THIEME

336 Original Research

Frequency Dependence Hearing Loss Evaluation

in Perforated Tympanic Membrane
Mohammed Radef Dawood1

1 Department of Otolaryngology, AL Mustansiriya University, College Address for correspondence Mohammed Radef Dawood, FCIMS,
of Medicine, Baghdad, Iraq Department of Otolaryngology, AL Mustansiriya University, College of
Medicine, Baghdad 10001, Iraq (e-mail: [email protected]).
Int Arch Otorhinolaryngol 2017;21:336–342.

Abstract Introduction Tympanic membrane perforation is a relatively common problem that

predisposes patients to varying degrees of conductive hearing loss.
Objective The objective of this study is to evaluate and analyze the frequency
dependence hearing loss in tympanic membrane perforation based on the size and
the site of perforation.
Methods For the study, I selected 71 patients’ (89) ears for the cross-sectional study
with tympanic membrane perforations; I examined the size and the site of perforations
under the microscope and classified them into small, moderate, large, and subtotal
perforations, and into anterior central, posterior central, malleolor central, and big
central perforations. I measured mean level of speech frequencies hearing loss, and its
relation with the site and the size of the perforation analyzed.
Results The mean hearing loss at different sizes of the perforation at all speech
frequencies was 37.4 dB, with ABG of 26.6 dB, and its maximum loss was detected in
subtotal perforation of 42.3 dB, with ABG of 33.7 dB, at 500 Hz frequency, while in
relation to the sites, it was 38.2 dB, with ABG of 26.8 dB, and its maximum loss was
detected in big central site perforation of 42.1 dB, with ABG of 33.6 dB, at 500 Hz
frequency.
Keywords Conclusions The hearing loss was proportionally related with the sizes of the
► tympanic membrane perforations, and the posterior site had greater impact on the hearing than anterior
► perforation site perforations. This was also applied to the frequency dependence hearing level, as
► hearing loss was detected to be worse at lower frequencies as 500 Hz, than those of 1000–2000 Hz.

Introduction when the movements of ossicles were hypothesized not im-

paired, this hypothesis proposed that the only auditory struc-
In the normal ear the difference in sound pressure that ture that is responsible for hearing loss in this case is the
develops between the external and the middle ear is respon- perforation of the tympanic membrane.2
sible for sound transmission that causes the vibratory move- In addition to this sound wave transmission across the
ments of the tympanic membrane and the lever action of the middle ear cleft, and of special concern, it serves a pro-
ossicles.1 tective function that shields the round window niche from
In case there is a perforation in the tympanic membrane, direct sound waves which is referred as “round window
this is going to reduce the surface area that is available for this baffle”.3
sound pressure transmission. Therefore, the sound pressure This shielding effect is a matter of debate. Some studies do
gradient becomes effectively insignificant, as is determined by not support the hypothesis of a shielding effect of the
the loss of the impedance of the tympanic membrane. Thus, tympanic membrane.1,4

received DOI https://doi.org/ Copyright © 2017 by Thieme Revinter

July 28, 2016 10.1055/s-0037-1598597. Publicações Ltda, Rio de Janeiro, Brazil
accepted ISSN 1809-9777.
November 3, 2016
published online
February 28, 2017
 Frequency Dependence Hearing Loss Evaluation in Perforated Tympanic Membrane Dawood 337

