(730) Impact of Point-of-Care EEG on Hospital Length of Stay and Functional Outcomes at Discharge

Introduction: Nonconvulsive seizures and status epilepticus are associated with poor clinical outcomes. Recent work by our group has shown that in patients admitted to the intensive care unit (ICU), there was an association between access to point-of-care electroencephalography (POC-EEG) and shorter length of stay (LOS) and better functional outcomes at discharge. We pose that use of POC-EEG (Ceribell, Inc.), regardless of ICU admission, will positively impact their overall hospital stay and outcomes at discharge.

Methods: This was a cohort sub-analysis of the retrospective SAFER-EEG trial, in three sites that had access to conventional EEG (convEEG) and POC-EEG. We included patients who received EEG monitoring at any point during their hospital stay, as part of their standard of care, and who had valid modified Rankin Scale (mRS) scores before EEG monitoring and at discharge. Abnormal EEG findings, extracted from the neurology report, were reviewed by the original research team and used in this analysis to identify patients who had seizures or epileptiform activity (EA) during their stay. Propensity-score matching was done to control for potential clinical confounders, and non-parametric tests were used to compare continuous variables.

Results: We included 491 patients (convEEG = 133, POC-EEG = 358). After matching, we found that median hospital LOS was significantly shorter for the POC-EEG group, compared to convEEG (8.6 d [IQR: 4.6, 18.1] vs. 11.8 d [6.3, 27.3]; p=0.009). When assessing the changes in mRS scores between discharge and pre-EEG in the matched cohorts, we found that POC-EEG had a smaller worsening of scores than the convEEG group (median delta: 2 [0, 3] vs. 3 [1, 4]; p = 0.001).
In the POC-EEG cohort, 152 patients (42.5%) had seizure or EA (Sz/EA), while 53 patients (39.8%) in the convEEG cohort were Sz/EA positive. After matching in these sub-groups, POC-EEG had a trend of shorter hospital LOS compared to convEEG (9.9 d [5.3, 21.9] vs. 17.8 [7.7, 29.7]; p = 0.09).

Conclusions: This analysis showed that access to POC-EEG monitoring, including outside the ICU, is associated with better clinical outcomes, like reduced hospital LOS and changes in functional disability scores at discharge. Future prospective studies need to confirm these findings.