International Journal of Pediatric Otorhinolaryngology 73 (2009) 1751–1756

The impact of chronic suppurative otitis media on children’s and teenagers’hearing

Luciana Fick Silveira Netto *, Sady Selaimen da Costa, Pricila Sleifer, Maria Elisa Luce Braga

Ear, Nose and Throat Service, Chronic Otitis Media Ambulatory Care Center, Hospital de Clınicas de Porto Alegre, Brazil

A R T I C L E I N F O

Article history:

Received 28 April 2009

Received in revised form 7 September 2009

Accepted 15 September 2009

Available online 23 October 2009

Keywords:

Chronic suppurative otitis media

Air-bone gap

Hearing thresholds

Pure tone audiometry

Children

Cholesteatoma

A B S T R A C T

Objective: Otitis media is the most common otological condition during childhood which compromises

sound conduction in the middle ear. In chronic cases, it is estimated that the degree to which hearing is

compromised is directly proportional to the damage caused to the middle ear’s structures. It means that

hearing thresholds may be influenced by factors such as the size and location of the tympanic

perforation, the presence of ossicular chain erosion or disarticulation as well as the presence of

cholesteatoma and its growth patterns. The goals of this study were to compare air conduction, bone

conduction thresholds and air-bone gaps of children and teenagers between those with chronic

suppurative otitis media with cholesteatoma and those without cholesteatoma. To compare air-bone

gap values for different cholesteatoma growth patterns. To verify the relationship between the number

of perforated quadrants and the size of the air-bone gap. To compare air-bone gap values between

tympanic perforations in posterior quadrants with those in anterior quadrants.

Methods: A transversal study involving 202 children and teenagers (287 ears), aged between 6 and 18,

with chronic suppurative otitis media with and without cholesteatoma, submitted to digital

videotoscopy and pure tone audiometry (PTA) was conducted.

Results and conclusions: Air conduction, bone conduction thresholds and air-bone gaps in children and

teenagers with CCOM are significantly greater. There were no significative differences between air-bone

gaps in epitympanic and posterior mesotympanic cholesteatomas. In NCCOM, the gap value is positively

correlated with the number of quadrants with tympanic perforation. There was no significative

difference between the air-bone gaps in tympanic perforations affecting the posterior and anterior

quadrants.

� 2009 Elsevier Ireland Ltd. All rights reserved.

Contents lists available at ScienceDirect

International Journal of Pediatric Otorhinolaryngology

journa l homepage: www.e lsev ier .com/ locate / i jpor l

1. Introduction

Otitis media is the most common otological condition duringchildhood which compromises sound conduction in the middle ear[1]. In chronic cases, hearing loss is often a permanent consequence,due to irreversible tissue changes in the auditory cleft.

It is estimated that the degree to which hearing is compromisedis directly proportional to the damage caused to the structures ofthe middle ear. The hearing thresholds may be influenced byfactors such as the size and location of the tympanic perforation,the presence of ossicular chain erosion or disarticulation as well asthe presence of cholesteatoma and its growth patterns [2,3].

The goals of this study were as follows: (1) to compare airconduction thresholds (AC), bone conduction thresholds (BC) andair-bone gaps in children and teenagers with chronic suppurative

* Corresponding author at: Rua Tomas Gonzaga, 430 casa 11, Bairro Boa Vista,

CEP: 91340-480, Porto Alegre, Brazil. Tel.: +55 51 99144660; fax: +55 51 21018164.

E-mail address: lunetto@portoweb.com.br (L.F. Silveira Netto).

0165-5876/$ – see front matter � 2009 Elsevier Ireland Ltd. All rights reserved.

doi:10.1016/j.ijporl.2009.09.033

otitis media with cholesteatoma (CCOM) and those withoutcholesteatoma (NCCOM); (2) to compare air-bone gaps betweendifferent cholesteatoma growth patterns; (3) to verify the relationbetween the numberofperforated quadrantsand the value ofthe air-bone gap; (4) to compare air-bone gaps between posterior quadranttympanic perforations and anterior quadrant tympanic perforations.

2. Patients and methods

2.1. Selected patients

We developed a prospective cross-sectional study, including allpatients with suppurative chronic otitis media diagnosis followedat Chronic Otitis Media Ambulatory Care Center of Hospital deClınicas de Porto Alegre (HCPA) during the period of November2003 and March 2005.

