Original research paper

Development of a safe dexamethasoneeluting electrode array for cochlear implantation Dimitra Stathopoulos 1,2 , Scott Chambers 1, Ya Lang Enke 3, Godofredo Timbol 3, Frank Risi3, Christopher Miller3, Robert Cowan 1, Carrie Newbold 1,2 1

The HEARing CRC, Carlton, Victoria, Australia, 2Department of Otolaryngology, University of Melbourne, East Melbourne, Victoria, Australia, 3Cochlear Limited, Macquarie University, New South Wales, Australia Objectives: Cochlear implantation can result in trauma leading to increased tissue response and loss of residual hearing. A single intratympanic application of the corticosteroid dexamethasone is sometimes used clinically during surgery to combat the potential effect of trauma on residual hearing. This project looked at the safety and efficacy of dexamethasone eluted from an intracochlear array in vivo. Methods: Three trials were conducted using normal hearing adult guinea pigs implanted with successive iterations of dexamethasone-eluting (DX1, DX2, and DX3) or non-eluting (control) intracochlear electrode arrays. The experimental period for each animal was 90 days during which hearing tests were performed at multiple time points. Results: There was no significant difference between matched control array and dexamethasone array groups in terms of spiral ganglion neuron density, organ of Corti condition, or fibrosis and ossification. A cochleostomy seal was present in all implanted cochleae. There were no differences in the degree of hearing threshold shifts between DX1 and DX3 and their respective control arrays. Cochleae implanted with DX2 arrays showed less hearing loss and marginally better spiral ganglion neuron survival than their control array counterparts. Post-explant inspection of the DX2 and DX3 arrays revealed a difference in pore density following dexamethasone elution. Conclusion: The dexamethasone doses used were safe in the guinea pig cochlea. Dexamethasone did not inhibit formation of a cochleostomy seal. The level of hearing protection afforded by dexamethasone eluting from an intracochlear array may depend upon the degree of elution and level of trauma inflicted. Keywords: Cochlear implantation, Dexamethasone, Corticosteroid, Drug delivery, Guinea pig, Hearing loss

Introduction In this study, we assess the biosafety and efficacy of a dexamethasone-eluting electrode array implanted into the guinea pig cochlea. Initially, cochlear implants were considered only for patients with total or profound hearing loss. Over time, implant design and sound coding have markedly improved recipients’ speech perception results (Davidson 2006; Leigh et al., 2011), thus broadening the criteria for implantation. Patients with lowfrequency residual hearing – a group previously lost in a diagnostic limbo with hearing loss too severe to benefit from hearing aids but too much residual hearing to risk surgical damage – can now benefit from a hybrid treatment using an implant in Correspondence to: Dimitra Stathopoulos, Department of Otolaryngology, 32 Gisborne Street, East Melbourne, Victoria 3002, Australia. Email: [email protected]

254

© W. S. Maney & Son Ltd 2014 DOI 10.1179/1754762813Y.0000000054

conjunction with a hearing aid in the same ear (Simpson et al., 2009). However, as with all invasive surgery, the risk of trauma due to the physical procedure remains – e.g. displacement of the basilar membrane; fracture or dislocation of the osseous spiral lamina; less commonly, damage to the lateral cochlear wall; and rupture of the basilar membrane with the electrode array entering the scala vestibuli (Nadol and Eddington, 2006). Post-surgical damage can be caused by the immune response to the implant as a foreign body and can also be caused by subsequent infection (Kiefer et al., 2004). The innate immune response forms a collagenous fibrous tissue sheath isolating the entire implant from the body. New bone growth occurs at and around the cochleostomy site and can also occur along the implant track to surround the electrode array (Nadol and Eddington, 2006). These traumas and subsequent

Cochlear Implants International

2014

VOL.

15

NO.

5

Stathopoulos et al.

