Using a simple rope-pulley system that mechanically couples the arms, legs, and treadmill reduces the metabolic cost of walking

Abstract

Background: Emphasizing the active use of the arms and coordinating them with the stepping motion of the legs may promote walking recovery in patients with impaired lower limb function. Yet, most approaches use seated devices to allow coupled arm and leg movements. To provide an option during treadmill walking, we designed a rope-pulley system that physically links the arms and legs. This arm-leg pulley system was grounded to the floor and made of commercially available slotted square tubing, solid strut channels, and low-friction pulleys that allowed us to use a rope to connect the subject's wrist to the ipsilateral foot. This set-up was based on our idea that during walking the arm could generate an assistive force during arm swing retraction and, therefore, aid in leg swing.

Methods: To test this idea, we compared the mechanical, muscular, and metabolic effects between normal walking and walking with the arm-leg pulley system. We measured rope and ground reaction forces, electromyographic signals of key arm and leg muscles, and rates of metabolic energy consumption while healthy, young subjects walked at 1.25 m/s on a dual-belt instrumented treadmill (n = 8).

Results: With our arm-leg pulley system, we found that an assistive force could be generated, reaching peak values of 7% body weight on average. Contrary to our expectation, the force mainly coincided with the propulsive phase of walking and not leg swing. Our findings suggest that subjects actively used their arms to harness the energy from the moving treadmill belt, which helped to propel the whole body via the arm-leg rope linkage. This effectively decreased the muscular and mechanical demands placed on the legs, reducing the propulsive impulse by 43% (p < 0.001), which led to a 17% net reduction in the metabolic power required for walking (p = 0.001).

Conclusions: These findings provide the biomechanical and energetic basis for how we might reimagine the use of the arms in gait rehabilitation, opening the opportunity to explore if such a method could help patients regain their walking ability.

Trial registration: Study registered on 09/29/2018 in ClinicalTrials.gov (ID-NCT03689647).

Keywords: Arms; Assistive device; Coordination; Energetics; Gait rehabilitation; Legs; Locomotion biomechanics; Walking.

Fig. 1

Arm-leg rope pulley system. Subjects walked on a split-belt force measuring treadmill while attached to a simple device that connects the ipsilateral arm and leg using a rope. The horizontal pulley bars are height adjustable, allowing for relative changes in rope length. Furthermore, the load cell is in series with the rope and used to measure rope tension during treadmill walking. Note that the reflective markers were attached to both sides while EMG sensors were placed only on the right side of the body due to a limited number of sensors available in our lab

Fig. 2

Mean ensemble forces demonstrating the mechanical demands during both normal (black line) and assisted (blue line) walking conditions. a During ~ 30–70% of the walking gait cycle, an assistive force was generated by the same-sided arm and leg rope connection (shaded gray area; n = 7). This assistive force is best understood as a forward force applied to the whole body. b In turn, this caused a decrease in propulsive and an increase in braking forces generated by the leg during assisted walking as compared to normal walking (n = 8). c The vertical GRF, which is required to support and accelerate the body, remained the same during both conditions (n = 8). Note that an assistive force was generated by the right arm and leg rope connection and another assistive force by the left arm and leg rope connection, but we only illustrate the assistive force generated by the right side

Fig. 3

Average electromyographic (aEMG) activity during assisted walking. We assessed a total of four upper and five lower limb muscles (mean ± SE; n = 8; see list of abbreviations). Each value is expressed relative to normal walking, representing a baseline of 100% (dashed black line). This data reveals a muscular shift characterized by greater arm and lesser leg muscle activity. Most notably, the arm’s triceps and biceps were the primary muscles to help transmit the assistive force onto the whole body during the propulsive phase, which reduced the leg’s medial gastrocnemius and soleus demand. * indicates a significant difference from baseline, p < 0.05

Fig. 4

Net metabolic power demand for walking. Using the arm-leg pulley system reduced the net metabolic power required to walk by 17% (mean ± SD, n = 8). Each line segment represents a subject, highlighting the observation that all subjects showed a reduction in net metabolic power (* indicates p < 0.05)

Fig. 5

Assisted walking. a During the propulsive phase, the right foot is planted onto the treadmill belt, which is moved backward. In turn, this movement pulled the right arm forward. The increased muscle activity of the biceps and triceps (bright red) helped stiffen the arm, which was necessary to transmit the force onto the whole body, eliciting a net forward force. The net forward force reduced the need for propulsion from the right leg and, therefore, reduced the muscular demand of the medial gastrocnemius and soleus (purple). b During the swing phase, the muscle activity of the arm’s biceps, triceps, and anterior deltoid (bright red) increased, but this activity did not coincide with a decrease in leg muscle activity. c During the braking phase, the right arm swung backward without causing any rope tension. At the same time, the left arm was helping to transmit the assistive force onto the whole body during the propulsive phase of the left leg. This forward force generated by the left arm and leg connection decreased the need for propulsion in the left leg but increased the demand for braking in the right leg. As such, an increase in the right tibialis anterior muscle activity occurred (bright red)