ORIGINAL RESEARCH

Gender and Limb Differences in Dynamic Postural Stability During Landing

Erik A. Wikstrom, MS, Mark D. Tillman, PhD, Kai J. Kline, BS, and Paul A. Borsa, PhD

Objective: To determine if gender and limb dominance aect dynamic postural stability and vertical ground reaction force data during jump landings. Secondary objective was to assess the reliability of the dynamic postural stability index (DPSI). Design: A mixed model (2 gender 2 limb) repeated measures design was used to determine the eects of gender and limb dominance on dynamic postural stability. Subjects were required to perform a two-legged jump to a height equivalent of 50% of their maximum vertical leap, land on a single-leg and balance for three seconds. Setting: Sports Medicine Research Laboratory. Participants: Forty healthy subjects (20 men, 20 women) participated in this investigation. Main Outcome Measures: The DPSI and its directional components quantied dynamic postural stability during a single-leg jump landing. Normalized vertical ground reaction force data quantied energy absorption. Results: DPSI values revealed that females had signicantly dierent dynamic postural stability as compared to males in the vertical plane [T (78) = 4.2, P <0.01], and in the composite score (dynamic postural stability index) [T (78) = 6.3, P <0.01]. In addition, females had signicantly higher peak vertical ground reaction forces [T (78) = 13, P = 0.01] than males. The DPSI also showed excellent reliability (ICC = 0.96), with a 95% condence interval ranging from 0.94 to 0.97. Conclusions: The results indicate that females have higher dynamic postural stability scores in the vertical direction as well as the composite score. This suggests that females used dierent dynamic postural stability strategies than males. There were no side-to-side dynamic postural stability dierences between healthy contralateral limbs. Key Words: jump landings, ground reaction forces, lower extremity (Clin J Sport Med 2006;16:311315)

Received for publication April 19, 2005; accepted March 10, 2006. From the Center for Exercise Science, University of Florida, Gainesville, FL. Reprints: Erik A. Wikstrom, University of Florida, Department of Applied Physiology and Kinesiology, PO Box 118205, Gainesville, FL 32611-8205; Phone: 352-392-0584, x1402 Fax: 352-392-5262 (e-mail: ewikstrom@hhp.u.edu) Copyright r 2006 by Lippincott Williams & Wilkins

ower extremity injuries are extremely common in todays athletic population.1 Female athletes have been reported to have a four to six times higher incidence of ACL injuries than males while participating in the same sporting activities2,3 and that female athletes may have a higher incidence of lateral ankle sprains than their male counterparts.4 Previous examinations of lower extremity gender dierences have shown that the participating females were ligament dominant (absence of muscle control of medio-lateral knee motion that results in high valgus knee torques and high ground reaction forces),5 have neuromuscular imbalances between dominant and non-dominant limbs,5 and use dierent landing strategies than males6 by examining neuromuscular control3,5 and joint kinematics.6 However, only one investigation has examined dynamic postural stability.7 Previous research also suggests that side-to-side dierences in static postural sway between contralateral limbs can increase the risk of injury,8 and that limb dominance may be a causal factor for these dierences.9 However, little research has examined limb dominance for dierences in more functional conditions such as a single-leg hopping task.8,9 Currently there are only two measures of dynamic postural stability; time to stabilization (TTS)10 and the dynamic postural stability index (DPSI).11 Both measures determine how well balance can be maintained while transitioning from a dynamic to static state by assessing the subjects ability to control ground reaction forces and maintain their center of gravity within their base of support. These are also functional measurements of neuromuscular control, because they are calculated during a single-leg-hop-stabilization maneuver. Measures of dynamic postural control are more informative than static measures of postural control because jump landings are commonly reported as a mechanism for lower extremity injury.12 However, TTS has several aws and only allows single plane examinations of dynamic postural stability (ie, vertical, anterior-posterior, mediallateral). The DPSI was designed to correct for these aws and it allows a multi-plane examination of dynamic postural stability. Therefore, the purpose of this investigation was to determine if gender and limb dominance aect dynamic postural stability as measured by the DPSI, peak vertical ground reaction force (vGRF) and time to peak vGRF (TPvGRF) during jump landings. Our secondary objective was to assess the reliability of the DPSI.

