IMPACT OF FOOTWEAR BENDING STIFFNESS ON LOWER-LIMB WORK REDISTRIBUTION AND RUNNING ECONOMY
Abstract
The bending stiffness of footwear impacts running efficiency and lower-limb work redistribution. However, studies integrating both aspects are scarce. This study aimed to examine the impact of footwear stiffness on lower-limb biomechanics, joint work, and overall metabolic efficiency. This study was performed on 12 male recreational runners to complete an experimental protocol while wearing two different running shoes with varying degrees of longitudinal bending stiffness. Paired-sample t tests were applied in this research. The stiffer footwear decreased the range of motion (ROM), angular velocity, negative and positive work of the metatarsophalangeal (MTP) joint, and ROM of the ankle joint. The running economy significantly improved with the stiffer footwear. However, no significant difference was observed in the joint work of other lower limbs. Shoes with increased bending stiffness significantly reduced the MTP joint ROM and angular velocity, thereby reducing negative power and work at the MTP joint and improving running economy for runners. This study provided relevant information for shoe designers, developers, and scientists conducting research on footwear midsole structures and designs.
1. Introduction
Sports performance is characterized as the time or energy expended per unit distance.1 Previous studies have shown that aerobic capacity is a prerequisite for excellent long-distance running performance.2,3,4 Running economy is considered the key indicator for evaluating an athlete’s aerobic capacity. It is also an essential factor in determining the endurance athletic performance of a runner.5
Footwear is a crucial factor influencing the running economy.6,7 The sandwich midsole structure, which consists of a compliant and resilient foam on both the top and bottom layers with a carbon plate in the middle, has been proven to enhance the running economy compared with conventional running shoes.8,9,10,11 From a material perspective, the amount of energy stored in a foam material depends on its compliance — the amount of compressive deformation the material undergoes when the runner touches the ground.11 The percentage of stored mechanical energy released is termed resilience (energy dissipated as heat by viscoelastic foam, which is inevitable12). Therefore, shoe midsoles that exhibit greater compliance and resilience reduce energy expenditure during running.13
From a biomechanical perspective, the improvement in footwear bending stiffness impacts lower-limb biomechanics, thus improving the running economy.9,10,14 However, a consensus on the functional mechanism has not yet been reached. Several studies suggested that the improvement in footwear bending stiffness should decrease energy dissipation and negative work at the metatarsophalangeal (MTP) joint by stiffening the joint and restricting dorsiflexion.7,15,16 However, some other studies14,17,18 reported no significant difference in negative work for footwear with varying degrees of stiffness. The augmentation of plantarflexion moment surpassing the reduction in dorsiflexion velocity could explain these findings. Thus, achieving an appropriate bending stiffness in footwear is essential to counterbalance plantarflexion joint movements and dorsiflexion angular velocities.17 For the ankle joint, the foot serves as a lever arm for the triceps surae muscles during running. This force point can be repositioned more anteriorly using a longitudinal stiffness plate.15,18 However, the impact on ankle joint movement is still unclear. Roy and Stefanyshyn14 demonstrated an increasing peak ankle joint moment with footwear featuring heightened bending stiffness. In contrast, several other studies found no difference in either peak or average ankle joint moment when altering longitudinal bending stiffness.18,19
Research on the mechanisms of the impact of footwear bending stiffness on hip-and-knee joints is limited. Cigoja et al.20 documented that increased footwear bending stiffness redistributed positive joint work from the knee to the MTP joint. Another study documented alterations in hip-joint mechanics with increased bending stiffness of footwear,21 reporting a specific increase in the hip moment arm. Numerous prior studies have individually explored the impacts of footwear bending stiffness on the biomechanical characteristics of lower limbs and running economy. However, research integrating both aspects is scarce. This study aimed to analyze the impact of footwear bending stiffness on lower-limb biomechanics, joint work, and overall metabolic efficiency. This study hypothesized that footwear bending stiffness would redistribute lower-limb work, and more proximal joints might compensate for the change in MTP joint work. Also, stiffer footwear would limit the range of motion (ROM) of the MTP joint and decrease negative power and work, thereby enhancing the running economy.
2. Methods
2.1. Participants
A total of 12 male recreational runners (mean age: 27.0±2.3 years), with an average height and weight of 174.0±3.2cm and 67.5±4.5kg, respectively, were recruited for this biomechanical and metabolic study. The participants were required to maintain a weekly running distance of at least 40km and achieve a personal best marathon performance of 3h and 30min. Their foot size (US 9.0±0.5) was confirmed using a Brannock foot-measuring device (The Brannock Device Co., NY, USA). The participants were screened to ensure that they had no lower extremity neuromuscular or musculoskeletal issues or experienced pain within the preceding 6 months. Ethical approval for the study was granted by the institutional ethics committee, guaranteeing compliance with established research protocols and standards.
