Clinical Biomechanics

Transcription

1 Clinical Biomechanics 26 (2011) Contents lists available at ScienceDirect Clinical Biomechanics journal homepage: Progressive resistance training and dynamic alignment in osteoarthritis: A single-blind randomised controlled trial Nasim Foroughi a,, Richard M. Smith a, Angela K. Lange a, Maria A. Fiatarone Singh a,b,c, Benedicte Vanwanseele a a Discipline of Exercise, Health, and Performance, Faculty of Health Sciences, University of Sydney, Sydney, NSW, Australia b Sydney Medical School, University of Sydney, Sydney, NSW, Australia c Hebrew SeniorLife and Jean Mayer USDA Human Nutrition Center on Aging at Tufts University, Boston, MA, USA article info abstract Article history: Received 19 May 2010 Accepted 31 August 2010 Keywords: Osteoarthritis Strength training Dynamic alignment Shank adduction angle Background: We hypothesised that high intensity progressive resistance training would improve lower limb dynamic alignment and function (lower knee adduction moment, increased muscle strength, and fewer knee osteoarthritis symptoms). Methods: Women (n=54) with osteoarthritis in at least one knee were randomised into a 6-month resistance training or a sham-exercise program. The primary outcomes were dynamic shank and knee adduction angles and knee adduction moment of the most symptomatic knee measured with quantitative gait analysis. Secondary outcomes were muscle strength, gait speed, and osteoarthritis symptoms. Findings: Dynamic alignment and knee adduction moment did not change over time or between groups. Muscle strength improved in both groups over time, but significantly more in the resistance training group (P=0.002). By contrast, gait velocity and pain improved over time (P 0.009) in both groups. Improvements in shank adduction angle were related to improvements in self-reported disability (r =0.381, P=0.015), but not to changes in muscle strength, gait velocity, or pain (all PN0.05). Interpretations: Although muscle strength improved significantly more in the PRT group, the hypothesised reduction in knee adduction moment, shank and knee adduction angles were not evident after either exercise modality. However, improvements in disability and shank adduction angle were significantly directly related. Initial alignment should be used to stratify this population into separate groups when designing future trials and alternative modes of training investigated to potentially enhance beneficial alterations in knee alignment Elsevier Ltd. All rights reserved. 1. Introduction Lower limb malalignment is a risk factor for osteoarthritis (OA) progression (Cicuttini et al., 2004; Sharma et al., 2001). Unlike static alignment measured using radiographs, dynamic alignment reflects the knee movement during locomotion providing useful information about the limb during the gait cycle rather than a single instance in time. Measuring dynamic alignment via gait analysis has been suggested (Hunt et al., 2008) to address the limitations of static radiographs (e.g. limited accessibility to clinicians, unnecessary exposure to radiation, financial costs) (Hinman et al., 2006). In addition, foot position (Hunt et al., 2006) and weight-bearing status Grant support: University of Sydney: Cumberland Grant Category B, January 2005; R & D Grant (S4201 U3301), January Corresponding author. Exercise, Health and Performance Faculty Research Group, Faculty of Health Sciences, University of Sydney, 75 East St, Lidcombe NSW 1825 Australia. address: nasim.foroughi@sydney.edu.au (N. Foroughi). during radiography affects measurement of alignment (Specogna et al., 2007) making standardisation difficult to achieve due to differences in facilities. Dynamic varus alignment in healthy knees is minimal; however, shank adduction angle is significantly greater in patients with medial knee OA compared to healthy controls (Foroughi et al., 2009). The dependent variables describing dynamic alignment are frontal plane shank (to provide the absolute orientation of the shank in space), thigh, and knee adduction angles. These variables show the limb alignment and its absolute position in the laboratory frame of reference. More pronounced varus knee is associated with impaired functional ability and insufficient use of muscle strength during the gait loading phase. The impact of muscle weakness on functional ability is greater in subjects with high varus valgus range of motion than those with lower ranges (Van Der Esch et al., 2008). Muscle forces and in vivo knee moments can be accurately estimated and predicted via a dynamic neuromusculoskeletal model developed recently (Lloyd et al., 2003). This model has been used to /$ see front matter 2010 Elsevier Ltd. All rights reserved. doi: /j.clinbiomech

