Résumé
Background. Kinematic assessment of upper limb motor recovery after stroke may be related to clinical scores while being more sensitive and reliable than clinical evaluation alone. Objective. To identify the potential of kinematics in assessing upper limb recovery early poststroke. Methods. Thirteen patients were included within 1 month poststroke and evaluated once a week for 6 weeks and at 3 months with (a) the Fugl-Meyer Assessment (FMA) and (b) kinematic analysis of reach-to-grasp movements. The link between clinical and kinematic data was identified using mixed model with random coefficient analysis. Results. Movement time, trajectory length, directness, smoothness, mean and maximum velocity of the hand were sensitive to change over time and distinguished between movements of paretic, nonparetic, and healthy control limbs. The FMA score increased with movement smoothness over time, explaining 62.5% of FMA variability. Conclusion. Kinematic analysis of reach-to-grasp movements is relevant to assess upper limb recovery early poststroke, and is linked to the FMA. Kinematics could provide more accurate real-time indicators of patients' recovery as compared with the sole use of clinical scores, although it remains challenging to establish the universality of the reaching model in relation to motor recovery after stroke.