Residual hearing and sound quality in cochlear implant patients

Abstract

Introduction:

Cochlear implantation (CI) improves speech perception in adults with hearing loss. However, patient-reported sound quality varies widely and correlates poorly with objective outcomes. The impact of residual acoustic hearing trajectories on subjective sound quality remains incompletely understood.

Methods:

In this prospective cross-sectional study, adult cochlear implant recipients (n = 31) were stratified into three groups based on low-frequency hearing: no residual hearing at baseline, preserved residual hearing following CI, and loss of residual hearing post-operatively. Subjective outcomes were assessed using sound quality and hearing questionnaires, compared with objective speech perception measures. Group differences were evaluated with covariate-adjusted analyses accounting for age, duration of deafness, and time since implantation.

Results:

Overall satisfaction and functional hearing scores were similar between groups. However, patients who lost residual hearing demonstrated lower sound quality ratings compared to those without baseline residual hearing, with trend-level differences persisting after adjustment. Patients with preserved hearing did not demonstrate superior subjective speech perception despite comparable objective gains, suggesting a dissociation between measured performance and perceived benefit. Objective speech recognition improvements were similar across all groups and showed weak associations with patient-reported outcomes. Among patients with preserved hearing, those using electroacoustic stimulation demonstrated descriptively improved sound quality, though utilization was limited.

Conclusion:

In this exploratory, cross -sectional analysis of a small cohort, loss of residual hearing following CI may negatively affect perceived sound quality, while preservation alone does not guarantee improved subjective outcomes without optimized EAS integration. These findings support the importance of incorporating patient-reported measures alongside objective testing and suggest that underutilization of electroacoustic stimulation may represent a modifiable gap in care. Given the design and small sample size, results warrant confirmation in larger, prospective studies.

Introduction

Cochlear implantation (CI) significantly improves speech perception in adults with moderate-to-severe hearing loss, with most patients demonstrating meaningful gains in word and sentence recognition compared to their preoperative performance with hearing aids (Dornhoffer et al., 2021; Hoppe et al., 2021). These improvements are substantial, with mean increases of 34-46% across different speech recognition tasks and remain stable over time (Carlson, 2020). CI candidacy has progressively expanded to include patients with residual low-frequency hearing, driven by evidence demonstrating significant benefits even in patients with better preoperative hearing (Tsai Do et al., 2024; Dietz et al., 2022). This expansion reflects a shift from viewing CI as a last resort for profound bilateral deafness to recognizing it as an effective intervention for a broader spectrum of hearing loss. However, patient-reported outcomes following CI remain variable and demonstrate weak to negligible correlations with objective audiologic metrics (Dorismond et al., 2023; McRackan et al., 2018). This dissociation indicates that speech recognition scores poorly predict sound quality and quality of life (McRackan et al., 2019; Li et al., 2019).

Modern surgical CI techniques, such as round-window insertion, have allowed for improved preservation of residual acoustic hearing post-implantation. In such patients, electroacoustic stimulation (EAS) combines electric stimulation from a cochlear implant for high frequencies with acoustic amplification of preserved low-frequency hearing in the same ear, providing auditory cues not reliably transmitted by electric stimulation alone (Irving et al., 2014; Reinhart et al., 2024). This hybrid approach demonstrates superior outcomes for speech perception in noise, spatial hearing, sound quality, and music appreciation compared to electric-only stimulation (Dalbert et al., 2016).

Even with advancements in surgical techniques, many patients still experience loss of residual hearing as a complication of surgery (Tsai Do et al., 2024). This presents three clinically relevant trajectories for CI recipients: (1) residual hearing preserved post-implant, (2) residual hearing lost because of surgery, and (3) no residual hearing at baseline. Existing literature largely focuses on hearing preservation rates and speech recognition outcomes across these subgroups, while there remains a limited understanding of how these different trajectories affect subjective sound quality and patient experience of CI hearing (Calvino et al., 2016).

Sound quality may be a particularly meaningful outcome as it reflects how patients experience hearing in everyday, real world listening conditions, such as conversation in background noise, music, traffic, which discrete, quiet-room speech recognition tests do not fully capture. Emerging evidence suggests sound quality may be a better predictor of quality of life than objective speech recognition scores (McRackan et al., 2018, 2019; Kral and O'Donoghue, 2010). CI sound differs inherently from acoustic hearing, particularly in its limitations in spectral resolution, temporal fine structure, and dynamic range (Limb and Roy, 2014; Plant et al., 2016). Consequently, prior acoustic hearing experience may influence speech perception outcomes after CI. Previous studies have demonstrated that individuals with shorter durations of profound hearing loss and those who acquire profound hearing loss post-lingually (vs. pre-lingually) report significantly improved word recognition and auditory memory (Nicastri et al., 2025; Imsiecke et al., 2020).

