit quietly so as not to introduce any unnecessary bodily noise,
thus, preventing accurate recordings of the emissions.

Types of Emissions

Emissions can be classified into two different, basic classes. They may occur
spontaneously (SOAEs) or in the presence of an evoking acoustic stimulus (EOAEs).
Bright (1996) defined SOAEs as toml signals of very low levels tint can be recorded in
the external ear canal without the use of any evoking stimulus. The prevalence of
recorded SOAEs varies depending on the instrumentation that is used (Bright, 1996).
Martin, Probst, and Lonsbury-Martin (1990) stated that SOAEs are present in
approximately 50% of normal hearing humans. Bright (1996) further stated that recent
data suggest that the prevalence of SOAEs is as high as 72% in persons with normal
hearing. In addition, it has been found that women exhibit a higher prevalence of SOAEs

than men (Bilger, Matthies, Hammel and Demorest, 1990; Martin et al., 1990; Whitehead

et al., 1993).

Gender differences as well as racial differences have been reported for SOAEs.
Whitehead et al. (1993) found significant differences in the SOAEs among three racial
groups. They found that Black individuals exhibited more SOAEs than either Asian or
White individuals. In addition, they noted a trend in which both the total number of
SOAEs and the average number of SOAEs per emitting ear increased with greater skin
pigmentation. They attributed their findings to different middle-ear conduction
properties between racial groups, specifically, differences in tympanometric gradient and
ear canal volume. The investigators acknowledged, however, that further study of these
differences was warranted. Although there is extensive literature on SOAEs, they are not
very useful clinically due to the lack of control over the frequency of the response and the
highly variable prevalence rate.

While SOAEs have not been found to be totally applicable in the clinic, EOAEs,
nevertheless, serve several clinical applications. There are currently three types of
EOAEs, transient evoked otoacoustic emissions (TEOAEs), stimulus frequency
otoacoustic emissions (SFOAEs), and distortion product otoacoustic emissions
(DPOAEs). There are notable difi‘erences among these classifications as to stimulus
parameters, latency of the response, and the degree to which a hearing loss can be
reliably detected.

Transient Evoked Otoacoustic Emissions (TEOAEs)

Transient evoked otoacoustic emissions (TEOAE) are aptly named for the

stimulus that is involved in its procedure. The stimulus consists of a transient acoustic

signal (e.g., click or short tone burst) and the response is recorded over a specific analysis

time of post-stimulation. The TEOAE analysis is based on a fast Fourier transform of the
response. TEOAEs are restricted to fiequencies fiom approximately 400 - 6000 Hz, and
are absent in patients with a hearing loss greater than 30 dB HTL (Lonsbury-Martin et al.,
1991). TEOAEs have a response-latency of approximately 5-20 ms from the time of the
stimulus and the response of the outer hair cells. In an earlier paper, Lonsbury-Martin
and Martin (1990) attributed the lag in time to a specific region within a hair cell that
responds to the evoking stimulus. Audiologists have found TEOAEs to be a very usefirl
screening tool for hearing impaired neonates, and in individuals who are unable or
unwilling to participate in conventional audiometric procedures.

Stimulus Frequency Otoacoustic Emissions (SFOAEs)

Stimulus Frequency Otoacoustic Emissions (SF OAEs) were explored as a
possible option for another test of cochlear function. The stimulus consist of a low-level
pure tone of continuous duration (Martin et al., 1990). The response to this stimulus is
recorded by slowly sweeping the frequency of the stimulus tone across a range of fiom
approximately 1000-2000 Hz (Lonsbury-Martin, Whitehead and Martin, 1991).
SFOAEs, however, are difficult to measure since the responses are embedded in the
stimulus. As SF OAEs were explored further, it was found that they are very similar to
TEOAEs. The similarities occurred within specific fiequency regions and at the level at
which the responses occurred. Due to the difficulty in measuring and interpreting
SFOAEs, as well as the similarity of SFOAEs to TEOAEs, SF OAEs were determined to
be a redundant test of cochlear function.

Distortion Product Otoacoustic Emissions (DPOAEs)

Distortion Product Otoacoustic Emissions (DPOAEs) are another type of evoked

OAE. The stimuli used to elicit DPOAEs are pure tones. What makes the DPOAE
different Item the TEOAE, other than the stimuli being pure tones, is that two tones are
introduced to the cochlea simultaneously. The two tones create an intermodulation
distortion response in which the signals are mechanically combined within the cochlea.
The response is not an exact replica of the stimulus, since it includes components that are
not present in the original stimuli (Lonsbury-Martin, Martin and Whitehead, 1996). The
pure tone stimuli that are used as eliciting stimuli are known as the “primary” tones.
Lonsbury-Martin et al. (1996, p.83) described these tones and their distortion products.

“...The lower-frequency pure tone is referred to as the fi ‘primary’, and its
[acoustic] level as L]; and the higher frequency pure-tone is referred to as the f2
‘primary’, and its [acoustic] level, as L2. The primaries are related in fiequency in that
the frequency separation of f2 from f1, commonly called the fz/f. ‘ratio’, is typically

around f; x 1.2 (i.e.: the primary tones are within one-third octave of each other). The
most fiequently measured acoustic intermodulation-distortion product is at the frequency

2fl-f2 (i.e.: the cubic difference tone), although the cochlea produces concurrent DPOAEs
at other fiequencies (e.g., fz-fi, 2f2-f., 3fi-2f2, etc.).”
It has been demonstrated that 2fl-f2 yields the largest emission in humans and other
mammals, which is why it is used as the defining distortion product (Lonsbury-Martin,
Whitehead and Martin, 1991). The distortion product emission occurs at 60-70 dB below
the level of the evoking stimuli (Lonsbury-Martin et al., 1996). The DPOAE, as well as
other OAEs, necessitate measurement by a small microphone placed in the ear canal.
DPOAEs can be detected in a frequency range fiom approximately 500 to 8000 Hz and
are absent in patients with hearing loss greater than approximately 50 dB HL (Lonsbury-
Martin et al., 1991).

While otoacoustic emissions can be interpreted as evidence of a normal

functioning cochlea (i.e., outer hair cell function), they are not a measure of hearing

sensitivity. Without knowing of an individual’s actual hearing threshold, only a

hypothesis can be made that pertains to threshold. Current applications of OAEs include
use as a screening tool in neonatal wards, screening for cochlear function while an
individual is taking ototoxic medication, and confirmation of pseudohypacusis.
Instrumentation

Instrumentation for DPOAEs is commercially available from a variety of
manufacturers (e. g., Bio-Logic Systems, Grason-Stadler Inc., Madsen Electronics,
Mimosa Acoustics and Otodynarnics Ltd.). Kemp presently holds the patent for
TEOAEs, which can only be found on the commercially available system from
Otodynamics Ltd. A few studies have used commercially available equipment to collect
data (Gorga et al., 1997; Homsby et al., 1996; Vinck et al., 1996). Other investigators
have used experimentally built systems to collect and report OAE data (Glattke and
Kujawa, 1991; Lasky, Snodgrass, and Hecox, 1994; Lonsbury-Martin, Harris, Stagner,
Hawkins and Martin, 1990; Lonsbury-Martin, McCoy, Whitehead, and Martin, 1993).
While a few investigators (Gorga et al., 1997; Homsby et al., 1996; Vinck et al., 1996)
conducted well-controlled studies, there was no information about the gender or race of
the subjects. More normative studies of DPOAEs are needed as pertains to race and

gender.

