Ultra-high resolution 7T MRI mapping of basal ganglia connectivity on an individual patient basis for Paediatric Deep Brain Stimulation (DBS)

Lead Research Organisation: King's College London
Department Name: Imaging & Biomedical Engineering


Aims of the Project:
This project will define clinical procedures to allow 7T imaging in children suffering from dystonia and collect preliminary data to address the question: 'does 7T MRI improve the localisation of basal ganglia nuclei and the of modelling the connectivity profile of implanted DBS electrodes, compared to standard-of-care 3T MRI'.

Dystonia is a painful disabling condition, characterised by uncontrollable muscle contractions. The commonest cause of dystonia in childhood is CP, affecting 2-3:1000 live births in the UK. Medications used to reduce dystonia are often ineffective, and significant side effects limit use. Symptoms of dystonia do not spontaneously remit for affected individuals, and so a lack of effective intervention leads to long term painful disability. DBS may be used to reduce dystonia in children with CP. Electrodes are implanted into the deep substance of the brain, connected to a neurostimulator placed under the skin (similar to a pace-maker). DBS delivers a continuous electrical stimulus, which effects the brain at the site of stimulation and beyond. Following DBS, improvements in dystonia seen in children with CP is variable. The basal ganglia nuclei targeted for electrode insertion are very small, with boundaries that can be difficult to see on 3T MRI. These nuclei have functional subdivisions, the motor area being the target for electrode insertion. Advanced neuroimaging techniques can allow us to examine how regions of the brain are connected, through "structural" and "functional" connectivity, which can allow us to identify the motor area. It is also possible to measure which areas of the brain are "connected" to DBS electrodes after they are implanted, helping us to understand how DBS works to improve dystonia.
The Evelina London Children's Hospital runs the leading DBS service for Children in the UK, and has pioneered novel treatments for children with dystonia.

Ultra-high field MRI
7T MRI scanners provide much higher neuroanatomical definition, and will define the target nuclei and their connectivity profiles with greater precision. However 7T imaging is technically challenging and detailed preliminary work is needed to make clinical imaging of children possible and to assess its value in clinical practice. We have recently established the Wellcome Trust London Ultra-high Field MRI Centre at the St Thomas' Campus of King's College London (PI Hajnal), with a specific paediatric research theme (PI Edwards). Our group is funded to develop appropriate sequences for imaging infants and children (PI Carmichael). We have an established track record of MR research in children, including the ongoing 15 million Euro Developing Human Connectome Project, funded by the European Research Council. This combination of imaging facilities and researchers with the DBS service offers a unique opportunity to make significant progress.

What we plan to do:
We will recruit 10 children with CP due to undergo DBS surgery and obtain MRI images from both 3T and 7T MRI scanners. We will use these images to build and compare 3-dimensional models of the basal ganglia, and the white matter pathways that connects them. We will compare how these images show us the connections from implanted electrodes to different regions of the brain, to help us understand how DBS effects the brain. Following surgery we will also perform a technique called DBS Evoked Potential (DBSEP). This technique will allow us to measure how accurately the connections we measure from MRI scans reproduce the actual wiring of the brain.

How will this help children:
A better understanding of how the basal ganglia are connected will help us to understand why children with CP experience dystonia. A better understanding of how DBS works in children with CP will help us better select children for surgery, and to better plan the placement of DBS electrodes to make sure each child experiences the best outcome possible.

Technical Summary

Dystonia is a painful disabling neurological condition characterised by involuntary muscle contractions. The commonest cause of dystonia in childhood is Cerebral Palsy (CP), affecting 2-3:1000 live births in the UK. Pharmacological treatments for dystonia in CP are often ineffective, and frequently produce significant side effects. This has led to increasing interest in Deep Brain Stimulation (DBS), which involves the neurosurgical placement of electrodes into one of the nuclei of the basal ganglia. Generally positive results are seen following DBS in children with CP, but the degree of improvement varies between children for reasons which are currently unclear. The success of DBS depends upon the accurate placement of stimulating electrodes. In children with CP, the boundaries of the target nuclei for DBS may be difficult to delineate on clinically available MRI scanners (1.5/3T field strength). The basal ganglia nuclei are known to be internally sub-divided into functional territories, the motor region being the target for DBS. Whilst the motor region cannot be visualised on structural MRI sequences, advanced imaging techniques can be used to define the region on the basis of measures of functional connectivity (resting state fMRI analysis) or structural connectivity (tractography analysis). Connectivity analysis can also be used to inform us of the distributed brain network through which implanted electrodes act, improving our understanding of the mechanisms of action of DBS. We hypothesise that the improved signal to noise ratio offered by ultra-high resolution 7T MRI will enable us to more accurately image the basal ganglia and its connections, and to map the brain network accessed by implanted DBS electrodes. We will perform a comparative study for a cohort of children undergoing DBS surgery, obtaining scans on both 3T and 7T MRI scanners. Following surgery, DBS Evoked Potentials will be used to help us begin to validate the results of connectivity analysis.

Planned Impact

The research produced by this project will be of interest to the clinical, neurosurgical, neuroimaging and research community. It will also be of direct interest and benefit to children with Cerebral Palsy (CP) and their parents and carers.

The clinical and neurosurgical community will find early, direct relevance of the comparison of 3T and 7T structural imaging sequences for the delineation of the boundaries of the basal ganglia. This will help inform the adoption of ultra high resolution MRI into routine clinical practise, and, in particular, the use of 7T MRI for direct neurosurgical planning. This project will provide a platform for future work which will enable the acquisition of a catalogue of ultra-high resolution MRI images of children and young people with motor disorders (beyond the immediate scope of this project). For centres involved in Deep Brain Stimulation (DBS) and other basal ganglia functional procedures without access to ultra high resolution MRI, the subcortical masks and template images evolving from this future work will have a direct application for neurosurgical planning.

Information provided on the brain network accessed by DBS will also be of interest to the clinical community, and the basic science and research community engaged in the study of movement disorder. This will provide information on the mechanism of action of DBS, and a potential avenue for improving targeting for electrode insertion.

The comparison of the performance 3T and 7T imaging in terms definition of the basal ganglia boundaries, parcellation of basal ganglia nuclei will be of direct interest to the neuroimaging community. This will help to define the advantages, and potential limitations of ultra high resolution MRI. Understanding where 3T MRI imaging offers equal benefit to 7T MRI imaging is of equal importance as determining where additional benefit is offered at ultra high resolution. Furthermore, the opportunity to correlate structural and functional connectivity profiles generated noninvasively through the measurement of DBS Evoked Potentials following surgery will provide an invaluable opportunity for the neuroimaging community. Neurosurgical procedures such as DBS provide a modality for vivo validation of these non-invasive techniques otherwise difficult to achieve.

CP effects 2-3:1000 live births in the UK. The effective management of dystonia represents a significant unmet need for patients with CP and their parents and carers. DBS is not a trivial intervention, carrying a 1% risk of mortality, 5% chance of peri-operative seizure and a 7-10% chance of infection of implanted hardware. Furthermore the procedure carries a cost of >£20,000 to the NHS for each child operated upon, as well as the cost for families and carers attend frequent appointments for assessment, time for recovery peri-operatively and the commitment to long term follow up. Improved neurosurgical targeting and a better understanding of the mechanism of DBS may help to optimise benefits seen for the individual patient, and so would carry a strong incentive for uptake to practitioners and service delivery managers.


10 25 50