Individuals who lose residual hearing after CI must transition from an acoustic hearing experience (often aided by hearing aids prior to CI) to a purely electric hearing experience post-implantation. We hypothesize that these individuals may experience a perceptual mismatch due to inherent differences in the mechanisms of sound transduction. Conversely, individuals with no residual hearing lack a recent acoustic reference and may therefore rate cochlear implant sound quality more favorably. Finally, individuals with preserved hearing post-implantation may experience the benefits of bimodal EAS, but may also experience interference in certain conditions, such as noisy environments (Imsiecke et al., 2020).

Many studies rely on speech perception tests (e.g., CNC, AzBio) as measures of success post-CI, but these fail to capture the perceptual qualities of sound. Patient-reported outcome measures for CI are essential for characterizing qualities of hearing that cannot be captured by objective tests; however, limited data, as well as measures, exist, with even less understanding on the impact of residual acoustic hearing has on these measures.

This study aims to evaluate the relationship between residual hearing status (preserved, lost, or absent) in relation to CI and subjective sound quality. Secondarily, subjective outcomes among these subgroups will be compared with objective outcomes, as will the role of EAS use within the preserved residual hearing subgroup. We hypothesize that patients who lose residual hearing will report worse sound quality than those with absent residual hearing or preserved hearing post-implantation. We also hypothesize that there will be a weak or absent correlation between objective and subjective hearing outcomes.

Materials and methods

Study design and participants

A prospective cross-sectional survey study of adult CI patients was conducted at Thomas Jefferson University Hospital from 11/1/2025 to 3/15/2026 after IRB approval (#2025-0217). Patients with unilateral or bilateral CI who had undergone CI at least 3 months prior to completion of surveys with an unaided post-operative audiogram were included. Patients without a documented post-operative unaided audiogram but with a preoperative unaided audiogram demonstrating no residual hearing were also included for the no residual hearing baseline cohort.

Residual hearing classification

Residual hearing status was defined using the implanted ear's low-frequency pure tone average (LFQ PTA), calculated as the mean of air conduction thresholds at 250, 500, and 1,000 Hz. A threshold of 80 dB HL was used to define the presence of residual hearing, consistent with prior literature defining residual hearing as pure-tone thresholds better than 80 dB (Reinhart et al., 2024; Helbig et al., 2016; Gantz et al., 2016). For unilateral CI patients, LFQ-PTA was derived from the implanted ear, and for bilateral CI recipients, LFQ-PTA was computed as the mean of both implanted ears since one survey was completed per participant with inability to distinguish between sides per participant.

Patients were classified into three mutually exclusive groups based on the pre- and subsequent post-operative implanted(s) ear LFQ-PTA: 1) Residual → Preserved (RP), 2) Residual → Lost (RL), and 3) No residual at baseline (NR). Among the RP group, patients were also grouped by utilization of electroacoustic stimulation (EAS) at the time of survey completion.

Outcome measures

Subjective sound quality was assessed using two surveys. The non-validated Subjective Quality Descriptors (SQuaD) Tool (MED-EL Corporation, Durham, US) consists of four questions regarding satisfaction, 14 questions regarding sound quality, and 12 questions regarding pitch, all measured on 1-5 Likert scale. Eight negatively described sound quality descriptor items (artificial/robotic, harsh, thin/hollow, nasally, muffled, muddy/smeared, distant/far away, and reverberant) were reverse-scored (6 – x) prior to computing the 14-item mean, such that higher scores consistently reflect better perceived sound quality. All 12 pitch descriptor items were similarly reverse-scored (6 – x) to yield a composite pitch mean in which higher values indicate more natural pitch perception. The second survey completed was the Speech, Spatial, and Qualities questionnaire (SSQ-12), (Gatehouse and Noble, 2004) with five questions regarding speech, three questions regarding spatial sound, and four questions regarding sound qualities, all measured on a 1-10 scale, with 10 indicating no difficulty. Post-operatively, the CI Quality of Life-35 (CIQOL-35) (McRackan et al., 2022) was assessed in patients and used as a secondary outcome and comparator instrument to examine the convergent validity of the surveys.

Additional variables collected from the electronic health record include temporal variables (time since surgery to date of survey completion, duration of deafness prior to CI, age at CI), device and accessory use (partner/mini mic use, contralateral device type), audiological variables including speech perception testing from pre- and post-operative testing, and demographics.

Statistical analysis

Statistical analyses were conducted in Python (version 3.13). Continuous variables are reported as mean (SD) and median. Bivariate associations between continuous predictors (age at surgery, time since surgery, duration of deafness, pre-operative LF-PTA, and CNC % change) and outcomes were assessed using Spearman rank correlations across the full cohort of patients (n = 31) and separately within each hearing preservation group. Hearing preservation group membership was coded as binary variables (1 = member, 0 = not) and included as predictors in the full-cohort correlation matrix.

