The overall purpose of this study was to investigate coupling within the distal foot during tasks of varying complexity and MTP positioning, with the goal of garnering greater understanding of the passive windlass mechanism’s role during dynamic movement. In support of our first hypothesis, the ratio of midtarsal to MTP motion increased from isolated MTP extension to dynamic MTP extension. Our kinematic results help determine limits on passive kinematic coupling, and are in line with a recent similar study (23), but with some important differences (detailed below). Regarding kinetic coupling, our hypo
The overall purpose of this study was to investigate coupling within the distal foot during tasks of varying complexity and MTP positioning, with the goal of garnering greater understanding of the passive windlass mechanism’s role during dynamic movement. In support of our first hypothesis, the ratio of midtarsal to MTP motion increased from isolated MTP extension to dynamic MTP extension. Our kinematic results help determine limits on passive kinematic coupling, and are in line with a recent similar study (23), but with some important differences (detailed below). Regarding kinetic coupling, our hypothesis that changes in MTP joint negative work would be proportional to changes in midtarsal positive work was also met, and the differences in kinematic and kinetic coupling provide additional insights into distal foot energy sources. And lastly, our investigation of foot structural differences suggests that there is not a strong relationship between foot structure and kinematic foot coupling.
hesis that changes in MTP joint negative work would be proportional to changes in midtarsal positive work was also met, and the differences in kinematic and kinetic coupling provide additional insights into distal foot energy sources. And lastly, our investigation of foot structural differences suggests that there is not a strong relationship between foot structure and kinematic foot coupling.
Kinematic Coupling and Task Complexity
Any midtarsal motion captured during Seated Passive MTP Extension is likely due almost entirely to the MTP extension itself (i.e., the windlass mechanism). When sitting in a non-weight bearing position, the plantar intrinsic foot muscles are inactive (33). While it is possible that even passive extension of the toes could evoke a small muscular response, care was taken to ensure investigators felt no active assistance or resistance when pushing the MTP joint into extension. If motion at the midtarsal joint was entirely due to the windlass mechanism during Seated Passive MTP Extension, then the coupling ratio for this condition can be used as a baseline to assess the contribution of the windlass mechanism to the coupling ratio during other tasks. For example, the DFCR for Neutral heel raise was 6 times that of the seated passive condition. Thus, roughly 5/6 of that motion is likely attributable to influences other than the windlass mechanism (e.g. muscle contractions or energy storage and return).
Comparing seated and standing conditions along with active and passive MTP extension provided some insights into the influence of the windlass mechanism. Seated versus Standing had an effect on the coupling ratio but passive versus active did not. For active conditions, we expected the toe extensor muscles to exert a small dorsiflexion moment on the midtarsal joint, thus slightly reducing arch rise. This did not occur, and it appears that active recruitment of the dorsal foot muscles to extend the MTP joint (i.e. extensor hallucis longus and brevis) has little effect on arch rise, either because the moment arms of these muscles are too small or because co-contraction of plantar muscles negated their effect. In contrast, we noted a difference between seated and standing conditions, with the standing DFCR for both passive and active conditions smaller than both the seated DFCRs. Sichting and Ebrecht did not find a difference in the change in navicular height per degree of MTP extension between their seated and standing passive MTP extension conditions (23). Possible explanations for these differences include measurement methodology (i.e. navicular height versus midtarsal angle as a measure of arch height) and potential differences in subject posture. We expect that the reason our standing conditions had smaller DFCR compared to seated may be due to the static loading experienced by the MLA during standing, which flattens the arch (Butler et al., 2008), increasing the distance between the MTP joint and calcaneus. This would either increase the tension in the PA or lengthen it, which could have an inhibiting or enhancing effect on the windlass mechanism (34). The resting length of the PA was not measured, but if it were slack when seated, then weight bearing should have removed this slack and allowed MTP extension to raise the arch more effectively. However, since the DFCR dropped slightly when standing, this is likely not the case. If the PA is already tensioned while seated, weight bearing may increase that tension, but its effect on the windlass mechanism may depend on its extensibility. If it acts as a rigid cable, the additional tension from weight bearing should still raise the arch to the same extent per degree of MTP extension; however, if it is flexible, MTP extension may stretch the aponeurosis instead of raising the arch. Given that the DFCR was smaller during standing, the PA likely stretched in response to weight bearing rather than acting as a rigid cable. Furthermore, the plantar intrinsic foot muscles provide postural support for the feet during standing (33), which may have increased midtarsal stiffness.
