The Role of Anatomy‐Based Fitting in Bimodal Rehabilitation: Choosing Electrode Arrays Based on Cochlear Length

ABSTRACT

Objectives

To evaluate the clinical impact of anatomy‐based fitting (ABF) in cochlear implants for patients with asymmetric hearing loss.

Study Design

Prospective cohort study.

Setting

Tertiary referral center specializing in hearing disorders.

Methods

Fifteen adult patients with asymmetric severe‐to‐profound sensorineural hearing loss or anacusis in the implanted ear, and residual low‐frequency hearing (≤ 65 dB HL at 125–500 Hz) in the better ear, were enrolled. All underwent cochlear implantation with bimodal rehabilitation (cochlear implant + hearing aid) and completed a 12‐month follow‐up. Cochlear duct length was measured preoperatively using OTOPLAN software to select the appropriate electrode array. Postoperative CT scans were analyzed with OTOPLAN to verify electrode positioning and generate the ABF map. Speech intelligibility in noise was assessed using the Italian Matrix sentence test in three listening conditions, comparing outcomes between the ABF and the default MAESTRO fitting maps.

Results

No statistically significant differences were found in speech perception scores between ABF and MAESTRO maps across test conditions. However, 86.7% of patients (13/15) preferred the ABF map for daily use, citing improved sound quality and listening comfort.

Conclusion

ABF, which aligns electrical stimulation with the patient's cochlear tonotopy by individualizing electrode selection and programming, may enhance the subjective auditory experience. While objective speech outcomes were comparable, patient preference strongly favored ABF, supporting its clinical relevance for optimizing cochlear implant performance in bimodal users.

Summary

Significant findings of the study

• Anatomy‐based fitting (ABF) was preferred by 86.7% of patients compared to clinically based fitting, despite no significant differences in objective speech outcomes.

• ABF improved subjective auditory experience, enhancing sound quality and listening comfort.

What this study adds

• This is the first prospective study evaluating ABF in asymmetric hearing loss with a 12‐month follow‐up.

• The study demonstrates that individualized electrode selection can optimize cochlear implant fitting without the need for postoperative imaging.

1. Introduction

In recent years, advancements in cochlear implant (CI) technology have significantly improved the rehabilitation of individuals with asymmetric hearing loss. Since 2010, the use of bimodal stimulation has increased notably [1]. This approach—combining a hearing aid (HA) in one ear and a CI in the contralateral ear affected by profound hearing loss—is widely regarded as the optimal solution for such patients.

Numerous advantages have been reported with the use of either a single HA or a CI, as each device contributes uniquely to auditory rehabilitation. HAs are particularly effective in amplifying low‐frequency sounds, while CIs—due to the placement of the electrode array in the basal turn of the cochlea—typically offer more consistent stimulation of high frequencies. However, the frequency map of adjacent spiral ganglion regions depends on the depth and position of the implanted electrode array within the cochlea [2].

As a result, CIs provide most of the speech information necessary for intelligibility, whereas HAs contribute valuable low‐frequency cues, including the perception of emotional prosody, consonant voicing, and suprasegmental features of speech [3, 4], leading to improved speech naturalness.

Many studies have investigated prognostic factors influencing outcomes following unilateral cochlear implantation. Key variables identified included age at onset of deafness, age at implantation, duration of deafness, years of CI use, etiology of hearing loss, and preoperative speech perception scores [5]. In contrast, recent reserch on bimodal stimulation has highlighted additional factors affecting speech perception, such as signal integration and interference, acoustic characteristics of speech, and the extent of residual acoustic hearing [1].

