Characteristics of Tympanometric Gradient and Shape in Normal Adults

Joo, Ga Young; Kim, Jin Sook; Lee, Jung Hak; Cho, Soo Jin; Kim, Kyu Sung

Original Article

Hearing scieneces

Korean Journal of Audiology 2003;7(2):123-130.

Characteristics of Tympanometric Gradient and Shape in Normal Adults

Ga Young Joo¹, Jin Sook Kim², Jung Hak Lee³, Soo Jin Cho², Kyu Sung Kim¹

¹Department of Otolaryngology-Head & Neck Surgery, College of Medicine, Inha University, Incheon
²Division of Speech Pathology Audiology, College of Natural Science, Hallym University, Chuncheon
³Department of Otolaryngology-Head & Neck Surgery,

정상 성인에서 고막운동계측도의 기울기와 형태적 특성

주가영^¹^, 김진숙^²^, 이정학^³^, 조수진^²^, 김규성^¹

^¹인하대학교 의과대학 이비인후과학교실
^²한림대학교 자연과학대학 언어청각학부
^³한림대학교 의과대학 이비인후과학교실

Keywords: Tympanometry;Tympanometric gradient;Tympanometric shape.

Correspondence Author：Ga Young Joo, 400-711 7-206, 3-Ga, Shin Heung-Dong, Jung-Gu, Incheon, Korea
Tel) (031) 890-2434, Fax) (031) 890-2430, E-mail：kyo-79@hanmail.net

Introduction

Tympanometry is an objective measure which evaluates middle ear status indirectly and has been used as an important component of audiologic evaluation. Tympanometry is suited to identifying the status of middle ear. Therefore, tympanometry is performed to screen for a middle ear disorder.
Various parameters including tympanometric peak pressure, the equivalent volume of the ear canal, compliance, gradient, and shapes were considered to analyze tympanometric results whether normal or abnormal. Previous studies suggested that these tympanometric parameters depended on frequency, rate, and direction of ear-canal pressure changes.¹⁾²⁾ Tympanometric gradient is an objective measure that describes the steepness of the slope of the tympanogram near the peak and also a quantitative expression of the shape of a tympanogram. Gradient was originally defined as the difference in "compliant" at the peak and where the tympanogram was 50 daPa wide on either side of the peak.³⁾⁴⁾ Previous work⁵⁾ suggested that tympanometric gradient and shape measures may be diagnostically and prognostically useful in the assessment and management of the middle ear disease. That is, the importance of them rests in their potential to give a more accurate estimate of middle ear effusion than either tympanometric peak pressure or static admittance.⁶⁾ The effects of frequency of ear-canal pressure changes on tympanometric gradient and shape have been studied by several investigators.⁵⁾⁷⁾ Actually, middle ear transmission is affected by the frequency of the acoustic stimulus.⁸⁾ Liden, et al.⁹⁾ noted two characteristic tympanogram patterns recorded using 800 Hz probe tone. The sharp notched tympanograms were recorded from ears with hypermobile eardrums and atrophic scarring, and broad, undulating peaked tympanograms. Those were recorded from ears with ossicular discontinuity and status poststapedectomy which were mass related middle ear problems. Y-226 Hz tympanogram recorded from these same ears and they were always single peaked patterns. That is, measures made with high-frequency probe tones provided information regarding the mass characteristics of the middle ear, particularly of the eardrum and ossicles. And measures using low-frequency probe tones provided information on the stiffness characteristics of the middle ear and on the volume of air medial to the probe tip. In the spite of the clinical value of tympanometry using high-frequency probe tones, generally low-frequency probe tone has been routine at clinical setting. The major reason was that the measures with high-frequency probe tone were more complex and not as easily understood as Y-226 Hz tympanometry.
Many investigators also noted that the effects of the direction and rate of ear-canal pressure changes in tympanometry. Margolis and Smith¹⁰⁾ have observed a higher incidence of tympanometric notching and deepening of the notch for ascending than for descending pressure changes. Creten and Van Camp¹¹⁾ have noted that abnormal tympanometric shape increased for fast rates, more than 30 daPa/s, with high frequency probe tones.
Tympanometric norms that are applied clinically have not considered gender or race factors until these days. However, as previous studies have suggested, tympanometric measurements may depend on gender and race,¹²⁾¹³⁾¹⁴⁾ and therefore, tympanometric norms which is suited for Korean should be established. As applying the results in clinical settings for gender and race specificity, we could improve test reliability and make more accurate interpretation of middle-ear disorders possible. Recently, the study of tympanometric measurements in Korean normal young adults was reported.¹⁵⁾ The study noted the effects of frequency, rate, and direction of stimuli on three tympanometric measures, ear canal volume, compliance, and tympanometric peak pressure. Unfortunately, the study did not analyze the tympanometric measurements using much higher frequency such as 1000 Hz.
The purpose of this study is to investigate the characteristics of tympanometric gradient and shape for native Korean. For this purpose, tympanograms obtained from 40 young adults with normal middle-ear transmission system were analyzed considering the effects of low and high frequencies, rate, and direction of ear-canal pressure changes.

