Cochlear implants are remarkable neural prostheses which have benefitted those with severe-profound sensorineural hearing loss achieve open set speech understanding. In recent years, research and continued design improvements have been made to the electrode arrays as well as to the processing strategies used to transmit auditory information to the electrodes. The newer designs have had the goal of allowing for atraumatic insertion with the goal of preserving intracochlear structure and residual acoustic hearing when present, while stimulating more discrete populations of spiral ganglion cells.
Among the commercially available electrode arrays available in the United States, there are precurved “modiolar hugging” (perimodiolar) arrays designed for closer position relative to the modiolus as well as straight “lateral wall” arrays designed to reside at the lateral wall of the cochlea. Additionally, mid-scalar designs are precurved electrodes intended to reside midway between the modiolus and lateral wall. There are proponents of each type with conflicting data to reinforce benefits and potential drawbacks related to both design approaches [see, for example, (1,2)].
The CI532 (Cochlear Ltd., Sydney, Australia) is a slim perimodiolar electrode with apical dimension of 0.35 and 0.475 mm at the basal end of the array with the last of 22 contacts at 14.40 mm and the first white band at 16.4 mm. Comparatively, the CI512 Contour Advance, also a perimodiolar electrode array by Cochlear has an apical dimension of 0.5 mm and basal dimensions of 0.8 mm with the last of 22 contacts at 14.25 mm and the first white band at 17.0 mm. Because perimodiolar electrodes assume a more medial position in the cochlea compared to lateral wall arrays, they tend to be shorter in total length than comparable lateral wall arrays. The CI522, Cochlear's slim-straight lateral wall array has the last contact at 19.10 mm while the shorter Hybrid-L array's last contact is at 15.0 mm.
While there is data to suggest shorter electrode length is more favorable for hearing preservation (3,4), there is also precedent for considering hearing preservation with full-length electrodes (5–7), similar in length to the electrode of study here as well as even longer electrodes (notably the Flex series by Med-El all of which have a last contact at 20 mm or greater) (8).
The Cochlear CI532 array was designed with the preferred features of both types of arrays—slim and flexible like outer wall arrays but pre-curved allowing for the benefits of the “perimodiolar advantage” which includes lower impedance levels, lower threshold and comfort levels (9), potential for more discrete stimulation of targeted spiral ganglion cells and decreased battery requirements.
The design of this array incorporates a novel mechanism for insertion with aspects of an advanced off stylet (AOS) in which the array is inserted into the basal turn of the cochlea and then advanced into space with the goal of avoiding direct contact with the outer wall of the cochlea (10). To accomplish this, the CI532 instead utilizes an external sheath rather than an internal stylet that allows for a significantly thinner overall array with a decrease in diameter of more than 60%. Years of research and temporal bone trials preceded its commercial release (11) to optimize its design for mainstream use. One potential issue identified in earlier iterations included the potential for tip fold overs which may be more of an issue with thinner perimodiolar electrodes because of the limited tactile feedback and very thin design (providing “room” for it to roll over). It has been our Center's protocol to confirm electrode placement intraoperatively with a plain film x-ray with all devices long preceding the introduction of this electrode (12). Since FDA approval in 2016, the CI532 has been used at our institution when a full length Cochlear array was chosen. Herein, we describe our surgical experience with this device, speech perception outcomes to date and data regarding preservation of acoustic hearing at the time of surgery as measured at initial activation.
All surgeries were performed between April 2016 and December 2017 at a single institution. Institutional IRB exemption was provided for this retrospective study. There was no consistent steroid protocol other than an intraoperative intravenous weight-based dose of Dexamethasone. During this time period, 237 CI532 devices were placed and data from all of these surgical procedures are included in this review. Of the 237 ears implanted, there were 52 patients (104 devices) implanted bilaterally, either simultaneously or sequentially with the CI532 during this study period.
Through June 2017, there were 163 patients for which there was preoperative and some postoperative data available. While the outcomes from surgery include all patients, prelingual children <5 years old at the time of the implantation were not included in the outcomes data for speech perception or hearing preservation (n = 49). A few patients were participants in clinical trials which required manipulation of sound coding strategies and so their data was excluded as well (n = 6). Other patients were implanted for off-label indications such as single-sided deafness (n = 8), or English was not their primary language and their results were not included in the outcomes analysis (n = 6). After excluding such patients, there were 94 patients with sufficient postoperative data for analysis.
