Journal of Applied Biomechanics, 2014, 30, 501-507 http://dx.doi.org/10.1123/jab.2013-0039 © 2014 Human Kinetics, Inc.
An Official Journal of ISB www.JAB-Journal.com ORIGINAL RESEARCH
Gait Biomechanics of a Second Generation Unstable Shoe Jacob K. Gardner,1 Songning Zhang,2 Max R. Paquette,3 Clare E. Milner,4 and Elizabeth Brock5 1Biola
University; 2University of Tennessee; 3University of Memphis; 4Drexel University; 5University of Oregon
The recent popularity of unstable shoes has sparked much interest in the efficacy of the shoe design. Anecdotal evidence suggests that earlier designs appear bulky and less aesthetically appealing for everyday use. The purpose of this study was to examine effects of a second generation unstable shoe on center of pressure (COP), ground reaction force (GRF), kinematics, and kinetics of the ankle joint during level walking at normal and fast speeds. In addition, findings were compared with results from the first generation shoe. Fourteen healthy males performed five successful level walking trials in four testing conditions: walking in unstable and control shoes at normal (1.3 m/s) and fast (1.8 m/s) speeds. The unstable shoe resulted in an increase in mediolateral COP displacement, first peak vertical GRF loading rate, braking GRF, ankle eversion range of motion (ROM), and inversion moment; as well as a decrease in anteroposterior COP displacement, second peak vertical GRF, ankle plantarflexion ROM, and dorsiflexion moment. Only minor differences were found between the shoe generations. Results of the generational comparisons suggest that the lower-profile second generation shoe may be as effective at achieving the desired unstable effects while promoting a smoother transition from heel contact through toe off compared with the first generation shoe. Keywords: biomechanics, unstable, kinematics, kinetics, toning The recent popularity of unstable shoes has sparked much interest into the efficacy of the unstable shoe design. The anteroposterior curved sole of unstable shoes is thought to make the shoes less stable than standard shoes. It has been proposed that the instability of the shoe would provide a training effect for muscles of the lower extremity that are important for maintaining static and dynamic postural stability during walking.1,2 These anticipated outcomes have spawned the terms toning shoes or fitness shoes to describe the functional roles of the shoe design. However, not enough research evidence exists to substantiate the claims of these types of shoes. The majority of published research has been done on the Masai Barefoot Technology (MBT) (Swiss Masai, Gossau, Switzerland) shoes.1–10 More recently, several other manufacturers have introduced their own models (eg, Skechers Shape-ups, Manhattan Beach, California; Dockers Active Balance, San Francisco, California; Reebok EasyTone, Canton, Massachusetts; New Balance True Balance, Boston, Massachusetts; and others). Thus, more research aimed to study other shoe designs and models is needed. Biomechanics studies have shown that MBT shoes can acutely alter spatiotemporal variables during gait.1,2,6 Walking in MBT shoes at a self-selected pace has resulted in decreased stride length and time, decreased step length, and decreased walking speed compared with a control shoe. It should be noted that the reported reductions Jacob K. Gardner is with the Department of Kinesiology, Health, and Physical Education at Biola University, La Mirada, CA. Songning Zhang is with the Department of Kinesiology, Recreation, and Sport Studies at the University of Tennessee, Knoxville, TN. Max R. Paquette is with the Department of Health and Sport Sciences at the University of Memphis, Memphis, TN. Clare E. Milner is with the Department of Physical Therapy and Rehabilitation Sciences at Drexel University, Philadelphia, PA. Elizabeth Brock is with the Lundquist College of Business at the University of Oregon, Eugene, OR. Address author correspondence to Jacob K. Gardner at [email protected].
in spatiotemporal characteristics are likely due to a decreased selfselected walking speed in MBT shoes. A recent study showed no significant spatiotemporal changes during gait between a control shoe and an MBT shoe when walking speed was controlled.11 Therefore, it is important to account for walking speed when making comparisons among shoes. Kinematic changes at the ankle have been detected while wearing MBT shoes; most notably, an increased dorsiflexion angle at heel strike,1,2 and either increased11 or decreased2 dorsiflexion in late stance have been observed. Our previous work has shown significant kinematic changes between a business casual unstable shoe (Active Balance; Genesco Inc., Nashville, Tennessee) and a standardized control shoe (Glacier; Genesco Inc., Nashville, Tennessee) during level walking in healthy men.12 The significant kinematic changes included an increased ankle dorsiflexion angle at heel strike and a decreased plantarflexion range of motion (ROM) during early stance. In addition, we also showed greater ankle eversion ROM in the unstable shoes. In addition to kinematics, a few studies have reported alterations in kinetics while walking in unstable shoes.11,12 Our previous study showed that despite similar walking speeds, the use of unstable shoes resulted in increased mediolateral center of pressure (COP) displacement and decreased anteroposterior COP displacement, decreased first peak vertical ground reaction force (GRF), and decreased ankle dorsiflexion moment.12 Taniguchi et al11 also found a significant decrease in ankle plantarflexion moment during late stance. In addition, our previous work also found increased loading rate of first peak vertical GRF, increased peak braking GRF, and increased peak ankle inversion moment in the unstable shoe.12 Recently, anecdotal evidence suggests that earlier designs appear too bulky and less aesthetically appealing for everyday use such as in business and office settings. It is important to be able to wear the shoe during everyday tasks, as researchers have suggested that unstable shoes may be beneficial for long-term use.4 For these reasons, the original unstable shoe (from Zhang et al12) was 501
Downloaded by W VA Univ Library on 09/16/16, Volume 30, Article Number 4
502 Gardner et al.
redesigned with a lower profile and a dressier style for daily wear in work environments. Therefore, the purpose of this study was to examine the effect of the redesigned (second generation) unstable shoe on COP, GRF, kinematics, and kinetics of the ankle joint during level walking at normal and fast speeds. We also wanted to compare the findings with the results from the first generation shoe. The aim was to determine if similar biomechanical changes compared with a control shoe would be maintained in the lower-profile second generation design. Based on the results of our previous study,12 we hypothesized that wearing the unstable shoes would result in a decrease in first peak vertical GRF, anteroposterior COP displacement, and ankle plantar flexion ROM; as well as an increase in mediolateral COP displacement, loading rate of first peak vertical GRF, peak braking GRF, ankle inversion ROM, and ankle inversion moment when compared with the control shoe. We also hypothesized that the biomechanical differences found to be significant in the first generation unstable shoe compared with the control shoe would be similar in the second generation shoe.
a
Methods Participants Fourteen healthy males participated in this study (age: 45.6 ± 8.4 y; height: 1.81 ± 0.07 m; weight: 80.6 ± 9.4 kg; BMI: 25.1 ± 2.8). All participants were free from any gait abnormality, neurologic disorders, or major surgeries, and free from any lower extremity injury for at least 6 months before the time of the study. All subjects provided written informed consent approved by the institutional review board before testing.
Instrumentation
b
A 9-camera three-dimensional (3D) motion capture system (240 Hz; Vicon Motion Systems, Oxford, United Kingdom) and a force platform (1200 Hz; Advanced Mechanical Technology Inc., Watertown, Massachusetts) were used to collect 3D kinematic and ground reaction force data, respectively. Retro-reflective anatomical markers were placed bilaterally on the acromion processes, iliac crests, greater trochanters, medial and lateral epicondyles, medial and lateral malleoli, and first and fifth metatarsal heads.13 Four tracking markers attached to rigid thermoplastic shells were placed bilaterally on the thighs and legs as well as on the pelvis and trunk. Foot tracking markers were placed directly on the posterior and lateral aspects of the heel-counter of the shoes. Kinematic and GRF data were recorded simultaneously using the Vicon Nexus software.
Figure 1 — Control (a) and unstable (b) shoe used for the study.
