Movement variability during single leg jump landings in individuals with and without chronic ankle instability

Clinical Biomechanics 27 (2012) 52–63

Contents lists available at ScienceDirect

Clinical Biomechanics j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / c l i n b i o m e c h

Movement variability during single leg jump landings in individuals with and without chronic ankle instability Cathleen Brown a,⁎, Bradley Bowser b, Kathy J. Simpson a a b

Department of Kinesiology Biomechanics Laboratory, University of Georgia, USA Department of Health and Nutritional Sciences, South Dakota State University, SD, USA

a r t i c l e

i n f o

Article history: Received 25 July 2010 Accepted 21 July 2011 Keywords: Mechanical instability Functional instability Motion analysis Kinematics Kinetics

a b s t r a c t Background: Repeated episodes of giving way at the ankle may be related to alterations in movement variability. Methods: Eighty-eight recreational athletes (39 males, 49 females) were placed in 4 groups: mechanically unstable, functionally unstable, copers, and controls based on ankle injury history, episodes of giving way, and joint laxity. Lower extremity kinematics and ground reaction forces were measured during single leg landings from a 50% maximum vertical jump in the anterior, lateral, and medial directions. Ensemble curves of 10 trials were averaged and coefficients of variation were identified for ankle, knee, hip, and trunk motion in 3 planes. A loge (ln) transformation was performed on the data. Mixed model analyses of variance (ANOVAs) with Tukey post-hoc tests were utilized with Bonferroni corrections to α ≤ 0.008. Findings: At the knee, controls were more variable than functionally unstable and copers for knee rotation before initial contact, and were more variable during stance than functionally unstable in knee rotation (P ≤ 0.008). Interactions during stance revealed controls were more variable than functionally unstable in lateral jumps for hip flexion, and than mechanically and functionally unstable in hip abduction in the anterior direction (P ≤ 0.008). Controls were more variable than all other groups in hip flexion and than mechanically unstable in hip abduction (P ≤ 0.008). Interpretation: Individuals with ankle instability demonstrated less variability at the hip and knee compared to controls during single leg jump landings. Inability to effectively utilize proximal joints to perform landing strategies may influence episodes of instability. © 2011 Elsevier Ltd. All rights reserved.

1. Introduction Lateral ankle sprains are one of the most common sports-related injuries (Fong et al., 2007; Hootman et al., 2007). It is estimated up to 73% of people who suffer lateral ankle sprains develop chronic ankle instability (Yeung et al., 1994), defined as subjective and repeated episodes of giving way at the ankle (Hertel, 2002). Some individuals with chronic ankle instability demonstrate physiologically lax lateral ligaments, and can be classified as mechanically unstable (Hertel, 2002). Alternately, some individuals are not ligamentously lax, but still complain of giving way, perhaps due to sensorimotor deficits. These individuals may be classified as functionally unstable (Hertel, 2002). History of ankle trauma, including lateral ankle sprains and chronic instability, is related to development of ankle osteoarthritis (Valderrabano et al., 2006). With a traumatic history, ankle osteoarthritis may be preventable if lateral ankle sprains and chronic ankle instability can be adequately treated (Valderrabano et al., 2006). ⁎ Corresponding author at: Department of Kinesiology, University of Georgia, 330 River Rd, Athens, GA 30602, USA. E-mail address: [email protected] (C. Brown). 0268-0033/$ – see front matter © 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.clinbiomech.2011.07.012

There is currently a lack of consensus on the criteria used to define those with chronic ankle instability, and what factors contribute to chronic instability (Delahunt et al., 2010). Factors contributing to functional instability, such as deficits in sensorimotor control, strength, or dynamic balance have been identified (Hertel, 2002; Hubbard et al., 2007; Wikstrom et al., 2007). Mechanical factors such as physiologic lateral ligament laxity, in combination with functional factors, have also been identified (Brown et al., 2008; Hubbard et al., 2007). Additionally, comparison groups with a history of sprain but no functional limitations have been identified as “copers,” and offer a relevant clinical model comparison (Brown et al., 2008; Hertel and Kaminski, 2005; Wikstrom et al., 2010a). We are attempting to separate individuals with chronic ankle instability into those with primarily functional deficits and those with primarily functional but also secondary mechanical deficits. We are attempting to tease out the contributions of primarily functional, but also secondary mechanical, deficiencies and compare performance on movement variability measures to copers without functional or mechanical deficits and an uninjured control group that has never been exposed to functional or mechanical contributing factors.

C. Brown et al. / Clinical Biomechanics 27 (2012) 52–63

Movement variability has been suggested as a factor contributing to developing and perpetuating injury (James, 2004; Konradsen, 2002; Konradsen and Voigt, 2002). Increased variability has been linked to risk of other musculoskeletal injuries including anterior cruciate ligament rupture (McLean et al., 1999), lower extremity overuse injury (James et al., 2000), patellofemoral pain (Heiderscheit et al., 2002), and injuries from falls (Hausdorff et al., 2001). There are a number of methods to quantify variability. One traditional, commonly used method is the coefficient of variation, the standard deviation normalized to the mean of the score distribution, representing relative or normalized variability converted to a percentage of the mean value (James, 2004). For the purposes of this study, the following operational definition of movement variability was used: variability in a single plane motion of a joint or body segment during time periods corresponding to transition from an unloaded to a loaded state (250 ms pre-initial contact to initial contact) and in transition from landing to single limb standing (1 s postinitial contact). The dependent variable, coefficient of variation, was calculated for those joints/segments in those specific windows of time associated with specific events, across multiple trials for each participant. Strengths of the coefficient of variation include its historical use, ease of interpretation and ability to compare performances with very different mean scores (James, 2004). Limitations of the measure include the influence of outlying or extreme data points and its reliance on the mean and standard deviation, particularly when the mean is close to zero (James, 2004). Additionally, this operational definition and dependent variable encompass only a single joint/segment at a discrete point in time, and does not compare coupling relationships in joints as others have done (Drewes et al., 2009). Few studies have determined if differences in motion variability exist in individuals with chronic ankle instability, particularly in proximal joints and during challenging sports-related movements. Applying dynamical systems theory, sensorimotor organization and coordination of degrees of freedom depends on the interaction of a number of factors, including task complexity, predictability of the environment, and the health of the person (James, 2004). Individuals adapt to a particular task using movement variability to deal with the personal, task, or environmental constraints (Davids et al., 2003). If ankle instability is considered a constraint on the system (Brown et al., 2009; McKeon and Hertel, 2008), the entire kinetic chain system (ankle, knee, hip, and trunk) should be assessed for variability. Specifically, variability during transitions, such as from jump landing to stance, may prove important in degree of movement variability with regards to injury. These periods of transition from unloaded to loaded state have been associated with ankle injury (Konradsen and Voigt, 2002). Individuals may prepare for transitions by increasing variability in certain joints while decreasing in others. Too much variability in the kinetic chain during transitions may hinder the ability of an individual with chronic ankle instability (CAI) to develop effective solutions to adapt to demands of specific sports-related tasks and the environmental constraints, indicating inability to coordinate and execute movement goals. Too little variability during transitions may hinder the ability of an individual with CAI to include a variety of degrees of freedom into an effective movement solution, indicating inability to adapt to new situations or changing situations. Proximal differences in joint motion and electromyography have been noted in chronic ankle instability populations (Bullock-Saxton, 1994; BullockSaxton et al., 1994; Caulfield and Garrett, 2002; Gribble and Robinson, 2009), and variability of motion may influence those findings. Evidence from a previous study indicated increased variability in a single plane at the ankle joint during a double limb landing transitioning from a run to stop jump to vertical jump (Brown et al., 2009). However, it is currently unclear what effects chronic ankle instability has on single joint movement variability in single limb landing. This study seeks to improve and expand on previous work associating variability with the constraints of chronic instability by utilizing transitions from single leg jump landing to single leg stance