Beyond its primary goal of restoring binaural hearing by optimizing auditory input in each ear, bimodal stimulation facilitates central auditory processing of interaural timing, phase, and level differences. This integration enhances speech perception in noise and improves sound source localization [6, 7]. Speech perception in both quiet and noisy environments is further improved when the acoustically stimulated ear retains low‐frequency hearing thresholds better than 60 dB [8]. This improvement is attributed to the auditory brainstem's ability to accurately compare interaural phase and timing differences at low frequencies, a key mechanism underlying binaural unmasking which supports speech understanding in background noise [6]. In response to these findings, all CI manufacturers strive to enhance frequency discrimination, particularly in the low‐frequency range, although they employ different strategies to achieve this goal.

An additional factor influencing speech perception outcomes is the precise knowledge of electrode position within the cochlea relative to the cochlear place of acoustic frequency stimulation [9]. It is generally assumed that the closer the frequency assigned to each electrode aligns with its corresponding cochlear place, the better the central integragation of frequency information. This alignment is associated with improved speech perception in noise due to more effective use of binaural cues [10]. Otoplan, a software developed by CAScination AG (Bern, Switzerland) in collaboration with MED‐EL Corporation (Innsbruck, Austria), enables measurement of the cochlear duct length using computed tomography imaging [11]. This tool assists surgeons in selecting the most appropriate electrode array for each patient and, postoperatively, in evaluating the position of each electrode contact via a second CT scan. Accurate knowledge of electrode positioning enables anatomy‐based fitting (ABF), which minimizes frequency mismatch between the electrically and the acoustically stimulated ears.

This study aims to compare CI fitting outcomes among patients with asymmetric hearing loss using bimodal stimulation, evaluating standard automatic frequency allocation provided by the MAESTRO software versus ABF, which accounts for individual electrode positioning within the cochlea.

2. Methods

2.1. Patients

This prospective study included 15 patients (10 males, 5 females), all of whom used a HA with a receiver‐in‐the‐ear (RITE) configuration on the non‐implanted side. The HAs, which varied across individuals, were the same devices used prior to cochlear implantation. None of the patients used bimodal streaming via AudioStream. Demographic and clinical characteristics are summarized in Table 1.

Table 1.

Demographic and clinical characteristics of cochlear implant patients.

Number	Gender	Age (years)	Aetiology	PTA aided ear (dB)	HA side	FF PTA with HA	HA use (years)
1	M	69	SNHL in otosclerosis	58	R	46	3
2	M	50	Otosclerosis	80	L	49	3
3	F	48	Congenital right anacusia and progressive left HL (ski slope)	60	L	42	5
4	M	81	Otosclerosis	58	L	60	5
5	M	58	Progressive HL	58	R	40	8
6	F	55	Meningitis	40	R	34	3
7	F	52	Congenital HL	70	L	35	34
8	M	50	Left SNHL and right progressive HL	50	R	40	4
9	M	72	Otosclerosis	60	R	45	5
10	F	35	Congenital HL	83	L	64	33
11	F	66	Bilateral Menière disease	42	R	33	5
12	M	54	Progressive HL	50	L	35	3
13	M	67	Cochleo‐vestibular paralysis	56	R	40	2
14	M	70	Ototoxicity	40	R	28	6
15	M	69	Menière disease	61	L	35	27

Abbreviations: FF, free field; HA, hearing aid; HL, hearing loss; PTA, pure tone average at 0, 500, 1000, 2000, 4000 Hz; SNHL, sensorineural hearing loss.

A sample size of 15 subjects was calculated to yield a two‐sided 95% confidence interval with a margin of error of 0.388 from the mean signal‐to‐noise ratio (SNR) difference between ABF and MAESTRO fitting methods, assuming a standard deviation (SD) of 0.700 for the paired differences.

All patients underwent unilateral cochlear implantation for asymmetric sensorineural hearing loss, defined by a pure‐tone average (PTA) > 70 dB HL (500–4000 Hz) in the implanted ear and a word recognition score (WRS) < 50% in aided free‐field conditions at 65 dB SPL. All participants completed a 12‐month follow‐up.

Inclusion criteria were: age > 18 years, PTA > 65 dB HL (125–500 Hz) in the better‐hearing ear, use of bimodal stimulation for auditory rehabilitation, functional activation of at least 10 CI electrodes.

