Abstract found on PubMed
Objective: Early differential diagnosis of psychogenic nonepileptic seizures (PNES) and epileptic seizures (ES) remains difficult. Self-reported psychopathology is often elevated in patients with PNES, although relatively few studies have examined multiple measures of psychopathology simultaneously. This study aimed to identify differences in multidimensional psychopathology profiles between PNES and ES patient groups.
Method: This was a retrospective case-control study involving patients admitted for video-EEG monitoring (VEM) over a two-year period. Clinicodemographic variables and psychometric measures of depression, anxiety, dissociation, childhood trauma, maladaptive personality traits, and cognition were recorded. Diagnosis of PNES or ES was determined by multidisciplinary assessment and consensus opinion. General linear mixed models (GLMMs) were used to investigate profile differences between diagnostic groups across psychometric measures. A general psychopathology factor was then computed using principal components analysis (PCA) and differences between groups in this ‘p’ factor were investigated.
Results: 261 patients (77 % with ES and 23 % with PNES) were included in the study. The PNES group endorsed greater symptomatology with GLMM demonstrating a significant main effect of group (?2p = 0.05) and group by measure interaction (?2p = 0.03). Simple effects analysis indicated that the PNES group had particularly elevated scores for childhood trauma (? = 0.78), dissociation (? = 0.70), and depression (? = 0.60). There was a high correlation between psychopathology measures, with a single p factor generated to explain 60 % variance in the psychometric scores. The p factor was elevated in the PNES group (? = 0.61). ROC curve analysis indicated that these psychometric measures had limited usefulness when considered individually (AUC range = 0.63-0.69).
Conclusion: Multidimensional psychopathological profile differences exist between patients with PNES and ES. Patients with PNES report more psychopathology overall, with particular elevations in childhood trauma, dissociation, and depression. Although not suitable to be used as a standalone screening tool to differentiate PNES and ES, understanding of these profiles at a construct level might help triage patients and guide further psychiatric examination and enquiry.