If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Are biomechanics during gait associated with the structural disease onset and progression of lower limb osteoarthritis? A systematic review and meta-analysis
Address correspondence and reprint requests to: N. D'Souza, Faculty of Medicine and Health, The University of Sydney, Susan Wakil Health Building, D 18 Western Avenue, Camperdown NSW 2006, Australia. Tel.: 61-2-8627-6970.
Graduate Programs in Rehabilitation Sciences, Faculty of Medicine, University of British Columbia, CanadaMotion Analysis and Biofeedback Laboratory, University of British Columbia, Canada
Motion Analysis and Biofeedback Laboratory, University of British Columbia, CanadaDepartment of Physical Therapy, Faculty of Medicine, University of British Columbia, Canada
To evaluate if gait biomechanics are associated with increased risk of structurally diagnosed disease onset or progression of lower limb osteoarthritis (OA).
Method
A systematic review of Medline and Embase was conducted from inception to July 2021. Two reviewers independently screened records, extracted data and assessed risk of bias. Included studies reported gait biomechanics at baseline, and either structural imaging or joint replacement occurrence in the lower limb at follow-up. The primary outcome was the Odds Ratio (OR) (95% confidence interval (CI)) of the association between biomechanics and structural OA outcomes with data pooled for meta-analysis.
Results
Twenty-three studies reporting 25 different biomechanical metrics and 11 OA imaging outcomes were included (quality scores ranged 12–20/21). Twenty studies investigated knee OA progression; three studies investigated knee OA onset. Two studies investigated hip OA progression. 91% of studies reported a significant association between at least one biomechanical variable and OA onset or progression. There was an association between frontal plane biomechanics with medial tibiofemoral and hip OA progression and sagittal plane biomechanics with patellofemoral OA progression. Meta-analyses demonstrated increased odds of medial tibiofemoral OA progression with greater baseline peak knee adduction moment (KAM) (OR: 1.88 [95%CI: 1.08, 3.29]) and varus thrust presence (OR: 1.97 [95%CI: 1.32, 2.96]).
Conclusion
Evidence suggests that certain gait biomechanics are associated with an increased odds of OA onset and progression in the knee, and progression in the hip.
Osteoarthritis (OA) is characterised by structural changes, such as bone marrow lesions (BMLs), osteophytes, cartilage loss and symptoms including pain and functional deficits
Global, regional, and national incidence, prevalence, and years lived with disability for 301 acute and chronic diseases and injuries in 188 countries, 1990–2013: a systematic analysis for the Global Burden of Disease Study 2013.
. To reduce this burden, understanding of modifiable risk factors of disease onset and/or progression is needed. A key emerging risk factor is theorised to be the loading environment of weight-bearing joints
In-vivo animal studies have shown a link between compressive joint loading and structural changes, which may contribute to increased OA onset and progression risk
. However, excessive compressive loading beyond a threshold may overwhelm the ability of the cartilage and subchondral bone to adapt, potentially causing tissue failure
. Excessive compressive or shear forces are therefore hypothesised to surpass the threshold of tissue capacity and can result in load being shifted to infrequently loaded joint areas
. For example, the external knee adduction moment (KAM) acts to rotate the tibia with respect to the femur in the frontal plane, and is widely considered a valid
proxy measure of medial to lateral knee joint load distribution during gait. Research alludes to an association between higher baseline KAM magnitudes during gait and greater odds of knee OA progression over time, with reported odds ratios (OR) of 1.3 [ 95%CI: 0.86, 1.98]
. Significant associations between gait biomechanical metrics at baseline and OA progression at follow-up have also been reported in people with hip OA. A recent study suggests that hip joint loading patterns, such as cumulative hip frontal plane moment, may be associated with hip OA progression
. Differences in gait biomechanics have also been identified between people with ankle OA and healthy controls, possibly alluding to increased risk of OA changes.
A comprehensive systematic evaluation of gait biomechanics at baseline and the association with subsequent OA onset and progression in lower limb weight-bearing joints is needed due to the potentially modifiable nature of gait. One systematic review, published in 2014
Is there a causal link between knee loading and knee osteoarthritis progression? A systematic review and meta-analysis of cohort studies and randomised trials.