To examine group differences among hearing preservation groups, known confounders were accounted for using covariate adjusted residuals for each of the 11 outcome measures (four independent questions regarding satisfaction [pleasantness, clarity, naturalness, overall satisfaction], composite satisfaction, composite pitch, composite sound quality descriptors, SSQ-12 spatial, SSQ-12 speech, SSQ-12 qualities, SSQ-12 composite). For each outcome, a linear regression model was fit with age at CI, time since surgery, and duration of deafness as predictors. The residuals from each model which represent variance in the outcome not explained by the three covariates were then added back to the grand mean of each outcome, so values remained on the original scale. Adjusted scores were then compared across hearing preservation groups using Kruskal Wallis tests (KW) for group differences and pairwise Mann-Whitney U tests for post-hoc comparisons. Benjamini-Hochberg false discovery correction was applied within each outcome to attempt to account for the multiple predictors tested.

Within the RP, differences between EAS and non-EAS patients were characterized descriptively using Cohen's d; no inferential tests were conducted for this comparison due to the small EAS subgroup size (n = 3). A significance threshold of α = 0.05 was used for all inferential tests, with trends noted at p < 0.10. Given the large number of outcome measures and subgroup analyses, these comparisons should be regarded as exploratory and hypothesis-generating.

Results

Thirty-one adult CI recipients with complete SQuaD survey data, and of those, thirty patients who also completed SSQ-12 were included. Demographics and clinical characteristics stratified by residual hearing group are presented in Table 1. The cohort was 55% female (n = 17), with a mean age at implantation of 68.8 years (SD 9.8). Mean duration of deafness prior to CI was 20.3 years (SD 15.0). The majority of patients had a contralateral hearing aid with bimodal pairing achieved (n = 26; 83.4%). Nine patients (29%) were using a Partner/Mini Mic remote microphone accessory at the time of survey completion. Three (27.3%) patients with residual hearing were using EAS mapping. Device manufacturers included Advanced Bionics (n = 18, 55%; 13 MidScala, 5 SlimJ), MED-EL (n = 8, 23%; 3 Flex28, 3 FlexSoft, 2 Flex34), and Cochlear Americas (n = 5, 13%; 3 Slim Modiolar, 1 Slim Straight, 1 Contour Advance). Post-operative CIQOL-35 data were available for 17 patients (mean global converted score 50.4, SD 14.1).

Using the implanted-ear LFQ-PTA with a 80 dB HL cutoff, patients were classified into three groups: RP (n = 9; pre-operative LFQ-PTA 47.4 ± 10.2 dB HL, post-operative LFQ-PTA 74.7 ± 8.5 dB HL), RL (n = 12; pre-operative LFQ-PTA 58.3 ± 21.6 dB HL, post-operative LFQ-PTA 102.0 ± 1.6 dB HL), and NR (n = 10; pre-operative LFQ-PTA 102.5 ± 8.3 dB HL). Pre- and post-operative LFQ-PTA differed significantly across groups [both p < 0.001, 95% CI effect size (0.45, 0.81) and (0.05, 0.57) respectively]. The three groups were well-matched on sex, race, ethnicity, duration of deafness, and time since surgery (all p > 0.05; Table 1). Age at CI differed across groups [p = 0.04, 95% CI effect size (0.00, 0.41)], with the RP being oldest (mean: 72.7 ± 8.4 years) and the NR group youngest (mean: 62.5 ± 8.7 years).

SQuaD and SSQ-12 outcomes

Across the full cohort (n = 31), mean SQuaD scores reflected moderate to high satisfaction, comprised of pleasantness 3.87 (SD: 1.15), clarity 3.58 (SD: 1.12), naturalness 3.68 (SD: 1.35), and overall satisfaction 3.94 (SD: 1.18) (Table 2). On the sound quality descriptor subscale, patients rated the sound as moderately good quality overall (SQ Mean 14-item: mean 3.43, SD: 0.91), with the “comfortable” descriptor rating highest on average (mean: 3.77), and “cartoon-like Darth Vader” lowest (mean: 1.53). On the pitch subscale, the most prevalent descriptor was “Tinny” (mean: 2.62) followed by “Metallic” (mean: 2.61).

Across the cohort (n = 30), mean SSQ-12 subscale scores showed moderate hearing difficulty across the three domains (Table 2). The sound qualities subscale showed the least reported difficulty (mean: 5.54, SD: 1.38), and the speech subscale showed the greatest (mean: 4.26, SD: 2.54), with a graduated trend observed for speech scores with NR median = 4.75 (mean: 5.55, SD: 2.31), RL median = 4.00 (mean: 4.33, SD: 2.34), RP median = 2.50 (mean: 3.02, SD: 2.48).

