...

THE CLINICAL VALUE OF THE AUDITORY STEADY STATE

by user

on
Category: Documents
8

views

Report

Comments

Transcript

THE CLINICAL VALUE OF THE AUDITORY STEADY STATE
University of Pretoria etd – Stroeb
bel, D (2006
7)
THE CLINICAL VALUE OF THE AUDITORY STEADY STATE
RESPONSE FOR EARLY DIAGNOSIS AND AMPLIFICATION
FOR INFANTS (0 – 8 months) WITH HEARING LOSS
Deidré Stroebel
Presented in partial fulfillment of the requirements for the degree M.
Communication Pathology
In the
Department Communication Pathology,
Faculty of Humanities
University of Pretoria
November 2005
University of Pretoria etd – Stroeb
bel, D (2006
7)
For Frederick, Nell and Mieke
Frederick Nell Stroebel
(23-08-1999 – 2-12-2004)
ACKNOWLEDGEMENTS
I am especially grateful to:
Dr
De
Wet
Swanepoel,
for
his
support,
guidance
and
encouragement – it has been a privilege to work with you
Ms Emily Groenewald, for her help, careful and thorough guidance
and support – your thoroughness has left an impression
The families and subjects participating in this study – thank you for
the privilege to work with you and the confidence you have in me
My friends and colleagues, for your support and encouragement
especially during the past year – thank you!
My help comes from the Lord
University of Pretoria etd – Stroeb
bel, D (2006
7)
SUMMARY
TITLE
:
THE CLINICAL VALUE OF THE AUDITORY STEADY
STATE RESPONSE FOR EARLY DIAGNOSIS AND
AMPLIFICATION FOR INFANTS (0 – 8 months) WITH
HEARING LOSS
NAME
:
DEIDRÉ STROEBEL
SUPERVISORS
:
DR.
DE WET SWANEPOEL
:
MS.
E. GROENEWALD
:
COMMUNICATION PATHOLOGY
DEPARTMENT
UNIVERSITY OF PRETORIA
DEGREE
:
M. COMMUNICATION PATHOLOGY
There has always been a need for objective tests that assess auditory
function
in
infants,
young
children,
and/or
any
patient
whose
development level precludes the use of behavioral audiometric
techniques. Although the Auditory Brainstem Response (ABR) is seen as
the ‘gold standard’ in the field of objective audiometry, it presents with its
own set of limitations. The Auditory Steady State Response (ASSR) has
gained considerable attention and is seen as a promising addition to the
AEP ‘family’ to address some of the limitations of the ABR. The ASSR
promises to estimate all categories of hearing loss (mild to profound) in a
frequency specific manner. It also indicates to the possibility to validate
hearing aid fittings by determining functional gain of hearing aids by
determining unaided and aided ASSR thresholds.
An exploratory research design was selected in order to compare
unaided thresholds, obtained through the use of three different
procedures – ABR, ASSR and behavioral thresholds. Aided thresholds were
also obtained and compared with two procedures – the aided ASSR
University of Pretoria etd – Stroeb
bel, D (2006
7)
(measured and predicted) and aided behavioral threshold. The results
indicated that both the ABR (tone burst and click) and ASSR provided a
reasonable
estimation
of
the
subsequently
obtained
behavioral
audiograms. The ASSR, however, approximated the behavioral thresholds
closer than the ABR and were furthermore able to quantify hearing
thresholds accurately for subjects with severe and profound hearing
losses. The result indicated further that the ASSR can be instrumental in the
validation process of hearing aid fittings in infants. These results
demonstrated
however,
that
the
ASSR
measured
thresholds
underestimate the aided behavioral thresholds and the aided ASSR
predicted thresholds overestimate the aided behavioral thresholds.
The research concluded that the ASSR is useful in estimating frequencyspecific behavioral thresholds accurately in infants and validating hearing
aid fittings. Until evidence is sufficient to recommend the ASSR as primary
electrophysiological measure of hearing in infants, the ASSR should be
used in conjunction with the ABR – following a test battery approach in
the diagnostic process of hearing loss in infants. The ASSR further shows
great promise in validating hearing aid fittings, but this specific application
of the ASSR needs further research evidence on large groups to validate
the procedure.
Key terms: Objective tests, estimate behavioral thresholds, auditory
brainstem response, auditory steady state response, frequency specific,
test battery, validation of hearing aids, ASSR measured thresholds, ASSR
predicted thresholds, auditory evoked potentials.
University of Pretoria etd – Stroeb
bel, D (2006
7)
OPSOMMING
TITEL
:
DIE KLINIESE WAARDE VAN DIE OUDITIEF
STANDHOUDENDE
RESPONS
VIR
VROEË
IDENTIFIKASIE VAN GEHOORVERLIES EN VROEË
PASSING VAN GEHOORAPPARATE IN DIE JONG
BABA (0 – 8 maande)
NAAM
:
DEIDRÉ STROEBEL
STUDIELEIERS
:
DR DE WET SWANEPOEL
:
ME E. GROENEWALD
:
KOMMUNIKASIEPATOLOGIE
DEPARTEMENT
UNIVERSITEIT VAN PRETORIA
GRAAD
:
M. KOMMUNIKASIEPATOLOGIE
In die veld van Oudiologie is daar ‘n voortdurende behoefte na
objektiewe oudiometriese prosedures om ouditiewe sensitiwiteit in babas,
jong kinders en/of enige pasiënte wie se ontwikkelingsvlak hul uitskakel
van gedragsoudiometrie, te bepaal. Die Ouditiewe Breinstam Respons
(OBR) word gesien as die “goue standaard” in die veld van objektiewe
oudiometrie, alhoewel die tegniek sy eie beperkinge voorhou. Die
Ouditiewe Standhoudende Respons (OSR), het aansienlike aandag begin
geniet en word gesien as ‘n belowende toevoeging tot die ‘familie’ van
Ouditief Ontlokte Potensiale (OOP), wat gevolglik sekere van die OBR se
tekortkominge kan aanspreek. Die OSR blyk ‘n ouditief ontlokte respons te
wees wat spesifiek geskik is om alle kategorieë van gehoorverlies
frekwensie-spesifiek te voorspel. Daar is ook aanduidings dat die OSR
geskik mag wees om passings van gehoorapparate by jong babs te
bevestig, deur beide onversterkte en versterkte OSR drempels te bepaal.
University of Pretoria etd – Stroeb
bel, D (2006
7)
‘n
Ondersoekende
navorsingsontwerp
is
gebruik
om
onversterkte
drempels, soos bepaal deur drie verskillende prosedures – OBR, OSR en
gedragsoudiometrie – te bepaal en te vergelyk. Versterkte drempels is
ook bepaal en vergelyk deur middel van twee prosedures, naamlik die
versterkte OSR (meting en voorspelling) en versterkte gedragsoudiometrie.
Die resultate het getoon dat beide die OBR (toonbreuk en klik) en die OSR
‘n redelike beraming van suiwertoon gedragsoudiometrie vertoon. Die
OSR het egter die suiwertoondrempels meer akkuraat beraam en was
daartoe
instaat
om
die
erge
en
uitermatige
gehoorverliese
te
kwantifiseer. Die resultate het verder daarop gedui dat die OSR ‘n rol kan
speel in die bevestiging/validasie van gehoorapparaatpassings in babas.
Die resultate het gedui daarop dat die OSR meting die versterkte
gedragsoudiometrie drempels onderskat, terwyl die OSR voorspelde
drempels die versterkte gedragsoudiometrie drempels oorskat.
Die navorsing het bevind dat die OSR nuttig is om frekwensie-spesifieke
suiwertoondrempels akkuraat vir babas te voorspel. Die OSR toon ook
waarde in die validasie-proses wanneer gehoorapparate gepas word.
Verdere navorsing is
egter nodig
alvorens
die
OSR
as
primêre
elektrofisiologiese prosedure aanbeveel kan word om gehoor van babas
te evalueer. Dit is duidelik dat die OSR deel van ‘n toets-battery
benadering moet wees om gehoorsensitiwiteit van babas te evalueer. Die
OSR dui verder daarop dat dit ‘n rol kan speel in die validasie-proses
wanneer gehoorapparate gepas word, maar dat hierdie toepassing van
die tegniek verdere navorsing benodig. Validasie daarvan op groot
groepe is nodig.
Sleutelwoorde:
Objektiewe
suiwertoondrempels,
ouditiewe
oudiometrie,
breinstam
beraming
respons,
van
ouditiewe
University of Pretoria etd – Stroeb
bel, D (2006
7)
standhoudende respons, frekwensie-spesifiek, toetsbattery, validasie van
gehoorapparate,
potensiale.
OSR
meting,
OSR
voorspelling,
ouditief
ontlokte
University of Pretoria etd – Stroeb
bel, D (2006
7)
TABLE OF CONTENT
PAGE
CHAPTER 1 BACKGROUND AND RATIONALE OF STUDY …………………………………
1
1.1
ORIENTATION TO THE STUDY………………………………………
1
1.2
BACKGROUND ……………………………………………………...
3
1.3
RATIONALE …………………………………………………………..
7
1.4
PROBLEM STATEMENT ……………………………………………...
10
1.5
DEFINITION OF TERMS ……………………………………………...
13
1.6
DIVISION OF CHAPTERS ……………………………………………
15
1.7
SUMMARY ……………………………………………………………
17
CHAPTER 2 –
CLINICAL APPLICATION OF AUDITORY EVOKED POTENTIAL IN INFANTS:
COMPARING THE AUDITORY BRAINSTEM RESPONSE AND AUDITORY
STEADY STATE RESPONSE ……………………………………………………….
18
2.1
INTRODUCTION ……………………………………………………..
18
2.2
EARLY INTERVENTION FOR INFANTS WITH HEARING LOSS …...
19
2.2.1
Early identification and diagnosis of hearing loss ……………
20
2.2.2
Early amplification for infants with hearing loss ………………
24
2.2.2.1
Approaches to pediatric hearing aid fitting ..........................
25
University of Pretoria etd – Stroeb
bel, D (2006
7)
PAGE
2.3
CRITICAL EVALUATION OF AEP’S IN PEDIATRIC
AUDIOLOGY …………………………………………………………
33
2.3.1
Auditory Brainstem Response ……………………………………
34
2.3.1.1
Detection of hearing loss …………………………………………
35
2.3.1.2
Diagnosis of hearing loss ………………………………………….
36
i.
ABR threshold evaluations using clicks ………………………...
36
ii.
ABR threshold evaluation using brief tones …………………...
39
2.3.1.3
The ABR in pediatric hearing aid fittings ………………………
43
2.3.1.4
Summary of the ABR application in Pediatric Audiology......
46
2.3.2
Perspectives on the Auditory Steady State Response ……...
47
2.3.2.1
Definition and development of Auditory Steady
State Response ……………………………………………………..
47
i.
Single stimuli vs. multiple stimuli ASSR …………………………..
51
2.3.2.2
Threshold determination ………………………………………….
53
2.3.2.3
Current clinical application of the ASSR in infants …………..
54
i.
Detection ……………………………………………………………
55
ii.
Diagnosis …………………………………………………………….
57
iii.
The ASSR in pediatric hearing aid fittings ……………………..
60
University of Pretoria etd – Stroeb
bel, D (2006
7)
PAGE
2.3.2.4
Critical evaluation of the ASSR ………………………………….
2.3.2.5
Summary of the ASSR application in Pediatric
61
Audiology ……………………………………………………………
65
2.4
CONCLUSION ……………………………………………………….
66
2.5
SUMMARY ……………………………………………………………
67
CHAPTER 3 –
RESEARCH METHODOLOGY ……………………………………………………
68
3.1
INTRODUCTION ……………………………………………………..
68
3.2
AIMS OF RESEARCH ………………………………………………..
69
3.2.1
Main aim ……………………………………………………………..
69
3.2.2
Sub aims ……………………………………………………………..
70
3.3
RESEARCH DESIGN …………………………………………………
70
3.4
ETHICAL CONSIDERATIONS ……………………………………….
73
3.4.1
Autonomy …………………………………………………………...
73
3.4.2
Beneficence ………………………………………………………...
75
3.4.3
Justice ………………………………………………………………..
77
3.5
SUBJECTS ……………………………………………………………..
77
3.5.1
Sampling ……………………………………………………………..
77
3.5.2
Selection criteria ……………………………………………………
78
University of Pretoria etd – Stroeb
bel, D (2006
7)
PAGE
3.5.2.1
Client status and record ………………………………………….
78
3.5.2.2
Hearing ability ………………………………………………………
78
3.5.2.3
Normal Middle Ear functioning ………………………………….
78
3.5.2.4
Age at time of identification …………………………………….
79
3.5.2.5
Neurological status ………………………………………………...
79
3.5.3
Subject Selection Apparatus …………………………………….
80
3.5.3.1
Hearing screening apparatus …………………………………...
80
3.5.3.2
Otoscopic examination …………………………………………..
81
3.5.3.3
Middle ear assessment ……………………………………………
81
3.5.4
Subject selection procedures …………………………………...
81
3.6
DESCRIPTION OF SUBJECTS ……………………………………….
82
3.7
MATERIAL AND APPARATUS ………………………………………
83
3.7.1
Hearing thresholds estimation apparatus ……………………..
83
3.7.2
Functional gain estimation apparatus ………………………...
88
3.7.3
Clinical audiometer ……………………………………………….
88
3.7.4
Test environment …………………………………………………...
89
3.7.5
Data collection sheet ……………………………………………..
89
3.8
PROCEDURE …………………………………………………………
89
3.8.1
Data collection procedure ………………………………………
89
3.8.1.1
Auditory Brainstem Response (ABR) …………………………....
91
University of Pretoria etd – Stroeb
bel, D (2006
7)
PAGE
3.8.1.2
Auditory Steady State Response (ASSR) ……………………….
91
3.8.1.3
Aided ASSR thresholds …………………………………………….
92
3.8.1.4
Unaided behavioral pure tone thresholds (BT)……………….
92
3.8.1.5
Aided behavioral thresholds …………………………………….
94
3.8.2
Procedures for data recording, processing and
Analysis ……………………………………………………………….
94
3.8.2.1
Recording of data …………………………………………………
94
3.8.2.2
Procedures for processing and analysis of data …………….
95
3.8.3
Validity and reliability ……………………………………………..
98
3.9
Summary ……………………………………………………………..
99
CHAPTER 4 –
RESULTS AND DISCUSSION ……………………………………………………..
101
4.1
INTRODUCTION ……………………………………………………..
101
4.2
RESULTS FOR SUB-AIM 1: TO INVESTIGATE THE POTENTIAL
CLINICAL VALUE OF THE ASSR IN EARLY DIAGNOSIS OF
HEARING LOSS IN A GROUP OF INFANTS BY DETERMINING
AND COMPARING UNAIDED ASSR, ABR AND BEHAVIORAL
THRESHOLDS …………………………………………………………
104
4.2.1
Individual subject results for sub-aim 1 ………………………...
104
4.2.1.1
Subject 1: Results for sub-aim 1 ………………………………...
104
University of Pretoria etd – Stroeb
bel, D (2006
7)
PAGE
4.2.1.2
Subject 2: Results for sub-aim 1 …………………………………
106
4.2.1.3
Subject 3: Results for sub-aim 1 …………………………………
109
4.1.2.4
Subject 4: Results for sub-aim 1 …………………………………
112
4.2.1.5
Subject 5: Results for sub-aim 1 …………………………………
115
4.2.1.6
Subject 6: Results for sub-aim 1 …………………………………
117
4.2.2
Collective results for all six subjects concerning
Sub-aim 1 …………………………………………………………….
119
4.2.2.1
Comparing the unaided ABR and unaided ASSR …………..
121
4.2.2.2
Unaided ASSR vs. unaided behavioral thresholds …………..
126
4.2.2.3
Unaided ABR vs. unaided behavioral thresholds ……………
131
4.3
RESULTS FOR SUB-AIM 2:
TO INVESTIGATE THE CLINICAL
VALUE OF THE ASSR FOR RELEVANT EARLY FITTING OF
HEARING
AIDS
COMPARING
IN
AIDED
INFANTS
ASSR
BY
AND
DETERMINING
AIDED
AND
BEHAVIORAL
THRESHOLDS …………………………………………………………
136
4.3.1
Individual subject results for sub-aim 2 ………………………...
137
4.3.1.1
Subject 1: Results for sub-aim 2 …………………………………
137
4.3.1.2
Subject 2: Results for sub-aim 2 …………………………………
139
4.3.1.3
Subject 3: Results for sub-aim 2 …………………………………
141
4.3.1.4
Subject 4: Results for sub-aim 2 …………………………………
143
4.3.1.5
Subject 5: Results for sub-aim 2 …………………………………
145
4.3.1.6
Subject 6: Results for sub-aim 2 …………………………………
147
University of Pretoria etd – Stroeb
bel, D (2006
7)
PAGE
4.3.2
Collective results from all six subjects concerning
sub-aim 2………………………………………………………….
149
4.4
DISCUSSION………………………………………………………
159
4.4.1
Sub-aim 1: To investigate the potential clinical value
of the ASSR in early diagnosis of hearing loss in a
group of infants by determining and comparing
unaided ASSR, ABR and behavioral thresholds…………...
160
4.4.1.1
ABR vs. ASSR……………………………………………………...
160
4.4.1.2
ASSR vs. behavioral measures………………………………..
161
4.4.1.3
ABR vs. behavioral measures…………………………………
164
4.4.2
Sub-aim 2: To investigate the potential clinical value
of the ASSR for relevant early fitting of hearing aids in
infants by determining and comparing aided ASSR
and aided behavioral thresholds……………………………
165
4.4.2.1
Unaided ASSR vs. aided ASSR responses…………………...
166
4.4.2.2
Aided ASSR responses vs. aided behavioral responses…
167
4.5
CONCLUSION…………………………………………………….
171
4.6
SUMMARY…………………………………………………………
172
University of Pretoria etd – Stroeb
bel, D (2006
7)
PAGE
CHAPTER 5 –
CONCLUSIONS AND IMPLICATIONS ………………………………………….
173
5.1
INTRODUCTION ……………………………………………………..
173
5.2
CONCLUSIONS ……………………………………………………...
174
5.2.1
Sub-aim 1: To investigate the potential clinical value of
the ASSR in early diagnosis of hearing loss in a group of
infants by determining and comparing unaided ASSR, ABR
and behavioral thresholds ……………………………………….
5.2.2
175
Sub-aim 2: To investigate the potential clinical value of
the ASSR for relevant early fitting of hearing aids in infants
by determining and comparing aided ASSR and aided
behavioral thresholds ……………………………………………..
176
5.3
THEORETICAL AND CLINICAL IMPLICATIONS …………………
177
5.4
CRITICAL EVALUATION OF THE CURRENT STUDY ……………..
180
5.5
RECOMMENDATION FOR FUTURE RESEARCH …………………
182
5.6
CONCLUSION ……………………………………………………….
184
REFERENCES
186
APPENDIX A
209
APPENDIX B
210
APPENDIX C
212
University of Pretoria etd – Stroeb
bel, D (2006
7)
LIST OF TABLES
PAGE
TABLE
2.1
Stages of hearing aid fitting process …………………………..
27
2.2
Advantages and limitations of the ABR ……………………….
46
2.3
Advantages and limitations of the ASSR ………………………
65
3.1
Description of subjects ……………………………………………
83
3.2
Protocol for click ABR ……………………………………………..
85
3.3
Protocol for tone burst ABR ………………………………………
86
3.4
Protocol for the ASSR ……………………………………………...
87
4.1
Background information and test results for subject 1 ……..
105
4.2
Background information and test results for subject 2 ……..
107
4.3
Background information and test results for subject 3 ……..
110
4.4
Background information and test results for subject 4 ……..
113
4.5
Background information and test results for subject 5 ……..
115
4.6
Background information and test results for subject 6 ……..
117
4.7
Summary of unaided thresholds for the six subjects
as determined by the ABR, ASSR and behavioral
assessment …………………………………………………………..
4.8
120
Average on all frequencies tested on three
procedures ………………………………………………………….
121
4.9
Statistical analysis of ABR and ASSR results ……………………
124
4.10
Statistical analysis of ASSR and behavioral measures ………
129
4.11
Statistical analysis of ABR and behavioral measures ……….
134
4.12
Unaided ASSR, aided ASSR and aided behavioral
thresholds measurements for subject 1 ………………………..
138
University of Pretoria etd – Stroeb
bel, D (2006
7)
PAGE
4.13
Unaided ASSR, aided ASSR and aided behavioral
thresholds measurements for subject 2 ………………………..
4.14
Unaided ASSR, aided ASSR and aided behavioral
thresholds measurements for subject 3 ………………………..
4.15
by
ASSR
and
behavioral
assessments
respectively …………………………………………………………
151
Aided ASSR measured responses vs. aided behavioral
responses …………………………………………………………….
4.20
148
Summary of aided thresholds for the six subjects as
determined
4.19
146
Unaided ASSR, aided ASSR and aided behavioral
thresholds measurements for subject 6 ………………………..
4.18
144
Unaided ASSR, aided ASSR and aided behavioral
thresholds measurements for subject 5 ………………………..
4.17
142
Unaided ASSR, aided ASSR and aided behavioral
thresholds measurements for subject 4 ………………………..
4.16
140
155
Aided ASSR predicted responses vs. aided behavioral
responses …………………………………………………………….
157
University of Pretoria etd – Stroeb
bel, D (2006
7)
LIST OF FIGURES
PAGE
FIGURE
2.1
Principles underlying the ASSR …………………………………..
48
2.2
A single tone and a modulated tone …………………………
49
2.3
Recording the ASSR ……………………………………………….
51
2.4
Multiple ASSR ……………………………………………………….
52
2.5
Polar Plot to Phase Coherence …………………………………
54
3.1
A schematic representation of the data collection
procedures ………………………………………………………….
90
4.1
Main-aim and sub-aims of study ………………………………..
102
4.2
Schematic representation of the ABR, ASSR predictions
and BT results for subject 1 ……………………………………….
4.3
Schematic representation of the ABR, ASSR predictions
and BT results for subject 2 ……………………………………….
4.4
118
Representation of comparative frequency thresholds
between the ABR and ASSR ……………………………………..
4.9
116
Schematic representation of the ABR, ASSR predictions
and BT results for subject 6 ……………………………………….
4.8
113
Schematic representation of the ABR, ASSR predictions
and BT results for subject 5 ……………………………………….
4.7
110
Schematic representation of the ABR, ASSR predictions
and BT results for subject 4 ……………………………………….
4.6
108
Schematic representation of the ABR, ASSR predictions
and BT results for subject 3 ……………………………………….
4.5
106
122
Relationship between 500 Hz tone burst ABR and ASSR
prediction bases on the measurement for seven ears ……..
125
University of Pretoria etd – Stroeb
bel, D (2006
7)
PAGE
4.10
Relationship between the click ABR and 2000 and 4000 Hz
ASSR predictions based on the measurement of seven
ears ……………………………………………………………………
4.11
125
Mean unaided ASSR thresholds and unaided behavioral
thresholds obtained at each frequency for all the ears
tested (n = 12) ………………………………………………………
4.12
127
Representation of comparative frequency thresholds
between the ASSR predicted thresholds and behavioral
thresholds ……………………………………………………………
4.13
Relationship between thresholds determined with ASSR
and behavioral responses for a specific number of ears…..
4.14
threshold
assessment
based
on
the
measurement for seven ears …………………………………….
threshold
assessment
based
on
the
measurement for seven ears respectively ……………………
Aided
results
from
subject
1
thresholds and ASSR thresholds
including
–
Aided
results
thresholds
and
from
ASSR
subject
2
thresholds
measured and
including
–
135
behavioral
predicted ……………………………………………………………
4.18
134
Relationship between click ABR and 2000 Hz and 4000 Hz
behavioral
4.17
132
Relationship between tone burst ABR and 500 Hz
behavioral
4.16
130
Representation of comparative frequency thresholds
between the ABR and behavioral thresholds ………………..
4.15
128
138
behavioral
measured
and
predicted ……………………………………………………………
140
University of Pretoria etd – Stroeb
bel, D (2006
7)
PAGE
4.19
Aided
results
thresholds
from
and
ASSR
subject
3
including
thresholds
–
behavioral
measured
and
predicted ……………………………………………………………
4.20
Aided
results
thresholds
from
and
ASSR
subject
4
including
thresholds
–
behavioral
measured
and
predicted ……………………………………………………………
4.21
Aided
results
thresholds
from
and
ASSR
subject
5
including
thresholds
–
Aided
results
thresholds
from
and
ASSR
subject
6
measured
including
thresholds
–
and
146
behavioral
measured
and
predicted ……………………………………………………………
4.23
144
behavioral
predicted ……………………………………………………………
4.22
142
148
Comparison of average aided results for all measured
ears based on aided behavioral assessment, measured
and ASSR predicted values ……………………………………...
4.24
Representation of comparative frequencies on aided
ASSR
measured
thresholds
and
aided
behavioral
thresholds ……………………………………………………………
4.25
predicted
thresholds
and
aided
behavioral
thresholds ……………………………………………………………
154
Relationship between aided behavioral thresholds and
aided
ASSR
measured
responses
based
on
the
measurements for six subjects …………………………………...
4.27
154
Representation of comparative frequencies on aided
ASSR
4.26
152
156
Relationship between aided behavioral thresholds and
aided
ASSR
predicted
responses
based
on
the
measurements for six subjects …………………………………..
158
University of Pretoria etd – Stroeb
bel, D (2006
7)
LIST OF ABBREVIATIONS
AABR
-
Automated Auditory Brainstem Response
ABG
-
Air Bone Gap
ABR(s)
-
Auditory Brainstem Response(s)
AC
-
Air Conduction
AEP(s)
-
Auditory Evoked Potential(s)
AER
-
Auditory Evoked Response
AM
-
Amplitude Modulation
ASSR(s)
-
Auditory Steady State Response (s)
BC
-
Bone conduction
BT
-
Behavioral Threshold
CF(s)
-
Carrier Frequency (s)
CNS
-
Central Nervous System
dB
-
Decibel
eCochG
-
Electrocochleography
EEG
-
Electro-Encephalo-Gram
EOAE
-
Evoked Oto-acoustic Emissions
FFR
-
Frequency Following Response
FFT
-
Fast Fourier Transform
FM
-
Frequency Modulation
HL
-
Hearing Level
Hz
-
Hertz
IAFM
-
Independent Amplitude and Frequency
Modulation
JCIH
-
Joint Committee on Infant Hearing
Kg
-
kilogram
kHz
-
Kilo Hertz
kOhms
-
Kilo Ohm
University of Pretoria etd – Stroeb
bel, D (2006
7)
L
-
Left
L-I
-
Latency Intensity
LLR
-
Late Latency Response
MASTER
-
Multiple auditory steady-state response
MF(s)
-
Modulation Frequency (s)
mg
-
milligram
MLR
-
Middle Latency Response
ms
-
Millisecond
n
-
Number
nHL
-
Normal Hearing Level
NICU
-
Neonatal Intensive Care Unit
NR
-
No Response
OAE(s)
-
Oto-Acoustic Emission (s)
PC
-
Phase Coherence
R
-
Right
s
-
Second
sec
-
Second
SD
-
Standard Deviation
SLR
-
Short Latency Responses
SN10
-
Slow-negative Potential
SNHL
-
Sensory Neural Hearing Loss
SPL
-
Sound Pressure Level
SSP
-
Steady State Potential
SSEP
Steady State Evoked Potential
TB
-
Tone Burst
UNHS
-
Universal Newborn Hearing Screening
USA
-
United States of America
WRS
-
Word Recognition Scores
University of Pretoria etd – Stroeb
bel, D (2006)
Cha p ter 1
INTRODUCTION
Cha p ter one introd uc es the p rob lem this stud y c onfronts; the ra tiona le
therefore, d esc rib es the term inolog y used , a nd p resents a n overview of
the c ontent a nd org a niza tion of the stud y.
1.1
ORIENTATION TO THE STUDY
The p ra c tic e of p ed ia tric a ud iolog y is b a sed on the p rinc ip le tha t the b est
outc om es for a c hild w ith hea ring loss a re a c hieved w hen his or her
hea ring sta tus is d eterm ined a s ea rly a s p ossib le, follow ed b y tim ely
intervention (Cone-Wesson, 2003:253). Hea ring loss in new b orn infa nts c a n
g o und etec ted until a s la te a s three yea rs of a g e w ithout sp ec ia lized
testing (Ha yes & Northern, 1997:4). When hea ring loss is d etec ted in the
new b orn p eriod , infa nts c a n b enefit m a xim a lly from
a m p lific a tion
(hea ring a id s) a nd intervention to fa c ilita te sp eec h a nd la ng ua g e
d evelop m ent (Sining er, Doyle & Moore, 1999:11). Evid enc e reg a rd ing
neura l d evelop m ent strong ly sup p orts suc h ea rly intervention for op tim a l
outc om es of c om m unic a tion a b ility a nd hea ring in infa nts (Sining er, Doyle
& Moore, 1999:11). Ped ia tric a ud iolog ists, therefore, should fa c ilita te ea rly
d etec tion of hea ring loss a nd intervention throug h sc reening p rog ra m s
a nd in-d ep th hea ring a ssessm ents, d eterm ining w hic h tec hnolog ies a nd
ha b ilita tion p rog ra m s a re b est suited to the need s of b oth the infa nt a nd
the fa m ily.
1
University of Pretoria etd – Stroeb
bel, D (2006)
In ord er to c a p ita lize on the p ositive a sp ec ts of ea rly id entific a tion, the
Yea r 2000 Position Sta tem ent on Infa nt Hea ring Sc reening p rod uc ed b y
the Joint Com m ittee on Infa nt Hea ring (JCIH), rec om m end tha t a ll infa nts’
hea ring should b e sc reened using objective, physiologic m ea sures in
ord er to id entify hea ring loss (JCIH, 2000:10). Aud iolog ic a nd m ed ic a l
eva lua tions should b e c ond uc ted b efore three m onths of a g e. Infa nts
w ith c onfirm ed hea ring loss should rec eive intervention b efore six m onths
of a g e from hea lth c a re a nd ed uc a tion p rofessiona ls w ith exp ertise in
hea ring loss a nd d ea fness in infa nts a nd young c hild ren (JCIH, 2000:10).
Infa nts id entified a c c ord ing to these rec om m end a tions a re too young for
the use of tra d itiona l a ud iom etric p roc ed ures to d eterm ine a n a ud iog ra m
(Cone-Wesson, 2003: 254). Instea d , elec trop hysiolog ic m ethod s suc h a s
Aud itory Evoked Potentia ls (AEP) m ust b e used to estim a te hea ring
threshold s. Ong oing eva lua tion of hea ring func tion is furtherm ore need ed
to m onitor the effec ts of ea rly intervention a nd AEP tests m a y a lso form a n
im p orta nt p a rt in this p roc ess (Cone-Wesson, 2003:270).
Ac c ord ing to Diefend orf a nd Web er (1994:56), four c riteria need to b e
a d d ressed b efore a m p lific a tion of a hea ring loss c a n oc c ur. The d eg ree,
the c onfig ura tion, the sym m etry a nd the typ e of hea ring loss m ust b e
d eterm ined . Am p lific a tion a nd ha b ilita tion stra teg ies, suc h a s c hoic e of
hea ring a id s, c oc hlea r im p la nts a nd c hoic e of m od e of c om m unic a tion,
a re b a sed on the a b ove c riteria – usua lly revea led b y the a ud iog ra m
(Ross, 2001:3). Elec trop hysiolog ic a l tests c a n a ssist in the p roc ess of
c ha ra c terizing hea ring loss for infa nts, sinc e these tests c a n m ea sure
a ud itory func tion ob jec tively - g iving freq uenc y-sp ec ific inform a tion
need ed to fit a p p rop ria te a m p lific a tion in this young p op ula tion.
2
University of Pretoria etd – Stroeb
bel, D (2006)
Ap a rt from a ssisting in the p roc ess of d ia g nosing the hea ring loss, it is
im p orta nt to verify tha t a sp ec ific hea ring a id p rovid es a d eq ua te
a m p lific a tion
(Pic ton, Dim itrijevic , Va n
Roon, Sa sha -John, Reed
&
Finkelstein, 2002:63). Ac c ord ing to Pic ton et a l. (2002:64), verific a tion of a
hea ring a id fitting p rovid es som e ind ic a tion of how w ell sound s a re hea rd
w hen the a id is used a t its p resc rib ed setting s. Elec trop hysiolog ic a l tests
c a n a ssist in this p roc ess of verific a tion in infa nts a nd young c hild ren.
Stud ies ha ve ind ic a ted tha t b oth the Aud itory Bra instem Resp onse (ABR)
a nd the Aud itory Stea d y Sta te Resp onse (ASSR) c a n b e used to estim a te
threshold w hen the stim uli is tra nsd uc ed b y a hea ring a id (Ga rnha m ,
Cop e, Durst, Mc Corm ic k, & Ma son, 2000:268; Pic ton, Durieux-Sm ith,
Cha m p a g ne, Whitting ha m , Mora n, Gig ueve & Bea ureg a rd , 1998:315).
Sinc e the ea rly id entific a tion of hea ring loss throug h new b orn hea ring
sc reening ha s resulted in a young er p op ula tion b eing served b y the
a ud iolog ist, elec trop hysiolog ic a l tests a re b ec om ing a m ore essentia l p a rt
of the req uired test-b a ttery.
1.2
BACKGROUND
The a ud iolog ist serves a s the p rofessiona l p rim a rily resp onsib le for the
a ssessm ent
a nd
non-m ed ic a l
d ia g nosis
of
a ud itory
im p a irm ent.
Assessm ent inc lud es, b ut is not lim ited to, the a d m inistra tion a nd
interp reta tion of b eha viora l, elec troa c oustic , a nd
elec trop hysiolog ic
m ea sures of the sta tus of periphera l a nd centra l a uditory nervous systems
(Sta c h, 1998:3). The m a in p urp ose of a hea ring eva lua tion is to d efine the
na ture a nd extent of hea ring im p a irm ent. A c om p rehensive d esc rip tion of
hea ring a b ility serves a s a first step in the reha b ilita tion of a hea ring
ha nd ic a p tha t result from a n im p a irm ent. An ind ivid ua l w ith a hea ring
3
University of Pretoria etd – Stroeb
bel, D (2006)
d isord er m ust therefore rely on a n a ssessm ent of a ud itory func tion a s the
found a tion of the reha b ilita tion p roc ess (Sw a nep oel, 2001:1).
Pure tone a ud iom etry rep resents the first a nd , a rg ua b ly, the m ost
fund a m enta l m ea sure of hea ring a c uity (Ha rrell, 2002: 71). The b eha viora l
p ure tone a ud iog ra m is a m ea sure of a ud itory threshold a s a func tion of
freq uenc y. The a ud iog ra m c onfig ura tion p rovid es fund a m enta l b a seline
inform a tion for the selec tion of a suita b le a m p lific a tion system (Sta c h,
1998:68). It is und ersta nd a b le then tha t p ure tone a ud iom etry ha s
rem a ined the a ud iom etric p roc ed ure of c hoic e. It em b od ies the g old
sta nd a rd for freq uenc y-sp ec ific threshold esta b lishm ent a g a inst w hic h a ll
other a ud iom etric m ea sures a re c om p a red (Sw a nep oel, 2001:3).
It is c lea r tha t the sta nd a rd c linic a l p ure tone tec hniq ue w ill not b e
effec tive for a ll c linic a l p op ula tions. The m ost ob vious is the p ed ia tric
p op ula tion. This is a lso true of the d evelop m enta lly d ela yed a nd , in som e
c a ses, the severely p hysic a lly im p a ired (Ha rrell, 2002:73). Cond itioned
a ud iom etric test tec hniq ues (suc h a s visua l reinforc em ent a ud iom etry),
tha t c a n p rovid e inc rea sing ly a c c ura te inform a tion in old er c hild ren (>6
m onths), a re not suita b le for q ua ntifying hea ring loss in very young infa nts
or for those w ith visua l or d evelop m enta l d isa b ilities (Ra nc e & Brig g s,
2002:237) a s sup ra threshold stim ula tion is req uired to elic it reflexive
resp onses.
These c ond itioned a ud iom etric m ethod s a re lim ited to the
d etec tion of hea ring loss g rea ter tha n 50 d B HL (Diefend orf & Web er,
1994:57).
Therefore, a s the a g e of id entific a tion is red uc ed the need for a c c ura te,
relia b le, objective
m ethod s for d eterm ining
hea ring
threshold s, is
b ec om ing m ore urg ent. When a n infa nt is id entified a s ha ving a hea ring
loss, there is a n im m ed ia te need to c ha ra c terize the d eg ree, the
4
University of Pretoria etd – Stroeb
bel, D (2006)
c onfig ura tion, a nd the typ e of loss, sinc e a p p rop ria te rec om m end a tions
reg a rd ing the selec tion a nd fitting of d evic es suc h a s hea ring a id s a nd
c oc hlea r im p la nts, c a n only b e m a d e w ith a n und ersta nd ing of b oth the
d eg ree a nd c onfig ura tion of a c hild ’ s hea ring loss (Ra nc e & Brig g s,
2002:237; Va nd er Werff, Brow n, Giena p p , & Sc hm id t Cla y, 2002:228).
Different tec hniq ues ha ve therefore b een d evelop ed over the yea rs in
ord er to a d d ress the p rob lem of q ua ntifying hea ring loss in young infa nts.
One suc h a m ea sure, tha t w a s rec eived p ositively for d ia g nostic p urp oses
in the p ed ia tric field , w a s otoa c oustic em issions – d isc overed in 1978 b y
Da vid Kem p . Otoa c oustic em issions (OAE’ s) a re sound s tha t orig ina te in
the c oc hlea a nd p rop a g a te throug h the m id d le ea r a nd into the ea r
c a na l, w here they c a n b e m ea sured using a sensitive m ic rop hone (Prieve
& Fitzg era ld , 2002:440). It ha s b een know n for the la st 20 yea rs tha t
ind ivid ua ls w ith sig nific a nt c oc hlea r hea ring loss ha ve no m ea sura b le
Evoked Otoa c oustic Em issions (EOAE’ s). Ha ll (1997:265) ha s ind ic a ted tha t
EOAE’ s a re either not m ea sura b le from sub jec ts w ith hea ring loss or a re
sub sta ntia lly
red uc ed
in
a m p litud e
c om p a red
to
norm a l hea ring
ind ivid ua ls. The m inim um level req uired for m ea sura b le EOAE’ s ra ng es
from 25 – 40 d B HL (Prieve & Fitzg era ld , 2002:452).
Given this sp ec ific p op ula tion a nd the need for sp ec ific inform a tion, w ith
reg a rd to d eg ree, c onfig ura tion, sym m etry a nd typ e of hea ring loss, w ork
in the a rea of OAE’ s d oes not p rovid e c om p elling evid enc e to ind ic a te
tha t EOAE’ s c ould b e used to p red ic t the na ture of a n infa nt’ s hea ring loss
(Prieve & Fitzg era ld , 2002:456). This p roc ed ure is, how ever, a p ow erful tool
for new b orn hea ring sc reening a nd should form a n integ ra l p a rt of the
d ifferentia l d ia g nosis test b a ttery in the d ia g nosis of hea ring loss in infa nts
(Ha ll, 2000:391).
5
University of Pretoria etd – Stroeb
bel, D (2006)
The ABR is a fa r field evoked p otentia l tha t is used freq uently to estim a te
a ud itory sensitivity in c hild ren too young to b e tested using sta nd a rd
b eha viora l m ethod s. The ABR to c lic k stim uli is the m ost c om m only used
c linic a l p roc ed ure in p red ic ting hea ring sensitivity (Va nd er Werff, Brow n,
Giena p p , & Sc hm id t Cla y, 2002:228). Althoug h the a b rup t onset of the
c lic k stim ulus is id ea l for g enera ting the ABR, the d ra w b a c k is tha t the c lic k
is a b roa d -sp ec trum sig na l, c onta ining energ y a c ross a w id e ra ng e of
freq uenc ies. Ow ing to this freq uenc y sp rea d , c lic ks c a nnot b e used to
a ssess sensitivity in sp ec ific freq uenc y reg ions, b ut ra ther to p rovid e a g ross
estim a te of hea ring sensitivity (Arnold , 2000:454).
ABR to b rief tones c a n b e used to ob ta in m ore freq uenc y-sp ec ific
threshold
inform a tion tha n a va ila b le
from
the
c lic k ABR. Sta p ells
(2000a :13) ha s show n a c ross stud ies tha t tone-ABR threshold s ha ve b een
found to b e a relia b le m ethod in ob ta ining freq uenc y sp ec ific inform a tion
in this young
p op ula tion. Sta p ells (2004: c onferenc e
p resenta tion)
m a inta ins tha t the tone-evoked a ud itory b ra instem resp onse (tone-ABR) is
c urrently the only m ea sure tha t c a n relia b ly p rovid e inform a tion w ith
reg a rd to severity, c onfig ura tion a nd na ture of hea ring loss in infa nts.
Som e stud ies ha ve how ever q uestioned the freq uenc y-sp ec ific ity a nd
relia b ility of threshold estim a tion w ith low freq uenc y tone-evoked ABR
(Cone-Wesson et a l., 2002:174). In a d d ition, toneb urst ABR w a veform s,
esp ec ia lly to low -freq uenc y stim uli, tend to b e less d istinc t a nd m ore
d iffic ult to id entify tha n the c lic k ABR (Arnold , 2000:459). Outp ut lim ita tions
a re a lso a c onc ern w ith tone b urst stim uli, p a rtic ula rly for low -freq uenc y
tone b ursts for w hic h threshold s a re eleva ted rela tive to b eha viora l
threshold s. These c onc erns m a y lim it the im p lem enta tion of toneb urst ABR
p rotoc ols (Va nd er Werff et a l., 2002:229).
6
University of Pretoria etd – Stroeb
bel, D (2006)
1.3
RATIONALE
In the p a st tw o d ec a d es, the Aud itory Stea d y Sta te Resp onse (ASSR) ha s
b een d evelop ed a s a n a lterna tive freq uenc y sp ec ific Aud itory Evoked
Potentia l (AEP) a p p roa c h to q ua ntify hea ring loss (Ra nc e et a l., 1998:499).
Va nd er Werff et a l. (2002:228) d efine the ASSR a s “ a n a lterna tive evoked
p otentia l tec hniq ue tha t uses c ontinuous ra ther tha n tra nsient stim uli to
elic it a resp onse from the a ud itory system ” . Unlike the ABR, the ASSR use
stim uli tha t is c ontinuous. The stim uli used to evoke ASSR, a re m od ula ted
tones, w hic h a re freq uenc y sp ec ific d ue to the fa c t tha t sp ec tra l energ y is
c onta ined only a t the freq uenc y of the c a rrier tone a nd the freq uenc y of
the m od ula tion (Hood , 1998:117). Resp onses from the neura l system tha t
resp ond s to the c ha ng es or m od ula tions in the stim uli a re rec ord ed . The
ASSR a p p ea rs to b e g enera ted b y the sa m e neura l a na tom ic a l reg ions
from w hic h the ABR evoked b y c lic ks or tone-b ursts is p rod uc ed (ConeWesson, 2003:267).
The ASSR show s p otentia l to a d d ress som e of the lim ita tions a ssoc ia ted
w ith ABR testing in the ea rly d ia g nosis a nd a m p lific a tion of infa nts. One of
the lim ita tions of ABR is the la c k of freq uenc y sp ec ific ity. The na ture of the
ASSR stim uli offers a d va nta g es over other short d ura tion stim uli (Ra nc e et
a l, 1998:499). Bec a use the threshold estim a tes ob ta ined from ASSR testing
a re freq uenc y sp ec ific , it a llow s for testing a c ross the a ud iom etric ra ng e
a nd for the g enera tion of evoked p otentia l a ud iog ra m s (Ra nc e et a l.,
1998:499). This fea ture w ill a d d ress the p rob lem of d eterm ining the
c onfig ura tion of hea ring loss in infa nts.
Due to the lim ita tions in m a xim um outp ut w ith the ABR, the a b senc e of
w a ve V in ABR test results d oes not inevita b ly im p ly the a b senc e of
hea ring (Arnold , 200:457). Ra nc e et a l. (1998:506) d em onstra ted the
7
University of Pretoria etd – Stroeb
bel, D (2006)
a d va nta g es of using ASSR’ s to d eterm ine resid ua l hea ring threshold s in
infa nts a nd c hild ren for w hom ABR’ s c ould not b e evoked a t 100 d BnHL.
The c ontinuously m od ula ted tone used to elic it the ASSR c a n b e
p resented a t levels a s hig h a s 120 d B HL. Ab senc e of a c lic k- or tone-b urst
ABR d oes not inevita b ly ind ic a te p rofound d ea fness, a nd ASSR tests m a y
revea l enoug h resid ua l hea ring to c onsid er the use of a m p lific a tion or
help d eterm ine the p referred ea r for c oc hlea r im p la nta tion (Ra nc e et a l.,
1998:506). The a d va nta g es offered b y the ASSR in this reg a rd a re m ost
b enefic ia l for infa nts a nd c hild ren a s the severity of the hea ring loss c a n
b e d eterm ined in a m ore a c c ura te m a nner.
Cone-Wesson et a l. (2002:270) m entions a third lim ita tion of the ABR for
a ud iolog ic a p p lic a tion a s the sub jec tive na ture of resp onse d etec tion.
Althoug h m ethod s for “ a utom a tic
d etec tion”
of ABR exist, these
a lg orithm s ha ve b een suc c essfully a p p lied for c lic k-evoked ABR. There a re
no p ub lished d a ta on the use of a utom a tic d etec tion c riteria for
d etec ting the resp onse to tona l stim uli. In c ontra st, there ha s b een
extensive resea rc h on the effic a c y for d etec ting a stea d y-sta te resp onse
a utom a tic a lly (Cone-Wesson et a l., 2002:175; Ra nc e et a l., 1995:499). The
ob jec tive na ture of resp onse d etec tion in the ASSR m ea sures m a y lea d to
m ore a c c ura te d ia g nosis of hea ring loss in infa nts.
Ac c ord ing to Ra nc e et a l. (1998:506) a further a d va nta g e of the ASSR a s
op p osed to the ABR, is the sp eed w ith w hic h a resp onse c a n b e
d etec ted . Althoug h Ha ll (2005:c onferenc e p resenta tion) d isp utes the
sp eed of the ASSR to b e fa ster tha n the ABR, severa l resea rc hers ha ve
c onc lud ed tha t ASSR offers the p ossib ility of estim a ting freq uenc y-sp ec ific
hea ring
threshold s in
b a b ies in
a
m ore
tim e-effic ient w a y (Luts,
Desloovere, Kum a r, Va nd erm eersc h & Wouters, 2004:995; Sw a nep oel,
2001:112; Ra nc e et a l., 1998:506). A c onsta nt unp red ic ta b le fa c tor in
8
University of Pretoria etd – Stroeb
bel, D (2006)
testing infa nts is tha t they m a y a w a ke a t a ny m om ent d uring the
p roc ed ure. The fa st d etec tion sp eed of the ASSR thus red uc es the need to
ha ve the infa nt a sleep or und er sed a tion for long p eriod s of tim e.
It is tec hniq ues suc h a s the
ASSR tha t und erlie
suc c essful ea rly
a m p lific a tion of hea ring nec essa ry to p rec lud e or lim it the a ud itory
sensory d ep riva tion effec ts (Ross, 1996:13). The a m p lific a tion p roc ess
b eg ins d irec tly a fter the d ia g nosis of a hea ring loss ha s b een m a d e.
Func tiona l eva lua tion of a hea ring a id seeks to d eterm ine w hether the
c hild b enefits from suc h a m p lific a tion. The func tiona l eva lua tion of
hea ring a id s is a s essentia l a s a n elec troa c oustic eva lua tion thereof. The
a id ed a ud iog ra m c a n eva lua te w hether the c hild is a b le to hea r soft
sound s w ithin exp ec ta tions, b a sed on the elec troa c oustic fitting of the
hea ring a id . Kuk (2004:1) a lso m a inta ins tha t using the levels ob ta ined
throug h func tiona l g a in m ust b e a rea ssura nc e to the p a rent to ensure
tha t the op tim a l op p ortunity is g iven to d evelop a c hild ’ s p otentia l.
Althoug h ABR m ea sures ha ve b een used in the p a st to a ssist in the fitting
of hea ring a id s in c hild ren, the c linic a l use of these p roc ed ures a re
tec hnic a lly c ha lleng ing (Ga rnha m et a l., 2000:268; Ma honey, 1985:351).
ASSR’ s ha ve
b een
used
to
d em onstra te
the
g a in
p rovid ed
by
a m p lific a tion (Pic ton et a l., 1998:315). ASSR’ s c a n b e ob ta ined in the
sound -field c ond ition – m ea suring a n una id ed resp onse a s w ell a s the
a id ed resp onse. The d ifferenc e in ASSR threshold ob ta ined in the a id ed
c ond ition is then used to p red ic t the func tiona l g a in of the hea ring a id
(Cone-Wesson, 2003:272). Ac c ord ing
to
Gloc kner in
Cone-Wesson
(2003:272), hea ring a id s a p p ea r to tra nsd uc e the m od ula ted tones w ith
g ood fid elity; the sp ec tra l c ha ra c teristic s of the m od ula ted tones p la yed
throug h a n a na log hea ring a id w ith no c om p ression a re w ell p reserved .
The fa c t tha t the stim uli a re m uc h m ore sta b le over tim e tha n b rief
9
University of Pretoria etd – Stroeb
bel, D (2006)
tra nsients m ea ns tha t they a re m ore relia b ly tra nsferred throug h the free
field sp ea kers a nd hea ring a id s – even w hen the hea ring a id s a re nonlinea r (Pic ton et a l., 2002:66). After b eing d ia g nosed w ith a hea ring loss
a nd fitted a p p rop ria tely w ith hea ring a id s, the a d eq ua c y of the fitting
need s to b e va lid a ted . Va lid a tion is a n ong oing p roc ess d esig ned to
ensure tha t the infa nt is rec eiving op tim a l sp eec h inp ut from others a nd
tha t
his or her ow n
sp eec h
is a d eq ua tely p erc eived
(Ped ia tric
Am p lific a tion Protoc ol, 2003:15).
1.4
PROBLEM STATEMENT
The d a w n of a n era of ea rly id entific a tion of hea ring loss in new b orns a nd
infa nts c onfronts a ud iolog ists w ith new c ha lleng es a nd op p ortunities. The
a d vent of universa l new b orn hea ring sc reening ha s m a d e it a ll the m ore
c om m on for a ud iolog ists to see infa nts less tha n tw o to three m onths of
a g e w ho ha ve b een id entified a s b eing a t risk of ha ving a hea ring loss.
The p roc ess of fitting a hea ring a id or d eterm ining the c a nd id a c y for
c oc hlea r im p la nta tion req uires d eta iled know led g e of these infa nts’
resid ua l hea ring a b ilities (Va nd er Werff et a l., 2002:228). For new b orns a nd
infa nts, evoked p otentia l estim a tes of a ud iom etric threshold s m a y b e the
only inform a tion a b out hea ring sta tus tha t is a va ila b le a t the tim e w hen
these c ritic a l d ec isions need to b e m a d e.
The tra nsform a tion of new d isc overies into p ra c tic a l c linic a l p roc ed ures
ha s b een a freq uent oc c urrenc e in a ud iolog ic a l test d evelop m ent over
the p a st three d ec a d es (Gorg a , 1999:29). Severa l rec ent stud ies ha ve
therefore exp lored the rela tionship b etw een ASSR elec trop hysiolog ic a l
threshold s a nd a ud iom etric b eha viora l threshold s for norm a l-hea ring a nd
hea ring im p a ired listeners (Dim itrijevic et a l., 2002:205; Herd m a n & Sta p ells,
10
University of Pretoria etd – Stroeb
bel, D (2006)
2001:41; Lins et a l., 1996:81; Ra nc e et a l., 1995:499). These investig a tors
ha ve rep orted find ing sig nific a nt c orrela tions b etw een ASSR threshold s
a nd b eha viora l a ud iom etric threshold s for ind ivid ua ls w ith a ra ng e of
hea ring losses. Other stud ies ha ve foc used on the c orrela tions b etw een
the ASSR a nd ABR a s threshold p red ic tion tec hniq ue (Cone-Wesson et a l.,
2002:173, Va nd er Werff et a l., 2002:227). Althoug h these results a p p ea r
p rom ising , it is d iffic ult to m a ke d efinite c onc lusions a b out the a p p lic a tion
of ASSR to the infa nt p op ula tion, a s these stud ies w ere b a sed on the
resp onses of a d ults or old er c hild ren ra ther tha n infa nts to eva lua te the
effic a c y of the ASSR a s a threshold estim a tion tool.
Sta p ells (2002a :14 & 2004: c onferenc e p resenta tion) c a utions tha t too few
stud ies a re a va ila b le c onc erning the infa nt p op ula tion to rec om m end the
ASSR m ethod for c linic a l use. How ever, the p otentia l a d va nta g es of the
ASSR tha t c om e from c ontinuous ra ther tha n tra nsient stim uli, inc lud ing
p otentia lly b etter freq uenc y-sp ec ific ity a nd the a b ility to ob ta in hig her
outp ut levels, w a rra nt further investig a tion of the c linic a l a p p lic a tion of the
ASSR in the infa nt p op ula tion.
In a d d ition to the need of a tool for freq uenc y sp ec ific estim a tions of
hea ring in infa nts the va lid a tion of a m p lific a tion ea rly on is a lso a n
essentia l c om p onent. Seew a ld
(2001:70) em p ha sizes the need
for
im p roving the q ua lity of p ed ia tric hea ring a id fitting , a s the c onseq uenc es
of d ec isions m a d e w ill b e w ith a c hild forever. Yet, a fter fitting hea ring a id s
on infa nts, va lid a tion of the fitting in m ost c a ses oc c urs throug h the use of
sub jec tive q uestionna ires a nd va ria b les b eing eva lua ted suc h a s a ud itory
a w a reness, sp eec h-p rod uc tion a b ilities, ra te of la ng ua g e a c q uisition a nd
soc ia l d evelop m ent (Sc ollie & Seew a ld , 2002:702) Aid ed threshold s a re
g enera lly d one only w hen the infa nt is m a ture enoug h to c om p lete
11
University of Pretoria etd – Stroeb
bel, D (2006)
b eha viora l a ud iom etry w hic h m a y b e severa l m onths a fter the initia l
fitting . In the a g e of ea rly id entific a tion, this need s to b e a d d ressed . A
lim ited num b er of stud ies on ASSR a nd func tiona l g a in ha ve b een
p erform ed (Gloc kner in Cone-Wesson, 2003:272; Pic ton et a l., 2002:63;
Pic ton et a l., 1998:315). These stud ies foc used on a d ults a nd old er c hild ren
a nd a lthoug h the results a re p rom ising , the a p p lic a tion p ossib ilities of the
ASSR in a d d ressing the sp ec ific need s of the infa nt p op ula tion need
further investig a tion.
Bess (2000:250) a nd Gra vel (2005:19) urg e a ud iolog ist to c ollec t, eva lua te
a nd integ ra te evid enc e a b out p roc ed ures in ord er to b ec om e evid enc eb a sed p ra c titioners1. This im p lies tha t a s new p roc ed ures b ec om e
a va ila b le, c linic ia ns m ust b e w illing to c ontinua lly eva lua te a nd m od ify
their c linic a l p rotoc ols. Therefore, w ith the a d vent of the ASSR in c linic a l
p ra c tic e a nd in lig ht of the c ruc ia l im p orta nc e of ea rly id entific a tion of
hea ring loss a nd of the intervention p roc ess tha t follow s, the q uestion tha t
a rises is:
Wha t is the clinica l va lue of Auditory Stea dy Sta te Response for ea rly
dia gnosis a nd for eva lua tion of a mplifica tion in infa nts with hea ring loss?