, aimed to study if the KAM biomechanical metric is associated with higher risk of knee OA progression. The authors did not find a significant association between higher KAM and OA progression, potentially due to the small number of studies (k = 4) and large variability in findings. Since 2014, additional studies have been conducted examining the relationship between KAM variables and knee OA progression
Is there a causal link between knee loading and knee osteoarthritis progression? A systematic review and meta-analysis of cohort studies and randomised trials.
was knee OA progression and one biomechanical metric, and did not provide data on OA outcomes at other lower limb joints or evaluate OA onset. This research is important as biomechanical metrics hypothesised to be associated with loading
Given the emerging research, a systematic and comprehensive risk evaluation is required for the hip, knee, and ankle joints. This review aims to determine and quantify if gait biomechanical metrics are associated with the onset or progression of structural OA changes in the major lower limb joints, defined by imaging-based changes or joint replacement occurrence.
Methods
Data sources and searches
We searched two electronic databases, Medline and Embase from inception to 12/07/2021. Key terms included “Osteoarthritis”, “Biomechanics/gait”, “Knee/Hip/Ankle”, “Disease onset/progression” and terms referring to OA structural changes (Appendix A). The search was restricted to humans, with no restriction on language, age, race, or geographical location. A bibliographic and citation search of included studies was conducted. This review was registered prospectively (Prospero CRD42019133920).
Inclusion/exclusion criteria
Studies
Studies were eligible if they were longitudinal and reported OA structural outcomes assessed through either imaging (e.g., radiographs or MRI: Magnetic Resonance Imaging) changes between baseline and follow-up or joint replacement surgery occurrence after baseline. Dissertations, conference proceedings, abstract, case studies, intervention studies with no observation/control group, and reviews (systematic or narrative) were excluded. Studies with participants who had history of knee/hip replacement prior to baseline, inflammatory arthritis, or neurological conditions affecting gait at baseline were excluded.
Exposure and outcome
All included studies reported gait biomechanics including kinetic, kinematic, and spatiotemporal metrics. The following terminology will be used throughout this review: we defined “biomechanical metrics” as umbrella kinetic, kinematic, or spatiotemporal measurements in the frontal, sagittal and transverse planes (e.g., KAM and hip flexion angle). We used the term “biomechanical variables” to define these metrics at specific time-points in the gait cycle (e.g., early stance peak KAM, or midstance hip flexion angle). Lower limb biomechanical metrics derived from marker-based motion capture, two-dimensional (2-d) videography, or pressure-sensing mats were extracted. Studies exclusively investigating static alignment or non-gait biomechanical metrics were excluded. Studies defined disease onset as any OA structural changes detected on imaging via semi-quantitative and quantitative scales from imaging-defined healthy joints at baseline. Studies defined lower limb OA progression as worsening of OA features on imaging, or the occurrence of joint replacement surgery. Occurrence of joint replacement surgery was chosen as an acceptable OA progression outcome as structural OA worsening on imaging over time has shown to predict the need for joint replacement surgery, inclusive of joint space narrowing (JSN) seen on radiographs
Three year joint space narrowing predicts long term incidence of knee surgery in patients with osteoarthritis: an eight year prospective follow up study.
All records were initially screened by title and abstract. Full-texts of relevant records were obtained and screened to determine eligibility. Two reviewers independently screened, and a third independent reviewer was available to resolve disagreements.
Risk of bias
Risk of bias was evaluated independently by two reviewers (ND, JC) with disagreements resolved by a third reviewer (JG). Quality was assessed using the risk of bias tool reported in Chapple et al.
, specifically designed for the assessment of OA prognostic studies (Appendix B). The tool contains 20 items divided into 4 subscales: study participation, study attrition, measurement and data presentation, and analysis and presentation of results. Each item was scored 0 (poor quality) or 1 (good quality), with a total score range of 0–21. We modified item N to have two separate points: one for blinding of the assessors and one for the use of standardised procedures for reading imaging or joint replacement occurrence. In reference to study attrition, studies were scored as “unable to determine” if they did not report the initial baseline sample of participants and only included participants who had follow-up data.
Quality of evidence
Two reviewers (ND, JC) independently assessed the quality of evidence using the Grading of Recommendations Assessment, Development and Evaluation (GRADE) for prognostic factors
GRADE Guidelines 28: use of GRADE for the assessment of evidence about prognostic factors: rating certainty in identification of groups of patients with different absolute risks.
