No established markers can effectively phenotype knee osteoarthritis (OA) patients into subgroups. Infrapatellar fat pad (IPFP) morphology data that can forecast disease symptoms, structural changes, and knee replacement (KR) are sparse and conflicting. This 96-month longitudinal exploratory study aimed to identify which IPFP morphological features were the most effective independent prognostic markers against these outcomes. This longitudinal study analyzed 1075 target knees (one knee per participant) from the Osteoarthritis Initiative (OAI) progression cohort. Structural changes include cartilage, bone marrow lesions (BMLs), and joint effusion volumes assessed using automated and quantitative magnetic resonance imaging systems (MRI). The IPFP global and signal (hyper- and hypo-) intensity volumes and areas were assessed using MRI combined with a newly developed, fully automated neuron-driven technology. Symptoms were evaluated using WOMAC scores. Data on KR was obtained from the OAI database. Data were collected at baseline and 12, 24, 48 and 96 months and analyzed using a mixed model for repeated measures (MMRM) or ANCOVA. The baseline characteristics were mild to moderate knee OA. Over time, disease symptoms (WOMAC), cartilage volume, IPFP global and hypointense signal volumes, and maximal and hypointense signal areas decreased (all p≤0.001). Joint effusion and hyperintense signal volume and area increased (both p≤0.001). Associations were found between IPFP morphologies at inclusion and changes in cartilage volume (hypointense and hyperintense volumes, 48, 96 months, p≤0.04), BML volume (global volume 48 months, p=0.05; hyperintense area, 12 months, p≤0.04), and effusion volume (hypointense volume 48 months and hyperintense volume 96 months, p≤0.05). At inclusion, smaller IPFP sizes (below median) were associated with cumulative KR at 96 months (global and hypointense volumes, p≤0.04 and maximum area, p=0.05). This longitudinal exploratory study, leveraging a fully automated technology, highlights that i) IPFP volume (global and both signals) is superior to area metrics in predicting long-term structural changes in OA, and ii) smaller IPFP volume and area are linked with reduced need for KR. These findings provide new insights into the usefulness of IPFP morphology as a predictive biomarker of knee OA outcomes, offering a new approach to stratifying knee OA patients.

Background: Although knee osteoarthritis (OA) is the most prevalent chronic musculoskeletal debilitating disease, there is no recognized marker to stratify OA patients. The infrapatellar fat pad (IPFP), although not yet as widely studied as other knee tissues, has been recognized as a likely key player in OA. However, there is sparse and conflicting data about which of this tissue’s morphology is the best forecasting marker of the disease’s symptoms, joint structural changes, and patient outcomes. Objectives: We hypothesized that some knee IFPP morphological features at baseline could be used as a prognostic marker of the OA disease symptoms, progression, and knee replacement. We also explored whether IPFP morphology changes are associated with disease symptoms and joint structure changes over time. Methods: By using a longitudinal study (0-96 months), the target knees (n=1075) of participants from the Osteoarthritis Initiative (OAI) progressor cohort were analyzed for their structural changes using X-ray for joint space width (JSW) and quantitative and automated MRI for cartilage, bone marrow lesions (BML) and effusion volumes, as well as the IPFP morphology (total volume and maximal area, and hyperintensity signal volume and area). The symptoms were evaluated using WOMAC and KOOS scores. The knee replacement was as in the OAI database. The analyses were performed at baseline and 12, 24, 48, and 96 months post-inclusion. Changes over time were calculated as the value at the follow-up minus the one at study inclusion, divided by the value at study inclusion. Overtime evolution was analyzed using the mixed model for repeated measures (MMRM) and association by ANCOVA. Results: At baseline, a significant association was found between the IPFP total volume and maximal area, as well as hyperintensity signal volume and area with the cartilage (p≤0.001), effusion (p≤0.001; except IPFP area) and BML (p≤0.027; except hyperintensity signal) volumes. Over time (0-96 months), there was a decrease in disease symptoms (WOMAC, KOOS; p≤0.001), IPFP total volume and maximal area (p≤0.001), JSW (p<0.001), cartilage volume (p<0.001), and an increase in effusion (p<0.001) and hyperintensity signal volume and area (p≤0.001). Importantly, significant associations were found between baseline hyperintensity signal volume with changes in JSW (96 months; p=0.008), cartilage (48 months; p=0.026) and effusion (96 months; p=0.038) volumes, as well as between the hyperintensity area with BML volume (12 months; p=0.044). Regarding the association for the changes of both IPFP morphology and knee structures, the most significant were the changes in the IPFP total volume with JSW (48 months; p<0.001), cartilage (24, 48 months; p≤0.049) and BML (24 months; p<0.02) volumes, and a trend toward significant difference at 12 and 48 months for BML and at 48 months for effusion volumes (p≤0.062). As for the disease symptoms or knee replacement, there was no association either at baseline or with the changes in the studied IPFP morphology. Conclusion: This extensive framework forecasting the progression of knee OA structural alterations revealed that the IPFP hyperintensity signal volume could be used as an early prognostic marker and that changes over time in the IPFP total volume were well associated with changes in the knee structural progression. This study offers a new approach for stratifying OA structural progressors. REFERENCES: NIL. Acknowledgements: The authors would like to thank the Osteoarthritis Initiative (OAI) participants and the Coordinating Center. The OAI is a public-private partnership funded by the National Institutes of Health. A special thanks to ArthroLab Inc. for providing the MRI data for this study. Disclosure of Interests: None declared.