Patient characteristics including etiology of hearing loss, duration of hearing loss/deafness, duration of hearing aid use, and age at cochlear implantation were documented.
The mean pure tone average (500, 1000, 2000 Hz) was calculated for the ear to be implanted preoperatively among those able to participate with testing of: 88.7 dB (N = 94) (Table 1). These patients are mostly comprised of standard cochlear implant candidates and not those specifically identified with residual acoustic hearing. Figure 1 shows the preoperative and mean unaided thresholds for these patients.
All patients in this series had a standard transmastoid posterior tympanotomy approach via a facial recess with an antero-inferior extended round window or peri-round window cochleostomy approach. Attempts were made to minimize insertional trauma by completing the opening of the endosteum using manual dissection rather than drilling, especially in cases where residual acoustic hearing was present. The electrode array was loaded in its sheath and with the fin facing antero-superiorly towards the dome of the lateral semicircular canal. While keeping the stopper “hubbed” at the cochleostomy opening, the electrode array was slowly advanced out of the electrode sheath. The electrode sheath was then removed, and the three white markers visible at the cochleostomy site at which time the cochleostomy was packed in the usual fashion using autologous tissue.
Intraoperative Testing and Confirmation of Configuration and Placement
All patients undergoing cochlear implantation at our center with any device or manufacturer have intraoperative impedances tested and manufacturer specific response telemetry performed. Additionally, an intraoperative plain film x-ray to confirm electrode placement and configuration is evaluated by the surgeon before completing the operation (12). If tip roll over was identified on x-ray, reloading and re-insertion of the same array was attempted with repeat x-ray. In cases of persistent tip fold over, a different electrode array was utilized as a backup device.
As part of our standard cochlear implant evaluation, each patient underwent pre- and postoperative audiological evaluations that included pure-tone air and bone thresholds in the unaided and aided setting at the time of initial evaluation, stimulation, at 3 months and 6 months poststimulation and at one year postoperatively. In addition, speech perception performance was evaluated at each of these time points using monosyllabic consonant-nucleus-constant (CNC) words which were used as the primary endpoint for speech perception in this study.
Preservation of Residual Acoustic Hearing
Additionally, preoperative unaided hearing in the implanted side alone was evaluated at individual low frequencies by computing a low frequency pure tone average (LF-PTA) by averaging thresholds at 250 and 500 Hz to assess for changes over time and to correlate with performance.
Patients with LF-PTA better than 80 dB were selected for subgroup analysis. Because of the inherent heterogeneity in assessing patients with low frequency residual hearing even among patients with LF-PTA better than 80 dB preoperatively, patients with measurable thresholds in these frequencies were divided into 2 groups—(1) those with LF-PTA of 65 dB or better (2) those with LF-PTA worse than 65 dB but ≤80 dB.
The rates of low frequency hearing preservation with the CI532 were calculated. The LF-PTA as described above was calculated for each patient preoperatively in the unaided condition, and with the device off (acoustic only) at the time of initial stimulation. Absolute change in thresholds as well as endpoint thresholds were noted. Given the limited duration of follow-up and in order to maximize the number of implantations analyzed, thresholds at initial stimulation only are included here as to assess the potential to implant the array with preservation of residual acoustic hearing.
Continuous variables were summarized by reporting means along with standard deviations (SD). A nonparametric equivalent of the two sample student's t-test (Wilcoxon's Rank Sum Test) was used to compare patients change in speech perception over different time points as the number of subjects with follow-up at a given time point varied. For the analyses presented, a p-value <0.05 was considered significant.
Speech perception data is from those patients contributing data with average age of 55.66 years (SD 26.79) and is shown in Table 2. Of the patients implanted with at least 6 months of CI experience, preoperative CNC-W scores were available; the mean preoperative CNC word score was 11.2% (SD 15.5). At three month postoperative evaluation, mean CNC word score was 57.2% (SD 21.9) for n = 71 patients with data available at this time point significantly improved from preoperative scores (t = −12.4, p < 0.00001). This remained significant over time for patients (n = 43) who completed a 6 month postimplantation evaluation at which time mean CNC word score was 59.9% (SD 19.5, t = 15.2, p < 0.0001). Of the 33 patients with one year data, mean CNC word score was 62.7% (SD 18.2, t = 15.1, p < 0.0001).
