Cerebral Activation during Vagus Nerve Stimulation: A Functional MR Study
Narayanan JT, Watts R, Haddad N, Labar DR, Li PM and Filippi CG;
Commented by , 28 Jan 2003
Background
Vagus nerve stimulation (VNS) is an effective nonpharmacological treatment in patients with medically refractory epilepsy who are not candidates for surgery. The neurophysiological effects of VNS have been described using positron emission tomography (PET), which showed acute and chronic changes in cortical and subcortical cerebral blood flow (CBF).
This included CBF increase bilaterally in the thalamus and hypothalamus and decreases bilaterally in the hippocampus and amygdala. The current study used functional MRI (fMRI) which has the advantage of being non-invasive and avoids the use of radio labeled tracers. Repeated measurements of CBF changes in short time intervals can be recorded offering novel information of brain activation.
Aim
To determine short-term VNS-related changes in CBF and cortical activation by using fMRI.
Methods
The authors studied five patients with medically refractory epilepsy who were undergoing VNS implantation. A routine MRI was performed on a 1,5 T magnet and fMRI was performed using EpiBOLD (echoplanar blood oxygenation level dependent) technique.
All patients had a nonfocal brain MRI. Just before starting the fMRI, the VNS was activated with a current of 0,5-2,0 mA. The VNS was programmed to function for intervals of 30 s, and time event was recorded as the beginning of an “on” cycle. Statistical parametric mapping (SPM) was used for data analysis (p<0,05). Each patient was analyzed individually.
Results
Brain activation was detected in bilateral thalami, ipsilateral more than contralateral, as well as bilateral insular cortices. Activation was also seen in ipsilateral basal ganglia and postcentral gyrus, and inferomedial occipital gyri, ipsilateral more than contralateral.
This pattern of activation was seen in all patients. The most robust activation was seen in the thalami and insular cortices. Areas of significant deactivation were heterogeneously distributed and did coincide between patients.
Discussion
Several neurophysiological studies have emphasized the central role of thalamocortical projections in generalized seizures, as well as in synchronization, propagation and secondary generalization of focal seizures.
PET studies have shown several subcortical areas of activation after VNS, among these bilateral thalami. The current study uses fMRI which has a higher spatial resolution than PET and very rapid temporal resolution of 1 ms. In this way the authors were able to study the very initial VNS related changes in brain activation.
Results showed that the most robust BOLD response was found in bilateral thalami and insular cortex. This, together with previous studies, suggests that at least part of the VNS effect may be due to thalamic activation. The study confirms that thalamocortical relay neurons are involved in the modulation of epileptic excitability of cortex.
Furthermore, the study provides significant information regarding new possible treatment strategies in medically refractory epilepsy. Among these are direct electric stimulation of deep brain structures, which has been successfully implemented in Parkinson’s disease and tremor, and might also be relevant in epilepsy.