Abstracts

Rationale: fMRI plays a key role in determining language lateralization pre-operatively in epilepsy surgery candidates involving language areas. Unlike primary language areas such as Broca’s or Wernicke’s area, supplementary motor area (SMA) language map is not well established either with fMRI or with direct cortical stimulation in children. Methods: Six patients (2M/4F, mean age 14.3, range 13-16) underwent pre-op fMRI using verb generation paradigm. Following intracranial EEG and cortical language mapping, all six patients had removal of epileptogenic frontal lobe sparing inferior frontal lobe. Post-op language outcome was correlated with the pre-op language fMRI and the extent of SMA resection. Results: All 6 patients had left frontal lobe resection. Four patients had left language dominance while two had right language dominance according to fMRI. All had language SMA activation with verb-generation paradigm in medial frontal area anterior to motor SMA activated by finger tapping. 3 out of 4 patients with dominant frontal lobe resection developed postoperative language deficit with complete recovery within 3 to 6 months. In all 3 patients, the resection margin included language SMA. In 1 patient without language deficit following the dominant frontal lobe resection, the language SMA was not included in the resection margin. Two patients with non-dominant frontal lobe resection had no language deficit. Conclusions: Post-op language deficit following SMA resection was associated with resection of dominant language SMA activation. Pre-operative identification of language SMA in addition to primary language area may help in reducing the risk of post-operative language deficit following frontal lobe epilepsy surgery.