Gait & Posture
Volume 35, Issue 1 , Pages 6-10, January 2012

Strategies used during a challenging weighted walking task in healthy adults and individuals with knee osteoarthritis

Center for Biomedical Engineering Research, University of Delaware, Newark, DE 19716, USA

Received 8 September 2010; received in revised form 10 July 2011; accepted 11 July 2011. published online 03 October 2011.

Article Outline

Highlights

► Healthy adults and those with knee OA walked at 1.0 m/s unweighted and with a weight vest. ► Healthy adults respond differently to weighted walking than individuals with knee OA. ► Weighted healthy group increases hip flexion at heel strike but knee OA does not. ► The knee OA group only made minor compensations during weighted walking.

Abstract 

Knee osteoarthritis (OA) is a disease that affects millions of people. While numerous gait differences have been identified between healthy adults and adults with knee OA under normal and challenging conditions, adults with knee OA have not been studied during a challenging weighted walking task. Investigation of the effect of weighted walking on the initial contact and loading response phases of gait was undertaken in 20 healthy and 20 knee OA subjects ages 40–85 years old walking at 1.0m/s while unweighted and weighted with 1/6th of their body weight in a weight vest. Subjects were grouped according to their Kellgren and Lawrence radiographic score and healthy subjects were age-matched to those with knee OA. ANOVA revealed significant effects for hip flexion angle at initial contact, step length, initial double support percent, and load rate. Post hoc t-tests revealed that subjects with knee OA had a larger initial double support percent and hip flexion angle at initial contact and a decreased load rate compared to unweighted, healthy adults. Also, both groups increased their initial double support percent in response to the challenging weighted walking task, but only the healthy adults increased their hip flexion angle at initial contact and decreased their load rate. During the weighted condition, the knee OA group had a shorter step length compared to the healthy group. Because the knee OA group only made minor compensations to their gait strategy, it appears that they may be unable or prefer not to adjust their gait mechanics due to underlying issues.

Keywords: Weight vest, Osteoarthritis, Knee, Challenge, Gait

 

Back to Article Outline

1. Introduction 

Knee osteoarthritis (OA) affects approximately 27 million adults in the United States and 25% of the population aged 25–74 [1]. The effects of knee OA can impact economic [2], social [3], psychological [4], physical [5], and functional [6] aspects of life. Biomechanical changes that occur in a person's gait pattern may lead to further progression of the disease [7], [8]. The knee adduction moment during gait has been the focus of much scrutiny because of its relation specifically to medial joint loading [9], [10]. Joint loading has been hypothesized to be the driving factor in the incidence and progression of knee OA [7].

Using challenging conditions to impose increased demand on the neuromusculoskeletal system can be beneficial to understanding movement strategies especially in an impaired population. Individuals with knee OA have been studied while walking quickly [11], [12], negotiating stairs [13], and crossing an obstacle [14]. Challenging these individuals to a task that is more strenuous than their normal activities of daily living can uncover underlying movement impairments which otherwise would not be seen and may help researchers discover underlying impairments to target with therapeutic intervention. For example, when subjects with knee OA are challenged to walk up stairs, severe knee OA subjects exhibit significant forward trunk lean, that would not be present during level walking, and requires adjustments to their hip and knee flexion moments to compensate for weak quadriceps [13] which may be addressed with strength training.

When healthy, young adults are subjected to load carrying, they respond with decreased single support, increased initial double support, and decreased swing with little other effects on sagittal plane motion during stance [15], [16]. There is also an increased metabolic demand and load rate as the percent of body weight carried is increased [17]. In other studies, older adults have used a weighted vest for exercise purposes (for example [18]). Salem et al. [18] found that the kinetic effects of various weight vest loads were joint specific and load dependent. In their study, when the load was increased from 0% body weight to 3% body weight, there was only an increase in the peak external knee flexion moment, but comparing 0% body weight to 5% body weight, there was a significant increase in the peak external ankle dorsiflexion moment and knee flexion moment during the stance phase of gait [18].

