How Does Foot Arch Type Affect Gait Biomechanics in Patients with Plantar Fasciitis?

Abstract

Plantar fasciitis (PFS) is a leading cause of heel pain, yet its clinical course varies widely. Although plantar fascia thickness (PFT) is often used as a pain marker, its prognostic value remains unclear. Objective: This study investigates whether foot arch morphology underlies distinct biomechanical profiles in PFS patients, potentially explaining the variability in its presentation. Methods: The cross-sectional study included 30 patients with PFS and 10 healthy controls. PFS patients were classified by arch type (pes rectus, pes planus, pes cavus) using the Arch Height Index (AHI). Baseline comparisons between healthy controls and PFS subgroups assessed PFT, Foot Function Index (FFI), joint stiffness ratio, and gait parameters. Results: PFT differed across groups but was not significantly associated with FFI scores (p = 0.233). The pes cavus group exhibited a lower metatarsophalangeal (MTP) stiffness ratio compared with healthy (p < 0.05). Pes planus and pes rectus groups showed excessive pronation, and the pes cavus group showed limited ankle dorsiflexion, indicating distinct gait mechanisms (p < 0.05). Conclusions: Foot arch morphology influences gait biomechanics, stiffness, and PFT in individuals with PFS. Incorporating individual arch types into clinical decision-making may facilitate more personalized interventions and improve treatment outcomes.

1. Introduction

Plantar fasciitis (PFS) is a common and burdensome cause of heel pain, accounting for roughly 1 million clinic visits annually. It represents 11–15% of all foot and ankle consultations and predominantly affects physically active, working-age adults [1,2,3]. The condition is often self-limiting, and more than 80% of patients achieve symptom resolution within one year [1,4]. However, a substantial minority continues to experience persistent pain beyond that period. In these chronic cases, the impact on quality of life can be substantial. Heel pain can persist in a substantial proportion of patients; in fact, about half continue to report symptoms even 5 years after the initial diagnosis [5]. Conservative care focuses on stretching, orthoses, and load management [6]. Despite these efforts, clinical outcomes vary widely, highlighting the importance of identifying individual factors that influence recovery [4,7]. That is, the occasional ineffectiveness of standard stretching emphasizes the need for a deeper understanding of PFS prognosis.

One pathological hallmark of PFS is a thickened plantar fascia (PF) observed on imaging. Ultrasound and MRI studies frequently show PF hypertrophy at the calcaneal origin in symptomatic individuals, sometimes exceeding 4 mm in thickness [8,9]. Increased thickness is thought to reflect fascia degeneration and chronic overload, and some have hypothesized that the degree of thickening could predict symptom severity or recovery [8,10,11]. Nevertheless, studies have shown that the utility of plantar fascia thickness (PFT) as a prognostic marker remains unclear. Recent findings indicate that while fascial thickness often decreases with treatment, neither the baseline thickness nor the magnitude of thinning correlates well with functional improvement [12,13]. In other words, a thick fascia can persist even after pain resolves, and thinner fascia does not guarantee better outcomes.

In light of these limitations, recent studies have shifted the focus to the biomechanical metrics that depict the functional condition of the foot. Shear-wave ultrasound elastography, for example, quantifies PF stiffness, a dynamic property markedly higher in PFS patients compared to healthy controls. Along with detailed analyses of foot kinematics, such stiffness or elasticity measures would provide a richer picture of tissue health than static thickness alone and would be promising for tracking clinical progress or prognosis [14,15]. Despite such advances, PFS is still managed as a homogeneous disorder [7]. Overuse-related micro-tearing at the calcaneal insertion is often cited as the primary mechanism of PFS. As a result, clinicians typically classify cases according to symptom chronicity or the presence of a heel spur, rather than underlying foot morphology [1]. This generalized approach may overlook important morphological differences between individuals that could influence disease presentation and progression.

