Objective

In stroke patients, motor impairments evaluated during the first few weeks usually predict upper limb motor outcomes after rehabilitation. However, a recent growing literature suggests that severe patients might not follow this principle . This study aimed at exploring predictors of motor recovery after intensive training in that sub population of stroke patients.

Material/Patients and methods

Thirty-eight patients with sub-acute hemiparesis (19 females, age 56 ± 17 [19–87] years; time from stroke, 55 ± 22 days) carried out 16 sessions (33 ± 12 days) of robot-assisted adaptive training using a shoulder/elbow device integrated into conventional care. Motor outcomes were assessed using the Fugl-Meyer Assessment (FMA) scale and kinematic measures at baseline and after treatment. Potential predictors were baseline neurological, cognitive (i.e. aphasia and neglect scores) status and kinematic measures. Age, gender, type of stroke, time since stroke, side of paresis and duration of training were also collected. Statistical analyzes explored potential predictors of motor changes and of favorable changes (Minimal Clinical Important Difference [MCID]).

Results

Analyzes didn’t show any clinical predictor of motor outcomes. Moreover, baseline patients’ clinical characteristics did not predict clinically important improvement. Baseline kinematic measures were poor predictors of change in FM scores after training.

Discussion – Conclusion

The results suggested that pre-therapy motor impairments, motion kinematic measures and associated cognitive deficits might not be reliable predictors of motor outcomes after an intensive rehabilitation program for the upper limb in severe stroke patients 2 months after onset. Intensive upper limb interventions could contribute to enhance motor recovery, especially in severe subacute stroke patients.