To determine the feasibility of a randomized controlled trial (RCT) examining outdoor walking on knee osteoarthritis (KOA) clinical outcomes and magnetic resonance imaging (MRI) structural changes. This was a 24-week parallel two-arm pilot RCT in Tasmania, Australia. KOA participants were randomized to either a walking plus usual care group or a usual care control group. The walking group trained 3 days/week. The primary outcome was feasibility assessed by changes being required to the study design, recruitment, randomization, program adherence, safety, and retention. Exploratory outcomes were changes in symptoms, physical performance/activity, and MRI measures. Forty participants (mean age 66 years (SD 1.4) and 60

Background Knee osteoarthritis is the most prevalent chronic musculoskeletal debilitating disease. Current treatments are only symptomatic, and to improve this, we need a robust prediction model to stratify patients at an early stage according to the risk of joint structure disease progression. Some genetic factors, including single nucleotide polymorphism (SNP) genes and mitochondrial (mt)DNA haplogroups/clusters, have been linked to this disease. For the first time, we aim to determine, by using machine learning, whether some SNP genes and mtDNA haplogroups/clusters alone or combined could predict early knee osteoarthritis structural progressors. Methods Participants (901) were first classified for the probability of being structural progressors. Genotyping included SNP genes TP63 , FTO , GNL3 , DUS4L , GDF5 , SUPT3H , MCF2L , and TGFA ; mtDNA haplogroups H, J, T, Uk, and others; and clusters HV, TJ, KU, and C-others. They were considered for prediction with major risk factors of osteoarthritis, namely, age and body mass index (BMI). Seven supervised machine learning methodologies were evaluated. The support vector machine was used to generate gender-based models. The best input combination was assessed using sensitivity and synergy analyses. Validation was performed using tenfold cross-validation and an external cohort (TASOAC). Results From 277 models, two were defined. Both used age and BMI in addition for the first one of the SNP genes TP63 , DUS4L , GDF5 , and FTO with an accuracy of 85.0%; the second profits from the association of mtDNA haplogroups and SNP genes FTO and SUPT3H with 82.5% accuracy. The highest impact was associated with the haplogroup H, the presence of CT alleles for rs8044769 at FTO , and the absence of AA for rs10948172 at SUPT3H . Validation accuracy with the cross-validation (about 95%) and the external cohort (90.5%, 85.7%, respectively) was excellent for both models. Conclusions This study introduces a novel source of decision support in precision medicine in which, for the first time, two models were developed consisting of (i) age, BMI, TP63 , DUS4L , GDF5 , and FTO and (ii) the optimum one as it has one less variable: age, BMI, mtDNA haplogroup, FTO , and SUPT3H . Such a framework is translational and would benefit patients at risk of structural progressive knee osteoarthritis.