In particular, foot morphology, particularly arch height, warrants more detailed consideration in this respect. Abnormal biomechanics of the foot can predispose individuals to PFS. For example, the pes planus group is characterized by a low medial arch and a tendency toward excessive pronation. In contrast, the pes cavus group maintains a high medial arch and tends to remain supinated, reducing shock absorption [16]. Although prior biomechanical research in healthy individuals consistently shows that the pes planus and the pes cavus groups exhibit markedly different patterns of plantar pressure and gait kinematics, it remains unclear whether similar distinctions hold true in PFS populations [16,17,18]. In particular, patients with PFS often present a combination of overpronation and limited ankle dorsiflexion [19,20,21], which may appear contradictory for a foot structure. However, no study has directly linked arch type to plantar-fasciitis presentation. This omission is notable. The inconsistent relationship between PFT and symptom severity, highlighted in previous studies, might be partly explained by this failure to account for foot morphology. It is plausible that the underlying biomechanical factors driving pain and fascial thickening are not uniform across all patients, as PFS is often managed as a homogeneous disorder. Instead, these factors may differ depending on the individual’s arch structure, which could in turn influence the clinical presentation and the specific relationship between PFT and symptoms, yet this remains unexplored.

The present study aimed to determine whether arch type is associated with differences in PFT, joint stiffness, and gait kinematics in individuals with PFS. We hypothesized that these biomechanical signatures would explain foot-type-specific mechanisms of plantar-fascial overload. By investigating the interplay between arch type and PF changes, this study seeks to fill a critical gap in understanding the biomechanical and anatomical factors that contribute to PFT. These findings carry important clinical implications and may constitute truly personalized rehabilitation strategies in patients with PFS. Customizing orthotic prescriptions and rehabilitation protocols to each patient’s specific arch morphology should enhance treatment outcomes. Ultimately, the present study aims to improve clinical decision-making and patient management by integrating structural foot considerations into the evaluation and treatment of PFT.

2. Materials and Methods

2.1. Participants

Forty individuals (10 healthy, 30 PFS) participated. All participants were fully informed about the study’s procedures, potential risks, and objectives. The recruitment period was between August 2024 and October 2024. We prospectively recruited a target cohort of 30 patients with clinical PFS. To create balanced subgroups for analysis, these 30 participants were classified using a post hoc relative grouping strategy based on their Arch Height Index (AHI) [22]. All 30 participants were ranked by their AHI score and divided into tertiles (3-quantiles). The bottom tertile (n = 10) was assigned to the pes planus group, the middle tertile (n = 10) to the pes rectus group, and the top tertile (n = 10) to the pes cavus group. This deliberate, balanced design was chosen to facilitate robust statistical comparisons among these distinct morphological subgroups. A cohort of 10 healthy controls was recruited to provide a normative baseline. Patients were eligible for inclusion if they had a clinical diagnosis of PFS from a hospital and had never experienced a complete rupture of the PF. Exclusion criteria included degenerative arthritis of the hindfoot or other joint diseases, any previous injury or surgery involving the foot or ankle, and inability to understand the study protocol due to language or cognitive limitations. Only the symptomatic limb was evaluated. If symptoms were bilateral, assessments were performed on the side that the participant rated as more painful.

2.2. Experimental Procedures

Participants were positioned supine with the knee fully extended and the ankle and metatarsophalangeal (MTP) joints positioned naturally. The rotational axis of the Plantar Fascia and Achilles Tendon Simultaneous Stretching (PASS) robot [23] aligned with the participant’s MTP and ankle joint axes, and the foot was secured to the bottom plate without restricting joint motion. PASS robot was used as a motorized test rig for measuring ankle and MTP joint range of motion (ROM) and stiffness. As detailed in [23], it is equipped with magnetic encoders and a torque sensor aligned with the MTP joint axis to measure joint angle and passive resistive torque, respectively. Prior to measurement, after mechanical axes were aligned, the angle sensors were initialized to 0° based on the foot’s neutral position, and the torque sensor was zeroed (tared) to offset the gravitational torque of the foot/ankle and the robot’s foot/ankle plate. Participants were then asked to remain relaxed while the robot passively dorsiflexed the MTP and ankle joint from 0° to the maximum tolerated angle (at 5°/s and 2.5°/s, respectively). This ‘maximum tolerated angle’ was operationally defined as the precise angle at which the participant first verbally reported any sensation of pain or significant stretch-related discomfort, and they were instructed to say ‘stop’ immediately upon reaching this threshold to ensure measurement consistency. Three valid trials were recorded for each foot; the torque-angle data from these trials were averaged for further analysis.