53

in a task that includes 3 different jump directions that provide different demands and constraints. An uninjured control group to assess “typical” variability is added and all 3 planes of motion and trunk kinematics will be included. Thus, the purpose of this study was to determine if differences in movement variability exist at the ankle, knee, hip, and trunk in 3 planes of motion between individuals with mechanical ankle instability, functional ankle instability, copers, and controls during a single leg jump landing movement in anterior, lateral, and medial directions. We hypothesized that individuals with mechanically and functionally unstable ankles would demonstrate greater variability compared to copers and controls at each joint during the 250 ms before and the 1 s time period after landing. We selected these time frames as important windows where transitions may be occurring between jumping and landing and between landing and single limb stance. We also hypothesized increased variability based on previous literature (Brown et al., 2009; Delahunt et al., 2006; Drewes et al., 2009; Konradsen and Voigt, 2002), though with this novel analysis, it was unclear how variability might change between joints and planes of motion. 2. Methods 2.1. Participants A previous study using coefficient of variation to assess kinematic variability of mechanically unstable, functionally unstable, and coper groups during stop jump maneuvers indicated that 25–140+ participants were necessary to achieve a power of 80 on ankle and knee sagittal and frontal plane motion, and hip sagittal and transverse plane motion during a stop jump (Brown et al., 2009). Another study analyzing gender differences during sidestep kinematics utilized mean standard deviation across 10 trials to measure variability and indicated 8–50 participants were necessary for a power of 80 for knee sagittal and frontal plane motion and hip transverse plane motion (McLean et al., 2004). As a preliminary study on single joint movement variability alterations in an unstable ankle population, we utilized different tasks and slightly different measures than previous studies. Thus, we used the a-priori power calculations as guidelines. Based on these calculations and on other dependent variables not reported in this study, group sizes of 20–24 were estimated to provide adequate power for most dependent variables. A total of 88 recreationally active participants, defined as performing at least 1.5 total hours of cardiovascular, resistance, sports-related, or other physical activity per week, were recruited for the study and placed into 1 of 4 groups. Mechanically and functionally unstable ankle individuals had a history of at least 1 moderate-severe ankle sprain at least 12 months ago that resulted in 3 days of immobilization, (Brown et al., 2009; Ross and Guskiewicz, 2006) and scored ≤25 on the Cumberland Ankle Instability Tool (CAIT) a self-reported measure of function (Hiller et al., 2006). Coper individuals had only 1 sprain meeting the same criteria and scored 25–28 on the CAIT and had no episodes of giving way (Brown et al., 2009). Mechanically and functionally unstable ankle individuals complained of ≥2 episodes of giving way in the past 12 months. Mechanically unstable individuals demonstrated laxity of lateral ankle ligaments to anterior drawer and/or talar tilt orthopedic tests (Beynnon et al., 2001; Hoppenfeld, 1976; Tropp et al., 1985); functionally unstable and coper individuals demonstrated no laxity (Brown et al., 2009). A single tester, with over 7 years of clinical experience, performed the orthopedic tests. Control individuals had no history of lateral ankle sprain nor did they display ligamentous laxity, and scored ≥28 on the CAIT. (Hiller et al., 2006) Exclusion criteria for all groups were history of fracture or surgery in either leg, swelling, pain or discoloration at the ankle at testing, or diagnosis of a vestibular or balance disorder, Charcot–Marie–Tooth disorder, or other neurologic disorder. Exclusion criteria also included

54

C. Brown et al. / Clinical Biomechanics 27 (2012) 52–63

Table 1 Participant demographics mean (SD). n

MAI Males Females FAI Males Females Copers Males Females Controls Males Females

Age (years)

Height (cm)

Weight (kg)

CAIT score

50% Jump height (cm) Anterior

Medial

Lateral

Range of motion (degrees) Plantarflexion

Dorsiflexion

Inversion

Eversion

8 13

18.6(3.3) 19.9(1.0)

177.1(6.2) 169.6(54)

72.8(9.1) 66.1(7.2)

18.6(3.3) 17.7(3.1)

17.3(4.6) 20.6(4.8)

17.3(7.9) 18.3(5.3)

15.7(4.1) 18.8(4.6)

61.0(9.3) 67.1(6.7)

6.4(3.2) 5.0(4.0)

15.1(4.4) 17.7(4.8)

8.8(2.0) 9.1(37)

11 12

20.5(1.7) 20.1(1.5)

179.8(7.8) 167.5(6.5)

78.5(9.4) 63.5(9.4)

19.0(3.6) 20.3(3.2)

19.0(5.3) 24.6(5.8)

17.3(6.1) 22.4(6.1)

16.8(6.4) 21.6(5.1)

55.6(7.6) 63.7(6.6)

7.0(4.8) 7.0(4.8)

13.9(7.2) 14.3(3.7)

7.2(3.3) 7.3(2.1)

8 12

19.8(1.2) 20.3(1.1)

184.1(5.2) 165.7(6.6)

79.3(5.7) 62.1(6.3)

26.8(3.0) 25.8(1.6)

23.6(6.4) 20.6(4.8)

20.0(4.3) 19.3(4.8)

19.8(4.6) 19.8(4.8)

60.4(6.0) 61.9(6.1)

14.5(18.5) 6.0(2.6)

15.4(5.9) 18.4(5.0)

6.9(2.2) 8.1(2.2)

12 12

19.8(13) 20.2(1.0)

175.5(6.3) 166.7(4.9)

69.6(8.1) 61.2(9.7)

28.8(1.7) 28.2(1.3)

22.6(5.3) 19.0(4.3)

20.3(56) 17.3(3.81)

20.3(5.6) 16.5(3.6)

57.8(6.9) 64.3(6.1)

9.2(4.0) 8.3(1.9)

17.7(6.3) 15.6(5.2)

6.7(2.7) 7.2(1.5)

n: number of participants; MAI: mechanical ankle instability; FAI: functional ankle instability.

an injury to the leg, other than at the ankle for the mechanically and functionally unstable ankle groups, in the last 3 months (Brown et al., 2009; Gribble and Robinson, 2009). Group demographics are reported in Table 1.

2004). Unsuccessful trials were not analyzed but a count was kept. The remaining 2 jump directions were completed with 5 min of rest in between directions while the Vertec height was adjusted and moved accordingly.

2.2. Instrumentation 2.4. Data processing, reduction, analysis, and interpretation An AMTI force platform (OR-6-6-0™, Advanced Medical Technologies Inc., Watertown, MA, USA) collected ground reaction force data at 1200 Hz and was synchronized with a Vicon Motion System (Vicon Motion Systems Ltd., Oxford, UK) which collected kinematic data at 240 Hz. Seven Vicon MX40 cameras and Workstation software (v5.2.4) captured the spatial location of retroreflective markers. The markers were placed utilizing a modified Helen-Hayes marker set (Chappell et al., 2002). An L-frame (Ergacal — 14 mm markers, Vicon Motion Systems Ltd., Oxford, UK) and a calibration wand (240 mm wand — 14 mm markers, Vicon Motion Systems Ltd., Oxford, UK) were utilized for static and dynamic calibration within the capture space. The global reference system was established similar to published guidelines (Wu et al., 2002). The 250 ms before and the first 1 s of data following initial contact (defined as N10 N vertical ground reaction force) were analyzed. 2.3. Data collection Participants completed a written informed consent as approved by the Institutional Review Board. Demographic data including limb dominance was measured (Hoffman et al., 1998), as was active ankle range of motion (Norkin and White, 1995). A 10 min warm-up, including stationary biking and self-directed stretching, was performed. Maximum vertical jump height was measured as previously reported (Ross and Guskiewicz, 2004) using a 2 legged jump for maximal height. Participants touched the flags of a vertical jump height measure (Vertec, Sports Imports, Columbus, OH, USA) with the reaching arm shoulder fully extended (Brown et al., 2004; Ross and Guskiewicz, 2004). The best of 3 trials was used. The other 2 jump directions were tested in the same manner except the participant jumped the same distance laterally or medially for the test limb. Jump direction order was randomized for testing, and the 50% height of the first direction's maximum vertical jump was calculated and set on the Vertec, adding 5% above and below the 50% mark to provide a target jump height range (Ross and Guskiewicz, 2004). Participants stood 70 cm from the center of an inground force platform, took off of 2 ft., fully extended 1 arm and touched the target range markers, landed on the test leg only with the test foot entirely on the force platform and stabilized (Ross and Guskiewicz, 2004). A minimum of 3 practice trials were completed, then 10 successful test trials. A successful trial included jumping 70 cm from the center of the platform, fully extending the shoulder to touch the markers, landing with the foot entirely on the force platform, and not moving or sliding the foot following landing (Ross and Guskiewicz,

Vicon Workstation was used to reconstruct the 3D marker trajectories. The MotionMonitor™ C3D Model Builder (Innovative Sports Training, Chicago, IL, USA) was used to import C3D data files from Workstation, and define joint centers and local coordinate systems from a static file. Joint kinematics for the ankle, knee, hip, and trunk in 3 planes were calculated. Data from the left leg were aligned to the right leg, and Euler angles were used similar to previously published guidelines (Wu et al., 2002). DataPac 2 K2 (Version 3.11, RUN Technologies, Mission Viejo, CA, USA) was used to determine initial contact (the forceplate registered N10 N of vertical force). A low-pass 4th-order, non-recursive Butterworth filter (cut-off frequency of 15 Hz) was applied to kinematic data (Yu et al., 1999). Each trial was normalized to two 101 point periods for the established periods before (250 ms) and after (1 s) initial contact (Fig. 1). Then the 10 normalized trials for 1 participant were combined to create average ensemble curves for each participant, 1 curve for pre-initial contact, and 1 curve for after initial contact (stance). The standard deviation for each data point on the mean ensemble curve for each individual in each time period were calculated before entry into the calculation of the average standard deviation for each group. Data were exported as ASCII files. A grand mean standard deviation (SD), the average standard deviation (SDavg) and average coefficient of variation (CVavg) were computed using an Excel spreadsheet (Microsoft Inc., Redmond, WA, USA) using previously published guidelines (Brown et al., 2009; James, 2004). Thus, the CVavg was calculated as a collapsed measure of all points within the curve. Statistical software (PASW, Version 18.0, SPSS Inc., Chicago, IL, USA) was used to conduct one-way analyses of variance (ANOVAs) to determine if the groups were statistically different in demographics,

Single Leg Jump Landing Time Definitions Initial Ground Contact Force platform >10N

-250ms -125ms 0 250ms Normalized to 101 Points

500ms 1 second Normalized to 101 points

Fig. 1. Single leg jump landing time definitions.