Residual hearing group comparisons

Across the SQuaD, the median overall satisfaction scores were comparable across hearing preservation groups: RP (median: 4.00), RL (median: 5.00), and NR (median: 4.50), with all pairwise Mann-Whitney comparisons being non-significant (all p > 0.14). Across the sound quality descriptor subscale, NR had the highest median and mean score (median: 4.21, mean: 3.85 ± 1.01), followed by RP (median: 3.64, mean: 3.43 ± 0.71), and RL (median: 2.96, mean: 3.01 ± 0.90) (Table 2). Higher scores reflect characteristics including more natural, full, clear, comfortable, and warm with less artificial, harsh, muffled or reverberant character. Pairwise comparisons showed a trend-level difference between NR and RL (p = 0.095, r = 0.44) on the sound quality descriptor scale. At the univariate level, NR group membership showed a weak non-significant positive association with the sound quality descriptor subscale (ρ = +0.32, p = 0.075), while RL group membership showed a negative association (ρ = −0.29, p = 0.109) (Figure 1). Across the pitch subscale, no significant between-group differences were observed (KW p = 0.435), with mean pitch scores being NR (mean: 4.16, SD: 0.90), RL (mean: 3.86, SD: 1.22), and RP (mean: 3.44, SD: 1.05). All pairwise comparisons were non-significant (NR vs. RL: p = 0.649; NR vs. RP: p = 0.230; RL vs. RP: p = 0.436).

Figure 1

Across SSQ-12 subscales, speech showed a notable pattern across hearing preservation groups, with a stepwise gradient of increasing scores (which indicate decreased difficulty), from RP (median: 2.50), RL (median: 4.00), NR (median: 4.75), though this did not reach statistical significance (KW p = 0.112) (Table 2). RP group membership was negatively associated with SSQ: Speech at the univariate level (ρ = −0.35, p = 0.056). To contextualize SSQ-12 speech findings, objective speech perception outcomes, including CNC and AzBio, were examined across the three residual hearing groups. All three groups achieved clinically meaningful improvements in objective speech perception, with CNC mean change in RP of 31.0 (SD: 20.6), RL of 45.2 (SD: 25.2), and NR of 46.0 (SD: 39.2), KW p = 0.475, and with AzBio mean change in RP of 34.6 (SD: 21.7), RL of 42.5 (SD: 28.4), and NR of 47.0 (SD: 44.9), KW p = 0.854. SSQ: Spatial showed no significant group differences (KW p = 0.945), with median scores of NR = 4.62, RL = 5.00, and RP = 6.00. SSQ: Qualities and SSQ-12 Total did not differ significantly across groups (all KW p > 0.39).

Accounting for variance effects of time since surgery to survey completion, age at implantation, and duration of deafness using linear regression residualization (Figure 2), no statistically significant group differences emerged. A notable pattern was observed in the SQuAD sound quality descriptor subscale, with RL showing lower adjusted scores (mean: 3.07) compared to both NR (mean: 3.77, p = 0.089) and RP (mean: 3.47, p = 0.418). A similar directional trend was observed for clarity, where RL scored the lower (mean: 3.41) compared to NR (mean: 3.90; p = 0.273). For SSQ-12 on the speech subscale, RP had the lowest adjusted scores (mean: 3.24) compared to NR (mean: 4.84, p = 0.073) and RL (mean: 4.73, p = 0.095). For SSQ: Spatial and Qualities, no group differences were observed, though RL and RP numerically outperformed NR (SSQ: Spatial adjusted means: RL = 5.06, RP = 5.64 vs. NR = 4.04; SSQ: Qualities adjusted means: RL = 6.04, RP = 5.10 vs. NR = 4.79; all p ≥ 0.18).

Figure 2

Of the nine patients in RP, three had an active EAS mapping component. These patients demonstrated medium-to-large effect size advantages over non-EAS patients (n=6) on sound quality descriptor ratings (d = +0.75) and perceived clarity (d = +0.73), with a moderate advantage in pitch perception (d = +0.58). Conversely, non-EAS patients rated naturalness higher (d=-0.73) and showed numerically better SSQ-12 Speech scores (d=-0.51), while spatial hearing was comparable between subgroups (d=-0.05). Given the small EAS subgroup size (n=3), these findings are descriptive.

Post-operative CIQOL-35 scores were available in only 17 patients (NR n = 4, RL n = 5, RP n = 8) and did not differ significantly across groups (KW H = 3.24, p = 0.198); however, a numerically descending pattern was observed from NR (mean: 60.1 ± 15.5) to RP (mean: 50.8 ± 6.3) to RL (mean: 41.7 ± 17.5).

