Abstract
Objective:
Little has been explored about the disparate contribution of medial longitudinal arch (MLA) and lateral longitudinal arch (LLA) to human gait and postural stability. This study aims to investigate the correlation of foot feature parameters including both MLA and LLA with postural stability.
Method:
Thirteen young and healthy subjects participated in this study. The newly developed FFMS extracted foot feature parameters in nonweight-bearing (NWB) and weight-bearing (WB) conditions along with postural stability parameters in single-leg-standing (SLS) condition. A bivariate correlation analysis was carried out to investigate the correlation between the foot characteristics and the postural stability parameters.
Results:
The foot length and width showed negative correlation with center of pressure (CoP) distance in medio-lateral (ML) and total direction, whereas the foot length in NWB and WB conditions, and the foot width in WB condition showed positive correlation with CoP distance in anterior-posterior (AP) direction. The height of the LLA curve and the area of the MLA were correlated with the postural stability parameters in AP direction. The ratios of the LLA height and area showed moderate correlation with the CoP distance in ML direction and total direction.
Conclusion:
The size of a foot, such as the length and width, is correlated with postural stability. Whereas the MLA features are associated with postural stability in AP direction, the LLA features are associated with that in ML and total direction.
Application:
The findings suggest that the roles and contributions of the MLA and LLA features in and to the postural control are different.
Keywords
Introduction
The human foot is a unique structure that has an important receptive field in human gait (Wright, Ivanenko, & Gurfinkel, 2012). In the functioning of the foot, the windlass mechanism is one of the most important factors. The plantar aponeurosis, the thick connective tissue supporting the foot arch, absorbs shock in movement and provides propulsion force at push-off by altering the height of the foot arch. This reduces the vertical displacement of the center of mass and allows human gait to be both natural and aesthetic (Hayot, Sakka, & Lacouture, 2013; Lin et al., 2013; Stolwijk et al., 2014).
As important as the foot mechanism, stability matters greatly in managing natural gaits. Postural stability, which can be defined as the ability to keep the body in equilibrium during certain movements, can be measured by observing the excursion of the center of pressure or CoP (de Kam, Kamphuis, Weerdesteyn, & Geurts, 2016; Hertel, Gay, & Denegar, 2002; Kim, Lim, & Yi, 2015; Marinho-Buzelli, Rouhani, Masani, Verrier, & Popovic, 2017; Wikstrom, Fournier, & McKeon, 2010; Zumbrunn, MacWilliams, & Johnson, 2011). Any lack of postural stability can challenge daily activities, which can be seen often among the elderly or patients with neurological disorders (Bizovska, Svoboda, Vuillerme, & Janura, 2017; Michikawa, Nishiwaki, Takebayashi, & Toyama, 2009; Vellas et al., 1997).
One element of the foot structure that can be highly associated with postural stability is the arch. The foot arch comprises the medial longitudinal arch (MLA), the lateral longitudinal arch (LLA), and transverse arch. The foot arch contributes to shock absorption, provides propulsion, and controls posture. The relation between foot arch and postural stability has been investigated, and several studies successfully demonstrated the impact of foot arch structure on the excursion of CoP. A study showed that people with cavus feet show larger CoP excursion compared with those with normal foot structure (Hertel et al., 2002), and another confirmed this finding by demonstrating that people with flatfoot had larger CoP excursion compared with those with neutral foot (Kim et al., 2015). In addition, there was a study which discovered the contribution of the foot architecture and physiology to the task of postural control in anterior-posterior (AP) direction (Wright et al., 2012). Even so, most studies have not taken the LLA into account and the experimental conditions were rather limited to static ones.
Compared with the attention that has been paid to the MLA, the LLA has been rather neglected largely due to the challenges derived from its measurement. As for measuring the MLA and the LLA especially in dynamic conditions, the LLA brings much more difficulties than the MLA because the LLA cannot be observed from the side but should be observed from the bottom. Yet, given their disparate contribution to shock absorption and movement reported (Fukano & Fukubayashi, 2009), taking not only the MLA but also the LLA into account is imperative. To address this measurement issue, a novel foot feature measurement system (FFMS) that can measure both the MLA and the LLA has recently been built and validated (Chun, Kong, Mun, & Kim, 2017).
Thus, this study aims to investigate the correlation of foot feature parameters including both the MLA and the LLA with postural stability using the developed FFMS. The distinctive roles of the MLA and LLA in postural control are to be investigated. The findings will help establish a notion that the roles and contributions of the MLA and LLA in and to postural stability are distinctive and break new ground in the field of ergonomics and applied physiology.