responses pose a risk to residual hearing by damaging cochlear structures and cells. Preserving usable residual hearing remains a priority for surgeons, clinicians, and patients. Along with soft surgery technique (Briggs et al., 2001; Culler et al., 1943; Friedland and Runge-Samuelson, 2009) and changes in implant design (Gantz et al., 2005), surgeons have used lubricants and anti-inflammatory steroids – such as dexamethasone – to reduce the potential impact of surgery and protect the cells and structures in the cochlea (Clark et al., 1995; Paasche et al., 2006, 2009). Surgeons first used a single systemic dose of antiinflammatory steroid to dampen the innate immune response, thereby potentially reducing the amount of fibrous tissue that may form around the electrode array. Their intent was to prevent or minimize the rise in electrical impedance seen clinically 2–3 weeks after implantation, and to maintain low-electrical impedance. This scheme was investigated and lower impedances were preserved at least 40 years after implantation (Paasche et al., 2006, 2009). Few animal studies have investigated the link between steroid application and impedance (Huang et al., 2007), with most studies assessing the effect of steroids on preserving residual hearing (Connolly et al., 2010; Eastwood et al., 2010; James et al., 2005; Lee et al., 2013; Maini et al., 2009). The latter studies all used a single application of dexamethasone either via local application on the round window at the time of surgery or via systemic delivery prior to surgery. While a single dose of dexamethasone might be sufficient to reduce inflammation at the time of surgery, sustained application may better mitigate delayed trauma resulting from the innate immune response. Long-term delivery via a mini-osmotic pump maintained lower auditory brainstem response (ABR) thresholds (Eshraghi et al., 2007; Vivero et al., 2008) and showed dexamethasone to be safe over the 4week delivery period. A drug-eluting electrode array is a clinically feasible solution for simple and efficient long-term drug delivery directly into the cochlea (Eshraghi et al., 2011). Dexamethasone-eluting implants have shown promising results for hearing protection over a period of 3 months (Braun et al., 2011), but no significant effect on tissue response. We investigated the effects of three prototype dexamethasone-eluting arrays implanted in the guinea pig cochlea for 3 months.

Materials and methods This study was approved by the Animal Research and Ethics Committee of the Royal Victorian Eye and Ear Hospital under Project numbers 07/151A, 08/170A, and 10/208AR, and was conducted in strict

Safety and efficacy of dexamethasone eluted from an intracochlear array

accordance with the Australian code of practice for the care and use of animals for scientific purposes (2004).

Trial experimental designs We performed three successive animal trials of dexamethasone-eluting electrode arrays, using normal hearing, tri-colour Dunkin-Hartley guinea pigs weighing 400–800 g. Details of groups and sample sizes are outlined in Fig. 1. The results of each trial led to changes and improvements in electrode array design for the succeeding trial.

Electrode arrays All electrode arrays comprised eight platinum ring electrodes embedded within a silicone rubber elastomer (Z60270, Cochlear Ltd, Sydney, Australia). A propriety form of micronized dexamethasone acetate (Cochlear Ltd; particle diameter less than 5 μm) was manually painted by an experienced technician to a subset of electrode arrays in each trial. Two different techniques were used to add dexamethasone from the tip of the array to the fifth electrode. The amount of dexamethasone per array was estimated from the total volume of coating applied. In Trial I, silicone rubber mixed with 20% w/w micronized dexamethasone acetate was painted onto electrode arrays to form a coating approximately 100 μm thick. Additional micronized dexamethasone was then ‘dusted’ onto the outer surface of the arrays before the final cure. These arrays, designated DX1, were estimated to contain 100 μg of dexamethasone. In Trial II, electrode arrays were again painted with silicone rubber mixed with 20% w/w micronized dexamethasone acetate, but no additional dexamethasone was dusted onto the surface before the final cure. These arrays, designated DX2, were estimated to contain 74 μg of dexamethasone. A second dexamethasone-eluting array type with a commercially manufactured coating was also implanted. As portions of the coating were found to have come away from the surface of the array, we deemed it unacceptable and no further analyses were performed. Due to the commercial sensitivity of this coating, this array is not described in this paper but is mentioned here to clarify the group numbers used in this trial. In Trial III, electrode arrays were painted with silicone rubber mixed with 24% w/w micronized dexamethasone acetate. These arrays, designated DX3, were estimated to contain 89 μg of dexamethasone. The remainder of the electrode arrays in each trial had no dexamethasone incorporated into the substrate silicone rubber; these were designated control arrays. All electrode arrays in Trial I were sterilized using a humidified hydrogen peroxide process (STERRAD).

Cochlear Implants International

2014

VOL.

15

NO.

5

255

Stathopoulos et al.

Safety and efficacy of dexamethasone eluted from an intracochlear array

Figure 1 Trial organization. Trial I: DX1 array (n = 6), control array (n = 6), unimplanted (n = 12). Trial II: DX2 array (n = 6), control array (n = 6), unimplanted (n = 6), X dexamethasone-eluting array not described in paper (n = 6). Trial III: DX3 array (n = 12), control array (n = 12).

All electrode arrays in Trials II and III were sterilized with ethylene oxide.