Patients with previous otological surgery, external auditorycanal alterations, congenital cholesteatomas or those withinconsistent pure tone audiometry (PTA) results were excludedfrom the sample.

Table 1Comparison of AC thresholds considering clinical diagnosis of CSOM (interquartile

range and median).

Frequency (Hz) NCCOM (n = 206) CCOM (n = 81) p-Value

Air conduction median (dBHL)

250 35 (25–45) 40 (30–55) 0.010

500 30 (20–40) 40 (25–55) <0.001

1000 25 (15–35) 40 (30–50) <0.001

2000 25 (15–30) 35 (20–45) <0.001

3000 25 (15–35) 35 (25–45) <0.001

4000 25 (20–40) 35 (25–45) <0.001

6000 30 (20–40) 40 (35–50) <0.001

8000 30 (20–40) 35 (24–45) <0.001

Mann–Whitney test. NCCOM = chronic otitis media without cholesteatoma;

CCOM = chronic otitis media with cholestetatoma; dBHL = decibel hearing level;

Hz = Hertz.

Table 2Comparison of BC thresholds considering clinical diagnosis of CSOM (interquartile

range and median).

Frequency (Hz) NCCOM (n = 206) CCOM (n = 81) p-Value

Bone conduction median (dBHL)

500 5 (0–10) 5 (0–10) 0.965

1000 5 (0–10) 5 (0–10) 0.243

2000 5 (0–10) 10 (5–15) 0.003

3000 10 (5–15) 10 (10–20) 0.001

4000 5 (0–15) 10 (5–15) <0.001

Mann–Whitney test. NCCOM = chronic otitis media without cholesteatoma;

CCOM = chronic otitis media with cholestetatoma; dBHL = decibel hearing level;

Hz = Hertz.

L.F. Silveira Netto et al. / International Journal of Pediatric Otorhinolaryngology 73 (2009) 1751–17561752

2.2. Data collection

Age, gender, race, otological history, symptom duration,familiar history of otologycal disease and affected ear werecollected for each selected patients, who were then submitted toan ENT examination, digital videotoscopy and PTA.

Examined ears were divided in to two groups: chronic otitismedia without cholesteatoma (group 1); and chronic otitis mediawith cholesteatoma (group 2).

Digital videotoscopy was performed with 08 optics of 3.6 mmdiameter connected to a microcamara. Images were captured inreal time by a Video Digital Recorder, using a Matrox PC-VCR boardwith a video resolution of 1024 � 768. The images were laterexamined by the same otolaryngologist for: (1) presence oftympanic membrane (TM) retraction or perforation; (2) quadrantsaffected by TM perforation (posterior/anterior); (3) number ofquadrants perforated (1–4); (4) presence of cholesteatoma; (5)cholesteatoma growth patterns (mesotympanic or epitympanic).

We used the ‘‘number of perforated quadrants’’ variable toachieve an estimate of tympanic perforation size, for lack of abetter instrument for assessing tympanic perforation.

Cholesteatoma growth patterns were labeled as posteriormesotympanic and epitympanic based on their likely locationsof origin, as suggested by videotoscopy. Cholesteatomas weredefined as epitympanic when originating in the pars flaccida of theTM (atical region); and as posterior mesotympanic whenoriginating in the posterior-superior quadrant of the pars tensa.Posterior and anterior epitympanic cholesteatomas were notdiscriminated due to the latter’s rarity and the similar pattern ofossicular damage.

All patients were submitted to PTA, determining AC thresholdsat 250, 500, 1000, 2000, 3000, 4000, 6000 and 8000 Hz, BCthresholds and air-bone gaps at 500, 1000, 2000, 3000 and 4000 Hzfrequencies. The air-bone gap was calculated using the differencebetween AC and BC thresholds. Interacoustic AD27 audiometerwith supraural TDH-39 earphones was used. For BC thresholds, abone-conduction transmitter placed on the mastoid bone wasused. Even for AC and BC, Narrow Band masking noise was appliedwhen needed.

In young children, which could arise problems to measurereliable thresholds, playful conditioning (play-conditioned audio-metry with supraural earphone) was performed. When necessary,the PTA could be concluded after two sessions in order to confirmits results.