Materials and Methods

Forty young adults (20 men, 20 women) whose average age was 25.38 years participated in this tympanometric examination. The population had normal hearing sensitivity, lower threshold than 20 dBHL and presence of an ipsilateral acoustic reflex reported at normal levels, less than 105 dBHL. The ear-canal was free from debris and tympanic membrane and middle ear did not appear any sign of infection. The population had no histories of any otologic disorders and familial hearing loss. All of them were native Korean.
Tympanograms were obtained using a Grason-Stadler, Inc (GSI) Tympstar Version1. At first, the status of ear canal and tympanic membrane, middle ear effusion or structual abnormality were identified by otoscopic examination. And then, tympanometric data was obtained with three frequency conditions；226, 678, and 1000 Hz, three rate conditions；12.5, 50.0, and 200 daPa/s, and two directions；ascending(-/+), and descending (+/-). As a result, the following 18 conditions were measured：
1) 226 Hz, ascending (-/+) at 12.5, 50.0, and 200 daPa/s；
2) 226 Hz, descending (+/-) at 12.5, 50.0, and 200 daPa/s；
3) 678 Hz, ascending (-/+) at 12.5, 50.0, and 200 daPa/s；
4) 678 Hz, descending (+/-) at 12.5, 50.0, and 200 daPa/s；
5) 1000 Hz, ascending (-/+) at 12.5, 50.0, and 200 daPa/s；
6) 1000 Hz, descending (+/-) at 12.5, 50.0, and 200 daPa/s.
The range of ascending direction (-/+) of ear canal pressure changes was from -400 daPa to +200 daPa and the range of descending direction (+/-) was from +200 to -400 daPa. Admittance tympanograms were recorded with all conditions, and at 50 daPa/s rate condition, susceptance and conductance tympanograms were also recorded. Previous studies noted that peak static admittance increases with repeated tympanometric runs.¹⁶⁾ To minimize the effect of repeated tympanometric runs, tympanograms were obtained after three trials. The data were analyzed with the significant level of .05 in the ANOVA by the SAS 8.0 software.

Results

Tympanometric gradient

Tympanometric gradient was recorded at 226 Hz which is usually used in clinical setting. Table 1. indicates the tympanometric results of descriptive statistics with various conditions. Tympanometric gradient means and standard deviations were larger with ascending than descending ear-canal pressure changes for all conditions (Table 1). Averaged in total participants, the tympanometric gradient value was 85.32 in descending and 101.25 daPa in ascending pressure changes. In comparison with gender, the mean difference values of female computed by directions were always larger than those of male at all three rate conditions. The mean difference value between ascending and descending were greatest at 12.5 daPa/s for both females and males. Tympanometric gradient values had a significant effect both on gender and direction (p<.05). Significant rate differences did not exist statistically. And the correlations between gender and direction, direction and rate, or rate and gender did not have statistically significant results.