Auditory brainstem response recording Animals were inducted using ketamine (Ilium-Ketamil, TroyLab, Sydney, Australia) (60 mg/kg) mixed with xylazine (Ilium-Xylazil, TroyLab) (4 mg/kg) (K/X) injected intramuscularly (IM). We recorded a click-evoked ABR for each ear as detailed previously (Xu et al., 1997). We used three 23-gauge needles subcutaneously (SC) as positive, negative, and ground electrodes. These were positioned, respectively, on top of the head, ∼5 mm rostral to the end of the pinnae; the neck, ∼25 mm caudal from the positive needle; the left thorax. The contralateral ear was blocked with ear mould compound (Otoform® K/c, Dreve, Germany). Acoustic stimuli were delivered with a Richard Allen DT-20 speaker placed 10 cm from the base of the ipsilateral pinna. The return signal was amplified by 105 and passed through a Krohn-hite 3750R band-pass filter, high pass at 150 Hz, 24 dB/octave, and low pass at 3 kHz, 6 dB/octave. The click-evoked ABR was followed by frequencyspecific tone-pip ABRs tested at 2, 8, 16, 24, and 32 kHz. All animals included in the experiment had normal hearing, i.e. a click-evoked threshold above 40 dBA. Click- and frequency-specific ABRs were repeated immediately after surgery, 30 days after surgery, and 90 days after surgery. ABRs were analysed using Igor Pro (WaveMetrics Inc., Oregon, USA) with a custom ABR analysis package (Dr James Fallon, Bionics Institute).

Surgery At the completion of the pre-surgery ABR, guinea pigs were injected with atropine sulphate at 0.06 mg/kg IM, enrofloxacin (Baytril®, Bayer, Leverkusen, Germany) at 10 mg/kg SC, Carprofen (Rimadyl®,

256

Cochlear Implants International

2014

VOL.

15

NO.

5

Pfizer, New York, USA) at 4 mg/kg, diluted 1 in 10 with 0.9% saline SC. Local anaesthetic, 2% Lignocaine hydrochloride (Ilium-Lignocaine, TroyLab), was injected SC along the incision site. Anaesthesia was maintained with additional injections of ketamine (40 mg/kg) and xylazine (4 mg/kg) as required. The mastoid bulla was exposed using a post-auricular approach through the platysma and splenius muscles. A ∼3 mm diameter bullostomy was drilled in the mastoid bulla to expose the round window and basal turn of the cochlea. A cochleostomy was carefully drilled into the basal turn using a 0.6 mm diamond burr. Excess bone dust and blood were cleared to allow a view of the cochlear endosteum, which was then ruptured. An array was inserted until resistance was encountered. The cochleostomy was sealed with fascia and the protruding length of the array super-glued to the edge of the bullostomy. The bullostomy was sealed with carboxylate cement (Durelon 3M ESPE). The muscle layers were sutured with absorbable monofilament (Ethicon PDS II, Johnson & Johnson, New Jersey, USA) and the external wound closed with stainless steel staples (Leukoclip SD, Smith + Nephew, England). An ABR was performed immediately after surgery, after which the animal was sent to recovery.

Histology After 90 days implantation, animals were anaesthetized and underwent a final ABR. While still under anaesthesia, animals were given a lethal dose of sodium-pentobarbital (Lethabarb, Virbac Animal Health, Australia) intraperitoneally; then, intracardially perfused with 500 ml of heparinized 0.9% saline at 37°C and 500 ml of 10% neutral buffered formalin at 4°C. The temporal bones were exposed and bullae were collected. The implanted bullae were opened with

Stathopoulos et al.

rongeurs to confirm the position of the electrode array in situ prior to explant. Cochleae were decalcified in 10% ethylenediaminetetraacetic acid in 10% neutral buffered formalin over 2 weeks. They were then trimmed of non-essential tissue, embedded in Spurr’s resin, and oriented for sectioning on a mid-modiolar plane. Sections were collected at 125 μm intervals and cut to a thickness of 2 μm. We measured the area of Rosenthal’s canal at each turn of the cochlea with NIH Scion Image (developed

Safety and efficacy of dexamethasone eluted from an intracochlear array

Table 1

Rating system used for organ of Corti assessment

Organ of Corti rating system 5 4 3 2 1

Most structure present, inner hair cell and outer hair cells visible, stereocilia visible Supporting cells and tunnel visible, some hair cells visible Tunnel visible, indeterminate cells visible, no hair cells Indeterminate cells visible on basilar membrane, no recognizable structure No cells visible on basilar membrane

at the US National Institutes of Health and available at http://rsb.info.nih.gov/nih-image/). Spiral ganglion neurons were counted and the density of the cells was calculated. We assessed the organ of Corti using a qualitative five-point grading scale based on morphology (Fig. 2, Table 1). Fibrous tissue and ossification were analysed using the NIH Scion image program. We measured the total area of fibrous tissue and new bone within the scala tympani for each section. We calculated the average area of fibrous tissue and new bone at the cochleostomy alone, and across the entire cochlea.

