Audiological and Speech Outcomes of Cochlear Implantation in Inner Ear Malformation Compared to the Normal Inner Ear

Abstract

To compare the overall clinical outcomes of cochlear implantation in children with structural inner ear abnormalities, with results of implantation in children with radiologically ‘normal’ inner ears. To study the incidence and performance outcomes of cochlear implantation in children with inner ear malformations (IEMs). It is a retrospective case control study of 57 normal and 57 abnormal inner ear patients operated for cochlear implant between Jan 1, 2014 and Aug 1, 2017, by Veria technique. The age range was between 12 months and 15 years. The prevalence of IEMs was 13.8%. Of the 57 cases with IEMs, 33.3% were of enlarged vestibular acqueduct, 7% were of isolated Incomplete Partition type-2, 21.1% cases were of Mondini’s dysplasia, 3.5% of Incomplete partition type-1 and 8.8% of cochlear nerve hypoplasia. Most commonly encountered malformation was EVA while IP-1 was the least common malformation. A CSF gusher was experienced in 11 cases. There was a statistically significant difference between the CAP and SIR scores of both IEM and control groups, both at 1 year and at 2 years (1 year CAP score p value < 0.001, 2 year CAP score p value 0.002 and 1 year SIR score, p value < 0.001, 2 year SIR score, p value 0.013). There was progressive improvement in the scores (of both groups) at the 2 year mark, compared to 1 year scores. Cochlear implantation is safe and surgically feasible in children with IEMs. However, the hearing outcome and speech perception outcomes are poor than those of patients with normal anatomy, with an exception seen in enlarged vestibular aqueduct.

Introduction

Congenital Hearing loss affects 2–3 infants per 1000 births [1]. Advances in audiometric technologies have enabled us to identify more new born with sensorineural hearing loss (SNHL). Of the children with congenital SNHL approximately 20% are found to have some or the other structural abnormality of the inner ear [2, 3]. These include malformations of the cochlea, cochlear nerve, semi-circular canals, and vestibular aqueducts, to name a few. Cochlear implantation is a procedure which can partially restore hearing for selected patients with SNHL. It has become a standard of care for auditory rehabilitation in paediatric patients with severe-to profound SNHL [4].

Patients with structural abnormalities of inner ear were earlier considered to be a contraindication to implantation due to various reasons which may have an impact on over all outcomes such as, concern about proper electrode insertion, array stability and risk of complications such as facial nerve injury, cerebrospinal fluid leak and a naturally depleted population of spiral ganglion cells. However better understanding of cochleo-vestibular malformations, improved cochlear devices, radiological imaging and surgical techniques, have resulted in better results [5].

It is still not clear to what extent the degree of structural abnormality affects the surgical and auditory outcomes, as various studies have demonstrated conflicting results [4, 6–10]. We have undertaken this study to compare the overall clinical outcomes of cochlear implantation in children with structural inner ear abnormalities, with results of implantation in children with radiologically ‘normal’ inner ears.

Materials and Methods

After obtaining permission from the institutional review board, this retrospective case control study was conducted in the Department of ENT, Civil Hospital, Ahmedabad. The data for the study was obtained from clinical and surgical records maintained by our Department’s medical records section.

The inclusion criteria we used for selecting cases are as follows-age more than 12 months and less than 15 years, presenting with bilateral severe to profound SNHL and underlying inner ear malformation, who had undergone cochlear implantation at our institute. Children who were not available for follow up for a minimum period of two years or did not have imaging or audiological data available, were excluded from the study. Also, children having comorbidities such as developmental delay or psychomotor diseases were excluded from the study.

We identified 57 such cases with varying degrees of inner ear malformations from total population of 413 patients who had undergone cochlear implantation by the Veria technique from 1st January 2014 to 31st December 2017. A control group (1:1) was also established from the remaining population who did not have inner ear structural abnormalities. The controls were all matched for age and duration of cochlear implant use and patients with co-morbidities were excluded from the control group as well. Audiological outcomes were compared by measuring Categories of auditory performance (CAP) and Speech intelligibility rating (SIR) (Tables 1, 2).