Time

Table 2 Mixed model ANOVA CVln for group by jump direction for pre-initial contact. Jump direction mean (standard deviation)

Main effect group between group

95% Confidence interval Variable

Group

Ankle plantarDorsiflexion

MAI

MAI

1.28 (0.60) FAI 1.34 (0.60) Coper 1.01 (0.26) Control 1.40 (0.77)

Ankle internal– MAI external rotation FAI

2.95 (0.65) 3.24 (1.02) Coper 2.82 (0.80) Control 3.09 (0.89)

Knee flexion– extension

MAI

4.25 (0.66) FAI 4.24 (0.85) Coper 3.85 (0.55) Control 4.56 (0.62)

95% Confidence interval

Lower limit

Upper limit

Medial Lower limit

Upper limit

Total

Lower Limit

Upper limit

Comparison Mean PDifference value

95% CI for Difference

1.63

2.12

2.00

2.54

2.75

2.36

MAI–FAI

0.032

0.786

2.10

2.56

1.92

2.44

2.00

2.32

MAI-coper

0.022

0.857

1.47

2.09

2.14

2.55

1.99

2.77

2.00

2.34

0.186

2.36

2.10

2.54

2.24

3.08

2.19

2.50

MAIcontrol FAI-coper

− 0.152

1.74

2.19 (0.65) 2.16 (0.59) 2.17 (0.71) 2.34 (0.80)

2.03

2.22

2.43 (0.71) 2.18 (0.61) 2.38 (0.83) 2.66 (0.98)

2.11

1.71

2.27 (0.600) 2.33 (0.54) 2.34 (0.44) 2.32 (0.52)

− 0.010

0.933

FAI-control

− 0.184

0.103

− 0.174

0.136

− 0.141

0.304

− 0.197 to 0.260 − 0.215 to 0.258 − 0.378 to 0.074 − 0.241 to 0.221 − 0.404 to 0.037 − 0.403 to 0.055 − 0.409 to 0.128 − 0.073 to 0.483 − 0.165 to 0.367 0.073 to 0.618 − 0.018 to 0.501 − 0.373 to 0.166 − 0.433 to 0.095 − 0.278 to 0.268 − 0.326 to 0.196 − 0.103 to 0.432 − 0.151 to 0.359 − 0.325 to 0.205 − 0.363 to 0.229 − 0.193 to 0.419 − 0.581 to 0.068 − 0.120 to 0.480 − 0.444 to 0.129 − 0.625 to − 0.041

1.00

1.55

1.08

1.6

0.89

1.13

1.07

1.72

2.65

3.25

2.80

3.68

2.44

3.19

2.71

3.46

3.95

4.55

3.87

4.6

3.60

4.11

4.30

4.82

2.11 (1.08) 2.16 (1.25) 1.97 (0.99) 1.61 (0.68)

3.88 (0.44) 3.90 (0.55) 3.68 (0.79) 3.81 (0.76)

3.86 (0.76) 4.02 (0.92) 3.96 (0.72) 4.00 (1.35)

1.62

2.60

1.62

2.70

1.50

2.43

1.32

1.90

3.67

4.08

3.66

4.13

3.31

4.05

3.49

4.13

3.51

4.21

3.63

4.42

3.62

4.30

3.43

4.57

1.55 (0.51) 1.86 (1.00) 1.35 (0.50) 1.63 (0.48)

3.88 (0.87) 4.08 (0.71) 4.22 (0.95) 4.00 (0.61)

3.92 (0.62) 3.98 (0.66) 3.88 (0.76) 4.15 (1.29)

1.32

1.78

1.43

2.30

1.11

1.58

1.43

1.84

3.49

4.27

3.78

4.39

3.78

4.67

3.74

4.26

3.64

4.21

3.70

4.27

3.53

4.24

3.60

4.69

1.65 (0.84) 1.79 (1.03) 1.44 (0.76) 1.54 (0.66)

3.57 (0.80) 3.74 (0.85) 3.57 (1.01) 3.63 (0.85)

4.01 (0.69) 4.08 (0.81) 3.90 (0.67) 4.24 (1.14)

1.45

1.84

Coper– control MAI–FAI

1.60

1.97

MAI-coper

0.205

0.148

1.24

1.64

0.101

0.454

1.36

1.73

MAIcontrol FAI-coper

0.345

0.013

FAI-control

0.242

0.068

− 0.104

0.449

− 0.169

0.208

3.38

3.76

Coper– control MAI–FAI

3.56

3.92

MAI-coper

− 0.005

0.972

3.38

3.77

− 0.065

0.626

3.46

3.81

MAIcontrol FAI-coper

0.164

0.227

0.104

0.421

− 0.060

0.656

3.80

4.23

FAI– control Coper– control MAI–FAI

− 0.067

0.655

3.88

4.28

MAI-coper

0.113

0.468

3.68

4.12

− 0.225

0.132

4.04

4.44

MAIcontrol FAI-coper

0.180

0.237

FAI-control

− 0.158

0.279

Coper– control

− 0.338

0.026

Comparison Mean Pdifference value

95% CI for difference

Cop-ant: Con-Ant

− 0.708

0.007 − 1.22 to − 0.194

(continued on next page)

55

Lateral

C. Brown et al. / Clinical Biomechanics 27 (2012) 52–63

Ankle inversion– eversion

95% Confidence interval

Upper limit

Anterior Lower limit

1.88 (0.55) FAI 1.97 (0.58) Coper 1.78 (0.66) Control 2.05 (0.74)

95% Confidence interval

Between group comparisons for jump direction

56

Table 2 (continued) Jump direction mean (standard deviation)

Main effect group between group

95% Confidence interval Variable

Group

Knee abduction– adduction

MAI

2.92 (1.26) 2.85 (1.12) Coper 2.53 (0.68) Control 3.19 (1.52)

Hip flexion

Hip abduction– adduction

MAI

3.45 (0.53) FAI 3.26 (0.65) Coper 3.28 (0.66) Control 3.84 (1.13)

MAI

3.16 (0.43) FAI 3.45 (0.68) Coper 3.29 (0.90) Control 3.76 (1.21)

Hip internal– MAI external rotation FAI

3.87 (1.07) 3.87

95% Confidence interval

Lateral

Lower limit

Upper limit

Medial Lower limit

Upper limit

Total

Lower Limit

Upper limit

Comparison Mean PDifference value

1.22

1.58

2.96

4.12

4.45

3.25

3.44

4.27

3.84

4.49

2.92

3.46

0.87

1.23

3.12

4.33

3.32

4.27

2.57

3.14

1.08

2.54

3.00

4.38

3.40

4.31

2.97 (1.49) 3.19 (1.45) 2.86 (1.61) 3.11 (1.75)

2.70

1.92

3.98 (1.03) 4.16 (0.75) 3.80 (1.02) 3.85 (1.08)

3.51

1.18

3.54 (1.28) 3.85 (0.96) 3.73 (1.30) 3.69 (1.63)

2.86

3.38

MAI– FAI MAIcoper MAIcontrol FAICoper FAIcontrol Coper– control MAI– FAI MAIcoper MAIcontrol FAI-Coper

2.35

3.49

2.36

3.33

2.21

2.84

2.55

3.84

3.20

3.69

2.98

3.54

2.97

3.59

3.37

4.32

2.97

3.36

3.16

3.74

2.86

3.71

3.24

4.27

3.38

4.35

3.47

4.27

1.81 (0.70) 2.02 (1.02) 1.74 (0.64) 2.40 (1.58)

4.08 (0.88) 3.98 (0.76) 4.13 (0.71) 4.55 (1.38)

