Energy reflectance and tympanometry in normal and otosclerotic ears

The major goal of this study was to examine differences in the middle ear mechano-acoustical properties of normal ears and ears with surgically confirmed otosclerosis using conventional and multifrequency tympanometry (MFT) as well as energy reflectance (ER). Second, we sought to compare ER, standard tympanometry, and MFT in their ability to distinguish healthy and otosclerotic ears examining both overall test performance (sensitivity and specificity) and receiver- operating characteristic analyses.

Sixty-two normal-hearing adults and 28 patients diagnosed with otosclerosis served as subjects. Tympanometric data were gathered on a clinical immittance machine, the Virtual 310 equipped with a high-frequency option. Two of the parameters, static admittance and tympanometric width, were measured automatically at a standard 226 Hz frequency. The remaining two parameters, resonant frequency and frequency corresponding to admittance phase angle of 45 degree (F45 degrees ), were derived from MFT, multicomponent tympanometry, using a mathematical approach similar to the method used in GSI Tympstar Version 2. ER data were gathered using Mimosa Acoustics (RMS-system v4.0.4.4) equipment.

Analyses of receiver-operating characteristic plots confirmed the advantage of MFT measures of resonant frequency and F45 degrees over the standard low-frequency measures of static admittance and tympanometric width with respect to distinguishing otosclerotic ears from normal ears. The F45 degrees measure was also found to be the best single index for making this distinction among tympanometric parameters. ER less than 1 kHz was significantly higher in otosclerotic ears than normal ears. This indicates that most of the incident energy below 1 kHz is reflected back into the ear canal in otosclerotic ears. ER patterns exceeding the 90th percentile of the normal ears across all frequencies correctly identify 82% of the otosclerotic ears while maintaining a low false alarm rate (17.2%); thus, this measure outperforms the other individual tympanometric parameters. Combination of ER and F45 degrees were able to distinguish all otosclerotic ears. Correlations and the individual patterns of test performance revealed that information provided by ER is supplemental to the information provided by conventional and MFT with respect to distinguishing otosclerotic ears from normal ears.

The present findings show that the overall changes of ER across frequencies can distinguish otosclerotic ears from normal ears and from other sources of conductive hearing loss. Incorporating ER in general practice will improve the identification of otosclerotic ears when conventional tympanometry and MFT may fail to do so. To further improve the false alarm rate, ER should be interpreted in conjunction with other audiologic test batteries because it is unlikely that signs of a conductive component, including abnormal middle ear muscle reflex and ER responses, would be observed in an ear with normal middle ear function.