Introduction

For the physiological process of hearing, sound pressure waves from the environment travel through the outer ear and the middle ear to the fluid filled chambers of the inner ear (the cochlea). The cochlea is a spiraled, hollow, conical bonny chamber, divided into three chambers. The scala vestibuli (SV) and scala media (SM) are divided by the Reissner membrane and SM from scala tympani (ST) by the basilar membrane (BM). One main characteristic of the cochlea is being a tuned organ so that high frequencies are represented at the base and low frequencies are represented within the apical turns. The hair cells, the sensory cells of the peripheral hearing organ, reside within the organ of Corti on the BM. The sound induced fluid vibrations are transmitted to the BM and induce the hair cell depolarization through the so called mechano-electrical transduction. These action potentials generated within the hair cells are then transmitted through the auditory nerve to the central auditory system [1].

An estimated 278 million people worldwide are living with disabling hearing impairment (at least moderate hearing loss in the better hearing ear) and this number is rising mainly due to a growing global population and longer life expectancies [2]. A large proportion of them show cochlear dysfunction due to outer hair cells dysfunction or loss and - in cases of profound hearing impairment, inner hair cells loss too. Outer hair cells have a filter and modulation function, which is necessary for frequency specific sound perception [3]. Traditionally, electrical neural stimulation has been used to bypass the nonfunctional peripheral sensory organ, the cochlea, to reasonably restore some auditory function. However, there is only incomplete compensation for these functional failures by conventional hearing aids and cochlear implants for speech perception in noisy environments and for more complex sounds of daily life. This incomplete compensation has been attributed to the lack of localized sensorineural activation across different frequency regions with these devices. Therefore, patients who have lost their outer hair cells but still have functional inner hair cells might benefit from the specific stimulation of the remaining inner hair cells and achieve, through this, a significant improvement in their hearing performance. We thought that this selective activation of the residual functional inner hair cells may be achieved through laser induced controlled vibration of the BM since laser light can be better focused in a clear watery environment when compared to electrical current.

Optoacoustic waves can be produced inside a medium when a short laser pulse is absorbed [4]. This is the result of transient thermal expansion due to laser heating the focused region, according to thermoelastic mechanism. For an optimal signal generation the so-called stress-confinement has to be fulfilled, ${\tau }_l\cdot {\mu }_a\cdot c_0\ll 1$ wherein τ_L is the pulse duration of a single pulse, µ_a is the optical absorption coefficient of the irradiated material, and c₀ is the local speed of sound, meaning that the laser pulse duration has to be short compared to the time the acoustic signal needs to propagate through the optical penetration depth at the speed of sound. In this case, no energy dissipation will occur during generation of the acoustic signal. In our experiments, we exploited the potential application of these optoacoustic waves to be used as a noncontact force that induces BM vibrations within the cochlea. For this, we applied laser pulses through an optical fiber onto the opened inner ear of an animal model. We used a microscope assisted Laser Doppler Vibrometer to prove, record and characterize the optoacoustic induced BM vibrations and their dependence on laser pulse energy.

Materials and methods

Pigmented guinea pigs (n = 6) (Charles River Laboratories, Solingen, Germany) of either sex (400 to 600 g) were used for our study according to the guidelines of the Animal care and use committee of the Medical University of Hannover and Lower Saxony. The animals were sacrificed under general anesthesia [5] and their cochleae were excised from the tympanic bulla. To minimize swelling and deformation of the cochlear tissues, the cochleae were placed into a Petri dish filled with artificial perilymph (solution of in mM: 145 NaCl, 2.7 KCl, 2 MgSO4, 1.2 CaCl2, 5 HEPES, pH: 7.4 osmolarity: 288 mOsm). The cochleae were then dissected in such a way that the BM of the basal turn was exposed, leaving the bony structures needed for anchoring in place (Fig. 1 ). A Nd:YAG laser (Quantel Brilliant BW, France) was used, with wavelengths at 532nm and 355 nm respectively (pulse duration 10 ns and repetition rate 10 Hz) e.g. controlled to emit up to 30 µJ/pulse.

 

Fig. 1 Schematic illustration of the experimental setup. (a) Diagram of a section through the organ of Corti. Three rows of outer hair cells and one row of inner hair cells reside on the basilar membrane. The basilar membrane is fixed at the osseous spiral lamina (OSL) medially and the spiral ligament laterally. Auditory nerve fibers (having the first neuron in the spiral ganglion) connect the hair cells with the brainstem. (b) Diagram of the microscopic view, from the scala tympani onto the exposed BM of one segment of the cochlea. The LDV is focused on the glass bead to measure the BM vibrations. Position S1, S2 and S3 represent laser stimulation sites on the OSL while the LDV measurements were done at a fixed position on the BM using the enhanced reflectance of a glass bead.