Statistical analysis Analyses were conducted using either Minitab 16 or Sigma Plot 12.0. Tests used were two-sample t-test, general linear model, repeated-measure analysis of variance, and regression.

Results Two animals in Trial II and three animals in Trial III died during surgery due to suspected myocardial infarction. Additional animals were used to replace those lost, keeping group numbers even. All other animals recovered well from surgery. We did not observe adverse effects related to neither repeated bouts of anaesthesia, nor any signs of post-surgical infection or inflammation during the 3-month implantation period.

Trial I

Figure 2 Examples for organ of Corti morphology used to determine scores as described in Table 1. Numbers on each photomicrograph pertain to the associated score.

Hearing thresholds for all groups dropped (i.e. hearing was lost) immediately after surgery. For the two implanted groups – control array and DX1 array – high-frequency hearing (16, 24, and 32 kHz) recovered after 3 months while low-frequency hearing (2 and 8 kHz) showed little change. Overall, we saw no difference in threshold shifts between the DX1 and control array at any time point, and only low levels of hearing loss were recorded for both groups. Mean threshold shifts for 2 and 32 kHz are shown in Fig. 3A. We recorded most fibrous tissue and new bone at the cochleostomy site, external to the basal scala tympani on the outer surface of the cochlea. In the scala tympani, fibrous tissue was mainly limited to a thin

Cochlear Implants International

2014

VOL.

15

NO.

5

257

Stathopoulos et al.

Safety and efficacy of dexamethasone eluted from an intracochlear array

Figure 3 Trial I results: (A) 2 and 32 kHz ABR threshold shifts, (B) tissue growth averages for the whole cochlea and cochleostomy alone, and (C) spiral ganglion neuron densities. The bars or lines in each graph show the group mean, with error bars signifying the standard error. The key at the bottom right applies to all graphs.

layer around the electrode array in the lower basal turn, with no reaction in more apical regions and no internal new bone. We measured the area of fibrous tissue within the area of the basal scala tympani and found the percentage of fibrous tissue to be 3% in the control array group and 4% in the DX1 array group. There was no significant difference in the amount of fibrous tissue or new bone at the cochleostomy site, or throughout the entire cochlea, between the DX1 and the control array implanted groups (Fig. 3B). While not statistically significant, we saw more fibrous tissue and new bone around the cochleostomy site in the DX1 array than the control array group (Fig. 4). We saw no signs of severe surgical trauma any of the cochleae. There was no difference in spiral ganglion neuron densities between the DX1 and the control array groups for any cochlear turn (Fig. 3C). Regression

258

Cochlear Implants International

2014

VOL.

15

NO.

5

analysis showed no relationship between the neuron densities and frequency-specific ABR results (lower basal turn assessed against 32 and 24 kHz; upper basal turn assessed against 8 kHz; and lower middle turn assessed against 2 kHz). Histological sections were matched to acoustic frequencies according to the tonotopic map published in Robertson and Gummer (1985). The organ of Corti was analysed for all turns of the cochlea. No sections were given a score lower than 4, and the few that received 4 (3.54% of total sections organs of Corti assessed) did not correlate with any fibrous tissue formation or damage. There was no difference between any groups.

Trial II In Trial I, although insignificant, we saw more fibrous tissue and new bone growth in the DX1 group than the control group. We hypothesized the additional ‘dusted’ dexamethasone might have changed the surface

Stathopoulos et al.