Experimental Protocol Before testing, participants walked in an indoor hallway for 5 minutes in a flat bottom control shoe (Gordon; Genesco Inc., Nashville, Tennessee; Figure 1a) and 5 minutes in an unstable curved bottom shoe (Active Balance, Second Generation; Genesco Inc., Nashville, Tennessee; Figure 1b) to familiarize themselves with both pairs of shoes. Before data collection, retro-reflective markers were placed on the participants as described above for 3D motion capture. Participants wore athletic clothing so that anatomical markers could be placed directly on the skin. Two separate calibration trials were performed for each shoe tested to define model segments and joint centers. During testing, each participant performed five successful
level walking trials in each of four testing conditions: walking in control and unstable at a normal walking speed of 1.3 m/s and a fast speed of 1.8 m/s. The walking speed was monitored by two pairs of photo cells and an electronic timer. A successful trial was defined as a trial in which the participant contacted the force platform with their right foot without targeting and while maintaining walking speeds within 5% of the desired speed. To reduce bias, the order of conditions was randomized first by shoe (control or unstable) and then by walking speed (1.3 m/s or 1.8 m/s). Each participant was given ample time to practice walking in each shoe and speed condition before data collection to become familiar with the testing protocol.
Data Processing and Analysis 3D kinematic and kinetic data were processed and analyzed using Visual 3D software suite (C-Motion, Inc., Germantown, Maryland). Kinematic and kinetic data were filtered with a second order lowpass digital Butterworth filter at 6 Hz and 50 Hz, respectively. Joint angles were calculated using an X-Y-Z Cardan rotation sequence. The conventions of joint kinematics and kinetics were determined using the right-hand rule, and all joint moments were calculated as internal moments. In addition, all GRF were normalized to each participant’s body weight (BW), and joint moments were normalized to body mass (N·m/kg). Customized computer programs (VB_V3D and VB_Table) were used to determine peaks in the curves of the
Downloaded by W VA Univ Library on 09/16/16, Volume 30, Article Number 4
Second Generation Unstable Shoe 503
3D kinematic and kinetic variables of interest from the output of Visual 3D and were also used to compute additional parameters and organize the variables for statistical analyses. The average of the five trials of the selected variables was used in subsequent statistical analysis. All variables of interest were analyzed during the stance phase of gait; from heel strike to toe off. Heel strike was determined when the vertical GRF magnitude reached a threshold of 10 newtons (N), and toe off was determined when the magnitude dropped below a threshold of 10 N. The kinematic and kinetic variables we chose to report were selected based on the previous study of the first generation rocker bottom shoe,12 which allowed us to make comparisons between the two shoe generations. To quantify shoe characteristics, the slope of the heel and forefoot regions of both the control and unstable test shoes were calculated using the following equations where SFT = forefoot slope percentage; Lft = forefoot slope length; f = forefoot length; SHL = heel slope percentage; Lhl = heel slope length; r = heel length; aFT = forefoot slope angle; t = toe height; aHL = heel slope angle; and h = heel height (Figure 2):12
SFT =
L ft L × 100 S HL = hl × 100 (1) f r
a FT =
t f
a HL =
h r
(2)
These measurements were also performed on the first generation shoe and its corresponding control shoe for generational comparisons. All measurements for the first and second generation shoes, as well as the control shoes, were computed as an average for shoe sizes 8, 9, and 11. A Shapiro-Wilk normality test was performed on each variable of interest. All variables were found to be normally distributed. Thus parametric statistical tests were performed for determining differences among conditions. For comparisons between the second generation shoe and its respective control shoe, a 2 × 2 (shoe × speed) repeated measures analysis of variance (ANOVA) was used to determine any differences between kinematic and kinetic variables (SPSS version 18.0; SPSS Inc., Chicago, Illinois). An alpha level of .05 was set a priori. When a shoe × speed interaction was found, a post hoc comparison using a paired samples t test with a Bonferroni adjustment for multiple comparisons was conducted to detect differences between shoes and/or speeds. In addition, for comparisons
Figure 2 — Diagram of measurement parameters used to compare the shoes.
between first12 and second generation shoes, subject demographic and select variables that were statistically different from their respective control shoes for each study were extracted and used for comparison (original data from the previous study conducted in our laboratory were obtained). The differences between the two shoe generations were compared using an independent samples t test for each variable of interest. Only shoe main effects in the normal walking speed were used for comparison. It is important to note that while the two studies were similar in methodology, the control shoe used for the first generation study was not the same control shoe used in the current study. In addition, the subject sample was not the same for each study, although there were two overlapping subjects. Independent samples t tests were performed on subject demographics excluding the two overlapped subjects and revealed that the age, height, mass and BMI were not different between the two studies.
Results For comparisons between the second generation shoe and the control shoe, the ANOVA results showed significant speed main effects for all COP and GRF variables (P < .001) except for mediolateral COP displacement (P = .277, Table 1). The unstable shoe showed
Table 1 Center of pressure and ground reaction force variables, mean ± SD Control Shoe
Unstable Shoe
Shoe
Speed
Interaction
Normal Speed
Fast Speed
Normal Speed
Fast Speed
P
P
P
Mediolateral COP displacement (m)a
–.05 ± .02
–.05 ± .02
–.06 ± .02
–.06 ± .02
.044
.277
.901
Anteroposterior COP displacement (m)a,b
.20 ± .04
.19 ± .04
.17 ± .03
.16 ± .03
< .001
< .001
.354
First peak vertical GRF (BW)b
1.18 ± .06
1.38 ± .07
1.17 ± .07
1.37 ± .07
.430
< .001
.855
Second peak vertical GRF (BW)a,b
1.14 ± .05
1.23 ± .08
1.11 ± .05
1.20 ± .07
.001
< .001
.963
Peak vertical GRF loading rate (BW/s)a,b
7.52 ± 1.05
11.48 ± 1.97
8.42 ± 2.09
12.24 ± 2.40
.024
< .001
.632
Peak braking GRF (BW)a,b
–.22 ± .02
–.29 ± .04
–.23 ± .02
–.30 ± .04
.012
< .001
.416
Peak propulsive GRF (BW)b
.23 ± .03
.31 ± .05
.24 ± .04
.31 ± .04
.693
< .001
.416
Variables
Abbreviations: COP, center of pressure; GRF, ground reaction force. a P < .05 = significant shoe main effect. b P < .05 = significant speed main effect.
Downloaded by W VA Univ Library on 09/16/16, Volume 30, Article Number 4
504 Gardner et al.
increased mediolateral COP displacement (P = .044), decreased anteroposterior COP displacement (P < .001), increased loading rate of the first peak vertical GRF (P = .024), decreased second peak vertical GRF (P = .001), and increased peak braking GRF (P = .012) compared with the control shoe. Ankle angle ensemble (average) curves for the sagittal and frontal planes during the normal and fast walking speeds are presented in Figures 3 and 4, respectively. A speed main effect was found for all ankle kinematic variables with the exception of ankle plantar flexion ROM (measured from contact to the peak plantar flexion angle, Table 2). In terms of shoe main effects, the unstable shoe showed an approximate 6° decrease in plantar flexion ROM (P < .001) and a decrease in total ankle joint ROM (P < .001). There
was also an approximate 2° increase in eversion ROM in the normal (P = .007) and fast (P = .001) walking speeds. Ankle moment ensemble (average) curves for the sagittal and frontal planes during the normal and fast walking speeds are presented in Figure 5 and 6, respectively. Speed main effects were significant for all ankle joint kinetic variables except for peak ankle inversion moment (Table 2). In terms of shoe main effects, the unstable shoe showed an approximate 40% decrease in peak ankle dorsiflexion moment (P < .001) and a 34% increase in peak ankle inversion moment (P = .003). Anteroposterior COP displacement was greater in the second generation shoe compared with the first generation shoe (P < .001, Table 3). In addition, the first peak vertical GRF was decreased in the second generation shoe compared
Figure 3 — Ensemble (average) ankle angle curves in the sagittal (top) and frontal (bottom) planes for the control (solid black line) and the unstable (dashed gray line) shoes at the normal walking speed.
Figure 4 — Ensemble (average) ankle angle curves in the sagittal (top) and frontal (bottom) planes for the control (solid black line) and the unstable (dashed gray line) shoes at the fast walking speed.
