Assessment of Speech Understanding After Cochlear Implantation in Adult Hearing Aid Users: A Nonrandomized Controlled Trial

Key Points

Question  Are cochlear implants safe and effective in adult hearing aid users with and without mild cognitive impairment (MCI)?

Findings  In this nonrandomized controlled trial of 96 adult hearing aid users, cochlear implantation resulted in 3 serious adverse events, and all resolved without sequelae. By 6 months of use, statistically significant and clinically important improvements in speech understanding in both quiet and noise and quality of life were observed in individuals with and without baseline MCI.

Meaning  The findings of this nonrandomized controlled trial seem to indicate that cochlear implantation is a safe and effective intervention for improving speech understanding and quality of life in a select group of adult hearing aid users with and without MCI.

Importance  Cochlear implants were approved for use in adults in the 1980s, but use remains low owing to a lack of awareness regarding cochlear implantation candidacy criteria and expected outcomes. There have been limited, small series examining the safety and effectiveness of cochlear implantation in adult hearing aid (HA) users with and without mild cognitive impairment (MCI).

Objective  To investigate the safety and effectiveness of a single-ear cochlear implant in a group of optimized adult HA users with and without MCI across a variety of domains.

Design, Setting, and Participants  In this nonrandomized controlled trial, a multicenter, prospective, repeated-measures investigation was conducted at 13 US institutions. The setting was academic and community-based cochlear implant programs. Eligible participants were 100 adults (aged >18 years) with postlinguistic onset of bilateral moderate sloping to profound or worse sensorineural hearing loss (≤20 years’ duration). Fluent English speakers underwent an optimized bilateral HA trial for at least 30 days. Individuals with aided Consonant-Vowel Nucleus-Consonant (CNC) word score in quiet of 40% or less correct in the ear to be implanted and 50% or less correct in the contralateral ear were offered cochlear implants. The first participant was enrolled on February 20, 2017, and the last participant was enrolled on May 3, 2018. The final follow-up was on December 21, 2018.

Interventions  Participants received the same cochlear implant system and contralateral HA.

Main Outcomes and Measures  The primary outcome measure was speech understanding in quiet (CNC word score) using both the cochlear implant and opposite ear HA. Secondary outcome measures included the following: adverse events; speech understanding in noise (AzBio signal-to-noise ratio of +10 db [+10 SNR]) Health Utilities Index Mark 3 (HUI3); Speech, Spatial, and Qualities of Hearing Questionnaire 49 (SSQ49); and Montreal Cognitive Assessment (MoCA).

Results  The median age at cochlear implantation of the 96 patients included in the trial was 71 years (range, 23-91 years), and 62 patients (65%) were male. Three serious adverse events requiring revision surgery occurred, and all resolved without sequelae. By 6 months after activation, the absolute marginal mean change in CNC word score and AzBio +10 SNR was 40.5% (95% CI, 35.9%-45.0%) and 24.1% (95% CI, 18.9%-29.4%), respectively. Ninety-one percent (87 of 96) of participants had a clinically important improvement (>15%) in the CNC word score in the implant ear. Mild cognitive impairment (MoCA total score ≤25) was observed in 48 of 81 study participants (59%) at baseline. Speech perception marginal mean improvements were similar between individuals with and without baseline MCI, with values of 40.9% (95% CI, 35.2%-46.6%) and 39.6% (95% CI, 31.8%-47.4%), respectively, for CNC word score and 27.5% (95% CI, 21.0%-33.9%) and 17.8% (95% CI, 9.0%-26.6%), respectively, for AzBio +10 SNR. Statistically significant and clinically important improvements in the HUI3 and SSQ49 were evident at 6 months.

Conclusions and Relevance  The findings of this nonrandomized controlled trial seem to indicate that cochlear implants are safe and effective in restoring speech understanding in both quiet and noise and improve quality of life in individuals with and without MCI.