. Four items: study limitations, inconsistency, indirectness, imprecision, and publication bias were evaluated to grade the overall quality of evidence for meta-analyses with ≥3 studies. Disagreements were resolved by a third reviewer (MS).
Data extraction and synthesis
Using a standardised template, two reviewers (ND, JC) independently extracted the following data: affected joint, baseline condition (healthy or OA), recruitment, participant characteristics, inclusion/exclusion, biomechanical assessment method, OA onset and progression definition, study design and attrition rate. The primary value extracted was the Odds Ratio (OR) (95%CI: 95% confidence interval) of the association between biomechanical variables and OA onset/progression. If this was unavailable, baseline biomechanical variables were extracted, which included mean (standard deviation (SD)) for the progression and non-progression subgroups or the reported regression coefficients. If a study only provided baseline biomechanical variables through graphs, data were extracted through the Webplot digitiser
. In this instance, baseline variables were entered into Comprehensive Meta-Analysis software which converted the values into an OR point estimate (95% CI).
Meta-analyses were performed for all biomechanical variables which could be grouped according to the consistent joint, plane of motion, direction/torque, and time/phase of the gait cycle. Analyses were conducted on all available outcomes of disease progression for the same biomechanical variable. If the data from the same cohort was presented in two studies, meta-analysis was conducted using the study reporting either the largest sample size and/or reported adjusted odds ratios. Heterogeneity was determined by evaluating the similarity in study methods (including imaging outcomes, follow-up times, and cohort similarities) as well as the I2 statistic (<40% suggesting low heterogeneity)
. Given that included studies in the meta-analyses had increased heterogeneity due to varying follow-up periods, I2 >40% and sample sizes, we used a random effects model for all analyses. Sensitivity analyses were conducted separately for outcomes by modes of imaging (radiographic and MRI), then grouped further for the specific outcome of disease progression (e.g.: joint space narrowing (JSN), BMLs). For meta-analyses where <3 studies were available, results are reported in Appendix C. Reported biomechanical metrics not included in a meta-analysis were presented in table format and reported during the narrative synthesis. Overall, a narrative summary of the number of studies reporting biomechanical metrics and the number of studies with positive, negative or no associations to OA pathology are provided for each joint.
Results
Our search yielded 2,914 independent records, of which 48 were retained for full-text screening, and 23 studies met the eligibility criteria for final inclusion (Fig. 1). Of the included studies, 21 evaluated the relationship between biomechanical metrics and knee OA
Higher knee flexion moment during the second half of the stance phase of gait is associated with the progression of osteoarthritis of the patellofemoral joint on magnetic resonance imaging.
Baseline knee adduction and flexion moments during walking are both associated with 5 year cartilage changes in patients with medial knee osteoarthritis.
Differences in baseline joint moments and muscle activation patterns associated with knee osteoarthritis progression when defined using a clinical versus a structural outcome.
Dynamic and static knee alignment at baseline predict structural abnormalities on MRI associated with medial compartment knee osteoarthritis after 2 years.
Longitudinal changes in tibial and femoral cartilage thickness are associated with baseline ambulatory kinetics and cartilage oligomeric matrix protein (COMP) measures in an asymptomatic aging population.
, and no study reported biomechanical metrics associated with ankle OA. Twenty-two studies (14 cohorts) investigated OA disease progression, and two of these studies (separate cohorts)
Longitudinal changes in tibial and femoral cartilage thickness are associated with baseline ambulatory kinetics and cartilage oligomeric matrix protein (COMP) measures in an asymptomatic aging population.
Community dwellings, sub-study (no Rx) of a previous RCT Age >50 Y, pain >3/10 VAS, KL 2–3.
3D Gait Analysis
MRI
Grade II = 53% Grade III = 7%
Progression: BML- Increase of 1 from a described semi-quantitative scale
Adjusted for age, gender, body mass index, MRI machine, static knee alignment, treatment group and baseline tibiofemoral cartilage defect score or BML scores
Baseline knee adduction and flexion moments during walking are both associated with 5 year cartilage changes in patients with medial knee osteoarthritis.
Differences in baseline joint moments and muscle activation patterns associated with knee osteoarthritis progression when defined using a clinical versus a structural outcome.
