Silver anniverSary iSSue vol. 21 noS. 1 & 2 January –June; July – december 2006

ORIGINAL ARTICLES

PhiliPPine Journal of otolaryngology-head and neck Surgery 5

ABSTRACT
Background: Pure tone audiometry is routinely used to determine conductive and sensorineural
hearing status. Ossicular discontinuity is usually assessed intra-operatively. If ossicular
discontinuity can be predicted by pure tone audiometry, perhaps the operative procedure of
choice and prognosis for hearing can also be anticipated.

Objective: To determine the predictive value of preoperative pure tone audiometry on the
presence of gross ossicular discontinuity in chronic otitis media.

Methods: Records of 205 patients, 7 to 75 years of age undergoing their first operation for
chronic otitis media were reviewed. Preoperative audiograms and operative records for
tympanomastoidectomy were evaluated. A total of 162 patients meeting inclusion criteria were
included in the study.

Likelihood ratios for positive and negative ossicular discontinuity for frequency-specific
air-bone gap cut-offs were determined. Multiple logistic regression analysis for pure tone
audiometry and operative findings to predict ossicular discontinuity was performed and a model
for predicting ossicular discontinuity using logistic regression obtained.

Results and Conclusion: Frequency-specific air bone gap (ABG) cut-off values can predict
ossicular discontinuity in chronic suppurative otitis media namely: < 20 dB ABG at 500 Hz predicts
absence of ossicular discontinuity while > 50 dB ABG at 500 Hz, >30 dB ABG at 2 KHz, and > 50 dB
ABG at 4 KHz best predict the presence of ossicular discontinuity in general.

In the absence of cholesteatoma, the air bone gaps of <30 dB at 500 Hz and <20 dB at 1 KHz
decrease probability of ossicular discontinuity from 32.97% to 2.54%. Combination of air bone
gaps of >50 dB at 500 Hz, >20 dB at 2 KHz and >40 dB at 4 KHz increase the probability of ossicular
discontinuity from 32.97% to 85.9%. These findings suggest that ossicular exploration may not
be necessary for the former while an evaluation of the ossicular chain may be mandatory for the
latter in the setting where cholesteatoma is not present or suspected.

Presence of cholesteatoma, granulation tissue and size of tympanic membrane perforation
are important factors to consider in predicting ossicular discontinuity.

Keywords: air bone gap; audiometry, pure tone; ossicular discontinuity; otitis media, suppurative;
logistic regression; likelihood ratio; predictive value.

CHRONIC otitis media is a common ear pathology operated on and a common cause of
hearing impairment in the local setting. Chronic infection may lead to middle ear structure
changes including ossicular discontinuity. Surgical intervention usually aims to eradicate disease
and restore hearing.

Restoration of hearing may be attained in well-planned ear surgery. Hearing tests (Pure
Tone Audiometry, Speech Testing) and imaging studies (x-rays, high resolution computed
tomography scanning) are used to predict serviceable hearing reserve and anatomical integrity,

Relationship of Pure Tone Audiometry and
Ossicular Discontinuity

in Chronic Suppurative Otitis Media

Ryner Jose C. Carrillo, MD1,
Nathaniel W. Yang, MD2,3,
Generoso T. Abes, MD, MPH 2,3

1Department of Otorhinolaryngology
Philippine General Hospital
University of the Philippines Manila

2Department of Otorhinolaryngology
College of Medicine – Philippine General Hospital
University of the Philippines Manila

3Philippine National Ear Institute
National Institutes of Health
University of the Philippines Manila

Correspondence: Ryner Jose C. Carillo, MD
Department of Anatomy, College of Medicine
University of the Philippines Manila
Pedro Gil St. Ermita, Manila 1000
Phone: (632) 526 4194

sENTro Head and Neck Medicine and Surgery
414 West East Center, 1336 Taft Ave., Ermita, Manila 1000
Phone: 632) 524 4455
E-mail: ryner_c@yahoo.com
Reprints will not be available from the author.

