Frameless Stereotaxy in Stereoelectroencephalography Using Intraoperative Computed Tomography
Loading...
Date
Publisher
MDPI
Abstract
Around one-third of all epilepsy patients are pharmacoresistant, often presenting with temporal lobe epilepsy (TLE), and 20% show extratemporal seizure onsets [1,2,3,4,5]. For carefully selected patients with focal pharmacoresistant epilepsy, resective surgery is a well-established and promising treatment option, showing excellent outcomes especially in TLE patients [4,6], but posing a greater challenge in extratemporal lobe epilepsy [7,8,9]. However, one major prerequisite for resective surgery is the clear and precise approximation of the epileptogenic zone (EZ). Especially in cases in which non-invasive multimodal diagnostics fail to accurately define the EZ due to inconclusive radiological, clinical, semiological, and neuropsychological findings, further invasive diagnostics utilizing intracranial electroencephalography (iEEG) might be useful [10,11].
The current common practice for performing iEEG is the so-called stereo EEG (SEEG), which is based on the implantation of depth electrodes. Following anatomo-electro-clinical hypotheses, SEEG allows for the identification of seizure origin and propagation and a definition of the EZ. SEEG depth electrode implantation can generally be performed using frame-based stereotaxy or frameless navigation, with or without robotic assistance [12,13,14,15]. Accuracy is crucial in determining the risk of intracranial complications and the likelihood of successful EEG recordings and localization of the epileptogenic zone [15], independent of the applied technique. Clinical accuracy is a multifactorial parameter. In stereotactic and navigational applications, overall or clinical accuracy, which is most important to the surgeon, can be roughly divided into application accuracy and intraoperative accuracy, and the surgeon tremendously depends on accuracy in every single step of the procedure. Application accuracy itself can be divided into three domains as follows: imaging, technical, and registration accuracy [16,17,18]. Imaging accuracy mainly concerns the precise and optimized multimodal planning of trajectories incorporating target and risk structures, and the modality also ensures geometrical accuracy. The technical accuracy of the navigational systems depends, e.g., on the technique used (frame-based vs. frameless); the intrinsic accuracy of the systems itself; or the tracking technique used, which is nowadays considered to be less than 3 mm in frameless systems and even less in frame-based stereotactic systems, aiming for submillimeter accuracy [16,19]. Especially in frameless setups, registration accuracy mainly influences the application accuracy, offering user-dependent methods for landmark- and surface-based techniques with mean target registration errors of up to 5 mm [20,21,22] or automated intraoperative imaging-based techniques with higher registration accuracy and with target registration errors (TREs) of less than 1 mm [18,23,24,25,26]. While the domains mentioned above are related to the presurgical phase, intraoperative events hamper overall accuracy during the course of surgery. Navigational accuracy is known to decrease over time, which is related to, e.g., the attachment of drapes, incision, trepanation/drilling, and the duration of the surgery itself. These impact the spatial relationship between the patient’s head and the reference unit (reference array vs. frame), affecting the non-linear deformations of the brain, which might be a minor issue compared to craniotomy cases.
Several methods for measuring accuracy have been introduced recently, and these are inconsistently used across different studies, dealing with the accuracy of depth electrodes in epilepsy patients. The most prominent ones are the Euclidean distance at the entry and target points, which describe the three-dimensional deviation between the planned and detected trajectory endpoints. However, especially in the case of the target points, the radial and depth errors can be considered instead, as the depth error, incorporated in the Euclidean distance, is surgeon-dependent, thereby not reflecting the accuracy of the used approach [15]. In addition to the radial error, the angular deviation between the planned and detected trajectory can also be considered. Another study suggested reporting the directional errors instead to assess the systematic error in the stereotactic system [27].
Despite a potential loss of implantation accuracy when using frameless implantation techniques instead of frame-based methods, this study aims to evaluate the effect of utilizing automated intraoperative imaging-based techniques for patient registration, which are known to offer higher accuracy, on the implantation accuracy of SEEG depth electrodes.
Metadata
Philipps-Universität Marburg
License
Except where otherwised noted, this item's license is described as Attribution 4.0 International