Trial Registration  ClinicalTrials.gov Identifier: NCT03007472

Worldwide, 466 million people have hearing loss, with expected growth to 630 million individuals by 2030 and to 900 million individuals by 2050.^1,2 Hearing loss creates life-changing disability by adversely changing sound awareness, communication ability, employment opportunities, and quality of life (QOL). It has been associated with many substantial comorbidities, including cognitive decline and dementia.³ The Lancet Commission recently identified hearing loss as the strongest modifiable risk factor for developing dementia.⁴ Therefore, meaningful interventions are needed to restore hearing for people across the globe.

Hearing aids (HAs) can improve sound detection and speech recognition for individuals who are hard of hearing. For those with more substantial hearing losses (moderate sloping to profound or worse sensorineural hearing loss), cochlear implants are able to improve both sound detection and speech recognition through neural stimulation, even for individuals with deafness. In practice, patients with hearing loss are first treated with HAs and progress to cochlear implants when amplification alone is not sufficient. There has been a growing trend to combine an HA in one ear with a cochlear implant in the other ear (ie, bimodal condition) to improve speech recognition in both quiet and noise, sound localization, and music appreciation.⁵ Unfortunately, cochlear implant use among eligible adults (<6% of candidates) remains low.⁶ Although the reasons are varied, this low penetration is in part associated with a reduced awareness of the benefits that cochlear implants can provide over conventional HAs.

The objective of the present study was to investigate the safety and effectiveness of a single-ear cochlear implant in a group of optimized adult HA users with and without mild cognitive impairment (MCI) across a variety of outcome domains, including speech recognition in quiet and noise, QOL, electrode position within the cochlea, and cognitive status. This study is unique in that it prospectively captures a comprehensive group of outcome measures in a large cohort of optimized HA users. Our results have practical implications for management and referral of patients with hearing loss.

In this nonrandomized controlled trial, a multicenter, prospective, repeated-measures investigation was conducted at 13 US institutions. The setting was academic and community-based cochlear implant programs. Eligible participants were 100 adults (aged >18 years) with postlinguistic onset of bilateral moderate sloping to profound or worse sensorineural hearing loss (≤20 years’ duration). The trial protocol (Supplement 1) was approved by the local institutional review boards. The aim of the trial was to gather long-term data on the US Food and Drug Administration–approved, commercially available CI532 implant (Cochlear Nucleus; Cochlear Ltd) and CP1000 sound processor (Cochlear Nucleus 7; Cochlear Ltd). The Consolidated Standards of Reporting Trials (CONSORT) reporting guideline was followed. Written informed consent was obtained by the site study coordinator from each individual after explaining the details of the study (rationale, aims, and objectives), the risks and benefits of the trial treatment (and alternative treatments), and the extent of patient involvement. The first participant was enrolled on February 20, 2017, and the last participant was enrolled on May 3, 2018. The final follow-up was on December 21, 2018. All participants received the same cochlear implant system and contralateral HA. The primary outcome measure was speech understanding in quiet (Consonant-Vowel Nucleus-Consonant [CNC] word score) using both the cochlear implant and opposite ear HA. Secondary outcome measures included the following: adverse events (AEs); speech understanding in noise (AzBio signal-to-noise ratio of +10 dB [+10 SNR]) Health Utilities Index Mark 3 (HUI3); Speech, Spatial, and Qualities of Hearing Questionnaire 49 (SSQ49); and Montreal Cognitive Assessment (MoCA).

Participants were 18 years or older with postlinguistic onset of bilateral moderate sloping to profound or worse sensorineural hearing loss. Participants offered cochlear implants were fluent English speakers with an aided CNC word score in quiet of 40% or less correct in the ear undergoing implantation and 50% or less in the contralateral ear. All participants underwent a minimum 30-day experience with optimized bilateral HAs fit to National Acoustics Laboratory (NAL-1) targets before baseline assessment.⁷

Individuals with prelinguistic onset of hearing loss before age 2 years, auditory neuropathy, retrocochlear pathology, structural cochlear anomalies, or severe to profound hearing loss exceeding 20 years were excluded. Individuals with prior cochlear implant experience, inability to complete study evaluations, or any other contraindication as assessed by the managing centers (listed in the Group Information section at the end of the article) were also excluded.