2 72 N. Foroughi et al. / Clinical Biomechanics 26 (2011) study forces in healthy joints during gait (Buchanan et al., 2005), to explain how the muscles work to stabilize the knee, and to determine the importance of muscles in joint stabilization (Lloyd et al., 2005). The quadriceps has the potential to support the frontal plane moments and is responsible for both static and dynamic stabilization in healthy knees (Lloyd et al., 2005). The knee ligaments are at greater risk of failure due to imbalanced muscular support and external loads. Quadriceps weakness and malalignment are risk factors for OA (Chang et al., 2004; Slemenda et al., 1997, 1998) and may be readily improved by resistance training (Lange et al., 2008). The benefits of exercise has been demonstrated for muscle strength, walking speed, disability (Alexander et al., 2001) and knee adduction moment (KAM) (Thorstensson et al., 2007), which is an indirect measure of the loads in medial compartment (Hurwitz et al., 1998; Jackson et al., 2004; Thorp et al., 2006; Wada et al., 2001; Zhao et al., 2007). Strength training may provide a wider range of motor control strategies for the knee such as improved balance (Orr et al., 2008), and biomechanics, which could potentially improve mobility and functional ability. More information is needed to describe the mechanics of the knee during walking and the mechanisms that allowed the reduction in pain and improvement in function shown previously in knee OA (Lange et al., 2008; O'reilly et al., 1999; Topp et al., 2002; Van Baar et al., 1998). The effects of these exercise programs on dynamic alignment have not been demonstrated. Therefore, we tested the following hypotheses: 1. A supervised progressive resistance training (PRT) program for older women with knee OA would lead to better symptomatic limb dynamic alignment and reduced knee adduction moment than sham exercise (minimal load, non progressive exercise). An improvement in alignment would be indicated by a reduction in the shank and knee adduction angles. 2. Improvements in dynamic alignment would be directly related to: Increases in muscle strength, Increases in gait speed, and Reductions in self-reported OA symptoms (measured by Western Ontario and McMaster Index (WOMAC, Likert VA3). 2. Methods 2.1. Design and setting An in-depth rationale for the design of the current study has been published previously (Lange et al., 2009). This study was a supervised, randomised, sham-exercise controlled trial approved by the Human Research Ethics Committee of the University of Sydney (13/12/2004, No /2/7848). The questionnaire data were double-blinded, the gait and physical performance testing were single-blinded (participant only). The subjects granted informed consent and were recruited from April 2005 to December 2006 via advertisements in local newspapers and word-of-mouth Eligibility criteria Women over the age of 40 in stable health with OA in at least one knee according to the American College of Rheumatology criteria (Altman, 1987) were eligible. Men were not included in this study due to gender difference in gait variables (Ferber et al., 2003). The exclusionary criteria were defined as having secondary OA (i.e. traumatic or post-surgical OA), other rare forms of arthritis, joint injury, injection, joint replacement or surgery within the past six months, participation in any structured exercise regimen more than once a week (past three months), disorders of the nervous system disrupting voluntary movement, severe functional limitation, cognitive impairment, and any contradictions to Magnetic Resonance Imaging (MRI) or resistance training Medical screening and MRI A telephone screening questionnaire was used to verify OA diagnosis and symptoms, other medical history, medications, contraindications to MRI, and habitual exercise participation. Upon completion, the study physician assessed the questionnaires to ascertain whether or not the participant met the eligibility criteria. If eligible, the study physician medically screened the participants, which included a history of arthritis onset, co-morbidities, previous trauma or surgery, sporting injuries, falls, pattern of pain, locking/ giving way, disability and a complete medical examination. The most severely affected knee was defined based on the clinical signs and symptoms and a 25-minute 3-Tesla MRI scan (Philips Medical Systems, Achieva 3 T, Netherlands) that was performed on that knee at baseline and after intervention. Severity of OA was graded based on Outerbridge Classification (Outerbridge, 1961) Exercise intervention Participants randomised into the experimental group performed high intensity PRT exercises at 80% of their peak muscle strength using Keiser machines. The exercises included unilateral knee extension, standing hip abduction and adduction; and bilateral knee flexion, leg press, and plantar-flexion. All exercises were performed for three sets of eight repetitions (6 9 s/repetition) with s rest between repetitions and 1 2 min rest between sets. One repetition maximum tests (1RM) were performed once every two weeks and a new 80% resistance was prescribed; in between strength tests, participants were prescribed 3% increments in resistance per session as tolerated. An intensity rating of on the Borg Rating of Perceived Exertion (RPE) scale (Borg, 1970) was considered optimal and was used to adjust the resistance between 1RM measurements to assure the intended continuous progressive overload. Exercises were modified daily according to participants' symptoms. Full range of motion was utilized unless limited by pain. In some participants, severe pain throughout the range of motion required substitution of isometric exercises of particular exercises intermittently during training period. The control group trained on the same equipment as the intervention group except