Journal Logo


Neuromuscular Exercise post Partial Medial Meniscectomy

Randomized Controlled Trial


Author Information
Medicine & Science in Sports & Exercise: August 2015 - Volume 47 - Issue 8 - p 1557-1566
doi: 10.1249/MSS.0000000000000596


Meniscal injury is a potent risk factor for knee osteoarthritis (21), and the medial meniscus is more commonly damaged compared with the lateral meniscus (9). Despite evidence questioning the efficacy of surgical treatment for meniscal tears (17,20), arthroscopic partial meniscectomy (APM) is a routine and commonly used surgical procedure. Estimates suggest that approximately 50% of people develop radiographic knee osteoarthritis within 10–20 yr from meniscal surgery (24). Middle-age people with degenerative meniscal tears have greater risk for development of symptomatic and radiographic knee osteoarthritis than younger individuals with traumatic meniscal tears (10). Therefore, middle-age APM patients are a subgroup who could benefit more from targeted strategies to prevent or delay knee osteoarthritis than younger individuals with traumatic tears.

Many authors agree that mechanical factors contribute to symptomatic and structural knee osteoarthritis (1,12,33). Knee osteoarthritis after medial APM is thought to relate to higher medial knee joint load. Using three-dimensional gait analysis, mediolateral knee joint load distribution is estimated using the knee external adduction moment (KAM). Peak KAM during walking seems higher in people after APM than that in controls (16,35) and increases over time after surgery (16). Other data show a relationship between higher KAM parameters and increased risk of knee pain onset (1) and structural disease progression in people with established knee osteoarthritis (3,25). Thus, KAM reduction could provide a plausible mechanism to prevent or delay the onset or progression of knee osteoarthritis after medial APM.

Exercise is an attractive option that targets neuromuscular features (e.g., coordination and control of the knee and surrounding muscles), and in particular could potentially reduce higher KAM during functional tasks. Neuromuscular exercises are commonly performed in weight-bearing functional postures. Importance is placed on features such as movement control and quality plus alignment of the lower limb and trunk. The focus on neutral alignment of the knee is the feature most likely to modify KAM. Results from pilot studies support the potential efficacy of neuromuscular exercise to reduce KAM in people with knee osteoarthritis during gait and a one-leg sit-to-stand rise (37,38). However, despite these pilot studies, other randomized controlled trials have found no change in KAM during gait after exercise programs in people with established knee osteoarthritis (4,13). Although gait is arguably the most clinically relevant task to reduce KAM, it is also of interest to evaluate the effect of exercise on more challenging tasks such as one-leg sit-to-stand or one-leg hop. Determining the effect of neuromuscular exercise on a spectrum of tasks with varying difficulty will provide better understanding of the potential of exercise to alter KAM, a key predictor of structural change, albeit during gait. Furthermore, a randomized controlled trial in people after APM reported improved physical function (11) and cartilage quality assessed using delayed gadolinium-enhanced magnetic resonance imaging (dGEMRIC) (29) after 4 months of neuromuscular exercise training. To date, no study has rigorously evaluated the effects of neuromuscular exercise on the KAM during tasks of varying difficulty in people who are at high risk of developing or progressing structural knee osteoarthritis.

The primary objective of this randomized controlled trial was to evaluate the effect of a 12-wk neuromuscular exercise program (similar to that aimed to ensure lower limb alignment (4) and shown to improve cartilage quality (29) and physical function (11)) during walking and one-leg sit-to-stand in people after a medial APM. The primary hypothesis was that peak KAM during normal-paced walking and one-leg sit-to-stand would reduce in a neuromuscular exercise group when compared with that in a nonintervention control group. This study investigated individuals with or without early signs of knee osteoarthritis, as the purpose of this intervention was to prevent or delay the onset or progression of structural disease in this population. Furthermore, individuals with or without minimal pain were included to exclude the potential role of pain reduction as a mediator of change in the primary outcome.



Sixty-two volunteers age 30–50 yr with an isolated medial APM within the past 3–12 months were recruited from May 2012 to July 2013, with follow-up completed in November 2013. All participants were identified by screening the surgical records of eight orthopedic surgeons, except for one participant who responded to a university advertisement. The institutional ethics committee approved the study, and participants provided a written informed consent. The main exclusion criteria included the following: (i) an average overall pain severity greater than three out of 10 in the past week on a numeric rating scale, (ii) moderate or severe radiographic osteoarthritis, defined as Kellgren–Lawrence grade 3 or 4 (22), (iii) lower limb surgery (other than one knee arthroscopy), (iv) complete anterior or posterior cruciate ligaments tears, (v) body mass index above 36 kg·m−2 (to reduce excessive soft tissue movement and minimize artifact in marker data acquisition), (vi) other forms of arthritis, diabetes, and cardiac circulatory conditions that limit everyday activities (32).