Discussion

This study examined the relationship between residual hearing preservation and patient reported sound quality and hearing outcomes in adult CI patients. With the small subgroup sample sizes, findings are exploratory and hypothesis-generating, with proposed mechanisms as explanations.

Sound quality and hearing preservation after implantation

Patients who lost residual hearing demonstrated trend level reductions in sound quality descriptor ratings compared to those with no residual hearing at baseline, and this pattern was maintained after covariate adjustment. It is hypothesized that patients with initial residual hearing who subsequently lose it may experience a perceptual mismatch or adaptation challenge as they transition from some mild acoustic processing to purely electric stimulation. If so, this shift may heighten the difference of CI sound's inherent spectral and temporal distortions (Carlson, 2020). In contrast with the NR group who lack recent auditory memory of acoustic sound for comparison, they may rate sound quality of electric-only stimulation more favorably, as they are comparing it primarily to their pre-implant memory of profound deafness rather than to acoustic hearing. This is consistent with evidence that sound quality perception involves both sensory processing and cognitive factors, including expectations and prior auditory experience. Berg et al. found that better pre-operative sound quality ratings predicted better early 1 month post operative sound quality ratings, suggesting that prior auditory experiences and expectations influence post-operative perceptions. CI patients undergo a variable period of acclimatization to electrical input as well, with sound quality improving over the first 3-6 months as the central auditory system adapts (Berg et al., 2026). The RL group may represent patients at varying stages of this adaptation process, with their auditory memory of acoustic hearing potentially prolonging the adaptation period or influencing their subjective quality ratings. However, because the timing of survey completion relative to surgery varied in this cohort, which may have captured patients at different stages of perceptual adaptation, differences in adaptation stage rather than residual hearing trajectory itself may account for part of these findings, though covariate adjustment analyses attempted to control for this factor.

Additionally, the divergence of the SQuaD sound quality descriptor ratings and SSQ-12: qualities scores in the RL group (Figure 2) may reflect the distinct qualities of each instrument. The SQuaD Descriptor Mean assesses the perceived character of CI-mediated sound like artificial, thin/hollow, richness, muddy/smeared, etc. which the RL group scored lowest. In contrast, SSQ-12: Qualities assesses functional hearing ability in everyday quality-related tasks, including sound segregation, sound identification, and listening effort, domains on which the RL group scored numerically highest.

These observations align with recent evidence that sound quality and not speech recognition is the primary driver of CI quality of life (Berg et al., 2025). RL's decreased sound quality descriptor scores combined with their numerically lower CIQOL-35 global scores, despite equivalent speech recognition gains, suggest that subjective experience of electric hearing is most adversely affected in patients who lost something perceptually meaningful at implantation.

Dissociation between objective and subjective outcomes

The trend toward lower SSQ-12 speech subscale scores in RP despite equivalent objective speech perception outcomes with CNC and AzBio represents a potentially clinically important dissociation between performance-based and patient-reported outcome measures.

One hypothesized potential mechanism contributing to the RP group's lower subjective speech scores relates to electric-acoustic interference in complex listening environments. Research has demonstrated that electric stimulation can mask acoustic hearing in CI recipients with ipsilateral residual hearing, with cognitive thresholds of acoustic tones significantly elevated in the presence of electric maskers (Perkins et al., 2021; Harrison et al., 2020; Mamelle et al., 2020). Importantly, this masking effect occurs regardless of whether patients are programmed with EAS, and is a physiological phenomenon arising from the interaction between electric and acoustic stimulation in the cochlea (Perkins et al., 2021; Spitzer et al., 2021). For RP patients using electric-only maps (67% of our RP cohort), this interference may degrade the utility of their preserved acoustic hearing in complex listening environments without providing the optimized integration that EAS programming offers, potentially explaining their paradoxically lower subjective speech perception despite intact audiometric thresholds. The SSQ-12 Speech subscale examines the multi-talker and reverberant complex listening scenarios, where a mismatch in the electric and acoustic stimulation of the cochlea may prove to be more apparent.

The lack of association between performance-based and patient-reported outcomes postoperatively suggests that these measures capture unique aspects of treatment benefit, highlighting the need to clinically implement both types of assessments for a more holistic evaluation of cochlear implant outcomes, consistent with prior literature (Dietz et al., 2022; Dorismond et al., 2023; McRackan et al., 2019).