without hip adduction, and performed knee extension bilaterally. These alterations were made to minimise recruitment of the adductors and quadriceps muscles and any stabilizing effects which might ensue even at low intensity. Minimal resistance was set on the machine with no progression. Exercise volume was reduced to two sets of eight repetitions, with same speed as in the PRT group. This regimen results in no significant improvements in muscle function (De Vos et al., 2005). Both groups trained three times per week for six months under the supervision of an exercise physiologist. The missed sessions due to illness were added onto the end of the intervention. Up to one month extension was allowed to complete the 78 sessions. Compliance was calculated as the percentage attended out of 78 sessions available Adverse events A weekly questionnaire was used to monitor adverse events and changes in health status. Adverse events were defined a priori as any musculoskeletal or cardiovascular event attributable to testing or training (e.g., inflammatory response in knee joint, cartilage/ligament/muscle tear, fracture, fall, angina, etc) (Loannidis et al., 2004). Pain during a training session that was self-limited and not considered consistent with an injury was not considered an adverse event but may have resulted in an adjustment of training protocol to accommodate limitations. Protocol deviations or adjustments occurred for both groups. The main deviation included changing from a dynamic to an isometric form of training (maximal intensity for PRT

3 N. Foroughi et al. / Clinical Biomechanics 26 (2011) and sub-maximal for controls) if the dynamic mode was causing pain, reducing the intensity for the PRT group, and/or limiting the range of motion Blinding All participants were blinded to the investigators' hypothesis. Physician screening was completed initially, followed by MRI. If participants were eligible following their MRI scan, the remainder of the baseline testing was completed. Participants were randomised at the completion of all baseline assessments so that all baseline testing was double-blinded and post-testing of muscle performance was single-blinded. Questionnaire data were double-blinded at both time points as it was self-reported. Both groups were trained at the same geographical location by the same trainers. At the follow-up assessment, participants were required to fill-out a Completion Questionnaire without the assistance of the assessor to assess the participants' perception of whether they felt they had been in the preferred group to increase muscle mass and strength, reduce pain, and improve physical function. This questionnaire enabled investigators to assess the success of participant blinding Assessments Primary outcomes gait testing and analysis procedure All measurements were performed on the most affected knee based on clinical signs and symptoms. A three-dimensional motion measurement system (Motion Analysis Corporation, USA) including 10 Eagle video cameras, the Expert Vision Advanced (EvaRT5.0) software package (Expert Vision Advanced Instruction Manual, 2000) and 2 sets of eight channel force plates (Kistler, Switzerland) were used to collect the gait data. The camera's shutter speed was 1/1000 of a second with the sampling rate of 100 frames per second (Hz) and the force data were collected at the sampling frequency of 1000 Hz. Thirty eight markers were attached bilaterally using a modified Helen Hayes marker set (Kadaba et al., 1990) on body segments. Each segment was defined using three markers and idealised as a rigid body with a local coordinate system defined to coincide with a set of anatomic axes. Subjects walked barefoot along a 10-meter walkway 5 times at their self-selected habitual speed with 2 3 min rest between the trials to minimise fatigue. A static trial was collected as a reference to determine body mass and positions of joint centers of rotation. The average values from five trials at both velocities were calculated for each subject permitting comparison of average values for each subject in the PRT or sham group. Gait velocity for each trial was calculated as the average velocity of the sacrum marker in the walking direction during two full strides and was averaged over the five trials. Segment angles relative to the laboratory and relative joint angles were calculated using joint coordinate systems (Wu and Cavanagh, 1995) at 30% stance phase where KAM was at its maximum value. Threedimensional internal moments were calculated using inverse dynamics via Kintrak 6 (University of Calgary, Canada) and were normalised to individual's percentage body weight and height (%BWx Ht) Primary outcomes The primary outcome was dynamic alignment and knee adduction moment of the most symptomatic leg measured during habitual gait before and after the intervention. We hypothesised that the dynamic alignment would change in a way that relieved the stress on the affected side of the knee joint. The dependent variables that describe the dynamic alignment would be shank and knee adduction angles. Together these variables show the alignment of the symptomatic limb and its absolute position in the laboratory frame of reference. In addition, we tested the hypothesis that PRT would significantly change the gait