The Foot Function Index (FFI) is a widely used, self-administered questionnaire designed to evaluate foot-related pain and disability. It consists of 23 questions divided into three subscales (pain, disability, and activity limitation) [24]. Each item is scored 0–10; the total FFI is the sum of all items (0–230). Higher FFI scores indicate greater pain or disability. The navicular-drop test (NDT) is measured to quantify static arch flexibility [25,26]. With the participant seated, the examiner marks the navicular tuberosity and records its vertical distance from the floor with a ruler. The participant then stands naturally, and the measurement is repeated. The NDT value (cm) equals the seated height minus the standing height; a larger value indicates greater midfoot collapse. Figure 1 shows the flow chart illustrating progression through the study.

Figure 1. Flow chart illustrating progression through the study. Abbreviations: AHI-Arch Height Index; MTP-metatarsophalangeal; PFS—Plantar fasciitis; PF—Plantar fascia; PFT—Plantar Fascia Thickness; PROM—Passive Range of Motion; IC—Initial Contact; TO—toe off; DoF—Degree of Freedom.

PFT was measured bilaterally using a handheld ultrasound device (SONON 300L, Healcerion, Seoul, Republic of Korea). Participants lay prone with the knee flexed to 90° and the ankle and MTP joints maintained in a neutral position [27,28]. The ultrasound probe was placed over the medial tubercle of the calcaneus, where stress concentration commonly leads to PFS symptoms. All PFT measurements were performed by a single experienced examiner to minimize inter-observer variability.

Participants performed a walking test by walking across seven 30 cm-wide stepping plates arranged at 70 cm intervals on the floor. They were instructed to maintain a consistent stride length and comfortable speed. Twenty-three reflective markers were placed on lower extremity landmarks based on a Plug-in Gait marker system, enabling three-dimensional motion capture and kinematic analysis [29]. In this model, the foot is treated as a single rigid segment; thus, ankle kinematics are defined as the rotation of the foot segment relative to the tibia (reporting ankle dorsiflexion/plantarflexion, inversion/eversion, and abduction/adduction over the gait cycle). 1st MTP kinematics were derived from the hallux relative to the foot.

2.3. Data Acquisition

The ankle and MTP joint dorsiflexion stiffness curve was generated from the measured torque and angle data via the PASS robot’s torque and angle sensors, which typically produced a cubic polynomial-like shape. Passive dorsiflexion range of motion (ROM) was defined as the maximum angle reached without pain during the PASS robot measurement; this point was defined as the ‘end’ point for analysis. The ‘start’ point was defined as 0°, and the ‘inflection point’ was identified as the local minimum of the first derivative of this curve. These three distinct points (start, inflection, and end) were used to calculate stiffness parameters. Total joint stiffness (${\mathit{K}}_{\mathit{t}\mathit{o}\mathit{t}}$) was calculated as the slope of the line connecting the start point and the end point, representing the overall change in torque relative to the change in angle. The stiffness of the initial interval (${\mathit{K}}_1$) was calculated from the start point to the inflection point, and the stiffness of the terminal interval (${\mathit{K}}_2$) was calculated from the inflection point to the end point. To reduce the influence of individual physical characteristics, the stiffness ratio (${\mathit{K}}_{\mathit{r}\mathit{a}\mathit{t}\mathit{i}\mathit{o}}$) was then calculated as the ratio of these two intervals (${\mathit{K}}_1$/${\mathit{K}}_2$) and used for further analysis. The mean value from three successful trials was used for statistical analyses.

A 12-camera VICON NEXUS system (Oxford Metrics Ltd., Oxford, UK), sampling at 100 Hz, collected all kinematic data during the walking test. Ground reaction forces were recorded using three force plates (Advanced Mechanical Technology, Inc., Watertown, MA, USA), sampling at 2000 Hz, embedded in the walkway. Gait events were defined at a vertical GRF threshold of 20 N. Gait cycles were defined from initial contact (vertical ground reaction force > 20 N) to the subsequent initial contact [30]. Toe off occurred when the vertical ground reaction force fell below 20 N. Kinematic data were processed using Visual 3D (C-Motion Inc., Germantown, MD, USA). Stance phase duration was calculated as the percentage of the gait cycle (0–100%) during which the vertical ground reaction force exceeded 20 N.

