Abstracts

RATIONALE: Previous studies have shown psychological testing may differentiate patients with psychogenic, non-epileptic seizures (NES) from those with epileptic seizures (ES). The Minnesota Multiphasic Personality Inventory (MMPI) has received the most attention as it commonly reveals an increased incidence of somatization, somatoform, and conversion profiles in NES patients. Other psychometric measures may further enhance diagnostic accuracy beyond the MMPI. Memory symptom validity testing, commonly used in neuropsychology, may help identify patients who exhibit suboptimal effort and motivation on neuropsychological examinations. Recent studies have shown patients with NES may be more likely to perform poorly on such tests. The objective of this study was to evaluate performance on the Test of Memory Malingering (TOMM) in patients with NES and ES, and determine its contribution to differential diagnosis of NES vs. ES.
METHODS: Subjects were drawn from a consecutive series of adults who underwent long-term video-EEG monitoring for purpose of differential diagnosis or to determine suitability for epilepsy surgery. Inclusion criteria were 1) definitive diagnosis of NES or ES made by a board certified clinical neurophysiologist based on ictal EEG findings and clinical semiology of the seizures, and 2) completion of both the TOMM and MMPI-2 as part of a routine neuropsychological screening evaluation. Patients with both seizure types (comorbid NES and ES), only subjective spells, or no typical spells recorded were excluded. 91 patients with NES and 168 patients with ES met these criteria.
RESULTS: Significant differences in performance were observed on the TOMM with 9% of the NES group and 1% of the ES group scoring in the invalid range ([lt]45) on trial 2 ([chi]²=9.6, p=.002). On the MMPI-2, mean scores of the NES group were significantly (p[lt].05) higher than ES group on clinical scales 1 to 8, with the greatest differences on scales 1 (p[lt].001) and 3 (p[lt].001). Stepwise logistic regression was used to select scores from the MMPI-2 and TOMM that contributed independently to diagnostic classification. The following variables were identified as significant predictors: MMPI-2 Hysteria scale (p[lt].001), performance on trial 1 of the TOMM (p[lt].01), and MMPI-2 Hypochondriasis scale (p[lt].05). These combined measures contributed to an overall classification rate of 79% (89% of patients with ES, and 60% of patients with NES were correctly classified)
CONCLUSIONS: Our findings reveal that the majority of patients with NES or ES demonstrate adequate effort and motivation on memory symptom validity tests. While an invalid TOMM performance indicating suboptimal effort or embellishment on neuropsychological exams may be rare in patients with epilepsy, it occurs in a significant minority of those with NES. Consequently, performance on symptom validity tests such as the TOMM may contribute to establishing a diagnosis in patients with seizures of unknown etiology.