CHAPTER 11
REVIEW OF THE LITERATURE

Since the discovery of otoacoustic emissions (OAEs) in 1978 by Kemp,
investigators have identified clinical applications for these measurements. Studies have
attempted to reveal new information regarding cochlear filnction and the role of hair cells
in hearing (Avan, Bonfils, Loth, Teysson and Menguy, 1993; Brownell, 1990; Kemp,
1986; Wake, Anderson, Takeno, Mount and Harrison, 1996), in addition to understanding
the process of OAE generation (Lonsbury-Martin, Whitehead and Martin, 1993).
Furthermore, there is a wealth of literature about possible clinical applications that
include, but are not limited to, newborn hearing screening, monitoring of ototoxic drug
effects, and the quantification of cochlear firnction. The first step in the use of OAEs is
to identify pathology as related to is the definition of normal OAEs.
Normative Data

Several investigators have collected information on adult human subjects with
respect to normative data, but two problems have emerged (Gorga, Neely, Ohlrich,
Hoover, Redner and Peters, 1997; Vinck, De Vel, Xu and Van Cauwenberge, 1996).
First, the normative data were based on human versus animal results and only age effects,
if any, were included. Second, the measurements were made in sound treated rooms that
did not account for ambient room noise. Emissions that are recorded clinically are most
likely to be recorded in a more reverberant or less-controlled acoustically environment.

Gorga et a1 (1997), Homsby, Kelly and Hall (1996), and Vinck et al. (1996) used
commercially available equipment to collect normative data on adults. Gorga and his

colleagues (1997) used the Bio-Logic Scout system, while Homsby et al. (1996)

compared five systems (Bio-Logic Scout, Grason-Statler 60, Madsen Celesta, Mimosa,
and Virtual 330). Both studies measured DPOAE responses using normal hearing subject
populations. Vinck et al. (1996) used the Otodynarnics Ltd. system to collect DPOAE
data also on a normal hearing population. Although all studies were well-controlled,
none of the investigations considered possible racial and gender effects among their
respective subject population.

There are several studies suggesting that race and gender may have significant
effects on otoacoustic emissions, which may affect demonstrated normative data (Cacace,
McClelland, Weiner and McFarland, 1996; Gaskill and Brown, 1990; Lonsbury-Martin,
Whitehead and Martin, 1991; Moulin, Collet, Veullet and Morgon, 1993; Whitehead,
Kamal, Lonsbury-Martin and Martin, 1993). None of these studies used comerically-
available equipment.

In light of these shortcomings, the present investigation was undertaken to explore
and answer the following questions: (1) Are there significant differences in the DPOAEs
recorded in different racial groups? (2) Are there significant differences between
gender?

Race and Gender Effects

Overfield (1995) stated that there are three levels fiom which to view human
variation: race and sex, race and age, and individual. Brues (1977, p.1) stated that race is
defined as “ a division of a species which differs from other divisions by the frequency
with which certain hereditary traits appear among its members.” Webster’s Desk
Didi—mm (1990) defines race as “(1) A division ofthe human species characterized by a

more or less distinctive combination of physical traits tint are transmitted in descent. (2)

A group of tribes or peoples forming an etimic stock. (3) Any group or class of persons.”
These definitions suggest possible differences may be related to clinical instrumentation
used to identify cochlear function or pathology.

The earliest audiologic studies that identified gender and racial differences dealt
with air-conducted pure-tone threshold measures. Corso (1959) investigated age and
gender differences in air conducted pure-tone thresholds. He found that women have
more sensitive hearing than men. Post (1964) compared the hearing sensitivity between
Blacks and Whites. He found that only adult subjects exhibited differences in threshold
sensitivity, and that Blacks exhibited better thresholds than Whites at fiequencies above
2000 Hz. He suggested that these difference were perhaps due to hereditary factors.
However, occupational environment and other subject factors were not taken into
account.

Shepherd, Goldstein and Rosenblitt (1964) also found racial differences in
auditory sensitivity measured using the method-of-limits and Békesy audiometry. While
the method-of-limits indicated that White men and White women had better thresholds
than their respective Black counterparts at fi'equencies of 1000 and 2000 Hz, Békesy
audiometry identified no significant racial differences among men at 1000 Hz. White
women demonstrated significantly better thresholds across fiequencies than Black
women. The limitations of this study were difference in sizes of the groups studied and
the educational level of participants was a variable in one group, but not the other. They
concluded that the threshold differences of the women could be attributed to a more
sensitive auditory system among White women, regardless of the socio-economic level or

cultural habits of the group.

Royster, Royster and Thomas (1980, p. 553) reported on pure-tone thresholds
during industrial hearing testing in North Carolina. They found significant effects on
thresholds for both race and gender. They cited reasons for the differences by quoting
others: “[difi‘erences] are due to ‘actual inborn differences between the biophysical
systems’ rather than to cultural and environmental influences.” Their findings thus
indicated significant differences between the two groups. Groups with the most sensitive
hearing (lowest thresholds) were Black women, followed by White women, Black men
and White men. The difference in mean thresholds between Black and White men was
between 10-20 dB for frequencies above 2000 Hz. In addition, the difference in mean
thresholds between Black and White women was smaller and found to be between 5-10
dB for frequencies above 2000 Hz. The investigators stated that their findings concurred
with those of Post (1964). Royster et al. (1980) cited also information from the National
Health Survey of 1960-1962. The latter study analyzed thresholds of 7,700 subjects who
were 18-79 years of age. They found that Black men had better mean thresholds than
White men across frequencies from 500-6000 Hz. They further stated that Black women
had bettermeanthresholdsat 500,1000, and6000 HzthanWhite women. Theactual
differences in mean threshold values were not noted.

Racial differences in anatomical structure have been identified (Hajnis, Farkas,
Ngim, Lee and Venkatadri, 1994). These investigators studied certain head and neck
anatomical differences among three racial groups, Asian, Black, and White. Of the
several measurements performed, the ones most relevant to the present study are those of
the head and ears. The largest and longest sizes of the circumference and length of the

head appeared among Blacks. The smallest and shortest dimensions were measured in

10

the Asian population. For the width of the head, Asians exhibited the largest size, while
Blacks exhibited the smallest. The pinnae were the most narrow among the Asian group,
the longest among the White group, and the shortest and widest among the Black group.
However, ear canal size and resonance characteristics of the external auditory canal were
not investigated.

Two groups of investigators have studied differences in cochlear length. This
may have implications for differences in DPOAEs between race and gender. Sato, Sando
and Takahashi (1991) noted differences in cochlear length between genders. The authors
studied the cochlea in the temporal bones of age-matched subjects for both male and
female subjects. The ages of the individuals ranged from 1 day to 76 years. They found
that the mean cochlear length for males was significantly longer when compared to the
length in females. The cochlear length as a function of gender did not vary with postnatal
age, prompting the investigators to conclude that the cochlea fully develops in-utero.
This suggested that male neonates have a longer cochlear length than female neonates.
These findings may explain why emissions are greater in amplitude among females than
in males. That is, the travel time fi'om the cochlea to the ear canal will have an impact on
the emissions. This may perhaps manifest itself in the dampening of the emissions in
males during the outward propagation of the waves to the outer ear canal.

Kimberly, Brown and Eggermont (1993) studied cochlear traveling time delay
using DPOAE phase measurements. They found that there were gender difl‘erences in
wave delay, with males exhibiting significantly longer delays than females. They too
attributed this to the anatomical difference in cochlear length between genders, as noted

by Sato et al. (1991). Differences in amplitude of the emission were not reported by

11

Kimberly et al. (1993).

There have been few studies on the effects of gender on DPOAEs. Gaskill and
Brown (1990) compared the amplitudes of DPOAEs to auditory threshold measures of
both men and women. They found that women exhibit larger amplitudes than men at
frequencies fiom 1000-5000 Hz. These investigators found also similarities between
auditory threshold and the DPOAE amplitude, stating that DPOAEs would be a good
predictor of acoustical function Moulin, Collet, Veuillet and Morgon (1993) also studied
the effects of gender on DPOAEs. They found a significant difference in amplitude at
only one fi'equency (2000 Hz). They stated that the emissions were of a lower amplitude
when the person does not exhibit SOAEs. This suggests that SOAEs may enhance
DPOAE amplitude at corresponding frequencies. Cacace, McClelland, Weiner and
McFarland (1996) observed gender effects on DPOAEs across frequencies, confirming
the results of Gaskill and Brown (1990) and Lonsbury-Martin et al. (1991). These
investigators used the Otodynamics Ltd. ILO-92 commercial system for stimulus
generation as well as for response recording, measurement and analyses.

