Abstract
Background Lower extremity overuse injuries are detrimental to military readiness. Extremes of arch height and heavy loads carried by military personnel are associated with increased risk for overuse injury. Little is known regarding the impact of load carriage on plantar pressure distributions during gait.
Objective The objective of this study was to determine how load carriage affects plantar pressure distributions during gait in individuals with varying arch types.
Design A cross-sectional, repeated-measures design was used for the study.
Methods The study was performed at a research laboratory at Fort Sam Houston, Texas. Service members who were healthy and weighing ≥70 kg were enrolled in the study. The participants (97 men, 18 women; mean age=31.3 years, SD=5.6; mean weight=86.0 kg, SD=11.0) were categorized as having high-, normal-, or low-arched feet on the basis of published cutoff values for the arch height index. Plantar pressure measurements were obtained with the use of an in-shoe pressure measurement system while participants wore combat boots and walked on a treadmill under 3 loaded conditions (uniform, 20-kg load, and 40-kg load). Maximal force (MaxF) and force-time integral (FTI) were assessed with the use of a 9-sector mask to represent regions of the foot. A 3 × 3 repeated-measures analysis of variance was used for analysis across the levels of load and arch type.
Results A significant interaction existed between arch type and load for MaxF and FTI in the medial midfoot, with greater force in low-arched feet. In the medial forefoot, MaxF and FTI were greatest in high-arched feet across all load conditions. In the great toe region, low-arched and normally arched feet had greater MaxF and FTI. The relative distribution of FTI increased proportionately in all regions of the foot regardless of arch type for all load conditions.
Limitations The influence of fatigue, greater loads, and different walking speeds was not assessed.
Conclusions Regardless of arch type, increases in load did not alter the relative distribution of force over the plantar foot during gait. Participants with high-arched feet had greater force in the medial forefoot region, whereas those with normally arched or low-arched feet had greater force in the great toe region, regardless of load. These differences in force distribution may demonstrate different strategies to generate a rigid lever during toe-off.
The US military has been involved in continuous wartime operations for more than 10 years, since the start of operations in Iraq in 2003. During this time frame, musculoskeletal complaints continued to be a primary contributor to disability claims on discharge1 and have been reported to cost up to $500 million annually.2 Although there are many factors that lead to musculoskeletal complaints during military service, the contemporary operating environment is known for requiring extended periods of walking and marching under heavy loads.3 During training and actual combat operations, members of the armed forces may carry very heavy loads for extended periods of time. One infantry unit that collected data during deployment to Afghanistan in 2003 reported an average fighting load of 29 kg, an approach march load (for more prolonged operations) of 46 kg, and an emergency approach march load (in which certain transportation resources were unavailable) of 60 kg across several missions.4
In training and combat environments, these loads have been associated with an increased risk of lower extremity overuse injuries such as foot blisters, metatarsalgia, stress fractures, knee pain, and general strains and sprains.5,6 The influence of load carriage on musculoskeletal injuries may be associated with increased vertical ground reaction force, increased cardiovascular demand, decreased balance, or impaired functional strength.7 Specifically, reduced stride length, increased cadence, increased hip flexion, and increased vertical ground reaction forces proportional to load carried during gait have been documented.5,8–11 Whereas load carriage within various world militaries has a long, documented history, its potential impact on military readiness and long-term disability related to musculoskeletal complaints continues to warrant further investigation.
Although overuse injuries are multifactorial,12,13 moderate evidence exists that extremes of arch height serve as an intrinsic risk factor for lower extremity injuries.14–17 Some researchers have noted that individuals with a rigid, high-arched foot may be at greater risk for ankle injuries, stress fractures, anterior knee pain, and injuries involving the lateral structures of the lower extremity,17–19 whereas those with a low-arched foot or an overpronatory foot have been at increased risk for medial tibial stress syndrome, knee pain, and other injuries involving the medial and soft tissue structures of the lower extremity.17,20,21 To help understand the association between static arch height and the dynamic foot during gait, multivariate models have been developed to describe the influence of arch height and foot shape on plantar pressure and force distributions during gait.22,23 Specifically, Teyhen and colleagues22,23 identified multivariate biomechanical models that link plantar pressure and force distributions predictive of arch height and arch type. It is unclear whether these alterations in plantar pressure and force distributions during gait help to explain the associated injury risk with those with extremely high- or low-arched feet or how load carriage would influence the distribution of pressure and force during gait in those with varying arch types.