This study was approved by the hospital’s Research EthicsCommittee. Parents or guardians of all patients signed a Free and

Informed Consent form prior to inclusion in the study.

2.3. Statistical analysis

The Statistical Package for Social Science (SPSS) for Windowsversion 14.0 was used for statistical analysis.

AC, BC thresholds and air-bone gaps were compared using theMann–Whitney nonparametric test. Qualitative variables werecompared in cross-tables using the chi-square test and statisticallysignificant correlation was found by calculating adjusted remain-ders. The relationship between the number of perforated quadrantsand air-bone gaps was assessed by the Spearman nonparametriccorrelation. A significance level of 5% (p � 0.05) was used.

3. Results

Two hundred and two children and teenagers aged 6–18(11 � 3.8 years old) were selected and examined. Eighty-five patientshad bilateral pathology, totaling 287 examined ears. Males con-stituted 51.6% of the sample.

Two hundred and six of the examined ears (71.8%) had NCCOM,while 81 ears (28.2%) had CCOM. In the group with cholesteatoma(group 2), 57 ears (70.5%) had mesotympanic cholesteatoma, while24 (29.5%) had epitympanic cholesteatoma.

Analysis of characteristics of included patients showed homo-geneneity between the groups.

3.1. Analysis of AC and BC thresholds

The analysis of AC medians is shown in Table 1. The comparisonbetween the groups showed significantly higher AC values in allexamined frequencies in those patients with cholesteatoma. The250 and 500 Hz frequencies were hardest affected in both groups.The difference between AC medians in both groups stand around10 dBHL.

The BC medians are seen in Table 2. The thresholds showedsignificantly higher in 2000, 3000 and 4000 Hz also in group 2.

3.2. Analysis of air-bone gaps

Group 1 showed smaller air-bone gap values in all examinedfrequencies. Both groups displayed higher air-bone gap values at500 Hz. Both groups showed a trend towards lower gap values astested frequencies were increased (Fig. 1) (Table 3).

Stratifying group 2 patients according to cholesteatoma growthpatterns, (mesotympanic or epitympanic), no significative differ-ences in air-bone gap values were observed as we can see inTable 4.

Patients of group 1 with TM perforation were selected andstratified according their number of perforated quadrants.Fig. 2(a)–(e) shows a significant positive correlation in the NCCOMgroup between the number of perforated quadrants of TM and air-bone gap values for all tested frequencies.

Group 1 patients who presented perforation of the pars tensa ofthe TM (n = 206) were stratified according to location of theperforation. Cases in which tympanic perforations were located

Fig. 1. Comparison of air-bone gaps beetwen NCCOM and CCOM cases.

Table 4Comparison of air-bone gaps (interquartile range and median) in mesotympanic

and epitympanic cholesteatomas.

Frequency (Hz) Mesotympanic (n = 57) Epitympanic (n = 24) p-Value

Air-bone gap median (dBHL)

500 35 (20–45) 40 (20–55) 0.231

1000 35 (20–45) 35 (22.5–47.5) 0.51

2000 20 (15–30) 25 (10–30) 0.326

3000 20 (10–30) 20 (10–30) 0.969

4000 25 (15–30) 22.5 (12.5–40) 0.491

Mann–Whitney test. dBHL = decibel hearing level; Hz = Hertz.

L.F. Silveira Netto et al. / International Journal of Pediatric Otorhinolaryngology 73 (2009) 1751–1756 1753

only in an anterior quadrant (n = 70; 34%) or only in a posteriorquadrant (n = 27; 13.1%) were selected. In TMs with anteriorperforations, the anterior-inferior quadrant was most frequentlyaffected. In cases of posterior quadrant perforations, the posterior-inferior quadrant perforation was most common.

No significant differences were found in average air-bone gapsfor any frequency when comparing single anterior and posteriortympanic perforations (p = 0.135), nor when comparing perfora-tions affecting both anterior quadrants versus those affecting bothposterior quadrants (p = 0.169).

4. Discussion

4.1. AC and BC thresholds

The analysis of results allows us to infer about the audiologicalprofile of children and teenagers with CSOM, and about the hearingthresholds and air-bone gap compared to different alterationsoccurred to the middle ear.

The AC thresholds have appeared significantly higher in allfrequencies CCOM patient group. Such behavior was alreadyexpected and commented by Austin [4] and Caldas [5]. Thatbecause, in the presence of cholesteatoma, the bone destruction isstill bigger than in the NCCOM cases [6].