Surgical Preservation of Residual Acoustic Hearing
We assessed the ability to preserve residual acoustic hearing after insertion of the electrode array by the methods described above. In this cohort, a total of 43 patients had preoperative unaided LF-PTA of 80 dB or better. Figure 2 plots preoperative and initial activation acoustic only thresholds for these patients. Of those 43 patients, 17 ears (39.5%) in 16/43 patients maintained a LF-PTA comprised of mean thresholds at 250 and 500 Hz better than 80 dB at time of initial stimulation with their cochlear implant. In assessing the absolute change in thresholds, these 17 cases lost an average of 20.1 dB at initial stimulation compared to their preoperative thresholds at these two frequencies. The mean age of these patients was 51.9 years (SD 27.98), not different than the overall cohort in terms of age at CI.
In considering the subset of patients with better preoperative “usable” or “aid-able” hearing, 22 of these 43 patients had preoperative acoustic thresholds equal or better than 65 dB at 250 and 500 Hz. Twelve of the 22 (54.5%) patients maintained both thresholds at 80 dB or better at initial stimulation. Six of these patients lost this residual acoustic hearing to profound levels at stimulation. Four patients maintained one of these thresholds at 80 dB or better at initial stimulation.
In our series, 11 of 237 (4.6%) electrode insertions had a tip rollover event during electrode insertion as in Table 3. These were all identified intraoperatively by plain film x-ray as per our protocol (e.g., Fig. 3) and re-inserted with proper configuration of the array confirmed prior to leaving the operating room. All were inserted via extended round window or peri-round window cochleostomies, never directly through the round window.
There was no resistance felt at the time of insertion in any of these cases. In all cases the intraoperative NRT and impedance testing did not show any abnormalities in any of the electrodes. Methods used to address an identified tip rollover varied by circumstance. In five of these cases, the rollover could be corrected by adjusting the insertion angle such that the trajectory of the tip of the array is closer to the modiolus after insertion of the sheath, or in other cases by widening the cochleostomy, which allows for a slightly different trajectory of the electrode array upon re-insertion. In two cases where reinsertion of the same electrode corrected the tip rollover, roll over persisted on the second insertion, and corrected on the third attempt.
In the remaining six cases, a backup device was used. In one case, S2, the backup device was used simply because the insertion sheath was not saved after the initial insertion on which a tip rollover was identified. This was early in our experience with the CI532 whereas we subsequently learned it is critical that after insertion, the sheath is not discarded and is kept sterile until the electrode is verified to be in good position as there is currently no separate sterile package of insertion sheaths available.
In the six cases where a different device was used, in only one instance S8 was a new CI 532 inserted, in all other cases a different Cochlear electrode array (the CI512 or 522) was used. In S8 and S10, before insertion, upon loading the array, attempting to draw back the electrode, it was noted the distal array traversed the slit in the sheath which may have predisposed to the tip foldover.
None of these patients had significant residual acoustic hearing and so the impact of re-insertion of hearing preservation is unclear. In terms of a possible impact on outcomes, this is a small series of patients with a great deal of variability in their demographics and indications including two single-sided deaf patients. Duration of deafness, a factor known to highly correlate with outcome measures varied widely as evidenced in Table 3. Early results, however, are consistent with the known variation in outcomes amongst cochlear implant recipients.
Changes in outcomes and expectations in cochlear implantation have come about from a number of reasons including improvement in electrode design, signal processing strategies, surgical techniques, and perhaps most importantly the patterns of hearing loss in patients being implanted. Device and electrode selection have become an increasingly complex discussion. Inherent in its design is the hope the CI532 may capture the benefits sought by current cochlear implant candidates presenting with increasing amounts of residual acoustic hearing while offering the benefits of perimodiolar positioning. While shorter arrays may offer higher rates of hearing preservation, the benefit of a long array with broader cochlear coverage is to obviate the need for a second surgery should residual acoustic hearing be lost and performance decline (13,14).
