Abstracts

Rationale: Accurate evaluation of postsurgical cognitive outcome is essential for informing families about potential cognitive risks/benefits associated with pediatric epilepsy surgery. To address this need, children with epilepsy were evaluated twice to develop reliable methods for assessing cognitive change. These methods were then used to examine the base-rate of cognitive change in a second group of children after epilepsy surgery. Methods: Sixty-three nonsurgical participants (6-16 years, mean age = 10.19, confirmed epilepsy with stable AED regimen, no neurosurgery or neurodegenerative disorder) completed neuropsychological evaluations about 9 months apart (the average pre- to post-epilepsy surgery assessment interval). The nonsurgical group's test scores were used to develop practice-adjusted Reliable Change Index (RCI) cutoff and Standardized Regression Based (SRB) change scores according to established methods. The resulting RCIs and SRBs were used to examine meaningful cognitive change on measures of intelligence (WISC-IV) and memory (CMS) in 156 children (mean age = 10.91) who underwent epilepsy surgery. Change scores for each surgical participant were calculated and were then classified as improved, declined, or unchanged based on RCIs and SRBs (90% confidence interval). Chi-square analyses were used to determine whether the proportion of observed changes differed significantly from expected proportions (i.e., 5% improvement, 5% decline). Results: Based on these RCIs, surgical patients changed more than expected by chance on all WISC-IV Indices except Perceptual Reasoning. A greater than expected number of patients declined (8-11%) and a greater than expected number improved (6-10%). Significantly greater change than expected based on chance was seen on all CMS Indices except Delayed Recognition. More children than expected declined on the Visual Immediate and Delayed Memory Indexes (13% and 15%). On the Verbal Immediate and Delayed Memory Indexes, more children than expected showed meaningful improvements (15% and 7%) and declines (16% and 11%). Many (20%) demonstrated significant improvements, while few (4%) showed declines the on the Attention/Concentration Index. Cognitive outcome in the surgical sample using SRBs revealed similar findings. Chi-square analyses comparing the proportion of patients classified as declined, improved, or unchanged using RCIs versus SRBs revealed no significant difference between the two methods. Conclusions: RCIs and SRBs provide empirically-based methods for evaluating clinically meaningful cognitive change following epilepsy surgery. While more children demonstrated score declines than expected by chance after epilepsy surgery, most improved or did not show a meaningful change in intellectual and memory scores. Reliable change methods indicate that epilepsy surgery is generally well-tolerated from a cognitive standpoint.