characteristics at the knee and clinical symptoms, evidenced by reduced peak KAM (at 30% stance phase) and improved WOMAC relative to the sham group Secondary outcomes muscle strength Digital Keiser pneumatic machines (K400, Keiser Sports Health Equipment, Fresno, CA, USA) were used to test the peak muscle strength of the lower limbs. Using established protocols (De Vos et al., 2005), 1RM tests were performed for all exercises twice at baseline (the higher of two results was used in analyses), and once at followup. Whole body total tonnage for the entire 6-month exercise period was calculated by summating the total summated loads of each exercises of each session and multiplying that by the total number of repetitions of all exercises across the 6 months Questionnaires Osteoarthritis-related symptoms and disability were assessed using an interviewer-administered WOMAC (Bellamy et al., 1988). It has been shown to be sensitive to change, with a change of 15 on the scale considered to be clinically meaningful improvement on this tool after knee replacement surgery (Escobar et al., 2007) Sample size We aimed to detect a 30% reduction in shank adduction angle of the PRT group compared to the sham-exercise group. Based on data from our previous study (Foroughi et al., 2009) a reduction of 30% corresponds with an absolute change of 1.5 degree shank adduction angle, with a SD estimated to be 0.8%. Based on these assumptions, a sample size of 44 patients was required with alpha set at 0.05, beta at 0.2, and a given power of 0.8. This was inflated to 54 to account for anticipated drop out of 10% Adherence Dropout was defined as failing to complete the intervention or one of the assessments. Discontinued intervention was defined as failing to complete the intervention, but completing all assessments. These discontinued cases were included in the complete case analysis, but dropouts were not, and missing data were not imputed Randomisation Following the baseline assessments, subjects were randomised to one of the two groups. A blinded co-investigator managed the concealed randomisation procedure. Subjects were stratified according to glucosamine and/or chondroitin use (current or within the past six months) and Physical Function (Section C; Disability) WOMAC score (b or N27) in order to equalise these potential confounders between groups Statistical analysis Data were inspected for normality, and expressed as mean and standard deviation or median and range. Non-normally distributed data were log-transformed prior to use with parametric statistics or used with non-parametric tests if assumptions of normality were not met despite transformation. A complete case primary analysis was performed due the novelty of our primary outcome. Comparisons between groups were made using repeated measures analysis of covariance (ANCOVA) for both time and group by time interactions. Covariates were included in the ANCOVA model if they were significantly different between the groups at baseline or if the covariate was significantly related to the dependent variable of interest. The Kruskal Wallis test or Mann Whitney U test were used

4 74 N. Foroughi et al. / Clinical Biomechanics 26 (2011) for non-normally distributed continuous data. Simple linear regression models were used to determine significance of potential relationships between variables of interest. All two-tailed p values of less than 0.05 were considered statistically significant (PASW 18.0, 2010, Chicago). 3. Results 3.1. Participants flow Fig. 1 and Table 1 present the flow and the baseline characteristics of participants. A total of 298 people were assessed for eligibility, 244 (82%) were excluded, and 63 (21%) people were eligible and interested. From the 63 eligible and interested people, 54 (86%) were enrolled and randomised into the PRT (n=26) and sham groups (n=28) which is comparable to a recent study with similar methods (Lim et al., 2008) Adherence The majority of subjects in both PRT (94%) and sham groups (73%) felt that they were in the preferred intervention group (P=0.563). Out of 78 exercise sessions in six months, the PRT and sham group attended an average of 67 (20) and 72 (21), (P=0.294). The percentage attendance (excluding dropouts) was 85 (25) % for the PRT and 92 (15) % for the sham group over the six-month (P=0.354) Numbers analysed The five participants who were lost to follow-up tended to have higher BMI (P=0.059) and had significantly greater fat-free mass (P=0.031) and percent body fat (P=0.008) at baseline than those who completed the intervention and/or the final assessments, but were similar in all other respects. Two out of 5 dropouts were related to the intervention (one in each group; due to knee and/or hip pain). The collected data for 6 subjects could not be retrieved due to Excluded (n=244) Medical (n=55) OA not observed by MRI (n=4) Not able to commit (n=185) Allocated to exercise group (n=26) Received allocated intervention (n=26) Did not receive allocated intervention (n=0) Lost to follow up (n=2) Discontinued intervention (n=3) OA-related (n=1) Medical (n=2) Exercise group analyzed (n=18) Excluded from analysis (n=6) Lost to follow up (n=2) Assessed for eligibility (n=298) Randomized (n=54) Completed trial for analysis (n=37) Allocated to Sham-exercise group (n=28) Received allocated intervention (n=27) Did not receive allocated intervention Family commitments (n=1) Lost to follow up (n=3) Discontinued intervention (n=1) OA-related (n=0) Medical (n=1) Sham-exercise group analyzed (n=19) Excluded from analysis (n=6) Lost to follow up (n=3) Fig. 1. Flow diagram of enrolment, allocation, follow-up, and analysis stages. Table 1 Baseline characteristics of the cohort. Characteristics technical problems. Subjects with laterally affected OA (PRT=2, Sham=4) were excluded from the statistical analysis. The primary analysis was performed on 37 complete cases Adverse events There were two minor adverse events in the sham group during testing which did not require any time-off from the training intervention. Knee pain was reported in 37.9% and 32.7% of intervention weeks by both sham and PRT groups. However, the difference in pain reporting between the groups was not significant (P=0.50) Primary outcomes Contrary to our hypothesis, the changes in knee and shank adduction angles, and knee adduction moment were not significantly different over time and did not improve more in the PRT group than sham-exercise group (P= , Table 2) Secondary outcomes Sham (n=28) PRT (n=26) Total (n=54) Mean (SD) Mean (SD) Mean (SD) Age (years) 64 (8) 64 (7) 64 (7) Mass (kg) 86.5 (23) 82.0 (14) 84.3 (18.9) Height (m) 1.6 (0.06) 1.6 (0.06) 1.6 (0.1) BMI (kg/m 2 )* 33.2 (8.1) 31.9 (5.2) 32.6 (6.9) Duration of OA (years) 10.8 (7.6) 9.5 (6.3) 10.2 (7) WOMAC Pain (range 0 20) 6.2 (3.1) 5.9 (3.5) 6 (3.3) Stiffness (range 0 8) 3.9 (1.8) 3.7 (1.3) 3.8 (1.6) Disability (range 0 68) 24.6 (10.7) 21.6 (11.4) 23.2 (11) Total (range 0 96) 34.7 (14.5) 31.2 (15.5) 33 (15) Knee extension 1RM (N-m) 38.2 (18.6) 38.9 (22.2) 38.5 (20.3) No. of chronic diseases (n) 1.8 (1.5) 1.3 (1.1) 1.5 (1.3) Gait velocity (m/s) 1.09 (0.19) 1.13 (0.15) 1.1(0.18) OA affected compartment (%) (%) Medial Lateral OA grade (Range1 4) 0.35 Mild Moderate Severe Very severe OA, Osteoarthritis; BMI, Body Mass Index; *, Adapted from World Health Organization (WHO) 1995, WHO 2000 and WHO 2004; WOMAC, Western Ontario and McMaster Osteoarthritis Index; lower score equals to less symptoms (Bellamy et al., 1988); Nm, Newton meters; 1RM, one repetition maximum;, MRI correlation of Modified Outerbridge Classification (Outerbridge, 1961); m/s, meters/seconds. Pain, stiffness, disability and total WOMAC score improved in both groups over time (P= ), but the interaction effect was not significant. The knee extension strength improved in both groups over time (P=0.001) with significantly greater absolute gains in the PRT group (P=0.002). The overall relative change in strength was 33.0 (68.3) % in the sham and 52.5 (59.6) % in the PRT group (P=0.082). The changes in alignment variables were not related to changes in total tonnage or strength (PN0.05). The changes in shank adduction angle were significantly and directly related to the changes in WOMAC disability in the whole cohort (Fig. 2), consistent with our hypotheses that reduced (improved) shank adduction angle would be associated with symptomatic improvement (r=0.381, P=0.015). Body mass and BMI did not change significantly over time or in either group following the intervention (Table 2). Gait speed increased significantly in both groups over time (P=0.001) with a P

5 N. Foroughi et al. / Clinical Biomechanics 26 (2011) Table 2 Characteristics of the sham and PRT group pre and post intervention. Gait characteristics Sham (n=19) PRT (n=18) P (T) P (T*G) Mean (SD) Mean (SD) Pre Post Pre Post Knee adduction moment 2.72 (0.96) 2.75 (0.90) 2.54 (1.41) 2.69 (1.30) Shank adduction angle 2.47 (8.26) 4.58 (3.95) 3.49 (8.52) 5.41 (2.95) Knee adduction angle 3.49 (5.32) 2.67 (4.78) 1.87 (6.90) 0.14 (7.58) Health and function Mass(kg) 84.9 (25.8) 80.9 (12.6) 85.0 (24.9) 79.7 (12.5) BMI(kg/m 2 )* 32.8 (9.4) 32.8 (9.0) 31.4 (5.4) 30.9 (5.3) WOMAC Pain (Range 0 20) 6.7 (3.5) 5.5 (3.6) 5.7 (3.3) 3.83 (2.7) Stiffness (Range 0 8) 3.9 (2.0) 3.2 (1.9) 3.5 (1.5) 2.6 (1.8) Disability (Range 0 68) 25.1 (11.5) 20.6 (12.9) 20.6 (11.4) (10.6) Total (Range 0 96) 35.3 (15.0) 29.3 (17.5) 29.9 (15.7) 19.5 (13.9) Knee extension 1RM (N-m) 35.9 (12.1) 44.2 (16.4) 42.8 (24.4) 60.2 (34.9) Velocity (m/s) 1.1 (0.19) 1.2 (0.17) 1.1 (0.17) 1.2 (0.17) Moments in (%BWxHt); angles in degrees G, Group; T, Time, knee adduction moment direction=positive; knee Adduction angle direction=negative; BMI, Body Mass Index; *, Adapted from World Health Organization (WHO) 1995, WHO 2000 and WHO 2004; WOMAC, Western Ontario and McMaster Osteoarthritis Index, lower score equals to less symptoms (Bellamy et al., 1988); Nm, Newton meters; 1RM, one repetition maximum; s, seconds; m, meters; G, Group; T, Time. percentage change of 8.6 (10) % in total cohort; however there was no interaction with group assignment. 