OP80 – 2596: Potential of spectral EEG analysis as academic performance predictor in young patients with hypoxic-ischaemic encephalopathy (HIE)

Objectives Currently there are no reliable assessment tools for academic performance prediction in HIE patients. We compared cortical electrical activity via spectral analysis of EEG of HIE subjects with that of the healthy control group. In previous studies we observed correlation between local maximum in the alpha spectrum and learning success in HIE adolescents. We expect that local maximum in the alpha spectrum is observed in the control group with learning success. Methods We studied two groups: 13 subjects (8 males) of mean age 21.7±0.9 years with mild (69.2%) to moderate HIE (30.8%), graded by Sarnat and Sarnat criteria, mean gestational age 36.3±3.4 weeks and body weight 2640±807 g. The group reached educational level of secondary school. Learning difficulties were reported in 6 (46.2%) subjects, 5 of them had problems with mathematics, one with additional ADHD problems. Young man with epilepsy needed global help due to psychomotor retardation. Second group was of 10 healthy subjects (5 males), mean age 23.2±1.1 years. EEG was performed after sleep deprivation with 30 electrodes, recorded at 256 Hz sampling rate. Data of 2 HIE subjects were not further analysed due to epileptiform changes. We decomposed EEG signals into brain rhythms. Gained high-dimensional data were further reduced with principal component analysis (PCA), which showed three significances, the rest we identified as background noise. Further spectral analysis was performed in Matlab with first PC further analysed because of its relation to the global alpha rhythm generator. Results The data from HIE individuals and control group with learning success showed local maximum of alpha rhythm at mean 7.0±1.0 Hz. Data of individuals within HIE group who reported learning difficulties showed absence of alpha's local maximum. Conclusions If observed with statistical significance in a bigger population, the method could help predicting academic success in HIE patients.