The presence of such ossicular erosions results in greaterhearing losses by the loss of the tympanic-ossicular systemconnection. Consequently, there is an impairment of both middleear’s sound amplification mechanism, hydraulic and lever, unlike

Table 3Comparison of air-bone gaps considering clinical diagnosis of CSOM (interquartile

range and median).

Frequency (Hz) NCCOM (n = 206) CCOM (n = 81) p-Value

Air-bone gap median (dBHL)

500 25 (15–35) 40 (20–50) <0.001

1000 20 (10–25) 35 (20–45) <0.001

2000 15 (10–25) 25 (15–35) 0.001

3000 15 (5–25) 20 (10–30) 0.017

4000 20 (10–30) 25 (15–35) 0.025

Mann–Whitney test. NCCOM = chronic otitis media without cholesteatoma;

CCOM = chronic otitis media with cholestetatoma; dBHL = decibel hearing level;

Hz = Hertz.

the alterations restricted only to the TM, which represent about80% of the NCCOM cases, where only the hydraulic mechanismwould be partly damaged. That, of course, as long as the connectionbetween the remnants parts of the TM and the oval windowremains intact [7].

The values found for the AC thresholds in both groups couldsuggest an association between NCCOM and mild hearing losses, andCCOM with a tendency to degrees of loss closer to moderate. Thedifferences between the thresholds obtained by the frequencybetween the groups have remained around 5 and 15 dBHL mainly at10 dBHL, according to what is suggested by Austin [4] and Caldas [5].

A series of studies has investigated the relation between theCSOM and the simultaneous cochlear alterations. The biggestchallenge, obviously, has been to prove that such alterations aredirect consequences of the middle ear pathology. The lowering ofBC thresholds, especially on high frequencies of 2000, 3000 and4000 Hz, has been widely reported in literature, mainly in CCOMpatients [8].

Our data is according to Bento et al. [9] and Papp et al. [10],Noordzij et al. [11], Kaplan et al. [12], Eisenman and Parisier [13]and McAndie and O’Reilly [14], which showed higher BC values at4000 Hz than at 500, 1000 and 2000 Hz.

Several hypotheses have been raised to explain such BCalterations. It is considered that it results from toxins passingfrom the middle ear to the cochlea through the round window,which has its permeability increased due to the chronicinflammatory process; damaging both the outer and inner haircells, located mainly in the basal region of the cochlear spiral [15].Thus, it is reasonable to infer that such damage prevail the most inthe presence of cholesteatoma.

Many authors refer, however, that the BC thresholds loweringshall be more of an expression of the middle ear damages than thereal cochlear condition. That is; that the modifications of the BCmay, in fact, be a result of the sound passing through the impairedossicular chain and, therefore, transmitted to the inner ear, duringthe brain’s own stimulation by the bone-conduction transmitter[12]. Hence, the greater the damage caused to the ossicular chain,the bigger would the impact be on the BC thresholds. Suchphenomenon has been described by Carhart [16], and called‘‘Carhart Notch’’.

4.2. Air-bone gaps behavior

Air-bone gaps were significantly greater for all frequencies inthe CCOM group, as expected. The greater hearing loss at lowfrequencies, especially 500 Hz, is explained by better-preservedBC thresholds and greater damage to AC thresholds. Ourobservations are corroborated by Jeng et al. [17], Durko [18]and Mehta et al. [19].

The difference in air-bone gap values, comparing the frequencyof both groups, was consistent, as expected, around 15 dBHL at 500and 1000 Hz, decreasing to values around 5 dBHL at highfrequencies. Such differences are clinically significant, especiallyfor school-aged children.

Fig. 2. Variation in air-bone gaps according to number of perforated quadrants.

L.F. Silveira Netto et al. / International Journal of Pediatric Otorhinolaryngology 73 (2009) 1751–17561754

Other studies showed different results such as Sakagami et al.[20]. In their study with 31 CCOM patients, they reported an air-bone gap (mean of 500, 1000 and 2000 Hz) of 18 dBHL. Theirfindings were lower than ours, most likely because the authorsselected only patients with intact ossicular chains or with minimalstructural changes, which would favor preservation of AC values.

