Scala vestibuli cochlear implant supported by 3D modeling of the inner ear

Sudden sensorineural hearing loss (SSNHL) is an unexplained, rapid loss of hearing either all at once or over a few days and often affects only one ear [1]. During 2006 and 2007, the annual incidence of SSNHL was 27 per 100,000 patients in the USA and this incidence rate increases with increasing age, ranging from 11 per 100,000 in patients younger than 18 years to 77 per 100,000 in patients 65 years and older [2]. An SSNHL can be the result of variety of reasons including infections, head trauma, autoimmune diseases, exposure to certain drugs that treat cancer or severe infections, blood circulation problems and neurological disorders such as multiple sclerosis [3]. Labyrinthitis ossificans is often a side effect of some of the SSNHL causes and creates surgical challenges in the placement of cochlear implant (CI) electrode.

Labyrinthitis ossificans and how far it extends inside the cochlear lumen can be visualized by computed tomography (CT) images; however, it could be challenging for young and inexperienced surgeons and or radiologists to compile the entire series of CT scans in bringing a three-dimensional (3D) representation of the anatomical structure in mind. 3D segmentation of the inner ear has been applied clinically in cases with inner ear malformations to accurately identify anatomical structures available, thereby simplifying the CI electrode placement [4‐8]. The same technique could be applied in cases with different degrees of scala tympani (ST) ossification to identify how deep it has extended and this helps surgeons to decide the placement of the CI electrode either in ST or in scala vestibuli (SV).

This article reports a case of ST ossification as identified by the 3D segmentation of the inner ear. We found the 3D segmentation of the inner ear structures clinically useful in making the decision of placing the electrode in SV and as well in the surgical planning of the cochleostomy to access the SV. The review of the relevant literature supported SV electrode placement in cases with difficult if not impossible ST electrode placement due to cochlear ossification.

A 45-year-old Austrian female was referred to the ear, nose and throat (ENT) department of the Medical University of Graz, Austria for CI, after she had been diagnosed as deaf in the right ear2 years ago due to SSNHL. The left ear showed no indications of hearing loss. All investigations performed elsewhere at the time of SSNHL, including MRI, showed normal results.

The CT image analysis of the right temporal bone applying 3D segmentation of the inner ear structures showed ossification in the basal portion of the cochlea as shown in Fig. 2 (preoperative analysis). The extent of ossification was seen only in the ST and it was estimated to be around 9 mm from the RW entrance (Fig. 2a–d). The cochlear size as measured by the diameter of the basal turn in the oblique coronal plane was 10 mm as shown in Fig. 2a. In contrast to the ossification of ST, the SV was completely patent as seen in both 2D images and the 3D model (Fig. 2b, d and f). The 3D model of the inner ear (green structure) showing the ossified portion of the ST (red structure) was useful in planning the location of cochleostomy anterior superior to the RW to reach the SV. Following all the regular CI surgical steps to reach the RW, it was found to be ossified as confirmed by gently touching it with a pointed surgical tool. Cochleostomy drilling was then performed approximately 1.5 mm anterior and superior to the RW niche. After identifying the SV lumen, a dummy insertion electrode provided by the company MED-EL (Innsbruck, Austria) was inserted to confirm the extent of SV lumen for a length of 30 mm and was followed by the insertion of a functional electrode array of length 28 mm to its full prescribed length inside the SV. Postoperatively, the patient felt well, without any dizziness or vertigo and postoperative CT images confirmed correct placement of the electrode covering an AID of 540° inside the SV (Fig. 2g, h).

The externally worn audio processor was activated 4 weeks following the CI surgery and the pure tone audiogram showed the hearing threshold ≤ 40 dB across all frequencies tested as shown in Fig. 3.

