**4. Discussion**

To our knowledge, this was the first comparative yet retrospective analysis to show that the IES CMD can facilitate the implantation of flexible lateral wall CI electrodes in patients with fibrotic, ossified, and malformed cochleae. Moreover, we showed that individual treatment with the IES CMD allowed a regular electrode array to be implanted without significantly impairing the performance of the patients.

Specifically, intraoperative impedance values of all three test groups were equally high and a presumed increase from insertion up to the FF was also evident in all three groups (Figure 2). At the timepoint M12, it can be stated that the median values of impedances of the IES and SPLIT groups were significantly higher than the median value of the REGULAR group. However, in relation to this, the median value of the SPLIT group was also significantly higher than the median value of the IES group.

In the REGULAR group, no significant increase in the median impedance values over time can be observed. Thus, based on the impedance values, an increased trauma due to the electrode insertion seemed unlikely here [15–17]. In contrast, there was an increase in impedances over time in the IES and SPLIT groups. A correlation between the level of fibrotic material and the impedance levels was found in preclinical cochlear implant models consisting of guinea pigs [18]. It could be possible that increased tissue formation due to insertion trauma may account for the increased impedance values observed in the IES and SPLIT patients. The increase in the IES group could be due to an increased amount of fibrous tissue growth around the electrode. However, it must be critically questioned whether this deviation, especially in comparison to the REGULAR group, is solely due to the use of the IES during surgery and the additional microtraumas that may have arisen. Patients in the IES group, unlike patients in the REGULAR group, already had increased cochlear damage (fibrosis, ossification, and malformation) before implantation and this damage may also explain the different impedance values. Another hypothesis therefore could be that the initial fibrotic tissue formation prior to implantation continues afterwards and leads to higher impedances. Furthermore, increased surgical trauma, such as cochlear drilling, as performed in some patients, could also trigger inflammation reaction with new tissue formation accounting for the increase in impedances [19]. In addition, it has not been evaluated whether other traumatic events, such as scalar shift, occurred after implantation, and whether they influenced the data. However, no scalar shift was described in the regular radiological evaluations of the postoperative CT images.

Further typical changes in electrode impedances occurred with the onset of electrical stimulation. It has been reported that the impedances of intracochlear electrodes are lower after stimulation compared to the levels before stimulation onset [20,21]. After implantation, a cell [22] and passivation [23] layer accumulated on the electrode surface of inactive electrode contacts. This layer was disrupted by the onset of electrical stimulation and resulted in a decrease in impedances, as clinically observed. Within 3 months after starting the electrical stimulation (after the FF), a decrease in impedances in all three groups, i.e., the IES group, the SPLIT group, and the REGULAR group, was observed in the present study. This decrease might corroborate studies showing that chronical electrical stimulation may lower electrode impedances [24–26].

As a result, it must be questioned whether patients in the SPLIT and IES groups have a shorter or more irregular wearing time of the implant processor to M12 and thus a nonregular electrical stimulation, thus contributing to an increase in impedances. However, this parameter was not recorded in our study.

Significantly increased impedance values 12 months after the FF were demonstrated in the groups with pre-damage of the cochlea (O and F groups) when compared to the REGULAR group (Figure 3). Pre-damage of the inner ear might have a significant influence on the further development of the electrode environment after implantation, leading to increased impedance values. No significant difference was found between patients who used the IES as a depth probe (the DM group) and patients in the REGULAR group, indicating that the IES is rather less traumatic when used in non-pre-damaged ears.

This assumption is contradicted by impedance values measured 12 months after implantation (M12), as depicted in Figure 4. While the impedances between all groups were still intraoperatively equal, the impedances of the IES group (FI and RI) were significantly increased from the FF onwards, as well as the values of the RR group. At M12, the values of the RR group dropped again. It is therefore assumed that without the use of the IES, the impedances will return to lower values over time. Thus, the IES could account for higher impedance values over time and should not be routinely used as a depth sensor, as the electrode environment appears to recover better without the use of the IES.

