Functional reorganization after surgical hemispherotomy for intractable epilepsy

Introduction: Epilepsy is a chronic neurological condition characterized by seizures that may lead to transient loss of consciousness (1). Hemispherotomy is a surgical procedure which isolates the affected hemisphere, hence leading to seizure freedom. Despite acceptable neurological outcome, cerebral functional reorganization especially of the isolated hemisphere remains debatable. We aimed at exploring plasticity changes after hemispherotomy. We expected no cortical inter-hemispheric transfer of functional connectivity as a result of the structural disconnection between the two hemispheres. We further hypothesized that plasticity would involve cerebellopontino-thalamocortical routes involving the remaining functional hemisphere. Methods: Patients showed drug-resistant lesional epilepsy associated with Rasmussen encephalitis and hemimegalencephaly. They were operated by a midline vertical hemispherotomy (2): complete callosotomy was performed, allowing access to the lateral ventricles. Perithalamic section of the white matter between the frontal and temporal horn disrupted the internal capsule, mbria, anterior commissure, but left the major intra-hemispheric bundles (superior and inferior longitudinal fasciculi, uncinated fasciculus, cingulum, external capsule) untouched. With this procedure the affected right hemisphere was isolated. MRI was performed 39m after surgery for patient1 (P1) and 31m after surgery for patient 2 (P2). Pseudo-Continuous Arti cial Spin Labelling (ASL) and 300 functional MRI volumes during resting condition were acquired on a GE Sigma 1.5T scanner. Age- and gender-matched control data were used from the National Database for Autism Research (NDAR) (http://ndar.nih.gov). ASL analysis included segmentation, brain extraction, normalization to a 7y-old brain Atlas obtained with Template-OMatic Toolbox, coregistration, realignment, normalization using the deformation elds obtained during the T1 normalization, and smoothing (10 mm isotropic filter).FMRI analysis included slice-time correction, realignment, segmentation of structural data, normalization of functional and structural data into standard MNI space, smoothing (6mm FWHM) and denoising (cCompCor, regression of motion paramaters, temporal filtering 0.008- 0.09Hz). Functional connectivity was performed using patient-specic spherical regions of interest at the cortical subcortical level. Results: In both patients, the disconnected right hemisphere (RH) showed general hypo-perfusion, and the preserved left hemisphere (LH) had hyper-perfusion mainly in motor and temporal areas. Neither patient showed inter-hemispheric transfer of functional connectivity. Thalamo-cortical connectivity was preserved only in the LHs. Intercerebellar connectivity was present but cerebello-cortical crossings were preserved only in the LHs. P2 further showed right-sided ipsilateral cerebello-cortical connectivity, which was mediated by the effect of the vascular system. Both patients' brain showed preserved yet lateralized network-level organization. Conclusions: After hemispherotomy the two hemispheres function in a disconnected manner. The isolated brain is resilient to support intrinsic functional connectivity which is mediated, yet not fully explained, by the effect of vascularization. These data are relevant for a better comprehension of functional organization after neurosurgery and raise inquiries about the cognitive role of the isolated right hemisphere