Presurgical motor fMRI in a 3-year-old child

Objectives: Functional magnetic resonance imaging (fMRI) is increasingly used for the non-invasive diagnostic work-up of patients prior to epilepsy surgery. In contrast to studies on visual or other sensory representations, the investigation of motor tasks requires active cooperation of the subject. This limits the application of this technique for pediatric patients. Here, we report on methods and results of a successful fMRI mapping of the hand motor representation in a 3-year-old boy with focal cortical dysplasia in the right central (Rolandic) cortex.

Material and Methods: In order to map the cortical activity during movements of his left hand, we made the boy squeeze a squeaking toy animal repetitively (activation condition); during the control condition, his mother (standing next to the child) took the animal and squeezed it in a similar manner, thus producing the same acoustic stimulation as in the activation condition. This “game“ of alternating “squeezing and squeaking“ allowed an fMRI investigation of active hand movements in a block design, with 8 activation periods and 8 control periods (30 seconds per period). During one such session (total 8 minutes), we acquired fMRI scans (EchoPlanarImaging, TE=50 ms, voxel size 3×3 x 4 mm3) every 6 seconds, with an acquisition time of 2.4 seconds, so that the scanning noise ceased for 3.6 seconds after each scan. This pause could be used to communicate with the child. Additional technical details: Siemens AVANTO, prospective motion correction (PACE), processing with SPM2, p<0.05 FWE.

Results: Two identical sessions reproducibly demonstrated a strong activation in the right Rolandic area, mostly immediately posterior to the dysplasia. The boy subsequently underwent invasive diagnostics via subdural grid electrodes; the results are not available yet.

Conclusions: Active motor fMRI is possible already at pre-school age when adapted protocols are applied.