G Model
GAIPOS-4358; No. of Pages 5 Gait & Posture xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Gait & Posture journal homepage: www.elsevier.com/locate/gaitpost
Foot posture, range of motion and plantar pressure characteristics in obese and non-obese individuals Paul A. Butterworth a,b,c, Donna M. Urquhart d,*, Karl B. Landorf a,b, Anita E. Wluka d, Flavia M. Cicuttini d, Hylton B. Menz b a
Department of Podiatry, La Trobe University, Bundoora, Victoria, Australia Lower Extremity and Gait Studies Program, La Trobe University, Bundoora, Victoria, Australia School of Health and Human Sciences, Southern Cross University, Bilinga, Queensland, Australia d Department of Epidemiology and Preventive Medicine, School of Public Health and Preventive Medicine, Monash University, Melbourne, Victoria, Australia b c
A R T I C L E I N F O
A B S T R A C T
Article history: Received 30 December 2013 Received in revised form 17 November 2014 Accepted 18 November 2014
Obesity is a world-wide health problem and is strongly associated with musculoskeletal disorders of the lower limb. The aim of this study was to evaluate plantar loading patterns in obese and non-obese individuals, while accounting for the contribution of foot structure, range of motion and walking speed. Sixty-eight participants (mean SD age, 52.6 8.0 years), including 47 females (69%), underwent assessments of body mass index, foot pain and foot structure. Plantar pressures were also obtained, using a floor-mounted resistive sensor mat system. Multiple regression analysis was used to determine which variables were most strongly associated with plantar loading patterns. Obese individuals exhibited flatter feet, reduced inversion–eversion range of motion, and higher peak plantar pressures when walking. After accounting for foot structure and walking speed, bodyweight was found to be significantly associated with elevated loading of the foot, particularly the forefoot and midfoot. These findings suggest that obesity increases the stresses applied to the foot directly, via increased bodyweight, and indirectly, via alterations to foot structure, which may partly explain the link between obesity and the development of foot pain. Clinicians dealing with foot problems should consider the effect of increased bodyweight on plantar loading in obese patients. ß 2014 Published by Elsevier B.V.
Keywords: Obesity Body mass index Foot Kinetics
1. Introduction Obesity is a world-wide health problem, and its prevalence is increasing in both developed and developing countries [1]. Previous estimates suggest that there are over 300 million people world-wide who are obese [2]. It is now well established that obesity is associated with diabetes and cardiovascular disease, however obesity is also associated with musculoskeletal disorders affecting the lower limb, such as knee and hip osteoarthritis [3]. More recent investigation has also found that the foot is not immune to the effects of obesity, with a recent systematic review of 25 studies involving 93,224 participants concluding that obesity is strongly associated with non-specific foot pain in the general population [4].
* Corresponding author at: School of Public Health and Preventive Medicine, Monash University, Alfred Centre, Commercial Road, Melbourne, Victoria 3004, Australia. Tel.: +61 3 9903 0994; fax: +61 3 9903 0556. E-mail address: [email protected] (D.M. Urquhart).
A possible explanation for the relationship between obesity and foot pain is that excess bodyweight leads to greater mechanical loading of the foot. Indeed, a link between increased force and pressure under the foot and obesity has been reported [5]. However, plantar loading patterns are affected by more than just weight. For example, variations in foot posture, joint range of motion and walking speed are also known to influence plantar loading patterns [6,7]. Accordingly, because obesity is associated with changes in foot structure [8] and walking speed [7], it is important to take these factors into account when comparing plantar pressures in obese versus non-obese individuals. Therefore, the aim of this study was to evaluate plantar loading patterns in obese and non-obese individuals, while accounting for the contribution of foot structure and walking speed. In doing so, our objective was to provide additional insights into foot function in people that are obese, and to explore the specific anatomical regions of the foot that may be subjected to increased plantar pressure in this population. We hypothesised that both bodyweight and foot structure would influence plantar loading patterns.
http://dx.doi.org/10.1016/j.gaitpost.2014.11.010 0966-6362/ß 2014 Published by Elsevier B.V.
