Deep brain stimulation for patients with movement disorders such as Parkinson's disease is now FDA approved treatment. While motor symptoms can be markedly improved, a drawback to this method is the production of non-motor side effects whose symptoms can overwhelm any benefits. The spatial sensitivity for optimal location of electrode contacts is within millimeters, and typically requires surgical implantation to be performed with the patient awake. However, many patients are adverse to awake neurosurgery, and there is a need for methods permitting accurate implantation with patients under anesthesia. Although some centers can perform such implantation based solely on anatomy, there is no information about functional location.

We describe a new method that addresses the need to functional information about DBS location in patients under general anesthesia, by using intra-operative fMRI while the DBS electrodes are electrically stimulated. This method is similar to one recently applied to epilepsy patients^1,2, and extensive dedicated safety tests were performed beforehand. Another group has recently published similar results using different parameters³.

Patient were scanned in a IMRIS intra-operative MRI at 1.5T, during DBS implantation while under general anesthesia. After electrode insertion using anatomic guidance, the MRI was placed over the patient and the electrodes connected to an external stimulator located in the control room, via a long shielded cable. After acquisition of an anatomic T1 sequence, a BOLD sensitive EPI sequence was acquired during stimulation of the electrodes in a block design of 30 seconds on/off. Stimulation parameters were 6 volts across bipolar contacts, at 130 Hz. A total of up to 4-5 DBS-fMRI sequences were obtained during the imaging session, each with variations of stimulation parameters such as voltages, contacts, and duration. Analysis was performed using AFNI.A total of 6 patients have been studied and there have been no adverse reactions. A variety of stimulation parameters were tested to explore parameter space, generally showing robust BOLD activation with voltages greater than 4 volts. Figure 1 shows maps of BOLD activation resulting from stimulation of electrodes in the thalamus (ViM) and subthalamic nucleus (STN) in two patients with movement disorders. Both positive and negative activation are demonstrated, which can be both proximal and distal to the electrode contacts, and the patterns are distinct. Figure 2 shows similar maps in the same patient, demonstrating the spatial sensitivity of the BOLD maps depending on a slight change of the stimulation contacts. Figure 3 shows the BOLD maps for stimulation of the STN in two patients with a different clinical response to DBS stimulation (as determined several months after implantation): the patient with significant clinical response clearly shows a wider extension of BOLD activation compared to the patient with no significant clinical response. These results shows promise for using this technique to predict the clinical response after implantation, and is also sensitive to slight changes in stimulation location.We present initial results from simultaneous stimulation of DBS electrodes and fMRI, performed in a Parkinson's patients at 1.5T while under general anesthesia. Robust BOLD activation can be easily elicited at voltages greater than 4V, whose patterns are both proximal and distal, with high spatial sensitivity, and whose patterns reflect clinical efficacy. This technique offers a possible alternative for patients wishing DBS implantation while under general anesthesia, in which the functional location of electrodes is desired to maximize clinical response and minimize side effects.