The effect of foot structure on the three-dimensional kinematic coupling behavior of the leg and rear foot.

Key Words: Foot alignment, Foot structure, Kinematic coupling, Radiographic classification, Rear-foot kinematics, Three-dimensional analysis.

Evidence suggests that a causal relationship exists between structural abnormalities of the foot and alterations in lower-extremity kinematics that may predispose an individual to characteristic musculoskeletal injuries.[1-8] Investigations of movement dysfunction related to the foot frequently focus on the subtalar joint (STJ) complex because of its role in force attenuation, as well as in the transfer of axial rotation of the leg to proration and supination of the foot during the support phase of gait.[3,9] Conversely, movements of proration and supination of the foot impart rotations to segments both proximal and distal to the STJ.[10,11] The magnitude of these rotations is determined primarily by the orientation or position of the STJ axis, the shape of the articulating surfaces, and the soft tissue and ligamentous constraints of the joint.[12-14]

The orientation of the axis relates to the function of the foot and the movement patterns that predominate in the lower extremity.[3,8,13,15-18] For a reference frame fixed in the foot, an STJ axis with a steeper inclination to the transverse plane purportedly results in a greater proportion of abduction and adduction motion of the foot when compared with other available motions.[6,13,16] Conversely, a low-inclined STJ axis would permit a greater proportion of inversion and eversion of the foot.[6,13,16] Variations in the orientation of the axis affect the torque and resultant STJ motion during standing, walking, and running[19] and may serve as the basis for characteristic musculoskeletal-related injuries of the lower extremity.

The estimation of STJ axis orientation and its relationship to interdependent rotations of the foot and lower extremity have presented challenges to investigators. Findings from both cadaver and two-dimensional and three-dimensional (3D) motion analysis studies have shown that the STJ axis changes its orientation throughout the stance phase of gait with proration and supination of the foot.[14,20] Variability has also been reported in the static orientation of the STJ axis in the sagittal, transverse, and frontal body planes in both human and cadaver models,[3,14,20-22] and has been largely attributed to the anthropometric differences in foot structure. Although the physical differences in foot structure have been recognized to contribute to the variability in findings among previous investigations, few researchers have categorized foot structures when studying kinematic behavior of the lower extremity.

One measurement that has been used to classify foot structure is the height of the medial longitudinal arch.[3,4,6,7,23-25] Both low- and high-arch foot structures have been linked to the occurrence of distinct injury patterns of the foot and lower extremity. In one study of 295 army recruits, Giladi and colleagues[26] found a higher incidence of stress fractures among soldiers with high arches (40%) than among soldiers with low arches (10%). In a follow-up study, Simkin et al[7] found that individuals with high arches had relatively more femoral and tibial stress injuries, whereas persons with low arches had more metatarsal injuries. In a similar study of US Army recruits, Cowan et al[1] found increased lower-extremity injuries in soldiers with high arches than in soldiers with low arches.

Although arch height is commonly used to provide an indirect estimate of STJ axis orientation and associated movement dysfunction, kinematic investigations that have studied the relationship between arch height and frontal-plane rotations of the rear foot have been inconclusive.[3,24,27,28] Biological variations in the shape of the bony and soft tissues of the foot frequently present difficulties with accurate identification and palpation of medial longitudinal arch landmarks and have raised concerns regarding the reliability and validity of this measurement as a predictor of foot function.[25,29] Furthermore, Nigg and colleagues[3] suggest that assessment of frontal-plane rotations alone may not be the best indicator of foot and lower-extremity function. These investigators proposed that the "coupling" between inversion and eversion of the foot and axial rotation of the leg may play an important role in predicting activity-related lower-extremity injuries. They described a "transfer coefficient," which is the ratio of the maximum amount of foot eversion to the maximum amount of medial rotation of the leg in runners with different medial longitudinal arch structures. Their findings supported the assumption that foot eversion is transferred or coupled to medial rotation of the tibia during the weight-acceptance portion of stance phase. The transfer coefficient was found to increase with increasing arch height. Only 27% of the variance in the transfer coefficient, however, was explained by arch height, suggesting that arch height alone may not be the best predictor for identifying a general foot type at risk of injury.[3] The height of the medial longitudinal arch has been one of the main criteria for defining foot structure. The utility of this measurement for predicting function has not been strongly supported by kinematic investigations.

Other measurements have been used to indicate positional relationships of the forefoot, rear foot, and lower leg in both non-weight-bearing and weight-bearing conditions. Many of these measurements have resulted in poor to moderate interrater and intrarater estimates of reliability.[30-33] In addition to the reliability issue, there is scant evidence to support the relationship between static measurements and the function of the foot during gait.[3,27,28] These studies suggest a need for more reliable and accurate means to characterize foot structure.

Radiographic evaluation of the foot has been considered by some investigators[25,34-40] to be a reliable technique for quantifying the structural configuration of the foot. Radiographic views of the foot in both the anteroposterior (AP) and lateral projections provide a means of identifying the relationship between the talus, calcaneus, navicular, and metatarsals in both transverse and sagittal planes, respectively. Weight-bearing views in the natural toe-out angle and base of support used during gait have been recommended to simulate the manner in which the foot functions in stance.[34,36,41] Despite the high reliability of radiographically obtained measurements for identifying foot structures, there are few reports relating radiographically classified foot types to lower-extremity kinematics.

The purpose of our investigation was to determine the effect of foot structure on the 3D kinematic behavior of the leg and rear foot during running in individuals with radiographically distinct foot types. Our focus was on the stance phase analysis of tibial medial rotation and lateral rotation, calcaneal eversion and inversion, and the coupled relationship between tibial and rear-foot rotations of the combined STJ and talocalcaneal joints.

