Vagus nerve stimulation in refractory epilepsy: SPECT activation study

Left-sided vagus nerve stimulation (VNS) is an efficacious treatment for pa tients with refractory epilepsy. The exact mechanism of action remains to b e elucidated. This study investigated the acute effects of initial VNS in p atients with refractory complex partial epilepsy with or without secondary generalization (complex partial seizures [CPS] +/- SG) by means of a perfus ion activation study with SPECT. Methods: Twelve patients (mean age, 32.2 /- 10.2 y; age range, 12-47 y) with a mean duration of CPS +/- SG of 19.8 /- 10.0 y (range, 5-33 y) received VNS. All patients were considered unsuit able candidates for resective surgery because of nonlocalizing findings on presurgical evaluation. VNS efficacy was evaluated for patients with at lea st 4-mo follow-up. VNS-induced regional cerebral blood flow alterations wer e studied by a Tc-99m-ethyl cysteinate dimer activation study with a single -day split-dose protocol before and immediately after an initial stimulatio n. Images were acquired on a triple-head camera with fanbeam collimators. A fter coregistration to a standardized template, both a semiquantitative ana lysis using predefined volumes of interest and a voxel-by-voxel analysis of the intrasubject activation (statistical parametric mapping) were performe d. Results: Seizure-frequency changes ranged from 100% decrease to 0% after VNS. The semiquantitative analysis revealed a consistent decrease of activ ity in the left thalamus (ratio stimulator on/off = 0.94 +/- 0.04; P = 0.00 5). These results were concordant with the voxel-by-voxel analysis in which a significant deactivation in the left thalamus was found with spread to t he ipsilateral hippocampus. There was no statistically significant correlat ion between initial VNS-induced thalamic hypoperfusion and seizure reductio n at maximum follow-up. Conclusion: Our findings are consistent with the hy pothesis that acute VNS reduces seizure onset or propagation through inhibi tion of the thalamic relay center. Differences with limited (H2O)-O-15 PET data may be associated with temporal effects caused by a stimulation-induce d local hemodynamic response and need further investigation. SPECT allows s tudy of cerebral physiopathologic effects of vagus nerve electrostimulation in complex partial epilepsy.