Determination of relationship between foot arch,hindfoot, and hallux motion using Oxford foot model: Comparison between walking and running

BackgroundThe foot arch plays an important role in propulsion and shock absorption during walking and running; however, the relationship among the foot arch, metatarsal locking theory, and nature of the windlass mechanism (WM) remain unclear. Research question: What are the differences in the kinematic relationship between the foot arch, hindfoot, and hallux during walking and running?MethodsRelative angles within the foot were measured in 18 healthy men using the Oxford foot model (OFM). Data for barefoot walking at a comfortable speed and rearfoot running at 2.0 m/s were collected. Angles of the forefoot relative to the hindfoot (OFM-arch), hallux relative to the forefoot (Hallux) on the sagittal plane, and hindfoot relative to the shank (Hindfoot) on three anatomical planes were obtained. The medial longitudinal arch (MLA) angle was calculated to verify that OFM-arch can substitute the MLA angle. Each parameter was subjected to cross-correlation analysis and Wilcoxon signed-rank tests to examine the relationship with OFM-arch and compare them during walking and running.ResultOFM-arch was similar to the conventional MLA projection angle in both trials (gait: 0.79, running: 0.96 p < 0.01). Synchronization of the OFM-arch and Hallux angles was higher in running than in walking (gait: −0.09, running: −0.75 p < 0.01). Hindfoot supination was unrelated to OFM-arch. Hindfoot angle on the transverse plane exhibited a moderate relationship with OFM-arch, indicating different correlations in walking and running (gait: 0.63, running: −0.68 p < 0.01).Significance: The elevation of the foot arch due to hallux dorsiflexion differed during walking and running; hence, other factors besides WM (such as intrinsic muscles) may affect the foot arch elevation during running. The hindfoot in the frontal plane does not contribute to arch raising and foot stability during running; it features different relationships with OFM-arch during walking and running.     

健康女大学生步行时左右足支撑期时相特征比较

目的:研究健康女大学生步行时支撑期时相参数特征的左右差别,为临床诊断、康复评定、矫形处方和运动训练等提供参考依据。方法:应用footscan?足底压力测量系统测试32例健康女大学生步行时的动态足底压力,计算支撑期足跟触地阶段、前足触地阶段、全足支撑阶段、前足蹬离阶段占足-地接触时间百分比,以及足跟触地和前足触地两个阶段内足跟内侧和外侧足底压力,分析上述参数5次有效测试数据左右侧差异及重测信度。结果:步行时左右侧足-地接触时间分别为618.3 ms和617.8 ms。左侧足跟触地阶段、前足触地阶段、全足支撑阶段、前足蹬离阶段占足-地接触时间的百分比分别为8.66%、9.30%、39.13%、42.91%,右侧为8.54%、9.25%、38.89%、43.30%。左侧足跟触地和前足触地两个阶段内足跟内侧和外侧足底压力分别为1157.57 N/cm2和1055.35 N/cm2,右侧分别为1240.25 N/cm2和1050.45 N/cm2。上述各参数组内相关系数均大于0.76,左右侧差异均无统计学意义。结论:健康女大学生步行时支撑期4个阶段的时间分布参数以及足跟触地和前足触地两个阶段内足跟内侧和外侧足底压力左右侧无明显差异。     

In-shoe plantar pressure distribution during running on natural grass and asphalt in recreational runners

The type of surface used for running can influence the load that the locomotor apparatus will absorb and the load distribution could be related to the incidence of chronic injuries. As there is no consensus on how the locomotor apparatus adapts to loads originating from running surfaces with different compliance, the objective of this study was to investigate how loads are distributed over the plantar surface while running on natural grass and on a rigid surface—asphalt. Forty-four adult runners with 4 ± 3 years of running experience were evaluated while running at 12 km/h for 40 m wearing standardised running shoes and Pedar insoles (Novel). Peak pressure, contact time and contact area were measured in six regions: lateral, central and medial rearfoot, midfoot, lateral and medial forefoot. The surfaces and regions were compared by three ANOVAS (2 × 6). Asphalt and natural grass were statistically different in all variables. Higher peak pressures were observed on asphalt at the central (p < 0.001) [grass: 303.8(66.7) kPa; asphalt: 342.3(76.3) kPa] and lateral rearfoot (p < 0.001) [grass: 312.7(75.8) kPa; asphalt: 350.9(98.3) kPa] and lateral forefoot (p < 0.001) [grass: 221.5(42.9) kPa; asphalt: 245.3(55.5) kPa]. For natural grass, contact time and contact area were significantly greater at the central rearfoot (p < 0.001). These results suggest that natural grass may be a surface that provokes lighter loads on the rearfoot and forefoot in recreational runners.     

