Perspective Chapter: Cochlear Implant Activation in the Immediate Postoperative Period at the Operating Room

Abstract

Cochlear implant (CI) activation usually takes place at about 30 days postoperative (PO). In our service, CI surgery is performed with local anesthesia and sedation, so Activation is possible with the patient’s cooperation, immediately after the CI surgery, still in the Operating Room (OR). The objective of this study was to provide the patient with hearing experience with the CI and to assess auditory perception immediately after surgery while still in the OR, and to compare Impedance Telemetry (IT), Neural Response Telemetry (NRT) and Comfort (C) level at two moments: in the OR and at the definitive Activation, approximately 30 days PO. Nine adults (12 ears) with acquired (post-lingual) deafness were included. Auditory perception was evaluated through Ling Sounds, musical instruments and clapping, presented in two different programming maps, elaborated using t-NRT, comparing between the two moments. We observed that while still in the OR, the patient can already present auditory detection and recognition responses. The values ​​of impedance, t-NRT and “C” level on both dates differed with statistical significance. We concluded that it is possible to provide the patient with an auditory experience with the CI immediately after surgery, and that the auditory experience and the values ​​of electrode IT, NRT and “C” level vary significantly between the two moments.

1. Introduction

Cochlear implants (CIs) are well established as a successful tool for treating deafness and providing access to sound in individuals with severe to profound bilateral hearing loss. The CI consists of an internal and an external unit. Among the internal unit components are: the receptor-stimulator, which includes the internal antenna, surgically placed next to the skull bone, behind the ear, under the skin, and the electrode array, which is positioned inside the cochlea. The external unit components are the microphone, speech processor and transmitting antenna. The CI partially replaces the functions of the cochlea, transforming sound energy into electrical signals. The auditory perception elicited by the CI depends on the amount of electrical current that passes through the system, and the amount of current needed for the sound stimulus to be perceived varies according to each individual and for each CI stimulation channel. The electrical stimulation parameters, therefore, must be individually adjusted to suit the needs of each patient.

The speech processor of the external unit is normally activated in the speech therapist’s office, about 30–40 days after surgery, time necessary for adequate healing of the surgical wound. For this activation the speech processor is coupled to the computer software and, often using data obtained during the surgery, certain parameters are adjusted: stimulus current level, speed and pulse width. The adjustment process is called programming or “mapping,” and the more accurate the mapping, the greater the potential for achieving adequate speech perception. This process takes time and must be performed regularly, as the use of the processor itself requires new adjustments, personalized programming for each individual, aiming at better hearing gain with increasingly clear, crisp and comfortable sound [1].

Mapping can be performed subjectively, through conditioning techniques and behavioral observation (clinical assessment) or objectively, through examinations. The common procedure in the programming of CI is the determination of the dynamic area for electrical stimulation. The dynamic area is the region between the amount of current that first induces an auditory sensation, that is, the threshold for electrical stimulation (T-level) and the maximum intensity sensation level that the patient will accept for electrical stimulation with comfort (C-level). The dynamic area is determined through psychophysical measures and the ease with which the speech therapist will obtain these levels varies according to a series of factors such as chronological age, mental conditions, time of deafness and other conditions of the patient’s development. In young children or individuals with other associated impairments, obtaining the dynamic area is part of a long and complicated process, especially in the initial periods of CI use. The limited auditory experience that these individuals had before CI surgery, as well as the cognitive and linguistic immaturity to perform the necessary procedures, makes the answers obtained many times inconsistent. Although conditioning techniques and behavioral observation can be used to obtain T and C levels, unfortunately, for a part of the population, these values are not obtained reliably. As a result, these levels are often arbitrarily adjusted. On the other hand, objective measures may include electrically evoked brainstem potentials (EABR), stapedial reflex in response to electrical stimulation in the cochlea or standard measures of acoustic impedance. A more direct way of measuring cochlear nerve function is the Electrically Evoked Compound Action Potential (ECAP). ECAP reflects the synchronized firing of cochlear nerve fibers and is in many ways similar to wave I found in EABR, occurring at a latency of less than 0.5 ms. The Neural Response Telemetry (NRT) is a technique that allows direct measurement of ECAP in implanted patients [1]. Behavioral assessment includes musical instruments of low, medium and high timbres and with weak, medium and strong intensity. These instruments are: rattle, agogô, bell, drum, coco, rattle, castanets, ganzá, reco-reco, caxixi, xylophone, triangle, black black, accordion, whistle, cymbals. There are also methods that are attributed to evaluation, such as knocking on the door and clapping hands [2]. Composing the behavioral assessment, it is possible to make use of the “Sounds of LING” proposed by Daniel Ling, which incorporates phonemes of low, medium and high frequencies, which typically occur in speech [3], a behavioral assessment that tests the effectiveness of the cochlear implant. The concept behind Daniel Ling’s Six Sound Test was to select familiar speech sounds that broadly represent the 250–8000 Hz speech spectrum. It is useful to address detection, discrimination and identification skills, but it is not a test of comprehension [4]. Auditory skills are: detection, which is the most basic level of sound perception, is awareness of the presence and absence of sound; discrimination is the ability to discriminate two or more stimuli, saying whether they are the same or different; the recognition skill that makes it possible to identify, classify and name what you have heard; and listening comprehension is considered the most complex, which allows the individual to understand the meaning of language in oral speech. Attention and memory processes permeate these skills and are essential for their development [4, 5, 6].