Clin J Sport Med

Volume 16, Number 4, July 2006

311

Wikstrom et al

Clin J Sport Med

Volume 16, Number 4, July 2006

METHODS Subjects and Design

A mixed model (2 gender 2 limb) repeated measures design was used to determine if gender and limb dominance aect dynamic postural stability. Forty subjects participated in this investigation (Table 1). During the rst test session, subjects read and signed the University-approved informed consent, and completed a medical history questionnaire to determine eligibility. All subjects were free from acute lower extremity and head injuries for the previous three months and did not suer from any equilibrium disorders. Additionally, all subjects were free from any chronic ankle or knee injuries.

and the jump stabilization maneuver used in this investigation are below 30 Hz (based on Fourier analysis). Therefore 200 Hz was selected so that the sampling frequency would be at least six times greater than the raw analog signal. The force plate data underwent an analog to digital conversion and was stored on a PC-type computer using the DATAPAC 2000 (Run Technologies, Laguna Hills, CA) analog data acquisition, processing, and analysis system.

Data Reduction Dynamic Postural Stability Index

Data was collected and exported from the Datapac software as comma delimited (.CSV) les. All data was reduced using a QuickBasic subroutine (version 4.5, Microsoft Corporation, Redmond, WA) which calculated the DPSI and its directional components (SIx, SIy, SIz). The SIx and SIy directional components assess the uctuations from zero along the X- (medial-lateral) and Y- (anterior-posterior) axes of the force plate, respectively. SIz assesses the uctuation from the subjects body weight (as a zero point) along the Z- (vertical) axis of the force plate. The DPSI is a composite of the SIx, SIy, and SIz axes and thus is sensitive to changes in postural stability in all three directions (Table 2). It is important to note that the DPSI is done separately from the directional components, using each array. The average values from the three successful trials on the rst day of testing were then used for analysis.

Data Collection Procedures

Subjects reported to a research laboratory for all test sessions. Each subject had both legs tested during three sessions all within a seven-day period and at least 24 hours between test sessions. The order of limb testing was random and counter balanced between sessions. The dominant limb was dened as the limb that the subject used to kick a soccer ball. Subjects had their maximum vertical jump (Vertmax) tested as described by Wikstrom et al.13 Baseline measures of body weight during a static stance were then recorded on a force plate during a 5 s window.10 Subjects then completed three successful jump protocol trials. Subjects started in a standing position 70 cm from the center of the force plate. Each subject was required to jump with both legs and touch an overhead marker placed at a position equivalent to 50% of the subjects maximum vertical leap before landing on the force plate.10,11,13 Each subject was to land on the test leg, stabilize as quickly as possible and balance for three seconds with their hands on their hips looking straight ahead.11 All subjects were instructed to jump with their head up and hands in a position to touch the designated marker. If a subject lost balance and touched the oor with the contra-lateral limb the trial was discarded and repeated. Likewise, if a short additional hop occurred upon landing the trial was discarded and repeated. Subjects then returned on two additional days and completed identical testing procedures. A Bertec tri-axial force plate (Bertec Corporation, Columbus, OH) was used to collect the baseline and jump landing ground reaction force data at 200 Hz.11 According to Ross and Guskiewicz,10 the raw analog signals of a single-leg stance

Ground Reaction Forces

The vGRFs, measured in Newtons were expressed as the magnitude of the peak force divided by the subjects body weight (BW), measured in Newtons, or multiples of BW, and the average of three successful trials was used for further analysis. The TPvGRF was determined as the time (in ms) that it took for the subject to reach peak vGRF from the time of impact. The average values from the three successful trials were used in subsequent analysis.