Currently, there are data on gender effects but none on racial effects as related to
DPOAE norms. However, one study did take race into account while investigating
SOAEs (Whitehead, Kamal, Lonsbury—Martin and Martin, 1993). These authors reported
significant differences in the occurrence of SOAEs between the two gender and the three
racial groups studied. They noted that Black individuals exhibited more SOAEs than
either the Asian or the White group. Black individuals emitted also more multiple
SOAEs. Since SOAEs are thought to enhance DPOAE amplitude (Moulin et al., 1993),

these racial and gender differences may also account for DPOAE differences. The

12

present study will investigate the effects of gender and race of both SOAEs and DPOAEs

in an attempt to further clarify this issue.

13

CHAPTER III
METHODS AND PROCEDURES
Subjects
The participants consisted of 40 normal hearing college-aged men and women.
Each subject participated on a volunteer basis and were members of the Michigan State
University Community. Four different groups were represented: Asian women, Black
women, White women, and White men. Each group consisted often participants. These
groups were chosen on the basis of past studies incorporating racial differences in
measurement of certain parts of the head (Hajnis et al., 1994; Overfield, 1985) and
differences found between SOAEs in different racial groups (Whitehead et al., 1993).
Comparison of the gender factor was based on measurements made of subjects of the
White group. Participants were between the ages of 18-35 years (mean = 23.03 years).
The average age for the four groups were similar (Asian women = 25.2 years, Black
women = 21.4 years, White women = 21.2 years, White men = 24.3 years). The test ear
was counterbalanced among participants to ensure equal distribution of left vs. right cars.
A typical test session lasted no more than 60 minutes.
General Experimental Protocol
Case History
All subjects completed a short case history to determine candidacy. The
questions investigated ethnic heritage, age, gender, height, hearing health history, family
history of hereditary hearing loss, significant noise exposure, history of significant

middle ear infections, and any past or current ototoxic drug use.

r3.

Otoscopy

Each subject underwent otoscopy using a video otoscope (Jedmed 80999-000,
Carrollton, TX). The otoscopic examination was performed to examine the conditions of
the outer ear canal and to confirm that it was fi'ee of occlusion due to cerumen or foreign
objects.
Middle Ear Testing

Tympanometry was performed using an immittance instrument to confirm normal
middle ear filnction (Grason-Stadler Inc., model 33, Milford, NH). Each subject was
seated in a comfortable chair and tests were performed in a quiet room. The immittance
meter recorded, tympanometry, static admittance and equivalent ear canal volume (V ea)-
Each subject exhibited normal values for all measurements (Margolis and Heller, 1987).
Audiometric Testing

Pure tone audiometry was performed to confirm that all participants demonstrated
hearing within normal limits. Testing was performed a sound-treated booth (Industrial
Acoustics Corporation, model 380, Bronx, NY). Each subject was seated in a
comfortable chair and headphones (TDH-49/Mx4lAR cushions) were placed over both
ears. Each subject was instructed to listen to a pure tone and indicate audibility by
raising their hand. Hearing thresholds were obtained at eight fiequencies (250, 500,
1000, 2000, 3000, 4000, 6000, and 8000 Hz), using air-conducted pure tone stimuli.
Hearing testing was performed using an audiometer (Grason-Stadler Inc., model 16,
Milford, NH), obtaining thresholds in increments of 5 or 10 dB steps (Hughson-Westlake
procedure). Hearing sensitivity was deemed normal when thresholds were better than 20

dB HL (ANSI, 1996).

15

Real Ear Measurements

A real car unoccluded response (REUR) and a real ear occluded response GIEOR)
were obtained using the CAB probe tip as the occlusion device (Hearing Aid Test System
Computer Controlled Real-Time Analyzer, F onix 6500, Frye Electronics Inc., Tigard,
OR; Probe Tip Variety Pack, cat. # G81 1700-9960, Gordon Stowe & Associates,
Wheeling, IL). These measurements demonstrated the resonance characteristics of
individual ears across the frequencies from 1000-8000 Hz, and were used to compare the
ear canal characteristics of each of the racial and gender groups. Each subject was seated
in a comfortable chair in a quiet room. A Velcro strip was placed around the head of the
subject and a small microphone was attached. A speaker was placed approximately one
meter away from the subject at a 45° angle to the microphone. The microphone measured
the amount of sound at the head and acted as the referent sound. A probe tube
microphone was then inserted into the ear canal of the subject to record the resonance
response with or without the emission probe tip inserted in the ear canal.
Distortion Product Otoacoustic Emission Testing Parameters

A commercially available EOAE instrument was used to measure DPOAEs
(ILO92, Otodynamics Ltd., London, England, UK). The software program was
controlled by a personal computer system (Zenith 486, St. Joseph, MI). The software is
capable of managing stimulus generation as well as recording and analyzing subsequent
emissions.

Two pure tone frequencies (f1 and f2) were delivered to the ear canal using a probe
tip device (Probe Tip Variety Pack, cat. G81 1700-9660, Gordon Stowe & Associates,

Wheeling, IL). The stimuli were presented simultaneously to the external ear canal

l6

through separate speakers to ensure that mixing of signals did not occur prior stimulating
the test car. The two frequencies elicited emissions at approximately 1000, 2000, 4000,
6000 and 8000 Hz, to produce the cubic distortion product (2f. -f2).

DPOAE stimulus levels (L. and L2, with L1> L2) were selected so that emissions
were elicited to unequal level stimuli as emission amplitude is larger to unequal intensity
levels (Lonsbury-Martin et al., 1996; Sun, Kim, Jung and Randolph, 1996). Sun et al.
(1996) noted also that emissions were more sensitive to cochlear dysfunction when the
stimuli are of unequal levels. Thus, the pure tone stimuli were presented at L. = 65 dB
SPL and L2 = 55 dB SPL, a ratio of 1.22 (Gorga et al., 1997).

DPOAEs were recorded using the “DP-gram” fornmt. This encompasses a wide
stimulus fiequency domain. For this study, each point of the DP-gram was obtained in 3
dB/per octave steps for the F2 fiequency between 1000 and 8000 Hz.

An emission was considered to be present if the response was 2 3 dB SPL greater
than the noise floor (Martin et al., 1990). The “probe fit function” of the software
assessed the fit of the probe in the ear canal to determine if the ambient noise level was
sufficient for accurate measurements to occur. Unlike tympanometry, OAEs do not
require a hermetic seal when obtaining emissions at or above 1000 Hz In order to obtain
accurate emissions for the purposes of this investigation, however, the probe was fit
snugly into the ear canal in order to maximally decrease the noise floor. The depth of the
probe was maintained constantly across all subjects to diminish variability of the data.
Spontaneous Otoacoustic Emission Testing Parameters

SOAEs were measured using the same commercially available equipment

(ILO92). The probe used to measure DPOAEs remained in the ear canal for purposes of

17

measuring SOAEs, and to keep the probe fit consistent. Measurement and frequency
analysis of emissions were completed using the same software program. An emission
was considered to present using the same criteria used for DPOAE measurements.
Data Analyses

Data were collected from racial and gender groups and were analyzed to identify
efi‘ects due to these differences. Amplitude of DPOAEs were combined within racial
and/or gender groups by fiequency, and analyzed using the General Linear Model.
Analysis of DPOAE response amplitudes over the 1000-8000 Hz fiequency range at the
distortion product intensity levels (65, 55 dB SPL) were performed in order to describe
the “DP-gram” response. All other data analysis was completed using a one-way
ANOVA. Statistical analysis of the DPOAE data was analyzed using SPSS (SPSS Inc.,
v.8.0, Chicago, IL). All other statistical analyses were computed using Minitab (Minitab
Inc., 1996, State College, Pa). Mean data were subsequemly graphed using Sigma Plot
(Jandel Corporation, 6.0, San Rafael, CA).