This was an assessor-blinded, randomized controlled trial. The detailed protocol for this study has been published (15) and summarized here. Potentially eligible individuals received an information letter from their orthopedic surgeon. Two weeks later, these individuals underwent telephone screening and were invited to participate. If eligible and interested, posteroanterior weight-bearing knee radiographs were obtained to confirm eligibility. Randomization was in permuted blocks of six or eight, stratified by gender, to either the exercise group (“ALIGN” program, a rehabilitation program that aims to ensure neutral alignment of the lower limb) or control group. An independent researcher prepared the randomization schedule using a computer-generated random number table. Group allocations were sealed in opaque, consecutively numbered envelopes by the same independent researcher and stored in a central locked location. The envelopes were opened in sequence by a different independent researcher not involved in recruitment or assessment of participants who subsequently disclosed the group allocation to the participant. Participants were not blinded to group allocation but were unaware of the study hypothesis.


Eligibility to deliver the intervention required that physiotherapists had experience in musculoskeletal physiotherapy and work in private practice. Seven physiotherapists (five male and two female physiotherapists) based in seven private practices around metropolitan Melbourne delivered the ALIGN program. They had an average of 8 yr (SD, 4.3; range, 3–20 yr) of clinical musculoskeletal experience. Three (43%) held postgraduate masters-level qualifications. Physiotherapists received a detailed treatment manual and attended a 3-h training session before trial involvement. Physiotherapists received payment for delivering the intervention.

Participants visited the physiotherapist of their choice eight times over the 12-wk period, as follows: twice in week 1, once in week 2, and fortnightly thereafter. The first session lasted 45 min, and the remaining seven sessions lasted approximately 30 min. The focus during the first session was to introduce the program and thereafter to progress the program as appropriate. Participants were asked to perform home exercises on both legs at least three times per week over 12 wk (15).

Neuromuscular exercises

The ALIGN exercise program is described in detail elsewhere (15). Participants performed six exercises with the aim of maintaining neutral mediolateral alignment of the lower limb while engaging the trunk muscles during functional exercises (Table 1) (see Appendix, Supplemental Digital Content 1, Details of the exercises used in the ALIGN neuromuscular exercise program, Exercises were selected on the basis of previous reports that these exercises or similar variations had improved cartilage quality and physical function in people after APM and reduced peak KAM in people with knee osteoarthritis (11,29,38). The participant and physiotherapist guided progression of exercise. Participants were encouraged to progress as they felt able to increase the volume of exercises performed, e.g., aiming to begin with two sets of 12 repetitions, progressing to two sets of 15 repetitions, three sets of 12 repetitions, and three sets of 15 repetitions. Progression of exercise by the physiotherapist aimed to increase the exercise intensity, which was achieved by the following: holding a medicine ball while performing the exercises, increasing the load by filling the medicine ball with water, increasing the duration of the hold phase of the exercise, changing the direction, and/or changing the supporting surface.

Summary of the neuromuscular exercise program “ALIGN” (see Appendix, Supplemental Digital Content 1, Details of the exercises used in the ALIGN neuromuscular exercise program,

Outcome Measures

The same blinded assessor performed measures at baseline (week 0) and follow-up (week 13) in the Movement Laboratory at the Centre for Health Exercise and Sports Medicine, University of Melbourne, Australia.

Three-dimensional movement analysis

Participants underwent three-dimensional movement analysis during the following three tasks: (i) walking (normal and fast pace), (ii) one-leg sit-to-stand, and (iii) one-leg hop for maximum distance, all wearing standardized footwear (Dunlop Volley, Pacific Brands, Australia). In each task, the external KAM and external knee flexion moment (KFM) were measured using a 12-camera motion analysis system (Vicon MX, Oxford, United Kingdom) and three force plates (AMTI Inc., Watertown, MA) embedded in the walkway. The peak KFM was assessed on the basis of in vivo evidence suggesting that medial tibiofemoral contact force may not reduce despite a reduction in peak KAM if the peak KFM increases simultaneously (39). A seven-segment model was constructed using Vicon Body Builder according to previously described methods (5). After the application of 33 reflective markers, participants completed a series of calibration trials that incorporated functional approaches to define hip centers and the knee joint flexion/extension helical axes according to Besier et al. (5). Marker trajectories and ground reaction forces (GRF) were low-pass-filtered at 6 Hz for walking (23) and one-leg sit-to-stand and at 12 Hz for one-leg hop using a dual-pass Butterworth filter. Inverse dynamics programmed in the Vicon Body Builder (5) was used to calculate net external moments in the shank coordinate frame. Test–retest reliability (coefficient of multiple determination, r2) of knee adduction–abduction moment and knee flexion–extension moment curves has been reported to be ≥0.75 (5). KAM and positive KAM impulse (positive area under the KAM–time curve) and KFM were normalized by dividing by body weight (N) times body height (m) and expressed as a percentage.

