Walking in Minimalist Shoes Is Effective for Strengthening Foot Muscles

Sarah T. Ridge; Mark T. Olsen; Dustin A. Bruening; Kevin Jurgensmeier; David Griffin; Irene S. Davis; A. Wayne Johnson


Med Sci Sports Exerc. 2019;51(1):104-113. 

In This Article


Subjects. Subjects were experienced runners between the ages of 18 to 34 yr. Inclusion criteria consisted of an average running mileage between 15 and 30 miles each week for at least 6 months before study participation. Subjects were excluded if they had any lower extremity injuries within the 3 months before beginning the study or if they had run barefoot or in minimalist shoes more than three times within the previous 3 months. A health evaluation screening form was used to verify inclusion criteria.

A total of 65 participants were recruited and randomly assigned to one of three groups: minimalist shoe walking (MSW), foot strengthening (FS) exercise, and control (C). Each subject signed an informed consent document approved by the Human Subjects Review Board at Brigham Young University. Eight participants dropped out of the study due to the time commitment (n = 4) or injury unrelated to the study (n = 4). Participant demographics for the 57 participants that completed the study are detailed in Table 1. An a priori power analysis was run on muscle size changes from a previously published study.[22] An expected change of 0.3 cm2 with a power of 80% resulted in a sample size of 16 subjects per group.

Intervention. All runners, regardless of group assignment, maintained their prestudy mileage in traditional running shoes throughout their participation in the study. The MSW and FS groups included additional interventions. Participants assigned to the MSW group were given a pair of minimalist shoes (Inov-8 Bare XF 210 or 260) to wear in place of their typical daily footwear and a pedometer (Omron HJ-720ITFFP). Both shoe styles were zero drop with no midsole, differing only by the closure (the XF 210 had elastic laces, whereas the XF 260 had Velcro which resulted in an additional 50 g of weight). Over an 8-wk period, they gradually increased the number of walking steps they took in the minimalist footwear while reducing the number of steps taken in their typical footwear to maintain typical daily activity. Beginning with weeks 1 to 2, participants walked 2500 steps per day in the minimalist shoes. Weeks 3 to 4 increased to a daily step count of 5000, and ultimately ended weeks 5 to 8 walking 7000 steps daily. Participants were asked to achieve the step count at least 5 d·wk−1. At no time during the study were participants allowed to run in the minimalist shoes. Participants assigned to the FS group were taught a series of exercises developed at the Spaulding National Running Center designed to strengthen their intrinsic and extrinsic foot muscles (Table 2). Each week, they performed these exercises at least 5 d·wk−1—once in the laboratory and four times at home. In accordance with progressive resistance exercises, new variations of the exercises were taught at weekly laboratory visits. Running mileage was recorded by all participants in an online form. Steps taken in minimalist footwear or exercise adherence was recorded on the same online form for the MSW and FS groups, respectively.

Data collection. All participants met with investigators before beginning the study to sign an informed consent which was approved by Brigham Young University's Human Subjects Review Board. During that visit, they also filled out the health history form and a questionnaire to confirm that they met the inclusion criteria for the study. They were also taught how to perform the doming movement strength test (see below) to decrease the learning effect of performing a novel movement during the first day of testing.

Ultrasound imaging and strength testing were completed at the beginning of the study (week 0), and at the end of weeks 4 and 8. Ultrasound imaging was always completed first, to mitigate any acute effects the strength testing might have on the imaging results.

Imaging of the intrinsic and extrinsic foot muscles was performed by a single trained researcher (9 yr imaging experience) who was blinded to group assignment throughout the data collection. Ultrasound images were collected at a frequency of 10 to 12 MHz using a GE LogiqP6 ultrasound unit (GE Healthcare, USA) with a multilinear array soundhead (ML6—15 MHz). Intratester reliability for the researcher was measured before collection and found to be high (ICC, 0.98–0.99, SEM = 0.003–0.015 cm2; 0.014–0.015 cm), consistent with other similar reported studies.[24–26] The CSA or thicknesses of seven muscles were measured for each participant's right foot.

