- Email: [email protected]
Effect of using truncated versus total foot length to calculate the arch height ratio
Available online at www.sciencedirect.com
The Foot 18 (2008) 220–227
Effect of using truncated versus total foot length to calculate the arch height ratio Thomas G. McPoil a,∗ , Mark W. Cornwall a , Bill Vicenzino b , Deydre S. Teyhen c , Joseph M. Molloy c , Douglas S. Christie c , Natalie Collins b a
Gait Research Laboratory, Department of Physical Therapy & Athletic Training, Northern Arizona University, Flagstaff, AZ 86011, USA b Department of Physiotherapy, University of Queensland, St. Lucia, Brisbane, Queensland 4072, Australia c U.S. Army-Baylor University Doctor Program in Physical Therapy, Foot Sam Houston, TX 78234, USA Received 23 January 2008; received in revised form 2 June 2008; accepted 4 June 2008
Abstract Objective: The purpose of this study was to determine the arch height ratio in a large cohort of subjects as well as to assess the reliability and validity of the foot measurements utilized in the study. Method: Eight hundred and fifty subjects, 393 women and 457 men, consented to participate in the study. The dorsal arch height, total foot length, and the truncated foot length were used to calculate two variations of the arch height ratio. In addition to determining within- and between-rater measurement reliability, radiographs were used to establish validity. Results: The truncated arch height ratio can be estimated using the total foot length, unless toe deformities are present in the individual being assessed. All foot measurements had high levels of intra- and inter-rater reliability and the validity of measuring the dorsal arch height while standing with equal weight on both feet was established. Conclusions: This investigation provides normative values from a large cohort of healthy female and male subjects for two variations of the arch height ratio. The arch height ratio is a reliable and valid measurement that may prove useful to clinicians and researchers for the classification of foot posture. © 2008 Elsevier Ltd. All rights reserved. Keywords: Foot; Posture; Clinical measurement; Reliability; Validity
1. Introduction The clinical diagnosis of foot disorders is often linked to abnormal foot posture. The assessment of static foot posture requires the use of a simple and minimally invasive technique that can easily be performed in the clinical setting with a high level of consistency and validity. The bony arch index (BAI), also described in the literature as the arch index and the arch height index, has been used as a technique by several authors to describe the posture of the medial longitudinal arch [1–6]. The bony arch index is different than the “arch index” that was originally described by Cavanagh and Rodgers, which was based on plantar surface contact area [7].
∗
Corresponding author. Tel.: +1 928 523 1499; fax: +1 928 523 9289. E-mail address: [email protected] (T.G. McPoil).
0958-2592/$ – see front matter © 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.foot.2008.06.002
Cowan et al. was one of the first to describe the BAI as a technique to quantify the height of the medial longitudinal arch using linear measurements rather than plantar surface contact area [1]. Cowan et al. defined the BAI as the ratio of navicular height to foot length, with foot length defined as the distance measured from the heel to the metatarsophalangeal joint. In their study, navicular height was defined as the linear distance from the navicular tuberosity to the supporting surface while the subject stood with full body weight on the foot that was being measured [1]. While the BAI has been widely used to assess the bony height of the medial longitudinal arch, different investigators have utilized different subject positions to obtain the measurement as well as to define foot length. Although Cowan et al. originally described the assessment of navicular height with the subject standing with full body weight on the foot being measured, other investigators have used 50% body
T.G. McPoil et al. / The Foot 18 (2008) 220–227
weight or 90% body weight [2,5,6]. With regard to foot length, while Cowan et al. originally described this measurement as the length from the heel to the metatarsophalangeal joint. Other investigators, however, have defined this as the distance from the most posterior aspect of the calcaneus to the end of the head of the first metatarsal or have used the distance from the posterior point on the calcaneus to the medial side of the first metatarsophalangeal joint [2,5]. These variations in the definition of the foot length used to calculate the BAI as well as the percentage of weight bearing the subject is positioned in for the measurement of navicular height make it very difficult to compare results amongst the various studies as well as to develop a normative database. Another issue related to the BAI is the consistency of measuring navicular height. Evans et al. is one of the few investigators who have reported good levels of reliability for the measurement of navicular height in adults [8]. Using four different raters with between 11 and 15 years of clinical experience, they reported that the within-rater reliability based on intraclass correlation coefficients (ICCs) for the 60 adults ranged from 0.80 to 0.85, with between-rater reliability of 0.76. While Weiner-Ogilvie and Rome reported that navicular height measurements demonstrated the smallest intraobserver and interobserver differences in a group of young adults, the mean differences in navicular height between the two observers who performed the measurements ranged from 17% to 22% [9]. Likewise, Menz and Munteanu found that when assessing navicular height in older adults using multiple raters, the test–retest reliability measured using ICCs was only 0.64 [10]. In a study quite similar to the one performed by Evans et al., Vinicombe et al. reported low to moderate reliability of navicular height measurements using five raters with 3–7 years of clinical experience [11]. The ICC values reported by Vinicombe et al. for within-rater reliability ranged from 0.33 to 0.62 with between-rater reliability between 0.54 and 0.76 over two different measurement sessions. Vinicombe et al. also reported standard error of the measurement (S.E.M.), which is a number in the same units as navicular height that represents how much variation would occur in the measurement if measured more than once [20]. Their S.E.M. results suggested that clinicians could expect a measurement error in the range of 12% to 28% when multiple clinicians are measuring navicular height [11]. One potential factor for the difference in the findings between Evans et al. and Vinicombe et al. could be related to the level of clinical experience of the raters used in the two studies. Evans et al. used raters with 11–15 years of clinical experience, while the raters in the Vinicombe et al. study had only 3–7 years of clinical experience [8,11]. Other factors that can influence the consistency of measuring navicular height between raters include difficulty in locating the navicular tuberosity secondary to anatomical variations among individuals as well as balance related issues that can occur as subjects intermittently change their standing position when attempting to maintain full or 90% of their body weight on the foot being measured [11,12].
