The Foot 24 (2014) 161–168

Contents lists available at ScienceDirect

The Foot journal homepage: www.elsevier.com/locate/foot

Can static foot posture measurements predict regional plantar surface area? Thomas G. McPoil a,∗ , Mathew Haager a , John Hilt a , John Klapheke a , Ray Martinez a , Cory VanSteenwyk a , Nicholas Weber a , Mark W. Cornwall b , Michael Bade a a b

Regis University, School of Physical Therapy, 3333 Regis Boulevard, G-4, Denver, CO 80221, United States Northern Arizona University, P.O. Box 15105, Flagstaff, AZ 86011, United States

a r t i c l e

i n f o

Article history: Received 30 May 2014 Received in revised form 16 July 2014 Accepted 19 July 2014 Keywords: Plantar surface area Foot posture Foot mobility Reliability

a b s t r a c t Background: The intent of this study was to determine if the use of a single or combination of static foot posture measurements can be used to predict rearfoot, midfoot, and forefoot plantar surface area in individuals with pronated or normal foot types. Methods: Twelve foot measurements were collected on 52 individuals (mean age 25.8 years) with the change in midfoot width used to place subjects in a pronated or normal foot mobility group. Dynamic plantar contact area was collected during walking with a pressure sensor platform. The 12 measures were entered into a stepwise regression analysis to determine the optimal set of measures associated with regional plantar surface area. Results: A two variable model was found to describe the relationship between the foot measurements and forefoot plantar contact area (r2 = 0.79, p < 0.0001). A four variable model was found to describe the relationship between the foot measurements and midfoot plantar contact area (r2 = 0.85, p < 0.0001) in those individuals with a 1.26 cm or greater change in midfoot width. Conclusions: The results indicate that clinicians can use a combination of simple, reliable and time efficient foot measures to explain 79% and 85% of the plantar surface area in the forefoot and midfoot, respectively. © 2014 Elsevier Ltd. All rights reserved.

1. Introduction Numerous studies have been conducted to determine if a relationship exists between the amount of plantar surface area and various clinical measurements of static foot posture. The reason for this research interest stems from studies conducted in the 1920s and 1930s that used footprints as a method of classifying the vertical height of the medial longitudinal arch of the foot as flat, normal, and high arched. Although Cureton et al. [1] questioned the validity of using plantar surface area to determine the vertical height of the arch, numerous researchers have continued to utilize plantar surface contact area in an attempt to predict the vertical height of the medial longitudinal arch. Several studies have been conducted to determine if a relationship exists between increased plantar surface area, associated with pes planus or flat foot, and the development of lower extremity overuse injuries. Kaufman et al. [2] and Levy et al. [3] assessed military populations and reported that individuals

∗ Corresponding author. Tel.: +1 303 964 5137; fax: +1 303 964 5474. E-mail address: [email protected] (T.G. McPoil). http://dx.doi.org/10.1016/j.foot.2014.07.003 0958-2592/© 2014 Elsevier Ltd. All rights reserved.

with increased plantar surface area associated with a pes planus or flatfoot type had a significantly greater number of lower extremity overuse injuries. Michelson et al. [4] assessed plantar surface area in a group of collegiate athletes and reported that individuals with a pes planus or flatfoot were not at greater risk of lower extremity injury. More recently, Knapik et al. [5] prospectively assessed over 2600 military recruits to determine if assigning running shoes based on plantar surface area was a factor in the development of injury during basic training. They reported that the shape of the plantar surface area had little influence on injury risk. It should be noted that Knapik et al. [5] used a visual assessment method to classify the plantar surface area of the foot rather than a quantifiable method such as footprints or a pressure sensor platform. In determining the validity of the plantar surface visual assessment technique used by Knapik et al. Swedler et al. [6] reported that 35% of the plantar surface shapes were misclassified when compared with measured arch height. Thus, it would appear that any study attempting to classify plantar surface area should consider using a quantifiable method rather than visual assessment. Since 1992, eight studies have attempted to predict the vertical height of the medial longitudinal arch based on measurements

