Abstracts

Rationale: It is well established that naming often declines following left (i.e., language dominant) but not right (i.e., nondominant) epilepsy surgery in adults. Although word retrieval abilities are critical for children who are in the process of developing expressive verbal skills, the fate of naming following epilepsy surgery in children is not as well understood, as most pediatric studies have used naming measures that may not be age appropriate, or have failed to account for hemispheric language dominance, potentially obscuring differences between those who had dominant vs nondominant hemisphere surgery. Using measures developed specifically for children, we examined postoperative auditory and visual naming in children following unilateral epilepsy surgery for refractory unilateral seizures involving the dominant or nondominant hemisphere, and in children with atypical (i.e., right or bilateral) language organization. We hypothesized that left hemisphere patients would decline postoperatively, whereas right hemisphere patients might exhibit improvement in naming (i.e., contralateral functioning).

Methods: Participants were 76 children who had epilepsy surgery (Left, n=29: resection: 15, laser ablation: 4, lesionectomy: 9, gamma knife: 1; Right, n=31: resection: 21, laser ablation: 5, lesionectomy: 5; Atypical n=16: resection: 10, laser ablation: 2, lesionectomy: 4) and pre and postoperative testing using the Children’s Auditory (ANT) and Visual Naming (VNT) tests. Performance was assessed using time-based Summary Scores (Number of items named rapidly [< 2 sec] – Number of items named after a delay [2-20 sec], i.e., a hallmark of naming difficulty, characterized by conscious instead of automatic processing). Language dominance was based on fMRI (n=33), right handedness and/or neuropsychological profile (n =20), dichotic listening (n=13), Wada (n=6), cortical mapping (n=3), or MEG (n=1). Multivariate ANOVA compared groups on demographic and clinical variables; paired T-tests compared pre and postoperative ANT and VNT performance.

Results: Across groups, there were no differences in age, IQ, or seizure outcome (left: age: 12.2 ± 2.7, IQ: 93.8 ±10.3, Engel outcome: 1.3 ±0.7); right: age: 12.0 ± 2.9, IQ: 90.3, ± 13.3, Engel outcome: 1.4 ± 0.9; Atypical: age: 11.7 ± 2.9, IQ: 87.7 ±11.9, Engel outcome: 1.6 ±0.8). Preoperatively, ANT scores were lower in those with left hemisphere seizures (p=.02), with no VNT differences across groups. However, postoperatively, VNT (but not ANT) performance declined after left hemisphere surgery (p = .02), whereas ANT (but not VNT) performance improved after right hemisphere surgery (p= .03). The Atypical group showed no naming changes.

Conclusions: Consistent with adult findings, children with left hemisphere seizures and surgery exhibited compromise to auditory naming preoperatively, yet postoperative decline was limited to visual naming. The dissociation between ANT and VNT performance, both pre and postoperatively, suggests that like adults, neurofunctional mechanisms underlying auditory and visual naming are separable in children.

Funding: NIH R01NS035140