2.4. Data Analysis

Given the sample size of the subgroups, non-parametric statistical tests were employed. Demographic data were compared between healthy controls and the overall PFS cohort using the Mann–Whitney U test, and among PFS arch-type subgroups using the Kruskal–Wallis H test [31]. For baseline comparisons, PFT, stance phase duration, ankle and MTP joint stiffness, and their stiffness ratios were compared among the four groups (healthy controls and the three PFS subgroups) using the Kruskal–Wallis test. When a significant overall difference was detected, pairwise post hoc comparisons were conducted using the Mann–Whitney U test with a Bonferroni correction applied to adjust the significance level (SPSS Statistics 26, IBM Corp., New York, NY, USA). Statistical Parametric Mapping (SPM) analysis [32] was applied to ankle and MTP joint angle time series to identify time--point--specific differences among the PFS subgroups. Only gait-cycle intervals maintaining significance (p < 0.05) over at least 10% of the cycle were reported, with significant intervals further examined via post hoc analysis.

3. Results

The demographic characteristics are summarized in Table 1. Ten healthy controls and 30 patients with PFS showed no differences in gender (60% and 63%), age (37.2 ± 7.1 and 35.2 ± 11.0 y; p = 0.398, U = 122.5, Mdiff = 2.00), or BMI (23.1 ± 4.0 and 25.8 ± 4.4 kg/m²; p = 0.184, U = 193.0, Mdiff = 2.66 kg/m²), but FFI was higher in PFS (0.0 ± 0.0 and 76.7 ± 33.5; p < 0.001, U = 300.0, Mdiff = 76.73). Within the PFS group, pes rectus, pes planus, and pes cavus groups did not differ in gender (60%, 70%, and 60%), age (32.3 ± 11.9, 36.0 ± 13.1, and 37.3 ± 7.9 y; p = 0.487, η² = 0.050, ε² = −0.021), BMI (26.1 ± 6.2, 26.2 ± 3.3, and 25.1 ± 3.4 kg/m²; p = 0.748, η² = 0.020, ε² = −0.053), or FFI (60.8 ± 25.8, 80.8 ± 40.6, and 88.6 ± 29.1; p = 0.233, η² = 0.101, ε² = 0.034), whereas AHI (0.35 ± 0.02, 0.25 ± 0.04, and 0.40 ± 0.02; p < 0.001, η² = 0.886, ε² = 0.877) differed. The AHI values decreased in the order of pes cavus, pes rectus, and pes planus, respectively. This is an expected result, as the participant subgroups were classified based on the AHI. Significant differences were also found in the NDT among the PFS subgroups (0.69 ± 0.24, 0.80 ± 0.16, and 0.33 ±0.13 cm; p < 0.001, η² = 0.642, ε² = 0.615). Post hoc results indicated that the pes cavus group had significantly smaller values compared to the pes rectus (p = 0.002, U = 95.5, Mdiff = 0.36 cm) and pes planus groups (p = 0.001, U = 98.5, Mdiff = 0.47 cm).

Table 1. Demographic data and characteristics of participants presented as mean (SD). Left p value, Mann–Whitney U test between healthy vs. PFS; Right p value, Kruskal–Wallis test among PFS subgroups. ***: p < 0.001. Abbreviations: PFS—Plantar fasciitis; BMI—Body Mass Index; FFI—Foot Functional Index; AHI—Arch Height Index; NDT—Navicular Drop Test.

	Healthy	PFS	p-Value	Pes Rectus	Pes Planus	Pes Cavus	p-Value
Participants, n	10	30		10	10	10
Gender: Male/Female, n (%)	6/4 (60/40%)	19/11 (63/37%)		6/4 (60/40%)	7/3 (70/30%)	6/4 (60/40%)
Age, y, mean (SD)	37.2 (7.1)	35.2 (11.0)	0.398	32.3 (11.9)	36.0 (13.1)	37.3 (7.9)	0.487
BMI, kg/m2, mean (SD)	23.1 (4.0)	25.8 (4.4)	0.184	26.1 (6.2)	26.2 (3.3)	25.1 (3.4)	0.748
FFI (0–230), mean (SD)	0 (0)	76.7 (33.5)	<0.001 ***	60.8 (25.8)	80.8 (40.6)	88.6 (29.1)	0.233
AHI, mean (SD)	0.36 (0.04)	0.33 (0.07)	0.416	0.35 (0.02)	0.25 (0.04)	0.40 (0.02)	<0.001 ***
NDT, cm, mean (SD)	0.68 (0.40)	0.61 (0.27)	0.875	0.69 (0.24)	0.80 (0.16)	0.33 (0.13)	<0.001 ***

