Time course of changes in neuromuscular properties following stroke.

To characterize the natural history of stroke effects on neuromuscular properties in elbow muscles, we tracked changes in elbow mechanical properties in hemiparetic stroke survivors after stroke. Using a parallel cascade system identification technique, we estimated intrinsic and reflex mechanical properties at 1, 2, 3, 6 and 12 months post stroke. At each time point, we examined neuromuscular changes during variations in mean elbow joint angle. Modulation of intrinsic and reflex properties was assessed using small amplitude pseudorandom positional perturbations at different mean elbow angles, over the entire range of motion. We identified two patterns of stroke effects on neuromuscular properties. In Group 1, intrinsic stiffness increased continuously after the stroke. In Group 2, it decreased continuously over this interval. Analogous results were recorded for reflex stiffness. These different recovery patterns may reflect the simultaneous emergence of two opposing mechanisms; i.e. brain recovery and secondary effects on neuromuscular properties. It follows that the progress of recovery may not reflect a single mechanism, and could depend on which mechanism is dominant at each time point.

Comparison of neuromuscular abnormalities between upper and lower extremities in hemiparetic stroke.

We studied the neuromuscular mechanical properties of the elbow and ankle joints in chronic, hemiparetic stroke patients and healthy subjects. System identification techniques were used to characterize the mechanical abnormalities of these joints and to identify the contribution of intrinsic and reflex stiffness to these abnormalities. Modulation of intrinsic and reflex stiffness with the joint angle was studied by applying PRBS perturbations to the joint at different joint angles. The experiments were performed for both spastic (stroke) and contralateral (control) sides of stroke patients and one side of healthy (normal) subjects. We found reflex stiffness gain (GR) was significantly larger in the stroke than the control side for both elbow and ankle joints. GR was also strongly position dependent in both joints. However, the modulation of GR with position was slightly different in two joints. GR was also larger in the control than the normal joints but the differences were significant only for the ankle joint. Intrinsic stiffness gain (K) was also significantly larger in the stroke than the control joint at elbow extended positions and at ankle dorsiflexed positions. Modulation of K with the ankle angle was similar for stroke, control and normal groups. In contrast, the position dependency of the elbow was different. K was larger in the control than normal ankle whereas it was lower in the control than normal elbow. However, the differences were not significant for any joint. The findings demonstrate that both reflex and intrinsic stiffness gain increase abnormally in both upper and lower extremities. However, the major contribution of intrinsic and reflex stiffness to the abnormalities is at the end of ROM and at the middle ROM, respectively. The results also demonstrate that the neuromuscular properties of the contralateral limb are not normal suggesting that it may not be used as a suitable control at least for the ankle study.

Mechanical abnormalities of the spastic ankle in chronic stroke subjects.

We examined the intrinsic and reflex contributions to ankle stiffness in people with chronic stroke and healthy subjects using the parallel system identification technique. Modulation of intrinsic and reflex stiffness was characterized by applying pseudorandom binary sequence (PRBS) perturbations to the ankle at different initial ankle joint over the entire range of motion (ROM). The experiments were performed for both paretic (stroke) and contralateral (control) side. Healthy (normal) subjects were used a secondary control. Reflex stiffness gain significantly increased in stroke than in control side at most positions. Intrinsic stiffness gain also increased significantly at dorsiflexing positions. These changes were position dependent. Thus, the abnormalities in intrinsic stiffness gain increased continuously from middle plantarflexion to full dorsiflexion while the major increase in reflex stiffness happened at the middle ROM. No significant changes were found in other intrinsic and reflex stiffness parameters. As compared to the normal ankle, the reflex stiffness gain of the control side was significantly larger, indicating that the control side is not normal. These findings demonstrate that both intrinsic and reflex stiffness contribute significantly to the mechanical abnormality associated with spastic ankle in hemiplegic stroke subjects. The results also suggest that the contralateral limb may not be used as a suitable control.