Hearing device configuration before cochlear implantation was 2 HAs; configuration after implantation was a cochlear implant and a contralateral HA (ie, bimodal condition). All study participants received the CI532 implant and were fitted with the CP1000 sound processor. After cochlear implantation, all participants were fitted with a ReSound HA (GN Hearing) for the nonimplant ear. Hearing aid use in the nonimplant ear was required until 6 months after cochlear implantation.

Participants were evaluated before and after receiving the cochler implant on a variety of measures. These included audiometric thresholds, speech recognition using recorded monosyllabic words, and sentence scores in noise conditions. Other measures were patient-reported outcomes regarding QOL, satisfaction with use of the hearing devices, and a cognitive function screening test. Outcomes for speech recognition and patient-reported measures were examined at the defined primary end point (6 months of cochlear implant use) and compared with performance at baseline.

Surgery consisted of a standard transmastoid facial recess approach to the cochlea. A round window or round window–related opening was created, and electrode insertion was performed according to the manufacturer’s specifications.⁸ An intraoperative radiograph was used to confirm expected coiling of the array. Postoperative computed tomography (CT) was used to identify the electrode location within the cochlea.⁹ Detailed findings at the time of surgery, complications, and results of the CT analysis will be reported in a future publication.

Speech recognition in quiet presented at 60 dB (weighted according to the A scale) in the soundfield (0-degree azimuth) was assessed before surgery and at 6 months in the monaural and binaural conditions using CNC monosyllabic words.¹⁰ The CNC lists consist of 50 word items scored as percentage correct identified through verbal responses per test condition. The AzBio sentence lists presented in background noise at +10 SNR were administered in the same conditions before and after cochlear implantation at the 6-month interval.¹¹ The AzBio sentence lists consist of 20 sentences, with each word in a sentence that was correctly identified scored as a percentage of the total words. At each test session, 2 CNC word lists and 1 AzBio sentence list were administered. When testing an individual ear, independent ear conditions included plugging of the opposite ear to minimize any contribution. The highest score evaluated in either the bimodal condition or the implant ear alone was considered the best-aided condition.

The HUI3 and SSQ49 were administered before surgery and at the 6-month interval.^12,13 Participants considered their everyday listening condition at the time of questionnaire completion.

The HUI3 is a widely used, validated instrument assessing a respondent’s level of function and QOL through the following modalities: vision, hearing, the capacity to be understood when speaking (speech domain), ambulation, dexterity, emotion, cognition, and pain.¹² Quantification of overall health status is calculated, with a value between 0 indicating death and 1.0 indicating perfect health. Changes of at least 0.03 in the multiattribute health index and at least 0.05 in single domains between test intervals are considered clinically important.¹³

The SSQ49 is a hearing ability functional scale with 49 questions designed to measure a range of hearing ability in a variety of daily listening conditions across 3 subcategories (speech understanding, spatial hearing, and quality of sounds).¹⁴ Participants respond to each item by rating their hearing ability on a visual analog scale ranging from 0 (not at all) to 10 (perfectly); higher ratings correspond to better self-reported hearing ability. Changes exceeding 1.0 in the mean subcategory or total scale scores are considered clinically important.¹⁵

Cognition was screened before surgery using the MoCA and retested again after 6 months of cochlear implant use.¹⁶ The MoCA is a rapid screening instrument designed to be a more sensitive screen for MCI compared with the Mini-Mental State Examination (MMSE) because it contains more cognitively demanding questions on memory recall and executive function.^16,17 The MoCA is 30-point test administered in approximately 10 minutes with components on short-term memory, visuospatial ability, executive function, language, concentration, and working memory. Scores of 26 and higher indicate normal cognition, whereas scores of 25 and lower indicate possible cognitive impairment.¹⁶ We categorized the participants into 2 groups with and without MCI based on the MoCA total score at baseline assessment. The MoCA total score was not used to assess cochlear implantation candidacy in this study.