3.49 (0.52) 3.34 (0.57) 3.29 (0.76) 3.55 (1.00)

2.59 (0.81) 2.81

1.49

2.13

1.58

2.46

1.44

2.03

1.73

3.07

3.68

4.48

3.66

4.31

3.80

4.47

3.96

5.13

3.25

3.72

3.09

3.58

2.93

3.64

3.13

3.98

2.22

2.96

2.33

3.30

2.04 (1.03) 1.68 (0.88) 1.86 (0.80) 2.49 (1.50)

4.03 (0.90) 3.91 (0.72) 4.08 (0.95) 4.63 (1.83)

3.68 (0.72) 4.08 (0.81) 3.93 (1.04) 4.12 (1.39)

2.48 (0.90) 2.39

1.57

2.51

1.30

2.06

1.49

2.23

1.86

3.13

3.62

4.43

3.60

4.22

3.63

4.52

3.85

5.40

3.35

4.01

3.73

4.42

3.45

4.42

3.53

4.71

2.08

2.89

2.05

2.74

2.25 (1.12) 2.18 (1.11) 2.04 (0.78) 2.70 (1.55)

3.85 (0.83) 3.72 (0.77) 3.83 (0.87) 4.34 (1.50)

3.44 (0.60) 3.62 (0.76) 3.50 (0.94) 3.81 (1.22)

2.98 (1.11) 3.03

1.97

2.54

1.91

2.45

1.75

2.39

2.43

2.96

− 0.215

0.272

0.116

0.566

− 0.143

0.459

0.331

0.095

0.072

0.704

− 0.260

0.186

0.073

0.711

0.214

0.294

− 0.441

0.024

0.141

0.480

FAI-control

− 0.514

*0.007

−0.655

*0.001

0.134

0.445

3.60

4.10

Coper– control MAI–FAI

3.48

3.96

MAI-coper

0.022

0.904

3.57

4.09

− 0.486

*0.006

4.10

4.57

MAIcontrol FAI-coper

− 0.112

0.528

FAI-control

− 0.620

*0.001

− 0.508

*0.004

− 0.177

0.255

3.22

3.66

Coper– control MAI–FAI

3.41

3.83

MAI-Coper

− 0.060

0.708

3.28

3.73

− 0.368

0.017

3.61

4.02

MAIcontrol FAI-coper

0.177

0.458

FAI-control

− 0.192

0.202

− 0.308

0.049

− 0.046

0.802

0.076

0.685

2.72

3.24

Coper– control MAI–FAI

2.78

3.27

MAI-coper

95% CI for Difference − 0.6 to 0.169 − 0.282 to 0.514 − 0.524 to 0.237 − 0.058 to 0.721 − 0.300 to 0.444 − 0.646 to 0.126 − 0.315 to 0.461 − 0.187 to 0.615 − 0.825 to − 0.057 −@@.252 to 0.534 − 0.889 to − 0.140 − 1.044 to − 0.266 − 0.212 to 0.480 − 0.336 to 0.38 − 0.828 to − 0.143 − 0.463 to 0.238 − 0.954 to − 0.286 − 0.855 to − 0.161 − 0.482 to 0.128 − 0.376 to 0.256 − 0.67 to − 0.066 − 0.193 to 0.426 − 0.487 to 0.103 − 0.614 to − 0.002 − 0.402 to 0.311 − 0.293 to

Comparison Mean Pdifference value

95% CI for difference

C. Brown et al. / Clinical Biomechanics 27 (2012) 52–63

Knee internal– MAI external rotation FAI

95% Confidence interval

Upper limit

Anterior Lower limit

1.40 (0.40) FAI 1.55 (0.86) Coper 1.05 (0.38) Control 1.81 (1.72)

95% Confidence interval

Between group comparisons for jump direction

Table 2 (continued) Jump direction mean (standard deviation)

Main effect group between group

95% Confidence interval Variable

Group

Anterior Lower limit

(0.93) 3.49 (0.78) Control 3.84 (1.09)

Coper

Upper limit

3.12

3.85

3.38

4.30

95% Confidence interval Lateral (1.12) 2.77 (1.02) 2.87 (1.19)

Lower limit

Upper limit

2.30

3.25

2.37

3.37

95% Confidence interval Medial Lower limit (0.80) 2.45 (1.14) 2.77 (1.38)

Upper limit

1.92

2.99

2.18

3.35

95% Confidence interval Total (1.13) 2.90 (1.07) 3.16 (1.30)

Lower Limit

Upper limit

Comparison Mean PDifference value

2.64

3.17

2.92

3.40

MAIcontrol FAI-coper

− 0.179 0.122

FAI-Control − 0.133

Trunk flexion

Trunk lateral rotation

3.78 (1.18) FAI 3.54 (1.09) Coper 3.37 (1.05) Control 3.57 (1.20)

MAI

3.13 (0.71) FAI 3.41 (0.98) Coper 3.16 (0.58) Control 3.37 (0.91)

MAI

3.76 (0.67) FAI 3.83 (0.68) Coper 3.92 (0.69) Control 3.88 (0.96)

3.24

4.32

3.07

4.01

2.88

3.86

3.07

4.08

2.81

3.45

2.99

3.84

2.89

3.43

2.98

3.75

3.46

4.06

3.54

4.13

3.60

4.24

3.47

4.28

3.54 (0.87) 3.23 (0.91) 3.35 (1.12) 3.56 (1.14)

3.79 (0.76) 4.34 (1.05) 3.97 (0.77) 3.81 (0.87)

3.29 (0.95) 3.63 (1.08) 3.42 (0.94) 3.35 (1.04)

3.15

3.94

2.84

3.63

2.83

3.87

3.08

4.04

3.44

4.14

3.89

4.79

3.61

4.33

3.44

4.17

2.86

3.73

3.16

4.09

2.98

3.86

2.91

3.79

3.13 (0.87) 3.05 (0.92) 3.32 (1.21) 3.43 (1.07)

3.58 (0.52) 3.99 (1.03) 3.92 (0.73) 3.44 (0.64)

3.01 (0.74) 2.95 (1.13) 2.84 (1.03) 2.92 (0.77)

2.73

3.53

2.65

3.44

2.76

3.89

2.97

3.88

3.34

3.81

3.55

4.44

3.58

4.26

3.17

3.70

2.67

3.34

2.46

3.44

2.35

3.32

2.60

3.25

3.48 (1.01) 3.27 (0.98) 3.35 (1.11) 3.52 (1.13)

3.50 (0.72) 3.91 (1.08) 3.68 (0.78) 3.54 (0.82)

3.35 (0.84) 3.47 (1.04) 3.39 (0.99) 3.38 (1.00)

− 0.255

3.22

3.75

Coper– control MAI–FAI

3.02

3.53

MAI-coper

0.135

3.08

3.62

− 0.036

3.27

3.77

MAIcontrol FAI-coper FAI-control

− 0.247 − 0.171

0.210

− 0.075

3.30

3.70

Coper– control MAI–FAI

3.72

4.11

MAI-coper

− 0.183

3.48

3.89

− 0.035

3.35

3.73

MAIcontrol FAI-coper FAI-control

0.380 0.148

− 0.415

0.232

3.13

3.58

Coper– control MAI–FAI

3.25

3.68

MAI-coper

− 0.037

3.16

3.62

− 0.031

3.17

3.60

MAIcontrol FAI-coper FAI-control

0.084

Coper– control

0.006

− 0.115

0.078

95% CI for Difference

0.446 − 0.532 to 0.174 0.507 − 0.24 to 0.483 0.447 − 0.483 to 0.24 0.161 − 0.613 to 0.103 0.256 − 0.153 to 0.574 0.480 − 0.241 to 0.512 0.843 − 0.396 to 0.324 0.688 − 0.443 to 0.293 0.168 − 0.598 to 0.105 0.355 − 0.536 to 0.193 *0.004 − 0.696 to − 0.133 0.217 − 0.474 to 0.108 0.805 − 0.313 to 0.244 0.110 − 0.053 to 517 *0.006 0.108 to 0.652 0.302 − 0.134 to 0.430 0.469 − 0.427 to 0.197 0.823 − 0.359 to 0.286 0.846 − 0.339 to 0.278 0.626 − 0.238 to 0.394 0.582 − 0.217 to 0.386 0.969 − 0.307 to 0.319

Comparison Mean Pdifference value

95% CI for difference

0.319

C. Brown et al. / Clinical Biomechanics 27 (2012) 52–63

Trunk lateral flexion

MAI

Between group comparisons for jump direction

* Significant at pb 0.008. MAI: mechanical ankle instability; FAI: functional ankle instability.