4. Discussion Our hypothesis that the dynamic alignment of the most symptomatic leg would adapt to PRT so as to relieve stress on the affected side of the knee joint was not supported by our results. That is, the changes in knee adduction moment, knee and shank adduction angles were not different following the intervention and the group by time interaction was not significant. Although no significant changes were observed in alignment and joint moments, muscle strength improved in the PRT group, just as it did in overall cohort. In a malaligned knee, such increased force-generating capacity could potentially increase the compression forces on the articular cartilage due to altered line of action of muscle force (Marks, 1993; Sharma et al., 2003). Increased medial compartment loading is associated with excessive varus malalignment whereas loading in the lateral compartment contributes to valgus malalignment (Mckellop et al., 1994). Previous literature is controversial in this regard. Lim et al. (2008) have shown in a cross-sectional study that greater varus alignment was associated with higher quadriceps strength. For example, each degree of greater varus alignment was associated with 3% higher in muscle strength (Lim et al., 2008). In a prospective study, higher quadriceps strength at baseline predicted greater OA progression in Fig. 2. Changes in WOMAC difficulty sub score versus changes in shank adduction angle of the total cohort (r=0.381, P=0.015). more varus malaligned knees (Sharma et al., 2003). However in another cross-sectional study, higher quadriceps strength was not associated with greater peak KAM (Lim et al., 2009). Thus, more studies are needed as clinical care often involves the prescription of muscle strengthening exercises in this cohort. Quadriceps muscles have the potential to support and stabilize the knee joint during both static and dynamic activities (Lloyd et al., 2005) and to reduce the varus thrust during gait which is a potential risk factor for OA progression (Chang et al., 2004). Following resistance training, the motor control system may learn to relieve the stress from the symptomatic side by stabilizing the knee; however, this mechanism may only apply to the more neutral knees, rather than knees with more malalignment. Varus alignment and varus thrust are correlated with KAM (Chang et al., 2004; Wada et al., 2001) and malalignment increases with disease severity (Mundermann et al., 2005). Gentle exercises in this group of subjects may reduce the pain and disability, and improve the physical function by producing a more controlled and aligned joint. Studies directly comparing prescriptive elements (modality, intensity, and volume of exercise) are critically needed to advance this field. We observed, for example, that both low and high intensity muscle training produced similar symptomatic improvements, despite less strength gain in the lower intensity group, and no significant biomechanical benefits in either group. Ultimately, large, long term prospective studies are required to identify whether improved joint alignment can be achieved with exercise, and can decrease the rate of OA progression. The peak KAM of the total cohort did not change significantly following our intervention, although it tended to decrease, preferentially in the PRT group ( 9% PRT vs. 13% Sham). The lack of significance in this study could be due to insufficient power, as the relative effect size was only 0.17, which would have required 545 subjects per group to reach significance, casting doubt on the meaningfulness of such small changes. Additionally, our results are in agreement with a recent study of larger cohort in which no significant difference was found in KAM after the intervention (Lim et al., 2008). Given the contradictory findings of several studies about the association of KAM with knee alignment, muscle strength, pain and progression of knee OA, the clinical importance of alterations in KAM requires further clarification. The cause of pain and loss of function in this disease is likely to be more complex than previous analyses have suggested. Pain reduction following analgesic medication treatment in OA has been reported to increase the KAM and therefore potentially increase

6 76 N. Foroughi et al. / Clinical Biomechanics 26 (2011) the risk of disease progression (Hurwitz et al., 2000). By contrast, we showed that strength training (both low and high intensity) is an effective intervention to improve pain in knee OA without worsening the abnormal loading on the knee joint. It is possible that analgesia without strength training allows individuals to load knees which are inherently unstable, by eliminating the warning signal of knee pain, and worsening joint forces. A non-exercise comparison group in our study may have identified this pattern more clearly. The intensity of exercises for both groups may not have been separated enough to differentiate between the loadings achieved by study groups. This may explain the lack of a group difference in KAM between the study groups. The sham-exercise group trained unloaded and had no progression in terms of resistance on the machines. However, some of the subjects in the sham-exercise group found it challenging to move the padded bar against the gravity, due to pain, or poor muscular condition at the beginning of the project. In the PRT group however, the subjects who had very severe OA may not have been able to progress the load as prescribed at each session due to high pain levels. In addition, it might have been too late for some of subjects with severe OA to start exercises because most of the cartilage had degenerated. Future studies should design a specific exercise program for the sham-exercise group by which the joint receives no