We hypothesized that two structurally distinct foot groups would demonstrate characteristic rotational patterns that favored tibial medial rotation and lateral rotation for a high rear-foot group and calcaneal eversion and inversion for a low rear-foot group, thereby providing an indirect estimation of the orientation of the STJ axis. We also hypothesized that the differences in angular rotations would be reflected in the coupling relationship between tibial axial rotation and calcaneal eversion and inversion during the stance phase of running gait. The rotation patterns unique to each foot group may provide a better understanding of injury mechanisms associated with foot structures similar to those studied in this investigation. This information may also serve as the basis for management strategies directed at controlling rotations in these foot structures.

Method

Subjects

Twenty recreational runners from a larger subject pool of 29 runners participated in this study. These individuals sought professional consultation from orthopedic specialists, podiatrists, or physical therapists for activity-related musculoskeletal pain in the lower limbs of at least 1 month's duration. The preliminary criteria for participation in the study were based on the subject's history and the clinical presentation of either a pes planus or pes cavus foot structure. Final determination for participation in the study was based on results of standard lateral and AP radiographic measurements of the foot described by Brand and Coleman.[34] Criteria for exclusion from the study included a history of ankle fracture, neuromuscular disease, or inability to meet radiographic measurement criteria. All participants read and signed an informed consent document approved by The University of Iowa Human Subjects Review Board and the Radiation Protection Committee.

Predetermined radiographic measurements of the lateral calcaneal inclination, lateral talometatarsal, and AP talometatarsal angles were used to classify subjects into either a low rear-foot profile group or a high rear-foot profile group (Tab. 1). Selection of these angles for categorization of the subjects into either group was based on reported values of normal and abnormal radiographic parameters[35,36,42-44] and results of a pilot reliability investigation of lateral and AP measurements.[45]

The radiographic criteria for classifying high and low rear-foot groups are indicated in Table 1. Fifteen of the 20 subjects met the criteria for all three measurements of the respective foot groups. The remaining 5 subjects met the criteria for two of the measurements and were less than 2 degrees away from the required value for the third measurement. Because this value was less than the expected deviation for between-day radiographic measurements, we considered these subjects appropriate candidates to continue the study.

Table 1.
Radiographic Measurement Criteria and Group Findings
for Lateral and Anteroposterior Measurements

Radiographic                   Radiographic
Angles                           Criteria

                           Low Rear         High Rear
                            Foot             Foot

Lateral calcaneal       [is less than     [is greater than
 inclination            or equal to]      or equal to
 [Sigma]                20 [degrees]      25 [degrees]

Lateral                 [is less than     [is greater than
 talometatarsal         or equal to]      or equal to]
 [Micro]                -4 [degrees]      0 [degree]

Anteroposterior         [is less than     [is greater than
 talometatarsal         or equal to]      or equal to]
 [Epsilon]              -2 [degrees]      +2 [degrees]

Radiographic                   Group
Angles                  Measurement ([degrees])             P

                       Low Rear        High Rear
                         Foot            Foot

                    [bar] X    SD    [bar] X    SD

Lateral calcaneal    16.6     3.1     31.3     4.1     .001([Alpha])
 inclination
 [Sigma]

Lateral              -5.8     2.4      6.0     5.0     .001([Alpha])
talometatarsal
 [Micro]

Anteroposterior      -2.3      6.3     5.2     4.4     .006([Alpha])
 talometatarsal
 [Epsilon]

([Alpha]) Significant group differences.

Ten subjects were identified for each foot group. The low rear-foot group consisted of 6 male and 4 female subjects with a mean age of 28.6 years (SD = 9.1, range = 18-48), a mean weight of 67.6 kg (SD = 13.5, range = 47-85.5), and a mean height of 174.5 cm (SD = 11.4, range = 158-191). The high rear-foot group consisted of 5 male and 5 female subjects with a mean age of 31.8 years (SD = 8.9, range = 17-48), a mean weight of 69.2 kg (SD = 9.1, range = 56-85.5), and a mean height of 175.0 cm (SD = 8.7, range = 157-188). Group comparisons for radiographic data are presented in Table 1. Significant between-group differences were demonstrated for all radiographic measurements.

Pre-experimental Protocol

Approximately 4 weeks prior to the test session, subjects were issued a pair of TEVA [TM] running sandals(*) and were instructed to progressively increase wearing time of the sandals. These commercially available sandals were designed to provide shock absorption for low-mileage running activities. All sandals had the same shock absorption and pain control characteristics. The midsole and heel height was 2.5 cm. The use of footwear with the same constructional features for all subjects minimized the potential effect of shoe design on movement. Sandals, rather than running shoes, were selected to allow for direct visualization of rear-foot motion during running via skin markers placed over the bony prominences of the rear foot and to minimize the restriction of motion by shoewear.

Subjects who had no previous experience with treadmill running were given a minimum of two training sessions on the treadmill prior to the data collection sessions. They practiced until they accommodated to running at self-selected speeds that were comparable to their recreational running pace. The subjects were pain-free during running at the time of testing.

Instrumentation

While the subjects ran, data were collected using three Panasonic 450AG high-shutter-speed, high-resolution cameras[dagger] operating at a rate of 60 frames per second and a shutter speed of 1/500 second. The Peak Performance Motion Analysis System [double dagger] was used to generate off-line coordinate data. Data transformation and smoothing and kinematic computations were carried out using customized software programs.