It w a s the p urp ose of this resea rc h end ea vor to find a nsw ers to this
p a rtic ula r q uestion.
1
Evid enc e Ba sed Pra c tic e is a n a p p roa c h to c linic a l servic e d elivery tha t ha s b ec om e
inc rea sing ly a d voc a ted in the p a st d ec a d e. EBP is d efined a s the ‘ c onsc ientious, exp lic it,
a nd jud ic ious use of c urrent b est evid enc e in m a king d ec isions a b out the c a re of
p a tients (Oxford -Centre for Evid enc e Ba sed Med ic ine, 2004: online).
12
University of Pretoria etd – Stroeb
bel, D (2006)
1.5
DEFINITION OF TERMS
For the p urp oses of this stud y, the follow ing term s w ill b e d efined a nd
d isc ussed in ord er to p rom ote a m utua l und ersta nd ing of the b a sic a nd
p rim a ry c onc ep ts d ea lt w ith in it. The term s p rovid ed a re a d a p ted from
the w ork of Mend el, Da nha uer & Sing h (1999) in the Illustra ted Dic tiona ry
of Aud iolog y, unless otherw ise sta ted .
Amplifica tion of a hea ring loss – a m p lific a tion refers to a n inc rea se in the
intensity of sound s. This is a c ollec tive term used for d evic es suc h a s
hea ring a id s. When referring to hea ring a id a ssessm ent, the term
func tiona l g a in is often used w hen va lid a ting a hea ring a id fitting .
Func tiona l g a in refers to the d ifferenc e in p erform a nc e b etw een a id ed
a nd una id ed threshold s m ea sures.
Auditory Evoked Potentia l – elec tric a l a c tivity evoked b y sound s a rising
from a ud itory p ortions of the p erip hera l or c entra l nervous system tra veling
from c ra nia l nerve VII to the c ortex, rec ord ed w ith elec trod es a nd a lso
know n a s a ud itory evoked resp onse. In this stud y the foc us w ill b e on the
follow ing evoked p otentia ls:
•
Aud itory Bra instem Resp onse (ABR) – a n ob jec tive test tha t
m ea sures the elec tric a l p otentia l p rod uc ed in resp onse to sound
stim uli b y the sync hronous d isc ha rg e of the first- throug h sixth- ord er
neurons in the a ud itory nerve a nd b ra instem ; a lso know n a s
b ra instem a ud itory evoked p otentia l (BAEP) a nd b ra instem a ud itory
evoked resp onse (BAER).
•
Aud itory Stea d y Sta te Resp onse (ASSR) – a n a ud itory evoked
p otentia l in w hic h the resp onse w a ve-form a p p roxim a tes the ra te
13
University of Pretoria etd – Stroeb
bel, D (2006)
of stim ula tion; a lso referred to a s stea d y-sta te evoked p otentia l
(SSEP).
Clinica l va lue – Aud iolog y is the hea lth-c a re p rofession d evoted to
hea ring . It is a c linic a l p rofession tha t ha s a s its uniq ue m ission the
eva lua tion of hea ring a b ility a nd the a m eliora tion of im p a irm ent tha t
results from hea ring d isord ers (Sta c h, 1998:2). Ped ia tric a ud iolog ists p la y a
c ruc ia l role in ea rly id entific a tion of hea ring im p a irm ent in infa nts a nd
eva lua tion of their hea ring a b ilities. In a d d ition, p ed ia tric a ud iolog ists
eva lua te the need
for hea ring a id
a m p lific a tion in the p ed ia tric
p op ula tion a nd m onitor the suc c ess of these fitting s. This stud y foc uses on
the p otentia l va lue of the ASSR a s a n a ssessm ent tool tha t c ould a id the
p ed ia tric a ud iolog ist in fulfilling his/ her c linic a l resp onsib ilities.
The ASSR ha ve b een used in a ud iolog y resea rc h c enters a round the
w orld . The results from the c linic a l stud ies ha ve show n tha t ASSR threshold s
c a n b e used to p red ic t p ure-tone threshold s in sleep ing infa nts a nd young
c hild ren. It ha s a lso show n suc c ess in eva lua ting hea ring a id fitting s b y
d eterm ining func tiona l g a in. As w ith other d isc overies in the field of
a ud iolog y
w here
tra nsform a tion
of
new
d isc overies into
c linic a l
p roc ed ures ha s oc c urred , this stud y investig a tes the a d op tion of the ASSR
into the c linic a l setting , c om p a ring this p rom ising tec hniq ue w ith the
tra d itiona l a p p roa c hes used in the d iffic ult-to-test p op ula tions.
Ea rly dia gnosis – The Hea lthy Peop le 2000 initia tive esta b lished the g oa l to
red uc e the a vera g e a g e a t w hic h c hild ren w ith sig nific a nt hea ring
im p a irm ent a re id entified to no m ore tha n 12 m onths of a g e b y the yea r
2000 (Diefend orf, 2002:469). With the im p lem enta tion of universa l hea ring
sc reening p rog ra m s, the Joint Com m ittee on Infa nt Hea ring (1994)
14
University of Pretoria etd – Stroeb
bel, D (2006)
rec om m end a tions w ere tha t a ll infa nts w ith hea ring loss should b e
id entified b y three m onths of a g e, a nd rec eive intervention b y six m onths
of a g e. In this stud y the tim e of id entific a tion va ried b etw een three
m onths of a g e a nd 6 m onths of a g e. Intervention (a m p lific a tion) w a s
im p lem ented im m ed ia tely a fter d ia g nosis – trying to c onform to the
g uid elines of the JCIH.
Infa nt - refers to a c hild d uring ea rliest p eriod of life – b efore a g e 1 a fter
the neona ta l p eriod (The c onc ise Oxford d ic tiona ry, 1982:512).
1.6
DIVISION OF CHAP TE R S
A resea rc h end ea vor, c onsisting of b oth a n em p iric a l a nd theoretic a l
c om p onent w a s c ond uc ted , in ord er to a nsw er the resea rc h q uestion
sta ted a b ove. The follow ing sec tion d elinea tes the d ivision of c ha p ters
a nd p rovid es a short sum m a ry of the c ontents of ea c h c ha p ter.
Cha pter one:
Ba ckground a nd ra tiona le
This c ha p ter p rovid es a n overview of the im p orta nc e of the need of
elec trop hysiolog ic a l p roc ed ures in the d ia g nostic p roc ess of hea ring loss
in young infa nts a nd the d iffic ult-to-test p op ula tion. The ABR is c ontra sted
w ith the ASSR tec hniq ue w ith reg a rd to its p otentia l for estim a ting p ure
tone b eha viora l threshold s. The use of ASSR in estim a ting func tiona l g a in in
the young p op ula tion is d isc ussed . The ra tiona le for the stud y a nd the
p rob lem sta tem ent is p rovid ed . Definitions of the term s a nd c onc ep ts
fund a m enta l to this stud y a re p rovid ed a nd c la rified .
15
University of Pretoria etd – Stroeb
bel, D (2006)
Cha pter two:
Clinica l a pplica tion of Auditory Evoked Potentia ls in
infa nts: Compa ring the a uditory bra instem response
a nd a uditory stea dy sta te response
The c urrent p roc ed ures of c hoic e for ea rly intervention for infa nts a re
d isc ussed – c onsid ering ea rly id entific a tion a nd d ia g nosis of hea ring loss
a nd a m p lific a tion for infa nts w ith hea ring loss. Attention is g iven to the
d ia g nostic p roc ess a nd hea ring a id fitting in the young infa nt p op ula tion foc using on the p rob lem s a nd c ha lleng es. A c ritic a l d isc ussion of AEP’ s in
p ed ia tric a ud iolog y w ill follow therea fter – c om p a ring the ABR m ethod
w ith the ASSR m ethod .
Cha pter three:
Resea rch Methodology
This c ha p ter d esc rib es the op era tiona l fra m ew ork im p lem ented to
c ond uc t this stud y. The a im s of this p resent stud y a re outlined . The
resea rc h d esig n a nd m ethod a re d isc ussed . The ethic a l issues rela ted to
this stud y a re c onsid ered . The sub jec ts, m a teria l a nd a p p a ra tus used in
the stud y a re d esc rib ed a s w ell a s the p roc ed ure tha t w a s follow ed to
c ond uc t this stud y.
Cha pter four:
Results a nd Discussion
The results a re p resented a c c ord ing to the sub -a im s stip ula ted in c ha p ter
three in ord er to a d d ress the m a in a im of the stud y. The results a re
p resented – utilizing the results from ea c h ind ivid ua l sub jec t. Therea fter the
c ollec tive results of the six sub jec ts w ill b e c onsid ered . Interp reta tion a nd
d isc ussion of the results a re p erform ed . The va lue a nd m ea ning of the
16
University of Pretoria etd – Stroeb
bel, D (2006)
resea rc h find ing s in rela tion to other stud ies a nd litera ture in this reg a rd is
d isc ussed .
Cha pter five:
Conclusions a nd Implica tions
The results from this stud y a re sum m a rized . This c ha p ter p rovid es a n outline
of the sig nific a nt results a nd the w a y they c ontrib ute to c urrent litera ture.
Using c ritic a l a p p ra isa l m ethod s, the resea rc h evid enc e a re a ssessed –
c onsid ering the va lue, va lid ity, relia b ility a nd releva nc e thereof. Future
resea rc h rec om m end a tions a re p rovid ed a nd a c onc lusion reg a rd ing the
stud y is form ula ted .
1.7
SUMMARY
This c ha p ter a im ed to p rovid e releva nt b a c kg round inform a tion in ord er
to foc us on the resea rc h end ea vor a nd to p rovid e a b roa d p ersp ec tive
of the ra tiona le und erlying the stud y. Attention w a s d ra w n to the infa nt
p op ula tion a s a d iffic ult-to-test p op ula tion a nd the sp ec ia l need for
ob jec tive a ud iom etric m ea sures in this p op ula tion a t a tim e w hen c ritic a l
d ec isions need to b e m a d e a b out intervention stra teg ies.
17
University of Pretoria etd – Stroeb
bel, D (2006
Cha p ter 2
CLINICAL APPLICATION OF AUDITORY EVOKED POTENTIALS IN
INFANTS: COMPARING THE AUDITORY BRAINSTEM RESPONSE AND
AUDITORY STEADY STATE RESPONSE
This c ha p ter a im s to p rovid e a theoretic a l b a c kg round to the em p iric a l
resea rc h a nd p rovid es a c ritic a l eva lua tion a nd interp reta tion of the
releva nt litera ture p erta ining to the sc op e of this stud y
2.1
INTRODUCTION
‘ From the m om ent tha t Aud itory Evoked Potentia ls (AEP) w ere first
rec ord ed , a ud iolog ists soug ht to exp loit the resp onses in ord er to eva lua te
the hea ring sta tus of p ersons d iffic ult to test’ (Jerg er, 1998: ed itoria l). The
use of AEP’ s for estim a tion of hea ring sensitivity a nd infa nt hea ring
sc reening ha s ha d a m a jor im p a c t on the a b ility to id entify hea ring
im p a irm ent in c hild ren, a s this p rovid es a n ob jec tive m ea ns of a ssessing
the integ rity of the p erip hera l a nd c entra l a ud itory system s (Sta c h,
1998:293). The Aud itory Bra instem Resp onse (ABR) ha s b ec om e the m ost
w id ely c linic a lly used AEP in estim a ting hea ring threshold s, b ut for the p a st
few d ec a d es a n evoked p otentia l, p a rtic ula rly suited for freq uenc ysp ec ific m ea surem ents, the Aud itory Stea d y Sta te Resp onse (ASSR), ha s
c om e und er c lose sc rutiny (Hood , 1998:117). In a d d ition to estim a ting
hea ring sensitivity in infa nts, the ASSR p rom ises to p rovid e a b etter
eva lua tion of hea ring a id p erform a nc e (Sw a nep oel, Sc hm ulia n & Hug o,
18
University of Pretoria etd – Stroeb
bel, D (2006
2002:52), w hic h is a n im p orta nt c om p onent in the va lid a tion of hea ring
a id fitting s.
This c ha p ter therefore exp lores the c linic a l a p p lic a tion of Aud itory Evoked
Potentia ls – c om p a ring the ABR a nd ASSR a s a n ob jec tive p roc ed ure in
the d ia g nosis of hea ring loss a nd va lid a tion of hea ring a id fitting in infa nts.
In the first sec tion, the c urrent p roc ed ures of c hoic e for ea rly intervention
for infa nts w ill b e d isc ussed und er the follow ing tw o sub -hea d ing s: Ea rly
identifica tion a nd dia gnosis of hea ring loss a nd a mplifica tion for infa nts
with hea ring loss. After la ying this found a tion, a c ritic a l d isc ussion of AEP’s
in pedia tric a udiology w ill follow .
2.2
EARLY INTERVENTION FOR INFANTS WITH HEARING LOSS
Aud iolog ists a re entering a p a rtic ula rly op tim istic era for the p rovision of
ea rly intervention servic es. There a re tec hnolog ic a l a d va nc es resulting in
m uc h
ea rlier
id entific a tion
of
c hild hood
hea ring
loss,
im p roved
a m p lific a tion d evic es p rovid ing enha nc ed a ud ib ility, a nd inc rea sed
op p ortunities for fa m ilies to rec eive interventions tha t a re resp onsive to
fa m ily-id entified need s (Moeller, 2001:109).
The Joint Com m ittee on Infa nts Hea ring (JCIH) therefore end orses ea rly
d etec tion of a nd intervention for infa nts w ith hea ring loss throug h
integ ra ted , interd isc ip lina ry system s of universa l hea ring
eva lua tion,
a nd
fa m ily-c entered
intervention
(Northern
sc reening ,
&
Dow ns,
2002:269). This very ea rly intervention m a xim izes the p rosp ec ts tha t these
p a tients w ill a c q uire the c om m unic a tion skills nec essa ry to a c hieve their
full p otentia l (Kirkw ood , 2002: ed itoria l).
19
University of Pretoria etd – Stroeb
bel, D (2006
2.2.1 Ea rly identifica tion a nd dia gnosis of hea ring loss
Hea ring loss is a n im p orta nt hea lth p rob lem in c hild hood tha t severely
im p a c ts on q ua lity of life. The id entific a tion of p erm a nent hea ring
im p a irm ent is the first step in a lifelong p roc ess for ea c h infa nt (Seew a ld ,
2000: vii). Ea rly id entific a tion of hea ring loss in c hild ren ha s a lw a ys b een a
long sta nd ing c linic a l p riority in a ud iolog y, a s hea ring loss tha t g oes
und etec ted
d evelop m ent
in
infa nts a nd
a nd
p ersona l
young
c hild ren
a c hievem ent
c om p rom ises op tim a l
(Diefend orf,
2002:469).
La ng ua g e a nd c om m unic a tion serve a s a found a tion for norm a l c hild
d evelop m ent, a nd d ela ys in the a c q uisition of these skills a ffec t litera c y,
a c a d em ic a c hievem ent, a nd soc ia l a nd p ersona l d evelop m ent (Ha yes &
Northern, 1997:4). Id entific a tion of a c hild ’ s hea ring loss a t a n ea rly a g e is
therefore the first step in a c om p rehensive p la n tha t a llow s for ea rly
m ed ic a l m a na g em ent, c onsid era tion of a c oustic a m p lific a tion, a nd
p la c em ent in a n ea rly intervention p rog ra m (Diefend orf & Web er,
1994:43).
With the p ositive effec t of ea rly id entific a tion, the Joint Com m ittee on
Infa nt Hea ring (2000) in the USA rec om m end s tha t, w henever p ossib le,
d ia g nostic testing should b e c om p leted a nd ha b ilita tion should b eg in b y
the tim e a n infa nt w ith a c ong enita l hea ring im p a irm ent rea c hes the a g e
of six m onths. The effec tiveness of the ea rly intervention p roc ess hing es on
the a ud iolog ist’ s a b ility to a c c ura tely p red ic t hea ring threshold s in the first
m onths of life. The p rim a ry ob jec tive in a ssessing the hea ring of a n infa nt
or young c hild is to ob ta in relia b le, ea r-sp ec ific a nd freq uenc y-sp ec ific
inform a tion on a ud itory func tion a s soon a fter b irth a s p ossib le (Ba c hm a nn
& Ha ll, 1998:4). This ob jec tive c a n c urrently only b e m et throug h the use of
a ud itory evoked p otentia ls (AEP) (Sining er & Cone-Wesson, 2002:298).
20
University of Pretoria etd – Stroeb
bel, D (2006
AEP’ s ha ve b een used in d ia g nostic a ud iolog y for m ore tha n three
d ec a d es a nd is b ec om ing inc rea sing ly p rom inent a s the a g e of hea ring
loss id entific a tion is b eing red uc ed sig nific a ntly d ue to Universa l New b orn
Hea ring Sc reening (UNHS) p rog ra m s (Roeser, Va lente & Hosford -Dunn,
2000:10).
The c ha lleng e of a c c ura tely d eterm ining the hea ring sta tus of a n infa nt or
young c hild is relia nt on sp ec ia lized tra ining a nd extensive c linic a l
exp erienc e (Ha yes & Northern, 1997:234). No sing le a ud itory test is p rec ise
enoug h to b e a p erfec t a nd c om p lete a ssessm ent tool. Defining the
na ture a nd d eg ree of a n infa nt’ s hea ring loss req uires the use of m ultip le
tests a nd tec hniq ues. The need for a test-b a ttery a p p roa c h in p ed ia tric
a ssessm ent c a n therefore not b e oversta ted (Diefend orf, 2002:473). The
b a sic p ed ia tric hea ring eva lua tion inc lud es a thoroug h d evelop m enta l
history, follow ed b y b eha viora l freq uenc y-sp ec ific threshold tests, a c oustic
im m itta nc e m ea surem ents, otoa c oustic em ission tests (OAE) a nd ABR a s
nec essa ry (Ha yes & Northern, 1997:234). The p ed ia tric hea ring eva lua tion
typ ic a lly is a n ong oing a c tivity a nd should b e a d a p ta b le to d ifferent
c irc um sta nc es (Ha yes & Northern, 1997:234).
With the a g e of id entific a tion d ec rea sing , b eha viora l c ond itioning of
neona tes a nd very young infa nts to sound field a ud itory stim uli is not
fea sib le (Diefend orf & Web er, 1994:56). An a c oustic im m itta nc e test
b a ttery c a n b e used to c a teg orize the na ture of the hea ring loss into
c ond uc tive, c oc hlea r, or b ra instem p a tholog y (Northern & Dow ns,
2002:211; Ha yes & Northern, 1997:251). Althoug h im m itta nc e c a n p rovid e
va lua b le inform a tion, it c a nnot p red ic t the d eg ree, c onfig ura tion, typ e
a nd sym m etry of the hea ring loss.
21
University of Pretoria etd – Stroeb
bel, D (2006
With the introd uc tion of c linic a l d evic es in 1988 for m ea suring evoked
otoa c oustic em issions, this tec hniq ue ha s b ec om e a rela tively rec ent
a d junc t
to
nonb eha viora l
p hysiolog ic -b a sed
a ud itory
resp onse
m ea surem ents (Ha ll, 2000:2). The p resenc e of EOAE’ s ha s p roven to b e
evid enc e of a norm a l func tioning c oc hlea a nd p erip hera l hea ring system .
How ever, Rob inette & Gla ttke (2000:506) c a utions tha t OAE’ s c a nnot b e
used to estim a te the a m ount of hea ring loss. The a p p lic a tion of OAE’ s
inc lud e the sc reening for hea ring loss in the new b orn a nd p ed ia tric
p op ula tion, a ug m enting b eha viora l test results in d iffic ult-to-test p a tients,
d evelop ing a true d ifferentia l d ia g nosis in term s of sep a ra ting hea ring loss
into “ sensory” a nd “ neura l” c om p onents a nd id entifying ind ivid ua ls w ith
sub tle a b norm a lities of CNS func tion (Rob inette & Gla ttke, 2000:506).
In ord er to ob jec tively m ea sure the neura l resp onses b eyond the sensory
resp onse of the c oc hlea , AEP’ s m ust b e em p loyed . Mend el, Da nha uer &
Sing h (1999:7) d efines AEP’ s a s elec tric a l a c tivity evoked b y sound s a rising
from a ud itory p ortions of the p erip hera l or c entra l nervous system tra veling
from c ra nia l nerve VIII to the c ortex – a lso know n a s a ud itory evoked
resp onses (AER). Althoug h inferenc es c a n b e m a d e a b out hea ring from
the evoked p otentia l d a ta , it should b e em p ha sized tha t it is not a test of
hea ring , b ut ra ther a test of sync hronous neura l func tion – the a b ility of
the c entra l nervous system to resp ond to externa l stim ula tion in a
sync hronous m a nner (Hood , 1998:95).
The c urrent m ost c om m on c la ssific a tion of AEP a c c ord ing to the la tenc y
ep oc h of the resp onse to b e exa m ined , w a s a d a p ted from the w ork of
Pic ton et a l. in 1974 a nd 1977 a nd Pic ton a nd Fitzg era ld in 1983 (Ferra ro &
Durra nt, 1994:318). The la te la tenc y resp onse (LLR) is the elec tric a l
p otentia ls em a na ting from the surfa c e of the sc a lp in resp onse to a n
22
University of Pretoria etd – Stroeb
bel, D (2006
a ud itory sig na l. These resp onses a re g enera ted b y the c ortex a t tim e
interva ls of 100 to 200 m sec a fter p resenta tion of a n a ud itory stim ulus
(Hood , 1998:4). These inc lud e the N1 c om p lex a nd the P300 (Ferra ro &
Durra nt, 1994:318). The m id d le la tenc y resp onse (MLR) oc c urs b etw een 10
a nd 80 m sec follow ing sig na l onset a nd a re thoug ht to a rise from tha la m ic
a nd p rim a ry c ortic a l p rojec tion a rea s (Hood , 1998:4). The m ost p rom inent
of these is the 40 Hz stea d y sta te p otentia l (SSP) (Ferra ro & Durra nt,
1994:318).
Those AEP’ s oc c urring w ithin the first 10 -15 m sec follow ing stim ulus onset
a re g enera lly referred to a s the “ ea rly” or short la tenc y resp onses (SLR).
The SLR inc lud es the ABR a nd a lso severa l c om p onents p rec ed ing the ABR
tha t a re rec ord ed via elec troc oc hleog ra p hy (ECoc hG) (Burka rd & Sec or,
2002:233). Other SLR inc lud e the slow -neg a tive p otentia l (SN10) a nd the
freq uenc y follow ing resp onse (FFR). The c linic a l use of b oth these SLR’ s ha s
b een oversha d ow ed b y tha t of other AEP’ s like the ABR.
The la te la tenc y resp onses a re p resent in infa nts a nd c hild ren, b ut a re
unrelia b le for threshold estim a tes in sleep ing ind ivid ua ls a nd the rec ord ing
a nd interp reta tion in c hild ren req uire c onsid era b le exp erienc e (Sta p ells,
2000a :13; Ha ll, 1992:107). The m id d le la tenc y resp onses a re not relia b ly
ob ta ined in infa nts a nd young c hild ren, a nd their a b senc e in a n otherw ise
norm a l sleep ing infa nt m a y b e c om p letely norm a l (Sta p ells, 2000a :13).
The ABR ha s none of these lim ita tions a nd ha s b ec om e the p roc ed ure of
c hoic e in the d ia g nostic a ssessm ent of the d iffic ult-to-test p op ula tions
(Sta p ells, 2002:14; Ba c hm a nn & Ha ll, 1998:41; Ha ll & Mueller, 1997:321).
Severa l rec ord ing m ethod s ha ve b een p rop osed in w hic h the ABR c a n b e
used to p red ic t the d eg ree, c onfig ura tion, typ e a nd sym m etry of the
23
University of Pretoria etd – Stroeb
bel, D (2006
hea ring loss (Hood , 1998:98). Ma ny rep orts exist d em onstra ting the
usefulness of these tec hniq ues in the d ia g nostic p roc ess of hea ring loss in
infa nts (Gorg a , 2002:49; Sta p ells, 2000a :16; Gorg a , 1999:31; Ba c hm a n &
Ha ll, 1998:41; Sta p ells & Oa tes, 1997:261). The ASSR ha ve rec ently g a ined
c onsid era b le a ttention a nd c a used exc item ent a m ong a ud iolog ists,
esp ec ia lly those involved in the a ssessm ent a nd sub seq uent hea ring a id
fitting of infa nts w ith hea ring loss (Sta p ells et a l., 2005:43).
2.2.2 Ea rly a mplifica tion for infa nts with hea ring loss
Onc e a hea ring im p a irm ent ha s b een id entified , a c om p lete a ssessm ent
m ust b e p erform ed in a va lid a nd tim ely m a nner. The find ing s from the
a ssessm ent a re used to d evelop the initia l c om p onents of the intervention
for the infa nt’ s entire life (Seew a ld , 2000: vii). Althoug h m a ny g uid elines,
suc h a s the Joint Com m ittee on Infa nt Hea ring (JCIH, 2000:10), c a ll for
a p p lic a tion of intervention p roc ed ures to b eg in no la ter tha n six m onths of
a g e, the c ha lleng e of m eeting suc h a n ob lig a tion is d a unting . The fitting
of hea ring a id s on infa nts ha s a lw a ys p resented p rob lem s d ue to the
lim ited c a p a b ility to utilize sta nd a rd b eha viora l testing tec hniq ues. With
infa nts, hea ring a id s a re fitted on the b a sis of only a few threshold s p er
ea r, w ith no sup ra threshold a ud itory p erc ep tion (Ped ia tric Working Group ,
1996:53). Even w ith the m ore rec ent a d va nc es in infa nt a ssessm ent, the
threshold
p red ic tions a re
useful, b ut
do
not
rep la c e
b eha viora l
a ud iom etry (Sc ollie & Seew a ld , 2002:687). The hea ring a id selec tion,
fitting , verific a tion a nd
va lid a tion p roc ess is therefore a n ong oing
c ha lleng e in this young p op ula tion.
24
University of Pretoria etd – Stroeb
bel, D (2006
2.2.2.1
Ap p roa c hes to p ed ia tric hea ring a id fitting
The im m ed ia te g oa l of sensory a ssista nc e to hea ring im p a ired c hild ren is
to p rovid e a s m uc h sensory inform a tion a s p ossib le w ith reg a rd s to the
sound p a tterns of sp eec h (Boothroyd , 1997:17). The long term g oa l of
enha nc ing sensory c a p a c ity is to inc rea se the sp eed a nd q ua lity of
d evelop m ent of sp oken la ng ua g e skills – to em p loy a d evelop m enta l
ra ther tha n rem ed ia l a p p roa c h (Ross, 1996:13). Suc c ess in m eeting this
long -term g oa l d ep end s not only on a id ed sensory c a p a c ity, b ut a lso on
c om m unic a tive exp erienc e, c om b ined w ith a p p rop ria te c linic a l a nd
ed uc a tiona l m a na g em ent (Boothroyd , 1997:17).
Onc e hea ring loss ha s b een c ha ra c terized , the next step is to d eterm ine
w hether a m p lific a tion should b e w orn (Lew is, 2000:150). Ac c ord ing to The
Ped ia tric Working Group (1996:54), “ threshold s eq ua l to or p oorer tha n 25
d B HL w ould ind ic a te c a nd id a c y for a m p lific a tion in som e form .” As
sta ted b efore, the g oa l of a m p lific a tion is to ensure a ud ib ility of the
sp eec h inp ut, verify tha t sound s a re not unc om forta b ly loud a nd to ensure
c onsistent a ud ib ility a nd hea ring a id p erform a nc e over tim e (Pa lm er,
2005:10; Kuk & Ma rc oux, 2002:504).
Althoug h sim ila r d ec isions a b out a m p lific a tion c ha ra c teristic s m ust b e
m a d e for the infa nt a s for the a d ult, the inform a tion on w hic h these
d ec isions a re b a sed a nd the need s of these tw o g roup s a re q uite d ifferent
(Pa lm er, 2005:11; Bea uc hine & Dona g hy, 1996:145). At the sim p lest level,
infa nts’ ea rs a re sm a ller tha n those of a d ults: a d ifferenc e tha t sig nific a ntly
im p a c ts a m p lific a tion-fitting d ec isions, suc h a s c hoic e of m ould s a nd
c hoic e of p resc rip tive ta rg ets (Pa lm er, 2005:11; Sc ollie & Seew a ld ,
25
University of Pretoria etd – Stroeb
bel, D (2006
2002:687; Dillon, 2001:410; Lew is, 2000:150; Bea uc hine
&
Dona g hy,
1996:145).
Moreover, a ud iolog ic a l inform a tion a va ila b le a t the tim e of hea ring
instrum ent fitting m a y b e lim ited in the c a se of infa nts. The p ed ia tric
a ud iolog ist need s to rely on threshold estim a tes a t the tim e w hen the
hea ring instrum ents a re selec ted . Dela ying a m p lific a tion until c om p lete
a ud iolog ic a l inform a tion is a va ila b le, m a y m ea n tha t the infa nt is w ithout
a m p lific a tion d uring c ritic a l p eriod s of la ng ua g e d evelop m ent (Sc ollie &
Seew a ld , 2002:685; Bea uc hine & Dona g hy, 1996:145).
Furtherm ore, the c om m unic a tion need s of a n infa nt w ho ha s a c ong enita l
hea ring loss a re a lso d istinc t from those of a n a d ult w ho ha s p rog ressive,
la te-onset hea ring loss. Infa nts d iffer from a d ults in how they use
a m p lific a tion. They listen to sp eec h from d ifferent d ista nc es a nd heig hts
a nd a m p lific a tion should a c c ount for these inp ut d ifferenc es. Infa nts a lso
d iffer from a d ults in tha t they use a m p lific a tion to a c q uire sp oken
la ng ua g e. They d o not ha ve the sa m e know led g e b a se tha t a d ults ha ve
w hen a ttem p ting to m a ke sense of a ud itory sig na ls tha t m a y b e d istorted ,
inc om p lete, or a ffec ted b y noise (Sc ollie & Seew a ld , 2002:685; Lew is,
2000:150; Bea uc ha ine & Dona g hy, 1996:145).
Ped ia tric
a m p lific a tion
fitting
p roc ed ures should
therefore
p rovid e
ob jec tive, va lid , a nd relia b le m ea sures of hea ring a id p erform a nc e for
sp eec h-level a nd hig h-level inp uts for the infa nt/ c hild (Pa lm er, 2005:12;
Sc ollie & Seew a ld , 2002:689, Dillon 2001:404). These m ea sures should ta ke
into a c c ount the need s of infa nts a nd c hild ren for a ud itory self-m onitoring
a nd the a c q uisition of a ud itory p roc essing a b ilities throug h a id ed sound .
26
University of Pretoria etd – Stroeb
bel, D (2006
The hea ring a id fitting p roc ess for infa nts c a n b e d esc rib ed a s five
seq uentia l sta g es (Ped ia tric Am p lific a tion Protoc ol, 2003:15; Sc ollie &
Seew a ld , 2002:685; Ped ia tric Working Group , 1996:53). These sta g es a re
sum m a rized in Ta b le 2.1.
Ta ble 2.1
Sta ges of hea ring a id fitting process
Sta ge
Process
¾The
Assessm ent
•
hea ring
loss is m ea sured , a nd
c a nd id a c y for
a m p lific a tion is d eterm ined
¾Num eric
Selec tion
•
ta rg et
for
hea ring
a id
elec troa c oustic
p erform a nc e a re c a lc ula ted , a nd a p p rop ria te hea ring
a id s a re c hosen
¾The hea ring a id s a re a d justed to p rovid e the d esired
Verific a tion
•
elec troa c oustic p erform a nc e
¾Aid ed a ud itory func tion is eva lua ted a nd c om p a red
Va lid a tion
•
w ith ha b ilita tive g oa ls
¾Orienta tion to hea ring a id s a re p rovid ed a nd hea ring a id
Inform a tiona l
•
Counseling
a nd
usa g e is m onitored
follow -up
A short d isc ussion of ea c h of these sta g es w ill follow :
•
Assessm ent
The effic a c y of hea ring a id fitting is p red ic a ted on the va lid ity of the
a ud iolog ic a l a ssessm ent. An essentia l g oa l of the c om p rehensive
a ud iolog ic a l a ssessm ent is to ob ta in ea r- a nd freq uenc y- sp ec ific
estim a tes of hea ring threshold for use a s a sta rting p oint in hea ring
instrum ent fitting a t the ea rliest op p ortunity (Roush, 2005:105; Ped ia tric
Working Group , 1996:54). Com p lete a ud iolog ic a l d a ta is seld om
ob ta ined w hen testing the very young c hild . In the a b senc e of a n
27
University of Pretoria etd – Stroeb
bel, D (2006
a ud iog ra m , hea ring a id fitting should p roc eed on the b a sis of
freq uenc y-sp ec ific ABR threshold estim a tions unless neurolog ic a l sta tus
c ontra -ind ic a tes suc h a c tion (Roush, 2005:105; Sc ollie & Seew a ld ,
2002:689; Ross, 1996:16; Diefend orf, Reitz, Esc ob a r & Wynne, 1996:125).
•
Selec tion
The
Ped ia tric
Working
Group
(1996:54)
rec om m end ed
tha t
infa nts/ c hild ren w ith threshold s p oorer tha n 25 d B HL b etw een 1000
a nd 4000 Hz should b e seen a s c a nd id a tes for a m p lific a tion – either
throug h the use of p ersona l hea ring a id s or som e other form of
a m p lific a tion
(Lew is,
2000:150).
Onc e
the
d ec ision
to
p rovid e
a m p lific a tion ha s b een m a d e, selec tion of hea ring a id s is a c om p lex
p roc ess (Sc ollie & Seew a ld , 2002:691; Bea uc ha ine & Dona g hy,
1996:145). Rec ent a d va nc em ents in hea ring instrum ent tec hnolog y
offer the p otentia l for sig nific a nt im p rovem ent in the la ng ua g e a nd
c om m unic a tion a b ilities a nd overa ll q ua lity of life of infa nts w ith
hea ring loss (Buerkli-Ha levy & Chec kley, 2000:77). It is im p orta nt to
selec t a m p lific a tion b a sed on the full ra ng e of uniq ue c ha ra c teristic s of
ea c h infa nt, inc lud ing the hea ring loss, the fa m ily, the ed uc a tiona l a nd
hom e environm ent, a nd a va ila b le hea ring a id tec hnolog y (BuerkliHa levy & Chec kley, 2000:77; Bea uc ha ine & Dona g hy, 1996:145).
•
Verific a tion
In the c ontext of ea rly intervention, infa nts w ill w ea r their hea ring a id s
a t fixed , c linic ia n- d eterm ined setting s for a long p eriod of tim e (Sc ollie,
2005:91). Rec ent c onsensus sta tem ents ha ve rec om m end ed tha t
hea ring a id p resc rip tion should b e d one in a n ob jec tive m a nner
28
University of Pretoria etd – Stroeb
bel, D (2006
(Ped ia tric Working Group , 1996; Ped ia tric Am p lific a tion Protoc ol, 2003).
At the verific a tion sta g e, ob jec tive hea ring a id p resc rip tions a re used
to p resc rib e sp ec ific a m p lific a tion c ha ra c teristic s (Sc ollie, 2005:91). The
hea ring a id s a re a d justed until they p rovid e the elec troa c oustic
p erform a nc e tha t is d eem ed a p p rop ria te for ea c h infa nt/ c hild (Sc ollie
& Seew a ld , 2002:698; Bea uc ha ine, 2002:106). The outp ut of the
instrum ent is m ea sured ob jec tively a c ross freq uenc y a nd inp ut ra ng es.
This p roc ed ure m ust c onfirm tha t the rea l-ea r p erform a nc e of the
instrum ent p rovid es outp ut levels tha t a re c om forta b le, sa fe, a nd
w ithout feed b a c k. The use of this ob jec tive a p p roa c h results in
c onsistent trea tm ent a c ross infa nts a nd c hild ren (Sc ollie, 2005:91).
•
Va lid a tion
Onc e the p resc rip tive p roc ed ure is c om p lete, a nd the setting s of the
hea ring a id s ha ve b een verified , the va lid a tion p roc ess b eg ins
(Ped ia tric Am p lific a tion Protoc ol, 2003:15). Va lid a tion of a id ed a ud itory
func tion is a c ritic a l c om p onent of the p ed ia tric a m p lific a tion p rovision
p roc ess. The p urp ose of va lid a ting a id ed a ud itory func tion is to
d em onstra te
the
b enefits/ lim ita tions of a n infa nt’ s/ c hild ’ s a id ed
listening a b ilities for p erc eiving sp eec h of others a s w ell a s his/ her ow n
sp eec h (Ped ia tric Am p lific a tion Protoc ol, 200315; Dillon, 2001:106; The
Ped ia tric Working Group , 1996:56). Va lid a tion is a c c om p lished , over
tim e, using inform a tion d erived throug h the a ura l ha b ilita tion p roc ess,
a s w ell a s the d irec t m ea surem ent of the infa nt’ s/ c hild ’ s a id ed a ud itory
p erform a nc e.
29
University of Pretoria etd – Stroeb
bel, D (2006
•
Inform a tiona l c ounseling a nd follow -up
Thoroug h a nd suita b le c ounseling , m onitoring a nd follow -up a re
essentia l in a p ed ia tric hea ring a id fitting p roc ess. Hea ring a id
orienta tion p rog ra m s should inc lud e a ll m em b ers w ho w ill b e a ssisting
the
infa nt (Bea uc hine, 2002:111). Typ ic a l a ud iolog ic a l follow -up
sc hed ules for infa nts a nd young c hild ren a re a t lea st every three
m onths to the a g e of three yea rs. More freq uent visits m a y b e req uired
w hen fitting infa nts young er tha n six m onths of a g e, (Bea uc ha ine,
2002:111).
In the p a st a ud iolog ists ha ve relied on a id ed a ud iog ra m s (a lso know n a s
func tiona l g a in m ea surem ents) a s the p rim a ry verifica tion tool for hea ring
a id fitting s in infa nts a nd young c hild ren (Stelm a c how ic z, Hoover, Lew is &
Brenm a n, 2002:38; Seew a ld , Mood ie, Sinc la ir & Cornelisse, 1996:165;
Hed ley-Willia m s, Thorp e & Bess, 1996:107). Tec hnic a lly, func tiona l g a in is
d efined a s the d ifferenc e in d B b etw een a id ed a nd una id ed sound -field
threshold s a s a func tion of freq uenc y. (Stelm a c how ic z et a l., 2002:38).
Typ ic a lly, the g oa l ha s b een to “ shift” threshold s into the ra ng e of 20-25
d BHL.
Over the yea rs, it ha s b een a c know led g ed tha t severa l lim ita tions a re
a ssoc ia ted w ith the use of func tiona l g a in a p p roa c hes for hea ring a id
verifica tion (Seew a ld , Mood ie, Sinc la ir & Cornelisse, 1996:178).
One serious lim ita tion of this p roc ed ure is rela ted to the form in w hic h the
p erform a nc e
c riteria
a re sp ec ified
(Stelm a c how ic z et a l., 2002:38;
Seew a ld et a l., 1996:178). When a p urely a ud iom etric -b a sed a p p roa c h is
ta ken to the selec tion p roc ess, it is not p ossib le to verify tha t the d esired
30
University of Pretoria etd – Stroeb
bel, D (2006
elec troa c oustic c ha ra c teristic s ha ve b een p rovid ed to the infa nt w ithout
va lid b eha viora l test results. Conseq uently, for infa nts, this a p p roa c h w ill
b e of lim ited use w hen im p orta nt selec tion-rela ted d ec isions need to b e
m a d e (Ped ia tric Am p lific a tion Protoc ol, 2003:13; Dillon, 2001:106). Another
c ritic ism of this p roc ed ure is the p oor test-retest relia b ility (Stelm a c how ic z
et a l., 2002:13).
Func tiona l
g a in
m ea surem ents ind ic a te
only
the
freq uenc y/ g a in
c ha ra c teristic s of a hea ring a id (Seew a ld et a l., 1996:178). There a re
a d d itiona l elec troa c oustic c ha ra c teristic s of hea ring a id s tha t should b e
c onsid ered w ithin the selec tion p roc ess. Consid era tion should b e g iven to
a sp ec ts suc h a s outp ut lim iting , c om p ression threshold s, c om p ression
ra tios a nd c ross-over freq uenc ies. Func tiona l g a in a lso d oes not sup p ly
freq uenc y sp ec ific inform a tion. It g ives inform a tion a c ross the freq uenc y
sp ec trum a t oc ta ve freq uenc ies, b ut the inter oc ta ve freq uenc ies a nd
troug hs a re overlooked . The freq uenc y resolution is therefore p oor (Dillon,
2001:106). Sm a ll c ha ng es in elec tro-a c oustic outp ut of the hea ring a id , or
a c oustic m od ific a tions m a y c rea te a ltera tions in the freq uenc y resp onse
a nd g a in c ha ra c teristic s of the hea ring a id . This w ill not nec essa rily b e
noted in the func tiona l g a in m ea surem ent.
Aid ed a ud iog ra m s d esc rib e hea ring a id func tion for very soft sound s only,
a nd then only a t a few freq uenc ies. In c a ses of severe to p rofound
hea ring loss, m inim a l or m ild loss, or w hen non-linea r sig na l p roc essing ,
d ig ita l noise red uc tion, or a utom a tic feed b a c k red uc tion c irc uitry is used ,
m islea d ing inform a tion m a y b e ob ta ined (Sc ollie & Seew a ld , 2002:688).
Due to the a b ove lim ita tions, c om p uterized rea l-ea r p rob e m ic rop hone
m ea surem ents ha ve b ec om e the p referred p roc ed ure to fit a nd a d just
31
University of Pretoria etd – Stroeb
bel, D (2006
hea ring a id s w ith infa nts. But func tiona l g a in m ea sures d o how ever p la y a
role in the ong oing p roc ess of va lida tion. Dillon (2001:419) em p ha sizes tha t
these m ea surem ents should b e a sup p lem ent to the elec troa c oustic
m ea surem ents. Func tiona l g a in m ea sures ha ve the follow ing uses:
•
It d em onstra tes to the p a rents tha t the c hild is c a p a b le of rea c ting
to sound (Dillon, 2001:419). Aid ed a nd una id ed sp eec h rec ep tion or
sp eec h a w a reness threshold s c a n d em onstra te the b enefit of
a m p lific a tion to p a rents of infa nts. It m a y a lso rule out the p ossib ility
of non-org a nic hea ring loss, neurolog ic a l c ond itions, or a ud itory
neurop a thy (Stelm a c how ic z et a l., 2002:39).
•
It d em onstra tes tha t the hea ring a id m a xim um outp ut exc eed s the
c hild ’ s hea ring
threshold
at
ea c h
freq uenc y
tested
(Dillon,
2001:419).
•
An a id ed threshold a t the level exp ec ted , g iven the hea ring a id
c oup ler g a in a nd una id ed hea ring threshold , p rovid es further
c onfirm a tion of the c hild ’ s una id ed threshold s (Dillon, 2001:419).
•
In the c a se of p rofound hea ring loss, a id ed threshold s a t the
exp ec ted levels c onfirm tha t the una id ed threshold s w ere not
b a sed solely on vib ra tory sensa tions (Dillon, 2001:419). Aid ed
threshold s a re a lso the b est w a y to d oc um ent p erform a nc e for
b one-c ond uc tion instrum ents, freq uenc y-tra nsp osition d evic es a nd
c oc hlea r im p la nts (Stelm a c how ic z et a l., 2002:42).
Va lida tion of a id ed a ud itory func tion is a d em onstra tion of the b enefits
a nd lim ita tions of a id ed hea ring a b ilities a nd b eg ins im m ed ia tely a fter the
fitting
a nd
ob jec tive
elec troa c oustic
verific a tion
of
a m p lific a tion
(Ped ia tric Am p lific a tion Protoc ol, 2003:15). Va lid a tion is a n ong oing
p roc ess d esig ned to ensure tha t the c hild is rec eiving op tim a l sp eec h
32
University of Pretoria etd – Stroeb
bel, D (2006
inp ut from others a nd tha t his or her ow n sp eec h is a d eq ua tely p erc eived
(Ped ia tric
Working
Group , 1996:56; Ped ia tric
Am p lific a tion Protoc ol,
2003:14). Func tiona l g a in is m ea sured b y find ing the hea ring threshold s in
a sound field w hile a p erson is una id ed a nd a g a in w hile a id ed – throug h
the use of b eha viora l a ud iom etric p roc ed ures (Dillon, 2001:106).
Infa nts a re how ever una b le to p rovid e c onc lusive b eha viora l inform a tion.
It m a y therefore b e nec essa ry to inc orp ora te sub jec tive non-tra d itiona l
eva lua tions, suc h a s p a rent q uestionna ires, to g a in b eha viora l inform a tion
a b out the fitting outc om e (Sc ollie & Seew a ld , 2002:701). Without the d a ta
d erived
from
b eha viora l a ssessm ents, it
is d iffic ult
to
a ssess the
p erform a nc e of hea ring a id s even w hen the theoretic a l a m p lific a tion
sp ec ific a tion is know n (Ga rnha m et a l., 2000:267). Ob jec tive m ea sures –
using AEP’ s - to a ssess hea ring a id p erform a nc e w ould p otentia lly a id the
m a na g em ent
of
these
d iffic ult-to-test
sub jec ts a s the
b eha viora l
func tiona l g a in m ea surem ents m a y only b e p erform ed a fter the infa nt ha s
rea c hed a n a p p rop ria te d evelop m enta l a g e w here a resp onse suc h a s
the hea d turn resp onse m a y b e utilized to m ea sure func tiona l g a in.
Therefore
AEP’ s m a y
p rovid e
useful inform a tion
w hen
b eha viora l
func tiona l g a in m ea surem ents a re not rea d ily a va ila b le d ue to the
sub jec t’ s a g e or d evelop m enta l inc a p a c ity. The next sec tion w ill therefore
foc us on AEP’ s in the field of p ed ia tric a ud iolog y.
2.3
CRITICAL EVALUATION OF AEP’s IN PEDIATRIC AUDIOLOGY
There ha s a lw a ys b een a need for ob jec tive tests tha t a ssess a ud itory
func tion
in
infa nts,
young
c hild ren
a nd / or
a ny
p a tient
w hose
d evelop m enta l level p rec lud ed the use of b eha viora l a ud iom etric
tec hniq ues. Althoug h severa l a p p roa c hes ha ve b een tried , for the p a st 25
33
University of Pretoria etd – Stroeb
bel, D (2006
yea rs, tha t need ha s b een m et p rim a rily b y the m ea surem ents of shortla tenc y a ud itory-evoked p otentia ls, p rim a rily the a ud itory b ra instem
resp onse (ABR) (Gorg a & Neely, 2002:49). In rec ent yea rs the Aud itory
Stea d y Sta te Resp onses ha s b ec om e a va ila b le a s a d ifferent tec hniq ue to
m ea suring the b ra in’ s resp onses to sound (Pic ton et a l., 2002:65). In
p ed ia tric a ud iolog ic a l p ra c tic e AEP’ s ha ve p roven to b e ind isp ensa b le for
d ia g nostic p urp oses b ut they ha ve a lso b eg un to d em onstra te the
p otentia l to a ssist b eyond the d ia g nostic p roc ess w ith the va lid a tion of
a m p lific a tion.
In the follow ing sec tion these tw o tec hniq ues w ill b e d isc ussed in term s of
their a p p lic a tion in the field of p ed ia tric a ud iolog y, b oth d ia g nostic a nd in
a m p lific a tion va lid a tion.
2.3.1 Auditory Bra instem Response
The ABR is m ostly used in the a ssessm ent of a ud itory func tion in infa nts,
c hild ren a nd a d ults w ho c a nnot p a rtic ip a te in volunta ry a ud iom etry a nd
is b y fa r the m ost w id ely used AEP in a ud iolog y (Arnold 2000:451; Hood ,
1998:96). The p op ula rity of the ABR stem s from the fa c t tha t it is a rob ust
resp onse tha t va ries very little b etw een ind ivid ua ls (inc lud ing infa nts),
m a king the resp onse fa irly ea sy to id entify und er m ost c irc um sta nc es (Ha ll,
1992:20). It is a lso hig hly sta b le – c ha ra c teristic s of the resp onse d o not
va ry b etw een w a kefulness a nd sleep a nd a re not a ffec ted b y m ost
m ed ic a tions, w hic h m ea n tha t c hild ren m a y b e tested relia b ly d uring
na tura l or sed a tion ind uc ed sleep (Arnold , 2000:455; Ra nc e et a l.,
1995:499). These c ha ra c teristic s ha ve m a d e it the m ost c om m only used
elec trop hysiolog ic a l tool to estim a te hea ring threshold s in d iffic ult-to-test
p op ula tions. The ABR w ill b e d isc ussed in term s of three a p p lic a tions in the
34
University of Pretoria etd – Stroeb
bel, D (2006
field of p ed ia tric a ud iolog y, na m ely: detection, dia gnosis a nd hea ring a id
fitting in infa nts.
2.3.1.1
Detec tion of hea ring loss
Sc reening , or ea rly d etec tion, of d isord ers ha s rec eived inc rea sing
a ttention in hea lth c a re over the la st q ua rter c entury (Feig htner, 1992:1).
The g enera l p rem ise for sc reening , or ea rly d etec tion, c lea rly m a kes sense.
Ea rly d etec tion offers the op p ortunity to rec og nize the c ond ition b efore
sym p tom s a p p ea r, a nd to p revent or d im inish suffering (Feig htner, 1992:2).
Hea ring loss is a n invisib le d isa b ility a nd is nea rly im p ossib le to d etec t
d uring a routine c linic a l exa m ina tion. Thus, if hea ring loss is not d etec ted
throug h new b orn hea ring sc reening p rog ra m s, it often g oes und etec ted
b efore 18 m onths of a g e (Diefend orf, 2002:469; Ha yes & Northern,
1997:214).
Althoug h the ABR is not a d irec t test of hea ring sensitivity, it ha s ea rned a
strong c linic a l rep uta tion a s a va lua b le tool to eva lua te the integ rity of the
a ud itory p a thw a ys (Diefend orf, 2002:471; Sta p ells, 2000a :13). Clic k evoked
ABR’ s c a n b e rec ord ed from infa nts a s young a s 27 w eeks g esta tion a g e,
a lthoug h resp onses m a y b e p oorly form ed (Ha ll, 1992:490). By 33 to 35
w eeks of g esta tion, resp onses a re m ore sta b le, a nd visua l d etec tion level
is c om p a ra b le to tha t of old er infa nts. Tra d itiona l ABR sc reening
d ep end ed on id entific a tion of w a ve V a t 30-40 d BnHL (Northern & Dow ns,
2002:285).
Autom a ted ABR (AABR) system s ha ve b een d evelop ed a nd used
sp ec ific a lly for hea ring sc reening p urp oses. The a utom a ted ABR system s
use a rule-d irec ted , sta tistic a l m ethod to d etec t a resp onse – thus
35
University of Pretoria etd – Stroeb
bel, D (2006
elim ina ting sub jec tive resp onse rec og nition (Cone-Wesson, 2003:266).
These a utom a tic d etec tion a lg orithm s w orks b y c om p a ring the online
resp onses from the infa nt w ith a ‘ norm a l’ tem p la te resp onse p a ttern
ob ta ined from a la rg e sa m p le p op ula tion of new b orns. If the test infa nt’ s
resp onses c orrela te w ith the norm a tive d a ta , the a utom a ted instrum ent
rend ers a ‘ p a ss’ d ec ision. If there is no c orrela tion b etw een the ‘ norm a l’
tem p la te a nd the test infa nt’ s resp onses, a ‘ refer’ resp onse is ob ta ined –
sug g esting the need for further testing (Northern & Dow ns, 2002:285).
These AABR system s a re entirely ob jec tive a nd a re p rog ra m m ed to
d eterm ine p a ss or refer c riteria for infa nts young er tha n six m onths of a g e.
A c lic k stim ulus is used w hen elic iting a n AABR. The c lic k ABR a c c ura tely
a p p roxim a tes b eha viora l p ure tone threshold s in the m id d le to hig h
freq uenc y reg ions (Sining er & Cone-Wesson, 2002:303) – therefore lim iting
d etec tion of hea ring loss in d ifferent freq uenc y ra ng es (Sta p ells, Gra vel &
Ma rtin, 1995:361). Inform a tion from this sing le intensity sc reening test is
insuffic ient to p red ic t d eg ree of hea ring im p a irm ent or the site of
d ysfunc tion (Ha yes & Northern, 1997:256). The a d va nta g es a nd lim ita tions
of the c lic k evoked ABR w ill b e d isc ussed in d eta il in the follow ing sec tion.
2.3.1.2
Dia g nosis of hea ring loss
i.
ABR threshold eva lua tions using c lic ks
The m ost w id ely used evoked p otentia l m ethod for eva lua ting a ud itory
threshold is the ABR to non-m a sked b roa d b a nd c lic ks (Sta p ells & Oa tes,
1997:258). The c lic k-evoked ABR c onsists of a series of seven p ositive-toneg a tive w a ves, oc c urring w ithin a b out 10 m s a fter stim ulus onset (Arnold ,
2000:451). It w a s not until the la te 1960’ s tha t elec tric a l p otentia ls
36
University of Pretoria etd – Stroeb
bel, D (2006
g enera ted b y the b ra instem w ere id entified in the la b ora tories of Jew ett
a nd c ollea g ues in the USA a nd Sohm er a nd Feinm esser in Isra el (Ha ll &
Mueller, 1997:322; Hood , 1998:5). Jew ett a nd c ollea g ues d em onstra ted
tha t neura l resp onses c ould b e rec ord ed from the b ra instem p a thw a ys –
show ing a resp onse c om p osed of a series of five to seven p ea ks (Burka rd
& Déc or, 2002:233). It is g enera lly a g reed tha t the ABR is g enera ted b y the
a ud itory nerve a nd sub seq uent fib er tra c ts a nd nuc lei w ithin the a ud itory
b ra instem p a thw a ys. A series of Rom a n num era ls (from I to VII) w ere
a ssig ned to the p ea ks. These d esig na tors ha ve b een used sinc e tha t tim e
to id entify the va rious c om p onents of the ABR (Hood , 1998:5). The m ost
w id ely used ABR m ea sure is the la tenc y of a c om p onent p ea k (Don &
Kw ong , 2002:274).
The c lic k-evoked ABR yield s the c lea rest ABR resp onse for threshold
estim a tion a s this rob ust resp onse va ries little b etw een ind ivid ua ls a nd is
ea sy to id entify (Ha ll, 1992:20; Arnold , 2000:455). In a ssessing hea ring
sensitivity, w a ve V of the ABR is used b ec a use it is the m ost rob ust of the
w a ves a nd
the one b est c orrela ted
w ith b eha viora l a ud iom etric
threshold s (Arnold , 2000:456). The low est c lic k level a t w hic h Wa ve V c a n
b e elic ited p rovid es inform a tion a b out the d eg ree of hea ring loss (Arnold ,
2000:456).
