One of the main findings of this study was that impedance at the initial fitting session measured in the 26 subjects whose devices were activated on the day after surgery (early switch-on) were significantly lower than that in the subjects who underwent conventional switch-on between 2 and 3 weeks after implantation of slim modiolar electrodes. The impedance values obtained intraoperatively were similar between the groups. However, the two groups showed interesting differences in the evolution of impedance from surgery to the initial fitting session. The average impedance level (MP2) in the conventional group increased by 34.8% (p = 0.001) at the initial switch-on session compared to the intraoperative level, while that of the early switch-on group had decreased by 54.9% (p < 0.001). This increase in impedance level for the slim modiolar electrode array during the first 2–3 weeks before the initial switch-on in the conventional group was similar with that reported in previous studies with different types of electrode arrays, including the Nucleus 24 contour electrode array and the Nucleus standard straight array21-23. As reported in a previous study using the Nucleus implant system, a significant reduction in impedance for all electrode contacts from slim modiolar electrodes was also observed from the early switch-on group in the present study18,24. In contrast, other studies using the MED-EL or the Advanced Bionics implant system showed no significant decrease in impedance at early switch-on19,20,25. Saunders et al. and Hu et al. have indicated that different electrode array designs show different trends in longitudinal impedance changes over time20,22. To the best of our knowledge, this is the first study to report the time course of the postoperative impedance levels for a slim modiolar electrode array.
The second important finding of this study was that the significantly lower electrode contact impedance in the early switch-on group compared to the conventional switch-on group lasted nearly two months. Notably, in the early switch-on group, the impedance of all electrode contacts gradually increased during the first month after device activation. This gradual increase observed in our study was in sharp contrast with previous reports in which electrode impedance decreased gradually with the use of electrical stimulation7,21,22,26. This discrepancy is presumed to stem from the degree of tissue that covered the electrode when electrical stimulation was initiated. After implantation, the adsorption of protein onto the electrode surface and cell adherence to a protein layer are elicited as a foreign body immune response27. An in vitro study showed that the protein layer and cell growth increased electrode impedance28 and also revealed no significant stimulation-induced impedance changes in clean electrodes without cell coverage29. In addition, in both an in vivo study7 and a clinical study26, the reduction in impedance caused by stimulation was found to be larger when the impedance levels at 2 weeks after surgery were high. The effects of an inflammatory response on stimulation-induced changes in impedance have also been clinically studied. Tykocinski et al. reported an increase in polarization impedance up to 2 weeks after implantation caused by changes in the electrode-electrode interface, and a reversal of those impedance changes once electrical stimulation was initiated6. Previous studies have suggested that electrical stimulation may cause protein desorption and/or cell loss via electroporation over the electrode7,28, while it may have no effect on fibrous tissue growth around the electrode6,9. Although impedance increased after early activation, the value measured at one month was significantly lower (p < 0.001) than the value measured during surgery. In addition, impedance at 1 month was significantly lower (p < 0.001) in the early switch-on group than in the conventional switch-on group. It is likely that the extent of the protein layer and cell growth in the early switch-on group continued to be lower than that in the conventional switch-on group for nearly 2 months. Thus, these findings indicate that early activation of the device partially prevented the increase in impedance associated with electrochemical reactions on the electrode surface. Empirically, providing low impedances in the early switch-on group at an early post-implant period was particularly beneficial for CI programming in a subgroup of patients who had abnormal findings on preoperative magnetic resonance imaging such as cochlear nerve deficiency (CND) or in elderly patients with a significantly smaller spiral ganglion neuron population. Indeed, in a bid to achieve auditory perception, CI recipients with CND required higher electrical charge per unit phase than those with normal anatomy30. In many cases, lowering impedance with the early switch-on protocol in such patients made it possible to reduce the number of out-of-compliance electrodes and allow for increased electric dynamic ranges from the early programming sessions.