No funding support was received for this study. The authors
signed a disclosure that they have no proprietary or financial
interest with any organization that may have a direct interest
in the subject matter of this manuscript, or in any product
used or cited in this study.

Presented at:

1. The 25th Politzer Society Meeting. Seoul, Korea, October
2005 (Poster).

2. Resident’s Research Forum, Reserch Information and
Dissemination Office, Philippine General Hospital,
University of the Philippines Manila November 2005.

3. Analytical Research Contest (2nd Place), Philippine
Society of Otolaryngology Head and Neck Surgery
49th Annual Convention, Westin Philippine Plaza Hotel,
Manila, December 1, 2005. Philipp J Otolaryngol Head Neck Surg 2006; 21 (1,2): 5-10 c Philippine Society of Otolaryngology – Head and Neck Surgery, Inc.

Silver anniverSary iSSue vol. 21 noS. 1 & 2 January –June; July – december 2006
ORIGINAL ARTICLES

6 PhiliPPine Journal of otolaryngology-head and neck Surgery

respectively.
In our setting, pure tone audiometry is a routine preoperative

procedure for tympanomastoidectomy while computed tomography
is expensive and less widely used. Further, ossicular integrity can be
assessed by high-resolution tomography, but can only be confirmed
intra-operatively1-5. Lacking prior information, patient prognosis and
surgical management techniques can only be determined during
surgery.

A previous study of the possible predictive value of audiograms
on ossicular status showed that the preoperative hearing tests did not
correlate with ossicular discontinuity6. However, studies of ossicular
dysfunction must be frequency-specific since the ossicular chain and
tympanic membrane have different efficiencies across frequencies7,8.

By far, pure tone audiometry is the only routine test to quantitatively
assess the hearing capacity of patients with otitis media undergoing
surgery9. Should this be able to predict ossicular chain status, the
conduct and the prognosis of ear surgery may be determined.

Hypothetical air bone gap distributions can be plotted as shown in
figure 1. A normal air bone gap will lie between -10 to 15 dB. Varying
degrees of ossicular chain dysfunction, i.e., ossicular fixation, functional
ossicular discontinuity and gross ossicular discontinuity may be
represented by varying degrees of conductive hearing loss represented
by the air bone gap.

Being able to determine minimum and maximum cut-off levels that
may represent absence or presence of particular ossicular defects will
allow a routine audiogram to predict specific defects in the hearing
mechanism.

Like the electrocardiogram predicting particular heart defects
through “lead-specific tracings”, a frequency-specific audiogram may
perhaps point to particular hearing deficits.

Correlations between pure tone audiometry and ossicular
discontinuity can be modified by the type and degree of tympanic
perforation, middle ear polyps, degree and bilaterality of hearing loss.
Granulation tissue and cholesteatoma may be independent predictors
of ossicular discontinuity but may alter results of hearing by direct

Frequency Percent

Gross ossicular discontinuity 92 / 162 56.8%

Cholesteatoma 91 / 161 56.2%

Granulation Tissue 85 / 162 52.5%

Tympanosclerosis 22 / 162 13.6 %

Attic Perforation 21 / 162 13.4%

Subperiosteal Abscess 2 / 161 1.2 %

Polyp in the middle ear 9 / 162 5.6%

Table 1. Operative Findings

Frequency Mean Standard
Air bone gap Deviation

500Hz 41.17 16.81

1KHz 35.77 17.34

2KHz 20.56 13.76

4KHz 34.17 15.30

Table 2. Average of frequency specific air bone gaps

Table 3. Likelihood Ratios (LR) and changes in Pretest to Post-test
Probabilities*

Likelihood Ratio Likelihood Ratio Generated Changes in
for a Positive test for a Negative test Pretest to Post-test
Probabilities

> 10 < 0.1 Large and conclusive changes

5-10 0.1-0.2 Moderate shifts

2-5 0.2-0.5 Small but sometimes important
shifts

1-2 0.5-1 Rarely important shifts

*Evidence Based Medicine Working Group. The Users’ Guides to Evidence-based Medicine11

Table 4. Air bone gap (ABG) cut off values at 500 Hz and corresponding
likelihood ratios (LR) for absence or presence of gross ossicular discontinuity
(OD).

