2.1.

Hearing Loss Animal Model Induced by Gentamicin

Healthy adult Sprague–Dawley rats ($N=11$, 8-week old, 180 to 200 g) with normal hearing were used as subjects. Gentamicin/furosemide-induced hearing loss models were established in eight rats, according to the modified method previously described.¹⁵ Briefly, the animals were treated daily by gentamicin $100\mathrm{mg}/\mathrm{kg}$ intravenously (i.v.) followed 10 min later by furosemide $90\mathrm{mg}/\mathrm{kg}$ i.v. for 2 days. The i.v. injection of ototoxic drugs was performed by inserting a syringe with an Angiocath Plus™ (BD, Franklin Lakes, New Jersey) into the tail vein of the animals with a very slow injection speed of 10 to $20\mu \text{l}/\mathrm{s}$. Forty-eight hours after gentamicin and furosemide treatment, hearing loss was confirmed with click auditory brainstem response (ABR).

Three rats were not included in the hearing loss model, in order to monitor if there was any problem with our hearing loss model and experimental technique. The same experimental technique [hearing measurement and scanning electron microscopy (SEM)] was applied to these rats, but the difference was that gentamicin/furosemide and laser were not given to these control animals (C ears, $n=3$). These control ears were expected to have normal hearing and normal cochlear hair-cell morphology.

2.2.

Auditory Brainstem Response Measurement

Thresholds of ABRs were determined from each ear before treatment (baseline values, day 0), after exposure to the ototoxic drug (day 2), and after the treatment with LLLT (day 12). ABRs were measured with an evoked response signal-processing system (System III, Tucker Davis Technologies, Alachua, Florida). The rats were anaesthetized and placed in a soundproof booth. Following anesthesia, needle electrodes were placed subcutaneously at the vertex (active electrode) and beneath each pinna (reference and ground electrodes, respectively). The click auditory stimuli were delivered through a tube inserted into the ear canal of the rat. Hearing thresholds were determined by assessment of the lowest stimulus level to elicit ABR peaks III at levels from 10 to 90 dB sound pressure level (SPL) in 5-dB steps. One thousand and twenty-four tone presentations were averaged.

2.3.

Laser Irradiation

In this in vivo study, 48 h after treatment with the ototoxic drug, the animals were irradiated by 830-nm diode laser (Hi-tech Optoelectronics, Beijing, China) for 60 min every day for 10 days. Laser irradiation was done only in the right ear ($\mathrm{GM}+\mathrm{L}$ ear, $n=8$), and the left ear (GM ear, $n=8$) served as the control. The total power output of the laser was set as 200 mW. The power of the laser was checked at the distal end of the optic fiber with a SOLO 2 laser power meter (Newport, Irvine, California) and a XLP12-1S-H2-DO detector head (Newport). The laser energy was delivered by inserting the hollow tube-surrounded laser fiber into the external auditory canal with a distance from the tip of the fiber to the surface of tympanic membrane of around 1 mm (Fig. 1). The core fiber of the optic fiber was 62.5 μm, and when the cladding was included, the diameter was 125 μm. Complete laser parameters are presented in Table 1 as recommended by Jenkins and Carroll.¹⁶ The axis of the optic fiber was evaluated through a multidetector computed tomography (CT) scan on a 64-detector scanner (LightSpeed, GE Healthcare, Milwaukee, Wisconsin). Coronal section images were obtained in 0.625-mm slice thicknesses. The optic fiber was positioned so that it was directly aimed at the cochlea. That is, the optic fiber and cochlea was in the same line as in Fig. 1. We have used a quantitatively measurable stereotactic laser aiming device that can consistently place the optic fiber in the external auditory canal. The stereotactic laser aiming system was composed of two measuring scales: a roll plane graduator and a yaw plane ruler. Before conducting the main experiment, we took several CT scans with various roll plane and yaw plane angles to figure out the correct roll plane and yaw plane angle which would aim the optic fiber to the cochlea.

Fig. 1

In vivo laser irradiation of the ear and computed tomography (CT) scan demonstrating the direction of the optic fiber. The gentamicin-treated rats were irradiated with 830-nm diode laser through the external auditory canal. (1) Current output of the laser, (2) optic fiber, (3) ear plug supporting the optic fiber, and (4) timer (a). The positional relationship between optic fiber, tympanic membrane, and cochlea was defined through a CT scan. (1) Tip of the optic fiber, (2) tympanic membrane, (3) ossicle in the middle ear, and (4) cochlea (b).