Participants performed two walking conditions: a self-selected normal-paced walk described as “a pace you would walk normally” and a self-selected fast-paced walk described as “a pace you would walk when in a hurry.” Two photoelectric beams measured speed. If the self-selected walking speed at follow-up was more than ±5% of baseline walking speed, participants were asked to adjust their walking speed accordingly, such that six trials were captured at a speed within ±5% of baseline speed (matched walking speed to baseline). The following three measures were obtained and averaged over six trials: peak KAM during the first half of stance, because it is typically the larger of two peaks (3), KAM impulse during the stance phase of walking, given its association with structural disease progression (3), and peak KFM throughout stance, considering its association with increased medial tibiofemoral joint contact force (39).

For the one-leg sit-to-stand task, participants sat in a standardized position at baseline and follow-up (15). Using one leg only, participants were instructed to rise up from the chair and return to seated position. The non-weight bearing leg was held slightly flexed in front of the body. Data collection for this task was defined as the period from when the GRF reached 100 N to the point when the GRF dropped below 100 N. The overall peak KAM, peak KFM, and KAM impulse during the rise-and-sit-down phase of this task were measured and averaged over three trials.

For the one-leg hop for distance task, participants were instructed to hop forward as far as possible and land steadily and the hop distance was measured (15). The peak KAM and peak KFM were calculated for the longest hop.

Self-reported pain and physical function

Self-reported pain during the past week was assessed on an 11-point numeric rating scale with terminal anchors of “no pain” (score, 0) and “worst pain possible” (score, 10) (2). Pain, other symptoms, activities of daily living function, sport and recreation, and knee-related quality of life were also assessed using the valid and reliable Knee Injury Osteoarthritis Outcome Score (KOOS) (30). At follow-up, participants were asked to rate their (i) overall change, (ii) change in pain, and (iii) change in physical function (compared with those at baseline) on a seven-point ordinal scale (1, much worse; 7, much better). Ratings were dichotomized such that “moderately better” or “much better” were considered to be improved.

Muscle strength

Maximal isometric and isokinetic concentric knee muscle strength were assessed using an isokinetic dynamometer (Humac NORM; CSMi, Lawrence, MA). For isometric quadriceps and hamstring testing, participants were securely seated with 60° knee flexion and the peak of three maximal contractions was recorded. Isokinetic quadriceps and hamstring strength testing was performed at 60° ·s−1, and the participant performed five maximal efforts. The peak of five trials was recorded. Isometric hip adductor and abductor muscle strength was assessed using a handheld dynamometer, with the participant lying supine (Lafayette Manual Muscle Test System 01163; Lafayette Instrument, Lafayette, Indiana) (28). The peak force measurements generated during two maximal efforts against the manual resistance were averaged as the measure of hip muscle strength (28). For all strength tests, participants received strong standardized encouragement. Values are reported as torque normalized to body mass (N·m·kg−1).

Physical performance tasks

Tests of physical performance included the following: (i) maximum number of controlled one-leg rises in 30 s (8), (ii) maximum number of knee bends in 30 s (8), and (iii) the one-leg hop for distance test with the farthest distance jumped over at least three maximal hops, with arms across the chest, recorded (15).

Other measures

The Kellgren–Lawrence grading system was used to determine radiographic osteoarthritis severity (22). Demographic information was collected at baseline. Adverse effects and cointerventions were assessed from participant logbooks and physiotherapist treatment notes. Adherence to the neuromuscular exercise program was determined by the number of home exercise sessions recorded as completed by participants in their logbooks. Home exercise adherence was expressed as the number of completed (and recorded) exercise sessions as a percentage of the 36 prescribed sessions (3 times per week for 12 wk). After the completion of the 12-wk program, physiotherapists also recorded their perceived impression of the participant’s overall adherence to the ALIGN program on an 11-point scale (0, “not at all”; 10, “completely as instructed”) (7). A global indication of physical activity was assessed using the Lower Extremity Activity Scale (36). A clinical measure of static frontal plane alignment of the knee in standing was measured at baseline using an inclinometer (18).