Tympanometric shape

The admittance tympanogram shape has been investigated as a function of three probe tone frequencies, two directions, and three rates. Also susceptance and conductance tympanogram shapes were analyzed with one rate condition, 50.0 daPa/s. It was founded that as the variables change, the tympanogram undergoes characteristic modification in four classes：Type A, notched, flat, and rising shape (Fig. 1).
Table 2 indicates the incidence of the four shapes as a function of frequency. At low probe tone frequency, usually normal type, Type A tympanograms were observed at all conditions. As the frequency of probe tone shifts, unuasual shaped tympanograms including notched, flat, and rising shapes were increased. At 226 Hz, only Type A tympanograms were observed. At 678 Hz, about 75% showed flat and rising with almost equal incidences. Notched shapes were found the least. At 1000 Hz, about 76% composed with notched and flat shapes with almost equal incidences. Type A tympanograms were found the least. For both frequencies, flat shape showed almost same composition, while more rising shapes at 678 Hz and more notched shapes at 1000 Hz were observed (Fig. 2). When compared with directions, at high frequencies, higher incidence of notched and rising shaped tympanograms were observed for ascending pressure changes. In contrast, flat shaped tympanograms were only observed for descending pressure changes.
The incidence of four shapes as a function of admittance components [Y (admittance), B (susceptance), G (conductance), B/G] was also analyzed only at 50 daPa/s rate condition. No matter what the admittance components were, only Type A tympanograms were observed at 226 Hz. It was noted that the conductance tympanograms presented quite high percentage of Type A, even though the incidence of unusual shaped tympanograms were increased as the frequency of probe tone shifts. According to stimulating frequency the incidence of Type A tympanograms was changed. The percentage of Type A tympanograms at 678 Hz was 18.8% in susceptance, 20.8% in admittance, 18.8% in susceptance/conductance, and 93.8% in conductance tympanogram. At 1000 Hz, the percentage of Type A tympanograms was 3.8%, 4.6%, 1.3%, and 85.0%, relatively (Fig. 3).
These results showed that at high frequencies such as 678 or 1000 Hz, conductance was more stable than any other admittance components. At 678 Hz, B, B/G, and Y components showed flat and rising shapes fairly. At 1000 Hz, more notched shapes were observed in B, B/G, and Y components, especially with more B and B/G components. When compared with directions, at 678 Hz, all admittance components have not showed notched shape with descending and flat shapes with ascending. At 1000 Hz, all admittance components have not showed rising shapes with descending and flat shapes with ascending.

Discussion

As the concern about early detection of hearing loss, treatment, and rehabilitation was increased, tympanometry has been widely used as a clinically useful measurement. However, the study about tympanometry test protocol which suits for native Korean adults is insufficient. For the accurate tympanometric evaluation as well as the standardized test protocol among clinical settings, tympanometry should be studied more closely. In addition, tympanometric gradient and shape should be included for the systematic analysis of middle ear characteristics. Through this study, we could perform reliable tympanometric evaluation in more flexible way for more exact interpretation of hearing loss. At the present study, the characteristics of tympanometric gradient and shape for native Korean adults were analyzed.
At first, typanometric gradient was analyzed as a function of the rate and direction of ear canal change with 226 Hz probe tone. Tympanometric gradient value was larger with ascending than descending ear-canal pressure changes. And the gradient value for male participants was larger than those for female participants. The reason of these significant differences is not certain, but it might be the size of ear-canal volume or the probe placement. Even though the rates were different in tympanometric results, it was negligibly small. In the past, the gradient data were not quantitatively comparable because of the differences in measurement units. It was not real physical units.⁵⁾ Instead, several previous investigators studied tympanometric width which is used to quantify tympanogram shape in the vicinity of the peak. Unfortunately, tympanometric width was not analyzed in the present study. Further studies were required the clinical usefulness of tympanometric width.
It was noted that tympanometric shapes generally progressed toward more complex patterns with increasing probe frequency.¹⁷⁾ The present study represented the same results. As probe frequencies increased from 226 Hz to 1000 Hz, various tympanometric shapes including notch, flat and rising were observed. About the effect of direction of ear-canal pressure, Wilson, et al.¹⁶⁾ reported a higher incidence of notched 678 Hz tympanograms for ascending than for descending pressure changes. Their results agree partially with the present study since a higher incidence of notched and rising tympanograms were noted for ascending and flat tympanograms were only noted for descending at 678 and 1000 Hz. In comparison among admittance components, the most stable and normal shapes have been observed for conductance tympanogram at even high frequencies. That is, the conductance magnitudes were less affected by frequency than other magnitudes such as admittance and susceptance. There are many studies which demonstrated the clinical usefulness of a high-frequency probe tone in identifying mass-related pathology measures. Hirsch, et al¹⁸⁾ have reported that infants, children and adults with confirmed middle ear disease represented apparently normal tympanograms with 226 Hz probe tone, while abnormal tympanograms at higher frequencies. Especially, at higher frequencies, the simple visual analysis of susceptance and conductance tympanograms could provide valuable information. Conclusively, when high frequency probe tone is used in clinical settings, the conductance tympanogram which represented the most normal and stable shapes in normal ears would be recommended. If the shape characteristics of conductance tympanogram studied additionally in pathologic ears, more useful test protocol would be established.