2.2. Shoe conditions
Two pairs of identical shoes in US size 9.0 with differing carbon plate structures were constructed based on the existing running shoe (Li-Ning Feidian, Li-Ning Company, Beijing, China). One model featured a midsole crafted from two layers of Pebax material without a mid-carbon plate, whereas the other model used a sandwich structure midsole comprising upper and lower layers of Pebax material with a mid-carbon plate (Fig. 1).

Fig. 1. Two experimental footwear designs. The left shoe has a carbon plate structure, whereas the right shoe does not have one.
2.3. Experimental protocol
Each participant was required to take two tests at the laboratory. In Test 1, the participants were made to run on the instrumented treadmill while wearing the expired-gas analysis system (MetaLyzer3B-R3; Cortex, Leipzig, Germany) at a self-selected pace to familiarize themselves with the two test shoes and experimental devices. The participants’ basic personal information such as foot size, weight, and the informed consent form were recorded.
Test 2 was conducted at least 24h after Test 1. A 10-camera infrared motion analysis system (Oxford Metrics Ltd., Oxford, UK, sampling 200Hz) was synchronized with an instrumented treadmill (Fully Instrumented Treadmill, Bertec Corp., OH, USA, sampling 1000Hz) to collect the lower-limb biomechanical variables. The participants were first made to warm up on a treadmill for 5min at their usual pace and then complete a 5-min test on a 1∘ treadmill at a speed of 3.61m/s.9,10 The lower-limb biomechanics and energy metabolism were recorded in the last 2min. Energy metabolism and biomechanical data were manually synchronized. During the inter-trial intervals, participants took a 10-min break to switch test shoes. A mirrored order was adopted for the experiment to counterbalance and randomly assign the test shoes. The results of the two trials for each shoe condition were averaged across all metrics.
2.4. Data processing
The biomechanical data underwent processing with Visual 3D (C-Motion Inc., MD, USA) to define body segments and calculate joint kinetic and kinematic variables. Marker and force data were filtered using a dual-pass second-order Butterworth filter set at a cutoff frequency of 50Hz. The Cardan angles were used to delineate joint movement, whereas an inverse dynamics method was employed to estimate sagittal joint moments. Mechanical joint powers were derived from the dot product of joint moment and angular velocity. Positive and negative mechanical work were determined by integrating the positive and negative joint power–time curves over the stance phase, respectively. The braking and propulsion periods were determined based on the zero-crossing point of anterior–posterior ground reaction force (GRF). The GRF and lower-limb kinetics were normalized by body weight.
For metabolic variables, the submaximal rates of oxygen uptake (VO2) and carbon dioxide production (VCO2) were recorded for the last 2min of each trial. The respiratory exchange ratio (RER) (RER=VCO2/VO2) was computed to verify that participants were exercising at a submaximal aerobic intensity (RER<1.0). The gross metabolic power was calculated using the Peronnet and Massicotte equation.22,23
2.5. Statistical analysis
The normal distribution of the data was initially validated using Shapiro–Wilk tests. Paired-sample t tests were conducted to examine the differences in joint angle, angular velocity, moment, power, work, and metabolism between the two shoe conditions. The significance level was established at 0.05, and the P values for individual comparisons were adjusted using the Benjamini and Hochberg methods. All statistical analyses were conducted using SPSS 20.0 (SPSS Inc., IL, USA).
3. Results
3.1. Kinematics and kinetics
The variations in lower-limb biomechanics are depicted in Fig. 2 and Table 1. In this study, the kinematic and kinetics of hip-and-knee joints showed a similar trend during a one-step gait between the two shoe conditions. No significant difference was observed in the ROM, angular velocity, moment, and power between the two shoe conditions. For the ankle joint, the angular velocity, moment, and power were similar. The footwear with the plate structure showed smaller peak sagittal ROM, and the dynamics of sagittal ROM was smaller during the 0–80% gait period. For the MTP joint, the moment was similar. Furthermore, the footwear with the plate structure showed smaller sagittal ROM, peak flexion angular velocity, and peak positive and negative power, and the dynamics of the four variables was smaller.

Fig. 2. Hip, knee, ankle, and MTP joint angle, angular velocity, moment, and power curve over the running stance phase for the two shoe conditions.