Electroacoustic stimulation

In this study, three of nine RP patients received active EAS mapping, and those patients demonstrated moderate advantages in the SQuaD sound quality descriptor subscale (D= + 0.75) and perceived clarity (d= +0.73) over non-EAS RP patients, suggesting that optimized electric acoustic fitting may attenuate sound quality character disadvantages in hearing preservation recipients (Turner et al., 2008; Gifford et al., 2017). The pitch perception advantage observed in EAS patients (d = +0.58) is also consistent with the benefit of EAS for musical sound quality (Blamey et al., 2015). EAS patients showed numerically lower SSQ-12 Speech scores (d = −0.51) and rated naturalness less favorably (d = −0.73) than non-EAS counterparts in RP, potentially reflecting the complexity of perceptual integration under different listening demands as well as likely attributable to small and unbalanced subgroup sizes.

The underutilization of EAS in our RP cohort may have limited their ability to capitalize on preserved hearing, potentially explaining the paradoxically lower subjective speech scores despite intact low-frequency hearing. This lower utilization rate is consistent with emerging literature documenting underutilization of EAS despite preserved hearing, (Perkins et al., 2021; Harrison et al., 2020; Mamelle et al., 2020) with identified barriers including physical discomfort with acoustic amplification, lack of perceived benefit, and difficulties adapting when low-frequency hearing deteriorates. This represents a critical gap between surgical hearing preservation success and clinical programming implementation that warrants further investigation.

Additional factors influencing outcomes

The RP group was significantly older at implantation (mean: 72.7 years) than the NR group (mean: 62.5 years), and age has been identified as a predictor of worse long-term sound quality ratings, suggesting older patients may experience different perceptual adaptation or expectations that influence subjective ratings (Berg et al., 2026). In our cohort, we attempted to control for covariates, including age at implantation (Figure 2), where trends continued to show, however, prior evidence suggests traditional patient-related factors demonstrate weak associations with speech recognition outcomes, underscoring the complexity of factors and unmeasured confounders influencing CI outcomes and limitations of traditional predictive models (Zhao et al., 2020).

Limitations and future directions

The sample of 31 patients divided across three groups with multiple predictors limits statistical power, therefore findings should be considered exploratory and hypothesis generating. The cross-sectional design rather than longitudinal design with pre- and post-operative SQuaD and SSQ-12 does not allow for causal inference. Additionally, the lack of standardized timing for survey completion relative to surgery introduces potential confounding from variable adaptation periods, though covariate adjusted analyses attempted to control for this factor. The heterogeneity in device manufacturers and electrode types may have introduced additional variability, though this reflects real-world clinical practice. Additionally, the timing and mechanism of hearing loss in the RL group are unknown, limiting our ability to determine whether hearing loss was immediate (surgical trauma) or delayed (progressive deterioration), which may have different implications for adaptation and sound quality perception. Further, the SQuaD, is a non-validated measure that warrants additional study and potential future validation.

Future research should examine sound quality and patient-reported outcome trajectories in patients with and without hearing preservation at standardized intervals, controlling for adaptation time and accounting for CI use via datalogging. Investigation of factors influencing EAS uptake and optimization in patients with preserved hearing is critical, as the low utilization rate (27%) in this cohort reflects a broader pattern documented in the literature and suggests a potential gap between surgical hearing preservation success and clinical programming implementation. The 80 dB threshold used to define residual hearing may include patients whose preserved hearing provides limited benefit from acoustic amplification, which is effective at thresholds of 70 dB or better, (Battmer et al., 2019) which may explain the limited EAS uptake observed. Studies should examine both patient-reported barriers (comfort, perceived benefit, adaptation challenges) and technical factors (electrode positioning, programming optimization, electric-acoustic interactions) that influence EAS utilization and outcomes. Additionally, studies examining the relationship between degree and configuration of residual hearing preservation and specific dimensions of sound quality would inform patient counseling and surgical planning.

Conclusion

This exploratory study demonstrates that loss of residual hearing following CI may negatively impact sound quality perception, a pattern that may be related to auditory memory and perceptual contrast effects, while preservation of residual hearing without optimized EAS programming may not translate to improved subjective speech perception despite equivalent objective outcomes. Electric-acoustic interference may contribute to the paradoxically lower subjective speech scores in patients with preserved hearing who are not using EAS, as electric stimulation can mask residual acoustic hearing even in the absence of EAS programming. Taken together, these preliminary findings support the importance of comprehensive outcome assessment incorporating both performance-based and patient-reported measures, as these capture distinct and complementary dimensions of cochlear implant benefit. The observed underutilization of EAS in patients with preserved hearing represents a gap between surgical success and clinical implementation that warrants targeted quality improvement efforts. Whether systematic EAS implementation and optimization in patients with preserved residual hearing improve subjective outcomes cannot be determined and warrant further larger prospective longitudinal studies to realize the full potential benefits of hearing preservation surgery. The weak correlations between objective speech recognition and patient-reported outcomes underscore the need for additional studies of validated patient-reported outcome measures to comprehensively assess treatment benefit and guide individualized rehabilitation strategies.