loading. Prospective studies have shown that increases in KAM increase the risk of chronic knee pain and radiographic progression (by 6-fold) 3 4 and 6 years later (Amin et al., 2004; Miyazaki et al., 2002). Reductions of 14, 13, and 8% in KAM have been reported during different tasks (e.g. one leg rise and gait) following resistance exercise (Thorstensson et al., 2007), valgus bracing (Pollo et al., 2002) and wedged insoles (Kerrigan et al., 2002). In contrast, other studies which used valgus bracing found no reduction in KAM following their intervention (Hewett et al., 1998; Kerrigan et al., 2002), and resistance training has thus far not been found to significantly alter KAM during gait in this or prior studies (Thorstensson et al., 2007; King et al., 2008). More studies are required to identify how different types and dosages of exercise affect the KAM during gait in this cohort. Furthermore, identifying the clinical importance of such a reduction in KAM should be a priority for future investigations. Acknowledgement The authors' acknowledgement goes to volunteers and The University of Sydney R & D Grant (S4201 U3301). References Alexander, M.J.L., Butcher, J.E., Macdonald, P.B., Effect of a water exercise program on walking gait, flexibility, strength, self-reported disability and other psychosocial measures of older individuals with arthritis. Physiother. Can. 53, Altman, R.D., Criteria for the classification of osteoarthritis of the knee and hip. Scand. J. Rheumatol. 65, Amin, S., Luepongsak, N., Mcgibbon, C.A., Lavalley, M.P., Krebs, D.E., Felson, D.T., Knee adduction moment and development of chronic knee pain in elders. Arth Rheum. 51, Bellamy, N., Buchanan, W., Goldsmith, C.H., Campbell, J., Stitt, L.W., Validation study of WOMAC: a health status instrument for measuring clinically important patient relevant outcomes to antirheumatic drug therapy in patients with osteoarthritis of the hip or the knee. J. Rheumatol. 15, Borg, G., Perceived exertion as an indicator of somatic stress. Scand. J. Rehab Med. 2, Buchanan, T.S., Lloyd, D.G., Manal, K., Besier, T.F., Buchanan, T.S., Lloyd, D.G., et al., Estimation of muscle forces and joint moments using a forward-inverse dynamics model. Med. Sci. Sports Exerc. 37, Chang, A., Hayes, K., Dunlop, D., Hurwitz, D., Song, J., Cahue, S., et al., Thrust during ambulation and the progression of knee osteoarthritis. Arth Rheum. 50, Cicuttini, F., Wluka, A., Hankin, J., Wang, Y., Longitudinal study of the relationship between knee angle and tibiofemoral cartilage volume in subjects with knee osteoarthritis. J. Rheumatol. 43, De Vos, N.J., Singh, N.A., Ross, D.A., Stavrinos, T.M., Orr, R., Fiatarone Singh, M.A., Optimal load for increasing muscle power during explosive resistance training in older adults. J. Gerontol. A Biol. Sci. Med. Sci. 60, Escobar, A., Quintana, J., Bilbao, A., Aróstegui, I., Lafuente, I., Vidaurreta, I., Responsiveness and clinically important differences for the WOMAC and SF-36 after total knee replacement. Osteoarthritis Cartilage 15, Expert Vision Advanced Instruction Manual, Motion Analysis Corporation. PN Copyright Santa Rosa, California, USA. Ferber, R., Davis, I.M., Williams, D.S., Gender differences in lower extremity mechanics during running. Clin. Biomech. 18, Foroughi, N., Smith, R.M., Lange, A.K., Baker, M.K., Fiatarone Singh, M.A., Vanwanseele, B., Dynamic alignment and its association with knee adduction moment in medial knee osteoarthritis. Knee doi: /j.knee Hewett, T.E., Noyes, F.R., Barber-Westin, S.D., Heckmann, T.P., Decrease in knee joint pain and increase in function in patients with medial compartment arthrosis: a prospective analysis of valgus bracing. Orthopedics 21, Hinman, R.S., May, R.L., Crossley, K.M., Is there an alternative to the full-leg radiograph for determining knee joint alignment in osteoarthritis? Arthritis Rheum. 55, Hunt, M.A., Birmingham, T.B., Giffin, J.R., Jenkyn, T.R., Associations among knee adduction moment, frontal plane ground reaction force, and lever arm during walking in patients with knee osteoarthritis. J. Biomech. 39, Hunt, M.A., Birmingham, T.B., Bryant, D., Jones, I.C., Giffin, J.R., Jenkyn, T.R., et al., Lateral trunk lean explains variation in dynamic knee joint load in patients with medial compartment knee osteoarthritis. Osteoarthritis Cartilage 16, Hurwitz, D.E., Sumner, D.R., Andriacchi, T.P., Sugar, D.A., Dynamic knee loads during gait predict proximal tibial bone distribution. J. Biomech. 31, Hurwitz, D.E., Ryals, A.R., Block, J.A., Sharma, L., Schnitzer, T.J., Andriacchi, T.P., Knee pain and joint loading in subjects with osteoarthritis of the knee. J. Orthop. Res. 18, Jackson, B.D., Teichtahl, A.J., Morris, M.E., Wluka, A.E., Davis, S.R., Cicuttini, F.M., The effect of the knee adduction moment on tibial cartilage volume and bone size in healthy women. Rheum 43, Kadaba, M.P., Ramakrishnan, H.K., Wootten, M.E., Measurement of lower extremity kinematics during level walking. J. Orthop. Res. 8, Kerrigan, D.C., Lelas, J.L., Goggins, J., Merriman, G.J., Kaplan, R.J., Felson, D.T., Effectiveness of a lateral-wedge insole on knee varus torque in patients with knee osteoarthritis. Arch. Phys. Med. Rehabil. 83, King, L.K., Birmingham, T.B., Kean, C.O., Jones, I.C., Bryant, D.M., Giffin, J.R., Resistance training for medial compartment knee osteoarthritis and malalignment. Med. Sci. Sports Exerc. 