Abstracts

Rationale: Stress is the most commonly reported seizure precipitant in multiple surveys, and increased self-reported stress has been associated with seizure occurrence in prospective diary studies. We performed a prospective, randomized, controlled trial (RCT) of stress reduction with smartphone monitoring of stress and mood variables in people with medication resistant focal epilepsy.Methods: Three centers recruited subjects with at least 2 focal onset seizures per month. Data collection included daily measures of precipitants such as stress, mood, sleep, and medication adherence. After a 2-month baseline, eligible subjects were randomized to a 3 month period of active treatment with 2-3 times daily progressive muscle relaxation versus sham treatment consisting of a focused exercise of extremity movements and writing. Subjects and evaluators were blinded to treatment allocation. Subjects recorded seizures and answered questions about stress and mood twice daily in a study smartphone. Primary efficacy outcome was change in average seizure frequency per month compared with baseline seizure frequency. Secondary analyses examined average seizure frequency per month (excluding month 5) compared to baseline seizure frequency and examined the primary efficacy outcome in a subgroup of patients who had a reduction of at least 1 seizure per 28 days. Wilcoxon rank sum and signed rank tests plus analysis of covariance (ANCOVA) after adjusting baseline seizure counts were used for data analysis. Intention to treat and per protocol analyses were performed.Results: 67 subjects completed the 5 month study. 94% of days all the diary entries were completed. Median seizure frequency was reduced by 29% in the active treatment arm (p=.006 compared to baseline) and by 25% in the sham arm (p=.064 compared to baseline). There was no difference between the active and sham arms when compared directly. A comparison of month 1 and 2 of baseline compared to just months 3 and 4 (excluding month 5) showed 33% seizure reduction in the active treatment arm (p=.009) compared to 18% reduction in the sham arm (p=.171). The difference between the active and sham treatments was significant in the subgroup of patients who had reduction of at least 1 seizure per 28 days. The intention to treat and per protocol analyses showed identical results.Conclusions: An RCT of stress reduction as treatment for medication resistant seizures is feasible, and subjects are highly compliant with treatment and daily diary entries. There was no significant difference between seizure reduction in the active and sham treatments in the overall group. Both groups demonstrated a significant decrease in seizures during the treatment, which may reflect a positive behavioral effect from sham treatment combined with daily diary maintenance. Secondary analyses showed that response to treatment was significant during the first two months of treatment, and identified a subgroup with a significant response to active stress reduction. The findings from this pilot study will be used to design future behavioral interventions for epilepsy treatment.