Abstracts

Rationale: Frontal lobectomy is the second most common type of resective epilepsy surgery, yet few studies have examined postoperative cognitive outcome in children. A 2011 review identified only four published studies with sample sizes ranging from 8-12 children. The current study used epilepsy-specific reliable change indices (RCIs) to examine cognitive outcome on a wide range of neuropsychological measures following frontal lobe resections (FLR) in a larger sample of children. Methods: Forty children [M(age)=10.0, SD=2.4; 63% male; dominant/nondominant resections=18/22] met selection criteria: aged 6-16 years, FLR for epilepsy between 1998 and 2015, pre- and post-operative neuropsychological evaluations [M(test-retest interval)=10.8 months, SD=6.3], and no prior neurosurgery. To determine clinically significant cognitive change after FLR, change scores for each cognitive variable were calculated and published RCIs for children with epilepsy were used to categorize patients as improved, no change, or declined. For variables without RCIs, a change score +/-1 standard deviation categorized patients. Using chi-squares, we determined whether base rate of cognitive change following FLR was greater than expected by chance alone (i.e., 5% improve, 5% decline) and examined outcome differences as a function of surgical side (dominant/non-dominant). Chi-square and ANOVA analyses identified variables that may be associated with greater than expected post-operative cognitive change (i.e., age at onset and surgery, duration of epilepsy, preoperative cognitive performance, pathology, and seizure outcome). Results: A larger than expected proportion of patients demonstrated clinically meaningful cognitive changes (improvement or decline) on over half of the measures assessed (?2>6.2, p < .05). Specifically, greater than expected declines were found on measures of intelligence, verbal fluency, perseverative responses, and visuomotor integration. In contrast, there were greater than expected improvements on measures of processing speed and contralesional manual dexterity. Interestingly, greater than expected change was observed in both directions on measures of sequencing/switching, planning, verbal memory, and ipsilesional manual dexterity (Figure). There were no differences in outcome as a function of surgical side. Better baseline performance conferred greater risk for postoperative decline for 4 of the 10 measures. Poor seizure outcome was associated with decline in verbal delayed memory. Later age at seizure onset was associated with decline in verbal fluency (Table). No associations were apparent between post-operative cognitive change and age at surgery, duration of epilepsy, or pathology. Conclusions: Approximately one-third of our sample demonstrated clinically meaningful declines in cognition after FLR. This was most notable on executive function and verbal memory measures. Interestingly, another third of the sample demonstrated postoperative improvements in these domains. Further research is needed to determine the role of resection site (e.g., orbitofrontal, lateral, mesial) and extent on cognitive outcomes in children and the role of cognitive outcome on quality of life to better inform physicians and families about cognitive risk with FLR. Funding: Support for this research was provided Epilepsy Foundation, Targeted Research Initiative for Youth (JSH & RMB: 246507) and Partnership for Pediatric Epilepsy Research (JSH & RMB: 189351) to RMB and JSH; National Institutes of Health (DPF: 1K23NS091344 - 01A1); and the Cleveland Clinic Epilepsy Center