Abstracts

Rationale: Although surgical resection is the treatment of choice for patients with drug-resistant mesial temporal lobe epilepsy (MTLE), a small fraction of patients are not candidates for this intervention. Hippocampal electrical stimulation (HS) may benefit these patients, but the evidence is inconclusive. In a double blind randomized controlled trial (RCT), we aimed to determine whether (HS) is safe and more effective than simply implanting an electrode in the hippocampus without electrical stimulation (HI). The trial was stopped due to poor recruitment. We present results of the data collected. Methods: A five-centre, parallel-group, double blind RCT allocated patients with MTLE to HI or HS (ratio 1:1 initially, then 1:2). Unilateral or bilateral HS was delivered using a 4-contact electrode implanted along the hippocampal longitudinal axis, and 135 Hz continuous cathodal stimulation of all electrodes involved in seizure generation. Outcomes were assessed during a seven-month period post-randomization. The primary outcome was change in seizure frequency between baseline and six months of follow-up. Secondary outcomes included safety, cognitive function, quality of life (QOLIE-89), and proportion with >50% seizure reduction. Non-parametric tests compared differences between groups (p < .05). Results: The target sample of 57 patients could not be recruited. The RCT was stopped after only six patients were randomized. Four patients were randomized to HI (ages 35 to 46 years, 2 left, 2 right), and two to HS (age 30 years, both bilateral). HI patients (median monthly baseline seizure frequency = 10) had a median seizure increase of 60%. HS patients (median monthly baseline seizure frequency = 12) had a mean seizure decrease of 45%. For convulsive (GTCS) and dyscognitive (CPS) seizures the baseline median monthly seizure frequencies were 1 in HI and 2 in HS patients, and the median change was 31.3% and 50% increase, respectively. None of the between-group differences was statistically significant. Only one patient (HS group) achieved a >50% in seizure reduction. The overall QOLIE-89 score at 7 months worsened by 13 points with HI and improved 3 points with HS (not significant). QOLIE-89 attention/concentration scores decreased by 4 points with HI and increased by 20 points with HS (borderline significant, p < 0.06). Subjective memory scores using QOLIE-89 memory scales decreased by 34 points with HI and increased by 10 points with HS (not significant). Scores of cognitive scales assessing recall (RAVLT, RCFT) were generally lower in the HS group compared to the HI group (not significant). Conclusions: Despite collective efforts and multicentre participation, the feasibility of adequately powered RCTs of HS for patients with MTLE remains enormously challenging. The HS paradigm used in this RCT in a few patients resulted in: 1) non-significant trend towards improvement with HS in the frequency of all seizures (but not of GTC or CPS), and in subjective memory function; 2) a borderline significant improvement in attention/concentration; 3) a non-significant worsening in recall function.