ABG OD OD LR LR P
500 Hz (dB) (+) (-) (+) (-) value

0.254

0.435

0.516

0.686

0.761

0.965

0.992

1.070

1.141

1.322

1.443

2.283

1.775

1.522

0.068

0.033

0.006

0.005

0.003

0.299

0.602

0-10

>10-20

>20-30

>30-40

>40-50

>50-60

>60-70

>70

Figure 1. Theoretical Air Bone Gap Distribution

Silver anniverSary iSSue vol. 21 noS. 1 & 2 January –June; July – december 2006

ORIGINAL ARTICLES

PhiliPPine Journal of otolaryngology-head and neck Surgery 7

conduction of sound. Inability to isolate a particular ear for testing may
alter the correlative process. Hence, patients with masking dilemma,
canal obstruction, head trauma, intracranial complications of otitis
media were excluded from the study.

The air-bone gap degree and pattern should be able to predict
an ossicular chain dysfunction. Such dysfunction can be in the form
of ossicular chain fixation or discontinuity. Ossicular discontinuity
may be further classified as gross or functional discontinuity. Gross
discontinuity is the absence of specific ossicular bones or a disruption
in the ossicular joints. Functional discontinuity is the absence of
unified ossicular movement on manipulation despite the presence of
gross ossicular chain continuity due to fibrosis or granulation tissue.
Functional discontinuity and ossicular fixation were not analyzed in this
study.

This study aims to determine the ability of pure tone audiometry in
predicting ossicular discontinuity in chronic suppurative otitis media.

MATERIALS AND METHODS
Records of 205 patients, 7 to 75 years of age whose first

tympanomastoidectomy was performed by a single surgeon (GTA) at
the Philippine General Hospital or Manila Doctors Hospital from June
1987 to July 2004 were reviewed. Preoperative audiograms within 12
months of surgery and operative records were analyzed.

Minimum mastoidectomy had to include an atticotomy with
visualization of the incudo-stapedial joint. Patients with masking
dilemma, obstructed canal lumen less than 3mm, congenital defects or
fractures of the temporal bone and ears with otitis media complications
were excluded. All patients were evaluated intraoperatively by the
same surgeon through direct ossicular chain visualization with an
operative microscope, and by palpation of the ossicular chain. Incudo-
stapedial joint continuity and pathology (granulation or fibrosis) over
the ossicular chain were visualized. The malleus handle and incudo-
stapedial joint were palpated further to assess chain continuity. Gross
ossicular discontinuity was defined as visualization of a disconnected or
absent part of the ossicular chain. Other forms of dysfunction (ossicular
fixation and functional dysfunction) were not included in this study.

Other findings including the size and location of tympanic
perforation, subperiosteal abscess, tympanosclerosis, cholesteatoma
and granulation tissue were tabulated.

Frequency-specific air bone gap cut off values were associated
with presence or absence of ossicular discontinuity with significance
determined using likelihood ratios (LR) and chi square test/Fisher’s
exact test.

Univariate analysis of independent factors and possible modifiers
predicting ossicular discontinuity was performed. The full model
variables (ossicular discontinuity (OD) associated with cholesteatoma,
granulation tissue, foul smelling discharge, aural polyp, size and type
of perforation, subperiosteal abscess, tympanosclerosis, bilaterality
and type of hearing loss, and frequency-specific air bone gap levels)
were subjected to the stepwise backward method for model estimation
(Wald Test) and Likelihood Ratio Tests in order to arrive at a parsimonious

Table 6. Air bone gap (ABG) cut off values at 2 KHz and corresponding likelihood
ratios (LR) for absence or presence of gross ossicular discontinuity (OD).

