Abstracts

Rationale: Psychogenic nonepileptic seizure (PNES) patients are often misclassified as having epilepsy, leading to a 7—10 year diagnostic delay. Accurate PNES screening tools are needed to quickly identify PNES patients and hasten referral for VEEG monitoring, the diagnostic gold standard. Studies have shown that PNES patients are more likely to present with chronic pain and multiple unexplained symptoms, such as tingling, numbness, dizziness, etc. The objective of this study is to determine if a standard patient-completed review-of-systems sheet can differentiate PNES from epilepsy patients. Methods: We sought to retrospectively review outpatient clinic charts of 100 consecutive adult epilepsy monitoring unit (EMU) patients undergoing VEEG evaluation for seizures at University Hospitals Case Medical Center. All referring epileptologists at our institution utilize an identical patient-completed review-of-symptoms checklist consisting of 66 symptoms. We used logistic regression to identify PNES-specific symptoms, according to highest odds ratios and significant p-values. We investigated the utility of three scores hypothesized to predict PNES, 1) Total number of checked-off symptoms, 2) Total number of checked-off pain symptoms (ten total; e.g., abdominal pain, chest pain, headache, etc.), and 3) Total number of PNES-specific symptoms. We generated post-estimation probability plots to identify critical scores on each of the three hypothesized measures, that correspond to high risk of having PNES. Results: We were not able to access charts for ten of the 100 EMU patients since they were referred by physicians outside of our institution, leaving 90 subjects for analysis. Using strict diagnostic criteria, there were 51 (56.7%) epilepsy subjects and 26 (28.9%) PNES subjects. For analysis, the six subjects who had both epilepsy and PNES were included in the PNES group and excluded from the epilepsy group, since the purpose of this study was to identify PNES patients. Logistic regression identified seven PNES-specific symptoms, 1) numbness/tingling (OR 4.6, p<0.01), 2) tiredness/fatigue (OR 3.5, p=0.01), 3) anxiety (OR=3.4, p=0.04), 4) headache (OR=3.2, p=0.02), 5) neck/back pain (OR=2.9, p=0.05), 6) depression (OR=2.9, p=0.08), and 7) trouble with walking/balance (OR=2.9, p=0.07). Post-estimation probability plots indicated that subjects checking off either three or more pain symptoms, three or more PNES-specific symptoms, or at least any five symptoms are at highest risk of having PNES.