Normalization
Based on previous investigations, it was expected that males would have greater mass and jump heights than females. In the current study, males did possess an additional 21 kg of mass and jumped 9.5 cm higher than females; therefore we proceeded by normalizing the dynamic stability index measurements and vGRFs by the energy dissipated during the landings performed by each individual. More specically, three variables were
TABLE 2. Formulas Used to Calculate the Dynamic Postural Stability Index and the Directional Components (SIx, SIy, SIz)
SIx = O [S(0Fx)2/number of samples] SIy = O [S(0Fy)2/number of samples] SIz = O [S(body weightFz)2/number of samples] DPSI = O [S(0Fx)2+S(0Fy)2+S(body weightFz)2/number of samples]

TABLE 1. Group Demographics

Measures N Age (years) Height (cm)* Weight (kg)* Jump height (cm)* Males 20 21.2 2.4 177.2 6.8 79.7 11.2 24.7 4.1 Females 20 21.1 1.4 163.1 9.4 59.1 5.7 15.2 2.5 T38 0.24 5.41 7.32 8.90 P 0.81 <0.01 <0.01 <0.01

*Indicates signicant dierence between males and females.

312

2006 Lippincott Williams & Wilkins

Clin J Sport Med

Volume 16, Number 4, July 2006

Gender and Limb Differences in Dynamic Postural Stability

TABLE 3. Results of Paired T-tests

Dependent Variables Dominant Non-Dominant 0.24 0.10 0.31 0.07 0.66 0.16 0.77 0.17 4.4 .57 33.38 7.12 T39 0.33 0.18 0.13 0.00 0.98 0.46 P 0.75 0.86 0.90 1.00 0.33 0.65 SIx 0.24 0.11 SIy 0.31 0.07 SIz 0.66 0.16 DPSI 0.77 0.16 vGRF 4.3 .58 Time to peak vGRF (ms) 32.75 6.77

RESULTS Energy and Normalized Values

Group means, standard deviations, t-scores and P-values for energy and normalized value comparisons appear in Table 4. Since our preliminary analysis found no signicant dierences between limbs, the dominant and non-dominant leg data was averaged for subsequent analysis. The study found signicant dierences in the vertical potential energy results. Males had signicantly greater potential energy than females. However, females had signicantly greater horizontal kinetic energy than males. The normalized SIy (SIy/KE) was signicantly dierent (increased) in males as compared to females. However, the normalized SIz (SIz/PE) and modied DPSI also indicated signicantly dierent (increased scores) dynamic postural stability between males and females. Female subjects landed with signicantly greater normalized vGRF than their male counterparts, while male subjects reached TPvGRF faster than females.

normalized to relevant energy values (horizontal kinetic energy (KE) or vertical potential energy (PE)). The SIy is a measure of variance of the GRF values in the anterior-posterior direction and is sensitive to the horizontal velocity and mass of the jumper. Therefore, SIy was normalized to the horizontal kinetic energy (0.5 mass horizontal velocity).2 Alternatively, the SIz was normalized to the potential energy (mass gravity jump height) dissipated during the landing because the SIz is a measure of variance in the vertical GRF. Similarly, vGRF were normalized to the vertical PE.

Reliability
A two-factor ICC revealed poor reliability for the SIx (ICC = 0.27) when measured alone. Reliability was good (ICC = 0.86) for SIy, while SIz had excellent reliability (ICC = 0.97). The DPSI also showed excellent reliability (ICC = 0.96), with a 95% condence interval ranging from 0.94 to 0.97. ICC was based on group means and standard deviations over three testing sessions (Day 1, 0.77 0.16, Day 2, 0.75 0.15, Day 3, 0.74 0.14).

Statistical Analysis
Subject demographics were compared using independent sample t-tests. Dependent variables (SIx, SIy, SIz, DPSI, vGRF, TPvGRF) were initially compared using a paired sample t-test with a Bonferroni adjustment to rule out leg dominance eects. Group means, standard deviations, t- and P-values for the leg dominance eects can be seen in Table 3. From our preliminary analysis we found no signicant leg dominance dierences, therefore we averaged the data across limbs for the energy and normalized variables (PE, KE, SIy/KE, SIz/PE, modied DPSI and vGRF/PE) and examined gender dierences using separate independent sample t-tests. Data was averaged and not collapsed so that the degrees of freedom were not articially inated and to maintain the assumption of independent observations. Test-re-test reliability for the SIx, SIy, SIz and DPSI measures were calculated using an intra-class correlation coecient (ICC(3,1)) formula prior to normalization.14 All reliability coecients were interpreted as follows; below 0.69 was poor, 0.70 to 0.79 was fair, 0.80 to 0.89 was good, and 0.90 to 1.00 was considered excellent.15 An alpha level of 0.05 was used for all statistical tests.

