Abstracts

RATIONALE: The extent of the mesial temporal resection, especially of the hippocampus is of great importance in term of seizure control by tailored temporal lobectomies, and controversy exists about the ideal posterior limit of hippocampal resection. Whatever limit is chosen, it can be difficult to ascertain the accuracy of the hippocampal resection when compared with the pre-planned surgical program. The aim of this study was 1) to compare the post-operative MRI data with the preoperative resection planning, as far as hippocampectomy is concerned, 2) To correlate these data with surgical outcome.
METHODS: 51 patients have been included in this study. All of them underwent a tailored temporal lobectomy for drug-resistant temporal lobe epilepsy. For each patient, the post-operative MRI was performed 2 months after surgery. It included coronal, bi-hippocampal, and sagittal slices. Only the T1 sequences were analysed in this study. The extent of hippocampal resection was better evaluated in the frontal plane. The resection was quoted as complete if no hippocampal remnant could be seen ahead the frontal slice located just anterior to the aqueductus mesencephali level. These MRI data were then compared with the per-operative data, namely with what the surgeon described in the operative report. The epileptological outcome was evaluated with a follow-up ranging from 6 to 60 months. For each group (complete or -not complete- hippocampectomy) the number of seizure-free patients (Engel class IA) was determined.
RESULTS: 1) According to the radiological criteria described above, the hippocampal resection was quoted as insufficient in 7 cases (13.7 %). In these cases, the hippocampal remnant length ranged from 1 to 2 cm anterior to the above described level in the coronal plane.
2) With respect to the visual analysis of post-operative MRI data, coronal slices proved to be the most appropriate views to detect a possible posterior hippocampal remnant.
3) We did not find any statistically significant difference, in term of epileptological outcome, between complete or not complete hippocampectomy.
CONCLUSIONS: We show here a very simple way to check the extent of a presumed total hippocampectomy, as the posterior limit of hippocampal resection is not always easy to identify intraoperatively and to remove without danger. However, the question remains, in term of post-operative results, to ascertain the usefulness of a very complete hippocampal resection, as the present study shows that the rate of cure is comparable, whatever complete is the posterior hippocampal resection.