162

T.G. McPoil et al. / The Foot 24 (2014) 161–168

derived from footprints obtained from inked mats, photographs, or pressure sensor platforms. Six of these studies [6–11] have used either the height of the tuberosity of the navicular bone or the dorsum of the foot as the indicator of the vertical height of the medial longitudinal arch. Needless to say the results of these studies have been disappointing, reporting that plantar surface area can only explain between 15% and 55% of the vertical height of the medial longitudinal arch. That would mean that only 50% of the total height of the medial longitudinal arch could be explained based on the plantar surface contact area. Teyhen et al. [12] were able to explain 60% of the medial longitudinal arch height using a set of five different plantar pressure parameters obtained while subjects walked over a pressure sensor platform. The five pressure parameters that were predictive of medial longitudinal arch height were; forefoot width, area between the foot axis and gait line, the lateral hindfoot force-time integral, the 1st metatarsal region force-time integral, and the mean pressure of metatarsals three, four and five. Only one study to date has attempted to predict plantar surface area, based on static measurements of foot posture and mobility. McPoil et al. [13] were able to utilize a three variable model that explained over 75% (r2 = 0.76) of the plantar surface contact area, excluding the toe region. The three variables in their predictive model included; heel width, midfoot width, and the arch height ratio (arch height divided by ball length) measured with equal weight on each lower extremity. While the results of this study would appear promising, the authors could only predict the total surface area of the plantar surface of the foot and failed to explain the amount of surface area in the rearfoot, midfoot, and forefoot regions. The ability for the clinician to be able to predict plantar surface area for specific regions of the foot, especially the midfoot, could be important when designing foot orthoses or footwear modifications. Recent research assessing the clinical effectiveness of foot orthoses for anterior knee pain reported that those individuals who had a 1.13 cm or greater change in midfoot width going from a non-weight bearing seated position to standing were more likely to benefit from the use of foot orthoses [14,15]. In addition, McPoil et al. [16] reported that those individuals with anterior knee pain who benefited from a contoured foot orthoses had a greater change in midfoot width in comparison to those who preferred a non-contoured orthoses. An important question based on the findings of these studies is whether the measured change in midfoot width is associated with an increase in midfoot plantar contact area. More recently, Queen et al. [17] reported that individuals with a pes planus or flat foot would appear to be at a greater risk for medial and lateral midfoot injuries during different athletic tasks. One could therefore hypothesize that one possible reason for this greater risk of injury is increased plantar surface area in the midfoot region. Methods available to the clinician to assess plantar surface area range from the use of foot imprints obtained from inked mats to more sophisticated and costly sensor platforms designed to measure plantar pressures as well as surface area. While an inked mat is cost effective, obtaining and analyzing the foot impressions can be time consuming. While commercially available pressure sensor platform systems permit the clinician to quickly obtain and analyze footprints, the equipment and software are expensive. Ideally, it would be most beneficial for the clinician to be able to predict a patient’s plantar surface area using a simple, reliable, and time efficient measurements of standing foot alignment. Thus, the purpose of our study was to determine if the use of a single or combination of static foot posture measurements can be used to predict rearfoot, midfoot, and forefoot plantar surface area in individuals with pronated or normal foot types. The 12 static foot posture measurements that were assessed included forefoot width (FFWid), midfoot width (MFWid), heel width (HLWid), and dorsal arch height (DAH) that were assessed while the participant stood

placing 10%, 50%, and 100% weight bearing on the tested lower extremity. 2. Methods 2.1. Participant characteristics One hundred and seventy individuals volunteered to have the width of their midfoot measured in both weight bearing and nonweight bearing to determine the amount of change in midfoot width. The change in midfoot width was defined as the difference between the width of the midfoot assessed in bilateral standing with equal weight on both feet and in non-weight bearing. The assessment in non-weight bearing was performed while the volunteer sat on the end of table so that both lower legs were hanging in a perpendicular position to the floor with the feet non-weight bearing and ankles slightly plantarflexed. Both the weight bearing and non-weight bearing assessment of midfoot width were assessed at 50% of the total foot length with a caliper (described in the Section 2.2) positioned so that the edges of the two plastic plates where aligned laterally and medially to the 50% length point on the dorsum of the foot. From this cohort of volunteers, 26 individuals had a change in midfoot width of greater than 1.26 cm and were placed in the pronated group. From the remaining 144 volunteers screened, 26 individuals matched by age and gender to those individuals in the pronated group with a change in midfoot width of less than or equal to 1.22 cm were placed in the normal group. The criteria for group assignment using the change in midfoot width was based on normative data published by McPoil et al. [18]. In their study, the mean change in midfoot width for 345 individuals was 1.01 cm with a standard error of the measurement (SEM) equal to 1.3 mm. Based on 95% confidence limits of the SEM (1.3 mm multiplied by 1.96), change in midfoot width values for the normal group would therefore be between 0.75 cm and 1.26 cm. The mean age of the 26 individuals (14 females, 12 males) in the pronated group was 26.3 ± 3.9 years with a range of 22–38 years. The mean age of the 26 individuals (14 females, 12 males) in the normal group was 25.6 ± 2.9 years with a range of 22–33 years. Participants were recruited from the Regis University population through community advertisements. In addition to the change in midfoot width, all participants met the following inclusion criteria: (1) no history of traumatic or overuse injury to either lower extremity in the 6 months preceding; (2) no congenital defect to either lower extremity; (3) no visible signs of foot pathology in both feet, including non-reducible claw or hammer toes, hallux valgus, hallux limitus and hallux rigidus. Each participant was instructed to conduct their normal activities of daily living prior to participation of the study. The Institutional Review Board of Regis University approved the protocol for data collection and all participants provided written informed consent prior to participation. 2.2. Instrumentation To obtain the foot measurements, a measurement platform that has been previously described was used (Fig. 1) [18]. The platform had modified grooves to allow for a plastic block to slide alongside the participant’s feet. The block contained a spring loaded reversible metal bar that was used to assess foot length. In addition to the platform, two addition instruments were constructed to allow for the measurements of arch height and the various foot widths. The weight bearing dorsal arch height gauge consisted of a digital caliper (Model #700-126, Mitutoyo America Corp., Aurora, IL 60502) with the fixed point attached to a 1.2 cm × 5.0 cm × 10.0 cm plastic block to hold the device in a vertical position. A sliding metal rod attached to the moving point of the caliper was used