Interestingly, the current study showed that impedance values stabilized much earlier in the early switch-on group than in the conventional switch-on group. For the patient group that underwent early switch-on, the measured values remained as low as 8.91 ± 0.15 kΩ (MP2) in period 3 (3–4 weeks after surgery) and stabilized thereafter. Conversely, in the conventional switch-on group, impedances decreased significantly from 14.44 ± 0.2 kΩ (MP2) in period 2 to 8.3 ± 0.13 kΩ (MP2) in period 6, and then remained relatively stable. Although the differences in impedance between the two groups were not significant for period 5, impedance remained slightly higher in the conventional switch-on group than in the early switch-on group (Figure 2). Among the impedance components, the access resistance is generally determined by the electrolyte resistivity around the electrode and can be increased by the development of a fibrous tissue sheath. Since previous studies have indicated that the access resistance was not affected by electrical stimulation7, we speculated that changes in this impedance component over time may not be different between the two groups. Therefore, the lack of a significant difference in the impedance between the two groups in period 5 suggests that the reversible component, such as polarization impedance, in the conventional group was reduced after 2 months of electrical stimulation. As described above, assuming that early activation has the effect of preventing fluctuation of polarization impedance, when stabilization of the impedance had occurred as early as period 3, access resistance also did not increase 3–4 weeks after surgery. One possible factor that could lead to the early stabilization of access resistance is the electrode array type. The current study used a slim modiolar electrode array, which is designed to cause minimal intracochlear trauma during insertion and can be placed in a position close to the modiolus through a high degree of pre-curvation. It is reasonable to presume that less traumatic insertion and closer perimodiolar proximity may lead to the development of less fibrous tissue around the electrode, especially at the medial aspect of the electrode array, where the electrode contacts are facing and electrical stimulation does occur. We further analyzed whether the amount of time spent wearing the CI device was associated with a difference in impedance evolution within the population in the early switch-on group. It was found that when the initial activation was performed early, the frequency of CI device usage did not significantly affect the impedance evolution. Based on this consistent result of early stabilization of impedances using the slim modiolar electrode with the early switch-on protocol, a stable map was usually achieved 3–4 weeks after activation of the device. As a result, early switch-on of the CI device not only provided a nonstop rehabilitation program, but also decreased the number of hospital visits for CI fitting.
Previous studies have identified that the time course of impedance values differ according to the type of electrode array, and even within the electrode array the trend of impedance changes varies across the electrode array18,20-22. However, longitudinal changes in impedance over the long-term for slim modiolar electrode arrays have not been previously investigated. In the present study, electrode impedance values measured intraoperatively were highest for basal electrodes and lowest for apical electrodes. This could be due to the presence of air bubbles around the basal electrode contacts, which is related to the characteristic insertion technique of the slim modiolar electrode array31. Since this electrode array is loaded into the external polymer sheath before insertion, air trapped inside this sheath is commonly inserted into the cochlea with the electrode. Although impedance did not differ significantly across the array, this trend was reversed in both groups in period 2 (Fig. 4). On average, the impedance of the apical electrodes was higher than that of the middle and basal electrodes. This may be related to the differences in the geometric surface area of the electrode contacts across the array, which gradually decreases in an apical direction (CI532, 0.19 mm2–0.16 mm2; CI632, 0.16 mm2–0.15 mm2). It has been reported that impedance is inversely related to the geometric surface area of the electrodes6. Figure 4 shows that this trend is reversed again about two months after surgery, with the basal electrodes having the highest value. This may be due to the combination of the two factors. First, it has been reported that there was more extensive fibrous tissue growth in the basal turn of the cochlea compared to the apical turn after implantation9,26. Second, the distance of the electrode in contact with the modiolar wall of the scala tympani was different. In this study, we used the round window insertion technique, in which the electrode was placed in a more medial position within the proximal basal turn by the crista fenestra at the inferior aspect of the round window compared to the electrode inserted via cochleostomy32. Although a favorable perimodiolar position of the electrode can be achieved by the slim modiolar electrode via the round window, a previous histological study has demonstrated that the distance of the electrode contact to the modiolar wall in the basal turn was more variable and farther apart than that in the apical and middle turns33. Thus, the impedance of the basal electrodes may be more vulnerable to the development of fibrous tissue because of its lateral placement.
As the safety and feasibility of the early activation of the CI device have been previously demonstrated, many centers have been encouraged to change their strategy to early switch-on. However, other than the financial benefits of reducing the waiting period and the number of hospital visits34, the functional benefits of early switch-on have not yet been clearly reported. In this study, we found that at the initial switch-on, significantly lower impedance was associated with early switch-on of the slim modiolar electrode than the conventional switch-on, and the effect lasted nearly 2 months. We also observed that impedance stabilization was achieved as early as one month after surgery with an early switch-on strategy, which was 2 months earlier than that with the conventional switch-on strategy. Since sound quality and power consumption can be affected by impedance, stable impedances are essential for efficient and safe electrical stimulation. However, we only investigated the impedance values in this study, which cannot lead to conclusions about the effects of the early switch-on strategy on clinical outcomes. Thus, further research with psychophysical data, including thresholds, comfort levels, and dynamic ranges, will be needed to validate the functional benefits of early impedance stabilization.
The present study demonstrates that when applying the early switch-on strategy as early as 1 d after surgery for the slim modiolar electrode array, a much earlier stabilization of impedance can be achieved compared to the conventional switch-on strategy performed 3–4 weeks postoperatively, potentially reducing the number of required mapping sessions in the early stages of rehabilitation. Additionally, the early switch-on strategy leads to comparatively lower impedance than the conventional switch-on strategy for at least 2 months postoperatively, offering a significant advantage in early mapping.