Xiaolu Li*Xingkuan Bu*Carlie Driscoll$

*Ear and Hearing Centre, Departmentof Otorhinolaryngology, JiangsuProvince Hospital, Nanjing MedicalUniversity, People’s Republic of China$Division of Audiology, School ofHealth and Rehabilitation Sciences,The University of Queensland,Australia

Key WordsChinese

Normative data

Schoolchildren

Tympanometry

AbbreviationsANOVA: analysis of variance

OME: otitis media with effusion

Peak Ytm: peak compensated static

acoustic admittance

SD: standard deviation

TPP: tympanometric peak pressure

TW: tympanometric width

VEA: equivalent ear canal volume

Original Article

International Journal of Audiology 2006; 45:55�/59

Tympanometric norms for Chinese

schoolchildren

Normas timpanometricas para escolares Chinos

AbstractCurrent tympanometric norms have acknowledged therelevance of age as an influencing factor. However, littleattention has been afforded to other potentialities such asthe non-pathological effects of gender, ear asymmetry,and racial heritage. This study aimed to examine norma-tive tympanometric findings in a large sample of Chineseschoolchildren. Using a Madsen 901 Middle Ear Analy-zer, data was collected from 269 children (538 ears),ranging in age from 6.2�/12.7 years (mean�/9.4 years,SD�/1.7), in Jiangsu province. Descriptive statisticswere calculated for the parameters of equivalent earcanal volume (x�/1.03, SD�/0.25, 90%�/0.68 �/1.46),peak compensated static acoustic admittance (x�/0.58,SD�/0.34, 90%�/0.26 �/1.13), tympanometric width(x�/112, SD�/36, 90%�/62�/156), and peak pressure(x�/�/25, SD�/30, 90%�/�/85�/�/10). Statistically sig-nificant ear asymmetry and grade/age effects were estab-lished, although differences found were minor. Incomparison with past studies in Caucasian paediatricpopulations, the Chinese normative data displayed mini-mal disparities.

SumarioSe ha reconocido en las normas timpanometricas actualesla relevancia de la edad como un factor de influencia. Sinembargo, se ha puesto poca atencion a otros factorespotenciales como el efecto no patologico del genero,la asimetrıa entre oıdos y la herencia racial. Esteestudio intenta examinar los hallazgos normativostimpanometricos en una gran muestra de escolaresChinos. Se utilizo un analizador de oıdo medio Madsen901 y se colectaron datos de 269 ninos (328 oıdos) conedades entre 6.2 y 12.7 anos (media�/9.4 anos, DS�/1.7)en la provincia de Jiangsu . Se calcularon estadısticasdescriptivas para los parametros de equivalenciadel volumen del conducto auditivo externo (x�/1.03,SD�/0.25, 90%�/0.68 �/1.46), la admitancia acusticaestatica pico compensada (x�/0.58, SD�/0.34, 90%�/

0.26 �/1.13), la amplitud del timpanograma (x�/112,SD�/36, 90%�/62�/156), y la presion pico (x�/�/25,SD�/30, 90%�/�/85�/�/10). Se establecieron los efectosestadisticamente significativos en cuanto a asimetrıa deoıdos y diferencia grado/edad, aunque las diferenciasencontradas fueron menores. Los datos normativosmostraron disparidades mınimas, en comparacion conotros estudios en la poblacion pediatrica China.

Several years ago, the American Speech-Language-Hearing

Association provided specific guidelines and referral criteria

for paediatric acoustic immittance screening/tympanometry

(ASHA, 1997). Such recommended norms clearly acknowledged

age as being an important non-pathological factor when

interpreting tympanometric data, as indeed has a multitude of

past research (De Chicchis et al, 2000; Haapaniemi, 1996; Hanks

& Rose, 1993; Palmu et al, 2001; Palmu & Rahko, 2003; Roush

et al, 1995; Silman et al, 1992). However, it should also be

considered that the current ASHA protocol, which is utilized

worldwide, reflects normative studies conducted in predomi-

nantly Caucasian children.

