HRCT-BASED PREDICTION FOR COCHLEAR IMPLANT OUTCOMES OF CASES WITH INNER EAR AND INTERNAL AUDITORY CANAL MALFORMATIONS
Introduction
Inner ear and internal auditory canal (IAC) malformations account for approximately 20–35% of congenital sensorineural hearing loss1,2 and an increasing number of children with inner ear and/or IAC malformations underwent cochlear implantation. According to Sennaroglu’s classification of inner ear malformations, which is the most widely accepted, the inner ear malformations are divided into labyrinth aplasia, cochlear aplasia, common cavity (CC), incomplete partition type I (IP-I), type II (IP-II), and type III (IP-III), cochlear hypo-plasia type I (CH-I), type II (CH-II), and type III (CH-III), and large vestibular aqueduct syndrome (LVAS).1,3 This classification is essential to investigate the etiology of the inner ear malformations, but with respect to predicting cochlear implant (CI) outcomes, it might not be enough, because it does not include IAC malformations such as narrow IAC (NIAC) and hypoplasia of the bony cochlear nerve canal (HBCNC). These IAC malformations are highly associated with cochlear nerve deficiency (CND), which has a negative impact to CI outcomes.4,5
The purpose of this study was to establish a new CT-based categorization which is simple and includes both inner ear and IAC malformations for predicting CI outcomes.
Materials and methods
Between 2004 and 2010, 98 subjects who were under 20 years old underwent cochlear implantation at Kobe City Medical Center General Hospital. Among them, CT revealed that 24 subjects had inner ear and/or IAC malformations at the implanted side.
We evaluated inner ear and IAC malformations at the implanted side based on CT findings. Sennaroglu’s classification was used to classify inner ear malformations and the IAC malformations were classified into NIAC and HBCNC. NIAC was diagnosed when the maximum diameter of the IAC was less than 2 mm.2 The width of the bony cochlear nerve canal (BCNC) was evaluated at the mid-portion between the base of the modiolus of a cochlea and the fundus of the IAC on axial images. When the diameter of the BCNC is less than 1.5 mm, it is diagnosed as HBCNC.5 CND was diagnosed when a cochlear nerve (CN) appeared smaller than the facial nerve on the parasagittal MR imaging.
We categorized inner ear and IAC malformations into four groups by two criteria: (1) the presence or absence of a bony modiolus in the cochlea; and (2) the diameters of IAC and BCNC. In this categorization, both Group 1 and Group 3 have a bony modiolus in the cochlea, while Group 2 and Group 4 lack this component. Both IAC and BCNC are normal in Group 1 and Group 2, but NIAC or HBCNC was observed in Group 3 and Group 4. Sennaroglu’s classification of inner ear malformations clearly discriminates between the presence and absence of a bony modiolus in the cochlea. According to his classification, a bony modiolus is present in IP-II, CH-III, LVAS, and a normal inner ear, while CC, IP-I, IP-III, CH-I, and CH-II have a cystic cavity without a bony modiolus.3
We evaluated CI outcomes by category of auditory performance (CAP) scores,6 hearing thresholds of pure-tone sounds, infant word speech discrimination scores, and monosyllabic word speech discrimination scores at one to three years after implantation. A subject with 0–4 CAP scores could not even understand common phrases without visual language and, therefore, we defined 5–7 CAP scores as a good CI outcome and 0–4 CAP scores as a poor one.
Results
We categorized our patients based on the two criteria as described above. In this study, there was no case categorized in Group 4. Group 1, Group 2, and Group 3 consisted of 11, 7, and 6 cases, respectively. MR imaging revealed CND in all cases of Group 3.
The post-operative CAP score was equal or over five in all cases of Group 1, but did not exceed four in all of Group 3. In Group 2, the post-operative CAP score was still four in two cases even after three years of CI usage, but reached to five or six in the remaining five cases. As shown in Figure 1, using our new categorization instead of the existing classifications, we can better discriminate between a good and poor outcome.