Please cite this article in press as: Butterworth PA, et al. Foot posture, range of motion and plantar pressure characteristics in obese and non-obese individuals. Gait Posture (2014), http://dx.doi.org/10.1016/j.gaitpost.2014.11.010
G Model
GAIPOS-4358; No. of Pages 5 P.A. Butterworth et al. / Gait & Posture xxx (2014) xxx–xxx
2
In order to classify participants as obese or non-obese, measures to determine body mass index (BMI) were taken. Weight was measured to the nearest 0.1 kg using electronic scales and height was measured to the nearest 0.1 cm using a stadiometer (with shoes, socks, and bulky clothing removed). From these data BMI was calculated, and participants with a BMI 30 kg/m2 were considered obese [10].
Foot posture was calculated using the Foot Posture Index (FPI), a tool developed and validated for use in clinical practice, with higher values indicating a more pronated foot. The FPI provides data from all three cardinal body planes (sagittal, transverse and frontal), thus providing a more detailed assessment of foot structure and function [16]. The FPI has been shown to have good reliability (ICC, 0.61) in older people [17] and high intra-rater and inter-rater reliability compared with other measures such as the longitudinal arch angle [18]. Arch structure, as measured by the Arch Index (AI), is correlated with pressures under the midfoot when walking [6,19]. Arch structure in this study was therefore measured by the AI, a tool that has been shown to have excellent reliability [20]. The AI score is a ratio of the middle third of a footprint relative to the total area excluding the toes [20], with a higher ratio indicating a flatter foot (0.28 low arch). Frontal plane (inversion and eversion) ankle joint range of motion may influence postural control while walking [21]. Ankle inversion and eversion range of motion was evaluated using a goniometer with the participant seated, as previously described by Menadue et al. [22]. This method has demonstrated high intraobserver reliability within a test session (ICC, 0.82–0.96) and interobserver measurements with participants in a seated position are more reliable than in a prone position [22].
2.3. Foot pain and disability assessment
2.5. Plantar pressures
Participants completed the Manchester Foot Pain and Disability Index (MFPDI) to document disabling foot pain at the time of the study. The MFPDI has undergone psychometric validation and has been used to determine the prevalence of disabling foot pain in population-based studies [11,12]. Using the original definition of foot pain as described by Garrow et al. [13], participants with a score of 1 were defined as having foot pain.
Dynamic plantar pressure data were collected using the MatScan1 (TekScan, USA) system, using similar methodology to previous studies [23]. The MatScan1 system has demonstrated moderate to good reliability in a previous study [24]. The MatScan1 platform was positioned in the centre of a flat walkway, and participants were asked to walk across the platform, striking the platform with their right foot (third step), and continuing to walk past the platform with three further steps. Data from the right foot were collected from three valid trials. The MatScan1 Research Software version 6.51 was used to construct foot masks for each participant. The masks were developed to produce six anatomical regions; whole foot, heel, midfoot, forefoot, hallux and toes (Fig. 1). The masking process for all participants was completed by the primary author, with previous research demonstrating good intraobserver reliability with this process [24]. Measures of contact area (cm2), maximum force (kg), peak pressure (kg/cm2) and contact time (ms) as a proxy measure of walking speed, were then calculated for each of the three trials, and an average value produced.
2. Methods 2.1. Participants Sixty-eight participants from a previous study [9] that assessed the relationship between obesity and musculoskeletal disease were invited to participate – all 68 agreed to participate in the present study. The aim of the previous study was to examine the relationship between body composition and foot pain, in a population ranging from healthy to obese. Initial recruitment, inclusion and exclusion criteria have been previously published [9]. The study was approved by the Alfred Health Human Ethics Committee and the Monash University Human Research Ethics Committee (project number 121/11) and all participants signed informed consent. 2.2. Body mass index
2.4. Foot structure A range of foot structure variables were measured as potential predictors of plantar loading patterns, based on the work of Menz and Morris [6]. The presence or absence of hallux valgus was determined using the Manchester Scale, a validated [14] method of assessing the severity of hallux valgus deformity, using standardised photographs [15]. Hallux valgus is an important foot deformity in the assessment of plantar pressures, as previous research has shown that hallux valgus negatively affects loading of the hallux [6]. contact area female (-0.30)* weight (0.29)*
peak pressure
maximum force
22
10
weight (0.29)*
hallux valgus (-0.39)** weight (0.37)**
hallux valgus (-0.33)**
32
59
47
35
weight (0.54)**
female (-0.48)** weight (0.22)*
7
50
46
weight (0.27)*
52 weight (0.77)**
female (-0.32)* weight (0.51)**
45
21
weight (0.69)**
weight (0.60)**
weight (0.66)**
52
39
weight (0.72)**
weight (0.61)**
Fig. 1. Results of multiple linear regression analyses for contact area, maximum force and peak pressure. Values displayed in mask regions represent R2 values as percentages, and values contained in brackets represent b-weights for each significant independent (‘predictor’) variable. *Significance of b-weight p < 0.05, **significance of b-weight p < 0.01.
