High-dose high-intensity arm neurorehabilitation in chronic stroke improves general motor control

BackgroundAfter stroke, patients can experience a rapid and generalised improvement in the control of movements, including movements that have not been trained (spontaneous recovery). This generalised improvement in motor control (or behavioural repair) is more effective in supporting functional recovery than task-specific training or behavioural compensation but has to date only been observed in the first few months after stroke. MethodsWe studied 81 chronic stroke patients at two time points, 3-weeks apart. 52 patients underwent a 3-week high-dose high-intensity upper limb neurorehabilitation programme (QSUL), and 29 chronic stroke patients receiving no treatment acted as control subjects (SC). At each time point, we assessed arm motor control kinematically using a 2D-planar reaching, a task which cannot be improved through compensation nor functional task training in 3D during neurorehabilitation. In addition, we measured strength, active range of joint motion and the upper extremity Fugl-Meyer score (FMA-UL). ResultsThe FMA-UL increased by 6 points (IQR 3-8) in the QSUL-group and 0 points (IQR -1-1) in the SC-group. There were significant improvements in smoothness, movement time and accuracy of 2D-planar reaching in the QSUL-group compared to SC-group (all group x timepoint interactions P<0.03), which were independent of changes in strength or active range of joint motion. ConclusionsChronic stroke patients retain the capacity for generalised improvement in motor control in response to high-dose high-intensity neurorehabilitation. Normal or closer to normal motor control should remain a therapeutic target for improving arm paresis at all stages after stroke.