221
In an attempt to circumvent the anatomical variability when attempting to palpate the navicular tuberosity as well as clinical experience issues that could effect the consistency of the navicular height measurement, Williams and McClay proposed measuring the height of the dorsum of the foot at 50% of foot length and dividing by either total foot length or truncated foot length [12]. Williams and McClay reported that the dorsum foot height divided by either total or truncated foot length had the highest ICC values of the seven measurements they evaluated. In their study, Williams and McClay assessed dorsal foot height in 10% and 90% weight bearing. Several other researchers have used the dorsal height at 50% foot length divided by the truncated foot length as a way to characterize arch height and have termed the measurement as the arch height ratio or arch height index [13–17]. While both within- and between-rater reliability for the arch height ratio or index has been reported to be high, inconsistencies in the way different researchers have defined the “truncated foot length” as well as the percentage of weight bearing used to obtain the dorsum foot height make comparison of the results from these various studies extremely difficult. Williams and McClay originally defined the truncated foot length as the distance from the most posterior portion of the calcaneus to the center of the first metatarsophalangeal joint, but Franettovich et al. and Vicenzino et al. defined the measure as the distance from the posterior heel surface to the first metatarsophalangeal joint line [12,15,16]. In large feet the difference between measuring to the first metatarsophalangeal joint line versus the center of the joint could be greater than a centimeter, which would affect the calculated ratio. With regard to percentage of weight bearing for measuring dorsal height, Williams and McClay performed the measurement in both 10% and 90% weight bearing and used radiographs to substantiate the validity of these two positions [12]. Several investigators have also reported measuring dorsal foot height with the subject in a relaxed standing position, assuming 50% body weight on each foot [15–17]. Franettovich et al. and Vinicombe et al. have suggested that performing foot measurements in 50% weight bearing improves measurement reliability by decreasing balance related issues that can occur as subjects intermittently change their foot position while attempting to stand with a partial amount of their body weight on the foot being assessed [11,17]. It would be logical to assume that measuring dorsal foot height in 50% weight bearing could affect the height measured in comparison to values obtained in 10% or 90% weight bearing. Unfortunately, Williams and McClay only validated the dorsal height measurement using radiographs in 10% and 90% weight bearing. The arch height ratio or index, using dorsal foot height measured at 50% of foot length, would appear to provide the clinician with a more reliable measurement than navicular height as a result of not having to palpate a bony landmark. The consistency and validity of the dorsal height measurement obtained in different weight-bearing conditions, however, warrants further investigation. Furthermore,
222
T.G. McPoil et al. / The Foot 18 (2008) 220–227
Table 1 Subject demographics (mean ± standard deviation) All subjects (n = 850) Females (n = 393) Males (n = 457)
Age (years)
Height (cm)
Weight (kg)
Body mass index
26.7 ± 6.4 26.1 ± 6.3 27.3 ± 6.5
171.7 ± 9.7 164.9 ± 7.2 171.7 ± 7.6
72.3 ± 14.1 63.2 ± 10.8 80.2 ± 11.7
24.4 ± 3.5 23.2 ± 3.5 25.4 ± 3.2
the largest group for whom the arch height index or ratio has been reported to date is 145 subjects. In their study, Zifchock et al. assessed 68 men and 77 women, using 10% and 50% weight bearing but did not validate the 50% weight-bearing condition [15]. Thus, descriptive data for a large cohort of subjects, including both women and men, would appear to be warranted in order to establish normative values for the arch height index or ratio. The purposes of this study were to (1) determine the reliability and validity of the arch height ratio when measured in bilateral resting standing posture, (2) determine the arch height ratio in a large cohort of subjects using both total foot length and truncated foot length, and (3) determine if extremity differences exist for dorsal arch height and foot or ball length measurements. For this study, the arch height ratio (AHR) was defined as the dorsal arch height at 50% of the total foot length, measured in bilateral resting standing posture, divided by either total foot or truncated foot length.
2. Methods 2.1. Subjects The left and right feet of 850 subjects, 393 women and 457 men, were assessed to establish a mean and standard deviation for a reference population of convenience. All subjects were recruited using advertisements throughout university and military communities. Each participant met the following inclusion criteria: (1) no history of congenital deformity in the lower extremity or foot; (2) no previous history of lower extremity or foot fractures; (3) no systemic diseases that could effect lower extremity or foot posture; (4) no history of trauma or pain to either foot, lower extremity, or lumbosacral region at least 12 months prior to the start of the investigation. Volunteers with any visual signs of hallux valgus, hallux limitus/rigidus, claw toes, or hammer toes were also excluded. Subject demographics are listed in Table 1. The study received ethical approval from the Northern Arizona University Institutional Review Board and the Brooke Army Medical Center Institutional Review Board. All subjects provided informed written consent prior to participation.
Fig. 1. Foot measurement platform.
be obtained (see Fig. 1). Prior to obtaining the standing measurements, each subject is positioned on the standing platform with both heels placed into left and right heel cups that were positioned 15.24 cm apart. Next, the sliding first metatarsophalangeal joint indicator was positioned over the medial prominence of the first metatarsal head. To ensure the proper placement of the indicator over the medial prominence of the first metatarsal head, the examiner ensured that the hallux could be extended without causing any displacement of the indicator (see Fig. 2). Once the first metatarsophalangeal joint indicator was properly positioned bilaterally, the subject was instructed to place equal weight on both feet so that the following weight-bearing measurements could be obtained.
2.2. Procedures Each subject was asked to stand on a specially constructed platform so that the total foot (heel-to-toe) length, the truncated foot (heel-to-ball) length, and dorsal arch height could
Fig. 2. Extending the first metatarsophalangeal joint to ensure proper placement of the indicator over the medial prominence of the first metatarsal head.
T.G. McPoil et al. / The Foot 18 (2008) 220–227
223
Fig. 3. Measurement of the heel-to-toe length using sliding bar.
Heel-to-toe length (HTL) was measured by placing the sliding bar on the centered metal ruler and moving the bar to just touch the longest toe, usually the hallux, of each foot (see Fig. 3). Heel-to-ball length (HBL) was recorded based on the position of the first metatarsophalangeal joint indicator in relation to offset metal rulers that were aligned with the centered metal ruler (see Fig. 4). Finally, the dorsal arch height (DAH) at 50% of total foot length was measured bilaterally. To obtain the 50% length of the foot, the previously measured HTL was divided in half and the dorsum of the foot was marked at the 50% length point using a water-soluble pen. Next, the sliding arm of a vertical digital caliper was posiFig. 5. (A) Alignment of the sliding arm of the caliper to the mark indicating 50% of the length of the foot. (B) Measurement of the dorsal arch height using the digital caliper.
tioned over the 50% height mark and vertical height from the floor to the dorsum of the foot was measured (see Fig. 5). Measurements were obtained on both feet and recorded in centimeters. Using the HTL and HBL, the following two ratios were determined using the DAH: (1) the total foot length arch height ratio (TFL-AHR) was calculated by dividing the DAH by the HTL and (2) the truncated arch height ratio (Trun-AHR) was calculated by dividing the DAH by the HBL. 2.3. Determination of reliability and validity
Fig. 4. Measurement of the heel-to-ball length.
To establish within- and between-rater reliability for the measurements, three raters were asked to assess the left and right feet of 12 randomly selected subjects. The raters performing the measurements were three physical therapists
224
T.G. McPoil et al. / The Foot 18 (2008) 220–227
with a minimum of 2 years clinical experience (mean: 16 years; range: 30–2 years). Each rater attended a 1-h training session to receive verbal instructions as well as to practice the techniques to make sure they were taking the measurements correctly. The reliability data collection consisted of two sessions, 1 week apart, in which each rater performed all three measurements on all 12 subjects. At each session, each rater measured both feet of each subject twice with at least 10 min separating the two sets of measurements. Each rater was blinded from the initial set of measurements to prevent rater bias. Although Williams and McClay have previously reported the validity for the dorsal arch height measurement described in this study, they only assessed radiographic images obtained while the subject stood with 10% and 90% of body weight placed on the foot being assessed [12]. Since the DAH in the current study was measured with equal weight on each foot, we believed it was necessary to establish validity for this method. Using the same foot measurement platform as previously described except with all metal removed, a digital radiograph of the right foot was obtained on 12 subjects. For each radiograph, the X-ray unit was positioned vertical to the supporting surface with the center of the X-ray beam placed just superior to the lateral malleolus. The distance from the X-ray tube to the foot was 101.6 cm and the exposure setting used was 150 mA at 54 kV. To prevent possible magnification and parallax errors when obtaining the measurements from the lateral radiograph, two metal pieces of known length were placed on the top and at the end of the X-ray film to serve as linear calibration references during data analysis. Three copies of each digital radiograph were produced and one of the authors measured the HBL and the DAH at 50% of the total foot length on each copy. Validity was established using the average of the three values for each of the two radiographic measurements compared with the values measured clinically on the same 12 subjects. 2.4. Statistical analysis Type (2,1) intraclass correlation coefficients (ICC) were calculated to determine the consistency of each rater to perform the measurements repeatedly both individually (within-rater) and in comparison to the other raters (betweenrater) [18]. The level of reliability for the ICC was classified using the characterizations reported by Landis and Koch
[19]. These characterizations were: slight, if the correlation ranged from 0.00 to 0.21; fair, if the correlation ranged from 0.21 to 0.40; moderate, if the correlation ranged from 0.41 to 0.60; substantial, is the correlation ranged from 0.61 to 0.80; and almost perfect, if the correlation ranged from 0.81 to 1.00. In addition to ICC values, the S.E.M. was also calculated as another index of the reliability of the measurement. In addition to descriptive statistics, t-tests were used to determine whether extremity differences for females and males existed for the foot length and dorsal height measurements or if differences existed between the clinical and radiographic measures of DAH and BHL. Pearson product moment correlation coefficients were calculated to assess the validity of the DAH and HBL obtained from radiographs with the clinical measurements as well as to examine the relationship between TFL-AHR and Trun-AHR. An alpha level of 0.05 was established for all tests of significance.