57

58

C. Brown et al. / Clinical Biomechanics 27 (2012) 52–63

jump height, or active ankle range of motion. Based on skewness of kinematic CV data, violations of assumptions for ANOVA occurred, so a loge (ln) transformation was performed CVln (Brown et al., 2009). Mixed model 4 (groups) × 3 (jump directions) ANOVAs were utilized to assess group differences in kinematic CVln variables. The main effects for group (collapsing the jump direction) and the between groups differences within each jump direction were analyzed. Tukey's Honestly Significant Difference post-hoc testing was applied with a Bonferroni adjustment for the alpha level to ≤0.008 for the 6 multiple comparisons of interest in each dependent variable. Main effects for jump direction (collapsing groups) were not compared, nor were group × jump direction interactions because they were not of interest. 3. Results The preliminary analysis indicated the groups were not statistically significantly different in age, height, weight, active ankle range of motion, or in height jumped in any of the 3 directions (P N 0.05; ηp2 b 0.10). The CAIT scores were significantly different between groups, with both the mechanically and functionally unstable ankle groups scoring significantly lower than both the coper and control groups, indicating decreased function (Table 1) (P b 0.05; ηp2 = 0.67). All following means and standard deviations are presented as transformed scores (CVln). 3.1. Ankle No main effects for group or group differences within a jump direction were noted for pre-initial contact or stance phases for any ankle variables (Tables 2 and 3). 3.2. Knee For pre-initial contact, a group main effect was noted for knee rotation. The functionally unstable and copers were less variable than controls. A group difference within the anterior jump direction was noted: the copers were less variable than controls for knee flexion. For the stance phase, a group main effect was noted that the functionally unstable group was less variable than controls for knee rotation. No other group main effects or group within jump direction differences were noted (Tables 2 and 3). 3.3. Hip For pre-initial contact, a group main effect was noted that mechanically unstable, functionally unstable, and copers were all less variable than controls for hip flexion. The mechanically unstable group was less variable than controls for hip abduction. There were also group differences within jump directions. The functionally unstable group was less variable than controls in hip flexion during the lateral jump. The mechanically and functionally unstable groups were less variable than controls in hip abduction during an anterior jump (Tables 2 and 3). 3.4. Trunk For pre-initial contact, the mechanically unstable and controls were less variable than functionally unstable in trunk lateral flexion. There were no other group main effects or group differences within jump direction (Tables 2 and 3). 4. Discussion The most notable result is that the mechanically and functionally unstable ankle groups, and occasionally copers, appear to have less variability in knee and hip joint motions during single leg jump

landing compared to an uninjured control group. Proximal joint variability appears to be different between these groups, specifically knee internal–external rotation and hip flexion–extension and abduction–adduction during pre-initial contact and stance. We report no differences in ankle motion and limited differences in trunk motion variability. Only mechanically unstable and control groups were less variable than functionally unstable in trunk frontal plane motion during pre-initial contact. Differences across the means of 10 trials were found between groups in terms of single joint movement variability. Taken individually, joints in the kinetic chain are behaving differently between groups in different planes. Specifically, at the knee, the control group appears to be more variable during knee flexion and rotation. At the hip, the control group was more variable in flexion and abduction. The differences indicate that the control group may be positioning these joints more variably in preparation for landing, compared to the previously injured groups. The control group may be less encumbered by the landing task constraints and with limitations from previous injury. Large ranges of motion are available in the sagittal plane for knee and hip, transverse plane for knee, and frontal plane for hip. The increased variability we noted may be due to the increased range available. If ankle instability and previous sprain is considered a constraint on the system, motions at other joints may be limited in their variability of responses and strategies to maintain upright stance during single leg landing, which required oscillation around a fixed foot. If the uninjured control group is considered “normal” with an ideal response, the other groups had too little variability at the knee and hip, and were thus unable to adapt efficiently to the demands of the single leg landing task which may indicate less robust neuromuscular control and sensorimotor response. The participants were still able to complete the task, and secondary analysis indicated there were no statistically significant differences in the number of failed trials between groups for any direction or overall. However, the unstable ankle groups and copers may not be as efficient or effective in finding movement solutions and utilizing proximal joints, as evidenced by increased variability in the CVln. The task required anterior translation, control of rotary stability from reaching for the jump height markers, and a fixed foot on landing. It seems logical that knee and hip flexion, hip abduction, and knee rotation would be required to maintain this position. Variability in these joints and planes prior to landing would likely influence the successful completion of the task. Our hypotheses were not supported, in that individuals with mechanical and functional ankle instability appeared to be less variable than controls, as did copers in certain cases. In the present study, increased variability at the knee and hip may be seen as a benefit: while all groups had to decrease the degrees of freedom at the ankle joint to maintain an upright stance within the confines of the task, only the control group was consistently more variable at the knee and hip. Individuals with ankle instability demonstrated differences in feedforward neuromuscular control during gait termination (Wikstrom et al., 2010b), and the presence of ankle instability is thought to affect supraspinal motor control (Hass et al., 2010). These differences in feedforward control may be caused in part by centrally mediated changes to motor control (Hass et al., 2010). If centrally mediated changes exist, that could explain differences in variability of motion at proximal joints and perhaps predisposition to repeated injury, particularly when assessing proximal kinematics. We imposed a state transition, when participants transitioned from an unloaded to a loaded state during landing. This required the acceptance of landing forces, potentially on tissue that had been compromised from previous ankle injury. Our participants with mechanical or functional instability appear to have a compromised ankle joint, as evidenced by decreased self-reported function. Previous research indicates proximal joints increase variability to avoid injury at distal joint (McLean et al., 2004). It is possible our injured groups were less likely to increase variability at proximal joints to avoid ankle injury in that distal joint. When numerous constraints

Table 3 Mixed model ANOVA CVln for group by jump direction for stance. Jump direction mean (standard deviation) 95% Confidence interval Variable

Group

Ankle plantardorsiflexion

MAI

Anterior Lower limit

4.12 (0.58) FAI 4.45 (0.55) Coper 4.36 (0.49) Control 4.19 (0.62)

2.39 (0.48) 2.16 (0.31) Coper 2.09 (0.37) Control 2.46 (0.61)

Ankle internal– external rotation

Knee flexion– extension

MAI

4.80 (0.91) FAI 5.17 (0.65) Coper 4.94 (0.56) Control 4.75 (0.81)

MAI

3.28 (0.35) FAI 3.12 (0.42) Coper 3.19 (0.39) Control 3.42 (0.78)

Knee abduction– MAI adduction

2.39 (0.50)

95% Confidence interval

95% Confidence interval

Upper limit

Lateral Lower limit

Upper limit

Medial Lower limit

Upper limit

Total

Lower limit

Upper limit

Comparison Difference P -value

3.85

4.39

3.56

4.42

4.65

4.33

MAI–FAI

− 0.14

0.220

3.72

4.42

4.11

4.71

4.15

4.47

MAI–coper

− 0.09

0.470

4.13

4.59

3.62

4.49

4.12

4.58

4.08

4.43

0.180

4.45

4.08

4.67

4.18

4.62

4.17

4.48

MAIcontrol FAI-coper

− 0.16

3.93

4.17 (0.73) 4.31 (0.70) 4.25 (0.67) 4.32 (0.62)

4.00

4.69

4.39 (0.58) 4.41 (0.69) 4.35 (0.49) 4.40 (0.52)

4.12

4.21

3.99 (0.94) 4.07 (0.81) 4.05 (0.93) 4.37 (0.70)

0.06

0.640

FAI-control

− 0.12

0.910

− 0.07

0.560

− 0.02

0.920

2.17

2.61

2.03

2.30

1.92

2.26

2.20

2.71

4.39

5.22

4.89

5.45

4.68

5.20

4.41

5.10

3.12

3.44

2.94

3.31

3.01

3.37

3.09

3.74

2.16

2.62

3.66 (1.68) 3.65 (1.50) 3.88 (1.62) 3.11 (1.06)

4.83 (0.75) 4.81 (0.81) 4.75 (0.94) 4.90 (0.86)

3.40 (0.38) 3.24 (0.37) 3.23 (0.42) 3.49 (0.77)

3.63 (0.61)

2.90

4.42

3.01

4.30

3.12

4.64

2.66

3.56

4.49

5.17

4.46

5.16

4.31

5.20

4.53

5.26

3.23

3.57

3.08

3.40

3.03

3.42

3.17

3.82

3.35

3.91

2.69 (0.60) 2.97 (0.89) 2.59 (0.55) 2.72 (0.61)

4.27 (0.85) 4.20 (0.58) 4.16 (0.56) 4.30 (0.58)

3.59 (0.60) 3.63 (0.57) 3.32 (0.64) 3.52 (0.81)

