Alteration in Global Motor Strategy After Ankle Sprain
Alteration in Global Motor Strategy After Ankle Sprain
All volunteers participating in this study were selected from the Canadian Forces (CF) military population. The LAS group was composed of 10 men with a diagnosis of acute unilateral LAS. They all received physiotherapy interventions within five days of injury and were discharged after a maximum of nine weeks. They were excluded if they had an ankle fracture (documented by X-ray), a third grade LAS or a high ankle sprain (tibiofibular sprain), or if they reported a history of neuromuscular or neurodegenerative disease. The control group consisted of a comparable group of 10 men without prior history of LAS or symptoms from the lumbar spine and the lower limbs, or a history of neuromuscular and neurodegenerative disease. The present study was approved by the ethics committees of the Quebec Rehabilitation Institute and the CF Health Services Group. All participants read and signed an informed consent form.
In this descriptive study, all participants were evaluated within a single session. Participants in the LAS group were selected in acute stage (within two days after injury) and were evaluated between eight and ten weeks after their injury. Personal and clinical characteristics of the participants were collected. Their functional ability was measured using two questionnaires: the Foot & Ankle Disability Index (FADI) and the Lower Extremity Function Scale (LEFS). Thereafter, they took part in a series of complex motor tasks and tests, where the testing order was kept the same for all participants.
During the SEBT, the participant had to: (1) touch the floor as far as possible with the tip of the foot of the reaching limb in three different directions with respect to the stance limb (anteromedial [AM], medial [M] and posteromedial [PM] directions) and (2) return to unipedal stance after each reaching movement while maintaining balance on the stance limb with the hands resting on the iliac crests. The AM, M and PM directions were the ones proposed by Hertel et al. because these directions were able to identify significant reach deficits associated with chronic ankle instability and provided complementary and non-redundant information. After a practice session (six trials/direction followed by a five-minute rest period), three successful trials, separated by a ten-second rest period, were recorded for each direction (Figure 1). Trials were rejected according to the criteria used in previous studies (1) weight-bearing by the reaching limb, (2) displacement of the stance limb, or (3) loss of balance. As shown in Figure 1, the directions to be followed by the reaching limb during the tests were clearly marked on the floor using graduated tape. The order of test conditions (two legs and three directions) was determined as follows: first the leg order was randomly selected, then a test direction was randomly determined. All conditions for one leg were completed starting with the determined direction followed by the next anticlockwise direction (right leg) or clockwise direction (left leg). The same condition order was then used for the second leg.
(Enlarge Image)
Figure 1.
Kinematics variables during the SEBT task, experimental setting and description of subtasks and events. Illustration of the six conditions at the SEBT (3 reaching directions per limb; AM: anteromedial, M: medial, PM: posteromedial) and profiles (mean ± 1 standard deviation; n = 3 trials) of the vertical position of the reaching foot (A), global body CoM (CoMgl) (B and C) and joint amplitude of motion of the stance limb (D) in a typical healthy subject in the PM reaching direction. The central grey rectangle, called the central phase around foot contact, represents the critical period used for the analyses. The lowest vertical position of the tip of the reaching foot was used to subdivide the task, which corresponds to the "foot contact" event (vertical line at 50% of the task). The first subtask is characterized by body lowering (B and D) and alignment of the reaching limb for foot contact (A). In the second subtask which occurred after foot contact, a rapid straightening up of the entire body (B and C) is performed while returning from perturbation in single-limb stance. The transition between subtasks represents a highly challenging period for stability and motor control as a change in movement direction takes place at the perceived limits of stability. This transition period, called the central phase, was defined as -1 s to +1 s after foot contact. Further analysis about the quality of motor control was performed during this phase.
During the SEBT, biomechanical data were collected to characterize both whole body and segmental strategies using a 3D motion capture system (Optotrak 3020; Northern Digital Inc., Waterloo, Ontario, Canada). Forty-five infrared markers placed in non-colinear triads were installed on the following body segments: lower limbs, head, and trunk. For uppers limbs, rings of three markers were placed around the wrists. Kinematic data was sampled at 50 Hz and then digitally filtered at 10Hz (low pass filter). Global body CoM (CoMgl) was estimated using the equations proposed by Winter, and adjusted to consider the influence of the upper limb CoM.
All individual kinematic data analyses were based on a mean of three trials per test condition. A mean of two trials was exceptionally used in a very few cases (5 out of 120 conditions). The maximal reach distance (MRD) on the SEBT was calculated from the horizontal plane coordinates (vectorial distance) of the tip of the reaching foot (probed point) at the time it touched the floor.