How ever, the ra p id onset of the c lic k, a nd its b roa d freq uenc y sp ec tra l
c ontent, results in a c tiva tion of a w id e a rea of the b a sila r m em b ra ne.
Sinc e a b roa d ra ng e of freq uenc ies is stim ula ted , it is not p ossib le to
ob ta in
a c c ura te
inform a tion
a b out
hea ring
sensitivity
at
d ifferent
freq uenc ies using a non-m a sked c lic k a lone (Sta p ells & Oa tes, 1997:248).
When using freq uenc y-sp ec ific stim uli, there is a tra d e-off b etw een
freq uenc y sp ec ific ity a nd neura l sync hrony (Hood , 1998:96; Ha ll 1992:123).
37
University of Pretoria etd – Stroeb
bel, D (2006
The a c oustic p rinc ip le und erlying this tra d e-off, involves the rela tionship
b etw een the d ura tion of the stim ulus a nd its freq uenc y c ontent – the
long er the d ura tion, the m ore freq uenc y sp ec ific it w ill b e.
Another a sp ec t influenc ing the freq uenc y sp ec ific ity of the c lic k ABR is the
tra nsd uc er. A 100-m ic rosec ond elec tric a l p ulse, im p ressed on a sta nd a rd
ea rp hone,
g enera tes a
b roa d b a nd
sig na l
(c lic k)
w hose
p rim a ry
freq uenc y em p ha sis is d eterm ined b y the resona nt freq uenc y of the
tra nsd uc er (Hood , 1998:96; Ha ll, 1992:123). Thus a c lic k, thoug h a
b roa d b a nd stim ulus, is nonetheless som ew ha t freq uenc y sp ec ific , b a sed
p rim a rily on the freq uenc y resp onse of the ea rp hones (Gorg a , 1999:31;
Hood , 1998:96). A c lic k therefore, w ith its a b rup t onset a nd b rief d ura tion,
is b etter to elic it a sync hronous neura l resp onse, b ut is not very freq uenc y
sp ec ific (Hood , 1998:97). The m a xim um energ y p ea ks a re in the freq uenc y
reg ion b etw een 1000 a nd 4000 Hz (Hood , 1998:96; Ha ll, 1992:107). The
g rea test a g reem ent w ith p ure-tone threshold s is in the 2000 to 4000 Hz
freq uenc y ra ng e. Clic k ABR’ s d o; how ever, p rovid e a g ross estim a te of
hea ring sensitivity a nd a n a ssessm ent of VIIIth nerve a nd a ud itory
b ra instem p a thw a y integ rity – a llow ing the c linic ia n to rule out p ossib le
neurolog ic a l involvem ent (Arnold , 2000:454; Gorg a , 1999:31; Sta p ells &
Oa tes, 1997:248).
Sta p ells & Oa tes (1997:258) c a utions tha t this m a y b e true, on a vera g e
a nd a c ross a la rg e g roup of p a tients w ith hea ring loss. It d oes not tra nsla te
into one b eing a b le to use the c lic k ABR threshold a s a relia b le estim a te of
2000-4000 Hz threshold for ind ivid ua l p a tients. These resea rc hers ha ve
d em onstra ted tha t a ny p a rtic ula r c lic k ABR threshold m a y rep resent a
w id e ra ng e of p ure-tone threshold s, m a king a c c ura te d eterm ina tion of
38
University of Pretoria etd – Stroeb
bel, D (2006
d eg ree of hea ring loss im p ossib le. This seem s esp ec ia lly true in the c a se of
slop ing hea ring losses.
The m a jor exp la na tion for the p rob lem s w ith the c lic k ABR for threshold
estim a tion lies w ith the b roa d -b a nd na ture of c lic ks, a nd the resulting
freq uenc y c ontrib utions to the c lic k-evoked ABR (Sta p ells & Oa tes,
1997:261). A norm a l c lic k ABR threshold d oes not nec essa rily im p ly norm a l
hea ring . It m a y only im p ly a n a rea of norm a l sensitivity b etw een 1000 a nd
4000 Hz (Perez-Ab a lo et a l., 2001:200; Ric ka rd s et a l., 1994:327). When a
hea ring im p a irm ent is restric ted to a p a rtic ula r freq uenc y reg ion, c lic kevoked ABR w ill often m iss the loss or sub sta ntia lly und erestim a te the
d eg ree of the loss (Sta p ells, 2000a :15; Sta p ells, Gra vel & Ma rtin, 1995:361).
This situa tion c a n oc c ur w ith hig h freq uenc y losses, low -freq uenc y losses or
im p a irm ents c onfined to the m id -freq uenc y reg ions (e.g . ‘ c ookie-b ite’
losses) (Sta p ells & Oa tes, 1997:261). As in b eha viora l a ud iom etry in old er
c hild ren, na rrow er b a nd stim uli m ust b e used in ord er to ob ta in ABR
threshold
estim a ted
for sp ec ific
freq uenc y reg ions. In c ontra st to
threshold s to c lic ks, ABR threshold s to b rief tona l stim uli p rovid e m ore
freq uenc y sp ec ific results.
ii.
ABR threshold eva lua tion using b rief tones
The c lic k-evoked ABR m a y b e useful a nd c linic a lly p ra c tic a l for estim a tion
of a ud itory func tion in the 1000 – 4000 Hz reg ion. This m ig ht b e a d eq ua te
for hea ring sc reening , b ut inform a tion on a ud itory sensitivity a c ross the
a ud iom etric ra ng e, esp ec ia lly the sp eec h freq uenc y reg ion (500 – 4000
Hz) is essentia l for a ud iolog ic a l m a na g em ent, suc h a s for the fitting of
hea ring a id s (Gorg a & Neely, 2002:50; Ha ll, 1992:107). The ABR to c lic ks
a lone c a n therefore not p rovid e inform a tion c onc erning hea ring sensitivity
39
University of Pretoria etd – Stroeb
bel, D (2006
for sp ec ific freq uenc ies (Gorg a , 1999:31; Sta p ells, Gra vel & Ma rtin,
1995:361). Sta p ells, Gra vel a nd Ma rtin (1995:361) a lso sta te tha t hea ring
loss restric ted to p a rtic ula r freq uenc y reg ions m a y b e und erestim a ted or
m issed entirely b y the c lic k-ABR threshold . It is therefore not p ossib le to
c ha ra c terize the sha p e of the hea ring loss from c lic k-evoked ABR a lone
even w ith c onsid era tion of the la tenc y/ intensity func tion (Sining er & ConeWesson, 2002:303). An estim a tion of low freq uenc y hea ring sta tus is
esp ec ia lly d esira b le in ord er to estim a te a ud itory func tion a c ross the
a ud iom etric ra ng e (Ha ll, 1992:107). Severa l typ es of stim uli a nd rec ord ing
m ethod s ha ve therefore b een p rop osed to p rovid e inform a tion for
na rrow er freq uenc y reg ions, suc h a s tone b ursts, filtered c lic ks, tone b ursts
a nd c lic ks m ixed w ith va rious typ es of noise, a nd hig h-p a ss m a sking of
c lic ks (Hood , 1998:98). These tec hniq ues a ll ha ve a d va nta g es a nd
lim ita tions. Tone b urst stim uli a re now w id ely a va ila b le on c om m erc ia l ABR
instrum enta tion, a nd a re therefore the m ost c om m only used typ e of
freq uenc y sp ec ific stim uli in ABR testing (Hood , 1998:98; Sta p ells & Oa tes,
1997:258).
In a ttem p ting to a p p roxim a te the b eha viora l p ure tone a ud iog ra m , it ha s
b ec om e fa irly c om m on to inc lud e b rief-d ura tion tona l stim uli a s p a rt of
the test p rotoc ol in ord er to estim a te the a ud iog ra m of young infa nts
(Sining er & Cone-Wesson, 2002:303; Sta p ells, 2002:11; Hood , 1998:96; Ha ll &
Mueller, 1997:360). This typ e of stim ulus is the result of a n a ttem p t to find
the “ b est c om p rom ise” tha t w ould m a xim ize freq uenc y sp ec ific ity a nd
neura l sync hrony (Hood , 1998:98). These stim uli ha ve na rrow er freq uenc y
sp ec tra tha n c lic ks b ut a re sub sta ntia lly b roa d er tha n the p ure tone stim uli
used for c onventiona l a ud iom etry, b ec a use of the b rief rise/ fa ll tim e (Ha ll,
1992:108).
40
University of Pretoria etd – Stroeb
bel, D (2006
Brief tone b ursts ha ve their c onc entra tion of energ y a t a nom ina l
freq uenc y of the tone (p red om ina nt energ y p ea k) a nd sid eb a nd s of
energ y a t low er a nd hig her freq uenc ies (Arnold , 2000:459; Oa tes &
Sta p ells, 1998:62). The sp rea d of stim ulus energ y to freq uenc ies other tha n
the nom ina l freq uenc y is know n a s sp ec tra l sp la tter. Bec a use the
sid eb a nd s a re less intense tha n the p ea k of energ y, the freq uenc y sp rea d
is m ore of a p rob lem a t hig h levels of stim ula tion (Arnold , 2000:459). The
d eg ree of sp ec tra l sp la tter is a lso influenc ed b y severa l p a ra m eters of the
stim uli, inc lud ing rise tim e, d ura tion, tem p ora l sha p ing a nd typ e of
tra nsd uc er used (Oa tes & Sta p ells, 1998:62).
Va rious ra m p ing or envelop e sha p ing tec hniq ues suc h a s Bla c km a n
ra m p ing ha ve b een im p lem ented a s a w a y to im p rove freq uenc y
sp ec ific ity of toneb urst stim uli. At hig h stim ulus intensities, stim ula tion c a n
how ever still sp rea d to a d ja c ent freq uenc y a rea s in p ersons w ith b etter
hea ring , d ue to b a sila r m em b ra ne m ec ha nic s (Arnold , 2000:459). An
a lterna tive w a y to ensure freq uenc y sp ec ific ity is to c om b ine d ifferent
m a sking m ethod s w ith the stim uli (Gorg a , 1999:29). The notc hed noise is
c urrently the m ost c linic a lly used m a sking tec hniq ue (Arnold , 2000:459).
Notc hed noise is sim ila r to w id e b a nd noise, c onta ining energ y a c ross the
freq uenc y sp ec trum , exc ep t w ithin a c erta in na rrow ra ng e of freq uenc ies
(the notc h). The freq uenc y, a t w hic h the notc h oc c urs, c orresp ond s to the
freq uenc y of the tone b urst. Thus, the sid e b a nd s of energ y p resent in the
tone b urst a re m a sked out, restric ting the a rea of stim ula tion to the
nom ina l freq uenc y of the tone b urst. This ensures tha t the ABR is
g enera ted b y neurons sensitive only to the test freq uenc y (Arnold ,
2000:459; Gorg a , 1999:36; Oa tes & Sta p ells, 1998:62).
41
University of Pretoria etd – Stroeb
bel, D (2006
Gorg a (1999:40) c onc lud ed in his resea rc h, tha t a c c ura te estim a tes of
threshold s a re p ossib le for a w id e ra ng e of freq uenc ies, using tone b urst
stim uli. Rea sona b ly a c c ura te estim a tes of the p ure tone b eha viora l
a ud iog ra m from 500 Hz – 4000 Hz c a n b e p rovid ed . Althoug h a rec ent
m eta -a na lysis of the tone b urst ABR litera ture b y Sta p ells (2000b :74) ha s
show n tha t a c ross stud ies, tone-ABR threshold s ha ve b een found to b e
b etw een 10 a nd 20 d BnHL in norm a l hea ring ind ivid ua ls a nd a re g enera lly
w ithin 15 d B of b eha viora l threshold for hea ring im p a ired ind ivid ua ls, som e
stud ies ha ve q uestioned the freq uenc y sp ec ific ity a nd relia b ility of
threshold estim a tion w ith low freq uenc y tone- evoked ABR (Va nd er Werff
et a l., 2002:228; Dim itrijevic et a l., 2002:206). The c red enc e is tha t the ABR
to 500 Hz tona l stim uli is p rim a rily g enera ted from the b a sa l end of the
c oc hlea , esp ec ia lly to hig her-intensity stim uli, a nd thus these threshold s
a re p oor p red ic tors of low -freq uenc y b eha viora l threshold s (Sta p ells &
Oa tes, 1997:261).
Furtherm ore, ABR to b oth c lic k a nd tone b urst stim uli d oes not a p p ea r to
b e a b le to d isting uish severe-to-p rofound hea ring losses in the ra ng e of 85
to 95 d B HL from those in the m ore p rofound ra ng es of 100 to 120 d B HL
(Sta p ells, 2000a :24). The p ossib ility of resid ua l hea ring a t these p rofound
levels c a n therefore not b e investig a ted throug h the use of ABR (Arnold ,
2000:454; Ra nc e, 1998:506). Another lim ita tion of the ABR is the sub jec tive
na ture of interp reting the results (Oa tes & Sta p ells, 1998:67; Ba c hm a nn &
Ha ll, 1998:42). Interp reting ABR w a ves – esp ec ia lly to low freq uenc y tone
b urst stim uli - is p rob lem a tic . Interp reta tion of these results req uires
exp erienc e a nd exp ertise (Sta p ells, 2000a :13). These tec hniq ues m a y a lso
b e tim e c onsum ing (Dim itrijevic et a l., 2002:206).
42
University of Pretoria etd – Stroeb
bel, D (2006
In c a rrying out c linic a l ABR tests on infa nts a nd young c hild ren, c linic ia ns
usua lly p roc eed w ith a n exp ec ta tion tha t the p a tient w ill w a ke up a t a ny
m om ent (Sta p ells, 2002:26). The a im in p ed ia tric a ud iolog y is therefore to
g a in a s m uc h inform a tion a s p ossib le in the tim e a va ila b le. ABR test
p rotoc ols,
therefore
a im
to
g a ther
freq uenc y-sp ec ific
threshold
inform a tion in the shortest p ossib le tim e (Sta p ells, 2000a :26; Arnold ,
2000:460). The d ura tion of a n ABR test session for infa nts a nd young
c hild ren is d eterm ined b y the a m ount of tim e they w ill rem a in a sleep
(Sta p ells, 2002:16). It is therefore essentia l to use a test p rotoc ol tha t is fa st,
effic ient, a nd
one tha t p rovid es the g rea test inc rea se in c linic a l
inform a tion w ith ea c h suc c essive step (Sta p ells, 2002:14). Althoug h the
c lic k ABR p rovid es im p orta nt inform a tion a b out a ud itory func tion, it d oes
not p rovid e suffic ient inform a tion to und ersta nd a ud itory func tion a c ross
the freq uenc y ra ng e (Gorg a , 1999:40). With low freq uenc y inform a tion,
p rovid ed throug h tone b urst ABR, a ud itory func tion c a n b e d efined w ith
g rea ter p rec ision. Ac q uisition of the hig h freq uenc y inform a tion p rovid ed
b y the c lic k ABR or 2000 Hz tone b urst, in c om b ina tion w ith low freq uenc y
inform a tion p rovid ed b y the tone b urst ABR, is nec essa ry to d efine the
c onfig ura tion of the hea ring loss (Arnold , 2000:461). This inform a tion is
essentia l in the d evelop m ent of a ha b ilita tive p rog ra m , inc lud ing the use
of p ersona l a m p lific a tion (Gorg a , 1999:40).
2.3.1.3
The ABR in p ed ia tric hea ring a id fitting s
Without the inform a tion from b eha viora l eva lua tions, it is d iffic ult to a ssess
the
p erform a nc e
a m p lific a tion
of
hea ring
sp ec ific a tions
a id s – even
a re
know n
w hen
(Ga rnha m ,
the
theoretic a l
Cop e,
Durst,
Mc Corm ic k & Ma son, 2000:267). Using elec trop hysiolog ic a l m ea sures to
a ssist in the hea ring a id fitting in infa nts is not a new id ea . Ac c ord ing to
43
University of Pretoria etd – Stroeb
bel, D (2006
Ma honey (1985:351) a ltered a ud itory evoked p otentia ls w ere m ea sured
b y Ra p in a nd Gra zia ni in 1967 und er a m p lific a tion. This p roc ed ure
involved the a d justm ent of the hea ring a id until the la tenc y of w a ve V of
the ABR d ec rea sed to w ithin norm a l lim its (Pic ton et a l., 1998:315).
Som e stud ies ha ve used the ABR threshold m ethod . Ac c ord ing to
Ma honey (1985:357), Mokotoff a nd Kreb s (1976) ob ta ined una id ed a nd
a id ed
ABR threshold s, a ud iom etric
threshold s a nd
elec troa c oustic
m ea sures on c oop era tive a d ult hea ring a id users a nd found fa vora b le
c orrela tions b etw een these p roc ed ures. Other stud ies (Cox & Metz, 1980;
Hec ox, 1983) m entioned in Ma honey (1985:359), sug g ested the use of ABR
w a ve V a b solute la tenc y a nd / or L-I slop e to p red ic t a p p rop ria te hea ring
a id sp ec ific a tions. The b a sic p rem ises w ere tha t norm a l w a ve V la tenc ies
req uire a n inta c t a ud itory system up to the neura l g enera tor, tha t norm a l
L-I slop e sug g ests norm a l d yna m ic loud ness func tion a nd tha t sp eec h
intellig ib ility a nd ABR la tenc y a re c orrela ted . It follow ed tha t if a hea ring
a id c a n b e a d justed in g a in, outp ut, a nd c om p ression c ha ra c teristic s to
g enera te a s norm a l a n ABR a s p ossib le in a p a tholog ic a l ea r, the
p roc ed ure ha d m erit a s a tool for the eva lua tion of a m p lific a tion. Another
ABR Hea ring Aid Eva lua tion m ethod w a s em p loyed b y Kiessling (1982)
(Ma honey, 1985:361). An una id ed ABR p rojec tion system b a sed on norm a l
a nd p a tholog ic a l a m p litud e g row th, to p resc rib e a p p rop ria te hea ring a id
g a in, c om p ression ra tio a nd c om p ression onset w a s used .
More rec ently Ga rnha m et a l. (2000:267) used the ABR a s a n ob jec tive
m ea sure to verify the a id ed hea ring threshold s in a g roup of c hild ren.
Ob jec tive d a ta w ere c ollec ted from the ABR a nd b eha viora l threshold s
w ere m ea sured b y use of a g e a p p rop ria te tests. When c om p a ring the
una id ed ABR c lic k threshold s to b eha viora l threshold s, the ABR threshold
44
University of Pretoria etd – Stroeb
bel, D (2006
w a s on a vera g e 9 d B low er. Using the sa m e c om p a rison for a id ed
resp onses, a d ifferenc e of <5 d B w a s ob served . This g roup of resea rc hers
c onc lud ed tha t a id ed ABR threshold s a re va lua b le in the m a na g em ent of
young c hild ren. How ever, w hen p erform ing these m ea surem ents, it is
essentia l to b e a w a re of the lim ita tions of the hea ring a id a nd the stim ulus.
Althoug h Ma honey (1985:356) illustra ted the fea sib ility of using ABR for
func tiona l g a in m ea surem ents, the w id esp rea d use of this tec hniq ue d id
not oc c ur. This p roc ed ure is tec hnic a lly c ha lleng ing d ue to four m a in
c onc erns. First, the c lic k stim ulus is very b rief a nd c a n b e sig nific a ntly
d istorted b oth in the sound field sp ea ker a nd in the hea ring a id . The
resulta nt stim ulus a rtifa c ts m a y ob sc ure interp reta tion of the resp onses
(Ga rnha m et a l., 2000:268). Sec ond , the m ost sig nific a nt lim ita tion
c onc erning this tec hniq ue stem s from the fa c t tha t hea ring a id s rea c t
d ifferently to ra p id ly c ha ng ing stim uli tha n to m ore c ontinuous stim uli
w hic h lea d s to d istortion of the stim ulus (Ma honey, 1985:368). Third , the
c lic k ABR is m a inly rela ted to hig h freq uenc y g a in a nd c orrela tion
b etw een w a ve V la tenc y a nd loud ness is low , p a rtic ula rly w hen there is a
slop ing hea ring loss (Pic ton et a l., 1998:316). Fourth, the b rief stim uli tha t
a re op tim a l for ABR rec ord ing s m a y not a c tiva te the hea ring instrum ent’ s
c om p ression c irc uitry in the sa m e w a y a s long er-d ura tion sp eec h sound s
(Brow n, Klein & Snyd ee, 1999:196) a nd m a y b e trea ted a s ‘ noise’ b y
hea ring instrum ents w ith sp eec h d etec tion a lg orithm s (Alc a ntra , Moore,
Kuhnel & La uner, 2003:40). For these rea sons a ttem p ts to use the ABR to
eva lua te hea ring instrum ents ha ve la rg ely b een a b a nd oned (Purd y,
Ka tsc h, Dillon, Storey, Sha rm a & Ag ung , 2005:116).
45
University of Pretoria etd – Stroeb
bel, D (2006
2.3.1.4
Sum m a ry of the ABR a p p lic a tion in p ed ia tric a ud iolog y
As a c onc lusion to this c ritic a l eva lua tion of the ABR, Ta b le 2.2 sum m a rizes
the a d va nta g es a nd lim ita tions of the ABR.
Ta ble 2.2
Adva nta ges a nd limita tions of the ABR
Adva nta ges
Limita tions
Dia gnosis
•
A noninva sive, sa fe a p p roa c h
•
Sta b le resp onse – resista nt to sta te of
•
c onsc iousness
•
of hig h freq uenc ies
•
Cha ra c teristic s sim ila r b etw een p eop le
– ea sy to id entify resp onse – even in
•
Rec ord a b le
No
d istinc tion b etw een severe
a nd
p rofound losses
•
infa nts
•
Clic k ABR p rovid es g enera l a ssessm ent
Stim uli c onta in energ y over ra ng e of
freq uenc ies a nd m a y evoke a resp onse
–
c lose
to
b eha viora l
a t a ny of these
threshold s
•
Tim e-c onsum ing
Tone b urst stim uli c a n b e used to
•
Sub jec tive interp reta tion of results
p rovid e m ore freq uenc y-sp ec ific
inform a tion
Va lida tion Process
•
Potentia l
to
inform a tion
c onc erning
func tiona l b enefit
p rovid e
ob jec tive
hea ring
•
a id
Clic k stim uli is very b rief a nd d istorts in
sp ea ker a nd / or hea ring a id
•
Hea ring a id s rea c t d ifferently to ra p id ly
c ha ng ing stim uli
•
Clic k ABR is m a inly rela ted to hig h
freq uenc y
g a in
a nd
c orrela tion
b etw een w a ve V la tenc y a nd loud ness
is low
•
Com p ression
c irc uitry
a c tiva ted
d ifferently from sp eec h stim uli
46
University of Pretoria etd – Stroeb
bel, D (2006
One tec hniq ue tha t ha s d em onstra ted
p rom ise in a d d ressing the
lim ita tions of the ABR in va lid a ting hea ring a id fitting s in infa nts is the
Aud itory Stea d y Sta te Resp onse (ASSR). This p roc ed ure a lso d em onstra tes
p rom ise in a d d ressing som e of the ABR lim ita tions in a ssessing hea ring
a b ilities in the d iffic ult-to-test p op ula tion (Sw a nep oel, Hug o & Rood e,
2004:531).
2.3.2
Perspectives on the Auditory Stea dy Sta te Response
In the p a st tw o d ec a d es, a n evoked p otentia l p a rtic ula rly suited to
freq uenc y-sp ec ific m ea surem ent, c om m only referred to a s the Aud itory
Stea d y Sta te Resp onse (ASSR) or Stea d y Sta te Evoked Potentia l (SSEP), ha s
b een und er c lose sc rutiny for c linic a l a p p lic a tion (Perez-Ab a lo et a l.,
2001:200).
2.3.2.1
Definition
a nd
Develop m ent
of
Aud itory
Stea d y
Sta te
Resp onse
The ASSR a re p eriod ic sc a lp p otentia ls tha t a rise in resp onse to reg ula rly
va rying stim uli suc h a s sinusoid a l a m p litud e a nd / or freq uenc y m od ula ted
tone (Ra nc e, Dow ell, Ric ka rd s, Beer & Cla rk, 1998:49). It yield s a w a veform
c losely follow ing the tim e c ourse of the stim ulus m od ula tion a nd a
resp onse sp ec ific to the freq uenc y of the c a rrier. By va rying the intensity of
the elic iting stim ulus a threshold resp onse c a n b e m ea sured (Jerg er, 1998:
ed itoria l).
The p rinc ip le und erlying the ASSR is b a sed on the follow ing c oc hlea r
m ec ha nic s
as
outlined
by
Lins,
Pic ton,
Bouc her,
Durieux-Sm ith,
Cha m p a g ne, Mora n, Perez-Ab a lo, Ma rtin a nd Sa vio (1996:84) a nd
47
University of Pretoria etd – Stroeb
bel, D (2006
illustra ted b y Fig ure 2.1: Sound w a ves p rod uc e a n effec t of p ola riza tion
a nd d ep ola riza tion of the inner ha ir c ells. Only the d ep ola riza tion of inner
ha ir c ells c a uses a ud itory nerve fib ers to tra nsm it a c tion p otentia ls. The
elec tric a l a c tion p otentia l outp ut of the c oc hlea therefore c onta ins a
rec tified version of the a c oustic stim uli. This rec tific a tion c a uses the outp ut
of the c oc hlea to ha ve a sp ec tra l c om p onent a t the freq uenc y a t w hic h
the c a rrier w a s m od ula ted . This c om p onent, w hic h is not p resent in the
sp ec trum of the stim uli, c a n b e used to a ssess the resp onse of the c oc hlea
to the freq uenc y of the c a rrier tone.
Figure 2.1
Principles underlying the ASSR (fro m Pic to n, 2005: c o nfe re nc e p re se nta tio n)
The stim uli used to evoke the ASSR a re a m od ula ted tone in the sta nd a rd
a ud iom etric ra ng e (Cone-Wesson, 2003:267). The tone c a n b e a m p litud e
(AM) or freq uenc y (FM) m od ula ted ; or b oth a m p litud e a nd freq uenc y
m od ula ted . The
stim uli c onsists of a
c a rrier freq uenc y (CF) (test
freq uenc y), m od ula ted over tim e in the a m p litud e d om a in a t a freq uenc y
of m od ula tion (MF) (Perez-Ab a lo, et a l., 2001:201). Fig ure 2.2 d em onstra tes
the m od ula tion of a p ure tone.
48
University of Pretoria etd – Stroeb
bel, D (2006
Figure 2.2
A single tone a nd a modula ted tone
(fro m Sw a ne p o e l, Sc hm ulia n & Hug o ,
2002:51).
Ac c ord ing to Dim itrijevic et a l. (2002:206), ASSR’ s w ere first sug g ested a s
a n ob jec tive m ea ns to a ssess hea ring b y Ga la m b os a nd c ollea g ues in
1981. These resea rc hers used m od ula tion freq uenc ies b etw een 35 a nd 55
Hz to a ssess hea ring threshold . They sub seq uently show ed tha t the 40-Hz
stea d y-sta te resp onse w a s ea sy to id entify a t intensities just a b ove
b eha viora l threshold s. How ever, som e lim ita tions for ob jec tive a ud iom etry
a re p resent w ith the 40-Hz stea d y-sta te resp onse suc h a s: (1) The resp onse
is unrelia b le in estim a ting threshold s in infa nts a nd young c hild ren
(Herd m a n & Sta p ells, 2001:41); (2) The resp onse d im inishes w hen sub jec ts
a re a sleep or sed a ted (Dim itrijevic , 2002:206 & Ra nc e, 1995:500); (3)
Resp onse a m p litud e d im inishes w hen severa l stim uli a re p resented
sim ulta neously (John, 1998:59).
Rec ent w ork ha s therefore foc used on a lterna tive ra tes of stim ula tion for
a ud iom etric p urp oses. Som e resea rc hers ha ve found tha t resp onses a re
rec ord ed c onsistently – d uring sleep , a nd a t low sound p ressure levels - in
a ll sub jec ts (inc lud ing infa nts) w hen a m od ula tion ra te of a b ove 70 Hz is
used (Sta p ells & Herd m a nn 2001:41; Lins et a l., 1996:82; Ra nc e et a l.,
1995:500; Ric ka rd s et a l., 1994:327). Therefore the ASSR elic ited b y c a rrier
freq uenc ies w ith hig her m od ula tion ra tes ha ve b een p rop osed a s a n
a lterna tive to ob jec tive freq uenc y sp ec ific a ud iom etry (Perez-Ab a lo et a l.,
2001:200). The c a rrier sine w a ve is the freq uenc y b eing tested a nd c a n b e
49
University of Pretoria etd – Stroeb
bel, D (2006
p resented a t a ny low or hig h freq uenc y tone a s in p ure tone testing
(Sw a nep oel, Sc hm ulia n & Hug o, 2002:51). These m od ula ted tones a re a s
freq uenc y sp ec ific a s p ure tones b ec a use sp ec tra l energ y is c onta ined
only a t the freq uenc y of the c a rrier tone a nd the freq uenc y of m od ula tion
(Cone-Wesson & Sining er, 2002:311; Hood , 1998:117).
Stud ies investig a ting the neura l sourc es of the ASSR ind ic a te they orig ina te
p rim a rily from b ra instem struc tures (Sta p ells, 2005:44; Kuw a d a et a l.,
2002:202) b ut this d ep end s on the ra te of m od ula tion a nd sub jec t sta te
(Cone-Wesson, 2003:267). Althoug h not yet c onfirm ed , it is p ossib le tha t
the ASSR a re ABR w a ve V to ra p id ly p resented stim uli (Sta p ells, 2005:44).
The ASSR is g enera ted w hen the c a rrier freq uenc y (test freq uenc y) is
p resented a t a ra te (m od ula tion freq uenc y) tha t is suffic ient to c a use a n
overla p p ing of tra nsient resp onses, thus b eing a susta ined resp onse
(Sw a nep oel, Sc hm ulia n & Hug o, 2002:51). A c a rrier freq uenc y stim ulus
trig g ers a sp ec ific reg ion of the b a sila r m em b ra ne, a c tiva ting ha ir c ells in
the c oc hlea in the reg ion tha t c orresp ond s p rim a rily to the tone
freq uenc y. As the resulting neura l a c tivity tra vels a long the a ud itory
p a thw a y, EEG a c tivity ‘ sync hronizes w ith’ or ‘ follow s’ the a m p litud e
m od ula tion freq uenc y
(Lins et a l., 1996:85). This m ea ns tha t the c a rrier
freq uenc y stim ula tes the c oc hlea w ith p oc kets of energ y a t the ra te of
the m od ula tion freq uenc y (Sw a nep oel, Sc hm ulia n & Hug o, 2002:51). The
energ y in the resulta nt resp onse is a t the freq uenc y of m od ula tion a nd its
ha rm onic s, a llow ing a na lysis of the resp onse in the freq uenc y d om a in
(Herd m a n & Sta p ells, 2001:41).
The ASSR is rec ord ed in a tim e-d om a in a nd m ust b e c onverted to a
freq uenc y-d om a in b y a Fa st Fourier Tra nsform (FFT) for a na lysis (Lins,
1996:85). In the freq uenc y d om a in, the resp onse to the c a rrier freq uenc y
50
University of Pretoria etd – Stroeb
bel, D (2006
c a n b e a ssessed b y the a m p litud e a nd p ha se of the FFT c om p onent
c orresp ond ing to the freq uenc y of m od ula tion of the c a rrier (Sw a nep oel,
Sc hm ulia n & Hug o, 2002:51). Com b ining resp onses w hilst m a inta ining b oth
p ha se a nd a m p litud e inform a tion ob ta in a n a vera g e resp onse (PerezAb a lo et a l., 2001:201). Fig ure 2.3 illustra tes this p roc ed ure.
Figure 2.3
i.
Recording the ASSR (fro m Sw a ne p o e l, Sc hm ulia n & Hug o , 2002:52).
Sing le stim uli vs. m ultip le stim uli ASSR
The ASSR c a n b e evoked using a sing le freq uenc y stim ulus (Ra nc e et a l.,
1995:501) or the ASSR c a n b e evoked using m ultip le-freq uenc y stim uli
p resented sim ulta neously (Lins et a l., 1996:81). With the la tter tec hniq ue, it
is p ossib le to p resent m ultip le a m p litud e-m od ula ted CF’ s sim ulta neously
a nd p erform a sep a ra te a na lysis for ea c h MF used in the c om p lex stim ulus
51
University of Pretoria etd – Stroeb
bel, D (2006
(Sining er & Cone-Wesson, 2002:313). Lins a nd Pic ton (1995:420) show ed
tha t it is p ossib le to p resent up to four CF’ s in ea rs, using 500, 1000, 2000
a nd 4000 Hz w ith eig ht d ifferent MF’ s. The MF’ s va ry for ea c h ea r a nd CF.
When sup ra threshold level (60 d B SPL) stim uli w ere used , there w ere no
d ifferenc e in resp onse a m p litud e for the sing le-tone-a lone c ond ition, four
stim uli c om b ined in one ea r, or four stim uli c om b ined in tw o ea rs (ConeWesson, 2003:271; Sining er & Cone-Wesson, 2002:313). On a vera g e, a n 18
d B d ifferenc e b etw een b eha viora l threshold s for the sing le tones a nd the
ASSR threshold s w a s found w hen tw o CF’ s w ere p resented sim ulta neously.
The m a jor a d va nta g e of this tec hniq ue is tha t b y sim ulta neously
p resenting m ultip le stim uli, (e.g . four stim uli in ea c h ea r for a tota l of eig ht),
m ultip le resp onses c a n b e rec ord ed d uring the tim e norm a lly req uired to
rec ord one (John et a l., 2002:247; Dim itrijevic et a l., 2000:207). Fig ure 2.4
illustra tes the m ulti freq uenc y ASSR.
Figure 2.4
Multiple ASSR (Fro m Sta p e lls, 2004: c o nfe re nc e p re se nta tio n).
52
University of Pretoria etd – Stroeb
bel, D (2006
2.3.2.2
Threshold d eterm ina tion
The p resenc e or a b senc e of a resp onse is d eterm ined a utom a tic a lly a nd
ob jec tively, using d etec tion p rotoc ols tha t c om p a re the resp onse to the
b a c kg round EEG a c tivity (Pic ton, 2002:65; Ra nc e, 1995:501). Autom a tic
resp onse d etec tion p rotoc ols rely on c om p uter a lg orithm s w hic h a re
a p p lied to the rec ord ed EEG sig na l to a na lyze the m a g nitud e a nd p ha se
of EEG a c tivity c orresp ond ing to the m od ula tion freq uenc y of the tone
a nd to d eterm ine the p resenc e or a b senc e of a n ASSR (Cone-Wesson &
Sining er, 2002:317).
Sa m p les of EEG a c tivity a re rec ord ed a nd a na lyzed a s the c ontinuous
m od ula ted tone is p resented . In ea c h EEG sa m p le, the m a g nitud e a nd
p ha se of the EEG a c tivity c orresp ond ing to the tone m od ula tion
freq uenc y is q ua ntified (Cone-Wesson & Sining er, 2002:317). The p ea ks in
the resulting sp ec trum , a nd the a m p litud e a nd p ha se of the sp ec tra l
p ea k, c a n b e m ea sured for p ha se c oherenc e (PC). The p ha se of the
m a jor p ea k c a n b e p lotted on p ola r c oord ina tes. The sine a nd c osine of
the a ng les form ed b y ea c h p ha se vec tor a re c a lc ula ted . PC va lues va ry
from 0.0 to 1.0 (Cone-Wesson & Sining er, 2002:317). When the sa m p le
p ha ses a re in p ha se w ith one a nother, there is a hig h c oherenc e, a nd the
va lue w ill b e c loser to 1.0. When the sa m p le p ha ses a re ra nd om , there is
low c oherenc e a nd va lues a re c loser to 0. Usua lly w hen a sig nific a nt level
of p < 0, 05 is ob ta ined , the nil hyp othesis is rejec ted , the sa m p les c a n b e
c onsid ered p ha se loc ked or c oherent, a nd a n evoked resp onse is
d eterm ined to b e p resent. Fig ure 2.5 show s a p ola r p lot of p ha se
c oherenc e.
53
University of Pretoria etd – Stroeb
bel, D (2006
Figure 2.5
Pola r Plot to Pha se Coherence (fro m Sining e r & Co ne -We sso n, 2002)
By rec ord ing resp onses a t d esc end ing intensities, a threshold or m inim um
resp onse level c a n b e ob ta ined a t the low est intensity elic iting a resp onse
(Sw a nep oel et a l, 2002:51).
2.3.2.3
Current Clinic a l Ap p lic a tion of the ASSR in Infa nts
The m a jor g oa l of evoked p otentia l a ud iom etry in infa nts is to p red ic t or to
estim a te a n infa nt’ s b eha viora l a ud iog ra m from evoked p otentia l d a ta –
w ithout a ny resp onse from the p a tient or sub jec tive interp reta tions of the
results b y a c linic ia n (Dim itrijev et a l., 2002:206; Gold stein & Ald ric h,
1999:109). Furtherm ore it is im p orta nt to seek a p roc ed ure tha t m a y g ive
the m ost inform a tion w ith reg a rd to freq uenc y ra ng e, sig na l m a g nitud e
ra ng e, resp onse relia b ility, c lea r c riteria for esta b lishing threshold a nd
va lid ity in term s of the p a tient’ s a c tua l a ud itory sensitivity. In the p a st tw o
d ec a d es, ASSR tec hniq ues ha ve b ec om e a va ila b le a s a n op tion for
ob jec tive hea ring testing (Ra nc e et a l., 1998:499). Severa l resea rc hers
54
University of Pretoria etd – Stroeb
bel, D (2006
found the ASSR to b e a relia b le m ethod to ob ta in freq uenc y sp ec ific
estim a tes of b eha viora l p ure tone threshold s in a d ults a nd old er c hild ren
(Dim itrijevic et a l., 2002:205; Herd m a n & Sta p ells, 2001:41; Lins et a l.;
1996:81 a nd Ra nc e et a l., 1995:499). Ric ka rd s et a l. (1994:327) d id resea rc h
on the a p p lic a tion of ASSR on w ell b a b ies a nd other resea rc hers d id
retrosp ec tive stud ies on the a p p lic a tion of ASSR on infa nts (Va nd er Werff
et a l., 2002:227; Cone-Wesson et a l., 2002:173) – c om p a ring the ABR results
w ith ASSR results. The c linic a l a p p lic a tion of the ASSR w ill now b e d isc ussed
– looking a t three a sp ec ts, na m ely detection, dia gnosis a nd hea ring a id
fitting in infa nts.
i.
Detec tion
‘ It is a lm ost a xiom a tic in the field of a ud iolog y tha t ea rly d etec tion a nd
ea rly intervention w ill yield a b etter func tioning hea ring im p a ired c hild ’
(Luterm a n,
1999:35).
p roc ed ures for
Over the
sc reening
p a st
new b orns,
thirty
yea rs, severa l d ifferent
inc lud ing
c a rd ia c
resp onse,
resp ira tion a ud iom etry, or a ltera tion of suc king a nd sta rtle resp onses ha ve
b een used , investig a ted a nd found w a nting (Luterm a n, 1999:37). Severa l
m ethod s of im p lem enta tion of the hig h risk reg ister a p p roa c h ha ve b een
used in the USA. It seem s to id entify a b out ha lf of new b orns w ith hea ring
loss (Northern & Ha yes, 1997:21). Rec ently the ABR ha s b een a utom a ted
a nd the EOAE ha s b een d evelop ed . Both these p roc ed ures c a n b e
ra p id ly a d m inistered , thus m a king
universa l sc reening
for hea ring
im p a irm ent fea sib le (Luterm a n, 1999:39).
ASSR’ s m a y ha ve a n a d va nta g e over the ABR a nd EOAE’ s in new b orn
sc reening (Sining er & Cone-Wesson, 2002:318). EOAE’ s a re thoug ht to
ha ve a n a d va nta g e over the c lic k-evoked ABR, b ec a use it is m ore
55
University of Pretoria etd – Stroeb
bel, D (2006
“ freq uenc y-sp ec ific ” . EOAE’ s a p p ea r to ind ic a te c oc hlea r integ rity for a t
lea st the 1000 – 4000 Hz hea ring ra ng e (Sining er & Cone-Wesson,
2002:318), how ever, EOAE’ s d o not test neura l func tion a nd c a nnot
p red ic t hea ring threshold (Ha ll, 2000:26). The AABR on the other ha nd , only
uses c lic k stim uli, lim iting estim a tion of hea ring loss in d ifferent freq uenc y
ra ng es (Sta p ells, Gra vel & Ma rtin, 1995:361).
ASSR tests op tim ized for sc reening m a y overc om e b oth the freq uenc y
lim ita tions of c lic k AABR a nd the site-of–lesion lim ita tions of EOAE. Sinc e
ASSR tests use tona l stim uli, the evoked p otentia l c a n b e effic iently
d etec ted w ith w ell-d oc um ented a lg orithm s, a nd a c c ura te threshold
estim a tes c a n b e ob ta ined (Sining er & Cone-Wesson, 2002:318; Ric ka rd s,
1994:327). The Ric ka rd s g roup rec ord ed ASSR’ s from 337 norm a l full-term
sleep ing new b orns to c om b ined a m p litud e a nd freq uenc y m od ula ted
tones. Resp onses w ere found m ost ea sily a nd c onsistently, rec ord ed a t
c a rrier freq uenc ies of 500 Hz, 1500 Hz a nd 4000 Hz w ith m od ula tion
freq uenc ies b etw een 60 Hz a nd 100 Hz. In this m od ula tion freq uenc y
ra ng e, the resp onse la tenc ies w ere b etw een 11 m s a nd 15 m s a nd the
m ea n resp onse threshold for the three c a rrier freq uenc ies w ere found to
b e 41.36 d B HL, 24.41 d B HL a nd 34.51 d b HL resp ec tively. These
resea rc hers sug g ested tha t the ASSR m a y b e useful for freq uenc y-sp ec ific
a utom a ted sc reening in new b orns w hen m od ula tion ra tes exc eed ed 60
Hz. Cone-Wesson et a l. (2002:276) used esta b lished tools (AABR a nd
EOAE’ s) a s the g old sta nd a rd a g a inst w hic h a n ASSR sc reening p rotoc ol
w a s c om p a red . It w a s found tha t a three-freq uenc y sc reening test (1000,
2000 a nd 4000 Hz) p rotoc ol c ould b e c om p leted w ithin tw o m inutes for
ea c h ea r. Althoug h the ASSR w ould seem to b e a n id ea l sc reening tool,
a p p rop ria te sc reening p erform a nc e d a ta (i.e., sensitivity a nd sp ec ific ity)
56
University of Pretoria etd – Stroeb
bel, D (2006
in
a p p rop ria te
c linic a l
sa m p les w ill
be
need ed
b efore
p ossib le
im p lem enta tion (Sta p ells, 2005:56).
Aud iog ra m estim a tion is c lea rly the m ost im p orta nt c linic a l a p p lic a tion of
the ASSR a t this tim e. The follow ing sec tion w ill foc us on the d ia g nosis of
hea ring loss in infa nts
ii.
Dia g nosis
Va rious exp erim ents ha ve d em onstra ted tha t the ASSR c a n b e relia b ly
rec ord ed a t intensities nea r b eha viora l threshold s in sed a ted a nd sleep ing
a d ults (Dim itrijevic et a l., 2002:205; Herd m a n & Sta p ells, 2001:41; Lins et a l.,
1996:81). Lins et a l. (1996:81) used a test tim e of 3.2 to 12.8 m inutes for
ea c h rec ord ing a nd found evoked resp onse threshold s tha t w ere
a p p roxim a tely 11 to 14 d B a b ove b eha viora l threshold s in the freq uenc y
ra ng e of 500 – 4000 Hz. ASSR threshold s a p p ea r to a p p roa c h b eha viora l
threshold s m ore c losely w ith hea ring losses of a p p roxim a tely 60 d B HL or
hig her. Ra nc e et a l. (1995:500) rec ord ed ASSR threshold s w ithin 11 to 20 d B
of the b eha viora l threshold s in a ra ng e 1 to 4 kHz a nd a p p roxim a tely 11 to
40 d B a t 500 Hz in sub jec ts w ith a hea ring loss of 60 d B or m ore. In sub jec ts
w ith hea ring losses b elow 60 d B HL, ASSR threshold s w ere found over a
w id er ra ng e.
Severa l investig a tors ob ta ined ASSR threshold s from infa nts w ho w ere not
a t risk for hea ring loss. There w ere som e d ifferenc es in a g e of the infa nts
b etw een the stud ies – Ric ka rd s et a l. (1994:327) tested infa nts young er
tha n 7 d a ys. This g roup of investig a tors found ASSR threshold s from 32 d B
SPL (1500 Hz) to 53 d B SPL (500 Hz). Lins et a l. (1996:81) tested the a g e
ra ng e of 1 to 10 m onths a nd found threshold s from 26 d B SPL (2000 Hz) to
57
University of Pretoria etd – Stroeb
bel, D (2006
58 d B SPL (500 Hz). Cone-Wesson et a l. (2002:260) tested a t a m ea n a g e of
11.5 m onths a nd ha d sim ila r results: threshold s va ried from 29 d B SPL (2000
Hz) to 45 d B SPL (500 Hz).
The ASSR evoked resp onses offers d efinite
a d va nta g es over tec hniq ues tha t req uire short d ura tion stim uli (Ra nc e et
a l., 1998:49). The ASSR is evoked b y freq uenc y-sp ec ific stim uli (ConeWesson, 2003:267 & Hood , 1998:117). This is b ec a use the stea d y sta te
stim uli a re c ontinuous tones tha t d o not suffer the sp ec tra l d istortion
p rob lem s a ssoc ia ted w ith b rief tone b ursts a nd c lic ks (Ra nc e et a l.,
1998:49). This sp ec ific ity a llow s testing a c ross the a ud iom etric ra ng e a nd
the g enera tion of evoked p otentia l a ud iog ra m s, w hic h in sub jec ts w ith
hea ring loss, c a n reflec t the c onfig ura tion of the loss a c c ura tely (Ra nc e et
a l., 1995:500).
Ra nc e et a l. (2005:297) a nd Ra nc e et a l. (1998:506) d em onstra ted the
a d va nta g es of using the ASSR to d eterm ine resid ua l hea ring threshold s for
those infa nts a nd c hild ren from w hom a c lic k ABR c ould not b e evoked
(a t 100 d BnHL). In the 1998 stud y, c om p leted b y Ra nc e et a l., ASSR’ s w ere
ob ta ined using CF’ s of 250-4000 Hz w ith MF’ s of 90 Hz. The a vera g e
d isc rep a nc y b etw een ASSR a nd b eha viora l threshold ra ng ed only 3 to 6
d B w ith la rg er d isc rep a nc ies found a t 250 a nd 500 Hz. ASSR threshold s
w ere w ithin 20 d B of p ure tone threshold s for 99% of the c om p a risons a nd
10 d B or less for 82% of the c om p a risons. Ra nc e et a l. (2005:297)
d em onstra ted results c onsistent w ith the p revious stud y. Overa ll, the
find ing s show ed a strong c orrela tion b etw een ASSR threshold a nd
b eha viora l hea ring threshold levels. Pea rson r c orrela tion c oeffic ient
va lues ra ng ed from 0.96 to 0.98 a c ross the test freq uenc ies in sub jec ts w ith
hea ring loss. These find ing s d em onstra ted the effic a c y of ASSR’ s for
estim a ting the a ud iog ra m in infa nts a nd c hild ren w ho c a n b enefit from
a m p lific a tion of their resid ua l hea ring (Sining er & Cone-Wesson, 2002:316).
58
University of Pretoria etd – Stroeb
bel, D (2006
The d eterm ina tion of a ir-c ond uc tion (AC) a nd b one-c ond uc tion (BC)
threshold s is a m a insta y of c linic a l a ud iolog y (Cone-Wesson, Ric ka rd s,
Poulis, Pa rker, Ta n & Polla rd , 2002:271). It is therefore im p orta nt to
d eterm ine the c ond uc tive c om p onent to a n infa nt’ s hea ring loss (Jeng ,
Brow n, Johnson & Va nd er Werff, 2004:68), p a rtic ula r in infa nts a nd young
c hild ren, w ho ha ve a hig h inc id enc e of m id d le ea r d isord ers, c a using
c ond uc tive hea ring loss (Cone-Wesson et a l., 2002:271). The ASSR c a n b e
p resented using b oth AC a nd BC tra nsd uc ers (Pic ton & John, 2004:542).
Jeng et a l. (2004;68) a nd Cone-Wesson et a l. (2002:271) ha ve show n a
strong c orrela tion b etw een tha t of the ASSR b one c ond uc tion g a p a nd
a ud iom etric estim a tes of a ir b one g a p . Using the ASSR in this m a nner
p rovid es a d d itiona l inform a tion a b out the na ture of the hea ring loss.
A further a d va nta g e of the ASSR, im p orta nt for the a p p lic a tion in infa nts,
a s c ited b y Ra nc e (1995:506), is the sp eed in w hic h a resp onse c a n b e
d etec ted . Resp onses c ould b e d etec ted w ithin 20 – 90s a fter onset of the
stim ulus. Va n d er Reijd en, Mens & Snik (2005:300) c onc lud ed in their
sum m a ry of test tim e in the infa nt p op ula tion tha t it a p p roxim a tely took
b etw een 3.2 to 12.8 m inutes p er ea r, if four c a rrier freq uenc ies w ere
tested . This fa st test tim e red uc es the need to ha ve the infa nt a sleep or
und er sed a tion for long p eriod s of tim e. As a result, the c linic ia n is m ore
likely to ob ta in a ll the inform a tion tha t is req uired b efore the sub jec t
a w a kens, a nd w ithin one testing p eriod (John et a l., 2004:551; Ra nc e et
a l., 1995:506).
A d isting uishing a nd a d va nta g eous fea ture of the ASSR tec hniq ue is tha t
ob jec tive d etec tion a lg orithm s ra ther tha n visua l d etec tion m ethod s a re
a lw a ys used to d eterm ine p resenc e or a b senc e of a resp onse (Sining er &
Cone-Wesson, 2002:316; Lins et a l., 1996:82). This is a p a rtic ula r a d va nta g e
59
University of Pretoria etd – Stroeb
bel, D (2006
for tec hniq ues c la im ing to b e “ ob jec tive”
in na ture a s a c c ura te
inform a tion w ith reg a rd s to the c onfig ura tion of the hea ring loss is
nec essa ry to d evelop a ha b ilita tion p rog ra m , suc h a s the use of
a m p lific a tion.
iii.
The ASSR in p ed ia tric hea ring a id fitting s
Another a p p lic a tion of the ASSR is w hen reha b ilita tion ha s sta rted a nd
hea ring a id s ha ve b een fitted a c c ord ing to the elec trop hysiolog ic ta rg ets.
Pic ton (1998:315) a nd Gloc kner in Cone-Wesson (2003:272) show ed tha t
ASSR’ s c ould b e rec ord ed w hen stim uli w ere p resented sim ulta neously
throug h a sound -field sp ea ker a nd a m p lified using a hea ring a id . Pic ton et
a l. (1998:315) rec ord ed resp onses a t c a rrier freq uenc ies of 500, 1000, 2000
a nd 4000 Hz in a g roup of 35 hea ring -im p a ired c hild ren using hea ring a id s.
The p hysiolog ic resp onses w ere rec ord ed a t intensities c lose to the
b eha viora l threshold s for sound s in the a id ed c ond ition, w ith a vera g e
d ifferenc es b etw een the p hysiolog ic
a nd
b eha viora l threshold s of
resp ec tively 17, 13, 13, a nd 16 d B for c a rrier freq uenc ies 500, 1000, 2000
a nd 4000 Hz. While there w ere d isc rep a nc ies b etw een b eha viora l (a id ed )
threshold a nd ASSR (a id ed ) threshold , it a p p ea red to b e no g rea ter tha n
those found w hen stim uli w ere tra nsd uc ed b y ea rp hones (Sining er &
Cone-Wesson, 2002:319). Their find ing s sug g est tha t it w ould b e p ossib le to
m ea sure func tiona l g a in of hea ring a id s on the b a sis of ASSR threshold
p red ic tions. The Pic ton g roup (1998) used a m ultip le-sim ulta neous stim ulus
tec hniq ue a nd for som e sub jec ts, resp onses w ere only found a t hig h
sup ra threshold levels or w ere a b sent. Retest w ith sing le AM tones in these
c a ses, show ed b etter c orresp ond enc e b etw een p ure tone a nd ASSR
threshold . This tec hniq ue show s g rea t p rom ise a s a w a y to a ssess a id ed
60
University of Pretoria etd – Stroeb
bel, D (2006
threshold s ob jec tively in sub jec ts w ho c a nnot relia b ly resp ond
to
b eha viora l testing .
Althoug h hea ring loss is c om m only a ssessed using p ure tone threshold s,
the m ost d eb ilita ting a sp ec t of a hea ring loss is d iffic ulty in sp eec h
p erc ep tion (Dim itrijevic , John & Pic ton, 2004:68). A nec essa ry first step in
the p erc ep tion of a w ord is to d isc rim ina te c ha ng es in the freq uenc y a nd
a m p litud e of a sound . The a b ility of the b ra in to d etec t c ha ng es in
freq uenc y a nd a m p litud e m a y b e a ssessed b y rec ord ing ASSR’ s to
m od ula tions in the freq uenc y a nd a m p litud e of sup ra -threshold tones
(Dim itrijevic , John & Pic ton, 2004:68). In this p a rtic ula r stud y ind ep end ent
a m p litud e a nd freq uenc y (IAFM) m od ula tion of tones stim ulus p a ra m eters
w ere a d justed to resem b le the a c oustic p rop erties of everyd a y sp eec h to
d eterm ine how w ell resp onses to these sp eec h-m od ula ted stim uli w ere
rela ted to w ord rec og nition sc ores (WRS). The c orrela tions b etw een WRS
a nd the num b er of IAFM resp onses rec og nized a s sig nific a ntly d ifferent
from the b a c kg round w ere b etw een 0.70 a nd 0.81 for the 40 Hz stim uli,
b etw een 0.73 a nd 0.82 for the 80 Hz stim uli, a nd b etw een 0.76 a nd 0.85 for
the c om b ined a ssessm ent of 40 Hz a nd 80 Hz resp onses. They c onc lud ed
their resea rc h, sta ting tha t IAFM resp onses a re sig nific a ntly c orrela ted w ith
WRS a nd tha t it m a y p rovid e a n ob jec tive tool for exa m ining the b ra in’ s
a b ility to p roc ess the a ud itory inform a tion need ed to p erc eive sp eec h.
2.3.2.4
Critic a l eva lua tion of the ASSR
The ASSR show s p rom ise in a d d ressing som e of the lim ita tions of the ABR;
how ever it still need s to b e va lid a ted in the c linic a l field – esp ec ia lly in the
p ed ia tric field , b efore it c a n b e rec om m end ed for c linic a l use.
