Abstracts

Rationale: Case reports of sudden unexpected death in Epilepsy (SUDEP) in Epilepsy Monitoring Units (EMUs), and more recently a comprehensive evaluation of cardiorespiratory arrests in EMUs (MORTEMUS study, 1), have highlighted the need for enhanced continuous supervision within this setting. Patients with frequent convulsive seizures are at highest risk for SUDEP. In addition, ictal arrhythmias, in particular bradycardia and asystole, are commonly encountered in the EMU. Therefore, it is critical that EMU staff are able to assess the (electrocardiogram) ECG continuously, which may be challenging particularly during motor seizures. Automatic detection systems are often lacking (2), but even if present require robust detection of the R wave on ECG. Lead I (Fig 1A) is currently the standard mode of ECG acquisition across many epilepsy monitoring units (EMUs). However, studies comparing the effectiveness of alternative ECG acquisition leads are lacking. We applied additional ECG electrodes to compare various leads of ECG acquisition for the purposes of assessing heart rhythm during seizures. The aim of the current study is to compare ECGs acquired using lead I (Fig 1A) and lead II (Fig 1B) by assessing whether R waves can be reliably detected by the human eye during noise-free (i.e. baseline) and noise-constrained (i.e. during onset and offset of generalised seizures) epochs. We hypothesize that the ECG signal acquired via lead II will allow for improved R detection rates and less obscuration by muscle during these periods when compared with lead I. References 1. Ryvlin, P., et al, The Lancet Neurology, 12(10), pp.966-977. 2. Rubboli, G. et al Epilepsy & Behavior, 44, pp.179-184. Methods: Two raters visually examined (by counting R waves) ECG traces from n = 17 GTCS. Traces were 40 second epochs of: random baseline; clinical seizure onset (including 20 seconds of onset clonic/tonic phase); and clinical seizure offset, as demarcated by a neurophysiologist, from Lead I and Lead II ECG (yielding 102 traces in total). Raters were blind as to Lead and epoch type. Intraclass correlation coefficients (ICC) were calculated to test for inter-rater reliability. A one-way analysis of variance (ANOVA) was carried out to assess whether significantly more R waves were detected with lead I compared with lead II ECG across the 3 epoch types. Results: ANOVA results revealed a significantly greater number of R waves were detected with Lead II ECG during seizure onset epochs (F=10.751, p=.002) and seizure offset epochs (F=4.007, p=.049) (see Fig 2). For responses to Lead I epochs, an ICC of 0.934 was observed, and for Lead II responses, an ICC of 0.918. These results indicate high inter-rater agreement. Conclusions: The Lead II ECG trace provides a greater resolution than lead I in revealing the R wave during periods of noise, particularly during ictal periods marred by muscle artefact. This therefore should facilitate visual monitoring of the ECG by EMU staff, improving safety. Improved R detection provided by Lead II also has the potential to advance our knowledge of heart rhythm dynamics in ictal periods, and may shed further light on the autonomic nervous system in epilepsy and SUDEP. Funding: This work was funded by NIH ?"National Institute of Neurological Disorders and Stroke (U01-NS090407-01 The Center for SUDEP Research) and undertaken at UCLH/UCL who receives a proportion of funding from the Department of Health's NIHR Biomedical Research Centres funding scheme.