Abstracts

Rationale: Cognitive function in patients with idiopathic generalized epilepsy (IGE) is commonly described as being within the normal range but lower than that of the general population. Limited sample size, methodological variability and the complex contribution of variables such as seizures, anti-epileptic drugs and inter-ictal epileptiform discharges have complicated an understanding of this heterogeneous condition. Neither a systematic review nor quantitative synthesis has been conducted to date. Juvenile myoclonic epilepsy (JME) has been of particular interest, with executive functions (EF) hypothesized as a specific deficit in this IGE subtype. Janz early observations of an immature, emotionally unstable and disinhibited personality style in JME, and subsequent reports of personality traits in JME resembling those of patients with frontal lobe lesions have encouraged investigation of the EF hypothesis. Whether EF deficits are specific to JME and distinct from generalized cognitive deficits remains to be demonstrated. Methods: We conducted a systematic search of Medline and Scopus databases. Eligible studies compared cognitive abilities in an IGE patient group with a healthy control group and were published in English in peer-reviewed journals from 1990 - 2013. Cognitive tests were grouped by cognitive factor-domain defined by the Cattell-Horn-Carroll model. Patient and healthy control group scores were standardized with control group scores used as estimated population parameters. Where studies employed multiple tests of the same cognitive factor, we derived a mean standard score for each factor within each study. A univariate random-effects meta-analysis (with DerSimonian-Laird estimators) was conducted on each cognitive factor, with separate analyses of IGE subtypes where applicable. Results: Of 24 eligible studies, nine studies reported patients with JME, four with CAE, two with IGE with generalized tonic-clonic seizures only, and nine with patients of mixed IGE subtypes. For all cognitive factors except visual processing, summary effects showed statistically significant lower scores in patients with IGE than healthy controls. Effect sizes ranged from 0.39-0.86 of a standard deviation, with speed of cognitive processing the most affected. Meta-analysis of scores on putative tests of executive function yielded a summary effect estimate of 0.69, with studies of JME patients showing the least discrepancy between patient and control group. Equivocal evidence of a trend towards differences between long-term memory ability in CAE and JME patients was observed. Conclusions: The average reduction in cognitive ability factors in patients with IGE compared to healthy controls across studies was as much as 0.86 standard deviation units. In support of the notion of a generalized reduction in cognitive function, all factors except visual processing showed significant impairment. Performance on measures of executive functioning was not more impaired than other cognitive factors, nor was a greater degree of impairment evident in JME samples.