Materials and Method
Foot Feature Measurement System
The newly developed foot feature measurement system (FFMS) was used for extracting foot feature parameters for this study. The system consists of a measurement unit which is a runway with a scanning stage installed underneath (Figure 1A) and analysis modules. The runway is a 2-m wooden pathway with a single depth camera (Intel Realsense F200) embedded in the middle. While the four uniaxial force sensors at each corner of the transparent acrylic panel track excursion of the CoP, the depth camera in the center captures the morphological characteristics of the foot and the geometric structure of the foot arch based on three-dimensional (3D) images (Figure 1B). The data about the CoP excursion and foot images are automatically time-synchronized and stored in the in-house software developed. Based on this information, the analysis module draws the geometric structure and CoP excursion data of the foot and extracts foot feature parameters such as foot length, foot width, and curves of the MLA and LLA (Figure 1B). A validation of this FFMS has been carried out, and the system was proven to be reliable, showing the average error for arch height is 0.53 mm (Chun et al., 2017).
Overall foot feature measurement system (FFMS): (a) The measurement system comprises a wooden structure of 200 cm (length) × 70 cm (width) × 45 cm (height), and the scanning stage is composed of a single depth camera with four uniaxial force sensors at the corners of the panel; (b) The analysis system can be used to analyze the geometric structure and CoP trajectory based on the obtained depth data; and (c) The defined foot feature parameters such as foot length, width, height, and area of MLA and LLA curves can be extracted from the analysis system.
Subjects
Thirteen healthy male subjects with the mean age of 28.08 years participated in this study (mean height: 174.31 ± 8.00 cm; mean weight: 77.15 ± 14.31 kg). No subject was suffering from foot injuries at the time of experiment, and subjects with abnormalities in both gait pattern and foot structure were excluded. All subjects gave written informed consent prior to the experiment, and there were no human rights violations reported throughout the entire experiment. This research complied with the tenets of the Declaration of Helsinki and was approved by the Institutional Medical Ethics Review Board at Korea Institute of Science and Technology. Informed consent was obtained from each participant.
Experimental Protocol
First, an experienced experimenter palpated to find and mark the calcaneus bone, navicular bone, first metatarsophalangeal (MTP) bone, and fourth MTP bone of each subject. Then three movement conditions, which were nonweight bearing (NWB), weight bearing (WB), and single-leg-standing (SLS), were given. In the NWB condition, all were instructed to sit on a height adjustable chair prepared in the middle of FFMS for 5 s with their food placed on the acrylic panel. The subjects were asked to keep their ankle in the neutral position while keeping their knee joint at an angle of 90°. As for the WB condition, all subjects were instructed to walk down the runway making sure that their foot stepped on the acrylic panel in their second stride. The foot images when the ground reaction force was the biggest were used for analysis. For SLS condition, all subjects were instructed to balance for 10 s on top of the acrylic panel with one leg lifted and bent keeping the knee and hip joints at an angle of 90°. All these experimental protocols were done for both lower limbs for each subject.
Foot Feature Parameters
From the obtained images, the analysis module extracted the foot feature parameters such as foot length and width (Figure 1B). The foot axis was defined as the line from the center of the heel to the tip of the second toe, which passes through the calcaneus bone; the length of this foot axis was set as the foot length. The foot width was defined as the line perpendicular to the foot axis, which meets the navicular bone (Figure 1C; Cavanagh & Rodgers, 1987). The ratio of foot length and width was defined as the foot length and width in the NWB condition divided by the corresponding values in the WB condition, respectively.
The MLA line was defined as the line that connects the calcaneus bone and the first MTP joint, whereas the LLA line was defined as the line that connects calcaneus and the fourth MTP joint (Figure 1C). Projecting the MLA and the LLA lines onto the plantar surface of the foot extracted the MLA and LLA curves. The height of the MLA was defined as the apex of the MLA curve; the same applied to the height of the LLA. The space area under the MLA and the LLA curve from the ground was defined as the area of the MLA and the area of the LLA, respectively (Figure 1C). The MLA height in NWB divided by the MLA height in WB became the ratio of MLA height, whereas the MLA curve area in NWB divided by the MLA curve area in WB became the ratio of MLA area. The same manner of calculation was applied to compute the ratio of LLA height and ratio of LLA area.
Postural Stability Parameters
From the data about the CoP excursion in SLS condition, postural stability was calculated. The postural stability parameters used in this study are as follows: standard deviation (CoP SD) in anterior-posterior (AP) and mediolateral (ML) direction, maximum moving area (CoPArea), and moving distance (CoPDistance) in AP, ML, and total direction (Michikawa et al., 2009; Vellas et al., 1997; Zumbrunn et al., 2011).