Unable to determine attrition (Follow-up in years (Radiographic: NPG: 7.3 (2.3) years PG: 6.8 (2.2) years TKA: NPG: 7.3 (2.0) years PG: 5.9 (2.6) years)
Retrospective secondary analysis of data from participants. Recruited from community and orthopaedic offices. Participants should have functional ability to jog 5 m, walk a city block and climb stairs.
Longitudinal changes in tibial and femoral cartilage thickness are associated with baseline ambulatory kinetics and cartilage oligomeric matrix protein (COMP) measures in an asymptomatic aging population.
Recruited from the MOST Study. Independent walking ability
Pressure walkway
MRI
Not provided
Progression: Increase in WORMS score >1 from 60 to 84 months to follow-up. Within grade WORMS score changes were also considered indicative of worsening.
Risk ratios adjusted for age, BMI and previous injury/surgery.
Dynamic and static knee alignment at baseline predict structural abnormalities on MRI associated with medial compartment knee osteoarthritis after 2 years.
Higher knee flexion moment during the second half of the stance phase of gait is associated with the progression of osteoarthritis of the patellofemoral joint on magnetic resonance imaging.
Recruited from the department of Orthopaedic surgery Kyoto University. Ability to walk without assistive devices
3D Gait Analysis
X-Ray
Minimum JSW at baseline: NPG:3.7 (1.4) mm PG: 2.9 (1.4) mm
Progression Reduction of 0.5 mm/year in JSW
Adjusted for age, gender, body weight, and minimum JSW (mm)
ACR = American college of rheumatology; BLOKS = Boston Leeds osteoarthritis knee score; BML = bone marrow lesions; BMI = body mass index; JSW = joint space width; K/L = Kellgren Lawrence; MAK = mechanical factors in arthritis of the knee; MOST = multicentre osteoarthritis study; MRI = magnetic resonance imaging; NR = not reported; OARSI-OMERACT = osteoarthritis research society international (OARSI)- outcome measures in rheumatology (OMERACT); PASE = physical activity scale for the elderly; PF = patellofemoral joint; Rx = intervention group; RCT = randomised controlled trial; SHOMRI = scoring hip osteoarthritis with MRI; TKA = total knee athroplasty; VAS = visual analog scale; WORMS = whole-organ MRI scoring; WOMAC = western Ontario and Mcmaster universities arthritis index.
ꝋ: Studies did not provide overall summary scores for Age, BMI, %female, therefore mean (SD) for subgroups were reported.
Baseline knee adduction and flexion moments during walking are both associated with 5 year cartilage changes in patients with medial knee osteoarthritis.
Higher knee flexion moment during the second half of the stance phase of gait is associated with the progression of osteoarthritis of the patellofemoral joint on magnetic resonance imaging.
Higher knee flexion moment during the second half of the stance phase of gait is associated with the progression of osteoarthritis of the patellofemoral joint on magnetic resonance imaging.
Baseline knee adduction and flexion moments during walking are both associated with 5 year cartilage changes in patients with medial knee osteoarthritis.
Differences in baseline joint moments and muscle activation patterns associated with knee osteoarthritis progression when defined using a clinical versus a structural outcome.
Dynamic and static knee alignment at baseline predict structural abnormalities on MRI associated with medial compartment knee osteoarthritis after 2 years.
Longitudinal changes in tibial and femoral cartilage thickness are associated with baseline ambulatory kinetics and cartilage oligomeric matrix protein (COMP) measures in an asymptomatic aging population.
. For OA structural outcomes, 14 studies conducted MRI evaluations, 8 studies used radiographic evaluations and two studies reported progression by total knee arthroplasty (TKA)
Differences in baseline joint moments and muscle activation patterns associated with knee osteoarthritis progression when defined using a clinical versus a structural outcome.
Baseline knee adduction and flexion moments during walking are both associated with 5 year cartilage changes in patients with medial knee osteoarthritis.
Longitudinal changes in tibial and femoral cartilage thickness are associated with baseline ambulatory kinetics and cartilage oligomeric matrix protein (COMP) measures in an asymptomatic aging population.
Differences in baseline joint moments and muscle activation patterns associated with knee osteoarthritis progression when defined using a clinical versus a structural outcome.
Dynamic and static knee alignment at baseline predict structural abnormalities on MRI associated with medial compartment knee osteoarthritis after 2 years.
Higher knee flexion moment during the second half of the stance phase of gait is associated with the progression of osteoarthritis of the patellofemoral joint on magnetic resonance imaging.