Although researchers have begun to address both the influence of load carriage and arch type on gait and its influence on lower extremity injuries, there is scant research that describes the influence of load carriage on distribution of forces over the plantar surface of the foot among those with various arch types. The purpose of this study was to describe the influence of load carriage on the distribution of forces in shod gait across different arch types as classified by arch height index (AHI).22,23 It was hypothesized that increasing loads would alter the plantar pressure and force distributions during gait and that these alterations would vary on the basis of arch type. Understanding how load carriage influences plantar pressure and force distributions during gait on the basis of arch type may contribute to a further understanding of lower extremity injury risk and may help to provide a theoretical foundation for new interventions to prevent and treat injuries associated with load carriage.
Method
Design Overview
A cross-sectional, repeated-measures design was used for the study. Active-duty service members who were healthy in the San Antonio, Texas, area were recruited on the basis of arch type to help ensure adequate representation of service members with high- and low-arched feet on the basis of the AHI. An a priori power analysis determined that 30 participants per group would allow for 80% power, assuming a medium effect size of 0.4. Plantar pressure measurements were obtained while the participants walked on a treadmill at 3.0 mph under each of 3 levels of load: Army combat uniform (ACU), 20 kg, and 40 kg.
Participants
Eligible participants were required to weigh >70 kg and be between 18 and 45 years of age, active duty military, and fluent in the English language. Emancipated minors who were 17 years of age also were allowed to volunteer for this study. Potential participants were excluded if they were pregnant, had an antalgic gait (per self-report or through visual observation), were limited in their occupational activities because of a lower extremity condition, were seeking medical care related to or complaining of a lower extremity injury, had an open wound or skin disease on the plantar surface of either foot, or had any other condition that would preclude them from load carriage of 40 kg. Individuals with a significant history of lower extremity trauma (fractures, surgery, or burns), major foot condition (ie, Charcot foot), back injury that resulted in an antalgic gait, or recurrent overuse injuries that resulted in an asymmetrical gait pattern assessed through visual observation also were excluded. Potential participants with forefoot deformities were excluded from the study. Although individuals under the weight of 70 kg typically carry a 40-kg load in the military, they were excluded to reduce the risk of injury while lifting or carrying the required 40-kg load during this exploratory study. A ceiling weight was not used as an entrance criterion on the basis of the military height and weight standards that all soldiers must maintain.
One hundred fifteen service members who were healthy (97 men, 18 women) were enrolled in this study. Twenty-eight participants had high-arched feet, 61 had normally arched feet, and 26 had low-arched feet. On average, they were 31.3 years of age (SD=5.6) and weighed 86.0 kg (SD=11.0, range=69.4–115.7). Their average height of 177.1 cm (SD=7.0) produced an average body mass index of 27.4 kg/m2 (SD=2.9). There were no differences in age, height, weight, or body mass index across arch type categories (Tab. 1). All participants provided informed consent before the study.
Participant Characteristicsa
Static Measurements of the Arch Type
The AHI is a clinical measure that yields reliable data and yields valid inferences to assess static arch type24,25 and injury risk.17,22 The development of AHI has helped to facilitate comparisons among individuals by normalizing dorsal arch height by foot length and has been found to correlate with specific lower extremity overuse injuries in runners.17 Weight-bearing measures of static arch height and heel-toe length were obtained with the participants standing on a specially constructed platform (Foot Assessment Platform System [FAPS]; eFig. 1).22,26,27 While standing, the participants placed their heels in heel cups spaced 15.2 cm (6 in) apart. A sliding indicator at the medial foot was used to ensure proper alignment of the rear and forefoot. Once properly positioned, the participant was asked to place equal weight on both feet to approximate 50% weight bearing on each limb and was confirmed by means of self-report. Heel-toe length was measured with the use of a centered metal ruler and a sliding bar as the distance from the most posterior aspect of the heel to the longest toe. Arch height was measured with the use of a vertical digital caliper from the platform to the dorsal surface of the foot at 50% heel-toe length (Mitutoyo Corporation, Aurora, Illinois). Arch height index was calculated by dividing arch height by foot length.22,24,25 Total foot length was used instead of truncated foot length because those with visually observable forefoot deformities (eg, claw toes, hammertoes) were excluded from the study. Although measurements were obtained bilaterally, only measurements of the right foot were used for this analysis.
