Intraoperative Impedance Gradient as a Physiological Indicator of Electrode-Neural Interface in Pediatric Cochlear Implantation

Abstract

Objective

To investigate the spatial gradient of intraoperative impedance across the cochlear electrode array in pediatric cochlear implant recipients and assess its potential as a physiological indicator for the electrode-neural interface.

Methods

A prospective observational study involving 56 pediatric patients underwent cochlear implantation with Cochlear Nucleus devices. Intraoperative polarized impedance and electrically evoked compound action potential (ECAP) threshold were recorded across all 1232 electrodes using AutoNRT software. Eight electrodes with open- or short-circuit were excluded, leaving 1,224 for analysis. Impedance values were categorized by cochlear region (basal, middle, apical), and electrodes with elevated impedance (10–20 kΩ) were analyzed for regional distribution and clinical relevance. Data were analyzed for spatial patterns and correlation with the ECAP threshold profiles.

Results

A consistent basal-to-apical increase in impedance was observed (7.7 ± 1.9, 9.2 ± 1.4, 10.8 ± 1.5 kΩ; p<0.001). Impedance and ECAP threshold were weakly correlated (ρ = –0.20, p<0.001; β = –1.26, p<0.001), with a positive association in the apical region (ρ = 0.12, p = 0.048). Electrodes with higher impedance (10–20 kΩ) were less likely to show elevated or absent TNRT (OR = 0.175, p = 0.02). The impedance gradient persisted across age groups and was significantly correlated with ECAP threshold patterns.

Conclusion

Intraoperative impedance monitoring reveals a strong and physiologically consistent gradient, with higher values in apical electrodes. This gradient reflects anatomical and tissue interface variations, which may offer a valuable physiological indicator for intraoperative electrode positioning and neural interface integrity.

1. Introduction

Cochlear implantation is a hallmark achievement in biomedical engineering. It merges advanced neuroprosthetic design with clinical otology to restore auditory function in individuals with severe to profound sensorineural hearing loss (Deep et al., 2019). Early implantation during the critical neurodevelopmental window among pediatric populations harnesses auditory system plasticity, leading to superior outcomes in speech perception, language acquisition, and neurocognitive development (Naik et al., 2021). As universal newborn hearing screening has enabled earlier identification of congenital hearing loss, intraoperative measures have become pivotal in optimizing cochlear implant (CI) placement and functionality (Athanasopoulos et al., 2024).

Traditionally, intraoperative telemetry in CI surgery has been employed to ensure device integrity through impedance testing and to verify auditory nerve activation via Electrically Evoked Compound Action Potential (ECAP) recordings. However, emerging evidence suggests that these measures are more than mere technical checks; they may reflect underlying biological conditions at the electrode-neural interface and offer predictive insight into long-term auditory outcomes (Tanamati et al., 2019).

Electrode impedance, the resistance to current flow at the electrode-tissue interface, is influenced by cochlear fluid conductivity, tissue encapsulation, ossification, and electrode positioning (Saoji et al., 2025). Notably, a spatial gradient of impedance is often observed intraoperatively, typically increasing from basal to apical electrodes. This gradient reflects the anatomical narrowing of the cochlear duct and changing electrode proximity to the modiolus, influencing both the spread of electrical current and the efficiency of neural excitation. Such variation is not random but is shaped by cochlear morphology, fluid composition, and peri-electrode tissue characteristics, suggesting its potential utility as a real-time biomarker for electrode positioning and neural accessibility (Davis et al., 2016).

Complementing impedance telemetry, ECAP measurements captured via Neural Response Telemetry (NRT) platforms, such as AutoNRT, provide objective data on auditory nerve excitability by measuring synchronized neural responses to electrical stimulation (Ries et al., 2024). ECAP parameters such as threshold, amplitude, and latency are integral to understanding the functional status of spiral ganglion neurons (He et al., 2017) and serve as foundations for initial mapping, particularly in preverbal children where behavioral audiometry is not feasible (Gärtner et al., 2021).