So the pressure is increased by the ratio of the two areas, An examination of the ears was done under a Carl Zeiss
and it is considered the virtual factor in establishing the microscope with 200 mm lenses. A single surgeon evaluated
impedance transformation, since it is larger than that of the the size of the perforation by applying the software package
stapes footplate which is 3.2 mm2, the pressure exerted on AutoCAD classic 2015 program, which measured the total
the stapes footplate is increased by 60/3.2 ¼ 18.75.5 area of tympanic membrane and perforation, and calculated
Whereas, in the case of a perforation in the tympanic the percentage of the perforation according to the following
membrane, the sound stimulus might be insufficiently formula:
coupled to the tympanic membrane and the impedance
transformer action might be lost. Then, the ability to move
the middle ear ossicles might be impaired.6
The perforation of the tympanic membranes is a relatively
common problem observed in ENT practice, and it usually Thus, the perforation of the tympanic membrane was
results from various factors such as infection, trauma, and classified as follows: small perforation when it constituted
sometimes iatrogenic causes. It can predispose to conductive less than 25% of the whole tympanic membrane; moderate
hearing loss, although its range usually does not exceed size perforation if constituted 25–50% of the whole
50 dB (this is not surprising because 60 dB is the maximal tympanic membrane; large perforation when constituted
conductive hearing loss). Moreover, hearing loss depends 50–75% of the whole tympanic membrane; and subtotal
upon the frequency, perforation size, and air space volume of perforation if constituted more than 75% of the whole
the middle ear.1 Various authors believe the site of the tympanic membrane. The site of perforation was classified
perforation has a significant impact on the level of the in relation to the handle of the malleus into: anterior
hearing loss,6 while other researchers believe that it has central, posterior central, malleolar central, and the big
no significant effect.7 central, when it involved all the quadrants of the tympanic
Given this divided conception, I set out to investigate the membrane, as shown in ►Fig. 1.
relationship between the site, the sizes of the tympanic I assessed patients’ hearing levels in decibel with AMBCO
membrane perforations, and the degree of the conductive 650 clinical diagnostic pure tone audiometry, which was
hearing loss, with a view to contribute to the body of recently calibrated according to international organization of
knowledge on this issue. standardization (ISO) standard in a soundproof room at
Hearing loss is a national health problem and has a frequencies of 250, 500, 1000, 2000, 4000, and 8000 Hz,
significant impact on the person’s physical and psychosocial and the air. I assessed bone conduction hearing threshold
condition. Therefore, it is essential for early recognition and with appropriate masking technique whenever required. I
effective management of the tympanic membrane perfora- determined hearing level by calculating the mean (average)
tion, since the untreated tympanic membrane perforation of air conduction and air-bone (ABG) at the speech frequen-
may lead to progressing destructive changes in the middle cies of 500, 1000, and 2000 Hz.
ear cavity, thus adding to further deterioration in the hearing The ethical and the scientific committee approved the
threshold.8 Also, the need for water protection to prevent study and participating patients gave their consent.
infections is sometimes very troublesome.
The aim of the study was to evaluate and analyze the
frequency dependence hearing loss in tympanic membrane
perforation based on the size and site of perforation.

Patients and Methods

This is a cross-sectional study conducted at the ENT clinic
from June 2015 to June 2016, which included 71 patients
(89) ears complaining of hearing difficulty, attributed to
their tympanic membrane perforations. I assessed the cases
based on relevant history and full ENT examination and the
selection was according to the following inclusion criteria;
age above 18 years, history of the perforation was within
one year of presentation, dry central type of tympanic
membrane perforation for more than 3 months, middle
ear cavity without abnormalities as cholesteatoma, polypi,
or granulation tissue, intact ossicular chain was checked by
paper patch test, functioning Eustachian tube was tested by
Toynbee’s test via the impedance audiometer, mastoid
status (no sign of mastoiditis), no middle ear surgery, a
conductive deafness (with adequate cochlear reserve, i.e., Fig. 1 Measurement of the size of the perforation. Abbreviations: P,
not mixed hearing loss). perforation; TM, tympanic membrane.

