Abstracts

Rationale: Many, but not all patients with refractory epilepsy associated with hippocampal sclerosis that go for video-EEG (VEEG) evaluation develop secondarily generalized tonic-clonic seizures (SGTCS) during ictal pre-surgical VEEG recording. Secondarily generalized tonic-clonic seizures during VEEG are precipitated by antiepileptic drug withdraw. Perhaps because of this fact the possible role of SGTCS during video-EEG as predictor of surgical outcome has not been adequately investigated. Our study was designed to evaluate if the presence of SGTCS during pre-surgical VEEG could be used as a predictor of unfavorable surgical outcome in patients with refractory mesial temporal lobe epilepsy related to hippocampal sclerosis (MTLE-HS). Methods: We revised clinical, imaging, VEEG, surgical, and anatomopathologic records of 286 consecutive patients with MTLE-HS submitted to anterior and mesial temporal lobectomy at Centro de Cirurgia de Epilepsia de Ribeirão Preto (CIREP) - Brazil. Presurgical parameters analyzed included sex, ethnicity, age at surgery, age at epilepsy onset, the duration of epilepsy, history of initial precipitating insult, epilepsy duration until surgery, family history of seizures in the first-degree offspring, monthly complex partial seizure frequency impairing awareness in the year before surgery, distribution of interictal spikes, neuroimaging, and side of surgery. VEEG data were reviewed, especially regarding the presence of SGTCS during the period of the pre-surgical evaluation. The magnitude of association between the presurgical parameters and seizure outcome was measured by odds ratio (OR) and respective 95% confidence interval (CI). Crude and adjusted ORs were estimated by logistic regression. Results: In our series 79% of the patients became seizure free after anterior and mesial temporal lobectomy. We did not observe differences between patients seizure free (Engel class I) and those who remain with seizures among all the studied variables, except for SGTCS during VEEG. One hundred forty-eight patients (51.7%) presented SGTCS during VEEG. Of those, 42 patients (39.62%) remained with seizures after surgery. One hundred thirty eight (48.3%) did not present generalized seizures during VEEG. Of those, only 19 patients (15.9%) remained non-seizure free. The chance of being seizure free in the group of patients without generalized seizures during VEEG was significantly higher than those patients that presented generalized seizures during pre-surgical evaluation. The observed association remained significant after logistic regression adjustments (O.R.=2.01; CI 95%=1.26-3.36; p<0.005). Conclusions: Our results suggest that the presence of SGTCS during VEEG monitoring predicts unfavorable surgical outcome in MTLE-HS. Our observation further suggests that selected clinical parameters easily observed during VEEG monitoring might be highly predictive of surgical outcome in MTLE-HS. This work was supported by FAPESP and CNPq.