The relationship between joint loading and knee OA [19], [20] and the change in knee kinematics during weight acceptance in normal walking [21], [22], [23] leads us to believe that there may be important strategies employed during loading which can influence the disease process. However, biomechanics during a challenging weighted walking condition have not been compared in healthy adults and adults with knee OA. The added load during a weighted condition will increase the forces on the lower extremity and may decrease the weight-bearing stability, possibly leading to compensatory biomechanical strategies. Therefore the objective of this study was to compare these strategies in individuals with knee OA and age-matched, healthy adults during the initial contact and loading response phases of gait.

Back to Article Outline

2. Methods 

Subjects between the ages of 40 and 85 years old were recruited (Table 1). All subjects had a radiograph taken while standing with their toes pointed forward, knees flexed to 30 degrees, and weight equally distributed between the feet. The tibiofemoral joint in all radiographs was evaluated using the Kellgren and Lawrence scale (K/L [24]) by a board certified radiologist and the knee OA group consisted of those subjects who had a K/L score of ≥2, while the healthy group consisted of those subjects who were age-matched to the knee OA group and had a K/L score of ≤1. Twenty subjects were used for each group based on an a priori power analysis of dependent variables.

Table 1. The inclusion and exclusion criteria for subjects used in our study.
Inclusion
• Age: 40–85
• BMI: <40
• Ambulatory (including those who use a cane or a walker)
• Able to walk for 5min at self-selected speed
• Able to walk up to 2×30s bursts at fastest tolerable speed on treadmill without assistance
Exclusion
• Congestive heart failure
• Any other heart problems or heart murmur
• Peripheral artery disease with claudication
• Cancer
• Pulmonary or renal failure
• Unstable angina
• Uncontrolled hypertension (>190/110mmHg)
• Dizziness and/or neurological disorder (stroke, Parkinson's, etc.)
• Pregnancy
• Joint replacement or pacemaker with metallic parts
• OA due to significant bony deformity
• ACL deficient knees
• Diagnosed arthritis of other lower extremity joints
• Any reason why they should not exert themselves physically
• Other orthopaedic condition affecting ambulation
• If answers YES to 2 of these:
Smokes
Diabetes
Family history of heart disease prior to age 55
Unmedicated high cholesterol
• Joint injections within 6 months
• Lower extremity surgical procedure that would affect ambulation

These criteria were formed in order to have an age group representative of the general knee OA population, however those subjects with other medical comorbidities were excluded so that the results were easier to interpret. Subjects were excluded if they were not physically capable of performing the challenges in the study or had previous or current cardiovascular, orthopaedic, or neurologic issues.

Each subject received a full explanation of the study including the risks of participation and provided written informed consent. This study was approved by the University of Delaware Human Subjects Review Board.

Subjects completed the Knee Injury and Osteoarthritis Outcome Survey (KOOS) which is a subjective questionnaire that has five subsections asking about symptoms, pain, activities of daily living, sports and recreation, and quality of life [25]. An isometric strength test was used to estimate knee flexor and extensor strength for each subject. Measurements were taken on a Biodex Systems 3 (Biodex Medical Systems, Shirley, NY, USA) dynamometer using a 60 degree isometric knee flexion and extension protocol. Each subject sat with the thigh and hips secured in the Biodex chair and performed three 3s bouts of isometric muscle contractions for each muscle group with 20s of rest in between. During the active muscle contraction, verbal encouragement was provided in order to maximize the force output.

Self-selected walking speed was measured over the middle 10m of a 30m segment of a hallway. Two walking passes were made and timed by a stopwatch. The average time was then used to determine the subject's self-selected walking speed in meters per second.

Gait data were collected with an 8 camera Motion Analysis system (Santa Rosa, CA) and an instrumented split-belt treadmill (Bertec Corporation, Worthington, OH). The kinematic data were collected at 60Hz and the kinetics at 1080Hz (EVaRT 5.0.4 and Cortex 1.0.0.198; Motion Analysis Corp., Santa Rosa, CA). A static standing trial was collected using 27 markers. Before any walking trials were completed, we removed the markers bilaterally from the medial knees and ankles and allowed the subjects 6min of accommodation time [26] at their self-selected speed only for the unweighted condition. After accommodation, two 30s walking trials at 1.0m/s unweighted and while wearing a weighted vest were collected.

