Author's Accepted Manuscript
Effects of footwear on three-dimensional tibiotalar and subtalar joint motion during running Cathryn D. Peltz, Jeffrey A. Haladik, Scott E. Hoffman, Michael McDonald, Nicole L. Ramo, George Divine, Matthew Nurse, Michael J. Bey
www.elsevier.com/locate/jbiomech
PII: DOI: Reference:
S0021-9290(14)00327-3 http://dx.doi.org/10.1016/j.jbiomech.2014.05.016 BM6671
To appear in:
Journal of Biomechanics
Accepted date: 24 May 2014 Cite this article as: Cathryn D. Peltz, Jeffrey A. Haladik, Scott E. Hoffman, Michael McDonald, Nicole L. Ramo, George Divine, Matthew Nurse, Michael J. Bey, Effects of footwear on three-dimensional tibiotalar and subtalar joint motion during running, Journal of Biomechanics, http://dx.doi.org/10.1016/j. jbiomech.2014.05.016 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Effects Of Footwear on Three-Dimensional Tibiotalar and Subtalar Joint Motion During Running 1
Cathryn D. Peltz, PhD* 1 Jeffrey A. Haladik 1 Scott E. Hoffman 1 Michael McDonald 1 Nicole L. Ramo 2 George Divine, PhD 3 Matthew Nurse, PhD 1 Michael J. Bey, PhD
1
Henry Ford Health Systems Department of Orthopaedic Surgery Bone and Joint Center 2
Henry Ford Health Systems Department of Public Health Sciences 3
Nike, Inc. Nike Sport Research Lab Beaverton, OR
* Corresponding Author: Henry Ford Hospital 2799 W. Grand Blvd E&R 2015 Detroit, MI 48202 Phone: 313-916-8680 Fax: 313-916-8812 Email: [email protected]
Word Count: 3398 Keywords: kinematics, barefoot, minimalist, motion control, dynamic
ABSTRACT Running is a popular form of recreation, but injuries are common and may be associated with abnormal joint motion. The objective of this study was to determine the effect of three footwear conditions – barefoot (BF), an ultraflexible training shoe (FREE), and a motion control shoe (MC) – on 3D foot and ankle motion. Dynamic, biplane radiographic images were acquired from 12 runners during overground running. 3D rotations of the tibiotalar and subtalar joints were quantified in terms of plantarflexion/dorsiflexion (PF/DF), inversion/eversion (IN/EV) and internal/external rotation (IR/ER). Across the early stance phase (defined as footstrike to heel-off), BF running demonstrated greater tibiotalar joint range of motion for PF/DF (28.2±8.3º) and IR/ER (7.0±1.4º) than the shod conditions
(FREE:
IR/ER=4.3±0.7º).
PF/DF=15.1±5.9º,
IR/ER=4.8±2.1º;
MC:
PF/DF=15.0±6.2º,
Also at the tibiotalar joint, BF running resulted in a position
significantly more plantarflexed (BF: 2.0±12.5º, FREE: 15.7±12.2º, MC: 16.5±9.3º) and internally rotated (BF: 12.9±4.5º, FREE: 10.7±4.3º, MC: 10.6±3.9º) at footstrike compared to both shod conditions. No differences were detected between the shod conditions at any point in the early stance phase at the tibiotalar joint. The MC condition demonstrated significant differences compared to FREE at several points throughout the early stance phase at the subtalar joint, with the greatest differences seen at 30% in PF/DF (MC -1.4±8.8º: FREE: -0.5±9.0º), IN/EV (MC -8.1±5.7º: FREE -6.3±5.5º) and IR/ER (MC -9.5±5.3º: FREE: -8.7±5.2º). These findings indicate that footwear has subtle effects on joint motion mainly between BF and shod conditions at the tibiotalar joint and between shod conditions at the subtalar joint.
INTRODUCTION Running for recreation and fitness has been very popular since the 1970’s.
Even
recently, participation in running and jogging in the United States increased 10.3% from 2010 to 2012, to 35.5 million (Rothschild 2012). However, running-related injuries are very common in this population. Previous research has reported that approximately 19% to 79% of runners are injured every year (Buist et al. 2010; Macera et al. 1989; van Gent et al. 2007). Pronation during running is believed to contribute to running-related injuries (Willems et al. 2007), but the complex relationships between joint motion and injury are not well understood. Consequently, it is not surprising that the running shoe industry offers a broad spectrum of models that range from shoes designed to minimize foot and ankle motion (i.e., motion control shoes) to shoes designed to mimic barefoot running by placing minimal constraints upon natural foot motion (i.e., minimalist shoes). In addition, barefoot running experienced a surge in popularity largely based on unsubstantiated claims of injury prevention (Robbins and Hanna 1987), though more recent publications strongly suggest that this surge has abated substantially (Nigg and Enders 2013).