Variable | With plate structure | Without plate structure |
---|---|---|
Hip | ||
ROM (∘) | 44.10±3.56 | 44.38±3.62 |
Peak extension angular velocity (∘/s) | 339.31±22.07 | 332.22±26.83 |
Peak extension moment (Nm/kg) | 2.68±0.51 | 2.47±0.65 |
Peak positive power (W/kg) | 4.42±1.28 | 4.34±1.22 |
Peak negative power (W/kg) | 5.64±2.04 | 5.96±1.46 |
Knee | ||
ROM (∘) | 23.26±2.24 | 23.77±2.03 |
Peak flexion angular velocity (∘/s) | 413.82±62.89 | 407.29±65.09 |
Peak extension moment (Nm/kg) | 3.09±0.39 | 3.16±0.46 |
Peak positive power (W/kg) | 6.20±1.98 | 6.73±2.06 |
Peak negative power (W/kg) | 14.25±3.25 | 14.42±3.74 |
Ankle | ||
ROM (∘) | 33.74±2.86* | 35.18±3.12* |
Peak plantarflexion angular velocity (∘/s) | 559.51±49.94 | 533.72±49.21 |
Peak plantarflexion moment (Nm/kg) | 2.94±0.44 | 2.89±0.41 |
Peak positive power (W/kg) | 10.97±2.89 | 11.60±2.26 |
Peak negative power (W/kg) | 6.47±2.40 | 6.18±2.26 |
MTP joint | ||
ROM (∘) | 16.66±2.24* | 22.28±2.17* |
Peak flexion angular velocity (∘/s) | 360.25±76.06* | 504.51±82.20* |
Peak flexion moment (Nm/kg) | 0.861±0.10 | 0.872±0.12 |
Peak positive power (W/kg) | 1.56±0.44* | 2.23±0.45* |
Peak negative power (W/kg) | 3.58±0.92* | 4.30±0.57* |
3.2. Lower-limb work
No significant difference in the positive work and negative work for hip, knee, and ankle joints was observed between the two shoe conditions. For the MTP joint, the footwear with the plate structure showed smaller positive and negative work (Table 2).
Variable | With plate structure | Without plate structure |
---|---|---|
Hip | ||
Positive work (J/kg) | 0.073±0.016 | 0.068±0.025 |
Negative work (J/kg) | 0.344±0.136 | 0.391±0.13 |
Knee | ||
Positive work (J/kg) | 0.305±0.081 | 0.338±0.093 |
Negative work (J/kg) | 0.424±0.092 | 0.450±0.108 |
Ankle | ||
Positive work (J/kg) | 0.630±0.107 | 0.696±0.112 |
Negative work (J/kg) | 0.351±0.111 | 0.330±0.118 |
MTP joint | ||
Positive work (J/kg) | 0.025±0.006* | 0.041±0.016* |
Negative work (J/kg) | 0.152±0.037* | 0.203±0.042* |
3.3. Metabolism
Significant differences in the variables were observed between the two shoe conditions (P<0.05), with lower values of VO2, oxygen cost of transport (O2COT), metabolic power, and energetic cost of transport (ECOT) in the footwear with the plate structure compared with that without the plate structure (Table 3).
Variable | With plate structure | Without plate structure |
---|---|---|
VO2 [mL/(kg⋅min)] | 40.92±3.68* | 42.06±3.65* |
O2COT [mL/(kg⋅km)] | 183.60±16.10* | 184.21±15.98* |
RER (VCO2/VO2) | 0.88±0.04 | 0.88±0.04 |
Metabolic power (W/kg) | 14.14±1.35* | 14.29±1.33* |
ECOT (J/kg/m) | 3.84±0.37* | 3.96±0.37* |
4. Discussion
This study aimed to analyze the impact of footwear stiffness on lower-limb biomechanics, joint work, and overall metabolic efficiency. The study hypothesized that the footwear bending stiffness would redistribute lower-limb work. Stiffer footwear would limit the ROM of the MTP joint, decrease negative power, and work, thereby enhancing running economy. The findings of this study suggested that stiffer footwear decreased the ROM, angular velocity, and both negative and positive work of the MTP joint, as well as the ROM of the ankle joint. Running economy was significantly improved by the plate structure footwear. However, no significant difference was observed for other lower-limb joint work.