T.G. McPoil et al. / The Foot 24 (2014) 161–168

163

Fig. 3. Measurement of midfoot width using digital caliper. Fig. 1. Measurement of total foot length using foot measurement platform.

to assess the arch height (Fig. 2). A second digital caliper (Model #700-126, Mitutoyo America Corp., Aurora, IL 60502) was modified to permit the assessment of forefoot, midfoot, and heel width by attaching 0.03 cm × 0.8 cm × 9.0 cm plastic plates to the fixed and moving points of the caliper (Fig. 3). A digital scale was used to determine subject’s total body weight as well as to facilitate foot posture measurements in 10% weight bearing. To obtain the dynamic plantar surface contact area during walking, an EMED-X floor mounted capacitance transducer platform (NOVEL USA Inc., Minneapolis, MN 55415) with an active sensor area of 32 cm × 47.5 cm was positioned at the midpoint of a 12m walkway. The platform had a matrix of 6080 sensors with a

Fig. 2. Digital gauge used to measure dorsal arch height.

density of four sensors per cm2 and a sampling rate of 100 Hz. The input pressure saturation point for the capacitance sensors used in the platform is 1270 kPa and none of the trials performed by the subjects in the current study exceeded this saturation point. 2.3. Procedures Upon arrival to the testing center, the participant’s height and weight was recorded and the foot with the greatest change in midfoot width was selected for data collection. Prior to obtaining the standing measurements, each subject was positioned on the foot measurement platform with both heels placed in left and right heel cups positioned 15.5 cm apart. Subjects were instructed to have their heels firmly against the back of the heel cups with medial calcanei touching the medial border of the heel cups. Once the subject was properly positioned on the platform, the subject was instructed to place equal weight on both feet, place their hands at their sides and look forward, so that the 50% weight bearing measurements could be obtained. Total foot length was measured first by placing the sliding metal bar on the centered metal ruler attached to the platform and moving the bar until touching the longest toe, usually the hallux, of each foot (Fig. 1). Next, the dorsal arch height at 50% of the total foot length was measured using the weight bearing arch height gauge previously described. To determine the point of 50% of total foot length, the total measured foot length was divided in half and the dorsum of the foot was marked at the 50% length point using a water-soluble ink pen. The sliding metal rod of the weight bearing height gauge was then positioned over the 50% length mark and the vertical height from the top of the platform to the dorsum of the foot was measured (Fig. 2). The midfoot width was then assessed by positioning the digital caliper so that the arms of the caliper were aligned laterally and medially to the 50% length point marked on the dorsum of the foot (Fig. 3). Once aligned, the arms were moved until they just made contact with the skin at the 50% length point. Following this measurement, the rater measured the subject’s forefoot width by assessing the widest portion of the forefoot by placing the arms of the digital caliper on either side of the metatarsal heads (Fig. 4). Next, the rater instructed the subject to step forward a few inches, out of the heel cups, in order to perform heel width measurements. With the subject in the same posture as previously described, hands at their sides and looking forward, the caliper was placed at a downward 45 degree angle around the

164

T.G. McPoil et al. / The Foot 24 (2014) 161–168

Fig. 4. Measurement of forefoot width using digital caliper.

base of the subject’s heel where heel width was recorded (Fig. 5). The subject was then instructed to step off the platform and onto an adjacent digital scale, where the forefoot width, midfoot width, dorsal arch height, and heel width were measured in 10% and 100% weight bearing on the subject’s previously selected foot. Prior to obtaining the standing measurements on the digital scale, each subject’s foot was positioned appropriately, 15.5 cm apart, with the foot to be measured placed on the digital scale and the other foot on the foot platform (Fig. 6). While the subject stood facing the digital scale readout, they were instructed to shift their body toward their non-selected limb so as to achieve a 10% body weight on the foot to be measured. Once the subject could consistently place 10% of their body weight on the foot to be measured while keeping their upper body upright, forefoot width, midfoot width, heel width, and dorsal arch height were measured. Next, the subject was instructed to shift their entire weight onto their selected

Fig. 6. Measurement of heel width in 100% weight bearing while subject stands on digital scale.

limb, maintaining the upright body position, so that the same four foot measures could be recorded with the participant placing 100% of their body weight on the foot to be measured. While standing on the foot being measured, the participants were instructed to keep their foot completely relaxed and were allowed to touch their fingertips to a supporting surface to prevent loss of balance. When the foot measurements were completed, each participant was then instructed to practice walking barefoot at a self-selected speed along the 12-m walkway for several minutes. In order to prevent targeting of the EMED-X platform, subjects were instructed not to look at the ground while walking. Walking speed was monitored using a digital stopwatch to time the subject as they walked between two lines positioned 6.1 m apart and were equidistant in relation to the platform. When between-trial walking speed was consistent (variation of less than 5% between trials), each subject was asked to walk barefoot over the walkway while data were recorded from the EMED-X platform for five trials on both the left and right foot. 2.4. Determination of reliability

Fig. 5. Measurement of heel width using digital caliper.