3. Results Descriptive statistics for all measurements are listed in Table 2. The intra- and inter-rater ICC and S.E.M. values for all three raters are shown in Table 3. The within-rater reliability ICC for HTL, HBL, and DAH for all raters ranged from 0.98 to 0.99 with S.E.M. values ranging from 0.03 to 0.05 cm. Between-rater reliability ICC for the same measurements ranged from 0.98 to 0.99 with S.E.M. values ranging from 0.04 to 0.07 cm. The mean values for DAH and HBL obtained from the radiographs were 6.72 ± 0.50 and 19.26 ± 1.53 cm, respectively. The mean values for DAH and HBL obtained from the clinical measurements were 6.58 ± 0.52 and 19.00 ± 1.39 cm, respectively. The correlation for DAH measured clinically in relation to the DAH measured from the radiographs was 0.929. Using a paired t-test, no significant difference (t = 0.74; d.f. = 1, 11; p < 0.47) was found between the two DAH measures. The correlation for BHL measured clinically in relation to the BHL measured from the radiographs was 0.985. No significant difference (t = −1.31; d.f. = 1, 11; p < 0.22) was found between the two BHL measures using a paired t-test. The arch height ratios calculated using the clinical measures in comparison to the radiographic measures were 0.300 and 0.352, respectively.
Table 2 Descriptive statistics for all foot measurements (mean ± standard deviation) All subjects
Males
Females
Right (n = 850)
Left (n = 850)
Combined (n = 1700)
Right (n = 457)
Left (n = 457)
Right (n = 393)
Left (n = 393)
HTL (cm) HBL (cm) DAH (cm)
25.78 ± 1.84 18.90 ± 1.39 6.50 ± 0.63
25.81 ± 1.83 18.88 ± 1.38 6.43 ± 0.60
25.79 ± 1.84 18.89 ± 1.39 6.47 ± 0.62
27.01 ± 1.30 19.83 ± 1.01 6.85 ± 0.54
27.04 ± 1.29 19.79 ± 1.01 6.76 ± 0.52
24.34 ± 1.24 17.83. ± 0.93 6.10 ± 0.45
24.39 ± 1.25 17.83 ± 0.93 6.05 ± 0.45
TFL-AHR Trun-AHR
0.253 ± 0.02 0.345 ± 0.03
0.249 ± 0.02 0.341 ± 0.03
0.251 ± 0.02 0.343 ± 0.03
0.254 ± 0.02 0.346 ± 0.03
0.250 ± 0.02 0.342 ± 0.03
0.251 ± 0.02 0.343 ± 0.03
0.248 ± 0.02 0.340 ± 0.03
T.G. McPoil et al. / The Foot 18 (2008) 220–227
225
Table 3 Intra- and inter-rater reliability values for all measurements (mean values in centimeters) Heel-to-toe length (HTL)
Heel-to-ball length (HBL)
Dorsal arch height (DAH)
ICC
Mean
S.E.M.
ICC
Mean
S.E.M.
ICC
Mean
S.E.M.
Intra-rater Rater 1 Rater 2 Rater 3
0.99 0.99 0.99
25.95 26.05 25.96
0.04 0.04 0.04
0.98 0.99 0.99
18.98 18.96 19.05
0.05 0.05 0.05
0.98 0.98 0.98
6.65 6.53 6.52
0.03 0.03 0.03
Inter-rater
0.99
25.99
0.06
0.99
19.00
0.07
0.98
6.52
0.04
S.E.M.: standard error of the measurement.
The results of the t-test indicated that there were no significant differences between the left and right feet for HTL (t = −0.57; d.f. = 1, 784; p < 0.57), HBL (t = 0.20; d.f. = 1, 784; p < 0.84), or DAH (t = 1.54; d.f. = 1, 784; p < 0.12) for the female subjects. For the male subjects, there was a significant difference between the left and right feet for DAH (t = 2.66; d.f. = 1, 912; p < 0.008) but not for HTL (t = −0.31; d.f. = 1, 912; p < 0.75) and HBL (t = 0.52; d.f. = 1, 912; p < 0.60). The difference between the left and right feet for DAH for the males was 0.09 cm. The correlation between HTL and HBL was 0.973 (r2 = 0.947) for the left foot and 0.971 (r2 = 0.943) for the right foot. The correlation between TFL-AHR and Trun-AHR for the left foot was 0.977 (r2 = 0.955) and 0.975 (r2 = 0.950) for the right foot. The correlation between HTL and HBL for all 1700 feet was 0.972 (r2 = 0.945). The correlation between DAH/HBL and DAH/HTL was 0.976 (r2 = 0.952). The regression model for predicting HBL using HTL for the LEFT foot is: −0.018941 + 0.7320418 × Left HTL. To predict the HBL using HTL for the RIGHT foot, the regression model is: −0.026049 + 0.7343511 × Right HTL.
