- Email: [email protected]
The Associations Between the Dominant and Nondominant Peak External Knee Adductor Moments During Gait in Healthy Subjects: Evidence for Symmetry
320
ORIGINAL ARTICLE
The Associations Between the Dominant and Nondominant Peak External Knee Adductor Moments During Gait in Healthy Subjects: Evidence for Symmetry Andrew J. Teichtahl, MBBS, B.Physio, Anita E. Wluka, MBBS, FRACP, PhD, Meg E. Morris, B.App Sci. (Physio), FACP, PhD, Susan R. Davis, MBBS, FRACP, PhD, Flavia M. Cicuttini, MBBS, FRACP, PhD ABSTRACT. Teichtahl AJ, Wluka AE, Morris ME, Davis SR, Cicuttini FM. The associations between the dominant and nondominant peak external knee adductor moments during gait in healthy subjects: evidence for symmetry. Arch Phys Med Rehabil 2009;90:320-4. Objectives: There is growing interest in the role of the knee adduction moment in the pathogenesis of knee pain and osteoarthritis. It is unclear whether the knee adduction moment is similar between the dominant and nondominant legs during locomotion. This study examined whether asymmetry exists in the peak knee adductor moments during gait in healthy adults. Design: Cross-sectional study. Setting: Musculoskeletal Research Centre, La Trobe University, Melbourne, Victoria, Australia. Participants: Three-dimensional Vicon gait analyses were performed for 17 healthy men and women. Interventions: Not applicable. Main Outcome Measures: The external dominant and nondominant peak knee adduction moments during early and late stance were analyzed to determine whether any significant differences occurred between limbs. Results: Peak knee adductor moments for dominant and nondominant limbs were significantly correlated during early (R⫽0.61, P⫽0.009) and late (R⫽0.72, P⫽0.001) stance. After adjustment for age and sex, there was an associated 0.58 (P⫽0.030) and 0.98 (P⫽0.009) unit increase in the peak knee adduction moment in the nondominant leg, for every 1 unit increase in the dominant leg during the early and late stance phases of gait, respectively. Further evidence for symmetry was provided by the symmetry index, which was 0.04% and 0.62% for early and late stance, respectively. Conclusions: In healthy subjects, the magnitude of the dominant limb peak external knee adduction moments during stance, and in particular late stance, appears representative of the magnitude of the moment in the nondominant limb. These findings imply symmetry between these moments and may have important implications when collecting data for limb analyses among healthy subjects. Whether gait symmetry protects against
From the Department of Epidemiology and Preventive Medicine (Teichtahl, Wluka, Cicuttini), and the National Health and Medical Research Council of Australia Centre of Clinical Research Excellence for the Study of Women’s Health (Davis), Monash University Medical School, Alfred Hospital; Baker Heart Research Institute (Wluka); and School of Physiotherapy, University of Melbourne (Morris), Melbourne, Australia. No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit on the authors or on any organization with which the authors are associated. Reprint requests to Flavia Cicuttini, MBBS, FRACP, PhD, Dept of Epidemiology and Preventive Medicine, Monash University, Alfred Hospital, Melbourne, Victoria 3004, Australia, e-mail: [email protected]. 0003-9993/09/9002-00136$36.00/0 doi:10.1016/j.apmr.2008.07.030
Arch Phys Med Rehabil Vol 90, February 2009
the onset of unilateral (or increases the risk for bilateral) pathological joint changes will need to be confirmed longitudinally. Key Words: Biomechanics; Gait; Knee; Rehabilitation. © 2009 by the American Congress of Rehabilitation Medicine PIDEMIOLOGIC STUDIES ARE increasingly examining E the role of the knee adduction moment, which concentrates joint load to the medial tibiofemoral compartment, in the pathogenesis of knee pain and osteoarthritis.1-4 For instance, the knee adduction moment has been associated with the longitudinal development of chronic knee pain1 and is a determinant of pain in people with2 and without radiographic knee osteoarthritis.3 The magnitude of the knee adduction moment is also associated with the size of the medial tibial plateau.4 Moreover, separate studies5,6 have confirmed that people with established knee pathology, including osteoarthritis, walk with larger than normal knee adduction moments. However, it is unclear whether the knee adduction moment occurring in 1 limb is representative of the moment occurring in the contralateral limb in people without lower-limb pathology. Determining whether biomechanic variables differ between the dominant and nondominant sides of the body is an important question for both clinical practice and research design. Gait asymmetries may reflect a biomechanic predisposition toward pathology, such as unilateral knee osteoarthritis. Moreover, asymmetric gait may result in a visible limp that some people find cosmetically unacceptable. Asymmetry between limbs is also an important methodologic consideration when collecting data. For instance, when data are collected for multiple joints or organs, deciding how best to analyze the dataset can be challenging. In contrast to single organs, such as the heart, limb analyses can be complicated by interlimb variability. Therefore, determining whether the dominant and nondominant limb peak external knee adductor moment are associated may enable a more validated and rationalized approach to limb selection when examining this commonly used biomechanic variable. The aim of this study was to determine whether the peak external knee adduction moments occurring in the dominant limb are representative of the same moments occurring in the contralateral limb during healthy adult locomotion.