We estimated that a sample size of 26 participants was needed to detect a mean (SD) difference of at least 15% (22%) in CNC word score in the cochlear implant ear with 90% power at the 2-sided α level of .05. Six months was chosen as the primary outcome measure end point because results for CNC word score plateau by this test interval. To control for potential confounding factors, we planned to enroll 100 participants. Standard descriptive statistics were used to describe baseline characteristics and outcome measures. Frequency and relative frequency were used to describe categorical variables. Normal distribution was tested using the Shapiro-Wilk test. Means (SDs) were used to describe normally distributed continuous variables, and medians (ranges) were used to describe nonnormally distributed variables. The McNemar test was used to compare the proportion of participants with MCI at baseline and 6-month visits. A mixed-effects model with participants as a random factor was used to explore the change in outcome measures between baseline and 6 months and to compare the change between participants with and without MCI after controlling for potential confounders. An unstructured covariance matrix was used to model within-participant error. Estimated marginal means (with 95% CIs) were plotted to show this change. A change of 15% or more in speech audiometry measures was considered clinically important.¹⁸ Statistical analysis used SAS, version 9.4, statistical software (SAS Institute Inc). A separate analysis of patients 65 years and older was conducted and is reported elsewhere.¹⁹

One hundred eligible participants who underwent unilateral cochlear implantation after HA optimization were enrolled in the study (Figure 1). Four individuals did not complete the study because of cochlear implant revision surgery (n = 3) (2 with electrode misplacement and 1 with aversive symptoms) or loss to follow-up (n = 1). Demographics and baseline characteristics are listed in Table 1. The median age at cochlear implantation in the 96 patients included in the trial was 71 years (range, 23-91 years), and 62 participants (65%) were men. Seventy participants (73%) were 65 years or older, and 13 participants (14%) were 80 years or older. The mean (SD) duration of hearing loss in the implant ear was 27 (14) years, with no statistically significant difference between ears (mean difference, 0.4 years; 95% CI, −0.6 to 1.4 years). The mean (SD) duration of HA use was 20 (12) years in the cochlear implant ear and 19 (11) years in the ear without the implant.

Figure 2 shows the range of auditory thresholds in the implant ear before surgery. Most patients had severe to profound hearing loss, although some low-frequency residual hearing did not preclude cochlear implantation candidacy. The mean (SD) 3-frequency, low-tone (0.5, 1.0, and 2.0 kHz) pure-tone average was a hearing level of 85 (15) dB in the ear with an implant and 78 (13) dB in the ear without an implant. After an optimized HA fitting (≥30 days’ use), speech recognition testing using 2 HAs revealed a median CNC word score of only 30% (range, 0%-68%) correct and a median AzBio +10 SNR of only 28% (range, 0%-91%) correct.

At surgery, all 22 electrodes were fully inserted in all participants. One electrode tip foldover (1%) was identified on intraoperative radiograph and was corrected without incident. Two electrode tip foldovers (2%) were identified after surgery and were revised with subsequent surgery.

Device-related or procedure-related AEs are listed in the eTable in Supplement 2. Three serious AEs were considered device related or procedure related and required revision surgery to resolve. All resolved without sequelae. These revisions included 1 case of aversive, nonauditory stimulation with hearing device use that eventually resolved after revision surgery. At the time of publication, this patient continues to use the cochlear implant device. In addition, postoperative CT revealed that 2 electrode tip foldovers that were not seen on intraoperative radiographs required revision implantation and were resolved uneventfully. Of the remaining AEs, all were minor and self-limited.

Postoperative CT showed that 82 of 92 electrode arrays (89%) with available CT images were placed fully in scala tympani, 3 (3%) were in scala vestibuli, and 7 (8%) crossed over from scala tympani to scala vestibuli. The mean (SD) insertion angle was 404° (60°).⁹

After device activation, all participants used their cochlear implant, with a mean (SD) duration of 12.7 (3.2) hours per day. Of 96 patients included in the trial, 94 (98%) continued to wear a contralateral HA (ie, bimodal condition).

Speech recognition measures included baseline, 6-month visit, and estimated marginal mean difference values for CNC word score. This primary outcome measure is summarized in Table 2.