DISCUSSION Gender Differences Dynamic Postural Stability

In this investigation, the data was normalized to the vertical PE and horizontal KE of the individual subjects, which account for mass, jump height, and horizontal velocity. It was discovered that males had almost twice as much potential energy (172 J) to dissipate as females (109.4 J), but that males had less horizontal KE (103.1 J) than females (115.9 J). These dierences indicate that during a sagittal plane jump protocol; females need to dissipate more KE than males because they were jumping with additional horizontal (anterior) velocity. The increased requirements for dissipating more kinetic

TABLE 4. Results of Independent T-tests

Dependent Variables Potential Energy (J)* Kinetic Energy (J)* SIy/KE* SIz/PE* Normalized DPSI* Normalized vGRF* Males 172.16 44.33 103.13 12.00 0.003 0.001 0.004 0.001 0.005 0.001 0.029 0.009 Females 109.35 29.00 115.87 14.98 0.002 0.001 0.006 0.001 0.006 0.001 0.044 0.009 T38 5.30 2.96 3.03 4.15 2.78 5.08 P <0.01 <0.01 <0.01 <0.01 <0.01 <0.01

*Indicates signicant dierence between males and females.

2006 Lippincott Williams & Wilkins

313

Wikstrom et al

Clin J Sport Med

Volume 16, Number 4, July 2006

energy (despite having lower mass) could increase the risk of ACL injury for females. Despite the higher vertical PE dissipation constraints of males, females had higher normalized dynamic postural stability index scores as exhibited by a higher normalized SIz, and DPSI. Similarly, we found that male subjects had higher SIy scores when normalized to the horizontal KE. Only one previous investigation has examined dynamic postural control and found no dierences between healthy males and females using TTS.7 While it is natural to state that one group had decreased dynamic postural stability, it must be noted that both males and female were able to successfully complete the jump landing protocol. Therefore, we must conclude that males and females used dierent landing strategies to stabilize and absorb energy after a jump landing. This explanation of group dierences is consistent with previous literature as Decker et al6 revealed that females landed with greater knee extension and ankle plantar exion than males, but exhibited greater ROM at the knee and ankle during the landing sequence. Another plausible explanation may be the required subject eort exerted. For both drop jumps and the hop stabilization test used in previous investigations, males may have an anthropometric advantage. During drop jumps, the height of the drop is almost always less (as percentage of height or leg length) for males. During this single-leg hop stabilization test the jump height is normalized to the individual subjects; however the jump length is standardized (70 cm) because of the increased height/leg length of males may make the protocol inherently easier. Wikstrom et al13 compared TTS during a drop jump and single-leg hop stabilization protocol. Specically, subjects indicated that the drop jump protocol was easier despite the longer TTS scores. The authors concluded that the subjects put forth less than maximal eort, but enough to complete the protocol. This conclusion was based on the work of McKinely et al16 who noted that the subjects with the shortest TTS had all three major muscles of the lower leg (gastrocnemius, soleus, anterior tibialis) contracted prior to landing, thus allowing faster reaction to the landing surface. Subjects with poor TTS showed little to no anticipatory contraction of the lower leg musculature as measured by electromyography, thus indicating an association between anticipatory contraction and improved dynamic postural stability. Therefore, a perception of ease by male subjects might have triggered a subconscious motor command to produce less or no anticipatory contraction and still feel condent in completing the task. In the current investigation we cannot state whether one group had signicantly worse dynamic postural stability as compared to another. This is because recent research in non-linear dynamics (a dynamic system that has a sensitive dependence on its initial conditions) also known as the chaos theory has challenged the traditional perspectives that associate high variability with decreased performance and pathology.17 The argument

stems from the capability of biological organisms to produce a variety of solutions to a particular task (ie, dierent landing strategies). This capability oers exibility to deal with unexpected constraints and is a major source of variability in movement patterns.18 This variability is suggested as being the cornerstone of healthy and adaptable physiological systems.19 Other investigators, however, suggest that both an increase or decrease in variability may be the result of disease depending on the specic dynamics being investigated.20 The authors of a review concerning the functional aspects of variability in postural control suggest the need for future investigations, yet strongly suggest that higher variability in postural control is a sign of a more stable individual.17

