Abstracts

Rationale: Effective treatment of seizures is dependent on the ability to quickly and accurately distinguish the event as either epileptic or nonepileptic. Psychogenic nonepileptic seizures (PNES) are prevalent among people without epilepsy and occasionally those can coexist with epileptic seizures (ES). Several semiologic signs including ictal eye closure,1 pelvic thrusting, and fluctuating course 2 have been described as useful diagnostic features in PNES. However, only a small number of studies has reported diagnostic sensitivity, specificity, and predictive values of various semiologic signs in PNES (Table 1). Even those studies are mostly based on small samples of patients. Against this backdrop, in this study, we aim to assess the significance of head shaking during an event as being indicative of PNES from a relatively large cohort of patients. Methods: The events of all patients who underwent video-EEG monitoring at the epilepsy monitoring unit of Monash Medical Centre, Australia, between May 2009 and November 2015 were retrospectively reviewed. Scalp EEG was recorded using the international 10-20 system of electrode placement. Sleep deprivation and reduction or withholding of antiepileptic drugs was routinely implemented to improve the diagnostic yield of the test. The final diagnosis of ES or PNES had been confirmed based on the consensus opinion of at least two epileptologists who had reviewed clinical and EEG data prior to this study. Two investigators (EM & HMAY), blind to the final diagnosis, independently reviewed all video-EEGs and tabulated the presence or absence of head shaking (side-to-side head movement) during the seizure. In cases of discrepancy, the video-EEG was reviewed by the third investigator (US) to arrive at a consensus. Patients with subclinical or non-motor seizures were excluded from the analysis. We calculated the diagnostic specificity, sensitivity, and predictive values of head shaking in PNES.  The Chi-square test was used to explore the relationship of the head shaking sign between ES and PNES. The data analysis was performed with IBM SPSS (version 21) statistical software package (IBM Corporation, New York, USA). Results: The monitoring of a total of 111 patients during the study period resulted in 60 patients being diagnosed with ES and the remaining 51 with PNES. Ages ranged between 19 and 87 years (mean age =34) and included 51 (45.9%) males and 60 (54.1%) females. A total of 585 seizures (ES 357, 61.0% & PNES 228, 39.0%) were captured with 72.1% of patients having more than 1 event during hospitalization.  Table 2 details the seizure types, corresponding number of patients, and evidence of head shaking. A significantly higher proportion of PNES demonstrated head shaking during seizures (62 PNES=27%) compared with 8.7% (31 of 357) ES (p=<0.0001). The diagnostic sensitivity of head-shaking sign was calculated at 0.27 and the specificity at 0.91. The positive and negative predictive values were 0.67 and 0.66, respectively. The positive likelihood ratio equalled 3 and negative likelihood ratio was 0.8. Conclusions: These findings suggest that head shaking is significantly more frequent during PNES and can be a useful sign to distinguish PNES from ES. The high specificity indicates a higher probability of the test being negative (absence of head shaking) in the absence of PNES. However, low sensitivity and moderate predictive values are drawbacks that should be borne in mind when interpreting this sign. Funding: No funding/sponsorship