gms | German Medical Science

Objective: Preoperative visualization of subcortical language pathways by means of DTI-FT is evolving as an important tool for surgical planning and decision-making in patients with language-eloquent brain tumors. Navigated nTMS cortical language mapping is an additional non-invasive method to further extend the basis for treatment choices. Efforts to incorporate nTMS data into DTI-FT are promising, but lack of standardized protocols make it hard to assess clinical usefulness. This work was conceived to evaluate the clinical usefulness of three distinct approaches to language pathway tractography.

Methods: 30 patients with left-hemispheric perisylvian lesions received preoperative nTMS language mapping and DTI. Language Tractography was conducted as follows.

• 1a: Tractography of five major language-relevant fascicles was carried out based on verified anatomical landmarks.
• 1b: Tracts with minimum tumor-distances greater than 20mm were hidden in order to reduce the representations complexity.
• 2a: All language-positive spots identified in nTMS were defined as a single ROI for DTI-FT.
• 2b: Streamlines located in the anatomical pathways of the five major language-relevant fascicles were isolated in order to reduce the representations complexity.
• 3: Tractograms resulting from Method 1b as well as individual language-positive spots identified by nTMS located below the projected craniotomy were defined as ROIs.

The results of each approach were rated by neurosurgeons from different hospitals in respect to usefulness for surgical planning and risk assessment, overall information content and perceived accuracy of tracts. Additionally, objective measures such as success to visualize specific tracts and tract volume comparisons were studied.

Results: Results reveal high usefulness ratings for anatomy-based approaches, as well as for the combined nTMS-enhanced approach. Post-processed representations had a more balanced perceived information content than unprocessed tractographies. The order of overall preferred methods was 1b>3>1a>2b>2a.

Conclusion: Results suggest that DTI-FT can be enhanced by nTMS data in a clinically meaningful way. Yet, relying solely on functional data for DTI-FT seeding cannot be recommended, as usefulness ratings were higher for all alternative approaches. Future work should explore the effect of different approaches on clinical outcome in order to guide the process of establishing meaningful standards.