4. Discussion The purposes of this study were threefold. They included (1) determine the reliability and validity of the arch height ratio when measured in bilateral resting standing posture, (2) determine the arch height ratio in a large cohort of subjects using both total foot length and truncated foot length, and (3) determine if extremity differences existed for dorsal arch height and foot or ball length measurements. Our first concern in interpreting the results was the reliability and validity of the measurement techniques used in this study. The ICC values for all three raters, regardless of the number of years of clinical experience, would be classified as “almost perfect” for both intra- and inter-rater reliability based on the characterizations provided by Landis and Koch [19]. Just as importantly the S.E.M. values, which denotes the way a score will vary if measured more than once, ranged from 0.03 to 0.07 cm for the dorsal arch height and for both foot length measurements. In addition to the high levels of measurement reliability, the assessment of the DAH when measured in relaxed standing with equal body
weight placed on each foot would appear to be valid based on using radiographs as the “gold standard” for comparison. The results indicate the clinical measurements of DAH and HBL when compared to radiographs are highly correlated. Based on these findings, the authors concluded that the reliability and validity of the measurement techniques used in the study were acceptable and that further analysis of the results could be performed. The results of the t-test indicate that the HTL, HBL, and DAH were not different between the left and right feet for the 393 female subjects. While HTL and HBL were also not different between the left and right feet for the 457 male subjects, DAH was significantly different between feet. The mean difference in DAH between feet for the males was less than 0.1 cm, which would not be considered a major difference from a clinical perspective. Thus, it can be concluded that minimal gender differences exist between the left and right feet for the measurements of HTL, HBL, and DAH. In discussing the relationship between the two arch height ratios calculated in this study using both total foot length and truncated foot length, since minimal differences exist between extremities, the left and right feet could be combined to create a larger data set. Menz, however, has previously reported on the several issues associated with the combining the left and right feet to create a larger data set for statistical analyses [21]. One of the issues noted by Menz is the counting of both the left and right feet as single independent observations so that the researcher can artificially increase the size of their data by counting the same subject twice [21]. Menz further states that it could be problematic to conduct research on individual feet rather than people since the manner in which an individual foot functions is at least partly dependent on the person attached to the foot [21]. As such, the remainder of the discussion on the results of this study will focus on data specific to the 850 left or right feet. For both the left and right feet, the HBL is highly correlated to HTL. HTL explains approximately 95% of the variance of the HBL for the left foot and 94% of the variance for the right foot. Based on these findings, the clinician could measure HTL and use the regression model provided in Section 3 to predict HBL unless the subject or patient had obvious digital deformities such as hallux valgus. In these cases, the clinician would have to specifically measure the
226
T.G. McPoil et al. / The Foot 18 (2008) 220–227
HBL rather than use the regression model provided to predict HBL using the HTL. The mean value for Trun-AHR was 0.345 for the left and 0.341 for the right. The mean value for the TFL-AHR was 0.249 for the left and 0.253 for the right. The results also indicate that the Trun-AHR is strongly correlated with TFL-AHR. The TFL-AHR explains approximately 95% of the variance of the Trun-AHR for both the left and right feet. Based on these findings, one would question the need to calculate the Arch Height Ratio using the truncated foot length since the arch height ratio calculated using the total foot length could be used to determine both ratios. When Cowan et al. originally described using the truncated length of the foot to calculate the bony arch index, they provided little justification for using this length other than stating that this measurement had the highest degree of association with risk of injury in military recruits [1]. They also provided minimal information on any of the other foot measurements they evaluated in their study. Williams and McClay noted that the use of the truncated foot length reduces the impact that toe deformities, such as claw toes and hallux valgus, can have on heel to longest toe foot length [12]. The fact that this study has shown that Trun-AHR is strongly correlated to TFL-AHR allows the clinician to determine an arch height ratio using either HTL or HBL, unless the subject or patient had obvious digital deformities such as hallux valgus or claw toes. It is important to point out that while the arch height ratio can be determined using either HTL or HBL, the arch height ratios calculated using HTL or HBL are different numerically and are not interchangeable. The Trun-AHR values reported in the current study are quite similar to values previously reported by Vicenzino et al. [16] and Zifchock et al. [15]. In the current study, the mean Trun-AHR was 0.345 for the right foot, 0.341 for the left foot. The mean Trun-AHR reported by Vicenzino et al. was 0.349, based on 17 subjects, while the Trun-AHR in the Zifchock et al. study was .340 for the 68 male subjects and 0.336 for the 77 females subjects. It is important to note both Vicenzino and Zifchock combined the left and right feet to create a larger data pool for analysis. In addition, Zifchock et al. could not measure both feet simultaneously with the subject standing since the measurement apparatus they used only permitted the measurement of one foot at a time. Previous investigations have not evaluated FL-AHR with subjects standing, placing equal weight on each foot. A limitation of the current study was the assumption by the authors that when subjects were asked to stand and place equal weight on each foot, that the subject was placing 50% of their body weight on each foot. While various methodologies could have been utilized to ensure that each subject was placing 50% of their body weight on each foot while the measurements in standing were being recorded, for example, having the subject stand with one foot on a scale, the methodology used in the current study can be easily replicated by clinicians. Tesser et al. has previously reported that the amount of asymmetry in weight distribution between extremities in relaxed standing is 4% or less in healthy sub-
jects [22]. Furthermore, while there could be slight variations in weight-bearing symmetry between extremities when a subject is asked to stand with equal weight placed on both feet, the high level of repeatability of the foot posture measurements utilized in this study would suggest that any degree of asymmetry is negligible. 5. Conclusions In summary, the findings of the current study provide normative values for the Trun-AHR and the TFL-AHR for the left and right foot in a large cohort of healthy male and female subjects. While all three foot measurements assessed in this study were found to have excellent levels of intraand inter-rater reliability, the validity of measuring DAH and the truncated foot length (HBL) while asking the subject to stand and place equal weight on both feet was also established. Based on these results, clinicians can assess the AHR using the total foot length to estimate the truncated foot length unless toe deformities were present in the individual being assessed. While the foot measurements assessed in the current study were found not to be different between the left and right feet in females, the DAH was significantly different between feet in males. It is important to note, however, that the mean difference in DAH between feet for the males was less than 0.1 cm. While further research is always warranted, the results of this study provide clinicians and researchers with normative values for the AHR in a large group of women and men that can be used for assessing and classifying foot structure. Conflict of interest None.
References [1] Cowan DH, Jones BH, Robinson JR. Foot morphologic characteristics and risk of exercise-related injury. Arch Fam Med 1993;2:773–7. [2] Saltzman CL, Nawoczenski DA, Talbot KD. Measurement of the medial longitudinal arch. Arch Phys Med Rehabil 1995;76:45–9. [3] Yen DY, Puffer JC, Schmalzried TP. Injuries in runners: a prospective study of alignment. Clin J Sport Med 1998;8:187–94. [4] Wearing SC, Urry S, Perlman PR, Dubois P, Smeathers JE. Serial measurement of the calcaneal pitch during midstance. J Am Podiatr Med Assoc 1999;89:188–93. [5] Lees A, Lake M, Klenerman L. Shock absorption during forefoot running and its relationship to medial longitudinal arch height. Foot Ankle Int 2005;26:1081–8. [6] 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. [7] Cavanagh P, Rodgers M. The arch index: a useful measure from footprints. J Biomech 1987;20:547–51. [8] Evans AM, Copper AW, Scharfbillig RW, Scutter SD, Williams MT. Reliability of the foot posture index and traditional measures of foot position. J Am Podiatr Med Assoc 2003;93:2003–213.
T.G. McPoil et al. / The Foot 18 (2008) 220–227 [9] Weiner-Ogilvie S, Rome K. The reliability of three techniques for measuring foot position. J Am Podiatr Med Assoc 1998;88:381–6. [10] Menz HB, Munteanu SE. Validity of 3 clinical techniques for the measurement of static foot posture in older people. J Orthop Sports Phys Ther 2005;35:479–86. [11] Vinicombe A, Raspovic A, Menz HB. Reliability of navicular displacement measurement as a clinical indicator of foot posture. J Am Podiatr Med Assoc 2001;91:262–8. [12] Williams DS, McClay IS. Measurements used to characterize the foot and the medial longitudinal arch: Reliability and validity. Phys Ther 2000;80:864–71. [13] Williams III DS, McClay IS, Hamill J. Arch structure and injury patterns in runners. Clin Biomech 2001;16:341–7. [14] Williams III DS, Davis IM, Scholz JP, Hamill J, Buchanan TS. Higharched runners exhibit increased leg stiffness compared it low-arched runners. Gait Posture 2004;19:263–9. [15] Zifchock RA, Davis I, Hillstrom H, Song J. The effect of gender, age, and lateral dominance on arch height and arch stiffness. Foot Ankle Int 2006;27:367–72.
227
[16] Vicenzino B, Franettovich M, McPoil T, Russell T, Skardoon G. Initial effects of anti-pronation tape on medial longitudinal arch during walking and running. Br J Sports Med 2005;39:939–43. [17] Franettovich MM, McPoil TG, Russell T, Skardoon G, Vicenzino B. The ability to predict dynamic foot posture from static measurements. J Am Podiatr Med Assoc 2007;97:115–20. [18] Shrout PE, Fleiss JL. Intraclass correlations: uses in assessing rater reliability. Psychol Bull 1979;86:420–8. [19] Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics 1977;33:159–74. [20] Rothstein JM. Measurement and clinical practice: the theory and application. In: Rothstein JM, editor. Measurement in physical therapy. New York: Churchill Livingstone; 1985. p. 1–46. [21] Menz HB. Two feet, or one person? Problems associated with statistical analysis of paired data in foot and ankle medicine. Foot 2004;14:2–5. [22] Tesser 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.
