gms | German Medical Science

Objective: To perform deep brain stimulation, an accurate localization of anatomical target structures as well as the precise placement of the electrodes within the target structures is required. To map preoperative data to the patient’s anatomy, stereotactic frame-based techniques are applied. As optimal frame positioning in relation to the scanner’s coordinate system might be difficult to achieve in some cases, localization accuracy might also be affected. In this study localization accuracy was evaluated with respect to deviations of the frame from optimal positioning corresponding to rotation, flexion and lateral bending of the head.

Methods: To analyze the accuracy of frame localization related to different frame positions, a stereotactic frame (Zamorano-Duchovny) including a CT localizer box was attached to a head phantom. Deviations from optimal frame positioning were revealed by adjusting the frame according to rotation, flexion or lateral bending or a combination of those (0°, 9° or 18° each). The localizer box was positioned in the iso-center of the AIRO^® iCT scanner (Brainlab, Munich, Germany). For localization a full-dose and super-low-dose protocol was used covering the whole localizer box. Frame localization was performed using the Stereotaxy Element (Brainlab, Munich, Germany). The localization accuracy was calculated as deviation of slice-by-slice detected points along each rod.

Results: In case of the full-dose protocol, reference localization accuracy was 0.17 ± 0.09 mm (rotation – flexion – lateral bending: 0-0-0). Successful localization was revealed in eight other cases with localization accuracies of 0.18 ± 0.09 mm (9-0-0), 0.18 ± 0.10 mm (18-0-0), 0.18 ± 0.09 mm (0-9-0), 0.21 ± 0.16 mm (0-0-9), 0.19 ± 0.09 mm (9-9-0), 0.18 ± 0.09 mm (9-0-9), 0.18 ± 0.09 mm (18-0-9) and 0.17 ± 0.09 mm (18-9-0), not significantly differing from the reference. In case of the super-low-dose protocol, reference localization accuracy was 0.14 ± 0.07 mm (0-0-0). Successful localization could be performed in eight cases with localization accuracies of 0.15 ± 0.08 mm (9-0-0), 0.16 ± 0.08 mm (18-0-0), 0.17 ± 0.08 mm (0-9-0), 0.17 ± 0.09 mm (0-0-9), 0.15 ± 0.08 mm (9-9-0), 0.14 ± 0.07 mm (9-0-9), 0.15 ± 0.07 mm (18-0-9) and 0.16 ± 0.09 mm (18-9-0), not significantly differing from the reference. However, the remaining scans could not be used for localization.

Conclusion: For frame-based stereotactic procedures, moderate deviations in all spatial planes are tolerated without loss of localization accuracy.