Written informed consent was obtained from all participants. The study was approved by the local Ethics Committee (Protocol n. 12/2021) and conducted in accordance with the Declaration of Helsinki.

Electrode length for each subject was selected based on individual cochlear dimensions. All patients underwent both preoperative and postoperative CT scans. Preoperative imaging was analyzed using Otoplan to determine cochlear size, while postoperative scans assessed insertion depth and electrode posizioning.

Two different CI frequency mapping methods were applied in all patients:

• 1.Clinically based fitting (CBF—MAESTRO): a generic mapping technique based on a predefined frequency distribution for the electrode array, initially adjusted according to the patient's subjective feedback during the first fitting session. This method does not account for individual anatomical variability of the cochlea [12];
• 2.ABF: a personalized mapping technique that uses patient's specific anatomical characteristics, preoperatively obtained by Otoplan, to align frequency allocation with the cochlear tonotopic map. Postoperative CT scans verify electrode placement (insertion depth and angle), enabling fine‐tuning of the stimulation map.

This approach aims to align the CI's frequency map with the cochlea's natural place‐pitch organization, improving fitting accuracy and bringing hearing outcomes closer to natural perception.

2.2. Intervention

Speech intelligibility in noise was assessed using the Italian Matrix sentence test (ITA Matrix) [13]. In this test, background noise was fixed at 65 dB SPL, while the signal intensity was adaptively adjusted based on the participant's responses.

The test material consisted of sentences composed of five phonetically balanced words, which the subject was instructed to repeat aloud. Scoring was based on the number of correctly repeated words and the outcome measure was the SNR at which the participant correctly recognized 50% of the words. For Italian‐speaking individuals with normal hearing, the reference SNR value is −7.3 ± 0.2 dB [13].

The ITA Matrix test was administered through two loudspeakers positioned one meter from the participant. Three spatial configurations were used to evaluate binaural hearing:

• Summation condition (S₀N₀): speech and noise both presented from the front speaker;

• Squelch condition (S₀N₉₀): speech presented from the front and noise directed toward the CI side;

• Head shadow condition (S₉₀N_‐90): speech presented to the CI ear and noise directed to the contralateral ear (HA side).

Each configuration was tested under two listening conditions: with the HA alone and with bimodal stimulation using both CI maps (ABF and MAESTRO), following the protocol described below.

CI activation occurred 4–6 weeks after surgery. Each patient was then randomly assigned to receive either the ABF or MAESTRO fitting map, which were labeled generically as “Map 1” or “Map 2” to ensure blinding. During the first 4 months, the two maps were alternated on a monthly basis; from the 5th month onward, they were alternated weekly to minimize long‐term adaptation to any single map.

The first evaluation took place 2 months after CI activation, testing with ITA Matrix all three spatial conditions (summation, squelch, head shadow) for both fitting maps. Further assessments occurred at 4 months post‐activation and monthly thereafter until the 9th month. During the 9th‐month assessment, patients were asked to independently choose their preferred map for continued use, based solely on subjective preference. A final evaluation was conducted at 12 months; at this stage, no further comparisons between the two maps were made, as participants had already adopted their preferred fitting strategy.

At the 12‐month follow‐up, pitch perception was also assessed using the MED‐EL Pitch Discrimination Test. This test evaluates the subject's ability to compare the frequency perception between the HA and CI ears. It involves three phases: two preliminary loudness‐balancing steps followed by the pitch discrimination task. Initially, with the CI turned off, two pure tones of 1000 Hz at varying intensities were presented in free field to the HA ear to identify the subject's most comfortable listening level. Once identified, loudness balancing between the HA and CI ears was then performed using pure tones with variable frequencies and intensities, both in free field and via Audiolink for the CI.