When interpreting the results, the usefulness of impedance values needs careful consideration. It is well accepted that electrode impedances may be a useful biomarker of inner ear pathology after cochlear implantation [27] and low impedances are desirable to minimize battery consumption. As such, impedances represent a non-invasive measuring method for obtaining information about the environment between electrodes and the respective neural interfaces [28]. It is believed that changes in the electrode impedances are related to the formation of a fibrous tissue matrix around the electrode array [18,22,29,30]. Foreign body immune responses may help to encapsulate the electrode array in fibrous tissue within the first few weeks of implantation [20,31]. Clinical studies on patients treated with cochlear implants show that, in the days and weeks following implantation, electrode impedances increase, forming a plateau after 4–6 weeks in situ [22,32].

Foggia et al. describes an inflammatory or fibrotic reaction as a response to the electrode array in the cochlea that occurs with every implantation [20]. Both the acute tissue response immediately after implantation (due to the insertion trauma) and the delayed response as a host-mediated foreign body response caused by nearly all biomaterials may help to explain the observed increase in impedance values [20]. Interestingly, a recent study has shown that impedances values do not correlate with speech understanding [33]. Considering the results of speech comprehension of the different groups 6 months after implantation, the patients of the REGULAR group achieved fairly good speech performance. The results obtained are consistent with those taken from other studies [34]. For the SPLIT group, poor results for speech perception were observed in our study, as corroborated by other studies [10]. Degeneration in spiral ganglion cells is particularly high in patients whose cause of deafness is bacterial meningitis [35]. Since many patients in the IES group had preoperative meningitis in their medical history, it can be assumed that a reduced number of spiral ganglion cells is one of the reasons for lower speech comprehension scores between the IES group and the REGULAR group. Therefore, as an important observation based on our results, it can be concluded that patients with implantation of a long lateral wall electrode using the IES CMD prior to insertion experience significant benefits in terms of speech understanding, as compared to those who were fitted with a Split-Array CMD (Figure 5). However, the case of a child implanted bilaterally with split arrays, which was excluded from our speech comprehension data, also shows that a long-term evaluation

is indispensable in order to be able to make further statements. The child did not show any measurable scores at time M6 of the speech comprehension data evaluation as the tests were not age-appropriate; however, measurable speech understanding has developed in the meantime. To conclude, if conditions permit, normal implantation with the aid of the IES CMD should be favoured over the implantation of a Split-Array CMD. Speech recognition after cochlear implantation is further dependent on the degree of spiral ganglion cell preservation.

Since there are other stiff electrodes available for cochlear implantation, questions surrounding the use of delicate electrodes, such as the FLEX series from MED-EL, arise since the recipients presented with an obliteration of the cochlea with no residual hearing. The FLEX electrode array series includes atraumatic devices with variably sized lengths up to 28 mm. As such, this series allows the length of the electrode array to be correlated with the size of the patient's cochlea, thus adapting to variations in cochlear geometry. An advantage of flexible electrode arrays is the avoidance of pronounced trauma to the cochlea during electrode insertion. Since patients in the IES group have previous damage up to the anatomically complete obstruction of the cochlea, there is no residual hearing worth protecting. Thus, the use of flexible electrodes does not seem beneficial from this point of view. However, several studies, such as those by Buchmann et al. [36] or Büchner et al. [34], have shown that speech understanding after implantation significantly depends on the insertion depth of the electrode and that the insertion of longer electrodes allows better speech perception. With this in mind, the electrodes used represent a promising means of achieving better speech understanding in affected patients, not due to their flexibility, but because of their potential increased cochlear coverage. Moreover, the various malformations in the IES group that we evaluated are cases in which the insertion length can often only be determined intraoperatively. The ability to adjust the insertion length accordingly is essential.

