Differential diagnosis between epileptic and psychogenic nonepileptic seizures through conversational analysis: A blinded prospective study in the Russian language

The current study examined the validity of conversational analysis (CA) in Russian patients with seizures, using a scoring table for the Simplified Linguistic Evaluation (SLE). The study sample was composed of 12 adult participants suffering either from epilepsy (ES) or psychogenic nonepileptic seizures (PNES) recruited in the Moscow Research and Clinical Center for Neuropsychiatry. Definitive diagnosis was established only after a habitual event was captured on vEEG. All participants with PNES or ES and at least one mental disorder underwent a 20-minute-long interview recorded on video. The interview then was evaluated by the external blinded physician already experienced in CA. Finally, that physician filled the SLE, consisting of 5 items analyzing the main characteristics of patient narrations. A score of 12 sug-gested a diagnosis of ES, while a score of <12 suggested a diagnosis of PNES. The blinded evaluator correctly identified 11 out of 12 cases. The concordance between the vEEG diag-nosis and the CA diagnostic hypothesis was 91.67%. The sensitivity of the scoring table was 100%, while the specificity was 80%. The positive and the negative predictive values were, respectively, 87.5% and 100%. Our results suggested that the differences in seizure descriptions between patients with PNES and patients with ES are similar across Indo-European language family and are independent of psychi-atric comorbidity. (c) 2021 Elsevier Inc. All rights reserved.

Automated summary

This study examined the validity of conversational analysis (CA) in Russian patients with seizures using a scoring table for the Simplified Linguistic Evaluation (SLE). The results showed that CA combined with SLE could effectively differentiate between patients with epilepsy and psychogenic nonepileptic seizures, demonstrating a high diagnostic accuracy.