To determine the reliability for the 12 measurements, three raters were asked to assess the first 10 participants (10 feet) from the pronated group and the first 10 participants (10 feet) from the normal group. The raters performing the measurements included two 2nd year graduate physical therapy students, with two months of clinical experience, and a physical therapist who had 10 years of experience. Each rater attended a single 1-h training session to receive verbal instructions as well as to practice the techniques to ensure that they were taking the measurements correctly. The reliability data were collected in two sessions, one-week apart. During each session, each rater performed all 12 measurements on each of the 20 participants. A different investigator recorded all

T.G. McPoil et al. / The Foot 24 (2014) 161–168

165

2.6. Statistical analysis To assess the reliability of the 12 measurements, intraclass correlation coefficients (ICC) were calculated to determine the consistency of each rater to perform the measurements between sessions (intra-rater ICC3,1 ) as well as the consistency among the three raters (intra-rater ICC2,3 ). The level of reliability for the ICC were classified based the characterizations reported by Landis and Koch [20]. These characterizations were: slight, if the correlation ranged from 0.00 to 0.20; fair, if the correlation ranged from 0.21 to 0.40; moderate, if the correlation ranged from 0.41 to 0.60; substantial, if the correlation ranged from 0.61 to 0.80; and almost perfect, if the correlation ranged from 0.81 to 1.00. Although the ICC is commonly used as an assessment of reliability, they are difficult to interpret clinically since ICC values are dependent on the variability of the group being assessed and may not transfer to different patient populations [21]. In light of this issue, the standard error of the measurement (SEM) was also calculated as another index of reliability. The SEM is a number in the same units as the original measurement that represents the way a single score would vary if the six-foot measurements used in this study were measured more than once [22]. In addition to descriptive statistics, a series of t-tests were performed to determine if there were differences in the plantar surface area between the FFPSA, MFPSA, and RFPSA as well for the Arch Index. The 12-foot measures (FFWid, MFWid, HLWid, DAH at 10%, 50%, 100% body weight) were entered into a stepwise forward linear regression to determine the most parsimonious set of variables associated with those plantar surface areas that were significantly different between the pronated group and the normal group. A significance level of p < 0.05 was required for entry into the model and p < 0.06 was used as the criteria for removal from the model. All statistical analyses were performed using JMP software, Version 8.0 (SAS Institute Inc., Cary, NC 27513). An alpha level of .05 was established for all tests of significance.

Fig. 7. Division of dynamic plantar surface area for the rearfoot (M01), midfoot (M02), forefoot (M03), and hallux/toes (M04).

measurements for each rater to prevent any possible measurement bias. In addition to inter-rater reliability, intra-rater reliability between the two sessions was also determined for each rater.

2.5. Data analysis To determine plantar surface area, a standardized four region masking model (Novel Automask, NOVEL USA Inc., Minneapolis, MN 55415) was used to divide the dynamic plantar surface area into four regions; rearfoot, midfoot, forefoot, and hallux/toes (see Fig. 7). The heel to midfoot and midfoot to forefoot regions were defined by using 73% and 45% of the entire foot length from the tip of the most forward toe, usually the hallux, to the posterior edge of heel, respectively. The forefoot to toe region was defined by using the pressure gradients around the peak pressures of the toes. The Novel Groupmask program (NOVEL USA Inc., Minneapolis, MN 55415) was used to determine the mean plantar surface area for the five trials collected on each participant’s foot that was selected for analysis. The plantar surface area was then determined for the rearfoot (RFPSA), midfoot (MFPSA), and forefoot (FFPSA) regions. The Arch Index, as defined by Cavanagh and Rodgers [19], was calculated by dividing the MFPSA by the sum of RFPSA, MFPSA, and FFPSA.

3. Results Demographic data for all subjects are listed in Table 1. The intra-rater and inter-rater ICC and SEM values are shown in Tables 2 and 3. The intra-rater reliability for all 12-foot measurements ranged from 0.90 to 0.98 for all three raters regardless of experience level. The intra-rater SEM values ranged from 0.08 to 0.22 cm and were all less than 5% of the actual mean measurement values. The inter-rater reliability ICC for the same measurements ranged from 0.92 to 0.98 for both day one and day two with SEM values all being less than 5% of the actual mean measurement values. Descriptive statistics for all measurements are listed in Table 4. The results of the t-tests indicated that mean values between the normal and pronated groups were significantly different for MFPSA (p = .0001), FFPSA (p = .0029) and the Arch Index (p = .0001). There was no difference in RFPSA between the two groups. As expected, there was a significant difference (p = .0001) between the pronated and normal groups for the change in midfoot width. The mean change in midfoot width was 1.52 + 0.24 cm (range from 1.27 to 2.24 cm) for the pronated group and 1.04 + 0.16 cm (range from 0.07 to 1.22 cm) for the normal group. As previously noted, a previous study was able to explain 76% of the plantar contact area of the foot excluding the toes. In the current study, using all 52 feet regardless of foot type classification, the stepwise forward regression analysis resulted in a five (5) variable model (F = 40.59; p < .0001) with r = 0.91 (r2 = 0.82). The five measurements that were included in the model were FFWid100%, DAH50%, MFWid100%, HLWid50%, and HLWid100%.

