Abstracts

Rationale: Deep brain stimulation has been used in an increasing frequency for treatment of selected patients with refractory epilepsy who were not adequate candidates for conventional resective surgery. Hippocampal DBS (Hip-DBS) has been offered in the treatment of refractory temporal lobe epilepsy. Although there were reports showing that Hip-DBS was indeed efficacious, little is known about the mechanism of action of Hip-DBS. We studied patients submitted to Hip-DBS in the pre- and post-electrode activation periods. Methods: Three adult patients with refractory temporal lobe epilepsy were studied. Patient 1 had refractory temporal lobe epilepsy (r-TLE) associated with left mesial temporal sclerosis (MTS) and was submitted to left Hip-DBS using a Soletra device. Patient 2 had r-TLE associated with post-herpetic encephalitis bilateral temporal atrophy. His video-EEG showed bitemporal interictal discharged, and ictal onset prevailing on the left side. He was submitted to bilateral Hip-DBS using a Kinetra device, although only left Hip-DBS was activated during this study. Patient 3 had r-TLE and normal MRI. His video-EEG showed bitemporal interictal discharged, and ictal onset prevailing on the left side. He was submitted to bilateral Hip-DBS using a Kinetra device, although only left Hip-DBS was activated during this study. All patients were submitted to pre- and (3 months) post-activation PET studies. PET findings were blindly reported. All patients had their electrode's location confirmed by post-operative CT scanning. Results: Seizure frequency was reduced in all patients: in Patient 1 from 6 to 0.5/month; in Patient 2 from 16 to 4 / month and in Patient 3 from 8 to 1 / month. Final parameters were 2.7V, 1.0V and 3.0V for patients 1, 2 and 3, respectively (300usec, 130Hz, continuous stimulation). In Patient 1, post-activation PET showed marked additional hypometabolism in both temporal lobes, mesial frontal cortices and thalamus. In Patient 2, post-activation PET showed additional hypometabolism within the stimulated temporal lobe. In Patient 3, post-activation PET was similar to the pre-activation one. Conclusions: This is very likely the first description of PET findings after Hip-DBS. Although the number of patients was very small and the etiology was heterogeneous, this study showed that Hip-DBS was able to generate additional hypometabolism within the stimulated temporal lobe and Papez circuit in some patients. At this point, we were not able to correlate these findings with clinical outcome regarding seizures, electrode positioning or stimuli intensity.