Despite their individual utility, the combined interpretation of the impedance and ECAP parameters remains underexplored in intraoperative contexts. The relationship between high impedance values and altered ECAP thresholds or waveform morphologies could provide insights into electrode-tissue distance, fibrotic obstruction, or suboptimal neural targeting. Conversely, electrodes demonstrating low impedance and robust ECAP responses may represent ideal loci for efficient neural stimulation. Investigating these interactions can inform surgical decisions, such as electrode positioning and array insertion depth, and postoperative processor programming and outcome prognostication.

This study undertakes a systematic, electrode-wise analysis of the intraoperative polarized component of impedance, as routinely provided by the cochlear implant clinical software and ECAP parameters across the cochlear array in pediatric CI recipients. By analyzing these metrics in tandem, the research aims to elucidate whether impedance serves as a biologically meaningful and spatially relevant indicator of the electrode-neural interface. Furthermore, it explores whether the integration of these parameters can enhance intraoperative decision-making and potentially serve as a predictor of neural response robustness and future auditory performance. It seeks to bridge the gap between intraoperative measurement and postoperative rehabilitation, advancing the precision and personalization of cochlear implant therapy in pediatric patients.

2. Materials and Methods

2.1. Study design and setting

This prospective observational study was conducted at the Department of ENT, Indira Gandhi Institute of Medical Sciences (IGIMS), Patna, between 2022 and 2025. The primary aim was to evaluate the intraoperative relationship between electrode impedance and ECAP parameters in children undergoing cochlear implantation. The study also explored whether these intraoperative measurements could provide meaningful insights into the functional status of the auditory nerve and assist in clinical decision-making.

2.2. Participants

A total of 56 pediatric patients, all aged 14 years or younger and diagnosed with bilateral severe to profound sensorineural hearing loss, were enrolled in the study. These children had failed to show sufficient improvement with hearing aids even after consistent use for at least six months. All participants were found suitable for cochlear implantation based on detailed audiological testing, including BERA, OAE, and ASSR, and imaging investigations like high-resolution CT and MRI of the temporal bone. Only those children who had a normal or near-normal cochlear structure and a visible cochlear nerve on MRI were considered for implantation. Patients with complete cochlear aplasia, cochlear nerve deficiency, inner ear malformations incompatible with electrode insertion, active ear infections, or neurological syndromes known to limit CI outcomes were excluded from the study.

2.3. Ethical considerations

The study received ethical approval from the Institutional Ethics Committee of IGIMS, Patna, with letter no. 965/IEC/IGIMS/2023. Patient confidentiality was maintained throughout the study. Consent forms and patient information sheets were provided in both English and Hindi to ensure informed participation by all families involved. Written informed consent was obtained from the parents or guardians of all participants.

2.4. Procedure

Each patient underwent cochlear implant surgery using a standard post-auricular approach with cortical mastoidectomy and posterior tympanotomy to access the round window niche. The electrode array used was the Cochlear™ Nucleus CI422 slim straight electrode array from Cochlear Ltd., Sydney, Australia, inserted fully into the scala tympani either through the round window or via a carefully placed cochleostomy, depending on the child’s cochlear anatomy. After insertion, the internal receiver-stimulator unit was secured in a bony well in the mastoid region. Intraoperative electrophysiological testing was conducted immediately after electrode insertion using the AutoNRT module within the Cochlear Custom Sound EP 6.0 software. This software automatically records two types of data for each of the 22 electrodes in the array.

2.5. Data collection

First, electrode impedance was measured to assess the flow of electrical current between the implant and the surrounding cochlear tissue. Impedance values were recorded in kilo-ohms (kΩ) and helped determine whether each electrode was functioning normally.

We excluded eight electrodes with open and short circuits from the analysis; consequently, 1224 of the 1232 electrodes (56 patients × 22 electrodes) were included in the final analysis.

2.5.1. Impedance measurement

The polarized component of impedance was recorded to evaluate electrode integrity and infer the effectiveness of current flow between the implant and the surrounding cochlear tissue. Impedance values between 1–20 kΩ are considered within the normal range according to Cochlear Nucleus device criteria (Cochlear Ltd., 2019).

