Motor recovery after striato-capsular stroke (SCS) depends on normal laterality of brain activation: a longitudinal PET study
Introduction: Although bilateral motor network overactivations during finger motion are documented after stroke, their relation with actual recovery is unknown. One longitudinal study reported a shift of activations from the affected to the unaffected hemisphere with elapsing time (Marshall et al, Stroke, 2000;31:656), but whether this shift was related to recovery was not assessed. Subjects and Methods: In this prospective, longitudinal study of 5 R-handed pts (59±13 yrs) with R-sided hemiparesis due to first-ever left SCS, we obtained two 3D-PET H2O15 studies(ECAT HR+), the 1st at 1–4 and the 2nd at ∼8 months after onset (PET1 and PET2, respectively), replicating 4 times 2 conditions: i) rest with eyes closed, metronome on; ii) metronome-cued right thumb-to-index tapping (TIT) at fixed-rate (1.26 Hz). Data were assessed individually with SPM96 and p< 0.05, corrected; controls were 7 healthy age-matched R-handed subjects. We computed correlations between R hand motor performance (max number of TITs in 15 s) and Laterality Index(LI = theaffected-to-unaffected-hemisphere ratio of activated voxels, Cramer et al, Stroke, 1997;28:2518). Results and comments: motor scores significantly improved from PET1 to PET2 (18±4 and 29±5, respectively, p<0.05, paired t-test); 2 pts exhibited mirror movement at PET1 and 1 at PET2. Significant overactivations at PET1 affected the whole motor network bilaterally, but were considerably less extensive at PET2, with a significant LI shift (.69±.11 and .08±.54, respectively, p<0.05). Shift magnitude was negatively correlated (Spearman rho= -.975, p<0.05) with clinical recovery (expressed as motor score changes), that is, the more the laterality shifted (i.e., deviated from normality), the lesser the recovery. Conclusion: The shift towards the unaffected hemisphere confirms earlier report, though performance was fixed here. This study is the first to document a correlation between recovery and brain activation patterns after stroke, such that the greater the shift, the worse the recovery.