In the final phase, pitch discrimination was evaluated using two target frequencies (500 and 1000 Hz). Pairs of tones—one presented in free field to the HA ear and one via Audiolink to the CI ear—were delivered sequentially, with one tone randomly chosen from the target frequencies. A pitch‐matching graph was generated to represent the frequency perceived through the CI. This test was performed twice per patient, once using the MAESTRO map and once using the ABF map.

2.3. Statistical Analysis

Descriptive stastistics were used to summarize clinical data, with results reported as means and SD. The Kolmogorov–Smirnov test was employed to assess the normality of continuous variables distributions. Comparisons between outcomes obtained with the two fitting strategies (ABF vs. MAESTRO) were conducted using the Wilcoxon‐Mann‐Whitney test for continuous variables and either the chi‐square (χ ²) test or Fisher's exact test for categorical variables, as appropriate. To enhance interpretability, the sample was stratified into quartiles based on SNR values. General linear models (GLMs) were applied to investigate potential differences in SNR outcomes across the three Matrix test configurations (summation, squelch, head shadow) for each fitting strategy. All statistical tests were two‐tailed, with a significance threshold set at p ≤ 0.05. Statistical analyses were performed using SAS software, version 9.4 (SAS Institute Inc., Cary, NC, USA).

3. Results

Patient characteristics are summarized in Table 2. The mean cochlear duct length across the cohort was (32.82 ± 3.3) mm. In 8 of the 15 patients (53.3%), auditory deprivation in the implanted ear exceeded 10 years. At the 9‐month follow‐up, 13 patients (86.7%) selected the ABF map as their preferred fitting strategy, while two patients opted to continue with the MAESTRO map.

Table 2.

Patients’ cochlear implant details.

Number	CI side	FF PTA CI (dB)	Deafness duration (years)	Cochlear length (mm)	Electrode insertion	Type of electrode
1	L	37.50	> 10	36.0	Incomplete (11–12 out)	Flex 28
2	R	36.25	< 10	37.4	Complete	Flex soft
3	R	42.00	> 10	35.7	Complete	Flex soft
4	R	45.00	< 10	26.0	Incomplete (12 out)	Flex 24
5	L	36.25	< 10	35.0	Complete	Flex soft
6	L	40.00	< 10	34.8	Complete	Flex soft
7	R	41.25	> 10	32.6	Complete	Flex soft
8	L	32.00	< 10	33.2	Complete	Flex soft
9	L	37.50	> 10	38.7	Complete	Flex soft
10	R	41.25	> 10	31.0	Complete	Flex 28
11	L	35.00	> 10	29.8	Incomplete (11–12 out)	Flex 28
12	R	33.50	> 10	34.4	Complete	Flex soft
13	L	31.25	< 10	33.1	Complete	Flex 28
14	L	33.75	> 10	32.5	Complete	Flex 28
15	R	37.50	< 10	37.1	Complete	Flex soft

Abbreviations: CI, cochlear implant; FF PTA, free‐field pure tone average at 0, 500, 1000, 2000, 4000 Hz.

Final speech perception outcomes, assessed 3 months after the preferred map selection, showed improved speech understanding in noise across all conditions when using bimodal stimulation compared to the HA alone. In the S₀N₀ condition (Summation), mean SNR improved from (4.44 ± 13.8) dB (HA alone) to (−1.54 ± 3.4) dB (bimodal), though the difference was not statistically significant (p = 0.250). In the S₉₀N₉₀ condition (Squelch), mean SNR improved from (5.82 ± 12.7) dB (HA alone) to (−2.82 ± 4.1) dB (bimodal), reaching statistical significance (p = 0.017). In the S₀N₉₀ condition (Head Shadow), mean SNR improved from (3.81 ± 11.0) dB (HA alone) to (−2.15 ± 4.2) dB (bimodal), though the difference was not statistically significant (p = 0.120). These findings indicate a general trend toward improved speech‐in‐noise performance with bimodal stimulation, particularly in the squelch condition, where the benefit reached statistical significance (Figure 1).