Please cite this article in press as: Butterworth PA, et al. Foot posture, range of motion and plantar pressure characteristics in obese and non-obese individuals. Gait Posture (2014), http://dx.doi.org/10.1016/j.gaitpost.2014.11.010
G Model
GAIPOS-4358; No. of Pages 5 P.A. Butterworth et al. / Gait & Posture xxx (2014) xxx–xxx
3
Table 1 Participant characteristics (values are the mean SD unless otherwise indicated).
2.6. Statistical analysis All continuous data were initially checked for distribution, with the following variables requiring logarithmic transformation because they were not normally distributed: maximum force (whole foot, hallux, toes and midfoot) and contact time. In addition, the FPI was Rasch-transformed into logit scores according to the method described by Keenan et al. [25]. Independent-samples t-tests and chi square tests were used to assess differences between groups (obese and non-obese) in relation to foot structure and function variables. Mann–Whitney U tests were used to assess differences between groups (obese and non-obese) in relation to the presence or absence of foot pain. A univariate general linear model, adjusted for total contact time, was used to evaluate the differences in plantar loading characteristics of the obese and non-obese participants. Correlations between foot structure and plantar loading with weight were assessed using Pearson’s bivariate correlation and Spearman’s rho bivariate correlations (adjusted for contact time). Correlations between plantar loading and foot structure were assessed using Pearson’s bivariate correlation (also adjusted for contact time) and effect sizes (Cohen’s d) were calculated from means and standard deviations. Finally, multivariate linear regression was undertaken to evaluate the relationship between foot structure and plantar pressures in the total sample. Foot structure variables found to be significantly correlated (p < 0.05) with plantar pressure variables were entered into a series of regression models to determine their relative importance in explaining variance. Only the most strongly associated variable was entered into the model. To avoid multicollinearity, only those independent variables that were not highly correlated (Pearson’s correlation
GAIPOS-4358; No. of Pages 5 Gait & Posture xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Gait & Posture journal homepage: www.elsevier.com/locate/gaitpost
Foot posture, range of motion and plantar pressure characteristics in obese and non-obese individuals Paul A. Butterworth a,b,c, Donna M. Urquhart d,*, Karl B. Landorf a,b, Anita E. Wluka d, Flavia M. Cicuttini d, Hylton B. Menz b a
Department of Podiatry, La Trobe University, Bundoora, Victoria, Australia Lower Extremity and Gait Studies Program, La Trobe University, Bundoora, Victoria, Australia School of Health and Human Sciences, Southern Cross University, Bilinga, Queensland, Australia d Department of Epidemiology and Preventive Medicine, School of Public Health and Preventive Medicine, Monash University, Melbourne, Victoria, Australia b c
A R T I C L E I N F O
A B S T R A C T
Article history: Received 30 December 2013 Received in revised form 17 November 2014 Accepted 18 November 2014
Obesity is a world-wide health problem and is strongly associated with musculoskeletal disorders of the lower limb. The aim of this study was to evaluate plantar loading patterns in obese and non-obese individuals, while accounting for the contribution of foot structure, range of motion and walking speed. Sixty-eight participants (mean SD age, 52.6 8.0 years), including 47 females (69%), underwent assessments of body mass index, foot pain and foot structure. Plantar pressures were also obtained, using a floor-mounted resistive sensor mat system. Multiple regression analysis was used to determine which variables were most strongly associated with plantar loading patterns. Obese individuals exhibited flatter feet, reduced inversion–eversion range of motion, and higher peak plantar pressures when walking. After accounting for foot structure and walking speed, bodyweight was found to be significantly associated with elevated loading of the foot, particularly the forefoot and midfoot. These findings suggest that obesity increases the stresses applied to the foot directly, via increased bodyweight, and indirectly, via alterations to foot structure, which may partly explain the link between obesity and the development of foot pain. Clinicians dealing with foot problems should consider the effect of increased bodyweight on plantar loading in obese patients. ß 2014 Published by Elsevier B.V.