4.1. Influence of footwear bending stiffness on redistributing lower-limb work
The MTP joint experienced dorsiflexion in the stance phase, entailing absorbing energy. Hence, it has been identified as a site of energy dissipation during running.17,24 Prior studies have demonstrated that enhancing the bending stiffness of footwear can mitigate energy absorption or negative work in the MTP joint by stiffening the joint and constraining dorsiflexion.8,9,20,24
These experimental results were in line with those of previous studies demonstrating that MTP ROM, peak flexion angular velocity, and negative work were all lower under the stiffer footwear condition.16,17 However, the experimental results revealed significantly lower positive power and work under the stiffer footwear condition, contradicting findings of other researchers. Previous studies have noted larger positive work with increased bending stiffness. They suggested that heightened bending stiffness facilitated an extended push-off phase, enabling MTP joint plantarflexion and the generation of positive power.17 Additionally, these findings offered evidence supporting the concept of elastic energy storage and release by the plate.20 Nonetheless, Cigoja et al.20 observed that the energy stored and subsequently released by the carbon fiber plate itself had a minimal impact, contributing to only about 0.3% of the positive work performed at the ankle. This indicated that the primary role of the plate was not as a spring.
Some studies on the mechanism of foot movement during running suggested that the foot served as a lever arm for the triceps surae muscles. The point of force application could be shifted more anteriorly by increasing footwear bending stiffness, thereby enhancing a runner’s pushing-off efficiency.21 Nevertheless, the impact on the ankle joint remains uncertain.15,18,21 The experimental results demonstrated that increasing footwear bending stiffness restricted the ROM of the ankle joint. No significant difference in moment, power, and work was observed, which aligned with the findings of most previous studies. These studies suggested that increasing footwear stiffness did not influence ankle joint peak or average moment.18,19,25 Furthermore, Willwacher et al.21 showed that the average ankle joint moment decreased when running in moderately stiff shoes compared with less stiff or the stiffest footwear. Thus, a preliminary conclusion could be made in this study that footwear bending stiffness might not have influenced the power or work of ankle joints.
The experimental results did not show a significant difference in the angle, angular velocity, moment, power, and work of the knee and hip joints, which was consistent with most previous studies.14,15,18 No consensus could be reached regarding the knee joint power. Cigoja et al.20 found that enhancing longitudinal bending stiffness redistributed positive joint work from the knee to the MTP joint. Conversely, Hoogkamer et al.16 did not observe any significant difference in knee power and work.
4.2. Influence of footwear bending stiffness on running economy
The characteristics of shoes, such as mass, upper material, and midsole compliance, have the potential to impact the running economy. In this study, two experimental footwear models were identical except for the midsole plate structure. The experimental findings indicated that enhancing footwear bending stiffness led to an improvement in running economy. Subsequently, diverse results were found regarding the impact of increased longitudinal bending stiffness on running economy. These findings ranged from slight deteriorations,26 no significant differences,7,19,27 to significant improvements.11,20 The discrepancy in previous findings could be attributed to the unique metabolically optimal longitudinal bending stiffness of an individual. Oh and Park15 suggested that each person had a personalized optimal stiffness, which was influenced by their inherent MTP joint flexion. Their study revealed a noteworthy enhancement in performance when individuals wore shoes closely tailored to their critical stiffness. Additionally, Hunter et al.28 suggested that the wide variation in responses (ranging from 0.0% to 6.4% change) observed could be attributed to the differing optimal shoe stiffness for individual runners.
Speed should also be taken into consideration as a crucial influencing factor. Previous studies26 explored the impact of longitudinal bending stiffness on running economy at various speeds, revealing different trends in running economy depending on the speed being tested. Overall, it appeared that athletes might not uniformly respond to heightened bending stiffness.
4.3. Limitations
This study had certain limitations. First, only two pairs of test shoes were used, which might not have provided a wide enough range of footwear bending stiffness options, potentially missing out on shoes more suitable for running economy. Additionally, the muscular effects of the shoes were not explored in detail. Hence, future research should incorporate electromyogram (EMG) testing to further understand the impact of footwear on muscle mechanics.
5. Conclusions
In conclusion, increasing shoe bending stiffness significantly reduced the MTP joint ROM and angular velocity, thereby reducing negative power and work at the MTP joint and improving running economy for runners. However, no significant differences in other joint work were noted. Therefore, it could not be concluded that changing the shoe bending stiffness would redistribute lower joint work. This insight can particularly benefit shoe designers, developers, and researchers investigating footwear midsole structures and designs.
Acknowledgment
The laboratory, experimental shoes, and testing equipment for this study were all provided by the Li-Ning Sports Science Research Center.
ORCID
Zuoliang Liu https://orcid.org/0009-0008-6420-6661
Fan Yang https://orcid.org/0000-0001-7240-1936