Table 1.

The revised categories of auditory performance (CAP) score described by Shepherd centre based on Nottingham cochlear implant program)

Category
0	Unaware of environmental sounds
1	Detects some environmental sounds
2	Responds to some speech sounds
3	Can identify some environmental sounds
4	Understands some spoken words with additional performatives e.g. ‘where is the duck that says quack quack’, ‘give me the car brmm’
5	Understands common phrases e.g. pick it up; it’s bath time
6	Understands some spoken words without performatives e.g. give me the duck’/ ‘go get the car’
7	Responds appropriately to simple questions e.g. what is it?
8	Understands conversations with familiar speakers
9	Understands conversations with unfamiliar speakers
10	Follows recorded stories
11	Uses the telephone with familiar speakers
12	Uses the telephone with unfamiliar speakers

Table 2.

Speech intelligibility rating (O’ Donoghue et al.)

Category	Criteria
5	Connected speech intelligible to all listeners. Child understood everyday context
4	Connected speech intelligible to all listener who has little experience of a deaf person’s speech
3	Connected speech intelligible to all listener who concentrates and lip- reads
2	Connected speech is unintelligible. Intelligible speech is developing in single words when context and lip reading cues are available
1	Connected speech is unintelligible. Pre- recognizable words in spoken language, primary mode of communication may be manual

Preoperative Evaluation

Preoperative work up of patients consisted of pure tone audiometry, otoacoustic emission tests, auditory steady state response and auditory brainstem audiometry. Aided and unaided thresholds of the children were determined by using visual response audiometry or play audiometry. Paediatric and psychiatric assessments were done to rule out developmental delays and other comorbidities. All patients undergoing cochlear implantation in our department had high-resolution CT scan through the petrous bone in axial sections. In addition to CT scan, they all had MRI scans demonstrating the presence of the cochlear nerve. MRI of the cochlea and IAC was also done. MRI examinations were performed on 1.5 T equipment. T1-weighted and turbo-spin echo (TSE) T2-weighted 3-mm axial images, and additional three-dimensional Fourier transform (3DFT) T2-weighted TSE sequences (TR/TE = 1500/250 ms) were obtained on three different planes (axial, perpendicular and parallel to IAC, i.e., oblique sagittal and coronal, respectively) for the purpose of cochlear nerve demonstration. Radiological abnormalities of the inner ear were classified according to the scheme of Sennaroglu [11] wherein he has classified the cochlear malformations into labyrinthine aplasia, cochlear aplasia, and common cavity, incomplete partition of the cochlea, hypoplasia and large vestibular aqueduct syndrome. All children were vaccinated against Streptococcus pneumonia and Meningococci.

Surgical Technique

All patients were implanted by the Veria technique. A cochleostomy of 1.2 mm was done using an electrical drill at the anterior inferior edge of the round window niche. Complete insertion depth was achieved in all cases (IEMs and controls). The cochleostomy was sealed with fascia from the temporal muscle and bone dust. Neural response audiometry (NRT) was carried out immediately after insertion. Patients were kept admitted under sterile conditions for a period of 10 days.

Follow Up Assessment

First tune-up of the implant was done 3 weeks after the operation. Thereafter routine follow up was done every 3 months and included post implantation speech perception and production outcomes measurement using Categories of auditory performance (CAP) and Speech intelligibility rating (SIR), along with audiological and speech training. All children were followed up for a minimum period of 2 years following initial activation.

CAP and SIR score were the audiological tests used to evaluate the efficacy of cochlear implantation in the malformed inner ears and these were compared with respective CAP and SIR scores of the control group to see if there was any statistically significant difference between the scores of the two groups. Statistical analyses were performed using SPSS statistics (version 20, IBM).

