Progressive Abduction Loading Therapy with Horizontal-Plane Viscous Resistance Targeting Weakness and Flexion Synergy to Treat Upper Limb Function in Chronic Hemiparetic Stroke: A Randomized Clinical Trial

Abstract

Background: Progressive abduction loading therapy has emerged as a promising exercise therapy in stroke rehabilitation to systematically target the loss of independent joint control (flexion synergy) in individuals with chronic moderate/severe upper-extremity impairment. Preclinical investigations have identified abduction loading during reaching exercise as a key therapeutic factor to improve reaching function. An augmentative approach may be to additionally target weakness by incorporating resistance training to increase constitutive joint torques of reaching with the goal of improving reaching function by "overpowering" flexion synergy. The objective was, therefore, to determine the therapeutic effects of horizontal-plane viscous resistance in combination with progressive abduction loading therapy.

Methods: 32 individuals with chronic hemiparetic stroke were randomly allocated to two groups. The two groups had equivalent baseline characteristics on all demographic and outcome metrics including age (59 ± 11 years), time poststroke (10.1 ± 7.6 years), and motor impairment (Fugl-Meyer, 26.7 ± 6.5 out of 66). Both groups received therapy three times/week for 8 weeks while the experimental group included additional horizontal-plane viscous resistance. Quantitative standardized progression of the intervention was achieved using a robotic device. The primary outcomes of reaching distance and velocity under maximum abduction loading and secondary outcomes of isometric strength and a clinical battery were measured at pre-, post-, and 3 months following therapy.

Results: There was no difference between groups on any outcome measure. However, for combined groups, there was a significant increase in reaching distance (13.2%, effect size; d = 0.56) and velocity (13.6%, effect size; d = 0.27) at posttesting that persisted for 3 months and also a significant increase in abduction, elbow extension, and external rotation strength at posttesting that did not persist 3 months. Similarly, the clinical battery demonstrated a significant improvement in participant-reported measures of "physical problems" and "overall recovery" across all participants.

Conclusion: The strengthening approach of incorporating horizontal-plane viscous resistance did not enhance the reaching function improvements observed in both groups. Data do not support the postulation that one can be trained to "overpower" the flexion synergy with resistance training targeting constitutive joint torques of reaching. Instead, flexion synergy must be targeted with progressive abduction loading to improve reaching function.

Trial registration: ClinicalTrials.gov, NCT01548781.

Keywords: exercise therapy; physical and rehabilitation medicine; physical therapy modalities; resistance training; robotics; stroke; stroke rehabilitation; upper extremity.

Figures

Figure 1

CONSORT flow diagram. Abbreviations: ITT, intention-to-treat; M, male; F, female; UE, upper extremity, FMA, Fugl-Meyer Assessment. Reasons for not meeting inclusion criteria: failed cognitive screen (2), low FMA (1), unsafe on robot due to motor impairment (2). Other reasons: participant moving permanently out of town (1).

Figure 2

Participant set up (top). Written informed consent was obtained from the pictured participant for use in publication and education materials. Visual display (bottom) illustrating the arm avatar viewed by the participant including the home target (gray) and the reaching target (red) with the reaching trajectory shown in white dots.

Figure 3

Diagram of top (A) and front (B) views illustrating the interface of the participant and device. The top view illustrates both the kinematics (extend/retract and axis rotation motion) and viscous resistance of the device in the horizontal plane during an outward reaching motion involving elbow extension and shoulder horizontal adduction. The front view illustrates both kinematics (extend/retract and up/down motion) and kinetics (upward and downward force) of the device. Regarding the frontal-plane kinetics, during a reaching task, if the required volitional abduction torque (abduction loading) was greater than the weight of the limb, the device would emulate a downward force making the limb heavier. In contrast, if the required abduction torque was less than the weight of the limb, the device would emulate an upward force partially unweighting the limb.

Figure 4

Example maximum reaching trajectories (left, indicated by a diamond) while supported on a haptic horizontal surface (top) and while abducting off of the horizontal surface at 50% of maximum abduction strength (bottom). Reaching trajectories were coached to be as fast and as far as possible in the direction of the target and only accepted if within the ±15° cone of tolerance (gray dotted line). The endpoint peak reaching velocity (right, indicated by a square) illustrates the peak reaching velocity associated with each reaching trajectory.

Figure 5

Normalized reaching distance (top) and velocity (bottom) and SEs for the comparison and experimental groups at pretesting, posttesting, and 3-month follow-up. There was a significant increase in reaching distance and velocity for all participants at posttesting that persisted at 3-month follow-up in all participants.