Objective assessment of autophony during phonation in the diagnosis of patulous Eustachian tube patients

A system enabling the objective assessment of the transmission of voice sounds to the external auditory canal (EAC) during phonation has recently been revised. Our aim was to evaluate the effectiveness of this new system in the diagnosis of patulous Eustachian tube (PET) patients by comparing the results obtained using this method with those obtained from conventional objective tests to diagnose PET. A prospective survey of medical records was included with definite PET, possible PET, and sensorineural hearing loss as control. The measurement system consists of a personal computer, an AD/DA converter (NI 6361, National Instruments), a probe microphone system for recording voice sound (ER-10C, Etymotic Research) and two microphones for measuring noise sound in the EAC (ER-10B+, Etymotic Research). Pronouncing the "Ni" sound for 5 s were recorded with these three microphones. The ratio of the maximum sound pressure of voice sound and noise sound in EAC (EAC/Voice) was simultaneously calculated, and results were displayed on a personal computer for diagnosing. Thirty-one patients of 42 ears with definite PET, 26 patients of 38 ears with possible PET, and 12 patients of 24 ears with sensorineural hearing loss as control were included. The EAC/Voice were 8.63 ± 5.43, 25.41 ± 32.63, and 25.87 ± 24.93 in the control, definite PET, and possible PET group respectively. The control group was significantly different from the definite PET (p < 0.05) and possible PET group (p < 0.05). ROC curve analysis confirmed 14.7 as the best diagnostic cut-off value of EAC/Voice (area under the curve=0.782, 95% CI 0.671–0.894). By adopting this cut-off point, 25 (56.8%) and 22 (61.1%) ears were determined as positive findings in the definite PET and possible PET group, respectively. There was no significant correlation between the positive findings judged by the current method and that of sonotubometry in the control (r = -0.63, p = 0.769), definite PET (r = 0.12, p = 0.451), and possible PET group (r = 0.12, p = 0.451). The current system is more useful in the objective assessment of autophony during phonation by calculating the ratio of voice sound and elicited noise sound transmitted in the EAC (EAC/Voice). This method seems promising because it is able to detect cases eluding conventionally used test methods such as sonotubometry performed without phonation, thereby increasing the accuracy of PET diagnoses.