Abstracts

Rationale:
Disruption of function in brain disease states, including epilepsy, can be demonstrated using functional connectivity MRI (fcMRI), based on task-free, resting state fMRI. In healthy controls, fcMRI reveals Resting State Networks (RSNs) characterized by specific spatial patterns. Key features of connectivity patterns are strong interhemispheric connectivity between symmetric, functionally homologous regions of the brain and intrahemispheric connections among cortical and subcortical regions. Functional MRI, both tasks-based and resting state, is routinely acquired along with other diagnostic studies as a part of presurgical evaluation for drug resistant epilepsy at our institution. We evaluated fcMRI connectivity and fMRI passive sensorimotor tasks in patients with hemimegalencephaly.
Method:
We retrospectively assessed fMRI findings in 9 pediatric (11 days to 17 years) hemimegaloencephaly patients evaluated during a nine-year period. Passive Motor (passive finger movement, 2-4 trials, four blocks, EPIBOLD sequence, TR=2.4s, TE=30 ms, flip angle = 90°, 3mm isotropic voxels) and Resting state (TR=2s, TE=27ms, flip angle = 78°, 4mm isotropic voxels) scans were performed under Propofol anesthesia using Siemens 3T Prisma. Task-based fMRI scans were processed using FEAT, part of FSL software, fcMRI processing was performed using 1000 Functional Connectomes Project scripts (utilized AFNI and FSL software). This study was approved by the Seattle Children’s Hospital Institutional Review Board.
Results:
Fig 1A. shows functional connectivity patterns in a patient with left hemimegalencephaly. Unaffected hemisphere demonstrated unilateral RSNs for Visual (V1, V5), Default Mode (PCC), Language (Broca’s) seeds while affected hemisphere revealed atypical connectivity within hemisphere and lack of connectivity to homologous areas in the opposite hemisphere. In all 9 cases we failed to find strong connectivity between homologous areas in opposite hemispheres (Table 1). Affected hemispheres lacked typical RSNs in all cases and instead demonstrated atypical connectivity patterns. Fig. 1B shows passive motor activation in the same patient that revealed bilateral activation pattern, displaced in the affected hemisphere. Passive motor task reveled variable results among patients: Five cases demonstrated focal SM activation in both hemispheres, 2 cases demonstrated activation in unaffected hemisphere only, and 2 cases had no detectable activation. However, the absence of activation may be inconclusive as it may reflect depth of anesthesia, medication effects, imaging artifacts due to physiologic cardiorespiratory and motion effects etc.
Conclusion:
Analysis in a small group of pediatric patients with hemimegalencephaly indicated that passive motor activation was detected in both affected and unaffected hemisphere in a majority of cases. Functional connectivity analysis revealed absence of typical RSNs patterns in affected hemisphere as well as dramatic reduction in interhemispheric connectivity. Resting state fcMRI findings could be investigated as to utility in predicting functional outcomes with resection.
Funding:
:
Funding:
was provided by Epilepsy Research Fund through the Center for Integrative Brain Research of Seattle Children's Hospital