Table 1

Complete parameters of laser treatment applied in the article.

In order to calculate the intracochlear penetration rate, we have measured the intracochlear laser power with a fresh rat head. After anesthesia, the rat was decapitated, and the head of the rat was sagitally cut in half. After removing the brain and nerve fibers, the cochlea was identified from the medial side (intracranial side) of the skull. In order to measure the intracochlear laser power, a hole (labyrinthotomy) was made in the cochlea from the medial side of the skull. The laser power meter detector head was placed right beside the labyrinthotomy site. Laser was irradiated from the external auditory canal in the same way that we have performed the main experiment. The power that was measured from the labyrinthotomy site (medial side of the skull) was considered as the laser power that had penetrated into the cochlea. We have performed the same penetration rate experiment in five different cochleae. From this pilot experiment, we have found that $6.20\%\pm 1.24\%$ of laser power had penetrated into the cochlea when the optic fiber was placed in the external auditory canal.

2.4.

Endoscopic Photograph of the Tympanic Membrane

In order to evaluate macroscopic changes of the tympanic membrane and the external auditory canal skin, an endoscopic photograph was taken before the laser irradiation and also every day after laser irradiation for 10 days. Coopix 990 (Nikon, Tokyo, Japan) and TL-1 light source (TiabloLaprairie, Quebec, Canada) was attached to a rigid 0-deg endoscope (130-303-100, Xion, Berlin, Germany). The diameter of the endoscope was 2.7 mm, and the working distance was $<10\mathrm{mm}$. The surface integrity and transparency of the tympanic membrane were evaluated. Perforation, inflammation, discharge, hemorrhage, swelling, and thickening of the tympanic membrane were considered a clinically significant adverse tissue reaction. Although not classified as a clinically significant adverse tissue reaction, vascular changes such as telangiectasia were also closely observed.

2.5.

Histologic Evaluation of the Tympanic Membrane and External Auditory Canal Skin

In order to evaluate microscopic changes of the tympanic membrane and external auditory canal skin, a histologic preparation was made after sacrificing the animals. On the last day of the experiment, after ABR recording, intracardiac perfusions were performed with 4% paraformaldehyde under general anesthesia. After perfusions, the deeply anesthetized animals were decapitated, and the bony external auditor canal and the tympanic membrane were harvested in one piece. The specimens were decalcified in RDO solution (Apex Engineering Products Corporation, Aurora, Illinois) and embedded in paraffin for 5-μm serial sections. The sections were stained with hematoxylin-eosin and examined under light microscopy (Olympus, Tokyo, Japan). The histologic findings were evaluated by one pathologist who was blinded to the group ($\mathrm{GM}+\mathrm{L}$ ear or GM ear). The epithelial integrity, cell morphology of the skin and subcutaneous tissue, and presence of inflammatory cells and vascular congestion were closely evaluated.

2.6.

Scanning Electron Microscopy

On the last day of the experiment, after the ABR recording, intracardiac perfusions and decapitations were performed using the same method mentioned above. The cochleae were removed immediately. The isolated cochleae were then decalcified in RDO solution for 20 min, after which the bony capsule was removed, and the lateral wall tissues (spiral ligament and stria vascularis) as well as the membranous structure were separated under a dissecting microscope. The dissected specimens were rinsed with 0.1 M Dulbecco’s phosphate-buffered saline (DPBS) and post-fixed in 1% osmium tetroxide for 15 min. Afterward, the specimens were gently rinsed with 0.1 M DPBS again and dehydrated in graded ethanol solutions. The specimens were transferred to hexamethyldisilazane for 15 min and dried at room temperature overnight. The dried specimens of the organ of Corti were attached to aluminum stubs with aluminum paint, and then coated with platinum-palladium using E-1030 PT-PD target assembly (Hitachi, Tokyo, Japan). The surfaces of the organs of Corti were examined using S-4300 SEM (Hitachi, Tokyo, Japan).

2.7.

Hair-Cell Counting

Quantitative SEM observations of the surface morphology of the organ of Corti were performed by counting retained hair cells from apical to basal turns of the cochleae, based on the method previously described.¹⁷ Different turns of the cochleae were defined according to the percent distance from the apex; 0.0% to 33.3%, 33.3% to 66.6%, and 66.6% and 100.0% from the apex were considered as apical, middle, and basal turns, respectively. Hair cells were counted over a 100-μm longitudinal distance from two separate regions in the representative images that were captured in the central area of each turn. A hair cell was considered absent if the stereociliary bundle was missing.