List of Abbreviations BMI CI SI
body mass index confidence interval symmetry index
DOMINANT AND NONDOMINANT PEAK EXTERNAL KNEE ADDUCTOR MOMENTS DURING GAIT, Teichtahl
METHODS Subjects Thirteen healthy women and 4 healthy men aged 20 to 50 years participated in this study. Subjects were recruited through the Jean Hailes Centre (a women’s health clinic in Australia) from an existing study examining healthy aging, and the remaining subjects were volunteers from our institution. Volunteers were excluded from the study if they (1) had a history of lower-limb arthropathies or pain in the month before testing; (2) had experienced significant pain or symptoms requiring medical treatment in the month before testing; (3) were currently using analgesic or anti-inflammatory medications; and (4) had been previously diagnosed with any neurologic, cardiovascular, or orthopedic conditions that may affect their gait. These exclusion criteria best isolated subjects with unimpaired gait patterns. The investigation was approved by the La Trobe University and Alfred Hospital Ethics Committee. All subjects were required to give written informed consent. Apparatus and Procedure Three-dimensional gait analyses were conducted in the Musculoskeletal Research Centre, La Trobe University, Australia. A 6-camera Vicon motion analysis systema was used to capture 3-dimensional kinematic data during 4 walking trials on both the right and left legs at a frequency of 50Hz. Ground reaction forces were measured by a Kistler 9281 force-platformb with a sampling frequency of 400Hz, which is the standard sampling frequency for several previous studies7,8 that have examined knee joint kinetics within our gait laboratory. Inverse dynamic analyses were performed by using PlugInGaitc software to obtain external joint moments calculated about an orthogonal axis system located in the distal segment of the joint. The peak external knee adduction moments during early and late stance (fig 1) were recorded while the subjects were instructed to walk barefoot at a normal pace along a 14-m walking strip to capture their natural gait patterns. The dominant limb was defined as the leg from which the individual stepped off when initiating gait. In this study, all subjects nominated their right limb as their dominant side, which was subsequently assessed and confirmed by a blinded observer witnessing each gait trial. Retroreflective markers and a knee-alignment device were placed in accordance with the specifications recommended by the Vicon Clinical Manager’s User Manual. Markers were placed on the left and right anterior superior iliac spine, thigh (lower lateral third), ankle (lateral malleoli), shank (lower third), forefoot (second metatarsal head on the midfoot side of the equines break
Fig 1. An example of an external knee adduction moment profile. NOTE. BW*HT body mass/weight (kg) multiplied by height (m). *Peak external knee adduction moment during early stance; **Peak external knee adduction moment during late stance.
321
between the forefoot and midfoot), heel (such that a line joining the forefoot markers reflected the long axis of the foot), and sacrum. The BMI (mass/height2 kg/m2) was calculated by measuring mass to the nearest 0.1kg by using a single pair of electronic scales and measuring height to the nearest 0.1cm by using a stadiometer (shoes and bulky clothing removed). Statistical Analysis Data were initially examined for features that would impede interpretation such as nonnormality, nonlinearity of the associations, and outlying observations. Knee moments were normalized to a percentage of body weight (kg) multiplied by height (m) (% BW*HT) for all analyses. There was no significant difference between the average walking speed of the dominant and nondominant limb trials. A paired-samples correlation was used to determine the association between knee adductor moments for the dominant and nondominant sides. Paired sample t tests were used to determine whether any significant differences existed between the dominant and nondominant peak knee adductor moments. Moreover, linear regression analyses were used to determine associations between the dominant and nondominant leg peak knee adduction moments during early and late stance before and after adjustment for age and sex. We also used a previously developed SI to further characterize the relationship between dominant and nondominant adductor moments,9,10 whereby SI (%) was calculated from the following equation: 100 ([nondominant moment ⫺ dominant moment]/0.5[nondominant moment ⫹ dominant moment]). An SI value of 0 represents perfect between-limb symmetry. Results with P values less than .05 (2 tailed) were considered to be statistically significant. All analyses were performed by using SPSS version 15.0.d RESULTS Thirteen women and 4 men participated in the study. The mean age of the cohort was 41.3⫾8.8 years, and their average BMI was 24.6⫾5.7kg/m2. The mean height and weight were 1.7⫾0.1m and 68.1⫾4.4kg, respectively, wheres the mean walking speed conformed with normative data for healthy gait (1.39⫾0.16 m/s).11 The mean values and SD of the peak external knee adductor moments during early and late stance are presented in table 1 and figure 2. There were no significant differences between the mean dominant and nondominant peak knee adductor moment values for either early or late stance (see table 1), and the peak knee adductor moments were moderately correlated between the dominant and nondominant legs during early (R⫽.61, P⫽.009) and late (R⫽.72, P⫽.001) stance (fig 3). In multivariate linear regression analyses, after adjustment for age and sex, we showed that there was an associated 0.58 (95% CI, 0.05–1.11; P⫽.030) and 0.98 (95% CI, 0.29 –1.67; P⫽.009) unit increase in the peak knee adduction moment in the nondominant leg, for every 1 unit increase in the dominant leg peak knee adductor moment during the early and late stance phases of gait, respectively (see table 1). Further evidence for symmetry was provided by the SI, which was 0.04% and 0.62% for early and late stance, respectively. DISCUSSION This study has shown that the dominant (right) and nondominant (left) peak external knee adduction moments are moderately correlated and significantly associated with one another during the stance phase, and in particular the late stance phase, of healthy adult locomotion. These results provide evidence for symmetrical peak knee adductor moments between limbs during the early and late stance phases of healthy adult walking. Arch Phys Med Rehabil Vol 90, February 2009
322
DOMINANT AND NONDOMINANT PEAK EXTERNAL KNEE ADDUCTOR MOMENTS DURING GAIT, Teichtahl Table 1: Comparison of Peak External Knee Adductor Moments in the Dominant and Nondominant Knees
Dominant Knee
Early stance Late stance
Nondominant Knee
Mean Difference Between Dominant and Nondominant Knees
Multivariate Analyses (95% CI)*
Mean
SD
Mean
SD
Mean Difference
P
Adjusted Mean Difference
P
4.10 2.32
1.20 0.82
4.09 2.25
1.17 0.60
0.01 0.07
.97 .60
0.58 (0.05–1.11) 0.98 (0.29–1.67)
.034 .009
NOTE. Knee moments are normalized to a percentage of body weight multiplied by height (% BW*HT). *Change in nondominant leg peak knee adduction moment (% BW*HT) per 1-unit increase in dominant leg peak knee adduction moment (% BW*HT) after adjustment for age and sex.