Previous research has focused on understanding the association between footwear and lower extremity kinematics (Bonacci et al. 2013; Brown et al. 1995; Willems et al. 2007; Willems et al. 2005; Williams et al. 2012), and these studies have typically relied upon optical motion capture techniques to track the three-dimensional (3D) position of markers secured to the subject’s shoes and/or skin. These studies have contributed significantly to our current understanding of footwear design and biomechanics, but the in-vivo accuracy of these conventional motion capture techniques is limited due to motion of the shoe and/or skin relative to the underlying bones.
For example,
Reinschmidt and colleagues reported that external markers typically overestimate
tibiocalcaneal rotations compared to bone markers, and that these overestimations can be up to 11º (Reinschmidt et al. 1997). Furthermore, this limitation often dictates that motions of the tibiotalar and subtalar joints be combined into a single composite measure of “ankle” joint motion, and that measures of inversion/eversion about an anterior-posterior axis do not represent the subtalar joint axis.
Investigators have
overcome these limitations by tracking the position of markers that are attached to bone pins that have been surgically implanted into the tibia and calcaneus (Stacoff et al. 1991; Stacoff et al. 1996), but the extent to which this invasive procedure alters normal motion is unclear.
The objective of this study was to determine the effect of footwear on 3D in-vivo foot and ankle motion. Our approach was to utilize a biplane x-ray imaging system to quantify tibiotalar joint rotations and subtalar joint rotations while healthy subjects ran barefoot, in ultraflexible training shoes, and in motion control shoes. We hypothesized that joint ranges of motion would be highest in the barefoot condition and lowest in the motion control condition, and that motion patterns in the ultraflexible training shoe would be most similar to motion patterns in the barefoot condition.
MATERIALS AND METHODS After obtaining Institutional Review Board approval and informed consent, a convenience sample of 12 subjects (6 male, 6 female, average age 24.2 ± 4.4 years, range 18-33 years) enrolled in this study. The subjects were all recreational runners who had averaged at least 25 miles per week of running for the year prior to testing. They were injury free, meaning they had taken no time off from running in the previous year due to injuries and had no history of previous surgery on either lower extremity. No
habitually barefoot runners were used in this study and all subjects performed the bulk of their training in conventional running shoes.
Prior to testing, subjects warmed up on a treadmill for 15 minutes at a self-selected pace in their own running shoes. After warm-up, subjects were tested in 3 footwear conditions: barefoot (BF), an ultraflexible training shoe (FREE, Nike Free 3.0 V4), and a motion control shoe (MC, Nike Structure Triax 15). The motion control shoe (Nike Structure Triax 15) is a stability shoe designed to help limit rearfoot motion in runners who pronate during the ground contact phase. To achieve this, it incorporates an articulated crash pad on the lateral rearfoot and has a medial post that extends along the medial mid-foot and heel. It has a heel-ball offset of 12mm, which is typical of traditional running shoes. The ultraflexible training shoe (Nike Free 3.0) is designed to provide the benefits of natural motion in a minimalist design by reducing the heel-ball offset to 4mm. The midsole features deep grooves that extend along the anterior-posterior and mediallateral directions with the intent of allowing greater flexibility compared to traditional running shoes. In addition, the shoe also has a minimally designed upper intended to allow the foot and toes to move more naturally.
Radiographic images were acquired at 120 Hz as subjects ran at a self-selected pace along a 15 m long custom elevated platform. The elevated platform allowed each subject’s left foot to be centered within the biplane x-ray system’s 3D imaging volume (Figure 1). A foot strike target of this 3D imaging volume was indicated on the platform, but subjects were specifically instructed to not look down or alter their gait in order to achieve this target. The location of each subject’s left foot relative to this foot strike target was monitored by laboratory personnel and the subjects’ starting position on the platform was modified as needed until their left foot was consistently centered within the
imaging volume. Three trials were acquired for each footwear condition, and the order of footwear testing was balanced across subjects. Each trial was 0.6 seconds long, which corresponded to 72 images per trial at the 120 Hz sampling rate. For each trial, the early stance phase was defined by identifying the frames from foot strike to heel-off in the dynamic radiographic images.
