Tetsuo Koyama, Midori Mochizuki, Yuki Uchiyama
Automated diffusion-tensor tractography can reliably detect microstructural alterations in alternative descending motor pathways originating from BA44, even in patients with mild stroke and favorable motor outcomes. This neuroimaging protocol offers a practical and clinically viable approach for assessing neural reorganization in stroke rehabilitation.
BACKGROUND: Although upper-limb motor recovery after stroke is strongly linked to the integrity of the primary corticospinal tract, alternative descending motor pathways originating from secondary motor-related cortical areas, such as Brodmann area 44 (BA44; Broca's area), may also contribute to functional recovery. This study aimed to establish a practical automated diffusion-tensor tractography protocol to evaluate the microstructural properties of corticospinal and corticoreticular projections (CS-P and CR-P) originating from BA44 in patients with mild stroke who achieved favorable motor outcomes.
METHODS: We retrospectively investigated patients with first-ever, unilateral thalamic or putaminal hemorrhagic stroke who achieved complete independence in activities of daily living and a Brunnstrom Recovery Stage of ≥5 in both the upper and lower limbs at discharge. Diffusion-tensor imaging was performed during the second week of hospitalization using a 3.0-T MRI scanner (MAGNETOM Trio; Siemens, Erlangen, Germany). The CS-P and CR-P were reconstructed using an automated tractography pipeline, with the cerebral peduncle and gigantocellular reticular nucleus serving as seed regions, respectively, and BA44 serving as the target region. Interhemispheric differences in fractional anisotropy (FA) values between the lesioned and non-lesioned hemispheres were evaluated using the Wilcoxon signed-rank test.
RESULTS: A total of 14 patients were enrolled. Both the CS-P and CR-P originating from BA44 were successfully reconstructed in all subjects. FA values in the lesioned hemisphere were significantly lower than those in the non-lesioned hemisphere for both the CS-P (median (IQR): 0.484 (0.452-0.514) vs. 0.508 (0.474-0.531), S = 39.5, P = 0.011) and the CR-P (0.442 (0.410-0.462) vs. 0.463 (0.446-0.474), S = 31.5, P = 0.049).
CONCLUSIONS: Automated diffusion-tensor tractography can reliably detect microstructural alterations in alternative descending motor pathways originating from BA44, even in patients with mild stroke and favorable motor outcomes. This neuroimaging protocol offers a practical and clinically viable approach for assessing neural reorganization in stroke rehabilitation.