In gymnastics, landings after an acrobatic exercise are

A three-dimensional shank-foot model to determine the foot motion during landings ADAMANTIOS ARAMPATZIS, GERT-PETER BRÜGGEMANN, and GASPAR MOREY KLAPSING German Sport University of Cologne, Institute for Biomechanics, 50933 Cologne, GERMANY ABSTRACT ARAMPATZIS, A., G-P. BRÜGGEMANN, and G. MOREY KLAPSING. A three-dimensional shank-foot model to determine the foot motion during landings. Med. Sci. Sports Exerc., Vol. 34, No. 1, 2002, pp. 130 138. Purpose: The purposes of this study were a) to develop a model of the foot capable of describing the foot motion during dynamic movements and b) to study the influence of different mats on foot motion during landing in gymnastics. Methods: Six female gymnasts (height: 1.63 0.04 m, weight: 58.21 3.46 kg) participated in this study. All six gymnasts carried out barefoot landings, falling from 80 and 115 cm onto three mats each with a different stiffness (hard, medium, and soft). Three synchronized digital high-speed video cameras (250 Hz) captured the motion of the left shank and foot. At the same time, the reaction forces between mat and foot at the forefoot and rearfoot were measured by two instrumented insoles (Paromed, 1000 Hz). The kinematics of the tibiotalar, talonavicular, and calcaneocuboid joints were examined. The lower leg and the foot were modeled by means of a multi-body system, comprising seven rigid bodies. For each joint, two joint coordinate systems attached on each of the connected segments were defined. Results: The mat stiffness did not show any influence on the maximal reaction forces or on the kinematics of the tibiotalar joint. For the soft mat, higher maximal eversion angles at the talonavicular and the calcaneocuboid joints were measured. Conclusions: The relative motion between forefoot and rearfoot was influenced by changing mat stiffness. Therefore, the construction of the mat influenced the motion of the foot. The observation of only the tibiotalar joint is not enough when studying the influence of different mats on foot motion. The functional benefit of the mechanical advantages of a soft mat (higher energy absorption) includes a decrease in stability. The surface of the landing mat should, therefore, be reinforced by a stabilizing mechanism. Key Words: MODELING, REARFOOT KINEMATICS, FOREFOOT KINEMATICS, EVERSION, FOOT STABILITY, MAT DEFORMATION, REACTION FORCES In gymnastics, landings after an acrobatic exercise are common tasks of daily training. During landings, the reaction forces reach values greater than 10 times body weight (14). This is much higher than those occurring while running (2.6 3.7 times body weight when running at velocities between 2.5 and 6.5 m s 1 ) as reported by Arampatzis et al. (1). Statistical data show that, in gymnastics, most (50 70%) of the acute injuries occur at the foot and the tibiotalar and knee joints (35). In the literature, it is often stated that an excessive eversion at the tibiotalar joint could lead to overload and injury at the foot and the knee (9,13,18,38). Many studies from different research groups have dealt with the motion of the tibiotalar joint during walking or running (3,9,20,22,28,30,31,36,37). Only a few observed the relative motion of the midfoot or the forefoot in relation to the rearfoot (11,19,29). More recent studies (33) mention that, for the understanding of orthotic effects, midfoot and forefoot movements may be more important than those of the calcaneus. Model calculations done for running (2) show that the maximal force values at the calcaneocuboid joint rise up to 72% of those reached at the tibiotalar joint. 0195-9131/02/3401-0130/$3.00/0 MEDICINE & SCIENCE IN SPORTS & EXERCISE Copyright 2002 by the American College of Sports Medicine Submitted for publication November 2000. Accepted for publication April 2001. 130 McNitt-Gray et al. (14 17) found that falling height and the landing mat itself influence the landing technique. Nigg (21) states that athletes change their motion to adjust to the different surfaces they interact with. It has also been reported that surfaces showing a high deformation are subjectively rated as more comfortable by the subjects (26). On the other hand, the different stiffness of landing mats has no (14) or just a very little (16) influence on maximal reaction force. However, the same reaction forces can produce different foot loadings when the foot kinematics are changed. Neither the foot motion during landing nor the effect on foot kinematics that the deformation of mats with a different stiffness may produce have been studied for now. The aims of this study were: a) to develop a model of the foot capable of describing the foot motion during dynamic movements and b) to study the influence of different mats on foot motion during landing in gymnastics. METHODS Experimental Protocol Six female gymnasts (height: 1.63 0.04 m, weight: 58.21 3.46 kg), active members of the university gymnastics team, participated in this study. Informed written consent was obtained from all subjects in accordance with the policy statement of the American College of Sports Medicine. All six gymnasts carried out barefoot landings, falling from 80 and 115 cm onto three mats each with a

