The Influence of Medial and Lateral Forefoot Height Discrepancy on Lower Limb Biomechanical Characteristics during the Stance Phase of Running

Abstract

Previous research has predominantly focused on the biomechanical effects of anterior–posterior foot motion during running, with comparatively less attention given to medial–lateral foot motion and its impact on lower limb biomechanical characteristics. We recruited 18 healthy runners who wore five different types of running shoes: regular shoes (NS), those with a 6 mm and 9 mm medial–lateral height difference in the forefoot (M6, M9), and those with a 6 mm and 9 mm lateral–medial height difference (L6, L9). Biomechanical parameters of lower limb joints during the stance phase of running, including range of motion, peak angular velocity, peak moment, power, and work, were analyzed. We used paired-sample t-tests and one-dimensional statistical parametric mapping (SPM1D) to compare joint biomechanics between shoes with varying height differences and NS. Under the L6 condition, notable differences occurred in the hip and knee flexion–extension moments during landing and push-off, accompanied by a significant increase in ankle dorsiflexion work and a significant decrease in inversion–eversion work. In contrast, the M9 condition resulted in decreased hip flexion–extension peak moment, power, and work in the sagittal plane. These findings indicate that varying forefoot medial–lateral height differences in running shoes significantly impact lower limb joint dynamics during the stance phase, particularly the L6 condition, potentially reducing knee injury risk and aiding gait improvement for overpronators. The findings offer valuable insights for sports injury prevention and athletic footwear design. However, further research is needed to understand the underlying mechanisms and practical implications for sports injury prevention and performance enhancement.

1. Introduction

Running, as a common form of exercise, is gaining increasing popularity among individuals. However, with the surge in running trends, running-related injuries also demonstrate an increase that cannot be ignored, contradicting the theoretical health benefits of active exercise [1]. Over millions of years, humans have evolved features related to energy efficiency, strength, stability, and temperature regulation, enabling exceptional performance in long-distance running [2]. However, reports indicate that the injury rates for long-distance runners range from 30% to 80% annually, with foot and ankle injuries constituting approximately one-fourth of these injuries [3]. Various factors, including lower limb joint characteristics [4], foot morphology [5], footwear features [6], and training parameters [7], contribute to the occurrence of running injuries [8,9,10]. Running is a complex movement involving multiple joints and muscles, with the hip, knee, and ankle working in concert to form a kinetic chain. The foot, a core component of the musculoskeletal system, plays a crucial role in running. Many running injuries have been associated with abnormal frontal plane motions of the foot [11].

Previous studies have highlighted the potential significance of height differences between the medial and lateral aspects of the foot in influencing abnormal frontal plane motion. There is evidence indicating a strong correlation between foot height asymmetry and running injuries, which is particularly notable among avid runners [12]. Furthermore, research has found that an increased height difference between the medial and lateral aspects of the foot leads to an asymmetric distribution of forces during foot strikes in runners, thereby increasing the load on lower limb joints and subsequently elevating the risk of injury [13]. Differences in foot morphology and the angle of the foot landing during running may impact the ankle, contributing to ankle injuries. Non-contact ankle sprains are reported to be common, and improper foot positioning during walking or running may lead to ankle injuries, representing a potential injury mechanism [14,15]. During running, abnormal frontal plane motion of the foot can lead to aberrant loading and stress distribution on the knee and hip joints, consequently increasing the risk of injury. Research indicates that frontal plane abnormalities alter the trajectory of lower limb movement, resulting in additional stress on the knee joint during motion, thereby heightening the probability of soft tissue injury in the knee joint [16]. Regarding the knee, excessive dynamic knee valgus has been linked to anterior cruciate ligament tears and patellofemoral pain syndrome, and dynamic knee valgus can be altered by ankle range of motion and hip strength [17]. The hip joint, the proximal link in the lower extremity kinetic chain, shared with the knee joint, is essential, and hip joint dysfunction may serve as the foundation for injuries such as anterior cruciate ligament tears, iliotibial band syndrome, and patellofemoral joint pain [18].

There are numerous methods to prevent running injuries resulting from variations in frontal plane motion of the foot, with the use of footwear featuring height discrepancies between the medial and lateral sides gradually gaining popularity for this purpose [19,20]. Although regular footwear offers protection and support for the feet, they may excessively cushion and restrict foot movement, potentially impacting lower limb muscle function, leading to decreased muscle functionality and increased risk of musculoskeletal injuries [21,22]. Therefore, understanding individual foot morphology and selecting appropriate footwear is crucial. Research indicates that the transverse arch, through the action of the tarsal bones, contributes to over 40% of longitudinal foot stiffness. The robust structure of the human foot effectively supports walking or running, representing a key evolutionary adaptation among bipedal animals [23]. As an integral component of the forefoot, even subtle variations in the skeletal morphology of the foot arch, such as changes in thickness and stiffness, exert a significant influence on the biomechanics of gait [13]. Most runners typically focus solely on forefoot cushioning or heel length when selecting running shoes, often overlooking that the running stride encompasses not only forward and backward motion but also crucial lateral movements. These lateral movements are integral to the running process and are influenced by factors such as foot morphology, footwear design, and the biomechanics of the running surface. Understanding the running surface is also one of the factors to consider when choosing footwear. Different surfaces have varying impacts on running injuries, such as terrain material, slope, and hardness [24]. Therefore, in this study, we compared the differences in lower limb joint biomechanical characteristics while running using standard running shoes versus shoes with varying heights between the medial and lateral sides of the forefoot. This comparison aims to explore the relationships between different height discrepancies in running shoes and the risk of sports injuries. Our research goal is to provide recommendations for the prevention and reduction of sports injuries and to offer valuable insights for the design of new running shoes.

Based on anatomical principles, the medial–lateral height discrepancy of the forefoot is expected to influence ankle joint motion, which, in turn, affects hip and knee joint movements, impacting running mechanics. We hypothesize that: (1) The medial–lateral height discrepancy of the forefoot is anticipated to influence the biomechanical characteristics of lower limb joints during the stance phase of running, particularly with a more pronounced impact on joint moments and work. (2) The biomechanical differences in lower extremity joints may not necessarily amplify with increasing height discrepancy.

2. Materials and Methods

2.1. Participant Recruitment and Ethics

The sample size was calculated using G*Power 3.1 (Franz Faul, Germany) with two tails, an effect size dz of 0.8, an α error probability of 0.05, and a power (1 − β) of 0.8, resulting in a minimum requirement of 15 participants for the study. Recruitment information for volunteers was disseminated online by members of the research team, where volunteers registered their names and contact details. Researchers collected necessary personal information including name, age, gender, and basic health status, and screened individuals based on trial inclusion and exclusion criteria. Throughout this period, volunteers remained uninformed about the specifics of the experiment, only being aware of a running task. Initially, we recruited a total of 23 participants. After screening, the final experiment recruited 18 healthy male running enthusiasts (age: 21.65 ± 1.41 years, height: 1.73 ± 0.03 m, weight: 67.70 ± 8.20 kg, body mass index (BMI): 22.40 ± 2.30 kg/m²) for the running test. Inclusion criteria for participants were as follows: (1) age between 18 and 60 years; (2) male; (3) absence of health issues and/or neuromuscular diseases and/or known gait abnormalities; (4) participants not sustaining lower limb injuries within the past 6 months; (5) right leg dominance (defined as the preferred leg for kicking a football); (6) running technique involving heel-to-toe movement. Exclusion criteria for participants were as follows: (1) age below 18 or above 60 years; (2) female; (3) presence of neuromuscular, musculoskeletal diseases, and gait abnormalities; (4) lower limb injury within the past six months; (5) involuntary participation in the trial; (6) left leg dominance; (7) running technique involving forefoot striking. Prior to the experiment, all participants received and signed informed consent forms approved by the Institutional Review Board. Ethical procedures for this study were approved by the Ethics Committee of Ningbo University, number RAGH20231210.