61
University of Pretoria etd – Stroeb
bel, D (2006
The lim ited d a ta b a se for infa nts w ith hea ring loss is a m a tter of g rea t
c onc ern. Ac c ord ing to Sta p ells (2004: c onferenc e p resenta tion), rela tively
few stud ies a re a va ila b le. Of these stud ies the tota l sa m p le size is not la rg e
– esp ec ia lly for the m ultip le ASSR. Of these stud ies c om p a risons w ere
m a d e w ith the c lic k ABR, w hic h is ina p p rop ria te a s this m ea sure d o not
g ive freq uenc y sp ec ific inform a tion. Only tw o stud ies c om p a red infa nt
ASSR to tone evoked ABR, b ut only for 500 Hz. Only a few stud ies inc lud ed
a c om p a rison b etw een the ASSR a nd b eha viora l threshold . All of a b ove
stud ies inc lud ed only Air Cond uc tion (AC) ASSR. No inform a tion is
a va ila b le on Bone Cond uc tion (BC) in infa nts w ith hea ring loss or infa nts
w ith c ond uc tive a nd / or m ixed hea ring losses (Sta p ells, 2004: c onferenc e
p resenta tion). Lim ited inform a tion is a va ila b le a b out infa nts w ith m ild or
m od era te hea ring loss.
Som e rec ent stud ies show ed the p ossib ility of sp urious/ a rtific ia l ASSR’ s a t
hig h intensity stim uli (Sm a ll & Sta p ells, 2004:611; Gorg a et a l., 2004:302;
Jeng et a l, 2004:67; Pic ton & John, 2004:541). ASSR threshold s w ere
m ea sured in sub jec ts w ho ha d no b eha viora l resp onses to sound a t the
lim its of p ure-tone a ud iom eters. It m a y thus a p p ea r tha t som e resp onses in
infa nts w ith p rofound SNHL m a y not b e a ud itory. Som e of these sp urious
resp onses m a y b e d ue to a lia sing , thus a sig na l p roc essing issue a nd other
sp urious resp onses a re likely p hysiolog ic a nd m a y b e a vestib ula r resp onse
(Sta p ells, 2004: c onferenc e p resenta tion). Clinic a lly this m a y b e of little
c onseq uenc e, a s these p a tients w ill in a ll likelihood rec eive c oc hlea r
im p la nts (Gorg a et a l., 2004:302). The m a nufa c turer w a s m a d e a w a re of
this p rob lem a nd c orrec tion w a s m a d e to the softw a re (Persona l
c orresp ond enc e: Biolog ic system s).
62
University of Pretoria etd – Stroeb
bel, D (2006
Althoug h Jeng et a l. (2004:67) a nd Cone-Wesson et a l. (2002:271),
rec ord ed ASSR using BC, w ith their results d em onstra ting a g ood
c orrela tion b etw een estim a ted a ir-b one g a p (ABG) using p ure tone
a ud iom etry a nd ASSR, the sub jec ts used in these stud ies w ere a d ults a nd
therefore no inform a tion on BC ASSR a re a va ila b le for infa nts. Da ta from
sub jec ts w ith p rofound hea ring loss a lso d em onstra ted tha t the levels
w here stim ulus a rtifa c ts b ec om e p rob lem a tic , w ere rela tively low (Jeng et
a l., 2004:67; Sm a ll & Sta p ells, 2004:611). Sm a ll & Sta p ells (2004:622)
c onc lud ed their stud y tha t a lthoug h ASSR’ s a p p ea r to b e p rom ising ,
b one-c ond uc tion ASSR’ s w ill not b e rea d y for c linic a l use until there a re
norm a tive threshold d a ta for infa nts of d ifferent a g es.
Op tim a l stim uli a nd a na lysis is not yet d eterm ined . Ac c ord ing to Sta p ells
(2004: c onferenc e p resenta tion), this, in itself, is not a p rob lem . How ever,
the sm a ll c linic a l d a ta b a se a va ila b le ha s used d ifferent p rotoc ols, e.g .
sing le vs. m ultip le ASSR’ s, F-test a na lysis, noise c riteria , stop p ing rules, to
na m e a few . Another c onc ern is the d ura tion of the stim ulus w hen
a ssessing the p rofound SNHL a s the d ura tion of hig h-intensity stim ula tion
c ould result in a c oustic tra um a (Sta p ells, 2004: c onferenc e p resenta tion).
Resea rc h stud ies (Ra nc e
et a l., 2005:297; Cone-Wesson, 2002:185;
Dim itrijevic et a l., 2002:205; Va nd er Werff, 2002:227) show tha t the ASSR
p erform in the c linic a l p ed ia tric setting a nd the results to the d a ta from
these resea rc h stud ies a re very p rom ising . The c onc erns m entioned a b ove
a re surm ounta b le a nd used in c onjunc tion w ith the ABR, the ASSR c a n
p rovid e a d d itiona l inform a tion a b out the c onfig ura tion a nd d eg ree of
a ny existing hea ring loss. Som e q uestions still rem a in:
63
University of Pretoria etd – Stroeb
bel, D (2006
The neura l g enera tors of the resp onse a re still in d isp ute, p a rtic ula rly a s a
func tion of MF. Cone-Wesson (2002:281) feels tha t this should not lim it
a d op tion of the ASSR in the c linic a s the p rec ise sites a nd struc tures
involved in the ABR ha ve not b een fully d efined either. The effec t of
neuro-d evelop m ent a nd neuro-m a tura tion insult on the ASSR is a c ritic a l
issue for investig a tion. A rela ted issue is the d efinition of norm a l “ threshold ”
for the ASSR a s a func tion of a g e – a s this is exp ec ted to va ry w ith b oth
m a tura tion of the a ud itory system p erip hery a nd the c entra l a ud itory
nervous system .
ASSR’ s ha ve not yet b een exp loited for neuro-otolog ic d ia g nosis. It is likely
tha t m ea sures of p ha se c oherenc e a nd a lso of la tenc y c ould b e used to
ind ic a te retroc oc hlea r a b norm a lities for sup ra threshold stim uli (Sining er &
Cone-Wesson, 2002:319).
Lins & Pic ton (1995:420) investig a ted the p hysiolog y und erlying the ASSR –
using m od ula tion ra tes b etw een 150-190 Hz. Eq ua l c ontrib utions b etw een
the b ra instem a nd c ortic a l a rea s w ere noted a t these hig her m od ula tion
ra tes. These resea rc hers hyp othesize tha t som e insig ht m a y b e g a ined into
p a tholog y of the a ud itory system up to c ortic a l level.
Resea rc h is still req uired to esta b lish w hether sing le m od ula ted tones offer
hig her freq uenc y sp ec ific ity a t hig h stim ula tion intensities. Gorg a , Neely,
Hoover, Dierking , Bea uc ha ine a nd Ma nning (2004:306) c a utions the
interp reta tion of hig h-level ASSR threshold m ea surem ents – using the m ultifreq uenc y system , a s it m a y not p rovid e inform a tion a b out p erip hera l
hea ring . Clinic a lly this m a y b e of little c onseq uenc e, a s these p a tients w ith
“ resp onses” ob served a t suc h hig h levels w ill in a ll likelihood rec eive
c oc hlea r im p la nts. Resea rc h is a lso req uired to esta b lish w hether a id ed
64
University of Pretoria etd – Stroeb
bel, D (2006
threshold s c a n b e ob ta ined from c oc hlea r im p la nt users, using a n
a d a p ter c a b le, to m a xim ize usa g e of elec trod e c onfig ura tions in the
m a p s (Ma ra is, 2003:37).
2.3.2.5
Sum m a ry of the ASSR a p p lic a tion in p ed ia tric a ud iolog y
As a c onc lusion to this c ritic a l eva lua tion of the ASSR, Ta b le 2.3 ind ic a tes
the a d va nta g es a s w ell a s the lim ita tions of the ASSR.
Ta ble 2.3
Adva nta ges a nd limita tions of the ASSR
Adva nta ges
Limita tions
Dia gnosis
•
Freq uenc y sp ec ific
– a p p roxim a te
•
p ure tones
•
Sta b le
–
Ob jec tive
resista nt
to
sta te
•
Disting uish
va lid a tion
–
¾Bone-c ond uc tion
of
¾Dura tion of hig h-intensity stim uli
a utom a tic
¾New eq uip m ent
d etec tion of
¾Sp urious/ a rtific ia l ASSR
resp onse
•
c linic a l
esp ec ia lly in the p ed ia tric field :
c onsc iousness
•
Req uires
b etw een
severe
a nd
•
Ca nnot
d ifferentia te
b etw een
p rofound losses
hea ring loss of p erip hera l orig in a nd
Rela tively fa st p roc ed ure
those w ith neura l tra nsm ission or
retroc oc hlea r orig in
Va lida tion Process
•
Provid es a b ility to eva lua te hea ring
a id s
•
Req uires c linic a l va lid a tion
¾Very lim ited resea rc h rep orts on
a p p lic a b ility
of
this
unc onventiona l a p p lic a tion of
the ASSR
65
University of Pretoria etd – Stroeb
bel, D (2006
It is evid ent tha t the ASSR show s g rea t p rom ise for the c linic a l field of
p ed ia tric a ud iolog y a s va rious resea rc hers ha ve d em onstra ted the
a d va nta g es of the ASSR, over other AEP tec hniq ues, suc h a s the ABR to
use a s a n ob jec tive p roc ed ure to id entify the na ture, d eg ree, sym m etry
a nd c onfig ura tion of the hea ring loss in infa nts a s w ell a s va lid a tion of
hea ring a id s. It is im p era tive how ever tha t m ore resea rc h va lid a te this
p roc ed ure a g a inst the ABR – the c urrent g old sta nd a rd in c linic a l p ra c tic e
for p ed ia tric a ud iolog y.
2.4
CONCLUSION
The need
for a tec hniq ue to estim a te freq uenc y-sp ec ific
hea ring
threshold s in a c linic a lly tim e-effic ient m a nner in the d iffic ult-to-test
p op ula tions ha s long b een a p riority in the field of p ed ia tric a ud iolog y
(Ha yes & Northern, 1997:234). Aud itory Evoked Potentia ls ha ve b een used
in d ia g nostic a ud iolog y for the p a st three d ec a d es a nd it is c lea r tha t in
the field of ob jec tive a ud iolog y, la rg e strid es ha ve b een m a d e in
a d d ressing this im p orta nt need .
The m ost w id ely used AEP tec hniq ue c urrently used to d eterm ine hea ring
threshold s in infa nts is the ABR. This tec hniq ue – using a c lic k stim ulus, c a n
p rovid e a g enera l eva lua tion of hea ring sensitivity in the hig h freq uenc y
reg ion (2 – 4 kHz). By using tone b urst stim uli, m ore freq uenc y sp ec ific
inform a tion w ill b e p rovid ed . Althoug h the ABR is a va lua b le tool, it
p resents w ith im p orta nt lim ita tions.
The ASSR ha ve b een used in a ud iolog y resea rc h c enters a round the w orld
for tw o d ec a d es a nd ha s d em onstra ted p rom ise in a d d ressing som e of
the lim ita tions of the ABR (Cone-Wesson et a l., 2002:273). The results from
66
University of Pretoria etd – Stroeb
bel, D (2006
c linic a l stud ies ha ve show n tha t ASSR threshold s c a n b e used to p red ic t
p ure-tone threshold in sleep ing infa nts a nd young c hild ren (John et a l.,
2004; Ra nc e et a l., 2002; Ra nc e et a l., 1998; Ra nc e et a l., 1995). ASSR
should therefore ha ve a n inc rea sing role in the follow -up a nd d ia g nostic
eva lua tion of infa nts w ho ha ve fa iled new b orn hea ring sc reening . Used in
c onjunc tion w ith ABR (AC a nd BC tone-evoked ABR), ASSR’ s p rovid e
a d d itiona l inform a tion a b out the c ontour a nd d eg ree of a ny existing
hea ring loss (Sta p ells, 2004: c onferenc e p resenta tion; Cone-Wesson et a l.,
2002:281). The ASSR a lso show s g rea t p rom ise a s a w a y to va lid a te hea ring
a id fitting s ob jec tively in sub jec ts w ho c a nnot relia b ly resp ond to
b eha viora l testing , b ut resea rc h d a ta is still lim ited .
2.5
SUMMARY
This c ha p ter a im ed to orienta te the rea d er on the top ic s of releva nc e a nd
to p rovid e a c ritic a l eva lua tion a nd interp reta tion of the releva nt
litera ture. In ord er to a c hieve this, the m ost w id ely used AER tec hniq ue for
estim a ting a ud itory threshold s in infa nts, na m ely the ABR w a s d esc rib ed ,
eva lua ted a nd d isc ussed . Sub seq uently the im p orta nc e of the hea ring
a id fitting p roc ess w a s d isc ussed – d esc rib ing the d ifferent, b ut eq ua lly
im p orta nt a sp ec ts of verific a tion a nd va lid a tion. The role of ea c h a sp ec t
in the hea ring a id fitting p roc ess w a s c la rified . La stly the ASSR w a s
d isc ussed a s a n AEP p rom ising to a d d ress the c urrent lim ita tions of the
ABR. Fina lly the g enera l id ea s of the c ha p ter w ere sum m a rized in the
c onc lusion.
67
University of Pretoria etd – Stroeb
bel, D (2006
Cha p ter 3
RESEARCH METHODOLOGY
This c ha p ter a im s to exp la in the m ethod used to c ond uc t the resea rc h
c om p onent of this stud y. This w ill b e d isc ussed in term s of a im s set for
this resea rc h, the resea rc h d esig n, ethic a l c onsid era tions, sub jec ts,
m a teria l a p p a ra tus a nd p roc ed ures used .
3.1
INTRODUCTION
“ Resea rc h ha s one end : the ultim a te d isc overy of truth” (Leed y & Om rod ,
2001: xviii). Its p urp ose is to lea rn w ha t ha s never b een know n b efore; to
a sk a sig nific a nt q uestion for w hic h no c onc lusive a nsw er ha s p reviously
b een found ; a nd , throug h the m ed ium of releva nt d a ta a nd their
interp reta tion, to find a n a nsw er to tha t q uestion (Leed y & Om rod ,
2001:xviii).
Cha p ter one introd uc ed the p rob lem surround ing this resea rc h end ea vor.
It a lso p rovid ed a ra tiona le for the stud y a nd exp la ined the resea rc h
q uestion. Cha p ter tw o p rovid ed a theoretic a l fra m ew ork, a s sup p ort for
the em p iric a l resea rc h c om p onent, c onc ep ts a nd c onstruc ts w ere then
sp ec ified . Cha p ter tw o a lso p rovid ed a n interp reta tion of the c urrent a nd
releva nt litera ture a va ila b le.
68
University of Pretoria etd – Stroeb
bel, D (2006
This c ha p ter a im s to exp la in the m ethod olog ic a l a p p roa c h im p lem ented
in c ond uc ting the em p iric a l c om p onent of the c urrent stud y.
3.2
AIMS OF RESEARCH
Sig nific a nt
c orrela tions
b etw een
ASSR threshold s
a nd
b eha viora l
a ud iom etric threshold s a s w ell a s c orrela tions b etw een the ASSR a nd ABR
a s a threshold p red ic tion tec hniq ue ha ve b een found b y severa l
resea rc hers (Dim itrijevic , 2002:205; Cone-Wesson et a l., 2002:173; Va nd er
Werff et a l., 2002:227; Herd m a n & Sta p ells, 2001:41; Lins et a l., 1996:81;
Ra nc e et a l., 1995:499). Althoug h these results ind ic a te the ASSR to b e a
p rom ising tec hniq ue in d eterm ining the a ud itory a b ility of a d ults, the need
a rises to va lid a te this p roc ed ure for the infa nt p op ula tion.
Sta p ells (2002:14 & 2004:c onferenc e p resenta tion) c a utioned a ud iolog ists
a b out the use of the ASSR in the c linic a l setting in the infa nt p op ula tion, a s
only a few stud ies ha d b een d one in this reg a rd . This p resent stud y
foc used on the use of ASSR in the d ia g nosis of hea ring loss a nd va lid a tion
of hea ring a id fitting in infa nts. The im p orta nc e of this tec hniq ue, should it
p rove to ha ve va lid c linic a l a p p lic a tion, is evid ent for the d iffic ult-to-test
p op ula tions. Therefore the a im s of the c urrent stud y a re a s follow s:
3.2.1
Ma in a im
The m a in a im of the stud y is to investig a te the c linic a l va lue of the ASSR for
ea rly d ia g nosis a nd for ea rly hea ring instrum ent fitting of hea ring loss in
infa nts.
69
University of Pretoria etd – Stroeb
bel, D (2006
3.2.2
Sub a ims
The follow ing sub -a im s w ere form ula ted in ord er to rea lize the m a in a im of
the stud y:
•
To investig a te the p otentia l c linic a l va lue of the ASSR for ea rly
d ia g nosis of a hea ring loss in a g roup of infa nts b y d eterm ining a nd
c om p a ring the:
•
Una id ed ABR threshold s (c lic k a nd toneb urst) a t the a g e of 3
– 6 m onths
•
Una id ed ASSR threshold s a t the a g e of 3 – 6 m onths
•
Una id ed b eha viora l threshold s a t the a g e of 8 – 12 m onths
(a fter a tim e la p se of 2 – 6 m onths follow ing d ia g nosis)
•
To investig a te the p otentia l c linic a l va lue of the ASSR for ea rly
hea ring a id fitting in a g roup of infa nts b y:
•
Determ ining a id ed ASSR a t the tim e of hea ring a id fitting
•
Com p a ring una id ed a nd a id ed ASSR a t the tim e of hea ring
instrum ent fitting
•
Determ ining a id ed b eha viora l threshold s a t the a g e of 8 – 12
m onths a nd c om p a ring these results w ith a id ed ASSR’ s.
3.3
RESEARCH DESIGN
Ba b b ie a nd Mouton (2002:72) sa id sc ienc e is a n enterp rise d ed ic a ted to
“ find ing out” . Resea rc h d esig n a d d resses the p la nning of sc ientific
enq uiry, d esig ning a stra teg y for find ing out som ething sp ec ific . The
d esig n is the c om p lete stra teg y of ta c kling the c entra l p rob lem . It p rovid es
the
struc ture
w ithin
w hic h
the
selec ted
va ria b les a re
c ontrolled ,
70
University of Pretoria etd – Stroeb
bel, D (2006
m a nip ula ted a nd m ea sured (Heg d e, 1987:135). The m ethod of resea rc h is
d efined b y Leed y a nd Orm rod (2001:100) a s the fra m ew ork to extra c t the
m ea ning from the d a ta c ollec ted .
In this sec tion the resea rc h p la n is d esc rib ed in term s of the g oa l set, the
a p p roa c h follow ed
a nd
the sp ec ific
resea rc h d esig n utilized . An
exp lora tory, c orrela tive-d esc rip tive stud y (Bellis, 2003:433) w ith a q ua siexp erim enta l d esig n (Leed y & Orm rod , 2005:231), im p lem enting a
q ua ntita tive resea rc h a p p roa c h, w a s selec ted to a c hieve the a im s of this
stud y.
The g oa l or p urp ose of this stud y w a s to exp lore, d esc rib e a nd c orrela te
(Bellis, 2003:433). Explora tory resea rc h is typ ic a lly used w hen a resea rc her
is exa m ining a new interest or w hen the sub jec t of stud y is itself rela tively
new a nd unstud ied (Ba b b ie, 1992:90). The ASSR is a rela tively new a d junc t
to the field of Aud iolog y a nd sp ec ific a lly need s va lid a tion in the p ed ia tric
field . The g oa l of descriptive resea rc h is to d esc rib e the c ha ra c teristic s of
a selec ted p henom enon (Bellis, 2003:436). In this stud y inform a tion w a s
c ollec ted w ith reg a rd s to d ifferent test m ethod s in a g roup of sub jec ts. The
results from this g roup ’ s p erform a nc e w a s rec ord ed a nd d esc rib ed . The
g oa l of the stud y a s reflec ted in the m a in- a nd sub -a im s w a s therefore to
exp lore, c orrela te a nd d esc rib e the c linic a l va lue of the ASSR a s
c om p a red w ith the ABR for d ia g nosis of hea ring loss. The role of the ASSR
a nd va lid a tion p roc esses of hea ring instrum ent in infa nts a re a lso exp lored
a nd d esc rib ed . A correla tive stud y exa m ines the extent to w hic h
d ifferenc es in one c ha ra c teristic or va ria b le a re rela ted to d ifferenc es in
one or m ore other c ha ra c teristic s or va ria b les (Leed y & Orm rod , 2001:191).
In this stud y a c orrela tion w a s m a d e b etw een the tw o d ifferent m ethod s
utilized to estim a te infa nts’ hea ring a b ilities. A further c orrela tion w a s
d ra w n b etw een the a id ed a nd una id ed p red ic ted threshold s d one
71
University of Pretoria etd – Stroeb
bel, D (2006
throug h ASSR tec hnolog y a nd the g old sta nd a rd of b eha viora l threshold
m ea sures – looking a t b oth the a id ed a nd una id ed b eha viora l threshold s.
A
resea rc h a p p roa c h w a s im p lem ented . Qua ntita tive
q ua ntita tive
resea rc h is used to a nsw er q uestions a b out rela tionship s a m ong m ea sured
va ria b les w ith the p urp ose of exp la ining , p red ic ting , a nd c ontrolling
p henom ena (Leed y & Orm rod , 2001:101). Qua ntita tive resea rc hers seek
exp la na tions a nd p red ic tions tha t w ill g enera lize to other p ersons a nd
p la c es (Leed y & Orm rod , 2001:102). Qua ntita tive d a ta c ollec tion m ethod s
w ere selec ted for this stud y d ue to the na ture of the d a ta to b e c ollec ted ,
na m ely threshold estim a tion va lues, b eha viora l threshold s, func tiona l g a in
estim a tions a nd
func tiona l g a in b eha viora l threshold s. Qua ntita tive
resea rc h is a lso used to a nsw er q uestions a b out rela tionship s a m ong
m ea sured va ria b les, w ith the p urp ose of exp la ining , p red ic ting a nd
c ontrolling p henom ena (Leed y & Orm rod , 2005:231). This stud y a im ed to
look a t the rela tionship b etw een d ifferent m ethod s used to p red ic t
threshold s
in
infa nts a nd
func tiona l
g a in
m ea surem ents.
During
q ua ntita tive resea rc h, sta nd a rd ized p roc ed ures a re used to c ollec t
num eric a l d a ta (Leed y & Orm rod , 2001:191). The va ria b les to b e stud ied
a re usua lly isola ted a nd extra neous va ria b les a re c ontrolled . This typ e of
d a ta c ollec tion a llow s for the use of sta tistic a l p roc ed ures to a na lyze a nd
interp ret the d a ta .
This stud y lend s itself to a q ua si-exp erim enta l d esig n (Drum m ond , 2003:32).
When c ond uc ting a qua si-experimenta l stud y, a ll c onfound ing va ria b les
c a nnot b e c ontrolled . Va ria b les a nd exp la na tions tha t ha ve not b een
c ontrolled for need to b e ta ken into c onsid era tion w hen d a ta is
interp reted (Leed y & Orm rod , 2005:227). Ac c ord ing to Mouton (2001:160),
a q ua si-exp erim enta l d esig n is usua lly q ua ntita tive in na ture, a im s to
72
University of Pretoria etd – Stroeb
bel, D (2006
p rovid e a c a usa l stud y of a sm a ll num b er of c a ses und er c ontrolled
c ond itions a s in this stud y. Sub seq uently this stud y la c ks the ing red ient of
ra nd om iza tion tec hniq ues of a true exp erim enta l d esig n.
A c ontrolled test environm ent, w ith uniform ity in test eq uip m ent a nd test
p rotoc ol, w a s selec ted to c ontrol environm enta l c ond itions. The va lid ity of
the exp lora tory stud y w a s enha nc ed b y the inc lusion of six sub jec ts.
3.4
ETHICAL CONSIDERATIONS
Sc ientists c onsid er resea rc h to b e a n ethic a l a c tivity. Resea rc hers seek
know led g e, solve p rob lem s, a nd d esig n new m ethod s of trea ting d isea ses
a nd d isord ers, b ut they ha ve the resp onsib ility of d oing a ll of this in a n
honest, resp onsib le, op en a nd ethica lly justifia b le m a nner (Heg d e,
1987:414). The b a sic tenet of ethic a l resea rc h is to p reserve a nd p rotec t
the hum a n d ig nity a nd rig hts of a ll sub jec ts involved in a resea rc h p rojec t
(Jenkins, Pric e & Stra ker, 2003:46).
The b a sic ethic a l p rinc ip les of a utonom y, b enefic enc e a nd justic e (Hyd e,
2005:297; Louw , 2004:1) w ere inc orp ora ted in this stud y.
3.4.1 Autonomy
Autonom y refers to the freed om of w ill, the rig ht to self-g overnm ent a nd
p ersona l freed om (Conc ise Oxford Dic tiona ry, 1984:59). In resea rc h,
a utonom y refers to stric tly volunta ry p a rtic ip a tion (Leed y & Orm rod ,
2001:107), to c hoose w hether or not to b e rec ip ients of sp ec ific a c tions
(Hyd e, 2005:297). The p erson involved m ust ha ve the leg a l c a p a c ity to
g ive c onsent (Jenkins, Pric e & Stra ker, 2003:47). The infa nt is a m inor a nd
73
University of Pretoria etd – Stroeb
bel, D (2006
therefore the p a rent or c a reg iver b ec a m e the a d voc a te for the infa nt. In
this stud y, the p a rent or c a reg iver ha d the resp onsib ility to a c t in the b est
interest of the infa nt.
•
Informed consent
Ea c h sub jec t’ s p a rent or c a reg iver w a s req uested to g ive w ritten
p erm ission for p a rtic ip a tion in this stud y. A letter of inform ed c onsent w a s
d ra w n up (Ap p end ix B). This letter exp la ined the p urp ose a nd na ture of
this stud y (Leed y & Orm rod , 2001:107). The letter inform ed the p a rents or
c a reg ivers of the infa nts of w ha t w a s exp ec ted of them a nd a b out their
a nd their infa nt’ s rig hts. Sub jec ts’ rig hts inc lud ed the follow ing :
Withd ra w a l of p a rtic ip a nts
The p a rents/ c a reg ivers w ere g iven the a ssura nc e tha t they ha d the rig ht
to w ithd ra w their b a b y a s a sub jec t from this stud y a t a ny tim e.
Priva c y, c onfid entia lity, a nonym ity
Pa rents’ / c a reg ivers’ p erm ission w a s req uested to use inform a tion in
p ersona l c lient rec ord s of a p riva te p ra c tic e, for resea rc h p urp oses. All
inform a tion used w a s c onfid entia l. The p riva c y of a ll sub jec ts w a s up held .
A letter sta ting the la tter w a s g iven to ea c h sub jec t (Ap p end ix B).
Disc losure of inform a tion
The p a rents/ c a reg ivers of sub jec ts w ere inform ed of the fa c t tha t the
results from this stud y m a y in future b e used in the p ub lishing of a sc ientific
74
University of Pretoria etd – Stroeb
bel, D (2006
a rtic le or c onferenc e or sem ina r p resenta tion. Inform a tion m ig ht b e
d isc ussed a t a c a d em ic g a thering s.
Deb riefing of resp ond ents
All inform a tion g a thered from this resea rc h w a s m a d e a va ila b le to the
p a rents or c a reg ivers. Resea rc h find ing s w ere sum m a rized in a letter a nd
sent to
ea c h
p a rtic ip a nt. Sinc e
these
letters c onta ined
p ersona l
inform a tion, no c op ies a re inc lud ed in the a p p end ix of this resea rc h
rep ort.
•
Ethica l Clea ra nce
This stud y ha d ethic a l c lea ra nc e from the Resea rc h Prop osa l a nd Ethic s
Com m ittee of the Fa c ulty of Hum a nities, University of Pretoria . A letter of
c onfirm a tion to this effec t is inc lud ed in Ap p end ix A. Sinc e the sub jec ts
w ere c lients of the resea rc her’ s ow n p riva te p ra c tic e, no ethic a l
c lea ra nc e or letter of inform ed c onsent to a nother institution w a s
req uired .
3.4.2 BENEFICENCE
Benefic enc e refers to a c ting in kind ness (Conc ise Oxford Dic tiona ry,
1984:83) or to the c onferra l of b enefits (Hyd e, 2005:297). Resea rc hers
should
not
exp ose
resea rc h
p a rtic ip a nts to
und ue
p hysic a l
or
p syc holog ic a l ha rm (Leed y & Orm rod , 2001:107; Ba b b ie, 1992:465). The risk
involved in p a rtic ip a ting in a stud y should not b e a p p rec ia b ly g rea ter
tha n the norm a l risks of d a y-to-d a y living . This a sp ec t w a s d ea lt w ith in the
follow ing m a nner:
75
University of Pretoria etd – Stroeb
bel, D (2006
•
Competency
The resea rc her w a s c om p etent to c a rry out the resea rc h d ue to her
p rofessiona l q ua lific a tion, a s w ell a s yea rs of exp erienc e in the field of
Aud iolog y. Tw o sup ervisors w ere involved in this p roc ess – ensuring a
suita b le resea rc h d esig n a nd g iving g uid a nc e in the p roc ess of resea rc h.
The resea rc her (STA 011037) a nd the sup ervisors w ere reg istered w ith the
Hea lth Professions Counc il of South Afric a .
•
Releva nce
The top ic of resea rc h w a s hig hly releva nt a t the tim e of d evelop m ent in
the Aud iolog ic field a s is ind ic a ted in the ra tiona le for the stud y.
•
Risks
Potentia l m ed ic a l risks involved in this stud y w ere c onsid ered a nd
a d d ressed . The usua l p roc ed ures a nd c a re m a inta ined in the c linic a l
p ra c tic e a p p lied . In c a ses w here sub jec ts need ed sed a tion, c hlora l
hyd ra te w a s p resc rib ed b y a p ed ia tric ia n a nd a d m inistered ora lly b y a
q ua lified a nd exp erienc ed p ed ia tric nurse. The p ed ia tric nurse m onitored
sub jec ts for oxyg en sa tura tion, resp ira tory ra te a nd hea rt ra te.
•
Discrimina tion
There w a s no d isc rim ina tion b etw een sub jec ts on the g round s of ra c e,
ec onom ic sta tus or g end er. Pa rents/ c a reg ivers w ere g iven the a ssura nc e
tha t their b a b y’ s sta tus a s c lient of the resea rc her’ s p riva te p ra c tic e w ould
not b e influenc ed b y their c onsent or refusa l to p a rtic ip a te in the stud y.
76
University of Pretoria etd – Stroeb
bel, D (2006
3.4.3 JUSTICE
In resea rc h ‘ justic e’ refers to honesty w ith p rofessiona l c ollea g ues (Leed y
& Orm rod , 2001:108). It a lso rela tes to fa irness in the d istrib ution or
a lloc a tion of b enefits a m ong m em b ers of soc iety (Hyd e, 2005:297).
Resea rc hers m ust rep ort their find ing s in a c om p lete a nd honest fa shion.
‘ Justic e’ w a s a d d ressed in the follow ing m a nner:
•
Dissemina tion of results
Resea rc h results w ere m a d e a va ila b le to a ll p a rtic ip a nts. The results w ere
m a d e a va ila b le to the p rofessiona ls in the field of Aud iolog y in ord er to
g a in know led g e of new d evelop m ents in the field a s w ell a s im p rove
servic e d elivery. Resea rc h find ing s w ere p ub lished in the form of a
resea rc h a rtic le, w hic h m a y b e used a nd d istrib uted b y the p ub lic .
3.5
SUBJECTS
Six infa nts w ith hea ring loss w ere id entified a s sub jec ts for this stud y.
3.5.1 Sa mpling
The sub jec ts inc lud ed in this stud y w ere selec ted , b a sed on a nonp rob a b ility c onvenienc e sa m p ling a p p roa c h (Ba b b ie, 1992:230). The
sub jec ts w ere selec ted from the c linic a l c a seloa d of the resea rc her’ s
p riva te p ra c tic e in Ca p e Tow n. These w ere infa nts referred for follow -up
eva lua tions a fter fa iling a sc reening eva lua tion.
77
University of Pretoria etd – Stroeb
bel, D (2006
3.5.2 Selection criteria
The sub jec ts w ere selec ted a c c ord ing to the follow ing c riteria :
3.5.2.1
Client sta tus a nd rec ord
Sub jec ts w ere c lients of the resea rc her’ s p riva te p ra c tic e of w hom
inform a tion is a va ila b le a nd w hose p a rent/ c a reg ivers ha d g iven their
c onsent for the b a b y to b e inc lud ed in the stud y (Ap p end ix B).
3.5.2.2
Hea ring a b ility
Sub jec ts w ere those w ho w ere referred for elec trop hysiolog ic a ssessm ent
a fter fa iling
a
c lic k-evoked
ABR sc reening
a nd
OAE’ s sc reening
a ssessm ent.
3.5.2.3
Norm a l Mid d le Ea r Func tioning
Sub jec ts w ere inc lud ed only if they show ed no evid enc e of m id d le ea r
p a tholog y in ord er to rule out a ny other fa c tors influenc ing tests results.
The m id d le ea r sta tus w a s d eterm ined b y otosc op ic exa m ina tion a nd hig h
freq uenc y tym p a nom etry – using a 1000 Hz p rob e tone a nd a n
exa m ina tion b y a n Ea r-Nose-a nd Throa t surg eon. A sing le-p ea ked hig h
freq uenc y (1000 Hz) tym p a nog ra m w a s ind ic a tive of norm a l m id d le ea r
func tion (Kei et a l., 2003:27).
78
University of Pretoria etd – Stroeb
bel, D (2006
3.5.2.4
Ag e a t tim e of id entific a tion
Infa nts2 w ere selec ted for this stud y, throug h referra l, fa iling a hea ring
sc reening p rotoc ol. At the tim e of id entific a tion, these infa nts w ere too
young to m ea sure hea ring a b ilities throug h tra d itiona l b eha viora l m ethod s
a nd it w a s therefore a p p rop ria te to use elec trop hysiolog ic m ea sures.
Onc e a n infa nt a c hieved a d evelop m enta l a g e of a p p roxim a tely six to
eig ht m onths, a ud iom etric inform a tion c ould b e ob ta ined effic iently using
a b eha viora l tec hniq ue, b a sed on p rinc ip les of c ond itioning (Diefend orf &
Web er, 1994:57). When the sub jec ts rea c hed this a g e, b eha viora l
m ethod s w ere used to d eterm ine una id ed a nd a id ed threshold s.
3.5.2.5
Neurolog ic a l sta tus
In ord er to rule out the p resenc e of a ud itory neurop a thy (neura l
tra nsm ission d isord er), b oth a n ABR a nd OAE eva lua tion w a s c ond uc ted .
In the c a se of a n a ud itory neurop a thy, the OAE or c oc hlea r resp onse
w ould still b e p resent w ith a n a b norm a l or a b sent ABR. Sub jec ts w ere
inc lud ed w hen the ABR a ssessm ent show ed no evid enc e of a neura l
tra nsm ission d isord er (Ra nc e & Ric ka rd s, 2002:237). This a sp ec t w a s further
a d d ressed b y m ea suring OAE’ s – the a b senc e of these OAE’ s c onfirm ed
the a b senc e of a ud itory neurop a thy.
Infa nt a s d efined b y Conc ise Oxford Dic tiona ry (1984:513): c hild d uring ea rliest p eriod of
life – b efore a g e 1.
2
79
University of Pretoria etd – Stroeb
bel, D (2006
3.5.3
Subject Selection Appa ra tus
The follow ing a p p a ra tus w ere used in the selec tion p roc ed ures of the
sub jec ts:
3.5.3.1
Hea ring Sc reening Ap p a ra tus
Sc reening
for hea ring
loss throug h Autom a ted
Aud itory Bra instem
Resp onse tec hniq ues (AABR) a nd Otoa c oustic Em issions (OAE) m ethod s
w ere d one on the ABAER from Biolog ic Aud iom etric System s, (c a lib ra ted
June 2004). The sound w a s tra nsd uc ed into the ea r c a na l b y the p rob e
m ic rop hone.
The ABR is a p hysiolog ic a l m ea sure of the a ud itory system to stim uli
p resented to the ea r. Short-d ura tion ‘ c lic k-stim uli’ w a s p resented to ea c h
ea r via the p rob e m ic rop hone a t 35 d BnHL. Rec ord ing w a s d one using a
three-elec trod e m onta g e of hig h forehea d a nd the m a stoid b one of
ea c h ea r. A m a xim um im p ed a nc e of 8 kOhm w a s a llow ed w ith a
m inim um d ifferenc e of 4 kOhm .
Otoa c oustic Em ission m ea surem ents w ere p erform ed on ea c h sub jec t.
Distortion Prod uc t Otoa c oustic Em issions using a 65/ 55 d B p rob e tone w a s
p erform ed on ea c h sub jec t (Ha ll & Mueller, 1997:247). A 2 – 5 kHz
sc reening p rotoc ol w ith ¾ p a ss ra te w a s used . The a b senc e of a resp onse
w ith these p a ra m eters w a s ind ic a tive of the p resenc e of a hea ring loss
g rea ter tha n 30 d B.
80
University of Pretoria etd – Stroeb
bel, D (2006
3.5.3.2
Otosc op ic Exa m ina tion
The otosc op ic
exa m ina tion of the externa l m ea tus a nd tym p a nic
m em b ra ne w ere p erform ed w ith a Heine m ini 2000 otosc op e.
3.5.3.3
Mid d le Ea r Assessm ent
Hig h freq uenc y tym p a nom etry a nd a c oustic reflex m ea surem ents – using
a 1000 Hz p rob e tone w ere p erform ed w ith the GSI Tym p sta r m id d le ea r
a na lyzer (c a lib ra ted June 2004).
Fow ler & Sha nks (2002:201) noted tha t tym p a nom etry – using a hig h
freq uenc y p rob e tone g ive m ore useful inform a tion w ith reg a rd s to the
m id d le ea r system of infa nts. A sing le-p ea ked hig h freq uenc y (1000 Hz)
tym p a nog ra m w a s ind ic a tive of norm a l m id d le ea r func tion (Kei et a l.,
2003:27). Further, the p resenc e of a n a c oustic reflex help ed to c onfirm a
norm a l m id d le ea r system . The a b senc e of the a c oustic reflex in the
p resenc e of norm a l tym p a nom etry sup p orted the p ossib le p resenc e of a
hea ring loss.
3.5.4
Subject Selection Procedures
The six sub jec ts (2 m a le a nd 4 fem a le) inc lud ed in this stud y w ere referred
to the p ra c tic e of the resea rc her for d ia g nostic elec trop hysiolog ic a l
a ssessm ent follow ing fa ilure on a c lic k-evoked ABR sc reening a ssessm ent
a nd a sub seq uent fa ilure on the OAE sc reen. Results of these sc reening
p roc ed ures ruled out the p ossib ility of a sub jec t p resenting w ith a neura l
tra nsm ission d isord er (a ud itory neurop a thy). These a ssessm ents w ere
a d m inistered b y the resea rc her herself.
81
University of Pretoria etd – Stroeb
bel, D (2006
All sub jec ts ha d norm a l m id d le ea r func tion a s d eterm ined b y a n
otosc op ic exa m ina tion a nd hig h freq uenc y tym p a nom etry – using a 1000
Hz p rob e tone. The otosc op ic exa m ina tion w a s p erform ed to insp ec t
w hether a ny visib le ob struc tion w a s p resent tha t c ould a ffec t the
c ond uc tion of sound to the tym p a nic m em b ra ne (Sta c h, 1998:174). Both
the otosc op ic exa m ina tion a nd hig h freq uenc y tym p a nom etry w ere
c ond uc ted on ea c h of the test oc c a sions (ABR sc reening a nd OAE
p reviously a nd the d a y of d ia g nostic a ssessm ent).
The
test
seq uenc e
sta rted
w ith
the
p erform a nc e
of
the
OAE
m ea surem ent. Fa ilure on this eva lua tion w a s follow ed w ith hig h freq uenc y
tym p a nom etry in ord er to exc lud e m id d le ea r p a tholog y. After show ing
norm a l im m itta nc e m ea surem ents, a n AABR a ssessm ent follow ed . The
infa nt w a s c onsid ered a sub jec t for this stud y, a fter fa iling the AABR.
3.6
DESCRIPTION OF SUBJECTS
The sub jec ts inc lud ed six infa nts w ith d ifferent d eg rees of hea ring loss. Tw o
m a le a nd four fem a le sub jec ts w ith a hea ring loss a nd a n a vera g e a g e of
five m onths (a g es ra ng ed from three to six m onths of a g e) w ere id entified .
These w ere b a b ies of w hom hea ring sc reening d a ta sinc e b irth w a s
a va ila b le to the resea rc her. Ta b le 3.1 inc lud es inform a tion reg a rd ing the
a g e of id entific a tion, g end er a nd the d eg ree of hea ring loss. Ad d itiona l
inform a tion
reg a rd ing
the
ind ivid ua l
sub jec ts
a c c om p a nies
the
d esc rip tion of ind ivid ua l results in Cha p ter 4.
82
University of Pretoria etd – Stroeb
bel, D (2006
Ta ble 3.1
Subject
Description of subjects
Gender
number
Age a t time
Degree of hea ring loss
of hea ring
loss
identifica tion
1
Ma le
3 m onths
Mod era tely Severe
2
Fem a le
5 m onths
Mod era tely Severe in rig ht ea r
Mod era te in left ea r
3
Fem a le
6 m onths
Severe in rig ht ea r
Profound in left ea r
3.7
4
Fem a le
6 m onths
Severe
5
Fem a le
4 m onths
Profound
6
Ma le
6 m onths
Profound
MATERIAL AND APPARATUS
The follow ing d a ta c ollec tion a p p a ra tus, m a teria ls a nd p roc ed ures
w ere used for the c ollec tion of d a ta :
3.7.1
The
Hea ring threshold estima tion a ppa ra tus
GSI Aud era from GSI (a d ivision of VIASYS), (c a lib ra ted Novem b er
2003), w a s used to p red ic t hea ring threshold s – using b oth c lic k evoked
a nd tone b urst Aud itory Bra instem Resp onse (ABR) a nd Aud itory Stea d y
Sta te Resp onse tec hniq ues (ASSR). The ABR a nd ASSR w ere rec ord ed
83
University of Pretoria etd – Stroeb
bel, D (2006
using b oth ip sila tera l a nd c ontra la tera l elec trod e m onta g es w ith hig h
forehea d p ositive, the m a stoid s neg a tive a nd the g round elec trod e
p ositioned on the low forehea d (Va nd er Werff et a l., 2002:229). The stim uli
w ere p resented via TIP 50 Insert HA-2 Tub ep hones w ith foa m ea rp lug s.
Elec trod e im p ed a nc e va lues w ere < 5 kOhm s a nd w ere w ithin 1.5 kOhm s
of ea c h other.
The p rotoc ol follow ed for c lic k ABR is rep resented in Ta b le 3.2. Ta b le 3.3
rep resents the p rotoc ol follow ed for tone b urst ABR a nd Ta b le 3.4
rep resents the p rotoc ol for the ASSR m ea surem ents.
84
University of Pretoria etd – Stroeb
bel, D (2006
Ta ble 3.2
Protocol for click ABR
Settings
Pa ra meters
Stimulus
Clic k
Ha ll & Mueller, 1997:334;
Dura tion
0.1 m s
Ha ll & Mueller, 1997:334;
Hood , 1998:54
Tra nsducer
Tip 50 insert ea rp hones
GSI-eq uip m ent
Pola rity
Ra refa c tion
Ha ll & Mueller, 1997:334;
Hood , 1998:52
Ra te
33.1/ sec .
Electrode
Ip sila tera l
pla cement
elec trod e m onta g es w ith:
Impeda nce
Hood , 1998:51
a nd
c ontra la tera l
•
Hig h forehea d - p ositive
•
Ma stoid s – neg a tive
•
Low forehea d – g round .
” N2KPV ZLWK GLIIHUHQFH EHWZHHQ
elec trod es
no
g rea ter
tha n
1.5
kOhm s.
85
University of Pretoria etd – Stroeb
bel, D (2006
Ta ble 3.3
Settings
Stimulus
Protocol for tone burst ABR
Pa ra meters
500 Hz,
Purd y & Ab b a s (2002:359);
Bla c km a n ra m p ing 2-1-2 c yc les
Sta p ells (2000a :17);
Gorg a (1999:37
Filter choice
30 – 1500 Hz
GSI-p rotoc ol
Tra nsducer
Tip 50 insert ea rp hones
GSI-eq uip m ent
Pola rity
Alterna ting
Minim izes a freq uenc y follow ing typ e
of resp onse (Sta p ells, 2000a :17)
Ra te
39.1/ sec
Electrode
Ip sila tera l
pla cement
elec trod e m onta g es w ith:
Impeda nce
Sta p ells (2000a :17)
a nd
c ontra la tera l
•
Hig h forehea d - p ositive
•
Ma stoid s – neg a tive
•
Low forehea d – g round .
” N2KPV ZLWK GLIIHUHQFH EHWZHHQ
elec trod es
no
g rea ter
tha n
1.5
kOhm s.
86
University of Pretoria etd – Stroeb
bel, D (2006
Ta ble 3.4
Protocol for the ASSR
Settings
Pa ra meters
Ca rrier
500, 1000, 2000, 4000 Hz
frequencies
Modula tion
500
1000
2000
4000
Cone-Wesson
frequencies
74
81
88
95
Va nd er Werff et a l. (2002:230).
AM
100%
a l.
(2002:178);
GSI Aud era p rotoc ol.
percenta ge
Cone-Wesson
FM
et
10%
et
a l.
(2002:178);
Va nd er Werff et a l. (2002:230).
percenta ge
Tra nsducer
Number
Tip 50 insert ea rp hones
of
16 (m inim um ) – 64 (m a xim um )
sweeps
Impeda nce
” N2KPV ZLWK GLIIHUHQFH EHWZHHQ
elec trod es no g rea ter tha n 1.5 kOhm s.
Electrode
Ip sila tera l a nd c ontra la tera l elec trod e
pla cement
m onta g es w ith:
•
Hig h forehea d - p ositive
•
Ma stoid s – neg a tive
•
Low forehea d – g round .
87
University of Pretoria etd – Stroeb
bel, D (2006
The stim uli used to evoke the ASSR c onsisted of c a rrier freq uenc ies of
500, 1000, 2000 a nd 4000 Hz tha t w ere 100 p erc ent a m p litud e
m od ula ted a nd 10 p erc ent freq uenc y m od ula ted a t m od ula tion
freq uenc ies of 74, 81, 88 a nd 95 Hz resp ec tively (Cone-Wesson et a l.,
2002:178; Va nd er Werff et a l., 2002:230). The Aud era d evic e a vera g ed
the ong oing EEG a c tivity a nd c om p uted the p ha se c oherenc e of the
sp ec tra l c om p onent of the resp onse a t the m od ula tion freq uenc y.
Sta tistic a l a na lysis w a s used to d eterm ine the p rob a b ility tha t the
ob served resp onse w a s d ue to c ha nc e. Betw een 16 a nd 64 sw eep s
w ere a na lyzed d uring ea c h rec ord ing . The test w a s term ina ted w hen
the p ha se c oherenc e rea c hed sta tistic a l sig nific a nc e or a t 64 sw eep s if
sig nific a nc e w a s not rea c hed . The sig nific a nt level w a s set a t 0.03 (GSI,
2001:3). The ASSR threshold s a re used to estim a te the p ure-tone
a ud iog ra m . This estim a tion utilizes a n a lg orithm b a sed on p ub lished
resea rc h from the University of Melb ourne in w hic h ASSR threshold s
m ea sured for p a tients w ith va rious a m ounts of hea ring loss w ere
c orrela ted w ith their b eha viora l a ud iog ra m s (GSI, 2001:6).
3.7.2
Functiona l Ga in Estima tion Appa ra tus
The GSI Aud era from GSI (c a lib ra ted Novem b er 2003), w a s used to p red ic t
func tiona l g a in. Stim uli w ere tra nsd uc ed throug h a RCA PRO-X33AV
loud sp ea ker in the free field . The loud sp ea ker w a s c a lib ra ted to p resent
stim uli a t 0º, 30 c m from the forehea d .
3.7.3
Clinica l a udiometer
Pure tone threshold s w ere ob ta ined using a GSI 61 Clinic a l Aud iom eter
(c a lib ra ted June 2004). Ac oustic stim uli w ere p resented throug h sound
88
University of Pretoria etd – Stroeb
bel, D (2006
field p resenta tion. Ea r sp ec ific inform a tion w a s rec ord ed using insert
ea rp hones (Sc ollie & Seew a ld , 2002:689). Na rrow b a nd s of noise w ere
used a s test stim uli, a s these infa nts w ere young er tha n 14 m onths of a g e
(Gra vel, 2002:40).
Func tiona l g a in w a s d eterm ined throug h sound field p resenta tion. Na rrow
b a nd s of noise w ere onc e a g a in used a s test stim uli.
3.7.4
Test environment
Beha viora l testing w a s c ond uc ted in a d oub le-w a lled , sound -a ttenua ting
room . Elec trop hysiolog ic a l testing w a s c ond uc ted in a q uiet sid e room of
the p riva te p ra c tic e.
3.7.5
Da ta collection sheet
The c ollec ted d a ta w a s ta b ula ted on a sum m a tive d a ta c ollec tion sheet
(Ap p end ix C).
3.8
PROCEDURE
The follow ing p roc ed ures w ere follow ed in ord er to ob ta in the nec essa ry
d a ta .
3.8.1
Da ta Collection Procedures
The a im w a s to c ollec t a t lea st five sets of d a ta on ea c h infa nt. The five
sets of d a ta from ea c h sub jec t inc lud ed the follow ing :
•
Una id ed ABR to c lic k a nd 500 Hz tone b urst stim uli
•
Una id ed ASSR to 4 freq uenc ies p er ea r
89
University of Pretoria etd – Stroeb
bel, D (2006
•
Aid ed ASSR to 4 freq uenc ies p er ea r
•
Una id ed p ure tone b eha viora l threshold s
•
Aid ed b eha viora l threshold s
Da ta c ollec tion w a s d one b y a q ua lified a ud iolog ist, reg istered w ith the
Hea lth Professions Counc il of South Afric a , w ith 14 yea rs of exp erienc e in
the
field
of p ed ia tric
a ud iolog y. The
d a ta
c ollec tion p roc ed ures
c onc erning the d ifferent typ es of d a ta w ill b e d isc ussed a c c ord ing to the
synop sis p resented in Fig ure 3.1.
Dia gnosis
ABR vs. ASSR
(3 – 6 m onths)
Comp a re
Beha viora l
Threshold s
(8 – 12 m onths)
Hea ring Aid Fitting
Within a month a fter
elec trop hysiolog ic a l eva lua tion
Va lida tion of Hea ring Aid fitting
Aid ed ASSR vs.
Una id ed ASSR
(3 – 6 m onths)
Figure 3.1
Comp a re
Aid ed Beha viora l
Threshold s
(8 – 12 m onths)
Schema tic representa tions of the da ta collection procedures
Sub jec ts und er the a g e of six m onths of a g e w ere nursed b y a p a rent a nd
w ere tested w hilst in a na tura l sleep . If sed a tion w a s need ed , sub jec ts
w ere sed a ted w ith c hlora l hyd ra te – p resc rib ed b y a p ed ia tric ia n
(50m g / kg , a d m inistered ora lly) a nd w ere m onitored for oxyg en sa tura tion,
resp ira tory ra te, a nd hea rt ra te throug hout the p roc ed ure b y a p ed ia tric
nurse.
90
University of Pretoria etd – Stroeb
bel, D (2006
3.8.1.1
Aud itory Bra instem Resp onse (ABR)
ABR testing w a s c om p leted a t referra l, a s show n in Fig ure 3.1. Initia lly,
c lic k-evoked ABR threshold s w ere rec ord ed b ila tera lly (Gorg a 1999:36).
ABR threshold s w ere then rec ord ed using 500Hz, 1000Hz a nd 2000Hz
toneb urst stim uli (Va nd er Werff et a l., 2002:230). At ea c h p resenta tion
level, a m inim um of 1200 sw eep s w ere a vera g ed . Inc rem ents of 10 d B
w ere used for sup ra threshold p resenta tions. Inc rem ents w ere red uc ed to 5
d B nea r threshold , a nd a m inim um of tw o rep lic a tions w ere rec ord ed a t
stim ula tion levels nea r threshold (Ra nc e & Ric ka rd s, 2002:238). The
threshold w a s d efined a s the low est level tha t resulted in a rep lic a b le ABR
w a ve V (Va nd er Werff et a l., 2002: 230; Cone-Wesson et a l., 2002:177).
3.8.1.2
Aud itory Stea d y Sta te Resp onse (ASSR)
The sa m e elec trod es used for the ABR testing w ere used for ASSR testing .
ASSR testing b eg a n a fter the ABR testing w a s c om p leted (see Fig ure 3.1).
ASSR testing w a s c ond uc ted a t 2000 Hz a nd 500 Hz in b oth ea rs. If tim e
p erm itted a nd the infa nt w a s still a sleep , ASSR testing a t 1000 Hz a nd 4000
Hz follow ed (Va nd er Werff et a l., 2002:228). Threshold s w ere ob ta ined
using a 10 d B d ow n a nd 5 d B up sea rc h p roc ed ure w ith a sta rting level of
50 d B. Threshold w a s d efined a s the m inim um level a t w hic h the p ha se
c oherenc e w a s sta tistic a lly sig nific a nt (Ra nc e & Ric ka rd s, 2002:238; ConeWesson et a l., 2002:178). When no ASSR c ould b e id entified a t m a xim um
p resenta tion levels, the run w a s rep ea ted (Va nd er Werff et a l., 2002:231).
The ASSR m ea sured threshold s w ere then used to estim a te the p ure-tone
a ud iog ra m . This estim a tion utilized a n a lg orithm b a sed on p ub lished
resea rc h from the University of Melb ourne in w hic h ASSR threshold s
m ea sured for p a tients w ith va rious a m ounts of hea ring loss w ere
91
University of Pretoria etd – Stroeb
bel, D (2006
c orrela ted w ith their b eha viora l a ud iog ra m s (Ra nc e et a l., 1995:499). This
p resent stud y refers to these estim a tions a s ‘ ASSR p red ic ted threshold s’ .
3.8.1.3
Aid ed ASSR threshold s
These tests c om m enc ed a p p roxim a tely a m onth a fter hea ring a id fitting
(Fig ure 3.1). Hea ring a id s w ere p rog ra m m ed a c c ord ing to ea c h infa nt’ s
hea ring loss – using p resc rip tive m ethod s. Resp onses w ere m ea sured a t
freq uenc ies of 500 Hz, 1000 Hz, 2000 Hz a nd 4000 Hz throug h the free field
sp ea ker (Pic ton et a l., 2002:69). A 10 d B d ow n, 5 d B up sea rc h p roc ed ure
w a s a g a in utilized . Sta rting levels w ere c om m enc ed a t 50 d B. Threshold for
func tiona l g a in p red ic tion w a s d efined a s the m inim um level a t w hic h the
p ha se c oherenc e w a s sta tistic a l sig nific a nt (Ra nc e & Ric ka rd s, 2002:238).
Dep end ing on the tec hnolog y of the hea ring a id s, c erta in fea tures of the
hea ring a id s need ed to b e d ea c tiva ted , suc h a s noise red uc tion system s
a nd feed b a c k m a na g em ent system s (Kuk, 2004:1).
