Abstracts

Rationale: Temporal lobe epilepsy (TLE) surgery is the most frequent type of surgical treatment offered to patients with drug-resistant focal seizures, and the only to have demonstrated superiority over medical treatment in randomized controlled trials (RCTs). However, the proportion of patients enjoying long-term post-operative seizure freedom remains sub-optimal. This situation partly recflects the lack of full understanding of the reasons underlying surgical failures, and limited capacity to predict such failures. Our purpose was to retrospectively analyse 190 drug-resistant patients who had undergone temporal lobectomy and identify the variables possibly influencing surgical outcome. Attention was focused on whether temporal plus seizures were associated to worse outcome compared to temporal seizures. Methods: We collected 190 patients from the database of two French Epilepsy Surgery Centres (Grenoble and Lyon) according to the following inclusion criteria: (i) absence of any detectable lesion on MRI, with the exception of hippocampal sclerosis; (ii) Stereoelectroencephalography (SEEG) recordings showing that seizures involved at least mesial and/or lateral temporal lobe (TL) structures or if SEEG was not available, scalp Video-EEG showing that at least the temporal lobe was involved by seizures; (iii) surgery performed according to the results of presurgical evaluation, either invasive or noninvasive, taking into account anatomical constraints and (iv) at least 12 months of post-operative follow-up. Results: In univariate model, poor seizure outcome was correlated with significant past medical history (i.e neonatal anoxia, traumatic brain injury or central nervous system infection, p=0.006), occurrence of secondary generalised tonic-clonic seizures (p=0.005), and diagnosis of "temporal plus" seizures (p< 0.001). Multivariate analyses showed that the only two statiscally significant variables predictive of poor outcome were a past medical history (p=0.046) and the type of seizures, namely "temporal plus" seizures (p< 0.001). Conclusions: In this study we confirm that risk of surgery failure is greater in patients with "temporal plus" epilepsy than in those with ‘'pure'' temporal epilepsy. Identifying those of temporal lobe epilepsy cases that are in fact suffering from temporal plus seizures might help to select candidates to invasive recordings before surgery, and to plan surgical strategies, thus influencing post-operative prognosis.