40, Lange, A.K., Vanwanseele, B., Fiatarone Singh, M.A., Progressive resistance training for treatment of osteoarthritis of the knee: a systematic review. Arthritis Rheum. 29, Lange, A.K., Vanwanseele, B., Foroughi, N., Baker, M.K., Shnier, R., Smith, R.M., et al., Resistive exercise for arthritic cartilage health (REACH): a randomized double-blind, sham-exercise controlled trial. BMC Geriatr. 9, 1. Lim, B.W., Hinman, R.S., Wrigley, T.V., Sharma, L., Bennell, K.L., Does knee malalignment mediate the effects of quadriceps strengthening on knee adduction moment, pain, and function in medial knee osteoarthritis? A randomized controlled trial. Arth & Rheum (Arthritis Care & Res). 59, Lim, B.W., Kemp, G.J., Metcalf, B., Wrigley, T.V., Bennell, K.L., Crossley, K.M., et al., The association of quadriceps strength with the knee adduction moment in medial knee osteoarthritis. Arth & Rheum (Arthritis Care & Res). 61, Lloyd, D.G., Besier, T.F., Lloyd, D.G., Besier, T.F., An EMG-driven musculoskeletal model to estimate muscle forces and knee joint moments in vivo. J. Biomech. 36, Lloyd, D.G., Buchanan, T.S., Besier, T.F., Lloyd, D.G., Buchanan, T.S., Besier, T.F., Neuromuscular biomechanical modeling to understand knee ligament loading. Med. Sci. Sports Exerc. 37, Loannidis, J.P.A., Evans, S.J.W., Gøtzsche, P.C., O'neill, R.T., Altman, D.G., Schulz, K., Better reporting of harms in randomized trials; An extension of the CONSORT statement. Ann. Intern. Med. 141, Marks, R., The effect of isometric quadriceps strength training in mid-range for osteo-arthritis of the knee. N. Z. J. Physiotherapy 21, Mckellop, H.A., Llinas, A., Sarmiento, A., Effects of tibial malalignment on the knee and ankle. Orthop. Clin. North Am. 25, Miyazaki, T., Wada, M., Kawahara, H., Sato, M., Baba, H., Shimada, S., Dynamic load at baseline can predict radiographic disease progression in medial compartment knee osteoarthritis. Ann. Rheum. Dis. 61, Mundermann, A., Dyrby, C.O., Andriacchi, T.P., Secondary gait changes in patients with medial compartment knee osteoarthritis: increased load at the ankle, knee, and hip during walking. Arth Rheum. 52, O'reilly, S.C., Muir, K.R., Doherty, M., Effectiveness of home exercise on pain and disability from osteoarthritis of the knee: a randomised controlled trial. Ann. Rheum. Dis. 58, Orr, R., Raymond, J., Fiatarone Singh, M.A., Efficacy of progressive resistance training on balance performance in older adults. Sports Med. 38, Outerbridge, R.E., The etiology of chondromalacia patella. J. Bone Joint Surg. 43B, Pollo, F.E., Otis, J.C., Backus, S.I., Warren, R.F., Wickiewicz, T.L., Reduction of medial compartment loads with valgus bracing of the osteoarthritic knee. Am. J. Sports Med. 30, Sharma, L., Song, J., Felson, D.T., Cahue, S., Shamiyeh, E., Dunlop, D.D., The role of knee alignment in disease progression and functional decline in knee osteoarthritis. JAMA. 286, Sharma, L., Dunlop, D.D., Cahue, S., Song, J., Hayes, K.W., Quadriceps strength and osteoarthritis progression in malaligned and lax knees. Ann. Intern. Med. 138,

7 N. Foroughi et al. / Clinical Biomechanics 26 (2011) Slemenda, C., Brandt, K.D., Heilman, D.K., Mazzuca, S., Braunstein, E.M., Katz, B.P., et al., Quadriceps weakness and osteoarthritis of the knee. Ann. Intern. Med. 127, Slemenda, C., Heilman, D.K., Brandt, K.D., Katz, B.P., Mazzuca, S.A., Braunstein, E.M., et al., Reduced quadriceps strength relative to body weight: a risk factor for knee osteoarthritis in women? Arth Rheum. 41, Specogna, A.V., Birmingham, T.B., Hunt, M.A., Jones, I.C., Jenkyn, T.R., Fowler, P.J., et al., Radiographic measures of knee alignment in patients with varus gonarthrosis: effect of weightbearing status and associations with dynamic joint load. Am. J. Sports Med. 35, Thorp, L.E., Wimmer, M.A., Block, J.A., Moisio, K.C., Shott, S., Goker, B., et al., Bone mineral density in the proximal tibia varies as a function of static alignment and knee adduction angular momentum in individuals with medial knee osteoarthritis. Bone 39, Thorstensson, C.A., Henriksson, M., Von Porat, A., Sjodahl, C., Roos, E.M., The effect of eight weeks of exercise on knee adduction moment in early knee osteoarthritis a pilot study. Osteoarth Cartilage 15, Topp, R., Woolley, S., Hornyak, W., Al, E., The effect of dynamic versus isometric resistance training on pain and functioning among adults with osteoarthritis of the knee. Arch. Phys. Med. Rehabil. 83, Van Baar, M.E., Dekker, J., Oostendorp, R.A., Bijl, D., Voorn, T.B., Lemmens, J.A., et al., The effectiveness of exercise therapy in patients with osteoarthritis of the hip or knee: a randomized clinical trial. J. Rheumatol. 25, Van Der Esch, M., Steultjens, M., Harlaar, J., Wolterbeek, N., Knol, D., Dekker, J., Varus valgus motion and functional ability in patients with knee osteoarthritis. Ann. Rheum. Dis. 67, 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. Rheum 40, Wu, G., Cavanagh, P.R., ISB recomendations for standardization in the reporting of kinematic data. J. Biomech. 28, Zhao, D., Banks, S.A., Mitchell, K.H., D'lima, D.D., Colwell, C.W.J.R., Fregly, B.J., Correlation between the knee adduction torque and medial contact force for a variety of gait patterns. J. Orthop. Res. 25,