ABG OD OD LR LR P
2K (dB) (+) (-) (+) (-) value

0.733

0.730

0.835

0.918

0.980

0.991

cnt

1.168

1.674

2.283

3.804

3.817

2.290

cnt

0.112

0.010

0.016

0.048

0.321

0.568

cnt

0-10

>10-20

>20-30

>30-40

>40-50

>50-60

>60-70

>70

Table 5. Air bone gap (ABG) cut off values at 1 KHz and corresponding likelihood
ratios (LR) for absence or presence of gross ossicular discontinuity (OD).

ABG OD OD LR LR P
1 KHz (dB) (+) (-) (+) (-) value

0.484

0.685

0.566

0.791

0.884

0.940

0.991

1.096

1.126

1.546

1.562

1.902

3.043

2.290

0.085

0.125

0.002

0.032

0.064

0.113

0.568

0-10

>10-20

>20-30

>30-40

>40-50

>50-60

>60-70

>70

Table 7. Air bone gap (ABG) cut off values at 4 KHz and corresponding likelihood
ratios (LR) for absence or presence of gross ossicular discontinuity (OD).

ABG OD OD LR LR P
4KHz (dB) (+) (-) (+) (-) value

1.014

0.592

0.666

0.774

0.890

0.937

cnt

0.999

1.141

1.446

2.092

2.435

9.925

cnt

0.604

0.072

0.012

0.006

0.043

0.031

cnt

0-10

>10-20

>20-30

>30-40

>40-50

>50-60

>60-70

>70

Table 8. Gross ossicular discontinuity (OD) and cholesteatoma

OD OD
(+) (-)

With Cholesteatoma

No Cholesteatoma

Odds ratio of 15.75, (p< 0.0001)

Silver anniverSary iSSue vol. 21 noS. 1 & 2 January –June; July – december 2006

ORIGINAL ARTICLES

8 PhiliPPine Journal of otolaryngology-head and neck Surgery

56.8% OD

75% OD
>75% OD

36.4% OD

500 Hz 1 KHz 2 KHz 4 KHz

Frequency

A
B
G

500 Hz 1 KHz 2 KHz 4 KHz

Frequency

A
B
G

32.97% OD

<14% OD

56% OD >51% OD

<19% OD

Table 9. No Cholesteatoma Group: Air bone gap (ABG) cut off values at 500
Hz and corresponding likelihood ratios (LR) for absence or presence of gross
ossicular discontinuity (OD).

ABG OD OD LR LR P
500 Hz (dB) (+) (-) (+) (-) value

0.407

0.156

0.313

0.581

0.651

0.982

0.983

1.053

1.228

1.510

1.564

2.588

1.356

2.033

0.351

0.021

0.004

0.026

0.005

0.535

0.553

0-10

>10-20

>20-30

>30-40

>40-50

>50-60

>60-70

>70

Table 10. No Cholesteatoma Group. Air bone gap (ABG) cut off values at 1
KHz and corresponding likelihood ratios (LR) for absence or presence of gross
ossicular discontinuity (OD).

ABG OD OD LR LR P
1 KHz (dB) (+) (-) (+) (-) value

0.203

0.339

0.395

0.723

0.806

0.931

1.156

1.277

1.871

1.779

2.287

3.050

0.066

0.031

0.001

0.044

0.051

0.199

0-10

>10-20

>20-30

>30-40

>40-50

>50-60

>60-70

>70

model.
Data were encoded using Epi Info 6.04d (Centers for Disease Control,

USA and World Health Organization), transformed to STATA file using
Stat transfer version 5 (Circle Systems, Inc, USA) and analyzed using
Intercooled Stata version 6 (Stata Corporation, USA) and EBMCAL v1.2
(Evidence Based Medicine Calculator).