2.2. Experimental Shoes

Five types of shoes were used in the experiment: neutral shoes (NS), running shoes with a 6 mm increase in the medial forefoot height (M6), running shoes with a 9 mm increase in the medial forefoot height (M9), running shoes with a 6 mm increase in the lateral forefoot height (L6), and running shoes with a 9 mm increase in the lateral forefoot height (L9). The shoes used in the experiment were of the Anta brand. Apart from the insoles, the shoes were identical in all other aspects, and all shoes were of European size 41. The differences in medial and lateral forefoot heights were achieved through specifically designed insoles, as illustrated in Figure 1. Each participant was required to test all five pairs of shoes, with the order of shoe wear randomly assigned. All testing was conducted during the same time of day to minimize potential variations in data due to circadian rhythms.

2.3. The Experimental Process

The experiment was conducted using a randomized crossover design. This study was conducted in the Biomechanics Laboratory at Ningbo University. Participants’ height and weight were assessed using a measuring tape and a calibrated scale, respectively. A motion capture system consisting of eight infrared cameras (Vicon Metrics Ltd., Oxford, UK) was employed to track movement trajectories at a frequency of 1000 Hz. Concurrently, a three-dimensional force plate (AMTI, Watertown, MA, USA) collected kinetic data at a frequency of 200 Hz. These systems were synchronized to allow for simultaneous data collection. An infrared timing light (Brower Timing System, Draper, UT, USA) was utilized to monitor the running time for each trial. The infrared light was placed at both ends of the force platform, spaced 3.3 m apart, and the running speed was calculated accordingly. Before the formal experiment, each subject performed five trial runs in neutral shoes to determine their average self-selected speed, which served as the baseline. During tests with different shoe conditions, a 5% speed variation was allowed to control the self-selected speed. Prior to the commencement of the official experiment, participants were required to warm up by running at a self-selected pace on a treadmill for 10 min, followed by a 5 min period to familiarize themselves with the experimental procedure and environment. Participants wore uniform compression garments and leggings.

Following previous research protocols, 38 reflective markers were placed on the participants’ trunks and lower limbs [25] (as shown in Figure 2a). After marker application, participants were instructed to stand on the force plate with feet together, maintaining a standard anatomical posture and facing forward to collect static data. During the dynamic phase of the experiment, participants were guided to run at a self-selected speed on a 10 m track with a 3D force plate positioned in the center. To accurately record the speed of each run, an infrared timing light was placed at the midpoint of the track, spaced 3.3 m apart. Throughout the experiment, participants selected their comfortable running pace and were required to step on the force plate with their right foot during the run, completing five valid trials per foot. A valid trial was defined as the heel making initial contact followed by toe-off, with the entire foot remaining within the boundaries of the force plate throughout the process. To ensure the accuracy of the data, participants were explicitly instructed to maintain their natural running posture and avoid any intentional alteration of their posture. A 30 s rest interval was provided between each trial to allow participants to adjust.

2.4. Data Preprocessing

This study focused on analyzing the kinematic and kinetic characteristics of the hip, knee, and ankle joints in the sagittal plane during running, as well as the performance of the hip and ankle joints in the coronal plane [26]. The existing literature has revealed significant variability in the sagittal and frontal plane movements of the lower extremities during running activities [26]. To accurately capture these variations, marker trajectory data were filtered using a zero-phase fourth-order Butterworth low-pass filter with a cutoff frequency of 12 Hz. Subsequently, Matlab R2018b software was utilized to convert the marker trajectory data from C3D file format to a format compatible with OpenSim 3.2016 software (.mot and .trc formats), and then import them into OpenSim for further data processing. The muscle–skeletal model provided by OpenSim (gait 2392) was employed during this process.

To ensure the accuracy of the model, static calibration was performed based on the positions of markers on the subjects and their respective body weights to adjust the model scale. The static weight of each marker was manually adjusted based on the root mean square (RMS) error between the experimental marker points and virtual marker points in the model, ensuring that the error was less than 0.02. Segment lengths and masses were scaled based on the subject’s anthropometric measurements. These included lengths of the femur, tibia, and foot segments, as well as their respective mass properties. The joint centers were defined using the positions of anatomical landmarks placed on the subjects. These landmarks were used to establish coordinate systems for each body segment and to define the location of joint centers. The adjusted scaled model was subsequently applied to appropriate analyses to calculate the data. Joint angles were computed using the inverse kinematics (IK) tool in OpenSim, while joint moments were estimated using the inverse dynamics (ID) tool in OpenSim. This methodology aims to provide more precise data support for understanding the kinematics and kinetics of lower limb joints during running.

2.5. Statistical Analysis

Statistical analysis was conducted using SPSS 26.0 software (SPSS, Chicago, IL, USA). Descriptive statistics, including mean and standard deviation (SD), were provided. Paired sample t-tests were employed to assess the changes in lower limb joint biomechanical parameters when wearing insoles with different height differences. MATLAB version 2018b software (MathWorks Inc., Natick, MA, USA) was utilized for statistical analysis. The significance level was set at α = 0.05, where p < 0.05 indicated statistical significance. During the stance phase of running, paired sample t-tests were conducted to analyze the primary effects of joint angles, angular velocities, peak moments, power, and work for the hip, knee, and ankle joints. Specifically, analyses were performed in the sagittal and coronal planes for the hip and ankle joints, and in the sagittal plane for the knee joint. This analysis aimed to explore differences between insoles with different height differences and regular insoles.

Given the one-dimensional time-varying nature of joint kinematics and kinetics, paired sample t-tests were conducted using the one-dimensional statistical parametric mapping (SPM1D) method. This method compared joint angles, angular velocities, moments, and power during the stance phase of running. The significance level was set at α = 0.05, where p < 0.05 indicated statistical significance.

3. Results

3.1. Hip Joint

The biomechanical characteristics of lower limb joints between shoes with different height differences and normal height difference running shoes were analyzed using paired samples t-test. According to Table 1, significant differences were found in the hip joint hip abduction range during the landing phase between M6 and M9 conditions compared to NS (p = 0.021, p = 0.039). Differences in hip joint flexion–extension power were observed under M6 (p = 0.021) and M9 conditions (p = 0.006), while hip joint extension peak moments differed under L6 (p = 0.045).

As shown in Table 2, during the push-off phase, significant differences in hip joint peak angular velocities of extension (p = 0.008) and abduction (p = 0.016) were noted under L6 compared to NS. The hip joint flexion–extension range differed under M6 (p = 0.014). Significant differences in hip joint minimum power (p = 0.006), external peak moment (p = 0.004), and external work (p = 0.022) were found under M6. Under M9, hip joint flexion–extension power (p = 0.002), external peak moment (p = 0.002), power (p = 0.001), and work (p = 0.001) showed significant differences. L6 showed differences in hip joint extension power (p = 0.011), peak moment (p = 0.009), work (p = 0.033), and flexion–extension power (p = 0.019). Significant differences in hip joint external power were observed under M9 (p = 0.042). All significant differences in the hip joint results were presented in Table 3.

