Feedback-controlled robotics-assisted treadmill exercise to assess and influence aerobic capacity early after stroke: a proof-of-concept study

Abstract

Purpose: The majority of post-stroke individuals suffer from low exercise capacity as a secondary reaction to immobility. The aim of this study was to prove the concept of feedback-controlled robotics-assisted treadmill exercise (RATE) to assess aerobic capacity and guide cardiovascular exercise in severely impaired individuals early after stroke. Method: Subjects underwent constant load and incremental exercise testing using a human-in-the-loop feedback system within a robotics-assisted exoskeleton (Lokomat, Hocoma AG, CH). Inclusion criteria were: stroke onset ≤8 weeks, stable medical condition, non-ambulatory status, moderate motor control of the lower limbs and appropriate cognitive function. Outcome measures included oxygen uptake kinetics, peak oxygen uptake (VO₂peak), gas exchange threshold (GET), peak heart rate (HRpeak), peak work rate (Ppeak) and accuracy of reaching target work rate (P-RMSE). Results: Three subjects (18–42 d post-stroke) were included. Oxygen uptake kinetics during constant load ranged from 42.0 to 60.2 s. Incremental exercise testing showed: VO₂peak range 19.7–28.8 ml/min/kg, GET range 11.6–12.7 ml/min/kg, and HRpeak range 115–161 bpm. Ppeak range was 55.2–110.9 W and P-RMSE range was 3.8–7.5 W. Conclusions: The concept of feedback-controlled RATE for assessment of aerobic capacity and guidance of cardiovascular exercise is feasible. Further research is warranted to validate the method on a larger scale.

Implications for Rehabilitation

• Aerobic capacity is seriously reduced in post-stroke individuals as a secondary reaction to immobility.

• Robotics-assisted walking devices may have substantial clinical relevance regarding assessment and improvement of aerobic capacity early after stroke.

• Feedback-controlled robotics-assisted treadmill exercise represents a new concept for cardiovascular assessment and intervention protocols for severely impaired individuals.

• Cardiovascular exercise
• constant load testing
• incremental exercise testing
• human-in-the-loop
• non-ambulatory

Acknowledgements

The authors thank Prof. Thierry Ettlin for facilitating and supporting the study, Dr Bernhard Spoendlin and Dr Andreas Rohner for the cardiac screening, Dr Niklaus Urscheler for his medical advice and Heike Rosemeyer for assistance during recruitment and exercise tests.