Abstracts

Rationale: Nonconvulsive status epilepticus (NCSE) is broadly defined as seizure activity without prominent motor manifestations lasting over 10 minutes. Previous literature indicates a high index of suspicion should be maintained for NCSE, as definitive diagnosis can only be made by electroencephalography (EEG). Delay in diagnosis can increase morbidity, mortality, length of hospital stay, and cost of care in intensive care units. At our institution, urgent EEG to exclude NCSE is ordered on encephalopathic patients in the medical ICU (MICU). Given the large volume of patients, this can result in EEG machine shortage, delaying evaluation of subsequent patients. Thus, characterizing clinical features of patients who are found to have NCSE would be an instrumental first step in developing criteria for triage of urgent EEG placement.

Methods: This is a retrospective study performed at the University of North Carolina of patients who received video EEG monitoring in the MICU. Variables collected include patient’s medical diagnoses and neuroimaging findings prior to EEG, descriptors of seizure-like activity, seizure history, and elements of neurologic exam. Length of stay and ultimate discharge location were recorded. Patients were determined to have seizures based on EEG report generated by board-certified neurophysiologists. For patients with electrographic seizures 2HELPS2B Score and duration of EEG until first seizure were calculated. Preliminary analysis examined 100 patients and used descriptive statistical measures.

Results: 48% of EEGs were placed for isolated encephalopathy. 8% of patients had electrographic evidence of seizures, 3 had clinical correlate. Please see attached table for comparison of features seen in patients who had seizure on EEG (n=8) and patients who did not (n=92). When comparing patients with (n=60) and without (n=40) acute or chronic brain injury on imaging using Fisher’s Exact Test, there was no significant association between incidence of seizure on EEG and evidence of brain injury (p=0.471). For the 8 with electrographic seizures, the average 2HELPS2B Score was 1.88, and average EEG duration until first seizure detected 26 hours. The average ICU length of stay was 14.5 days and ultimately 87.5% were deceased at discharge.

Conclusions: Preliminary analysis suggests that although almost half of EEGs are placed for isolated encephalopathy, the majority of patients found to have electrographic seizures presented with clinical seizure-like activity beyond isolated encephalopathy prior to EEG placement. Variables such as ocular movement abnormalities and abnormal MRI or CT findings may help predict NCSE and distinguish situations where there could be benefit for longer duration of EEG despite 2HELPS2B Score suggesting otherwise. We plan to collect additional data and use forward stepwise logistic regression for creation of a tool to aid in selection of encephalopathic patients for urgent EEG.

Funding: Please list any funding that was received in support of this abstract.: None.