Table 1. Descriptive analysis of the range of motion, peak angular velocity, peak moment, power, and work of the hip, knee, and ankle joints during the landing phase (defined as 0–42% of the stance phase) of running while wearing running shoes with different forefoot height differences.

Joint	Variables		Neutral	Medial		Lateral
				6 mm	9 mm	6 mm	9 mm
Hip	Sagittal	ROM (°)	3.26 ± 1.29	3.24 ± 1.41	3.17 ± 1.39	2.99 ± 0.88	3.06 ± 0.79
		Peak angular velocity (rad/s)	2.74 ± 0.85	2.50 ± 0.66	2.31 ± 0.32	2.59 ± 0.86	2.69 ± 0.62
		Peak moment (Nm/kg)	−3.33 ± 1.27	3.60 ± 0.69	−3.31 ± 0.73	−4.39 ± 0.95 *	−3.94 ± 1.14
		Maximum power (w/kg)	8.96 ± 3.97	7.30 ± 1.78	8.43 ± 5.75	8.32 ± 5.41	9.89 ± 5.85
		Minimum power (w/kg)	−7.70 ± 4.81	−6.06 ± 2.21	−5.08 ± 1.69	−7.06 ± 2.71	−6.58 ± 3.30
		Positive work (J/kg)	0.22 ± 0.08	0.25 ± 0.13	0.28 ± 0.18	0.26 ± 0.16	0.29 ± 0.17
		Negative work (J/kg)	−0.38 ± 0.23	−0.29 ± 0.14	−0.24 ± 0.12	−0.36 ± 0.17	−0.33 ± 0.21
	Frontal	ROM (°)	12.57 ± 4.03	10.69 ± 2.76 *	10.51 ± 3.78 *	11.19 ± 4.03	11.80 ± 3.10
		Peak angular velocity (rad/s)	3.95 ± 1.29	3.39 ± 1.00	3.31 ± 1.18	3.61 ± 1.13	3.68 ± 0.93
		Peak moment (Nm/kg)	−1.82 ± 0.31	−1.78 ± 0.32	−1.72 ± 0.21	−1.95 ± 0.21	−1.98 ± 0.61
		Maximum power (w/kg)	0.29 ± 0.42	0.23 ± 0.60	0.21 ± 0.71	0.13 ± 0.63	0.20 ± 1.08
		Minimum power (w/kg)	−6.33 ± 3.26	−4.20 ± 1.40 *	−4.09 ± 2.95 *	−5.90 ± 2.32	−6.12 ± 2.89
		Positive work (J/kg)	0.01 ± 0.01	0.01 ± 0.02	0.01 ± 0.02	0.01 ± 0.01	0.01 ± 0.02
		Negative work (J/kg)	−0.43 ± 0.19	−0.33 ± 0.09	−0.34 ± 0.17	−0.52 ± 0.15	−0.55 ± 0.23
Knee	Sagittal	ROM (°)	33.57 ± 3.19	32.70 ± 3.66	31.81 ± 4.77	31.81 ± 3.56	32.97 ± 2.66
		Peak angular velocity (rad/s)	−11.03 ± 0.74	−10.92 ± 0.64	−10.74 ± 0.64	−11.06 ± 1.05	−10.85 ± 0.62
		Peak moment (Nm/kg)	−0.64 ± 0.27	−0.80 ± 0.38	−0.52 ± 0.50	−1.19 ± 0.47 *	−0.81 ± 0.41
		Maximum power (w/kg)	3.84 ± 1.99	5.10 ± 2.38	4.04 ± 2.40	6.83 ± 2.10 *	5.53 ± 2.83 *
		Minimum power (w/kg)	−15.64 ± 2.65	14.04 ± 1.55	−14.55 ± 3.21	−8.36 ± 4.25 *	−13.53 ± 9.20
		Positive work (J/kg)	0.20 ± 0.14	0.22 ± 0.10	0.21 ± 0.18	0.38 ± 0.16 *	0.30 ± 0.28
		Negative work (J/kg)	−1.06 ± 0.24	−0.93 ± 0.22	−0.95 ± 0.28	−0.57 ± 0.31 *	−1.00 ± 0.73
Ankle	Sagittal	ROM (°)	11.62 ± 2.64	11.67 ± 1.49	13.41 ± 3.96	10.05 ± 1.52	9.90 ± 1.33
		Peak angular velocity (rad/s)	−5.87 ± 1.03	−6.48 ± 1.31	−6.00 ± 1.53	−6.15 ± 1.78	−5.63 ± 1.38
		Peak moment (Nm/kg)	−2.30 ± 0.39	−2.46 ± 0.43	−2.44 ± 0.42	−2.69 ± 0.48	−2.73 ± 0.82
		Maximum power (w/kg)	2.03 ± 1.79	2.67 ± 2.47	1.42 ± 1.74	3.87 ± 2.89	1.48 ± 1.42
		Minimum power (w/kg)	−14.99 ± 4.09	−13.85 ± 2.27	−14.60 ± 4.16	−15.57 ± 4.22	−15.65 ± 5.27
		Positive work (J/kg)	0.06 ± 0.05	0.10 ± 0.10	0.05 ± 0.06	0.06 ± 0.02	0.06 ± 0.05
		Negative work (J/kg)	−0.67 ± 0.22	−0.82 ± 0.24	−0.82 ± 0.24	−0.79 ± 0.29	−0.81 ± 0.35
	Frontal	ROM (°)	6.63 ± 1.69	7.21 ± 3.75	5.84 ± 1.55	8.11 ± 3.85	8.04 ± 3.69
		Peak angular velocity (rad/s)	−4.08 ± 1.30	−4.69 ± 2.16	−4.18 ± 1.70	−3.02 ± 0.86 *	−4.97 ± 2.57
		Peak moment (Nm/kg)	0.59 ± 0.53	0.63 ± 0.60	0.67 ± 0.60	0.69 ± 0.59	0.71 ± 0.69
		Maximum power (w/kg)	1.91 ± 1.59	1.07 ± 1.04	1.73 ± 1.50	0.93 ± 0.44	0.91 ± 0.74
		Minimum power (w/kg)	−0.90 ± 0.62	−1.48 ± 1.50	−0.96 ± 0.60	−0.65 ± 0.62	−1.20 ± 1.18
		Positive work (J/kg)	0.07 ± 0.07	0.04 ± 0.05	0.07 ± 0.06	0.04 ± 0.05	0.03 ± 0.05
		Negative work (J/kg)	−0.04 ± 0.03	−0.10 ± 0.12	−0.05 ± 0.03	−0.04 ± 0.04	−0.10 ± 0.12

Note: Bold and * in the table indicate statistically significant differences compared to NS, with a significance level set at p < 0.05.

Table 1. Descriptive analysis of the range of motion, peak angular velocity, peak moment, power, and work of the hip, knee, and ankle joints during the landing phase (defined as 0–42% of the stance phase) of running while wearing running shoes with different forefoot height differences.