Electrodes with very high values (above 20 kΩ) were considered “open circuits” (Cochlear Ltd., 2019)., indicating a possible extracochlear electrodes, trapped air bubbles, ossification, or fibrotic tissue around the electrode, and very low values (below 1 kΩ) suggested “short circuits” possibly due to an unintended electrical coupling between electrode contacts, insulation failure, or internal device malfunction were excluded from the data analysis. Impedance telemetry was performed for all 1232 electrodes in monopolar MP1+2 configuration using the built-in clinical protocol. The impedance was calculated using biphasic pulses at a fixed current level of 80 current units (cu) and pulse width of 25 µs/phase. Impedance values were computed automatically by the implant system based on voltage at the electrode–tissue interface.

2.5.2. ECAP Measurement

Second, intraoperative ECAP recordings were obtained using the AutoNRT module in Custom Sound EP (Cochlear Ltd.). Recordings were performed while the patient remained under general anaesthesia. The forward-masking subtraction paradigm was employed to minimize stimulus artifact by alternating masker–probe and probe-only traces.

The following stimulus and recording parameters were used for intraoperative ECAP acquisition: biphasic pulse width of 25 µs/phase, stimulation mode MP1+2, stimulation rate 250 Hz, masker–probe interval 122 µs, gain 50 dB, and artifact reduction achieved through averaging of 35 sweeps.

Although the AutoNRT module provides amplitude and latency information, only threshold ECAP threshold values (TNRT) were analyzed in the present study, in accordance with the primary study objective.

The electrodes were grouped into three regions for analysis: basal (electrodes 1–5), middle (9–14), and apical (18–22). This allowed for comparison of how impedance and ECAP responses varied across the spiral of the cochlea.

2.6. Data Analysis

Data management and statistical analyses were performed using Microsoft Excel (v16.89; Microsoft Corporation, Redmond, Washington) and XLSTAT Life Sciences (v2024.4.0.1424 Addinsoft, Paris, France). Descriptive statistics such as mean, range, and standard deviation were calculated for impedance and ECAP values in each cochlear region. To examine relationships between these parameters, statistical tests were performed. Spearman’s rank correlation was used to assess whether impedance was associated with basal, mid, apical, and average ECAP thresholds. The higher impedance within normal range were analyzed to see their effect on ECAP threshold using binary logistic regression. Chi-square tests and post-hoc analysis using adjusted residuals helped determine whether specific electrode categories (basal, mid and apical) were more likely to produce recordable ECAP responses. Linear regression analysis was also applied to explore whether impedance values could predict ECAP behavior across electrodes. For all tests, a p-value less than 0.05 was considered statistically significant.

3. Results

A total of 56 pediatric patients underwent cochlear implantation during the study period. The age of the participants ranged from 1 to 14 years, with a mean age of 5.009 years and a standard deviation of 3.214 years, indicating a predominance of early implantation. Specifically, 39 children (69.64%) were under six years of age, underscoring the emphasis on early auditory intervention (Fig.1). Among them, 38 (68%) were male and 18 (32%) were female.

Figure 1.

Histogram showing the age distribution of 56 cochlear implant recipients included in the study. Most participants were between 2 and 6 years of age, with fewer cases observed in older age groups.

All surgeries were completed successfully without intraoperative complications. In every case, full insertion of the electrode array was achieved and confirmed intraoperatively using depth markers on the electrode array under direct visualization. This was later confirmed postoperatively using transorbital X-ray imaging. Impedance and TNRT were not used to determine insertion depth.

3.1. Impedance Distribution

The mean intraoperative impedance values revealed statistically significant gradient across the cochlear spiral. For the basal electrodes, the average impedance was 7.7 ± 1.9 kΩ; for the middle electrodes, the value was 9.2 ± 1.4 kΩ; and for the apical electrodes, the impedance increased further to 10.8 ± 1.5 kΩ (Kruskal–Wallis test, K = 235.5, p < 0.001; post-hoc comparisons: apical vs basal, apical vs mid, and mid vs basal, p < 0.0001). Importantly, this basal-to-apical impedance gradient was observed regardless of patient age, gender, or laterality of implantation, suggesting that it reflects anatomical and physiological properties of the cochlea rather than inter-individual variability (Fig.2).