FIGURE 1 Medial, front, and lateral view of the marker placement to define the shank-foot model. 16. Calcaneus medial (more posterior) 17. Calcaneus lateral 18. Fascias tibiae The video sequences were digitized using the automatic tracking option from the Peak Motus system. When marker reflection was not good enough for automatic tracking, those points were digitized manually. Those trials where a marker was hidden (for example by arm movement), were excluded from the analysis. All three cameras covered a volume of approximately 60 60 60 cm 3. For the calibration of the cameras, a 30 30 30 cm 3 cube was used. different stiffness (hard, medium and soft). The mechanical properties (stiffness and damping) of the mats were similar to those of the mats used in gymnastic training and competition. Two different landing techniques (6) were performed (soft and hard landing). The hard and soft landing differed most in the amplitude of the knee angle. For the hard landing, the subjects were instructed to bend the knees as little as possible. This was subjectively controlled by video. The size of all three mats was 200 200 30 cm 3. Three synchronized digital high-speed video cameras (250 Hz) captured the motion of the left shank and foot. At the same time, the reaction forces between mat and foot at the forefoot and rearfoot were measured by two instrumented insoles (Paromed, 1000 Hz; Paromed Medizintechnik GmbH, Neubeuren, Germany). To synchronize the cameras and the insoles, a common trigger for all four systems was used. Eighteen reflective markers were attached to the left shank and foot (Fig. 1). Eight markers were fixed on predefined anatomical landmarks to allow the definition of the joint coordinate systems. The rest of the markers were placed on not strictly defined anatomical positions but rather on locations where lesser skin movements were expected because of lesser subcutaneous soft tissue and little movement of tendon observed at these locations. Below is a list of the marker locations: Bony landmarks 1. Caput metatarsale I (most medial point) 2. Caput metatarsale V (most lateral point) 3. Tuberositas naviculare (most medial point) 4. Os cuboideum (diagonal superior to the basis of the 5th metatarsal) 5. Malleolus medialis (most medial point) 6. Malleolus lateralis (most lateral point) 7. Condylus medialis tibiae (most medial point) 8. Caput fibula (most lateral point) Other markers 9. Caput metatarsale I (medial superior) 10. Caput metatarsale V (lateral superior) 11. Caput metatarsale II III (between 2nd and 3rd metatarsal heads) 12. Metatarsus I (more proximal) 13. Metatarsus V (more proximal) 14. Os naviculare (on the instep) 15. Calcaneus medial (more anterior) Reference Measurement Foot motion was determined with reference to a neutral position (Fig. 1). For the definition of the neutral position, the left foot and shank from each subject were placed in an inertial reference coordinate system (RCS). A mechanic frame was used for placing the lower leg and the foot in the neutral reference position. Once the position was attained, the frame was retired, and three cameras (50 Hz) registered this position. The RCS had axis 1 pointing forward, axis 2 in mediolateral direction, and axis 3 vertically pointing toward the ground. The plane defined by axis 1 and 2 was parallel to the ground. The neutral position of the shank and foot was defined as follows: the shank was vertical (tip of the lateral malleolus directly below the tip of the caput fibula with no lateral inclination) corresponding to the direction of the third axis of the RCS. The foot was flat on the ground, its longitudinal axis (tuberositas calcanei-caput metatarsale III) parallel to the first axis of the RCS and the medial aspect pointing in the direction of axis 2 from the RCS. All 18 markers were filmed in this neutral position by using three cameras (50 Hz) to calculate their 3D coordinates. The Shank-Foot Model The lower leg and the foot were modeled by means of a multi-body system, comprising seven rigid bodies (Fig. 2). The software used for modeling purposes was the simulation software Alaska (advanced Lagrangian solver in kinetic analysis, version 3.0, Chemnitz). For the elaboration of the model, care was taken to account for the functional anatomy of the foot (19), as most of the existing models (11,29) do not account for the medial and lateral parts of the foot separately. The definition of the model is presented in Table 1. For this study, the talocalcaneal and the two metatarsophalangeal joints were not used. For the whole motion process, the following constraints were imposed: no motion at these three joints was allowed. For each joint, two joint coordinate systems (JCS) attached on each of the connected segments were defined. The joint coordinate systems were defined in the neutral position by using the above mentioned bony landmarks (markers 1 8). Below, the definition of each JCS is explained. A THREE-DIMENSIONAL SHANK-FOOT MODEL Medicine & Science in Sports & Exercise 131