166

T.G. McPoil et al. / The Foot 24 (2014) 161–168

Table 1 Subject characteristics with values presented as mean, standard deviation (SD) and 95% confidence intervals (CI). Mean

SD

95% CI

Table 3 Inter-rater reliability coefficients (ICC) and standard error of the measurement (SEM) for day 1 and day 2. DAY 1

All feet (n = 52) Age (Years) Height (cm) Weight (kg) Change in Midfoot Width (cm)

25.94 168.28 67.49 1.28

3.46 8.26 12.50 0.32

24.90–26.50 166.00–170.60 64.10–70.97 1.19–1.37

Pronated group (n = 26) Age (Years) Males (n = 12) Females (n = 14) Height (cm) Males (n = 12) Females (n = 14) Weight (kg) Males (n = 12) Females (n = 14) Change in midfoot width (cm) Males (n = 12) Females (n = 14)

26.30 27.25 25.50 167.86 173.14 162.57 69.04 79.03 59.04 1.53 1.59 1.46

3.90 4.81 2.96 6.36 6.56 6.15 8.79 11.11 6.48 0.24 0.29 0.19

24.72–27.88 24.02–30.48 23.79–27.21 165.30–170.40 168.70–177.50 159.00–166.10 65.49–72.59 71.57–86.49 55.30–62.78 1.43–1.63 1.40–1.79 1.35–1.57

Normal group (n = 26) Age (Years) Males (n = 12) Females (n = 14) Height (cm) Males (n = 12) Females (n = 14) Weight (kg) Males (n = 12) Females (n = 14) Change in midfoot width (cm) Males (n = 12) Females (n = 14)

25.81 26.11 25.50 171.37 178.64 164.09 66.61 74.17 59.04 1.06 1.10 1.01

3.15 3.33 2.96 6.55 9.28 3.82 9.21 11.93 6.48 0.15 0.12 0.17

24.54–27.08 23.87–28.35 23.79–27.21 168.70–174.00 172.40–184.90 161.90–166.30 62.89–70.33 66.16–82.18 55.30–62.78 0.99–1.21 0.97–1.23 0.91–1.11

DAY 2

ICC

Since RFPSA was not found to be significantly different between the normal and pronated groups, a stepwise forward regression analysis was only conducted on MFPSA and FFPSA for the pronated and normal groups as well as all feet. To predict MFPSA for the pronated group, the regression analysis resulted in a four variable model (FFWid100%, MFWid100%, HLWid50%, HLWid100%) with r = 0.92 (r2 = 0.85). For the normal group, the stepwise forward regression analysis to predict MFPCA resulted in a two (2) variable model (DAH100%; HLWid100%) with r = 0.66 (r2 = 0.43). For all 52 feet, the regression analysis resulted in a two (2) variable model (DAH50%; MFWid100%) with r = 0.78 (r2 = 0.61).

SEM (in cm)

ICC

SEM (in cm)

Forefoot width 10% 0.94 50% 0.97 100% 0.92

0.15 0.11 0.19

0.96 0.94 0.97

0.12 0.18 0.13

Dorsal arch height 10% 0.97 50% 0.98 100% 0.97

0.13 0.09 0.11

0.98 0.94 0.97

0.09 0.15 0.11

Midfoot width 10% 0.96 50% 0.98 100% 0.95

0.13 0.11 0.15

0.96 0.98 0.95

0.14 0.10 0.10

Rearfoot width 10% 0.94 50% 0.95 100% 0.95

0.15 0.13 0.13

0.95 0.97 0.96

0.14 0.10 0.13

To predict FFPCA for the pronated group, the regression analysis only selected a single measurement (FFWid100%) with r = 0.85 (r2 = 0.72). For the normal group, the stepwise forward regression analysis to predict FFPCA resulted in a two (2) variable model (FFWid 100%; HLWid 100%) with r = 0.88 (r2 = 0.78). For all 52 feet, the regression analysis also resulted in a two (2) variable model (FFWid 100%; HLWid100%) with r = 0.89 (r2 = 0.79). 4. Discussion Previous research has demonstrated that individuals with anterior knee pain who demonstrate a change in midfoot width of greater than 1.13 cm may have a better outcome using a contoured foot orthosis [14,15]. An important question based on the findings of these studies is whether the measured change in midfoot width is associated with an increase in midfoot plantar surface area which could assist in explaining the need for a contoured foot orthosis. Previous attempts to utilize simply, clinically efficient and Table 4 Mean and standard deviation (SD) for the foot measurements, plantar contact areas (PCA) and Arch Index by groups.

Table 2 Intra-rater reliability coefficients (ICC), mean (in cm) and standard error of the measurement (SEM).