Exercise therapy is recommended as first line treatment for knee osteoarthritis (OA), but it remains to be sub-optimally applied (1). Movement-evoked pain is a potential barrier to exercise adherence, but recent evidence suggests that such pain can be improved by training (2). Walking programs are low-cost, easily adopted and can be performed outdoors which can minimize the risk of SARS-CoV-2 transmission when in a group (3).To explore the acute pain trajectories of individuals with knee OA during a 24-week outdoor walking intervention. In addition, to explore the effect of pain trajectories and/or baseline characteristics on retention and adherence.Individuals with clinical knee OA and bone marrow lesions (BMLs) on magnetic resonance imaging (MRI) were asked to follow a 24-week walking program. Every week consisted of two one hour supervised group sessions at various outdoor locations and one unsupervised session. At the start and end of every supervised group walk, knee pain was self-reported by participants to their trainer using a numerical rating scale (NRS) (0-10). The difference between the NRS pain values was considered as an acute pain change evoked by that walk. At baseline, the most affected knee of each participant was assessed using the Visual Analogue Scale (VAS) pain, the Western Ontario and McMasters Universities Osteoarthritis Index (WOMAC) pain, stiffness and function, wellbeing (3 questionnaires) and the Osteoarthritis Research Society International (OARSI) recommended strength and performance measures.In total, N = 24 participants started the program of whom N = 7 (29%) withdrew. Pain at the start of each walk decreased from NRS 2.5 (SD 1.6) at the first walk (N = 24) to NRS 0.9 (SD 0.8) at the final walk (N = 17). This pain was estimated to decrease on NRS by -0.04 (95% CI -0.05 to -0.02) per supervised session, p < 0.001 during the first 12 weeks and -0.01 (95% CI -0.02 to -0.004), p = 0.004 during the second twelve weeks of the program. The number (%) of participants who experienced an acute increase in pain decreased from 11 (45.8%) at the first walk to 4 (23.5%) at the last walk.At baseline, non-adherent participants (<70% of group sessions) (N = 11) had lower physical performance scores, including the 30s Chair Stand Test (mean 10 (SD 1.7) stands versus mean 12.0 (SD 1.7) stands, p = 0.011), Fast Past Walk Test (1.23 (SD 0.14) meter per seconds (m/s) vs 1.50 (SD 0.20) m/s, p = 0.001), Six Minute Walk Test (418.8 (SD 75.9) m vs 529 (SD 72.6) m, p = 0.002), compared to adherent participants (N = 13). Non-adherent participants also had less severe self-reported symptoms including WOMAC stiffness (90.7 (SD 44.5) mm vs 121.5 (SD 17.0) mm, p = 0.031), compared to adherent participants. During the first two weeks of walking, acute increases in pain on average (mean ≥0.5 NRS) were reported by a greater number of non-adherent (N = 5 (45.5%)) than adherent participants (n = 4 (30.8%)).This was an exploratory study and results need to be interpreted with caution due to the small sample size. The walking program resulted in clinically important improvements (MCIIs) (≥ 1 on NRS) (4) in start pain and acute pain changes. Improvements in start pain during the first 12-weeks were comparable to improvements measured in the NEMEX program (2) and may suggest that 12 weeks of exercise is sufficient to achieve MCIIs in pain. Improvements in acute changes in pain were smaller, which may have been related to a floor effect (5). Lower physical performance scores at baseline and more acute increases in pain during the first two weeks was associated with non-adherence. Participants with these characteristics may benefit from a lighter introduction to exercise.[1]Bennell KL, et al. The Lancet Regional Health-Western Pacific. 2021;12:100187.[2]Sandal LF, et al. Osteoarthritis and cartilage. 2016;24(4):589-92.[3]Bulfone TC, et al. The Journal of infectious diseases. 2021;223(4):550-61.[4]Perrot S, et al. Pain. 2013;154(2):248-56.[5]McHorney CA, et al. Quality of life research. 1995;4(4):293-307.We thank the participants who made this study possible. We would like to acknowledge the research staff, Kate Probert, Lizzy Reid, Simone Fitzgerald, Claire Roberts, Jasmin Ritchie, Dawn Simpson, and Tim Albion. We also thank Hamish Newsham-West for his contribution to the study design.Stan Drummen: None declared, Saliu Balogun: None declared, Lieke Scheepers Grant/research support from: Competitive Grant Program Inflammation ASPIRE 2020 Rheumatology International Developed Markets from Pfizer, Employee of: previously worked as an Associate Director Epidemiology at the Medical Evidence Observational Research Department at AstraZeneca., Ishanka Munugoda: None declared, aroub lahham: None declared, Kim Bennell: None declared, Rana Hinman: None declared, Michele Callisaya: None declared, Guoqi Cai: None declared, Petr Otahal: None declared, Tania Winzenberg Consultant of: received payment to create educational material by AMGEN, Zhiqiang Wang: None declared, Benny Antony: None declared, Johanne Martel-Pelletier Shareholder of: ArthroLab Inc., Jean-Pierre Pelletier Shareholder of: ArthroLab Inc., François Abram Consultant of: ArthroLab Inc., Employee of: Arthrolab Inc., Graeme Jones Speakers bureau: received payment for a speakers bureau from Novartis, Dawn Aitken: None declared