We used the direct linear transformation (DLT) method to obtain the 3D coordinates of each target marker during the running trials.[46] This method requires a minimum of two cameras to record the event and a calibration structure or control object with known 3D coordinates. The calibration structure contained 40 known targets and was used to set the DLT parameters for coordinate information in an effective calibration space (78 x 55 x 55 cm).

The Metrecom Skeletal Analysis System [TM][sections][6] was used to acquire coordinate information from selected anatomical landmarks and target markers on the leg and rear foot.[47,48] This six degree-of-freedom digitizer was used to directly digitize, via a hand-held probe, selected anatomical landmarks and six target markers on the leg and foot segments. From these anatomical and target marker data, separate anatomical and target marker Cartesian coordinate systems were generated for each of the leg and rear-foot segments.[49] A transformation matrix, which related the orientation of the target marker coordinate system to the anatomical coordinate system, was later used to transform the coordinate data generated from the target markers recorded during the running trials to clinically relevant anatomical coordinate systems, as defined by the anatomical landmark data.[3,50-52] Construction and transformation of the coordinate systems in this manner allowed for description of the rotation of the tibia and lower leg relative to the rear foot about axes defining dorsiflexion and planter flexion, abduction and adduction, and medial and lateral rotation.[49] In the weight-bearing foot, tibial abduction and adduction may be considered equivalent to calcaneal eversion and inversion, the frontal-plane component of STJ proration and supination, respectively. For the purposes of this report, tibial abduction and adduction will be described by the more familiar clinical rotations of calcaneal eversion and inversion. The anatomical coordinate systems and axes orientation of leg and rear-foot segments are illustrated in Figure 1.

[Figure 1 ILLUSTRATION OMITTED]

Experimental Procedure

Kinematics were determined from six spherical 13-mm-diameter markers attached to the bony landmarks at the tibial tuberosity, the fibula head, and the lateral malleolus of the leg. The three rear-foot markers were placed on the skin overlying the posterosuperior calcaneus, the posteroinferior calcaneus, and the lateral calcaneus (Fig. 1). Prior to the test session, the calibration structure was filmed and a laboratory coordinate system was established such that the X axis was directed mediolaterally, the Y axis was directed anteroposteriorally, and the Z axis was directed vertically. Three-dimensional coordinate data were also collected from selected anatomical and target landmarks on the leg and foot segments using the Metrecom Skeletal Analysis System [TM]. The anatomical and target landmarks were digitized with the subject standing with a comfortable base of support and toe-out angle. The angles between the body segments in this position were defined as the standing neutral position.

Subjects were given a 5-minute (minimum) warm-up period to reach their self-selected running speeds on the treadmill. Data were then collected over a 2-minute running period. A light-emitting diode, visible to all cameras, was used to synchronize the film record for subsequent digitization. The stance phase of three successive steps of the tested lower extremity was selected from the middle of the 2-minute running period and used in subsequent analysis.

Data Reduction

The target markers were placed over bony landmarks of the lower leg and rear foot to minimize the error associated with underlying soft tissue movement. The orthogonal coordinate systems constructed from the target markers were not necessarily aligned with the anatomical axes of the leg and rear foot, thus limiting meaningful clinical interpretation of relative segmental rotations. To describe the rotations about clinically relevant axes, anatomical coordinate systems were generated for each of the leg (tibia) and foot segments, as previously described. Two 3 x 3 transformation matrices ([T.sub.t] and [T.sub.f]), which related the assumed constant orientation of the target marker coordinate system to the anatomical coordinate system for the tibia and foot segments, respectively, were then determined. These matrices were used to transform the target marker coordinate system to the anatomical coordinate system of each segment for every frame of the digitized running trial.

Two additional transformation matrices ([R.sub.f]/L and [R.sub.t]/L) were determined at each sample of the stance period and were based on the target marker data measured in the laboratory coordinate system. These matrices relate the foot and tibia target marker coordinate systems to the laboratory coordinate system (L), respectively. The final computation ([R.sub.t/f]) to determine the anatomical orientation of the tibia with respect to the foot was then determined and can be summarized in the following equation:

(1) [R.sub.t]/f = [([R.sub.t]/L) x [T.sub.f][sup -1] x ([R.sub.t]/L) x [T.sub.t]

In order to extract angular information of the tibia with respect to the rear foot from the transformation matrix ([R.sub.t]/f), a Cardan angle system of three ordered rotations (Z-Y-X) was used to define the relative orientation of the two body segments.[51,52] The Z-Y-X convention of this order-dependent rotation system can be described by [Alpha], [Beta], and [Gamma], indicating medial and lateral, abduction and adduction, and dorsiflexion and plantar-flexion rotations of the leg with respect to the rear foot, respectively.

(2) R = [R [Alpha]] [R [Beta]] [R [Gamma]]

Marker data were digitally filtered using a fourth-order, zero phase shift Butterworth filter with a cutoff frequency of 7 Hz.[53]

Data Analysis

The assumptions underlying our investigation were that the movements of dorsiflexion and planter flexion, eversion and inversion, and medial and lateral rotation do not occur exclusively at one joint.[22,54-56] Rather, these movements result from combined rotations at the talocrural joint and the STJ. Because the talocrural joint makes a greater contribution to dorsiflexion and planter flexion than the STJ does,[22,54] this rotational component was not a primary consideration in this investigation. The focus of the analysis was directed to the primary STJ component rotations of eversion and inversion and medial and lateral rotation.