Researchers who obtained the static foot measurements were graduate students trained by a senior researcher on this project. Interrater reliability coefficients (intraclass correlation coefficient [ICC]) and standard error of the measurement (SEM) were calculated on a sample of 30 participants (ICC [2,1] ≥.94; SEM ≤1.02 mm). These values are comparable to prior published reliability values.24,25
In-Shoe Plantar Pressure Measurements of the Foot During Gait
Maximum force (MaxF) and force-time integral (FTI) were recorded with the use of capacitance-based pressure-sensing insoles (Pedar, Novel Electronics Inc, St Paul, Minnesota). The Pedar-x has been reported to yield reliable data for in-shoe measurement of plantar pressures.28–31 Each pressure-sensing insole was 2.5 mm thick, with a sampling rate of 50 Hz. Before each participant's arrival for data collection, the insoles were calibrated across a pressure range of 0 to 60 N/cm2 in intervals of 5 N/cm2 by means of the Trublu calibration device (Novel Electronics Inc).
Plantar pressure measurements were obtained while the participants wore their combat boots. Before data collection, any custom-made or off-the-shelf inserts were removed from the participant's boots, and the appropriate size plantar pressure insert was positioned in the boot. One researcher helped the participants don their boots to ensure that appropriate positioning was maintained. The insert position was verified by appropriate plantar pressure maps before initiating data recording. After a period of familiarization on the treadmill, all participants walked for approximately 30 seconds at 4.8 kph (3.0 mph) on a treadmill (Biodex Rehabilitation Treadmill, Biodex Medical Systems, Shirley, New York) under each of 3 levels of load: ACU, 20 kg of load, and 40 kg of load (eFig. 2). The ACU condition consisted of the participant's body weight plus the weight of the ACU. The 20-kg load condition added 20 kg to the ACU condition in the form of body armor, a simulated weapon built to the same weight and dimensions of an M16, a Kevlar helmet, and lead weights representing ammunition. The 40-kg load condition added an additional 20 kg in the form of a weighted Army-issue backpack. The sequence of loaded conditions was counterbalanced. To allow time for participants to accommodate to treadmill walking under each load condition, the treadmill was slowly ramped up to 4.8 kph (3 mph), and the participant was allowed sufficient time to establish a constant walking speed before recording began.
Data Analysis
Descriptive statistics were calculated to summarize the demographic characteristics of the sample. To ensure sufficient representation of the extremes of arch height, participants were recruited and categorized on the basis of arch type of their right foot (high >0.267, normal, and low <0.229) with the use of previously established cutoff values for AHI.22 Maximum force and FTI were calculated to provide an understanding of the maximal forces and the impulse (area under the force-time curve) for the plantar foot. A 9-sector mask was used that consisted of the following regions: medial and lateral hindfoot; medial and lateral midfoot; medial, middle, and lateral forefoot; great toe; and toes 2 through 5 (eFig. 3). On the basis of the results of prior research,22,23,32 the focus of the analysis was on the following foot regions: entire foot region, great toe, medial forefoot, medial midfoot, and lateral forefoot regions. Changes in MaxF and FTI were analyzed with a 3 × 3 repeated-measures analysis of variance across the levels of load carriage and arch type. If the interaction was significant, the interaction was plotted, implications were discussed, and the simple main effects for each arch type were assessed. If the interaction was not significant, the main effects of load across all arch types were assessed. Bonferroni-adjusted t tests were used for post hoc analysis.
On the basis of previous research, we expected that the FTI values would increase as load increased and the distribution of the load would differ on the basis of arch type; therefore, relative distributions of FTI (FTI region/total FTI plantar surface of the foot × 100%) also were calculated for each of the 9 sectors of the mask across all load conditions. All statistical analyses were performed with the use of SPSS software, version 16.0 (SPSS Inc, Chicago, Illinois).
Role of the Funding Source
This work was funded through the Bone Health Research Program, Telemedicine and Advanced Technology Research Center (TATRC), US Army Medical Research and Material Command, Fort Detrick, Maryland.