Four intrinsic foot muscles (the flexor hallucis brevis [FHB]), the ABDH, flexor digitorum brevis [FDB], and quadratus plantae [QP]) were imaged while the subject sat in a semireclined position with the knee supported on a pillow. The hip was abducted and externally rotated, the knee flexed to 90° and the ankle plantarflexed to 30°, allowing access to the plantar surface of the foot. The image of the FHB was recorded by aligning the probe with the shaft of the first metatarsal, using the head of the metatarsal as a bony landmark (Figure 1). Once the shaft and head of the first metatarsal were visualized concurrently with tendon of the flexor hallucis longus (FHL), the image was recorded. The thickness of the FHB was measured, perpendicularly to the FHL tendon, at 2 cm from the first metatarsal head between the outer border of FHL tendon and the shaft of the first metatarsal (Figure 1). The CSA of the ABDH was assessed transversely across the medial border of the foot aligned with the navicular tuberosity to ensure repeatability of the ABDH measurement. (Figure 1). After imaging the ABDH the probe was slid to the plantar surface of the foot and perpendicular to the sole of the foot while maintaining probe alignment with the navicular tuberosity, until the full CSA of the FDB and QP were visualized (Figure 1). The subject was asked to contract these muscles by gently flexing the toes and then returning to rest. A retrospective cine-loop was recorded, from which still images were selected for measurement purposes. The cine-loop provided a way to help determine fascia borders of FDB and QP for the later measurement of muscle CSA.

Figure 1.

Ultrasound images of each of the seven measured muscles.

To image the extrinsic lower leg muscles (tibialis anterior [TA], tibialis posterior [TP], and FDL), the leg was extended and the thigh was supported on a pillow with the knee slightly flexed thus unweighting the subjects calf. The CSA of the FDL was measured with the ultrasound probe held transversely at a distance of 50% from the medial knee joint line to medial malleolus. The subject was again asked to gently flex the toes to visualize the borders of the FDL fascia. The thicknesses of the TA and TP were measured at 30% of the distance from the lateral knee joint line to the lateral malleolus with the ultrasound probe positioned transversely to the leg. First, the TA was visualized with sufficient gel and light pressure to limit variability in the measurement due to compressing the superficial muscle (Figure 1). Next, the deeper TP was visualized deep to the interosseous membrane. The subjects were asked to gently invert their foot and then to relax to produce contraction of the TP and help with visualization of the fascial borders of the muscle. This contraction was record with a cine-loop for later analysis. Images were captured when the interosseus membrane appeared horizontal across the ultrasound monitor (Figure 1).

Ultrasound measurements were performed by a separate researcher who was also blinded to group assignment. The manufacturer supplied ultrasound software was used to measure CSA or thickness for each muscle. Measurements from the two recorded images for each muscle were then averaged for statistical analysis.

Strength testing. Three measures of foot muscle strength were collected: GT, lateral toes (LT) flexion, and doming. All measurements were conducted on custom-built apparatuses using reliable techniques (GT ICC, 0.903; GT SEM, 0.703 kg; LT ICC, 0.924; LT SEM, 0.655 kg; doming ICC, 0.949; doming SEM, 0.883 kg).[27] The apparatuses and data collection procedures were described in detail previously.[27] To ensure repeatable foot placement and alignment with the dynamometer used for toe flexion, during the first day of testing, each subject's foot was traced onto a piece of paper which rested on top of the platform. During subsequent testing sessions, the foot was realigned with the tracing on the paper. Position settings for the foot and dynamometer used for doming strength testing were also recorded during the initial testing session and used for all subsequent testing.

Peak force measurements were determined using a custom LabView (National Instruments, Inc., Austin, TX) program. A trained researcher visually inspected each force curve, choosing the peak of the largest plateau during the trial. Six data points around the peak were averaged to generate an average peak force for the trial. Peak forces from three trials of each test were averaged together for use in the statistical analysis.

Statistical analysis. Mixed model ANOVA (three groups × three times) were run for each dependent variable to determine if groups changed differently over time (interaction effect). Main effects for group or time were not reported independently, as these were not of interest in this study. When there was a significant interaction between group and time, post hoc repeated-measures ANOVA were run on each group separately to determine if there were significant increases in muscle size and strength between weeks 0, 4, and 8. To determine if any increases were significantly different between groups, the changes in each variable between weeks 0 to 4, 4 to 8, and 0 to 8 were analyzed using ANOVA. Significance for all ANOVA was set at α = 0.05. The Benjamini–Hochberg method of correcting for multiple comparisons was used with a false discovery rate of 0.10.[28]