Abstracts

Rationale: It has been hypothesized, that motor recovery after hemispherectomy is mediated by the compensatory reinforcement of contralesional uncrossed fibers of the pyramidal tract (PT_c), but also of the bilaterally organized cortico-rubro-spinal system. While the importance of so-called alternate motor fibers (aMF), which most likely comprise the cortico-rubro-spinal system, has been demonstrated in motor recovery after stroke, no white matter correlates of motor recovery in patients after hemispherectomy have been observed up to now. In the current study, we used diffusion tensor imaging (DTI) to examine PT_c and aMF in these patients by relating their microstructural properties to functional motor outcome. Methods: 20 patients after functional hemispherectomy (twelve women; mean age at scan 19.6±3.0 years; mean age at surgery 10.5±5.8 years; etiology: porencephaly, n=10 | neurodevelopmental disorders, n=5 | Rasmussen's encephalitis, n=5) and ten healthy controls (six women; mean age at scan 22.1±2.1) underwent DTI at 3 T (1.72 mm³ voxel size; 60 directions with b=1000s/mm², 6 b0). Patient's motor impairment of the upper extremity was assessed with Motricity Index scores (MI). Using FSL, we applied probabilistic tractography to reconstruct PT_c and aMF separately (fig. 1). Contralesional seed regions were defined in the basis pontis (for PT_c) and in the tegmentum pontis (for aMF). Two contralesional waypoint masks were defined in the internal capsule and the primary motor cortex. Tract-specific fractional anisotropy (FA) values and measures of probabilistic connectivity were extracted for statistical analyses. To specifically assess probabilistic connectivity of cortico-rubral fibers, tractography was re-run using a seed region in the contralesional red nucleus and the waypoint masks as described above. Results: The group mean MI was 49.3±17.1. MI were found to be inversely related to age at surgery (r=-0.589, p=0.006). aMF exhibited a trend towards higher FA values (p=0.056) in patients relative to controls, whereas PT_c did not show any difference (p=0.645). Probabilistic connectivity of cortico-rubral fibers was found to be higher in the patient group relative to controls (p<0.001). Partial correlation analyses with age at surgery as control variable yielded strong positive correlations for FA of PT_c (r=0.650, p=0.003) and aMF (r=0.597, p=0.007; fig. 2). A final stepwise regression analysis with MI as dependent variable resulted in a significant model (F(2,19)=11.714, p=0.001) with FA of aMF (p=0.007) and age at surgery (p=0.014) as predictor variables. Conclusions: The inverse correlation between age at surgery and MI suggests that earlier surgery is related to better motor outcome. Their associations with MI indicate that observed diffusivity alterations are functionally meaningful and may represent the result of preceding compensatory remodeling. Higher probabilistic connectivity of cortico-rubral fibers as well as results of the stepwise regression analysis suggest a crucial role of the cortico-rubro-spinal system in this plastic process. This system may be targeted in future rehabilitation trials.