Kerlinger (1964) stated that the use of parametric statistics is preferable over non-
parametric statistics due to their increased power in finding significant differences. He
defines power as the probability that the null hypothesis will be rejected when it is
actually false. While parametric statistics are favorable, certain conditions must be met
before being applied. According to McCall (1994), there are four assumptions
underlying the use of the ANOVA: (1) the sample is made up of randomly chosen
independent samples, (2) the groups of scores are independent of each other, (3)
homogeneity of variance within the groups, and (4) each sample is drawn fiom a normal

population. All conditions named were met in the present investigation prior to the use of

18

the one-way AN OVA and the General Linear Model.

19

CHAPTER IV
RESULTS
There is considerable promise for the use of distortion product otoacoustic
emissions (DPOAE) as an audiological diagnostic tool. The purposes of present study
were to compare DPOAEs among different racial groups and gender using a
commercially available instrument and to establish normative DPOAE data. This study
explored and answered the following questions: (1) Are there significant differences in
DPOAEs recorded in different racial groups? (2) Are there significant differences in
DPOAEs between genders? Participants included individuals from three different racial
groups: Asian, Black and White. The question about gender was answered by comparing
DPOAE data between White men and White women.
Pure Tone Hearing Thresholds
Pure tone data were analyzed by a one-way AN OVA to determine whether there
were significant differences in hearing thresholds between groups. There was a
statistically significant difference between groups (p < 0.05) (see Figures 1 and 2). Post
hoc testing using the Tukey pairwise comparison (p < 0.05, df = 3, F = 4.39) indicated
that the difference existed at 6000 Hz between the men and all groups of women. There
were no significant difference in threshold between the groups of women. While
significant differences were observed between genders, each individual that was included

in the study demonstrated hearing thresholds within normal limits.

20

Mean Hearing Threshold (dB HL)

 

 

 

40*

504'

60.....

 

70 ‘ ' ' + Asian Women

 

 

 

 

+ Black Women 5 ' g A

80"." +WhiteWomen
90y — Criterion for Normal Hearing Thresholds ,_ ,,,,,,,,,,
100 ; } § } } i . ;

 

250 500 1000 2000 3000 4000

Frequency (Hz)

6000 8000

Figure 1. Mean pure tone hearing thresholds among racial groups of women

21

 

Mean Hearing Threshold (dB HL)

 

~10

 

 

 

 

 

 

 

0.4

10 ~~

201

404

504 -~

60- -.

7O-..+AsianWomen . ......
+BlackWomen 5 i i

80 ~ w + White Women
+WhiteMen

90 t ' — Criterion for Normal Hearing Thresholds

100 7" i T i ; l % ;

 

 

250 500 1000 2000 3000 4000 6000 8000

Frequency (Hz)

Figure 2. Mean pure tone hearing thresholds between men and women.

22

Equivalent Ear Canal Volume (V...)

These data were analyzed for ear canal volume using a one-way AN OVA.
Admittance values from tympanometric measurements were compared among the various
racial groups of women. There were statistically significant differences between groups
(p < 0.001, F = 10.82, df= 3) in equivalent ear canal volume (Va). Post hoc testing
(Tukey pairwise comparison, p <0.05) indicated a difference between men and all groups
of women (see Figure 3). There were no differences between the groups of women.
Height

Analysis of height using one-way ANOVA demonstrated a significant difference
between groups (p < 0.001). Post hoc testing (Tukey pairwise comparison) indicated a
significant difference between men and women at p < 0.05 level (see Figure 4). There
were no significant differences between the groups of women.

Distortion Product Otoacoustic Emissions

Of the 40 cars that were tested, each ear demonstrated definable emissions of
similar configuration (see Figure 5). Measurement of these emissions progressed
quickly, taking less than three minutes to complete. Data were recorded at nine response
frequencies across all groups. The frequencies were 1184, 1477, 1868, 2368, 2996, 3748,
4724, 5957 and 7495 Hz. There was a statistically significant difference between groups
(p < 0.05) (see Figure 6). Post hoc testing (Scheffé) isolated the difference to be between
White men and all women at the distortion product frequency 4724 Hz (see Figure 7).

There were no statistically significant differences among the racial groups of women.

23

Estimated Ear Canal Volume (cm3)

 

Groups

Figure 3. Equivalent ear canal volume for all groups, including standard
deviation bars.

24

Mean Height (in)

 

80

 

 

 

 

 

 

 

 

 

 

LL

7o - -
60- fl ,, ............
50_.. . . . . .. ...........
4o-
30... ., ., .. . . .......... . ................
20-. -Asian Women . . . ...............

mBlackWomen

-WhiteWomen
10‘ [:lWhiteMen ' """
0 ,L

Group

Figure 4. Mean height for all groups including standard deviation bars.

25

 

DPOAE Amplitude (dB SPL)

 

20

 

 

 

181~ .+ AsianWomen ........ 3 ......... I .......... .......... .........
—I— BlackWomen - ' j } f
16TH i +WhiteWomen ........ i ,,,,, ...... ......... .........
+WhiteMen ' . . ; -
14—-- -

 

 

 

 

 

I T I I I I I I

1184 1477 1868 2368 2996 3748 4724 5957 7495

Frequency (Hz)

Figure 5. Mean DPOAE amplitude for all groups across fiequencies.

26

 

 

18 “ ' + Asian Women
+ Black Women
16 ‘ ‘ + White Women

 

 

 

DPOAE Amplitude (dB SPL)

 

 

0 I I I I I I I I I

 

 

1184 1477 1868 2368 2996 3748 4724 5957 7495

Frequency (Hz)

Figure 6. Mean DPOAE amplitude among racial groups of women.

27

 

14

DPOAE Amplitude (dB SPL)

 

 

 

 

 

 

I I I I I I I r I

 

 

1184 1477 1868 2368 2996 3748 4724 5957 7495

Frequency (Hz)

Figure 7. Mean DPOAE amplitude between men and women.

28

..
11‘

. \
\Vk

..iKiQ

In addition to the above analyses, the data were analyzed by accounting for the noise
floor of the DPOAE responses (see Figure 8). This increased the emission amplitude for
all frequencies.

Real Ear Unoccluded Response

Real ear unoccluded responses were analyzed by one-way AN OVA to determine
differences in ear canal resonance. Statistical analysis revealed significant differences
between groups (p < 0.01, F = 4.67, df= 3) (see Figure 9). Post hoc testing (Tukey
pairwise comparison) found the differences to exist between men and all groups of
women at 2000 Hz (p < 0.05) (see Figure 9). There were no statistically significant
difi‘erences between the groups of women (see Figure 10). Upon closer examination at
2000 Hz, no significant difference is confirmed at this fiequency (see Figure 11). At first
glance, one may see a significant difference at 8000 Hz among the women; however, the
difference did not reach statistical significance (see Figure 12).