Heshiki et al. [21] verified air-bone gaps in 80 unilateral NCCOMpatients. The following values were found, respectively, for 500,

1000, 2000 and 4000 Hz: 35.1, 31.4, 28.9 and 31.4 dBHL. All thesevalues were slightly higher than ours.

Direct comparison of our findings with those of the studiesmentioned above is difficult. Results are presented in differentformats. Most studies include a wide range of patient ages, do notdistinguish between CCOM and NCCOM, and use averages of500–4000 Hz frequencies for reporting air-bone gaps. Weconsider it important to analysis according to frequency, since

L.F. Silveira Netto et al. / International Journal of Pediatric Otorhinolaryngology 73 (2009) 1751–1756 1755

we believe, in agreement with Mehta et al. [19] and Ravicz et al.[22], that conductive hearing loss from TM perforations arefrequency-dependent.

Based on our findings, we hypothesize that air-bone gaps inchildren and teenagers with CCOM are usually within the 20–40 dBHL range, significantly higher than air-bone gaps in patientswith NCCOM, which typically have values in the 15–25 dBHLrange. The lower the frequency, the greater the gap.

4.3. Cholesteatoma growth patterns

Swartz [23] claims that bone destruction is the most commoncomplication of cholesteatomas, and that the damage caused to theossicular chain depends on the origin of the cholesteatoma and itsform of expansion. Ginsberg and White [24] agree that differentgrowth patterns could lead to different alterations in the ossicularchain. Swartz [23] reports that the ossicular chain is intact inapproximately 26% of epitympanic cholesteatomas, and in only10% of mesotympanic cholesteatomas.

There were, however, no significant differences between air-bone gaps associated with either cholesteatoma growth pattern inour study. Our data agrees with the findings of Durko [18]. Thisauthor obtained 29.5 dBHL values for air-bone gaps in epitym-panic cholesteatomas and 24.8 dBHL for mesotympanic choles-teatomas (p = 0.25) (mean of 500–4000 Hz). Results allow us toinfer that the air-bone gap seems, therefore, not to be influencedby the growth patterns of this pathology, contrary to ourexpectations.

4.4. Tympanic membrane perforations

Mehta et al. [19] reported that conductive hearing loss becomesmore serious the larger the size of the tympanic perforation. Wemade a similar observation in our study. This positive associationat all frequencies for NCCOM patients between the number ofquadrants affected by tympanic perforation and the air-bone gapvalue was also observed by Prasansuk and Hinchcliffe [25] and byBhusal et al. [26].

An experimental study with rats [27] found a systematicdecrease in the umbo’s velocity as the pars tensa tympanicperforation grows in size, in the presence of an intact ossicularchain. This decrease was greater for low-frequency sound stimuli.Perforation decreases the functional vibratory area of TM, butdoes not nullify it completely. An intact, articulate and mobileossicular chain can use the remaining portion of the TM to amplifythe sound energy that is received. Obviously, the larger theperforation, the less viable this hydraulic mechanism of soundamplification.

A series of authors report that not only the size of thetympanic perforation but also its location determines theincrease in hearing loss. It is believed that limited perforationsto anterior quadrants, with no concurrent alterations to theossicular chain, lead to almost meaningless hearing loss.Perforations of posterior quadrants, on the other hand, especiallythe posterior-inferior, are more relevant to hearing loss, mainlydue to the exposure of the round window. Without protectionfrom the TM, sounds from the external acoustic meatus arriveconcurrently to the oval and round windows. This ends upnullifying the movement of the endocochlear fluid column[2,5,28].

Our results, however, agree with Mehta et al. [19], whichsuggest that hearing loss does not depend on the location of thetympanic perforation, contradicting much of the literature on thesubject. Voss et al. [29] and Mehta et al. [19] believe that the mainmechanism involved in assessing hearing loss in the presence ofTM perforation is the reduction of sound pressure on both sides

of the MT (directly related to the size of the perforation), and notthe nullifying effect of the round window.

4.5. The ossicular chain

The ossicular chain seems to perform a primary role indetermining hearing thresholds. Caldas [5] reports that, with thedevelopment of ossicular chain damage, hearing loss becomesmore significant, regardless of the size of tympanic perforation.

