Background The amplitude-integrated electroencephalogram (aEEG) is reliable for early prediction of outcome in asphyxiated neonates. Hypothermia influences the positive predictive value (PPV) of early aEEG delaying normalisation of background pattern.

Methods Forty-one neonates with hypoxic ischaemic encephalopathy (HIE) Sarnat stage II (n = 24 normothermia (NT), n = 17 hypothermia (HT)) were included into analysis.

Firstly aEEGs of the first 3 days (d) of life were analysed for both groups (background pattern, sleep-wake cycling and seizure activity; descriptive analysis and calculation of a combined aEEG-score).

Secondly aEEG parameters were correlated with postnatal clinical parameters, severity of neuroimaging abnormalities and neurodevelopmental outcome.

Finally, the PV of aEEGs was compared between the groups.

Results The rate of pathological aEEGs on d1 showed no significant difference between the groups (pathological aEEG: 64% HT, 71% for NT). On d3, the rate was significantly lower in the HT group than in the NT group (18% HT, 57% for NT, p = 0.045).

There was a significant difference in outcome: in the HT group 82% showed a normal outcome, compared to 43% in the NT group.

The PPV of a combinded aEEG-score was higher on d1 in the HT group than in the NT group and increased in both groups from d1-d3.

The lack of sleep-wake cycling in children treated with HT significantly correlated with severity of neuroimaging abnormalites (p = 0.05) and pathological outcome (p = 0.03). In contrast, there was no correlation between seizure activity and outcome.

Conclusion Our results reflect published data, underlining the importance of aEEG as early outcome predictor in neonatal HIE.