Safety and efficacy of dexamethasone eluted from an intracochlear array

Figure 4 Examples of tissue reactions (circled) to the control array (top row) and DX1 array (bottom row). Both groups show fibrous tissue and ossification. However, a visible but statistically insignificant increase in tissue growth was present for the DX1 arrays.

properties of the array, making it more adherent for cells. This led to the removal of the extra dexamethasone ‘dusting’ step for the manufacture of the arrays for Trial II. ABR results for the DX2 and the control array groups both exhibited negative threshold shifts immediately after surgery at all frequencies tested (Fig. 5A). Thresholds partially recovered by 30 days for both groups. At 90 days, we saw a second occurrence of hearing loss in the control array group, whereas the DX2 group remained comparatively stable. This trend, while present at all frequencies tested, did not translate into a statistically significant difference between the DX2 and the control array groups at any time point. We saw fibrous tissue and new bone at the cochleostomy site, external to the cochlea in both implanted groups. Within the scala tympani, fibrous tissue was limited to a thin layer of tissue around the electrode array, with no sign of reaction in more apical regions. Only in one cochlea, implanted with a control array, was the basal turn filled with fibrous tissue in all sections. There was no indication of major surgical trauma or infection in this animal. There was no significant difference in the amount of fibrous tissue or new bone at the cochleostomy site, or throughout the entire cochlea, between the DX2 and the control array groups (Fig. 5B). However, there tended to be less tissue in the DX2 group compared with the control array group. We measured scala tympani area in the basal turns and found 6% fibrous tissue in the control array group and 2% in the DX2 array group. There was no significant difference in spiral ganglion neuron densities between the treatment groups for any turn of the cochlea (Fig. 5C). There was no difference in organ of Corti condition between the DX2 and the control array groups.

Almost all sections had an intact organ of Corti, with hair cells and cilia visible in all turns. Only one cochlea – implanted with a control array – lacked a full complement of hair cells and received scores of 2 and 3 in the lower basal and upper basal turns. In this cochlea, hair cell loss was confined to the basal turn, in a region where the osseous spiral lamina had been broken during array insertion. There were no significant differences between groups for organ of Corti scores in any turn. In this trial, cochleae implanted with DX2 arrays showed less hearing loss, marginally better spiral ganglion neuron survival, and less fibrous tissue growth than cochleae implanted with control arrays.

Trial III Following promising results from Trial II, we wished to confirm the possible benefits of the dexamethasone-coated array with a similar prototype. We increased group numbers for robust statistical analysis. The arrays were assumed to be of the same effectiveness as Trial II. Fig. 6A shows changes in hearing thresholds following implantation. The DX3 and the control array groups both lost hearing immediately after implantation. Hearing partially recovered for both the groups by 30 days, and changed little between 30 and 90 days. In this trial, there was no difference in hearing thresholds between the DX3 and the control array groups, for any frequency or time point. Fibrous tissue and new bone occurred at the cochleostomy and fibrous tissue grew as a loose capsule around the electrode array in the lower basal turn. There was no difference in the amount of fibrous tissue or new bone at the cochleostomy, or throughout the entire cochlea, between the DX3 and the control array groups (Fig. 6B). We measured scala tympani area in the basal turn and found 3%

Cochlear Implants International

2014

VOL.

15

NO.

5

259

Stathopoulos et al.

Safety and efficacy of dexamethasone eluted from an intracochlear array

Figure 5 Trial II results: (A) 2 and 32 kHz ABR threshold shifts, (B) tissue growth averages for the whole cochlea and cochleostomy alone, and (C) spiral ganglion neuron densities. The bars or lines in each graph show the group mean, with error bars signifying the standard error. The key at the bottom right applies to all graphs.

fibrous tissue in the control array group and 2% in the DX3 array group. Histological results were similar to those for Trials I and II. There was little difference in spiral ganglion neuron densities (Fig. 6C) between the DX3 and the control array groups. There was no significant difference between groups for organ of Corti scores. Of the total sections organs of Corti assessed, 3.6% from the DX3 array group and 4.5% from the control array group were given a score of 4. A score of 3 was given to 1.1% of DX3 array organs of Corti assessed and 0.7% to the control array group.

Trials overall ABR threshold shifts and spiral ganglion neuron densities were not correlated when trials were assessed individually. However, when data points from all three trials were assessed together, there was a small but significant correlation between the threshold shift

260

Cochlear Implants International

2014

VOL.

15

NO.

5

at 32 kHz from pre-surgery to day 90 and spiral ganglion neuron densities in the lower basal turn of the cochlea (P = 0.011). This relationship was not related to the amount of fibrous tissue or new bone. No correlations were detected between any other measured variables. Overall, we observed very little fibrous tissue and new bone within implanted cochleae: collectively, only 2–6% of the basal turn of cochleae implanted with dexamethasone and control arrays contained fibrous tissue. The outer surface of the explanted DX2 and DX3 arrays were inspected with an environmental scanning electron microscope revealing a pore density difference between these two arrays (Fig. 7). On an average, DX2 arrays contained 360 pores per 100 μm2, while DX3 arrays contained 170 pores per 100 μm2. These pore counts were significantly different (P < 0.0001).