For the weighted condition, a front and back loaded weight vest (MiR Vest, Inc., San Jose, CA, USA) was used. The vest was placed over the head of each subject and an elastic strap was used to secure it. The weight was added to the trunk so as to allow unimpeded motion of the lower extremity during gait. One static trial on a single forceplate imbedded into the split-belt treadmill (Bertec Corporation, Worthington, OH) was used to get the most accurate weight of each subject while unweighted. The weight for the challenging condition was added in 2lbs increments until closest to 1/6th body weight and verified on the forceplate. This weight was moderately heavy for most subjects and was selected to pose a challenge but not completely disrupt gait. Subjects had no acclimatization time in the vest and walked with the weighted vest until two 30s walking trials at 1.0m/s had been completed.

Inverse dynamics were computed in Orthotrak 6.3.5 (Motion Analysis Corp., Santa Rosa, CA). From sequential gait cycles for each subject's 30s walking trial, the final five full gait cycles on each leg were extracted and averaged into one representative gait cycle for each subject normalized to 100% of the gait cycle. The peak kinematic and kinetic values were taken from the averaged trial but the spatiotemporal parameters were calculated from the last five gait cycles and averaged. For the knee OA group, only data from the more severely affected limb is reported and random legs were chosen for the healthy group.

A two-way repeated measures analysis of variance (ANOVA) was performed in SPSS 17.0 (SPSS Inc., Chicago, IL) to determine the effect of knee OA and the weighted condition on walking patterns. The independent variables were group (Healthy or Knee OA) and condition (Unweighted or Weighted). The dependent variables were ankle, knee and hip flexion angles at initial contact, knee flexion excursion, step length, load rate, peak braking force, peak hip and knee flexion moments and initial double limb support. We also tested for an interaction effect to analyze the differences between the changes of each group due to the additional weight (group×condition). Significance was set at p<0.05. If significance was found with the two-way repeated measures ANOVA, then two-tailed, paired t-tests were done within groups and two-tailed, Student's t-tests were done between groups to find exactly where the difference was occurring. Significance for the t-tests was also set at p<0.05.

Back to Article Outline

3. Results 

The subjects in our study were grouped according to their K/L score from a flexed knee radiograph and the healthy group was age-matched to the knee OA group (Table 2). The knee OA group had a higher weight (p=0.001), increased BMI (p<0.001), decreased isometric knee flexor (p=0.003) and extensor strength (p=0.001), decreased self-selected walking speed (p=0.006), and poorer reported KOOS scores regarding their pain (p<0.001) and symptoms (p<0.001). There was no significant difference between the age (p=0.713) and height (p=0.976) of the two groups. Specific to our study, the percent which each subject was weighted was significantly different (p=0.002), but the overall weight put on each subject was not (p=0.087).

Table 2. A comparison of general subject characteristics.
General subject characteristics
HealthyKnee OA
Number2020
Percent female (%)60.070.0
Age (years)63.6±10.062.45±8.73
Height (m)1.69±0.081.69±0.10
Weight (kg)*70.6±12.986.8±15.4
BMI (kg/m2)*24.7±3.6030.5±4.31
Weighted BMI (kg/m2)29.1±4.0335.3±4.82
Quadriceps strength (Nm/kg)*1.81±0.421.31±0.50
Hamstring strength (Nm/kg)*1.02±0.320.73±0.24
Self-selected walking speed (m/s)*1.32±0.171.17±0.13
KOOS pain*97.5±4.8265.8±23.0
KOOS symptoms*96.8±3.9363.9±24.9
Percent weighted (%)*17.8±2.1815.9±1.42
Amount weighted (kg)12.49±2.0913.70±2.27

Data is presented as mean±1 standard deviation. An asterisk denotes a significant difference between the groups. Significance was set at p≤0.05.