Figure 2.

Impedance Profile Across Electrode Array. Line graph showing mean electrode impedance (in kΩ) across individual cochlear implant channels, categorized into basal (CH 01–05), mid (CH 09–14), and apical (CH 18–22) segments. Error bars indicate standard error. CH: Channel.

3.2. TNRT Distribution

The mean intraoperative TNRT values displayed a clear and statistically significant gradient across the cochlear spiral. For the basal electrodes (E1–E5), the average TNRT was 187.6 ± 14.4 cu; for the mid electrodes (E9–E14), the value was 188.4 ± 15.0 cu; and for the apical electrodes (E18–E22), the TNRT recorded further to 168.9 ± 15.3 cu (K = 198.8, p < 0.001; post-hoc comparisons: apical vs basal and apical vs mid, p < 0.0001) (Fig.3).

Figure 3.

Threshold Neural Response Telemetry (TNRT) Across Electrode Array. Line graph showing mean TNRT (current units) across cochlear implant channels, categorized into basal, mid, and apical segments.

3.3. ECAP Threshold (TNRT) and Impedance Correlation

Threshold Neural Response Telemetry (TNRT) was successfully recorded in 1212 of 1224 electrodes (99.02%). The remaining 12 electrodes failed to yield clear ECAP waveforms, and a majority of these (58.33%) were located in the apical third of the array, where higher impedance values were more frequently observed.

The chi-square test yielded significant association between electrode segment and TNRT categories among electrodes demonstrating normal impedance values (χ² (6) = 12.74, p = 0.047). Table 1 presents the data showing that electrodes with normal impedance predominantly exhibited normal TNRT values, with 1204 out of 1224 electrodes falling into this category. A small number of these electrodes had high (6), low (2), or unrecordable (12) TNRT readings (Park et al., 2023). The adjusted residuals from post hoc analysis revealed that low TNRTs were significantly more frequent in the apical segment (adjusted residual = +2.07, p < 0.05), indicating greater neural responsiveness in this region.

Table 1. Contingency table showing the association between Impedance and Threshold Neural Response Telemetry (TNRT) categories. There is a statistically significant association between electrode segments and TNRT categories with normal impedance.

Impedance	TNRT (no. of electrodes, n=1232)
	High	Low	Normal	Not Recordable	Total	χ², p value
Normal	6	2	1204	12	1224	12.74, 0.047
Basal	2	0	437	7
Mid	4	0	382	2
Apical	0	2	385	3
Open Circuit	0	0	0	4	4
Short Circuit	0	0	4	0	4

The heatmap in Fig. 4 illustrates the correlation between impedance and TNRT across basal, mid, and apical segments as well as for average values. Spearman’s correlation revealed region-dependent effects: a weak positive correlation in the apical region (ρ = 0.12, p = 0.048), no significant association in the mid region (ρ = 0.07, p = 0.21), and a weak inverse association in the basal region (ρ = –0.15, p = 0.01). The overall impedance–TNRT association across all electrodes showed a weak negative trend (ρ = –0.20, p < 0.001), which was supported by linear regression (β = –1.26, 95% CI: –1.74 to –0.77, p < 0.001). Although statistically significant, impedance explained only a small proportion of TNRT variance (R² = 0.021), suggesting that other physiological factors play a larger role in neural response variability.

Figure 4.

Heatmap displaying the association between impedance and TNRT (Threshold Neural Response Telemetry) across basal, mid and apical categories as well as average values. Green indicates observed positive correlation, while red indicates negative correlation. Strongest positive associations were observed in the apical region.

