Download Real Ear Unaided Responses in Ears with Tympanic Membrane

J Am Acad Audiol 3 : 60-65 (1992)
Real Ear Unaided Responses in Ears with
Tympanic Membrane Perforations
Colleen L. Moryl*
Jeffrey L. Danhauer*
Joseph R. DiBartolomeot
Abstract
This study evaluated 13 adult patients with tympanic membrane (TM) perforations in one or
both ears . Probe microphone real ear unaided responses (REURs) for mean peakfrequency
and peak dB gain showed no substantial differences among patients, across ears with intact
TMs, ears with TM perforations (regardless of size), or between measures taken before and
after closure of the perforations . Inspection of individual differences in the patients' REURs,
however, produced an interesting finding. The REURs of ears having small TM perforations
did not differ from those with intact TMs, but the REURs for ears having larger perforations
consistently revealed bimodal responses (two prominent peaks separated by a valley of at
least 10 dB lower gain) . Thus, size of the perforation affected the REURs.
Key Words: Hearing aids, real ear probe microphone measures, real ear unaided
response (REUR), tympanic membrane perforations
ir-conduction hearing aids are frequently considered for patients with persistA ent tympanic membrane (TM) perforations when surgery is not an option . The use of
probe microphone measures for hearing aid
assessment has become routine in clinic and
research settings . Measurement of the real ear
unaided response (REUR) "refers to the measured frequency response in dB SPL with a probe
microphone at a specified point within the unaided, unoccluded external auditory canal for a
specified sound field" (Schweitzer et al, 1990) .
The REUR is often the first measure taken
when fitting hearing aids by evaluating the real
ear insertion response (REIR) with probe microphone systems.
The REUR may also be helpful in determining if physical and acoustic characteristics of
the ear canal are altered by TM perforations
and whether corresponding compensations must
'Department of Speech and Hearing Sciences, University of California, Santa Barbara, California
t The Ear Foundation, Santa Barbara ; and Department
of Otolaryngology, University of California, Los Angeles,
California
Reprint requests : Jeffrey L . Danhauer, Department of
Speech and Hearing Sciences, Snidecor Hall, University of
California Santa Barbara, Santa Barbara, CA 93106
60
be made in fitting hearing aids to patients
having perforations . Generally, an unoccluded
ear without a TM perforation has a natural
resonance near 2700 Hz (i .e ., in the range of
2500 to 3000 Hz) on the order of 10 to 20 dB
(Hawkins and Mueller, 1986 ; Libby,1986,1987 ;
Dirks and Kincaid, 1987 ; Hawkins, 1987 ;
Gallagher, 1989). These characteristics, however, have not been documented for ears of
patients having TM perforations . TM perforations may alter acoustic and physical characteristics of the ear and produce an REUR that
may differ from that of an ear with an intact TM .
Typically, inserting a hearing aid into an
ear reduces the natural gain observed in the
REUR . Hearing aids are often selected to provide additional gain in the 2500 to 3000 Hz
region to compensate for this insertion loss . If a
TM perforation causes the insertion loss to
occur at different frequencies than typically
observed for the intact TM, and if the dispenser
is unaware of this, then the compensation provided by the hearing aid may be at the wrong
frequencies for a given patient. Therefore, a
knowledge of alterations in the REUR produced
by TM perforation is important because proportional compensations may be necessary or desirable in the frequency response of hearing
aids fitted to these patients .
Tympanic Membrane Perforations/Moryl et al
The purpose of this study was to perform
probe microphone measurements on a sample
of patients having TM perforations to document the effects ofperforation size on the REUR
(peak frequency and peak amplitude) .
METHOD
Patients
Thirteen patients (six males and seven females) ranging from 17 to 88 (mean = 54) years
of age were obtained from an otolaryngology
practice over a 3-month period . All were diagnosed with TM perforations (10 unilateral and
3 bilateral) by an otolaryngologist (JRD) . The
perforations varied in size . Two patients had
tympanoplasties and were tested before and
after surgical closure. One patient's perforation
healed naturally and was also analyzed before
and after closure.