Real ear occluded response measurement verified the probe fit in the ear canal. If
resonance qualities of the ear canal decreased, the probe fit was considered adequate.
Visual inspection of these data insured a good probe fit, thus, no firrther analyses seemed
appmpfiate-

Spontaneous Otoacoustic Emissions

Of the 40 participants tested, 58% (23/40) exhibited SOAEs. Of the women, 70%
(21/30) exhibited SOAEs at fi'equencies ranging fiom 745 Hz to 5301 Hz , and at levels
from -30.2 dB SPL to 1.1 dB SPL (see Figure 13). The White men who exhibited
SOAEs showed multiple emissions, i.e., more than one emission occurred across the

fiequency spectrum Among the Asian women, six individuals exhibited SOAEs. Of this

29

group, three of the six individuals exhibited multiple SOAEs, thus, four of the Asian
women did not exhibit any SOAEs. Among the thk women, eight individuals
demonstrated SOAEs, and seven of these exhibited multiple SOAEs. Among the White
women, seven individuals demonstrated SOAEs, while five exhibited multiple SOAEs.
Comparing the number of SOAEs present in each of the four groups, no statistically
significant differences were found (see Figure 14). It is interesting to note, however, that
among all responses, there was only one fiequency match of the SOAEs to DPOAE
responses. This occurred for the Black women at the frequency of 1477 Hz, with an
amplitude of — 4.7 dB SPL. There were, what appear to be 14 close matches in fiequency
among all groups, with more matches among the women than among men due to the
increased fiequency of SOAEs in these groups (within 8-71 Hz of a DPOAE response).
The majority of SOAEs occurred within the middle to lower fiequency range (1000-2000
Hz) (see Figure 13). . Additionally, a comparison of DPOAE amplitude was made
between individuals who have SOAEs and those who did not exhibit such a response (see

Figure15). This indicates that DPOAEs are enhanced when SOAEs are present.

30

 

28“

26-
2.,_.__ .. . .

22_.. i. .. .......

     
     

 

+ Asian women
' —I- Black women
+ White women
- + White men

DPOAE Amplitude (dB SPL)

 

 

1184 1447 1868 2368 2996 3748 4724 5957 7495

Frequency (Hz)

Figure 8. Mean DPOAE amplitudes for all four groups across an frequencies
Accounting for the noise floor.

31

Real Ear Unoccluded Response (dB SPL)

24
22
20
18
16
14
12
1O

owe-moo

 

 

 

 

 

 

 

 

 

“ """ + Asian Women
_ ...... —I— Black Women

+ White Women
‘ """ + White Men

: .. .....——-————'
T "‘T. - : T
250 500 1000 2000 4000 8000
Frequency (Hz)

Figure 9. Mean real car unoccluded responses among all four groups.

32

 

Real Ear Unoccluded Response (dB SPL)

 

 

 
 
 
 

20 4- + Asian Women
+ Black Women
18 " + White Women

 

 

 

 

 

 

250 500 1000 2000 4000 8000

Frequency (Hz)

Figure 10. Mean real car unoccluded response among all racial groups of women.

33

 

20

18 - .......... 1. ._ , g, M. , ............

1,. i . , i l ...........

12

 

.................................................................

 

 

1
,LL
r_i

l O Asianwomen

: ._ 1 1 I Blackwomen

104 ......... .......... ‘ Whitewomen
- ~ ‘ ' O Whitemen

 

 

Real Ear Unoccluded Response (dB SPL)

 

 

 

 

 

 

8 i i 1' i i i i

1991 1992 1993 1994 1995 1996 1997 1998 1999

Frequency (Hz)

Figure l 1. REUR response of all four groups at 2000 Hz, data points are ofi‘set
for greater clarity.

34

Real Ear Unoccluded Response (dB SPL)

 

N
O

 

 

 

 

 

 

 

 

 

 

 

15 ... l ....... l. . T .........
A
,0- l __ ................................
fl l
5 a ......................................................................................
‘5 0 Asian women
0 z ............... I Black women ..... ‘ L ..........
A White women g
‘_L 0 White men
-5 . , T i e i ;
1991 1992 1993 1994 1995 1996 1997 1998 1999
Frequency (Hz)

Figure 12. REUR response of all four groups at 8000 Hz, data points are offset

for greater clarity.

35

SOAE Amplitude (dB SPL)

 

 

 

 

 

 

 

 

0- . .............. O AsianWomen
3 . ' 9 I BlackWomen
i . I ' O WhiteWomen
-5 u .... .- ........................... . ..... . White Men
3 c
-10 _ 1'.i| .,,.., .. H .........
1' 9 . I I: I
-15.. ............ i. .E.. ....... .L ..., ......
I - O ‘
O ,0 .
o '. i
-20 - .h . ..lbi,. ‘.
‘5 .0I . .
, O _
25 '0:Ii l 1 .
. ' I I i
. Oi .
-30. .3 ......... . .................... . .......... 5 .............
-35 i i f i i
0 1000 2000 3000 4000 5000
Frequency (Hz)

Figure 13. SOAE amplitude distribution across frequency among all groups.

36

6000

Number of SOAEs Present

30

 

25*

20-

- Asian Women
[:1 Black Women

- White Women
C] White Men

 

 

 

 

 

 

Figure 14. Number of SOAEs occurring among all groups.

37

Group

 

 

26

 

 

 

 

 

 

 

 

3

LL

(0

CO

3

a)

‘o

3

"d

E

<

UJ

<

0

CL

0 E .

: : : : 5 —¢— SOAEs present

12 _,, . I, ., ...... .. ...... ‘ . NOSOAESpI'esenI
10 i i i

1184 1447 1868 2368 2996 3748 4724 5957 7495

Frequency (Hz)

Figure 15. Comparison of DPOAE amplitude, including the noise floor, between
persons that do or do not demonstrate SOAEs.

38

Normative Data

While it would not be prudent to over emphasize the overall normative data found
in this study due sample size, a comment relating DPOAE results of college age women
can be stated. Normative data that has been compiled by researchers has included both
men and women. Gorga and colleagues (1997) included subjects within a wide age range
(1.3-96.5 years) for their study. The data fi'om the current study fall within the normative
data compiled by other investigators (Gorga et al, 1997; Vinck et aL 1996). Figin'e 16
demonstrated the DPOAE amplitude across frequencies (+/- one standard deviation).
The response amplitudes appear to be high, but this is not surprising. This study
incorporated more women than men (30/40), and the previous data indicated that women
have larger amplitude emissions than men. In addition, the ages of the women are within

a more specific range, other researchers included individuals with a wider age.

39

 

20

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

i i
i l
A 15 _‘ ‘T’ ....................
.1
CL _
‘0 l
m l
B
Q) 10 ., .............. .... ..................
'o
3
E
< 5 d ..................................................
LIJ
<
O
0.
D O~ ...........................................
+ Men
+ Women
‘5 I I I I I I I I I

 

 

1184 1477 1868 2368 2996 3748 4724 5957 7495

Frequency (Hz)

Figure 16. Normative DPOAE amplitudes across frequency for women with one
standard deviation indicated.

40

CHAPTER V

DISCUSSION AND SUMMARY

Summary of the Study

The present study investigated the effects of racial and gender differences on
DPOAE response amplitude. It was surmised that racial and gender differences might
need to be accounted for when establishing normative data. Due to the differences in the
shapes of the facial area among racial groups, it was believed that differences in ear canal
resonance could contribute to differences in DPOAE responses (Hajnis, F arkas, Ngim,
Lee and Venkatadri, 1994). Three racial groups were selected to take part in the study;
Asian, Black and White individuals. These groups were chosen based upon prior
research that indicated that racial and gender differences can be seen in SOAEs. Results
reported in the literature indicated gender effects are present in the DPOAE amplitude.
Thus far, explanations as to these differences have been speculative as to why these
occur. In order to compare gender differences, White men were included as participants
in the study.
Discussion of Relevant Findings

Data analysis of pure tone hearing thresholds found no significant differences
among racial groups of women. Although these data are in contrast to what was found in
prior research, it is not known if individuals in these studies demonstrated thresholds
within normal limits as did the subjects in this current study (Post, 1964; Shepherd,

Goldstein and Rosenblilt, 1964; Royster, Royster and Thomas, 1980). Post (1964) found

41

that Blacks exhibited better overall thresholds than their white counterparts. Shepherd,
Goldstein and Rosenblilt (1964) reported conflicting results. They found that White men
and White women demonstrated better hearing thresholds than Black men and Black
women. Royster, Royster and Thomas (1980) observed similar data of that of Post
(1964), i.e., Black individuals demonstrated better hearing thresholds than White
individuals. The results reported in the above studies were not an agreement with the
results of the present study. This difference could be due to hearing losses found in the
persons included in these prior studies, or to the differences between measurement
techniques and equipment used. In addition, the small sample size that was used in the
current study was smaller than any of the previous investigations. No racial differences
were found in hearing thresholds for the current data.

