Bilateral cochlear implantation in the ferret: a novel animal model for behavioral studies

Abstract

Bilateral cochlear implantation has recently been introduced with the aim of improving both speech perception in background noise and sound localization. Although evidence suggests that binaural perception is possible with two cochlear implants, results in humans are variable. To explore potential contributing factors to these variable outcomes, we have developed a behavioral animal model of bilateral cochlear implantation in a novel species, the ferret. Although ferrets are ideally suited to psychophysical and physiological assessments of binaural hearing, cochlear implantation has not been previously described in this species. This paper describes the techniques of deafening with aminoglycoside administration, surgical implantation of an intracochlear array and chronic intracochlear electrical stimulation with monitoring for electrode integrity and efficacy of stimulation. Experiments have been presented elsewhere to show that the model can be used to study behavioral and electrophysiological measures of binaural hearing in chronically implanted animals. This paper demonstrates that cochlear implantation and chronic intracochlear electrical stimulation are both safe and effective in ferrets, opening up the possibility of using this model to study potential protective effects of bilateral cochlear implantation on the developing central auditory pathway. Since ferrets can be used to assess psychophysical and physiological aspects of hearing along with the structure of the auditory pathway in the same animals, we anticipate that this model will help develop novel neuroprosthetic therapies for use in humans.

Fig. 1

Diagram of an electrode assembly suitable for bilateral implantation in the ferret (all dimensions in mm). The intracochlear portion of the array consists of 7 platinum ring electrodes with an inter-electrode separation of ∼0.4 mm. At the skull fixation point, a wide piece of Dacron mesh was fixed to the lead wire to protect it from the titanium skull fixation clip.

Fig. 2

High resolution micro-focus radiograph in anterior–posterior view of a ferret skull showing bilateral intracochlear arrays inserted to an even depth into the basal turn of both cochleas relative to the round window niche (RWN) marked with a fine wire. The apical electrode in the electrode array is marked with an arrowhead.

Fig. 3

(A) Representative ECAP waveforms from a ferret at the stimulus amplitudes indicated. (B) P1–N1 amplitude plotted as a function of stimulus level for the same recording with best-fitting linear regression (solid black line). P1–N1 amplitudes of less than 30 μV (dashed line) were considered likely to be within the noise floor. Therefore they were excluded from threshold estimation analysis.

Fig. 4

(A) Nucleus ESPrit 3G speech processor (1) attached to a modified Nucleus Cochlear implant CI24RE emulator (2). Pockets were incorporated within the neckline of a detachable ‘backpack’ (3) that was attached to a jacket made from a ferret harness and elasticated tubular bandage (4). (B) Ferrets carried their cochlear implants within the jacket that held the microphone of the left and right speech processor immediately posterior to the ipsilateral pinna and enabled animals to carry on with their normal activities during chronic stimulation.

Fig. 5

(A) ITDs measured from the external ear canals of sixteen adult ferrets (red lines) plotted as a function of lateral angle. Here, a negative lateral angle denotes a position to the animal's left. (B and C) ILDs measured from the external ear canals of adult ferrets (red lines; n = 16) for sounds filtered between 0.75 and 1.5 kHz (B) and 4 and 8 kHz (C) are plotted as a function of lateral angle. The confidence interval for the group mean is represented by the grey shaded area (5–95%). ITDs (D) and ILDs separated by frequency (E and F) measured from the “pockets” of a custom-made jacket worn by three adult ferrets are plotted in blue (mean + s.d.) as a function of lateral angle. For comparison, the confidence intervals for the external ear canal measurements are re-plotted in grey. Mean (+s.d.) ITDs (G) and ILDs separated by frequency (H and I) are shown for the external ear canal measurements in red and for the jacket pockets in blue.

Fig. 6

Sound evoked head and jacket orienting responses. The upper panels show the orienting movements of a ferret performing a free-field sound localization task recorded from an infrared camera (60 frames s⁻¹) using a reflective strip attached to the animal's head. The lower panel shows how the horizontal angle of the animal's head (red circles; mean + s.d.; n = 1110 trials) and jacket (blue circles; mean + s.d.; n = 932 trials) changed after a stimulus was presented from the +90° speaker position.

Fig. 7

To model the effects of head movement on binaural cues, the animal's head orientation (n = 1) was moved within the chamber in increments of 30° from the 0° speaker position towards the +90° speaker position. In each orientation of the animal, (A) ILDs and (B) ITDs were measured from the external ear canals of an adult ferret (red lines) plotted as a function of lateral angle of the sound source. The thickness of the line represents the head position of the animal within the chamber, with decreasing line thickness being associated with head positions further away from the 0° speaker position. (C) ILDs and (D) ITDs measured from microphones positioned within the jacket pockets as a function of lateral speaker position and head position within the chamber. For reference, data recorded from the external ear canals are replotted in gray (C and D).

Fig. 8

Data from figure are replotted to show variation between successive ILD and ITD measurements within the same animal. (A) ITDs and (B) ILDs (mean + s.d.) measured from the external ear canals (red circles) and jacket pockets (blue circles) of an adult ferret positioned four times within the chamber, plotted as a function of lateral angle away from the zero crossing for each recording (F₀).

Fig. 9

Impedance measurements for (i) one representative animal (F0866; 1st and 3rd columns; measurements from the left and right ears in blue and red, respectively), and (ii) averaged across both ears of all four animals (2nd and 4th columns), plotted as a function of intracochlear electrode position (AE1–AE7) and post-operative day. In the bottom right panel, mean impedance measurements (±s.d.) are shown for each electrode position, for all 8 ears (n = 4 animals).

Fig. 10

(A) Mean ECAP thresholds across all measurements (+s.d.) plotted as a function of electrode position, for all 8 ears (n = 4 animals). (B) Mean ECAP thresholds for the left (blue) and right (red) ears are also shown for individual animals as a function of electrode position.

Fig. 11

ECAP thresholds plotted for each electrode position as a function of post-operative day with best-fitting linear regressions. Each circle represents one ear of the four animals.