Abstracts

Rationale: Systematic evaluations of polygraphic FIET analysis revealed variations in spatiotemporal characteristics at variance with conventional FIET modeling by one peak at one instant of time representing one single dominant generator. This exploratory study was designed to examine time domain of FIET in further detail with polygraphic channel overlay (PGCO) display. When baseline is controlled, PGCO can reveal subtle time-domain features prior to and following base-peak-trough-wave (BPTW) segment as well as complexity of FIET peaks (Fig). BPTW waveform template (Epilepsia 2005: 46 [Suppl 8]: 138) employed initial segment BP as reference. PGCO envelope resembles BPTW template but displays true phase reversals. Methods: FIET samples consisted of 72 single spikes and sharp waves (SSW) from 121 mixed samples of SSW and wicket spikes (Abstract 3.117, 2011 AES Annual Meeting, www.aesnet.org). FIET were reformatted in 23 common average derivations with high and low frequency filters set at 70 Hz and 0.16 Hz. PGCO were prepared manually with monitor screen images of 2-s FIET derivations with display gain controlled. PGCO were inspected with special attention to pre-base shift, and compared against blinded polygraphic FIET waveform ranking and BPTW scores. Results: SSW could be differentiated against background-baseline with exception of 5 ranked at lower end. FIET ranking did not differ significantly between PGCO and blinded polygraphic review. Pre-base shift was seen in 38 of 72 FIET, but its onset was not always definable. Conclusions: SSW detection from PGCO display was more conservative, and may improve reliability of clinical EEG analysis. PGCO can reveal baseline deviation prior to BPTW segment, when display condition is favorable. Target signal magnitude and background noise-signal, of either cerebral or extracerebral origin, affected confidence of SSW detection. Display settings may have to be tailored. This method may improve clinical differentiation of SSW from FIET not predictive of epilepsy.