Results
An average of 9.8 consecutive error-free steps (SD=0.6) were analyzed for each participant in each load condition. The interaction between load × arch type for MaxF of the entire foot was not significant (P=.92). However, there was a main effect for load (P<.001) for all arch types across all loads that demonstrated increased MaxF values as the load increased (Tab. 2). There was a significant interaction between load × arch type for MaxF (P=.004) in the medial forefoot region. Maximum force was greater in the high-arched feet relative to normally arched and low-arched feet (P<.001) across all loads (eFig. 4). Normally arched feet had higher maximal force values than low-arched feet only for the ACU and ACU+20 kg conditions (P≤.025). The reverse was true at the great toe region, in which low-arched and normally arched feet demonstrated greater MaxF (P≤.004) compared with high-arched feet (eFig. 5). Additionally, there was a significant interaction between load × arch type for MaxF (P=.001) in the medial midfoot region, resulting in greater MaxF in the medial midfoot region in those with low arches (Fig. 1). There were no differences between normally arched and high-arched feet (P≥.95). Beyond increases in MaxF resulting from increased load, there were no significant differences in MaxF for the central forefoot, lateral midfoot, and medial and lateral hindfoot regions.
Post Hoc Analyses of Maximal Force (N)a
Both maximal force (MaxF) and force-time integral (FTI) had a significant load × arch type interaction (P≤.001) in the medial midfoot region. Low-arched feet had higher MaxF and FTI values compared with normally arched and high-arched feet (P≤.005) under all load conditions. There were no differences between normally arched and high-arched feet (P≥.95). ACU=Army combat uniform.
No interaction effects were found for FTI for the entire foot (Tab. 3). However, a significant interaction existed between arch type and load for FTI (P≤.001) in the medial midfoot, with greater impulse in those with low-arched feet. In the medial forefoot, FTI (P=.002) was greatest in high-arched feet across all load conditions, and normally arched feet had higher FTI values compared with low-arched feet in ACU and ACU+40 kg conditions (P≤.04; Fig. 1). In the great toe region, low-arched and normally arched feet had greater FTI (P=.03) under ACU and ACU+20 kg conditions. In general, as load increased, FTI increased, regardless of arch type (P<.001). Although there were differences in absolute values of the FTI on the basis of load and arch type, the relative distribution of FTI in the 9 regions of the foot as a percentage of total FTI for the entire foot increased proportionally under all load conditions regardless of arch type (Fig. 2). Outside of increases in FTI resulting from increased load, there were no significant differences in FTI for the central forefoot, lateral midfoot, and medial and lateral hindfoot regions.
Post Hoc Analyses of Force-Time Integral (N-s)a
Relative distribution of the force-time integral (FTI) values for each area of the foot. As the total load increased, indicated by the change to brighter colors in the image, the relative distribution (percentage of total FTI) during gait remained virtually constant (P>.05). The figure is from a composite pressure map of 9.8 error-free steps for a single participant with normally arched feet under all 3 load conditions. The data overprint is mean percentage of FTI for the group with normally arched feet. The relative distributions (percentage of total FTI) for high- and low-arched types were similarly unchanged over the 3 levels of load. ACU=Army combat uniform.
Discussion
Although previous researchers have identified that increasing load results in increased ground reaction forces, decreased stride length, increased cadence, and increased hip flexion,7 scant knowledge exists to describe how increasing load influences the force distributed throughout the foot and the influence of arch type on that distribution. The results of this study provide preliminary evidence on the impact of load on the distribution of force in the plantar foot during gait. In general, individuals with low-arched feet had greater force in the medial midfoot region. Although not measured in this study, it can be theorized that this pattern might be related to increased pronation relative to normally arched and high-arched feet or a combination of pronation and soft tissue contacting the capacitive insoles. Normally arched and low-arched feet had greater force in the great toe region, whereas high-arched feet had greater force in the medial forefoot region. Future research should determine whether these differences indicate different strategies used by individuals with normally arched and low-arched feet to generate a rigid lever arm during toe-off.