Means and standard deviations were obtained for the kinematic variables describing [Alpha] and [Beta] rotations about [Z.sub.t] and [Y.sub.t], respectively. Rotations about [Z.sub.t] ([Alpha]) defined tibial medial and lateral rotation relative to the rear foot, and rotations about [Y.sub.t] ([Beta]) defined calcaneal eversion and inversion.

Individual axis rotations about [Z.sub.t] and [Y.sub.t] were assessed for two different phases of the stance period. Peak-to-peak rotations were examined over the entire stance period generating peak-to-peak values as well as for the period of stance from heel contact to maximum eversion and maximum medial rotation. This period from heel contact to maximum eversion (HCEV) and medial rotation (HCMR) has been defined by previous investigators[57] as the period of maximum pronation.

A coupling ratio was used to describe the proportional relationship between the axial rotation of the tibia and the frontal-plane rotation of the foot. This ratio has been used to provide additional information about the interdependent kinematic behavior of the rear foot and leg[3] that could be related to injury patterns in the lower extremity. This ratio was also defined for the different phases of the stance period. Table 2 summarizes the kinematic variables studied in this investigation.

Table 2.
Description of Kinematic Variables

Kinematic Variable      Definition(a)

[Alpha]                 Medial and lateral rotation of leg (tibia/
                        fibula) with respect to the rear foot,
                        determined as a rotation about the axis
                        [Z.sub.dagger]. A laterally rotated position
                        of the leg is defined as negative; a
                        medially rotated position of the leg is
                        defined as positive.

[[Alpha].sub.p-p]       Total medial and lateral range of motion,
                        measured as the difference between peak
                        lateral rotation and peak medial rotation
                        of the leg with respect to the rear foot
                        during stance phase.

[[Alpha].sub.HCMR]      Medial rotation of leg with respect to the
                        rear foot, determined from heel contact to
                        peak medial rotation.

[Beta]                  Abduction and adduction of leg with respect
                        to the foot, determined as a rotation about
                        the axis [Y.sub.dagger]. For the purposes of
                        this article, an abducted position of the
                        leg is defined as positive calcaneal
                        eversion; an adducted position of the leg is
                        defined as negative calcaneal inversion.

[[Beta].sub.p-p]        Total eversion and inversion range of
                        motion, measured as the difference between
                        peak eversion and peak inversion of the leg
                        with respect to the rear foot during stance
                        phase.

[[Beta].sub.HCEV]       Eversion range of motion of leg with respect
                        to the rear foot, determined from heel
                        contact to peak eversion.

[[Beta].sub.HCEV]/      Coupling ratio defining the relationship
[[Alpha].sub.HCMR]      between calcaneal eversion and inversion and
                        leg medial and lateral rotation from heel
                        contact to peak eversion and medial
                        rotation.

(a) All units are degrees.

Independent sample t tests were used to evaluate group differences for the kinematic variables. Prior to analysis of the kinematic data, the stance phase was normalized in time and the data were expressed as percentages of normalized stance. For each kinematic variable, the mean of three trials was used in the analysis. The significance was set at P [is less than] .05. Statistical procedures were carried out using the Statistical Analysis System.[parallel]

Results

Selection of the radiographic angles for categorization of the subjects into either group was based on reported values of normal and abnormal radiographic variables[35,36,42-44] and results of a pilot reliability investigation of lateral and AP measurements taken from 19 subjects. The intraclass correlation coefficients for repeated measures (type 1,1)[45] were .89 for the AP talometatarsal angle, .96 for the lateral talometatarsal angle, and .97 for the calcaneal inclination angle. The coefficients were determined for one tester (DAN), who made all measurements in the study.

The reconstruction accuracy of the DLT method used to obtain the 3D coordinates of each target marker during the running trials was assessed by computing the root-mean-square (RMS) error between the true coordinate values of the points on the calibration structure and those estimated using the DLT method. The average values for the RMS error in the X, Y, and Z directions of the laboratory coordinate system were 0.41, 0.64, and 0.42 mm, respectively, for all subjects tested.

There were no differences in running speeds between groups. The normalized kinematic profiles of a representative subject from each of the low and high rear-foot groups are presented in Figures 2A and 2B, respectively. Both individuals were tested at the same average running speed of 3.1 m/s. The figures illustrate the relative rotations of tibial medial and lateral rotation ([Alpha]) and calcaneal eversion and inversion ([Beta]) for three trials over the total stance period. The standing neutral position for each subject is noted by the horizontal line in the graphs. The kinematic patterns indicate that at heel contact, or 0% of stance, the calcaneus was inverted and the tibia was laterally rotated relative to the calcaneus. The calcaneus then everted and the tibia medially rotated until both angular components reached their maximum value between 30% and 50% of the stance. In the last half of stance, there was a reversal of direction demonstrated by calcaneal inversion and tibial lateral rotation.

[Figure 2 ILLUSTRATION OMITTED]

Figures 3A and 3B demonstrate the coupling pattern between tibial medial and lateral rotation ([Alpha]) and calcaneal eversion and inversion ([Beta]) throughout the stance period for a representative subject in each foot group. The axes are scaled similarly to demonstrate the differences in both the magnitudes and kinematic patterns of rotations for each subject. The subject in the low rear-foot group demonstrated greater magnitudes of total calcaneal eversion and inversion rotation compared with tibial medial and lateral rotation for the total stance period. Conversely, the subject in the high rear-foot group showed magnitudes of rotations favoring tibial medial and lateral rotation over calcaneal eversion and inversion.

