Abstract
Ilyas A, Vilella L, Restrepo CE, Johnson J, Pati S, Lacuey N, Lhatoo S, Thompson SA, Tandon N. Epilepsia. 2024;65(3):641–650. Objective: Stereo-electroencephalography (SEEG) is the preferred method for intracranial localization of the seizure-onset zone (SOZ) in drug-resistant focal epilepsy. Occasionally SEEG evaluation fails to confirm the pre-implantation hypothesis. This leads to a decision tree regarding whether the addition of SEEG electrodes (two-step SEEG [2sSEEG]) or placement of subdural electrodes (SDEs) after SEEG (SEEG2SDE) would help. There is a dearth of literature encompassing this scenario, and here we aimed to characterize outcomes following unplanned two-step intracranial electroencephalogram (iEEG). Methods: All 225 adult SEEG cases over 8 years at our institution were reviewed to extract patient data and outcomes following a 2-step evaluation. Three raters independently quantified the benefits of additional intracranial electrodes. The relationship between the 2-step iEEG benefit and clinical outcome was then analyzed. Results: Fourteen patients underwent 2sSEEG and 9 underwent SEEG2SDE. In the former cohort, the second SEEG procedure was performed for these reasons-precise localization of the SOZ (36%); defining margins of eloquent cortex (21%); and broadening coverage in the setting of nonlocalizable seizure onsets (43% of cases). Sixty-four percent of 2sSEEG cases were consistently deemed beneficial (Light's κ = 0.80). 2sSEEG performed for the first 2 indications was much more beneficial than when onsets were not localizable (100% vs 17%, P = .02). In the SEEG2SDE cohort, SDEs identified the SOZ and enabled delineation of margins relative to the eloquent cortex in all cases. Significance: The 2-step iEEG is useful if the initial evaluation is broadly concordant with the original electroclinical hypothesis, where it can clarify onset zones or delineate safe surgical margins; however, it provides minimal benefit when the implantation hypothesis is erroneous, and we recommend that 2sSEEG not be generally utilized in such cases. SDE implantation after SEEG minimizes the need for SDEs and is helpful in delineating surgical boundaries relative to ictal-onset zones and eloquent cortex.
Commentary
The journey culminating in epilepsy surgery is a long struggle against therapeutic inertia. After several noninvasive tests, the ultimate step of intracranial electroencephalogram (iEEG) evaluation probes the grit of up to a third of drug-resistant epilepsy patients ultimately undergoing epilepsy surgery. 1 Planning iEEG is the epitome of personalized, precision, interdisciplinary medicine. It requires hours of painstaking multimodal data analysis, especially when stereo-electroencephalography (SEEG) has become the preferred and predominant mode of iEEG evaluation. 2 Despite the patient's tenacity and the epilepsy caregiving team's intellectual prowess, 20% to 35% of patients undergoing SEEG do not undergo resective epilepsy surgery because the information remains insufficient to guide a resection.3,4 Having come this far, and before denying the chance of seizure freedom, the novel information gleaned from SEEG can provide insights, which, if subsequently explored, may eventually lead to a resection. In the paper, Ilyas et al 5 share their experience of such scenarios where they followed up SEEG with additional iEEG evaluation to salvage a surgical solution. 5
The author's contribution to the literature is critical because there has been a tremendous increase in iEEG monitoring, which is not followed by a resection in US epilepsy centers. 6 `Besides the previously discussed reasons for this trend, 7 a critical contributing factor could be the SEEG technology's inherent limitation in spatial sampling. Although it samples 20% more gray matter volume compared to the subdural grid (SDG), a single SEEG contact only captures EEG activity within a 5 mm distance (volume of 30 mm3). Thus, a typical evaluation with 150 contacts using 10 to 15 electrodes only samples 4.5 cm3, ie, 0.5% of total human brain cortical volume. 8 Therefore, SEEG's interpretation is perennially haunted by the “missing electrode” paradigm. Therefore, recording typical ictal onsets using strategic sampling of cortical space that allows for the recognition of pattern similarity and consistent latency to define the putative epileptogenic zone is crucial for SEEG's success. 2 Not uncommonly, the electrode's restricted sampling space and viewing angle of the EEG activity can make the ictal-onset patterns appear less than ideal and rarely undetectable despite the seizure onset zone (SOZ) being not more than a few centimeters away far from the sampled cortex. 8
Overcoming such limitation requires experience in planning and interpreting SEEG, which may be hard to come by when the median number of iEEG evaluations at level 4 comprehensive epilepsy centers in the US is substantially lower than 1 per month. 6 Conversely, large volume centers performing several SEEGs per month accrue knowledge and soft skills faster through iterative learning.3,5 Leveraging such experience, Ilyas et al 5 report that 10% of their 225 SEEG implantations over 8 years required an additional iEEG evaluation. Among these 23 patients, 14 (6%) received additional depth electrodes for a supplemental SEEG during the same admission. The rest were readmitted a median of a month after the initial SEEG for a follow-up SDG evaluation. In this latter subgroup, the initial SEEG identified the SOZ, and the subsequent left hemispheric SDG was performed solely to define the resection's safe margin, which it did successfully. Supplemental SEEG is currently more relevant to most centers practicing iEEG, and hence, a detailed review is warranted.
