Abstracts

Rationale: Although mapping of sensorimotor cortex in patients with perirolandic lesions undergoing neurosurgery is critical, successful sensorimotor cortex mapping remains a significant challenge for patients who cannot be compliant during functional MRI or magnetoencephalography (MEG), including young children and patients with developmental delay. In such cases, as a surrogate to motor mapping, somatosensory mapping in MEG is performed under sedation. However, transcranial magnetic stimulation (TMS), allows for direct motor mapping even in very young children without requiring sedation. We examined whether MEG under sedation or TMS was more successful at localizing primary sensory (S1) and motor (M1) areas respectively in a large pediatric cohort.  Methods: We identified 59 patients with epilepsy or brain tumor (mean age 5.9+-3.8 years, range 0.8 to 16 years, 30 females) who underwent sensorimotor localization with MEG under sedation (S1) and TMS (M1) at our institution. For MEG, the sedation was induced by either propofol (n=25) or propofol and dexamedetomidine (n=34) and followed in all patients by a continuous infusion of propofol (5-10mg/kg/hr). Localization of S1 was attempted with MEG during pneumatic tactile stimulation applied to the index finger or thumb, localizing the source of activity associated with the peak of the early or middle latency response using the single equivalent current dipole analysis. The TMS M1-hand mapping was performed under MRI guidance while recording electromyography from small muscles of the hand and forearm muscles. The cortical locations of motor evoked potentials and cortical silent periods elicited by TMS were recorded.  Results: The MEG data in 8 patients was not usable due to artifacts from VNS, dental caps or VP shunt. Of the 51 patients, S1 was localized in at least one hemisphere in only 6 patients, with 2 patients (4.4 and 13 years) having both left and right cortices mapped and 4 patients (3.7 - 6.7 years) having successful localization in one hemisphere. TMS localized M1 in both hemispheres in 45 patients (including 8 patients with metal) and interhemispheric motor reorganization was identified in 7 patients. In 5 patients, M1 was found in only one hemisphere and TMS failed to identify any motor area in 2 patients (0.84 and 2.2 years-old). In particular, TMS was successful in 18 (including 4 children with motor reorganization) of the 20 children younger than 3.7 years in this cohort in all of whom MEG was unsuccessful. The success rate of MEG was similar for the two sedation protocols and the lesion location did not influence MEG or TMS outcomes.  Conclusions: TMS had a very high success rate in this pediatric patient cohort and should be considered as a reliable test for identifying motor areas in patients undergoing presurgical evaluation. Somatosensory mapping using pneumatic tactile stimulation in MEG under sedation appears to have a lower yield in localizing S1, in part due to low intensity of the stimulus as well as depth of sedation decreasing the responsiveness to tactile stimulation. Median nerve stimulation may be better suited to localize S1 in MEG under sedation. Additionally, modifications to sedation protocols should be further explored to increase success in S1 mapping in MEG. However, in patients in whom MEG fails to successfully map S1 (in particular children under 4 years of age), or is not feasible (metal implants), TMS offers a safe, well tolerated and reliable alternative method for mapping motor cortex towards successful surgical planning and minimizing post-operative deficits.  Funding: No funding