Welcome to the 9th EAPS Congress Programme Scheduling

The congress will officially run on Barcelona Time (GMT+2)
To convert the congress times to your local time Click Here

Displaying One Session

Session Type
Pre-Recorded Oral Session
Date
10/06/2022
Session Time
08:00 AM - 11:59 PM
Room
Pre-Recorded Oral

EXPLORING THE CHALLENGES OF AMPLITUDE INTEGRATED EEG FOR SEIZURE DETECTION IN PEDIATRIC PATIENTS

Presenter
  • Laura Macdarby (Ireland)
Date
10/06/2022
Session Time
08:00 AM - 11:59 PM
Session Type
Pre-Recorded Oral Session
Presentation Type
Abstract Submission
Lecture Time
08:00 AM - 08:10 AM
Duration
10 Minutes

Abstract

Background and Aims

Amplitude integrated EEG (aEEG) is a mainstay of care in neonatal intensive care units (NICU). Despite increasing use outside the NICU, there are knowledge gaps in relation to the normative characteristics of aEEG, and its accuracy in identifying seizures in older children. Our objective is to evaluate the sensitivity and specificity of neonatal aEEG seizure identification criteria for seizure-recognition in older children.

Methods

120 pediatric EEG recordings were assembled, comprising 30 studies with seizures and 90 without. Seizure-containing studies were annotated in detail to describe number of seizures, their duration, distribution and spread. Two-channel aEEG (C3-P3, C4-P4) recordings were generated, blinded and independently reviewed without reference to the raw EEG trace. Seizures were reported according to the Hellstrom-Westas criteria for detection of neonatal seizures. Diagnostic accuracy of aEEG for seizure recognition was calculated. Logistic regression was used to identify factors associated with correct seizure identification on aEEG.

Results

Median patient age was 6.1 years. Abnormal recordings featured 123 discrete seizures in total. Status epilepticus (SE) was evident by EEG in 13 cases. Using neonatal seizure recognition criteria, aEEG had a sensitivity of 43% and specificity of 75%. aEEG identified SE in only 38% of the 13 reported cases, although seizures were correctly identified in 84% of this subpopulation.Seizures were more likely to be missed by aEEG when < 1 minute.

Conclusions

aEEG neonatal seizure detection criteria have a poor sensitivity for detecting aEEG seizures in older children mainly due to short duration of seizures in this population. Time compression eliminates detection of short seizures.

Hide