Background: There are many challenges in the treatment of seizures in neonates, because of difficulties in diagnosis and lack of universal consensus on what to treat. There are concerns regarding the safety and adverse effects of the antiepileptic drugs (AED) used, including effects on brain development and neurological outcomes. Hence, it is imperative to rigorously evaluate the basis of AED usage in neonates.

Methods: We analysed the data regarding AED use, electroencephalography (EEG) and clinical findings recorded during a diagnostic accuracy study comparing 24 hours of conventional video-EEG (VEEG) (not used for clinical decision making in real time) with amplitude-integrated EEG (aEEG) (used for clinical decision making in real time) in 35 term or near-term neonates with suspected or at-risk of seizures. Correlation between seizure burden and number of AEDs was assessed using Pearson correlation.

Results: Twelve out of 35 infants received AEDs during the study based on seizures detected on aEEG (nine infants) or clinical events alone (three infants). Out of nine infants with aEEG seizures (74 seizure episodes), seven infants had seizures on VEEG (86 electrographic-only seizures and 83 electro-clinical seizures) while two infants did not have seizures on VEEG. There was no correlation between seizure burden on VEEG with number of anticonvulsants used during the study period (R²=0.0111).

Conclusion: Neonates are mostly treated with AEDs based on clinical and/or aEEG findings. Both of these modalities have a risk of under-treatment or over-treatment. The use of AEDs correlates poorly with the true seizure burden.