International Archives of Otorhinolaryngology Vol. 21 No. 4/2017

338 Frequency Dependence Hearing Loss Evaluation in Perforated Tympanic Membrane Dawood

Table 1 Tympanic membrane perforation sizes Results

Size No. Percentage The current study comprised of 71 patients (89 ears): 53
Small 16 17.9% (74.6%) patients with unilateral ear involvement and 18
Medium 21 23.5% (25.3%) patients with bilateral ear involvement. There were
48 men (67.6%) and 23 women (32.3%); mean age was 27.579
Large 23 25.8%
(8.649) years. With regards to the etiology of the perforation,
Subtotal 29 32.5% 73 ears (82%) were due to CSOM, and 16 ears (17.9%) resulted
from old traumatic tympanic membrane perforation.
The distribution of frequencies of the sizes and the sites of
Table 2 Tympanic membrane perforation sites
the tympanic membrane perforation are shown in ►Table 1
and ►Table 2, respectively.
Site No. Percentage
The mean hearing loss, irrespective to the sizes and the sites
Anterior central 18 20.2% of the perforations and at all speech frequencies, was 37.8 dB.
Posterior central 21 23.5% I analyzed the relation between the mean hearing loss and
Central malleolar 24 26.9% the size of the tympanic membrane perforation, which re-
vealed a mean hearing loss of 32.2 dB in small size perforation,
Big central 26 29.2%
36.4 dB in medium perforation, and 39.6 dB in large size
perforation, whereas in subtotal perforation it was 41.5 dB.
The relation between the mean hearing loss and the site of
the tympanic membrane perforation revealed that the mean
Statistical Analysis hearing loss in anterior central perforation was 34.7 dB, in
I performed the statistical analysis using the software posterior central perforation was 37.8 dB, and in malleolar
Statistical Package for Social Sciences (version 17; SPSS central perforation was 39.2 dB, while in the big central
Inc, Chicago, Illinois, USA). The sites and sizes of the perforation it was 41.2 dB.
tympanic membrane perforations were separately corre- The relation between the mean hearing loss at different
lated with the magnitude of hearing losses through Pear- speech frequencies in 89 ears irrespective to the sizes and the
son’s test. I express data as mean standard deviation sites of the tympanic membrane perforations was: 38.8 dB at
(SD) for parametric data and as numbers and percent 500 Hz frequency, 38.1 dB at 1000 Hz, and 37.8 dB at 2000 Hz.
for non-parametric data. I applied the t-test wherever I analyzed the relation between the mean hearing loss at
necessary, with the p value <0.05 being considered sta- different speech frequencies and the sizes of the tympanic
tistically significant. membrane perforation (►Table 3). The mean hearing loss at

Table 3 Hearing level distribution with speech frequencies among the sizes of the perforations

Size Hearing loss Hearing loss Hearing loss Mean hearing p value
at 500 Hz at 1000 Hz at 2000 Hz loss
Small
mean 34 33.1 29.5 32.2 0.025
SD  3.273 6.942 11.791 5.35836
Medium
mean 37.1 36.3 35.8 36.4 0.003
SD  9.328 8.103 8.343 6.34011
Large
mean 39.7 39.6 39.5 39.6 0.006
SD  7.710 11.060 8.427 6.65332
Subtotal
mean 42.3 41.4 40.8 41.5 0.005
SD  8.168 11.392 9.996 8.21201
Total
mean 38.3 37.6 36.4 37.4 0.001
SD  9.527 11.183 9.361 8.08674

Abbreviations: SD, standard deviation; Hz., Hertz.

International Archives of Otorhinolaryngology Vol. 21 No. 4/2017

 Frequency Dependence Hearing Loss Evaluation in Perforated Tympanic Membrane Dawood 339

Table 4 Hearing level distribution with speech frequencies among perforation sites

Site Hearing loss at 500 Hz Hearing loss at 1000 Hz Hearing loss at 2000 Hz Mean hearing loss p value
Anterior central
mean 36.4 35.9 31.8 34.7 0.04
SD  3.361 2.653 6.251 2.01765
Posterior central
mean 38.7 37.8 36.9 37.8 0.003
SD  7.473 12.554 9.129 8.11767
Central malleolar
mean 40 39.1 38.5 39.2 0.002
SD  7.192 7.415 9.416 4.89134
Big central
mean 42.1 41.2 40.3 41.2 0.001
SD  8.289 11.648 11.569 8.7291
Total
mean 39.3 38.5 36.9 38.2 0.0001
SD  8.527 10.183 8.360 7.18465

Abbreviations: SD, standard deviation; Hz., Hertz.