It is estimated that Chinese people comprise approximately

one quarter of the world’s population (US Census Bureau,

1999). Wan and Wong (2002) have demonstrated that tympano-

metric measures in Chinese young adults vary significantly from

those commonly obtained in Western subjects. These investiga-

tors, along with Robinson and Allen (1984), suggested that the

observed normative differences might be related to anatomical

variations in the Eustachian tube between peoples of different

races. As noted by Rushton et al (1997), it is entirely feasible

that the tympanometric dissimilarities also extend to paediatric

populations, particularly in view of the lower prevalence of

otitis media with effusion (OME) in Chinese children (Chen

et al, 2003; Lien et al, 1985). Furthermore, inappropriate

application of Caucasian norms to Chinese children could result

in massive misdiagnosis of middle ear pathology (Tong, 1999).

Yet, no normative values for tympanometric testing in Chinese

children are available.

Due to the absence of normative data for this population, the

present study is aimed at obtaining and examining normative

tympanometric findings in a large sample of Chinese school-

children. The plausibly influencing factors of age/grade, ear

asymmetry (Haapaniemi, 1996; Hanks & Rose, 1993) and gender

(Margolis & Heller, 1987; Roup et al, 1998; Wan & Wong, 2002)

were given due attention. It is hoped that the subsequent analysis

of 90% range values will contribute towards an increase in the

clinical accuracy and efficiency of immittance screening in

Chinese schoolchildren.

Methods

ParticipantsA total of 269 children, studying in a rural primary school in

Nanjing, Jiangsu Province, People’s Republic of China, were

included in the present study. Participants ranged in age from

6.2�/12.7 years (mean�/9.4 years, SD�/1.7), were selected from

every grade in the school (Grades 1�/6), and represented both

ISSN 1499-2027 print/ISSN 1708-8186 onlineDOI: 10.1080/14992020500377881# 2006 British Society of Audiology, InternationalSociety of Audiology, and Nordic Audiological Society

Carlie Driscoll, PhDDivision of Audiology, School of Health and Rehabilitation Sciences,The University of Queensland, St. Lucia, Queensland 4072, Australia.E-mail: carlie.driscoll@uq.edu.au

Received:June 7, 2005Accepted:August 26, 2005

Int J

Aud

iol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

ichi

gan

Uni

vers

ity o

n 11

/05/

14Fo

r pe

rson

al u

se o

nly.

genders in equal proportions. Refer to Table 1 for mean age, age

ranges, and number of participants per grade. In the current

investigation, the terms ‘age’ and ‘grade’ may be considered as

synonymous, as the age distributions of the participants and

their grades corresponded well (e.g. 90% of Grade 1 participants

were approximately 7 years of age, that is, born in the same

year). All participants took part in the study as volunteers

and only children who had returned consent forms signed by

their primary caregivers were included. Three inclusion criteria

were applied to obtain the 269 participants. Firstly, a negative

history of otologic pathology or familial hearing loss was

required. Secondly, otoscopic examination was unremarkable

bilaterally, including no sign of: outer or middle ear pathologies,

structural abnormality, cerumen occlusion, or canal debris.

Thirdly, participants were required to display normal audio-

grams bilaterally, in accordance with WHO (2002) grades of

hearing loss (four frequency average of 500, 1000, 2000, and

4000 Hz air conduction thresholds 5/20 dB HL). Originally, the

sample included 317 children. However, application of the

inclusion criteria resulted in a final working database of 269

participants.

Procedure & MaterialsAll testing was completed on-site, in quiet rooms within the

school and during normal attendance hours. Ambient noise

levels during testing, as measured by a ND2 Xing-Qiu sound

level meter, ranged from 35�/40 dB A. The testing team was led

by a chief otolaryngologist/professor from a major university

hospital and included one otolaryngologist, three audiologists,

and three nurses, all from the same university hospital and all

with specific training in their required duties.

Pure-tone data was obtained using Madsen Orbiter and

Danplex audiometers fitted with TDH-39 headsets and cali-

brated as per ANSI S3.6-1996 specifications (ANSI, 1996).

