Reconstruction and validation of a short version of the scale for suspicion of psychogenic non-epileptic seizures in a Japanese cohort

PURPOSE: Functional/dissociative seizures (FDS) are challenging to diagnose. Video-electroencephalography monitoring, the gold standard for the diagnosis, is impractical for all patients. To improve FDS diagnosis, a scale for suspicion of psychogenic non-epileptic seizures (SS-PNES) was developed in Brazil, consisting of 15 items, on a scale of 0-3 for each item, combining epileptological questions, current and past psychiatric characteristics. A total score of 20 or more is considered as FDS-like. Here, the reliability and validity of SS-PNES were evaluated in a Japanese cohort. METHODS: The subjects were 250 of 326 patients who received a comprehensive epilepsy evaluation at our institution. Factor analyses were applied to the SS-PNES data. RESULTS: A two-factor model with six items was found to be optimal for FDS diagnosis, demonstrating acceptable reliability coefficients and validity to identify FDS. The two factors were named "emotional stress" and "psychiatric characteristics." Most patients with FDS and comorbid epilepsy showed high scores for both factors. However, nearly 40 % of epileptic seizure-only patients were misdiagnosed as FDS-like. CONCLUSION: In our Japanese cohort, the original 15-item scale did not show reproducibility, so a shortened version with a two-factor structure was developed and evaluated. The new SS-PNES-short had high sensitivity but low specificity for identifying FDS. The addition of epilepsy-specific factors may improve the specificity of SS-PNES.