Neurosurgery
Clinical fMRI
Neurosurgery is the one field of medicine that stands to benefit most from fMRI-based function localization. Surgical removal of disruptive brain pathology, such as tumors, requires knowledge of the specific location of eloquent regions as well as their connections (fibers). The most reliable method of obtaining this is electrical stimulation of cortex (electrocortical stimulation, ECS) and fiber tracts during surgery, which often requires patients to be awake. fMRI can be used to identify cortical foci that mediate a particular function such as hand/arm movement, language production or working memory. fMRI is fundamentally different from ECS so function localization differs, requiring a level of human interpretation of fMRI activity maps. At the UMC, fMRI is frequently conducted for purely clinical reasons. Especially in the case of low-grad gliomas, which are brain tumors that are not (yet) malicious and are not (yet) accompanied by functional deficits, it is of essence to avoid loss of functions during tumor removal, but also to remove as much as possible to obtain a better prognosis. At present, fMRI is used to decide whether or not surgery is feasible (function within or very adjacent to the tumor reduces chances of successful removal), and as a dependable indicator of regions that need to be stimulated for further verification. Starting in 2010 we are exploring the potential of high-field fMRI (at 7 Tesla) for better dientification of function-,critical regions. At this fieldstrength the conrtibution of larger bloodvessels (which contaminate the fMRI activity maps) is strongly reduced.
We also acquire 32-direction DTI images for a study on accuracy of fMRI-informed fiber tracking (PhD student Gert Kristo). This study is conducted in collaboration with the University of Tilburg (Bea de Gelder) and Saint Elizabeth Hospital Tilburg (Geertjan Rutten). Neurostimulation of fiber tracts during surgery is registered (coordinates of positive and negative sites) and compared to DTI-based fiber tracts obtained presurgically. Fiber tracts are not revealed to the surgeon. One of the major challenges in this study is to correct for deformation of the brain due to tissue removal and CSF displacement ('brain shift').
Another important endeavor is the Braincarta project. Many clinical centers are not able to organize the manpower and skills to implement fMRI paradigms and analyze the data. To make fMRI more accessible, we have developed a set of tasks that can be administered in a simple fashion, with instructions for EPI parameters and data analysis. Goal is to achieve a standard set of tasks and acquisition approach across academic and non-academic neurosurgery centers with which multicenter trials can be conducted. A network of centers has been formed in recent years and is now formalized as the ELGGN (European LowGrade Glioma Network) initiated in 2006 by Hugues Duffau for the purpose of standardizing surgical and non-surgical treatment and research methods. A first generation standard fMRI task set has now been applied by 6 centers, and we are currently evaluating differences in results, and ways of reducing differences by optimizing analysis techniques.
