The aim of our study was to analyze the performance of bedside aEEG interpretation to detect seizures when interpreted by NICU clinical staff in real-time. 

Adequate detection and management of seizures is imperative, as undertreated seizures may cause neurological damage. The clinical detection of seizures is challenging in neonates. Seizure monitoring via amplitude-integrated EEG (aEEG) is standard of care in many NICUs; however, conventional EEG (cEEG) is the gold standard for seizure detection. 

Neonates requiring seizure monitoring received aEEG and cEEG in parallel. Clinical events and aEEG were interpreted at bedside. Subsequently, aEEG and cEEG were independently analyzed by experienced neonatology and neurology readers. Sensitivity and specificity of bedside aEEG as compared to expert aEEG interpretation and cEEG were evaluated.

Thirteen neonates were monitored for an average duration of 33 hours. Fourteen seizure-like events were detected by clinical observation alone, and 12 others by bedside aEEG analysis. One of the clinical, and none of the bedside aEEG events were confirmed as seizures on cEEG. Post-hoc expert aEEG interpretation revealed eight suspected seizures, all different from the ones detected by the bedside aEEG team, of which one was confirmed via cEEG. Eight seizures were recorded on cEEG. Expert aEEG interpretation had a sensitivity of 13% with 46% specificity for individual seizure detection, and a sensitivity of 50% with 46% specificity for detecting patients with seizures. 

Real-world bedside aEEG monitoring failed to detect all seizures evidenced via cEEG, while misclassifying other events as seizures. Even post-hoc expert aEEG interpretation provided limited sensitivity and specificity. Combining aEEG and cEEG is feasible. However, considering the poor sensitivity and specificity of bedside aEEG interpretation, combined monitoring may provide limited clinical benefit.