BACKGROUND:Vastus medialis intramuscular fat has been proposed to be a modifiable determinant of knee cartilage loss in patients with knee osteoarthritis. The objective was to determine whether vastus medialis intramuscular fat relates to osteoarthritis severity and quadriceps muscle strength in patients with non-traumatic and post-traumatic knee osteoarthritis. METHODS:For this cross-sectional study, participants with knee osteoarthritis were classified into two groups: non-traumatic (n = 22; mean age = 60 years) and post-traumatic (n = 19; mean age = 56 years). Healthy adults were included (n = 22; mean age = 59 years). A 3-Tesla magnetic resonance imaging was used to measure vastus medialis cross-sectional area and intramuscular fat. Isometric knee extensor muscle torque was assessed using an isokinetic dynamometer and normalized to body mass (Nm/kg). Knee osteoarthritis severity was assessed using standing antero-posterior radiographs (Kellgren-Lawrence scores). Regression analyses examined relationships between 1) vastus medialis intramuscular fat with knee osteoarthritis severity and osteoarthritis group, after accounting for sex and body mass index, and 2) knee extensor muscle torque with vastus medialis intramuscular fat, after accounting for sex and vastus medialis cross-sectional area. FINDINGS:Vastus medialis intramuscular fat was positively associated with body mass index (B = 0.321, P < 0.001), but not with osteoarthritis severity or group (P > 0.05). Higher vastus medialis intramuscular fat was associated with reduced knee extensor muscle torque (B = -0.040, P = 0.018). INTERPRETATION:Greater vastus medialis intramuscular fat was associated with lower quadriceps muscle strength in patients with knee OA. It is unclear whether this is due to the accumulation of vastus medialis intramuscular fat or other potential factors, such as diet and physical inactivity.

Background Knee osteoarthritis is the most prevalent chronic musculoskeletal debilitating disease. Current treatments are only symptomatic and to improve this, we need a robust prediction model to stratify patients at an early stage according to the risk of joint structure disease progression. Some genetic factors, including single nucleotide polymorphism (SNP) genes and mitochondrial (mt)DNA haplogroups/clusters, have been linked to this disease. Objectives For the first time, we aim to determine, by using machine learning, whether some SNP genes and mtDNA haplogroups/clusters alone or combined could predict early knee osteoarthritis structural progressors. Methods Participants (901) were first classified for the probability of being structural progressors. Genotyping included SNP genes TP63, FTO, GNL3, DUS4L, GDF5, SUPT3H, MCF2L, TGFA, mtDNA haplogroups H, J, T, Uk, others, and clusters HV, TJ, KU, C-others. They were considered for prediction with major risk factors of osteoarthritis, namely, age and body mass index (BMI). Seven supervised machine learning methodologies were evaluated. The support vector machine was used to generate gender-based models. The best input combination was assessed using sensitivity and synergy analyses. Validation was performed using 10-fold cross-validation as well as an external cohort (TASOAC). Results From 277 models, two were defined. Both used age and BMI in addition for the first one of the SNP genes TP63, DUS4L, GDF5, FTO with an accuracy of 85.0%; the second profits from the association of mtDNA haplogroups and SNP genes FTO and SUPT3H with 82.5% accuracy. The highest impact was associated with the haplogroup H, the presence of CT alleles for rs8044769 at FTO , and the absence of AA for rs10948172 at SUPT3H . Validation accuracy with the cross-validation (about 95%) and the external cohort (90.5%, 85.7%, respectively) was excellent for both models. Conclusion This study introduces a novel source of decision support in precision medicine in which, for the first time, two models were developed consisting of i) age, BMI, TP63, DUS4L, GDF5, FTO and ii) the optimum one as it has one less variable: age, BMI, mtDNA haplogroup, FTO, SUPT3H. Such a framework is translational and would be of benefit to patients at risk of structural progressive knee osteoarthritis. Acknowledgements The authors would like to thank the Osteoarthritis Initiative (OAI) participants and Coordinating Center for their work in generating the clinical and radiological data of the OAI cohort and for making them publicly available. The OAI is a public-private partnership comprised of five contracts (N01-AR-2-2258; N01-AR-2-2259; N01-AR-2-2260; N01-AR-2-2261; N01-AR-2-2262) funded by the National Institutes of Health, a branch of the Department of Health and Human Services, and conducted by the OAI Study Investigators. Private funding partners include Merck Research Laboratories; Novartis Pharmaceuticals Corporation, GlaxoSmithKline; and Pfizer, Inc. Private sector funding for the OAI is managed by the Foundation for the National Institutes of Health. This manuscript was prepared using an OAI public use data set and does not necessarily reflect the opinions or views of the OAI investigators, the NIH, or the private funding partners. None of the authors are part of the OAI investigator team. Moreover, the authors are also grateful to the TASOAC participants. A special thanks to ArthroLab Inc. for having provided the MRI data used for classifying structural progressors for each individual. Disclosure of Interests Hossein Bonakdari: None declared, Jean-Pierre Pelletier Shareholder of: ArthroLab Inc., Grant/research support from: Work supported in part by the Osteoarthritis Research Unit of the University of Montreal Hospital Research Centre and the Chair in Osteoarthritis from the University of Montreal., Francisco J. Blanco: None declared, Ignacio Rego-Perez: None declared, Alejandro Durán-Sotuela: None declared, Dawn Aitken: None declared, Graeme Jones: None declared, Flavia Cicuttini: None declared, Afshin Jamshidi Grant/research support from: Received a bursary from the Canada First Research Excellence Fund through the TransMedTech Institute in Canada., François Abram Employee of: was an employee of ArthroLab Inc., Johanne Martel-Pelletier Shareholder of: ArthroLab Inc., Grant/research support from: Work supported in part by the Osteoarthritis Research Unit of the University of Montreal Hospital Research Centre and the Chair in Osteoarthritis from the University of Montreal.