We examined speech discrimination scores of 22 cases except for two cases of Group 3 whose response to voice was poor. The correct percentage of the closed-set infant word discrimination test was ≥ 80 in all cases of Group 1, while the score ranged from 40 to 60 in tested cases of Group 3. The correct percentage of Group 2 widely varied between cases, ranging from 55 to 100. The open-set monosyllabic word discrimination test is much more difficult than the closed-set infant word discrimination test and, therefore, only 17 of 24 patients, who were over five years old and used their CI for more than two years, underwent this examination. All tested cases of Group 1 and 3 cases of Group 2 could answer correctly in equal or over 80% of accuracy. The correct percentage of the remaining cases, including all tested cases of Group 3, was ≤ 30.
Fig. 1. A. The post-operative CAP score of each type of malformations based on the existing classifications. One case with both CH-III and HBCNC is plotted twice (*). B. The post-operative CAP score of each group of our new categorization. In both graphs, the members of Group 1, Group 2, and Group 3 are represented by a circle, diamond, and triangle, respectively.
In this study, we established a new CT-based categorization including both the inner ear and IAC malformations. This categorization is defined by two criteria; (1) the presence or absence of a bony modiolus in the cochlea; and (2) the diameters of IAC and BCNC. We focused on these structures because the bony modio-lus contains spiral ganglion cells, the major target of CI-mediated electrical stimulation, and their axons go through BCNC and IAC.
Group 1, which is defined by the presence of a bony modiolus of the cochlea with a normal IAC and BCNC, showed the best CI-aided hearing performance among three groups. The high proportion of post- or peri-lingually deaf cases might also contribute to the high CI outcomes of this group.7 Group 2 is defined by the absence of a bony modiolus with a normal diameter of IAC. The CAP score and speech discrimination score varied widely between cases in this group, but five out of seven cases could understand common phrases without visual languages. Group 3 is defined by the presence of a bony modiolus in the cochlea with NIAC or HBCNC and their post-operative improvement of hearing performance was limited. Visual languages were necessary for them to understand common phrases even after long usage of their CI. MR imaging revealed CND in all cases of Group 3, which might be responsible for their poor outcomes.
Conclusion
Our new CT-based categorization, which was based on the presence or absence of a bony modiolus in the cochlea and the diameters of IAC and BCNC, was effective in predicting CI outcomes of children with inner ear and/or IAC malformations. The CI outcomes were the best in Group 1, followed by Group 2 and Group 3. All cases of Group 1 showed good CI outcomes and could communicate orally. On the other hand, all cases of Group 3 showed poor CI outcomes and used lip-reading or sign language to understand common phrases. The CI outcomes of Group 2 varied between cases, but many of them showed good CI-aided hearing performance.
References
1.Sennaroglu L, Saatci I. A new classification for cochleovestibular malformations. Laryngoscope 112:2230–2241, 2002
2.Papsin BC. Cochlear implantation in children with anomalous cochleovestibular anatomy. Laryngoscope 115:1–26, 2005
3.Sennaroglu L. Cochlear implantation in inner ear malformations – a review article. Cochlear Implants Int 11:4–41, 2010
4.Song MH, Bae MR, Kim HN, Lee WS, Yang WS, Choi JY. Value of intracochlear electrically evoked auditory brainstem response after cochlear implantation in patients with narrow internal auditory canal. Laryngoscope 120:1625–1631, 2010
5.Miyasaka M, Nosaka S, Morimoto N, Taiji H, Masaki H. CT and MR imaging for pediatric cochlear implantation: emphasis on the relationship between the cochlear nerve canal and the cochlear nerve. Pediatr Radiol 40:1509–1516, 2010
6.Archbold S, Lutman ME, Marshall DH. Categories of Auditory Performance. Ann Otol Rhinol Laryngol Suppl 166:312–314, 1995
7.Niparko JK. Cochlear implants: Principles & practices. Philadelphia: Wolters Kluwer Health/Lippincott Williams & Wilkins, 2009
Address for correspondence: Yasushi Naito MD, PhD, Department of Otolaryngology Kobe City Medical Center General Hospital, 650–0047 2–1-1 Minatojima Minamimachi Chuo-ku, Kobe City, Japan. naito@kcho.jp
Cholesteatoma and Ear Surgery – An Update, pp. 371–373
Edited by Haruo Takahashi
2013 © Kugler Publications, Amsterdam, The Netherlands