Study quality ranged from 12 to 20 points out of a maximum 21, and a mean (SD) score of 16.78 (2.55) (Table II). Studies performed strongly (k = 15) with regards to measurement, and data presentation as well as analysis and presentation of results. Majority of studies had appropriate representation of the OA population, with the study setting/baseline characteristics adequately described. Biomechanical metrics were clearly defined, and structural imaging outcomes used reliable scoring systems with blinded assessors. Seven studies
Differences in baseline joint moments and muscle activation patterns associated with knee osteoarthritis progression when defined using a clinical versus a structural outcome.
Dynamic and static knee alignment at baseline predict structural abnormalities on MRI associated with medial compartment knee osteoarthritis after 2 years.
Baseline knee adduction and flexion moments during walking are both associated with 5 year cartilage changes in patients with medial knee osteoarthritis.
Differences in baseline joint moments and muscle activation patterns associated with knee osteoarthritis progression when defined using a clinical versus a structural outcome.
Longitudinal changes in tibial and femoral cartilage thickness are associated with baseline ambulatory kinetics and cartilage oligomeric matrix protein (COMP) measures in an asymptomatic aging population.
Dynamic and static knee alignment at baseline predict structural abnormalities on MRI associated with medial compartment knee osteoarthritis after 2 years.
Higher knee flexion moment during the second half of the stance phase of gait is associated with the progression of osteoarthritis of the patellofemoral joint on magnetic resonance imaging.
Dynamic and static knee alignment at baseline predict structural abnormalities on MRI associated with medial compartment knee osteoarthritis after 2 years.
(91%) reported at least one significant association between a biomechanical variable at baseline and OA disease onset or progression. Eighteen studies found an association between higher values of baseline gait biomechanics and OA progression in the knee and hip, signifying an increased risk of OA progression
Higher knee flexion moment during the second half of the stance phase of gait is associated with the progression of osteoarthritis of the patellofemoral joint on magnetic resonance imaging.
Baseline knee adduction and flexion moments during walking are both associated with 5 year cartilage changes in patients with medial knee osteoarthritis.
Differences in baseline joint moments and muscle activation patterns associated with knee osteoarthritis progression when defined using a clinical versus a structural outcome.
Longitudinal changes in tibial and femoral cartilage thickness are associated with baseline ambulatory kinetics and cartilage oligomeric matrix protein (COMP) measures in an asymptomatic aging population.
found a negative association between higher values of hip adduction moment, toe-out angle and step length (observed in one quintile) with the risk of OA progression in the knee, suggesting a potentially protective association with OA structural progression.
Table IIISummary of the confirmed associations between baseline biomechanical metrics and variables with risk of OA onset or progression in the lower limb regions
Study
Lower limb region
Biomechanical variables/metrics associated with onset and/or progression
Baseline knee adduction and flexion moments during walking are both associated with 5 year cartilage changes in patients with medial knee osteoarthritis.
Differences in baseline joint moments and muscle activation patterns associated with knee osteoarthritis progression when defined using a clinical versus a structural outcome.
Longitudinal changes in tibial and femoral cartilage thickness are associated with baseline ambulatory kinetics and cartilage oligomeric matrix protein (COMP) measures in an asymptomatic aging population.
Dynamic and static knee alignment at baseline predict structural abnormalities on MRI associated with medial compartment knee osteoarthritis after 2 years.
Higher knee flexion moment during the second half of the stance phase of gait is associated with the progression of osteoarthritis of the patellofemoral joint on magnetic resonance imaging.
Appendix D represents all reported biomechanical metrics and variables and their associations with OA disease onset and/or progression. A total of 25 biomechanical metrics were studied for the hip and knee joints, out of which 83 biomechanical variables were evaluated during specific gait cycle time-points.
The majority of evidence supported an association between frontal, sagittal and transverse plane lower limb joint biomechanics and knee OA progression. For OA onset and progression in the medial tibiofemoral joint and individual regions within this compartment, positive associations were found with peak KAM values (5/8 cohorts), higher peak KAM
Longitudinal changes in tibial and femoral cartilage thickness are associated with baseline ambulatory kinetics and cartilage oligomeric matrix protein (COMP) measures in an asymptomatic aging population.