Figure 1.

Mean signal‐to‐noise ratio (SNR) outcomes (± SD) in summation (A), squelch (B), head shadow (C) listening conditions for anatomy‐based fitting (ABF) and MAESTRO fitting strategies.

Interim analyses were conducted across all time points prior to the final evaluation, comparing SNR values from the ITA Matrix test across the three spatial configurations and both fitting strategies (ABF and MAESTRO). Median SNR values recorded from the first assessment (2 months post‐activation) through to the seventh control visit (9 months post‐activation) are presented in Table 3, along with a comparative analysis of overall performance differences between fitting strategies.

Table 3.

Italian matrix sentence (ITA Matrix) test results collected across different follow‐up sessions (from the first to the seventh control), comparing performance in the three listening conditions for each fitting strategy (ABF and MAESTRO).

Number	Map	Signal‐to‐noise ratio (dB, Mean ± SD)			p‐value
		Squelch	Summation	Head shadow
1	ABF	1.22 ± 8.2	1.08 ± 4.3	2.45 ± 4.8	0.958
	MAESTRO	1.46 ± 8.6	1.05 ± 5.4	2.45 ± 5.8
2	ABF	−1.47 ± 6.3	0 ± 4.9	1.21 ± 2.8	0.794
	MAESTRO	−0.63 ± 6.4	−0.36 ± 3.5	−0.03 ± 2.8
3	ABF	−1.63 ± 4.7	−1.23 ± 3.5	−0.76 ± 3.9	0.873
	MAESTRO	−1.78 ± 4.6	−1.26 ± 3.1	−1.00 ± 4.2
4	ABF	−3.18 ± 3.8	−2.08 ± 3.4	−1.07 ± 4.3	0.965
	MAESTRO	−2.93 ± 4.0	−2.14 ± 2.7	−1.36 ± 3.7
5	ABF	−2.03 ± 3.7	−2.25 ± 2.7	−1.74 ± 3.4	0.803
	MAESTRO	−1.85 ± 3.9	−1.79 ± 3.3	−2.04 ± 3.8
6	ABF	−3.40 ± 4.6	−1.95 ± 4.1	−1.87 ± 5.5	0.763
	MAESTRO	−3.52 ± 3.6	−2.13 ± 3.9	−2.53 ± 4.6
7	ABF	−3.44 ± 3.8	−2.26 ± 3.2	−2.47 ± 3.3	0.994
	MAESTRO	−3.64 ± 3.6	−2.51 ± 2.8	−2.03 ± 4.0

Abbreviation: ABF, anatomy‐based fitting.

To characterize the SNR distribution, the dataset of measurements was divided into quartiles:

• Q1 (n = 161): SNR < −4.1 dB;

• Q2 (n = 161): −4.1 dB≤ SNR < −1.9 dB;

• Q3 (n = 156): −1.9 dB≤ SNR < 1.2 dB;

• Q4 (n = 158): SNR ≥ 1.2 dB.

A summary of the quartiles distribution is presented in Table 4.

Table 4.

Italian matrix sentence (ITA Matrix) test results stratified by quartiles, with comparison of the two fitting strategies (ABF and MESTRO) across the three listening conditions (head shadow, summation, and squelch).

Map	Quartile 1 (dB, Mean ± SD)			Quartile 2 (dB, Mean ± SD)
	Head shadow	Summation	Squelch	Head shadow	Summation	Squelch
ABF	−6.11 ± 1.7	−5.12 ± 0.7	−6.67 ± 1.8	−2.55 ± 0.5	−3.08 ± 0.6	−2.80 ± 0.6
MAESTRO	−6.14 ± 1.4	−5.08 ± 1.1	−6.10 ± 1.5	−3.11 ± 0.6	−3.05 ± 0.7	−2.86 ± 0.6
p‐value	0.870	0.380	0.130	0.005	0.860	0.770