References

• 1

  BattmerR. D.ScholzS.GeisslerG.ErnstA. (2019). Electric acoustic stimulation (EAS) with the Naída CI Q90 sound processor in experienced cochlear implant users. Cochlear Implants Int.20, 331–340. doi: 10.1080/14670100.2019.1659468

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 2

  BergK. A.BirkyH. M.SevichV. A.MoberlyA. C.TamatiT. N. (2025). Sound quality, not speech recognition, explains cochlear implant-related quality of life outcomes. JASA Express Lett.5:104401. doi: 10.1121/10.0039069

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 3

  BergK. A.TamatiT. N.MoberlyA. C. (2026). Predictors of early and long-term sound quality ratings in adult cochlear implant recipients. Laryngoscope136, E273–E280. doi: 10.1002/lary.70007

  • CrossRef
  • Google Scholar
  • View reference in article

• 4

  BlameyP. J.MaatB.BaşkentD.MawmanD.BurkeE.DillierN.et al. (2015). A retrospective multicenter study comparing speech perception outcomes for bilateral implantation and bimodal rehabilitation. Ear Hear.36, 408–416. doi: 10.1097/AUD.0000000000000150

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 5

  CalvinoM.GavilánJ.Sánchez-CuadradoI.LassalettaL.Pérez-MoraR.MuñozE.et al. (2016). Using the HISQUI29 to assess the sound quality levels of Spanish adults with unilateral cochlear implants and no contralateral hearing. Eur. Arch. Oto-Rhino-Laryngol. 273, 1233–1239. doi: 10.1007/s00405-015-3789-0

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 6

  CarlsonM. L. (2020). Cochlear implantation in adults. N. Engl. J. Med.382, 1531-1542. doi: 10.1056/NEJMra1904407

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 7

  DalbertA.HuberA.BaumannN.VeraguthD.RoosliC.PfiffnerF.et al. (2016). Hearing preservation after cochlear implantation may improve long-term word perception in the electric-only condition. Otol. Neurotol.37:1193. doi: 10.1097/MAO.0000000000001193

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 8

  DietzA.HeinrichA.TörmäkangasT.Iso-MustajärviM.MiettinenP.WillbergT.et al. (2022). The effectiveness of unilateral cochlear implantation on performance-based and patient-reported outcome measures in finnish recipients. Front. Neurosci.16:786939. doi: 10.3389/fnins.2022.786939

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 9

  DorismondC.PatroA.HolderJ. T.PerkinsE. L. (2023). Correlation between quality of life and speech recognition outcomes following cochlear implantation. Otol. Neurotol.43, 1349–1357. doi: 10.1097/MAO.0000000000004029

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 10

  DornhofferJ. R.ReddyP.MeyerT. A.Schvartz-LeyzacK. C.DubnoJ. R.McRackanT. R.et al. (2021). Individual differences in speech recognition changes after cochlear implantation. JAMA Otolaryngol. Head Neck Surg. 147:5094. doi: 10.1001/jamaoto.2020.5094

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 11

  GantzB. J.DunnC.OlesonJ.HansenM.ParkinsonA.TurnerC.et al. (2016). Multicenter clinical trial of the nucleus hybrid S8 cochlear implant: final outcomes. Laryngoscope126:25572. doi: 10.1002/lary.25572

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 12

  GatehouseS.NobleI. (2004). The Speech, Spatial and Qualities of Hearing Scale (SSQ). Int. J. Audiol.43, 85–99. doi: 10.1080/14992020400050014

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 13

  GiffordR. H.DavisT. J.SunderhausL. W.DormanM. F. (2017). Combined electric and acoustic stimulation with hearing preservation: effect of cochlear implant low-frequency cutoff on speech understanding and perceived listening difficulty. Ear. Hear.39, 1214–1225. doi: 10.1097/AUD.0000000000000418

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 14

  HarrisonL.ManjalyJ. G.EllisW.et al. (2020). Hearing preservation outcomes with standard length electrodes in adult cochlear implantation and the uptake of electroacoustic stimulation. Otol. Neurotol.41, 274–281. doi: 10.1097/MAO.0000000000002702

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 15

  HelbigS.AdelY.RaderT.StoverT.BaumannU. (2016). Long-term hearing preservation outcomes after cochlear implantation for electric-acoustic stimulation. Otol. Neurotol.37, 1157–1165. doi: 10.1097/MAO.0000000000001066

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 16

  HoppeU.HockeT.HastA.IroH. (2021). Cochlear implantation in candidates with moderate-to-severe hearing loss and poor speech perception. Laryngoscope. 131:lary.28771 doi: 10.1002/lary.28771

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 17

  ImsieckeM.BüchnerA.LenarzT.NogueiraW. (2020). Psychoacoustic and electrophysiological electric-acoustic interaction effects in cochlear implant users with ipsilateral residual hearing. Hear. Res.2020:386. doi: 10.1016/j.heares.2019.107873