Following testing, computed tomography (CT) scans of the distal tibia and entire foot were acquired (LightSpeed16, GE Medical Systems, Piscataway, New Jersey). The scans were acquired with a slice thickness of 1.25 mm and an in-plane resolution of approximately 0.6 mm per pixel. The calcaneus, talus and distal tibia were manually segmented from surrounding bones and soft tissue and reconstructed into 3D bone models (Mimics, Materialise, Leuven, Belgium).
For each trial, the 3D motions of the tibia, talus and calcaneus were determined for every frame from foot strike to heel-off of the biplane x-ray images using a model-based tracking technique (Bey et al. 2011). Briefly, this technique uses the CT-based bone models to determine the position and orientation of each bone based on its 3D shape and radiographic density.
Anatomical coordinate systems were created for each
subject’s tibia, talus, and calcaneus from the CT-based bone models. The anatomical axes were defined such that the x-axis was nominally aligned in the anterior/posterior direction, the y-axis was nominally aligned in the superior/inferior direction and the z-axis was nominally aligned in the medial/lateral direction for each bone (Figure 2), consistent with the coordinate systems recently described by Gutenkust and colleagues (Gutekunst et al. 2013; Schon et al. 1998; Steel et al. 1980; Wu et al. 2002). 3D rotations of the tibiotalar joint (i.e., talus relative to the tibia) and subtalar joint (i.e., calcaneus relative to the talus) were calculated for every frame in terms of plantarflexion/dorsiflexion (PF/DF)
about
the
z-axis,
inversion/eversion
(IN/EV)
about
the
x-axis
and
internal
rotation/external rotation (IR/ER) about the y-axis (Wu et al. 2002). Joint ranges of motion (maximum-minimum) were calculated for each trial. In addition, the 3D joint rotations were normalized in 10% increments of the early stance phase which was defined as from foot strike (0%) to heel-off (100%). The data from all 3 trials were then averaged for each subject.
Differences in measures of joint motion were assessed with footwear condition as the main effect using a one-way ANOVA with repeated measures for each range of motion value (across the entire trial) as well as at each 10% interval (GraphPad Prism 6, GraphPad Software, La Jolla, CA and SAS version 9.4, Cary, North Carolina). If the ANOVA was significant, Bonferroni post-hoc tests corrected for multiple comparisons were then calculated in order to compare between each pair of footwear conditions. For ANOVA analysis, significance was set at p
Effects of footwear on three-dimensional tibiotalar and subtalar joint motion during running Cathryn D. Peltz, Jeffrey A. Haladik, Scott E. Hoffman, Michael McDonald, Nicole L. Ramo, George Divine, Matthew Nurse, Michael J. Bey
www.elsevier.com/locate/jbiomech
PII: DOI: Reference:
S0021-9290(14)00327-3 http://dx.doi.org/10.1016/j.jbiomech.2014.05.016 BM6671
To appear in:
Journal of Biomechanics
Accepted date: 24 May 2014 Cite this article as: Cathryn D. Peltz, Jeffrey A. Haladik, Scott E. Hoffman, Michael McDonald, Nicole L. Ramo, George Divine, Matthew Nurse, Michael J. Bey, Effects of footwear on three-dimensional tibiotalar and subtalar joint motion during running, Journal of Biomechanics, http://dx.doi.org/10.1016/j. jbiomech.2014.05.016 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Effects Of Footwear on Three-Dimensional Tibiotalar and Subtalar Joint Motion During Running 1
Cathryn D. Peltz, PhD* 1 Jeffrey A. Haladik 1 Scott E. Hoffman 1 Michael McDonald 1 Nicole L. Ramo 2 George Divine, PhD 3 Matthew Nurse, PhD 1 Michael J. Bey, PhD
1
Henry Ford Health Systems Department of Orthopaedic Surgery Bone and Joint Center 2
Henry Ford Health Systems Department of Public Health Sciences 3
Nike, Inc. Nike Sport Research Lab Beaverton, OR
* Corresponding Author: Henry Ford Hospital 2799 W. Grand Blvd E&R 2015 Detroit, MI 48202 Phone: 313-916-8680 Fax: 313-916-8812 Email: [email protected]
Word Count: 3398 Keywords: kinematics, barefoot, minimalist, motion control, dynamic
ABSTRACT Running is a popular form of recreation, but injuries are common and may be associated with abnormal joint motion. The objective of this study was to determine the effect of three footwear conditions – barefoot (BF), an ultraflexible training shoe (FREE), and a motion control shoe (MC) – on 3D foot and ankle motion. Dynamic, biplane radiographic images were acquired from 12 runners during overground running. 3D rotations of the tibiotalar and subtalar joints were quantified in terms of plantarflexion/dorsiflexion (PF/DF), inversion/eversion (IN/EV) and internal/external rotation (IR/ER). Across the early stance phase (defined as footstrike to heel-off), BF running demonstrated greater tibiotalar joint range of motion for PF/DF (28.2±8.3º) and IR/ER (7.0±1.4º) than the shod conditions
(FREE:
IR/ER=4.3±0.7º).