3.95 (0.82)

2.41

2.96

2.59

3.36

2.33

2.84

2.46

2.97

3.88

4.65

3.95

4.45

3.90

4.42

4.05

4.54

3.32

3.87

3.38

3.87

3.02

3.62

3.18

3.86

3.58

4.32

2.91 (1.18) 2.93 (1.18) 2.85 (1.25) 2.76 (0.82)

4.63 (0.87) 4.73 (0.79) 4.62 (0.78) 4.65 (0.79)

3.42 (0.47) 3.33 (0.50) 3.25 (0.49) 3.48 (0.77)

3.32 (0.93)

2.67

3.15

Coper– control MAI-FAI

2.70

3.16

MAI-coper

0.06

0.730

2.61

3.10

0.15

0.370

2.54

2.99

MAIcontrol FAI-coper

0.08

0.650

FAI-control

0.17

0.310

0.09

0.590

− 0.93

0.480

4.45

4.82

Coper– control MAI–FAI

4.55

4.91

MAI-coper

0.02

0.900

4.43

4.81

− 0.01

0.920

4.47

4.82

MAIcontrol FAI-coper

0.11

0.410

FAI-control

− 0.03

0.810

− 0.03

0.810

0.09

0.350

3.28

3.57

Coper– control MAI–FAI

3.19

3.47

MAI-coper

0.18

0.090

3.10

3.39

− 0.05

0.590

3.34

3.61

MAIcontrol FAI-coper

0.81

0.420

FAI-control

− 0.15

0.130

Coper– control MAI–FAI

− 0.23

0.020

− 0.10

0.510

3.10

3.54

95% CI for difference

Comparison Mean PDifference value

95% CI for Difference

− 0.38 to 0.09 − 0.33 to 0.15 − 0.38 to 0.07 − 0.18 to 0.29 − 0.24 to 0.21 − 0.30 to 0.16 − 0.35 to 0.32 − 0.29 to 0.41 − 0.18 to 0.48 − 0.26 to 0.42 − 0.15 to 0.49 − 0.24 to 0.43 − 0.35 to 0.17 − 0.25 to 0.28 − 0.27 to 0.24 − 0.15 to 0.37 − 0.17 to 0.33 − 0.29 to −@@.23 − 0.1 to 0.29 − 0.28 to 0.38 − 0.25 to 0.14 − 0.12 to 0.28 − 0.34 to 0.04 − 0.43 to − 0.03 − 0.41 to 0.20

C. Brown et al. / Clinical Biomechanics 27 (2012) 52–63

Ankle inversion– MAI eversion FAI

95% Confidence interval

59

(continued on next page)

60

Table 3 (continued) Jump direction mean (standard deviation) 95% Confidence interval Variable

Group

Anterior Lower limit

FAI

2.63 (0.54) Coper 2.26 (0.50) Control 2.93 (1.51)

Hip flexion

Hip abduction– adduction

Hip internal– external rotation

MAI

2.99 (0.74) FAI 2.99 (0.79) Coper 2.98 (0.62) Control 3.47 (1.33)

MAI

3.80 (0.36) FAI 3.59 (0.42) Coper 3.81 (0.74) Control 4.26 (1.16)

MAI

3.84 (0.58) FAI 3.84 (0.49) Coper 4.10 (0.77) Control 4.52 (1.19)

MAI FAI Coper

4.57 (0.90) 4.61 (0.82) 4.48 (0.62)

95% Confidence interval

95% Confidence interval

Upper limit

Lateral Lower limit

Upper limit

Medial Lower limit

Upper limit

Total

Lower limit

Upper limit

Comparison Difference P -value

95% CI for difference

2.40

2.87

3.40

4.12

4.19

3.63

MAI-Coper

0.23

0.140

3.20

3.82

3.11

3.86

2.86

3.31

0.270

3.57

3.24

4.36

3.34

4.14

3.29

3.70

MAIcontrol FAI-coper

− 0.17

2.29

3.42 (0.90) 3.09 (0.88) 3.49 (1.32)

3.21

2.50

3.88 (0.71) 3.49 (0.80) 3.74 (0.94)

3.57

2.03

3.76 (0.83) 3.51 (0.67) 3.80 (1.32)

0.34

0.030

− 0.08 to 0.55 − 0.47 to 0.13 0.03 to 0.65

FAI-control

− 0.07 − 0.40 0.120

− 0.36 0.23 0.010 − 0.71 − 0.01 0.490 − 0.23 0.48 0.690 − 0.29 0.44 0.030 − 0.75 − 0.05 0.780 − 0.41 0.31 *0.003 − 0.87 − 0.18 0.009 − 0.83 − 0.12 0.560 − 0.22 0.41 0.690 − 0.26 0.39 *0.003 − 0.79 − 0.17 0.870 − 0.35 0.29 *b0.001 − 0.88 − 0.27 *0.001 − 0.86 − 0.23 0.690 − 0.34 0.22 0.420 − 0.41 0.17 *0.003 − 0.70 − 0.14 0.670 − 0.35 0.22 0.009 − 0.63 − 0.09 0.040 − 0.58 − 0.02 0.910 − 0.31 0.35 0.670 − 0.27 0.41 0.400 − 0.46 0.18

2.65

3.33

2.65

3.34

2.69

3.27

2.91

4.03

3.64

3.96

3.41

3.77

3.47

4.16

3.77

4.75

3.58

4.11

3.63

4.05

3.74

4.46

4.02

5.02

4.16

4.98

4.25

4.96

4.19

4.77

2.26 (0.84) 2.24 (0.66) 2.20 (0.71) 2.71 (1.67)

4.26 (0.61) 4.01 (0.60) 4.16 (1.06) 4.82 (1.26)

4.93 (0.48) 5.06 (0.49) 4.91 (0.53) 5.39 (0.94)

2.95 (0.94) 3.06 (0.88) 3.02 (1.09)

1.88

2.64

1.95

2.52

1.87

2.53

2.00

3.41

3.98

4.53

3.75

4.27

3.66

4.65

4.29

5.36

4.71

5.15

4.85

5.27

4.67

5.16

4.99

5.78

2.53

3.38

2.68

3.44

2.50

3.53

2.56 (0.86) 2.21 (0.56) 2.41 (1.01) 2.84 (1.60)

4.17 (0.61) 4.34 (0.57) 4.05 (0.92) 4.58 (1.64)

4.77 (0.80) 4.82 (0.63) 4.89 (0.83) 4.90 (1.36)

2.96 (0.93) 2.76 (0.80) 2.77 (1.06)

2.17

2.96

1.97

2.45

1.94

2.88

2.17

3.51

3.89

4.44

4.09

4.59

3.62

4.49

3.89

5.27

4.41

5.14

4.55

5.09

4.50

5.28

4.32

5.47

2.54

3.39

2.41

3.11

2.28

3.27

2.60 (0.86) 2.48 (0.76) 2.53 (0.85) 3.01 (1.55)

4.07 (0.57) 3.98 (0.61) 4.01 (0.91) 4.56 (1.37)

4.52 (0.79) 4.57 (0.75) 4.63 (0.81) 4.94 (1.21)

3.49 (1.19) 3.47 (1.16) 3.42 (1.20)

2.35

2.86

Coper– control MAI–FAI

2.24

2.73

MAI-coper

0.070

2.27

2.79

− 0.400

2.77

3.25

MAIcontrol FAI-coper

− 0.050

FAI-control

− 0.530 − 0.480

3.85

4.30

Coper– control MAI–FAI

3.76

4.20

MAI-coper

0.070

3.77

4.24

− 0.048

4.34

4.77

MAIcontrol FAI-coper FAI-control

− 0.580 − 0.550

0.090

− 0.030

4.32

4.72

Coper– control MAI–FAI

4.38

4.77

MAI-coper

− 0.120

4.43

4.84

− 0.420

4.75

5.13

MAIControl FAI-coper

− 0.060

FAI-control

− 0.360 − 0.300

− 0.060

3.26

3.73

Copercontrol MAI-FAI

3.25

3.70

MAI-coper

0.070

3.18

3.66

MAIcontrol

− 0.140

0.020

0.670

Comparison Mean PDifference value

95% CI for Difference

to to to to to to to to to

FAI-Lat: Con-Lat

− 0.81

0.003 − 1.339 to − 0.280

MAI-Ant: Con-Ant FAI-Ant: Con-Ant

− 0.68

0.006 − 1.158 to − 0.196 0.005 − 1.150 to − 0.210

to to to to to to to to to to to to to to

−0.68

C. Brown et al. / Clinical Biomechanics 27 (2012) 52–63

Knee internal– external rotation

95% Confidence interval

Table 3 (continued) Jump direction mean (standard deviation) 95% Confidence interval Variable