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 18

  IrvingS.GillespieL.RichardsonR.RoweD.FallonJ. B.WiseA. K.et al. (2014). Electroacoustic stimulation: now and into the future. Biomed. Res. Int. 2014, 1-8. doi: 10.1155/2014/350504

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 19

  KralA.O'DonoghueG. M. (2010). Profound deafness in childhood. N. Engl. J. Med.363:11225. doi: 10.1056/NEJMra0911225

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 20

  LiC.KuhlmeyM.KimA. H. (2019). Electroacoustic stimulation. Otolaryngol. Clin. North Am. 52:008. doi: 10.1016/j.otc.2018.11.008

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 21

  LimbC. J.RoyA. T. (2014). Technological, biological, and acoustical constraints to music perception in cochlear implant users. Hear. Res.2014:308. doi: 10.1016/j.heares.2013.04.009

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 22

  MamelleE.GrangerB.SterkersO.MosnierI.VanierA.BouccaraD.et al. (2020). Long-term residual hearing in cochlear implanted adult patients who were candidates for electro-acoustic stimulation. Eur. Arch. Oto-Rhino-Laryngol.277, 381–388. doi: 10.1007/s00405-019-05745-6

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 23

  McRackanT. R.BauschardM.HatchJ. L.Franko-TobinE.DroghiniR.NguyenS. A.et al. (2018). Meta-analysis of quality-of-life improvement after cochlear implantation and associations with speech recognition abilities. Laryngoscope. 128, 982–990. doi: 10.1002/lary.26738

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 24

  McRackanT. R.HandB. N.ChidaralaS.VelozoC. A.DubnoJ. R. (2022). Normative Cochlear Implant Quality of Life (CIQOL)-35 profile and CIQOL-10 global scores for experienced cochlear implant users from a multi-institutional study. Otol. Neurotol.41, 895–903. doi: 10.1097/MAO.0000000000003596

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 25

  McRackanT. R.HandB. N.VelozoC. A.DubnoJ. R. (2019). Association of demographic and hearing-related factors with cochlear implant-related quality of life. JAMA Otolaryngol Head Neck Surg.145, 431–438. doi: 10.1001/jamaoto.2019.0055

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 26

  NicastriM.Dincer D'AlessandroH.ConsolinoP.et al. (2025). Timing still matters: Auditory attention and verbal memory in children implanted before versus after age one. Int. J. Pediatr. Otorhinolaryngol.2025:198. doi: 10.1016/j.ijporl.2025.112605

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 27

  PerkinsE.LeeJ.ManzoorN.GiffordR. H.DwyerR. T.LabadieR. F.et al. (2021). The reality of hearing preservation in cochlear implantation: who is utilizing EAS?Otol. Neurotol.42, 47–55. doi: 10.1097/MAO.0000000000003074

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 28

  PlantK.McDermottH.Van HoeselR.DawsonP.CowanR. (2016). Factors predicting postoperative unilateral and bilateral speech recognition in adult cochlear implant recipients with acoustic hearing. Ear Hear. 37, 664–676. doi: 10.1097/AUD.0000000000000233

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 29

  ReinhartP.ParkinsonA.GiffordR. H. (2024). Hybrid cochlear implant outcomes and improving outcomes with electric–acoustic stimulation. Otol. Neurotol. 45:4305. doi: 10.1097/MAO.0000000000004305

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 30

  SpitzerE. R.WaltzmanS. B.LandsbergerD. M.FriedmannD. R. (2021). Acceptance and benefits of electro-acoustic stimulation for conventional-length electrode arrays. Audiol. Neurotol.25, 121-129. doi: 10.1159/000507975

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 31

  Tsai DoB. S.BushM. L.WeinreichH. M.SchwartzS. R.AnneS.AdunkaO. F.et al. (2024). Clinical practice guideline: age-related hearing loss. Otolaryngol Head Neck Surg.170, S1–S54. doi: 10.1002/ohn.750

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 32

  TurnerC. W.ReissL. A. J.GantzB. J. (2008). Combined acoustic and electric hearing: preserving residual acoustic hearing. Hear. Res.240, 1-9. doi: 10.1016/j.heares.2007.11.008

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article

• 33

  ZhaoE. E.DornhofferJ. R.LoftusC.NguyenS. A.MeyerT. A.SladenD. P.et al. (2020). Association of Patient-Related Factors with Adult Cochlear Implant Speech Recognition Outcomes: a Meta-analysis. JAMA Otolaryngol - Head Neck Surg.146, 613–620. doi: 10.1001/jamaoto.2020.0662

  • Pubmed Abstract
  • CrossRef
  • Google Scholar
  • View reference in article