The prevalence of ossicle damage in CCOM patients couldexplain the greater air-bone gaps found in this group. We believethat this reinforces the idea that, in CCOM, alterations in ossicularchain function may be the primary agents responsible for hearingloss. This is in contrast to cases of NCCOM, in which the ossicularsystem is intact and it is the status of the TM that determineshearing thresholds.

5. Conclusions

Based on our data results, we concluded that children andteenagers with CCOM presented significantly greater AC, BCthresholds and air-bone gaps than those with NCCOM. Lowerfrequencies were more compromised than higher ones.

There were no significant differences in air-bone gaps in thepresence of epitympanic and posterior mesotympanic cholestea-tomas.

In patients with NCCOM, there was correlation at allfrequencies between the number of quadrants affected bytympanic perforation and air-bone gap values and there wereno significant differences in air-bone gaps for tympanic perfora-tions affecting posterior quadrants versus those affecting anteriorquadrants.

References

[1] C.D. Bluestone, Epidemiology and pathogenesis of CSOM: implications forprevention and treatment, Int. J. Pediatr. Otorhinolaryngol. 42 (3) (1998)207–223.

[2] S.S. Costa, L.C.A. Souza, Otite media cronica nao-colesteatomatosa, in: C.A.H.Campos, H.O.O. Costa (Eds.), Tratado de otorrinolaringologia: Doencas – otologiae base do cranio, Roca, Sao Paulo, 2003, pp. 72–92.

[3] I. Fukuchi, D.P. Cerchiari, E. Garcia, et al., Timpanoplastias: resultados cirurgicos eanalise dos fatores que podem interferir no seu sucesso, Revista Brasileira deOtorrinolaringologia 72 (2) (2006) 267–271.

[4] D.F. Austin, Chronic otitis media, in: J.J. Ballanger, J.B. Snow, Jr. (Eds.), Otorhino-laryngology—Head and Neck Surgery, Williams & Wilkins, PA, USA, 1996.

[5] N. Caldas, Otite media cronica nao-colesteatomatosa, in: N. Caldas, S. CaldasNeto, T. Sih (Eds.), Otologia e Audiologia em Pediatria, Revinter, Rio de Janeiro,1999 .

[6] S.S. Costa, L. Schmidt, M.M. Smith, et al., Sensorineural hearing loss in patientswith chronic otitis media, in: AAO-HNSF Annual Meeting & OTO EXPO 2005, vol.131, New York, (2004), pp. 231–1231.

[7] S. Caldas Neto, Anatomia da orelha humana, in: N. Caldas, S. Caldas Neto, T. SIH(Eds.), Otologia e Audiologia em Pediatria, Revinter, Rio de Janeiro, 1999.

[8] E. Vartiainen, J. Vartiainen, Age and hearing function in patients with chronicotitis media, J. Otorhinolaryngol. 24 (6) (1995) 336–339.

[9] R.F. Bento, S.S. Grasel, L.K. Ishida, A. Miniti, Estudo clınico da perda auditivaneurossensorial em doentes com otite media cronica, Revista Brasileira de Otor-rinolaringologia 59 (1) (1993) 12–16.

[10] Z. Papp, S. Rezes, I. Jokay, I. Sziklai, Sensorineural hearing loss in chronic otitismedia, Otol. Neurotol. 24 (2003) 141–144.

[11] J.P. Noordzij, E.E. Dodson, R.A. Ruth, H.A. Arts, P.R. Lambert, Chronic otitis mediaand sensorineural hearing loss: is there a clinically significant relation? Am. J.Otol. 16 (4) (1995) 420–423.

[12] D.M. Kaplan, D.M. Fliss, M. Kraus, R. Dagan, A. Leiberman, Audiometric findings inchildren with chronic suppurative otitis media without cholesteatoma, Int. J.Pediatr. Otorhinolaryngol. 35 (1996) 89–96.

[13] D.J. Eisenman, S.C. Parisier, Is chronic otitis media with cholesteatoma associatedwith neurosensory hearing loss? Am. J. Otol. 19 (1998) 20–25.

[14] C. McAndie, B.F. O’Reilly, Sensorineural hearing loss in chronic otitis media, Clin.Otolaryngol. Allied Sci. 24 (3) (1999) 220–222.

[15] S. Cureoglu, P.A. Sachchern, M.M. Paparella, B.R. Lindgren, Cochlear changes inchronic otitis media, Laryngoscope 114 (2004) 622–626.