The hallmark of osteoarthritis (OA), the most prevalent musculoskeletal disease, is the loss of cartilage. By using machine learning (ML), we aimed to assess if baseline knee bone curvature (BC) could predict cartilage volume loss (CVL) at one year, and to develop a gender-based model. BC and cartilage volume were assessed on 1246 participants using magnetic resonance imaging. Variables included age, body mass index, and baseline values of eight BC regions. The outcome consisted of CVL at one year in 12 regions. Five ML methods were evaluated. Validation demonstrated very good accuracy for both genders (R ≥ 0.78), except the medial tibial plateau for the woman. In conclusion, we demonstrated, for the first time, that knee CVL at one year could be predicted using five baseline BC region values. This would benefit patients at risk of structural progressive knee OA.

Aim: In osteoarthritis (OA) there is a need for automated screening systems for early detection of structural progressors. We built a comprehensive machine learning (ML) model that bridges major OA risk factors and serum levels of adipokines/related inflammatory factors at baseline for early prediction of at-risk knee OA patient structural progressors over time. Methods: The patient- and gender-based model development used baseline serum levels of six adipokines, three related inflammatory factors and their ratios (36), as well as major OA risk factors [age and bone mass index (BMI)]. Subjects (677) were selected from the Osteoarthritis Initiative (OAI) progression subcohort. The probability values of being structural progressors (PVBSP) were generated using our previously published prediction model, including five baseline structural features of the knee, i.e. two X-rays and three magnetic resonance imaging variables. To identify the most important variables amongst the 47 studied in relation to PVBSP, we employed the ML feature classification methodology. Among five supervised ML algorithms, the support vector machine (SVM) demonstrated the best accuracy and use for gender-based classifiers development. Performance and sensitivity of the models were assessed. A reproducibility analysis was performed with clinical trial OA patients. Results: Feature selections revealed that the combination of age, BMI, and the ratios CRP/MCP-1 and leptin/CRP are the most important variables in predicting OA structural progressors in both genders. Classification accuracies for both genders in the testing stage (OAI) were >80%, with the highest sensitivity of CRP/MCP-1. Reproducibility analysis showed an accuracy ⩾92%; the ratio CRP/MCP-1 demonstrated the highest sensitivity in women and leptin/CRP in men. Conclusion: This is the first time that such a framework was built for predicting knee OA structural progressors. Using this automated ML patient- and gender-based model, early prediction of knee structural OA progression can be performed with high accuracy using only three baseline serum biomarkers and two risk factors. Plain language summary Machine learning model for early knee osteoarthritis structural progression Knee osteoarthritis is a well-known debilitating disease leading to reduced mobility and quality of life – the main causes of chronic invalidity. Disease evolution can be slow and span many years; however, for some individuals, the progression/evolution can be fast. Current treatments are only symptomatic and conventional diagnosis of osteoarthritis is not very effective in early identification of patients who will progress rapidly. To improve therapeutic approaches, we need a robust prediction model to stratify osteoarthritis patients at an early stage according to risk of joint structure disease progression. We hypothesize that a prediction model using a machine learning system would enable such an early identification of individuals for whom osteoarthritis knee structure will degrade rapidly. Data were from the Osteoarthritis Initiative, a National Institute of Health (United States) databank, and the robustness and generalizability of the developed model was further evaluated using osteoarthritis patients from an external cohort. Using the supervised machine learning system (support vector machine), we developed an automated patient- and gender-based model enabling an early clinical prognosis for individuals at high risk of structural progressive osteoarthritis. In brief, this model employed at baseline (when the subject sees a physician) easily obtained features consisting of the two main osteoarthritis risk factors, age and bone mass index (BMI), in addition to the serum levels of three molecules. Two of these molecules belong to a family of factors names adipokines and one to a related inflammatory factor. In brief, the model comprising a combination of age, BMI, and the ratios CRP/MCP-1 and leptin/CRP were found very robust for both genders, and the high accuracy persists when tested with an external cohort conferring the gender-based model generalizability. This study offers a new automated system for identifying early knee osteoarthritis structural progressors, which will significantly improve clinical prognosis with real time patient monitoring.