No previous study has examined whether the peak knee adduction moment is associated between dominant and nondominant limbs. Previously, the peak knee adduction moment during late stance was found to be an important determinant of medial tibiofemoral joint load12 as well as knee symptoms including pain1-3 and joint structures, such as medial tibial bone size.4 Therefore, determining whether the adductor moment occurring at 1 knee is representative of the same moment occurring at the contralateral limb during the late stance phase of gait would be of particular biomechanic interest. In the current study, we found a significant correlation (r⫽0.72, P⫽0.001) between the dominant and nondominant peak knee adduction moment during late stance. This suggests that over half the variance (R2⫽.52) in the peak knee adduction moment occurring during late stance is accounted for by the variance in the same moment in the contralateral knee. Therefore, although other variables may influence adductor moment variability, the peak moment occurring in 1 knee is the predominant determinant of the variability of the moment occurring in the contralateral knee. Moreover, this result may infer that the same factors within a person impact symmetrically on gait, accounting for 52% of the variability. Taken together, this implies symmetry between knee adductor moments. This is further substantiated by SI values of 0.04% and 0.62% for early and late stance, respectively. Although no other study has examined the SI between the dominant and nondominant adductor moments during healthy gait, these values indicate a high degree of symmetry in the knee adduction moments of our subjects. In
addition, linear regression analyses also provide conclusive evidence for symmetry, showing that for every unit increase in the peak knee adduction moment in the dominant leg in late stance, there was an associated 0.98-unit increase in the peak adductor moment in the contralateral leg after adjustment for age and sex (P⫽.009). Although our cross-sectional results imply symmetry between the peak knee adductor moments, it is impossible to determine whether this protects against unilateral knee pathologies, such as knee osteoarthritis. Although an increased knee adduction moment has been shown among people with knee osteoarthritis,5,6 it is unclear whether this is a cause or result of the disease process. It is possible that people at risk of osteoarthritis, such as those with a genetic predisposition toward the disease,13-15 may show asymmetric gait patterns that contribute toward the development of unilateral joint pathology. Previously, we have shown that the offspring of people with medial tibiofemoral osteoarthritis walk with less foot rotation than normal controls.16 Whether gait symmetry protects against the onset of unilateral (or increases the risk for bilateral) pathologic joint changes will need to be confirmed longitudinally. The results of this study have direct implications for research design. Nonindependence of the units being analyzed needs to be considered when examining data from multiple organs.17 An elementary approach to overcoming this problem was to analyze 1 unit per person. For example, limb studies have often selected the dominant side, given
Fig 2. Boxplots comparing the dominant and nondominant peak external knee adductor moments during both early and late stance.
Arch Phys Med Rehabil Vol 90, February 2009
DOMINANT AND NONDOMINANT PEAK EXTERNAL KNEE ADDUCTOR MOMENTS DURING GAIT, Teichtahl
323
Fig 3. Scatterplots of the association between the dominant and nondominant peak knee adduction moments during the (A) early and (B) late stance phases of gait.
that it may have the most functional significance. There does not appear to be empiric evidence validating this approach, and several studies have tended to perform separate analyses of each limb to better account for interlimb variability. Spreading the data over 2 analyses reduces statistical power, potentially leading to inaccurate reporting of datasets and inappropriate rejection of the null hypothesis. This study is the first to provide evidence that random selection of a knee may be representative of the peak knee adductor moments occurring on the contralateral side when analyzing the stance phase of healthy adult locomotion. Study Limitations A potential limitation of this study was the relatively small sample size. Nevertheless, the mean values for the peak knee adductor moments in this study are comparable to previously published normative values for knee joints without significant pathology.5,6 This implies that even though our cohort was small, it was representative of the normal population for the peak adductor moment values. Additionally, although there was no significant difference between walking speeds of dominant and nondominant trials, we did not control for other variables that may influence joint moments, such as potential leg length discrepancies or other asymmetries that may have been apparent. By analyzing the natural walking velocity and without making subjects walk at a prescribed qualitative pace, we were able to simulate functional rather than unnatural gait patterns, which may have otherwise affected the results and potentially predated asymmetry. Furthermore, although our aim was to compare the dominant versus the nondominant knees, all dominant knees were right sided. Therefore, our results could also be interpreted in the context of right and left leg symmetry. Finally, because we examined healthy subjects, our results cannot be generalized to people with lower-limb pathology, including any of the arthropathies.