Results

Between Jan 1 2014 to Aug 1 2017, 413 cochlear implantations were done out of which 57 cases and 57 controls were selected for the study.

Cases with IEMs

Out of 57 cases, 68.3% (41) were males and 31.7% (19) female. Median age at cochlear implantation was 4 years (range 2–9 years).

Of the 57 cases with IEMs, 19 (33.3%) were of enlarged vestibular acqueduct, 4 (7%) were of isolated IP2, 12 (21.1%) cases were of Mondini’s dysplasia, 2 (3.5%) of IP1 and 5 (8.8%) of cochlear nerve hypoplasia. Most commonly encountered malformation was EVA while IP-1 was the least common malformation (Table 3).

Table 3.

Incidence of various IEMs

Type of IEM	Frequency	Percentage
CN hypoplasia	5	8.8
EVA	19	33.3
IP-1	2	3.5
IP-2	4	7
LO	4	7
Mondinis dysplasia	12	21
Narrow IAC	4	7
SCD	7	12.3
Total	57	100

IEM inner ear malformation, CN cochlear nerve, EVA enlarged vestibular aqueduct, IP incomplete partition, LO labyrinthine ossification, IAC internal auditory canal, SCD semi-circular canal dysplasia

A CSF gusher was experienced in 11 cases (8 with EVA and 1 with IP-1 and 2 with Mondini’s dysplasia), which was controlled by placing a winding temporalis fascia graft along the electrode, after its complete insertion. Aside from complete bed rest and head end elevation no additional precautions, such as hyperosmotic agents or lumbar drainage were required during or following surgery. None of the children with the CSF gusher experienced dizziness, vomiting or CSF leak in the post-operative period. All patients were implanted using the Veria technique. Complete electrode insertion was done through a cochleostomy in all the cases, except in the 4 cases of labrynthitis ossificans, where in the electrode placement proved to be difficult. Out of 57 patients, 29 patients were implanted with SONATA TI100 + standard, 9 patients were implanted with SONATA form and 7 patients with SONATATI compressed, while 5 patients with SONATA flex and 9 patients with COCHLEAR CI24RE (ST).

The mean CAP score at 1 year and 2 year for the cases was 3.9 and 6.1 (range 1–7 and 1–10), respectively, while the mean SIR score at 1 year and 2 year was 2.4 and 3.8 (range 1–4 and 1–5) (Table 4).

Table 4.

Comparative 1 and 2 year CAP and SIR scores of cases and controls

Normal or Abnormal cochlea	CAP score @ 1 year	CAP score @ 2 year	SIR score @ 1 year	SIR score @ 2 year
Normal cochlea
Mean	5.09	7.21	3.44	4.42
Median	5.00	7.00	4.00	4.00
SD	1.353	1.634	0.627	0.625
Range	6	6	2	2
Minimum	2	4	2	3
Maximum	8	10	4	5
Abnormal cochlea
Mean	3.91	6.12	2.47	3.84
Range Median	4.00	6.00	3.00	4.00
SD	1.491	1.852	0.734	1.031
Range	6	9	3	4
Minimum	1	1	1	1
Maximum	7	10	4	5

Control Group

Out of 57 controls, 68.3% (39) were males and 31.7% (18) female. Median age at cochlear implantation was 4 years (range 2–9 years). There was no statistically significant difference between age of cochlear implantation for cases and controls. The mean CAP score at 1 year and 2 year for the controls was 5.09 and 7.21 (range 2–8 and 4–10), respectively, while the mean SIR score at 1 year and 2 year was 3.4 and 4.4 (range 2–4 and 3–5) (Table 4).