Joint	Variables		Neutral	Medial		Lateral
				6 mm	9 mm	6 mm	9 mm
Hip	Sagittal	ROM (°)	3.26 ± 1.29	3.24 ± 1.41	3.17 ± 1.39	2.99 ± 0.88	3.06 ± 0.79
		Peak angular velocity (rad/s)	2.74 ± 0.85	2.50 ± 0.66	2.31 ± 0.32	2.59 ± 0.86	2.69 ± 0.62
		Peak moment (Nm/kg)	−3.33 ± 1.27	3.60 ± 0.69	−3.31 ± 0.73	−4.39 ± 0.95 *	−3.94 ± 1.14
		Maximum power (w/kg)	8.96 ± 3.97	7.30 ± 1.78	8.43 ± 5.75	8.32 ± 5.41	9.89 ± 5.85
		Minimum power (w/kg)	−7.70 ± 4.81	−6.06 ± 2.21	−5.08 ± 1.69	−7.06 ± 2.71	−6.58 ± 3.30
		Positive work (J/kg)	0.22 ± 0.08	0.25 ± 0.13	0.28 ± 0.18	0.26 ± 0.16	0.29 ± 0.17
		Negative work (J/kg)	−0.38 ± 0.23	−0.29 ± 0.14	−0.24 ± 0.12	−0.36 ± 0.17	−0.33 ± 0.21
	Frontal	ROM (°)	12.57 ± 4.03	10.69 ± 2.76 *	10.51 ± 3.78 *	11.19 ± 4.03	11.80 ± 3.10
		Peak angular velocity (rad/s)	3.95 ± 1.29	3.39 ± 1.00	3.31 ± 1.18	3.61 ± 1.13	3.68 ± 0.93
		Peak moment (Nm/kg)	−1.82 ± 0.31	−1.78 ± 0.32	−1.72 ± 0.21	−1.95 ± 0.21	−1.98 ± 0.61
		Maximum power (w/kg)	0.29 ± 0.42	0.23 ± 0.60	0.21 ± 0.71	0.13 ± 0.63	0.20 ± 1.08
		Minimum power (w/kg)	−6.33 ± 3.26	−4.20 ± 1.40 *	−4.09 ± 2.95 *	−5.90 ± 2.32	−6.12 ± 2.89
		Positive work (J/kg)	0.01 ± 0.01	0.01 ± 0.02	0.01 ± 0.02	0.01 ± 0.01	0.01 ± 0.02
		Negative work (J/kg)	−0.43 ± 0.19	−0.33 ± 0.09	−0.34 ± 0.17	−0.52 ± 0.15	−0.55 ± 0.23
Knee	Sagittal	ROM (°)	33.57 ± 3.19	32.70 ± 3.66	31.81 ± 4.77	31.81 ± 3.56	32.97 ± 2.66
		Peak angular velocity (rad/s)	−11.03 ± 0.74	−10.92 ± 0.64	−10.74 ± 0.64	−11.06 ± 1.05	−10.85 ± 0.62
		Peak moment (Nm/kg)	−0.64 ± 0.27	−0.80 ± 0.38	−0.52 ± 0.50	−1.19 ± 0.47 *	−0.81 ± 0.41
		Maximum power (w/kg)	3.84 ± 1.99	5.10 ± 2.38	4.04 ± 2.40	6.83 ± 2.10 *	5.53 ± 2.83 *
		Minimum power (w/kg)	−15.64 ± 2.65	14.04 ± 1.55	−14.55 ± 3.21	−8.36 ± 4.25 *	−13.53 ± 9.20
		Positive work (J/kg)	0.20 ± 0.14	0.22 ± 0.10	0.21 ± 0.18	0.38 ± 0.16 *	0.30 ± 0.28
		Negative work (J/kg)	−1.06 ± 0.24	−0.93 ± 0.22	−0.95 ± 0.28	−0.57 ± 0.31 *	−1.00 ± 0.73
Ankle	Sagittal	ROM (°)	11.62 ± 2.64	11.67 ± 1.49	13.41 ± 3.96	10.05 ± 1.52	9.90 ± 1.33
		Peak angular velocity (rad/s)	−5.87 ± 1.03	−6.48 ± 1.31	−6.00 ± 1.53	−6.15 ± 1.78	−5.63 ± 1.38
		Peak moment (Nm/kg)	−2.30 ± 0.39	−2.46 ± 0.43	−2.44 ± 0.42	−2.69 ± 0.48	−2.73 ± 0.82
		Maximum power (w/kg)	2.03 ± 1.79	2.67 ± 2.47	1.42 ± 1.74	3.87 ± 2.89	1.48 ± 1.42
		Minimum power (w/kg)	−14.99 ± 4.09	−13.85 ± 2.27	−14.60 ± 4.16	−15.57 ± 4.22	−15.65 ± 5.27
		Positive work (J/kg)	0.06 ± 0.05	0.10 ± 0.10	0.05 ± 0.06	0.06 ± 0.02	0.06 ± 0.05
		Negative work (J/kg)	−0.67 ± 0.22	−0.82 ± 0.24	−0.82 ± 0.24	−0.79 ± 0.29	−0.81 ± 0.35
	Frontal	ROM (°)	6.63 ± 1.69	7.21 ± 3.75	5.84 ± 1.55	8.11 ± 3.85	8.04 ± 3.69
		Peak angular velocity (rad/s)	−4.08 ± 1.30	−4.69 ± 2.16	−4.18 ± 1.70	−3.02 ± 0.86 *	−4.97 ± 2.57
		Peak moment (Nm/kg)	0.59 ± 0.53	0.63 ± 0.60	0.67 ± 0.60	0.69 ± 0.59	0.71 ± 0.69
		Maximum power (w/kg)	1.91 ± 1.59	1.07 ± 1.04	1.73 ± 1.50	0.93 ± 0.44	0.91 ± 0.74
		Minimum power (w/kg)	−0.90 ± 0.62	−1.48 ± 1.50	−0.96 ± 0.60	−0.65 ± 0.62	−1.20 ± 1.18
		Positive work (J/kg)	0.07 ± 0.07	0.04 ± 0.05	0.07 ± 0.06	0.04 ± 0.05	0.03 ± 0.05
		Negative work (J/kg)	−0.04 ± 0.03	−0.10 ± 0.12	−0.05 ± 0.03	−0.04 ± 0.04	−0.10 ± 0.12

Note: Bold and * in the table indicate statistically significant differences compared to NS, with a significance level set at p < 0.05.

Table 2. Descriptive analysis of the range of motion, peak angular velocity, peak moment, power, and work of the hip, knee, and ankle joints during the push-off phase (defined as 43–100% of the stance phase) while wearing running shoes with different forefoot height differences.