The Foot 18 (2008) 220–227
Effect of using truncated versus total foot length to calculate the arch height ratio Thomas G. McPoil a,∗ , Mark W. Cornwall a , Bill Vicenzino b , Deydre S. Teyhen c , Joseph M. Molloy c , Douglas S. Christie c , Natalie Collins b a
Gait Research Laboratory, Department of Physical Therapy & Athletic Training, Northern Arizona University, Flagstaff, AZ 86011, USA b Department of Physiotherapy, University of Queensland, St. Lucia, Brisbane, Queensland 4072, Australia c U.S. Army-Baylor University Doctor Program in Physical Therapy, Foot Sam Houston, TX 78234, USA Received 23 January 2008; received in revised form 2 June 2008; accepted 4 June 2008
Abstract Objective: The purpose of this study was to determine the arch height ratio in a large cohort of subjects as well as to assess the reliability and validity of the foot measurements utilized in the study. Method: Eight hundred and fifty subjects, 393 women and 457 men, consented to participate in the study. The dorsal arch height, total foot length, and the truncated foot length were used to calculate two variations of the arch height ratio. In addition to determining within- and between-rater measurement reliability, radiographs were used to establish validity. Results: The truncated arch height ratio can be estimated using the total foot length, unless toe deformities are present in the individual being assessed. All foot measurements had high levels of intra- and inter-rater reliability and the validity of measuring the dorsal arch height while standing with equal weight on both feet was established. Conclusions: This investigation provides normative values from a large cohort of healthy female and male subjects for two variations of the arch height ratio. The arch height ratio is a reliable and valid measurement that may prove useful to clinicians and researchers for the classification of foot posture. © 2008 Elsevier Ltd. All rights reserved. Keywords: Foot; Posture; Clinical measurement; Reliability; Validity
1. Introduction The clinical diagnosis of foot disorders is often linked to abnormal foot posture. The assessment of static foot posture requires the use of a simple and minimally invasive technique that can easily be performed in the clinical setting with a high level of consistency and validity. The bony arch index (BAI), also described in the literature as the arch index and the arch height index, has been used as a technique by several authors to describe the posture of the medial longitudinal arch [1–6]. The bony arch index is different than the “arch index” that was originally described by Cavanagh and Rodgers, which was based on plantar surface contact area [7].
∗
Corresponding author. Tel.: +1 928 523 1499; fax: +1 928 523 9289. E-mail address: [email protected] (T.G. McPoil).
0958-2592/$ – see front matter © 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.foot.2008.06.002
Cowan et al. was one of the first to describe the BAI as a technique to quantify the height of the medial longitudinal arch using linear measurements rather than plantar surface contact area [1]. Cowan et al. defined the BAI as the ratio of navicular height to foot length, with foot length defined as the distance measured from the heel to the metatarsophalangeal joint. In their study, navicular height was defined as the linear distance from the navicular tuberosity to the supporting surface while the subject stood with full body weight on the foot that was being measured [1]. While the BAI has been widely used to assess the bony height of the medial longitudinal arch, different investigators have utilized different subject positions to obtain the measurement as well as to define foot length. Although Cowan et al. originally described the assessment of navicular height with the subject standing with full body weight on the foot being measured, other investigators have used 50% body
T.G. McPoil et al. / The Foot 18 (2008) 220–227
weight or 90% body weight [2,5,6]. With regard to foot length, while Cowan et al. originally described this measurement as the length from the heel to the metatarsophalangeal joint. Other investigators, however, have defined this as the distance from the most posterior aspect of the calcaneus to the end of the head of the first metatarsal or have used the distance from the posterior point on the calcaneus to the medial side of the first metatarsophalangeal joint [2,5]. These variations in the definition of the foot length used to calculate the BAI as well as the percentage of weight bearing the subject is positioned in for the measurement of navicular height make it very difficult to compare results amongst the various studies as well as to develop a normative database. Another issue related to the BAI is the consistency of measuring navicular height. Evans et al. is one of the few investigators who have reported good levels of reliability for the measurement of navicular height in adults [8]. Using four different raters with between 11 and 15 years of clinical experience, they reported that the within-rater reliability based on intraclass correlation coefficients (ICCs) for the 60 adults ranged from 0.80 to 0.85, with between-rater reliability of 0.76. While Weiner-Ogilvie and Rome reported that navicular height measurements demonstrated the smallest intraobserver and interobserver differences in a group of young adults, the mean differences in navicular height between the two observers who performed the measurements ranged from 17% to 22% [9]. Likewise, Menz and Munteanu found that when assessing navicular height in older adults using multiple raters, the test–retest reliability measured using ICCs was only 0.64 [10]. In a study quite similar to the one performed by Evans et al., Vinicombe et al. reported low to moderate reliability of navicular height measurements using five raters with 3–7 years of clinical experience [11]. The ICC values reported by Vinicombe et al. for within-rater reliability ranged from 0.33 to 0.62 with between-rater reliability between 0.54 and 0.76 over two different measurement sessions. Vinicombe et al. also reported standard error of the measurement (S.E.M.), which is a number in the same units as navicular height that represents how much variation would occur in the measurement if measured more than once [20]. Their S.E.M. results suggested that clinicians could expect a measurement error in the range of 12% to 28% when multiple clinicians are measuring navicular height [11]. One potential factor for the difference in the findings between Evans et al. and Vinicombe et al. could be related to the level of clinical experience of the raters used in the two studies. Evans et al. used raters with 11–15 years of clinical experience, while the raters in the Vinicombe et al. study had only 3–7 years of clinical experience [8,11]. Other factors that can influence the consistency of measuring navicular height between raters include difficulty in locating the navicular tuberosity secondary to anatomical variations among individuals as well as balance related issues that can occur as subjects intermittently change their standing position when attempting to maintain full or 90% of their body weight on the foot being measured [11,12].