Regional plantar pressure during walking, stair ascent and descent

Regional plantar pressures during stair walking may be injurious in at risk populations. However, limited data are available examining the reliability of plantar pressure data collected during stair walking. The aims of this study were three fold; to assess the reliability of the plantar pressure data recorded during stair walking, to assess the effects of level ground and stair walking on plantar loading, and to develop regression equations to predict regional plantar pressures in stair walking from those collected on level ground. Fifteen subjects without conditions affecting their ability to walk on level surfaces or stairs were recruited. Each participant performed at least 10 steps in level ground and stair walking while plantar pressure data were recorded in six foot regions. Reliability was assessed using Intraclass Correlation Coefficient. A repeated measures ANOVA was used to assess the effect of activity on plantar pressure, and a linear regression was used to predict forefoot loading during stair walking. A reliability of 0.9 was achieved within 10 steps in all foot regions, with the forefoot requiring fewer steps. Plantar pressures were influenced by both, foot region and activity, with the heel and forefoot regions generally experiencing lower peak pressures and maximal forces during stair walking than level ground walking. The regression equations predicting peak pressure during stair walking accounted for between 37% and 70% of the variance of the stair walking data. These findings establish the reliability of plantar pressure data collected during stair walking. Future studies should investigate these parameters in clinical populations.     

Radiographic-directed local coordinate systems critical in kinematic analysis of walking in diabetes-related medial column foot deformity

Diabetic foot deformity onset and progression maybe associated with abnormal foot and ankle motion. The modified Oxford multi-segmental foot model allows kinematic assessment of inter-segmental foot motion. However, there are insufficient anatomical landmarks to accurately representation the alignment of the hindfoot and forefoot segments during model construction. This is most notable for the sagittal plane which is referenced parallel to the floor, allowing comparison of inter-segmental excursion but not capturing important sagittal hind-to-forefoot deformity associated with diabetic foot disease and can potentially underestimate true kinematic differences. The purpose of the study was to compare walking kinematics using local coordinate systems derived from the modified Oxford model and the radiographic directed model which incorporated individual calcaneal and 1st metatarsal declination pitch angles for the hindfoot and forefoot. We studied twelve participants in each of the following groups: (1) diabetes mellitus, peripheral neuropathy and medial column foot deformity (DMPN+), (2) DMPN without medial column deformity (DMPN−) and (3) age- and weight-match controls. The modified Oxford model coordinate system did not identify differences between groups in the initial, peak, final, or excursion hindfoot relative to shank or forefoot relative to hindfoot dorsiflexion/plantarflexion during walking. The radiographic coordinate system identified the DMPN+ group to have an initial, peak and final position of the forefoot relative to hindfoot that was more dorsiflexed (lower arch phenotype) than the DMPN− group (p < .05). Use of radiographic alignment in kinematic modeling of those with foot deformity reveals segmental motion occurring upon alignment indicative of a lower arch.     

The biomechanical effects of pronated foot function on gait. An experimental study

The relationship between foot kinematics and the development of lower extremity musculoskeletal disorders (MSD) has been the focus of recent attention. However, most studies evaluated static foot type and not dynamic foot function. The purpose was to compare lower limb and foot kinematics, and plantar pressures during gait in physically active individuals with pronated and non-pronated foot function. Foot function in 154 adult participants was documented as pronated (n = 63) or neutral (n = 91) using 2 established methods: The Foot Posture Index and the Center of Pressure Excursion Index. Difference between the groups in triplanar motion of the lower limb during barefoot gait was evaluated using a 3D motion capture system incorporating the Oxford Foot Model. Dynamic parameters of plantar pressure were recorded using a pressure platform. Anterior-posterior pelvic tilt range of motion (ROM), peak knee internal rotation, forefoot dorsiflexion ROM, peak forefoot abduction, and rearfoot eversion were all increased in those with pronated foot function. Hallux contact time and time to peak force under the medial forefoot were increased with pronated foot function, and maximal force under the lateral forefoot was reduced. Pronated foot function affected the whole lower limb kinematic chain during gait. These kinematic alterations could increase the risk of developing MSD. Further studies should elucidate the relationship between pronated foot function and MSD, and, if confirmed, foot function should be evaluated in clinical practice for patients with lower limb and low back pain.     