[Figure 3 ILLUSTRATION OMITTED]

Means, standard deviations, and results from the t tests for the dependent variables are summarized in Table 3. Group differences were demonstrated for the kinematic variables describing the stance phase coupling ratio of [Beta] max/[Alpha] max. A coupling ratio that is equal to 1.0 would represent an equal number of rotations about [Y.sub.t] and [Z.sub.t] for calcaneal eversion and inversion and tibial medial and lateral rotation, respectively. The ratio of 1.5 (SD = 1.3) for the low rear-foot group indicates greater calcaneal eversion and inversion relative to tibial medial and lateral rotation. The ratio of 0.91 (SD = 0.65) for the high rear-foot group favors tibial medial and lateral rotation relative to calcaneal eversion and inversion. There was also a difference between groups for the rotations that occurred about [Z.sub.t] from heel contact to maximum tibial medial rotation ([[Alpha].sub.HCMR]), particularly in the first 50% of the stance period. Both groups demonstrated rotations of similar magnitude for calcaneal eversion during this period of stance.

Table 3.
Results of Independent Sample t Tests for Kinematic
Variables(a)

                         Low Rear-          High Rear-
Kinematic                Foot Group(a)     Foot Group(b)
Variable               [bar] X     SD    [bar] X     SD      P

[[Alpha].sub.p-p]       10.4      5.5      14.6     7.6     .23

[[Beta].sub.p-p]        12.2      5.0      10.2     4.2     .90

[[Alpha].sub.p-p]
/[[Alpha].sub.p-p]       1.5      1.3       0.9     0.6     .04(b)

[[Alpha].sub.HCMR]       6.4      4.1      10.4     7.6     .04(b)

[[Beta].sub.HCEV]        8.8      4.4       8.2     3.9     .85

[[Beta].sub.HCEV]/
[[Alpha].sub.HCMR]       1.8      1.1       1.1     0.7     .29

(a) Means and standard deviations (in degrees of rotation),

(b) Significant group differences (P <.05).

Discussion

Although structural differences of the foot have been reported to account for kinematic variability found in previous investigations,[13,14,20] no prior study has evaluated lower-extremity kinematic behavior of distinct foot subtypes classified according to rigorous radiographic criteria. In our study, each subject had a foot structure that would have been characterized, by clinical appearance, as either as pes cavus or pes planus. Continued participation in the study was based on outcomes of radiographic measurements.

The use of similar running sandals allowed the leg and rear-foot rotations to be described without the potentially confounding effects of footwear such as reinforced heel counters, dual-density midsoles, and other motion-control components of the shoe. These constructional features have been shown to alter rear-foot kinematic responses in previous investigations. (58-61) The features of the running footwear unique to this study also allowed the rear-foot markers to be visualized directly for measurement. Markers were placed over bony prominences to minimize the influence of underlying soft tissue or muscle movement associated with skin marker use.

This investigation modeled the combined subtalar and talocrural joints as a single joint connecting the leg to the rear foot. Using a single-axis hinge model, the amount of rotation that occurs about anatomically defined coordinate axes provides an approximate and indirect assessment of the axis orientation. The orientation of the STJ axis influences not only the amount and direction of talocalcaneal joint motion but also the amount and direction of motion extrinsic to the joint.[6,10,11,13,16,62-63] Inman[13] contended that the inclination of the axis of the STJ from the transverse plane influences the relationship between frontal- and transverse-plane motions at the STJ. Van Langelaan[14] and Engsberg and colleagues[20,22] also reported axis orientations that varied according to the configuration (position) of the STJ. The predominant rotations demonstrated by each foot group in our study suggest an effective, combined STJ and talocrural joint axis orientation that favors calcaneal eversion and inversion for the low rear-foot structure and tibial medial and lateral rotation for the high rear-foot structure.

The values for total tibial medial and lateral rotation and eversion and inversion found in this study were appreciably smaller in magnitude than values reported by Nigg et al[24] for runners with high and low arches. These investigators observed 31.6 degrees of foot eversion and inversion in the high-arch group and 27.8 degrees of eversion and inversion in the low-arch group. These values can be compared with calcaneal eversion and inversion rotations of 10.2 and 12.2 degrees for high and low rear-foot groups, respectively, in our investigation. For tibial medial and lateral rotation, Nigg and colleagues reported values of 29.3 and 13.7 degrees for high- and low-arch groups, as compared with our findings of 14.6 and 10.4 degrees, respectively. A likely factor contributing to the greater magnitudes of rotation observed in the previous investigation was the use of the entire shoe and foot segment in analysis of relative motion between the leg and foot. Inclusion of the forefoot may have resulted in an increase in the magnitude of rotation between the lower leg and the foot, as midtarsal joint rotation ranging between 17 and 22 degrees has been documented in both cadaver and in vivo investigations.[10,58,64] In addition, investigations have shown that the magnitudes of foot rotation may represent the motion of the shoe rather than the foot,[59-61] thereby overestimating actual rear-foot motion.

A frequently discussed feature of foot biomechanics is the mechanism of coupling of leg rotation with pronation and supination of the foot.[3,10,11,55,62] As the calcaneus moves from its position of relative inversion at heel contact to maximum eversion, its motion is coupled with that of tibial medial rotation. Investigators[3] have proposed that it is not necessarily the magnitude of calcaneal eversion that is related to injury in the lower extremity, but rather the coupling of calcaneal eversion to leg rotation that may pose a more serious threat. Extremes in axis orientation may adversely affect the magnitude of this coupled response.

In our study, group differences were found for the coupling ratio for the total rotations about [Y.sub.t] and [Z.sub.t] ([Beta] max/[Alpha] max). If the rotations about each axis were equal, then a perfect coupling ratio of 1.0, or a one-to-one relationship, would be described. These ratios were 1.53 for the low rear-foot group and 0.91 for the high rear-foot group and are reflected in the kinematic profiles for representative subjects shown in Figures 3A and 3B.