Joint	Variables		Neutral	Inside		Outside
				6 mm	9 mm	6 mm	9 mm
Hip	Sagittal	ROM (°)	39.46 ± 6.59	38.51 ± 4.59	38.79 ± 6.12	37.24 ± 6.15	37.95 ± 6.24
		Peak angular velocity (rad/s)	−3.00 ± 1.34	−2.37 ± 1.30	−3.53 ± 1.31	−1.93 ± 1.90 *	−3.14 ± 1.36
		Peak moment (Nm/kg)	−3.10 ± 0.48	−2.89 ± 0.46	−2.83 ± 0.73	−3.80 ± 0.56 *	−3.26 ± 0.86
		Maximum power (w/kg)	15.06 ± 3.78	15.31 ± 4.71	13.20 ± 6.05	19.89 ± 6.04 *	17.32 ± 7.22
		Minimum power (w/kg)	7.90 ± 4.00	3.92 ± 1.61 *	5.20 ± 4.71 *	5.40 ± 5.43 *	7.61 ± 2.70
		Positive work (J/kg)	3.48 ± 1.06	3.10 ± 1.04	2.91 ± 1.75	4.52 ± 1.28 *	3.98 ± 1.61
		Negative work (J/kg)	/	/	/	/	/
	Frontal	ROM (°)	9.52 ± 4.00	7.16 ± 3.98 *	7.19 ± 4.41 *	8.05 ± 3.88	8.29 ± 3.88
		Peak angular velocity (rad/s)	1.61 ± 0.47	1.80 ± 0.54	1.68 ± 0.58	2.13 ± 0.64 *	1.90 ± 0.55
		Peak moment (Nm/kg)	−1.61 ± 0.13	−1.35 ± 0.24 *	−1.34 ± 0.24 *	−1.60 ± 0.18	−1.76 ± 0.43
		Maximum power (w/kg)	2.19 ± 0.84	1.52 ± 0.86	1.17 ± 0.33 *	1.48 ± 0.51 *	1.67 ± 0.49
		Minimum power (w/kg)	−2.43 ± 0.75	−2.32 ± 0.67	−2.16 ± 1.06	−3.24 ± 0.74 *	−3.22 ± 1.51 *
		Positive work (J/kg)	0.24 ± 0.09	0.16 ± 0.11 *	0.13 ± 0.07 *	0.17 ± 0.10	0.19 ± 0.08
		Negative work (J/kg)	−0.14 ± 0.08	−0.11 ± 0.06	−0.13 ± 0.07	−0.17 ± 0.06	−0.16 ± 0.07
Knee	Sagittal	ROM (°)	25.55 ± 2.95	24.89 ± 4.53	24.39 ± 2.83	22.18 ± 3.96 *	22.84 ± 2.47
		Peak angular velocity (rad/s)	6.89 ± 0.65	6.29 ± 1.17	6.43 ± 0.60	6.23 ± 0.99 *	6.35 ± 0.54
		Peak moment (Nm/kg)	1.85 ± 0.27	1.61 ± 0.50	1.83 ± 0.57	1.10 ± 0.54 *	1.65 ± 1.08
		Maximum power (w/kg)	5.21 ± 1.80	4.50 ± 2.90	5.04 ± 3.17	2.16 ± 1.38 *	4.34 ± 3.07
		Minimum power (w/kg)	−5.99 ± 1.54	−5.02 ± 1.30	−4.20 ± 1.98 *	−5.25 ± 1.13	−5.51 ± 2.19
		Positive work (J/kg)	0.45 ± 0.25	0.38 ± 0.32	0.55 ± 0.47	0.15 ± 0.11 *	0.38 ± 0.36
		Negative work (J/kg)	−0.48 ± 0.10	−0.41 ± 0.11	−0.34 ± 0.23 *	−0.55 ± 0.14	−0.47 ± 0.32
Ankle	Sagittal	ROM (°)	23.45 ± 5.83	27.82 ± 7.15	29.08 ± 5.96 *	22.56 ± 4.58	24.13 ± 5.38
		Peak angular velocity (rad/s)	7.31 ± 0.90	6.39 ± 0.63 *	6.32 ± 1.27 *	6.64 ± 0.98 *	6.34 ± 0.95 *
		Peak moment (Nm/kg)	−3.08 ± 0.25	−3.19 ± 0.39	−3.17 ± 0.41	−3.33 ± 0.35	−3.51 ± 0.94
		Maximum power (w/kg)	17.28 ± 4.31	16.40 ± 6.44	17.96 ± 6.73	14.06 ± 3.23 *	17.61 ± 6.69
		Minimum power (w/kg)	−16.71 ± 3.95	−16.26 ± 3.46	−15.47 ± 4.67	−18.71 ± 3.63	−17.84 ± 5.89
		Positive work (J/kg)	1.62 ± 0.53	1.80 ± 0.87	2.16 ± 1.06	1.42 ± 0.40	1.88 ± 0.72
		Negative work (J/kg)	−1.54 ± 0.33	−1.46 ± 0.45	−1.23 ± 0.50 *	−1.86 ± 0.35 *	−1.59 ± 0.44
	Frontal	ROM (°)	6.84 ± 2.04	7.89 ± 2.09	8.30 ± 2.49	5.75 ± 1.73	5.58 ± 1.94
		Peak angular velocity (rad/s)	3.68 ± 1.10	3.48 ± 0.90	3.41 ± 0.89	0.92 ± 0.31 *	3.01 ± 0.90
		Peak moment (Nm/kg)	0.58 ± 0.66	0.66 ± 0.73	0.66 ± 0.72	0.72 ± 0.67	0.78 ± 0.73
		Maximum power (w/kg)	2.02 ± 1.49	2.07 ± 1.50	1.89 ± 1.37	0.43 ± 0.33	1.36 ± 1.83
		Minimum power (w/kg)	−1.42 ± 1.04	−1.29 ± 1.01	−0.81 ± 0.59 *	−0.43 ± 0.39 *	−1.26 ± 0.97
		Positive work (J/kg)	0.16 ± 0.13	0.20 ± 0.16	0.19 ± 0.15	0.04 ± 0.03 *	0.13 ± 0.12
		Negative work (J/kg)	−0.10 ± 0.07	−0.09 ± 0.07	−0.06 ± 0.04 *	−0.03 ± 0.03 *	−0.07 ± 0.06

Note: Bold and * in the table indicate statistically significant differences compared to NS, with a significance level set at p < 0.05.

Table 2. Descriptive analysis of the range of motion, peak angular velocity, peak moment, power, and work of the hip, knee, and ankle joints during the push-off phase (defined as 43–100% of the stance phase) while wearing running shoes with different forefoot height differences.