Table 2 Ankle kinematic and kinetic variables, mean ± SD Control Shoe Variable Sagittal contact angle (°)a PF ROM
(°)b
Unstable Shoe
Shoe
Speed Interaction
Normal Speed
Fast Speed
Normal Speed
Fast Speed
P
P
P
.21 ± 4.03
2.12 ± 4.31
–.44 ± 3.44
.85 ± 3.76
.228
< .001
.237
–14.49 ± 3.47 –15.35 ± 2.92
–8.85 ± 3.42
–9.21 ± 2.58
< .001
.113
.176
Peak DF angle (°)a
11.19 ± 3.29
9.64 ± 3.37
9.87 ± 4.02
8.13 ± 4.80
.088
.006
.629
Stance phase joint ROM (°)a,b
25.47 ± 3.63
22.86 ± 4.53
19.28 ± 4.01
16.49 ± 3.98
< .001
< .001
.745
–5.59 ± 3.11
–4.47 ± 2.59
–6.99 ± 3.55
–6.71 ± 3.82
.001
.015
.025
.38 ± .05
.49 ± .07d
.22 ± .05e
.31 ± .07d,e
< .001
< .001
.050
–1.39 ± .09
–1.51 ± .15
–1.37 ± .12
–1.48 ± .15
.197
< .001
.819
.11 ± .05
.12 ± .05
.15 ± .04
.16 ± .05
.003
.160
.930
Eversion ROM
(°)a,b,c
Peak DF moment (N⋅m/kg)a,b Peak PF moment
(N⋅m/kg)a
Peak inversion moment
(N⋅m/kg)b
Abbreviations: PF, plantarflexion; ROM, range of motion; DF, dorsiflexion. a P < .05 = significant speed main effect. b P < .05 = significant shoe main effect. c P < .05 = significant interaction of shoe and speed. d Significant difference between speeds for same shoe in post hoc comparison. e Significant difference between shoes at same speed in post hoc comparison.
Downloaded by W VA Univ Library on 09/16/16, Volume 30, Article Number 4
Second Generation Unstable Shoe 505
Figure 5 — Ensemble (average) ankle moment curves in the sagittal (top) and frontal (bottom) planes for the control (solid black line) and the unstable (dashed gray line) shoes at the normal walking speed.
Figure 6 — Ensemble (average) ankle moment curves in the sagittal (top) and frontal (bottom) planes for the control (solid black line) and the unstable (dashed gray line) shoes at the fast walking speed.
Table 3 Ground reaction force and ankle joint variable differences between shoe generations, mean ± SD ML COP Disp. (m)
AP COP Disp. (m)a
First Peak Vertical GRF (BW)a
Vertical GRF Loading Rate (BW/s)
Peak Braking GRF (BW)
Second Peak Vertical GRF (BW)
Eversion ROM (°)
Peak Inversion Moment (N⋅m/kg)
First generation
–.06 ± 0.02
.18 ± 0.04
1.16 ± .08
8.27 ± 2.10
–.23 ± .02
1.11 ± .05
–6.87 ± 3.45
.14 ± .04
Second generation
–.07 ± 0.02
.26 ± 0.02
1.11 ± .05
7.69 ± 1.97
–.21 ± .02
1.08 ± .03
–8.81 ± 2.34
.14 ± .05
.651
< .001
.042
.447
.093
.057
.091
.729
P value
Abbreviations: ML, mediolateral; COP, center of pressure; Disp., displacement; AP, anteroposterior; GRF, ground reaction force; ROM, range of motion. Note. First generation unstable shoe results from Zhang et al 2012. Second generation unstable shoe results from the current study. a P < .05 = significant difference between shoe generations.
with the first generation shoe (P = .042). No other variables were significantly different between the two shoe generations. Dimensional characteristics of the first and second generation shoes, along with their respective control shoes, are shown (Table 4). In the second generation shoe, the heel slope angle was 13.7° larger and the forefoot slope angle was about 2.5° larger than the control shoe. In comparing the two unstable shoe generations, the first generation shoe had larger heel and forefoot slope percentages compared with the second generation shoe. In addition, the heel slope angle was greater in the first generation shoe, while the forefoot angle was smaller compared with the second generation shoe (Table 4).
Discussion One of the purposes of this study was to determine the biomechanical differences between a low-profile unstable shoe and a standard dress shoe during level walking at two different speeds. Several studies have examined biomechanical characteristics of gait in unstable shoes. However, this is the first study, to our knowledge, that has analyzed a low-profile unstable shoe design. The second
Table 4 Shoe measurements of the first and second generation unstable shoes and their respective control shoes αHL
αFT
23.4°
17.1°
104.0% 102.9%
8.1°
12.7°
Second generation unstable shoe 104.4% 102.8%
13.7°
18.9°
0°
16.4°
SHL
SFT
First generation unstable shoe
108.8% 105.5%
First generation control Second generation control
100.0% 103.0%
Abbreviations: SHL, heel slope percentage; SFT, forefoot slope percentage; αHL, heel slope angle; αFT, forefoot slope angle.
purpose was to compare two generations of the unstable shoes from the same manufacturer to detect gait biomechanical differences as a result of changes in the shoe designs. When compared with the control shoe, the second generation unstable shoe resulted in a 19% significant increase in mediolateral COP displacement at normal walking speed. An increase in the mediolateral COP displacement is consistent with our previous findings.12 This increased mediolateral COP movement was thought
Downloaded by W VA Univ Library on 09/16/16, Volume 30, Article Number 4
506 Gardner et al.
to be related to the softer midsole materials in the heel region of the unstable shoes.12 We also found a decrease in the anteroposterior COP displacement in the unstable shoe compared with the control shoe. The decreased COP movement is supported by an increase of 15° and 4.5% in the heel slope angle and heel slope percentage compared with the control shoe. The decreased anteroposterior COP displacement may indicate that wearing the unstable shoes results in a more anterior heel contact upon foot strike.12 This may also explain why the loading rate of the first peak vertical GRF was increased in the unstable shoe when compared with the control shoe, and may have caused the increased peak braking GRF during weight acceptance.12 Even though the loading rate of the first peak vertical GRF was increased in the unstable shoe, there was no difference in the first peak vertical GRF between the unstable and control shoes. However, the second peak (push-off peak) did decrease in the unstable shoes. The curvature of the forefoot region of the unstable shoes may contribute to the reduced second peak vertical GRF. This finding does not affect propulsion; as no difference was found in peak propulsive GRF between the shoes. The result is in agreement with the finding in our previous study of the first generation unstable shoe.12 Furthermore, this finding was also supported by the lack of difference in the peak plantar flexion moment between the shoes during push-off. In addition, anecdotal reports from our participants suggest that the unstable shoe did not feel different than the control shoe after the initial familiarization period. In terms of ankle joint kinematics, the decreased dorsiflexion moment in the unstable shoe during weight acceptance is consistent with the findings of the MBT unstable shoe studied by Taniguchi et al.11 The peak plantar flexion moment during late stance was not reduced in the unstable shoe compared with the control shoe, which is different from the result of the previous study.12 This lack of difference could be related to the reduced profile in the second generation shoe. An even greater reduction of 21% in peak ankle plantar flexion moment was found by Taniguchi et al.11 This decrease may be related to the slightly, but nonsignificant, slower walking speed (self-selected) in the MBT in the study. In addition, the participants in the study by Taniguchi et al11 were trained for 30 minutes before testing on how to properly walk in the unstable shoes. Our participants, on the other hand, were provided with a 5 minute familiarization period. Whether the amount of initial exposure to the shoes has an effect on biomechanical variables during a one-time walking session remains unclear. However, Nigg et al1 did not show any significant difference of plantar flexion impulse between the MBT and a running shoe in walking after a 2-week accommodation period. Finally, since Taniguchi et al11 only reported sagittal plane joint variables, it is not clear if the increased peak inversion moment found in our study is also observed in MBT shoes. When comparing biomechanical variables between unstable shoe generations, the decreased anteroposterior COP displacement in the current study is consistent with the results from our first generation unstable shoe study.12 However, the displacement was significantly greater in the second generation design. The greater anteroposterior COP displacement was likely a result of the smaller heel angle of the shoe sole in the second generation shoe. This was the largest difference between the two shoe designs and was almost 10° less in the second generation shoe (Table 4). The GRF variables were similar between shoe generations, but a significant difference was found in the first peak vertical GRF, as the second generation shoe showed a 0.05 BW smaller peak vertical GRF compared with its first generation counterpart. While this difference is small, it may
