Diffusion tensor imaging of the visual sensory pathway: are we there yet?

N THIS ISSUE OF THE JOURNAL, YAMAMOTO AND COLleagues 1 have used diffusion tensor imaging (DTI) to visualize the three-dimensional architecture of the three major groups of fiber bundles—anterior, central, and posterior—in the optic radiation in vivo. Although their results on the central and posterior bundles were in agreement with the classic study of the optic radiation performed by Ebeling and Reulen and published in 1988, 2 Yamamoto and associates found that Meyer’s loop as depicted by tractography was located substantially more posteriorly than reported by Ebeling and Reulen, who obtained their results by studying 25 formalin-fixed human hemispheres using either Klingler’s fiber dissection technique or consecutive frontal sections, enabling them to gain precise data on the three-dimensional course, length, and other measurements of the radiation. Why the discrepancy? DTI uses properties of water diffusion to provide information about the integrity, location, and orientation of white-matter tracts in the brain 3‐5 and represents an exciting new opportunity not only for examining the microstructure of the brain in various diseases but also for studying functional connectivity. DTI exploits the fact that diffusion of water molecules in white matter is constrained, or anisotropic. It moves preferentially along the primary axes of fiber tracts. Diffusion tensor data are represented in each voxel as a three-dimensional ellipsoid, reflecting the rate of diffusion along the ellipsoid’s three principal axes. Voxels along fiber pathways tend to form lines along these pathways. One, thus, can trace the structure of any fiber tract by connecting the long axis of each ellipsoid between given starting and end points. There are, however, several problems with quantitative DTI tractography that need to be addressed before this method becomes a clinical reality, and these problems may be responsible for the discrepancy between the findings of Yamamoto and associates 1 and those of Ebeling and Reulen. 2 For instance, there is as yet no reference standard for in vivo tractography. DTI also suffers from the same artifacts and limitations as those associated with diffusionweighted imaging (DWI), including motion artifactinduced ghosting, eddy current misregistration errors, and loss of signal intensity from variations in susceptibility. In particular, it has been shown in several DWI studies that partial volume effects from cerebrospinal fluid (CSF) result in overestimation of the apparent diffusion coefficient (ADC) and underestimate diffusion anisotropy in regions prone to partial volume effects, such as may be the case with Meyer’s loop as it is adjacent to the temporal horn of the lateral ventricle. Many of these same studies also show that suppressing CSF signals by incorporating fluid-attenuated inversion recovery (FLAIR) techniques eliminates many of the inaccuracies in ADC and anisotropy measurements in susceptible brain regions.6‐8 In a recent study, Chou and associates9 demonstrated that performing DTI with CSF suppression provided a more accurate picture of fiber tract position and volume in periventricular regions than DTI without CSF suppression.