A single clinical tympanometer (Madsen Zodiac 901 Middle Ear

Analyzer), calibrated in accordance with ANSI S3.39-1987

specifications (ANSI, 1987), was used to collect all tympano-

metric data. Tympanometry was generally performed once in

each ear per participant and tympanograms were successfully

obtained from all participants. In the event that a type B

tympanogram was collected [as per Jerger’s (1970) classification

system], testing was repeated to verify that the result was not

related to inappropriate probe placement. Testing was per-

formed in sweep mode, whereby a 226 Hz probe tone and a

pump speed of 400 daPa/s were automatically produced through

a pressure range of �/200 to �/400 daPa, following detection of

an hermetic seal. Four measurement parameters were recorded;

1. equivalent ear canal volume (Vea): admittance measured at

�/200 daPa,

2. peak compensated static acoustic admittance (Peak Ytm):

peak admittance minus value at �/200 daPa,

3. tympanometric width (TW): 50% of its amplitude from peak

to tail, and

4. tympanometric peak pressure (TPP): the ear canal pressure

corresponding to peak admittance.

Analysis of the resulting data was completed using the

Statistical Package for the Social Sciences (SPSS) software

(version 9.0.0). Descriptive statistics were provided and a

factorial model, which included three factors [gender (male/

female), ear (left/right) and grade (1�/6)] and all interactions, was

fitted to the data to investigate the effects of these variables on

the measurement parameters. The significance of any term was

assessed using the analysis of variance (ANOVA) at a 95%

confidence level. Post-hoc analysis was performed using Tukey’s

HSD Test.

Results

Table 2 displays the mean, standard deviation, and 90% range

values for the test parameters of Vea, Peak Ytm, TW, and TPP,

respectively. Values presented for TW and TPP have been

rounded to the nearest whole number in keeping with the format

of data provided by the test instrument. In view of the following

significant effects (ear asymmetry and grade) being minimal in

absolute magnitude, the normative data presented in Table 2 has

been given in a collapsed format.

Results from the ANOVA revealed a statistically significant

ear effect for the Vea parameter [F(1,538)�/7.759, p�/.006].

Specifically, mean volumes for left ears were marginally greater

than for right ears. A significant grade effect was also found for

Vea [F(5,538)�/6.437, pB/.001], TPP [F(5,538)�/3.890, p�/.002]

and Peak Ytm [F(5,538)�/2.864, p�/.015]. Post-hoc analysis

showed that children in grades 1 and 2 had significantly lower

mean Vea values than those in grades 4�/6 and significantly more

negative TPP mean values than those in grade 4 (pB/.05).

Additionally, children in grade 1 had significantly lower mean

Peak Ytm values than those in grade 5 (pB/.05) (see Table 3). No

further gender, ear, or grade effects and interactions were

evident.

For comparative purposes, Table 4 contains findings from

studies of tympanometric data in Caucasian paediatric popula-

tions.

Table 1. Mean age, age range (in years), and number ofparticipants per grade

Grade x 100% Range N

1 6.9 6.2�/8.0 40

2 8.0 7.2�/9.2 45

3 8.6 7.2�/10.0 46

4 9.9 8.9�/12.6 42

5 10.9 9.9�/12.7 56

6 11.8 10.8�/12.7 40

Table 2. Normative tympanometric data for 538 ears ofChinese schoolchildren (where Vea, Peak Ytm, TW, and TPPrefer to equivalent ear canal volume, peak static acousticadmittance, tympanometric width, and tympanometric peakpressure, respectively)

Vea (cc) Peak Ytm

(mmho)

TW

(daPa)

TPP

(daPa)

Mean 1.03 0.58 112 �/25

SD 0.25 0.34 36 30

90%

range

0.68�/1.46 0.26�/1.13 62�/156 �/85�/�/10

56 International Journal of Audiology, Volume 45 Number 1

Int J

Aud

iol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

ichi

gan

Uni

vers

ity o

n 11

/05/

14Fo

r pe

rson

al u

se o

nly.

Discussion

The present investigation examined normative tympanometric

findings in a large sample of Chinese schoolchildren. A very

small, yet significant, ear effect was found, with larger equivalent

ear canal volumes noted in left ears. This finding is unlikely to be

of any clinical relevance, as ear canal volumes are thought to be

generally similar, though not identical, between left and right

ears (Fowler & Shanks, 2002). As suggested by Haapaniemi

(1996), who also reported an ear asymmetry effect in his sample

of Caucasian schoolchildren, the disparity could reflect the

handedness of the testers. That is, the tympanometry apparatus

may have been held differently between left and right ears as a

result of the hand used by the tester. No physiological basis for

this ear asymmetry effect has been suggested in the literature to

date.