‘ ASSR m ea sured threshold ’ for ea c h freq uenc y w a s d efined a s the low est
HL a t w hic h a tria l w a s jud g ed to b e a ‘ resp onse’ . ‘ ASSR p red ic ted
threshold ’ for ea c h freq uenc y referred to the estim a ted b eha viora l
a ud iog ra m – using the University of Melb ourne a lg orithm (Ra nc e et a l.,
1995:499). Both these m ea surem ents w ere used d uring the c om p a rison
b etw een a id ed ASSR’ s a nd a id ed b eha viora l threshold s a s the norm a tive
d a ta from w hic h p red ic ted threshold s w ere c a lc ula ted w ere c om p iled b y
d a ta not d erived for a id ed ASSR’ s.
3.8.1.4
Una id ed b eha viora l p ure tone threshold s (BT)
This eva lua tion w a s c ond uc ted a t the a g e a t w hic h the infa nts w ere
m a ture enoug h to c om p lete the a ud iom etric testing . There w a s a tim e
92
University of Pretoria etd – Stroeb
bel, D (2006
d ela y of four to six m onths b etw een the tim es of the evoked p otentia l a nd
a ud iom etric testing (see Fig ure 3.1). Infa nts w ho w ere 3 – 6 m onths of a g e
a t the tim e of elec trop hysiolog ic a l testing w ere a t lea st 7 – 9 m onths old a t
the tim e of the b eha viora l testing a nd w ere therefore m a ture enoug h to
c om p lete the b eha viora l a ssessm ent.
The follow ing Visua l Resp onse Aud iom etry Protoc ol w a s follow ed (Gra vel,
2000:39):
Na rrow b a nd s of noise w ere used a s test stim uli. Assessm ent b eg a n w ith
sound field p resenta tions of the stim ulus (500 Hz) a t 30 d B HL. If the sub jec t
oriented tow a rd the loud sp ea ker, the hea d turn resp onse w a s reinforc ed
a nd a nother stim ulus a t the sa m e level w a s p resented . A hea d turn w a s
a g a in reinforc ed a nd the threshold sea rc h (d esc end ing ) w a s initia ted . If
no resp onse oc c urred , a fter tw o p resenta tions a t 30 d B HL, sig na l level w a s
inc rea sed in 20 d B step sizes until a n orienta tion tow a rd s the loud sp ea ker
oc c urred . Tw o resp onses a t the sa m e level w a s the sta rting level for
threshold sea rc h. The initia l d esc ent step size for threshold sea rc h w a s 10
d B a nd rem a ined 10 d B for the up -d ow n threshold sea rc h p roc ed ure.
Sound field threshold s w ere ob ta ined a c ross the freq uenc y ra ng e from
500 Hz throug h 4000 Hz. Test ord er w a s 500 Hz, follow ed b y 2000 Hz, 4000
Hz, a nd then 1000 Hz. Threshold w a s c a lc ula ted from the levels of the
three resp onse reversa ls follow ing the first m iss on the initia l d esc ent.
In c a ses w here there w a s no resp onse to sound field stim uli a t 80 d B HL, a
b one-c ond uc ted sig na l(na rrow b a nd noise a t 250 or 500 Hz tha t w a s
intense enoug h to b e felt) w a s used to tea c h the sub jec t the hea d -turn
resp onse.
93
University of Pretoria etd – Stroeb
bel, D (2006
Insert ea rp hones w ere used to d eterm ine ea r sp ec ific threshold s – using
the sa m e ord er of test freq uenc ies a s used in sound field p resenta tions.
Insert ea rp hones w ere p referred a s they a re lig htw eig ht, d o not inhib it the
hea d -turn resp onse a nd p rovid e g ood intera ura l a ttenua tion in the c a ses
of a sym m etric a l hea ring loss. For threshold sea rc h und er these c ond itions,
the step size w a s red uc ed to 5 d B. Threshold sea rc h w a s id entic a l to tha t
d esc rib ed for sound field a ssessm ent.
3.8.1.5
Aid ed b eha viora l threshold s
Testing w a s c ond uc ted w ith ea c h sub jec t’ s ow n hea ring a id s. Freq uenc ies
of 500 Hz, 1000 Hz, 2000 Hz a nd 4000 Hz w ere tested – using b roa d b a nd
sig na ls w ith the sp ea kers p ositioned a t 0º. Assessm ent of a id ed resp onses
follow ed the sa m e p roc ed ure a s d esc rib ed a b ove w ith sound field
p resenta tions a s d isc ussed in 3.8.1.4. (Also see Fig ure 3.1).
3.8.2
Procedures for da ta recording, processing a nd a na lysis
The follow ing p roc ed ures w ere follow ed to rec ord , p roc ess a nd a na lyze
the d a ta .
3.8.2.1
Rec ord ing of d a ta
The follow ing inform a tion w a s rec ord ed for ea c h sub jec t on a d a ta sheet
(Ap p end ix C).
Da ta yield ed from ea c h sub jec t inc lud ed :
•
Una id ed ABR c lic k threshold (d eterm ined a t tim e of d ia g nosis)
•
Una id ed ABR 500 Hz tone b urst threshold (d eterm ined a t tim e of
d ia g nosis)
94
University of Pretoria etd – Stroeb
bel, D (2006
•
Una id ed ASSR threshold for a t lea st 2000 Hz a nd 500 Hz (d eterm ined
a t tim e of d ia g nosis)
•
Aid ed ASSR threshold s (d eterm ined w ithin a m onth a fter d ia g nosis)
•
Una id ed Beha viora l Threshold s (d eterm ined 4 – 6 m onths a fter
d ia g nosis)
•
Aid ed Beha viora l Threshold s (d eterm ined 4 – 6 m onths a fter
d ia g nosis)
3.8.2.2
Proc ed ures for p roc essing a nd a na lysis of d a ta
“ Sta tistic s a re a m ong the m ost p ow erful tools in the resea rc her’ s toolb ox” ,
(Leed y & Orm rod , 2001:252). These tools inc lud e d esc rip tive a nd inferentia l
sta tistic s. Descriptive sta tistics enta ils ord ering a nd sum m a rizing the d a ta
b y m ea ns of ta b ula tion a nd g ra p hic rep resenta tion a nd the c a lc ula tion
of d esc rip tive m ea sures. In this w a y the inherent trend s a nd p rop erties of
the ob served d a ta em erg e c lea rly (Steyn, Sm it, Du Toit & Stra sheim ,
2003:5). Inferentia l sta tistics on the other ha nd serve a d ifferent p urp ose. It
d ra w s c onc lusions a b out the p op ula tion from w hic h the sa m p le w a s
d ra w n b y c om p a ring d esc rip tive m ea sures tha t ha ve b een c a lc ula ted .
(Leed y & Orm rod , 2005:252). In this stud y the op tions a va ila b le for relia b le
a nd va lid a na lysis w a s to a c erta in extent lim ited b y the rela tively sm a ll
sa m p le. A sta tistic ia n a t the Dep a rtm ent of Sta tistic s a nd Ac tua ria l
Sc ienc e of the University of Stellenb osc h w a s c onsulted in his p riva te
c a p a c ity d uring the p la nning of this stud y.
In ord er to d eterm ine the c linic a l va lue of the ASSR m ethod in infa nts,
ea c h
sub jec t’ s ind ivid ua l
p erform a nc e
w a s d esc rib ed
on
ea c h
p roc ed ure. The results ob ta ined d uring the una id ed ASSR eva lua tion w ere
c om p a red w ith the una id ed ABR results a t the tim e of d ia g nosis a nd
95
University of Pretoria etd – Stroeb
bel, D (2006
sub seq uently w ith the results ob ta ined d uring the una id ed b eha viora l
a ssessm ent. The sa m e p roc ed ure w a s follow ed w ith the a id ed ASSR a nd
a id ed b eha viora l a ssessm ent results. Therea fter the c ollec tive results for a ll
sub jec ts w ere a na lyzed . The foc us w a s not on c om p a ring the d a ta of
d ifferent sub jec ts a s suc h, b ut to c om p a re the d a ta for the d ifferent ea rs
a s it w a s rec ord ed throug h the use of three d ifferent m ea suring
tec hniq ues b a sed on a p a rtic ula r stim ulus. In this w a y the d a ta c ollec ted
throug h the use of the ASSR c ould b e c om p a red to the d a ta c ollec ted
throug h the use of other m ea suring tec hniq ues. The c ollec ted d a ta w ere
ta b ula ted in ra w d a ta ta b les a nd g ra p hic a lly c om p a red in fig ures
resem b ling a n a ud iog ra m .
In ord er to g et a b roa d er p ersp ec tive on the resea rc h find ing s, the d a ta
for a ll sub jec ts c ollec tively w a s further a na lyzed using descriptive sta tistics.
In these a na lyses three a sp ec ts w ere ta ken into a c c ount, na m ely: p oints
of c entra l tend enc y (m ea n); extent of d isp ersion (ra ng e/ sta nd a rd
d evia tion); a nd the extent to w hic h d ifferent va ria b les w ere rela ted to
one a nother (c orrela tion) (Leed y & Orm rod , 2005:257; Drum m ond ,
2003:112). Da ta p roc essing a nd a na lysis of the c ollec tive d a ta enta iled
the follow ing :
•
Ca lc ula ting the ra nge of threshold va lues for the 12 ea rs p er
stim ulus freq uenc y a s rec ord ed b y ea c h m ea suring tec hniq ue.
•
Ca lc ula ting
the
sta nda rd
devia tion (SD)
for the
m ea sured
threshold s for ea c h stim ulus freq uenc y in a d d ition to the ra ng e.
How ever, the d ec ision to inc lud e the SD w a s ta ken w ith full
know led g e tha t the lim ited num b er of d a ta p oints inc rea ses the
p ossib ility tha t the SD m a y b e und uly influenc ed b y a sing le d a ta
96
University of Pretoria etd – Stroeb
bel, D (2006
p oint. With this in m ind , the SD d a ta for resp onses to a sing le
stim ulus freq uenc y w a s c onsid ered w ith extrem e c a re.
•
Determ ining the extent of the difference b etw een threshold s
rec ord ed for a p a rtic ula r stim ulus freq uenc y b y c onsid ering the
num b er of threshold s tha t fa lls w ithin 10 d B, 15 d B, 20 d B a nd m ore
tha n 20 d B from ea c h other. This c a n b e seen a s a c a teg oric a l
c om p a rison of d ifferenc es.
•
Determ ining the mea n of the threshold va lues for the 12 ea rs p er
stim ulus freq uenc y, rec ord ed b y ea c h m ea surem ent tec hniq ue.
•
Ca lc ula ting the difference b etw een the m ea n of threshold s (of 12
ea rs p er stim ulus freq uenc y) rec ord ed b y tw o sp ec ific m ea suring
tec hniq ues.
•
Esta b lishing the sta tistica l significa nce of the d ifferenc es b etw een
the m ea n of threshold s for 12 ea rs p er stim ulus freq uenc y b y using
inferentia l sta tistic s. Tw o-sa m p le c om p a risons b etw een the una id ed
ASSR results vs. the una id ed ABR results; the una id ed ASSR results vs.
the una id ed b eha viora l results; a nd the a id ed ASSR results vs. the
a id ed b eha viora l results w ere d one b y a p p lying the Wilc oxon
Sig ned -Ra nk Test (Steyn et a l., 1991:594). This test is va lid w ith the
a ssum p tion tha t the sa m p les w ere d ra w n ind ep end ently from tw o
sets of d a ta w ith d istrib utions of sim ila r sha p e (Steyn et a l.,
1994:594). This test is p ow erful b ec a use it uses the size (m a g nitud e)
of d ifferenc es a s w ell a s the d irec tion (p ositive or neg a tive). The size
of the d ifferenc es is ind ic a ted b y ra nking the d ifferenc es for the
c om b ined sc ores (Drum m ond , 1998:128). The Wilc oxon Sig ned Ra nk test
d eterm ines a
p -va lue, d eterm ining
the
sta tistic a l
sig nific a nc e of the d ifferenc e b etw een tw o d a ta sets. For tw o sets
of d a ta to ha ve a sta tistic a l sig nific a nt d ifferenc e, the p -va lue
97
University of Pretoria etd – Stroeb
bel, D (2006
should b e sm a ller tha n 0.05. The d a ta c onc erning this sp ec ific
a na lysis a re rep resented in the form of ta b les show ing the p -va lue.
•
In a n effort to b roa d en the sc op e of the sta tistic a l d esc rip tion of
results, the m ea n a nd the SD for threshold s for a ll m ea surem ents
(inc lud ing m ea surem ents for a ll stim ulus freq uenc ies a nd a ll ea rs),
eva lua ted w ith a p a rtic ula r p roc ed ure, w ere c a lc ula ted . The
m otiva tion for d oing so w a s to inc lud e a la rg er num b er of d a ta
p oints a nd sub seq uently m inim izing the effec t tha t one sing le d a ta
p oint m a y ha ve on the SD va lue. This a p p roa c h how ever is a lso not
w ithout its p rob lem s, sinc e one p roc ed ure m a y p rove to b e m ore
sensitive to hig her or low er stim ulus freq uenc ies a nd intensities tha n
a nother, resulting in a c a nc eling effec t. These SD va lues w ere
therefore a lso interp reted w ith extrem e c a ution.
•
In the fina l insta nc e, d eterm ining the correla tion coefficient of
threshold va lues (p er ea r, p er stim ulus freq uenc y) p rovid ed b y tw o
sp ec ific
m ea suring
tec hniq ues. The
c orrela tion
va lues w ere
interp reted a c c ord ing to c a teg oric a l g uid elines p rovid ed b y
Koenker (in Leed y, 1981:115).
3.8.3 Va lidity a nd Relia bility
The va lidity of a m ea surem ent is the extent to w hic h the instrum ent
m ea sures w ha t it is sup p osed to m ea sure (Leed y & Orm rod , 2005:31).
Resea rc h litera ture revea ls m a ny va lid a tion p roc ed ures (Ba iley, 1982:69;
Ba b b ie, 1992:132):
•
Interna l va lid ity a sks w hether a d ifferenc e exists a t a ll in a ny g iven
c om p a rison (Ba iley, 1982:72). In this p resent stud y it w ould a sk
w hether or not a n a p p a rent d ifferenc e b etw een results c a n b e
98
University of Pretoria etd – Stroeb
bel, D (2006
exp la ined a s m ea surem ent a rtifa c ts. It w a s therefore im p orta nt to
use the sa m e test p rotoc ol w ith the eq uip m ent set up in the
p resc rib ed m a nner w ith ea c h ind ivid ua l sub jec t.
•
Externa l va lid ity is the p rob lem of interp reta tion (Ba iley, 1982:73). In
this p resent stud y this a sp ec t w ould refer to the interp reta tion of the
d ifferent test results a s ob ta ined throug h d ifferent m ea surem ents.
This a sp ec t w a s a d d ressed b y the fa c t tha t the resea rc her w a s
c om p etent to c a rry out the d ifferent p roc ed ures d ue to her
q ua lific a tion a nd yea rs of exp erienc e.
Relia bility of a m ea sure is sim p ly its c onsistenc y (Ba b b ie, 1992:135;
Drum m ond , 2003:79). The resea rc her used d ifferent tec hniq ues to
m ea sure the sa m e c onc ep t – in this c a se una id ed a nd a id ed
threshold s m ea sures. These tec hniq ues w ere a d m inistered to the sa m e
sub jec ts, using the sa m e test p rotoc ol a nd test environm ent in ea c h
ind ivid ua l c a se.
3.9
SUMMARY
This c ha p ter p rovid ed a c om p rehensive d esc rip tion of the p roc ed ures
im p lem ented in the resea rc h m ethod olog y to ob ta in the d a ta a c c ord ing
to the sub -a im s of this stud y. This w a s d one in ord er to a c hieve the m a in
a im of the stud y. The need for c linic a l va lid a tion of the ASSR in the
p ed ia tric p op ula tion w a s the m otiva tion for this stud y. The exp erim enta l
d esig n w a s d esc rib ed , follow ed b y the d isc ussion of the sub jec ts in term s
of selec tion c riteria , p roc ed ures involved in selec tion a nd a p p a ra tus used
for selec ting sub jec ts. Sub seq uently a d esc rip tion w a s p rovid ed of the
sub jec ts. The m a teria l a nd a p p a ra tus used for the c ollec tion of d a ta a nd
the a na lysis thereof a s w ell a s the p roc ed ures for d a ta a na lysis w ere
99
University of Pretoria etd – Stroeb
bel, D (2006
d isc ussed , follow ed
by a
d esc rip tion of the p roc ed ures for d a ta
p roc essing a nd a na lysis. The c ha p ter c onc lud ed w ith a review of va lid ity
a nd relia b ility a s it rela te to the c urrent stud y.
100
University of Pretoria etd – Stroeb
bel, D (2006
Cha p ter 4
RESULTS AND DISCUSSION
This c ha p ter a im s to p resent the results of the em p iric a l resea rc h a nd
eluc id a tes the m ea ning a nd sig nific a nc e thereof w ithin the c urrent b od y
of know led g e
4.1
INTRODUCTION
The d a w n of a n era of ea rly id entific a tion of hea ring loss in new b orns a nd
infa nts p oses new c ha lleng es a nd offers new op p ortunities to a ud iolog ists.
With the a d vent of universa l new b orn hea ring sc reening , it is c om m on for
a n a ud iolog ist to see infa nts less tha n tw o to three m onths of a g e w ho,
d uring the new b orn p eriod , ha ve b een id entified a s b eing a t risk for
hea ring loss. It is therefore essentia l to find evid enc e in ord er to esta b lish a
p rotoc ol tha t w ould yield the m ost inform a tion w ith reg a rd to resid ua l
hea ring a b ilities in this p op ula tion. Resea rc h, a s initia ted in this stud y, is
essentia l to the im p lem enta tion of a p p rop ria te d ia g nostic p rotoc ols in this
sp ec ific p op ula tion.
The m ethod olog ic a l a p p roa c h, sp ec ified in c ha p ter 3 ha s p rovid ed the
op era tiona l fra m ew ork for extra c ting the nec essa ry d a ta for a d d ressing
the m a in a im of this stud y. The m a in a im of this stud y, to esta b lish the
c linic a l va lue of the ASSR for ea rly d ia g nosis a nd a m p lific a tion of infa nts
101
University of Pretoria etd – Stroeb
bel, D (2006
w ith hea ring loss, w a s a d d ressed throug h the rea liza tion of tw o sub -a im s.
These a im s a re sc hem a tic a lly sum m a rized in Fig ure 4.1.
Ma in Aim
To esta b lish the c linic a l va lue of
the ASSR for ea rly d ia g nosis a nd
a m p lific a tion of infa nts w ith
hea ring loss
Sub Aim 1
Sub Aim 2
To investig a te the c linic a l va lue of
the ASSR in ea rly d ia g nosis of
hea ring loss in a g roup of infa nts b y
d eterm ining a nd c om p a ring
una id ed ASSR, ABR a nd b eha viora l
threshold s.
To investig a te the c linic a l va lue of
the ASSR for releva nt ea rly fitting
of hea ring a id s in infa nts b y
d eterm ining a nd c om p a ring
a id ed ASSR a nd a id ed b eha viora l
threshold s
Figure 4.1
Ma in-a im a nd sub-a ims of study
Ana lyzed results for the c urrent stud y a re g roup ed , rep orted , interp reted
a nd sub seq uently d isc ussed in rela tion to releva nt a nd c om p a ra b le
litera ture. The first sub -a im w a s a c hieved b y d eterm ining a nd c om p a ring
the una id ed ASSR a nd ABR threshold s a t the tim e of d ia g nosis a t a young
a g e (3-6 m onths of a g e). These ASSR a nd ABR threshold s w ere then
c om p a red w ith una id ed b eha viora l threshold s ob ta ined a t a la ter
d evelop m enta l a g e, w hen sub jec ts w ere a b le to p rovid e relia b le
b eha viora l resp onses (8 – 14 m onths of a g e).
The sec ond sub -a im w a s a d d ressed b y d eterm ining a id ed ASSR threshold s
w ithin a m onth a fter d ia g nosis of hea ring loss a nd a fter ea c h sub jec t w a s
fitted w ith hea ring a id s. The a id ed ASSR threshold s w ere then c om p a red
102
University of Pretoria etd – Stroeb
bel, D (2006
w ith
a id ed
b eha viora l threshold s w hen
the
sub jec ts rea c hed
a
d evelop m enta l a g e a llow ing relia b le b eha viora l resp onses to b e elic ited .
The results a re p resented a nd d esc rib ed a c c ord ing to ea c h of the sub a im s. The results from ea c h ind ivid ua l sub jec t a re initia lly c onsid ered ,
follow ed b y a c ollec tive a na lysis of the results for the six sub jec ts. In the
sec ond p a rt of this c ha p ter, a d isc ussion of results a long sid e c urrent
litera ture w ill follow . In the fina l sec tion of this c ha p ter, g enera l c onc lusions
from the stud y a re d ra w n a nd the m a in resea rc h q uestion is a nsw ered .
In ord er to d eterm ine the clinica l va lue of the ASSR method in ea rly
dia gnosis of a
hea ring
loss in infa nts, ea c h
sub jec t’ s ind ivid ua l
p erform a nc e w ill b e d esc rib ed on ea c h eva lua tion p roc ed ure. The results
ob ta ined d uring the una id ed ASSR eva lua tion w ill b e c om p a red w ith the
una id ed ABR results a t the tim e of d ia g nosis a nd sub seq uently b oth these
p roc ed ures w ill b e c om p a red w ith the una id ed b eha viora l a ssessm ent
results ob ta ined . Follow ing p resenta tions of ea c h ind ivid ua l c a se, the
results for the six sub jec ts c ollec tively w ill b e c onsid ered . In the c ollec tive
a na lysis of the d a ta , the foc us w ill b e on a c om p a rison of the threshold
d a ta for a ll 12 ea rs (of the six sub jec ts) a s it w a s rec ord ed throug h the use
of three d ifferent m ea suring tec hniq ues. The d esc rip tive a nd inferentia l
sta tistic s from the g roup w ill b e rep orted .
In ord er to d eterm ine the clinica l va lue of the ASSR in the va lida tion of
hea ring a id fitting, the sec ond p a rt of the results w ill p resent ea c h
sub jec t’ s ind ivid ua l p erform a nc e on the a id ed ASSR – c om p a ring una id ed
ASSR va lues w ith the a id ed ASSR va lues a nd sub seq uently w ith results
ob ta ined d uring a id ed b eha viora l a ssessm ent. Therea fter the results of a ll
six sub jec ts w ill b e a na lyzed c ollec tively, a s it w a s rec ord ed throug h the
103
University of Pretoria etd – Stroeb
bel, D (2006
use of these tw o m ea suring tec hniq ues. The d esc rip tive a nd inferentia l
sta tistic s from this g roup of six sub jec ts w ill b e p resented .
4.2.
RESULTS FOR SUB-AIM 1: TO INVESTIGATE THE POTENTIAL CLINICAL
VALUE OF THE ASSR IN EARLY DIAGNOSIS OF HEARING LOSS IN A
GROUP OF INFANTS BY DETERMINING AND COMPARING UNAIDED
ASSR, ABR AND BEHAVIORAL THRESHOLDS.
Bila tera lly c lic k-evoked ABR resp onses w ere rec ord ed first. Therea fter the
tone b urst ABR a ssessm ent w a s c a rried out, follow ed b y the ASSR
a ssessm ent. Beha viora l threshold s w ere ob ta ined from ea c h sub jec t a t the
d evelop m enta l a g e
w hen
they
c ould
rend er relia b le
b eha viora l
resp onses. The results for ea c h ind ivid ua l sub jec t a re d esc rib ed in the
follow ing sec tion.
4.2.1 Individua l subject results for sub-a im 1
In ord er to a id the interp reta tion of the ind ivid ua l results, a short sum m a ry
of ea c h sub jec t’ s b a c kg round inform a tion is a d d ed to the una id ed ABR,
ASSR a nd b eha viora l a ssessm ent results sum m a rized in ta b le form a t (see
Ta b les 4.1 – 4.6).
4.2.1.1
Sub jec t 1:
Results for sub -a im 1
The b a c kg round inform a tion a nd test results for sub jec t 1 a re p resented in
Ta b le 4.1 a nd Fig ure 4.2.
104
University of Pretoria etd – Stroeb
bel, D (2006
Ta ble 4.1
Ba ckground informa tion a nd test results for subject 1
Sex
Male
Risk factors
Born at 34 weeks gestation age.
Diagnosed with cytomegalovirus
Age at time of hearing loss identification
3 months
Degree of hearing loss
Moderately severe sensory neural hearing loss in the
right ear.
No response could be measured at maximum
intensities of equipment in the left ear.
Age at time of hearing aid fitting
4 months
Type of hearing aid
Digital hearing aid on right ear
Age at time of behavioral assessment
10 months
ABR results
Tone burst
Click
R = 50 dBnHL
R = 65 dBnHL
L = NR
L = NR
ASSR predicted results
Behavioral assessment results
500 Hz
1000 Hz
2000 Hz
4000 Hz
R = 55 dB
R = 55 dB
R = 65 dB
R = 70 dB
L = NR
L = NR
L = NR
L = NR
500 Hz
1000 Hz
2000 Hz
4000 Hz
R = 50 dB
R = 55 dB
R = 65 dB
R = 75 dB
L = NR
L = NR
L = NR
L = NR
NR = No Response
Right Ea r
Frequency (kHz)
dB HL
0.5
0
10
20
30
40
50
60
70
80
90
100
110
120
1
2
4
BTH
ASSR
ABR
105
University of Pretoria etd – Stroeb
bel, D (2006
Figure 4.2
Schema tic representa tions of the ABR, ASSR predictions a nd
BT results for subject 1
For this ind ivid ua l c a se the tone b urst ABR w a s 5 d B low er tha n the 500Hz
ASSR p red ic ted threshold . The c lic k ABR yield ed the sa m e threshold a s the
2000 Hz ASSR p red ic ted threshold a nd there w a s only a 5 d B d ifferenc e
c om p a red to the 4000 Hz ASSR p red ic ted threshold , w ith the ABR ha ving
the low er va lue. In c om p a rison w ith the b eha viora l threshold s m ea sured
a t a la ter sta g e, the ASSR p red ic tion threshold s c losely follow ed the
c onfig ura tion of the b eha viora l threshold s – a d ifferenc e of only 5 d B a t
500 Hz a nd 4000 Hz w a s noted . Threshold s c orresp ond ed a t 1000 Hz a nd
2000 Hz on these tw o p roc ed ures. When c om p a ring the results from the
ABR w ith the b eha viora l threshold s, id entic a l threshold s w ere m ea sured
w ith the tone b urst ABR a nd a t 500 Hz. The c lic k ABR a nd 2000 Hz
b eha viora l resp onse yield ed the sa m e threshold s a nd a t 4000 Hz the
b eha viora l resp onse w a s 10 d B low er tha n the c lic k ABR.
No resp onse c ould b e m ea sured on a ny of the three m ea suring
tec hniq ues in the left ea r a t m a xim um intensity of the eq uip m ent.
4.2.1.2
Sub jec t 2:
Results for sub -a im 1
The b a c kg round inform a tion a nd test results for sub jec t 2 a re p resented in
Ta b le 4.2 a nd Fig ure 4.3.
Ta ble 4.2
Ba ckground informa tion a nd test results for subject 2
Sex
Female
Risk factors
Born
at
36
weeks
gestational
age
through
106
University of Pretoria etd – Stroeb
bel, D (2006
emergency caesarian;
Low birth weight;
Admitted to NICU.
Age at time of hearing loss identification
Five months
Degree of hearing loss
Moderately severe sensory neural loss in right ear;
Moderate sensory neural hearing loss in left ear
Age at time of hearing aid fitting
Five months
Type of hearing aid
Digital hearing aids binaurally
Age at time of behavioral assessment
14 months
ABR results
Tone burst
Click
R = 60 dBnHL
R = 75 dBnHL
L = 70 dBnHL
L = 60 dBnHL
500 Hz
1000 Hz
2000 Hz
4000 Hz
R = 60 dB
R = 50 dB
R = 65 dB
R = NR
L = 40 dB
L = 50 dB
L = 55 dB
L = 50 dB
500 Hz
1000 Hz
2000 Hz
4000 Hz
R = 70 dB
R = 65 dB
R = 80 dB
R = 95 dB
L = 50 dB
L = 60 dB
L = 75 dB
L = 95 dB
ASSR predicted results
Behavioral assessment results
NR = No Response
Right Ea r
Left Ear
Frequency (kHz)
1
2
4
0.5
0
10
0
10
20
30
20
30
40
50
60
40
50
60
70
80
BTH
ASSR
ABR
dB HL
dB HL
0.5
Frequency (kHz)
70
80
90
100
90
100
110
120
110
120
Figure 4.3
1
2
4
BTH
ASSR
ABR
Schema tic representa tions of the ABR, ASSR predictions a nd
BT results for subject 2
107
University of Pretoria etd – Stroeb
bel, D (2006
The ABR a nd ASSR ind ic a ted c om p a ra b le results in the rig ht ea r. On b oth
the tone b urst ABR a nd 500 Hz ASSR p red ic tion, a threshold estim a tion of
60 d B w a s m ea sured . A d ifferenc e of only 10 d B w ith the c lic k ABR
threshold a nd 2000 Hz ASSR p red ic ted threshold w a s noted . This sub jec t
w oke up b efore c om p leting the 4000 Hz ASSR in the rig ht ea r a nd
therefore no result is a va ila b le on tha t sp ec ific m ea surem ent. Beha viora l
resp onses w ere m ea sured a t 14 m onths of a g e. These threshold s w ere
eleva ted b y 10 to 15 d B a t the resp ec tive freq uenc ies for b oth the ASSR
a nd ABR m ea surem ents.
In the left ea r the tone b urst ABR threshold w a s 30 d B hig her tha n the 500
Hz ASSR p red ic ted threshold . A d ifferenc e of 5 to 10 d B w a s p resent
b etw een the ASSR p red ic ted threshold s for 2000 a nd 4000 Hz in
c om p a rison w ith the c lic k evoked ABR threshold . The ASSR ha d the low er
va lue. The b eha viora l threshold s yield ed resp onses w ith a d ifferenc e of 10
d B a t 500 Hz a nd 1000 Hz in c om p a rison w ith the ASSR p red ic ted
threshold s a t the sa m e freq uenc ies (the ASSR a g a in ha d the low er va lues).
The hig h freq uenc ies (2000 a nd 4000 Hz) show ed b ig d isc rep a nc ies
b etw een the ASSR p red ic ted threshold s a nd b eha viora l threshold s (± 20 to
45 d B) w ith the ASSR ha ving the low er va lues.
The tone b urst ABR
threshold s w ere 20 d B hig her tha n the 500 Hz b eha viora l threshold , b ut the
hig h freq uenc ies of 2000 Hz a nd 4000 Hz b eha viora l threshold s w ere 15 to
35 d B hig her tha n the c lic k ABR threshold s.
4.2.1.3
Sub jec t 3:
Results for sub -a im 1
The b a c kg round inform a tion a nd test results for sub jec t 3 a re p resented in
Ta b le 4.3 a nd Fig ure 4.4.
108
University of Pretoria etd – Stroeb
bel, D (2006
Ta ble 4.3
Ba ckground informa tion a nd test results for subject 3
Sex
Female
Risk factors
One of a twin, born at 32 weeks gestational age with
a family history of congenital deafness.
Age at time of hearing loss identification
Six months
Degree of hearing loss
Severe sensory neural hearing loss in right ear;
Profound sensory neural hearing loss in left ear.
Age at time of hearing aid fitting
Six months
Type of hearing aid
Digitally programmable analogue hearing aids
binaurally
Age at time of behavioral assessment
12 months
ABR results
Tone burst
Click
R = 90 dBnHL
R = 70 dBnHL
L = 90 dBnHL
L = 95 dBnHL
500 Hz
1000 Hz
2000 Hz
4000 Hz
R = 90 dB
R=105 dB
R = 95 dB
R = 80 dB
L = 95 dB
L = 95 dB
L = 85 dB
L = 80 dB
500 Hz
1000 Hz
2000 Hz
4000 Hz
R = 80 dB
R = 70 dB
R = 70 dB
R = 80 dB
L = 80 dB
L = 75 dB
L = 90 dB
L = 80 dB
ASSR predicted results
Behavioral assessment results
Right Ea r
Left Ear
Frequency (kHz)
1
2
4
0
10
20
30
40
50
60
70
80
90
100
110
120
Figure 4.4
0.5
BTH
ASSR
ABR
dB HL
dB HL
0.5
Frequency (kHz)
0
10
20
30
40
50
60
70
80
1
2
4
BTH
ASSR
ABR
90
100
110
120
Schema tic representa tions of the ABR, ASSR predictions a nd
BT results for subject 3
109
University of Pretoria etd – Stroeb
bel, D (2006
The rig ht ea r’ s resp onses c a n b e d esc rib ed a s follow s: resp onses in the
rig ht ea r for the tone b urst ABR a nd 500 Hz ASSR p red ic ted threshold s
yield ed the sa m e threshold . The c lic k ABR threshold w a s 25 d B low er tha n
the 2000 Hz ASSR p red ic ted threshold . The d ifferenc e b etw een the c lic k
ABR threshold a nd 4000 Hz ASSR p red ic ted threshold w a s 10 d B, w ith the
c lic k ABR ha ving the low er va lue. When c om p a ring the ASSR p red ic ted
threshold s a nd b eha viora l threshold s, a d ifferenc e of 10 d B w a s noted a t
500 Hz a nd a d ifferenc e of 35 d B a t 1000 Hz. The 2000 Hz c om p a rison
b etw een these tw o m ea surem ents show ed a 25 d B d ifferenc e. In a ll of
these insta nc es the ASSR p red ic ted threshold s ha d the hig her va lue. At
4000 Hz the threshold s b etw een these tw o m ea surem ents c orresp ond ed
w ell. A 10 d B d ifferenc e w a s noted b etw een the tone b urst ABR a nd 500
Hz b eha viora l threshold s, w ith the tone b urst ABR ha ving the hig her va lue.
The c lic k ABR yield ed the sa m e threshold a s the 2000 Hz b eha viora l
threshold . A 10 d B d ifferenc e w a s p resent b etw een the c lic k ABR a nd 4000
Hz b eha viora l threshold w ith the b eha viora l resp onse b eing the low er
va lue.
The threshold s from the left ea r c orresp ond ed b etter b etw een the
d ifferent m ea surem ents. A d ifferenc e of 5 d B w a s noted b etw een the
tone b urst ABR a nd 500 Hz ASSR p red ic ted threshold s w ith the ASSR ha ving
the hig her va lue. A sim ila r result w a s ob ta ined in the hig h freq uenc ies –
w ith a d ifferenc e of 10 d B b etw een the c lic k ABR a nd 2000 Hz ASSR
p red ic ted threshold s, a nd 15 d B d ifferenc e b etw een the c lic k ABR a nd
4000 Hz ASSR p red ic ted threshold s. In this insta nc e the c lic k ABR ha d the
hig her va lue. When c om p a ring the ASSR p red ic ted threshold s a nd
b eha viora l threshold s in the low freq uenc ies (500 Hz a nd 1000 Hz),
b eha viora l threshold s w ere 15 to 20 d B low er tha n the ASSR threshold s. The
110
University of Pretoria etd – Stroeb
bel, D (2006
ASSR p red ic ted threshold s yield ed 5 to 10 d B low er threshold s in the hig h
freq uenc ies of 2000 Hz a nd 4000 Hz. When c om p a ring the tone b urst ABR
threshold w ith the b eha viora l threshold , a 10 d B d ifferenc e is noted
b etw een these tw o m ea surem ent tec hniq ues – the tone b urst ABR b eing
the hig her va lue. The c om p a rison b etw een the c lic k ABR threshold a nd
b eha viora l threshold show s a 5 d B d ifferenc e w ith the 2000 Hz c om p a rison
a nd a 15 d B d ifferenc e w ith the 4000 Hz c om p a rison – in b oth c a ses the
ABR ha ving the hig her va lue.
4.2.1.4
Sub jec t 4:
Results for sub -a im 1
The b a c kg round inform a tion a nd test results for sub jec t 4 a re p resented in
Ta b le 4.4 a nd Fig ure 4.5.
111
University of Pretoria etd – Stroeb
bel, D (2006
Ta ble 4.4
Ba ckground informa tion a nd test results for subject 4
Sex
Female
Risk factors
Twin of subject 3, born at 32 weeks gestational age
with a family history of congenital deafness.
Age at time of hearing loss identification
Six months
Degree of hearing loss
Severe sensory neural hearing loss bilaterally
Age at time of hearing aid fitting
Six months
Type of hearing aid
Digitally programmable analogue hearing aids
binaurally
Age at time of behavioral assessment
12 months
ABR results
Tone burst
Click
R = 75 dBnHL
R = 75 dBnHL
L = 75 dBnHL
L = 75 dBnHL
500 Hz
1000 Hz
2000 Hz
4000 Hz
R = 90 dB
R = 85 dB
R = 85 dB
R = 80 dB
L = 80 dB
L = 85 dB
L = 85 dB
L = 70 dB
500 Hz
1000 Hz
2000 Hz
4000 Hz
R = 80 dB
R = 80 dB
R = 75 dB
R = 90 dB
L = 85 dB
L = 75 dB
L = 80 dB
L = 80 dB
ASSR predicted results
Behavioral assessment results
Right Ea r
Left Ear
Frequency (kHz)
1
2
4
0.5
0
10
0
10
20
30
40
50
60
20
30
40
50
60
BTH
ASSR
70
80
90
100
ABR
110
120
Figure 4.5
dB HL
dB HL
0.5
Frequency (kHz)
70
80
90
100
1
2
4
BTH
ASSR
ABR
110
120
Schema tic representa tions of the ABR, ASSR predictions a nd
BT results for subject 4
112
University of Pretoria etd – Stroeb
bel, D (2006
The results from this sub jec t show ed a g ood c om p a rison b etw een the
d ifferent p roc ed ures. The rig ht ea r show ed a d ifferenc e of 15 d B w hen
c om p a ring the tone b urst ABR threshold w ith the 500 Hz ASSR p red ic ted
threshold . In this c a se the ABR ha d the low er threshold . The c lic k ABR
threshold w a s 10 d B low er tha n the threshold for the 2000 Hz ASSR
p red ic ted threshold a nd 5 d B low er tha n the threshold for 4000 Hz ASSR
p red ic tion. The c om p a rison b etw een the ASSR p red ic ted threshold s a nd
b eha viora l threshold s show ed a n a vera g e d ifferenc e of 5 to 10 d B w ith
the ASSR ha ving the low er threshold a t a ll freq uenc ies exc ep t a t 4000 Hz.
The tone b urst ABR threshold w a s 5 d B low er tha n the 500 Hz b eha viora l
threshold . The c lic k ABR threshold ha d the sa m e va lue a s the 2000 Hz
b eha viora l threshold a nd w a s 5 d B low er tha n the 4000 Hz b eha viora l
threshold .
Sim ila r results w ere found in the left ea r. The tone b urst ABR threshold w a s 5
d B low er tha n threshold for the 500 Hz ASSR p red ic tion. The threshold for
the c lic k ABR w a s 10 d B low er tha n the threshold a t 2000 Hz on the ASSR
p red ic tion a nd 5 d B low er tha n the 4000 Hz threshold on the ASSR
p red ic tion. The ASSR p red ic ted threshold s d iffered w ith 5 to 10 d B from
those of the b eha viora l a ssessm ent a c ross the freq uenc y ra ng e. A 10 d B
d ifferenc e w a s p resent b etw een threshold s of the tone b urst ABR a nd the
500 Hz b eha viora l - w ith the tone b urst ABR ha ving the low er va lue. The
c lic k ABR threshold w a s 5 d B low er tha n the 2000 Hz a nd 4000 Hz
b eha viora l threshold s.
4.2.1.5
Sub jec t 5:
Results for sub -a im 1
113
University of Pretoria etd – Stroeb
bel, D (2006
The b a c kg round inform a tion a nd test results for sub jec t 5 a re p resented in
Ta b le 4.5 a nd Fig ure 4.6.
Ta ble 4.5
Ba ckground informa tion a nd test results for subject 5
Sex
Female
Risk factors
Born at 26 weeks gestational age; Admitted to
NICU for 2 months.
Age at time of hearing loss identification
Four months
Degree of hearing loss
Profound sensory neural hearing loss bilaterally
Age at time of hearing aid fitting
Five months
Type of hearing aid
High power digitally programmable analogue
hearing aids binaurally
Age at time of behavioral assessment
12 months
ABR results
Tone burst
Click
114
University of Pretoria etd – Stroeb
bel, D (2006
ASSR predicted results
Behavioral assessment results
R = NR
R = NR
L = NR
L = NR
500 Hz
1000 Hz
2000 Hz
4000 Hz
R=105 dB
R=115 dB
R=105 dB
R=100 dB
L=105 dB
L=105 dB
L=105 dB
L=110 dB
500 Hz
1000 Hz
2000 Hz
4000 Hz
R = 95 dB
R=105 dB
R=100 dB
R=100 dB
L = 95 dB
L=105 dB
L=110 dB
L=110 dB
NR = No Response
Right Ea r
Left Ear
Frequency (kHz)
1
2
4
0.5
0
10
0
10
20
30
20
30
40
50
60
40
50
60
BTH
ASSR
70
80
ABR
dB HL
dB HL
0.5
Frequency (kHz)
2
4
70
80
90
100
90
100
110
120
110
120
Figure 4.6
1
BTH
ASSR
ABR
Schema tic representa tions of the ABR, ASSR predictions a nd
BT results for subject 5
No resp onse c ould b e m ea sured on the ABR a t m a xim um outp ut (90
d BnHL) of the eq uip m ent on b oth the tone b urst ABR a nd the c lic k ABR.
The ASSR show ed resp onses a c ross the freq uenc y ra ng e of 500 Hz to 4000
Hz. Beha viora l resp onses w ere a lso m ea sured a t a ll the freq uenc ies.
When c om p a ring the results of the rig ht ea r b etw een the ASSR p red ic ted
threshold s a nd b eha viora l a ssessm ent, a 5 to 10 d B d ifferenc e w a s noted
115
University of Pretoria etd – Stroeb
bel, D (2006
a t 500, 1000 a nd 2000 Hz, w ith the b eha viora l threshold s b eing low er. The
threshold s a t 4000 Hz c orresp ond ed on these tw o p roc ed ures.
The left ea r ha d sim ila r results. A d ifferenc e of 10 d B w a s noted a t 500 Hz
b etw een the ASSR p red ic ted threshold s a nd b eha viora l threshold s - w ith
the b eha viora l threshold b eing low er. A d ifferenc e of 5 d B w a s p resent a t
2000 Hz w ith the ASSR p red ic ted threshold s b eing the low er va lue in this
insta nc e. The freq uenc ies of 1000 Hz a nd 4000 Hz yield ed the sa m e results
on these tw o m ea surem ents.
4.2.1.6
Sub jec t 6:
Results for sub -a im 1
The b a c kg round inform a tion a nd test results for sub jec t 6 a re p resented in
Ta b le 4.6 a nd Fig ure 4.7.
Ta ble 4.6
Ba ckground informa tion a nd test results for subject 6
Sex
Male
Risk factors
None
Age at time of hearing loss identification
Six months
Degree of hearing loss
Profound sensory neural hearing loss bilaterally
Age at time of hearing aid fitting
Six months
Type of hearing aid
High power digital hearing aids binaurally
Age at time of behavioral assessment
8 months
ABR results
Tone burst
Click
R = NR
R = NR
L = NR
L = NR
ASSR predicted results
Behavioral assessment results
500 Hz
1000 Hz
2000 Hz
4000 Hz
R=105 dB
R=105 dB
R=105 dB
R = NR
L=105 dB
L=115 dB
L=115 dB
L = NR
500 Hz
1000 Hz
2000 Hz
4000 Hz
R=105 dB
R=110 dB
R=105 dB
R = NR
116
University of Pretoria etd – Stroeb
bel, D (2006
L=105 dB
L=110 dB
L=110 dB
L = NR
NR = No Response
Right Ea r
Left Ea r
Frequency (kHz)
1
2
4
0.5
0
10
20
30
0
10
20
30
40
50
60
70
80
40
50
60
70
80
BTH
ASSR
ABR
dB HL
dB HL
0.5
Frequency (kHz)
90
100
90
100
110
120
110
120
Figure 4.7
1
2
4
BTH
ASSR
ABR
Schema tic representa tions of the ABR, ASSR predictions a nd
BT results for subject 6
No resp onses for sub jec t 6 c ould b e m ea sured on either the tone b urst
ABR or the c lic k ABR a t the m a xim um outp ut of the eq uip m ent (90 d BnHL).
Both the ASSR a nd b eha viora l m ea sures yield ed no resp onse a t 4000 Hz.
The threshold p red ic tion on the ASSR a nd the m ea sured b eha viora l
threshold s in the rig ht ea r d iffered w ith only 5 d B a t 1000 Hz - w ith the ASSR
ha ving the low er va lue. Results a t the other freq uenc ies yield ed the sa m e
threshold va lues.
Sim ila r results w ere ob ta ined in the left ea r, w ith a 5 d B d ifferenc e a t 1000
Hz a nd 2000 Hz b etw een the ASSR p red ic ted threshold s a nd b eha viora l
threshold s. In this c a se the b eha viora l threshold s w ere the low er levels.
117
University of Pretoria etd – Stroeb
bel, D (2006
In c onc lusion, w hen c onsid ering the results of the ind ivid ua ls, ind ic a tions
a re tha t the ASSR m a y p rove to b e a very useful a d d ition to the p ed ia tric
a ud iolog y test b a ttery – 80.5 % of the freq uenc ies p red ic ted b y the ASSR,
estim a ted b eha viora l threshold s w ithin 10 d B a s op p ose to the 57% of the
ABR. Yet, it is only w hen c onsid ering the results for a num b er of ind ivid ua ls
tha t a p a rtic ula r trend m a y b e id entified . In the follow ing sec tion the
results tha t c onc ern the ea rly d ia g nosis a s it is b a sed on m ea surem ents for
a ll six sub jec ts (12 ea rs), a re d esc rib ed , c om p a red a nd d isc ussed .
4.2.2 Collective results for a ll six subjects concerning sub-a im 1
The c ollec tive results for a ll six the sub jec ts c onc erning sub -a im 1 a re
sum m a rized in Ta b le 4.7. Foc using on the c ollec tive results for a ll ea rs
m ea sured , a further c om p a rison of the three eva lua tion p roc ed ures w ere
d one ta king into a c c ount the d isp ersion, the c entra l tend enc y a nd the
rela tion of the c ollec tive d a ta p rovid ed b y the d ifferent eva lua tion
p roc ed ures. The a b solute threshold m ea surem ents of ea c h ea r m ea sured
a nd the a rithm etic m ea n va lues for the num b er of ea rs m ea sured , p er
stim ulus freq uenc y, d eterm ined b y ea c h of the three p roc ed ures, a re a lso
inc lud ed in Ta b le 4.7, a s w ell a s the c a lc ula ted ra ng e a nd the sta nd a rd
d evia tion of the a b solute threshold va lues a nd the num b er of ea rs
m ea sured for a p a rtic ula r stim ulus freq uenc y. Ta b le 4.8 sum m a rizes the
m ea n of the resp onses to a ll stim ulus freq uenc ies p resented p er ea r, a s
rec ord ed b y the three d ifferent p roc ed ures, a s w ell a s the sta nd a rd
d evia tion, a nd the num b er of d a ta p oints used for this c a lc ula tion.
118
University of Pretoria etd – Stroeb
bel, D (2006
Ta ble 4.7
Summa ry of una ided thresholds for the six subjects a s determined by the ABR, ASSR a nd BT.
ABR
ASSR
Behavioral thresholds
Tone Burst
Click
500 Hz
1000 Hz
2000 Hz
4000 Hz
500 Hz
1000 Hz
2000 Hz
4000 Hz
R=65dBnHL
R=50dBnHL
R=55dB
R=55dB
R=65dB
R=70dB
R=50dB
R=55dB
R=65dB
R=75dB
L =NR
L=NR
L=NR
L=NR
L=NR
L=NR
L=NR
L=NR
L=NR
L=NR
R=60dBnHL
R=75dBnHL
R=60dB
R=50dB
R=65dB
R=NR
R=70dB
R=65dB
R=80dB
R=95dB
L=70dBnHL
L=60dBnHL
L=40dB
L=50dB
L=55dB
L=50dB
L=50dB
L=60dB
L=75dB
L=95dB
Subject 1
Subject 2
Subject 3
R=90dBnHl
R=70dBnHL
R=90dB
R=105dB
R=95dB
R=80dB
R=80dB
R=70dB
R=70dB
R=80dB
L=90dBnHL
L=95dBnHL
L=95dB
L=95dB
L=85dB
L=80dB
L=80dB
L=75dB
L=90dB
L=80dB
R=75dBnHL
R=75dBnHL
R=90dB
R=85dB
R=85dB
R=80dB
R=80dB
R=80dB
R=75dB
R=90dB
L=75dBnHL
L=75dBnHL
L=80dB
L=85dB
L=85dB
L=70dB
L=85dB
L=75dB
L=80dB
L=80dB
R=NR
R=NR
R=105dB
R=115dB
R=105dB
R=100db
R=95dB
R=105dB
R=100dB
R=100dB
L=NR
L=NR
L=105dB
L=105dB
L=105dB
L=110dB
L=95dB
L=105dB
L=110dB
L=110dB
R=NR
R=NR
R=105dB
R=110dB
R=105dB
R=NR
R=105dB
R=110dB
R=105dB
R=NR
L=NR
L=NR
L=105dB
L=115dB
L=115dB
L=NR
L=105dB
L=110dB
L=110dB
L=NR
Mean
75dBnHL
71.4dBnHL
84.6dB
87.3dB
87.7dB
87.5dB
81.4dB
82.7dB
87.3dB
87.8dB
Range
30
45
65
65
60
40
55
55
45
35
SD
11.5
14.06
23.07
24.73
19.54
16.69
18.99
20.90
16.49
11.76
7
7
11
11
11
8
11
11
11
9
Subject 4
Subject 5
Subject 6
Number of
ears
measured
NR = No Response
ASSR = ASSR predictions
SD = Sta nda rd Devia tion
120
University of Pretoria etd – Stroeb
bel, D (2006
Ta ble 4.8
Avera ge of a ll frequencies tested on the three procedures
ABR
ASSR
BT
predictions
Mea n
73.2
86.71
82.74
SD
12.5
20.78
18.75
21
41
42
Num b er of d a ta
p oints
Clic k ABR results w ere c om p leted on 12 ea rs (six sub jec ts). Of those 12
ea rs, five ha d no resp onse to c lic ks a t the m a xim um intensity lim it (90
d BnHL) of the eq uip m ent. Toneb urst ABR to 500 Hz w a s c om p leted on a ll
12 ea rs to w hic h five ha d no resp onse a t the lim its of the eq uip m ent (90
d BnHL).
ASSR’ s m ea surem ents w ere c om p leted on a ll 12 ea rs. Only one ea r ha d
no resp onse a t a ny freq uenc y of the ASSR exc ep t 4000 Hz, w here a nother
three ea rs ha d no resp onse a t the m a xim um intensity of the eq uip m ent.
The b eha viora l a ssessm ent show ed one ea r w ith no resp onse a t a ll
freq uenc ies on the b eha viora l testing . Another tw o ea rs ha d no resp onse
a t 4000 Hz. The results from a ll the sub jec ts a re show n in Ta b le 4.7.
4.2.2.1
Com p a ring the una id ed ABR a nd una id ed ASSR
As ind ic a ted in Ta b le 4.7, the ra nge of the a b solute m ea surem ents va lues
for the 500 Hz ASSR’ s is 35 d B b roa d er tha n the ra ng e for the tone b urst
ABR. The ra ng e for the 2000 Hz ASSR p red ic tions is 20 d B b roa d er tha n the
ra ng e for the c lic k ABR. The ra ng e of the 4000 Hz ASSR p red ic tions is
how ever 5 d B sm a ller tha n tha t of the c lic k ABR. The num b er of ea rs ta ken
into a c c ount is a g a in seven on the ABR, 11 on the 2000 Hz ASSR, b ut eig ht
121
University of Pretoria etd – Stroeb
bel, D (2006
on the 4000 Hz ASSR p red ic tion. Althoug h the ra ng e of the threshold tha t
w a s d eterm ined w ith the ASSR seem s in m ost c a ses b roa d er tha n the
ra ng e of m ea surem ents d eterm ined w ith the ABR, it w ould a lso seem a s if
the d ifferenc e b etw een the ra ng e m a y b e influenc ed b y the num b er of
ea rs m ea sured (See Ta b le 4.7). It is therefore d iffic ult to d ra w a ny
c onc lusions b a sed on the ra ng e of the threshold m ea sured w ith the ASSR
a nd ABR. The SD va lues seem to c onfirm the ra ng e va lues. It is how ever
risky to d ra w a ny c onc lusions from the SD d a ta sinc e the num b er of d a ta
p oints for the ABR m ea surem ents w ere lim ited to seven a nd c onsid ering
the inevita b le ind ivid ua l d ifferenc es, va ria tion in resp onses c a n b e
exp ec ted to b e hig h a nd w ould inevita b ly ha ve a n a ffec t on the SD
va lues for suc h a sm a ll sa m p le.
Com p a ring the ABR results for b oth tone b ursts a nd c lic ks w ith the 500 Hz,
2000 Hz a nd 4000 Hz ASSR, it w a s noted tha t the m a jority of c om p a ra b le
threshold s - 14 of the 21 (67%) – show ed a d ifferenc e of 10 d B or less. These
Compa ra tive frequency
thresholds
results a re sum m a rized in Fig ure 4.8.
16
14
12
10
8
6
4
2
0
” 10 d B
15 d B
20 d B
> 20 d B
Difference in dB
Figure 4.8
Representa tion of compa ra tive frequency thresholds between
the ABR a nd ASSR
122
University of Pretoria etd – Stroeb
bel, D (2006
Further insig ht w a s g a ined b y c a lc ula ting the mea n a nd c onsid ering the
d ifferenc e b etw een the m ea n of the a b solute va lues for ea c h of the
eva lua tion p roc ed ures. The m ea n of the una id ed c lic k ABR’ s ra ng ed from
71.4 d BnHL for the c lic k ABR a nd 75 d BnHL for the una id ed tone b urst ABR.
The m ea n ASSR p red ic ted threshold s levels ra ng ed from 84.6 d B HL to 87.7
d B HL. The m ea n of a ll the tone b urst ABR threshold s w ere 10 d B low er
tha n the m ea n of a ll the 500 Hz ASSR p red ic ted threshold s (see Ta b le 4.7).
When c om p a ring the c lic k ABR w ith the 2000 Hz ASSR a nd 4000 Hz ASSR,
the m ea n results ind ic a te tha t the ABR m ea surem ents a g a in ha d the
low er resp onse level, w ith a d ifferenc e of a p p roxim a tely 17 d B. It is
releva nt thoug h, to ta ke into a c c ount the num b er of ea rs tested w ith
ea c h p roc ed ure (Ta b le 4.7). Only seven ea rs rep resent the results on the
ABR, w here 11 a re rep resented on the ASSR a vera g e. The five ea rs not
rep resented on the ABR results a re the ones tha t ha d no resp onse on this
p roc ed ure a nd therefore fa ll in a c a teg ory m ore severe tha n c ould b e
m ea sured b y the ABR, thus infla ting the c a lc ula ted m ea n of the ASSR
m ea surem ents.