Sample Size

The primary end points were peak KAM during the stance phase of normal-paced walking and a one-leg sit-to-stand task. This study was powered to detect a between-group difference in change in peak KAM of 10%, equating to an approximate reduction of 0.20 N·m/(body weight (BW) × height (HT))% for peak KAM with an anticipated SD of the change score of 0.3 N·m/(BW × HT)% in the ALIGN group and no change in the control group. Although the minimal clinically important difference to be detected for a change in KAM indices remains unknown, we speculated that a 10% reduction could plausibly be associated with a significant reduction in risk of disease progression (25). On this basis, a sample of 27 participants per group was required for a two-tailed comparison of the groups using ANCOVA, adjusting for baseline values as covariates, with 80% power and an alpha level of 0.05. To allow for 15% dropout, 31 participants were recruited per group.

Statistical Analysis

Main comparative analyses between groups were performed blinded to group allocation using an intention-to-treat approach. Two-tailed significance was set at P < 0.05. For continuous outcome measures, differences in mean change (follow-up minus baseline) were compared between groups using ANCOVA, adjusting for baseline scores of the outcome variable. For normal- and fast-paced walking, analysis was performed for walking speeds that were unmatched and matched to baseline walking (within ±5% of baseline walking speed). Model diagnostic checks used residual plots. Results are presented as estimated differences with 95% confidence intervals (CI). The Pearson chi tests were used to compare medication use and cointerventions between the groups. Log binomial regression was used to compare likelihood of improvements overall and in function and pain.


Outcome measures

Of the 415 individuals identified as potentially eligible, 156 (38%) were ineligible, 152 (37%) were not interested in participating, and 34 (8%) could not be contacted. Sixty-two participants (31 ALIGN and 31 control participants) were randomized, and 60 (31 (100%) ALIGN and 29 (94%) control participants) completed the follow-up assessment (Fig. 1). Two participants in the control group withdrew from the study because of relocation overseas and excessive time commitments. At baseline, participant characteristics were similar between groups (Table 2). The cohort was predominantly male, had minimal radiographic osteoarthritis, and was slightly overweight. According to the KOOS subscales, the cohort had similar levels of function and symptoms to those of a population-based age-matched reference group (27).

Demographic and clinical characteristics of the ALIGN and control groups, described as mean (SD) or n (%) unless otherwise stated.
Flow diagram of study protocol.

There were no significant differences between the ALIGN and control groups for absolute changes in peak KAM during normal-paced gait within ±5% of baseline walking speed (mean difference, 0.22 (95% CI, −0.11 to 0.55) N·m/(BW × HT)%; P = 0.19) or peak KAM during one-leg sit-to-stand (−0.01 (95% CI, −0.33 to 0.31) N·m/(BW × HT)%; P = 0.95) (Table 3). Similarly, no between-group differences were observed for changes in peak KAM for walking speed unmatched to baseline (see Table, Supplementary Digital Content 2, Results of gait variables using unmatched walking speed to baseline assessment, Neither group showed significant within-group change in peak KAM during either normal-paced gait or one-leg sit-to-stand (Table 3).

Mean (SD) of groups, mean (SD) change within groups, and mean (95% CI) difference in change between groups, adjusted for baseline scores.

There were no between-group differences for changes in any of the secondary outcomes (Table 3). In the ALIGN group, significant improvements were observed in the number of knee bends and one-leg rises performed in 30 s, maximum one-leg hop, and knee-related quality of life (KOOS). In addition, in the ALIGN group, walking speed increased for normal-paced gait and peak KFM increased during a one-leg hop for distance. In the control group, significant improvements were found in the number of one-leg knee bends, number of one-leg rises performed in 30 s, and knee-related quality of life (KOOS) and walking speed increased for normal-paced gait.

Because not all participants reported pain or physical dysfunction before the trial, the seven-point scale of improvement in pain was only completed by 24 (77%) ALIGN participants and 27 (87%) control participants and the scale of physical improvement was only completed by 23 (74%) ALIGN participants and 26 (84%) control participants. Of those reporting problems at baseline, improvement in pain at follow-up was not different between the groups and was reported by 7/25 (28%) of ALIGN participants and 4/25 (16%) controls (relative risk (95% CI), 0.57 (0.09–1.71); P = 0.32). Participants in the ALIGN group were more likely to perceive an improvement in physical function, with improvement reported by 10/23 (43%) ALIGN participants compared with only 3/24 (13%) control participants (relative risk (95% CI), 0.29 (0.09–0.91); P = 0.04). Participants in the ALIGN group were more likely to report overall improvement (14/30 (47%) ALIGN participants and 5/29 (17%) control participants (relative risk (95% CI), 0.38 (0.16–0.92); P = 0.03)).