In shoe pressure measurements during different motor tasks while wearing safety shoes: The effect of custom made insoles vs. prefabricated and off-the-shelf

Health and safety regulations in many countries require workers at risk to wear safety shoes in a factory environment. These shoes are often heavy, rigid, and uncomfortable. Wearing safety shoes daily leads to foot problems, discomfort and fatigue, resulting also in the loss of numerous working days. Currently, knowledge of the biomechanical effects of insoles in safety shoes, during working activities, is very limited.Seventeen workers from a metalworking factory were selected and clinically examined for any foot conditions. Workers feet were 3D scanned, with regards to their plantar view, and the images used to design 34 custom-insoles, based on foot and safety shoe models.Three insoles were blind-tested by each worker: custom (CUS); prefabricated with the safety-shoe (PSS), and off-the-shelf (OTS). Foot-to-insole pressure distribution was measured in seven motor tasks replicating typical working activities: single and double-leg standing; weight lifting; stair ascending and descending; normal and fast walking.Wearing CUS within safety shoes resulted in a greater uniform pressure distribution across plantar regions for most of the working activities. Peak pressure at the forefoot during normal walking was the lowest in the custom insole (CUS 275.9 ± 55.3 kPa; OTS 332.7 ± 75.5 kPa; PSS 304.5 ± 54.2 kPa). Normal and fast walking were found to be the most demanding activities in terms of peak pressure.Wearing safety shoes results in high pedobarographic parameters in several foot regions. The use of custom insoles designed on the foot morphology helps decrease peak pressure and pressure-time integral compared to prefabricated featureless insoles.     

Effects of an auditory biofeedback device on plantar pressure in patients with chronic ankle instability

Chronic ankle instability (CAI) patients have been shown to have increased lateral column plantar pressure throughout the stance phase of gait. To date, traditional CAI rehabilitation programs have been unable to alter gait. We developed an auditory biofeedback device that can be worn in shoes that elicits an audible cue when an excessive amount of pressure is applied to a sensor. This study determined whether using this device can decrease lateral plantar pressure in participants with CAI and alter surface electromyography (sEMG) amplitudes (anterior tibialis, peroneus longus, medial gastrocnemius, and gluteus medius). Ten CAI patients completed baseline treadmill walking while in-shoe plantar pressures and sEMG were measured (baseline condition). Next, the device was placed into the shoe and set to a threshold that would elicit an audible cue during each step of the participant's normal gait. Then, participants were instructed to walk in a manner that would not trigger the audible cue, while plantar pressure and sEMG measures were recorded (auditory feedback (AUD FB) condition). Compared to baseline, there was a statistically significant reduction in peak pressure in the lateral midfoot–forefoot and central forefoot during the AUD FB condition. In addition, there were increases in peroneus longus and medial gastrocnemius sEMG amplitudes 200 ms post-initial contact during the AUD FB condition. The use of this auditory biofeedback device resulted in decreased plantar pressure in the lateral column of the foot during treadmill walking in CAI patients and may have been caused by the increase in sEMG activation of the peroneus longus.     

Pressure characteristics in painful pes cavus feet resulting from Charcot-Marie-Tooth disease

Charcot–Marie–Tooth (CMT) disease often presents with peripheral muscle imbalance associated with a painful cavus (medial high-arched) foot deformity which becomes increasingly severe and rigid as the disease progresses. The purpose of this study was to investigate the effect of pes cavus on foot pain and dynamic plantar pressure in CMT, and to explore the relationships between plantar pressure and pain. Sixteen participants diagnosed with CMT and painful pes cavus were assessed for foot posture, ankle dorsiflexion range of motion, levels of foot pain, functional impairment, health-related quality of life and plantar pressure distribution while walking. Plantar pressure parameters (mean pressure, peak pressure, pressure–time integral) and contact duration were measured using the Novel Pedar^® in-shoe capacitance transducer system and the foot was divided into rearfoot, midfoot and forefoot regions for analysis. Increasing cavus foot deformity was associated with more widespread foot pain and increased pressure under the forefoot and midfoot regions. In contrast, peak pressure decreased under the rearfoot. Neither relationship was found between foot pain intensity and any of the pressure variables, nor was ankle dorsiflexion range of motion correlated with pain location, intensity or degree of pes cavus. Although pes cavus in CMT is associated with substantial pain and dysfunction, there is no clear link between foot pain and plantar pressure. The more severe the degree of pes cavus, however, the more pressure develops under the lateral margin of the foot; probably as a result of the changed foot–ground contact seen during gait.     