Ad d itiona l a na lyses p rovid ed a c ollec tive view of threshold s to a ll stim ulus
freq uenc ies a s d eterm ined b y a sp ec ific p roc ed ure. As ind ic a ted in Ta b le
4.8, the mea n of a ll freq uenc ies tested on the ABR w a s 73.2 d BnHL in
c om p a rison w ith a m ea n of the 86.71 d B on the ASSR. The sta nd a rd
d evia tion on these tw o m ea suring tec hniq ues d iffered - w ith the SD 12.5
d B on the ABR a nd 20.78 d B on the ASSR, ind ic a ting a w id er d isp ersion of
ASSR m ea surem ents. Due to the outp ut lim ita tions of eq uip m ent, the ABR
c ould not rend er resp onses on a ll sub jec ts resulting in a sm a ller num b er of
a va ila b le ABR m ea surem ents (21), a s op p osed
to a va ila b le ASSR
m ea surem ents (41). Althoug h a hig her num b er of d a ta p oints w ere
123
University of Pretoria etd – Stroeb
bel, D (2006
a va ila b le the SD va lues for a ll m ea surem ents p er p roc ed ure should a g a in
b e interp reted w ith c a ution.
Sta tistic a l a na lyses of the m ea n d a ta – using the Exa c t Wilc oxon Ra nk Sum
Test - ind ic a ted tha t no sta tistic a lly sig nific a nt d ifferenc e exist b etw een the
m ea n threshold s m ea sured w ith the ABR a nd the ASSR. Ta b le 4.9
sum m a rizes the results of the inferentia l sta tistic a l a na lysis of a vera g e for a ll
the ea rs m ea sured w ith the tone b urst ABR vs. 500 Hz ASSR a nd the c lic k
ABR vs. 2000 Hz a nd 4000 Hz ASSR.
Ta ble 4.9
Sta tistica l a na lysis of ABR a nd ASSR predicted results
Stimulus
P va lue
•
Click ABR vs. 2000 Hz ASSR
P = 0.4074
•
Click ABR vs. 4000 Hz ASSR
P = 1.0000
•
500 Hz tone burst vs. 500 Hz ASSR
P = 0.4991
For a d ifferenc e to b e sig nific a nt the p -va lue should b e sm a ller tha n 0.05
(Steyn, Sm it, Du Toit & Stra sheim , 2003:596). In this c a se none of the p va lues w ere sm a ller tha n 0.05 a nd therefore no sig nific a nt sta tistic a l
d ifferenc e w a s noted b etw een threshold s d eterm ined b y the una id ed
ABR a nd b y the una id ed ASSR.
The results p rovid ed b y the ABR a nd ASSR w ere a lso c om p a red w ith
reg a rd to its rela tion. Fig ure 4.9 show s the rela tionship or c orrela tion
c oeffic ient b etw een the 500 Hz toneb urst ABR (TB) a nd the ASSR
p red ic ted threshold , using a 500 Hz c a rrier freq uenc y. It is im p orta nt to
note tha t a la rg e p rop ortion of ea rs (5 of 12) ha d no resp onse to the 500
124
University of Pretoria etd – Stroeb
bel, D (2006
Hz tone b urst ABR a t 90 d BnHL; therefore, only 7 ea rs a re rep resented in
the eq ua tion. The d a ta ind ic a te tha t there is a m od era te to m a rked
p ositive c orrela tion b etw een ASSR threshold s a t 500 Hz a nd the 500 Hz
tone b urst ABR threshold s (r = .77).
120
y = 1.4063x - 32.612
2
R = 0.5875
100
500 Hz ASSR
80
60
40
20
0
0
10
20
30
40
50
60
70
80
90
100
Tone Burst ABR
Figure 4.9
Rela tionship between 500 Hz tone burst ABR a nd ASSR
prediction ba sed on the mea surement for seven ea rs
120
120
y = 0.4199x + 41.922
2
R = 0.3021
100
100
80
80
4000 Hz ASSR
2000 Hz ASSR
y = 0.5783x + 35.12
2
R = 0.3084
60
60
40
40
20
20
0
0
0
10
20
30
40
50
Click ABR
60
70
80
90
100
0
10
20
30
40
50
60
70
80
90
100
Click ABR
Figure 4.10 Rela tionship between the click ABR a nd 2000 a nd 4000 Hz
ASSR prediction ba sed on the mea surement of seven ea rs
Fig ure 4.10 show s a c om p a rison m a trix b etw een the threshold s for seven
ea rs a s ob ta ined w ith the c lic k ABR to the 2000 Hz a nd 4000 Hz ASSR. It is
im p orta nt to note tha t 5 of the ea rs tested , ha d no resp onse to c lic ks a t 90
125
University of Pretoria etd – Stroeb
bel, D (2006
d BnHL; therefore only 7 ea rs a re rep resented in fig ure 4.8. The d a ta
ind ic a te tha t there is a fa ir d eg ree of p ositive c orrela tion b etw een the
c lic k ABR a nd the 2000 Hz ASSR threshold (r = .56). A sim ila r d eg ree of
p ositive c orrela tion is found b etw een the 4000 Hz ASSR threshold a nd the
c lic k-evoked ABR threshold (r = .57). The c orrela tion results a re therefore a
c onfirm a tion of w ha t w a s ind ic a ted b y the inferentia l sta tistic s.
To sum m a rize:
•
Results ind ic a te tha t it w a s in m ore insta nc es p ossib le to d eterm ine
threshold s w ith the ASSR tha n w ith the ABR
•
In 67% of the freq uenc ies tested the threshold s b etw een the ABR
a nd ASSR c orresp ond ed w ithin 10 d B of ea c h other.
•
ASSR threshold s for the six sub jec ts show a b ig g er va ria tion tha n the
ABR threshold s, b ut it is im p ossib le to c om e to a c lea r c onc lusion a s
to w ha t this m a y ind ic a te.
•
Differenc es b etw een m ea n threshold s m ea sured w ith the ASSR a nd
the ABR exist, b ut it show s no sta tistic a l sig nific a nc e
•
Results c onfirm tha t there is a m od era te to fa ir p ositive c orrela tion
b etw een the
threshold s d eterm ined
b y the
ASSR a nd
ABR
resp ec tively (Leed y & Orm rod , 2005:306).
4.2.2.2
Una id ed ASSR vs. una id ed b eha viora l threshold s
The num b er of ea rs tested w a s sim ila r on these tw o a p p roa c hes. When
a na lyzing the ra nge inform a tion on these tw o m ea surem ents, the ra ng e
of the results a lso seem s sim ila r (see Ta b le 4.7). A d ifferenc e of 5 d B in the
ra ng e of the 4000 Hz c om p a rison is p resent, w ith a d ifferenc e of 15 d B a t
2000 Hz, 10 d B a t 1000 Hz a nd 10 d B a t 500 Hz. The SD va lues seem to
126
University of Pretoria etd – Stroeb
bel, D (2006
c onfirm the ra ng e va lues – the SD va lues on the ASSR va ried from 16.69 to
24.73. The SD va lues on the b eha viora l threshold s va ried from 11.76 to
20.90. Interp reta tion of the SD va lues on suc h a sm a ll sa m p le how ever is
risky.
Fig ure 4.11 illustra tes the mea n una id ed ASSR p red ic ted threshold s a nd
una id ed b eha viora l threshold s ob ta ined a t ea c h freq uenc y for a ll the
ea rs tested (n=12). The m ea n una id ed ASSR p red ic ted threshold levels
ra ng ed from 84.6 d B HL to 87.7 d B HL. The m ea n b eha viora l threshold
levels ra ng ed from 81.4 d B HL to 87.8 d B HL (Ta b le 4.7). One ea r show ed
no resp onse on either of the tw o p roc ed ures a t a ll freq uenc ies. Another
three ea rs ha d no resp onse a t 4000 Hz on the ASSR a nd tw o of these ea rs
ha d no resp onse a t 4000 Hz on the b eha viora l eva lua tion.
Avera ge results
Frequency (kHz)
dB HL
0.5
0
10
20
30
40
50
60
70
80
90
100
110
120
1
2
4
BTH
ASSR
Figure 4.11 Mea n una ided ASSR thresholds a nd una ided beha viora l
thresholds obta ined a t ea ch frequency for a ll the ea rs tested
(n=12).
127
University of Pretoria etd – Stroeb
bel, D (2006
A d ifferenc e of 5 d B w a s noted a t 500 Hz a nd 1000 Hz a nd 10 d B a t 4000
Hz b etw een the mea n of the threshold s d eterm ined b y the ASSR
p red ic tion a nd b eha viora l m ea surem ents. In the low freq uenc ies the
b eha viora l threshold s w ere slig htly low er a nd a t 4000 Hz the ASSR
p red ic ted threshold s w ere m inim a lly low er. Results show tha t the a vera g es
of threshold s for a ll the ea rs, d eterm ined w ith the ASSR a nd b eha viora l
a ssessm ents, w ere very sim ila r (see Ta b le 4.8).
Com p a ring the ASSR p red ic ted threshold s w ith b eha viora l threshold s for
a ll freq uenc ies tested , it w a s noted tha t the m a jority of c om p a ra b le
threshold s - 33 of the 41 (80.5%) – show ed a d ifferenc e of 10 d B or less.
Compa ra tive frequency
thresholds
These results a re sum m a rized in Fig ure 4.12.
33
30
27
24
21
18
15
12
9
6
3
0
” 10 d B
15 d B
20 d B
> 20 d B
Difference in dB
Figure 4.12 Representa tion of compa ra tive frequency thresholds between
the ASSR predicted thresholds a nd beha viora l thresholds
Sta tistic a l a na lyses (Exa c t Wilc oxon Ra nk Sum Test) of the m ea n d a ta a re
sum m a rized in Ta b le 4.10. No sta tistic a l d ifferenc e b etw een the threshold s
d eterm ined b y the una id ed ASSR a nd una id ed b eha viora l a ssessm ent
w a s found , a s a ll p -va lues w ere m ore tha n 0.05.
128
University of Pretoria etd – Stroeb
bel, D (2006
Ta ble 4.10
Sta tistica l a na lysis of ASSR predictions a nd
beha viora l
mea sures
Stimulus
P - va lue
•
500 Hz ASSR vs. beha vioura l threshold
P = 0.8128
•
1000 Hz ASSR vs. beha vioura l threshold
P = 0.7475
•
2000 Hz ASSR vs. beha vioura l threshold
P = 0.7440
•
4000 Hz ASSR vs. beha vioura l threshold
P = 0.5039
The results p rovid ed b y the ASSR a nd b eha viora l a ssessm ent w ere a lso
c om p a red w ith reg a rd to its rela tion. The follow ing sc a tter p lots in Fig ure
4.13 rep resent the rela tionship or c orrela tion c oeffic ient b etw een ea c h
freq uenc y tested d uring the ASSR eva lua tion a nd the sub seq uent
b eha viora l m ea surem ent. A hig hly d ep end a b le to m od era te p ositive
c orrela tion is id entified for three of the test freq uenc ies, na m ely r = .93 a t
500 Hz; r = .82 a t 1000 Hz; r = .79 a t 2000 Hz, d eterm ined w ith the ASSR a nd
b eha viora l a ssessm ents. Threshold s d eterm ined w ith the tw o p roc ed ures
ind ic a tes a m od era te to fa ir d eg ree of p ositive c orrela tion a t 4000 Hz (r =
.59).
129
University of Pretoria etd – Stroeb
bel, D (2006
120
120
y = 0.6958x + 22.006
y = 0.7669x + 16.529
2
2
R = 0.678
100
1000 Hz Beha viora l a ssessment
500 Hz Beha viora l a sssessment
R = 0.8684
80
60
40
20
0
100
80
60
40
20
0
0
10
20
30
40
50
60
70
80
90
100
110
120
0
10
20
30
40
50
500 Hz ASSR
120
70
80
90
100
110
120
130
120
y = 0.6631x + 29.101
y = 0.4295x + 49.295
R2 = 0.3436
2
R = 0.6176
4000 Hz Beha viora l a ssessment
2000 Hz Beha viora l a ssessment
60
1000 Hz ASSR
100
80
60
40
20
100
80
60
40
20
0
0
0
10
20
30
40
50
60
70
80
90
100
110
120
130
0
Figure 4.13 Rela tionship
10
20
30
40
50
60
70
80
90
100
110
120
4000 Hz ASSR
2000 Hz ASSR
between thresholds determined
with ASSR
predictions a nd beha viora l responses for a specific number of
ea rs.
In sum m a ry:
•
The num b er of freq uenc ies w here threshold s c ould b e d eterm ined
w ith ASSR c om p a res fa vora b le
w ith tha t of the
b eha viora l
a ssessm ents.
•
In 80.5% of the freq uenc ies tested , the threshold s b etw een the ASSR
a nd b eha viora l a ssessm ent c orresp ond ed w ithin 10 d B of ea c h
other.
•
The
ra ng e
of the
m ea surem ents d eterm ined
w ith
the
tw o
p roc ed ures c om p a res w ell.
130
University of Pretoria etd – Stroeb
bel, D (2006
•
There is no sta tistic a l d ifferenc e b etw een a vera g es d eterm ined w ith
the ASSR p red ic tions a nd b eha viora l threshold s
•
Results ind ic a te a hig hly d ep end a b le to fa irly p ositive c orrela tion
b etw een
threshold s
d eterm ined
by
ASSR
a nd
b eha viora l
a ssessm ents.
4.2.2.3
Una id ed ABR vs. una id ed b eha viora l threshold s
As ind ic a ted in Ta b le 4.7, the ra nge of the a b solute m ea surem ents va lues
for the 500 Hz b eha viora l threshold is 25 d B b roa d er tha n the ra ng e for the
tone b urst ABR. The ra ng e for the c lic k ABR a nd 2000 Hz b eha viora l
resp onse a re b oth 45 d B. The ra ng e for the c lic k ABR w a s 10 d B b roa d er
tha n the ra ng e for the 4000 Hz b eha viora l threshold s. The num b er of ea rs
ta ken into a c c ount is seven on the ABR a nd 11 on the 2000 Hz b eha viora l
threshold s a ssessm ent, b ut nine on the 4000 Hz b eha viora l a ssessm ent. It
w ould seem a s if the ra ng e of the threshold tha t w ere d eterm ined w ith the
500 Hz b eha viora l threshold a ssessm ent is b roa d er tha n the ra ng e of the
tone b urst ABR. This is not the c a se how ever w ith the ra ng e of threshold
d eterm ina tion b etw een the c lic k ABR a nd 2000 Hz b eha viora l threshold
a ssessm ent. It is therefore d iffic ult to d ra w a ny c onc lusions b a sed on the
ra ng e of the threshold s m ea sured w ith the ABR a nd b eha viora l threshold
a ssessm ents. It is a g a in risky to d ra w c onc lusion from the SD va lues. These
va lues seem to c onfirm the ra ng e va lues. The num b er of d a ta p oints for
the ABR m ea surem ents w ere lim ited to seven a nd c onsid ering the
inevita b le ind ivid ua l d ifferenc es, va ria tion in resp onses c a n b e exp ec ted
to b e hig h a nd w ould ha ve a n a ffec t on SD va lues for suc h a sm a ll
sa m p le.
131
University of Pretoria etd – Stroeb
bel, D (2006
Com p a ring the tone b urst a nd c lic k evoked ABR threshold s w ith the
b eha viora l threshold s, it w a s noted tha t only 12 of the 21 (57%) of the
c om p a ra b le threshold s, show ed a d ifferenc e of 10 d B or less. These results
Compa ra tive frequency
thresholds
a re sum m a rized in Fig ure 4.14.
33
30
27
24
21
18
15
12
9
6
3
0
” 10 d B
15 d B
20 d B
> 20 d B
Difference in dB
Figure 4.14 Representa tion of compa ra tive frequency thresholds between
the ABR a nd beha viora l thresholds
Further insig ht w a s g a ined
b y a g a in c a lc ula ting
the
mea n a nd
c onsid ering the d ifferenc e b etw een the m ea n of the a b solute va lues for
ea c h of the eva lua tion p roc ed ures. The m ea n of the ABR ra ng ed from
71.4 d BnHL for the c lic k ABR to 75 d BnHL for the tone b urst ABR. The m ea n
of the b eha viora l threshold s ra ng ed from 81.4 d B HL to 87.8 d B HL. The
m ea n of the tone b urst ABR threshold s d iffered w ith 6.4 d B from the 500 Hz
b eha viora l threshold s (see Ta b le 4.7). The tone b urst ABR ha d the low er
va lue. When c om p a ring the c lic k ABR w ith the 2000 Hz a nd 4000 Hz
b eha viora l threshold s, the a vera g e results ind ic a te tha t the ABR a g a in
ha d the low er resp onse level – w ith a d ifferenc e of a p p roxim a tely 16.3 d B.
It is releva nt thoug h, to c onsid er a g a in the num b er of ea rs tested on ea c h
132
University of Pretoria etd – Stroeb
bel, D (2006
p roc ed ure (Ta b le 4.7). Only seven ea rs rep resent the results on the ABR,
w here 11 a re rep resented on the 500 Hz a nd 2000 Hz b eha viora l threshold
m ea surem ent a nd nine ea rs a re rep resented on the 4000 Hz b eha viora l
threshold m ea surem ent. The five ea rs not rep resented on the ABR results
a re the ones tha t ha d no resp onse on this p roc ed ure a nd therefore fa ll in
a c a teg ory m ore severe tha n c ould b e m ea sured b y the ABR.
Ad d itiona l a na lyses p rovid ed a c ollec tive view of threshold s to a ll stim ulus
freq uenc ies a s d eterm ined b y a sp ec ific p roc ed ure. As ind ic a ted in Ta b le
4.8, the m ea n of a ll freq uenc ies tested on the ABR w a s 73.2 d BnHL in
c om p a rison w ith the m ea n of 82.74 d B on the b eha viora l a ssessm ent. The
SD on these tw o m ea suring tec hniq ues d iffered – w ith the SD 12.5 d B on
the ABR a nd 18.75 on the b eha viora l a ssessm ent, ind ic a ting a w id er
d isp ersion of b eha viora l threshold s. Due to the outp ut lim ita tions of
eq uip m ent, the ABR c ould not rend er resp onse on a ll sub jec ts resulting in
a sm a ller num b er of a va ila b le ABR threshold s (21) a s op p osed to a va ila b le
b eha viora l threshold s (42). Althoug h a hig her num b er of d a ta p oints w ere
a va ila b le, the SD va lues for a ll m ea surem ents p er p roc ed ure should a g a in
b e interp reted w ith c a ution.
Sta tistic a l a na lysis of the m ea n d a ta – using the Exa c t Wilc oxon Ra nk Sum
Test – ind ic a ted tha t no sta tistic a l sig nific a nt d ifferenc e exists b etw een the
m ea n threshold s m ea sured w ith the ABR a nd b eha viora l threshold
a ssessm ents. Ta b le 4.11 sum m a rizes the results of the inferentia l sta tistic a l
a na lysis of the a vera g e for a ll the ea rs m ea sured w ith the tone b urst ABR
vs. 500 Hz b eha viora l a ssessm ent a nd the c lic k ABR vs. 2000 Hz a nd 4000
Hz b eha viora l threshold a ssessm ent.
133
University of Pretoria etd – Stroeb
bel, D (2006
Ta ble 4.11
Sta tistica l a na lysis of ABR a nd beha viora l mea sures
Stimulus
P - va lue
•
500 Hz beha vioura l threshold vs. tone
burst ABR
P = 0.1563
•
2000 Hz beha vioura l threshold vs. click
ABR
P = 0.5000
•
4000 Hz beha vioura l threshold vs. click
ABR
P = 0.2188
The results p rovid ed b y the ABR a nd b eha viora l a ssessm ent w ere a lso
c om p a red w ith reg a rd to its rela tion. Fig ure 4.15 show s the rela tionship or
c orrela tion c oeffic ient b etw een the tone b urst ABR a nd the 500 Hz
b eha viora l threshold a ssessm ent. It is im p orta nt to note tha t a la rg e
p rop ortion of the ea rs (5 of 12) ha d no resp onse to the tone b urst ABR a t
90 d BnHL; therefore, only seven ea rs a re rep resented in this eq ua tion. The
d a ta ind ic a te tha t there is a m od era te to m a rked p ositive c orrela tion
b etw een b eha viora l threshold a ssessm ent a t 500 Hz a nd the tone b urst
ABR (r = .77).
500 Hz Beha viora l a ssessment
120
y = 1.125x - 16.518
2
R = 0.5981
100
80
60
40
20
0
0
10
20
30
40
50
60
70
80
90
100
Tone Burst ABR
Figure 4.15 Rela tionship between tone burst ABR a nd 500 Hz beha viora l
threshold a ssessment ba sed on the mea surement for seven
ea rs
134
University of Pretoria etd – Stroeb
bel, D (2006
Fig ure 4.16 show s a c orrela tion m a trix b etw een the threshold s for seven
ea rs a s ob ta ined w ith the c lic k ABR to the 2000 Hz a nd 4000 Hz b eha viora l
threshold a ssessm ent. Ag a in it should b e noted tha t only seven ea rs a re
rep resented in the eq ua tion a s five ea rs ha d no resp onse to the ABR a t 90
d BnHL. The d a ta ind ic a te tha t there is a d ep end a b le p ositive c orrela tion
b etw een the c lic k ABR threshold a nd 2000 Hz b eha viora l threshold
a ssessm ent (r = .89). A fa ir d eg ree of p ositive c orrela tion is a lso found
b etw een the 4000 Hz b eha viora l threshold a nd the c lic k ABR (r = .40). The
c orrela tion results a re therefore a c onfirm a tion of w ha t w a s ind ic a ted b y
the inferentia l sta tistic s.
120
y = 0.5331x + 36.205
R2 = 0.7997
4000 Hz Beha viora l a ssessment
2000 Hz Beha viora l a ssessment
120
100
80
60
40
20
y = 0.1536x + 68.313
2
R = 0.1632
100
80
60
40
20
0
0
0
10
20
30
40
50
60
70
80
90
100
Click ABR
0
10
20
30
40
50
60
70
80
90
100
Click ABR
Figure 4.16 Rela tionship between click ABR a nd 2000 Hz a nd 4000 Hz
beha viora l threshold a ssessment ba sed on the mea surement
for seven ea rs respectively
To sum m a rize:
•
Results ind ic a te tha t it w a s not p ossib le to d eterm ine threshold s w ith
the ABR in a ll the c a ses.
•
In only 57% of the freq uenc ies tested the threshold s b etw een the
ABR a nd b eha viora l a ssessm ent c orresp ond ed w ithin 10 d B of ea c h
other.
135
University of Pretoria etd – Stroeb
bel, D (2006
•
Differenc es b etw een a vera g es of threshold s m ea sured w ith the ABR
a nd b eha viora l threshold a ssessm ents exist, b ut it show s no sta tistic a l
sig nific a nt d ifferenc e.
•
Results c onfirm tha t there is a fa ir to d ep end a b le p ositive c orrela tion
b etw een the threshold s d eterm ined b y the ABR a nd b eha viora l
threshold a ssessm ent resp ec tively.
Consid ering the c om p a ra tive results d esc rib ed in 4.2.2.1, 4.2.2.2 a nd
4.2.2.3, it seem s tha t the ASSR m ea surem ent c om p a re w ell to the
m ea surem ents d one w ith the other tw o p roc ed ures, a lthoug h there seem s
to b e a slig htly hig her c orrela tion b etw een the ASSR a nd the b eha viora l
a ssessm ents tha n w ha t exist b etw een the ABR a nd the b eha viora l
a ssessm ents. 80.5% of the freq uenc ies tested throug h the use of ASSR
c orresp ond ed w ithin 10 d B w ith the b eha viora l threshold s. Only 57% of the
freq uenc ies tested throug h the use of the ABR c orresp ond ed w ithin 10 d B
of the b eha viora l threshold s – even thoug h no c orrec tions w ere m a d e for
the ABR threshold s.
4.3
RESULTS FOR SUB-AIM 2: TO INVESTIGATE THE CLINICAL VALUE OF THE
ASSR FOR RELEVANT EARLY FITTING OF HEARING AIDS IN INFANTS BY
DETERMINING
AND
COMPARING
AIDED
ASSR
AND
AIDED
BEHAVIORAL THRESHOLDS.
Exc ep t in the c a se of sub jec t 1, resp onses w ere rec ord ed w hile the
sub jec t w a s w ea ring b ina ura l hea ring a id s. The threshold s rec ord ed a re
therefore a n ind ic a tion of the a id ed threshold s of the b est resp onse a t
ea c h freq uenc y of the b est ea r.
136
University of Pretoria etd – Stroeb
bel, D (2006
Resp onses for the ASSR w ere rec ord ed a t c a rrier freq uenc ies of 500, 1000,
2000 a nd 4000 Hz in this g roup of six hea ring im p a ired infa nts using hea ring
a id s. The sa m e freq uenc ies w ere eva lua ted d uring the b eha viora l
a ssessm ent. The results from the ind ivid ua l sub jec ts w ill b e d isc ussed first,
follow ed b y the c ollec tive results.
4.3.1 Individua l subject results for sub-a im 2
Ea c h ind ivid ua l sub jec t’ s a id ed results w ill now b e rep orted on. In the
ind ivid ua l ta b les (See Ta b les 4.12 to 4.17) a n ind ic a tion is g iven of b oth the
una id ed ASSR results – the m ea sured threshold s a nd the p red ic ted
threshold s. Both these threshold va lues w ill b e ta ken into a c c ount a s the
norm a tive d a ta from w hic h p red ic ted threshold s a re c a lc ula ted , w ere not
c om p iled for a id ed ASSR’ s. Therefore a true c om p a rison c a n b e m a d e
b etw een the una id ed a nd a id ed ASSR results. The b eha viora l results w ill
b e c om p a red w ith b oth va lues on the a id ed ASSR.
4.3.1.1
Sub jec t 1: Results for sub -a im 2
The a id ed test results for sub jec t 1 a re p resented in Ta b le 4.12 a nd Fig ure
4.17.
137
University of Pretoria etd – Stroeb
bel, D (2006
Ta ble 4.12
Una ided ASSR, a ided ASSR a nd a ided beha viora l thresholds
mea surements for subject 1
Unaided ASSR results
Unaided ASSR predicted results
Aided ASSR results
Aided ASSR predicted results
Aided Behavioral assessment results
500 Hz
1000 Hz
2000 Hz
4000 Hz
70 dB
65 dB
75 dB
80 dB
500 Hz
1000 Hz
2000 Hz
4000 Hz
55 dB
55 dB
65 dB
70 dB
500 Hz
1000 Hz
2000 Hz
4000 Hz
NR
50 dB
30 dB
45 dB
500 Hz
1000 Hz
2000 Hz
4000 Hz
NR
30 dB
20 dB
15 dB
500 Hz
1000 Hz
2000 Hz
4000 Hz
25 dB
20 dB
25 dB
30 dB
Aided results
Frequency (kHz)
0.5
1
2
4
dB HL
0
10
20
30
40
50
60
70
80
90
100
110
120
Aided BTH
Aided ASSR m
Aided ASSR p
Figure 4.17 Aided results for subject 1 including beha viora l thresholds a nd
ASSR thresholds – mea sured a nd predicted
A rec og niza b le d ifferenc e w a s noted b etw een the una id ed a nd a id ed
ASSR resp onses (see Ta b le 4.12). When c om p a ring the m ea sured
threshold s, a d ifferenc e of b etw een 15 to 45 d B a c ross the freq uenc y
ra ng e w a s noted . When c onsid ering the d ifferenc e in the a id ed a nd
una id ed p red ic ted ASSR threshold s, the d ifferenc e is a p p roxim a tely 10 d B
138
University of Pretoria etd – Stroeb
bel, D (2006
m ore. No a id ed resp onse c ould b e m ea sured a t the m a xim um outset of
the eq uip m ent a t 500 Hz.
When c om p a ring the a id ed ASSR w ith the a id ed b eha viora l threshold s, a
d ifferenc e of 30 d B w a s noted b etw een the m ea sured a id ed ASSR a nd
the b eha viora l threshold a t 1000 Hz. A d ifferenc e of 5 d B w a s p resent for
the sa m e c om p a rison a t 2000 Hz a nd a 15 d B d ifferenc e w a s p resent for
the 4000 Hz c om p a rison. On a ll of these c om p a risons, the a id ed
b eha viora l threshold s ha d the low er va lue.
When c om p a ring the a id ed ASSR – using the p red ic ted threshold s w ith the
a id ed b eha viora l threshold s, a d ifferenc e of 10 d B w a s noted a t 1000 Hz, 5
d B a t 2000 Hz a nd 15 d B a t 4000 Hz. In this c a se the ASSR ha d the low er
va lues for the 2000 Hz a nd 4000 Hz c om p a rison.
4.3.1.2
Sub jec t 2: Results for sub -a im 2
The a id ed test results for sub jec t 2 a re p resented in Ta b le 4.13 a nd Fig ure
4.18.
Ta ble 4.13
Una ided ASSR, a ided ASSR a nd a ided beha viora l thresholds
139
University of Pretoria etd – Stroeb
bel, D (2006
mea surements for subject 2
Unaided ASSR results
Unaided ASSR predicted results
Aided ASSR results
Aided ASSR predicted results
Aided Behavioral assessment results
500 Hz
1000 Hz
2000 Hz
4000 Hz
60 dB
60 dB
60 dB
65 dB
500 Hz
1000 Hz
2000 Hz
4000 Hz
40 dB
50 dB
55 dB
50 dB
500 Hz
1000 Hz
2000 Hz
4000 Hz
50 dB
35 dB
35 dB
30 dB
500 Hz
1000 Hz
2000 Hz
4000 Hz
25 dB
20 dB
20 dB
5 dB
500 Hz
1000 Hz
2000 Hz
4000 Hz
30 dB
35 dB
35 dB
40 dB
Aided results
Frequency (kHz)
0.5
1
2
4
0
10
dB HL
20
30
40
50
60
70
80
90
100
110
120
Aided BTH
Aided ASSR m
Aided ASSR p
Figure 4.18 Aided results from subject 2 including beha viora l thresholds
a nd ASSR thresholds – mea sured a nd predicted
A rec og niza b le d ifferenc e w a s noted b etw een the una id ed a nd a id ed
ASSR threshold s (see Ta b le 4.13). When c om p a ring the m ea sured
threshold s, a d ifferenc e of b etw een 10 to 35 d B a c ross the freq uenc y
ra ng e w a s noted . When c onsid ering the d ifferenc e in the a id ed a nd
una id ed p red ic ted ASSR threshold s, the d ifferenc e b etw een the va lues
w ere 15 to 45 d B.
140
University of Pretoria etd – Stroeb
bel, D (2006
When c om p a ring the a id ed ASSR threshold w ith the a id ed b eha viora l
threshold s, a d ifferenc e of 20 d B w a s noted b etw een the m ea sured a id ed
ASSR threshold a nd the b eha viora l threshold a t 500 Hz. No d ifferenc e w a s
p resent a t 1000 Hz a nd 2000 Hz. At 4000 Hz a 10 d B d ifferenc e w a s noted .
For the 500 Hz c om p a rison, the a id ed b eha viora l threshold s ha d the low er
va lue. For 4000 Hz c om p a rison, the ASSR va lue ha d the low er va lue.
When c om p a ring the a id ed ASSR – using the p red ic ted threshold s w ith the
a id ed b eha viora l threshold s, a d ifferenc e of 5 d B w a s noted a t 500 Hz, 15
d B a t 1000 Hz, 15 d B a t 2000 Hz a nd 35 d B a t 4000 Hz. In this c a se the ASSR
ha d the low er va lues a c ross the freq uenc y ra ng e.
4.3.1.3
Sub jec t 3: Results for sub -a im 2
The a id ed test results for sub jec t 3 a re p resented in Ta b le 4.14 a nd Fig ure
4.19.
Ta ble 4.14
Una ided ASSR, a ided ASSR a nd a ided beha viora l thresholds
mea surements for subject 3
141
University of Pretoria etd – Stroeb
bel, D (2006
Unaided ASSR results
Unaided predicted ASSR results
Aided ASSR results
Aided predicted ASSR results
Aided Behavioral assessment results
500 Hz
1000 Hz
2000 Hz
4000 Hz
100 dB
100 dB
90 dB
90 dB
500 Hz
1000 Hz
2000 Hz
4000 Hz
90 dB
95 dB
85 dB
80 dB
500 Hz
1000 Hz
2000 Hz
4000 Hz
NR
50 dB
60 dB
50 dB
500 Hz
1000 Hz
2000 Hz
4000 Hz
NR
35 dB
45 dB
35 dB
500 Hz
1000 Hz
2000 Hz
4000 Hz
35 dB
35 dB
40 dB
40 dB
Aided results
Frequency (kHz)
dB HL
0.5
0
10
20
30
40
50
60
70
80
90
100
110
120
1
2
4
Aid ed BTH
Aid ed ASSR m
Aid ed ASSR p
Figure 4.19 Aided results from subject 3 including beha viora l thresholds
a nd ASSR thresholds – mea sured a nd predicted
Ag a in a rec og niza b le d ifferenc e w a s noted b etw een the una id ed a nd
a id ed ASSR threshold s (see Ta b le 4.14). When c om p a ring the m ea sured
ASSR threshold , a d ifferenc e of b etw een 40 to 50 d B a c ross the freq uenc y
ra ng e w a s noted . When c onsid ering the d ifferenc e in the a id ed a nd
una id ed p red ic ted ASSR threshold s, the d ifferenc e b etw een the va lues
w ere 25 to 60 d B. No a id ed ASSR resp onse a t 500 Hz c ould b e m ea sured
on this sub jec t.
142
University of Pretoria etd – Stroeb
bel, D (2006
When c om p a ring the a id ed ASSR threshold w ith the a id ed b eha viora l
threshold s, a d ifferenc e of 15 d B w a s noted b etw een the m ea sured a id ed
ASSR threshold a nd the b eha viora l threshold a t 1000 Hz. A d ifferenc e of
20 d B w a s noted a t 2000 Hz a nd a 30 d B d ifferenc e a t 4000 Hz. In this
c om p a rison the a id ed b eha viora l threshold s ha d the low er va lue.
When c om p a ring the a id ed ASSR – using the p red ic ted threshold s w ith the
a id ed b eha viora l threshold s, no d ifferenc e w a s noted a t 1000 Hz, 5 d B a t
2000 Hz a nd 15 d B a t 4000 Hz. In this c a se the ASSR ha d the hig her va lue
for 2000 Hz a nd the low er va lue for 4000 Hz.
4.3.1.4
Sub jec t 4: Results for sub -a im 2
The a id ed test results for sub jec t 4 a re p resented in Ta b le 4.15 a nd Fig ure
4.20.
Ta ble 4.15
Una ided ASSR, a ided ASSR a nd a ided beha viora l thresholds
mea surements for subject 4
Unaided ASSR results
500 Hz
1000 Hz
2000 Hz
4000 Hz
143
University of Pretoria etd – Stroeb
bel, D (2006
Unaided predicted ASSR results
Aided ASSR results
Aided predicted ASSR results
Aided Behavioral assessment results
90 dB
90 dB
90 dB
80 dB
500 Hz
1000 Hz
2000 Hz
4000 Hz
80 dB
85 dB
85 dB
70 dB
500 Hz
1000 Hz
2000 Hz
4000 Hz
NR
50 dB
60 dB
50 dB
500 Hz
1000 Hz
2000 Hz
4000 Hz
NR
35 dB
45 dB
35 dB
500 Hz
1000 Hz
2000 Hz
4000 Hz
30 dB
35 dB
25 dB
25 dB
Aided results
Frequency (kHz)
dB HL
0.5
0
10
20
30
40
50
60
70
80
90
100
110
120
1
2
4
Aid e d BTH
Aid e d ASSR m
Aid e d ASSR p
Figure 4.20 Aided results from subject 4 including beha viora l thresholds
a nd ASSR thresholds – mea sured a nd predicted
Ag a in a rec og niza b le d ifferenc e w a s noted b etw een the una id ed a nd
a id ed ASSR resp onses (see Ta b le 4.15). When c om p a ring the m ea sured
threshold s, a d ifferenc e of b etw een 30 to 40 d B a c ross the freq uenc y
ra ng e w a s noted . When c onsid ering the d ifferenc e in the a id ed a nd
una id ed p red ic ted ASSR threshold s, the d ifferenc e b etw een the va lues
w ere 35 to 50 d B. No a id ed ASSR resp onse a t 500 Hz c ould a g a in b e
m ea sured on this sub jec t.
144
University of Pretoria etd – Stroeb
bel, D (2006
When c om p a ring the a id ed ASSR w ith the a id ed b eha viora l threshold s, a
d ifferenc e of 15 d B w a s noted b etw een the m ea sured a id ed ASSR a nd
the b eha viora l threshold a t 1000 Hz. A d ifferenc e of 35 d B w a s noted a t
2000 Hz a nd a 25 d B d ifferenc e a t 4000 Hz. In this c om p a rison the a id ed
b eha viora l threshold s ha d the low er va lue.
When c om p a ring the a id ed ASSR – using the p red ic ted threshold s w ith the
a id ed b eha viora l threshold s, no d ifferenc e w a s noted a t 1000 Hz, 20 d B a t
2000 Hz a nd 10 d B a t 4000 Hz. In this c a se the ASSR ha d the hig her va lue
for 2000 Hz a nd 4000 Hz.
4.3.1.5
Sub jec t 5: Results for sub -a im 2
The a id ed test results for sub jec t 4 a re p resented in Ta b le 4.16 a nd Fig ure
4.21.
145
University of Pretoria etd – Stroeb
bel, D (2006
Ta ble 4.16
Una ided ASSR, a ided ASSR a nd a ided beha viora l thresholds
mea surements for subject 5
Unaided ASSR results
Unaided predicted ASSR results
Aided ASSR results
Aided predicted ASSR results
Aided Behavioral assessment results
500 Hz
1000 Hz
2000 Hz
4000 Hz
110 dB
110 dB
110 dB
105 dB
500 Hz
1000 Hz
2000 Hz
4000 Hz
105 dB
105 dB
105 dB
100 dB
500 Hz
1000 Hz
2000 Hz
4000 Hz
NR
60 dB
70 dB
80 dB
500 Hz
1000 Hz
2000 Hz
4000 Hz
NR
45 dB
60 dB
65 dB
500 Hz
1000 Hz
2000 Hz
4000 Hz
45 dB
50 dB
55 dB
60 dB
Aided results
Frequency (kHz)
0.5
1
2
4
0
10
dB HL
20
30
40
50
60
70
80
90
100
110
120
Aide d BTH
Aide d ASSR m
Aide d ASSR p
Figure 4.21 Aided results from subject 5 including beha viora l thresholds
a nd ASSR thresholds – mea sured a nd predicted
Ag a in a rec og niza b le d ifferenc e w a s noted b etw een the una id ed a nd
a id ed ASSR resp onses (see Ta b le 4.16). When c om p a ring the m ea sured
threshold s, a d ifferenc e of b etw een 30 to 50 d B a c ross the freq uenc y
ra ng e w a s noted . When c onsid ering the d ifferenc e in the a id ed a nd
una id ed p red ic ted ASSR threshold s, the d ifferenc e b etw een the va lues
146
University of Pretoria etd – Stroeb
bel, D (2006
w ere 35 to 60 d B. No a id ed ASSR resp onse a t 500 Hz c ould a g a in b e
m ea sured on this sub jec t.
When c om p a ring the a id ed ASSR w ith the a id ed b eha viora l threshold s, a
d ifferenc e of 10 d B w a s noted b etw een the m ea sured a id ed ASSR a nd
the b eha viora l threshold a t 1000 Hz. A d ifferenc e of 15 d B w a s noted a t
2000 Hz a nd a 20 d B d ifferenc e a t 4000 Hz. In this c om p a rison the a id ed
b eha viora l threshold s ha d the low er va lue.
When c om p a ring the a id ed ASSR – using the p red ic ted threshold s w ith the
a id ed b eha viora l threshold s, a 5 d B d ifferenc e w a s noted a t a ll the
freq uenc ies m ea sured (1000 Hz – 4000 Hz). In this c a se the ASSR ha d the
hig her va lue for 2000 Hz a nd 4000 Hz.
4.3.1.6
Sub jec t 6: Results for sub -a im 2
The a id ed test results for sub jec t 4 a re p resented in Ta b le 4.17 a nd Fig ure
4.22.
147
University of Pretoria etd – Stroeb
bel, D (2006
Ta ble 4.17
Una ided ASSR, a ided ASSR a nd a ided beha viora l thresholds
mea surements for subject 6
Unaided ASSR results
Unaided predicted ASSR results
Aided ASSR results
Aided predicted ASSR results
Aided Behavioral assessment results
500 Hz
1000 Hz
2000 Hz
4000 Hz
110 dB
110 dB
110 dB
NR
500 Hz
1000 Hz
2000 Hz
4000 Hz
105 dB
105 dB
105 dB
NR
500 Hz
1000 Hz
2000 Hz
4000 Hz
50 dB
40 dB
70 dB
NR
500 Hz
1000 Hz
2000 Hz
4000 Hz
25 dB
25 dB
60 dB
NR
500 Hz
1000 Hz
2000 Hz
4000 Hz
40 dB
50 dB
65 dB
NR
Aided Results
Frequency (kHz)
0.5
1
2
4
0
10
dB HL
20
30
40
50
60
70
80
90
100
Aid e d BTH
110
120
Aid e d ASSR p
Aid e d ASSR m
Figure 4.22 Aided results from subject 6 including beha viora l thresholds
a nd ASSR thresholds – mea sured a nd predicted
A rec og niza b le d ifferenc e w a s noted b etw een the una id ed a nd a id ed
ASSR resp onses (see Ta b le 4.17). When c om p a ring the m ea sured
threshold s, a d ifferenc e of b etw een 40 to 70 d B a c ross the freq uenc y
ra ng e w a s noted . When c onsid ering the d ifferenc e in the a id ed a nd
una id ed
p red ic ted
ASSR threshold s, the
d ifferenc e
b etw een
the
148
University of Pretoria etd – Stroeb
bel, D (2006
threshold s w ere 45 to 80 d B. No a id ed ASSR resp onse a t 4000 Hz c ould b e
m ea sured on this sub jec t.
When c om p a ring the a id ed ASSR w ith the a id ed b eha viora l threshold s, a
d ifferenc e of 10 d B w a s noted b etw een the m ea sured a id ed ASSR
threshold a nd the b eha viora l threshold a t 500 Hz. A d ifferenc e of 10 d B
w a s noted a t 1000 Hz a nd 2000 Hz. In this c om p a rison the a id ed
b eha viora l threshold s ha d the low er va lue for 500 a nd 2000 Hz.
When c om p a ring the a id ed ASSR – using the p red ic ted threshold s w ith the
a id ed b eha viora l threshold s, a 15 d B d ifferenc e w a s noted a t 500 Hz, 25
d B a t 1000 Hz a nd 5 d B a t 2000 Hz. In this c a se the ASSR ha d the low er
threshold s for the freq uenc ies tested .
Looking a t these a id ed results of the ind ivid ua l sub jec ts, it w ould seem
tha t the ASSR m a y p roof a va lua b le c ontrib ution to the p roc ess of
p ed ia tric hea ring a id fitting s. In the follow ing sec tion, the results tha t
c onc ern va lid a tion of hea ring a id fitting s in infa nts a s it is b a sed on the
m ea surem ents for a ll six sub jec ts a re d esc rib ed , c om p a red a nd d isc ussed .
4.3.2 Collective results for a ll six subjects concerning sub-a im 2
All of the sub jec ts show ed rec og niza b le a id ed ASSR resp onses a b ove their
una id ed ASSR threshold s. In Ta b le 4.18 the results of the a id ed ASSR – the
m ea sured
threshold
a nd
the p red ic ted
ASSR threshold
(using the
p red ic tion form ula e d evised b y Melb ourne University: Ra nc e et a l., 1995)
a s w ell a s the a id ed b eha viora l threshold s a re p rovid ed . Four sub jec ts
show ed no resp onse on the ASSR a t 500 Hz a id ed resp onse a t the
m a xim um outp ut of the sp ea ker (77,7d B). Only one sub jec t ha d no
149
University of Pretoria etd – Stroeb
bel, D (2006
resp onse to 4000 Hz a id ed ASSR (94,9d B). Resp onses w ere rec ord ed a t
c a rrier freq uenc ies of 500, 1000, 2000 a nd 4000 Hz in this g roup of six
hea ring im p a ired infa nts using hea ring a id s. The sa m e freq uenc ies w ere
tested d uring the b eha viora l a ssessm ent, na m ely 500 Hz, 1000 Hz, 2000 Hz
a nd 4000 Hz.
150
University of Pretoria etd – Stroeb
bel, D (2006
Ta ble 4.18
Summa ry of a ided thresholds for the six subjects a s determined by ASSR a nd beha viora l
a ssessments respectively.
Aided ASSR (measured)
Aided ASSR (predicted)
Aided BT
500
1000
2000
4000
500
1000
2000
4000
500
1000
2000
4000
Hz
Hz
Hz
Hz
Hz
Hz
Hz
Hz
Hz
Hz
Hz
Hz
Subject 1
NR
50 dB
30 dB
45 dB
NR
30 dB
15 dB
15 dB
25 dB
20 dB
25 dB
30 dB
Subject 2
50 dB
35 dB
35 dB
30 dB
25 dB
20 dB
20 dB
5 dB
30 dB
35 dB
35 dB
40 dB
Subject 3
NR
50 dB
60 dB
70 dB
NR
35 dB
45 db
55 dB
35 dB
35 dB
40 dB
40 dB
Subject 4
NR
50 dB
60 dB
50 dB
NR
35 dB
45 dB
35 dB
30 dB
35 dB
25 dB
25 dB
Subject 5
NR
50 dB
70 dB
80 dB
NR
45 dB
60 dB
65 dB
45 dB
50 dB
55 dB
60 dB
Subject 6
50 dB
45 dB
70 dB
NR
25 dB
25 dB
60 dB
NR
40 dB
50 dB
65 dB
NR
Mean
50 dB
46.7 dB
54.2 dB
55 dB
25 dB
31.7 dB
40.8 dB
35.5 dB
34.2 dB
37.5 dB
40.8 dB
39 dB
Range
O
15
40
50
0
25
45
60
20
30
40
35
Number
2
6
6
5
2
6
6
5
6
6
6
5
NR = No Response
P = prediction
151
University of Pretoria etd – Stroeb
bel, D (2006
The ra nge of resp onse d eterm ined w ith the ASSR (m ea sured a nd
p red ic ted ) a nd b eha viora l m ea surem ents w a s sim ila r b etw een d ifferent
m ea surem ents. At 500 Hz the ra ng e w a s the sa m e b etw een the tw o
d ifferent ASSR results, a s only tw o ea rs ha d resp onses a nd the resp onse
level w a s the sa m e for the ea rs. The ra ng e w a s 20d B on the b eha viora l
m ea surem ent, b ut six va lues a re c a lc ula ted a s op p osed to tw o. The ra ng e
a t 1000 Hz w a s 15d B on the m ea sured ASSR, 25d B on the p red ic ted ASSR
a nd 30d B on the b eha viora l a ssessm ent. At 2000 Hz the sa m e ra ng e w a s
noted for the m ea sured ASSR a nd b eha viora l a ssessm ent. The p red ic ted
ASSR w a s 5 hig her tha n these m ea sures. At 4000 Hz the m ea sured ASSR
ha d a ra ng e of 50d B, the p red ic ted ASSR 60d B a nd the b eha viora l
a ssessm ent 35d B.
Avera ge Aided Results (n 6)
Frequency (kHz)
dB HL
0.5
0
10
20
30
40
50
60
70
80
90
100
110
120
1
2
4
Aide d BTH
Aide d ASSR m
Aide d ASSR p
Figure 4.23 Compa rison of a vera ge a ided results for a ll mea sured ea rs
ba sed on a ided beha viora l a ssessment, mea sured a nd ASSR
predicted va lues
152
University of Pretoria etd – Stroeb
bel, D (2006
Fig ure 4.23 rep resents the a vera g e a id ed results. The mea n a id ed ASSR
m ea sured threshold s ra ng ed from 0 to 55 d B HL. The m ea n a id ed ASSR
p red ic ted threshold s ra ng ed from 25 to 40.8 d B HL a nd the m ea n a id ed
b eha viora l threshold s ra ng ed from 34.2 to 40.8 d B HL. A rec og niza b le
d ifferenc e w a s noted b etw een the m ea n una id ed a nd m ea n a id ed ASSR
threshold s. When using the m ea sured va lues, a n a vera g e d ifferenc e of
b etw een 20 to 40 d B a c ross the freq uenc y ra ng e w a s noted . When
looking a t the d ifferenc e in the p red ic ted va lues, the d ifferenc es b etw een
the a id ed a nd una id ed va lues w ere 45 to 60 d B.
When c om p a ring the mea n a id ed ASSR m ea sured threshold s w ith the
a id ed b eha viora l threshold s, a d ifferenc e of 15.8 d B w a s noted b etw een
the a id ed ASSR m ea sured threshold s a nd the b eha viora l threshold a t 500
Hz. A d ifferenc e of 9.2 d B w a s noted a t 1000 Hz, 13.4 d B a t 2000 Hz a nd 16
d B a t 4000 Hz. In this c om p a rison the a id ed b eha viora l threshold s ha d the
low er va lue.
When c om p a ring the mea n a id ed ASSR – using the p red ic tion va lues w ith
the a vera g e a id ed b eha viora l threshold s, a 9.2 d B d ifferenc e w a s noted
a t 500 Hz, 5.8 d B a t 1000 Hz, no d ifferenc e a t 2000 Hz a nd 4 d B d ifferenc es
a t 4000 Hz. In this c a se the ASSR ha d the low er va lues for the freq uenc ies
tested .
Com p a ring the a id ed ASSR m ea sured threshold s w ith a id ed b eha viora l
threshold s for a ll the freq uenc ies tested , it w a s noted tha t only 8 out of 19
c om p a ra b le a id ed threshold s c orresp ond ed w ithin 10 d B of ea c h other.
These results a re rep resented in Fig ure 4.24.
153
Compa ra tive frequency
thresholds
University of Pretoria etd – Stroeb
bel, D (2006
10
8
6
4
2
0
” 10 d B
15 d B
20 d B
> 20 d B
Difference in dB
Figure 4.24 Representa tion of compa ra tive frequencies on a ided ASSR
mea sured thresholds a nd a ided beha viora l thresholds
Com p a ring the a id ed ASSR p red ic ted threshold s w ith a id ed b eha viora l
threshold s for a ll the freq uenc ies tested , it w a s noted tha t 11 of 19
c om p a ra b le a id ed threshold s c orresp ond ed w ithin 10 d B of ea c h other.
Compa ra tive frequency
thresholds
Fig ure 4.25 rep resents these results.
12
10
8
6
4
2
0
” 10 d B
15 d B
20 d B
> 20 d B
Difference in dB
Figure 4.25 Representa tion of compa ra tive frequencies on a ided ASSR
predicted thresholds a nd a ided beha viora l thresholds
154
University of Pretoria etd – Stroeb
bel, D (2006
Sta tistic a l a na lyses (Exa c t Wilc oxon Ra nk Sum Test) of the results for
d ifferenc es in a vera g e threshold s a s d eterm ined b y the Aid ed ASSR
m ea sured threshold s a nd b eha viora l threshold s a re sum m a rized in Ta b le
4.19. No
sta tistic a l d ifferenc e
b etw een a ny of the
a id ed
results
d eterm ined w ith these tw o p roc ed ures w a s found , a s a ll p -va lues w ere
m ore tha n 0.05. How ever the p -va lue on the 2000 Hz show ed a sm a ller
va lue tha n the other freq uenc y va lues. It w ould seem tha t a lthoug h the p va lue still ind ic a tes no sta tistic a lly sig nific a nt d ifferenc e, there seem s to b e
a
tend enc y tow a rd s a
d ifferenc e b eing
p resent on this sp ec ific
m ea surem ent. No a na lyses c ould b e m a d e a t 500 Hz a s resp onses on the
a id ed ASSR c ould b e m ea sured only on tw o sub jec ts.
Ta ble 4.19
Aided ASSR mea sured responses vs. a ided beha viora l
responses
Stimulus
P - va lue
•
Aided 500 Hz ASSR vs. a ided beha vioura l
threshold
•
Aided 1000 Hz ASSR vs. a ided beha vioura l
threshold
P = 0.1875
•
Aided 2000 Hz ASSR vs. a ided beha vioura l
threshold
P = 0.0625 **
•
Aided 4000 Hz ASSR vs. a ided beha vioura l
threshold
P = 0.1250
N.A. ( only 2 va lues)
N.A. not a pplica ble
** Tendency towa rd difference
The results p rovid ed b y the a id ed ASSR m ea sured threshold s a nd a id ed
b eha viora l threshold s w ere a lso c om p a red w ith reg a rd to its rela tion. The
follow ing sc a tter p lots in Fig ure 4.26 rep resent the rela tionship or
c orrela tion c oeffic ient b etw een ea c h freq uenc y tested d uring the a id ed
ASSR m ea sured
eva lua tion a nd
the sub seq uent a id ed
b eha viora l
155
University of Pretoria etd – Stroeb
bel, D (2006
m ea surem ent. A p ositive c orrela tion w a s noted on ea c h ind ivid ua l
freq uenc y tested . A m od era te to m a rked c orrela tion w a s noted a t 2000
Hz (r = .70) a nd a t 4000 Hz (r = .63). A c ha ng e rela tionship b etw een the
results of the p roc ed ures is ind ic a ted a t 1000 Hz (r = .07).
70
2000 Hz Aided Beha viora l
a ssessment
1000 Hz Aided Beha viora l
a ssessment
60
y = 0.0968x + 32.903
2
50
R = 0.0057
40
30
20
10
0
y = 0.6603x + 5.0685
2
R = 0.502
60
50
40
30
20
10
0
0
10
20
30
40
50
60
70
0
10
1000 Hz Aided ASSR mea sured response
20
30
40
50
60
70
80
2000 Hz Aided ASSR measured response
4000 Hz Aided Beha viora l
a ssessment
70
60
y = 0.4219x + 15.797
2
R = 0.3955
50
40
30
20
10
0
0
10
20
30
40
50
60
70
80
90
4000 Hz Aided ASSR mea sured response
Figure 4.26 Rela tionship between a ided beha viora l thresholds a nd a ided
ASSR mea sured responses ba sed on the mea surements for six
subjects
Sta tistic a l a na lyses (Exa c t Wilc oxon Ra nk Sum Test) of the results for
d ifferenc es in a vera g e threshold s a s d eterm ined b y the Aid ed ASSR
p red ic ted va lues a nd b eha viora l threshold s a re sum m a rized in Ta b le 4.20.
No sta tistic a l d ifferenc e b etw een a ny of the a id ed results d eterm ined w ith
these tw o p roc ed ures w a s found , a s a ll p -va lues w ere m ore tha n 0.05. No
156
University of Pretoria etd – Stroeb
bel, D (2006
a na lyses c ould b e m a d e a t 500 Hz a s resp onses on the a id ed ASSR c ould
b e m ea sured only w ith tw o sub jec ts.