Baseline knee adduction and flexion moments during walking are both associated with 5 year cartilage changes in patients with medial knee osteoarthritis.
Differences in baseline joint moments and muscle activation patterns associated with knee osteoarthritis progression when defined using a clinical versus a structural outcome.
observed through structural worsening. For the patellofemoral joint, two studies investigated and reported positive associations for dynamic joint stiffness
Higher knee flexion moment during the second half of the stance phase of gait is associated with the progression of osteoarthritis of the patellofemoral joint on magnetic resonance imaging.
were associated with hip OA progression in one study each.
Results of meta-analyses
Due to variability of biomechanics and OA outcomes reported, meta-analyses could only be conducted on three biomechanical variables and their association with OA disease progression in the medial tibiofemoral joint. These included peak KAM, KAM impulse and varus thrust. For the KAM variable, three studies reported peak KAM
Differences in baseline joint moments and muscle activation patterns associated with knee osteoarthritis progression when defined using a clinical versus a structural outcome.
There is low quality evidence to suggest that early stance peak KAM was associated with increased odds of medial tibiofemoral OA progression. There was risk of bias present in the assessment of individual studies, and increased incosistency/heterogenity in the meta-analysis. There was dispersion in the sample sizes and therefore, publication bias was detected.
Differences in baseline joint moments and muscle activation patterns associated with knee osteoarthritis progression when defined using a clinical versus a structural outcome.
There is moderate quality evidence to suggest that early stance peak KAM was associated with increased odds of radiographic outcomes of medial tibiofemoral OA progression. There was risk of bias present in individual studies, and moderate inconsistency present in the meta-analsysis. However, the exposure and outcome methods were relavant to the clinical OA population. Whilst publication bias was difficult to assess due to the small number of studies, we decided that there was low bias present as the studies assessed other prognostic factors.
Radiographic outcomes of medial tibiofemoral progression
Differences in baseline joint moments and muscle activation patterns associated with knee osteoarthritis progression when defined using a clinical versus a structural outcome.
There is moderate quality evidence to suggest that varus thrust is associated with increased odds of medial tibiofermoral OA progression. The meta-analysis had a moderate-large effect size, with individual studies having low risk of bias in the assessment. The prognostic factor and imaging outcome were also relevant to the clinical OA population. It was difficult to assess for publication bias given the small number of studies, however, the studies were dispersed and additional secondary analyses were conducted in large samples, therefore publication bias was undetected.
∗ There is low quality evidence to suggest that early stance peak KAM was associated with increased odds of medial tibiofemoral OA progression. There was risk of bias present in the assessment of individual studies, and increased incosistency/heterogenity in the meta-analysis. There was dispersion in the sample sizes and therefore, publication bias was detected.
† There is moderate quality evidence to suggest that early stance peak KAM was associated with increased odds of radiographic outcomes of medial tibiofemoral OA progression. There was risk of bias present in individual studies, and moderate inconsistency present in the meta-analsysis. However, the exposure and outcome methods were relavant to the clinical OA population. Whilst publication bias was difficult to assess due to the small number of studies, we decided that there was low bias present as the studies assessed other prognostic factors.
‡ There is moderate quality evidence to suggest that varus thrust is associated with increased odds of medial tibiofermoral OA progression. The meta-analysis had a moderate-large effect size, with individual studies having low risk of bias in the assessment. The prognostic factor and imaging outcome were also relevant to the clinical OA population. It was difficult to assess for publication bias given the small number of studies, however, the studies were dispersed and additional secondary analyses were conducted in large samples, therefore publication bias was undetected.
Greater magnitudes of early stance peak KAM were associated with increased odds of overall OA progression in the medial tibiofemoral joint (OR: 1.88 [95%CI: 1.08,3.29]; k = 5; average follow-up = 45months; n = 601; I2 = 73%, Fig. 2, low quality evidence) as determined by medial tibiofemoral BMLs, cartilage damage, cartilage defects, KL grade worsening and medial tibiofemoral JSN. Greater magnitudes of early stance peak KAM were associated with increased odds of radiographically-defined OA progression in the medial tibiofemoral joint (OR:3.53 [95%CI: 1.47, 8.48]; k = 3; average follow-up = 64 months; n = 207; I2 = 46%, Fig. 3, moderate quality evidence). Varus thrust presence was associated with increased odds of overall OA disease progression in the medial tibiofemoral joint (OR: 1.97 [95%CI: 1.32, 2.96]; k = 3; average follow-up = 54 months; n = 8,059; I2 = 76%, Fig. 4, moderate quality evidence).