2.8.

Statistical Analysis

Audiometry and hair-cell counting were performed by a single-blinded investigator. Data were analyzed statistically using the Statistical Package for the Social Sciences (SPSS, Inc., an IBM Company, Chicago, Illinois) software. Wilcoxon signed ranks test was used to compare the hair-cell damage and hearing loss between the experimental groups. Null hypotheses of no difference were rejected if $p$-values were less than 0.05.

3.1.

Location and Direction of the Optic Fiber

The location and direction of the optic fiber were identified by CT scan. As shown in Fig. 1, the laser fiber, tympanic membrane, and cochlea shared a similar axis with each other. This ensured that the inner ear was directly within the irradiation axis. The CT scan also proved that the tip of the optic fiber did not touch the tympanic membrane. The distance from the tip of the optic fiber to the tympanic membrane was approximately 1 mm. The correct angle that can precisely aim the cochlea was $-7.5\mathrm{deg}$ in the roll plane and 14.5 deg in the yaw plane.

3.2.

Endoscopic Findings of the Tympanic Membrane

Until the end of the experiment, all of the tympanic membranes were intact without any clinically significant adverse tissue reaction. That is, there was no evidence of perforation, inflammation, discharge, hemorrhage, swelling, or thickening of the tympanic membrane. But after daily LLLT, the vessels on the tympanic membrane and external auditory canal started to become hyperemic. The vascular congestion was noticed from the first to third days of LLLT and kept on progressing until the end of the experiment. Meanwhile, there was no change in the vasculature of the nonirradiated left ear (Fig. 2).

Fig. 2

Endoscopic findings of the tympanic membrane. The tympanic membrane was photographed every day after laser irradiation with an endoscope. Throughout the experiment, there was no evidence of perforation, inflammation, discharge, hemorrhage, swelling, or thickening of the tympanic membrane. But after daily laser irradiation, the vessels in the tympanic membrane and external auditory canal started to become hyperemic. The vascular congestion was noticed from the first to third days of laser irradiation and progressed until the end of the experiment. Meanwhile, there was no change in the vasculature of the nonirradiated left ear.

3.3.

Histologic Evaluation of the Tympanic Membrane and External Auditory Canal Skin

After 10 days of LLLT, there was no significant adverse tissue reaction when the tissues were evaluated under the light microscope (Fig. 3). That is, there was no clinically significant difference in the epithelial integrity of the skin between the $\mathrm{GM}+\mathrm{L}$ and GM ears. The morphologies of the cells in the epithelium and subcutaneous tissue were not different. There was no evidence of inflammation or infection in both the $\mathrm{GM}+\mathrm{L}$ and GM ears. When the experimental groups were blinded, the pathologist was not able to distinguish between the $\mathrm{GM}+\mathrm{L}$ and GM ears based on the histologic findings.

Fig. 3

Histologic evaluation of the tympanic membrane and skin of the external auditory canal. After 10 days of low-level laser treatment, there was no significant adverse tissue reaction when the tissue was evaluated under the light microscope. That is, there was no clinically significant difference in the epithelial integrity of the skin between the GM ears [(a) low power and (c) high power] and $\mathrm{GM}+\mathrm{L}$ ears [(b) low power and (d) high power]. Although there was no clinically significant histologic difference between the GM and $\mathrm{GM}+\mathrm{L}$ ears, a subtle difference was noticed. The tympanic membrane seemed slightly thicker in the $\mathrm{GM}+\mathrm{L}$ ears (b) when compared with the GM ears (a). Also, the vessels in the $\mathrm{GM}+\mathrm{L}$ ears seemed slightly hyperemic (d) when compared with those of the GM ears (c).

Although there was no clinically significant histologic difference between the $\mathrm{GM}+\mathrm{L}$ and GM ears, a subtle difference was noticed. The tympanic membrane seemed slightly thicker in the $\mathrm{GM}+\mathrm{L}$ ears when compared with the GM ears. Also, the vessels in the $\mathrm{GM}+\mathrm{L}$ ears seemed slightly hyperemic when compared with the GM ears. But these findings were not substantial and were not noticed in every specimen of the $\mathrm{GM}+\mathrm{L}$ ears. The pathologist noticed these findings, but considered it a normal variation. In general, it seemed that there was no difference between the $\mathrm{GM}+\mathrm{L}$ and GM ears.

