Saturday, December 6, 2008�Poster Session 1�1:00 p.m.-6:00 p.m.�Clinical Neurophysiology

Abstract

Abstract Nicholas S. Abend*,†, Dennis Dlugos*,†, S. Herman†, A. Topjian‡,§, M. Donnelly*, R. Ichord*,†, M. Helfaer‡,§, V. Nadkarni‡,§ and R. Clancy*,†*Neurology, The Children's Hospital of Philadelphia, Philadelphia, PA; †Neurology, The University of Pennsylvania School of Medicine, Philadelphia, PA; ‡Critical Care Medicine, The Children's Hospital of Philadelphia, Philadelphia, PA and §Pediatrics, The University of Pennsylvania School of Medicine, Philadelphia, PA Rationale: The incidence and characteristics of acute seizures in children undergoing therapeutic hypothermia following pulseless cardiac arrest (CA) are unknown. Since a high proportion of seizures may be subclinical, diagnosis requires continuous electroencephalography (EEG). We hypothesized that (i) EEG seizures are common during therapeutic hypothermia following CA and (ii) the background patterns would evolve during hypothermia and re-warming, and (iii) the background abnormalities would be associated with survival and short-term outcome. Methods: As part of a feasibility study of therapeutic hypothermia in children, we performed a prospective, pilot, observational study of 10 consecutive children undergoing continuous EEG monitoring during epochs of hypothermia (24 hrs), re-warming (12–24 hours), and an additional 24 hrs. EEG seizures were identified and the background during each epoch was described in a standardized manner. A standardized neurological exam was performed on day 5–7 to provide a cerebral injury score (CIS), rated from normal to profoundly abnormal. Results: Ten patients were studied from March to December 2007 with a mean age 19.9 ± 17.5 mo (range 2.2–51.2 mo). EEG monitoring began within 4 hours of hypothermia initiation in all. Over 720 hours of EEG were reviewed. Electrographic seizures occurred in 5/10 (50%) patients and began during the second 12 hours of hypothermia in 3 and during re-warming in 2. Most seizures were subclinical and electrographically generalized. Eight patients with initially mild/moderate EEG background abnormalities (discontinuity, slowing, attenuation) improved during warming; 2 patients with initially severe background abnormalities (burst suppression) had EEG worsening (longer duration suppression and status epilepticus). Background abnormalities did not predict seizures. Of the two patients with severe background abnormalities, one survivor had a severely abnormal CIS and one died. Of the eight patients with initial mild/moderate background abnormalities, 2 survivors did not have a CIS, 3 had a mild/moderate CIS, and 2 had a severe CIS (1 died). Thus, 1 of 2 (50%) with severe background abnormalities died while 1 of 8 (13%) with mild/moderate background abnormalities died. Also, death or severe CIS occurred in 2 of 2 (100%) with severely abnormal EEG background but only 2 of 6 (33%) with mild/moderate background abnormalities. Conclusions: Seizures occurred in half of those undergoing therapeutic hypothermia after CA and were mostly subclinical and generalized. The EEG background may improve or worsen during hypothermia and re-warming, suggesting that the earliest EEG results may not be useful for prognostication. However, the trend suggests more severe abnormalities are associated with worse survival and short term outcome. Further study correlating continuous EEG background patterns with short- and long-term outcomes in children treated with therapeutic hypothermia is warranted.