Abstracts

Rationale: Conventional video-EEGbased on the international 10-20 system, modified for neonates, is the standard as recommended by the American Clinical Neurophysiology Society (ACNS). Several neonatal montages with reduced electrode arrays are suggested by the ACNS guidelines as acceptable alternatives, particularly given the high risk of skin breakdown in this population. Whether the reduced array of electrodes retains the high sensitivity for seizure detection remains to be examined. This attribute is particularly important in neonates with hypoxic-ischemic encephalopathy (HIE) as the presence of seizures predicts poor prognosis and impacts clinical care. The objective of this study was to compare the efficacy of electrographic seizure detection in a reduced array of EEG recording electrodes in neonates undergoing the standard hypothermia protocol at our institution for HIE. Methods: We retrospectively reviewed video-EEG recordings of in neonates with HIE who underwent therapeutic hypothermia at Texas Children's Hospital in 2013. We identified patients with seizures based on the EEG report. All neonates underwent conventional multichannel video-EEG recordings using the 10-20 system. The extended array consisted of 16 scalp electrodes: Fp1, Fp2, F3, F4, F7, F8, C3, C4, P3, P4, T7, T8, P7, P8, O1, and O2. Midline electrodes (Fz, Cz, Pz) were not included to provide access for head ultrasounds. We then simulated the reduced array using the NicVue software to an 8 electrodes montage as in the ACNS's guideline except minus the midline electrodes: Fp1, Fp2, C3, C4, T7, T8, O1, and O2. An electrographic seizure was defined as sudden and evolving repetitive stereotyped waveforms with a definite start, middle, and end, lasting at least 10 seconds at least one EEG channel. Seizure number was counted as the number of seizure events recorded in the monitoring periods. Results: Thirty-five neonates with HIE underwent therapeutic hypothermia in 2013. Of these neonates, 8 had seizures detected by the extended EEG array. Comparison of the extended vs. the reduced array in 4 patients, revealed a total of 329 seizure episodes using the extended array. In contrast, the reduced array recorded only 231 seizure episodes. In 2 of 3 neonates, the Fp1 and 2 leads failed to detect 57% of seizures arising from the F3, F4, or F8 leads. The reduced array was able to detect all seizures arising from other brain regions as compared with the extended array. Conclusions: Our findings suggest that the reduced array of electrodes using Fp leads without inclusion of F3 and F4 lacks the sensitivity in detecting frontal seizures. Therefore, in instances where midline electrodes are unavailable, the F3 and F4 leads appear superior to the Fp or frontopolar leads in detecting seizures arising from the frontal regions. Studies are ongoing to verify our observations in a larger group of study subjects and to explore the use of the F3 and F4 frontal leads in the reduced array of electrodes.