3.4.

SEM Hair-Cell Count

The hair cells were observed to be absent, and the number of remaining hair cells was calculated in a 100-μm length of each turn (apical, middle, and basal turns) in the SEM images (Fig. 4). The hair-cell morphology was normal in the C ears; there was no absent hair cell. As for the GM ears, the hair cells in the basal turn were mostly damaged ($10.0\pm 4.8\mathrm{cells}/100\mu \text{m}$), while the hair cells were partially conserved in the middle turn ($26.7\pm 21.2\mathrm{cells}/100\mu \text{m}$) and relatively intact in the apical turn ($60.4\pm 23.5\mathrm{cells}/100\mu \text{m}$). This gradual severity of hair-cell loss was similar to that of the previous reports, which also used gentamicin to induce hearing loss. In the $\mathrm{GM}+\mathrm{L}$ ears, a similar finding was found as in the GM ears. That is, most of the hair cells in the basal turn were damaged ($13.6\pm 7.0\mathrm{cells}/100\mu \text{m}$), while the hair cells were partially conserved in the middle turn ($49.7\pm 24.0\mathrm{cells}/100\mu \text{m}$) and relatively intact in the apical turn ($66.1\pm 5.9\mathrm{cells}/100\mu \text{m}$). But the number of hair cells was significantly larger in the $\mathrm{GM}+\mathrm{L}$ ears in the middle ($p=0.028$) and basal turns ($p=0.042$) when compared with that of the GM ears. There was no difference in the number of hair cells in the apical turn between the two groups. When the number of hair cells was summated for all three turns [Fig. 5(a)], it was $97.1\pm 40.9\mathrm{cells}/300\mu \text{m}$ in the GM ears, $129.4\pm 34.3\mathrm{cells}/300\mu \text{m}$ in the $\mathrm{GM}+\mathrm{L}$ ears, and $197.0\pm 12.5\mathrm{cells}/300\mu \text{m}$ in the C ears. The number of hair cells was significantly larger in the $\mathrm{GM}+\mathrm{L}$ ears when compared with that of the GM ears ($p=0.034$).

Fig. 4

Scanning electron microscopy (SEM) and quantitative hair-cell count results. The hair-cell morphology was normal in the C ears: there were no absent hair cells. As for the GM and the $\mathrm{GM}+\mathrm{L}$ ears, most of the hair cells in the basal turn were damaged, while the hair cells were partially conserved in the middle turn and relatively intact in the apical turn. But, the number of hair cells was significantly larger in the $\mathrm{GM}+\mathrm{L}$ ears in the middle turn ($p=0.028$) and basal turn ($p=0.042$) when compared with that of the GM ears. $\text{Scale bar}=20\mu \text{m}$

Fig. 5

Total hair-cell count and hearing outcome. When the number of hair cells was summated for all the three turns (a), the total number of hair cells was significantly larger in the $\mathrm{GM}+\mathrm{L}$ ears when compared with that of the GM ears ($p=0.034$). As for hearing outcome (b) before gentamicin/furosemide treatment, the hearing threshold was normal in all the animals ($<30\mathrm{dB}$ SPL). After administration of gentamicin/furosemide (day 2), the hearing threshold significantly increased; After daily irradiation treatments for 10 days (day 12), the hearing was significantly better in the $\mathrm{GM}+\mathrm{L}$ ears when compared with that of the GM ears ($p=0.023$).

3.5.

Hearing Outcome

Before gentamicin/furosemide treatment, the hearing threshold was normal in all the animals ($<30\mathrm{dB}$ SPL). After administering gentamicin/furosemide (day 2), the hearing threshold increased to $54.3\pm 11.0\mathrm{dB}$ SPL in the GM ears and $52.9\pm 12.5\mathrm{dB}$ SPL in the $\mathrm{GM}+\mathrm{L}$ ears [Fig. 5(b)]. There was no difference in the degree of hearing loss before starting the LLLT. After daily irradiation treatments for 10 days (day 12), the hearing threshold in the $\mathrm{GM}+\mathrm{L}$ ears improved to $44.3\pm 12.7\mathrm{dB}$ SPL. But, the hearing threshold in the GM ears did not improve ($57.1\pm 18.0\mathrm{dB}$ SPL). The hearing was significantly better in the $\mathrm{GM}+\mathrm{L}$ ears when compared with that of the GM ears ($p=0.023$).