Variations in plantar pressure variables across five cardiovascular exercises

PURPOSE: To quantify variations in plantar pressure variables in healthy adults across five cardiovascular exercises. METHODS: Ten young (19-35 yr old) and 10 middle-aged (45-60 yr old) individuals participated. After equipment familiarization, plantar pressure data were recorded during walking, running, elliptical training, stair climbing, and recumbent biking. Separate one-way analyses of variance with repeated measures identified significant differences in pressure variables across exercises and between age groups under the forefoot, arch, and heel. RESULTS: Forefoot: Peak pressures were higher during walking (253 kPa), running (251 kPa), and elliptical training (213 kPa) than stair climbing (130 kPa) and recumbent biking (41 kPa; P < or = 0.001). Biking pressures were lower than all other conditions (P < 0.001). Arch: Pressures were higher during running (144 kPa) compared with all other conditions (P < or = 0.001). Intermediate-level pressures during walking (119 kPa) and elliptical training (102 kPa) exceeded those during stair climbing (80 kPa; P < or = 0.002). Pressures were lowest during recumbent biking (33 kPa; P < 0.001). Heel: Pressures were highest during walking (215 kPa) and running (188 kPa), exceeding those recorded during all other activities (P < 0.001). Moderate elliptical training pressures (94 kPa) surpassed stair climbing values (66 kPa; P = 0.014). Pressures were lowest during recumbent biking (25 kPa; P < 0.001). The only significant difference identified between age groups was a larger arch contact area in the young compared with middle-aged, when averaged across exercises (P = 0.011). CONCLUSIONS: When protection of the forefoot is important (e.g., diabetic foot neuropathies), biking and stair climbing offer optimal pressure reductions. If protecting the heel from high pressures and forces is warranted, recumbent biking, stair climbing, and elliptical training provide greater relief.     

Effects of flexible and rigid rocker profiles on in-shoe pressure

Rocker profiles are commonly used in the prevention of diabetic foot ulcers. Rockers are mostly stiffened to restrict toe plantarflexion to ensure proper offloading. It is also described that toe dorsiflexion should be restricted. However, the difference in effect on plantar pressure between rigid rockers that restrict this motion and flexible rockers that do not is unknown. In-shoe plantar pressure data were collected for a control shoe and the same shoe with rigid and flexible rockers with the apex positioned at 50% and 60%. For 29 healthy female adults peak plantar pressure (PP), maximum mean pressure (MMP) and force-time integral (FTI) were determined for seven regions of the foot. Generalized estimate equation was used to analyse the effect of the different shoes on the outcome measures for these regions. Compared to the control shoe a significant increase of PP and FTI was found at the first toe for both rigid rockers and the flexible rocker with the apex positioned at 60%, while MMP was significantly increased in rockers with an apex position of 60% (p < 0.001). PP at the first toe was significantly lower in flexible rockers when compared to rigid rockers (p < 0.001). For both central and lateral forefoot PP and MMP were significantly more reduced in rigid rockers (p < 0.001), while for the medial forefoot no differences were found. The use of rigid rockers results in larger reductions of forefoot plantar pressures, but in worse increase of plantar pressures at the first toe compared to rockers that allow toe dorsiflexion.     

