Comparison of 1000 Hz-, 226 Hz-probe tone tympanometry and magnetic resonance imaging in evaluating the function of middle ear in infants

Abstract

Introduction: Universal newborn hearing screening proved to be an effective measure to screen out hearing-disability infants at a younger age, from which testified the existence of middle ear fluid or effusion was one major source of referred screening result. Some reports had verified the comparative advantage of using the 1000-Hz probe tone for tympanometry in infants, while little was known about the accordance to imaging manifestations, especially the findings of Magnetic Resonance Imaging (MRI).

Objective: To investigate the diagnostic consistency between 226 Hz-, 1000 Hz- probe tone tympanometry and MRI in infants after hearing screening.

Method: 226 Hz and 1000 Hz probe-tone tympanometry were tested followed by MRI in 58 infants. According to MRI results, patients were divided into Group 1 with abnormal middle ears with fluid, Group 2 with normal middle ears. Tympanometry included tympanograms, volume of ear cannal (Vec, mL), compliance of peak (Y, mmho), the width at the pressure of ±50dapa (TW, daPa), the pressure of the peak (TPP, daPa). The data were analyzed by statistic software SPSS19.0.

Results: In group 2, 92.41% were negative tympanograms(n = 73), 7.59% was positive tympanograms(n = 6) at 1000 Hz; while 91.14% was negative tympanograms (n = 72), 8.86% was positive tympanograms (n = 7) at 226 Hz. In group 1, 80.00% were positive tympanograms(n = 28), 20.00% was negative tympanograms (n = 7) at 1000 Hz; while 80.00% was negative tympanograms (n = 28), 20.00% was positive tympanograms (n = 7) at 226 Hz. McNemar test demonstrated that there was no significant difference between 1000 Hz and MRI diagnosis (P = 1.000), while significant difference between 226 Hz and MRI test (P = 0.001). Diagnostic concordance rate between 1000 Hz tympanometry and MRI (Kappa = 0.730, p = 0.000) was significantly higher than 226 Hz tympanometry (Kappa = 0.134, p = 0.095). Normal data of 1000 Hz tympanometry showed Vec was 0.71 ± 0.28ml, Y was 0.82 ± 0.49mmho, TW was 124.60 ± 27.68dapa, and TPP was 11.83 ± 73.73dapa.

Conclusion: 1000 Hz tympanometry significantly had a better diagnostic agreement with MRI than 226 Hz. Clinically, 1000 Hz but not 226 Hz tympanometry is recommended to diagnose otitis middle ear fluid in infants.

Keywords: MRI; Magnetic resonance imaging; OME; Otitis middle ear fluid; Probe tone; Tympanometry.