Objective: For more than a century, positive and/or negative ear canal air pressure (ECP) has often been employed to simulate negative middle ear pressure (MEP) in exploring the latter's effect on hearing sensitivity and outcomes of various physiological assessments of the auditory system. However, systematic investigation is lacking on validation of these practices. This study was aimed at comparing these air pressure variations in humans in terms of effect on distortion product otoacoustic emissions (DPOAEs) and discussing certain issues pertaining to the middle ear transfer function.

Design: The 2f1 - f2 DPOAE was measured for nine f2 frequencies from 600 to 8000 Hz in 27 adult ears under four air pressure conditions: normal MEP, negative MEP, and positive and negative ECPs. The subjects voluntarily induced negative MEPs with magnitudes ranging from -40 to -420 daPa, as estimated by the tympanometric peak pressure. For each negative MEP, positive and negative ECPs were applied, respectively, at the same magnitude in absolute value as the negative MEP after it was equalized. Negative MEP and ECP variations were compared in terms of change in DPOAE level.

Results: Positive ECP resembled negative MEP, showing a distinct frequency-specific model of effect on DPOAE level: (1) DPOAEs were attenuated the greatest for frequencies at and below 1000 Hz, which increased from 4-6 to 10-12 dB with increasing the pressure for the tested range; (2) DPOAE attenuation decreased with increasing frequency and was minimal at 2000 Hz; and (3) DPOAE level significantly declined for the frequencies between 2000 and 6000 Hz and tended to increase for high frequencies. Compared with a negative MEP, an equivalent negative ECP yielded a smaller reduction of DPOAE levels for frequencies below 2000 Hz, as well as 3000 Hz, but greater reduction for frequencies above 4000 Hz. One phenomenon that occurred under all three air pressures was a minimal change of the DPOAE level at 2000 Hz. This resulted in a peak at 2000 Hz in the DPOAE level change versus frequency function.

Conclusions: Effects of negative MEP and ECP variations on DPOAEs in human ears are comparable, to a great extent, to findings from previous studies on hearing sensitivity in humans and tympanic membrane vibration at the umbo in human temporal bones. The present study demonstrates that positive ECP can be used to simulate negative MEP in research on the middle ear function in live humans. Results also suggest that long-lasting beliefs regarding the ECP effect on the middle ear conduction should be amended: (1) only positive ECP, not negative ECP, attenuates sound transmission more for low frequencies than for high frequencies, and (2) positive ECP has a greater effect than negative ECP only for low frequencies and not for high frequencies. Discussion of the present results together with those from previous studies sheds light on the middle ear dynamics under diverse pressure changes across the tympanic membrane and proposes that distorted configuration of the tympanic membrane and ossicular chain is the key factor in the effect of MEP or ECP on the middle ear sound transmission for low frequencies.