Objective:

We hypothesized that utilization of the Ceribell Rapid Response EEG (RR-EEG) can reduce time to diagnosis in a level one trauma center in the Department of Defense. A secondary aim was to evaluate utilization practices and cost savings.

Background:

Non-convulsive seizures (NCS) and non-convulsive status epilepticus (NCSE) have been found in up to 27% of patients that are critically ill and often go unrecognized or have delays in recognition. Electroencephalography (EEG) is the gold standard diagnostic tool but is a limited resource, requires trained technicians and specialized neurologists, thereby potentially leading to delays in diagnosis and treatment.

Design/Methods:

We obtained 50 rr-EEG headsets from the manufacturer Ceribell Inc as part of quality improvement effort. From September 2021 through April 2022, Neurology residents triaged consults concerning for NCS or NCSE and based on clinical suspicion performed rr-EEG or ordered conventional EEG. Data was collected to include time of consult, time of initiation, time of interpretation, provider’s clinical impressions, and final diagnosis of seizure or not.

Results:

Average time to initiation of data collection for rr-EEG was 48.94 minutes compared to 128.05 minutes for conventional EEG  (p-value <0.001). Average time to interpretation of rr-EEG was 58.59 minutes compared to 246.29 minutes time of interpretation after downloading the conventional EEG. We compared time of interpretation of the rr-EEG (58.59 minutes) to the initiation of the conventional EEG (128.05 minutes) due to delays associated with work-flow and formal reading of the EEGs. This yielded a p-value of <0.001.

Conclusions:

rr-EEG reduced time to initiation and interpretation by over 50% compared to conventional EEG as utilized by neurologists at a busy Level 1 trauma center. rr-EEG also reduced cost per EEG while increasing utilization and availability of EEG technology.