FIGURE 2 Shank-foot model. Tibiotalar Joint (TTJ) JCS1. Its origin was set at the midpoint of the line joining both malleoli. Axis 3 was defined by the line joining the origin with the midpoint between landmarks 7 and 8. Axis 3 pointed downward. Axis 2 was orthogonal to axis 3 and was contained in the plane defined by axis 3 and the line joining the malleoli. Axis 2 pointed toward medial. Axis 1 resulted from the cross-product of axes 3 and 2. JCS2. JCS2 resulted from moving the reference coordinate system to the origin as defined for JCS1. Talonavicular Joint (TNJ) JCS1. JCS1 resulted from moving the reference coordinate system to the origin, which was located at the level of landmark 3, at the midpoint of the 3/5 of the mediolateral distance between tuberositas naviculare and os cuboideum. JCS2. JCS2 had the same origin as JCS1. Axis 2 corresponded to axis 2 of the RCS transported to the origin. Axis 1 was orthogonal to axis 2 and contained in the plane defined by axis 2 and the line joining landmarks 1 and 3. Axis 3 resulted from the cross-product of axes 1 and 2. Calcaneocuboid Joint (CCJ) JCS1. JCS1 resulted from moving the reference coordinate system to the origin, which was located at the level of the os cuboideum, at the midpoint of the 2/5 of the mediolateral distance between Os cuboideum and tuberositas naviculare. JCS2. JCS2 had the same origin as JCS1. Axis 2 corresponded to axis 2 of the RCS transported to the origin. Axis 1 was orthogonal to axis 2 and contained in the plane defined by axis 2 and the line joining landmarks 2 and 4. Axis 3 resulted from the cross-product of axes 1 and 2. Markers 9 18 were fixed onto the corresponding segments of the model, using their 3D coordinates from the reference measurement in neutral position. These markers (markers 9 18) plus both malleoli were used for the dynamic tracking. The remaining six markers were detached before performing the landings. The 12 markers kept were tracked in space during the analysis of the landings. Their 3D coordinates allowed the calculation of the model motion and, hence, the motion at the tibiotalar, talonavicular, and calcaneocuboid joints. Segment 1 contained markers 5, 6, and 18; segment 3 markers 15 17; segment 4 markers 9, 12, and 14; and segment 5 markers 10, 11, and 13. In an attempt to account for the relative motion of the markers caused by movements and deformation of the soft tissues at the foot, the markers were attached to the model by using 3D linear spring damping elements. This method does not harm the constraints of the model. From 10 6 N m 1 on, the spring constant chosen for the tracking gives back stable coordinates without oscillation. Basing on experience, the damping was set 100-fold lower. For this application, the spring constant was set to k 10 6 N m 1 and the damping constant was set to 10 4 N s m 1. Data Analysis The kinematics of the tibiotalar, talonavicular, and calcaneocuboid joints were defined by the orientation of the second joint coordinate system with respect to the first one. Eversion-inversion, dorsi-plantar flexion, and adduction-abduction are defined by the Bryant angles (,, ) as follows: (Eversion-inversion). This was defined as the angle between the 2nd axis of JCS1 (2 JCS1 ) and the nodal line 3 JCS2 1 JCS1. 0 when (2 JCS1,3 JCS2 1 JCS1,1 JCS1 ) form a right-handed system. This motion corresponds to a rotation about the 1st axis of JCS1 (1 JCS1 ). This rotation transfers JCS1 in a new coordinate system (JCS1,2). (Dorsal-plantar flexion). This was defined as the angle between 3 JCS1 and 3 JCS2. 0 when ( 3 JCS1,2, 3 JCS2, 2 JCS1,2 ) form a right-handed system. This motion corresponds to a rotation about 2 JCS1,2. TABLE 1. Definition of the segments and joints of the shank-foot model. Segments Bones Joints Connected Segments Degree of Freedom Segment 1 Tibia and fibula Free joint Space-segment 1 6 (free joint) Segment 2 Talus Tibiotalar joint Segment 1-segment 2 3 (ball-socket joint) Segment 3 Calcaneus Talocalcaneal joint Segment 2-segment 3 1 (rotational joint) Segment 4 Os naviculare, three cuneiformi, and metatarsals I,II,III Talonavicular joint Segment 2-segment 4 3 (ball-socket joint) Segment 5 Cuboid and metatarsals IV,V Calcaneocuboid joint Segment 3-segment 5 3 (ball-socket joint) Segment 6 Phalanges I,II,III Metatarsophalangeal joint 1 Segment 4-segment 6 1 (rotational joint) Segment 7 Phalanges IV,V Metatarsophalangeal joint 2 Segment 5-segment 7 1 (rotational joint) 132 Official Journal of the American College of Sports Medicine http://www.acsm-msse.org