Figure 2 summarizes stance phase duration, PFT and stiffness. Stance phase duration (p = 0.009, η² = 0.294, ε² = 0.236) was longer in healthy controls than in the pes rectus (p = 0.044, U = 58.0, Mdiff = 1.835%) and pes planus groups (p = 0.022, U = 11.0, Mdiff = 2.113%). PFT (p < 0.001, η² = 0.570, ε² = 0.534) was significantly greater in the pes rectus (p = 0.005, U = 95.0, Mdiff = 1.19 mm), pes cavus groups (p = 0.001, U = 99.0, Mdiff = 1.49 mm) than in healthy controls. PFT in the pes cavus group also exceeded that in the pes planus group (p = 0.034, U = 13.0, Mdiff = 0.77 mm). The ankle stiffness ratio ($K_{ratio-ankle}$; p = 0.0293, η² = 0.231, ε² = 0.167) showed a significant overall difference among groups; however, post hoc analysis did not reveal significant differences between any individual groups. The MTP stiffness ratio ($K_{ratio\_MTP}$; p = 0.0102, η² = 0.290, ε² = 0.230) was lower in the pes cavus group than in healthy controls (p = 0.013, U = 9.0, Mdiff = 0.441). No other stiffness variables differed among the groups in total ankle stiffness ($K_{tot\_ankle}$; p = 0.163, η² = 0.132, ε² = 0.059) or total MTP stiffness ($K_{tot\_MTP}$; p = 0.424, η² = 0.072, ε² = −0.006).

Figure 2. Comparison of stance phase duration, PFT, stiffness and stiffness ratio of ankle/MTP joint. Error bars represent standard error. *: p < 0.05; **: p < 0.01; ***: p < 0.001. Abbreviations: PFT-Plantar Fascia Thickness; ${\mathit{K}}_{\mathit{r}\mathit{a}\mathit{t}\mathit{i}\mathit{o}\_\mathit{a}\mathit{n}\mathit{k}\mathit{l}\mathit{e}}$-stiffness ratio of ankle joint; MTP-metatarsophalangeal; ${\mathit{K}}_{\mathit{r}\mathit{a}\mathit{t}\mathit{i}\mathit{o}\_\mathit{M}\mathit{T}\mathit{P}}$-stiffness ratio of MTP joint; ${\mathit{K}}_{\mathit{t}\mathit{o}\mathit{t}\_\mathit{a}\mathit{n}\mathit{k}\mathit{l}\mathit{e}}$-stiffness of ankle joint; ${\mathit{K}}_{\mathit{t}\mathit{o}\mathit{t}\_\mathit{M}\mathit{T}\mathit{P}}$-stiffness of MTP joint.

The pes cavus group exhibited limited ankle dorsiflexion during early stance phase (1–33% of the gait cycle) and reduced peak dorsiflexion in swing phase (59–92%) compared with the pes planus group (Figure 3a). Statistical analysis confirmed significant differences among groups during the early stance phase (p = 0.003, η² = 0.298, ε² = 0.250). Post hoc testing revealed that the pes cavus group showed significantly less dorsiflexion compared to the pes rectus (p = 0.031, U = 121.0, Mdiff = 6.27°), pes planus (p = 0.044, U = 119.0, Mdiff = 5.52°), and healthy groups (p = 0.021, U = 21.0, Mdiff = 7.20°) (Figure 3c). Significant differences were also found among groups during the swing phase (p = 0.021, η² = 0.207, ε² = 0.153), with the pes cavus group showing less dorsiflexion than the pes planus group (p = 0.048, U = 118.0, Mdiff = 8.78°) (Figure 3d).

Figure 3. (a,b): Ankle/MTP joint sagittal plane rotation during the walking test. The gait cycle (0–100%) is divided into the stance phase (0–60%) and the swing phase (60–100%), separated by a vertical dotted line indicating toe off. The blue, orange, magenta and black lines represent pes rectus, pes planus, pes cavus, and healthy, respectively. Shaded bands denote the 95% confidence interval. The gray-shaded area corresponds to sections where the Kruskal–Wallis test results indicate a statistically significant difference (p < 0.05). Below each graph, bars compare group differences at each time point: pes rectus vs. pes planus, pes rectus vs. pes cavus and pes planus vs. pes cavus. The color of bars corresponds to the group with larger values. (c–e): comparison of mean angle values within the gray-shaded area. Error bars represent standard error. *: p < 0.05; **: p < 0.01. Abbreviations: IC—Initial contact; MTP—metatarsophalangeal.

