Abstracts

Rationale: Dravet Syndrome (DS) is a pediatric epileptic encephalopathy caused by haploinsufficiency of the SCN1A gene, which codes for the sodium channel Nav1.1 and is associated with significant premature mortality. The incidence of sudden unexpected death in epilepsy (SUDEP) among children with DS may be higher than among children with other forms of epilepsy. Recent work in a mouse model of DS implicates abberant peri-ictal parasympathetic activity as a critical component of SUDEP pathophysiology. We sought to examine if interictal autonomic function is altered in patients with DS by analyzing prolonged ECG recordings. We hypothesize that, compared to non-DS epilepsy controls, patients with DS will demonstrate altered markers of vagal activity on analysis of heart-rate variability (HRV). Methods: We performed a retrospective case-control study examining continuous single-channel ECG recordings derived from patients who had undergone video-EEG recording from February 1, 2013 to March 1, 2017. We identified 20 DS patients and 54 epilepsy controls matched for sex and age (within ±10%).  For each tracing, automated QRS complex detection was performed, followed by computation of the following time-domain and frequency-domain HRV measures: mean interbeat interval (IBI); grand heart rate; SDNN (standard deviation of IBIs); sdHR (standard deviation of HRs); RMSSD (root mean sum-squared difference of IBI across sequential beats); pNN50 (percentage of successive intervals differing by at least 50 ms); SDNN index (mean of standard deviations across all 1 minute intervals); frequency (in Hz) of the very-low frequency (VLF), low-frequency (LF) and high-frequency (HF) peaks; power of the VLF, LF, and HF peaks; LF/HF ratio; and SD1 and SD2, the standard deviations of the Poincare plot along its 2 axes. Matlab and the HRVAS toolkit (https://sourceforge.net/projects/hrvas/) were used to compute these. The results were compared with a two-tailed independent samples t-test. Results: Of the measures listed, only the frequency domain measures of LF and HF power, and LF/HF ratio differed significantly between the 2 groups, and this was seen only in the sleep state.  In the sleep state, DS patients had reduced LF power (0.33 vs 0.45 normalized units, p=0.01), increased HF power (0.66 vs 0.55 normalized units) and a reduced LF/HF ratio (0.61 vs 1.01, p=0.04) compared to controls. Conclusions: In this study, frequency-domain measures of HRV suggest that patients with DS have increased parasymapathetic activity relative to sympathethic activity interictally (at least during sleep) compared to epilepsy patients without DS. Further study is needed to determine this difference in interictal vagal tone is specific to the genetic mutation or is due to other disease-related factors such as epilepsy severity or medications. Funding: No funding was received which supported this abstract.