Residual low-frequency hearing after early device activation in cochlear implantation

The cochlear implant (CI) is a standard treatment for patients with severe to profound hearing loss [1, 2]. The usage of a CI improves the speech recognition [3] in everyday life as well as the quality of life [4]. Since the introduction of the CI, the indication criteria have been expanded. Initially only completely deaf patients were treated with a cochlear implant [5], later also patients with low-frequency residual hearing [6]. Also the treatment procedure has changed over time. Until a few years ago, the sound processor was initially activated after a standard healing phase of 3–6 weeks [1, 2, 7, 8], to ensure that the wound healing process was completed. With advances in surgical technique and instrumentation, as well as the use of the small incision technique [9], the wound is smaller and heals faster, with less postoperative pain [10] and wound swelling [11]. This allows earlier initial sound processor activation within only 2–3 days after surgery. This significantly shortens the time period between surgery and first fitting [12], thus hearing rehabilitation may start much sooner. Previously, we demonstrated the feasibility and safety of early processor activation after CI surgery also after long-term follow-up [12, 13]. As in other studies [3, 14, 15] it was shown that the early fitting of the sound processor is a safe and effective procedure with no known additional medical risks or complications. Facilitating early fitting, comparable speech recognition in quiet was achieved compared to standard fitting [12, 13]. Despite these advantages, until now it is unknown if early sound processor activation may influence residual hearing in patients following cochlear implantation. Hearing preservation today is a general aim in CI surgery. For postoperative hearing preservation, an atraumatic surgical technique for structure preservation is required [5, 16, 17]. Helbig et al. [16] presented data on long-term acoustic hearing preservation after CI und concluded that preservation is feasible in patients fitted 4–6 weeks after implantation (standard healing phase).

The advantage of using a hearing preservation technique for CI surgery [18‐20] is that the acoustic low-frequency residual hearing can be used for electric-acoustic stimulation (EAS). It is a combination of acoustic stimulation via hearing aid function in the apical area of the cochlear and electrical stimulation of the basal parts of the cochlear [5, 21]. The use of low-frequency residual hearing in EAS patients leads to an improvement in speech recognition [22]. The EAS/hybrid usage of the sound processor has become established in patients with severe high-frequency hearing loss [5, 20].

The investigation of hearing preservation associated with the early initial processor activation is of interest because it is not clear whether early electrical stimulation has an effect on residual acoustic hearing function.

Intracochlear healing processes are completed after a period of 4–5 weeks [2]. It remains unclear, whether acoustic and electrical stimulation after traumatic electrode insertion could lead to apoptosis of the hair cells and thus to the loss of residual hearing. The insertion of the electrode is potentially traumatic for the sensitive hearing organ [23]. Therefore, it is unknown whether early processor activation affects hair cell function and whether the immune response induced by electrode insertion [24], in combination with early electrical stimulation, may deteriorate hearing thresholds in the long term. In case of other traumatic events, such as noise trauma or acute sudden hearing loss, acoustic overstimulation with high stimulation levels is avoided. For example, the German audiological association ADANO recommends avoiding examinations such as BERA (brainstem evoked response audiometry), CERA (cortical evoked response audiometry), stapedius reflex and electrocochleography with high sound pressure levels within eight days after sudden hearing loss [25].

The aim of this study therefore was to examine whether early fitting of the sound processor has an impact on the postoperative low-frequency residual hearing as a potential negative side effect following the early fitting procedure. Therefore, long term data on residual hearing loss and of speech recognition in quiet and in noise were compared over a period of 12 months between patients who received an early device activation (study group) and patients whose processor was initially activated after the standard healing phase (control group). It is assumed that long term preservation of the residual hearing is possible after early initial processor activation.

To our knowledge, there are currently no studies that have examined the postoperative residual hearing preservation after CI surgery in association with the early sound processor activation within a long term follow-up of 12 months.