Among the 14 patients undergoing supplemental SEEG, 50% were lesional cases. None had a symptomatic complication. A supplemental SEEG could be considered in 3 broad scenarios:
Refining SOZ localization: Initial SEEG's ictal-onset pattern was suboptimal but consistent with the pre-implant hypothesis, suggesting that SOZ was nearby. It led to supplemental SEEG in 5 patients, deemed “beneficial” as it improved SOZ localization. All 5 underwent resection, with outcomes of Engel I in one patient, Engel II in one, and Engel III in the rest. Delineating resection margins: Initial SEEG localized the SOZ but insufficiently sampled the cortical space to guide resection margins, particularly near the eloquent cortex. It led to supplemental SEEG in 3 patients, deemed “beneficial” as it allowed resection, with all achieving Engel I outcomes at the last follow-up. Incorrect pre-implantation hypothesis: When the pre-implant hypothesis proved incorrect and inconsistent with initial SEEG findings, suggesting the SOZ was outside the sampled regions. Six patients underwent supplemental SEEG for this indication. It was “beneficial” for only 1, who had resection and achieved an Engel I outcome. In the rest, SOZ was not localized despite a longer supplemental SEEG duration. Among them, 1 did not have surgery, 2 had vagus nerve stimulator placement, and 2 had resection without improved seizure control during a brief follow-up.
This study offers 2 key takeaways, essentially the sides of the same coin: scenarios when supplemental SEEG can lead to surgical intervention and when it's unlikely to be helpful. The fulcrum of these 2 outcomes is the robustness of the pre-implantation hypothesis because the information gained from SEEG is only as good as the hypothesis guiding the cortical sampling. In scenarios a and b above, even the initial sampling can be considered partially “beneficial” because it, more or less, confirmed the SOZ. The supplemental SEEG was required only to refine the SOZ localization and resection margin. In contrast, if the initial SEEG evaluation is based on a weak pre-implantation hypothesis or overly relies on a single network hypothesis, it fails to sample less likely but plausible networks contributing to the SOZ. Consequently, as scenario c suggests, adding further electrodes is less likely to lead to a successful resection. This can occur due to missed semiological signs, overlooked subtle magnetic resonance imaging findings, lack of information about the potential networks or focal irritative zones from tests such as SPECT and MEG, respectively, poor synthesis of pre-invasive information, or complex epilepsy not amenable to resectable localization. Although SEEG's success is highly operator-dependent, prognostic model like 5-SENSE highlights some of the electroclinical, neuroimaging, and neuropsychological features that makes an SEEG less likely to identify a focal (sublobar) SOZ.
9
This finding of Ilyas et al 5 supports the empirical knowledge that the die of SEEG's success is cast during the pre-implant hypothesis generation, which guides the cortical sampling. Trainees should especially heed this fundamental fact because the analysis of the SEEG recording is intellectually enticing and stimulating. It can potentially pull their focus and energy away from learning the fundamentals of scalp electroclinical correlations and synthesis of robust, hierarchical, and testable pre-implantation hypotheses. There is no substitute for the latter, especially once the Rubicon of SEEG implantation is crossed.
In another single-center study, supplemental SEEG performed for defining resection margin led to surgery in 5 out of 6 patients but only in 4 out of 8 patients when used for investigating an alternate hypothesis. Two patients in each subgroup achieved Engel I outcome. These outcomes are only slightly better than Ilyas et al, 5 potentially due to the former group performing more than twice the number of SEEG evaluations during a similar study period. 10 Despite small sample sizes, these studies suggest that we can consider supplemental SEEG when it can refine the SOZ localization or resection margin. Its benefit in investigating a revised or new SOZ hypothesis after initial SEEG fails to confirm the pre-implantation hypothesis is less promising. Nonetheless, supplemental SEEG's application in the latter scenario deserves further evaluation because it is a safe practice, and even if it leads to resection in a fraction of patients, it can be positively transformative for those individuals.