Dynamic plantar pressure patterns in children and adolescents with Charcot-Marie-Tooth disease

BackgroundThe dynamic plantar pressure patterns of children and adolescents with Charcot-Marie-Tooth (CMT) disease and its relationship to musculoskeletal alterations may help to understand the natural history of the disease and improve therapeutic interventions.Research questionThe study compared dynamic plantar pressure patterns in children and adolescents with and without CMT. It also tested the associations between isometric muscle strength (IMS), passive range of motion (ROM), foot posture and dynamic plantar pressure patterns in CMT.MethodsThis cross-sectional study compared children and adolescents (aged 8–18 years) with CMT (n = 40) with a typical group (n = 40). The plantar pressure distribution during gait was recorded, and the contact area (CA), peak pressure (PP), contact time (CT) and pressure-time integral (PTI) in five foot regions (rearfoot, midfoot lateral, midfoot medial, lateral forefoot and medial forefoot) were analysed. The IMS of the dorsiflexors and plantar flexors, passive ROM, and foot posture were also recorded.ResultsPP (medial midfoot and medial forefoot) and PTI (rearfoot, lateral midfoot and medial forefoot) were higher in children with CMT compared with the typical group. The adolescents with CMT presented a less CA (whole foot) and a higher CT (medial midfoot) when compared with typical group. For CMT, in the medial midfoot, plantar flexor IMS associated with PP (β=-11.54, p = 0.01) and PTI (β=-3.38, p = 0.04); supinated foot posture associated with PP (β = 33.89, p = 0.03) and PTI (β = 12.01, p = 0.03).SignificanceChildren with CMT showed clear changes in most of the dynamic plantar pressure variables, while adolescents with CMT showed changes mostly in CA and CT. This information together with the associations established between supinated foot, dorsiflexion ROM and plantar flexions IMS can be useful for guiding rehabilitation professionals in their therapies.     

Understanding the role of foot biomechanics on regional foot orthosis deformation in flatfoot individuals during walking

BackgroundFoot orthoses (FOs) are one of the most common interventions to restore normal foot mechanics in flatfoot individuals. New technologies have made it possible to deliver customized FOs with complex designs for potentially better functionalities. However, translating the individuals’ biomechanical needs into the design of customized FOs is not yet fully understood.Research questionOur objective was to identify whether the deformation of customized FOs is related to foot kinematics and plantar pressure during walking.MethodsThe kinematics of multi-segment foot and FOs contour were recorded together with plantar pressure in 17 flatfoot individuals while walking with customized FOs. The deformation of FOs surface was predicted from its contour kinematics using an artificial neural network. Plantar pressure map and deformation were divided into five anatomically based regions defined by the corresponding foot segments. Forward stepwise linear mixed models were built for each of the four gait phases to determine the feet-FOs interaction.ResultsIt was observed that some associations existed between foot kinematics and pressure with regional FOs deformation. From heel-strike to foot-flat, longitudinal arch angle was associated with FOs deformation in forefoot. From foot-flat to midstance, rearfoot eversion accounted for variation in the deformation of medial FOs regions, and forefoot abduction for the lateral regions. From midstance to heel-off, rearfoot eversion, longitudinal arch angle, and plantar pressure played significant role in deformation. Finally, from heel-off to toe-off, forefoot adduction affected the deformation of forefoot and midfoot.SignificanceThis study provides guidelines for designing customized FOs. Flatfoot individuals with excessive rearfoot eversion or very flexible medial arches require more support on medial FOs regions, while the ones with excessive forefoot abduction need the support on lateral regions. However, a compromise should be made between the level of support and the level of increase in plantar pressure to avoid stress on foot structures.     

Effect of a carbon reinforcement for maximizing shoe outsole bending stiffness on plantar pressure and walking comfort in people with diabetes at high risk of foot ulceration

BackgroundDifferent shoe design features can reduce peak plantar pressure to help prevent foot ulcers in people with diabetes. A carbon reinforcement of the shoe outsole to maximize bending stiffness is commonly applied in footwear practice, but its effect has not been studied to date.Research questionWhat is the effect of a carbon shoe-outsole reinforcement on peak plantar pressure and walking comfort in people with diabetes at high risk of foot ulceration?MethodsIn 24 high-risk people with diabetes, in-shoe regional peak pressures were measured during walking at a comfortable speed in two different shoe conditions: an extra-depth diabetes-specific shoe with a non-reinforced outsole and the same type of shoe with a 3-mm-thick full-length carbon reinforcement of the outsole. The same custom-made insole was worn in both shoe conditions. Walking comfort was assessed using a Visual Analogue Scale (0–10, 10 being highest possible comfort).ResultsSignificantly lower metatarsal head peak pressures (by a median 10–22 kPa) were found with the reinforced shoe compared to the non-reinforced shoe (p < .001). In >83% of cases with the reinforced shoe and >71% with the non-reinforced shoe metatarsal head peak pressures were <200 kPa. At the hindfoot, peak pressures were significantly higher (by a median 24 kPa) with the reinforced shoe (p = .001). No significant shoe effects were found for the toes. No significant shoe effects were found for walking comfort: median 6.1 for the reinforced shoe versus 5.6 for the non-reinforced shoe.SignificanceAdding a full-length carbon reinforcement to the outsole of a diabetes-specific shoe significantly reduces peak pressures at the metatarsal heads, where ulcers often occur, in high-risk people with diabetes, and this does not occur at the expense of patient-perceived walking comfort.     