[16] R. Carhart, The clinical application of bone conduction audiometry, Arch. Otolar-yngol. 51 (1950) 789–808.

L.F. Silveira Netto et al. / International Journal of Pediatric Otorhinolaryngology 73 (2009) 1751–17561756

[17] F.C. Jeng, M.H. Tsai, C.J. Brown, Relationship of preoperative findings and ossiculardiscontinuity in chronic otitis media, Otol. Neurotol. 24 (2003) 29–32.

[18] M. Durko, Air-bone gap and hearing impairment level predictive value in pre-operative assessment of cholesteatoma localization in the tympanic cavity,Otolaryngol. Pol. 58 (1) (2004) 73–77.

[19] R.P. Mehta, J.J. Rosowski, S.E. Voss, et al., Determinants of hearing loss inperforations of tympanic membrane, Otol. Neurotol. 27 (2006) 136–143.

[20] M. Sakagami, T. Seo, M. Node, et al., Cholesteatoma otitis media with intactossicular chain, Auris Nasus Larynx 26 (1999) 147–151.

[21] Z. Heshiki, J.V. Tagliarini, A.C. Javaroni, Repercussao da otite media cronica nao-colesteatomatosa e sem complicacao pregressa sobre a audicao ossea, RevistaBrasileira de Otorrinolaringologia 60 (3) (1994) 174–176.

[22] M.E. Ravicz, J.J. Rosowski, S.N. Merchant, Mechanisms of hearing loss resultingfrom middle-ear fluid, Hear. Res. 195 (2004) 103–130.

[23] J.D. Swartz, Cholesteatoma of the middle ear: diagnosis, etiology and complica-tions, Radiol. Clin. North Am. 22 (1984) 15–34.

[24] I.A. Ginsberg, T.P. White, Consideracoes otologicas em audiologia, in: J. Katz (Ed.),Tratado de audiologia clınica, 4th ed., Manole, Sao Paulo, 1999.

[25] S. Prasansuk, R. Hinchcliffe, Tympanic membrane perforation descriptors andhearing levels in otitis media, Audiology 21 (1) (1982) 43–51.

[26] C.L. Bhusal, R.P. Guragain, R.P. Shivastav, Size of tympanic membrane perforationand hearing loss, J. Nepal Med. Assoc. 45 (161) (2006) 167–172.

[27] D.C. Bigelow, P.B. Swanson, J.C. Saunders, The effect of tympanic membraneperforation size on umbo velocity in the rat, Laryngoscope 106 (1) (1996) 71–76.

[28] O.L.M. Cruz, S.S. Costa, Otite media cronica simples, in: S.S. Costa, O.L.M. Cruz,J.A.A. Oliveira (Eds.), Otorrinolaringologia: princıpios e pratica, Artes Medicas,Porto Alegre, 1994.

[29] S.E. Voss, J.J. Rosowski, S.N. Merchant, et al., How do tympanic membraneperforations affect human middle ear sound transmission? Acta Otolaryngol.121 (2001) 169–173.

Luciana Fick Silveira Netto Audiologist, specialist in Hospital Speech Therapy, MS

in Medical Sciences: Pediatrics, Universidade Federal do Rio Grande do Sul.

Researcher at the Brazilian Otitis Media Center – Hospital de Clınicas de Porto

Alegre (HCPA), Brazil.

Sady Selaimen da Costa ENT, Doctorate in Surgery. Assistant Professor,

Ophthalmology and ENT Department, Universidade Federal do Rio Grande do

Sul, School of Medicine and Graduate Programs in Medical Sciences: Pediatrics and

Surgery, Brazil.

Pricila Sleifer Audiologist at the Hospital de Clınicas de Porto Alegre; Doctorate and

MS in Medical Sciences: Pediatrics, Universidade Federal do Rio Grande do Sul.

Specialist in Clinical Audiology; Professor at Universidade Federal do Rio Grande do

Sul. Researcher at the Brazilian Otitis Media Center – HCPA, Brazil.

Maria Elisa Luce Braga Audiologist, specialist in Clinical Audiology, Master’s

candidate in Medical Sciences: Pediatrics, Universidade Federal do Rio Grande do

Sul. Researcher at the Brazilian Otitis Media Center – HCPA, Brazil.