The hip flexion angle at initial contact (Table 3, Fig. 1A) showed a significant effect for condition (p=0.001), and post hoc tests revealed that only the healthy group (p<0.001) increased their hip flexion angle at initial contact in response to the load while the knee OA group did not (p=0.225). The effect of group (p=0.051) was not significant nor was there a significant interaction effect for group×condition (p=0.138).

Table 3. The variables that were analyzed during this study.
Gait variablesUnweighted healthyWeighted healthyUnweighted knee OAWeighted knee OA
Ankle dorsiflexion angle at initial contact (degrees)1.04±3.081.21±2.442.27±4.142.13±4.68
Knee flexion angle at initial contact (degrees)3.35±6.743.73±6.886.12±10.106.89±10.14
Hip flexion angle at initial contact (degrees)27.37±6.89b29.27±7.35a33.29±9.6134.08±9.81
Step length (cm)45.58±3.6045.58±3.60d42.30±6.4141.82±5.42b
Initial double support percent (percent)13.92±1.22b,c15.20±1.32a,d15.89±2.22a,d16.72±2.02b,c
Peak braking force (N/kg)−0.14±0.02−0.13±0.01−0.14±0.03−0.14±0.03
Peak external hip flexion moment (Nm/kg)0.53±0.130.52±0.090.60±0.180.56±0.21
Peak external knee flexion moment (Nm/kg)0.41±0.150.38±0.140.38±0.170.39±0.19
Knee flexion excursion (degrees)13.34±4.5013.41±5.5412.06±4.4711.82±4.61
Load rate (N/kg/s)7.23±1.71c6.65±1.166.09±1.01a6.11±1.50

The values are shown as mean±1 standard deviation. Significance during post hoc t-tests is denoted by letters representing which group was compared (a, unweighted healthy; b, weighted healthy; c, unweighted knee OA; d, weighted knee OA). Significance was set at p<0.05.

  • View full-size image.
  • Fig. 1. 

    (A) Hip flexion angle at initial contact. The y-axis represents the positive hip flexion angle in degrees (°). The weighted, healthy group and the unweighted, knee OA group had a significantly larger hip flexion angle at initial contact compared to the unweighted, healthy group. (B) Step length with the more affected limb forward. The y-axis is the length of the step in centimeters (cm). The weighted, knee OA group had a significantly shorter step length compared to the weighted, healthy group. (C) Initial double limb support percent with the more affected limb forward. The y-axis is the percent (%) of the gait cycle (out of 100%) that the subject is in double limb support with their more affected limb forward. The weighted, healthy group and the unweighted, knee OA group had a significantly longer double support percent compared to the unweighted, healthy group. Also, the weighted, healthy group and the unweighted, knee OA group had significantly shorter double support percent compared to the weighted, knee OA group. (D) Load rate, or the slope of the vertical GRF curve between heel strike and the first peak in vertical GRF. The y-axis is the change in force normalized to the subject's weight divided by the change in time (N/kg/s). The weighted, healthy group and the unweighted, knee OA group had a significantly lower load rate compared to the unweighted, healthy group. Data are shown with mean±1 standard deviation. Brackets indicate statistically significant differences where p<0.05. Results are presented for four groups: (1) black=unweighted, healthy group; (2) white=weighted, healthy group; (3) gray=unweighted, knee OA group; and (4) stripes=weighted, knee OA group.

The step length data (Table 3, Fig. 1B) showed no significant differences for condition (p=0.525) or for the group×condition effect (p=0.530). There was a significant difference for the group effect (p=0.025). Post hoc analysis revealed that there was no significant difference between the healthy and knee OA groups in the unweighted condition (p=0.272), but the knee OA group had a shorter step length compared to healthy in the weighted condition (p=0.027).