be more desirable for load reduction during the weight acceptance phase of gait. The ankle kinematic and kinetic parameters showed no statistical differences between shoe generations. This is an important finding as it shows that the second generation shoe still provides similar unstable effects as the first generation shoe, even though it has a lower-profile design. Besides aesthetics, one of the reasons for changing the shoe design was to reduce the profile to elicit a smoother transition between heel strike and toe off. Based on the dependent variables selected, the findings from the lower-profile construction of the second generation shoe did not greatly differ from that of its first generation counterpart. The largest difference between the two shoe generations was the slope of the heel, which was decreased by about 4.4% in the second generation shoe and was accompanied by a 9.7° decrease in heel slope angle. As mentioned previously, the difference in heel slope was likely the reason for the decreased first peak vertical GRF between the two shoes. Interestingly, no change occurred in the peak braking GRF even though the heel slope was smaller. It is unclear if the small but significant changes noted in this study would produce clinically meaningful outcomes. To achieve a training effect, one would expect increases in joint ROM and/or joint moments (ie, increased muscular effort). The decreases in sagittal plane ROM and moments in this study suggest a possible decrease in training effect, which is just the opposite of what we would initially expect to see. However, in the frontal plane, the increase in eversion ROM and inversion moment, while small, may be beneficial for mediolateral movement and postural control (ie, sway). Thus, it is possible that wearing unstable shoes on a regular basis might help to strengthen the lower extremity muscles such as the tibialis anterior and peroneus longus for ankle eversion movement and aid in ankle contribution to sway control; however, it would probably not aid in strengthening antigravity muscles (eg, gastrocnemius and soleus) responsible for lower extremity movements in sagittal plane during gait. Our results and the results of a previous study,1 suggest that subjects may eventually get accustomed to the shoes and effects may diminish over time. It is possible that unstable shoes may be advantageous for variables other than muscular training. One study showed that when subjects with knee osteoarthritis wore the unstable shoes they saw a decrease in total knee pain after 12 weeks of use.4 However, the control group wearing standard shoes also saw a decrease in knee pain and there was no statistical difference in total pain between groups. In addition, pain during walking was reduced in both groups, but the control group experienced a slightly larger reduction compared with the test group. Thus, it is difficult to determine if there was a true cause and effect relationship between the shoes and symptom relief in the patient population. More prospective studies with focus on effects on secondary planes are needed to determine the effectiveness of the unstable shoe as a training device. It should be noted that since the shoes used in this study were designed for use by men only, it is unclear if the observed results would hold true for women. Previous literature suggests that men and women use different strategies to control the stability at the ankle joint when wearing unstable shoes.5 Thus, it may be necessary to account for gender differences when evaluating the effects of shoe designs. In addition, it is unclear how the unstable shoes designed for dress or office use would differ from other unstable shoe types designed for exercise. Lastly, since the foot tracking markers were placed directly on the shoe and not on the skin of the foot, it is possible that the true foot movements within the shoes were not fully described.
Second Generation Unstable Shoe 507
Downloaded by W VA Univ Library on 09/16/16, Volume 30, Article Number 4
The results from the comparisons between the second generation unstable shoe compared with the control shoe suggest that the unstable shoe caused changes in ankle kinematics, kinetics, GRF, and COP variables. It is evident that these changes exist, but in general, differences are small. It is not clear if these small changes produce clinically meaningful results. In addition, the results of the generational comparisons suggest that the lower-profile second generation unstable shoe may be as effective at achieving the desired unstable effects while promoting a smoother transition from heel contact through toe off compared with the first generation shoe. While this study did not evaluate other brands of unstable shoes, it provides evidence to suggest that large, bulky soles may not be necessary, and that a lower-profile sole design may yield similar biomechanical results. More research is needed to better understand the effects of these types of shoes on toning or fitness improvements as advertised by several manufacturers. Furthermore, more research is warranted on the effects of unstable shoes in populations affected by conditions such as osteoarthritis and diabetic neuropathy. Acknowledgments Funding for this study was provided by Genesco Inc.
References 1. Nigg B, Hintzen S, Ferber R. Effect of an unstable shoe construction on lower extremity gait characteristics. Clin Biomech (Bristol, Avon). 2006;21(1):82–88. PubMed doi:10.1016/j.clinbiomech.2005.08.013 2. Romkes J, Rudmann C, Brunner R. Changes in gait and EMG when walking with the Masai Barefoot Technique. Clin Biomech (Bristol, Avon). 2006;21(1):75–81. PubMed doi:10.1016/j.clinbiomech.2005.08.003 3. Landry SC, Nigg BM, Tecante KE. Standing in an unstable shoe increases postural sway and muscle activity of selected smaller extrinsic foot muscles. Gait Posture. 2010;32(2):215–219. PubMed doi:10.1016/j.gaitpost.2010.04.018 4. Nigg BM, Emery C, Hiemstra LA. Unstable shoe construction and reduction of pain in osteoarthritis patients. Med Sci
Sports Exerc. 2006;38(10):1701–1708. PubMed doi:10.1249/01. mss.0000228364.93703.53 5. Nigg BM, G KE, Federolf P, Landry SC. Gender differences in lower extremity gait biomechanics during walking using an unstable shoe. Clin Biomech (Bristol, Avon). 2010;25(10):1047–1052. PubMed doi:10.1016/j.clinbiomech.2010.07.010 6. Buchecker M, Wagner H, Pfusterschmied J, Stoggl TL, Muller E. Lower extremity joint loading during level walking with Masai barefoot technology shoes in overweight males. Scand J Med Sci Sports. 2012;22(3):372-380.. PubMed 7. Ramstrand N, Andersson CB, Rusaw D. Effects of an unstable shoe construction on standing balance in children with developmental disabilities: a pilot study. Prosthet Orthot Int. 2008;32(4):422–433. PubMed doi:10.1080/03093640802339403 8. Ramstrand N, Thuesen AH, Nielsen DB, Rusaw D. Effects of an unstable shoe construction on balance in women aged over 50 years. Clin Biomech (Bristol, Avon). 2010;25(5):455–460. PubMed doi:10.1016/j.clinbiomech.2010.01.014 9. Boyer KA, Andriacchi TP. Changes in running kinematics and kinetics in response to a rockered shoe intervention. Clin Biomech (Bristol, Avon). 2009;24(10):872–876. PubMed doi:10.1016/j.clinbiomech.2009.08.003 10. Stöggl T, Haudum A, Birklbauer J, Murrer M, Muller E. Short and long term adaptation of variability during walking using unstable (Mbt) shoes. Clin Biomech (Bristol, Avon). 2010;25(8):816–822. PubMed doi:10.1016/j.clinbiomech.2010.05.012 11. Taniguchi M, Tateuchi H, Takeoka T, Ichihashi N. Kinematic and kinetic characteristics of Masai Barefoot Technology footwear. Gait Posture. 2012;35(4):567–572. PubMed doi:10.1016/j.gaitpost.2011.11.025 12. Zhang S, Paquette MR, Milner CE, Westlake C, Byrd E, Baumgartner L. An unstable rocker-bottom shoe alters lower extremity biomechanics during level walking. Footwear Sci. 2012;4(3):243–253. doi:10.1 080/19424280.2012.735258 13. Zhang S, Wortley M, Silvernail JF, Carson D, Paquette MR. Do ankle braces provide similar effects on ankle biomechanical variables in subjects with and without chronic ankle instability during landing? J Sport Health Sci. 2012;1(2):114–120. doi:10.1016/j.jshs.2012.07.002
An Official Journal of ISB www.JAB-Journal.com ORIGINAL RESEARCH
Gait Biomechanics of a Second Generation Unstable Shoe Jacob K. Gardner,1 Songning Zhang,2 Max R. Paquette,3 Clare E. Milner,4 and Elizabeth Brock5 1Biola