Group

Anterior Lower limit

Control 4.60 (0.75)

Trunk flexion

Trunk lateral rotation

3.53 (0.60) FAI 3.36 (0.68) Coper 3.51 (0.73) Control 3.48 (0.78)

MAI

4.70 (0.73) FAI 4.87 (0.81) Coper 5.07 (0.74) Control 4.76 (0.76)

MAI

4.56 (0.84) FAI 4.52 (0.79) Coper 4.50 (0.63) Control 4.49 (0.76)

95% Confidence interval

95% Confidence interval

Upper limit

Lateral Lower limit

Upper limit

Medial Lower limit

Upper limit

Total

Lower limit

Upper limit

Comparison Difference P -value

95% CI for difference

4.92

3.43 (1.22)

3.94

2.87 (1.16)

3.36

3.63 (1.28)

3.41

3.85

FAI-coper

0.050

0.760

FAI-control

− 0.160

0.320

− 0.210

0.210

0.140

0.200

− 0.28 to 0.38 − 0.47 to 0.16 − 0.54 to 0.12 − 0.07 to 0.35 − 0.23 to 0.21 − 0.19 to 0.22 − 0.36 to 0.07 − 0.33 to 0.08 − 0.19 to 0.24 − 0.38 to 0.08 − 0.46 to 0.02 − 0.19 to 0.27 − 0.30 to 0.17 − 0.03 to 0.42 0.03 to 0.49

3.26

3.81

3.07

3.66

3.17

3.85

3.15

3.81

4.37

5.03

4.52

5.23

4.72

5.41

4.44

5.08

4.18

4.94

4.17

4.86

4.21

4.80

4.17

4.81

3.27 (0.45) 3.14 (0.42) 3.14 (.45) 3.32 (0.60)

5.12 (0.71) 5.30 (0.59) 5.34 (0.69) 5.10 (0.61)

3.41 (0.70) 3.60 (0.53) 3.70 (0.70) 3.75 (0.84)

2.92

3.06

3.47

2.96

3.32

2.92

3.35

3.07

3.57

4.80

5.44

5.04

5.55

5.02

5.66

4.84

5.36

3.09

3.72

3.37

3.83

3.37

4.03

3.40

4.10

3.24 (0.52) 3.12 (0.41) 3.42 (0.97) 3.19 (0.51)

5.08 (0.57) 5.18 (0.55) 5.15 (0.69) 4.91 (0.61)

3.39 (0.68) 3.34 (0.60) 3.44 (0.92) 3.22 (0.62)

2.38

3.00

3.48

2.95

3.30

2.97

3.87

2.98

3.41

4.82

5.34

4.94

5.42

4.83

5.47

4.65

5.17

3.08

3.70

3.09

3.60

3.00

3.87

2.96

3.48

3.35 (0.54) 3.21 (0.52) 3.36 (0.75) 3.33 (0.64)

4.97 (0.69) 5.12 (0.68) 5.19 (0.70) 4.92 (0.67)

3.78 (0.92) 3.82 (0.82) 3.88 (0.88) 3.82 (0.90)

3.20

3.50

Coper– control MAI–FAI

3.06

3.36

MAI-coper

− 0.008

0.940

3.20

3.51

0.020

0.880

3.19

3.47

MAIcontrol FAI-coper

− 0.150

0.180

FAI-control

− 0.120

0.240

0.020

0.820

− 0.150

0.200

4.80

5.14

Coper– control MAI–FAI

4.96

5.28

MAI-coper

− 0.220

0.080

5.01

5.36

0.040

0.700

4.77

5.08

MAIcontrol FAI-coper

− 0.670

0.570

FAI-control

0.200

0.090

0.260

0.030

− 0.040

0.780

3.61

3.96

Coper– control MAI–FAI

3.65

3.99

MAI-coper

− 0.100

0.460

3.70

4.07

− 0.040

0.780

3.65

3.99

MAIcontrol FAI-Coper

− 0.060

0.640

FAI-control

0.001

0.100

Coper– Control

0.060

0.630

− 0.29 0.21 − 0.35 0.16 − 0.28 0.21 − 0.31 0.19 − 0.24 0.24 − 0.19 0.31

to to to to

Comparison Mean PDifference value

95% CI for Difference

C. Brown et al. / Clinical Biomechanics 27 (2012) 52–63

Trunk lateral flexion

MAI

4.28

95% Confidence interval

to to

* Significant at pb 0.008. MAI: mechanical ankle instability; FAI: functional ankle instability.

61

62

C. Brown et al. / Clinical Biomechanics 27 (2012) 52–63

were required to complete the jump landing task, the injured groups may have been unable to propagate adequate solutions for success. With the inverted pendulum landing constraints, multiplanar shear forces had to be endured at the joints. We found no differences in ankle variability; this may indicate limited variability was available based on task constraints However, participants could modulate variability at other joints to accept these forces and control shear. While we did not directly measure feed-forward control, this concept may provide rationale for the current findings. When using an ankle strategy to balance in this inverted pendulum scenario, increasing variability at proximal joints may increase the demands on soft tissue surrounding the ankle to stabilize the joint. If the unstable ankle groups and copers had compromised soft tissue, which is likely, they may not have been as effective in stabilizing the ankle joint. Decreasing variability at the proximal joints could reduce stability demands on the injured ankle. Jump direction did not appear to have a considerable or systematic effect on variability in proximal joints. We found relatively few group differences within jump directions. While the direction of the jump was different, the overall motion and landing constraints were consistent. Different jump directions may have changed demands on joints in terms of moments, shear forces, and kinematics. Our lack of differences may be attributable to low power or the fact the jump direction did not change demands very much. We thought including multiple jump directions was relevant to capture a more realistic, encompassing task that reflected sports activity. The CV is a unitless measure, and represents relative or normalized variability, the variability converted to a percentage of the mean value (James, 2004). Lower variability is indicated by lower values of the CV. To our knowledge, there is no agreed upon acceptable range for the CV, particularly with regards to kinematic or kinetic variables. For use in diagnostic testing, CV values may be less than 20% or lower, depending on the precision of the test. However, for human movement, those same guidelines may not be applicable. Additionally, because our data were skewed, a natural log transformation was performed. Using similar methods in a previous study, the CVln values ranged from 2.08 to 3.56 for ankle, knee and hip kinematic variables during the stance phase of a 2-legged stop jump (Brown et al., 2009). Our data fit this range similarly, and exceed it in some cases. Taking stance in the anterior jump, the closest matching task, the current study's ankle, knee, and hip sagittal plane CVln are all greater than the previous study. The ankle and knee frontal plane CVln are smaller, while the hip transverse plane CVln is greater. The differences are not large, and may be attributable to the difference in nature of the task: double vs. single leg landing, transitioning to a vertical jump vs. stabilized landing, and the jump height performed. It is likely we reported greater variability in the single leg jump landing, because task complexity was greater as well as degrees of freedom. Overall, the ranges of reported data are similar, with some increased means in the current study. It is difficult to compare the current study's preinitial contact data, as the previous study did not report it (Brown et al., 2009). Effect sizes for differences between group means ranged from negligible to medium. When using the ln transformation, it becomes more difficult to determine if a statistically significant difference was behaviorally significant. That is, were true motion variability differences clinically relevant? The small to medium effect sizes may indicate that while statistically significant, the differences may not be clinically relevant. Observable behavior during landing may not be truly more or less variable. The limitations of this study include use of the CV variable that is sensitive to means close to zero, as well as the natural log transformation, which is not representative of the true spread of the data and is difficult to interpret clinically. Additionally, differences in sample sizes between genders and lack of participant matching across groups can be considered limitations. A clinical measure of ankle laxity, not an instrumented measure, was used, and participants thus had a range of laxity and dysfunction. The cross-sectional study design