CONCLUSIONS In healthy subjects, the average peak external knee adduction moment during stance, and in particular late stance, occurring in the dominant limb appears representative of those occurring in the nondominant limb. These results provide evidence for symmetric peak knee adductor moments during early and late stance phase during healthy adult walking. Whether altered biomechanics, including the asymmetry shown among people with unilateral joint disease5,6 are a cause or result of disease needs to be examined longitudinally. The random selection of 1 person’s knee when analyzing the peak knee adduction moment in healthy subjects appears to be a valid indicator of the average peak moments being exerted on the contralateral side during the stance phase of gait. Acknowledgments: We thank the staff at the Musculoskeletal Research Centre, La Trobe University for assisting with data collection including Tim Bach, MSc, PhD; Dong Cheng, BSci, PhD; Kate Webster, BSci (Hons), PhD; and Jo Wittwer, BAppSci (Physio), Grad Dip Physio (Neuro). We especially thank the study participants who made this study possible. References 1. Amin S, Luepongsak N, McGibbon CA, LaValley MP, Krebs DE, Felson DT. Knee adduction moment and development of chronic knee pain in elders. Arthritis Rheum 2004;51:371-6. 2. Hurwitz DE, Ryals AR, Block JA, Sharma L, Schnitzer TJ, Andriacchi TP. Knee pain and joint loading in subjects with osteoarthritis of the knee. J Orthop Res 2000;18:572-9. 3. Teichtahl AJ, Wluka AE, Morris ME, Davis SR, Cicuttini FM. The relationship between the knee adduction moment and knee pain in middle-aged women without radiographic osteoarthritis. J Rheumatol 2006;33:1845-8. 4. Jackson BD, Teichtahl AJ, Morris ME, Wluka AE, Davis SR, Cicuttini FM. The effect of the knee adduction moment on tibial cartilage volume and bone size in healthy women. Rheumatology (Oxford) 2004;43:311-4. Arch Phys Med Rehabil Vol 90, February 2009
324
DOMINANT AND NONDOMINANT PEAK EXTERNAL KNEE ADDUCTOR MOMENTS DURING GAIT, Teichtahl
5. Sharma L, Hurwitz DE, Thonar EJ, et al. Knee adduction moment, serum hyaluronan level, and disease severity in medial tibiofemoral osteoarthritis. Arthritis Rheum 1998;41:1233-40. 6. Schipplein OD, Andriacchi TP. Interaction between active and passive knee stabilizers during level walking. J Orthop Res 1991; 9:113-9. 7. Webster KE, Gonzalez-Adrio R, Feller JA. Dynamic joint loading following hamstring and patellar tendon anterior cruciate ligament reconstruction. Knee Surg Sports Traumatol Arthrosc 2004;12: 15-21. 8. Webster KE, Wittwer JE, O’Brien J, Feller JA. Gait patterns after anterior cruciate ligament reconstruction are related to graft type. Am J Sports Med 2005;33:247-54. 9. Robinson RO, Herzog W, Nigg BM. Use of force platform variables to quantify the effects of chiropractic manipulation on gait symmetry. J Manipulative Physiol Ther 1987;10:172-6. 10. White SC, Lifeso RM. Altering asymmetric limb loading after hip arthroplasty using real-time dynamic feedback when walking. Arch Phys Med Rehabil 2005;86:1958-63. 11. Bohannon RW. Comfortable and maximum walking speed of adults aged 20-79 years: reference values and determinants. Age Ageing 1997;26:15-9. 12. Andriacchi TP. Dynamics of knee malalignment. Orthop Clin North Am 1994;25:395-403.
Arch Phys Med Rehabil Vol 90, February 2009
13. Spector TD, Cicuttini F, Baker J, Loughlin J, Hart D. Genetic influences on osteoarthritis in women: a twin study. BMJ 1996;312:940-3. 14. Uitterlinden AG, Burger H, Huang Q, et al. Vitamin D receptor genotype is associated with radiographic osteoarthritis at the knee. J Clin Invest 1997;100:259-63. 15. Cicuttini FM, Baker JR, Spector TD. The association of obesity with osteoarthritis of the hand and knee in women: a twin study. J Rheumatol 1996;23:1221-6. 16. Teichtahl AJ, Morris ME, Wluka AE, Bach TM, Cicuttini FM. A comparison of gait patterns between the offspring of people with medial tibiofemoral osteoarthritis and normal controls. Clin Exp Rheumatol 2003;21:421-3. 17. Sutton AJ, Muir KR, Jones AC. Two knees or one person: data analysis strategies for paired joints or organs. Ann Rheum Dis 1997;56:401-2. Suppliers a. Six-camera Vicon motion analysis system; Vicon Motion Systems, 14 Minns Business Park, West Way, Exford OX2 0JB UK. b. Kistler 9281 force-platform; Kistler Instrumente AG, Verkauf Schweiz, Eulachstrasse 22, 8408 Winterthur, Switzerland. c. PlugInGait software; Oxford Metrics, 14 Minns Business Park, West Way, Oxford, UK OX2-0JB. d. SPSS version 15.0; SPSS Inc, 233 S Wacker Dr, 11th Fl, Chicago, IL 60606.
