RESUMO
BACKGROUND: Kinematic indices (KIs) are frequently used as objective measures to assess the upper extremities motor performance in post stroke patients. The clinimetric analysis of these indices has been mostly limited to their averaged values over different directions of reaching movements. Recent studies indicate direction dependencies of such motor performances due to neural and/or biomechanical causes. The direction dependencies of such indices and their clinimetric parameters remains to be investigated. METHODS: An apparatus was built to perform and measure planar point-to-point reaching tasks in 8 directions using a virtual reality environment. 24 stroke and 18 healthy individuals participated in the study. 24 kinematic indices were calculated. Reliability (ICC), construct validity (Spearman correlation), and responsiveness (paired t-test pre and post intervention) were analyzed in each direction. RESULTS: The clinimetric parameters were found highly direction dependent. The reliability of the indices were strongest when moving away and towards the body. The validity (Spearman>0.75) and responsiveness (p<0.05) were most pronounced when moving in the NW-SE direction. These findings are in compliance with some previous neuro-musculoskeletal observations. CONCLUSION: While smoothness parameters are relatively uniform in all directions, speed and accuracy are direction dependent. The clinimetrics of the kinematic indices also depend on the direction and show stronger values in the NW-SE direction which is therefore proposed as the most accurate and responsive direction for kinematic assessment in stroke patients.