Statistically significant and clinically important improvements in CNC word score were achieved at 6 months compared with baseline (eFigure 1 and eFigure 2 in Supplement 2). Most participants (79 [82%]) had a CNC word score in the cochlear implant ear higher than 40% at the 6-month visit. The absolute marginal mean changes between 6 months and baseline were 46.3% (95% CI, 41.7%-50.9%) in the implant ear and 40.5% (95% CI, 35.9%-45.0%) in the bimodal condition. Apical electrode insertion angle was a substantial confounder of CNC word score change, but the magnitude of this association was small (<10% change). Ninety-one percent (87 of 96) of participants had a clinically important improvement (>15%) in the CNC word score in the implant ear. Statistically significant improvements for speech recognition in noise (AzBio +10 SNR) were seen at 6 months compared with baseline, with marginal mean differences of 27.9% (95% CI, 22.2%-33.6%) for the implant ear and 24.1% (95% CI, 18.9%-29.4%) for the bimodal condition (eFigure 1 and eFigure 2 in Supplement 2). The CNC word score in the best-aided condition improved by a mean of 37.3% (95% CI, 32.0%-42.5%) for individuals 65 years or older compared with a mean of 50.4% (95% CI, 41.8%-59.6%) for younger individuals. Similarly, AzBio +10 SNR in the best-aided condition improved by a mean of 22.6% (95% CI, 16.6%-28.7%) for individuals 65 years or older compared with 33.0% (95% CI, 23.8%-43.7%) in younger individuals.

Statistically significant and clinically important improvement was noted at 6 months compared with baseline in the domains of hearing (mean difference, 0.30; 95% CI, 0.25-0.36), speech (mean difference, 0.08; 95% CI, 0.04-0.12), and the multiattribute health index (mean difference, 0.18; 95% CI, 0.14-0.22). The change in the emotional domain was also statistically significant, but the mean difference was smaller than the minimally clinically important difference of 0.05 (Table 2).

Statistically significant and clinically important improvement was noted in subjective hearing ability in a variety of daily listening conditions across 3 subcategories of the SSQ49. These subcategories included speech understanding, spatial hearing, and quality of sounds in addition to the total score (Table 2). Specifically, the SSQ49 total score improved by an estimated marginal mean difference of 2.5 (95% CI, 2.2-2.9) at the 6-month interval compared with baseline. Similar magnitudes of improvement were seen for the subcategories.

Forty-eight of 81 study participants (59%) screened at baseline had MCI (MoCA total score ≤25). At 6 months, there was a statistically significant (albeit small) improvement (estimated marginal mean difference, 0.6; 95% CI, 0.0-1.3) in the MoCA total score compared with baseline, and only 34 of 81 participants (42%) had a MoCA total score of 25 or lower, for a proportion reduction of 17% (95% CI, 5%-29%). Only 27 of 48 participants (56%) with baseline impairment were identified as having cognitive impairment at 6 months. Among 33 participants with no cognitive impairment at baseline, 7 (21%) had cognitive impairment at 6 months (eFigure 3 in Supplement 2).

Figure 3 shows that the changes in CNC word score and AzBio +10 SNR were not statistically significantly different between participants with and without MCI at baseline. Specifically, speech perception marginal mean improvements were similar between individuals with and without baseline MCI, with values of 40.9% (95% CI, 35.2%-46.6%) and 39.6% (95% CI, 31.8%-47.4%), respectively, for CNC word score and 27.5% (95% CI, 21.0%-33.9%) and 17.8% (95% CI, 9.0%-26.6%), respectively, for AzBio +10 SNR. Statistically significant and clinically important improvements in the HUI3 and SSQ49 were evident at 6 months.

Cochlear implants were approved by the US Food and Drug Administration in 1984 for adult patients with profound hearing loss who gained no benefit from HAs,²⁰ but use remains low owing to a lack of awareness regarding cochlear implantation candidacy criteria and expected outcomes. There have been limited, small series examining the safety and effectiveness of cochlear implantation in adult HA users with and without MCI.^21-25 The criteria have broadened to include individuals with better hearing thresholds rather than deafness but who still benefit only partially from their HAs.²⁶ Since the earlier trials,^27,28 there have been only modest increases in cochlear implant adoption among the candidate population, with as few as 6% of adult candidates receiving the technology.⁶ This finding is explained in part by a broad lack of awareness regarding the indications, safety, and effectiveness of cochlear implantation among physicians of all types, hearing aid specialists, and the population at large. Our study results indicate that today, individuals with moderate sloping to profound or worse sensorineural hearing loss benefit from cochlear implantation when HAs are not sufficient. Cochlear implantation surgery is safe and greatly improves a person’s ability to understand speech in quiet and noise, thus improving QOL. These results underscore the fact that individuals with moderate sloping to profound or worse sensorineural hearing loss should be offered a cochlear implant evaluation when their HAs are not providing sufficient rehabilitation for speech understanding.