Arch rise was much greater during dynamic heel raises compared to the isolated MTP extension conditions, closely matching the results of Sichting and Ebrecht, who also found that arch rise was significantly greater during walking compared to passive MTP extension during sitting and standing (23). The difference between dynamic and isolated MTP extension likely has a large contribution from the plantar intrinsic and extrinsic foot muscles. These muscles are active during the propulsive phase of walking (35, 36) and a likely source of tension across the arch (24). As foot and ankle mechanics are similar between push-off of walking and heel raises (32), these muscles are likely active in a similar manner during heel raises. While we did not see any differences among the three heel raise conditions, this may be due in part to the statistical treatment, where we had numerous pairwise comparisons. Without an adjustment for multiple comparisons, ToeExt had a significantly larger DFCR than Neutral (p = 0.03), which became insignificant after adjustment (p = 0.13). With a larger sample size or more direct comparison between heel raise conditions, this relationship may have been significant. If weight bearing results in some stretch of the PA, starting heel raises with the toes extended should increase this stretch, potentially removing any extensibility and resulting in greater arch rise. As mentioned, when compared to the passive seated and passive standing conditions, only about 1/6 of the arch rise seen during heel raises is likely attributable to the windlass mechanism. Combined with the minimal effect of starting with the toes extended, it is likely that the windlass mechanism plays a secondary role in dynamic arch rise compared to the role of active muscle contractions.
Kinetic Coupling During Heel Raises
Although the kinematic coupling ratio was not statistically significant among the dynamic heel raise conditions, the amount of joint work at the MTP and midtarsal joints was substantially affected by the starting position of the MTP joint in these tasks. Starting with the toes extended resulted in a greater amount of work being absorbed at the MTP joint as well as generated at the midtarsal joint. Conversely, starting with the toes flexed resulted in less work being absorbed at the MTP joint and generated at the midtarsal joint. These results are in support of our hypothesis that the work generated at the midtarsal joint would change proportionally to the work absorbed at the MTP joint during heel raises (as indicated by the consistency of distal to hindfoot work across conditions). This hypothesis was based on previous investigations that found kinetic coupling in walking and running. As walking speed increases, both MTP negative work and midtarsal positive work increase (18). Likewise, when comparing the power profiles of runners with varying foot strikes, Bruening et al. found that forefoot strikers had greater MTP negative work concurrent with greater midtarsal positive work (5).
Considering the results of the current study in conjunction with the work from which we formed our hypothesis, it is evident that there is a functional coupling between the MTP and midtarsal joints that may be even greater in kinetics than in kinematics. Yet, our results suggest that this kinetic coupling is likely due only in small measure to the windlass mechanism. Previous research has noted that during the propulsive phase of walking, the arch quickly rises despite a likely decrease in tension within the PA (34, 37), concurrent with continued MTP extension (38). Biarticular plantar intrinsic and extrinsic foot muscles may play a large role, both actively and passively through spring-like properties. An investigation of the flexor digitorum brevis muscle (a biarticular muscle spanning the plantar aspect of the foot in parallel to the PA) found that during loading of the arch the muscle tendon stretches while the muscle fascicle is active isometrically (25). During heel raises, the intrinsic foot muscles may be a source of power generation at the midtarsal joint and furthermore, stretch of these muscle tendons may facilitate energy storage and return between the MTP and midtarsal joints.
