Functional Magnetic Resonance Imaging of Episodic Memory in Left Temporal Lobe Epileptic Children: Evidence for controlateral involvement
Abstract number :
2.121;
Submission category :
5. Human Imaging
Year :
2007
Submission ID :
7570
Source :
www.aesnet.org
Presentation date :
11/30/2007 12:00:00 AM
Published date :
Nov 29, 2007, 06:00 AM
Authors :
S. Martins1, 4, C. Chiron1, 4, I. Jambaque3, B. Guillery-Girard2, E. Eustache2, L. Hertz-Pannier1, 4
Rationale: Temporal lobe epilepsy (TLE) is known to impair episodic memory development in childhood (Jambaque et al, 2007). However, no study has investigated the neural correlates of memory impairment in left TLE children using functional Magnetic Resonance (fMRI). Our aim was 1) to identify mesial temporal lobe (MTL) dysfunctions in left TLE children and 2) to search for potential reorganisation in episodic memory encoding in left TLE (Martins et al., 2007).Methods: Five left TLE children (age: 13,6 years; σ= 0,8) and 8 matched healthy children (age =12,1 years; σ= 1,3) with normal non verbal intelligence participated in a slow event related 1.5T fMRI episodic memory paradigm. During fMRI acquisition, children intentionally encoded 60 comic scenes, and explored 60 corresponding scrambled pictures. To ensure children’s participation, they were asked to answer if a suggested item was present in the previous encoded comic. Responses were recorded, allowing to control for absence of seizure during the experiment. Ten minutes later, the subsequent memory recognition test took place out of the magnet. Children had to recognise the encoded comics among lures (Correct Recognition Ratio = CRecog). fMRI group analysis of whole brain activation was performed using a random-effect model (SPM2, p< 0.05, corrected for multiple comparisons). TLE and healthy children were compared using a Two sample T-test, corrected for multiple comparisons. We then explored the correlations of activations with CRecog in TLE patients.Results: Behavioural data confirmed episodic memory impairments in the left TLE group, when compared with healthy children (CRecog: z= 2,5; p. <0.01). In both groups, activations showed the usual network sustaining episodic memory encoding (Menon et al 2006). However, group comparisons revealed restricted differences: in left TLE children, activations were significantly weaker bilaterally in parahippocampal gyri, precentral gyri and cerebellum, in the left fusiform gyrus, left posterior cingulum, but also right hippocampal uncus (p. <0.001) (figure 1). Moreover, in the TLE group, correlations analyses between activations and CRecog showed greater involvement of right hippocampus and right thalamus with performance but no ispsilateral homologous regions (R2>0.9; p. <0.01) (figure 2). Finally, left and right middle orbito-frontal gyri, as left inferior temporal gyrus, right middle temporal gyrus and right middle cingulum were negatively correlated to CRecog (R2>0.9; p. <0.01). Conclusions: Our fMRI findings show bilateral MTL impairment (ie, parahippocampal gyri) in left TLE children during episodic memory encoding. Moreover, correlations analyses showed greater involvement of right hippocampus and right thalamus in efficient encoding suggesting controlateral reorganisation to sustain episodic memory in left temporal lobe epilepsy. Sources: Jambaque et al. Neuropsychologia(2007) in press Martins et al. OHBM, to be published Menon et al. Cogn. Brain. Res (2005) 379-385
Neuroimaging