where the deformation was also bigger when compared with the rearfoot (Tables 2 and 3). The kinematics of the tibiotalar, talonavicular, and calcaneocuboid joints are shown in Figure 6. During the landing, the tibiotalar joint everted, dorsiflexed, and abducted (Fig. 6), which means that the foot pronates. For all three joints, the standard deviation for the adduction-abduction was higher than the standard deviation for eversion-inversion or dorsi-plantarflexion. Former studies also reported a higher FIGURE 3 Reaction forces under the left forefoot vs mat deformation for the three mats (N 6, soft landing). (Adduction-abduction). This was defined as the angle between 2 JCS1,2 and 2 JCS2. 0 when (2 JCS1,2,2 JCS2, 3 JCS2 ) form a right-handed system. This motion corresponds to a rotation about 3 JCS2. The mat deformation at the forefoot area was calculated as the mean value of the coordinates of points 9, 10, and 11. In the same manner, the mat deformation in the rearfoot area was calculated as the mean value from points 15, 16, and 17. The mat deformation was determined as the difference between the height of the unloaded mat and the average value of the vertical coordinates of the markers. The differences between the three different mats were checked using a nonparametric test for several dependent samples (Friedman test). At those parameters where differences were found, a nonparametric test for two dependent samples (Wilcoxon test) was applied to assess the differences between the mats. The level of significance was set at P 0.05. RESULTS Figure 3 illustrates the force-deformation characteristics of the three different mats during landings from 80 cm. The graph demonstrates how the different mat constructions correspond to the different stiffness. In Figures 4 and 5, the reaction forces and the deformation of the mats at the forefoot and the rearfoot area are illustrated. The three mats did not significantly influence the maximal values of the reaction forces (Tables 2 and 3). The differences were not statistically significant (P 0.05), neither at the landings from height 1 (80 cm) nor at those from height 2 (115 cm). The maximal force values were changed by means of the two different landing techniques. The hard landing showed significant (P 0.05) higher values for all three mats (Tables 2 and 3). The maximal deformation was influenced as much by the landing technique (soft or hard) as by the mat construction (Tables 2 and 3). The soft mat and the hard landing technique produced higher maximal deformation values. The differences were bigger at the forefoot area FIGURE 4 Force-time curves (mean SD, N 6) for the three mats during soft landings (H1, 80 cm; mat 1, hard; mat 2, medium; mat 3, soft). A THREE-DIMENSIONAL SHANK-FOOT MODEL Medicine & Science in Sports & Exercise 133

FIGURE 5 Mat deformation over time (mean SD, N 6) under the left fore- and rear-foot during soft landing on the three mats (H1, 80 cm; M1, hard; M2, medium; M3, soft). variability for adduction-abduction in running (31) and walking (12). The range of motion at the tibiotalar joint was not influenced by the different mats (Table 4). The maximal eversion at the tibiotalar joint reached values between 3 and 8. No statistical differences could be found for the eversion at the tibiotalar joint when compared with the different landing techniques (Table 4). At the hard landings, the gymnasts had lower maximal dorsiflexion values when landing from height 1 (80 cm, Table 4). Differences in maximal eversion were found at the talonavicular and the calcaneocuboid joints. The soft mat (mat 3) showed statistically significant (P 0.05) higher values (Tables 5 and 6), reaching 13 and 18 at the talonavicular and the calcaneocuboid joints, respectively. The landing technique used had no obvious influence in the range of motion of either of the two above-mentioned joints. When comparing soft and hard landings, no statistically significant (P 0.05) difference could be found either in maximal eversion or maximal dorsiflexion at the talonavicular or the calcaneocuboid joints (Tables 5 and 6). DISCUSSION The maximal reaction forces reached values between 2.7 and 4.2 times body weight for the left foot when falling from 80 and 115 cm. McNitt-Gray et al. (14,16) reported similar TABLE 2. Maximal reaction forces and maximal deformation (mean SD, N 6) at the left foot during landings from height 1 (80 cm). F max sum (N kg 1 ) 28.63 (2.94) 27.14 (2.89) 27.27 (5.59) 37.71* (5.27) 35.04* (5.27) 35.72* (4.23) F max rearfoot (N kg 1 ) 15.89 (2.72) 15.73 (2.72) 14.77 (3.73) 20.85* (4.25) 16.64 1 (3.60) 17.65 1 (3.01) F max forefoot (N kg 1 ) 12.88 (1.66) 11.80 (2.22) 13.05 (2.43) 17.61* (3.77) 17.58* (2.32) 17.72* (2.81) S rearfoot (cm) 7.90 (1.50) 8.57 (0.91) 9.38 1 (1.56) 9.20* (1.19) 9.58* (1.26) 10.83* (1.88) S forefoot (cm) 8.72 (0.92) 9.47 1 (0.89) 11.61 1,2 (2.10) 10.04* (1.67) 11.55 1 (1.72) 13.77*,1,2 (2.10) F max sum, maximal force (sum); F max rearfoot, maximal force (rearfoot); F max forefoot, maximal force (forefoot); S rearfoot, mat deformation at the rearfoot area; S forefoot, mat deformation at the forefoot area. 134 Official Journal of the American College of Sports Medicine http://www.acsm-msse.org