Summary and Conclusion

The present study showed the characteristics of tympanometric gradient and shape as a function of frequency, direction, and rate of middle ear pressure changes. The results were as follows：First, tympanometric gradient was affected by direction of middle ear pressure changes. That is, the gradient mean value was larger for ascending direction change. Second, there were gender differences in tympanometric gradient. The mean gradient value for male was larger than that of female. Third, as the frequency of probe tone shifted, high incidence of unusual types of tympanometric shapes such as notched, flat, and rising were observed. Fourth, at high frequencies, the conductance (G) was less affected by frequency than admittance (Y) and susceptance (S).
Based on the present findings, it is suggested that the tympanometric gradient and shape should be considered for reliable evaluation as well as accurate interpretation of middle ear status using tympanometry. The additional studies about tympanometry variables on young infants and adults who had middle ear disorders are advisable.

REFERENCES

Feldman R, Fria T, Palfrey C, Dellecker C. Effects of rate of air pressure change on tympanometry. Ear Hear 1984; 5:91-5.
Wilson RH, Shanks JE, Kaplan SK. Tympanometric changes at 226 Hz and 678 Hz across 10 trials and two directions of ear canal pressure change. J Speech Hear Res 1984;27(2):257-66.
Brooks D. An objective method of determining fluid in the middle ear. Intern Audiol 1968;7:280-6.
Brooks D. The use of electro-acoustic bridge in the assessment of middle ear function. Intern Audiol 1969;8:563-5.
Koebsell KA, Margolis RH. Tympanometric gradient measured from normal preschool children. Audiol 1986;25: 149-57.
De Jonge R. Normal tympanometric gradient: A comparison of three methods. Audiol 1986;25:299-308.
Colletti V. Multifrequency tympanometry. Audiol 1977; 16:278-87.
Pickles JO. An Introduction to the Physiology of Hearing (2nd ed.). London, Academic Press, 1991.
Liden G, Petersin JL, Bjorkman G. Tympanometry: A method for analysis of middle ear function. Acta Otolaryngol 1970;263:218-24.
Margolis RH, Goycoolea HG. Multifrequency tympanometry in normal adults. Ear Hear 1993;14(6):408-13.
Creten WL, Van Camp KJ. Transient and quasi-static tympanometry. Scand Audiol 1974;3:39-42.
Wan IKK, Wong LLN. Tympanometric norms for chinese young adults. Ear Hear 2002;23(5):416-21.
Roup CM, Wiley TL, Safady SH, Stoppenbach DT. Tympanometric screening norms for adults. J Am Acad Audiol 1998;7:55-60.
Hall JW. Effects of age and sex on static compliance. Arch Otorhinolaryngol 1979;105:153-6.
Kim JS. Effects of frequency, rate, and direction of stimuli on tympanometric measures. Korean J Commun Disoreders 2003;8(2):179-95.
Wilson RH, Shanks JE, Kaplan SK. Tympanometric changes at 226 Hz and 678 Hz across 10 trials and two directions of ear canal pressure change. Jf Speech Hear Res 1984;27(2):257-66.
Holte L, Margolis RH, Cavanaugh RM. Developmental changes in multifrequency tympanograms. Audiol 1991; 30:1-24.
Hirsch JE, Margolis RH, Rykken JR. A comparison of acoustic reflex and auditory brainstem response screening of high-risk infants. Ear Hear 1991;13:181-6.