Intracochlear Pressure Measurements in Scala Media Inform Models of Cochlear Mechanics

In the classic view of cochlear mechanics, the cochlea is comprised of two identical fluid chambers separated by the cochlear partition (CP). In this view the traveling wave pressures in the two chambers mirror each other; they are equal in magnitude and opposite in phase. A fast pressure mode adds approximately uniformly. More recent models of cochlear mechanics take into account the structural complexity of the CP and the resulting additional mechanical modes would lead to distinct (non-symmetric) patterns of pressure and motion on the two sides of the CP. However, there was little to no physiological data that explored these predictions. To this aim, we measured intracochlear fluid pressure in scala media (SM), including measurements close to the sensory tissue, using miniaturized pressure sensors (∼ 80 μm outer diameter). Measurements were made in-vivo from the basal cochlear turn in gerbils. SM pressure was measured at two longitudinal locations in different preparations. In a subset of the experiments SM and ST (scala tympani) pressure data was measured at the same longitudinal location. Traveling wave pressures were observed in both SM and ST, and showed the relative phase predicted by the classical theory. However, SM pressure showed spatial variations that had not been observed in ST, which points to a relatively complex CP motion on the SM side. These data both underscore the first-order validity of the classic cochlear traveling wave model, and open a new view to CP mechanics. INTRODUCTION A physiologically healthy cochlea shows non-linear, sharply tuned basilar membrane responses [? ]. The active mechanism within the cochlea amplifies responses to low and moderate level sounds in a frequency dependent manner, extending the dynamic range and sharpening the tuning of mammalian hearing. Theoretical models of active cochlear mechanics incorporate outer hair cell electromechanics into a model of the passive mechanics. The classic passive traveling wave model predicts equal magnitude and opposite phase of the pressure on either side of the cochlear partition (CP), where the CP is treated as a simple beamed structure separating scala tympani (ST) and scala vestibuli (SV) [? ]. However, modern models of passive cochlear mechanics include a multi-compartment CP and predict different frequency-dependent pressure variations on either side of the CP [? ]. We measured SM and ST pressure in response to pure tone stimulation in gerbil. Presently our preparations are approximately passive (due to the SM cochleostomy) and are useful for testing the predictions of passive cochlear models. METHODS Animal use was approved by Columbia University’s IACUC. Experiments were performed in young adult gerbils anesthetized with pentobarbital. Surgery was performed to expose the cochlea and a hole of diameter ∼ 90 μm was hand-drilled through the cochlear bone in order to access SM and in some experiments ST. In order to penetrate the lateral wall of SM, after opening the bone, a minutiae pin connected to an electric cautery delivered a short pulse of current, resulting in a clean opening. Pure tones at frequencies from 1 to 50 kHz were applied at the ear canal at levels of 40 to 90 dB SPL. Measurements solely in ST showed the nonlinear compression that has been reported previously [? ]. However, even seemingly atraumatic opening of SM led to linear responses in both scalae, most likely due to loss of endocochlear potential. Pressure sensors were inserted and pressure was measured as the sensor approached the sensory tissue. The micro-pressure sensors used in this study were a modified version of the sensor introduced in [? ]: light is delivered by a fiber optic to a sensing membrane, whose vibration varies the amount of light returning to a photodetector. The sensors here employed single-mode optical-fiber coupled to a super-luminescent diode. With this design, the outer diameter of the sensor could be reduced to ∼ 80 μm. Sensors were calibrated in water and air, and the calibrations were flat in frequency. Their sensitivity extended down to a level of ∼60 dB SPL with 1 second of signal averaging. Because of middle ear and intracochlear pressure gain, this corresponds to an ear canal pressure of ∼ 35 dB N or m al iz ed (r eE C ) p re ss ur e (d B ) P ha se (r eE C ) i n cy cl es Frequency (kHz) 10 20 30 40 50 60 C 0 5 10 15 20 25 30 35 40 −2.5 −2 −1.5 −1 −0.5 0 0.5 D Frequency (kHz) In-vivo G0415 post-mortem G0401