Limb Restoration Rehabilitation
Neuroplasticity (includes neuroscience)
In patients with mild-to moderate stroke, an average of 70% of the recovery happens within the 3-6 months post-stroke (Prabhakaran et al., 2008; Krakauer & Marshall, 2015; Winters at al., 2016; Stinear et al., 2017) and upper limb recovery remains incomplete in most patients despite therapy (Kwakkel et al., 2002; Krakauer & Marshall, 2015; Lang et al., 2016; Winstein et al., 2016; see Stinear et al., 2017 for a review of recent trials). However, even when rehabilitation is provided at the chronic stage of stroke, some individuals do benefit from treatment at the activity level. In this talk, we will use whole brain white matter analysis to show which pathways were found to be predictive of recovery of arm function in chronic stroke in those who benefited from a 8-week upper-limb task specific training.
To investigate the neuroanatomical correlates of upper limb recovery following task-specific training.
Design : Pre-post intervention trial
Setting : Participants underwent diffusion tensor image (DTI) scans before and after training.
Participants (or Animals, Specimens, Cadavers) : Thirty participants with mild and moderate chronic stroke. The analysis was performed on the entire sample and on two sub-groups: Responders [ΔARAT ≥ 6, N=13] and Non-Responders [ΔARAT < 6, N=17].
Interventions : 1h of upper limb task-specific training for 4 days/week during 8 weeks.
Main Outcome Measure(s) : FA values were extracted using tract-based spatial statistics (TBSS). The recovery from activity limitation was quantified by the change in the Action Research Arm Test (ΔARAT).
While ARAT scores increased following training (p<0.0001, ΔARAT = 5.53 ± 5.20), change in FA was not significant. However, the TBSS analysis revealed significant positive correlations between baseline FA and ΔARAT but only for Responders (r=0.85, p=0.0002). Importantly, there were no significant group differences between baseline ARAT scores or baseline FA. Twenty-two of the 72 tracts of the HCP-842 atlas had streamlines that contained voxels predictive of ΔARAT and they were located in the contralesional hemisphere.
Our results show that: 1) absence of a baseline difference in FA values between Responders and Non-Responders suggests that white matter integrity is not the only factor predicting recovery in chronic stroke; 2) in Responders, contralesional hemisphere may be part of functional brain reorganization. Additional analyses are needed to characterize the differences between Responders and Non-Responders (e.g. CST damage, FA ratio between ipsi- and contralesional CST).