Cochlear implant insertion should be as least traumatic as possible in order to reduce trauma to the cochlear sensory structures. The force applied to the cochlea during array insertion should be controlled to limit insertion-related damage. The relationship between insertion force and histological traumatism remains to be demonstrated. Twelve freshly frozen cadaveric temporal bones were implanted with a long straight electrodes array through an anterior extended round window insertion using a motorized insertion tool with real-time measurement of the insertion force. Anatomical parameters, measured on a pre-implantation cone beam CT scan, position of the array and force metrics were correlated with post-implantation scanning electron microscopy images and histological damage assessment. An atraumatic insertion occurred in six cochleae, a translocation in five cochleae and a basilar membrane rupture in one cochlea. The translocation always occurred in the 150- to 180-degree region. In the case of traumatic insertion, different force profiles were observed with a more irregular curve arising from the presence of an early peak force (30 ± 18.2 mN). This corresponded approximately to the first point of contact of the array with the lateral wall of the cochlea. Atraumatic and traumatic insertions had significantly different force values at the same depth of insertion (p < 0.001, two-way ANOVA), and significantly different regression lines (y = 1.34x + 0.7 for atraumatic and y = 3.37x + 0.84 for traumatic insertion, p < 0.001, ANCOVA). In the present study, the insertion force was correlated with the intracochlear trauma. The 150- to 180-degree region represented the area at risk for scalar translocation for this straight electrodes array. Insertion force curves with different sets of values were identified for traumatic and atraumatic insertions; these values should be considered during motorized insertion of an implant so as to be able to modify the insertion parameters (e.g axis of insertion) and facilitate preservation of endocochlear structures.

Hypothesis: An inappropriate insertion axis leads to intracochlear trauma during cochlear implantation (CI). Background: Few studies assessed the relationship between the insertion axis and the electrode scalar location. Methods: Preimplantation cone-beam CT (CBCT) was performed on 12 human temporal bones. In five temporal bones, an optimal insertion axis was planned, due to the impossibility to attain the ST centerline from the posterior tympanotomy, because of facial canal position. In the seven other temporal bones, an inaccurate insertion axis was intentionally planned (optimal axis+15 degrees). Automated CI array insertion according to the planned axis was performed with a motorized insertion tool driven by a navigated robot-based arm. The cochlea and basilar membrane were segmented from the preimplantation CBCT and the array segmented from the postimplantation CBCT to construct a merged final three-dimensional (3D) model. Microscopical and 3D analysis were performed to determine the intracochlear trauma at the level of each electrode. Results: A good agreement was observed in determining electrode position between microscopic analysis and the 3D model (Cohen's kappa k = 0.67). The angle of approach to the ST centerline was associated with the number of electrodes inserted into the ST (r = −0.65, p = 0.02, [95% CI −0.90 to −0.11] Spearman's rank correlation). Conclusion: A 3D reconstruction model was effective in determining the array position in the cochlea scalae. Our data indicate that the angle of approach to the ST centerline is a critical factor in intracochlear trauma. Additional studies should be conducted to assess the importance of the insertion axis with other array designs.