Download Full Size | PDF

Laser pulses were presented through a fiber (100 µm core diameter), directed to the osseous spiral lamina (OSL) at an angle of approximately 50 degrees (Fig. 1(a)). The laser beam on the target was 200 µm in diameter along its short axis, and 300 µm along its long axis. Three laser stimulation sites are shown on the OSL (Fig. 1(b)). A Laser Doppler Vibrometer (LDV) (Polytec, OFV 551, Germany) was used to record the BM vibrations. It has been reported that BM reflectivity is extremely low, being only around 2.1 × 10⁻⁵ [6]. In order to increase the reflection of LDV light, a silver coated glass bead of 30 µm in diameter (Microsphere Technology Ltd, UK) was placed on the BM. The density of the glass bead is 2 g/cm³, which is only two times the water density and therefore, its mass effect on the BM can be neglected. A weak CW red laser from the LDV head was focused on the glass bead using the same microscope objective which was used to collect the bead scattered red laser light from velocity measurement. In order to enhance the signal-to-noise ratio, the recorded signals of 512 trials were averaged. All measurements were done within 4 hours after the excision of the cochleae.

Results

We applied the laser pulses first on the OSL at the site S1 (Fig. 1) whose center is about 250 µm from the glass bead on which we performed the LDV recordings.

The BM vibrations that we recorded consecutive to the laser irradiation induced two types of vibrations: a fast vibration (FV) and a slow vibration (SV). The FV, having a frequency around 100 kHz, started with the first positive peak followed by the first negative peak after laser stimulation (see arrow in Fig. 2 ). This pattern reflects the character of the optoacoustic waves as described by Wang et al. [7]. We assumed that the OSL absorbs the laser energy, resulting in optoacoustic waves which diffuse away from the source. A certain amount of the energy and momentum of the OSL would induce the BM vibrations that could be recorded by the LDV [8].

 

Fig. 2 An example of BM vibration following the 532 nm laser-pulse application at position S1 (as marked in Fig. 1) within the basal turn of a guinea pig cochlea. The arrow represents t = 0, the start of the laser pulse.

Download Full Size | PDF

The SV following the FV had a frequency of about 20 kHz for recordings that we performed in basal turn areas, corresponding to the resonant frequencies represented in this cochlear segment. After the optoacoustic wave stimulation, the SV of the BM became weaker and weaker most probably due to the damping of the liquid environment surrounding it [9]. The measured vibration lasted more than 7 cycles in 380 μs, disappearing into the noise level afterwards.

The magnitude of the BM vibrations depended on the applied laser pulse energy (Fig. 3 ), suggesting that optoacoustic induced BM vibrations could be modulated by adjusting the applied laser-pulse energy.

 

Fig. 3 Input/output functions of the laser pulse energy applied at the site S1, versus the FV magnitude for (a) 532 nm laser pulses as well as for (b) 355 nm laser pulses. The results fit to a square-root curve (solid lines).

Download Full Size | PDF

The experimental data were well fitted by a square-root function:$A(E)=A_0\sqrt{E-E_{th}}$, wherein A is the magnitude, A₀ is a constant, E is the laser pulse energy, and E_th is the minimal laser energy required for the excitation of the BM vibrations. From the fittings, we obtained the threshold of 370 nJ per pulse for the 355 nm laser and 4.2 μJ per pulse for the 532 nm laser. The difference between these two thresholds indicated that 355 nm laser was more absorbed than 532 nm laser. Higher absorption induced stronger optoacoustic waves which in turn induced the increase in vibration magnitudes. Since the OSL is a bony structure, our results are in accordance with the results from Bahari et al. who studied the absorption spectrum of tibia bone for a wide light source from UV to IR, observing a strong absorption in the 200 – 400 nm spectrum [10].

As mentioned in the introduction, the cochlea acts as a spatial frequency analyzer which exhibits resonant vibrations at characteristic frequencies that vary with position along the BM, giving rise to a topographical mapping of auditory frequency. The main advantage of optical-compared to electrical stimulation being the localized sensorineural activation across different frequency regions would require localized vibrations of the BM. To assess if localized activation could be achieved with our system, we moved the optoacoustic source away from the original stimulation site S1 by moving the laser pulse on the OSL in the longitudinal direction along the BM. We then measured the vibration on the BM where the glass bead was located (Fig. 1(b)) and analyzed the correlation between the vibration magnitude and the distance of the optoacoustic source to S1. At S1, which is the closest site to the BM where vibration was measured, the laser induced the largest BM vibration magnitude (Fig. 4 ). The vibration magnitude decreased quickly when we moved the laser away from S1 due to the attenuation of the optoacoustic waves. When the laser was at 1mm distance from S1, a decrease down to 50% of the original magnitude on S1 was observed in one cochlea. In two other cochleae, the magnitude decreased down to 10% of the original magnitude on S1 at 1 mm distance from S1. Therefore, the vibrations within the cochlea induced by our optoacoustic source were dependent on the location of the optoacoustic source, meaning a selective stimulation of the BM.