A grade effect was also evident (refer to Table 3); the current

group displayed increasing equivalent ear canal volumes, in-

creasing peak compensated static acoustic admittance, and less

negative tympanometric peak pressure with advancing grades.

These effects, associated with the changing physical dimensions

and efficiency of the middle ear system with advancing age, are

well documented in the literature (De Chicchis et al, 2000;

Fowler & Shanks, 2002; Haapaniemi, 1996; Margolis & Heller,

1987; Palmu et al, 2001; Palmu & Rahko, 2003; Roush et al,

1995) and highlight the importance of utilizing age-specific

normative data. No gender effects were revealed in the current

study. Such is in keeping with Wan and Wong’s (2002) report of

minimal gender differences in Chinese young adults but appears

in contrast to some Caucasian studies that have reported

higher admittance values in boys than in girls (Haapaniemi,

1996; Margolis & Heller, 1987; Osterhammel & Osterhammel,

1979).

In view of the 90% range of normative values obtained in the

present study (refer to Table 2), Vea values B/0.68 or �/1.46 cc,

Peak Ytm values B/0.26 or �/1.13 mmho, TW values �/156

daPa, and TPP values B/�/85 daPa would be considered

abnormal in this sample. Hence, the ASHA (1997) immittance

guidelines for determining the need for medical referral of

children aged 1�/8 years (Peak Ytm B/0.3 mmho or TW]/200)

may not provide the most appropriate criteria for Chinese

elementary schoolchildren. This is particularly the case for

Chinese children in grades 1 and 2, who are likely to be of an

age that is covered by the ASHA guidelines.

In comparison with other tympanometric studies in Cauca-

sian populations (refer to Table 4), the data of the present

investigation best corresponds with that of Haapaniemi (1996).

Minimal difference was noted in Peak Ytm and TPP values

between the two studies, which were conducted in similarly aged

samples. The only divergence was in the Vea parameter, with

Haapaniemi’s subjects displaying lower values than seen in the

present study. The data of the current study was also closely

comparable to that of Hanks and Rose (1993). Although these

authors utilized a slower pump speed of 50 daPa/s and included

subjects with hearing loss in their sample, little difference was

apparent in Vea and Peak Ytm 90% ranges between the two

Table 3. Mean (and standard deviation) tympanometric datafor 538 ears of Chinese schoolchildren per grade, where L, R,Vea, Peak Ytm, and TPP refer to left ear, right ear, equivalentear canal volume, peak static acoustic admittance, and tympa-nometric peak pressure, respectively

Grade

LVea

(cc)

RVea

(cc)

Peak Ytm

(mmho)

TPP

(daPa)

1 0.97 (0.21) 0.95 (0.16) 0.50 (0.20) �/31 (37)

2 0.96 (0.26) 0.92 (0.22) 0.51 (0.23) �/34 (28)

3 1.04 (0.22) 1.00 (0.17) 0.54 (0.33) �/24 (25)

4 1.09 (0.24) 1.08 (0.22) 0.63 (0.55) �/18 (26)

5 1.09 (0.33) 1.06 (0.32) 0.63 (0.29) �/22 (31)

6 1.11 (0.23) 1.08 (0.22) 0.62 (0.29) �/22 (26)

Table 4. Selected tympanometric studies in paediatric populations

N

(ears)

Ages

(years)

Vea

(cc)

Peak Ytm

(mmho)

TW

(daPa)

TPP

(daPa)

Current Study 538 6�/13

Mean 1.03 0.58 112 �/25

90% range 0.68�/1.46 0.26�/1.13 62�/156 �/85�/�/10

Silman et al (1992) 42 3�/10

Mean

90% range 0.35�/1.25 55�/180

Nozza et al (1992) 260 3�/16

Mean 0.90 0.78

90% range 0.40�/1.39

Haapaniemi (1996) 471 6�/15

Mean 0.80 0.50 �/5

90% range 0.50�/1.20 0.30�/1.10 �/80�/�/25

Hanks & Rose (1993) 316 6�/15

Mean 1.00 0.70 �/7

90% range 0.60�/1.50 0.30�/1.50 �/65�/�/20

Tympanometric norms for Chineseschoolchildren

Li/Bu/Driscoll 57

Int J

Aud

iol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

ichi

gan

Uni

vers

ity o

n 11

/05/

14Fo

r pe

rson

al u

se o

nly.

studies. Hanks and Rose presented a slightly higher peak Ytm

mean value and a substantially less negative TPP value, but such

would be expected given their inclusion of older children than in

the current investigation.