ORIGINAL ARTICLE
The Associations Between the Dominant and Nondominant Peak External Knee Adductor Moments During Gait in Healthy Subjects: Evidence for Symmetry Andrew J. Teichtahl, MBBS, B.Physio, Anita E. Wluka, MBBS, FRACP, PhD, Meg E. Morris, B.App Sci. (Physio), FACP, PhD, Susan R. Davis, MBBS, FRACP, PhD, Flavia M. Cicuttini, MBBS, FRACP, PhD ABSTRACT. Teichtahl AJ, Wluka AE, Morris ME, Davis SR, Cicuttini FM. The associations between the dominant and nondominant peak external knee adductor moments during gait in healthy subjects: evidence for symmetry. Arch Phys Med Rehabil 2009;90:320-4. Objectives: There is growing interest in the role of the knee adduction moment in the pathogenesis of knee pain and osteoarthritis. It is unclear whether the knee adduction moment is similar between the dominant and nondominant legs during locomotion. This study examined whether asymmetry exists in the peak knee adductor moments during gait in healthy adults. Design: Cross-sectional study. Setting: Musculoskeletal Research Centre, La Trobe University, Melbourne, Victoria, Australia. Participants: Three-dimensional Vicon gait analyses were performed for 17 healthy men and women. Interventions: Not applicable. Main Outcome Measures: The external dominant and nondominant peak knee adduction moments during early and late stance were analyzed to determine whether any significant differences occurred between limbs. Results: Peak knee adductor moments for dominant and nondominant limbs were significantly correlated during early (R⫽0.61, P⫽0.009) and late (R⫽0.72, P⫽0.001) stance. After adjustment for age and sex, there was an associated 0.58 (P⫽0.030) and 0.98 (P⫽0.009) unit increase in the peak knee adduction moment in the nondominant leg, for every 1 unit increase in the dominant leg during the early and late stance phases of gait, respectively. Further evidence for symmetry was provided by the symmetry index, which was 0.04% and 0.62% for early and late stance, respectively. Conclusions: In healthy subjects, the magnitude of the dominant limb peak external knee adduction moments during stance, and in particular late stance, appears representative of the magnitude of the moment in the nondominant limb. These findings imply symmetry between these moments and may have important implications when collecting data for limb analyses among healthy subjects. Whether gait symmetry protects against
From the Department of Epidemiology and Preventive Medicine (Teichtahl, Wluka, Cicuttini), and the National Health and Medical Research Council of Australia Centre of Clinical Research Excellence for the Study of Women’s Health (Davis), Monash University Medical School, Alfred Hospital; Baker Heart Research Institute (Wluka); and School of Physiotherapy, University of Melbourne (Morris), Melbourne, Australia. No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit on the authors or on any organization with which the authors are associated. Reprint requests to Flavia Cicuttini, MBBS, FRACP, PhD, Dept of Epidemiology and Preventive Medicine, Monash University, Alfred Hospital, Melbourne, Victoria 3004, Australia, e-mail: [email protected]. 0003-9993/09/9002-00136$36.00/0 doi:10.1016/j.apmr.2008.07.030
Arch Phys Med Rehabil Vol 90, February 2009
the onset of unilateral (or increases the risk for bilateral) pathological joint changes will need to be confirmed longitudinally. Key Words: Biomechanics; Gait; Knee; Rehabilitation. © 2009 by the American Congress of Rehabilitation Medicine PIDEMIOLOGIC STUDIES ARE increasingly examining E the role of the knee adduction moment, which concentrates joint load to the medial tibiofemoral compartment, in the pathogenesis of knee pain and osteoarthritis.1-4 For instance, the knee adduction moment has been associated with the longitudinal development of chronic knee pain1 and is a determinant of pain in people with2 and without radiographic knee osteoarthritis.3 The magnitude of the knee adduction moment is also associated with the size of the medial tibial plateau.4 Moreover, separate studies5,6 have confirmed that people with established knee pathology, including osteoarthritis, walk with larger than normal knee adduction moments. However, it is unclear whether the knee adduction moment occurring in 1 limb is representative of the moment occurring in the contralateral limb in people without lower-limb pathology. Determining whether biomechanic variables differ between the dominant and nondominant sides of the body is an important question for both clinical practice and research design. Gait asymmetries may reflect a biomechanic predisposition toward pathology, such as unilateral knee osteoarthritis. Moreover, asymmetric gait may result in a visible limp that some people find cosmetically unacceptable. Asymmetry between limbs is also an important methodologic consideration when collecting data. For instance, when data are collected for multiple joints or organs, deciding how best to analyze the dataset can be challenging. In contrast to single organs, such as the heart, limb analyses can be complicated by interlimb variability. Therefore, determining whether the dominant and nondominant limb peak external knee adductor moment are associated may enable a more validated and rationalized approach to limb selection when examining this commonly used biomechanic variable. The aim of this study was to determine whether the peak external knee adduction moments occurring in the dominant limb are representative of the same moments occurring in the contralateral limb during healthy adult locomotion.
