Electrocorticographic evidence in mesial temporal epilepsy.
Electrocorticographic evidence of different physiopathologic types in mesial temporal epilepsy. Surgical implications.
Torres Díaz C, Pastor Gómez J, Mar@n P, García Navarrete E, Navas García, Gil Simoes R, M; García de Sola,R.
National Reference Center for Epilepsy Surgery
Comunicación al Congreso Mundial de Neurocirugía Funcional (WSSFN): Tokio del 27 al 31 de Mayo del 2013.
Hospital Universitario. HOSPITAL LA PRINCESA MADRID
The concept of epileptogenic neural networks was characterized over the past two decades, through imaging, neurophysiological and histological studies, indicating that seizures of patients with partial epilepsy do not originate from a single brain area, the «focus ictal» but in several cortical and subcortical structures highly interconnected, forming a «epileptogenic neural network.» In the case of mesial temporal lobe epilepsy, there have been studies suggesting that the nodes involved in the generation of seizures may be spread over several extrahippocampal and extratemporal limbic structures.
In previous studies, the analysis of the intraoperative electrocorticographic recordings using linear correlation and phase synchronization, allowed the quantifycation of the lateral temporal lobe intracortical interactions. These data showed the existence of a synchronous activity emerging from specific areas of the neocortex, which clearly differ from the rest of the lateral temporal cortex. Resection of the nodes with high local synchronization in the ECoG correlated with a good postoperative outcome, while the sparing of these nodes was associated with unfavorable outcome despite complete resections of mesial structures.
We hypothesize that intraoperative electrocorticography might reflect the neurophysiological changes that occur in the epileptic network when the surgical resection is being performed, allowing an individualised surgical strategy for each patient. The aim of this study is to confirm this hypothesis and to assess the efficacy of an optimized surgical approach based on it.
We prospectively studied a series of 30 patients that underwent temporal lobe resections for refractory mesial temporal epilepsy in our Epilepsy Surgery Unit between 2010 and 2012. Patients were evaluated according to our protocol including VEEG assessment with foramen ovale electrodes They were diagnosed with mesial temporal epilepsy and estimated candidates for resective surgery (Spencer’s technique).
Patients were intraoperatively evaluated by electrocorticography (ECoG) with 20 subdural electrode grids over the lateral temporal cortex, and an 8 electrode mesial strip, under low doses of sevoflurane (0.5%) and remifentanil (0.1 mg/kg/min). Fifteen minutes after the start of the ECoG, 0,1 mg/kg of iv etomidate were administered. The mesial strip was maintained until the anterior medial temporal resection was performed, and a second dosage of etomidate was given 10 minutes after its finalization. A tailored resection of the mesial structures was carried out in some of the patients according to the ECoG results.
Fifteen patients were male, and 16 procedures were performed on the right side. Mean age at time of surgery was 42 years (rage 22-61). Mean epilepsy duration before surgery was 26 Semiology was suggestive of mesial temporal seizures in all of the patients. 21 patients had MRI suggestive of HS, VEEG showed irritative mesial spikes, and acute ictal abnormalities were found in all of the patients coming from the mesial temporal structures.
In six out of the 30 patients, the irritative interictal mesial spikes disappeared following the anterior medial temporal resection, and mesial structures were spared. All of the patients are in Engel’s class I at the last follow-up (mean 23 months after the surgery).
In the remaining patients, the irritative mesial activity increased or was unchanged, and the amygdale and the hippocampus were excised. Eighty-three percent of these patients are in Engel’s Class I, at the last follow-up (mean 16 months after the resection).
Ictal Pattern: FOE
This is the VEEG with foramen ovale electrodes corresponding to one of the patients undergoing an exclusive anterolateral resection. In the first image, the patient is performing his normal activities, prior to the onset of a seizure. In the second picture, the patient loses consciousness and begins to present with oral and superior limbs automatisms. The electroencephalographic origin of the seizures is located in the left foramen ovale electrodes recording. In the third image, automatisms in superior limbs continue and the seizure spreads. The last image shows a postictal left ear scratching.
Neuropsychological assessment of the patients that underwent standard lateral and mesial resections, did not show any statistically significant difference between the preoperative and postoperative IQ (mean 100.5 / 97.2 in right side procedures, and 97.2, 99, 9 in left side interventions, respectively). The verbal and visual memory is postoperatively increased in patients operated on the right side, while it is reduced in patients undergoing left side resection.
In patients undergoing lateral resections, IQ, and verbal and visual memory improve after surgery on both sides.
This table shows the patients’ clinical characteristics and preoperative evluations in the subgroup of patients who underwent limited resections. Semiology does not show any specific features other than those characteristically shown in MTLS. In 3 cases, MRI findings were suggestive of mesial temporal sclerosis, and consistent with the VEEG lateralization. Two patients had normal MRI, and in one case was there was a bilateral ETM associated with a tumor.
20-30% of patients undergoing temporal lobectomy and amygdalohippocampectomy fail to achieve seizure freedom, despite an adequate identification of the seizure onset in mesial region, and hippocampal extensive resections.
On the other hand, limited lateral resections produce good long-term results in 38 and 50% of patients which is an only slightly lower percentage than that obtained in series where amygdala and hippocampus are resected. Moreover, the resection of the hippocampus is associated with some degree of cognitive sequelae in 3-42% of patients after right-sided procedures and in 16-80% of patients undergoing left side resections. Some authors have proposed a classification of the ELTM in different subtypes, but this is still the subject of debate, and adequate diagnostic tests have been established to classify patients into the differet subtypes.
In previous studies, it has been suggested that the resection of certain hyperconnected areas of the lateral temporal cortex, identified by a postoperative analysis of the intraoperative ECOG or intracranial electrodes recordings is essential for the resolution of seizures in patients with MTLE.
Electrocorticography has been routinely used for guiding the resection of the lateral cortex in ELTM. Although its value is controversial, it can provide continuous information of the generated intracortical interactions in TLE. However, most of these studies have focused on the neocoticales abnormalities without specifically addressing hippocampal ECoG findings.
In some of the studies in which the structures were monitored, using ECoG mesial to determine the extent of the hippocampal resection, a correlation was found between the resection of hippocampal interictal spikes and the postoperative outcome, irrespective of the size of resection area, which is consistent with our results.
- Intraoperative EcoG might be helpful to
- Our results strongly suggest the existence of physiopathologic differences within the mesial temporal epilepsy syndrome
- There is a need for more studies to
- The identification of these different subtypes is fundamental for an individualized surgical approach and for the consideration of new therapeutic strategies differentiate mesial temporal epilepsy subtypes confirm these results