Abstracts

To determine which factors are associated with functional outcome and seizure reduction in patients who underwent resective surgery or multiple subpial transections (MST) in the Rolandic area.
All patients who underwent Rolandic cortical resection or MST from 1979 to 2003 at the London Health Sciences Centre were identified. Patients were included if they had seizures originating in the Rolandic area recorded with subdural electrodes with adequate coverage, or if they had scalp recorded seizures and a discrete epileptogenic lesion on MRI in the Rolandic area. All patients had electrocorticography (ECoG) at the time of surgery. Data collected include preoperative and postoperative neurological deficits, MRI findings, interictal and ictal scalp EEG, interictal and ictal subdural data, ECoG findings, type and extent of surgery, neuropathologic diagnosis, and follow-up seizure frequency.
58 patients were identified, 33 men and 24 women. Etiologies included tumour (30), dysplasia (5), stroke (3), cavernoma (3), AVM (2), Rasmussen encephalitis (3), trauma (1), infection (1), cerebral palsy (1), porencephalic cyst (1), meningioangiomatosis (1), mitochondrial cytopathy (1), and undetermined (6). Patients were taking an average of 1.8 antiepileptic drugs (AED), range 0-4. Scalp EEG telemetry was obtained in 49 of the 58 patients. Interictal spikes were detected from the central region in only 14 (28.6%), were multifocal in 9 (18.4%), and there were no interictal spikes in 12 (24.5%). On scalp EEG focally-originating seizures arose from the central region exclusively in 13 (26.5%) patients, independently from the central region and other loci in 3 (6%), from other loci but not central in 6 (12%), multifocal onset in 2 (4%), ambiguous or diffuse in 10 (20.4%), and no seizures were captured in 17 (34.7%). Neuroimaging showed a focal lesion involving or abutting the Rolandic area in most patients. MRI was normal in 3, showed multifocal lesions in 4, and a diffuse hemispheric abnormality in 1. In patients from whom follow-up information was available (51 of 58), worthwhile improvement in seizures (Engel classes 1-3) occurred in 38 (74.5%), but only 17 (33%) were completely free of disabling seizures. Better outcomes were more common when the entire epileptogenic MRI lesion could be removed. New deficits were seen in 50% of patients.
Worthwhile seizure improvements can be achieved through well-planned resection and MST in the Rolandic cortex. New deficits can be expected in 50% of patients, and can be moderate to severe in half to two thirds of these. Scalp EEG provided little useful information to guide surgery. Further studies are required to determine whether new techniques, such as functional neuroimaging and improved functional neurophysiology, may permit better delineation between eloquent and epileptogenic cortex.