Ta ble 4.20
Aided ASSR predicted responses vs. a ided beha viora l
responses
Stimulus
P - va lue
•
Aided 500 Hz ASSR vs. a ided beha vioura l
threshold
•
Aided 1000 Hz ASSR vs. a ided beha vioura l
threshold
P = 0.1249
•
Aided 2000 Hz ASSR vs. a ided beha vioura l
threshold
P = 0.2438
•
Aided 4000 Hz ASSR vs. a ided beha vioura l
threshold
P = 0.2504
N.A. ( only 2 va lues)
N.A. not a pplica ble
The c om p a rison w ith reg a rd to the rela tion b etw een the a id ed ASSR
p red ic ted threshold s a nd a id ed b eha viora l threshold s a re rep resented in
the follow ing sc a tter p lots. Fig ure 4.27 rep resent the rela tionship or
c orrela tion c oeffic ient b etw een ea c h freq uenc y tested d uring the a id ed
ASSR p red ic ted
eva lua tion a nd
the sub seq uent a id ed
b eha viora l
m ea surem ent. A p ositive c orrela tion w a s noted on ea c h ind ivid ua l
freq uenc y tested . A m a rked c orrela tion w a s noted a t 2000 Hz (r = .76) a nd
a t 4000 Hz (r = .61). A slig ht rela tionship b etw een the results of the
p roc ed ures in ind ic a ted a t 1000 Hz (r = .28).
157
University of Pretoria etd – Stroeb
bel, D (2006
70
2000 Hz Aided Beha viora l a ssessment
1000 Hz Aided Beha viora l a ssessment
60
y = 0.3329x + 26.765
R2 = 0.0803
50
40
30
20
10
y = 0.6392x + 14.733
R2 = 0.5787
60
50
40
30
c
20
10
0
0
0
5
10
15
20
25
30
35
40
45
0
50
10
1000 Hz Aided ASSR predicted measurement
20
30
40
50
60
70
2000 Hz Aided ASSR predicted measurement
4000 Hz Aided Beha viora l a ssessment
70
y = 0.3094x + 28.015
2
R = 0.3664
60
50
40
30
20
10
0
0
10
20
30
40
50
60
70
80
4000 Hz Aided ASSR predicted measurement
Figure 4.27 Rela tionship between a ided beha viora l thresholds a nd a ided
ASSR predicted responses ba sed on the mea surements for six
subjects
To sum m a rize:
•
The sa m e a m ount of freq uenc ies tested , d em onstra ted results w ith
a id ed ASSR m ea sured a nd the a id ed ASSR p red ic ted threshold s.
•
In only 42% of the a id ed freq uenc ies tested , the a id ed ASSR
m ea sured
threshold s
a nd
a id ed
b eha viora l
threshold s
c orresp ond ed w ithin 10 d B of ea c h other.
•
In 58% of the a id ed freq uenc ies tested , the a id ed ASSR p red ic ted
threshold s a nd a id ed b eha viora l threshold s c orresp ond ed w ithin 10
d B of ea c h other.
158
University of Pretoria etd – Stroeb
bel, D (2006
•
No sta tistic a lly sig nific a nt d ifferenc es w ere evid ent b etw een a id ed
a vera g es of ASSR m ea sured threshold s a nd a id ed b eha viora l
threshold s.
•
No sta tistic a l sig nific a nt d ifferenc es w ere evid ent b etw een a id ed
a vera g es of ASSR p red ic ted threshold s a nd a id ed b eha viora l
threshold s.
•
Results c onfirm tha t there is a m od era te to c ha ng e c orrela tion
b etw een the a id ed ASSR m ea sured resp onse a nd a id ed b eha viora l
a ssessm ent.
•
Results c onfirm tha t there is a fa ir to m od era te p ositive c orrela tion
b etw een the a id ed threshold s d eterm ined b y the b eha viora l
a ssessm ent a nd the ASSR p red ic ted resp onse.
Ana lysis of the d a ta led to c om p a ra tive results w hic h ind ic a ted tha t
b oth the a id ed ASSR (m ea sured a nd p red ic ted ) results c om p a re
fa vora b ly to tha t of a id ed b eha viora l a ssessm ents, a lthoug h there is a
hig her c orrela tion b etw een the a id ed ASSR p red ic tions a nd the a id ed
b eha viora l a ssessm ents.
4.4
DISCUSSION
The p urp ose of this stud y w a s to d eterm ine the c linic a l va lue of the ASSR
for ea rly d ia g nosis a nd a m p lific a tion of infa nts w ith hea ring loss. This w a s
d one b y using b oth ABR a nd ASSR m ea surem ents to p red ic t hea ring
threshold s a nd to c om p a re these results ob ta ined in infa nts w ith hea ring
loss. Aid ed ASSR threshold s w ere m ea sured in ord er to va lid a te the
hea ring
a id
fitting . These
results w ere
c om p a red
w ith b eha viora l
m ea surem ents. In the follow ing sec tions the results of this stud y w ill b e
d isc ussed a c c ord ing to the sub -a im s.
159
University of Pretoria etd – Stroeb
bel, D (2006
4.4.1 Sub-a im 1: To investiga te the potentia l clinica l va lue of the ASSR in
ea rly dia gnosis of hea ring loss in a group of infa nts by determining
a nd compa ring una ided ASSR, ABR a nd beha viora l thresholds.
Sub -a im 1 w ill b e d isc ussed in the follow ing sec tion.
4.4.1.1
ABR vs. ASSR
Althoug h som e d isc rep a nc ies w ere noted b etw een these tw o m ea suring
tec hniq ues in the ind ivid ua l sub jec ts (sub jec t 2 a nd sub jec t 4: tone b urst
ABR vs. 500 Hz ASSR a nd sub jec t 3: c lic k ABR a nd 2000 Hz ASSR), the results
from this stud y show tha t the tone b urst ABR a nd 500 Hz ASSR ha ve a
strong p ositive c orrela tion (r = .77). Sim ila r results w ere ob ta ined w ith the
c lic k ABR a nd 2000 Hz ASSR c om p a rison (r = .62). Johnson a nd Brow n (in
Va nd er Werff et a l., 2002:233), tested a sm a ll g roup of hea ring im p a ired
a d ults w ith a ra ng e of hea ring losses a nd c om p a red toneb urst ABR
threshold s w ith ASSR threshold s. These resea rc hers found a strong p ositive
c orrela tion of r = .91 b etw een ABR a nd ASSR threshold s. The stud y b y
Va nd er Werff et a l. (2002:233) a g rees w ell w ith the p revious stud y
m entioned . The stud y c ond uc ted b y Cone-Wesson et a l. (2002:184)
c onc lud ed tha t tone-ABR a nd ASSR c ould b oth b e used to estim a te
hea ring threshold s a s p ositive c orrela tions w ere found b etw een these tw o
m ea surem ents. This is c onfirm ed b y the results of the c urrent stud y. In this
c urrent stud y how ever, no c orrec tion w ere m a d e for the ABR results.
The p op ula tion for w hom ASSR threshold estim a tion p roc ed ures m a y
p rove p a rtic ula rly b enefic ia l is c hild ren w ith severe to p rofound hea ring
losses. The c ontinuous tones used to elic it the ASSR resem b le the stim uli
used in b eha viora l testing a nd c a n therefore p resented a t hig her levels
160
University of Pretoria etd – Stroeb
bel, D (2006
tha n the ABR. The ASSR is therefore w ell suited to q ua ntify hea ring loss in
the severe to p rofound ra ng e (Ra nc e et a l., 2005:298). This p resent stud y
rep orted on five ea rs for w hic h no c lic k ABR or 500 Hz tone b urst ABR w a s
rec ord ed a t the m a xim um stim ula tion levels. Four of these ea rs ha d
m ea sura b le ASSR threshold s a t 500 a nd 2000 Hz. Tw o ea rs a lso ha d
resp onses a t 4000 Hz on the ASSR. Only one ea r ha d no resp onse on either
of the m ea surem ents. These find ing s of p otentia l a d va nta g es of ASSR over
ABR for severe to p rofound losses a re c onsistent w ith results of p reviously
rep orted stud ies (Ra nc e et a l., 2005:294; Sw a nep oel et a l., 2004:534;
Va nd er Werff, 2002:233; Ra nc e et a l., 1998:57; Ra nc e et a l., 1995:505).
These stud ies ha ve show n tha t error in p red ic tion of hea ring loss d ec rea ses
w ith inc rea sing d eg ree of hea ring loss. The evid enc e from this stud y further
ind ic a tes tha t a b sent ASSR im p lies no usa b le hea ring a t tha t freq uenc y.
Tha t is not true of ABR, for w hic h evid enc e ha s show n tha t a b sent ABR
d oes not rule out useful resid ua l hea ring (Ra nc e et a l., 1998:48).
Both the ASSR a nd tone b urst ABR ha ve d em onstra ted c linic a l va lue for
estim a ting the p ure-tone a ud iog ra m in infa nts w ith hea ring loss (ConeWesson et a l., 2002:185). The d a ta from this p resent stud y a nd those of
other stud ies (Stueve & O’ Rourke, 2003; Va nd er Werff, 2002) sug g est tha t
there a re no sig nific a nt d ifferenc es in threshold d eterm ina tion b etw een
the tw o tec hniq ues.
4.4.1.2
ASSR vs. Beha viora l m ea sures
The results from this stud y show tha t the ASSR p roc ed ure c a n a c c ura tely
id entify a nd q ua ntify hea ring loss in infa nc y. For these sub jec ts there w a s a
strong rela tionship b etw een the ASSR threshold s ob ta ined d uring infa nc y
a nd their sub seq uently esta b lished b eha viora l a ud iog ra m s. The d ifferenc e
161
University of Pretoria etd – Stroeb
bel, D (2006
b etw een the a vera g e ASSR threshold p red ic tion a nd the a vera g e
b eha viora l threshold w a s 0 – 10 d B (Fig ure 4.11), w ith c orrela tion va lues of
.93 a t 500 Hz; .82 a t 1000 Hz; .79 a t 2000 Hz a nd .59 a t 4000 Hz. In stud ies
tha t ha ve c om p a red the ASSR w ith b eha viora l threshold s, very strong
p ositive c orrela tions w ere a lso found b etw een these tw o m ea sures
(Ra nc e et a l., 2005:295).
In a stud y to d eterm ine the effec t of a ud iom etric c onfig ura tion on
threshold s a nd sup ra threshold ASSR, a hig hly sig nific a nt c orrela tion
b etw een p ure-tone b eha viora l a nd ASSR threshold s for ind ivid ua ls w ith
either slop ing or fla t a ud iom etric c onfig ura tions w a s revea led
(Va nd er
Werff & Brow n: 2005:319). In the p resent stud y of 12 ea rs, it w a s found tha t
the ASSR results w ere a c c ura te in d eterm ining the c onfig ura tion of the
loss. As w ith the stud y b y Ra nc e et a l. (1998:58) a nd Ra nc e et a l.
(2005:295), the find ing s for ind ivid ua l freq uenc ies tra nsla ted into a c c ura te
d esc rip tions of the sub jec ts’ hea ring losses. The d ifferenc e in threshold s
d iffered b etw een 0 – 20 d B, w ith the ASSR in m ost of the c a ses b eing
slig htly hig her tha n the b eha viora l threshold – esp ec ia lly in the low
freq uenc ies (exc lud ing sub jec t 2 & 3). Ra nc e et a l. (1998:58) found a
sim ila r p a ttern
overestim a ting
in
their sub jec ts w ith
the
the b eha viora l levels a nd
ASSR threshold s slig htly
m irroring the a ud iog ra m
c onfig ura tion. These find ing s a re sim ila r to the find ing s of Lins et a l.
(1996:95) w hen they found a sig nific a nt d ifferenc e in m ea n threshold a t
500 Hz in a g roup of a d ult sub jec ts. These resea rc hers a lso show ed a
g enera l tend enc y a c ross freq uenc y for ASSR threshold s in infa nts to b e
hig her tha n for a d ults.
In this p resent stud y, the results from sub jec t 2 show ed ina c c ura te
threshold s p red ic tions. The ABR threshold s a nd ASSR p red ic tion threshold s
162
University of Pretoria etd – Stroeb
bel, D (2006
c orrela ted w ell a t tim e of d ia g nosis, b ut sub seq uent b eha viora l threshold s
w ere 10 d B to 35 d B low er a t d ifferent freq uenc ies tha n p revious
elec trop hysiolog ic a l results. There is evid enc e of d eteriora tion in hea ring
level in this sub jec t. This a sp ec t is b eing eva lua ted further. The results from
the b eha viora l a ssessm ent im p a c t neg a tively on the results of this stud y.
The elec trop hysiolog ic a l a ssessm ent in sub jec t 3 ind ic a ted to a g rea ter
hea ring loss tha n w ha t w a s sub seq uently d eterm ined w ith b eha viora l
a ud iom etry. No a p p a rent rea son for these d isc rep a nc ies c ould b e found .
A p ossib le influenc e m a y b e the p resenc e of a b norm a l tuning c urves in
the c oc hlea , c a used b y im p a irm ent. Pic ton et a l. (1998:329) found tha t
the p resenc e of a b norm a l tuning c urves in the c oc hlea c a used the
im p a ired system to ha ve p la c e a nd freq uenc y sp ec ific ity d isc rep a nc ies.
This m ec ha nism m ig ht not lea d to w ell sync hronized stea d y sta te
resp onses a nd the p hysiolog ic threshold s m a y b e eleva ted rela tive to the
b eha viora l threshold s (Pic ton et a l., 1998:329).
The a ud iog ra m s show n of ea c h ind ivid ua l sub jec t a lso reflec ts one of the
p a rtic ula r a d va nta g es of the ASSR a ssessm ent in sub jec ts w ith m inim a l
a m ounts of resid ua l hea ring (sub jec t 5 & 6). The c ontinuous tones used to
elic it the ASSR resem b le the stim uli used to elic it b eha viora l resp onses a nd
c a n b e p resented a t hig her levels tha n is p ossib le for b rief stim uli. The ASSR
is therefore esp ec ia lly w ell suited for q ua ntifying hea ring loss of a severe to
p rofound na ture (Ra nc e et a l., 2005:298). Of the five hund red a nd fifty-six
sub jec ts w ith either norm a l hea ring or sensorineura l hea ring loss, only four
show ed ASSR threshold s a t levels > 10 d B low er tha n their sub seq uently
esta b lished b eha viora l threshold s.
In a rec ent stud y Pic ton, Dim itrijevic , Perez-Ab a lo a nd Va n Roon
(2005:154) c onc lud ed their rep ort b y sta ting tha t the a c c ura c y of
163
University of Pretoria etd – Stroeb
bel, D (2006
threshold estim a tion d ep end s on the va ria b ility of threshold estim a tion
ra ther tha n on a ny m ea n d ifferenc e b etw een p hysiolog ic a l a nd
b eha viora l threshold s. The results from their exp erim ents d em onstra ted
severa l w a ys to im p rove the a c c ura c y of estim a ting b eha viora l threshold s
from the ASSR – the m a in fa c tor b eing to red uc e b a c kg round noise. With
tim e lim ita tion in the c linic a l setting , the results w ill typ ic a lly b e threshold s
tha t ha ve a sta nd a rd d evia tion of 10 d B – w hic h is sim ila r to the va ria b ility
ob ta ined using tone b urst ABR (Sta p ells, 2000b :74).
This p resent stud y sup p orts the find ing s of the p revious stud ies. The ASSR
a ssessm ent d em onstra tes the c linic a l va lue for estim a ting the p ure-tone
a ud iog ra m in infa nts w ith hea ring loss. It c a n thus b e seen a s a very useful
step in the eva lua tion p roc ess for these ea rly-id entified infa nts – a llow ing
the b eha viora l a ud iog ra m to b e p red ic ted a nd intervention p roc esses to
b e im p lem ented .
4.4.1.3
ABR vs. Beha viora l m ea sures
This stud y a lso show s tha t rea sona b ly a c c ura te estim a tes of 500 Hz a nd
2000 a nd 4000 Hz p ure tone b eha viora l threshold s c a n b e ob ta ined b y
rec ord ing tone b urst ABR a nd c lic k ABR. A m a rked c orrela tion (r = .77) w a s
found b etw een the 500 Hz b eha viora l a ssessm ent a nd tone b urst ABR.
Sim ila r find ing s a re w ell rep orted on in severa l stud ies (Sta p ells, Gra vel &
Ma rtin, 1995:361; Sta p ells, 2000a :20; Gorg a , 1999:29). The c lic k ABR
show ed a d ep end a b le c orrela tion w ith the 2000 Hz b eha viora l a ssessm ent
a nd only a fa ir d eg ree of p ositive c orrela tion w ith 4000 Hz b eha viora l
a ssessm ent. This find ing a g rees w ith the notion tha t the c lic k ABR threshold
rep resents hea ring in the 2000 to 4000 Hz freq uenc y reg ion.
164
University of Pretoria etd – Stroeb
bel, D (2006
How ever it w a s evid ent tha t the severe to p rofound sensory neura l
hea ring loss w ill not b e id entified a nd eva lua ted w ith the ABR. This w a s
evid ent a s only seven of the 12 ea rs c ould b e eva lua ted using the ABR.
These find ing s a re c onsistent w ith p reviously rep orted stud ies (Ra nc e et a l.,
2005; Va nd er Werff et a l., 2002; Ra nc e et a l., 1998; Ra nc e et a l., 1995).
Sum m a ry:
•
This stud y c onc lud es tha t b oth the ABR a nd ASSR c a n b oth b e used
to estim a te hea ring threshold s – a s p ositive c orrela tions w ere found
b etw een these tw o m ea surem ents. How ever the ASSR p roved to
b e m ore b enefic ia l in the severe to p rofound hea ring loss
p op ula tion to q ua ntify their hea ring losses.
•
This stud y ind ic a tes tha t the ASSR p roc ed ure c a n a c c ura tely
id entify a nd q ua ntify hea ring loss in infa nts a s a strong rela tionship
w a s noted b etw een the ASSR threshold s ob ta ined d uring infa nc y
a nd their sub seq uently ob ta ined b eha viora l a ud iog ra m s.
•
Althoug h the tone b urst ABR a nd c lic k evoked ABR ind ic a ted to
p rovid e rea sona b ly a c c ura te estim a tes of the 500 Hz, 2000 Hz a nd
4000 Hz b eha viora l a ud iog ra m , it w a s evid ent tha t the severe to
p rofound sensory neura l hea ring losses w ill not b e id entified a nd
eva lua ted throug h the use of the ABR.
4.4.2 Sub-a im 2: To investiga te the clinica l va lue of the ASSR for releva nt
ea rly fitting of hea ring a ids in infa nts by determining a nd compa ring
a ided ASSR a nd a ided beha viora l thresholds.
Sub -a im 2 w ill b e d isc ussed in the follow ing sec tion.
165
University of Pretoria etd – Stroeb
bel, D (2006
4.4.2.1
Una id ed ASSR vs. a id ed ASSR resp onses
All of the sub jec ts show ed rec og niza b le a id ed ASSR resp onses a b ove their
una id ed ASSR threshold s. In this stud y b oth the m ea sured ASSR a nd the
ASSR using the p red ic tion form ula e d evised b y Melb ourne University
(Ra nc e et a l., 1995) w a s used . On the m ea sured ASSR, the resp onses w ere
w ithin 20 - 40 d B a b ove the una id ed m ea sured ASSR. Using the p red ic tion
form ula e, the a vera g e d ifferenc e b etw een the una id ed ASSR a nd a id ed
ASSR w a s 45 – 60 d B. In only tw o sub jec ts how ever c ould a n a id ed ASSR
resp onse b e m ea sured a t 500 Hz a t the m a xim um outp ut of the sp ea ker
(77,7 d B).
The ina b ility in this p resent stud y to d eterm ine m ore a id ed ASSR threshold s
a t 500 Hz m ig ht b e exp la ined b y hea ring a id c ha ra c teristic s a nd the
outp ut of the c a lib ra ted sp ea ker. Aid ed hea ring threshold s a re lim ited b y
the outp ut of the hea ring a id (Ga rnha m et a l., 2000:277). Sa tura tion a nd
d istortion of the outp ut sig na l w hen using a hig h-intensity stim uli in
c onjunc tion w ith m od era te to hig h g a ins w ere rep orted in the stud y b y
Ga rnha m a nd his c ollea g ues. Distortion introd uc es a d d itiona l sid eb a nd
freq uenc ies to those in the inp ut stim uli, thus d ec rea sing the freq uenc y
sp ec ific ity of the resp onse – influenc ing the resp onse m ea surem ent of the
ASSR.
An a sp ec t tha t m ig ht ha ve p la yed a further role in the ina b ility to ob ta in
m ore a id ed ASSR threshold s a t 500 Hz m ig ht b e the test environm ent.
These m ea surem ents w ere ob ta ined in a q uiet room in the p ra c tic e of the
resea rc her. Results from p reviously rep orted stud ies (Perez-Ab a lo et a l.,
2001:210; Sw a nep oel, 2001:120; Lins et a l., 1996:95) ind ic a ted tha t a c oustic
a m b ient b a c kg round noise exerts a sig nific a nt influenc e on the ASSR
166
University of Pretoria etd – Stroeb
bel, D (2006
results. Noise levels w ithin this sp ec ific test room m a y not ha ve b een
suffic iently q uiet to esta b lish threshold s in the sound field a t 500 Hz (Ha rrell,
2002:75).
Another p ossib le influenc e m a y b e the p resenc e of a b norm a l tuning
c urves in the c oc hlea , c a used b y im p a irm ent. Pic ton et a l. (1998:329)
found tha t the p resenc e of a b norm a l tuning c urves in the c oc hlea
c a used the im p a ired system to ha ve p la c e a nd freq uenc y sp ec ific ity
d isc rep a nc ies. Desp ite a m p lific a tion, the sound s - in this insta nc e 500 Hz m a y b e p roc essed throug h a rea s of the c oc hlea tha t a re not p la c e
sp ec ific for 500 Hz. This m ec ha nism m ig ht not lea d to w ell sync hronized
stea d y sta te resp onses a nd the p hysiolog ic threshold s m a y b e eleva ted
rela tive to the b eha viora l threshold s (Pic ton et a l., 1998:329).
4.4.2.2
Aid ed ASSR resp onses vs. a id ed b eha viora l resp onses
In the g roup of six sub jec ts, the a id ed ASSR m ea sured resp onses w ere on
a vera g e b etw een 9.2 d B a nd 16 d B hig her tha n the a id ed b eha viora l
threshold s. These results a re sim ila r to the d ifferenc es rep orted b y Pic ton et
a l. (1998:327), w here the a id ed ASSR resp onses w ere on a vera g e b etw een
13 a nd 17 d B hig her tha n the b eha viora l threshold s. The Pic ton g roup of
resea rc hers investig a ted the p ossib le use of the MASTER (m ultip le a ud itory
stea d y-sta te resp onse) tec hniq ue in the a ssessm ent of a id ed threshold s in
the sound field on 38 c hild ren (a g es 11 – 17 yea rs) w ith hea ring
im p a irm ent. Most c hild ren in their stud y show ed rec og niza b le resp onses
w ithin 10 a nd 30 d B a b ove their b eha viora l threshold s w ith their hea ring
a id s. The p hysiolog ic threshold s w ere q uite c losely rela ted to the
b eha viora l threshold s exc ep t a t 4000 Hz w here there w a s a sig nific a ntly
g rea ter
va ria b ility
in
the
rela tion
b etw een
the
b eha viora l
a nd
167
University of Pretoria etd – Stroeb
bel, D (2006
p hysiolog ic a l threshold s. In severa l of the a id ed sub jec ts, no resp onses
w ere found a t 4000 Hz even w hen stim uli w ere sig nific a ntly a b ove
b eha viora l threshold s. Pic ton et a l. (1998:322) ob ta ined b etter threshold s –
using the sa m e stim uli – p resented sing ly. The rela tions b etw een the
p hysiolog ic a nd b eha viora l threshold s b ec a m e c loser. There w ere no
sig nific a nt d ifferenc es in the p hysiolog ic -b eha viora l d ifferenc es a m ong
the d ifferent a ud iom etric freq uenc ies. The c onc lusion for their find ing s w a s
tha t the p hysiolog ic -b eha viora l d ifferenc e w a s p rob a b ly rela ted to
rec ruitm ent – the resp onse rea c hes a level w here it is rec og niza b le a t
intensity c loser to threshold .
In the c urrent g roup of six sub jec ts, the a id ed ASSR p red ic ted threshold s
w ere on a vera g e b etw een 4 d B a nd 9.2 d B low er tha n the a id ed
b eha viora l threshold s. These results d iffer from the Pic ton g roup results
(1998:327); how ever, Pic ton et a l. (1998:327) d id not use p red ic tion
form ula e to d eterm ine a id ed ASSR threshold levels. The a id ed ASSR
threshold s in their rep ort w ere the a c tua l m ea sured threshold s – using the
MASTER.
Com p a ring the m ea n a id ed ASSR m ea sured a nd p red ic ted threshold s in
c om p a rison w ith the a id ed b eha viora l threshold s, it w ould seem a s if the
c orrela tion b etw een the a id ed ASSR p red ic ted va lues a nd a id ed
b eha viora l threshold va lues a re m ore p ositive b y a sm a ll m a rg in –
esp ec ia lly a t 1000 a nd 2000 Hz. How ever in the ind ivid ua l c a ses va ria tions
a re noted : the a id ed ASSR p red ic ted va lues for sub jec t 1, 3, 4 a nd 5
c losely a p p roxim a ted the a id ed b eha viora l threshold va lues m ore so tha n
the a id ed ASSR m ea sured va lues. The results from sub jec t 2 a nd 6
ind ic a ted to the ASSR m ea sured va lues to a p p roxim a te the a id ed
b eha viora l threshold s m ore c losely tha n the ASSR p red ic ted va lues.
168
University of Pretoria etd – Stroeb
bel, D (2006
The d ifferenc es b etw een ASSR (m ea sured threshold s) a nd b eha viora l
threshold s va ried b etw een 0 a nd 20 d B. The d ifferenc es b etw een ASSR
(p red ic ted va lues) a nd b eha viora l threshold s va ried b etw een 5 a nd 25
d B. The va ria nc e m ig ht b e exp la ina b le b y inter-sub jec t d ifferenc es.
Sub jec t 1, 2 a nd 6 show ed low er resp onses on the a id ed ASSR p red ic ted
va lues a s on the a id ed b eha viora l m ea surem ent. These w ere the sub jec ts
w ho w ere fitted w ith d ig ita l hea ring a id s. Sub jec t 3, 4 a nd 5 w ere fitted
w ith d ig ita lly p rog ra m m a b le hea ring a id s. Their a id ed ASSR p red ic ted
resp onses w ere b etw een 0 a nd 20 d B hig her tha n the a id ed b eha viora l
threshold s. This ra ng e is fa r from op tim a l; how ever w here there is no other
inform a tion
a b out
a id ed
threshold s,
this d eg ree
of
a c c ura c y
is
a c c ep ta b le (Pic ton et a l., 1998:327). It is c lea r how ever, tha t the results
from this stud y ind ic a te to the a id ed ASSR m ea sured resp onses to b e not
sp ec ific
enoug h
a nd
tha t
the
a id ed
ASSR p red ic ted
threshold s
overestim a te threshold s. New c orrec tion fig ures m a y b e need ed for the
ASSR to b e used for the p urp ose of estim a ting func tiona l g a in a nd la rg er
sc a le stud ies a re need ed to va lid a te this a p p roa c h.
This stud y ha s show n tha t a id ed ASSR a re va lua b le in the va lid a tion of the
a id ed p erform a nc e in som e sub jec ts a nd c a n p rovid e va lua b le func tiona l
inform a tion. In this g roup of six sub jec ts, it w a s the first c lea r resp onse
rec ord ed on these infa nts. It a lso c lea rly ind ic a ted p ossib le c oc hlea r
im p la nta tion c a nd id a c y for sub jec t 5 a nd 6 a nd c onfirm ed tha t the
una id ed threshold s w ere not b a sed solely on sp urious/ a rtific ia l AASR’ s a t
hig h intensity stim uli (Sm a ll & Sta p ells, 2004:611; Gorg a et a l., 2004:302;
Jeng et a l, 2004:67; Pic ton & John, 2004:541; Dillon, 2001:419). How ever,
som e lim ita tions to a id ed ASSR w ere found :
169
University of Pretoria etd – Stroeb
bel, D (2006
•
Only the linea r op era tion of the a id c a n b e tested (Ga rnha m et a l.,
2000:277). Ad va nc ed
p roc essing
fea tures, suc h a s feed b a c k
m a na g ing system s a nd noise red uc tion system s, w ere d ea c tiva ted
on the d ig ita l hea ring a id s.
•
Aid ed threshold s a re not uninform a tive – c lea rly if threshold s a re
b elow the sp eec h intensities, the a id c a nnot im p rove sp eec h
p erc ep tion. Pic ton et a l. (2002:68) c a utions tha t the a ssessm ent of
a id ed threshold s is oc c urring a t levels tha t a re not releva nt to the
p erc ep tion of a m p lified sp eec h. Ob jec tive a ssessm ent of hea ring
a id m ea surem ents a t c om fort levels m a y b e a m ore effic ient
a p p roa c h to fitting of hea ring a id s tha n d eterm ining a id ed
threshold s (Pic ton et a l., 1998:328).
Althoug h these lim ita tions a re p resent, a id ed ASSR’ s w ere found to b e
va lua b le – esp ec ia lly in the c a ses of sub jec t 5 a nd 6 w here c oc hlea r
im p la nt c a nd id a c y w a s d eterm ined a t suc h a young a g e. Aid ed ASSR
m ea sures m a y b ec om e m ore va lua b le a s the need a rise to d eterm ine
c oc hlea r im p la nta tion c a nd id a c y a t ea rlier a g es a nd to m a na g e infa nts
w ith hea ring loss m ore effec tively. How ever, w hen p erform ing a id ed
hea ring a id threshold m ea surem ents it is essentia l to b e a w a re of
lim ita tions in b oth the hea ring a id s a nd the stim uli used to evoke a
resp onse.
Sum m a ry:
•
All sub jec ts show ed rec og niza b le a id ed ASSR resp onses a b ove their
una id ed ASSR threshold s. There w a s a n ina b ility to d eterm ine a id ed
ASSR’ s a t 500 Hz in four sub jec ts.
170
University of Pretoria etd – Stroeb
bel, D (2006
•
In the g roup of six sub jec ts, the a id ed ASSR m ea sured threshold s
w ere on a vera g e b etw een 9.2 d B a nd 16 d B hig her tha n the a id ed
b eha viora l threshold s. The a id ed ASSR p red ic ted w ere on a vera g e
b etw een 4 d B a nd 9.2 d B low er tha n the a id ed b eha viora l
threshold s – ind ic a ting to the a id ed m ea sured threshold s to
und erestim a te b eha viora l threshold s a nd the a id ed p red ic ted
threshold s to overestim a te the a id ed b eha viora l threshold s.
4.5
CONCLUSION
The results from the c urrent stud y ind ic a te g ood c orrela tion b etw een the
ABR a nd ASSR a s m ethod to p red ic t hea ring threshold s in this g roup of
infa nts. The ASSR how ever d oes ha ve the a d va nta g e over the ABR in
ind ivid ua ls w ith a severe to p rofound hea ring loss. Resp onses c ould b e
m ea sured in these c a ses throug h the use of ASSR in the a b senc e of a ny
ABR resp onses. Furtherm ore, the a b senc e of ASSR resp onses a t m a xim um
levels w a s a relia b le ind ic a tor of p rofound or tota l hea ring loss. The ASSR
thus a llow ed for g rea ter d eg rees of hea ring im p a irm ent to b e eva lua ted .
The freq uenc y sp ec ific ity of the stim ulus tones a llow ed a ssessm ent of
resid ua l hea ring a c ross the a ud iom etric freq uenc y ra ng e.
The ASSR find ing s for ind ivid ua l freq uenc ies tra nsla ted into a c c ura te
d esc rip tions of the sub jec ts’ hea ring losses in c om p a rison w ith b eha viora l
threshold s. The c onfig ura tion of the hea ring loss c ould b e p red ic ted
throug h the use of ASSR. Results suc h a s these c a n p rovid e the b a sis for
ea rly intervention suc h a s fitting of hea ring a id s or d eterm ining c a nd id a c y
for c oc hlea r im p la nta tion.
171
University of Pretoria etd – Stroeb
bel, D (2006
Hea ring a id fitting in the infa nt p op ula tion rem a ins a c ha lleng e a nd a id ed
ASSR ha ve the p otentia l to p rovid e ob jec tive inform a tion w ith reg a rd s to
hea ring a id func tiona l b enefit in the va lid a tion p roc ess. Aid ed ASSR
threshold inform a tion is va lua b le a nd im p orta nt in the m a na g em ent of
c ha lleng ing c hild ren. In this stud y a id ed ASSR threshold s p rovid ed
a d d itiona l inform a tion.
It w ould therefore seem a s if the ASSR ha s g ot c linic a l va lue in the ea rly
d ia g nosis of hea ring loss in infa nts a s the una id ed ASSR va lues c orrela ted
w ell w ith the ABR a t the tim e of d ia g nosis a nd sub seq uently w ith the
una id ed b eha viora l threshold s.
Furtherm ore it w ould seem a s if the ASSR ha s a n a d d itiona l c linic a l va lue in
the va lid a tion of hea ring a id fitting s for infa nts a s the a id ed ASSR
m ea sured a nd p red ic ted va lues c orrela ted w ell w ith the a id ed b eha viora l
threshold s.
4.6
SUMMARY
This c ha p ter rep orted a nd d isc ussed the results ob ta ined in this stud y
a c c ord ing to the tw o sub -a im s. These sub -a im s w ere selec ted in a n
a ttem p t to a nsw er the m a in a im of this stud y. The results p erta ining to
ea c h sub -a im w ere d isc ussed a nd integ ra ted w ith litera ture to ensure the
va lid ity thereof. Conc lusions w ere d ra w n from the results in ea c h sub -a im
a nd sum m a rized a t the end of the c ha p ter in ord er to a nsw er the m a in
a im of the stud y.
172
University of Pretoria etd – Stroeb
bel, D (2006)
Cha p ter 5
CONCLUSIONS AND IMPLICATIONS
This c ha p ter a im s to d ra w g enera l c onc lusions a nd im p lic a tions from the
resea rc h, c ritic a lly eva lua te find ing s, a nd m a ke rec om m end a tions for
future resea rc h
5.1
INTRODUCTION
Within a rela tively short p eriod of tim e, there ha s b een rem a rka b le a nd
revolutiona ry c ha ng es in the field of p ed ia tric a ud iolog y tha t d em a nd
p rofessiona ls to rethink d ia g nostic a nd intervention p a ra d ig m s (KurtzerWhite & Luterm a n, 2001: introd uc tion). ‘ Evid enc e Ba sed Pra c tic e’ (EBP) is
therefore a n a p p roa c h to c linic a l servic e d elivery tha t ha s b ec om e
inc rea sing ly a d voc a ted (Gra vel, 2005:17). EBP refers to ‘ c onsc ientious,
exp lic it, a nd jud ic ious use of c urrent b est evid enc e in m a king d ec isions
a b out the c a re of p a tients’ (Oxford -Centre for Evid enc e Ba sed Med ic ine,
2004: online). The p rim a ry elem ent of EBP is the m a jor role of sc ientific
evid enc e in c linic a l d ec ision-m a king (Gra vel, 2005:17). This sentim ent ha s
b een the und erlying d riving forc e b ehind the resea rc h end ea vor of this
stud y.
There ha s a lw a ys b een a need for ob jec tive tests tha t a ssess a ud itory
func tion
in
infa nts,
young
c hild ren,
a nd / or
a ny
p a tient
w hose
d evelop m enta l level p rec lud ed the use of b eha viora l a ud iom etric
tec hniq ues (Gorg a & Neely, 2002:49). The ASSR ha ve therefore g a ined
173
University of Pretoria etd – Stroeb
bel, D (2006)
c onsid era b le a ttention a nd is seen a s a p rom ising a d d ition to the AEP
‘ fa m ily’ . This stud y p rop osed to g a ther evid enc e w ith reg a rd s to the
c linic a l va lue of the ASSR in infa nts. It is thus log ic a l to eva lua te ‘ b est
evid enc e’ throug h c ritic a l a p p ra isa l of this resea rc h end ea vor (Hill &
Sp ittlehouse, 2005:1). Critic a l a p p ra isa l is a n essentia l p a rt of evid enc eb a sed c linic a l p ra c tic e tha t inc lud es the p roc ess of system a tic a lly find ing ,
a p p ra ising a nd a c ting on evid enc e of effec tiveness. Critic a l a p p ra isa l is a
system a tic p roc ess, exa m ining resea rc h evid enc e to a ssess its va lid ity,
results a nd releva nc e. This p roc ess a llow s m a king sense of resea rc h
evid enc e a nd thus b eg ins to c lose the g a p b etw een resea rc h a nd
p ra c tic e (Hill & Sp ittlehouse, 2005:1).
The p urp ose of this c ha p ter is therefore to d ra w releva nt c onc lusions from
the results rep orted a nd d isc ussed in c ha p ter 4. A c ritic a l eva lua tion of the
stud y
is
sub seq uently
p rovid ed
to
id entify
the
inherent
a nd
m ethod olog ic a l lim ita tions of this stud y, follow ed b y rec om m end a tions for
future resea rc h. Fina lly a c onc lusion a nd sum m a ry of the c ha p ter is
p rovid ed .
5.2
CONCLUSIONS
The need for resea rc h to p rovid e evid enc e to justify c linic a l p ra c tic es is
a c know led g ed b y m ost c linic ia ns (Jenkins, Pric e & Stra ker, 2003:4). This
exp lora tory stud y w a s c ond uc ted a c c ord ing to tw o sub -a im s, w hic h
resulted in the sum m a rized c onc lusions tha t follow b elow .
174
University of Pretoria etd – Stroeb
bel, D (2006)
5.2.1 Sub-a im 1: To investiga te the potentia l clinica l va lue of the ASSR in
ea rly dia gnosis of hea ring loss in a group of infa nts by determining
a nd compa ring una ided ASSR, ABR a nd beha viora l thresholds
•
This stud y c onc lud ed tha t b oth the ABR a nd ASSR c ould b oth b e
used to estim a te hea ring threshold s – a s p ositive c orrela tions w ere
found b etw een these tw o m ea surem ents. How ever the ASSR
p roved to b e m ore b enefic ia l in the severe to p rofound hea ring loss
p op ula tion to q ua ntify their hea ring losses.
•
This stud y ind ic a ted tha t the ASSR p roc ed ure c a n a c c ura tely
id entify a nd q ua ntify hea ring loss in infa nts a s a strong rela tionship
w a s noted b etw een the ASSR threshold s ob ta ined d uring infa nc y
a nd their sub seq uently ob ta ined b eha viora l a ud iog ra m s.
•
Althoug h the tone b urst ABR a nd c lic k evoked ABR ind ic a ted to
p rovid e rea sona b ly a c c ura te estim a tes of the 500 Hz, 2000 Hz a nd
4000 Hz b eha viora l a ud iog ra m , it w a s evid ent tha t the severe to
p rofound sensory neura l hea ring losses w ill not b e id entified a nd
eva lua ted throug h the use of the ABR.
The ASSR ha s the p otentia l to p rovid e a c c ura te p red ic tions of the
b eha viora l a ud iog ra m a nd b e used suc c essfully w ith p op ula tions w ith
severe to p rofound losses.
175
University of Pretoria etd – Stroeb
bel, D (2006)
5.2.2 Sub-a im 2: To investiga te the clinica l va lue of the ASSR for releva nt
ea rly fitting of hea ring a ids in infa nts by determining a nd compa ring
a ided ASSR a nd a ided beha viora l thresholds
•
All sub jec ts show ed rec og niza b le a id ed ASSR resp onses a b ove their
una id ed ASSR threshold s. There w a s a n ina b ility to d eterm ine a id ed
ASSR’ s a t 500 Hz in four sub jec ts.
•
In the g roup of six sub jec ts, the a id ed ASSR m ea sured threshold s
w ere on a vera g e b etw een 9.2 d B a nd 16 d B hig her tha n the a id ed
b eha viora l threshold s. The a id ed ASSR p red ic ted w ere on a vera g e
b etw een 4 d B a nd 9.2 d B low er tha n the a id ed b eha viora l
threshold s – ind ic a ting to the a id ed m ea sured threshold s to
und erestim a te b eha viora l threshold s a nd the a id ed p red ic ted
threshold s to overestim a te the a id ed b eha viora l threshold s.
The ASSR ha s the p otentia l to d eterm ine a id ed ASSR threshold s. This
p roc ed ure c a n therefore b e used to d eterm ine func tiona l g a in a nd thus
p la y a role in the ong oing p roc ess of va lid a ting hea ring a id fitting s in
infa nts.
The ASSR, d esp ite som e lim ita tions id entified , d em onstra ted g rea t p rom ise
for ea rly d ia g nosis a nd a m p lific a tion of infa nts w ith hea ring loss. The
d isc ussions a c c ord ing to the sp ec ified sub -a im s, revea led va lua b le
theoretic a l a nd c linic a l im p lic a tions a nd m a d e rec om m end a tions for
p rotoc ols to serve a s a g uid e for future use of the ASSR in the c linic a l
setting .
176
University of Pretoria etd – Stroeb
bel, D (2006)
5.3
THEORETICAL AND CLINICAL IMPLICATIONS
A m a jor justific a tion for elec trop hysiolog ic a ud iom etry is tha t rea sona b le
m ea sures of hea ring threshold s in a freq uenc y sp ec ific m a nner c a n b e
ob ta ined in ord er to c onstruc t a n a ud iog ra m (Gold stein & Ald ric h
(1999:3). Neona tes p rovid e the p rim e exa m p le. At p resent the toneevoked ABR is the only tec hniq ue tha t c a n p rovid e b oth the a ir- a nd
b one- c ond uc tion results req uired for ea rly intervention for c hild ren w ith
c ond uc tive or sensorineura l hea ring loss. The tone-evoked ABR ha s
suffic ient resea rc h, c linic a l d a ta b a se, a nd c linic a l history to rec om m end it
a s the prima ry technique for threshold estim a tion in infa nts (Sta p ells,
2005:55).
This p resent stud y ha s p roved how ever tha t b oth the ASSR a nd ABR
d em onstra ted effic a c y for estim a ting the p ure-tone a ud iog ra m in infa nts
w ith hea ring loss. No sig nific a nt d ifferenc e in threshold d eterm ina tion w a s
found b etw een these tw o tec hniq ues. The ASSR d id how ever ha ve the
a d va nta g e over the ABR in d eterm ining resid ua l hea ring in the severe to
p rofound g roup .
It is therefore evid ent tha t b oth tec hniq ues ha ve its ow n a d va nta g es a nd
its d isa d va nta g es. As ind ic a ted b y the review of the c urrent litera ture, the
evid enc e is la c king a nd not yet suffic ient to rec om m end the ASSR a s the
prima ry elec trop hysiolog ic
m ea sure of hea ring
in infa nts (Sta p ells,
2005:56). These tw o tec hniq ues should p rob a b ly b e used in c onjunc tion
w ith ea c h other (Ha ll, 2005: c onferenc e p resenta tion). Jerg er & Ha yes
(1976) in Diefend orf (2002:473) p rom oted the c onc ep t of a test b a ttery
a p p roa c h so tha t no sing le test w ill b e interp reted in isola tion, b ut va rious
tests a c t a s a c ross-c hec k on the fina l outc om e. Ina p p rop ria te or
177
University of Pretoria etd – Stroeb
bel, D (2006)
inc om p lete
d ia g nostic
c onc lusions
w ill
lea d
to
ina p p rop ria te
m a na g em ent a nd the c onseq uenc es thereof w ill b e w ith the c hild forever
(Seew a ld , 2001:70). By using these tec hniq ues in c om b ina tion, a m ore
solid found a tion for intervention w ill b e p rovid ed .
When c onsid ering tw o of the m ost im p orta nt ‘ truths’ in EBP (Oxford -Centre
for Evid enc e Ba sed Med ic ine: online), na m ely:
•
Pra c tic e m ust a lw a ys b e c onsid ered in view of the need s, c ulture
a nd p referenc es of the ind ivid ua l;
•
There is the rea l p rob a b ility tha t som e of the evid enc e-b a se
sup p orting c urrent p ra c tic e w ill c ha ng e or, ind eed , b e entirely
refuted b y evid enc e tha t w ill em erg e in the future,
there is a need to c ontinua lly re-exa m ine the c urrent a p p roa c h to
eva lua te hea ring a b ilities in infa nts.
The ASSR a nd ABR p resent w ith uniq ue q ua lities tha t c a n b e c om b ined to
p rovid e c om p lem enta ry results, w hic h w ill serve to verify results ob ta ined
w ith ea c h p roc ed ure (Sw a nep oel, 2001:114). Tim e is lim ited w hen w orking
w ith infa nts. It is therefore essentia l to use a test p rotoc ol tha t is fa st,
effic ient, a nd
one
tha t p rovid es the g rea test a m ount of c linic a l
inform a tion w ith ea c h suc c essive step ta ken (Sta p ells, 2002a :14) for ea c h
ind ivid ua l infa nt (Oxford -Centre for Evid enc e Ba sed Med ic ine: online).
Althoug h Sta p ells (2004: c onferenc e) ha s c a lled for the c lic k ABR to b e
a b olished , the c lic k ABR ha s p roven itself over the la st three d ec a d es a s a
relia b le p red ic tor of a ud itory sensitivity in the hig h freq uenc y reg ion
d esp ite its la c k of freq uenc y-sp ec ific ity (Sw a nep oel, 2001:115). It ha s
178
University of Pretoria etd – Stroeb
bel, D (2006)
rem a ined the m ost c om m only used elec trop hysiolog ic m ea sure b ec a use
of the c lea r resp onse, the hig h rep rod uc ib ility a nd sta b ility of the resp onse
(Arnold , 2000:455). The c lic k ABR is a lso the only tec hniq ue a t p resent to
a ssess the p resenc e of a ud itory neurop a thy (AN) – a lso know n a s a ud itory
d ys-sync hrony
(Tha rp e
&
Ha ynes,
2005:271).
Both
p roc ed ures
a p p roxim a ted the b eha viora l threshold s w ell in this stud y – how ever the
ASSR a p p roxim a ted b eha viora l threshold s c loser tha n the ABR (g roup
results). This a sp ec t w a s influenc ed b y the fa c t tha t few er ea rs c ould b e
tested w ith the ABR tha n w ith the ASSR. Althoug h a d d itiona l resea rc h on
ASSR testing in infa nts w ith hea ring loss is need ed (Sta p ells, 2005:55), b y
using the ASSR in a d d ition to the ABR, useful inform a tion m a y a lrea d y b e
p rovid ed to help d isting uish b etw een infa nts w ith severe a nd p rofound
losses (Roush, 2005:105).
These results sug g est a test-b a ttery a p p roa c h to ob jec tive a ud iom etry.
These tw o tec hniq ues a re ind ep end ent m ea sures of a ud itory sensitivity
tha t a re a b le to p rovid e d ifferent, thoug h c om p lem enta ry inform a tion.
The need s a nd p referenc e of ea c h infa nt w ill b e a c c om m od a ted b y
using this test-b a ttery a p p roa c h. Not only w ill a c ross-c hec k p rinc ip le b e
a d va nta g eous to ea c h ind ivid ua l infa nt, b ut the sp ec ific a d va nta g es of
ea c h p roc ed ure w ill g ive the m ost c om p rehensive a ssessm ent nec essa ry
to ensure tha t a true reflec tion of ea c h infa nt’ s a ud itory sta tus is a va ila b le
from w hic h reha b ilita tive d ec isions c a n b e m a d e (Roush, 2005:105).
After hea ring loss is d ia g nosed , fitting of hea ring instrum ents c a n oc c ur
w hen infa nts a re a s young a s five w eeks old (Yoshina g o-Ita no, 2004:451).
Ob jec tive m ea sures suc h a s AEP’ s offer the p ossib ility of eva lua ting the
effec tiveness of hea ring instrum ents in infa nts. This p resent stud y d id not
eva lua te the ABR’ s a b ility to d eterm ine hea ring instrum ent effec tiveness
179
University of Pretoria etd – Stroeb
bel, D (2006)
a s the litera ture ha s show n tha t the b rief stim uli tha t a re op tim a l for ABR
rec ord ing s m a y b e c onta m ina ted b y stim ulus a rtifa c ts. This sp ec ific
p roc ed ure w a s a lso seen a s c om p lic a ted a nd a ttem p ts to use the ABR to
eva lua te hea ring instrum ents ha ve la rg ely b een a b a nd oned (Purd y,
2005:116). This stud y ind ic a ted to the ASSR b eing a relia b le m ethod to
d eterm ine a id ed threshold s to ensure a ud ib ility of sp eec h sound s. The
results from the a id ed ASSR m a y sug g est the need to c onsid er a lterna tive
m a na g em ent – suc h a s in the c a se of tw o sub jec ts in this stud y w ho b oth
ha d p rofound sensory neura l hea ring losses a nd w ere fitted w ith hig hp ow ered
hea ring
a id s.
The
d ec ision
to
p roc eed
w ith
c oc hlea r
im p la nta tion w a s exp ed ited . The id ea tha t the ASSR c a n b e used to
va lid a te hea ring instrum ent fitting s is rea sona b le, b ut is yet to b e va lid a ted
a s a p roc ed ure.
5.4
CRITICAL EVALUATION OF THE CURRENT STUDY
Critic a l a p p ra isa l of a n em p iric a l resea rc h end ea vor is essentia l to
d eterm ine the va lue of the results ob ta ined a nd is a n essentia l p a rt of
evid enc e-b a sed c linic a l p ra c tic e. Relia b ility a nd va lid ity of the results a s
w ell a s the influenc e of id entified lim ita tions, inherent to the stud y, is
req uired to ensure the a p p rop ria te interp reta tion thereof. Severa l a sp ec ts
d eserving c ritic a l a p p ra isa l w ill b e d isc ussed in the follow ing p a ra g ra p hs.
The first a sp ec t to b e c onsid ered is the sa m p ling size of the c urrent stud y.
The b a sic rule is, the la rg er the sa m p le, the b etter (Leed y & Orm rod ,
2005:207). The sa m p ling size nec essa ry for a stud y d ep end s on the typ e of
stud y a nd is req uired to p rovid e a rep resenta tive p op ula tion from w hic h
inferenc es c a n b e d ra w n reg a rd ing a sp ec ific p henom enon in a sp ec ific
p op ula tion. Althoug h the sa m p le in the c urrent stud y w a s rep resenta tive
180
University of Pretoria etd – Stroeb
bel, D (2006)
of b oth sexes a nd c overed a ra ng e of a g es in infa nts, the sa m p le size w a s
not sig nific a ntly rep resenta tive of hea ring im p a ired infa nts. This w a s
how ever a n exp lora tory stud y – only the sec ond rep orting on a id ed
ASSR’ s a nd the first of its na ture on infa nts.
The sec ond a sp ec t tha t need s to b e ta ken into c onsid era tion is the test
environm ent. All b eha viora l threshold s (a id ed
a nd
una id ed ) w ere
ob ta ined in a d oub le w a lled , sound -a ttenua ted b ooth, w hile the
elec trop hysiolog ic a l a ssessm ents w ere c om p leted in a q uiet room w ithout
a ny sound a ttenua tion. The a c oustic a l a m b ient b a c kg round noise levels
w ere not m ea sured a nd therefore d id not a llow for c om p a rison b etw een
a c oustic noise levels b etw een the d oub le-w a lled , sound -a ttenua ted
b ooth a nd the q uiet room . The p ossib le d ifferenc e w a s not c onsid ered
w hen interp reting the results. This noise fa c tor m ig ht ha ve p la yed a role –
esp ec ia lly in ob ta ining a id ed ASSR results. Hig her levels of a m b ient
a c oustic noise in the q uiet room m ig ht ha ve c a used eleva ted threshold s
a nd the a b senc e of the rep orted a id ed 500 Hz ASSR threshold s. Thus the
threshold d ifferenc es c ould b e infla ted on a c c ount of the va ria b ility in the
test environm ents (Perez-Ab a lo et a l., 2001:210; Sw a nep oel, 2001:120; Lins
et a l., 1996:95).
A third a sp ec t id entified in the c ritic a l a p p ra isa l of the c urrent stud y is the
la c k of test-retest relia b ility m ea sures. Ac c ord ing to Sta p ells (2000a :13),
one of the lim ita tions w ith the ABR is the ina p p rop ria te interp reta tion of
w a veform s. A w a y to im p rove relia b ility of a test is to ha ve tw o
a d m inistra tors c orrela ting the results of the sa m e p roc ed ure. This m a y b e
of va lue in b oth the interp reta tion of ABR a nd b eha viora l threshold
a ssessm ent. The resp onses m ea sured d uring this stud y, w a s interp reted b y
the resea rc her a lone.
181
University of Pretoria etd – Stroeb
bel, D (2006)
A fourth a sp ec t tha t need s to b e ta ken into c onsid era tion, is the fa c t tha t
a c lic k-evoked ABR a nd only a 500 Hz tone b urst w ere used to c om p a re
w ith the ASSR. Na rrow freq uenc y reg ions (ASSR) w ere therefore c om p a red
w ith those from b roa d a nd unc erta in freq uenc y reg ions (c lic k ABR).
Id ea lly a c om p a rison should b e m a d e b etw een the infa nts’ ASSR
threshold s to their tone-evoked ABR – the c urrent ‘ g old sta nd a rd ’ infa nt
threshold m ea sure (Hyd e, 2005:287; Sta p ells, 2002:14).
The c ritic a l eva lua tion of the litera ture, c urrent stud y a nd c onsid era tion of
sig nific a nc e of the results ob ta ined
ha s revea led
future resea rc h
im p lic a tions tha t a re d isc ussed in the follow ing p a ra g ra p h.
5.5
RECOMMENDATION FOR FUTURE RESEARCH
Clea rly, there is a n im p orta nt role for the ASSR in estim a ting hea ring
threshold s a nd va lid a ting hea ring a id fitting s of infa nts. How ever, a
resea rc h q uestion a nsw ered ra ises new q uestions to b e a nsw ered . The
results ob ta ined in a nd c onc lusions d ra w n from this p resent resea rc h
end ea vor, revea led a sp ec ts tha t req uire further investig a tion. These a re
p resented to p rovid e sug g estions for future resea rc h end ea vors.
In ord er to va lid a te the ASSR p roc ed ure in the infa nt p op ula tion, it w ill b e
of va lue to c om p a re the ASSR ob ta ined a t a ll freq uenc ies, w ith tone b urst
ABR – using d ifferent freq uenc y tone b ursts. This d a ta w ill not only p rovid e
c om p a ra tive d a ta to the a c c ura c y of threshold d eterm ina tion, b ut a lso
relia b ility a nd tim e-effic ienc y of ea c h p roc ed ure.
In ord er to further va lid a te the ASSR p roc ed ure in the infa nt p op ula tion, it
w ill b e of va lue to d eterm ine b one c ond uc tion ASSR. By d eterm ining the
182
University of Pretoria etd – Stroeb
bel, D (2006)
BC ASSR, p ossib le m id d le ea r involvem ent w ill b e ruled out d uring the
a ssessm ent a nd a true p ic ture of the hea ring loss w ill em erg e.
Althoug h the va st m a jority of resea rc h ha s foc used on threshold
d eterm ina tion a nd op tim a l d etec tion stra teg ies, this p resent stud y a nd a
stud y from Pic ton (1998) exp lored the use of ASSR a nd hea ring a id
p erform a nc e. The results from this stud y a re very p rom ising , b ut the
p roc ed ures need to b e va lid a ted on a la rg er g roup of infa nts a s w ell a s
on c hild ren of other a g es – a s this p roc ed ure w ill p rob a b ly b e of use to the
d iffic ult-to-test p op ula tion, inc lud ing old er c hild ren w ith d evelop m enta l
d ela ys. Different p red ic tion form ula e m ig ht a lso b e nec essa ry to b e
d evelop ed for the a p p lic a tion of the ASSR for this p urp ose.