OBJECTIVE:By using machine learning, our study aimed to build a model to predict risk and time to total knee replacement (TKR) of an osteoarthritic knee.METHODS:Features were from the Osteoarthritis Initiative (OAI) cohort at baseline. Using the lasso method for variable selection in the Cox regression model, we identified the 10 most important characteristics among 1,107 features. The prognostic power of the selected features was assessed by the Kaplan-Meier method and applied to 7 machine learning methods: Cox, DeepSurv, random forests algorithm, linear/kernel support vector machine (SVM), and linear/neural multi-task logistic regression models. As some of the 10 first-found features included similar radiographic measurements, we further looked at using the least number of features without compromising the accuracy of the model. Prediction performance was assessed by the concordance index, Brier score, and time-dependent area under the curve (AUC).RESULTS:Ten features were identified and included radiographs, bone marrow lesions of the medial condyle on magnetic resonance imaging, hyaluronic acid injection, performance measure, medical history, and knee-related symptoms. The methodologies Cox, DeepSurv, and linear SVM demonstrated the highest accuracy (concordance index scores of 0.85, Brier score of 0.02, and an AUC of 0.87). DeepSurv was chosen to build the prediction model to estimate the time to TKR for a given knee. Moreover, we were able to decrease the features to only 3 and maintain the high accuracy (concordance index of 0.85, Brier score of 0.02, and AUC of 0.86), which included bone marrow lesions, Kellgren/Lawrence grade, and knee-related symptoms, to predict risk and time of a TKR event.CONCLUSION:For the first time, we developed a model using the OAI cohort to predict with high accuracy if a given osteoarthritic knee would require TKR, when a TKR would be required, and who would likely progress fast toward this event.

Background/Aim The clinical relevance of MRI knee abnormalities in athletes is unclear. This study aimed to determine the prevalence of MRI knee abnormalities in Australian Rules Football (ARF) players and describe their associations with pain, function, past and incident injury and surgery history. Methods 75 male players (mean age 21, range 16-30) from the Tasmanian State Football League were examined early in the playing season (baseline). History of knee injury/surgery and knee pain and function were assessed. Players underwent MRI scans of both knees at baseline. Clinical measurements and MRI scans were repeated at the end of the season, and incident knee injuries during the season were recorded. Results MRI knee abnormalities were common at baseline (67% bone marrow lesions, 16% meniscal tear/extrusion, 43% cartilage defects, 67% effusion synovitis). Meniscal tears/extrusion and synovial fluid volume were positively associated with knee symptoms, but these associations were small in magnitude and did not persist after further accounting for injury history. Players with a history of injury were at a greater risk of having meniscal tears/extrusion, effusion synovitis and greater synovial fluid volume. In contrast, players with a history of surgery were at a greater risk of having cartilage defects and meniscal tears/extrusion. Incident injuries were significantly associated with worsening symptoms, BML development and incident meniscal damage. Conclusions MRI abnormalities are common in ARF players, are linked to a previous knee injury and surgery history, as well as incident injury but do not dictate clinical symptomatology.