List of Abbreviations BMI CI SI
body mass index confidence interval symmetry index
DOMINANT AND NONDOMINANT PEAK EXTERNAL KNEE ADDUCTOR MOMENTS DURING GAIT, Teichtahl
METHODS Subjects Thirteen healthy women and 4 healthy men aged 20 to 50 years participated in this study. Subjects were recruited through the Jean Hailes Centre (a women’s health clinic in Australia) from an existing study examining healthy aging, and the remaining subjects were volunteers from our institution. Volunteers were excluded from the study if they (1) had a history of lower-limb arthropathies or pain in the month before testing; (2) had experienced significant pain or symptoms requiring medical treatment in the month before testing; (3) were currently using analgesic or anti-inflammatory medications; and (4) had been previously diagnosed with any neurologic, cardiovascular, or orthopedic conditions that may affect their gait. These exclusion criteria best isolated subjects with unimpaired gait patterns. The investigation was approved by the La Trobe University and Alfred Hospital Ethics Committee. All subjects were required to give written informed consent. Apparatus and Procedure Three-dimensional gait analyses were conducted in the Musculoskeletal Research Centre, La Trobe University, Australia. A 6-camera Vicon motion analysis systema was used to capture 3-dimensional kinematic data during 4 walking trials on both the right and left legs at a frequency of 50Hz. Ground reaction forces were measured by a Kistler 9281 force-platformb with a sampling frequency of 400Hz, which is the standard sampling frequency for several previous studies7,8 that have examined knee joint kinetics within our gait laboratory. Inverse dynamic analyses were performed by using PlugInGaitc software to obtain external joint moments calculated about an orthogonal axis system located in the distal segment of the joint. The peak external knee adduction moments during early and late stance (fig 1) were recorded while the subjects were instructed to walk barefoot at a normal pace along a 14-m walking strip to capture their natural gait patterns. The dominant limb was defined as the leg from which the individual stepped off when initiating gait. In this study, all subjects nominated their right limb as their dominant side, which was subsequently assessed and confirmed by a blinded observer witnessing each gait trial. Retroreflective markers and a knee-alignment device were placed in accordance with the specifications recommended by the Vicon Clinical Manager’s User Manual. Markers were placed on the left and right anterior superior iliac spine, thigh (lower lateral third), ankle (lateral malleoli), shank (lower third), forefoot (second metatarsal head on the midfoot side of the equines break
Fig 1. An example of an external knee adduction moment profile. NOTE. BW*HT body mass/weight (kg) multiplied by height (m). *Peak external knee adduction moment during early stance; **Peak external knee adduction moment during late stance.
321
between the forefoot and midfoot), heel (such that a line joining the forefoot markers reflected the long axis of the foot), and sacrum. The BMI (mass/height2 kg/m2) was calculated by measuring mass to the nearest 0.1kg by using a single pair of electronic scales and measuring height to the nearest 0.1cm by using a stadiometer (shoes and bulky clothing removed). Statistical Analysis Data were initially examined for features that would impede interpretation such as nonnormality, nonlinearity of the associations, and outlying observations. Knee moments were normalized to a percentage of body weight (kg) multiplied by height (m) (% BW*HT) for all analyses. There was no significant difference between the average walking speed of the dominant and nondominant limb trials. A paired-samples correlation was used to determine the association between knee adductor moments for the dominant and nondominant sides. Paired sample t tests were used to determine whether any significant differences existed between the dominant and nondominant peak knee adductor moments. Moreover, linear regression analyses were used to determine associations between the dominant and nondominant leg peak knee adduction moments during early and late stance before and after adjustment for age and sex. We also used a previously developed SI to further characterize the relationship between dominant and nondominant adductor moments,9,10 whereby SI (%) was calculated from the following equation: 100 ([nondominant moment ⫺ dominant moment]/0.5[nondominant moment ⫹ dominant moment]). An SI value of 0 represents perfect between-limb symmetry. Results with P values less than .05 (2 tailed) were considered to be statistically significant. All analyses were performed by using SPSS version 15.0.d RESULTS Thirteen women and 4 men participated in the study. The mean age of the cohort was 41.3⫾8.8 years, and their average BMI was 24.6⫾5.7kg/m2. The mean height and weight were 1.7⫾0.1m and 68.1⫾4.4kg, respectively, wheres the mean walking speed conformed with normative data for healthy gait (1.39⫾0.16 m/s).11 The mean values and SD of the peak external knee adductor moments during early and late stance are presented in table 1 and figure 2. There were no significant differences between the mean dominant and nondominant peak knee adductor moment values for either early or late stance (see table 1), and the peak knee adductor moments were moderately correlated between the dominant and nondominant legs during early (R⫽.61, P⫽.009) and late (R⫽.72, P⫽.001) stance (fig 3). In multivariate linear regression analyses, after adjustment for age and sex, we showed that there was an associated 0.58 (95% CI, 0.05–1.11; P⫽.030) and 0.98 (95% CI, 0.29 –1.67; P⫽.009) unit increase in the peak knee adduction moment in the nondominant leg, for every 1 unit increase in the dominant leg peak knee adductor moment during the early and late stance phases of gait, respectively (see table 1). Further evidence for symmetry was provided by the SI, which was 0.04% and 0.62% for early and late stance, respectively. DISCUSSION This study has shown that the dominant (right) and nondominant (left) peak external knee adduction moments are moderately correlated and significantly associated with one another during the stance phase, and in particular the late stance phase, of healthy adult locomotion. These results provide evidence for symmetrical peak knee adductor moments between limbs during the early and late stance phases of healthy adult walking. Arch Phys Med Rehabil Vol 90, February 2009
322
DOMINANT AND NONDOMINANT PEAK EXTERNAL KNEE ADDUCTOR MOMENTS DURING GAIT, Teichtahl Table 1: Comparison of Peak External Knee Adductor Moments in the Dominant and Nondominant Knees
Dominant Knee
Early stance Late stance
Nondominant Knee
Mean Difference Between Dominant and Nondominant Knees
Multivariate Analyses (95% CI)*
Mean
SD
Mean
SD
Mean Difference
P
Adjusted Mean Difference
P
4.10 2.32
1.20 0.82
4.09 2.25
1.17 0.60
0.01 0.07
.97 .60
0.58 (0.05–1.11) 0.98 (0.29–1.67)
.034 .009
NOTE. Knee moments are normalized to a percentage of body weight multiplied by height (% BW*HT). *Change in nondominant leg peak knee adduction moment (% BW*HT) per 1-unit increase in dominant leg peak knee adduction moment (% BW*HT) after adjustment for age and sex.