Our findings indicate that each arch type had characteristic force distribution patterns (Fig. 3). These results are supported by previous research that identified arch type in both shod and unshod conditions through the use of parameters derived from plantar pressure measurements.22,32 Similar to the predictive algorithm developed by Teyhen et al,22 greater forces and impulse (FTI) across the first metatarsal (medial forefoot region) was characteristic of high-arched feet. Our results also were similar to results from previous research that identified greater force and impulse across the medial midfoot as being characteristic of low-arched feet and higher force in the great toe indicative of normally arched and low-arched feet.22,32 However, unlike those prior studies, we did not identify increased force along the lateral foot regions in high-arched feet. This discrepancy may be related to the fact that this study recorded plantar pressures in the shod condition with the use of the capacitive-based insoles, whereas the prior study measured the unshod condition with the use of a capacitive-based platform. The results of this study add to the growing body of knowledge describing the relationship between arch height and foot posture and plantar pressure and force distribution.22,23,32
Arch types as categorized by arch height index had characteristic distributions of force-time integral (FTI). Normally arched and low-arched feet displayed significantly higher FTIs in the great toe region, whereas high-arched feet demonstrated higher forces in the medial forefoot. Low-arched feet demonstrated higher FTIs in the medial midfoot compared with normally arched and high-arched feet. The figure is from a composite pressure map of 10 error-free steps from a representative participant for each arch type in the weight-bearing condition. The data overprint is FTI (in newton-seconds). NA=not assessed.
Ultimately, the interest in assessing plantar pressure and force distribution associated with load carriage is to determine how they may be associated with injuries. Although the current study cannot directly answer this question, the results of this study provide data that can influence research in this area. Future research should determine whether the increased force in the medial midfoot in low-arched feet, the increased force in the great toe for normally arched and low-arched feet, and the increased force in the medial forefoot in high-arched feet are associated with the typical injuries associated with load carriage (blisters, metatarsalgia, stress fractures, knee pain, and general strains and sprains).5,6 Additionally, the relative distribution of the forces, as measured by the FTI, was consistent because the load applied to the soldier increased regardless of arch type (Fig. 2). Thus, for loads up to 40 kg, the increased load applied to the soldier resulted in a similar increase in force over the plantar surface of the foot during gait. Theoretically, on the basis of the 2 main findings in this study, future combat boot or orthotic devices may need to be unique to arch type and designed to accommodate increased load. However, some of the design features of these devices to mitigate injury risk may be similar for both the loaded and unloaded gait.
Previous examinations of the biomechanical effects of load carriage have primarily focused on greater ground reaction forces and the alteration of mechanics at the joints extrinsic to the foot. Some authors8,9 have indicated that larger loads led to greater hip flexion and associated trunk lean, which theoretically would increase load over the forefoot region. Although we did not measure hip flexion and trunk lean in this study, the loads used in this study did not result in an increase in pressure or force in the forefoot region. Although the force increased in the entire foot with increased load, the relative distribution of these forces did not alter on the basis of load or arch type (Fig. 2). Future research should determine how the altered trunk position and hip and knee angles typically associated with load carriage to maintain center of gravity over the base of support influences the plantar pressure and force distribution for a given arch type. Loads >40 kg may be required to adequately assess this relationship. A greater understanding of how load influences gait and the distribution of force throughout the foot on the basis of arch type may help to guide orthotic and footwear design.
A number of limitations to this study relate to the data collection procedures. There were no statistically significant differences between the right foot and the left foot. Therefore, the analysis used in this study was on the right foot. However, these findings may not generalize to those with varying foot types. Furthermore, the FAPS board used in determination of AHI provides a support surface underneath the entire foot; thus not allowing the midfoot to descend beneath the level of the forefoot and heel. In individuals with very mobile arches, it is possible that measurement of arch height on a device in which the forefoot and heel are elevated off the surface could result in a different categorization of arch type.
Another limitation of this study relates to the brief duration of load carriage during which measurement was conducted. Although the relative distribution of the force was consistent with increasing loads, it is unknown whether this pattern would persist with greater loads, distances, or if the participant was assessed while walking overground instead of on a treadmill. Fatigue associated with load carriage for many hours or even days may have an impact on the distribution of force that could affect function and the risk of injury that were undetectable in this project. Researchers have found changes in pressure distribution in unshod runners after a marathon that indicated reduced pressure in the toe regions compared with pre-race measurements.33,34 The impact of fatigue on load carriage ultimately may have an effect on other body tissues or alignment, resulting in stresses in specific regions of the foot that could exceed tissue tolerances and result in musculoskeletal injury. Future prospective studies should explore the incidence of tendonopathy of the flexor hallucis longus and examine the effects of fatigue on medial midfoot pressures. Future research also should examine the effects of fatigue on load and arch types and the ability of orthotics or footwear to mitigate these effects. Finally, future research should explore the relationships between plantar pressure and 3-dimensional kinematics of the foot.