Forefoot plantar shear stress distribution in hallux valgus patients

BackgroundHallux valgus is a common foot disorder often experienced with secondary callosities and metatarsalgia. Many factors including improper shoes might be responsible in the pathophysiology of the problem. Hallux valgus deformity has been shown to alter the biomechanics of the whole foot rather than affecting only the great toe. Due to changes in the biomechanical functioning of the first ray, other regions of the forefoot area have been shown to bear abnormal loads with increased vertical loading on medial, central and lateral forefoot regions. The purpose of this study was to investigate the pattern of forefoot plantar shear loading in hallux valgus patients and compare these results with those of control subjects.MethodsA total of 28 subjects were recruited for the study of which 14 were clinically diagnosed with hallux valgus. A custom built platform was used to collect peak pressure and shear data. A repeated measures analysis of variance was used to analyze the recorded data.FindingsAntero-posterior shear was significantly lower in the deformity group (p < 0.05). The lateral forefeet of the patients, however, experienced slightly higher shear loads (p > 0.05).InterpretationPropulsive shear force generation mechanism under the medial forefoot was impaired in the disorder group. In general, shear loading of the plantar feet shifted laterally. Previously hypothesized higher medio-lateral shear magnitudes under the hallux were not confirmed.     

Effect of gait speed changes on foot loading characteristics in children

Gait speed has been shown to influence foot loading patterns in adults but the mechanism has not been investigated in children. The present study investigated the effects of changes in gait speed on foot loading characteristics in 20 typically developing children who participated in plantar pressure measurements at normal, slow and fast walking speeds. In spite of shorter contact times in the fast walking speed condition, significantly increased foot loading was seen in the hindfoot, medial and central forefoot and toes while it slightly decreased in the lateral midfoot and forefoot. The results generally confirm the findings in adults that gait speed does not uniformly affect foot loading characteristics and that these effects should be kept in mind when comparing different subject groups or children at repeated measurement occasions.     

Changes to stance limb peak,cumulative, and regional plantar foot forces among normal walking and three mobility aids in healthy older adults

BackgroundMobility aids are commonly prescribed to offload an injured lower extremity. Device selection may impact stance foot loading patterns and foot health in clinical populations at risk of foot ulceration.Research questionsTwo questions motivated this study: How does device selection influence peak plantar and regional (rearfoot, mid foot and forefoot) foot forces on the stance foot? Does device selection influence peak, cumulative, and regional plantar forces within a 200 m walking trial?MethodsTwenty-one older adults walked 200 m at self-selected pace in four randomized conditions for this prospective crossover study. Participants used a walker, crutches, wheeled knee walker (WKW), and no assistive device (control condition). Plantar forces were measured using a wireless in-shoe system (Loadsol, Novel Inc., St. Paul, MN). Repeated measures analyses of variance were used to determine differences in peak and cumulative total and regional forces among walking conditions. Paired sample t-tests compared forces during first and last 30 s epochs of each condition to determine device influence over time.ResultsThe WKW reduced peak net forces by 0.29 and 0.35 bodyweight (BW) when compared to the walker or control condition with similar trends in all foot regions. Crutch use had similar peak forces as control. There were no differences in the number of steps taken within devices comparing first and last epochs. Crutches had a 0.04 and 0.07 BW increase in peak net and forefoot forces during the last epoch. Walker use had 66.44 BW lower cumulative forefoot forces in the last epoch.SignificanceCrutches had similar stance foot loading as normal walking while a walker lowered forefoot forces at the expense of increased steps. A WKW may be the best choice to ‘protect’ tissues in the stance foot from exposure to peak and cumulative forces in the forefoot region.