Gray Matter Sampling Differences Between Subdural Electrodes and Stereoelectroencephalography Electrodes

Abstract

Objective: Stereoelectroencephalography (SEEG) has seen a recent increase in popularity in North America; however, concerns regarding the spatial sampling capabilities of SEEG remain. We aimed to quantify and compare the spatial sampling of subdural electrode (SDE) and SEEG implants. Methods: Patients with drug-resistant epilepsy who underwent invasive monitoring were included in this retrospective case-control study. Ten SEEG cases were compared with ten matched SDE cases based on clinical presentation and pre-implantation hypothesis. To quantify gray matter sampling, MR and CT images were coregistered and a 2.5mm radius sphere was superimposed over the center of each electrode contact. The estimated recording volume of gray matter was defined as the cortical voxels within these spherical models. Paired t-tests were performed to compare volumes and locations of SDE and SEEG recording. A Ripley's K-function analysis was performed to quantify differences in spatial distributions. Results: The average recording volume of gray matter by each individual contact was similar between the two modalities. SEEG implants sampled an average of 20% more total gray matter, consisted of an average of 17% more electrode contacts, and had 77% more of their contacts covering gray matter within sulci. Insular coverage was only achieved with SEEG. SEEG implants generally consist of discrete areas of dense local coverage scattered across the brain; while SDE implants cover relatively contiguous areas with lower density recording. Significance: Average recording volumes per electrode contact are similar for SEEG and SDE, but SEEG may allow for greater overall volumes of recording as more electrodes can be routinely implanted. The primary difference lies in the location and distribution of gray matter than can be sampled. The selection between SEEG and SDE implantation depends on sampling needs of the invasive implant.

Keywords: depth electrodes; epileptogenic zone; intracranial electrodes; intracranial monitoring; stereoelectroencephalography; subdural grid.

Figure 1

The recording volume of gray matter covered by implants, projected to an anatomical pial surface (A,C) or an inflated pial surface (B,D) of subjects in Pair 1 (A,B from Subject S1; C,D from Subject G1). On the inflated pial surface images, the lighter areas represent the gray matter over the convexity (crest of gyrus); while the darker areas represent the gray matter within sulci.

Figure 2

Violin plot of volume of gray matter recorded by SEEG and SDE implants in gyri, sulci, and 7 cortical regions of interest (ROIs) for the entire cohort. Statistically significant differences (*) between pair-matched SEEG and SDE cases were found in gyri (paired t-test, p < 0.0001), sulci (paired t-test, p < 0.0001), insula cortex (Wilcoxon test, p = 0.0156), and mesial temporal cortex (Wilcoxon test, p = 0.0273). Increased sampling over the lateral temporal lobe with SDE was not statistically significant.

Figure 3

Ripley's k-function for SEEG and SDE over distance d. The Y-axis represents the K-function values on a logarithmic scale, while the X-axis represents the distance from an arbitrary contact in millimeters. The dashed lines represent the 95% confidence intervals of the k-function values. SEEG implants had denser coverage than SDE over distances <10 mm and sparser coverage than SDE over distances >10 mm.