Preliminary clinical outcomes with the CI532 compare favorably with speech perception outcomes and hearing preservation rates from other Cochlear arrays as well as those of other manufacturers. Generally, there is great variability in speech perception in terms of outcomes measures and time points as well as changes over time possibly related to changes in candidacy considerations. Results tend to be better in patients with residual hearing preoperatively. When considering mean postoperative CNC words, CI512 outcomes range between 61.5 and 76% in more recent studies (1,15). Rates for Med El Standard and Flex 28 electrodes from some recently published data range from CNC-W scores of 43 and 47% (2). For the HiRes90K by Advanced Bionics, CNC word scores of 62.4% have been reported (16).
Comparing hearing preservation rates is even more challenging. There is a wide range of hearing preservation rates noted in recent literature and reported outcomes are highly dependent on preoperative hearing and how hearing preservation is defined. This makes comparisons across studies, let alone across arrays, difficult. Most studies consider any amount of measurable acoustic hearing, though others use a more stringent acoustic threshold in the “aid-able” range, itself an ill-defined feature. Others consider together those with widely varying preoperative acoustic thresholds. There is also a range as to the time points at which hearing was measured. In the current study, hearing preservation with the slim perimodiolar electrode among those with preoperative LF-PTA of 80 dB was 39.5% before 1 year. Among the Cochlear arrays, reported rates with the Contour Advance (17) are from 24 to 43% (18). For the AB Hi Focus Mid Scala rates of aidable hearing at one year after activation of 31% were recently reported (17) and for the Flex EAS system from Med El rates of 35% preservation to within 10 dB were reported in the first two months after surgery (19).
Some may argue that the CI532 electrode introduces complexity and considerations not previously relevant or necessary with other arrays. Our experience suggests that these challenges may be easily overcome and may allow us to maximize outcomes to better meet ever rising patient expectations. Unique among the perimodiolar options available from Cochlear Corporation, the CI532 electrode array is “reloadable” in cases where a tip fold over is detected and can be re-inserted rather than utilizing the backup device. Training on the technique of insertion in a laboratory setting along with the ability to confirm placement with intraoperative x-ray are particularly important with this electrode given the lack of tactile feedback. Our institutional experience both implanting the electrode and in training other surgeons in laboratory exercises (manuscript in press) suggests that there is a learning curve associated with this electrode's use with a majority of the tip roll overs to date occurring within the first 100 cases.
At present, intraoperative imaging is the only way to detect such issues with tip fold over, although spread of excitation studies and electric field imaging may prove to be a reliable way of detecting tip rollovers in the future (20). Of note, the manufacturer only recommends insertion a maximum of two times prior to resorting to the backup device. It is possible that multiple reinsertions may change the mechanics of the electrode. Early in the experience, it was noted that some arrays out of the package were misaligned with the sheath and electrode such that when the electrode is drawn back into the sheath it does not straighten appropriately. Further experience has shown that it is possible that rotating the sheath and electrode alignment before insertion could rectify this, reducing the likelihood of a tip rollover; however, this requires further study.
In a majority of the cases, the surgeon did not anticipate a tip rollover based on preoperative imaging or intraoperative findings. Given the unknown element as to why these events occurred in these particular cases, a different array was often chosen in the event there was an intracochlear anatomic variation predisposing to this issue with this particular array. It is notable that in some of the patients in this series with tip rollovers, the CI532 was a simultaneous or sequential implant with no identified issues with implant surgery on the contralateral side. One subject did have a Mondini malformation. There are, however, a number of subjects in our series with this malformation that have been successfully implanted with a CI532 electrode and this in fact is now the electrode we use most often for such cases. In other malformations that may introduce more variability in the anatomy of the basal turn, alternate electrode arrays should be carefully considered.
Our experience demonstrates the ability to preserve residual acoustic hearing with surgical insertion of this electrode at stimulation. As demonstrated by the preoperative thresholds of the patients studied here, this electrode has been used as our standard Cochlear full-length electrode and as such most of these patients are standard CI candidates with minimal residual acoustic hearing, not candidates chosen specifically based on the degree of preoperative residual acoustic hearing. Additionally, as has been noted with virtually all electrode arrays to date including short arrays designed for hearing preservation in those with significant amounts of preoperative residual acoustic hearing (21), loss of hearing over time has been noted in these cohorts as well and remains an important phenomenon to be understood. As such, a limitation of this study is the lack of long term data on preservation of acoustic hearing with this new electrode.
The nature of the electrode array dictates that it is not appropriate for all cases. Specifically, this electrode is not for revision cases, abnormal anatomy such as most cochlear malformations, or cases involving cochlear ossification.