A review of the relevant literature from PubMed search identified 13 articles dated from 1990 until 2018 reporting on the SV placement of CI electrode array due to ossification in ST ([9‐21]; Table 1). The etiology of hearing loss reported in these 13 articles ranged from meningitis, temporal bone fracture, otosclerosis, sudden SNHL, autoimmune disease, Noonan’s syndrome, Cogan syndrome, Meniere’s disease and inner ear malformations. The electrode array types implanted in all these cases mainly belonged to the older generation types that were intended to cover the basal turn (360° of angular depth) of the cochlea. The one case from Kiefer et al. reported on a full insertion of a STANDARD electrode from MED-EL that offers an insertion depth of 30 mm [11]. Although not given in Table 1, cases reported in all these studies showed an improved hearing performance after CI procedure with electrode in SV, compared to the preoperative hearing scores.

The SV electrode placement was first reported by Steenerson et al. in 1990 as an alternative solution, when the ST is occluded with ossification [9]. Since then there have been 13 reports up to 2018, that have demonstrated the SV electrode placement whenever ST was not available due to ossification [9‐21]. Kiefer et al. [11] and Lin et al. [18] were the only reports to show deep insertion of an older generation electrode in the SV. Other than these two reports, all the other reports given in Table 1, reported on the short electrode array placement in the SV that literally covered not more than 360° of AID. To the best of the authors’ knowledge, this is the first report showing the full insertion of the flexible version of a longer length electrode array (28 mm) covering an AID of 540° in SV due to ossification in the ST.

The 3D segmentation of the inner ear (green) and the ossified portion (red) that were overlaid on the 2D image slice gave us a clear picture of where to drill the cochleostomy to access the SV. Fig. 4 showcases the 3D segmented image of the right ear along with the ossified portion in the basal turn of the ST. The white circle around the black dot in Fig. 4 pointed by the black arrow is the location identified for cochleostomy drilling to access the SV. The left cochlea did not show any traces of ossification anywhere within the cochlea.

Although there were earlier studies that reported on the clinical application of 3D segmentation of the inner ear especially in the identification of inner ear malformation types [4‐8], to the best of our knowledge ours is the first report that applied 3D segmentation of the inner ear from the CT images, in accurately identifying the ossified portion of ST as shown in Figs. 2 and 4. The time required to perform the 3D segmentation is not more than 10–15 min and it was very helpful to us as operating surgeons in accurately accessing the SV by drilling the cochleostomy 1.5 mm anterior superior to the RW niche. The insertion of the dummy insertion electrode to a length of 30 mm inside the SV convinced us to choose a 28 mm long electrode and flexible array. Also, for the reason that this patient was profoundly deaf across all frequencies tested preoperatively, electrical stimulation across the entire frequency range was thought to be necessary and as supported by recent reports [22‐24].

The postoperative image and the 3D segmentation of the implanted electrode confirmed the full insertion of the 28 mm long electrode array in SV without deviating back into the ST at any location along the cochlea lumen. The postoperative pure tone audiometry showing the hearing thresholds ≤ 40 dB across all frequencies tested confirmed the electrical stimulation of the neuronal elements from the SV.

The electrode placement in the SV due to ST ossification described in this report is in accordance with the literature findings. We felt that the 3D segmentation of the inner ear was clinically useful in understanding the extent of ST ossification, identification of the cochleostomy location to access SV for the electrode placement and as well for teaching our resident doctors. One of the limitations associated with this report is that no detailed information on the postoperative speech performance of the case presented was reported; however, subjectively the patient was highly satisfied with her overall hearing with CI in the right ear and natural hearing in the left ear. The literature from the past has taught us that the required levels of electrical stimulation for subjects implanted via the SV were similar to those required for subjects who had the standard ST insertion of the electrode array [19, 21]. Also, the literature has demonstrated that SV insertion of CI electrode offers hearing function comparable to ST insertion and SV is often available when the ST is not [14, 19, 21]. Although our case is only few weeks with the audio processor activated, the hearing threshold of ≤ 40 dB as measured from the pure tone audiometry is a clue that the patient will benefit from the CI procedure.

This report demonstrates the usefulness of 3D segmentation of the inner ear in accurately identifying the ossified portion of the ST and the cochleostomy site to access SV. Insertion of the dummy insertion electrode to check the extent of SV availability was useful in deciding the length of the flexible electrode array. This report could add to the growing literature supporting SV electrode placement in cases with ossification in ST that cannot be drilled out completely.