PF/DF=15.1±5.9º,
IR/ER=4.8±2.1º;
MC:
PF/DF=15.0±6.2º,
Also at the tibiotalar joint, BF running resulted in a position
significantly more plantarflexed (BF: 2.0±12.5º, FREE: 15.7±12.2º, MC: 16.5±9.3º) and internally rotated (BF: 12.9±4.5º, FREE: 10.7±4.3º, MC: 10.6±3.9º) at footstrike compared to both shod conditions. No differences were detected between the shod conditions at any point in the early stance phase at the tibiotalar joint. The MC condition demonstrated significant differences compared to FREE at several points throughout the early stance phase at the subtalar joint, with the greatest differences seen at 30% in PF/DF (MC -1.4±8.8º: FREE: -0.5±9.0º), IN/EV (MC -8.1±5.7º: FREE -6.3±5.5º) and IR/ER (MC -9.5±5.3º: FREE: -8.7±5.2º). These findings indicate that footwear has subtle effects on joint motion mainly between BF and shod conditions at the tibiotalar joint and between shod conditions at the subtalar joint.
INTRODUCTION Running for recreation and fitness has been very popular since the 1970’s.
Even
recently, participation in running and jogging in the United States increased 10.3% from 2010 to 2012, to 35.5 million (Rothschild 2012). However, running-related injuries are very common in this population. Previous research has reported that approximately 19% to 79% of runners are injured every year (Buist et al. 2010; Macera et al. 1989; van Gent et al. 2007). Pronation during running is believed to contribute to running-related injuries (Willems et al. 2007), but the complex relationships between joint motion and injury are not well understood. Consequently, it is not surprising that the running shoe industry offers a broad spectrum of models that range from shoes designed to minimize foot and ankle motion (i.e., motion control shoes) to shoes designed to mimic barefoot running by placing minimal constraints upon natural foot motion (i.e., minimalist shoes). In addition, barefoot running experienced a surge in popularity largely based on unsubstantiated claims of injury prevention (Robbins and Hanna 1987), though more recent publications strongly suggest that this surge has abated substantially (Nigg and Enders 2013).
Previous research has focused on understanding the association between footwear and lower extremity kinematics (Bonacci et al. 2013; Brown et al. 1995; Willems et al. 2007; Willems et al. 2005; Williams et al. 2012), and these studies have typically relied upon optical motion capture techniques to track the three-dimensional (3D) position of markers secured to the subject’s shoes and/or skin. These studies have contributed significantly to our current understanding of footwear design and biomechanics, but the in-vivo accuracy of these conventional motion capture techniques is limited due to motion of the shoe and/or skin relative to the underlying bones.
For example,
Reinschmidt and colleagues reported that external markers typically overestimate
tibiocalcaneal rotations compared to bone markers, and that these overestimations can be up to 11º (Reinschmidt et al. 1997). Furthermore, this limitation often dictates that motions of the tibiotalar and subtalar joints be combined into a single composite measure of “ankle” joint motion, and that measures of inversion/eversion about an anterior-posterior axis do not represent the subtalar joint axis.
Investigators have
overcome these limitations by tracking the position of markers that are attached to bone pins that have been surgically implanted into the tibia and calcaneus (Stacoff et al. 1991; Stacoff et al. 1996), but the extent to which this invasive procedure alters normal motion is unclear.
The objective of this study was to determine the effect of footwear on 3D in-vivo foot and ankle motion. Our approach was to utilize a biplane x-ray imaging system to quantify tibiotalar joint rotations and subtalar joint rotations while healthy subjects ran barefoot, in ultraflexible training shoes, and in motion control shoes. We hypothesized that joint ranges of motion would be highest in the barefoot condition and lowest in the motion control condition, and that motion patterns in the ultraflexible training shoe would be most similar to motion patterns in the barefoot condition.
MATERIALS AND METHODS After obtaining Institutional Review Board approval and informed consent, a convenience sample of 12 subjects (6 male, 6 female, average age 24.2 ± 4.4 years, range 18-33 years) enrolled in this study. The subjects were all recreational runners who had averaged at least 25 miles per week of running for the year prior to testing. They were injury free, meaning they had taken no time off from running in the previous year due to injuries and had no history of previous surgery on either lower extremity. No
habitually barefoot runners were used in this study and all subjects performed the bulk of their training in conventional running shoes.