Figure 4 shows individual subject data for both low and high rear-foot groups indicating the coupled relationship between calcaneal eversion and inversion ([Beta] max) and tibial medial and lateral rotation (a max) for the entire stance period of running gait. A one-to-one coupling relationship is illustrated by a line drawn from the origin (0,0) to the upper right corner of the graph. Using the analogy to the Inman model of the STJ axis,[13] this line would correspond to an axis with an average inclination of 45 degrees to the transverse plane. Data points that lie on the line indicate an equal amount of rotation for calcaneal eversion and inversion and tibial medial and lateral rotation. In our study, subjects in the high rear-foot group demonstrated greater magnitudes of tibial axial rotation ([Alpha]) and this rotation predominated over calcaneal eversion and inversion, when compared with subjects in the low rear-foot group. The predominant rotational component for subjects in the low rear-foot group favored calcaneal eversion and inversion.

[Figure 4 ILLUSTRATION OMITTED]

Figure 5 demonstrates individual subject data for the coupling responses for the period of stance between heel contact and maximum proration. This is the period of stance frequently addressed in pathology because of the high loads sustained over a relatively short period of time.[65] The magnitude of calcaneal eversion was similar for both groups for this period of stance. These findings are in contrast to the common clinical notion that subjects with high arches have more rigid feet and subjects with low arches have hypermobile feet in terms of their inversion-aversion range of motion.[64-67] These findings are consistent with data from a previous 3D investigation.[3] The primary kinematic differences, shown in the coupling responses between groups, can be related to the accompanying tibial medial rotation, which was greater for subjects in the high rear-foot group. This study suggests that associated musculoskeletal problems of the high rear-foot group are not necessarily due to a lack of motion; rather, a greater proportion of tibial axial rotation may be the precipitating cause of problems. The "imbalance" of rotations, as illustrated in the coupling ratios, may provide improved understanding about lower-extremity kinematic behavior and pathologies that are not apparent during examination of single-axis rotations alone.

[Figure 5 ILLUSTRATION OMITTED]

The coupling ratios determined for each foot group in this study may lend support to the theory proposed by Simkin and colleagues[7,8] that feet with relatively low calcaneal inclination angles absorb more "energy" during the stance phase of locomotion than feet with high angles; therefore, less motion is transferred to the bones of the shank and thigh. These investigators related their observations to the high incidence among army recruits of stress fractures in feet with low inclination angles. Conversely, the greater magnitude of tibial medial rotation that is coupled to calcaneal eversion may provide an explanation for characteristic clinical pathologies of the knee and hip associated with the cavus foot-type structure.[1,2,7]

Conclusion

If the proportion of rotation is considered to be an estimate of an "average" axis of the combined talocalcaneal joint and STJ, then the values found in this study reflect a distinct axis orientation for each foot group. The predominant rotations suggest a combined talocalcaneal joint and STJ axis to favor calcaneal eversion for the low rear-foot group and tibial medial and lateral rotation for the high rear-foot group. The individual axis rotations and coupling ratios varied for the different phases of stance chosen for analysis. The findings of this study also suggest that traditional frontal-plane assessment of foot and leg motion may not be adequate to describe the kinematics of this region. An assessment of the coupling between rear foot and leg, in combination with frontal-plane assessments, may improve our understanding of injuries related to foot structures similar to those studied in this investigation. Orthotic and footwear choices can be directed to controlling those rotations that predominate in these common clinical profiles.

Acknowledgments

We acknowledge Gary Soderberg, PhD, PT, FAPTA, Professor James G Andrews, Warren G Darling, PhD, Trudy L Burns, PhD, Bing Yu, PhD, Shun-Hwa Wei, PhD, PT, and Paula Ludewig, PhD, PT, for their input throughout various phases of this project.

(*) Deckers Corp, 495-A, S Fairview Ave, Goleta, CA 93117.

[dagger] Panasonic Co, 1 Panasonic Way. Secaucus, NJ 07094.

[double dagger] Peak Performance Technologies Inc, 7388 S Revere Pkwy, Suite 603, Englewood, CA 80112.

[sections] Faro Medical Technologies Inc, 125 Technology Park, Lake Mary, FL 32746.

[parallel] SAS Institute Inc, PO Box 8000, Cary, NC 27511.

References

[1] Cowan DN, Jones BH, Robinson JR. Foot morphologic characteristics and risk of exercise-related injury. Arch Fam Med. 1993;2:773-777.

[2] James SL, Bates BT, Osternig LR. Injuries to runners. Am J Sports Med. 1978;6:40-50.

[3] Nigg BM, Cole GK, Nachbauer W. Effects of arch height of the foot on angular motion of the lower extremities in running. J Biomech. 1993;26:909-916.

[4] Olerud C, Rosendahl Y. Torsion-transmitting properties of the hind foot. Clin Orthop. 1987;214:285-294.

[5] Radin EL, King HY, Riegger C, et al. Relationship between lower limb dynamics and knee joint pain. J Orthop Res. 1991;9:398-405.

[6] Root ML, Orien WP, Weed JH. Normal and Abnormal Function of the Foot: Clinical Biomechanics. Los Angeles, Calif: Clinical Biomechanics Corp; 1977:29-33.

[7] Simkin A, Leicher I, Giladi M, et al. Combined effect of foot arch structure and an orthotic device on stress fractures. Foot Ankle. 1989;10:25-29.