Comparing and contrasting the present results with other

studies, such as Nozza et al (1992) and Silman et al (1992), is less

appropriate due to discrepancies in the age ranges of subjects.

For instance, Nozza reported similar Vea values and higher peak

Ytm values than the current study. However, this difference is

not inconceivable given their higher upper age range. Likewise,

although the findings of Silman et al (1992) were similar to those

of the current investigation, the former’s subjects were of a lower

mean age and provided wider 90% ranges for peak Ytm and TW

values.

Comparison of normative tympanometric data between

differing racial populations is a difficult task considering the

paucity of prior large-scale investigations, variations in study

methodology, equipment and age ranges, and limitations in

statistical analysis. In particular, it should be noted that the 90%

ranges provided in the present study may not necessarily reflect

the best, nor only, criteria to use when assessing Chinese

schoolchildren. Firstly, they are reflective of the distribution of

values observed in a healthy group and do not include those with

middle ear dysfunction. Future investigations of tympanometric

data in Chinese children could involve OME validation techni-

ques (such as pneumatic otoscopy or, optimally, myringotomy)

in order to assess the sensitivity and specificity of the currently

described criteria. Secondly, as minimal difference appeared to

exist between the data obtained in this study and that obtained

by Haapaniemi (1996) in his Caucasian sample, it may be

discovered upon replication and expansion of this investigation,

that racially specific normative criteria are unnecessary for

immittance screening of Chinese schoolchildren. This future

examination could compare the accuracy and efficiency of the

Chinese and Caucasian criteria against a gold standard OME

validation technique. Finally, future studies could also examine

the appropriateness of using Chinese adult normative criteria in

screening of Chinese schoolchildren. Wan & Wong (2002)

described Peak Ytm values for their Southern Chinese young

adult sample (mean age�/23 yrs) that were not substantially

different to those presently provided for Northern Chinese

schoolchildren. Again, only further investigation utilizing

OME validation measures will clarify the clinical need for

separate norms.

Conclusion

The current study of tympanometric normative data in Chinese

schoolchildren has revealed some statistically significant ear

asymmetry and grade/age effects. Examination of the 90%

ranges for tympanometric variables suggests that the referral

criteria that are commonly utilized for Western populations in

the younger grades may not be appropriate if applied to Chinese

children. However, the Chinese normative data provided in this

study did not dramatically differ from that of two previous

investigations in Caucasian schoolchildren, which implies that

racially specific criteria may not be required. Additional

investigations will be needed to address this matter and should

include OME validation techniques in order to explore the test

performance of the contained normative ranges.

Acknowledgements

The current investigation was supported by a research grant

from the University of Queensland. The authors would like to

sincerely thank Mr. Zhou Lulin (headmaster of LongDu

Primary School), Mr. Nicholas Culbert (research assistant),

Dr. Ross Darnell (statistician), Ms. Poren Kwong and Ms. Wen-I

Lee (translators) for their assistance during the course of

this study. Thanks are also extended to the journal reviewers,

whose expert advice greatly improved the clinical utility of this

paper.

References

American National Standards Institute. 1987. Specifications for instru-ments to measure aural acoustic impedance and admittance. ANSIS3.39-1987. New York: ANSI.

American National Standards Institute. 1996. Specifications for audio-meters. ANSI S3.6-1996. . New York: ANSI.

American Speech-Language-Hearing Association (ASHA). 1997. Guide-lines for audiologic screening . Rockville MD: American Speech-Language-Hearing Association.

