Development of a Clinically Useful Algorithm for Tractography-Based Ventral Intermediate Nucleus Targeting
Year: 2017
Type: article
Abstract: Essential tremor, characterized by a persistent, bilateral kinetic and postural tremor, is a common neurological disorder with a prevalence of 0.9% in the general population, and up to 4.6% in those over age 65. Deep brain stimulation (DBS) and thalamotomy of the ventral intermediate nucleus (Vim) of the thalamus can successfully control upper limb tremor. Treatment efficacy depends on precise targeting of the Vim, particularly areas with a high degree of connectivity among regions implicated in the “tremor network,” including the ipsilateral motor cortex and contralateral cerebellum. Indirect targeting is a common method used to locate the Vim target that superimposes atlas-derived coordinates onto a patient's unique magnetic resonance imaging (MRI) scan. Unfortunately, this approach does not account for interindividual anatomic variability, and MRI limits visualization of individual thalamic nuclei. Individualized tractography may provide more robust anatomic data for targeting thalamic nuclei. A recent study by Sammartino et al1 presents a novel method for determining the tractography-based Vim target (T-Vim) for ultimate use in the clinical treatment of essential tremor. T-Vim uses the pyramidal tract (PT) and medial lemniscus (ML) as landmarks for the identification of the region of interest (ROI; Figure A). The location of the T-Vim-based ROI was compared to the Vim location identified by conventional indirect targeting methods in 18 patients with essential tremor and 22 healthy controls. For both patients and controls, the T-Vim target was more anterolateral than the conventional coordinate-determined Vim target. Tractography was also utilized to visualize projections from the T-Vim, which reliably showed projections to the primary motor cortex and in a few participants, projections to premotor cortex (Figure B). Tractography, however, was associated with a potential imaging artifact that depicted apparent ipsilateral fibers from the cerebellar hemisphere to the dentate-rubro-thalamic tract (DRT).Figure.: A, The Vim ROI was determined by creating 2 lines (blue lines) at the borders of the ML (blue outline) and PT (red outline) as identified using the tractography technique. One line is drawn at the medial border of the PT, and the other at the anterior border of the ML. At the intersection of these lines a 3-mm ROI representing the Vim is placed (yellow circle). B, Coronal and sagittal views of a reconstruction of the long range projections of the PT (red), ML (blue), and those afferent and efferent to the Vim identified using the tractography method. C, Microelectrode stimulation outcomes with 7 mm trajectory determined by conventional intraoperative targeting methods. DRT and ML are indicated by yellow and blue, respectively. The results of microelectrode stimulation (tremor reduction, tremor arrest, or if there were sensory side effects, shading of the region corresponding to sensation) at coordinates along the trajectory are plotted at the right side of the image. The tip of the electrode (blue) corresponds to the most proximal point in the 7 mm trajectory, 5 mm proximal to the target area, which is at 0. The dashed blue lines represent the trajectory through the DRT up to the distal end of the trajectory 2 mm below the target. D, Microelectrode stimulation outcomes with 6 mm trajectory determined by the novel tractography targeting method. Reprinted from Sammartino et al,1 licensed under CC BY 4.0. TR, tremor reduction; TA, tremor arrest.A post hoc analysis was then performed on a separate cohort of 6 essential tremor or tremor-predominant Parkinson's disease patients who underwent DBS implantation or microelectrode recording-guided Vim thalamotomy. The position of the ML and PT were concordant within <1 mm between tractography and conventional intraoperative localization methods. The Euclidian distance between the original surgical Vim and the T-Vim target was a mean of 1.6 mm (standard deviation, 1.1). In 1 patient, the tremor reduction with microelectrode stimulation along trajectories determined by T-Vim and conventional methods was compared. The trajectory determined by conventional methods had a smaller zone of tremor efficacy (4 mm) compared to the T-Vim trajectory (5 mm) within the DRT. Complete tremor arrest was achieved at the target of the T-Vim trajectory, while sensory side effects were elicited in the target of the conventional trajectory due to off-target sensory thalamus stimulation (Figures C and D). These findings support the use of deterministic tractography that relies on the ML and PT internal landmarks to locate Vim in clinical settings. In particular, this new algorithm is an attractive alternative to the existing probabilistic tractography algorithms that are computationally demanding and less adept for mapping long neural pathways. Although this study identified potential limitations of deterministic tractography, such as imaging artifacts, it may be a convenient and accurate method to target the Vim for therapy. This could have significant implications for surgical outcomes as suboptimal electrode placement is considered a major cause of inadequate or failed tremor reduction. In the future, this algorithm could be used to more precisely locate DBS targets for essential tremor or other surgical indications, and thus facilitate more successful surgical outcomes.
Source: Neurosurgery
Institutions: Johns Hopkins University, Johns Hopkins Medicine
Language: English
Subfield: Neurology
Field: Medicine
Domain: Health Sciences
Open Access status: hybrid
OpenAlex ID: W2609297423