Fig. 2Meta-analysis for the association between early stance peak KAM and medial tibiofemoral OA progression. ∗ = bone marrow lesions and cartilage damage; Combined ꝋ = bone marrow lesions and cartilage defects; JSN = joint space narrowing; KAM = knee adduction moment; KL= Kellgren Lawrence grade.
Fig. 3Meta-analysis for the association between early stance peak KAM and radiographic outcomes of medial tibiofemoral OA progression. JSN = Joint space narrowing; KAM = knee adduction moment; K&L = Kellgren Lawrence grade.
Fig. 4Meta-analysis for the association between varus thrust and medial tibiofemoral OA progression. BML = bone marrow lesions; JSN = joint space narrowing.
Findings from sensitivity analyses with <3 studies are reported in Appendix C. For early stance peak KAM, positive associations remained for MRI- defined BMLs (OR: 1.30 [95%CI: 1.02,1.66]; k = 2; average follow-up = 18 months; n = 394), but not for medial tibiofemoral JSN and cartilage damage or defects in the medial tibiofemoral joint. For KAM impulse, positive associations remained for increased medial tibiofemoral BMLs size (OR: 2.07 [95%CI: 1.17,3.68]; k = 2; average follow-up = 18 months; n = 394), but not for cartilage damage or defects. For the onset of medial tibiofemoral joint OA, varus thrust presence was not associated (OR: 1.08 [95%CI: 0.86,1.36]; k = 2 studies; average follow-up = 72 months; n = 7,822).
Discussion
This systematic review confirms that gait biomechanics are associated with increased odds of OA progression in the hip and knee joints over time. Results confirm that higher early stance peak KAM is associated with 1.88 greater odds of OA progression in the medial tibiofemoral joint (k = 5; n = 601; low quality evidence). We also identified that people who exhibited varus thrust had 1.97 greater odds of medial tibiofemoral OA progression (n = 8,059). Our review also suggests associations between biomechanics and OA progression in the hip and patellofemoral joints.
Meta-analyses demonstrated that peak KAM, KAM impulse and varus thrust were positively associated with OA disease progression. The previous review by Henriksen et al.
Is there a causal link between knee loading and knee osteoarthritis progression? A systematic review and meta-analysis of cohort studies and randomised trials.
, found no significant association between peak KAM and medial tibiofemoral OA progression (OR: 1.90 [95%CI:0.85, 4.25]). The pooled meta-analysis for this previous review included a study
which had an active intervention present after baseline, and therefore was excluded from our review. Since 2014, additional studies have investigated the KAM
Baseline knee adduction and flexion moments during walking are both associated with 5 year cartilage changes in patients with medial knee osteoarthritis.
Differences in baseline joint moments and muscle activation patterns associated with knee osteoarthritis progression when defined using a clinical versus a structural outcome.
Longitudinal changes in tibial and femoral cartilage thickness are associated with baseline ambulatory kinetics and cartilage oligomeric matrix protein (COMP) measures in an asymptomatic aging population.
, and our meta-analysis (k = 5) concluded that peak KAM was positively associated with OA progression in the medial tibiofemoral joint (low-quality evidence). Due to diversity in the KAM variables reported, our meta-analyses were limited to only include five studies. Overall, seven cohorts (9 out of 10 studies) which investigated the KAM variables found a greater risk of knee OA onset
Longitudinal changes in tibial and femoral cartilage thickness are associated with baseline ambulatory kinetics and cartilage oligomeric matrix protein (COMP) measures in an asymptomatic aging population.
Baseline knee adduction and flexion moments during walking are both associated with 5 year cartilage changes in patients with medial knee osteoarthritis.
Differences in baseline joint moments and muscle activation patterns associated with knee osteoarthritis progression when defined using a clinical versus a structural outcome.
with higher KAM values. The KAM impulse considers the magnitude and duration of the KAM over stance and may represent a better indicator of medial tibiofemoral joint loading. Whilst we identified a positive association between KAM impulse and 2.07 greater odds of worsening BMLs in the medial tibiofemoral joint, as only two studies were pooled, the strength of the predictive relationship is uncertain. Greater OA progression odds due to higher KAM values is supported by Trepcyznski et al.
