Abstracts

Rationale: We propose a novel method for mapping brain connectivity within the epileptogenic zone (EZ) by combining intraoperative fMRI with simultaneous SEEG depth electrode stimulation. We hypothesize that SEEG-fMRI can more accurately localize and anatomically define the brain areas responsible for the generation and early propagation of recurrent seizures. Methods: Four consecutive SEEG implanted patients with non-lesional MRIs and difficult to localize medically intractable focal epilepsy underwent extra-operative cortico-cortico evoked potential stimulation (CCEPS), with cathode stimulation in depth electrode contacts located in the hypothetical EZ. Subsequently, patients underwent similar SEEG stimulation (stimulation parameters: biphasic pulses at 20Hz, pulse width of 0.3ms, up to 8mA in intensity) during fMRI sequences. The obtained BOLD activation maps were compared with the anatomical location of the SEEG-defined EZ, the electrophysiological connectivity mapped by CCEPS and with post-operative seizure outcome at 6 months.Results: The fMRI images obtained during the stimulation of SEEG-defined epileptic areas resulted in strong BOLD activated areas (locally and distally to the stimulation site) in 3 out of 4 patients. Resection of BOLD activated areas resulted in seizure freedom in all 3 patients. In the remaining patient with no BOLD activation, seizures persisted after surgery. Stimulation of non-epileptic areas (controls) resulted in weak or no BOLD activation in 3 out of 4 patients and strong BOLD activation in one patient. BOLD images anatomically correlated with areas with maximal connectivity obtained from the CCEPS studies. Pathology from BOLD areas revealed focal cortical dysplasia in all the patients.Conclusions: The preliminary data suggests that SEEG-fMRI images may reveal cortical and subcortical areas with high connectivity within the epileptic network and, consequently, more accurately localize and anatomically define brain areas responsible for the generation and early propagation of recurrent seizures. Nevertheless, further long-term post operative seizure outcome studies using the proposed method are necessary to validate this conclusion.