Map	Quartile 3 (dB, Mean ± SD)			Quartile 4 (dB, Mean ± SD)
	Head shadow	Summation	Squelch	Head shadow	Summation	Squelch
ABF	−0.28 ± 0.8	−0.47 ± 1.0	−0.24 ± 0.9	5.48 ± 5.6	6.23 ± 4.6	7.81 ± 6.6
MAESTRO	−0.47 ± 0.9	−0.86 ± 0.7	−0.76 ± 0.9	5.75 ± 6.6	6.83 ± 5.5	8.47 ± 7.6
p‐value	0.364	0.145	0.089	0.706	0.600	0.823

Note: Data are reported as signal‐to‐noise ratio (SNR) medians in decibe, stratified by quartiles. Lower SNR values indicate better speech perception in noise. Abbreviation: ABF, anatomy‐based fitting.

At the 9‐month visit, patients were asked to select their preferred CI map for continued use based solely on subjective preference: 13 patients (86.7%) selected the ABF map as their preferred fitting strategy, reporting improved sound quality and greater listening comfort, while two patients (13.3%) opted to continue with the MAESTRO map. A binomial test indicated that this preference for ABF was statistically significant (p = 0.004), suggesting that the observed difference was unlikely due to chance.

In the pitch discrimination analysis at 500 Hz, congruence between acoustic and electric pitch discrimination was observed in 10 cases. Of these, six cases (60.0%) were associated with the ABF map and four cases (40.0%) with the MAESTRO map. Although the ABF strategy showed a trend toward improved pitch matching, the difference was not statistically significant (p =  0.392).

At 1000 Hz, correct pitch discrimination occurred in 6 of 22 measurements (27.3%), with five cases (83.3%) linked to the ABF map and one case (16.7%) to the MAESTRO map. This difference, while suggestive, also failed to reach statistical significance (p =  0.149).

4. Discussion

This study aimed to compare the standard MAESTRO CI fitting method with the ABF approach in a cohort of patients undergoing bimodal auditory rehabilitation. Speech in noise performance was evaluated across three listening conditions—head shadow, summation, and squelch—but no statistically significant differences emerged between the two fitting strategies.

Our inclusion criteria, which excluded candidates with low‐frequency thresholds exceeding 65 dB HL in the better‐hearing ear, resulted in a heterogeneous patient cohort. This variability was reflected in the broad range of SNR values. To improve interpretability, the data were stratified into quartiles based on SNR; however, even with this approach, no significant differences were observed between ABF and MAESTRO fittings (Table 4).

These results contrast with existing literature emphasizing the importance of accurate electrode‐to‐place frequency alignment. Prior studies have shown that mismatches between cochlear tonotopy and electrode mapping—often arising from disparities in cochlear duct length and electrode insertion depth—can adversely affect speech perception, particularly in cases of asymmetric hearing loss [14, 15]. In such scenarios, interaural mismatches may further complicate binaural integration, especially in noisy environments where low‐frequency synchrony is critical for speech understanding [6].

Several recent investigations have reported that tailoring frequency allocation to individual cochlear anatomy using ABF leads to better speech recognition and faster adaptation in postlingually deafened adults [16, 17, 18, 19]. While these studies support the theoretical rationale for ABF, our findings differ—possibly due to a key distinction in the study design. In our prospective protocol, electrode type and length were selected preoperatively based on cochlear duct length measurements obtained via Otoplan software. Therefore, this anatomical matching likely led to a default frequency allocation under MAESTRO that closely resembled ABF, thus minimizing the tonotopic mismatch even without a postoperative map reallocation. In our dataset, the median absolute difference in SNR between ABF and MAESTRO maps was less than 1 dB across listening conditions, indicating minimal divergence between the two fitting methods. The similarity in performance likely reflects the common use of anatomically selected electrodes, rather than any ongoing patient‐driven tuning of the MAESTRO map, which was adjusted only once during the initial fitting session. In contrast, many earlier studies applied a one‐size‐fits‐all approach to electrode selection. For instance, Maro et al. [18] reported a mean cochlear duct length of 41.34 mm (range: 37.5–46.8 mm), yet 90% of patients received a Flex28 electrode. Similarly, Fan et al. [19] used Flex28 arrays in nearly half their patients (43.8%), despite average cochlear duct lengths exceeding 40 mm.