University; 2University of Tennessee; 3University of Memphis; 4Drexel University; 5University of Oregon
The recent popularity of unstable shoes has sparked much interest in the efficacy of the shoe design. Anecdotal evidence suggests that earlier designs appear bulky and less aesthetically appealing for everyday use. The purpose of this study was to examine effects of a second generation unstable shoe on center of pressure (COP), ground reaction force (GRF), kinematics, and kinetics of the ankle joint during level walking at normal and fast speeds. In addition, findings were compared with results from the first generation shoe. Fourteen healthy males performed five successful level walking trials in four testing conditions: walking in unstable and control shoes at normal (1.3 m/s) and fast (1.8 m/s) speeds. The unstable shoe resulted in an increase in mediolateral COP displacement, first peak vertical GRF loading rate, braking GRF, ankle eversion range of motion (ROM), and inversion moment; as well as a decrease in anteroposterior COP displacement, second peak vertical GRF, ankle plantarflexion ROM, and dorsiflexion moment. Only minor differences were found between the shoe generations. Results of the generational comparisons suggest that the lower-profile second generation shoe may be as effective at achieving the desired unstable effects while promoting a smoother transition from heel contact through toe off compared with the first generation shoe. Keywords: biomechanics, unstable, kinematics, kinetics, toning The recent popularity of unstable shoes has sparked much interest into the efficacy of the unstable shoe design. The anteroposterior curved sole of unstable shoes is thought to make the shoes less stable than standard shoes. It has been proposed that the instability of the shoe would provide a training effect for muscles of the lower extremity that are important for maintaining static and dynamic postural stability during walking.1,2 These anticipated outcomes have spawned the terms toning shoes or fitness shoes to describe the functional roles of the shoe design. However, not enough research evidence exists to substantiate the claims of these types of shoes. The majority of published research has been done on the Masai Barefoot Technology (MBT) (Swiss Masai, Gossau, Switzerland) shoes.1–10 More recently, several other manufacturers have introduced their own models (eg, Skechers Shape-ups, Manhattan Beach, California; Dockers Active Balance, San Francisco, California; Reebok EasyTone, Canton, Massachusetts; New Balance True Balance, Boston, Massachusetts; and others). Thus, more research aimed to study other shoe designs and models is needed. Biomechanics studies have shown that MBT shoes can acutely alter spatiotemporal variables during gait.1,2,6 Walking in MBT shoes at a self-selected pace has resulted in decreased stride length and time, decreased step length, and decreased walking speed compared with a control shoe. It should be noted that the reported reductions Jacob K. Gardner is with the Department of Kinesiology, Health, and Physical Education at Biola University, La Mirada, CA. Songning Zhang is with the Department of Kinesiology, Recreation, and Sport Studies at the University of Tennessee, Knoxville, TN. Max R. Paquette is with the Department of Health and Sport Sciences at the University of Memphis, Memphis, TN. Clare E. Milner is with the Department of Physical Therapy and Rehabilitation Sciences at Drexel University, Philadelphia, PA. Elizabeth Brock is with the Lundquist College of Business at the University of Oregon, Eugene, OR. Address author correspondence to Jacob K. Gardner at [email protected].
in spatiotemporal characteristics are likely due to a decreased selfselected walking speed in MBT shoes. A recent study showed no significant spatiotemporal changes during gait between a control shoe and an MBT shoe when walking speed was controlled.11 Therefore, it is important to account for walking speed when making comparisons among shoes. Kinematic changes at the ankle have been detected while wearing MBT shoes; most notably, an increased dorsiflexion angle at heel strike,1,2 and either increased11 or decreased2 dorsiflexion in late stance have been observed. Our previous work has shown significant kinematic changes between a business casual unstable shoe (Active Balance; Genesco Inc., Nashville, Tennessee) and a standardized control shoe (Glacier; Genesco Inc., Nashville, Tennessee) during level walking in healthy men.12 The significant kinematic changes included an increased ankle dorsiflexion angle at heel strike and a decreased plantarflexion range of motion (ROM) during early stance. In addition, we also showed greater ankle eversion ROM in the unstable shoes. In addition to kinematics, a few studies have reported alterations in kinetics while walking in unstable shoes.11,12 Our previous study showed that despite similar walking speeds, the use of unstable shoes resulted in increased mediolateral center of pressure (COP) displacement and decreased anteroposterior COP displacement, decreased first peak vertical ground reaction force (GRF), and decreased ankle dorsiflexion moment.12 Taniguchi et al11 also found a significant decrease in ankle plantarflexion moment during late stance. In addition, our previous work also found increased loading rate of first peak vertical GRF, increased peak braking GRF, and increased peak ankle inversion moment in the unstable shoe.12 Recently, anecdotal evidence suggests that earlier designs appear too bulky and less aesthetically appealing for everyday use such as in business and office settings. It is important to be able to wear the shoe during everyday tasks, as researchers have suggested that unstable shoes may be beneficial for long-term use.4 For these reasons, the original unstable shoe (from Zhang et al12) was 501
Downloaded by W VA Univ Library on 09/16/16, Volume 30, Article Number 4
502 Gardner et al.
redesigned with a lower profile and a dressier style for daily wear in work environments. Therefore, the purpose of this study was to examine the effect of the redesigned (second generation) unstable shoe on COP, GRF, kinematics, and kinetics of the ankle joint during level walking at normal and fast speeds. We also wanted to compare the findings with the results from the first generation shoe. The aim was to determine if similar biomechanical changes compared with a control shoe would be maintained in the lower-profile second generation design. Based on the results of our previous study,12 we hypothesized that wearing the unstable shoes would result in a decrease in first peak vertical GRF, anteroposterior COP displacement, and ankle plantar flexion ROM; as well as an increase in mediolateral COP displacement, loading rate of first peak vertical GRF, peak braking GRF, ankle inversion ROM, and ankle inversion moment when compared with the control shoe. We also hypothesized that the biomechanical differences found to be significant in the first generation unstable shoe compared with the control shoe would be similar in the second generation shoe.
a
Methods Participants Fourteen healthy males participated in this study (age: 45.6 ± 8.4 y; height: 1.81 ± 0.07 m; weight: 80.6 ± 9.4 kg; BMI: 25.1 ± 2.8). All participants were free from any gait abnormality, neurologic disorders, or major surgeries, and free from any lower extremity injury for at least 6 months before the time of the study. All subjects provided written informed consent approved by the institutional review board before testing.
Instrumentation
b
A 9-camera three-dimensional (3D) motion capture system (240 Hz; Vicon Motion Systems, Oxford, United Kingdom) and a force platform (1200 Hz; Advanced Mechanical Technology Inc., Watertown, Massachusetts) were used to collect 3D kinematic and ground reaction force data, respectively. Retro-reflective anatomical markers were placed bilaterally on the acromion processes, iliac crests, greater trochanters, medial and lateral epicondyles, medial and lateral malleoli, and first and fifth metatarsal heads.13 Four tracking markers attached to rigid thermoplastic shells were placed bilaterally on the thighs and legs as well as on the pelvis and trunk. Foot tracking markers were placed directly on the posterior and lateral aspects of the heel-counter of the shoes. Kinematic and GRF data were recorded simultaneously using the Vicon Nexus software.
Figure 1 — Control (a) and unstable (b) shoe used for the study.