Prior to testing, subjects warmed up on a treadmill for 15 minutes at a self-selected pace in their own running shoes. After warm-up, subjects were tested in 3 footwear conditions: barefoot (BF), an ultraflexible training shoe (FREE, Nike Free 3.0 V4), and a motion control shoe (MC, Nike Structure Triax 15). The motion control shoe (Nike Structure Triax 15) is a stability shoe designed to help limit rearfoot motion in runners who pronate during the ground contact phase. To achieve this, it incorporates an articulated crash pad on the lateral rearfoot and has a medial post that extends along the medial mid-foot and heel. It has a heel-ball offset of 12mm, which is typical of traditional running shoes. The ultraflexible training shoe (Nike Free 3.0) is designed to provide the benefits of natural motion in a minimalist design by reducing the heel-ball offset to 4mm. The midsole features deep grooves that extend along the anterior-posterior and mediallateral directions with the intent of allowing greater flexibility compared to traditional running shoes. In addition, the shoe also has a minimally designed upper intended to allow the foot and toes to move more naturally.
Radiographic images were acquired at 120 Hz as subjects ran at a self-selected pace along a 15 m long custom elevated platform. The elevated platform allowed each subject’s left foot to be centered within the biplane x-ray system’s 3D imaging volume (Figure 1). A foot strike target of this 3D imaging volume was indicated on the platform, but subjects were specifically instructed to not look down or alter their gait in order to achieve this target. The location of each subject’s left foot relative to this foot strike target was monitored by laboratory personnel and the subjects’ starting position on the platform was modified as needed until their left foot was consistently centered within the
imaging volume. Three trials were acquired for each footwear condition, and the order of footwear testing was balanced across subjects. Each trial was 0.6 seconds long, which corresponded to 72 images per trial at the 120 Hz sampling rate. For each trial, the early stance phase was defined by identifying the frames from foot strike to heel-off in the dynamic radiographic images.
Following testing, computed tomography (CT) scans of the distal tibia and entire foot were acquired (LightSpeed16, GE Medical Systems, Piscataway, New Jersey). The scans were acquired with a slice thickness of 1.25 mm and an in-plane resolution of approximately 0.6 mm per pixel. The calcaneus, talus and distal tibia were manually segmented from surrounding bones and soft tissue and reconstructed into 3D bone models (Mimics, Materialise, Leuven, Belgium).
For each trial, the 3D motions of the tibia, talus and calcaneus were determined for every frame from foot strike to heel-off of the biplane x-ray images using a model-based tracking technique (Bey et al. 2011). Briefly, this technique uses the CT-based bone models to determine the position and orientation of each bone based on its 3D shape and radiographic density.
Anatomical coordinate systems were created for each
subject’s tibia, talus, and calcaneus from the CT-based bone models. The anatomical axes were defined such that the x-axis was nominally aligned in the anterior/posterior direction, the y-axis was nominally aligned in the superior/inferior direction and the z-axis was nominally aligned in the medial/lateral direction for each bone (Figure 2), consistent with the coordinate systems recently described by Gutenkust and colleagues (Gutekunst et al. 2013; Schon et al. 1998; Steel et al. 1980; Wu et al. 2002). 3D rotations of the tibiotalar joint (i.e., talus relative to the tibia) and subtalar joint (i.e., calcaneus relative to the talus) were calculated for every frame in terms of plantarflexion/dorsiflexion (PF/DF)
about
the
z-axis,
inversion/eversion
(IN/EV)
about
the
x-axis
and
internal
rotation/external rotation (IR/ER) about the y-axis (Wu et al. 2002). Joint ranges of motion (maximum-minimum) were calculated for each trial. In addition, the 3D joint rotations were normalized in 10% increments of the early stance phase which was defined as from foot strike (0%) to heel-off (100%). The data from all 3 trials were then averaged for each subject.
Differences in measures of joint motion were assessed with footwear condition as the main effect using a one-way ANOVA with repeated measures for each range of motion value (across the entire trial) as well as at each 10% interval (GraphPad Prism 6, GraphPad Software, La Jolla, CA and SAS version 9.4, Cary, North Carolina). If the ANOVA was significant, Bonferroni post-hoc tests corrected for multiple comparisons were then calculated in order to compare between each pair of footwear conditions. For ANOVA analysis, significance was set at p