Adherence, adverse events, medication use and cointerventions

From a maximum of eight, the number of physiotherapy sessions attended by the ALIGN group ranged from 0 to 8 with a median (interquartile range (IQR)) of 8 (1). From a maximum of 36 required home exercise sessions, the ALIGN group completed a median (IQR) of 29 (10) sessions with a range of 0–34 among the 29/31 participants who returned the logbook. One participant attended no physiotherapy sessions, as the person withdrew from receiving the intervention because of increased knee pain for reasons unrelated to the study. One participant scored zero for adherence to home exercises because the participant did not begin the program. Physiotherapist-perceived impressions of the participants’ overall adherence to the ALIGN program ranged from 3 to 10, with a median (IQR) of 10 (2).

There were no major adverse events. Thirty-nine minor adverse events were reported by 16/31 (52%) ALIGN participants and predominately related to increased knee pain and back pain. Cointerventions and medication use during the trials were similar for the ALIGN and control groups (Table 4).

Medications and cointerventions reported as n.


The aim of this assessor-blinded randomized controlled trial was to evaluate the effects of a physiotherapist-guided, home-based neuromuscular exercise program on the peak KAM during walking and one-leg sit-to-stand in middle-age people with no to mild pain after medial APM. We found no evidence that the 12-wk neuromuscular exercise program investigated here alters the peak KAM in this group with no or minimal symptoms, who are at high risk of developing or progressing early knee osteoarthritis. Some between-group differences were observed in improvement for self-reported pain, physical function (in those with minor pain and physical dysfunction at baseline), and overall improvement, favoring those in the ALIGN group.

Peak KAM during normal-paced walking was selected as a primary outcome, given its association with progression of medial tibiofemoral osteoarthritis (25) and the relevance of medial joint loading to this medial APM group. We anticipated that KAM would reduce in the ALIGN group because the exercises were designed to encourage neutral frontal plane alignment of the lower limb and this could plausibly be expected to reduce one of KAM’s two primary determinants, the length of the knee–GRF lever arm (19). Contrary to our hypothesis, we found no difference in change in peak KAM during walking between the groups. This was despite excellent adherence to the ALIGN program, and scope to reduce peak KAM during gait (i.e., peak KAM was approximately 26% larger than that in healthy controls (16)). Failure to reduce KAM in this study might be explained by poor transfer of the trained skills to gait. No exercises in the ALIGN program resemble the heel–toe action of gait, and this degree of task specificity may be required to modify gait.

Although not statistically significant, the difference in change in peak KAM during normal- and fast-paced walking between the groups was approximately 7% (Table 2). The current study was powered to detect a 10% difference in change in peak KAM between the two groups. Thus, a larger sample size may have produced a statistically significant difference in support of our hypothesis. The nonsignificant difference might also reflect KAM measurement error (6). Importantly, the magnitude and direction of peak KAM change (i.e., approximately 5% decrease in ALIGN and 2% increase in controls) are consistent across both normal- and fast-paced walking. It is of interest to note that we observed no between-group change in peak KFM during gait, to alter medial tibiofemoral joint contact force (39).

Peak KAM during one-leg sit-to-stand was also included as a primary outcome, because an uncontrolled pilot study found a 14% reduction in peak KAM during this task after an 8-wk neuromuscular exercise program in people (n = 13) with early knee osteoarthritis (38). Despite the similarity of the current sample in terms of age and radiographic disease severity and using a similar neuromuscular training program, we did not find a reduction. There are three issues to consider in relation to this finding. First, unlike peak KAM during walking, it is unknown whether people after APM have a higher peak KAM during one-leg sit-to-stand than that of controls. Thus, it is unknown whether participants had scope to reduce peak KAM during this task. Second, KAM can be altered by foot progression angle (34), knee flexion, and hip rotation (14). The reduction in peak KAM observed in the pilot study could be explained by these aforementioned variations during testing, rather than the 8-wk neuromuscular exercise program. Third, the clinical relevance of reduced peak KAM during one-leg sit-to-stand remains unknown.