Joint	Variables		Neutral	Inside		Outside
				6 mm	9 mm	6 mm	9 mm
Hip	Sagittal	ROM (°)	39.46 ± 6.59	38.51 ± 4.59	38.79 ± 6.12	37.24 ± 6.15	37.95 ± 6.24
		Peak angular velocity (rad/s)	−3.00 ± 1.34	−2.37 ± 1.30	−3.53 ± 1.31	−1.93 ± 1.90 *	−3.14 ± 1.36
		Peak moment (Nm/kg)	−3.10 ± 0.48	−2.89 ± 0.46	−2.83 ± 0.73	−3.80 ± 0.56 *	−3.26 ± 0.86
		Maximum power (w/kg)	15.06 ± 3.78	15.31 ± 4.71	13.20 ± 6.05	19.89 ± 6.04 *	17.32 ± 7.22
		Minimum power (w/kg)	7.90 ± 4.00	3.92 ± 1.61 *	5.20 ± 4.71 *	5.40 ± 5.43 *	7.61 ± 2.70
		Positive work (J/kg)	3.48 ± 1.06	3.10 ± 1.04	2.91 ± 1.75	4.52 ± 1.28 *	3.98 ± 1.61
		Negative work (J/kg)	/	/	/	/	/
	Frontal	ROM (°)	9.52 ± 4.00	7.16 ± 3.98 *	7.19 ± 4.41 *	8.05 ± 3.88	8.29 ± 3.88
		Peak angular velocity (rad/s)	1.61 ± 0.47	1.80 ± 0.54	1.68 ± 0.58	2.13 ± 0.64 *	1.90 ± 0.55
		Peak moment (Nm/kg)	−1.61 ± 0.13	−1.35 ± 0.24 *	−1.34 ± 0.24 *	−1.60 ± 0.18	−1.76 ± 0.43
		Maximum power (w/kg)	2.19 ± 0.84	1.52 ± 0.86	1.17 ± 0.33 *	1.48 ± 0.51 *	1.67 ± 0.49
		Minimum power (w/kg)	−2.43 ± 0.75	−2.32 ± 0.67	−2.16 ± 1.06	−3.24 ± 0.74 *	−3.22 ± 1.51 *
		Positive work (J/kg)	0.24 ± 0.09	0.16 ± 0.11 *	0.13 ± 0.07 *	0.17 ± 0.10	0.19 ± 0.08
		Negative work (J/kg)	−0.14 ± 0.08	−0.11 ± 0.06	−0.13 ± 0.07	−0.17 ± 0.06	−0.16 ± 0.07
Knee	Sagittal	ROM (°)	25.55 ± 2.95	24.89 ± 4.53	24.39 ± 2.83	22.18 ± 3.96 *	22.84 ± 2.47
		Peak angular velocity (rad/s)	6.89 ± 0.65	6.29 ± 1.17	6.43 ± 0.60	6.23 ± 0.99 *	6.35 ± 0.54
		Peak moment (Nm/kg)	1.85 ± 0.27	1.61 ± 0.50	1.83 ± 0.57	1.10 ± 0.54 *	1.65 ± 1.08
		Maximum power (w/kg)	5.21 ± 1.80	4.50 ± 2.90	5.04 ± 3.17	2.16 ± 1.38 *	4.34 ± 3.07
		Minimum power (w/kg)	−5.99 ± 1.54	−5.02 ± 1.30	−4.20 ± 1.98 *	−5.25 ± 1.13	−5.51 ± 2.19
		Positive work (J/kg)	0.45 ± 0.25	0.38 ± 0.32	0.55 ± 0.47	0.15 ± 0.11 *	0.38 ± 0.36
		Negative work (J/kg)	−0.48 ± 0.10	−0.41 ± 0.11	−0.34 ± 0.23 *	−0.55 ± 0.14	−0.47 ± 0.32
Ankle	Sagittal	ROM (°)	23.45 ± 5.83	27.82 ± 7.15	29.08 ± 5.96 *	22.56 ± 4.58	24.13 ± 5.38
		Peak angular velocity (rad/s)	7.31 ± 0.90	6.39 ± 0.63 *	6.32 ± 1.27 *	6.64 ± 0.98 *	6.34 ± 0.95 *
		Peak moment (Nm/kg)	−3.08 ± 0.25	−3.19 ± 0.39	−3.17 ± 0.41	−3.33 ± 0.35	−3.51 ± 0.94
		Maximum power (w/kg)	17.28 ± 4.31	16.40 ± 6.44	17.96 ± 6.73	14.06 ± 3.23 *	17.61 ± 6.69
		Minimum power (w/kg)	−16.71 ± 3.95	−16.26 ± 3.46	−15.47 ± 4.67	−18.71 ± 3.63	−17.84 ± 5.89
		Positive work (J/kg)	1.62 ± 0.53	1.80 ± 0.87	2.16 ± 1.06	1.42 ± 0.40	1.88 ± 0.72
		Negative work (J/kg)	−1.54 ± 0.33	−1.46 ± 0.45	−1.23 ± 0.50 *	−1.86 ± 0.35 *	−1.59 ± 0.44
	Frontal	ROM (°)	6.84 ± 2.04	7.89 ± 2.09	8.30 ± 2.49	5.75 ± 1.73	5.58 ± 1.94
		Peak angular velocity (rad/s)	3.68 ± 1.10	3.48 ± 0.90	3.41 ± 0.89	0.92 ± 0.31 *	3.01 ± 0.90
		Peak moment (Nm/kg)	0.58 ± 0.66	0.66 ± 0.73	0.66 ± 0.72	0.72 ± 0.67	0.78 ± 0.73
		Maximum power (w/kg)	2.02 ± 1.49	2.07 ± 1.50	1.89 ± 1.37	0.43 ± 0.33	1.36 ± 1.83
		Minimum power (w/kg)	−1.42 ± 1.04	−1.29 ± 1.01	−0.81 ± 0.59 *	−0.43 ± 0.39 *	−1.26 ± 0.97
		Positive work (J/kg)	0.16 ± 0.13	0.20 ± 0.16	0.19 ± 0.15	0.04 ± 0.03 *	0.13 ± 0.12
		Negative work (J/kg)	−0.10 ± 0.07	−0.09 ± 0.07	−0.06 ± 0.04 *	−0.03 ± 0.03 *	−0.07 ± 0.06

Note: Bold and * in the table indicate statistically significant differences compared to NS, with a significance level set at p < 0.05.

According to the SPM1D results presented in Figure 3 and Figure 4, significant differences were observed compared to the NS condition. Under the M6 condition, hip flexion angular velocity differed significantly at 21–24% (p = 0.021) and hip abduction angular velocity at 81–90% (p = 0.014). For the M9 condition, significant differences were found in the hip flexion angle at 0–15% (p = 0.001) and hip flexion angular velocity at 22–32% (p = 0.002). Under the L6 condition, significant differences were observed in hip flexion angular velocity at 17–19% (p = 0.037). In terms of moments and power, under the M6 condition, the hip abduction moment showed significant differences at 50–54% (p = 0.016) and hip abduction power at 83–91% (p < 0.001). For the M9 condition, significant differences were found in the hip abduction moment at 52–74% (p < 0.001) and hip abduction power at 78–90% (p < 0.001). Under the L6 condition, significant differences were observed in the hip flexion moment at 25–81% (p = 0.001, p = 0.027, p < 0.001), hip abduction moment at 13–16% (p = 0.036), hip flexion power at 17–20% (p = 0.019), and hip abduction power at 79–83% (p = 0.007).

3.2. Knee Joint

According to Table 1, compared to NS, different shoe conditions did not show statistically significant main effects on knee joint kinematics during the landing phase. However, significant differences were found under the L6 condition in the knee joint extension peak moment (p = 0.003), maximum power (p = 0.002), minimum power (p = 0.001), flexion work (p = 0.012), and extension work (p = 0.001). Under the L9 condition, the knee joint maximum power (p = 0.044) also showed significant differences.

According to Table 2, compared to NS, during the push-off phase, significant differences were observed under the L6 condition in the knee joint flexion–extension range (p = 0.006) and peak angular velocities of knee joint extension (p = 0.013). Under the M6 condition, no significant differences were noted in knee joint parameters. Under the M9 condition, significant differences were found in knee joint extension power (p = 0.015) and work (p = 0.018). Under the L6 condition, knee joint extension peak moment (p = 0.001), power (p = 0.001), and work (p = 0.001) showed significant differences. Additionally, under the M9 condition, knee joint flexion–extension power (p = 0.044) exhibited significant differences. All significant differences in the knee joint results were presented in Table 3.

According to the SPM1D results presented in Figure 3, under the L6 condition, significant differences were observed in knee extension angular velocity at 18–21% (p = 0.037), the knee flexion moment at 24–61% (p < 0.001), and knee flexion power at 27–35% and 65–76% (p = 0.002, p < 0.001). Under the M9 condition, significant differences were found in knee extension power at 30–38% (p = 0.001).

3.3. Ankle Joint

According to Table 1, compared to NS, significant differences were found in the peak angular velocity of ankle joint inversion–eversion under the L6 condition during the landing phase of the stance phase (p = 0.001).