, where medial to lateral loading imbalance was correlated with increased medial compressive forces in individuals who later underwent TKA.
Our meta-analysis of low-quality evidence indicates that higher peak KAM values were associated with 1.88 increased odds of medial tibiofemoral OA progression, with the association size varying between studies. This may be due to follow-up time variability, or sensitivity of imaging outcomes. For instance, larger OR's were reported where follow-up time was greater than 24 months (OR range = 4.3–6.4)
that generally reported higher KAM values in participants also identified larger OR's. In this instance, developing a sensitivity threshold value for peak KAM may help identify individuals at greater risk of OA progression. Miyazaki et al.
. As this was the only study to provide a threshold, further studies are needed to confirm threshold values for evaluating OA progression risk. The variability in findings may also be explained by 1) different imaging methods, 2) different joint regions examined, 3) different joint structures (e.g., cartilage defects, BMLs), 4) variability in the imaging outcomes used and 5) time-frames to detect structural changes. For instance, Chang et al.
, found associations between higher peak KAM values with cartilage loss in the subregions of the knee but did not find an association for the medial tibiofemoral joint as a whole. Further research using standardised imaging outcome measurements are needed to quantify the size of the increased odds more accurately.
Varus thrust is proposed to acutely increase medial tibiofemoral loading via an abrupt lateral shift of the knee during stance, followed by return to a less varus alignment during swing
of moderate quality evidence demonstrated a positive association between varus thrust and increased medial tibiofemoral loading with a 97% increased odds for medial tibiofemoral OA progression. These studies visually identified, and dichotomously categorised varus thrust (present or absent). Two of these studies
, used MRI outcomes to define disease onset and confirmed an association between varus thrust and presence of BMLs in the medial tibiofemoral joint over 24 months, though the same was not found for other imaging outcomes
, used radiographic assessment and defined disease onset as incident KL grade ≥2 and found no association with disease onset over 84 months. As the studies used different imaging tools to assess and define disease onset, it was difficult to combine these results. The WORMS scoring tool used in Wink et al.
,. Further studies using a standardised imaging tool are required to determine if an association exists between varus thrust presence and knee OA onset.
Sagittal and transverse plane biomechanical metrics were also found to be associated with increased risk of medial tibiofemoral OA progression; however, data could not be pooled for meta-analyses due to the small number of findings. Three studies evaluated transverse plane metrics and two of these
Differences in baseline joint moments and muscle activation patterns associated with knee osteoarthritis progression when defined using a clinical versus a structural outcome.
Differences in baseline joint moments and muscle activation patterns associated with knee osteoarthritis progression when defined using a clinical versus a structural outcome.
with medial tibiofemoral JSN. This supports findings from in vitro studies which demonstrate shear loading and the negative effect on cartilage integrity
Baseline knee adduction and flexion moments during walking are both associated with 5 year cartilage changes in patients with medial knee osteoarthritis.
Differences in baseline joint moments and muscle activation patterns associated with knee osteoarthritis progression when defined using a clinical versus a structural outcome.
Baseline knee adduction and flexion moments during walking are both associated with 5 year cartilage changes in patients with medial knee osteoarthritis.
identified an association between maximum knee flexion angle and increased femoral anterior-posterior displacement with medial tibiofemoral OA progression, suggesting that kinematic changes in heel strike and midstance may increase medial knee contact forces
Higher knee flexion moment during the second half of the stance phase of gait is associated with the progression of osteoarthritis of the patellofemoral joint on magnetic resonance imaging.
Higher knee flexion moment during the second half of the stance phase of gait is associated with the progression of osteoarthritis of the patellofemoral joint on magnetic resonance imaging.
Higher knee flexion moment during the second half of the stance phase of gait is associated with the progression of osteoarthritis of the patellofemoral joint on magnetic resonance imaging.
to be associated with patellofemoral progression. It has been postulated that a higher KFM is associated with increased patellofemoral joint reaction forces
whilst a larger step length may have protective effects, however this was only identified in the second largest step length quintile and therefore possibly a chance finding
00941-9&id=gr1.jpg){kind=link}
00941-9&id=gr2.jpg){kind=link}
00941-9&id=gr3.jpg){kind=link}
00941-9&id=gr4.jpg){kind=link}