This is the first randomized controlled trial to evaluate the effects of neuromuscular exercise on KAM in a population with good function and minimal symptoms but were at risk for development or progression of knee osteoarthritis. Despite failure to find a difference between groups, this study makes an important contribution to knowledge. A recent study on the effect of neuromuscular exercise on KAM during walking in people with painful established knee osteoarthritis speculated that KAM may not alter because of the lack of task specificity (4) or that changes in peak KAM might only be revealed in more demanding tasks. Our study argues against these issues, as we found no evidence that neuromuscular training alters peak KAM during a demanding one-leg sit-to-stand task despite its inclusion in the ALIGN program. This finding questions the efficacy of this neuromuscular exercise program to reduce KAM during such functional tasks. Other forms of neuromuscular training that target different aspects of function may be effective, but this requires further study. It has also been speculated previously that training both the affected and unaffected leg may accentuate the effect of neuromuscular exercise on KAM (4). Although participants in the current study performed the ALIGN program on both limbs, peak KAM did not reduce in our study. Overall, our findings concur with and further substantiate evidence from clinical trials that report no effect of exercise, regardless of type, on KAM in people with knee osteoarthritis (4,13).

Although KAM measurement has excellent reliability (6), it also has limitations as a primary outcome. Peak KAM is widely used because it has been shown to relate to structural cartilage change (25) but is only a surrogate measure of the latter. Because we did not quantify structural change in this study, we cannot conclude that neuromuscular exercise does not delay or prevent osteoarthritic structural change in APM patients. Indeed, a supervised 16-wk neuromuscular exercise program found improved cartilage quality assessed using dGEMRIC in people 3–5 yr after medial APM (29). In that study, dGEMRIC was associated with osteoarthritic changes in the same cohort 11 yr later (26), which supports dGEMRIC as a clinically relevant indicator of adverse structural change.

The strengths of our study include rigorous study design with methodological features to minimize bias such as the following: concealed allocation, excellent participant retention, and good adherence to the ALIGN program. Several limitations of this study warrant consideration. First, only 15% of those identified as potentially eligible from private surgical records participated and this questions the potential generalizability of the results. The most frequent reason that people declined was lack of interest to participate. Although our participants had similar KOOS scores to that of other APM populations 3–5 yr after surgery, they reported less difficulty in performing activities associated with sport and recreation (15,29,31,35). Such observations may suggest better baseline neuromuscular control with less scope for improvement. Second, the lack of participant blinding requires consideration. Although a placebo intervention would provide a more optimal comparator, design of a credible placebo treatment for neuromuscular exercise is difficult. Furthermore, because the placebo effect on objective outcomes is minimal in knee osteoarthritis (40), we deemed a placebo treatment unnecessary, given the objective nature of our primary outcome (peak KAM). Third, we did not formally assess the adherence of the physiotherapists to the standardized ALIGN intervention protocol.

In conclusion, our results show no change in peak KAM during walking or one-leg sit-to-stand after neuromuscular exercise in an APM cohort, who reported good function and few symptoms. Future studies are required to assess long-term effects of neuromuscular exercise on structural measures of osteoarthritis onset and progression and the effect of more task-specific protocols before the disease-modifying effect of neuromuscular exercise can be conclusively determined.

The study physiotherapists providing the physiotherapy treatments were Ian McFarland, Riley Bodger, Laurie McCormack, John Pompei, Anthony Feron, Josh Heery, and Alison Harding. We thank the following orthopedic surgeons and their staff for assisting with participant recruitment: Mr. Tim Whitehead, Prof. Julian Feller, Mr. Rohan Price, Mr. Cameron Norsworthy, Mr. Robert Steele, Mr. Chris Kondogiannis, and Mr. Peter Gard. The authors thank Mr. Joel Martin and Ms. Penny Campbell for randomizing the participants and managing the home exercise equipment and Mr. Ben Metcalf for providing the randomization schedule. The authors also thank Ms. Janine Topp for her contribution to data acquisition by assisting with data collection and processing.

This study was funded by an Australian National Health and Medical Research Council Program grant (#631717). K. L. B. and R. S. H. were partly funded by Australian Research Council Research Future fellowships (#FT 0991413 and #FT 130100175), and P. H. is supported by a National Health and Medical Research Council fellowship (#APP1002190). M. H. was supported by a Ph.D. scholarship from a National Health and Medical Research Council Program grant (#631717). The study sponsor did not play any role in the study design, collection, and analysis or interpretation of data or in the writing of the article or decision to submit the manuscript for publication. None of the funders had any role in the study other than to provide funding.

The authors’ contributions were as follows: M. H., significant manuscript writer, concept and design, data acquisition, data analysis, and data interpretation; R. S. H., concept and design, significant manuscript reviewer, data interpretation, and statistical expertise; T. V. W., significant manuscript reviewer, concept and design, and data analysis and interpretation; E. M. R., significant manuscript reviewer, concept and design, and data interpretation; P. W. H., obtained funding, significant manuscript reviewer, concept and design, and data interpretation; M. P. S., significant manuscript reviewer, statistical analysis, and data interpretation; K. L. B., obtained project funding, significant manuscript reviewer, concept and design, and data interpretation. All authors provided feedback on the draft of this article and read and approved the final article.