[8] Simkin A, Leichter I. Role of the calcaneal inclination in the energy storage capacity of the human foot: a biomechanical model. Med Biol Eng Comput. 1990.28:149-132.

[9] Moseley L, Smith R, Hunt A, Gant R. Three-dimensional kinematics of the rearfoot during the stance phase of walking in normal adult males. Clin Biomech. 1996;11:39-45.

[10] Lundberg A, Svensson O, Bylund C, et al. Kinematics of the ankle/foot complex, part 2: proration and supination. Foot Ankle. 1989;9:248-253.

[11] Lundberg A, Svensson I. Bylund C, Selvick G. Kinematics of the ankle/foot complex, part 3: influence of leg rotation. Foot Ankle. 1989;9:304-309.

[12] Benink RJ. The constraint mechanism of the human tarsus. Acta Orthop Scand. 1985;56(S215):49-68.

[13] Inman VT. The Joints of the Ankle. Baltimore, Md: Williams & Wilkins; 1976:35-43.

[14] Van Langelaan E. Relative talotibial movements and relative tarsal movements. Acta Orthop Scand. 1983;204:135-265.

[15] Novick A, Kelley D. Position and movement changes of the foot with orthotic intervention during the loading response of gait. J Orthop Sports Phys Ther. 1990;11:301-312.

[16] Root NIL, Weed JP, Sgarlato T, Bluth DR. Axis of motion of the subtalar joint. J Am Podiatr Med Assoc. 1966;36:149-15.5.

[17] Scott S. Winter D. Talocrural and talocalcaneal joint kinematics and kinetics during the stance phase of walking. J Biomech. 1991;24: 743-752.

[18] Wright DG, Desai SM, Henderson WH. Action of the subtalar and ankle-joint complex during the stance phase of walking. J Bone Joint Surg Am. 1964;46:361-382.

[19] Kirby KA. Methods for determination of positional variations in the subtalar joint axis. J Am Podiatr Med Assoc. 1987;77:228-234.

[20] Engsberg JR, Andrews JG. Kinematic analysis of the talocalcaneal/ talocrural joint during running support. Med Sci Sports Exerc. 1987;19: 275-284.

[21] Ball P, Johnson GR. Technique for the measurement of hindfoot inversion and eversion and its use to study a normal population. Clin Biomech. 1996:11:165-169.

[22] Engsberg JR, Grimston SK, Wackwitz JH. Predicting talocalcaneal joint orientations from talocalcaneal/talocrural joint orientations. J Orthop Res. 1988;6:749-757.

[23] Hawes M, Nachbauer W, Sovak D, Nigg B. Footprint parameters as a measure of arch height. Foot Ankle. 1992;13:22-26.

[24] Nigg BM, Nachbauer W, Cole GK. Effects of arch height of the foot on angular motion of the lower extremities in running. In: Proceedings of NACOB II, the Second North American Congress on Biomechanics, Chicago, IL. Chicago, Ill: American and Canadian Society of Biomechanics; 1992:233-234.

[25] Saltzman CL, Nawoczenski DA, Talbot KD. Measurement of the medial longitudinal arch. Arch Phys Med Rehabil. 1995;76:45-49.

[26] Giladi M, Milgrom C, Stein M, et al. The low arch, a protective factor in stress fractures: a prospective study of 295 military recruits. Orthop Rev. 1985;14:709-712.

[27] Hamill J, Bates B, Knutzen K, Kirkpatrick G. Relationship between selected static and dynamic lower extremity measures. Clin Biomech. 1989;4:217 -225.

[28] Kernozek T, Richard M. Foot placement angle and arch type: effect on rearfoot motion. Arch Phys Med Rehabil. 1990;71:988-991.

[29] Cowan DN, Robinson JR, Jones BH, et al. Consistency of visual assessments of arch height among clinicians. Foot Ankle. 1994;15: 213-217.

[30] Elveru RA, Rothstein JM, Lamb RL. Goniometric reliability in a clinical setting: subtalar and ankle joint measurements. Phys Ther. 1988;68:672-677.

[31] Lattanza L, Gray G, Kantner R. Closed versus open kinematic chain measurements of subtalar joint eversion: implications for clinical practice. J Orthop Sports Phys Ther. 1988;9:310-314.

[32] Picciano AM, Rowlands MS, Worrell T. Reliability of open and closed kinetic chain subtalar joint neutral positions and navicular drop test. J Orthop Sports Phys Ther. 1993;18:553-558.

[33] Smith-Oricchio K, Harris B. Interrater reliability of subtalar neutral, calcaneal inversion and eversion. J Orthop Sports Phys Ther. 1990;12: 10-15.

[34] Brand PW, Coleman WC. A standard for dorsal-planter and lateral radiographic projections of the feet. Orthopedics. 1987;10:117-120.

[35] Cobey JC, Sella E. Standardizing methods of measurement of foot shape by including the effects of subtalar rotation. Foot Ankle. 1981;2: 30-36.

[36] DiGiovanni JE, Smith SD. Normal biomechanics of the adult rearfoot: a radiographic analysis. J Am Podiatr Med Assoc. 1976;66: 812-824.

[37] Gould N. Graphing the adult foot and ankle. Foot Ankle. 1982;2: 213-219.

[38] Kitaoka HB, Lundberg A, Luo ZP, An KN. Kinematics of the normal arch of the foot and ankle under physiologic loading. Foot Ankle. 1995;16:492-499.

[39] Saltzman Cl., Brandser EA, Berbaum KS, et al. Reliability of standard foot radiographic measurements. Foot Ankle. 1994;15:661-665.