The ANOVA yielded a main effect of condition (p<0.001) with both the healthy (p<0.001) and knee OA (p=0.018) groups showing a larger percent of gait in initial double support during the weighted condition compared to themselves during the unweighted condition (Table 3, Fig. 1C). There was also a main effect of group (p=0.002), and post hoc t-tests also showed a significant difference between the groups in both conditions (unweighted p=0.002; weighted p=0.027) with the knee OA group having a significantly longer initial double support percent in both conditions. No group×condition effect (p=0.216) was observed.

Load rate (Table 3, Fig. 1D) showed no significant difference in the ANOVA model for condition (p=0.057), but there was a significant difference for the group (p=0.038) and group×condition interaction effects (p=0.043). Between group differences existed during the unweighted condition (p=0.009) with the knee OA group having a lower load rate than their healthy, age-matched counterparts, but there was no statistical difference between groups during the weighted condition (p=0.211).

Back to Article Outline

4. Discussion 

The knee OA subjects were heavier, with higher BMI, increased pain and symptoms, and decreased isometric knee flexor and extensor strength on their more affected limb which is consistent with previous studies [22], [27]. Also, consistent with reports that knee OA subjects compensate for pain and loading by decreasing walking speed (for example [28]), our knee OA group had a slower self-selected walking speed. However, when walking unweighted at a control speed of 1.0m/s, their load rate was less than healthy, age-matched adults.

The current results also reveal an increased hip flexion angle at initial contact, due to both an increase in the pelvic tilt and an increase in the thigh flexion calculated post hoc, and increased initial double support percent with the more affected limb forward in the unweighted knee OA group and the weighted healthy group compared to the unweighted healthy group. A recent study of young healthy soldiers has shown an increase in hip range of motion under a load that is roughly 20% of their body weight [29]. We did not report range of motion in our study, but the healthy group did increase their hip flexion angle at initial contact in response to roughly 15% of their body weight while the knee OA subjects did not. The increase in motion at the hip seen in soldiers [29] could be the best available and preferred response to walking under a loaded condition. This strategy may have been implemented by the healthy group but was not selected by the knee OA group possibly because they already use this compensation due to their increased body weight compared to healthy adults.

When the healthy group was weighted, they more closely resembled the unweighted knee OA group in terms of BMI and gait parameters including initial double support percent, hip flexion at initial contact and load rate. The increase in initial double support percent has been seen in older adult populations who have a fear of falling [30] and could be in response to decreased stability. The decreased load rate is contrary to previous studies involving young adults during weighted walking tasks [15], but may be due to our normalization of the kinetics to the weight of the person plus the vest during the weighted condition.

The knee OA group increased their initial double support percent when weighted, and they had a significantly smaller step length compared to the weighted healthy group. Similar to the weighted healthy group, the weighted knee OA group could be increasing their initial double support percent for stability reasons. The smaller step length could be a result of pain potentially induced during the challenging task which may have caused the knee OA subjects to modify their walking pattern.

While undergoing this challenging weighted walking task, the healthy adults responded with different biomechanics during the weight acceptance phase of gait. The results suggest that the control strategy responsible for reduced load rate (as employed by healthy adults) was either not available to or not preferred by individuals with knee OA and this limited compensatory strategy could potentially be a factor contributing to disease progression. Also, there may be some underlying limitations present in knee OA subjects, namely decreased relative strength, that prevented them from changing some of their gait mechanics when challenged. Knee OA subjects are forced to use a higher percent of their peak isometric force during normal and challenged walking which could limit the compensations they can make in these conditions. Altered neuromuscular control resulting in enhanced co-contraction and stiffness in knee OA subjects [27], [31] may further explain the limited range of compensatory strategies observed.

