LAUSR

Learning Objectives: Sudden unexpected death (SUDEP) is a significant cause of mortality in epilepsy. Lack of seizure control and history of status epilepticus (SE) are known risk factors. Furthermore, cardiac alterations have been observed, which may contribute to SUDEP. However, temporal evolution of the cardiac changes is unknown. Here we sought to investigate whether cardiac alterations develop over time in children with epilepsy who may be at risk for SUDEP. Methods: Children admitted to the pediatric intensive care unit for seizures or SE between 4/2014 and 7/2017 were prospectively identified. Children were included if they had at least 1 EKG study in the electronic medical record database and did not have any pre-existing cardiac conditions. Demographic data including the timing and indication of EKG studies were obtained. The primary outcomes were the presence of altered electrophysiology (EP) or arrhythmias. Altered EP was defined as QRS axis deviation, altered PR, QRS and QTc intervals, or ST segment and T wave changes. Arrhythmias were classified as atrial, ventricular, junctional rhythm, or heart block. We performed Student t or MannWhitney test for continuous variables; Fisher exact or Pearson χ2 for categorical variables; and logistic regression analysis to identify factors associated with EKG alterations. Results: 244 children met the study criteria with 719 EKGs. 88 children had no history of epilepsy (control, 216 EKGs), 156 children had epilepsy (503 EKGs). Compared with controls, the epilepsy group was more likely to have altered EP (1.4 [1.02–1.94], OR [95% CI], p < 0.05) and had more arrhythmias (PVC: 0.47% vs. 1.64%, junctional rhythm: 0% vs. 1.02%, heart block: 0% vs. 0.41%, control vs. epilepsy, p < 0.05). Within the epilepsy group, abnormal EKG studies occurred at an older age as compared with normal EKGs (66.6 ± 4.0 vs. 97.4 ± 5.0 months, normal vs. abnormal EKG, p < 0.01). In the control group no differences in age were observed (56.1 ± 5.8 vs. 54.2 ± 6.4 months, normal vs. abnormal EKG). Accordingly, logistic regression revealed that age was independently associated with abnormal EKG only in the epilepsy group (p < 0.01). Conclusions: Our findings suggest that altered cardiac EP and arrhythmias may develop over time in this vulnerable epileptic cohort. Therefore, cardiac surveillance in a select group of children with epilepsy may be warranted.