all frequencies irrespective to the sizes of the tympanic the maximum ABG was detected with big central perforation
membrane perforation was 37.4 dB. The maximum hearing 33.6 dB at 500 Hz frequency, and minimum ABG was detected
loss detected in subtotal perforation at 500 Hz frequency with anterior central perforation 18.8 dB at 2000 Hz
was 42.3 dB, while the minimum hearing loss detected in frequency.
small size perforation at 2000 Hz frequency was 29.5 dB.
I studied the relation between the mean of the hearing
Discussion
loss at different speech frequencies and sites of the tympanic
membrane perforation (►Table 4). The mean hearing loss at The tympanic membrane plays an important role in the
all frequencies irrespective to the sites of the tympanic tympano-ossicular system for sound transmission in the
membrane perforation was 38.2 dB. I detected a maximum middle ear.
hearing loss of 42.1 dB in big central perforation at 500 Hz Perforation of the tympanic membrane is one of the main
frequency, and minimum hearing loss of 31.8 dB in anterior causes of conductive hearing impairment, and its magnitude
site perforation at 2000 Hz frequency. is usually less than 50 dB.1
I also used another audiometric parameter in the assess- In the current study, 82% of the perforations resulted from
ment of the hearing loss, which was the air-bone gap (ABG). CSOM. This high incidence was probably related to patients’
The results showed mean ABG level at all speech frequencies poor cultural and socioeconomic level, as well as their
irrespective to the sizes and the sites of the perforation of life style. This finding is in agreement with that from Biswas
26.7 dB. et al.9
The relation of the ABG measurement in different speech The current study’s results for maximum frequencies of
frequencies for comparison with different sizes of the tym- tympanic membrane perforation detected in subtotal sized
panic membrane perforation was shown in ►Table 5, where and big central located perforations as well as for mean
the the maximum ABG level irrespective to the all speech hearing loss in all enrolled ears, irrespective to the site and
frequencies was 33.5 dB in subtotal perforation, and mini- the size parameters, was almost comparable with those of
mum ABG level was 18.5 dB in small size perforation. As for Nahata et al10 and Ristovska et al.11
the relation to specific speech frequency, the maximum ABG The current study had revealed a linear correlation be-
detected with subtotal size perforation 33.7 dB at 500 Hz tween the sizes of the perforation and the hearing loss, as the
frequency, and minimum ABG detected with small size highest degree of hearing loss detected with subtotal size
perforation was 17.8 dB at 2000 Hz frequency. perforation with the air conduction was 41.5 dB with an ABG
ABG measurements for comparison with different sites of of 33.5 dB. Thus, the greater the increase in the size of the
the tympanic membrane perforation in relation with the tympanic membrane perforation meant a steeper decline in
speech frequencies are shown in ►Table 6. The maximum auditory perception. This observation is supported by several
ABG detected irrespective of the frequencies was 33.4 dB in studies,2,8,12,13 and the explanation for it is attributable to
big central perforation, and minimum ABG was 19.4 dB in the impedance matching impairment of the hydraulic action
anterior central perforation. As for specific speech frequency; of the tympanic membrane, as it is related to the reduction in

International Archives of Otorhinolaryngology Vol. 21 No. 4/2017

340 Frequency Dependence Hearing Loss Evaluation in Perforated Tympanic Membrane Dawood

Table 5 Mean ABG with different perforation sizes in relation to speech frequencies

Size ABG ABG ABG Mean ABG loss p value

at 500 Hz at 1000 Hz at 2000 Hz
Small
mean 19.3 18.4 17.8 18.5 0.046
SD  3.142 4.811 5.670 4.22725
Medium
mean 25.7 24.8 23.9 24.8 0.05
SD  6.436 5.201 5.451 3.42011
Large
mean 30.1 29.2 28.9 29.4 0.003
SD  7.047 10.281 8.975 7.19070
Subtotal
mean 33.7 33.6 33.2 33.5 0.002
SD  9.158 12.392 10.996 9.11081
Total
mean 27.2 26.5 26 26.6 0.0001
SD  8.527 10.183 8.360 7.08465

Abbreviations: ABG, air bone gap; Hz, Hertz; SD, standard deviation.

its surface area as compared with that of the stapedial threshold. In the current study, the level of the hearing loss
footplate. Thus, there was reduction in the ossicular coupling detected in posterior central sited perforation was 37.8 dB,
due to the sound pressure difference that had been abolished with ABG of 25.6 dB, while that at the anterior central site
across the tympanic membrane, which resulted in a decrease perforation was 34.7 dB, with ABG of 19.4 dB. This result was
of sound amplification, and therefore had significant impact statistically significant with P value of 0.001. The central
on auditory perception.14 malleolar and the big central were not included in this
However, there is little consensus between various authors comparison, since they involved both areas. Our finding was
on the effect of the site of the perforation on the hearing in agreement with other studies,4,10,15 and results were likely

Table 6 Mean ABG with different perforation sizes in relation to speech frequencies

Site ABG ABG ABG Mean ABG loss p value

at 500 Hz at 1000 Hz at 2000 Hz
Anterior central
mean 20.1 19.3 18.8 19.4 0.06
SD  4.583 3.875 3.471 3.22987
Posterior central
mean 26.3 25.6 24.9 25.6 0.002
SD  6.582 11.665 8.239 7.23879
Central malleolar
mean 28.9 28.8 28.4 28.7 0.003
SD  7.192 7.415 9.416 4.89134
Big central
mean 33.6 33.4 33.2 33.4 0.007
SD  8.298 11.648 11.569 8.72391
Total
mean 27.2 26.8 26.4 26.8 0.0001
SD  8.527 10.183 8.361 7.08464

Abbreviations: ABG, air bone gap; Hz, Hertz; SD, standard deviation.