Keywords: Obesity Body mass index Foot Kinetics
1. Introduction Obesity is a world-wide health problem, and its prevalence is increasing in both developed and developing countries [1]. Previous estimates suggest that there are over 300 million people world-wide who are obese [2]. It is now well established that obesity is associated with diabetes and cardiovascular disease, however obesity is also associated with musculoskeletal disorders affecting the lower limb, such as knee and hip osteoarthritis [3]. More recent investigation has also found that the foot is not immune to the effects of obesity, with a recent systematic review of 25 studies involving 93,224 participants concluding that obesity is strongly associated with non-specific foot pain in the general population [4].
* Corresponding author at: School of Public Health and Preventive Medicine, Monash University, Alfred Centre, Commercial Road, Melbourne, Victoria 3004, Australia. Tel.: +61 3 9903 0994; fax: +61 3 9903 0556. E-mail address: [email protected] (D.M. Urquhart).
A possible explanation for the relationship between obesity and foot pain is that excess bodyweight leads to greater mechanical loading of the foot. Indeed, a link between increased force and pressure under the foot and obesity has been reported [5]. However, plantar loading patterns are affected by more than just weight. For example, variations in foot posture, joint range of motion and walking speed are also known to influence plantar loading patterns [6,7]. Accordingly, because obesity is associated with changes in foot structure [8] and walking speed [7], it is important to take these factors into account when comparing plantar pressures in obese versus non-obese individuals. Therefore, the aim of this study was to evaluate plantar loading patterns in obese and non-obese individuals, while accounting for the contribution of foot structure and walking speed. In doing so, our objective was to provide additional insights into foot function in people that are obese, and to explore the specific anatomical regions of the foot that may be subjected to increased plantar pressure in this population. We hypothesised that both bodyweight and foot structure would influence plantar loading patterns.
http://dx.doi.org/10.1016/j.gaitpost.2014.11.010 0966-6362/ß 2014 Published by Elsevier B.V.
Please cite this article in press as: Butterworth PA, et al. Foot posture, range of motion and plantar pressure characteristics in obese and non-obese individuals. Gait Posture (2014), http://dx.doi.org/10.1016/j.gaitpost.2014.11.010
G Model
GAIPOS-4358; No. of Pages 5 P.A. Butterworth et al. / Gait & Posture xxx (2014) xxx–xxx
2
In order to classify participants as obese or non-obese, measures to determine body mass index (BMI) were taken. Weight was measured to the nearest 0.1 kg using electronic scales and height was measured to the nearest 0.1 cm using a stadiometer (with shoes, socks, and bulky clothing removed). From these data BMI was calculated, and participants with a BMI 30 kg/m2 were considered obese [10].
Foot posture was calculated using the Foot Posture Index (FPI), a tool developed and validated for use in clinical practice, with higher values indicating a more pronated foot. The FPI provides data from all three cardinal body planes (sagittal, transverse and frontal), thus providing a more detailed assessment of foot structure and function [16]. The FPI has been shown to have good reliability (ICC, 0.61) in older people [17] and high intra-rater and inter-rater reliability compared with other measures such as the longitudinal arch angle [18]. Arch structure, as measured by the Arch Index (AI), is correlated with pressures under the midfoot when walking [6,19]. Arch structure in this study was therefore measured by the AI, a tool that has been shown to have excellent reliability [20]. The AI score is a ratio of the middle third of a footprint relative to the total area excluding the toes [20], with a higher ratio indicating a flatter foot (0.28 low arch). Frontal plane (inversion and eversion) ankle joint range of motion may influence postural control while walking [21]. Ankle inversion and eversion range of motion was evaluated using a goniometer with the participant seated, as previously described by Menadue et al. [22]. This method has demonstrated high intraobserver reliability within a test session (ICC, 0.82–0.96) and interobserver measurements with participants in a seated position are more reliable than in a prone position [22].
2.3. Foot pain and disability assessment
2.5. Plantar pressures
Participants completed the Manchester Foot Pain and Disability Index (MFPDI) to document disabling foot pain at the time of the study. The MFPDI has undergone psychometric validation and has been used to determine the prevalence of disabling foot pain in population-based studies [11,12]. Using the original definition of foot pain as described by Garrow et al. [13], participants with a score of 1 were defined as having foot pain.
