Simulation Software Enables Accurate Placement of Electrode Arrays in Cochlear Implants

Noran Engineering, Westminster, California

Cochlear implants have been used successfully to provide auditory information for bilaterally profoundly deaf patients. This is achieved by electrically stimulating auditory nerve fibers via an electrode array, which is surgically implanted into the scala tympani of the cochlea (inner ear). It is important that the electrode array does not cause damage to the fine intra-cochlear structures during insertion, as this can result in the loss of spiral ganglion cells, which are necessary for the implant to evoke auditory percepts.

Accurate and consistent placement of electrode arrays is important in sound reproduction, since their position affects the level of electrical impulses transmitted to the hearing nerves. The Nucleus standard straight array has been used for many years, and when it is inserted into the cochlea (which is shaped like a snail's shell), the electrode lies along the outer wall of the scala tympani.

Figure 1. a **Pre-Curved Contour** electrode array.

To improve speech perception in patients, the Contour array was developed to lie closer to the nerve cells located centrally near the modiolus. The Contour array is produced precurved, but held straight for surgical insertion with an interior platinum (Pt) stylet, inserted in the silicone carrier of the array and removed after completion of the electrode insertion. The array then recovers its pre-curved shape within the cochlea. On removal of the stylet, it is desirable for the final resting position of the Contour array to be close to the modiolus.

Experiments involving insertion of the Contour array in a number of temporal bones and in patient trials have shown significant variation in their final placement after the stylet had been removed, which may be related to insertion technique. Since experiments with temporal bones and limited patient trials are hard to replicate, a theoretical approach has been adopted to assess the placement of the Contour array and to study the effects of different insertion procedures on its final placement. A numerical model similar to that used previously to evaluate the insertion trajectories of the Nucleus straight array has been used to study the curling behavior of the Contour array, and to evaluate the effects of different insertion procedures on its final resting position in the cochlea.

NEiNASTRAN, a finite element (FE) analysis software, was used in the present study to predict the variation in trajectories of the Contour array as it curls on withdrawal of the stylet. The FE model was used to predict the final resting position of the Contour array with different insertion depths. The effects of varying the standard insertion procedure by inserting or retracting the array after the stylet has been completely removed were also evaluated.

Figure 2. **Predicted Curling Behavior** of the Contour array for Case 3 (left) and Case 4,

The FE analysis is performed in two stages. The first stage involves modeling the insertion of the array into the cochlea. The second stage, which is the subject of the present study, involves modeling the curling behavior of the Contour array as the stylet is removed. Four cases were modeled (see Table). Cases 1 and 2 were set up to compare the curling behavior of the electrode array inserted to two different insertion depths. The stylet was withdrawn in two steps. The third and the fourth cases were set up to examine the effects of varying the insertion procedure on the final position of the electrode array after the stylet had been removed. In Case 3, the steps, which were used in Case 1, were followed with an extra step added to push the electrode array a further 1 mm into the cochlea. Conversely, in Case 4, the steps used in Case 2 were followed with an extra step added to retract the electrode array by 1 mm. Since the external ends of the electrode array in Cases 1 and 2 were 2 mm apart, the external ends of the electrode array in Cases 3 and 4 are expected to converge to the same position.

A **Description of Steps** assumed in the modeled cases.

Figure 2 shows the final position of the electrode array in Case 3 (left) with the electrodes in the back half of the array pushed away from the modiolus as a result of inserting the external end of the array a further 1 mm. There appears to be a small amount of forward sliding with the front tip moving slightly deeper into the cochlea but slightly further from the modiolus.

At right in Figure 2 is the final position of the electrode array in Case 4 with the external end of the array retracted by 1 mm. In contrast to the previous case (Case 3), the electrodes in the back half of the array are pulled closer to the modiolus, but the front end of the array moves away from the modiolus and lies against the outer wall of the cochlea.

Using the FE analysis enabled accurate positioning of the electrode arrays, and provided vital information on the curling behavior of the arrays.

This work was performed using finite element analysis software from Noran Engineering. For more information, visit http://info.ims.ca/5658-122 .

ORIGINAL URL - /Briefs/May06/5658_122.html