Abstracts

Rationale: Surgery in the rolandic cortex is challenging due to the potential new deficits that might arise after cortical resection of this eloquent area. On the other hand, adequate mapping might lead to seizure freedom with minimal or no deficits.Methods: Nineteen patients submitted to rolandic cortex resection were studied. All ;atients presented with lateralized simple partiaL (motor or somato-sensitive) seizures. MRI was normal in 9, and in 10 patients MRI showed a discrete lesion in this area (4 FCD type II, 3 cavernoma, 3 tumors). Patients with normal MRI were submitted to invasive monitoring using subdural grids. Patients with discrete MRI-visible lesions were submitted to surgery under general anesthesia and intraoperative motor strip mapping.Results: Twelve patients had foci in the dominant hemisphere. Five of the nine patients with normal MRI become seizure free; nine of the 10 patients with discrete lesions did so. Seven patients with normal MRI needed additional cortical resection outside the rolandic area; only one patient with a discrete lesion did so. Preservation of rolandic vasculature was achieved in all patients. Resection of fibers below the depth of the central sulcus was avoided in patients submitted to surgery within the dominant hemisphere. Twelve patients had resection of the motor cortex. Patients who receive resection restricted to the post-central gyrus presented with numbness of the face for 1-3 days, and numbness of the arm for 2-3 weeks. In those who received pre-central gyrus resection, facial asymmetry lasted for 2-3 weeks in all except one patients, who persisted with visible asymmetry. No resection was ;performed in the hand motor area. Three patients presented with transient (3-4 weeks) conduction dysphasia.Conclusions: Surgery can be safely performed within the rolandic cortex as far as a strict microsurgical technique is used. When needed, speech area mapping is better performed using implanted subdural grids. No surgery was performed in awaken patients.