did not allow for a causal link between movement variability and injury. In an attempt to accurately describe variability, we used a large number of variables and analyses. While we corrected for family-size error rate, there is a possibility that a Type II error was made and power was low on some variables. The anterior and lateral jump directions appear to elicit the most group differences, and future research should focus on other knee and hip motions in addition to the sagittal plane that may have relevance to episodes of giving way. Future studies should emphasize alternate measures of variability such as relative phase and curve analysis. 5. Conclusions Previous ankle injury appears to be a constraint on the movement system and those with ankle instability demonstrate decreased variability at the knee and hip. This may indicate central motor programming differences that decrease landing efficiency. Acknowledgements This study was funded by the University of Georgia Research Foundation. The sponsors had no involvement in the study design, collection, analysis and interpretation of data, in the writing of the manuscript; or in the decision to submit the manuscript for publication. References Beynnon, B.D., Renstrom, P.A., Alosa, D.M., Baumhauer, J.F., Vacek, P.M., 2001. Ankle ligament injury risk factors: a prospective study of college athletes. J. Orthop. Res. 19, 213–220. Brown, C.N., Ross, S.E., Mynark, R., Guskiewicz, K.M., 2004. Assessing functional ankle instability with joint position sense, time to stabilization, and electromyography. J. Sport Rehabil. 13, 122–134. Brown, C.N., Padua, D.A., Marshall, S.W., Guskiewicz, K.M., 2008. Individuals with mechanical ankle instability exhibit different motion patterns than those with functional ankle instability and ankle sprain copers. Clin. Biomech. 23, 822–831. Brown, C.N., Padua, D.A., Marshall, S.W., Guskiewicz, K.M., 2009. Variability of motion in individuals with mechanical or functional ankle instability during a stop jump maneuver. Clin. Biomech. 24, 762–768. Bullock-Saxton, J.E., 1994. Local sensation changes and altered hip muscle function following severe ankle sprain. Phys. Ther. 74, 17–31. Bullock-Saxton, J.E., Janda, V., Bullock, M.I., 1994. The influence of ankle sprain injury on muscle activation during hip extension. Int. J. Sports Med. 15, 330–334. Caulfield, B.M., Garrett, M., 2002. Functional instability of the ankle: differences in landing patterns of ankle and knee movement prior to and post landing in a single leg jump. Int. J. Sports Med. 23, 64–68. Chappell, J.D., Yu, B., Kirkendall, D.T., Garrett, W.E., 2002. A comparison of knee kinetics between male and female recreational athletes in stop-jump tasks. Am. J. Sports Med. 30, 261–267. Davids, K., Glazier, P., Araujo, D., Bartlett, R., 2003. Movement systems as dynamical systems: the functional role of variability and its implications for sports medicine. Sports Med. 33, 245–260. Delahunt, E., Monaghan, K., Caulfield, B., 2006. Altered neuromuscular control and ankle joint kinematics during walking in subjects with functional instability of the ankle joint. Am. J. Sports Med. 34, 1970–1976. Delahunt, E., Coughlan, G.F., Caulfield, B., Nightingale, E.J., Lin, C.W., Hiller, C.E., 2010. Inclusion criteria when investigating insufficiencies in chronic ankle instability. Med. Sci. Sports Exerc. 42, 2106–2121. Drewes, L.K., McKeon, P.O., Paolini, G., Riley, P., Kerrigan, D.C., Ingersoll, C.D., et al., 2009. Altered ankle kinematics and shank-rear-foot coupling in those with chronic ankle instability. J. Sport Rehabil. 18, 375–388. Fong, D.T., Hong, Y., Chan, L.K., Yung, P.S., Chan, K.M., 2007. A systematic review on ankle injury and ankle sprain in sports. Sports Med. 37, 73–94. Gribble, P., Robinson, R., 2009. Differences in spatiotemporal landing variables during a dynamic stability task in subjects with CAI. Scand. J. Med. Sci. Sports 20, e63–e71. Hass, C.J., Bishop, M.D., Doidge, D., Wikstrom, E.A., 2010. Chronic ankle instability alters central organization of movement. Am. J. Sports Med. 38, 829–834. Hausdorff, J.M., Rios, D.A., Edelberg, H.K., 2001. Gait variability and fall risk in community-living older adults: a 1-year prospective study. Arch. Phys. Med. Rehabil. 82, 1050–1056. Heiderscheit, B.C., Hamill, J., van Emmerik, E.A., 2002. Variability of stride characteristics and joint coordination among individuals with unilateral patellofemoral pain. J. Appl. Biomech. 18, 110–121. Hertel, J., 2002. Functional anatomy, pathomechanics, and pathophysiology of lateral ankle instability. J. Athl. Train. 37, 364–375. Hertel, J., Kaminski, T.W., 2005. Second international ankle symposium summary statement. J. Orthop. Sports Phys. Ther. 35, A2–A6.

C. Brown et al. / Clinical Biomechanics 27 (2012) 52–63 Hiller, C.E., Refshauge, K.M., Bundy, A.C., Herbert, R.D., Kilbreath, S.L., 2006. The cumberland ankle instability tool: a report of validity and reliability testing. Arch. Phys. Med. Rehabil. 87, 1235–1241. Hoffman, M., Schrader, J., Applegate, T., Koceja, D., 1998. Unilateral control of the functionally dominant and nondominant extremities of healthy subjects. J. Athl. Train. 33, 319–322. Hootman, J.M., Dick, R., Agel, J., 2007. Epidemiology of collegiate injuries for 15 sports: summary and recommendations for injury prevention initiatives. J. Athl. Train. 42, 311–319. Hoppenfeld, S., 1976. Physical Examination of the Spine and the Extremities. Prentice Hall, London. Hubbard, T.J., Kramer, L.C., Denegar, C.R., Hertel, J., 2007. Contributing factors to chronic ankle instability. Foot Ankle Int. 28, 343–354. James, C.R., 2004. Considerations of movement variability in biomechanics research. In: Stergiou, N. (Ed.), Innovative Analyses of Human Movement. Human Kinetics, Champaign, IL, pp. 29–62. James, C.R., Dufek, J.S., Bates, B.T., 2000. Effects of injury proneness and task difficulty on joint kinetic variability. Med. Sci. Sports Exerc. 32, 1833–1844. Konradsen, L., 2002. Sensori-motor control of the uninjured and injured human ankle. J. Electromyogr. Kinesiol. 12, 199–203. Konradsen, L., Voigt, M., 2002. Inversion injury biomechanics in functional ankle instability: a cadaver study of simulated gait. Scand. J. Med. Sci. Sports 12, 329–336. McKeon, P.O., Hertel, J., 2008. Spatiotemporal postural control deficits are present in those with chronic ankle instability. BMC Musculoskelet. Disord. 9, 76. McLean, S.G., Neal, R.J., Myers, P.T., Walters, M.R., 1999. Knee joint kinematics during the sidestep cutting maneuver: potential for injury in women. Med. Sci. Sports Exerc. 31, 959–968. McLean, S.G., Lipfert, S.W., van den Bogert, A.J., 2004. Effect of gender and defensive opponent on the biomechanics of sidestep cutting. Med. Sci. Sports Exerc. 36, 1008–1016.

63

Norkin, C.C., White, D.J. (Eds.), 1995. Measurement of Joint Motion: A Guide to Goniometry. F.A. Davis, Philadelphia. Ross, S.E., Guskiewicz, K.M., 2004. Examination of static and dynamic postural stability in individuals with functionally stable and unstable ankles. Clin. J. Sports Med. 14, 332–338. Ross, S.E., Guskiewicz, K.M., 2006. Effect of coordination training with and without stochastic resonance stimulation on dynamic postural stability of subjects with functional ankle instability and subjects with stable ankles. Clin. J. Sports Med. 16, 323–328. Tropp, H., Odenrick, P., Gillquist, J., 1985. Stabilometry recordings in functional and mechanical instability of the ankle joint. Int. J. Sports Med. 29, 180–182. Valderrabano, V., Hintermann, B., Horisberger, M., Fung, T.S., 2006. Ligamentous posttraumatic ankle osteoarthritis. Am. J. Sports Med. 34, 612–620. Wikstrom, E.A., Tillman, M.D., Chmielewski, T.L., Cauraugh, J.H., Borsa, P.A., 2007. Dynamic postural stability deficits in subjects with self-reported ankle instability. Med. Sci. Sports Exerc. 39, 397–402. Wikstrom, E.A., Tillman, M.D., Chmielewski, T.L., Cauraugh, J.H., Naugle, K.E., Borsa, P.A., 2010a. Dynamic postural control but not mechanical stability differs among those with and without chronic ankle instability. Scand. J. Med. Sci. Sports. 20 (1), e137–144. Wikstrom, E.A., Bishop, M.D., Inamdar, A.D., Hass, C.J., 2010b. Gait termination control strategies are altered in chronic ankle instability subjects. Med. Sci. Sports Exerc. 42 (1), 197–205. Wu, G., Siegler, S., Allard, P., Kirtley, C., Leardini, A., Rosenbaum, D., et al., 2002. ISB recommendation on definitions of joint coordinate system of various joints for the reporting of human motion — part I: ankle, hip, and spine. J. Biomech. 35, 543–548. Yeung, M.S., Chan, K.M., So, C.H., Yuan, W.Y., 1994. An epidemiological survey on ankle sprain. Br. J. Sports Med. 28, 112–116. Yu, B., Gabriel, D., Noble, L., An, K.N., 1999. Estimate of optimum cutoff frequency for a low-pass digital filter. J. Appl. Biomech. 15, 318–329.