 

Fig. 4 Localization measurements of the optoacoustic stimulation on the BM. The y-axis represents the normalized vibration magnitude and the x-axis represents the distance of the stimulation site to the original site S1. Data from three cochleae is represented by different markers and thin lines. Each data point represents the averaged value of 3 measurements within the same cochlea at each distance with error bars. Bold solid line: average of the animals.

Download Full Size | PDF

Discussion

Izzo et al. reported the application of pulsed IR laser on spiral ganglion cells within the cochlea [3,11, and 12,]. Laser absorption is supposed to induce a transient increase in the temperature of the irradiated spiral ganglion cells, the first neuron on the auditory pathway. Temperature sensitive ion channels in their cellular membrane would then be activated inducing a neural action potential which is then transferred to the central auditory system. Therefore, the excitation of spiral ganglion cells generating auditory nerve impulses is most probably based on a photo-thermal ion channel activation that we could define as photo-thermo-electrical effect. However, our target patients group is different from that of Izzo et al, and for this reason we propose a different stimulation method. Our goal is to find a stimulation strategy for patients with mid to highly sensorineural hearing impairment having residual hearing due to still functional inner hair cells but non-functional or missing outer hair cells. Our final goal is to mimic the natural hearing process by stimulating these remaining functional inner hair cells using optoacoustic waves. The optoacoustic induced BM vibrations would induce the depolarization of the hair cells through a mechano-electrical effect (1), which differs from the thermo-electric effect induced by infrared laser pulses described by Izzo et al. The experiment proving that this stimulation method can induce action potentials that are transmitted to the central nervous system has been previously reported by our group [13]. The auditory brainstem responses, as a result of optoacoustic cochlear activation, (OABR) were similar to ones induced by sound stimulation (ABR) [13]. The experiments reported herein however are proving the principle behind the mechanism underlying the activation of the cochlea using green laser light, in vivo. Therefore, both LDV and ABR measurements support the hypothesis that green laser light pulses could be used for a novel intra-cochlear, optical auditory prosthesis.

These experiments have been performed using a fiber with 100 µm core diameter directed to the OSL at an angle of approximately 50 degrees. We estimate that the laser irradiation area have been about 300 μm × 200 μm in diameter. However, the future laser beam applications are going to be perpendicularly to the target and using a thinner fiber (e.g., a fiber with a core diameter 50 μm). For these improved irradiation conditions we expect that the irradiation diameter could be reduced to less than 100 μm x 100 μm. This would determine a 70% decrease in BM vibration magnitude at a distance of 0.5 mm from the optoacoustic source and therefore an increase in vibration localization and frequency specific activation of the cochlea. Specially designed fiber endings eventually with attached miniaturized lenses might be considered as well to increase even more the focusing capability of the light and through this inducing even more localized vibrations of the basilar membrane.

In order to apply the laser pulses at different levels within the cochlea, the insertion of a fiber bundle within the ST will be necessary. Our preliminary study [14] demonstrates that the insertion depth and force of a silicone coated fiber bundle (8 fibers with a core diameter of 20 μm) were comparable to what has been achieved with conventional electrode arrays of cochlear implant that are currently implanted in patients. These findings are encouraging as to the safety of fiber implantation within the cochlea and the ability to reduce insertion forces through different types of fiber coatings and variations of the fiber diameter.

For the generation of the optoacoustic waves, part of the laser energy was converted into the optoacoustic energy. The characteristics of the induced optoacoustic waves depended on the laser pulse length, beam size, as well as on the optical and thermo-physical properties of the irradiated tissue. During and immediately after the generation of the optoacoustic waves, the optoacoustic waves propagated away from their origin, changing their original characteristics (e.g., wave pattern and energy) due to attenuation, dispersion, and diffraction within the tissue. The pattern in time domain of the optoacoustic waves reaching the BM became broader as demonstrated by the FV of the BM vibrations. The physical mechanism behind the square-root relation in Fig. 3 between the vibration magnitude and the laser energy is not clear yet. The threshold difference between 355 nm and 532 nm laser suggests a difference in the laser absorption of these two wavelengths within the irradiated tissue. Further experiments investigating the optical properties of the different cochlear structures (e.g. BM, OSL) in a large wavelength range and optimizing the optoacoustic wave generation within the cochlea, are currently being performed in our lab.

Conclusions

Pulsed laser irradiation induces optoacoustic waves within the cochlea and localized vibrations of the BM in close proximity to the optoacoustic wave’s source. The vibration magnitude can be modulated by adjusting the laser pulse energy. We propose that this cochlear stimulation method could be used for a novel optical cochlear implant device, with improved frequency specific cochlear activation and consequently improved speech perception for mid- to highly hearing impaired patients with residual hearing.