At the MTP joint, the pes cavus group showed excessive dorsiflexion during stance phase (10–44%) relative to the pes rectus and pes planus groups (Figure 3b). However, the comparison of mean angles within this statistically significant interval did not reveal significant differences among the groups (p = 0.131, η² = 0.120, ε² = 0.060) (Figure 3e).

In the ankle’s frontal and transverse planes, the pes rectus and pes planus groups demonstrated greater eversion and external rotation than the other groups (Figure 4). In the frontal plane, these groups showed excessive ankle eversion between 43–60% and 66–79% of the gait cycle versus the pes cavus group and healthy controls. Significant differences among groups were found during the stance phase interval (43–60%) (p = 0.004, η² = 0.288, ε² = 0.239). Post hoc analysis indicated that the pes planus group exhibited significantly greater eversion compared to the pes cavus group (p = 0.037, U = 120.0, Mdiff = 6.64°) and the healthy group (p = 0.044, U = 119.0, Mdiff = 4.40°) (Figure 4c). No significant group differences were found during the swing phase interval (66–79%) (Figure 4d). In the transverse plane, the pes rectus and pes planus groups exhibited increased ankle external rotation versus the pes cavus group from terminal stance through pre--swing (53–71%). Significant differences were found among groups during this interval (p = 0.011, η² = 0.237, ε² = 0.185), with post hoc analysis showing significantly greater external rotation in the pes planus group compared to the pes cavus group (p = 0.018, U = 127.0, Mdiff = 4.29°) (Figure 4e).

Figure 4. (a,b): Ankle joint frontal/transverse plane rotation during the walking test. The gait cycle (0–100%) is divided into the stance phase (0–60%) and the swing phase (60–100%), separated by a vertical dotted line indicating toe off. The blue, orange, magenta and black lines represent pes rectus, pes planus, pes cavus, and healthy, respectively. Shaded bands denote the 95% confidence interval. The gray-shaded area corresponds to sections where the Kruskal–Wallis test results indicate a statistically significant difference (p < 0.05). Below each graph, bars compare group differences at each time point: pes rectus vs. pes planus, pes rectus vs. pes cavus and pes planus vs. pes cavus. The color of bars corresponds to the group with larger values. (c–e): comparison of mean angle values within the gray-shaded area. Error bars represent standard error. *: p < 0.05; **: p < 0.01. Abbreviations: IC—Initial contact; MTP—metatarsophalangeal.

4. Discussion

In this study, we found that PFT, traditionally considered the primary marker of PFS, did not necessarily correlate with clinical symptom severity. We found differences in fascial thickness between the pes planus and pes cavus groups. However, there was no link between thickness and pain levels (FFI). Thus, arch type can be considered when assessing PFS burden. Furthermore, our results align with previous studies suggesting that PFT alone should not be the sole determinant for treatment planning [12,13].

One such factor is the MTP joint stiffness ratio, which was notably lower in the pes cavus group. A lower stiffness ratio indicates that the tissue stretches more easily at smaller angles but becomes abruptly rigid at higher angles [23]. This pattern likely stems from the high-arched foot’s rigid structure, which is compliant at small angles but resists elongation as the angle increases. During toe off, the MTP joint reaches peak dorsiflexion, and this sudden rise in stiffness may exacerbate PF pain, prolonging or intensifying symptoms.

Motion analysis revealed that individuals in the pes cavus group often demonstrated limited ankle dorsiflexion but excessive dorsiflexion at the MTP joint. Limited ankle dorsiflexion from initial contact to mid-stance is consistent with findings from previous studies on asymptomatic cavus feet [16,33]. Plantar pressure studies on healthy cavus feet confirm this, showing a significantly reduced contact area, which results in higher peak pressures under the heel and lateral forefoot compared to neutral feet [16]. This restriction reduces shock absorption and can exacerbate pain [34,35]. Particularly in patients with PFS, in whom inflammation has already occurred, this limitation becomes a factor that perpetuates pain and symptoms. Moreover, in our study, the dorsiflexion deficit persisted from toe off through swing phase, pointing to a pain-mediated alteration of mechanics. The prolonged limitation may arise from a premature heel-off strategy that shortens forward tibial progression. It may also reflect the intrinsically high stiffness of the Achilles tendon and PF in pes cavus, which physically restricts additional dorsiflexion during terminal stance. Lastly, the stiffness ratio indicates that, in mid-stance (approximately 0–15° range in MTP joint) [36], the MTP joint in the pes cavus group moved with minimal resistance, likely due to the tissue’s slack at lower angles.