Stimulus and Instrumentation
The speech-weighted composite signal (80
sinusoids from 100 to 8000 Hz spaced at 100 Hz
intervals) provided by the Frye 6500 real ear
probe microphone system was used as the stimulus for this study. This system consists of a
reference microphone, a probe microphone coupled to a 0.95-mm outer diameter probe tube of
soft, flexible silicone, and a 77-mm diameter
soundfield loudspeaker located on an adjustable stand.
Procedure
Probe microphone measures were conducted
in a room within an otolaryngology office . The
patient was seated 18 inches from the soundfield
loudspeaker. The loudspeaker was adjusted in
a horizontal plane to the patient's ear at 45
degrees azimuth (re: the patient's nose). The
reference microphone was attached to a Velcro
headband placed just above the patient's pinna
and facing forward. The probe tube was marked
at 22 mm from the end of the probe tip. The
probe tube was inserted 22 mm into the ear
canal with the marked portion referenced to the
opening ofthe meatus . Thus, assuming an average ear canal length of 24 mm (Zemplenyi et al .,
1985), the measures were made at a distance of
about 2 mm from the TM . The insertion depth
was verified by otoscopic inspection of the ear
canal. This probe-tube position was used for all
measures .
Patients were instructed to remain still
while the real ear measurements were obtained .
When the reference and probe microphones
were both in place, the probe tube was inserted
into the patient's ear canal and the system was
leveled (calibrated) using the composite signal .
The REUR was then obtained for the ear having
the TM perforation using the composite signal
at 70 dB SPL. Each REUR was immediately
stored and printed out in both graphic and
numeric data forms. The probe tube was then
removed from the patient's ear and reinserted
to the same insertion depth in the same ear
canal for a second measurement. All test-retest
measures for each patient were repeatable
within 2 dB of each other (except at 200 and 300
Hz) and were, therefore, considered reliable .
For the 10 patients with one intact eardrum, an REUR was also obtained for that ear
as a control using the same procedures as above.
Thus, four REURs were collected for each patient. Both ears were tested for the three patients with bilateral TM perforations (i .e ., they
did not have a control ear) . This resulted in a
total of 16 perforations for analyses . Three of
the patients (two after surgical closure, and one
after his perforation healed naturally) were reevaluated using the same procedures as used
for the first evaluation .
RESULTS
P
eak frequency (in Hz) and peak amplitude
(in dB gain) values for the ears having TM
perforations and for the control (intact TM) ears
are shown in Table 1. The table shows essentially no differences between the mean peak
frequencies and mean peak dB gain of the
perforated and control ears when the perforations are treated without regard to size . However, the table shows some individual variability among ears . The range for peak frequency
was 1700 to 3700 Hz and that for dB gain was 8.7
to 25 .3 dB for the perforated ears . For the
control ears, the range was somewhat smaller
(2300 to 2900 Hz) for peak frequency, but greater
for peak dB gain (9 .5 to 31 .9 dB).
Not shown in the table, is the fact that
several of the REURs for the perforated ears
had two prominent peaks separated by a valley
of at least 10 dB reduced gain (referred to here
as a bimodal response). Figure 1 is an example
of one such case . Note that, for this patient, the
valley actually resulted in negative gain . This
bimodal response was noted for 7 of the 16 ears
having perforated TMs. Responses were only
61
Journal of the American Academy of Audiology/Volume 3, Number 1, January 1992
Table 1
Peak Frequency and dB Gain Points
in the REURs for Each Ear
Peak Frequency (Hz)
Patient
Perforation
R
L
R
L
R
L
Mean =
SD =
Control
2700
3400
1900
2300
2400
2200
1700
2500
3000
3700
2600
2600
2300
2100
2100
3000
2600
2531 .25
535.06
2570
170 .29
2500
2300
2800
2900
2500
2500
2600
2500
2500
Peak Gain (dB)
Perforation Control
22 .3
23 .1
24 .7
17 .6
24 .1
12 .4
25 .3
22 .4
22 .2
16 .0
9 .1
9 .8
8 .7
18 .5
25 .0
21 .9
18 .94
5 .98
23 .5
12 .9
23 .5
19 .6
18 .7
31 .9
9 .5
16 .7
16 .1
22 .3
Small
Patient
1
2
3
4
5
6
7
8
9
10
11
12
13
Mean =
19 .47
6 .30
Note : 10 of the patients had one intact eardrum that
served as a control ear ; three had bilateral perforations and
responses were obtained for both right (R) and left (L) ears .
analyzed up to 4000 Hz since data points above
that cut-off are considered artifactual .