Analysis of pure tone thresholds found statistically significant differences
between gender at only one frequency (6000 Hz). One might assume noise exposure to
be a factor between these two groups, however, all participants demonstrated hearing
thresholds within normal limits. Most participants across all groups did indicate
exposure to loud sounds at one time or another (e.g., at loud concerts). The results of the
literature state that there is a gender effect among hearing thresholds across fiequencies.
Corso ( 1959) found that women have more sensitive hearing than men. Additionally,
Royster, Royster and Thomas (1980) also found that women exhibited better thresholds
than men. Again, these reports may be speculative in comparison to the current study
due to technological advances and stricter guidelines established since these studies were
performed.

In the present study, men exhibited a larger ear canal volume than women. This

42

is consistent with previous findings (Zwislocki, 1970). Zwislocki stated that the average
ear canal length is approximately 5% greater in men than women and the cross-sectional
area is approximately 10% greater. Further, it has been documented that a gender effect
related to ear canal volume is real (Margolis and Heller, 1987). White men have
DPOAEs of lower amplitude than White women, and this could be due to the larger ear
canal volume. While the DPOAE is of cochlear origin, ear canal volume by itself may
interact with the amplitude of the DPOAE causing them to be significantly reduced in
men. At the present time there is no information published in the literature relating ear
canal volume to the gender effect seen in the present DPOAE amplitude.

Time were no significant difl‘erences in the DPOAE amplitude among the racial
groups of women. There were, however, significant gender effects, and these findings
concur with the results in the literature. Gender differences in these data were seen at the
frequency 4724 Hz. Even though a statistical difference was not documented at multiple
fiequencies, one can identify a definite difference of 2-7 dB between the DPOAE
amplitude responses across all frequencies between men and women (see Figm'e7).
There is still some discussion as to why gender efiects occur. Hall, Baer, Chase and
Schwaber (1994) attribute gender effects evident in DPOAEs to minor differences in
detection threshold. While it could be argued that a decrease in hearing thresholds might
contribute to differences in DPOAEs, this does not explain the DPOAE difierence seen in
these data, as there was only one fiequency in the pure tone audiograrn that was
significantly different between genders. If this was indeed true, one would expect the
DPOAE difference to occur in close proximity to the frequency where the hearing

threshold differences occurred. Moulin et al. (1993) attribute gender effects evident in

43

TEOAEs and DPOAEs to a higher incidence of SOAEs in women (58%) than men
(22%). Similarly, Bilger, Mathies, Hammel and DeMorst (1990) agree, stating that more
women than men exhibit SOAEs, which can contribute to increased DPOAEs. A
comparison of DPOAE amplitude was made between individuals who have SOAEs and
those who did not exhibit such a response (we Figure15). These data support the notion
that SOAEs do increase the DPOAE amplitude. The results of the present study
indicated that women have more SOAEs than men. This was true among the racial
groups tested, in that each female racial group demonstrated more SOAEs than the White
men. It would be of interest to select men from the other corresponding racial groups to
determine if this may be true for all racial groups. Perhaps the difference in DPOAE
amplitude is due to the increased ear canal volume in men, which could decrease the
amplitude of the emission before reaching the probe tip in the outer ear canal. Another
plausible explanation to be considered is cochlear length of men is longer than in women
that (Sato, Sando and Takahashi, 1991). These investigators suggested a dampening of
the emission on outward propagation due to the longer outward travel time of the
emission. However, this effect could be attributable to the combination of all of these
factors that contribute to the difference in DPOAE amplitude observed in the present
study. The overall gender effect found in the DPOAE amplitude data concur with Gaskill
and Brown (1990) who found gender effects for the fiequencies between 1000-5000 Hz,
with DPOAE amplitudes being larger in women than in men.

Data analysis of real car measurement found no statistically significant effects
among racial groups of women. Since no racial effects were evident in the DPOAE data,

no further analysis of ear canal resonance seemed warranted.

It has been shown that probe placement is crucial to accurate results with real ear
measurement (Dirks, Ahlstrom and Eisenber, 1996; Shaw, 1974). In this study, the probe
microphone was placed in each participant’s ear at the mid-canal position. The probe
microphone was inserted until the participant felt it was uncomfortable, typically the
bony portion of the ear canal. The probe was then pulled out slightly for the participants’
comfort and the REUR/REOR measurements were made. The results of these data are
not likely to be relevant to the differences evident in the DPOAEs. If the differences in
resonance were evident at higher frequencies, ear canal resonance may imve an effect on
DPOAE responses. If differences were seen at more than one frequency, between all
groups, one would be more inclined to consider ear canal resonance characteristics to
play a role in affecting DPOAE responses.

The present study did not find any racial differences 5 regards SOAEs. These
findings do not agree with those of Whitehead et al. (1993) who did find a racial effect.
Perhaps this could be attributed to a difference in subject populations. Whitehead et al.
(1993) used both men and women of different racial groups, whereas this study included
only women. In addition, they studied both ears of all subjects, while in the present study
only one ear was tested.

Although the majority of SOAEs fell within the middle to lower frequency range
(1000-2000 Hz) (see Figure 10), SOAEs do have an effect on DPOAE amplitude. While
only one response out of all 73 SOAEs measured occurred at the same frequency as the
DPOAE response, there appears to be an overall amplifying effect when SOAEs are
present (see Figure 15). The perfect match between SOAEs and DPOAEs occurred at

1477 Hz at an amplitude of - 4.7 dB SPL. In order for the DPOAE to increase in

45

amplitude, the SOAE has to occur at the same fiequency or within close proximity to the
response fiequency, thus creating a superimposition of waves. This finding is consistent
with that found by Bilger et al. (1990), Penner and Zhang (1997), and Whitehead et a1.
(1993). The overall SOAE records exhibited multiple responses at a variety of lower
frequencies, rather than at the high frequencies. While it could be argued that the SOAE
response seen here affect the DPOAE response, the fi'equencies at which the SOAEs
occurred did not entirely match the DPOAE response fi'equencies. Yet, there is a definite
indication that SOAEs tend to increase the DPOAE amplitude.

Whether the presence of SOAEs contribute to the increased DPOAE amplitude
among the women is still speculative. Weir, Pasanen and McFadden (1988) state that
SOAEs increase the amplitude of the DPOAE response when the frequencies are in close
proximity. They found this difference to occur while using low levels of the evoking
stimuli for the DPOAE. Weir et al. (1988) studied DPOAE using a growth function
format, which measures the emission at different intensity levels at only one specific
frequency. They stated that the SOAE has an effect on the DPOAE when they are within
approximately 100 Hz of each other. This present investigation found fourteen SOAEs to
be present within 100 Hz of the DPOAE response among all participants. This
assumption could explain the gender efl‘ect demonstrated in the current data for the lower
frequencies. While most women had multiple SOAEs, the majority of these SOAEs
occurred at the mid-range of the spectrum (1000-2000 Hz), while the significant
difference seen between men and women occurred at a higher frequency, there was an
overall difierence between genders. Of the 40 participants, 23 of whom demonstrated

SOAEs, only one person had an exact frequency match between SOAE and DPOAE

46

frequency. The DPOAE primary levels in this study were not similar, nor were the
measuring conditions of the SOAE response, thus, these data cannot be readily compared
to the results observed by Weir et al. (1988).
Conclusions

The development of otoacoustic emissions has greatly aided the audiologist’s role
in identifying site-of-lesions in persons with hearing loss. It is important to establish
normative data on new equipment in order to compare data for accurate conclusions
about possible pathology. This study examined the possible racial and gender effects on
DPOAEs using normal hearing participants from a university population. The following
questions were asked:
(1) Are there significant differences among DPOAEs recorded in different racial groups?
(2) Are there significant differences among DPOAEs recorded between gender?

