How long should continuous EEG monitoring be performed in critically ill patients? Temporal dynamics of electrographic seizure detection in a prospective cohort

Abstract

Background: Continuous electroencephalographic monitoring is increasingly used in intensive care units to detect electrographic seizures and nonconvulsive status epilepticus. However, the optimal duration of monitoring required for reliable seizure detection remains uncertain. Defining the temporal dynamics of seizure detection is essential for optimizing monitoring protocols and improving resource utilization in neurocritical care. Objective: To determine the temporal distribution of electrographic seizure detection during continuous electroencephalographic monitoring in critically ill patients. Methods: Prospective cohort study including 173 critically ill patients undergoing continuous electroencephalographic monitoring for clinical indications such as unexplained alteration of consciousness, suspected seizures, abnormal movements, or evaluation of encephalopathy. Continuous recordings were obtained using the international 10–20 electrode system and interpreted by experienced neurophysiologists. In patients with seizures, the time interval between the start of monitoring and the first electrographic seizure was recorded. Results: Electrographic seizures occurred in 35 patients (20.2%). The mean time to seizure detection was 13.3 hours and the median time was 8 hours. Seizures were detected as early as 2 hours and as late as 60 hours after monitoring initiation. Approximately half of seizures were detected within the first 8 hours of monitoring and most occurred within the first 12 hours. However, delayed seizures occurring after prolonged monitoring were also observed. Conclusion: Most electrographic seizures in critically ill patients are detected within the first 12 hours of continuous electroencephalographic monitoring. Nevertheless, delayed seizure detection occurring up to 60 hours suggests that prolonged monitoring may be necessary in selected high-risk patients. These findings have direct implications for optimizing monitoring duration and resource allocation in intensive care settings.