The data were separated according to small
(3 mm or less), medium (3 to 5 mm), and large
(greater than 5 mm) TM perforations to determine whether size of the perforation produced
differences in the REURs . The peak frequency
and dB gain points for the REURs for the three
perforation sizes are shown in Table 2. Aside
from the lower values obtained for the medium
sized perforations (note: only an n of 3), no
apparent differences were observed among the
mean values for size . However, seven of the
ears with perforations exhibited bimodal
REURs such as that noted earlier in Figure 1 :
the REURs for all six ears having large perforations and one of the three ears having medium
perforations .
Figures 2, 3, and 4 show the REURs for
three patients before and after closure of the
perforations . Patient 6 (shown in Figure 2) had
a small pinhole perforation, which healed naturally, and there was no large difference between
the REURs for his ear with perforated and
intact (control) TMs. After closure, the REUR
for the ear with the healed TM and his control
ear were exactly the same . The variation observed between this patient's pinhole perforation and his control TM was no greater than
might be expected for two intact TMs of any
individual .
62
Table 2 Peak Frequency and dB Gain Points
in the REURs for Small, Medium, and
Large Perforations
Hz
dB Gain
2700
3400
22 .3
23 .1
2400
24 .1
2500
3000
22 .4
22 .2
2100
2100
18 .5
21 .9
2600
22 .1
Large
Medium
Hz
dB Gain
1900
2300
24 .7
17 .6
1700
25 .3
Hz
dB Gain
2200
12 .4
3700
2600
2600
2300
16 .0
9 .1
9 .8
8 .7
3000
25 .0
2733 .3 13 .5
1966 .7 22 .5
Perforation sizes : small = 3 mm or less ; medium = 3
to 5 mm ; and large = greater than 5 mm .
The other two patients (10 and 13), shown
in Figures 3 and 4, respectively, had large
perforations and received tympanoplasties.
Before surgery, both patients' REURs demonstrated the bimodal configuration; after surgery they both exhibited only one prominent
peak . After surgery, Patient 10's ear with the
healed TM revealed a similar response to that
with his intact TM . Recall that Patient 13 had
bilateral TM perforations (one small and one
large) . As noted earlier, his ear with the large
perforation exhibited a bimodal REUR before
surgery. After surgery, his REUR revealed only
one peak characteristic of ears with intact TMs.
dB
---
30 1
--
20~---------------------------------10
.25
.5
1
2
4 KHz
Figure 1 Example of a bimodal REUR for a patient
having a large TM perforation. Note the two large peaks
in the response separated by a valley of low gain.
dtfji: li
Iih
1
hi -ai11Ml
Tympanic Membrane Perforations/Moryl et al
dB
10
00
-10
-20
.25
(a)
.5
1
2
4 KHz
dB
.25
(b)
.5
1
2
4 KHz
Figure 2 REURs for Patient 6 before (a) and after (b)
closure of the TM perforation .
DISCUSSION
Ithough small perforations produced sinAgle-peaked REURs similar to those for
ears with intact TMs, ears for one medium and
all large perforations in this study exhibited
bimodal REURs (see Fig. 1) . For the two large
perforations in this study that were monitored
through closure, the REURs for the ears with
healed TMs were also similar to those with
intact TMs. However, the responses observed
for ears with one of the medium and all of the
large perforations suggest that, if hearing aids
were fitted to these ears, these perforations
could create noticeable changes in the REIR.
Specifically, the two peaks observed in the
bimodal REURs for these patients would create
insertion gain and loss at different frequencies
from those typically noted for ears with intact
TMs. Normally, a smooth insertion response is
desirable in hearing aid fittings . Although a
smooth response can also be obtained for ears
.25
(b)
.5
1
2
4 KHz
Figure 3 REURs for Patient 10 before (a) and after (b)
closure of the TM perforation
having TMs with larger perforations, the additional peaks and valley in the bimodal response
can present a challenge in hearing aid fittings .