[40] Shereff M, DiGiovanni L, Bejjani F, et al. A comparison of non-weight-bearing and weight-bearing radiographs of the foot. Foot Ankle. 1990;10:306-311.

[41] Steel Ml,Johnson K, Dewith M, Ilstrup D. Radiographic measurements of the normal adult foot. Foot Ankle. 1980;1:151-158

[42] Gamble F, Yale I. Clinical Foot Roentgenology. New York, NY: Robert E Krieger; 1975.

[43] Hunt G. Subtalar Joint Pronation: The Influence of Balanced Insoles. College Park, Md: University of Maryland; 1980. Master's thesis.

[44] Weissman SD. Radiology of the Foot. Baltimore, Md: Williams Wilkins; 1989.

[45] Shrout P, Fleiss J. Intraclass correlations: uses in assessing rater reliability. Psychol Bull. 1979;86:420-428.

[46] Abdel-Aziz Y, Karara H. Direct linear transformation from comparator coordinates into object space coordinates in close-range photogrammetry. In: American Society of Photogrammetry Symposium on Close Range Photogrammetry. [Falls Church, Va: American Society of Photogrammetry; 1971.

[47] Norton BJ, Ellison JB. Reliability and concurrent validity of the Metrecom for length measurements on inanimate objects. Phys Ther. 1993;73:266-274.

[48] Smidt GL, McQuade KJ, Wei SH. Evaluation of the Metrecom and its use in quantifying skeletal landmark locations. J Orthop Sports Phys Ther. 1992;16:182- 188.

[49] Siegel KL, Kepple TM, O'Connell PG, et al. A technique to evaluate foot function during the stance phase of gait. Foot Ankle. 1995; 16: 764-770.

[50] Nawoczenski DA, Wei SH, McQuade KJ, Smidt GL. Use of a six degree-of freedom digitizer to augment three-dimensional kinematic video analysis techniques. Phys Ther. 1993;73:S101. Abstract.

[51] Craig J. Robotics: Mechanics and Control. Reading, Mass: Addison-Wesley Publishing Co; 1989:48-51.

[52] Wei SH, McQuade KJ, Smidt GL. Three-dimensional joint range of motion measurements from skeletal coordinate data. J Orthop Sports Phys Ther. 1993;18:687-691

[53] Yu B, Hay JG,. Angular momentum and performance in the triple jump: a cross-sectional analysis. J Appl Biomech. 1995;11:81-102.

[54] Areblad M, Nigg BM, Ekstrand J, et al. Three dimensional measurement of rearfoot motion during running. I Biomech. 1990;23:933-940.

[55] Siegler S, (then J, Schneck C. The three-dimensional kinematics and flexibility characteristics of the human ankle and subtalar joints, part I: kinematics. J Biomed Eng. 1988;110:364-373.

[56] Soutas-Little R, Beavis G, Verstraete M, Markus T. Analysis of foot motion during running using a joint coordinate system. Med Sci Sports Exerc. 1987;19:285-293.

[57] Mann R, Baxter D, Lutter L. Running Symposium. Foot Ankle. 1981;1:190-224.

[58] Ouzounian TJ, Shereff MJ. In vivo determination of midfoot motion. Foot Ankle 1989;10:140-146.

[59] Reinschmidt C, Stacoff A, Stussi E. Heel movement within a court shoe. Med Sci Sports Exerc. 1992;24:1390-1395.

[60] Stacoff A, Kalin X, Stussi E. The effects of shoes on the torsion and rearfoot motion in running. Med Sci Sports Exerc. 1991;23:482-490.

[61] Stacoff A, Reinschmidt C, Stussi E. The movement of the heel within a running shoe. Med Sri Sports Exerc. 1992;24:695-701.

[62] Hinterman B, Nigg BM, Cole GK Influence of selective arthrodesis on the movement transfer between calcaneus and tibia in vitro. Clin Biomech. 1994;9:356-361.

[63] Manter J. Movements of subtalar and transverse tarsal joints. Anat Rec. 1941;80:397-410.

[64] Hicks J. The mechanics of the foot, I: the joints. J Anat. 1953;87:345-357.

[65] Close J, Inman V, Poor P, Todd B. The function of the subtalar joint, Clin Orthop. 1967;50:159-177.

[66] Subotnick SI. The cavus foot. The Physician and Sportsmedicine. 1980;8(7):53-55.

[67] Subotnick SI. The flat foot. The Physician and Sportsmedicine. 1981;9(8):85-91.

DA Nawoczenski, PhD, PT, is Associate Professor, Department of Physical Therapy, Ithaca College-University of Rochester Campus, 300 E River Rd, Suite 1-102, Rochester, NY 14623 (USA) ([e-mail]). This study was completed in partial fulfillment of the requirements for Dr Nawoczenski's Doctor of Philosophy degree at The University of Iowa. Address all correspondence to Dr Nawoczenski.

CL Saltzman, MD, is Associate Professor, Department of Orthopaedic Surgery, The University of Iowa Hospitals and Clinics, Iowa City, Iowa.

TM Cook, PhD, PT, is Associate Professor, Department of Preventive Medicine and Environmental Health and Physical Therapy Graduate Program, The University of Iowa, Iowa City, Iowa.

The study protocol was approved by The University of Iowa Human Subjects Review Board.

This research was presented at the Summer Meeting of the American Orthopaedic Foot and Ankle Society;July 3-5, 1994; Coeur d'Alene, Idaho.

The study was supported by a grant from the Foundation for Physical Therapy Inc.

This article was submitted December 23, 1996, and was accepted October 1, 1997.