RESULTS
Of the 162 patients included in the study, 92 (56.8%) had gross

ossicular discontinuity, 91 (56.2%) and 85 (52.5%) had cholesteatoma
and granulation tissue respectively. Other operative findings are
summarized in Table 1.

Audiometry revealed 112 (69%) had bilateral hearing loss while 115
(71%) had mixed hearing loss. Averages of frequency-specific air bone
gaps were narrowest at 2 KHz and widest at 500Hz. (Table 2)

Frequency-specific cut-off values were analyzed for the presence
or absence of gross ossicular discontinuity. Corresponding p values
were taken using either chi square or Fisher’s exact test, and respective
likelihood ratios for positive or negative results were obtained.
Significant likelihood ratios (LR) were defined as small, moderate or
large results. (Table 3)

For an air bone gap (ABG) <20 dB at 500 Hz the probability of
ossicular discontinuity of 56.8% decreased to 36.4%, predicting absence
of gross ossicular discontinuity (OD). For an ABG > 50 dB at 500 Hz there
were increased chances from 56.8% to 75% of predicting the presence
of gross ossicular discontinuity. (Table 4)

Air bone gaps >30 dB and > 40 dB generated statistically significant
but small effects on the chance of having ossicular chain discontinuity.
(Table 5)

Air bone gaps > 30 dB or >40 dB at 2 KHz shifted the baseline
probability of OD from 56.8% to 75% and 83.34%, respectively. (Table
6)

Air bone gaps > 50 dB at 4 KHz increased the chance of OD from
56.8% to 76.2%. Increasing air bone gap cut-off values showed
progressive increase in likelihood ratios for a positive test. (Table 7)

Combinations of air bone gaps of > 50 dB at 500 Hz, >30 dB at 2 KHz
and >50 dB at 4 KHz increased the probability of ossicular discontinuity
from a baseline probability of 56.8% to 94.35%.

There was no significant predictive value for air bone gap levels on
the status of ossicular chain continuity in the cholesteatoma group.
Cholesteatoma was associated with ossicular discontinuity with an
odds ratio of 15.75 (p<0.0001). (Table 8)

Among subjects without cholesteatoma, significant cut-off values

Figure 2. Air bone gap and
ossicular discontinuity,
with or without
cholesteatoma

Figure 3. No Cholesteatoma
Group: Air bone gap and
ossicular discontinuity

Silver anniverSary iSSue vol. 21 noS. 1 & 2 January –June; July – december 2006

ORIGINAL ARTICLES

PhiliPPine Journal of otolaryngology-head and neck Surgery 9

were obtained. Air bone gaps <30 dB at 500 Hz decreased the probability
of ossicular discontinuity from 32.97% to 7.67% while air bone gaps >
50dB at 500 Hz increased the probability of ossicular discontinuity from
32.97% to 56%. (Table 10)

In the absence of cholesteatoma, ABG < 20 dB at 1 KHz decreased
probability of ossicular discontinuity from 32.97% to 14.29%. (Table
10)

In the no cholesteatoma group, ABG >20 dB at 2 KHz increased the
probability of ossicular discontinuity from 32.97% to 51.4%. (Table 11)

In absence of cholesteatoma, ABG > 40 dB at 4 KHz increased the
probability of gross ossicular discontinuity from 32.97% to 52%. (Table
12)

In the no cholesteatoma group, a combination of air bone gaps of
<30 dB at 500 Hz and <20 dB at 1 KHz decreased probability of ossicular
discontinuity from 32.97% to 2.54%. Combination of air bone gaps
of >50 dB at 500 Hz, >20 dB at 2 KHz and >40 dB at 4 KHz increased
probability from 32.97% to 85.9%.

Air bone gap cut off levels were plotted with frequency and
probability of ossicular discontinuity. (Figure 2 and Figure 3). The
baseline chance of ossicular discontinuity in general was 57%. In the
absence of cholesteatoma, the chance of ossicular discontinuity was
33%.