Experimental Protocol Before testing, participants walked in an indoor hallway for 5 minutes in a flat bottom control shoe (Gordon; Genesco Inc., Nashville, Tennessee; Figure 1a) and 5 minutes in an unstable curved bottom shoe (Active Balance, Second Generation; Genesco Inc., Nashville, Tennessee; Figure 1b) to familiarize themselves with both pairs of shoes. Before data collection, retro-reflective markers were placed on the participants as described above for 3D motion capture. Participants wore athletic clothing so that anatomical markers could be placed directly on the skin. Two separate calibration trials were performed for each shoe tested to define model segments and joint centers. During testing, each participant performed five successful
level walking trials in each of four testing conditions: walking in control and unstable at a normal walking speed of 1.3 m/s and a fast speed of 1.8 m/s. The walking speed was monitored by two pairs of photo cells and an electronic timer. A successful trial was defined as a trial in which the participant contacted the force platform with their right foot without targeting and while maintaining walking speeds within 5% of the desired speed. To reduce bias, the order of conditions was randomized first by shoe (control or unstable) and then by walking speed (1.3 m/s or 1.8 m/s). Each participant was given ample time to practice walking in each shoe and speed condition before data collection to become familiar with the testing protocol.
Data Processing and Analysis 3D kinematic and kinetic data were processed and analyzed using Visual 3D software suite (C-Motion, Inc., Germantown, Maryland). Kinematic and kinetic data were filtered with a second order lowpass digital Butterworth filter at 6 Hz and 50 Hz, respectively. Joint angles were calculated using an X-Y-Z Cardan rotation sequence. The conventions of joint kinematics and kinetics were determined using the right-hand rule, and all joint moments were calculated as internal moments. In addition, all GRF were normalized to each participant’s body weight (BW), and joint moments were normalized to body mass (N·m/kg). Customized computer programs (VB_V3D and VB_Table) were used to determine peaks in the curves of the
Downloaded by W VA Univ Library on 09/16/16, Volume 30, Article Number 4
Second Generation Unstable Shoe 503
3D kinematic and kinetic variables of interest from the output of Visual 3D and were also used to compute additional parameters and organize the variables for statistical analyses. The average of the five trials of the selected variables was used in subsequent statistical analysis. All variables of interest were analyzed during the stance phase of gait; from heel strike to toe off. Heel strike was determined when the vertical GRF magnitude reached a threshold of 10 newtons (N), and toe off was determined when the magnitude dropped below a threshold of 10 N. The kinematic and kinetic variables we chose to report were selected based on the previous study of the first generation rocker bottom shoe,12 which allowed us to make comparisons between the two shoe generations. To quantify shoe characteristics, the slope of the heel and forefoot regions of both the control and unstable test shoes were calculated using the following equations where SFT = forefoot slope percentage; Lft = forefoot slope length; f = forefoot length; SHL = heel slope percentage; Lhl = heel slope length; r = heel length; aFT = forefoot slope angle; t = toe height; aHL = heel slope angle; and h = heel height (Figure 2):12
SFT =
L ft L × 100 S HL = hl × 100 (1) f r
a FT =
t f
a HL =
h r
(2)
These measurements were also performed on the first generation shoe and its corresponding control shoe for generational comparisons. All measurements for the first and second generation shoes, as well as the control shoes, were computed as an average for shoe sizes 8, 9, and 11. A Shapiro-Wilk normality test was performed on each variable of interest. All variables were found to be normally distributed. Thus parametric statistical tests were performed for determining differences among conditions. For comparisons between the second generation shoe and its respective control shoe, a 2 × 2 (shoe × speed) repeated measures analysis of variance (ANOVA) was used to determine any differences between kinematic and kinetic variables (SPSS version 18.0; SPSS Inc., Chicago, Illinois). An alpha level of .05 was set a priori. When a shoe × speed interaction was found, a post hoc comparison using a paired samples t test with a Bonferroni adjustment for multiple comparisons was conducted to detect differences between shoes and/or speeds. In addition, for comparisons
Figure 2 — Diagram of measurement parameters used to compare the shoes.
between first12 and second generation shoes, subject demographic and select variables that were statistically different from their respective control shoes for each study were extracted and used for comparison (original data from the previous study conducted in our laboratory were obtained). The differences between the two shoe generations were compared using an independent samples t test for each variable of interest. Only shoe main effects in the normal walking speed were used for comparison. It is important to note that while the two studies were similar in methodology, the control shoe used for the first generation study was not the same control shoe used in the current study. In addition, the subject sample was not the same for each study, although there were two overlapping subjects. Independent samples t tests were performed on subject demographics excluding the two overlapped subjects and revealed that the age, height, mass and BMI were not different between the two studies.
Results For comparisons between the second generation shoe and the control shoe, the ANOVA results showed significant speed main effects for all COP and GRF variables (P < .001) except for mediolateral COP displacement (P = .277, Table 1). The unstable shoe showed
Table 1 Center of pressure and ground reaction force variables, mean ± SD Control Shoe
Unstable Shoe
Shoe
Speed
Interaction
Normal Speed
Fast Speed
Normal Speed
Fast Speed
P
P
P
Mediolateral COP displacement (m)a
–.05 ± .02
–.05 ± .02
–.06 ± .02
–.06 ± .02
.044
.277
.901
Anteroposterior COP displacement (m)a,b
.20 ± .04
.19 ± .04
.17 ± .03
.16 ± .03
< .001
< .001
.354
First peak vertical GRF (BW)b
1.18 ± .06
1.38 ± .07
1.17 ± .07
1.37 ± .07
.430
< .001
.855
Second peak vertical GRF (BW)a,b
1.14 ± .05
1.23 ± .08
1.11 ± .05
1.20 ± .07
.001
< .001
.963
Peak vertical GRF loading rate (BW/s)a,b
7.52 ± 1.05
11.48 ± 1.97
8.42 ± 2.09
12.24 ± 2.40
.024
< .001
.632
Peak braking GRF (BW)a,b
–.22 ± .02
–.29 ± .04
–.23 ± .02
–.30 ± .04
.012
< .001
.416
Peak propulsive GRF (BW)b
.23 ± .03
.31 ± .05
.24 ± .04
.31 ± .04
.693
< .001
.416
Variables
Abbreviations: COP, center of pressure; GRF, ground reaction force. a P < .05 = significant shoe main effect. b P < .05 = significant speed main effect.
Downloaded by W VA Univ Library on 09/16/16, Volume 30, Article Number 4
504 Gardner et al.
increased mediolateral COP displacement (P = .044), decreased anteroposterior COP displacement (P < .001), increased loading rate of the first peak vertical GRF (P = .024), decreased second peak vertical GRF (P = .001), and increased peak braking GRF (P = .012) compared with the control shoe. Ankle angle ensemble (average) curves for the sagittal and frontal planes during the normal and fast walking speeds are presented in Figures 3 and 4, respectively. A speed main effect was found for all ankle kinematic variables with the exception of ankle plantar flexion ROM (measured from contact to the peak plantar flexion angle, Table 2). In terms of shoe main effects, the unstable shoe showed an approximate 6° decrease in plantar flexion ROM (P < .001) and a decrease in total ankle joint ROM (P < .001). There
was also an approximate 2° increase in eversion ROM in the normal (P = .007) and fast (P = .001) walking speeds. Ankle moment ensemble (average) curves for the sagittal and frontal planes during the normal and fast walking speeds are presented in Figure 5 and 6, respectively. Speed main effects were significant for all ankle joint kinetic variables except for peak ankle inversion moment (Table 2). In terms of shoe main effects, the unstable shoe showed an approximate 40% decrease in peak ankle dorsiflexion moment (P < .001) and a 34% increase in peak ankle inversion moment (P = .003). Anteroposterior COP displacement was greater in the second generation shoe compared with the first generation shoe (P < .001, Table 3). In addition, the first peak vertical GRF was decreased in the second generation shoe compared
Figure 3 — Ensemble (average) ankle angle curves in the sagittal (top) and frontal (bottom) planes for the control (solid black line) and the unstable (dashed gray line) shoes at the normal walking speed.
Figure 4 — Ensemble (average) ankle angle curves in the sagittal (top) and frontal (bottom) planes for the control (solid black line) and the unstable (dashed gray line) shoes at the fast walking speed.