Meningitis may occur more frequently in patients implanted with a CI either due to local infections or as a result of the actual surgical intervention. Other risk factors such as congenital or acquired anatomical defects, previous meningitis, or immunodeficiencies are described [37]. In the IES group, no postoperative meningitis occurred at the time of evaluation, although many of the risk factors apply to the evaluated group (Table 1). Furthermore, the IES CMD does not appear to increase the risk for meningitis. Vertigo and vestibular dysfunction may occur postoperatively after cochlear implantation [38]. Patients of the IES group already showed symptoms of this category preoperatively, and the number of affected patients slightly decreased postoperatively (Table 2). Here, it must be considered that the pre- and postoperatively affected patients do not necessarily correlate with one another. However, it seems that the use of the IES CMD does not result in a strongly increased trauma risk, since clinical symptoms and adverse events are to be regarded as indicators. Postoperative facial stimulation is observed in patients of all three groups. In the REGULAR and IES groups, facial stimulation could be controlled and improved by the use of triphasic pulses, while no change occurred in the SPLIT group (Table 2).

There are several limitations associated with the present study. First of all, the small number of patients included in this study limits the conclusions that can be drawn from the data. Additionally, the inhomogeneous patients, especially in the IES group, make it difficult to compare the data to the other study groups. Despite these limitations, we were able to demonstrate significant differences between groups and we could derive some important observations from our data.

## **5. Conclusions**

In summary, this study shows that the IES CMD can successfully treat patients who would otherwise be non-users or would only be able to receive a split-array CMD or an insufficient number of inserted electrode contacts. The IES CMD offers a method to insert long flexible lateral-wall electrodes into the cochlea with a concomitant low risk of clinical complications.

The above evaluation also shows the broad applicability of the IES CMD, as it is a tool that can be used in almost all age groups and for a wide range of diseases. The IES CMD forms an important safe surgical aid for special cases, which does not greatly prolong surgical intervention and makes successful implantation possible.

Nevertheless, the IES CMD should not be applied as a standard instrument for cochlear implantation because its use leads to higher postoperative impedances, possibly due to a more invasive and traumatic implantation when compared to the FLEX arrays without the IES CMD. If possible, imaging techniques, or, more specifically, manufactured insertion electrodes, should be used to determine cochlear length, eliminating any negative influence on hearing results using the IES. In the future, it may be interesting to evaluate whether steroid administration via an inner ear catheter [15] can lead to a further reduction in impedances after implantation with the IES CMD. Furthermore, the IES CMD also provides a means of overcoming restrictions in the cochlea, allowing more electrode contacts to be inserted when only partial insertion would be possible without its use. The question around whether speech understanding can be improved also needs to be clarified further in future research endeavours.

**Author Contributions:** Conceptualization, A.W. and R.B.S.; data curation, J.A.C.H., A.W. and R.B.S.; formal analysis, J.A.C.H. and R.B.S.; funding acquisition, T.L.; investigation, J.A.C.H., A.W. and R.B.S.; methodology, J.A.C.H., A.W. and R.B.S.; project administration, A.W. and R.B.S.; resources, M.E.T., E.K., N.K.P., L.G. and T.L.; software, E.K. and L.G.; supervision, A.W., T.L. and R.B.S.; validation, E.K. and L.G.; visualization, J.A.C.H., A.W. and R.B.S.; writing—original draft, J.A.C.H.; writing—review and editing, J.A.C.H., A.W., M.E.T., E.K., N.K.P., L.G., T.L. and R.B.S. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Institutional Review Board Statement:** For the evaluation of anonymized intra-departmental and routine patient data, there was no obligation to submit the data to the Ethics Committee of Hannover Medical School in principle. Protocol code: 1897-2013. Date of approval: 9 July 2013. The study was conducted in accordance with the Declaration of Helsinki.

**Informed Consent Statement:** Informed consent was obtained from all subjects involved in the study.

**Data Availability Statement:** The data presented in this retrospective analysis are available on request from the corresponding author. The data are not publicly available as they contain personal and sensitive patient data. All data are stored on the servers of Hannover Medical School, Department of Otorhinolaryngology.

**Acknowledgments:** We are grateful to Cornelia Batsoulis and Max Fröhlich, both from the MED-EL Hannover Research Center, for their scientific support in this retrospective study.

**Conflicts of Interest:** The authors declare no conflict of interest.