Key Variables Used to Characterize Motor Control. The SEBT task was divided into two main subtasks: the going-to- and the return-from-MRD (Figure 1A). The transition between these subtasks has been called the central phase and was defined as -1 s to +1 s after foot contact (Figure 1: central grey rectangle). Four variables describing the behaviour of the CoMgl, called global variables, have been chosen as key indicators of global body strategy during this task. The maximal displacement of the CoMgl vertically; peak-to-peak CoMgl velocities along the vertical axis were calculated during the central phase (Figure 2A); distance between the centre of the foot and the mean position of the horizontal CoMgl when the foot contacted the floor and the horizontal excursion of the CoMgl during the central phase were also calculated (Figure 2C).
(Enlarge Image)
Figure 2.
Global body strategies along the vertical axis (A) and horizontal plane (C) in the central phase. A: the profiles for global CoM lowering (#) and for the peak-to peak velocity of the global CoM (# #) are illustrated for one limb of both groups (mean ± 1 standard deviation for healthy group and mean for LAS group; n = 10 trials per group) during the medial reaching direction on the SEBT. B: mean values (+1 standard deviation) for global CoM lowering and peak-to peak of CoM vertical velocity of both groups. C: CoM displacement in the horizontal plane (# # #) during the different conditions (directions and limbs) in both groups. Horizontal resultant lines of global CoM position at foot contact (doted circles) were calculated and used for further statistical analyses. D: mean values (+1 standard deviation) for global CoM in the horizontal plane. Asterisks in figures B and D represent a significant difference between groups (MANOVA; p < 0.05; n = 20 limbs per group).
In addition, the following variables representing the limb and trunk strategies during the task were quantified: maximal trunk lowering relative to the pelvis, maximal pelvis lowering, and magnitude of hip abduction of the reaching limb. Stance limb strategy was further analysed using the maximal peak of flexion of three joints (hip, knee and ankle) as well as their angular velocity in the sagittal plane (peak to peak value). Finally, for the reaching limb, angular velocity in the frontal plane at the hip was also computed.
Statistical Analyses. All statistical analyses (α = 0.05) were conducted using SPSS for Windows version 12.0. Parametric and descriptive analyses were conducted to meet the research objectives. Non-parametric tests (Mann–Whitney exact test) were used for the comparison of personal characteristics between groups. Descriptive statistics were calculated from the global strategy variables (Objective 1). Multivariate analyses were used to measure group and limb effects on global and segmental strategy variables. No results of univariate ANOVA were reported if the multivariate statistics were not found statistically significant (Objective 1). The horizontal excursion of the CoMgl was analysed using graphical representations (Objective 1). Mean difference (MD) and 95% confidence interval (CI) were calculated for statistically significant results. Regression analyses were carried out to determine which combination of two global variables along the vertical axis was providing the best estimate of SEBT performance for each test condition (using standardized Beta coefficients [β]) (Objective 3).
As the majority of the LAS participants (70%) had injured their dominant limb, performance and strategy variables in all test conditions of the LAS participants were compared, for the second research objective, to those obtained from the dominant limb of the healthy group. A bar graph representing all regression coefficient combinations was used to compare each component's contributions to global lowering for each reaching direction (see Figure 3, Objective 2). Peak-to-peak angular velocities in the stance limb were compared through multivariate analyses to measure the group and limb effects (Objective 1). The hip abduction velocity of the reaching limb was analysed at three times of measure (25, 50 and 75% of the central phase duration) to look for potential group effects for the injured and the uninjured limbs (repeated measured ANOVA) (Objective 1).
(Enlarge Image)
Figure 3.
Segmental motor strategy variables derived from global motor strategy variables. A: Examples (mean profiles ±1 standard deviation) for a typical participant of different contributions of the pelvis and trunk lowering to the global CoM lowering between AM and PM directions in the central phase. B: Contribution of segmental strategy variables to the global CoM lowering represented by beta standardized coefficients (β) for each reaching direction and each lower limb. Each bar graph represents a combination of the dominant limb of the healthy group with either the uninjured limb (D-UI limb) or the injured limb (D-I limb) of the LAS group (n = 20). C: Hip ADD angular velocities of the reaching limb for the injured limb (dotted line) and the control group (black line and shaded area; mean ± 1 standard deviation) in each direction and a summary (mean velocity ±1 SD) at three different time points in the central phase for the medial direction. Asterisks represent a significant difference between groups (MANOVA; p < 0.05).
