Effects of diffusion time on diffusion tensor derived parameters measured on the rat brain at ultrahigh magnetic field

Introduction: Over the last decade, Diffusion Tensor Imaging (DTI) became an MR technique routinely used in clinical environments. The importance of preclinical studies at ultrahigh magnetic field resulted in an increasing number of publications focused on the rodent brain. Several studies have shown that the choice of sequence parameters (diffusion gradient sampling scheme, number of directions, diffusion gradient duration, diffusion time, region of interest) as well as the intrinsic specifications of the system (magnetic field strength) can have a huge impact on the derived tensor quantifications [1-3], but this work has never been done at ultrahigh magnetic field. In this context, the aim of this work was to study the influence of td and brain microstructures on diffusion tensor derived parameters in the rat brain at 9.4T. Materials and Methods: All experiments were performed on an actively-shielded 9.4T/31cm magnet (Varian/Magnex) equipped with 12-cm gradient coils (400mT/m, 120μs) with a quadrature transmit-receive 18-mm surface RF coil. The rat (n = 4) was lying prone, its head secured via ear bars and continuously anesthetized under a flow of 1.5-2% isoflurane in oxygen. Body temperature was maintained at 37±0.5°C using thermoregulated water circulation. After automatic adjustment of first and second order shims (FASTMAP [4] water half-height linewidth ranged between 18 and 22 Hz), 3 repeated Diffusion Tensor Echo Planar Images (4 shots) were performed with td = 10, 25 and 39 ms respectively. A semi-adiabatic double spin echo sequence was used [5] and diffusion gradients were applied around the first 180° with the same polarity for short td or around the two 180° with inverted polarity for long td, resulting in a b-value set to 1000 s.mm. Diffusion gradients were applied along 42 spatial directions: Icosahedral 21 directions as well as the 21 opposite directions to cancel b-value cross terms [6]. For the three measurements, image parameters were: FOV = 23 × 15 mm, matrix size = 128 × 64 zero-filled to 256 × 168, 8 slices of 0.8 mm thickness in the axial plane, 18 averages with TE/TR = 50/2000 ms. Note that, both TE and the b-value were kept constant (50 ms and 1000 s.mm respectively) for the three different td, allowing an accurate assessment of the effects of the diffusion time only. Using homemade Matlab (Mathworks, Natick, MA) software, diffusivity values (ADC, D// and D⊥) as well as fractional anisotropy (FA) was derived from the tensor. On the direction encoded color maps, ROIs were drawn in the corpus callosum and in the cortex for the 8 different slices of the rat brain (fig. 1) in order to evaluate the variation of diffusion tensor derived parameters function of the td and along the rat brain. Collected data were submitted to a Friedman non-parametric test.