Abstracts

Rationale: Surgical treatment of refractory temporal lobe epilepsy (TLE) is highly effective, with 60-80% of patients achieving long term seizure freedom after cortico-amygdalohippocampectomy (CAH; Elsharkawy et. al., J Neurosurg 2009). Seizure onsets are usually within the mesial temporal lobe, a finding that has encouraged conservative surgical approaches, such as selective amygdalohippocampectomy and laser interstitial thermal therapy (LITT). There is concern that outcome is less good following LITT in TLE patients with nonlesional MRI (Tao et. al., J Neurol Neurosurg Psychiatry 2018) and failure following LITT may be due to involvement of structures outside the medial temporal lobe (Tandon et. al. Epilepsia 2018). Stereo-electroencephalographic (SEEG) studies demonstrate the frequent involvement of the temporal pole (TP) in TLE (Charabdes et. al., Brain 2005); the TP's role in seizure generation may therefore explain some of the selective procedures' failures. Here we present a small series of TLE patients with SEEG-proven temporal pole disease and highlight particular findings suggesting significant polar involvement in TLE. Methods: We compiled the ictal semiology, imaging findings, and interictal scalp EEG and SEEG findings of four TLE patients with SEEG-proven temporal pole involvement (Table). Electrodes in all patients sampled both the hippocampus (Hip) and the temporal pole (TP), in addition to other structures dictated by the pre-implant hypothesis. Results: Patient symptoms and objective semiology was variable; only one patient reported a consistent aura, which was experiential with polymodal sensory components. However, all patients had semiology compatible with 'temporal lobe epilepsy' at some phase of their habitual seizures. All patients had interictal spikes on scalp EEG maximal in the mid- or anterior temporal regions. Three patients were nonlesional on MRI though all were PET-positive; one patient had imaging findings of dual pathology in the same temporal lobe. On SEEG, all patients exhibited spiking at one or more TP contacts, independent of the hippocampus. The majority of seizures involved the TP simultaneously with hippocampal structures and some were clearly of isolated TP onset, though several seizures involved the TP only secondarily. All TP- and TP-Hip simultaneous ictal onsets were of the low-voltage fast activity (LVFA) type. Conclusions: We present a small series of TLE patients who underwent SEEG evaluation with electrodes sampling both the temporal pole and the ipsilateral hippocampus. Indeed, SEEG was undertaken in these patients to explore the contribution of the TP (in addition to other structures of the TLE network) owing to uncertainty in ascribing the entire epileptogenic zone to the mesial structures. 'Independent' involvement of the TP was diagnosed by interictal spiking that did not necessarily involve the hippocampus; all patients subsequently exhibited seizures that were clearly of TP onset, or involved the TP and ipsilateral Hip simultaneously. All such seizure onsets were of LVFA type; the repetitive spiking pattern that is specific to hippocampal-onset seizures (Perucca et. al., Brain 2014) was not seen. Interestingly, though all patients had a semiology that was compatible with temporal lobe epilepsy, none reported an abdominal aura. A larger study is under way that compares TLE patients with and without significant TP involvement on SEEG, to identify the noninvasive features that predict TP involvement in patients with TLE. Funding: No funding