Authors :
Presenting Author: Suyi Ooi, MBBS FRACP – The Florey, University of Melbourne
Graeme Jackson, MBBS PhD FRACP – Clinical Director of The Florey, Neurologist, Professorial Fellow of the University of Melbourne, The Florey, University of Melbourne; Naoto Nagino, MBBS – PhD student, The Florey, University of Melbourne; Chris Tailby, BA/BSc (Hons) PhD Psych (Clin Neuro) – Cognition Group Lead, The Florey Epilepsy Imaging Division, The Florey, University of Melbourne; David Vaughan, MBBS PhD FRACP – Neurologist, Senior Research Fellow, Deputy Director at the Florey Node for the National Imaging Facility, Epilepsy Imaging Division of The Florey, The Florey, University of Melbourne
Rationale:
Applicability of a prediction tool for clinical outcomes after the first seizure relies on accurate diagnoses made at the First Seizure Clinic (FSC). Our aims were to (1) assess factors influencing diagnostic accuracy and the development of epilepsy, and (2) identify areas where a clinical prediction tool is needed.
Methods:
We conducted a retrospective analysis of consecutive patients seen at Austin Health FSC (Melbourne, Australia) between 2016 and 2021 andfollowed up for at least six months.
Results:
Out of 1056 FSC patients, (mean age 41.4, standard deviation 20.4 years old, 54% male), 478 (45%) whom had minimum 6 months follow-up were analyzed and were divided into three categories: (1) A ‘correct diagnosis’ was defined as a sustained diagnosis with six or more months of follow-up and those who developed epilepsy from first unprovoked seizure, comprising 83% of patients; (2) An ‘incorrect diagnosis’ where a diagnostic revision occurred in 5% of cases; (3) A ‘complicated’ group (12%), which included patients with an initial unclassified diagnosis who were later reclassified (9%), and patients who developed epilepsy from an initial diagnosis of acute symptomatic seizure (3%). In categories (2) and (3) whom had a diagnostic revision or diagnosis determined, 33% of patients were diagnosed with epilepsy after an initial event that was not provoked or unprovoked seizure.
Of the 1056 patients, 42% (n=73/172) developed epilepsy after first unprovoked seizure (median follow-up 12.5, interquartile range 32 months, median time to seizure recurrence 5.8 [95% confidence interval 4.7 – 8.2] months). The most common reason for the development of epilepsy were further seizures (74%), followed by an abnormal EEG (27%) and MRI (10%) after the initial FSC appointment.
Conclusions:
In this large retrospective analysis of consecutive first seizure clinic patients, the majority of initial diagnosis were sustained over time.
However, one in three individuals were reclassified to epilepsy from an initial diagnosis of seizure mimic, highlighting caution in misdiagnosis of seizure mimics.
The high rate of seizure recurrence to diagnose epilepsy highlights the importance of longitudinal follow-up and early investigations, but importantly reinforces the need for more robust multimodal clinical prediction tools after the first seizure.
Funding: Not applicable