Methods
Participant Selection
All volunteers participating in this study were selected from the Canadian Forces (CF) military population. The LAS group was composed of 10 men with a diagnosis of acute unilateral LAS. They all received physiotherapy interventions within five days of injury and were discharged after a maximum of nine weeks. They were excluded if they had an ankle fracture (documented by X-ray), a third grade LAS or a high ankle sprain (tibiofibular sprain), or if they reported a history of neuromuscular or neurodegenerative disease. The control group consisted of a comparable group of 10 men without prior history of LAS or symptoms from the lumbar spine and the lower limbs, or a history of neuromuscular and neurodegenerative disease. The present study was approved by the ethics committees of the Quebec Rehabilitation Institute and the CF Health Services Group. All participants read and signed an informed consent form.
Study Design
In this descriptive study, all participants were evaluated within a single session. Participants in the LAS group were selected in acute stage (within two days after injury) and were evaluated between eight and ten weeks after their injury. Personal and clinical characteristics of the participants were collected. Their functional ability was measured using two questionnaires: the Foot & Ankle Disability Index (FADI) and the Lower Extremity Function Scale (LEFS). Thereafter, they took part in a series of complex motor tasks and tests, where the testing order was kept the same for all participants.
SEBT Procedure
During the SEBT, the participant had to: (1) touch the floor as far as possible with the tip of the foot of the reaching limb in three different directions with respect to the stance limb (anteromedial [AM], medial [M] and posteromedial [PM] directions) and (2) return to unipedal stance after each reaching movement while maintaining balance on the stance limb with the hands resting on the iliac crests. The AM, M and PM directions were the ones proposed by Hertel et al. because these directions were able to identify significant reach deficits associated with chronic ankle instability and provided complementary and non-redundant information. After a practice session (six trials/direction followed by a five-minute rest period), three successful trials, separated by a ten-second rest period, were recorded for each direction (Figure 1). Trials were rejected according to the criteria used in previous studies (1) weight-bearing by the reaching limb, (2) displacement of the stance limb, or (3) loss of balance. As shown in Figure 1, the directions to be followed by the reaching limb during the tests were clearly marked on the floor using graduated tape. The order of test conditions (two legs and three directions) was determined as follows: first the leg order was randomly selected, then a test direction was randomly determined. All conditions for one leg were completed starting with the determined direction followed by the next anticlockwise direction (right leg) or clockwise direction (left leg). The same condition order was then used for the second leg.
(Enlarge Image)
Figure 1.
Kinematics variables during the SEBT task, experimental setting and description of subtasks and events. Illustration of the six conditions at the SEBT (3 reaching directions per limb; AM: anteromedial, M: medial, PM: posteromedial) and profiles (mean ± 1 standard deviation; n = 3 trials) of the vertical position of the reaching foot (A), global body CoM (CoMgl) (B and C) and joint amplitude of motion of the stance limb (D) in a typical healthy subject in the PM reaching direction. The central grey rectangle, called the central phase around foot contact, represents the critical period used for the analyses. The lowest vertical position of the tip of the reaching foot was used to subdivide the task, which corresponds to the "foot contact" event (vertical line at 50% of the task). The first subtask is characterized by body lowering (B and D) and alignment of the reaching limb for foot contact (A). In the second subtask which occurred after foot contact, a rapid straightening up of the entire body (B and C) is performed while returning from perturbation in single-limb stance. The transition between subtasks represents a highly challenging period for stability and motor control as a change in movement direction takes place at the perceived limits of stability. This transition period, called the central phase, was defined as -1 s to +1 s after foot contact. Further analysis about the quality of motor control was performed during this phase.
Data Collection
During the SEBT, biomechanical data were collected to characterize both whole body and segmental strategies using a 3D motion capture system (Optotrak 3020; Northern Digital Inc., Waterloo, Ontario, Canada). Forty-five infrared markers placed in non-colinear triads were installed on the following body segments: lower limbs, head, and trunk. For uppers limbs, rings of three markers were placed around the wrists. Kinematic data was sampled at 50 Hz and then digitally filtered at 10Hz (low pass filter). Global body CoM (CoMgl) was estimated using the equations proposed by Winter, and adjusted to consider the influence of the upper limb CoM.
Data Analysis
All individual kinematic data analyses were based on a mean of three trials per test condition. A mean of two trials was exceptionally used in a very few cases (5 out of 120 conditions). The maximal reach distance (MRD) on the SEBT was calculated from the horizontal plane coordinates (vectorial distance) of the tip of the reaching foot (probed point) at the time it touched the floor.
