Cognitive functioning in people with epilepsy plus severe learning disabilities: a systematic analysis of predictors of daytime arousal and attention.
Espie CA., Paul A., McColl JH., McFie J., Amos P., Gray J., Hamilton DS., Jamal GA.
In spite of the high prevalence of epilepsy and the importance of preserving cognitive function in people with learning disabilities, this population has received relatively little research attention. This study sets out systematically to investigate possible predictive factors of inter-ictal states of arousal and attention. The daytime function of 28 people with epilepsy and severe learning disabilities was assessed by performance on a two-choice reaction time vigilance task, behavioural analysis of time-sampled video recordings taken in naturalistic settings, and carer ratings on visual analogue scales. This methodology yielded eight discrete functional measures, from which two further index measures were derived after principal components analysis. A range of clinical and psychosocial assessments was completed and subjects had 36 hour ambulatory EEG and sleep EEG monitoring. Regression models identified significant predictors of cognitive function from a range of potential explanatory variables i.e. demographic, clinical, pharmacological, background EEG rhythms and sleep parameters. Results indicated that greater severity of learning disability, longer bedtime periods, poor sleep efficiency, frequent seizures and antiepileptic drug polytherapy were significant predictor variables. Explained variance (adjusted R2) was greater than 50% for six of 10 outcome variables (range up to 85%). Furthermore, significant regression equations (P < 0.05) were obtained for all but one variable. Thus, these results appear reasonably robust. Results support an interactional model of daytime arousal and attention in people with epilepsy plus severe learning disabilities. Inter-ictal cognitive function appears to be mediated by a combination of organic, circadian (sleep wake), clinical and pharmacological factors.