By Sean Nagel, MD; Caio Matias, MD; Michael Phillips, MD; Stephen E. Jones, MD, PhD; and Andre Machado, MD, PhD
Cleveland Clinic is a non-profit academic medical center. Advertising on our site helps support our mission. We do not endorse non-Cleveland Clinic products or services Policy
Real-time or near-real-time intraoperative MRI is poised to revolutionize deep brain stimulation (DBS) surgery by simplifying the operation and enhancing the patient experience.
Intraoperative MRI is now routinely offered and used as an option to guide electrode placement during DBS surgery in patients at Cleveland Clinic. Results of a preliminary evaluation of DBS leads placed in our intraoperative MRI suite indicate that actual lead location closely matched intended placement targets. This article reviews the rationale for the use of intraoperative MRI in DBS as well as the next steps in our assessment of this exciting treatment option.
Microelectrode recording and the awake patient
DBS is a well-established treatment for Parkinson disease and essential tremor and is also FDA-approved under a humanitarian device exemption to treat refractory obsessive-compulsive disorder and dystonia.
Microelectrode recording (MER) has been routinely used during stereotactic procedures for several decades to improve the localization of subcortical structures for DBS or for ablative treatments such as thalamotomy or pallidotomy. This standard technique has been refined gradually to improve the safety profile and increase the accuracy and precision of functional neurosurgical procedures for movement disorders and other applications. MER is typically carried out in the awake patient (Figure 1, left), along with evoked potentials. The recordings are then compared to surgical atlases and the patient’s anatomy to infer electrode location.
In contrast, intraoperative MRI improves accuracy of lead location by assessing the relationship with anatomic landmarks instead of the physiological recordings obtained by MER. An important advantage of intraoperative MRI is that it allows the surgery to be done under general anesthesia, without any awake physiological measurements.
Building the intraoperative MRI suite
Cleveland Clinic constructed its intraoperative MRI suite with a special infrastructure consisting of two adjacent rooms — a leading-edge operating room and an MRI suite (Figure 1, right) — separated by mechanical doors. At any time during the operation, the surgeon can request an MRI to assess the location of DBS leads and make corrections as necessary. The mechanical doors open and the MRI machine moves on ceiling-mounted rails into the surgical field. Once images are acquired, they can be overlaid on the initial surgical plan to assess any deviation between the actual location of DBS leads and the intended plan.
Figure 1. Options for electrode placement during DBS surgery are evolving from traditional microelectrode recording in the awake patient (left) to intraoperative MRI in the patient under general anesthesia (right).
The likeliest candidates
Patients who are offered or opt for asleep DBS with intraoperative MRI may have incapacitating claustrophobia, heightened anxiety, large-amplitude tremor or another medical comorbidity that increases the risk for awake surgery, such as respiratory disease, airway compromise or increased risk of seizures. Additional patients may simply elect to have surgery under general anesthesia because they are not comfortable with the idea of being awake for part of the surgery. This option is far more tolerable from a patient comfort standpoint, and we anticipate that most patients and surgeonguide the workflow
Although the workflow from one intraoperative MRI case to the next may vary based on multiple patient factors, the principles are unchanging. For many patients, the workflow proceeds as follows:
- A head-frame or frameless system is fitted to the patient’s head, and a CT or MRI is completed.
- The patient is secured to the fixed intraoperative MRI table. Using the preplanned trajectory based on the preoperative MRI coregistered with the stereotactic images, the first lead is implanted.
- The doors are opened, the MRI machine is brought over the patient and a volumetric MRI (1.5T) is completed.
- These images are merged with the preoperative stereotactic CT using the planning software, and the actual trajectory is compared with the intended trajectory (Figure 2).
- After adjustments are made to account for brain shift, a second lead is implanted, followed by repeat intraoperative MRI imaging.
Figure 2. Axial (left) and coronal (right) T1-weighted intraoperative MRIs following implantation of a unilateral right-sided DBS electrode into the globus pallidus internus with preplanned trajectory overlay (dashed blue line, which shows the entirety of the trajectory, including portions lying outside the planes). The actual position of the electrode crossing into these two planes is shown by the hypointense (dark) focus denoted by the arrows. Note how the actual lead tracks with the intended trajectory in both planes. There was no appreciable error in targeting, and adjusting the target for the contralateral side was not necessary.
Promising preliminary outcomes assessment
Our preliminary results, described below, indicate that outcomes are similar with preserved safety. Long-term follow-up will be necessary before definitive conclusions can be drawn.
We recently conducted an assessment of lead location in 24 patients who underwent DBS lead implantation in the intraoperative MRI suite. We used stereotactic planning software (iPlan® 3.0, Brainlab, Munich, Germany) to fuse preoperative MRI and CT scans with intraoperative MRI images. All images were normalized to two midline anatomic landmarks, the anterior commissure and the posterior commissure, which allowed a comparison between the intended preoperative x, y and z coordinates and the actual intraoperative electrode coordinates. In the stereotactic space, the x-axis represents the medial-lateral position, the y-axis represents the anterior-posterior position and the z-axis represents the superior-inferior position.
The average differences between the intended target and the actual electrode tip were as follows:
- 1.1 ± 0.9 mm for the x-axis
- 0.7 ± 0.04 mm for the y-axis
- 0.8 ± 0.9 mm for the z-axis
These differences are within the expected error range for stereotactic neurosurgical operations. Furthermore, since the electrode used for DBS has a diameter of 1.27 mm, errors of approximately 1 to 2 mm are considered acceptable.
Next steps and implications
The next step will be to compare motor and nonmotor outcomes between patients with Parkinson disease undergoing DBS with intraoperative MRI vs. MER.
We anticipate that the role of intraoperative MRI for DBS will expand over the next several years, for a couple of reasons. The number of potential candidates for DBS is expected to increase, not only as a result of new indications but also because those previously not offered awake DBS may now be candidates for this new method using intraoperative MRI under general anesthesia. In addition, real-time lead evaluation will likely enhance patient safety.
Dr. Nagel is a staff neurosurgeon in the Center for Neurological Restoration in Cleveland Clinic’s Neurological Institute.
Dr. Matias was a research fellow in the Center for Neurological Restoration when this article was written.
Dr. Phillips is a staff physician in the Center for Neuroimaging and the Department of Neurosciences as well as Vice Chair of Research and Academic Affairs in Cleveland Clinic’s Imaging Institute.
Dr. Jones is a staff physician in the Center for Neuroimaging.
Dr. Machado is Director of the Center for Neurological Restoration.