The CI532 uses a novel insertion technique to facilitate placement of a thin perimodiolar array. There is a learning curve to its use and intraoperative x-rays are of value to confirm for optimal placement. Speech perception outcomes are similar to other full-length arrays with the added potential for at least short-term preservation of residual acoustic hearing. Limited results in standard CI patients with residual acoustic hearing suggests that it is possible to preserve hearing during surgery and for an extended time postoperatively, while in others the mechanism for loss of hearing overtime remains to be understood.
The authors thank Emily Spitzer AuD for assistance in revisions to the manuscript.
1. Holden LK, Finley CC, Firszt JB, et al. Factors affecting open-set word recognition in adults with cochlear implants. Ear Hear
2. O’Connell BP, Cakir A, Hunter JB, et al. Electrode location and angular insertion depth are predictors of audiologic outcomes in cochlear implantation. Otol Neurotol
3. Roland JT Jr, Gantz BJ, Waltzman SB, et al. United States multicenter clinical trial of the cochlear nucleus hybrid implant system. Laryngoscope
4. Gantz BJ, Dunn C, Oleson J, et al. Multicenter clinical trial of the Nucleus Hybrid S8 cochlear implant
: final outcomes. Laryngoscope
5. Friedmann DR, Peng R, Fang Y, et al. Effects of loss of residual hearing on speech performance with the CI422 and the Hybrid-L electrode. Cochlear Implants Int
6. Van Abel KM, Dunn C, Sladen DP, et al. Hearing preservation
among patients undergoing cochlear implantation. Otol Neurotol
7. Svrakic M, Roland JT Jr, McMenomey SO, et al. Initial operative experience and short-term hearing preservation
results with a mid-scala cochlear implant
electrode array. Otol Neurotol
8. Pillsbury HC 3rd, Dillon MT, Buchman CA, et al. Multicenter US Clinical Trial with an electric–acoustic stimulation (EAS) system in adults: final outcomes. Otol Neurotol
9. Shepherd RK, Hatsushika S, Clark GM. Electrical stimulation of the auditory nerve: the effect of electrode position on neural excitation. Hear Res
10. Roland JT Jr. A model for cochlear implant
electrode insertion and force evaluation: results with a new electrode design
and insertion technique. Laryngoscope
11. Briggs RJ, Tykocinski M, Lazsig R, et al. Development and evaluation of the modiolar research array—multi-centre collaborative study in human temporal bones. Cochlear Implants Int
12. Cosetti MK, Troob SH, Latzman JM, et al. An evidence-based algorithm for intraoperative monitoring during cochlear implantation. Otol Neurotol
13. Fitzgerald MB, Sagi E, Jackson M, et al. Reimplantation of hybrid cochlear implant
users with a full-length electrode after loss of residual hearing. Otol Neurotol
14. Carlson ML, Archibald DJ, Gifford RH, et al. Reimplantation with a conventional length electrode following residual hearing loss in four hybrid implant recipients. Cochlear Implants Int
15. Holden LK, Firszt JB, Reeder RM, et al. Factors affecting outcomes in cochlear implant
recipients implanted with a perimodiolar
electrode array located in scala tympani. Otol Neurotol
16. Bierer JA, Spindler E, Bierer SM, et al. An examination of sources of variability across the consonant-nucleus-consonant test in cochlear implant
listeners. Trends Hear
17. Hunter JB, Gifford RH, Wanna GB, et al. Hearing preservation
outcomes with a mid-scala electrode in cochlear implantation. Otol Neurotol
18. Cosetti MK, Friedmann DR, Zhu BZ, et al. The effects of residual hearing in traditional cochlear implant
candidates after implantation with a conventional electrode. Otol Neurotol
19. Adunka OF, Dillon MT, Adunka MC, et al. Hearing preservation
and speech perception outcomes with electric-acoustic stimulation after 12 months of listening experience. Laryngoscope
20. Grolman W, Maat A, Verdam F, et al. Spread of excitation measurements for the detection of electrode array foldovers: a prospective study comparing 3-dimensional rotational x-ray and intraoperative spread of excitation measurements. Otol Neurotol
21. Roland JT Jr, Gantz BJ, Waltzman SB, et al. Long-term outcomes of cochlear implantation in patients with high-frequency hearing loss. Laryngoscope