The conductive hearing loss due to an experimentally induced middle ear effusion alters the interaural level and time difference cues to sound location

Abstract

Otitis media with effusion (OME) is a pathologic condition of the middle ear that leads to a mild to moderate conductive hearing loss as a result of fluid in the middle ear. Recurring OME in children during the first few years of life has been shown to be associated with poor detection and recognition of sounds in noisy environments, hypothesized to result due to altered sound localization cues. To explore this hypothesis, we simulated a middle ear effusion by filling the middle ear space of chinchillas with different viscosities and volumes of silicone oil to simulate varying degrees of OME. While the effects of middle ear effusions on the interaural level difference (ILD) cue to location are known, little is known about whether and how middle ear effusions affect interaural time differences (ITDs). Cochlear microphonic amplitudes and phases were measured in response to sounds delivered from several locations in azimuth before and after filling the middle ear with fluid. Significant attenuations (20-40 dB) of sound were observed when the middle ear was filled with at least 1.0 ml of fluid with a viscosity of 3.5 Poise (P) or greater. As expected, ILDs were altered by ~30 dB. Additionally, ITDs were shifted by ~600 μs for low frequency stimuli (<4 kHz) due to a delay in the transmission of sound to the inner ear. The data show that in an experimental model of OME, ILDs and ITDs are shifted in the spatial direction of the ear without the experimental effusion.

FIG. 1

Experimental setup for instilling silicone oil into the middle ear space. A coated silver wire electrode is placed on the round window (RW) to record the cochlear microphonic (CM) and compound action potential (CAP) waveforms. Different viscosities of silicone oil are injected into the middle ear space through an inferior filling tube located in the posterior portion of the bulla. A vent hole is drilled in the superior portion of the bulla to let air escape as the fluid is instilled from the bottom of the bulla. The indicated fluid volumes are based on anatomical examination and are thus approximately anatomically correct.

FIG. 2

Example of CM amplitude and threshold measurements for baseline condition (closed circles, no fluid) and middle ear fluid condition (1.5 ml of 100 P silicone oil, open circles) in one animal with a 4-kHz tone stimulus. The CM amplitude is substantially reduced in the presence of middle ear fluid. Thresholds of CM responses are shifted towards higher stimulus levels by as much as ~40 dB SPL. Comparable amplitude and thresholds shifts were seen at all frequencies tested (250 Hz to 12 kHz).

FIG. 3

Across-animal mean (n = 13 ears) attenuation of sounds presented at 0 ° azimuth using four different viscosities of silicone oil at four different volumes (0.5, 1.0, 1.5 ml, and bulla full). Positive attenuation values indicate a larger CHL. Attenuation remained nearly consistent across viscosities above 3.5 P (A, 3.5 P; B, 100 P; C, 600 P). There was an effect of fluid volume on sound attenuation for all three viscosities tested, with concomitant significant increases in CM attenuation for all fluid volumes >0.5 ml. Symbols and error bars indicate the mean attenuation and ±1 SD, respectively.

FIG. 4

Baseline interaural level differences for eight different frequencies (0.25–12 kHz) as a function of sound source azimuth (−90 ° = left side) in the baseline condition without fluid in the middle ear space as measured by the cochlear microphonic. Due to the method by which ILD was computed, the ILDs were 0 dB at the midline and symmetrical across azimuth; therefore, data are only shown for one half of the azimuthal plane. Symbols and error bars indicate the across-animal mean ILD (n = 13 ears) and ±1 SD, respectively.

FIG. 5

Baseline interaural time differences measured from cochlear microphonic measurements in the baseline condition without any fluid in the middle ear space for eight different frequencies (0.25–12 kHz). Due to the method by which ITDs were computed, the ITDs were 0 μs at the midline and were symmetrical across azimuth; therefore, data are only shown for half of the azimuthal plane. Symbols and error bars indicate the across-animal mean ITDs (n = 13 ears) and ±1 SD, respectively.

FIG. 6

Across-animal mean and individual animal baseline ILDs (thin red = mean across animals, thin gray = individual animals) and experimental (thick red = mean across animals, thin black = individual animals) 1.5 ml of 100 P silicone oil) as a function of sound source azimuth (n = 3 ears). The simulated middle ear effusion due to silicone oil causes an ILD shift of ~30 dB across frequencies with a larger shift evident at higher frequencies both for individual animals and across animals.

FIG. 7

Example cochlear microphonic (CM) waveform in response to a 500-Hz tone pip stimulus presented at 0 ° azimuth. Left panel CM waveform in the baseline condition (without fluid). In the baseline condition, the waveforms overlap due to the assumption of head symmetry, resulting in a 0-μs ITD. Right panel CM waveform after the middle ear is filled with 1.5 ml of 100 P silicone oil. Addition of the silicone oil results in a ~500-μs delay in the CM waveform.

FIG. 8

Simulated middle ear effusion induced delays (in microseconds) in transmission of sound to the inner ear across four frequencies (0.25–2 kHz) for different volumes (0.5, 1.0, 1.5 ml, and bulla full) and viscosities of silicone oil (n = 13 ears); negative value indicate delays in the CM waveforms re: baseline. The induced delays remained nearly consistent for fluid viscosities above 3.5 P (A, 3.5 P; B, 100 P; C, 600 P). There was an effect of fluid volume on sound transmission delay for all three viscosities tested, with significant increases in delay for all fluid volumes >0.5 ml. Symbols and error bars indicate the across-animal mean delay and ±1 SD, respectively.

FIG. 9

Across-animal mean and individual animal baseline (thin red = mean across animals, thin gray = individual animals) and experimental (thick red = mean across animals, thin black = individual animals; 1.5 ml of 100 P silicone oil) interaural time differences for four frequencies (250 Hz to 2 kHz; indicated in upper left of panels) across sound source azimuth (n = 3 ears). The simulated middle ear effusion due to silicone oil causes a shift in the spatial dependence of the ITD cue.

FIG. 10

Compound action potential responses pre- and post-TTX administration to the round window membrane. Post-TTX measurements were taken ~3 h after TTX administration. A CAP amplitudes after TTX administration (red line) are significantly reduced compared to CAP amplitudes before TTX administration (black line). B Normal baseline (closed circles) and TTX-treated baseline (open circles) ITD measurements as a function of azimuth (−90 ° = left side) for a 500-Hz stimulus. Data are only shown for one hemisphere in the azimuthal plane.