This study has several other limitations. Each participant wore his or her combat boots. It is possible that differences in boot type and manufacturer introduced another source of variance, but it was thought that control of this variable with a standard boot would have introduced another source of variance associated with an unfamiliar, ill-fitting, stiff new boot that may have altered gait. Additionally, the speed of gait in this study was set at 4.8 kph (3 mph) to replicate the typical marching pace of military units. Many studies have used either self-selected walking speed or calculated a walking speed on the basis of leg length. Although walking speed is a potential source of variance, the researchers elected to standardize in a manner reflective of the setting soldiers would typically march under load. Finally, individuals with foot pathology and low body weight were excluded from this study. It is possible that a load of >57% body weight or foot pathology could have yielded alterations in plantar pressure distributions dissimilar to the reported findings. The research team elected to exclude these potential participants to reduce risk of injury in slight individuals and out of deference to eliminating the confounding influence of gait pattern alterations associated with pain. Future research might consider similar analysis of the pathological foot.
Conclusions
Increase in load corresponded with increased force across the plantar foot, regardless of arch type. Although the force increased with increased load, the relative distribution of the force across the plantar surface of the foot was consistent. However, differences in force in the medial midfoot, medial forefoot, and the great toe region were able to differentiate individuals on the basis of arch type. Future research should assess the role of orthoses and footwear design to mitigate the increased force associated with arch type.
Footnotes
Dr Goffar, Mr Reber, Mr Naylor, Dr Walker, and Dr Teyhen provided concept/idea/research design. All authors provided writing and data collection. Dr Goffar, Mr Reber, Mr Miller, Mr Naylor, Ms Rodriguez, Dr Walker, and Dr Teyhen provided data analysis. Dr Goffar, Mr Reber, Mr Naylor, and Dr Teyhen provided project management. Dr Goffar and Dr Teyhen provided fund procurement. Mr Christiansen, Mr Naylor, Ms Rodriguez, and Dr Teyhen provided study participants. Dr Goffar, Ms Rodriguez, and Dr Teyhen provided facilities and equipment. Dr Teyhen provided institutional liaisons. Mr Christiansen, Mr Miller, Ms Rodriguez, and Dr Teyhen provided clerical support. Mr Reber and Dr Teyhen provided consultation (including review of manuscript before submission). The authors thank the members of the Foot Assessment Algorithms for Soldiers in Training (FAAST) research group from the US Army–Baylor University Doctoral Program in Physical Therapy class of 2008 for their contributions to the body of knowledge pertaining to plantar pressure distributions and foot posture that made this work possible. The authors acknowledge the ongoing support of Susan Diekrager and Maria Pasquale from Novel Corporation for their technical support.
This research was presented at Army Medical Department, Graduate School Research Day, July 2010. Platform presentations of this research were given at the Combined Sections Meeting of the American Physical Therapy Association; February 9–11, 2011; New Orleans, Louisiana, and the 2nd International Congress on Soldiers' Physical Performance; May 2011; Jyväskylä, Finland. A poster presentation of this research was given at the Texas Physical Therapy Association Annual Conference, October 22, 2010, Arlington, Texas.
This study was approved by the Brooke Army Medical Center Institutional Review Board.
This work was funded through the Bone Health Research Program, Telemedicine and Advanced Technology Research Center (TATRC), US Army Medical Research and Material Command, Fort Detrick, Maryland.
The views expressed in this article are those of the authors and do not reflect the official policy or position of Brooke Army Medical Center, the US Army Public Health Command, the US Army Medical Department, the US Army Office of the Surgeon General, the Department of the Army, the Department of the Air Force, the Department of Defense, or the US Government.
- Received March 12, 2012.
- Accepted April 8, 2013.
- © 2013 American Physical Therapy Association
