Whether patients with a neocortical temporal lobe epilepsy need an additional amygdalo-hippocampectomy is a frequently asked question in clinical practice. We investigated if spikes and HFOs recorded during intraoperative ECoG (ioECoG) can guide the neurosurgeon in this decision. Methods: We selected patients with a temporal neocortical lesion who underwent an ioECoG-tailored resection without an amygdalo-hippocampectomy from the RESPect database. We visually marked spikes, ripples (80-250Hz) and fast ripples (FRs; 250-500Hz) on neocortical and mesiotemporal channels before and after the resection. We looked for differences in event rates and resection ratios between good (Engel 1A) and poor outcome groups and performed a logistic regression to identify outcome predictors. Results: Fourteen out of 24 included patients had a good outcome. The poor-outcome patients showed a younger age at epilepsy onset (p = 0.01) and higher rates of ripples on neocortical channels distant from the resection in pre- and post-ioECoG (ppre = 0.04, ppost = 0.048). Post-ioECoG FRs were found only in poor-outcome patients (N = 3). Factors predictive for seizure outcome were neocortical and mesiotemporal post-ioECoG ripples (ORneo = 3.68, pneo = 0.055; ORmesio = 0.13, pmesio = 0.03) and age at epilepsy onset (OR = 1.76, p = 0.03). Two poor-outcome patients underwent repeated surgery for removal of mesial structures, but their outcome did not improve. Conclusion: High rates of post-ioECoG ripples distant from the resection and the presence of residual FRs are indicators of an incomplete resection. HFOs in ioECoG may help to inform the neurosurgeon on the chance of success but do not specifically indicate that mesiotemporal structures should additionally be removed.

Introduction: For 1.3-1.9 million patients suffering from drug-resistant focal epilepsy in Europe alone, neurosurgery is often the only path to seizure freedom. The delineation between healthy and epileptogenic tissue relies on MRI and other diagnostics and can be further refined through intraoperative electrocorticography (ioECoG). Complete removal of tissue generating interictal electric discharges (IEDs, <80Hz) and high frequency oscillations (HFOs, 80-500Hz) in ioECoG correlates with a good surgical outcome. The precise link between IEDs/HFOs and MRI lesions, however, is unknown. We aim to identify this link and improve our understanding of how ioECoG and MRI findings should be weighed against each other to accurately delineate epileptogentic tissue. Methods: We retrospectively included patients from the HFO Trial (Van ’t Klooster, 2015) who underwent ioECoG-tailored surgery for a single lesion and achieved seizure freedom (Engel 1a). All lesions and resection cavities were segmented based on pre- and postoperative MRI scans. We automatically detected and visually verified IEDs and HFOs in 1-minute ioECoG epochs. Using intraoperative photographs, the MRI-coordinates of all ioECoG electrodes were determined on a rendering of the cortical surface. Results: 33 patients met our inclusion criteria. We will compare the number and rate of IEDs and HFOs per electrode with the volume of, and distance from the MRI lesion and subsequent resection cavity. We aim to find underlying functional relationships between those variables through regression analyses to better integrate MRI and ioECoG in our current model for tissue delineation. Conclusions: This study is ongoing, results are expected by mid-year 2021.