Table 2 Ankle kinematic and kinetic variables, mean ± SD Control Shoe Variable Sagittal contact angle (°)a PF ROM
(°)b
Unstable Shoe
Shoe
Speed Interaction
Normal Speed
Fast Speed
Normal Speed
Fast Speed
P
P
P
.21 ± 4.03
2.12 ± 4.31
–.44 ± 3.44
.85 ± 3.76
.228
< .001
.237
–14.49 ± 3.47 –15.35 ± 2.92
–8.85 ± 3.42
–9.21 ± 2.58
< .001
.113
.176
Peak DF angle (°)a
11.19 ± 3.29
9.64 ± 3.37
9.87 ± 4.02
8.13 ± 4.80
.088
.006
.629
Stance phase joint ROM (°)a,b
25.47 ± 3.63
22.86 ± 4.53
19.28 ± 4.01
16.49 ± 3.98
< .001
< .001
.745
–5.59 ± 3.11
–4.47 ± 2.59
–6.99 ± 3.55
–6.71 ± 3.82
.001
.015
.025
.38 ± .05
.49 ± .07d
.22 ± .05e
.31 ± .07d,e
< .001
< .001
.050
–1.39 ± .09
–1.51 ± .15
–1.37 ± .12
–1.48 ± .15
.197
< .001
.819
.11 ± .05
.12 ± .05
.15 ± .04
.16 ± .05
.003
.160
.930
Eversion ROM
(°)a,b,c
Peak DF moment (N⋅m/kg)a,b Peak PF moment
(N⋅m/kg)a
Peak inversion moment
(N⋅m/kg)b
Abbreviations: PF, plantarflexion; ROM, range of motion; DF, dorsiflexion. a P < .05 = significant speed main effect. b P < .05 = significant shoe main effect. c P < .05 = significant interaction of shoe and speed. d Significant difference between speeds for same shoe in post hoc comparison. e Significant difference between shoes at same speed in post hoc comparison.
Downloaded by W VA Univ Library on 09/16/16, Volume 30, Article Number 4
Second Generation Unstable Shoe 505
Figure 5 — Ensemble (average) ankle moment curves in the sagittal (top) and frontal (bottom) planes for the control (solid black line) and the unstable (dashed gray line) shoes at the normal walking speed.
Figure 6 — Ensemble (average) ankle moment curves in the sagittal (top) and frontal (bottom) planes for the control (solid black line) and the unstable (dashed gray line) shoes at the fast walking speed.
Table 3 Ground reaction force and ankle joint variable differences between shoe generations, mean ± SD ML COP Disp. (m)
AP COP Disp. (m)a
First Peak Vertical GRF (BW)a
Vertical GRF Loading Rate (BW/s)
Peak Braking GRF (BW)
Second Peak Vertical GRF (BW)
Eversion ROM (°)
Peak Inversion Moment (N⋅m/kg)
First generation
–.06 ± 0.02
.18 ± 0.04
1.16 ± .08
8.27 ± 2.10
–.23 ± .02
1.11 ± .05
–6.87 ± 3.45
.14 ± .04
Second generation
–.07 ± 0.02
.26 ± 0.02
1.11 ± .05
7.69 ± 1.97
–.21 ± .02
1.08 ± .03
–8.81 ± 2.34
.14 ± .05
.651
< .001
.042
.447
.093
.057
.091
.729
P value
Abbreviations: ML, mediolateral; COP, center of pressure; Disp., displacement; AP, anteroposterior; GRF, ground reaction force; ROM, range of motion. Note. First generation unstable shoe results from Zhang et al 2012. Second generation unstable shoe results from the current study. a P < .05 = significant difference between shoe generations.
with the first generation shoe (P = .042). No other variables were significantly different between the two shoe generations. Dimensional characteristics of the first and second generation shoes, along with their respective control shoes, are shown (Table 4). In the second generation shoe, the heel slope angle was 13.7° larger and the forefoot slope angle was about 2.5° larger than the control shoe. In comparing the two unstable shoe generations, the first generation shoe had larger heel and forefoot slope percentages compared with the second generation shoe. In addition, the heel slope angle was greater in the first generation shoe, while the forefoot angle was smaller compared with the second generation shoe (Table 4).
Discussion One of the purposes of this study was to determine the biomechanical differences between a low-profile unstable shoe and a standard dress shoe during level walking at two different speeds. Several studies have examined biomechanical characteristics of gait in unstable shoes. However, this is the first study, to our knowledge, that has analyzed a low-profile unstable shoe design. The second
Table 4 Shoe measurements of the first and second generation unstable shoes and their respective control shoes αHL
αFT
23.4°
17.1°
104.0% 102.9%
8.1°
12.7°
Second generation unstable shoe 104.4% 102.8%
13.7°
18.9°
0°
16.4°
SHL
SFT
First generation unstable shoe
108.8% 105.5%
First generation control Second generation control
100.0% 103.0%
Abbreviations: SHL, heel slope percentage; SFT, forefoot slope percentage; αHL, heel slope angle; αFT, forefoot slope angle.
purpose was to compare two generations of the unstable shoes from the same manufacturer to detect gait biomechanical differences as a result of changes in the shoe designs. When compared with the control shoe, the second generation unstable shoe resulted in a 19% significant increase in mediolateral COP displacement at normal walking speed. An increase in the mediolateral COP displacement is consistent with our previous findings.12 This increased mediolateral COP movement was thought
Downloaded by W VA Univ Library on 09/16/16, Volume 30, Article Number 4
506 Gardner et al.
to be related to the softer midsole materials in the heel region of the unstable shoes.12 We also found a decrease in the anteroposterior COP displacement in the unstable shoe compared with the control shoe. The decreased COP movement is supported by an increase of 15° and 4.5% in the heel slope angle and heel slope percentage compared with the control shoe. The decreased anteroposterior COP displacement may indicate that wearing the unstable shoes results in a more anterior heel contact upon foot strike.12 This may also explain why the loading rate of the first peak vertical GRF was increased in the unstable shoe when compared with the control shoe, and may have caused the increased peak braking GRF during weight acceptance.12 Even though the loading rate of the first peak vertical GRF was increased in the unstable shoe, there was no difference in the first peak vertical GRF between the unstable and control shoes. However, the second peak (push-off peak) did decrease in the unstable shoes. The curvature of the forefoot region of the unstable shoes may contribute to the reduced second peak vertical GRF. This finding does not affect propulsion; as no difference was found in peak propulsive GRF between the shoes. The result is in agreement with the finding in our previous study of the first generation unstable shoe.12 Furthermore, this finding was also supported by the lack of difference in the peak plantar flexion moment between the shoes during push-off. In addition, anecdotal reports from our participants suggest that the unstable shoe did not feel different than the control shoe after the initial familiarization period. In terms of ankle joint kinematics, the decreased dorsiflexion moment in the unstable shoe during weight acceptance is consistent with the findings of the MBT unstable shoe studied by Taniguchi et al.11 The peak plantar flexion moment during late stance was not reduced in the unstable shoe compared with the control shoe, which is different from the result of the previous study.12 This lack of difference could be related to the reduced profile in the second generation shoe. An even greater reduction of 21% in peak ankle plantar flexion moment was found by Taniguchi et al.11 This decrease may be related to the slightly, but nonsignificant, slower walking speed (self-selected) in the MBT in the study. In addition, the participants in the study by Taniguchi et al11 were trained for 30 minutes before testing on how to properly walk in the unstable shoes. Our participants, on the other hand, were provided with a 5 minute familiarization period. Whether the amount of initial exposure to the shoes has an effect on biomechanical variables during a one-time walking session remains unclear. However, Nigg et al1 did not show any significant difference of plantar flexion impulse between the MBT and a running shoe in walking after a 2-week accommodation period. Finally, since Taniguchi et al11 only reported sagittal plane joint variables, it is not clear if the increased peak inversion moment found in our study is also observed in MBT shoes. When comparing biomechanical variables between unstable shoe generations, the decreased anteroposterior COP displacement in the current study is consistent with the results from our first generation unstable shoe study.12 However, the displacement was significantly greater in the second generation design. The greater anteroposterior COP displacement was likely a result of the smaller heel angle of the shoe sole in the second generation shoe. This was the largest difference between the two shoe designs and was almost 10° less in the second generation shoe (Table 4). The GRF variables were similar between shoe generations, but a significant difference was found in the first peak vertical GRF, as the second generation shoe showed a 0.05 BW smaller peak vertical GRF compared with its first generation counterpart. While this difference is small, it may