Previous studies^29-32 have demonstrated that hearing loss is a risk factor for development of dementia. In a 2017 review article by the Lancet Commission,⁴ social isolation and hearing loss were identified as 2 modifiable risk factors for dementia, with hearing loss being the largest contributor. In that study, the presence of hearing loss increased the risk of developing dementia 9 to 17 years later (risk ratio, 1.94; 95% CI, 1.38-2.73). In our study, MCI (MoCA total score ≤25) was identified before cochlear implantation in 59% of patients. This finding is not surprising given the median age of our study population (71 years) and the substantial degree of hearing loss present for more than 25 years in most participants. By 6 months after cochlear implantation, there was a statistically significant (albeit small) improvement (estimated marginal mean difference, 0.6; 95% CI, 0.0-1.3) in the MoCA total score compared with baseline, and 42% had a MoCA total score of 25 or lower (a proportion reduction of 17%; 95% CI, 5%-29%). Notably, individuals with MCI demonstrated similar degrees of improvement in speech recognition in both quiet and noise. Taken together, these data suggest that cochlear implantation is not contraindicated in individuals with cognitive decline. That some individuals with preexisting cognitive impairment show small improvements in cognitive performance could be explained by learning associations from MoCA retesting.³³ Results of related studies^21-25 using various assessment tools suggest that cochlear implantation can improve cognitive ability. Further study with longer follow-up is warranted to more clearly delineate such improvement.

Strengths of this study include the use of a variety of metrics to comprehensively describe the benefit of cochlear implantation in adults. Speech recognition measures have long been used to demonstrate the effectiveness of cochlear implantation. More recently, QOL measures have been adopted to understand how this technology improves daily function. Enrolled prospectively, we believe that the large cohort and the broad set of outcome domains that we assessed provide a holistic view of the benefit of cochlear implantation.

This study also has some limitations. Cochlear implantation trials (eg, those by Bassim et al²⁷ and by Balkany et al²⁸) have relied on single-subject experimental designs rather than randomized, placebo-controlled designs for many years owing to the progressive, irreversible, and heterogeneous nature of hearing loss as well as the inability to conceal the presence of the hearing device after surgery.³⁴ Another limitation of the present study is the short follow-up period. For both patient-reported outcome measures (HUI3 and SSQ49) and cognitive measures (MoCA), longer follow-up is needed to clarify the associations observed herein more precisely. For cognitive measures specifically, longer-term and more extensive cognitive assessments using a variety of tools are needed to further delineate the improvements. In addition, because this study evaluated a single implant type, the findings may not be generalizable to patients who receive other cochlear implantation products.

Cochlear implants are safe and effective in restoring speech understanding in both quiet and noise and improve QOL in individuals with and without MCI.

Accepted for Publication: June 19, 2020.

Corresponding Author: Craig A. Buchman, MD, Department of Otolaryngology–Head & Neck Surgery, Washington University School of Medicine in St Louis, 660 S Euclid Ave, Campus Box 8115, St Louis, MO 63110 (buchmanc@wustl.edu).

Published Online: August 27, 2020. doi:10.1001/jamaoto.2020.1584

Author Contributions: Dr Buchman and Dr Kallogjeri had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Buchman, Herzog, McJunkin, Firszt.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: Buchman, Herzog, Firszt, Kallogjeri.

Critical revision of the manuscript for important intellectual content: Buchman, McJunkin, Wick, Durakovic, Kallogjeri.

Statistical analysis: Kallogjeri.

Obtained funding: Buchman.

Administrative, technical, or material support: Buchman, Herzog, McJunkin, Durakovic.

