Abstracts

Rationale: The prevalence of psychiatric comorbidities in patients with epilepsy is estimated to be 32%; psychogenic nonepileptic seizures are usually a result of an underlying psychiatric condition. The purpose of this study was to evaluate whether existing psychiatric comorbidities and psychiatric medications at the time of admission to the epilepsy monitoring unit (EMU) correlate with discharge diagnosis. Methods: We reviewed medical records of patients admitted to the adult EMU at Temple University Hospital from January 2016 to March 2018. Of 185 patients admitted to EMU within this timeframe, 165 patients (93 females) were admitted for characterization of their events. Age range was 18 to 85 years (mean 44 years). EMU length of stay ranged from 1 to 11 days (mean 4 days). Excluding 1 patient with a physiological event, 101 patients had at least one event captured on video EEG categorized by epileptologists as epileptic seizures (ES), psychogenic nonepileptic seizures (PNES) or a combination of epileptic and psychogenic nonepileptic seizures (ES+PNES). Results: Of 101 patients, 38 (37.6%) had ES, 52 (51.5%) had PNES and 11 (10.9%) had ES+PNES. Sixty-three (62.4%) patients (63.5% females and 36.5% males) had at least one psychiatric diagnosis at the time of admission to EMU. They had a higher rate of PNES diagnosis with statistical significance compared to the group without a psychiatric diagnosis (66.7% versus 26.3%; p<0.01). The group with psychiatric comorbidity had a significantly lower rate of ES diagnosis compared to the group without psychiatric comorbidity (20.6% versus 65.8%; p<0.01). There was no statistically significant difference in both groups for diagnosis of ES+PNES (p=0.45). Fifty-three (52.5%) patients were on psychiatric medications at the time of admission to EMU. This group had significantly more patients diagnosed with PNES (67.9% versus 33.3%; p<0.01) and significantly fewer patients diagnosed with ES (20.8% versus 56.3%; p<0.01) compared to the group without psychiatric medications. There was no statistically significant difference in both groups for the diagnosis of ES+PNES (p=0.88). Conclusions: To our knowledge, this is the first study to report the likelihood of diagnosis of ES, PNES or both based on prior established psychiatric diagnosis and use of psychiatric medications in the EMU setting. Our study demonstrates that patients admitted to EMU for event characterization with any psychiatric diagnosis or use of psychiatric medications are more likely to be to be diagnosed with psychogenic nonepileptic seizures. Conversely, patients lacking comorbid psychiatric diagnoses who are not on psychiatric medications are more likely to have epileptic seizures. This is a single center study with small sample size. Prospective multicenter studies are needed to help differentiate epileptic versus psychogenic nonepileptic seizures. Funding: None