Authors :
Presenting Author: Freya Prentice, BA, MSc – University College London
Oliver Browning, BA – University College London; Maria Eriksson, BSc, MSc – University College London; Patricia Martin-SanFilippo, C Psychol – UCL Great Ormond Street Institute of Child Health – University College London & Great Ormond Street Hospital; Sara Shavel-Jessop, QiCN – UCL Great Ormond Street Institute of Child Health – University College London & Great Ormond Street Hospital; Torsten Baldeweg, PhD – UCL Great Ormond Street Institute of Child Health – University College London
Rationale:
Deficits in verbal memory recall have been consistently associated with left but not right temporal lobe epilepsy (TLE) in adults (Helmstaeder & Elger, 2009). The pediatric literature on side-specific deficits is mixed, with several studies showing poorer delayed verbal memory in children with mesial compared to lateral pathologies regardless of the side of TLE (Roubicek et al., 2020). The task-specificity model (Saling, 2009) predicts greater memory deficits in those with mesial compared to lateral pathologies on memory measures requiring arbitrary but not semantic associations of information, due to the role of the hippocampus in episodic memory (Vargha-Khadem et al., 1997). In this study, we examined whether there are differences in delayed arbitrary- and semantically-associated verbal memory based on the side and lesion location in TLE in children, evaluating side- and task-specificity.
Methods:
We included 76 children assessed at Great Ormond Street Hospital (GOSH) between 2002 and 2022 as part of the epilepsy surgery program. Patients were included if they had a clinical diagnosis of TLE and a preoperative assessment on the delayed word pairs (WP) and stories subtests of the Children’s Memory Scale (CMS). Scaled memory scores were converted to z-scores. Lesion location was categorized based on radiology reports and confirmed by the visual inspection of the MRI scan by an experienced neuroanatomist. All statistical analyses were performed in R version 4.2.3. A 2x2x2 mixed design ANOVA was used to compare between-group (side: left vs right; location: mesial vs lateral) and within-group differences (subtest: WP vs stories) in memory scores.
Results:
There was a significant main effect of the subtest used (F[1, 72]=15.53, p< .001) but not side or location (see group distributions in Figure 1). Post-hoc pairwise comparisons showed significantly lower scores on the stories compared to WP subtests in the left lateral (t[14]=2.47, p=.027) and right lateral groups only (t[6]=4.05, p=.007; see Table 1 for descriptive statistics).