Abstracts

Rationale: The Amsterdam Resuscitation Study (ARREST, PLoS ONE 7(8): e42749. doi:10.1371/journal.pone.0042749) and Oregon Sudden Unexpected Death Study (Circ Arrhythm Electrophysiol 2013;6:912-916) revealed that risk of death due to sudden cardiac arrest is 3-fold greater in patients with epilepsy than in the general population. However, the factors responsible for precipitating cardiac arrest and adequate methods for risk detection require further elucidation. We hypothesized that the attendant stresses associated with EMU admission would provide an opportune environment to disclose latent cardiac electrical instability. T-wave alternans (TWA), a beat-to-beat fluctuation in the ST-segment and T wave, which is highly correlated with sudden cardiac death in several nonepileptic populations, could prove valuable in uncovering ictal and interictal markers of elevated risk for cardiac arrhythmia.  Methods: Twenty patients with clinically indicated causes for EMU admission were enrolled.  On the first day of hospital admission, EKG recording began using extended continuous 14-day patch monitors, which reduce noise and movement artifact associated with standard cardiac telemetry with wired leads, especially during clinical seizures (upper left figure).  The signals were pre-processed for further noise and baseline wander removal.  TWA was assessed using FDA-cleared modified moving average analysis software (GE Healthcare, Milwaukee WI).  EKGs were continuously recorded throughout the EMU stay without interruption during clinical interventions and antiepileptic medication adjustments.  All patients also had continuous EEG (cEEG) monitoring using a digital monitoring system (XLTEK, Natus Medical Inc., Pleasanton, CA), including a single channel EKG (lead I) as part of the standard EEG recording protocol. Results: On admission day, maximum interictal TWA was elevated in 15 of 20 patients with chronic epilepsy to >60 µV, which exceeds the 47-µV cutpoint of abnormality associated with a >3-fold risk of sudden cardiac death in the general population. Of the three patients with lower maximum interictal TWA levels, one was found not to have epilepsy; the second was receiving a beta-adrenergic blocking drug, which reduces TWA in parallel with sudden cardiac death risk in general populations; and the third had received a vagus nerve stimulation device over 17 years ago. The 10 patients who experienced a seizure during their EMU stay exhibited higher maximum interictal TWA levels on admission day than the 10 patients without seizures (65+-2.5 µV vs 41.5+-4.2, p<0.001, means +- SEM) (upper right figure). The greatest increase in maximum ictal TWA was observed during generalized tonic-clonic seizures (GTCs) compared to partial seizures (PS) (51.3+-3.8 vs 31.4+-4.5 μV, p<0.02) (lower left figure). One patient experienced ventricular premature beats (VPBs) and a burst of couplets at the peak of the partial-seizure induced TWA surge (lower right figure). Concurrent EKG recordings with standard clinical leads during the seizure were affected by excessive artifact and did not detect these arrhythmias.  Conclusions: The EMU provides an important setting in which to uncover cardiac electrical instability associated with chronic, drug-resistant epilepsy especially on admission day, presumably due to behavioral stress associated with admission to the EMU and with the occurrence of acute seizures. In our cohort, generalized seizures were associated with the greatest TWA elevations compared to focal seizures. Use of wireless EKG patch monitors permitting analysis of microvolt levels of TWA, not only during interictal periods but also during acute seizures, offers a new approach to dynamic tracking of cardiac electrical instability. These findings also shed light on the cardiac response of patients with chronic epilepsy and the potential mechanisms underlying elevated risk for sudden cardiac death in this vulnerable population. Funding: Funding for this study was received from LivaNova PLC.