Supervision: Buchman, Herzog.

Conflict of Interest Disclosures: Dr Buchman reported serving as a consultant for Advanced Bionics, Cochlear Ltd, Iotamotion, and Envoy Medical; receiving research support from Advanced Bionics, Cochlear Ltd, and MedEL; obtaining grants from the US Department of Defense; having a patent to US9,072,468B2; and owning stock in Advanced Cochlear Diagnostics, LLC. Drs Herzog, McJunkin, and Firszt reported serving as consultants for Cochlear Ltd. Dr Herzog reported receiving grants from Cochlear Americas. Dr McJunkin reported receiving consultant fees for a video from the Cochlear Corporation. Dr Wick reported receiving research support from the Cochlear Corporation and serving as a consultant for Stryker. Dr Firszt reported serving as a consultant for Advanced Bionics, receiving grants and personal fees from Cochlear Americas, and receiving personal fees from Advanced Bionics. Dr Kallogjeri reported serving as a consultant for and owning stock in PotentiaMetrics. No other disclosures were reported.

Funding/Support: This work was supported by Cochlear Ltd.

Role of the Funder/Sponsor: The funder worked with Dr Buchman on study design, protocol adherence, centralized data collection, and monitoring. Data collection, analysis, interpretation, and writing of the report were the sole responsibility of the authors. Surgical procedures were paid for by the patient’s insurance carrier or Medicare. Payments from the funder were only for those tests that were deemed investigative (Health Utilities Index Mark 3; Speech, Spatial, and Qualities of Hearing Questionnaire 49; Montreal Cognitive Assessment; and postoperative computed tomography analysis) and beyond the scope of routine audiological care.

Group Information: The CI532 Study Group comprises investigators who performed the patient evaluations, surgical procedures, and data collection at the various institutions included in this trial. The CI532 Study Group members are as follows: Washington University in St Louis and Center for Hearing and Balance Disorders (now a part of the Washington University in St Louis program), St Louis, Missouri (Craig A. Buchman, Jacques A. Herzog, Jonathan L. McJunkin, Cameron C. Wick, Nedim Durakovic, Jill B. Firszt, Dorina Kallogjeri, Andrew Drescher, Laura Holden, Noel Dwyer, Lydia Beyer, Susan Rathgeb, Lisa Potts, and Karen Mispagel); Dallas Ear Institute, Dallas, Texas (Bob Peters, Yoav Hahn, Kristin King, and Leslie Lianos); Ear Medical Group/Cochlear Hearing Center, San Antonio, Texas (Brian Perry, Susan King, Jerome Evans, and Linda Luduena); Hearts for Hearing, Oklahoma City, Oklahoma (Mark Wood, Stan Baker, Mila Duke, Sara Neumann, and Jace Wolfe); Midwest Ear Institute, Kansas City, Missouri (Robert Cullen, Joe Ursick, Kristen Lewis, Sarah Zlmoke, and Morgan Nelson); New York University, New York (Susan Waltzman, Tom Roland, Daniel Jethanamest, David Friedman, Laurel Mahoney, Alison Rigby, and Bill Shapiro); Ohio State University, Columbus (Oliver F. Adunka, Aaron Moberly, Edward Dodson, and Kara Vasil); Rocky Mountain Ear Center, Denver, Colorado (David Kelsall, Eric Lupo, and Allison Biever); Spokane ENT, Spokane, Washington (Neil Giddings, Jonathan Ziegler, and Pam Polennsky-Boner); University of California, San Francisco (Charles Limb, Aaron Tward, and Kurt Kramer); University of Iowa, Iowa City (Bruce Gantz, Marlan Hansen, Camille Dunn, Jill Beecher, Tanya Van Voorst, and Sue Karsten); and University of Michigan, Ann Arbor (Terry Zwolan, Hussam El-Kashlan, Steve Telian, and Heidi Slager).

Disclaimer: Dr Kallogjeri is the Statistics Editor for JAMA Otolaryngology–Head & Neck Surgery but was not involved with any of the decisions regarding review of the manuscript or its acceptance.

Data Sharing Statement: See Supplement 3.