Multiple logistic regression analysis was done using gross ossicular
discontinuity as dependent variable. The most parsimonious model
consisted of cholesteatoma, granulation, ABG at 1 KHz and size of
perforation as the most important independent variables. The simplest
model may be used for predicting ossicular discontinuity and the
baseline chance of ossicular chain discontinuity was only applicable in
this population. (Table 13)

1
Probability of OD = --------------------------------------------------------------------

1 + e - (-4.386 + 1.66(Cholesteatoma) + 2.07(Granulation) +0.032(ABG at 1KHz) +0.0278(Size of Perforation))

Cholesteatoma (0-absent, 1-present)
Granulation (0-absent, 1-present)
ABG at 1 KHz (continuous variable, 0-60)
Size of Perforation (continuous variable, 0-100)

DISCUSSION
Frequency-specific air-bone gap (ABG) cut-off values can predict

ossicular discontinuity in chronic suppurative otitis media. Generally,
ABG < 20 dB at 500 Hz predict absence of ossicular discontinuity while
ABG > 50 dB at 500 Hz, >30 dB at 2 KHz, and > 50 dB at 4 KHz best
predict the presence of ossicular discontinuity in general.

Pre-operative audiograms best detect the absence or presence
of ossicular chain discontinuity in the absence of cholesteatoma.

Table 11. No Cholesteatoma Group. Air bone gap (ABG) cut off values at 2
KHz and corresponding likelihood ratios (LR) for absence or presence of gross
ossicular discontinuity (OD).

ABG OD OD LR LR P
2KHz (dB) (+) (-) (+) (-) value

0.763

0.555

0.850

0.965

0.959

1.154

2.153

2.372

2.033

6.00

0.263

0.003

0.082

0.401

0.330

0-10

>10-20

>20-30

>30-40

>40-50

>50-60

>60-70

>70

Table 13. Final Model for Patients showing important variables after logistic
regression analysis:

Coefficient P
value

Cholesteatoma

Granulation

Air bone gap 1 KHz

Size of perforation

_constant

1.66

2.07

.032

.028

-4.39

0.016

0.000

0.037

0.048

0.001

Table 12. No Cholesteatoma Group. Air bone gap (ABG) cut off values at 4
KHz and corresponding likelihood ratios (LR) for absence or presence of gross
ossicular discontinuity (OD).

ABG OD OD LR LR P
4KHz (dB) (+) (-) (+) (-) value

0.813

0.271

0.478

0.732

0.944

1.017

1.238

1.657

2.218

1.627

0.579

0.032

0.008

0.024

0.335

0-10

>10-20

>20-30

>30-40

>40-50

>50-60

>60-70

>70

Silver anniverSary iSSue vol. 21 noS. 1 & 2 January –June; July – december 2006

ORIGINAL ARTICLES

10 PhiliPPine Journal of otolaryngology-head and neck Surgery

REFERENCES:
1. Banerjee A,et al. Computed tomography in suppurative ear disease: does it influence

management? J Laryngol Otol. 2003 Jun;117(6):454-8.
2. Fuse T, et al. Diagnosis of ossicular chain in the middle ear by high-resolution CT.

Nippon Jibiinkoka Gakkai Kaiho. 1992 Feb;95(2):247-52.
3. Jackler, RK, et al. Computed tomography in suppurative ear disease: a correlation of surgical and

radiographic findings. Laryngoscope. 1984 Jun;94(6):746-52.
4. Leighton SE, et al. The role of CT imaging in the management of chronic suppurative otitis media.