International Archives of Otorhinolaryngology Vol. 21 No. 4/2017

 Frequency Dependence Hearing Loss Evaluation in Perforated Tympanic Membrane Dawood 341

due to the impairment of the potential effect of the “round location. The same applied to the frequency dependence of
window baffle” on hearing threshold. Other studies,12,13,16 hearing loss, as detected at lower speech frequencies. The
however, did not detect any significant differences in hearing results of the study indicated that hearing loss was worse at
loss in anterior versus posterior site of the perforations. Mehta lower frequencies, such as 500 Hz, than at higher frequen-
et al2 in their study also stated that hearing loss did not vary cies, such as 1000–2000 Hz, irrespective of the sizes and sites
substantially with site of the perforation and any such varia- of the tympanic membrane perforations.
tion was negligible. Thus, the shielding effect remains a subject
of debate.4
Most of the research that analyzes the relationship be- Conflicts of interest
tween the perforations in the tympanic membrane and Author declares no conflict of interest.
hearing thresholds at various frequencies concludes that
hearing loss is frequency dependent, with most maximum
losses detected at the lower frequencies.17 The current study
reveals the same observations, that is, hearing loss was References
1 Voss SE, Rosowski JJ, Merchant SN, Peake WT. Middle-ear function
greater at lower frequencies, and decreased as the frequency
with tympanic-membrane perforations. I. Measurements and
increased, as the maximum hearing loss at 500 Hz frequency mechanisms. J Acoust Soc Am 2001;110(3 Pt 1):1432–1444
was 38.8 dB in air conduction hearing loss, and 27.2 dB in 2 Mehta RP, Rosowski JJ, Voss SE, O’Neil E, Merchant SN. Determi-
ABG level. Our results were in agreement with international nants of hearing loss in perforations of the tympanic membrane.
studies, as those of Nahata et al,10 Ristavska et al,11 those on Otol Neurotol 2006;27(02):136–143
temporal bones by Ahmad and Ramani,14 Bigelow et al18 on 3 Ogisi FO, Adobamen P. Type 1 Tympanoplasty in Benin: a 10-year
review. Niger Postgrad Med J 2004;11(02):84–87
rats, in the clinical studies of Bhusal et al,19 Al-Omari, and Al-
4 Voss SE, Rosowski JJ, Merchant SN, Peake WT. How do tympanic-
Doski.20 membrane perforations affect human middle-ear sound trans-
Lerut et al,21 in their study, revealed that the frequency mission? Acta Otolaryngol 2001;121(02):169–173
pattern was similar to an “inverted V shape” at the audio- 5 Pickles JO. Physiology of hearing. In: Scott-browns Otolaryngol-
gram, with a turning point around 2000Hz, as the ABG ogy. Michael Gleeson, George GB, Martin JB, Ray C, John H,
Nicholas SJ, et al editors. 7th ed. London: Hodder Arnold; 2008:
(hearing loss) at frequencies below the 2000Hz frequency
3181
were larger than those above 2000 Hz frequency. Simlarly,
6 Peake WT, Rosowski JJ, Lynch TJ III. Middle-ear transmission:
Nahata et al10 concluded that the inherent frequency of the acoustic versus ossicular coupling in cat and human. Hear Res
tympanic membrane had been calculated to beat 2000 Hz, 1992;57(02):245–268
since the tympanic membrane vibrates the most at this 7 Roland NJ, McRae RDR, Mc Combe AW. Chronic suppurative
frequency. Röösli et al,22 however, observed that hearing otitis media. In: Key topics in Otolaryngology and head and
neck surgery. 2nd ed. Wales, Bios scientific publishers; 2001:
loss at 2000 Hz frequency cannot be explained by tympanic
38–41
membrane perforation alone. Nonetheless, it is worth 8 Pannu KK, Chadha S, Kumar D, Preeti. Evaluation of hearing loss in
noting here that I did not include the frequencies above tympanic membrane perforation. Indian J Otolaryngol Head Neck
2000 Hz in the calculation of the hearing threshold in the Surg 2011;63(03):208–213
current study. 9 Biswas SS, Hossain A, Alam M, Atiq T, Al-Amin Z. Hearing
The study by Austin23 disagreed with our finding, as he evaluation after myringoplasty. Bangladesh J Otorhinolaryngol.
2010;16(01):23–28
concluded that the presence of the perforation did not have a
10 Nahata V, Patil CY, Patil RK, Gattani G, Disawal A, Roy A. Tympanic
significant impact on frequency dependent hearing loss. His membrane Perforation: Its correlation with hearing loss and
audiometric pattern revealed a flat curve pattern at all frequency affected - An analytical study. Indian J of Otology.
speech frequencies (500, 1000, 2000 Hz), as well as in 2014;20(01):10–15
relation to each size of tympanic membrane perforation. 11 Ristovska L, Jachova Z, Filipovski R, Atanasovska N. Correlation
between tympanic membrane perforation. JSER 2016;17(02):
Since there was no further detailed information on his study,
36–49
it is difficult to devise an explanation for the differences in
12 Rafique M, Farrukh MS, Shaikh FS. Assessment of hearing loss in
results for both studies. tympanic membrane perforation at tertiary care hospitals.
The outcome of this study will provide a better under- JLUMHS 2014;13(01):32–36
standing of hearing loss attributed to tympanic membrane 13 Park H, Hong SN, Kim HS, et al. Determinants of conductive
perforation. This can help specialists manage such cases hearing loss in tympanic membrane perforation. Clin Exp Otor-
hinolaryngol 2015;8(02):92–96
more efficiently, given that hearing loss represents a major
14 Ahmad SW, Ramani GV. Hearing loss in perforations of the
health problem and this is the time to educate people on its tympanic membrane. J Laryngol Otol 1979;93(11):1091–1098
severe consequences. 15 Nepal A, Bhandary S, Mishra SC, Singh I, Kumar P. The morphology
of central tympanic membrane perforations. Nepal Med Coll J
2007;9(04):239–244
Conclusions 16 Ibekwe TS, Ijaduola GT, Nwaorgu OG. Tympanic membrane
perforation among adults in West Africa. Otol Neurotol 2007;
The level of the hearing loss was proportionately related with
28(03):348–352
the size of the tympanic membrane perforation. Moreover, 17 McArdle FE, Tonndorf J. Perforations of the tympanic membrane
the posterior sited perforation had greater impact on the and their effects upon middle-ear transmission. Arch Klin Exp
threshold of the hearing than those with anterior sited Ohren Nasen Kehlkopfheilkd 1968;192(02):145–162