In our experience, we have not had difficulty fitting hearing aids to ears with small
perforations . However, Figure 5a provides an
example of a REUR from a patient with a large
TM perforation showing a bimodal response
having two prominent peaks (at about 800 and
2500 Hz) and a large valley (at about 1200 Hz) .
Interestingly, the REIR in Figure 5b shows
three corresponding large peaks at about 400,
1200, and 2500 Hz and two valleys (at about 800
and 2000 Hz). This patient complained of having loudness intolerance and speech discrimination problems with several behind-the-ear
(BTE) hearing aids that were fitted to his loss .
These complaints probably arose because the
REIR was not smooth . We suspect that the
patient tried to turn up the volume on the
hearing aid to fill in the information that was
missing from the valleys in the spectrum . In
63
Journal of the American Academy of Audiology/Volume 3, Number 1, January 1992
.25
(a)
.25
(b)
.5
.5
1
2
1
2
4 KHz
.25
.5
1
2
4 KHz
4 KHz
Figure 4 REURs for Patient 13 before (a) and after (b)
closure of the TM perforation.
doing so, excess gain was present in the spectrum where the peaks occurred, thus, causing
discomfort and potential distortion of speech.
The use of filters in the ear hook and in the
earmold tubing helped somewhat . Here, it is
important to have a hearing aid with a flexible
response . Some digitally programmable hearing aids may allow the dispenser to program a
response that closely approximates a chosen
prescription fitting. This type of case demonstrates the importance of using probe microphone measures in hearing aid fittings, specifically when TM perforations are present.
Electroacoustic test box or functional gain measures might fail to identify the peaks and the
valley in the unaided response due to the perforation causing the patient's complaints .
In summary, our results show that the ears
havinglarge perforations had abnormal REURs.
In fitting hearing aids, the increased cavity size
produced by large TM perforations would likely
require greater amounts of gain at some fre64
(a)
(b)
Figure 5 REUR of a patient with a large TM perforation (a). REIR for a hearing aid fit (b) to the ear in (a). The
dark line is the target curve. These measures were taken
with a Madsen IGO system rather than the Frye 6500
used in the other figures.
quencies and less at others than would typically
be needed for ears having intact TMs. This
suggests that special compensations may be
needed in fitting hearing aids to patients having large TM perforations . Although this conclusion is based on a small number of patients,
all those having large TM perforations in this
study exhibited similar bimodal REURs, which
differed from REURs seen in ears with intact
TMs and in those with small perforations .
Acknowledgment. This research was supported by a
University ofCalifornia Santa Barbara General Research
Grant (8-587529-19900-7), a Graduate Student Humanities/Social Sciences Research Grant, and the Ear Foundation of Santa Barbara.
The authors acknowledge David F. Henry and Maxine
DiBartolomeo for their assistance in acquiring patients
for this study. Special appreciation is expressed to H.
Gustav Mueller, Lucille Beck, Dennis J. Arnst, Larry
Revit, and Kimberly J. Danhauer for their valuable comments on this manuscript.
Tympanic Membrane Perforations/Moryl et al
REFERENCES
Libby ER . (1986) . The shift toward real ear measurements. Hear Instr 37 : 6, 50 .
Dirks DD, Kincaid GE . (1987) . Basic acoustic considerations of ear canal probe measurements . Ear Hear 8:6067S.
Libby ER. (1987) . Real ear considerations in hearing aid
selection. Hear Instr 38 : 14-16.
Gallagher G. (1989) . Real-ear measurement . Hear J 42 :
11-17.
Hawkins DB . (1987) . Clinical ear canal probe tube measurements . Ear Hear 8: 74-818 .
Hawkins DB, Mueller HG . (1986) . Some variables affecting the accuracy of probe tube measurements . Hear Instr
37 : 8, 48 .
Schweitzer HC, Sullivan RF, Beck LB, Cole WA . (1990) .
Developing a consensus for "real ear" hearing instruments terms. Hear Instr 41 : 28, 46 .
Zemplenyi J, Gilman S, Dirks D. (1985) . Optical method
for measurement of ear canal length. JAcoust Soc Am 78 :
2146-2148.