Ground Reaction Forces

Similar to our dynamic postural stability ndings, our ground reaction force data suggest that males and females use dierent landing strategies. Males reached TPvGRF faster than females, but females had decreased energy absorption as exhibited by higher vGRF. However, Lephart et al7 indicated that there were no gender dierences in vGRF during a hop and step-down test, but did indicate that females did land with less knee exion than their male counterparts. Similar to the current investigation, mass was greater in males than females, but the step-down test had a standardized jump height (20 cm) and the forward hop tests did not have a minimum jump height requirement. During the current investigation, females jumped to 15.2 cm, and while smaller than the height of the stepdown test used by Lephart et al,7 a previous investigation has shown that forward hop tests are perceived to be more dicult and create higher vertical ground reaction force (4.03 BW>2.06 BW) than a step-down test.13 Therefore, females enrolled by Lephart et al7 may not have been stressed to the same extent as females in the current investigation, possibly explaining the dierences in results. Previous investigations have reported higher vGRF in females as compared to males, indicating neuromuscular control decits possibly due to ligament dominance when compared to males.5 Ligament dominance implies that ground reaction forces dictate the motion of the lower extremity, especially the knee during landing. While speculative, as no kinematic data were collected, it is reasonable to assume that the females in this investigation expressed ligament dominance because the force data are similar to previous investigations.5

Limb Dominance
In previous investigations, limb imbalances have been measured using strength tests and single-limb balance tests. These studies have shown that poor single-limb postural stability leads to higher injury incidence.21 However, these deciencies are not consistent within the literature2123 and little research has been performed to quantify bilateral limb discrepancies in more functional clinical tests such as a single-leg hopping task.9,24 Similar to the results of our investigation, Ross
r

314

2006 Lippincott Williams & Wilkins

Clin J Sport Med

Volume 16, Number 4, July 2006

Gender and Limb Differences in Dynamic Postural Stability

et al9 and Colby et al24 reported no bilateral decits for TTS and vGRF, and TTS respectively. Ross et al9 did indicate dierences in kinematic measurements and TPvGRF, which is contrary to our results as we found no dierences in TPvGRF. They suggested that this was due to altered strategy between limbs. While we have no way of disputing their claims, our data indicates that while healthy contralateral limbs might use dierent strategies, they control dynamic postural stability equally well.

Dynamic Postural Stability Index

The reliability of the DPSI was shown to be excellent between test sessions (0.96).15 Similarly, the reliability of the DPSI is at least equivalent to those described in previous research. Colby et al24 reported reliability coecients for dominant and non-dominant limb TTS scores from a hop stabilization test were similar to the one used in this investigation as ranging from (0.872 to 0.971). Additionally, reliability reports of TTS range from (0.61 to 0.80) while using an identical singleleg jump stabilization maneuver to the one used in this investigation.25

Clinical Significance
Dierences in dynamic postural stability were noted between genders, yet it is unclear if these dierences are the result of dierent postural control strategies or impairments in neuromuscular control. Because we used all healthy subjects, it is likely that these dierences were due to dierent postural control strategies and it is possible that the strategy used by female subjects gives rise to an increased probability for injury. One reason for the potential increased probability of injury is the fact that females do not absorb vGRF as well as males. Clinicians should focus on instructing proper landing technique and implementing neuromuscular control training programs in their female populations. These techniques have been shown to reduce vGRF and minimize neuromuscular imbalances, noted in previous research.5 Specically, they have improved neuromuscular control at the knee to minimize valgus forces, improved hamstring to quadriceps strength ratios, and minimize limb dominance strength and coordination dierences in female subjects. In addition, clinicians and researchers should be aware of the potential advantage that males have during jump tests and account for the dierences in KE and PE. In addition, lower extremity limb dominance does not appear to be a factor that causes dynamic postural stability dierences. Therefore, clinicians should use the DPSI as a benchmark for return to play decisions after unilateral lower extremity injury because the measure is objective and reliable. REFERENCES
1. Frank C, Jackson D. The science of reconstruction of the anterior cruciate ligament. J Bone Joint Surg Am. 1999;79:15561576.