Dynamic plantar pressure data were collected using the MatScan1 (TekScan, USA) system, using similar methodology to previous studies [23]. The MatScan1 system has demonstrated moderate to good reliability in a previous study [24]. The MatScan1 platform was positioned in the centre of a flat walkway, and participants were asked to walk across the platform, striking the platform with their right foot (third step), and continuing to walk past the platform with three further steps. Data from the right foot were collected from three valid trials. The MatScan1 Research Software version 6.51 was used to construct foot masks for each participant. The masks were developed to produce six anatomical regions; whole foot, heel, midfoot, forefoot, hallux and toes (Fig. 1). The masking process for all participants was completed by the primary author, with previous research demonstrating good intraobserver reliability with this process [24]. Measures of contact area (cm2), maximum force (kg), peak pressure (kg/cm2) and contact time (ms) as a proxy measure of walking speed, were then calculated for each of the three trials, and an average value produced.
2. Methods 2.1. Participants Sixty-eight participants from a previous study [9] that assessed the relationship between obesity and musculoskeletal disease were invited to participate – all 68 agreed to participate in the present study. The aim of the previous study was to examine the relationship between body composition and foot pain, in a population ranging from healthy to obese. Initial recruitment, inclusion and exclusion criteria have been previously published [9]. The study was approved by the Alfred Health Human Ethics Committee and the Monash University Human Research Ethics Committee (project number 121/11) and all participants signed informed consent. 2.2. Body mass index
2.4. Foot structure A range of foot structure variables were measured as potential predictors of plantar loading patterns, based on the work of Menz and Morris [6]. The presence or absence of hallux valgus was determined using the Manchester Scale, a validated [14] method of assessing the severity of hallux valgus deformity, using standardised photographs [15]. Hallux valgus is an important foot deformity in the assessment of plantar pressures, as previous research has shown that hallux valgus negatively affects loading of the hallux [6]. contact area female (-0.30)* weight (0.29)*
peak pressure
maximum force
22
10
weight (0.29)*
hallux valgus (-0.39)** weight (0.37)**
hallux valgus (-0.33)**
32
59
47
35
weight (0.54)**
female (-0.48)** weight (0.22)*
7
50
46
weight (0.27)*
52 weight (0.77)**
female (-0.32)* weight (0.51)**
45
21
weight (0.69)**
weight (0.60)**
weight (0.66)**
52
39
weight (0.72)**
weight (0.61)**
Fig. 1. Results of multiple linear regression analyses for contact area, maximum force and peak pressure. Values displayed in mask regions represent R2 values as percentages, and values contained in brackets represent b-weights for each significant independent (‘predictor’) variable. *Significance of b-weight p < 0.05, **significance of b-weight p < 0.01.
Please cite this article in press as: Butterworth PA, et al. Foot posture, range of motion and plantar pressure characteristics in obese and non-obese individuals. Gait Posture (2014), http://dx.doi.org/10.1016/j.gaitpost.2014.11.010
G Model
GAIPOS-4358; No. of Pages 5 P.A. Butterworth et al. / Gait & Posture xxx (2014) xxx–xxx
3
Table 1 Participant characteristics (values are the mean SD unless otherwise indicated).
2.6. Statistical analysis All continuous data were initially checked for distribution, with the following variables requiring logarithmic transformation because they were not normally distributed: maximum force (whole foot, hallux, toes and midfoot) and contact time. In addition, the FPI was Rasch-transformed into logit scores according to the method described by Keenan et al. [25]. Independent-samples t-tests and chi square tests were used to assess differences between groups (obese and non-obese) in relation to foot structure and function variables. Mann–Whitney U tests were used to assess differences between groups (obese and non-obese) in relation to the presence or absence of foot pain. A univariate general linear model, adjusted for total contact time, was used to evaluate the differences in plantar loading characteristics of the obese and non-obese participants. Correlations between foot structure and plantar loading with weight were assessed using Pearson’s bivariate correlation and Spearman’s rho bivariate correlations (adjusted for contact time). Correlations between plantar loading and foot structure were assessed using Pearson’s bivariate correlation (also adjusted for contact time) and effect sizes (Cohen’s d) were calculated from means and standard deviations. Finally, multivariate linear regression was undertaken to evaluate the relationship between foot structure and plantar pressures in the total sample. Foot structure variables found to be significantly correlated (p < 0.05) with plantar pressure variables were entered into a series of regression models to determine their relative importance in explaining variance. Only the most strongly associated variable was entered into the model. To avoid multicollinearity, only those independent variables that were not highly correlated (Pearson’s correlation
