*3.5. Speech Comprehension*

Due to the retrospective analysis and the resulting limited availability of datasets, both patients whose speech tests were conducted in the free field as well as in direct coupling were included in the following speech data evaluation.

To avoid bilateral benefits, based on the results of the audiogram, an additional measurement in direct coupling was carried out if it was suspected that the non-test ear was influenced. Therefore, it can be assumed that the influence of the non-test ear was selected entirely from the speech results (Figure 5).

For the SPLIT group, only a few datasets were available since not every patient could complete each of the tests. As a consequence, the number of subjects varied between tests at measurement point after 6 months (M6). Furthermore, the results of a bilaterally implanted child in the SPLIT group were excluded from an evaluation of the speech data, as speech development was not possible due to child's age. The exclusion allowed an unbiased trend of hearing performance in patients implanted with a split-array CMD.

**Figure 5.** The Freiburg monosyllabic word test (FMT), and the HSM sentence test in quiet and in noise. Median scores (in % correct) after 6 months of device use. Comparing patients from the IES group (red dots), the REGULAR group (black dots), and the SPLIT group (blue dots). Asterisks mark significant differences between groups.

After 6 months of device use, for FMT, the subjects belonging to the IES group scored a median of 30%. The median was 60% after 6 months in the REGULAR group and 0% in the SPLIT group. There were significant differences between the IES and REGULAR groups (IES/REGULAR *p* < 0.01). Due to the small size of the test group, no statistical evaluation of the SPLIT group was carried out.

For the HSM test in quiet, the subjects of the IES group scored a median of 34.9% after 6 months of device use. In comparison, the median values of the REGULAR group were 95% after 6 months and 0% after 6 months in the SPLIT group. An intergroup comparison between the IES and REGULAR groups showed a significant difference in the results of the HSM test in quiet (*p* < 0.01).

For the HSM test in noise, patients of the IES and SPLIT groups achieved median values of 0% after 6 months of device use, as opposed to the REGULAR group (63.49%; *p* < 0.001).

## *3.6. Clinical Data*

The clinical data presented here are based on a retrospective analysis of the patients' medical history before and after cochlear implantation. The numbers of patients suffering from tinnitus, vertigo, or facial stimulation at the respective times are shown in Table 2. Furthermore, seven patients in the IES group and two patients in the SPLIT group suffered from meningitis before surgery. None of the patients in the REGULAR group had a pre-existing meningitis in their medical history. After surgery, there was no evidence of postoperative meningitis in all three groups.

In order to alleviate facial stimulation in affected patients, triphasic pulses were utilized in four of the affected cases (Table 2), but were only temporarily sufficient for one of the cases. In the three remaining patients, individual electrode contacts had to be switched off and the stimulation level was lowered below the facial nerve stimulation (FNS) threshold in order to remedy the symptoms. The remaining two cases were not further documented because one patient was lost for follow-up (patient in the SPLIT group) or the other patient was currently no longer fitted with a CI (patient in the IES group).


**Table 2.** Clinical data.

<sup>1</sup> *n* = 8 patients were evaluated in the SPLIT group; *n* = 28 patients were evaluated in the REGULAR group; and *n* = 33 and *n* = 29 patients were preoperatively and postoperatively evaluated in the IES group, respectively.
