Robot-assisted training compared with an enhanced upper limb therapy programme and with usual care for upper limb functional limitation after stroke: the RATULS three-group RCT.

Natasha Alvarado,H. Rodgers,L. Vale,S. Andole,Steven Hogg,S. Wilkes,C. Price,F. van Wijck,D. Howel,Richard Francis,Lisa Shaw,H. Bosomworth,J. Dawson,H. Krebs,T. Finch,N. Wilson,Cristina Fernandez-Garcia,Lydia Aird,D. Turner,D. Cohen,G. Ford,N. Hughes,L. Ternent

DOI: https://doi.org/10.3310/hta24540

IF: 4.106

2020-10-01

Health Technology Assessment

Abstract:BACKGROUND Loss of arm function is common after stroke. Robot-assisted training may improve arm outcomes. OBJECTIVE The objectives were to determine the clinical effectiveness and cost-effectiveness of robot-assisted training, compared with an enhanced upper limb therapy programme and with usual care. DESIGN This was a pragmatic, observer-blind, multicentre randomised controlled trial with embedded health economic and process evaluations. SETTING The trial was set in four NHS trial centres. PARTICIPANTS Patients with moderate or severe upper limb functional limitation, between 1 week and 5 years following first stroke, were recruited. INTERVENTIONS Robot-assisted training using the Massachusetts Institute of Technology-Manus robotic gym system (InMotion commercial version, Interactive Motion Technologies, Inc., Watertown, MA, USA), an enhanced upper limb therapy programme comprising repetitive functional task practice, and usual care. MAIN OUTCOME MEASURES The primary outcome was upper limb functional recovery 'success' (assessed using the Action Research Arm Test) at 3 months. Secondary outcomes at 3 and 6 months were the Action Research Arm Test results, upper limb impairment (measured using the Fugl-Meyer Assessment), activities of daily living (measured using the Barthel Activities of Daily Living Index), quality of life (measured using the Stroke Impact Scale), resource use costs and quality-adjusted life-years. RESULTS A total of 770 participants were randomised (robot-assisted training, n = 257; enhanced upper limb therapy, n = 259; usual care, n = 254). Upper limb functional recovery 'success' was achieved in the robot-assisted training [103/232 (44%)], enhanced upper limb therapy [118/234 (50%)] and usual care groups [85/203 (42%)]. These differences were not statistically significant; the adjusted odds ratios were as follows: robot-assisted training versus usual care, 1.2 (98.33% confidence interval 0.7 to 2.0); enhanced upper limb therapy versus usual care, 1.5 (98.33% confidence interval 0.9 to 2.5); and robot-assisted training versus enhanced upper limb therapy, 0.8 (98.33% confidence interval 0.5 to 1.3). The robot-assisted training group had less upper limb impairment (as measured by the Fugl-Meyer Assessment motor subscale) than the usual care group at 3 and 6 months. The enhanced upper limb therapy group had less upper limb impairment (as measured by the Fugl-Meyer Assessment motor subscale), better mobility (as measured by the Stroke Impact Scale mobility domain) and better performance in activities of daily living (as measured by the Stroke Impact Scale activities of daily living domain) than the usual care group, at 3 months. The robot-assisted training group performed less well in activities of daily living (as measured by the Stroke Impact Scale activities of daily living domain) than the enhanced upper limb therapy group at 3 months. No other differences were clinically important and statistically significant. Participants found the robot-assisted training and the enhanced upper limb therapy group programmes acceptable. Neither intervention, as provided in this trial, was cost-effective at current National Institute for Health and Care Excellence willingness-to-pay thresholds for a quality-adjusted life-year. CONCLUSIONS Robot-assisted training did not improve upper limb function compared with usual care. Although robot-assisted training improved upper limb impairment, this did not translate into improvements in other outcomes. Enhanced upper limb therapy resulted in potentially important improvements on upper limb impairment, in performance of activities of daily living, and in mobility. Neither intervention was cost-effective. FUTURE WORK Further research is needed to find ways to translate the improvements in upper limb impairment seen with robot-assisted training into improvements in upper limb function and activities of daily living. Innovations to make rehabilitation programmes more cost-effective are required. LIMITATIONS Pragmatic inclusion criteria led to the recruitment of some participants with little prospect of recovery. The attrition rate was higher in the usual care group than in the robot-assisted training or enhanced upper limb therapy groups, and differential attrition is a potential source of bias. Obtaining accurate information about the usual care that participants were receiving was a challenge. TRIAL REGISTRATION Current Controlled Trials ISRCTN69371850. FUNDING This project was funded by the National Institute for Health Research (NIHR) Health Technology Assessment programme and will be published in full in Health Technology Assessment; Vol. 24, No. 54. See the NIHR Journals Library website for further project information.

Medicine,Engineering