2. Arendt EA, Agel J, Dick R. Anterior cruciate ligament injury patterns among collegiate men and women. J Athl Train. 1999; 34:8692. 3. Hewett T, Stroupe A, Nance T, et al. Plyometric training in female athletes: decreased impact forces and increased hamstring torques. Am J Sport Med. 1996;24:765769. 4. Hosea T, Carey C, Harrer M. The gender issue: epidemiology of ankle injuries in athletes who participate in basketball. Clin Orthop. 2000;372:4549. 5. Hewett T, Paterno M, Myers G. Strategies for enhancing proprioception and neuromuscular control of the knee. Clin Ortho Rel Res. 2002;402:7694. 6. Decker M, Torry M, Wyland D, et al. Gender dierences in lower extremity kinematics, kinetics and energy absorption during landing. Clin Biomech. 2003;18:662669. 7. Lephart S, Ferris C, Riemann B, et al. Gender dierences in strength and lower extremity kinematics during landing. Clin Orthop Rel Res. 2002;401:162169. 8. Soderman K, Alfredson H, Pietila T, et al. Risk factors for leg injuries in female soccer players: a prospective investigation during one out-door season. Sport Med. 2001;9:313321. 9. Ross S, Guskiewicz K, Prentice P, et al. Comparison of biomechanical factors between the kicking and stance limbs. J Sport Rehabil. 2004;13:135150. 10. Ross S, Guskiewicz K. Time to stabilization: a method for analyzing dynamic postural stability. Athl Ther Today. 2003;8:3739. 11. Wikstrom E, Tillman M, Smith A, et al. New force plate technology measure of dynamic postural stability: the dynamic postural stability index. J Athl Train. 2005;40:305309. 12. McKay G, Goldie P, Payne W, et al. Ankle injuries in basketball: injury rate and risk factors. Br J Sports Med. 2001;35:10335;108. 13. Wikstrom E, Tillman M, Borsa P. Detection of dynamic stability decits in subjects with functional ankle instability. Med Sci Sports Exerc. 2005;37:169175. 14. Portney L, Watkins M. Foundations of Clinical Research: Applications to Practice. Norwalk, CT, Appleton & Lange; 1993:509516. 15. Denegar CR, Ball DW. Assessing reliability and precision of measurement: An introduction to intraclass correlation and standard error of measurement. J Sport Rehabil. 1993;2:3542. 16. McKinely P, Pedotti A. Motor strategies in landing from a jump: the role of skill in task execution. Exp Brain Res. 1992;90:427441. 17. van Emmerik R, van Wegen E. On the functional aspects of variability in postural control. Exerc Sport Sci Rev. 2002;30: 177183. 18. Latash M, Scholz J, Schoner G. Motor control strategies revealed in the structure of motor variability. Exerc Sport Sci Rev. 2002;30: 2631. 19. Lipsitz L. Dynamics of stability: the physiologic basis of functional health and fraility. J Gerontol A Biol Sci Med Sci. 2002;57: B115B125. 20. Vaillancourt D, Newell K. Changing complexity in human behavior and physiology through aging and disease. Neurobiol Aging. 2002; 23:111. 21. Tropp H, Ekstrand J, Gillquist J. Stabilometry in functional instability of the ankle and its value in predicting injury. Med Sci Sports Exerc. 1984;16:6466. 22. Homan M, Schrader J, Applegate T, et al. Unilateral postural control of the functionally dominant and nondominant extremities of healthy subjects. J Athl Train. 1998;33:319322. 23. Tanaka T, Hashimoto N, Nakata M, et al. Analysis of toe pressures under the foot while dynamic standing on one foot in healthy subjects. J Orthop Sports Phys Ther. 1996;23:188193. 24. Colby S, Hintermeister R, Torry M, et al. Lower limb stability with ACL impairment. J Orthop Sports Phys Ther. 1999;29: 444451. 25. Ross S, Guskiewicz K. Examination of static and dynamic postural stability in individuals with functionally stable and unstable ankles. Clin J Sport Med. 2004;14:332338.

2006 Lippincott Williams & Wilkins

315