ALL Feet (n = 52)

Pronated (n = 26)

Normal (n = 26)

Mean

Mean

Mean

SD

SD

SD

SEM (cm)

ICC

Mean (cm)

SEM (cm)

ICC

Mean (cm)

SEM (cm)

Forefoot width (in cm) 10% 9.34 50% 9.51 100% 9.59

Forefoot width 10% 0.98 9.41 50% 0.96 9.62 100% 0.94 9.75

0.10 0.15 0.16

0.94 0.98 0.96

9.44 9.65 9.69

0.15 0.10 0.14

0.94 0.92 0.98

9.64 9.74 9.87

0.16 0.20 0.09

Dorsal arch height (in cm) 10% 6.38 50% 6.30 100% 6.15

0.59 0.59 0.57

6.29 6.18 6.07

0.59 0.55 0.59

6.48 6.41 6.24

0.58 0.61 0.54

Dorsal arch height 10% 0.98 6.12 50% 0.97 6.10 100% 0.98 5.97

0.09 0.10 0.09

0.97 0.96 0.97

6.29 6.24 6.15

0.12 0.13 0.12

0.98 0.90 0.97

6.15 6.17 5.95

0.09 0.21 0.11

Midfoot width (in cm) 10% 8.30 50% 8.44 100% 8.43

0.79 0.81 0.77

8.66 8.89 8.83

0.81 0.78 0.75

7.95 8.10 8.05

0.61 0.59 0.60

Midfoot width 10% 0.98 8.34 50% 0.99 8.70 100% 0.98 8.65

0.10 0.09 0.12

0.98 0.91 0.95

8.71 8.99 8.87

0.11 0.22 0.16

0.95 0.99 0.98

8.60 8.91 8.83

0.15 0.09 0.10

Rearfoot width (in cm) 10% 6.35 50% 6.51 100% 6.51

0.51 0.49 0.53

6.42 6.67 6.66

0.54 0.50 0.54

6.28 6.35 6.36

0.48 0.43 0.48

Rearfoot width 10% 0.95 6.46 50% 0.96 6.61 100% 0.94 6.70

0.13 0.12 0.14

0.91 0.94 0.97

6.53 6.77 6.79

0.17 0.14 0.11

0.97 0.98 0.98

6.10 6.42 6.51

0.10 0.08 0.08

RF PSA (in cm2 ) MF PSA (in cm2 ) FF PSA (in cm2 ) Arch index

4.39 11.04 6.17 0.07

34.50 30.12 51.28 0.26

3.92 9.82 6.07 0.04

33.07 17.90 46.32 0.18

4.80 8.62 5.31 0.07

Rater 1 ICC

Rater 2 Mean (cm)

Rater 3

33.78 24.01 48.8 0.22

0.66 0.71 0.66

9.58 9.76 9.83

0.66 0.68 0.64

9.10 9.25 9.35

0.57 0.65 0.60

T.G. McPoil et al. / The Foot 24 (2014) 161–168

reliable foot measurements to predict plantar surface area have only explained plantar surface area for the entire foot, excluding the toes [13]. In addition, previous studies assessing plantar surface area have made no attempt to subgroup individuals based on foot mobility. Thus, the intent of our study was to determine if the use of a single or combination of static foot posture measurements, assessed in different percentages of weight bearing, could be used to predict rearfoot, midfoot, and/or forefoot plantar surface area in individuals with pronated or normal foot types. Individuals selected to participate in the study were classified prior to testing as pronated or normal based on the change in midfoot width with the foot classification re-assessed after testing using the Arch Index technique described by Cavanagh and Rodgers [19]. 4.1. Assessment of reliability The first step in interpreting our data prior to performing any prediction analyses was to assess the results of the intra- and interrater reliability of the 12 foot measurements utilized in the study. The ICC values for all three (3) raters, regardless of the number of years of clinical experience, were all classified as “almost perfect” for both intra-rater and inter-rater reliability using the characterizations outlined by Landis and Koch [20]. In addition, the SEM values for both intra-rater and inter-rater reliability were also quite small (less than 5% of the mean) for all of the foot measurements used in this study. Based on these findings, the authors believe that the reliability of the measurement techniques that were used in this study were acceptable in order to proceed with further analysis of the results. 4.2. Prediction of specific plantar contact areas Prior to performing further analysis of the data, the authors also believed it was important to determine if the pronated and normal groups were meaningfully different based on the change in midfoot width and the Arch Index. The results of the t-test results confirmed that both the change in midfoot width and the Arch Index were significantly different between the two groups. In addition, based on 95% confidence limits of the SEM previously published the change in midfoot width for the normal group should be between 0.75 cm and 1.26 cm. In the current study, the change in midfoot width for the normal group was between 0.70 and 1.22 cm, while the change in midfoot width for the pronated group was between 1.27 and 2.24 cm. With regard to the Arch Index, Cavanagh and Rodgers [19] stated that a normal arch would be between 0.21 and 0.26. In the current study, the mean value for the Arch Index of the normal group was 0.18 and 0.26 for the pronated group. Based on these findings, the authors concluded that the normal and pronated groups in this study were meaningfully different and that further analysis could be undertaken. As previously noted, an earlier investigation was able to explain 76% of the plantar surface area of the foot, excluding the toes using a sample of 155 individuals. The three (3) foot measurements that were required were midfoot width, heel width, and the arch height ratio, with all three recorded while the individuals stood with 50% body weight on each foot. Since in the current study only 52 feet were included in the analysis, the authors first determined if they could achieve the same degree of prediction for the plantar surface area of the foot, excluding the toes. Based on the results using all 52 feet regardless of foot type classification, the regression analysis was able to explain over 80% of the plantar contact area, excluding the toes. The five measurements that were included in the model were FFWid100%, DAH50%, MFWid100%, HLWid50%, and HLWid100%. Thus, it would appear that the addition of foot measurements with 100% weight bearing on the foot being assessed