Stathopoulos et al.

Safety and efficacy of dexamethasone eluted from an intracochlear array

Figure 6 Trial III results: (A) 2 and 32 kHz ABR threshold shifts, (B) tissue growth averages for the whole cochlea and cochleostomy alone, and (C) spiral ganglion neuron densities. The bars or lines in each graph show the group mean, with error bars signifying the standard error. The key at the bottom right applies to all graphs.

Figure 7 Scanning electron micrographs of the surfaces of post-explant dexamethasone-eluting arrays in (A) Trial II and (B) Trial III. An average of 360 pores and 170 pores per 100 μm2 were measured in Trials II and III, respectively.

Cochlear Implants International

2014

VOL.

15

NO.

5

261

Stathopoulos et al.

Safety and efficacy of dexamethasone eluted from an intracochlear array

Discussion Our study investigated the viability of three prototype dexamethasone-eluting arrays.

Fibrous tissue growth When an electrode array is inserted into the scala tympani, fibrous tissue is deposited around it as part of the innate immune response to a foreign object. This fibrous tissue layer impedes the flow of electrical charge from the electrodes to the spiral ganglion neurons, thereby increasing the power requirements of the cochlear implant (Newbold et al., 2010). Dexamethasone treatment at the time of implantation reduces fibrous tissue growth in animal models (Connolly et al., 2010; James et al., 2005; Lee et al., 2013). In human cochlear implant recipients, for whom histological data are lacking, a single local dose of corticosteroids administered at the time of implantation provided a prolonged drop in electrode impedance, presumably due to reduced fibrous tissue growth over the electrodes (Paasche et al., 2006, 2009). In light of these results, we expected to see less fibrous tissue in cochleae implanted with dexamethasone-eluting electrode arrays. However, there was no significant difference in the amount of fibrous tissue between control arrays and dexamethasone arrays. We observed little fibrous tissue and new bone within implanted cochleae regardless of array type – an average of 2–6% of the basal turn. In contrast, Lee et al. (2013) reported that fibrous tissue occupied, on an average, 22% of the basal scala tympani of cochleae implanted with non-eluting electrode arrays. This figure dropped to 9% when dexamethasone was given systemically. Work on dexamethasone-eluting arrays have shown a significant reduction in inflammatory cells after 13 days (Farhadi et al., 2013), but longer term studies over 3 months such as this one have not seen any significant effects on fibrous tissue growth (Braun et al., 2011).

Residual hearing Several studies (Connolly et al., 2010; Eastwood et al., 2010; James et al., 2005; Lee et al., 2013; Maini et al., 2009) saw preservation of hearing thresholds using various modes of dexamethasone application. In contrast, we saw no difference between our dexamethasone-eluting and control arrays. Subsequent analysis of these studies showed a relationship between the amount of fibrosis and several other parameters including: damage to the osseous spiral ligament; outer hair cell loss; presence of foreign body giant cells; and hearing threshold shifts (O’Leary et al., 2013). We did not see such relationships in our trials for either the dexamethasone or the control array groups. We did not record extensive fibrous tissue reactions beyond the cochleostomy. Only 1 of

262

Cochlear Implants International

2014

VOL.

15

NO.

5

36 implanted cochleae in our study showed surgically induced structural damage. James et al. (2008) speculate that dexamethasone affords greater protection when trauma is more severe or there is greater resistance to array insertion, even if this does not culminate in extensive trauma. The low level of trauma we inflicted may inform why we saw no statistically significant changes in hearing threshold shifts with the provision of dexamethasone. We note in Trial II, the greater difference in threshold shifts between the DX2 and the control array group coincides with a greater difference in tissue reaction (Fig. 5), a relationship we did not see in Trials I and III.

