Abnormal synergies and associated reactions post-hemiparetic stroke reflect the neuroanatomy of brainstem motor pathways

Abstract

Individuals with moderate-to-severe post-stroke hemiparesis have difficulty controlling proximal and distal joints of the arm independently because they are constrained to stereotypical movement patterns called the flexion and extension synergies. Over the last three decades, we and others have quantitatively characterized these patterns and have provided evidence that they emerge because of an increased influence of diffusely-projecting brainstem motor pathways following stroke-induced damage to corticospinal and corticobulbar pathways. In our recent work that has focused on how they influence post-stroke hand function, we observed three notable aspects of synergy expression that we have never studied systematically: (1) paretic wrist and finger muscles were often activated maximally while individuals contracted muscles at a different joint, not during a maximal voluntary contraction of the wrist and finger muscles themselves; (2) there were differences in the magnitude of synergy expression when elicited via contraction of proximal muscles vs. distal muscles; (3) there was consistent movement resembling flexion or extension synergy patterns in the paretic limb during maximal efforts with the non-paretic limb (a phenomenon described clinically as an associated reaction), and the strength of these movement seemed to differ based on which muscles in the non-paretic limb were being activated. In the current study, we investigated the above behaviors systematically during maximal isometric contractions of shoulder, elbow and wrist/finger muscles, specifically focusing on differences between proximal vs. distal joints and flexor vs. extensor muscles. Our overall hypothesis is that the muscle-dependent nature of the behaviors we have observed is consistent with how the muscles are impacted by corticofugal damage and the upregulation of brainstem motor pathways that results. That is, we expected that our findings would reflect the fact that the greatest proportion of descending neural control comes from the corticospinal tract for distal muscles and from brainstem motor pathways for proximal muscles. We further expected that findings would reflect the fact that the reticulospinal tract, thought to underlie the flexion synergy, has bilateral effects in the upper limbs and favors facilitation of flexor muscles on the ipsilateral side. Supporting this hypothesis, we found that for some participants, joint torque and muscle activation generated during maximal voluntary contractions were lower than during maximal synergy-induced contractions. This was more prevalent and more severe in magnitude at the wrist and fingers than at the shoulder and elbow. We also found that synergy-driven contractions were strongest when elicited via proximal joints and weakest when elicited via distal joints. Finally, we found that associated reactions in the paretic wrist/finger flexors were stronger than those of other paretic muscles and were the only ones whose response depended on whether the contralateral contraction was at a proximal or distal joint. Our results provide indirect evidence for how an increased reliance on brainstem motor pathways post-stroke contributes to abnormal motor behaviors and demonstrate the need to examine whole-limb behavior when studying or seeking to rehabilitate the paretic upper limb.