Clin Otolaryngol. 1993 Feb;18(1):23-9.
5. Swartz, J, et al. Ossicular erosions in the dry ear: CT diagnosis. Radiology. 1987 Jun;163(3):763-5.
6. Jeng Fuh-Cheng, et al. Relationship of Preoperative Findings and Ossicular Discontinuity in

Chronic Otitis Media. Otology and Neurotology. Vol. 24, No. 3, 2003.
7. Cummings, CW, et al. Otolaryngology Head and Neck Surgery. 3rd Edition. 1998.
8. Ruckenstien, M. Comprehensive Review of Otolaryngology. 2004
9. PSO-HNS Clinical Practice Guideline on Chronic Suppurative Otitis Media, 1986.
10. Karja J, et al. Destruction of ossicles in chronic otitis media. J Laryngol Otol. 1976 Jun;90(6):509-

18.
11. Roman Jaeschke, Gordon H. Guyatt, David L. Sackett, and the Evidence Based Medicine Working

Group. The Users’ Guides to Evidence-based Medicine and reproduced with permission from
JAMA. (1994;271(5):389-391) and (1994;271(9):703-707). Copyright 1995, American Medical
Association.

12. McGrew, BM., et al. Impact of mastoidectomy on simple tympanic membrane perforation repair.
The Laryngoscope. 2004; 114: 506-511.

13. Sakagami, M. et al. Cholesteatoma otitis media with intact ossicular chain. Auris Nasus Larynx.
1999: (26) 147-151.

Likelihood-ratio analysis showed cut-off values in air-bone gaps at
500 Hz, 2 KHz and 4 KHz are more reliable determinants of ossicular
status. In the absence of cholesteatoma, air-bone gaps of <30 dB
at 500 Hz and <20 dB at 1 KHz decrease the probability of ossicular
discontinuity from 32.97% to 2.54%. Air-bone gaps of >50 dB at 500
Hz, >20 dB at 2 KHz and >40 dB at 4 KHz increase the probability of
ossicular discontinuity from 32.97% to 85.9%. These findings suggest
that ossicular exploration may not be necessary for the former while an
evaluation of the ossicular chain may be mandatory for the latter in the
setting where cholesteatoma is not present or suspected.

Patterns suggesting an intact ossicular chain dictate a more
conservative procedure, e.g., Wullstein type I tympanoplasty while
air-bone gap levels that suggest discontinuity warrant thorough
inspection and possible reconstruction of the ossicular chain. Routine
mastoidectomy for simple tympanic perforations may be avoided,
although it may have its advantages12.

The presence of cholesteatoma, granulation tissue and size of
tympanic membrane perforation are important factors to consider
in predicting ossicular discontinuity. The prevalence of ossicular
discontinuity and cholesteatoma in this population was 56.8% and
58.2%, respectively. Cholesteatoma presence is an important risk factor
for ossicular chain discontinuity (odds ratio of 15.75) and may warrant

ossicular exploration with even simple or small tympanic perforations.
The poor correlation of audiograms in this study with ossicular chain
discontinuity in the presence of cholesteatoma may be due to the
latter’s ability to transmit sound. Even with an intact ossicular chain,
subjects with cholesteatoma merit mandatory ossicular inspection
by atticotomy and/or mastoidectomy with posterior tympanotomy,
followed by ossiculoplasty as needed. For such cases, the advantage of
Wullstein type III tympanoplasty over type I has been cited in previous
studies13.

Although ossicular evaluation is routinely performed
intraoperatively, prior knowledge of ossicular chain status gleaned
from audiograms may influence the planned surgical technique
(simple tympanoplasty, atticotomy with attic wall reconstruction,
mastoidectomy with posterior tympanotomy and an ossiculoplasty)
and implications of treatment, complementing computed tomography
(or, as in our setting, replacing it).

Averaging results across frequencies may dilute the predictive value
of audiometry on ossicular chain dysfunction. Frequency-specific air-
bone gap cut-off values can predict gross ossicular discontinuity and
the true magnitude of this relationship can be better established in a
prospective study. Moreover, varying degrees of ossicular dysfunction
may be investigated.