TABLE 3. Maximal reaction forces and maximal deformation (mean SD, N 6) at the left foot during landings from height 2 (115 cm). F max sum (N kg 1 ) 36.66 (4.57) 36.22 (5.56) 35.21 (7.47) 39.82 (6.51) 39.41* (7.02) 42.25* (7.36) F max rearfoot (N kg 1 ) 21.51 (3.34) 19.50 (3.34) 19.06 (4.02) 22.45 (4.69) 21.80* (4.01) 23.10 (4.58) F max forefoot (N kg 1 ) 15.67 (2.70) 17.53 (2.99) 16.61 (3.85) 18.40 (2.17) 18.21 (4.09) 19.84 (2.37) S rearfoot (cm) 9.48 (1.16) 10.22 (1.04) 10.77 (1.45) 9.93 (1.05) 11.25* (1.50) 12.08*,1 (1.12) S forefoot (cm) 10.00 (0.77) 11.12 (0.92) 12.12 1 (2.12) 10.53 (0.83) 12.58 1 (2.01) 13.35*,1 (2.04) F max sum, maximal force (sum); F max rearfoot, maximal force (rearfoot); F max forefoot, maximal force (forefoot); S rearfoot, mat deformation at the rearfoot area; S forefoot, mat deformation at the forefoot area. values for landings from comparable heights. The construction of the mat (hard, medium, soft) had no influence on the maximal force values for either of the two landing techniques. The technique itself (soft or hard landing) did clearly influence the maximal values of the reaction forces. McNitt- Gray et al. (14) also found no differences in the maximal reaction forces when comparing two different stiff mats. A later study of the same group (16) showed little but significant differences in the maximal force values for landings on a soft and a hard mat. Similar and to some extent controversial results are often reported in the literature regarding shoe research. Many studies reported no differences in maximal force values when altering shoe sole stiffness (3,23,25,27,40). Other studies report how changes in surface stiffness (4,10) or shoe sole geometry (28) do influence the maximal force values. The lacking difference in the maximal force values when comparing the three mats with different stiffness can be explained by the ability of the human body to adapt to different surfaces (7,8,21). Nigg (21) reports that athletes adapt to different surfaces changing their motion. Farley et al. (7) and Ferris and Farley (8) found an increased leg stiffness when hopping and running on softer surfaces. The stiffness of the lower extremities and, therefore, the landing technique used have an influence on the maximal force values. This influence was observed in former studies (5,6,24,41) and was also confirmed by this study (Tables 2 and 3). The maximal deformations and the deformation curves for the three mats were different for both soft and hard landing technique (Tables 2 and 3, Fig. 5). The energy absorption by the mat increased steadily with the reduction of the mat stiffness (Fig. 3). A higher energy absorption by the mat at the same falling height is a positive sign when looking at the functionality of landing mats, because while landing, the aim of the athlete is to reduce his kinetic energy to zero. High deforming surfaces can also increase the subjective comfort of the athlete (26). The maximal eversion angles at the tibiotalar joint reach values ranging between 3 and 8. This value could only be compared with those of walking and running literature, because no landing studies referring to eversion values were found. In the walking and running research, values ranging from 2 to 25 are reported (23,28 31,36,39). This high variability can be explained by the individual differences between humans (30,31,33) but also by the different measuring and calculation methods (3,11,20,30,31,34). In our study, we used a 3D model and external anatomical bony FIGURE 6 Eversion-inversion, dorsiflexion-plantarflexion, and adduction-abduction for the three joints (mean SD, N 6) during soft landings from 80 cm of the hard mat; TTJ, tibiotalar joint; TNJ, talonavicular joint; CCJ, calcaneocuboid joint. A THREE-DIMENSIONAL SHANK-FOOT MODEL Medicine & Science in Sports & Exercise 135