Chen, C.H., Lin, C.J., Hwang, Y.H. & Ku, C.J. 2003. Epidemiologyof otitis media in Chinese children. Clin Otolaryngol , 28, 442�/

445.De Chicchis, A.R., Todd, N.W. & Nozza, R.J. 2000. Developmental

changes in aural acoustic admittance measures. J Am Acad Audiol ,11, 97�/102.

Fowler, C.G. & Shanks, J.E. 2002. Tympanometry. In J. Katz (ed.),Handbook of Clinical Audiology-5th Ed . Baltimore, Maryland:Lippincott Williams & Wilkins, pp. 175�/204.

Haapaniemi, J.J. 1996. Immittance findings in school-aged children. EarHear, 17, 19�/27.

Hanks, W.D. & Rose, K.J. 1993. Middle ear resonance and acousticimmittance measures in children. J Sp Hear Res , 36, 218�/222.

Jerger, J.F. 1970. Clinical experience with impedance audiometry. ArchOtolaryngol , 92, 311�/324.

Lien, C-F., Goan, M-L., Huang, J-L., Shiao, A-S. & Chang, P. 1985. Asurvey of otitis media among primary school children in Fu-ShienShang and Wulai Shang. Taiwan J Otolaryngol Head Neck Surg , 20,13�/20.

Margolis, R.H. & Heller, J.W. 1987. Screening tympanometry: Criteriafor medical referral. Audiol , 26, 197�/208.

Nozza, R.J., Bluestone, C.D., Kardatze, D. & Bachman, R. 1992. Towardthe validation of aural acoustic immittance measures for diagnosisof middle ear effusion in children. Ear Hear, 13, 442�/453.

Osterhammel, D. & Osterhammel, P. 1979. Age and sex variations for thenormal stapedius reflex thresholds and tympanometric compliancevalues. Scand Audiol , 8, 153�/158.

Palmu, A., Puhakka, H., Huhtala, H., Takala, A.K. & Kilpi, T. 2001.Normative values for tympanometry in 7- and 24-month-oldchildren. Audiol , 40, 78�/184.

Palmu, A. & Rahko, T. 2003. Normative values for tympanometry in 4-5-year-old children. Int J Audiol , 42, 327�/330.

Robinson, D.O. & Allen, D.V. 1984. Racial differences in tympanometricresults. J Sp Hear Dis , 49, 140�/144.

Roup, C.M., Wiley, T.L., Safady, S.H. & Stoppenbach, D.T. 1998.Tympanometric screening norms for adults. J Audiol , 7, 55�/60.

Roush, J., Bryant, K., Mundy, M., Zeisel, S. & Roberts, J. 1995.Developmental changes in static admittance and tympanometricwidth in infants and toddlers. J Am Acad Audiol , 6, 334�/338.

Rushton, H.C., Tong, M.C.F., Yue, V., Wormald, P.J. & van Hasselt,C.A. 1997. Prevalence of otitis media with effusion in multiculturalschools in Hong Kong. J Laryngol Otol , 111, 804�/806.

Silman, S., Silverman, C.A. & Arick, D.S. 1992. Acoustic-immittancescreening for the detection of middle-ear effusion in children. J AmAcad Audiol , 3, 262�/268.

Tong M.C.F. 1999. Otitis media with effusion in Chinese pre-school andschool children in Hong Kong. Is the disease different from that in the

58 International Journal of Audiology, Volume 45 Number 1

Int J

Aud

iol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

ichi

gan

Uni

vers

ity o

n 11

/05/

14Fo

r pe

rson

al u

se o

nly.

West? Unpublished doctoral dissertation : Chinese University of HongKong, Hong Kong.

US Census Bureau 1999. Report WP/98, World Population Profile: 1998 .Washington DC: US Government Printing Office.

Wan, I.K.K. & Wong, L.L.N. 2002. Tympanometric norms for Chineseyoung adults. Ear Hear, 23, 416�/421.

World Health Organization. 2002. Grades of hearing impairment .www.who.int/pbd/pdh/Docs/GRADESTable-DEFs.pdf.

Tympanometric norms for Chineseschoolchildren

Li/Bu/Driscoll 59

Int J

Aud

iol D

ownl

oade

d fr

om in

form

ahea

lthca

re.c

om b

y M

ichi

gan

Uni

vers

ity o

n 11

/05/

14Fo

r pe

rson

al u

se o

nly.