No authors have a conflict of interest.

The results of the present study do not constitute endorsement by the American College of Sports Medicine.


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 (3): 371–6.
2. Bellamy N. Osteoarthritis clinical trials: candidate variables and clinimetric properties. J Rheumatol. 1997; 24 (4): 768–78.
3. Bennell KL, Bowles KA, Wang YY, Cicuttini F, Davies-Tuck M, Hinman RS. Higher dynamic medial knee load predicts greater cartilage loss over 12 months in medial knee osteoarthritis. Ann Rheum Dis. 2011; 70 (10): 1770–4.
4. Bennell KL, Kyriakides M, Metcalf B, et al. Neuromuscular versus quadriceps strengthening exercise in people with medial knee osteoarthritis and varus malalignment: a randomised controlled trial. Arthritis Rheumatol. 2014; 66 (4): 950–9.
5. Besier TF, Sturnieks DL, Alderson JA, Lloyd DG. Repeatability of gait data using a functional hip joint centre and a mean helical knee axis. J Biomech. 2003; 36 (8): 1159–68.
6. Birmingham TB, Hunt MA, Jones IC, Jenkyn TR, Giffin JR. Test-retest reliability of the peak knee adduction moment during walking in patients with medial compartment knee osteoarthritis. Arthritis Rheum. 2007; 57 (6): 1012–7.
7. Borkovec TD, Nau SD. Credibility of analogue therapy rationales. J Behav Ther Exp Psy. 1972; 3 (4): 257–60.
8. Bremander AB, Dahl LL, Roos EM. Validity and reliability of functional performance tests in meniscectomized patients with or without knee osteoarthritis. Scand J Med Sci Sports. 2007; 17 (2): 120–7.
9. Englund M, Guermazi A, Roemer FW, et al. Incidental meniscal findings on knee MRI in middle-aged and elderly persons. N Engl J Med. 2008; 359 (11): 1108–15.
10. Englund M, Roos EM, Lohmander LS. Impact of type of meniscal tear on radiographic and symptomatic knee osteoarthritis: a sixteen-year followup of meniscectomy with matched controls. Arthritis Rheum. 2003; 48 (8): 2178–87.
11. Ericsson YB, Dahlberg LE, Roos EM. Effects of functional exercise training on performance and muscle strength after meniscectomy: a randomized trial. Scand J Med Sci Sports. 2009; 19 (2): 156–65.
12. Felson DT. Osteoarthritis as a disease of mechanics. Osteoarthritis Cartilage. 2013; 21 (1): 10–5.
13. Foroughi N, Smith RM, Lange AK, Baker MK, Fiatarone Singh MA, Vanwanseele B. Lower limb muscle strengthening does not change frontal plane moments in women with knee osteoarthritis: a randomized controlled trial. Clin Biomech (Bristol, Avon). 2011; 26 (2): 167–74.
14. Fregly BJ, D’Lima DD, Colwell CW Jr. Effective gait patterns for offloading the medial compartment of the knee. J Orthop Res. 2009; 27 (8): 1016–21.
15. Hall M, Hinman RS, Wrigley TV, et al. The effects of neuromuscular exercise on medial knee joint load post-arthroscopic partial medial meniscectomy: ‘SCOPEX’ a randomised control trial protocol. BMC Musculoskel Disord. 2012; 13: 23.
16. Hall M, Wrigley TV, Metcalf BR, et al. A longitudinal study of strength and gait after arthroscopic partial meniscectomy. Med Sci Sports Exerc. 2013; 45 (11): 2036–43.
17. Herrlin S, Wange P, Lapidus G, Hållander M, Werner S, Weidenhielm L. Is arthroscopic surgery beneficial in treating non-traumatic, degenerative medial meniscal tears? A five year follow-up. Knee Surg Sports Traumatol Arthrosc. 2013; 21 (2): 358–64.
18. Hinman RS, May RL, Crossley KM. Is there an alternative to the full-leg radiograph for determining knee joint alignment in osteoarthritis? Arthritis Rheum. 2006; 55 (2): 306–13.
19. Hunt MA, Birmingham TB, Giffin JR, Jenkyn TR. Associations among knee adduction moment, frontal plane ground reaction force, and lever arm during walking in patients with knee osteoarthritis. J Biomech. 2006; 39 (12): 2213–20.
20. Katz JN, Brophy RH, Chaisson CE, et al. Surgery versus physical therapy for a meniscal tear and osteoarthritis. N Engl J Med. 2013; 368 (18): 1675–84.
21. Katz JN, Martin SD. Meniscus-friend or foe: epidemiologic observations and surgical implications. Arthritis Rheum. 2009; 60 (3): 633–5.