Objective assessment includes tests such as Impedance Telemetry that aims to assess the integrity and functionality of the electrode array and Neural Response Telemetry (NRT), which allows recording of the electrically evoked compound action potential (ECAP) of the distal portion of the auditory nerve, using the implant itself to elicit the stimulus and record the ECAP responses, evaluating the functional characteristics of ganglion cells and other neural structures [7, 8], which can be useful for programming the speech processor during the first postoperative adjustment [9] in patients who do not provide feedback, such as in young children. NRT is obtained in approximately 80% of the evaluated individuals and its technique can be a valuable tool in confirming the integrity of the internal device, in the objective determination of which electrodes can be included in a given map, the best stimulation speeds and speech coding strategies, as well as in the estimation of the dynamic area, with extreme clinical importance [10]. ECAP thresholds can be useful to predict the minimum and maximum levels that determine the dynamic area for electrical stimulation, these levels can be named and defined differently for the different brands of CI on the market. The dynamic area is the region between the amount of current that first induces the auditory sensation, that is, the threshold for electrical stimulation (T-level) and the maximum intensity sensation level that the patient will accept for electrical stimulation (C-level). This is done so that the CI is programmed within the loudness range which allows speech sounds and other sounds to be audible but not uncomfortable [11]. It is generally determined through psychophysical measurements, although objective or electrophysiological measurements of hearing can be used [12]. However the correlation between ECAP thresholds and psychophysical thresholds is affected by many factors, and analysis showed that there is only weak evidence to support the use of eCAP data for CI fitting purposes; and they are also an equally weak predictor for both T- and C-levels [13].

With the use of local anesthesia and sedation technique for CI surgery in our service [14], at the end of the surgery, still in the operating room, the patients are already awake and when the NRT is performed, they report listening to the stimuli that occur during the testing and sometimes may question if that is what they will hear after activating the CI.

With this report, we realized that it would be possible, even in the operating room, to activate the speech processor with patient cooperation, allowing the patient auditory perception immediately after surgery, still in the Operating Room. In this way we mapped the activation using the NRT thresholds to determine and test the dynamic field (T and C levels). After this activation, dressing of the wound was performed and the definitive activation was scheduled for about 30 days after this date, when adequate healing is achieved.

Therefore, the objective of this work was to carry out activation immediately after surgery while still in the Operating Room, providing the patient with an auditory experience with the CI, determining the dynamic field (T and C levels) and performing Impedance telemetry (IT), and Neural response telemetry (NRT) which is based on the measure of the Electric Compound Action Potential (ECAP) thresholds, using two different programming maps (Map 1 and 2) and comparing them at the moment of surgery and at the definitive Activation, 30 days later.

