Functional implications of impaired control of submaximal hip flexion following stroke

Abstract

Introduction: We quantified submaximal torque regulation during low to moderate intensity isometric hip flexion contractions in individuals with stroke and the associations with leg function.

Methods: Ten participants with chronic stroke and 10 controls performed isometric hip flexion contractions at 5%, 10%, 15%, 20%, and 40% of maximal voluntary contraction (MVC) in paretic, nonparetic, and control legs.

Results: Participants with stroke had larger torque fluctuations (coefficient of variation, CV), for both the paretic and nonparetic legs, than controls (P < 0.05) with the largest CV at 5% MVC in the paretic leg (P < 0.05). The paretic CV correlated with walking speed (r2 = 0.54) and Berg Balance Score (r2 = 0.40). At 5% MVC, there were larger torque fluctuations in the contralateral leg during paretic contractions compared with the control leg.

Conclusions: Impaired low-force regulation of paretic leg hip flexion can be functionally relevant and related to control versus strength deficits poststroke.

Figure 1

Single participant data of hip flexion torque traces at the 5% load level. The relative magnitude of torque fluctuations (coefficient of variation) and error from the target torque (root mean square error) were calculated as performance measures at each load. Frequency content of the torque signal (power density spectrum) was also analyzed.

Figure 2

Group means ± standard error of the CV (ratio) for the paretic (black), non-paretic (grey) and control (white) legs at each of the load level conditions (percent of MVC). As a main effect, the magnitude of CV varied as follows: paretic > non-paretic > control. There was an interaction between the leg and load condition where the paretic leg CV was largest at the 5% load condition.

Figure 3

Percent area of frequency power for each load level in each frequency bin. Note that at the 5% load level there is less power in the lowest frequency bin and more power in the 7–15 Hz bin.

Figure 4

Group averages of the CV for the contralateral leg at the 5% load level. The control leg had significantly smaller CV as compared to the non-paretic and control legs (P < 0.05).

Figure 5

Associations between the paretic and non-paretic hip flexor CV at the 5% load and self-selected walking speed (A) and Berg Balance Score (B). Participants with larger torque fluctuations in their paretic leg walked slower and had worse balance (P < 0.05). The associations for the non-paretic leg were not significant (P > 0.05).