22. Kellgren JH, Lawrence JS. Radiological assessment of osteo-arthrosis. Ann Rheum Dis. 1957; 16 (4): 494–502.
23. Kirkely C. Clinical Gait Analysis: Theory and Practice. 1st ed. London (United Kingdom): Churchill Livingstone; 2006.
24. Lohmander LS, Englund PM, Dahl LL, Roos EM. The long-term consequence of anterior cruciate ligament and meniscus injuries: osteoarthritis. Am J Sports Med. 2007; 35 (10): 1756–69.
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. 2002; 61 (7): 617–22.
26. Owman H, Ericsson YB, Englund M, et al. Association between delayed gadolinium-enhanced MRI of cartilage (dGEMRIC) and joint space narrowing and osteophytes: a cohort study in patients with partial meniscectomy with 11 years of follow-up. Osteoarthritis Cartilage. 2014; 22 (10): 1537–41.
27. Paradowski PT, Bergman S, Sunden-Lundius A, Lohmander LS, Roos EM. Knee complaints vary with age and gender in the adult population. Population-based reference data for the Knee injury and osteoarthritis outcome score (KOOS). BMC Musculoskelet Disord. 2006; 7: 38.
28. Pua YH, Wrigley TW, Cowan SM, Bennell KL. Intrarater test-retest reliability of hip range of motion and hip muscle strength measurements in persons with hip osteoarthritis. Arch Phys Med Rehabil. 2008; 89 (6): 1146–54.
29. Roos EM, Dahlberg L. Positive effects of moderate exercise on glycosaminoglycan content in knee cartilage—a four-month, randomized controlled trial in patients at risk of osteoarthritis. Arthritis Rheum. 2005; 52 (11): 3507–14.
30. Roos EM, Roos HP, Ekdahl C, Lohmander LS. Knee injury and osteoarthritis outcome score (KOOS)—validation of a Swedish version. Scand J Med Sci Sports. 1998; 8 (6): 439–48.
31. Roos EM, Roos HP, Ryd L, Lohmander LS. Substantial disability 3 months after arthroscopic partial meniscectomy: a prospective study of patient-relevant outcomes. Arthroscopy. 2000; 16 (6): 619–26.
32. Sangha O, Stucki G, Liang MH, Fossel AH, Katz JN. The Self-Administered Comorbidity Questionnaire: a new method to assess comorbidity for clinical and health services research. Arthritis Rheum. 2003; 49 (2): 156–63.
33. Segal NA, Anderson DD, Iyer KS, et al. Baseline articular contact stress levels predict incident symptomatic knee osteoarthritis development in the MOST cohort. J Orthop Res. 2009; 27 (12): 1562–8.
34. Shull PB, Silder A, Shultz R, et al. Six-week gait retraining program reduces knee adduction moment, reduces pain, and improves function for individuals with medial compartment knee osteoarthritis. J Orthop Res. 2013; 31 (7): 1020–5.
35. Sturnieks DL, Besier TF, Hamer PW, et al. Knee strength and knee adduction moments following arthroscopic partial meniscectomy. Med Sci Sports Exerc. 2008; 40 (6): 991–7.
36. Terwee CB, Bouwmeester W, van Elsland SL, de Vet HCW, Dekker J. Instruments to assess physical activity in patients with osteoarthritis of the hip or knee: a systematic review of measurement properties. Osteoarthritis Cartilage. 2011; 19 (6): 620–33.
37. Thorp LE, Wimmer MA, Foucher KC, Sumner DR, Shakoor N, Block JA. The biomechanical effects of focused muscle training on medial knee loads in OA of the knee: a pilot, proof of concept study. J Musculoskelet Neuronal Interact. 2010; 10 (2): 166–73.
38. Thorstensson CA, Henriksson M, von Porat A, Sjodahl C, Roos EM. The effect of eight weeks of exercise on knee adduction moment in early knee osteoarthritis—a pilot study. Osteoarthritis Cartilage. 2007; 15 (10): 1163–70.
39. Walter JP, D’Lima DD, Colwell CW, Fregly BJ. Decreased knee adduction moment does not guarantee decreased medial contact force during gait. J Orthop Res. 2010; 28 (10): 1348–54.
40. Zhang W, Robertson J, Jones AC, Dieppe PA, Doherty M. The placebo effect and its determinants in osteoarthritis: meta-analysis of randomised controlled trials. Ann Rheum Dis. 2008; 67 (12): 1716–23.


Supplemental Digital Content

© 2015 American College of Sports Medicine