Objective: To examine if relationships between knee osteoarthritis (OA) progression with knee moments and muscle activation during gait vary between patients with non-traumatic and post-traumatic knee OA. Design: This longitudinal study included participants with non-traumatic (n = 17) and post-traumatic (n = 18) knee OA; the latter group had a previous anterior cruciate ligament rupture. Motion capture cameras, force plates, and surface electromyography measured knee moments and lower extremity muscle activation during gait. Cartilage volume change were determined over 2 years using magnetic resonance imaging in four regions: medial and lateral plateau and condyle. Linear regression analysis examined relationships between cartilage change with gait metrics (moments, muscle activation), group, and their interaction. Results: Measures from knee adduction and rotation moments were related to lateral condyle cartilage loss in both groups, and knee adduction moment to lateral plateau cartilage loss in the non-traumatic group only [b = -1.336, 95% confidence intervals (CI) = -2.653 to -0.019]. Generally, lower levels of stance phase muscle activation were related to greater cartilage loss. The relationship between cartilage loss in some regions with muscle activation characteristics varied between non-traumatic and posttraumatic groups including for: lateral hamstring (lateral condyle b = 0.128, 95%CI = 0.003 to 0.253; medial plateau b = 0.199, 95%CI = 0.059 to 0.339), rectus femoris (medial condyle b = -0.267, 95% CI = -0.460 to -0.073), and medial hamstrings (medial plateau; b = -0.146, 95%CI = -0.244 to -0.048). Conclusion: Findings indicate that gait risk factors for OA progression may vary between patients with non-traumatic and post-traumatic knee OA. These OA subtypes should be considered in studies that investigate gait metrics as risk factors for OA progression. (c) 2021 Osteoarthritis Research Society International. Published by Elsevier Ltd. All rights reserved.

OBJECTIVE:To examine whether joint line tenderness and patellofemoral grind from physical examination were associated with cartilage volume loss, worsening of radiographic osteoarthritis, and the risk of total knee replacement.METHODS:This study examined 4,353 Osteoarthritis Initiative participants. For each measurement of joint line tenderness and patellofemoral grind, the patterns were defined as no (none at baseline and at 1 year), fluctuating (present at either time point), and persistent (present at both time points). Cartilage volume loss and worsening of radiographic osteoarthritis over 4 years were assessed using magnetic resonance imaging and radiographs, and total knee replacement over 6 years was assessed.RESULTS:A total of 35.0% of participants had joint line tenderness, and 15.8% had patellofemoral grind. Baseline patellofemoral grind, but not joint line tenderness, was associated with increased cartilage volume loss (1.08% per year versus 0.96% per year; P = 0.02) and an increased risk of total knee replacement (odds ratio [OR] 1.55 [95% confidence interval (95% CI) 1.11-2.17]; P = 0.01). While the patterns of joint line tenderness were not significantly associated with joint outcomes, participants with persistent patellofemoral grind had an increased rate of cartilage volume loss (1.30% per year versus 0.90% per year; P < 0.001) and an increased risk of total knee replacement (OR 2.10 [95% CI 1.30-3.38]; P = 0.002) compared with those participants without patellofemoral grind.CONCLUSION:Patellofemoral grind, but not joint line tenderness, may represent a clinical marker associated with accelerated cartilage volume loss over 4 years and an increased risk of total knee replacement over 6 years. This simple clinical examination may provide clinicians with an inexpensive way to identify those at higher risk of disease progression who should be targeted for surveillance and management.

Background: Intra-articular corticosteroid injections (IACI) are commonly used for the treatment of symptomatic knee osteoarthritis (OA) and therapeutic guidelines have recommended their use. However, their safety regarding the evolution of structural changes remains unknown. Objectives: This study explored the effects of IACI on the evolution of knee OA structural changes assessed by magnetic resonance imaging (MRI). Methods: Participants were selected from the Osteoarthritis Initiative database. In this nested case-control design study, participants who received one treatment with IACI and had MRI exams available at the yearly follow-up visits before (pre-treatment), during (treatment), and after (post-treatment) were defined as “cases”. Each case was matched with one control for age, gender, body mass index (BMI), height, joint space width (JSW), cartilage volume, bone marrow lesion (BML), meniscal extrusion, and Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) pain at baseline. Ninety-three (93) participants fulfilling the inclusion criteria were selected and matched to controls ( n= 93). The study structural variables were MRI (cartilage volume, meniscal thickness, bone marrow lesion (BML), bone curvature), X-rays (JSW), and symptoms (WOMAC pain), assessed at the yearly consecutive visits and changes measured within the follow-up periods. Results: At baseline, the control and treatment groups were balanced. In the pre-treatment period, the cartilage volume loss in the medial compartment was significantly greater in the IACI (p=0.006) compared to the control group, with a numerical trend (p=0.071) in the lateral compartment. In the treatment period, the cartilage loss was not different between groups, with the exception of a significantly greater loss in the lateral compartment in the IACI group (p=0.041). In the post-treatment period there was no difference in the cartilage loss between the groups in both compartments. For the meniscal thickness loss in the pre-treatment period, no difference was found between groups; however, there was a significantly greater loss (p=0.007) during the treatment period in the IACI. In the post-treatment period, the loss of the medial meniscus was similar in both groups. For the lateral meniscus, there was no significant difference at any time between the two groups. The loss in JSW in the pre- and post-treatment periods was not different between groups, but was significantly greater (p=0.011) in the IACI group in the treatment period. The changes in the BML sizes over time were small and similar between groups. For the bone curvature, IACI group showed a smaller change compared to the control (p=0.037) at the treatment period. The WOMAC pain changes in both groups were small and unlikely to be clinically relevant. Conclusion: This study provides evidence that in knee OA, IACI were not associated with the occurrence of any deleterious effect on knee structures post-treatment, including cartilage volume and loss. The increase in the rate loss of medial meniscal thickness, which was associated with a loss of JSW, was a transient phenomenon and its clinical relevance unknown at that time. Acknowledgments: This initiative was funded by a grant from La Chaire en arthrose de l’Université de Montréal, and by ArthroLab Inc. (both in Montreal, Quebec, Canada). Disclosure of Interests: Jean-Pierre Pelletier Shareholder of: ArthroLab Inc., Grant/research support from: TRB Chemedica, Speakers bureau: TRB Chemedica and Mylan, Jean-Pierre Raynauld Consultant of: ArthroLab Inc., François Abram Employee of: ArthroLab Inc., Marc Dorais Consultant of: ArthroLab Inc., Patrice Paiement Employee of: ArthroLab Inc., Johanne Martel-Pelletier Shareholder of: ArthroLab Inc., Grant/research support from: TRB Chemedica