TABLE 4. Maximal joint motion (degrees, mean SD, N 6) at the tibiotalar joint of the left foot during landings from height 1 (80 cm) and height 2 (115 cm). max (height 1) 4.62 (1.98) 3.05 (2.70) 4.75 (2.10) 6.52 (3.23) 5.17 (2.92) 5.12 (3.11) max (height 1) 15.77 (6.01) 17.49 (5.67) 10.72 (5.89) 10.41 (5.76) 7.34* (4.65) 3.32* (4.58) Fmax (height 1) 6.86 (2.92) 8.14 (3.11) 7.69 (3.42) 5.82 (2.86) 5.44* (2.76) 4.98 (2.39) max (height 2) 5.29 (3.34) 6.72 (3.19) 5.44 (3.34) 5.73 (2.93) 7.58 (3.24) 6.62 (3.42) max (height 2) 18.23 (5.23) 15.50 (6.34) 13.16 (5.55) 14.39 (6.76) 11.12 (6.04) 12.24 (6.81) Fmax (height 2) 9.01 (3.28) 9.16 (3.45) 7.38 (3.21) 7.59 (3.97) 9.21 (3.14) 5.62 (2.98) max, maximal eversion; max, maximal dorsiflexion; Fmax, adduction-abduction amplitude until force maximum. landmarks. We tried to account for skin movements by attaching the markers to the model by means of linear spring damping elements. The maximal eversion angles obtained with our model were in accordance with those of 3D studies considering only the tibiotalar joint (20,32,36,37) and other studies that modeled the foot as a multiple body system (11,12,29). They were also in the range from values reported by studies using intracortical bone pins (30,31,33). Neither did the mats show an influence on the kinematics of the tibiotalar joint (Table 4) nor was the maximal eversion influenced by changing the landing technique. Similar results are reported in papers studying the effect of shoe soles with different stiffness (22,23,37). Nigg et al. (22) found that changes in the material properties had only a minimal effect on foot eversion. Despite the fact that at the landings the maximal forces are clearly higher, the maximal eversion angles are similar to those for walking or running (12,20,31,36). The maximal reaction force values at running velocities from 2.5 to 4.5 m s 1 are approximately 2.4 3.0 times body weight (1). For the studied (two legged) landings, the maximal reaction forces were up to 4.2 times body weight on only the left foot (Tables 2 and 3). These results show that the motion at the tibiotalar joint is a steady motion and remains constant despite of considerable changes in mat deformation and reaction forces (Tables 2 and 3). As expected, the maximal dorsiflexion at falling height 1 (80 cm) is higher for the soft landing than for the hard one. At falling height 2 (115 cm), no significant difference was found between soft and hard landings (Table 4). This can be due to the fact that changes in landing technique (soft, hard) are mainly controlled by changes in knee and hip flexion (5,7,41). During the landing, the talonavicular and the calcaneocuboid joints evert and dorsiflex with respect to the rearfoot. The eversion at the calcaneocuboid joint is higher (Fig. 5). The range of motion at the talonavicular joint reaches values from 5 to 13. At the calcaneocuboid joint, these values range between 8 and 18 (Tables 5 and 6). The forefoot motion with respect to the rearfoot behaves in a similar way, as reported by Rattanaprasert et al. (29) in a study about walking with healthy subjects. They (29) state that the forefoot moves with respect to the rearfoot starting from an inverted position at the beginning of the contact phase toward an eversion until 60% of the contact time. However, the maximal eversion angles are clearly lower during walking (29) when compared with landing. The three different mats influence the eversion in both the talonavicular and the calcaneocuboid joints. For mat 3 (soft), the one having the greatest deformation, significant higher eversion values were found compared with mat 1 (hard) and mat 2 (medium). It seems that the depth of deformation and therefore also the deformation hollow, which for the soft mat is mainly located at the forefoot area (Figure 5, Tables 2 and 3), were responsible for the higher eversion of the forefoot with respect to the rearfoot. The magnitude of the load on inner structures caused by these changes in the eversion at the talonavicular and the calcaneocuboid joints cannot be calculated by means of this version of the model. Anyway, it is likely that the range of motion of the whole foot was influenced and that this could lead to overloads. By means of model calculations (2), the maximal forces at the calcaneocuboid joint during running reach values up to 8 times body weight. Correspondingly, at the tibiotalar joint, we get forces up to 11 times body weight (2). This illustrates that TABLE 5. Maximal joint motion (degrees, mean SD, N 6) at the talonavicular joint of the left foot during landings from height 1 (80 cm) and height 2 (115 cm). max (height 1) 5.05 (3.21) 6.99 (3.13) 12.67 1,2 (4.11) 4.54 (3.11) 8.90 (3.55) 12.82 1 (3.93) max (height 1) 6.14 (2.52) 2.88 (2.21) 4.10 (1.42) 5.92 (2.12) 4.36 (2.17) 2.28 (2.01) Fmax (height 1) 3.11 (1.99) 4.05 (2.01) 6.33 (1.99) 5.85 (3.01) 5.67 (1.98) 4.87 (2.12) max (height 2) 7.33 (3.34) 7.54 (2.98) 13.41 1,2 (3.63) 8.46 (3.06) 6.82 (3.27) 11.42 1,2 (2.18) max (height 2) 5.09 (2.38) 5.31 (2.70) 4.63 (2.47) 5.60 (2.37) 5.09 (2.27) 3.77 (2.01) Fmax (height 2) 5.17 (2.62) 5.49 (3.22) 3.90 (2.95) 4.32 (2.68) 4.46 (2.39) 5.42 (3.08) max, maximal eversion; max, maximal dorsiflexion; Fmax, adduction-abduction amplitude until force maximum. 136 Official Journal of the American College of Sports Medicine http://www.acsm-msse.org