We compared the 1 year and 2 year CAP and SIR scores of cases versus the controls. We performed the Mann–Whitney U test of significance since the data was skewed. There was a statistically significant difference in the CAP as well as SIR scores between IEM and control groups. This was noted at both 1 year and at 2 years (1 year mean CAP score for IEM vs control was 3.9 vs 5.09 respectively, p value < 0.001, 2 year mean CAP score 6.1 vs 7.2, p value 0.002 and 1 year mean SIR score 2.4 vs 3.4, p value < 0.001, 2 year mean SIR score 3.8 vs 4.4, p value 0.013) (Tables 5, 6). There was progressive improvement in the scores (of both groups) at 2 years, compared to the 1 year scores.

Table 5.

Test of significance for 1 year and 2 years CAP scores for cases and controls

	CAP score @ 1 year	CAP score @ 2 years
Mann–Whitney U	924.500	1085.500
Wilcoxon W	2577.500	2738.500
Z	− 4.040	− 3.096
Asymp. Sig. (2-tailed)	< 0.001	0.002

Table 6.

Test of significance for 1 year and 2 years SIR scores for cases compared to controls

	SIR score @ 1 year	SIR score @ 2 years
Mann–Whitney U	589.500	1114.500
Wilcoxon W	2242.500	2767.500
Z	− 6.252	− 3.083
p value	< 0.001	0.002

However on comparing CAP and SIR scores within the sub-groups of IEM, there was no statistical difference in case of EVA patients, p > 0.001. While poor speech perception was seen in cochlear nerve hypoplasia and IP-1.

Discussion

Our present study, reflects a significant percentage i.e. 14% (57 out of a total population of 417) of patients with inner ear anomalies and this percentage is within the range reported worldwide. These patients underwent cochlear implantation at our institute. Reports from various other studies have indicated that the incidence of inner ear malformations ranges from 6.9 to 32% [12–14]. Buchman et al. [4] demonstrated that the prevalence of inner ear anomalies among patients who received cochlear implants in their study was 8.8%.

In the current study 68.3% (n = 39) were males and 31.7% (n = 18) were females. A retrospective study by Xiuyong ding showed that, out of 1237 patients undergoing cochlear implant, 59.6% were males [15], demonstrating male prepondarance.

In this study, we have also analysed the individual frequencies of various inner ear malformations according to the Sennaroglu and Saatci classification. The anomaly that was most commonly encountered in our study was, the isolated vestibular aqueduct, accounting for 33.3% (19 cases), the Mondini deformity was seen in 21.1% cases (n = 12). 7% (n = 4) cases were that of isolated IP2 (incomplete partition defect—type 2) and cochlear nerve hypoplasia in 8.8% cases (n = 5). A literature review indicated that the most common anomalies and the frequencies of various anomalies varied among different regions and studies. In a 2010 review article on inner ear anomalies, Sennaroglu [2] reported the following subtypes and frequencies:IP 41% (IP-I/cystic cochleovestibular malformation, 20% IP-II [Mondini deformity] 19%, and IP-III [X-linked deafness] 2%), and LVA 15%.

Complication

Traditionally, cochlear implant surgeries are fraught with complications such as facial nerve paresis and CSF leaks. In our study out of 57 patients, 11 (19.29%) had CSF gusher which was controlled using cochleostomy packing. In a study conducted by Buchman et al. [4] they had CSF gusher in 21% patients all of which were sealed effectively.

Luntz et al. [16] did 10 cochlear implants in 10 patients with inner ear malformations in which he had CSF gusher in 6 (60%) patients.

We did not encounter abnormal facial nerve anatomy in any of our cases. Though there are studies in the literature where a significant number of patients with abnormal facial nerve have been encountered. Buchmann et al. identified facial nerve anomalies in nine (32%) of 28 patients and were almost always associated with SCC aplasia or CC [4]. This can explained as Sennaroglu has recently suggested that six groups of IEM had a higher risk to be associated with a FN anomaly [17]: complete labyrinthine aplasia, rudimentary otocyst, cochlear aplasia, common cavity, cochlear hypoplasia, and incomplete partition of the cochlea type III and our study only have 2 cases of these six anomalies. The facial nerve monitor was not routinely done in our center if the patient had a normal anatomy, but use of a nerve monitor in this group of patients is strongly recommended.

