Post-stroke patients often suffer from a hemiparesis affecting the functional abilities of lower and/or upper extremities. Improving walking and everyday activities are therefore important rehabilitation goals for these patients.

Robotic assisted rehabilitation (RAR) and functional electrostimulation (FES) can, among others, be used as a supplement to conventional rehabilitation in post-stroke patients. The purpose of this report was to evaluate whether there is an additional clinical benefit of using RAR or FES in stroke rehabilitation when compared to standard rehabilitation alone.

Methods: For lower limb stroke rehabilitation, we conducted two systematic reviews to evaluate the potential clinical benefit of RAR and FES with regard to functional outcomes such as ability to walk and activities of daily living. For upper limb stroke rehabilitation, we have summarised the evidence identified by a recent high quality AWMF S3 guideline.

Results: For lower limb rehabilitation, the evidence consisted of eleven RCTs for RAR and further 17 RCTs for FES. The evidence support the use of RAR (especially end-effectors) in a rehabilitation programme. Insufficient evidence was found to prove that any of the FES interventions combined with standard rehabilitation was superior to standard rehabilitation alone, although the evidence suggests that some sub-interventions of FES (tilt sensor FES systems) are non-inferior when compared to ankle-foot-orthoses in patients with drop foot.

For upper limb stroke rehabilitation, the AWMF S3 guideline found evidence consisting of one Cochrane SR and 16 RCTs for RAR and nine RCTs for FES that fulfilled our inclusion criteria. The evidence identified by the guideline supports arm robot therapies including both exoskeletons and other electromechanical active robotic devices especially for patients in the subacute stadium. For FES, the AWMF S3 guideline found low quality evidence demonstrating that FES, indicated for patients with severe incomplete hand paresis and partially preserved proximal motoric function (movement and holding function), can be used for inducing grasping and releasing or finger and hand extension combined with training of everyday activities.

Conclusion: The identified evidence indicates that RAR may yield a clinical benefit in stroke rehabilitation in the subacute stadium. The evidence is insufficient to show superiority or inferiority of FES and standard rehabilitation in comparison to standard rehabilitation alone (although some devices were proven non-inferior).

In light of numerous therapeutic options available in stroke rehabilitation, often with limited proven benefit, but increased costs, health economic evaluations for those interventions that showed a certain clinical benefit or at least non-inferiority are recommended. Here, the focus should be on relieving the physiotherapist's workload (both in terms of time and physical). For such an evaluation, it is essential to consider the general conditions or the organisational setting and the severity of the stroke. On the other hand, a disinvestment in treatment modalities that are not proven by evidence or are not cost-effective should be considered.