Statistics
For statistical analysis, SPSS software (SPSS, Chicago, IL) was employed and the bivariate correlation analysis on the foot feature parameters and postural stability parameters was carried out. The correlation coefficient r which shows the level of correlation was classified into very high (.9–1), high (.7–.9), moderate (.5–.7), low (.3–.5), and very low (.1–.3). When the coefficient was greater than .05, the two parameters were considered to be correlated. In this study, the significance level was set to p < .05.
Results
Correlation of Foot Length and Width With Postural Stability
The extracted foot feature parameters (Table 1) and postural stability parameters (Table 2) showed that both foot length and width in NWB and WB conditions have low and moderate correlations with postural stability with the coefficient r varying from .454 to .571 (Table 3).
Mean and Standard Deviation of Foot Size and Foot Arch Between NWB and WB Condition
Note. NWB = nonweight bearing; WB = weight bearing; MLA = medial longitudinal arch; LLA = lateral longitudinal arch.
Mean and Standard Deviation of the Postural Stability Parameters During SLS Condition
Note. SLS = single leg standing; CoP = center of pressure; AP = anterior-posterior; ML = mediolateral.
Correlation Coefficients and Significances Between Foot Size and Postural Stability
Note. CoP = center of pressure; AP = anterior-posterior; ML = mediolateral; NWB = nonweight-bearing; WB = weight-bearing.
p < .05. **p < .01.
As for the correlation of foot length and width with CoP distance, while the foot length in NWB and WB conditions showed moderate and high correlation (r = .541 to .543) with CoP distance in AP direction, the foot width in WB condition showed low correlation with it (r = .442). The foot length and width, however, showed negative correlation with CoP distance in ML and total direction (r = −.602 to −.750). The ratios of the foot length and width in WB condition did not differ from those in NWB condition. No significant correlation of these ratios with the postural stability parameters was observed (Table 3).
Correlation of the MLA and LLA With Postural Stability
Whereas the height of the MLA curve in both the NWB and WB conditions did not show any significant correlation with postural stability parameters (Table 4), the height of the LLA curve in the WB condition showed a moderate correlation with the postural stability parameters particularly in AP direction (r = .519). The area of the MLA in both NWB and WB conditions showed low correlation with the postural stability parameters in AP direction (r = .471 and .43, respectively). The ratios of the LLA height and area showed moderate correlation with the CoP Distance in ML direction (r = .503 and .558) and total direction (r = .538 and .552). The changes in the MLA and LLA curves under NWB and WB conditions are presented in Figure 2, and scatter plots showing the correlation between the foot features and postural stability were shown in Figure 3.
Correlation Coefficients and Significances Between Foot Arch Parameters and Postural Stability
Note. CoP = center of pressure; AP = anterior-posterior; ML = mediolateral; MLA = medial longitudinal arch; NWB = nonweight-bearing; WB = weight-bearing; LLA = lateral longitudinal arch.
p < .05. **p < .01.
Medial longitudinal arch and lateral longitudinal arch curves and their standard deviations in static and dynamic conditions. The black line shows the averaged arch curves of 13 subjects, and the blue line shows the standard deviations (mm).
Scatter plots showing the correlation of (a) center of pressure (CoP) distance in anterior-posterior (AP) and medial longitudinal arch (MLA) area in weight-bearing (WB) condition, and (b) CoP distance in mediolateral (ML) and ratio of lateral longitudinal arch (LLA) height.
Discussion
This study aimed to investigate the correlation of foot feature parameters in both static and dynamic conditions with postural stability based on 13 young and healthy subjects. Assisted by the newly developed FFMS, foot feature parameters such as foot length and width along with the MLA and LLA under NWB and WB conditions were extracted. The extracted parameters were compared with postural stability parameters to explore the correlation between the two. Although the foot length and width were found to be positively related to the postural stability in AP direction, those were negatively related to the postural stability in ML and total direction. As for the MLA and LLA, the MLA area in NWB and WB conditions and the LLA height in WB condition were positively correlated with postural stability in AP direction, and the ratio of LLA height and ratio of LLA area were positively correlated with postural stability in ML and total direction. The findings confirm the claims arguing the role and contribution of the MLA and LLA are distinctive in and to the postural stability of different directions.