2. Materials and methods

This was a prospective analytical, longitudinal study, approved by the Institutional Review Board under number 12855619.9.0000.5529, carried out in nine adult patients (six unilateral and three bilateral cases, 12 ears total), all with acquired (post-lingual) hearing loss, who underwent CI surgery under local anesthesia and sedation according to our standard protocol published elsewhere 14 and who consented for the CI activation in the Operating Room (OR). Surgery was performed in the Operating Room of a tertiary hospital in the city of Curitiba, Parana, Brazil.

Patients either already were using Hearing Aids and underwent CI surgery (unilateral cases) or had a CI in one ear and underwent sequential CI surgery (bilateral cases). The sample included both sexes. Exclusion criteria were not consenting for the activation in the OR, not undergoing intraoperative NRT or failing (having no response) in assessment of impedance telemetry, or in obtaining intraoperative ECAP.

The CI model was CI24RE (CA) and CI422. Measurements were obtained during CI surgery through computer software NRT Custom Sound EP 3.2 (3.2.3855), connected to the portable programming unit and speech processor and the transmission antenna of the CI (software developed by Cochlear Corporation).

The NRT system allows pacing and recording on any pair of electrodes, in monopolar or bipolar modes. Normally, monopolar mode is used. The stimulation pulses are presented to a specific intracochlear electrode that has the MP1 extracochlear electrode as a reference, positioned below the temporal muscle flap. Another intracochlear electrode located in the vicinity is used as a recording electrode, having as reference another extracochlear electrode (MP2), located on the receptor-stimulator. Generally, the intracochlear electrode used for ECAP recording is located approximately 1.5 mm more apical (space of two electrodes) in relation to the position of the pacing electrode. For example, if the pacing electrode selected is 5, the recording electrode will be 7. The pair of active and reference electrodes used for pacing must be different from the pair of electrodes used for recording in order to reduce artifact. Electrodes with high impedance values or electrodes that are out of compliance should not be used for measurements.

Firstly, in the OR, Impedance Telemetry (IT) was performed to assess integrity and functionality of the electrodes. Impedances were measured on the 22 electrodes in monopolar MP1, monopolar MP2, monopolar MP1 + 2 and Common Ground (CG) modes. Values were considered normal when between 1.5 kΩ and 20 kΩ in MP1, MP2 and MP1 + 2 modes and between 0.7 kΩ and 20 kΩ in CG mode. Only electrodes that presented impedance within normal limits, according to software standards, were used for recording Neural Response Telemetry (NRT). Then, measurements of intraoperative t-NRT (NRT threshold) were performed in all electrodes, and we used the response in at least five electrodes for analysis, dividing the cochlea into four regions: electrodes 01 to 05, 06 to 10, 11 to 15 and 16 to 22. The current level (CL) in each electrode initiated at 150CL, with an interval of 6CL between one stimulus and the other, up to the maximum stimulation of 255CL or until reaching t-NRT. Interpulse interval was fixed at 500 μs, stimulation speed at 80 Hz in series of 25 μs per phase.

We used Cochlear Corporation Custom Sound EP software to obtain objective measurements of IT and NRT, and Custom Sound to assemble maps and perform Activation following surgery. With the Nucleus 5 - CP 810 processor, two maps were created with stimulation speed of 900 Hz, 8 maxima, volume 6, sensitivity 12, with different levels of “T” and “C”:

• Map 1: Created with “C” levels by subtracting 10 current units from tNRT;

• Map 2: Created with “C” levels equal to tNRT.

And “T” levels were estimated at 40 current units below “C” level.

Afterwards, each patient was evaluated in the OR, with either map, by:

• Ling sounds (/a/, /i/, /u/, /m/, /s/, /sh/);

• Instrumental sounds: bell rattle (2KHz to 6KHz), coconut shells (600 Hz to 3KHz), bell (4KHz to 8KHz) and drum (250 Hz to 600 Hz) (Figure 1);

• Claps (one or two claps).

Figure 1.

Instruments sounds used: Bell rattle, coconut shells, bell and drum.

Presentations were performed in a closed set, the patient being informed about which sounds would be presented: Ling Sounds, musical instruments or clapping. All were performed approximately 60 cm from the speech processor. The patient was still lying on the surgical table. Responses were observed and noted. The analysis of the auditory perception responses, immediately after surgery, in the two different maps tested, was done by observational analysis. The behavioral tests were evaluated based on the auditory skills: detection, discrimination and recognition.

