38 Are there any factors that predict the diagnosis of epilepsy or Psychogenic Non-Epileptic Seizures (PNES) in patients admitted to a specialist epilepsy unit?

Abstract

Introduction Differentiating between epilepsy and psychogenic non-epileptic seizures (PNES) can be difficult. Although clearly not a substitute for taking a careful history, certain patient characteristics may assist the clinicians towards diagnosis. The population of patients referred to an epilepsy specialist centre represent a complex and distinct group of patients and it is not clear which factors, if any, could point towards a diagnosis of epilepsy or PNES. Aims/Objectives To investigate if the diagnosis of epilepsy or PNES is predicted by baseline demographic and clinical variables, including Hospital Anxiety and Depression (HADS) scores and medication prescription, in patients admitted to a specialist adult epilepsy centre. Methods We conducted an observational retrospective cohort of consecutive patients admitted to the William Quarrier Scottish Epilepsy Centre (WQSEC) over a period of one year (01/09/16–01/09/17). Chosen predictor variables at baseline included: sex, age, employment education or training (EET), Scottish Index of Multiple Deprivation 2016 (SIMD 2016) rank status, attack frequency, length of index admission, number of anti-epileptic agents prescribed, prescription of benzodiazepines, of analgesia, or of psychotropic medications, and HADS scores. Outcome measures were diagnosis of epilepsy or PNES, from diagnosis made by expert clinicians on discharge from index admission. Because of the presence of dual diagnosis, two multivariable binary logistic regression models were built – one for the epilepsy and one for the PNES diagnosis outcomes. Results 50/73 (69%) of patients admitted were diagnosed with epilepsy and 39/73 (53%) with PNES. These respective groups include 16/73 (22%) who had a dual diagnosis of both epilepsy and PNES. The model to predict epilepsy showed that significant individual predictor variables included number of antiepileptic agents prescribed (Odds Ratio (OR)=3.59 (95%CI: 1.37, 9.42), p=0.010), prescription of psychotropic medications (OR=0.19 (95%CI: 0.04, 0.91), p=0.038), and length of index admission (OR=0.89 (95%CI: 0.81, 0.98), p=0.018). The model to predict presence of PNES revealed only one significant individual predictor variable, which was EET status (OR 0.13 (95%CI: 0.02, 0.86), p=0.035). Conclusions Baseline clinical and demographic factors may be of some utility to the clinician in anticipating a diagnosis of epilepsy or of PNES.