To determine the utility of suggestive seizure induction for inpatient work‐up of suspected psychogenic nonepileptic seizures (PNES).
Prospective study of epilepsy center inpatient admissions with suspected PNES. Patients were randomized to undergo suggestive induction first (group A) and then, if necessary, long‐term video–electroencephalography (EEG) monitoring, or vice versa (group B). Diagnostic pathways were compared. Potential clinical predictors for diagnostic success were evaluated.
Length of in‐hospital stay did not significantly differ between groups. Suspicion of PNES was confirmed in 43 of 77 (56%) patients, evenly distributed between group A (22 of 39) and group B (21 of 38). In nine patients, recorded habitual seizures were epileptic and in 25 cases, no diagnostic event could be recorded. Diagnosis of PNES was ascertained primarily by recording a typical seizure through suggestive induction in 24 patients and through long‐term monitoring in 19 patients. In group A (induction first), monitoring was not deemed necessary in 21% of cases. In group B (monitoring first), 13% would have remained inconclusive without suggestive induction. Patients who reported triggers to their habitual seizures were not more likely to have spontaneous or provoked PNES during monitoring or suggestive inducion, respectively. Patients with subjective seizure prodromes (auras) were significantly more likely to have a PNES during suggestive induction than those without (odds ratio [OR] 3.4, 95% confidence interval [CI] 1.1‐10.4). There was no significant difference in seizure frequency between patients with spontaneous PNES during long‐term monitoring and those with nondiagnostic monitoring sessions.
Our results support the notion that suggestive seizure induction can reduce the number of inconclusive inpatient workups, and can obviate resource‐intensive long‐term monitoring in one fifth of cases. Patients who are aware of prodromes might have a higher chance of having seizures induced through suggestion.