Key Variables Used to Characterize Motor Control. The SEBT task was divided into two main subtasks: the going-to- and the return-from-MRD (Figure 1A). The transition between these subtasks has been called the central phase and was defined as -1 s to +1 s after foot contact (Figure 1: central grey rectangle). Four variables describing the behaviour of the CoMgl, called global variables, have been chosen as key indicators of global body strategy during this task. The maximal displacement of the CoMgl vertically; peak-to-peak CoMgl velocities along the vertical axis were calculated during the central phase (Figure 2A); distance between the centre of the foot and the mean position of the horizontal CoMgl when the foot contacted the floor and the horizontal excursion of the CoMgl during the central phase were also calculated (Figure 2C).
(Enlarge Image)
Figure 2.
Global body strategies along the vertical axis (A) and horizontal plane (C) in the central phase. A: the profiles for global CoM lowering (#) and for the peak-to peak velocity of the global CoM (# #) are illustrated for one limb of both groups (mean ± 1 standard deviation for healthy group and mean for LAS group; n = 10 trials per group) during the medial reaching direction on the SEBT. B: mean values (+1 standard deviation) for global CoM lowering and peak-to peak of CoM vertical velocity of both groups. C: CoM displacement in the horizontal plane (# # #) during the different conditions (directions and limbs) in both groups. Horizontal resultant lines of global CoM position at foot contact (doted circles) were calculated and used for further statistical analyses. D: mean values (+1 standard deviation) for global CoM in the horizontal plane. Asterisks in figures B and D represent a significant difference between groups (MANOVA; p < 0.05; n = 20 limbs per group).
In addition, the following variables representing the limb and trunk strategies during the task were quantified: maximal trunk lowering relative to the pelvis, maximal pelvis lowering, and magnitude of hip abduction of the reaching limb. Stance limb strategy was further analysed using the maximal peak of flexion of three joints (hip, knee and ankle) as well as their angular velocity in the sagittal plane (peak to peak value). Finally, for the reaching limb, angular velocity in the frontal plane at the hip was also computed.
Statistical Analyses. All statistical analyses (α = 0.05) were conducted using SPSS for Windows version 12.0. Parametric and descriptive analyses were conducted to meet the research objectives. Non-parametric tests (Mann–Whitney exact test) were used for the comparison of personal characteristics between groups. Descriptive statistics were calculated from the global strategy variables (Objective 1). Multivariate analyses were used to measure group and limb effects on global and segmental strategy variables. No results of univariate ANOVA were reported if the multivariate statistics were not found statistically significant (Objective 1). The horizontal excursion of the CoMgl was analysed using graphical representations (Objective 1). Mean difference (MD) and 95% confidence interval (CI) were calculated for statistically significant results. Regression analyses were carried out to determine which combination of two global variables along the vertical axis was providing the best estimate of SEBT performance for each test condition (using standardized Beta coefficients [β]) (Objective 3).
As the majority of the LAS participants (70%) had injured their dominant limb, performance and strategy variables in all test conditions of the LAS participants were compared, for the second research objective, to those obtained from the dominant limb of the healthy group. A bar graph representing all regression coefficient combinations was used to compare each component's contributions to global lowering for each reaching direction (see Figure 3, Objective 2). Peak-to-peak angular velocities in the stance limb were compared through multivariate analyses to measure the group and limb effects (Objective 1). The hip abduction velocity of the reaching limb was analysed at three times of measure (25, 50 and 75% of the central phase duration) to look for potential group effects for the injured and the uninjured limbs (repeated measured ANOVA) (Objective 1).
(Enlarge Image)
Figure 3.
Segmental motor strategy variables derived from global motor strategy variables. A: Examples (mean profiles ±1 standard deviation) for a typical participant of different contributions of the pelvis and trunk lowering to the global CoM lowering between AM and PM directions in the central phase. B: Contribution of segmental strategy variables to the global CoM lowering represented by beta standardized coefficients (β) for each reaching direction and each lower limb. Each bar graph represents a combination of the dominant limb of the healthy group with either the uninjured limb (D-UI limb) or the injured limb (D-I limb) of the LAS group (n = 20). C: Hip ADD angular velocities of the reaching limb for the injured limb (dotted line) and the control group (black line and shaded area; mean ± 1 standard deviation) in each direction and a summary (mean velocity ±1 SD) at three different time points in the central phase for the medial direction. Asterisks represent a significant difference between groups (MANOVA; p < 0.05).
Source...