221
In an attempt to circumvent the anatomical variability when attempting to palpate the navicular tuberosity as well as clinical experience issues that could effect the consistency of the navicular height measurement, Williams and McClay proposed measuring the height of the dorsum of the foot at 50% of foot length and dividing by either total foot length or truncated foot length [12]. Williams and McClay reported that the dorsum foot height divided by either total or truncated foot length had the highest ICC values of the seven measurements they evaluated. In their study, Williams and McClay assessed dorsal foot height in 10% and 90% weight bearing. Several other researchers have used the dorsal height at 50% foot length divided by the truncated foot length as a way to characterize arch height and have termed the measurement as the arch height ratio or arch height index [13–17]. While both within- and between-rater reliability for the arch height ratio or index has been reported to be high, inconsistencies in the way different researchers have defined the “truncated foot length” as well as the percentage of weight bearing used to obtain the dorsum foot height make comparison of the results from these various studies extremely difficult. Williams and McClay originally defined the truncated foot length as the distance from the most posterior portion of the calcaneus to the center of the first metatarsophalangeal joint, but Franettovich et al. and Vicenzino et al. defined the measure as the distance from the posterior heel surface to the first metatarsophalangeal joint line [12,15,16]. In large feet the difference between measuring to the first metatarsophalangeal joint line versus the center of the joint could be greater than a centimeter, which would affect the calculated ratio. With regard to percentage of weight bearing for measuring dorsal height, Williams and McClay performed the measurement in both 10% and 90% weight bearing and used radiographs to substantiate the validity of these two positions [12]. Several investigators have also reported measuring dorsal foot height with the subject in a relaxed standing position, assuming 50% body weight on each foot [15–17]. Franettovich et al. and Vinicombe et al. have suggested that performing foot measurements in 50% weight bearing improves measurement reliability by decreasing balance related issues that can occur as subjects intermittently change their foot position while attempting to stand with a partial amount of their body weight on the foot being assessed [11,17]. It would be logical to assume that measuring dorsal foot height in 50% weight bearing could affect the height measured in comparison to values obtained in 10% or 90% weight bearing. Unfortunately, Williams and McClay only validated the dorsal height measurement using radiographs in 10% and 90% weight bearing. The arch height ratio or index, using dorsal foot height measured at 50% of foot length, would appear to provide the clinician with a more reliable measurement than navicular height as a result of not having to palpate a bony landmark. The consistency and validity of the dorsal height measurement obtained in different weight-bearing conditions, however, warrants further investigation. Furthermore,
222
T.G. McPoil et al. / The Foot 18 (2008) 220–227
Table 1 Subject demographics (mean ± standard deviation) All subjects (n = 850) Females (n = 393) Males (n = 457)
Age (years)
Height (cm)
Weight (kg)
Body mass index
26.7 ± 6.4 26.1 ± 6.3 27.3 ± 6.5
171.7 ± 9.7 164.9 ± 7.2 171.7 ± 7.6
72.3 ± 14.1 63.2 ± 10.8 80.2 ± 11.7
24.4 ± 3.5 23.2 ± 3.5 25.4 ± 3.2
the largest group for whom the arch height index or ratio has been reported to date is 145 subjects. In their study, Zifchock et al. assessed 68 men and 77 women, using 10% and 50% weight bearing but did not validate the 50% weight-bearing condition [15]. Thus, descriptive data for a large cohort of subjects, including both women and men, would appear to be warranted in order to establish normative values for the arch height index or ratio. The purposes of this study were to (1) determine the reliability and validity of the arch height ratio when measured in bilateral resting standing posture, (2) determine the arch height ratio in a large cohort of subjects using both total foot length and truncated foot length, and (3) determine if extremity differences exist for dorsal arch height and foot or ball length measurements. For this study, the arch height ratio (AHR) was defined as the dorsal arch height at 50% of the total foot length, measured in bilateral resting standing posture, divided by either total foot or truncated foot length.
2. Methods 2.1. Subjects The left and right feet of 850 subjects, 393 women and 457 men, were assessed to establish a mean and standard deviation for a reference population of convenience. All subjects were recruited using advertisements throughout university and military communities. Each participant met the following inclusion criteria: (1) no history of congenital deformity in the lower extremity or foot; (2) no previous history of lower extremity or foot fractures; (3) no systemic diseases that could effect lower extremity or foot posture; (4) no history of trauma or pain to either foot, lower extremity, or lumbosacral region at least 12 months prior to the start of the investigation. Volunteers with any visual signs of hallux valgus, hallux limitus/rigidus, claw toes, or hammer toes were also excluded. Subject demographics are listed in Table 1. The study received ethical approval from the Northern Arizona University Institutional Review Board and the Brooke Army Medical Center Institutional Review Board. All subjects provided informed written consent prior to participation.
Fig. 1. Foot measurement platform.
be obtained (see Fig. 1). Prior to obtaining the standing measurements, each subject is positioned on the standing platform with both heels placed into left and right heel cups that were positioned 15.24 cm apart. Next, the sliding first metatarsophalangeal joint indicator was positioned over the medial prominence of the first metatarsal head. To ensure the proper placement of the indicator over the medial prominence of the first metatarsal head, the examiner ensured that the hallux could be extended without causing any displacement of the indicator (see Fig. 2). Once the first metatarsophalangeal joint indicator was properly positioned bilaterally, the subject was instructed to place equal weight on both feet so that the following weight-bearing measurements could be obtained.
2.2. Procedures Each subject was asked to stand on a specially constructed platform so that the total foot (heel-to-toe) length, the truncated foot (heel-to-ball) length, and dorsal arch height could
Fig. 2. Extending the first metatarsophalangeal joint to ensure proper placement of the indicator over the medial prominence of the first metatarsal head.
T.G. McPoil et al. / The Foot 18 (2008) 220–227
223
Fig. 3. Measurement of the heel-to-toe length using sliding bar.
Heel-to-toe length (HTL) was measured by placing the sliding bar on the centered metal ruler and moving the bar to just touch the longest toe, usually the hallux, of each foot (see Fig. 3). Heel-to-ball length (HBL) was recorded based on the position of the first metatarsophalangeal joint indicator in relation to offset metal rulers that were aligned with the centered metal ruler (see Fig. 4). Finally, the dorsal arch height (DAH) at 50% of total foot length was measured bilaterally. To obtain the 50% length of the foot, the previously measured HTL was divided in half and the dorsum of the foot was marked at the 50% length point using a water-soluble pen. Next, the sliding arm of a vertical digital caliper was posiFig. 5. (A) Alignment of the sliding arm of the caliper to the mark indicating 50% of the length of the foot. (B) Measurement of the dorsal arch height using the digital caliper.
tioned over the 50% height mark and vertical height from the floor to the dorsum of the foot was measured (see Fig. 5). Measurements were obtained on both feet and recorded in centimeters. Using the HTL and HBL, the following two ratios were determined using the DAH: (1) the total foot length arch height ratio (TFL-AHR) was calculated by dividing the DAH by the HTL and (2) the truncated arch height ratio (Trun-AHR) was calculated by dividing the DAH by the HBL. 2.3. Determination of reliability and validity
Fig. 4. Measurement of the heel-to-ball length.
To establish within- and between-rater reliability for the measurements, three raters were asked to assess the left and right feet of 12 randomly selected subjects. The raters performing the measurements were three physical therapists
224
T.G. McPoil et al. / The Foot 18 (2008) 220–227
with a minimum of 2 years clinical experience (mean: 16 years; range: 30–2 years). Each rater attended a 1-h training session to receive verbal instructions as well as to practice the techniques to make sure they were taking the measurements correctly. The reliability data collection consisted of two sessions, 1 week apart, in which each rater performed all three measurements on all 12 subjects. At each session, each rater measured both feet of each subject twice with at least 10 min separating the two sets of measurements. Each rater was blinded from the initial set of measurements to prevent rater bias. Although Williams and McClay have previously reported the validity for the dorsal arch height measurement described in this study, they only assessed radiographic images obtained while the subject stood with 10% and 90% of body weight placed on the foot being assessed [12]. Since the DAH in the current study was measured with equal weight on each foot, we believed it was necessary to establish validity for this method. Using the same foot measurement platform as previously described except with all metal removed, a digital radiograph of the right foot was obtained on 12 subjects. For each radiograph, the X-ray unit was positioned vertical to the supporting surface with the center of the X-ray beam placed just superior to the lateral malleolus. The distance from the X-ray tube to the foot was 101.6 cm and the exposure setting used was 150 mA at 54 kV. To prevent possible magnification and parallax errors when obtaining the measurements from the lateral radiograph, two metal pieces of known length were placed on the top and at the end of the X-ray film to serve as linear calibration references during data analysis. Three copies of each digital radiograph were produced and one of the authors measured the HBL and the DAH at 50% of the total foot length on each copy. Validity was established using the average of the three values for each of the two radiographic measurements compared with the values measured clinically on the same 12 subjects. 2.4. Statistical analysis Type (2,1) intraclass correlation coefficients (ICC) were calculated to determine the consistency of each rater to perform the measurements repeatedly both individually (within-rater) and in comparison to the other raters (betweenrater) [18]. The level of reliability for the ICC was classified using the characterizations reported by Landis and Koch
[19]. These characterizations were: slight, if the correlation ranged from 0.00 to 0.21; fair, if the correlation ranged from 0.21 to 0.40; moderate, if the correlation ranged from 0.41 to 0.60; substantial, is the correlation ranged from 0.61 to 0.80; and almost perfect, if the correlation ranged from 0.81 to 1.00. In addition to ICC values, the S.E.M. was also calculated as another index of the reliability of the measurement. In addition to descriptive statistics, t-tests were used to determine whether extremity differences for females and males existed for the foot length and dorsal height measurements or if differences existed between the clinical and radiographic measures of DAH and BHL. Pearson product moment correlation coefficients were calculated to assess the validity of the DAH and HBL obtained from radiographs with the clinical measurements as well as to examine the relationship between TFL-AHR and Trun-AHR. An alpha level of 0.05 was established for all tests of significance.
