Intrinsic foot muscles have the capacity to control deformation of the longitudinal arch

Abstract

The human foot is characterized by a pronounced longitudinal arch (LA) that compresses and recoils in response to external load during locomotion, allowing for storage and return of elastic energy within the passive structures of the arch and contributing to metabolic energy savings. Here, we examine the potential for active muscular contribution to the biomechanics of arch deformation and recoil. We test the hypotheses that activation of the three largest plantar intrinsic foot muscles, abductor hallucis, flexor digitorum and quadratus plantae is associated with muscle stretch in response to external load on the foot and that activation of these muscles (via electrical stimulation) will generate sufficient force to counter the deformation of LA caused by the external load. We found that recruitment of the intrinsic foot muscles increased with increasing load, beyond specific load thresholds. Interestingly, LA deformation and muscle stretch plateaued towards the maximum load of 150% body weight, when muscle activity was greatest. Electrical stimulation of the plantar intrinsic muscles countered the deformation that occurred owing to the application of external load by reducing the length and increasing the height of the LA. These findings demonstrate that these muscles have the capacity to control foot posture and LA stiffness and may provide a buttressing effect during foot loading. This active arch stiffening mechanism may have important implications for how forces are transmitted during locomotion and postural activities as well as consequences for metabolic energy saving.

Keywords: electromyography; foot stiffness; multi-segment foot model.

Figure 1.

Experimental set-up. Foot motion, ground reaction forces and intramuscular electromyography were recorded during incremental loading (experiment 1) and independent electrically evoked contractions of the three major plantar intrinsic foot muscles (experiment 2). Loads ranging from 0% to 150% of body mass were added to a loading device, which was secured to the distal aspect of the participants right thigh. The participant's foot was placed on the centre of a force plate and four motion analysis cameras were positioned to record three-dimensional motion of the shank and two individual foot segments during each task.

Figure 2.

Location of electrodes within the intrinsic foot muscles. Schematic of the anatomical location of (a) abductor hallucis (AH), (b) flexor digitorum brevis (FDB) and (c) quadratus plantae (QP) from the plantar aspect of a right foot. Fine wire pairs of electromyography (EMG) electrodes (black lines with hooked ends) were inserted under ultrasound guidance, with one pair being inserted proximally and one pair distally to the muscle belly. The proximal electrode pair was used for the EMG recordings in experiment 1, whereas one wire from each of the proximal and distal pairs was connected to a constant current electrical stimulator, which delivered trains of electrical stimulation to each muscle independently in experiment 2. (Online version in colour.)

Figure 3.

Multi-segment foot model marker locations. Retro-reflective skin markers for three-dimensional motion capture were applied to the right foot of each subject over defined anatomical landmarks in order to create a multi-segment foot model [1]. The figure depicts views from the anterior (a), medial (b) and lateral (c) aspects of the right foot. Markers were attached to rigid plastic discs that were secured to the skin with double-sided adhesive tape.

Figure 4.

Group means ± s.d. for (a) change in longitudinal arch (LA) height, (b) change in muscle tendon unit length and (c) normalized electromyographic (EMG) root mean square (r.m.s.) plotted as a function of load applied to the thigh during the incremental loading task. For each participant, muscle length and arch height were normalized to the resting unloaded values. The EMG r.m.s. amplitude was normalized to the maximal value recorded during the 150% body mass trial. Open circles (red) represent abductor hallucis, open squares (blue) represent flexor digitorum brevis and open triangles (green) represent quadratus plantae muscle. (Online version in colour.)

Figure 5.

(a) Diagram of the measurements of longitudinal arch (LA) length and height. (b) Group mean ± s.e. for LA length and height with 50% (open) and 100% (filled) body mass loading for abductor hallucis (AH, red), flexor digitorum brevis (FDB, blue) and quadratus plantae (QP, green) muscles. LA length and height values are shown in response to loading (squares) and stimulation (circles). Length and height of the LA are presented as a percentage change from the resting unloaded LA values (mean unloaded LA length = 156.7 ± 18.2 mm, mean unloaded LA height = 53.5 ± 4.7 mm). Stimulation of AH, FDB and QP resulted in a significant reduction in LA length and increase in LA height for all conditions (all p ≤ 0.05). (Online version in colour.)

Figure 6.

Depiction of foot motion changes occurring owing to stimulation of abductor halluces (AH). The position of the foot segments under load is represented by the grey-shaded image and the stimulated position is represented by the red outlined image. The movements include (a) calcaneal extension and metatarsal flexion in the sagittal plane (b) calcaneal abduction and metatarsal adduction in the axial plane and (c) calcaneal inversion in the frontal plane. This combination of segment movements lead to a reduction in length and an increase in height of the longitudinal arch. (Online version in colour.)

Figure 7.

Changes in calcaneal and metatarsal segment angles owing to passive loading and intrinsic foot muscle stimulation. Group means ± s.e. for changes in calcaneal and metatarsal segment angles owing to loading, 50% (open) and 100% (filled) body mass as well as the subsequent changes in segment angles occurring with stimulation of abductor hallucis (AH, red), flexor digitorum brevis (FDB, blue) and quadratus plantae (QP, green) muscles. Segment angles are shown in response to loading (squares) and stimulation (circles). Angular rotations are defined relative to the laboratory coordinate system (x-lateral, y-anterior, z-upward) and according to an x–y–z Cardan sequence of rotations, with extension–flexion (positive extension) as the rotation about the x-axis, inversion–eversion (positive inversion) as the rotation about the y-axis and abduction–adduction (positive adduction) as the rotation about the z-axis. Segment angles are normalized to the seated, unloaded segment angle, such that zero degrees equals the unloaded segment angle for all axes. β indicates significant effect of load (100% versus 50% body mass) on segment angle. Asterisk indicates significant change in segment angle owing to muscle stimulation. (Online version in colour.)

Figure 8.

Changes in centre of pressure (COP) position owing to intrinsic foot muscle stimulation. Mean ± s.e. for COP in the mediolateral (COP_ML, x-coordinate) and anteroposterior (COP_AP, y-coordinate) directions occurring owing to electrically evoked contractions in abductor hallucis (red circle), flexor digitorum brevis (blue square) and quadratus plantae (green triangle) with both 50% (open) and 100% (filled) loading conditions. Changes in COP position were calculated by subtracting the COP position immediately prior to stimulation from the subsequent maximum COP displacement that occurred during muscle stimulation, such that 0,0 (x,y) represents the COP position prior to muscle stimulation, for all conditions. Stimulation of AH, FDB and QP produced significant changes in COP position in both loading conditions (all p ≤ 0.05). (Online version in colour.)