gms | German Medical Science

Objective: So far, neurosurgically applied fiber tractography (FT) based on Diffusion Tensor Imaging (DTI) is well established. However, multiple representation models of High Angular Resolution Diffusion Imaging (HARDI) have become a focus of interest, improving the reconstruction of fiber bundles in brain areas with disturbed diffusion properties or crossing fibers. So far, High Resolution Fiber Tractography (HRFT) required elaborate image computing, which limited the clinical applications to a few reports in neurosurgery. With this prospectively designed study, we present our initial experience using an analytical q-ball approach (QBI) as HRFT model for the reconstruction of the language-associated pathways in comparison with conventional DTI-FT.

Methods: For 11 patients with gliomas in the dominant hemisphere, who underwent surgery under awake craniotomy conditions, probabilistic QBI-, and DTI-FT was performed for the major subcomponents of the language-associated fiber bundles on the Medical Imaging Interaction Toolkit (MITK) software. The data was derived from a routine diffusion sensitized magnetic resonance imaging (MRI) sequence (b=1000m/s, 64 gradient directions). Quantitative analysis in MITK and statistical comparison evaluated tract density (TD) and tract volume (TV).

Results: Quantitative analysis showed significantly larger mean TVs using QBI-FT. This holds true for overall fiber reconstructions (p=0.003) as well as for the superior longitudinal fascicle (SLF) subcomponent (p=0.018) while trends were observed for the inferior frontooccipital fascicle (IFOF) or medial and inferior longitudinal fascicle (MLF/ILF) (p=0.053 and p=0.089) (Table 1 [Tab. 1]). Bland-Altman analysis revealed systematically lager TVs via QBI for all reconstructed pathways (subcomponents and total object). The results were independent of tumor volume or peritumoral edema. TD was not significantly different for QBI or DTI.

Conclusion: Quantitative analysis of FT for language-associated pathways reveals larger TVs using QBI compared with DTI-FT (Figure 1 [Fig. 1]). As DTI generally tends to underdetermine TV, larger TVs can be interpreted as a favourable result, resolving fibers no only around the centreline, but also marginal fibers. Moreover, QBI-FT can be well integrated in the clinical routine. Rather than conventional DTI-FT, sophisticated HRFT techniques should be considered for preoperative planning and future intraoperative applications in selected cases of eloquent glioma surgery.