Utilization of continuous EEG (cEEG) has been limited in many settings due to logistical or economic constraints. It has been previously demonstrated that 1) patients with initial 30-minute EEG showing only generalized slowing are at low risk of seizures, 2) patients with epileptiform findings were most likely to have seizures within 16 hours of monitoring, and 3) the 2HELP2B score can predict seizure probability based on history and EEG findings. We developed a protocol that incorporates this research to strategically allocate cEEG to patients most likely to benefit from it. The purpose of this project was to analyze the effects of this protocol at our institution.

We collected data on encephalopathic or comatose inpatients that underwent EEG May-August 2019 and compared to May-August 2017. Patients with 2HELPS2B >0 on initial 30-minute EEG were monitored with cEEG for a targeted duration of 16 hours. If no seizures were detected, cEEG was discontinued. Findings of interest included utilization of cEEG, frequency of seizure diagnoses, discharge disability, and the relationship of 2HELPS2B to cEEG findings.

We included 43 and 56 patients from 2017 and 2019, respectively. Mean age was 55 and 57 years; 74% and 93% were in the ICU. cEEG was done in 14% and 50% of patients; seizures were detected in 0% and 17%. All patients with seizures had 2HELPS2B >0, and 5 of 6 had scores >1. Among cEEG patients, seizures were detected in 5 of 17 with mRS 5-6 and 0 of 12 with mRS 0-4.  

Our cEEG protocol resulted in greater utilization of cEEG. The increased frequency of seizure diagnoses suggests improved efficiency of patient selection. The majority of patients with seizures had 2HELPS2B >1, supporting its utility as a prediction tool. Seizures were more likely to be detected in patients with very poor outcome.