Figure 2

Region of interest volumes computed from longitudinal in vivo MRI and high-resolution ex vivo MRI. (a, b) Representative tracings of the left (L) and right (R) striatum (STR) in repeated, longitudinal in vivo anatomical MRI scans acquired from the same animal at postnatal day 23 (PD23; a) and PD95 (b). (c) Representative tracings of the left and right STR, thalamus (THAL) and hippocampus (HIPP) in an ex vivo high-resolution (High-Res) anatomical magnetic resonance imaging (MRI) scan acquired from the same PD95 animal as in a and b. (d) The color legend for a–c is shown here, including an example three-dimensional (3D) reconstruction of each traced region computed from the High-Res scan as in c. (e, f) Graphs depict differences in raw volumes of the left (e) and right (f) striatum between animals treated for 70 days (PD24–94) with either lisdexamfetamine (LDX) or vehicle (VEH; N=12 per group). No statistical differences were found between the groups at PD23. At PD95, relative to VEH-treated controls, the animals treated with LDX showed significant hypertrophy of the left but not right striatum, with left striatal volume showing a significant treatment by time point interaction (P=0.04). (g) Graph depicts t-score differences in normalized regional volumes between treatment groups computed from longitudinal in vivo MRI (left) and cross-sectional ex vivo High-Res MRI (right). Longitudinal in vivo MRI: no statistical differences in striatal volume were found at PD23. At PD95, relative to VEH-treated controls, animals treated with LDX showed significant hypertrophy of the left (P=0.0002) and right (P=0.02) striatum. Only the left striatum showed a significant treatment by time point interaction. High-Resolution ex vivo MRI: relative to VEH-treated controls, the animals treated with LDX showed significant hypertrophy of the left (P=0.005) and right (P=0.04) striatum. Neither the thalamus nor the hippocampus showed a significant group difference in volume. (h) Exploratory analysis of LDX effects on subregional striatal volume at PD95 (ex vivo High-Res MRI). The illustration to the left depicts the approximate coronal location of segments comprising the dorsal, mid and ventral striatum as derived from FSL-based segmentation of subject-specific tracing masks (see the 'Materials and methods' section). Partial eta-squared effect sizes were computed from between-group t-test comparisons of normalized volume of each subregion within each hemisphere. As shown in the graph on the right, these analyses revealed that within both hemispheres (i.e., left and right striatum), LDX treatment yielded the largest effect on striatal hypertrophy in the mid and dorsal striatum, with a relatively smaller effect on the ventral aspect of this structure. MRI, magnetic resonance imaging.