Abstracts

Rationale: Amplitude-integrated EEG (aEEG) is widely used in neonatal intensive care. The aEEG background is predictive of outcome in acute neonatal encephalopathy. Abnormalities of conventional EEG (CEEG) background are associated with increased risk of neonatal seizures. There are currently two aEEG background classification systems, which have not previously been directly compared: (i) a simple voltage criteria classification proposed by al Naqeeb [1] and (ii) a more complex pattern recognition classification described by Hellström-Westas [2]. Objectives: To determine the agreement between CEEG background classification and the simple and complex aEEG classification systems. To determine if aEEG interpreter experience affects agreement with CEEG. To establish the inter-rater agreement using the two aEEG background classification systems. Methods: The EEG background was classified for 144 neonatal CEEGs (traditional criteria for normal/mildly abnormal, moderately abnormal, or severely abnormal background). A single channel from each record (derived from C3-C4) was converted to aEEG. These traces were independently interpreted by four individual neonatologists and one paired team of aEEG experts according to the simple system described by al Naqeeb and the more complex system described by Hellström-Westas.Results: There was better inter-reader agreement using the simple than the complex classification system (multi-rater κ 0.66 and 0.44, respectively). For each individual reader, agreement with CEEG was better using the simple system than the complex system. However, there was similar fair-to-moderate agreement between both aEEG classification systems and the CEEG background classification (simple: κ 0.34-0.45; complex: κ 0.36-0.45). The agreement between CEEG and aEEG background classification did not vary significantly with aEEG interpreters’ experience.Conclusions: This is the first study to directly compare CEEG background assessment with the two most widely used aEEG background classification systems. Regardless of degree of expertise with aEEG, the readers consistently had higher agreement when using the simpler al Naqeeb system. This has implications for patient selection in future neonatal seizure studies. References: [1] al Naqeeb N, Edwards AD, Cowan FM, Azzopardi D. Assessment of neonatal encephalopathy by amplitude-integrated electroencephalography. Pediatrics. 1999;103(6):1263-71. [2] Hellström-Westas L, Rosen I, de Vries LS, Greisen G. Amplitude-integrated EEG classification and interpretation in preterm and term infants. Neoreviews. 2006;7(2):e72-e87.