The Clinical Utility of BioEP in Diagnostic Decision Making in Epilepsy (CITADEL)

A Multi-centre, Prospective, Two-arm Randomised Controlled Trial to Determine the Clinical Utility of BioEP in Diagnostic Decision Making in Epilepsy

We plan a prospective multisite trial to determine the clinical utility of BioEP in the context of diagnostic decision making. We will use findings from the trial to improve user experience of our Neuronostics platform (the tool which clinicians use to obtain a BioEP score from EEG and the aligned report). The data coming from the trial will also enable us to iterate the BioEP algorithms and so improve future performance.

We will conduct a two-arm randomised control trial with two groups: (i) Usual Care, (ii) Usual Care + BioEP score and report. In this context, usual care involves clinical/ patient history, eye-witness accounts and any standard tests ordered (EEG, MRI, blood tests etc.). We will recruit adults attending first seizure clinics where there is a suspicion that epilepsy is the causing symptoms and so a first EEG is recommended. Participants will be sourced from 10 sites. The trial duration will be 1 year recruitment, and 2 year follow up period. Randomisation: Consenting individual patients will be randomly assigned to either Usual Care or Usual Care + BioEP in a 1:1 ratio. Randomisation will be stratified by centre, with eligible and consenting participants allocated in random blocks. The allocation schedule will be generated by the trial statistician, and concealed from the clinicians assessing eligibility and recruiting patients.

When a consenting participant has their first EEG, they will be assigned with a unique participant study identifier before being uploaded. The centre will convert the EEG to EDF format (or upload the total file) and upload this to the Neuronostics platform. The EEG will be stored on the Neuronostics database management system, within which the computational and mathematical analysis will be carried out. This study involves data analysis and mathematical modelling of the EEG recordings of each patient individually to generate a computational analysis score (BioEP).

The working diagnosis timepoint is defined as the diagnostic decision of the clinician after the continuous process of information gathering, integration, and interpretation and deciding that sufficient information has been gathered to make a definitive judgment. Initial suspicions of a provoked or isolated seizure leading to a watch-and-wait strategy are not regarded as a working diagnosis. We propose that the addition of BioEP information will result in a reduction in the median time to a working diagnosis versus usual care, due to the additional available information provided by the BioEP score.

Inclusion Criteria: Adult (age 18 and above) presenting with first suspected seizure(s) Able to give informed consent Patient receives EEG following clinicians' decision to refer for an EEG based on clinical history and seizure description taken during first seizure clinic. Exclusion Criteria: Participants unable to tolerate an EEG test so no EEG data were gathered Participants with a known hepatic/renal encephalopathy Participants that upon history taking have a clear clinical diagnosis of a physical condition other than epilepsy (e.g. vasovagal syncope)