Going Beyond Diffusion Tensor Imaging Tractography in Eloquent Glioma Surgery–High-Resolution Fiber Tractography: Q-Ball or Constrained Spherical Deconvolution?

Abstract

As a result of the resolution of intravoxel fiber crossing, high-resolution fiber tractography (HRFT) provides advantages over conventional diffusion tensor imaging (DTI) for fiber tractography (FT). Nevertheless, neurosurgically applied FT is still predominantly based on DTI. Although the application of HRFT is evolving, there is still a lack of data about which method should be preferred. With this prospectively designed study, we present our initial experience comparing an analytical Q-ball imaging (QBI) approach with constrained spherical deconvolution (CSD) and conventional DTI-FT considering a particularly neurosurgical perspective. For 18 patients with eloquent gliomas in the dominant hemisphere, probabilistic FT based on QBI, CSD, and DTI was performed for the major components of the language-associated pathways using a routine diffusion-weighted sequence. Quantitative analysis evaluated tract density, tract volume (TV), tract length (TL), number of fibers, and tract surface (TS) of the fiber object. Both HRFT models showed a significantly larger mean TV, TL, and TS compared with DTI (for QBI vs. DTI: TV (P= 0.0000), TL (P= 0.0048), and TS (P= 0.0129); for CSD vs. DTI: TV (P= 0.0000), TL (P= 0.0008), and TS (P= 0.0010)). However, results of QBI versus CSD did not differ significantly for these variables: TV (P= 0.1415), TL (P= 0.2837), and TS (P= 0.3692). Bland-Altman analysis supports these findings, suggesting systematically higher values for TV, TL, and TS with HRFT but no relevant differences of either QBI or CSD. Neither tumor volume nor peritumoral edema influenced FT results. Our quantitative analysis showed no significant differences regarding TV, TL, and TS for the HRFT methods; however, it suggested advantages over DTI-FT in terms of the display of marginal and terminal fibers. In our recently established setting, QBI-FT shows greater potential for integration into the clinical workflow.