Binary logistic regression analysis was performed to evaluate the effect of elevated electrode impedance on the likelihood of obtaining high or non-recordable ECAP responses. The dependent variable represented the presence or absence of high or non-recordable TNRT (>236 cu), while the predictor variable was higher impedance (10–20 kΩ) with low-mid range impedance (1–9 kΩ) serving as the reference category. The model indicated that electrodes with higher impedance had an 82.5% lower likelihood of exhibiting high or non-recordable ECAPs compared with those having normal impedance (β = –1.8, OR = 0.175, 95% CI: 0.04–0.76, p = 0.02). Although statistically significant, the model showed a small explanatory effect (R² = 0.046), reinforcing that impedance alone accounted for a limited proportion of the variance in TNRT outcomes.

4. Discussion

This study offers an in-depth electrode-wise evaluation of intraoperative impedance and ECAP parameters in a pediatric cochlear implant cohort, presenting one of the largest region-specific datasets. The results reinforce and extend previous research suggesting that impedance telemetry, when interpreted alongside ECAP characteristics, holds significant diagnostic and prognostic value during cochlear implantation. Beyond its traditional role as a technical check of electrode function, impedance demonstrates consistent biological relevance in reflecting tissue-electrode interaction, cochlear health, and potentially auditory nerve health.

4.1. Impedance as a Physiological Marker

The observed statistically significant increase in impedance from the basal to apical cochlear regions in our study is consistent with earlier reports by Wagner et al. (2023), who demonstrated spatial impedance gradients that correlated with cochlear fluid distribution and electrode proximity to the modiolus. Similarly, Ohlemiller and Gagnon (2004) observed this gradient in adult recipients and attributed it to anatomical narrowing and denser tissue encapsulation at the apical end of the cochlea. The apical turn, due to its tighter curvature and longer insertion path, is also more susceptible to fibrosis and trauma-related cellular deposition, which are known to increase impedance (Geerardyn et al., 2023).

The uniformity of this gradient across our entire pediatric cohort, regardless of age, gender, or laterality, suggests that it is likely rooted in consistent anatomical properties rather than patient-specific variability. This highlights impedance as a spatial biomarker of cochlear microstructure and supports its real-time use in assessing insertion depth, trauma likelihood, and array-tissue interface quality.

4.2. Impedance and ECAP Interdependence

One of the central findings of this study is the significant correlation between elevated impedance values and higher TNRT values, especially in the apical region. This supports earlier work by Li et al. (2017), who reported similar trends in adult users.

In pediatric studies, He et al. (2017) emphasized the role of ECAP threshold patterns in assessing auditory nerve responsiveness, but few have correlated this data with impedance telemetry. Although all analyzed impedances were within the functional range (<20 kΩ), higher values (10–20 kΩ) were significantly associated with reduced likelihood of TNRT presence (OR = 0.175, p = 0.02). This likely reflects device-level electrical constraints rather than neural degeneration. Specifically, as impedance increases, so does the voltage required to deliver current (V = I × R). If this voltage exceeds the stimulator’s compliance limit, current delivery becomes clipped or insufficient, leading to absent ECAPs or TNRTs despite intact neural pathways (Saoji et al., 2021). While impedance and TNRT correlated, this association likely reflects shared cochlear-region factors rather than a causal relationship (Fatani et al., 2024). Moreover, standard clinical impedance telemetry lacks sensitivity to detect inter-electrode faults or partial shorts, as it does not assess current spread or channel interaction. Advanced diagnostics such as electrical field imaging or scalp potential mapping are required in such cases (Eitutis et al., 2022).

This relationship between the physical (impedance) and physiological (ECAP) metrics has also been emphasized in more recent studies using modern imaging and modeling. For instance, Aebischer et al. (2021) combined impedance telemetry with 3D cochlear models and found that regions of high impedance often overlapped with post-insertion fibrosis or trauma zones. Our findings similarly imply that poor electrode-neural coupling, reflected by both high impedance and poor ECAP, is more common in the apical cochlear turns, where mechanical insertion is more challenging.

4.3. Clinical Utility and Surgical Decision-Making

The clinical implications of integrating impedance and ECAP data during surgery are multifaceted. First, detecting electrodes with poor functional coupling intraoperatively allows immediate intervention, such as modifying insertion depth, withdrawing and reinserting the array, or marking electrodes for postoperative deactivation. Studies by Warren and Atcherson (2023) have suggested that early identification and exclusion of non-functional electrodes lead to more efficient and comfortable processor programming in children.