be more desirable for load reduction during the weight acceptance phase of gait. The ankle kinematic and kinetic parameters showed no statistical differences between shoe generations. This is an important finding as it shows that the second generation shoe still provides similar unstable effects as the first generation shoe, even though it has a lower-profile design. Besides aesthetics, one of the reasons for changing the shoe design was to reduce the profile to elicit a smoother transition between heel strike and toe off. Based on the dependent variables selected, the findings from the lower-profile construction of the second generation shoe did not greatly differ from that of its first generation counterpart. The largest difference between the two shoe generations was the slope of the heel, which was decreased by about 4.4% in the second generation shoe and was accompanied by a 9.7° decrease in heel slope angle. As mentioned previously, the difference in heel slope was likely the reason for the decreased first peak vertical GRF between the two shoes. Interestingly, no change occurred in the peak braking GRF even though the heel slope was smaller. It is unclear if the small but significant changes noted in this study would produce clinically meaningful outcomes. To achieve a training effect, one would expect increases in joint ROM and/or joint moments (ie, increased muscular effort). The decreases in sagittal plane ROM and moments in this study suggest a possible decrease in training effect, which is just the opposite of what we would initially expect to see. However, in the frontal plane, the increase in eversion ROM and inversion moment, while small, may be beneficial for mediolateral movement and postural control (ie, sway). Thus, it is possible that wearing unstable shoes on a regular basis might help to strengthen the lower extremity muscles such as the tibialis anterior and peroneus longus for ankle eversion movement and aid in ankle contribution to sway control; however, it would probably not aid in strengthening antigravity muscles (eg, gastrocnemius and soleus) responsible for lower extremity movements in sagittal plane during gait. Our results and the results of a previous study,1 suggest that subjects may eventually get accustomed to the shoes and effects may diminish over time. It is possible that unstable shoes may be advantageous for variables other than muscular training. One study showed that when subjects with knee osteoarthritis wore the unstable shoes they saw a decrease in total knee pain after 12 weeks of use.4 However, the control group wearing standard shoes also saw a decrease in knee pain and there was no statistical difference in total pain between groups. In addition, pain during walking was reduced in both groups, but the control group experienced a slightly larger reduction compared with the test group. Thus, it is difficult to determine if there was a true cause and effect relationship between the shoes and symptom relief in the patient population. More prospective studies with focus on effects on secondary planes are needed to determine the effectiveness of the unstable shoe as a training device. It should be noted that since the shoes used in this study were designed for use by men only, it is unclear if the observed results would hold true for women. Previous literature suggests that men and women use different strategies to control the stability at the ankle joint when wearing unstable shoes.5 Thus, it may be necessary to account for gender differences when evaluating the effects of shoe designs. In addition, it is unclear how the unstable shoes designed for dress or office use would differ from other unstable shoe types designed for exercise. Lastly, since the foot tracking markers were placed directly on the shoe and not on the skin of the foot, it is possible that the true foot movements within the shoes were not fully described.
Second Generation Unstable Shoe 507
Downloaded by W VA Univ Library on 09/16/16, Volume 30, Article Number 4
The results from the comparisons between the second generation unstable shoe compared with the control shoe suggest that the unstable shoe caused changes in ankle kinematics, kinetics, GRF, and COP variables. It is evident that these changes exist, but in general, differences are small. It is not clear if these small changes produce clinically meaningful results. In addition, the results of the generational comparisons suggest that the lower-profile second generation unstable shoe may be as effective at achieving the desired unstable effects while promoting a smoother transition from heel contact through toe off compared with the first generation shoe. While this study did not evaluate other brands of unstable shoes, it provides evidence to suggest that large, bulky soles may not be necessary, and that a lower-profile sole design may yield similar biomechanical results. More research is needed to better understand the effects of these types of shoes on toning or fitness improvements as advertised by several manufacturers. Furthermore, more research is warranted on the effects of unstable shoes in populations affected by conditions such as osteoarthritis and diabetic neuropathy. Acknowledgments Funding for this study was provided by Genesco Inc.
References 1. Nigg B, Hintzen S, Ferber R. Effect of an unstable shoe construction on lower extremity gait characteristics. Clin Biomech (Bristol, Avon). 2006;21(1):82–88. PubMed doi:10.1016/j.clinbiomech.2005.08.013 2. Romkes J, Rudmann C, Brunner R. Changes in gait and EMG when walking with the Masai Barefoot Technique. Clin Biomech (Bristol, Avon). 2006;21(1):75–81. PubMed doi:10.1016/j.clinbiomech.2005.08.003 3. Landry SC, Nigg BM, Tecante KE. Standing in an unstable shoe increases postural sway and muscle activity of selected smaller extrinsic foot muscles. Gait Posture. 2010;32(2):215–219. PubMed doi:10.1016/j.gaitpost.2010.04.018 4. Nigg BM, Emery C, Hiemstra LA. Unstable shoe construction and reduction of pain in osteoarthritis patients. Med Sci
Sports Exerc. 2006;38(10):1701–1708. PubMed doi:10.1249/01. mss.0000228364.93703.53 5. Nigg BM, G KE, Federolf P, Landry SC. Gender differences in lower extremity gait biomechanics during walking using an unstable shoe. Clin Biomech (Bristol, Avon). 2010;25(10):1047–1052. PubMed doi:10.1016/j.clinbiomech.2010.07.010 6. Buchecker M, Wagner H, Pfusterschmied J, Stoggl TL, Muller E. Lower extremity joint loading during level walking with Masai barefoot technology shoes in overweight males. Scand J Med Sci Sports. 2012;22(3):372-380.. PubMed 7. Ramstrand N, Andersson CB, Rusaw D. Effects of an unstable shoe construction on standing balance in children with developmental disabilities: a pilot study. Prosthet Orthot Int. 2008;32(4):422–433. PubMed doi:10.1080/03093640802339403 8. Ramstrand N, Thuesen AH, Nielsen DB, Rusaw D. Effects of an unstable shoe construction on balance in women aged over 50 years. Clin Biomech (Bristol, Avon). 2010;25(5):455–460. PubMed doi:10.1016/j.clinbiomech.2010.01.014 9. Boyer KA, Andriacchi TP. Changes in running kinematics and kinetics in response to a rockered shoe intervention. Clin Biomech (Bristol, Avon). 2009;24(10):872–876. PubMed doi:10.1016/j.clinbiomech.2009.08.003 10. Stöggl T, Haudum A, Birklbauer J, Murrer M, Muller E. Short and long term adaptation of variability during walking using unstable (Mbt) shoes. Clin Biomech (Bristol, Avon). 2010;25(8):816–822. PubMed doi:10.1016/j.clinbiomech.2010.05.012 11. Taniguchi M, Tateuchi H, Takeoka T, Ichihashi N. Kinematic and kinetic characteristics of Masai Barefoot Technology footwear. Gait Posture. 2012;35(4):567–572. PubMed doi:10.1016/j.gaitpost.2011.11.025 12. Zhang S, Paquette MR, Milner CE, Westlake C, Byrd E, Baumgartner L. An unstable rocker-bottom shoe alters lower extremity biomechanics during level walking. Footwear Sci. 2012;4(3):243–253. doi:10.1 080/19424280.2012.735258 13. Zhang S, Wortley M, Silvernail JF, Carson D, Paquette MR. Do ankle braces provide similar effects on ankle biomechanical variables in subjects with and without chronic ankle instability during landing? J Sport Health Sci. 2012;1(2):114–120. doi:10.1016/j.jshs.2012.07.002