The second assessment was done at the day of definite activation, 30 days postop, with IT and NRT.

Results were described as mean, standard deviation (SD), median, minimum and maximum. To compare NRT and IT between the two evaluation moments, Student’s t test was used for paired samples. The normality condition was evaluated by the Shapiro-Wilk test. Values ​​of p < 0.05 were considered of statistical significance. Data were analyzed using the computer program Stata/SE v.14.1. StataCorpLP, USA. Detection, recognition and non-detection to sound were considered for the observational analysis of responses to Ling Sounds, instrumental and clapping sounds.

3. Results

The analysis performed was based on data from 12 ears of nine patients (six unilateral and three bilateral cases). The CI electrode bundle CI24RE (CA) was used for 10 ears and two ears with CI422. Based on the presented tNRT, Maps 1 and 2 were tested. We based the analysis in that sound detection precedes auditory recognition, so all percentage calculations were based on the total detection and recognition being 100%, and from this we calculated the percentage of recognition.

For the detection, non-detection and recognition of Ling Sounds in the 12 ears, we found that with Map 1: the 12 ears detected the phoneme /a/, but only 3 ears recognized it, the 12 ears detected /i/, but only 2 recognized it, 11 detected /u/, 3 recognized it, 11 detected /m/, 12 detected /sh/ and one recognized it, 12 detected /s/. With Map 2, for the phoneme /a/ 12 detected and 5 recognized it, for /i/ 12 detected and 2 recognized it, for /u/ 12 detected and 6 recognized it, for /m/ 12 detected and 2 recognized it, for /sh/ 12 detected and 4 recognized it, and for /s/ 12 detected and 1 recognized it. Based on these data we can say that we obtained 12.85% of recognition of LING sounds for Map 1 and 27.77% with Map 2.

For the detection, non-detection and auditory recognition of musical instruments with Map 1, there was detection of the bell rattle, coconut shells, bell and drum for the 12 ears, 3 of which recognized the bell and 3 recognized the drum. As for Map 2, there was detection of the bell rattle for 11 ears, for the other instruments there was detection with the 12 ears, and 2 ears recognized the bell, and 3 the drum. Therefore, it was possible to observe 25% of recognition of the bell and drum for Map 1 and 16.66% of recognition for the bell and 25% for the drum with Map 2.

For the detection, non-detection and recognition of one or two claps we verified that palms were possible to be detected and recognized. With Map 1 we found 41.66% of recognition for one or two claps and with Map 2 the recognition was 33.33%. With both Map 1 and Map 2, some patients reported detecting and discriminating the sound, but they did not recognize it (they did not know what they were hearing). They detected but made mismatches between Ling Sounds, between the instruments, and named instruments such as “hiss,” “beat,” “a thin sound,” or “papapa.” With Map 1, the rattle (that was not presented) was also mentioned, and one patient reported hearing well and one reported hearing it low. With Map 2, three patients reported being too loud, one patient reported discriminating between low and high sounds. All these responses were considered detection. Only those who recognized the sound being presented were considered recognition. Results were similar for all patients in both maps, although discomfort was reported with Map 2 (stronger) by three patients.

Regarding tNRT, Table 1 presents descriptive statistics of tNRT in the operating room and on activation and the mean difference between tNRT in the two situations.

Electrode	Variable	Mean ± SD	Median (min-max)	Mean reduction (OR – activation)	p*
Electrode 1	tNRT OR	180.7 ± 19.6	182.5 (145–205)	14.8	<0.001
	tNRT activation	165.8 ± 21.6	169.5 (126–191)
Electrode 6	tNRT OR	187.3 ± 14.9	184.5 (162–215)	15.2	<0.001
	tNRT activation	172.1 ± 10.8	173 (155–194)
Electrode 11	tNRT OR	193.3 ± 16.5	199,5 (165–217)	13.1	0.015
	tNRT activation	180.3 ± 20.7	189 (134–203)
Electrode 16	tNRT OR	184.1 ± 20.4	185 (141–220)	11.2	0.016
	tNRT activation	172.9 ± 16.4	176 (135–191)
Electrode 22	tNRT OR	170.9 ± 19.2	175 (140–204)	13.8	0.091
(n = 11)	tNRT activation	157.1 ± 18.6	154 (129–190)

Table 1.