167

slightly enhances the ability to predict total plantar surface area, excluding the toes. The primary purpose for performing this study was to determine if the plantar surface area of the rearfoot, midfoot, and forefoot regions of the foot could be determined. The results of t-tests indicated that MFPSA and FFPSA were significantly different between the pronated and normal groups, but that there was no difference between the groups for RFPSA. Based on these results, further analysis was only conducted on the midfoot and forefoot regions. 4.3. Regression analyses Based on the findings of the regression analysis, the use of FFWid100%, MFWid100%, HLWid50%, HLWid100% can explain 85% of the variance of MFPSA for the pronated group. The use of DAH50% and MFWid100% could only explain 61% of the variance for MFPSA for all feet, while the use of DAH100% and HLWid100% could only explain 43% of the variance for the normal group. These results indicate that the ability of the clinician to accurately predict the MFPSA using the static foot measurements assessed in this study is limited to those individuals with a 1.26 cm or greater change in midfoot width. Using data from the current study, the clinician can predict the MFPSA based on the selected foot measurements using the following formula: −56.71 + (5.74 × FFWid100%) + (9.34 × MFWid100%) + (−22.60 × HLWid50%) + (14.80 × HLWid100%) Using this prediction formula, the mean value was calculated for the predicted MFPSA (30.43 cm2 ) and compared using a t-test to the actual measured mean obtained for MFPSA (30.12 cm2 ) for the pronated group. The results of the t-tests were not significant (p = .71) and the error of the mean (1.82 cm2 ) was quite small which further validates the prediction formula for those individuals with a change in midfoot width equal to or greater than 1.26 cm. Initial t-tests to determine differences in regional plantar contact area also indicated that FFPSA was also significantly different between the pronated and normal groups with the pronated group demonstrating an increase in plantar surface area. Since an increase in the spread or width of the forefoot of the foot is often associated with pes planus or flat foot, this difference in FFPSA is not surprising. Based on the findings of the regression analysis, the use of FFWid100% and HLWid100% could explain 79% of the variance of FFPCA for all feet and 78% of the variance of FFPSA for the normal group. The use of FFWid100% was found to explain 72% of the variance of FFPSA for the pronated group. As a result of the similarity of the regression analyses for all feet as well as the pronated and normal groups, the authors would recommend using the variables selected for all feet when attempting to predict FFPSA. Thus, the clinician can predict the FFPSA based on the selected foot measurements using the following formula: −34.87 + (5.79 × FFWid100%) + (4.35 × HLWid100%) Using this prediction formula, the mean value was calculated for the predicted FFPSA (48.97 cm2 ) and compared using a t-test to the actual measured mean obtained for FFPSA (48.8 cm2 ) for all feet. The results of the t-tests were not significant (p = .68) and the error of the mean (0.79 cm2 ) was quite small which further validates the prediction formula for all individuals irrespective their change in midfoot width. The overall findings of this study indicate that the clinician can use the measurements of FFWid100% and HLWid100% to explain 79% of the plantar surface area of the forefoot in all individuals and with the addition of two more measurements, MFWid100% and HLWid50%, explain 85% of the plantar surface area of the midfoot

168

T.G. McPoil et al. / The Foot 24 (2014) 161–168

in those individuals with a 1.26 cm of greater change in midfoot width. These four (4) measurements are simple to perform, have been shown to be reliable irrespective of rater experience, and are time efficient for the clinician to perform. To the best of the authors’ knowledge, this is the first study that has been able to predict plantar surface area for specific plantar regions of the foot.

Conflict of interest None of the authors listed for this study had any financial or personal relationships with other people or organizations that could have inappropriately influenced the results reported. References

4.4. Limitations A possible limitation of the current study is that the plantar contact area was acquired as participants walked across a pressure sensor platform. Urry and Wearing [23] have noted that the foot surface area mappings recorded using pressure sensor platforms with low sensor resolution are not the same as foot impressions obtained using an inked mat system in static standing. While the pressure sensor platform used in the current study had a greater sensor resolution than the sensor platform used by Urry and Wearing [23], care should be used when comparing the results reported in this study with measures of plantar surface area obtained using an inked mat system in static standing. The authors believe that acquisition of pressure sensor data dynamically, in activities such as walking, provides a more functional representation of plantar surface area in comparison to static standing. Another limitation of the current study was the assumption by the authors that participants, when asked to stand with equal weight on both feet, were actually placing 50% of their body weight on each foot. While several methods could have been employed to ensure that the participants were placing 50% of their body weight on each foot, for example having the participant stand with one foot on a scale, the methodology used by the authors in the current study can be easily reproduced by clinicians. Teser et al. [24] have previously reported that the amount of asymmetry in weight distribution between extremities in relaxed standing is 4% or less in healthy individuals. Furthermore, while their could have been slight variations in the amount of weight placed on each extremity when the subject was asked to place equal weight on each foot, the high level of consistency for all 12 foot measurements used in this study over multiple days would suggest that any degree of asymmetry was minimal. Finally, it is important to note that the external validity of this study’s findings is limited to the demographic characteristics of the population that participated in this study. 5. Conclusion While further research is always necessary, the results of this study indicate that the clinician can use the measurements of FFWid100% and HLWid100% to explain 79% of the plantar surface area of the forefoot in all individuals and with the addition of two more measurements, MFWid100% and HLWid50%, explain 85% of the plantar surface contact area of the midfoot in those individuals with a 1.26 cm of greater change in midfoot width. Prediction equations are provided to permit the practitioner to calculate either the midfoot or forefoot plantar surface area depending on clinical need or interest. Financial disclosure There were no external or internal sources of funding used to support this study.