Our individualized electrode selection may explain the minimal performance gap observed between the two fitting strategies. This interpretation is supported by Dietreux et al. [20], who found that smaller cochleae were associated with deeper electrode insertion angles and reduced mismatch. Their findings suggest that preoperative imaging alone may be sufficient to guide optimal electrode selection and reduce frequency‐place discrepancies.

These results carry important clinical implications, particularly for pediatric patients. A recent review on otologic planning tools cited limited ABF application in children due to concerns about radiation exposure from postoperative CT scans [21]. Our findings suggest that with anatomically guided electrode placement, it may be possible to achieve effective frequency alignment—and thus favorable auditory outcomes—without the need for postoperative imaging, supporting safer and more streamlined care in pediatric settings.

Nevertheless, the finding that 86.7% of participants ultimately opted for the ABF map is noteworthy. This statistically significant preference reinforces the perceptual relevance of ABF, even in the absence of objective differences. While objective speech‐in‐noise results showed no significant advantage, patients may have perceived improvements in pitch naturalness or listening comfort under ABF. Although our pitch discrimination test did not show statistically significant differences between the two fitting strategies, a greater proportion of congruent pitch matches were observed under ABF at both 500 Hz and 1000 Hz.

A key strength of this study is its prospective design, making it the first to assess ABF in the context of asymmetric hearing loss and anatomically matched electrode selection. However, limitations include the relatively small sample size and the absence of subjective outcome measures such as quality‐of‐life assessments or patient satisfaction surveys. Including these tools might have provided further insight into the clear patient preference for ABF.

5. Conclusions

In summary, our findings indicate that when electrode selection is guided preoperatively by patient‐specific cochlear anatomy, no significant differences in speech‐in‐noise perception are observed between the ABF and standard MAESTRO fitting strategies. This highlights the critical role of individualized electrode selection in optimizing auditory outcomes and achieving alignment between electrical stimulation and the cochlea's natural tonotopic organization ‐ even in the absence of postoperative imaging.

Although ABF did not confer a statistically significant advantage in objective speech‐in‐noise measures, the strong patient preference (86.7%) for ABF suggests potential perceptual benefits beyond what is captured by standard testing—such as improved sound quality or listening comfort. ABF remains a valuable clinical tool, particularly in scenarios where preoperative electrode selection is less precise, where full array insertion is not achieved, or in anatomically atypical cochleae (e.g., with unusually large cochlear duct lengths and reduced insertion angles). In such cases, ABF can help mitigate frequency‐place mismatches and contribute to more natural and effective auditory perception.

Author Contributions

Andrea Canale: overall design of the research, drafting of the manuscript. Arianna Daquino: clinical investigator. Maria Gragnano: clinical investigator. Claudia Filippini: statistical analysis. Marco Boldreghini: clinical investigator. Giuseppe Riva: drafting and reviewing of the manuscript. Andrea Albera: drafting and reviewing of the manuscript.

Funding

The authors received no specific funding for this work.

Ethics Statements

The study was conducted in accordance with the Declaration of Helsinki and approved by the local Ethics Committee of the Università degli Studi di Torino (Protocol No. 12/2021). Written informed consent was obtained from all participants prior to their inclusion in the study.

Conflicts of Interest

The authors declare no conflicts of interest.

Data Availability Statement

The data supporting the findings of this study are openly available in Albera A, Canale A. Anatomy‐based Fitting in Cochlear Implantation—Database. Zenodo; 2024, at 10.5281/zenodo.13748503.