Although using a controlled speed of 1.0m/s allowed for a better statistical comparison, walking at a speed slower than normal could have influenced the results. In addition, because we wanted to explore global group differences in compensatory strategies for weighted walking, we chose not to match by gender or weight. Consequently, there was a higher percentage of women in the knee OA group than healthy group which could influence the results. Body weight was not matched but rather accounted for through normalization. Furthermore, we did not have an equivalent relative weighted condition due to the limitations of our weight vest (i.e., 2lbs increments). As a result, the changes seen may be related to absolute weight added which was slightly but not significantly higher for OA subjects. Future studies might also consider using varying levels of challenging weighted walking conditions as only one weight was added in this study and may not have provided the optimal scenario for investigating compensations. It is possible that the knee OA subjects prefer to keep their gait pattern consistent even under a weighted condition because of an innate control preference underlying a desire to decrease pain and increase function. Subjective measures of pain were not recorded in this study and could suggest possible motives for selected compensatory strategies.

Overall, the knee OA group did not respond the same way as the healthy adults to the weighted condition. The healthy adults responded to the challenging weighted walking condition by reducing load rate, and increasing initial double support percent and hip flexion angle at initial contact while the knee OA group only prolonged their initial double support percent. These significant findings lead us to believe that individuals with knee OA are performing close to capacity during unweighted walking and that only subtle compensatory strategies are employed when challenged.

Back to Article Outline

Acknowledgements 

We gratefully acknowledge the input of Katherine Rudolph, Kurt Manal and Stephen Messier to the thesis on which this manuscript is based. Additionally, we would like to acknowledge Joe Zeni, Jr. for his assistance during data collections. Funding provided by NIH P20-RR16458.

Conflict of interest

The authors have no conflict of interest in this study.