Comparison between tNRT values: in the operating room (OR) and at the activation day (after 30 days).

^*

Student’s t test for paired samples, p < 0.05.

OR = operating room, activation = activation day; min-max = minimum and maximum values; SD = standard deviation.

Based on Table 1, we can see that there was a statistically significant difference between tNRT obtained in the OR and on the day of definitive activation for electrodes 1, 6, 11 and 16, only not for electrode 22 (but p value suggests that there is a tendency for a statistically significant difference).

Table 2 shows descriptive statistics of tNRT in the operating room and the measurements of “C” level informed by the patient on the day of activation and the mean difference between them. There was a statistical difference between tNRT measurement performed in the OR and the measurement of the “C” level informed by the patient on the day of activation as a comfort level. We can observe that the values of the “average and median” of the current levels for each electrode decrease between one situation and another.

Electrode	Variable	Mean ± SD	Median (min-max)	Mean reduction (OR – “C”)	p*
Electrode 1	tNRT OR	180.7 ± 19.6	182.5 (145–205)	40.7	<0.001
	“C” level activation	139.9 ± 21.1	140 (102–175)
Electrode 6	tNRT OR	187.3 ± 14.9	184.5 (162–215)	43.5	<0.001
	“C” level activation	143.8 ± 14.1	146 (122–162)
Electrode 11	tNRT OR	193.3 ± 16.5	199.5 (165–217)	46.4	<0.001
	“C” level activation	146.9 ± 12.7	147.5 (130–165)
Electrode 16	tNRT OR	184.1 ± 20.4	185 (141–220)	38.2	<0.001
	“C” level activation	145.8 ± 12.3	147 (126–165)
Electrode 22	tNRT OR	170.9 ± 19.2	175 (140–204)	29.4	0.001
(n = 11)	“C” level activation	141.5 ± 11.4	138 (127–165)

Table 2.

Comparison between tNRT in the operating room (OR) and the “C” level assessment (activation with responses) (after 30 days).

^*

Student’s t test for paired samples, p < 0.05.

OR = operating room, activation = activation day; min-max = minimum and maximum values; SD = standard deviation.

We can see clearly how in the three situations (tNRT on the day of surgery, tNRT on the day of definitive activation and “C” level on the day of activation) current level decreases.

Table 3 presents descriptive statistics of IT in the OR and at activation 30 days later and the average difference between IT in the two situations. There was a statistically significant difference between the two moments. It is clear how Impedance values increased consecutively.

Electrode	Variable	Mean ± SD	Median (min-max)	Mean increase (ativ – CC)	p*
Electrode 1	IT OR	57 ± 1.5	5.7 (3.9–8.7)	8.3	<0.001
	IT activation	13.9 ± 2.2	14 (9.6–17)
Electrode 6	IT OR	5.1 ± 2.1	5.1 (2.5–9.7)	7.9	<0.001
	IT activation	13 ± 2.2	13.1 (7.7–15.7)
Electrode 11	IT OR	5.6 ± 2.2	5.2 (2.9–10.9)	6.2	<0.001
	IT activation	11.8 ± 2	12.2 (7.6–13.8)
Electrode 16	IT OR	5.8 ± 2.0	5 (3.6–10.1)	5.6	<0.001
	IT activation	11.4 ± 2.8	12.3 (6.6–15.1)
Electrode 22	IT OR	6.3 ± 2.6	5.3 (3.2–10.6)	5.4	<0.001
	IT activation	11.8 ± 3.3	12 (4.4–16.8)

Table 3.

Comparison between electrode impedance telemetry (IT) in the operating room (OR) and in the activation day (after 30 days).

^*

Student’s t test for paired samples, p < 0.05.

OR = operating room, activation = activation day; min-max = minimum and maximum values; SD = standard deviation.