TABLE 6. Maximal joint motion (degrees, mean SD, N 6) at the calcaneocuboid joint of the left foot during landings from height 1 (80 cm) and height 2 (115 cm). max (height 1) 8.52 (3.18) 9.56 (2.65) 12.04 1 (2.36) 8.36 (3.11) 12.06 (3.55) 17.52 1,2 (3.96) max (height 1) 9.13 (3.30) 6.70 (3.53) 9.11 (1.42) 8.93 (1.54) 9.24 (3.22) 9.42 (3.04) Fmax (height 1) 1.94 (1.03) 2.90 (1.07) 1.56 (0.93) 3.34 (1.23) 3.15 (1.49) 2.01 (1.01) max (height 2) 9.74 (1.76) 11.17 (3.28) 15.72 1,2 (3.54) 10.60 (3.23) 12.35 (3.57) 15.07 1 (2.99) max (height 2) 8.22 (2.67) 8.67 (1.76) 10.17 (2.56) 9.66 (2.65) 9.47 (3.29) 7.84 (2.96) Fmax (height 2) 2.32 (1.09) 3.33 (1.50) 1.47 (0.89) 2.40 (1.73) 3.41 (1.80) 2.57 (1.04) max, maximal eversion; max, maximal dorsiflexion; Fmax, adduction-abduction amplitude until force maximum. the load on the midfoot is high and sufficient to lead to injury. From this point of view, it seems that the motion at the mid- and fore-foot is an important issue for the collision between athlete and mat. Maybe its observation can lead to a better understanding of foot injury mechanisms (35). In conclusion, mats with a different stiffness did not significantly alter the magnitude of the peak reaction forces. The deformation and therefore also the energy absorption by the mat was influenced by the material properties of the mat. Soft mats showed higher deformations and absorbed more energy. Also, mat construction did not influence the kinematic of the tibiotalar joint. The observation of the tibiotalar joint alone was not enough for determining the effect of different mats on the foot motion. The motion at the forefoot in relation to the rearfoot was influenced by mats with a REFERENCES 1. ARAMPATZIS, A., G.-P. BRÜGGEMANN, and V. METZLER. The effect of speed on leg stiffness and joint kinetics in human running. J. Biomech. 32:1349 1353, 1999. 2. CIDDINGS, V. L., G. S. BEAUPRÉ, R. T. WHALEN, and D. R. CAPTER. Calcaneal loading during walking and running. Med. Sci. Sports Exerc. 32:627 634, 2000. 3. COLE,G.K,B.M.NIGG,G.H.FICK, and M. M. MORLOCK. Internal loading of the foot and ankle during impact in running. J. Appl. Biomech. 11:25 46, 1995. 4. DE WITT, B., D. DE CLERCQ, and M. LENOIR. The effect of varying midsole hardness on impact forces and foot motion during foot contact in running. J. Appl. Biomech. 11:395 406, 1995. 5. DEVITA, P., and W. A. SKELLY. Effect of landing stiffness on joint kinetics and energetics in the lower extremity. Med. Sci. Sports Exerc. 24:108 115, 1992. 6. DUFEK, J. S., and B. T. BATES. The evaluation and prediction of impact forces during landings. Med. Sci. Sports Exerc. 22:370 377, 1990. 7. FARLEY, C. T., R. BLICKHAN, J. SAITO, and C. R. TAYLOR. Hopping frequency in humans: a test of how springs set stride frequency in bouncing gaits. J. Appl. Physiol. 71:2127 2132, 1991. 8. FERRIS, D. P., and C. T. FARLEY. Interaction of leg stiffness and surface stiffness during human hopping. J. Appl. Physiol. 82:15 22, 1997. 9. HINTERMANN, B., and B. M. NIGG. Pronation in runners implications for injuries. Sports Med. 26:169 176, 1998. 10. LAFORTUNE, M. A., E. M. HENNIG, and M. J. LAKE. Dominant role of interface over knee angle for cushioning impact loading and regulating initial leg stiffness. J. Biomech. 29:1523 1529, 1996. 11. LEARDINI, A., M. G. BENEDETTI, F. CATANI, L. SIMONCINI, and S. GIANNINI. An anatomically based protocol for the description of foot segment kinematics during gait. Clinical Biomech. 14:528 536, 1999. 12. LIU, W,S.SIEGLER, H.HILLSTROM, and K. WHITNEY. Three-dimensional, six-degrees-of-freedom kinematics of the human hindfoot different stiffness. The maximal eversion angles at the talonavicular and the calcaneocuboid joints were significantly greater for soft mats. In this manner, the mat construction influenced the motion and in consequence also the stability of the forefoot. In light of these results, it seems necessary to combine the mechanical advantages of a softer mat (high energy-absorption capability) with a better functionality of the biological system (athlete). This could be achieved by implementing a stabilizing mechanism at the surface of the mat. The mat should be able to provide an orthotic effect for the mediolateral motion of the forefoot. Address for correspondence: Dr. Adamantios Arampatzis, German Sport University of Cologne, Institute for Biomechanics, Carl Diem Weg 6, 50933 Cologne, Germany; E-mail: Arampatzis@hrz.dshs-koeln.de. during the stance phase of level walking. Hum. Mov. Sci. 16:283 298, 1997. 13. MCCLAY, I., and K. MANAL. Coupling parameters in runners with normal and excessive pronation. J. Appl. Biomech. 