No previous study has examined whether the peak knee adduction moment is associated between dominant and nondominant limbs. Previously, the peak knee adduction moment during late stance was found to be an important determinant of medial tibiofemoral joint load12 as well as knee symptoms including pain1-3 and joint structures, such as medial tibial bone size.4 Therefore, determining whether the adductor moment occurring at 1 knee is representative of the same moment occurring at the contralateral limb during the late stance phase of gait would be of particular biomechanic interest. In the current study, we found a significant correlation (r⫽0.72, P⫽0.001) between the dominant and nondominant peak knee adduction moment during late stance. This suggests that over half the variance (R2⫽.52) in the peak knee adduction moment occurring during late stance is accounted for by the variance in the same moment in the contralateral knee. Therefore, although other variables may influence adductor moment variability, the peak moment occurring in 1 knee is the predominant determinant of the variability of the moment occurring in the contralateral knee. Moreover, this result may infer that the same factors within a person impact symmetrically on gait, accounting for 52% of the variability. Taken together, this implies symmetry between knee adductor moments. This is further substantiated by SI values of 0.04% and 0.62% for early and late stance, respectively. Although no other study has examined the SI between the dominant and nondominant adductor moments during healthy gait, these values indicate a high degree of symmetry in the knee adduction moments of our subjects. In
addition, linear regression analyses also provide conclusive evidence for symmetry, showing that for every unit increase in the peak knee adduction moment in the dominant leg in late stance, there was an associated 0.98-unit increase in the peak adductor moment in the contralateral leg after adjustment for age and sex (P⫽.009). Although our cross-sectional results imply symmetry between the peak knee adductor moments, it is impossible to determine whether this protects against unilateral knee pathologies, such as knee osteoarthritis. Although an increased knee adduction moment has been shown among people with knee osteoarthritis,5,6 it is unclear whether this is a cause or result of the disease process. It is possible that people at risk of osteoarthritis, such as those with a genetic predisposition toward the disease,13-15 may show asymmetric gait patterns that contribute toward the development of unilateral joint pathology. Previously, we have shown that the offspring of people with medial tibiofemoral osteoarthritis walk with less foot rotation than normal controls.16 Whether gait symmetry protects against the onset of unilateral (or increases the risk for bilateral) pathologic joint changes will need to be confirmed longitudinally. The results of this study have direct implications for research design. Nonindependence of the units being analyzed needs to be considered when examining data from multiple organs.17 An elementary approach to overcoming this problem was to analyze 1 unit per person. For example, limb studies have often selected the dominant side, given
Fig 2. Boxplots comparing the dominant and nondominant peak external knee adductor moments during both early and late stance.
Arch Phys Med Rehabil Vol 90, February 2009
DOMINANT AND NONDOMINANT PEAK EXTERNAL KNEE ADDUCTOR MOMENTS DURING GAIT, Teichtahl
323
Fig 3. Scatterplots of the association between the dominant and nondominant peak knee adduction moments during the (A) early and (B) late stance phases of gait.
that it may have the most functional significance. There does not appear to be empiric evidence validating this approach, and several studies have tended to perform separate analyses of each limb to better account for interlimb variability. Spreading the data over 2 analyses reduces statistical power, potentially leading to inaccurate reporting of datasets and inappropriate rejection of the null hypothesis. This study is the first to provide evidence that random selection of a knee may be representative of the peak knee adductor moments occurring on the contralateral side when analyzing the stance phase of healthy adult locomotion. Study Limitations A potential limitation of this study was the relatively small sample size. Nevertheless, the mean values for the peak knee adductor moments in this study are comparable to previously published normative values for knee joints without significant pathology.5,6 This implies that even though our cohort was small, it was representative of the normal population for the peak adductor moment values. Additionally, although there was no significant difference between walking speeds of dominant and nondominant trials, we did not control for other variables that may influence joint moments, such as potential leg length discrepancies or other asymmetries that may have been apparent. By analyzing the natural walking velocity and without making subjects walk at a prescribed qualitative pace, we were able to simulate functional rather than unnatural gait patterns, which may have otherwise affected the results and potentially predated asymmetry. Furthermore, although our aim was to compare the dominant versus the nondominant knees, all dominant knees were right sided. Therefore, our results could also be interpreted in the context of right and left leg symmetry. Finally, because we examined healthy subjects, our results cannot be generalized to people with lower-limb pathology, including any of the arthropathies.