A total of 57 patients were enrolled in the prospective study. The early fitting group (EF) included 32 patients (18 males, 14 females). In this group, the sound processor was initially activated within a maximum of 15 days after CI surgery. In the control group (CG), which included 25 patients (8 males, 17 females) the processor was first activated after a standard healing phase of 3–6 weeks. The CI surgery was carried out in approximately the same time span in both patient groups. All patients had a minimum age of 18 years and fulfilled the inclusion criteria for an EAS/hybrid CI surgery. Therefore, a sufficient residual low-frequency hearing was necessary. The criterion was set at a maximum hearing loss of 70 dB HL at a test frequency of 500 Hz. The assignment to the respective study group (EF or CG) was non-randomized and based on the willingness of the patients to participate in the early fitting process. Patients who did not want an early processor activation were assigned to the CG. The demographic data of the patients are shown in Table 1. In both study groups, the “small incision” [9] technique was applied for CI surgery. A DVT (digital volume tomography) was performed postoperatively in all patients to confirm the placement and position of the electrode. The first fitting interval with initial device activation was accomplished in three fitting appointments. After initial device activation the two additional fitting appointments during the first fitting interval took place within two weeks. In both groups, regular follow-up visits were carried out after 3, 6 and 12 months following initial processor activation. In all clinic appointments, the wound healing status was assessed by a physician, the sound processor was programmed by an audiologist and audiometric assessments were performed. For data analysis, the air conduction thresholds were regarded, which were measured via headphones (Telephonics TDH-39P, Farmingdale, NY, United States). To assess low-frequency residual hearing, the PTA_low (pure tone average) was calculated as the average hearing loss at 125 Hz, 250 Hz and 500 Hz. Furthermore, speech recognition in quiet was measured using the Freiburg multisyllabic and monosyllabic word test at 65 dB SPL presentation level in free field condition. In addition, the 50% speech reception threshold (SRT) of the adaptive Oldenburg Sentence Test (Oldenburger Satztest, OlSa) [26] in quiet was determined in free field at a starting level of 55 dB SPL. To assess speech recognition in noise, the OlSa was carried out with adaptive noise level (olnoise signal) [26] in free field condition. A fixed speech level of 65 dB SPL and a noise starting level of 60 dB SPL was used to determine SRT. Speech and noise signal were presented from the front (S₀N₀). Both OlSa tests were performed in a closed setup. The study data were collected preoperatively, at the third day of the first fitting interval and at the follow-up intervals after 3 and 12 months following CI surgery. The study procedure is shown schematically in Fig. 1.

The early fitting procedure allowed a much earlier initial activation of the sound processor than after awaiting the standard healing phase. The EF (n = 32) was first fitted after 3.0 days (median) and thus significantly earlier (U = 1.000, Z = -6.451, p < 0.001, Mann–Whitney-U-Test) than the CG (n = 25) after 28.0 days (median). The study data were collected over a period of 12 months. Since not all patients were able to attend their follow-up appointments after 3 and 12 months following initial device activation for personal or therapeutic reasons (e.g. ongoing stationary rehabilitation), not all data sets were complete.

The demographic data (see Table 1) showed a comparable age of both experimental groups (EF 58.1/CG 59.9 years, mean). At 27 years (mean), the CG had a longer duration of profound hearing loss than the EF, which had a duration of 17.5 years. In addition, CG patients had more hearing aid experience (25.6 years) than EF patients (14.5 years). However, the duration of profound hearing loss can often not be precisely determined. The patient’s memory of the onset of the hearing loss is often vague or cannot be clearly determined because of progressive hearing loss, especially if residual hearing is still present.

Nevertheless, it is known that the duration of profound hearing loss has an impact on the performance with CI [27]. In the EF subject group 8 patients were implanted with Flex24 (6) and Flex26 (2) electrodes, in the CG subject group 9 patients with Flex20 (1), Flex24 (7) and Flex26 (1). Those are shorter and more flexible electrodes that can promote the preservation of postoperative residual hearing [28]. In patients fitted with a device manufactured by Cochlear, the straight electrode arrays CI522/CI622 (EF 11/5, CG 12) and the perimodiolar flexible CI532 (EF 3) were used. No differences in hearing preservation between CI522 and CI532 were expected. Preservation of residual hearing is feasible with both electrode arrays [29].

As expected, initial processor activation of patients who received an early fitting procedure took place 3 days after surgery, whereas in the CG an average of 28 days was observed. Thus, the EF patients were able to gain listening experience much earlier.

Early fitting of the sound processor had no negative effect on the preservation of the postoperative low-frequency residual hearing. In both EF and CG, long term preservation of the low-frequency hearing was possible. Both groups showed a comparable development of EAS/hybrid CI aided speech recognition within the first year after CI surgery. Due to the sufficient long term preservation of the residual low-frequency hearing, EF patients could benefit to the same extent from EAS/hybrid usage as patients who were fitted after the standard healing phase. The results of this study showed that early device activation is possible in EAS/hybrid CI patients with sufficient residual low-frequency hearing.