The findings of this study agree with those of the previous one that discovered the foot length in WB condition was longer than that in NWB condition (Stolwijk et al., 2014). The study also found that bigger foot size is related with better postural stability in ML and total direction but not with that in the AP direction. As Table 1 and Figure 2 shows, the arch height in WB condition was lower than that in NWB condition. This decrease in height was due to the shock caused by the downward force from the body weight (Lin et al., 2013; Samson et al., 2014). Reportedly, the mean difference of arch height between NWB and WB is around 4mm (Bandholm, Boysen, Haugaard, Zebis, & Bencke, 2008; McPoil et al., 2008) but that of the current study was around 2 mm. The experimental protocol that limited the walking distance may have resulted in this reduction.
It’s also worth noting that the height of MLA showed no association with postural stability, whereas the curve area of the MLA was associated with the CoP in AP direction. The reported flattening of the foot arch in AP direction derived from tibia tilting forward during double limb standing condition (Wright et al., 2012) may have influenced the results of the current study. This suggests that the curve area of the MLA rather than the height of the MLA can be a consideration when investigating the correlation between foot features and postural stability. As for the postural stability in ML and total direction, the characteristics of LLA were found to be associated. The ratio of LLA height and area showed a positive correlation with the CoP in ML direction and total direction. Given the noticeable finding from Fukano and his colleague—which claimed the different deformation pattern between MLA and LLA against landing condition and demonstrated the MLA had larger translational motion while the LLA had larger rotational motion during shock absorption (Fukano & Fukubayashi, 2009)—it can be argued that the characteristics of LLA may contribute to postural control in ML and overall direction. In addition, our previous study, which showed the LLA characteristics and MLA characteristics have distinctive roles in human gait, may support the finding of this study (Mun, Song, Chun, & Kim, 2018).
This study can find its contribution in that it investigated the interrelationship between foot anatomical-mechanical features and postural stability not only in static conditions but also in dynamic conditions. Thanks to the newly developed FFMS, the study was able to measure the LLA and trace the changes of the MLA and the LLA made in dynamic conditions. To our knowledge, this study is the first to show that the postural stability in AP direction is related to the MLA features whereas the postural stability in ML and total direction is related to the LLA features.
Be that as it may, this study bears some inevitable limitations. First, the subject number was small. In examining only 13 young and healthy subjects, this study was not able to recruit various arch types or the foot features of the elderly or the injured. That limits this study to be considered preliminary; future studies involving a larger number of subjects with various arch type and physical quality should follow. Second, the gait temporospatial parameters, such as stride time, step time, stride length, step length, gait velocity, and cadence, were not taken into account as indicators that represent gait performance. For studies that involve these parameters, joint range of motions and muscle activation should follow.
Conclusion
Through this study, we conclude that (1) the length and width of a foot are correlated with postural stability and (2) roles and contributions of the MLA and LLA features in and to the postural stability are disparate. We found that the length and width of a foot are correlated with postural stability in that bigger foot size is related to better postural stability in ML and total direction but not with that in the AP direction. The data also proved the disparate roles and contributions of the MLA and LLA features in and to the postural stability, in that the MLA features are associated with postural stability in AP direction whereas the LLA features are associated with that in ML and total direction.
Key Points
Foot anatomical-mechanical features and postural stability were measured by a newly developed foot feature measurements system.
This study found that the length and width of a foot are correlated with postural stability, in that bigger foot size is related with better postural stability in ML and total direction but not with that in the AP direction.
It has also proved the disparate roles and contributions of the MLA and LLA features in and to the postural stability, in that the MLA features are associated with postural stability in AP direction while the LLA features are associated with that in ML and total direction.
Footnotes
Acknowledgements
This research project was supported by the High-tech based national athletic performance improvement (Winter), Korea Sports Promotion Foundation, and the Korea Institute of Science and Technology (KIST) Institutional Program (Project No. 2E29450).
Kyung-Ryoul Mun is a senior research scientist at the Korea Institute of Science and Technology (KIST). His research interests include biomechanics, human motion analysis, gait rehabilitation robotics, and motion analysis with artificial neural networks.
Sungkuk Chun is a senior researcher in the Korea Photonics Technology Institute. His research interests include computer vision and machine learning based human motion analysis and natural user interaction.
Junggi Hong is currently the Dean of Sports Medicine Graduate School at CHA University. His primary research interest is to examine neuromuscular mechanisms of sports injuries and to investigate whether sports injuries could be prevented with various types of sports performance training.
Jinwook Kim is currently a principal research scientist at the Korea Institute of Science and Technology (KIST). His research interests include real-time physics-based simulation, human motion analysis, and human–computer interaction.