The population used for the present investigation included men and women of a
specific age group. Although no effects among the racial groups was evident in the
DPOAE responses, this study confirmed that gender effects are important to consider and
should be studied within a larger population. It has been suggested that gender efi‘ects
can be accounted for by the noted difference in cochlear length, which may decrease the
amplitude of an emission during outward propagation (Sato et al., 1991). Moreover, the
gender effect may be influenced by increased ear canal volume seen between men and
women. That is, the increase in ear canal size may have an effect on the recording of the
OAE in the ear canal, thus, resulting in the gender differences that have been previously
documented. The effect may also be influenced by the presence of SOAEs, which was

observed in the current study. All of these factors, either single or in combination, may

47

influence the amplitude of the DPOAE emissions between genders. Thus, gender
differences need to be considered when establishing normative data for clinical use.
Implications for Further Research

The purpose of this study was to explore the effecst of race and gender on
DPOAEs in order to establish normative data for clinical use. The present study used
commercially available software and test equipment in order to elicit, record and measure
otoacoustic emissions. In addition, all of the methods used in the present investigation
included those used in routine clinical practice. Based on these findings, there are several
suggestions for firture research:

1) The sample population should include male participants of the several racial

groups.

2) EOAE measurements of both ears of all participants should be included.

3) DPOAE measures should include DP-growth function measures at several

intensity levels at a single frequency to more closely correlate those data to SOAE

data.

4) Real ear measurements should employ a more accurate method for

measurement of ear canal resonance in order to determine much smaller

differences that may be present, since at this time one is unable to measure it

accurately without causing physical stress to the participant.

5) Measurements of children similar to those used in the present study to identify

a gender effect at younger ages and to establish clinical norms.

48

APPENDIX

Table 1. One-Way ANOVA of Hearing Thresholds (dB HL): All Groups.

Fr uenc

250
500
1000
2000
3000
4000
6000
8000

*" p<0.01

wwwwwwwwfi;

50

E
1.82
1.19
0.60
0.45
1.21
2.80
4.39
1.67

P
0.161
0.328
0.622
0.720
0.321
0.053
0.010”
0.191

 

Table 2. One-Way ANOVA of Ear Canal Volume (cm3): All Groups.

F P

Source ._ _.
10.82 0.000"

Group

WIS;

.. p<0.001

51

Table 3. One-Way ANOVA of Height (in): All Groups.

F P

Source _ __
29.27 0.000"

Group

Calf};

** p<0.001

52

Table 4. General Linear Model method of DPOAE Amplitude (dB SPL): All Groups.

Source if E .13
Group 3 3.835 0.018

* p<0.05

53

Table 5. One-Way ANOVA of Real Ear Unoccluded Response: By Frequency.

Frequency (Hz)
250
500
1000
2000
4000
8000

* p<0.01

MMMMMU’IIQ.’

54

E
1.69
1.18
0.69
4.67
0.09
1.27

P
0.187
0.330
0.562
0.007*
0.967
0.298

Table 6. One-Way ANOVA of the number of SOAEs.

 

 

F

Source

 

or IE;

 

 

Group 2.54

 

55

 

 

'3 'l"U

Table 7. General Linear Model of DPOAEs with misc floor at 2 SD.

 

 

Source

F

 

Group

 

d__r
3

 

2.535

'5 1‘1!

 

 

56

 

Table 8. General Linear Model of DPOAEs with and without SOAEs.

 

 

Source df F

I'U

 

 

 

 

Group 1 5.678 0.022

 

* p<0.05

57

 

BIBLIOGRAPHY

BIBLIOGRAPHY

American National Standards Institute (1996). Specifications for audiometers
(ANSI S3.6-1989). New York, NY: ANSI.

Avan, P., Bonfils, P., Loth, D., Teyssou, M., & Menguy, C., (1993). Exploration
of cochlear function by otoacoustic emissions: relationship to pure-tone audiometry.
Proggss in Brain Research 97. 67-75.

Bilger, R.C., Matthies, M.L., Hammel, DR, and Demorest, ME. (1990). Genetic
implications of gender differences in the prevalence of spontaneous otoacoustic emissions.
Journal of Speech and Hem Research. 33. 418-432.

Bright, KE., (1996). Spontaneous otoacoustic emissions. In M.S. Robinette &
T.J.Glattke (Eds), Otoacoustic emissions: clinical applications (pp.46-59). New York,
NY: Thienre.

Brownell, W.E., (1990). Outer hair cell electromotility and otoacoustic emissions.
Ear and He_ar_mg' , 11(2), 82-92.

Brues, AM. (1977). People and Races. New York: Macmillian.

Cacace, A.T., McClelland, W.A., Weiner, J., and McFarland, DJ. (1996).
Individual differences and the reliability of 2fl -fl distortion-product otoacoustic
emissions: Effects of time-of -day, stimulus variables, and gender. Journ_al of Speech and

Hearing Research, 39,1138-1148.

Cianfione,G., Mattia, M., Altissirni, G., and Turchetta, R. (1990). Distortion
product otoacoustic emissions and spontaneous otoacoustic emission suppression in
hunmns. In F. Grandori, G. Cianfione & D.T. Kemp (Eds), Cochlear mechanisms m
otoacoustic emissipm (pp. 126-138). Basel: Karger

Corso, LP. (1959). Age and sex differences in pure-tone thresholds. Journal of the
AcousthaI Society of Ameripg 31(4), 498-507.

Culpepper, NE. (1996). Neonatal screening via evoked otoacoustic emissions. In

M.S Robinette & TJ. Glattke (Eds), Otoacofiic emissions: clinical applications (pp.
233-270). New York, NY: Thieme.

Dirks, D.D., Ahlstrom, J.B., & Eisenberg, LS. (1996). Comparison of probe

insertion methods on estimates of ear canal SPL. Journal of th_e American Academy of
Audiology, 7, 31-38.

59

Doyle, K.J., Burggraaff, B., Fujikawa, S., Kim, J., & MacArthur, C.J. (1997).
Neonatal hearing screening with otoscopy, auditory brainstem response and otoacoustic
emissions. Otplarypgology Head and Neck Surgery. 116(6 pt. 1), 597-603.

Gaskill, SA, & Brown, AM. (1990). The behavior of the acoustic distortion
product, 2fl -Q, from the human ear and its relation to auditory sensitivity. Jom of the
' S i of ' 88(2), 821-839.

Gaskill, S.A., & Brown, AM. (1993). Comparing the level of the acoustic
distortion product 2fi-f2 with behavioral threshold audiograms from annual-hearing and
hearing-impaired ears. British 19mm of Audiology, 27(6), 397-407.

Glattke,T.J., & Kujawa, S.G., (1991). Otoacoustic emissions. A_m§rican Jom of
Audiology, 1, 29-40.

Gold, T. (1948). Hearing 11. The physical basis of the action of the cochlea.
Proceedings of the Royal Society of 14mm, Series B 135, 492-498.

Gorga, M.P., Neely, S.P., Ohh‘ich, 8., Hoover, B., Redner, J., & Peters, J. (1997).
From laboratory to clinic: A large scale study of distortion product otoacoustic emissions
inears with norml hearing and ears withhearing loss. Bar and Hm’ , 18(6), 440—455.

Hajnis, K., Farkas, L.G., Ngim, R.C.K, Lee, S.T., & Venkatadri, G. (1994).
Racial and ethnic morphometric differences in the craniofacial complex. In L.G. Farkas
(Ed). What-1mm (pp. 201-218). New York. NY: Raven Press.

Homsby, 8., Kelly, T., & Hall, J.W. (1996). Normative data for five FDA-
approved distortion product OAE systems. H ' Jo 49(9), 39-46.

Hurley, RM., & Musiek, RE. (1994). Effectiveness of transient-evoked
otoacoustic emissions (TEOAEs) in predicting hearing level. Journal of th_e American
Academy of Audiplogy, 5, 195-203.