Table 2 shows that during the push-off phase, significant differences were observed in the ankle joint flexion–extension range under the M9 condition (p = 0.009). Significant differences were also found in the peak angular velocities of ankle joint dorsiflexion under the M6, M9, L6, and L9 conditions (p = 0.003, p = 0.001, p = 0.012, p = 0.001), and in ankle joint inversion peak angular velocity under the L6 condition (p = 0.001). Under the M9 condition, significant differences were observed in ankle joint inversion–eversion power (p = 0.041) and work (p = 0.025). Under the L6 condition, significant differences were found in ankle joint flexion–extension power (p = 0.037), work (p = 0.045), and inversion–eversion power (p = 0.002) and work (p = 0.002) compared to NS. All significant differences in the ankle joint results were presented in Table 3.

According to the SPM1D results presented in Figure 3 and Figure 4, under the M6 condition, significant changes were observed in ankle dorsiflexion angular velocity at 26–30% (p = 0.014) and ankle plantar flexion power at 26–30% (p = 0.018). Under the L6 condition, significant differences were noted in ankle dorsiflexion angular velocity at 39–43% (p = 0.003), ankle inversion angular velocity at 5–14% and 32–56% (p < 0.001), ankle dorsiflexion moment at 5–23% (p < 0.001), ankle dorsiflexion power at 5–7% (p = 0.041), and ankle inversion power at 35–45% and 68–78% (p < 0.001, p < 0.001). Under the L9 condition, significant differences were observed in ankle eversion angular velocity at 37–51% (p < 0.001) and ankle eversion power at 77–81% (p = 0.012). Significant differences were also noted in the hip adduction moment at 13–18% (p = 0.021). Under the M9 condition, significant differences were found in ankle eversion power at 70–73% (p = 0.032).

Table 3. Summary of all significant differences in the range of motion, peak angular velocity, peak moment, power, and work of the hip, knee, and ankle joints between different shoe conditions and the neutral shoes as presented in Table 1 and Table 2.

Joint	Phase	Condition	Variables
Hip	Landing	M6	Frontal	ROM (°)
			Frontal	Minimum power (w/kg)
		M9	Frontal	ROM (°)
			Frontal	Minimum power (w/kg)
		L6	Sagittal	Peak moment (Nm/kg)
	Push-Off	M6	Sagittal	Minimum power (w/kg)
			Frontal	ROM (°)
				Peak moment (Nm/kg)
				Positive work (J/kg)
		M9	Sagittal	Minimum power (w/kg)
			Frontal	ROM (°)
				Peak moment (Nm/kg)
				Maximum power (w/kg)
				Positive work (J/kg)
		L6	Sagittal	Peak angular velocity (rad/s)
				Peak moment (Nm/kg)
				Maximum power (w/kg)
				Minimum power (w/kg)
				Positive work (J/kg)
		L9	Frontal	Minimum power (w/kg)
Knee	Landing	L6	Sagittal	Peak moment (Nm/kg)
				Maximum power (w/kg)
				Minimum power (w/kg)
				Positive work (J/kg)
				Negative work (J/kg)
		L9	Sagittal	Maximum power (w/kg)
	Push-Off	L6	Sagittal	ROM (°)
				Peak angular velocity (rad/s)
				Peak moment (Nm/kg)
				Maximum power (w/kg)
				Positive work (J/kg)
				Minimum power (w/kg)
				Negative work (J/kg)
Ankle	Landing	L6	Frontal	Peak angular velocity (rad/s)
	Push-off	L6	Sagittal	Peak angular velocity (rad/s)
				Maximum power (w/kg)
				Negative work (J/kg)
			Frontal	Peak angular velocity (rad/s)
				Minimum power (w/kg)
				Positive work (J/kg)
				Negative work (J/kg)
		M9	Sagittal	ROM (°)
				Peak angular velocity (rad/s)
				Negative work (J/kg)
			Frontal	Minimum power (w/kg)
				Negative work (J/kg)

Table 3. Summary of all significant differences in the range of motion, peak angular velocity, peak moment, power, and work of the hip, knee, and ankle joints between different shoe conditions and the neutral shoes as presented in Table 1 and Table 2.

Joint	Phase	Condition	Variables
Hip	Landing	M6	Frontal	ROM (°)
			Frontal	Minimum power (w/kg)
		M9	Frontal	ROM (°)
			Frontal	Minimum power (w/kg)
		L6	Sagittal	Peak moment (Nm/kg)
	Push-Off	M6	Sagittal	Minimum power (w/kg)
			Frontal	ROM (°)
				Peak moment (Nm/kg)
				Positive work (J/kg)
		M9	Sagittal	Minimum power (w/kg)
			Frontal	ROM (°)
				Peak moment (Nm/kg)
				Maximum power (w/kg)
				Positive work (J/kg)
		L6	Sagittal	Peak angular velocity (rad/s)
				Peak moment (Nm/kg)
				Maximum power (w/kg)
				Minimum power (w/kg)
				Positive work (J/kg)
		L9	Frontal	Minimum power (w/kg)
Knee	Landing	L6	Sagittal	Peak moment (Nm/kg)
				Maximum power (w/kg)
				Minimum power (w/kg)
				Positive work (J/kg)
				Negative work (J/kg)
		L9	Sagittal	Maximum power (w/kg)
	Push-Off	L6	Sagittal	ROM (°)
				Peak angular velocity (rad/s)
				Peak moment (Nm/kg)
				Maximum power (w/kg)
				Positive work (J/kg)
				Minimum power (w/kg)
				Negative work (J/kg)
Ankle	Landing	L6	Frontal	Peak angular velocity (rad/s)
	Push-off	L6	Sagittal	Peak angular velocity (rad/s)
				Maximum power (w/kg)
				Negative work (J/kg)
			Frontal	Peak angular velocity (rad/s)
				Minimum power (w/kg)
				Positive work (J/kg)
				Negative work (J/kg)
		M9	Sagittal	ROM (°)
				Peak angular velocity (rad/s)
				Negative work (J/kg)
			Frontal	Minimum power (w/kg)
				Negative work (J/kg)

Figure 3. Descriptive results of the angles, angular velocities, moments, and powers in the sagittal plane obtained from SPM during the stance phase of running while wearing shoes with different forefoot medial–lateral height differences at the hip, knee, and ankle joints. Here, A represents the neutral shoe (NS), B represents M6, C represents M9, D represents L6, and E represents L9.

Figure 3. Descriptive results of the angles, angular velocities, moments, and powers in the sagittal plane obtained from SPM during the stance phase of running while wearing shoes with different forefoot medial–lateral height differences at the hip, knee, and ankle joints. Here, A represents the neutral shoe (NS), B represents M6, C represents M9, D represents L6, and E represents L9.

Figure 4. Descriptive results of the angles, angular velocities, moments, and powers in the frontal plane obtained from SPM during the stance phase of running while wearing shoes with different forefoot medial–lateral height differences at the hip, knee, and ankle joints. Here, A represents the neutral shoe (NS), B represents M6, C represents M9, D represents L6, and E represents L9.

Figure 4. Descriptive results of the angles, angular velocities, moments, and powers in the frontal plane obtained from SPM during the stance phase of running while wearing shoes with different forefoot medial–lateral height differences at the hip, knee, and ankle joints. Here, A represents the neutral shoe (NS), B represents M6, C represents M9, D represents L6, and E represents L9.