3. Results Descriptive statistics for all measurements are listed in Table 2. The intra- and inter-rater ICC and S.E.M. values for all three raters are shown in Table 3. The within-rater reliability ICC for HTL, HBL, and DAH for all raters ranged from 0.98 to 0.99 with S.E.M. values ranging from 0.03 to 0.05 cm. Between-rater reliability ICC for the same measurements ranged from 0.98 to 0.99 with S.E.M. values ranging from 0.04 to 0.07 cm. The mean values for DAH and HBL obtained from the radiographs were 6.72 ± 0.50 and 19.26 ± 1.53 cm, respectively. The mean values for DAH and HBL obtained from the clinical measurements were 6.58 ± 0.52 and 19.00 ± 1.39 cm, respectively. The correlation for DAH measured clinically in relation to the DAH measured from the radiographs was 0.929. Using a paired t-test, no significant difference (t = 0.74; d.f. = 1, 11; p < 0.47) was found between the two DAH measures. The correlation for BHL measured clinically in relation to the BHL measured from the radiographs was 0.985. No significant difference (t = −1.31; d.f. = 1, 11; p < 0.22) was found between the two BHL measures using a paired t-test. The arch height ratios calculated using the clinical measures in comparison to the radiographic measures were 0.300 and 0.352, respectively.
Table 2 Descriptive statistics for all foot measurements (mean ± standard deviation) All subjects
Males
Females
Right (n = 850)
Left (n = 850)
Combined (n = 1700)
Right (n = 457)
Left (n = 457)
Right (n = 393)
Left (n = 393)
HTL (cm) HBL (cm) DAH (cm)
25.78 ± 1.84 18.90 ± 1.39 6.50 ± 0.63
25.81 ± 1.83 18.88 ± 1.38 6.43 ± 0.60
25.79 ± 1.84 18.89 ± 1.39 6.47 ± 0.62
27.01 ± 1.30 19.83 ± 1.01 6.85 ± 0.54
27.04 ± 1.29 19.79 ± 1.01 6.76 ± 0.52
24.34 ± 1.24 17.83. ± 0.93 6.10 ± 0.45
24.39 ± 1.25 17.83 ± 0.93 6.05 ± 0.45
TFL-AHR Trun-AHR
0.253 ± 0.02 0.345 ± 0.03
0.249 ± 0.02 0.341 ± 0.03
0.251 ± 0.02 0.343 ± 0.03
0.254 ± 0.02 0.346 ± 0.03
0.250 ± 0.02 0.342 ± 0.03
0.251 ± 0.02 0.343 ± 0.03
0.248 ± 0.02 0.340 ± 0.03
T.G. McPoil et al. / The Foot 18 (2008) 220–227
225
Table 3 Intra- and inter-rater reliability values for all measurements (mean values in centimeters) Heel-to-toe length (HTL)
Heel-to-ball length (HBL)
Dorsal arch height (DAH)
ICC
Mean
S.E.M.
ICC
Mean
S.E.M.
ICC
Mean
S.E.M.
Intra-rater Rater 1 Rater 2 Rater 3
0.99 0.99 0.99
25.95 26.05 25.96
0.04 0.04 0.04
0.98 0.99 0.99
18.98 18.96 19.05
0.05 0.05 0.05
0.98 0.98 0.98
6.65 6.53 6.52
0.03 0.03 0.03
Inter-rater
0.99
25.99
0.06
0.99
19.00
0.07
0.98
6.52
0.04
S.E.M.: standard error of the measurement.
The results of the t-test indicated that there were no significant differences between the left and right feet for HTL (t = −0.57; d.f. = 1, 784; p < 0.57), HBL (t = 0.20; d.f. = 1, 784; p < 0.84), or DAH (t = 1.54; d.f. = 1, 784; p < 0.12) for the female subjects. For the male subjects, there was a significant difference between the left and right feet for DAH (t = 2.66; d.f. = 1, 912; p < 0.008) but not for HTL (t = −0.31; d.f. = 1, 912; p < 0.75) and HBL (t = 0.52; d.f. = 1, 912; p < 0.60). The difference between the left and right feet for DAH for the males was 0.09 cm. The correlation between HTL and HBL was 0.973 (r2 = 0.947) for the left foot and 0.971 (r2 = 0.943) for the right foot. The correlation between TFL-AHR and Trun-AHR for the left foot was 0.977 (r2 = 0.955) and 0.975 (r2 = 0.950) for the right foot. The correlation between HTL and HBL for all 1700 feet was 0.972 (r2 = 0.945). The correlation between DAH/HBL and DAH/HTL was 0.976 (r2 = 0.952). The regression model for predicting HBL using HTL for the LEFT foot is: −0.018941 + 0.7320418 × Left HTL. To predict the HBL using HTL for the RIGHT foot, the regression model is: −0.026049 + 0.7343511 × Right HTL.