Kemp, D.T. (1978). Stimulated acoustic emissions fiom within the human auditory
system. 152mm,! of the Amuig Society of Artie—r191 fl, 1386-1391.

Kemp, D.T., (1986). Otoacoustic emissions, travelling waves and cochlear
mechanisms. MW 95-104.

Kemp, D.T. (1996). Otoacoustic emissions in perspective. In M.S. Robinette &

T.J. Glattke (Eds), Otoacoustic emissions: c_lmlc_a_l' ' applicatm' 113 (pp. 1-21). New York,
NY: Thieme.

 

Kemp, D.T., Ryan, S. and Bray, P. (1990). Guide to the effective use of
otoacoustic emissions. Ear a_n_d Hm’ , 11(2), 93-105.

60

Kerlinger, F .N. (1965). Foundations of behavioral research. New York, NY: Holt,
Rinehart and Winston.

Kimberly, B.P., Brown, D.K., & Eggermont, J.J. (1993). Measuring human
cochlear traveling wave delay using distortion product emission phase responses. Journal
of the Acousthzal Society of Amerigg, 94(3), 1343-1350.

Lasky, RE., Snodgrass, E., and Hecox, K. (1994). Distortion-product otoacoustic
emission input/output functions as a firnction of frequency in human adults. Journal pf the
American Academy of Audiology, 5, 183-194.

Leonard, G., Smurzynski, J., Jung, M.D., & Kim DO. (1990). Evaluation of
distortion product otoacoustic emissions as a basis for the objective clinical assessment of
cochlear function. In F. Grandori, G. Cianfrone, & D.T. Kemp (Eds), Advances in
Audiology: Vol 7. CochlegMechanisms and Otoacoustip Emissions (pp. 139-148).
Basel, Karger.

 

Lonsbury-Martin, B.L., Harris, F .P., Stagner, B.B., Hawkins, M.D., and Martin,
GK. (1990). Distortion-product emissions in humans: 1.. Basic properties in normally

hearing subjects. Annals of Otology, Rhinology and L_armgology, 99(5, pt. 2) (Suppl.
147), 3-14.

Lonsbury-Martin, B.L., & Martin, GK, (1990). The clinical utility of distortion-
product otoacoustic emissions. Bar and Hm’ , 11(2), 144-154.

Lonsbury-Martin, B.L., Whitehead, M.L., & Martin, GK. (1991). Clinical
applications of otoacoustic emissions. Journal of Speech and ng' Research. 34. 964-
981.

Lonsbury-Martin, B.L., McCoy, M.J., Whitehead, M.L. & Martin, GK. (1993).
Clinical testing of distortion-product otoacoustic emissions. Bar and Hear_1ng° , 1(1), 11-22.

Lonsbury-Martin, B.L., Whitehead, M.L., & Martin, OK. (1993). Distortion-
product otoacoustic emission in normal and impaired ears: insight into generation
processes. Progress in Brain Research, 97, 77-90.

Lonsbury-Martin, B.L., Martin, G.K., & Whitehead, ML. (1996). Distortion
product otoacoustic emissions. In M.S. Robinette & T.J. Glattke (Eds), Otoacoustic
emissions: clinical applications (pp. 83-109). New York, NY: Thieme.

McCall, Robert B., (1994). Fundamental statistics for behavioral sciences (6th ed.).
New York: Hardcourt Brace.

61

Margolis, RH. & Heller, J.W. (1987). Screening tympanometry: Criteria for
medical referral. Audiology, 26, 197-208.

Martin, G.K., Probst, R., and Lonsbury-Martin, BL. (1990). Otoacoustic
emissions in human ears: normative findings. Ear and Hear_mg' , 11(2), 106-120.

Moulin, A., Collet, L., Veullet, E., & Morgon, A. (1993). Interrelations between
transiently evoked otoacoustic emissions, spontaneous otoacoustic emissions and acoustic
distortion products in normally hearing subjects. Hearing Research, 65, 216-233.

Moulin, A., Bera, J., & Collet, L. (1994). Distortion product otoacoustic emissions
and sensorineural hearing loss. Audiology, 33, 305-326.

Nakamura, M., Yarnasoba, T., and Kaga, K. (1997). Clumges in otoacoustic
emissions in patients with idiopathic sudden deafness. Audiology, 36, 121-135.

Norton, S.J., & \Vlden, J.E. (1990). Evoked otoacoustic emissions in normal
hearing infants and children: emerging data and issues. Ear and Hear—mg' , 11(2),121-127.

Overfield, T. (1985). Biologic Variation in Health and Illness: race, age and sex
differences Menlo Park, CA: Addison-Wesley.

Overfield, T. (1995). Biologic Variation in Healtlfld Iflnpss: race, age, and sex
differences (Rev. ed.). Boca Raton, FL: CRC Press.

Patuzzi, R. (1993). Otoacoustic emissions and the categorization of cochlear and
retro-cochlear lesions. British Journal of Audiology, 27, 91-95.

Penner, M.J. & Zhang, T. (1997). Prevalence of spontaneous otoacoustic
emissions in adults revisited. ng' Research 103, 28-34.

Post, RH. (1964). Hearing acuity variation among negros and whites. Eugenics
QuAteriy, 11,(2), 65-81.

Probst, R., Harris, F.P., & Hauser, R., (1993). Clinical monitoring using
otoacoustic emissions. Brit'gh Journ_al of Audiology, 27, 85-90.

Royster, L.H., Royster, J.D., & Thomas, W.G. (1980). Representative hearing
levels by race and sex in North Carolina industry. Journ_al of the American Acoustical
Society of Ameripa, 68(2), 551-566.

Sato, H., Sando, I., & Takahashi, H. (1991). Sexual dimorphism and development
of the human cochlea. Acta OtoMgologica (Stockholm), 111, 1037-1040.

62

Shepherd, D.C., Goldstein, R. and Rosenbltlt, B. (1964). Race difference in
auditory sensitivity. Journal of Smech and Hearr__ng° Research. 7. 389-397.

Strickland, E., Burns, EM. and Tubis, A (1985). Incidence of spontaneous
otoacoustic emissions in children and infants. Journal of the Acoustical Society of
America,78, 931-935.

Sun, X., Kim, D.O., Jlmg, M.D., & Randolph, KJ. (1996). Distortion product
otoacoustic emission test of sensorineural hearing loss in humans: comparison of unequal-
and equal-level stimuli. Annals of OtologyLRhinology and thhrgology, 105(12), 982-990.

 

Tanaka, Y., Suzuki, M., & Inoue, T. (1990). Evoked otoacoustic emissions in
sensorineural hearing impairment: its clinical implications. Bar and Hm’ , 11(2), 134-
143.

Ueda, H., Yamarnoto, Y., and Yanagita, N. (1996). Effect of aspirin on transiemly
evoked otoacoustic emissions in guinea pigs. ORL 58 61-67.

 

Vinck, B.M., De Vel, E., Xu, 2., & Van Cauwenberge, PB. (1996). Distortion
product otoacoustic emissions: a normative study. Audiology, 35, 231-245.

Wake, M., Anderson, J., Takeno, S., Mormt, R.J., & Harrison, R.V. (1996).
Otoacoustic emission amplification after inner hair cell damage. Wig,
116(3), 374-81.

Webster’s desk dictionhry of the lish e. (1990). New York, NY:
Portland House.

Weir, C.C., Pasanen, E.G., & McFadden, D. (1988). Partial dissociation of
spontaneous otoacoustic emissions and distortion products during aspirin use in humans.
Journ_al of the Acoustical Society of America. 84(1), 230-237.

Whitehead, M.L., Kamal, N., Lonsbury-Martin, B.L., & Martin, GK. (1993).
Spontaneous otoacoustic emissions in different racial groups. Scandinavian Audiolpgy, 22,
3-10.

Zwislocki, J. (1970). An acoustic coupler for eggphone cah'bration (Special Report
no. LCS-S-7), Syracuse University, New York.

63

"iiiiiliiiiiiiir