4. Discussion

The aim of this study was to investigate the effects of different forefoot medial–lateral height differences on lower limb joint kinematics and kinetics during the stance phase of running. The running shoes used in the experiment had height differences only in the medial–lateral aspect, with all other structural aspects being identical, and the running speed was a self-selected comfortable pace. According to the experimental results, the forefoot height difference significantly affected the biomechanical characteristics of the lower limb joints. Compared to the neutral shoe (NS), different height difference conditions exhibited distinct differences. Of note, the changes in lower limb joint parameters did not linearly increase with increasing height difference values, suggesting a certain biomechanical adaptability of the human body. However, excessive variations might lead to diminished responses or require longer adaptation periods, confirming our initial hypotheses. Particularly, under the L6 condition, significant differences in hip and knee flexion–extension moments during landing and push-off were observed, which align with similar biomechanical changes reported in their study involving altered footwear [27], along with a significant increase in ankle dorsiflexion work and a significant decrease in inversion–eversion work. Under the M9 condition, the hip flexion–extension peak moment, power, and work decreased in the sagittal plane. Additionally, significant differences in ankle dorsiflexion angular velocity were observed under the M6, M9, L6, and L9 conditions, with minimal impact on joint angles.

In this experiment, the most pronounced alterations in lower limb joint parameters occurred when subjects wore shoes with varying forefoot height discrepancies, particularly under the L6 condition. These findings suggest adaptive adjustments in the lower limb kinematic chain, possibly driven by shifts in foot contact patterns [28]. The increased forefoot height in the L6 design facilitated initial lateral foot contact during running, which likely influenced foot posture. Throughout the landing and propulsion phases, there was a notable tendency for the foot to adopt a dorsiflexed position, thereby enhancing dorsiflexion range and intensity. This heightening of lateral foot height also altered the distribution of foot loading, potentially augmenting the engagement of lateral structures while alleviating stress on medial structures such as the calcaneus and metatarsals. These biomechanical adaptations may have consequential implications for ankle joint moments across both sagittal and coronal planes. Studies have shown that forefoot shoes significantly increase ankle dorsiflexion moments and Achilles tendon load but effectively reduce the load on the patellofemoral joint during the support phase of running. However, it is not recommended for novices and beginners to wear forefoot shoes [29]. We speculate that the increase in dorsiflexion work may be associated with a decrease in eversion work. As more of the load is concentrated on the lateral forefoot region, the muscles of the plantar foot may be more involved in dorsiflexion movements, thereby reducing the demand for eversion motion and work. This has important implications for running shoe design and gait adjustment. Selecting appropriate shoe designs and adjusting running posture can reduce the risk of ankle injuries. Research has shown that using lateral wedge insoles can reduce the response time of chronic ankle instability and improve dynamic balance [30]. It is worth noting that in the analysis of results, we found significant differences in peak ankle angular velocity under four different conditions (M6, M9, L6, L9) compared to NS. We believe that the main reason is mechanical adaptation, and the significant differences in peak ankle angular velocity under different conditions may reflect immediate adaptations of the body to changes in shoe design. Although angular velocity changes occur, the body may maintain the stability of movement angles by adjusting other kinetic parameters (such as muscle strength utilization or coordination of other joints), thus optimizing movement efficiency. This design can enhance runners’ adaptability under specific conditions, such as uneven surfaces or inclined tracks, where running in such unstable conditions may stimulate muscle groups related to the lower limbs and enhance the coordination ability of the ankle joint and surrounding muscles. Studies have shown that the unstable design of biomimetic shoes helps strengthen muscle control, enhance postural stability, and proprioceptive ability [31].

Under such conditions of unbalanced load distribution, the muscles, ligaments, and joints of the lower limb need to adapt to the altered mechanical environment in different ways, which may lead to changes in the overall distribution pattern of moment and power. Through the analysis of knee joint moment during the landing and push-off phases, we found that, compared to NS, there were significant differences in knee flexion moment under the L6 condition, with a significant reduction in the peak moment during the push-off phase, potentially reducing the risk of knee cartilage damage and contributing positively to the prevention of knee pain and degenerative diseases such as osteoarthritis. As demonstrated in a study, 6 mm lateral wedge insoles significantly reduced pain in patients with grade II and III knee osteoarthritis during walking, standing, resting, and knee joint movement [32], and it has been shown that patients with knee osteoarthritis have unique plantar pressure patterns during walking, with lateral wedge insoles being one of the treatment options [33]. We believe that the design of L6 may shift the lower limb load from the knee joint to other joints, such as the hip or ankle joint. Studies have shown that when running in minimalist shoes, joint work is redistributed from the knee to the ankle [34], which is consistent with the increased ankle dorsiflexion moment described earlier. Such shoe design may assist runners in moving more efficiently. As demonstrated in the results shown in Table 2, knee joint extension work significantly decreased under the L6 condition compared to NS. This transfer of energy contributes to overall running efficiency improvement and reduces the risk of knee joint injury.

In the L6 condition, we observed significant differences in the hip joint’s flexion–extension moment during the stance phase, with notable increases in peak moments during both the landing and push-off phases. Our analysis of experimental results revealed a significant increase in hip joint extension work. Besides the previously mentioned load distribution, we speculate that the design featuring a higher lateral forefoot alters the force line during foot landing and propulsion, potentially requiring more hip joint force to maintain balance and stability, especially necessitating increased extension work during propulsion. The increase in work implies a greater transfer of power to the hip joint musculature to accommodate the biomechanical changes induced by the new shoe condition. Additionally, under the M9 condition, we observed reductions in the hip joint’s abduction–adduction peak moment, power, and work in the sagittal plane. Such design alterations may enhance foot stability, reducing the additional force required to maintain balance during hip abduction. Studies suggest that reducing the abduction moment at the hip joint during standing is crucial for preventing the progression of hip osteoarthritis [35]. Lowering the load during abduction movements can decrease the risk of damage to surrounding soft tissues, such as hip muscles and ligaments, thereby reducing the incidence of muscle strains and ligament injuries. These findings have important implications for the health and long-term running habits of runners.

While this study provides valuable insights into the impact of forefoot height difference on lower limb joint biomechanics, there are several limitations to consider. Firstly, the participants recruited for this experiment were all healthy males; thus, the findings may not be generalizable to females and injured runners. Secondly, the study did not account for the potential impact of different self-selected running speeds and the lack of control for participants’ habitual footwear on the results. Thirdly, due to the limited joint degrees of freedom provided by the muscle–skeletal model (gait 2392) from OpenSim used in the data preprocessing stage, our analysis in this study was confined to the sagittal and coronal planes of the hip and ankle joints, as well as the sagittal plane of the knee joint. Additionally, this study only focused on the stance phase of running, as it has been reported that the stance phase of running is closely associated with running-related injuries [36,37]. Lastly, in future research designs, consideration could be given to exploring different population groups, varying speed gradients, and using more advanced computational methods to enhance the accuracy of biomechanical analysis. For example, employing finite element-based computational methods could reduce costs and provide more accurate model analysis results [38,39].

5. Conclusions

Compared to NS, wearing running shoes with different height differences appears to affect the joint moments and work during the stance phase of running, while having little to no discernible impact on joint angles. Particularly, under the L6 condition, it may contribute to a reduction in the risk of knee joint injuries and exhibit a positive effect on gait improvement in individuals with pronated feet. However, these findings are preliminary and should be interpreted with caution. This study provides initial insights into the potential biomechanical effects of running shoes with different height differences on lower limb joints during the stance phase. The limitations of our study, including the homogeneous participant group, self-selected running speeds, and habitual footwear effects, highlight the necessity for further in-depth research. Further in-depth research is necessary to elucidate the underlying mechanisms of different shoe types on running biomechanics and to evaluate their practical potential in preventing sports injuries and enhancing athletic performance. Future studies should also address the limitations of the current research to validate and expand upon these findings.