Tympanometry in infants with middle ear effusion having been identified using spiral computerized tomography☆

Abstract 

Purpose

This study was carried out to evaluate the diagnostic value of 226 and 1000 Hz probe-tone tympanometry in infants with effusion in the middle ear.

Methods

For this study, we recruited 52 infants with ages 42 days to 6 months as clinical subjects in a hearing-ability screening program. After a spiral computerized tomography (CT) scan of the patients, we tested their hearing using tympanometry of 2 probe-tone frequencies: 226 and 1000 Hz. The patients were divided into 2 groups according to the results of the CT scan: group 1 patients had normal middle ears without fluid, and group 2 patients had ears with fluid. We recorded the tympanograms and their percentage of every type and compared the tympanometric results with CT to get the concordance rate between tympanometry and CT diagnose while obtaining the normal values of 1000 Hz tympanometric measures.

Results

The 226 Hz probe-tone tympanograms of middle ears with fluid differed greatly from those without. At 226 Hz, their tympanograms were single-peaked tympanograms (51.06%), double-peaked tympanograms (44.68%), flat tympanograms (2.13%), and negative-pressure tympanograms (2.13%) for the group with normal middle ears, but single-peaked tympanograms (77.19%), double-peaked tympanograms (19.30%), and flat tympanograms (3.51%) for the group with middle ear effusion. The 1000 Hz probe-tone tympanograms included single-peaked or flat-type tympanograms in both the normal middle ear group and the group with middle ear effusion. The group with normal middle ears was identified by spiral CT, and its tympanograms mostly had a single peak (97.87%) during 1000 Hz tympanometry. Tympanograms of the middle ear effusion group mostly had a flat curve (98.25%). When the Liden/Jerger classification system was used to evaluate the tympanograms, normal tympanograms were single-peaked, and flat tympanograms indicated middle ear effusion. According to this standard, the concordance rate between the 1000 Hz tympanometry (98.08%) and CT diagnosis was higher than when 226 Hz tympanometry (25%) (P < .05) was performed, and the value of κ was equal to 0.961 between 1000 Hz tympanometry and CT diagnosis.

Conclusions

In clinical practice, 1000 Hz tympanometry, not 226 Hz, is recommended to determine the presence of middle ear fluid in infants younger than 6 months.