The outcomes of CI is associated with age at implantation [18], onset and duration of hearing loss [19], inner ear anatomy [20], presence of neurodevelopmental disorders [21], and level of psychosocial support [21] and rehabilitation efforts [22] following CI. Keeping all the above parameters constant, except inner ear anatomy, we analysed the outcomes of cochlear implantation for patients with inner ear malformation and those with anatomically normal inner ears.

Theoretically, the outcome of CI might be impeded in patients with cochlear malformations due to possible disorganization or misplacement of the remaining neural structures (REF). Consistent with this, the cases in our study (children with inner ear malformation) had a poorer CAP score at 1 year and 2 year compared to the children with anatomically normal inner ears (mean 3.9 and 6.1 vs 5.09 and 7.2, p value < 0.001 and 0.002). The difference between the CAP and SIR scores was statistically significant, suggesting that the underlying malformation may impact the auditory outcomes after CI in a negative way. Similar outcomes were seen with SIR scores at 1 year and 2 years (mean 2.4 and 3.8 vs 3.4 and 4.4, p value < 0.001 and 0.013). We used the CAP and SIR for assessment of outcomes, as it is well recognized in the international literature [23].

SIR score is used to measure the outcome of cochlear implantation with respect to speech, measuring the intelligibility of speech and the quality, which might be recognizable by the listener.

Despite some variation in the literature, our study results are similar to other studies. Our findings are consistent with a study by Isaiah et al. [24], which showed significant differences in hearing outcomes following surgery for children with cochlear anomalies. One exception noted in their study is that children with EVA had results comparable to children with normal anatomy. In a study by Bille et al. [25], cochlear abnormalities did not significantly impact long term category of auditory performance (CAP) and speech intelligibility rating (SIR) scores. Most studies reported good outcomes in mild anomalies or labyrinthine anomalies (REF). In our study too, audiological results postoperatively demonstrated that patients with EVA had comparable outcomes as compared to patients without any inner ear malformations. Their CAP score ranged from 7 to 9 postoperatively. Buchmann [4] in his study noted several interesting conclusions. First, children with the constellation of IP, EVA, and a dilated vestibule (i.e.,Mondini’s malformation) perform very well on speech perception testing, with at least 10 (63%) of 16 patients achieving some degree of open-set speech recognition. Children with isolated EVA likewise perform well, with eight (89%) of nine achieving openset recognition.

This result corroborates with the result of our study in which audiological results were not affected by the presence of cochleovestibular anomalies. In our study results of patients with LVA is comparable to children with normal anatomy, however those with common cavity, Mondini’s dysplasia or IP2 anomaly do not fare so well.

The challenges in case of IEMs are the uncertain position of the electrode and the possibility of misplacement of the electrode. In all cases, we were able to insert a standard electrode with complete insertion depth, which was also the experience by Sennaroglu et al. [6].

Thus, CI should certainly be offered to patients with inner ear malformations on the same conditions as in other patients.

Conclusion

The data presented in this study reaffirms that patients with inner anomalies may be a suitable candidates for cochlear implantation, though postoperative hearing outcomes may be poorer for children with inner ear anomalies compared to those with normal anatomy. The outcomes may show progressive improvement with subsequent increase in years of use.

In conclusion, this study suggests that the presence of inner ear abnormalities impacts hearing outcomes following cochlear implant surgery. The severity of such anomalies did not significantly affect postoperative hearing outcomes in this population. Given the inconsistent findings in the literature, future studies should consider a prospective, multisite approach with standardized testing and a larger sample size.

Declarations

Conflict of interest

All the authors declare that they have no conflict of interest.

Ethical Approval

All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.

Informed Consent

Informed consent was obtained from all individual participants included in the study.