CONCLUSIONS In healthy subjects, the average peak external knee adduction moment during stance, and in particular late stance, occurring in the dominant limb appears representative of those occurring in the nondominant limb. These results provide evidence for symmetric peak knee adductor moments during early and late stance phase during healthy adult walking. Whether altered biomechanics, including the asymmetry shown among people with unilateral joint disease5,6 are a cause or result of disease needs to be examined longitudinally. The random selection of 1 person’s knee when analyzing the peak knee adduction moment in healthy subjects appears to be a valid indicator of the average peak moments being exerted on the contralateral side during the stance phase of gait. Acknowledgments: We thank the staff at the Musculoskeletal Research Centre, La Trobe University for assisting with data collection including Tim Bach, MSc, PhD; Dong Cheng, BSci, PhD; Kate Webster, BSci (Hons), PhD; and Jo Wittwer, BAppSci (Physio), Grad Dip Physio (Neuro). We especially thank the study participants who made this study possible. References 1. Amin S, Luepongsak N, McGibbon CA, LaValley MP, Krebs DE, Felson DT. Knee adduction moment and development of chronic knee pain in elders. Arthritis Rheum 2004;51:371-6. 2. Hurwitz DE, Ryals AR, Block JA, Sharma L, Schnitzer TJ, Andriacchi TP. Knee pain and joint loading in subjects with osteoarthritis of the knee. J Orthop Res 2000;18:572-9. 3. Teichtahl AJ, Wluka AE, Morris ME, Davis SR, Cicuttini FM. The relationship between the knee adduction moment and knee pain in middle-aged women without radiographic osteoarthritis. J Rheumatol 2006;33:1845-8. 4. Jackson BD, Teichtahl AJ, Morris ME, Wluka AE, Davis SR, Cicuttini FM. The effect of the knee adduction moment on tibial cartilage volume and bone size in healthy women. Rheumatology (Oxford) 2004;43:311-4. Arch Phys Med Rehabil Vol 90, February 2009
324
DOMINANT AND NONDOMINANT PEAK EXTERNAL KNEE ADDUCTOR MOMENTS DURING GAIT, Teichtahl
5. Sharma L, Hurwitz DE, Thonar EJ, et al. Knee adduction moment, serum hyaluronan level, and disease severity in medial tibiofemoral osteoarthritis. Arthritis Rheum 1998;41:1233-40. 6. Schipplein OD, Andriacchi TP. Interaction between active and passive knee stabilizers during level walking. J Orthop Res 1991; 9:113-9. 7. Webster KE, Gonzalez-Adrio R, Feller JA. Dynamic joint loading following hamstring and patellar tendon anterior cruciate ligament reconstruction. Knee Surg Sports Traumatol Arthrosc 2004;12: 15-21. 8. Webster KE, Wittwer JE, O’Brien J, Feller JA. Gait patterns after anterior cruciate ligament reconstruction are related to graft type. Am J Sports Med 2005;33:247-54. 9. Robinson RO, Herzog W, Nigg BM. Use of force platform variables to quantify the effects of chiropractic manipulation on gait symmetry. J Manipulative Physiol Ther 1987;10:172-6. 10. White SC, Lifeso RM. Altering asymmetric limb loading after hip arthroplasty using real-time dynamic feedback when walking. Arch Phys Med Rehabil 2005;86:1958-63. 11. Bohannon RW. Comfortable and maximum walking speed of adults aged 20-79 years: reference values and determinants. Age Ageing 1997;26:15-9. 12. Andriacchi TP. Dynamics of knee malalignment. Orthop Clin North Am 1994;25:395-403.
Arch Phys Med Rehabil Vol 90, February 2009
13. Spector TD, Cicuttini F, Baker J, Loughlin J, Hart D. Genetic influences on osteoarthritis in women: a twin study. BMJ 1996;312:940-3. 14. Uitterlinden AG, Burger H, Huang Q, et al. Vitamin D receptor genotype is associated with radiographic osteoarthritis at the knee. J Clin Invest 1997;100:259-63. 15. Cicuttini FM, Baker JR, Spector TD. The association of obesity with osteoarthritis of the hand and knee in women: a twin study. J Rheumatol 1996;23:1221-6. 16. Teichtahl AJ, Morris ME, Wluka AE, Bach TM, Cicuttini FM. A comparison of gait patterns between the offspring of people with medial tibiofemoral osteoarthritis and normal controls. Clin Exp Rheumatol 2003;21:421-3. 17. Sutton AJ, Muir KR, Jones AC. Two knees or one person: data analysis strategies for paired joints or organs. Ann Rheum Dis 1997;56:401-2. Suppliers a. Six-camera Vicon motion analysis system; Vicon Motion Systems, 14 Minns Business Park, West Way, Exford OX2 0JB UK. b. Kistler 9281 force-platform; Kistler Instrumente AG, Verkauf Schweiz, Eulachstrasse 22, 8408 Winterthur, Switzerland. c. PlugInGait software; Oxford Metrics, 14 Minns Business Park, West Way, Oxford, UK OX2-0JB. d. SPSS version 15.0; SPSS Inc, 233 S Wacker Dr, 11th Fl, Chicago, IL 60606.