Considerations The pore density of DX2 arrays was almost double that of the DX3 arrays (P < 0.0001). However, the estimated amount of dexamethasone in the coating of these arrays was similar, 74 and 89 μg, respectively. The reason for the difference in pore density between DX2 and DX3 is unclear. Silicone not loaded with dexamethasone has no pores, therefore, we speculate that the pores may be the result of dexamethasone eluting from the array. If this is the case, it is possible that the greater pore density in the DX2 arrays may have been due to greater elution of dexamethasone. The method and timing of dexamethasone application may also be of importance. It is possible that the protective effect of dexamethasone may be greater when applied prior to drilling of the cochleostomy. In our study, the dexamethasone eluted upon insertion of the array following creation of the cochleostomy. A number of studies saw hearing protection where dexamethasone was applied prior to surgical implantation, either systemically or locally on the round window (Chang et al., 2009; Connolly et al., 2010; James et al., 2008; Lee et al., 2013). One parameter overlooked during the course of our study was the use of the systemic non-steroidal antiinflammatory analgesic, Carprofen (Rimadyl®, Pfizer) as per standard operating procedures for surgery. Whether or not this impacted on the results and effectiveness of the dexamethasone and to what degree are unknown.

Conclusions In some circumstances, incorporating dexamethasone into an intracochlear electrode array could protect against the inflammatory response following cochlear implantation and consequently might help preserve residual hearing. Whether dexamethasone treatment is effective or not may depend on how much trauma is inflicted upon the cochlea during the implantation procedure. The doses of dexamethasone administered in our trials did not adversely affect spiral ganglion neuron and organ of Corti health, or the formation

Stathopoulos et al.

of a cochleostomy seal suggesting that the levels delivered were safe after 3 months in situ.

Acknowledgements We acknowledge the financial support of the HEARing CRC, established and supported under the Australian Government’s Cooperative Research Centres Programme. The Igor Pro ABR analysis package was created by, and used with permission from, Robert Shepherd’s research group currently funded by the National Institutes of Health contract NIDCD HHS-N-263-2007-0053c. We would like to thank James Fallon, Helen Feng, Rodney Millard, Elisa Borg, Sue Pierce, and the staff of the Biological Research Centre. We also thank Andrew Chang, Hayden Eastwood, Jacqueline Andrew, Ricki Minter, Brianna Flynn, Leon Heffer, and Leon Winata for their valued contribution to this work.

References Braun S., Ye Q., Radeloff A., Kiefer J., Gstoettner W., Tillein J. 2011. Protection of inner ear function after cochlear implantation: compound action potential measurements after local application of glucocorticoids in the guinea pig cochlea. ORL: Journal for Oto-Rhino-Laryngology and its Related Specialties, 73: 219–228. Briggs R.J.S., Tykocinski M., Saunders E., Hellier W., Dahm M.C., Pyman B.C., et al. 2001. Surgical implications of perimodiolar cochlear implant electrode design: avoiding intracochlear damage and scala vestibuli insertion. Cochlear Implants International, 2: 135–149. Chang A., Eastwood H., Sly D., James D., Richardson R., O’Leary S. 2009. Factors influencing the efficacy of round window dexamethasone protection of residual hearing post-cochlear implant surgery. Hearing Research, 255: 67–72. Clark G.M., Shute S.A., Shepherd R.K., Carter T.D. 1995. Cochlear implantation: osteoneogenesis, electrode-tissue impedance, and residual hearing. The Annals of Otology, Rhinology & Laryngology. Supplement, 166: 40–42. Connolly T.M., Eastwood H., Kel G., Lisnichuk H., Richardson R., O’Leary S. 2010. Pre-operative intravenous dexamethasone prevents auditory threshold shift in a guinea pig model of cochlear implantation. Audiology and Neuro-otology, 16: 137–144. Culler E., Coakley J.D., Gross N. 1943. A revised frequency-map of the guinea-pig cochlea. The American Journal of Psychology, 56: 475–500. Davidson L.S. 2006. Effects of stimulus level on the speech perception abilities of children using cochlear implants or digital hearing aids. Ear and Hearing, 27: 493–507. Eastwood H., Chang A., Kel G., Sly D., Richardson R., O’Leary S.J. 2010. Round window delivery of dexamethasone ameliorates local and remote hearing loss produced by cochlear implantation into the second turn of the guinea pig cochlea. Hearing Research, 265: 25–29. Eshraghi A.A., Adil E., He J., Graves R., Balkany T.J., Van De Water T.R. 2007. Local dexamethasone therapy conserves hearing in an animal model of electrode insertion traumainduced hearing loss. Otology & Neurotology, 28: 842–849. Eshraghi A.A., Dinh C.T., Bohorquez J., Angeli S., Abi-Hachem R., Van De Water T.R. 2011. Local drug delivery to conserve hearing: mechanisms of action of eluted dexamethasone within the cochlea. Cochlear Implants International, 12(Suppl 1): S51–S53.