[1] Cureton TK, Wickens JS, Haskell PE. The validity of footprints as a measure of the vertical height of the arch and functional efficiency of the foot. Res Q 1935;6:70–80. [2] Kaufman KR, Brodine SK, Shaffer RA, Johnson CW, Cullison TR. The effect of foot structure and range of motion on musculoskeletal overuse injuries. Am J Sports Med 1999;27:585–93. [3] Levy JC, Mizel MS, Wilson LS, Fox W, McHale K, Taylor DC, et al. Incidence of foot and ankle injuries in West Point cadets with pes planus compared to the general cadet population. Foot Ankle Int 2006;27:1060–4. [4] Michelson JD, Durant DM, McFarland E. The risk associated with pes planus in athletes. Foot Ankle Int 2002;23:629–33. [5] Knipik JJ, Brosch LC, Venuto M, Swedler DI, Bullock SH, Gaines LS, et al. Effect of injuries of assigning shoes based on foot shape in air force basic training. Am J Prev Med 2010;38:S197–211. [6] Swedler DI, Knapik JJ, Grier T, Jones BH. Validity of plantar surface visual assessment as an estimate of foot arch height. Med Sci Sports Exerc 2010;42: 375–80. [7] Hawes MR, Nachbauer W, Sovak D, Nigg BM. Footprint parameters as a measure of arch height. Foot Ankle 1992;13:22–6. [8] Chu WC, Lee SH, Chu W, Wang TJ, Lee MC. The use of arch index to characterize arch height: a digital image processing approach. IEEE Trans Biomed Eng 1995;42:1088–93. [9] Shiang TY, Lee SH, Lee SJ, Chu WC. Evaluating different footprint parameters as a predictor of arch height. IEEE Eng Med Biol Mag 1998;17:62–6. [10] McPoil TG, Cornwall MW. Use of plantar contact area to predict medial longitudinal arch height during walking. J Am Podiatr Med Assoc 2006;96: 489–94. [11] Queen RM, Mall NA, Hardaker WM, Nunley JA. Describing the medial longitudinal arch using footprint indices and a clinical grading system. Foot Ankle Int 2007;28:456–62. [12] Teyhen DS, Stoltenberg BE, Collinsworth KM, Giesel CL, Williams DJ, Kardouni CH, et al. Dynamic plantar pressure parameters associated with static arch height index during gait. Clin Biomech 2009;24:391–6. [13] McPoil TG, Vicenzino B, Cornwall MW, Collins N. Can foot arthropometric measurements predict dynamic plantar surface contact area? J Foot Ankle Res 2009;2:28–36. [14] Vicenzino B, Collins N, Cleland J, McPoil T. A clinical prediction rule for identifying patients with patellofemoral pain who are likely to benefit from foot orthoses: a preliminary determination. Br J Sports Med 2010;44: 862–6. [15] Mills K, Blanch P, Dev P, Martin M, Vicenzino B. A randomized control trial of short term efficacy of in-shoe foot orthoses compared with a wait and see policy for anterior knee pain and the role of foot mobility. Br J Sports Med 2012;46:247–52. [16] McPoil TG, Vicenzino B, Cornwall MW. Effect of foot orthoses contour on pain perception in individuals with patellofemoral pain. J Am Podiatr Med Assoc 2011;101:7–16. [17] Queen RM, Mall NA, Nunley JA, Chuckpaiwong B. Difference in plantar loading between flat and normal feet during different athletic tasks. Gait Posture 2009;29:582–6. [18] McPoil TG, Vicenzino B, Cornwall MW, Collins N, Warren M. Reliability and normative values for the foot mobility magnitude: a composite measure of vertical and medial-lateral mobility of the midfoot. J Foot Ankle Res 2009;2:1–12. [19] Cavanagh PR, Rodgers MM. The arch index: a useful measure from footprints. J Biomech 1987;20:547–51. [20] Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics 1977;33:159–74. [21] Rohner-Spengler M, Mannion AF, Babst R. Reliability and minimal detectable change for the figure-of-eight-20 method of measurement of ankle edema. J Orthop Sports Phys Ther 2007;37:199–205. [22] Rothstein JM. Measurement and clinical practice: the theory and application. In: Rothstein JM, editor. Measurement in physical therapy. New York: Churchill Livingstone; 1985. p. 1. [23] Urry SR, Wearing S. Arch indexes from ink footprints and pressure platforms are different. Foot 2005;15:68–73. 24 Teser S, Hagstrom N, Fallang B. Weight distribution in standing and sitting positions, and weight transfer during reaching tasks, in seated stroke subjects and healthy subjects. Physiother Res Int 2007;12:82–94.