Abstracts

Rationale: Mesial temporal lobe epilepsy (MTLE) is the most common focal epilepsy in adults and frequently do not respond well to antiepileptic drugs. Patients with refractory MTLE may present progressive atrophy of limbic structures and cognitive deficits. Neuropsychological assessment (NPA) allows the measurement of intensity of cognitive impairment and magnetic resonance imaging (MRI) can assess the presence of structural damage. We proposed to investigate the cognitive performance of patients undergoing surgery for MTLE and associate it with lateralization, seizure frequency and volume of the surgical lacunae (VSL). Methods: We evaluated 49 patients (26 operated on the left side; 23 on the right), using the Engel classification; 21 were in Engel IA (seizure-free) and 28 in other Engel classifications (IB-III). They were submitted to pre/postoperative NPA, including the Wechsler Memory Scale-Revised, Rey auditory Verbal Learning Test (RAVLT) and Wechsler Adult Intelligence Scale-Revised. We used a 3D-T1-weighted MRI acquisition for volumetry. The manual volumetry of VSL was performed using software DISPLAY (David McDonald, www.bic.mni.mcgill.ca/software). We used statistical tests: t-Student test or Wilcoxon test for comparing the performance of NPA between groups pre/postoperatively, unpaired t-test to compare the performance NPA between patients groups operated on the right and left side and to access the differences in cognitive performance between groups Engel IA and Others; Pearson correlation between NPA and VSL. We considered p<0.05. Results: Statistical difference between groups pre/postoperative subtests: intelligence coefficient- IQ, general, verbal and visual memory, delayed recall, indicating better performance in preoperative period (table 1). There were statistical differences between patients operated on the right and on the left side: general (t=3.3, p=0.002) and verbal memory (t=3.92, p=0.001), delayed recall (t=2.73, p=0.009), RAVLT-A7 (t=4.95, p=0.001) and RAVLT-recognition (t=2.72, p=0.009), showing the worst performance for left-sided surgery. Among the patients, 43% was allocated the Engel IA scale, 8% Engel IB, 21% Engel IC, 10% Engel ID, 4% Engel IIA, 6% Engel IIB and IID and 2% Engel IIIA. There were significant differences between groups Engel IA and Others for performance tests: IQ (t=2.33, p=0.025), general (t=2.18, p=0.035) and visual memory (t=2.35, p=0.024). These results indicate better performance of Engel IA. In other subtests, although not statistically significant, there was a better performance of patients classified in Engel IA. We found association between VSL right with IQ (r=0.55, p=0.006) and visual memory (r=0.45, p=0.03). Conclusions: The data suggest that surgical resection accentuate the cognitive deficits in patients with MTLE, being more pronounced in the left-sided resection and in those with persistent seizures postoperatively. Paradoxically, the greater the VSL the lesser cognitive impairment, probably due to the relationship between more extensive surgical resection and better seizure control.