Conversely, the pes planus and pes rectus groups showed excessive ankle eversion and external rotation with toe off, commonly referred to as overpronation. This pattern mirrors that reported in asymptomatic planus feet, where the medial arch collapses under load and the subtalar joint remains everted through terminal stance [37]. In this study, however, even pes rectus behaved like pes planus. Pain-avoidance and altered loading may reduce posterior tibialis activation, compromising medial-arch stability in otherwise rectus feet; this interpretation aligns with posterior tibial tendon dysfunction mechanisms, although causal evidence in PFS is limited [38,39,40]. When this muscle cannot work well, even pes rectus feet lose arch stability, so pes rectus feet overpronate like pes planus. This overpronation allows the midfoot move too much, stretches the PF, and can cause small repeated tears at its heel attachment, keeping the heel painful. This mechanism is consistent with plantar pressure studies in healthy individuals, which demonstrate that pes planus feet experience increased loading in the medial and midfoot regions as the arch collapses [16].

Overall, our findings point to two distinct overload paradigms. Pes planus and pes rectus fit an “excessive motion” model: overpronation and midfoot collapse lengthen the PF and create tensile micro-trauma. Pes cavus fits a “restricted motion” model: limited ankle dorsiflexion and windlass tightening focus impact forces on a stiff PF. Biomechanically, limited ankle dorsiflexion [19,21] and excessive pronation [20,41] are both recognized risk factors for PFS, and our findings underscore how these factors can intersect to overload the PF in different ways. Ultimately, these differences highlight the importance of individualized treatment approaches guided by arch morphology.

Our findings indicate that arch height is a clinically meaningful factor for stratifying patients with PFS, rather than a merely descriptive feature, and therefore supports arch-specific rehabilitation. For pes planus/rectus, priorities include frontal-plane control (intrinsic foot training [42]) and medial-arch support (taping or medially posted orthoses) to reduce prolonged pronation. For pes cavus, emphasis should be on gastro-soleus and plantar fascia stretching, heel lifts or rocker-bottom footwear to reduce first-MTP demand, and cues that facilitate tibial progression and dorsiflexion [43]. These arch-specific needs align well with emerging rehabilitation technologies. For instance, the ‘restricted motion’ pes cavus group could benefit from consistent, high-intensity stretching delivered by robotic devices [44], while the ‘excessive motion’ pes planus/rectus groups could be aided by soft wearable systems designed to dynamically control overpronation [45]. Prospective, arch-stratified trials are warranted to determine whether these targeted, technology-assisted protocols can improve outcomes beyond standard care.

Several limitations should be considered when interpreting these findings. First, the passive measurement constraint inherent in our stiffness protocol relied on a self-reported ‘maximum tolerated angle’ based on pain or discomfort. This means the endpoint might reflect individual pain tolerance rather than the true physiological end-range of motion, potentially underestimating the terminal stiffness and influencing the calculated total stiffness and stiffness ratio. Although a familiarization trial aimed to mitigate this, subjective variability likely remained. Second, inherent inter-trial variability persists despite averaging multiple repetitions; subtle variations in participant gait patterns, marker placement, or effort levels across trials can contribute to this variability. Third, we used Plug-in Gait, which treats the foot as a single rigid segment; therefore, rearfoot–forefoot–hallux relative motions could not be fully resolved. Future work with multi-segment foot model (e.g., the Oxford Foot Model) is warranted. Fourth, while our subgroup design was balanced (n = 10 per group) and baseline comparisons showed no significant differences in potential confounders like BMI or gender distribution among the PFS subgroups, this modest sample size limits statistical power, particularly for detecting subtle differences. Future validation with larger, more diverse samples, incorporating multi-segment foot models and plantar pressure analyses, is recommended; such pressure data might specifically help elucidate why pain severity did not correlate with PFT in our study, potentially revealing stronger links between pain and dynamic loading patterns.

5. Conclusions

This study confirms that PFS biomechanics vary systematically with foot arch type. Distinct joint stiffness and gait deviations highlight the importance of integrating arch morphology into assessment and rehabilitation planning. Tailoring orthoses, exercise, and load management strategies to the individual arch is likely to improve clinical outcomes. Future longitudinal work should test whether arch-specific interventions can prevent recurrence and accelerate recovery in diverse patient populations.