The discrepancy between kinematics and kinetics effects in our study also suggests a large muscular role in distal foot coupling. To better understand the factors contributing to this coupling, we plotted the mean angle, moment, and power vs time graphs (Figure 4) for each condition. For angles, although there was greater midtarsal plantarflexion in ToeExt at the start of heel raises, this did not result in a greater peak angle or angular velocity compared to Neutral (Figure 4A). Instead, the midtarsal plantarflexion moment increased throughout the movement for ToeExt (Figure 4B). This increased moment appears to be the main contributor to the increased peak power and work values in ToeExt (Figure 4C) and is likely due to an anteriorly shifted center of pressure (COP). Shifting the COP in either an anterior or posterior direction should affect both MTP and midtarsal joints fairly equally, with associated proportional increases in active or passive tension at both joints and may be a main factor in this kinetic coupling. In addition, the altered joint moments could also change the muscle force-length positioning, with a more advantageous length in the ToeExt starting position. The ToeFlex condition, in contrast, exhibited less peak midtarsal plantarflexion and slightly lower moment compared to Neutral, perhaps due to being placed in a disadvantageous position. However, more research investigating the extent of energy transfer between these two joints could provide greater insight as it is still unclear if any of the energy absorbed at the MTP joint is transferred to the midtarsal joint through muscle tendons and the PA or simply dissipated.
Arch Flexibility and Arch Height Versus Kinematic Coupling
We based our hypothesis that high arch flexibility and low arches would be related to a less efficient windlass mechanism (i.e., smaller DFCR) on the results from a study done by Lucas and colleagues (27). In partial support of our hypothesis, there was a negative correlation between arch flexibility and DFCR during Seated Active MTP extension, with DFCR decreasing with increasing flexibility. Similarly, there was a negative correlation between arch height and DFCR during Standing Active MTP extension, with DFCR decreasing with increasing arch height. Contrary to our hypothesis, though, there was a positive correlation between arch flexibility and DFCR during both Neutral and ToeFlex heel raises. It is possible that the relationship between arch structure and the windlass mechanism is more prominent in isolated movements; however, the lack of significant correlations across all other isolated tasks suggests something more equivocal. The research exploring the relationship between static structure and dynamic foot function in walking and running is mixed. A number of researchers have found no correlation between foot structure and range of motion during stance (e.g. (39, 40)). Yet, isolated studies such as Magalhães et al. provide some reason for continued research. They found that individuals with greater foot mobility had increased range of motion in both the frontal and sagittal planes at the midfoot joint complex during walking compared to individuals with less foot mobility (41). Thus, the relationship between foot structure and function is nuanced and warrants further investigation. Our results suggest that arch height or arch flexibility alone may not be adequate predictors of dynamic foot function.
A traditional clinical assumption is that high arches are stiff and low arches are flexible (42, 43). However, recent work demonstrates that many arch flexibility types exist within arch height types (31). The current study supports this notion, as we found that 21% of our participants had both stiff and low arches while 10% had both flexible and high arches. To classify the arches of our participants for this tally, we used the classifications of Zifchock et al. (31) for arch flexibility, grouping the ‘very-stiff’ and ‘stiff’ categories into one category called ‘stiff’ (similar grouping was done for the ‘very-flexible’ and ‘flexible’ categories). For arch height, the average of the cut-offs specified by Hillstrom et al. (44) and Williams et al. (45) was used. Perhaps if we recruited individuals that had both stiff and high arches or flexible and low arches, or more clinically extreme foot structures, a stronger correlation between foot structure and DFCR would have been observed. Future studies could explore these specific populations as it may provide useful insight for clinical applications
There are some limitations to our work. First, our measure of arch flexibility calculated using the AHIMS arch height index measurement system may not be an accurate measure of functional arch flexibility as it is calculated from static positions. Future work could explore the relationship between foot function and arch flexibility or stiffness calculated during dynamic movement. Secondly, we did not control for possible anterior-posterior COP differences during our heel raise conditions. We had subjects focus on reaching the same height between conditions, which likely helped with any possible anterior-posterior leaning. However, it is possible that the COP was different between conditions which would affect inverse dynamic calculations and future investigations could control for this possibility