International Archives of Otorhinolaryngology Vol. 21 No. 4/2017

342 Frequency Dependence Hearing Loss Evaluation in Perforated Tympanic Membrane Dawood

18 Bigelow DC, Swanson PB, Saunders JC. The effect of tympanic 21 Lerut B, Pfammatter A, Moons J, Linder T. Functional correlations
membrane perforation size on umbo velocity in the rat. Laryngo- of tympanic membrane perforation size. Otol Neurotol 2012;
scope 1996;106(1 Pt 1):71–76 33(03):379–386
19 Bhusal CL, Guragain RP, Shrivastav RP. Frequency dependence of 22 Röösli C, Sim JH, Chatzimichalis M, Huber AM. How does closure of
hearing loss with perforations. JNMA J Nepal Med Assoc 2007; tympanic membrane perforations affect hearing and middle ear
46(168):180–184 mechanics? An evaluation in a patient cohort and temporal bone
20 Al-Omari AF, Al-Doski FS. Effect of the size of tympanic mem- models. Otol Neurotol 2012;33(03):371–378
brane perforation on hearing. Tikrit J Pharma Sci. 2013;9(01): 23 Austin DF. Sound conduction of the diseased ear. J Laryngol Otol
154–161 1978;92(05):367–393

International Archives of Otorhinolaryngology Vol. 21 No. 4/2017