Back to Article Outline

References 

  1. Lawrence RC, Felson DT, Helmick CG, Arnold LM, Choi H, Deyo RA, et al. Estimates of the prevalence of arthritis and other rheumatic conditions in the United States – part II. Arthritis Rheum. 2008;58(1):26–35
  2. Gupta S, Hawker GA, Laporte A, Croxford R, Coyte PC. The economic burden of disabling hip and knee osteoarthritis from the perspective of individuals living with this condition. Rheumatology. 2005;44:1531–1537
  3. Holman HR, Lorig KR. Overcoming barriers to successful aging: self-management of osteoarthritis. West J Med. 1997;167(4):265–268
  4. Creamer P, Lethbridge-Cejku M, Hochberg MC. Factors associated with functional impairment in symptomatic knee osteoarthritis. Rheumatology. 2000;39(5):490–496
  5. Jinks C, Jordan K, Croft P. Osteoarthritis as a public health problem: the impact of developing knee pain on physical function in adults living in the community: (KNEST 3). Rheumatology. 2007;46(5):877–881
  6. Davis MA, Ettinger WH, Neuhaus JM, Mallon KP. Knee osteoarthritis and physical functioning – evidence from the NHANES-I epidemiologic follow-up study. J Rheumatol. 1991;18(4):591–598
  7. Andriacchi TP, Mündermann A. The role of ambulatory mechanics in the initiation and progression of knee osteoarthritis. Curr Opin Rheumatol. 2006;18(5):514–518
  8. Jackson BD, Wluka AE, Teichtahl AJ, Morris ME, Cicuttini FM. Reviewing knee osteoarthritis – a biomechanical perspective. J Sci Med Sports. 2004;7(3):347–357
  9. Zhao D, Banks SA, Mitchell KH, D’Lima DD, Colwell CW, Fregly BJ. Correlation between the knee adduction torque and medial contact force for a variety of gait patterns. J Orthop Res. 2007;25(6):789–797
  10. Hurwitz DE, Sumner DR, Andriacchi TP, Sugar DA. Dynamic knee loads during gait predict proximal tibial bone distribution. J Biomech. 1998;31(5):423–430
  11. Zeni JA, Higginson JS. Differences in gait parameters between healthy subjects and persons with moderate and severe knee osteoarthritis: a result of altered walking speed?. Clin Biomech. 2009;24:372–378
  12. Landry SC, McKean KA, Hublery-Kozey CL, Stanish WD, Deluzio KJ. Knee biomechanics of moderate OA patients measured during gait at a self-selected and fast walking speed. J Biomech. 2007;40:1754–1761
  13. Asay JL, Mündermann A, Andriacchi TP. Adaptive patterns of movement during stair climbing in patients with knee osteoarthritis. J Orthop Res. 2009;325–329
  14. Chen HL, Lu TW, Wang TM, Huang SC. Biomechanical strategies for successful obstacle crossing with the trailing limb in older adults with medial compartment knee osteoarthritis. J Biomech. 2008;41:753–761
  15. Tilbury-Davis DC, Hooper RH. The kinetic and kinematic effects of increasing load carriage upon the lower extremity. Hum Mov Sci. 1999;18:693–700
  16. Kinoshita H. Effects of different loads and carrying systems on selected biomechanical parameters describing walking gait. Ergonomics. 1985;28(9):1347–1362
  17. Puthoff ML, Darter BJ, Nielsen DH, Yack HJ. The effect of weight vest walking on metabolic responses and ground reaction forces. Med Sci Sports Exerc. 2006;38(4):746–752
  18. Salem GJ, Wang MY, Azen SP, Young JT, Greendale GA. Lower extremity kinetic response to activity program dosing in older adults. J App Biomech. 2001;17:103–112
  19. Maly MR. Abnormal and cumulative loading in knee osteoarthritis. Curr Opin Rheumatol. 2008;20(5):547–552
  20. Wada M, Maezawa Y, Baba H, Shimada S, Sasaki S, Nose Y. Relationships among bone mineral densities, static alignment and dynamic load in patients with medial compartment knee osteoarthritis. Rheumatology. 2001;40(5):499–505
  21. Zeni JA, Higginson JS. Dynamic knee joint stiffness in subjects with a progressive increase in severity of knee osteoarthritis. Clin Biomech. 2009;24:366–371
  22. Childs JD, Sparto PJ, Fitzgerald GK, Bizzini M, Irrgang JJ. Alterations in lower extremity movement and muscle activation patterns in individuals with knee osteoarthritis. Clin Biomech. 2004;19:44–49
  23. Gök H, Ergin S, Yavuzer G. Kinetic and kinematic characteristics of gait in patients with medial knee arthrosis. Acta Orthop Scand. 2002;73(6):647–652
  24. Kellgren J, Lawrence J. Radiologic assessment of osteoarthritis. Ann Rheum Dis. 1957;16:494–501
  25. Roos EM, Roos HP, Lohmander LS, Ekdahl C, Beynnon BD. Knee injury and osteoarthritis outcome score (KOOS) – development of a self-administered outcome measure. J Orthop Sports Phys Ther. 1998;28(2):88–96
  26. Zeni JA, Higginson JS. Gait parameters and stride-to-stride variability during familiarization to walking on a split-belt treadmill. Clin Biomech. 2010;25(4):383–386
  27. Rudolph KS, Schmitt LC, Lewek MD. Age-related changes in strength, joint laxity, and walking patterns: are they related to knee osteoarthritis?. Phys Ther. 2007;87(11):1422–1432
  28. Mündermann A, Dyrby CO, Hurwitz DE, Sharma L, Andriacchi TP. Potential strategies to reduce medial compartment loading in patients with knee osteoarthritis of varying severity – reduced walking speed. Arthritis Rheum. 2004;50(4):1172–1178
  29. Qu X, Yeo JC. Effects of load carriage and fatigue on gait characteristics. J Biomech. 2011;44(7):1259–1263
  30. Chamberlin ME, Fulwider BD, Sanders SL, Medeiros JM. Does fear of falling influence spatial and temporal gait parameters in elderly persons beyond changes associated with normal aging. J Gerontol. 2005;60A(9):1163–1167
  31. Zeni JA, Rudolph K, Higginson JS. Alterations in quadriceps and hamstrings coordination in persons with medial compartment knee osteoarthritis. J Electromyogr Kinesiol. 2010;20(1):148–154

PII: S0966-6362(11)00235-9

doi:10.1016/j.gaitpost.2011.07.012

Gait & Posture
Volume 35, Issue 1 , Pages 6-10, January 2012

Article Tools

* Image Usage & Resolution