Abstracts

Rationale: Lennox-Gastaut syndrome (LGS), a severe epilepsy phenotype, is characterised by drug-refractory seizures, cognitive impairment and scalp EEG findings such as tonic seizures and generalised paroxysmal fast activity (PFA). Seizures are usually refractory to anti-epileptic medications, with novel treatment approaches required. The ESTEL (Electrical Stimulation of Thalamus in Epilepsy of the Lennox-Gastaut phenotype; NHMRC project #1108881) study is currently evaluating deep brain stimulation (DBS) as a treatment in patients with LGS. To date, 16 adult patientshave undergone bilateral implantation to the centromedian nucleus of thalamus (CM). Formal analysis of the blinded data awaits complete enrolment; however, seizure suppression (following 3-6 months of stimulation) has been seen in some patients. Seizure diaries, considered the 'gold standard' in determining response to therapy, have been shown to be inaccurate. To provide a parallel estimation of seizure frequency, we evaluated ESTEL patients using 24h ambulatory EEG. We aimed to assess whether burden of PFA correlated withelectrographic seizure counts and monthly seizure diaries. Methods: Baseline data (i.e., pre-DBS insertion) of five patients enrolled in ESTEL (ages 19-37; 5 females) were evaluated. Parents/caregivers were asked to complete a standardised seizure diary for least three months prior to DBSinsertion, to determine baseline seizure frequency. Only ‘countable’ seizures (e.g., tonic, generaliZed tonic-clonic, drop attacks) were tallied to determine an average seizure/day estimate per patient. Overnight ambulatory EEG (International 10-20 system) was recorded during the baseline period for an average of 23.51 hours (range 20.4-25.3). EEG data weremanually evaluated for (i) total number of electrographic seizures/recording; and (ii) total number and duration (secs)of PFA discharges between 1200-0200hours on each recording. For two patients, seizure diaries and ambulatory EEG were assessed at baseline andagain atpost-implantationweek 12, 24 and 36. Results: The mean (across patients) average number of seizures/day reported on baseline seizure diaries was 0.84. In comparison, mean electrographic seizure rate was 3.11/hour (i.e., 75/day) and mean PFA discharge rate was 1.90/minute (i.e., 114/hour). During two continuous hours of overnight EEG, PFA comprised on average 3.2% of the record. Excluding data from one patient whose seizures/hour rate was considered to be an outlier, we found a trend for average seizures/hour on EEG to correlate with total PFA duration(Pearson’s r=0.72, p=0.27, n=4; Figure 1) and PFA discharges/min (Pearson’s r=0.85, p=0.15, n=4). Conclusions: Seizure diaries are a vastly inaccurate estimate of seizure burden in LGS, underestimating true frequency by 50-100 fold. 24h ambulatory EEG allows a much more accurate measure of epileptic activity, confirming a very high burden of seizures (>3/hour) and generalized paroxysmal fast activity (1-2/minute). The high burden of measured discharges, and ability to objectively measure changes in rate, should allow a more accurate estimation of treatment effect. Further analysis of ESTEL data will confirm whether reductions in PFA predict reductions in seizures, and whether PFA burden can be used as a ‘biomarker’ of treatment effect in LGS. Funding: National Health and Medical Research Council project grant #1108881