4. Discussion

We observed that in our CI patients because of the local anesthesia and sedation technique [14] it was possible to perform CI activation at the immediate period of surgery, while still in the Operating Room (OR). We also found that we can use tNRT as a “C” level of stimulation.

In our study, we found that there was a statistically significant difference in telemetry (both IT and tNRT comparing OR day and day of activation), in that IT increased after surgery, and tNRT and C level decreased after surgery; in the OR there was detection in behavioral tests with Ling Sounds, musical instruments and clapping, but discomfort to the sound was also reported. Due to this discomfort, we therefore suggest using a lower current level for activation than the one found on tNRT at the day of surgery.

Lai et al. [15] showed that intraoperative NRT data were generally stable enough to be used to assist in initial speech processor mappings, and it was not possible to predict changes in the map’s subjective threshold or comfortable loudness levels based on changes observed in the NRT data. In our study, we warn that this should be done with caution, because there was a statistically significant difference when comparing tNRT responses on the day of surgery with the measurement of “C” level (comfort threshold) at the day of activation. These values decreased, tNRT in OR was higher than tNRT on the day of definitive activation, and this in turn was higher than the “C” level measured on the day of definitive activation. Unlike tNRT, we observed that Impedance values ​​increased from the day of surgery to the day of activation, but it should be noted that IT was the first procedure performed and the stimulation current had still passed. Often, when we perform CI activation in young children, who do not cooperate or who do not allow tNRT to be performed, we use tNRT performed in the OR as basis for building the activation map. It is important to know, although it is information given by an adult, how much the tNRT data performed on the day of surgery can help but also be uncomfortable in the listening experience. In this study, we observed that the map with the “C” level at the tNRT threshold, although considered uncomfortable by two patients, because it was higher, allowed the detection of Ling Sounds, clapping and the detection of instruments only reporting whether bass or treble. Of course, with children we should rely much more on observing behavioral responses and make use of other objective measures such as the investigation of the electrically evoked stapedial reflex threshold [16]. It is important to emphasize that when programming levels are determined based on ECAP thresholds, the stimulation can be uncomfortably high, particularly in the basal electrodes [17, 18, 19]. We could observe that there was a decrease of about 40cu for the basal electrodes between the ECAP thresholds (tNRT) on the day of surgery (OR) and the “C” level reported by the patient at the Activation day.

More research is needed to affirm whether the eCAP can be used to predict current levels of individual CI recipients at the day of activation [13]. But our findings can be of help.

Behavioral measures, even if minimally observable, are important for CI programming. Objective electrophysiological measurements help predict behavioral levels, but these alone cannot replace the accuracy of a behavioral map [20]. Research has revealed a stronger correlation between ECAP threshold and “C” level than between ECAP threshold and “T” level [21]. We believe that our patients found it easy to detect the sound and even to recognize it, because they had all post-lingual hearing loss and already wore a CI in one ear or hearing aids. We have previously studied pre- and post-lingual differences in programming elsewhere [22]. It was possible to observe with Map 2, level “C” at the NRT threshold, a higher percentage of detection and recognition of Ling sounds when compared to Map 1. Regarding the instruments, we used instruments with different sound spectrum: for the bell rattle, detection was considered the fact that they identified the sound and reported being strong or weak and being high or low, we could observe that only for one ear with Map 2 the bell rattle was not detected.

In this research, it was possible to observe that even immediately after the insertion of the electrode bundle in the cochlea, while still in the OR, the patient can already present auditory detection and recognition responses with both objective and behavioral measures. We believe in the importance of activation as early as possible, but we agree that the healing period must be respected. This auditory experience may make them calmer and hopeful to wait for the definitive Activation date in 30 days and this observation may also help in the programming of the CI.

5. Conclusion

This chapter shows that CI activation in the Operating Room, immediately after surgery with local anesthesia and sedation, is possible. We could provide the patient with a hearing experience with the CI while still in the OR with auditory detection and auditory discrimination of different types of stimuli, but with poor recognition. Maps with higher current offer better results, but also provide auditory discomfort. Impedance telemetry values increased from the day of surgery to the day of definitive activation, at 30 days, and Neural Response Telemetry values decreased for the same period, and both were statistically significant.

Conflicts of interest

None to declare.