Deficits of reach-to-grasp coordination following stroke: Comparison of instructed and natural movements.

The present work evaluates whether stroke-induced deficits of reach-to-grasp movements, established by typical laboratory paradigms, transfer unconditionally to more natural situations. Sixteen patients with a stroke to the motor-dominant left hemisphere and 16 age- and gender-matched healthy control subjects executed grasping movements with their left (ipsilesional, non-dominant) hand. All movements started in the same position, were aimed at the same object positioned in the same location, and were followed by forward displacement of that object along the same path. Twenty movements were performed as a repetitive, externally triggered task executed for their own sake (context L, as in typical laboratory tasks). Twenty movements were performed as part of a self-initiated action sequence aimed at winning a reward (context E, similar to many everyday situations). The kinematics and dynamics of the transport, grasp and manipulation component of each reach-to-grasp movement were quantified by 41 parameters. Analyses of variance yielded a significant effect of Context for 29 parameters, a significant effect of Group for 9 parameters (mostly related to the coupling of hand transport and grip aperture), and a significant interaction for 5 parameters (all related to the coupling of hand transport and grip aperture). The interaction reflected the fact that stroke patients' movement parameters were more abnormal in context E than in context L. Our data indicate that unilateral stroke degrades the grasp-transport coupling, and that stroke-related motor deficits may be more pronounced in a natural than in a laboratory context. Thus, for stroke patients, assessments and rehabilitation regimes should mainly use activities that are as natural as possible.