13:109 124, 1997. 14. McNITT-GRAY, J. L. Kinematics and impulse characteristics of drop landings from three heights. Int. J. Sports Biomech. 7:201 204, 1991. 15. McNITT-GRAY, J. L. Kinetics of the lower extremities during drop landings from three heights. J. Biomech. 26:1037 1046, 1993. 16. McNITT-GRAY, J. L., T. YOKOI, and C. MILLWARD. Landing strategies used by gymnasts on different surfaces. J. Appl. Biomech. 10:237 252, 1994. 17. McNITT-GRAY, J. L., T. YOKOI, and C. MILLWARD. Landing strategy adjustments made by female gymnasts in response to drop height and mat composition. J. Appl. Biomech. 9:173 190, 1993. 18. MESSIER, S. P., and K. A. PITTALA. Etiologic factors associated with selected running injuries. Med. Sci. Sports Exerc. 20:501 505, 1988. 19. MORLOCK, M., and B. M. NIGG. Theoretical considerations and practical results on the influence of the representation of the foot for the estimation of internal forces with models. Clin. Biomech. 6:3 13, 1991. 20. MOSELEY, L., R. SMITH, A.HUNT, and R. GANT. Three-dimensional kinematics of the rearfoot during the stance phase of walking in normal young adult males. Clin. Biomech. 11:39 45, 1996. 21. NIGG, B. M. The validity and relevance of tests used for the assessment of sports surfaces. Med. Sci. Sports Exerc. 22:131 139, 1990. 22. NIGG, B. M., A. KHAN, V. FISHER, and D. STEFANYSHYN. Effect of shoe insert construction on foot and leg movement. Med. Sci. Sports Exerc. 30:550 555, 1998. 23. NIGG, B. M., and M. MORLOCK. The influence of lateral heel flare of running shoes on pronation and impact forces. Med. Sci. Sports Exerc. 19:294 302, 1987. A THREE-DIMENSIONAL SHANK-FOOT MODEL Medicine & Science in Sports & Exercise 137

24. NIGG, B. M., and W. LIU. The effect of muscle stiffness and damping on simulated impact force peaks during running. J. Biomech. 32:849 856, 1999. 25. NIGG, B. M., H. A. BAHLSEN, S.M.LUETHI, and S. STOKES. The influence of running velocity and midsole hardness on external impact forces in heel-toe running. J. Biomech. 20:951 959, 1987. 26. NIGG, B. M., M. R. YEADON, and W. HERZOG. The influence of construction strategies of sprung surfaces on deformation during vertical jumps. Med. Sci. Sports Exerc. 20:396 402, 1988. 27. NIGG, B. M., W. HERZOG, and L. J. READ. Effect of viscoelastic shoe insoles on vertical impact forces in heel-toe running. Am. J. Sports Med. 16:70 76, 1988. 28. PERRY, S. D., and M. A. LAFORTUNE. Influences of inversion/ eversion of the foot upon impact loading during locomotion. Clin. Biomech. 10:253 57, 1995. 29. RATTANAPRASERT, U., R. SMITH, M. SULLIVAN, and W. GILLEARD. Three-dimensional kinematics of the forefoot, rearfoot, and leg without the function of tibialis posterior in comparison with normals during stance phase of walking. Clin. Biomech. 14:14 23, 1999. 30. REINSCHMIDT, C., A. J. VAN DEN BOGERT,A.LUNDBERG,B.M.NIGG, N. MURPHY, A. STACOFF, and A. STANO. Tibiofemoral and tibiocalcaneal motion during walking: external vs. skeletal markers. Gait Posture 6:98 109, 1997. 31. REINSCHMIDT, C., A. J. VAN DEN BOGERT,N.MURPHY,A.LUNDBERG, and B. M. NIGG. Tibiocalcaneal motion during running, measured with external and bone markers. Clin. Biomech. 12:8 16, 1997. 32. SIEGEL, K. L., T. M. KEPPLE, P. G. O CONNELL, L. H. GERBER, and S. J. STANHOPE. A technique to evaluate foot function during the stance phase of gait. Foot Ankle Int. 16:764 770, 1995. 33. STACOFF, A., C. REINSCHMIDT,B.M.NIGG, et al. Effects of foot orthoses on skeletal motion during running. Clin. Biomech. 15:54 64, 2000. 34. STACOFF, A., C. REINSCHMIDT, and E. STÜSSI. The movement of the heel within a running shoe. Med. Sci. Sports Exerc. 24:695 701, 1992. 35. THEISS, P., H. LOHRER, and W. WILLAM. Talentiert was tun? Chancen und Risiken des Frauenkunstturnens (Talented what to do? Chances and risks in female gymnastics). Leistungssport 4:5 12, 1999. 36. VAN GHELUWE, B., D. KERWIN, P. ROOSEN, and R. TIELEMANS. The influence of heel fit on rearfoot motion in running shoes. J. Appl. Biomech. 15:361 372, 1999. 37. VAN GHELUWE, B., R. TIELEMANS, and P. ROOSEN. The influence of heel counter rigidity on rearfoot motion during running. J. Appl. Biomech. 11:47 67, 1995. 38. VAN MECHELEN, W. Running injuries: a review of the epidemiological literature. Sports Med. 14:320 335, 1992. 39. VAN WOENSEL, W., and P. R. CAVANAGH. A perturbation study of lower extremity motion during running. Int. J. Sports Biomech. 8:30 47, 1992. 40. WRIGHT, I. C., R. R. NEPTUNE, A.J.VAN DEN BOGERT, and B. M. NIGG. Passive regulation of impact forces in heel-toe running. Clin. Biomech. 13:521 531, 1998. 41. ZHANG, S. N., B. T. BATES, and J. S. DUFEK. Contributions of lower extremity joints to energy dissipation during landings. Med. Sci. Sports Exerc. 32:812 819, 2000. 138 Official Journal of the American College of Sports Medicine http://www.acsm-msse.org