An a id ed threshold sup p lies c erta in inform a tion a b out a ud ib ility of sound s,
b ut no inform a tion a b out p erc ep tion of sound s is g iven. Stud ies b y
Dim itrijevic et a l. (2004:68) used the ASSR to p red ic t sup ra threshold
a ud itory a b ilities suc h a s w ord
d isc rim ina tion. Multip le
c a rriers of
ind ep end ently m od ula ted freq uenc y a nd a m p litud e (‘ IAFM’ ) stim uli ha ve
b een m od eled to ha ve sim ila r a c oustic sp ec tra to sp eec h. Using these
sp eec h-m od eled
stim uli,
sig nific a nt
c orrela tions
b etw een
w ord
d isc rim ina tion a nd d etec tion of IAFM w ere found in norm a l-hea ring a nd
hea ring -im p a ired sub jec ts (Dim itrijevic et a l., 2004:84). Althoug h ASSR’ s
rep resent a rela tively low level of a ud itory p roc essing , IAFM m a y b e used
to d eterm ine w hether or not the a ud itory system ha s suffic iently p roc essed
the nec essa ry inp ut req uired for sp eec h p erc ep tion a t a la ter a nd hig her
level of p roc essing (Sta p ells, 2005:56).
183
University of Pretoria etd – Stroeb
bel, D (2006)
5.6
CONCLUSION
AEP’ s a re a n id ea l tool for investig a ting a ud itory func tion in young infa nts,
a s they p rovid e a n ob jec tive m ea sure of the b ra in’ s resp onse to sound
(Purd y
et
a l.,
2005:115).
Rec ent
tec hnolog ic a l
a nd
resea rc h
a d va nc em ents ha ve a id ed the d evelop m ent of this field , ensuring the
c ontinua tion of end ea vors g enera ting tec hniq ues tha t a p p roxim a te the
a c c ura c y,
relia b ility,
freq uenc y-sp ec ific ity
a nd
tim e
effic ienc y
of
b eha viora l p ure tone a ud iom etry (Sw a nep oel, 2001:121) – b oth una id ed
a nd a id ed .
This investig a tion of the c linic a l va lue of the ASSR in infa nts ha s
d em onstra ted the ASSR’ s a b ility to estim a te b eha viora l p ure tone
threshold s rea sona b ly w ell. It ha s a lso show n tha t the ASSR ha s the
p otentia l to p la y a role in the ong oing p roc ess of hea ring instrum ent fitting
in infa nts a s a id ed ASSR threshold s c om p a red rea sona b ly w ell w ith a id ed
b eha viora l threshold s. How ever, w hile a d d itiona l resea rc h on ASSR testing
in infa nts w ith hea ring loss is need ed , it is im p orta nt to c ritic a lly c onsid er
c urrently a va ila b le p roc ed ures a long sid e the new . In his c losing a d d ress of
A Sound Found a tion throug h Ea rly Am p lific a tion c onferenc e in 1998 Bess
c ha lleng ed the c linic ia ns to b ec om e m ore evid enc e b a sed w ith the
follow ing w ord s: ‘ Effec tive c linic ia ns p rod uc e im p roved tec hniq ues a nd
c onsta ntly q uestion a nd eva lua te evid enc e, m ethod s, a nd p roc ed ures,
d isc a rd ing the unp rod uc tive, a nd d evelop ing a nd testing the new ’ (Bess,
2000:250).
This b ec om es essentia l in ord er to im p lem ent tec hniq ues in a c c ord a nc e
to the a d va nta g es a nd d isa d va nta g es of ea c h p roc ed ure. Evid enc e from
the c urrent stud y ind ic a ted tha t the ASSR p resented w ith uniq ue
184
University of Pretoria etd – Stroeb
bel, D (2006)
c ha ra c teristic s tha t should b e inc orp ora ted in a test-b a ttery a p p roa c h
a nd therefore ha s c linic a l va lue for ea rly d ia g nosis a nd a m p lific a tion of
infa nts w ith hea ring loss.
‘ if w e truly d esire to a fford the b est p ossib le servic es to c hild ren a nd their
fa m ilies, w e m ust b e w illing to c ontinua lly m od ify our c linic a l p rotoc ols a s
new evid enc e em erg es ’ (Bess, 2000:250)
185
University of Pretoria etd – Stroeb
bel, D (2006)
REFERENCE LIST
Alc a ntra , J.L., Moore, B.C., Kuhnel, V., & La uner, S. 2003. Eva lua tion of the
Noise Red uc tion System in a Com m erc ia l Dig ita l Hea ring Aid .
Interna tiona l Journa l of Aud iolog y 42(1): 34-42.
Am eric a n Ac a d em y of Aud iolog y, 2003: Ped ia tric Am p lific a tion Protoc ol.
Arnold , S.A. 2000. The Aud itory Bra in Stem Resp onse. In: Roeser, R.J.,
Va lente, M., Hosfored -Dunn, H. (ed s.), Aud iolog y Dia g nosis (p p . 451470). New York: Thiem e.
Ba b b ie, E. 1992. The Pra c tic e of Soc ia l Resea rc h. Ca lifornia : Wa d sw orth
Pub lishing Com p a ny.
Ba b b ie, E., a nd Mouton, J. 2002. The Pra c tic e of Soc ia l Resea rc h. Ca p e
Tow n: Oxford University Press Southern Afric a .
Ba c hm a nn, K.R., a nd Ha ll III, J.W. 1998. Ped ia tric Aud itory Bra instem
Resp onse Assessm ent: The Cross-c hec k Princ ip le Tw enty yea rs la ter.
Sem ina rs in Hea ring 12 (1): 41–59.
Ba iley, K.D. 1982. Method s of Soc ia l Resea rc h. Lond on: Collier Ma c m illa n
Pub lishers.
Bea uc ha ine, K.L. 2002. An Am p lific a tion Protoc ol for Infa nts. In: R.C.
Seew a ld a nd J.S. Gra vel (ed s.), A Sound Found a tion throug h Ea rly
Am p lific a tion: Proc eed ing s of the Sec ond Interna tiona l Conferenc e
(p p . 105-112). Stä fa , Sw itzerla nd : Phona k AG.
1
University of Pretoria etd – Stroeb
bel, D (2006)
Bea uc ha ine, K.L., a nd Donog hy, K.F. 1996. Am p lific a tion Selec tion
Consid era tions in the Ped ia tric Pop ula tion. In: F.H. Bess, J.S. Gra vel
a nd A.M. Tha rp e (ed s.), Am p lific a tion for Child ren w ith Aud itory
Defic its (p p . 145–160). Na shville: Bill Wilkerson Center Press.
Bellis, T.J. 2003. Assessm ent a nd Ma na g em ent of Centra l Aud itory
Proc essing Disord ers in the Ed uc a tiona l Setting from Sc ienc e to
Pra c tic e 2nd Ed ition. Ca na d a : Sing ula r Pub lishing .
Bess, F.H., 2000. Ea rly Am p lific a tion for Child ren: Im p lem enting Cha ng e. In:
R.C.
Seew a ld
(ed .),
A
Sound
Found a tion
throug h
Ea rly
Am p lific a tion: Proc eed ing s of a n Interna tiona l Conferenc e (p p . 247251). Stä fa , Sw itzerla nd : Phona k AG.
Bess, F.H., & Ha ll III, J.W. 1992. Sc reening Child ren for Aud itory Func tion.
Na shville: Bill Wilkerson Center Press.
Boothroyd , A. 1997. Aud itory Ca p a c ity of Hea ring Im p a ired Child ren using
Hea ring Aid s a nd Coc hlea r Im p la nts: Issues of effic a c y a nd
a ssessm ent. Sc a nd ina via n Aud iolog y, 26(46): 17-25.
Brow n, E., Klein, A.J., a nd Snyd ee, K.A. 1999. Hea ring -a id -p roc essed Tonep ip s: Elec troa c oustic
a nd ABR c ha ra c teristic s. Journa l of the
Am eric a n Ac a d em y of Aud iolog y 10: 190-197.
Buerkli-Ha levy, O., a nd Chec kley, P.C. 2000. Ma tc hing Tec hnolog y to the
Need s of Infa nts. In: R.C. Seew a ld (ed .), A Sound Found a tion
throug h
Ea rly
Am p lific a tion: Proc eed ing s of a n
Interna tiona l
Conferenc e (p p . 77-86). Stä fa , Sw itzerla nd : Phona k AG.
2
University of Pretoria etd – Stroeb
bel, D (2006)
Burka rd , R.F., a nd Sec or, C. 2002. Overview of Aud itory Potentia ls. In: J.Ka tz
(ed .), Ha nd b ook of Clinic a l Aud iolog y 5th Ed . (p p . 233-248).
Ba ltim ore: Lip p inc ott, Willia m s & Wilkins.
Cone-Wesson, B. 2003. Elec trop hysiolog ic Assessm ent of Hea ring in Infa nts:
Com p ound
Nerve
Ac tion, Aud itory Bra instem
Resp onse, a nd
Aud itory Stea d y Sta te Resp onse. The Volta Review 103(4): 253-279.
Cone-Wesson, B., Dow ell, R.C., Tom lin, D., Ra nc e, G., a nd Ming , W.J. 2002.
The Aud itory Stea d y-Sta te Resp onse: Com p a risons w ith the Aud itory
Bra instem Resp onse. Journa l of Am eric a n Ac a d em y of Aud iolog y
13(4): 173-187.
Cone-Wesson, B., Ric ka rd s, F., Poulis, C., Pa rker, J., Ta n, L., a nd Polla rd , J.
2002. The Aud itory Stea d y Sta te Resp onse: Clinic a l Ob serva tions
a nd Ap p lic a tions in Infa nts a nd Child ren. Journa l of the Am eric a n
Ac a d em y of Aud iolog y 13(5): 270-282.
Cone-Wesson, B., Pa rker, J., Sw id erski, N., a nd Ric ka rd s, F. 2002. The
Aud itory
Stea d y-Sta te
Resp onse:
Full-Term
a nd
Prem a ture
Neona tes. Journa l of the Am eric a n Ac a d em y of Aud iolog y 13(5):
260-269.
Cone-Wesson, B., Vohr, B.R., Sining er, Y., Wid en, J.E., Folsom , R.C., Gorg a ,
M.P., a nd Norton, S.J., 2000. Id entific a tion of Neona ta l Hea ring
Im p a irm ent: Infa nts w ith Hea ring Loss. Ea r a nd Hea ring 21(5): 488507.
3
University of Pretoria etd – Stroeb
bel, D (2006)
De Vos, A.S., Stryd om , H., Fouc hé, C.B., Pog g enp oel, M. a nd Sc hurink, E.W.
2002. Resea rc h a t Gra ss Roots 2nd Ed . Pa a rl: Va n Sc ha ik.
Diefend orf, A.O. 2002. Detec tion a nd Assessm ent of Hea ring Loss in Infa nts
a nd Child ren. In: J. Ka tz (ed .), Ha nd b ook of Clinic a l Aud iolog y 5th
Ed . (p p . 469-480). Ba ltim ore: Lip p inc ott, Willia m s & Wilkins,
Diefend orf, A.O., Reitz, P.S., Esc ob a r, M.W., a nd Wynne, M.K. 1996.
Initia ting Ea rly Am p lific a tion: TIPS for Suc c ess. In: F.H. Bess, J.S. Gra vel
a nd A.M. Tha rp e (ed s.), Am p lific a tion for Child ren w ith Aud itory
Defic its (p p . 123-144). Na shville: Bill Wilkerson Center Press.
Diefend orf, A.O. a nd Web er, B.A. 1994. Id entific a tion of Hea ring Loss:
Prog ra m m a tic a nd Proc ed ura l Consid era tions. In: J. Roush a nd N.D.
Ma tkin (ed s.), Fa m ily Centered Assessm ent a nd Intervention (p p . 4366). Ba ltim ore: York Press, Inc .
Dillon, H. 2001. Hea ring Aid s. Syd ney: Boom era ng Press.
Dim itrijevic , A., Sa sha John, M., Va n Roon, P., Purc ell, D.W., Ad a m onis, J.,
Ostroff, J., Ned zelski, J.M., a nd Pic ton T.W. 2002. Estim a ting the
Aud iog ra m Using Multip le Aud itory Stea d y Sta te Resp onses. Journa l
of the Am eric a n Ac a d em y of Aud iolog y 13(4): 205-224.
Dim itrijevic , A., Sa sha John, M., a nd Pic ton, T.W. 2004. Aud itory Stea d ySta te Resp onses a nd Word Rec og nition Sc ores in norm a l-Hea ring
a nd Hea ring -Im p a ired Ad ults. Ea r a nd Hea ring 25(1): 68-84.
4
University of Pretoria etd – Stroeb
bel, D (2006)
Don, M., a nd Kw ong , B. 2002. Aud itory Bra instem Resp onse: Differentia l
Dia g nosis. In: J. Ka tz (ed .), Ha nd b ook of Clinic a l Aud iolog y 5th Ed .
(p p . 274-297). Ba ltim ore: Lip p inc ott, Willia m s & Wilkins,
Drum m ond , A. 2003. Resea rc h Method s for Thera p ists. United King d om :
Nelson Thornes Ltd .
Feig htner, J.W. 1992. Sc reening in the 1990’ s: Som e Princ ip les a nd
Guid elines. In: F.H. Bess a nd J.W. Ha ll iii (ed s.), Sc reening Child ren for
Aud itory Func tion (p p . 1-16). Na shville: Bill Wilkerson Center Press.
Ferra ro, J.A., a nd Durra nt, J.D. 1994. Aud itory Evoked Potentia ls: Overview
a nd Ba sic
Princ ip les. In: J. Ka tz (ed .), Ha nd b ook of Clinic a l
Aud iolog y 4th Ed . (p p . 317-338). Ba ltim ore: Lip p inc ott, Willia m s a nd
Wilkins.
Fow ler, C.G., a nd Sha nks, J.E. 2002. Tym p a nom etry. In: J. Ka tz (ed .),
Ha nd b ook of Clinic a l Aud iolog y 5th Ed . (p p . 175-204). Ba ltim ore:
Lip p inc ott, Willia m s a nd Wilkins.
Ga rnha m , J., Cop e, Y., Durst, C., Mc Corm ic k, B., a nd Ma son, S.M. 2000.
ABR a ssessm ent of a id ed threshold s b efore c oc hlea r im p la nta tion.
British Soc iety of Aud iolog y: 267-278.
Gold stein, R., a nd Ald ric h, W.M. 1999. Evoked Potentia l Aud iom etry:
Fund a m enta ls a nd Ap p lic a tions. Boston: Allyn & Ba c on.
5
University of Pretoria etd – Stroeb
bel, D (2006)
Gorg a , M.P., Neely, S.T., Hoover, B.M., Dierking , D.M., Bea uc ha ine, K.L.,
a nd Ma nning , C. 2004. Determ ining the Up p er Lim its of Stim ula tion
for Aud itory Stea d y Sta te Resp onse Mea surem ents. Ea r a nd Hea ring
25(3): 302-307.
Gorg a , M.P., a nd Neely, S.T. 2002. Som e Fa c tors tha t Ma y Influenc e the
Ac c ura c y of Aud itory Bra instem Resp onse Estim a tes of Hea ring Loss.
In: R.C. Seew a ld a nd J.S. Gra vel (ed s.), A Sound Found a tion throug h
Ea rly Am p lific a tion: Proc eed ing s of the
Sec ond
Interna tiona l
Conferenc e (p p . 49-61). Stä fa , Sw itzerla nd : Phona k AG.
Gorg a , M.P. 1999. Pred ic ting Aud itory Sensitivity from Aud itory Bra instem
Resp onse Mea surem ents. Sem ina rs in Hea ring 20(1): 29-43.
Gorg a , M.P., a nd Thornton, R.T. 1989. The Choic e of Stim uli for ABR
Mea surem ents. Ea r a nd Hea ring 10(4): 217-230.
Gra vel, J.S. 2005. Evid enc e-Ba sed Pra c tic e in Ped ia tric Aud iolog y. In: R.C.
Seew a ld a nd J.M. Ba m ford (ed s.), A Sound Found a tion throug h
Ea rly
Am p lific a tion:
Proc eed ing s
of
the
Third
Interna tiona l
Conferenc e (p p . 17– 26). Stä fa , Sw itzerla nd : Phona k AG.
Gra vel, J.S. 2000. Aud iolog ic Assessm ent for the Fitting of Hea ring
Instrum ents: Big Cha lleng es from Tiny Ea rs. In: R.C. Seew a ld (ed .), A
Sound Found a tion throug h Ea rly Am p lific a tion: Proc eed ing s of a n
Interna tiona l Conferenc e (p p . 33-46). Stä fa , Sw itzerla nd : Phona k
AG.
6
University of Pretoria etd – Stroeb
bel, D (2006)
Ha ll III, J.W. 2005. Ped ia tric c onferenc e: Pretoria , South Afric a .
Ha ll III, J.W. 1992. Ha nd b ook of Aud itory Evoked Resp onses. Boston: Allyn
a nd Ba c on.
Ha ll III, J.W. 1999. Aud itory Evoked Resp onse a nd Otoa c oustic Em ission
Ha nd s-on Workshop . Ja nua ry 21-23, 1999, Va nd erb ilt University
Med ic a l Centre, Na shville, Tennessee.
Ha ll III, J.W. 2000. Ha nd b ook of Otoa c oustic Em issions. , Ca lifornia : Sing ula r
Pub lishing Group .
Ha ll III, J.W. a nd Mueller, H.G. 1997. Aud iolog ist’ Desk Referenc e, Volum e I.
Ca lifornia : Sing ula r Pub lishing Group .
Ha ll III, J.W., a nd Mueller, H.G. 1998. Aud iolog ist’ Desk Referenc e, Volum e
II. Ca lifornia : Sing ula r Pub lishing .
Ha rrell, W.R. 2002. Pure tone Eva lua tion. In: J. Ka tz (ed .), Ha nd b ook of
Clinic a l Aud iolog y 5th Ed . (p p . 71-87). Ba ltim ore: Lip p inc ott, Willia m s
a nd Wilkins.
Ha yes, D., a nd Northern, J. 1997. Infa nts a nd Hea ring . Sa n Dieg o:
Sing ula r Pub lishing g roup .
Heg d e, M.N. 1987. Clinic a l Resea rc h in Com m unic a tive Disord ers:
Princ ip les a nd Stra teg ies. Boston: Colleg e-Hill Pub lic a tion.
7
University of Pretoria etd – Stroeb
bel, D (2006)
Hed ley-Willia m s, A., Tha rp e, A.M., a nd Bess, F.H. 1996. Fitting Hea ring Aid s
in the Ped ia tric Pop ula tion: A survey of p ra c tic e p roc ed ures. In: F.H.
Bess, J.S. Gra vel a nd A.M. Tha rp e (ed s.), Am p lific a tion for Child ren
w ith Aud itory Defic its (p p . 107-122). Na shville: Bill Wilkerson Center
Press.
Herd m a n, A.T., a nd Sta p ells, D.R. 2003. Aud itory Stea d y-Sta te Resp onse
Threshold s of Ad ults w ith
Sensorineura l Hea ring
Im p a irm ents.
Interna tiona l Journa l of Aud iolog y (42): 237-248.
Herd m a n, A.T., a nd Sta p ells, D.R. 2001. Threshold Determ ina tion using the
Monotic a nd Dic hotic Multip le Aud itory Stea d y-Sta te Resp onse
Tec hniq ue in Sub jec ts. Sc a nd ina via n journa l of Aud iolog y (30): 4149.
Hill, A., a nd Sp ittlehouse, C. 2005. Wha t is c ritic a l a p p ra isa l?
w w w .evid enc e-b a sed -m ed ic ine.c o.uk; a c c essed Novem b er 2005.
Hood , L.J. 1998. Clinic a l Ap p lic a tions of the Aud itory Bra instem Resp onse.
Sa n Dieg o: Sing ula r Pub lishing g roup .
Hyd e, M. 2005. Evid enc e-Ba sed Pra c tic e, Ethic s a nd EHDI Prog ra m Qua lity.
In: R.C. Seew a ld a nd J.M. Ba m ford (ed s.), A Sound Found a tion
throug h Ea rly Am p lific a tion: Proc eed ing s of the Third Interna tiona l
Conferenc e (p p . 281– 301). Stä fa , Sw itzerla nd : Phona k AG.
Hyd e, M., Sining er, Y.S., a nd Dom , M. 1998. Ob jec tive Detec tion a nd
a na lysis of Aud itory Bra instem Resp onse: An historic a l Persp ec tive.
Sem ina rs in Hea ring 19(1): 97-113.
8
University of Pretoria etd – Stroeb
bel, D (2006)
Ja c ob son, J.T., 1985. The Aud itory Bra instem Resp onse. Sa n Dieg o:
Colleg e-Hill Press.
Jeng , F-C., Brow n, C.J., Johnson, T.A., a nd Va nd er Werff, K.R. 2004.
Estim a ting Air-Bone Ga p s Using Aud itory Stea d y-Sta te Rep onses.
Journa l of the Am eric a n Ac a d em y of Aud iolog y 15(1): 67-78.
Jerg er, J. 1998. The Aud itory Stea d y Sta te Resp onse: Ed itoria l. Journa l of
the Am eric a n Ac a d em y of Aud iolog y (ed itoria l).
Jenkins, S., Pric e, C.J., a nd Stra ker, L. 2003. The Resea rc hing Thera p ist.
Ed inb urg h: Churc hill Living stone.
John, M.S., Brow n, D.K., Muir, P.J., a nd Pic ton, T.W. 2004. Rec ord ing
Aud itory Stea d y Sta te Resp onse in Young Infa nts. Ea r a nd Hea ring
25(6): 539-553.
John, M.S., Dim itrijevic , A., a nd Pic ton, T. W. 2002. Aud itory Stea d y-Sta te
Resp onses to Exp onentia l Mod ula tion Envelop es. Ea r a nd Hea ring
23(2): 106-117.
John, M.S., Lins, O.G., Bouc her, B.L., a nd Pic ton, T.W. 1998. Multip le
a ud itory stea d y-sta te resp onses (MASTER): Stim ulus a nd rec ord ing
p a ra m eters. Aud iolog y: 59-82.
Joint Com m ittee on Infa nt Hea ring Sc reening , 1994. 1994 Position
Sta tem ent. ASHA (36): 38-42.
Joint Com m ittee on Infa nt Hea ring , 2000. Yea r 2000 Position Sta tem ent:
9
University of Pretoria etd – Stroeb
bel, D (2006)
Princ ip les a nd
Guid elines for
Ea rly
Hea ring
Detec tion
a nd
Intervention Prog ra m s. Am eric a n Journa l of Aud iolog y, (9): 9-29.
Ka tz, J. 2002. Clinic a l Aud iolog y 5th Ed . (p p . 3-8). Ba ltim ore: Lip p inc ott,
Willia m s a nd Wilkins.
Kei, J., Allison-Levic k, J., Doc kra y, J., Ha rrys, R., Kirkeg a rd , C., Wong , J.,
Ma urer, M., Heg a rty, J., Young , J., a nd Tud ehop e, D. 2003. Hig hFreq uenc y (1000 Hz) Tym p a nom etry in Norm a l Neona tes. Journa l of
the Am eric a n Ac a d em y of Aud iolog y 14(1): 20-28.
Kirkw ood , D.H. 2002. Ca ring for our Young est Pa tients: A uniq ue
Op p ortunity a nd Cha lleng e. Hea ring Journa l 55(11): ed itoria l.
Kuk, F. 2004. Persona l Corresp ond enc e: Ped ia tric s a nd func tiona l g a in.
Kuk, F., a nd Ma rc oux, A. 2002. Fa c tors Ensuring Consistent Aud ib ility in
Ped ia tric Hea ring Aid Fitting . Journa l of the Am eric a n Ac a d em y of
Aud iolog y 13(9): 503-520.
Kurtzer-White, E., a nd Luterm a n, D. 2001. Ea rly Child hood Dea fness.
Tim onium : York Press.
Kuw a d a , S., And erson, J.S., Ba tra , R., Fitzp a tric k, D.C., Teissier, N. &
D’ Ang elo, W.R. 2002. Sourc es of the Sc a lp -Rec ord ed Am p litud eMod ula tion Follow ing Resp onse. Journa l of the Am eric a n Ac a d em y
of Aud iolog y 13(4):188-204.
Leed y, P.D., a nd Orm rod , J.E. 2005. Pra c tic a l Resea rc h: Pla nning a nd
10
University of Pretoria etd – Stroeb
bel, D (2006)
Desig n. New Jersey: Merrill Prentic e Ha ll.
Leed y, P.D., a nd Om rod , J.E. 2001. Pra c tic a l Resea rc h: Pla nning a nd
Desig n. New Jersey: Merill Prentic e Ha ll.
Leed y, R. 1997. Pra c tic a l resea rc h: Pla nning a nd Desig n. Ohio: Colum b us.
Leed y, R. 1981. How to rea d resea rc h a nd und ersta nd it. New York:
Ma c m illa n.
Lew is, D.E. 2000. Hea ring Instrum ent Selec tion a nd Fitting in Child ren. In: M.
Va lente, H. Hosford -Dunn, a nd
R.J. Roeser (ed s.), Aud iolog y
Trea tm ent (p p . 149-212). New York: Thiem e.
Lins, O.G., Pic ton, T.W., Bouc her, B.L., Durieux-Sm ith, A., Cha m p a g ne, S.C.,
Mora n, L.M., Perez-Ab a lo, M.C., Ma rtin, V., a nd Guillerm o, S. 1996.
Freq uenc y-Sp ec ific Aud iom etry Using Stea d y-Sta te Resp onses. Ea r
a nd Hea ring 1796(2): 81-96.
Lins, O.G., a nd Pic ton, T.W. 1995. Aud itory Stea d y-Sta te Resp onses to
Multip le Sim ulta neous Stim uli. Elec troenc ep ha log ra p hy a nd Clinic a l
Neurop hysiolog y 96: 420-432.
Louw , B., 2004 Lec ture: Resea rc h Ethic s in Com m unic a tion Pa tholog y.
Luterm a n, D.M. w ith Kurtzer-White, E. a nd Seew a ld , R.C., 1999. The Young
Dea f Child . Ba ltim ore: York Press Inc .
Luts, H., Desloovere, C., Kum a r, A., Va nd erm eersc h, E., a nd Wouters, J.
11
University of Pretoria etd – Stroeb
bel, D (2006)
2004.
Ob jec tive
Threshold s
in
Assessm ent
Ba b ies.
of
Freq uenc y-sp ec ific
Interna tiona l
Journa l
of
Hea ring
Ped ia tric
Otorhinola ryng olog y (68): 915-926.
Ma honey, T.M., 1985. Aud itory Bra instem Resp onse Hea ring Aid
Ap p lic a tions. In: J.T. Ja c ob son (ed .), The Aud itory Bra instem
Resp onse (p p . 349-370). Sa n Dieg o: Colleg e-Hill Press.
Ma ra is, C.C., 2003. Tra nsd uc er influenc e on Aud itory Stea d y Sta te Evoked
Potentia ls.
Unp ub lished
M
Com m unic a tion
Pa tholog y
thesis,
University of Pretoria , South Afric a .
Mend el, L.L., Da nha uer, J.L., a nd Sing h, S. 1999. Sing ula r’ s Illustra ted
Dic tiona ry of Aud iolog y. Sa n Dieg o: Sing ula r Pub lishing Group , Inc .
Moeller, M.P. 2001. Intervention a nd Outc om es for Young Child ren Who
a re Dea f a nd Ha rd of Hea ring a nd their Fa m ilies. In: E. Kurtzer-White
a nd D. Luterm a n (ed s.), Ea rly Child hood Dea fness (p p . 109-138).
Tim onium : York Press.
Moeller, M.P., 2000. Ea rly Intervention a nd La ng ua g e Develop m ent in
Child ren w ho a re Dea f a nd Ha rd of Hea ring . Ped ia tric s, 106(E43).
Mouton, J. 2001. How to suc c eed in your Ma ster’ s & Doc tora l Stud ies.
Pretoria , South Afric a : Va n Sc ha ik Pub lishers.
Na tiona l Institutes of Hea lth Consensus Sta tem ent, 1993. Ea rly
12
University of Pretoria etd – Stroeb
bel, D (2006)
Id entific a tion of Hea ring Im p a irm ent in Infa nts a nd Young Child ren.
In: D. Ha yes a nd J. Northern (ed s.), Infa nts a nd Hea ring (a p p end ix
E). Sa n Dieg o: Sing ula r Pub lishing Group .
Nea ult, M.W. 2001. After Sc reening : The Dia g nostic Proc ess in Ea rly
Chilhood Dea fness. In: E. Kurtzer-White a nd D. Luterm a n (ed s.), Ea rly
Child hood Dea fness (p p . 29-48). Sa n Dieg o: York Press.
Neum a n, W.L. 1997. Soc ia l Resea rc h Method s: Qua lita tive a nd
Qua ntita tive Ap p roa c hes. Boston: Alla n Ba c on.
Northern, J.L., a nd Dow ns, M.P. 2002. Hea ring in Child ren. Ba ltim ore:
Lip p inc ott, Willia m s a nd Wilkins.
Oa tes, P., a nd Sta p ells, D.R. 1998. Aud itory Bra instem Resp onse Estim a tes
of the Pure-Tone Aud iog ra m : Current Sta tus. Sem ina rs in Hea ring
19(1): 61-85.
Orla nd o, M.S., a nd Prieve, B.A. 1998. Mod els for Universa l New b orn
Hea ring
Sc reening
Prog ra m s in
Universa l
New b orn
Hea ring
Sc reening . In: L.G. Sp iva k (ed .), Universa l New b orn
Hea ring
Sc reening . New York: Thiem e.
Oxford -Centre for Evid enc e Ba sed Med ic ine 2004.
w w w .c eb m .net/ eb m _is_isnt.a sp ; a c c essed Oc tob er 2005.
Pa lm er, C. 2005. In Fitting Kid s w ith Hea ring Aid s, Ensuring Sa fety a nd
13
University of Pretoria etd – Stroeb
bel, D (2006)
Aud ib ility is a g ood w a y to sta rt. Hea ring Journa l 58(2): 10-17.
Ped ia tric Working Group of the Conferenc e on Am p lific a tion for Child ren
w ith Aud itory Defic its, 1996. Am p lific a tion for Infa nts a nd c hild ren
w ith hea ring loss. Am eric a n Journa l of Aud iolog y 5(1): 53-68.
Perez-Ab a lo, M.C., Sa vio, G., Torres, A., Ma rtin, V., Rod rig uez, E., a nd
Ga la n, L. 2001. Stea d y Sta te resp onses to Multip le Am p litud e
Mod ula ted Tones: An op tim ized m ethod to test Freq uenc y-sp ec ific
threshold s in hea ring
im p a ired
c hild ren a nd
norm a l hea ring
sub jec ts. Ea r a nd Hea ring 22(3): 200-211.
Pic ton, T.W. 2005. Ob jec tive Aud iom etry: Prob lem s a nd Prog ress. Ora l
p resenta tion
at
the
2nd
Europ ea n
Conferenc e
on
Ped ia tric
Am p lific a tion Solutions: Sound for a Young Genera tion.
Pic ton, T.W., Dim itrijevic , A., Perez-Ab a lo, M-C., a nd Va n Roon, P. 2005.
Estim a ting
Aud iom etric
Threshold s Using
Aud itory Stea d y-Sta te
Resp onses. Journa l of the Am eric a n Ac a d em y of Aud iolog y 16(3):
140–154.
Pic ton, T.W., a nd John M.S. 2004. Avoid ing Elec trom a g netic Artifa c ts When
Rec ord ing
Aud itory
Stea d y-Sta te
Resp onses.
Journa l
of
the
Am eric a n Ac a d em y of Aud iolog y 15(8): 541-554.
Pic ton, T.W., Dim itrijevic , A., Va n Roon, P., Sa sha -John, M., Reed , M., a nd
14
University of Pretoria etd – Stroeb
bel, D (2006)
Finkelstein, H. 2002. Possib le
Roles for Aud itory Stea d y-Sta te
Resp onses in fitting Hea ring Aid s. In: R.C. Seew a ld a nd J.S. Gra vel
(ed s.),
A
Sound
Found a tion
throug h
Ea rly
Am p lific a tion:
Proc eed ing s of the Sec ond Interna tiona l Conferenc e (p p . 63-73).
Stä fa , Sw itzerla nd : Phona k AG.
Pic ton, T.W., John, M.S., a nd Dim itrijevic , A. 2002. Possib le Roles for the
Aud itory Stea d y Sta te Resp onses in Id entific a tion, Eva lua tion a nd
Ma na g em ent of Hea ring Loss. Aud iolog y Tod a y (14): 29-34.
Pic ton, T.W., Durieux-Sm ith, A., Cha m p a g ne, S.C., Whitting ha m , J., Mora n,
L.W., Gig ueve, C., a nd Bea ureg a rd , Y. 1998. Ob jec tive Eva lua tion
of Aid ed Threshold s using Aud itory Stea d y-Sta te Resp onse. Journa l
of the Am eric a n Ac a d em y of Aud iolog y (9): 315-331.
Prieve, B.A., a nd Fitzg era ld , T.S. 2002. Otoa c oustic Em issions. In: J. Ka tz
(ed .), Ha nd b ook of Clinic a l Aud iolog y 5th Ed . (p p . 440-466).
Ba ltim ore: Lip p inc ott, Willia m s a nd Wilkins.
Purd y, S.C., Ka tsc h, R., Dillon, H., Storey, L., Sha rm a , M., a nd Ag ung , K.,
2005. Aid ed
Cortic a l Aud itory Evoked
Potentia ls for Hea ring
Instrum ent Eva lua tion in Infa nts. In: R. C. Seew a ld a nd J.M. Ba m ford
(ed s.),
A
Sound
Found a tion
throug h
Ea rly
Am p lific a tion:
Proc eed ing s of the Third Interna tiona l Conferenc e (p p . 115-128).
Stä fa , Sw itzerla nd : Phona k AG.
Purd y, S.C., a nd Ab b a s, P.J. 2002. ABR Threshold s to Toneb ursts Ga ted w ith
15
University of Pretoria etd – Stroeb
bel, D (2006)
Bla c km a n a nd Linea r Wind ow s in Ad ults w ith Hig h-freq uenc y Sensori
neura l Hea ring Loss. Ea r a nd Hea ring 23(4): 358-368.
Ra nc e, G., Rop er, R., Sym ons, L.M., Poulis, C., Dourla y, M., a nd Kelly, T.
2005. Hea ring Threshold s Estim a tion in Infa nts Using Aud itory Stea d ySta te Resp onses. Journa l of the Am eric a n Ac a d em y of Aud iolog y
16(5): 291 – 300.
Ra nc e, G., a nd Brig g s, R.J.S. 2002. Assessm ent of hea ring level in infa nts
w ith sig nific a nt hea ring loss: the Melb ourne exp erienc e w ith stea d ysta te
evoked
p otentia l threshold
testing . Anna ls of Otolog y
Rhinolog y a nd La ryng olog y 111 (Sup p l. 189): 22 -28.
Ra nc e, G., a nd Ric ka rd s, F. 2002. Pred ic tion of Hea ring Threshold in Infa nts
Using Stea d y-Sta te Evoked Potentia ls. Journa l of the Am eric a n
Ac a d em y of Aud iolog y 13(5): 236-245.
Ra nc e, G., Beer, D.E., Cone-Wesson, B., Shep herd , R.K., Dow ell, R.C., King ,
A.M., Ric ka rd s, F.W., a nd Cla rk, G.M. 1999. Clinic a l Find ing s for a
Group of Infa nts a nd Young Child ren w ith Aud itory Neurop a thy. Ea r
a nd Hea ring 20(3): 238-252.
Ra nc e, G., Dow ell, R.C., Ric ka rd s, F.W. Beer, D.E., a nd Cla rk, G.M. 1998.
Stea d y Sta te evoked p otentia ls a nd b eha viora l hea ring threshold s
in a
g roup
of c hild ren w ith a b sent c lic k-a ud itory b ra instem
resp onse. Ea r a nd Hea ring 19(1): 48-61.
Ra nc e, G., Ric ka rd s, F.W., La w renc e, T.C., De Vid i, S., a nd Cla rk, G.M.
16
University of Pretoria etd – Stroeb
bel, D (2006)
1995. The Autom a ted Pred ic tion of Hea ring Threshold s in Sleep ing
Sub jec ts Using Aud itory Stea d y-Sta te Evoked Potentia ls. Ea r a nd
Hea ring 16(5): 499-507.
Ric ka rd s, F.W., Ta n, L.E., La w renc e, T.C., Wilson, O.J., Drew , J.H., a nd Cla rk,
G.M. 1994. Aud itory Stea d y-Sta te Evoked Potentia l in New b orns.
British Journa l of Aud iolog y 28: 327-337.
Referra l Guid elines for Coc hlea r Im p la nts, 2004. Southern ENT.
Rob inette, M.S., a nd Ga ltke, T.J. 2000. Otoa c oustic Em issions. In: R.J.
Roeser, M. Va lente a nd H. Hosford -Dunn (ed s.), Aud iolog y Dia g nosis
(p p . 503-526). New York: Thiem e.
Roeser, R.J., Va lente, M., a nd Hosford -Dunn, H. 2000. Aud iolog y Dia g nosis.
New York: Thiem e.
Ross, M. 2001. Som e Reflec tion on Ea rly Child hood Dea fness. In: E. KurzerWhite a nd D. Luterm a n (ed s.), Ea rly Child hood Dea fness (p p . 1-12).
Tim onium : York Press, Inc .
Ross, M. 1996. Am p lific a tion for Child ren: The Proc ess Beg ins. In: F. Bess, J.S.
Gra vel a nd A.M. Tha rp e (ed s.), Am p lific a tion for Child ren w ith
Aud itory Defic its (p p . 1-28). Na shville: Bill Wilkerson Center Press.
Roush, J., a nd Ma tkin, N.D. 1994. Infa nts a nd Tod d lers w ith Hea ring Loss:
Fa m ily Centered Assessm ent a nd Intervention. Tim onium : York Press,
Inc .
17
University of Pretoria etd – Stroeb
bel, D (2006)
Roush, P.A., 2005. Hea ring Aid Fitting in Infa nts: Pra c tic a l Consid era tions
a nd Cha lleng es. In: R.C. Seew a ld a nd J.M. Ba m ford (ed s.), A Sound
Found a tion throug h Ea rly Am p lific a tion: Proc eed ing s of the Third
Interna tiona l Conferenc e (p p . 105-114). Stä fa , Sw itzerla nd : Phona k
AG.
Sc ollie, S.D., 2005. Presc rip tive Proc ed ures for Infa nts a nd Child ren. In: R.C.
Seew a ld a nd J.M. Ba m ford (ed s.), A Sound Found a tion throug h
Ea rly
Am p lific a tion:
Proc eed ing s
of
the
Third
Interna tiona l
Conferenc e (p p . 91-104). Stä fa , Sw itzerla nd : Phona k AG.
Sc ollie, S., a nd Seew a ld , R.C. 2002. Hea ring Aid Fitting a nd Verific a tion
Proc ed ures for Child ren. In: J. Ka tz (ed .), Ha nd b ook of Clinic a l
Aud iolog y 5th Ed . (p p . 687-706). Ba ltim ore: Lip p inc ott, Willia m s a nd
Wilkins.
Seew a ld , R.C. 2001. Current Issues in Ped ia tric Hea ring Aid Fitting . In: E.
Kurtzer-White a nd D. Luterm a n (ed s.), Ea rly Child hood Dea fness (p p .
63-72). Tim onium : York Press.
Seew a ld , R.C. 2000. A Sound Found a tion Throug h Ea rly Am p lific a tion:
Proc eed ing s of a n Interna tiona l Conferenc e (ed itoria l). Stä fa ,
Sw itzerla nd : Phona k AG.
Seew a ld , R.C., Mood ie, K.S., Sinc la ir, S.T., a nd Cornelisse, L.E. 1996.
18
University of Pretoria etd – Stroeb
bel, D (2006)
Tra d itiona l a nd Theoretic a l Ap p roa c hes to Selec ting Am p lific a tion
for Infa nts a nd Young Child ren. In: F.H. Bess J.S. Gra vel a nd A.M.
Tha rp e (ed s.), Am p lific a tion for Child ren w ith Aud itory Defic its (p p .
161-192). Na shville: Bill Wilkerson Center Press.
Sining er, Y., Ma rsh, R., Wa ld en, B., a nd Wilb er, L.A. 2003. Guid elines for
Ethic a l Pra c tic e in Resea rc h for Aud iolog ists. Aud iolog y Tod a y
15(6):14-17.
Sining er, Y.S., a nd Cone-Wesson, B. 2002. Threshold Pred ic tion Using
Aud itory Bra instem Resp onse a nd Stea d y-sta te Evoked Potentia ls
w ith Infa nts a nd Young Child ren. In: J. Ka tz (ed .), Ha nd b ook of
Clinic a l Aud iolog y 5th Ed . (p p . 298-322). Ba ltim ore: Lip p inc ott,
Willia m s a nd Wilkins.
Sining er, Y.S., Doyle, K.L., a nd Moore, J.K. 1999. The Ca se for Ea rly
Id entific a tion
of
d evelop m ent,
Hea ring
Loss in
exp erim enta l
Child ren.
a ud itory
Aud itory
system
d ep riva tion,
a nd
d evelop m ent of sp eec h p erc ep tion a nd hea ring . Ped ia tric Clinic s
of North Am eric a 46(1): 1-14.
Sm a ll, S.A., a nd Sta p ells, D.R. 2004. Artifa c tua l Resp onses When Rec ord ing
Aud itory Stea d y-Sta te Resp onses. Ea r a nd Hea ring 25(6): 611-623.
Sta c h B.A. 1998. Clinic a l Aud iolog y a n Introd uc tion. Sa n Dieg o: Sing ula r
Pub lishing Group .
Sta p ells, D.A., Herd m a n, A., Sm a ll, S.A., Dim itrijevic , A., a nd Ha tton, J.,
19
University of Pretoria etd – Stroeb
bel, D (2006)
2005. Current Sta tus of the Aud itory Stea d y-Sta te Resp onse for
Estim a ting a n Infa nt’ s Aud iog ra m . In: R.C. Seew a ld a nd J.M.
Ba m ford (ed s.), A Sound Found a tion throug h Ea rly Am p lific a tion:
Proc eed ing s of the Third Interna tiona l Conferenc e (p p . 43-60).
Stä fa , Sw itzerla nd : Phona k AG.
Sta p ells, D.R. 2004 Current sta tus of the Aud itory Stea d y Sta te Resp onse
For Estim a ting a n Infa nt’ s Aud iog ra m . Ora l p resenta tion a t the third
interna tiona l c onferenc e, A Sound
Found a tion Throug h Ea rly
Am p lific a tion Conferenc e, Chic a g o.
Sta p ells, D.R. 2002. The Tone-evoked ABR: Why it’ s the m ea sure of c hoic e
for young infa nts. The Hea ring Journa l 55(11): 14-17.
Sta p ells, D.R. 2000a . Freq uenc y-Sp ec ific Evoked Potentia l Aud iom etry in
Infa nts. In: R.C. Seew a ld (ed .), A Sound Found a tion throug h Ea rly
Id entific a tion: Proc eed ing s of a n Interna tiona l Conferenc e (p p . 1332). Stä fa , Sw itzerla nd : Phona k AG.
Sta p ells, D.R. 2000b . Threshold Estim a tion b y the Tone-evoked Aud itory
Bra instem Resp onse: a litera ture m eta -a na lysis. Journa l of Sp eec h
La ng ua g e Pa tholog y a nd Aud iolog y 24: 74-83.
Sta p ells, D.R., a nd Oa tes, P. 1997. Estim a tion of the Pure-Tone Aud iog ra m
b y the Aud itory Bra instem Resp onse: A Review . Aud iolog y a nd
Neuro-Otolog y 2(5): 257-280.
Sta p ells, D.R., Gra vel, J.S., a nd Ma rtin, B.A. 1995. Threshold for Aud itory
20
University of Pretoria etd – Stroeb
bel, D (2006)
Bra in Stem Resp onses to Tones in notc hed Noise from Infa nts a nd
Young Child ren w ith Norm a l Hea ring or Sensorineura l Hea ring Loss.
Ea r a nd Hea ring , 16(4): 361-371.
Sta p ells, D.R., Pic ton, T.W., Durieux-Sm ith, A., Ed w a rd s, C.G., a nd Mora n,
L.M. 1990. Threshold for Short-La tenc y Aud itory-Evoked Potentia ls to
Tones in Notc hed Noise in Norm a l-Hea ring a nd Hea ring -Im p a ired
Sub jec ts. Aud iolog y 29: 262-274.
Stelm a c how ic z, P.A. 2000. How d o w e know w e’ ve g ot it rig ht?
Elec troa c oustic a nd Aud iom etric Mea sures. In: R.C. Seew a ld (ed .),
A Sound Found a tion Throug h Ea rly Am p lific a tion: Proc eed ing s of a n
Interna tiona l Conferenc e (p p . 109-118). Stä fa , Sw itzerla nd : Phona k
AG.
Stelm a c how ic z, P.A., Hoover, B., Lew is, D.E., Brenm a n, M. 2002. Is
Func tiona l Ga in rea lly Func tiona l? The Hea ring Journa l 55(11): 38-42.
Steyn, A.G.W., Sm it, C.F., Du Toit, S.H.C., a nd Stra sheim , C. 1994. Mod ern
Sta tistic s in Pra c tic e. Pretoria : JL Va n Sc ha ik Pub lishers.
Stueve, M.P., a nd O’ Rourke, C. 2003. Estim a tion of hea ring loss in c hild ren:
Com p a rison of a ud itory stea d y-sta te resp onse, a ud itory b ra instem
resp onse, a nd b eha viora l test m ethod s. Am eric a n Journa l of
Aud iolog y 12(2): 125–136.
Sw a nep oel, D-W., Hug o, R., a nd Rood e, R. 2004. Aud itory Stea d y-Sta te
21
University of Pretoria etd – Stroeb
bel, D (2006)
Resp onses for Child ren w ith Severe to Profound Hea ring Loss.
Arc hives Otola ryng olog y Hea d a nd Nec k Surg ery 130(5): 531-535.
Sw a nep oel, D.W., Sc hm ulia n, D., a nd Hug o, R., 2002. The Effec tiveness of
the Aud itory Stea d y Sta te Resp onse in Dia g nosing Hea ring Loss in
Infa nts. Hea lth SA Gesond heid 7(4): 47-56.
Sw a nep oel, D.W., 2001. Estim a ting Pure Tone Beha viora l Threshold s w ith
the Dic hotic Multip le Freq uenc y Aud itory Stea d y Sta te Resp onse
Com p a red to a n Aud itory Bra instem Resp onse Protoc ol in Norm a l
Hea ring Ad ults, Unp ub lished M Com m unic a tion Pa tholog y thesis,
University of Pretoria , South Afric a .
Tha rp e, A.M., a nd Ha ynes, D.S. 2005. Aud itory Neurop a thy/ Dys-sync hrony:
A Mounta in or a Molehill. In: R.C. Seew a ld a nd J.M. Ba m ford (ed s.),
A Sound Found a tion throug h Ea rly Am p lific a tion: Proc eed ing s of the
Third Interna tiona l Conferenc e (p p . 271-278). Stä fa , Sw itzerla nd :
Phona k AG.
Va lente, M., Hosford -Dunn, H., a nd Roeser, R.J., 2000. Aud iolog y
Trea tm ent. New York: Thiem e.
Va ld es, J.L., Perez-Ab a lo, M.C., Ma rtin, V., Sa vio, G., Sierra , C., Rod rig uez,
E., a nd Lins, O. 1997. Com p a rison of Sta tistic a l Ind ic a tors for the
Autom a tic Detec tion of 80 Hz Aud itory Stea d y Sta te Resp onses. Ea r
a nd Hea ring 18(5): 420-429.
Va n d er Reijd en, C.S., Mens, L.H.M., a nd Snik, A.F.M. 2005. EEG Deriva tions
22
University of Pretoria etd – Stroeb
bel, D (2006)
Provid ing Aud itory Stea d y Sta te Resp onses w ith Hig h Sig na l-to-Noise
Ra tios in Infa nts. Ea r a nd Hea ring 26(3): 299-309.
Va nd er Werff, K.R., a nd Brow n, C.J. 2005. Effec t of Aud iom etric
Config ura tion on Threshold a nd Sup ra threshold Aud itory Stea d ySta te Resp onses. Ea r a nd Hea ring 26(3): 310 – 326.
Va nd er Werf, K.R., Brow n, C. J., Giena p p , B.A., a nd Sc hm id t Cla y, K.M.
2002. Com p a rison of Aud itory Stea d y-Sta te Resp onse a nd Aud itory
Bra instem Resp onse Threshold s in Child ren. Journa l of the Am eric a n
Ac a d em y of Aud iolog y 13(5): 227-235.
Yoshina g o-Ita no, C. 2004. Levels of Evid enc e: Universa l New b orn Hea ring
Sc reening (UNHS) a nd Ea rly Hea ring Detec tion a nd Intervention
System s (EHDI). Journa l of Com m unic a tion Disord ers 37(5): 451-465.
23
University of Pretoria etd – Stroeb
bel, D (2006)
Ap p end ix A
1
University of Pretoria etd – Stroeb
bel, D (2006)
Ap p end ix B
Re se a rc he r: De id ré Stro e b e l
Te l: 021 930 3136
21 July 2004
To Whom It Ma y Concern:
Proposed resea rch project:
The clinica l Applica tion of ASSR in the Dia gnosis and fitting of Hea ring Aids in Infa nts (0 – 8 months)
Tha nk yo u fo r c o nsid e ring fo r yo ur c hild to b e p a rt o f this re se a rc h p ro je c t. The p o sitive re sults o f e a rly
id e ntific a tio n o f he a ring lo ss in infa nts o n d iffe re nt a sp e c ts o f the ir d e ve lo p m e nt a re w e ll kno w n. Diffe re nt
m e tho d s a re use d to o b ta in info rm a tio n a b o ut infa nts’ he a ring sta tus. The se m e tho d s d iffe r fro m the tec hniq ue s
use d o n a d ults. As te c hno lo g y im p ro ve s, ne w m e tho d s b e c o m e a va ila b le tha t sho w s a lo t o f p ro m ise in the
fie ld o f p e d ia tric a ud io lo g y.
I a m c urre ntly p la nning a re se a rc h stud y in this reg a rd a s p a rt o f the re q uire m e nts fo r a m a ste r’ s d e g re e a t the
Unive rsity o f Pre to ria . The p ro p o se d p ro je c t invo lve s d e te rm ining the c linic a l va lue o f Aud ito ry Ste a d y Sta te
Re sp o nse s (ASSR) a s a w a y to p re d ic t he a ring thre sho ld s, a nd to e va lua te he a ring a id s in yo ung infa nts. The
re sults fro m the he a ring a sse ssm e nts w ill b e m o nitore d a nd c o m p a re d w ith the re sults o f tw o c linic a lly p ro ve n
p ro c e d ure s, fre q ue ntly use d to d e te rm ine he a ring thre sho ld s, na m e ly the Aud ito ry Bra inste m Re sp o nse (ABR)
a nd Pure to ne Aud io m e try, fo r a p e rio d o f tim e . The stud y w ill b e c o nd uc te d und e r the sup e rvisio n o f p e rso nne l
a t the De p a rtm e nt o f Co m m unic a tio n Pa tho lo g y.
Pro c e d ure s c urre ntly inc lud e d in the sta nd a rd te st p ro to c o l use d to a sse ss infa nts in m y p riva te p ra c tic e , invo lve
the fo llo w ing :
•
The d ia g no stic se ssio n – inc lud ing ABR a nd ASSR.
•
M e a suring the g a in fro m he a ring a id s thro ug h ASSR
•
Be ha vio ra l te sting a fte r the a g e o f 6 m o nths.
As a c lie nt o f this p ra c tic e the se p ro c e d ure s w ill a lso b e use d to e va lua te a nd m o nito r yo ur b a b y’ s hea ring . All
o f the a b o ve p ro c e d ure s a re no n-inva sive , no p a in is invo lve d a nd the ABR a nd ASSR p ro c e d ure s a re no rma lly
d o ne w hile the b a b y is sle e p ing . If se d a tio n sho uld b e ne e d e d , this w ill b e d o ne in c o nsulta tio n w ith a
p e d ia tric ia n a nd w ith the ne c e ssa ry m e d ic a l sup e rvisio n. No a d d itio na l c o sts w ill b e c ha rg e d fo r the Aud ito ry
Ste a d y Sta te Re sp o nse te st, a s the va lue o f this test is still b e ing re se a rc he d .
I w o uld like to re q ue st yo ur c o nse nt fo r yo ur b a b y’ s p a rtic ip a tio n; p e rm issio n to use the re sults o f yo ur b a b y’ s
ro utine he a ring te sts; a s w e ll a s p e rm issio n to use info rm a tio n fro m yo ur b a b y’ s re c o rd s fo r this re sea rc h p ro je c t.
Yo u ha ve m y a ssura nc e tha t no unne c e ssa ry te sts w ill b e d o ne . Yo u m a y a lso w ithd ra w yo ur c hild fro m the
stud y a t a ny tim e .
M yse lf, m y sup e rviso r, M r. De We t Sw a ne p o e l, o r Prof. B. Lo uw , he a d o f the De p a rtm e nt o f Co m m unic a tio n
Pa tho lo g y a t the Unive rsity o f Pre to ria m a y b e c o nta c te d , sho uld yo u ne e d a ny furthe r info rm a tio n.
Tha nk yo u fo r yo ur a ssista nc e .
De id ré Stro e b e l
Resea rcher
M r. De We t Sw a ne p o e l
Supervisor
Pro f. B. Lo uw
HEAD: Depa rtment of Communica tion Pa thology
2
University of Pretoria etd – Stroeb
bel, D (2006)
Surna m e: __________________________ Na m e: ____________________________
I ha ve rea d the letter of inform a tion reg a rd ing Mrs. D. Stroeb el’ s p rop osed
resea rc h stud y.
I und ersta nd w ha t is involved a nd g ive p erm ission tha t the test results of
m y c hild __________________________________ m a y b e used .
____________________________
___________________________________
Sig na ture
Da te
3
University of Pretoria etd – Stroeb
bel, D (2006)
Ap p end ix C
DATA RECORDING SHEET
Sub jec t: ___________________________
ABR
Clic k
Tone Burst
Da te:
Ag e a t tim e of a ssessm ent:
500 Hz
ASSR
1000 Hz
2000 Hz
4000 Hz
Una id ed
Mea sured
Una id ed
Pred ic ted
Aid ed
Mea sured
Una id ed
Pred ic ted
Da te (una id ed ):
Da te (a id ed ):
Ag e a t tim e of a ssessm ent:
Ag e a t tim e of a ssessm ent:
500 Hz
Beha viora l thresholds
1000 Hz
2000 Hz
4000 Hz
Una id ed BT
Aid ed BT
Da te (una id ed ):
Da te (a id ed ):
Ag e a t tim e of a ssessm ent:
Ag e a t tim e of a ssessm ent:
4
Fly UP