Safety and efficacy of dexamethasone eluted from an intracochlear array

Farhadi M., Jalessi M., Salehian P., Ghavi F.F., Emamjomeh H., Mirzadeh H., et al. 2013. Dexamethasone eluting cochlear implant: histological study in animal model. Cochlear Implants International, 14: 45–50. Friedland D.R., Runge-Samuelson C. 2009. Soft cochlear implantation: rationale for the surgical approach. Trends in Amplification, 13: 124–138. Gantz B.J., Turner C., Gfeller K.E., Lowder M.W. 2005. Preservation of hearing in cochlear implant surgery: advantages of combined electrical and acoustical speech processing. Laryngoscope, 115: 796–802. Huang C.Q., Tykocinski M., Stathopoulos D., Cowan R. 2007. Effects of steroids and lubricants on electrical impedance and tissue response following cochlear implantation. Cochlear Implants International, 8: 123–147. James C., Albegger K., Battmer R., Burdo S., Deggouj N., Deguine O., et al. 2005. Preservation of residual hearing with cochlear implantation: how and why. Acta Otolaryngol, 125: 481–491 James D.P., Eastwood H., Richardson R.T., O’Leary S.J. 2008. Effects of round window dexamethasone on residual hearing in a Guinea pig model of cochlear implantation. Audiology & Neuro-otology, 13: 86–96. Kiefer J., Gstoettner W., Baumgartner W., Pok S.M., Tillein J., Ye Q., et al. 2004. Conservation of low-frequency hearing in cochlear implantation. Acta Oto-laryngologica, 124: 272–280. Lee J., Ismail H., Lee J.H., Kel G., O’Leary J., Eastwood H., et al. 2013. Effect of both local and systemically administered dexamethasone on long term hearing and tissue response in a guinea pig model of cochlear implantation. Audiology & Neurotology, 18: 392–405. Leigh J., Dettman S., Dowell R., Sarant J. 2011. Evidence-based approach for making cochlear implant recommendations for infants with residual hearing. Ear and Hearing, 32: 313–322. Maini S., Lisnichuk H., Eastwood H., Pinder D., James D., Richardson R.T., Chang A., et al. 2009. Targeted therapy of the inner ear. Audiology & Neuro-otology, 14: 402–410 Nadol J.B., Jr., Eddington D.K. 2006. Histopathology of the inner ear relevant to cochlear implantation. Advances in Otorhinolaryngology, 64: 31–49. Newbold C., Richardson R., Millard R., Huang C., Milojevic D., Shepherd R., et al. 2010. Changes in biphasic electrode impedance with protein adsorption and cell growth. Journal of Neural Engineering, 7: 056011. O’Leary S., Monksfield P., Kel G., Connolly T., Souter M., Chang A., et al. 2013. Relations between cochlear histopathology and hearing loss in experimental cochlear implantation. Hearing Research, 298: 27–35. Paasche G., Bockel F., Tasche C., Lesinski-Schiedat A., Lenarz T. 2006. Changes of postoperative impedances in cochlear implant patients: the short-term effects of modified electrode surfaces and intracochlear corticosteroids. Otology & Neurotology, 27: 639–647. Paasche G., Tasche C., Stover T., Lesinski-Schiedat A., Lenarz T. 2009. The long-term effects of modified electrode surfaces and intracochlear corticosteroids on postoperative impedances in cochlear implant patients. Otology & Neurotology, 30: 592–598. Robertson D., Gummer M. 1985. Physiological and morphological characterization of efferent neurones in the guinea pig cochlea. Hear Research, 20: 63–77. Simpson A., McDermott H.J., Dowell R.C., Sucher C., Briggs R.J. 2009. Comparison of two frequency-to-electrode maps for acoustic-electric stimulation. International Journal of Audiology, 48: 63–73. Vivero R.J., Joseph D.E., Angeli S., He J., Chen S., Eshraghi A.A., et al. 2008. Dexamethasone base conserves hearing from electrode trauma-induced hearing loss. Laryngoscope, 118: 2028–2035. Xu J., Shepherd R.K., Millard R.E., Clark G.M. 1997. Chronic electrical stimulation of the auditory nerve at high stimulus rates: a physiological and histopathological study. Hearing Research, 105: 1–29.

Cochlear Implants International

2014

VOL.

15

NO.

5

263

Development of a safe dexamethasone-eluting electrode array for cochlear implantation.

Cochlear implantation can result in trauma leading to increased tissue response and loss of residual hearing. A single intratympanic application of th...
583KB Sizes 0 Downloads 4 Views