Second, among electrodes with normal impedance, 98.4% (1204/1224) showed normal TNRTs, with low thresholds occurring more frequently in the apical segment (adjusted residual = +2.07, p < 0.05), consistent with the established apical–basal gradient in cochlear neural responsiveness (Cosetti et al., 2012; Bayrak et al., 2019). The significant association between electrode segment and TNRT category (p = 0.047), together with region-specific impedance–TNRT relationships (weakly positive apically, non-significant mid-array, weakly negative basally), indicates that neural responsivity follows a systematic tonotopic pattern rather than random variation. The greater prevalence of low TNRTs in the apical region may reflect improved spiral ganglion survival and closer electrode–neuron proximity in this region (Fatani et al., 2024). In contrast, the inverse basal trend may be related to insertion-related trauma or broader current spread, both of which can reduce neural coupling efficiency (DeVries et al., 2016). Collectively, these findings support the combined use of impedance and TNRT as physiologic indicators of electrode–neuron interface quality and as practical guides for intraoperative mapping and early programming, particularly in preverbal pediatric recipients (He et al., 2017). In preverbal children, where behavioral responses are unavailable, this real-time parameter combination can serve as a foundation for early speech processor mapping and help anticipate regions requiring higher current levels or special stimulation strategies.

Recent advancements have also examined the dynamic nature of impedance over time. Research by Schraivogel et al. (2023) observed that impedance typically decreases after activation, possibly due to perilymphatic stabilization and electrical conditioning. Our study focused on intraoperative measurements, but the elevated impedance observed in apical electrodes may represent transient early-phase tissue responses that resolve with stimulation.

Moreover, studies using machine learning and impedance-ECAP datasets, such as Mo et al. (2025), are exploring predictive modeling for cochlear implant outcomes. These approaches depend on the kind of granular, region-specific data that our study provides, and thus our findings may contribute to the development of AI-driven decision tools in future CI programming platforms.

4.4. Limitations

While the study has considerable strengths in sample size, electrode-level granularity, and focus on the pediatric population, certain limitations should be acknowledged. First, the lack of long-term postoperative speech perception or auditory outcome data precludes definitive conclusions on the predictive utility of impedance-ECAP profiles. Second, this study exclusively used Cochlear Nucleus® devices, which limits applicability to systems with different electrode array designs, telemetry algorithms, and stimulation strategies (e.g., Med-El or Advanced Bionics).

Third, we did not utilize postoperative imaging (e.g., cone-beam CT or rotational tomography) to confirm electrode position relative to the modiolus, nor did we measure scalar location, which could refine our understanding of impedance variation. Finally, while AutoNRT provides automated ECAP threshold estimation, manual waveform validation may offer better insight into subtle waveform morphology changes.

4.5. Future Directions

Future studies should aim for longitudinal follow-up correlating intraoperative telemetry with auditory development metrics (Skidmore et al., 2023) such as speech recognition thresholds, Categorical Auditory Performance (CAP) scores, or language comprehension. Comparative analyses across different implant platforms and insertion techniques (round window vs. cochleostomy) would also enhance generalizability. Additionally, integrating imaging data and machine learning models with intraoperative metrics may offer new dimensions for predictive analytics in cochlear implant programming.

5. Conclusion

Intraoperative impedance measurements reveal a predictable and clinically informative gradient along the cochlear spiral in pediatric cochlear implant recipients. The basal-to-apical rise in impedance correlates significantly with ECAP parameters, supporting the role of impedance as a real-time physiological indicator of electrode-neural interface quality. Integrating impedance analysis into surgical protocols may enhance electrode placement strategies and auditory outcome prediction.

Biography

E-mail: maruti.nandan20@gmail.com E-mail: amit.ks.sharma@gmail.com E-mail: rksent5@gmail.com

Funding Statement

This research did not receive any specific grant from funding agencies in the public, commercial or not-for-profit sectors.