Objectives: The aim was to identify the most important features of structural knee osteoarthritis (OA) progressors and classification using machine learning methods. Methods: Participants, features and outcomes were from the Osteoarthritis Initiative. Features were from baseline (1107), including articular knee tissues (135) assessed by quantitative magnetic resonance imaging (MRI). OA progressors were ascertained by four outcomes: cartilage volume loss in medial plateau at 48 and 96 months (Prop_CV_48M, 96M), Kellgren–Lawrence (KL) grade ⩾ 2 and medial joint space narrowing (JSN) ⩾ 1 at 48 months. Six feature selection models were used to identify the common features in each outcome. Six classification methods were applied to measure the accuracy of the selected features in classifying the subjects into progressors and non-progressors. Classification of the best features was done using an automatic machine learning interface and the area under the curve (AUC). To prioritize the top five features, sparse partial least square (sPLS) method was used. Results: For the classification of the best common features in each outcome, Multi-Layer Perceptron (MLP) achieved the highest AUC in Prop_CV_96M, KL and JSN (0.80, 0.88, 0.95), and Gradient Boosting Machine for Prop_CV_48M (0.70). sPLS showed the baseline top five features to predict knee OA progressors are the joint space width, mean cartilage thickness of the medial tibial plateau and sub-regions and JSN. Conclusion: In this comprehensive study using a large number of features ( n = 1107) and MRI outcomes in addition to radiological outcomes, we identified the best features and classification methods for knee OA structural progressors. Data revealed baseline X-ray and MRI-based features could predict early OA knee progressors and that MLP is the best classification method.

Objective. The infrapatellar fat pad (IPFP) has been associated with knee osteoarthritis onset and progression. This study uses machine learning (ML) approaches to predict serum levels of some adipokines/related inflammatory factors and their ratios on knee IPFP volume of osteoarthritis patients.Methods. Serum and MRI were from the OAI at baseline. Variables comprised the 3 main osteoarthritis risk factors (age, gender, BMI), 6 adipokines, 3 inflammatory factors, and their 36 ratios. IPFP volume was assessed on MRI with a ML methodology. The best variables and models were identified in Total-cohort (n = 678), High-BMI (n = 341) and Low-BMI (n = 337), using a selection approach based on ML methods. Results. The best model for each group included three risk factors and adipsin/C-reactive protein combined for Total-cohort, adipsin/chemerin; High-BMI, chemerin/adiponectin HMW; and Low-BMI, interleukin-8. Gender separation improved the prediction (13–16%) compared to the BMI-based models. Reproducibility with osteoarthritis patients from a clinical trial was excellent (R: female 0.83, male 0.95). Pseudocodes based on gender were generated.Conclusion. This study demonstrates for the first time that the combination of the serum levels of adipokines/inflammatory factors and the three main risk factors of osteoarthritis could predict IPFP volume with high reproducibility, with the superior performance of the model accounting for gender separation.