Introduction: The purpose of this study was to compare the following three treatments of upper limb motor impairment in chronic moderate/severely impaired stroke survivors: 1) Customized whole limb motor learning (Whole ML), including shoulder/elbow robotics (ROB) and functional electrical stimulation (FES); 2) prescribed shoulder/elbow ROB + motor learning (S/E ROB ML); and 3) prescribed wrist/hand FES + ML (W/H FES ML).
Methods: Subjects with chronic stroke were stratified according to relative level of impairment (Fugl-Meyer coordination Test) in wrist/hand versus shoulder/elbow. Their stratification category dictated the treatment group to which they were allocated, as follows: 1) WHOLE ML (n=18), equal distal and proximal impairment; 2) S/E ROB ML (n=10 ), greater shoulder/elbow impairment; and 3) FES ML (n=8), greater wrist/hand impairment. Treatment was 5 days/wk, 5 hrs/day, 60 sessions. Primary measure was the Arm Motor Assessment Tool, 13 complex tasks, Functional Scale (AMAT-F), acquired at baseline and at follow-up, 3 months after last treatment. Secondary measures were subscales, AMAT-F Shoulder/Elbow (AMAT-F S/E) and AMAT Wrist/Hand (AMAT-F W/H). Statistical analyses were Kruskal Wallace model to account for group size differences; group comparisons and within-group pre-/follow-up comparisons were analyzed (and multiple comparisons corrections).
Results: All three groups had significant within-group gain for AMAT-F (p=.0007; p=.039; p=.047, respectively; similar results were shown for AMAT S/E subscale. Only the Whole ML group had significant gain in AMAT W/H (p=.007). Gain for all three groups exceeded the threshold for clinically significant gain in AMAT-F (.21 points), as follows: Whole ML (.66 points, 3 x’s the threshold); S/E ROB ML (.53 points, 2.5 x’s); W/H FES ML (.46 points, 2 x’s).
Conclusion: For the customized whole limb motor learning group there was greater improvement in upper limb function versus the two groups that received prescribed, focused training for either shoulder/elbow or wrist/hand, according to the AMAT-F clinically significant threshold and according to the AMAT-F W/H subscale.