4. Discussion The purposes of this study were threefold. They included (1) determine the reliability and validity of the arch height ratio when measured in bilateral resting standing posture, (2) determine the arch height ratio in a large cohort of subjects using both total foot length and truncated foot length, and (3) determine if extremity differences existed for dorsal arch height and foot or ball length measurements. Our first concern in interpreting the results was the reliability and validity of the measurement techniques used in this study. The ICC values for all three raters, regardless of the number of years of clinical experience, would be classified as “almost perfect” for both intra- and inter-rater reliability based on the characterizations provided by Landis and Koch [19]. Just as importantly the S.E.M. values, which denotes the way a score will vary if measured more than once, ranged from 0.03 to 0.07 cm for the dorsal arch height and for both foot length measurements. In addition to the high levels of measurement reliability, the assessment of the DAH when measured in relaxed standing with equal body
weight placed on each foot would appear to be valid based on using radiographs as the “gold standard” for comparison. The results indicate the clinical measurements of DAH and HBL when compared to radiographs are highly correlated. Based on these findings, the authors concluded that the reliability and validity of the measurement techniques used in the study were acceptable and that further analysis of the results could be performed. The results of the t-test indicate that the HTL, HBL, and DAH were not different between the left and right feet for the 393 female subjects. While HTL and HBL were also not different between the left and right feet for the 457 male subjects, DAH was significantly different between feet. The mean difference in DAH between feet for the males was less than 0.1 cm, which would not be considered a major difference from a clinical perspective. Thus, it can be concluded that minimal gender differences exist between the left and right feet for the measurements of HTL, HBL, and DAH. In discussing the relationship between the two arch height ratios calculated in this study using both total foot length and truncated foot length, since minimal differences exist between extremities, the left and right feet could be combined to create a larger data set. Menz, however, has previously reported on the several issues associated with the combining the left and right feet to create a larger data set for statistical analyses [21]. One of the issues noted by Menz is the counting of both the left and right feet as single independent observations so that the researcher can artificially increase the size of their data by counting the same subject twice [21]. Menz further states that it could be problematic to conduct research on individual feet rather than people since the manner in which an individual foot functions is at least partly dependent on the person attached to the foot [21]. As such, the remainder of the discussion on the results of this study will focus on data specific to the 850 left or right feet. For both the left and right feet, the HBL is highly correlated to HTL. HTL explains approximately 95% of the variance of the HBL for the left foot and 94% of the variance for the right foot. Based on these findings, the clinician could measure HTL and use the regression model provided in Section 3 to predict HBL unless the subject or patient had obvious digital deformities such as hallux valgus. In these cases, the clinician would have to specifically measure the
226
T.G. McPoil et al. / The Foot 18 (2008) 220–227
HBL rather than use the regression model provided to predict HBL using the HTL. The mean value for Trun-AHR was 0.345 for the left and 0.341 for the right. The mean value for the TFL-AHR was 0.249 for the left and 0.253 for the right. The results also indicate that the Trun-AHR is strongly correlated with TFL-AHR. The TFL-AHR explains approximately 95% of the variance of the Trun-AHR for both the left and right feet. Based on these findings, one would question the need to calculate the Arch Height Ratio using the truncated foot length since the arch height ratio calculated using the total foot length could be used to determine both ratios. When Cowan et al. originally described using the truncated length of the foot to calculate the bony arch index, they provided little justification for using this length other than stating that this measurement had the highest degree of association with risk of injury in military recruits [1]. They also provided minimal information on any of the other foot measurements they evaluated in their study. Williams and McClay noted that the use of the truncated foot length reduces the impact that toe deformities, such as claw toes and hallux valgus, can have on heel to longest toe foot length [12]. The fact that this study has shown that Trun-AHR is strongly correlated to TFL-AHR allows the clinician to determine an arch height ratio using either HTL or HBL, unless the subject or patient had obvious digital deformities such as hallux valgus or claw toes. It is important to point out that while the arch height ratio can be determined using either HTL or HBL, the arch height ratios calculated using HTL or HBL are different numerically and are not interchangeable. The Trun-AHR values reported in the current study are quite similar to values previously reported by Vicenzino et al. [16] and Zifchock et al. [15]. In the current study, the mean Trun-AHR was 0.345 for the right foot, 0.341 for the left foot. The mean Trun-AHR reported by Vicenzino et al. was 0.349, based on 17 subjects, while the Trun-AHR in the Zifchock et al. study was .340 for the 68 male subjects and 0.336 for the 77 females subjects. It is important to note both Vicenzino and Zifchock combined the left and right feet to create a larger data pool for analysis. In addition, Zifchock et al. could not measure both feet simultaneously with the subject standing since the measurement apparatus they used only permitted the measurement of one foot at a time. Previous investigations have not evaluated FL-AHR with subjects standing, placing equal weight on each foot. A limitation of the current study was the assumption by the authors that when subjects were asked to stand and place equal weight on each foot, that the subject was placing 50% of their body weight on each foot. While various methodologies could have been utilized to ensure that each subject was placing 50% of their body weight on each foot while the measurements in standing were being recorded, for example, having the subject stand with one foot on a scale, the methodology used in the current study can be easily replicated by clinicians. Tesser et al. has previously reported that the amount of asymmetry in weight distribution between extremities in relaxed standing is 4% or less in healthy sub-
jects [22]. Furthermore, while there could be slight variations in weight-bearing symmetry between extremities when a subject is asked to stand with equal weight placed on both feet, the high level of repeatability of the foot posture measurements utilized in this study would suggest that any degree of asymmetry is negligible. 5. Conclusions In summary, the findings of the current study provide normative values for the Trun-AHR and the TFL-AHR for the left and right foot in a large cohort of healthy male and female subjects. While all three foot measurements assessed in this study were found to have excellent levels of intraand inter-rater reliability, the validity of measuring DAH and the truncated foot length (HBL) while asking the subject to stand and place equal weight on both feet was also established. Based on these results, clinicians can assess the AHR using the total foot length to estimate the truncated foot length unless toe deformities were present in the individual being assessed. While the foot measurements assessed in the current study were found not to be different between the left and right feet in females, the DAH was significantly different between feet in males. It is important to note, however, that the mean difference in DAH between feet for the males was less than 0.1 cm. While further research is always warranted, the results of this study provide clinicians and researchers with normative values for the AHR in a large group of women and men that can be used for assessing and classifying foot structure. Conflict of interest None.
References [1] Cowan DH, Jones BH, Robinson JR. Foot morphologic characteristics and risk of exercise-related injury. Arch Fam Med 1993;2:773–7. [2] Saltzman CL, Nawoczenski DA, Talbot KD. Measurement of the medial longitudinal arch. Arch Phys Med Rehabil 1995;76:45–9. [3] Yen DY, Puffer JC, Schmalzried TP. Injuries in runners: a prospective study of alignment. Clin J Sport Med 1998;8:187–94. [4] Wearing SC, Urry S, Perlman PR, Dubois P, Smeathers JE. Serial measurement of the calcaneal pitch during midstance. J Am Podiatr Med Assoc 1999;89:188–93. [5] Lees A, Lake M, Klenerman L. Shock absorption during forefoot running and its relationship to medial longitudinal arch height. Foot Ankle Int 2005;26:1081–8. [6] 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. [7] Cavanagh P, Rodgers M. The arch index: a useful measure from footprints. J Biomech 1987;20:547–51. [8] Evans AM, Copper AW, Scharfbillig RW, Scutter SD, Williams MT. Reliability of the foot posture index and traditional measures of foot position. J Am Podiatr Med Assoc 2003;93:2003–213.
T.G. McPoil et al. / The Foot 18 (2008) 220–227 [9] Weiner-Ogilvie S, Rome K. The reliability of three techniques for measuring foot position. J Am Podiatr Med Assoc 1998;88:381–6. [10] Menz HB, Munteanu SE. Validity of 3 clinical techniques for the measurement of static foot posture in older people. J Orthop Sports Phys Ther 2005;35:479–86. [11] Vinicombe A, Raspovic A, Menz HB. Reliability of navicular displacement measurement as a clinical indicator of foot posture. J Am Podiatr Med Assoc 2001;91:262–8. [12] Williams DS, McClay IS. Measurements used to characterize the foot and the medial longitudinal arch: Reliability and validity. Phys Ther 2000;80:864–71. [13] Williams III DS, McClay IS, Hamill J. Arch structure and injury patterns in runners. Clin Biomech 2001;16:341–7. [14] Williams III DS, Davis IM, Scholz JP, Hamill J, Buchanan TS. Higharched runners exhibit increased leg stiffness compared it low-arched runners. Gait Posture 2004;19:263–9. [15] Zifchock RA, Davis I, Hillstrom H, Song J. The effect of gender, age, and lateral dominance on arch height and arch stiffness. Foot Ankle Int 2006;27:367–72.
227
[16] Vicenzino B, Franettovich M, McPoil T, Russell T, Skardoon G. Initial effects of anti-pronation tape on medial longitudinal arch during walking and running. Br J Sports Med 2005;39:939–43. [17] Franettovich MM, McPoil TG, Russell T, Skardoon G, Vicenzino B. The ability to predict dynamic foot posture from static measurements. J Am Podiatr Med Assoc 2007;97:115–20. [18] Shrout PE, Fleiss JL. Intraclass correlations: uses in assessing rater reliability. Psychol Bull 1979;86:420–8. [19] Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics 1977;33:159–74. [20] Rothstein JM. Measurement and clinical practice: the theory and application. In: Rothstein JM, editor. Measurement in physical therapy. New York: Churchill Livingstone; 1985. p. 1–46. [21] Menz HB. Two feet, or one person? Problems associated with statistical analysis of paired data in foot and ankle medicine. Foot 2004;14:2–5. [22] Tesser 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.